COUV PDF
25/06/03
16:54
Page 1
Mycosphaerella leaf spot
diseases of bananas:
present status
and outlook
Proceedings of the 2nd International workshop
on Mycosphaerella leaf spot diseases held in San José,
Costa Rica, 20-23 May 2002
L. Jacome, P. Lepoivre, D. Marin, R. Ortiz, R. Romero
and J.V. Escalant, editors
Document2
25/06/03
17:00
Page 1
The mission of the International Network for the Improvement of Banana and Plantain (INIBAP) is
to sustainably increase the productivity of banana and plantain grown on smallholdings for domestic consumption and for local and export markets.
The programme has four specific objectives:
• To organize and coordinate a global research effort on banana and plantain, aimed at the development, evaluation and dissemination of improved cultivars and at the conservation and use of
Musa diversity
• To promote and strengthen collaboration and partnerships in banana-related research activities
at the national, regional and global levels
• To strengthen the ability of NARS to conduct research and development activities on bananas and
plantains
• To coordinate, facilitate and support the production, collection and exchange of information and
documentation related to banana and plantain.
INIBAP is a programme of the International Plant Genetic Resources Institute (IPGRI), a Future
Harvest centre.
The International Plant Genetic Resources Institute is an autonomous international scientific
organization, supported by the Consultative Group on International Agricultural Research (CGIAR).
IPGRI’s mandate is to advance the conservation and use of genetic diversity for the well being of
present and future generations. IPGRI’s headquarters is based in Rome, Italy, with offices in another
19 countries worldwide. It operates through three programmes: (1) the Plant Genetic Resources
Programme, (2) the CGIAR Genetic Resources Support Programme, and (3) the International Network
for the Improvement of Banana and Plantain (INIBAP).
The international status of IPGRI is conferred under an Establishment Agreement which, by January
2000, had been signed and ratified by the Governments of Algeria, Australia, Belgium, Benin, Bolivia,
Brazil, Burkina Faso, Cameroon, Chile, China, Congo, Costa Rica, Côte d’Ivoire, Cyprus, Czech Republic,
Denmark, Ecuador, Egypt, Greece, Guinea, Hungary, India, Indonesia, Iran, Israel, Italy, Jordan, Kenya,
Malaysia, Mauritania, Morocco, Norway, Pakistan, Panama, Peru, Poland, Portugal, Romania, Russia,
Senegal, Slovakia, Sudan, Switzerland, Syria, Tunisia, Turkey, Uganda and Ukraine.
Cover illustration: Microscope photo of the causal agent of black leaf streak disease, Paracercospora fijiensis
(photo: J. Carlier, CIRAD).
Citation: Jacome L., P. Lepoivre, D. Marin, R. Ortiz, R. Romero and J.V. Escalant (eds). 2003. Mycosphaerella leaf spot
diseases of bananas: present status and outlook. Proceedings of the Workshop on Mycosphaerella leaf spot diseases held in San Jose, Costa Rica on 20-23 May 2002. The International Network for the Improvement of
Banana and Plantain, Montpellier, France.
INIBAP-ISBN: 2-910810-57-7
© International Plant Genetic Resources Institute 2003
IPGRI Headquarters
Via dei Tre Denari 472/a
000 57 Maccarese (Fiumicino)
Rome, Italy
INIBAP
Parc Scientifique Agropolis II
34397 Montpellier Cedex 5
France
MyLsd 17x24
25/06/03
15:15
Page 1
Mycosphaerella leaf spot
diseases of bananas:
present status
and outlook
Proceedings of the 2nd International workshop
on Mycosphaerella leaf spot diseases held in San José,
Costa Rica, 20-23 May 2002
L. Jacome, P. Lepoivre, D. Marin, R. Ortiz, R. Romero
and J.V. Escalant, editors
MyLsd 17x24
25/06/03
15:15
Page 3
Acknowledgments
INIBAP would like to thank all those who helped in the organization of the 2nd
International workshop on Mycosphaerella leaf spot diseases and contributed to
the publication of these proceedings.
CATIE, Chiquita, CORBANA, Dole, EARTH, Lapanday, Syngenta,TADECO and Total Fina
Elf for their financial support to the organization of the meeting and the publication
of these proceedings.
Jorge A. Sandoval (CORBANA), Galileo Rivas (CATIE), Franklin Rosales and Luis
Pocasangre (INIBAP-LAC) and Ronald Madrigal and Arllen Carpio (EARTH) for
helping organize the workshop.
Luis Jacome, Philippe Lepoivre, Douglas Marin, Rodomiro Ortiz and Ronald Romero
for efficiently chairing the sessions and for their work as scientific editors.
Jean-Vincent Escalant and Claudine Picq for overseeing the organization of the
workshop and the production of these proceedings.
Andrew Entwistle and Anne Vézina for the technical editing of these proceedings.
3
MyLsd 17x24
25/06/03
15:15
Page 5
Contents
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction to workshop
Overview of progress and results since the first international workshop
on Mycosphaerella leaf spot diseases of bananas in 1989
7
X. MOURICHON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Session 1 – Impact of Mycosphaerella leaf spot diseases of bananas
Introduction - The spread, detection and impact of black leaf streak disease
and other Mycosphaerella species in the 1990s
R. A. ROMERO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
The distribution and importance of the Mycosphaerella leaf spot diseases of banana
D. R. JONES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Integrating morphological and molecular data sets on Mycosphaerella,
with specific reference to species occurring on Musa
P. W. CROUS, J. Z. GROENEWALD, A. APTROOT, U. BRAUN, X. MOURICHON and J. CARLIER . . . . . . . . . . . . . . . 43
Improved PCR-based detection of Sigatoka disease and black leaf streak disease
in Australian banana crops
J. HENDERSON, K. GRICE, J. PATTEMORE, R. PETERSON and E. AITKEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Impact of minor Mycosphaerella pathogens on bananas (Musa) in South Africa
A. VILJOEN, A. K. J. SURRIDGE and P. W. CROUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Economic impact and management of black leaf streak disease in Cuba
L. PÉREZ VICENTE, J. M. ALVAREZ and M. PÉREZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Management of black leaf streak disease in tropical Asia
A. B. MOLINA and E. FABREGAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Impact of Mycosphaerella spp. in Brazil
Z. J. MACIEL CORDEIRO and A. PIRES DE MATOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Poster - Fungi associated with banana foliage in South Africa
A. K. J. SURRIDGE, A. VILJOEN and F. C. WEHNER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Session 2 – Population biology and epidemiology
Introduction - Population biology and epidemiology
L. H. JÁCOME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Airborne dispersal of Mycosphaerella fijiensis
P. J. A. BURT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Genetic differentiation in Mycosphaerella leaf spot pathogens
J. CARLIER, H. HAYDEN, G. RIVAS, M.-F. ZAPATER, C. ABADIE and E. AITKEN . . . . . . . . . . . . . . . . . . . . . . . . 123
Development and application of molecular markers
in Mycosphaerella populations in Colombia
C. MOLINA, S. APONTE, A. GUTIÉRREZ, V. NÚÑEZ and G. KAHL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Poster - An electrophoretic karyotype for Mycosphaerella fijiensis
L. CONDE-FERRÁEZ, CECILIA M. RODRÍGUEZ, L. PERAZA-ECHEVERRÍA and A. JAMES . . . . . . . . . . . . . . . . . . . . 141
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
5
MyLsd 17x24
25/06/03
15:15
Page 6
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Session 3 – Host-pathogen interactions
Introduction - Banana–Mycosphaerella fijiensis interactions
P. LEPOIVRE, J. P. BUSOGORO, J. J. ETAME, A. EL HADRAMI, J. CARLIER, G. HARELIMANA,
X. MOURICHON, B. PANIS, A. STELLA RIVEROS, G. SALLÉ, H. STROSSE and R. SWENNEN . . . . . . . . . . . . . . . . . 151
Efficiency and durability of partial resistance against black leaf streak disease
C. ABADIE, A. ELHADRAMI, E. FOURÉ and J. CARLIER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Poster - Early evaluation of black leaf streak resistance by using mycelial suspensions
of Mycosphaerella fijiensis
Y. ALVARADO CAPÓ, M. LEIVA MORA, M. A. DITA RODRÍGUEZ, M. ACOSTA, M. CRUZ,
N. PORTAL, R. GÓMEZ KOSKY, L. GARCÍA, I. BERMÚDEZ and J. PADRÓN . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
Session 4 – Genetic improvement for the management of resistance
Introduction - Genetic improvement for a sustainable management of resistance
K. CRAENEN and R. ORTIZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Conventional breeding of bananas
C. JENNY, K. TOMEKPÉ, F. BAKRY and J.V. ESCALANT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Transgenic approaches for resistance to Mycosphaerella leaf spot diseases in Musa spp.
R. SWENNEN, G. ARINAITWE, B.P.A. CAMMUE, I. FRANCOIS, B. PANIS, S. REMY, L. SÁGI , E. SANTOS,
H. STROSSE and I. VAN DEN HOUWE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Mutagenesis and somaclonal variation to develop new resistance
to Mycosphaerella leaf spot diseases
N. ROUX, A. TOLOZA, J. P. BUSOGORO, B. PANIS, H. STROSSE, P. LEPOIVRE,
R. SWENNEN and F. J. ZAPATA-ARIAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Reaction of banana genotypes to black leaf streak disease in the State of Acre in Brazil
M. DE J. B. CAVALCANTE, A. DA S. LEDO, F. H. S. COSTA, Z. J. M. CORDEIRO and A. P. MATOS . . . . . . . . . . . . 251
The International Musa testing programme (IMTP): a worldwide programme
to evaluate elite Musa cultivars
J.-V. ESCALANT
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Session 5 – Integrated disease management
Management of Mycospharella leaf spot diseases in Australia
R. PETERSON, K. GRICE and S. DE LA RUE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Spread and management of black leaf streak disease in the Dominican Republic
P. E. JORGE and T. POLANCO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Microbiological control of black leaf streak disease
A. STELLA RIVEROS, C. INÉS GIRALDO and A. GAMBOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Precision agriculture to improve management decisions and field research
E. SPAANS and L. QUIROS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Poster - The role of managing resistance to fungicides in maintaining
the effectiveness of integrated strategies to control black leaf streak disease
S. KNIGHT, M. WIRZ, A. AMIL, A. HALL and M. SHAW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
List of participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
6
MyLsd 17x24
25/06/03
15:15
Page 7
Foreword
The rapid expansion in the 1980s of black leaf streak disease, which is caused by
Mycosphaerella fijiensis, resulted in such damage to small producers that it
encouraged INIBAP to organize the 1st International workshop on Sigatoka leaf spot
diseases of Banana held in San José, Costa Rica in March 1989. Coming 13 years
after this meeting, the 2nd International workshop on Mycosphaerella leaf spot diseases
of bananas provided a timely opportunity to analyse the current situation regarding
Mycosphaerella leaf spot diseases at the global level.
Black leaf streak disease has been spreading for more than 20 years and is now
reported in most parts of the world. During that time, considerable research
efforts have been expended to develop alternatives to allow small producers to
continue producing banana and plantain. Efforts to create new resistant varieties were
part of a broad spectrum of activities including classical and modern tools for genetic
improvement. Studies are ongoing to develop a better understanding of host-pathogen
interactions. The epidemiology, distribution and population structure of the
Mycosphaerella pathogens are being investigated at national, regional and international levels. Research is also conducted to develop new methods to control the
disease based on a rational use of fungicides.
All those who are involved and interested in the sustainability of the small banana
and plantain producers know that the state of the research and the impact of
Mycospharella leaf spot diseases have radically changed over the last decade. Sigatoka
disease (caused by Mycosphaerella musicola) is still important in some parts of the
world and a previously undescribed leaf spot disease, eumusae leaf spot disease
(caused by Mycosphaerella eumusae), has recently been discovered in southern and
southeastern Asia.
By organizing this workshop, in collaboration with EARTH, CORBANA and CATIE
and in the framework of PROMUSA, INIBAP hopes to strengthen collaborations to
ensure that the benefits of the research efforts reach the smallholders and to accelerate
the creation of new varieties resistant to Mycospharella leaf spot diseases.
7
MyLsd 17x24
25/06/03
15:15
Page 9
Editorial note
Throughout the proceedings, black leaf streak disease, also known as black Sigatoka, is used to
refer to the disease caused by Mycosphaerella fijiensis, Sigatoka disease, also known as yellow
Sigatoka, refers to the disease caused by Mycosphaerella musicola, and eumusae leaf spot disease,
which was called Septoria leaf spot disease when first identified, refers to the disease caused by
Mycosphaerella eumusae.
MyLsd 17x24
25/06/03
15:15
Page 11
Introduction to workshop
Overview of progress and results
since the first international workshop
on Mycosphaerella leaf spot diseases
of bananas in 1989
X. Mourichon
At the San José workshop, 1989, the main topics about banana pathogens were:
1) improvements in knowledge,
2) geographical distribution,
3) epidemiology,
4) mechanisms of host-parasite interactions,
5) sources of resistance and genetic improvement in Musa and
6) efficacy of new fungicides and their use.
This paper reviews and evaluates progress made in recent years, especially the
major developments, and highlights topics in which efforts have perhaps not been
sufficiently sustained.
In general, the main results obtained on Mycosphaerella leaf spot diseases over
the past 10 years have been widely published in refereed journals together with several
reviews. The last book edited by CABI (Jones 2000) is a very good synthesis of present
knowledge of Mycosphaerella leaf spot diseases of bananas.
CIRAD-AMIS, Montpellier, France
11
MyLsd 17x24
5/08/03
7:51
Page 12
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Identification, taxonomy and diagnosis
In 1989, the discussions at the San José meeting concentrated on the availability of
diagnostic methods for the pathogens causing Sigatoka disease, black leaf streak disease
and black Sigatoka. The main observation was the risk of confusing the species when
diagnosis was based on the observation of symptoms alone.
In situ or in vitro observation of the anamorph was considered to be the most reliable
method to identify the diseases:
1) the anamorph Cercospora musae, now Pseudocercospora musae, for Sigatoka
disease caused by Mycosphaerella musicola,
2) the anamorph Cercospora fijiensis, now Paracercospora fijiensis, for black leaf
streak disease caused by Mycosphaerella fijiensis,
3) the two anamorphs for black Sigatoka which at the time were attributed
to Mycosphaerella fijiensis var. difformis were in reality the anamorphs of
M. fijiensis and M. musicola. The species was described only in Latin America.
There has been much work on these aspects in recent years, using the methods
developed for molecular taxonomy on different populations of the pathogens
(Carlier et al., 1994, 2000; Johanson and Jeger, 1993; Johanson et al., 1994).
Molecular markers are highly sensitive at discriminating between fungal
species and have clarified the taxonomy of the banana pathogens.
Molecular markers have made it possible to distinguish clearly between
M. fijiensis and M. musicola and to confirm that M. fijiensis and M. fijiensis var.
difformis are synonymous (Carlier et al., 1994).
More recently, markers have been used to identify a new species pathogenic to
banana, Mycosphaerella eumusae. Initially, M. eumusae was thought to have a
Septoria anamorph (Carlier et al., 2000) therefore the new disease was called
Septoria leaf spot disease. Additional detailed work on the morphotaxonomy
attributed M. eumusae with a Pseudocercospora anamorph and the name was
changed to eumusae leaf spot disease, ELSD (Crous and Mourichon, 2002).
Other methods based on serology are still being developed. The methods are
intended above all to be sufficiently simple to diagnose the early stages of disease,
for example, within the framework of preventive control measures with rational
use of fungicides (Etienne et al., 1995). More specific methods to identify
pathogens as well as opportunist or non-pathogenic species of Mycosphaerella are
also being developed.
The use of molecular markers has provided a great deal of information in recent
years. In particular, molecular markers have made it possible to perform many
analyses of genetic diversity, mainly in M. fijiensis.
A high level of genetic diversity has been demonstrated in M. fijiensis
particularly in the genetic structure of populations at a macrogeographical
scale (Carlier et al., 1996). The diversity and geographical distribution of popu12
MyLsd 17x24
25/06/03
15:15
Page 13
Introduction to workshop
X. Mourichon
lations are mainly explained by genetic recombination in the teleomorph of
Mycosphaerella. Genetic differences have also been observed at smaller scales e.g.
at a field scale (Müller et al., 1997).
M. musicola also shows considerable intraspecific diversity, again involving sexual
reproduction (Hayden et al., 2000, 2002).
There is universal agreement about the extent of genetic diversity in M. fijiensis
and M. musicola, and the implications for the capacity of the two fungi to evolve.
Thus, genetic diversity must be taken into account when devising strategies to
improve disease resistance in banana. A knowledge of the genetic variation in
different geographical populations of the pathogens is important for the management
of resistance genes.
Geographical distribution of Mycosphaerella
Black leaf streak disease (BLSD) and Sigatoka disease (SD) are widespread in the main
banana production zones, particularly Southeast Asia, which is the zone of origin
of the pathogens, and the Pacific, and also in Latin America and Africa (Jones, 2000).
Since the San José workshop, the main change has been the rapid spread of BLSD
in Latin America from Central America northwards to Mexico and Florida, USA, and
southwards to Colombia, Peru, Venezuela, Bolivia and Brazil. BLSD has spread to
the Caribbean e.g. Cuba, Jamaica, the Dominican Republic and Haiti, and the rest
of the Caribbean arc is threatened.
BLSD has also spread to the western central and eastern parts of the African
continent and, recently, to Madagascar. M. fijiensis is no longer confined to northern
Australia and is a new constraint that has to be managed in other Australian
commercial plantations.
Very little information is available on the geographical distribution of the new
species, M. eumusae, cause of ELSD. The species appears to be centred in India but
probably has a wider distribution and requires detailed study.
Finally, a fourth species, M. musae, causing speckle disease, is widespread
throughout the world but generally causes little damage except in the sub-tropical
areas of Australia and South Africa.
Epidemiology of Mycosphaerella leaf spot diseases
At the San José meeting, it was recognized that knowledge of the epidemiology of
Mycosphaerella leaf spot diseases was weak. Just one recommendation regarding
“the urgent need to concentrate efforts on a better understanding of the different
epidemiological components” was made.
Nevertheless, important research (unpublished) had been done on various epidemiological aspects of BLSD in Latin America and Africa. The work was mainly on the effects
of abiotic factors on the different phases of monocyclic infection by M. fijiensis i.e.
infection processes, incubation period, rates of symptom development, reproduction and
dispersal (Gauhl, 1994; Fouré, 1992; Rutter et al., 1998; Smith et al., 1997).
At the time, this work aimed to make the control strategies more scientific but
the question is whether the results were sufficiently exploited and used.
13
MyLsd 17x24
25/06/03
15:15
Page 14
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Other biotic factors e.g. genetic host resistance also affect parts of the infection
cycle (see: management of genetic resistance).
Host-pathogen interactions
In the past, host resistance was evaluated in the field under natural, and hence variable,
pathogen pressure, and in environmental conditions that differed considerably between
locations. Environmental factors can have major effects on the expression of resistance,
especially partial resistance.
Variability was reduced by using tissue-cultured plants inoculated with different
isolates of Mycosphaerella. The plants could be maintained in controlled environmental
chambers but the system was criticised for such things as using host plants that were
too young, which could give rise to resistant phenotypes that were not true-to-type
and results that were not reproducible. A miniaturized technique was thus developed.
Using leaf fragments kept under artificial survival conditions, made it possible to work
with older leaves, to work under closely controlled environmental conditions, and to
allow the different banana genotypes to express the type and level of resistance to
M. fijiensis typical of those expressed in the field. Crucially, all stages of development
in a monocyclic infection were represented in the miniaturised system. The method can
be used to analyse the nature of host-parasite interactions and to study the variability
in virulence and aggressiveness in different populations of the pathogens.
Research over the last 10 years has made use of various models: banana plants in
natural conditions, young plants or leaves under artificial survival conditions, providing
complementary data about the nature of compatible and incompatible interactions.
Host-pathogen interactions occur in nature as the following phenotypes: very or
highly resistant banana cultivars and partially resistant bananas with resistance
varying from very marked partial resistance to very susceptible. All banana germplasm
can be classified by these types of behaviour (Fouré et al., 2000). Only a few wild, and
cultivated diploid and triploid acuminata are resistant to BLSD. High resistance in the
triploid acuminata is present only in cultivars of the Ibota subgroup.
Phenotypes with different levels of partial resistance appear to be widely distributed
among all diploid and triploid acuminata and balbisiana genotypes. Genome B seems
to give higher levels of partial resistance.
The idea that the relationships between Mycosphaerella and banana were based on
compatible and incompatible interactions was proposed at the San José workshop.
Compatible interactions refer to susceptible, or partially resistant bananas that display
different degrees of partial resistance. M. fijiensis can complete its entire infection cycle
under this type of interaction. It was also suggested that partial resistance might result
from a constitutive polyphenolic compound.
The various cytological, ultrastructural and biochemical studies, and genetic
analyses of different varieties of banana with different levels of partial resistance, clearly
show that proanthocyanidins play a role in partial resistance. It is also
clear that the compounds are not essential for the expression of partial resistance.
Proanthocyanidin is probably involved in the rate of lesion elongation, but other factors
may be involved at other stages of the infection monocycle and should be identified
(Beveraggi et al., 1995; Mourichon et al., 2000).
14
MyLsd 17x24
25/06/03
15:15
Page 15
Introduction to workshop
X. Mourichon
Using a model under controlled conditions, it is fairly easy to dissect partial resistance
and evaluate the importance of certain sequences of monocyclic infection, e.g.
incubation period, rate of lesion development, effectiveness of infection, latent periods
and different parameters of sexual and asexual sporulation. Recent studies suggest that
the presence of several components of resistance act on different stages of the infection
cycle. Depending on the banana cultivar, partial resistance may be the result of different
mechanisms. Thus, similar expressions of partial resistance could depend on different
genetic interactions, e.g. efficacy of infection or level of sexual reproduction. These
possibilities should be taken into account by breeders, for whom partial resistance is a
major objective.
Partial resistance in banana is important because it is considered to be more durable.
However, the great diversity of pathogen populations and their capacity to evolve should
be taken into account. Specific interactions between the pathogen and the host plant
must be established for certain sequences of the infectious monocycle (Abadie et al.,
2001a, b). It is possible that some specific interactions may select for more aggressive
pathotypes of Mycosphaerella. The result would be gradual erosion, rather than a sudden
decrease, of partial resistance.
Incompatible interactions. Banana cultivars that are very resistant rapidly block
progress of the fungus in the early stages of disease. At the San José workshop, it was
suggested that such behaviour could be governed by an active defence mechanism.
Studies of host-parasite interactions using cytology, particularly at the ultrastructural
scale, provided accurate images of the interactions that occur after inoculation. There
is clear evidence of active mechanisms such as cell collapse occurring after penetration
of stomata (Beveraggi et al., 1995; Mourichon et al., 2000). Similarly, hypersensitive
reactions have been elicited, and necrosis induced experimentally by fungal compounds
of high molecular weight.
Other research reported at the San José workshop demonstrated that phytotoxic
compounds or toxins were released by M. musicola and M. fijiensis. This raised the
question of the role of these compounds in the infection process. Breeders were interested
in the use of toxic compounds in schemes for the early selection of bananas resistant
to M. fijiensis, in particular.
During the last decade, a large number of phytotoxic compounds produced by
M. fijiensis have been described in the literature, e.g. juglone which displays a high
level of biological activity (Stierle et al., 1991).
Several research projects have shown that such compounds are not primary
determinants of the disease but are secondary determinants of pathogenicity (Harelimana
et al., 1997). The role of these compounds as agents in the selection of resistance, as
originally considered, deserves discussion, in particular for studies on partial resistance.
Breeding for resistance to Mycosphaerella leaf spot
diseases
At the San José meeting, it was stated that “little is known of the genetics and
inheritance of resistance to Sigatoka diseases”. Several strategies and programmes
for genetic improvement were presented but these had been developed mainly
for the control of Fusarium wilt.
15
MyLsd 17x24
25/06/03
15:15
Page 16
Mycosphaerella leaf spot diseases of bananas: present status and outlook
A great deal of effort has been made over the past decade by several
institutions that started programmes to breed for resistance to BLSD in dessert
and cooking bananas. The priority of the breeding programmes has been to search
for high levels of partial resistance, which is considered to be more durable in
the presence of diverse and evolving populations of a pathogen. In some breeding
programmes, molecular marker–assisted selection was developed to introduce
resistance. Several partially resistant hybrids were produced and tested in multisite setups such as INIBAP’s International Musa testing programme. Some hybrids
survived the validation stage and were distributed more widely.
Other breeding approaches were developed using biotechnology, and in
particular the production of transgenic plants using genes coding for antifungal
proteins (AFPs) (see session 4: R. Swennen). However, this approach has the
disadvantage of conferring monogenic resistance and is considered unstable in the
presence of diverse populations of Mycosphaerella species. Nevertheless, the strategy
deserves further study. For example, introducing specific resistance genes in
bananas that possess a high level of partial resistance might be an attractive
approach.
Control strategies
In San José, there was much discussion about the potential of new fungicides, rational
ways of using them, the advantages of forecasting systems, and the management
of resistance to fungicides. It was emphasized that effective and rational control
required a greater knowledge of different aspects of the epidemiology of the
pathogens.
It is generally agreed that this theme is probably the one which received the least
attention and hence could still produce important results. The use of fungicides
remains the strategy by which other strategies are compared. In the past, the selection
pressure by different active ingredients has given rise to the disastrous situation where
fungicide-resistant pathotypes are continuously selected. This strategy is no longer
acceptable in a society increasingly concerned about the environment (Romero, 2000).
In future, the aim will be to propose alternatives to chemical control but without
excluding them completely. Integrated control strategies are needed that combine
several different methods of control, methods which individually are only partially
effective. Integrated control strategies would have to be adapted to the different
farming systems of banana production on a large scale and production of plantains
and cooking bananas on smallholdings where rational chemical control is difficult.
1) Chemical control can still be considered, provided that its use is strictly limited.
Forecasting methods should be improved or adapted to different environmental
conditions.
2) Control measures based on cultural practices are known to affect inoculum
pressure in the field e.g. leaf removal, methods of irrigation, management of
planting density. There is potential for the information to be used more
effectively.
16
MyLsd 17x24
25/06/03
15:15
Page 17
Introduction to workshop
X. Mourichon
3) The use of genetic resistance is important for the future. With the production
of resistant varieties as an objective, it is necessary to consider strategies to
manage resistance, in order to maximize the durability of resistance.
The key link between these three aspects is the need to obtain more information
about the epidemiology of these pathogens, to make better use of what has been
achieved in the past and to propose new lines of research into qualitative and
quantitative aspects of epidemiology. A particular effort in modelling is expected
to take into account the dynamics of pathogen populations.
References
Abadie C., A. El Hadrami and J. Carlier. 2001a. Banana partial resistance against
Mycosphaerella fijiensis: studies of efficiency and durability. P. 43 in Symposium on
Durable Disease Resistance, Wageningen, The Netherlands, November 2001.
Abadie C., A. El Hadrami, G. Rivas, M.F. Zapater and J. Carlier. 2001b. Studies of
Mycosphaerella fijiensis population structure and partial resistance of bananas. INFOMUSA
10(1): XIV-XV.
Beveraggi A., X. Mourichon and G. Salle. 1995. Etude des interactions hôte-parasite chez
des bananiers sensibles et résistants inoculés par Cercospora fijiensis (Mycosphaerella
fijiensis) responsable de la maladie des raies noires. Canadian Journal of Botany 73:13281337.
Carlier J., X. Mourichon, D. Gonzales de León, M.F. Zapater and M.H. Lebrun. 1994. DNA
restriction fragment length polymorphisms in Mycosphaerella species causing banana leaf
spot diseases. Phytopathology 84:751-756.
Carlier J., M.H. Lebrun, M.F. Zapater, C. Dubois and X. Mourichon. 1996. Genetic structure
of the global population of banana black leaf streak fungus Mycosphaerella fijiensis.
Molecular Ecology 5:499-510.
Carlier J., X. Mourichon and D.R. Jones. 2000. Black leaf streak. The causal agent.
Pp. 46-56 in Diseases of Banana, Abacá and Enset. (D.R. Jones, ed.). CABI Publishing,
Wallingford, UK.
Carlier J., M.F. Zapater, F. Lapeyre, D.R, Jones and X. Mourichon. 2000. Septoria leaf spot
of banana: a newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90:884-890.
Crous P.W and X. Mourichon. 2002. Mycosphaerella eumusae and its anamorph
Pseudocercospora eumusae spp. nov., causal agent of Eumusae Leaf Spot Disease of Banana.
Sydowia 54:35-43.
Etienne J.L., A. Binder, H. Steiner and J.F. Rodriguez. 1995. Detection of black Sigatoka disease
in banana leaves using Elisa immuno diagnostics. Pp. 213-218 in Proceedings of the
XI Acorbat meeting. (V. Morales Soto, ed.), CORBANA, San Jose, Costa Rica.
Fouré E. and A. Moreau. 1992. Contribution à l’étude épidémiologique de la cercosporiose
noire dans la zone du Mungo au Cameroun. Fruits 47:3-16.
Fouré E., X. Mourichon and D.R. Jones. 2000. Black leaf streak. Host reaction. Evaluating
germplasm for reaction to black leaf streak. Pp. 62-67 in Diseases of Banana, Abacá and
Enset. (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.
Gauhl F. 1994. Epidemiology and Ecology of Black Sigatoka (Mycosphaerella fijiensis Morelet)
on Plantain and Banana in Costa Rica, Central America. Translation of a PhD thesis
originally in German. INIBAP, Montpellier, France, 120pp.
17
MyLsd 17x24
25/06/03
15:15
Page 18
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Harelimana G., P. Lepoivre, H. Jijakli and X. Mourichon. 1997. Use of Mycosphaerella fijiensis
toxins for the selection of banana cultivars resistant to black leaf streak. Euphytica
96(1):125-128.
Hayden H.L., J. Carlier and E.A.B. Aitken. 2000. The population genetics of Mycosphaerella
musicola in Australia. 2nd International Symposium on Molecular and Cellular Biology
of Banana. Brisbane, Australia, 20 October-3 November 2000.
Hayden H.L., J. Carlier and E.A.B. Aitken. In press. The genetic structure of Mycosphaerella
fijiensis from Australia, Papua New Guinea and the Pacific Islands. Plant Pathology.
Johanson A. and M.J. Jeger. 1993. Use of PCR for detection of Mycosphaerella fijiensis and
M. musicola, the causal agents of Sigatoka leaf spots in banana and Plantains. Mycological
Research 97:670-674.
Johanson A., R.N. Crowhurst, E.H.A. Rikkerink, R.A. Fullerton and M.D. Templeton. 1994.
The use of species-specific DNA probes for the identification of Mycosphaerella fijiensis
and M. musicola, the agents of Sigatoka diseases of banana. Plant Pathology 44:701-707.
Jones D.R. 2000. Diseases of Banana, Abacá and Enset. (D.R. Jones ed.). CABI Publishing,
Wallingford, UK, 544pp.
Mourichon X., P. Lepoivre and J. Carlier. 2000. Black leaf streak. Host-pathogen interactions.
Pp. 67-72 in Diseases of Banana, Abacá and Enset. (D.R. Jones, ed.). CABI Publishing,
Wallingford, UK.
Müller R., C. Pasberg-Gauhl, F. Gauhl, J. Ramser and G. Kahl. 1997. Oligonucleotide
fingerprinting detects genetic variability at different levels in Nigerian Mycosphaerella
fijiensis. Journal of Phytopathology 145:25-30.
Romero R.A. 2000. Black leaf streak. Control. Pp. 72-79 in Diseases of banana, Abacá and
Enset. (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.
Rutter J., P.J.A. Burt and F. Ramirez. 1998. Movement of Mycosphaerella fijiensis spores and
Sigatoka disease development on plantain close to an inoculum source. Aerobiologia
14:201-208.
Smith M.C., J. Rutter, P.J.A. Burt, F. Ramirez and O.E.H. Gonzales. 1997. Black Sigatoka disease
of banana: spatial and temporal variability in disease development. Annals of applied
Biology 131:63-77.
Stierle A.A., R. Upadhyay, J. Hershenhorn, G.A. Strobel and G. Molina. 1991. The phytotoxins
of Mycosphaerella fijiensis, the causative agent of black Sigatoka disease of bananas and
plantains. Experientia 47:853-859.
18
MyLsd 17x24
25/06/03
15:15
Page 19
Session 1
Impact of Mycosphaerella
leaf spot diseases of bananas
MyLsd 17x24
25/06/03
15:15
Page 21
Session 1
R.A. Romero
Introduction
The spread, detection and impact of
black leaf streak disease and other
Mycosphaerella species in the 1990s
R. A. Romero
Abstract
By the end of the 1980s, black leaf streak disease caused by Mycosphaerella fijiensis was present
in all continents where bananas or plantains were grown, although distribution in some regions
was limited to a few countries. In this presentation, the spread of the disease during the 1990s
in several countries and regions, and the important socio-economic consequences are discussed.
From 1990 to 1999, new records of black leaf streak disease were reported from six countries in
Africa, eight in Asia, eight in Latin America and the Caribbean, and one from Australasia/Oceania.
M. fijiensis has also spread within countries to ecological niches that were previously occupied
by M. musicola, the causal agent of Sigatoka disease, thus threatening the survival of this
pathogen. This presentation also discusses the methods, developed in the last decade, to
identify the species of Mycosphaerella that cause leaf spot diseases in banana.The methods were
used to confirm the synonymy of M. fijiensis var. difformis and M. fijiensis, and have provided the
opportunity to study the genetic diversity of pathogen populations among isolates from
different geographical regions.
Resumen - Propagación, detección e impacto de la Sigatoka negra y otras enfermedades
foliares causadas por Mycosphaerella de bananos en la década de los 90
Al finales de la década de los 80, la Sigatoka negra, causada por Mycosphaerella fijiensis, ya se
encontraba presente en todos los continentes donde se cultivan bananos y plátanos, pero su
distribución en algunas regiones estaba limitada a unos pocos países. En esta presentación, se
discutirá brevemente la propagación de la enfermedad durante los 90 abarcando varios países
y regiones, y su impacto socioeconómico. De 1990 a 1999, se reportaron nuevos registros de la
enfermedad desde seis países en Africa, ocho en Asia, ocho en América Latina y el Caribe, y solo
un informe de un país en Australasia/Oceania. M. fijiensis también ha estado progresando en
los países para llegar a los nichos ecológicos que anteriormente estaban ocupados solo por
Chiquita Brands, San José, Costa Rica
21
MyLsd 17x24
25/06/03
15:15
Page 22
Mycosphaerella leaf spot diseases of bananas: present status and outlook
M. musicola, agente causal de la Sigatoka amarilla, amenazando la supervivencia de este
patógeno. La presentación también discutirá el desarrollo de los métodos más precisos para
identificar las diferentes especies de Mycosphaerella que causan las enfermedades de las
manchas foliares en banano en la última década, que han mejorado nuestra habilidad de detectar
el patógeno. Los mismos métodos permitieron confirmar la sinonimia entre M. fijiensis var. difformis
y M. fijiensis, así como el estudio de la diversidad genética de las poblaciones del patógeno entre
los aislados de diferentes regiones geográficas.
Résumé - La propagation, la détection et l’impact de la maladie des raies noires et
d’autres espèces de Mycosphaerella dans les années 1990
A la fin des années 1980, la maladie des raies noires, causée par Mycosphaerella fijiensis, était
présente sur tous les continents où les bananes et les plantains étaient cultivés, bien que dans
certaines régions sa distribution était limitée à quelques pays seulement. Dans cet exposé, nous
présenterons la propagation de la maladie dans plusieurs pays et régions dans les années 1990
ainsi que les prinicipales répercutions socio-économiques. Entre 1990 et 1999, de nouveaux cas
de maladie des raies noires ont été enregistrés dans six pays d’Afrique, huit d’Asie, huit d’Amérique
latine et des Caraïbes et un d’Australasie/Océanie. Dans certains pays, M. fijiensis, s’est également
répandu dans les niches écologiques occupées au préalable par M. musicola, l’agent provoquant
la maladie de Sigatoka, mettant ainsi en danger la survie de ce pathogène. Nous présenterons
également les méthodes développées dans la dernière décennie qui permettent d’identifier les
espèces responsables des maladies foliaires causées par les Mycosphaerella chez la banane. Ces
méthodes ont été utilisées afin de confirmer la synonymie entre M. fijiensis var. difformis et M.
fijiensis, et ont permis l’étude de la diversité génétique des populations de pathogènes parmi des
isolats de différentes régions géographiques.
Introduction
This paper describes the most important events that have characterized the situation
of Mycosphaerella leaf spot diseases in the decade 1990-2000. Emphasis is given to
black leaf streak disease because of its greater importance in comparison with other
Mycosphaerella leaf spot diseases. Black leaf streak disease continued to spread to
new areas and remains a threat to other countries and regions. From the end
of the 1980s to 1999, black leaf streak disease was newly reported in six countries
in Africa, eight in Asia, eight in Latin America and in one location in
Australasia/Oceania. In America, the French West Indies and the Windward Islands,
two regions that produce bananas for export are threatened by black leaf streak
disease.
In the near future, it is possible that black leaf streak disease may attain a
distribution similar to that of Sigatoka disease caused by M. musicola. Details of
the spread and the distribution of Mycosphaerella leaf spot diseases are described
by Pasberg-Gauhl et al. (2000) and as updated by D. Jones elsewhere in this volume.
The continued spread of black leaf streak disease poses two questions 1) are
there preventative measures that could delay or prevent further spread of the disease?
and 2) are those countries that are threatened by the disease taking measures to
prevent the entry of the pathogen?
During the last decade there have been considerable advances in the development
of precise techniques to detect Mycosphaerella spp. pathogens (Carlier et al., 1994;
Johanson and Jeger, 1993a, 1993b). However, there are few examples where these
22
MyLsd 17x24
25/06/03
15:15
Page 23
Session 1
R.A. Romero
molecular methods are being used to prevent or delay the spread of these pathogens
to new areas. Little is known about the integration of these methods with strategies
to prevent the dissemination of the diseases caused by Mycosphaerella spp. A possible
limitation to the use of molecular methods is a lack of adequate infrastructure and
trained personnel in developing countries.
The arrival of black leaf streak disease to a new area has resulted in the replacement of Sigatoka disease as the predominant disease of bananas and plantains. The
ability of M. fijiensis to displace M. musicola has not been adequately studied.
However, it is known that M. fijiensis has several biological characteristics that make
it more competitive than M. musicola, e.g. greater ascospore production, more sexual
cycles a year, and a higher rate of colonization of host tissue (Stover, 1980; Mouliom
Pefoura et al., 1996). Little information is available about differences in pathogenicity
between the two species.
The replacement of M. musicola by M. fijiensis, i.e. the elimination of one species
by another in a short time, is interesting from an ecological and evolutionary point
of view. Unfortunately, there are no studies that show whether M. musicola has in
fact disappeared. It is possible that M. musicola coexists with M. fijiensis at low
frequencies that are difficult to detect. If both pathogens survive in mixed
populations, with M. musicola at a low frequency and M. fijiensis as the predominant
species, it would be preferable to refer to a complex of Mycosphaerella leaf spot
diseases rather than to black leaf streak disease. Breeding strategies for resistance
would need to take this possibility into consideration because resistance to
M. fijiensis does not necessarily imply resistance to M. musicola. Similarly,
resistance genes incorporated by genetic engineering would need to be evaluated
against both pathogens. Current molecular techniques may be able to clarify whether
M. musicola survives at low frequencies with M. fijiensis.
Reports describe how M. musicola is better adapted than M. fijiensis to low
temperatures, and how M. musicola is prevalent at higher altitudes where
temperatures are cooler (Mouliom Pefoura et al., 1996; Romero and Gauhl, 1988;
Tapia, 1993). However, over the years, M. fijiensis has been replacing M. musicola
at higher altitudes in Costa Rica (Gauhl et al., 2000) suggesting that M. fijiensis may
be adapting to the environment. Plantains and other cooking bananas commonly
grown at higher altitudes, sometimes in combination with coffee, can be severely
affected due to the greater aggressiveness of black leaf streak disease in comparison
with Sigatoka disease.
The socio-economic impact of black leaf streak disease has continued to increase
as the pathogen reaches new areas. The impact has also increased as the disease
becomes more difficult to control because of increased resistance of the pathogen
to new fungicides. Thus, fungicide resistance is increasingly an important constraint
to control black leaf streak disease in plantations dedicated to the export market.
Recently M. eumusae, a new species pathogenic to bananas, has been described
(Anon., 1995; Carlier et al., 2000). Little is known about the biology and epidemiology
of this species, and hence it is not possible to evaluate whether or not it represents
a threat to small-scale and large–scale banana production. Research is urgently
needed to characterize the pathogenicity and aggressiveness of M. eumusa. In
particular, studies are needed on its competitive ability with respect to M. musicola
23
MyLsd 17x24
25/06/03
15:15
Page 24
Mycosphaerella leaf spot diseases of bananas: present status and outlook
and M. fijiensis to determine the potential of M. eumusae as a pathogen of bananas
and other Musa genotypes.
During the course of this workshop, we will have the opportunity to examine
how these events have taken place or are currently underway. We may be able to
learn from the difficulties and progress on technical assistance and research and hence
increase our ability to improve the control of these diseases in the future.
References
Anonymous. 1995. Musanews. INFOMUSA 4(2):26-30.
Carlier J., X. Mourichon, D. Gonzales de León, M.F. Zapater and M.H. Lebrun. 1994. DNA
restriction fragment length polymorphisms in Mycosphaerella species causing banana leaf
spot diseases. Phytopathology 84:751-756.
Carlier J., X. Mourichon and D.R. Jones. 2000. Fungal diseases of the foliage. Sigatoka like
leaf spots, Septoria leaf spot. Pp. 93-96 in Diseases of Bananas, Abacá and Ensete (D.R.
Jones, ed.). CAB International, Wallingford, UK.
Gauhl F., C. Pasberg-Gauhl and D.R. Jones. 2000. Fungal diseases of the foliage. Sigatoka
leaf spots. Black leaf streak, disease cycle and epidemiology. Pp. 56-62 in Diseases of
Bananas, Abacá and Ensete (D.R. Jones, ed.). CAB International, Wallingford, UK.
Johanson A. and M.J. Jeger. 1993a. Use of PCR for detection of Mycosphaerella fijiensis and
M. musicola, the causal agents of Sigatoka leaf spots in banana and plantain. Mycological
Research 97:670-674.
Johanson A. and M.J. Jeger. 1993b. Detection of Mycosphaerella fijiensis and M. musicola
in banana leaf tissue using the polymerase chain reaction. Pp. 227-236 in Breeding Banana
and Plantain for Resistance to Diseases and Pests. Proceedings of an International
symposium on Genetic improvement of bananas for resistance to diseases and pests. CIRAD
and INIBAP, France.
Mouliom Pefoura A., A. Lassoudière, J. Foko and D.A Fontem. 1996. Comparison of
development of Mycosphaerella fijiensis and Mycosphaerella musicola on banana and
plantain in the various ecological zones in Cameroon. Plant Disease 80:950-953.
Pasberg-Gauhl C., F. Gauhl and D.R. Jones. 2000. Fungal diseases of the foliage. Sigatoka
leaf spots. Black leaf streak, distribution and economic importance. Pp. 37-44 in Diseases
of Bananas, Abacá and Ensete (D.R. Jones, ed.). CAB International, Wallingford, UK.
Romero R.A. and F. Gauhl. 1988. Determinación de la severidad de la Sigatoka negra
(Mycosphaerella fijiensis var. difformis) en bananos a diferentes altitudes sobre el nivel
del mar. Revista de la Asociación Bananera Nacional (ASBANA), San José, Costa Rica
12(29):7-10.
Stover R. H. 1980. Sigatoka leaf spots of bananas and plantains. Plant Disease 64:750-755.
Tapia A. 1993. Distribución altitudinal de la Sigatoka amarilla (Mycosphaerella musicola) y
la Sigatoka negra (Mycosphaerella fijiensis) en Costa Rica. Tesis. Universidad de Costa
Rica, 76pp.
24
MyLsd 17x24
25/06/03
15:15
Page 25
Session 1
D.R. Jones
The distribution and importance of
the Mycosphaerella leaf spot diseases
of banana
D. R. Jones
Abstract
The three main fungal leaf spot pathogens of banana are Mycosphaerella musicola, M. fijiensis and
M. eumusae. All result in serious economic damage to cultivars in the Cavendish subgroup grown
for local consumption and export. M. musicola was the first foliar pathogen to be identified as a
major problem. Its spread from the Southeast Asian/Pacific region, where it was established by the
1920s, to North and South America in the 1930s, resulted in widespread disruption to the export
trade. Since the 1960s, M. musicola has been steadily and largely replaced worldwide by M. fijiensis.
This pathogen is more damaging to Cavendish cultivars than M. musicola and attacks a wider range
of banana clones. It was first recognised in the Pacific region and has since become a major problem
in most banana-growing areas. The susceptibility of plantain and other subsistence banana types
is of special concern in Africa. By the late 1980s, reports in the scientific literature of records of
M. fijiensis and M. musicola in Asia led to some confusion as to the true distribution of the two
pathogens in this region. Many specimens collected in the region were identified as M. eumusae,
a new leaf spot pathogen of banana.This fungus was observed damaging Cavendish cultivars and
plantain. Most records for M. eumusae have been from Asia but the pathogen has also been found
on islands in the Indian Ocean and in West Africa. Because of the similarity of symptoms caused
by M. musicola, M. fijiensis and M. eumusae, it is proposed that the diseases caused by these three
fungi be known collectively as Mycosphaerella leaf spot diseases.
Resumen - Distribución e importancia de las enfermedades de las manchas foliares
causadas por Mycosphaerella en banano
Los tres principales patógenos fungosos de las manchas foliares del banano son Mycosphaerella
musicola, M. fijiensis y M. eumusae. Todos ellos producen serios daños económicos a los cultivares
en el subgrupo Cavendish cultivados para el consumo local y para la exportación. M. musicola fue
el primer patógeno foliar identificado como el principal problema en la región productora de bananos
de América Latina y el Caribe. Se propagó de la región de Sudeste asiático y el Pacífico, donde se
había establecido allá por la década de los 20, al Nuevo Mundo en la década de los 30, resultando
en una quiebra del comercio de exportación. Desde los años 60, M. musicola fue reemplazada firme
y extensamente en todo el mundo por M. fijiensis. Este patógeno es más dañino para los cultivares
Consultant in International Agriculture,Worcestershire, United Kingdom
25
MyLsd 17x24
25/06/03
15:15
Page 26
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Cavendish que M. musicola y ataca un rango más amplio de cultivares. Por primera vez esta
enfermedad fue reconocida en el Pacífico. Actualmente, M. fijiensis es uno de los principales
problemas en la mayoría de las áreas productoras de banano y ha reemplazado a M. musicola como
patógeno dominante de la mancha foliar. La susceptibilidad del plátano y otros tipos de bananos
de subsistencia representa la principal preocupación en Africa. A finales de la década de los 80, los
informes encontrados en la literatura científica de los registros de M. fijiensis y M. musicola en el
Sudeste de Asia llevaron a algunas confusiones en cuanto a la verdadera distribución de los dos
patógenos en esta región. Muchos especímenes recolectados en la región fueron identificados como
M. eumusae, un nuevo patógeno de la mancha foliar del banano. Este hongo se observó como dañino
para los cultivares Cavendish y plátano. Aunque hasta la fecha la mayoría de los registros sobre la
M. eumusae provienen de las regiones de Sudeste y Sur de Asia, el patógeno también fue encontrado
en las islas del Océano Indico y en Africa Occidental. Debido a la similitud de los síntomas causados
por M. musicola, M. fijiensis y M. eumusae, se sugiere que las enfermedades causadas por estos tres
hongos se denominen manchas foliares causadas por Mycosphaerella.
Résumé - La distribution et l’importance des maladies foliaires causées par les
Mycosphaerella
Les trois principaux pathogènes responsables des maladies foliaires de la banane sont
Mycosphaerella musicola, M. fijiensis et M. eumusae. Tous trois provoquent de sérieux dégâts
économiques aux cultivars du sous-groupe Cavendish, cultivés pour la consommation locale et
l’exportation. M. musicola a été le premier pathogène de Mycosphaerella a être identifié
provoquant des problèmes majeurs. Sa propagation de l’Asie du Sud-Est/région du Pacifique, où
il était établi dans les années 1920, à l’Amérique du Nord et du Sud dans les années 1930, a
sérieusement affecté le commerce international. Depuis les années 1960, M. musicola a été
graduellement et largement remplacé à travers le monde par M. fijiensis. Ce pathogène cause
plus de dommages aux Cavendish que M. musicola et attaque un spectre plus large de cultivars
de bananiers. Il a d’abord été identifié dans la région du Pacifique avant de devenir un problème
majeur dans la plupart des zones cultivées de bananes. La sensibilité du plantain et d’autres types
de bananes servant d’aliment de base est particulièrement préoccupante en ce qui a trait à
l’Afrique. A la fin des années 1980, les études rapportant la présence de M. fijiensis et de M. musicola
en Asie, ont prêté à confusion quant à la véritable distribution des deux pathogènes dans cette
région. De nombreux échantillons récoltés dans la région ont été identifiés comme étant
M. eumusae, un nouveau pathogène responsable d’une maladie foliaire des bananiers. Il a été
observé que ce champignon attaquait les Cavendish et le plantain. C’est en Asie que la majorité
des observations de M. eumusae ont été faites mais ce pathogène a également été trouvé dans
des îles de l’Océan Indien et en Afrique de l’Ouest. Du fait des symptômes similaires causés par
M. musicola, M. fijiensis et M. eumusae, nous proposons que les pathologies causées par ces trois
champignons soient connues collectivement en tant que maladies foliaires causées par les
Mycosphaerella.
Introduction
The most serious leaf spot diseases of banana are caused by three species of
Mycosphaerella. All three were discovered and became important constraints to
commercial production in the 20th century. One has long since reached the limits
of its distribution; a second is now approaching the limits of its distribution and
a third is possibly only now beginning its global spread. The evidence suggests
that all three may have arisen in the Southeast Asian/Australasian region, which
is the centre of origin of Musa species and also the centre of evolution of cultivated
banana (Jones, 2000). Some accessions of M. acuminata ssp. banksii, which is a
wild diploid banana that has contributed genetic components to most edible banana
26
MyLsd 17x24
25/06/03
15:15
Page 27
Session 1
D.R. Jones
clones (Carreel, 1995), are known to be susceptible to at least two of the
Mycosphaerella species causing leaf spot (Carlier et al., 2000a) and so coevolution is a strong possibility.
It is not clear why the Mycosphaerella leaf spot pathogens were not disseminated
during the first movements of banana planting material out of the Southeast
Asian/Australasian region, as they could have been carried as infections of scale
leaf tissue associated with sword suckers (Stover, 1978). Possibly, the fungi were
not widespread pathogens on subspecies of M. acuminata or cultivated banana.
Alternatively, the fungi, which initially were saprophytes surviving by colonizing
dead leaf tissue, may have evolved to become pathogenic quite recently. Stover
(1978) speculated that M. fijiensis might have evolved on a susceptible wild diploid
and then spread to edible cultivars. The histories of the three Mycosphaerella fungi
that continue to cause so much damage to cultivated banana are contrasted and
compared in chronological order of discovery in the following text.
Mycosphaerella musicola, the cause of Sigatoka disease
First records
Mycosphaerella musicola was the first leaf pathogen to become a serious problem
on commercial banana plantations. The fungus, originally named Cercospora musae
from its imperfect stage, was first described as a pathogen of banana on the island
of Java in Indonesia at the beginning of the 20th century (Zimmermann, 1902).
However, it wasn’t until 10 years later in Fiji that it became prominent as the cause
of an important disease. The pathogen was first found in Fiji in the Sigatoka
(pronounced ‘Singatoka’) Valley on the main island Viti Levu and quickly became
an important constraint to production (Philpott and Knowles, 1913; Massee, 1914).
The disease became known as Sigatoka disease, and more recently as yellow Sigatoka,
to distinguish it from the disease caused by M. fijiensis, which is widely known as
black leaf streak disease or black Sigatoka.
Global spread
The subsequent spread of M. musicola, as it was called after the perfect stage was
discovered (Leach, 1941), around the world to practically all banana-growing regions
is well documented (Meredith, 1970) (Table 1). The chronological sequence of disease
occurrences led to speculation that ascopores borne on high altitude wind currents
may have been responsible for the intercontinental dissemination of the pathogen
from Australia to Africa and Central America (Stover, 1962). However, the latest
information on survivability and movement of windblown ascospores, which are the
spores implicated in long distance spread, indicate that this hypothesis is unlikely
(Parnell et al., 1998). Stover (1980) changed his opinion on the distances ascospores
of Mycosphaerella pathogens could spread by stating that wind dispersal of
M. fijiensis from small areas of infection to new areas was probably slight when
distances exceeded 50 km. Natural ultraviolet radiation, which would be a factor
limiting high-altitude dispersal on clear days, kills ascospores of M. fijiensis within
27
MyLsd 17x24
25/06/03
15:15
Page 28
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 1. Year Sigatoka disease was first reported in a given country until 1970 (modified from information published
in Table 2 of Meredith,1970).
Region/Country
Australasia/Oceania
Fiji
Australia (Queensland)
Australia (New South Wales)
Solomon Islands
Papua New Guinea
New Caledonia
Indonesia (Irian Jaya)
Norfolk Island
Wallis Island
USA (Hawaii)
Samoa
American Samoa
French Polynesia
Tonga
Asia
Indonesia (Java)
Sri Lanka
Philippines
Malaysia (Peninsula)
China
India (Assam)
Taiwan
Malaysia (Sabah and Sarawak)
Cambodia
Thailand
Hong Kong
Vietnam
Laos
Brunei
Latin America/Caribbean
Guadeloupe
Surinam
Trinidad
Guyana
Honduras
Jamaica
Belize
Grenada
Martinique
Mexico
Colombia
Costa Rica
Dominican Republic
Guatemala
Haiti
Year
1912
1924
1927
1946
1951
1951
1953
1954
1954
1958
1961
1961
1962
1965
1902
1919
1921
1933
1936
1946
1953
1959
1960
1962
1966
1966
1967
1968
1932 ?
1933
1933
1935
1935
<1936
1936
1936
1936
1936
1937
1937
1937
1937
1937
Region/Country
Nicaragua
Panama
St. Vincent
Venezuela
Cuba
Dominica
Puerto Rico
St. Lucia
Brazil
El Salvador
Peru
Ecuador
Montserrat
USA (Florida)
Bolivia
Africa
Uganda
Tanzania
(mainland and possibly Zanzibar)
Kenya
Cameroon
Congo
Mozambique
Guinea
Nigeria
South Africa
Zimbabwe
Ghana
Malawi
Sierra Leone
Madagascar
Côte d’Ivoire
Nigeria
Mauritius
Angola
Somali Republic
Zambia
Year
1937
1937
1937
1937-41
1938
1938
1938-39
1938
1944
1944
1946
1952
1955
1955
1968
1938
1939
?
1941
1948
1948
1952
?
1954
1954
1955
1955
1956
1957
1959
?
1960
1962
1962
1966
?: Pathogen recorded as present, but uncertainty over year of first
detection.
Note: Meredith 1970 indicates that there is a possibility that some
reports of Sigatoka disease in the Pacific area (such as in Hawaii in
1958) could have been black leaf streak disease. Sigatoka disease has
still not been reported in the Canary Islands, Egypt and Israel bananagrowing areas. The dry summer climate in these countries may not be
conducive for the establishment of the disease.
6 hours (Parnell et al., 1998). Ascospores may not spread more than a one or two
hundred kilometres and then only in strong winds and heavy cloud, or at night. In
summary, many pathologists now believe that airborne ascospores of Mycosphaerella
leaf spot pathogens spread disease between growing area within countries and
between countries, but that airborne spread between continents is unlikely.
28
MyLsd 17x24
25/06/03
15:15
Page 29
Session 1
D.R. Jones
Movement between continents and between isolated countries, such as those found
in the Pacific, is probably the result of the transfer of diseased material by humans.
The appearance of the disease in Australia in the mid-1920s may have been due
to the same movements of planting material from the South Pacific that are believed
to have introduced bunchy top disease (Magee, 1953). Another possibility is that
diseased banana leaves may have been used as packing material for goods shipped
into Australia from the South Pacific.
Sigatoka disease may have been introduced to East Africa in the late 1930s by
the movement of planting material/diseased leaves from Asia. During the colonial
period, there was much migration to East Africa from India and settlers and/or British
administrators undoubtedly transported infected suckers of their favourite cultivars
from one location to another.
The appearance of Sigatoka disease in European colonies in the eastern Caribbean
region in the mid-1930s could also have been due to colonial-inspired movements
of Musa germplasm from Asia. The almost simultaneous discovery of the disease at
several locations may have been due to multiple introductions of the pathogen or
to rapid spread from one point after an undetected build-up of inoculum. Prevailing
landward winds off the Caribbean Sea would have carried spores of M. musicola
across the north of South America and into Central America.
Today, Sigatoka disease is regarded as having a worldwide distribution, although
it has not been recorded in the Canary Islands, Egypt or Israel (Meredith, 1970). The
dry summer climates of these countries may make the local environment unsuitable
for disease establishment.
Impact
The most efficient leaves for photosynthesis on a vegetatively growing banana are
the second to fifth counting down from the top of the plant. Therefore, if optimal
assimilation potential of the plant is to be maintained, it is important that leaves
2-5 are free of excessive shade, severe leaf tearing and disease. In a vigorous plant
growing in the tropics, this critical leaf area is renewed monthly (Robinson, 1996)
and the pathogen does not cause enough damage to have an appreciable effect on
growth (Leach, 1946). Damage comes after bunch emergence when leaf production
ceases and leaf tissue cannot be renewed. The greater the damage on remaining leaves
and the earlier it occurs after shooting, the greater the effects on yield. Sigatoka
disease also affects the physiology of developing fruits causing premature ripening
(Meredith, 1970). This occurs in the field if the plant is severely diseased or in transit
to markets if moderately affected. For these reasons it was important to control
Sigatoka disease in commercial plantations.
The economic impact of Sigatoka disease has been twofold. First, there was the
direct effect on production when the disease first became established and control
measures were being developed. Between 1937 and 1941, production in Mexico was
halved as a direct result of Sigatoka disease. In Honduras, production declined to
less than one third of the pre-disease level (Meredith, 1970). These were enormous
losses. Secondly, there were the on-going costs associated with leaf spot control once
effective measures were developed and adopted. The struggle against Sigatoka disease
29
MyLsd 17x24
25/06/03
15:15
Page 30
Mycosphaerella leaf spot diseases of bananas: present status and outlook
in this regard has been well documented (Meredith, 1970; Stover 1972, 1990; Carlier
et al., 2000a). Bordeaux mixture, the first effective fungicide, had to be applied as
a high volume spray and large pipeline systems were installed at great expense to
deliver the chemical in plantations. Later, petroleum oil applied as a low-volume
spray from aircraft proved effective and costs were reduced, especially when combined
with forecasting systems. Protectant fungicides, such as dithiocarbomates, and
systemic fungicides, such as benomyl, improved the standard of control. However,
control measures, which included labour intensive cultural practices, such as
pruning old diseased leaves, was an expense that was borne by growers. In 1990 in
Queensland, where Sigatoka disease is still the dominant leaf spot disease, control
measures were estimated as 14% of total production costs.
There are no real figures on the impact Sigatoka disease had on small-scale
growers in developing countries producing fruit for families or local markets.
Plantains and other cooking bananas favoured by many subsistence farmers are
resistant to Sigatoka disease in lowland areas, and therefore the impact in these cases
would be minimal. Those growing the more susceptible dessert banana cultivars, such
as in the Pome subgroup (AAB), would either have had to change to cooking types,
accepted a loss in yield, or invested in backpack sprayers. However, the worst was
yet to come.
Mycosphaerella fijiensis, the cause of black leaf streak
disease
First records
The first report of M. fijiensis causing damage was in the same Sigatoka valley on
the island of Vitu Levu in Fiji where M. musicola was first recognised as a major
pathogen of banana fifty years earlier. In February 1963, the disease caused by
M. fijiensis was reported to be spreading rapidly in the Sigatoka Valley (Rhodes,
1964) and was predicted to affect the whole island by the end of 1964 (Leach, 1964a).
The causal organism was also described for the first time from material collected in
Fiji (Leach, 1964b). The disease caused by M. fijiensis was called black leaf streak
disease by Rhodes (1964). Leach (1964b) described the risk of spread of this new
disease of banana as “a grave threat” and feared that the abundance of airborne
ascospores produced by the pathogen may lead to dissemination around the world
faster than for Sigatoka disease. He also warned that an outbreak of a new, perhaps
worse, disease might happen again in the future in another location.
The problem caused by M. fijiensis in Fiji became apparent when the mist-sprays
of light mineral oil being used to control M. musicola lost their effectiveness. The
recognition of yet another important banana pathogen in Fiji before anywhere else
can probably be attributed to the fact that at this location there were sizeable
plantations of susceptible dessert banana cultivars and an efficient plant protection
service experienced in banana problems.
Surveys undertaken after black leaf streak disease was discovered in Fiji led to
the conclusion that the pathogen had most likely been present in the Pacific and
parts of the Pacific rim for many years (Meredith, 1970; Stover, 1976; Stover, 1978;
30
MyLsd 17x24
25/06/03
15:15
Page 31
Session 1
D.R. Jones
Long, 1979). It was suggested that M. fijiensis may have been in the Hawaiian Islands
in 1958 (Meredith and Lawrence, 1969). An analysis of herbarium specimens by
Stover (1976), showed that M. fijiensis was present in Papua New Guinea by at
least 1957 and in Taiwan as early as 1927. The similarity of symptoms with those
of Sigatoka disease most likely masked the arrival of this new disease in many
countries. Because of this, the year that black leaf streak disease was first
discovered in many countries in this region (Table 2) does not reflect the order of
spread of the pathogen.
Global spread
When the black leaf streak pathogen was first found in Honduras in 1972, it was
thought from its morphology to be a variant of M. fijiensis and was named
M. fijiensis var. difformis (Mulder and Stover, 1976). However, it was later shown
that M. fijiensis and M. fijiensis var. difformis were synonymous (Pons, 1987). Black
leaf streak disease has precedence as the common name for the disease caused by
M. fijiensis and was adopted by Carlier et al. (2000a) in the publication ‘Diseases
of Banana, Abacá and Enset’, however it is widely known in as black Sigatoka or
Sigatoka negra in Spanish. The choice of which name to use in publications is one
of personal preference.
A measure of the rate of spread of M. fijiensis between countries can be
gained from an examination of the records from the Latin American/Caribbean
region (Table 2). Within three years of its detection in Honduras in 1972, M. fijiensis
was reported in Belize to the north and by 1977 had arrived in Guatemala to the
west. Local spread was quicker in the direction of prevailing winds from the east
and northeast (Stover, 1980). In 1979, it appeared in El Salvador, Nicaragua and
Costa Rica and by 1981 had spread north to Mexico and south to Panama and
northern Colombia (Carlier et al., 2000a). Spread was believed to have been
accelerated in Central America by the movement of diseased banana leaves and
leaf trash across international boundaries with road-transported banana and
plantain fruit (Stover, 1980). By 1986, commercial plantations in northern Ecuador
were affected and plantains in western Venezuela succumbed in 1991. Spread to
northern Peru occurred in 1994 and to Bolivia in 1996. The first report from western
Brazil came in 1996 and since then M. fijiensis has been advancing in a
southwesterly direction towards the Brazilian coast. In 2001, movements of banana
fruit and associated banana leaves from inland areas where M. fijiensis was found
to coastal cities were being controlled in an effort to delay spread (R. S. Moreira,
Brazil, personal communication).
In the Caribbean, black leaf streak disease was first found in Cuba in 1990,
Jamaica in 1995 and the Dominican Republic in 1996. The first authenticated report
from Haiti, which has a dry climate, was in 1999. Natural spread to other
Caribbean countries is inevitable, but may be slowed considerably by the prevailing
winds, which blow from the east. Recent investigations involving the analysis of
isolates have revealed that the source of inoculum for the outbreak in Jamaica may
have come form Central America and not windblown from Cuba as originally
suspected (G. Rivas, Costa Rica, personal communication). The outbreak in the
31
MyLsd 17x24
25/06/03
15:15
Page 32
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 2. Countries where black leaf streak disease has been detected (modified from Carlier et al., 1999 with information
supplied by X. Mourichon and D. R. Jones).
Region/Country
Australasia/Oceania
Solomon Islands
Papua New Guinea
Fiji
French Polynesia
Micronesia
New Caledonia
Vanuatu
Tonga
Samoa
USA (Hawaii)
American Samoa
Cook Islands
Niue
Norfolk Island
Australia (Torres Strait/Cape York)
Wallis and Fortuna Islands
Australia (North Queensland,
currently under eradication)
Asia
Taiwan
Philippines (Luzon)
Singapore
Philippines (Mindanao)
Malaysia (Peninsula)
Thailand
Indonesia (Java)
Indonesia (Halmahera)
China (Hainan)
Bhutan
China (Guangdong)
China (Yunnan)
Vietnam
Indonesia (Sumatra)
Indonesia (Kalimantan)
East Malaysia (Sarawak)
Latin America/Caribbean
Honduras
Belize
Guatemala
Nicaragua
Costa Rica
El Salvador
Mexico
Panama
Year1
1957 (1946)
1957 (1951)
1963
1964-67
1964-67
1964-67
1964-67
1965
1965
1969 (1958)
1975
1976
1976
1980
1981
1996
2001
1927
1964
1964-67
1965
1965
1969
1969
1970
1980
1985
1990
1993
1993
1993
1996
1996
1972 (1969)
1975
1977
1979
1979
1979 2
1980
1981
Region/Country
Colombia
Ecuador
Cuba
Venezuela
Peru
Jamaica
Bolivia3
Dominican Republic
Brazil
USA (Florida)
Haiti
Africa
Zambia
Gabon
Cameroon (south-east)
Cameroon (south-west)
São Tomé
Côte d’Ivoire
Congo
Nigeria
Ghana
Rwanda
Burundi
Tanzania (inc. Pemba and Zanzibar)
Democratic Republic of Congo
(highlands)
Democratic Republic of Congo
(lowlands)
Togo
Kenya
Malawi
Uganda
Benin
Comoros
Mayotte
Central African Republic
Madagascar
Year1
1981
1986
1990
1991
1994
1995
1996
1996
1998
1998
1999
1973 4
1978
1980
1983
1983
1985
1985
1986
1986
1986
1987
1987
1987
1988
1988
1988
1990
1990
1993
1993
1993
1996
2000
1
Most likely year of introduction, but no definite record published at
this time.
2 Year disease was first reported or year present from herbarium
specimens (earliest year disease was believed present in hindsight
within brackets). 3 In 2001, black leaf streak disease was spreading
south-eastwards across Brazil from the Mato Grosso towards the
Atlantic coast. 4 Authenticity of this record has been challenged.
Dominican Republic has also been linked to Central America, though the evidence
is more circumstantial. In both examples, the disease appeared shortly after
banana fruit was shipped to the islands. Could inoculum on the surface of fruit or
in associated leaf trash have introduced black leaf streak to these countries?
32
MyLsd 17x24
25/06/03
15:15
Page 33
Session 1
D.R. Jones
The first record of black leaf streak disease in Africa was from Zambia in 1973
(Raemaekers, 1975). The publication of this outbreak is convincing but the identity
of the pathogen could not be confirmed from specimens sent to the UK, therefore
doubt remains as to the authenticity of the report (Dabek and Waller, 1990). The
next record was from Gabon in 1978. Frossard (1980) believed it might have been
introduced on planting material from Asia. The disease then spread steadily through
Central and West Africa reaching Côte d’Ivoire, Nigeria and Ghana in 1985-1986,
and Uganda and Malawi in 1990 (Table 2). A second, separate introduction of
M. fijiensis into Africa is thought to have occurred in 1987 on the island of Pemba.
This outbreak is believed to have led to the pathogen spreading to the island of
Zanzibar and coastal areas of Tanzania and Kenya (Carlier et al., 2000a). In 2000,
M. fijiensis was recorded in Madagascar for the first time.
The Australian experience
Stover (1978) believed that M. fijiensis may have originated in the Papua New
Guinea-Solomon Islands area and disseminated around the South Pacific with
banana leaves or planting material. This possibility is suggested by the discovery
that isolates of M. fijiensis are more diverse in the Papua New Guinea /Philippines
region than elsewhere, an indication that the area may be the centre of origin of
the pathogen (Carlier et al., 2000a). Therefore, it is likely that M. fijiensis may have
been present on banana on islands in the Torres Strait and on the tip of Cape York
Peninsula, Australia long before its discovery on the first plant pathological survey
of the area in 1981 (Jones and Alcorn, 1982). The pathogen may not have spread
further south in Australia because of the barrier presented by the Cape York
Peninsula, which is a large, remote area of native bush with comparatively few
communities and banana plants. After 1981, better land and air communications,
which encouraged more tourists and people seeking an alternative lifestyle, led to
a higher risk of spread. During the 1990s, M. fijiensis was regularly eradicated from
isolated outbreaks on small plantings within the Peninsula. In all cases, the origin
of the inoculum could not be positively determined. In 2000, an outbreak occurred
on a commercial banana planting at Daintree on the northern fringe of the more
heavily populated coastal strip centred on Cairns. The grower was compensated for
the destruction of his crop by the Australian banana industry. Although eradicated,
the close proximity of this outbreak to the main banana growing area was worrying.
Towards the end of the wet season in April 2001, M. fijiensis was detected on
unmanaged (feral) banana plants and also on cultivated plants in an adjacent farm
in the Tully Valley, which is in the heart of the commercial banana-growing area
in North Queensland centred south of Cairns. Subsequently, the pathogen was
reported from other locations in the same area. An eradication campaign was
immediately mounted. This campaign gathered momentum when the governments
of banana-growing states and the Commonwealth Government pledged funds.
Measures included: (1) establishment of a special banana quarantine area, (2) a ban
on the movement of fruit from this area to other banana-growing areas in Australia,
(3) close monitoring of crops and the diagnosis of any leaf spots detected,
(4) destruction of fields where affected plants were found, (5) drastic pruning of all
33
MyLsd 17x24
25/06/03
15:15
Page 34
Mycosphaerella leaf spot diseases of bananas: present status and outlook
banana plants in the growing area, (6) regular application of systemic fungicides
and (7) zero tolerance for leaf spot disease. The campaign was conducted during
the 2001 dry season, which also markedly reduced the chances of spore release and
infection, with the co-operation of most growers.
A total of 25 plants have been found infected with M. fijiensis in the Tully area.
The last seven plants were either growing in private gardens or were unmanaged.
At the time of the International Sigatoka Workshop on 20-23 May 2002, M. fijiensis
had not been detected for over five and a half months. There are hopes for the
successful eradication of M. fijiensis, which will be the first time that this will have
been achieved anywhere in the world.
Impact
Black leaf streak disease is the major constraint to cultivation in commercial
plantations producing dessert banana fruit for export. It is also a limiting factor
for small-scale and subsistence farmers growing plantain. The disease has had a
much greater impact than Sigatoka disease because the life cycle and epidemiology
of the causal pathogen makes it more difficult to control and it attacks a wider
range of banana clones. M. fijiensis attacks younger leaves on more susceptible
banana clones than those affected by M. musicola. On dessert clones in the
Cavendish subgroup, which are extremely susceptible, the pathogen can kill much
more leaf tissue in the critical 2-5-leaf range than M. musicola. After flowering,
remaining leaves are rapidly killed resulting in premature ripening and large
reductions in yield.
The arrival of black leaf streak disease in Latin America coincided with the
introduction of systemic fungicides to control Sigatoka disease on plantations.
The fungicides were also effective against black leaf streak disease, but many more
applications were needed to maintain control with corresponding increases in
production costs. It also became much more important to prevent the disease
building-up too much as loss of control could have serious consequences. Good
disease control management became essential and those that couldn’t were in
serious difficulties.
Initial effects on commercial banana production soon after the arrival of the
disease were devastating. In the South Pacific, only 49% of unsprayed fruit reached
export quality (Firman, 1972). Fiji ceased exporting banana fruit in 1974 and
Samoa in 1984. Exports also dropped in Tonga and the Cook Islands because
export quality could not be achieved (Fullerton, 1987). In Central America, export
banana cultivation has survived, but at a price. Costs of control in Costa Rica
are now US$900-1500 hectare/year. The overuse of fungicides with the same mode
of action, and the use of fungicides below recommended doses has led to increased
resistance to fungicides in M. fijiensis populations. Different strategies such as
alternation of fungicides with different modes of action and the use of fungicide
mixtures had to be adopted. Monitoring the resistances of local populations of
M. fijiensis to a range of fungicides is now routine. In Costa Rica, the rising costs
of labour and leaf spot control are making commercial banana cultivation increasingly uneconomical. Other dessert bananas grown for local consumption, such
34
MyLsd 17x24
25/06/03
15:15
Page 35
Session 1
D.R. Jones
as clones in the Pome subgroup (AAB) and ‘Silk’ (AAB), which are popular in
Brazil and India, are also susceptible to black leaf streak disease.
Plantain cultivation has also been seriously affected by M. fijiensis. The pathogen
has caused a considerable decrease in the availability of fruit for local consumption
with a corresponding substantial increase in market price. In many areas of Central
America, growers have either gone out of business or have formed cooperatives to
share resources to cover the costs of spray equipment and chemicals. In Panama,
plantain production was estimated to have decreased by 69% and prices to have
increased by 50% between 1979 and 1984 (Bureau, 1990). In Costa Rica, black leaf
streak disease was calculated to have reduced production by 40% by 1982 (Romero,
1986). Similar effects occurred in the plantain industry in Colombia, South America.
After the introduction of black leaf streak disease, plantains became scarce and
expensive and consumers changed to cheaper foods (Belalcazar, 1991). Black leaf streak
disease also affects plantain production in Africa and endangers food security for
many poor people. On poor sandy soils in West Africa, it has been estimated that
yield losses are 33% and 76% during the plant and ratoon cropping cycle respectively
(Mobambo et al., 1996). This has led to small-scale farmers abandoning plantain
cultivation.
Plantain is not the only cooking banana type grown by subsistence farmers and
other small-scale growers to be affected by black leaf streak. In the Great Lakes area
of Africa, the East African highland cultivars in the Lujugira-Mutika subgroup (AAA)
are susceptible. Losses of 37% due to the combined affects of black leaf streak disease
and Cladosporium speckle have been reported (Tushemereirwe, 1996). In the Pacific,
cultivars in the popular Maia Maoli-Popoulu subgroup (AAB) are susceptible (Carlier
et al., 2000a).
Interactions between Mycosphaerella fijiensis
and Mycosphaerella musicola
Soon after M. fijiensis was discovered on Fiji, it was reported to be displacing
M. musicola as the dominant leaf spot. Displacement occurred in many coastal areas
in the tropics particularly in Latin America. At elevation, M. musicola has an
advantage being suited to cooler conditions and there are a number of reports (include
references here) of the two pathogens co-existing at heights of around 1200m to
1500m with M. musicola dominating at higher altitudes and M. fijiensis at lower
ones. There is also evidence that M. fijiensis may be slowly adapting to higher
elevations (Carlier et al., 2000a).
M. musicola has probably not disappeared completely from banana in coastal
areas dominated by M. fijiensis. Evidence of small amounts of survival within the
leaf spot population comes from Nigeria where ‘SH-3362’ (AA), a hybrid resistant
to M. fijiensis and susceptible to M. musicola, was found to be affected by
M. musicola at planting (C. Pasberg-Gauhl and F. Gauhl, Nigeria, personal
communication). In the Philippines, M. musicola has also still been reported to be
present despite the dominance of M. fijiensis in commercial plantations. This may
be because the great genetic diversity of cultivated banana in this country has meant
some clones, like ‘Amas’ (AA, syn. ‘Sucrier’), which is more susceptible to M. musicola
35
MyLsd 17x24
25/06/03
15:15
Page 36
Mycosphaerella leaf spot diseases of bananas: present status and outlook
than M. fijiensis, have maintained the former pathogen. In other areas of Southeast
Asia, the situation has been more difficult to understand. This is discussed more fully
under “Mycosphaerella eumusae”.
Mycosphaerella eumusae, the cause of eumusae
leaf spot
Confusion between leaf spots in Asia
The latest of the three Mycosphaerella leaf spot pathogens to affect banana was only
recognised in the mid-1990s (Carlier et al., 2000a). This was because of uncertainties
concerning the distribution of Sigatoka disease and black leaf streak disease in
Southeast and South Asia. The uncertainty arose because the usual rapid displacement
of M. musicola by M. fijiensis in tropical lowland areas, as had occurred in the
Americas and West Africa, did not seem to have happened in Java (Indonesia), West
Malaysia and Thailand (Jones, 1990). Although M. fijiensis had first been recorded
at these locations in the mid to late 1960s, M. musicola was still present near Bogor
in Java and Kuala Lumpur in West Malaysia in 1976 (Stover, 1976). From
observations ten years later in 1988, the author believed Sigatoka disease was still
the dominant leaf spot occurring in Java, West Malaysia and Thailand. If M. fijiensis
was present, what was stopping it from becoming the dominant leaf spot? There
was also a problem concerning the leaf spot situation in South Asia. If M. fijiensis
had been recorded in Bhutan in 1985 (Peregrine, 1989), then why hadn’t it since
been found in neighbouring India?
First records
It became possible between 1992 and 1995 for the author, who was employed at
the time by the International Network for the Improvement of Banana and Plantain
(INIBAP), to collect specimens of leaf spot in the Southeast Asian/South Asian
region during visits for diagnosis. He hoped that this would help determine the
distribution of the two main Mycosphaerella leaf spot pathogens, which may help
explain the apparent lack of expansion of M. fijiensis in the region. Specimens,
which were thought to be mainly of M. musicola, were collected at different
locations and on different banana clones in southern India, Sri Lanka, West
Malaysia and Thailand. The specimens were sent by courier to Drs Xavier
Mourichon and Jean Carlier at the Centre de coopération internationale en
recherche agronomique pour le développement (CIRAD) in Montpellier for
identification. Unexpectantly, M. musicola was not identified from any of the
specimens collected. Some specimens from Johore in West Malaysia were found
to be M. fijiensis, but the majority of leaf spots from all countries were caused
by a fungus that was unknown. This fungus had Mycosphaerella as the perfect
(teleomorph) stage and what first appeared to be Septoria as its imperfect
(anamorph) stage (Anon., 1995; Carlier et al., 2000b). Initially, it was believed
that the pathogen might have been Phaeoseptoria musae (Anon.,1995), which has
been reported to have Mycosphaerella as a perfect stage and is fairly widespread
36
MyLsd 17x24
25/06/03
15:15
Page 37
Session 1
D.R. Jones
(Carlier et al., 2000a). However, this turned out not to be the case (Carlier et al.,
2000b). The new fungus was named M. eumusae (Carlier et al., 2000b).
Because of the similarity of symptoms with those of Sigatoka disease (Carlier et al.,
2000a), it is likely that the leaf spot caused by M. eumusae was seen by the author in
Malaysia and Thailand in 1988. Evidence suggested that it might have been the common
leaf spot of banana in the South Asia and parts of Southeast Asia. If so, M. eumusae
competed effectively with M. fijiensis and prevented it from becoming dominant.
A specimen of a leaf spot collected in the Mekong Delta in Vietnam by Ivan
Buddenhagen in 1995 was later identified as M. eumusae at CIRAD, as were specimens
from Mauritius in 1997 (Carlier et al., 2000b) and Réunion in 2000 (X. Mourichon,
France, personal communication). Re-examination of specimens collected at Onne,
Nigeria in 1989 and 1990, when M. fijiensis was presumed to be present, also revealed
the presence of M. eumusae (Carlier et al., 2000b). Details of findings of M. eumusae
that are documented by CIRAD are summarised in Table 3.
Table 3. Countries where Mycosphaerella eumusae, the causal agent of eumusae leaf spot disease, has been detected
in chronological order of records (modified from Table 1 of Carlier et al., 2000b with additional information supplied
by X. Mourichon). All identifications made by J. Carlier and M.F. Zapater, CIRAD.
Country
Nigeria (Onne)
Nigeria (Onne)
India (Bangalore)
Malaysia (Johor State)
Thailand (Sukothai)
Thailand (Surat Thani)
Thailand (Tha Yang)
India (Kannara)
Sri Lanka (Gannoruwa)
Sri Lanka (Nugahena)
Vietnam (Mekong Delta)
Mauritius
Réunion
Banana host
AAB clone, most likely in plantain subgroup
AAB clone, most likely in plantain subgroup
‘Grande naine’ (AAA, Cavendish subgroup)
‘Pisang kapas’ (AA or AAB)
‘Grande naine’ (AAA, Cavendish subgroup)
‘Williams’ (AAA, Cavendish subgroup)
‘Kluai hom thong’ (AAA)
‘Grande naine’ (AAA, Cavendish subgroup)
AAA clone in Cavendish subgroup
‘Anamala’ (AAA, syn. ‘Gros Michel’)
‘Sucrier’ (AA)
‘Grande naine’ (AAA, Cavendish subgroup)
‘Grande naine’ (AAA, Cavendish subgroup)
Year specimen was
collected and collector
1989 (IITA)
1990 (IITA)
1992 (D.R. Jones, INIBAP)
1993 (D.R. Jones, INIBAP)
1994 (D.R. Jones, INIBAP)
1994 (D.R. Jones, INIBAP)
1994 (D.R. Jones, INIBAP)
1995 (D.R. Jones, INIBAP)
1995 (D.R. Jones, INIBAP)
1995 (D.R. Jones, INIBAP)
1995 (I. Buddenhagen)
1997 (S.P. Beni Madhu, AREU)
2000 (C. Lavigne, CIRAD)
Recently, a re-examination of M. eumusae specimens and cultures has revealed
that the imperfect stage of the fungus is likely to be Pseudocercospora and not
Septoria as originally thought. However, the fungus can be distinguished from
M. musicola and M. fijiensis on morphological grounds (Crous and Mourichon, 2002)
and by ITS sequence analysis (Carlier et al., 2000b). A change in the common name
of the disease from Septoria leaf spot disease to eumusae leaf spot disease has been
proposed (Crous and Mourichon, 2002).
Impact
The economic impact of M. eumusae as a leaf pathogen of banana has still to be
evaluated. However, it is known to seriously affect cultivars in the important AAA
Cavendish and AAB plantain subgroups in southern India. It has also been observed
37
MyLsd 17x24
25/06/03
15:15
Page 38
Mycosphaerella leaf spot diseases of bananas: present status and outlook
causing large areas of necrosis on leaves of ‘Anamala’ (AAA, syn. ‘Gros Michel’) in
Sri Lanka. Other cultivars/clones seen with symptoms are ‘Kluai lep mu nang’ (AA),
‘Pisang mas’ (AA), ‘Pisang kapas’ (AA) and ‘Mysore’ (AAB) (Carlier et al., 2000a).
More research needs to be undertaken on the effect of M. eumusae on important
clones.
Summary and discussion
The three Mycosphaerella species causing leaf spot diseases of banana reported
above are serious pathogens. M. musicola and M. fijiensis were well documented
in the scientific literature because of their steady global or near global spread and
impact on banana cultivation. However, M. fijiensis, the more recent has had a
far greater impact. It is more difficult to control on plantations of dessert bananas
for export and local consumption, and also affects the production of cooking
bananas grown by resource-poor farmers. The arrival of M. fijiensis in West Africa
and the perceived threat that it posed to the livelihood of the peoples there led to
the formation of INIBAP and the plantain breeding programmes of the International
Institute of Tropical Agriculture (IITA) and the Centre africain de recherches sur
bananiers et plantains (CARBAP).
The third Mycosphaerella leaf spot pathogen is not so well known because it
has been overlooked until fairly recently. Evidence on distribution suggests that
it arose, like the others, in the Southeast Asian/Australasian region. It is not known
how long M. eumusae has affected banana in South and Southeast Asia, nor the
extent of its distribution. Extensive surveys of leaf spots in the region would help
clarify the situation. An estimate of the severity of the disease and the name of
the clone affected would help determine host reactions. Basic information on the
biology and epidemiology of the pathogen is also needed.
M. eumusae has been present in Onne, Nigeria for at least 13 years but was
not found in a recent thorough survey of leaf spot organisms in neighbouring
Cameroon (C. Abadie, France, personal communication). Therefore, in West Africa,
the pathogen may still be confined to southeast Nigeria. If so, one must speculate
on how it got here in the first place. Was M. eumusae introduced with planting
material from Asia?
M. eumusae seems to be an important pathogen and may be able to compete
with M. fijiensis. The apparent dominance of M. eumusae in parts of Asia suggests
that it was established before the introduction of M. fijiensis and thus perhaps
able to resist intrusion by the latter. The situation at Onne, Nigeria, may be different
with M. eumusae because arrival was after M. fijiensis became dominant over
M. musicola and subsequent competitive interactions favoured the established
pathogen. Work is needed to determine if this speculation has any basis in fact.
The identification of the leaf spot pathogens found on plantain at Onne may help
determine relative competitiveness on the main susceptible host in the area. The
identification of leaf spot pathogens on cultivars in the IITA germplasm collection
may help to determine the nature of leaf spot interactions on other clones.
When M. fijiensis first appeared, some believed that it might have arisen in
Fiji by mutation from M. musicola. Stover (1969) initially considered M. fijiensis
38
MyLsd 17x24
25/06/03
15:15
Page 39
Session 1
D.R. Jones
to be a physiological strain of M. musicola. The latter hypothesis is now considered
to be unlikely. How do we view the appearance of M. eumusae? How are the leaf
spot pathogens evolving? Recent phylogenetic studies indicate that M. musicola,
M. fijiensis and M. eumusae may once have had a common origin (P. Crous, South
Africa, personal communication). All may have arisen from similar saprophytic
or weakly pathogenic fungi growing on damaged or weakened leaf tissues of
banana. Stover (1969) reported M. minima as a saprophytic co-inhabitant with
M. musicola in leaf spots. He has also recorded M. musae was an endophyte in
Sigatoka leaf spots (Stover, 1994). Other fungi could be evolving as parasites in
senescing leaf tissue. Further investigations of speciation in Mycosphaerella and
other related genera found on banana leaves in the centre of origin of banana
and elsewhere may prove interesting.
The similarity of the necrotic symptoms caused by M. musicola, M. fijiensis
and M. eumusae and the fact that all three have Mycosphaerella as the perfect
stage suggests that the diseases should be grouped together as Sigatoka leaf spot
diseases. These diseases warrant fundamental research to clarify their evolution
and adaptability, and to help to find ways of breeding resistant bananas.
Acknowledgements
The author thanks the Australian Banana Growers’ Council for financial support to attend
the workshop on Mycosphaerella leaf spot pathogens.
References
Anonymous 1995. Musanews. INFOMUSA 2(2):26-30.
Belalcazar S.L. 1991. El Cultivo del Plátano en el Trópico. Pp. 288-289 in Manual de Assistencia
Técnica No. 50. Instituto Colombiano Agropecuario, Armenia, Colombia.
Bureau E. 1990. Adoption of a forecasting system to control black Sigatoka (Mycosphaerella
fijiensis Morelet) in plantain plantations in Panama. Fruits 45:329-338.
Carlier J. E. Fouré, F. Gauhl, D.R. Jones, P. Lepoivre, X. Mourichon and C. Pasberg-Gauhl.
2000a. Fungal diseases of the foliage. Pp. 37-142 in Diseases of Banana, Abacá and Enset
(D.R. Jones, ed.). CABI Publishing, Wallingford, UK.
Carlier J., M.F. Zapater, F. Lapeyre, D.R. Jones and X. Mourichon. 2000b. Septoria leaf spot
of banana: a newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90:884-890.
Carreel F. 1995. Etude de la diversité génétique des bananiers (genre Musa) à l’aide des
marqueurs RFLP. PhD thesis, Institut National Agronomique, Paris-Grignon, France.
Crous P. and X. Mourichon. 2002. Mycosphaerella eumusae and its anamorph Pseudocercospora
eumusae spp. nov.: causal agent of eumusae leaf spot disease of banana. Sydowia 54:
35-43.
Dabek A.J. and J.M. Waller. 1990. Black leaf streak and viral streak: new banana diseases in
east Africa. Tropical Pest Management 36(2):157-158.
Firman I.D. 1972. Susceptibility of banana cultivars to fungus diseases in Fiji. Tropical
Agriculture (Trinidad) 49:189-196.
Frossard P. 1980. Apparition d’une nouvelle et grave maladie foliaire des bananiers et
plantains au Gabon: la maladie des raies noires, Mycosphaerella fijiensis Morelet. Fruits
35:443-453.
39
MyLsd 17x24
25/06/03
15:15
Page 40
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Fullerton R.A. 1987. Banana production in selected Pacific islands. Pp. 57-62 in Banana
and Plantain Breeding Strategies, Proceedings of an International Workshop held in
Cairns, Australia, 13-17 October 1986 (G.J. Persley and E.A. De Langhe, eds). ACIAR
Proceedings 21, Australian Centre for International Agricultural Research, Canberra,
Australia.
Jones D.R. 1990. Black Sigatoka in the Southeast Asian–Pacific region. Musarama 3 (1):2-5.
Jones D.R. 2000. Introduction to banana, abacá and enset. Pp. 1-36 in Diseases of Banana,
Abacá and Enset (D.R. Jones, ed.). CABI Publishing, Wallingford, UK.
Jones D.R. and J.L. Alcorn. 1982. Freckle and black Sigatoka diseases of banana in far north
Queensland. Australasian Plant Pathology 11:7-9.
Leach R. 1941. Banana leaf spot Mycosphaerella musicola, the perfect stage of Cercospora
musae Zimm. Tropical Agriculture (Trinidad) 18:91-95.
Leach R. 1946. Banana Leaf Spot (Mycosphaerella musicola) on the Gros Michel Variety in
Jamaica. Investigations into the Aetiology of the Disease and Principles of Control by
Spraying. Bulletin, Government Printer, Kingston, Jamaica.
Leach R. 1964a. Report on Investigations into the Cause and Control of the New Banana
Diseases in Fiji, Black Leaf Streak. Council Papers Fiji 38, Suva.
Leach R. 1964b. A new form of banana leaf spot in Fiji, black leaf streak. World Crops 16:D6064.
Long P.G. 1979. Banana black leaf streak disease (Mycosphaerella fijiensis) in Western Samoa.
Transactions of the British Mycological Society 72:299-310.
Magee C.J.P. 1953. Some aspects of the bunchy top disease of banana and other Musa spp.
Journal and Proceedings of the Royal Society of New South Wales 87:3-18.
Massee G. 1914. Fungi exotici: XVIII. Kew Bulletin 1914, 159.
Meredith D.S. 1970. Banana Leaf Spot Disease (Sigatoka) caused by Mycosphaerella musicola
Leach. Phytopathological Papers, No. 11, Commonwealth Mycological Institute, Kew,
Surrey, UK, 147pp.
Meredith D.S. and J.S. Lawrence. 1969. Black leaf streak disease of bananas (Mycosphaerella
fijiensis): Symptoms of disease in Hawaii, and notes on the conidial state of the causal
fungus. Transactions of the British Mycological Society 52:459-476.
Mobambo K.N., F. Gauhl, R. Swennen and C. Pasberg-Gauhl. 1996. Assessment of the cropping
cycle effects of black leaf streak severity and yield decline of plantain and plantain hybrids.
International Journal of Pest Management 42:1-8.
Mulder J.L. and R.H. Stover. 1976. Mycosphaerella species causing banana leaf spot.
Transactions of the British Mycological Society 67:77-82.
Parnell M., P.J.A. Burt and K. Wilson. 1998. The influence of exposure to ultraviolet radiation
in simulated sunlight on ascospores causing Black Sigatoka disease of banana and plantain.
International Journal of Biometeorology 42:22-27.
Peregrine W.T.H. 1989. Black leaf streak of banana Mycosphaerella fijiensis Morelet. FAO Plant
Protection Bulletin 37(3):130.
Philpott J. and C.H. Knowles. 1913. Report on a Visit to Sigatoka. Phamphlet of the Department
of Agriculture, Fiji 3, Suva.
Pons N. 1987. Notes on Mycosphaerella fijiensis var. difformis. Transactions of the British
Mycological Society 89:120-124.
Raemaekers R. 1975. Black leaf streak disease in Zambia. PANS 21:396-400.
Rhodes P.L. 1964. A new banana disease in Fiji. Commonwealth Phytopathological News 10:
38-41.
Robinson J.C. 1996. Bananas and Plantains. Crop Production Science in Horticulture 5, CAB
International, Wallingford, UK, 238pp.
40
MyLsd 17x24
25/06/03
15:15
Page 41
Session 1
D.R. Jones
Romero C. R. 1986. Impacto de Sigatoka negra y roya del cafeto en actividad platanera
nacional. Revista de la Asociación Bananera Nacional (ASBANA), San José, Costa Rica
12(9):7-10.
Stover R.H. 1962. Intercontinental spread of banana leaf spot (Mycosphaerella musicola Leach).
Tropical Agriculture (Trinidad) 29:327-338.
Stover R.H. 1969. The Mycosphaerella species associated with banana leaf spots. Tropical
Agriculture (Trinidad) 46:325-332.
Stover R.H. 1972. Banana, Plantain and Abaca Diseases. Commonwealth Mycological
Institute, Kew, Surrey, UK, 316pp.
Stover R.H. 1976. Distribution and cultural characteristics of the pathogens causing leaf spot.
Tropical Agriculture (Trinidad) 53:111-114.
Stover R.H. 1978. Distribution and probable origin of Mycosphaerella fijiensis in Southeast
Asia. Tropical Agriculture (Trinidad) 55:65-68.
Stover R.H. 1980. Sigatoka leaf spots of banana and plantain. Plant Disease 64:750-755.
Stover R.H. 1990. Sigatoka leaf spots: thirty years of changing control strategies: 1959-1989.
Pp. 66-74 in Sigatoka Leaf Spot Diseases of Bananas, Proceedings of an International
Workshop held in San José, Costa Rica, March 28-April 1, 1989 (R.A. Fullerton and
R.H. Stover, eds). INIBAP, Montpellier, France.
Stover R.H. 1994. Mycosphaerella musae and Cercospora “non-virulentum” from Sigatoka
leaf spots are identical. Fruits 49:187-190.
Tushemereirwe W.K. 1996. Factors influencing the expression of leaf spot diseases of highland
bananas in Uganda. PhD thesis, University of Reading, UK, 197 pp.
Zimmerman A. 1902. Uber einige tropischer Kulturpflanzen beobachtete Pilze. Zentralblatt
für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene 8:219 (abs).
41
MyLsd 17x24
25/06/03
15:15
Page 43
Session 1
P.W. Crous et al.
Integrating morphological and
molecular data sets on Mycosphaerella,
with specific reference to species
occurring on Musa
P. W. Crous1, J.Z. Groenewald1, A. Aptroot2, U. Braun3,
X. Mourichon4 and J. Carlier4
Abstract
The genus Mycosphaerella (= Sphaerella) is one of the largest genera of ascomycetes, containing
more than 3000 named taxa. Approximately 23 anamorph genera have been linked to
Mycosphaerella via cultural studies. Several of these anamorph genera have recently been
reduced to synonymy based on phylogenetic studies derived from ITS1, 5.8S and ITS2 DNA sequence
data. In addition, several genera not previously associated with Mycosphaerella, have also been
shown to cluster within Mycosphaerella, which has proved to be largely monophyletic. From these
results, as well as a re-evaluation of the criteria upon which anamorph genera are distinguished
in this complex, a reduced set of informative criteria and genera are proposed. The degree of scar
thickening, darkening and refraction, as well as the presence or absence of pigmentation in
conidiophores and conidia appear to be useful features delimiting anamorph genera of
Mycosphaerella. Species, however, are still separated on a combination of characters such as
conidiomatal structure, the nature and arrangement of conidiophores, conidiogenesis, dehiscence
scars and pigmentation. For the species that occur on Musa, anamorph morphology appears to
be more informative than the more conserved teleomorph morphology, and can be used to
separate the major pathogens, namely M. fijiensis (the causal agent of black leaf streak disease),
M. musicola (the causal agent of Sigatoka disease), M. eumusae (the causal agent of eumusae
leaf spot disease), as well as other reputedly less important pathogens.
Resumen - Integración de los conjuntos de datos morfológicos y moleculares en
Mycosphaerella, con referencia específica a las especies que ocurren en Musa
El género Mycosphaerella (= Sphaerella) es uno de los géneros más grandes de ascomicetos, que
contiene más de 3000 taxa nombrados. Aproximadamente 23 géneros anamorfos diferentes han
estado vinculados con Mycosphaerella mediante estudios culturales. Varios de estos géneros
anamorfos fueron reducidos recientemente debido a la sinonimia basada en estudios filogenéticos
1
University of Stellenbosch, Matieland, South Africa
Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands
Martin-Luther-Universität, Halle (Saale), Germany
4 CIRAD, Montpellier, France
2
3
43
MyLsd 17x24
25/06/03
15:15
Page 44
Mycosphaerella leaf spot diseases of bananas: present status and outlook
derivados de los datos de las secuencias de ADN ITS1, 5.8S y ITS2. En adición, se ha demostrado que
varios géneros que no estaban asociados previamente con Mycosphaerella, actualmente están
agrupados dentro del género Mycosphaerella, el cual está comprobado, que es extensamente
monofilético. De estos resultados, así como de una nueva evaluación de los criterios de acuerdo a
los cuales se distinguen géneros anamorfos en este complejo, se propone un conjunto reducido de
criterios y géneros informativos. El grado del espesor, oscurecimiento y refracción de las cicatrices,
así como la presencia o ausencia de pigmentación en conidióforos y conidios parecen representar
características útiles que delimitan los géneros anamorfos de Mycosphaerella. Sin embargo, las
especies aún están separadas en una combinación de caracteres como la estructura conidiomatal,
la naturaleza y el arreglo de los conidióforos,conidiogénesis,cicatrices de dehiscencia y pigmentación.
Para las especies que ocurren en Musa, la morfología anamorfa parece ser más informativa que la
morfología teleomorfa más conservada, y puede ser utilizada para separar los principales patógenos,
a saber M. eumusae (agente causal del ELSD), M. musicola (agente causal de la enfermedad de
Sigatoka), M. fijiensis (agente causal de la raya negra de la hoja).
Résumé - L’intégration des données morphologiques et moléculaires sur Mycosphaerella,
particulièrement celles des espèces présentes sur Musa
Le genre Mycosphaerella (= Sphaerella) est un des genres les plus représentés des ascomycètes
avec plus de 3000 taxa. Environ 23 genres anamorphes ont été liés à Mycosphaerella à l’aide
d’études sur les cultures. Des études phylogénétiques à partir de séquences d’ADN ITS1, 5.8S
et ITS2, ont permis d’identifier les synonymes parmi ces genres anamorphes. De plus,
plusieurs genres qui n’étaient pas associés auparavant à Mycosphaerella, se regroupent dans
ce genre qui s’est avéré être principalement monophylétique. A partir de ces résultats ainsi
que par la réévaluation des critères à partir desquels on peut distinguer les genres
anamorphes, nous proposons une réduction du nombre de critères pertinents et de genres.
Le degré d’épaississement, de noircissement et de réfraction des cicatrices, ainsi que la
présence ou l’absence de pigmentation dans les conidiophores et les conidies, semblent être
des critères intéressants pour délimiter les genres anamorphes de Mycosphaerella. Les
espèces se distinguent toutefois par une combinaison de caractères tels que la structure des
conidioma, la nature et la disposition des conidiophores, la conidiogénèse, les cicatrices de
déhiscence et la pigmentation. Pour les espèces qui se trouvent sur Musa, la morphologie
de l’anamorphe est plus instructive que la morphologie moins variable du téléomorphe et
peut être utilisée pour distinguer les principaux pathogènes, soit M. fijiensis (l’agent causal
de la maladie des raies noires), M. musicola (l’agent causal de la maladie de Sigatoka),
M. eumusae (l’agent causal de l’ELSD, eumusae leaf spot disease), ainsi que d’autres pathogènes
considérés moins importants.
What is Mycosphaerella?
Corlett (1991, 1995) lists more than 2000 species belonging to the genus
Mycosphaerella Johanson (Dothideales: Mycosphaerellaceae) (including the fungi
described in Sphaerella Ces. et De Not.). This makes it one of the largest genera
of ascomycetes known. Furthermore, this genus has been confirmed to have
anamorphs in at least 23 different genera (Crous et al., 2000), including Cercospora
Fres. and Septoria Sacc., which alone encompass several thousand species (Pollack,
1987; Sutton, 1980). Some species are saprobes, but most are known from necrotic
lesions that are associated with leaves, stems or fruit of various hosts (Park and
Keane, 1984). Host specificity still plays a major role in the identification of species,
where taxa are chiefly compared with those that occur on a specific host genus
or family of phanerogamic plants (Chupp, 1954; Braun, 1995). Von Arx (1949)
44
MyLsd 17x24
25/06/03
15:15
Page 45
Session 1
P.W. Crous et al.
regarded some species as polyphagous but the concept of host specificity is still
strongly adhered to, especially with species shown to be plant pathogens.
Species of Mycosphaerella are usually defined as having small, black, immersed
or erumpent to almost superficial pseudothecial ascomata, with various degrees
of stromatic tissue surrounding the ascomata, and pale to brown superficial
mycelium being present or absent, smooth or verruculose. The ascomatal wall
consists of 3–6 layers of pseudoparenchyma cells but in some taxa this can also
be more prominently developed, especially around the ostiole. Ostioles are
singular, central, and usually lined with periphyses; in some taxa, however, this
develops further, giving the impression of periphysoids. The hamathecium dissolves
at maturity, and no stromatic tissue remains between the asci. Asci are bitunicate,
8-spored, sessile, arranged in a cluster, hyaline, ovoid to obovoid, ellipsoidal or
cylindrical, rarely clavate. Ascospores are bi to multi-seriate, thin to thick-walled,
guttulate or not, 1-septate, constricted at the septum or not, usually hyaline
and smooth, rarely with a mucous sheath, fusoid to obovoid, ellipsoid or elongate.
Anamorphs are highly variable, including numerous hyphomycete and coelomycete
genera, namely Cercospora Fres., Cercosporella Sacc., Cladosporium Link, Clypeispora Ramaley, Colletogloeopsis Crous et M.J. Wingf., Dissoconium De Hoog,
Oorschot et Hijwegen, Fusicladiella Höhn., Miuraea Hara, Passalora Fr.,
Phaeophleospora Rangel, Phloeospora Wallr., Pseudocercospora Speg., Pseudocercosporella Deighton, Ramularia Unger, Septoria Sacc., Sonderhenia H.J. Swart
and J. Walker, Stenella Syd., Thedgonia B. Sutton, Uwebraunia Crous et M.J. Wingf.,
Xenostigmina Crous (Crous et al., 2000; 2001).
Given the wide range of variation among anamorphs, it was not clear whether
Mycosphaerella was monophyletic, paraphyletic or polyphyletic. Crous (1998)
hypothesized that Mycosphaerella could be split into groups as indicated by the
various anamorph genera. Based on the revision of the monograph on
Mycosphaerella by A. Aptroot (CBS, The Netherlands), several sections are recognized
(modified from Barr 1972):
• Section Mycosphaerella is characterized by cylindrical asci and mostly
uniseriate, thin-walled, often small ascospores that are constricted at the septum
and inequilateral, with rounded upper ends. Anamorphs: typically Ramularia with
Asteromella spermatial states. Representative species: the common polyphagous
M. punctiformis (Pers. : Fr.) Starb.
• Section Tassiana M.E. Barr is characterized by pyriform asci and irregularly
arranged, thick-walled ascospores that are often large and constricted at the
septum and nearly equilateral, relatively broad with rounded ends. Anamorph:
typically Cladosporium. Representative species: the common polyphagous species
M. tassiana (de Not.) Joh. (with large ascospores) and M. longissima (Fuck.) Lindau
(with small ascospores). Further research is still required to determine whether
the teleomorphs of Cladosporium subgen. Heterosporium (David, 1997) can also
be accommodated in this section. Preliminary data suggest, however, that
Cladosporium may fall outside Mycosphaerella (Crous et al., 2001, unpublished
data).
45
MyLsd 17x24
25/06/03
15:15
Page 46
Mycosphaerella leaf spot diseases of bananas: present status and outlook
• Section Caterva M.E. Barr is characterized by cylindrical asci and irregularly
arranged, thin-walled, often medium-sized ascospores that are rarely constricted
at the septum and inequilateral, with more or less pointed ends. Asteroma and
Asteromella spermatial forms are typical. Representative species: the common
polyphagous M. subradians (Fr. : Fr.) Schroeter.
• Section Longispora M.E. Barr is characterized by cylindrical asci with
aggregated, thin-walled, long and slender ascospores that are rarely constricted
at the septum and mostly equilateral, long but slender ascospores, characteristically
with rounded upper and pointed lower ends. Anamorphs: Phloeospora or Septoria
sensu lato. Representative species: M. millegrana (Cooke) Schroeter (with short
spores), M. latebrosa (Cooke) Schroeter (with longer spores) and M. populi
(Auersw.) J. Schröt. (with the longest spores in the genus). The genus Sphaerulina
Sacc., which differs only by additional septa, may be synonymous.
• Section Fusispora M.E. Barr is characterized by pyriform asci and irregularly
arranged, thin-walled ascospores that are rarely constricted at the septum and
mostly equilateral, fusiform, pointed ascospores. Anamorphs have not been proved.
• Section Plaga M.E. Barr (including Section Macula M.E. Barr) includes endophytic
species sporulating on leaf spots, many of which are described as plant pathogens.
This section is characterized by obovoid to ellipsoidal or cylindrical asci, small to
medium sized ascopores, fusiform to obovoid with rounded ends. Many species have
been described in these groups, and the majority originate from warm-temperate
and tropical areas. Anamorphs include Colletogloeopsis, Mycovellosiella, Phaeophleospora, Pseudocercospora, Pseudocercosporella, Sonderhenia, Stenella Syd., Dissoconium, Uwebraunia and possibly others. Representative species: listed by Crous (1998)
on Eucalyptus.
How do we separate species in this complex based
on morphology?
Mycosphaerella encompasses species with a saprobic, plant pathogenic as well
as a hyperparasitic habit. In general, however, most species are found to be
associated with leaf spots. Some species have been isolated as endophytes (Crous,
1998) but this is not the norm. Importantly, up to four taxa have been reported
from the same lesion on leaves of Eucalyptus (Crous, 1998) suggesting that some
act as primary and others as secondary or weak pathogens.
In the past, the taxonomy of Mycosphaerella relied mostly on aspects such as
host, symptom type, and teleomorph morphology (pseudothecium, ascus and
ascospore morphology). Very few studies focused on cultures, therefore ascospore
germination patterns and cultural characteristics are unknown for most species.
Furthermore, no ex-type cultures are available for molecular studies.
Most links to anamorphs have also been based on association. Given
the concept of sympatric colonisation of host tissue discussed above, this has led
to numerous erroneous reports of anamorph/teleomorph links. Anamorph
46
MyLsd 17x24
25/06/03
15:15
Page 47
Session 1
P.W. Crous et al.
morphology has focused strongly on aspects such as conidiomatal structure, and
mode of conidiogenesis (von Arx, 1983; Sutton and Hennebert, 1994). An
important overlap has been observed between different conidiomatal types (Nag
Raj, 1993; Braun 1995), hence species have been transferred from one anamorph
genus to another (Sutton et al., 1996; Braun, 1998). The separation of some
coelomycete and hyphomycete anamorphs of Mycosphaerella is, therefore,
debatable.
Subsequent to the wide taxonomic concept employed for the cercosporoid fungi
by Chupp (1954), Deighton (1973, 1976, 1987, 1990) recognised the value of
pigmentation, conidiogenesis and conidium release. Different types of dehiscence
scars (David, 1993) have subsequently been recognised to separate genera
such as Cladosporium, Cercospora and Stenella. Conidiogenesis is variable in this
complex (Verkley, 1997), as are dehiscence scars (Crous, 1998), but it still remains
a useful feature to separate species (Braun, 1995).
Integrating morphological and molecular data sets
The issue of integrating morphological and molecular data sets in Mycosphaerella
is beset with numerous problems. Several hypotheses have been proposed to divide
the genus into separate parts, groups, sections or genera. The molecular work to
date, however, mainly supported one major clade of Mycosphaerella (Table 1), as
well as two minor clades, typified by Dissoconium and Cladosporium (Crous et
al., 2000, 2001). That anamorph concepts have not always correlated with different
groups in Mycosphaerella, but have been shown to have evolved more than once
within the genus, has caused confusion. Sometimes, however, integration has
simplified or reduced the genera. For example, species of Mycovellosiella
(superficial mycelium; conidia in chains), Phaeoramularia (no superficial
mycelium, conidia in chains), and Passalora (no superficial mycelium, conidia
solitary) consistently cluster together, leading to synonymy of the genera under
the older name Passalora (Crous et al., 2001). Pseudocercospora is another
confusing example. Published work (Crous et al., 2001) indicates that several other
genera are also clustered here, including Pseudophaeoramularia U. Braun, and
Stigmina, while the position of Stenella was still not clearly resolved. As more
taxa have been added to our analyses (unpublished data), it has become clear that
Stenella and Stigmina are in fact good genera, whereas Pseudophaeoramularia
clusters in Pseudocercospora. These findings are still in agreement with those of
Crous et al. (2001), namely that conidial catenulation is not a good feature at the
generic level, and that the degree of thickening of spore scars is important, rather
than just presence or absence.
Molecular data will also influence the way we define the teleomorph.
Preliminary data suggest that ascospore septation, and the presence of
pseudoparenchymatal stromatic tissue around the ostiole is of lesser taxonomic
value at the generic level (unpublished data). This will result in many older names
having to be re-evaluated, and may even eventually require the conservation of
Mycosphaerella, to safely preserve the name for future use over older, but lesser
known names.
47
MyLsd 17x24
25/06/03
15:15
Page 48
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Integrating morphology and molecular data sets on a species level is also beset
with problems. Firstly, as discussed above, most species are known from dried
herbarium material only and there are no reliable reference strains. Most strains
that can be ordered from culture collections are sterile (a common phenomenon
associated with species of Mycosphaerella) therefore their identities cannot be
confirmed. Many of the species already sequenced in GenBank originate from such
strains, and thus these data that are routinely used by plant pathologists obscure
the issue even further, even for the few species presently known on Musa.
Table 1. Fungal isolates included for ITS sequence analysis
Accession no.
Teleomorph
Anamorph
U04234
Leptosphaeria microscopica
Unknown
PCR18
PP15
Mycosphaerella cruenta
M. berkeleyi
Pseudocercospora cruenta
Passalora personata
458
487
PF7
PF8
01A
009
PFD9
M. eumusae
M. eumusae
M. fijiensis
M. fijiensis
M. fijiensis
M. fijiensis
M. fijiensis var. difformis
PM10
PM11
121
090
CA1
MA12
M. musicola
M. musicola
M musicola
M musicola
Mycosphaerella
state unknown
Mycosphaerella
state unknown
Unknown
Pseudocercospora eumusae
Pseudocercospora eumusae
Paracercospora fijiensis
Paracercospora fijiensis
Paracercospora fijiensis
Paracercospora fijiensis
Paracercospora fijiensis
var. difformis
Pseudocercospora musae
Pseudocercospora musae
Pseudocercospora musae
Pseudocercospora musae
Cercospora apii
IMI 271341
Unknown
CH5, CH6
Cercospora hayi
Mycocentrospora acerina
Phaeoseptoria musae
Origin
ATCC 42652 (Saccharum
officinarum, Kenya)
ATCC 262271 (Vigna, Puerto Rico)
MPPD L2121 (Arachis,
Oklahoma, U.S.A.)
Musa, Malaysia
Musa, Thailand
ATCC 22116 (Musa, Philippines)
ATCC 22117 (Musa, Hawaii)
Musa, Philippines
Musa, Ntoum, Gabon
ATCC 36054 (Musa, Honduras)
ATCC 22115 (Musa, Philippines)
ATCC 36143 (Musa, Honduras)
Musa, Indonesia, West Java
Musa, Armenia, Colombia
ATCC 12246
ATCC 12234
(Musa, Cuba)
ATCC 34539 (Daucus carota,
Norway)
Musa, Honduras
Re-evaluating species occurring on Musa
Mycosphaerella eumusae, M. fijiensis and M. musicola
The Mycosphaerella state of M. eumusae is morphologically very similar to that
of M. musicola. As expected, reports from literature also suggest that these two
pathogens have commonly been confused in the past, thereby also questioning
the value of much of the published literature (and distribution records) on this
disease complex. Leaf spot symptoms attributed to M. eumusae were first seen
after a survey conducted in Southeast Asia during 1992–1995 (Carlier et al.,
2000b). The anamorph of M. eumusae was initially regarded as a species of
48
MyLsd 17x24
25/06/03
15:15
Page 49
Session 1
P.W. Crous et al.
Septoria (Carlier et al., 2000a, b). Pseudocercospora eumusae, the anamorph of
M. eumusae, is characterized by having predominantly epiphyllous sporodochia
that form on dark brown substomatal stromata. The sporodochia are mingled with
developing spermatogonia. Young sporodochia are subepidermal and substomatal, and initially produce conidia that appear to be exuding from a
subepidermal, substomatal pycnidium. In cross section, however, the subepidermal
and substomatal structure is seen to be a sporodochium, not a pycnidium. As more
stromatal tissue is formed, conidiophores become erumpent, and sporodochia burst
through the epidermis, almost appearing acervular, but in fact being subepidermal
sporodochia. Conidiophores are subhyaline to pale olivaceous, becoming pale
brown at the base, subcylindrical, 0 to 3-septate, 10 to 25 x 3–5 µm, with
conidiogenous cells terminating in truncate ends. Sporodochia of M. eumusae
develop in a similar fashion to those of M. musicola but the conidiophores are
much longer and more septate in the former. Conidia of P. eumusae are
subhyaline to pale olivaceous, subcylindrical, (18–)30–50(–65) x (2–)2.5–3 µm,
3 to 8-septate, and have subtruncate ends without visible scars. Conidia can be
distinguished from those of M. musicola by their more cylindrical shape,
subtruncate ends, and shorter dimensions (Crous and Mourichon, 2002). Based
on ITS sequence data (Figure 1), the two species are also very closely related.
Isolates of M. fijiensis (anamorph: P. fijiensis) are easily distinguished from P.
musae and P. eumusae by their minutely thickened spore scars (Deighton, 1979).
These scars have been shown to be of lesser phylogenetic importance in the
cercosporoids (Stewart et al., 1999), and Paracercospora should be merged back
into Pseudocercospora (Crous et al., 2000, 2001), but they are still valuable at the
species level and should be used to separate taxa. Nevertheless, the anamorph of
M. fijiensis should now correctly be referred to as Pseudocercospora fijiensis.
Other species of Mycosphaerella
As can be seen below, numerous additional species of Mycosphaerella (or their
anamorphs) have been described from Musa. Little is known about many of these
taxa, but they are expected to occur on lesions typically associated with the major
pathogens discussed above. Some species, e.g. Cercospora hayi, appear to have
a wider host range than just Musa. From general morphology, C. hayi resembles
what is currently referred to as the Cercospora apii sensu lato morphotype. It is
suspected that Cercospora apii is a weak or secondary pathogen with a wide host
range, and has been described from numerous hosts worldwide. Based on
morphology, C. hayi is indistinguishable from C. apii. From the ITS sequence data
(Figure 1), the two species are clearly similar, and should be regarded as
conspecific, with preference being given to the older name, C. apii (Braun et al.,
unpublished data). Cladosporium musae, and several other species of Cladosporium
have also been recorded from Musa. The generic affinities, and circumscription
of genera within the Cladosporium-complex, however, remain to be resolved
pending a full morphological and molecular study (Braun et al. unpublished data).
Mycosphaerella musae is another interesting species and is commonly associated
with Mycosphaerella speckle. From an analysis of published literature, and
49
MyLsd 17x24
25/06/03
15:15
Page 50
Mycosphaerella leaf spot diseases of bananas: present status and outlook
purported anamorphs associated with this species (Stover, 1994; Carlier et al.,
2000a), it is clear that several different biological species have in the past been
treated as representative of M. musae.
Mycocentrospora acerina MA12
Phaeoseptoria musae IMI271341
100
Leptosphaeria microscopica U04234
M. musae 122
Mycosphaerella musae
M. berkeleyi PP15
100
C. hayi CH5
100
63
C. apii CA1
61
Cercospora apii sensu lato
C. hayi CH6
M. cruentaPCR18
69
100
99
79
89
88
Ps. musae PM11
Ps. musaePM10
Mycosphaerella musicola
M. musicola 121
M. musicola 090
M. eumusae 458
100
Mycosphaerella eumusae
M. eumusae 487
51
92
M. fijiensis 01A
M. fijiensis 009
54
Pa. fijiensis PF8
Mycosphaerella fijiensis
74
Pa. fijiensis var.
difformis PFD9
10 changes
Pa. fijiensis PF7
Figure 1. One of eight most parsimonious trees (length = 697 steps, CI = 0.803, RI = 0.795, RC = 0.639). Bootstrap
support from 1000 replicates is shown above the lines. Mycocentrospora acerina, Phaeoseptoria lysae and
Leptosphaeria microscopica were used as outgroups.
M. = Mycosphaerella, Ps. = Pseudocerospora and Pa. = Paracercospora.
50
MyLsd 17x24
25/06/03
15:15
Page 51
Session 1
P.W. Crous et al.
Major Mycosphaerella diseases
Black leaf streak disease
Mycosphaerella fijiensis M. Morelet, Ann. Soc. Sci. Nat. Archéol. Toulon Var. 21:105.
1969.
= Mycosphaerella fijiensis var. difformis J.L. Mulder & R.H. Stover, Trans. Brit. Mycol.
Soc. 67:82. 1976.
Anamorph: Pseudocercospora fijiensis (M. Morelet) Deighton, Mycol. Pap. 140:144.
1976.
≡ Cercospora fijiensis M. Morelet, Ann. Soc. Sci. Nat. Archéol. Toulon Var. 21:105.
1969.
≡ Paracercospora fijiensis (M. Morelet) Deighton, Mycol. Pap. 144:51. 1979.
= Cercospora fijiensis var. difformis J.L. Mulder & R.H. Stover, Trans. Brit. Mycol.
Soc. 67:82. 1976.
≡ Paracercospora fijiensis var. difformis (J.L. Mulder & R.H. Stover) Deighton, Mycol.
Pap. 144:52. 1979.
Host(s) and Distribution: Musa acuminata, Musa spp.; American Samoa,
Australia, Belize, Benin, Bhutan, Bolivia, Brazil, Burundi, Cameroon, Central African
Republic, China, Colombia, Comoros, Congo, Cook Islands, Costa Rica, Côte d’Ivoire,
Cuba, Dominican Republic, Ecuador, El Salvador, Fiji, French Polynesia, Gabon,
Ghana, Guatemala, Guinea-Bissau, Guyana, Haiti, Honduras, Indonesia, Jamaica,
Kenya, Malawi, Malaysia, Mayotte, Mexico, Micronesia, Netherlands Antilles, New
Caledonia, Nicaragua, Niger, Nigeria, Niue, Norfolk Island, Northern Mariana
Islands, Panama, Papua New Guinea, Peru, Philippines, Rwanda, Samoa, São Tomé
and Principe, Singapore, Solomon Islands, Somalia, Tahiti, Taiwan, Tanzania
(Pemba, Zanzibar), Thailand, Togo, Tonga, Uganda, USA (FL, HI), Vanuatu,
Venezuela, Vietnam, Wallis and Futuna Islands, Zambia.
Eumusae leaf spot disease
Mycosphaerella eumusae Crous et X. Mourichon, Sydowia 54:36. 2002.
Anamorph: Pseudocercospora eumusae Crous et X. Mourichon, Sydowia, 54:36. 2002.
Leaf spots amphigenous, initially visible as faint brown streaks, developing
into oval or elliptical light brown lesions with grey centres and dark brown
borders, coalescing to form large, brown, necrotic areas under favourable
conditions. Grey spots and patches are visible in necrotic areas, and lesions are
surrounded by a chlorotic yellow zone. Pseudothecia amphigenous, predominantly
hypophyllous, black, subepidermal, becoming slightly erumpent, globose, up to
80 µm diam., apical ostiole 10–15 µm wide; wall consisting of 2–3 layers of
medium brown textura angularis. Asci aparaphysate, fasciculate, bitunicate,
subsessile, obovoid, straight or slightly incurved, 8-spored, 30–50 x 9–15 µm.
Ascospores tri- to multiseriate, overlapping, hyaline, guttulate, thick-walled,
51
MyLsd 17x24
25/06/03
15:15
Page 52
Mycosphaerella leaf spot diseases of bananas: present status and outlook
straight, obovoid with obtuse ends, widest in the middle of apical cell, medianly
1-septate or basal cell slightly longer than apical cell, tapering towards both ends,
but with a more prominent taper towards lower end, (11–)12–13(–16.5) x
(3–)3.5–4(–4.5) µm. Spermogonia predominantly hypophyllous, subepidermal,
substomatal, globose, dark brown, up to 75 µm diam. Spermatia hyaline, rodshaped, 3–6 x 1–2 µm. Mycelium internal, pale brown, consisting of septate,
branched, smooth hyphae, 1.5–2.5 µm wide. Caespituli sporodochial, subepidermal,
substomatal, predominantly epiphyllous, grey, up to 100 µm wide. Conidiophores
aggregated in dense fascicles arising from the upper cells of a brown stroma up
to 70 µm wide; conidiophores subcylindrical, smooth, hyaline or pale brown below,
0–3-septate, straight to geniculate-sinuous, unbranched or branched below,
10–25 x 3–5 µm. Conidiogenous cells terminal, unbranched, hyaline, smooth,
tapering to flat-tipped apical loci, proliferating sympodially, or 1–4 times
percurrently near the apex, 10–20 x 3–4 µm; scars inconspicuous. Conidia solitary,
subhyaline to pale olivaceous, thick-walled, smooth, subcylindrical, apex obtuse,
base subtruncate, straight to variously curved, 3–8-septate, (18–)30–50(–65) x
(2–)2.5–3 µm; hila inconspicuous. Cultural characteristics: colonies pale olivaceous
grey (23””’d according to Rayner, 1970) to rosy vinaceous (7”d) (surface), and
brown vinaceous (5”’m) (bottom), with even margins and moderate aerial
mycelium, obtaining 10 mm diam. after 2 months at 25°C in the dark (Crous and
Mourichon, 2002).
Host(s) and Distribution: Musa spp.; Southern India, Sri Lanka, Thailand,
Malaysia, Vietnam, Mauritius, Nigeria.
Sigatoka disease
Mycosphaerella musicola R. Leach ex J.L. Mulder, Trans. Brit. Mycol. Soc. 67:77.
1976.
≡ Mycosphaerella musicola R. Leach, Trop. Agric. 18:92. 1941. (nom. nud.).
Anamorph: Pseudocercospora musae (Zimm.) Deighton, Mycol. Pap. 140:148. 1976.
≡ Cercospora musae Zimm., Centralbl. Bakteriol. Parasitenk. 2. Abth. 8:219. 1902.
= Cercospora musae Massee, Bull. Misc. Inform. 28:159. 1914.
Host(s) and Distribution: Musa acuminata, M. banksii, M. basjoo, M. liukiuensis,
M. paradisiaca, M. textiles; widely distributed, including American Samoa, Angola,
Antigua and Barbuda, Argentina, Australia, Bahamas, Barbados, Belau, Belize, Bolivia,
Brazil, Brunei Darussalam, Bhutan, Cambodia, Cameroon, Cape Verde, Cayman
Islands, China, Colombia, Congo, Cook Islands, Costa Rica, Côte d’Ivoire, Cuba,
Dominica, Dominican Republic, Ecuador, El Salvador, Ethiopia, Fiji, French Guiana,
French Polynesia, Gabon, Ghana, Grenada, Guadeloupe, Guam, Guatemala, Guinea,
Guinea-Bissau, Guyana, Haiti, Honduras, Hong Kong, India, Indonesia, Jamaica,
Kenya, Kiribati, Laos, Madagascar, Malawi, Malaysia, Martinique, Mauritius, Mexico,
Micronesia, Montserrat, Mozambique, Nepal, New Caledonia, Nicaragua, Nigeria,
Niue, Norfolk Island, Panama, Papua New Guinea, Peru, Philippines, Puerto Rico,
Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Samoa,
52
MyLsd 17x24
25/06/03
15:15
Page 53
Session 1
P.W. Crous et al.
Sao Tome and Principe, Sierra Leone, Solomon Islands, Somalia, South Africa,
Sri Lanka, Surinam, Taiwan, Tanzania, Thailand, Togo, Tonga, Trinidad and Tobago,
Tuvalu, Uganda, USA (FL, HI), Vanuatu, Venezuela, Vietnam, Wallis and Futuna
Islands, Yemen, Zambia, Zimbabwe.
Other diseases caused by species of Mycosphaerella
and its anamorphs
Cercospora hayi Calp., Studies on the Sigatoka Disease of Bananas and its Fungus
Pathogen, Atkins Garden and Research Laboratory, Cuba, p. 63. 1955.
Host(s) and Distribution: Musa paradisiaca, Musa sp.; Brazil, Cuba, Philippines.
Notes: Part of the C. apii sensu lato species complex.
Cercospora musae var. paradisiaca Bat. et R. Garnier, Revista Agric. (Recife), 3:54.
1963.
Host(s) and Distribution: Musa paradisiaca; Brazil.
Notes: Status unknown, has not been treated.
Cercospora pingtungensis T.Y. Lin et J.M. Yen, Bull. Soc. Mycol. France 87:427. (1971)
1972.
Host(s) and Distribution: Musa acuminata, M. cavendishii; China, Taiwan.
Notes: Conidia pigmented with thickened hila, not a Cercospora.
Cladosporium bisporum, Matsush., Icones microfungorum a Matsushima lectorum
(Kobe):33. 1975.
Host(s) and Distribution: Musa paradisiaca; Japan.
Cladosporium leaf speckle
Cladosporium musae E.W. Mason, Mycol. Pap. 13:2. 1945.
Host(s) and Distribution: Musa paradisiaca, M. textiles, Musa spp.; Bangladesh,
Brunei, Burundi, Cameroon, Congo, Côte d’Ivoire, Cuba, Ecuador, Egypt, Ethiopia,
France, Ghana, Ghana, Guinea, Guinea, Honduras, Honduras, Hong Kong, Indonesia,
Jamaica, Malaysia, Nepal, Papua New Guinea, Rwanda, Sierra Leone, Solomon Islands,
South Africa, Sri Lanka, Sudan, Uganda, Thailand, Togo, Uganda, Vietnam, Western
Samoa, Zimbabwe.
Cladosporium oxysporum Berk. et M.A. Curtis, J. Linn. Soc. Bot. 10:362. 1868.
Host(s) and Distribution: Musa paradisiaca; Brazil, Venezuela.
Mycosphaerella formosana T.Y. Lin et J.M. Yen, Rev. Mycol. 35:323. 1971.
Host(s) and Distribution: Musa sp.; Taiwan.
Mycosphaerella henriquesiana G. Winter, Bol. Soc. Brot. 4:13. 1886.
Host(s) and Distribution: Musa sp.; Africa.
53
MyLsd 17x24
25/06/03
15:15
Page 54
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Mycosphaerella liukiuensis Sawada, Special Publ. Coll. Agric. Natl. Taiwan Univ.
63. 1959.
Host(s) and Distribution: Musa formosana (?), M. liukiuensis; Taiwan.
Mycosphaerella speckle
Mycosphaerella musae (Speg.) Syd. et P. Syd., Phillipp. J. Sci. 8:482. 1913.
≡ Sphaerella musae Speg., Anal. Mus. Nac. Hist. Nat. Buenos Aires 19:354. 1909.
= Sphaerella musae Sacc., Atti Accad. Sci. Veneto-Trentino-Instriana, Ser. 3, 10:67.
1917, homonym.
Host(s) and Distribution: Musa acuminata, M. banksii, M. cavendishii, M. coccinea,
M. paradisiaca, M. paradisiaca, M. textilis, M. uranoscopos, Musa sp.; Argentina,
Australia, Malaysia, Philippines, Puerto Rico, USA (HI), Virgin Islands.
Pseudocercospora fengshanensis (T.Y. Lin & J.M. Yen) J.M. Yen & S.K. Sun,
Cryptogam. Mycol. 4:197. 1983.
≡ Cercospora fengshanensis T.Y. Lin & J.M. Yen, Rev. Mycol. 35:320. (1970) 1971.
Host(s) and Distribution: Musa acuminata; China, Taiwan.
Pseudocercospora musae-sapienti (A.K. Kar & M. Mandal) U. Braun & Mouch.,
New Zealand J. Bot. 37:317. 1999.
≡ Cercospora musae-sapienti A.K. Kar & M. Mandal, Norweg. J. Bot. 22:105. 1975.
Host(s) and Distribution: Musa paradisiaca; India, Wallis.
Pseudocercospora musicola U. Braun, New Zealand J. Bot. 37:317. 1999.
≡ Cercospora musicola Sawada (musaecola), Rep. Gov. Agric. Res. Inst. Taiwan 85:116.
1943 (nom. inval.).
≡ Cercospora musicola Goh & W.H. Hsieh, Cercospora and similar fungi from Taiwan
(1990, p. 242). (nom. inval.).
= Cercospora musae-liukiuensis Sawada, Special Publ. Coll. Agric. Natl. Taiwan Univ.
8:221. 1959. (nom. nud.).
Host(s) and Distribution: Musa acuminata, M. basjoo, M. liukiuensis,
M. paradisiaca; China, Japan, Taiwan.
Excluded from Mycosphaerella
Deightoniella leaf spot
Teleomorph state unknown (presumed not Mycosphaerella)
Anamorph: Deightoniella torulosa (Syd.) M.B. Ellis, Mycol. Pap. 66:7 (1957).
≡ Brachysporium torulosum Syd., Hedwigia 49: 83 (1909).
= Cercospora musarum S.F. Ashby, Bull. Dept. Agric. Jamaica N.S. 2:109. 1913.
Host(s) and Distribution: Musa paradisiaca, Musa spp.; Australia, Egypt, Jamaica,
Malaysia, Sierra Leone, Somalia, Taiwan, Thailand, Vietnam.
54
MyLsd 17x24
25/06/03
15:15
Page 55
Session 1
P.W. Crous et al.
Current and future research prospects
1. Given the obvious problems surrounding Mycosphaerella research discussed
above, it is imperative that all researchers agree that they are working with the
same disease. This needs to be firmly established by means of molecular techniques.
Primers have been developed to identify M. musicola and M. fijiensis (Johanson
et al., 1994) but little is known about similar, closely related species, and whether
they could be separated using these primers which are based on ITS sequence data.
Although ITS has thus far proved to be very valuable in Mycosphaerella systematics,
additional genes also need to be sequenced, as ITS alone indicates similarity, not
necessarily conspecificity.
2. The data presently available in GenBank for species occurring on Musa indicate
some variation within well-known taxa. This could either be due to sequencing
errors, intraspecific variation, or to different researchers working with different
species. To standardize the taxonomy and pathology research being conducted
on these organisms, we need to agree on what they are. This can only be achieved
by designating epitypes of the various species following a thorough taxonomic
study. These cultures should then be lodged in culture collections and be readily
available to those wishing to study the organisms occurring on Musa. In all these
instances, it is best for mycologists to derive the cultures from fresh specimens,
so that the identity can be confirmed on the host material, and once again in
culture.
3. To fully understand species within Mycosphaerella, we need to collect both
species and populations. We need to address questions relating to host specificity,
speciation and, in respect to plant pathology, epidemiology, fungicide sensitivity,
the importance of the different morphs and mechanisms of variation and
dispersal. We also need to learn how the species are migrating around the world.
To address these questions, we need to develop the correct molecular markers that
can be used to for investigations at a species and a population level. Once again,
these populations need to be deposited in reputable culture collections (i.e. CBS,
IMI or ATCC) so that they can be studied and re-studied in years to come.
References
Arx J.A. von 1949. Beitrage zur Kenntnis der Gattung Mycosphaerella. Sydowia 3:28–100.
Arx J.A. von 1983. Mycosphaerella and its anamorphs. Proc. K. Nederl. Akad. Wet. Ser. C
86:15–54.
Barr M.E. 1972. Preliminary studies on the Dothideales in temperate North America. Contrib.
Univ. Michigan Herb. 9:523–638.
Braun U. 1995. A monograph of Cercosporella, Ramularia and allied genera (Phytopathogenic
Hyphomycetes). Vol. 1. IHW-Verlag, Eching, Germany.
Braun U. 1998. A monograph of Cercosporella, Ramularia and allied genera (Phytopathogenic
Hyphomycetes). Vol. 2. IHW-Verlag, Eching, Germany.
55
MyLsd 17x24
25/06/03
15:15
Page 56
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Carlier J., E. Fouré, F. Gauhl, D.R. Jones, P. Lepoivre, X. Mourichon, Pasberg-Gauhl C. and
R.A. Romero. 2000a. Fungal diseases of the foliage. Pp. 37–141 in Diseases of Banana,
Abacá and enset. (D.R. Jones, ed.). CAB International, Wallingford, UK.
Carlier J., M.F. Zapater, F. Lapeyre, D.R. Jones and X. Mourichon. 2000b. Septoria leaf spot
of banana: a newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90:884-890.
Chupp C. 1954. A monograph of the fungus genus Cercospora. Ithaca, New York. Published
by the author.
Corlett M. 1991. An annotated list of the published names in Mycosphaerella and Sphaerella.
Mycol. Mem. 18:1–328.
Corlett M. 1995. An annotated list of the published names in Mycosphaerella and Sphaerella:
corrections and additions. Mycotaxon 53:37–56.
Crous P.W. 1998. Mycosphaerella spp. and their anamorphs associated with leaf spot diseases
of Eucalyptus. Mycol. Mem. 21:1–170.
Crous P.W., A. Aptroot, J.C. Kang, U. Braun and M.J. Wingfield. 2000. The genus
Mycosphaerella and its anamorphs. in Molecules, morphology and classification: towards
monophyletic genera in the Ascomycetes. (K.A. Seifert, W. Gams, P.W. Crous and G.J.
Samuels, eds). Stud. Mycol. 45:107-121.
Crous P.W., J.C. Kang, and U. Braun. 2001. A phylogenetic redefinition of anamorph genera
in Mycosphaerella based on ITS rDNA sequence and morphology. Mycologia 93:1081–1101.
Crous P.W. and X. Mourichon. 2002. Mycosphaerella eumusae and its anamorph
Pseudocercospora eumusae spp. nov.: causal agent of eumusae leaf spot disease of banana.
Sydowia 54:35–43.
David J.C. 1993. A revision of taxa referred to Heterosporium Klotzsch ex Cooke (Mitosporic
Fungi). PhD Dissertation, University of Reading, UK.
David J.C. 1997. A contribution to the systematics of Cladosporium: revision of fungi
previously referred to Heterosporium. Mycol. Pap. 172:1–157.
Deighton F.C. 1973. Studies on Cercospora and allied genera. IV. Cercosporella Sacc.,
Pseudocercosporella gen. nov. and Pseudocercosporidium gen. nov. Mycol. Pap. 133:1–62.
Deighton F.C. 1976. Studies on Cercospora and allied genera. VI. Pseudocercospora Speg.,
Pantospora Cif. and Cercoseptoria Petr. Mycol. Pap. 140:1–168.
Deighton F.C. 1979. Studies on Cercospora and allied genera. VII. New species and
redispositions. Mycol. Pap. 144:1–56.
Deighton F.C. 1987. New species of Pseudocercospora and Mycovellosiella, and new
combinations into Pseudocercospora and Phaeoramularia. Trans. Brit. Mycol. Soc.
88:365–391.
Deighton F.C. 1990. Observations on Phaeoisariopsis. Mycol. Res. 94:1096-1102.
Johanson A., R.N. Crowhurst, E.H.A. Rikkerink, R.A. Fullerton and M.D. Templeton. 1994.
The use of species-specific DNA probes for the identification of Mycosphaerella fijiensis
and M. musicola, the causal agents of Sigatoka disease of banana. Plant Pathol.
44:701–707.
Nag Raj T.R. 1993. Coelomycetous anamorphs with appendage-bearing conidia. Mycologue
Publications, Waterloo, ON, Canada.
Park R.F. and P.J. Keane. 1984. Further Mycosphaerella species causing leaf diseases of
Eucalyptus. Trans. Brit. Mycol. Soc. 83:93–105.
Pollack F.G. 1987. An annotated compilation of Cercospora names. Mycol. Mem. 12:1–212.
Stewart E.L., Z. Liu, P.W. Crous and L. Szabo. 1999. Phylogenetic relationships among some
cercosporoid anamorphs of Mycosphaerella based on rDNA sequence analysis. Mycol. Res.
103:1491–1499.
56
MyLsd 17x24
25/06/03
15:15
Page 57
Session 1
P.W. Crous et al.
Stover R.H. 1994. Mycosphaerella musae and Cercospora ‘non-virulentum’ from Sigatoka leaf
spots are identical. Fruits 49:187–190.
Sutton B.C. 1980. The Coelomycetes. Commonwealth Mycological Institute, Kew, England.
Sutton B.C. and G.L. Hennebert. 1994. Interconnections amongst anamorphs and their possible
contribution to Ascomycete systematics. Pp. 77–98. in Ascomycete Systematics. Problems
and perspectives in the nineties. (D.L. Hawksworth, ed.). Plenum Press, New York.
Sutton B.C., S.F. Shamoun and P.W. Crous. 1996. Two leaf pathogens of Ribes spp. in North
America, Quasiphloeospora saximontanensis (Deighton) comb. nov. and Phloeosporella
ribis (J.J. Davis) comb. nov. Mycol. Res. 100:979–983.
Verkley G.J.M. 1997. Ultrastructural evidence for two types of proliferation in a single
conidiogenous cell of Septoria chrysanthemella. Mycol. Res. 102:368-372.
57
MyLsd 17x24
25/06/03
15:15
Page 59
Session 1
J. Henderson et al.
Improved PCR-based detection of
Sigatoka disease and black leaf streak
disease in Australian banana crops
J. Henderson1, K. Grice2, J. Pattemore1, R. Peterson2 and E. Aitken1
Abstract
Accurate diagnosis of black leaf streak disease is often complicated by the presence of other fungal
pathogens and in particular by the morphological similarity of the related species Mycosphaerella
musicola, the causal agent of Sigatoka disease. In addition, high rainfall often washes away fungal
structures making microscopic identification difficult. Starting in 1998, the Queensland
Department of Primary Industries has been using molecular methods to help diagnose black leaf
streak disease. A polymerase chain reaction (PCR) assay was used, but the protocol was found
to lack specificity when applied to Australian isolates of the fungi. In July 2000, a project aimed
at improving the sensitivity and specificity of the PCR as well as streamlining the assay was
initiated. Various components of the PCR test were targeted for improvement. Homogenization
of banana leaf tissue has eliminated possible cross-contamination while tripling batch
throughput. An improved DNA extraction method produces cleaner DNA in less than half the
time of the prior extraction method. Flexibility and sensitivity of the PCR has been improved by
introducing a new enzyme while the new format PCR thermal cyclers have increased sample
throughput. Importantly, specificity has been enhanced with the design of new diagnostic primers.
These changes produce a definitive result during the first PCR in more than 98% of samples while
increasing daily throughput more than eight-fold.
Resumen - Mejoramiento de la detección de la Sigatoka negra y Sigatoka amarilla
basada en PCR en los cultivos bananeros de Australia
A menudo el diagnóstico preciso de la Sigatoka negra es complicado debido a la presencia de
otros patógenos fungosos y en particular por la similitud morfológica de la especie relacionada
Mycosphaerella musicola, agente causal de la Sigatoka amarilla. En adición, fuertes precipitaciones
a menudo se llevan las estructuras fungosas dificultando la identificación microscópica. El QDPI
ha estado utilizando métodos moleculares para confirmar el diagnóstico de la Sigatoka negra
desde 1998. Se utilizó un ensayo de la reacción en cadena de polimerasa (PCR), sin embargo se
descubrió que al protocolo le faltaba la especificidad al aplicarlo a los aislados australianos de
los hongos. En julio de 2000, se inició un proyecto dirigido a mejorar la sensibilidad y especificidad
del PCR así como la modernización del ensayo. Se seleccionaron varios componentes del examen
1
2
Cooperative Research Centre for Tropical Plant Protection, Indooroopilly, Australia
Queensland Department of Primary Industries, Mareeba, Australia
59
MyLsd 17x24
25/06/03
15:15
Page 60
Mycosphaerella leaf spot diseases of bananas: present status and outlook
PCR para ser mejorados. La homogenización del tejido foliar del banano con la ayuda de una
micromano de mortero plástica dentro del tubo ha eliminado una posible contaminación
cruzada al triplicar el rendimiento de los lotes. Un método mejorado de extracción de ADN produce
un ADN más limpio en menos de la mitad del tiempo, que el método de extracción anterior. La
flexibilidad y sensibilidad de PCR fueron mejoradas introduciendo una nueva enzima, mientras
que nuevos variadores térmicos para el formateo de PCR han aumentado el rendimiento de las
muestras. La especificidad de PCR ha sido mejorada a través del diseño de nuevos iniciadores de
diagnóstico. Combinadas, estas mejoras producen un resultado definitivo durante el primer ensayo
de PCR en más del 98% de las muestras, mientras que el rendimiento diario de la muestra es
8 veces mayor.
Résumé - Détection améliorée basée sur la PCR de la maladie de Sigatoka et de la
maladie des raies noires dans les plantations de bananes en Australie
Le diagnostic exact de la maladie des raies noires est souvent rendu plus difficile par la présence
d’autres pathogènes fongiques, et en particulier par la similarité morphologique d’une espèce
voisine Mycosphaerella musicola, l’agent causal de la maladie de Sigatoka. De plus, de fortes
précipitations éliminent souvent des structures fongiques ce qui rend l’identification au
microscope difficile. En 1998, le Queensland Department of Primary Industries (QDPI) a commencé
à utiliser des méthodes moléculaires afin de mieux identifier la maladie des raies noires. Un essai
basé sur la réaction en chaîne par polymérase (PCR, polymerase chain reaction) a été utilisé mais
ce protocole était peu spécifique quant aux isolats australiens du champignon. En juillet 2000,
une étude a été initiée afin d’augmenter la sensibilité et la spécificité de la PCR et pour
rationaliser le protocole. Divers composants du test PCR ont été ciblés afin d’être améliorés.
L’homogénéisation des tissus de la feuille de banane a permis d’éliminer les contaminations
extérieures tout en triplant le débit. Une méthode améliorée d’extraction de l’ADN permet
d’obtenir un ADN plus pur en deux fois moins de temps. La flexibilité et la sensibilité de la PCR
ont été améliorées grâce à l’utilisation d’une nouvelle enzyme. De plus, le nouveau format des
thermocycleurs PCR a permis d’accroître le débit. Il est important de noter que la spécificité a
été mise en valeur par la conception de nouvelles amorces diagnostiques. Ces changements
produisent un résultat définitif dans la première PCR dans plus de 98% des cas et multiplient
par huit le nombre d’échantillons traités par jour.
The Tully 2001 black leaf streak disease outbreak
The value of the Australian banana industry is estimated to be A$357 million (US$193
million) per year. In 2000, nearly 250 000 tonnes of bananas were produced by 100
growers in Australia. All Australian bananas are produced for consumption locally
and 85% are of the ‘Cavendish’ variety. The majority of bananas are grown in
northern Queensland, with 67% of the crop concentrated in Tully, Cairns and Innisfail
(Figure 1).
In April 2001, the Australian banana industry suffered a potentially devastating
outbreak of black leaf streak disease (caused by Mycosphaerella fijiensis) in Tully
(Figure 1). This is the pathogen’s first incursion in a major commercial region in
Australia; failure to control the pathogen would have far reaching effects on the
industry.
The Queensland Department of Primary Industries (QDPI) has had considerable
success eradicating previous outbreaks of black leaf streak disease by plant
destruction and replacement with resistant varieties. Since the initial discovery of
black leaf streak disease in 1981 at Bamaga, an Aboriginal community located 40km
60
MyLsd 17x24
25/06/03
15:15
Page 61
Session 1
J. Henderson et al.
from the tip of the Cape York Peninsula, black leaf streak disease has been detected
and eradicated eight times in far north Queensland. This ninth outbreak was in Tully
where crops are estimated to be worth A$119 million per year (US$64 million).
Figure 1. Map of Queensland (Australia)
and inset showing location of major
commercial banana growing region
and Tully, the site of the 2001 black leaf
streak disease outbreak.
Diagnosis of black leaf streak disease in Australia
Banana crops are routinely surveyed for black leaf streak disease by QDPI scientists
at the Centre for Tropical Agriculture, Mareeba. Accurate diagnosis of black leaf streak
disease is complicated by the morphological similarity of M. fijiensis to a related species
M. musicola, which causes Sigatoka disease. Usually, experienced plant pathologists
distinguish the two diseases by the development of symptoms and microscopical
features of the fungi. In Tully, conidia were absent because of prolonged rainfall, and
identification of morphological characters was not possible. Therefore, molecular
methods were used for diagnosis.
The QDPI has used the polymerase chain reaction (PCR) to confirm diagnoses of
Mycosphaerella leaf spot diseases since 1998 (Johanson, 1997). Approximately 10%
of laboratory samples required confirmation by PCR. However, the method was slow
and lacked specificity to some Australian and Torres Strait Island isolates of the fungi.
The lack of specificity was possibly due to the high variability among the Southeast
Asian populations of the pathogen. Populations of M. fijiensis from the Torres Strait
were found to differ from those of the Pacific (Hayden, 2001). However, the Torres
Strait populations and Pacific populations were found to be related to those from Papua
New Guinea where there is a considerable diversity (Carlier, pers. comm.). In the original
61
MyLsd 17x24
25/06/03
15:15
Page 62
Mycosphaerella leaf spot diseases of bananas: present status and outlook
study by Johanson et al. (1994) isolates from the Torres Strait were not included and
it is possible that these isolates could be the source of the variability not detected by
the original PCR primers.
In July 2000, a project between QDPI and the Cooperative Research Centre for
Tropical Plant Protection (CRCTPP) was initiated with the aim of improving the
specificity of the diagnostic procedure and increasing throughput in readiness for
outbreaks of the disease1. Aspects of the PCR diagnostic procedure that were targeted
for improvement included sample excision, homogenisation of banana leaf tissue, DNA
extraction, PCR protocol, PCR primer design and equipment.
Flame sterilised cork-borers 4 mm in diameter have replaced scalpels for the removal
of suspect lesions from banana leaves. The method is quick and simple and there is
no cross-contamination between samples. Plastic micropestles have replaced ceramic
mortar and pestles for homogenising leaf tissue. Micropestles have reduced the potential
for cross-contamination between samples, eliminated transfer from mortar to tube, and
have tripled the throughput. The rapid cetyltrimethyl ammonium bromide (CTAB) DNA
extraction method (Stewart and Via, 1993) was adopted. It produces cleaner DNA in
half the time.
New PCR primer sequences specific to M. musicola and M. fijiensis, and a
modification of published ribosomal gene primer sequences (White et al., 1990) to
improve specificity to increase duplex stability of the primers with the target DNA
(Rychlik, 1993), has improved specificity of the PCR assay. A size difference was also
included in the PCR assay with the specific primer for M. fijiensis designed to the ITS1
and the specific primer for M. musicola designed to the ITS2 (Figure 2). Flexibility
and sensitivity was improved by introducing the hot-start enzyme, TaqGold DNA
polymerase (PE Biosystems). TaqGold DNA polymerase requires heat-activation before
amplification can proceed. Therefore, non-specific amplification products are reduced
and reactions can be left at 4°C until the addition of template. New equipment at the
Centre for Tropical Agriculture has also improved throughput of assays. New format
PCR thermal cyclers have increased tube capacity from 30 to 192 per run, and new
electrophoresis equipment can analyse 52 samples for Sigatoka disease and black leaf
streak disease at the same time.
The new methods produce a high quality DNA preparation and provide a definitive
result during the first PCR in more than 98% samples. In addition, extraction time is
more than halved and daily throughput increased by more than eight-fold.
Application of new molecular test
Use of the new methods in April 2001 coincided with Australia’s most severe
outbreak of black leaf streak disease. This was the first outbreak in a commercial
growing area; previous outbreaks had been further north and in places where
containment was easy. Fungal structures were absent on banana samples because
of high rainfall at Tully. Therefore, diagnosis of up to 50% of samples was confirmed
using the PCR assay. The PCR assay provided the Australian government and
1
The project is co-managed by Ron Peterson (Principal Plant Pathologist, Mareeba QDPI) and Juliane Henderson (Research
Officer, CRCTPP).
62
MyLsd 17x24
25/06/03
15:15
Page 63
Session 1
J. Henderson et al.
Sigatoka disease
MMfor
1 8S (partial)
ITS1
ITS2
5.8S
MFfor
R635
25S (partial)
R635mod
Black leaf streak disease
Figure 2. Location of primers for the diagnostic PCR assay.The black leaf streak disease specific forward primer
(MFfor) is located on the ITS1 region while the Sigatoka disease specific forward primer (Mmfor) is located
on the ITS2 region.
banana industry the confidence to start a A$20 million (US$10.8 million)
surveillance and eradication plan in Tully, and more than 2500 PCR tests have
been done.
The CRCTPP and QDPI continue to monitor and improve the PCR diagnostic.
Thus, scientists from the CRCTPP improved homogenization of banana tissue by
the use of glass beads shaken at high speeds. The commercially available
“FastPrep” Instrument (Q-Biogene) processes 12 samples in 45 seconds and
eliminates cross-contamination between samples by single-use O-ring tubes. Use
of the method at the Centre for Tropical Agriculture is dependant on funds.
Opportunities to automate and improve specificity and sensitivity of the assay
are being studied as part of the CRCTPP’s plan to use new technology. Development
of a real-time PCR assay to detect and differentiate M. fijiensis, M. musicola and
M. eumusae is in progress. A fluorescent PCR format increases sensitivity and
specificity, reduces cross-contamination, and increases throughput because postPCR processing is not required.
To ensure the robustness of the PCR diagnostic and to facilitate development
of new diagnostic assays for Mycosphaerella leaf spot diseases in Australia, the
sequence variability in Australasian isolates of M. musicola and M. fijiensis will
be studied2. First, the region incorporating the ITS1, 5.8S ribosomal gene and ITS2
will be cloned and sequenced from Mycosphaerella isolates from Australia, the
Torres Strait Islands and Fiji. In collaboration with other groups studying sequences
pertaining to the disease, we will compare our database with overseas isolates.
If further sequence information is required, other conserved fungal genes, e.g.
ß-tubulin, histone-4, glyceraldehyde-3-phosphate, will be investigated. The
information from this study will help us to understand how the 2001 Australian
outbreak arose, e.g. whether from one or several sources.
2
Drs Elizabeth Aitken (Department of Botany, University of Queensland) and Juliane Henderson will use joint funding from
the Australian Banana Industry Protection Board (BIPB) and Horticulture Australia Limited (HAL) to investigate sequence
variability of Mycosphaerella causing leaf spot diseases on banana.
63
MyLsd 17x24
25/06/03
15:15
Page 64
Mycosphaerella leaf spot diseases of bananas: present status and outlook
The status of black leaf streak disease in Australia is yet to be confirmed and
the future application of diagnostic tests is uncertain. The method could be used
to maintain Australia’s disease-free status, as far as black leaf streak disease is
concerned, or to monitor pathogen populations for control measures should the
disease become endemic. Either would ensure that the best diagnostic assay is
available to the Australian banana industry.
References
Hayden H. 2001. Genetic variability in populations of pathogens causing black and yellow
Sigatoka diseases of bananas. PhD Thesis, University of Queensland, Australia.
Johanson A. 1997. Detection of Sigatoka Leaf Spot Pathogens of Banana by the Polymerase
Chain Reaction. Natural Resources Institute, Chatham, UK.
Rychlik W. 1993. Selection of primers for polymerase chain reaction. Pp. 31-40 in Methods
in Molecular Biology, Vol. 15: PCR Protocols: Current Methods and Applications (B.A.
White, ed.). Humana Press Inc, Totowa, New Jersey.
Stewart C.N. and L.E. Via. 1993. A rapid CTAB DNA isolation technique useful for RAPD
fingerprinting and other PCR applications. BioTechniques 14(5):748-750.
White T.J., T. Bruns, S. Lee and J.W. Taylor. 1990. Amplification and direct sequencing of
fungal ribosomal RNA genes for phylogenetics. Pp. 315-322 in PCR Protocols: A Guide
to Methods and Applications (M.A. Innis, D.H. Gelfand, J.J. Sninsky and T.J. White, eds.).
Academic Press Inc., San Diego, USA.
64
MyLsd 17x24
25/06/03
15:15
Page 65
Session 1
A. Viljoen et al.
Impact of minor Mycosphaerella
pathogens on bananas (Musa) in
South Africa
A. Viljoen1, A.K.J. Surridge1 and P.W. Crous2
Abstract
Of the species of Mycosphaerella known to occur on bananas, only M. musicola and M. musae
occur in South Africa. Since both species are less damaging than M. fijiensis and M. eumusae, they
are considered minor Mycosphaerella pathogens of this host. However, both M. musicola and
M. musae can cause significant damage to bananas in the subtropics. For several years, M. musicola
seemed to be the dominant pathogen of banana leaves in South Africa. It was very severe in
banana plantations in Southern KwaZulu-Natal in the early 1990s, and caused losses of up to
50% in Cavendish bananas due to early ripening and lower yields in the Komatipoort area in 1999
and 2000. A highly coordinated disease management programme involving severe deleafing and
fungicidal sprays has reduced the impact of Sigatoka disease in the country since 2001. However,
Mycosphaerella speckle now appears to have replaced Sigatoka disease as the dominant leaf
pathogen in all banana growing areas of South Africa. Management strategies for Sigatoka disease
seem to be less effective against Mycosphaerella speckle. Although this fungus primarily affects
older leaves, the disease has become very severe in southern KwaZulu-Natal during 2002. Its
economic impact and epidemiology, however, still have to be determined.
Resumen - Impacto de los patógenos de Mycosphaerella de menor importancia sobre
los bananos (Musa) en Africa del Sur
De las varias especies de Mycosphaerella que ocurren en los bananos, solo M. musicola y M. musae,
ocurren en Africa del Sur. Ya que ambas especies causan menores daños que M. fijiensis y
M. eumusae, ellas se consideran patógenos de Mycosphaerella de menor importancia en este
hospedante. Sin embargo, tanto M. musicola como M. musae pueden causar daños significativos
a los bananos en los subtrópicos. Durante varios años, M. musicola se consideró el patógeno
dominante en las hojas de los bananos en Africa del Sur. A principios de la década de los 90, este
patógeno afectó severamente las plantaciones bananeras en el sur de KwaZulu-Natal, y causó
pérdidas de hasta 50% en los bananos Cavendish, debido a una maduración precoz y bajos
rendimientos en el área de Komatipoort en 1999 y 2000. Un programa coordinado de manejo
de la enfermedad, que incluyó deshoje y rociados de funguicidas redujo el impacto de la
1
2
University of Pretoria, Pretoria, South Africa
University of Stellenbosch, Matieland, South Africa
65
MyLsd 17x24
25/06/03
15:15
Page 66
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Sigatoka amarilla en el país desde 2001. Sin embargo, la mancha causada por Mycosphaerella
parece haber reemplazado actualmente la Sigatoka amarilla como patógeno foliar dominante
en todas las zonas productoras de banano en Africa del Sur. Parece que las estrategias de manejo
de la Sigatoka amarilla son menos eficaces contra la mancha causada por Mycosphaerella. Aunque
este hongo afecta principalmente las hojas viejas, la enfermedad se hizo muy severa en el sur
de KwaZulu-Natal durante el año 2002. No obstante, aún falta determinar su impacto económico
y la epidemiología.
Résumé - Impact des pathogènes mineurs de Mycosphaerella sur les bananiers (Musa)
en Afrique du Sud
De toutes les espèces connues de Mycosphaerella affectant les bananiers, seules M. musicola
et M. musae se trouvent en Afrique du Sud. Vu que ces deux espèces provoquent moins de dégâts
que M. fijiensis et M. eumusae, elles sont considérées comme étant des pathogènes mineurs de
cet hôte. Toutefois, M. musicola et M. musae peuvent toutes deux provoquer des dégâts
significatifs aux cultures de bananes dans la zone subtropicale. Pendant plusieurs années, il
semblait que M. musicola était le pathogène dominant des feuilles de bananiers en Afrique du
Sud. L’infection était même très grave dans les plantations du sud du KwaZulu-Natal au début
des années 1990 et en 1999 et 2000 a provoqué chez les bananiers Cavendish de la région de
Komatipoort des pertes pouvant aller jusqu’à 50% dues à un mûrissement prématuré des fruits
et à des rendements réduits. Un programme de gestion de la maladie parfaitement coordonné
impliquant un défeuillage massif ainsi que des traitements fongicides a réduit l’impact de la
maladie de Sigatoka dans le pays depuis 2001. Toutefois, Mycosphaerella speckle semble
maintenant avoir remplacé la maladie de Sigatoka et se trouve être le pathogène dominant dans
les régions de culture de la banane en Afrique du Sud. Les stratégies de gestion de la maladie
de Sigatoka semblent moins efficaces envers le Mycosphaerella speckle. Bien que ce pathogène
affecte en premier les feuilles les plus âgées, la maladie est devenue très grave dans le sud du
KwaZulu-Natal en 2002. Son impact économique ainsi que son épidémiologie restent encore à
être déterminés.
Introduction
Fungi that cause disease on leaves of banana and plantain include Mycosphaerella
fijiensis M. Morelet (the causal agent of black leaf streak disease), M. musicola
Leach ex J.L. Mulder (the causal agent of Sigatoka disease), M. eumusae Crous et
X. Mourichon (the causal agent of eumusae leaf spot disease) and M. musae (Speg.)
Syd. et P. Syd. (the causal agent of Mycosphaerella speckle). M. fijiensis is the most
virulent and economically important of the four Mycosphaerella spp. Since its
discovery in Fiji in 1963, M. fijiensis has replaced M. musicola as the main leaf
pathogen in all tropical countries that produce banana (Jones, 2000). However,
Sigatoka disease is still the main leaf disease in the subtropics and at higher altitudes
in the tropics. In 1995, a new disease, eumusae leaf spot, was reported on Musa
(Carlier et al., 2000). Eumusae leaf spot has only been found in Southeast Asia and
in Nigeria, West Africa, (Jones, these proceedings) but is very damaging there. Black
leaf streak disease, Sigatoka disease and eumusae leaf spot disease comprise the
Mycosphaerella leaf spot disease complex on banana. Mycosphaerella speckle is not
considered to be important on banana. It is limited to the subtropics and is severe
only on Cavendish bananas in Australia (Jones, 2000).
The dominant Mycosphaerella spp. pathogens that occur in the subtropics are
M. musicola and M. musae. Since neither consistently damage banana leaves, they
66
MyLsd 17x24
25/06/03
15:15
Page 67
Session 1
A. Viljoen et al.
are considered minor pathogens. The objective of this manuscript is to report the
impact of such minor Mycosphaerella pathogens on bananas in South Africa.
Banana production in South Africa
Bananas are produced in six areas in South Africa (Figure 1). The crop was
introduced from India at the beginning of the 19th century. Production began along
the southern and northern sections of the KwaZulu-Natal (KZN) coast, then
introduced in the former Transvaal province and planted in Kiepersol, Tzaneen
and Levubu. The largest production area, the Onderberg (4600 ha), became
important in the 1990s. The total area of commercial banana production is
12 500 ha, and is with Cavendish cultivars only. Almost 90% of new plantings
are from tissue culture, and transport of banana plants between production areas
is strictly controlled. All bananas are consumed locally, but there is a possibility
of export to the Middle East.
Leaf diseases of banana in South Africa
Since 1999, regular surveys of areas where banana is cultivated have shown that
Sigatoka disease, Mycosphaerella speckle and Cordana leaf spot (caused by
Cordana musae [Zimm.] Höhn.) are present in all production areas. Cladosporium
speckle (Cladosporium musae E.W. Mason) was found only in Levubu. Sigatoka
disease and Mycosphaerella speckle were the most important.
The banana leaf diseases in the southern part of Africa have not been studied
very much. Black leaf streak disease is present in most tropical African countries
(Jones, 2000), and has been reported as far south as northern Malawi (Ploetz,
1992). However, black leaf streak disease is not known in Zimbabwe (which borders
the Levubu area) and Mozambique (which borders the Onderberg area) (Figure 1).
Figure 1.
The six banana production areas
of South Africa: Levubu, Letaba
(Tzaneen), Hazyview (Kiepersol),
Onderberg, and northern and
southern KwaZulu-Natal.
67
MyLsd 17x24
25/06/03
15:15
Page 68
Mycosphaerella leaf spot diseases of bananas: present status and outlook
A severe outbreak of Sigatoka disease in 1999-2000, prompted an investigation
into the identity of the causal agent. Samples were collected in all production areas,
and the fungi identified using morphological and molecular techniques (Surridge et
al., these proceedings). PCR primers developed by Johansen and Jeger (1994) were used
to distinguish between Sigatoka disease and black leaf streak disease. Isolates from
single conidia were further sequenced (ITS region) and compared with sequence data
of M. musicola, M. fijiensis and M. eumusae from GenBank. All local isolates proved
to be M. musicola, which causes Sigatoka disease. There was no evidence of
M. fijiensis and M. eumusae, suggesting they have not been introduced in South Africa.
The severity of the outbreaks was attributed to favourable weather conditions and
increases in the amount of inoculum.
The life cycles of Mycosphaerella leaf spot diseases have a sexual (teleomorph) stage,
which produces ascospores, and an asexual (anamorph) stage, which produces conidia
(Jones, 2000). Conidia are the main spore produced by M. musicola (Meredith, 1970).
Conidia are dispersed within the leaf canopy and to neighbouring plants by rain, which
dislodges and washes them onto adjacent leaves. Ascospores are forcefully discharged
and spread by wind currents over bigger distances than conidia. Both types of spore
require moisture for production, release, infection, growth and sporulation. Most stages
in the life cycle take place over a wide range of temperatures; however, minimum night
temperatures of 18oC and 21oC are needed for the production of conidia and
ascospores of M. musicola, respectively (Meredith, 1970). Conidia are produced on both
leaf surfaces, while ascospore production is almost three times greater on the upper
(adaxial) than lower (abaxial) leaf surface (Meredith, 1970).
Climatic conditions in South Africa
The banana production areas of South Africa are located in the east between
25o and 30o latitude and 30o and 32o longitude. The areas have a subtropical
climate with cool, dry winters and warm, wet summers. Rainfall and temperature
data for the Onderberg over a period of 10 years showed that November and
March were the most favourable months for infection and disease development
(Figure 2). During this time minimum night temperatures exceed 18oC, which is
necessary for the production of conidia. Minimum night temperatures exceed 21oC
only in January and February, therefore the period for ascospore production is
short. Disease development is most rapid between November and March but slows
substantially during the cooler months from May through September. Climatic
conditions in South Africa provide ample opportunities for the management of
Sigatoka disease.
The impact of Mycosphaerella diseases
Van den Boom and Kuhne (1969) first reported Sigatoka disease in South Africa, although
the disease was also mentioned in 1954 (Meredith, 1970). The first report of Mycosphaerella speckle in South Africa was in 1973 (Brodrick, 1973). Despite these late reports,
both diseases have been associated with banana leaves for as long as producers can
remember.
68
25/06/03
15:15
Page 69
A. Viljoen et al.
10
40
5
20
0
0
Ave. max. temperature (°C)
be
r
be
cem
De
No
vem
to b
Oc
mb
t
p te
Se
Au
gu
s
Ju l
Ju n
Ap
Ma
Feb
ru a
ua
Jan
Ave. min. temperature (°C)
r
60
er
15
er
80
y
20
e
100
Ma
y
25
ril
120
rch
30
ry
140
ry
35
Rainfall
Session 1
Temperature
MyLsd 17x24
Ave. rainfall (mm)
Figure 2. Average annual temperature range and rainfall in the banana growing areas of South Africa.
Sigatoka disease first became severe in South Africa during the 1960’s (Van den
Boom and Kuhne, 1969). The disease devastated production in southern KZN in the
1990s and in the Onderberg in 2000 (Viljoen, unpublished data). Mycosphaerella musicola
infects the first three leaves of the banana plant. The symptoms first become visible on
the third or fourth leaf (Jones, 2000). Under favourable weather conditions and with
large amounts of inoculum, M. musicola can destroy all leaves after the stage when
bunches are produced. This is what happened during the 1999-2000 outbreaks of
Sigatoka disease in the Onderberg. Damage included smaller fruits, smaller bunches,
and premature fruit ripening in the field and in boxes. Farmers reported losses of up
to 50% of the crop. An extensive disease management programme was implemented
in October 2000 to halt the devastation. All leaves with Sigatoka lesions were cut and
turned over on the plantation floor to limit the release of air-borne ascospores. Many
bunches were sacrificed, in one instance amounting to nearly 18 000 bunches on a
farm of about 40 ha. A fungicidal spray programme with protectant and systemic
fungicides when the rainy season started and night temperatures exceeded 18oC was
recommended to growers. A total of six sprays of systemic fungicide were recommended,
interrupted with a protectant fungicide after every second application of systemic
fungicide (Peterson et al. these proceedings).
None of the farmers applied the recommended number of sprays, and only 2-4 sprays
were applied in total. The cost of fungicides, therefore, was small compared with the
costs of fungicide sprays used to control black leaf streak disease in the tropics. Sigatoka
disease was almost absent from banana fields in 2001, and current control strategies
are now limited to cutting leaves and the application of one or two sprays, dependant
on pre-seasonal leaf spot incidence, per year.
Mycosphaerella musae infects older leaves of banana plants (Jones, 2000).
Mycosphaerella speckle is rarely visible above the fifth fully open leaf, and seldom affects
fruit quality and quantity after bunching. Since 2000, Mycosphaerella speckle has become
69
MyLsd 17x24
25/06/03
15:15
Page 70
Mycosphaerella leaf spot diseases of bananas: present status and outlook
more severe, and now is the main leaf disease of banana in South Africa. The symptoms
are leaf yellowing (chlorosis) and death (necrosis). Necrosis is most visible on the older
leaves, but, in 2002, chlorosis affected leaves as young as the third leaf after bunching
in southern KZN. The effects on yield have not yet been determined. Control strategies
are similar to those for Sigatoka disease, and include removing leaves and applying
fungicides.
Conclusion
Mycosphaerella musicola and M. musae are the only Mycosphaerella leaf pathogens
of banana in South Africa. They are considered to be minor pathogens, but become
damaging under favourable weather conditions and in the presence of large
amounts of inoculum. Subtropical climatic conditions and a clear understanding of
the biology and epidemiology of M. musicola make the management of Sigatoka
disease relatively easy. The increased severity of Mycosphaerella speckle may result
from the management of Sigatoka disease. The quantity of M. musicola ascospores
released into the air is reduced by placing leaves upside down on the ground, but
this probably increases the quantity of M. musae ascospores, which are mainly
released from the lower leaf surface (Jones, 2000). A better understanding of the
biology and epidemiology of M. musae is needed to develop the necessary
management practices for Mycosphaerella speckle in the subtropics.
References
Brodrick H.T. 1973. Spikkelblaar. Banana Series Journal J4:1-2.
Carlier J., M.F. Zapater, F. Lapeyre, D.R. Jones, and X. Mourichon. 2000. Septoria leaf spot
of banana: A newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90(8):884-890.
Johanson A. and M.J. Jeger. 1993. Use of PCR for detection of Mycosphaerella fijiensis and
M. musicola, the causal agents of Sigatoka leaf spots in banana and plantain. Mycological
Research 96(6):670-674.
Jones D.R. 2000. Fungal diseases of the foliage. Pp. 37-141 in Disease of banana, abacá and
enset. (D.R. Jones ed.) CAB International, Wallingford, UK, 544pp.
Meredith D.S. 1970. Banana leaf spot disease (Sigatoka) caused by Mycosphaerella musicola
Leach. Phytopathological Papers no 11, Commonwealth Mycological Institute, Kew, Surrey,
UK. 147pp.
Ploetz R.C. 1992. A current appraisal of banana and plantain diseases in Malawi. Tropical
Pest Management 38:36-42.
Van den Boom T. and F.A. Kuhne. 1969. First report of Sigatoka disease of banana in South
Africa. Citrus Journal 428:17-18.
70
MyLsd 17x24
25/06/03
15:15
Page 71
Session 1
L. Pérez Vicente et al.
Economic impact and management
of black leaf streak disease in Cuba
L. Pérez Vicente, J.M. Alvarez and M. Pérez
Abstract
Black leaf streak disease (caused by Mycosphaerella fijiensis) is the most damaging disease
found in Musa plantations in Cuba. Four years after the appearance of the disease in 1990,
it had replaced Sigatoka disease (caused by M. musicola) in all areas of the country. During
1991 and 1992, plantations of ‘Cavendish’ bananas received 23 fungicide applications per year,
which accounted for 15% of total production costs. A system of integrated control based on
cultural practices, a bioclimatic warning system for timing fungicide application and the use
of systemic fungicides and mineral oil, reduced the number of applications to 13-15 per year
and the cost of control to less than 40%. Black leaf streak disease has seriously affected the
production of susceptible varieties. In 1989, more than 40 000 ha of plantain (Musa cv. AAB)
and 14 000 ha of ‘Cavendish’ (Musa cv. AAA) were treated with fungicide. However, by
the end of 1995 the areas had decreased by 69% and 51% respectively. Since 1994, ‘FHIA-18’,
‘FHIA-03’, ‘FHIA-01-1’, ‘FHIA-02’ and ‘FHIA-21’ with partial resistance to black leaf streak
disease were introduced into Cuba. Currently, there are 10 000 ha planted with these clones
resulting in an 80% reduction in the use of fungicide. The severity of black leaf streak disease
on ‘FHIA-18’ has been inversely correlated with the availability of total K and with the ratio
K/(K+Ca+Mg) in soil and foliage. Variability in the pathogenicity of M. fijiensis populations
has been studied in order to identify possible changes that result from large scale cultivation
of FHIA hybrids with partial resistance.
Resumen - Impacto económico y manejo de la enfermedad de la raya negra en Cuba
La enfermedad de la raya negra o Sigatoka negra, causada por Mycosphaerella fijiensis, es la
enfermedad más nociva presente en las plantaciones de musáceas en Cuba. Cuatro años
después de su aparición en 1990, reemplazó a la Sigatoka amarilla (M. musicola) en todas las
áreas del país. Durante 1991 y 1992 se realizaron hasta 23 aplicaciones de fungicida por año
en plantaciones de banano ‘Cavendish’ con un costo de protección alcanzando el 15% del costo
total de la producción. Un sistema de manejo integrado basado en prácticas culturales y
pronóstico bioclimático de los momentos de tratamiento permitió reducir a 13-15 los
tratamientos con funguicidas sistémicos y aceite mineral al año en las principales plantaciones
de producción con un costo menor de un 40%. La Sigatoka negra ha tenido un serio impacto
en la producción de plátanos susceptibles. En 1989, existían más de 40 000 ha de plátanos
INISAV, Cuba
71
MyLsd 17x24
25/06/03
15:15
Page 72
Mycosphaerella leaf spot diseases of bananas: present status and outlook
(Musa cv. AAB) y 14 000 ha de banana ‘Cavendish’ (Musa cv. AAA) bajo protección fúngica. A
finales de 1995 se habían reducido en un 69 y 51% respectivamente. A partir de 1994, se
introdujeron los clones ‘FHIA-18’,‘FHIA-03’,‘FHIA-01-1’,‘FHIA-02’ y ‘FHIA-21’ con resistencia parcial
a Sigatoka negra. En la actualidad existen 10 000 ha plantadas de estos clones y se ha reducido
el consumo de fungicidas en un 80%. Se ha observado una correlación inversa entre la
severidad de los ataques de Sigatoka negra en el clon FHIA-18 y la disponibilidad total de K
y con la relación K/(K+Ca+Mg) en suelo y hojas. Se desarrollan estudios de la variabilidad
genética de las poblaciones de M. fijiensis con el objetivo de determinar cambios de
patogenicidad a causa del cultivo a gran escala de clones con resistencia parcial.
Résumé - Impact économique et gestion de la maladie des raies noires à Cuba
La maladie des raies noires, causée par Mycosphaerella fijiensis, est la maladie la plus
destructrice des plantations de bananes à Cuba. Quatre années après l’apparition de la
maladie en 1990, elle avait remplacé dans tout le pays la maladie de Sigatoka, causée par
M. musicola. Pendant les années 1991 et 1992, les plantations de bananes ‘Cavendish’ ont reçu
23 traitements de fongicides par an , représentant 15% des coûts totaux de production. Un
système de lutte intégrée basé sur les pratiques culturales, un système de prévision
bioclimatique pour déterminer le moment d’application des fongicides et l’utilisation de
fongicides systémiques et d’huile minérale, ont permis de réduire le nombre de traitements
à 13-15 par an et les coûts de plus de 60%. La maladie des raies noires a sérieusement affecté
la production de variétés sensibles. En 1989, plus de 40 000 ha de plantain (Musa cv. AAB) et
14 000 ha de bananes ‘Cavendish’ (Musa cv. AAA) ont été traités avec des fongicides. Toutefois,
fin 1995, les surfaces traitées ont été réduites respectivement de 69% et 51%. Depuis 1994,‘FHIA18’, ‘FHIA-03’, ‘FHIA-01-1’, ‘FHIA-02’ et ‘FHIA-21’, des hybrides possédant une résistance partielle
à la maladie des raies noires, ont été introduits à Cuba. Actuellement, 10 000 ha sont plantés
avec ces clones ce qui entraîné une baisse de 80% de l’utilisation de fongicides. La gravité
de la maladie des raies noires sur ‘FHIA-18’ a été inversement corrélée avec la disponibilité du
K total et avec le rapport K/(K+Ca+Mg) dans le sol et le feuillage. La variabilité du pouvoir
pathogène des populations de M. fijiensis a été étudiée afin d’identifier des modifications
potentielles qui pourraient découler de la culture à grande échelle d’hybrides FHIA partiellement
résistants.
Introduction
Many fungal, bacterial and viral diseases affect Musa but undoubtedly the one
that has had the most socio-economic impact at a world level has been black leaf
streak disease caused by the fungus Mycosphaerella fijiensis Morelet.
Black leaf streak disease was first reported in Cuba in 1991 (Vidal, 1992).
Previously, the main banana and plantain disease in Cuba was Sigatoka disease
caused by M. musicola Leach ex Mulder, for which a warning system (Ganry and
Meyer, 1972b) for timing the application of fungicides in oil emulsions had been
established (Perez, 1983, 1989). Mycosphaerella musicola causes considerable
economic losses in ‘Cavendish’ cultivars (AAA) and occasionally in plantains
(AAB), which generally show an acceptable level of resistance to the pathogen
(Vakili, 1968; Perez et al., 1981).
This article reviews the economic impact of black leaf streak disease on banana
and plantain production in Cuba, and summarises studies on the epidemiology,
disease management and resistance of cultivars carried out over several years
in Cuba.
72
MyLsd 17x24
25/06/03
15:15
Page 73
Session 1
L. Pérez Vicente et al.
Impact of disease
Protection cost
In the 1980s, 15 to 16 applications of mineral oil were carried out each year to control
M. musicola. A quarter of applications contained mixtures of dithiocarbamate
fungicides and two or three contained benomyl or propiconazole. By the end of the
80s, the use of a bioclimatic model to forecast the treatments, based on the method
of scoring the rate of development of the disease (Ganry and Meyer, 1972a, b), led
to a 30% reduction in the cost of controlling M. musicola (Perez, 1989). The cost per
hectare in ‘Cavendish’ plantations varied between US $134 and US $241 (Figure 1).
In 1991 and 1992, the first and second years after the first outbreak of black leaf
streak disease, the cost per hectare rose to US$640 and US $801. The adoption of a
warning system to time the applications of oil-based systemic fungicides (Perez et
al., 1993a, 2002) reduced the cost in later years (Perez et al., 1993b, 1997, 2000a, b;
Porras and Perez, 1997; Perez, 1998).
Changes in areas occupied by the cultivars
Banana and plantain production in Cuba is for local consumption due to a lack
of export markets and to insufficient production levels to satisfy the high demand.
The areas planted with different cultivars are shown in Figure 2. In 1990, more
than 14 000 ha were planted with ‘Cavendish’ cultivars and treated with fungicide
to control Sigatoka disease. In 2001, ten years after the arrival of black leaf streak
disease, the area was reduced to 15% of that figure. In 1990, more than
43 000 ha were planted with plantains but black leaf streak disease reduced this
to 18% in later years.
The ‘Cavendish’ plantations have been replanted with resistant FHIA hybrids,
particularly ‘FHIA-23’ and ‘FHIA-18’ (AAAB). As a result, the amount of fungicide
imported for controlling black leaf streak disease in ‘Cavendish’ has declined
(Figure 3). The plantains have been replaced by ‘Burro CEMSA 3/4’ (ABB) and
‘FHIA-03’ (AABB).
Epidemiology of black leaf streak disease:
relation betwen weather and disease development
Tables 1 and 2 show the correlation matrix between weather variables – the cumulated
quantity and duration of rain over a 10 or 14-day period, the weekly Piche
evaporation, the weekly accumulated hours with relative humidity over 95%
(RH>95%) - and the state of development of the disease (Fouré, 1988) recorded weekly
for eight weeks in ‘Grande naine’ and ‘CEMSA 3/4’. High significative correlations
were found between the accumulated quantity and duration of rain and RH>95%,
and the state of development of the disease in banana and plantain.
In ‘Cavendish’ bananas, Perez et al. (1993b, 2000a) found the highest correlations
between the cumulated quantity and duration of rain over a 10, 14 or 28-day period,
and the state of development of the disease four to six weeks later. Regression
73
25/06/03
15:15
Page 74
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Protection cost
in US dollars per hectare
800
700
600
500
400
300
200
100
50
Nueva Paz, 1994
Menendez, 1995
Menendez, 1994
Limoncito, 1994
La Cuba, 1992
La Cuba, 1991
Horquita, 1990
Contramaestre, 1990
Sagua, 1990
Sagua, 1989
Artemisa, 1987
Banao, 1984
0
Figure 1. The cost of protection per hectare in various banana producing enterprises before and after the
first report of Mycosphaerella fijiensis in Cuba. Bars for black leaf streak disease represent the costs in the
first and second years after the disease appeared in each plantation.
60
50
Planted area (x 1000 ha)
MyLsd 17x24
40
30
20
0
Cavendish (AAA)
Plantains (AAB)
1990
1995
ABB cultivars
1997
1999
FHIA cultivars
2000
Figure 2. Surface area planted with various types of banana and plantain in Cuba, 1990-2000.
74
25/06/03
15:15
Page 75
Session 1
L. Pérez Vicente et al.
12
2500
10
2000
8
1500
6
1000
4
Planted area (X 1000 ha)
Cost of fungicides (x 1000 US dollars)
MyLsd 17x24
2
0
1989 1990
1991
1992 1993 1994 1995
Cost of fungicides
1996 1997
1998
1999 2000 2001
0
Planted area
Figure 3. Cost of fungicides to protect ‘Cavendish’ plantations and area planted with FHIA hybrids that are
partially resistant to black leaf streak disease.
equations were used for develop models to predict the development of black leaf
streak disease as a function of the duration and intensity of rainfall. The observed
and estimated state of development of the disease in ‘Grande naine’ as a function
of rainfall four weeks before is shown in Figure 4.
In ‘CEMSA 3/4’, Perez et al. (2000b), found high correlations between the state
of development of the disease and the cumulated quantity and duration of rain over
a 10 or 14-day period four to six weeks before, and the weekly Piche evaporation
(PwEv) three to six weeks before (Table 2). The highest correlation coefficient was
obtained between the cumulated quantity of rain over a 14-day period and the state
of development of the disease five weeks later. In ‘CEMSA 3/4’, a highly significant
negative correlation was found between PwEv and the state of development of black
leaf streak disease three to five weeks later.
In general, the state of development of the disease in any week of the year is highly
dependent on leaf wetness four and five weeks before. Many of the biological process
of the fungus, such as mating between compatible isolates and pseudothecia
development (Mourichon and Zapater, 1990), conidia development, ascospore and
conidia germ tube growth (Porras y Perez, 1997) and ascospore release from
pseudothecia (Stover 1980), are dependent on the presence of a water layer on the
leaf surface or on high relative humidity.
In ‘CEMSA 3/4’, the cumulated duration of rain over a 14-day period and the state
of development of the disease four weeks later are shown in Figure 5. The progressions
and regressions of the disease were highly correlated with rainfall and leaf wetness.
75
MyLsd 17x24
25/06/03
15:15
Page 76
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 1. Correlation matrix between climatic factors and the state of development of black leaf streak disease in ‘Grande
naine’, La Cuba 1995–1996. (Adapted from Perez et al., 2000a).
Number of weeks after the recording of the independent variable
Dependent
variable
Independent
variable
SD4L
Rf7mm
RfD7 min
Rf10 mm
RfD10 min
Rf14 mm
RfD14 min
H7
H10
H14
RH>95%
PwEv
T Med
0
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
-0.27 *
0.32 *
1
0.34*
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
-0.29 *
n.s.
2
3
n.s.
n.s.
0.38 *
0.37 *
0.45 **
0.38 *
n.s.
0.55*
n.s.
n.s.
n.s.
0.29 *
0.56 ***
0.45 **
0.54 ***
0.39 *
0.64 ***
0.51 **
n.s.
n.s.
n.s.
0.316*
n.s.
0.29 *
4
0.61 ***
0.48 **
0.80 ***
0.74 ***
0.77 ***
0.75 ***
0.72 **
0.79**
n.s.
0.276*
n.s.
n.s.
5
0.74 ***
0.77 ***
0.71 ***
0.73 ***
0.69 ***
0.73 ***
0.71 **
0.70 **
n.s.
0.308*
n.s.
n.s.
6
7
0.41 *
0.52 **
n.s.
n.s.
n.s.
n.s.
0.53 *
n.s.
n.s.
0.36**
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s
n.s.
n.s.
0.36**
n.s.
n.s.
* Statistically significant at probability 0.05. ** Statistically significant at probability 0.01. *** Statistically significant at probability 0.001.
n.s.: Not statistically significant.
SD4L: State of development of black leaf streak disease in the four youngest leaves. Rf(n)mm: Cumulated quantity, in mm, of rain over a
period of n days. RfD(n)min: Cumulated duration, in min, of rain over a period of n days. H(n): Cumulated quantity, in mm, of water on the
leaves over a period of n days. PwEv: Weekly Piche evaporation.
Table 2. Correlation matrix between climatic factors and the state of development of black leaf streak disease in ‘CEMSA
3/4’ (Musa spp., AAB), La Cuba 1996. (Adapted from Perez et al., 2000b).
Number of weeks after the recording of the independent variable
Dependent
variable
Independent
variable
SD4L
Rf7 mm
RfD7 min
Rf10 mm
RfD10 min
Rf14 mm
RfD14 min
H7 (mm)
H10 (mm)
H14 (mm)
PwEv
T Med
0
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0.49*
0.57**
-0.44 **
n.s.
1
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0.50*
0.57**
-0.58 ***
n.s.
2
3
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0.58*
-0.60***
n.s.
0.46 **
0.34 *
0.51 **
0.40 *
0.52 **
0.41 *
n.s.
n.s.
n.s.
-0.70***
0.40 *
4
0.48 **
0.39 *
0.59 ***
0.50 **
0.65 ***
0.58 ***
n.s.
0.73**
0.81**
-0.74***
0.45 **
5
0.60 ***
0.64 ***
0.74 ***
0.74 ***
0.74 ***
0.72 ***
n.s.
n.s.
0.55*
-0.75***
0.48 **
6
0.58 ***
0.55 **
0.62 ***
0.57 ***
0.64 ***
62 ***
n.s.
n.s.
n.s.
-0.52**
0.51 **
7
0.55 **
0.54 **
0.40 *
0.38 *
0.39 *
0.38 *
n.s.
n.s.
n.s.
n.s.
0.43 **
* Statistically significant at probability 0.05. ** Statistically significant at probability 0.01. *** Statistically significant at probability 0.001.
n.s.: Not statistically significant.
SD4L: State of development of black leaf streak disease in the four youngest leaves. Rf(n)mm: Cumulated quantity, in mm, of rain over a
period of n days. RfD(n)min: Cumulated duration, in min, of rain over a period of n days; H(n): Cumulated quantity, in mm, of water on the
leaves over a period of n days. PwEv: Weekly Piche evaporation.
76
25/06/03
15:15
Page 77
Session 1
L. Pérez Vicente et al.
2000
2500
1750
State of development
of the disease 4 weeks later
2000
1500
1250
1500
1000
R2 = 0.56*
1000
750
500
500
State of development of the disease
Cumulated duration of rain over a 14-day period
Duration of rainfall
250
47 49 51
3
1
5
7
9
11 13 15 17 19 21 23 25 27 29 31 33 35
Weeks
0
Figure 4. Correlation between the cumulated duration of rain, in minutes, over a 14-day period and the state
of development of black leaf streak disease in the four youngest leaves (SD4L) of ‘Grande naine’ 4 weeks later.
(From Perez et al., 2000b).
2500
2750
2250
2000
1750
Duration of rainfall
2500
State of development
of the disease 4 weeks later
2250
2000
r = 0.72***
1750
1500
1500
1250
1250
1000
1000
750
750
500
500
250
250
State of development of disease
Cumulated duration of rain over a 14-day pediod (min)
MyLsd 17x24
0
48 50 52
2
4
6
8
10
12
14 16 18 20 22 24 26 28 30
Weeks
Figure 5. Curves of the cumulated duration of rain, in minutes, over a 14-day period and the state of
development of black leaf streak disease in the four youngest leaves (SD4L) of ‘CEMSA 3/4’ 5 weeks later. (From
Perez et al., 2000b).
77
25/06/03
15:15
Page 78
Mycosphaerella leaf spot diseases of bananas: present status and outlook
The curves of the observed and the estimated state of development of the
disease using the model SD4L = 6.24 (S14d Rfmm) – 198.2 PwEv + 1812 are shown
in Figure 6. The model predicts the progress of the disease three weeks later.
State of development of the disease (SD4L)
MyLsd 17x24
3000
Observed state of development
2500
Estimated state of development
R2 = 0.85*
2000
1500
1000
500
35 37 39 41 43 45 47 49 51
1
3
5
7
9
11
13 15
17 19 21
Weeks
Figure 6. Observed and estimated state of development of the disease in the four youngest leaves (SD4L).
The estimated values were derived using the model SD4L = 6.24 (S14d Rfmm) – 198.2 EvPp + 1812. (From Perez
et al., 2000b).
Between 1993 and 1996, 13 to 15 treatments of systemic fungicides per year were
applied in ‘Cavendish’ plantations using the pattern of rainfall as the variable on
which to base the decision of using fungicides. For example, the state of development
of black leaf streak disease and the timing of fungicide treatments, using the
bioclimatic model, are shown in Figure 7 for La Cuba in 1994.
Between 8 and 10 treatments of fungicides are required each year in plantain
plantations in Cuba, to achieve an adequate control of black leaf streak disease. The
treatments carried out using a bio-climatic model for the cultivar ‘CEMSA 3/4’ in
La Cuba plantations during 1996 are shown in Figure 8. However, the low yields of
plantain cultivars and the low prices paid for the product do not cover the costs of
controlling black leaf streak disease.
A comparison of the cost of controlling black leaf streak disease using the
bioclimatic model for timing applications with the cost of using a predetermined schedule
of fungicide applications are shown in Table 3. The use of the bioclimatic model led to
a 40% reduction in total costs, which resulted in an important reduction of the quantity
of fungicides imported in Cuba to control black leaf streak disease in ‘Cavendish’
plantations (Perez et al., 1993a, b, 1997, 2000a, b; Porras and Perez, 1997; Perez, 1998).
The model for timing applications of fungicides has shown to be effective in
regions where the total annual rainfall is under 2000 mm. The system depends on
the use of systemic fungicides able to inhibit the evolution of infections already
established in the host at the time of the application.
78
25/06/03
15:15
Page 79
Session 1
L. Pérez Vicente et al.
1200
16
14
1000
12
800
10
600
8
6
400
4
200
0
Youngest leaf spotted (YLS)
State of development of the disease (SD4L)
2
1
5
9
13
17
21
25
WEEKS
SD4L
SD4L
33
37
YLS
41
45
49
53
0
YLS
Figure 7. Black leaf streak disease control in a ‘Grande naine’ (AAA) plantation using the bioclimatic model
for timing applications in La Cuba in 1994 (fields 1 and 2). Arrows indicate the moments of the application.
(Adapted from Perez, 1998).
Bi
3000
14
12
2500
Te
Te
10
2000
1500
Pr
Oil
Tr
Be
8
6
Oil
1000
4
500
Youngest leaf spotted (YLS)
State of development of the disease (SD4L)
MyLsd 17x24
2
0
0
33 36
39 42
45 48
51
2
5
8
11
14
17
20 23 26 29 32
Weeks
SD4L
YLS
Figure 8. Black leaf streak disease control in a ‘CEMSA 3/4’ plantation using the bioclimatic model for timing
applications in La Cuba, 1995–1996. Arrows indicate the moments of the application with tebuconazole (Te),
propiconazole (Pr), mineral oil (Oil), benomyl (Be) and tridemorph (Tr). (Adapted from Perez et al. 2000b).
Resistance of cultivars
Different studies have been reported on the resistance of banana and plantains
to M. fijiensis (Meredith and Lawrence 1970, Firman 1972, Fouré et al.1984, Fouré
et al. 1990). Fouré et al. (1990) reported two different types of resistant reaction
79
MyLsd 17x24
25/06/03
15:15
Page 80
Mycosphaerella leaf spot diseases of bananas: present status and outlook
against black leaf streak disease in Musa: the hypersensitive reaction observed
in ‘Yangambi km 5’, and the partial resistance, that is expressed by the duration
of the cycle of evolution of the disease and a reduction in reproduction of the
pathogen. Studies done in 1995 (Hernandez and Perez, 2001), showed the
reaction and components of the resistance to black leaf streak disease of various
FHIA hybrids and Musa genotypes from the Cuba collection. Table 4 shows a
significant increase in the incubation period (from leaf emergence and the
appearance of the first symptoms), and in the transition period (from streak
stage to necrotic spots), as well as a significant reduction in the production of
pseudothecia. As a result, a reduction of the logistic rates of increment of infection
(typical of partial resistance) takes place and the plants reach the flowering stage
with a greater number of functional leaves.
Table 3. Comparison of the number and the cost of fungicide applications in ‘Cavendish’ plantations using a bioclimatic model and following a pre-determined schedule. (Adapted from Perez, 1998).
Timing of application
Plantations
Using a predetermined schedule Using a bioclimatic model
Year
Number
Total cost
Year
Number
Total cost
of treatments
(US$)
of treatments
(US$)
La Cuba
1991
1992
21
23
801.24
619.66
1993
1994
1995
1996
Limoncito
1994
22
1995
1996
1995
1996
1994
1995
18
23
1994
23
13
13
8
4
up to June
246.48
288.72
172.39
71.21
412.09
518.49
Sola
Nueva Paz
Guines
568.74
303.29
299.33
269.52
476.19
Quemado
De Guines
Menendez
15
13
12
12
599.66
1994
1995
1996
1996
1995
1996
11
12
11
13
13
10
221.56
282.78
237.31
326.56
308.96
219.78
(Hurricane)
A comparison of the infection indices and the youngest leaf spotted (Stover
and Dickson, 1970) are shown for FHIA hybrids plantations from four locations
in Cuba, during the most favourable periods for disease development in each
locality (Table 5).
80
MyLsd 17x24
25/06/03
15:15
Page 81
Session 1
L. Pérez Vicente et al.
Table 4. Reaction to black leaf streak disease of a group of FHIA hybrids in Cuba without fungicide protection. (Adapted
from Hernandez and Perez, 2001).
Clones
FHIA-1-1
FHIA-02
FHIA-03
FHIA-18
SH 3436
Grande naine
Incubation period
(days)
Transition period
(days)
Number of functional
leaves at harvest
February
June
February
June
Mother
plant
First
follower
43.5 b
46.9 b
60.4 a
52.8 ab
35.8 c
27.9 c
28.2 a*
31.0 a
24.1 b
28.7 a
28.0 a
16.7 c
76.4**
>150**
>150**
>150**
84.3*
36.6
75.5
86.6
107.7
119.0
80.2
43.0
9
8
10
12
10
1
8
10
8
9
7
0
Spot size
(mm)
17.3
13.3
15.5
14.3
12.7
17.5
Number of
pseudothecia
15.9
34.8
31.6
35.0
9.5
73.6
* Different letters indicate significant differences at probability 0.05.
** Most spots stopped developing at Fouré’s stage 3 (Fouré, 1984).
Table 5. Infection index (II) and youngest leaf spotted (YLS) in FHIA hybrids and ‘Cavendish’ at the moment of maximum
disease severity.
Cultivar
Locality and year
II (%)
YLS
FHIA-23
La Cuba, 1996
La Cuba, 2001
Alquízar, 2001
20.3
16.3
21.7
6.8
7.8
6.5
FHIA-18
La Cuba, 1996
La Cuba, 2000 (Fca. Berlier)
La Cuba, 2000 (Fca. Cozola)
La Cuba, 2001
Alquizar, 2001
Baracoa, 2002
2.1
28.2
23.2
10.3
14.2
15.8
11.0
5.6
6.4
7.6
10.6
9.6
Cavendish
La Cuba, 1996
La Cuba, 2001
Alquizar, 2001
Baracoa, 2002
35.6
36.3
59.8
4.3
5.3
4.3
In the last two years, there has been an increase in the severity of black leaf streak
disease in experimental plots in Ciego de Avila compared with the levels observed
in 1996, despite the fact that in 1996 the plots were largely surrounded by fields of
‘Cavendish’ bananas. Increased disease severity in Ciego de Avila has been specially
marked on the ‘FHIA-18’ and ‘FHIA-23’ hybrids. The causes of the changes in the
susceptibility are under study. A negative correlation has been observed between
the level of K in soil and the severity of black leaf streak disease. Table 6 shows
values for the concentration of K and the K/K+Ca+Mg ratio in the soil of farms with
red latosolic soils planted with the cultivar ‘FHIA-18’ and not sprayed against black
leaf streak disease, and disease severity expressed as the youngest leaf spotted.
Farms with the lowest content of K in the soil had the highest severity of black
leaf streak disease (Table 6). Potassium seems to have an important role in the defence
mechanisms of banana plants of the FHIA cultivars to pathogens. Peng et al. (1999)
81
MyLsd 17x24
25/06/03
15:15
Page 82
Mycosphaerella leaf spot diseases of bananas: present status and outlook
reported that soils deficient in K are conducive to Fusarium oxysporum f.sp. cubense
in Australia. Different stress factors have been shown to have an influence on
peroxidase and phenylalanine ammonia lyase enzyme systems which are at the same
time related to defence mechanism in plants to pathogens (Aguiar et al., 2000).
A nationwide survey of pathogenic variability of populations of M. fijiensis has
been undertaken based on the widespread use of FHIA hybrids in Cuba. A collection
of single ascospore isolates of M. fijiensis from different cultivars collected at
locations where FHIA hybrids have been extensively planted and from regions where
they have not been introduced has been carried out. Artificial inoculations on selected
hybrids under controlled conditions are in progress.
Table 6. Disease severity expressed as the youngest leaf spotted (YLS) in farms planted with ‘FHIA-18’ and availability
of potassium in the soil, La Cuba, Ciego de Avila, in Nov 2000.
Farms
YLS
K
(meq/100g soil)
K/(K+Ca+Mg)
El Berlier
Cozola
El Transformador (Cuba 3)
El Colorado (Farm 10)
La pista (Tin)
Cooperative
5.6
6.4
7.0
8.0
10.5
11.2
0.44
0.52
0.65
0.85
1.5
1.2
1/59
1/56
1/38
1/39
1/14
1/15
Conclusion
1. Black leaf streak disease has had a strong impact on the economy of growers
and on banana and plantain production since the first appearance in Cuba. The
costs/ha of control of black leaf streak disease increased fourfold due to the increase
of the number of fungicide treatments. The annual cost of fungicides in Cuba for
black leaf streak disease control in the first year after disease outbreak in Cuba reached
US$2 million.
2. The area planted with ‘Cavendish’ cultivars and with plantains had been reduced
to 15% and 18% respectively of the existing area previous to the introduction of
black leaf streak disease. At the same time, the area planted with FHIA hybrids that
are resistant to black leaf streak disease is steadily increasing, leading to a 19%
reduction in the amounts of fungicides imported during the two first years following
the arrival of black leaf streak disease.
3. The development of the disease in Cuba is highly correlated with the cumulated
quantity and duration of the rain over a 14-day period, four weeks before in the
case of ‘Cavendish’, and five weeks before in the case of plantains (AAB). These
relationships can be useful for timing fungicide treatments in banana and plantains.
4. From 1993 to 1997, bioclimatic warnings were used in the main ‘Cavendish’
production plantations in Cuba, which allowed an optimization of control measures
and a 40% reduction in the costs of protecting against black leaf streak disease.
82
MyLsd 17x24
25/06/03
15:15
Page 83
Session 1
L. Pérez Vicente et al.
5. Resistance of the different FHIA hybrids is expressed as a longer period of
incubation, a longer period of transition from streaks to spots, and a reduction of
the production of pseudothecia, all of which are typical of partial resistance. The
severity of black leaf streak disease on the cultivar ‘FHIA-18’ is currently higher in
soils with low K contents.
6. Studies are in progress to determine the potential for M. fijiensis populations to
change pathogenicity as a result of the extensive planting of resistant hybrids.
References
Aguilar E.A., D.W. Turner and K. Sivasithamparam. 2000. Fusarium oxysporum f.sp. cubense
inoculation and hypoxia alter peroxidase and phenylalanine ammonia lyase enzyme
activities in nodal roots of banana cultivars (Musa sp.) differing in their susceptibility to
Fusarium wilt. Australian Journal of Botany 48:589–596.
Firman I.D. 1972. Susceptibility of banana cultivars to fungus diseases in Fiji. Tropical
Agriculture Trinidad 49:189-196.
Fouré E. 1988. Stratégies de lutte contre la Cercosporiose noire des bananiers et des plantains
provoquée par Mycosphaerella fijiensis Morelet. L’avertissement biologique au Cameroun.
Evaluation des possibilités d’amélioration. Fruits 43(5):269-274.
Fouré E., M. Grisoni and R. Zurfluh. 1984. Les Cercosporioses du bananier et leurs traitements.
Comportement des variétés. II. Etude de la sensibilité des bananiers et plantains
á Mycosphaerella fijiensis Morelet et des quelques caractéristiques biologiques de la maladie
des raies noires au Gabon. Fruits 39:365-378.
Fouré E. A. Moulioum Pefoura and X. Mourichon. 1990. Etude de la sensibilité variétale des
bananiers et des plantains à Mycosphaerella fijiensis Morelet au Cameroun. Caractérisation
de la résistance au champ des bananiers appartenant à divers groupes génétiques. Fruits
45:339-345.
Ganry J. and J.P. Meyer. 1972a. La lutte contrôlée contre la Cercosporiose aux Antilles. Bases
climatiques de l’avertissement. Fruits 27:665-676.
Ganry J. and J.P. Meyer. 1972b. La lutte contrôlée contre le Cercospora aux Antilles.
Application de techniques d’observation et numération de la maladie. Fruits 27:767-774.
Hernandez A. and L. Perez. 2001. Reaction of banana and plantain cultivars to black Sigatoka
disease caused by Mycosphaerella fijiensis, Morelet. Epidemiological components of
the resistance. Fitosanidad 5(3):9-15.
Meredith D.S. and J.S. Lawrence. 1970. Black leaf streak of bananas (Mycosphaerella fijiensis).
Susceptibility of cultivars. Tropical Agriculture Trinidad 47:275-287.
Peng H.X., K. Sivasithamparam and D.W. Turner. 1999. Chlamydospore germination and
Fusarium wilt of banana plantlets in suppressive and conducive soils are affected by
physical and chemical factors. Soil Biology and Biochemistry 31(10):1363-1374.
Pérez L. 1983. Epifitiología de la mancha de la hoja del plátano (Sigatoka) causada por
Mycosphaerella musicola. Factores que influyen en el período de incubación y el desarrollo
de la enfermedad en Cuba. Agrotecnia de Cuba 15(1):55–64.
Pérez, L. 1989. Sistema de pronóstico climático fenológico de los tratamientos contra la
mancha de la hoja (Sigatoka) causada por Mycosphaerella musicola en plátano fruta (Musa
acuminata AAA). Agrotecnia de Cuba 21(2):35–46.
Pérez L., 1998. Black Sigatoka disease control in banana and plantains plantations in Cuba.
Management of the disease based on an integrated approach. INFOMUSA. Vol. 7(1):27-30.
83
MyLsd 17x24
25/06/03
15:15
Page 84
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Pérez L., C. Torres, M. Delgado and F. Mauri. 1981. Resistencia de diferentes clones de plátano
a la Sigatoka causada por Mycosphaerella musicola Leach. Agrotecnia de Cuba
13(2):51–66.
Pérez L., F. Mauri, B. Barranco and G. García. 1993a. Efficacy of EBI’s fungicides in the
control of Mycosphaerella fijiensis Morelet on banana and plantains with treatments based
on stage of evolution of the disease (biological warnings) in Cuba. P.55 in Proceedings
of the 6th International Congress of Plant Pathology, Montreal.
Pérez L., F. Mauri, A. Hernández and A. Porras. 1993b. Efficacy of a biological warning system
for timing fungicide treatments for the control of black Sigatoka disease (Mycosphaerella
fijiensis Morelet) in banana plantations in Cuba. Proceedings of the 6th International
Congress of Plant Pathology, Montreal.
Pérez L., A. Hernández, A. Porras, E. Abreu, A. Guzmán, J. Montero, A. Méndez, H. Martínez,
A. Aguirre y R. Pupo. 1997. Generalización del manejo integrado de Sigatoka negra en
bananos y plátanos. Balance de cuatro años de su aplicación en áreas de producción.
XII Fórum de Ciencia y Técnica.
Pérez L., F. Mauri, A. Hernández, E. Abreu y A. Porras. 2000a. Epidemiología de la Sigatoka
negra (Mycosphaerella fijiensis Morelet) en Cuba. Pronóstico bio-climático de los
tratamientos en bananos (Musa acuminata AAA). Revista Mexicana de Fitopatología
18(1):15–26.
Pérez L., A. Hernández y A. Porras. 2000b. Epidemiología de la Sigatoka negra (Mycosphaerella
fijiensis Morelet) en Cuba. Pronóstico bio-climático de los tratamientos en plátanos (Musa
spp. AAB). Revista Mexicana de Fitopatología 18:27–35.
Porras A. y L. Pérez. 1997. Efecto de la temperatura en el crecimiento de los tubos germinativos
de las ascósporas de Mycosphaerella spp. causantes de Sigatoka en plátanos. Cálculo de
las sumas de velocidades de desarrollo para el pronóstico de los tratamientos a partir de
la temperatura máxima y mínima diarias en Cuba. INFOMUSA 6(2):27-31.
Stover R.H. and J.D. Dickson. 1970. Leaf spot of bananas caused by Mycosphaerella musicola
Leach. Methods of measuring spotting prevalence and severity. Tropical Agriculture
Trinidad 47: 289-302.
Vakili N.G. 1968. Response of Musa acuminata species and edible cultivars to infection by
Mycosphaerella musicola. Tropical Agriculture Trinidad 45:13-22.
Vidal A. 1992. Sigatoka negra en Cuba. En nuevos focos de plagas y enfermedades. Boletín
Fitosanitario de la FAO 40:1-2.
84
MyLsd 17x24
25/06/03
15:15
Page 85
Session 1
A.B. Molina and E. Fabregar
Management of black leaf streak
disease in tropical Asia
A. B. Molina1 and E. Fabregar2
Abstract
In the Philippines, diseases are the major production constraint in the region. Leaf spot diseases
cause significant fruit yield and quality reduction in both commercial banana plantations and
in small-scale farms. Black leaf streak disease is the most important banana leaf spot disease
in the region. Although eumusae leaf spot disease has been reported in India, Sri Lanka,Thailand
and Malaysia, the extent of damage has not yet been established. Sigatoka disease is no longer
a major concern.The damage due to black leaf streak disease affects mostly smallholders as they
generally do not implement any systematized disease management programme. Disease
management in commercial plantations is based on the use of fungicides together with cultural
practices that reduce the inoculum of the disease and therefore its development. Changes in
the cropping/production system in the commercial plantations of Southern Philippines have had
an impact on the management of Mycosphaerella leaf spot diseases. The adoption of annual
cropping reduced disease pressure and the need for year-round fungicide applications.
Resumen - Estado del manejo de la Sigatoka negra en Asia Tropical
El las Filipinas, las enfermedades representan la principal limitación en la región. Las enfermedades
de las manchas foliares provocadas por el género Mycosphaerella causan una reducción
significativa del rendimiento y calidad de la fruta tanto en las plantaciones comerciales como
en pequeñas fincas bananeras. La Sigatoka negra es el problema de las manchas foliares más
importante en la región. Aunque se ha reportado la presencia de Mycosphaerella eumusae (mancha
foliar eumusae) en India, Sri Lanka, Tailandia y Malasia, la dimensión de los daños que ella causa
aún no se ha establecido. La Sigatoka amarilla ya no representa la principal preocupación. El daño
debido a la Sigatoka negra se observa principalmente al nivel de los pequeños agricultores, ya
que, básicamente, ellos no implementan ningún programa de manejo sistematizado en
comparación con las plantaciones comerciales. El manejo de las enfermedades empleado en las
plantaciones comerciales es un programa basado en fungicidas reforzado por prácticas culturales
que reduce el inóculo de la enfermedad y reduce así su desarrollo. Los cambios en los sistemas
de cultivo y producción tuvieron cierto impacto sobre el manejo de la mancha foliar de Sigatoka
en las plantaciones comerciales en el sur de Filipinas. La adopción del cultivo anual redujo la presión
de la enfermedad y así se evita la necesidad de la aplicación de fungicidas durante todo el año.
1 INIBAP-Asia
2
and the Pacific, Los Baños, Philippines
Lapanday Fruits Development Corporation, Philippines
85
MyLsd 17x24
25/06/03
15:15
Page 86
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Résumé - Gestion de la maladie des raies noires en Asie tropicale
Aux Philippines, les maladies constituent la principale contrainte de la production. Les maladies
foliaires causées par les Mycosphaerella réduisent le rendement et la qualité des fruits dans les
plantations commerciales, ainsi que dans les petites fermes. Dans la région, la maladie des raies
noires est la plus importante maladie foliaire attaquant la banana. Bien que l’ELSD (eumusae leaf
spot disease) ait été observée en Inde, au Sri Lanka, en Thaïlande et en Malaisie, l’étendue des
dégâts n’a pas encore été établie. La maladie de Sigatoka n’est plus considérée comme importante.
Les dégâts dus à la maladie des raies noires affectent principalement des petits fermiers car ils
ne peuvent généralement pas mettre en place de programme systématique de gestion des
maladies. La gestion des maladies dans les grandes plantations est basée sur l’utilisation de
fongicides ainsi que sur des pratiques culturales visant à réduire la quantité d’inoculum et donc
le développement de la maladie. Des changements dans le système de production dans les
plantations commerciales du sud des Philippines ont eu un impact sur la gestion des maladies
foliaires causées par les Mycosphaerella. Le fait d’avoir adopté une culture annuelle a réduit la
pression de la maladie et le recours aux fongicides tout au long de l’année.
Introduction
Banana (Musa spp.), a group of plants that comprises many different types of sweet
dessert bananas, cooking bananas and plantains, is an important fruit crop in Asia.
Bananas are grown largely by smallholder farmers, traded by local entrepreneurs
and consumed locally. Thus, banana plays a major role in food security and is
a source of income for the rural poor. The Philippines is the only major banana-exporting
country in Asia and bananas generate important foreign exchange. In summary, banana
is an important food and source of income for local farmers and for the region.
The main constraints to banana production and threats to the industry in the region
are from pests and diseases. The region is the centre of origin of Musa, and hence many
serious pests and diseases affect the crop, e.g. leaf spot diseases caused by Mycosphaerella
spp. are responsible for a reduction in fruit yield and quality in commercial plantations
and on small-scale farms.
Three species of Mycosphaerella are present in Asia. M. fijiensis, responsible for black
leaf streak disease, is the most important pathogen because of its wide distribution and
its virulence which gives rise to epidemics. M. eumusae, a newly reported pathogen, is
potentially devastating and has been reported in India, Sri Lanka, Thailand and Malaysia
but not, so far, in the Philippines. M. musicola, the causal agent of Sigatoka disease,
is present in Southeast Asia but is no longer of major importance.
In commercial plantations, even if the disease does not affect yield, it can still reduce
fruit quality and render the fruit unfit for export. Leaf spot diseases also cause fruits
to ripen prematurely during transport to the market.
Disease management practices
Banana production systems in Southeast Asia are classified as follows (1) backyard,
(2) mixed crop, (3) commercial smallholder plantation and (4) large commercial
export-oriented plantation. The first three are intended for local markets and
production is on a small scale. Management of leaf spot diseases varies according
to the system of production.
86
MyLsd 17x24
25/06/03
15:15
Page 87
Session 1
A.B. Molina and E. Fabregar
Small-scale farming
Leaf spot disease management by small-scale banana growers ranges from minimal
to none, that is there is no systematic management of the disease. The disease reduces
fruit size on susceptible cultivars but the fruits are still acceptable to the local market.
Local consumers are not exacting in terms of fruit quality, size and ripening
characteristics, unlike the commercial export market. Thus, small-scale banana
growers make a certain amount of profit even when leaf spot diseases are not controlled.
In addition, small-scale banana growers plant many varieties of banana. Figure 1
shows the various local cultivars grown in the Philippines and their proportion relative
to total production. This variety of cultivars caters to the local demand for different
uses or consumption of bananas. Some varieties are used cooked, processed or as fresh
banana. The planting of different varieties provides genetic diversity against black leaf
streak disease. Several important local varieties, e.g. ‘Saba’ and ‘Pelipita’ in the
Philippines and ‘Pisang kepok’ in Indonesia, are resistant to Mycosphaerella leaf spot
diseases. Hence, for banana growers who specialize in cooking bananas, leaf spot
diseases are not important. ‘Lakatan’, ‘Pisang berangan’, and some ‘Cavendish’
varieties are susceptible to black leaf streak disease, but still produce yields that are
acceptable to the local market.
Latundan
8%
Bungulan
5%
Others
3%
Saba
39%
Lakatan
13%
Cavendish-type
32%
Figure 1. Most popular banana cultivars grown in the Philippines.
Chemical control is not practised by small-scale growers because of the expense.
Better-off local growers may remove severely diseased leaves as a means of reducing
inoculum and infection but none use fungicides. The variety of cultivars planted and
the diversity of growing conditions reduce the risk of disease in comparison with largescale monocrop commercial plantations.
Commercial plantations
A high level of disease control is required in commercial plantations for the export
market. In Southeast Asia, only the Philippines export large quantities of bananas.
87
MyLsd 17x24
25/06/03
15:15
Page 88
Mycosphaerella leaf spot diseases of bananas: present status and outlook
The high standards of fruit quality and the characteristically large-scale monocultures
used in the production of banana for export require intensive disease management.
Thus, the cultivation of a single banana variety ‘Cavendish’ and monocropping make
disease management in commercial plantations very challenging. However, the value
of the export market supports the use of expensive disease management practices.
The high levels of disease control achieved by a mixture of fungicides and cultural
practices reduce inoculum and conditions that favour disease development. In the
Philippines, fungicides may be applied up to 40 times per year (Table 1). The control
programme includes contact fungicides, which are alternated or mixed with systemic
fungicides applied in water or as an oil-water emulsion. The fungicides and doses used
by plantation owners are approved by the local regulatory agencies. In addition, the
fungicide doses are within the tolerances permitted by the importing countries. Table
2 lists the fungicides used in a typical commercial plantation in the Philippines.
Table 1. Spray programme of LADC, Davao, Philippines.
Number of cycles
Fungicide
1997
1998
1999
2000
2001
Strobilurin
Triazole
Dithane/Vondoze/Maneb
Daconil
Calixin
0
9
4
11
7
0
11
4
8
9
4
8
11
7
8
6
8
12
6
8
6
8
12
5
7
Total cycles
31
32
38
40
38
Oil used (L//ha/yr)
76
82
147
119
130
Table 2. Fungicides used by commercial growers in the Philippines.
Trade name
Chemical name
Volume/hectare
Systemic
Bankit 25 SC
Baycor 300 EC
Bumper 25 EC
Calixin 75 EC
Folicur 250 EC
Indar 2F
Siganex
Sico 250 EC
Tega 075 EC
Tilt 250 EC
Azoxystrobin
Bitertanol
Propiconazole
Tridemorph
Tebuconazole
Fenbuconazole
Pyrazophos
Propiconazole
Trifloxystrobin
Propiconazole
0.4 L
0.5 L
0.4 L
0.6 L
0.23 L
0.4 L
0.5 L
0.4 L
1.0 L
0.4 L
Protectant
Daconil 720 SC
Dithane F448
Dithane M-45
Dithane OS
Vondozeb 42 SC
Chlorothalonil
Mancozeb
Mancozeb
Mancozeb
Mancozeb
1.38 L
4.0 L
2.5 kg
1.75 L
4.0 L
The high dependence on the very specific fungicide propiconazole in the late 80s
to early 90s resulted in a considerable increase in insensitivity of Mycosphaerella.
88
MyLsd 17x24
25/06/03
15:15
Page 89
Session 1
A.B. Molina and E. Fabregar
However, the loss of effectiveness was less and occurred later than in Central America.
The prolonged effectiveness in the Philippines is possibly because spray programmes
have always been based on the principle of rotation and/or combinations of fungicides,
avoiding block or consecutive applications of the same products. The fungicide Benomyl
(Benlate) was used unwisely in the late 70s and resulted in a high degree of fungicide
resistance. Since then, Benlate has been withdrawn from use in the Philippines. The
introduction of newer chemicals such as Azoxystrobin has provided a much-needed
opportunity for fungicide control in combination with triazoles.
Commercial plantations also integrate cultural and other practices to their disease
management in order to reduce conditions that favour disease development. Good
drainage and irrigation practices receive attention. The removal and destruction of
severely affected leaves also reduce inoculum.
Monitoring of disease severity on plants with or without a flower is regularly
practiced in commercial plantation. In some plantations, early detection and
quantification of symptoms are also done. The data are used to schedule fungicide
treatments as well as a guide for harvest to reduce the risk of the effects of leaf spot
diseases on fruit quality. Frequently, when disease is severe and reduces the numbers
of functional leaves, fruits are harvested earlier to avoid the risk of premature fruit
ripening during transit to the market. This also removes a source of inoculum from
the plantation.
Changes in cropping/production systems have also had important effects on the
management of leaf spot diseases in commercial plantations in the southern Philippines.
Adoption of annual cropping reduced disease pressure and the need for year round
fungicide application. In Taiwan, where annual cropping was introduced more than
a decade ago, black leaf streak disease is no longer a problem. Expanded banana
commercial production in the Philippines included a few thousand plantations of
‘Lakatan’ to supply the lucrative local market, but not controlling leaf spot diseases
as intensely is providing an abundant source of inoculum for the export ‘Cavendish’
plantations, making control programmes more difficult. The export market has also
increased considerably during the last decade (Figure 2). The increased area devoted
to monoculture crops has increased the risk of epidemics.
Use of resistant varieties
As mentioned before, some popular local cultivars are resistant to black leaf streak
disease. However, many of the acuminata dessert type bananas, e.g. ‘Cavendish’, ‘Lakatan’
and‘Pisang berangan’, are susceptible to black leaf streak disease.
Several of the varieties provided by INIBAP through its International Musa Testing
Programme (IMTP) proved to be resistant to black leaf streak disease in Southeast Asia.
In particular, the FHIA series are very resistant to this disease in field trials. The high
yield potential and disease resistance of these varieties make them attractive to tropical
Asian farmers and consumers. It is worth noting that, being the centre of origin of
banana, Asian consumers in the tropics are used to eating different kinds of banana
of different size, colour and taste. Moreover, Asians prepare bananas for a variety of
uses. Hence, disease-resistant, high-yielding hybrids have the potential to increase
productivity of bananas in tropical Asia.
89
25/06/03
15:15
Page 90
Mycosphaerella leaf spot diseases of bananas: present status and outlook
120
100
80
Million boxes
MyLsd 17x24
50
40
20
0
1985
1990
1995
1998
1999
2000
2001
Year
Figure 2. Number of 13-kg boxes exported by the Philippines.
Literature consulted
Carlier J., M.F. Zapater, F. Lapeyre, D.R. Jones and X. Mourichon. 2000. Septoria leaf spot of
banana: A newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90:884:890.
Jones D.R., S.H. Jamaluddin and N.H. Nik Masdek. 1994. Banana disease survey of west
Malaysia. Pp. 49-62 in Proceedings of the Fourth Meeting of the Regional Advisory
Committee of INIBAP-Asia and the Pacific Network held in Taiwan Banana Research
Institute, Chiuju, Pingtung, Taiwan, November 21-25, 1994.
Magnaye, L.V. and L.E. Herradura. 1995. Rescue and Conservation of the Southeast Asian
Regional Banana Germplasm Collection (A Terminal Report of the INIBAP/IPGRI-funded
project of the same title). 47pp.
Molina G.C. and V.N. Villegas. 2001. Etiology and survey of banana leaf spot in the Philippines.
A final report of a collaborative project submitted to INIBAP Asia and the Pacific.
Qi Pei-Kun, Jiang Zi-De and Xi Ping-Gen. 2001. Etiology and preliminary survey of banana
leaf spot diseases in Guangdong Province in China. A final report of a collaborative project
submitted to INIBAP-Asia and the Pacific. College of resources and environmental sciences,
South China Agricultural University, Guangzhou, China.
Selvarajan R., S. Uma and S. Sathiamoorthy. 2000. Etiology and survey of banana leaf spot
diseases in India. Pp. 94-102 in Advancing banana and plantain R&D in Asia and the
Pacific, Vol. 10 (A.B. Molina, V.N. Roa and M.A.G. Maghuyop, eds) INIBAP-ASPNET, Los
Baños, Laguna, Philippines.
Sirisena J.A. 1997. Status of banana production in Sri Lanka. Pp. 160-180 in Proceedings of
the seventh meeting of the Regional Advisory Committee of INIBAP-ASPNET held in VASI,
Hanoi, Vietnam, October 21-23, 1997.
90
MyLsd 17x24
25/06/03
15:15
Page 91
Session 1
Z.J. Maciel Cordeiro and A. Pires de Matos
Impact of Mycosphaerella spp.
in Brazil
Z.J. Maciel Cordeiro and A. Pires de Matos
Abstract
Brazil ranks second in the world for banana production. Bananas are grown throughout the country,
mainly by smallholders. Of the Mycosphaerella species present in Brazil, M. musicola (anamorph
Pseudocercospora musae), the causal agent of Sigatoka disease, and M. fijiensis (anamorph
Pseudocercospora fijiensis), the causal agent of black leaf streak disease, are the most serious. They
can cause a 100% yield loss on susceptible varieties from the Cavendish (Musa cv. AAA) and Prata
(Musa cv. AAB) groups. Sigatoka disease is present in all banana growing areas of the country. Yield
loss is dependent on environmental conditions but is estimated at 50% on average. Black leaf streak
disease is still confined to the north region of the country plus the State of Mato Grosso (centrewest region). It causes a 100% yield loss in dessert bananas and around a 70% yield loss in plantain,
a crop which is very important as a staple food in northern Brazil. Since 1999, susceptible cultivars
have been gradually replaced by resistant cultivars, e.g.‘Caipira’ (AAA),‘Thap maeo’ (AAB),‘FHIA-18’
(AAAB) and ‘Pacovan Ken’(AAAB),especially in the State of Amazonas,following the recommendations
of the Brazilian agriculture research corporation (Embrapa).Traditional varieties have been replaced
by cultivars resistant to black leaf streak disease because of the high yield losses and the lack of
opportunity for growers to use fungicides. In addition there is the socio-economic impact of the
ban by the Federal and State authorities on the transport of banana fruits from infected areas,
designed to prevent the spread of black leaf streak disease to other banana growing regions of the
country.
Resumen - Impacto de Mycosphaerella spp. en el banano en Brasil
La producción bananera es una actividad muy importante en Brasil, el segundo productor del
mundo de este cultivo. El banano se cultiva en todo el país, principalmente por pequeños
agricultores. Entre las especies de Mycosphaerella que se dan en Brasil, M. musicola (anamorpho
Pseudocercospora musae), agente causal de la Sigatoka amarilla, y M. fijiensis (anamorpho
Paracercospora fijiensis), agente causal de la Sigatoka negra, son las más serias. Ellas han sido
responsables por pérdidas de rendimiento de hasta 100% en las variedades susceptibles como
las que pertenecen a los grupos ‘Cavendish’ (Musa cv. AAA) y Prata (Musa cv. AAB). La Sigatoka
amarilla se encuentra en todas las áreas productoras de banano del país causando pérdidas
de rendimiento que varía de acuerdo a las condiciones ambientales prevalecientes. En
Embrapa Mandioca e Fruticultura, Cruz das Almas, Brazil
91
MyLsd 17x24
25/06/03
15:15
Page 92
Mycosphaerella leaf spot diseases of bananas: present status and outlook
promedio, las pérdidas debido a la Sigatoka amarilla en Brasil se estiman en un 50%. La Sigatoka
negra está aún confinada a la región norte, más, el estado de Mato Grosso (región central
occidental). Además de causar pérdidas de rendimiento del 100% en los bananos de postre,
la Sigatoka negra también causa pérdidas de rendimiento de un 70% en plátano, muy
importante para la región norte de Brasil. En esta región, especialmente en el estado de
Amazonas, a partir de 1999, los cultivares susceptibles han sido reemplazados gradualmente
por los cultivares resistentes, recomendados por la Corporación Brasileña de Investigación
Agrícola (Embrapa) como: ‘Caipira’ (AAA), ‘Thap maeo’ (AAB), ‘FHIA-18’ (AAAB) y ‘Pacovan Ken’
(AAAB). El reemplazo de las variedades tradicionales fue forzado tanto por las altas pérdidas
de rendimiento, como por la inhabilidad de los productores de insertar los agroquímicos como
parte del sistema de producción. Además de la sostenibilidad ecológica de variedades
resistentes, también se observan los impactos socioeconómicos en las acciones de las
organizaciones federales y estatales que previenen el transporte de las frutas de banano de
las áreas infectadas, dirigidas a proteger las regiones del país libres de las enfermedades
producto de la diseminación de la Sigatoka negra.
Résumé - Impact des Mycosphaerella spp. au Brésil
Le Brésil est le deuxième producteur mondial de banane. Les bananiers sont cultivés dans tout
le pays surtout par des petits producteurs. De toutes les espèces de Mycosphaerella présentes
au Brésil, M. musicola (anamorphe Pseudocercospora musae), l’agent causal de la maladie de
Sigatoka, et M. fijiensis (anamorphe Pseudocercospora fijiensis), l’agent causal de la maladie des
raies noires, sont les plus sérieuses. Elles peuvent provoquer jusqu’à 100% de perte de production
sur des variétés sensibles comme celles des groupes Cavendish (Musa cv. AAA) et Prata (Musa
cv. AAB). La maladie de Sigatoka est présente dans toutes les régions productrices de bananes
du pays. La réduction du rendement dépend des conditions environnementales mais en
moyenne elle est estimée à 50%. La maladie des raies noires est encore confinée dans la région
Nord du pays ainsi que dans l’état du Mato Grosso (région centre-ouest). Elle provoque des pertes
de 100% chez la banane dessert et d’environ 70% chez la banane plantain, un aliment de base
important dans le nord du Brésil. Depuis 1999, les cultivars sensibles ont été progressivement
remplacés par des cultivars résistants, par exemple ‘Caipira’ (AAA), ‘Thap maeo’ (AAB), ‘FHIA-18’
(AAAB) et ‘Pacovan Ken’ (AAAB), particulièrement dans l’état de l’Amazone suivant les
recommandations de la société brésilienne de recherche en agriculture (Embrapa). Étant donné
les réductions importantes du rendement et l’impossibilité qu’ont les cultivateurs à utiliser des
fongicides, les variétés traditionnelles ont été remplacées par des cultivars résistants à la maladie
des raies noires. De plus, l’interdiction, par les autorités fédérales, de transporter les bananes
provenant de zones infectées, afin de limiter la propagation de la maladie des raies noires vers
d’autres régions du pays, a un impact socioéconomique certain.
Introduction
The cultivation of banana in Brazil plays an important economic and social role in large
and small plantations. Smallholders grow bananas mainly as a subsistence crop. Plantains
are grown as a staple food, mainly in the north and northeastern regions of the country.
Brazil is the world’s second largest producer of bananas and production is estimated
at six million metric tons per year. The total area cultivated with banana is about
533 000 hectares, distributed throughout the country: 8% in the south, 28% in the
southeast, 8% in the center west, 32% in the northeast and 24% in the north.
Of the various phytosanitary problems that affect banana production in Brazil, the
most important are those related to leaf diseases caused by Mycosphaerella musicola
and M. fijiensis, wilt diseases caused by Fusarium oxysporum f. sp. cubense and Ralstonia
92
MyLsd 17x24
25/06/03
15:15
Page 93
Session 1
Z.J. Maciel Cordeiro and A. Pires de Matos
solanacearum race 2, and soft rot caused by Erwinia carotovora. Nematodes, mainly
Radopholus similis, also affect the banana crop, and the weevil borer Cosmopolites
sordidus is perhaps the most important insect pest. Due to the increasing demands by
the consumer for fruits of better quality and appearance, pre and postharvest fruit diseases
and damage to fruit by arthropods, e.g. thrips, have become increasingly important.
Because they reduce the quantity and quality of bananas, and are difficult and expensive
to control, leaf spot diseases caused by M. musicola and M. fijiensis are considered to
be the most serious diseases affecting the Brazilian banana industry.
Impact of Mycosphaerella musicola
Sigatoka disease, caused by M. musicola Leach (anamorph Pseudocercospora musae
(Zumm) Deighton) was first reported in Java, in 1902. Sigatoka disease is found in all
banana growing regions except for Israel, Egypt and the Canary Islands. The disease
was first reported in 1944 in the Amazon region of Brazil. Eight years later, the disease
was found in the southeast region, the largest banana production region of Brazil. Today,
Sigatoka disease is widespread throughout Brazil causing severe yield losses in those
regions where environmental conditions favour disease development.
In Brazil, Mycosphaerella musicola results in an average yield loss of about 50%
which may be higher in particular regions of the country. In banana orchards affected
by Sigatoka disease, the number of bunches and of hands per bunch is lower and the
fruits, in addition to being smaller and lighter, ripen prematurely. A high incidence of
Sigatoka disease causes early decline of the banana orchard. Plant vigour declines and
yield is reduced in later crop cycles (Cordeiro and Matos, 2001) and orchards have to
be replanted at shorter intervals than usual.
According to Cordeiro (1990), by the end of the 80s, the cost of control using five
applications per year of systemic fungicides plus mineral oil accounted for 9% of the
total production cost estimated at US$1350/ha/year. In recent years, the cost has
increased because of the need to increase the numbers of applications to seven per
year. Today, the cost of controlling Sigatoka disease amounts to about 10% of the
total production cost. In areas where fungicide control measures are not used, yield
losses can exceed 80%
and fruit quality is very
poor (Figure 1).
Figure 1.
Damage from Sigatoka
disease on ‘Prata aña’ in the
absence of control measures.
93
MyLsd 17x24
25/06/03
15:15
Page 94
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Impact of Mycosphaerella fijiensis
Black leaf streak disease was first reported in 1963, in the Fiji Islands, district of
Sigatoka. In America, the disease was first reported in Honduras in 1972, where it
was called black Sigatoka. The disease spread from Honduras throughout the banana
growing areas of Central and South America. In Brazil, black leaf streak disease was
first discovered in 1998 (Pereira et al., 1998; Cordeiro et al., 1998), in banana orchards
located in the municipalities of Tabatinga and Benjamim Constant, State of
Amazonas, on the borders of Brazil, Colombia and Peru.
Distribution of black leaf streak disease in Brazil
Following the discovery of black leaf streak disease in Brazil, surveys were carried out
to follow its spread. By the end of 1998, black leaf streak disease was found in the
State of Acre, probably introduced accidentally from Bolivia. A survey in early 1999,
revealed high levels of black leaf streak disease in Rodônia, in the northern region of
the state, and Mato Grosso, in the central western region (Cordeiro et al., 2000). Recent
surveys in the northern region of Brazil showed that the disease had spread throughout
the region, e.g. the States of Pará, Amapá, Roraima, Amazonas, Acre and Rondônia
(Gasparotto et al. 2001a). Figure 2 shows the distribution of black leaf streak disease
in northern Brazil, inclu0ding the State of Mato Grosso, and the central western region.
Phytosanitary defence strategy
The yellow band in Figure 2 corresponds to a buffer zone, and includes the States
of Mato Grosso do Sul, Goiás, Tocantins and Maranhão, where a phytosanitary
defence strategy has been put in place to delay as much as possible the spread
of M. fijiensis to other banana growing areas currently free of the disease
(Figure 2). Planned actions include training for technicians and banana growers
with emphasis on the recognition of the symptoms of black leaf streak disease,
field evaluations of resistant varieties, e.g. ‘Caipira’ (AAA), ‘Thap Maeo’ (AAB),
‘FHIA-18’ (AAAB), ‘Pacovan Ken’ (AAAB) which are recommended by Embrapa.
Dependent on the agreement of growers, orchards in the buffer zone will be
replanted with varieties resistant to black leaf streak disease in order to establish
a barrier to prevent the spread of the pathogen. The movement of bananas from
areas affected by black leaf streak disease to disease-free areas is restricted; a
phytosanitary certificate of origin is required before permitting transport of fruits
to other areas.
Impact of black leaf streak disease
It is not yet possible to quantify exactly the impact of black leaf streak disease
on banana production in Brazil. In general, banana cultivation is characterized
by a low level of technology, mainly subsistence cultivation without the use of
fungicidal control measures. Nevertheless, field evaluations of fungicides to control
black leaf streak disease in the Amazon region have shown that approximately
94
MyLsd 17x24
25/06/03
15:15
Page 95
Session 1
Z.J. Maciel Cordeiro and A. Pires de Matos
26 applications with systemic fungicide and up to 52 applications of protectant
fungicides would be necessary to give adequate control of black leaf streak disease
in areas where weather favours the disease (Gasparotto et al., 2001b; Pereira and
Gasparotto, 2001). The large numbers of fungicide applications are ten times higher
than the total fungicide applied for the control of Sigatoka disease in some areas
of Brazil. In areas with clear dry periods, 15 fungicide applications give sufficient
control. Even in these conditions, the numbers of applications is twice that needed
to control Sigatoka disease.
Figure 2. Distribution of black leaf streak disease in Brazil (April 2002).
The biggest impact of black leaf streak disease in the Amazon region is the
need to change the varieties that are cultivated. According to growers of ‘Prata
comum’ in the Amazon region, banana production fell almost to zero after the
arrival of black leaf streak disease in the municipalities of Benjamim Constant
and Tabatinga in the State of Amazonas. The only way to continue banana
production in the area would be the use of varieties resistant to black leaf streak
disease. To support the use of resistant varieties, the government should acquire
micropropagated plantlets for distribution to growers. At present, however, a clear
decision has yet to be made about the preferred resistant cultivar, based on
information about yield performance and consumer acceptance. Since 1999, the
Embrapa tissue culture laboratory has sold 1 384 003 plantlets of varieties resistant
to black leaf streak disease to the states affected by the disease (Table 1).
95
MyLsd 17x24
25/06/03
15:15
Page 96
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 1. Varieties resistant to black leaf streak disease and number of plantlets sold to the area of incidence of the
disease in Brazil.
Variety
Number of plantlets
sold by 2001
Thap maeo (AAB)
Caipira (AAA)
FHIA-01 (AAAB)
FHIA-18 (AAAB)
SH36-40 (AAAB)
FHIA-21 (AAAB)
FHIA-03 (AABB)
FHIA-10
Ouro (AA)
Prata zulu (ABB)
PV03-44 (AAAB)
PV42-85 (AAAB)
Total
Number contracted
from 2002
Total number
of plantlets
301 968
648 100
42 000
163 735
128 150
9 000
5 900
3 000
4 650
3 000
74 500
50
350 000
200 000
300 000
150 000
-
651 968
848 100
42 000
463 735
128 150
9 000
5 900
3 000
4 650
150 000
74 500
50
1 384 053
1 000 000
2 384 053
Research activities
Before the discovery of black leaf streak disease in Brazil, all banana genotypes
generated by Embrapa’s breeding programme were sent to Costa Rica to evaluate
their resistance in collaboration with CATIE, INIBAP and CORBANA. At present,
genotypes have been evaluated in the states of Acre and Amazonas, in
collaboration with Embrapa Acre and Embrapa Western Amazon. In addition to
supporting the banana breeding programme in Brazil, the knowledge generated
by researchers has also helped the phytosanitary defence strategy. One of the most
important contributions to fight black leaf streak disease in Brazil has been the
delivery of resistant varieties. In addition, experiments have showed that spores
of M. fijiensis can survive for 60 days on several types of surface including clothes,
fruits, wood and iron (Hanada et al., 2000). That observation gave strong support
to the legal actions that ban the transport and sale of fruits from affected to disease
free areas.
References
Cordeiro Z.J.M. 1990. Economic impact of Sigatoka disease in Brazil. Pp. 56-60 in Sigatoka
leaf spot diseases of bananas (R.H. Stover and R. Fullerton, eds). Proceedings of an
international workshop, San José, Costa Rica, March 28–April 1,1989.
Cordeiro Z.J.M and A.P. de Matos. 2001. Sigatoka-amarela no Norte de Minas Gerais. Simpósio
Norte Mineiro sobre a cultura da banana, Nova Porteirinha, 6-9 novembro. Anais I,
pp. 238-247.
Cordeiro Z.J.M, A.P. de Matos and S. De O. Silva. 1998. Black Sigatoka confirmed in Brazil.
INFOMUSA 7(1):31.
Cordeiro Z.J.M., A.P. de Matos, L. Gasparotto and M. de J.B. Cavalcante. 2000. Disseminação
da Sigatoka-negra no Brasil. Summa Phytopathologica 26:110. (Abstract 062).
96
MyLsd 17x24
25/06/03
15:15
Page 97
Session 1
Z.J. Maciel Cordeiro and A. Pires de Matos
Gasparotto L., J.C.R. Pereira and D.R. Trindade. 2001a. Situação atual da Sigatoka negra da
bananeira. Fitopatologia brasileira (suplemento) 26:449. (Abstract 692).
Gasparotto L., J.C.R. Pereira, M.M. Costa and M.C.N. Pereira. 2001b. Fungicidas para o controle
da Sigatoka negra da bananeira. Fitopatologia brasileira (suplemento) 26:434. (Abstract
636).
Hanada R.E., L.Gasparotto and J.C.R. Pereira. 2000. Sobrevivência de conídios de
Mycosphaerella fijiensis em diferentes materiais. Fitopatologia brasileira (suplemento)
25:380. (Abstract 303).
Pereira, J. C. R., L. Gasparotto, A. F. Da S. Coelho, and A.F. Urben. 1998. Ocorrência da Sigatoka
negra no Brasil. Fitopatologia brasileira (suplemento) 23:295.
Pereira, J. C. R. and L. Gasparotto. 2001. Sigatoka negra da bananeira. Simpósio Norte Mineiro
sobre a cultura da banana. Pp. 102-104 in Nova Porteirinha, 6-9 de novembro, Anais I.
97
MyLsd 17x24
25/06/03
15:15
Page 99
Session 1
A.K.J. Surridge et al.
Poster
Fungi associated with banana foliage
in South Africa
A.K.J.Surridge, A. Viljoen and F.C. Wehner
Abstract
A comprehensive investigation was conducted to determine the identity, distribution and
importance of fungi associated with banana leaves in South Africa. Banana leaves were randomly
collected from the five banana growing areas in the country. Spores were isolated from leaf lesions
following surface sterilization and incubation in moisture chambers or taken directly collected
from lesions. Single spores were then cultured on half-strength potato dextrose agar. Both
molecular and morphological techniques were applied to identify the isolates. Four main
diseases were found in the different banana growing areas. Yellow Sigatoka (caused by
M. musicola), Mycosphaerella speckle (caused by M. musae) and Cordana leaf spot (caused by
Cordana musae) were present in all five areas, whereas, Cladosporium speckle (caused by
Cladosporium musae) only occurred in Levubu. Many other fungi, predominantly saprophytes and
endophytes, were also isolated. The most common species include (in order of predominance)
Nigrospora sacchari, N. sphaerica and N. oryzae.
Resumen - Hongos asociados con el follaje del banano en Africa del Sur
En Africa del Sur, se llevó a cabo una amplia investigación con el fin de determinar la identidad,
distribución e importancia de los hongos asociados con las hojas de banano. Las hojas de banano
se recolectaron al azar en cinco zonas bananeras del país. Los aislados se hicieron de las lesiones
foliares después de la esterilización de su superficie, incubación en cámaras húmedas, o las esporas
se recolectaron directamente de las lesiones. Luego, las esporas individuales fueron cultivadas
en el agar de dextrosa de patata de fuerza media. Para identificar los aislados se emplearon
técnicas tanto moleculares como morfológicas. En diferentes zonas productoras de banano se
descubrieron cuatro enfermedades principales. La Sigatoka amarilla (causada por M. musicola),
la mancha Mycosphaerella (causada por M. musae) y la mancha foliar Cordana (causada por
Cordana musae) se encontraron presentes en todas las cinco áreas, mientras que la mancha por
Cladosporium (causada por Cladosporium musae) solo se encontró en Levubu. También se
aislaron muchos otros hongos, predominantemente saprofíticos y endofíticos. Las especies más
comunes incluyen (en orden de predominancia) Nigrospora sacchari, N. sphaerica y N. oryzae.
University of Pretoria, Pretoria, South Africa
99
MyLsd 17x24
25/06/03
15:15
Page 100
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Résumé - Champignons associés au feuillage du bananier en Afrique du Sud
Une recherche détaillée a été faite afin de déterminer l’identité, la répartition et l’importance des
champignons associés aux feuilles de bananier en Afrique du Sud. Des feuilles de bananiers ont
été récoltées au hasard dans les cinq régions productrices du pays. Des spores ont été isolées des
lésions foliaires après stérilisation de la surface foliaire et mise en incubation dans des chambres
humides ou prélevées directement sur les lésions. Des spores isolées ont ensuite été cultivées sur
un milieu gélifié à l’agar à demi-concentration de dextrose de pomme de terre. Des techniques
moléculaires et morphologiques ont été appliquées afin d’identifier les isolats. Quatre maladies
principales ont été trouvées dans les différentes régions de culture de la banane. La maladie de
Sigaoka (causée par M. musicola),le Mycosphaerella speckle (causé par M. musae) et le Cordana (causé
par Cordana musae) étaient présents dans les cinq zones, alors que le Cladosporium (causé par
Cladosporium musae) n’a été trouvé qu’à Levubu. De nombreux autres champignons, surtout des
saprophytes et des endophytes, ont également été isolés. Les espèces les plus communes sont, dans
l’ordre d’importance, Nigrospora sacchari, N. sphaerica et N. oryzae.
Introduction
Among the various fungi associated with the foliage of banana plants, pathogens
such as Mycosphaerella musicola, M. fijiensis and M. eumusae cause significant
losses. Others, e.g. M. musae, Cladosporium musae and Cordana musae, can become
damaging under certain climatic conditions. In addition to these pathogens, various
species of endophytic fungi have also been reported on Musa species. The most
commonly isolated are Colletotrichum gloeosporioides, Nigrospora oryzae,
Pestalotiopsis palmarum and Phoma spp. (Brown et al., 1998 ; Photita et al., 2001).
Some of these, e.g. N. oryzae, are also known to cause minor disease on their host
plant (Ellis, 1971).
In South Africa, the fungal pathogens previously reported on banana include
M. musicola (Van den Boom and Kuhne, 1969), M. musae (Brodrick, 1973) and
Cordana musae (Roth, 1965). However, these reports were based solely on symptom
sightings and not on isolation of the causal organisms. This study was conducted
to determine, and update existing knowledge of, the identity of fungi associated with
banana leaves in South Africa.
Materials and methods
Banana leaves were randomly collected from the five banana growing areas in
South Africa during 2000-2001, namely Levubu, Tzaneen, Kiepersol, Komatipoort
and southern Kwa-Zulu Natal. Samples were taken from diseased leaves incubated
in moisture chambers and cultured on half-strength potato-dextrose agar. Isolates
were identified morphologically. The identity of the causal organism was confirmed
using diagnostic PCR and the species-specific primers of Johansen and Jeger (1993).
Results
Four leaf diseases were identified. Sigatoka disease, Mycosphaerella speckle and
Cordana leaf spot were present in all five areas. Cladosporium speckle occurred
only in Levubu, the most northern of the areas. M. musicola was the most
100
MyLsd 17x24
25/06/03
15:15
Page 101
Session 1
A.K.J. Surridge et al.
commonly isolated pathogen, followed by M. musae (Table 1). The presence of
M. musicola and the absence of other Mycosphaerella pathogens was confirmed
by PCR. Various saprobes and endophytes representing 23 species were also
isolated from banana leaf material (Table 1). Nigrospora oryzae was the species
isolated most frequently.
Table 1. Fungi associated with banana foliage in South Africa.
Species
Number of isolates
Alternaria alternata
Alternaria cf. citri
Alternaria tenuissima
21
3
4
Bipolaris cynodontis
Cladosporium musae
Colletotrichum gloeosporioides
Colletotrichum musae
Cordana musae
Curvularia lunata
Curvularia pallescens
Diaporthe sp.
Drechslera dematoidea
Drechslera phlei
Epicoccum nigrum
Exserohilum rostratum
Harpographium sp.
Mycosphaerella musae
Mycosphaerella musicola
Myrothecium verrucaria
Nigrospora oryzae
Nigrospora sacchari
Nigrospora sphaerica
Pestalotiopsis guepinii
Phoma glomerata
Phyllosticta sp.
Pithomyces sacchari
Selenophoma asterina
Selenophoma juncea
3
5
5
1
30
1
1
3
1
1
3
1
1
30
66
1
81
10
25
9
7
1
1
10
5
Discussion
M. musicola is the most common and severe pathogen of banana foliage in South
Africa. It was identified morphologically and its presence was confirmed using
molecular markers (Johanson and Jeger, 1993). The second most prevalent pathogen
was M. musae which in some cases caused severe symptoms resulting in leaf death.
The subtropical conditions in the banana plantations of South Africa appear to be
conducive to Sigatoka disease. Cordana musae is considered to be a minor/secondary
leaf pathogen and was most often observed infecting in conjunction with Sigatoka
disease or a speckle. Its presence and that of Cladosporium musae was
morphologically confirmed, as well as according to lesion appearance.
101
MyLsd 17x24
25/06/03
15:15
Page 102
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Most fungi isolated were strictly saprobes. The genus Nigrospora, particularly
N. oryzae, was the most commonly isolated. This conforms to literature of endophytes
isolated from banana leaves in Hong Kong (Brown et al., 1998). All species isolated
are the first recordings on banana leaves in South Africa. Colletotrichum
gloeosporioides, C. musae, N. oryzae, and some Curvularia, Pestalotiopsis, Phoma,
Phyllosticta species have previously been reported from Musa in Thailand (Photita
et al., 2001), Hong Kong and Northern Queensland, Australia (Brown et al., 1998).
However, no report could be traced referring to the presence of Alternaria cf. citri,
A. tenuissima, Bipolaris cynodontis, Diaporthe sp., Drechslera dematoidea, D. phlei,
Exserohilum rostratum, Harpographium sp., Myrothecium verrucaria, N. sacchari,
N. sphaerica, Pithomyces sacchari, Selenophoma asterina and S. juncea on Musa
species.
References
Brodrick H. T. 1973. Spikkelblaar. Banana Series Journal J4:1-2.
Brown K. B., K.D. Hyde and D.I. Guest. 1998. Preliminary studies on endophytic fungal
communities of Musa acuminata species complex in Hong Kong and Australia. Fungal
Diversity 1:27-51.
Ellis M. B. 1971. Dematiaceous Hyphomycetes. Commonwealth Mycological Institute, Kew,
Surrey, United Kingdom.
Johanson A. and M.J. Jeger. 1993. Use of PCR for detection of Mycosphaerella fijiensis and
M. musicola, the causal agents of Sigatoka leaf spots in banana and plantain. Mycological
Research 96:670-674.
Photita W., S. Lumyong, P. Lumyong and K.D. Hyde. 2001. Endophytic fungi of wild
banana (Musa acuminata) at Doi Suthep Pui National Park, Thailand. Mycological Research
105: 1508-1513.
Roth G. 1965. A new leaf spot disease of dwarf Cavendish banana in South Africa. South
African Journal of Agricultural Science 8:87-92.
Van den Boom T. and F.A. Kuhne. 1969. First report of Sigatoka disease of banana in South
Africa. Citrus Journal 428:17-18.
102
MyLsd 17x24
25/06/03
15:15
Page 103
Session 1
Recommendations
Recommendations of session 1
Dispersal of Mycosphaerella spp.
Mycosphaerella fijiensis continues to spread to new areas. It is the dominant leaf spot pathogen
in West Africa. In 2000-2002, the pathogen was identified for the first time in Madagascar,
the Bahamas, the Galapagos Islands of Ecuador and in the north Queensland banana growing
area where eradication is being attempted. The encroaching threat of M. fijiensis in the eastern
Caribbean is of concern. It has been estimated that 40% of banana growers in the French
Antilles would stop cultivating banana if the pathogen became established. The effects in the
Windward Islands would also be significant.
Quarantine may need to be either strengthened or reinforced to prevent entry. The
introduction of an appropriate monitoring system to detect any incursion of M. fijiensis
should be encouraged.
M. eumusae is currently limited in extent throughout most of Asia, although there is some
evidence that the pathogen may have reached Africa. Eumusae leaf spot disease has been
observed on ‘Mysore’ (AAB) in Sri Lanka. As this clone is highly resistant to M. musicola
and M. fijiensis, there is some cause of concern. Information suggests that Cavendish and
plantain cultivars are also very susceptible. The dynamics of the disease are not fully
understood. In order to prepare adequate disease control strategies, a detailed knowledge of
the epidemiology of this pathogen is urgently required.
The exact distribution of M. eumusae needs to be known. Further surveys in South and
Southeast Asia, to determine where M. musicola, M. fijiensis and M. eumusae occur,
are necessary.
More information on the effect of M. eumusae on the growth and yield of banana clones
is needed.
More laboratory research needs to be undertaken with M. eumusae to determine its
optimum growth temperature and other relevant biological information. This data would
underpin epidemiological studies.
The name of the banana clone affected, an indicator of the severity of the leaf spot and
local environmental data would be useful as this may help explain distribution. IMTP
trials are seen as ideal locations for assessing the reaction of different clones to the
different leaf spot pathogens. The collection and diagnosis of specimens of leaf spot from
IMTP trials sites needs to be continued. The cooperation and collaboration of scientists
in South and Southeast Asia is viewed as essential. Identification tools should be provided
to enable diagnoses to be undertaken locally.
103
MyLsd 17x24
25/06/03
15:15
Page 104
Mycosphaerella leaf spot diseases of bananas: present status and outlook
M. eumusae was also identified from leaf spot specimens collected in 1989 and 1990 at Onne
in Nigeria. Eumusae leaf spot disease has therefore been at this location for at least 13 years.
The IITA germplasm collection and clones in local banana farms in and around Onne
should be surveyed for leaf spot pathogens. The results would indicate the relative
competitiveness of M. fijiensis and M. eumusae on different cultivars in West Africa.
Although M. eumusae has not been identified from many isolates of leaf spot pathogens
collected in neighbouring Cameroon, surveys of leaf spot in Nigeria and other countries
in West Africa may indicate if spread has occurred
Taxonomy
Anamorph morphology is more important than teleomorph morphology in distinguishing
the Mycosphaerella leaf spot pathogens. It has been proposed that the anamorph stage of
M. fijiensis be renamed Pseudocercospora fijiensis as the phylogenetic studies do not support
keeping the name Paracercospora fijiensis.
Diagnostics
Several fungi diseases that attack leaves have been reported on Musa and other related species.
A greater knowledge of Mycosphaerella pathogens/saprophytes, and of those in related genera,
is a prerequisite to the development of rapid diagnostic tests to distinguish leaf spot pathogens.
Diagnostic tools depend on the development of species-specific primers, such as microsatellites
and ITS-sequences, tested on Mycosphaerella isolates from all over the world.
More taxonomic information about species of Mycosphaerella and other related genera
that either form or occur in banana leaf lesions would be beneficial.
Diagnostic tools specific to the three main species of Mycosphaerella pathogen on Musa:
M. fijiensis, M. musicola and M. eumusae should be developed.
The currently available molecular methods should be assessed for their specificity.
A manual with descriptions of symptoms and morphological characters should be
produced.
Protocols for collection and analysis of samples should be developed.
PROMUSA participants should be trained on the different technologies required:
collecting and sampling, single-ascosopore cultures, and molecular markers.
Resistant cultivars
Reports of black leaf streak disease on two resistant FHIA hybrids are of concern. Stress caused
by adverse growing conditions may be responsible.
Factors responsible for breakdowns need to be investigated, including the possibility that
resistance is being eroded.
104
MyLsd 17x24
25/06/03
15:15
Page 105
Session 2
Population biology
and epidemiology
MyLsd 17x24
25/06/03
15:15
Page 107
Session 2
L.H. Jácome
Introduction
Population biology and epidemiology
L. H. Jácome
Introduction
Mycosphaerella leaf spot diseases comprise one of the most important disease complex
limiting banana and plantain production. Sigatoka disease is still important in some
areas. In the 1990s, black leaf streak disease was spreading and is now reported from
most parts of the world. Recently, a new disease, eumusae leaf spot was reported in
South and Southeast Asia.
Black leaf streak disease reached most South American countries by the early
1990s and was reported in Bolivia in 1996 and Brazil in 1998. The disease was
reported in the Caribbean in the Dominican Republic in 1996 and in Haiti in 2000.
It threatens banana production in Puerto Rico and the Lesser Antilles. The disease
was reported in the United States in 1998. In 2000, two new outbreaks of black leaf
streak disease were reported in the Caribbean and Indian Ocean region. In Latin
America the latest report was in the Galapagos Islands in early 2001. In April 2001,
it was reported for the first time in a commercial production area near Tully in North
Queensland, Australia. This illustrates the spreading capacity of this disease.
The continued spread of black leaf streak disease in the tropics within the last
decade has made the disease the most economically important disease of bananas
and plantains. Except in the Philippines, Sigatoka disease has been replaced by black
leaf streak disease. Sigatoka disease is better adapted to cooler areas and dominates
at altitudes above 1200 metres above sea level. Therefore, knowledge of the
distribution and variability of the pathogens is needed to ensure the efficient
introduction and sustainability of resistant hosts.
Restriction enzyme analysis has revealed unique differences in the banding pattern
of DNA fragments. Genetic variability among populations of Mycosphaerella from
different geographical regions has been identified using DNA restriction fragment
Pathotec, La Lima, Honduras
107
MyLsd 17x24
25/06/03
15:15
Page 108
Mycosphaerella leaf spot diseases of bananas: present status and outlook
length polymorphism (RFLP) markers. RFLPs are used as markers on physical and
genetic linkage maps. To complement RFLP tests, methods based on polymerase
chain reaction (PCR) could be used to obtain DNA profiles. Differentiation of the
pathogens by PCR could be useful in determining their distribution, spread and
prevalence. Additional work is now needed to obtain accurate information on the
geographical distribution of genetic variants of M. fijiensis, and to test hypotheses
about the origin and spread of the pathogen, and its population structure.
Given that the disease caused by M. fijiensis is more serious than the one caused
by M. musicola, research is needed on the pathogenic variability of M. fijiensis,
the distribution of its pathogenic variants and the identification of sources of
resistance. Studies on genetic diversity in M. fijiensis using RFLP have shown high
levels of polymorphism. Genetic variability was greatest in the Philippines and Papua
New Guinea. Isolates from Africa, the Pacific Islands and Latin America formed
genetically homologous groups that were specific to each region. Groups from the
Pacific and Latin America also appeared to be related. Are these independent
introductions of the pathogen? Are we dealing with the effects of genetic drift on
the population structure?
Gametic disequilibrium analysis among RFLP loci has shown that genetic
recombination plays an important role in the population structure of M. fijiensis.
Gene pyramiding may not be durable. Mixing varieties or partial resistance could
be more appropriate. Observations indicate that resistance to black leaf streak disease
may be breaking down in some hybrids. It is not known whether this is due to more
pathogenic variants or to favourable environmental conditions resulting in severe
disease pressure. In addition to studying the effects of climatic conditions on the
infection process, managing the inoculum using cultural practices should be
considered. The fitness of fungicide-resistant Mycosphaerella deserves attention, as
does the potential contribution of Paracercospora fijiensis to epidemics of black
leaf streak disease in some areas.
Disease dynamics has been studied by calculating rates of disease development
in relation to climatic factors. Epidemiological studies demonstrated that ascospores
were the predominant inoculum of black leaf streak disease and their release was
correlated with rainfall. Therefore, it should be possible to predict future rates of
disease development from previous patterns of spore release in relation to
temperature and rainfall. However, no consistency has been observed between
ascospore release and disease development, probably because the conditions for spore
release are not always conducive to infection.
There is considerable variation in disease progress curves and the relative duration
of epidemics in pathosystems of Musa-Mycosphaerella spp. and plantation
management (treated with fungicide versus untreated). For each epidemic, it is
possible to determine the time of disease onset, the initial amount of disease, the
rate of disease increase, the area under the disease progress curve, the shape of the
curve, the maximum disease, final amount of disease and the duration of the
epidemic. In tropical and semitropical climatic zones, for epidemics that are not
curtailed by the harvest of an annual crop, e.g. black leaf streak disease, the
progressive and regressive phases of an epidemic correspond mainly to seasonal
changes in weather conditions.
108
MyLsd 17x24
25/06/03
15:15
Page 109
Session 2
L.H. Jácome
At one level, a temporal analysis seeks gross comparisons between experimental
treatments, e.g. fungicide spray schedules in order to evaluate strategies for disease
management. At a second, more complex level, changes in specific environmental
factors, pathogens or host resistance lead to changes in the epidemic that are reflected
by changes in the disease progress curves. A third level of analysis corresponds to
comparative epidemiology, where the purpose is to identify similarities and
differences between epidemics based on the shape of the disease progress curves and
to look for the elements that serve as primary determinants. Geographical populations
of Mycosphaerella, with marked genetic differentiation, could be considered as
separate epidemiological units requiring independent disease management. This
suggests that studies and modelling of the epidemiology, distribution and population
structure of the three Mycosphaerella leaf spot pathogens should be undertaken at
the national, regional and global levels.
In general, there is little information on the biology, population structure and
epidemiology of the Mycosphaerella leaf spot pathogens. As a result, areas requiring
further investigation are pathogen variability, distribution of variants, sources of
resistance, epidemiology and population structure. The papers and posters in this
session have been selected to improve our understanding of the population biology
and epidemiology of the Mycosphaerella pathogens of Musa. Aspects of the
aerobiological pathway of M. fijiensis ascospores and conidia at small-scale and
mesoscale levels are discussed. Such studies are needed to clarify the temporal and
spatial patterns of the spread of leaf spot diseases within or across banana cropping
sequences, and in relation to environmental and host factors. Several forecasting
schemes for black leaf streak disease have been introduced and are based on those
developed for Sigatoka disease in the French Antilles. The forecasting schemes
combine local weather and ascospore trapping data. The value of ascospore trapping
for forecasting black leaf streak disease is discussed.
Aspects of the genetic structure and evolution of Mycosphaerella pathogens at
the global, regional and local levels are discussed. Genetic differentiation and
independency of introductions of the pathogens according to the region are also
described. Knowledge of the variability within Mycosphaerella is necessary for
breeding and management of disease resistance. The usefulness of microsatellite
markers for the study of fungal populations having high evolutionary rate is presented
by an analysis of the introduction and spread of black leaf streak disease in Mexico.
The presence of polymorphisms in chromosome length between molecular karyotypes
of M. fijiensis is discussed. Understanding the organization of the genome could
lead to the development of new strategies for disease control management.
Given that pathogens can evolve to break down host resistance, further research
on the evolution of the three Mycosphaerella pathogens on resistant hosts is needed.
As stated in the Musa disease fact sheet No. 8, published by INIBAP, population
studies of Mycosphaerella pathogens of banana are required to determine whether
the banana-producing regions correspond to one or several epidemiological units.
The studies should use molecular markers and determine pathogenicity. Techniques
are also needed to determine quickly and reliably host-pathogen interactions in
controlled conditions. The variability in pathogenicity in genetically differentiated
populations of the three pathogens could then be evaluated by means of a standard
109
MyLsd 17x24
25/06/03
15:15
Page 110
Mycosphaerella leaf spot diseases of bananas: present status and outlook
set of Musa cultivars. Population studies should help define, for the different regions,
a set of Mycosphaerella isolates that are representative of the variability in virulence
and aggressiveness for use in resistance screening. Knowing the components of partial
resistance which greatly reduce the rate of disease development in the field is also
important. Finally, the evolution of the pathogen population in response to the
selection pressure exerted by resistant cultivars should be evaluated if durable
resistance is to be achieved.
110
MyLsd 17x24
25/06/03
15:15
Page 111
Session 2
P.J.A. Burt
Airborne dispersal
of Mycosphaerella fijiensis
P. J. A. Burt
Abstract
Dispersal of Mycosphaerella fijiensis ascospores and conidia within, above and outside unsprayed
banana plantations was studied in a series of field experiments at CATIE in Costa Rica. Laboratory
experiments were also conducted in Costa Rica and the UK, to estimate the potential viability
of spores dispersing through the atmosphere.Three field regimes were used to assess windborne
spore dispersal on a local and mesoscale, in relation to wet and dry seasons. Spore catches were
analysed in relation to the time of day of capture, temperature, rainfall and behaviour of the wind.
Results showed that ascospores and conidia are windborne within infected plantations and up
to several tens of kilometres away from disease sources. Laboratory studies of simulated spore
release under field conditions showed that significantly fewer spores entered the air than might
be expected on the basis of field surveys. The reason for this is unclear. There is also evidence
that a major constraint on the airborne dispersal of viable spores is their duration of exposure
to ultraviolet radiation in sunlight. A greater understanding of the microscale processes occurring
on the surface of an infected banana leaf is required in order to resolve the role of the wind in
the epidemiology of black leaf streak disease. A more accurate quantification of the numbers of
spores undertaking long-distance dispersal (with assessments of their viability in the field) is
also essential. Future research needs are discussed.
Resumen - Dispersión aérea de Mycosphaerella fijiensis
Se han investigado los aspectos de la vía aerobiológica de las ascosporas y conidias de
Mycosphaerella fijiensis en experimentos en el campo en CATIE, Costa Rica. También se realizaron
investigaciones en el laboratorio en Costa Rica y Reino Unido con el fin de evaluar la viabilidad
potencial de la dispersión de esporas en la atmósfera. Se utilizaron tres regímenes de campo para
evaluar la dispersión de esporas por el viento a escalas local y media en relación con las
estaciones húmeda y seca. La captura de esporas fue analizada basándose en la hora del día de
la captura, temperatura, precipitación y comportamiento del viento. Los resultados mostraron
que las ascosporas y conidias se propagan por el viento dentro de las plantaciones infectadas y
hasta decenas de kilómetros fuera de las fuentes de la enfermedad. Sin embargo, de los estudios
de laboratorio que simularon la liberación de las esporas bajo condiciones de campo, está claro
que se podría esperar que una cantidad significativamente menor de esporas penetre en el aire
en base a las encuestas en el campo. La razón de este hecho no está clara. También existen
Natural Resources Institute, Kent, United Kingdom
111
MyLsd 17x24
25/06/03
15:15
Page 112
Mycosphaerella leaf spot diseases of bananas: present status and outlook
evidencias de que la duración de su exposición a la radiación ultravioleta del sol es una de las
principales limitaciones para la dispersión aérea. Se requiere un mayor entendimiento de los
procesos a pequeñas escalas que ocurren en la superficie de una hoja de banano infectado con
el fin de resolver con mayor profundidad el papel del viento en la epidemiología de la Sigatoka
negra. También es esencial realizar una cuantificación más precisa de las cantidades de esporas
que se dispersan a largas distancias (con la evaluación de su viabilidad en el campo). Se debe
discutir estas futuras investigaciones.
Résumé - Dispersion aérienne de Mycosphaerella fijiensis
La dispersion des ascospores et des conidies de Mycosphaerella fijiensis a été étudiée au cours
d’une série d’essais en champ au CATIE, au Costa Rica. Des essais en laboratoire ont également
été effectués au CATIE et en Grande-Bretagne, afin d’estimer la viabilité potentielle des spores
au cours de leur dispersion dans l’atmosphère.Trois régimes en champ ont été utilisés pour évaluer
la dispersion par le vent des spores à l’échelle locale et moyenne, en relation avec la saison sèche
et humide. Les spores récoltées ont été analysées en relation avec l’heure de capture, la
température, la pluviométrie, et la force et la direction du vent. Les résultats ont montré que les
ascospores et les conidies sont transportées par le vent au sein des plantations infectées et jusqu’à
plusieurs dizaines de kilomètres des sources d’infection. Les simulations en laboratoire de la
dissémination des spores en conditions naturelles ont montré qu’un nombre de spores
significativement plus faible que celui attendu sur la base des mesures en champ entrait dans
l’air. Les raisons de ce phénomène ne sont pas claires. Il existe également des preuves qu’une
contrainte majeure à la dispersion aérienne de spores viables est la durée de leur exposition aux
rayons ultraviolets du soleil. Une meilleure compréhension des processus à l’échelle microscopique
qui ont lieu à la surface d’une feuille de bananier infectée est nécessaire, afin de comprendre le
rôle du vent dans l’épidémiologie de la maladie des raies noires. Une quantification plus précise
du nombre de spores qui sont dispersées à longue distance (avec une évaluation de leur viabilité
en champ) est également essentielle. Les besoins futurs de recherches sont discutés.
Introduction
Banana and plantain are major subsistence crops for small-scale farmers in the
developing world, and production is increasing worldwide. Crops are affected by a
range of pests, including Mycosphaerella musicola, the causal agent of Sigatoka
disease and M. fijiensis, the causal agent of black leaf streak disease. The infective
agents are ascospores and conidia. Black leaf streak disease has gradually replaced
Sigatoka disease in most banana growing areas and can reduce yields by up to 50%
(Stover and Simmonds, 1987). Fungicides can control black leaf streak disease but
they are too expensive for small-scale farmers and affect the environment. Black
leaf streak disease is currently absent from most of the Caribbean. Its arrival would
be disastrous for the smallholders. Quarantine regulations being strict, the risk of
black leaf streak disease arriving in the Caribbean by windborne dispersion requires
quantification.
Previous studies have implicated wind and water in the release of ascospores and
conidia of M. fijiensis, but there is disagreement in the literature about their relative
importance. Conidial dispersal appears to occur primarily in water, either as runoff
in dew or by rain splash (Leach, 1946; Stover, 1968, 1972; Meredith et al., 1973;
Stover and Simmonds, 1987; Gauhl, 1989) whereas ascospores are primarily removed
from diseased leaves by wind (Leach, 1946). There is also evidence that the conidia
may be blown off infected leaves (Stover and Simmonds, 1987), and that ascospores
112
MyLsd 17x24
25/06/03
15:15
Page 113
Session 2
P.J.A. Burt
are released by water (Leach, 1941; Meredith, 1962; Meredith and Lawrence, 1970;
Stover, 1970; Meredith et al., 1973). One thing is sure: within infected plantations,
ascospores and conidia are present in the air in varying amounts (Gauhl, 1989).
Long-distance airborne dispersal is known for some fungal pathogens e.g.
Peronospora tabacina, the causal agent of tobacco blue mould (Davies et al., 1985);
Cochliobolus heterostrophus, which causes corn leaf blight (Pedgley, 1982);
Melampsora spp., the causal agent of poplar rust; and Hemileia vastatrix, which causes
coffee leaf rust (Burdekin, 1960; Bowden et al., 1971; Pedgley, 1982). There are no
records of M. fijiensis ascospores or conidia high in the atmosphere. In the absence
of such records, it is necessary to look at disease incidence, but this can be unreliable,
as the infective agents may enter an area long before the disease is first observed
or reported. The global pattern of Sigatoka disease suggests windborne dispersal
from east to west. The pattern is less obvious for black leaf streak disease but disease
records suggest that some intra-continental spread may have been windborne (Burt,
1994).
Even assuming that airborne spores travel in large enough numbers to overcome
dilution in the atmosphere and that they enter an area where there are suitable hosts
present, they may be affected by environmental conditions during transport. Of these,
temperature and ultraviolet radiation in sunlight are probably the most significant.
High temperatures destroy spore walls and denature DNA (Parnell et al., 1998).
Ultraviolet (UV) radiation, depending on wavelength, also denatures DNA.
Wavelengths below 320 nm, especially 250-270 nm (which do not reach the ground),
are the most lethal (Setlow, 1974; Rotem et al., 1985; Chuang and Su, 1988; Rotem
and Aust, 1991). UV radiation at wavelengths above 290 nm does not reach the
ground, and kills spores within a few hours (Maddison and Manners, 1972; Bashi
and Aylor 1983; Rotem et al. 1985). The ascospores of M. fijiensis have thin walls
and are hyaline in contrast to the thick-walled spores of many windborne fungi.
It is clear from the literature that ascospores and conidia of M. fijiensis have
the capacity to be windborne over long and short distances, and that spores are
present in the air within diseased plantations. Water (dew and rain) has been
implicated in short-distance dispersal, although the precise role of wind in spore
dispersal is unclear.
The aerobiology of M. fijiensis was studied in 1992-1995 by investigating spore
dispersal within a plantation; spore movement across a small experimental site;
mesoscale dispersal; potential inoculum loads and spore viability. The studies were
conducted at the Centro Agronómico Tropical de Investigación y Enseñanza (CATIE),
Costa Rica, and in the UK. This paper summarizes the results.
Methods
Dispersal of spores within a plantation
Airborne spores of M. fijiensis were monitored at four sites using rotorod spore traps
within a naturally infected plantation at CATIE; samples were taken below the canopy,
at mid-canopy and above the canopy (Burt et al., 1997). Wind speed, temperature,
relative humidity and rainfall were recorded. Spore concentrations were meas113
MyLsd 17x24
25/06/03
15:15
Page 114
Mycosphaerella leaf spot diseases of bananas: present status and outlook
ured with a Burkard volumetric spore trap (Burkard Manufacturing Company,
Rickmansworth, UK) at a height of 1.6 m above ground level. Disease incidence and
disease development were also recorded (Burt et al., 1997). No attempt was made to
control black leaf streak disease by treatment with fungicide.
Small-scale dispersal of ascospores and conidia
A small-scale field investigation was undertaken between May and August 1995 in
order to clarify some of the results obtained by spore trapping and to investigate
the role of wind and water in spore dispersal (Rutter et al., 1998). Disease
development and spore movement was monitored in a small experimental plot in
which 100 healthy plantain plants were arranged 2 m apart (Figure 1). Data were
collected from quadrats and analysed. The numbers of ascospores and conidia in
and around the plot were measured using four sets of rotorods at 0.5 m and 1.5 m
above ground level for 20 minutes at 07.00 local time, the time when spore
concentration is highest (Rutter and Burt, 1997). Wind, temperature, relative
humidity and rainfall were also recorded. The ambient spore concentration was
recorded continuously using a Burkard spore trap set at 6 m above ground level.
An inoculum made up of a bunch of diseased leaves was placed in the centre of the
plot (Figure 1) and the progress of the disease measured across the plot.
B
Q2
N
Q3
R3/4
X
R7/8
R5/6
Q4
Q1
Figure 1. Small plot experimental design. Plants, shown by circles, were 2 m apart. R refers to the position
and number of rotorod traps, X the location of the source of inoculum, the two rotorod traps and the weather
instruments. The Burkard trap (B) was located outside the plot. Q refers to quadrats. (From Rutter et al., 1998).
114
MyLsd 17x24
25/06/03
15:15
Page 115
Session 2
P.J.A. Burt
Mesoscale dispersal
Mesoscale dispersal of spores was investigated by sampling along a 5 km transect
across the floor and up one side of a valley outside CATIE, between April and August
1995 with 3 continuously recording Burkard spore traps. Trap 1 was located in the
middle of the plantation, trap 2 in the small experimental plot and trap 3 was 5 km
north-north-west of the plantation, on the side of the valley at an elevation of 1000
m (Burt et al., 1998). There were no obstacles between trap 3 and the other two traps
and there were no sources of inoculum close to trap 3 (the nearest large source was
the plantation at CATIE). Daily spore counts were analysed in relation to the speed
and direction of the wind at the three sites (Burt et al., 1998).
Assessment of inoculum
Disease surveys suggest that the infected plants in a plantation would act as a vast
reservoir of inoculum, but trapping data suggest that, whilst both ascospores and
conidia are windborne, they are not abundant in the air (particularly ascospores).
Ten banana leaves showing necrosis on at least 16% of the surface of the leaf
(using the modified Stover scale to assess the levels of infection) were selected at
random from the CATIE plantation during two periods: October 1993–February 1994
and April–September 1995. Each leaf was divided into three parts, each with different
amounts of necrosis (Figure 2). Necrotic tissue from each part was excised, the area
measured and the perithecia counted. A regression equation relating the number of
perithecia to the necrotic area was calculated (Burt et al., 1999).
A random selection of the leaf sections was exposed to a regime of wetting and
drying under simulated field conditions (21°C and 100% relative humidity and 28°C
at 60% relative humidity) in order to assess spore release (Burt et al., 1999). The
released spores were deposited on agar in Petri dishes and counted using a
compound microscope. A second regression equation was derived, relating the number
of ascospores to perithecia in infected leaf tissue (Burt et al., 1999).
Plant
Area A
Area B
Area C
1/3 leaf length
1/3 leaf length
Typical leaf dimensions: 50 - 120 cm long, 25 - 50 cm wide
Figure 2. Leaf parts used for counts of perithecia. (From: Rutter et al., 1998).
115
MyLsd 17x24
25/06/03
15:15
Page 116
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Viability of dispersed spores
Spore viability under conditions of temperatures and solar radiation likely to be
encountered during atmospheric transport was investigated (Parnell et al., 1998).
Ascospores released from wetted leaf sections were exposed to simulated sunlight
with UV wavelengths between 270 and 800 nm typical of the tropics, and
incubated under a light/dark cycle at 24-26°C (Parnell et al. 1998). Control plants
were subjected to the same regime minus the UV radiation.
Results and discussion
Dispersal of spores within a plantation
Spore numbers at all sampling heights were generally low and the majority were
found at sunrise, the time of dew evaporation (Table 1). More conidia than
ascospores were caught at all heights within the plantation. Spores were also present
after rain in the afternoon, again in low numbers (Burt et al., 1997). More conidia
than ascospores were found at all heights within the plantation (Table1).
The number of spores that got into the air was much smaller than the one expected
from looking at disease incidence alone. These low numbers made it impossible to
investigate the relationship between spore capture and weather, especially rainfall
which was low during sampling. It was also unclear whether the larger number of
conidia in the air is atypical or indicates that dispersal is mainly by conidia. At a
lowland site, Gauhl (1989) had found more ascospores than conidia. The different
results obtained at CATIE suggest that ascospore trapping may not be a reliable
method to forecast black leaf streak disease, at least at highland sites.
Table 1. Summary of spore trapping of M. fijiensis within an infected plantation, December 1993–February 1994.
Position relative
to the canopy
Ascospores
Mean spore count
Number
(min-max)
of samples
Conidia
Mean spore count
(min-max)
Number of
samples
Above
1.35 (0-6)
40
5.75 (0-54)
40
Middle
1.15 (0-63)
1.18 (0-5)
39
40
6.59 (0-63)
7.65 (0-44)
39
40
Bottom
1.20 (0-8)
40
8.52 (0-64)
40
From Burt et al., 1997
Small-scale dispersal of ascospores and conidia
More conidia than ascospores were caught in the rotorod traps within and just
above the plants in the small plot, but catches were again low (Figure 3). Initially,
it appeared that many more ascospores than conidia were recorded in the Burkard
spore trap but when the data were for sampling volume , the Burkard trap caught
fewer spores overall (Rutter et al., 1998). There was no evidence that spores entered
the plantation from outside and caused disease.
116
25/06/03
15:15
Page 117
Session 2
P.J.A. Burt
Disease progress across the small plot was uneven; plants in separate quadrats
showed different rates of disease progression (Figure 4). Wind direction was not
consistent but northerly winds were the most frequent. At these times, plants in
quadrat 1 were directly downwind of the inoculum source, which might account
for the higher spore catches there (Figure 3). Otherwise, wind direction and the
appearance of symptoms were not related (Rutter et al., 1998). Southerly winds
were the second most common and would have blown inoculum over quadrat 3.
However neither the pattern of disease spread nor the rotorod catches reflected
this. Southerly winds blowing across quadrat 1 might explain the high spore
catches by rotorods in quadrat 4. Patterns of disease spread in the small plot
showed no evidence of splash dispersal. Importantly, symptoms were not seen in
the plants immediately surrounding the source of innoculum until the middle of
June, when the upper leaves of the plants were affected. Rotorod traps 50 m
downwind of the plot had no ascospores, and only a few conidia (up to 12 per
sampling run) (Rutter et al., 1998).
180
150
Catch
120
Ascospores
90
Conidia
60
30
0
R1&2 (centre)
R3&4 (Q2/Q3)
R5&6 (Q4)
R7&8 (Q1)
Trap number and quadrat
Figure 3. Total spore catch in the small plot during the sampling period 1 May-3 August 1995. (From: Rutter
et al., 1998).
% of infected plants
MyLsd 17x24
150
Q3
Q4
Q1
Q2
100
50
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20
Sample number (every 4 days starting 2/5/95)
Figure 4. Appearance of disease symptoms in the quadrats of the small plot. (From: Rutter et al., 1998).
Mesoscale dispersal
Ascospores and conidia of M. musicola and M. fijiensis were present in the air at
all sites, usually between 0530-0830 in the morning with an abrupt cut-off at 08.30
117
MyLsd 17x24
25/06/03
15:15
Page 118
Mycosphaerella leaf spot diseases of bananas: present status and outlook
(Burt et al., 1998). Many more spores of M. fijiensis than of M. musicola were caught.
The analysis was therefore based on M. fijiensis only (Table 2). Despite the apparent
relationship between spore numbers in traps 1 and 3 (Table 2) the numbers of spores
caught in each trap were not statistically related to each other (Burt et al., 1998).
No differences in wind direction were recorded at the three trapping sites. The
winds blew most frequently from north to south, or vice versa, along the valley
axis, as might be expected (Burt et al., 1998). Winds blowing from between the
north-east and south-east would have crossed over trap 3 and the large plantation
at CATIE, the nearest large source. But there was no evidence that the spores were
being blown directly from this source, either directly or after allowing for a lag
between times of release and capture (Burt et al., 1998).
There were no other large sources of inoculum near trap 3. However, there are
many large plantations on the Atlantic coast, 40 km east of the trapping sites.
NE and SE winds could have blown spores from the plantations to the study area
but the cut-off in spore captures around 0830 in the morning does not support
this. It is possible that spores were descending at night at the time when the
nocturnal inversion would have been breaking up.
Table 2. Mean daily spore catches of M. fijiensis during the sampling period of 29 April–30 August 1995.
Mean daily catches of ascospores
(standard error of the mean)
Mean daily catched of conidia
(standard error of the mean)
n
Trap 1
153.5
(14.6)
72
(12.1)
116
Trap 2
124.2
(17.4)
8.5
(1..2)
127
Trap 3
144.6
(16.4)
8.2
(1..2)
118
From: Burt et al., 1998
Assessment of potential innoculum
The numbers of perithecia was related to the area of necrotic leaf tissue at different
stages of disease development (Burt et al., 1999):
ln(perithecial number) = [1.173 x ln(necrotic area)] + 4.624
A second regression equation was calculated from data from the release and rewetting
experiments (Burt et al., 1999):
ln(no. ascospores) = [1.173 x ln(necrotic area)] + 6.128
where necrotic area is measured in cm2.
This gave a mean of 4.5 ascospores per perithecium, a result which does not resolve
the ambiguities surrounding the number of ascospores present in each perithecium
reported in the literature, with values ranging from between 1 and 27 (Stahel, 1937;
Stover, 1972) and up to 160 (Meredith and Lawrence, 1970; Stover, 1970) being
reported. Even if there were some errors in the counting of the released ascospores,
it is clear that the number of ascospores available for release is not being represented
in spore trap catches.
118
MyLsd 17x24
25/06/03
15:15
Page 119
Session 2
P.J.A. Burt
Ascospore viability
Exposure to UV radiation at tropical sunlight levels killed ascospores after six hours
continuous exposure (Parnell et al., 1998). Although spores in the air are unlikely
to be exposed to continuous sunlight or temperature for these lengths of time, this
gives some indication of the potential for long-distance transport of viable spores.
A 12-hour transport time in winds of, for example, 10 m/s, would permit transport
over distances up to 400 km. This might not preclude the transport of viable spores
over that distance. In Costa Rica, however, the prevailing winds are off the sea on
both coasts, so windborne transport towards the Caribbean is unlikely. This
combination of unfavourable winds and viability may explain why black leaf streak
disease has not reached the Caribbean in significant amounts, and why localised
disease outbreaks there have not spread. Looking at spore viability in relation to
distance (Burt, 1994), suggests that wind dispersal of M. fijiensis ascospores over
longer distances is unlikely. For example, a trans-Atlantic transport time of even 5
days–which is probably the shortest period possible (Rosenberg and Burt, 1999)–is
significantly longer than the time spores are likely to remain viable, even assuming
that they were carried under cloudy skies all the time.
Suggestions for future research
It is clear that more research is required if the full contribution of airborne dispersal to
the epidemiology of M. fijiensis is to be fully understood. It is clear that much more
inoculum is being produced on the leaves than is entering the air. Consequently, there
is a need to understand more fully the relationship between wind patterns and spore
movement on all scales, but particularly updraughts in relation to spore movement inside
and out of a canopy.
Detailed studies of the leaf-surface are also required, to measure runoff (the number
of spores transported and their destination) and plant architecture and micrometeorology.
Rather than relying on rewetting experiments, which may be prone to error, detailed
histological investigations of necrotic tissue at various stages of the disease may also
reveal more information about the supply of inoculum (especially the number of
ascospores).
Finally, a more complex field investigation is needed to resolve whether or not spores
actually travel over long distances. This should be undertaken in association with a
study to determine the effect of natural environmental conditions on the viability of
spores found at various distances from known sources.
Acknowledgements
The material summarised in this paper was prepared over a period of seven years and involved
many people. Specifically, the support and assistance of Dr Elkin Bustamente, Principal
Pathologist, CATIE, Costa Rica, is most gratefully acknowledged. The following also made
valuable contributions to the research: John Rutter, Herbert Gonzales, Francisco Ramirez,
Kate Wilson, Mark Parnell, Margaret Smith, Jane Rosenberg, Sheila Green, John Sherington
and Philip Shannon. Funding was provided by the Crop Protection Programme of the
Department of International Development of the United Kingdom, who can accept no
responsibility for any information provided, nor views expressed.
119
MyLsd 17x24
25/06/03
15:15
Page 120
Mycosphaerella leaf spot diseases of bananas: present status and outlook
References
Bashi E. and D.E. Aylor 1983. Survival of detached sporangia of Peronospora destructor and
Peronospora tabacina. Phytopathology 73:1135-1139.
Bowden J., P.H. Gregory and C.G. Johnson. 1971. Possible wind transport of coffee rust across
the Atlantic Ocean. Nature 229:500-501.
Burdekin D.A. 1960. Wind and water dispersal of coffee leaf rust in Tanganyika. Kenya Coffee,
212-213.
Burt P.J.A. 1994. Windborne dispersal of Sigatoka leaf spot pathogens. Grana 33:108-111.
Burt P.J.A., J. Rutter and H. Gonzales. 1997. Short-distance windborne dispersal of the fungal
pathogens causing Sigatoka diseases in banana and plantain. Plant Pathology 46:451-458.
Burt P.J.A., J. Rutter and F. Ramirez. 1998. Airborne spore loads and mesoscale dispersal of
the fungal pathogens causing Sigatoka diseases in banana and plantain. Aerobiologia
14:209-214.
Burt P.J.A., L.J. Rosenberg, J. Rutter, F. Ramirez and H. Gonzales. 1999. Forecasting the airborne
spread of Mycosphaerella fijiensis, a cause of black Sigatoka disease on bananas:
estimations of numbers of perithecia and ascospores to aid forecasting. Annals of Applied
Biology 135:367-377.
Chuang T.Y. and H.J. Su. 1988. Physiological study of Fusarium oxysporum f.sp. cubense.
Memoirs of the College of Agriculture, National Taiwan University 28:19-26.
Davies J.M., C.E. Main and W.C. Nesmith. 1985. The biometeorology of blue mold of tobacco:
Part 2, the evidence for long-range sporangiospore transport. Pp. 473-498 in Movement
and dispersal of agriculturally-important biotic agents (Mackenzie, Barfield, Kennedy,
Berger and Taranto, eds).
Gauhl F. 1989. Untersuchingen zur Epidemiologie und Okologie der Schwarzen SigatokaKrankheit (Mycosphaerella fijiensis Morelet) an Kochbananen (Musa sp.) in Costa Rica.
PhD thesis, University of Gottingen, Germany.
Leach R. 1941. Banana leafspot (Mycosphaerella musicola) the perfect stage of Cercospora
musae Zimm. Tropical Agriculture Trinidad 18:91-95.
Leach R. 1946. Banana leafspot (Mycosphaerella musicola) on the Gros Michel variety in
Jamaica. Kingston, Jamaica: The Government Printer.
Maddison A.C. and J.G. Manners. 1972. Sunlight and viability of cereal rust uredospores.
Transactions of the British Mycological Society 59:429-443.
Meredith D.S. 1962. Some components of the air-spora in Jamaican banana plantations. Annals
of Applied Biology 50:577-594.
Meredith D.S. and J.S. Lawrence. 1970. Black leaf streak disease of bananas (Mycosphaerella
fijiensis): symptoms of disease in Hawaii, and notes on the conidial state of the causal
fungus. Transactions of the British Mycological Society 52:459-476.
Meredith D.S., J.S. Lawrence and I.D. Firman. 1973. Ascospore release and dispersal in black
leaf streak of bananas (Mycosphaerella fijiensis). Transactions of the British Mycological
Society 60:547-554.
Parnell M., P.J.A. Burt and K. Wilson. 1998. The influence of exposure to ultraviolet radiation
in simulated sunlight on ascospores causing black Sigatoka disease of banana and plantain.
Int. J. Biometeorol. 42(1):22-27.
Pedgley D.E. 1982. Windborne pests and diseases. Chichester: Ellis Horwood.
Rosenberg L.J. and P.J.A. Burt. 1999. Windborne displacements of Desert Locusts from Africa
to the Caribbean and South America. Aerobiologia 15:167-175.
Rotem J. and H.J. Aust. 1991. The effect of ultraviolet and solar radiation and temperature
on survival of fungal propagates. Journal of Phytopathology 133:76-84.
120
MyLsd 17x24
25/06/03
15:15
Page 121
Session 2
P.J.A. Burt
Rotem J., B. Wooding and D.E. Aylor 1985. The role of solar radiation, especially ultraviolet,
in the mortality of fungal; spores. Phytopathology 75:510-514.
Rutter J. and P.J.A. Burt. 1997. An assessment of the levels of inoculum of Mycosphaerella
fijiensis in a banana plantation. Pp. 229-242 in Aerobiology (S.N. Agashe ed.), Oxford
and IBH Publishing Co. Pvt Ltd, New Delhi.
Rutter J., P.J.A. Burt and F. Ramirez. 1998. Movement of Mycosphaerella fijiensis spores and
Sigatoka disease development on plantain close to an inoculum source. Aerobiologia
14:201-208.
Setlow R.B. 1974. The wavelength in sunlight effective in producing skin cancer: a theoretical
analysis. Proceedings of the National Academy of Science 71:3363-3366.
Stahel G. 1937. Notes on Cercospora leaf spot of bananas (Cercospora musae). Tropical
Agriculture Trinidad 14:2587-264.
Stover R.H. 1968. Leaf spot of bananas caused by Mycosphaerella musicola: perithecia and
sporodochia production in different climates. Tropical Agriculture Trinidad 45:1-12.
Stover R.H. 1970. Leaf spot of bananas caused by Mycosphaerella musicola: role of conidia
in epidemiology. Phytopathology 60:856-860.
Stover R.H. 1972. Banana, plantain and abaca diseases. Kew: Commonwealth Mycological
Institute.
Stover R.H. and N.W. Simmonds. 1987. Bananas. Harlow, England: Longman
121
MyLsd 17x24
25/06/03
15:15
Page 123
Session 2
J. Carlier et al.
Genetic differentiation in
Mycosphaerella leaf spot pathogens
J. Carlier1, H. Hayden2, G. Rivas3, M.-F. Zapater1, C. Abadie1 and E. Aitken4
Abstract
Black leaf streak disease and Sigatoka disease, caused respectively by two related fungi,
Mycosphaerella fijiensis and M. musicola, are the most important leaf spot diseases of bananas.
Understanding the genetic structure of the populations and the evolution of these pathogens is
an important aid in breeding and managing disease resistance. The population structure of each
pathogen was analysed using molecular markers mainly at the global and continental scales.
Features common to both were observed: 1) the centre of diversity is located in Southeast Asia
and founder events accompanying the introduction of the pathogens in other regions led to a
reduction of genetic diversity; 2) genetic diversity is maintained in all populations and is also present
at the scale of the plant; 3) genetic recombination played an important role in the genetic structure
of both pathogens; 4) genetic differentiation exists between populations from the global to the
local level. The main difference between the two species had to do with the measures of genetic
differentiation. Whereas the African populations of M. fijiensis were significantly different from
the Latin American/Caribbean ones, no significant difference was observed between the African
and Latin American/Caribbean populations of M. musicola.This suggests independent introductions
of M. fijiensis but not of M. musicola. Except for this situation, the genetic differentiation observed
between populations at the global and continental scales indicate an important effect of genetic
drift and limited gene flow.
Resumen - Diferenciación genética en los patógenos de la mancha foliar Mycosphaeralla
La enfermedades de la mancha foliar de los bananos más importantes se deben a dos hongos
relacionados: Mycosphaerella fijiensis y M. musicola, los agentes causales de la enfermedad de la
raya negra de la hoja y de la enfermedad de Sigatoka, respectivamente. El entendimiento, tanto
de la estructura genética de la población, como de la evolución de estos patógenos proporciona
información importante para brindar asistencia al mejoramiento y manejo de la resistencia a la
enfermedad. La estructura de la población de ambos patógenos fue analizada utilizando
marcadores moleculares a escalas global y continental. Se observaron las siguientes características
comunes: 1) el centro de la diversidad está localizado en el Sudeste de Asia y los eventos de
1
CIRAD, Montpellier, France
Institute of Land and Food Resources, Victoria, Australia
3 CATIE, Costa Rica
4
CRCTPP, Brisbane, Australia
2
123
MyLsd 17x24
25/06/03
15:15
Page 124
Mycosphaerella leaf spot diseases of bananas: present status and outlook
colonización que acompañaron la introducción de los patógenos en otras regiones han llevado a
una reducción de la diversidad genética en comparación con el Sudeste de Asia; 2) la diversidad
genética se mantiene en todas las poblaciones y se distribuye a escala de la planta, 3) la
recombinación genética desempeña un papel importante en la estructura genética de ambos
patógenos; 4) existe una diferenciación genética entre las poblaciones a escalas de global a local.
La principal diferencia observada es la existencia de una diferenciación genética entre las
poblaciones africanas y las poblaciones latinoamericanas y caribeñas de M. fijiensis pero no de
M. musicola. Este resultado sugiere introducciones independientes de M. fijiensis pero no de
M. musicola. Con excepción de la situación descrita arriba, la diferenciación genética observada
en ambos patógenos entre las poblaciones a escalas global y continental indica un efecto
importante de la genética y un flujo de genes bajo.
Résumé - Différenciation génétique chez les Mycosphaerella pathogènes
La maladie des raies noires et la maladie de Sigatoka, respectivement causées par deux
champignons apparentés, Mycosphaerella fijiensis et M. musicola, sont les maladies foliaires les
plus importantes chez le bananier. La compréhension de la structure génétique des populations
et de l’évolution de ces pathogènes représente une aide importante pour l’amélioration et la gestion
de la résistance à ces maladies. La structure de la population de chaque pathogène a été analysée
en utilisant des marqueurs moléculaires, principalement à l’échelle globale et continentale. Des
caractéristiques communes aux deux pathogènes ont été observées : 1) leur centre de diversité
est localisé en Asie du Sud-est et des événements fondateurs accompagnant l’introduction des
pathogènes dans d’autres régions ont conduit à une réduction de la diversité génétique ; 2) la
diversité génétique est maintenue dans toutes les populations et est également présente à l’échelle
de la plante ; 3) la recombinaison génétique a joué un rôle important dans la structure génétique
des deux pathogènes ; 4) une différentiation génétique existe entre populations, du niveau global
au niveau local. La principale différence entre les deux espèces concerne les niveaux de la
différentiation génétique. Alors que les populations africaines de M. fijiensis sont significativement
différentes de celles d’Amérique latine/Caraïbes, aucune différence significative n’a été observée
entre les populations de M. musicola originaires d’Afrique et d’Amérique latine/Caraïbes. Ceci suggère
des introductions indépendantes de M. fijiensis, mais pas de M. musicola. A part cette situation,
la différentiation génétique observée entre les populations à l’échelle globale et continentale indique
un effet important de la dérive génétique et des flux géniques limités.
Introduction
Black leaf streak disease, caused by Mycosphaerella fijiensis, and Sigatoka disease,
caused by M. musicola, are the most important leaf spot diseases of bananas (Jones,
2000). The fungi are haploid and heterothallic. The anamorph and teleomorph stages
are both present on infected leaves, and the ascospores produced during the sexual
stage play an important role in epidemics. The first species to be described was
M. musicola, in Java in 1902. The rapid dissemination of Sigatoka disease round the
world in the 1930s, led to speculations that the spores were carried by air currents
between continents: from Asia to the Pacific, from the Pacific to Australia, from
Australia to Africa and from Africa to Latin America (Stover, 1962). In 1962,
M. musicola was present in all banana-producing regions, making Sigatoka disease
one of the most important plant diseases. Black leaf streak disease was reported in
Fiji in 1964 and since then has been reported throughout the Pacific and Asia. The
chronology of records suggests that M. fijiensis originated, as M. musicola, in
Southeast Asia (Mourichon and Fullerton, 1990), which is also the centre of origin
124
MyLsd 17x24
25/06/03
15:15
Page 125
Session 2
J. Carlier et al.
of the host genus Musa. Starting in the 1970s, M. fijiensis spread to Africa, Latin
America and the Caribbean. Being more aggressive, M. fijiensis replaced M. musicola
as the dominant leaf spot pathogen in many areas. Although the distribution of both
pathogens is well documented in Australia, the Pacific region, Africa, Latin America
and the Caribbean, it is still not well understood in Southeast Asia.
Why and how to analyse populations of pathogens?
Knowledge of the genetic population structure and evolution of the pathogens is an
important aid in breeding and managing disease resistance. The main objective of
such study is to provide information on the level and distribution of variability.
Molecular markers are often used to analyse population structure at different
geographical scales. It should make it possible to identify potential sources of
resistance, which are expected to be in areas where the diversity of pathogens and
host are high. It should also ensure that the diversity of pathogens used when screening
for resistance is representative of the one in the regions where resistant hosts are
intended to be used. Pathogens can evolve to break down total resistance or erode
partial resistance. The evolution of pathogen populations depends on mutation,
recombination, genetic drift, gene flow and the selection pressure exerted by the host.
It should be possible to limit and restrict the evolution of pathogenicity by varying
host resistance in space and time. Such strategies should improve the durability of
the types of resistance used and ensure the durability of the culture.
Another objective of pathogen population studies is to evaluate the relative
importance of the evolutionary factors on the evolution of pathogens. Such
information would make it possible to model and test the effect of different
management strategies on the evolution of the pathogen. Analysing population
structure through space allows us to evaluate the effects of genetic recombination,
genetic drift and gene flow on the evolution of the pathogen. This paper reviews the
results obtained at global and local scales for M. musicola and M. fijiensis. A second
approach consists in studying the evolution of the pathogen in fields of resistant hosts
by using molecular markers and by characterizing pathogenicity. This allows us to
evaluate the effect of the selection pressure exerted by the host on the pathogen. This
second approach is described in another paper (see Abadie et al. in this volume).
Global population structure
RFLP markers were used to study the genetic structure of the global population of
M. musicola and of M. fijiensis (Carlier et al., 1996; Hayden et al., in prep.) Random
single-locus probes were used on samples from Southeast Asia, Australia, the Pacific
Islands, Africa, Latin America and the Caribbean. Features common to both pathogens
were observed.
Southeast Asia has the highest level of gene diversity (Table 1) and the majority
of alleles found in this region were also present in the other regions. This supports
the hypothesis that the pathogens originated in Southeast Asia. Founder events
accompanying the introduction of the pathogens to other regions have led to a reduction
in genetic diversity in comparison with Southeast Asia. Nevertheless, genetic diversity
125
MyLsd 17x24
25/06/03
15:15
Page 126
Mycosphaerella leaf spot diseases of bananas: present status and outlook
is maintained in all populations. Ecological conditions being favourable for disease
development and banana cultivation almost year round in most growing areas, low
genetic drift in large pathogen populations can maintain the high levels of genetic
diversity observed. Therefore, a certain level of variability in pathogenicity might also
be maintained in pathogen populations, allowing the pathogen to attack newly
introduced resistant genotypes, as has been observed with M. fijiensis on ‘Paka’ and
‘T8’ in Rarotonga, Cook Islands (Fullerton and Olsen, 1995). The existence of specific
interactions between the host and M. fijiensis isolates was suggested for highly resistant
genotypes (Fullerton and Olsen, 1995). Variability in aggressiveness was evaluated for
two M. fijiensis samples from Cameroon and the Philippines by inoculating five partially
resistant cultivars using a leaf piece assay (El Hadrami, 2000). Variability was similar
but low for both countries, however genetic diversity in the Philippines was much higher
(Carlier et al., 1996). Specific interactions between the isolates and the cultivar were
not detected. Only susceptible hosts are cultivated in these countries, and the lack of
a selection pressure being exerted by the host on the pathogens could explain the results.
The potential of pathogen populations to overcome partial resistance should be analysed
by following their evolution in fields of resistant hosts (see Abadie et al., this
proceedings).
Genetic recombination plays an important role in the genetic structure of
M. musicola and M. fijiensis. Genetic markers were statistically independent therefore
characteristics of pathogenicity could not be related to RFLP genotypes. With regards
to breeding programmes, introducing specific resistance genes in individual cultivars
(pyramiding) may not be a strategy for durable resistance in banana. Mixing varieties
or using partially resistant hosts might be more appropriate.
Table 1. Nei’s gene diversity estimates for populations of M. fijiensis (Carlier et al., 1996) and M. musicola (Hayden
et al., in prep.) from different geographical regions.
Species
Asia
Africa
M. fijiensis
M. musicola
0.57
0.41
0.25
0.27
Latin America and Caribbean
0.40
0.21
Australia and Pacific
0.28
0.33
A high level of genetic differentiation was observed between populations of
M. musicola and M. fijiensis from different regions (Figure 1). The Fst parameter (Wright,
1951) estimated for all loci between pairs of populations varied between 0.14 and 0.58
for M. fijiensis and between 0.025 and 0.55 for M. musicola. But whereas the African
populations of M. fijiensis were significantly different from the Latin
American/Caribbean ones (Fst = 0.49), no significant difference was observed between
the African and Latin American populations of M. musicola (Fst = 0.025, not significant).
This suggests a separate introduction of M. fijiensis in each region but a common one
for M. musicola.
On the other hand, the high levels of genetic differentiation observed between
Australian and African populations of M. musicola (Fst = 0.47) does not support the
hypothesis of Stover (1962) whereby spores of M. musicola were carried by air currents
from Australia to Africa. In general, the high level of genetic differentiation of both
pathogens at a global scale suggests occasional migration events between continents.
126
MyLsd 17x24
25/06/03
15:15
Page 127
Session 2
J. Carlier et al.
Long distance dissemination of the disease around the world was more likely to have
occurred by movement of infected plant material, as proposed by Mourichon and
Fullerton (1990).
Indonesia
Latin America
Latin America/
Caribbean
Africa
Africa
Pacific
Islands
Philippines
A
Papua New Guinea
Australia
B
Figure 1. Global population structure of M. fijiensis (Carlier et al., 1996) and M. musicola (Hayden et al., in prep.).
Additive trees constructed from estimates of Wright’s Fst among pairs of geographical populations.
Continental and local population structures
Population structures at a continental scale were studied in Australia for
M. musicola (Hayden, 2000). Collections of isolates from twelve sites along the
east coast were analysed using 15 RFLP markers. The level of gene diversity (Nei,
1978), varied between 0.14 and 0.37. On a plant, the pathogen isolated from a
given lesion would often be a haplotype not found in the other lesions, meaning
that diversity is also present at a fine scale. Low to high levels of genetic
differentiation were observed between populations (Fst = 0.04-0.45). There was
no apparent correlation between genetic and geographical distances as high levels
of genetic differentiation were observed between neighbouring populations and
low levels were observed in populations separated by long distances.
The population structure of M. fijiensis was analysed in Africa, Latin America
and the Caribbean (Rivas et al., subm.). Samples from different countries were
characterized using CAPS (Cleaved Amplified Polymorphic Sequence) markers
(Zapater et al., subm.). The results obtained for both continents were similar. The
value of gene diversity varied between 0.19 and 0.38 for Africa, and 0.16 and
0.36 for the Latin America/Caribbean region. The low levels detected in some
populations suggest that founder effects occurred during the spread of the disease
on both continents. In the Latin America and Caribbean region, the highest levels
are observed in populations from Honduras and Costa Rica, supporting the
suggestion that the pathogen entered the continent in this area. In one locality
in Cameroon, the values of gene diversity estimated at the scale of the field and
of the plant are similar suggesting, as with M. musicola, that diversity is distributed
at a fine scale.
Important levels of genetic differentiation were detected between most of the
populations (Fst = 0.04-0.45 for Africa and 0.01-0.56 for Latin America/Caribbean).
127
MyLsd 17x24
25/06/03
15:15
Page 128
Mycosphaerella leaf spot diseases of bananas: present status and outlook
There is sufficient differentiation between populations in the Caribbean islands
to support the hypothesis that there was more than one introduction from Latin
America. In Cameroon, genetic differentiation was detected between localities 300
km apart but not between localities 200 km apart.
Finally, the genetic structure of M. fijiensis was studied in the Australasia/
Pacific region using RFLP markers (Hayden et al., subm.). Genetic differentiation
was detected between the Torres Straits Islands, Pacific Islands and Papua New
Guinea. At a local scale, there was no differentiation between two sites in the
small Mer island.
The levels of genetic differentiation observed at the continental scale for both
pathogens suggest an important effect of genetic drift on population structure
and limited gene flow. Thus, spread of the diseases within continents could be
due to the movement of infected plants and/or very restricted ascospore dispersal.
As populations are probably not in genetic equilibrium, gene flow resulting from
ascospore dispersal may be underestimated. However, preliminary results from
an epidemiological study of black leaf streak disease suggests that dispersal of
the pathogen is more restricted than previously thought. The results suggest a
dispersal gradient of about 25 m from an inoculum source (Abadie et al., this
proceedings).
Conclusion and perspectives
The population structures of M. musicola and M. fijiensis are now better known
at different geographical scales. However, at a regional scale few samples from
Southeast Asia have been analysed. A new pathogen, Mycosphaerella eumusae,
was recently discovered and detected mainly in Southeast Asia (Carlier et al., 2000).
Southeast Asia is not only the centre of origin of all three pathogens but also of
the host genus Musa. The distribution of the pathogens and their population
structure should now be determined in detail for this region. Host-pathogen
interactions could differ for each pathogen. One hypothesis to explain the
continued presence of the three pathogens in Southeast Asia is the high diversity
of host species. The hypothesis could be tested by surveying the fungal species
in relation to host diversity. If host-pathogen interactions differ, the resistance
genes introduced to produce new varieties could be more or less efficient
depending on the pathogen they are exposed to. Their utilization should take into
account the distribution of pathogen species. Zones of co-evolution for the three
pathogens could be localized in Southeast Asia. This area is a potential source
of resistance, therefore a study of pathogen populations in natural systems should
provide us with information to complement evaluations of the relative importance
of the different evolutionary forces.
The results to date suggest that genetic drift has an important effect on the
structure of pathogen populations and that gene flow is limited. The limit of
ascospore dispersal should be estimated indirectly at a country scale using genetic
models such as the isolation by distance model (Rousset, 1997). However, we can
already predict that the improvement of quarantine measures at the continental
scale might limit the risk of introducing the disease to new areas, and limit the
128
MyLsd 17x24
25/06/03
15:15
Page 129
Session 2
J. Carlier et al.
exchange between existing pathogen populations from different countries. At a
country and local level, geographical obstacles could also limit exchange between
populations from different fields by playing on gene flow. Such a measure could
have an impact on the durability of the resistances and of the management
strategies.
References
Carlier J., M.F Zapater, F. Lapeyre, D.R. Jones and X Mourichon. 2000. Septoria leaf spot of
banana: A newly discovered disease caused by Mycosphaerella eumusae (anamorph
Septoria eumusae). Phytopathology 90(8):884-890.
Carlier J., M.H. Lebrun, M.F. Zapater, C. Dubois and X. Mourichon. 1996. Genetic structure
of the global population of bananas black leaf streak fungus Mycosphaerella fijiensis.
Molecular Ecology 5:499-510
El Hadrami A. 2000. Caractérisation de la résistance partielle des bananiers à la maladie des
raies noires et évaluation de la variabilité de l’agressivité de l’agent causal, Mycosphaerella
fijiensis. Thèse d’Université. Faculté Universitaire des Sciences Agronomiques de Gembloux,
Belgium. 153pp.
Fullerton R.A. and T.L. Olsen. 1995. Pathogenic variability in Mycosphaerella fijiensis Morelet
cause of black Sigatoka in banana and plantain. New Zealand Journal of Crop and
Horticultural Science 23:39-48.
Hayden H.L., J. Carlier and E.A.B Aitken. (In preparation). Population differentiation in the
banana leaf spot pathogen Mycosphaerella musicola, examined at a global scale.
Hayden H.L., J. Carlier and E.A.B. Aitken. (Submitted). The genetic structure of Mycosphaerella
fijiensis from Australia, Papua New Guinea and the Pacific Islands.
Hayden H.L. 2000. Population genetic studies of Mycosphaerella species infecting banana.
Thesis, University of Queensland, Australia.
Jones D. 2000. Diseases of Banana, Abacá and Enset. CAB International, Wallingford, UK.
Mourichon X. and R.A. Fullerton. 1990. Geographical distribution of the two species
Mycosphaerella musicola Leach (Cercospora musae) and M. fijiensis Morelet (C. fijiensis),
respectively agents of Sigatoka disease and black leaf streak disease in Bananas and
Plantains. Fruits 45:213-218.
Nei M. 1978. Estimation of average heterozygosity and genetic distances from a small number
of individuals. Genetics 89:583-590.
Rivas G. G., M.-F. Zapater, C. Abadie and J. Carlier. (Submitted). Founder effect and stochastic
dispersal at the continental scale of Mycosphaerella fijiensis, a tropical fungal pathogen
of bananas that has recently spread in Latin America, the Caribbean and Africa.
Rousset F. 1997. Genetic differentiation and estimation of gene flow from F-statistic under
isolations by distance. Genetics 145:1219-1228.
Stover R.H. 1962. Intercontinental spread of banana leaf spot (Mycosphaerella musicola Leach).
Tropical Agriculture Trinidad 39(4):327-338.
Wright S. 1951. The genetical structure of populations. Annals of Eugenics 15:323-354.
Zapater M.F., A. Rakotonantoandro, F. Cohen, J. Carlier. (Submitted). CAPS (Cleaved
Amplified Polymorphic Sequence) markers for the fungal banana pathogen Mycosphaerella
fijiensis.
129
MyLsd 17x24
25/06/03
15:15
Page 131
Session 2
C. Molina et al.
Development and application of
molecular markers in Mycosphaerella
populations in Colombia
C. Molina1,2, S. Aponte1, A. Gutiérrez1, V. Núñez1 and G. Kahl2
Abstract
In order to design effective strategies against the Mycosphaerella banana pathogens M. fijiensis
and M. musicola, it is essential to have information on genetic diversity and population
composition. Information to understand the population dynamics of these banana pathogens
was based on microsatellite markers. The present study reports tests on 48 primer pairs
designed for M. musicola, of which 26 proved to be polymorphic and four were transferable
to M. fijiensis. Based on microsatellites, a comparison was made of the genetic variability in
M. fijiensis and M. musicola populations from Colombia, Costa Rica and Venezuela. Dendograms
for each species were based on the Dice similarity algorithm and grouped with the UPGMA
clustering method. With the exception of a few isolates, most clusters coincided with the
geographical locations (sampling sites).
Resumen - Desarrollo y aplicación de los marcadores moleculares en las poblaciones
de Mycosphaerella en Colombia
Con el fin de diseñar estrategias eficaces contra los patógenos de Mycosphaerella, M. fijiensis
y M. musicola, en el banano, es esencial disponer de la información sobre la diversidad
genética y composición de sus poblaciones. La información, con ayuda de la cual se entienden
las dinámicas de las poblaciones de estos patógenos de banano, se basa en los marcadores
de microsatélites. El estudio actual presenta los informes de las pruebas de 48 pares de
iniciadores diseñados para M. musicola, de los cuales 26 resultaron ser polimórficos y cuatro
transferibles a M. fijiensis. Basándose en los microsatélites, se hizo la comparación de la
variabilidad genética en las poblaciones de M. fijiensis y M. musicola procedentes de Colombia,
Costa Rica y Venezuela. Los dendogramas para cada especie se basaron en el algoritmo de
similitud de Dice y se agruparon con el método de análisis de conglomerados UPGMA. Con
excepción de unos pocos aislados, la mayoría de los conglomerados coincidieron con las
localizaciones geográficas (sitios de muestreo).
1
2
CORPOICA, Bogotá, Colombia
University of Frankfurt, Frankfurt/Main, Germany
131
MyLsd 17x24
25/06/03
15:15
Page 132
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Résumé - Développement et application des marqueurs moléculaires chez des populations de Mycosphaerella en Colombie
Afin de développer des stratégies efficaces contre les Mycosphaerella pathogènes du bananier,
M. fijiensis et M. musicola, il est essentiel d’avoir des informations sur la diversité génétique
et la composition des populations. Les informations visant à comprendre la dynamique des
populations de ces pathogènes du bananier ont été basées sur les marqueurs microsatellites.
Cette étude présente les tests effectués avec 48 paires d’amorces conçues pour M. musicola,
dont 26 se sont avérées polymorphes et quatre ont été transférables à M. fijiensis. En se basant
sur les microsatellites, une comparaison de la variabilité génétique a été faite chez des
populations de M. fijiensis et M. musicola originaires de Colombie, du Costa Rica et du
Venezuela. Les dendrogrammes pour chaque espèce ont été basés sur l’algorithme de
similarité de Dice et groupés avec la méthode d’agrégation UPGMA. A l’exception de quelques
isolats, la plupart des agrégats coïncidaient avec les localisations géographiques (lieux de
collecte).
Introduction
Mycosphaerella fijiensis and M. musicola, the two most severe fungal pathogens
of plantain and banana, are the major cause of economic losses in commercial
plantations and in numerous smallholdings. Both pathogens spread around the
world through demographic events such as founder effects. Population bottlenecks
have been caused by increased doses of fungicides, the introduction of partly
resistant host varieties, and isolation by distance and geographical barriers between
populations.
M. musicola was first reported from Java in 1902 (Mourichon et al., 1997),
and from the Sigatoka Valley, Fiji in 1912 (Leach, 1941), where it caused an
epidemic. About 50 years later a more aggressive pathogen, M. fijiensis, was
detected in the same region (Leach, 1964). Both pathogens rapidly colonized the
South Pacific Islands, Asia, Africa and America (Stover, 1976). M. fijiensis has
been reported from sites where M. musicola was formerly present, suggesting a
gradual displacement of M. musicola to higher altitudes (inter-Andean valley
populations). Therefore, the dynamics of population structure of both pathogens
is in some way interdependent and most likely to be influenced by parameters
common to both.
Direct comparison of the populations of both pathogens would help understand
local and regional genetic diversity and differentiation, and the influence of
environmental pressures on the spread of the disease to new sites. It could also
help predict the behaviour of new epidemics.
Molecular markers have become important tools for the investigation of the
genetic composition of fungal populations (Groppe and Boller, 1997; Bucheli et
al., 2001). Restriction Fragment Length polymorphism (RFLP) markers were
developed for the M. fijiensis genome, and used to characterize populations of
M. fijiensis at a regional and global scale (Carlier et al., 1994, 1996; Müller et
al., 1997). More recently, simple sequence repeat (SSR) markers have been
established for M. fijiensis (Neu et al., 1999) and M. musicola (Molina et al., 2001)
which, together with other PCR-based DNA profiling methods, provide a new
method to compare the genetic diversity of both pathogens.
132
MyLsd 17x24
25/06/03
15:15
Page 133
Session 2
C. Molina et al.
In the present study, dendograms based on Dice similarity algorithms were obtained
for each species. With the exception of a few isolates that grouped outside the main
clusters, the majority of individuals grouped according to their geographical location.
Materials and methods
Sampling sites and fungal material
Eighty isolates of Mycosphaerella fijiensis and 64 of M. musicola were collected from
12 locations. Isolates from Colombia were from two adjacent locations on the
Caribbean coast (Santa Marta), and from three sites in the inner valleys of the Andes,
Caldas, La Mesa and Cachipay. Isolates from Costa Rica and Venezuela were included
for comparison. Costa Rican isolates were collected in “Valle Central” and those from
Venezuela were collected in Mérida, a mid-altitude region located in the eastern
branch of the Andes. There are two types of banana and plantain crop management
in these areas (i) extensive banana plantations that are typical of the Colombian
Atlantic coast and (ii) smallholdings that are characteristic of higher altitudes such
as “Valle Central” in Costa Rica and the inner Andean valleys of Venezuela and
Colombia (Price 1999).
DNA isolation
Infected plantains and banana leaves with advanced stages of lesion development
were transferred to a humid chamber to allow ascospores to discharge onto 1.5%
water agar. Single ascospores were identified microscopically, transferred to V8
medium and incubated for 12 days at 25ºC. DNA was isolated from mycelium using
a FastDNA Kit, Bio 101. DNA preparations were further purified by phenol:chloroform
extraction (24:1, v/v) and ethanol precipitation. After washing with 70% ethanol,
the final pellets were dissolved in an appropriate volume of 10 mM Tris-HCl, 1 mM
EDTA, pH 8.
SSR design for M. musicola
The single ascospore culture MmCol-LM9.5.1 (collected from plantations severely
affected by Sigatoka disease in a mid-altitude region of Colombia) was used as
source material for the construction of a genomic library. Fungal DNA was isolated
according to Weising et al. (1991) and purified by cesium chloride gradient
centrifugation. Microsatellite enriched libraries were constructed and screened
according to Fischer and Bachmann (1998). Detailed information about methodology, primer sequences and genebank accession numbers for the 26 polymorphic
M. musicola primers can be found in Molina et al. (2001).
Microsatellite analysis
PCR amplifications were performed in a Perkin Elmer 9700 thermocycler in 10 µl
reactions containing 5 ng of genomic DNA template, 0.5 µM of each forward and
133
MyLsd 17x24
25/06/03
15:15
Page 134
Mycosphaerella leaf spot diseases of bananas: present status and outlook
reverse primer, following PCR conditions according to Neu et al. (1999) and Molina
et al. (2001). Products were separated on 6% sequencing gels and autoradiographed
(Sambrook et al., 1989).
For microsatellite allelic data, an initial matrix containing allele sizes was
constructed, from which a 0/1 matrix was also derived. Dice similarity coefficient
matrices and their corresponding dendrograms, grouped by the UPGMA agglomerative method, were calculated with the NTSYS software package (Rohlf, 1993).
Results
SSR design for M. musícola
Out of 1029 colonies screened, 205 yielded a positive hybridization signal (enrichment
efficiency: 19.9%). Sixty-four clones were sequenced, and primers could be designed
for 48 clones. Primer functionality was tested on a set of
24 template DNAs comprising 18 M. musicola isolates from a single field in Colombia,
four isolates from Costa Rica and two from Mexico. The original clone served as a
positive control. Primer transferability was tested with three M. fijiensis isolates:
Mf-Col-LD9.1 (Colombia), Mf-Mex-015 (Mexico), Mf-PNG-294 (Papua New-Guinea).
A total of 48 primer pairs were tested and 26 yielded single polymorphic bands of
the expected size. The characteristics are summarized in Table 1.
Successful cross-priming with M. fijiensis DNA was observed at four loci. The
availability of polymorphic microsatellite markers specific for M. musicola makes
it possible to study the population structure of the pathogen in areas infested with
Sigatoka disease, and to compare the pathogens using the same marker system.
SSR typing
SSR markers were used to type all isolates from both species, nine for M. fijiensis
and eleven for M. musicola, showing high levels of polymorphism (Neu et al., 1999;
Molina et al., 2001). Although the markers proved to be highly informative for the
species for which they were developed, they were of limited value when transferred
to other species. For example, M. fijiensis marker Mf-SSR-061 resulted in
monomorphic patterns when used for typing M. musicola populations. The same was
true for M. musicola marker Mm-SSR-024 in M. fijiensis isolates. An average of 3.4
alleles were observed for M. fijiensis and 4.0 alleles for M. musicola. Polymorphic
loci were 80.0% and 90.9% for M. fijiensis and M. musicola respectively.
Cluster analysis based on Dice similarity index
Dendograms based on Dice similarity indexes were produced for each species. The
M. musicola tree was based on 42 SSR markers. Isolates grouped into six clusters
(I to VI in Figure 1), which mostly paralleled the geographic locations. For example,
in cluster I all isolates belonged to Cachipay and La Mesa, Colombia. These two
locations are close to each other, have similar weather conditions and similar crop
management practices. Cluster II mostly contained isolates from La Mesa, with some
134
MyLsd 17x24
25/06/03
15:15
Page 135
Session 2
C. Molina et al.
Table 1. Characteristics of microsatellites cloned from Mycosphaerella musicola.
Locus
Repeat of cloned allele
Primer sequences (5’ – 3’)
Mm SSR 01
(CA)9 G (CA)11
TAGTTGCAACCGAACAGG/ CTCCGTAGGTATGATGGTGT
Mm SSR 03
(GA)4 (GC)2 (GA)32
CTCCGTAGGTATGATGGTGT/ GCTTCGTCAAGACCCTTAC
Mm SSR 05
((GT)4 (C))3
CCTCTTACGAAGTCTGTGGT/ TATCTCGGGAGACCAGACTA
Mm SSR 06
(GA)8
CGAACAGGACGAAAGAATAG/ GTTTGTTCCAGTTCGTCAAG
Mm SSR 07
(CA)50
ACGAGGTTTCAGAAGCAATA/ TCTTTCACCGAAGAAACCT
Mm SSR 09
((GT)6 AT (GT)3 GCn(GT)5 (GC)
AGGGACGAACAAAACAGAG/ CCATGGTTTTCAAGCATATT
Mm SSR 10
(CA)30
GAGAGCATGAAAAGTGGAAA/ CGTGACACTCGTCAGTTACA
Mm SSR 14
(CA)7 CAA (CA)19
ATTTGGTGAATGGGGTAAG/ ACAGAGGGAAGCAAGTTTTT
Mm SSR 15
(CA)27
CTACTGAGGCAGTCGCTAAC/ GGAGAGGTGGAAAAAGAAGT
Mm SSR 16
(GA)6 AAA (GA)17
CCATCTGCCTTGAGATAGTC/ GAATTTATTCCAGCGAAGC
Mm SSR 18
(GA)n
ATCTGATTCGTATGGTGGAG/ TTGCTACTACTGGTGCTTCTC
Mm SSR 21
(CTT)9
GTCGACCTCCATGACACTC/ TGCATGCAATCTGTAACCT
Mm SSR 22
(GAA)9
CCAAAGCTTGAGTTGCTATT/ ACAACTTTTTGAGGAAAATGTAA
Mm SSR 23
(CTT)27
CGACCTAGTCGAGGATGATA/ CGAAGACTTCTGAAAGGTCA
Mm SSR 24
(GAA)2 GG (GAA)3GG (GAA)12
TCAAGAGGAGGAGAAGTTGA/ GGTTCTGATCAAGAGGAGGA
Mm SSR 26
(CAA)8
ATATCTCTTCGTGTTTTGCG/ AAGTGTGGTCACAGCAAGTT
Mm SSR 30
(CA)28
TGATGTTAAGTTGACGGACA/ CTAAGCCAAACCTCAATCAG
Mm SSR 31
(AC)27
AACCACATCTTCGATCAGG/ CACATGGAATATCCTTGGTC
Mm SSR 34
(CA)19
CTCGCTGCCTGATTATTCT/ AGATGGCATCGCTTCAC
Mm SSR 35
(CA)4 AA (CA)26
TAACAATGTCCCTGAGAAGC/ GCCTTATCTGGAAAGTATCGT
Mm SSR 36
(CA)13
ATTCCAGGTACGGCTACAC/ ATTCAGATCTGGTCTGGTTG
Mm SSR 38
((GT)n (CG))3
GAGAGTGAGATCGGTAGCAA/ CGGGATTAAGGTCTACCAA
Mm SSR 39
(CA)19
TGCGAATTCCATTGATATG/ CGTGTGCTGACGAGAGAT
Mm SSR 41
(GT)14
GGTGAGGTCGTTATTGTTGT/ GCTTTAGAGGTTTCGTTCTTC
Mm SSR 44
(CA)9 (CT)14
CCTCACTCTCGCTCATACA/ AGAATGGACGAAAAACACTG
Mm SSR 46
(CT)6 (GT)38
CGTGGACCTATTGTCAACTC/ TGGGTTACATTTACGAGAGAA
isolates from Costa Rica and one from Venezuela. For clusters III and IV, all isolates
were from Venezuela with each cluster having isolates from a single location. Cluster
III represents isolates from Mérida and cluster IV represents isolates from Santa Rosa.
Isolates from clusters V and VI were respectively from Costa Rica and Colombia.
The M. fijiensis tree, based on 42 polymorphic SSR markers, shows four clusters
(Figure 2). Cluster I comprises, almost exclusively, Santa Marta isolates, with a few
isolates from Costa Rica and one from Caldas (Colombia). Clusters II and IV mostly
comprise isolates from different locations in Costa Rica, with some isolates from
135
MyLsd 17x24
25/06/03
15:15
Page 136
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Colombia. Cluster III comprises 15 isolates from Caldas (Colombia), two isolates from
Costa Rica and four from Santa Marta (Colombia). The 80 isolates used represent 51
haplotypes. One particular haplotype was found in 10 isolates from Santa Marta
(Colombia).
0.00
0.25
Dice similarity index
0.50
0.75
1.00
M. musicola SSR
Figure 1. Dendogram for Mycosphaerella musicola populations, based on Dice similarity indexes. Colombian
populations are Mm-Col-CH: Clachipay and Mm-Col-LM: La Mesa; Venezuelan populations are Mm-Ven-Md:
Merida and Mm-Ven-SR: Santa Rosa; Costa Rican populations are Mm-CR-DE: el Descanso, Mm-CR-ME: Ma.
Eugenia and Mm-CR-QB: Quebrador.
Discussion
In the M. fijiensis and M. musicola dendograms, most clusters correspond to the
original sampling sites and show a correlation between clusters and discrete populations. Studies of genetic diversity at a worldwide level (Carlier et al., 1996) and a
136
MyLsd 17x24
25/06/03
15:15
Page 137
Session 2
0.00
C. Molina et al.
0.25
Dice similarity index
0.50
0.75
1.00
M. fijiensis SSR
Figure 2. Dendogram for Mycosphaerella fijiensis populations, based on Dice similarity indexes. Colombian
populations are Mf-Col-SM: Santa Marta and Mf-Col-CD: Caldas; Costa Rican populations are Mf-CR-DE: el
Descanso, Mf-CR-ME: Ma. Eugenia, Mf-CR-TR: Trsissia and Mf-CR-SR: San Rafael.
regional level in Africa (Müller et al., 1997) suggest that in most situations
colonization by M. fijiensis involves founder effects, a few individuals from an
original population representing the haplotypic pool of a derived population.
In the M. musicola tree, all but one of the Venezuelan isolates are grouped in two
discrete clusters (III and IV). This is particularly true for the isolates from Mérida, which
show very similar to identical DNA profiles. Only one Venezuelan isolate (Mm-VenMd-20) is in a separate group, with Colombian isolates from La Mesa (cluster II). M.
137
MyLsd 17x24
25/06/03
15:15
Page 138
Mycosphaerella leaf spot diseases of bananas: present status and outlook
musicola populations are mainly confined to mid-altitude regions of the Andes where
bananas are grown in smallholdings. These smallholdings are separated by deep valleys
and high mountains which are geographical barriers to gene flow. This may explain
the separation of Colombian and Venezuelan populations in the M. musicola tree, despite
being located in the same mountain range.
In the M. fijiensis tree, it is important to highlight that the majority of isolates from
the Atlantic coast (Santa Marta) are grouped in cluster I, whereas isolates from the midAndean region (Caldas) are in cluster III. Both regions are isolated from each other by
physical distance and geographical barriers that prevent gene flow between populations.
In addition, populations from the Atlantic coast are constantly under environmental
pressure (high dosages of fungicides) whereas Andean populations are rarely treated
with chemicals, keeping those areas as reservoirs of genetic diversity.
The colonization dynamics of M. fijiensis for Central America has been documented
(Unión de Países Exportadores de Banano, 1994). Colonization started from Honduras
and continued southwards to Costa Rica and Colombia. Black leaf streak disease was
first reported in 1981 in Urabá, Colombia, where it was confined for four years, until
a first outbreak occurred along the shores of the Atrato River and then spread to the
Atlantic coast and mid-Andean regions. Forty four isolates from Santa Marta (Atlantic
Coast) corresponded to only 23 haplotypes, whereas 16 isolates from Caldas (mid-Andes)
represented 10 haplotypes. This could be an indication of higher genetic diversity in
the Caldas populations. In the Atlantic coast of Colombia, the genetic consequences of
the founder effect could have been enhanced by the pressure exerted by high doses of
fungicide and strict regulations on the transport of plant material between populations.
M. fijiensis and M. musicola followed the same route, that is from Central America
to Colombia. In cluster II of each phenogram (Figure 1 and 2), Costa Rican isolates are
found in the same group with Colombian isolates. The genetic similarity of these two
distant populations is consistent with the fact that Central American populations are
ancestral and that the similarity cannot be explained by gene flow.
Acknowledgements
Plant Genetic Resources and Biotechnology (CORPOICA, Colombia) appreciates the
support from COLCIENCIAS (223-95). C. Molina appreciates a fellowship from UNESCO,
Paris (No. Sc-206.668.1) and Stiftung für Internationale Wissenschaftliche Zusammenarbeit
(Frankfurt/Main, Germany). We would like to thank Luisa Pérez for commenting on the
manuscript.
References
Bucheli E., B. Gautschi and J.A. Shykoff. 2001. Differences in population structure of the anther
smut fungus Microbotryum violaceum on two closely related host species, Silene latifolia
and S. dioica. Molecular Ecology 10:285-294.
Carlier J., X. Mourichon, D. Gonzalez-de-Leon and M.H. Lebrun. 1994. DNA restriction fragment
length polymorphisms in Mycosphaerella species that cause banana leaf spot diseases.
Phytopathology 84:751–756.
Carlier J., M.H. Lebrun, M.F. Zapater, C. Dubois and X. Mourichon. 1996. Genetic structure of
the global population of banana black leaf streak fungus, Mycosphaerella fijiensis.
Molecular Ecology 5:499–410.
138
MyLsd 17x24
25/06/03
15:15
Page 139
Session 2
C. Molina et al.
Fischer D. and K. Bachmann. 1998. Microsatellite enrichment in organisms with large genomes
(Allium cepa). Biotechniques 24:796–802.
Groppe K. and T. Boller. 1997. PCR assay based on a microsatellite-containing locus for detection
and quantification of Epichloe endophytes in grass tissue. Applied and Environmental
Microbiology 63:1543-1550.
Leach R. 1941. Banana leaf spot, Mycosphaerella musicola, the perfect stage of Cercospora musae
Zimm. Tropical Agriculture 18:91-95
Leach R. 1964. A new form of banana leaf spot in Fiji, black leaf streak. Wild Crops 16:60-64.
Molina C., D. Kaemmer, S. Aponte, K. Weising and G. Kahl. 2001. Microsatellite markers for
the fungal banana pathogen Mycosphaerella musicola. Molecular Ecology Notes 1:137-139.
Mourichon X., J. Carlier and E. Fouré. 1997. Sigatoka Leaf Spot Diseases. INIBAP Musa disease
fact sheet No.8. 4pp.
Müller R., C. Pasberg-Gauhl, F. Gauhl, J. Ramser and G. Kahl. 1997. Oligonucleotide
fingerprinting detects genetic variability at different levels in Nigerian Mycosphaerella
fijiensis. Journal of Phytopathology 145:25-30.
Neu C., D. Kaemmer, G. Kahl, D. Fischer and K. Weising. 1999. Polymorphic microsatellite markers
for the banana pathogen Mycosphaerella fijiensis. Molecular Ecology 8:523–525.
Price N. 1999. Highland Bananas in Colombia. INFOMUSA 8(2):26-28.
Rohlf J. 1993. Numerical Taxonomy and Multivariate Analysis System. Version 1.8.
Rozen S. and H.J. Skaletsky. 1997. Primer3. Available at http://www-genome.wi.mit.edu/
genome_software/other/primer3.html.
Sambrook J., E.F. Fritsch and T. Maniatis. 1989. Molecular Cloning: a Laboratory Manual.
2nd Edition. Cold Spring Harbor Laboratory Press, New York.
Stover R.H. 1976. Distribution and cultural characteristics of the pathogens causing banana
leaf spot. Tropical Agriculture 53:111-115.
Unión de Países Exportadores de Banano (UPEB). 1994. The main lines of the banana industry
in Latin America. INFOMUSA 5(1):14-19.
Weising K.B. Beyermann, J. Ramser and G. Kahl. 1991. Plant DNA fingerprinting with
radioactive and digoxigenated oligonucleotide probes complementary to simple repetitive
DNA sequences. Electrophoresis 12:159–169.
139
MyLsd 17x24
25/06/03
15:15
Page 141
Session 2
L. Conde-Ferráez et al.
Poster
An electrophoretic karyotype
for Mycosphaerella fijiensis
L. Conde-Ferráez, C. M. Rodríguez, L. Peraza-Echeverría and A. James
Abstract
In view of the current problems caused by black leaf streak disease in banana production, a
knowledge and understanding of the genetics and organization of the genome of Mycosphaerella
fijiensis could lead to the development of new control strategies. Regarding the former, mycelium
was obtained from isolates of M. fijiensis from three sites in Mexico (Veracruz, Colima and Chiapas)
in order to estimate the size of the genome by using the CHEF (Contour clamped Homogeneous
Electric Field) system. Different conditions of pulse field electrophoresis enabled the separation
of M. fijiensis chromosomes and a preliminary estimate of the karyotype of each isolate was
obtained. Isolates from Colima and Chiapas had bands corresponding to at least 10 chromosomes
in the size range 0.71 to 2.2 Mb. The Veracruz isolate had at least 14 chromosomes in a size range
of 0.67 to 5.6 Mb. Genome size calculated for the Veracruz isolate was at least 28 Mb, which is
comparable to that of some ascomycete fungi. Attempts were made to estimate the genome
size of the Colima and Veracruz isolates. Differences in the principal band suggested the
presence of polymorphisms in chromosome length between the isolates studied, as reported
for other species of fungi.
Resumen - Cariotipo molecular de Mycosphaerella fijiensis
Ante la problemática actual ocasionada por la Sigatoka negra en la producción de banano, el
conocimiento y la comprensión de la genética y la organización del genoma de Mycosphaerella
fijiensis podrían conducir al desarrollo de nuevas estrategias para su control. Considerando lo
anterior, se propuso obtener el cariotipo molecular de tres aislados de M. fijiensis por medio del
sistema CHEF (Contour clamped Homogeneous Electric Field), así como estimar su tamaño
genómico. Para ello, se utilizó el micelio de aislados procedentes de tres diferentes lugares de
México (Veracruz, Colima y Chiapas). Se ensayaron diferentes condiciones de electroforesis de
campo pulsante que permitieron separar los cromosomas de M. fijiensis. Se obtuvo una estimación
preliminar del cariotipo de cada aislado. En los aislados de Colima y Chiapas se observaron bandas
correspondientes a por lo menos 10 cromosomas, en un rango de tamaño entre 0.71 y 2.2 Mb. En
Centro de Investigación Científica de Yucatán, Mérida, México.
141
MyLsd 17x24
25/06/03
15:15
Page 142
Mycosphaerella leaf spot diseases of bananas: present status and outlook
el aislado de Veracruz se observaron por lo menos 14 cromosomas en un rango de tamaño entre
0.67 y 5.6 Mb. El tamaño del genoma calculado para el aislado de Veracruz es de al menos 28 Mb
lo cual es comparable con algunos hongos ascomicetos reportados en la literatura. Se pretende
realizar experimentos para estimar el tamaño genómico de los aislados de Colima y Veracruz.
Diferencias observadas en el patrón de bandeo sugieren la existencia de polimorfismos en la
longitud de los cromosomas entre los aislados estudiados, lo cual también ha sido reportado en
otras especies de hongos.
Résumé - Un karyotype électrophorétique pour Mycosphaerella fijiensis
Du fait des problèmes posés à l’heure actuelle par la maladie des raies noires pour la production
de bananes, la connaissance et la compréhension de la génétique et de l’organisation du
génome de M. fijiensis pourrait conduire au développement de nouvelles stratégies de contrôle.
En ce qui concerne ce dernier point, du mycelium a été obtenu à partir d’isolats de M. fijiensis
provenant de trois sites au Mexique (Veracruz, Colima et Chiapas) afin d’estimer la taille du
génome en utilisant le système CHEF (Contour clamped Homogeneous Electric Field). Des
conditions différentes d’électrophorèse en champ pulsé ont permis la séparation des chromosomes
de M. fijiensis et une estimation préliminaire du karyotype de chaque isolat a été obtenue. Les
isolats provenant du Colima et du Chiapas avaient des bandes correspondant à au moins 10
chromosomes dont la taille variait entre 0,71 et 2,2 Mb. L’isolat de Veracruz avait au moins 14
chromosomes dont la taille variait entre 0,67 et 5,6 Mb. La taille du génome calculée pour l’isolat
provenant de Veracruz était d’au moins 28 Mb, ce qui est comparable à celle de certains
champignons ascomycètes. Des essais ont été réalisés pour estimer la taille du génome des isolats
provenant de Colima et de Veracruz. Les différences observées pour la bande principale suggèrent
la présence de polymorphismes dans la longueur des chromosomes entre les isolats étudiés,
comme cela a été rapporté chez d’autres espèces de champignons.
Introduction
The chromosomes of many fungi are too small to be identified by cytological
methods therefore the detailed karyotype and genome size of most fungal species
are unknown. But starting with the successful separation of Saccaromyces
cerevisiae chromosomes (Schwartz and Cantor, 1984), Pulsed Field Gel
Electrophoresis (PFGE) has been used to obtain the molecular karyotypes of
important fungi, including pathogenic fungi. New technologies and electrophoresis
apparatus have since been developed, resulting in improved techniques for
separating large DNA fragments and chromosomes.
The CHEF (contour-clamped homogeneous electric field), electrophoresis system
(Chu et al., 1986), has been used to separate fungal chromosomes and estimate
genome size. Karyotyping procedures using PFGE generally involve the production
of protoplasts or sphaeroplasts. However a different technique (McCluskey et al., 1990)
permits the preparation of chromosome-sized DNA without the need for such
laborious procedures.
Even though it affects Musa worldwide, there are no data on the karyotype or
genome size of Mycosphaerella fijiensis. Knowing and understanding the genetics
and genome organization of fungal pathogens could lead to the development of new
strategies in controlling black leaf streak disease. This paper describes the use of
PFGE to obtain the molecular karyotype of M. fijiensis and estimates of genome
size.
142
MyLsd 17x24
25/06/03
15:15
Page 143
Session 2
L. Conde-Ferráez et al.
Methods
Leaves displaying symptoms of black leaf streak disease were collected from
Veracruz, Colima and Chiapas in Mexico. Fungal isolates were obtained from single
ascospores and grown on PDA medium. The identity of isolates was verified by the
morphology of mycelium, Koch´s postulates and PCR (Johanson, 1997; Neu et al.,
1999). For the preparation of agarose plugs, isolates were grown in liquid shake
culture (PDB medium) for 6 days at 26oC and 100 rpm. The mycelium was separated
by centrifugation, ground, washed with buffer, mixed with melted agarose at 45°C
and transferred to plug moulds. Plugs were incubated in proteinase K at 60°C for
48 h, and washed and stored in 0.5 M EDTA at 4°C. Empirical methods were tried
to optimize the resolution of chromosomes of different size ranges, using PFGE with
CHEF DR II and CHEF DR III. Finally, genome size was estimated by adding the values
assigned to each band resolved in pulsed field gels.
Results and discussion
We obtained a preliminary estimation of the karyotype of each isolate. For the
Veracruz isolate, 12 bands were resolved, representing 14 chromosomes in a size
range of 0.67 to 5.6 Mb (Figures 1 and 2). At least 10 chromosomes were resolved
in the Colima and Chiapas isolates, in a size range of 0.71 to 2.2 Mb. The estimated
genome size for the Veracruz isolate is at least 28 Mb (Figure 3), which is
comparable to that reported for other ascomycetes (Cooley and Caten, 1991;
McDonald and Martinez, 1991). The genome sizes of the other two isolates are to
be determined in further experiments. Differences observed in banding patterns
suggest the existence of chromosome length polymorphisms, which has been
reported for other fungi. A broader study with different and well characterized
haplotypes of M. fijiensis is now underway at the Centro de Investigación Científica
de Yucatán.
Conclusion
This is the first estimate of the karyotype and genome size of M. fijiensis, a useful
tool for constructing a physical and genetic map or for calculating the required size
of a genomic library
References
Chu G., D. Vollrath and R.W. Davis. 1986. Separation of large DNA molecules by contourclamped homogeneous electric fields. Science, 234:1582-1585.
Cooley R.N. and C. Caten. 1991. Variation in electrophoretic karyotype between strains of
Septoria nodorum. Molecular and General Genetics 228:17-23.
Johanson A. 1997. Detection of Sigatoka leaf spot of banana by the Polymerase Chain
Reaction. Natural Resources Institute. The University of Greenwich, U.K. 37pp.
McCluskey K., B.W. Russell and D. Mills. 1990. Electrophoretic karyotyping without the need
for generating protoplasts. Current Genetics 18:385-386.
143
MyLsd 17x24
25/06/03
15:15
Page 144
Mycosphaerella leaf spot diseases of bananas: present status and outlook
McDonald B.A. and J.P. Martinez. 1991. Chromosome length polymorphisms in Septoria tritici
population. Current Genetics 19:265-271.
Neu C., D. Kaemmer, G. Kahl, D. Fischer and K. Weising. 1999. Polymorphic microsatellite
markers for the banana pathogen Mycosphaerella fijiensis. Molecular Ecology 8:523-525.
Schwartz, D.C. and C.R. Cantor. 1984. Separation of yeast chromosome-sized DNAs by Pulsed
Field Gradient Gel Electrophoresis. Cell 37:67-75.
1
2
3
4
2200
+
1600
5
1900
1500
1300
1200
1070
1125
1050
1020
945
830
740
670
825
785
750
680
610
450
365
285
225
Figure 1. CHEF gel stained with SYBR Green showing separation of medium size chromosomes. Lane 1:
S. cerevisiae chromosome size standards. Lanes 2-5: M. fijiensis isolate from Veracruz. Numbers on the left give
the siez of the standards. Numbers on the right correspond to the nine bands resolved. Numbers in bold
represent comigrating bands.
144
MyLsd 17x24
25/06/03
15:15
Page 145
Session 2
L. Conde-Ferráez et al.
1
2
3
4
5.7
4.6
5
5.7
4.6
4.0
4.6
Figure 2. CHEF gel stained with SYBR Green showing separation of largest chromosomes. Lane 1: S. pombe
chromosome size standards. Lanes 2-5: M. fijiensis isolate form Veracruz. Numbers on the left give the size
of the standards. Numbers on the right correspond to the three bands resolved. Numbers in bold represent
comigrating bands.
145
MyLsd 17x24
25/06/03
15:15
Page 146
Mycosphaerella leaf spot diseases of bananas: present status and outlook
1
2
3
2200
4
5
2200
+
1600
1400
1055
1125
945
1020
945
805
785
750
825
715
785
750
680
610
450
Figure 3. CHEF gel stained with SYBR Green showing separation of largest chromosomes. Lanes 1-4: M. fijiensis
isolates from Veracruz, Colima, Chiapas #1 and Chiapas #2 respectively. Lane 5: S. cerevisiae chromosome size
standards. Numbers on the right give the size of the standards. Numbers on the left correspond to the eight
bands resolved. Numbers in bold represent comigrating bands.
146
MyLsd 17x24
25/06/03
15:15
Page 147
Session 1
Recommendations
Recommendations of session 2
Information on the epidemiology and population structure of the main Mycosphaerella species
(M. fijiensis, M. musicola and M. eumusae) at the national, regional and international levels
are needed to better understand the distribution and the spread of the pathogens, to anticipate
the evolution of pathogen populations and to define resistance management strategies. Such
studies are particularly necessary in Asia, the centre of diversity of the three pathogens and
where little research has so far been conducted.
Distribution of Mycosphaerella spp.
Knowing the name of the banana clones affected, the severity of the leaf spot diseases
and the local environmental conditions would help explain the distribution of M. fijiensis,
M. musicola and M. eumusae. IMTP trials are seen as ideal locations for assessing the reaction
of the banana clones to various leaf spot pathogens.
The collection and diagnosis of specimens from IMTP trials sites should be continued.
Identification tools should be provided to enable diagnoses to be undertaken locally.
The exact distribution of M. eumusae needs to be known.
Further surveys in South and Southeast Asia are necessary to determine where
M. musicola, M. fijiensis and M. eumusae occur. The cooperation and collaboration of
scientists in South and Southeast Asia is viewed as essential. The INIBAP regional office
for Asia and the Pacific should strengthen and facilitate any exchange between Asian
partners and the rest of the PROMUSA community.
National and international collections
The creation of national collections of strains of Mycosphaerella pathogens is of special
relevance to the understanding of population structure. The collections must be based on
single-spore cultures with an in vitro characterization of the anamorph stage (in vitro
sporulation of conidia). Diagnostic tools would help the development of collections of
Mycosphaerella isolates. It has been recommended to provide the participants with a
protocol to sample, establish and maintain the collections (see the “Diagnostics” section in
session 1 recommendations).
A reliable, rapid test to distinguish M. musicola, M. fijiensis, M. eumusae and possible
other Mycosphaerella pathogens/saprophytes needs to be developed. Information on how
to distinguish the three pathogens on morphological characteristics also needs to be
produced and circulated to banana scientists. INIBAP was asked to address this need.
The establishment of a national collection should be promoted and facilitated through
the organization of a training course; especially for those countries that develop breeding
programmes, but also in places where banana resistant hybrids are used on an industrial
scale, and where the high diversity of Musaceas has likely produced a similar diversity
of pathogens.
147
MyLsd 17x24
25/06/03
15:15
Page 148
Mycosphaerella leaf spot diseases of bananas: present status and outlook
It was recommended to develop an international core collection of M. fijiensis,
M. musicola and M. eumusae. The different strains should be conserved as fungal mycelia
and DNA. CIRAD was suggested as host of the international collection, using a similar
mechanism as INIBAP developed with KULeuven regarding Musa germplasm. INIBAP
was asked to address this needs in collaboration with CIRAD.
Genetic population structure
The study of the genetic population structures of Mycosphaerella pathogens is already ongoing at the national, regional and international levels. However, the number of countries
involved at the national level should be increased to refine regional and international studies.
The sampling protocols should be standardized and widely distributed. INIBAP and CIRAD
agreed to work together in the preparation of this information which should include
several detailed illustrations of the different pathogens and their anamorph stages. This
information should also be part of the IMTP guidelines.
More molecular markers, such as SSR and CAPS, should be developed to improve the
understanding of the different populations structures.
Pathogenicity characterization
In vitro and in vivo inoculation systems exist to evaluate the pathogenicity of the various
Mycosphaerella strains. The different pathogens and their relation to their host need to be
compared under controlled conditions using these methods.
The methodologies that currently exist should be standardized. The in vitro inoculation
on leaf fragment developed at CIRAD should be distributed together with the methodology
to isolate, cultivate and produce the inoculum of the different pathogens. INIBAP and
CIRAD have been requested to compile in a technical document, all the different
information already published on these methods.
Dispersal of Mycosphaerella spp.
More research is necessary to understand spore dispersal.
Disease incidence data should be collected from the field and the scientific literature.
Laboratory methods to understand the mechanism of spore release, and spore survival
in the atmosphere should be developed.
The potential for windborne dispersal suspected from laboratory studies should be verified
and assessed at the plantation level (as opposed to dispersal through other means, such
as infected planting material).
148
MyLsd 17x24
25/06/03
15:15
Page 149
Session 3
Host-pathogen interactions
MyLsd 17x24
25/06/03
15:15
Page 151
Session 3
P. Lepoivre et al.
Introduction
Banana–Mycosphaerella fijiensis
interactions
P. Lepoivre1, J. P. Busogoro1, J. J. Etame1, A. El Hadrami1,2,
J. Carlier2, G. Harelimana1, X. Mourichon2, B. Panis3,
A. Stella Riveros4, G. Sallé5, H. Strosse3 and R. Swennen3
Abstract
Using standard testing procedures, banana genotypes were classified as 1) highly resistant
cultivars characterized by an early blockage of leaf infection (incompatible interactions), 2) partially
resistant cultivars exhibiting a slow rate of symptom development (compatible reactions)
and 3) susceptible cultivars, characterized by rapid development of necrotic lesions (compatible
reaction).
Most information on incompatible reactions comes from observations of early necrosis of
stomatal guard cells and the deposit of electron-dense compounds around the penetration sites
of M. fijiensis on the cultivar ‘Yangambi km5’. Such rapid death of a few host cells, associated with
the blockage of the progression of the infecting agent is usually defined as a hypersensitive
reaction. Such a reaction often operates within a gene-for-gene relationship and as a consequence
the resulting resistance may be unstable.
As regards compatible interactions, cytological studies showed that M. fijiensis behaves first as
a biotrophic parasite which colonizes exclusively the intercellular spaces without the formation
of haustoria. Two main mechanisms have been investigated to explain the slow development
of a lesion in partially resistant genotypes: preformed antifungal compounds and tolerance to
putative toxin(s) produced by M. fijiensis.
The mechanisms will be presented in relation to their possible use as early screening markers
for selecting banana genotypes for durable resistance to M. fijiensis.
1
Faculté Universitaire des Sciences Agronomiques, Gembloux, Belgium
CIRAD, Montpellier, France
3
Katholieke Universiteit Leuven, Leuven, Belgium
4
CATIE, Costa Rica
5 Université Pierre et Marie Curie, Paris, France
2
151
MyLsd 17x24
25/06/03
15:15
Page 152
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Resumen - Interacciones banano – Mycosphaerella fijiensis
Utilizando los procedimientos de evaluación estándar, los genotipos de banano fueron clasificados
en tres categorías: 1) cultivares altamente resistentes caracterizados por un bloqueo temprano de
la infección foliar (interacciones incompatibles), 2) cultivares parcialmente resistentes que exhiben
una evolución lenta de los síntomas (reacciones compatibles), y 3) cultivares susceptibles,
caracterizados por un desarrollo rápido de las lesiones necróticas (reacción compatible).
La mayor parte de la información sobre las reacciones incompatibles proviene de los estudios del
cultivar ‘Yangambi km5’. Se observaron la necrosis de las células de guarda estomatales y los depósitos
de los compuestos con alta densidad de electrones alrededor de los sitios de penetración de M. fijiensis.
La muerte tan rápida de unas cuantas células hospedantes asociada con el bloqueo de la progresión
del agente infectante se define usualmente como una reacción hipersensible. Esta reacción a menudo
opera dentro de una relación de gen por gen y podría convertir la resistencia en inestable.
Con las reacciones compatibles,los estudios citológicos revelaron que M. fijiensis se comporta primero
como un parásito biotrófico que coloniza exclusivamente los espacios intercelulares sin formar los
haustorios. Dos mecanismos principales podrían estar involucrados en el desarrollo lento de las
lesiones observado en los genotipos resistentes parcialmente:compuestos antifungosos sintetizados
de manera constitutiva o tolerancia a la(s) toxina(s) putativa(s) producidas por M. fijiensis.
Estos mecanismos se presentarán en relación con su posible utilidad como marcadores de cribado
temprano en la selección de los genotipos de banano con respecto a la resistencia duradera a
M. fijiensis.
Résumé - Interactions bananier–Mycosphaerella fijiensis
En utilisant des procédures de test standard, des génotypes de bananier ont été classés en :
1) cultivars hautement résistants caractérisés par un blocage rapide de l’infection foliaire
(interactions incompatibles), 2) cultivars partiellement résistants montrant un développement
lent des symptômes (réactions compatibles) et 3) cultivars susceptibles caractérisés par un
développement rapide de lésions nécrotiques (réaction compatible).
L’essentiel des informations sur les réactions incompatibles provient d’observations de nécrose
précoce des cellules de garde des stomates et du dépôt de composés denses en électrons autour
des sites de pénétration de M. fijiensis chez le cultivar ‘Yangambi km5’. La mort aussi rapide d’un
petit nombre de cellules hôtes, associée avec le blocage de la progression de l’agent infectieux,
est habituellement définie comme une réaction hypersensible. Une telle réaction se produit souvent
dans le cadre d’une relation gène pour gène et, en conséquence, la résistance qui en résulte peut
être instable.
Pour ce qui concerne les interactions compatibles, les études cytologiques ont montré que M. fijiensis
se comporte d’abord comme un parasite biotrophique qui colonise exclusivement les espaces
intercellulaires sans formation d’haustoria. Deux mécanismes principaux ont été étudiés pour
expliquer le développement lent des lésions chez les génotypes partiellement résistants : des
composés antifongiques préformés et la tolérance à une(des) toxine(s) putative(s) produite(s) par
M. fijiensis.
Les mécanismes sont présentés en relation avec leur utilisation possible comme marqueurs lors
de criblage précoce pour sélectionner des génotypes de bananiers possédant une résistance durable
à M. fijiensis.
Introduction
Black leaf streak disease is the most devastating disease of banana and plantain
worldwide. The fungus induces foliar leaf streaks which, in highly susceptible
cultivars, leads to the total collapse of the plant.
Just as host plants evolved several defence mechanisms, pathogens have ways
to evade or suppress these defence mechanisms. The response of the host and the
152
MyLsd 17x24
25/06/03
15:15
Page 153
Session 3
P. Lepoivre et al.
pathogen are crucial to the outcome of infection. A knowledge of the interactions
is increasingly important for the rational selection of genotypes resistant to plant
pathogens. The interactions between banana and Mycosphaerella fijiensis remained
unknown for a long time.
Although the field performance of an accession is the ultimate reference for
evaluating its resistance, the method is not suitable for the study of host pathogen
interactions. Nevertheless, the reaction to black leaf streak disease of about 50 Musa
species belonging to various genetic groups was studied under natural infection
conditions (Fouré et al., 1990). The study led to the grouping of the banana genotypes in three categories: 1) highly resistant (HR) cultivars characterized by an early
blockage of leaf infection (incompatible interactions); 2) partially resistant (PR)
cultivars exhibiting slow rates of symptom development (compatible interactions);
and 3) susceptible (S) cultivars characterized by a rapid development of necrotic
lesions (compatible interactions). Later, banana-M. fijiensis-interactions were
studied under controlled conditions of inoculation (Mourichon et al., 1987) which
reproduced the behaviour in the field of three reference cultivars: ‘Yangambi km5’
(AAA; HR), ‘Fougamou’ (ABB; PR) and ‘Grande naine’ (AAA; S).
These preliminary results, which were presented at the International workshop
held in San José in 1989, were the start of investigations into banana-M. fijiensis
interactions. They began with the microscopic events that take place in banana
tissues and were followed by the analysis of the biochemical processes that culminate in the expression of resistance or susceptibility.
Host-pathogen interactions
Cytological studies of the interactions between M. fijiensis and the three reference
cultivars ‘Yangambi km5’, ‘Fougamou’ and ‘Grande naine’ revealed that M. fijiensis
enters banana leaves by the stomata.
In compatible interactions (‘Grande naine’ and ‘Fougamou’ inoculated with
M. fijiensis, strain 049 HND from Honduras), the pathogen colonized exclusively
the intercellular spaces between mesophyll cells, without forming haustoria. There
was a long period of biotrophy before the observation of the first cytological
alterations to the mesophyll cells. Hyphae were observed between living cells ahead
of the necrotic zone, a faster growth rate of hyphae being the main difference between
susceptible (‘Grande naine’) and partially resistant (‘Fougamou’) cultivars (Beveraggi,
1992; Beveraggi et al., 1995).
In contrast, early necrosis of stomata guard cells and appositions around the
penetration sites were observed with incompatible interactions (‘Yangambi km5’
inoculated with M. fijiensis strain 049HND) (Beveraggi et al., 1995).
The behaviour of partially and highly resistant genotypes of banana can be linked
to major groups of plant-parasite interactions. The rapid death of only a few host
cells, associated with the blockage of the progression of the infecting agent in the
highly resistant cultivar ‘Yangambi km5’, is usually defined as an hypersensitive
reaction. These often operate within a gene-for-gene relationship giving rise to
resistance that is unstable. In comparison, the partial resistance of the reference
cultivar ‘Fougamou’, for example, is usually considered polygenic and durable.
153
MyLsd 17x24
25/06/03
15:15
Page 154
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Incompatible interactions: Highly resistant cultivars
The hypersensitive reaction operating within a gene-for-gene relationship is
generally explained, either by the presence of a specific avirulence factor (or elicitor)
or by the coordinated action of non-specific elicitor(s) and a specific suppressor (de
Wit, 1992; Atkinson, 1993). In banana–M. fijiensis interactions, there is no
experimental evidence of a gene-for-gene relationship because it is difficult to study
the genetics of triploid genotypes such as ‘Yangambi km5’. But that such a
relationship exists is supported by laboratory tests in which isolates of M. fijiensis
overcame the resistance of ‘Yangambi km5’ (Fullerton and Olsen, 1995).
Riveros and Lepoivre (1994) did preliminary experiments to identify the elicitors
that induce resistance. Intercellular fluids (IF) from leaves of ‘Yangambi km5’
(incompatible) and ‘Grande naine’ (compatible) inoculated with 049HND, and crude
eliciting fractions (CEF) prepared from germinating spores of virulent and
avirulent M. fijiensis isolates, elicited necrosis and appositions in banana
cultivars (Riveros and Lepoivre, 1994).
Regardless of the eliciting preparation (IFs or CEFs prepared with avirulent
or virulent M. fijiensis isolates), the reaction was more intense and quicker
in ‘Yangambi km5’ than in the susceptible ‘Grande naine’. The behaviour
of ‘Yangambi km5’ cannot be explained by a race-specific eliciting activity
in the IFs or the CEFs. However, the eliciting activity present in the IFs of
‘Yangambi km5’ inoculated with the avirulent strain 049 HND appeared to be
higher than that in the compatible relationship between ‘Grande naine’ and the
same isolate. Thus, we speculate that ‘Yangambi km5’ could have a higher
sensitivity to the elicitor(s) but could also have a host-mediated effect on the
release, production or stability of specific elicitor(s) produced by the fungal isolates.
Such host-mediated effects have been reported in soybean tissues where plant
enzymes are responsible for the release of elicitors from hyphal walls of
Phytophthora megasperma (Boller, 1987).
A wide range of fungal compounds have been implicated as elicitors of HR:
polysaccharides (Sharp et al., 1984), glycoproteins (Schaffrath et al., 1995),
peptides (de Wit et al., 1985) and hydrolytic enzymes (Boller, 1987). In banana–
M. fijiensis interactions, polysaccharide compounds may be involved in eliciting
activity (Riveros, unpublished data).
Evidence of a hypersensitive-like-reaction to M. fijiensis represents a first step
towards a better characterization of that reaction. Independently of the
mechanisms of resistance, there is the problem of the durability of resistance.
Durability is of outmost importance because breakdown of resistance for a staple
crop such as plantain would have dramatic effects. Because of the difficulties
inherent in improving triploid bananas, breeding for resistance to black leaf streak
disease often did not take into account host-pathogen interactions. The existence
of M. fijiensis isolates able to overcome resistant ‘Yangambi km5’ in laboratory
tests (Fullerton and Olsen, 1995) shows that highly resistant parents should not
be used without appropriate management procedures, such as mixtures of
cultivars, choice of the “right gene combination” and co-ordinated regional
deployment of genes.
154
MyLsd 17x24
25/06/03
15:15
Page 155
Session 3
P. Lepoivre et al.
Compatible reaction: partial resistant cultivars
With partially resistant cultivars (compatible interactions), cytological studies showed
that M. fijiensis behaves at first as a biotrophic parasite that colonizes exclusively the
intercellular spaces (Beveraggi et al. 1995). Two possible mechanisms have been
investigated to explain slow lesion development in partially resistant genotypes:
preformed antifungal compounds and tolerance to putative toxin(s) produced by
M. fijiensis.
Constitutively synthesised antifungal compounds
Many antimicrobial compounds produced by plants play an important role in the
response to infection by cellular pathogens. Defence compounds may be classified
into phytoanticipins, which are constitutive, and phytoalexins, which are synthesised
in response to microorganisms. The two groups of secondary metabolites include
a wide range of chemical families. However, phytoanticipins are primarily involved
in non-host rather than varietal resistance.
For ‘Fougamou’, histological analysis revealed the presence, in mesophyll
layers, of many specialized cells containing vacuoles rich in polyphenol. The contents
of the vacuoles were released into the intercellular spaces. The contents had a high
affinity for fungal cell walls and their presence around hyphae seemed to be
correlated with the slow growth of mycelium in parenchyma tissues (Beverragi et
al., 1992, 1995). Gire (1994) identified soluble phenols in the leaf tissues of several
banana cultivars with different levels of partial resistance. He also observed a close
correlation between flavane (protoanthocyanidins) content and the level of partial
resistance. However, a study conducted on a larger number of genotypes suggested
that the role of these constitutive compounds in partial resistance is restricted to
a limited number of cultivars (El Hadrami, 1997).
The role of toxins in pathogenesis
Pathogen toxins could constitute an alternative technique for rapidly screening
resistant banana genotypes as in vitro plant tissues or young plants. The symptoms
of black leaf streak disease suggest a possible involvement of phytotoxic compounds.
Such compounds were found in culture filtrates of M. fijiensis (Molina and Krausz,
1989; Lepoivre and Acuna, 1989; Upadhyay et al., 1990; Strobel et al., 1993). Stierle
et al. (1991) reported that 2,4,8-tetrahydroxytetralone and juglone were the most
abundant and most phytotoxic compounds.
If toxins are involved in the development of black leaf streak disease, it may be
possible to use them to identify resistant genotypes. During the previous workshop
at San José, the use of M. fijiensis toxins for screening had four major limitations:
1) a lack of quantitative and sensitive bioassays to measure the effects of M. fijiensis
metabolites on banana genotypes; 2) insufficient characterization of the variability
in toxin production of M. fijiensis; 3) a lack of experimental evidence for the role
of the metabolites in the disease; and 4) the assurance that the susceptibility and/or
resistance of cultured tissues reflected the reaction of the whole plant.
155
MyLsd 17x24
25/06/03
15:15
Page 156
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Bioassay to quantify the effect of M. fijiensis
metabolites
A set of bioassays was developed to quantify the toxic effects of the metabolites
obtained from M. fijiensis culture filtrates. The induction of necrosis by a leaf
puncture bioassay on detached banana leaves, or the injection of ethyl acetate crude
extract (EaCE) into the leaves are easy to perform but neither method is sensitive
(injections of 250 ppm EaCE are required for ‘Grande naine’) or quantitative.
The electrolyte leakage assay represented a quantitative but rather insensitive
assessment of the toxicity of M. fijiensis metabolites. The test did not distinguish
between cultivars. The most sensitive and accurate toxin assay was based on the
measurement of chlorophyll fluorescence. The vitality index seemed to be the most
sensitive method for early assessment of the effects of EaCE and a specific indicator
of photosynthetic activity.
Purification of the EaCE revealed the presence of different fractions with similar
properties to the crude extracts. Juglone, a purified metabolite previously shown
to be present in extracts of M. fijiensis culture filtrates, was identified in the extracts
of all the strains analyzed. Injection of juglone into banana leaf tissues gave similar
results to EaCE for ranking cultivars (Etame, unpublished data).
Chloroplasts as target site of juglone
The involvement of the photosynthetic apparatus in reaction to EaCE and juglone
is in agreement with observation of light-dependent toxicity irrespective of the
bioassay. The observation of swelling chloroplasts as the first abnormality observed
by electron microscopy of EaCE-treated leaves also fits this pattern.
Busogoro (unpublished data) developed a bioassay using isolated chloroplasts and
measuring their capacity to reduce 2,6- dichlorophenoindolphenol (DCPIP) as a marker
of the Hill reaction, which expresses electron transport from water to any electron
acceptor by intact chloroplasts when exposed to light (Allen and Holmes, 1986).
Juglone inhibited the Hill reaction in suspensions of banana chloroplasts. In
addition, ‘Fougamou’ chloroplasts appeared to be less affected by juglone than
‘Grande Naine’ chloroplasts. These results suggest that chloroplasts are one of the
primary action sites of juglone.
The role of toxins in banana-M. fijiensis interactions
The electrolyte leakage assay and chlorophyll fluorescence were used to compare the
sensitivity to EaCE of different banana cultivars with their behaviour in the field (highly
resistant, sensitive or partially resistant) as scored using the rank of the youngest leaf
spotted with necrotic lesions.
These toxin assays confirmed that the incompatible interactions of ‘Yangambi km5’
were not related to resistance to EaCE. The toxicity of EaCE preparations was
independent of the virulence of the strain (unpublished data). Mechanisms of resistance
in highly resistant cultivars were definitely not related to the action of these toxic
metabolites.
156
MyLsd 17x24
25/06/03
15:15
Page 157
Session 3
P. Lepoivre et al.
Considering just the susceptible ‘Grande naine’ and the partially resistant
‘Fougamou’ cultivars, susceptibility to EaCE was correlated with sensitivity to
infection, suggesting that slow lesion development is associated with a lower
sensitivity to M. fijiensis toxins.
Such quantitative assessment is difficult to interpret because the concentrations
of toxin(s) that were used in the bioassay could exceed the in vivo concentration
and affect the mode of action of EaCE, hence affecting the rating of the cultivars.
The hypothesis that M. fijiensis metabolites have a secondary role as determinants
of pathogenicity agrees with cytological studies, which showed no evidence of an
early effect of toxic compounds in the long period of biotrophy before observing
the first cytological alterations in the mesophyll cells.
Selection of banana tissues resistant to juglone
The work was done with ‘Three hand planty’, a genotype susceptible to black leaf
streak disease, and juglone for which an embryo cell suspension was available.
Juglone was toxic to embryogenic cell suspensions and somatic embryos of the
cultivar. Necrosis of all cell suspensions and somatic embryos was quickly observed
at 50 ppm or more of juglone, with the exception of some somatic embryos that
continued development after treatment with 50 ppm of juglone.
The plants regenerated from the surviving embryos showed a higher resistance
to juglone: 250 ppm was required to induce necrosis in the leaf puncture bioassay
with selected plantlets in comparison with 100 ppm for non selected plants.
However, the selected plants did not show higher resistance to black leaf streak disease
than the mother Three hand planty’ genotype following inoculation with M. fijiensis
(El Hadrami, unpublished data).
Daub (1986) advised caution when using metabolites from pathogens to screen
tissue cultures for resistance. Nevertheless, fungal toxins have been proposed to screen
banana in vitro (Strobel et al., 1993). There have been claims that resistant material
has been produced by selecting callus of banana that survived increasing
concentrations of M. fijiensis toxins (Okole and Shulz, 1993). Our results confirm
the possibility of selecting banana plants resistant to M. fijiensis metabolites but
this approach did not result in higher resistance to black leaf streak disease.
References
Allen J.F. and N.G. Holmes. 1986. Electron transport and Redox Titration. Pp. 103-141
in Photosynthesis energy transduction. A practical approach. (M.F. Hipkins and N.R. Baker,
eds). IRL PRESS, Washington.
Atkinson M.M. 1993. Molecular mechanisms of pathogen recognition by plants. Advances
in Plant Pathology 10:35-59.
Beveraggi A. 1992. Etude des interactions hôte-parasite chez des bananiers sensibles et
résistants inoculés par Cercospora fijiensis responsables de la maladie des raies noires.
Thèse de 3ème cycle, Université de Montpellier II, USTL.
Beveraggi A., X. Mourichon and G. Salle. 1995. Etude comparée des premières étapes de
l’infection chez les bananiers sensibles et résistants infectés par Cercospora fijiensis
157
MyLsd 17x24
25/06/03
15:15
Page 158
Mycosphaerella leaf spot diseases of bananas: present status and outlook
(Mycosphaerella fijiensis), agent responsable de la maladie des raies noires. Canadian
Journal of Botany 73:1328-1337.
Boller T. 1987. Hydrolytic enzymes in plant disease resistance. Pp. 385-413 in Plant-Microbe
Interactions, vol.2 (Kosuge, T. and E.W. Nester, E. W. eds.). MacMillan, New York.
Daub M.E. 1986. Tissue culture and the selection of resistance to pathogens. Annual Review
of Phytopathology 24:159-186.
de Wit P.G.J.M. 1992. Molecular characterization of gene-for-gene systems in plant fungus
interactions and the application of avirulence genes in control of plant pathogens. Annual
Review of Phytopathology 30:391-418.
de Wit P.G.J.M., E.E. Hofman, G.C.M. Velthius and J. Kuc. 1985. Isolation and characterization
of an elicitor of necrosis isolated from intercellular fluids of compatible interactions of
Cladosporium fulvum (syn. Fulvia fulva) and tomato. Plant Physiology 77:642-647.
El Hadrami A. 1997. Protoanthocyanidines constitutifs des feuilles de bananiers et résistance
partielle vis-à-vis de Mycosphaerella fijiensis, l’agent causal de la maladie des raies noires.
Thèse de DEA, Faculté universitaire des Sciences Agronomiques de Gembloux.
Fullerton R.A. and T.L. Olsen. 1995. Pathogenic variability in Mycosphaerella fijiensis Morelet
cause of black Sigatoka in banana and plantain. New Zealand Journal of Crop and
Horticultural Science 23:39-48.
Fouré E., A. Mouliom Pefoura and X. Mourichon. 1990. Etude de la sensibilité variétale des
bananiers et plantains à Mycosphaerella fijiensis Morelet au Cameroun. Caractérisation
de la résistance au champ de bananiers appartenant à divers groupes génétiques. Fruits
45:339-345.
Gire A. 1994. Relation entre la résistance partielle du bananier à Cercospora fijiensis et une
composante cellulaire constitutive de nature polyphénolique. DEA, Université Montpellier II.
Lepoivre P. and C.P. Acuna. 1989. Production of toxins by Mycosphaerella fijiensis var.
difformis and induction of antimicrobial compounds in banana: their relevance in breeding
for resistance to black Sigatoka. Pp. 201-207 in Sigatoka Leaf Spot Diseases of Bananas.
Proceedings of an international workshop, San José, Costa Rica, March 28-April 1, 1989
(Fullerton, R.A. and R.H. Stover eds). INIBAP Montpellier, France.
Molina G., and J.P. Krausz. 1989. A phytotoxic activity in extracts of broth cultures of
Mycosphaerella fijiensis var. difformis and its use to evaluate host resistance to Black
Sigatoka. Plant Disease 73:142-144.
Mourichon X., D. Peter and M.F. Zapater. 1987. Inoculation expérimentale de M. fijiensis
Morelet sur jeunes plantules de bananier issues de culture in vitro. Fruits 42:195-198.
Okole B.N. and F.A. Shulz. 1993. Selection of banana and plantain (Musa spp.) resistant to
toxins produced by Mycosphaerella species using in vitro culture techniques. Pp. 378 in
Breeding banana and plantain for resistance to diseases and pests. Proceeding of the
International Symposium on Genetic Improvement of Bananas to Diseases and Pests.
(Ganry J. ed.). CIRAD-FHLOR, Montpellier, France, 7-9 septembre 1992.
Riveros A.S. and P. Lepoivre. 1994. Early induction of non specific cultivar glucanase eliciting
activity in the apoplastic fluids of banana infected by Mycosphaerella fijiensis. Arch. Int.
Physiol. Bioch. Biophys. 102(5):9.
Schaffrath U., H. Schelinpflug and H.J. Reisner. 1995. An elicitor from Pyricularia oryzae
induces resistance response in rice: isolation, characterization and physiological properties.
Physiol. and Mol. Plant Pathol. 46:293-307.
Sharp J.K., M. McNeil and P. Albersheim. 1984. The primary structures of one elicitor-active
and seven elicitor-inactive hexa (ß-D-glucopyranosyl)-D-glucitols isolated from mycelial
walls of Phytophthora megasperma f. sp. glycinea. J. Biol. Chem. 259:11321-11336.
158
MyLsd 17x24
25/06/03
15:15
Page 159
Session 3
P. Lepoivre et al.
Stierle, A.A., R. Upadhyay, J. Hershenhorn, G.A. Strobel and G. Molina. 1991. The phytotoxins
of Mycosphaerella fijiensis, the causative agent of black Sigatoka disease of bananas and
plantains. Experientia 47:853-859.
Strobel G.A., A.A. Stierle, R. Upadhyay, J. Hershenhorn and G. Molina. 1993. The phytotoxins
of Mycosphaerella fijiensis, the causative agent of black sigatoka disease, and their potential
use in screening for disease resistance. Pp. 93-103 in Biotechnology applications for banana
and plantain improvement, 27-31 janvier 1992. INIBAP, Montpellier, France.
Upadhyay, R., G.A. Strobel and S. Coval. 1990. Some toxins of Mycosphaerella fijiensis.
Pp. 231-236 in Sigatoka Leaf Spot Diseases of Bananas: Proceedings of an international
workshop San José, Costa Rica, March 28-April 1, 1989 (Fullerton, R.A. and R.H. Stover
eds). INIBAP Montpellier, France.
159
MyLsd 17x24
25/06/03
15:15
Page 161
Session 3
C. Abadie et al.
Efficiency and durability
of partial resistance against
black leaf streak disease
C. Abadie1,2, A. Elhadrami2, E. Fouré1 and J. Carlier2
Abstract
Black leaf streak disease caused by Mycosphaerella fijiensis is the most destructive leaf disease
of bananas and plantains. Genetic improvement for resistance appears as the most appropriate
tool to control the disease. As a high level of diversity is maintained in pathogen populations,
breeders prefer working with partial resistance, which is thought to be durable, instead of total
resistance. Our aim is to evaluate the efficiency and durability of partial resistance. To achieve
this objective, three complementary approaches were undertaken:
1) Partial resistance was characterized by measuring various variables over the life cycle of the
fungus under field and controlled conditions. The evaluation of 13 partially resistant varieties
revealed the existence of several components acting at various stages of the infectious cycle.
2) The efficiency of two resistant varieties which differ for two resistance components (infection
efficacy, ascospores production) were studied. No difference in disease dispersal and incidence
was observed between resistant varieties during the first year whereas small differences in disease
severity, increasing over time, were measured. These results could be explained by differences
in endogeneous inoculum production. Experiments are conduced to measure endogeneous
inoculum in each field to confirm this hypothesis.
3) The durability of resistance is being studied by analyzing the evolution of pathogen populations.
Molecular characterization using CAPS markers was used on populations isolated after 6 and 25
months of cultivation. No significant difference between the populations taken from susceptible
and resistant bananas was observed after 6 months. Pathogenicity variability was undergone
to assess an eventual selective effect of hosts.
Resumen - Eficacia y durabilidad de la resistencia parcial contra la enfermedad de
la raya negra de la hoja
La enfermedad de la raya negra de la hoja causada por Mycosphaerella fijiensis es la enfermedad
foliar más destructiva de bananos y plátanos. El mejoramiento genético con respecto a la
resistencia parece convertirse en la medida de control más apropiada. Debido a un alto nivel de
diversidad que se mantiene en las poblaciones del patógeno, los programas de mejoramiento
se basan en la resistencia parcial que se considera hasta más durable que la resistencia total. El
propósito de estos estudios consiste en evaluar la eficacia y durabilidad de la resistencia parcial
1
CARBAP, Douala, Cameroon
2CIRAD, Montpellier, France
161
MyLsd 17x24
25/06/03
15:15
Page 162
Mycosphaerella leaf spot diseases of bananas: present status and outlook
para ayudar a los programas de mejoramiento. Para lograr este objetivo, se examinaron tres
enfoques complementarios:
1) La caracterización de la resistencia parcial se estudió a través de la medición de diferentes
parámetros del ciclo de vida del hongo en condiciones de campo y controladas. La evaluación de
13 variedades diferentes con resistencia parcial reveló la existencia de varios componentes de
resistencia que actúan en diferentes etapas del ciclo de infección.
2) La eficacia de dos variedades resistentes en las cuales los dos componentes de resistencia (la
eficacia de la infección y la producción de ascosporas) se estudiaron. Durante el primer año no
se observaron diferencias entre las dos variedades en cuanto a dispersión espacial o su incidencia.
Sin embargo, las mediciones indicaron bajas diferencias en la severidad de la enfermedad, la cual
aumentó en los años siguientes. Las diferencias en la producción del inóculo endógeno podrían
explicar los resultados arriba mencionados. Varios experimentos que se están llevando a cabo
actualmente para estudiar el inóculo endógeno podrían confirmar la hipótesis.
3) Se está estudiando la durabilidad de la enfermedad a través de la evolución de las poblaciones
del patógeno. La caracterización molecular con los marcadores CAPS se utilizó en las poblaciones
aisladas después de 6 y 25 meses de cultivo. Después de seis meses no se observaron diferencias
significativas entre las dos poblaciones provenientes de los bananos susceptibles y resistentes.
Se estudió la variabilidad en el poder patógeno para evaluar un eventual efecto selectivo de los
hospedantes.
Résumé - Efficacité et durabilité de la résistance partielle contre la maladie des raies
noires
La maladie des raies noires, causée par Mycosphaerella fijiensis, est la maladie foliaire la plus
destructrice chez les bananiers et les plantains. L’amélioration génétique pour la résistance apparaît
comme le moyen le mieux approprié pour contrôler la maladie. Comme un niveau élevé de diversité
est maintenu dans les populations de pathogènes, les sélectionneurs préfèrent travailler avec
une résistance partielle, qui est considérée comme étant plus durable, plutôt qu’avec une
résistance totale. Trois approches complémentaires ont été suivies pour atteindre cet objectif :
1) La résistance partielle a été caractérisée en mesurant différentes variables pendant le cycle
vital du champignon au champ et en conditions contrôlées. L’évaluation de 13 variétés partiellement
résistantes a révélé l’existence de plusieurs composants agissant à des stades différents du cycle
infectieux.
2) L’efficience de deux variétés résistantes qui différaient pour leurs composants de résistance
(efficacité de l’infection, production d’ascospores) a été étudiée. Aucune différence dans la
dispersion et l’incidence de la maladie n’a été observée entre les variétés résistantes pendant la
première année, alors que de petites différences ont été notées pour la sévérité de la maladie,
qui augmentait avec le temps. Ces résultats pourraient être expliqués par des différences dans
la production d’inoculum endogène. Des essais sont en cours pour mesurer l’inoculum endogène
dans chaque champ pour confirmer cette hypothèse.
3) La durabilité de la résistance a été étudiée en analysant l’évolution des populations de
pathogènes. La caractérisation moléculaire avec des marqueurs CAPS a été utilisée sur des
populations isolées après 6 et 25 mois de culture. Aucune différence significative n’a été trouvée
entre des populations prélevées sur des bananiers résistants et sensibles. La variabilité de la
pathogénicité a été étudiée pour évaluer la possibilité d’un effet sélectif des hôtes.
Introduction
Black leaf streak disease, caused by the ascomycete fungus Mycosphaerella
fijiensis, is the most important foliar disease of banana worldwide (Jones, 2000).
Indeed, the main varieties cultivated industrially and by smallholders, and
belonging to the Cavendish and plantains groups, are very susceptible to black
162
MyLsd 17x24
25/06/03
15:15
Page 163
Session 3
C. Abadie et al.
leaf streak disease. Chemical control strategies based on biological forecasting
system were first developed in industrial plantations (Fouré, 1988). Although
efficient, chemical control negatively affects the environment and human health,
and is too expensive for poor smallholders. Managing black leaf streak disease
must integrate the use of resistant varieties.
Two types of resistance against black leaf streak disease have been described
(Fouré, 1992). High resistance, characterized by a blockage of symptom expression
and an absence of sporulation, and partial resistance, characterized by moderate
disease expression and a normal but slow development of symptoms up to necrosis.
High resistance is believed to be similar to hypersensitivity and under the control
of a mono or oligogenic system which may be easily circumvented by pathogens
(Fouré et al., 2000). For example, the high resistance of ‘Paka’ is no longer effective
in the Cook Islands and studies of pathogenic variability showed that it was
overcome by some virulent strains after 8 years of cultivation (Fullerton and Olsen,
1995).
Analysis of population genetic structure of M. fijiensis at various geographical
scales revealed high levels of gene diversity and genetic differentiation (Carlier
et al., 1996 and in this volume). Such structures suggest a high adaptation potential
of the pathogen, hence the use in the banana breeding programmes of CARBAP
and CIRAD of partial resistance instead of high resistance.
Partial resistance is a complex character which could include several components
corresponding to different stages of pathogen infectious cycle (Young, 1996). Until
recently, only one variable, youngest leaf spotted (YLS), was used to evaluate partial
resistance of bananas in screening trials. Such evaluation was useful to characterise
germplasm but not sufficient to identify the components of partial resistance. For
example, sporulation, which could have an important effect in the case of a
polycyclic epidemic like black leaf streak disease, is not measured. Identifying the
components coming into play at various stages of the infectious cycle can be
conducted by inoculation under controlled conditions.
The efficiency of the components of partial resistance can only be evaluated
in the framework of an epidemiological study in a field of one variety. Pathogen
populations could evolve and erode resistance depending on the evolutionary forces
at work. Strategies based on the evolutionary potential of the pathogen could
improve the durability of resistance. To define such strategies, the relative
importance of the evolutionary forces acting on the pathogen. The effect of genetic
recombination, genetic drift and gene flow on pathogen evolution can be evaluated
by analyzing population structure (Carlier et al, 1996 and this volume).
Resistant hosts exert a selection pressure on the pathogens. The evolution of
the pathogen can be studied in fields of resistant hosts by using molecular markers
and characterizing its pathogenicity. The durability of partial resistance will be
studied by following over time the evolution of pathogens in fields of resistant
hosts.
The objectives of this study are to identify components of partial resistance to
black leaf streak disease in bananas, and to evaluate their efficiency and durability
in controlling the disease. A three-step experimental approach was developed: 1)
characterization of partial resistance of various cultivars under controlled
163
MyLsd 17x24
25/06/03
15:15
Page 164
Mycosphaerella leaf spot diseases of bananas: present status and outlook
conditions, 2) epidemiological study using one variety to evaluate the efficiency
of the components of partial resistance, and 3) analysis of the pathogen population
over time to evaluate the durability of partial resistance in relation to the selection
pressure exerted by the host.
Components of partial resistance
To characterize partial resistance, variables corresponding to the different stages of
an infectious cycle were estimated and compared among various partially resistant
varieties. Evaluations of resistance were conducted in Cameroon on 7 partially
resistant and susceptible cultivars under field conditions and on 10 cultivars under
controlled conditions. In the latter, pieces of banana leaf maintained on a culture
media were inoculated (El Hadrami et al., 1998). Fifteen strains were used.
The infectious cycle was dissected in eight stages (Figure 1) including incubation
period; spore efficacy (number of lesions); size and growth rate of lesions; asexual
and sexual latency period; asexual and sexual sporulation capacity.
Many variables were significantly different between resistant varieties under field
and controlled conditions. For example, the lesions were significantly smaller in
‘Zebrina’ and ‘Pisang Ceylan’ than in other resistant varieties (Figures 2a and 2b).
The number of lesions and the incubation period also varied significantly with the
highest number of lesions found on ‘Pisang Berlin’ and ‘Pisang Ceylan’ and the
Production of ascospores
Production of conidia
Number, size & growth rate of lesions
Ascopores
Conidia
First lesions
Necrosis
Sporulating necrosis
Sporulating lesions
with ascospores
with conidia
Penetration
Asexual phase
Epiphyl phase
Sexual phase
Incubation period
Asexual latency period
Sexual latency period
Infectious ascospores
Figure 1. Infectious cycle of Mycosphaerella fijiensis (from El Hadrami, 2000). The components of partial
resistance being characterized are in italics.
164
25/06/03
15:15
Page 165
Session 3
14
C. Abadie et al.
Controlled conditions
a
b
10
c
8
d
de
e
ef
6
ef
fg
4
g
S
PCE
FOU
ZEB
PR
Field conditions
30
a
ab
25
20
TAN
PKW
TDE
PBE
PMA
0
ROS
2
CAV
Size of lesion
(% of a 25-cm2 piece of leaf)
12
Size of lesion
(% of spotted area)
bc
ab
bc
15
c
c
10
5
S
PCE
FOU
ZEB
PBE
FSO
PMA
0
CAV
MyLsd 17x24
PR
Figure 2. Size of lesions in partially resistant (PR) and susceptible (S) varieties under (a) controlled conditions
and under (b) field conditions. (Cav: Cavendish (AAA), FSO: French sombre (AAB), PMA: Pisang madu (AAcv),
ROS: Rose d’Ekona, PBE: Pisang Berlin (AAcv), TDE: Thong det (AAcv), PKW: Pisang klutuk wulung (BBw),
TAN (BBw), ZEB: Zébrina (AAw), FOU: Fougamou (ABB), PCE: Pisang C eylan (ABB)).
longest incubation period observed on ‘Fougamou’ and ‘Tani’ (data not shown). No
difference was observed in the production of asexual spores whereas significant
differences were observed in the production of sexual spores (Figure 3). ‘Pisang madu’
and ‘Zebrina’ produced 3 to 8 times less than the other three partially resistant
varieties, including ‘Pisang Berlin’. Thus, although ‘Pisang madu’ and ‘Pisang
Berlin’ behave similarly in field trials as regards YLS, ‘Pisang madu‘ produced four
times less perithecia/cm2 of necrosis than ‘Pisang Berlin’.
165
25/06/03
15:15
Page 166
Mycosphaerella leaf spot diseases of bananas: present status and outlook
This first step allowed us to identify components of partial resistance which act
at different stages of the infectious cycle : early stage (e.g. small lesions on ‘Zebrina’
and ‘Pisang Ceylan’), intermediate stages (e.g. long incubation period on ‘Fougamou’
and ‘Tani’) or end of the cycle (e.g. low production of sexual spores on ‘Pisang madu’
and ‘Zebrina’).
a
Field conditions
1600
Number of perithecia/cm2 of necrosis
1200
b
800
bc
bc
bc
400
c
c
S
PCE
FOU
ZEB
PBE
FSO
PMA
0
CAV
MyLsd 17x24
PR
Figure 3. Production of sexual spores in five partially resistant (PR) and two susceptible (S) varieties under
field conditions.(Cav: Cavendish (AAA), FSO: French sombre (AAB), PMA: Pisang madu (AAcv), PBE: Pisang Berlin
(AAcv), ZEB: Zébrina (AAw), FOU: Fougamou (ABB), PCE: Pisang Ceylan (ABB)).
Efficiency of components
An epidemiological study was conducted on two varieties to evaluate the efficiency
of two components of partial resistance: spore efficacy and production of sexual
spores. ‘Pisang madu’, ‘Pisang berlin’ and a susceptible control (‘Grande naine’) were
cultivated in three rectangular plots containing 150 bananas/plot. Measures were
taken over 3 cropping cycles.
During the first year, no difference in disease dispersal and disease incidence was
observed between the 3 plots. No spatial auto-correlation was measured at plot scale
(Moran index not significant) and disease incidence was similar between plots (data
not shown).
Disease severity (visual quantification of percentage of spotted surface per plant)
was significantly different in resistant varieties (about 20 % for ‘Pisang Berlin’) and
susceptible varieties (an average of 40%) (Table 1). This result shows the efficient
role of partial resistance in controlling black leaf streak disease. Significant
166
MyLsd 17x24
25/06/03
15:15
Page 167
Session 3
C. Abadie et al.
differences were also observed between the two resistant varieties, differences
which increased over time due to the declining disease severity on ‘Pisang madu’
(Table 1). Production of sexual spores, on the other hand, was stable over the three
cropping cycles. This result could be due to differences in components of resistance
or an increase in resistance ‘Pisang madu’ over time. To test the first hypothesis,
secondary inoculum was measured by using spore trap plants in each resistant plot.
The precision of this method needs to be improved to be able to show eventual
differences in auto-inoculum production.
Table 1. Efficacy of partial resistance expressed as disease severity (% of spotted area/banana) on susceptible (S)
and partially resistant (PR) varieties over three cropping cycles.
Varieties
Disease severity (% of spotted area)
1st
Grande naine (S)
Pisang Berlin (PR)
Pisang madu (PR)
2nd cycle
cycle
37.5 a
22.7 b
17.5 c
39 a
18.6 b
7.1 c
3rd cycle
36 a
17.2 b
0.8 c
Durability of components partial resistance
The population structures of about 50 isolates of M. fijiensis taken from resistant
(‘Pisang madu’) and susceptible (‘Grande naine’) after 6 and 25 months of cultivation
were analysed using molecular markers and a pathogenic test to evaluate the relative
importance of genetic drift and selection by the host.
Seven CAPS (cleaved amplified polymorphism sequences) neutrals markers were
used and the pathogenicity of the isolates was measured by inoculating leaf pieces
of ‘Pisang madu’ and ‘Grande naine’ maintained on a culture media (El Hadrami
et al., 1998).
No significant difference in genetic differentiation was detected between the
samples isolated from resistant and susceptible bananas. The estimate of Wright’s
Fst parameter over all loci between pairs of populations was low and ranged from
0.007 to 0.0432. This absence of differentiation between populations suggests a low
genetic drift effect during colonisation and/or important gene flow between fields.
No clear evidence of differences in aggressiveness were observed between the
two pathogens after 6 months of cultivation. Samples isolated from resistant and
susceptible bananas seemed to have the same pathogenic behaviour. Further
analyses will be done on populations after 25 months of cultivation.
Conclusion and perspectives
This study revealed the existence of various components of partial resistance under
controlled and field conditions. Then, the efficiency of two components of
resistance (spore efficacy, production of sexual spores) on disease control were
tentatively evaluated. A selective effect of partial resistance components on
pathogen population was not detected but aggressiveness of isolates were evaluated
only after six months of cultivation. We are analysing pathogens population after
167
MyLsd 17x24
25/06/03
15:15
Page 168
Mycosphaerella leaf spot diseases of bananas: present status and outlook
two years of cultivation. However, the absence of genetic differentiation between
fields containing different varieties could be the result of high gene flow. High
gene flow can counteract changes in gene frequency as a result of selection by
the host. To clarify this point, spore dispersal needs to be specified directly in
epidemiological studies or indirectly ing population genetic models.
Computer models can now simulate epidemics using parameters corresponding
to the different stages of the pathogen’s infectious cycle. The effect and the
importance of the components of partial resistance can then be tested by
comparing simulated and observed epidemics. Computer models can help in
designing future experiments and in defining spatio-temporal management
strategies of resistant varieties at different scales (from plot to regional scale) by
testing different scenarios.
References
Carlier J., M.H. Lebrun, M. F. Zapater, C. Dubois and X. Mourichon. 1996. Genetic structure
of the global population of bananas black leaf streak fungus Mycosphaerella fijiensis.
Molecular Ecology 5:499-510.
El Hadrami A., M.F. Zapater, F. Lapeyre, C. Abadie and J. Carlier. 1998. A leaf disk assay to
assess partial resistance of banana germplasm and agressiveness of Mycosphaerella fijiensis,
the causal aget of black leaf streak disease. 7th International Congress of Plant Pathology,
Edinburgh, Scotland. BSPP vol. 2, p.1.1.24.
El Hadrami, A. 2000. Caractérisation de la résistance partielle des bananiers à la maladie des
raies noires et évaluation de la variabilité de l’agressivité de l’agent causal, Mycosphaerella
fijiensis. Thèse d’Université. Faculté Universitaire des Sciences Agronomiques de Gembloux,
Belgique. 153pp.
Fouré E. 1988. Stratégies de lutte contre la cercosporiose noire des bananiers et plantains
provoquée par Mycosphaerella fijiensis Morelet. L’avertissement biologique au Cameroun.
Evaluation des possibilités d’amélioration. Fruits, vol 43(5): 269-274.
Fouré E. 1992. Characterization of the reactions of bananas cultivars Mycosphaerella fijiensis
Morelet in Cameroon and genetics of resistance. Pp. 159-170 in Breeding banana
and plantain for resistance to diseases and pests, Proceedings of the International
Symposium on Genetic Improvement of Bananas for Resistance to Diseases and Pests
7-9 september 1992, Montpellier, France.
Fouré E., X. Mourichon and D. Jones. 2000. Evaluating germplasm for reaction to black leaf
streak. P 62-72 in Diseases of Banana, Abaca and Enset. (Jones D. ed.), CAB International,
Wallingford, UK, 544pp.
Fullerton R.A. and T. L. Olsen. 1995. Pathogenic variability in Mycosphaerella fijiensis Morelet
cause of black Sigatoka in banana and plantain. New Zealand Journal of Crop and
Horticultural Science 23:39-48.
Jones D. 2000. Diseases of Banana, Abaca and Enset. CAB International, Wallingford, UK,
544pp.
Young N.D. 1996. QTL mapping and quantitative disease resistance in plants. Annual review
of phytopathology, 34:479-501.
168
MyLsd 17x24
25/06/03
15:15
Page 169
Session 3
Y. Alvarado Capó et al.
Poster
Early evaluation of black leaf streak
resistance by using mycelial
suspensions of Mycosphaerella
fijiensis
Y. Alvarado Capó, M. Leiva Mora, M. A. Dita Rodríguez,
M. Acosta, M. Cruz, N. Portal, R. Gómez Kosky,
L. García, I. Bermúdez and J. Padrón
Abstract
A standardized method for the early evaluation of resistance to black leaf streak on in vitro Musa
plants was developed using mycelial suspensions of Mycosphaerella fijiensis. Seven cultivars:‘FHIA18’, ‘FHIA-01’, ‘FHIA-21’, ‘Grande naine’, ‘Yangambi’, ‘Calcutta 4’ and ‘Niyarma yik’ were tested in a
greenhouse. Inoculum was adjusted to 105 cfu/ml and applied to the lower surfaces of the first
three open leaves. Plants were evaluated 15 days after inoculation and at 15-day intervals until
60 days. A standardized scale of leaf symptoms ensured consistency between evaluators. All
cultivars except ‘Yangambi’, showed a similar response to M. fijiensis in natural conditions. Partial
resistance expressed in FHIA cultivars was characterized by a slow rate of symptom development
with ‘Calcutta 4’ the slowest. ‘Grande naine’ and ‘Niyarma yik’ gave a susceptible reaction and
their symptoms were more severe. Artificial inoculation of in vitro plants with mycelial suspensions
was an easy, rapid and practicable method to determine resistance to M. fijiensis. An inoculum
adjusted to an appropriate concentration gave uniform symptoms on the inoculated leaf. The
method has promise for the evaluation of in vitro plants in breeding programmes.
Resumen - Evaluación temprana de la resistencia a la raya negra de la hoja mediante
el uso de la suspensión del micelio de Mycosphaerella fijiensis
Se describen los métodos de normalización para la evaluación temprana de la resistencia a la
raya negra de la hoja en las plantas in vitro de Musa mediante el uso de la suspensión del micelio
Instituto de Biotecnología de las Plantas, Santa Clara, Cuba
169
MyLsd 17x24
25/06/03
15:15
Page 170
Mycosphaerella leaf spot diseases of bananas: present status and outlook
de Mycosphaerella fijiensis. Siete cultivares:‘FHIA-18’,‘FHIA-01’,‘FHIA-21’,‘Grande naine’,‘Yangambi’,
‘Calcutta 4’ y ‘Niyarma Yik’ fueron utilizados para la prueba de inoculación en el invernadero. La
concentración del inóculo fue ajustada a 105 ufc/ml y las primeras tres hojas en abrirse fueron
inoculadas en la superficie inferior. El período de evaluación empezó a los 15 días y termino 60
días después de la inoculación. En la evaluación de las hojas se utilizó una escala (unidad
experimental) que permitió evitar la confusión del evaluador y describir el desarrollo de los
síntomas de mejor manera. Con excepción del ‘Yangambí’, los cultivares mostraron un
comportamiento similar contra los patógenos en condiciones naturales. La resistencia parcial
expresada en los cultivares de la FHIA se caracterizó por una lenta evolución de los síntomas. El
cultivar ‘Calcutta 4’mostró el tiempo más lento de desarrollo de los síntomas. Los cultivares ‘Grande
naine’ y ‘Niyarma yik’ mantuvieron reacciones susceptibles y sus síntomas alcanzaron grados
mayores de afectación. La inoculación artificial de las plantas in vitro utilizando la suspensión
de micelio resultó ser un método fácil, rápido y factible para conocer la expresión de la resistencia
de las plantas contra M. fijiensis. La utilización del inóculo con una concentración ajustable permitió
obtener síntomas homogéneos y uniformes de la hoja inoculada. El mismo representa una
herramienta útil para la evaluación de las plantas in vitro en los programas de mejoramiento.
Résumé - Evaluation précoce de la résistance à la maladie des raies noires au moyen
de suspensions mycéliennes de Mycosphaerella fijiensis
Une méthode standardisée d’évaluation précoce de la résistance à la maladie des raies noires
sur des vitroplants de Musa a été développée en utilisant des suspensions mycéliennes de
Mycosphaerella fijiensis. Sept cultivars, ‘FHIA-18’, ‘FHIA-01’, ‘FHIA-21’, ‘Grande naine’, ‘Yangambi’,
‘Calcutta 4’et ‘Niyarma yik’ont été étudiés en serre. L’inoculum a été ajusté à 105 cfu/ml et appliqué
sur la surface inférieure des trois premières feuilles ouvertes. Les plantes ont été évaluées 15 jours
après l’inoculation, puis à intervalles de 15 jours jusqu’à 60 jours. Une échelle standardisée de
symptômes foliaires a permis d’assurer la consistance entre évaluateurs. Tous les cultivars sauf
‘Yangambi’ ont montré une réponse similaire à M. fijiensis en conditions naturelles. La résistance
partielle exprimée par les cultivars FHIA a été caractérisée par un faible taux de développement
des symptômes, qui était le plus bas chez ‘Calcutta 4’. ‘Grande naine’ et ‘Niyarma yik’ ont donné
une réaction susceptible et leurs symptômes étaient plus sévères. L’inoculation artificielle de
vitroplants avec des suspensions mycéliennes s’est avérée une méthode simple, rapide et
praticable pour déterminer la résistance à M. fijiensis. Un inoculum ajusté à une concentration
appropriée a donné des symptômes uniformes sur la feuille inoculée. Cette méthode est
prometteuse pour l’évaluation de vitroplants dans des programmes d’amélioration.
Introduction
Screening for resistance to black leaf streak in field condition is time-consuming
and expensive. Early evaluation in controlled conditions is an important
requirement to increase success and evaluate feasibility. Stover (1987)
recommended the use of standard varieties and the development of standardized
methods for preparing inoculum and measuring disease response in controlled
environments.
Mourichon et al. (1987) developed a greenhouse method for artificial
inoculation with Mycosphaerella fijiensis conidia and mycelium onto three
reference cultivars with different levels of resistance (‘Grande naine’, ‘Fougamou’
and ‘Yangambi km 51/2’). Romero and Sutton (1997) used suspensions of conidia
to evaluate the reaction of four Musa genotypes at three temperatures with isolates
from different geographical regions. Jones (1995) used hyphal fragments of
M. musicola for the rapid assessment of different Musa spp., and Balint-Kurti et
170
MyLsd 17x24
25/06/03
15:15
Page 171
Session 3
Y. Alvarado Capó et al.
al. (2001) developed inoculation techniques using transgenic strains of M. fijiensis
expressing GFP by using conidia and mycelia.
The development of artificial inoculation techniques is also necessary to
improve and simplify selection procedures in breeding programmes. Even though
ascospores and conidia are commonly used for artificial inoculation, mycelial
fragments may provide an alternative for evaluating disease development in
controlled conditions.
The objective of the work was to prepare a standardized method for the rapid
evaluation of resistance to black leaf streak disease in in vitro Musa plants using
mycelial suspensions of M. fijiensis, and its application in breeding programmes.
Material and methods
Plant material
In vitro plants of ‘FHIA-18’ (AAAB), ‘FHIA-01’ (AAAB), ‘FHIA-21’ (AAAB), ‘Grande
naine’ (AAA), ‘Yangambi’ (AAA), ‘Calcutta 4’ (AA) and ‘Niyarma yik’ (AA) were
inoculated in a greenhouse. Plants were acclimatized for eight weeks on a substrate
comprising 50% casting, 30% compost and 20% zeolite. Plants were grown in
plastic pots 100 cm in diameter.
Preparation of mycelial suspension
The pathogenic monoascosporic strain CC-IBP-1 (isolated from ‘Grande naine’)
was used. Colonies of M. fijiensis were grown on potato dextrose agar (PDA) at
28oC for 14 days. Pieces fragments were transferred to an Erlenmeyer flask
containing 200 ml of V8 liquid medium (200 ml V8 juice, 0.3 g CaCO3, 800 ml
water, pH 6.0). The flask was incubated at 28oC in a shaker (130 rpm) for 15 days.
The mycelium was then blended and 1 ml was transferred to a Petri dish containing
PDA to obtain homogeneous growth over the surface of the dish. The culture was
incubated in the dark at 28oC for 15 days. Two 1-cm2 discs of mycelium were
transferred to a 1-litre flask containing 400 ml of liquid V8, and shaken at 130
rpm at 28o C for 15 days. The mycelium (10 g in 900 ml of sterile distilled water)
was blended and filtered through two layers of gauze to remove large fragments
of hyphae. The concentration was determined with a haemacytometer and
adjusted to 105 cfu/ml. Gelatin at 1% was added to the final inoculum.
Inoculation
Plants 20 cm in height and with four active leaves were selected for inoculation.
There were 15 plants of each cultivar.
The first three open leaves of each plant were inoculated on the under surface
using a brush. Leaves were marked on the upper side. The plants were allowed
to dry for 2 h, and the relative humidity maintained at 90-100% for the first three
days by spraying continuously with water. Afterwards, the humidity was saturated
during the night but spraying was suspended during the day. Sunlight was used.
171
MyLsd 17x24
25/06/03
15:15
Page 172
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Evaluation
Inoculated leaves were examined every 15 days starting on the 15th day after
inoculation and ending on the 60th. Symptoms corresponded approximately to
the descriptions of Fullerton and Olsen (1995) for in vitro plants inoculated with
conidia. Table 1 describes the scale used to evaluate symptom development and
the classification of genotypes according to the stage of symptom development.
Table 1. Stages of symptom development in in vitro Musa plants inoculated with mycelial suspensions of
Mycosphaerella fijiensis in the greenhouse.
Stage
Description
0
1
2
3
4
Leaf symptoms mostly absent.
Reddish flecks on lower leaf surface. No symptoms on the upper surface.
Regular or irregular reddish circular spots on the lower leaf surface. No symptoms on the upper surface.
Regular or diffuse light brown circular spots oin the upper leaf surface.
Black or brown circular spots, possibly with a yellow halo or chlorosis of adjacent tissues, on the upper leaf
surface. Areas of green tissue sometimes present.
Black spots with dry centre of grey colour. Leaf completely necrotic, sometimes hanging down.
5
Classification of genotypes according to symptom development
Resistant: stages 0-1
Partially resistant: stages 2-3
Susceptible: stages 4-5
In vitro plants of ‘Grande naine’ and ‘FHIA-21’, obtained from IBP breeding
programmes (by mutagenesis), were evaluated as described above. Figure 1
describes the work schedule for the method.
Results and discussion
As described by Mourichon et al. (1987), symptoms developed on the leaves of
in vitro plants artificially inoculated with mycelial suspensions of M. fijiensis.
The appearance of symptoms was similar to those observed on suckers in the field.
The necrotic spots were slightly circular, possibly because young plants derived
from tissue culture have limited vein development and black leaf streak lesions
tend to be spherical (Mourichon et al., 2000). The majority of lesions were often
observed on the same leaf at the same stage of development. All cultivars, except
‘Yangambi’, responded to M. fijiensis in a similar manner to that observed in
natural conditions (Table 2).
The reaction of ‘Yangambi’ was characterized by the presence of symptoms
after the first 15 days, symptoms which reached stages 2 and 3 in 30-45 days
similar to the behaviour of susceptible genotypes. Some leaves had necrotic spots.
Mourichon et al. (1987) had reported a hypersensitive reaction for in vitro plants
of ‘Yangambi’ inoculated with conidia. In contrast, Fullerton and Olsen (1995)
reported a typical susceptible response in ‘Yangambi’ when in vitro plants were
inoculated with the conidia of a virulent strain of M. fijiensis.
172
MyLsd 17x24
25/06/03
15:15
Page 173
Session 3
Y. Alvarado Capó et al.
Schedule
In vitro plants
Mycosphaerella fijiensis
Acclimatized 8 weeks
20cm, 4 active leaves
Mycelial suspension
1x105 cfu/ml
INOCULATION
Under surface of first 3 open leaves
Humidity 90-100% for 3 days-sunlight
EVALUATION
0
1
2
3
4
5
15, 30, 45 and 60 days
Stages 0-5
Figure 1. Schedule of the method for artificial inoculation of in vitro plants of Musa with mycelial suspensions
of Mycosphaerella fijiensis in the greenhouse.
Table 2. Reaction of seven Musa cultivars to artificial inoculation with mycelial suspensions of Mycosphaerella fijiensis
in the greenhouse.
Cultivars
Grande naine
Niyarma yik
FHIA-01
FHIA-18
FHIA-21
Calcutta 4
Yangambi
Reaction in the field
Symptom stage
Susceptible
Susceptible
Partially resistant
Partially resistant
Partially resistant
Resistant
Resistant
15 d*
30 d
45 d
60 d
1
1
0
0
0
0
1
1
1
0
0
1
0
2
2
1
1
1
1
1
3
4
4
2
2
2
1
3
*d = days after inoculation.
The partial resistance expressed by FHIA cultivars was characterized by a slow
development of symptoms. Only after 60 days were reddish spots seen on the upper
surface of the leaves and the majority of leaves remained free of symptoms.
Romero and Sutton (1997) reported similar results when they examined the
173
MyLsd 17x24
25/06/03
15:15
Page 174
Mycosphaerella leaf spot diseases of bananas: present status and outlook
response of ‘FHIA-01’ and ‘FHIA-02’ inoculated with conidia. They pointed out
that although the mechanism of resistance to black leaf streak is not known, a
low density of stomata, and increased production of cuticular wax, phytoalexin,
suberin and lignin, or resistance to phytotoxins may be associated with partial
resistance.
‘Calcutta 4’ showed the slowest rate of symptom development. Most leaves had
stage 1 symptoms although stages 2 and 3 were observed on a few leaves. ‘Grande
naine’ and ‘Niyarma yik’ reacted, as expected. They were susceptible and the
symptoms developed to the fullest extent. On ‘Grand naine’, symptoms were mostly
at stages 4 and 5 whereas they were mostly at stages 3 and 4 on ‘Niyarma yik’. High
humidity in the greenhouse caused the leaves of ‘Calcutta 4’, ‘Niyarma yik’,
‘Yangambi’ and ‘FHIA-21’ to senesce rapidly.
As for the in vitro plants from the IBP mutagenesis breeding programmes,
differences in response were found between lines of the same genotype (‘FHIA-21’)
and between ‘Grande naine’ and the control (Table 3).
Table 3. Reaction of in vitro plants obtained from IBP mutagenesis breeding programmes to artificial inoculation with
mycelial suspensions of Mycosphaerella fijiensis in the greenhouse.
Cultivar
Symptom stage
‘Grande naine’
Line-GN-A1
4
‘FHIA- 21’
Line-F21-A
Line-F21-B
Line-F21-C
Line-F21-D
Line-F21-E
3
3
2
1
2
Control
‘Grande naine’
‘FHIA-21’
5
2
Conclusion
Artificial inoculation of in vitro plants with mycelial suspensions was easy, rapid and
practical for determining the expression of resistance of plants to M. fijiensis. Its use
in breeding programs could reduce time and work.
The standardized scale of symptoms was useful in order to ensure consistency
between evaluators and to improve the description of symptom development. The
use of inoculum with an adjusting concentration resulted in homogeneous and uniform
symptoms on inoculated leaves. There was no interference from saprophytic fungi.
Large quantities of mycelial fragments were produced in a short time and could solve
the problem of loss of conidial production in vitro, which occurs with some isolates
of M. fijiensis. An additional advantage is that mycelial fragments can be produced
at different periods of the year. The method may have uses in other studies on hostpathogen interactions.
174
MyLsd 17x24
25/06/03
15:15
Page 175
Session 3
Y. Alvarado Capó et al.
References
Balint-Kurti P.J., G.D. May and A. Churchill. 2001. Development of a Transformation System
for Mycosphaerella Pathogens of Banana. FEMS Microbiology Letters 195:9-15.
Fullerton R.A. and T.L. Olsen. 1995. Pathogenic variability in Mycosphaerella fijiensis Morelet,
cause of black Sigatoka in banana and plantain. New Zealand Journal of Crop and
Horticultural Science 23:39-48.
Jones D.R. 1995. Rapid assessment of Musa for reaction to Sigatoka disease. Fruits 50(1):11-22.
Mourichon X., D. Peter and M. Zapater. 1987. Inoculation expérimentale de Mycosphaerella
fijiensis Morelet sur de jeunes plantules de bananiers issues de culture in vitro. Fruits
42:195-198.
Mourichon X., P. Lepoivre and J. Carlier. 2000. Host-pathogens interactions. Chapter 2. Fungal
disease of the foliage. Pp. 67-72 in Diseases of Banana, Abacá and Enset. (D.R. Jones,
ed.). CABI publishing, Wallingford, Oxford, UK.
Romero R.A. and T.B. Sutton. 1997. Reaction of four Musa genotypes at three temperatures
to isolates of Mycosphaerella fijiensis from different geographical regions. Plant Disease
81:1139-1142.
Stover R.H. 1987. Measuring response of Musa cultivars to Sigatoka pathogens and proposed
screening procedures. Pp. 114-118 in Banana and Plantain breeding strategies. Proceedings
of an international workshop, Cairns, Australia, 13-17 October 1986. (G.J. Persley and
E.A. de Langhe, eds). ACIAR Proceeding No. 21.
175
MyLsd 17x24
25/06/03
15:15
Page 176
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Recommendations of session 3
Several cases of an unexpected level of susceptibility to black leaf streak disease have been
reported. Although different reasons have been offered to explain the phenomenon (poor
nutrition, environmental stress), the problem of the erosion of resistance cannot be ignored
and requires a precise characterization of the pathogen population. A greater understanding
of the mechanisms involved in plant-pathogen interactions continues to be needed to ensure
the long term success of breeding programmes. The current programmes based on a priori
hypothesis have shown their limits.
Mechanisms of pathogenicity
Other pathosystems (such as Magnaporthe grisea) have shown the powerful nature of the
genetic approach to identify, without any a priori, the pathogenicity factors. These approaches
include the study of gene expression during infection (differential display, DNA chip, SSH,
etc.), production of pathogenicity mutants, comparative genomic and gene function validation
techniques. A technique for the transformation of M. fijiensis is already available at the Boyce
Thompson Institute (USA). The genetics and physical mapping of M. fijiensis genome in
Mexico, in collaboration with PRI, Netherlands, should speed up the expected progress of
these experimental approaches.
It was recommended to develop genetic and molecular biology tools for M. fijiensis in
collaboration with groups working on M. graminicola.
It was recommended launching a genomic initiative to access to genomic tools (EST
collection, physical map, genome sequence) and set up a genomic-wide comparison of
M. fijiensis to M. graminocola
Mechanisms involved in partial resistance
We recommend studying differences in resistance and susceptibility in hosts which have
a similar genetic background. Future work should concentrate on segregating populations
to evaluate critically possible mechanisms of resistance.
The genetic approach recommended to study pathogenicity factors should also be adopted
to identify the mechanisms of partial resistance (patterns of genes expression during
infection of resistant cultivars). These genetic studies should be accompanied by a
dissection of the infection cycle to identify the components of resistance.
176
MyLsd 17x24
25/06/03
15:15
Page 177
Session 13
Recommendations
Vertical resistance
It was considered that characterization of the different resistance genes represents a
prerequisite to evaluate several strategies of gene deployment (pyramiding, mixture,
multilines). Detection and identification of resistance genes in host plants relies on the
availability of different isolates of the pathogen exhibiting different virulence phenotypes.
It was recommended to collect pathogenic isolates on resistant cultivars for evaluation
[genetic (to evaluate the genetic control of pathogenicity in resistant cultivars) and
controlled inoculation on resistant cultivars (to evaluate the genetic control of resistance
in banana)]. In this respect genetic crossings between isolates showing different behaviors
on resistant cultivars are recommended to identify the genetic controls of virulence on
resistant cultivars, thus leading to a better identification of genes for resistance.
It was recommended to develop marker-assisted breeding (both for partial resistance
and resistance genes). This will be facilitated by the activities of CIRAD on the genetic
map of banana.
177
MyLsd 17x24
25/06/03
15:15
Page 179
Session 4
Genetic improvement for the
management of resistance
MyLsd 17x24
25/06/03
15:15
Page 181
Session 4
K. Craenen and R. Ortiz
Introduction
Genetic improvement for a sustainable
management of resistance
K. Craenen1 and R. Ortiz2
Abstract
In the 1990s, innovative cross-breeding and classic genetic analysis of segregation ratios allowed
advances in the understanding of host plant response to black leaf streak disease. Partial resistance
owing to a recessive major gene (bs1) coupled with at least two additive minor genes (bsri) appears
to be durable because this genetic system slows disease development in the host plant. As a
consequence, resistant hybrids show more healthy leaves, i.e. greater photosynthetic leaf area,
than their susceptible full-sibs, which may partially account for their high yield. Although other
breeding approaches such as genetic transformation, mutagenesis and somaclonal variation are
advocated to develop new resistance to Mycosphaerella leaf spot diseases in Musa, farmers today
are only adopting the research products from the so-called conventional breeding, i.e. tetraploid
or triploid resistant hybrids from interspecific interploidy crosses. Recent findings on pathogenicity
with molecular and cellular biology tools are providing new knowledge on host plant – pathogen
interactions, which may result in science-led approaches for deploying resistance against
sigatoka diseases within a holistic integrated disease management framework. For example,
cultivar mixtures and gene pyramiding may be alternatives for potential durable resistance to
sigatoka diseases of plantain and banana.
Resumen - Mejoramiento genético para un manejo sostenible de la resistencia
El cruzamiento innovador y el análisis genético clásico de las proporciones de segregación en las
poblaciones resultantes de estos permitió avanzar en el entendimiento de la respuesta de la planta
hospedante a la Sigatoka negra en Musa en la década de los 90. La resistencia parcial debido a
un gen principal recesivo (bs1) acoplado al menos a dos genes secundarios aditivos (bsri) parece
ser duradera, debido a que este sistema genético retarda el desarrollo de la enfermedad en la
planta hospedante. Como consecuencia, los híbridos resistentes muestran mayor cantidad de
hojas sanas, es decir, una mayor área foliar fotosintética que sus hermanos completos susceptibles,
lo que puede explicar su alto rendimiento. Aunque se afirma que otros enfoques de mejoramiento
como la transformación genética, mutagénesis y variación somaclonal desarrollan una nueva
resistencia a las enfermedades de las manchas foliares causadas por Mycosphaerella en Musa,
actualmente los agricultores están adoptando solo los productos de investigaciones del tal
1
Belgian Technical Cooperation, Kampala, Uganda
International Institute of Tropical Agriculture, Nigeria
2
181
MyLsd 17x24
25/06/03
15:15
Page 182
Mycosphaerella leaf spot diseases of bananas: present status and outlook
llamado mejoramiento “convencional”, es decir, híbridos tetraploides o triploides resistentes de
los cruces interespecíficos, interploidia. Los recientes descubrimientos sobre la patogenecidad
con la ayuda de las herramientas de la biología molecular y celular están proporcionando nuevos
conocimientos sobre las interacciones planta hospedante-patógeno, que pueden resultar en
enfoques guiados por la ciencia para desarrollar resistencia contra las enfermedades de la Sigatoka
en el marco holístico de manejo integrado de la enfermedad. Por ejemplo, las mezclas de los
cultivares y la construcción piramidal de genes podrían representar alternativas para una
resistencia duradera potencial a las enfermedades de Sigatoka de bananos y plátanos.
Résumé – L’amélioration génétique pour une gestion durable de la résistance
Dans les années 1990, des croisements novateurs et l’analyse génétique classique des ratios de
ségrégation ont permis de mieux comprendre la réaction des plantes-hôtes à la maladie des raies
noires. Une résistance partielle due à un gène récessif principal (bs1) couplée à au moins deux
gènes mineurs additifs (bsri) semble durable étant donné que le système génétique ralentit le
développement de la maladie dans la plante-hôte. Par conséquent, les hybrides résistants ont
plus de feuilles fonctionnelles, c.-à-d. une plus grande surface photosynthétique, que leurs parents
susceptibles ce qui expliquerait en partie leur rendement élevé. Même si d’autres méthodes
d’amélioration, comme la transformation génétique, la mutagénèse et la variation somaclonale
sont prônées pour mettre au point de nouvelles résistances aux maladies foliaires causées par
Mycosphaerella spp., de nos jours les fermiers adoptent seulement les produits issus des
méthodes traditionnelles d’amélioration, c.-à-d. des hybrides tétraploïdes ou triploïdes résistants
issus de croisements interploïdes interspécifiques. Des recherches récentes sur la pathogénicité
en utilisant des outils de la biologie cellulaire et moléculaire nous renseignent sur les interactions
plante-pathogène qui pourraient mener à des méthodes scientifiques pour déployer la résistance
aux cercosporioses dans un cadre global de lutte intégrée aux maladies. Par exemple, l’assortiment
de cultivars et le cumul des gènes (gene pyramiding) pourraient être des options pour créer une
résistance durable aux cercosporioses qui affectent les bananiers.
Dedication. To Dirk R. Vuylsteke (1958-2000) the ‘father’ of genetic-led Musa improvement,
with whom we learned some of the issues discussed in this article. Dirk himself, his many
articles and book chapters and research products ensuing from his breeding work will be
always a source of inspiration to us and the new generation of plantain and banana breeders
worldwide. We miss him greatly but Dirk will always live among us and those who share his
humanitarian view of improving the livelihoods of the rural poor, particularly in Africa.
Breeding for disease resistance
Knowledge about the kind of disease, its effects and epidemics should be acquired
before launching an efficient plant breeding program for disease resistance
(Simmonds and Smartt, 1999). Throughout the breeding process information may
be gathered on the kinds of resistance and the genetics of the host-plant reaction.
At the same time research products, i.e., resistant germplasm from this approach,
are made available for eco-friendly and sustainable disease control. The breeding
strategy for durable host plant resistance needs to consider the populations and
their genetic diversity (for both plant host and pathogen) and reliable screening
methods.
Interaction between the product of a resistance gene R in the plant host and the
avirulence (avr) gene encoded by a given pathogen isolate, results in the specific
recognition of the pathogen (Flor, 1971). If R or avr is absent, the pathogen continues
182
MyLsd 17x24
25/06/03
15:15
Page 183
Session 4
K. Craenen and R. Ortiz
colonising the plant host, reproduces and ultimately causes the disease (Holub, 2001).
However, when R matches avr, the plant host recognises the pathogen, and a series
of intracellular signal events occur in the plant host. The most common genetic
interpretation for such an interaction claims that R products are receptors for avrencoded ligands, and that this recognition often leads to rapid, localized, cell death
of those penetrated by the pathogen, i.e. hypersensitivity.
Recent reports suggest that R genes are usually organized as clusters in plant
genomes (Fluhr, 2001; Leister et al., 1998), which provides a comparative advantage
for pyramiding specific resistance genes that may protect an individual plant host
against many pathogen isolates (Dangl and Holub, 1997). Likewise, pyramiding will
benefit the plant host because it may have a genetic reservoir from which new specific
resistance may evolve. Advances in molecular breeding can assist in monitoring and
accelerating the introgression of R genes into the host plant (Rommens and Kishore,
2000).
The crop
Bananas and plantains are giant perennial herbs that thrive in the humid tropics
and subtropics. The edible cultivars, in order of decreasing numerical importance,
are triploid, diploid or tetraploid, and belong to the Eumusa series of the genus Musa.
The warm, humid conditions required for banana and plantain also favour the
development of Mycosphaerella fijiensis, the causal agent of black leaf streak disease.
All plantain cultivars and most triploid bananas are susceptible to black leaf streak
disease. In the long-term, identification of resistant landraces or resistance breeding
are generally considered as the most appropriate strategies to control the disease
(Craenen et al., 2000).
The disease and pathogen
Black leaf streak disease has become a major constraint to expanding the cultivation
of edible Musa. The causal pathogen of black leaf streak disease, Mycosphaerella
fijiensis Morelet, is a fungus that attacks the leaves. The fungal spores are
disseminated by wind and infect the leaves as they unroll. The disease develops faster
where humidity and rainfall are high. It has spread rapidly to all major banana and
plantain growing areas and the spread is still continuing. Chemical control strategies
exist, but are environmentally unsound and socio-economically inappropriate,
particularly within the framework of the resource-poor smallholders that grow the
crop in Africa (Craenen et al., 2000).
Diversity and pathogenecity
Isolates of M. fijiensis from different geographical origins were assayed with
restriction fragment length polymorphisms (RFLP) (Carlier et al., 1994, 1996).
Australasia and Southeast Asia isolates showed the greatest variation, suggesting the
pathogen originated there. Genetically homogenous groups with low variation were
observed in Africa, the Pacific Islands, and Latin America. These results indicate a
183
MyLsd 17x24
25/06/03
15:15
Page 184
Mycosphaerella leaf spot diseases of bananas: present status and outlook
few introductions of small size in each of these regions from Southeast Asia (Carlier
et al., 1994). Genetic distances between regions were high but the isolates from the
Pacific Islands and Latin America seem related, suggesting that the introductions to
Latin America came from the Pacific Islands. Founder effects accompanied the
introductions to Latin America, some Pacific Islands and Africa (Carlier et al., 1999),
thereby reducing genetic diversity in these regions.
Recent analysis with microsatellite polymorphisms revealed variation in the
fingerprint patterns of M. fijiensis populations from Nigeria (Muller et al., 1995, 1997).
Microsatellite (oligonucleotide) fingerprinting appears to be a reliable technique for
assessing genetic variation among individuals as well as for defining clusters of related
genotypes, i.e. detecting intraspecific variation even on a microgeographical scale.
Fullerton and Olsen (1993) evaluated the pathogenic diversity within populations
of M. fijiensis. They used as differential hosts most of the standard cultivars currently
being used in the International Musa Testing Programme. A wide host response to the
whole range of strains was reported. The most susceptible across all isolates were ‘Grande
naine’ and ‘SF 215’, while ‘Calcutta 4’, the widest source of alleles resistant to black
leaf streak disease, was susceptible to some strains, particularly those collected in the
Pacific Islands and Papua New Guinea. However, the host reaction to the pathogen must
be tested with adult plants and, as their results suggest, for probably more than one
year in order to detect strains present at low frequency. Perhaps, resistant germplasm
needs to be tested over several years in locations with very virulent strains in order to
evaluate the durability of resistance.
The data in Fullerton and Olsen (1993) were re-analysed using simple linear regression
models to determine the stability of resistance to black leaf streak disease in Musa
acuminata, and using principal component analysis to study the pattern of strain and
genotype variation in the pathogen-host plant interaction (Ortiz et al., 2000). ‘Tuu Gia’
was regarded as having non-specific resistance to the 33 strains of M. fijiensis used in
the experiment. Furthermore, it seems that its resistance does not break down even under
the pressure of high virulent strains. In contrast, the resistance of ‘Calcutta 4’ was very
unstable and may break down in environments where highly virulent strains have
evolved.
Principal component analysis revealed that isolates of M. fijiensis from Papua New
Guinea were the most virulent, and that ‘Calcutta 4’ accounted for most of the genotype
x strain interaction. Strains collected from the same country could be clustered together
(e.g. Nigeria or most of the Pacific Islands’ isolates), or had a continuous virulence
distribution (e.g. Papua New Guinea), or were completely distinct (e.g. Central America).
The latter suggests a different geographical origin for the introduced M. fijiensis or a
change in virulence genes in the strains from Central America as a result of the intensive
use of fungicides to control black leaf streak disease.
Incidence and severity of black leaf streak disease
Quantification of black leaf streak disease is necessary to evaluate resistance and to
determine yield loss, the importance of black leaf streak disease in particular areas,
and the efficacy of control measures (Craenen, 1998). Fouré (1985) described in detail
the experimental methods to characterize the different host responses. A high level of
184
MyLsd 17x24
25/06/03
15:15
Page 185
Session 4
K. Craenen and R. Ortiz
resistance to black leaf streak disease is characterized by hypersensitivity. Different
levels of partial resistance, ranging from strong partial resistance to susceptibility have
been observed in Musa germplasm (Fouré, 1994).
Damage resulting from disease can be evaluated accurately only by measuring its
incidence and severity (Gauhl et al., 1994; Jones, 1994). Severity of the disease relates
to the intensity of damage to individual plants, while incidence deals with the
percentage of plants affected in a population.
The incidence and severity of black leaf streak disease on Musa can be assessed
in the laboratory or in the field. The laboratory method involves determining
ascospore and conidia production with the aid of a microscope, but this method is
tedious, very time-consuming and not very accurate.
Field evaluation under conditions of natural infection is the most common and
preferred method to assess incidence and severity of black leaf streak disease (Craenen,
1997). The nature and amount of lesions and the rate of their development on the
leaves is observed in the field. This method does not require a full understanding of
host-pathogen interactions, nor plant population systems and is therefore suitable for
field workers trained in symptom recognition.
Although field-screening methods are relatively simple, they are also timeconsuming and influenced by environmental factors, such as weather and soil, which
affect symptom expression. Therefore, it is recommended to compare the test plants
with reference cultivars and to gather a large number of observations to validate the
results.
Laboratory evaluation
The production of ascospores can be estimated by taking leaf samples with pseudothecia
from the same plant at different times, as described in Stover (1976). For conidiophores,
leaves at stage 2 of symptom development are collected in the field (Fouré, 1982).
However, results can vary widely from leaf to leaf; spots on each leaf can differ
enormously in number of spore-producing organs, and subjective errors are the source
of wrong results. Not being very valuable, this method is not described further.
An inoculation technique using leaf pieces under controlled conditions has proven
to be very simple and has many advantages (El Hadrami et al., 1998a, b). This technique
enables the production of the sexual and asexual phases of M. fijiensis and is very
useful to study host-parasite interactions (El Hadrami et al., 2000). The infection patterns
and symptom evolution were the same as those observed in the field, and the leaf
piece assay allowed the expression of resistant phenotypes. Furthermore, this method
may allow the epidemiological appraisal of partial resistance and the variability of
virulence in M. fijiensis.
Field evaluation
Field evaluation of disease severity requires knowledge not only of the stages of symptom
development and the percentage of leaf area spotted, but also of the different stages
of unrolling of the leaf (Craenen, 2001). Disease severity is evaluated by recording the
percentage of the leaf area that is spotted using a 7-category scale (modified from Stover
185
MyLsd 17x24
25/06/03
15:15
Page 186
Mycosphaerella leaf spot diseases of bananas: present status and outlook
and Dickson, 1970), which ranges from 0 (no symptoms) to 6 (51-100% leaf area with
symptoms) (Craenen 1997, 1998).
Epidemiology
In order to gain insight into the epidemiology of the disease, it is necessary to quantify
the amount of inoculum in an area. At the IITA High Rainfall Station in Onne, Nigeria,
spore-trapping and weather data were recorded for three years. This research indicated
that spore concentrations in the air were lower during the dry season and higher in the
rainy season. Ascospores were more frequent than conidiospores. During the dry season,
ascospore concentration was only three times as high as conidiospore concentration,
whereas during the rainy season ascospore concentration was found to be 40 times
higher (IITA, 1995).
Measuring the different characteristics of disease development on parents and their
hybrids in epidemiological studies led to the identification of different types of host
response and to the classification of Musa germplasm into different categories
according to the response to the disease. The different characteristics to be recorded
assume that M. fijiensis infects first the unfolded (> 10 cm) cigar leaf of the host plant
(Fullerton, 1994; Jones,1994). These characteristics are:
• Incubation time: calculated as the number of days between cigar leaf emergence
(Brun’s stage 2) and the appearance of the initial chlorotic fleck symptoms (i.e. depigmentation spot) relating to symptom stage 1 of Fouré’s scale. Brun’s stage 2 refers
to an upright cigar leaf, still strongly rolled and free from the petiole of the preceding
leaf, but not reaching its full length. At the first stage of symptom development only
minute yellowish specks (< 1 mm in length) are seen on the lower (abaxial) surface
of the leaf. They are not visible in translucent light.
• Evolution time: calculated as the number of days between first symptoms (Fouré’s
stage 1) and the occurrence of mature lesions (Fouré’s stage 6). At this last stage of
symptom development, the centre of the spot dries out and fades into a clear gray.
Often a black ring, surrounded by a yellow halo, encircles the gray centre. Necrotic
spots remain visible after the leaf has dried up completely.
• Disease development time: defined as the number of days elapsing between Brun’s
stage 2 of leaf emergence and Fouré’s stage 6 of symptoms, i.e. incubation time plus
evolution time.
• Lifetime of the leaf: recorded as the number of days between leaf emergence (Brun’s
stage 2) and leaf death.
• Youngest leaf spotted (YLS) at flowering: recorded as the first leaf (counting
downwards from the first top unfurled leaf) that shows spots (equal to or more than
10) with a necrotic dry center (Vakili, 1968).
The higher the YLS the more fully functional leaves on the plant, and hence, greater
resistance to the fungus (Craenen and Ortiz, 1997). The YLS score correlates significantly
with disease development time and other parameters to assess host plant response to
black leaf streak disease (Craenen, 1994). The YLS is also very easy to score, heritable,
and after scoring the level of host response can be defined, thereby allowing grouping
186
MyLsd 17x24
25/06/03
15:15
Page 187
Session 4
K. Craenen and R. Ortiz
of genotypes. Four distinct levels of host response to black leaf streak disease were defined
by Ortiz and Vuylsteke (1994) as follows:
susceptible (< 8 leaves without spots before or at flowering),
less susceptible (8–10),
partially resistant (> 10)
and highly resistant (none)
These epidemiological characteristics do not only depend on the amount of
inoculum present, but also on climatological factors that may affect the development
of the disease. Hence, it is essential to monitor environmental factors. Daily readings
from a weather station near to the experimental site are required. If there is no weather
station available, daily readings can be taken (if possible early in the morning) for the
following factors:
– rainfall (with a simple rain gauge);
– minimum and maximum temperature (with a minimum/maximum thermometer);
– relative humidity (with a hygrometer).
Temperature and relative humidity can also be recorded continuously with a
mechanical hygrothermograph.
Variability may be introduced in data sets from different locations due to
environmental or genetic causes (Ortiz et al., 1993). To minimise these effects the host
response was re-defined as the ratio between the YLS and the total number of standing
leaves (NSL) at the time of scoring. The index of non-spotted leaves (INSL) to assess
the host response to black leaf streak disease is calculated as follows:
INSL = 100 – 100 x [(NSL – YLS + 1)/NSL]
The INSL is the proportion of standing leaves without typical late stage symptoms
of the disease (i.e. spots with a necrotic centre). This index provides an estimation of
the available photosynthetic leaf area prior to fruit filling.
Early screening
Breeding resistance requires methods able to discriminate resistant and susceptible
genotypes at different stages of plant development (Leproive et al., 1993). In vitro
selection or field assessment using young plant materials are among some of the early
screening methods.
Inoculating Musa leaf tissue with a crude extract of M. fijiensis was suggested as
an early screening method (Hernández 1995). The crude extract screening method allows
a rapid (48 hours in greenhouse plants and 72 hours in callus tissue) identification of
host-plant resistance. For example, leaf tissue of susceptible cultivars such as ‘Grande
naine’ or ‘Currare’ showed the highest levels of phenolic compounds while resistant
germplasm (e.g. ‘Yangambi km5’, ‘Calcutta 4’ or ‘Saba’) showed low phenol content,
after being inoculated with a crude extract of M. fijiensis.
Pino (1997) indicated that 120 hours after inoculating in vitro plants with M. fijiensis
toxins, lesions in susceptible banana cultivar ‘Grande naine’ were larger than in putative
resistant mutants of ‘Grande naine’ (24.12 – 49.22 mm2 vs. 1.07 mm – 4.07 mm2,
respectively). Similarly, Okole and Schulz (1997) reported as promising an in vitro
selection technique using microsections (or callus cultures) of banana and plantain using
187
MyLsd 17x24
25/06/03
15:15
Page 188
Mycosphaerella leaf spot diseases of bananas: present status and outlook
a double selection system. The selection system uses in the first stage raw filtrate
concentrations (10-100 mg/L) of the fungus, and then a specific purified fungal toxin
(2,4,8-trihydroxyltetralone or 2,4,8-tht) isolated from M. fijiensis. This toxin plays an
important role in the development of necrotic leaf symptoms that causes host-specific
reactions depending on their concentration at different pathogenesis stages in tissue
culture materials (Hoss, 1998). A Musa genotype resistant to black leaf streak disease
with increased 2,4,8-tht content, produced the hypersensitive reaction and elicited
postinfectional defense reactions in the host plant, which led to its incompatibility with
the pathogen. A susceptible Musa genotype showed toxic doses of 2,4,8-tht but only
after the establishment of a compatible interaction, which first helped the biotrophic
nutrition of the pathogen, and acted as a virulence factor at the necrotrophic phase of
pathogenesis.
Toxin-resistant plantlets of two Cavendish cultivars (‘Williams’ and ‘Petite naine’)
and of ‘Horn plantain’ were regenerated using the above-mentioned method (Okole et
al., 2000). Rooted plants were further transferred to soil infected with suspensions of
M. fijiensis spores (0.3 g/ml). About 11 to 19% of the plantlets resistant to the toxin
were resistant to M. fijiensis in this culture chamber test, which reproduces the symptoms
of black leaf streak disease. However, the plants that withstood the toxin injection to
their tissues and the double selection procedure have not yet been field-tested.
As pointed out by Harelimana et al. (1996), screening with toxins to select resistant
germplasm has two major limitations. First, the lack of experimental evidence on the
role of toxins in disease development, and second, the susceptibility or resistance of
the cultured tissues do not reflect those of the adult plant because of the mode of action
of the toxin. It has not been demonstrated that toxins of M. fijiensis participate in the
initiation of infection or in the hypersensitive reaction of highly resistant adult plants.
Nonetheless, toxins could play a secondary role in pathogenicity, e.g. in disease
development in partially resistant cultivars. Research showed that chloroplasts could
be a precocious site of action of the toxins, suggesting that in vitro heterotrophic Musa
tissues may not be suitable for early screening.
Another early screening method for host response consists in using natural
inoculum on young Musa plants (3-month-old micropropagated plants). This method
confirmed the resistance of plantain hybrids and the susceptibility of their female plantain
parent to this disease (Mobambo et al., 1994). Furthermore, the characteristics associated
with disease development in these young plants were similar to those observed on adult
plants of the same genotypes (Mobambo et al., 1997). This method was also faster,
cheaper, less labour intensive and required less field space than screening of adult plants.
Highly susceptible germplasm can be rejected at a precocious stage with this early
screening method, thus reducing the sample size (and associated) costs for testing adult
plants in the field.
Host plant response to black leaf streak disease
Genetics of resistance
In plantains and bananas, resistance to black leaf streak disease is genetically
controlled. Genetic analysis has been carried out on diploid and tetraploid progenies
188
MyLsd 17x24
25/06/03
15:15
Page 189
Session 4
K. Craenen and R. Ortiz
obtained from triploid (plantain) x diploid crosses (Vuylsteke et al., 1993). The diploid
male parent was the resistant true-breeding line ‘Calcutta 4’ (Ortiz et al., 1998b).
Thus, the populations produced can be considered to be genetically equivalent to
test-crosses for the host response to this disease. Resistance to black leaf streak disease
is mainly the result of the interaction of three independent alleles: a recessive allele
at a major locus (bs1) and the alleles of at least two independent minor, modifying
genes with additive effects (bsri) (Ortiz and Vuylsteke, 1994). These genes have a
strong dosage effect at the tetraploid level that results in higher levels of resistance
in tetraploid than in diploid hybrids. Resistance genes are present in the genome of
susceptible plantains, but their expression is masked by the dominant effect of the
major gene for susceptibility (Ortiz and Vuylsteke, 1994).
These results were further confirmed by investigating tetrasomic segregration in
a cross between a resistant and a susceptible tetraploid hybrid (Ortiz, 2000). The
resistant maternal genotype was a nulliplex for the major resistance locus and the
paternal susceptible genotype was a duplex for the corresponding host response. Both
parents were balanced diallelic for the two minor modifier loci with additive effects.
The segregating tetraploid population from this cross showed a tri-modal frequency
distribution in the population, which was not significantly different to the expected
ratio (1.7 resistant : 1 susceptible) from the early genetic model defined by Ortiz
and Vuylsteke (1994).
Using the gene-for-gene hypothesis (Flor, 1971), a host-plant resistance system
based on recessive alleles is difficult to overcome by the pathogen as this requires
a mutation to the dominant allele of the virulence locus (Ortiz and Vuylsteke, 1994).
Since such mutations are rare (Simmonds, 1979), resistance based on recessive alleles
may prove to be durable.
Resistance genotypes are expressed phenotypically in the plants. Highly resistant
plants exhibit the longest incubation time and leaf life span as well as a hypersensitive
reaction to black leaf streak disease (Craenen and Ortiz, 1998). This extremely resistant
polygenic response (Ortiz and Vuylsteke, 1994) blocks disease development at an
early stage, thereby impeding the occurrence of mature necrotic lesions in the leaves.
In contrast, susceptible cultivars have a short incubation time, evolution time and
disease development time. This indicates that after infection, disease symptoms evolve
quickly into necrotic spots, resulting in extensive leaf death and defoliation.
Results from multilocational trials in Africa showed that all partially resistant
hybrids had a homeostatic host response to black leaf streak disease (Ortiz et al.,
1997). Also some of them achieved high and stable yields across environments due
to their resistance (Ortiz and Vuylsteke, 1995), even under low organic matter inputs
(Ortiz et al., 1995).
Mechanisms of resistance
A sample of 20 euploid Musa hybrids of various ploidy, exhibiting a range of resistant
and susceptible responses, were used to investigate the role of stomatal density, stomatal
length and the thickness of epicuticular wax in resistance to M. fijiensis (Craenen et
al., 1997). The female parents of these hybrids were susceptible plantains, while the
male parent was a wild, non-edible resistant banana (‘Calcutta 4’). Stomatal length was
189
MyLsd 17x24
25/06/03
15:15
Page 190
Mycosphaerella leaf spot diseases of bananas: present status and outlook
negatively correlated with the initial development (incubation time) of black leaf streak
disease in the leaves of young diploids but not in those of polyploid hybrids. Stomatal
density on the abaxial surface of young leaves was negatively correlated with
incubation time only in polyploids. Incubation time was positively correlated with the
accumulation of epicuticular wax in both diploid and polyploid hybrids. Although the
resistant male parent lacked epicuticular wax, derived hybrids possessed epicuticular
wax of various thickness which enhanced resistance. Hence, the two minor additive
modifier genes (bsri) which enhanced resistance may control decreased stomatal density
and increased leaf waxiness. Both characteristics may be two resistance mechanisms
that lengthen the incubation time of the disease in the leaves.
Craenen and Ortiz (1997) determined the role of the major gene for resistance (bs1)
in a sample of euploid hybrids from triploid-diploid crosses of two French plantains
and a diploid wild banana, and with a known genotype for the bs1 locus. Their host
response was assessed in the humid forest zone of Nigeria. Analysis of frequency
distribution in each segregating population showed that almost all the traits displayed
a normal distribution across ploidy level. This suggests that additive gene action plays
an important role in the host plant response to the fungus. However, the environment
and the genotype x environment interaction significantly affected the host response,
which explains the low reproducibility of all traits. Intrafamily variation was larger than
interfamily variation, and most of the genetic variation in each family depended on
individual genotypes, regardless of their ploidy. The additive effect of, and the
intralocus interaction at, the bs1 locus were established by one-way analysis of variance
and regression analysis. Intralocus interaction at the bs1 locus apparently regulates the
appearance of symptoms on the leaf surface, whereas the additive effect and the
intralocus interaction of the bs1 locus affect disease development in the host plant.
Therefore, the gene action(s) at the bs1 locus may provide durable resistance by slowing
down disease development.
Effect of black leaf streak disease on agronomic traits
Yield loss due to black leaf streak disease is 33 to 50% in plantain (Stover, 1983;
Mobambo et al., 1993) as a result of a reduced number of fruits per bunch and a lower
fruit weight. Black leaf streak disease has no effect on plant height and suckering, but
delays flowering and harvest by more than one month. The disease also causes premature
fruit ripening (Stover, 1980; Mobambo et al., 1993). In plantain landraces, normal fruit
filling or ripening time, i.e. time from flowering to harvest, is 91 days. Black leaf streak
disease significantly reduces this time. This premature fruit ripening is expected to have
adverse effects on postharvest characteristics, such as a reduced shelf life. In addition,
fruits are shorter and thinner, which generally results in lower quality fruit and lower
market value. Black leaf streak disease thus has a negative impact on fruit bulking,
probably as the result of a reduction in healthy leaf area (Mobambo et al., 1993).
Fruits of susceptible plantain hybrids are unable to bulk fully. Less susceptible
and partially resistant plantains are bigger, longer and heavier. Resistant plants
have increased bunch weight due to complete fruit filling as they have more
functional leaves for photosynthesis during the period between flowering and
harvest (Craenen and Ortiz, 1998).
190
MyLsd 17x24
25/06/03
15:15
Page 191
Session 4
K. Craenen and R. Ortiz
Genetic analysis of the plantain genome is difficult due to triploidy and high sterility.
As shown earlier, ploidy manipulations (scaling up and down the number of
chromosomes) and interspecific plantain-banana hybridization opened the path for the
genetic amelioration of the crop and for the investigation of its genome. There are several
associated effects of ploidy, parthenocarpy and resistance to black leaf streak disease
on growth and yield characteristics of euploid hybrids. The number of copies of the
resistance allele (bs1) and of the parthenocarpy gene (P1), as well as their intralocus
interaction and ploidy level, have all been found to significantly affect bunch and fruit
characteristics of euploid hybrids (Ortiz et al., 1998a). Epistasis significantly affected
fruit weight and size in one cross but not in another. Significant multiple regression
models combining ploidy and genetic markers explained 15% to 85% of quantitative
trait variation (QTV). The amount of QTV accounted by ploidy and genetic markers varied
according to the characteristic and cross in which the markers were examined.
Linear and multiple regression models, coefficients of determination, and DurbinWatson statistics were used by Craenen and Ortiz (1996) to determine the single and
combined effects of the major locus for resistance to black leaf streak disease (bs1)
and of ploidy on bunch weight, fruit weight, fruit length and fruit girth in the
progenies derived from crosses between a resistant diploid wild banana source and
the susceptible French plantain landraces. Differences in yield were mainly due to
changes in weight and circumference of the fruit, which are affected by the disease.
The combined effect of ploidy and resistance to black leaf streak disease was partially
responsible for QTV in yield. As a result of the gene interaction in the locus for
resistance (bs1), the partially resistant phenotypes showed higher yield than their
more susceptible full sibs.
The performance of 20 euploid hybrids was compared with that of their parents
to determine the influence of the disease on growth parameters and components of
yield (Craenen and Ortiz, 1998). There were significant differences among the hybrids
for all components of resistance, growth parameters, and yield components. For
diploid hybrids, which often had a short growth cycle or early flowering, or both,
the disease incubation time was significantly correlated with days to fruit filling
(P < 0.05). However, for tetraploid hybrids that had a long growth cycle and delayed
flowering, the correlation was not significant (P > 0.05). For diploid and tetraploid
hybrids, disease evolution time and disease development time were both correlated
(P < 0.05) with days to fruit filling. Bunch weight of tetraploid hybrids was correlated
(P < 0.05) with disease development time as scored by the youngest leaf spotted at
flowering (r = 0.933; P < 0.001). This result confirms that resistant hybrids with
potentially high yield could be selected efficiently by recording the youngest leaf
spotted at flowering.
Diploid or polyploid breeding
Most banana and all plantain cultivars grown by farmers in the tropics are
triploids. Euploid hybrids derived from triploid-diploid crosses are either mostly
diploids but some are tetraploids. They are seldom triploids. The most important
goal in genetic population improvement programs for managing disease resistance
in large populations is to enhance the frequency of favourable alleles controlling
191
MyLsd 17x24
25/06/03
15:15
Page 192
Mycosphaerella leaf spot diseases of bananas: present status and outlook
the desired characteristics. Recurrent selection methods are the most common for
increasing the frequency of favourable alleles in a cyclic fashion.
Theoretically, sq2 recessive individuals will be selected out from a large diploid
population where q is the allele frequency of a recessive gene and s is the intensity
of selection against the recessive genotype. Similarly, in a large tetraploid population,
sq4 represents the proportion of recessive genotypes that are selected out. Hence,
the change of allele frequency appears to be faster at the diploid than at the tetraploid
level. This occurs because the recessive allele is only hidden in the heterozygous
genotype at the diploid level while it will be included in the triplex, duplex and
simplex at the tetraploid level. For example, if q = 0.5 and selection efficiency s =1
at both ploidy levels, then the change in allele frequency after selection will be four
times larger in the diploids than that for the tetraploids (25% vs 6.25%, respectively).
Population genetics theory also suggest that the smaller the s, the lower response
to selection. When there are no escapes during screening, i.e. maximum selection
efficiency, increases in the frequency of favourable allele will be maximum. With
increased rates of escapes, breeding becomes inefficient and to the point that it may
become worthless. As pointed out by Mendoza (1988) “the degree of success in altering
the genotypic structure of the population, by modifying its gene frequency, is a function
of the precision in identifying and isolating the individuals carrying the attributes
under selection. Any errors or ‘escapes’ during the process, depending on their
magnitude, could alter the response to selection… A breeding effort can only be as
efficient as the screening procedure permits.”
Selection appears to be more effective in the early cycles when the frequency of
the favourable allele is low, especially at the diploid level. However, if there are
escapes owing to unreliable screening methods, particularly for a small population
size, then a lowering of the response to selection will occur at a very low frequency
of the favourable allele. With tetraploids, when the frequency of the favourable allele
exceeds 0.4, the response to selection falls rapidly. When the frequency reaches 0.8
in the tetraploid population, the response to selection becomes practically nil for
breeding purposes.
Outlook
The results of this research support the early views of Musa breeders (Ortiz, 1997;
Vuylsteke 2000), who claimed that a broad-based, improved Musa germplasm with
pest and disease resistance was necessary to achieve the sustainable production of
this vegetatively propagated perennial crop. This germplasm was obtained by using
conventional crossbreeding and may be further enhanced with the utilization of
innovative methods for the introduction of additional genetic variation, e.g. ideotype
breeding, polycross mating design or marker-aided introgression (Ortiz, 2001). In
short, “the prospects of banana and plantain breeding are unlimited and increased
efforts will at once initiate a new phase of Musa evolution” (Vuylsteke, 2001).
Manipulation of the Musa genome for its genetic betterment will be also facilitated
by the available knowledge on the inheritance of most important characteristics in
plantain and banana (Ortiz, 2000). Likewise, the information regarding fungal
diseases, such as black leaf streak disease, and the interactions between the
192
MyLsd 17x24
25/06/03
15:15
Page 193
Session 4
K. Craenen and R. Ortiz
pathogen and its host plant, provide the basis for a rational integrated management
strategy to control the disease. For example, the partial resistance provided by the
bs1 gene (Craenen and Ortiz, 1997; Ortiz and Vuylsteke, 1994) can be easily
incorporated into mixed cultivar systems common among the resource-poor farmers
in the tropics (IITA, 1998). These farmers prefer cropping systems that provide
intraspecific (cultivar mixtures) and interspecific (inter-cropping) diversity to
maximise land, use labour efficiently and minimise the risk of crop failure.
Deploying resistant hybrids in farmers’ cropping systems in association with their
own landraces is regarded as non-disruptive (IITA, 1999). In this suggested cultivar
mixture, the resistant hybrids serve as inoculum traps that reduce the spread of the
disease to the susceptible plantain landraces and may increase the bunch weight of
the landraces that are preferred by farmers due to their culinary and rheological
characteristics (IITA, 2000). On-farm participatory research undertaken by IITA will
provide more insights into this proposal for deployment of resistance to black leaf
streak disease using a cultivar mixture system (IITA, 2001). Data are still being
recorded in a farm in south-eastern Nigeria.
A cultivar mixture system would preserve genetic diversity and provide new, highyielding hybrids that may be incorporated in the local diet through novel processing
methods. Introducing new cultivars may lead to losses of diversity in farmer’s fields
(Sharrock et al., 2000), particularly when single-cultivar plantations are preferred
over mixed farming. Furthermore, diseases, e.g. black leaf streak disease, may spread
quickly into single-cultivar plantations of susceptible germplasm or when resistance
breaks down in improved germplasm. Hence, cultivar mixtures may provide an
“insurance” for a sustainable farming system.
Pyramiding genes from distinct germplasm sources may also enhance partial
resistance in plantain and banana. For example, IITA hybrids, which show this kind
of resistance, have alleles for resistance to black leaf streak disease from two sources:
triploid plantains and diploid bananas. The resistance alleles are masked by intra
and interlocus interactions in highly susceptible plantain parent landraces. The
resistance alleles are mainly from the wild banana accession ‘Calcutta 4’ or the diploid
banana cultivar ‘Pisang lilin’ (Hartman and Vuylsteke, 1999). The search for other
sources of resistance against a wide range of strains (e.g. Fullerton and Olsen, 1991,
1993) appears mandatory to develop a strategy for durable resistance to black leaf
streak disease. This requires urgent attention because resistance in at least two
cultivars (‘Paka’ and ‘T8’) broke down after eight years of cultivation in the Cook
Islands (Hartman and Vuylsteke, 1999).
The incorporation of different resistance types in the same genotype could
potentially confer durable resistance to black leaf streak disease. Indeed, resistance
alleles may be more stable depending on their mode of action and the particular
resistance they control. For example, ‘Calcutta 4’ has a polygenic hypersensitive
response that stops all development of the pathogen, but it also possesses recessive
alleles controlling partial resistance in its plantain hybrids. This partial resistance
simply slows disease development and may be more difficult to circumvent than
the hypersensitive response, which already failed when screening young plants with
Papua New Guinea strains. Hypersensitive responses are often associated with a genefor-gene host-pathogen interaction (Flor, 1971), but this hypothesis has not been
193
MyLsd 17x24
25/06/03
15:15
Page 194
Mycosphaerella leaf spot diseases of bananas: present status and outlook
tested in Musa. Resistance to black leaf streak disease provided by the recessive bs1
gene may be stable in the host-plant because virulence requires a rare mutation to
the dominant allele in the respective locus of the pathogen.
In conclusion, improved propagules with partial resistance to black leaf streak
disease coming from distinct genetic sources, along with crop husbandry techniques
are part of a holistic approach for long-term, sustainable productivity in Musa
farming systems.
Acknowledgements
To Sarah and Yannick Vuylsteke, who kindly allowed the senior author to take some
of their time in order to write some of the reports included in this review article. The
Directorate General of International Cooperation (DGIC, Belgium) supported this
research at the International Institute of Tropical Agriculture (IITA, Nigeria).
References
Carlier J., M.H. Lebrun, M.F. Zapater, C. Dubois and X. Mourichon. 1996. Genetic structure
of the global population of banana black leaf streak fungus Mycosphaerella fijiensis.
Molecular Ecology 5:499-510.
Carlier J., X. Mourichon, D. González de León, M.F. Zapater and M.H. Lebrun. 1994. DNA
restriction fragment length polymorphism in Mycosphaerella fijiensis that cause banana
leaf spot diseases. Phytopathology 84:751-756.
Carlier J., A. El Hadrami, H. Hayden, M.F. Zapater and F. Lapeyre. 1999. Population study of
Mycosphaerella fijiensis and genetic improvement of bananas for resistance to black leaf
streak disease. Pp. 40 in Abstracts of the International Symposium on the Molecular and
Cell Biology of Banana, Cornell University, New York.
Craenen K. 1994. Assessment of black sigatoka resistance in segregating progenies. MusAfrica
4:4-5.
Craenen K. 1997. Technical Manual on Black Sigatoka Disease of Banana and Plantain. IITA,
Ibadan, Nigeria. 23pp.
Craenen K. 1998. Black Sigatoka Disease of Banana and Plantain: A Reference Manual. IITA,
Ibadan, Nigeria. 60pp.
Craenen K. 2001. Black Sigatoka Resistance in Plantain-Banana Hybrids: Assessment,
Genetics, Resistance Mechanisms and their Effect on Yield. PhD Thesis, Katholieke
Universiteit Leuven, Belgium. 155pp.
Craenen K., J. Coosemans and R. Ortiz. 1997. The role of stomatal traits and epicuticular wax
in resistance to Mycosphaerella fijiensis in banana and plantain Musa spp. Tropicultura
15:136-140.
Craenen K. and R. Ortiz. 1996. Effect of the black sigatoka resistance locus bs1 and ploidy
level on fruit and bunch traits of plantain-banana hybrids. Euphytica 87:97-101.
Craenen K. and R. Ortiz. 1997. Effect of the bs1 gene in plantain and banana hybrids in response
to black sigatoka. Theoretical and Applied Genetics 95:497-505.
Craenen K. and R. Ortiz. 1998. Influence of black sigatoka disease on the growth and yield
of diploid and tetraploid plantains. Crop Protection 17:13-18.
Craenen K., R. Ortiz, E. Karamura and D. Vuylsteke (eds). 2000. Proceedings of the
First International Conference on Banana and Plantain for Africa. Acta Horticulturae 540.
589pp.
194
MyLsd 17x24
25/06/03
15:15
Page 195
Session 4
K. Craenen and R. Ortiz
Dangl J. and E. Holub. 1997. La Dolce Vita: a molecular feast in plant-pathogen interactions.
Cell 91:17-24.
El-Hadrami A., M.F. Zapater, F. Lapeyre, X. Mourichon and J. Carlier. 1998a. A leaf disk assay
to assess partial resistance of banana germplasm and aggressiveness of Mycosphaerella
fijiensis, the causal agent of black leaf streak disease in The 7th International Congress
of Plant Pathology, ICPP98, Vol. 2, Edinburg, Scotland, 9-16 August 1998.
El-Hadrami A., M.F. Zapater, F. Lapeyre, C. Abadie, X. Mourichon and J. Carlier. 1998b.
Evaluation sur fragments foliaires en survie de la résistance partielle du bananier et de
l’agressivité de Mycosphaerella fijiensis, agent causal de la maladie des raies noires in
Recontres de Mycologie-Phytopatologie, Aussois, 27 September – 1 October 1998. 1p.
El-Hadrami A., C. Abadie and J. Carlier. 2000. Evaluation de la résistance partielle du bananier
à Mycosphaerella fijiensis (maladie des raies noires) en conditions contrôlées et au champ
in Rencontres de Mycologie-Phytopatologie, Aussois, 3-9 March 2000. 1p.
Flor H.H. 1971. Current status of the gene-for-gene concept. Annual Review Phytopathology
9:275-296.
Fluhr R. 2001. Sentinels of disease. Plant resistance genes. Plant Physiology 127:1367-1374.
Fouré E. 1982. Les cercosporioses du bananier et leurs traitements. Etude de la sensibilité
variétale des bananiers et plantains à Mycosphaerella fijiensis Morelet au Gabon (maladie
des raies noires). I. Incubation et évolution de la maladie. Fruits 37:749-771.
Fouré E. 1985. Les cercosporioses du bananier et leurs traitements. Etude de la sensibilité
variétale des bananiers et plantains à Mycosphaerella fijiensis Morelet au Gabon (maladie
des raies noires). III. Comportement des variétés. Fruits 40:393-399.
Fouré E. 1994. Leaf spot diseases of banana and plantain caused by Mycosphaerella musicola
and M. fijiensis. Pp. 37-46 in The Improvement and Testing of Musa: A Global
Partnership (D.R. Jones, ed.). INIBAP, Montpellier, France.
Fullerton R.A. 1994. Sigatoka leaf diseases. Pp. 12-14 in Compendium of Tropical Fruit
Diseases (R.C. Ploetz, G.A. Zentmeyer, W.T. Nishijima, K.G. Rohrbach and H.D. Ohr, eds).
APS Press, St. Paul, Minnesota.
Fullerton R.A. and T.L. Olsen. 1991. Pathogen variability in Mycosphaerella fijiensis Morelet.
Pp. 105-114 in Banana Diseases in Asia and the Pacific (R.V. Valmayor, B.E. Umaldi and
C. Bejosano, eds). INIBAP, Los Baños, Philippines.
Fullerton R.A. and T.L. Olesen. 1993. Pathogen diversity of Mycosphaerella fijiensis Morelet.
Pp. 201-211 in Breeding Banana and Plantain for Resistance to Diseases and Pests. (J.
Ganry, ed.). CIRAD – INIBAP, Montpellier, France.
Gauhl F., C. Pasberg-Gauhl, D. Vuylsteke and R. Ortiz.1994. Multilocational Evaluation of
Black Sigatoka Resistance in Banana and Plantain. IITA Research Guide 47. IITA, Ibadan,
Nigeria. 59 pp.
Harelimana G., P. Leproive, H. Jijakli and X. Mourichon.1996. Use of Mycosphaerella fijiensis
toxins for the selection of banana cultivars resistant to black leaf streak. Pp. 171-175 in
Meeting on Tropical Plants, Montpellier, France, 11-15 March 1996. EUCARPIA,
Montpellier, France.
Hartman J. and D. Vuylsteke. 1999. Breeding for fungal resistance in Musa. Pp. 83-92 in
Genetics and Breeding for Crop Quality and Resistance (G.T. Scarascia-Mugnozza,
E. Porceddu and M.A. Pagnotta, eds). Kluwer Academic Press, Dordrecth.
Hernández N.R. 1995. In vitro and greenhouse selection of Musa resistance in black sigatoka
(Mycosphaerella fijiensis Morelet). INFOMUSA 4(1):15-16.
Holub E.B. 2001. The arms race is ancient history in Arabidopsis, the wild flower. Nature
Reviews 2:516-527.
195
MyLsd 17x24
25/06/03
15:15
Page 196
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Hoss R. 1998. Untersuchugen zur Funktion and Spezifitat pilzlicher Sekundamertaboliten im
Pathosystem “schwarze Sigatokakrankheit” der Banane (Musa sp. – Mycosphaerella
fijiensis). PhD Thesis, 123 pp. (English abstract in Musarama 1999, 12(1):50)).
IITA. 1995. Plant Health Management Division Annual Report 1994. IITA, Cotonou, Benin
Republic.
IITA. 1998. Project 7 – Improving plantain- and banana-based systems. Annual Report 1997.
IITA, Ibadan, Nigeria.
IITA. 1999. Project 7 – Improving plantain- and banana-based systems. Annual Report 1998.
IITA, Ibadan, Nigeria.
IITA. 2000. Project 7 – Improving plantain- and banana-based systems. Annual Report 1999.
IITA, Ibadan, Nigeria.
IITA. 2001. Project 2 – Improving plantain- and banana-based systems. Annual Report 2000.
IITA, Ibadan, Nigeria.
Jones, D.R. (ed.). 1994. The Improvement and Testing of Musa: A Global Partnership. INIBAP,
Montpellier, France. 303pp.
Leister D., J. Kurth, D.A. Laurie, M. Yano, T. Sasaki, K. Devos, A. Graner and P. Schulze-Lefert.
1998. Rapid reorganization of resistance gene homologues in cereal genomes. Proceedings
National Academy Sciences (USA) 95:370-375.
Leproive P., C.P. Acuña and A.S. Riveros. 1993. Screening procedures for improving resistance
to banana black leaf streak disease. Pp. 213-220 in Breeding Banana and Plantain for
Resistance to Diseases and Pests. (J. Ganry, ed.). CIRAD – INIBAP, Montpellier, France.
Mendoza. H.A. 1988. Progress in resistance breeding in potatoes as a function of efficiency
of screenig procedures. Pp. 39-64 in Bacterial Diseases of the Potato. Centro Internacional
de la Papa, Lima, Perú.
Mobambo K.N., F. Gauhl, D. Vuylsteke, R. Ortiz, C. Pasberg-Gauhl and R. Swennen. 1993.
Yield loss in plantain from black sigatoka leaf spot and field performance of resistant
hybrids. Field Crops Research 35:35-42.
Mobambo K.N., C. Pasberg-Gauhl, F. Gauhl and K. Zuofa. 1994. Early screening for black
leaf streak/black sigatoka disease resistance under natural inoculation conditions.
INFOMUSA 3(2):14-16.
Mobambo K.N., C. Pasberg-Gauhl, F. Gauhl and K. Zuofa. 1997. Host response to black sigatoka
in Musa germplasm of different ages under natural inoculation conditions. Crop
Protection 16:359-363.
Muller R., C. Pasberg-Gauhl, F. Gauhl, D. Kaemmer and G. Kahl. 1995. Tracing microsatellite
polymorphisms within the Nigerian populations of Mycosphaerella fijiensis. INFOMUSA
4(1):9-11.
Muller R., C. Pasberg-Gauhl, F. Gauhl, D. Kaemmer and G. Kahl. 1997. Oligonucleotide
fingerprinting detects genetic variability at different levels in Nigerian Mycosphaerella
fijiensis. Journal of Phytopathology 145:25-30.
Okole B., C. Memela, S. Rademan, K.J. Kunert and M. Brunette. 2000. Non-conventional
breeding approaches for banana and plantain against fungal diseases at AECI. Acta
Horticulturae 540:207-214.
Okole B. and F.A. Schultz. 1997. Selection of Mycosphaerella fijiensis – resistant cell lines
from micro-cross sections of bananas and plantains. Plant Cell Reports 13:339-342.
Ortiz R. 1997. Secondary polyploids, heterosis and evolutionary crop breeding for further
improvement of the plantain and banana genome. Theoretical and Applied Genetics
94:1113-1120.
Ortiz R. 2000. Understanding the Musa genome: an update. Acta Horticulturae 540:157-168.
196
MyLsd 17x24
25/06/03
15:15
Page 197
Session 4
K. Craenen and R. Ortiz
Ortiz R. 2001. Dedication: Dirk. R. Vuylsteke: Musa scientist and humanitarian. Plant Breeding
Reviews 21:1-25.
Ortiz R., K. Craenen and D. Vuylsteke. 1998a. Ploidy manipulations and genetic markers as
tools for analysis of quantitative trait variation in progeny derived from triploid plantain.
Hereditas 126:255-259.
Ortiz R., J.H. Crouch, D.R. Vuylsteke, R.S.B. Ferris and J. Okoro, 2000. Cultivar development,
genotype x environment interaction and multi-site testing of improved plantain and
banana germplam in sub-Saharan Africa. Pp. 84-106 in Genotype-by-environment
interaction analysis of IITA mandate crops in sub-Saharan Africa (I.J. Ekanayake and
R. Ortiz, eds). IITA, Ibadan, Nigeria..
Ortiz R., J. Okoro, R. Apanisile and K. Craenen. 1995. Preliminary assessment of the yield
potential of Musa hybrids under low external organic matter input. MusAfrica 7:15-17.
Ortiz R. and D. Vuylsteke. 1994. Inheritance of black sigatoka resistance in plantain-banana
(Musa spp.) hybrids. Theoretical and Applied Genetics 89:146-152.
Ortiz R. and D. Vuylsteke. 1995. Genotype-by-environment interaction in Musa germplasm
revealed by multi-site evaluation in sub-Saharan Africa. HortScience 30:795.
Ortiz R., D. Vuylsteke and J.H. Crouch. 1998b. Musa genetics, ‘Calcutta 4’ and scientific ethics:
reply to Shepherd’s letter. INFOMUSA 7(2):31-32.
Ortiz R., D. Vuylsteke, R.S.B. Ferris, J.U. Okoro, A. N’Guessan, O.B. Hemeng, D.K. Yeboah, K.
Afreh-Nuamah, E.K.S. Ahiekpor, E. Fouré, B.A. Adelaja, M. Ayodele, O.B. Arene, F.E.O.
Ikiediugwu, A.N. Agbor, A.N. Nwongu, E. Okoro, G.O. Kayode, I.K. Ipinmoye, S.A. Akele
and A. Lawrence. 1997. Developing new plantain varieties for Africa. Plant Varieties and
Seeds 10:39.57.
Ortiz R., D. Vuylsteke, J.U. Okoro, R.S.B. Ferris, O.B. Hemeng, D.K. Yeboah, C.C: Anojulu, B.A.
Adelaja, O.B. Arene, A.N. Agbor, A.N. Nwongu, G. Kayode, I.K. Ipinmoye, S.A. Akele and
A. Lawrence. 1993. Host response to black sigatoka across West and Central Africa.
MusAfrica 3:8-10.
Pino J.A. 1997. Selección temprana de mutantes de banana y plátano resistentes a
Mycosphaerella fijiensis mediante fitotóxinas. Agrotecnia de Cuba 27:89-91.
Rommens C.M. and G.M. Kishore. 2000. Exploiting the full potential of disease resistance
genes for agricultural use. Current Opinion in Biotechnology 11:120-125.
Sharrock S.L., J.-P. Horry and E. Frison. 2000. The state of use of Musa diversity. Pp. 223244 in Broadening the Genetic Base of Crop Production (H.D. Cooper, C. Spillane and T.
Hodgkin, eds). CABI Publishing – FAO – IPGRI, Wallingford.
Simmonds N.W. 1979. Principles of Crop Improvement. Longman, London and New York.
Simmonds N.W. and J. Smartt. 1999. Principles of Crop Improvement (2nd edition). Blackwell
Science, Oxford. Pp.227-261.
Stover R.H. 1976. Distribution and cultural characteristics of the pathogen causing banana
leaf spot. Tropical Agriculture (Trinidad) 53:111-114.
Stover R.H. 1980. Sigatoka leaf spot diseases of bananas and plantains. Plant Disease 64:
750-755.
Stover R.H. 1983. Effet de la cercosporiose noire sur les plantains en Amérique Centrale. Fruits
38:326-329.
Stover R.H. and J.D. Dickson. 1970. Leaf spot of banana caused by Mycosphaerella musicola:
methods of measuring spotting prevalence and severity. Tropical Agriculture (Trinidad)
47:289-302.
Vakili N.G. 1968. Responses of Musa acuminata species and edible cultivars to infection by
Mycosphaerella musicola. Tropical Agriculture (Trinidad) 45:13-22.
197
MyLsd 17x24
25/06/03
15:15
Page 198
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Vuylsteke D. 2000. Breeding bananas and plantains: from intractability to feasibility. Acta
Horticulturae 540:149-156.
Vuylsteke D. 2001. Strategies for utilization of genetic variation in plantain improvement.
PhD Thesis, Katholieke Universiteit Leuven, Belgium.
Vuylsteke D., R. Swennen and R. Ortiz. 1993. Development and performance of black sigatokaresistant tetraploid hybrids of plantains (Musa spp., AAB group). Euphytica 65:33-42.
198
MyLsd 17x24
25/06/03
15:15
Page 199
Session 4
C. Jenny et al.
Conventional breeding
of bananas
C. Jenny1, K. Tomekpé2, F. Bakry3 and J.V. Escalant4
Abstract
Whereas ancestral bananas are fertile diploids, the main groups of bananas grown today are clones
of plants, mostly triploids, which are reproduced entirely vegetatively and consequently difficult
to breed. Conventional breeding techniques have yielded new varieties conventional breeding can
accomplish only so much. Not all genetic combinations necessarily lead to useful hybrids. This
communication presents the strategies which have been and are still being used in this
field–namely the 3x/2x scheme and the creation of triploid hybrids from ancestral diploid
varieties–and draws attention to their strengths and limitations.
Resumen - Mejoramiento convencional de los bananos
Mientras que los bananos ancestrales son diploides fértiles, los principales grupos de bananos
que se cultivan actualmente son clones de las plantas, en su mayoría triploides, que se reproducen
solo vegetativamente y, en consecuencia, son difíciles de mejorar. Las técnicas de mejoramiento
convencional han producido nuevas variedades y el mejoramiento convencional tiene sur
límites. Todas las combinaciones genéticas no llevan necesariamente a generar híbridos útiles.
En este trabajo se presentan las estrategias que todavía están siendo utilizadas en este campo,
a saber, el esquema 3x/2x y la creación de los híbridos triploides a partir de las variedades
ancestrales diploides, y se destacan sus fortalezas y debilidades.
Résumé - Amélioration conventionnelle des bananiers
Alors que les bananiers ancestraux sont des diploïdes fertiles, les principaux groupes de
bananiers cultivés aujourd’hui sont des clones de plantes, principalement triploïdes, qui se
reproduisent selon un mode entièrement végétatif et sont donc difficiles à améliorer. Les
techniques classiques d’amélioration ont produit de nouvelles variétés, et l’amélioration
conventionnelle a ses limites. Toutes les combinaisons génétiques ne conduisent pas
nécessairement à des hybrides utiles. Cette communication présente les stratégies qui ont été,
et sont encore utilisées dans ce domaine, c’est-à-dire le schéma 3x/2x et la création d’hybrides
triploïdes à partir de variétés ancestrales diploïdes, et insiste sur leurs avantages et leurs limites.
1CIRAD-FHLOR, Guadeloupe
2CARBAP, Douala, Cameroon
3
CIRAD, Montpellier, France
4INIBAP, Montpellier, France
199
MyLsd 17x24
25/06/03
15:15
Page 200
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Introduction
Bananas and plantains are difficult crops to breed because most of the important
and popular varieties are highly sterile and therefore do not produce seeds.
Furthermore, compared to many other important food crops, there is a relative lack
of knowledge on Musa genetics and cytogenetics. Despite these constraints,
important progress has been made in the genetic improvement of Musa in recent
years and new varieties are now becoming available from breeding programmes.
Major breeding programmes that use conventional breeding methodologies are
located at the Fundación Hondureña de Investigación Agrícola (FHIA) in Honduras,
the Centre de Coopération Internationale en Recherche Agronomique pour le
Développement (CIRAD-FLHOR) in France and Guadeloupe, the International
Institute of Tropical Agriculture (IITA) in Nigeria and Uganda, the Centre Africain
de Recherches sur Bananiers et Plantains (CARBAP) in Cameroon and the Empresa
Brasiliera de Pesquisa Agropecuaria (EMBRAPA) in Brazil.
In the last decades new banana varieties were mainly created by using
conventional breeding techniques (Bakry and Horry, 1992; Jenny et al., 1994;
Menendez and Shepherd, 1975; Rowe and Rosales, 1992; Shepherd, 1968; Soares
Filho et al., 1992; Swennen and Vuylsteke, 1990; Tomekpé et al., 1998). One of the
peculiarities of bananas is the need to adapt these techniques to the genetics of
polyploid plants.
This communication presents the strategies which have been and are still being
used in this field, and draws attention to their strengths and limitations.
Constraints to the improvement of bananas
Whilst ancestral bananas are fertile diploids, the main groups of bananas grown
today are clones of plants, mostly triploids, which are reproduced entirely
vegetatively. For producers and consumers, this feature presents two advantages:
triploidy gives the plant vigour, making it easier to grow than diploids; clonal
propagation assures uniformity which facilitates management, both in the field and
throughout the distribution and sale chain. Finally, triploidy ensures sterility of the
fruit, enabling it to be eaten.
On the other hand, the nature of these plants also presents potential dangers
for their cultivation, and obstacles to their improvement. First, the genetic
uniformity of these plants facilitates the spread of diseases and increases the impact
of the latter on banana plants. For example, the ‘Cavendish’ varieties throughout
the world are all susceptible to leaf spot diseases. The same is true for plantains
with regard to black leaf streak disease, whether in Africa, South America or Asia.
Furthermore, the sterility of the clones currently grown is a considerable hindrance
to their genetic improvement.
It is therefore clear that the primary needs in terms of breeding have to do with
the various diseases which affect the crop. Among the most important is Fusarium
wilt (caused by Fusarium oxysporum f.sp. cubense), and Sigatoka disease (caused
by Mycosphaerella musicola) and black leaf streak (caused by Mycosphaerella
fijiensis). However there are also other improvement criteria, especially for dessert
200
MyLsd 17x24
25/06/03
15:15
Page 201
Session 4
C. Jenny et al.
bananas for export, which have to do with fruit quality and post-harvest
characteristics.
One of the first ideas was to draw on the natural existing genetic resources
to find solutions for replacement varieties. Thus, during the 20th century, all of the
‘Gros Michel’, which was traditionally grown on commercial plantations, was
gradually replaced by ‘Cavendish’ which is resistant to race 1 of Fusarium wilt. In
certain areas, attempts were made to introduce ‘Pisang awak’ and ‘Bluggoe’ to replace
or supplement the plantains susceptible to black leaf streak. These solutions are
unfortunately of limited value because the natural varieties suitable for cultivation
are very few, and are susceptible to other parasites, such as nematodes and weevils.
Often the fruit produced is not the kind favoured locally and fails to gain acceptance.
Finally, the problem of potential susceptibility ends up being passed on to the new
clones through the creation of a new genetic uniformity which can easily be
circumvented by the pathogens. The need for genuine genetic improvement is
therefore real.
An improvement strategy: the 3x/2x scheme
The prerequisite to any genetic improvement strategy is to analyse the available tools.
For conventional methods, it is necessary to identify the required qualities of fertility
and useful characters of the germplasm which can be used. Since the 1920s, the
questions tackled have been: which cultivars can be improved, and what
characteristics can be introduced into them. Detailed analysis of the fertility of the
main triploid cultivars resulted in the identification of a certain number of triploid
clones which had retained residual female fertility, like ‘Gros Michel’ (Musa cv. AAA)
and the French-type plantains (Musa cv. AAB). Conversely, certain clones turned
out to be completely sterile, like ‘Cavendish’1 (Musa cv. AAA).
The search for sources of resistance to the main diseases was pursued by
looking for wild varieties present in the area of origin of bananas, Southeast Asia.
In this way, the variety Musa acuminata ssp. burmannicoïdes ‘Calcutta’ 4, among
others, was identified, notably for its resistance to M. musicola and to
M. fijiensis.
Armed with these tools, breeders crossed the various parents among themselves
to improve the cultivated triploid clones, by combining the residual female fertility
of the latter with the strong male fertility of the wild varieties (Figure 1). Detailed
analysis of the mechanisms brought into play during these hybridizations has
shown that the fertile triploid parent produced gametes during meiosis with a v
ariable chromosome number, ranging from 11 to 66. Among these gametes, those
with 33 chromosomes, i.e. with a true restitution nucleus, were particularly useful.
Among the progeny the first priority was to look for tetraploid hybrids, the result
of fusion between this restitution nucleus and a normal haploid gamete from the
wild parent. The final result resembled more a fusion than a true hybridization,
because recombination on the triploid side is low, and on the wild parent, being
very homozygous, produces very homogeneous gametes. In this way hybrid
1
It is now known that under certain stress conditions, it is possible to obtain pollen from ‘Cavendish’ clones.
201
MyLsd 17x24
25/06/03
15:15
Page 202
Mycosphaerella leaf spot diseases of bananas: present status and outlook
tetraploids were produced with a genome similar to the complete genome of the
triploid parent one is seeking to improve, supplemented by a wild haploid genome
contributing sources of resistance. A certain degree of recombination would however
explain the sometimes mediocre quality of the tetraploid hybrids obtained.
Male parent
Female parent
2n = 3x = 33
2n = 2x = 22
MEIOSIS
Male parent
Female gametes
n = variable
n = 11
Some non reduced gametes
GAMETE FUSION
Progeny: mix of euploid and eneuploid embryos
2n = 22, 33 … 44 ...77
Tetraploid hybrid
Figure 1. Scheme for creating tetraploid hybrids from triploid and diploid parents.
Two complementary methods exist to improve this strategy:
• It is possible to produce diploid hybrids which appear to be useful, and which
can themselves be used in an improvement programme.
• The produced tetraploid hybrids, being more fertile than the triploid parent, can
be reintroduced in the crossing schemes with a view to creating secondary triploid
hybrids. One must not however forget that in this latter case the genetic gain obtained
from nuclear restitution will be reduced by recombination which will occur during
the meiosis of this tetraploid.
Nevertheless, many research organizations throughout the world produced hybrids
using this scheme. As for cooking banana hybrids, the cultivars ‘FHIA-21’ (FHIA,
Honduras), ‘CRBP-39’ (CARBAP, Cameroon) and ‘Bita-3’ (IITA, Nigeria) (Ortiz and
Vuylsteke, 1998) should be mentioned. In the majority of cases, these hybrids were
found to be resistant to both Sigatoka disease and black leaf streak disease, which was
202
MyLsd 17x24
25/06/03
15:15
Page 203
Session 4
C. Jenny et al.
the primary aim of their breeding. This immediate success can be attributed to the
type of resistance used. Very often the wild male parent chosen was ‘Calcutta 4’ which
is highly resistant (HR) to leaf spot diseases. It is generally accepted that this type of
resistance depends on a smaller number of genes than partial resistance (PR), and
consequently can be more easily transmitted during hybridization. The result however
poses two questions: the exact transmission mechanism of these resistances is not
known; and as a corollary, one cannot predict the durability of these resistances over
time. Logically one would prefer to try to introduce into these hybrids partial resistance,
known to be more durable but more complex, which has, for example, been tried by
using parents such as M-53.
The 3x x 2x strategy described here also suffers from several limitations.
• From a purely technical point of view, this strategy depends on the existence of
valuable triploid clones exhibiting exploitable female fertility. There are relatively
few of these, limiting the possibilities of enlarging the genetic base of these crossings.
• The hybrid populations obtained by crossing are usually small.
• The percentage of tetraploid hybrids being small, the possibilities for selection
are very limited. This technique cannot be used to work simultaneously on a large
number of characters to be improved.
More recently, the realization of the existence of potentially active sequences of
banana streak virus (BSV) contained within the balbisiana genome has further reduced
the possibilities for using this strategy. Moreover, the most fertile triploid genomes
are often of the AAB or ABB types. Such germplasm should not be used for breeding
as long as the activation mechanisms of BSV are unknown. One might nevertheless
use it to produce various diploid type AA hybrids, notably plantain diploids which
constitute 50% of the 3x/2x descendants produced by certain plantain cultivars.
Several dozen of these hybrids produced at CARBAP were found to be negative for
IC-PCR and have never expressed any BSV symptoms in the field even under stress
conditions in which nearly all tetraploid hybrids frequently present symptoms.
Lastly, the tetraploid nature of the hybrids formed often leads to problems of fruit
quality. In dessert-type hybrids the firmness of the pulp is less in the tetraploids.
This problem also exists in cooking hybrids, albeit less pronounced. In numerous
polyploid plants, it has been found that the water content of the cells increases with
ploidy level. If this proves true for bananas, it might explain the phenomenon.
Moreover, at the tetraploid level, female fertility might be restored, often leading to
the presence of seed in the fruit, which is unacceptable to present-day consumers.
Recent developments with biotechnological tools and a better understanding of
the evolutionary processes of bananas have led to the introduction of another
breeding strategy aimed at the production of triploid hybrids.
Creation of triploid hybrids from ancestral diploids
The natural emergence of triploid cultivars derived from ancestral diploid varieties
is due to the accidental production of unreduced gametes in one of the diploid
parents during hybridization (Simmonds, 1962). The 4x x 2x strategy is a copy of
203
MyLsd 17x24
25/06/03
15:15
Page 204
Mycosphaerella leaf spot diseases of bananas: present status and outlook
this natural evolutionary process. The meiotic error leading to these unreduced
gametes is replaced by a chromosome doubling in one of the parents using
colchicine (Vakili, 1967; Stover and Buddenhagen, 1986; Bakry et al., 1997).
Unlike the scheme just described, this strategy does not attempt to improve
existing varieties, but rather to create new improved varieties, close to the
established targets, using ancestral varieties. These new hybrids should therefore
combine all the classic characteristics of the banana requiring improvement, plus
the improved characters for which the strategy was introduced.
The triploid hybrids are obtained by simple hybridisation between a diploid
parent and a tetraploid parent (Figure 2). This hybridisation has to be the end of
the procedure, since the product obtained is almost completely sterile, and can
thus no longer be improved by conventional means. The tetraploid parent has
previously been obtained by doubling with colchicine an ancestral diploid or an
improved diploid. After treatment, the purely tetraploid nature of the parent is
checked by flow cytometry. The success of the strategy rests on the judicious choice
of the parents. These can either be natural diploid cultivars or improved diploid
cultivars.
Tetraploid development
AAw
x
AAcv
colchicine
AAcv
AAcv
x
AAcv
AAcv
x
BBw
AAAAcv
colchicine
ABcv
AABBcv
Triploid development
AAAAcv
AAAAcv
AAw
x
x
AAAAcv
AAAAcv
BBw
x
AAAAcv
AAcv
x
AABBcv
AABcv
AABcv
AAw
x
AABBcv
AABcv
AAcv
1. Nearly 20 clones developed at CARBAP and CIRAD.
2. Currently stopped because of BSV concern.
3. Annual hybrid population size of nearly 400 plants in the field.
4. 98% of the progeny is triploid.
5. Currently stopped because of SV concern.
Figure 2. Strategy for creating triploid hybrids from diploid material.
204
(3,4)
(3,5)
(1)
(2)
MyLsd 17x24
25/06/03
15:15
Page 205
Session 4
C. Jenny et al.
At CIRAD, where this method has been favoured for several years, progress has
been made in characterizing available genetic resources and understanding the
relationships between ancestral and cultivated varieties, thus facilitating the
definition of pools of parent lines according to the desired results (Jenny et al., 1999).
Moreover, the development of molecular tools has made, and continues to make,
this strategy more efficient. Among the most important results, it has been possible
to demonstrate the uniparental inheritance of cytoplasmic organelles, which helps
in identifying the phylogeny of bananas (Fauré et al., 1994). Musa is one of the few
species with biparental cytoplasmic inheritance: paternal inheritance of mitochondria
and maternal inheritance of chloroplasts.
The extent of genetic variability within the acuminata genome has been related
to the variability in fruit quality in the main cultivated groups. The most striking
example is probably the strict relationship between the sub-species M. acuminata
banksii and cooking bananas. It has thus been possible to produce triploid cooking
banana hybrids of purely acuminata origin. The variability of the acuminata genome
also permits variation in the type of fruit type and plant obtained, whether dessert
or cooking, but also with regards to the fruit’s sweetness or acidity, its length, the
plant’s number of suckers, and its yield, etc.
One of the main attractions of this strategy rests on making use of highly fertile
parents, thus leading to a large progeny. In preliminary results, CARBAP identified
in the 100 or so individuals of the progeny of a BB x AAA cross, about 20% of
hybrids having useful resistance to black leaf streak disease (BLSD).
To sum up, the 4x x 2x strategy presents a certain number of undeniable
advantages in genetics:
• The number of available parents is only limited by our knowledge of the
germplasm, knowing that subsequently pools of parents that produce the targeted
results will be formed.
• Using a tetraploid parent allows better control of the heritability of characters
due to limited recombination within the polyploid parent.
• Highly fertile parents lead to the production of large populations in which it is
possible to set up a true selection programme, possibly based on several
improvement criteria.
In order to improve this strategy, it would be valuable to further enlarge the
diploid crossing base either by crossing known diploids to obtain improved
diploids or by more collection missions in the regions of interest. CARBAP is
currently developing, from diploid plantain hybrids resistant to BLSD, second and
third generation improved diploids (secondary and tertiary diploids obtained by
crossing diploid plantain hybrids with different sources of resistance to BLSD) which
are in the process of chromosome doubling so as to be integrated into this strategy.
In the near future, the use of molecular markers should facilitate selection at the
parent level (identification of the genes to be transferred) and hybrid level
(identification of the genes effectively transferred). Nowadays for example, three
QTL (Quantitative Trait Loci) have already been localized in relation to resistance
to BLSD (Persley and George, 1999). Their use remains dependent on the completion
of the genetic map being created by CIRAD. Among the future challenges,
205
MyLsd 17x24
25/06/03
15:15
Page 206
Mycosphaerella leaf spot diseases of bananas: present status and outlook
a molecular characterization of fruit quality will be particularly important for the
creation of dessert-type varieties which are competitive in the world market.
Once again however, BSV acts as a brake on the best use of this strategy. The
B genome is excluded from breeding schemes because of possible activation of
integrated viral DNA sequences within the genome. It is known that activation is
linked to certain stresses including hybridisation and in vitro multiplication. These
two stresses do not necessarily activate the same integrated sequences, and could
therefore have additive effects (Lheureux et al., 2003). It is therefore very important
to study the B genome in more detail in order to be able, as soon as possible, to
use it again for creating varieties.
• The range of banana species is not so rich that we can manage for their
improvement without the one or two species from which cultivated bananas
originated.
• In particular, the use of M. balbisiana in crossing schemes confers resistance, in
particular to Mycosphaerella musicola and M. fijiensis, on the hybrids produced.
This increased vigour renders the plants less susceptible to growth stresses.
• Finally, the high natural fertility of M. balbisana is an advantage for the
production of large numbers of progeny.
Among the avenues to explore, one should mention collecting in the regions of
origin of the species M. balbisiana, the analysis of the germplasm present in
collections and genetic methods of improvement such as for example the extraction
of the B genome from interspecific clones. In each case, it will be essential to gather
international expertise and competence on the subject, and the PROMUSA workshop
on the diversity of the M. balbisiana genome held in Bangkok in 2002 was a good
starting point.
Conclusion
At a more technical level, it emerges from this presentation that conventional breeding
can accomplish only so much. Not all genetic combinations necessarily lead to useful
hybrids. Crossing cooking bananas with dessert ones, for example, generally leads
to intermediate hybrids of no real value. Not all combinations are possible, and many
simply do not work, when using conventional hybridization techniques.
Conventional breeding methods should be viewed as just a part - admittedly a
significant part - of a more general genetic improvement strategy. Unconventional
techniques can usefully complete this arsenal (Novak, 1992; Sagi et al., 1995;
Sharrock et al., 2000). For example, protoplast fusion is one of the ways which could
increase the possibilities of overcoming certain fertility barriers in combining parental
lines. This technique also allows transmission of intact genotypes by bypassing the
recombination phenomena associated with meiosis, and also results in modification
at the cytoplasmic level of the cells, thus potentially leading to novel results.
Mutagenesis and genetic transformation - aside from the arguments as to their
appropriateness - might improve either the parents or, at the other end of the chain,
the hybrids created by conventional hybridization and which lack certain characters.
A complementary approach, recombinant DNA, has led to the production of the first
206
MyLsd 17x24
25/06/03
15:15
Page 207
Session 4
C. Jenny et al.
transgenic banana and plantain, but also to the generation of a large number of
transgenic lines with agronomically useful genes. Transgenic plants transformed with
genes encoding antifungal proteins are currently available for field-testing.
It is essential that Musa breeders and biotechnologists work together to accelerate
improvement. In view of the limited resources being devoted to research into Musa
improvement, and knowing the scale of the problems to be overcome, it is important
to strengthen collaboration between the various institutions working on the problem,
and to take advantage of all the resources already available to develop research on
Musa. The Global Programme for Musa Improvement (PROMUSA) aims to bring
together all the scientists working on Musa genetic improvement, involving in the
same programme geneticists, biotechnologists, but also pathologists and physiologists.
References
Bakry F. and J.P. Horry. 1992. Tetraploid hybrids from interploid 3x/2x crosses in cooking
bananas. Fruits 47:641-655.
Bakry F., F. Carreel, M.L. Caruana, F.X. Côte, C. Jenny and H. Tézenas du Montcel 1997.
Les bananiers. Pp. 109-140 in L’amélioration des plantes tropicales (A. Charrier, M. Jacquot,
S. Hamon and D. Nicolas, eds). CIRAD and ORSTOM, Paris and Montpellier, France.
Fauré S., J.L. Noyer, F. Carreel, J.P. Horry, F. Bakry and C. Lanaud. 1994. Maternal inheritance
of chloroplast genome ant paternal inheritance of mitochondrial genome in bananas (Musa
acuminata). Current Genetics 25:265-269.
Jenny C., E. Auboiron and A. Beveraggi. 1994. Breeding plantain-type hybrids at CRBP.
Pp. 176-187 in The improvement and testing of Musa: a global partnership (D.R. Jones,
ed.). Proceedings of the first global conference of the IMTP held at FHIA, Honduras, 2730 April 1994. INIBAP, Montpellier, France.
Jenny C., F. Carreel, K. Tomekpé, X. Perrier, C. Dubois, J.P. Horry and H. Tézenas du Montcel.
1999. Les bananiers. Pp. 113-139 in Diversité génétique des plantes tropicales cultivées
(P. Hamon, M. Seguin, X. Perrier and J.C. Glaszmann, eds). CIRAD, Montpellier, France.
Lheureux F., F. Carreel, C. Jenny, B.E.L. Lockhart and M.L. Iskra-Caruana. 2003. Identification
of genetic markers linked to banana streak disease expression in interspecific Musa hybrids.
Theor. and Appl. Gen. 106(4):594-598.
Menendez T. and K. Shepherd. 1975. Breeding new bananas. World crops (May/June):
104-112.
Novak F.J. 1992. Musa (bananas and plantains). Pp. 449-487 in Biotechnology of perennial
fruit crops (F.A. Hammerschlag and R.E Litz, eds). CAB International, Wallington, UK.
Ortiz R. and D. Vuylsteke. 1998. ‘Bita-3’: a starchy banana with partial resistance to black
Sigatoka and tolerance to streak virus. HortScience 33:358-359.
Persley G.J. and P. George (eds). 1999. Banana, Breeding and Biotechnology - Commodity
advances through banana improvement project research, 1994 - 1998. The World Bank,
Washington D.C. 62pp.
Rowe P. and F. Rosales. 1992. Genetic improvement of bananas, plantains and cooking bananas
in FHIA, Honduras. Pp. 243-266 in Breeding bananas and plantains (J. Ganry, ed.).
Proceedings of an International Symposium on Genetic Improvement of Bananas for their
Resistance to Diseases and Pests. CIRAD-FLHOR, Montpellier, France.
207
MyLsd 17x24
25/06/03
15:15
Page 208
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Sagi L., B. Panis, S. Remy, H. Schoofs, K. de Smet, R. Swennen and B.P.A. Cammue. 1995.
Genetic transformation of banana and plantain (Musa spp.) via particle bombardment.
Biotechnology 13:481-485.
Sharrock S., J.P. Horry and E.A. Frison. 2001. The state of the use of Musa diversity. Pp. 223243 in Broadening the genetic base of crop production (H.D. Cooper, C. Spillane and
T. Hodgkin, eds). IPGRI/FAO, Rome , Italy.
Shepherd K. 1968. Banana breeding in the West Indies. Pest articles and news summaries
14:370-379.
Simmonds N.W. 1962. The evolution of the bananas. Longman, Green & Co, London, UK .
Soares Filho W., S. Dos, Z.J.M. Cordeiro, K. Shepherd, J.L.L. Dantas, S. de Oliveira e Silva
and M.A.P. da Cunha. 1992. The banana genetic improvement programme at
CNPMF/EMBRAPA, Brazil. Pp. 339-346 in Breeding bananas and plantains (J. Ganry, ed.).
Proceedings of an International Symposium on Genetic Improvement of Bananas for their
Resistance to Diseases and Pests. CIRAD-FLHOR, Montpellier, France.
Stover R.H. and I.W. Buddenhagen. 1986. Banana breeding: polyploidy, disease resistance
and productivity. Fruits 41:175-191.
Swennen R. and D. Vuylsteke. 1990. Aspects of plantain breeding at IITA. Pp. 252-266
in Sigatoka leaf spot disease of bananas (R.A. Fullerton and R.H. Stover, eds). Proceedings
of an international workshopheld at San José, Costa Rica, 28 March-1 April 1989. INIBAP,
Montpellier, France.
Tomekpé K., N. Noupadja, C. Abadie, E. Auboiron and J. Tchango Tchango. 1998. Genetic
improvement of plantains at CRBP: performance of black Sigatoka resistant plantain
hybrids. Pp. 45-50 in Actas: Seminario Internacional sobre Producción de Plátano.
4-8 de Mayo 1998, Armenia, Quindío. CORPOICA, Colombia.
Vakili N.G. 1967. The experimental formation of polyploidy and its effect in the genus Musa.
Amer. J. Bot. 54:24-36.
208
MyLsd 17x24
25/06/03
15:15
Page 209
Session 4
R. Swennen et al.
Transgenic approaches for resistance
to Mycosphaerella leaf spot diseases
in Musa spp.
R. Swennen1, G. Arinaitwe1, B.P.A. Cammue2, I. François2, B. Panis1,
S. Remy1, L. Sági1, E. Santos1, H. Strosse1 and I. Van den Houwe1
Abstract
In smallholdings, average banana and plantain yields per unit have not increased significantly
in the last 30 years. Increases in production are due almost exclusively to an increase in the area
under cultivation. Increasing pest and disease pressure, especially from leaf spot diseases, and
the deteriorating natural resource base have collectively been responsible for these low yields.
Resistant high yielding bananas have been bred and supplied to smallholders in the 1990s after
nearly 70 years of conventional breeding. This very slow progress was due to the high sterility,
poor seed germination rate, need for interploidy crosses, the long generation cycle, which are
inherent to bananas and plantains. A breeding program can only supply a few promising
hybrids per year for further evaluation. The few selected hybrids are high yielding and resistant
to some diseases but have usually lost other desired characteristics such as shelf life or pulp
texture. Genetic transformation tools offer an opportunity for plant breeders to overcome the
constraints imposed by the high level of sterility of the most popular cultivars. Good progress
has been made in the development of a molecular toolbox for bananas and plantains in the areas
of 1) cell suspension, 2) genetic transformation (particle bombardment and Agrobacteriummediated transformation), 3) high expression of foreign genes, 4) insertion of multiple genes and
5) identification of genes for resistance to fungal disease.
Resumen - Enfoques transgénicos para la resistencia a las enfermedades de las
manchas foliares en banano (Musa spp.)
Durante los últimos 30 años, los rendimientos promedio de los bananos y plátanos no han
aumentado significativamente en las pequeñas fincas y los aumentos de producción se deben
casi exclusivamente a un aumento del área bajo cultivo. El aumento de la presión de plagas y
enfermedades, especialmente de las enfermedades de las manchas foliares, y el deterioro de la
base de recursos naturales han sido responsables de manera colectiva de estos rendimientos tan
bajos. En la década de los 90, se seleccionaron bananos resistentes de alto rendimiento los cuales
1
Laboratory of Tropical Crop Improvement, KULeuven, Leuven, Belgium
Centre for Microbial and Plant Genetics, KULeuven, Leuven, Belgium
2
209
MyLsd 17x24
25/06/03
15:15
Page 210
Mycosphaerella leaf spot diseases of bananas: present status and outlook
fueron suministrados a los pequeños productores, después de casi 70 años de mejoramiento
convencional. Este progreso tan lento se debió a una alta esterilidad, una tasa pobre de
germinación de las semillas, una necesidad de cruzamientos interploídicos, un largo ciclo de
regeneración, etc., inherentes a los bananos y plátanos. Básicamente, un programa de
mejoramiento puede proporcionar solo unos pocos híbridos prometedores por año para realizar
las evaluaciones consiguientes. Los pocos híbridos seleccionados son de alto rendimiento y
resistentes a algunas enfermedades, pero usualmente pierden otras características deseadas como
vida verde, textura de la pulpa, etc. Las herramientas de la transformación genética ofrecen una
oportunidad a los fitomejoradores para vencer las limitaciones impuestas por el alto nivel de
esterilidad en las variedades más populares. También se alcanzó un buen progreso en el
desarrollo de una serie de herramientas moleculares para los bananos y plátanos en las áreas
de (1) desarrollo de las suspensiones celulares; (2) tecnologías de transformación genética
(bombardeo con partículas o transformación con Agrobacterium); (3) alta expresión de genes
foráneos; (4) inserción de genes múltiples; (5) identificación de genes para la resistencia a
enfermedades fungosas.
Résumé - Approches transgéniques de la résistance aux maladies foliaires causées
par les Mycosphaerella chez les Musa spp.
Dans les exploitations de petite taille, les rendements en bananes et bananes plantain n’ont pas
significativement augmenté au cours des 30 dernières années. L’augmentation de la production est
due presque exclusivement à une augmentation de la surface cultivée. L’accroissement de la pression
des maladies et ravageurs, et particulièrement des maladies foliaires, et la détérioration de la base
de la ressource naturelle ont été collectivement responsables de ces faibles rendements. Des
bananiers résistants et à rendement élevé ont été produits et distribués aux petits producteurs dans
les années 90, après près de 70 ans d’amélioration conventionnelle. Ces progrès très lents sont dus
à la stérilité élevée,au faible taux de germination des semences,au besoin de réaliser des croisements
interploïdes et au long cycle de génération, qui sont propres aux bananiers et aux bananiers
plantain. Un programme d’amélioration ne peut produire que quelques hybrides prometteurs par
an pour leur évaluation ultérieure. Les quelques hybrides sélectionnés ont une production élevée et
sont résistants à certaines maladies mais ont généralement perdu d’autres caractéristiques désirées,
telles que la durée de conservation ou la texture de la pulpe. Les outils de transformation génétique
offrent une opportunité aux sélectionneurs de surmonter les contraintes imposées par le niveau élevé
de stérilité des cultivars les plus populaires.Des progrès importants ont été faits dans le développement
d’une boîte à outils moléculaires pour les bananiers et les bananiers plantain dans les domaines :
1) des suspensions cellulaires ; 2) de la transformation génétique (bombardement de particules et
transformation avec Agrobacterium ; 3) du niveau d’expression élevé de gènes étrangers ;
4) de l’insertion de gènes multiples et 5) de l’identification de gènes de résistance aux maladies
fongiques.
Introduction
The predicted increase of the world’s population to 8 billion people by 2025 (Harris,
1996) will require developing nations to dramatically increase crop yields.
Technologies such as the application of fertilizers or pesticides will have to
contribute, but the most environmentally safe and sustainable approach is the
production and delivery of stress resistant high yielding cultivars. Until recently,
new cultivars were produced by cross-breeding or the selection of induced or
natural mutations. With the rapid advances in molecular biology, the genetic
modification of tropical crops needs to be envisaged to accelerate and focus genetic
improvement.
210
MyLsd 17x24
25/06/03
15:15
Page 211
Session 4
R. Swennen et al.
Bananas are one of the first domesticated crops (De Langhe and De Maret, 1999).
Some 3000 years ago, between 3 to 8 plantain cultivars were introduced in Africa
(De Langhe et al., 1995) (Mbida et al., 2001). Somatic mutations gave rise to about
120 plantain cultivars (Swennen, 1990) which are all susceptible to black leaf streak
disease. Clearly, plantains have a very narrow genetic basis but this is also true for
the entire Musa genus despite the existence of about 1200 accessions (Van den houwe
et al., 2000).
Average yields of bananas and plantains, hereafter called bananas, have not
increased significantly in the last 30 years and increases in production are due almost
exclusively to an increase in the area under cultivation. Average yields on
smallholdings remain below 8 t/ha/yr but yields up to 80 t/ha/yr are possible. The
gradual decrease in yields in the major banana growing regions has been attributed
to increased pest and disease pressure and a deteriorating natural resource base. As
a result, many rural communities in Africa are now unable to meet their basic needs
for food and income.
Resistant high yielding bananas have been bred and supplied to smallholders in
the 1990s after nearly 70 years of conventional breeding (Vuylsteke et al., 1993a,
1993b, 1993c, 1994, 1995; Rowe, 1984; Rowe and Rosales, 1990). This extremely
slow progress is due to high sterility, poor seed germination rate, the need for
interploidy crosses (Swennen and Vuylsteke, 1993; Vuylsteke and Swennen, 1993;
Ortiz and Vuylsteke, 1995) and the long generation cycle. Basically, a breeding
program supplies only a few promising hybrids per year for further evaluation. Only
0.1% of the selected hybrids are high yielding and resistant to some diseases but
they have lost other desired characteristics such as shelf life or pulp texture. Genetic
transformation tools offer an opportunity for plant breeders to overcome the
constraints imposed by the high level of sterility of the most popular cultivars
(Swennen, 1994; Sági et al., 1995a, 1995c, 1998a, 1998b).
In this article, we discuss the different molecular tools available for banana
improvement, i.e. 1) embryogenic cell suspensions, 2) gene transfer technologies,
3) expression of foreign genes, 4) insertion of multiple genes and 5) gene identification. The current technology has the potential of producing several hundreds
transgenic plants per day, in contrast to conventional breeding methods. Possible
future scenarios for the production of bananas resistant to leaf spot diseases, such
as strategies relying on meristem transformation, R-genes from banana and pathogen
inducible promoters, are presented.
Transgenic bananas have been available at KULeuven since 1994. Yet, until now
they could not be tested in the field because of a lack of national laws in African
plantain-growing countries to regulate their release. This causes unnecessary delays
in the further testing, fine-tuning and delivering of resistant bananas and plantains
to smallholders.
Cell and tissue culture
In many plant species, genetic transformation is very simple and genes are
transferred to callus obtained, for example, from wounded leaves as is the case with
apples (De Bondt 1995). Even simpler is the flower dip method in Arabidopsis, which
211
MyLsd 17x24
25/06/03
15:15
Page 212
Mycosphaerella leaf spot diseases of bananas: present status and outlook
does not necessitate an in vitro process for transformation (Bent, 2000; Clough and
Bent, 1998). In bananas, embryogenic cell suspensions are still needed and the
procedure is far from routine (Schoofs et al., 1999). Unlike most dicots (De Vries et
al., 1988; Meijer et al., 1999) and seedbearing monocots (Vasil and Vasil, 1986; Vasil,
1987), bananas are highly recalcitrant to embryogenesis. Four main procedures have
been developed, each relying on different explants: zygotic embryos (Cronauer and
Krikorian, 1988; Escalant and Teisson, 1989), rhizome slices and leaf sheaths (Novak
et al., 1989), immature (fe)male flowers (Escalant et al., 1994; Grapin et al., 1996;
Grapin et al., 1998) and proliferating meristem cultures (Dhed’a et al., 1991; Schoofs,
1997).
Most embryogenic suspensions are produced from meristems or flowers, each
method having advantages and disadvantages. For example, the former depends on
extensive preparation of material before induction of embryogenesis, whereas the
latter requires direct access to flowering banana plants.
At KULeuven the ‘scalp’ method (Schoofs, 1997) relies on rapidly proliferating
cultures initiated from a shoot-tip meristem cultured on a medium containing high
levels of cytokinin. Embryogenesis-competent scalps contain a high number of tiny
white meristems with only a small amount of corm or leaf tissue. The shoot-tip is
first screened for endophytes and if found positive either cleaned-up or replaced by
an endophyte-free shoot-tip (Van den houwe et al., 1998; Van den houwe and
Swennen, 2000). The scalp method involves: 1) preparation of embryogenesis
competent explants (scalps), which takes 5 to 14 months; 2) embryogenesis
induction, which takes 4 to 7 months; and 3) suspension initiation and upgrading,
which takes 3 to 6 months (Swennen et al., 1998). Hence 12 to 27 months, depending
on the cultivar (Schoofs, 1997), are needed before a suspension is ready for
transformation.
The production of suspensions from East African Highland bananas is particularly
cumbersome (Strosse et al. in press, Table 1). A broad range of cytokinins at varying
concentrations was explored for scalp induction and it was found that TDZ (thidiazuron)
was a good alternative to BAP (benzylaminopurine) (Table 2 and Figure 1). In fact,
10 mM TDZ could reduce by threefold the embryogenesis induction time (Strosse
et al., in press). The embryogenic response was found to depend on the genotype
(Figures 2 and 3) and even on the selected line and the experiment, and varied between
0 and 22.2% (Strosse et al., in press). Homogeneous complexes consisting of a high
proportion of embryogenic callus and early-stage transparent embryos are preferred
as inoculum but embryogenic cell suspensions remain more or less heterogeneous
(Georget et al., 2000).
Once cell suspensions are produced, they undergo quality control measurements
at repeated intervals on regeneration potential, health status (Van den houwe et al.,
1998), DNA content (Roux et al., in press a), true-to-typeness, etc. Between 104 to
105 somatic embryos per ml settled cell volume can be obtained. Hence, a ‘Grande
naine’ cell suspension can produce 14 580 to 100 980 plants while 27 000 to 117 000
plants can be regenerated from an ‘Orishele’ suspension (Strosse et al., in press)
(assuming an inoculum of 1.5% settled cell volume in a 60-ml cell suspension
maintenance medium and a twofold increase of cell volume after a two-week
subculture). DNA content is assessed through flow cytometry and can show the loss
212
0160
0084
0086
0111
0109
0517
1259
0570
1256
213
16056
1872
2040
1296
1872
456
2808
1104
1128
864
576
336
1200
504
Number of
inoculated
scalps
273
3
3
29
240
99
112
11
16
Responsive
scalps*
1.1
0.5
0.9
2.4
3.5
5.5
0.8
0.9
Frequency
(%)**
3.8
0.5
2
5.8
22.2
11.7
2.9
4.2
Highest
frequency
in a single
experiment
yes
yes
yes
yes
yes
yes
Yes
EC
yes
yes
yes
yes
yes
yes
yes
ECS
yesf (1)b
yes (1) b
yes (4) b
yes (7)b
yes (4)b
Ready for
first applications
Session 4
*
**
EC
ECS
a
b
wild diploid
Cavendish
Cavendish
Cavendish
Cavendish
Cavendish
highland
highland
highland
plantain
plantain
plantain
cooking
ITC
code
15:15
Number of scalps forming embryogenic complexes
% of scalps forming embryogenic complexes (RS/NS3)x100, with NS3 being the total number of scalps longer than 3.5 months in culture (long enough for first embryogenic complexes to form)
Embryogenic complex
Embryogenic cell suspension
Derived via zygotic embryo rescue. Seeds obtained from IITA, Nigeria
Number of independently established cell suspension lines ready for first applications at the end of 2001
AA
AAA
AAA
AAA
AAA
AAA
AAA-h
AAA-h
AAA-h
AAB-p
AAB-p
AAB-p
ABB
Calcutta 4a
Grande naine
GN FHIA
GN JD
Williams BSJ
Williams JD
Ingarama
Mbwazirume
Nyamwihogora
Agbagba
Obino l’ewai
Orishele
Burro cemsa
Type
25/06/03
Total
Mean
Genome
Cultivar
Table 1. Preliminary results using the scalp method (Jan 1998–Dec 2001).
MyLsd 17x24
Page 213
R. Swennen et al.
wild diploid
Cavendish
highland
plantain
cooking
cooking
cooking
0365
0083
0111
0010
0643
0643
ITC code
96
960
264
144
144
144
120
1872
Number of
inoculated scalps
4.2
2.8
2.8
1
4
4
22
1080
1.7
1.8
Frequency (%)***
13
Responsive
scalps**
24
720
24
24
144
144
NS3*
2.3
4.2
5.6
5.6
2.8
Highest frequency
in a single experiment
yes
yes
yes
yes
EC
yes
ECS
15:15
Total number of scalps longer than 3.5 months in culture (long enough for first embryogenic complexes to form)
Number of scalps forming embryogenic complexes
% of scalps forming embryogenic complexes (RS/NS3)x100
Embryogenic complex
Embryogenic cell suspension
Derived via zygotic embryo rescue. Seeds obtained from IITA, Nigeria
Scalps derived from 1 µM TDZ meristem cultures instead of 10µM TDZ meristem cultures as in all other cases
AA
AAA
AAA-h
AAB-p
ABB
ABB
ABB
Calcutta4a
Williams
Igisahira gizanswe
Agbagba
Bluggoe
Cachacob
Cachaco
Total
Mean
Type
25/06/03
*
**
***
EC
ECS
a
b
Genome
Cultivar
Table 2. Preliminary results using TDZ scalps (Jan 1998–Dec 2001).
MyLsd 17x24
Page 214
Mycosphaerella leaf spot diseases of bananas: present status and outlook
214
MyLsd 17x24
5/08/03
7:57
Page 215
Session 4
R. Swennen et al.
of even 1 chromosome (Dolozel et al., 1999). Rate of somaclonal variation can vary
from very low (2%) (Côte et al., 2000) to very high (99%) (data not published). Since
high quality suspensions are rare, they are cryopreserved for backup purposes (Panis
et al., in press; Panis et al., 1990).
Figure 1.
Highly proliferating meristem cultures
of ‘Williams’ (AAA) four months after
inoculation of a 5-mm explant (apical
dome fully covered by three to four leaf
primordia and a few millimeters of corm
beneath the apical dome) on MS based
medium supplemented with 10 µM TDZ,
bar = 769 µm.
Figure 2.
Highly regenerable embryogenic
cell suspensions of ‘Grande naine’
(AAA). Embryogenic cell clusters
observed with (left) light microscope, bar = 95 µm and (right)
germinating embryos one month
after culturing on regeneration
medium, bar = 370 µm.
Figure 3.
Highly regenerable embryogenic
cell suspensions of ‘Orishele’ (AAB).
Embryogenic cell clusters observed
with (left) light microscope, bar =
95 µm and (right) germinating
embryos one month after culturing
on regeneration medium, bar =
370 µm.
Gene transfer methods
Regenerable embryogenic cell suspension (ECS) cultures are the material of choice
(Dhed’a et al., 1991; Escalant et al., 1994; Côte et al., 1996, 1997) for the genetic
engineering of bananas via particle bombardment-mediated transformation (PMT)
and Agrobacterium-mediated transformation (AMT) (Sági et al., 2000). The former
uses the biolistic gun device (Sági et al. 1995a), whereas the latter uses cocultivation
215
MyLsd 17x24
25/06/03
15:15
Page 216
Mycosphaerella leaf spot diseases of bananas: present status and outlook
with Agrobacterium (Pérez Hernández, 2000). Different methods and cultivars are
being used (Sági et al., 1995a; May et al., 1995; Remy et al., 1998a, 1998b; Becker
et al., 2000; Ganapathi et al., 2001). Our comparative study involved ECS of four
cultivars in the exponential growth phase (4+1 days after subculture). DNA plasmid,
PMT and plant regeneration were according to Sági et al. (1995a, 1995b). For AMT
transformation, evaluation of transient gene expression and selection and
regeneration of transformants was according to Pérez Hernández (2000), Pérez
Hernández et al. (1999) and Arinaitwe et al. (in press).
Results indicate that the efficiency of the two gene transfer methods is quite similar
at the transient level with different promotors. In contrast, AMT was more efficient
than PMT after selection (Arinaitwe et al., in press) (Table 3). The higher number of
plants regenerated using the AMT system in comparison with the PMT system
confirmed AMT as the method of choice for transforming plant cells, as reported by
Newell (2000).
Table 3. Shoot regeneration of four cultivars using the AMT and PMT methods.
Cultivar
AMT
PMT
Grande naine
Obino l’ewai
Orishele
Three hand planty
117
118
93
96
76
41
05
98
ECSs of four cultivars were co-cultivated with an Agrobacterium tumefaciens
strain: AGLO harbouring the binary plasmid pUbi-sgfpS65T; and EHA101 harbouring
the binary plasmid pFAJ3000. Plasmid pFAJ3000 contains a gusA (ß-glucuronidase)
gene driven by the CaMV 35S promoter and a neo gene under the control of the
NOS promoter. Plasmid pUbi-sgfpS65T contains a gfp (green fluorescent protein)
gene driven by the ubiquitin promoter (Arinaitwe et al., in press). There was a
difference in the expression of the two reporter genes used (Figure 4). This was,
probably, due to differences in efficiency of the two A. tumefaciens strains used
and the variable embryogenesis.
The effect of infection time on transformation frequency was investigated in
‘Grande naine’ and ‘‘Three hand planty’ (AAB) by using transient gus expression
(TGE) and transient green fluorescent protein expression (TGFPE). In both cultivars
and both marker genes, transient expression increased with increasing infection time
(Table 4). With ‘Grand naine’, maximum transient expression was reached after
8 and 12 hours for TGE and TGFPE, respectively. More TGE was observed in Three
hand planty’ than ‘Grande naine’, possibly due to differences in the quality of the
cell line.
Variable volumes (ml) of ECS were plated and uniformly spread over a 50-mm
nylon mesh. Transient GFP expression indicates that T-DNA transfer was highest
at 100±50 ml (Arinaitwe et al., in press) but dropped sharply when the volume of
ECS was increased to 300 and 600 ml. A decreased attachment and access to
individual embryogenic cells or cell clusters by Agrobacterium is considered to be
the cause. High TGFPE in small volumes is attributed to increased exposure of ECSs
216
25/06/03
15:15
Page 217
Session 4
R. Swennen et al.
60
GUS
% Regeneration
50
GFP
40
30
20
10
0
Grande naine
Three hand planty
Obino l'ewai
Orishele
Figure 4. Percentage of regenerated shoots of four cultivars transformed via Agrobacterium (AGLO, pUbisgfpS65T with gfp gene; EHA 101, pFAJ 3000 with gusA gene).
to Agrobacterium since efficient spreading of thin layers of cells is achieved during
the co-cultivation phase.
Upon subculture an ECS starts to multiply but its growth (Schoofs et al. 1999)
and cell cycle (Roux et al. in press b) changes with age. Hence an effect of ECS age
on transformation frequency is expected and could be confirmed (Figure 5). Cell
competence for transformation increased from day 1 until day 7, beyond which it
dropped. This period is thought to coincide with the exponential growth phase of
the ECS (Sági et al. 1995a, b). Efficient transformation of 7-day-old ECSs has been
reported in cultivar ‘Rasthali’ (AAB) (Ganapathi et al. 2001).
2500
Number of blue foci
MyLsd 17x24
1st trial
2nd trial
2000
1500
1000
500
0
1
3
5
7
9
Age (days)
Figure 5. Effect of age of embryogenic cell suspension on transformation frequency: transient gus expression
in cultivar ‘Obino l’ewai’ transformed via Agrobacterium (EHA 101; pFAJ 3000) (n=3).
217
MyLsd 17x24
25/06/03
15:15
Page 218
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 4. The effect of infection time on transient gene expression frequency (Mean±SE of the number of spots of
gene expression per cell sample of 50 mg (fresh weight); at least 4-5 replications).
Marker
gene
Cultivar
Infection time (hrs)
8
10
>1500
>1500
12
>1500
14
>1500
gus
Grande naine
4
794±87.9
6
881±91.8
gfp
Grande naine
209.7±24
272±41
365.7± 28
922.7±21
>1500
>1500
gfp
Three hand planty
1169.3±150
1311.7±95
>1500
>1500
>1500
>1500
The combined use of these and other factors resulted in a five-fold increase in
transient expression compared to the original procedure. Representative results of
these experiments are shown in Figure 6. One hundred mg (fresh weight) of control
banana cells showed no background transient expression of the gus reporter gene
(Figure 6A). In contrast, an average of 1500 blue foci was observed in the same
amount of a cell line of the dessert banana ‘Grande naine’ (Figure 6B) using the
improved method, in comparison with about 250 blue foci using the standard
protocol. The uniform distribution of the transiently transformed cells also indicates
the high efficiency of the improved procedure. Similarly increased transient gus
expression rates have been observed in several bananas cultivars. Experiments are
now in progress to determine if increased transient gene expression improves the
yield of transgenic plants.
Figure 6. Transient GUS expression in a cell suspension culture of ‘Grande naine’ after co-cultivation
with Agrobacterium tumefaciens EHA105 harbouring the GUS-intron containing binary vector pFAJ3000.
A) 100 mg control cells showing no GUS expression, B) the same amount of transformed cells after six days
of co-cultivation using the improved method.
Following confirmation that banana could be transformed (Sági et al., 1995a;
May et al., 1995), several groups looked for suitable promoters to improve the
expression of heterologous genes. More than 25 heterologous and 1 homologous
promoters have been tested (Table 5). In general strong transcription is obtained when
genes are driven by the constitutive promoters such as the promoter from the maize
ubiquitin, or the rice actin gene and promoters from the pregenomic RNA of banana
streak badnavirus. Few tissue-specific promotors have been identified in banana but
some promoter regions from the banana bunchy top nanavirus (BBTV) seem to have
a potential for expression in vascular tissue.
218
MyLsd 17x24
25/06/03
15:15
Page 219
Session 4
R. Swennen et al.
Table 5. Promoters used in banana genetic transformation.
Promoter
Source of the promoter
CaMV35S
Cauliflower Mosaic Virus
CaMV35S a
35S-AMV
Cauliflower Mosaic Virus
Cauliflower Mosaic Virus
Alfalfa Mosaic Virus
Cauliflower Mosaic Virus
35S-35S b
35S-35S-AMV
Reference
Sági et al., 1992; Sági et al., 1994; Sági et al., 1995b;
Sági et al., 1995c; CIRAD, 2001; Dugdale et al., 1998;
Pérez Hernández et al., 1998; Sági, 1998; Más et al.,
1999; Schenk et al., 1999; Becker et al., 2000;
Dugdale et al., 2001; Schenk et al., 2001
Más et al., 2000
Sági et al., 1995c
Emu c
Cauliflower Mosaic Virus
Alfalfa Mosaic Virus
Recombinant ARE-ocs-adh1
Act-1
Act-1
Ubi
Rice actin gene act1D
Banana actin gene
Maize ubiquitin
(ocs)3mas d
Recombinant ocs-mas
Sc
Sugarcane bacilliform badnavirus
(ScBV)
Banana streak badnavirus
My from Mysore cv. BSV
Cv from Cavendich cv. BSV
Go from Goldfinger cv. BSV
Banana Bunchy Top Virus
Badnavirus
My
Cv
Go
BBTV DNA 1 to 6
S1
S2
BT1, BT2, BT3, BT4, BT5 e
BT6.1 f
a
b
c
d
e
f
Moy et al., 1998; Moy et al., 1999; Schenk et al., 2001;
Remy et al., 1998b; Sági et al., 1995a;
Sági et al., 1995b; Sági et al., 1995c
Sági et al., 1994; Sági et al., 1995a; Sági et al., 1995b;
Sági et al., 1995c; Sági et al., 1998a
Sági et al., 1995a; Sági et al., 1995b;
Sági et al., 1998a; Remy, 2000
May et al. 1995; Sági et al., 1998a
Hermann et al., 2001b
Grapin, 1995; Sági et al., 1995a; Grapin et al., 1996;
CIRAD, 2001; Sági et al., 1998a; Dugdale et al., 1998;
Moy et al., 1998; Remy et al., 1998b; Moy et al., 1999;
Schenk et al., 1999; Becker et al., 2000;
Pérez Hernández, 2000; Ganapathi et al., 2001;
Schenk et al., 2001
Remy et al., 1998b; Moy et al., 1998; Moy et al., 1999;
Remy, 2000; Ganapathi et al., 2001
Schenk et al., 1999
Sági, 1998a; Schenk et al., 2001; Remans et al. 2000
Dugdale et al., 1998; Becker et al., 2000;
Hermann et al., 2001a
Dugdale et al., 2000
Dugdale et al., 2001
OCS enhancer plus Cauliflower Mosaic Virus promoter plus rice act1 untranslated sequence
Enhanced Cauliflower Mosaic Virus promoter
Six copies of the 41-bp ARE (anaerobic responsive element) plus four copies of the 40-bp ocs (octopine synthase) enhancer plus the 5’
end of a truncated adh1 promoter linked to its first intron
A tandem of three upstream activating sequences (UAS) of the octopine synthase gene (ocs) and a promoter/activator region of the mannopine
synthase gene (mas)
Banana Bunchy Top Virus DNA 1-5 intergenic regions plus maize ubiquitin (ubi1) intron
Banana Bunchy Top Virus DNA-6 intergenic region plus intron mediated enhancement of maize ubi1, maize adh1, rice act1 and sugarcane
rbcs genes
Transformation for resistance to leaf spot diseases
Plants have developed a range of defense mechanisms against pathogens such
as, the rapid death of the first infected cells (Colligne and Slusarenko, 1987) which
prevents further pathogen spread. Other defense mechanisms include increased
219
MyLsd 17x24
25/06/03
15:15
Page 220
Mycosphaerella leaf spot diseases of bananas: present status and outlook
lignification of the cell wall (Vance et al., 1980), the synthesis of phytoalexines
(Hahlbrock and Scheel, 1989) and the production of reactive oxygen species (Mehdy,
1994). The biochemical complexity of these mechanisms makes it difficult to develop
molecular methods to breed for fungus resistance.
Of the large number of genes activated upon pathogen recognition by the plant
is a group that encodes for pathogenesis-related (PR) proteins (Van Loon et al., 1987;
Linthorst, 1991). PR proteins have been defined as proteins encoded by the host plant
but induced only under pathological or related stress conditions (Antoniw et al.,
1980). To date, more than 10 families of PR proteins have been classified, which
include chitinases. In addition, plants express numerous other genes that encode for
proteins with antimicrobial properties (Broekaert et al., 1997; García-Olmedo et al.,
1998). These are small, stable and cysteine-rich peptides isolated from seeds of diverse
plant species (Broekaert et al., 1992; Cammue et al., 1992; Terras et al., 1992a, b;
Cammue et al., 1995; Osborn et al., 1995). These plant defensins, or antifungal
peptides, are highly active against a broad spectrum of phytopathogenic fungi. For
example, two of these proteins (Rs-AFP1 and Rs-AFP2) have been isolated from radish
(Raphanus sativus) seeds. The latter appears to be the most potent with IC50 values
ranging from 0.4 to 25 µg/mL (Terras et al., 1992a, b). Several of these antifungal
peptide genes were shown to inhibit the growth of Mycosphaerella spp. under in
vitro conditions (Cammue et al., 1993) but were not toxic to human fibroblasts and
erythrocytes (Terras et al., 1992a, b; Cammue et al., 1995) or to banana cells (Cammue
et al., 1993). We therefore focused our work on the insertion of these antifungal
or antimicrobial proteins in banana with a different mode of action. Morphogenic
defensins, like Rs-AFP2 from Raphanus sativus, cause a reduction in hyphal
elongation and an increase in hyphal branching, whereas non-morphogenic ones,
such as Dm-AMP1 from Dahlia merckii (Osborn et al. 1995), slow down hyphal
elongation without a visible morphological effect.
Since it was demonstrated in tobacco that disease resistance can be increased by
simultaneously integrating different antifungal proteins (Jach 1995), the frequency
of co-transformations with particle bombardment was evaluated in three independent
experiments (Remy et al. 1998a). In experiment 1, the selectable marker gene (gene
A) and an antifungal peptide gene (gene B) were introduced into embryogenic cells
of the plantain cultivar ‘Three hand planty’ in a linked position, i.e. the two genes
were present on the same plasmid. In experiments 2 and 3, the plasmid with genes
A and B were co-transformed with another plasmid that carried a different
antifungal protein gene (gene C) which was thus not linked to genes A or B.
Transgenic shoots were then regenerated from all three experiments and analysed
by PCR for the presence of each foreign gene. Integration of these genes was also
confirmed by Southern gel blot hybridisation in a number of selected plants from
each experiment.
The number of plants carrying both gene A and gene B or C was used to calculate
the co-transformation frequencies of linked genes and according to the following
equation:
{No. of (A+B or C)+ / No. of A+} x 100
As one can expect, the linked genes co-existed in the transgenic plants at a high
frequency that ranged between 90 to 100% in the different experiments. Similarly,
220
MyLsd 17x24
25/06/03
15:15
Page 221
Session 4
R. Swennen et al.
as expected, the unlinked gene showed a lower co-transformation frequency than
the linked genes. However, this frequency was still remarkably high, in the range
of 70 to 80%, probably due to efficient co-precipitation of the two plasmids on
microparticles. Similarly to the results obtained by Hadi et al. (1996), this observation
indicates that simultaneous bombardment of different plasmid molecules may be a
convenient way for the introduction of multiple genes into crop plants.
Co-transformation frequency with Agrobacterium was also estimated with one
of the several possible methods. Two Agrobacterium strains were used to introduce
two reporter genes (gfp and luc) into bananas. The elegance of this experimental
setup is the simultaneous and live observation of gene expression by imaging.
This way, co-transformation frequencies were directly measured by co-expression
frequency. As was expected, co-transformation frequency was relatively low because
of the low probability that two bacterial cells will deliver their T-DNA molecules to
the same plant cell. In our case, the average frequency of gfp and luc cotransformation in four cultivars was around 3% after three weeks and 4% two months
after transformation (Ahmed, unpublished data).
The expression of antimicrobial peptides in transgenic bananas was analysed
by ELISA using specific antibodies. Out of more than 150 single transformants,
i.e. transgenic plantains expressing only one antimicrobial peptide, more than 10%
had a relatively high concentration (between 0.05-0.12% of total soluble protein)
in the leaf. In contrast, out of 16 double transformants, i.e. plants expressing a
Dm-AMP1 and another antimicrobial peptide from onions (Ac-AMP1), 6 (37%)
accumulated one or both peptides to at least four times the background level
(Remy, 2000).
In order to assess tolerance to fungus in transgenic lines, a simple, sensitive and
reproducible leaf disc bioassay has been developed (Remy, 2000). A 5-cm leaf disc
was excised from transgenic plants grown in a greenhouse and inoculated in situ
with fungi. Four days after infection, a differential disease response was observed
between independent transformants, whereas a non-pathogenic fungus (e.g. Fusarium
sp.) was unable to induce disease symptoms, indicating that the assay is specific for
host-pathogen interactions. Transformations with different promoter-gene constructs
resulted in a wide range of tolerance to fungus among the independent transgenic
plants. Computer image capturing and software-based area calculation has been used
to precisely measure the area of infected leaf and to classify independent
transformants according to their tolerance. This procedure was used to screen 42
independent transgenic plantain lines expressing Ac-AMP1 (Pérez Hernández,
2000), among which 6 lines had 2 to 3 times less necrosis upon infection with
Colletotrichum musae than the untransformed controls.
Towards an improved transformation technology
Efficient transformation techniques exist in banana but they rely on labourintensive cell suspension technology. Moreover, controlled transgene expression needs
to be developed for banana relying, among others, on native or heterologous
developmental and tissue specific promoters and especially pathogen-inducible ones.
This is needed because constitutive resistance leads to a decrease in fitness (Heil and
221
MyLsd 17x24
25/06/03
15:15
Page 222
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Baldwin, 2002). However, relatively few promoters are available for the regulated
control of target gene expression.
Meristem transformation
Existing transformation technologies rely on cell suspensions, an elaborate, expensive
and time-consuming process. A transformation procedure based instead on tissue
culture of cell cultures would be more efficient and less dependent on the genotype.
To this end, more than 12 000 explants of ‘Grande naine’ and ‘Williams’ were infected
with Agrobacterium with or without sonication. Selection was performed either on
a liquid or solid medium using different growth regulators. Independently of
treatment, the frequency of putative transgenic cultures was about 0.5% in both
cultivars after 2 to 3 months. This consistently low transformation frequency indicates
that proliferating meristematic cultures may not be the ideal target for genetic
transformation. This could be attributed to the heterogenous structure of the
meristem containing few embryogenic cells. Meristem transformation may produce
chimeras that are not easily identified or dissociated (Roux et al., 2001). In addition,
based on DNA behaviour in mouse embryonic stem cells (Lei et al., 1996),
embryogenic plant tissues in their early development may be more suitable for gene
targeting than tissues that underwent several divisions and differentiation (Kumar
and Fladung, 2001). However, given the little effort directed to the optimization of
meristem transformation in banana, this avenue should be further explored.
Positive selectable marker genes
Once foreign genes are delivered to banana suspensions, transformed cells need to
be harvested. With PMT, about 1 to 7 cells per 100 µl of a 33% settled cell volume
suspension (± 25 mg fresh weight cells) are transformed, and up to 100 transgenic
cells from twice the same volume with AMT. Because of this very low transformation
frequency, selectable marker genes, e.g. the neomycin phosphotransferase gene (Fraley
et al., 1983) which confers resistance to aminoglycoside-type antibiotics such as
kanamycin, neomycin and G-418 (geneticin), are used. Occasionally herbicide
resistance genes are also used. These are negative selection systems. In banana,
antibiotic resistance genes should pose no concern to the environment because there
is no pollen in edible bananas, yet there is concern that such genes in genetically
modified food organisms pose a hazard to human health (Fuchs et al., 1992). Research
is conducted to completely remove selectable marker genes (Puchta 2000).
In positive selection systems, transformed cells can convert a physiologically
inert substance into a compound that stimulates growth. Hence, transgenic cells
overgrow non-transformed cells that are starved rather than killed, e.g. the gusA
(ß-glucuronidase) gene from Escherichia coli that hydrolyzes benzyladenineglucuronide (Okkels et al. 1997) into active cytokinin and thus stimulates growth
of transgenic cells. Other examples include the use of phosphomannose isomerase
(PMI, or mannose-6-phosphate isomerase), an enzyme catalyzing the reversible
isomerization of mannose-6-phosphate to fructose-6-phosphate, which serves as a
precursor for the glycolytic pathway, and xylose isomerase (or D-xylose ketol222
MyLsd 17x24
25/06/03
15:15
Page 223
Session 4
R. Swennen et al.
isomerase) that interconverts D-xylose to D-xylulose, which after phosphorylation
by xylulokinase enters the pentose phosphate pathway. The former enzyme is used
to give transgenic cells a growth advantage over non-transgenic ones because most
plant species cannot metabolize mannose (or mannose-6-phosphate) and in plant
tissue cultures mannose has been known to be unable to support growth (Malca et
al., 1967). The manA gene of E. coli (Miles and Guest, 1984), which codes for PMI,
has been successfully used in transformation experiments with cassava, maize, rice,
sugar beet and wheat (see Suprasanna et al., 2002 for references). No adverse
nutritional effects were found (Reed et al., 2001). Xylose, like mannose, cannot be
metabolized by plant cells, whereas they can utilize D-xylulose because they express
xylulokinase. So far, transformation with the xylose isomerase gene (xylA) (Wong
et al., 1991; Lee et al., 1990) as a selectable marker was demonstrated in potato,
tobacco and tomato (Haldrup et al., 1998a, b). This enzyme is widely recognized as
safe, since it is commercially used in the starch industry and for food processing.
The use of positive selectable markers genes has the additional advantage that
it can increase the transformation frequency dramatically because no toxic substances
are released from dying cells. Positive selection systems have proven their value in
several crops (sugar beet, cassava, maize, wheat and rice). For an overview the reader
is referred to Suprasanna et al. (2002). These novel selectable marker systems are
being tested in banana in order to increase the transformation frequency but also
to allow repeated transformation operations, and minimize the use of antibiotic and
herbicide selectable marker genes.
Tagging, isolation and characterization of novel promoters
and genes
The isolation of promoters of differentially expressed or inducible genes can be
accomplished indirectly or directly. In the first approach, the promoter is isolated
in parallel or subsequently to the characterization of a gene of interest by molecular
techniques. However, when dealing with multigene families and pseudogenes or
with genes that are developmentally regulated or exert a cell-specific pattern of
expression this approach is likely to be extremely difficult.
Promoters can also be identified directly within the genome via tagging by
transformation with a promoterless reporter gene and screening for individual
transformants, in which reporter gene expression is activated. After plasmid rescue
of the respective region from the genome or via direct genome walking (e.g. by
inverse PCR or various anchored PCR techniques), the promoter region is isolated
and sequenced. Via a combined screening of a population of transgenic plants for
different parameters (e.g. abiotic and biotic stress factors, development and tissuespecific expression), several promoters for various genes can simultaneously be
identified and thoroughly characterized without the a priori isolation and analysis
of the corresponding coding sequence(s).
At KULeuven, several thousands of transgenic cultures can be produced in a
relatively short time by using Agrobacterium-mediated transformation of
embryogenic cell suspension cultures. Such a large number of transgenic cultures
provides a reasonable chance of tagging interesting promoters. At present, the target
223
MyLsd 17x24
25/06/03
15:15
Page 224
Mycosphaerella leaf spot diseases of bananas: present status and outlook
cultivars are ‘Grande naine’ and ‘Three hand planty’. The tagging construct
contains a promoterless luciferase gene linked to a T-DNA border and, so far,
approximately 2500 and 4000 transgenic cultures respectively have been selected
after transformation. The expression of the luciferase gene (luc) (De Wet et al., 1985)
is not destructive and therefore provides the opportunity for continuous detection
of reporter gene activity in plants using low-light imaging techniques (Ow et al.,
1986; Millar et al., 1992; Chia et al., 1994). Screening for activated luciferase
expression has been carried out under a liquid nitrogen-cooled CCD camera coupled
to a sophisticated image capture and analysis system (Remy et al., in press). A liquid
nitrogen cooling system is used since it reduces the dark current to less than
1 electron per pixel per hour allowing long exposures of up to tens of minutes.
LUC activity could be detected 80 minutes after bombardment and was clearly
visible 40 minutes later with the codon-modified luc+gene (Sherf and Wood, 1994).
The luc+gene showed a much higher and faster LUC activity than the wild type
luc gene (Remy et al., in press).
As a wide range of LUC activities was detected (from less than 5 to more than
300 relative grey levels/pixel), it is clear that this simple, fast and sensitive in vivo
reporter gene assay can become a valuable tool in gene expression studies of
bananas. In parallel with the multiplication of the tagged population, preliminary
screenings have been performed for constitutive activation and for promoters
inducible by temperature shock, salt stress and herbicide treatment. The frequency
of cultures with detectable (constitutive) activation has been around 10% in different
tagging experiments with ‘Three hand planty’ (Figures 7A and B). On the other
hand, as expected, the frequency of inducible activation by specific conditions has
so far been well below 1% (Figure 7C and D for salt-induced activation).
Figure 7.
Luciferase imaging of transgenic
cultures of the plantain cultivar
‘Three hand planty’.
A) Light image of a 24-well plate with
cultures transformed with a 35Sluciferase construct for constitutive
expression.
B) The same cultures screened for
luciferase expression.
C) Light image of a 24-well plate
with cultures transformed with a
promoterless luciferase construct for
tagging.
D)The same cultures screened for
luciferase activation after salt stress
(arrow indicates a positive culture).
In future, selected plants will be analysed by southern hybridization, as well as
via a novel anchored PCR technique, to screen for individuals containing single
insertions. Then, TAIL-PCR will be used to recover and clone the plant DNA flanking
the luciferase gene. The cloned fragments will be sequenced, compared to sequence
databases and analysed with standard bioinformatic tools for conserved regions and
224
MyLsd 17x24
25/06/03
15:15
Page 225
Session 4
R. Swennen et al.
putative transcription factor binding sites. Based on the sequence obtained, internal
fragments of each promoter candidate will be hybridized to DNA purified from the
original transformant as well as an untransformed control to confirm their genuine
nature. Finally, for functional testing promoter-gfp fusions will be re-introduced
to banana to confirm the constitutive or wound and pathogen pattern of expression
as well the tissue specificity in an ectopic situation.
Confirmed promoters will then be utilized for disease control, for example with
antifungal genes, and associated genes isolated and characterized to understand
their role in plant-pathogen interactions. Except for promoter tagging of genes
related to nematode feeding structures (Bartels et al., 1997; Puzio et al., 1999), this
method has not been applied to studies on plant-pathogen interactions. Since
salicylic acid (SA) and its analogs play an important role in the defense response
of many plant species to pathogen attack, promotorless luc tagging in banana would
be useful to unravel upregulated genes that are involved in widely different
metabolic pathways including pathogen defense. SA treatment mimics osmotic and
oxidative stress, mediates the oxidative burst that leads to cell death in the
hypersensitive response, and acts as a signal for the development of systemic acquired
resistance (Shirasu et al., 1997).
Resistance genes are often proposed to control pathogen attack. Based on a
“guard model” it is proposed that R genes, which are generated randomly, most
likely through a birth-and-death process (Michelmore and Meyers, 1998), stand a
better chance to induce resistance if identified from the plant family of which a
certain cultivar needs to acquire additional resistance (Van der Hoorn et al., 2002).
Resistance genes from banana are currently unavailable for banana transformation
although techniques are available and some sources do exist. A series of resistance
gene analogs (RGAs) were isolated, using degenerate PCR primers targeting highly
conserved regions in proven plant resistance genes (e.g. leucine-rich repeat
sequences) (Wiame et al., 2000). For an overview of the current situation and a
proposed strategy to correct this situation, the reader is referred to Kahl (in press).
In any case, there is a need not to focus only on a few resistance genes but on the
simultaneous detection, identification and quantification of all transcripts at a given
time and monitoring of gene expression patterns at various developmental stages
or after specific treatments (Matsumara et al., 1999) correlated to physiological or
developmental processes. Much is expected from The Global Musa Genomics
Consortium.
Increased expression
The search in badnaviruses for useful promoters to drive transgene expression in
banana (Schenk et al., 1999) resulted in two novel DNA fragments of 2105 bp (My)
and 1322 bp (Cv) amplified from the upstream region of the coding sequence of
two Australian banana streak badnavirus (BSV) isolates. Evaluation of the My and
Cv promoters in transgenic banana demonstrated that these promoters could drive
high-level expression of either the gusA or the gfp reporter gene in different tissues
during vegetative development. For instance, gus activity in transgenic in vitro plants
of the plantain ‘Three hand planty’ containing the My promoter were up to seven
225
MyLsd 17x24
25/06/03
15:15
Page 226
Mycosphaerella leaf spot diseases of bananas: present status and outlook
times higher in leaf tissue and up to four times stronger in root and corm tissue
compared to plants harbouring the maize ubiquitin promoter (Schenk et al., 2001).
The Cv promoter showed activities that were similar to the maize ubiquitin
promoter in transgenic in vitro plants, but was significantly reduced in larger
glasshouse-grown plants.
Enhancement of transgene expression levels by translational fusion of transgenes
has been observed by different research groups. In an experiment, the sequence
coding for a naturally occurring plant linker peptide was used to connect the
sequence coding for two antimicrobial proteins (AMPs) in a polyprotein construct
that was transformed to Arabidopsis thaliana (François et al., 2002a, 2002b). The
linker peptides used were based on the fourth linker peptide of the IbAMP
polyprotein precursor isolated from the seed of Impatiens balsamina (Tailor et al.,
1997). The heterologous polyprotein precursors were demonstrated to be cleaved
post-translationally in A. thaliana thereby releasing the two AMPs (François et al.,
2002a, 2002b). Cleavage appeared to be complete as no immunoreactive polyprotein
precursor could be detected in the transformed A. thaliana plants. A striking
observation from the experiments was that the expression levels of the first protein
were several times higher in plants transformed with the polyprotein constructs
compared to plants transformed with the single protein construct. Expression levels
as high as 3.1% of total protein content, as seen in some lines transformed with
polyprotein constructs, have so far never been reported in literature for the nuclear
expression of a transgene in leaves of transgenic plants.
In another experiment, enhanced expression of the gene coding for the
antimicrobial peptide sarcotoxin IA was studied by fusing translationally the coding
sequence of this gene to that of E. coli b-glucuronidase (GUS) (Okamoto et al., 1998).
Western blot analysis of transgenic tobacco plants demonstrated that the amounts
of sarcotoxin IA present in the form of sarcotoxin IA-GUS fusion proteins were
considerably higher than in tobacco plants transformed with the single sarcotoxin
IA peptide construct.
It is assumed that a high transcription of genes coding for proteins that control
fungi under in vitro conditions will increase resistance in plants. One of these
strategies relies on strong promoters. However, much higher transgene expression
levels can be achieved with chloroplast genetic engineering (Daniell et al., 2002)
because chloroplasts are polyploid. Thousands of copies of foreign genes per plant
cell will generate extraordinarily high levels of foreign protein. Consequently,
chloroplast transgenic plants can show a 25-fold increase in the accumulation of
foreign gene products than nuclear transgenic plants (Lee et al., in press; Daniell
et al. 2002). In tobacco this resulted in the accumulation of 45.3% foreign protein
of total soluble protein (De Cosa et al., 2001).
In related experiments, 21.5% of total soluble protein was demonstrated to be
foreign and resulted in the protection against a fungal pathogen (DeGray et al.,
2001). Sidorov et al. (1999) and Ruf et al. (2001) achieved expression levels up to
5% and 50%, respectively. Moreover, in contrast to nuclear transformation where
the integration of a transgene is random and in unpredictable numbers, chloroplast
transformation facilitates the controlled integration in a pre-determined site,
thereby influencing the expression of the transgene (Kumar and Fladung, 2001).
226
MyLsd 17x24
25/06/03
15:15
Page 227
Session 4
R. Swennen et al.
Field testing
The genetic modification of many tropical and subtropical crops offers the prospect
of faster plant improvement (Ortiz, 1998; Sharma et al., 2000). Banana may be next.
Since 1994, putative fungal resistant transgenic plantains have been obtained at
KULeuven and analysed in partnership with scientists from Cuba, Ecuador, India and
Uganda. The non-toxicity of the expressed proteins in fruits, and in feeding tests
with rats, suggests that these plants should be field-tested in the tropics for
confirmation of resistance and biosafety evaluation. Contained fields and nurseries
have been put in place, but the LMOs (living modified organism) have not been
exported due to the absence of competent national authorities to approve the request
for import and risk assessment studies (Sági et al., 1998b).
The absence of a regulatory framework in most tropical developing countries is
delaying the evaluation of LMO bananas and plantains and the cultivation of resistant
plantains by smallholders. This is occurring despite the ratification of the Cartagena
Protocol (Cartagena Protocol, 2000) and article 19(3) of the Convention on Biological
Diversity (CBD) (Convention on Biological Diversity, 1994). The objective of the
Cartagena Protocol (adopted in January 2000) is to ensure an adequate level of
protection in the field for the safe transfer, handling and use of LMOs resulting from
modern biotechnology that may have adverse effects on the conservation and
sustainable use of biological diversity, taking also into account risks to human health,
and specifically focusing on transboundary movements.
Currently, public opinion is influenced by feelings about “Frankenstein food” and
by the “precautionary principle”. The former calls for more scientific data whereas
the latter should allow for approvals for field-testing. Indeed, the “precautionary
principle” should not be used to stop field-testing but to guide scientists under what
circumstances field-testing should be conducted. Besides, scientific data from the
field can be used to further improve current requirements. Edible bananas are
particularly suitable since they are both seed and pollen sterile. Thus, the introduced
gene(s) remain confined to the transformed plant. Banana LMO plants should have
been among the first plants to be tested in developing countries.
The development and field release of transgenic plants have been much debated.
It is clear that the deployment of transgenic plants should be safe (Custers, 2001).
Therefore, ecological risk assessment studies need to be conducted on matters dealing
with the invasiveness of the transgenic crop (can it become a weed in the natural
habitat?), on the invasiveness of the transgene itself (gene flow into wild relatives)
and the environmental side effects of the transgenic products (on non-target
organisms, for example) (Amman, 2001). To avoid any type of invasiveness, research
is conducted to introduce “reproductive isolation barriers” into crop plants, the
biosafety of transgenic crops being one of the driving forces. Examples are male
sterility (there are no viable pollen, hence no outcrossing) and the terminator
technology (seeds cannot germinate without chemical application). Complementary
strategies rely on the cultivation of sexually incompatible crops and respecting
isolation barriers, i.e. crops that can intercross are separated by a crop that cannot
intercross (Obrycki et al., 2001). The industry and environmentalists favour
reproductive isolation barriers, but the strategy would seriously handicap future
227
MyLsd 17x24
25/06/03
15:15
Page 228
Mycosphaerella leaf spot diseases of bananas: present status and outlook
breeding as it would require that an interesting gene be inserted in each cultivar
separately, which is very cumbersome and costly. However, a transgenic plant of
interest should become part of a conventional breeding programme and used for
further crossing (Dodds et al., 2001). In the case of banana, this means that diploids
should also be transformed for use in the current breeding programmes.
Conclusion
The predicted increase of the world’s population in the coming years poses many
challenges to developing countries to feed their population as more than 90% of
the population increase will occur there. But since agricultural productivity is
currently low, there are many opportunities to improve that situation. Technical
solutions, such as biotechnology, are not the only solution but form part of a package
to be used in synergy with an agro-ecological approach. The technology being in
the plant material, biotechnology ensures benefits to smallholders without changing
local cultural practices, as long as the appropriate features are considered. The 2001
United Nations Human Development Report unequivocally states that biotechnology
offers “the hope of crops with higher yields, pest- and drought-resistant properties
and superior nutritional characteristics - especially for farmers in ecological zones
left behind by the green revolution” (UNDP, 2001).
Many opponents raise ethical questions but blocking the development and
application of biotechnology can also be construed as unethical. Zero risk does not
exist. The important point is that biotechnology poses risks that are equal to the
risks encountered in conventional breeding (NRC, 2000). “A process that is safer
shouldn’t be given up because it cannot be elevated to an impossible standard of
absolute safety” (Trewavas, 2000). In the end, what counts is zero harm.
Breeding has long been used to suppress plant diseases and pathogen-resistant
cultivars quickly became popular and grown as homogeneous crops. Pathogens can
eventually overcome resistance and become epidemic, forcing breeders to introduce
a cultivar with a new resistance trait. The battle never ends as pathogens always try
to circumvent recognition by resistant plants.
To protect crops better, plant cultivars that differ in their resistance mechanisms
should be mixed, as in natural plant populations (Dangl and Jones, 2001). For
example, the deployment of different rice cultivars resulted in a 94% reduction in
the occurrence of rice blast (Zhu et al., 2000). Smallholders would be best served
by interplanting into their existing banana plot banana plants resistant to leaf spot
diseases consisting of cultivar(s) in which a single or a combination of foreign genes
have been integrated. Given that it should be possible target proteins both intraand intercellularly, a broad-spectrum resistance to banana leaf spot should be
achievable.
References
Antoniw J.F., C.E. Ritter, W.S. Piepont and L.C. Van Loon. 1980. Comparison of three
pathogenesis-related proteins from plants of two cultivars of tobacco infected with TMV.
J. Gen. Virol. 47:79-87.
228
MyLsd 17x24
25/06/03
15:15
Page 229
Session 4
R. Swennen et al.
Amman K. 2001. Safety of genetically engineered plants: an ecological risk assessment of
vertical gene flow. Custers R. (ed.). Safety of Genetically Engineered Crops. VIB, Belgium
pp. 61-87. http://www.vib.be/frame.cfm
Arinaitwe G., S. Remy, H. Strosse, R. Swennen and L. Sági. in press. Agrobacterium-and particle
bombardment-mediated transformation of a wide range of banana cultivars (S.M. Jain
and R. Swennen, eds) Banana Improvement: Cellular, Molecular Biology and Induced
Mutations, Kluwer.
Bartels N., F.M. van der Lee, J. Klap, O.J. Goddijn, M. Karimi, P. Puzio, F.M. Grindler,
S.A. Ohl, K. Lindsey, L. Robertson, W.M. Robertson, M. Van Montagu, G. Gheysen and
P.C. Sijmons. 1997. Regulatory sequences of Arabidopsis drive reporter gene expression
in nematode feeding structures. Plant Cell 9:2119-2134.
Becker D.K., B. Dugdale, M.K. Smith, R.M. Harding and J.L. Dale. 2000. Genetic transformation
of Cavendish banana (Musa spp. AAA group) cv ‘Grand Nain’ via microprojectile
bombardment. Plant Cell Rep. 19:229-234.
Bent A.F. 2000. Arabidopsis in planta transformation. Uses, mechanisms, and prospects for
transformation of other species. Plant Physiology 124:1540-1547.
Broekaert W.F., X. Mariën, F.R.G. Terras, M.F.C. De Bolle, P. Proost, J. Van Damme, L. Dillen,
M. Claeys, S.B. Rees, J. Vanderleyden and B.P.A. Cammue. 1992. Antimicrobial peptides
from Amaranthus caudatus seeds with sequence homology to the cysteine/glycine-rich
domain of chitin-binding proteins. Biochemistry 31:4308-4314.
Broekaert W.F., B.P.A. Cammue, M.F.C. De Bolle, K. Thevissen, G.W. De Samblaux and
R.W. Osborn. 1997. Antimicrobial peptides of plants. Crit. Rev. Plant Sci. 16:297-323.
Cammue B.P.A., M.F.C. De Bolle, F.R.G. Terras, P. Proost, J. Van Damme, S.B. Rees,
J. Vanderleyden and W.F. Broekaert. 1992. Isolation and characterisation of a novel class
of plant antimicrobial peptides from Mirabilis jalapa L. seeds. J. Biol. Chem. 267:22282233.
Cammue B.P.A., M.F.C. De Bolle, F.R.G. Terras and W.F. Broekaert. 1993. Fungal disease control
in Musa: application of new antifungal proteins. Pp. 221-225 in Breeding Banana and
Plantain for Resistance to Diseases and Pests. (Ganry J., ed.) Montpellier, 7-9 September
1992, CIRAD, Montpellier, France.
Cammue B.P., K. Thevissen, M. Hendrikx, K. Eggermont, I.J. Goderis, P. Proost, J. Van Damme,
R.W. Osborn, F. Guerbette, J.C. Kader and W.F. Broekaert. 1995. A potent antimicrobial
protein from onion (Allium cepa L.) seeds showing strong sequence homology to plant
lipid transfer proteins. Plant Physiol. 109:445-455.
Cartagena Protocol on Biosafety to the Convention on Biological Diversity. 2000. Secretariat
of the Convention on Biological Diversity, Text and Annexes, Montreal.
http://www.biodiv.org/biosafety/ and http://www.biodiv.org/doc/legal/cartagena-protocolen.pdf
Chia T.-F., Y.S. Chan and N.H. Chua. 1994. The firefly luciferase gene as a non-invasive reporter
for Dendrobium transformation. Plant J. 6:441-449.
CIRAD. 2001. Banana genetic transformation. Plant genomics at CIRAD. At http://www.cirad.fr/
presentation/programmes/biotrop/resultats/biositecirad/transfo/bananatg.htm
Clough S.J. and A.F. Bent. 1998. Floral dip: a simplified method for Agrobacterium-mediated
transformation of Arabidopsis thaliana. The Plant Journal 16(6):735-743.
Colligne D.B. and A.J. Slusarenko. 1987. Plant gene expression in response to pathogens.
Plant Mol. Biol 9:389-410.
Côte F.X., R. Domergue, S. Mommarson, J. Schwendiman, C. Teisson and J.V. Escalant. 1996.
Embryogenic cell suspension from the male flower of Musa AAA cv. Grand nain. Physiol.
Plant. 97:285-290.
229
MyLsd 17x24
25/06/03
15:15
Page 230
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Côte F.X., T. Legavre, A. Grapin, B. Valentin, O. Frigout, J. Babeau, D. Meynard, F. Bakry and
C. Teisson. 1997. Genetic transformation of embryogenic cell suspension in plantain (Musa
AAB) using particle bombardmen. Proceedings of the International symposium on
biotechnology of tropical and subtropical species: Part 1. (Drew R.A., A. Sasson, eds). Acta
Hort. 460.
Côte F.X., O. Goue, R. Domergue, B. Panis and C. Jenny. 2000. In-field behaviour of banana
plants (Musa AA spp.) obtained after regeneration of cryopreserved embryogenic cell
suspensions. Cryoletters 21:19-24.
Convention on Biological Diversity. 1994. Convention on Biological Diversity Text and
Annexes. Geneva: Interim Secreteriat for the Convention on Biological Diversity.
http://www.biodiv.org/convention/articles.asp?lg=0
Cronauer S.S. and A.D. Krikorian. 1988. Plant regeneration via somatic embryogenesis in the
seeded diploid banana Musa ornata Roxb. Plant Cell Reports 7:23-25.
Custers R. 2001. Safety of Genetically Engineered Crops. VIB publication: 160pp.
http://www.vib.be/frame.cfm
De Bondt A. 1995. Molecular Breeding of Apple (Malus domestica Borkh.) via Agrobacterium
tumefaciens for increased fungal resistance. PhD thesis No 297, Katholieke Universiteit
Leuven, Belgium. 103pp.
De Langhe E., R. Swennen and D. Vuylsteke. 1995. Plantain in the Early Bantu World. Sutton
J. (ed.). Azania XXIX-XXX 1994-1995. Special volume on ‘The Growth of Farming
Communities in Africa from the Equator Southwards. Proceedings of a conference of the
British Institute in Eastern Africa. Cambridge, 4-8 July 1994:147-160.
De Langhe E. and P. De Maret. 1999. Tracking the banana: its significance in early agriculture.
Pp. 377-396 in The prehistory of food. Appetites for change (C. Gosden and J. Hather,
eds). Routledge, London and New York.
Dangl J.L. and J.D.G. Jones. 2001. Plant pathogens and integrated defence responses to
infection. Nature 411(6939):826-833.
Daniell H., M.S. Khan and L. Allison. 2002. Milestones in chloroplast genetic engineering:
an environmentally friendly era in biotechnology. Trends in Plant Science 7:84-91.
De Cosa B., W. Moar, Seung-Bum Lee, M. Miller and H. Daniell. 2001. Overexpression of
the Bt Cry2Aa2 operon in chloroplasts leads to formation of insecticidal crystals. Nat.
Biotechnol. 19: 71-74.
DeGray G., K. Rajasekaran, F. Smith, J. Sanford and H. Daniell. 2001. Expression of an
antimicrobial peptide via the chloroplast genome to control phytopathogenic bacteria and
fungi. Plant Physiol. 127:852-862.
De Vries S.C., H. Booij, P. Meyerink, G. Huisman, H.D. Wilde, T.L. Thomas and A. van Kammen.
1988. Acquisition of embryogenic potential in carrot cell-suspension cultures. Planta
176:196-204.
De Wet J.R, K.V. Wood, D.R. Helinski and M. DeLuca. 1985. Cloning of firefly luciferase cDNA
and the expression of active luciferase in Escherichia coli. Proc Natl Acad Sci USA 82:78707873.
Dhed’a D., F. Dumortier, B. Panis, D. Vuylsteke and E. De Langhe. 1991. Plant regeneration
in cell suspension cultures of cooking banana cv. Bluggoe (Musa spp. ABB group). Fruits
46:125-135.
Dodds J.H., R. Ortiz, J.H. Crouch, V. Mahalasksmi and K.K. Sharma. 2001. Biotechnology,
the Gene Revolution, and Proprietary Technology in Agriculture: A Strategic Note for the
World Bank. Strategy Today 2:29pp. http://www.biodevelopments.org/ip/ipst2.pdf or
http://www.swiftt.cornell.edu
230
MyLsd 17x24
25/06/03
15:15
Page 231
Session 4
R. Swennen et al.
Dolezel J., M.A. Lysak, M. Dolezelova, M. Valarik, H. Simkova and J. Vrana. 1999. Analysis
of Musa genome using flow cytometry and molecular cytogenetics. Pp.85-93 in Report
of the 3rd research co-ordination meeting of FAO/IAEA/BADC co-ordinated research project
‘Cellular biology and biotechnology including mutation techniques for creation of new
useful banana genotypes, Colombo, Sri Lanka, 4-8 October 1999.
Dugdale B., P.R. Beetham, D.K. Becker, R.M. Harding and J.L. Dale. 1998. Promoter activity
associated wit intergenic regions of banana bunchy top virus DNA-1 to –6 in transgenic
tobacco and banana cells. Journal of General Virology 79:2301-2311.
Dugdale B., D.K. Becker, P.R. Beetham, R.M. Harding and J.L. Dale. 2000. Promoters derived
from banana bunchy top virus DNA-1 to DNA-5 direct vascular-associated expression in
transgenic banana (Musa spp.) Plant Cell Rep. 19:810-814.
Dugdale B., D.K. Becker, R.M. Harding and J.L. Dale. 2001. Intron-mediated enhancement of
the banana bunchy top virus promoter in banana (Musa spp.) embryogenic cells and plants.
Plant Cell Rep. 20:220-226.
Escalant J.V. C. Teisson and F. Côte. 1994. Amplified somatic embryogenesis from male flowers
of triploid banana and plantain culivars (Musa spp.), In vitro Plant Cellular and
Developmental Biology 30:181-186.
Escalant J.V. and C. Teisson. 1989. Somatic embryogenesis from immature zygotic embryos
of the species Musa acuminata and Musa balbisiana. Plant Cell Rep. 7:665-668.
Fraley R.T., S.G. Rogers, R.B. Horsch, P.R. Sanders, J.S. Fick, S.P. Adams, M.L. Bittners,
L.A. Brand, C.L. Fink, Y.S. Fry, G.R. Galluppi, S.B. Goldberg, N.L. Hoffmann and S.C. Woo.
1983. Expression of bacterial genes in plant cells. Proc. Natl. Acad. Sci USA 80:4803-4807.
François I.E.J.A., M.F.C. De Bolle, I.J.W.M. Goderis, P. Verhaert, P. Proost, B.P.A. Cammue and
W.F. Broekaert. 2002a. Transgenic expression in Arabidopsis thaliana of a polyprotein
construct leading to production of two different antimicrobial proteins. Plant Physiol.
128:1346-1358.
François I.E.J.A., G.I. Dwyer, M.F.C. De Bolle, I.J.W.M. Goderis, P. Proost, P. Wouters, W.F.
Broekaert and B.P.A. Cammue. 2002b. Processing in transgenic Arabidopsis thaliana plants
of polyproteins with linker peptide variants derived from the Impatiens balsamina
antimicrobial polyprotein precursor. Plant Physiol. Biochem. 40(10):871-879.
Fuchs R.L., J.E. Ream, B.G. Gammond, M.W. Naylor, R.M. Leimgruber and S.A. Berberich.
1992. Safety assessment of the neomycin phosphotransferase II (NPTII) protein.
Bio/Technology 11:1543-1547.
Ganapathi T.R., N.S. Higgs, P.J. Balint-Kurti and J. Van Eck. 2001. Agrobacterium-mediated
transformation of embryogenic cell suspensions of the banana cultivar Rasthali (AAB).
Plant Cell Rep. 20:157-162.
Garcia-Olmedo F., A. Molina, J.M. Alamillo and P. Rodriguez-Palenzuela. 1998. Plant defense
peptides. Biopolymers 47:479-491.
Georget F., R. Domergue, N. Ferriere and F.X. Côte. 2000. Morphohistological study of the
different constituents of a banana (Musa AAA, cv. Grande naine) embryogenic cell
suspension. Plant Cell Reports 19:748-754.
Grapin A. 1995. Régénération par embryogénèse somatique en milieu liquide et transformation
génétique par biolistique de bananiers di- et triploïdes. (Regeneration by somatic
embryogenesis in a liquid medium and genetic transformation of diploid and triploid
bananas by biolistics.) ENSAM, Montpellier (FRA) 93pp.
Grapin A., J. Schwendiman and C. Teisson. 1996. Somatic embryogenesis in plantain banana.
In vitro Plant Cellular and Developmental Biology 32:66-71.
Grapin A., O. Frigout, S. Monmarson, T. Legavre and F. Côte. 1996. GUS reporter gene
expression in transformed embryogenic cell suspensions of banana (Musa sp.). Meeting
231
MyLsd 17x24
25/06/03
15:15
Page 232
Mycosphaerella leaf spot diseases of bananas: present status and outlook
on tropical plants. Communications and posters. - EUCARPIA, European Association for
Research on Plant Breeding, Wageningen (NLD) 294pp.
Grapin A., J.L. Ortiz, R. Domergue, J. Babeau, S. Monmarson, J.V. Escalant, C. Teisson and
F. Côte. 1998. Establishment of embryogenic callus initiation and regeneration of
embryogenic cell suspensions from female and male immature flowers of Musa.
INFOMUSA 7(1):13-15.
Hadi M.Z., M.D. McMullen and J.J. Finer, 1996. Transformation of 12 different plasmids into
soybean via particle bombardment. Plant Cell Rep. 15:500-505.
Hahlbrock K. and D. Scheel. 1989. Physiology and molecular biology of phenylpropanoid
metabolism. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40:347-369.
Haldrup A., S.G. Petersen and F.T. Okkels. 1998a. Positive selection: a plant selection principle
based on xylose isomerase, an enzyme used in food industry. Plant Cell Rep. 18:76-81.
Haldrup A., S.G. Petersen and F.T. Okkels. 1998b. The xylose isomerase gene from
Thermoanaerobacterium thermosulfurogenes allows effective selection of transgenic plant
cells using D-xylose as the selection agent. Plant Mol. Biol. 37:287-296.
Harris J.M. 1996. World agricultural futures: regional sustainability and ecological limits.
Ecological Economics 17:95-115.
Heil M. and I.T. Baldwin. 2002. Fitness costs of induced resistance: emerging experimental
support for a slippery concept. Trends in Plant Science 7:61-67.
Hermann S.R., D.K. Becker, R.M. Harding and J.L. Dale. 2001a. Promoters derived from banana
bunchy top virus-associated components S1 and S2. Plant Cell Rep. 20:642-646.
Hermann S.R., R.M. Harding and J.L. Dale. 2001b. The banana actin 1 promoter drives nearconstitutive transgene expression in vegetative tissues of banana (Musa spp.). Plant Cell
Rep. 20:525-530.
Jach G., B. Gornhardt, J. Moundy, J. Logemann, E. Pinsdorf, R. Leah, J. Schell and C. Maas.
1995. Enhanced quantitative resistance against fungal disease by combinatorial expression
of different barley antifungal proteins in transgenic tobacco. Plant J. 8:97-109.
Kahl G. in press. The banana genome in focus: a technical perspective. In Banana
Improvement:. Cellular, Molecular Biology and Induced Mutations (Jain S.M. and
Swennen R., eds). Kluwer.
Kumar S. and M. Fladung. 2001. Controlling transgene integration in plants. Trends in Plant
Science 6:155-159.
Lee C., M. Bagdasarian, M. Meng and J.G. Zeikus. 1990. Catalytic mechanism of xylose
(glucose) isomerase from Clostridium thermosulfurogenes: Characterization of the structural
gene and function of active site histidine. J. Biol. Chem. 265:19082-19090.
Lee S.B., H.B. Kwon, S.J. Kwon, S.C. Park, M.J. Jeong, S.E. Han, H. Daniell and M.O. Byun.
In press. Drought tolerance conferred by the yeast trehalose-6 phosphate synthase gene
engineered via the chloroplast genome. Transgenic Research.
Lei H., S.P. Oh, M. Okano, R. Jüttermann, K.A. Goss and R. Jaenisch. 1996. De novo DNA
cytosine methyltransferase activities in mouse embryonic stem cells. Development
122:3195-3205.
Linthorst H.J.M. 1991. Pathogenesis-related proteins of plants. Crit. Rev. Plant Sci. 10:123150.
Lysak M.A., M. Dolezelova, J.P. Horry, R. Swennen and J. Dolozel. 1999. Flow cytometric
analysis of nuclear DNA content in Musa. Theoretical and Applied Genetics 98:1344-1350.
Matsumara H., S. Nirasawa and R. Terauchi. 1999. Technical advance: Transcript profiling
in rice (Oryza sativa L.) seedlings using serial analysis of gene expression. The Plant J.
20:719–726.
232
MyLsd 17x24
25/06/03
15:15
Page 233
Session 4
R. Swennen et al.
Malca I., R.M. Endo and M.R. Long. 1967. Mechanism of glucose counteraction of inhibition
of root elongation by galactose, mannose and glucosamine. Phytopathology 57:272-278.
Más L., S. Martinez, G. Aguero, M. Reyes, R. Gomez, G. De la Riva, A. Coego, R. Vasquez,
B. Ocaña and V. Gill. 1999. Expression of foreign genes in somatic embryos of banana
cv ‘Gran Enano’, via gene gun. P. 51 in The International Symposium on the Molecular
and Cellular Biology of Banana.
Más L., G. Agüero, V. Gil, M. Reyes, R. Gómez, B. Ocaña and S. Martínez. 2000. Optimización
de parámetros en la transformación de embriones somáticos de banano utilizando pistola
de genes. Biotecnología Vegetal 1:51-54.
May G., R. Afza, H. Mason, A. Wiecko, F. Novak and C. Arntzen. 1995. Generation of transgenic
banana (Musa acuminata) plants via Agrobacterium-mediated transformation. Bio/
Technology 13:486-492.
Mbida M. C., H. Doutrelepont, L. Vrydaghs, R. Swennen, R.J. Swennen, H. Beeckman, E. De
Langhe and P. De Maret. 2001. First archaeological evidence of banana cultivation in
central Africa during the third millenium before present. Vegetation History and
Archaeobotany 10:1-6.
Mehdy M.C. 1994. Active oxygen species in plant defense against pathogens. Plant Physiol.
105:467-472.
Meijer E.A., S.C. de Vries and A.P. Mordhorst. 1999. Co-culture with Daucus carota somatic
embryos reveals high 2,4-D uptake and release rates of Arabidopsis thaliana cultured cells.
Plant Cell Reports 18(7-8):656-663.
Michelmore R.W. and B.C. Meyers. 1998. Clusters of resistance genes in plants evolve by
divergent selection and a birth-and-death process. Genome Res. 8:1113-1130.
Miles J S. and J.R. Guest. 1984. Nucleotide sequence and transcriptional start point of the
phosphomannose isomerase gene (manA) of Escherichia coli. Gene 32:41-48.
Millar A.J., S.R. Short, K. Hiratsuka, N.-H. Chua and S.A. Kay. 1992. Firefly luciferase as a
reporter of regulated gene expression in plants. Plant Mol Biol Rep 10:324-337.
Moy Y., B. Hanson, N. Gutterson, and D. Engler. 1998. Development of an efficient banana
transformation method and analysis of the expression of some constitutive promoters in
young transgenic plants. Plant biotechnology and in vitro biology in the 21st century:
abstracts. IAPTC, International Association for Plant Tissue Culture, Jérusalem, Israel:140.
Moy Y., B. Hanson, D. Engler and N. Gutterson. 1999. Analysis of uidA gene expression from
transgenes in field-grown ‘Grand Nain’ (AAA) banana plants. The 1st International
symposium on the molecular and cellular biology of banana.
Newell C.A. 2000. Plant transformation technology: Developments and applications. Molecular
Biotechnology 16:53-65.
Novak F.J., R. Afza, M. Van Duren, M. Perea-Dallos, B.V. Conger and T. Xiaolang. 1989. Somatic
embryogenesis and plant regeneration in suspension cultures of dessert (AA and AAA)
and cooking (ABB) bananas (Musa spp.). Bio/Technology 46:154-159.
NRC (National Research Council). 2000. Genetically modified pest-protected plants: science
and regulation. Washington, DC: National Academy Press. HTTP://WWW.NAP.EDU/BOOKS/
0309069300.HTML
Obrycki J.J., J.E. Losey, O.R. Taylor and L.C.H. Jesse. 2001. Analysis of transgenic insecticidal
corn developed for lepidopteran pests reveals that the potential benefits of crop genetic
engineering for insect pest management may not outweigh the potential ecological and
economic risks. BioScience 51(5):353.
233
MyLsd 17x24
25/06/03
15:15
Page 234
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Okkels F. T., J.L. Ward and M. Joersbo. 1997. Synthesis of cytokinin glucuronides for the
selection of transgenic plant cells. Phytochemistry 46:801-804.
Okamoto M., I. Mitsuhara, M. Ohshima, S. Natori and Y. Ohashi. 1998. Enhanced expression
of an antimicrobial peptide sarcotoxin IA by GUS fusion in transgenic tobacco plants.
Plant Cell Physiol. 39:57-63.
Ortiz R. 1998. Critical role of plant biotechnology for the genetic improvement of food crops:
perspectives for the next millennium. Electronic Journal of Biotechnology 1(3).
http://www.ejb.org/content/vol1/issue3/full/7
Ortiz R. and D. Vuylsteke. 1995. Factors influencing seed set in triploid Musa spp. L. and
production of euploid hybrids. Annals of Botany 75:151-155.
Osborn R.W., G.W. De Samblanx, K. Thevissen, I. Goderis, S. Torrekens, F. Van Leuven, S.
Attenborough, S.B. Rees and W.F. Broekaert. 1995. Isolation and characterisation of plant
defensins from seeds of Asteraceae, Fabaceae, Hippocastanaceae and Saxifragaceae. FEBS
Lett. 368:257-262.
Ow D.W., K.V. Wood, M. DeLucia, J.R. De Wet, D.R. Helinski and S.H. Howell. 1986. Transient
and stable expression of the firefly luciferase gene in plant cells and transgenic plants.
Science 234:856-859.
Panis B., L.A. Withers and E. De Langhe. 1990. Cryopreservation of Musa suspension cultures
and subsequent regeneration of plants. Cryo-Letters 11:337-350.
Panis B., H. Strosse, S. Remy, L. Sági and R. Swennen. in press. Cryopreservation of banana
tissues: support for germplasm conservation and banana improvement. In Banana
Improvement: Cellular, Molecular Biology and Induced Mutations (S.M. Jain and
R. Swennen, eds). Kluwer.
Pérez Hernández J.B. 2000. Development and application of Agrobacterium-mediated genetic
transformation to increase fungus-resistance in banana (Musa spp). PhD thesis No. 442,
Katholieke Universiteit Leuven, Belgium. 187pp.
Pérez Hernández J.B., S. Remy, V. Galán Saúco, R. Swennen, and L. Sági. 1998. Chemotaxis,
attachment and transgene expression in the Agrobacterium-mediated banana
transformation system. Med. Fac. Landbouww. Univ. Gent 63 (4b):1603-1606.
Pérez Hernández J.B., S. Remy, V. Galan Sauco, R. Swennen and L. Sági. 1999. Chemotactic
movement and attachment of Agrobacterium tumefaciens to single cells and tissues of
banana. Journal of Plant Physiology 155:245-250.
Puchta H. 2000. Removing selectable marker genes: taking the short cut. Trends Plant Sci.
5:273-274.
Puzio P.S., J. Lausen, J. Almeida-Engler, D.G. Cai, G. Gheysen and F.M.W. Grundler. 1999.
Isolation of a gene from Arabidopsis thaliana related to nematode feeding structures. Gene
239:163-172.
Reed J., L. Privalle, L. Powell, M. Meghji, J. Dawson, E. Dunder, J. Suttie, A. Wenck, K. Launis,
C. Kramer, Y.-F. Chang, G. Hansen and M. Wright. 2001. Phosphomannose isomerase: an
efficient selectable marker for plant transformation. In Vitro Cell. Dev. Biol. 37:127-132.
Remans T., L. Sági, A.R. Elliotts, R.G. Dietzgen, R. Swennen, P. Ebert, C.P.L. Grof, J.M. Manners
and P.M. Schenk. 2000. Banana streak virus promoters are highly active in transgenic
banana and other monocot and dicot plants. 2nd International Symposium on the Molecular
and Cellular Biology of Banana. Byron Bay, Australia, October-November 2000.
Remy S. 2000. Genetic transformation of banana (Musa spp.) for disease resistance by
expression of antimicrobial proteins. PhD thesis No. 420, Katholieke Universiteit Leuven,
Belgium. 341pp.
234
MyLsd 17x24
25/06/03
15:15
Page 235
Session 4
R. Swennen et al.
Remy S., I. François, B.P.A. Cammue, R. Swennen and L. Sági. 1998a. Co-transformation as
a potential tool to create multiple and durable resistance in banana (Musa spp.). Acta
Hort. 461:361-365.
Remy S., A. Buyens, B.P.A Cammue, R. Swennen and L. Sági. 1998b. Production of
transgenic banana plants expressing antifungal proteins. Acta Hort. 490:433-436.
Remy S., G. De Weerdt, I. Deconinck, R. Swennen and L. Sági. in press. An ultrasensitive
luminescent detection system in banana biotechnology: from promotor tagging to
southern hybridization. Banana Improvement: Cellular, Molecular Biology and Induced
Mutations (S.M. Jain and R. Swennen, eds) Kluwer
Roux N., H. Strosse, A. Toloza, B. Panis and J. Dolozel. In press a. Detecting ploidy level
instability of banana embryogenic cell suspension cultures by flow cytometry in
Banana Improvement: Cellular, Molecular Biology and Induced Mutations (S.M. Jain and
R. Swennen, eds). Kluwer.
Roux N., A. Toloza, J. Dolozel and B. Panis. In press b. Usefulness of embryogenic cell
suspension cultures for the induction and selection of mutants in Musa spp. in Banana
Improvement: Cellular, Molecular Biology and Induced Mutations (S.M. Jain and
R. Swennen, eds). Kluwer.
Roux N., Dolezel J., Swennen R. and F.J. Zapata-Arias. 2001. Effectiveness of three
micropropagation techniques to dissociate cytochimeras in Musa spp. Plant Cell, Tissue
and Organ Culture 66:189-197.
Rowe P. 1984. Breeding bananas and plantains. Plant Breeding Reviews 2:135-155.
Rowe P. and F. Rosales. 1990. Breeding bananas and plantains with resistance to black sigatoka.
Pp. 243-251 in Sigatoka Leaf Spot Disease of Bananas. (R. Fullerton and R. Stover, eds).
INIBAP, Montpellier, France.
Ruf S., M. Hermann, I.J. Berger, H. Carrer and R. Bockl. 2001. Stable genetic transformation
of tomato plastids and expression of a foreign protein in fruit. Nat. Biotechnol. 19:870875.
Sági L. 1998. Control of flowering time in banana (Musa spp.), a tropical monocot, by genetic
transformation. Acta Agronomica Hungarica 46:439-448.
Sági L., S. Remy, B. Panis, G. Volckaert and R. Swennen. 1992. Transient gene expression in
banana protoplasts. Banana Newsletter 15:42.
Sági L., S. Remy, B. Panis, R. Swennen and G. Volckaert. 1994. Transient gene expression in
electroporated banana (Musa spp., cv. ‘Bluggoe’, ABB group) protoplasts isolated from
regenerable embryogenic cell suspensions. Plant Cell Rep. 13:262-266.
Sági L., B. Panis, S. Remy, H. Schoofs, K. De Smet, R. Swennen and B.P.A. Cammue. 1995a.
Genetic transformation of banana and plantain (Musa spp.) via particle bombardment.
Bio/Technology 13:481-485.
Sági L., S. Remy, B. Verelst, B. Panis, B.P.A. Cammue, G. Volckaert and R. Swennen. 1995b.
Transient gene expression in transformed banana (Musa cv. ‘Bluggoe’) protoplasts and
embryogenic cell suspensions. Euphytica 85:89-95.
Sági L., B. Panis, S. Remy, B.P.A. Cammue and R. Swennen. 1995c. Stable and transient genetic
transformation of banana (Musa spp. cv. ‘Bluggoe’, ABB-group) using cells and protoplasts.
Pp. 85-102.in Proceedings of the 11th ACORBAT Meeting (Morales Soto V. ed.). San José,
Costa Rica, 13-18 February 1994
Sági L., G.D. May, S. Remy and R. Swennen. 1998a. Recent developments in biotechnological
research of banana (Musa spp.). Pp. 313-327 in Biotechnology and Genetic Engineering
Reviews (M.P. Tombs, ed.). Intercept Ltd, Andover, England.
235
MyLsd 17x24
25/06/03
15:15
Page 236
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Sági L., S. Remy and R. Swennen. 1998b. Genetic transformation for the improvement of
bananas - A critical assessment. Focus Paper II. Pp. 33-35 in Networking Banana and
Plantain: INIBAP Annual Report 1997. INIBAP, Montpellier, France.
Sági L., S. Remy, J.B. Pérez Hernández, B.P.A. Cammue and R. Swennen. 2000. Transgenic
banana (Musa Species). pp. 255-268 in Biotechnology in Agriculture and Forestry,
Transgenic Crops II (Y. P. S. Bajaj, ed.). Vol. 47. Springer, Berlin, Heidelberg, New York.
Schenk P.M., L. Sági, T. Remans, R.G. Dietzgen, M.J. Bernard, M.W. Graham and J.M. Manners.
1999. A promoter from sugarcane bacilliform badnavirus drives transgene expression in
banana and other monocot and dicot plants. Plant Molecular Biology 39:1221-1230.
Schenk P.M., T. Remans, L. Sági, A.R. Elliott, R.G. Dietzgen, R. Swennen, P. Ebert, C.P.L. Grof
and J.M. Manners. 2001. Promoters for pregenomic RNA of banana streak badnavirus are
active for transgene expression in monocot and dicot plants. Plant Molecular Biology
47:399-412.
Schoofs H. 1997. The origin of embryogenic cells in Musa. PhD thesis No. 330, Katholieke
Universiteit Leuven, Belgium. 257p.
Schoofs H., B. Panis., H. Strosse, A. Mayo Mosqueda, J. Lopez Torres, N. Roux, J. Dolozel
and R. Swennen. 1999. Bottlenecks in the generation and maintenance of morphogenic
banana cell suspensions and plant regeneration via somatic embryogenesis therefrom.
INFOMUSA 8(2):3-7.
Sharma H.C., K.K. Sharma, N. Seetharama and R. Ortiz. 2000. Prospects for using transgenic
resistance to insects in crop improvement. Electronic Journal of Biotechnology 3(2).
http://www.ejb.org/content/vol3/issue2/full/3/
Sherf B.A. and K.V. Wood. 1994. Firefly luciferase engineered for improved genetic reporting.
Promega Notes 49:14-21.
Shirasu K., H. Nakajima, K. Rajashekar, R.A. Dixon and C. Lamb. 1997. Salicylic acid potentiates
an agonist-dependent gain control that amplifies pathogen signal in the activation of
defense mechanisms. Plant Cell 9:261-270.
Sidorov V.A., D. Kasten, S.Z. Pang, P.T.J. Hajdukiewicz, J.M. Staub and N.S. Nehra. 1999.
Technical advance: stable chloroplast transformation in potato: use of green fluorescent
protein as a plastid marker. Plant J. 19:209-216.
Strosse H., I. Van den houwe and B. Panis. in press. Banana cell and tissue culture review.
In Banana Improvement: Cellular, Molecular Biology and Induced Mutations (S.M. Jain
and R. Swennen, eds). Kluwer.
Suprasanna P., L. Sági and R. Swennen. 2002. Positive selectable marker genes for routine
plant transformation. In Vitro Cell Dev Biol Plant 38:125-128.
Swennen R. 1990. Limits of morphotaxonomy: names and synonyms of plantain in Africa
and elsewhere. Pp. 172-210 in The identification of genetic diversity in the genus Musa.
Proceedings of an International Workshop (R. L. Jarret, ed.). Los Baños, Philippines, 510 September 1988. INIBAP, Montpellier, France.
Swennen R. 1994. De veredeling van de banaan voor resistentie tegen de bladschimmel
Mycosphaerella fijiensis. Mededeling der Zittingen Koninklijke Academie voor Overzeese
Wetenschappen 39:567-576.
Swennen R., I. Van den houwe, S. Remy, L. Sági and H. Schoofs. 1998. Biotechnological
approaches for the improvement of Cavendish bananas, Acta Horticultura 490:415-423.
Swennen R. and D. Vuylsteke. 1993. Breeding black Sigatoka resistant plantains with a wild
banana. Tropical Agriculture (Trinidad) 70(1):74-77.
Tailor R.A., D.P. Acland, S. Attenborough, B.P.A. Cammue, I.J. Evans, R.W. Osborn, J. Ray,
S.B. Rees and W.F. Broekaert. 1997. A novel family of small cysteine-rich antimicrobial
236
MyLsd 17x24
25/06/03
15:15
Page 237
Session 4
R. Swennen et al.
peptides from seeds of Impatiens balsamina is derived from a single precursor protein. J.
Biol. Chem. 272:24480-22487.
Terras F.R.G., I.J. Goderis, F. Van Leuven, J. Vanderleyden, B.P.A. Cammue and W.F.
Broekaert. 1992a. In vitro antifungal activity of a radish (Raphanus sativus L.) seed protein
homologous to nonspecific lipid transfer proteins. Plant Physiol. 100:1055-1058.
Terras F.R.G., H. Schoofs, M.F.C. De Bolle, F. Van Leuven, S.B. Rees, J. Vanderleyden, B.P.A.
Cammue and W.F. Broekaert. 1992b. Analysis of two novel classes of plant antifungal
proteins from radish (Raphanus sativus L.) seeds. J. Biol. Chem. 267:15301-15309.
Trewavas A. 2000. GM is the best option we have. Agbioview. http://agbiovieuw.listbot.com/
UNDP. 2001. Human Development Report 2001. Making new technologies work for human
development. http://www.undp.org/hdr2001/
Van den houwe I., J. Guns and R. Swennen. 1998. Bacterial contamination in Musa shoot
tip cultures. Acta Horticultura 490:485-492.
Van den houwe I. and R. Swennen. 2000. Characterization and control of bacterial
contaminants in in vitro cultures of banana (Musa spp.). Acta Horticulturae 530:69-79.
Van den houwe I., B. Panis and R. Swennen. 2000. The in vitro germplasm collection at the
Musa INIBAP Transit Centre and the importance of cryopreservation. Pp.255-260 in
Cryopreservation of tropical plant germplasm, Current research progress and applications
(F. Engelmann and H. Takagi, eds). IPGRI, Rome, Italy .
Van der Hoorn R.A.L., P.J.G.M De Wit and M.H.A.J Joosten. 2002. Balancing selection favors
guarding resistance proteins. Trends Plant Science 7:67-71.
Van Loon L.C., Y.A.M. Gerritsen and C.E. Ritter. 1987. Identification, purification and
characterization of pathogenesis-related proteins from virus-infected Samsun NN tobacco
leaves. Plant Mol. Biol. 9:593-609.
Vance C.P., T.H. Kirk and R.T. Sherwood. 1980. Lignification as a mechanism of disease
resistance. Annu. Rev. Phytopathol. 18:259-288.
Vasil I.K. 1987. Developing cell and tissue culture systems for the improvement of cereal and
grass crops. Journal of Plant Physiology 128:193-218.
Vasil V. and I.K. Vasil. 1986. Plant regeneration from friable embryogenic callus and cell
suspension cultures of Zea mays L. Journal of Plant Physiology 124:399-408.
Vuylsteke D. and R. Swennen. 1993. Genetic improvement of plantains: the potential of
conventional approaches and the interface with in vitro culture and biotechnology.
Pp. 169-176 in Biotechnology applications for banana and plantain improvement. San
José, Costa Rica, 27-31 January 1992.
Vuylsteke D., R. Swennen and R. Ortiz 1993a. Registration of 14 improved tropical Musa
plantain hybrids with black Sigatoka resistance. HortScience 28(9):957-959.
Vuylsteke D., R. Swennen and R. Ortiz. 1993b. Development and performance of
black Sigatoka-resistant tetraploid hybrids of plantain (Musa spp., AAB group). Euphytica
65:33-42.
Vuylsteke D., R. Ortiz and S. Ferris. 1993c. Genetic and agronomic improvement for
sustainable production of plantain and banana in Sub-saharan Africa. African Crop Science
Journal 1(1):1-8.
Vuylsteke D., R. Ortiz and R. Swennen. 1994. Breeding plantain hybrids for resistance to Black
Sigatoka. IITA Research 8:9-14.
Vuylsteke D., R. Ortiz, S. Ferris and R. Swennen. 1995. ‘PITA-9’: a black-sigatoka-resistant
triploid hybrid from the ‘False Horn’ plantain gene pool. HortScience 30(2):395-397.
Wiame I., R. Swennen and L. Sági. 2000. PCR-based cloning of candidate disease resistance
genes from banana (Musa acuminata). Acta Horticulturae 521:51–57.
237
MyLsd 17x24
25/06/03
15:15
Page 238
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Wong H. C., Y. Ting, H.-C. Lin, F. Reichert, K. Myambo, K. Watt, P.L. Toy, and R.J. Drummond.
1991. Genetic organization and regulation of the xylose degradation genes in Streptomyces
rubiginosus. J. Bacteriol. 173:6849-6858.
Zhu Y., H. Chen, J. Fan, Y. Wang, Y. Li, J. Chen, J. Fan, S. Yang, L. Hu, H. Leung, T.W. Mew,
P.S. Teng, Z. Wang and C.C. Mundt. 2000. Genetic diversity and disease control in rice.
Nature 406:718-722.
238
MyLsd 17x24
25/06/03
15:15
Page 239
Session 4
N. Roux et al.
Mutagenesis and somaclonal
variation to develop new resistance
to Mycosphaerella leaf spot diseases
N. Roux1, A. Toloza1, J.P. Busogoro2, B. Panis3,
H. Strosse3, P. Lepoivre2, R. Swennen3 and F. J. Zapata-Arias1
Abstract
Mycosphaerella leaf spot diseases can reduce fruit yield by up to 50%. Chemical strategies exist
to combat these diseases, but they are environmentally unsound, hazardous and very expensive
for many farmers. The only sustainable means to reduce the use of pesticides is breeding for
tolerant cultivars. Whereas breeders are looking for genetic variability to develop new varieties,
edible Musa cultivars are multiplied vegetatively, polyploid and sterile. Spontaneous somatic
mutation has already contributed largely to obtaining new cultivars of Musa. Nevertheless, the
rate of occurrence is too low to satisfy practical breeding needs. Mutations can be induced by
physical or chemical mutagens. With the development of tissue culture techniques, in vitro
mutagenesis and somaclonal variation raised hopes in the 1980-1990s. In spite of this, very few
useful and stable mutants/somaclones were obtained. The multicellular structure of meristems
which leads to chimerism is certainly an impeding factor. Additionally, the random process in
mutation induction calls for the screening of several thousand plants after treatment. Recently,
following a five-year FAO/IAEA/DGIC coordinated research project, it has been possible to
overcome these two barriers by mutagenic treatment of embryogenic cell suspensions and by
establishing an early mass screening method resting on infiltration of juglone, a toxic metabolite
of Mycosphaerella fijiensis. After screening approximately 4000 plants, 15 putative mutants
showed tolerance to this metabolite. These plants must be evaluated for their resistance to
M. fijiensis infection under controlled conditions and field experiment.
Resumen - Mutagénesis y variación somaclonal para desarrollar nueva resistencia a
las enfermedades de la mancha foliar por Mycosphaerella
Las enfermedades de las manchas foliares causadas por Mycosphaerella spp. afectan
significativamente el cultivo bananero y puede reducir el rendimiento de la fruta en hasta un
1
International Atomic Energy Agency Laboratories, Seibersdorf, Austria
University Gembloux, Gembloux, Belgium
3Katholieke Universiteit Leuven, Leuven, Belgium
2Agricultural
239
MyLsd 17x24
25/06/03
15:15
Page 240
Mycosphaerella leaf spot diseases of bananas: present status and outlook
50%. Existen estrategias de control químico para combatir estas enfermedades, pero estas causan
daños al ambiente, son peligrosas y muy costosas para muchos agricultores. El único medio
sostenible para reducir el uso de plaguicidas es el mejoramiento de los cultivares tolerantes. Los
mejoradores que están investigando la variabilidad genética para desarrollar nuevas variedades,
toman en cuenta que las variedades comestibles de Musa se multiplican vegetativamente, son
poliploides y estériles. La mutación somática espontánea es una fuente de variación que ya ha
contribuido en gran medida en la obtención de las nuevas variedades en Musa spp. Sin embargo,
la tasa de ocurrencia es muy baja para satisfacer las necesidades prácticas de mejoramiento. Las
mutaciones pueden ser inducidas por mutágenos físicos o químicos. Con el desarrollo de las
técnicas del cultivo de tejidos, la mutagénesis y variación somaclonal in vitro elevaron las
esperanzas en las décadas de los 80 y 90. A pesar de esto, se obtuvieron pocos mutantes o
somaclones útiles y estables. La estructura multicelular de los meristemas que conduce al
quimerismo es ciertamente un factor de impedimento. Adicionalmente, el proceso aleatorio para
inducir la mutación requiere el cribado de varios miles de plantas después del tratamiento.
Recientemente, después de realizar un proyecto de investigación de cinco años coordinado por
FAO/IAEA/DGCI, fue posible vencer estas dos barreras mediante un tratamiento mutagénico de
las suspensiones de células embriogénicas y el establecimiento de un método de cribado
masivo temprano que se basa en la infiltración de juglone, un metabolito tóxico de Mycosphaerella
fijiensis.. Después de realizar el cribado de aproximadamente 4000 plantas, 15 mutantes putativos
mostraron tolerancia a este metabolito.
Résumé – Mutagénèse et variation somaclonale pour développer de nouvelles
résistances aux maladies foliaires causées par Mycosphaerella spp.
Les maladies foliaires causées par Mycosphaerella spp. peuvent réduire le rendement de jusqu’à
50%. Des moyens chimiques existent pour lutter contre ces maladies, mais ils sont nocifs pour
l’environnement, dangereux et très coûteux pour nombre de fermiers. La seule manière durable
de réduire le recours aux insecticides est de créer des cultivars tolérants. Sauf que pour y arriver,
les sélectionneurs ont besoin de variabilité génétique et que les cultivars de bananiers sont
polyploïdes, stériles et multipliés végétativement. Les mutations somatiques spontanées ont
déjà passablement contribuées à l’obtention de nouveaux cultivars, mais leur fréquence est trop
faible pour satisfaire les besoins des sélectionneurs. Des agents mutagènes chimiques et
physiques peuvent provoquer des mutations. Dans les années 1980-1990, le développement des
techniques de culture de tissus, mutagénèse in vitro et variation somaclonale ont soulevé bien
des espoirs. Malgré cela, peu de mutants/somaclones ont été obtenus. La structure multicellulaire
des méristèmes, qui produit des chimères, est sans doute un facteur limitant. De plus,
l’induction de mutations est un processus aléatoire qui nécessite le criblage de plusieurs milliers
de plants. Récemment, suite à un projet de recherche de cinq ans coordonné par FAO/IAEA/DGIC,
il a été possible de contourner ces deux obstacles en faisant subir un traitement mutagène à
des suspensions de cellules embryogéniques et en mettant au point une méthode de criblage
précoce basée sur l’infiltration de juglone, un métabolite toxique de Mycosphaerella fijiensis.
Après avoir criblé 4000 plants, 15 mutants présomptifs ont montré une tolérance à ce
métabolite. Ces plants doivent être évalués pour leur résistance à M. fijiensis sous des conditions
contrôlées et en champ.
Introduction
Bananas and plantains (Musa spp.) are a staple food of millions of people and rank
among the top five food commodities. However, Mycosphaerella leaf spot diseases
can reduce fruit yield by 50% (Mourichon et al., 1997). Chemical control stategies
exist to combat these diseases, but they are environmentally unsound, hazardous
and very expensive for many farmers (Persley and George, 1999). Breeding of resistant
240
MyLsd 17x24
25/06/03
15:15
Page 241
Session 4
N. Roux et al.
cultivars is the only sustainable mean to reduce the use of pesticides. But edible Musa
are difficult to breed since they are polyploid and sterile.
Spontaneous mutations are at the origin of almost all of the edible banana and
plantain cultivars (Buddenhagen, 1987). The best example is the spontaneous banana
mutant ‘Cavendish’ from Vietnam, which is resistant to Fusarium wilt (race 1) and
which replaced ‘Gros Michel’ in the 1950s and 1960s (Ploetz, 1994). The discovery
of this banana mutant saved the banana industry from collapsing but as a
consequence, the export trade relies on a very narrow genetic base since only one
or two triploid cultivars of the subgroup Cavendish dominate the export market
(Risède and Tézenas du Montcel, 1997).
Banana is probably the best example in the history of agriculture of the
pathological perils of monoclone culture. Indeed, without clonal diversification, the
trade can hardly be expected to survive indefinitely and the generation and use of
genetic variability may be the only remaining option. Spontaneous somatic mutation
has already contributed largely in obtaining new cultivars in Musa spp. Nevertheless,
the rate of occurrence is too low to satisfy practical breeding needs. Mutations can
be induced by tissue culture (somaclonal variation) and/or by physical or chemical
mutagens (induced mutants). With the development of tissue culture techniques, in
vitro mutagenesis and somaclonal variation raised hopes in the 1980 and 1990s. So
far, however, very few useful and stable mutants/somaclones have been obtained.
The multicellular structure of meristems, which leads to chimerism is certainly an
impeding factor. Additionally, the random process of mutation induction calls for
the screening of several thousand plants after treatment. Recently, following a fiveyear FAO/IAEA/DGIC coordinated research project, it has been possible to overcome
these two barriers. This paper presents the improved methodology and the potential
use of mutants in genetic improvement programmes.
Advantages and limitations of induced mutation
and somaclonal variation
Somaclonal variants
In vitro propagated plants are not necessarily true to type. Off-type plants might
differ permanently (i.e. somaclonal variation) or temporarily from the source plant
as a result of an epigenetic or physiological effect. The term ‘somaclonal variation’
was introduced to describe the genetic variation in plants regenerated from any form
of cell culture. Larkin and Scowcroft (1981) advocate the view that somaclonal
variation represents a new source of variability and therefore constitutes a powerful
tool to the breeder. Nevertheless somaclonal variation from micropropagated banana
and plantain should not be overestimated as a source of novel variability for use in
genetic improvement (Vuylsteke et al., 1991). A narrow spectrum of variants has
been obtained through somaclonal variation. It is becoming increasingly clear that
somaclonal variation is usually undesirable (Vuylsteke et al., 1996).
Some off-types have improved agronomical traits, such as the higher yield of
the ‘French reversion’ variant plantain and the short stature of dwarfs (Vuylsteke et
al., 1996). Regarding disease resistance, a ‘Cavendish’ banana was recovered in Taiwan
241
MyLsd 17x24
25/06/03
15:15
Page 242
Mycosphaerella leaf spot diseases of bananas: present status and outlook
for its tolerance to Fusarium wilt (Hwang et al., 1993). More recently Trujillo et al.
(1996) reported the selection of a somaclonal variant CIEN BTA-03 tolerant to
Sigatoka disease. But cytogenetic and molecular characterization revealed that CIEN
BTA-03 was in fact a tetraploid clone and was not part of the Cavendish subgroup
to which the parental ‘Williams’ (AAA) belongs (Gimenez et al., 2001).
Induced mutants
The Musa mutation induction system based on in vitro techniques to recover mutant
plants and micropropagate desirable mutants was developed by Novák et al. (1990)
in the FAO/IAEA Laboratories. It is now applied in a few Musa breeding programmes.
Gamma irradiation is the main physical mutagen used to induce genetic variation.
More recently, Roux (1997) standardized the methodology to provide guidelines to
mutation induction programmes in Musa spp. Shoot-tips excised from clones
representing the different genomic constitutions of the genus Musa were treated with
10 doses from 10 to 100 Gy of a 60Co gamma irradiation source at a dose rate of 44
Gy/min. For each Musa accession, 200 explants were treated for sensitivity testing
and 20 non-irradiated explants were used as control. Radiation sensitivity and postirradiation recovery were assessed by measuring the survival rate, the propagation
rate, the shoot height and the fresh weight.
The different Musa accessions showed different responses according to their ploidy
level and genomic constitution. The following ranges of doses are recommended:
• 10 to 20 Gy of gamma irradiation for diploid clones ‘Calcutta 4’ (AA) and
‘Tani’ (BB);
• 30 to 40 Gy of gamma irradiation for the triploids ‘Three hand planty’ (AAB),
‘Grande naine’ (AAA), ‘Williams’ and ‘Kamaramasenge’ (triploid, formerly
classified as AB);
• 40 to 50 Gy of gamma irradiation for the triploid ‘Cachaco’ (ABB).
From the FAO/IAEA mutant varieties database, two banana accessions were
registered as improved mutant varieties: ‘Novaria’ for early flowering and ‘Klue hom
thong KU1’ for its bunch size and cylindrical shape from which larger bananas can
be selected.
Other desirable variants/putative mutants have been identified for release or
further confirmation trials. Examples are shown in Table 1.
Most of the improved characteristics are agronomic features. Disease resistance
seems to be difficult to obtain through mutation induction techniques. Consequently,
Smith et al. (1995) used an original strategy. Instead of irradiating an agronomically
superior but susceptible genotype, they irradiated ‘Dwarf parfitt’, an extra dwarf
Cavendish banana that has shown a high level of resistance to race 4 of Fusarium
wilt. Following radiation, 35 M1V3 (M: Mutagenic treatment; V: vegetative generation)
out of 500 explants irradiated at 20 Gy were recovered that possessed improved
agronomic characteristics (taller plant size, increased yield and no choking). Most
importantly these selections appeared to retain the resistance to race 4 derived from
the mother plant ‘Dwarf parfitt’.
242
MyLsd 17x24
25/06/03
15:15
Page 243
Session 4
N. Roux et al.
Bhagwat and Duncan (1998) used gamma irradiation (8 to 20 Gy) and chemical
mutagens to make ‘Highgate’ (AAA) tolerant to Fusarium oxysporum f.sp. cubense.
Twelve weeks after inoculation in the greenhouse, 0.3 to 0.9% of the regenerated
plants derived from irradiated explants and 1.9 to 6.1% of chemically treated explants
had less than 10% vascular invasion in their corms with no external symptoms. These
plants were considered tolerant to the fungus and were multiplied, ex vitro, for field
screening.
Table 1. List of putative mutants obtained.
Country
Cuba
Parent clone
SH-3436
Parecido al Rey
Malaysia
Pisang rastali
Novaria
Philippines Lakatan
Latundan
Sri Lanka Embul
Austria
Grande naine
(IAEA**)
Selected clone
SH-3436-L9
6.44
Mutiara
LK-40
LT-3
Embul-35 Gy
GN35-I to
GN35-VIII
Selected traits
Reduced height
Reduced height
FOC* tolerance
FOC tolerance
Reduced height
Larger fruit size
Earliness
Tolerance to toxin
from Mycosphaerella
fijiensis
Technique
Place of induction
Gamma rays
Gamma rays
Somaclonal variation
Somaclonal variation
Gamma rays
Gamma rays
Gamma rays
Gamma rays
Cuba
IAEA
Malaysia
Malaysia
IAEA
IAEA
Sri Lanka
IAEA
* FOC: Fusarium oxysporum f.sp. cubense
**IAEA: International Atomic Energy Agency.
Even though the traditional shoot tip mutation induction technique has permitted
to obtain useful mutants, the following limitations are impeding its wider use:
• The treatment of shoot-tips with mutagenic agents (physical or chemical) results
in a high degree of chimerism. This is a serious obstacle to mutation techniques
since it is not yet possible to distinguish mutated cells from none mutated cells.
• Since mutation induction is a random process, efficiency requires the need to
treat and screen as many plants as possible. However, a bottleneck occurs due
to the time spent on field screening. The current methods of field screening are
also site-specific and involve considerable resources: large numbers of technicians,
hours of work, fertilizer, logistic support and high cost.
Recent technical achievements
Origin of embryogenic cell suspensions
In order to screen efficiently for characters such as disease resistance, an efficient
method to overcome chimerism after mutagenic treatment is needed (Roux et al.,
2001). Considering this, somatic embryogenesis is the most promising method since
somatic embryos are assumed to be of single cell origin (Halperin, 1966). In some
species, histological studies confirmed the single cell origin of somatic embryos. They
develop either directly from an explant or as secondary embryos at the surface of
older somatic embryos (Litz and Gray, 1992). Grapin et al. (1998) stipulated from
cytological studies on somatic embryo ontogenesis in Musa that a unicellular origin
was more than likely. However such studies can only be performed on few somatic
243
25/06/03
15:15
Page 244
Mycosphaerella leaf spot diseases of bananas: present status and outlook
embryos and thus the extrapolation of these findings to a large number of somatic
embryos is risky.
Colchicine treatment and ploidy analysis with flow cytometry, used by Roux et
al. (2001) to monitor the efficiency of 3 different micropropagation techniques in
dissociating chimeras, was applied to verify the unicellular origin of somatic embryos
from ECS. After treating cell suspensions with colchicine (Figure 1), the embryos
were subsequently transferred on a regeneration medium in test tubes. As soon as
green plantlets with shoot and roots were obtained, leaves pieces of 0.5 cm2 were
excised and their ploidy measured through flow cytometry before acclimatization
(Table 2). The majority of the regenerated plants were triploid. Among the treated
cells, the proportion of regenerated hexaploid plants (5.3%) remained very low
compared to triploid plants. We think that triploid cells have a comparative
advantage over induced hexaploid cells during culture. In contrast to shoot-tip
cultures that were treated with colchicine (Roux et al., 2001), no mixoploids among
the regenerated plants were observed, which confirms the single cell origin of
embryos. Thus embryogenic cell suspensions seem to be the material of choice for
mutagenic treatments.
100
Ploidy frequency distribution (% of nuclei)
MyLsd 17x24
90
80
70
60
50
40
30
20
10
0
0%
3x
0.05%
6x
0.1%
0.2%
Colchicine concentration
Figure 1. Ploidy frequency distribution of ‘Williams’ suspension cells (cell-line 124T) with triploid (3x) and
hexaploid (6x) nuclear DNA content observed 15 days after colchicine treatment (w/v).
244
MyLsd 17x24
25/06/03
15:15
Page 245
Session 4
N. Roux et al.
Table 2. Ploidy distribution of regenerated plants from colchicines-treated ECS of the triploid cell line 124T.
Colchicine
concentration (%)
0
0.05
0.1
0.2
Number of regenerated
plants
Ploidy
3n
6n
Chimeric mixoploidy (3n+6n)
108
63
37
88
108
58
36
84
0
5
1
4
0
0
0
0
296
286
10
0
Mutagenic treatment of embryogenic cell suspensions
Two cell lines of the cultivars ‘Williams’ and ‘Three hand planty’ were sieved using a
1000-mm pore size mesh to obtain a fine suspension. The ECS were then subcultured
in the maintenance medium (ZZ) in 100-ml flasks at a concentration of about 3% of
settled cell volume (SCV). Three days after subculture 0.5 ml of cells were transferred
to a sterile Petri dish and the medium was removed. The cells aggregates were then
irradiated at doses ranging from 0 to 250 Gy with 25 Gy intervals using a 60Co gamma
source at a dose rate of 30 Gy/min. After irradiation, the cells were resuspended in
maintenance ZZ medium in centrifuge tubes and transferred to 100-ml Erlenmeyer flasks
at different quantities according to the parameter to be analyzed. To study the effect
of gamma radiation on the growth of ECS, fresh weight gain and regeneration capacity
were determined. The results were expressed in percentage of the control (non-irradiated
cells) at all doses.
The two radiosensitivity curves for ‘Williams’ and ‘Three hand planty’ are quite similar.
After 28 days, at 75 Gy, the cells’ weight gain (CWG75=0,84g) was 50% of the control
(CWG0= 1,68g) (Figure 2A).
To measure the regeneration capacity, green plantlets were counted and transferred
to Magenta GA7 boxes containing semi-solid R3 regeneration medium for further growth
before acclimatization. The radiosensitivity curve was obtained by comparing the number
of regenerated plantlets for each dose with the control plants (from non-irradiated ECS)
(Figure 2B). Radiation at a low level seems to stimulate the regeneration capacity
especially in ‘Williams’. We must, however, take into account that in control plants, the
density of embryos in the temporary immersion system vessels was too high and hence,
a considerable number of small plantlets could not develop. In both genotypes no plants
regenerated above 200Gy. The number of regenerated plants drops drastically above
50 Gy for ‘Williams’ whereas for ‘Three Hand Planty’ the number of regenerated plantlets
decreases less dramatically. The regenerated plants were transferred to the greenhouse
for early mass screening for tolerance to black leaf streak disease.
Establishment of an early mass screening method
Genetic variability is a prerequisite before selecting for disease resistance. A technique
that can reliably identify resistant plants is then adopted to screen the populations
(Lepoivre et al., 1993). The Plant Pathology Unit at the Faculté Universitaire des Sciences
245
25/06/03
15:15
Page 246
Mycosphaerella leaf spot diseases of bananas: present status and outlook
120
100
WIL
THP
80
60
40
20
0
0
50
100
150
Dose (Gy)
200
Regeneration capacity (% of control)
140
Fresh weight gain (% of control)
MyLsd 17x24
120
80
60
40
20
0
0
250
WIL
THP
100
50
100
150
Dose (Gy)
200
250
Figure 2. Effect of gamma radiation on : (A) fresh weight gain 28 days after irradiation and (B) regeneration
capacity at 60 to 90 days after irradiation. All parameters are expressed as a percentage of the non-irradiated
ECS (control). Cell suspension cultures of the cultivars ‘Willliams’ (WIL) and ‘Three hand planty’ (THP) were
used. For each treatment the equivalent of four Petri dishes were measured.
Agronomiques de Gembloux (FUSAGx) developed a rapid and early screening protocol
for tolerance to black leaf streak disease in banana and plantain. The method is based
on the infiltration of juglone (5-hydroxy-1,4-naphthoquinone), a toxic metabolite of
Mycosphaerella fijiensis. After developing different bioassays, it was concluded that slow
lesion development in cultivars exhibiting a partial resistance to black leaf streak disease
was correlated with lower sensitivity to M. fijiensis toxins (Lepoivre, 1995). The toxin
is probably not involved in the initiation of the infection but could serve as a secondary
determinant of the pathogenecity, contributing to the lesion expansion in cultivars
exhibiting partial resistance to black leaf streak disease (Harelimana et al., 1997). The
prospect of utilizing plant tissue cultures to generate and identify novel genetic variants
has sparked the interest of researchers for many years (Dix, 1996). Nevertheless in Musa,
in vitro heterotrophic tissues are not suitable targets to perform the screening with such
toxin (Harelimana et al., 1997). Our first goal was thus, under greenhouse conditions,
to determine the lowest concentration of juglone which enables the differentiation
between the susceptible cultivar ‘Grande naine’ and the partially resistant cultivar
‘Fougamou’. After performing 10 assays with 4 to 8 replicates we concluded that 25
ppm was the most suitable concentration of juglone to distinguish between a partially
resistant and a susceptible plant. This dose was thus further used to screen the plants
regenerated from irradiated shoot tips.
Four batches (100 meristems/batch) of ‘Grande naine’ were irradiated at 35 Gy and
propagated over four subcultures. The plants were then acclimatized in the greenhouse.
The early mass screening method can be divided in three steps:
1. Plant preparation: The acclimatized plants which reached the six leaf stage are
maintained at 90 to 100% relative humidity for 48 hours to open the stomata.
246
MyLsd 17x24
25/06/03
15:15
Page 247
Session 4
N. Roux et al.
2. Infiltration: The second open leaf of each plant was infiltrated on its under surface
with juglone (25 ppm), crude extract (500 ppm) and 10% methanol (control). The
amount of each infiltration was 20µl.
3. Necrose observation: 48 hours after infiltration, the plants were observed
for necrosis and compared with control plants (non-irradiated ‘Grande naine’
= positive control; non-irradiated ‘Fougamou’ = negative control).
To date, from a total of 3 728 ‘Grande naine’ plants screened, 15 putative mutants
(0.4%) were selected for their tolerance to 25 ppm of juglone (Table 3).
Table 3. Inventory of irradiated meristems from the cultivar ‘Grande naine’, multiplied, regenerated into plants, screened
and retained for their partial resistance to 25ppm of juglone.
Batch
Irradiated
meristems
Number. of
shoots in M1V1
Number of plants
screened in M1V4
Number of
plants retained
A (ST)
B (ST)
C (MA)
D (MA)
TOTAL
100
100
100
100
400
142
105
110
140
497
780
512
1351
1085
3728
8
0
4
3
15
ST: propagated by shoot tip culture; MA: propagated by the multi-apexing culture technique
Mx: Mutagenic treatment; Vx: Vegetative generation.
The two first batches were propagated by shoot-tip culture and the two second
batches were propagated by the multi-apexing technique to dissociate more efficiently
chimeras (Table 3). Among the young banana plants selected for their tolerance to
juglone, some were showing an increased content of anthocyanin (Figure 3).
Even though we may not have directly obtained mutant genes controlling for
resistance, genes responsible for anthocyanin biosynthesis may have been activated
and indirectly provided tolerance to black leaf streak disease. Atanassova et al. (2001)
studied the effect of mutations affecting anthocyanin biosynthesis during tomato and
pepper development under stress conditions. They found four genes, which had a kind
of universal effect on tomato and pepper germination as they increased the germination
potential of the individual accession under a relatively large range of stresses.
Figure 3.
Regenerated plants from irradiated
shoot tips of the cultivar ‘Grande naine’:
A) susceptible to 25 ppm of juglone;
B) putative mutant, tolerant to 25 ppm
of juglone.
247
MyLsd 17x24
25/06/03
15:15
Page 248
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Thus the establishment of morphological marker correlated with tolerance to stress
even at one plant growth stage, could be useful. The pre-selected plants and their
suckers were transferred to the FUSAGx for further screening by inoculation under
control conditions.
Conclusion and future directions
Over the last five years, considerable efforts have been assembled to speed up
the in vitro mutagenesis process and make it more efficient. Mutation induction
techniques as well as other genetic improvement strategies will benefit from the
use of embryogenic cell suspensions and the establishment of early mass
screening techniques. Tolerance to Mycosphaerella fijiensis obtained after
mutagenic treatment still needs to be confirmed under field conditions. Such
mutants may not have the required agronomic characters but they would be very
useful in genetic studies and may help in discovering genes responsible for
resistance or susceptibility.
Mutation induction should no longer be seen as an independent genetic
improvement strategy but more as a tool which can contribute to functional
genomics and genetic improvement programmes based on cross-breeding or
genetic transformation. For example, a disease resistant mutant from a diploid
Musa, could be used as a parent line in cross-breeding programmes and also help
in understanding the mechanism of resistance and permit the isolation of genes
to be used in genetic transformation.
Acknowledgements
The authors wish to thank Ms. Ines Van den Houwe, (INIBAP) for providing the vegetative
clones of Musa. This work was supported by a Joint FAO/IAEA/GDIC (Belgian General
Direction for International Cooperation) Coordinated Research Project. The study was
undertaken as part of the Global Programme for Musa Improvement (PROMUSA).
References
Atanassova B., S. Daskalov, L. Shtereva and E. Balatcheva. 2001. Anthocyanin mutations
improving tomato and pepper tolerance to adverse climatic conditions. Euphytica
120:357-365.
Bhagwat B. and E.J. Duncan. 1998. Mutation breeding of Highgate (Musa acuminata, AAA)
for tolerance to Fusarium oxysporum f.sp. cubense using gamma irradiation. Euphytica
101:143-150.
Buddenhagen I.W. 1987. Disease susceptibility and genetics in relation to breeding of bananas
and plantains. Pp. 95-109 in Banana and Plantain Breeding Strategies (Persley G.J. and
E.A. De Langhe , eds). ACIAR Proceedings 21, Canberra, Australia.
Gimenez C., E. Garcia, N.X. Enrech and I. Blanca. 2001. Somaclonal variation in banana:
Cytogenetic and molecular characterization of the somaclonal variant CIEN BTA-03.
In Vitro Cell. Dev. Biol.-Plant 37:217-222.
248
MyLsd 17x24
25/06/03
15:15
Page 249
Session 4
N. Roux et al.
Grapin A., J.L Ortiz, R. Domergue, J. Babeau, S. Monmarson, J.V. Escalant, C. Teisson and F.
Côte. 1998. Establishment of embryogenic callus and initiation and regeneration of
embryogenic cell suspensions from female and male immature flowers of Musa.
INFOMUSA 7(1):13-15.
Halperin W. 1966. Alternative morphogenetic events in cell suspensions. American Journal
of Botany 53:443-453.
Harelimana G., P. Lepoivre, H. Jijakli and X. Mourichon. 1997. Use of Mycosphaerella fijiensis
toxins for the selection of banana cultivars resistant to Black Leaf Streak. Euphytica 96:125128.
Hwang S.C., W.H. Ko and C.P. Chao. 1993. GCTCV-215-1: a promising Cavendish clone resistant
to race 4 of Fusarium oxysporum f.sp. cubense. Pp. 62-74 in Recent Developments in
Banana Cultivation Technology (R.V. Valmayor, S.C. Hwang, R. Ploetz, S.W. Lee and
V.N. Roa, eds). Proceedings, International Symposium, Chiuju, Pingtung, Taiwan, 14-18
December 1992. INIBAP, Los Banos, Philippines.
Larkin P.J. and W.R. Scowcroft. 1981. Somaclonal variation - a novel source of variability
from cell culture for plant improvement. Theoretical and Applied Genetics 60:197-214.
Lepoivre P. 1995. Development of screening procedures for resistance to black leaf streak
disease in banana and plantain. End of mission report, IAEA, Vienna 16pp.
Lepoivre P., C.P. Acuna and A.S. Riveros. 1993. Screening procedures for improving resistance
to banana black leaf streak disease. Pp. 213-220 in Breeding banana and Plantain for
Resistance to Diseases and Pests (J. Ganry, ed.). CIRAD, Montpellier, France.
Litz R.E. and D.J. Gray. 1992. Organogenesis and Somatic Embryogenesis. Pp. 3-34 in
Biotechnology of perennial Fruit Crops (Hammerschlag F.A. and Litz R.E. eds). CAB
International, Wallingford, Oxon. U.K.
Mourichon X., J. Carlier and E. Fouré. 1997. Les cercosporioses. Musa Disease Fact Sheet
no8, INIBAP, Montpellier, France.
Novak F.J., R. Afza, M. van Duren and M.S. Omar. 1990. Mutation induction by gamma
irradiation of in vitro cultured shoot-tips of banana and plantain (Musa cvs). Tropical
Agriculture (Trinidad) 67(1):21-28.
Persley G.J. and P. George. 1999. Commodity Advances through Banana Improvement
Research, 1994-1998. Environmentally and socially sustainable development, agricultural
research and extension group series, The World Bank, Washington, D.C. 62pp.
Ploetz R.C. 1994. Panama disease: Return of the first banana menace. International Journal
of Pest Management 40:326-336.
Risede J.M. and H. Tezenas du Montcel. 1997. Banana monocultures and environmental
protection: assessment and perspectives. Fruits 52(4):225-232.
Roux N. 1997. Improved methods to increase diversity in Musa using mutation and tissue
culture techniques. Pp. 49-56 in Report of the second Research Co-ordination Meeting
of FAO/IAEA/BADC Co-ordinated Research Project, Kuala Lumpur. IAEA, Vienna, Austria.
Roux N.S., J. Dolezel, R. Swennen and F.J. Zapata-Arias. 2001. Effectiveness of three
micropropagation techniques to dissociate cytochimeras in Musa sp. Plant Cell, Tissue
and Organ Culture 66:189-197.
Sharrock S. and E. Frison. 1999. Musa production around the world - trends, varieties and
regional importance. Pp. 42-47 in INIBAP Annual Report 1998, focus paper 2. INIBAP,
Montpellier, France.
Smith M.K., S.D. Hamill, P.W. Langdon and Pegg, K.G. 1995. In vitro mutation breeding for
the development of bananas with resistance to race 4, fusarium wilt (Fusarium oxysporum
f.sp. cubense). Pp. 37-44 in Final reports of FAO/IAEA Co-ordinated research programme,
TECDOC-800, IAEA, Vienna, Austria.
249
MyLsd 17x24
25/06/03
15:15
Page 250
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Trujillo I. and E. Garcia. 1996. Strategies for obtaining somaclonal variants resistant to yellow
sigatoka (Mycosphaerella musicola). INFOMUSA 5:12-13.
Vuylsteke D., R. Swennen and E. De Langhe. 1991. Somaclonal variation in plantains (Musa
spp., AAB group) derived from shoot-tip culture. Fruits 46:429-439.
Vuylsteke D., R. Swennen and E. De Langhe. 1996. Field performance of somaclonal variants
of plantain (Musa spp., AAB group). Journal of the American Society for Horticultural
Science 121(1):42-46.
250
MyLsd 17x24
25/06/03
15:15
Page 251
Session 4
M. de J.B. Cavalcante et al.
Reaction of banana genotypes
to black leaf streak disease
in the State of Acre in Brazil
M. de J. B. Cavalcante1, A. da S. Ledo1,
F. H. S. Costa1, Z. J. M. Cordeiro2, A. P. Matos2
Abstract
Black leaf streak disease (caused by Mycosphaerella fijiensis Morelet) is the most severe disease
affecting commercial varieties of banana of economic importance in the world. Its occurrence
was verified in Brazil in 1998, in the State of Amazonas, and it has spread in the plantations through
out the State of the Acre, severely attacking the varieties of the Terra subgroup. The objective
of this study was to evaluate seven banana genotypes from the Embrapa Mandioca e Fruticultura,
under two cultivation systems (with and without cultural practices) with the goal of obtaining
low environmental impact alternatives to control the disease. The evaluations regarding severity
were accomplished on a monthly basis, in ten plants of each genotype, using a descriptive scale.
‘FHIA-01’, ‘FHIA-02’, ‘Caipira’, ‘FHIA-21’, ‘PV 42-85’ and ‘Thap maeo’ presented resistance to black
leaf streak disease whereas ‘SH 36-40’ proved to be susceptible. There was no significant effect
of the cultivation system on the severity of black leaf streak disease.
Resumen - Reacción de los genotipos de banano a la Sigatoka negra en el estado de
Acre, Brasil
La Sigatoka negra (causada por Mycosphaerella fijiensis Morelet) es la enfermedad más severa
que afecta las variedades comerciales de banano de importancia económica en todo el mundo.
La enfermedad se confirmó en Brasil en 1998, en el Estado de Amazonas, y luego se propagó a
las plantaciones a través del Estado de Acre, atacando severamente las variedades del subgrupo
Terra. El objetivo de este estudio consistió en evaluar siete genotipos de banano de Embrapa
Mandioca e Fruticultura, en dos sistemas de cultivo (con y sin empleo de prácticas culturales),
con el fin de obtener alternativas de control para la enfermedad con un bajo impacto ambiental.
Las evaluaciones con respecto a la severidad se realizaron mensualmente en diez plantas de
cada genotipo, utilizando una escala descriptiva. Los resultados mostraron que los genotipos
‘FHIA-01’,‘FHIA-02’,‘Caipira’,‘FHIA-21’,‘PV 42-85’ y ‘Thap maeo’ resultaron resistentes mientras que
el genotipo ‘SH36-40’ fue susceptible a la Sigatoka negra en los dos sistemas de cultivo.
1Embrapa, Rio
2Embrapa
Branco, Brazil
Mandioca e Fruticultura, Cruz das Almas, Brazil
251
MyLsd 17x24
25/06/03
15:15
Page 252
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Résumé – Réaction de génotypes de bananiers à la maladie des raies noires dans l’Etat
de Acre, Brésil
La maladie des raies noires (causée par Mycosphaerella fijiensis Morelet) est la plus sévère des
maladies qui affectent les cultivars commerciaux économiquement importants. Sa présence au
Brésil a été confirmée en 1998, dans l’Etat d’Amazonas, d’où elle s’est propagée à l’Etat de Acre,
affectant sévèrement les variétés du sous-groupe Terra. L’objectif de cette étude a été d’évaluer
sept génotypes de bananiers de l’Embrapa Mandioca e Fruticultura sous deux systèmes de culture
(avec ou sans pratiques culturales) dans le but de développer des méthodes de lutte contre la
maladie dont l’impact sur l’environnement serait faible. Les criblages furent réalisés sur une base
mensuelle, 10 plants par génotype, et utilisant une échelle descriptive. ‘FHIA-01’, ‘FHIA-02’,
‘Caipira’, ‘FHIA-21’, ‘PV 42-85’ et ‘Thap maeo’ ont présenté une résistance à la maladie des raies
noires, tandis que ‘SH 36-40’ s’est avéré susceptible. Il n’y a pas eu d’effet significatif du système
de culture sur la sévérité de la maladie des raies noires.
Introduction
Black leaf streak disease is the most devastating disease of banana worldwide. It
can cause losses up to 100% if no control measure is taken (Cordeiro et al., 1998).
In the State of Acre, banana is the most consumed fruit and is considered a staple
food among the poor populations. It is also exported to other States. The disease
was observed for the first time in Brazil in early 1998 (Cordeiro et al.,1998), in the
municipalities of Tabatinga and Benjamim Constant, and in Rio Branco and
Acrelândia at the end of 1998 (Ritzinger et al., 1999; Cavalcante et al., 1999).
Resistant varieties are not only less expensive to control pathogens than
fungicides, they are also preferable from an environmental point of view (Pereira et
al., 1999).
The present work aims to evaluate the resistance to black leaf streak disease of
banana cultivars under the weather and soil conditions found in Acre and using two
cultivation systems.
Material and methods
The research was conducted in Embrapa’s experimental station in Acre, Rio Branco.
Seven genotypes (‘PV-4285’, ‘FHIA-21’, ‘Caipira’, ‘FHIA-01’, ‘FHIA-02’, ‘SH-3640’ and
‘Thap maeo’) were evaluated for their resistance to black leaf streak disease under
two systems of cultivation: a traditional system (weeding) and a more intensive system
(weeding, trimming, shedding and fertilization).
A randomized complete block design (7 genotypes x 2 cultivation systems) with
5 replications was used and data were recorded on 10 leaves/plant.
Disease severity was assessed on a monthly basis, starting from the sixth month
after planting. The disease was observed in individual leaves, using the following
scale (Stover, 1971 modified by Gauhl, 1994):
0 = absence of symptoms
1 = less than 1% of lamina with symptom (presence of streaks and/or > 10 spots)
2 = 1 to 5% of lamina with symptoms
3 = 6 to 15% of lamina with symptoms
4 = 16 to 33% of lamina with symptoms
252
25/06/03
15:15
Page 253
Session 4
M. de J.B. Cavalcante et al.
5 = 34 to 50% of lamina with symptoms and
6 = 51 to 100% of lamina with symptoms
The variables were submitted to an analysis of variance (F test) and the averages
were compared using Scott & Knott’s test (1974) at 1% of significance.
Results
According to the analysis of variance (Table 1) the genotype had the most significant
effect on disease severity of black leaf streak disease. There was no interaction
between the cultivation system and the genotypes nor isolated effects of the
cultivation system on the severity of the disease.
Table 1. Analysis of variance of disease severity of black leaf streak disease in seven genotypes grown under two
cultivation systems.
Source of Variation
DF
Average Square
Block
Genotypes (G)
Cultivation system (CS)
Interaction (G X CS)
Error
4
6
1
6
52
51.2714
1356.9333*
66.0571ns
10.3238ns
24.8868
VC (%)
15.30
*Statistically significant at probability 0.01.
As presented in Figure 1, disease severity was highest on ‘SH-3640’ (55,10%),
followed by ‘Thap maeo’ (39%) and ‘FHIA-21’ (33,3%). The genotypes ‘Caipira’,
‘FHIA-01’ and ‘FHIA-02’ were similar whereas the hybrid ‘PV-4285’ presented the
lowest severity (19,70%) (Figure 2).
55,1a
60
50
39b
Disease severity (%)
MyLsd 17x24
40
33,5c
30
25,8d
26,1d
FHIA-02
FHIA-01
29,3d
19,7e
20
10
0
PV-4285
Caipira
FHIA-21
Thap maeo SH 36-40
Figure 1. Average disease severity of black leaf streak disease in seven genotypes.
253
MyLsd 17x24
25/06/03
15:15
Page 254
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Figure 2.
‘PV 42-85’free of black leaf streak
disease.
Figure 3.
‘SH 36-40’ showing signs of black leaf
streak disease.
The cultivation system (weeding, shedding, trimming and fertilization) did not
influence the severity of black leaf streak disease in ‘PV-4285’, ‘Caipira’, ‘FHIA-01’,
‘FHIA-02’, ‘Thap maeo’ and ‘SH-3640’.
For the data recording shooting, it was observed that the genotypes ‘FHIA-01’,
‘FHIA-02’, ‘Caipira’, ‘FHIA-21’, ‘PV 42-85’ and ‘Thap maeo’ were more resistant to
black leaf streak disease than ‘SH 36-40’, which was the most susceptible to the
disease (Figure 3).
254
MyLsd 17x24
25/06/03
15:15
Page 255
Session 4
M. de J.B. Cavalcante et al.
Conclusion
The genotypes ‘FHIA-01’, ‘FHIA-02’, ‘Caipira’, ‘FHIA-21’, ‘PV 42-85’ and ‘Thap maeo’
presented resistance to black leaf streak disease. The genotype ‘SH 36-40’ is
susceptible to black leaf streak disease. There was no significant effect of the
cultivation system on the severity of black leaf streak disease.
References
Cavalcante M.J.B., T.M.S Gondim, Z.J.M. Cordeiro, A.P. Matos, J.L. Hessel and F.R.V. Sampaio.
1999. Ocorrência da sigatoka-negra em dez municípios do Estado do Acre. Rio Branco:
EMBRAPA-CPAF/AC. p.1-2. (EMBRAPA-CPAF/AC. Comunicado Técnico 107).
Cordeiro Z.J.M., A.P. Matos and S. de O Silva. 1998. La Sigatoka negra en Brasil. INFOMUSA
7(1):30-31.
Gauhl F. 1994. Epidemiology and ecology of black Sigatoka (Mycosphaerella fijiensis Morelet)
on plantain and banana (Musa spp) in Costa Rica, Central América. INIBAP, Montpellier,
120pp.
Pereira L.V., Z.J.M. Cordeiro, A. Figueira, R. H. Hinz and A.P. Matos. 1999. Doenças da
bananeira. Informe Agropecuário, Belo Horizonte 20(196):37-47.
Ritzinger C.H.S.P, R. Ritzinger, Z.J.M. Cordeiro and M.J.B. Cavalcante. 1999. Ocorrência de
sigatoka negra da bananeira em Rio Branco, AC, Brasil. Fitopatologia Brasileira, v.24
(Suplemento), p.450.
Stover R. H. 1971. A proposed international scale for estimating intensity of banana leaf spot
(Mycosphaerella musicola). Tropical Agriculture 48:185-196.
255
MyLsd 17x24
25/06/03
15:15
Page 257
Session 4
J.V. Escalant
The International Musa testing
programme (IMTP): a worldwide
programme to evaluate elite Musa
cultivars
J.V. Escalant
Abstract
The International Musa Testing Programme (IMTP) is a collaborative effort coordinated by INIBAP
to evaluate, in suitable sites worldwide, elite Musa cultivars produced by breeding programmes
and promising accessions from the INIBAP collection. Established in 1989, IMTP trials are
designed to be replicated anywhere in the world and aim to evaluate elite clones for resistance
and/or tolerance to black leaf streak disease, Sigatoka disease and Fusarium wilt. Phase II of
IMTP started in 1996 when 15 countries initiated their field plots using 9 elite clones from
Honduras, Taiwan, Brazil and Cuba. This presentation summarizes the results. The analysis of
the results indicates that disease development time is not a reliable parameter for evaluating
resistance levels and that the infection index is a more reliable parameter. Among the
cultivars tested, ‘FHIA-23’ and ‘SH-3436-9’ displayed a good level of resistance using ‘Pisang
Ceylan’ as a resistant reference. This conclusion is consistent with the youngest leaf spotted
score obtained in most countries. A correlation was found between the infection index at bunch
emergence and the average finger weight across sites and genotypes. So far, 23 research
institutes in 21 countries are participating in IMTP III.
Resumen - Programa Internacional de Evaluación de Musa (IMTP): un programa
mundial para evaluar las variedades elite de Musa
El Programa Internacional de Evaluación de Musa (IMTP) es un esfuerzo colaborativo
coordinado por INIBAP cuyo fin es evaluar las variedades elite de Musa en sitios apropiados
alrededor de todo el mundo. Los ensayos del IMTP son diseñados para poder replicarlos en
cualquier lugar del mundo. El IMTP primero se desarrolló para realizar evaluaciones detalladas
de material nuevo con el fin de obtener información sobre su resistencia o tolerancia a las
Sigatokas negra y amarilla y al marchitamiento por Fusarium. La fase II del IMTP empezó en
1996 cuando 15 países diferentes establecieron sus parcelas en el campo utilizando
INIBAP, Montpellier, France
257
MyLsd 17x24
25/06/03
15:15
Page 258
Mycosphaerella leaf spot diseases of bananas: present status and outlook
9 diferentes cultivares elite de Honduras, Taiwán, Brasil y Cuba. En esta presentación se brinda
una apreciación global de los resultados. Un análisis de los resultados obtenidos indicó que
el tiempo de desarrollo de esta enfermedad no es un parámetro confiable para evaluar los
niveles de resistencia. Inversamente, los índices de infección parecen ser un parámetro más
confiable. Dentro del material evaluado, ‘FHIA-23’ y ‘SH-3436-9’ mostraron un buen nivel de
resistencia en comparación con la referencia resistente ‘Pisang Ceylan’. Esta conclusión es
consistente con el conteo de la hoja más joven manchada obtenido en la mayoría de los países.
Se observó una buena correlación entre el índice de infección durante la emergencia de
racimos y el peso promedio del dedo por el sitio y por el genotipo. Actualmente, 23 institutos
de investigaciones en 21 países están participando en la fase III del IMTP.
Résumé – Le programme international d’évaluation des Musa (IMTP) : un programme
international pour évaluer les cultivars d’élite Musa
Le programme international d’évaluation des Musa (IMTP) est un effort de collaboration
coordonné par l’INIBAP pour évaluer, dans des sites appropriés du monde entier, des cultivars de
bananiers produits par les programmes d’amélioration et des accessions de la collection de l’INIBAP.
Etablis en 1989, les essais IMTP sont conçus pour être répliqués n’importe où à travers le monde
et vise à évaluer les clones d’élites pour leur résistance et/ou tolérance à la maladie des raies noires,
la maladie de Sigatoka et la fusariose. La phase II de l’IMTP a débuté en 1996 lorsque 15 pays ont
mis en place leurs parcelles expérimentales pour 9 clones d’élites provenant du Honduras,Taïwan,
Brésil et Cuba. Cette présentation résume les résultats. L’analyse des résultats indique que le temps
de développement de la maladie n’est pas un paramètre fiable pour évaluer les niveaux de
résistance et que l’indice d’infection est plus fiable. Parmi les cultivars testés, ‘FHIA-23’ et ‘SH3436-9’ ont démontré un bon niveau de résistance par comparaison au témoin ‘Pisang Ceylan’.
Cette conclusion est consistante avec le score obtenu pour la plus jeune feuille nécrosé obtenu
dans la plupart des pays. Une corrélation a été observée entre l’indice d’infection à l’émergence
du régime et le poids moyen des doigts pour tous les sites et génotypes. Jusqu’à maintenant, 23
instituts de recherche dans 21 pays participent à la phase III de l’IMTP.
Introduction
The International Musa Testing Programme (IMTP) is a collaborative effort
coordinated by INIBAP to evaluate, in suitable sites worldwide, elite Musa cultivars
produced by breeding programmes and promising accessions from the INIBAP
collection. Taking into account local conditions, IMTP trials are designed to be
replicated anywhere in the world. The programme was developed to evaluate new
germplasm in order to obtain information on their resistance/tolerance to black leaf
streak disease and Sigatoka disease, caused respectively by Mycosphaerella fijiensis
and M. musicola, and to Fusarium wilt, caused by Fusarium oxysporum fsp. cubense.
IMTP trials can also be used to conduct basic research on the pathogen and its host,
such as epidemiological studies, host-pathogen relationships of the different strains
of a pathogen, and adaptability and productivity studies.
Two protocols have been developed in response to the demand from national
programmes to evaluate germplasm under local conditions while recognizing
the need for more detailed research at a limited number of sites. The two types of
evaluation are:
1) performance evaluations which use a simplified protocol to obtain data on cultivar
or hybrid performance under local conditions and basic data on disease resistance
or tolerance;
258
MyLsd 17x24
25/06/03
15:15
Page 259
Session 4
J.V. Escalant
2) and in-depth evaluations which are more complete disease resistance evaluations
carried out at a smaller number of sites. These sites are used to screen new
improved hybrids and, if requested by breeding programmes, parental breeding
lines. They can also provide opportunities for basic research on host-pathogen
interactions. A standard procedure for data management and statistical analysis
has been developed and the guidelines, which have been revised in the wake of
phase II of the programme, are made available in English, French or Spanish to
the participating programmes.
Phase I
The programme was established in 1989 as a partnership between National
Agricultural Research Systems (NARS), INIBAP breeders and pathologists from several
institutes. The objective was to use multilocation trials to identify banana and plantain
hybrids meeting local requirements which small-scale farmers could use to replace
susceptible cultivars. A second objective of IMTP was to stimulate Musa breeding
programmes by providing information on the response of their improved cultivars
to pathogens. A indirect effect of IMTP has been to increase the capacity of national
organisations to carry out research on banana and plantain.
The programme began by evaluating germplasm from the Fundación Hondureña
de Investigación Agrícola (FHIA) for resistance to black leaf streak disease. Seven
tetraploid hybrids from a wide variety of genetic backgrounds were tested along with
several reference diploid clones (wild and edible) that represented the whole range
of reactions to black leaf streak disease, i.e. from highly resistant to highly
susceptible. The trials were conducted in six countries. Site managers were trained,
and provided with technical guidelines and funding to carry out the trials. Four years
later, the detailed results were published and three hybrids were recommended for
distribution: ‘FHIA-01’ and ‘FHIA-02’, two dessert banana cultivars with outstanding
performance and high resistance to black leaf streak disease, and ‘FHIA-03’, a cooking
banana also with excellent performance and resistance to black leaf streak disease.
Over the last ten years these clones have been distributed to more than 50 countries.
In view of the success of the programme, INIBAP was asked to develop the programme
further.
Phase II
In IMTP II, germplasm was assessed for resistance to M. fijiensis and M. musicola.
Four breeding programmes contributed germplasm (Table 1) and the number of test
sites was increased to 37, even though the trials were financed by the participating
institutes. The majority of IMTP II trials were set up in 1996-1997. The complete
report includes results on resistance to black leaf streak disease from sites in
Cameroon, Costa Rica, Honduras, Nigeria, the Philippines,Tonga and Uganda and to
Sigatoka disease from one site in Colombia. A complete analysis was not possible
because of missing data, for reasons that include natural catastrophes, in places like
Cameroon, Costa Rica, Tonga, Thailand, Cuba and India. This presentation summarises
the IMTP II results.
259
25/06/03
15:15
Page 260
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Table 1. Improved cultivars included in the resistance trials of IMTP II.
Clone
Origin1
Genome
PV-03.44
PA-03.22
SH-3436-9
FHIA-01
FHIA-03
FHIA-17
FHIA-23
GCTCV-119
GCTCV-215
EMBRAPA
EMBRAPA
INIVIT
FHIA
FHIA
FHIA
FHIA
TBRI
TBRI
AAAB
AAAB
AAAA
AAAB
AABB
AAAA
AAAA
AAA
AAA
1Breeding
programme
EMBRAPA: Empresa Brasileira de Pesquisa Agropecuaria, Brazil; FHIA: Fundación Hondureña de Investigación Agrícola, Honduras;
INIVIT: Instituto Nacional de Investigación en Viandas Tropicales, Cuba; TBRI: Taiwan Banana Research Institute, Taiwan.
Resistance to black leaf streak disease
The data presented come from the trials in Costa Rica, Cameroon (Figure 1) and Tonga
which are considered as representative of Latin America, Africa and Pacific Asia
respectively. The response of the clones to black leaf streak disease varied according
to the biotic and abiotic factors present in each country.
40
12
Infection Index
YLS
35
10
30
20
6
15
YLS
8
25
Infection Index
4
10
2
Local cultivar
PV 02-44
0
PA 03-22
Pisang Berlin
SH3436-9
FHIA 23
Pisang Ceylan
Yangambi Km5
0
Saba
5
Calcutta 4
MyLsd 17x24
Figure 1. Infection index of black leaf streak disease and youngest leaf spotted (YLS) of different genotypes
in Cameroon.
260
MyLsd 17x24
25/06/03
15:15
Page 261
Session 4
J.V. Escalant
‘FHIA-23’ and ‘SH-3436-9’ from Honduras, had an average infection index similar
to that of the resistant reference cultivar ‘Pisang Ceylan’ suggesting that ‘FHIA-23’
and ‘SH-3436-9’ are resistant to black leaf streak disease (Table 2). The conclusion
is consistent with the youngest leaf spotted (YLS) score obtained in most countries.
Table 2. Infection index of black leaf streak disease and youngest leaf spotted (YLS) of various genotypes in Costa
Rica and Tonga.
Costa Rica
Clone
Tonga
Infection Index
YLS
36.1
34.4
24.6
18.4
8.3
8.2
7.4
7.8
PV-03.44
PA-03.22
SH-3436-9
FHIA-23
Clone
PV-03.44
PA-03.22
SH-3436-9
FHIA-23
References
Calcutta 4
Yangambi Km5
Pisang Ceylan
Saba
Local cultivar
Infection Index
YLS
29.9
34.8
13.5
61
7
7
11
11
1.8
20
19
20.3
30
11
10
9
5
References
10.6
19.1
18.7
32.3
45.6
7.9
7.9
8.5
8
4.9
Calcutta 4
Yangambi Km5
Pisang Ceylan
Saba
Local cultivar
Agronomic performance
Bunches of ‘FHIA-23’ and ‘SH 3436-9’ weighed on average 30.6 kg and 22.3 kg,
with a maximum of 39.4 kg in Cameroon and 28.8 kg in Tonga. Although, the local
cultivars differed between countries, their average bunch weight was 16.5 kg, with
a maximum bunch weight of 22.8 kg in Tonga (Figure 2). This substantiates the
improved performance of FHIA hybrids. However, FHIA and INIVIT hybrids had
longer growth cycles than local reference cultivars with an average of 474 and 420
days for ‘FHIA-23’ and ‘SH-3436-9’, respectively.
Discussion
Disease development time (DDT) was not a reliable measure of resistance possibly
because of difficulties in interpreting leaf symptoms under certain conditions. For
example, when disease pressure is high, stage 1 lesions may coalesce due to their
high number and appear similar to a stage 6 necrotic lesion. The infection index
seems to be a more reliable parameter. Besides being comparable between countries,
the results can be used to classify the new hybrids.
The same clone in different country can display a different tolerance to black
leaf streak disease. Tolerance being influenced by many factors, e.g. management,
soil fertility, pathogen pressure, presence of other pathogens and climatic conditions,
it is not possible to generalize the results. The effect of these factors on yield is not
easy to demonstrate or quantify. However, work at the Centre africain de recherches
sur bananiers et plantains (CARBAP) in Cameroon has demonstrated an effect of
261
25/06/03
15:15
Page 262
Mycosphaerella leaf spot diseases of bananas: present status and outlook
black leaf streak disease on bunch weight. A correlation between the infection index
at bunch emergence and fruit weight averaged across sites and genotypes (r=-0.71)
was found using IMTP data (Figure 3). The correlation suggests that the numbers of
characteristics to record could be reduced, thus simplifying data collection and
management. It would also reduce the need for visual interpretation of symptoms
in the field.
40
35
30
25
Weight (Kg)
MyLsd 17x24
20
15
10
5
0
Bago Oshiro
Cameroon
Costa Rica
Honduras
Tonga
Uganda
Country
FHIA-23
SH-3436-9
PV-03.44
PA-0.33-22
local
Figure 2. Average bunch weight of various hybrids in different locations.
Unexpectedly, in some sites, the highly resistant clones ‘Calcutta 4’ and ‘Yangambi
km5’ had stage 6 necrotic lesions. Further investigations are needed to determine
whether the effect was due to aggressive strains of M. fijiensis or to a new pathogen.
Conclusion
FHIA hybrids had consistently the best yields in trials. With few exceptions, their
bunches were heavier than those from other improved and local cultivars. FHIA
hybrids also responded well to careful management and to fertilizer. In summary,
FHIA hybrids performed well under a range of conditions; the better the conditions
the better their performance. The improved hybrid ‘FHIA-23’ had the best performance
in all the trials to evaluate resistance to Sigatoka diseases.
262
MyLsd 17x24
25/06/03
15:15
Page 263
Session 4
J.V. Escalant
Across clones
250
Finger Weight (g)
200
R2 = 0,5064
150
100
50
0
0
10
20
30
40
50
60
70
Infection index
Figure 3. Correlation between the infection index at bunch emergence and fruit weight averaged across sites
and genotypes.
Cultivar ‘GCT CV 119’ deserves special attention as it had the lowest discoloration
scores for two races of F. oxysporum f.sp. cubense and high bunch weight under
conditions of good crop husbandry. It is important to stress that resistance alone is
not useful. It needs to be combined with good production, and acceptable post-harvest
and organoleptic traits. Improved banana varieties contribute not only to reducing
disease incidence but also to improving food production. The complete data and
statistical analysis will be published shortly1.
IMTP II database
All the information from IMTP II, including agronomic trains and host plant
response, for all genotypes and all sites, is compiled in a database to facilitate
access to the data on new Musa germplasm compiled throughout the world. The
database is also included in the CD-ROM, ‘Evaluation of Musa germplasm for
resistance to Sigatoka diseases and Fusarium wilt’. The CD-ROM also contains
the technical guidelines, a comprehensive analysis of phase II results entitled
‘Evaluating bananas: a global partnership’, a catalogue of candidate and reference
clones, including those for IMTP III trials, and the transfer agreement to obtain
genetic material. The CD-ROM contains all the information needed to participate
in IMTP III.
IMTP III
In 2001, 450 consignments of germplasm accessions were sent from the INIBAP
Transit Centre (ITC) to 23 institutes in 21 countries participating in IMTP III. Thirtyfive cultivars including plantains, cooking bananas and dessert bananas are available
1
To receive the final report please contact Jean-Vincent Escalant, the IMTP coordinator at INIBAP.
263
MyLsd 17x24
25/06/03
15:15
Page 264
Mycosphaerella leaf spot diseases of bananas: present status and outlook
for evaluation. Eleven institutes are carrying out “in-depth” studies to evaluate
improved varieties, which will involve epidemiological and ecological research.
Thirteen institutes will be evaluating the performance of varieties against local
diseases and conditions following a more simplified format of data gathering (Table
3). For the first time two private companies will be carrying out evaluations. The
first results are expected in 2003.
Table 3. Institutes and countries involved in IMTP III
Country
Institute
Phase III
Australia
Bangladesh
Burundi
Queensland Horticultural Institute (QHI)
Bangladesh Agricultural Research Institute (BARI)
Institut de recherches agronomique et zootechnique de la
communauté économique des pays des grands lacs (IRAZ)
Centre africain de recherches sur bananiers et plantains (CARBAP)
South China Agricultural University (SCAU)
Corporación Colombiana de Investigación Agropecuaria (CORPOICA)
Corporación Bananera Nacional (CORBANA)
Centro para el Desarrollo Agropecuario y Forestal (CEDAF)
Not confirmed
Performance
Performance
Performance
In-depth
Performance
In-depth
In-depth
Performance
Performance
In-depth
Philippines
Philippines
Rwanda
Sri Lanka
Uganda
Institut Interaméricain de Coopération pour l’agriculture (IICA)
Fundación Hondureña de Investigación Agrícola (FHIA)
National Research Center on Banana (NRCB)
Central Research Institute for Horticulture (CRIH)
Malaysian Agricultural Research and Development Institute (MARDI)
Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP)
Universidad de León (UNAN León)
Servicio Nacional de Sanidad Agraria (SENASA)
Bureau of Plant Industry – Davao National Crop Research
and Development Center (BPI – DNCRDC)
Dole Asia Research ; Stanfilco
Lapanday Fruit Company
Institut des sciences agronomiques du Rwanda (ISAR)
Agricultural Research Station (ARS)
National Agricultural Research Organization (NARO)
Vietnam
Vietnam Agricultural Science Institute (VASI)
Cameroon
China
Colombia
Costa Rica
Dominican
Republic
Haiti
Honduras
India
Indonesia
Malaysia
Mexico
Nicaragua
Peru
Philippines
Performance
In-depth
Performance
In-depth
In-depth
Performance
In-depth
In-depth
Performance
Performance
In-depth/
Performance
Performance
All except two institutes asked to evaluate leaf spot diseases demonstrating the
worldwide impact of leaf spot diseases on banana production. Black leaf streak disease
is the main leaf spot of banana in the world. Leaf spots are also caused by M. musicola
and M. eumusae, the latter recently discovered in Southeast Asia and easily confused
with M. fijiensis. In the framework of IMTP, a training course for IMTP III
participants was organized by INIBAP and CIRAD, and hosted by the Malaysian
Agricultural Research and Development Institute (MARDI). The aims were to
standardize methods of data collection and evaluation of leaf spot diseases and to
264
MyLsd 17x24
25/06/03
15:15
Page 265
Session 4
J.V. Escalant
train researchers to identify leaf spot diseases by morphological differences in the
anamorph stages of the fungi.
Data entry forms
265
MyLsd 17x24
25/06/03
15:15
Page 266
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Recommendations of session 4
Progress has been made towards the creation of new varieties resistant to black leaf streak
disease, either through conventional and/or modern technologies. New tetraploid hybrids
resistant to black leaf streak disease are already available and some of these are widely grown
around the world. Good progress has also been made in the development of a molecular
toolbox for bananas and plantains in the area of the genetic transformation.
Musa balbisiana genome
The presence of the activable form of the banana streak virus (BSV) in interspecific hybrids
(AxB), hinders the production of a new generation of triploid hybrids. Access to all the
balbisiana diversity is important because of the BSV related problem but also to get a better
knowledge of the existing diversity in the B genome and its contribution in the interspecific
hybrids.
It is recommended to study the diversity of the Musa balbisiana genome with both
morphological and molecular traits. It is also recommended to promote and facilitate
new collecting missions.
Breeding for resistance
All possible sources of resistance to pests and diseases are needed to genetically improve
Musa cultivars. Previous studies report on resistance to Mycosphaerella spp. of Musa
schizocarpa (genome S) and Musa textilis (genome T). Cultivars containing T and S genomes
have also been reported in Papua New Guinea as being highly resistant. Prospecting for new
sources of resistance to Mycosphaerella spp., as well as other diseases, should be easier since
all Musa species, except Musa textilis, are covered by the facilitated access provision in the
recently signed International Treaty on Plant Genetic Resources for Food and Agriculture
Treaty.
It is recommended to anticipate the needs of genetic improvement programmes by
screening the T and S genome species as well as subspecies of Musa acuminata and
other Musa spp. to detect new possible sources of resistance to pests and diseases.
Durability of resistance
The erosion of resistance is a problem which should be addressed to ensure the durability of
resistant improved cultivars.
The pathogen populations should be characterized in areas where the resistance of hybrids
appears to be decreasing.
266
MyLsd 17x24
25/06/03
15:15
Page 267
Session 4
Recommendations
Computer modelling of black leaf streak disease epidemics is a way to predict the durability
of resistance and to evaluate different disease management strategies. However, the
development of such a model requires more quantitative parameters to describe disease
epidemics and the evolution of the pathogen population in response to the selection pressure
exerted by resistant hosts.
It is recommended to study the structure of the population at field and regional level
in mixed-cultivars plots combining vertical and horizontal resistances.
Mutation induction
Mutation induction techniques should no longer be seen as an independent genetic
improvement strategy but more as a tool that can contribute to cross-breeding programmes
by increasing genetic diversity in parental lines. For example, the barley MLO gene that confers
complete resistance to powdery mildew was obtained by mutagenesis. Mutants could also
help in understanding the mechanism of resistance. Induced deletion mutants and aneuploids
in particular will be useful to map or locate genes of interest and molecular markers.
Genetic transformation
Triploid cultivars of banana are often pollen and seed sterile and as such they should benefit
from simpler risk assessment protocols regarding geneflow.
It is recommended to encourage the development of national legislation to allow field
testing of transgenic Musa plants, to collect data, to guide further research and regulation.
It is also recommended to continue the development of transgenic banana plants to allow
in-depth studies on plant development and on plant-pathogen interactions and to increase
resistance to Mycosphaerella.
Genetic transformation is also recommended to identify and isolate genes of resistance
using the ‘knockout’ strategy on resistant cultivars. This will also be very useful to study
host-pathogen interaction.
267
MyLsd 17x24
25/06/03
15:15
Page 269
Session 5
Integrated disease management
MyLsd 17x24
25/06/03
15:15
Page 271
Session 5
R. Peterson et al.
Management of Mycosphaerella
leaf spot diseases in Australia
R. Peterson1, K. Grice1 and S. De La Rue1
Abstract
Mycosphaerella musicola is the major leaf spot pathogen affecting banana in the tropics, whereas
Mycosphaerella musae and M. musicola dominate in the sub-tropical areas of Australia. Control
strategies include chemical and cultural practices. In tropical areas, sprays are applied at intervals
of 10-14 days during the wet season and 21-28 days during the dry season. Annually, 18-24 protectant
and systemic fungicide sprays are applied with petroleum oil (5L). Since the early 1980s, the banana
industry has fought to exclude this disease from the production areas by monitoring and creating
a barrier of resistant plant material. Black leaf streak disease has been detected in the Cape York
area eight times in the past 20 years and was limited to a few plants and in all cases was successfully
eradicated.
In April 2001, black leaf streak disease was identified in the main production area of North
Queensland. An intensive survey (April-June) indicated the disease was restricted to the Tully area.
Black leaf streak disease was found on 14 commercial properties and on 11 clumps of unmanaged
plants. An eradication programme commenced in September 2001 across approximately 4500 ha
of banana plants and involved deleafing commercial plantings to zero disease, weekly fungicide
applications for six months and the destruction of all non-managed plants. Black leaf streak disease
has not been detected on commercial plantations for eight months and on non-managed plants
for four months. It was not detected in 1550 samples assessed during January to April 2002.
Manejo de las Mycosphaerella en Australia
La Sigatoka amarilla (causada por Mycosphaerella musicola) es la principal enfermedad foliar que
afecta a los bananos en los trópicos, mientras que la mancha foliar (Mycosphaerella musae) y la
Sigatoka amarilla predominan en las áreas subtropicales de Australia. La estrategias de control
incluyen la aplicación de los químicos y las prácticas culturales. En las áreas tropicales, los rociados
se aplican a intervalos de 10-14 días durante la estación húmeda y de 21-28 días durante la estación
seca. Anualmente, se aplican 18-24 rociados de fungicidas protectores y sistémicos mezclados con
el aceite mineral (5L). Desde inicios de la década de los 80, la industria bananera ha luchado para
erradicar esta enfermedad en las áreas de producción monitoreando y creando un barrera de material
vegetal resistente. La Sigatoka negra fue detectada en el área de Cabo York ocho veces durante los
últimos 20 años y fue limitada a unas pocas plantas y todos los casos fueron erradicados
exitosamente.
1
Queensland Department of Primary Industries, Mareeba, Australia.
271
MyLsd 17x24
25/06/03
15:15
Page 272
Mycosphaerella leaf spot diseases of bananas: present status and outlook
En abril de 2001, la Sigatoka negra fue identificada en la principal área de producción de Queensland
del Norte. Una encuesta intensiva (abril-junio) indicó que la enfermedad estaba restringida al área
de Tully. La Sigatoka negra fue descubierta en 14 propiedades comerciales y en 11 grupos de plantas
sin manejar. En septiembre de 2001 empezó un programa de erradicación en aproximadamente
4500ha de bananos e incluyó el deshoje total de las siembras comerciales, aplicaciones semanales
de funguicidas durante seis meses y la destrucción de todas las plantas sin manejo. La Sigatoka
negra no fue detectada en las plantaciones comerciales durante ocho meses y en las plantas sin
manejo durante cuatro meses. Tampoco se detectó la Sigatoka negra en 1550 muestras evaluadas
durante enero-abril de 2002.
Résumé - Gestion des Mycosphaerella en Australie
Mycosphaerella musicola est le principal agent pathogène des maladies foliaires dans les tropiques,
alors que Mycosphaerella musae et Mycosphaerella musicola dominent dans les régions subtropicales d’Australie. Les stratégies de contrôle incluent des pratiques chimiques et culturales. Dans
les zones tropicales, les pulvérisations sont appliquées à des intervalles de 10-14 jours pendant la
saison humide et de 21-28 jours pendant la saison sèche. Sur une année, 18-24 pulvérisations de
fongicides protecteurs et systémiques sont appliquées avec de l’huile de pétrole (5L). Depuis le début
des années 1980, l’industrie bananière a lutté pour exclure cette maladie des zones de production
par une surveillance et la création d’une zone tampon de matériel résistant. La maladie des raies
noires a été détectée huit fois au cours des 20 dernières années dans la zone de Cape York ; elle
était limitée à quelques plantes et, chaque fois, elle a été éradiquée avec succès.
En avril 2001, la maladie des raies noires a été identifiée dans la principale zone de production du
North Queensland. Un inventaire intensif (avril-juin) a indiqué que la maladie était restreinte à la
région de Tully. La maladie des raies noires a été trouvée dans 14 plantations commerciales et dans
11 groupes de plantes n’ayant subi aucun traitement. Un programme d’éradication a commencé en
septembre 2001 sur environ 4500 ha de bananeraies ; il comprenait la suppression des feuilles dans
les plantations commerciales jusqu’au niveau zéro de la maladie, des applications hebdomadaires
de fongicides pendant 6 mois et la destruction de toutes les plantes non traitées. La maladie des
raies noires n’a pas été détectée dans les plantations commerciales pendant huit mois et pendant
quatre mois sur les plantes non traitées. Elle n’a pas été détectée dans les 1550 échantillons analysés
entre janvier et avril 2002.
Introduction
Mycosphaerella musicola, the cause of Sigatoka disease, was first recorded in Australia
in 1924 (Benson, 1925) and is still the predominant leaf disease of banana in tropical
Australia. In the sub-tropical areas of south Queensland and New South Wales, the
major leaf disease is Mycosphaerella speckle (caused by Mycosphaerella musae) followed
by Sigatoka disease. In the Pacific region Sigatoka disease was reported from the
Sigatoka Valley in Fiji in 1912 (Philpott and Knowles, 1913). Black leaf streak disease,
caused by Mycosphaerella fijiensis, was recorded later, in 1963 (Rhodes, 1964), but
has since become the dominant leaf disease in the region.
The integrated approach used to control Mycosphaerella pathogens involves the
use of cultural practices and fungicides. In wet tropical areas, fungicide sprays are
applied at 10-14 day intervals during the wet season (December–April) and 14-28 days
during the remainder of the year, for a total of 18-24 applications per year. Mancozeb
is the main protectant fungicide used, and the application rates and length of
withholding periods vary depending on the formulation used. Systemic fungicides are
used predominately during the wet season when conditions are more conducive to
272
MyLsd 17x24
25/06/03
15:15
Page 273
Session 5
R. Peterson et al.
disease development. Current registrations include propiconazole (Tilt®, Bumper® and
Aurora®), tebuconazole (Folicur®) and benomyl (Benlate®). Cultural practices used to
minimize the risk of disease development include good drainage and air circulation,
location (not adjacent to a permanent body of water, buildings or rainforest) and most
importantly inoculum reduction (deleafing).
Loss of sensitivity to fungicides
Sensitivity of M. musicola populations to systemic fungicides was assessed using a
technique based on the Fungicide Resistance Action Committee (FRAC) guidelines, but
using conidia instead of ascospores. Conidia are produced at an earlier stage of disease
development than ascospores and are readily available during the winter and spring
months. Ascospores are produced in summer and autumn and are not readily available
during the remainder of the year due to phytosanitary regulations and deleafing practices.
For the triazole fungicides (propiconazole and tebuconazole) germtube growth after
72 hours was expressed on a sensitivity graph (germtube growth at 4 fungicide
concentrations in relation to growth in the absence of the fungicide). For benomyl,
percent germination was recorded after 48 hours. Sensitivity graphs were used to establish
the effective concentration required to give a 50% reduction in germtube growth (EC50).
Baseline data was established for each fungicide by averaging the EC50 of more
than 20 populations of M. musicola. These were selected from unsprayed plants at least
25 km away from the nearest commercial/sprayed block of bananas. The sensitivity of
populations from fungicide-treated plots was assessed by comparing the graph and the
EC50 with the baseline average. An 8 to 16-fold increase in the EC50 is considered a
moderate shift while a more than 16-fold increase is a severe shift in sensitivity.
A shift in sensitivity was first detected with propiconazole in 1995. In limited surveys
over the last 2 to 3 years, moderate to severe shifts in sensitivity to triazole fungicides
were detected:
• Shifts in sensitivity to tebuconazole were detected in populations that had not been
sprayed with tebuconazole, but with propiconazole;
• Cross resistance from propiconazole to tebuconazole is 100%, but the reverse is
variable;
• Shifts in sensitivity to triazoles occurred in isolated plantations and was linked to
excessive use (8-12 applications), the number of consecutive applications (3-5) and
applications to heavily diseased plants (no deleafing);
• Shifts in sensitivity were reversed when the product or similar products were withheld
for 6-12 months;
• Resistance to benomyl was high across most populations.
Strategies to minimise the risk of resistance to fungicides based on FRAC guidelines
were developed and implemented in 1996 in consultation with growers and government
representatives. Some of the cultural and chemical related strategies included:
• Regular deleafing to remove heavily infested leaf material prior to the application
of a systemic fungicide;
273
MyLsd 17x24
25/06/03
15:15
Page 274
Mycosphaerella leaf spot diseases of bananas: present status and outlook
• A maximum of 6 triazole fungicides to be used in any one season;
• A maximum of 2 consecutive applications of triazoles;
• A 4-month triazole-free period from July to October.
Black leaf streak disease
Black leaf streak disease was first identified in Australia in 1981 at Bamaga on the tip
of the Cape York Peninsula and throughout the Torres Strait area. It has since been
found at 7 other locations in the sparsely populated Cape York area. Areas affected
have ranged from only a few plants to an entire residential area (Weipa, population
3000), and to a 36-ha commercial plantation of organically grown bananas. In all cases
the outbreaks were eradicated by destroying all plant tissue at the site of the infection
and within a surrounding buffer area. Plants were destroyed by burying, ploughing
or burning. In most cases the destroyed plants were replaced with resistant/tolerant
banana cultivars. Regular intense surveys have failed to detect black leaf streak disease
at any of these sites except at Bamaga. Bamaga was the first attempt at eradication
and the program has been refined considerably over the past 15-20 years.
In April 2001, M. fijiensis was detected in the Tully area where approximately 55%
of northern Queensland bananas are produced. An intensive survey indicated the disease
was restricted to the Tully Valley area and was only a recent introduction (6-12 months).
The identification of the organism was initially complicated by the lack of conidia
and sporodochia due to heavy rainfall in the area. Drying and wetting of leaf material
failed to produce the sporulating structures required for microscopic identification. All
samples were therefore visually assessed and suspicious samples were analysed by using
the PCR (Polymerase Chain Reaction) method. Dr Juliane Henderson from the
Cooperative Research Centre for Tropical Plant Protection in Brisbane modified and
refined the PCR protocol based on methods published by the Natural Resource Institute
(Johanson, 1997). In the 12 months to 31st March 2002, more than 8600 banana leaf
samples have been assessed and 2979 samples have been PCR tested. Black leaf streak
disease has been positively identified on only 13 commercial properties and on 12
clumps of unmanaged banana plants.
An eradication programme was started in September 2001 and was based on:
• M. fijiensis having no dormant structures;
• M. fijiensis only affects Musa species. There are no alternative hosts;
• M. fijiensis survives in leaf tissue:
– >20 weeks in leaf in canopy (Gauhl, 1994);
– 4-8 weeks in leaf tissue in contact with the ground (Peterson et al., 1998);
• Deleafing and placing leaves in piles reduces by about 80% the potential of inoculum
production.
Programme consisted of three components:
• Maintaining all commercial banana plantations (4500 ha) at zero visible disease levels
for 6-8 months to ensure all inoculum is destroyed;
274
MyLsd 17x24
25/06/03
15:15
Page 275
Session 5
R. Peterson et al.
• Weekly spray programme (systemic and protectant) for at least 6 months to prevent
any ascospore release from plant remains establishing new infections;
• Eradication/destruction of all non-managed plants (not sprayed or deleafed) in the
entire area.
Results
In commercial plantations:
• All blocks were monitored (every 2nd row) at 4 to 6-week intervals;
• >95% of the area have had zero visible disease since December 2001;
• Over 8600 samples have been collected and examined:
– No black leaf streak disease has been recorded since August 2001.
In un-managed plants:
• > 28 000 stems and > 22 000 suckers have been destroyed;
• No black leaf streak disease has been recorded since November 2001;
• All land plots (8500) have been visited and revisited between January to March
2002, and 6382 stems and 6610 suckers have been destroyed;
• No black leaf streak disease has been found in the 6382 stems.
Spray programme:
• 13 systemic and 14 protectant fungicide sprays have been applied from August
2001 to February 2002;
• Oil caused damage when applied in hot dry conditions to plants that were waterstressed;
• Trifloxystrobin (temporary registration) caused considerable damage when applied
in hot conditions. Damage was reduced when applied early morning or late
afternoon.
The success of this programme can be attributed to the full participation of growers
in combination with the unseasonable dry weather conditions experienced between
August 2001 and April 2002. Conditions during this period were not conducive to
the development of Sigatoka diseases.
Future plan – establish disease-free areas
Sentinel plant network:
• >130 sites of 8-10 plants have been established throughout the area on a 1-km
grid near sites where black leaf streak disease has been confirmed. Spacing was
increased to a 10-km grid when >15 km from a known black leaf streak disease
site. All plants are inspected and sampled for disease identification at 4-week
intervals.
275
MyLsd 17x24
25/06/03
15:15
Page 276
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Commercial plantations:
• Monitoring of commercial plantations is to be conducted at 6 to 8-week intervals
and all diseased leaves are to be sampled;
• Sites where un-managed plants have been destroyed are to be inspected for
regrowth. All diseased tissue is to be collected for identification;
• A period of 12 months, including an average wet season, without an outbreak
should demonstrate an area free of black leaf streak disease.
References
Benson A.H. 1925. Leaf spot disease of bananas. Qld Agric. J. 24:392-393.
Gauhl F. 1994. Epidemiology and ecology of black Sigatoka (Mycosphaerella fijiensis Morelet)
on plantain and banana (Musa spp.) in Costa Rica, Central America. INIBAP, Montpellier,
France, 120pp.
Johanson A. 1997. Detection of Sigatoka leaf spot pathogens of banana by the Polymerase chain
reaction. Natural Resource Institute, Chatham, UK.
Peterson R.A., K.R.E. Grice and A. Wunsch (eds). 1998. Survival of M. musicola in leaf tissue.
Report, Department of Primary Industries, Mareeba, Australia.
Philpot J. and C.H. Knowles. 1913. Report on a visit to Sigatoka. Pamphlet Dep. Agric. Fiji 3.
Rhodes P.L. 1964. A new banana disease in Fiji. Commonw. Phytopath. News 10:38-41.
276
MyLsd 17x24
25/06/03
15:15
Page 277
Session 5
P.E. Jorge and T. Polanco
Spread and management
of black leaf streak disease
in the Dominican Republic
P.E. Jorge and T. Polanco
Abstract
In the Dominican Republic, black leaf streak disease, caused by Mycosphaerella fijiensis, was
first identified in 1996 in the province of Montecristi, on the Northwest side of the country,
a region with prevalent dry conditions. It was then identified in Hato Mayor on the
southeast side on July, 1998. The disease appeared in 1999 in the Provinces of Sánchez
Ramírez, Samaná, Dajabón, Santiago Rodríguez and Monte Plata, located on the Northwest
and East sides, suggesting spread to these areas from the first two identified regions. In
the year 2000, it was first identified in the Southwest, in the Provinces of Azua and San Juan
de la Maguana. Today, black leaf streak disease continues to spread, moving to the central
portion of the country, the largest plantain growing area, where favorable environmental
conditions are common.
Previous to the appearance of black leaf streak disease, measures regarding management
of Sigatoka disease were not very intensive. After the appearance of black leaf streak disease
in 1996 growers have changed their practices to compensate for the presence of the disease.
Currently, the management methods adopted by growers include deleafing, application of
fungicides, fertilization and, to a lesser extent, planting tolerant-resistant materials, mostly
FHIA hybrids. Deleafing and application of fungicides are done on a routine basis, determined
by climatic conditions. Deleafing is usually done weekly during the rainy seasons and monthly
otherwise. A similar pattern is observed for the application of fungicides. Fertilization is
implemented to assure a prompt recovery from the stress induced by the disease. Few
growers rely on climatic/biological disease forecasting strategies.
Resumen - Propagación y manejo de la Sigatoka negra en República Dominicana
En República Dominicana, la Sigatoka negra, causada por Mycosphaerella fijiensis, fue
identificada por primera vez en 1996 en la provincia de Montecristi, en el noroeste del país,
una región con condiciones secas prevalecientes. Se identificó en Hato Mayor en la parte
sudeste en julio de 1998. En 1999, la enfermedad apareció en las provincias de Sánchez
Instituto Dominicano de Investigaciones Agropecuarias y Forestales, Santo Domingo, Dominican Republic
277
MyLsd 17x24
25/06/03
15:15
Page 278
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Ramírez, Samaná, Dajabón, Santiago Rodríguez y Monte Plata, localizadas en el noroeste y
oriente, sugiriendo su propagación a estas áreas de las dos regiones identificadas
inicialmente. En el año 2000, la enfermedad fue identificada por primera vez en el sudoeste,
en las provincias de Azua y San Juan de la Maguana. Actualmente, la Sigatoka negra continua
propagándose, moviéndose a la parte central del país, la principal área bajo el cultivo de
plátano, donde existen condiciones ambientales favorables.
Previas a la aparición de la Sigatoka negra, las medidas con respecto al manejo de la Sigatoka
amarilla no fueron muy intensivas. Después de la aparición de la Sigatoka negra en 1996
los productores han cambiado sus prácticas de cultivo para compensar la presencia de la
enfermedad. Actualmente, los paquetes de manejo de la enfermedad adoptados por los
productores consisten en el deshoje, aplicación de fungicidas, fertilización y, en un menor
grado, siembra de materiales tolerantes o resistentes, principalmente los híbridos de la FHIA.
El deshoje y la aplicación de funguicidas son actividades que se realizan sobre una base
habitual determinada por las condiciones climatológicas. Los deshojes se realizan
aproximadamente cada semana durante las estaciones lluviosas, y mensualmente durante
las estaciones secas. Un comportamiento similar se observa para la aplicación de los
fungicidas. La fertilización se implementa para asegurar una rápida respuesta y recuperación
del estrés inducido por la enfermedad. Pocos productores confían en estrategias de preaviso
biológico y climatológico de la enfermedad.
Résumé - Propagation et gestion de la maladie des raies noires en République
Dominicaine
En République Dominicaine, la maladie des raies noires, causée par Mycosphaerella fijiensis,
a été identifiée pour la première fois en 1996 dans la province de Montecristi, dans la partie
nord-ouest du pays, une région où les conditions de sécheresse prévalent. Elle a été ensuite
identifiée à Hato Mayor sur la côte sud-est en juillet 1998. La maladie est apparue en 1999
dans les provinces de Sánchez Ramírez, Samaná, Dajabón, Santiago Rodríguez et Monte Plata,
situées dans les parties nord-ouest et est, ce qui suggère une propagation vers ces zones
depuis les deux premières régions identifiées. En 2000, la maladie a d’abord été identifiée
dans le sud-ouest, dans les provinces d’Azua et San Juan de la Maguana. Aujourd’hui, la
maladie des raies noires continue de s’étendre, en se dirigeant vers la partie centrale du pays,
la plus grande zone de culture des bananes plantain, dans laquelle des conditions
d’environnement favorables sont fréquemment rencontrées.
Avant l’apparition de la maladie des raies noires, les mesures visant à contrôler la maladie
de Sigatoka n’étaient pas très intensives. Après l’apparition de la maladie des raies noires
en 1996, les planteurs ont modifié leurs pratiques pour compenser la présence de la
maladie. Aujourd’hui, les méthodes de gestion adoptées par les planteurs incluent
l’effeuillage, l’application de fongicides l’application d’engrais et, dans une moindre mesure,
l’utilisation de plants résistants/tolérants, principalement des hybrides de la FHIA. L’effeuillage
et l’application de fongicides sont réalisés de manière routinière, en fonction des conditions
climatiques. L’effeuillage est effectué de façon hebdomadaire pendant la saison des pluies,
et mensuellement le reste du temps. La même périodicité est utilisée pour l’application des
fongicides. La fourniture d’engrais est réalisée afin d’assurer une récupération rapide du stress
induit par la maladie. Seul un petit nombre de planteurs ont recours à des stratégies de
prévision du développement de la maladie basées sur des facteurs climatiques/biologiques.
Introduction
The Dominican Republic occupies the eastern two thirds of the Caribbean island of
Hispaniola and has an area of 48 422 km2. The Republic of Haiti occupies the
remainder of the island. Both countries are divided by the Cordillera Mountains which
278
MyLsd 17x24
25/06/03
15:15
Page 279
Session 5
P.E. Jorge and T. Polanco
reach 3175 meters and separate the southern and northern parts of the island. The
mountains are important barriers for the natural movement of inoculum of black
leaf streak disease (teleomorph: Mycosphaerella fijiensis Morelet; anamorph:
Paracercospora fijiensis (Morelet) Deighton), and the movement of plant material
between the two regions. Banana and plantain production is mainly in the
southwestern and northern parts of the country. Thus, production of Musa species
is isolated and growers from either side of the mountains do not exchange planting
material and there is little commercial exchange of fruits.
The country is divided into eight regions by the Ministry of Agriculture and Musa
production occurs in all regions. Banana production is mainly in the Southwest and
the Northwest and plantain production is mainly in the Cibao Central, including the
Northeast, North and North Central regions, where a large proportion of the
population lives. Plantain is the third most important vegetable crop, after rice and
beans. Plantain is consumed within the country, whereas a large part of the bananas,
whether grown by using conventional or organic methods, are exported.
The average rainfall from 1961 to 1990 (Secretaría de Estado de Agricultura, 1998)
from the different regions are shown on Table 1. Full expression of black leaf streak
disease in the Dominican Republic is dependent on rainfall and humidity, as
temperatures in banana and plantain areas do not limit the development of the
disease. Figure 1 shows symptoms of the disease in dry and humid regions. Generally,
the disease is limited to the first stages in dry regions and fully expressed in humid
regions.
Humid regions
Dry regions
Figure 1. Symptoms of black leaf streak disease in the dry and humid regions of the Dominican Republic.
Table 1. Average rainfall in the Dominican Republic by region for the period between 1961 and 1990.
Region
Rainfall (mm/year)
North central
Northeast
North
East
Central
Northwest
South
Southwest
1200-1350
1200-1350
1200-1350
1342-1583
1400-1800
< 600
< 600
665
Source: Secretaría de Estado de Agricultura, 1998. Anuario estadistico de planificación sectorial agropecuaria.
279
MyLsd 17x24
25/06/03
15:15
Page 280
Mycosphaerella leaf spot diseases of bananas: present status and outlook
The distribution of plantain and banana crops is shown in Table 2 (Secretaría de
Estado de Agricultura, 1998). More information on farm size and number of farmers
can be found in Perez (2000), Departamento de Sanidad Vegetal (2000) and
Secretaría de Estado de Agricultura (2000).
Table 2. Area planted with plantain and banana in the Dominican Republic.
Region
South
Southwest
Central
East
North Central
Northeast
North
Northwest
Total
Plantain (ha)
Banana (ha)
12 500*
5764
3750
625b
14 333
3139
9000
2876
51 987
3105
5722
2062
4280
3585
6201
24 955
Source: Secretaría de Estado de Agricultura, 1998
* Personal communication, Secretaría de Estado de Agricultura.
Pathogen identification
The spread of black leaf streak disease in the Americas, and the appearance of the
disease in Cuba in 1990 (Vidal, 1992) and in Jamaica in 1994-1995 alerted the
Dominican authorities. In 1995, the Department of Plant Protection, Secretaría de
Estado de Agricultura de la República Dominicana, started a project to monitor the
disease, in collaboration with Haiti’s phytosanitary authorities and the Animal and
Plant Health Inspection Service of the United States Department of Agriculture
(APHIS/USDA).
Black leaf streak disease was first recorded in 1996 in Guayubín, province of
Montecristi, at a farm of 125 ha planted with banana and plantain. Distribution within
the farm was limited mainly to plantain. At the time, M. fijiensis was identified by
the Plant Pathology Laboratory at CIRAD/AMIS, Montpellier, France. In May 1997,
the pathogen was also identified, with the assistance of Mary E. Palm from the United
States Department of Agriculture, as P. fijiensis based on the morphology of the
conidia and conidiogenous cells in her report to the Secretaría de Estado de
Agricultura. This was the first official diagnosis of the causal agent of black leaf
streak disease in the Dominican Republic. Also in 1997, Fouré surveyed the different
production areas to determine the distribution and incidence of black leaf streak
disease in the country (Fouré, 1997).
In 1999, T. Polanco, in collaboration with Jean Carlier and Marie Zapater from
CIRAD, isolated M. fijiensis from the northwestern, northeastern and central eastern
regions of the Dominican Republic (Polanco, 1999). Rivas et al. (2001) compared the
isolates with those from South America, Central America and the Caribbean. The
isolate from the Dominican Republic was found to be closely related to the Cuban
isolate.
280
MyLsd 17x24
25/06/03
15:15
Page 281
Session 5
P.E. Jorge and T. Polanco
Spread of black leaf streak disease
The spread of black leaf streak disease in the Dominican Republic up to May 2002
is shown in Figure 2.
Infected zones
Figure 2. Spread of black leaf streak disease in the Dominican Republic up to May 2002.
Northwestern region
The northwest region has approximately 6250 ha of banana and 2876 ha of plantain
(Table 2). Bananas are mostly grown for export to Europe, where there is a market
for traditional and organic fruits.
The northwestern region of the Dominican Republic where black leaf streak disease
was first identified in 1996, is mostly characterized by environmental conditions
that do not favour the development of the disease. The rainy-humid conditions that
favour disease development prevail for only short periods, usually in May and
November-December. There were less than 600 mm of rain/year for the period 19611990, usually distributed over 70 days per year (Secretaría de Estado de Agricultura,
1998). Thus, most of the region is considered to be low risk for black leaf streak
disease because of the climate.
In June 1999, black leaf streak disease was identified in the Provinces of Valverde,
Dajabón, Santiago Rodriguez and other communities of Montecristi (Polanco 1999).
Eastern region
The eastern region includes the provinces of Hato Mayor, La Altagracia, El Seibo,
La Romana and Higuey. Rainfall is an average of 1342-1583 mm per year distributed
281
MyLsd 17x24
25/06/03
15:15
Page 282
Mycosphaerella leaf spot diseases of bananas: present status and outlook
over 122-137 days (Secretaría de Estado de Agricultura, 1998). Because of the
favourable conditions, black leaf streak disease has caused considerable damage in
the region. Many growers have changed from plantain to another crop and the
plantains harvested are very small.
Black leaf streak disease was identified in the eastern region in the Provinces of
Hato Mayor and El Seibo in the communities of Hato Mayor, Sabana de la Mar and
El Valle, respectively (Figure 2). The disease was identified in 1998, i.e. two years
after the original identification in the northwest and it appears that the disease had
been introduced through planting material.
Before the occurrence of black leaf streak disease in 1998, approximately 1875
ha of plantain were grown in Hato Mayor and El Seibo but today, only 625 ha are
planted with the crop because of the disease. Even though the area and the production
are small, plantain is of economic and social importance for the region. Previously
the crop supplied the regional market and provided food and economic support to
several thousand small-scale farmers. Today, demand for plantain is satisfied by
supply from other producing areas.
The hybrid ‘FHIA-21’ is being introduced to this region as an alternative, but
acceptance has been limited.
Central region
The central region includes the provinces of Monte Plata, San Cristóbal and Santo
Domingo-Distrito Nacional and production of plantain is approximately 3750 ha
(Table 2). Rainfall is an average of 1400-1800 mm/year (Table 1).
The disease appeared in 1999 in the three provinces, and affected the communities
of Bayaguana, Monte Plata, San Cristobal, Yamasá and Villa Mella. The occurrence
of the disease appeared to have been caused by hurricane George in September 1998.
Hurricane George moved from the southeast to the southwest and winds blew strongly
into the middle of the island, suggesting that introduction was from the south region,
which was already affected.
Southwestern region
The southwestern region includes the provinces of Azua and San Juan de la Maguana.
Approximately 5722 ha of banana, mostly organic (Azua) and 5764 ha of plantain
are grown in the region (Secretaría de Estado de Agricultura, 1998; Departamento
de Sanidad Vegetal, 2000).
Annual rainfall for the region was an average of 665 mm in 1961-1990
(Table 1) and was mostly distributed over 62 days, mostly in November-December.
Thus, the region is at low risk from outbreaks of black leaf streak disease (Secretaría
de Estado de Agricultura, 1998).
Black leaf streak disease was reported on plantain from a small locality in the
province of San Juan de la Maguana in January 2000; however, production in the
province is not important. The disease was identified on organic bananas in Azua,
in February 2000. Spread of the disease in Azua has been limited by the weather
conditions; in contrast to Sigatoka disease which is moderately prevalent.
282
MyLsd 17x24
25/06/03
15:15
Page 283
Session 5
P.E. Jorge and T. Polanco
North, north-central and northeastern regions (Cibao Central)
The Cibao Central includes the north-central, northeastern and north regions, and
includes the provinces of La Vega, Espaillat, Sánchez, Samaná, Puerto Plata and others.
Production of plantain is 26 472 ha (Table 2) and of banana 9927 ha. Both crops are
mostly produced without irrigation, and are dependent on rainfall for the water supply.
These regions have an average rainfall of 1200-1350 mm/year, with 122-173 days
of rain per year (Secretaría de Estado de Agricultura, 1998). These conditions suggest
that black leaf streak disease has the potential to cause considerable damage to plantain.
In August 1998, black leaf streak disease was found on plantain in the northeastern
region in the provinces of Sánchez and Samaná but not in the north-central and
northern regions. In 2001, symptoms were recognized on plantain in the community
of La Isabela, province of Puerto Plata in the northern region. The disease has caused
considerable damage to the point where crops have been abandoned, or deleafing and
fungicide treatments implemented to ensure production, as was the case in La Isabela.
In March 2002, black leaf streak disease was identified in the north-central region
in the community Hoya del Camú, province of La Vega, and in April 2002 in the north
in the community of Moca, province of Espaillat. In both instances the disease affected
plantain. In Moca, symptoms were diagnosed at a farm of 1.5 ha. Information from
the growers suggests that the disease was introduced through plant material from
infected areas.
Southern region
In the provinces of Barahona and Neiba, most of the area planted with plantains is
irrigated. In this region, 5807 hectares of plantain are irrigated. Rainfall in 1961-1990
was on average less than 600 mm/year and was distributed over 48-69 days
(Secretaría de Estado de Agricultura, 1998). As a dry region, there is little likelihood
of serious damage from black leaf streak disease, unless favourable conditions were
to occur for prolonged periods.
Management of black leaf streak disease
In general, management of black leaf streak disease in the Dominican Republic depends
on the pressure exerted by the disease. The type of crop (banana or plantain), the size
of the production unit, whether production is conventional or organic and the prevailing
environmental conditions determine the techniques used to control black leaf streak
disease. A technical package was only adopted in response to the spread of the disease
within the country. Limited preventive measures had been taken when the disease was
not present.
Cultural practices
Banana growers and, to a lesser extent, plantain growers have adopted the practice
of deleafing. This is done every 7-10 days during the rainy season when the disease
pressure is high, and every 30 days in the dry season when disease pressure is low.
283
MyLsd 17x24
25/06/03
15:15
Page 284
Mycosphaerella leaf spot diseases of bananas: present status and outlook
In the eastern region, where environmental conditions favour the disease and
the pathogen is present, only growers who have adopted deleafing and biological
monitoring of the diseased have survived. Production of plantain in the region
declined from 1875 ha before the appearance of black leaf streak disease in 1998
to 625 ha today.
Chemical control
Conventional banana production is dependent on the use of fungicides, e.g.
triazoles, dithiocarbamates, benzimidazoles, mineral oils and to a lesser extent,
strobilurins (Polanco, 1998). Organic banana growers use mineral and vegetable oils
and organic products, e.g. citric acid and waxes. Preliminary data obtained by Polanco
(personal communication) showed that there was no difference between mineral oil
alone and mixtures of oil with the organic products. This work is to be repeated.
Conventional and organic banana growers spray between 7 to 10 times a year,
and sometimes up to 15 times, depending on rainfall. Sometimes treatments are not
effective or may have been unnecessary because of faulty forecast or monitoring.
Large-scale banana plantations, conventional or organic, are usually sprayed by
plane in the northwestern and south-western regions. Small-scale growers use
motorized high-pressure pumps. In general, plantain growers do not use fungicides.
Resistant hybrids
FHIA hybrids, especially ‘FHIA-21’ and to a lesser extent ‘FHIA-20’, were introduced
before the identification of the disease in the country. At present, there are 600 000
to 800 000 ‘FHIA-21’ plants in affected and not affected areas, representing less than
one percent of the total planted area. Production is for the fresh and industrial markets.
Research plots have been established to compare native plantains, ‘FHIA-21’ and
‘Rulo’ for their response to the technical package needed for a successful crop of ‘FHIA21’ and to different management practices including deleafing and fertilizer treatment.
When black leaf streak disease is absent, cultural practices and chemical control are
not implemented, unless if there is a high incidence of Sigatoka disease.
As a preventive measure, the replacement or planting of new areas with the resistant
clone ‘FHIA-21’ has recently been recommended and plants are available from the
Ministry of Agriculture. Acceptance of ‘FHIA-21’ by growers has been limited mainly
because the most important plantain production areas do not have the disease or the
spread is very limited, and consumers prefer the fruit of local cultivars.
Little has been done to diversify the gene pool of banana, and production is entirely
with ‘Cavendish’ cultivars. This is mainly because of the requirements of the banana
market and the prevalence of dry conditions in the banana production areas.
References
Departamento de Sanidad Vegetal. 2000. Proyecto manejo integrado de la Sigatoka negra
(Mycosphaerella fijiensis Morelet) en los cultivos de musáceas de la República Dominicana.
Secretaría de Estado de Agricultura, Santo Domingo, República Dominicana. 42pp.
284
MyLsd 17x24
25/06/03
15:15
Page 285
Session 5
P.E. Jorge and T. Polanco
Fouré E. 1997. La maladie des raies noires des bananiers et plantains en République
Dominicaine. Distribution, incidence et méthodes de contrôle. CIRAD-FHLOR. Rapport de
mission en République Dominicaine du 28 août au 5 septembre 1997.
Palm M.E. 1997. Trip report to the Dominican Republic.
Pérez Vicente L. 2000. Informe de la misión del consultor en manejo integrado de Sigatoka
negra, Dr Luis Pérez Vicente realizada en la Secretaría de Agricultura (SEA) de República
Dominicana del 28 de agosto al 29 de septiembre del 2000. FAO. 32pp.
Polanco T. 1998. La Sigatoka negra del plátano y guineo: reconocimiento y manejo.
Departamento de Sanidad Vegetal, Secretaría de Estado de Agricultura, Santo Domingo,
Republica Dominicana. 14pp.
Polanco T. 1999. Informe de estancia, Laboratorio de fitopatología CIRAD-AMIS, Montpellier,
Francia. Departamento de Sanidad Vegetal, Secretaría de Estado de Agricultura, Santo
Domingo, Dominican Republic. 14pp.
Rivas G.G., M.F. Zapater and J. Carlier. 2001. Estructura de poblaciones de Mycosphaerella
fijiensis en América Latina. Congreso Internacional de Sigatoka. SERBANA. San José, Costa
Rica, Abril 2001.
Secretaría de Estado de Agricultura. 1998. Anuario estadístico de planificación sectorial
agropecuaria. Subsecretaría de Estado de Planificación Sectorial Agropecuaria, Santo
Domingo, República Dominicana. 126pp.
Secretaría de Estado de Agricultura. 2000. Registro Nacional de Productores Agropecuarios,
Santo Domingo, República Dominicana.
Vidal A. 1992. Sigatoka negra en Cuba. En nuevos focos de plagas y enfermedades. Boletin
Fitosanitario de la FAO 40:1-2.
285
MyLsd 17x24
25/06/03
15:15
Page 287
Session 5
A.S. Riveros et al.
Microbiological control of black leaf
streak disease
A. S. Riveros, C. I. Giraldo and A. Gamboa
Abstract
Isolates of bacteria from the phyllosphere of tomato and banana were obtained from CATIE: Bacillus
cereus, Bacillus sp., Serratia marcescens, Serratia entomophila, unidentified strains with glucanolytic
and chitinolytic capacity (GS2-GS3-GC1-GBC2,SE/PO2,White) and one isolate of Mycosphaerella fijiensis
collected recently in Turrialba, Costa Rica. Crude culture filtrates of some microorganisms inhibited
ascospore germination and the growth in vitro of M. fijiensis colonies. The two filtrates with the
greatest effect resulted in changes to the ultrastructure of M. fijiensis hyphae, when examined under
a scanning electronic microscope, in comparison with untreated tissue.
Resumen - Control microbiológico de la Sigatoka negra
Los aislados de la colección del CATIE obtenidos de la filosfera de tomate y hojas de banano: Bacillus
cereus, Bacillus sp., Serratia marcescens, Serratia entomophila y cepas no identificadas con
capacidad glucanolítica y quitinolítica (GS2-GS3-GC1-GBC2, SE/PO2, White) y un aislado de
Mycosphaerella fijiensis recolectado recientemente en Turrialba, Costa Rica, fueron utilizados para
preparar filtrados de los cultivos líquidos. Los filtrados crudos de estos microorganismos se
evaluaron bajo condiciones in vitro con el fin de determinar la germinación de las ascosporas y
el crecimiento de las colonias de M. fijiensis (agente causal de la Sigatoka negra en banano y
plátano). Los resultados muestran un efecto inhibitorio importante de algunos de estos filtrados
en diferentes etapas de desarrollo de Mycosphaerella. La observación, bajo un microscopio
electrónico de barrido, de las estructuras del hongo tratado con dos filtrados prometedores,
muestra claras alteraciones de ultraestructura en el tejido tratado en comparación con el
testigo sin tratamiento.
Résumé – Lutte microbiologique contre la maladie des raies noires
Des isolats de bactéries de la phyllosphère de tomates et de bananiers ont été obtenus du CATIE :
Bacillus cereus, Bacillus sp., Serratia marcescens, Serratia entomophila, des souches non identifiées
ayant une capacité glucanolytique et chitinolytique (GS2-GS3-GC1-GBC2, SE/PO2, White) et un
isolat de Mycosphaerella fijiensis collecté récemment à Turrialba, au Costa Rica. Les filtrats bruts
de certains micro-organismes ont inhibé la germination des ascospores et la germination in vitro
de colonies de M. fijiensis. Les deux filtrats qui ont eu le plus d’effet ont induit des changements
de l’ultrastructure des hyphes de M. fijiensis, en microscopie électronique à balayage, par rapport
aux tissus non traités.
CATIE, Turrialba, Costa Rica.
287
MyLsd 17x24
25/06/03
15:15
Page 288
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Introduction
During the seventies, agriculture was characterized by an undiscriminated use of
agrochemicals. The situation has changed little since then, but international and national
regulations have imposed changes aimed at reducing pollution and making agriculture
sustainable. There are two types of agriculture: 1) high input agriculture characterized
by high productivity that is limited by environmental factors and 2) low input agriculture
with production that, in addition to environmental factors, is limited by pests, diseases
and weeds.
Changes in the market and the influence of globalization are forcing a reconsideration
of research in agriculture. The preferences of consumers play an increasingly important
role that affects the work of multidisciplinary teams made up of researchers, ecologists,
producers and industry.
The Tropical Agricultural Research and Higher Education Center (CATIE) has
defined one of its research objectives as the implementation of methodologies focused
on the biological control of the most common diseases and pests affecting economically
important tropical crops, such as those belonging to the Musaceae family.
CATIE started studies on Musa and M. fijiensis with a project financed by AID/ ROCA
(USA) the first phase of which started in July 1984. Other projects which followed were
financed by RENARM (USA), CIRAD (France), INIBAP, INCO-Musa, Natural Resource
Institute (NRI; UK), CINVESTAV (Mexico) and FONTAGRO.
These collaborative projects not only spurred research on biological control but also
on the control of black leaf streak disease, a disease that was already threatening banana
and plantain production. Other outcomes were the development of systems for somatic
embryogenesis, cell suspensions, plant pathology, cryopreservation, genetic transformation and the genetics of M. fijiensis populations.
Since then, the Plant Protection Unit at CATIE has developed integrated pest
management (IPM) practices for black leaf streak disease based on the preservation of
the environment, reduced risks to farmers, the rural population and consumers, and
the sustainability of traditional agriculture. Countries included within the CATIE mandate
have a rich biodiversity which may contain materials or products, e.g. genes of wild
plants or biopesticides, that might be useful in IPM programmes.
Research into the biological control of black leaf streak disease at CATIE encouraged
researchers involved in the AID/ROCAP-USA project, e.g. Dr Elkin Bustamante and his
team who were the first to work on the project. After a careful study of the different
aspects of the parasitic relationship between Musa and M. fijiensis: the biology and
morphology of the pathogen, the phenology and physiology of the plant, the
phyllosphere, soil (importance of rhizobacteria, endophytic fungi and organic
amendments), and methods of internal and external inoculation, a research programme
was constructed to better study these aspects (Figure 1).
Step 1. Identification of antagonistic microorganisms
The purpose was to isolate microorganisms antagonistic to M. fijiensis and to evaluate
their effectiveness under greenhouse and field conditions. One hundred and twenty
isolates with chitinolytic activity were obtained from plants of cv. ‘Grande naine’ coming
288
MyLsd 17x24
25/06/03
15:15
Page 289
Session 5
A.S. Riveros, C.I. Giraldo and A. Gamboa
Identification
of antagonistic
microorganisms
Effect of substrate
on antagonistic
microorganisms
External vs internal
application
of antagonistic
microorganisms
1
2
3
Induced resistance “ISR”,
substrates and promotion
of plant growth
Induced resistance and
promotion of plant growth
5
4
Figure 1. Steps involved in research on biological control at CATIE (1986-2001).
from two different locations: an area of high incidence of black leaf streak disease and
an area of low incidence of the disease. The ‘low incidence site’ provided the highest
population of microorganisms, which was evaluated on chitin agar. Thirteen bacterial
strains were selected on the basis of their chitinolytic activity (Serratia marcescens,
Serratia entomophyla and Bacillus spp.). Under greenhouse and field conditions the level
of control of M. fijiensis was 40% in comparison with a level of 60% using fungicides
(González, 1994; González et al. 1996).
Regarding glucanolytic activity, 196 strains of bacteria were collected from plants of
cv. ‘Grande naine’ of which 37 belonged to the genus Bacillus. The microorganisms were
purified and evaluated on glucan-agar and glucan nutrient-agar media. Seven strains
with glucanolytic activity were selected. GS2, GBC2, BS3 and BC1 showed antagonistic
effects, inhibiting germination of M. fijiensis ascospores in 25% of the cases and reducing
germination tube length in 47%. Four strains were tested in the presence and absence
of glucan. Commercial glucan being expensive, a common source of glucan from
agricultural waste was used (Talavera-Sevilla, 1996; Talavera et al., 1998a, b).
Step 2. Effect of substrate on antagonistic bacteria
The effect of different substrates on the growth and survival of antagonistic bacteria
were investigated. The aim was to modify the physical and nutritional conditions in
order to inhibit germination and establishment of the pathogen and favour antagonistic
organisms.
The bacterial strains used were: Serratia marcescens R1, Serratia entomophila A100
and Bacillus cereus A30. The substrates tested, singly or in combination, were leaf extract,
milk, foliar fertilizers, molasses, cassava starch, glucan and chitin. The highest recovery
level of bacteria was observed in molasses which had positive effects on antagonistic
microorganisms. A combination of milk and molasses increased multiplication and
survival of R1 and A30 (Ruiz-Silvera et al., 1997a, b).
Plants treated with a combination of chitin, yeast and calcium nitrate alternated
with commonly used fungicides, reduced the number of fungicide treatments by 40%
in comparison with fungicides alone (Arango-Ospina, 2000).
289
MyLsd 17x24
25/06/03
15:15
Page 290
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Step 3. Internal vs external application of antagonistic
microorganisms
The objectives of the study were to evaluate the effects of R1 and A30 applied
externally in combination with Silwet L-77, Nu-Film 17 and mineral oil, and to
evaluate an endophytic inoculation method via the roots. R1 was compatible with
stickers, mainly mineral oil. Mineral oil in combination with antagonistic
microorganisms reduced disease severity in comparison with the controls. The best
colonization of internal tissues was when A30 was applied directly inside the plant
(Miranda, 1996).
Step 4. The phenomenom of induced resistance and promotion
of plant growth
Stimulation by pathogens, non-pathogenic microorganisms, and by substances
of biological or non-biological origin can induce resistance in susceptible plants
Induced resistance to disease and growth promotion have potential for controlling
disease (Figure 2).
Four bacterial and one fungal suspensions were applied to the rhizosphere;
KH2PO4 and K2HPO4 solutions were applied to the leaves as abiotic exogenous
inducers. In a second experiment, microorganisms were evaluated with the addition
of sugarcane pulp, sugarcane filter press and coffee husks to the rhizosphere.
Pseudomonas fluorescens (PRA25), P. cepacia (AMMD) and Trichoderma harzianum
(Th) plus substrates increased growth the most and reduced disease in comparison
with the controls (water and substrates). However, the lowest percentage was
obtained with propiconazole (Tilt®). There was a significant and negative
correlation between them (Gutiérrez, 1996).
KH2PO4
Chemical inductors
{
Serratia marcescens R1
Bacillus cereus (A30)
Pseudomonas fluorescens (PRA25)
Pseudomonas cepacia (AMMD)
Trichoderma harzianum (Th)
R1 - A30
PRA25 - AMMD
Th
+
Substrates
sugarcane pulp
sugarcane filter press
coffee busk
Figure 2. Illustration of the concept of induced resistance to control black leaf streak disease (© S. Belalcázar,
2002).
290
MyLsd 17x24
25/06/03
15:15
Page 291
Session 5
A.S. Riveros, C.I. Giraldo and A. Gamboa
Step 5. Induced systemic resistance (ISR)
Investigations improved the understanding of the use of organic amendments,
antagonistic microorganisms, substrates as energy sources in combination with
mycorrhizal fungi and organic extracts known as efficient microorganisms, EMs
(Okumoto et al., 2001; Ayuso, 2000; Sanchez Garita et al., 1998; Okumoto, 1992).
Induced systemic resistance (ISR) to disease results from the inoculation of lower
leaves, or roots, with restrictive pathogens, non-pathogenic races of pathogens,
non-pathogens, products of pathogen or non-pathogens, and organic or inorganic
chemicals. ISR is also referred to in the scientific literature as SIR or SAR (Ku,?
2001).
The objective of the research was to evaluate, under greenhouse conditions,
the response of cv. ‘Grande naine’, as an example of a banana cultivar susceptible
to black leaf streak disease, and of ‘FHIA-23’ as an example of a clone resistance
to the disease, in the presence of 3 resistance inducers and one foliar substrate as
an energy source. The inducers were PRA25 bacteria and culture filtrate from
germinating spores of M. fijiensis strains according to Riveros and Lepoivre (1998),
and Acilbenzolar-S-metil (BION®,), a synthetic inducer provided by Syngenta. ISR
was higher in ‘FHIA-23’ than in cv. ‘Grande naine’.
Vermicompost increased ISR. BION® resulted in high ISR in both cultivars.
Rhizobacteria and M. fijiensis filtrate induced resistance only in ‘FHIA-23’ when in
the presence of an energy source. In the field, BION® reduced disease incidence in
cv. ‘Grande naine’ in comparison with conventional control measures (Patiño, 2001).
Massive applications of antagonistic bacteria or fungi on crops could have
unforeseen effects on the environment. The objectives of the study were to evaluate
the in vitro biological activity of microbiological filtrates on a M. fijiensis ascospore
preparation, the growth of M. fijiensis colonies, and the effects of two filtrates on
the cell structure of M. fijiensis. Emphasis was put on the isolation of strains with
glucanolytic and chitinolytic activity.
Materials and methods
Bacterial strains used in this study were obtained from CATIE’s Plant Protection Unit
Collection:
• Bacillus cereus (A30), isolated from tomato leaves (Lycopersicon pimpinelli),
Turrialba, Costa Rica (Okumoto, 1992).
• Serratia marcescens (R1), isolated from banana leaves (Musa sp.), Limon province,
Costa Rica (González, 1994).
• Serratia entomophila (SE), isolated from Canterbury Valley (New Zealand) from
the digestive tract of the scarabid Costelytra zealandica (donated by Trevor Jackson
from the AgResearch Lincoln Laboratory in 1994).
• GS2, GS3, GC1, GBC2, bacteria with glucanolytic activity isolated from banana
leaves, Indiana farm, Siquirres, Costa Rica (Talavera-Sevilla, 1996).
• SE/PO2, isolated from deep well water, Carmen de Siquirres, Costa Rica (Gamboa,
personal communication).
291
MyLsd 17x24
25/06/03
15:15
Page 292
Mycosphaerella leaf spot diseases of bananas: present status and outlook
• White, chitinolytic bacteria, isolated from plantain leaves, La Montaña farm, CATIE,
Costa Rica (Arango-Ospina, 2000).
• Extracts of the fluid obtained from suspensions of the conidia of M. fijiensis isolated
from La Montaña farm, Turrialba, Costa Rica.
Microbiological extracts were prepared on Petri dishes containing nutrient agar
medium (DIFCO). Two boxes per bacterial strain were inoculated with 20 ml bacterial
solution previously maintained at 4°C. The bacterial suspension was uniformly
distributed over the medium using a glass handle and the boxes were incubated at 30°C
for 2 days. Once the bacteria started growing, sterile distilled water was added to the
medium and the bacterial suspension removed with a scalpel; approximately 20 ml of
each bacterial suspension was transfered to sterile vials. The absorbency at 600 nm of
a 3-ml solution was measured with a spectrophotometer. One ml of each suspension
with an optical density of 1.2 was transferred to 200 ml of sterile nutrient medium
(DIFCO) and incubated for 12 hours at 30°C and 150 rpm. Absorbency was measured
again at 600 nm and gave values of 1.2 ±0.02 after 12 hours
Suspensions were adjusted to an absorbency of 1.1 by adding sterile nutrient medium.
Cell-free extracts were obtained by centrifugation at 5000 rpm for 40 minutes followed
by vacuum filtration on 0.22 mm membranes. Extracts were kept at 4°C in sterile flasks
and protected from the light.
A suspension of 2x105 conidia/ml of M. fijiensis in 700 ml of sterile distilled water
was agitated at 100 cycles/min for 48 hours in darkness and then filtered using ethamine
and Whatman paper. The residue was lyophilized to obtain 0.341 mg of powder, which
was diluted in 70 ml sterile distilled water to obtain a final 20x concentration and filtered
through 0.22 µm Nalgene Disposable Filterware filters. The filtrate was protected from
light and kept at 4°C until its utilization.
Samples of plantain leaves with black leaf streak disease were transferred to La
Montaña farm, which belongs to CATIE. Using a magnifying glass, fragments of viable
perithecia were removed and transferred to the laboratory in paper bags. The samples
were checked using a stereomicroscope, and sections with abundant sporulating lesions
were cut into 2x2 cm pieces. Two to four of these pieces were stapled to pieces of paper
and incubated in a humid chamber for 24 hours at room temperature. They were then
transferred to sterile distilled water for 5 minutes to hydrate the perithecia. Ascospores
discharged on water agar (4% w/v). Treatments were 0.5, 0.1 and 0.01 ppm dilutions
of microbiological filtrates and the controls were without microbiological filtrate or with
fungicide.
Diluted culture filtrates were mixed with 15 ml of V8 medium, with constant agitation
and then transferred to Petri dishes; there were 3 replicates per treatment. Seven-dayold sub-cultures of a strain of M. fijiensis isolated from La Montaña were used. Twenty
colonies of 1–1.5 cm in diameter were excised with a scalpel and transferred to an assay
tube with 3 ml of 0.05% (v/v) Tween 20 and agitated in a vortex. The assay tubes were
left to rest for 10 minutes and 10 drops of liquid were transferred to each Petri dish
and spread with a glass handle. Dishes were sealed, and incubated in darkness for 5
days at 26°C.
After 5 days of incubation, the diameters of M. fijiensis colonies were measured
using a micrometer in the 4x field of a microscope. Thirty readings were taken,
292
MyLsd 17x24
25/06/03
15:15
Page 293
Session 5
A.S. Riveros, C.I. Giraldo and A. Gamboa
frequency intervals of amplitude 10 were done and only the 10 data from the
interval with higher frequency were registered to conduct the analysis of the
inhibiting effect of the microbiological filtrates.
The longest germination tubes were measured using a micrometer in the 10x
or 40x field of a microscope. One hundred readings were taken per treatment.
Since bacteria were cultured in nutrient medium, bioassays included a treatment
with this medium.
Initial data were multiplied by a correction factor to transform the values to
microns, 10 and 2.5 for 10x and 40x, respectively. The difference in germ tube
length between the treatment and the control was used to calculate the inhibiting
effect of the microbiological filtrates.
Results and discussion
Figure 3A shows preliminary results obtained for growth inhibition of M. fijiensis
ascospores discharged onto different concentrations of microorganism filtrates. The
percentage of inhibition was almost 50% and in some cases higher when the medium
included filtrates of bacteria with glucanolytic activity (GBC2) and chitinolytic activity
(SE/PO2 and White).
Regarding colony diameter (Figure 3B) the general tendency remained similar
except that another bacterium with glucanolytic activity (GC1) showed a higher
inhibition which was not fully revealed during the sexual phase of ascospore
development.
The crude M. fijiensis filtrate (FCMf) also revealed, a clear inhibiting effect on
ascospore and colony growth at a concentration of 0.5 ppm but not at the other
concentrations. This poses the question as to whether the toxin(s) produced by
M. fijiensis spores during the germination process can inhibit the pathogen in a
“suicidal” type of action.
After five days of incubation, physical growth had stopped in the cultures with
the crude filtrates of GBC2 and SE/PO2 at the 0.1 ppm concentration in comparison
to the absolute control (water). Electronic transmission microscopy revealed
modifications at the level of cell organelles with a strong presence of electrodense
osmophylic globules that were not found neither in the control nor in the transversal
longitudinal cut (Figure 4).
The preliminary results suggest that liquid culture filtrates of four bacterial strains
with glucanolytic or chitinolytic activity, and the liquid filtrate of germinating spores
of a Costa Rican strain of M. fijiensis inhibited the growth of M. fijiensis germ tubes
and colonies.
Crude liquid preparations diluted from antibiotic(s) or toxin(s) and without
bacterial or fungal cells had similar effects as the fungicide Tilt®.
The promising microbiological filtrates need to be evaluated under greenhouse
and field conditions with or without adjuvant applications.
Acknowledgements
The authors thank INIBAP and INIBAP-LAC for the financial support to conduct this
research.
293
25/06/03
15:15
Page 294
Mycosphaerella leaf spot diseases of bananas: present status and outlook
140.0
A
% of inhibition
100.0
60.0
20.0
- 20.0
- 60.0
Tilt
White
GBC2
GBC2
GBC2
GC1
GS3
GS2
SE
R1
- 140.0
A30
- 100.0
Microbiological filtrates
0.5 ppm
0.1 ppm
0.01 ppm
120.0
B
100.0
% of inhibition
80.0
60.0
40.0
20.0
Tilt
White
GBC2
GBC2
GBC2
GC1
GS3
GS2
SE
- 20.0
R1
0.0
A30
MyLsd 17x24
Microbiological filtrates
A30:Bacillus cereus; R1:Serratia marcescens; SE: Serratia entomophila; GS2, GS3, GC1 and GBC2:
bacteria with glucanolytic activity; SEPO2 and White: bacteria with chitinolytic activity; FCMf:
conidial fluid; Tilt: triazole fungicide.
Figure 3. Effect of three dilutions of nine antagonistic microbiological filtrates on A) germ tube growth
and B) colony diameter of M. fijiensis . Arrows indicate where crude filtrates affected growth.
294
MyLsd 17x24
25/06/03
15:15
Page 295
Session 5
A.S. Riveros et al.
GBC2
Control
0.1 ppm
(water)
SE/P02
0.1 ppm
Figure 4. Cytological changes revealed by transmission electron microscopy of M. fijiensis hyphae tissues
treated with crude filtrates of bacteria with glucanolytic (CBC2) and chitinolytic (SE/PO2) activity.
References
Arango-Ospina M.E. 2000. Manejo de sustratos para el control biológico de Sigatoka negra
(Mycosphaerella fijiensis) en el cultivo de banano. Tesis Mag. Sc. Biblioteca ORTON-CATIE,
Turrialba, Costa Rica. 102pp.
Ayuso F. 2000. Influencia de enmiendas orgánicas y un hongo endomicorricico sobre el
nematodo Radopholus similis, en banano Musa (AAA). Tesis Mag. Sc. Biblioteca ORTONCATIE, Turrialba, Costa Rica. 114pp.
González R. 1994. Efecto de microorganismos quitinolíticos en el desarrollo de Sigatoka negra
(Mycosphaerella fijiensis) en banano. Tesis Mag. Sc. Biblioteca ORTON-CATIE, Turrialba,
Costa Rica. 97pp.
González R., E. Bustamante, Ph. Shannon, S. Okumoto and G. Leandro. 1996. Selección de
microorganismos quitinolíticos en el control de Sigatoka negra (Mycosphaerella fijiensis)
en banano. Manejo Integrado de Plagas (Costa Rica) 40:6-11.
Gutiérrez FA. 1996. Estudio de factores en la inducción de resistencia a Mycosphaerella fijiensis
y promoción de crecimiento en plantas de banano. Tesis Mag. Sc. Biblioteca ORTON-CATIE,
Turrialba, Costa Rica. 91pp.
Ku J. 2001. Concepts and direction of induced systemic resistance in plants and its
application. European Journal of Plant Pathology107:7-12.
295
MyLsd 17x24
25/06/03
15:15
Page 296
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Miranda J.E. 1996. Evaluación de microorganismos antagonistas al hongo Mycosphaerella
fijiensis Morelet, colocados en el interior y exterior de la planta de banano. Tesis Mag.
Sc. Biblioteca ORTON-CATIE, Turrialba, Costa Rica. 101pp.
Okumoto S. 1992. Efecto de enmiendas sobre bacterias antagónicas a Alternaria solani en
tomate (Lycopersicon esculentum Mill). Tesis Mag. Sc. Biblioteca ORTON-CATIE, Turrialba,
Costa Rica. 114pp.
Okumoto S., E. Bustamante and A. Gamboa. 2001. Actividad de cepas de bacterias
quitinolíticas antagonistas a Alternaria solani in vitro. Manejo Integrado de Plagas (Costa
Rica) 59:58-62.
Patiño L.F. 2001. Efecto de una fuente de energía, tres inductores de resistencia y un sustrato
foliar sobre Sigatoka negra en banano. Tesis Mag. Sc. Biblioteca ORTON-CATIE, Turrialba,
Costa Rica. 91pp.
Riveros A.S. and P. Lepoivre. 1998. Alternativas bioquímicas para el control indirecto de
Sigatoka en Musáceas. Pp. 436-447. Resúmenes. XIII Reunión ACORBAT, Guayaquil,
Ecuador.
Ruiz-Silvera C., E. Bustamante, F. Jimenez, J.L. Saunders, S. Okumoto and R. Gonzalez. 1997a.
Efecto de sustratos sobre crecimiento y supervivencia de bacterias antagonistas a
Mycosphaerella fijiensis. Manejo Integrado de Plagas (Costa Rica) 45:1-8.
Ruiz-Silvera C., E. Bustamante, F. Jimenez, J.L. Saunders, S. Okumoto and R. Gonzalez. 1997b.
Sustratos y bacterias antagonistas para el manejo de Mycosphaerella fijiensis en banano.
Manejo Integrado de Plagas (Costa Rica) 45:9-17.
Sánchez Garita V., E. Bustamante and R. Shattock. 1998. Selección de antagonistas para el
control biológico de Phytophthora infestans en tomate. Manejo Integrado de Plagas (Costa
Rica) 48:25-34.
Talavera-Sevilla M.E. 1996. Determinación de ß-glucano en subproductos agrícolas y
evaluación del efecto de microorganismos glucanoliticos sobre Mycosphaerella fijiensis
en banano. Selección de antagonistas. Tesis Mag. Sc. Biblioteca ORTON-CATIE, Turrialba,
Costa Rica. 80pp.
Talavera M., E. Bustamante, R. González and V. Sanchez. 1998a. Selección y evaluación en
laboratorio y campo de microorganismos glucanoliticos antagonistas a Mycosphaerella
fijiensis. Manejo Integrado de Plagas (Costa Rica) 47:24-30.
Talavera M., F. Lopez, E. Bustamante and R. González. 1998b. Extracción y cuantificación
de beta-glucano a partir de sustratos comunes en el trópico. Manejo Integrado de Plagas
(Costa Rica) 47:31-36.
296
MyLsd 17x24
25/06/03
15:15
Page 297
Session 5
E. Spaans and L. Quiros
Precision agriculture to improve
management decisions and field
research
E. Spaans and L. Quiros
Abstract
Precision agriculture helps farmers to improve management decisions. Since standard
management practices do not take into account variability in the environment, resources may
be wasted at certain stages of production and insufficient in others. Instead of calculating an
average over the whole farm, there is a need for a detailed profit analysis of agricultural
enterprises. It should be done at a high spatial resolution in order to make decisions that take
into account local conditions within a field. At the Commercial Farm of EARTH University, we are
mapping harvests at a spatial resolution of 4 ha within a 110 ha plantation.The network of railways
in the plantation is used to georeference the origin of the fruit which, together with the weight
of the fruits, is stored in a database. The costs are divided into fixed and variable costs, and
subtracted from the income to produce a map of profits. The spatial variability of the harvest
was enormous, with a greater than 300% difference within the field. To investigate the possible
causes of the variability, we are monitoring parameters in the field that affect the growth of
banana, e.g. soil fertility, plant nutrition, functional roots and age of plantation. Correlation
coefficients between production and each of the parameters were calculated. The coefficients
can be used to decide exactly what needs to be done in the areas not generating sufficient profit.
This scheme of intensive and systematic data acquisition, and interpretation provides
opportunities for field research and hence to improve management practices including the control
of Sigatoka disease. The project is in progress and we present the preliminary results we have
obtained.
Resumen - Agricultura de precisión como base para mejorar las decisiones de manejo
e investigación en la finca
La agricultura de precisión es una herramienta eficaz para ayudar a los agricultores a tomar
mejores decisiones de manejo. Debido a que las prácticas de manejo normalizadas no reconocen
la variabilidad del ambiente, se gastan recursos en algunas áreas, mientras que en otras no se
invierte lo suficiente. Este hecho requiere realizar un análisis de ganancias detallado de nuestra
empresa agrícola; de esta manera, no es el promedio global de toda la finca, sino una resolución
Escuela de Agricultura de la Región Tropical Húmedo (EARTH University), San José, Costa Rica
297
MyLsd 17x24
25/06/03
15:15
Page 298
Mycosphaerella leaf spot diseases of bananas: present status and outlook
espacial alta, la que permite hacer tomas de decisiones que respondan a las condiciones locales
en el campo. Los ingresos provienen de la cosecha, así, que hacemos un mapa de la cosecha con
una resolución espacial de 4 ha dentro de una plantación de 110 ha en la Finca Comercial de la
Universidad EARTH. La red de ferrocarril en la plantación sirve para hacer referencia geográfica
al origen de la fruta, que, junto con su peso, medido en la planta empacadora, se almacena en
una base de datos. Los costos se dividen en costos fijos y variables, y se restan del ingreso para
producir un mapa de ganancias. Se descubrió que la variabilidad espacial de la cosecha fue enorme,
con más de 300% de diferencias dentro del campo. Para investigar las causas posibles de esta
variabilidad, empezamos a monitorear varios parámetros en el campo que afectan el crecimiento
de los bananos, como la fertilidad del suelo, nutrición de la planta, raíces funcionales y la edad
de la plantación, y luego calculamos los coeficientes de correlación entre la productividad por
un lado y cualesquiera de los parámetros del campo por otro. Este coeficiente puede ser utilizado
como una guía para decidir las necesidades exactas que deben ser cumplidas en cada una de
las áreas que no generan ganancias suficientes. Este esquema de adquisición de datos intensivo
y sistemático y su interpretación también permiten que las investigaciones en el campo mejoren
eficazmente las prácticas de manejo, incluyendo el control de la Sigatoka. Este proyecto se
encuentra en progreso y se discutirán los resultados preliminares.
Résumé - L’agriculture de précision pour améliorer les décisions de gestion et la
recherche en champ
L’agriculture de précision aide les agriculteurs à améliorer leurs décisions de gestion. Les pratiques
de gestion standard ne prennent pas en compte la variabilité de l’environnement, et des ressources
pourraient donc être dilapidées à certains stades de la production mais être insuffisantes à d’autres.
Au lieu de calculer une moyenne sur toute l’exploitation,une analyse détaillée du profit est nécessaire
pour les entreprises agricoles. Elle devrait être réalisée avec une résolution spatiale élevée afin de
prendre des décisions qui tiennent compte des conditions locales au sein du champ. Dans la ferme
commerciale de l’Université EARTH, nous cartographions les récoltes avec une résolution spatiale de
4 ha sur une plantation de 110 ha. Le réseau des voies ferrées est utilisé pour géoréférencer l’origine
des fruits qui, avec le poids des fruits, est stocké dans une base de données. Les coûts sont divisés en
coûts fixes et variables, et soustraits du revenu pour produire une cartographie des bénéfices. La
variabilité spatiale de la récolte s’est avérée énorme,avec une différence de plus de 300% à l’intérieur
d’un même champ. Afin de rechercher les causes possibles de cette variabilité, nous faisons le suivi
des paramètres en champ qui affectent la croissance des bananiers, tels que la fertilité du sol, la
nutrition des plants, les racines fonctionnelles et l’âge de la plantation. Les coefficients de corrélation
entre la production et chacun des paramètres ont été calculés. Les coefficients peuvent être utilisés
pour décider exactement ce qui doit être fait dans les zones qui ne génèrent pas de bénéfices suffisants.
Ce schéma intensif et systématique d’acquisition de données, ainsi que son interprétation, offrent
des occasions de recherche en champ qui pourraient permettre d’améliorer les pratiques de gestion,
y compris la lutte contre les cercosporioses. Le projet est en cours et nous présentons les premiers
résultats que nous avons obtenus.
Introduction
Globalization has opened up markets and increased competition between agricultural
producers worldwide. In order to continue being competitive, farmers must improve
the efficiency of their production systems, i. e. reduce costs while maintaining or
even improving the production as well as the social and environmental impact of
their agricultural enterprise.
Productivity has increased considerably over the last century due to technological
advances (e.g. in plant nutrition and fertilization, genetic improvement, mechaniza298
MyLsd 17x24
25/06/03
15:16
Page 299
Session 5
E. Spaans and L. Quiros
tion and pest management) that have allowed the agricultural community to respond
to the increasing demands for agricultural products. These advances and the
improved understanding of plant-environment interactions have resulted in the
development of technological packages for major crops. A technological package is
a recommended set of agricultural practices that describe all stages of production,
e.g. soil preparation, crop planting, pest and disease management, soil fertilization,
and harvesting and packing methods. These generalized recommendations were
developed for average conditions and do not take into account specific conditions
encountered in the field. Inevitably, resources are wasted in some areas, hence raising
production costs and increasing the risk of environmental contamination. Similarly,
not enough resources may be invested in certain areas, resulting in suboptimal growth
and loss of income.
The reason is that resources (soil, weather, water, etc.) are not homogenous
throughout the farm and over time. Moreover, the socioeconomic conditions of the
enterprise (the nature of the market, prices, policies and standards of certification,
amongst others) are also in a constant state of change.
This calls for a more entrepreneurial approach to agriculture. We need to improve
decision making, to take more precise ones to fine-tune the management of the
resources to what is really needed. That is precisely what precision agriculture is all
about: doing the right thing, at the right time and at the right place. To answer the
fundamental question of what is the right thing to do, we need to acquire detailed
information about the production system at the adequate spatial resolution.
Traditional applications of precision agriculture involve high-tech data acquisition:
a Global Positioning System (GPS) for georeferencing the data and monitoring sensors
mounted on harvest equipment to gather high resolution data while harvesting. A
Geographic Information System (GIS) is used to manage the large amount of data
and map them. These requirements have hindered the implementation of precision
agriculture in many Latin American production systems where harvest is often done
manually and access to and support for technology is sometimes limited. We think
it is more useful to reflect on the basic principle of precision agriculture and then
creatively develop the implementation, considering the specific conditions of each
production system.
This paper presents our interpretation of precision agriculture and its
implementation at the commercial banana farm of EARTH University. Its relevance
to this workshop is that it improves decision-making and provides a unique
opportunity for farm research, including the study of the effectiveness of agricultural
practices to control the leaf spot diseases caused by Mycosphaerella.
Implementation of precision agriculture
Being competitive means optimizing profits, not harvest. Considering the spatial and
temporal variability of resources throughout the farm, it is to be expected that profits
vary as well, particularly if the same technological package is used over the entire
farm. Thus an overall financial analysis of the farm is not sufficient; we need a
detailed analysis of costs and benefit throughout the entire farm, in order to allow
for precise management.
299
MyLsd 17x24
25/06/03
15:16
Page 300
Mycosphaerella leaf spot diseases of bananas: present status and outlook
As an example we describe banana production on a 110 ha farm located on the
campus of EARTH University in the Humid Tropics of the Atlantic Zone of Costa
Rica. Since the financial benefit results from the harvested banana, a precise
monitoring of the harvest is needed to precisely monitor income. The farm was divided
into blocks of 4 ha (100 m wide by 400 m long). The railway that transports the
fruits runs through the center of each block, and fruits are collected on a distance
of 50 m on each side of the tracks. Each block has an identification, and printed
labels are provided to the harvesters to identify the geographical origin of every
bunch. At the packing plant, each bunch is weighed and the information, together
with the geographical origin (spatial component) and the day of harvest (temporal
component), is stored in a database. This mapping system is inexpensive and lowtech, and does not require GPS since the railway network serves as the geographical
reference within the plantation.
The bananas are still attached to the stalk when they are weighed. To obtain the
weight of the bananas, the weight of the bunch is reduced by 10% to account for
the weight of the stalk. Finally, a general value of 12% is used to account for the
rejected banana (because of inadequate quality), and the resulting weight is divided
by 18.14 to obtain the number of boxes exported. Data were collected for the period
February to October of the year 2001.
The costs of banana production were divided into fixed and variable costs. Fixed
costs were equally divided over the entire productive surface area of the plantation,
while the variable costs were equally divided over the weight of the bunches.
Costs are currently recorded for the entire farm, thus we do not have the same
spatial resolution as for the production costs. The quantity of rejected bananas is
also an average value reported by the packing plant. These factors limit data analysis,
but the situation shows how in practice the implementation of precision agriculture
is a process of gradual improvements in data acquisition and interpretation as the
entire team familiarizes itself with the process. This year, for example, we started
recording production cost per block.
The cost and harvest data are used to determine the profit of every block within
the plantation. In order to explain and eventually reduce the variability in harvest,
we also monitor field parameters that we know affect plant growth, like soil fertility,
plant nutritional status, nematode infestation (as expressed in % of functional roots),
and age of plantation. This last parameter was included because the plantation was
planted about 40 years ago and recently some areas have been renewed. To determine
the degree to which each field parameter affects the production of bananas,
correlation coefficients were calculated between the harvest and each of the field
parameters.
Results and discussion
The productivity of bananas varied widely over the entire farm, ranging from
943 to 3040 boxes/ha/year, with an average of 1538 boxes/ha/year. These numbers
translated into a loss of US$2655/ha/year in the least productive block, and a profit
of US$2316/ha/year in the most productive block. Overall, the financial loss of the
plantation was US$1245/ha/yea. This shows the usefulness of precision agriculture,
300
25/06/03
15:16
Page 301
Session 5
E. Spaans and L. Quiros
since we can now locate the problem areas and quantify their impact on the financial
return of the farm.
So what should we do with this information, and where and when? To answer
these questions we analysed the data statistically. Although we found a range of
soil fertility, plant nutritional status and nematode infestation, none of these
parameters were significantly correlated (P<0.05) with the harvest, except for foliar
Mg (r = 0.52) and Cu (r = 0.50). This means that when the harvest is good, levels
of foliar Mg and Cu are high, and when harvest is bad, levels of foliar Mg and Cu
are low. But although significant, the correlation was not very strong. Figure 1 shows
what really happened in the plantation. All the areas that produced profit are the
ones that had been renewed, and all the areas with old plant material produced losses.
From this (preliminary) analysis we recommended that the right thing to do was to
speed up the rejuvenation of the plantation, starting with the least productive block.
The next obvious question is when should a plantation be renewed. There is no
standard answer. The moment will depend on cultivar, soil conditions and
management practices. The answer will be provided by the continuous and precise
profit analysis of the plantation over time.
Interestingly, interpretation of the foliar analyses in comparison with optimum
published values, suggested that the entire plantation was deficient in N, even the
blocks that produced 3040 boxes/ha/year. Without this careful spatial analysis of
production and field parameters, the recommendation would have probably been to
increase nitrogen fertilization, but this would probably not have solved the problem.
Instead, it would have probably increased production costs and enhanced the risk
of nitrate leaching to groundwater.
3600
= 40-year-old block
3000
= 2 to 4-year-old block
2500
Yield (boxes/ha/yr)
MyLsd 17x24
Break-even
2000
1500
1000
500
0
Block
Figure 1. Ranking of blocks in order of increasing productivity. The break-even point represents the production
needed to cover the costs (zero profit).
301
MyLsd 17x24
25/06/03
15:16
Page 302
Mycosphaerella leaf spot diseases of bananas: present status and outlook
We propose that this type of on-farm research also be used to identify farmspecific optimum values or tolerance levels, which could lead to a better interpretation
of field data. Furthermore, this implementation of precision agriculture could be used
to rapidly and effectively evaluate alternative practices, by selecting those blocks
that meet certain criteria and applying different treatments to them. The monitoring
system provides us with continuous and timely feedback about the plant response
to the different treatments. We are now including the monitoring of leaf spot diseases
in the plantation, which will allow us to investigate the interaction between the
disease and other field parameters, and determine the tolerance levels for the desired
production.
Conclusion
In this paper we presented the principles of precision agriculture, and discussed a
case where we implemented precision agriculture on a commercial banana plantation.
The implementation is a gradual process, and we are still far away from a precisely
managed farm. However, we have shown that the system has already provided
valuable information which we have been able to use to take effective management
decisions. The continuous and precise data collection inherent to precision agriculture
should make it possible to rapidly and effectively investigate alternative practices.
302
MyLsd 17x24
25/06/03
15:16
Page 303
Session 5
S. Knight et al.
Poster
The role of managing resistance
to fungicides in maintaining
the effectiveness of integrated
strategies to control
black leaf streak disease
S. Knight1, M. Wirz1, A. Amil2, A. Hall2 and M. Shaw3
Abstract
Fungicide programmes are essential for commercial production of banana in all regions where
Mycosphaerella fijiensis is prevalent. A key factor influencing the design of fungicide programmes is
the importance of following resistance management principles, in order to preserve long-term
effectiveness. Resistance management is based on the appropriate limitation and alternation of
products that have a site-specific mode of action. The introduction in 1997 of the first strobilurin
fungicide, azoxystrobin, represented a significant step forward in the integrated control of black leaf
streak disease, because of its efficacy and favourable environmental and toxicological profile. In
recognition of its site-specific mode of action,anti-resistance management guidelines were developed
by the Fungicide Resistance Action Committee (FRAC) before its commercial introduction. The first
strobilurin-resistant individuals were documented in 2000 in Costa Rica,and resistance to strobilurin
has reached high levels on some farms in the main banana production zones of Costa Rica.
Molecular characterization of resistant isolates has identified the cause of resistance as a single point
mutation in the fungal target protein, cytochrome b. A large-scale population dynamics study is in
progress to examine the evolution of resistance in the field, using molecular techniques (PCR). The
factors that influence this evolution (disease pressure, climate, fungicide spray programme) will be
examined, and the extent of migration of the resistant population will be estimated. The study will
enable recommendations to be validated or improved,and will support efforts to limit the proliferation
of resistance to this important group of fungicides.
1Syngenta
Costa Rica SA, San José, Costa Rica
Hill International Research Centre, Bracknell, UK
3
University of Reading, Reading, UK
2Jealott’s
303
MyLsd 17x24
25/06/03
15:16
Page 304
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Resumen - Papel del manejo de la resistencia a funguicidas en el mantenimiento de la
eficacia de las estrategias integradas para el control de la Sigatoka negra
Los programas de funguicidas siguen siendo esenciales para la producción bananera comercial en
todas las regiones donde prevalece Mycosphaerella fijiensis. Un factor clave que influye sobre el diseño
de los programas de aplicación de funguicidas es la importancia de seguir los principios del manejo
de la resistencia, con el fin de preservar la eficacia a largo plazo. El manejo de la resistencia se basa
en la limitación y alternación apropiadas de los productos que tienen un modo de acción específico
en el sitio. La introducción en 1997 del primer funguicida strobilurina, azoxistrobina, representó un
significativo paso hacia adelante en el control integrado de la Sigatoka negra, debido a su excelente
eficacia y perfil ambiental y toxicológico favorable.En reconocimiento de este modo de acción específico
del sitio, se desarrollaron guías de manejo anti-resistencia por el Comité de Acción de Resistencia a
los Fungicidas (Fungicide Resistance Action Committee (FRAC) antes de su introducción comercial. Los
primeros individuos resistentes a la strobilurina fueron documentados en 2000 en Costa Rica, y la
resistencia a la strobilurina ha alcanzado niveles altos en algunas fincas en las principales zonas
productoras de banano de Costa Rica. La caracterización molecular de los aislados resistentes ha
identificado como la causa de la resistencia una mutación puntual individual en la proteína diana
fungosa, citochromo b. Un estudio a gran escala de las dinámicas de la población se está llevando a
cabo actualmente para examinar la evolución de la resistencia en el campo, utilizando técnicas
moleculares (PCR). Se examinarán los factores que influyen sobre esta evolución (presión de la
enfermedad, clima, programa de rociado de fungicidas), y se estimará la extensión de migración de
la población resistente. El estudio permitirá validar o mejorar las recomendaciones para el manejo
de la resistencia, y apoyará los esfuerzos que se realizan para limitar la proliferación de la resistencia
en este importante grupo de fungicidas.
Résumé – Le rôle de la gestion de la résistance aux fongicides dans le maintien de
l’efficacité des stratégies de lutte intégrée contre la maladie des raies noires
Les plantations commerciales de bananes sont extrêmement dépendantes des applications de
fongicides partout où Mycosphaerella fijiensis prévaut. Un facteur important qui influence la
conception des programmes d’arrosage est le respect des principes de gestion de la résistance aux
fongicides afin de préserver leur efficacité.La gestion de la résistance repose sur la restriction de l’usage
et l’alternance de produits qui ont un mode d’action spécifique. L’introduction en 1997 du premier
fongicide à base de strobilurine, l’azoxystrobine, représente une avancée importante dans la lutte à
la maladie des raies noires étant donné son efficacité et son profil environnemental et toxicologique
favorable. En reconnaissance de son mode d’action spécifique, des lignes directrices pour empêcher
le développement de résistance ont été élaborées par le Fungicide Resistance Action Committee (FRAC)
avant sa distribution commerciale.Les premiers cas de résistance à la strobilurine ont été documentés
en 2000 au Costa Rica,et les niveaux de résistance ont atteint des taux très élevés sur certaines fermes
des principales zones de production bananière au Costa Rica. La caractérisation d’isolats résistants
a permis de remonter à la source de la résistance, une mutation isolée dans une protéine ciblée du
champignon, le cytochrome b. Une étude de population à grande échelle utilisant des techniques
moléculaires (PCR) est en cours afin de suivre l’évolution de la résistance en champ. Les facteurs qui
influencent cette évolution (la pression de la maladie, le climat, le programme d’application des
fongicides) seront examinés et l’étendue de la migration de la population résistante estimée. L’étude
permettra de valider ou améliorer les recommandations et participera à limiter la prolifération de la
résistance à ce groupe important de fongicides.
Introduction
A key factor influencing the design of programmes to control Mycosphaerella fijiensis
is the importance of following resistance management guidelines to ensure the long
term effectiveness of site-specific fungicides.
304
MyLsd 17x24
25/06/03
15:16
Page 305
Session 5
S. Knight et al.
The introduction in 1997 of the first strobilurin fungicide, Bankit
(azoxystrobin), represented a significant step forward in the integrated control
of black leaf streak disease, due to its excellent efficacy and favourable
environmental and toxicological profile. It has a site-specific mode of action, the
inhibition of electron transport at the Qo site of cytochrome bc1. Anti-resistance
management guidelines were developed by the Fungicide Resistance Action
Committee (FRAC) before commercial introduction, and a global sensitivity
monitoring programme was initiated.
The first strobilurin-resistant individuals were documented in 2000 in Costa
Rica, and resistance to strobilurin has reached high levels on some farms in the
main banana production zones of Costa Rica. Molecular characterization of
resistant isolates has identified the cause of resistance as a single point mutation
in the fungal target protein, cytochrome b, known as G143A. Resistant isolates can
be detected via quantitative polymerase chain reaction analysis.
A large-scale population dynamics study was initiated to examine the
evolution of resistance in the field by using molecular analysis. The factors that
influence this evolution (disease pressure, climate, fungicide spray programme)
will be examined, and the extent of migration of the resistant population will be
estimated.
Materials and methods
In vitro bioassay
Sensitivity of M. fijiensis to azoxystrobin, was evaluated using the in vitro
methodology recommended by the FRAC.
Sporulating tissue collected from the field was allowed to discharge onto
fungicide-amended agar, and elongation of ascospore germ tubes was measured
at 1 ppm and 10 ppm of azoxystrobin. A 75% growth or more at the discriminating
rate of 10 ppm relative to the control indicated resistance to strobilurin.
Molecular detection of strobilurin resistance
The presence of the G143A mutation was detected using a diagnostic primer for
G143A. This primer is extended only when the G143A mutation is present in the
sample.
Field population dynamics in Costa Rica
Repeated sampling of infected foliar tissue by using a multilayered hierarchical
sampling structure is in progress in Costa Rica. Plantations were selected to meet
the following criteria:
• Presence/absence of strobilurin selection pressure;
• Presence/absence of resistance mutation;
• High/low levels of relative disease pressure.
305
MyLsd 17x24
25/06/03
15:16
Page 306
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Results and discussion
Global resistance monitoring results
The results of the global sensitivity monitoring for azoxystrobin are shown in Table 1.
Table 1. Prevalence of resistance to azoxystrobin (measured as >75% growth at 10 ppm relative to control) in Central
and South America (Number of subpopulations tested given in brackets).
Country
Mexico
Belize
Honduras
Guatemala
Nicaragua
Costa Rica
Panama
Colombia
Ecuador
Cameroon
Prevalence of resistance (% of resistant spores)
1999
2000
0.0 (2)
0.0 (4)
-
0.0 (2)
0.0 (3)
0.0 (4)
11.3 (33)
0.0 (5)
0.0 (4)
-
2001
0.0 (30)
0.0 (25) *
0.0 (3)
11.1 (78)
3.3 (15)
0.03 (25)
0.01 (28)**
0.0 (12)
*Some resistant individuals were detected in two farms through PCR analysis.
**False positive (PCR analysis demonstrated that resistance was not present).
Molecular analysis
A high degree of correlation (P≤0.01) was observed between extensive germ tube
growth at 10 ppm of zoxystrobin and detection of the G143A resistance mutation
(Figure 1). The correlation between the PCR data and bioassay data at 1 ppm was
lower (0.8 compared to 0.94). In other words, extensive germ tube growth, normally
associated with resistance, is occasionally detected on 1 ppm agar in the absence
of G143A. This may indicate an alternative mechanism of resistance, and studies are
in progress to address this question.
Table 2 gives the prevalence of resistance in populations of M. fijiensis sampled
from 16 different sites in Costa Rica. Ten samples showed varying levels of G143A
detection, whilst 6 sample populations remained sensitive.
Conclusion
The ongoing population dynamics study will enable the validation or improvement
of the management guidelines regarding resistance to strobilurin, and will support
efforts to limit the proliferation of resistance to this important group of fungicides.
Fungicide programmes are expected to remain indispensable for commercial
banana production for the foreseeable future. A significant increase in the number
of alternative chemical modes of action to control of Sigatoka diseases is unlikely
within the next 5-10 years. Therefore resistance management will remain a prime
consideration in the design of control programmes, underpinned by appropriate
monitoring efforts and a judicious revision of the guidelines.
306
25/06/03
15:16
Page 307
Session 5
S. Knight et al.
100
Baseline
Farm 1
Farm 2
Farm 3
75
% of ascopores
MyLsd 17x24
PCR
25%
PCR
11%
50
PCR
1%
PCR
<0.01%
25
0
0
1
2
3
1 ppm
4
1
2 3 4 0 1 2 3
Germ tube length class
4 0
1
2
3
4
10 ppm
Figure 1. Correlation between in vitro growth in the presence of azoxystrobin and detection of
the G143A resistance mutation. (Germ tube length class: 0 means not germinated, 1 means <25%
growth relative to control, 2 means between 25 and 50% growth, 3 means between 50 and 75%
growth, and 4 means over 75% growth).
Table 2. Prevalence of resistance in populations of M. fijiensis estimated by determining the presence of the G143A
resistance mutation at sixteen different sites in Costa Rica.
Farm number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15 (untreated area)
16
Region
Prevalence of resistance (%)
Guápiles
Guápiles
Guápiles
Limón
Guápiles
Siquirres
Siquirres
Limón
Sarapiquí
Siquirres
Guápiles
Siquirres
Siquirres
Siquirres
Siquirres
Guápiles
1
3
25
0
27
0
2
0
6
0.6
5
0.6
0
0
0
11
307
MyLsd 17x24
25/06/03
15:16
Page 308
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Recommendations of session 5
Integrated disease management
Strategies to control black leaf streak disease and Sigatoka disease can, according to the
country and the scale of production, include not only chemical and cultural practices but
also the use of mixed crops or resistant clones. The important inhibitory effect of some natural
substances derived from microorganisms antagonistic to fungi, have also been reported as
effective in reducing, in vitro, the development of M. fijiensis.
It was recommended to integrate working groups from different disciplines to develop
an achievable IPM approach to manage Sigatoka diseases.
Chemical strategies and/or the use of improved hybrids should always be used jointly
with adequate agricultural practices to maximize yield and efficacy of the
management practices.
It was recommended to study natural/synthetic substances capable of promoting or
activating systemic acquired resistance in the broad sense.
It was recommended to evaluate the feasibility of precision agriculture farming to
optimize disease management.
It was recommended to assess different crop systems with potential positive impact
on disease management.
It was recommended to include the FRAC’s Banana Working Group Guidelines
(http://www.gcpf.org/frac) for fungicide resistance management in order to broaden
the knowledge of such guidelines.
It was recommended to develop alternative/improved methods/equipments for groundbased applications that can be used by smallholders.
308
MyLsd 17x24
25/06/03
15:16
Page 311
List of participants
Catherine Abadie*
CARBAP
BP 832 Douala
Cameroon
* Current address
CIRAD-FLHOR
Avenue Agropolis TA 40/02
34398 Montpellier Cedex 5
France
Tel.: +33 467616529
Fax: +33 467615793
E-mail: catherine.abadie@cirad.fr
Juan Fernando Aguilar
FHIA
P.O. Box 2067
San Pedro Sula
Honduras
Tel.: +504 6 682470
Fax: +504 6 682313
E-mail: jaguilar@fhia.org.hn
Maria Elena Aguilar
CATIE
Lab. de Biotecnología
Apdo 7170, Turrialba
Costa Rica
Tel.: +506 5566455
Fax: +506 5562626
E-mail: aguilarm@catie.ac.cr
Yelenis Alvarado Capó
Instituto de Biotecnología de las Plantas
Carretera a Camajuaní km 5
Santa Clara, Villa Clara
Cuba
Tel.: +53 42 281257
Fax: +53 42 281329
E-mail: yalvarado@uclv.edu.cu
Sergio Mauricio Aponte
CORPOICA
Km 14 via Mosquera
A.A. 240142
Las Palmas, Bogotá D.C.
Colombia
Tel.: +57 1 3443000
Fax: +57 1 3441435
E-mail: apontesergio@yahoo.com
Maria J. Barbosa Cavalcante
EMBRAPA
Rod. BR-364 Km 14
Caixa Postal 321
Rio Branco
Acre
Brazil
Tel.: +55 68 2123200
Fax: +55 68 2123200
E-mail: maju@cpafac.embrapa.br
Silvio Belalcázar Caravajal
INIBAP Honorary Research Fellow
Latin America and Caribbean Regional Office
C/o CATIE
7170 Turrialba
Costa Rica
Tel.: +506 5562431
Fax: +506 5562431
E-mail: belcar@armenia.multi.net.co
Peter Burt
Natural Resources Institute
University of Greenwich
Chatham, Kent ME4 4TB
United Kingdom
Tel.: +44 634 883231
Fax: +44 634 880066/77
E-mail: P.J.A.Burt@greenwich.ac.uk
Russell Caid
Consultant
Chiquita Brands International
Kentucky
USA
E-mail: RussCaid@aol.com
Jean Carlier
CIRAD-AMIS
Avenue Agropolis TA 40/02
34398 Montpellier Cedex 5
France
Tel.: +33 467616529
Fax: +33 467615793
E-mail: jean.carlier@cirad.fr
311
MyLsd 17x24
25/06/03
15:16
Page 312
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Arllen Carpio Brenes
EARTH
Apdo Postal 4442-1000
San José
Costa Rica
Tel.: +506 7 130000
Fax: +506 7 130001
E-mail: afcarpio@earth.ac.cr
Laura Conde
CICY
Calle 43
Col. Churburna de Hidalgo
CP 97200 Merida
Mexico
Tel.: +52 99 813914
Fax: + 52 99 813900
E-mail: laura@cicy.mx
Zilton Cordeiro
EMBRAPA
Rua EMBRAPA s/n
Caixa Postal 007
Cruz das Almas
44380-000 Bahia
Brazil
Tel.: +55 021 756212120
Fax: +55 021 756211118
E-mail: zilton@cnpmf.embrapa.br
Pedro Crous
Department of Plant Pathology
University of Stellenbosch
P.Bag X1
Matieland 7602
South Africa
Tel.: +27 218 084 796
Fax: +27 218 084 956
E-mail: PWC@sun.ac.za
Fritz Elango
EARTH
Apartado 4442-1000
San José
Costa Rica
Tel.: +506 7 130000
Fax: +506 7 130133
E-mail: felango@earth.ac.cr
Jean-Vincent Escalant
INIBAP
Parc Scientifique Agropolis II
34397 Montpellier Cedex 5
France
Tel.: +33 467611302
Fax: +33 467610334
E-mail: j.escalant@cgiar.org
Emily Fabregar
Lapanday Agricultural & Development
Corporation
Maryknoll Road
Bo Pampanga
Davao City
The Philippines
Tel.: +63 82 2352551
Fax: +63 82 2342359
E-mail: egf@skyinet.net
Henry Fagan
WIBDECO
Manoel Street, Compton Bldg
PO Box 115
Castrie, Saint Lucia
W.I.
Tel.: +1 758 4522411
Fax: +1 758 4514601
Pedro Ferreira
CATIE
7170 Turrialba
Costa Rica
Tel.: +506 5566431
Fax: +506 5561533
E-mail: CATIE@catie.ac.cr
Emile Frison
INIBAP
Parc Scientifique Agropolis II
Montpellier 34397 Cedex 5
France
Tel.: +33 467611302
Fax: +33 467610334
E-mail: e.frison@cgiar.org
José Garza
INIFAP
Campo Exp. Tecoman
Km 35 Carretera Colima Manzanillo
Colima, Mexico
Tel.: +31 33 240133
E-mail:
tecoman@cirpac.inifap.conacy.mx
312
MyLsd 17x24
25/06/03
15:16
Page 313
List of participants
Friedhelm Gauhl
Chiquita
PO Box 025216-1582
Miami, Florida
33102-5216
USA
Tel.: +1 506 2042001
Fax: +1 506 2042397
E-mail: fgauhl@chiquita.com
Kathy Grice
Centre for Tropical Agriculture
28 Peters St
Mareeba Qld 4880
Australia
Tel.: +61 740 928555
Fax: +61 740 923593
E-mail: Kathy.Grice@dpi.qld.gov.au
Mauricio Guzman
CORBANA SA
La Rita de Pococi
Limón
Costa Rica
Tel.: +506 7 633176
Fax: +506 7 633055
E-mail: mguzman@corbana.co.cr
Juliane Henderson
CRCTPP
Molecular Diversity
et Diagnostics Research Laboratory
80 Meiers Rd
Indooroopilly Qld 4068
Australia
Tel.: +61 738 969341
Fax: +61 738 969533
E-mail:
juliane.henderson@dpi.qld.gov.au
Luis Jacome
PATHOTEC, S.A.
Local # 5 Interior, Hotel Lima
La Lima, Cortes
Honduras
Tel.: +504 6685610
Fax: +504 6685613
E-mail: lhjacome@infovia.hn
Erwan Jade
Total Fina Elf
Special fluids
51, Esplanade du Général de Gaulle
La Défense 10
92907 Paris La Défense Cedex
France
Tel.: +33 141356123
Fax: +33 141355134
Email: Erwan.Jade@TotalFinaElf.com
Andrew James
Unidad de Biotecnología, División Biología
Vegetal
CICY
Calle 43 No. 130
Colonia Churburná de Hidalgo
CP 97200, Merida
Mexico
Tel.: +52 99 813914
Fax: +52 99 813900
E-mail: andyj007@cicy.mx
Ramiro Jaramillo
Honorary guest
Former INIBAP regional coordinator
Apdo 4824-1000
San José
Costa Rica
Christophe Jenny
CIRAD-FHLOR
Station de Neufchâteau
Sainte Marie
F-97130 Capesterre Belle-Eau
Guadeloupe
Tel.: +33 590861768
Fax: +33 590868077
E-mail: christophe.jenny@cirad.fr
David Jones
12 Charlotte Brontë Drive
Droitwich Spa
Worcestershire WR9 7HU
United Kingdom
Tel.: +44 190 4462098
Fax: +44 190 4462250
E-mail: bananadoctor@email.msn.com
313
MyLsd 17x24
25/06/03
15:16
Page 314
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Pedro E. Jorge
IDIAF
Rafael Augusto Sanchez #89
Santo Domingo
Dominican Republic
Tel.: +1 809 5678999
Fax: +1 809 5679199
E-mail: pejorge@hotmail.com
Dieter Kaemmer
CICY
Calle 43 No. 130
Colonia Churburná de Hidalgo
CP 97200
Mexico
Tel.: +52 99 813914
Fax.: +52 99 813900
E-mail: dieter@cicy.mx
Gert H.J. Kema
IPO-DLO
P.O.Box 9060
6700 GW Wageningen
The Netherlands
Tel.: +31 317 476149
Fax: +31 317 410113
E-mail: G.H.J.Kema@IPO.DLO.NL
Susan Knight
Syngenta
Centro de Ciencia y Tecnología Ultrapark
Edificio 7B Segundo Piso
La Aurora de Heredia
Heredia
Costa Rica
Tel.: +506 2 939500
Fax: +506 2 931628
E-mail: susan.knight@syngenta.com
Marc Henri Lebrun
UMR 1932 CNRS INRA
Aventis Cropscience
14-20 rue Pierre Baizet
B.P. 9163
69263 Lyon, Cedex 09
France
Tel.: +33 472852481
Fax: +33 472852297
E-mail: marc-henri.lebrun@aventis.com
Philippe Lepoivre
Unité de Phytopathologie
University of Gembloux
Passage des Déportés, 2
B-5030 Gembloux
Belgium
Tel.: +32 81 622437
Fax: +32 81 610126
E-mail: lepoivre.p@fsagx.ac.be
Ronald Madrigal Barrantes
EARTH University
Las Mercedes
Guácimo
Limón
Costa Rica
Tel.: +506 7 130000
Fax: +506 7 130005
E-mail: rmadriga@ns.earth.ac.cr
Douglas Marin
Delmonte
Apdo 4084-1000
San José
Costa Rica
Tel.: +506 7103674
Fax: +506 7103627
E-mail: marin.douglas@freshdelmonte.co.cr
Sergia Milagrosa
Del Monte Fresh
Davao City
Philippines
Tel.: +63 82 2331838
Fax: +63 82 2340438
E-mail: SMilagrosa@freshdelmonte.com.ph
Gus Molina
INIBAP
Asia-Pacific Regional Office
c/o IRRI Rm 31
Khush Hall
Los Baños, Laguna 4031
Philippines
Tel.: +63 2 8450563
Fax: +63 2 8911292
E-mail: a.molina@cgiar.org
314
MyLsd 17x24
25/06/03
15:16
Page 315
List of participants
Xavier Mourichon
CIRAD-AMIS
Av. Agropolis TA 40/02
34398 Montpellier Cedex 5
France
Tel.: +33 467615869
Fax: +33 467615581
E-mail: xavier.mourichon@cirad.fr
Luis Pérez Vicente
INISAV
Gaveta 634
11300, Playa
Ciudad Habana
Cuba
Tel.: +537 8 782 420
Fax: +537 2 405 35
E-mail: lperez@inisav.cu
Juan Luis Ortiz
CATIE
7170 Turrialba
Cartago
Costa Rica
Tel.: +506 5 566455
E-mail: jortiz@catie.ac.cr
Claudine Picq
INIBAP
Parc Scientifique Agropolis II
Montpellier 34397 Cedex 5
France
Tel.: +33 467611302
Fax: +33 467610334
E-mail: c.picq@cgiar.org
Rodomiro Ortiz
IITA
Carolyn House 26
Dingwall Road
Croydon CR9 3EE
Nigeria
Tel.: +234 2 2412626
Fax: +234 2 2412221
E-mail: r.ortiz@cgiar.org
Luis Pocasangre
INIBAP
Latin America and Caribbean Regional Office
C/o CATIE
7170 Turrialba
Costa Rica
Tel.: +506 5562431
Fax: +506 5562431
E-mail: lpoca@catie.ac.cr
Luis Fernando Patiño
AUGURA
CENIBANANO
Calle 3 Sur No. 41-65, Edif. Banco de Occidente
Medellin
Colombia
Tel.: +57 4 3211333
Fax: +57 4 8236606
E-mail: lpatino@augura.com.co
Leticia Peraza
CICY
Calle 43 No. 130
Colonia Churburná de Hidalgo
CP 97200
Mexico
Tel.: +52 99 813966
Fax: +52 99 813900
E-mail: latyperaza@yahoo.com
Tania Polanco
IDIAF
Rafael Augusto Sanchez #89
Santo Domingo
Dominican Republic
Tel.: +809 5678999
Fax: +809 5679199
E-mail: tpolanco@hotmail.com
Vivencio L. Quiñon
TADECO
A.O. Floriendo
Papabo City
Davao
Philippines
Tel.: +63 84 8220541
Edwin Raros
Dole Asia Research
c/o Stanfilco
Davao City
Philippines
315
MyLsd 17x24
25/06/03
15:16
Page 316
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Nicolas Roux*
IAEA Plant Breeding Unit
A-2444 Seibersdorf
Austria
Galileo Rivas
CATIE
7170 Turrialba
Costa Rica
Tel.: +506 5560232
Fax: +506 5566480
E-mail: grivas@catie.ac.cr
*Current address
INIBAP
Parc Scientifique Agropolis II
34397 Montpellier Cedex 5
France
Tel.: +33 467611302
Fax: +33 467610334
E-mail: n.roux@cgiar.org
Mauricio Rivera Canales
FHIA
Apdo Postal 2067
San Pedro Sula
Honduras
Tel.: +504 6682470
Fax: +504 6682313
E-mail: mrivera@fhia.org.hn
Laszlo Sági
KULeuven
Kasteelpark Arenberg 13
B-3001 Leuven
Belgium
Tel.: +32 16 321681
Fax: +32 16 321993
E-mail: laszlo.sagi@agr.kuleuven.ac.be
Alba Stella Riveros Angarita
UTOLIMA - CATIE
Unidad de Fitoprotección
7170 Turrialba
Costa Rica
Tel.: +506 5566021
Fax: +506 5560606
E-mail: asrivero@catie.ac.cr
Jorge Sandoval
CORBANA
Apdo 390-7210
Guápiles, Limón
Costa Rica
Tel.: +506 7633176
Fax: +506 7633055
E-mail: jsandoval@corbana.co.cr
Ronald Romero
Chiquita
Building D, FORUM
Santa Ana
Costa Rica
Tel.: +506 2 042 001
Fax: +506 2 042 399
E-mail: rromero@chiquita.com
Hans Sauter
Edificio Del Monte
200 Metros al Este del Periódico La Republica
Barrio Tournon
San Jose
Costa Rica
Fax: +506 222-9769
E-mail:
sauter.hans@freshdelmonte.co.cr
Franklin Rosales
INIBAP
Latin America and Caribbean
Regional Office
C/o CATIE
7170 Turrialba
Costa Rica
Tel.: +506 5562431
Fax: +506 5562431
E-mail: frosales@catie.ac.cr
R. Selvarajan
NRCB
44 Ramalinga Nagar
Vayalur Road
Trichy 620 017
India
Tel.: +91 431 771299
Fax: +91 431 770564
E-mail: selvarama@yahoo.com
316
MyLsd 17x24
25/06/03
15:16
Page 317
List of participants
Ronald Vargas
CORBANA
Diagonal a la Casa Presidencial
A.A. 6504-1000
San José
Costa Rica
Tel.: +506 7633176
Fax: +506 7633055
E-mail: rvargas@corbana.co.cr
Moises Soto Ballestero
Corporación S y M SA
Apartado Postal 479-1007
San José
Costa Rica
Tel.: +506 336670
Fax: +506 2904941
E-mail: agbenig@racsa.co.cr
Egbert Spaans
EARTH
Apdo 4442-1000
San Jose,
Costa Rica
Tel.: +506 7 130000
Fax: +506 7 130001
E-mail: espaans@earth.ac.cr
Altus Viljoen
FABI
University of Pretoria
Pretoria 0002
South Africa
Tel.: +27 12 4203856
Fax: +27 12 4203960
E-mail: aviljoen@postino.up.ac.za
Rony Swennen
KULeuven
Laboratory of Tropical Crop Improvement
Kasteelpark Arenberg 13
3001 Leuven
Belgium
Tel.: +32 16 321420
Fax: +32 16 321993
E-mail: Rony.Swennen@agr.kuleuven.ac.be
Manuel Wirz
Syngenta
Centro de Ciencia y Tecnología Ultrapark
Edificio 7B Segundo Piso
La Aurora de Heredia
Heredia
Costa Rica
Tel.: +506 2939500
Fax: +506 2931628
E-mail: manuel.wirz@syngenta.com
Ana Cecilia Tapia
CATIE
Apdo 7170, Turrialba
Costa Rica
Tel.: +506 5560232
Fax: +506 556 6480
E-mail: tapiaa@catie.ac.cr
317
24/06/03
9:50
Page 1
Mycosphaerella leaf spot diseases of bananas: present status and outlook
Mycosphaerella leaf spot
diseases of bananas:
present status
and outlook
Proceedings of the 2nd International workshop
on Mycosphaerella leaf spot diseases held in San José,
Costa Rica, 20-23 May 2002
L. Jacome, P. Lepoivre, D. Marin, R. Ortiz, R. Romero
and J.V. Escalant, editors
isbn 2-910810-57-7
View publication stats
Couv MEP OK