Mycosphaerella leaf spot diseases of bananas - CBS

Mycosphaerella leaf spot 

diseases of bananas: 

present status 

and outlook 

Proceedings of the 2 nd 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

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 

INIBAP 

Via dei Tre Denari 472/a 

Parc Scientifique Agropolis II 

000 57 Maccarese (Fiumicino) 34397 Montpellier Cedex 5 

Rome, Italy 

France

Mycosphaerella leaf spot 

diseases of bananas: 

present status 

and outlook 

Proceedings of the 2 nd 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

Acknowledgments 

INIBAP would like to thank all those who helped in the organization of the 2 nd 

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

Contents 

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 

Introduction to workshop 

Overview of progress and results since the first international workshop 

on Mycosphaerella leaf spot diseases of bananas in 1989 

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 


5

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 


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 


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 


List of participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 

6

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 1 st 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 2 nd 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

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.


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


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. musae. 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 popu- 

12


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


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


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


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


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 73:1328- 

1337. 

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


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. 2 nd 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

Session 1 

Impact of Mycosphaerella 

leaf spot diseases of bananas

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


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

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


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

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,itis 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


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 20 th 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

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 20 th 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


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 Year Region/Country Year 

Australasia/Oceania 

Fiji 1912 

Australia (Queensland) 1924 

Australia (New South Wales) 1927 

Solomon Islands 1946 

Papua New Guinea 1951 

New Caledonia 1951 

Indonesia (Irian Jaya) 1953 

Norfolk Island 1954 

Wallis Island 1954 

USA (Hawaii) 1958 

Samoa 1961 

American Samoa 1961 

French Polynesia 1962 

Tonga 1965 

Asia 

Indonesia (Java) 1902 

Sri Lanka 1919 

Philippines 1921 

Malaysia (Peninsula) 1933 

China 1936 

India (Assam) 1946 

Taiwan 1953 

Malaysia (Sabah and Sarawak) 1959 

Cambodia 1960 

Thailand 1962 

Hong Kong 1966 

Vietnam 1966 

Laos 1967 

Brunei 1968 

Latin America/Caribbean 

Guadeloupe 1932 ? 

Surinam 1933 

Trinidad 1933 

Guyana 1935 

Honduras 1935 

Jamaica

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


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

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


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 Year 1 Region/Country Year 1 

Australasia/Oceania 

Solomon Islands 1957 (1946) 

Papua New Guinea 1957 (1951) 

Fiji 1963 

French Polynesia 1964-67 

Micronesia 1964-67 

New Caledonia 1964-67 

Vanuatu 1964-67 

Tonga 1965 

Samoa 1965 

USA (Hawaii) 1969 (1958) 

American Samoa 1975 

Cook Islands 1976 

Niue 1976 

Norfolk Island 1980 

Australia (Torres Strait/Cape York) 1981 

Wallis and Fortuna Islands 1996 

Australia (North Queensland, 

currently under eradication) 2001 

Asia 

Taiwan 1927 

Philippines (Luzon) 1964 

Singapore 1964-67 

Philippines (Mindanao) 1965 

Malaysia (Peninsula) 1965 

Thailand 1969 

Indonesia (Java) 1969 

Indonesia (Halmahera) 1970 

China (Hainan) 1980 

Bhutan 1985 

China (Guangdong) 1990 

China (Yunnan) 1993 

Vietnam 1993 

Indonesia (Sumatra) 1993 

Indonesia (Kalimantan) 1996 

East Malaysia (Sarawak) 1996 

Latin America/Caribbean 

Honduras 1972 (1969) 

Belize 1975 

Guatemala 1977 

Nicaragua 1979 

Costa Rica 1979 

El Salvador 1979 2 

Mexico 1980 

Panama 1981 

Colombia 1981 

Ecuador 1986 

Cuba 1990 

Venezuela 1991 

Peru 1994 

Jamaica 1995 

Bolivia 3 1996 

Dominican Republic 1996 

Brazil 1998 

USA (Florida) 1998 

Haiti 1999 

Africa 

Zambia 1973 4 

Gabon 1978 

Cameroon (south-east) 1980 

Cameroon (south-west) 1983 

São Tomé 1983 

Côte d’Ivoire 1985 

Congo 1985 

Nigeria 1986 

Ghana 1986 

Rwanda 1986 

Burundi 1987 

Tanzania (inc. Pemba and Zanzibar) 1987 

Democratic Republic of Congo 

(highlands) 1987 

Democratic Republic of Congo 

(lowlands) 1988 

Togo 1988 

Kenya 1988 

Malawi 1990 

Uganda 1990 

Benin 1993 

Comoros 1993 

Mayotte 1993 

Central African Republic 1996 

Madagascar 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

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


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

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


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

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 Banana host Year specimen was 

collected and collector 

Nigeria (Onne) AAB clone, most likely in plantain subgroup 1989 (IITA) 

Nigeria (Onne) AAB clone, most likely in plantain subgroup 1990 (IITA) 

India (Bangalore) ‘Grande naine’ (AAA, Cavendish subgroup) 1992 (D.R. Jones, INIBAP) 

Malaysia (Johor State) ‘Pisang kapas’ (AA or AAB) 1993 (D.R. Jones, INIBAP) 

Thailand (Sukothai) ‘Grande naine’ (AAA, Cavendish subgroup) 1994 (D.R. Jones, INIBAP) 

Thailand (Surat Thani) ‘Williams’ (AAA, Cavendish subgroup) 1994 (D.R. Jones, INIBAP) 

Thailand (Tha Yang) ‘Kluai hom thong’ (AAA) 1994 (D.R. Jones, INIBAP) 

India (Kannara) ‘Grande naine’ (AAA, Cavendish subgroup) 1995 (D.R. Jones, INIBAP) 

Sri Lanka (Gannoruwa) AAA clone in Cavendish subgroup 1995 (D.R. Jones, INIBAP) 

Sri Lanka (Nugahena) ‘Anamala’ (AAA, syn. ‘Gros Michel’) 1995 (D.R. Jones, INIBAP) 

Vietnam (Mekong Delta) ‘Sucrier’ (AA) 1995 (I. Buddenhagen) 

Mauritius ‘Grande naine’ (AAA, Cavendish subgroup) 1997 (S.P. Beni Madhu, AREU) 

Réunion ‘Grande naine’ (AAA, Cavendish subgroup) 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


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

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 



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


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:D60- 

64. 

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. 


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

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

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. Crous 1 ,J.Z. Groenewald 1 ,A.Aptroot 2 ,U.Braun 3 , 

X. Mourichon 4 and J. Carlier 4 

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 

2 

Centraalbureau voor Schimmelcultures,Utrecht, The Netherlands 

3 

Martin-Luther-Universität, Halle (Saale), Germany 

4 

CIRAD, Montpellier, France 

43


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

Session 1 


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


• 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

Session 1 


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


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 Origin 

U04234 Leptosphaeria microscopica Unknown ATCC 42652 (Saccharum 

officinarum, Kenya) 

PCR18 Mycosphaerella cruenta Pseudocercospora cruenta ATCC 262271 (Vigna,Puerto Rico) 

PP15 M. berkeleyi Passalora personata MPPD L2121 (Arachis, 

Oklahoma, U.S.A.) 

458 M. eumusae Pseudocercospora eumusae Musa, Malaysia 

487 M. eumusae Pseudocercospora eumusae Musa, Thailand 

PF7 M. fijiensis Paracercospora fijiensis ATCC 22116 (Musa, Philippines) 

PF8 M. fijiensis Paracercospora fijiensis ATCC 22117 (Musa, Hawaii) 

01A M. fijiensis Paracercospora fijiensis Musa, Philippines 

009 M. fijiensis Paracercospora fijiensis Musa, Ntoum, Gabon 

PFD9 M. fijiensis var. difformis Paracercospora fijiensis ATCC 36054 (Musa, Honduras) 

var. difformis 

PM10 M. musicola Pseudocercospora musae ATCC 22115 (Musa,Philippines) 

PM11 M. musicola Pseudocercospora musae ATCC 36143 (Musa, Honduras) 

121 M musicola Pseudocercospora musae Musa, Indonesia, West Java 

090 M musicola Pseudocercospora musae Musa, Armenia, Colombia 

CA1 Mycosphaerella Cercospora apii ATCC 12246 

state unknown 

CH5, CH6 Mycosphaerella Cercospora hayi ATCC 12234 

state unknown 

(Musa, Cuba) 

MA12 Unknown Mycocentrospora acerina ATCC 34539 (Daucus carota, 

Norway) 

IMI 271341 Unknown Phaeoseptoria musae 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

Session 1 


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


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 

100 

Phaeoseptoria musae IMI271341 

Leptosphaeria microscopica U04234 

M. musae 122 

Mycosphaerella musae 

100 

M. berkeleyi PP15 

C. hayi CH5 

63 

100 

61 

C. apii CA1 

Cercospora apii sensu lato 

C. hayi CH6 

69 

M. cruenta PCR18 

100 

Ps. musae PM11 

79 

99 

89 

88 

Ps. musae PM10 

M. musicola 121 

M. musicola 090 

Mycosphaerella musicola 

100 

51 

M. eumusae 458 

M. eumusae 487 

Mycosphaerella eumusae 

92 

M. fijiensis 01A 

10 changes 

54 

74 

M. fijiensis 009 

Pa. fijiensis PF8 

Pa. fijiensis var. 

difformisPFD9 

Pa. fijiensis PF7 

Mycosphaerella fijiensis 

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

Session 1 


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


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

Session 1 


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


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

Session 1 


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


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 



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

Session 1 


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

Session 1 

J. Henderson et al. 

Improved PCR-based detection of 

Sigatoka disease and black leaf streak 

disease in Australian banana crops 

J. Henderson 1 ,K.Grice 2 ,J.Pattemore 1 ,R.Peterson 2 and E. Aitken 1 

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 

Cooperative Research Centre for Tropical Plant Protection, Indooroopilly, Australia 

2 

Queensland Department of Primary Industries, Mareeba, Australia 

59


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

Session 1 


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


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 disease 1 . 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

Session 1 


Sigatoka disease 

MMfor 

R635 

ITS1 

ITS2 

1 8S (partial) 5.8S 25S (partial) 

MFfor 

R635mod 


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 post- 

PCR 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 studied 2 . 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


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

Session 1 

A. Viljoen et al. 

Impact of minor Mycosphaerella 

pathogens on bananas (Musa) in 

South Africa 

A. Viljoen 1 ,A.K.J. Surridge 1 and P.W. Crous 2 

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 

University of Pretoria, Pretoria, South Africa 

2 

University of Stellenbosch, Matieland, South Africa 

65


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

Session 1 


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 19 th 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


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 18 o C and 21 o C 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 

25 o and 30 o latitude and 30 o and 32 o 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 18 o C, which is 

necessary for the production of conidia. Minimum night temperatures exceed 21 o C 

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

Session 1 


35 

140 

30 

120 

25 

100 

Temperature 

20 

15 

10 

5 

0 

January 

February 

March 

April 

May 

June 

July 

August 

September 

October 

November 

December 

80 

60 

40 

20 

0 

Rainfall 

Ave. min. temperature (°C) Ave. max. temperature (°C) 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 18 o C 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


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. 



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

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


(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,‘FHIA- 

18’,‘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

Session 1 


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


800 

700 

600 

500 

400 

300 

200 

100 

50 

0 

Banao, 1984 

Artemisa, 1987 

Sagua, 1989 

Sagua, 1990 

Contramaestre, 1990 

Protection cost 

in US dollars per hectare 

Horquita, 1990 

La Cuba, 1991 

La Cuba, 1992 

Limoncito, 1994 

Menendez, 1994 

Menendez, 1995 

Nueva Paz, 1994 

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) 

40 

30 

20 

0 

Cavendish (AAA) Plantains (AAB) ABB cultivars FHIA cultivars 

1990 1995 1997 1999 2000 

Figure 2. Surface area planted with various types of banana and plantain in Cuba, 1990-2000. 

74

Session 1 


2500 

12 

Cost of fungicides (x 1000 US dollars) 

2000 

1500 

1000 

10 

8 

6 

4 

2 

Planted area (X 1000 ha) 

0 

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 

0 

Cost of fungicides 

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


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 Independent 

variable variable 0 1 2 3 4 5 6 7 

SD4L Rf7mm n.s. 0.34* n.s. 0.56 *** 0.61 *** 0.74 *** 0.41 * n.s. 

RfD7 min n.s. n.s. n.s. 0.45 ** 0.48 ** 0.77 *** 0.52 ** n.s. 

Rf10 mm n.s. n.s. 0.38 * 0.54 *** 0.80 *** 0.71 *** n.s. n.s. 

RfD10 min n.s. n.s. 0.37 * 0.39 * 0.74 *** 0.73 *** n.s. n.s. 

Rf14 mm n.s. n.s. 0.45 ** 0.64 *** 0.77 *** 0.69 *** n.s. n.s. 

RfD14 min n.s. n.s. 0.38 * 0.51 ** 0.75 *** 0.73 *** n.s. n.s. 

H7 n.s. n.s. n.s. n.s. 0.72 ** 0.71 ** 0.53 * n.s 

H10 n.s. n.s. 0.55* n.s. 0.79** 0.70 ** n.s. n.s. 

H14 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. 

RH>95% n.s. n.s. n.s. 0.316* 0.276* 0.308* 0.36** 0.36** 

PwEv -0.27 * -0.29 * n.s. n.s. n.s. n.s. n.s. n.s. 

TMed 0.32 * n.s. 0.29 * 0.29 * n.s. n.s. 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 Independent 

variable variable 0 1 2 3 4 5 6 7 

SD4L Rf7 mm n.s. n.s. n.s. 0.46 ** 0.48 ** 0.60 *** 0.58 *** 0.55 ** 

RfD7 min n.s. n.s. n.s. 0.34 * 0.39 * 0.64 *** 0.55 ** 0.54 ** 

Rf10 mm n.s. n.s. n.s. 0.51 ** 0.59 *** 0.74 *** 0.62 *** 0.40 * 

RfD10 min n.s. n.s. n.s. 0.40 * 0.50 ** 0.74 *** 0.57 *** 0.38 * 

Rf14 mm n.s. n.s. n.s. 0.52 ** 0.65 *** 0.74 *** 0.64 *** 0.39 * 

RfD14 min n.s. n.s. n.s. 0.41 * 0.58 *** 0.72 *** 62 *** 0.38 * 

H 7 

(mm) n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. 

H 10 

(mm) 0.49* 0.50* n.s. n.s. 0.73** n.s. n.s. n.s. 

H 14 

(mm) 0.57** 0.57** 0.58* n.s. 0.81** 0.55* n.s. n.s. 

PwEv -0.44 ** -0.58 *** -0.60*** -0.70*** -0.74*** -0.75*** -0.52** n.s. 

TMed n.s. n.s. n.s. 0.40 * 0.45 ** 0.48 ** 0.51 ** 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

Session 1 


2500 

2000 

Cumulated duration of rain over a 14-day period 

2000 

1500 

1000 

500 

Duration of rainfall 

State of development 

of the disease 4 weeks later 

R 2 = 0.56* 

1750 

1500 

1250 

1000 

750 

500 

250 

State of development of the disease 

47 49 51 1 3 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). 

Cumulated duration of rain over a 14-day pediod (min) 

2500 

2250 

2000 

1750 

1500 

1250 

1000 

750 

500 

250 

r = 0.72*** 

Duration of rainfall 

State of development 

of the disease 4 weeks later 

48 50 52 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 

Weeks 

2750 

2500 

2250 

2000 

1750 

1500 

1250 

1000 

750 

500 

250 

0 

State of development of disease 

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


The curves of the observed and the estimated state of development of the 

disease using the model SD4L = 6.24 (S 14d 

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) 

3000 

2500 

2000 

1500 

1000 

500 

R2 = 0.85* 

Observed state of development 

Estimated state of development 

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 (S 14d 

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

Session 1 



1200 

1000 

800 

600 

400 

200 

0 

1 5 9 13 17 21 25 33 37 41 45 49 53 

16 

14 

12 

10 

8 

6 

4 

2 

0 

Youngest leaf spotted (YLS) 

WEEKS 

SD4L SD4L YLS 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). 


3000 

2500 

2000 

Pr 

1500 

1000 

500 

0 

33 36 

Te 

Te 

Tr 

Be 

Oil 

39 42 45 48 51 2 5 8 

Weeks 

Bi 

Oil 

11 14 17 20 23 26 29 32 

14 

12 

10 

8 

6 

4 

2 

0 

Youngest leaf spotted (YLS) 

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


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 21 801.24 

1992 23 619.66 

1993 15 568.74 

1994 13 303.29 

1995 12 299.33 

1996 12 269.52 

Limoncito 1994 22 476.19 

1995 13 246.48 

1996 13 288.72 

Quemado 1995 8 172.39 

De Guines 1996 4 71.21 

up to June 

Menendez 1994 18 412.09 

1995 23 518.49 

Sola 1994 11 221.56 

1995 12 282.78 

1996 11 237.31 

Nueva Paz 1994 23 599.66 1996 13 326.56 

Guines 1995 13 308.96 

1996 10 219.78 

(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

Session 1 


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 Incubation period Transition period Number of functional 

(days) (days) leaves at harvest 

February June February June Mother First 

plant follower 

Spot size 

(mm) 

Number of 

pseudothecia 

FHIA-1-1 43.5 b 28.2 a* 76.4** 75.5 9 8 17.3 15.9 

FHIA-02 46.9 b 31.0 a >150** 86.6 8 10 13.3 34.8 

FHIA-03 60.4 a 24.1 b >150** 107.7 10 8 15.5 31.6 

FHIA-18 52.8 ab 28.7 a >150** 119.0 12 9 14.3 35.0 

SH 3436 35.8 c 28.0 a 84.3* 80.2 10 7 12.7 9.5 

Grande naine 27.9 c 16.7 c 36.6 43.0 1 0 17.5 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 20.3 6.8 

La Cuba, 2001 16.3 7.8 

Alquízar, 2001 21.7 6.5 

FHIA-18 La Cuba, 1996 2.1 11.0 

La Cuba, 2000 (Fca. Berlier) 28.2 5.6 

La Cuba, 2000 (Fca. Cozola) 23.2 6.4 

La Cuba, 2001 10.3 7.6 

Alquizar, 2001 14.2 10.6 

Baracoa, 2002 15.8 9.6 

Cavendish La Cuba, 1996 35.6 4.3 

La Cuba, 2001 - - 

Alquizar, 2001 36.3 5.3 

Baracoa, 2002 59.8 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


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 K/(K+Ca+Mg) 

(meq/100g soil) 

El Berlier 5.6 0.44 1/59 

Cozola 6.4 0.52 1/56 

El Transformador (Cuba 3) 7.0 0.65 1/38 

El Colorado (Farm 10) 8.0 0.85 1/39 

La pista (Tin) 10.5 1.5 1/14 

Cooperative 11.2 1.2 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

Session 1 


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


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 6 th 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 6 th 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

Session 1 

A.B. Molina and E. Fabregar 

Management of black leaf streak 

disease in tropical Asia 

A. B. Molina 1 and E. Fabregar 2 

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 and the Pacific, Los Baños, Philippines 

2 

Lapanday Fruits Development Corporation, Philippines 

85


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

Session 1 


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. 

Lakatan 

13% 

Latundan 

8% 

Bungulan 

5% 

Others 

3% 

Saba 

39% 

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


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 0 0 4 6 6 

Triazole 9 11 8 8 8 

Dithane/Vondoze/Maneb 4 4 11 12 12 

Daconil 11 8 7 6 5 

Calixin 7 9 8 8 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 Azoxystrobin 0.4 L 

Baycor 300 EC Bitertanol 0.5 L 

Bumper 25 EC Propiconazole 0.4 L 

Calixin 75 EC Tridemorph 0.6 L 

Folicur 250 EC Tebuconazole 0.23 L 

Indar 2F Fenbuconazole 0.4 L 

Siganex Pyrazophos 0.5 L 

Sico 250 EC Propiconazole 0.4 L 

Tega 075 EC Trifloxystrobin 1.0 L 

Tilt 250 EC Propiconazole 0.4 L 

Protectant 

Daconil 720 SC Chlorothalonil 1.38 L 

Dithane F448 Mancozeb 4.0 L 

Dithane M-45 Mancozeb 2.5 kg 

Dithane OS Mancozeb 1.75 L 

Vondozeb 42 SC Mancozeb 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

Session 1 


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


120 

100 

80 

Million boxes 

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

Session 1 

Z.J. Maciel Cordeiro and A. Pires de Matos 

Impact of Mycosphaerella spp. 

in Brazil 


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


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

Session 1 


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


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

Session 1 


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


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 Number contracted Total number 

sold by 2001 from 2002 of plantlets 

Thap maeo (AAB) 301 968 350 000 651 968 

Caipira (AAA) 648 100 200 000 848 100 

FHIA-01 (AAAB) 42 000 - 42 000 

FHIA-18 (AAAB) 163 735 300 000 463 735 

SH36-40 (AAAB) 128 150 - 128 150 

FHIA-21 (AAAB) 9 000 - 9 000 

FHIA-03 (AABB) 5 900 - 5 900 

FHIA-10 3 000 - 3 000 

Ouro (AA) 4 650 - 4 650 

Prata zulu (ABB) 3 000 150 000 150 000 

PV03-44 (AAAB) 74 500 - 74 500 

PV42-85 (AAAB) 50 - 50 

Total 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

Session 1 


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

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


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

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 21 

Alternaria cf. citri 3 

Alternaria tenuissima 4 

Bipolaris cynodontis 3 

Cladosporium musae 5 

Colletotrichum gloeosporioides 5 

Colletotrichum musae 1 

Cordana musae 30 

Curvularia lunata 1 

Curvularia pallescens 1 

Diaporthe sp. 3 

Drechslera dematoidea 1 

Drechslera phlei 1 

Epicoccum nigrum 3 

Exserohilum rostratum 1 

Harpographium sp. 1 

Mycosphaerella musae 30 

Mycosphaerella musicola 66 

Myrothecium verrucaria 1 

Nigrospora oryzae 81 

Nigrospora sacchari 10 

Nigrospora sphaerica 25 

Pestalotiopsis guepinii 9 

Phoma glomerata 7 

Phyllosticta sp. 1 

Pithomyces sacchari 1 

Selenophoma asterina 10 

Selenophoma juncea 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


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. 



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

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


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

Session 2 

Population biology 

and epidemiology

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


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

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


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

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


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

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 meas- 

113


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 

N 

Q2 

Q3 

R3/4 

X 

R7/8 

R5/6 

Q1 

Q4 

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

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


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 Ascospores Conidia 

to the canopy 

Mean spore count Number Mean spore count Number of 

(min-max) of samples (min-max) samples 

Above 1.35 (0-6) 40 5.75 (0-54) 40 

Middle 1.15 (0-63) 39 6.59 (0-63) 39 

1.18 (0-5) 40 7.65 (0-44) 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

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). 

Catch 

180 

150 

120 

90 

60 

30 

0 

R1&2 (centre) R3&4 (Q2/Q3) R5&6 (Q4) R7&8 (Q1) 

Trap number and quadrat 

Ascospores 

Conidia 

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 

150 

100 

50 

Q1 

Q2 

Q3 

Q4 

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


(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 Mean daily catched of conidia n 

(standard error of the mean) 

(standard error of the mean) 

Trap 1 153.5 72 116 

(14.6) (12.1) 

Trap 2 124.2 8.5 127 

(17.4) (1..2) 

Trap 3 144.6 8.2 118 

(16.4) (1..2) 

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 cm 2 . 

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

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. 


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


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 Sigatoka- 

Krankheit (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. 


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

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

Session 2 

J. Carlier et al. 

Genetic differentiation in 

Mycosphaerella leaf spot pathogens 

J. Carlier 1 ,H.Hayden 2 ,G.Rivas 3 ,M.-F.Zapater 1 ,C.Abadie 1 and E. Aitken 4 

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 


2 

Institute of Land and Food Resources, Victoria, Australia 

3 

CATIE, Costa Rica 

4 

CRCTPP, Brisbane, Australia 

123


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

Session 2 


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


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 Latin America and Caribbean Australia and Pacific 

M. fijiensis 0.57 0.25 0.40 0.28 

M. musicola 0.41 0.27 0.21 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

Session 2 


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). 

Africa 

Latin America 

Latin America/ 

Caribbean 

Indonesia 

Pacific 

Islands 

Africa 

A 

Philippines 

Papua New Guinea 

B 

Australia 

Figure 1. Global population structure of M. fijiensis (Carlier et al., 1996) and M. musicola (Hayden etal., 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


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

Session 2 


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. 


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

Session 2 

C. Molina et al. 

Development and application of 

molecular markers in Mycosphaerella 

populations in Colombia 

C. Molina 1,2 , S. Aponte 1 ,A.Gutiérrez 1 ,V.Núñez 1 and G. Kahl 2 

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 

CORPOICA, Bogotá, Colombia 

2 

University of Frankfurt, Frankfurt/Main, Germany 

131


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

Session 2 


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. 


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


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

Session 2 


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


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). 

Dice similarity index 

0.00 0.25 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

Session 2 


Dice similarity index 

0.00 0.25 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-Ven- 

Md-20) is in a separate group, with Colombian isolates from La Mesa (cluster II). M. 

137


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 mid- 

Andean 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. 


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

Session 2 


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 



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. 

2 nd 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

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


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

Session 2 


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 26 o C 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. 


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


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 5 

2200 

+ 

1600 

1900 

1500 

1300 

1125 

1200 

1070 

1050 

1020 

945 

825 

785 

750 

680 

610 

450 

365 

285 

225 

830 

740 

670 

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

Session 2 


1 2 3 4 5 

5.7 

4.6 

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


1 2 3 4 5 

2200 

1400 

2200 

+ 

1600 

1055 

1125 

945 

805 

785 

750 

715 

1020 

945 

825 

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

Session 1 



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


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

Session 3 

Host-pathogen interactions

Session 3 

P. Lepoivre et al. 

Introduction 

Banana–Mycosphaerella fijiensis 

interactions 

P. Lepoivre 1 ,J.P.Busogoro 1 ,J.J.Etame 1 ,A.El Hadrami 1,2 , 

J. Carlier 2 ,G.Harelimana 1 ,X.Mourichon 2 ,B.Panis 3 , 

A. Stella Riveros 4 ,G.Sallé 5 ,H.Strosse 3 and R. Swennen 3 

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 

2 


3 

Katholieke Universiteit Leuven, Leuven, Belgium 

4 

CATIE, Costa Rica 

5 

Université Pierre et Marie Curie, Paris, France 

151


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

Session 3 


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


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

Session 3 


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


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

Session 3 


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 3 ème 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


(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 



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

Session 3 


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

Session 3 

C. Abadie et al. 

Efficiency and durability 

of partial resistance against 

black leaf streak disease 

C. Abadie 1,2 ,A.Elhadrami 2 ,E.Fouré 1 and J. Carlier 2 

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 

2 


161


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

Session 3 


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


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 

Penetration 

First lesions 

Sporulating lesions 

with conidia 

Necrosis 

Sporulating necrosis 

with ascospores 

Epiphyl phase 

Asexual 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

Session 3 


Size of lesion 

2 

(% of a 25-cm piece of leaf) 

14 

12 

10 

8 

6 

4 

2 

a 

de 

d 

Controlled conditions 

b 

e 

ef 

c 

fg 

ef 

g 

0 

CAV 

a 

Field conditions 

bc 

bc 

ab 

c 

ab 

c 

CAV 

FSO 

PMA 

PBE 

ZEB 

FOU 

PCE 

PMA 

ROS 

PBE 

TDE 

PKW 

TAN 

ZEB 

FOU 

PCE 

S 

PR 

30 

Size of lesion 

(% of spotted area) 

25 

20 

15 

10 

5 

0 

S 

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/cm 2 of necrosis than ‘Pisang Berlin’. 

165


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’). 

1600 

a 

Field conditions 

2 

Number of perithecia/cm of necrosis 

1200 

800 

400 

bc 

bc 

b 

bc 

c 

c 

0 

CAV 

FSO 

PMA 

PBE 

ZEB 

FOU 

PCE 

S 

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

Session 3 


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) 

1 st cycle 2 nd cycle 3 rd cycle 

Grande naine (S) 37.5 a 39 a 36 a 

Pisang Berlin (PR) 22.7 b 18.6 b 17.2 b 

Pisang madu (PR) 17.5 c 7.1 c 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


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. 


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. 7 th 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 



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

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:‘FHIA- 

18’, ‘FHIA-01’, ‘FHIA-21’, ‘Grande naine’, ‘Yangambi’, ‘Calcutta 4’ and ‘Niyarma yik’ were tested in a 

greenhouse. Inoculum was adjusted to 10 5 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


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 10 5 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 10 5 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 5 1/2 ’). 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

Session 3 


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 

28 o C 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 CaCO 3 

, 800 ml 

water, pH 6.0). The flask was incubated at 28 o C 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 28 o C for 15 days. Two 1-cm 2 discs of mycelium were 

transferred to a 1-litre flask containing 400 ml of liquid V8, and shaken at 130 

rpm at 28 o 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 10 5 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


Evaluation 

Inoculated leaves were examined every 15 days starting on the 15 th day after 

inoculation and ending on the 60 th . 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 Leaf symptoms mostly absent. 

1 Reddish flecks on lower leaf surface. No symptoms on the upper surface. 

2 Regular or irregular reddish circular spots on the lower leaf surface. No symptoms on the upper surface. 

3 Regular or diffuse light brown circular spots oin the upper leaf surface. 

4 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. 

5 Black spots with dry centre of grey colour. Leaf completely necrotic, sometimes hanging down. 

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. 


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

Session 3 


Schedule 

In vitro plants 

Mycosphaerella fijiensis 

Acclimatized 8 weeks 

20cm, 4 active leaves 

Mycelial suspension 

1x10 5 cfu/ml 

INOCULATION 

Under surface of first 3 open leaves 

Humidity 90-100% for 3 days-sunlight 

EVALUATION 

0 1 2 

15, 30, 45 and 60 days 

Stages 0-5 

3 4 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 Reaction in the field Symptom stage 

15 d* 30 d 45 d 60 d 

Grande naine Susceptible 1 1 2 4 

Niyarma yik Susceptible 1 1 1 4 

FHIA-01 Partially resistant 0 0 1 2 



Calcutta 4 Resistant 0 0 1 1 

Yangambi Resistant 1 2 3 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


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 3 

Line-F21-B 3 

Line-F21-C 2 

Line-F21-D 1 

Line-F21-E 2 

Control 

‘Grande naine’ 5 

‘FHIA-21’ 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

Session 3 


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, 



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



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

Session 13 


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

Session 4 

Genetic improvement for the 

management of resistance

Session 4 

K. Craenen and R. Ortiz 

Introduction 

Genetic improvement for a sustainable 

management of resistance 

K. Craenen 1 and R. Ortiz 2 

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 (bs 1 

) coupled with at least two additive minor genes (bsr i 

) 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 (bs 1 

) acoplado al menos a dos genes secundarios aditivos (bsr i 

) 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 

2 

International Institute of Tropical Agriculture, Nigeria 

181


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 (bs 1 

) couplée à au moins deux 

gènes mineurs additifs (bsr i 

) 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

Session 4 


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


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

Session 4 


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


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

Session 4 


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 mm 2 vs. 1.07 mm – 4.07 mm 2 , 

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


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

Session 4 


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


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 (bsr i ) 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 (bs 1 ) 

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 bs 1 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 bs 1 locus were established by one-way analysis of variance 

and regression analysis. Intralocus interaction at the bs 1 locus apparently regulates the 

appearance of symptoms on the leaf surface, whereas the additive effect and the 

intralocus interaction of the bs 1 locus affect disease development in the host plant. 

Therefore, the gene action(s) at the bs 1 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

Session 4 


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 (bs 1 ) and of the parthenocarpy gene (P 1 ), 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 Durbin- 

Watson 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 (bs 1 ) 

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 (bs 1 ), 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


the desired characteristics. Recurrent selection methods are the most common for 

increasing the frequency of favourable alleles in a cyclic fashion. 

Theoretically, sq 2 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, 

sq 4 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

Session 4 


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 

bs 1 

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


tested in Musa. Resistance to black leaf streak disease provided by the recessive bs 1 

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. 


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 


of the global population of banana black leaf streak fungus Mycosphaerella fijiensis. 


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 bs 1 

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 bs 1 

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

Session 4 


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 7 th 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


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. 







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 



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

Session 4 


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. 223- 

244 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 (2 nd 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


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

Session 4 

C. Jenny et al. 

Conventional breeding 

of bananas 

C. Jenny 1 ,K.Tomekpé 2 ,F.Bakry 3 and J.V. Escalant 4 

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. 

1 

CIRAD-FHLOR, Guadeloupe 

2 

CARBAP, Douala, Cameroon 

3 


4 

INIBAP, Montpellier, France 

199


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

Session 4 


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 20 th 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


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. 

Female parent 

2n = 3x = 33 

Male parent 

2n = 2x = 22 

MEIOSIS 

Female gametes 

n = variable 

Some non reduced gametes 

Male parent 

n = 11 

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

Session 4 


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


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 

AAcv 

AAcv 

x 

x 

x 

AAcv 

AAcv 

BBw 

AAcv 

ABcv 

colchicine 

colchicine 

AAAAcv 

AABBcv 

(1) 

(2) 

Triploid development 

AAcv 

AAw 

x 

x 

AAAAcv 

AAAAcv 

AAAAcv 

AAAAcv 

(3,4) 

BBw 

AAcv 

AAw 

x 

x 

x 

AAAAcv 

AABBcv 

AABBcv 

AABcv 

AABcv 

AABcv 

(3,5) 

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

Session 4 


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


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

Session 4 


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, 27- 

30 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


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. 223- 

243 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, 


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

Session 4 

R. Swennen et al. 

Transgenic approaches for resistance 


in Musa spp. 

R. Swennen 1 ,G.Arinaitwe 1 ,B.P.A.Cammue 2 ,I.François 2 ,B.Panis 1 , 

S. Remy 1 ,L.Sági 1 ,E.Santos 1 ,H.Strosse 1 and I. Van den Houwe 1 

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 

2 

Centre for Microbial and Plant Genetics, KULeuven, Leuven, Belgium 

209


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

Session 4 


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


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 10 4 to 

10 5 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

Session 4 


Table 1. Preliminary results using the scalp method (Jan 1998–Dec 2001). 

Cultivar Genome Type ITC Number of Responsive Frequency Highest EC ECS Ready for 

code inoculated scalps* (%)** frequency first applications 

scalps in a single 

experiment 

Calcutta 4 a AA wild diploid 1872 

Grande naine AAA Cavendish 1256 2040 112 5.5 11.7 yes yes yes (4) b 

GN FHIA AAA Cavendish 1296 11 0.8 2.9 yes yes 

GN JD AAA Cavendish 1872 16 0.9 4.2 Yes yes 

Williams BSJ AAA Cavendish 0570 456 

Williams JD AAA Cavendish 2808 99 3.5 22.2 yes yes yes (7) b 

Ingarama AAA-h highland 0160 1104 

Mbwazirume AAA-h highland 0084 1128 240 

Nyamwihogora AAA-h highland 0086 864 

Agbagba AAB-p plantain 0111 576 3 0.5 0.5 yes yes yes f (1) b 

Obino l’ewai AAB-p plantain 0109 336 3 0.9 2 yes yes yes (1) b 

Orishele AAB-p plantain 0517 1200 29 2.4 5.8 yes yes yes (4) b 

Burro cemsa ABB cooking 1259 504 

Total 16056 273 

Mean 1.1 3.8 

* 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) 

EC Embryogenic complex 

ECS Embryogenic cell suspension 

a Derived via zygotic embryo rescue. Seeds obtained from IITA, Nigeria 

b Number of independently established cell suspension lines ready for first applications at the end of 2001 

213


Table 2. Preliminary results using TDZ scalps (Jan 1998–Dec 2001). 

Cultivar Genome Type ITC code Number of NS3* Responsive Frequency (%)*** Highest frequency EC ECS 

inoculated scalps scalps** in a single experiment 

Calcutta4 a AA wild diploid 96 24 

Williams AAA Cavendish 0365 960 720 13 1.8 2.8 yes yes 

Igisahira gizanswe AAA-h highland 0083 264 24 

Agbagba AAB-p plantain 0111 144 24 1 4.2 4.2 yes 

Bluggoe ABB cooking 0010 144 144 4 2.8 5.6 yes 

Cachaco b ABB cooking 0643 144 144 4 2.8 5.6 yes 

Cachaco ABB cooking 0643 120 

Total 1872 1080 22 

Mean 1.7 2.3 

* 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 

EC Embryogenic complex 

ECS Embryogenic cell suspension 

a Derived via zygotic embryo rescue. Seeds obtained from IITA, Nigeria 

b Scalps derived from 1 µM TDZ meristem cultures instead of 10µM TDZ meristem cultures as in all other cases 

214

Session 4 


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


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 117 76 

Obino l’ewai 118 41 

Orishele 93 05 

Three hand planty 96 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

Session 4 


% Regeneration 

60 

50 

40 

30 

20 

GUS 

GFP 

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). 

Number of blue foci 

2500 

2000 

1500 

1000 

1 st trial 

2 nd trial 

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


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 Cultivar Infection time (hrs) 

gene 4 6 8 10 12 14 

gus Grande naine 794±87.9 881±91.8 >1500 >1500 >1500 >1500 

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

Session 4 


Table 5. Promoters used in banana genetic transformation. 

Promoter Source of the promoter Reference 

CaMV35S Cauliflower Mosaic Virus 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 

CaMV35S a Cauliflower Mosaic Virus Más et al., 2000 

35S-AMV Cauliflower Mosaic Virus Sági et al., 1995c 

Alfalfa Mosaic Virus 

35S-35S b Cauliflower Mosaic Virus 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 

35S-35S-AMV Cauliflower Mosaic Virus Sági et al., 1994; Sági et al., 1995a; Sági et al., 1995b; 

Alfalfa Mosaic Virus 

Sági et al., 1995c; Sági et al., 1998a 

Emu c Recombinant ARE-ocs-adh1 Sági et al., 1995a; Sági et al., 1995b; 

Sági et al., 1998a; Remy, 2000 

Act-1 Rice actin gene act1D May et al. 1995; Sági et al., 1998a 

Act-1 Banana actin gene Hermann et al., 2001b 

Ubi Maize ubiquitin 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 

(ocs) 3 

mas d Recombinant ocs-mas Remy et al., 1998b; Moy et al., 1998; Moy et al., 1999; 

Remy, 2000; Ganapathi et al., 2001 

Sc Sugarcane bacilliform badnavirus Schenk et al., 1999 

(ScBV) 

Badnavirus Banana streak badnavirus Sági, 1998a; Schenk et al., 2001; Remans et al. 2000 

My 

My from Mysore cv. BSV 

Cv 

Cv from Cavendich cv. BSV 

Go 

Go from Goldfinger cv. BSV 

BBTV DNA 1 to 6 Banana Bunchy Top Virus Dugdale et al., 1998; Becker et al., 2000; 

S1 

Hermann et al., 2001a 

S2 

BT1, BT2, BT3, BT4, BT5 e Dugdale et al., 2000 

BT6.1 f Dugdale et al., 2001 

a 

OCS enhancer plus Cauliflower Mosaic Virus promoter plus rice act1 untranslated sequence 

b 

Enhanced Cauliflower Mosaic Virus promoter 

c 

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 

d 

Atandem of three upstream activating sequences (UAS) of the octopine synthase gene (ocs) and a promoter/activator region of the mannopine 

synthase gene (mas) 

e 

Banana Bunchy Top Virus DNA 1-5 intergenic regions plus maize ubiquitin (ubi1) intron 

f 

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


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 IC 50 

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

Session 4 


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


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 ketol- 

222

Session 4 


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


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

Session 4 


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


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

Session 4 


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


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

Session 4 


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. 267:2228- 

2233. 

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


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 82:7870- 

7873. 

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

Session 4 


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 3 rd 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 diet 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


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. 10:123- 

150. 

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

Session 4 


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 21 st 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 1 st 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


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. 2 nd 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

Session 4 


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 


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 


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. 19:870- 

875. 

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


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, 5- 

10 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

Session 4 


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


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

Session 4 

N. Roux et al. 

Mutagenesis and somaclonal 

variation to develop new resistance 


N. Roux 1 ,A.Toloza 1 ,J.P.Busogoro 2 ,B.Panis 3 , 

H. Strosse 3 ,P.Lepoivre 2 ,R.Swennen 3 and F. J. Zapata-Arias 1 

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 

2 

Agricultural University Gembloux, Gembloux, Belgium 

3 

Katholieke Universiteit Leuven,Leuven, Belgium 

239


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

Session 4 


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


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 60 Co 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 M 1 

V 3 

(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

Session 4 


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 Parent clone Selected clone Selected traits Technique Place of induction 

Cuba SH-3436 SH-3436-L9 Reduced height Gamma rays Cuba 

Parecido al Rey 6.44 Reduced height Gamma rays IAEA 

Malaysia Pisang rastali Mutiara FOC* tolerance Somaclonal variation Malaysia 

Novaria FOC tolerance Somaclonal variation Malaysia 

Philippines Lakatan LK-40 Reduced height Gamma rays IAEA 

Latundan LT-3 Larger fruit size Gamma rays IAEA 

Sri Lanka Embul Embul-35 Gy Earliness Gamma rays Sri Lanka 

Austria Grande naine GN35-I to Tolerance to toxin Gamma rays IAEA 

(IAEA**) GN35-VIII from Mycosphaerella 

fijiensis 

* 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


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 cm 2 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) 

90 

80 

70 

60 

50 

40 

30 

20 

10 

0 

0% 

0.05% 0.1% 

0.2% 

3x 

6x 

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

Session 4 


Table 2. Ploidy distribution of regenerated plants from colchicines-treated ECS of the triploid cell line 124T. 

Colchicine Number of regenerated Ploidy 

concentration (%) 

plants 

3n 6n Chimeric mixoploidy (3n+6n) 

0 108 108 0 0 

0.05 63 58 5 0 

0.1 37 36 1 0 

0.2 88 84 4 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 60 Co 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 (CWG 75 

=0,84g) was 50% of the control 

(CWG 0 

= 1,68g) (Figure 2A). 

To measure the regeneration capacity, green plantlets were counted and transferred 

to Magenta GA7 boxes containing semi-solid R 3 

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


140 

Fresh weight gain (% of control) 

120 

100 

80 

60 

40 

20 

WIL 

THP 

0 

0 50 100 150 200 250 

Dose (Gy) 

Regeneration capacity (% of control) 

120 

100 

80 

60 

40 

20 

WIL 

THP 

0 

0 50 100 150 200 250 

Dose (Gy) 

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

Session 4 


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 Number. of Number of plants Number of 

meristems shoots in M 1 

V 1 

screened in M 1 

V 4 

plants retained 

A(ST) 100 142 780 8 

B (ST) 100 105 512 0 

C (MA) 100 110 1351 4 

D (MA) 100 140 1085 3 

TOTAL 400 497 3728 15 

ST: propagated by shoot tip culture; MA: propagated by the multi-apexing culture technique 

M x 

: Mutagenic treatment; V x 

: 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


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. 


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

Session 4 


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 96:125- 

128. 

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 

n o 8, 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, 


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


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

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. Cavalcante 1 ,A.da S. Ledo 1 , 

F. H. S. Costa 1 ,Z.J.M.Cordeiro 2 ,A.P.Matos 2 

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. 

1 

Embrapa, Rio Branco, Brazil 

2 

Embrapa Mandioca e Fruticultura, Cruz das Almas, Brazil 

251


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 



252

Session 4 


5 = 34 to 50% of lamina with symptoms and 


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 4 51.2714 

Genotypes (G) 6 1356.9333* 

Cultivation system (CS) 1 66.0571ns 

Interaction (G X CS) 6 10.3238ns 

Error 52 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). 

60 

55,1a 

50 

39b 

Disease severity (%) 

40 

30 

20 

19,7e 

25,8d 26,1d 

29,3d 

33,5c 

10 

0 

PV-4285 

FHIA-02 FHIA-01 Caipira FHIA-21 Thap maeo SH 36-40 

Figure 1. Average disease severity of black leaf streak disease in seven genotypes. 

253


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

Session 4 


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

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


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 ‘SH- 

3436-9’ 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

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


Table 1. Improved cultivars included in the resistance trials of IMTP II. 

Clone Origin 1 Genome 

PV-03.44 EMBRAPA AAAB 

PA-03.22 EMBRAPA AAAB 

SH-3436-9 INIVIT AAAA 

FHIA-01 FHIA AAAB 

FHIA-03 FHIA AABB 

FHIA-17 FHIA AAAA 

FHIA-23 FHIA AAAA 

GCTCV-119 TBRI AAA 

GCTCV-215 TBRI AAA 

1 

Breeding 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 Infection Index 

YLS 

35 

12 

10 

30 

Infection Index 

25 

20 

15 

10 

5 

8 

6 

4 

2 

YLS 

0 

0 

Calcutta 4 

Yangambi Km5 

Pisang Ceylan 

FHIA 23 

SH3436-9 

Pisang Berlin 

Saba 

PA 03-22 

PV 02-44 

Local cultivar 

Figure 1. Infection index of black leaf streak disease and youngest leaf spotted (YLS) of different genotypes 

in Cameroon. 

260

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 

Tonga 

Clone Infection Index YLS Clone Infection Index YLS 

PV-03.44 36.1 8.3 PV-03.44 29.9 7 

PA-03.22 34.4 8.2 PA-03.22 34.8 7 

SH-3436-9 24.6 7.4 SH-3436-9 13.5 11 

FHIA-23 18.4 7.8 FHIA-23 61 11 

References 

References 

Calcutta 4 10.6 7.9 Calcutta 4 1.8 

Yangambi Km5 19.1 7.9 Yangambi Km5 20 11 

Pisang Ceylan 18.7 8.5 Pisang Ceylan 19 10 

Saba 32.3 8 Saba 20.3 9 

Local cultivar 45.6 4.9 Local cultivar 30 5 

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


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) 

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

Session 4 

J.V. Escalant 

250 

Across clones 

200 

Finger Weight (g) 

150 

100 

50 

R 2 = 0,5064 

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 shortly 1 . 

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 

0 

0 10 20 30 40 50 60 70 

Infection index 

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


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 Queensland Horticultural Institute (QHI) Not confirmed 

Bangladesh Bangladesh Agricultural Research Institute (BARI) Performance 

Burundi Institut de recherches agronomique et zootechnique de la 

communauté économique des pays des grands lacs (IRAZ) 

Performance 

Cameroon Centre africain de recherches sur bananiers et plantains (CARBAP) In-depth 

China South China Agricultural University (SCAU) Performance 

Colombia Corporación Colombiana de Investigación Agropecuaria (CORPOICA) In-depth 

Costa Rica Corporación Bananera Nacional (CORBANA) In-depth 

Dominican Centro para el Desarrollo Agropecuario y Forestal (CEDAF) Performance 

Republic 

Haiti Institut Interaméricain de Coopération pour l’agriculture (IICA) Performance 

Honduras Fundación Hondureña de Investigación Agrícola (FHIA) Performance 

India National Research Center on Banana (NRCB) In-depth 

Indonesia Central Research Institute for Horticulture (CRIH) Performance 

Malaysia Malaysian Agricultural Research and Development Institute (MARDI) In-depth 

Mexico Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP) In-depth 

Nicaragua Universidad de León (UNAN León) Performance 

Peru Servicio Nacional de Sanidad Agraria (SENASA) Performance 

Philippines Bureau of Plant Industry – Davao National Crop Research In-depth 

and Development Center (BPI – DNCRDC) 

Philippines Dole Asia Research ; Stanfilco In-depth 

Philippines Lapanday Fruit Company In-depth 

Rwanda Institut des sciences agronomiques du Rwanda (ISAR) Performance 

Sri Lanka Agricultural Research Station (ARS) Performance 

Uganda National Agricultural Research Organization (NARO) In-depth/ 

Performance 

Vietnam Vietnam Agricultural Science Institute (VASI) 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

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



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

Session 4 


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

Session 5 

Integrated disease management

Session 5 

R. Peterson et al. 

Management of Mycosphaerella 

leaf spot diseases in Australia 

R. Peterson 1 , K. Grice 1 and S. De La Rue 1 

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


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

Session 5 


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


•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 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 31 st 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

Session 5 


•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 2 nd 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


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

Session 5 

P.E. Jorge and T. Polanco 

Spread and management 

of black leaf streak disease 

in the Dominican Republic 


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


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 km 2 . The Republic of Haiti occupies the 

remainder of the island. Both countries are divided by the Cordillera Mountains which 

278

Session 5 


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 1200-1350 

Northeast 1200-1350 

North 1200-1350 

East 1342-1583 

Central 1400-1800 

Northwest < 600 

South < 600 

Southwest 665 

Source: Secretaría de Estado de Agricultura, 1998. Anuario estadistico de planificación sectorial agropecuaria. 

279


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 Plantain (ha) Banana (ha) 

South 12 500* 3105 

Southwest 5764 5722 

Central 3750 - 

East 625b - 

North Central 14 333 2062 

Northeast 3139 4280 

North 9000 3585 

Northwest 2876 6201 

Total 51 987 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

Session 5 


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 1961- 

1990, 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


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

Session 5 


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


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 ‘FHIA- 

21’ 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

Session 5 


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

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/PO 2 

,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/PO 2 

,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/PO 2 

,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


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

Session 5 

A.S. Riveros, C.I. Giraldo and A. Gamboa 

Identification 

of antagonistic 

microorganisms 

Effect of substrate 

on antagonistic 


External vs internal 

application 

of antagonistic 


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


Step 3. Internal vs external application of antagonistic 


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; 

KH 2 

PO 4 

and K 2 

HPO 4 

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

Session 5 


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. 


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


• 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 2x10 5 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

Session 5 


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. 


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. 


The authors thank INIBAP and INIBAP-LAC for the financial support to conduct this 

research. 

293


140.0 

100.0 

A 

% of inhibition 

60.0 

20.0 

- 20.0 

- 60.0 

- 100.0 

- 140.0 

A30 

R1 

SE 

GS2 

GS3 

GC1 

GBC2 

GBC2 

GBC2 

White 

Tilt 

Microbiological filtrates 

0.5 ppm 0.1 ppm 0.01 ppm 

120.0 

100.0 

B 

% of inhibition 

80.0 

60.0 

40.0 

20.0 

0.0 

- 20.0 

A30 

R1 

SE 

GS2 

GS3 

GC1 

GBC2 

GBC2 

GBC2 

Microbiological filtrates 

White 

Tilt 

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

Session 5 

A.S. Riveros et al. 

GBC2 

0.1 ppm 

Control 

(water) 

SE/P0 2 

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 ORTON- 

CATIE, 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


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, 


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, 


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, 


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

Session 5 

E. Spaans and L. Quiros 

Precision agriculture to improve 

management decisions and field 

research 


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


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, mechaniza- 

298

Session 5 


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


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. 


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

Session 5 


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


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

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. Knight 1 ,M.Wirz 1 ,A.Amil 2 ,A.Hall 2 and M. Shaw 3 

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. 

1 

Syngenta Costa Rica SA, San José, Costa Rica 

2 

Jealott’s Hill International Research Centre, Bracknell, UK 

3 

University of Reading, Reading, UK 

303


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

Session 5 


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 bc 1 

. 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 G 143 

A. 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. 


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 G 143 

A mutation was detected using a diagnostic primer for 

G 143 

A. This primer is extended only when the G 143 

A 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



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 

Prevalence of resistance (% of resistant spores) 

1999 2000 2001 

Mexico - 0.0 (2) - 

Belize - - - 

Honduras 0.0 (2) 0.0 (3) 0.0 (30) 

Guatemala 0.0 (4) 0.0 (4) 0.0 (25) * 

Nicaragua - - 0.0 (3) 

Costa Rica - 11.3 (33) 11.1 (78) 

Panama - 0.0 (5) 3.3 (15) 

Colombia - 0.0 (4) 0.03 (25) 

Ecuador - - 0.01 (28)** 

Cameroon - - 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 G 143 

A 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 G 143 

A. 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 G 143 

A 

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

Session 5 


100 

Baseline 

Farm 1 

Farm 2 

Farm 3 

% of ascopores 

75 

50 

PCR 



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

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 


France 

Tel.: +33 467616529 

Fax: +33 467615793 

E-mail: jean.carlier@cirad.fr 

311


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 



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 


Costa Rica 

Tel.: +506 5566431 

Fax: +506 5561533 

E-mail: CATIE@catie.ac.cr 

Emile Frison 

INIBAP 


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


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 

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 

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 

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


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 



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


Xavier Mourichon 

CIRAD-AMIS 

Av. Agropolis TA 40/02 


France 

Tel.: +33 467615869 

Fax: +33 467615581 

E-mail: xavier.mourichon@cirad.fr 

Juan Luis Ortiz 

CATIE 


Cartago 

Costa Rica 

Tel.: +506 5 566455 

E-mail: jortiz@catie.ac.cr 

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 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 



CP 97200 

Mexico 

Tel.: +52 99 813966 

Fax: +52 99 813900 

E-mail: latyperaza@yahoo.com 

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 

Claudine Picq 

INIBAP 


Montpellier 34397 Cedex 5 

France 

Tel.: +33 467611302 

Fax: +33 467610334 

E-mail: c.picq@cgiar.org 

Luis Pocasangre 

INIBAP 

Latin America and Caribbean Regional Office 

C/o CATIE 


Costa Rica 

Tel.: +506 5562431 

Fax: +506 5562431 

E-mail: lpoca@catie.ac.cr 

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


Galileo Rivas 

CATIE 


Costa Rica 

Tel.: +506 5560232 

Fax: +506 5566480 

E-mail: grivas@catie.ac.cr 

Mauricio Rivera Canales 

FHIA 

Apdo Postal 2067 

San Pedro Sula 

Honduras 

Tel.: +504 6682470 

Fax: +504 6682313 

E-mail: mrivera@fhia.org.hn 

Alba Stella Riveros Angarita 

UTOLIMA - CATIE 

Unidad de Fitoprotección 


Costa Rica 

Tel.: +506 5566021 

Fax: +506 5560606 

E-mail: asrivero@catie.ac.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 

Franklin Rosales 

INIBAP 

Latin America and Caribbean 

Regional Office 

C/o CATIE 


Costa Rica 

Tel.: +506 5562431 

Fax: +506 5562431 

E-mail: frosales@catie.ac.cr 

Nicolas Roux* 

IAEA Plant Breeding Unit 

A-2444 Seibersdorf 

Austria 

*Current address 

INIBAP 



France 

Tel.: +33 467611302 

Fax: +33 467610334 

E-mail: n.roux@cgiar.org 

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 

Jorge Sandoval 

CORBANA 

Apdo 390-7210 

Guápiles, Limón 

Costa Rica 

Tel.: +506 7633176 

Fax: +506 7633055 

E-mail: jsandoval@corbana.co.cr 

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 

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


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 

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 

Ana Cecilia Tapia 

CATIE 

Apdo 7170, Turrialba 

Costa Rica 

Tel.: +506 5560232 

Fax: +506 556 6480 

E-mail: tapiaa@catie.ac.cr 

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 

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 

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 

317

isbn 2-910810-57-7

Mycosphaerella leaf spot diseases of bananas - CBS

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?