Investigation of entomopathogenic fungi for control of false
codling moth, Thaumatotibia leucotreta, Mediterranean fruit
fly, Ceratitis capitata and Natal fruit fly, C. rosa in South
African Citrus.
A thesis submitted in the fulfilment of
the requirements for the degree of
MASTER OF SCIENCE
of
RHODES UNIVERSITY
by
Tarryn Anne Goble
December 2009
i
Abstract
The biology of key citrus pests Thaumatotibia leucotreta Meyrick (Lepidoptera: Tortricidae),
Ceratitis capitata Wiedemann (Diptera: Tephritidae) and Ceratitis rosa Karsch (Diptera:
Tephritidae) includes their dropping from host plants to pupate in the soil below citrus trees.
Since most EP fungi are soil-borne microorganisms, the development and formulation of
alternative control strategies using these fungi as subterranean control agents, targeted at larvae
and pupae in the soil, can potentially benefit existing IPM management of citrus in South Africa.
Thus, a survey of occurrence of entomopathogenic fungi was undertaken on soils from citrus
orchards and natural vegetation (refugia) on conventionally and organically managed farms in
the Eastern Cape Province in South Africa. A method for baiting soil samples with citrus pest T.
leucotreta and C. capitata larvae, as well as with the standard bait insect, Galleria mellonella
Linnaeus
(Lepidoptera:
Pyralidae),
was
implemented.
Sixty-two
potentially
useful
entomopathogenic fungal isolates belonging to four genera were collected from 288 soil samples,
an occurrence frequency of 21.53%. The most frequently isolated entomopathogenic fungal
species was Beauveria bassiana (Balsamo) Vuillemin (15.63%), followed by Metarhizium
anisopliae var. anisopliae (Metschnikoff) Sorokin (3.82%). Galleria mellonella was the most
effective insect used to isolate fungal species (χ2=40.13, df=2, P≤ 0.005), with a total of 45
isolates obtained, followed by C. capitata with 11 isolates, and T. leucotreta with six isolates
recovered. There was a significantly (χ2=11.65, df=1, P≤ 0.005) higher occurrence of
entomopathogenic fungi in soil samples taken from refugia compared to cultivated orchards of
both organically and conventionally managed farms. No significant differences were observed in
the recovery of fungal isolates when soil samples from both farming systems were compared.
The physiological effects and host range of 21 indigenous fungal isolates obtained in the Eastern
Cape were investigated in the laboratory to establish whether these isolates could be effectively
used as biological control agents against the subterranean life stages of C. rosa, C. capitata and
T. leucotreta. When these pests were treated with a fungal concentration of 1 x 107 conidia ml-1,
the percentage of T. leucotreta adults which emerged in fungal treated sand ranged from 5 to
60% (F=33.295; df=21; P=0.0001) depending on fungal isolate and the percentage of pupae with
visible signs of mycosis ranged from 21 to 93% (F= 96.436; df=21; P=0.0001). Based on fungal
isolates, the percentage adult survival in C. rosa and C. capitata ranged from 30 to 90% and 55
to 86% respectively. The percentage of C. rosa and C. capitata puparia with visible signs of
ii
mycosis ranged from 1 to 14% and 1 to 11% respectively. Deferred mortality due to mycosis in
C. rosa and C. capitata adult flies ranged from 1 to 58% and 1 to 33% respectively, depending
on fungal isolate. Entomopathogenic fungal isolates had a significantly greater effect on the
adults of C. rosa and C. capitata than they did on the puparia of these two fruit fly species.
Further, C. rosa and C. capitata did not differ significantly in their response to
entomopathogenic fungi when adult survival or adult and pupal mycosis were considered.
The relative potency of the four most virulent Beauveria isolates as well as the commercially
available Beauveria bassiana product, Bb Plus® (Biological Control Products, South Africa),
were compared against one another as log-probit regressions of mortality against C. rosa, C.
capitata and T. leucotreta which all exhibited a dose-dependent response. Against fruit flies the
estimated LC50 values of all five Beauveria isolates ranged from 5.5 x 1011 to 2.8 x 1012
conidia/ml-1. There were no significant differences between the relative potencies of these five
fungal isolates. When T. leucotreta was considered, isolates: G Moss R10 and G 14 2 B5 and Bb
Plus® were significantly more pathogenic than G B Ar 23 B3 and FCM 10 13 L1. The estimated
LC50 values of the three most pathogenic isolates ranged from 6.8 x 105 to 2.1 x 106 conidia/ml-1,
while those of the least pathogenic ranged from 1.6 x 107 to 3.7 x 107 conidia/ml-1.
Thaumatotibia leucotreta final instar larvae were exposed to two conidial concentrations, at four
different exposure times (12, 48, 72 and 96 hrs) and showed an exposure time-dependant
relationship (F=5.43; df=3; P=0.001). At 1 x 107conidia/ml-1 two Beauveria isolates: G Moss
R10 and G 14 2 B5 were able to elicit a response in 50% of test insects at 72 hrs (3 days)
exposure.
Although a limited amount of mycosis was observed in the puparia of both fruit fly species,
deferred adult mortality due to mycosis was high. The increased incidence of adult mortality
suggests that post emergence mycosis in adult fruit flies may play a more significant role in field
suppression than the control of fruit flies at the pupal stage. The increased incidence of pupal
mortality, as well as the relatively low concentrations of conidia required to elicit meaningful
responses in T. leucotreta pupae may suggest that pre-emergent control of false codling moth
will play a more significant role in field suppression than the control of adult life stages using
indigenous isolates of entomopathogenic fungi. Various entomopathogenic fungal application
techniques targeted at key insect pests within integrated pest management (IPM) systems of
citrus are discussed.
iii
To my parents,
Clive and Jen
Thank you both for your incredible love of nature
and your incredible nature of love
iv
Declaration
The following thesis has not been submitted to any university other than Rhodes University,
Grahamstown, South Africa. The work presented here is that of the author.
Date:
v
Acknowledgements
The road to scientific understanding is a journey of growth. There are many people along the
way that inspire and teach you, and who make opportunities available. I would sincerely like to
thank the following people for guiding me through this important time in my life.
To my supervisor, Martin Hill, thank you for being so enthusiastic about Entomology and
fascinating my imagination with all the possibilities that this profession holds. Your interest in
this subject inspires interest in others.
To Jo Dames, my microbial co-supervisor, thank you for being so willing to take me on as a
student in your lab. You always offer sound advice and friendship.
I thank Sean Moore for keeping a close watch on my work for the last three years. You are really
the reason I have found myself in Invertebrate Pathology and loving it! Your guidance and
support is very much appreciated, thank you.
I would also like to thank Kate Benyon for being the most wonderful financial manager a girl
could ask for. I really appreciate your help and friendship.
My deepest gratitude goes to Julie Coetzee and Martin Villet for all your help with statistics, the
department is so lucky to have the two of you and I really appreciate the time you have both
taken to explain and help me understand stats!
I would sincerely like to thank the Citrus Academy for all the help and support I have received.
Jacomien and Desiree you ladies are fantastic and you really have both gone above and beyond
to help me.
I am truly grateful to Citrus Research International for the technical and financial support I have
received. Tim Grout thank you for all your help and advice with this project. I would like to
thank Aruna Manrakhan, Craig Chambers and Rooikie for assisting me with technical aspects of
the project.
vi
A special thank you to Million, Milton and Simon Apane. I am also extremely grateful to Terry
Butterworth for his technical assistance.
I would really like to thank the following citrus farmers for allowing me to collect soil samples
from orchards and being so enthusiastic about this project: Keith Finnemore, Dave Gerber, Bruce
Knott, Paul Marais, Rob & Steve Moss.
I would like to thank the following organisations for their financial support, Deutscher
Akademischer Austauschdienst (DAAD), National Research Foundation (NRF) and the Ernest &
Ethel Eriksen Trust.
I thank my friends James Robey and Elodie Heyns for your continual help, support and
encouragement, and for the hours of time you put into helping me with lab work. I thank Colleen
Hepburn (Ma Gumboots) for all your help with reading the manuscript and being there for me in
every aspect of my life, I appreciate your friendship and opinion in this profession.
Finally to my family: Clive, Jen, Jess, Megan and David. Thank you for your love,
encouragement and support always. I am so lucky to have you guys in my life and I love you all.
I would like to thank an important member of my family, Mike Strohman, your encouragement
and enthusiasm in everything we do is so special, thank you!
vii
Table of Contents
Abstract ......................................................................................................................................ii
Declaration ................................................................................................................................. v
Acknowledgements ................................................................................................................... vi
Table of Contents ................................................................................................................... viii
List of Tables ............................................................................................................................ xi
List of Figures ..........................................................................................................................xii
List of Abbreviations ............................................................................................................... xv
CHAPTER I-GENERAL INTRODUCTION ............................................................................ 1
1.1 CITRUS IN SOUTH AFRICA ...................................................................................... 1
1.1.1 The genus Citrus ..................................................................................................... 1
1.1.2 Citrus production areas ........................................................................................... 1
1.1.3 Climatic requirements ............................................................................................. 3
1.1.4 Soil requirements .................................................................................................... 3
1.1.5 Rootstocks ............................................................................................................... 4
1.1.6 Irrigation systems .................................................................................................... 4
1.1.7 The South African citrus export industry................................................................ 5
1.1.8 Citrus pests .............................................................................................................. 6
1.1.9 Control of citrus pests - towards an integrated pest management approach ........... 8
1.2 TARGET PESTS ........................................................................................................... 9
1.2.1 False codling moth, Thaumatotibia leucotreta ....................................................... 9
1.2.2 Fruit flies ............................................................................................................... 14
1.2.3 Justification for alternative control tactics ............................................................ 17
1.3 ENTOMOPATHOGENIC FUNGI .............................................................................. 18
1.3.1 General taxonomy ................................................................................................ 18
1.3.2 Fungal infection process and environmental cycling............................................ 19
1.3.3 Fungi as biological control agents ........................................................................ 20
1.3.4 Genus Beauveria ................................................................................................... 23
1.3.5 Genus Metarhizium ............................................................................................... 26
1.3.6 Genus Conidiobolus .............................................................................................. 28
1.3.7 Factors influencing the efficacy of EP fungi ........................................................ 29
1.3.8 Compatibility of fungi with agrochemicals .......................................................... 33
1.3.9 Compatibility with natural enemies and other biological control agents ............. 36
1.3.10 Safety of EP fungi towards vertebrates ............................................................... 38
viii
1.4 AIMS OF THE STUDY .............................................................................................. 38
CHAPTER II-SURVEYING FOR INDIGENOUS ISOLATES OF ENTOMOPATHOGENIC
FUNGI ..................................................................................................................................... 40
2.1 INTRODUCTION ....................................................................................................... 40
2.2 METHODS & MATERIALS ...................................................................................... 46
2.2.1 Insect Cultures ...................................................................................................... 46
2.2.2 Soil sampling ........................................................................................................ 46
2.2.3 Baiting procedures ................................................................................................ 48
2.2.4 Isolation and identification of fungi ...................................................................... 49
2.2.5 Quick screening method to detect entomopathogenic fungi ................................. 50
2.2.6 Data analysis ......................................................................................................... 51
2.3 RESULTS .................................................................................................................... 52
2.3.1 Bait insects and fungal species ............................................................................. 52
2.3.2 Abundance and geographic distribution of entomopathogenic fungi ................... 52
2.3.3 Farming systems and field margins ...................................................................... 55
2.4 DISCUSSION .............................................................................................................. 58
2.4.1 Bait insects and fungal species ............................................................................. 58
2.4.2 Fungal ecology ...................................................................................................... 59
2.4.3 Potential for biological control ............................................................................. 64
CHAPTER III-ASSESSING THE VIRULENCE OF ENTOMOPATHOGENIC FUNGI
AGAINST CITRUS PESTS .................................................................................................... 69
3.1 INTRODUCTION ....................................................................................................... 69
3.2 METHODS & MATERIALS ...................................................................................... 75
3.2.1 Fungal cultures ...................................................................................................... 75
3.2.2 Preparation of conidial suspensions ...................................................................... 77
3.2.3 Inoculation of insects ............................................................................................ 77
3.2.4 Concentration-response of citrus pests to EP fungi .............................................. 78
3.2.5 Exposure time-response trial of Thaumatotibia leucotreta to EP fungi ............... 79
3.2.6 Statistical analyses ................................................................................................ 79
3.3 RESULTS .................................................................................................................... 80
3.3.1 The effects of EP fungi on Thaumatotibia leucotreta adult emergence and pupal
mycosis ............................................................................................................................ 80
3.3.2 The effects of EP fungi on Ceratitis rosa and C. capitata adult survival ............. 83
3.3.3 The effects of EP fungi on Ceratitis rosa and C. capitata adult and pupal mycosis
.......................................................................................................................................... 85
3.3.4 The effect of fungal concentration on fruit fly mortality ...................................... 91
ix
3.3.5 The effect of fungal concentration on Thaumatotibia leucotreta mortality ......... 94
3.3.6 Exposure time-mortality relationship study between EP fungi and Thaumatotibia
leucotreta ......................................................................................................................... 95
3.4 DISCUSSION .............................................................................................................. 97
3.4.1 The effects of EP fungi on fruit flies, Ceratitis rosa and C. capitata ................... 97
3.4.2 The effects of EP fungi on false codling moth, Thaumatotibia leucotreta ......... 104
CHAPTER IV-GENERAL DISCUSSION............................................................................ 108
4.1 Integrated Pest Management strategies for the use of EP fungi in citrus .............. 108
4.2 How can soil application of EP fungi work in citrus? ........................................... 114
4.3 Autoinoculation devices for the use of EP fungi in citrus IPM ............................. 120
4.4 Conclusion ............................................................................................................. 125
4.5 Future work ............................................................................................................ 126
CHAPTER V-REFERENCES ............................................................................................... 129
x
List of Tables
Table 1.1 Important insect and mite pests of citrus in South Africa ................................................ 7
Table 1.2 EP fungi used in the control of citrus pests worldwide .................................................. 22
Table 2.2.1 Agronomic data of orchards of the six farms in the Eastern Cape from which soils
were sampled and used in insect baiting. ........................................................................................ 50
Table 2.3.1 Distribution and occurrence frequency (% positive samples) of entomopathogenic
fungi in 288 soils sampled from orchards and refugia on three organically and three
conventionally managed citrus farms in the Eastern Cape Province. ............................................. 54
Table 2.3.2 The location and ecological data of Beauvaria bassiana collected from citrus
orchards and refugia soils in the Eastern Cape Province, South Africa. ......................................... 56
Table 2.3.3 The location and ecological data of Conidiobolus coronatus, Lecanicillium
psalliotae, Metarhizium anisopliae and M. flavoviride collected from citrus orchards and refugia
soils in the Eastern Cape Province, South Africa............................................................................ 57
Table 3.2.1 Twenty one indigenous EP fungal isolates tested against Thaumatotibia leucotreta,
Ceratitis capitata and C. rosa final instar larvae ............................................................................ 76
Table 3.3.1 The calculated probit data of each of the five fungal isolates tested against Ceratitis
rosa, C. capitata and Thaumatotibia leucotreta. ............................................................................ 93
xi
List of Figures
Figure 1.1 The citrus producing regions of South Africa and Swaziland showing the total citrus
production per province and the total area planted to various citrus cultivars per province............. 2
Figure 1.2 The geographical distribution of false codling moth, Thaumatotibia leucotreta in
Africa and neighbouring islands ..................................................................................................... 10
Figure 1.3 Thaumatotibia leucotreta (A) egg laid singly on the rind of fruit (B) fifth instar
larva with characteristic red colouring (C) sclerotised brown pupae (D) mottled grey adult
moth. ................................................................................................................................................ 12
Figure 1.4 (A) Pupa, larva and eggs of Ceratitis capitata (B) Adult C. capitata fruit fly (C)
Adult female C. rosa fruit fly .......................................................................................................... 16
Figure 1.5 (A) Schematic diagram of an EP fungus invading the host insect cuticle (B)
Cordyceps sp. fungus growing out of the thorax of an unidentified fly species ............................. 20
Figure 1.6 (A) A photomicrograph showing three conidiogenous cells on a hypha of Beauveria
bassiana. (B) B. bassiana grown on Sauboraud dextrose agar (SDA) at 25 °C ............................. 24
Figure 1.7 The phylogenetic relationships of the genus Beauveria based on several
independent Bayesian and parsimony analyses .............................................................................. 25
Figure 1.8 (A) A transmission electron micrograph showing conidiogenous cells on the hyphae
of Metarhizium anisopliae (B) M. anisopliae grown on Sauboraud dextrose agar (SDA) at 25
°C (C) Conidia of M. flavoviride at 1000X under oil immersion (D) M. flavoviride grown on
Sauboraud dextrose agar (SDA) at 25°C ........................................................................................ 27
Figure 1.9 (A) A photomicrograph of Conidiobolus coronatus showing characteristic conidia
with prominent papilla (B) C. coronatus grown on Sauboraud dextrose agar (SDA) at 25°C ...... 29
Figure 1.10 A schematic diagram summarizing the compartments in temperate terrestrial
ecosystems in which B. bassiana and M. anisopliae occur naturally ............................................. 34
Figure 2.2.1 Map of the Sunday‟s River Valley citrus growing region and surrounding areas in
the Eastern Cape Province, South Africa, with the locations of the six farms, marked with a
bullet (•), where soils were sampled in this study. .......................................................................... 47
Figure 2.2.2 Sampling points in each orchard and the adjacent field margin (refugia). ................ 48
Figure 2.2.3 Galleria mellonella larvae infected with conidia from various EP fungal species
during the employed quick-screening method ................................................................................ 51
x ii
Figure 2.3.1 The total number of EP isolates and species recovered from the three different bait
insects used. ..................................................................................................................................... 53
Figure 2.3.2 Mean (% ± SE) of the frequency (% of samples) of entomopathogenic fungi in
soils sampled from orchards and refugia of organically and conventionally farmed soils ............. 58
Figure 2.4.1 (A, B, D) Metarhizium anisopliae var. anisopliae (C) Metarhizium flavoviride ..... 65
Figure 2.4.2 (E, F) Metarhizium anisopliae
(G, H) Beauveria bassiana ................................... 66
Figure 2.4.3 (I, J, K, L) Beauveria bassiana .................................................................................. 66
Figure 2.4.4 (M, N, O, P) Beauveria bassiana ............................................................................... 67
Figure 2.4.5 (Q, R, S, T) Beauveria bassiana ................................................................................ 67
Figure 2.4.6 (U) Beauveria bassiana ............................................................................................. 68
Figure 3.1 (A, B) Different types of culture colonies of Metarhizium anisopliae indicating
sterile sectorization and the restoration of conidiation in some areas. ............................................ 72
Figure 3.2.3 (A) Cut plastic bottles used as emergence chambers placed on Pearlite during
bioassay experiments. (B) During the scoring of mycosis dead insects were placed onto SDA
agar .................................................................................................................................................. 79
Figure 3.3.1 (A) A Thaumatotibia leucotreta wandering larva infected with Metarhzium
anisopliae var. anisopliae (B) An adult T. leucotreta moth infected with Beauveria bassiana ..... 81
Figure 3.3.2 Pathogenicity of isolates of Beauveria bassiana and Metarhizium anisopliae to
Thaumatotibia leucotreta: Mean (% ± SE) of adult emergence after treatment with a
concentration of 1 x 107 conidia ml-1............................................................................................... 81
Figure 3.3.3 Pathogenicity of Beauveria bassiana and Metarhizium anisopliae to
Thaumatotibia leucotreta: Mean (% ± SE) of pupal mycosis after treatment with a
concentration of 1 x 107 conidia ml-1............................................................................................... 82
Figure 3.3.4 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis rosa (black bars) and C. capitata (grey bars): Mean
(% ± SE) of fruit fly adult survival after treatment with a concentration of 1 x 107 conidia ml-1 ... 83
Figure 3.3.5 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis rosa (black bars) and C. capitata (grey bars): Mean
(% ± SE) of fruit fly adult survival after treatment with a concentration of 1 x 108 conidia ml-1 ... 85
Figure 3.3.6 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis rosa: Mean (% ± SE) of Ceratitis rosa adult (black bars)
and pupal (grey bars) mycosis after treatment with a concentration of 1 x 107 conidia ml-1 .......... 86
xiii
Figure 3.3.7 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis rosa: Mean (% ± SE) of Ceratitis rosa adult (black bars)
and pupal (grey bars) mycosis after treatment with a concentration of 1 x 108 conidia ml-1. ......... 87
Figure 3.3.8 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis capitata: Mean (% ± SE) of Ceratitis capitata adult (black
bars) and pupal (grey bars) mycosis after treatment with a concentration of 1 x 107 conidia
ml-1.. ................................................................................................................................................ 88
Figure 3.3.9 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and
Metarhizium anisopliae to Ceratitis capitata: Mean (% ± SE) of Ceratitis capitata adult (black
bars) and pupal (grey bars) mycosis after treatment with a concentration of 1 x 108 conidia
ml-1.. ................................................................................................................................................ 89
Figure 3.3.10 (A) Ceratitis capitata pupa infected with Metarhizium flavoviride (B) Ceratitis
rosa adult infected with Beauveria bassiana (C) C. capitata infected with M. anisopliae ............ 90
Figure 3.3.11 The log-probit regressions of mortality (pupal and adult combined) caused by
five isolates of Beauveria bassiana against Ceratitis rosa and C. capitata.................................... 92
Figure 3.3.12 The log-probit regressions of mortality (pupal and adult combined) caused by
five isolates of Beauveria bassiana against Thaumatotibia leucotreta........................................... 95
Figure 3.3.13 Thaumatotibia leucotreta total adult and pupal mycosis after final instar larvae
were exposed to three fungal isolates at two concentrations, for 12, 48, 72 and 96 hrs time
periods. ............................................................................................................................................ 96
Figure 3.4.1 (A) Female adult C. rosa fruit fly showing B. bassiana growing on the abdominal
segments (B) Female fly showing B. bassiana growth the dorsal and lateral thoracic scelrites
(C) Male adult C. rosa fruit fly showing B. bassiana growing on the abdominal segments. ....... 101
Figure 3.4.2 (A) Thaumatotibia leucotreta pupae spin cocoons around themselves and may
become attached to one another in Petri-dishes. (B) Beauveria bassiana fungal mycelia seen
growing from two adjacent T. leucotreta pupae. .......................................................................... 106
Figure 3.4.3 (A, B) Beauveria bassiana isolate FCM 10 13 L1 showing in vitro culture
degeneration (fluffy-mycelia growth). .......................................................................................... 107
x iv
List of Abbreviations
χ2
μl
%
ºC
BCA
CGA
cc
cm
CRI
EP
EU
FCM
F
g
hrs
ha
IGR
IPM
ITS
kg
L
LC
50
chi-squared
micro litre
percent
degrees Celsius
biological control agent
Citrus Growers Association
centimetre cubed
centimetre
Citrus Research International
entomopathogenic
European Union
false codling moth
frequency
grams
hours
hectare
insect growth regulator
integrated pest management
internal transcribed spacer
kilograms
litre
medial lethal concentration
LD
medial lethal dose
LT
50
medial lethal exposure time
metre
milligrams per litre
minute
milliliter
millimeter
mega Pascal
number of replicates
nano metre
potential of hydrogen ions
sterile insect technique
sensu stricto
United Kingdom
United States of America
United States Dollar
ultraviolet
South African Rand
50
m
mg/L
min
ml
mm
MPa
n
nm
pH
SIT
s.s.
UK
USA
US$
UV
ZAR
xv
I
GENERAL INTRODUCTION
1.1 CITRUS IN SOUTH AFRICA
1.1.1 The genus Citrus
Citrus (Rutaceae) Linnaeus is a genus of flowering, fruit-bearing trees which originated in the
tropical and subtropical areas of Southeast Asia (McCoy et al. 2007) and then spread to other
continents. The genus has had a long history in South Africa with some of the first orange and
lemon orchards planted over three centuries ago, in the 1650s when seedling trees, originating
from St. Helena Island, were planted in the gardens of the Dutch East India Company in Cape
Town, over 100 years before citrus orchards were planted in California (Mather & Rowcroft
2004). The precise number of natural species is unknown and genetic evidence suggests that
some wild, true-breeding species are of hybrid origin thus leaving the taxonomy and systematics
of the genus complicated (Nicolosi et al. 2000). Historically studies on the relationships between
citrus genera and species have been largely based on morphological characteristics. However,
recent molecular studies reveal that citrus can be divided into two subgenera, citrus and Papeda
(non-edible species) (Nicolosi et al. 2000). Almost all cultivated citrus belongs to the sub-genus
citrus, and the numerous cultivars are derived from three true breeding species based on genetic
studies where citron, pummelo and mandarin were placed into three distinct groups (Nicolosi et
al. 2000).
1.1.2 Citrus production areas
Citrus production occurs mainly in the Western and Eastern Cape, Mpumalanga, KwaZuluNatal, and Limpopo provinces (Bedford 1998) (Fig. 1.1). There are important differences
between production regions in South Africa based on climate and farm structure. The Western
Cape and Eastern Cape are considered to be „cooler‟ citrus growing areas and production is
1
focused on navel oranges and lemons (Mather 2003). The cooler climate has allowed farmers to
respond to consumer demand for easy peelers (soft citrus) such as Clementines and Satsuma
mandarins, and most of the country‟s easy peelers are produced in these two regions (Fig. 1.1).
The farm sizes in the Western and Eastern Cape Provinces are small, and most citrus is packed
by privatised cooperatives within facilities that are amongst the largest in the world (Mather
2003). In Mpumalanga, Limpopo and KwaZulu-Natal the climate is warmer and better suited to
the cultivation of grapefruit and Valencia oranges (Mather 2003) (Fig. 1.1). Navel orange trees,
however, tend to set smaller crops in the warmer parts of the citrus areas (Agricultural Research
Council 2003). Farm sizes in these warmer regions are larger and most farmers pack in smaller,
privately owned facilities. In terms of area under citrus cultivation (Ha), the Limpopo and
Western Cape Provinces have the largest total amounts of land under cultivation, followed by the
Eastern Cape and Mpumalanga (Fig. 1.1).
Figure 1.1 The citrus producing regions of South Africa and Swaziland showing the total citrus
production per province and the total area planted to various citrus cultivars per province (CGA
Annual Report 2008/9).
2
1.1.3 Climatic requirements
Climatic conditions vary considerably among the different citrus producing regions and even
within regions, and the effects of temperature can have significant impacts on production
(Bedford 1998). Production in South Africa is confined to areas with mild and almost frost-free
winters where temperatures rarely drop below -2°C (Agricultural Research Council 2003). Citrus
does well in most areas where the temperatures are between 10 to 30°C (the optimum
temperature between 20 to 28°C). However, temperature limits growth above 37.7°C and
prolonged temperatures above 30°C can cause fruit drop (Wellington 1960; Veldman 2001).
Night temperatures of between 10 to 14°C in June and July induce more intense blossom, and
mean minimum temperatures below 13°C in autumn (early March to early June), induce external
fruit colour (Veldman 2001). Most cultivars start to bear during the third year, although in hotter
climes it can be earlier and in colder areas as much as a year later. Lemons start bearing fruit a
year earlier because of their vigorous growth habit (Veldman 2001). Most citrus cultivar
production tends to peak after 8-12 years (Veldman 2001). Citrus (excluding lemons) require
shorter days and cooler temperatures in winter for a normal production rhythm. In the more
tropical areas the flowering pattern is much less clearly defined and main-season crops tend to be
considerably smaller (Agricultural Research Council 2003).
1.1.4 Soil requirements
Citrus can be grown in a wide range of soil types provided they are well-drained; usually sandy
loams of colluvial or alluvial origin with the most productive localities situated on the floors or
slopes of river valleys (Wellington 1960). Fertile, well-aerated soils with a pH of between 6 and
6.5 are ideal (Agricultural Research Council 2003). In the drier parts of the citrus growing
regions in South Africa, and in particular the Sundays River Valley in the Eastern Cape, soils of
a loam or clay nature may become alkaline (pH 9) due to lower rainfall, although careful
drainage and the addition of compost can counteract these effects (Wellington 1960). An unusual
condition in South African citrus soils is the low phosphorus content. The high acid-sugar ratio
of citrus fruit is thought to be as a result of this soil condition (Wellington 1960). An ideal citrus
soil with respect to optimum water holding capacity would be red, yellow-brown or brown, have
a clay content of between 10 to 40% and no clay or structural layers within one meter of the soil
surface (Agricultural Research Council 2003).
3
1.1.5 Rootstocks
All citrus produced in South Africa are budded onto rootstocks, and the majority of these are
certified trees (Veldman 2001). Farmers should have certification of their plantings which would
include the type of cultivar and the rootstock used, as well as the nursery information from where
they obtained their trees. This certification names the budwood, which is used to produce the
trees, and it is a requirement of Eurogap that the trees are true to type and are cleared of any
foreign pathogens (Lee 2009). Certification of trees and rootstocks means that they have various
degrees of resistance to soil- borne diseases and viruses (Lee 2009). Rootstocks influence tree
and fruit characteristics and are used to regulate tree height and improve tree age. They can
further improve some physiological problems associated with some cultivars, such as creasing,
cold or drought resistance and some stocks can sustain flooding situations (Veldman 2001). The
commercial rootstocks available in South Africa include the most widely used rootstock,
Swingle citrumelo; Rough lemon is often used on virgin soils and areas that induce high internal
fruit quality, it is also widely used on lemons and is adaptable to high pH soils (Veldman 2001).
Troyer/Carrizo citrange can be used on a wide range of soil types and cultivars; the rootstock
X639 is semi-dwarfed and used for high pH soils. The following rootstocks are used to a lesser
extent in South Africa: C35 citrange, Minneola x Trifoliate, Yuma citrange and Benton citrange
(Veldman 2001). For further reading on the importance of certification of citrus in integrated
pest management (IPM) it is suggested that Lee (2009) be consulted.
1.1.6 Irrigation systems
Citrus production requires vast amounts of water, especially when the canopy volume (tree age)
increases (Veldman 2001). Water is particularly important during the blossoming and fruitsetting periods, which occur from August to November in South Africa. Citrus will grow where
there is available water equivalent to a rainfall of 760-860 mm a year (Wellington 1960). The
mean annual rainfall in South African citrus areas varies from as much as 743 mm at Letaba and
949 mm at Richmond in the summer rainfall areas to 237 mm at Clanwilliam in the winter
rainfall area (Bedford 1998). Summer rainfall can be extremely unpredictable in the Eastern
Cape over these important phenological periods, so irrigation is particularly important.
Wellington (1960) however reported that a slight drying out of the soil from February to June
results in better quality fruits.
4
The most commonly used systems are: (1) overhead irrigation, where stationary sprayers are
situated approximately 15 m apart and wet the overall area and (2) under foliage irrigation which
involves three techniques; open hydroponic systems (OHS), micro irrigation and drip irrigation.
OHS irrigation is very technical and the whole system is computerized. Usually the irrigation
scheduling is pulsed with trees receiving water and nutrients at intervals each day as per the
requirement for any given day, but this is an extremely expensive method (Veldman 2001). With
micro irrigation, small cone sprays which are used to concentrate the wetting area are placed
under each tree. Wider sprayers are used at a later stage to enlarge the wetting area. However,
wider sprayers increase the likelihood of evaporation and use a vast amount of water (Veldman
2001). This factor motivated a move away from this system to drip irrigation, especially in water
scarce areas. Drip irrigation consists of 1-4 inline drippers that can also be used for fertigation
(applying liquid fertilizers to trees) which makes the system cost effective and also ensures a
better nutrient uptake (Veldman 2001). Throughout the world, farmers are increasingly adopting
the use of drip irrigation. As the quality and automation of these irrigation systems improves, the
feasibility of using them to apply insecticides, growth regulators and nutrients also increases
(Grout & Stephen 2005).
Understanding the growth and environmental parameters of a particular cropping system is
crucial for the successful implementation and use of microbial biological control agents.
Tritrophic interactions between crop plants, their pests and microbial control agents applied to
control pests cannot be ignored and must be considered together. A greater understanding of the
biology and ecology of crops will enhance our understanding of the environments in which
microbial control agents are required and better adapted strains or isolates of these control agents
may be applied.
1.1.7 The South African citrus export industry
The South African citrus industry was established in the early 1800s. However, it was not until
1926, with the founding of the single-desk exporter, the South Africa Co-operative Citrus
Exchange, that the industry started realising its enormous potential in the export market (Mather
& Greenberg 2003; Mather & Rowcroft 2004). Exports of citrus to the United Kingdom (UK)
started at the turn of the century; by the 1960s, South Africa was exporting over half of all
southern hemisphere fresh citrus produce and was ranked amongst the top five fresh citrus
5
exporters in the world (Mather 2003). In 1995, the citrus industry exported 43 million cartons of
citrus to 60 countries with the gross value of ZAR 1.6 billion (FABI 1998). New market
legislation was passed in 1996 which broke the export monopoly and allowed growers to choose
independent export agents, which inspired a strong private sector response (Mather 1999; Mather
& Greenberg 2003).
The South African citrus industry is now the second largest exporter of citrus in the world after
Spain (FAO: citrus Fruit Annual Statistics 2006), with Europe, Japan, China and the USA being
the largest export markets (CGA Annual Report 2008/9). In 2008, up to 89 million cartons (68%
of total citrus production in South Africa) of citrus fruit were exported, which generated an
annual income of approximately ZAR 5.1 billion for South Africa (Edmonds, pers. comm.).
Citrus varieties such as Valencia and Navel oranges are the most popular on the overseas
markets with approximately 41 million cartons and 20 million cartons of 15 kg each exported
respectively. These are followed by grapefruit (13 million cartons), lemons (8 million cartons)
and soft citrus (7 million cartons) (CGA Annual Report 2007/8).
1.1.8 Citrus pests
Citrus is host to as many as 875 insects and mites, however only 144 are well known major pest
species worldwide (Smith & Peña 2002). South Africa is known to harbor over 100 citrus pest
species of economic importance (Bedford 1998). Approximately 30-60% of known citrus pests
constitute the Hemiptera, such as aphids, scale insects, mealybugs, leafhoppers and whiteflies.
These are followed by the Lepidoptera (fruit boring moths, piercing moths, leaf rollers and leaf
miners), Acari, Coleoptera, Homoptera, Diptera and Thysanoptera (Smith & Peña 2002) (Table
1.1). As mentioned climatic conditions vary among production areas which restrict the
distribution of some citrus pests. Some pests are more sensitive to extremes in temperature and
humidity, thus each citrus producing region has its own complex of major and minor pests
(Bedford 1998). Citrus pests are generally more abundant and more difficult to control in the
hotter, low-lying areas of the Northern Province and Mpumalanga. In other areas such as
Citrusdal in the Western Cape, where temperature extremes vary considerably during winter and
summer, very few pests require regular control strategies (Bedford 1998).
6
Table 1.1 Important insect and mite pests of citrus in South Africa
Citrus pests
Common name
Scientific name
Armoured scale
Red scale*
Circular purple scale ▲
Citrus mussel scale
Aonidiella aurantii*
Chrysomphalus aonidium ▲
Cornuaspis becki
Soft scale
Soft brown scale
Soft green scale
Coccus hesperidum
Pulvinaria aethiopica
Wax scale
Citrus wax scale
White wax scale
Gascardia brevicauda
Gascardia destructor
Cottony cushion scale
Mealybugs
Cottony cushion scale ▲
Citrus mealybug ▲
Karoo thorn mealybug
Oleander mealybug
Spherical mealybug ▲
Icerya purchasi ▲
Planococcus citri ▲
Nipaecoccus vastator
Paracoccus burnerae
Nipaeoccus viridis ▲
Aphids
Black Citrus aphid ▲
Cotton aphid
Toxoptera citricidus ▲
Aphis gossypii
Leafhoppers
Citrus leafhopper ▲
Green Citrus leafhopper
Penthimeola bella ▲
Empoasca citrusa
Blackflies
Citrus blackfly
Spiny blackfly ▲
Aleurocanthus woglumi
Aleurocanthus spiniferus ▲
Psyllids
Thrips
Caterpillars
Citrus psylla*
Citrus thrips*
Citrus swallowtail
False codling moth*
Apple leaf roller
Citrus leaf roller
Citrus looper
African bollworm
Trioza erytreae*
Scirtothrips aurantii*
Papilio demodocus
Thaumatotibia leucotreta*
Tortrix capensana
Archips occidentalis
Ascotis selenaria reciprocaria
Helicoverpa armigera
Fruit flies
Mediterranean fruit fly *
Natal fruit fly *
Marula fruit fly
Ceratitis capitata *
Ceratitis rosa *
Ceratitis cosyra
Mites
Citrus bud mite ▲
Citrus grey mite
Citrus rust mite ▲
Citrus red mite ▲
Red spider mite
Citrus flat mite
Citrus silver mite ▲
Oriental spider mite ▲
Aceria sheldoni ▲
Calacarus citrifolii
Phyllocoptruta oleivora ▲
Panonychus citri ▲
Tetranychus cinnabarinus
Brevipalpus californicus
Polyphagotarsonemus latus ▲
Eutetranychus orientalis ▲
* represents key pests of citrus, ▲ represents major or occasional important pests of citrus, all
other pests are minor or sporadic pests of citrus (Annecke & Moran 1982; Bedford 1998; Smith
& Peña 2002).
7
1.1.9 Control of citrus pests - towards an integrated pest management approach
Pesticides used in the early days of citriculture in South Africa had very short residual action and
included: resinwash, lime sulphur, sulphur dust and nicotine sulphate (Bedford 1998). Until
1948, citrus pest control relied substantially on annual fumigation of trees which were enclosed
under tarpaulins, using the highly toxic gas, hydrogen cyanide (HCN) (Annecke & Moran 1982).
From 1948 onwards, fumigation and sulphur dust were replaced by parathion, an
organophosphate insecticide, which was much quicker to apply to trees and was used to combat
Californian red scale, Aonidiella aurantii and also controlled other important citrus pests
(Bedford 1998). It was noted however that resurgence of secondary pests (soft brown scale,
Coccus hesperidum and citrus red mite, Panonychus citri) occurred mainly as a result of the
disappearance of natural enemies; this was due to the toxic effect of the organophosphate sprays
(Annecke & Moran 1982). As a result of the resurgence of insect pests at this time, the cycle of
injudicious spraying of chemical insecticides continued which eventually lead to insecticide
resistance. This prompted the use of new insecticides (chlorpyrifos, mercaptothion and
methidathion), with newer active ingredients and the combining of chemical pesticides to kill
insects (Bedford 1998). As a result of this „pesticide treadmill‟ in an effort to control red scale,
citrus growers were forced to move toward an integrated pest management (IPM) approach and
the use of narrow range petroleum oils replaced organophosphates (Bedford 1998). Synthetic
pyrethroids were registered in 1988 to control citrus thrips, Scirtothrips aurantii, and were highly
successful to a point but were also found to disrupt populations of natural enemies and biological
control agents released to control scale insects. At this time insect growth regulators (IGR‟s)
were applied to citrus in an attempt to amalgamate control tactics as part of an IPM strategy to
control resurgences of red scale, following failed oil sprays (Bedford 1998). Tartar emetic and
bait sprays (methiocarb and formetanate) were introduced at this time as part of an IPM strategy
against thrips in citrus orchards. Further application of new products such as Confidor®
(imidacloprid) (Bayer CropScience, Germany), a stem applied systemic which was considered
more effective and safer to natural enemies than foliar sprays, already showed potential
disruption of the biological control of Icerya purchasi (Bedford 1998).
IPM had became an integral part of citrus production in the 1980s but later in the 1990s there
was a dramatic increase in the use of thripicides in citrus which disrupted the biological control
of red scale and mealybugs. The increased use of thripicides together with the reductions in
organophosphate usage led to drastic increases in mealybug populations during the 1990s
(Bedford 1998).
8
The control of citrus in South Africa has relied almost exclusively on chemical control in the last
100 years and biological control had been largely ignored or unstudied which is ironic because as
early as 1892, before the extensive use of chemicals in citrus, the Vedalia ladybird beetle
Rodolia cardinalis, was introduced to control I. purchasi and was highly successful (Bedford
1998). Bedford (1998) suggested that “the only way to avoid pesticide-induced outbreaks of
pests is to strive towards integrated control”. The term integrated control refers to the combined
usage of the advantageous aspects of both chemical and biological control, thus killing insect
pests with minimal disruption to natural enemies and maintaining a greater permanence of pest
suppression (Orr 2003). The South African citrus industry has now developed standardised
testing procedures, in collaboration with the Agricultural Research Council, to screen new plant
protection agents to ensure that they comply with citrus IPM programmes (Bedford 1998).
Further, numerous studies on the implementation of predators and parasitoids in citrus orchard
systems have been undertaken, reiterating the need for a reduction in pesticides which are highly
potent and have persistent residual action (Bedford 1998).
1.2 TARGET PESTS
Tephritid fruit flies, Ceratitis capitata (Wiedemann) and C. rosa (Karsch) constitute major pests
of citrus in South Africa and are also considered economically important fruit pests worldwide
(Ekesi et al. 2002; Smith & Peña 2002; Dimbi et al. 2003). Similarly, false codling moth,
Thaumatotibia leucotreta (Meyrick), is considered a major problem in most cultivated fruits,
vegetables and nuts in South Africa and other countries (Kirkman & Moore 2007; Stibick 2008).
The highly phytophagous nature of all three of these insect pests, their relative phytosanitary
importance and the economic importance of fruit loss makes it vital to control these pests in
citrus.
1.2.1 False codling moth, Thaumatotibia leucotreta
The false codling moth, Cryptophlebia (=Thaumatotibia) leucotreta (Lepidoptera: Tortricidae),
revised by Komai (1999), is regarded as a major insect pest on citrus in South Africa (Newton
1998; Van den Berg 2001). This moth was first described as a pest on citrus in KwaZulu-Natal
by Fuller in 1901, who called it the Natal codling moth; it was later found in the northern areas
9
of South Africa and named the orange codling moth (Newton 1998). False codling moth (T.
leucotreta) is found throughout sub- Saharan Africa (where it originates) as well as the
neighboring islands of the Indian and Atlantic Oceans (Stibick 2008) (Fig. 1.2). The species has
a broad host range, being recorded from over 80 different species of trees and shrubs, which
includes important economic crops such as avocados, citrus, cotton, litchis, macadamias, pome
and stone fruit (Van den Berg 2001; Venette et al. 2003; Kirkman 2007; Kirkman & Moore
2007; Stibick 2008). In citrus, navels appears to be the variety most heavily damaged by T.
leucotreta. Grapefruits and naartjies are less susceptible to attack by T. leucotreta, and larval
development is rarely, if ever, completed in lemons and limes, possibly due to the acidity content
in this cultivar (Newton 1998). Thaumatotibia leucotreta larvae are capable of developing in
hard, green fruit before control measures can be implemented and once the fruit is damaged, it
becomes susceptible to fungal infection and scavengers and drops to the ground (Annecke &
Moran 1982).
Figure 1.2 The geographical distribution of false codling moth, Thaumatotibia leucotreta in
Africa and neighbouring islands (Stibick 2008).
10
1.2.1.1 Life cycle of false codling moth
The life stages of T. leucotreta are described in detail by Stibick (2008). Female moths lay pearlwhite, oval eggs (which turn slightly reddish with a black spot shortly before neonate larvae
hatch) singly on the rind of fruit. At higher insect densities, more than one egg is laid on a single
fruit (Newton 1998). The eggs (Fig. 1.3 A) are flattened with a reticulate surface and measure
approximately 0.77 mm in length and 0.60 mm in width (Van den Berg 2001); egg incubation
varies according to temperature but generally lasts for up to 12 days in winter months (Kirkman
2007). There are five recognized larval instars which are determined by the width of the head
capsule (Kirkman 2007). Larval development varies according to temperature but averages 35
days in summer and 67 days in winter (Kirkman 2007). Neonate larvae are cream-white with a
dark brown head and measure approximately 1.5 mm in length and generally bore into fruit
within 24 hrs of hatching (Wysoki et al. 2002). The entire larval stage progresses inside the fruit
until the final instar larval stage, which is pinkish-red (Fig. 1.3 B) and measures 12-15 mm in
length, exits the fruit to pupate in the soil (Van den Berg 2001). Final instar larvae drop to the
ground below host plants and begin spinning cocoons with silk and soil particles (Wysoki et al.
2002; Stibick 2008). The pupal stage (Fig. 1.3 C), like both the egg and larval stages, is also
temperature dependent and can last from 21 to 80 days; there is no diapause stage (Stibick 2008).
Adult moths (Fig. 1.3 D) emerge with mottled dark grey forewings and pale, fringed hind wings
and generally live for 2 to 3 weeks in the field. The sex ratio of field populations is generally 1:1
(Newton 1998). Males are usually smaller than females and have a grayish anal tuft of scales and
a scent organ near the anal angle of each hind wing (Van den Berg 2001). A female moth can lay
up to 450 eggs during her lifetime, and as many as 5 or 6 generations a year may occur, however
this is influenced by several factors, including temperature, food availability and quality,
photoperiod, humidity, latitude and the effect of predators and diseases (Stibick 2008).
1.2.1.2 Economic damage
T. leucotreta causes annual losses in excess of ZAR 100 million to the South African citrus
industry, mainly as a result of pre-harvest fruit loss caused by cryptic, internal feeding of the
larvae, and as a result of post-harvest decay of fruit (Kirkman & Moore 2007). Larval feeding
and development can affect fruit development at any stage, causing premature ripening and
infested fruit may drop as early as November when they are no more than 15-20 cm in diameter
(Newton 1998). Estimates of long term crop losses in citrus in Nelspruit during 1971 were
recorded as 24-45 fruit per tree (Newton 1998). Foreign market rejection of fruit due to pest
11
presence is considered to be the most important factor pertaining to this recognized phytosanitary
pest and contributes to further loss of revenue for the industry (Moore 2002). False codling moth
has as yet not established in the United States of America despite numerous interceptions at port
entries. It does represent a significant import threat to the USA particularly because large areas in
the southern and south western United States are under agricultural production and these areas
have similar climate parameters to South Africa; establishment of the pest in the USA would
result in US$ billions in economic losses (Stibick 2008). Large export markets such as China
will only accept fruit from guaranteed T. leucotreta-free orchards and sensitive markets such as
the USA will reject large consignments of fruit due to the presence of T. leucotreta, which
constitutes further losses in revenue for the export industry (Kirkman 2007).
Figure 1.3 Thaumatotibia leucotreta (A) egg laid singly on the rind of fruit (B) fifth instar larva
with characteristic red colouring (C) sclerotised brown pupae (D) mottled grey adult moth. Photo
Credits: (A, C, D) River Bioscience (B) A.M. Varela, ICIPE
12
1.2.1.3 Control of false codling moth
Control of T. leucotreta in citrus in South Africa has included:
1.) Pre-harvest monitoring: The Lorelei trap which contains a female-based pheromone
dispenser and an adhesive is used to ascertain the numbers of moths present in or near orchards;
the threshold value of trap catches is 10 males per trap per week (Moore et al. 2008).
2.) Chemical control: Registered chemicals used include, two chitin synthesis inhibitors,
Alsystin® (Bayer CropScience, Germany) and Nomolt® (Cyanamid, South Africa). These
chemicals however are known to cause sterilization in predatory beetles and result in population
explosions of citrus red mite and oriental mite. Further, these chemicals may not be used on fruit
destined for export to the USA (Kirkman 2007). An organophosphate insecticide called PenncapM® (Elf Atochem, Netherlands) which kills larvae on contact and two pyrethroids, Meothrin
(Sanachem, South Africa) and cypermethrin (Agropharm, South Africa) are also registered for
use on T. leucotreta in citrus (Kirkman 2007).
3.) Cultural control: Practices such as orchard sanitation, can contribute largely to the control of
T. leucotreta; as much as 75% of fruit picked up weekly from orchard floors can still be infested
with T. leucotreta larvae (Moore & Kirkman 2009).
4.) Behavioural control: Mating disruption using synthetic female pheromones to confuse or
disorientate males finding females for mating can subsequently lead to a reduction of fertile eggs
in the field (Moore 2002). The attract and kill technique is another mating disruption method
which attracts male moths to a synthetic female pheromone. When the male moths make contact
with an incorporated insecticide they are killed (Kirkman 2007).
5.) Genetic control: Sterile insect technique (SIT) is a control measure that employs gamma
irradiation to induce sterility in male moths. This is followed by mass releases of sterilized male
moths into citrus areas where mating with females of wild populations results in infertile eggs
(Kirkman 2007).
6.) Post harvest control: includes cold sterilization of exported fruit which maintains fruit
sterility on route to lucrative markets such as China, Japan and the USA (Kirkman 2007).
7.) Biological control: The use of natural enemies against T. leucotreta includes a suite of insects
(mainly Hymenoptera) and microorganisms which vary in their respective levels of control.
Moore (2002) lists 14 hymenopteran egg larval and pupal parasitoids, of which
Trichogrammatoidea cryptophlebiae, Agathis bishopi and Apophua leucotreta are considered to
be common and elicit effective suppression of T. leucotreta. Three Tachinid fly species are
known to parasitize the larvae and pupae of T. leucotreta (Moore 2002). Further, three
13
hemipterous predators (two Orius sp. and Rhynocoris albopunctatus) and a mite (Pediculoides
sp.) exert limited natural control on T. leucotreta populations in the field (Newton 1998). The
use of microorganisms such as viruses and fungi has also contributed to the control of T.
leucotreta. Viruses in particular have been very successful, the Cryptophlebia leucotreta
granulovirus (CrleGV), is highly pathogenic to T. leucotreta and can cause extensive epizootics
in nature (Moore 2002). Likewise the entomopathogenic fungus Beauveria bassiana (Balsamo)
has been recorded on T. leucotreta pupae amongst the leaf litter in Zebediela Estate, although
attempts to culture and apply this fungal pathogen have been met with little success to date
(Newton 1998).
1.2.2 Fruit flies
Mediterranean fruit fly, Ceratitis capitata
Ceratitis capitata (Diptera: Tephritidae) is one of the world‟s most damaging fruit pest species
and originates from the tropical zones of Central Africa (Van den Berg 2001). This species is
widespread throughout the world and occurs in all South African provinces (Du Toit 1998).
Ceratitis capitata is a highly polyphagous insect species, recorded from over 260 different plant
species including some important economic crops such as avocado, coffee, citrus, custard apple,
granadilla, guavas, litchi, mangos and persimmons (Thomas et al. 2001; Van den Berg 2001).
Adult Mediterranean fruit fly are approximately 4 mm long with a black scutellum and defined
ivory markings which do not enclose three black areas, as in the Natal fruit fly (Fig. 1.4 B). The
males are easily recognised by the two spatulate setae on the head (Du Toit 1998).
Natal Fruit Fly, Ceratitis rosa
Ceratitis rosa (Diptera: Tephritidae) is an economically important fruit fly species with a
widespread distribution throughout Africa and the islands of Mauritius and Réunion (Botha et al.
2004). Its distribution in South Africa is more humidity dependent than Ceratitis capitata and is
confined to the subtropical regions of KwaZulu-Natal (where it was first found), the Northern
Province, Mpumalanga and along the coastal regions of the Western and Eastern Cape Provinces
(Ware 2002; de Meyer et al. 2008). Ceratitis rosa is a pest of over 100 different plant species,
which include important agricultural species such as apple, apricot, avocado, blackberry, citrus,
coffee, custard apple, fig, guava, Kei-apple or umkokolo, loquat, mango, papaya, peach, pear,
14
persimmon, plum, quince, roseapple, soursop, and Chrysophyllum natalense, Eugenia cordata,
Garcinia livingstonei, Opuntia sp., Rawsonia lucida, Solanum auriculatum, and Solanum
giganteum (Weems 2002; Botha et al. 2004). The appearance of this fruit fly looks very similar
to that of the Mediterranean fruit fly; however it has a larger body size (length of the fly 4 to 5
mm). Females can be separated from other fruit fly species by their thoraxes which bear ivory
bands enclosing three black areas on the scutellum (Fig. 1.4 C). Males have characteristic dark
bristles on the tibia of the mid-pair of legs (Weems 2002).
1.2.2.1 Life cycle of fruit flies
The life cycles of both Mediterranean and Natal fruit flies are similar (Du Toit 1998) and are
discussed together. Ceratitis females puncture the rind of the fruit and oviposit clusters of about
10-20 small, white, sausage-shaped eggs in the rind tissue. Larvae hatch and burrow through the
flesh (Smith & Peña 2002) (Fig. 1.4 A). The life cycle is temperature dependant and is shorter in
warmer climates and takes longer in cooler environments. Fruit flies lay eggs on mature green
and ripening fruit with some species laying eggs in unripe fruit (Botha et al. 2004). Two to four
days after oviposition the eggs hatch. The larvae emerge as white maggots and pass through
three larval instars, which usually take 5 to 11 days (Thomas et al. 2001) (Fig. 1.4 A). After 4 to
17 days the third instar maggots leave the fruit, making holes in the skin, and drop to pupate in
the soil; this pupal stage may last for 10 to 20 days (Botha et al. 2004). The pupae are white,
brown or black and approximately 4 to 6 mm long (Du Toit 1998). They are found in the soil 2
to 5 cm beneath the host plant (Thomas et al. 2001). Adult flies emerge from the pupae 10 to 15
days after pupation (Du Toit 1998). Oviposition may take place as early as four to five days after
emergence in warm temperatures but can take several days after emergence at 20°C; and adult
longevity is 2 to 3 months depending on the season (Thomas et al. 2001). Some Ceratitis species
populations can produce up to 10 generations per annum depending on the availability of host
plants and suitable climatic conditions (Thomas et al. 2001; Weems 2002).
15
Figure 1.4 (A) Pupa, larva and eggs of Ceratitis capitata (B) Adult C. capitata fruit fly
(C) Adult female C. rosa fruit fly. Photo Credits: (A) California Department of Food and
Agriculture (B) Scott Bauer, USDA (C) Citrus Research International.
1.2.2.2 Economic damage
The economic importance of fruit flies has been well described by Du Toit (1998); Ekesi et al.
(2002) and Dimbi et al. (2003, 2004). Fruit flies affect the production of quality fruits by
infestation which causes both direct and indirect loss of product. Direct loss occurs through fruit
flies feeding on citrus which causes fruit drop and renders the fruit inedible because of fungal
infections which invade the fruit after puncture marks are created by ovipositing females (Du
Toit 1998). Indirect loss arises from phytosanitary restrictions imposed by importing countries to
prevent entry and establishment of fruit flies (Ekesi et al. 2007).
1.2.2.3 Control of fruit flies
Control methods against fruit flies in South African citrus include:
1.) Pre-harvest monitoring: Includes the use of Sensus® traps which contain synthetic attractants
to ascertain the number of fruit flies present in or near orchards (Smith & Peña 2002). Suitable
economic thresholds have not been determined for the control of fruit fly in all citrus areas but a
recommendation for producers to try and reduce fly numbers to an average of one female fly per
trap per week when Questlure® (Quest Developments CC, Brits) is used as an attractant is
suggested by Moore et al. 2008. Treatment thresholds for use with the Sensus trap are four fruit
flies per trap per week when Capilure® is used. If Ceratitislure® is used the thresh-hold is either
zero female fruit flies or four male fruit flies per trap per week (Moore et al. 2008).
2.) Cultural control: Orchard sanitation involves the collection and destruction of all senesced
citrus fruit in orchards (Ekesi et al. 2007). Fruit flies are notorious for infesting decaying fruit
16
and piles of fallen fruit can cause unnecessary build up of fruit fly numbers in orchards (Moore
& Kirkman 2009).
3.) Behavioural control: The most common method used to control fruit flies in South Africa is
the attract and kill technique, whereby a protein attractant is mixed with a toxicant in a bait
station (Ware et al. 2003). The male annihilation technique involves the arrangement of high
density trapping stations in which a female-based attractant is combined with an insecticide; the
aim is to reduce the male population so that low level mating occurs (Ekesi et al. 2007).
4.) Genetic control: Sterile insect technique (SIT) includes the genetic sterilization of male fruit
flies but does not affect their ability to mate; the aim is to introduce a genetic mutation in the
male sperm which induces sterility in the sperm but does not affect sperm function, males then
mate but no eggs can be fertilized (Ekesi et al. 2007).
5.) Chemical control: In citrus orchards chemicals are usually applied as weekly applications of
toxic bait in the form of large droplets over half the canopy of every alternate tree. Baits consist
of mixtures of mercaptothion or trichlorfon and protein hydrolysate (Du Toit 1998). The method
targets adult flies, mainly females (because they require a protein meal before oviposition) and
aims to attract and poison them before they cause damage to fruit
6.) Biological control: The use of natural enemies such as predators or parasitoids has always
been appealing because they are safe and self perpetuating. There are three fruit fly parasitoids
which occur in South Africa, Opius concolor and O. humilis (Braconidae) which both attack the
larvae, and a pupal parasitoid, Trichopria capesis (Diapriidae). However, these natural enemies
are unable to control fruit fly in citrus (Du Toit 1998). The role of entomopathogenic fungi as it
is known from these target insect pests species is such that epizootics of fungal disease of fruit
fly in nature is uncommon and natural infection rates are low (Ekesi et al. 2007). There have
only been a few reports of natural infection of Ceratitis capitata adults by the entomopathogenic
fungus Entomophthora muscae and E. schizophorae in Israel (Ekesi et al. 2007). Ekesi et al.
(2007) reported that Isaria fumosorosea and Beauveria bassiana also caused natural infections in
C. capitata populations in Spain.
1.2.3 Justification for alternative control tactics
The production of citrus in South Africa plays a major role in providing food security for the
country; it further provides employment and opportunity for foreign export and trade. As it
currently stands, the South African citrus industry is the second largest exporter of citrus in the
world after Spain (FAO: Citrus Fruit Annual Statistics 2006) and generates an annual income of
17
approximately ZAR R5.1 billion in currency for South Africa (Edmonds, pers. comm.). Present
control of insect pests on citrus still relies on insecticide use because the cosmetic integrity of
this crop is vital for export. Citrus exports to the European Union (EU) however have now been
restricted by the recent lowering of maximum residue limits of certain insecticides, such as
dimethoate (Grout & Stephen 2005). This is due to the EU‟s drive to reduce unnecessary and
harmful chemical pesticide use (Mather 1999) and, to a lesser extent, the recent consumer
interest in organically produced foods, which prohibits the use of pesticides (Zehnder et al.
2007). Further large export markets will continue to impose strict phytosanitary restrictions on
key citrus pests such as Natal fruit fly and false codling moth because these pests have not as yet
become established in Europe, USA, China or Japan (Botha et al. 2004; Stibick 2008), which
makes their control in citrus orchards crucial. South African citrus growers exporting fruit to
these overseas markets are obliged to reconsider their pest control practices, and in particular the
use of some chemical pesticides. As a consequence, sustainable agriculture may rely increasingly
on alternative control strategies that are environmentally friendly and reduce the amount of
human contact with pesticides; further the use of these alternative control strategies within IPM
programmes must be considered. To maintain a stronghold in overseas export markets, it has
become important for the South African citrus industry to investigate other facets of IPM, which
includes the study and use of microbial control agents, such as entomopathogenic fungi
(hereafter referred to as EP fungi). The biology of the key citrus pests, T. leucotreta; C. capitata
and C. rosa, includes their dropping from their host plants to pupate in the soil or leaf litter
below citrus trees. Since most EP fungi are soil-borne microorganisms, the development and
formulation of alternative control strategies using these fungi as subterranean control agents,
targeted at larvae and pupae in the soil, can potentially benefit existing IPM management of
citrus in South Africa.
1.3 ENTOMOPATHOGENIC FUNGI
1.3.1 General taxonomy
Fungi are a diverse group of unicellular (yeasts) or hyphal (filamentous fungi), eukaryotic
microorganisms that are all heterotrophic, meaning that they have absorptive nutrition and
reproduce either sexually or asexually by means of spores (Madigan et al. 2005). The most
common characteristic used to assign fungi to their four divisions, based on genetic relationships,
relies on sexual fruiting structures (Inglis et al. 2001). Until recently, it was thought that
18
Hyphomycete EP fungi had lost the ability to produce sexual spores. These fungi were
traditionally placed in the division, Deuteromycota within the superficial class Hyphomycete.
They are characterized by mycelia forms that produce asexual spores called conidia which are
borne on specialized conidiogenous cells (Inglis et al. 2001). However, with the advent of
molecular technologies, some EP taxa have now been connected with a sexual state and most of
these taxa exhibit ascomycetous affinities which include characteristics such as septate hyphae
with lamellate hyphal walls with a thin electron-dense outer layer and a thicker electrontransparent inner layer (Inglis et al. 2001; Rehner & Buckley 2005). When a mitotic fungus has
been connected with a sexual state it is called the „anamorph‟. The sexual or perfect state is
called the „teleomorphic state‟. For example, Beauveria bassiana (the anamorphic state) has been
linked to the teleomorphic species Cordyceps bassiana (Rehner & Buckley 2005). Most EP
fungi now fall under the order Hypocreales which belongs to the division Ascomycota and the
most common EP hyphomycetous genera include: Aspergillus (Micheli), Beauveria (Balsamo),
Culicinomyces, Hirsutella (Patouillard), Metarhizium (Metschnikoff), Nomuraea (Yasuda),
Isaria (=Paecilomyces) (Samson), Tolypocladium and Lecanicillium (=Verticillium) (Gams and
Zare) (Inglis et al. 2001).
1.3.2 Fungal infection process and environmental cycling
All EP fungi have the same basic mode of action when infecting insect hosts and are unique in
that they can infect insect hosts through the cuticle (Goettel 1995). The conidia, or asexual
spores, come into direct contact with the host and adhere to the cuticle via nonspecific
hydrophobic mechanisms (Inglis et al. 2001). Under specific environmental conditions the
conidia germinate and a germ tube, or appressorium (penetration-peg structure), are produced
(Inglis et al. 2001) (Fig. 1.5 A). The cuticle is penetrated by a combination of mechanical
pressure from the appressorium and the action of cuticle-degrading enzymes, such as trypsin,
metalloproteases and aminopeptidases (Bidochka & Small 2005). The fungus grows by
vegetative growth in the host haemocoel and external conidia are produced upon death of the
host when fungal hyphae exit through the less sclerotised areas of the cuticle (Inglis et al. 2001)
(Fig. 1.5 B). Some EP fungal species, such as Metarhizium and Beauveria, produce powerful
insecticidal cyclic peptides called destruxins and a toxic metabolite called oosporein (Goettel et
al. 1989; Inglis et al. 2001). Comprehensive reviews on the function and mode of action of
destruxins and oosporeins are reported in the literature (Zimmermann 2007a, b). Fungal strains
which produce these toxins are sought after for commercialisation because although fungi grow
19
sparsely in the insect haemocoel, they kill the insect relatively quickly. Strains which do not
produce these toxins grow profusely in the insect haemocoel but take much longer to kill their
hosts (Bidochka & Small 2005). The host cuticle is the first line of defense against fungal
infection and has a central role in determining fungal specificity. If the fungus breaches the
cuticle, successful infection can only occur if the fungus can overcome the innate immune
response of the insect (Hoffman et al. 1999). Insects respond in both a cellular and humoral
manner to fungal infection, with immune activation occurring as early as the point of cuticle
degradation during the penetration step (Gillespie et al. 2000). Fungi overcome host defense
responses by developing cryptic growth forms that are effectively masked from the insect
defense responses, or by the production of immune-modulating substances that suppress the host
defense system as mentioned (Bidochka & Small 2005) (Fig. 1.5 A).
Figure 1.5 (A) Schematic diagram of an EP fungus invading the host insect cuticle
(B) Cordyceps sp. fungus growing out of the thorax of an unidentified fly species. Photo Credits:
(A) Thomas & Read 2007 (B) Lawrence Gilbert, University of Texas Austin
1.3.3 Fungi as biological control agents
Soil-inhabiting EP fungi are an important and widespread component of most terrestrial
ecosystems and play a key role in regulating some soil-dwelling insect populations (Meyling &
Eilenberg 2007; Quesada-Moraga et al. 2007). The majority of fungi that naturally regulate
insect and mite populations in the field belong to the orders Hypocreales and Entomophthorales.
20
However, the latter has not been possible to mass produce and commercialisation or inundative
application of this fungal order has not occurred, unlike the Hypocreales which are amenable to
mass production and commercialisation (Dolinski & Lacey 2007).
The use of microbial control agents specifically EP fungi, have been investigated for the control
of a wide range of orchard pests (Cross et al. 1999; Puterka 1999; Lacey & Shapiro-Ilan 2003;
Alves et al. 2005; Castrillo et al. 2005; Dolinski & Lacey 2007; Lacey & Shapiro-Ilan 2008).
Further the use of EP fungi against various citrus pests has been successful (Table 1.2).
Numerous entomopathogens (fungi, viruses, nematodes and microsporidia) impose a high level
of natural control on citrus pests throughout much of the humid, warmer climates of the world.
EP fungi appear to be the most prevalent pathogens attacking citrus pests; to date over 24
mycopathogens have been recorded (McCoy et al. 2007).
The use of EP fungi in the form of mycoinsecticides has many desirable traits, EP fungi leave no
toxic residues on crops, and are generally harmless to beneficial and other non target insects
(Goettel 1995; Zimmermann 2007a, b). They also pose minimal risks to humans and animals and
there is no damage to the environment because ground water and riparian habitats are not
contaminated by EP fungi, as is the case with pesticides (Zimmerman 2007a). Their host specific
nature enhances their potential role in IPM, as it preserves natural enemies which then make a
greater contribution to the overall regulation of pests (Inglis et al. 2001). The main impact of EP
fungi on pest populations however may only occur once economic thresholds have been
surpassed. Because chemical control can be really effective at immediately suppressing a
growing pest population it is unlikely that EP fungi will ever supersede chemical control and it is
highly unlikely that the disuse of chemicals will occur (Inglis et al. 2001). Thus, it becomes more
important to select fungal strains which are pesticide tolerant and can survive in the cropping
systems for which they are required. The use of EP fungi in cropping systems will require
augmentation of isolates, most often applied as a prophylactic treatment rather than a once off
application (Ekesi et al. 2007). Some of the most versatile and widely used fungal biological
control agents are from the order Hypocreales and some important genera are discussed further
below.
21
Table 1.2 EP fungi used in the control of citrus pests worldwide (Dolinski & Lacey 2007)
Common name
Pest species
Citrus rust mite
Phyllocoptruta oleivora Beauveria bassiana
Hirsutella thompsonii
Alves et al. (2005)
McCoy (1996)
Citrus red mite
Panonychus citri
B. bassiana
Shi & Feng (2006)
Citrus broad mite
Polyphagotarsonemus
latus
H. thompsonii,
B. bassiana, Isaria
fumosoroseus
Peña et al. (1996)
False spider mite
Brevipalpus phoenicis
H. thompsonii
Rossi-Zalaf & Alves (2006)
Whiteflies
Dialeurodes citri
Aschersonia spp.
Dolinski & Lacey (2007)
Citrus black fly
Aleurocanthus woglumi Aschersonia aleyrodis
Elizondo & Quezada (1990)
Scales
Parlatoria ziziphi
El-Choubassi et al. (2001)
Aphids
Fungal control
A. aleyrodis
A. goldiana
References
Selenaspidus articulates A. aleyrodis
Gravena et al. (1988
Chrysomphalus
aonidum
Lepidosaphes beckii
Podonectria coccicol
Pseudomicrocera
henningsii,
Sphaerostilbe
aurantiicola
Yen & Tsai (1969)
Toxoptera citricida
B. bassiana
Poprawski et al. (1999)
Diaprepes root weevil Diaprepes abbreviatus B. bassiana
McCoy et al. (2007)
Fruit flies
B. bassiana
M. anisopliae
Ekesi et al. (2002; 2003;
2005)
Dimbi et al. (2003a; 2003b)
Castillo et al. (2000)
I. fumosoroseus
B. brongniartii
Konstantopoulou &
Mazomenos (2005)
Ceratitis capitata
22
1.3.4 Genus Beauveria
Beauveria (Balsamo) Vuillemin (Ascomycota: Hypocreales) is a cosmopolitan genus of soil
borne EP moulds, first discovered as a devastating disease which affected European silkworms
in the 18th and 19th centuries, known as the white muscardine disease (Rehner 2005). Although
the initial investigation of Beauveria was instigated from a need to protect domesticated insects
such as silkworms and honey bees, it is also an important natural pathogen of insects, capable of
infecting more than 700 species of arthropods (Lord 2005). Its hosts include many economically
important pests and its wide variation in virulence towards different insect hosts makes it one of
the more versatile candidates of EP fungi for the biological control of insect pests (Rehner 2005).
Other aspects which have contributed to the large amount of research focused on this fungal
genus are firstly, its broad distribution in nature which makes Beauveria easy to recognise and
has lead to it being noted as the most frequently encountered of all EP fungi (Rehner 2005).
Secondly, Beauveria is an extremely traceable organism as it can be easily isolated from insect
cadavers or from the soil using simple media, antibiotics and selective agents (Rehner 2005).
Thirdly, the genus can be easily cultured in the laboratory on simple media and can be conserved
by storing conidia in glycerol solutions at ultra-low temperatures or by freeze-drying (Rehner
2005). Furthermore, Beauveria is amenable to large-scale production and formulation as a
mycoinsecticide. Despite two centuries of research and several comprehensive taxonomic
studies, there are still significant problems in the identification, taxonomy and nomenclature of
species in this genus and progress in elucidating the genetics and evolution of Beauveria is just
beginning (Rehner 2005; Rehner & Buckley 2005).
1.3.4.1 Taxonomy
Vuillemin (1912) formally described the genus Beauveria, assigning Botrytis bassiana as the
type species, this later became Beauveria bassiana in recognition of J. Beauverie in 1914, after
his invaluable contribution to the study of white muscardine disease (Rehner 2005). The genus is
easy to distinguish morphologically; its most distinctive features are the sympodial to whorled
clusters of short-globose to flask-shaped conidiogenous cells that produce a succession of onecelled, sessile and hyaline conidia on an elongating „zigzag‟ rachis (Domsch et al. 2007) (Fig.
1.6 A).
23
In culture, Beauveria species typically produce white coloured mycelium and conidia (Fig. 1.6
B), although some isolates may produce yellow pigment in the older, central parts of the colony
(Domsch et al. 2007). Colony growth tends to be rapid, and the texture of the mycelium is
typically lanate to woolly and synnemalike projections can occasionally be observed (Rehner
2005). Beauveria conidial production can often be profuse, frequently creating a chalky, mealy
or powdery appearance on the colony surface (Rehner 2005). Many isolates can excrete a red
pigment into the culture medium, although this does not occur under all culture conditions
(Rehner 2005).
Figure 1.6 (A) A photomicrograph showing three conidiogenous cells on a hypha of Beauveria
bassiana. The asexual conidia are produced in acropetal succession (youngest at the tip) (B) B.
bassiana grown on Sauboraud dextrose agar (SDA) at 25 °C. Photo Credits: (A) George Barron,
University of Guelph, Canada (B) T. Goble, Rhodes University, South Africa.
The conidial form is the only morphological feature of the genus that has proven helpful for
species delineation, as conidia vary from globose, ellipsoid, cylindrical and vermiform and range
in sizes from 1.8 to 6.0 µm in their greatest dimensions (Rehner 2005; Domsch et al. 2007). In
total 49 species have been classified in the genus of which 22 are currently considered valued
(Rehner, 2005). Overlapping variation in the size of conidia among many species has raised
questions of the validity of routine species identifications. Further, cultural features of Beauveria
are influenced by culture conditions and are thus unreliable as taxonomic features (Rehner 2005).
Six morphological species can be discerned in the genus Beauveria: B. amorpha, B. bassiana, B.
brongniartii, B. caledonica, B. velata and B. vermiconia (Fig. 1.7). There is however much
confusion with regard to the species B. bassiana for which many morphologically similar species
have been described; as it stands seven species are synonymous with B. bassiana (Rehner 2005).
24
Current molecular techniques based on the comparison of available internal transcribed spacer
(ITS) sequences and other molecular markers among species, has shown great promise in
elucidating the genetic taxonomy of this genus (Rehner & Buckley 2005; Meyling et al. 2009).
An important inference obtained using molecular analyses was the discovery of the close
relatedness between the asexual or anamorphic genus Beauveria, and the sexual or teleomorphic
genus Cordyceps. Beauveria bassiana s.s. reproduces asexually by mitosporic conidia and for
some time it has been presumed to be exclusively clonal (Meyling et al. 2009). Phylogenetic and
developmental studies have linked B. bassiana s.s. to the Asian sexual species Cordyceps
bassiana which provides evidence that B. bassiana is in fact facultatively sexual (Meyling et al.
2009).
Figure 1.7 The phylogenetic relationships of the genus Beauveria based on several independent
Bayesian and parsimony analyses. Clades A-E are labelled according to their morphological
species identification and micrographs of representative conidia (Rehner 2005).
25
1.3.5 Genus Metarhizium
Metarhizium (Metschnikoff) (Hypocreales: Ascomycota) is a haploid cosmopolitan genus of soil
borne EP moulds. Elie Metchnikoff, a Russian microbiologist, was the first person to isolate and
name the green muscardine fungus, Entomophthora anisopliae in the 1880s (now known as
Metarhizium anisopliae) which was found infecting the grain beetles he worked on (Lord 2005).
Metarhizium species have a wide range of virulence and are known to infect more than 200
different insect species, many of which are major agricultural pests, such as sugarcane stemborers, scarab grubs and grasshoppers (Bidochka & Small 2005). Further Metarhizium is capable
of infecting insects from over 14 different orders including some non target orders such as
Malacostrata (Amphipods), Acari, Ephemeroptera and Dermaptera (Zimmermann 2007a).
1.3.5.1 Taxonomy
As with Beauveria, the earlier classification of this genus was based on morphological
characteristics, but because these characteristics are fairly limited it has complicated the
taxonomy (Bidochka & Small 2005). The genus is morphologically defined on the basis of the
arrangement of the phialides which bear chains and columns of green, dry and slightly ovoid
conidia (Zimmermann 2007a) (Fig. 1.8 A,B). Tulloch used the morphological characteristics of
the conidia to distinguish between the two forms of M. anisopliae; the short-spore (5-8 µm) form
called M. anisopliae var. anisopliae and the long-conidial form (10-16 µm) called M. anisopliae
var. majus (Zimmermann 2007a). Since then nine additional taxa have been described, although
each may not be formally recognised as a separate Metarhizium species (Bidochka & Small
2005). Metarhizium flavoviride (Gams and Rozsypal) (Fig. 1.8 C, D) differs from M. anisopliae
by having wider, greyish to yellow conidia (Zimmermann 2007a).
A population genetic analysis by St Leger et al. in 1992, using allozyme markers however
discounted any evidence for a tight association between morphological characteristics such as
conidial shape and colour (Zimmermann 2007a). DNA and other molecular markers have
recognised areas of consensus concerning the taxonomy of Metarhizium. (1) Metarhizium
anisopliae is recognised as a monophyletic group but that the designation of varieties or cryptic
species within the group is still unclear. (2) Metarhizium anisopliae is related to M. flavoviride
but can be distinguished using molecular markers. There seems to be genetically distinguishable
clades or cryptic species within the M. anisopliae and M. flavoviride. (3) It is estimated that 1015 different cryptic species comprise the genus Metarhizium and M. anisopliae; M. flavoviride
26
and M. album are recognised as separate species. (4) An association of M. anisopliae genotypes
occurs with habitat type in temperate and polar regions and (5) Southeast Asia is the probable
origin of the evolution and diversity of M. anisopliae (Bidochka & Small 2005; Zimmermann
2007a). Until 1991, there was no teleomorph of the genus Metarhizium, until Cordyceps taii was
discovered which had an anamorphic state called Metarhizium taii, later Cordyceps
brittlebankisoides was isolated which proved the teleomorph of M. anisopliae var. majus
(Zimmermann 2007a). A taxonomic revision of Metarhizium is greatly needed however this
requires larger data sets as well as informative nucleotide characteristics which are not available
at this time (Bidochka & Small 2005).
Figure 1.8 (A) A transmission electron micrograph showing conidiogenous cells on the hyphae
of Metarhizium anisopliae (B) M. anisopliae grown on Sauboraud dextrose agar (SDA) at 25 °C
(C) Conidia of M. flavoviride at 1000X under oil immersion (Zeiss Imager A1 DIC microscope
with Canon Powershot (D) M. flavoviride grown on Sauboraud dextrose agar (SDA) at 25°C.
Photo Credits: (A) Fumio Ihara, National Institute of Fruit Tree Science, Japan (B, D) T. Goble,
Rhodes University, South Africa (C) Dan Johnson, University of Lethbridge, Canada
27
Several commercial endeavours have registered strains of Metarhizium as the active ingredient of
various products for insect pest management namely: Metarhizium 50® (AgoBiocontrol) for
garden pest control in Columbia; Biogreen® and Green Guard® in Australia for red-headed
cockchafer, Adoryphouse couloni and locusts respectively, BIO 1020® in Germany for vine
weevils, Otiorhynchus sulcatus and Green Muscle® (Biological Control Products, South Africa)
against locusts, Locustana pardalina and grasshoppers (Bidochka & Small 2005). Many other
registered Metarhizium products are currently available and a comprehensive list can be found in
Faria & Wraight (2007). There are many criteria which become important in selecting a strain of
Metarhizium for commercialisation: a high level of virulence towards the target pest, the
utilisation of indigenous strains is preferable because exotic strains could be politically and
ecologically problematic. Finally, the ability to distinguish and track the formulated strain
against background populations of EP fungi, either through genotyping or genetic tagging
becomes crucial to ascertaining the effectiveness of the applied strain (Bidochka & Small 2005).
1.3.6 Genus Conidiobolus
Conidiobolus is the only genus of the entomophthorales which is commonly isolated from the
soil environment (Domsch et al. 2007). The vegetative mycelium consists of short hyphal
bodies. Colonies grow rapidly and are flat, cream in colour becoming folded and covered by a
fine powdery surface mycelium and conidiophores (Fig. 1.9 B). The colony colour becomes tan
with age. The conidia are forcibly discharged by the conidiophores which are simple, forming
solitary, terminal conidia which are spherical (10-25 µm in diameter), single celled and have
prominent papilla (Domsch et al. 2007) (Fig. 1.9 A). The conidia may also produce hair-like
structures called villae, and when conidia germinate they produce either (1) single or multiple
hyphal tubes that may also become conidiophores or (2) produce multiple short conidiophores,
each bearing a small secondary conidium (Domsch et al. 2007).
28
Figure 1.9 (A) A photomicrograph of Conidiobolus coronatus showing characteristic conidia
with prominent papilla (B) C. coronatus grown on Sauboraud dextrose agar (SDA) at 25°C
(photo credits: (A) Dr. K. Salfelder, Universidad de Los Andes, Venezuela (B) T. Goble, Rhodes
University, South Africa
1.3.7 Factors influencing the efficacy of EP fungi
A variety of factors may influence fungal host range and the specificity of fungal pathogens, for
example: the fungal strain, the host‟s physiological state, nutrition, defense, host mechanisms
and the insect cuticle (Inglis et al. 2001). Further, the development of epizootics is controlled by
a variety of environmental factors and is influenced by the distribution of both the host and
pathogen populations (Goettel 1995). Fungal genera with broad host ranges, such as Beauveria
and Metarhizium are generally facultative pathogens which are ubiquitous in the soil and are
prime candidates for inundative biological control of insect pests (Goettel 1995).
1.3.7.1 The insect host
There is a complex array of morphological and physiological factors which influence the
susceptibility of insect pests to EP fungi. One of the most important concepts however is that
„stressed insects‟, either through overcrowding, poor nutrition, exposure to chemicals or harsh
environmental factors, are more susceptible to fungi than non-stressed insects (Inglis et al.
2001). Inadequate insect nutrition from feeding on plants which produce volatiles or alkaloids
and stress the insect, can greatly enhance the efficacy of EP fungi (Inglis et al. 2001); conversely
insects may feed on plants such as cow-pea and obtain plant-derived fungistatic compounds
which greatly reduces fungal efficacy (Ekesi et al. 2000). Another important factor is the
developmental stage of an insect pest as not all stages in an insect‟s life cycle are equally
susceptible to fungal infection. In some cases, as with the susceptibility of immature stages of the
29
legume flower thrips, Megalurothrips sjostedti, to the EP fungus Metarhizium anisopliae, the
adult stage was found to be more susceptible to infection than the larval and pupal stages (Ekesi
& Maniania 2000). In other cases the adults of the western flower thrips, Frankliniella
occidentalis were more susceptible to L. lecanii than the nymphs (Vestergaard et al. 1995).
Differential susceptibility in insect host stage is likely further related to loss of infection through
ecdysis, as the insect molts during instar development, so it may loose fungal inocula as the
cuticle is removed. Insect density is also very important for the development of epizootics, as
insect density increases there is a higher probability of insects coming into contact with infected
individuals. The behaviour of insects can also greatly affect the persistence of fungal disease, for
example sun-basking (to increase body temperature) and crowding behaviour (to elicit a
behavioural fever to raise the body temperature) in locusts has been shown to slow the infection
rate of Metarhizium anisopliae var. acridum (Wilson et al. 2002). Grooming behaviour in
termites can spread the conidia of EP fungi to other individuals (Inglis et al. 2001), but this
sanitation behaviour can also remove inoculum and limit the spread of disease and instill
resistance in some social insect groups (Yanagawa et al. 2008, 2009).
1.3.7.2 The environment
Environmental factors such as solar radiation, temperature, rainfall and wind interact with each
other and have dramatic effects on the efficacy of fungi and the stability of EP fungi in the field.
Solar radiation
The effect of irradiance, particularly in epigeal habitats, is one of the most important
environmental parameters for the persistence of EP fungal spores; exposure to sunlight for even
a few hours, particularly the UV-B portion of the spectrum (285 - 315 nm) can completely
inactivate the infectious conidia of Metarhizium (Braga et al. 2001). There are significant
differences in susceptibility to irradiation among fungal strains and species, and mechanisms for
resistance against irradiation should be considered when selecting candidate EP fungi for product
formulation (Inglis et al. 2001). Microhabitats play an important role in enhancing the
persistence of fungal conidia by providing shade under tree canopies for example (Inglis et al.
2001). The rapid inactivation of propagules by sunlight is a major hindrance to the successful
commercialisation of entomopathogens and thus much research has focused on enhancing
conidial persistence in the field by using solar blockers or UV-B absorbing substances
30
(sunscreens such as clay) and optical brighteners, such as Tinopal™ (Ciba Chemical Company,
Switerland) or stilbene (Ibargutxi et al. 2008). Unlike the canopy layer, the soil environment is
less prone to UV-B irradiance penetration and there are less dramatic fluctuations in temperature
(Ansari et al. 2007), making this habitat the seeming ideal place for the persistence of conidia
when considering commercial product application (Ekesi et al. 2003). However, temperature
conditions in the top few centimeters of the soil layer may be well over 40°C in some instances,
therefore incorporation of conidia into soil to enhance persistence will depend on climatic factors
as well as shading or depth of conidia in the soil (Inglis et al. 2001).
Temperature
Ambient temperature influences the rate of insect infection and may restrict the use of EP fungi
in biological control. Most EP fungi are mesophilic (they can grow at a range of temperatures)
but grow optimally at temperatures between 25 and 35°C. Species such as Beauveria bassiana
can grow at a wide range of temperatures (8 to 35°C) with a maximum threshold of 37°C
(Fernandes et al. 2008). Selecting EP fungal strains which can withstand environmental
temperature fluctuations is an important consideration in the development of a commercial
product and often strains that come from hotter areas perform better at higher temperatures and
likewise those that come from cold climates perform better at lower temperatures (Inglis et al.
2001). Recent evidence suggests that even when Beauveria strains were found in the same
geographic location their variability to thermotolerance was great, indicating that microhabitat
adaption is just as important as the macro-environment factor (Fernandes et al. 2008). However,
M. anisopliae isolates found in agricultural soils were found to be better adapted to heat and UVB exposure than isolates which were found in cold, forested habitats of similar latitude
(Bidochka et al. 2001, 2002; Fernandes et al. 2008). There is general agreement that fungal
isolates from higher latitudes are more cold-tolerant than isolates from nearer the equator
(Rangel et al. 2005). Behaviour such as basking and fevers which increase host temperature may
also influence the rate and severity of fungal infection and thus selecting commercial fungal
strains which elicit better infection at higher temperatures should be considered.
Relative humidity
Environmental moisture is a very important parameter for the persistence and germination of
conidia in nature. Conidiogenesis on insect surfaces requires high moisture (Inglis et al. 2001), if
31
this factor is low, conidia production slows and the horizontal transmission of spores is reduced,
slowing the transfer of infection. A number of studies have indicated that dry conditions
immediately after the application of EP fungi are less injurious than previously thought as
Beauveria and Metarhizium strains were able to germinate and effectively infect insects at low
ambient humidity, provided that there was sufficient moisture within the microhabitats (Inglis et
al. 2001). The requirement for high environmental moisture was lessened with the breakthrough
technology of formulating fungal spores in oil (Bateman & Alves 2000). The conidia remain
active inside the oil droplets and research now concludes that oil enhances the insect infectivity
at low humidity levels (Inglis et al. 2001).
Rainfall
Precipitation can play a vital role in the dispersal of conidia from substrates, but there is strong
evidence to suggest that significant amounts of fungal conidia may be „washed off‟ the insect
cuticle during rainfall despite the lipophilic mechanisms of conidial adhesion (Inglis et al. 2000).
Despite these evidences, confounding factors like canopy density and architecture as well as
insect behaviour on the persistence of fungal conidia in these environments, remains to be
resolved (Inglis et al. 2001). Research on the use of stickers in the persistence of fungal conidia
has been limited but their use is cautioned because they may retard the transfer of conidia from
plant leaves to insects in a natural environment (Inglis et al. 2000).
Soil factors
Many EP fungi are soil-borne microorganisms and this environment represents a complex set of
biotic and abiotic factors which can affect the efficacy and persistence of fungi. These abiotic
factors include: soil type (texture, cation exchange capacities, organic matter content, pH etc),
moisture and the occurrence of other soil microflora (Ali-Shtayeh et al. 2002; Ekesi et al. 2003;
Jaronski 2007; Quesada-Moraga et al. 2007). Soil moisture has been shown to adversely affect
the persistence of B. bassiana, M. anisopliae and L. lecanii because vertical movement of
conidia in the soil is affected and loss of inoculum occurs (Inglis et al. 2001). Soil texture and
organic matter appear to be the most important factors governing the vertical movement of fungi
with sandy-textured soils low in organic matter retaining fewer propagules than clay-textured
and organic soils (Jaronski 2007). Mietkiewski et al. (1997) observed that the isolation of B.
bassiana from arable soils in the UK was usually associated with natural undisturbed habitats
32
which were higher in organic matter and Quesada-Moraga et al. (2007) suggested that soils with
greater organic matter content have higher cation exchange capacities with greater organic matter
enhancing conidia absorption. Quesada-Moraga et al. (2007) found that M. anisopliae
predominated in soils with a pH lower than 7 and was better adapted to slightly more acidic soils
than B. bassiana. This said however, the effects of pH seem to be minimal (Inglis et al. 2001).
Management practices such as tillage may also affect the presence of EP fungi substantially,
because during these processes conidia may be moved to the surface of the soil, where they
become exposed to UV-B radiation and higher temperatures (Jaronski 2007; Meyling &
Eilenberg 2007). The effects of biotic factors, such as other soil microorganisms, may have an
effect on the persistence of EP fungal species. Meyling & Eilenberg (2006b) found that organic
matter content and subsequent biological activity in the soil adversely affected the persistence of
this fungal species due to antagonistic effects. Mainly because both Beauveria and Metarhizium
are poor competitors for organic resources compared to opportunistic saprophytic fungi that are
ubiquitous in the soil (Meyling & Eilenberg 2006b). Understanding the ecology of EP fungi in
the environment, particularly the soil, is an important prerequisite to applying these fungi
correctly in the field. Meyling & Eilenberg (2007) provide a schematic diagram which looks at
the ecology of both Beauveria and Metarhizium in the soil and how these species are dispersed
and cycle in the environment. (Fig. 1.10).
1.3.8 Compatibility of fungi with agrochemicals
It is important to realize that EP fungi will not displace chemical pesticides in all commercial
crop production systems, often because livelihoods may be at stake and chemical control may be
needed to hurriedly suppress a rapidly expanding pest population or simply because EP fungi
may not target all insect pests (Inglis et al. 2001). Much of the research on EP fungi has focused
on the influence and compatibility of agrochemicals and the germination of EP fungi in vitro and
their inhibitory properties which vary within and among chemical classes. Agrochemical
insecticides, herbicides and fungicides are often applied to conventional agricultural systems and
can either have a significantly deleterious effect (Poprawski & Majchrowicz 1995) on the
occurrence of EP fungi or work synergistically with fungal biological control agents (Purwar &
Sachan 2006).
33
Figure 1.10 A schematic diagram summarizing the compartments in temperate terrestrial
ecosystems in which B. bassiana and M. anisopliae occur naturally. Compartments are
connected with lines to indicate dispersal and infection pathways. Plant associations are included
as potential compartments for the natural occurrence of B. bassiana and M. anisopliae (Meyling
& Eilenberg 2007).
EP fungi can be adversely affected by chemical pesticides in several ways. In addition to direct
toxic effects on the fungi themselves, chemicals may have an indirect effect of killing the
potential fungal hosts (Jaros-Su et al. 1999). In general, it appears that germinating EP fungal
spores are more sensitive to pesticides than the fungal mycelia (Sterk et al. 2002). Agrochemical
pesticides can inhibit vegetative growth, conidiogenesis, fungal sporulation and cause genetic
mutations which can result in low conidial viability and virulence towards insect pests (Mochi et
al. 2006).
Imidacloprid, the active ingredient of Confidor® (Bayer CropScience, Germany), is a
neonicotinoid (neuro-active insecticide modeled after nicotine) which is often used in citrus to
control whiteflies and scale insects. James & Elzen (2001) investigated the dual use of this
insecticide with B. bassiana and found antagonism in that B. bassiana inhibited the effectiveness
34
of imidacloprid. The authors hypothesize that B. bassiana caused a behavioural response that
reduced insect feeding and uptake of imidacloprid. Beauveria bassiana spore germination and
colony formation were not inhibited by this chemical but B. bassiana did not adsorb or degrade
imidacloprid in a tank mix either. However, McCoy et al. (2007) found that B. bassiana, the
active ingredient of the commercial product, Mycotrol® (BioWorks, New York), formulated in
oil or a combination of fungus and a sub-lethal dose of imidacloprid were efficacious against the
neonate larvae and adults of root feeding weevils, Diaprepes species, when applied as a soil
barrier in citrus.
Insecticide treatments against citrus rust mite, Phyllocoptruta oleivora (Ashmead), on navel
oranges have also been shown to significantly reduce the incidence of Hirsutella thompsonii,
while methidathion and chlorobenzilate were previously shown to be moderately toxic to other
EP fungi (Puterka 1999). The fungicide, azoxystrobin, is commonly used to control citrus black
spot in orchards; it possesses the broadest spectrum of activity of all presently known antifungals
(Anonymous 2001). Sterk et al. (2002) looked at the toxicity of this compound against four
beneficial fungal species namely: I. fumosoroseus, L. lecanii, B. bassiana and T. harzianum and
found azoxystrobin to be less toxic than expected. Azoxystrobin also seemed to be compatible
with these beneficial fungi when there were gaps of a few days between applications of EP fungi
and fungicides (Sterk et al. 2002). Majchrowicz & Poprawski (1993) examined the in vitro
effects of fungicides on EP fungi and concluded that the susceptibility of fungi to fungicides
varied widely depending on the chemicals and species of fungus. They found that
dithiocarbamate derivatives zineb and copper oxychloride together or mancozeb alone
completely inhibited germination of the fungi tested, including B. bassiana. The inhibitory
effects were directly proportional to the dose of the active ingredient. Mochi et al. (2006)
investigated the effects of various pesticides in the soil which could affect the pathogenicity of
the EP fungus M. anisopliae which was used against Ceratitis capitata in orchard systems. Two
fungicides based on the chemicals: chlorothalonyl and tebuconazol, an acaricide based on
abamectin, an insecticide based on trichlorfon and a herbicide based on ametrin were
incorporated into soil at the manufacturers recommended doses. The pesticides had negligible
influences on the action of M. anisopliae but the fungicides effected fungal conidia during the
pupal stage and decreased the pathogenicity of the fungus against C. capitata (Mochi et al.
2006). In South African citrus orchards, the use of the insecticide chlorpyrifos is often employed
to control African bollworm (Helicoverpa armigera), however the fungicidal effects of this
insecticide have been noted by de Oliveira et al. (2003), who found that B. bassiana germination
35
was 100% inhibited by the insecticide when compatibility tests were carried out in Brazilian
coffee plantations.
1.3.9 Compatibility with natural enemies and other biological control agents
The contribution that predators, parasitoids and indigenous pathogens provide in cropping
systems is crucial for the suppression of many insect pest populations and is an important aspect
to consider in IPM strategies. Reports on the natural prevalence of EP fungi and particularly the
hyphomycetes have shown that these fungi occur widely in soils and on insects in the aerial
environment illustrating the long evolutionary history of coexistence which EP fungi share with
other microorganisms (Zimmermann 2007b). Some hyphomycetes have antagonistic properties
against other microorganisms and similarly other microorganisms may have a suppressive effect
on EP fungi which may affect their efficacy in the field (Inglis et al. 2001). The main concern in
terms of inundatively applying mycoinsecticides to systems is their potential to displace natural,
non-target populations of microorganisms and similarly these biological control agents have to
persist in the soil themselves and should not be affected by other microbes. Metarhizium
anisopliae and B. bassiana have been shown to suppress phytopathogenic fungi such as
Ophistoma ulmi, the causal agent of Dutch Elm disease (Zimmermann 2007a). Further B.
bassiana has been shown to have antagonistic effects and reduces the disease incidence of
Phoma betae, a common soil fungus which attacks weak plants and is the causal agent of
blackleg of beet. Beauveria bassiana was also able to reduce Rhizotonia solani (damping-off of
tomato) in greenhouses and protect cotton against a seedling disease complex (Zimmermann
2007b). The compatibility of M. anisopliae and other EP fungi with the mycoparasites,
Gliocladium (now Clonostachys), Trichoderma harzianum and Lecanicillium lecanii, have
recently been investigated and it was shown that M. anisopliae and B. bassiana were highly
susceptible (suppressed growth of these EP fungi) to all mycoparasites tested, but that coapplication of mycoparasites and EP fungi did not affect their biological control efficacy in vivo
(Zimmermann 2007a). The inundative application of strains of B. bassiana against the masson
pine moth, Dendrolimus punctatus, in China showed that during the course of a year, indigenous
and exotic strains were reisolated, and it was found that indigenous strains continued to persist
and were not displaced by the exotic B. bassiana strains (Zimmermann 2007b). In studying the
vegetative compatibility of indigenous and mass-released strains of B. bassiana in North
America, it was suggested that the likelihood of genetic exchange between these strain types was
36
very low due to a large number of vegetative compatibility groups which had self/non-self
recognition systems (Castrillo et al. 2004).
It is surprising that very little attention has been given to the co-application of EP fungi and other
entomopathogens such as viruses, nematodes and bacteria, as the perceived enhancement of
biological control efficacy seems promising, particularly because these microbial organisms have
different modes of action, which would reduce inter-species antagonisms. In a study by Rao et
al. (2006) the co-application of two EP fungi, B. bassiana and N. rileyi, were assessed against
the second instar larvae of Spodoptera litura and no synergistic effects on insect mortality were
found. In other studies synergistic effects were observed when M. anisopliae and Serratia
entomophila (an Enterobacteriaceae agent which causes amber disease in grass grubs) were coapplied against the early instar grass grub, Costelytra zealandica (Inglis et al. 2001). Inglis et al.
(1997) demonstrated the effective use of „fungal cocktails‟ when they tested the hypothesis that
co-applied B. bassiana and M. flavoviride could be used to increase the temperature ranges over
which the fungi on their own would elicit an infection. In the hot temperature environment, less
mortality was observed for B. bassiana (3%) than for M. flavoviride (52%). Conversely, in the
cool temperature environment, less mortality was observed for M. flavoviride (46%) than for B.
bassiana (100%). Co-application induced greater mortality than if M. flavoviride was applied
alone in a simulated hot-temperature environment or B. bassiana alone in a simulated cooltemperature.
The safety of EP fungi towards non-target arthropods, earthworms and pollinators is also a very
important consideration for the use of mycoinsecticides because of the broad host range of some
members, such as B. bassiana, which is capable of infecting over 700 different insect species
across 14 insect and amphipod orders (Rehner 2005; Zimmermann 2007b). At this stage it is
important to note that there are differences in the physiological and ecological host ranges where
pathogens are concerned. The physiological host range demonstrates which organisms can be
infected in the laboratory, which is often much broader because insects may become stressed and
larger concentrations of conidia are used (Zimmermann 2007a). While the ecological host range
demonstrates the organisms which may be infected in nature, which is often narrower because
pathogens may not occupy the same ecological niches as some organisms or infect all organisms
in the environment (Zimmermann 2007a). The capability of B. bassiana to infect non-target
organisms such as collembolans, carabid beetle predators and coccinellids has been demonstrated
under stress conditions however in these studies, earthworms and honeybees were not affected
37
(Zimmermann 2007b). Metarhizium anisopliae has been shown to infect a wide range of insects
in laboratory experiments and some non-target effects have been observed in the field, however
lower non-target effects were seen in the field (Zimmermann 2007a). This EP fungus showed
pathogenicity towards two natural enemies in laboratory experiments: the green lace wing,
Chrysoperla carnea and a mirid predator, Dicyphus tamaninii. The mirid predator showed
greater susceptibility than C. carnea with corrected mortalities of 10 and 4%, respectively
(Thungrabeab & Tongma 2007). The literature concludes that EP fungi have minimal impact on
non-target organisms, especially when isolate selection and spacio-temporal factors are
considered, and are generally safe and compatible with a number of beneficial insects (Inglis et
al. 2001; Hajek & Goettel 2007; Thungrabeab & Tongma 2007; Zimmermann 2007a, b).
1.3.10 Safety of EP fungi towards vertebrates
The registration requirements of a new mycoinsecticide product are such that detailed vertebrate
safety tests have to be included; all safety tests using EP fungi such as Beauveria, Metarhizium
and Paecilomyces carried out against fish, amphibians, reptiles, birds and small mammals
revealed that no serious detrimental effects were observed in these groups (Zimmermann 2007a,
b). Allergies in humans are rare but have been recorded and may be as a result of extended
exposure to EP fungi, particularly in production facilities (Zimmermann 2007b). Some orders in
the class Zygomycete, which includes the EP fungi in the order Entomophthorales, are
pathogenic to humans. Pathogens such as Conidiobolus coronatus cause various forms of
zygomycosis, which causes serious nasal infection and is considered a possible cause of rhino
facial swelling in immigrants from equatorial Africa, Asia, and America (Fischer 2008).
1.4 AIMS OF THE STUDY
Based on the literature there appears to be great interest in the development of mycopathogens
for IPM in cropping systems because the natural incidence of these microorganisms in
agricultural environments has been shown to significantly suppress pest populations. By
inundatively applying conidia of EP fungi to the soil in agricultural systems for the control of
pests with subterranean life stages, citrus growers and other IPM cropping systems are offered
another tool for insect control which is environmentally friendly and safe. A greater
understanding of the ecology of EP fungi in nature and in cropping systems will enhance our
38
knowledge of environmental persistence and aid in strain selection which would lead to the
enhancement of biological control efficacy. Thus, the aims of this study were (1) to recover and
identify indigenous isolates of EP fungi from soil samples taken from organically and
conventionally managed citrus orchards and (2) to compare the recovery of fungal isolates and
species from the bait insects, Galleria mellonella (Lepidoptera: Pyralidae), T. leucotreta and C.
capitata. This study also served to (3) compare both the distribution and abundance of EP fungi
in conventionally versus organically farmed citrus soils, and (4) to compare cultivated orchards
and orchard margin (refugia) areas in the Sunday‟s River Valley citrus growing region in the
Eastern Cape Province in South Africa.
Further aims were (5) to compare the pathogenicity of 21 obtained indigenous isolates of EP
fungi against final instar larvae of T. leucotreta, C. capitata and C. rosa using sand-conidial
suspension incorporation bioassays. Screening trials were undertaken to determine the most
virulent fungal isolates, which had broader host ranges and were capable of infecting all three
insect test species. Further aims were (6) to test the four most virulent isolates with a
commercially produced Beauveria product, Bb Plus® (Biological Control Products, South
Africa), against the three citrus pests using various concentrations of fungal conidia incorporated
into sand to ascertain the LC50 and LC90 values of each isolate and (7) to test the three most
virulent fungal isolates which were highly pathogenic to T. leucotreta in concentration-response
bioassays against the final instar larvae of T. leucotreta using four different exposure times, 12,
48, 72 and 96 hrs and two conidial concentrations.
39
II
SURVEYING FOR INDIGENOUS ISOLATES OF
ENTOMOPATHOGENIC FUNGI
2.1 INTRODUCTION
Soil-inhabiting entomopathogenic (EP) fungi are among the many natural enemies of arthropod
pests in natural and agricultural systems; numerous fungal genera have cosmopolitan
distributions and have been recognised as playing a key role in pest regulation and suppression
(Ali-Shtayeh et al. 2002; Meyling & Eilenberg 2007). Studies on the natural occurrence and
distribution of EP fungi have been conducted in Canada (Bidochka et al. 1998), China (Bing-Da
& Xing-Zhong 2008), Denmark (Meyling & Eilenberg 2006b), Mauritius (Sookar et al. 2008),
Norway (Klingen et al. 2002), Pacific Northwest of the USA (Bruck 2004), Palestine (AliShtayeh et al. 2002), Spain (Asensio et al. 2003; Quesada-Moraga et al. 2007), Switzerland
(Keller et al. 2003; Rodrigues et al. 2005) and the United Kingdom (Chandler et al. 1997).
An enhanced understanding of the ecology of indigenous populations of EP fungi within
cropping systems and their surrounds is an important prerequisite for the evaluation of their
contributions to pest control (Meyling & Eilenberg 2007). It is also important to understand the
impact of agricultural practices and management strategies on naturally occurring EP fungal
populations. Many economically important ascomycete fungi, such as Beauveria bassiana and
Metarhizium anisopliae are facultative organisms capable of both saprophytic growth and
survival in the soil environment, and show pathogenicity towards insect hosts which they
colonise (Bidochka et al. 2001; Meyling & Eilenberg 2007). Understanding the nature of these
interactions could facilitate more effective exploitation of fungi for biological pest control
strategies (Vega et al. 2009).
Note:
This chapter has been published in BioControl on line: DOI 10.1007/s10526-009-9259-0
40
Studies on M. anisopliae revealed that this fungal species had rhizosphere competence (the
ability of a micro-organism to colonize and become established in the narrow zone of soil which
is subject to the influence of living plant roots) (St Leger 2008). Beauveria bassiana has been
included in the spectrum of fungi with endophytic activity; studies suggest that the presence of
this fungus in vascular plant tissues is an adaptive protection against herbivorous insects and its
natural occurrence has been found within corn, cocoa, poppy, coffee and tomato (Meyling &
Eilenberg 2007). This facultative ability has enormous potential for biological control; if insect
hosts become scarce in a particular agricultural habitat, these EP fungi can adapt and persist in
the environment until hosts become available again. Furthermore, EP fungi play additional roles
in nature including being antagonists of plant pathogens (Goettel et al. 2008) and plant growth
promoters (Vega et al. 2009).
There is a present paradigm that assumes that the genotype of an EP fungus coevolves with a
particular insect host taxon which it infects (Bidochka et al. 2002). Both Beauveria bassiana and
Metarhizium anisopliae are capable of infecting hundreds of different insect species (Bidochka
& Small 2005; Rehner 2005) and both, are considered facultative pathogens (Meyling &
Eilenberg 2007). A recent study in Canada analysed various genetic populations of B. bassiana
taken from forested, agricultural and Canadian Arctic habitats. Three distinct genetic groups of
B. bassiana were observed which were distinctly correlated with these habitats. Further, within
each group, recombining population structures as well as clonally reproducing lineages were
noted (Bidochka et al. 2002). Beauveria bassiana genotypes isolated from forested and Arctic
habitats exhibited cold tolerances and were able to grow at low temperatures (8°C), while those
genotypes isolated from agricultural habitats were able to grow at 37°C and exhibited UV
tolerances (Bidochka et al. 2002). Interestingly, there were no clear associations between genetic
groups of B. bassiana and the ability to infect different lepidopteran and coleopteran insect hosts
(Bidochka et al. 2002). This was further observed in a population genetic study with the EP
fungal species, M. anisopliae sampled from forested (showed cold-active growth) and
agricultural habitats (grew at higher temperatures and had UV tolerance) (Bidochka et al. 2001).
Two distinct, reproductively isolated, genetic groups were observed which were strongly
associated with different habitats yet no associations with insect hosts were found (Bidochka et
al. 2001). These genetic studies suggest that M. anisopliae may be an assemblage of cryptic
species (reproductively isolated species with similar/identical morphology) which may each be
adapted to various environmental conditions. The elucidation of fungal genetics will continue to
41
increase our understanding of the taxonomy of these microbes as well as the important
ecological functions they perform.
One method of determining EP fungal population densities and occurrences includes isolating
fungi on selective media and quantifying the number of colony forming units (CFU) per gram of
soil by diluting the soil sample (Keller et al. 2003). Most EP fungi and a range of bacteria can
grow on artificial media in vitro, and specific media have been developed to select for certain
groups of microorganisms (Veen & Ferron 1966). The development of selective media for the
isolation of EP fungi includes the inhibition of bacteria by the application of broad-spectrum
antibiotics such as chloramphenicol, tetracycline or streptomycin. Hypocrealean fungi grow
relatively slowly in comparison to the ubiquitous opportunistic saprophytic fungi found in the
soil environment. Thus, the contents of the media need to include substances that prevent these
fungi from out-competing and overgrowing the species of interest (Meyling 2007). The
disadvantages of using soil suspensions on selective media include overgrowth of opportunistic
soil fungi on media plates, and often small soil samples are used (1 g), therefore the risk of not
sampling the fungus is increased because EP fungi have a clumped distribution in the soil
(Meyling 2007). Furthermore, the use of the soil dilution plating method is convoluted because
of the preparation of soil subsamples, the determination of water content, dilution and plating of
suspensions, as well as the initial production of media (Jaronski 2007). Often lower numbers of
samples are included which limits the sample area and the true reflection of occurrence of fungi.
Presently most studies of EP fungal biodiversity in natural and agricultural systems are
conducted with the use of standard bait insects, such as Galleria mellonella (Lepidoptera) or
Tenebrio molitor (Coleoptera) (Klingen et al. 2002; Meyling et al. 2009). The method was
originally developed by Zimmermann (1986) to isolate EP nematodes from the soil, but fungi
were also isolated. Insect baiting was suggested to be a more sensitive method than traditional
soil suspensions plated on selective media and could be employed as a standard technique for
obtaining EP fungi from soil environments (Meyling 2007). Standard bait insects are easily
obtained and cultured on artificial diet, and are also susceptible to fungal infection (Chandler et
al. 1997). Often a large number of fungal isolates are obtained using these bait insects, but fungal
species diversity is usually low (Klingen et al. 2002), and it has been suggested that lepidopteran
or colepteran-specific isolates may be obtained whilst employing the standard bait insects
(Chandler et al. 1997). Some perceived disadvantages to employing this technique are: a) that
some insect bait species may select for specific fungal pathogens and thus may not entirely be
42
representative of all the potential fungal species present in the soil environment; b) baiting may
be regarded as a semi-quantitative measure at best (Jaronski 2007); c) moist soils may enhance
the infection of nematodes which can also utilise the bait insects; and d) it may be difficult to
quantify inoculum levels (Meyling 2007).
If soils baited with G. mellonella are incubated at specific temperatures, it may enhance or
decrease selection of specific fungal species present in the sample pot, for example Jaronski
(2007) showed that when forest soil samples were incubated at 18°C, B. bassiana, Isaria
farinosus and I. fumosoroseus were observed. However, when baiting was done at 25°C, M.
anisopliae predominated while I. farinosus was greatly reduced. Baiting soil samples with larvae
of G. mellonella is a widely applied screening tool for the determination of the distribution and
occurrence of indigenous species of EP fungi (Ali-Shtayeh et al. 2002; Klingen et al. 2002;
Asensio et al. 2003; Keller et al. 2003; Rodrigues et al. 2005; Sookar et al. 2008; Meyling et al.
2009). It is unlikely however that any single protocol will be an adequate method of estimating
the total diversity of EP fungi in soil environment because isolation techniques are selective in
one way or another (Klingen et al. 2002). Therefore Klingen et al. (2002) suggested that if a
more representative range of EP fungi is to be sampled, it may be necessary to use a selection of
insect species or bait methods with complimentary susceptibilities to fungal pathogens.
Biodiversity is usually evaluated by the assessment of species diversity, however the assessment
of EP fungal populations and communities has been difficult because fungal taxonomy has been
limited and fungal taxa usually comprise cryptic species (Meyling & Eilenberg 2007). Meyling
et al. (2009) inferred species diversity and reproductive potential using molecular techniques on
Beauveria within a single arable field (~ 28 ha in size) and bordering hedgerow in Denmark.
Seven phylogenetic species were discovered including five phylogenetic species within B.
bassiana senso stricto. Mating-type polymerase chain reaction (PCR) assays showed that all five
B. bassiana s.s. species possessed bipolar out-crossing mating systems. These analyses
demonstrated the genetic diversity of B. bassiana within a relatively limited geographic area
(Meyling et al. 2009) and the greatest genetic diversity was observed in the semi-natural habitat
of the hedgerow (~ 0.35 ha in size) compared to the arable field. The simultaneous occurrence of
phylogenetically similar members of B. bassiana from populations below and above ground
suggests that EP fungi may cycle between the soil environment and the phylloplanes of
vegetation and insect hosts in natural habitats (Meyling & Eilenberg 2007; Meyling et al. 2009).
These authors concluded that natural habitats offered structural diversity to groups of organisms
43
(fungi and their insect hosts) and were important reservoirs for fungal species diversity compared
to the cultivated fields. Evidence suggests that maintaining landscape heterogeneity in
agricultural habitats is crucial for the maintenance of fungal species diversity which allows
fungal environmental adaptation, which in turn contributes to pest management in sustainable
agriculture (Meyling & Eilenberg 2007).
The effects of agricultural practices can have an important impact on the distribution and
occurrence of EP fungi. Annually cropped vegetable systems for example are highly disturbed
environments due to tillage regimes and the fungal communities in arable fields are different to
those communities of less disturbed habitats (Meyling & Eilenberg 2007). Meyling & Eilenberg
(2007) showed that systems with no-till or strip-till versus those that were ploughed or disked
had higher B. bassiana and M. anisopliae densities. Tillage practices turn the soil over and
expose EP fungal conidia to UV-B and temperature increases, both of which inactivate fungal
germination (Braga et al. 2001). Temperature, moisture and UV radiation appear to be the most
important factors concerning EP fungal survival in the soil (Meyling & Eilenberg 2007) (cf
Chapter I). In citrus orchards these environmental factors are considerably reduced depending on
tree age and canopy size which provide shade and reduces temperature and UV-B radiation
compared to annual vegetable cropping systems (McCoy et al. 2007). Furthermore, citrus
orchards are perennial habitats and may support larger, more stable microbial communities than
annual vegetable cropping systems. The use of irrigation systems and the need to frequently
water citrus, offers continuous soil moisture for soil inhabiting EP fungi (Ali-Shtayeh et al.
2002). Reducing areas of bare ground through the use of mulches may reduce the sizes of pest
populations because these areas then become favourable for ground-dwelling predators (Meyling
& Eilenberg 2007). Mulching however may be unfavourable for EP fungi because increased
amounts of organic matter in the soil have been shown to increase the antagonistic effects of
opportunistic saprophytic fungi and other microbes against EP fungi, which have a slower growrate and are poorer competitors for organic resources (Meyling & Eilenberg 2007).
The development of virulent EP fungal biopesticides against insect pests has been the sole focus
of much of the literature, which has reduced our ability to improve biological control efficacy
and persistence in the field (Wang et al. 2005), but these microbial agents have the potential to
play multiple roles in crop protection against plant pathogens and insects (Vega et al. 2009).
These findings indicate that the ecological role of EP fungi in the environment is not yet fully
understood and limits our ability to employ them successfully for pest management. Thus, the
44
continuation of exploratory collection studies allows information about the natural history of
these microbes to be explored, particularly in a context outside of their insect hosts. Ecological
data can aid in isolate selection and the development of application strategies for the
implementation of EP fungi (Castrillo et al. 2005). It further ensures that microbial control
efforts are not limited to currently available mycoinsecticides and their inundative releases.
Meyling & Eilenberg (2007) discuss the concept of conservation biological control (CBC): „the
manipulation of farming management practices to enhance the living conditions of specific
natural enemies of a pest, with the primary focus being suppression of that pest species.‟ This
concept serves to conserve any natural enemies/microbial populations already present, rather
than inundating an area with maladapted, non-indigenous strains or species, reiterating the need
for ecological collection studies of EP fungi.
Habitat influences the distribution and abundance of insect species and it also drives the
distribution of EP fungi; any indication of insect/host related population structure within EP
fungi should be seen principally as the result of the habitat in which the insects and fungi cooccur (Bidochka et al. 2001) and not as a result of the occurrence of insect hosts alone.
Increasing evidence suggests that habitat selection, and not insect host selection, drives the
population structure of EP fungi and that adaption to a specific habitat type is a more important
selection criterion for identifying potential biological control candidates (Bidochka et al. 2002).
While it may be important to obtain highly infective and virulent strains of EP fungi to control
pests, it is even more important to look for isolates which are also capable of surviving and
persisting in the environments and habitats for which they are required. For the abovementioned
reasons it was thought pertinent to investigate the distribution and occurrence of indigenous
isolates of EP fungi from citrus growing regions and their surrounds, for the subsequent control
of citrus pests.
Thus, the aims of this chapter were (1) to recover and identify indigenous isolates of EP fungi
from soil samples, so as to assess the diversity and occurrence of fungi from citrus orchards and
natural habitats; and (2) to compare the recovery of fungal isolates and species from the bait
insects, Galleria mellonella (Lepidoptera: Pyralidae), Thaumatotibia leucotreta (Lepidoptera:
Tortricidae) and Ceratitis capitata (Diptera: Tephritidae). The purpose being that by baiting with
these key citrus pests, one would obtain fungal isolates which may be more virulent towards T.
leucotreta and C. capitata. This study also served to (3) compare both the distribution and
abundance of EP fungi in conventionally versus organically farmed citrus soils; and (4) compare
45
cultivated orchards versus natural (refugia) areas in the Sunday‟s River Valley citrus growing
region and surrounding areas in the Eastern Cape Province in South Africa.
2.2 METHODS & MATERIALS
2.2.1
Insect Cultures
Thaumatotibia leucotreta larvae were obtained as final instar larvae from a continuous
laboratory culture held at River Bioscience, Addo, South Africa. Ceratitis capitata larvae were
obtained from a continuous laboratory culture held at Citrus Research International (CRI),
Nelspruit, South Africa, and maintained on artificial diet until final instar larvae were needed.
Larvae of the greater wax moth, Galleria mellonella, are often used as a standard bait insect for
the isolation of EP fungi, because the larvae are susceptible to infection by these fungi (Chandler
et al. 1997; Meyling 2007). Galleria mellonella larvae were obtained from absconded bee hives
held at Grahamstown (33°23′54″S; 26°25′41″E), cultured on honeybee wax (Meyling 2007) and
maintained in constant darkness at 24°C.
2.2.2 Soil sampling
A total of 288 soil samples were collected from three conventional citrus farms and three organic
citrus farms (as defined by the EU Council Regulation 2092/91) in the Eastern Cape Province,
South Africa (Fig. 2.2.1). The locations of the sampled soils were recorded using global
positioning system (GPS) equipment (Garmin: E-Trex). Conventional citrus sites included,
Arundel Farm near Addo (33°30′57″S; 25°39′11″E), Mosslands Farm near Grahamstown
(33°23′54″S; 26°25′41″E) and J & B citrus near Cookhouse (32°45′56″S; 25°45′46″E). The
organic sites included: Rosedale Farm near Addo (33°32′21″S; 25°41′39″E); Hippo Pools Farm
near Kirkwood (33°24′42″S; 25°24′34″E) and Olifantskop near Addo (33°37′14″S; 25°40′49″E)
(Fig. 2.2.1).
46
Figure 2.2.1 Map of the Sunday‟s River Valley citrus growing region and surrounding areas in
the Eastern Cape Province, South Africa, with the locations of the six farms, marked with a
bullet (•), where soils were sampled in this study.
At each of the six farms, three orchards were selected. A total of 12 soil samples were collected
per orchard, along two intersecting transects, until the end of the orchard was reached (Fig.
2.2.2). Additionally, indigenous vegetation within 1-2 km of orchard sites was randomly
sampled at 12 points, which were at least 18 m apart; hereafter referred to as refugia (Fig. 2.2.2).
Soil samples were collected with the use of a cylindrical soil auger (7 cm x 14 cm) volume 538
cm3 to a depth of 15 cm. Surface litter was removed and samples were placed in clear plastic
bags (31 × 20 cm) and labelled. Orchard soil samples were collected from under the tree
canopies at the beginning of autumn (March) 2008, stored at 4°C and individually baited within
two months of the sampling period. All orchards had trees of similar sizes and the height of the
tree canopies was consistent. Agricultural soil type data and other agronomic information were
obtained from farm owners (Table 2.2.1). The classification of South African soils is based on
47
the recognition of diagnostic topsoils and soil horizons, and soil forms are defined in terms of
these sequences (Agricultural Research Council 2009).
Figure 2.2.2 Sampling points in each orchard and the adjacent field margin (refugia).
2.2.3 Baiting procedures
In the laboratory, each soil sample was thoroughly mixed and then sieved through a metal sieve
with a mesh size of 4 mm. Soil samples were divided into three 200 ml portions and transferred
to 400 ml transparent plastic pots which were then sealed with perforated lids. If soil samples
were too dry they were moistened with distilled water to maintain humidity during baiting. EP
fungi were isolated from soil samples using variations of the „Galleria bait method‟
(Zimmermann 1986; Chandler et al. 1997; Ali-Shtayeh et al. 2002; Klingen et al. 2002; Keller et
al. 2003; Meyling & Eilenberg 2006b). Three separate experiments were run to test T.
leucotreta, C. capitata and G. mellonella larvae individually as bait insects. The Galleria bait
method was conducted using 10 final instar larvae which were placed on the surface of each soil
sample container. Galleria mellonella larvae were heat-treated according to Meyling (2007) to
48
prevent excessive webbing in the soil. The same protocol was undertaken using 10 T. leucotreta
and 10 C. capitata larvae per soil sample; however there was no heat-treatment applied to these
species because excessive webbing in the soil was not experienced with these insects. All soil
samples were incubated at 22°C in the dark. Containers were inverted daily for the first week to
make bait insects penetrate the soil as much as possible. After initial baiting, samples were
checked for the presence of dead larvae every 3-4 days for three weeks. All dead larvae were
surface sterilised with 70% ethanol prior to incubation in a moisture chamber (Petri dish with
moistened filter paper) to prevent opportunistic external saprophytic fungi from growing on the
cadavers. Sporulating larvae and/or pupae were placed on appropriate media to isolate fungal
cultures.
2.2.4 Isolation and identification of fungi
Fungal isolations were made using selective media adapted from Meyling (2007) with the
following composition and preparation: 60 g SDA (Sabouraud Dextrose Agar, Merck)
supplemented with 1 ml Dodine, 50 mg/L Chloramphenicol, 50 mg/L Ampicillin or 50 mg/L
Rifampicin. Plates were incubated at 22°C in the dark. All potential EP fungi were identified
microscopically using tape mounts, according to morphological characteristics described in
taxonomic keys and other relevant literature (Barnett 1960; Domsch et al. 2007). Selected strains
were sent to the Mycology Unit at the Plant Protection Research Institute (PPRI) in Pretoria,
South Africa, for morphological identification. Once identified each isolate was allocated an
accession number and stored in the South African National Collection of Fungi held at PPRI.
The PPRI identification and accession numbers for these cultures were in turn used as a
reference catalogue. All saprophytic fungi arising from the baiting procedures were discarded.
49
Table 2.2.1 Agronomic data of orchards of the six farms in the Eastern Cape from which soils
were sampled and used in insect baiting.
Farm Name Orchard No. Soil Type
Arundel
Mosslands
J&B Citrus
Rosedale
Hippo Pools
Olifantskop
Citrus Cultivar
Size (Ha)
Rootstock
Fungicides and Other Additives
1
Loam
Navel Oranges
1.00
Troyer citrange
Abamectin, Tartox, Cryptogran
17
Loam
Navel Oranges
0.65
Carizzo citrange
Abamectin, Tartox, Cryptogran
23
Loam
Navel Oranges
0.50
Swingle citrumelo
Imidacloprid, Cryptogran
10
Oakleaf
Navel Oranges
0.82
Rough lemon
Abamectin
11
Oakleaf
Navel Oranges
0.25
Rough lemon
Abamectin
14
Glenrosa
Navel Oranges
0.98
Rough lemon
Copper oxychloride, Abamectin,
Chlorphyripos, Mercaptothion
14
Oakleaf
Cara caras
2.17
Carizzo citrange
20
Oakleaf
Cara caras
1.90
Carizzo citrange
22
Sterkspruit
Late Navels
1.63
Carizzo citrange
3
Oakleaf
Navel Oranges
0.90
Carizzo citrange
15
Oakleaf
Clementine‟s
0.90
Carizzo citrange
20
Oakleaf
Lane Late Navels
1.10
Carizzo citrange
14
Loam
Palmer Navels
0.40
Rough Lemon
21
Loam
Palmer Navels
0.45
Swingle citrumelo
24
Loam
Navel Oranges
0.50
Carizzo citrange
13
Loam
Palmer Navels
0.80
Rough lemon
15
Loam
Palmer Navels
0.80
Rough lemon
19
Loam
Palmer Navels
0.50
Rough lemon
Imidacloprid, Methidathion, Mancozeb,
Abamectin, Azoxystrobin
Imidacloprid, Methidathion, Mancozeb,
Abamectin, Azoxystrobin
Imidacloprid, Methidathion, Mancozeb,
Abamectin, Azoxystrobin
Copper spray, Neem & Pyrethrum,
Cryptogran, Broadband, compost extract
Neem & pyrethrum, Cryptogran, Copper
spray, Broadband, compost extract
Neem & Pyrethrum, Cryptogran, compost
extract
Neem & Pyrethrum, copper spary, fish
emulsion, mineral oil, compost tea
Neem & Pyrethrum, copper spary, fish
emulsion, mineral oil, compost tea
Neem & Pyrethrum, copper spary, fish
emulsion, mineral oil, compost tea
Neem & Pyrethrum, copper spray, fish
emulsion, mineral oil, compost tea
Neem & Pyrethrum, copper spray, fish
emulsion, mineral oil, compost tea
Neem & Pyrethrum, copper spray, fish
emulsion, mineral oil, compost tea
2.2.5 Quick screening method to detect entomopathogenic fungi
To separate EP fungi from opportunistic saprophytic fungal species, a rapid screening method
was adapted from Ali-Shtayeh et al. (2002). Isolated fungal cultures were first grown on SDA
media for 10-12 days until sporulation was reached. Fungal suspensions were prepared by
incising small pieces of hyphae and conidia from sporulating cultures and placing them into
sterilized 1.5 μl microcentrifuge tubes, containing 1ml of sterilized distilled water and 0.05%
Triton X, then vortex mixed. Five final instar G. mellonella larvae were dipped into each fungal
suspension for two seconds. The larvae were then placed into moisture chambers and incubated
in the dark at 22°C. Petri-dishes were checked daily for the presence of dead larvae. Dead larvae
were placed onto SDA agar until sporulation was again detected. Koch‟s postulates (Madigan et
50
al. 2005) were employed and Galleria larvae were re-infected with spores from these cultures
(Fig. 2.2.3). Upon insect mortality, larvae were again placed onto SDA agar until sporulation
was detected. Sporulating EP fungal cultures were labelled and sent to the Mycology Unit at
PPRI in Pretoria, for morphological identification. Cultures that were not sent to PPRI were
morphologically identified using the methods described above.
Figure 2.2.3 Galleria mellonella larvae infected with conidia from various EP fungal species
during the employed quick-screening method (all images were taken with a Leica EZ 4D
dissecting microscope at ~ X16) (Photo Credits: T. Goble, Rhodes University).
2.2.6 Data analysis
Only EP fungal species were considered for all statistical analyses. Chi-squared (χ2) tests were
used to compare the recovery of EP isolates from the three bait insects used and specifically the
recovery of Beauveria bassiana from these insects. Chi-squared tests were also used to compare
the occurrence of EP fungi in soils sampled from organically versus conventionally farmed soil,
and between soils sampled in cultivated orchards versus refugia. This test was also used to
compare the recovery of B. bassiana from organic versus conventionally managed soils.
Infections were registered qualitatively per sample pot. This meant that whether one or several
larvae infected with the same EP fungal species were observed in the same sample pot, this was
51
registered as one infection. All analyses are based on 288 soil samples from six localities
(farms); 108 from organically farmed soil; 108 from conventionally farmed soil and 72 from
refugia. In all analyses these ratios were adjusted accordingly.
2.3 RESULTS
2.3.1 Bait insects and fungal species
The fungal species and the number of fungal isolates obtained during this study differed
significantly according to the three insects used as bait (Fig. 2.3.1). When the standard bait
insect, G. mellonella, was used, a total of 45 EP fungal isolates were obtained, this was
significantly higher (χ2=40.13, df=2, P≤ 0.005) than those recovered from either C. capitata (11
isolates) or T. leucotreta (6 isolates) (Fig. 2.3.1). Beauveria bassiana was isolated significantly
more frequently (χ2=43.72, df=2, P≤ 0.005) than any other fungal species from all bait insects
(Fig. 2.3.1). More Metarhizium anisopliae var. anisopliae isolates were obtained using G.
mellonella than any other bait insect and the only isolate of Metarhizium flavoviride Gams and
Rozsypal was also obtained from this bait insect species. Ceratitis capitata was the only bait
insect to isolate the fungal species, Conidiobolus coronatus (Constantin) Batko and
Lecanicillium psalliotae Treschew. Lecanicillium psalliotae is recognised as a nematophagous
fungal species (Gan et al. 2007), but was included as an EP fungal species because of its
capability of infecting arthropods (Pirali-Kheirabadi et al. 2006) and because of its close
relatedness to Lecanicillium lecanii (a well-known EP fungus) (Zare & Gams 2008). The genus
Metarhizium was not isolated from C. capitata (Fig. 2.3.1).
2.3.2 Abundance and geographic distribution of entomopathogenic fungi
The distribution and frequency (% positive samples) of EP fungi in soil sampled from cultivated
orchards and refugia on six citrus farms in the Eastern Cape Province are shown in Table 2.3.1.
In total, 62 EP fungal isolates belonging to 4 genera were recovered from 288 soil samples baited
with each of the three insect bait species at 22°C, an occurrence frequency of 21.53% (Table
2.3.1). The most frequently isolated EP fungal species was B. bassiana, which was recovered
from 15.63% of all soil samples baited. This was followed by M. anisopliae var. anisopliae,
which was found to occur in fewer soil samples at a lower frequency of 3.82%. Lecanicillium
psalliotae was isolated with a frequency of 1.39%, while other species such as M. flavoviride
52
(0.35%) and C. coronatus (0.35%) were only isolated once from soil samples (Table 2.3.1). Only
1.38% (4 out of 288) of the soil samples collected yielded two EP fungal species in the same
sample.
Figure 2.3.1 The total number of EP fungal isolates and species recovered from the three
different bait insects used. Letters above a column denote significant differences using χ2 test,
P≤0.005.
53
Table 2.3.1 Distribution and occurrence frequency (% positive samples) of EP fungi in 288 soil samples, from orchards and refugia on three
organic and three conventionally managed citrus farms, in the Eastern Cape Province, South Africa.
Conventional farms (% F)
EPF species*
Mosslands
Arundel
Organic farms (% F)
J&B Citrus
Rosedale
Hippo Pools
Olifantskop
% F **
N= 36
N=12
N= 36
N=12
N= 36
N=12
N= 36
N=12
N= 36
N=12
N= 36
N=12
orchards
refugia
orchards
refugia
orchards
refugia
orchards
refugia
orchards
refugia
orchards
refugia
All species
27.7
66.5
8.2
8.3
5.5
24.9
22.2
33.3
5.5
16.6
33.1
58.2
21.5
B. bassiana
11.1
41.6
5.5
0.0
5.5
16.6
22.2
33.3
5.5
8.3
27.7
41.6
15.6
C. coronatus
0.0
0.0
0.0
0.0
0.0
8.3
0.0
0.0
0.0
0.0
0.0
0.0
0.3
M. anisopliae
16.6
8.3
2.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.7
16.6
3.8
M. flavoviride
0.0
8.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
L. psalliotae
0.0
8.3
0.0
8.3
0.0
0.0
0.0
0.0
0.0
8.3
2.7
0.0
1.4
*Percentage frequency of isolates found per farm
** Percentage frequency % F based on the total number of isolates found on all farms/288 soil samples
54
The greatest diversity of EP fungal species was recorded from the conventional citrus farm,
Mosslands. Eighteen isolates belonging to four fungal species (three genera) were obtained
(Table 2.3.1). The only isolate of M. flavoviride was obtained from one soil sample taken from
refugia at this farm. At Arundel, four isolates belonging to three species of EP fungi were
recovered. Five isolates belonging to two EP fungal species were retrieved from J & B citrus and
the only isolate of C. coronatus found in this study was retrieved from a soil sample taken from
refugia at this location. The farm, Rosedale, yielded the lowest diversity of EP fungal species.
Only B. bassiana was retrieved from soil samples at this farm. However, a large number of
isolates were obtained (12 isolates). At Hippo Pools, four fungal isolates belonging to two
species were recovered. At Olifantskop, the largest number of EP fungal isolates was recovered
(19 isolates) but the diversity of fungal species (three species) was lower than was observed at
Mosslands. The sixty two isolates of EP fungi which were collected from cultivated orchards and
refugia, from the six farms in the Eastern Cape are listed in Tables 2.3.2 and 2.3.3.
2.3.3 Farming systems and field margins
More EP fungal isolates were recovered from organic (35 isolates) than conventionally (25
isolates) farmed soil samples although this was not significant (Fig. 2.3.2). Beauveria bassiana
was isolated significantly more frequently in samples from organically farmed soils (30 isolates)
than from samples of conventionally farmed soils (15 isolates) (χ2=5.00, df=1, P≤ 0.05). In
contrast, M. anisopliae var. anisopliae was isolated more frequently in samples from
conventionally cultivated soils (8 isolates) than organically farmed soil samples (3 isolates) but
this result was not significant. Lecanicillium psalliotae occurred equally in both organically and
conventionally farmed soil samples. The single occurrences of M. flavoviride and C. coronatus
were not considered further. When comparing cultivated lands and refugia, significantly more EP
fungi were found in refugial habitats than cultivated land habitats (χ2=11.65, df=1, P ≤ 0.005)
(Fig. 2.3.2). However, when comparing cultivated lands and refugia separately, there were no
significant differences observed in the occurrence of EP fungi from the soil samples when both
farming systems were compared (Fig. 2.3.2).
55
Table 2.3.2 The location and ecological data of Beauvaria bassiana collected from citrus
orchards and refugia soils in the Eastern Cape Province, South Africa.
Farm
Arundel
Arundel
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
J&B Citrus
J&B Citrus
J&B Citrus
J&B Citrus
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Olifantskop
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Rosedale
Hippo Pools
Hippo Pools
Hippo Pools
Type
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
conventional
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
organic
NA = not apllicable,
Habitat
Area (Ha)
Soil Type
orchard
orchard
orchard
refugia
refugia
refugia
refugia
refugia
orchard
orchard
orchard
orchard
orchard
refugia
refugia
orchard
orchard
orchard
orchard
orchard
orchard
orchard
orchard
orchard
orchard
refugia
refugia
refugia
refugia
refugia
orchard
orchard
orchard
orchard
orchard
orchard
orchard
orchard
refugia
refugia
refugia
refugia
orchard
orchard
refugia
0.50
0.50
0.98
2.00
2.00
2.00
2.00
2.00
0.82
0.25
0.25
0.74
1.11
2.00
2.00
0.80
0.80
0.80
0.80
0.50
0.50
0.50
0.50
0.50
0.50
2.00
2.00
2.00
2.00
2.00
0.90
0.90
0.90
0.90
1.10
1.10
1.10
1.10
1.20
1.20
1.20
1.20
0.40
0.40
1.00
Loam
Loam
Oakleaf Richley
Unknown
Unknown
Red sandy
Red sandy
Red sandy
Oakleaf Caledon
Oakleaf Caledon
Oakleaf Caledon
Oakleaf Caledon
Sterkspruit
Unknown
Unknown
Loam
Loam
Loam
Loam
Loam
Loam
Loam
Loam
Loam
Loam
Unknown
Unknown
Unknown
Unknown
Unknown
Oakleaf Letaba
Oakleaf Letaba
Oakleaf Letaba
Oakleaf Letaba
Oakleaf Limpopo
Oakleaf Limpopo
Oakleaf Limpopo
Oakleaf Limpopo
Unknown
Unknown
Unknown
Unknown
Loam
Loam
Unknown
Rootstock
carizzo citrange
carizzo citrange
rough lemon
NA
NA
NA
NA
NA
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
NA
NA
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
rough lemon
NA
NA
NA
NA
NA
troyer citrange
troyer citrange
troyer citrange
carizzo citrange
carizzo citrange
carizzo citrange
carizzo citrange
carizzo citrange
NA
NA
NA
NA
rough lemon
rough lemon
NA
Bait Type
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
FCM
FCM
FCM
Galleria
FF
FF
FF
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
Galleria
FCM
FF
FF
FF
Accession no.
RHO Ar 17 L6
PPRI 9679
PPRI 9555
PPRI 9774
PPRI 9773
PPRI 9554
PPRI 9557
PPRI 9593
PPRI 9680
PPRI 9592
PPRI 9594
PPRI 9678
PPRI 9693
PPRI 9556
PPRI 9553
RHO OL 13 L1
PPRI 9688
RHO OL 15 L5
RHO OL 15 B3
PPRI 9689
RHO OL 19 B1
RHO OL 19 B2
RHO OL 19 B3
RHO OL 19 B5
RHO OL 19 B6
PPRI 9772
PPRI 9771
PPRI 9692
PPRI 9691
PPRI 9690
RHO Rose. 3 L2
PPRI 9681
RHO Rose. 3 B2
PPRI 9682
RHO Rose.23 L1
PPRI 9687
RHO Rose 23 L4
RHO Rose 23 L5
PPRI 9683
PPRI 9686
PPRI 9684
PPRI 9685
RHO HP 14 L5
RHO HP 14 B5
PPRI RHO R1
RHO = Rhodes University accession numbers
FCM = Thaumatotibia leucotreta,
FF = Ceratitis capitata, PPRI = Plant Protection Research Institute accession numbers
56
Table 2.3.3 The location and ecological data of Conidiobolus coronatus, Lecanicillium psalliotae, Metarhizium anisopliae and M. flavoviride
collected from citrus orchards and refugia soils in the Eastern Cape Province, South Africa.
Farm
Type
J&B Citrus
Area (Ha)
conventional
refugia
2.00
Unknown
NA
FF
Conidiobolus coronatus
PPRI 9695
Arundel
conventional
NA
FF
Lecanicillium psalliotae
conventional
Red sandy
NA
FF
Lecanicillium psalliotae
Olifantskop
organic
Loam
rough lemon
FF
Lecanicillium psalliotae
Hippo Pools
organic
1.00
2.00
0.80
1.00
Unknown
Mosslands
refugia
refugia
orchard
refugia
Unknown
NA
FF
Lecanicillium psalliotae
PPRI 9768
PPRI 9767
PPRI 9766
RHO 14 B4
Arundel
conventional
swingle citrange
FCM
M. anisopliae group
conventional
Oakleaf Richley
rough lemon
FCM
M. anisopliae var. anisopliae
Mosslands
conventional
Oakleaf Richley
rough lemon
Galleria
M. anisopliae var. anisopliae
Mosslands
conventional
Oakleaf Richley
rough lemon
Galleria
M. anisopliae var. anisopliae
Mosslands
conventional
Oakleaf Richley
rough lemon
Galleria
M. anisopliae var. anisopliae
Mosslands
conventional
Oakleaf Richley
rough lemon
Galleria
M. anisopliae var. anisopliae
Mosslands
conventional
Oakleaf Caledon
rough lemon
Galleria
M. anisopliae var. anisopliae
Mosslands
conventional
0.50
0.98
0.98
0.98
0.98
0.98
0.25
2.00
Loam
Mosslands
orchard
orchard
orchard
orchard
orchard
orchard
orchard
refugia
Unknown
NA
Galleria
M. anisopliae var. anisopliae
Olifantskop
organic
Galleria
M. anisopliae var. anisopliae
Unknown
NA
Galleria
M. anisopliae var. anisopliae
Olifantskop
organic
0.50
2.00
2.00
rough lemon
organic
orchard
refugia
refugia
Loam
Olifantskop
Unknown
NA
Galleria
M. anisopliae var. anisopliae
PPRI 9802
PPRI 9800
PPRI 9801
Mosslands
conventional
refugia
2.00
Unknown
NA
Galleria
Metarhizium flavoviridae
PPRI 9560
NA = not apllicable,
Soil Type
Rootstock
Bait Type
Fungal genus or species
Accession No.
Habitat
PPRI 9561
RHO F 14 2 B5
PPRI 9558
RHO 14 2 L6
PPRI 9562
RHO G 14 2 B5
PPRI 9803
PPRI 9559
RHO = Rhodes University accession numbers
FCM = Thaumatotibia leucotreta,
FF = Ceratitis capitata, PPRI = Plant Protection Research Institute accession numbers
57
60.0
Occurrence (% of samples)
50.0
b
b
40.0
30.0
a
a
20.0
10.0
0.0
cultivated land
refugia
Organic
Conventional
Figure 2.3.2 Mean (% ± SE) of the frequency (% of samples) of EP fungi occurring in soil
samples from orchards and refugia of organic and conventionally managed farms. Letters above
a column denote significant differences using χ2 test, P ≤ 0.005.
2.4 DISCUSSION
2.4.1 Bait insects and fungal species
This study confirmed the sensitivity of the Galleria bait method, as significantly more EP fungal
isolates were obtained using this insect species than either T. leucotreta or C. capitata. The
observed differences in isolation of EP fungal species and isolate numbers in this study may in
part be attributed to the biology of the bait insect species used (Smith 1937; Thomas et al. 2001;
Smith & Peña 2002; Stibick 2008), as well as their relative susceptibilities to fungal infection
(Vilcinskas et al. 1997; Konstantopoulou & Mazomenos 2005; Begemann 2008). Beauveria
bassiana was isolated significantly more often from G. mellonella than was any other EP fungal
species from any other bait insect. This fungal species was first described from South African
citrus orchards at Zebediela in the Northern Province by Schoeman in 1960, who found it
58
growing on pupae of the citrus measuring worm, Ascotis selenaria receprocaria (Begemann
2008). In an extensive survey undertaken by Hatting et al. (2004), B. bassiana was the most
commonly isolated EP fungus (87%) from South African soils baited with G. mellonella.
Various other authors have also reported B. bassiana to be the most commonly recovered EP
fungal species, using the bait insect G. mellonella, from soil samples (Chandler et al. 1997;
Meyling & Eilenberg 2006b; Quesada-Moraga et al. 2007). Klingen et al. (2002) investigated
and compared the use of G. mellonella and Delia floralis as bait insects. No significant
differences were observed in the number of isolates of EP fungi recovered from both insect
species but the fungal species, Tolypocladium cylindrosporum, was isolated more frequently
from D. floralis (Anthomyiidae) than G. mellonella. When C. capitata was used as a bait insect
in the present study, it was the only insect to isolate the fungal species, L. psalliotae. The larvae
of G. mellonella are known to be highly susceptible to insect pathogens (Chandler et al. 1997;
Vilcinskas et al. 1997; Meyling 2007) and this may explain the large number of EP fungal
isolates which were recovered from this insect bait species when compared to the other bait
insects used. The soil environment is an integral habitat for the completion of the lifecycles of
both T. leucotreta and C. capitata, therefore these bait species may be better adapted to the
adversities of this environment, which may include lower susceptibility to fungal infections,
compared to G. mellonella, to which the soil is not a natural habitat. One feature of the biology
of T. leucotreta is their ability to spin a cocoon made from silk and sand particles prior to
pupation (Stibick 2008). This cocoon may minimize the possibility of the spores of EP fungi
coming into direct contact with the cuticle of the insect. The spores of EP fungi may adhere to
the cocoon but because of the lack of direct contact with the insect cuticle may fail to germinate.
Perhaps future baiting procedures using this insect species should consider heat-treating the
larvae, as is done with G. mellonella, but this warrants further investigation.
2.4.2 Fungal ecology
The 21.53% occurrence of five species of EP fungi in soils from the Sunday‟s River Valley
citrus producing area and surrounding areas in Eastern Cape Province, South Africa, is lower
than the expected range when compared to studies in various other countries. Ali-Shtayeh et al.
(2002) found insect-pathogenic fungi to occur in 33.6% of the soil samples studied in irrigated
vegetable fields and citrus orchards in Palestine. Other typical recovery rates include: 38.6% in
Mauritius (Sookar et al. 2008), 52% in the Pacific North West, USA (Bruck 2004), 55.5% in
China (Bing-Da & Xing-Zhong 2008), 71.7% in Spain (Quesada-Moraga et al. 2007), 91% in
59
Ontario, Canada (Bidochka et al. 1998), and 96% in Switzerland (Keller et al. 2003). Chandler et
al. (1997) reported that EP fungi were common inhabitants of the soil biota, but the diversity of
species was low, usually with one or two species occurring frequently; this observation proved
true for the present study and has been noted by other authors (Keller et al. 2003; QuesadaMoraga et al. 2007).
The occurrence of L. psalliotae was low and was only recovered from C. capitata larvae.
Literature is scarce on the distribution and occurrence of this nematophagous fungal species
(Gan et al. 2007) which can also be EP and capable of infecting other arthropod hosts (Evans &
Whitehead 2005; Pirali-Kheirabadi et al. 2006). Metarhizium flavoviride was only isolated once
from a soil sample in this study however its‟ isolation has been recorded in other studies and this
fungal species is considered to have a wide geographic distribution (Braga et al. 2001). Meyling
& Eilenberg (2006b) for example found M. flavoviride to occur commonly (21.9%) in soils of a
Danish research farm in Copenhagen. Conidiobolus coronatus was also isolated once in this
study. Hatting et al. (1999) conducted a survey on the fungal pathogens of aphids from South
Africa and revealed that C. coronatus was not isolated from field-collected cereal aphids in
South Africa; in fact the only record of this fungal species was obtained from a mycosed aphid
cadaver in an aphid-rearing facility in Bethlehem, South Africa. The study however did not
include soil samples and thus the limited occurrence of C. coronatus must be carefully
considered. However, Ali-Shtayeh et al. (2002) found C. coronatus to be the most frequently
isolated EP fungus (31.4%) in citrus orchard soils in Palestine. The limited number of fungal
isolates obtained from the species M. flavoviride, C. coronatus and L. psalliotae may mean that
these fungal species are locally rare or simply that they may not have been detected by the bait
method protocol for various reasons which are discussed further below. However, based on the
studies mentioned, these species of fungi do occur with more prevalence worldwide.
The low incidence of soil samples which yielded more than one EP fungal species in the same
sample (1.38%) may be an artifact or limitation of the method of qualitative quantification of
infection constantly used in the Galleria baiting technique. As long as there were more than one
morphologically identifiable fungal species within a sample pot, detection of multiple EP fungal
species was easy. However, genetically different Beauveria species within the same pot may
have been missed due to the extreme difficulty in determining genetically distinct Beauveria
species based on morphology alone. Meyling et al. (2009) inferred species diversity and
reproductive potential using molecular techniques on Beauveria within a single arable field and
60
bordering hedgerow in Denmark. Seven phylogenetic species were discovered including five
phylogenetic species within B. bassiana sensu stricto. These analyses demonstrate the genetic
diversity of B. bassiana within a relatively limited geographical area with the greatest genetic
diversity observed in the semi-natural habitat of the hedgerow compared to the arable field.
Furthermore, it was found that Beauveria phylogenetic species also occupied overlapping niches
within the hedgerow habitat. Based on these results it may be possible that some genetically
distinct Beauveria isolates could have been overlooked in a particular sample during the fungal
recovery process using the insect bait technique.
This study showed a greater likelihood of finding EP fungi in refugia than in cultivated habitats,
but no differences were observed in the recovery of fungi when organic and conventional
farming systems were considered. Klingen et al. (2002) however, found more EP fungi in
cultivated fields of organically farmed soils and, no significant differences in the occurrence of
EP fungi were found between the refugia of the two cropping systems (Klingen et al. 2002).
There was a common result in that a positive relationship existed between occurrence of fungi
and organically farmed soils in the present study. Beauveria bassiana and M. anisopliae var.
anisopliae were the most commonly isolated EP fungal species in soils from the Eastern Cape
Province. On five of the six farms, Arundel Farm being the only exception, B. bassiana was
recovered more often from soil samples taken from refugia than from cultivated lands.
Furthermore, there was a significant difference in the recovery of B. bassiana from organic
compared to conventionally farmed soils.
Mietkiewski et al. (1997) and Chandler et al. (1997) in the United Kingdom both showed that
the frequency of occurrence of EP fungi in intensively cultivated soils was lower than in
marginal habitats. Meyling et al. (2009) attributed greater Beauveria diversity in seminatural
habitats of hedgerows in Denmark to increased humidity, reduced ultra-violet light, long-term
environmental stability and a greater abundance and diversity of insect hosts. Natural habitats are
expected to host more diverse and stable insect communities compared to cultivated lands, due to
the absence of various pest control tactics, and the structurally more complex, shaded and
undisturbed composition of these habitats. Meyling & Eilenberg (2007) suggested that insect
hosts should still be considered the primary source of organic matter for fungal population build
up. An increase in organic matter content (from surrounding vegetation) in the soils of natural
habitats may also favour the prevalence of EP fungi. Quesada-Moraga et al. (2007) suggested
that soils with greater organic matter content have higher cation exchange capacities with greater
61
organic matter enhancing conidia adsorption. Meyling & Eilenberg (2006b) however found that
organic matter content and subsequent increase in the biological activity in the soil adversely
affected the persistence of EP fungal species due to antagonistic effects. Mainly because both
Beauveria and Metarhizium are poor competitors for organic resources compared to
opportunistic saprophytic fungi that are ubiquitous in the soil (Meyling & Eilenberg 2007).
The natural occurrence of EP fungi in the soil is influenced by a complicated set of abiotic and
biotic factors such as soil structure (Jaronski 2007), organic matter in the soil (Quesada-Moraga
et al. 2007), ultra-violet light (UV-B) (Braga et al. 2001), cultural practices: pesticide
applications, tillage regimes, crop cover and rotations, fertilizers (Filho et al. 2001; Saenz-deCabez Irigaray et al. 2003; Jaronski 2007; Meyling & Eilenberg 2007) and insect hosts. In light
of some of the abovementioned factors, it appears that organically farmed soil may be a more
suitable habitat for insect-pathogenic fungi compared to cultivated, conventionally farmed soils
(Klingen et al. 2002; Meyling & Eilenberg 2006b), in which the presence of insecticides and
especially fungicides may have a direct killing effect on EP fungi (Khalil et al. 1985; Klingen et
al. 2002; Meyling & Eilenberg 2007). One of the conventional farms in this study, J&B Citrus,
regularly applied the fungicide mancozeb to sample orchards for the control of citrus Black Spot,
Guignardia citricarpa. Lagnaoui and Radcliffe (1998) however, found mancozeb to be strongly
inhibitory of germination of EP fungal conidia. Majchrowicz and Poprawski (1993) examined
the in vitro effects of fungicides on fungi and found that dithiocarbamate derivatives zineb and
copper oxychloride applied together or mancozeb applied alone, completely inhibited the
germination of six fungi tested, including B. bassiana. Further the uses of the broad-spectrum
fungicide azoxystrobin, and the organophosphate insecticide methidathion, within the orchards at
J&B Citrus may also have contributed, in part, to lower incidences of fungal isolates from this
farm because both compounds have been shown to negatively affect EP fungi in vitro (Puterka
1999; Jaronski 2007).
Cultivated citrus orchards are perennial habitats and may support more diverse and stable
microbial communities compared to annual cropping systems (Ali-Shtayeh et al. 2002); tree
canopies provide shade and maintain humidity levels in the soil, which in turn reduce harmful
UVB rays from penetrating the soil and desiccating fungal spores (McCoy et al. 2007). The use
of compost teas (and manure) or organic fertilizers may have a positive effect on the occurrence
of EP fungal species because an increased carbon load in the soil is favourable to soil inhabiting
insects, which themselves, are potential hosts for fungi (Ali-Shtayeh et al. 2002; Klingen et al.
62
2002). This together with the fact that conventional farming practices in the Eastern Cape are
considered „soft‟ with emphasis placed on IPM strategies rather than intensive pesticide
applications, may be one reason for the limited differences between farming systems when
comparing cultivated orchards. This was shown by the large recovery of EP fungal isolates (18
isolates) from Mosslands Farm. Further, the conventional farm Arundel, showed proportionally
more diversity of EP fungi in the soils. Very few agrochemicals and other additives were applied
in citrus orchards at this location. The soils at Arundel Farm are predominantly sandy loam to
which supplements of humic acid and compost were made periodically. The locality of this farm
is such that the Sundays River flows through it, thus the soils could be considered younger in age
and more underdeveloped (less clay content) compared to those further away from the river, such
as at Rosedale (Finnemore, pers. comm.). Loamy soils are composed of sand, silt, and clay in a
relatively even concentration (about 40-40-20% respectively), and generally contain more
nutrients and humus than sandy soils and have better infiltration and drainage than silt soils (Soil
Classification Working Group 1991). In contrast, the organic farm Hippo Pools yielded a low
incidence of fungal isolates; the owner reported that the farm was initially managed as a
conventional citrus farm with an intense pesticide regime prior to his acquisition of the property.
It had been managed as an organic citrus farm for the past six years. Thus, it is also important to
consider the history of a sampling area during collection studies. Similarly, at the organic farm
Olifantskop, where a large number of fungal isolates were obtained (19 isolates), the farmer
reported that his orchards were all first time plantings on virgin soils, which essentially were
natural habitats, and this may explain the high occurrence of fungi at this farm.
The direct application of EP fungi, in the form of microbial products, in organic and some
conventional farming practices may also have a significant influence on the occurrence and
recovery of fungi in these systems. The organic citrus farm, Rosedale for example, made use of a
Beauveria bassiana formulated product called Bb Plus® (Biological Control Products (BCP),
South Africa) which was applied to two of the orchards from which soil samples were collected,
six months prior to the sampling period. Interestingly enough only four Beauveria isolates were
recovered from the two orchards augmented with the Bb Plus® product. The rest of the isolates
were recovered from the only orchard to which the product had not been applied, and from the
natural areas. Thus, the incidence of occurrence of B. bassiana isolates from the cultivated
orchard sites 3, and 15 at Rosedale may or may not be attributed to residual spores from the
direct application of Bb Plus®. The only way of determining the presence of residual Bb Plus®
conidia would be to genetically analyze the isolates. However, this was not the focus of this
63
study and there were only four isolates obtained from the areas where the product had been
applied. Vänninen et al. (2000) in discussing the persistence of augmented M. anisopliae and B.
bassiana in Finnish agricultural soils, found lower abundances (a 44% reduction) of fungal
conidia in peat soil sites compared to both sandy (a 37% reduction) and clay (a 14% reduction)
soils, just one year after application. Beauveria bassiana conidia were generally less persistent in
all soil types compared to M. anisopliae (Vänninen et al. 2000). The agricultural soils at
Rosedale were all classified as Oakleaf form, which comprise 20% clay content and are
calcareous (Finnemore, pers. comm.). These soils drain well but are relatively depauperate of
nutrients and organic matter; soils are annually supplemented with compost and compost teas are
applied every two months (Finnemore, pers. comm). This practice however could stimulate the
growth of saprophytic fungi which could subsequently antagonise EP fungal persistence in the
soil. The conidia of the applied Bb Plus® product may still have been present in the soil of
orchards 3 and 15, however determination of conidial persistence is complex (Vänninen et al.
2000; Jaronski 2007).
Increasing evidence suggests that habitat selection and not insect host selection drives the
population structures of deuteromycete EP fungi (Bidochka et al. 2001; Bidochka et al. 2002)
particularly in light of recent studies which advocate that some EP fungal genera are facultative
pathogens, capable of multiple ecological roles (Goettel et al. 2008; Vega et al. 2009) and that
adaption to a specific habitat type is a more important selection criterion for identifying potential
biological control candidates (Bidochka et al. 2002). While it may be important to obtain highly
infective and virulent strains of fungi to control insect pests, it is even more important to look for
isolates which are also capable of surviving and persisting in the environments and habitats for
which they are required.
2.4.3 Potential for biological control
Twenty-one of the 62 collected indigenous fungal isolates were selected for testing against the
subterranean life stages of T. leucotreta, C. capitata and C. rosa in conidial-sand incorporated
bioassays to evaluate their potential as biological control agents (cf Chapter III). Indigenous
isolates were selected so as to represent an even geographic spread from the six farms in the
Eastern Cape. The EP fungal isolates selected for conidial-sand incorporation bioassays are
shown in Figs 2.4.1-2.4.6. Fungal isolates were grown on SDA supplemented with 1 ml/L
Dodine, 50 mg/L Chloramphenicol, 50 mg/L Ampicillin or 50 mg/L Rifampicin in Petri dishes
64
and maintained at an ambient temperature of 25°C. Fungal isolates were labelled according to
the bait insect species they were originally isolated from, followed by the farm name and then
the orchard or refugia sample number.
Figure 2.4.1 A, B, D- Metarhizium anisopliae var. anisopliae C- Metarhizium flavoviride
65
Figure 2.4.2 E, F- Metarhizium anisopliae
G, H- Beauveria bassiana
Figure 2.4.3 I, J, K, L- Beauveria bassiana
66
Figure 2.4.4 M, N, O, P- Beauveria bassiana
Figure 2.4.5 Q, R, S, T- Beauveria bassiana
67
Figure 2.4.6 U- Beauveria bassiana (Photo Credits, T. Goble, Rhodes University)
68
III
ASSESSING THE VIRULENCE OF
ENTOMOPATHOGENIC FUNGI AGAINST CITRUS
PESTS
3.1 INTRODUCTION
The ecological distribution and occurrence of indigenous isolates of EP fungi from citrus
orchards and refugia was examined in Chapter II and their potential use as biological control
agents (BCA‟s) against fruit flies and false codling moth was recognized. This chapter will
investigate the physiological effects and host specificity of various indigenous fungal isolates
obtained in the Eastern Cape, to establish whether these isolates can be effectively used as
control agents against the subterranean life stages of C. capitata, C. rosa and T. leucotreta. The
biology of these key citrus pest species includes their dropping directly from fruit trees to pupate
in the soil or leaf litter below (Thomas et al. 2001; Ekesi et al. 2007; Begemann 2008, Stibick
2008). Since most EP fungi are soil-borne microorganisms, the investigation of a management
strategy option in citrus, targeted at the pre-pupating larvae and pupae of these insect pest species
in the soil, needs to be investigated. For this reason, it was decided to explore soil as an
inoculation medium for citrus pest control. Unlike arborial habitats, the soil environment is less
prone to dramatic environmental fluctuations in temperature and humidity, which can have
negative effects on the recycling and persistence of EP fungi (Ansari et al. 2007). Van den Berg
(2001) reported that natural infection by Beauveria bassiana was often recorded in T. leucotreta
pupae found amongst the leaf litter in Zebediela Citrus Estate, Northern Province, South Africa.
Fungi with broad host ranges, such as B. bassiana, are frequently even more specific under field
conditions because of complex abiotic and biotic factors, thus reports from Van den Berg (2001)
that natural infections have already been observed in T. leucotreta illustrates a potential for
control strategies in these orchard environments. Begemann (2008) studied the potential use of
B. bassiana against the subterranean life stages of T. leucotreta in South Africa and concluded
69
that, under natural conditions, the final instar migrating larvae become exposed to the fungus
when leaving the fruit to pupate in the soil. Despite limited reports of epizootics of fungal
disease of fruit flies in nature (Ekesi et al. 2007) numerous studies have demonstrated the
potential success of soil treatments with EP fungi for control management strategies against fruit
flies (Ekesi et al. 2002; Ekesi et al. 2005; Mochi et al. 2006; Ekesi et al. 2007).
Accurate confirmation of fungal performance through the employment of bioassays is an
important step in the development of a mycoinsecticide. A „biological assay‟ is defined as “the
measure of the potency of a stimulus (chemical, physical or biological) by means of the reactions
that it produces in living matter” (Hatting & Wraight 2007) and is employed to test the
physiological host range of a particular fungal isolate against insect hosts under controlled
conditions (Finney 1979). When using EP fungi as the stimulus against insect pests, the most
common infectious units employed are the conidia or the yeast-like hyphal bodies (blastospores)
and the intensity of response among the treated insects is used as the measure of the effect of this
stimulus or “the disease power of the microorganism” (Hatting & Wraight 2007). The median
lethal dose (LD50) is the most precise expression of virulence. However, this is very difficult to
determine and often fungal virulence is expressed as a concentration (Hatting & Wraight 2007)
and most commonly the median lethal concentration (LC50) required to kill 50% of the test insect
population is employed. Median lethal time (LT50) is used to quantify the activity of virulent
fungi and is important in expressing the time until insect death (which is important to farmers).
Hatting and Wraight (2007) report that lethal time is proposed as “the period of exposure to a
pathogenic stimulus which will produce death in half the test subjects. The length of exposure is
a direct measure of the dosage, and an increase in the period of exposure results in an increased
uptake and true dose in the same ratio...” Further understanding of bioassay procedure and the
practice of statistical inference can be obtained in Finney (1979).
Experimental accuracy is absolutely crucial in bioassay procedures and the effects of what may
go wrong are discussed in detail by Hatting & Wraight (2007). Not only is the accuracy of the
bioassay important, but there are essential factors which need to be considered in conjunction
with these assays. These factors include issues such as attenuation of fungi due to repeated subculturing in the laboratory. Wang et al. (2005) reported spontaneous phenotypic degeneration
which resulted in sterile fungal culture sectors when sub-culturing Metarhizium anisopliae on
artificial media in vitro (Fig. 3.1). Fungal strains which are continually grown on artificial media
can show culture degradation which results in either a loss or reduction in sporulation and
70
virulence, and cultures are often observed with a fluffy-mycelium type growth (Wang et al.
2005). Unfortunately, fungal culture degradation is irreversible and inheritable and can result in
extensive commercial losses (Wang et al. 2005). These authors attempted to grow sterile fungal
cultures (those indicative of colony sectorization) on host-related media. However, this did not
result in the recovery of wild-type characteristics nor was rejuvenation experienced. They
concluded that perhaps these data may alleviate the long standing dispute as to whether passage
through insect hosts can or cannot restore the virulence of attenuated EP fungal strains. Hatting
& Wraight (2007) however, reported that numerous studies have demonstrated successful
restoration of virulence in M. anisopliae, B. bassiana and Isaria farinosa, following passage
through an insect host. Furthermore, Dimbi et al. (2004) reported that the virulence of
Metarhizium strains was maintained by regular passage through insect hosts during sub-culturing
and maintenance of fungal strains. However, in the laboratory, isolates of fungi with broad host
ranges are generally considered most virulent to the host insect from which they were first
isolated (Goettel 1995). So the question of whether passage through any insect host would
maintain this virulence needs to be considered. Brownbridge et al. (2001) reported that the B.
bassiana GHA strain, which is the active ingredient of the commercial products, BotaniGard®
and Mycotrol®, after being passed through three different insect host species, maintained its
virulence and stability in vitro. Further, Aemprapa (2007) noted that no fungal isolates of
dipteran origin were used in the screening of fungi against Bactrocera dorsalis in Thailand, and
yet all fungi were highly pathogenic towards this fly species. This suggests that fungi from any
insect origin would have the potential to elicit a high level of pathogenicity against test insects,
provided they are susceptible.
Standardization of the growth substrate is important to reduce variability among fungal isolates
during bioassay procedure (Hatting & Wraight 2007), and it is important that EP fungal species
receive the best nutrition in vitro at temperatures optimal for growth, germination and
sporulation. The use of Sabouraud Dextrose Agar (SDA) which contains peptone (10 g),
dextrose (40 g) and agar (10 g), is often used in the isolation and maintenance of most
hyphomycete fungal strains, and can be further supplemented with 0.1% yeast extract and
glucose (10 g) (Brownbridge et al. 2001; Meyling 2007).
71
Figure 3.1 A, B Different types of culture colonies of Metarhizium anisopliae indicating sterile
sectorization and the restoration of conidiation in some areas (Wang et al. 2005).
The ratio of antibiotics and fungicides in the media, which are used to reduce the amount of
bacteria and saprophytic fungal species growth when culturing EP fungi in vitro, are also
essential. Hu & St Leger (2002) found that by using Veen‟s semi-selective media minus the
fungicide cyclohexamide, more species of EP fungi were observed from collected soil samples
than when the fungicide was included in the medium. The discovery that dodecylguanidine
monoacetate (dodine) can be used to selectively inhibit non-EP fungi has simplified isolation of
EP fungi and enhanced their maintenance on artificial media (Jaronski 2007). The addition of
strong concentrations (greater than 50 mg/L) of the antibiotic, Rifampicin, can reduce mycelia
growth of indigenous isolates of EP fungi (personal observation). For more information
regarding the isolation and culturing of EP fungi in vitro and the effects various media have on
fungal growth factors, it is suggested that more classic works be referred to (Chase et al. 1986;
Li & Holdom 1995; Ibrahim et al. 2002; Meyling 2007; Safavi et al. 2007).
Insect hosts factors such as development stage, age, sex and activity may influence insect
response to EP fungi. Test insects must be selected carefully to limit the effects of these factors
so as not to confound the effects of the conidial concentrations applied during bioassay
procedures (Van Ark 1995). In a study by Begemann (2008), various concentrations of B.
bassiana in suspension and as dry preparations were tested against the final instar larvae of T.
leucotreta. The author found that non-migrating larvae illustrated a dose-dependant response
with LC50 and LC90 values in the range of 5.2 x 104 to 1.0 x 106 and were more susceptible to the
72
fungus. There was however no dose-dependant response observed for migrating T. leucotreta
larvae and this was attributed to lowered susceptibility by migrating larvae, possibly due to the
moulting process already taking place.
When considering fruit flies as tests insects during conidial-sand incorporation bioassays, it is
crucial to use third instar larvae which are „jumping‟. This indicates fitness of the test insect and
ensures that the correct life stage has been utilised. Hatting & Wraight (2007) further suggest
that homogenisation (thorough mixing) of stock suspensions of conidia or blastospores, and the
investigation of the viability of conidia (generally measured in terms of percentage germination),
are also important deliberations during bioassay preparation. Consideration of all the
abovementioned factors may reduce the amount of variation within and among fungal isolates
tested in bioassays.
As a group, fungi have the widest host ranges among insect pathogens, but host ranges vary
considerably depending on fungal species (Inglis et al. 2001). An attribute of good fungal BCA‟s
includes that their host ranges are not too specific, as is case with some of the entomophthoralean
species of fungi which infect only one genus of insect, as in Formica ants (Boer, 2008) or the
fungus Nomuraea rileyi which infects Lepidoptera in the Noctuidae (Inglis et al. 2001). Fungal
genera such as Beauveria and Metarhizium are known for their broad host ranges which is
desirable in being able to knock down more than one species of insect (Goettel 1995).
Concurrently, insect pests should exhibit susceptibility to fungal pathogens and occupy similar
niches in the environment so as to increase the advantage of fungal BCA‟s and their use (Ekesi et
al. 2002).
Other important considerations are virulence (which is the quantitative amount of disease a
pathogen can stimulate in insect hosts) and subsequent speed of kill. Fungal strains which are
more virulent will require less inoculum in the environment and production of these strains will
therefore be cheaper (Inglis et al. 2001). The use of genetic engineering by the directional
addition of one or more genes coding for pathogenicity is the most common approach to shifting
the physiology of EP fungi to enhance virulence and has been described in greater detail by St
Leger & Screen (2001). A high level of sporulation of fungal cultures, either on host cadavers or
on artificial media, is also considered a desirable trait because each pupa dying in the soil
constitutes an infection site. The more prolific the conidial growth on the dying host, the more
easily the increased inoculum is dispersed (Ekesi et al. 2002) and the more desirable the fungal
73
isolate becomes when commercialisation is considered (Brownbridge et al. 2001). Increasing
evidence suggests that while it may be important to obtain highly infective and virulent strains of
fungi to control insect pests, it is even more important to look for isolates which are also capable
of surviving and persisting in the environments and habitats for which they are required
(Bidochka et al. 2001; Bidochka et al. 2002). A comprehensive list of environmental
considerations of fungal BCA‟s is listed in Chapter I.
Soil moisture is considered an important environmental factor in determining efficacy of fungal
isolates when controlling subterranean life stages. At 20-30°C, mortality in pupae was highest at
a water potential of -0.1 and -0.01 mega Pascal (MPa) and lowest at a water potential of -0.0055
and -0.0035 MPa (Ekesi et al. 2003). This is important in the use of EP fungi as an integrated
pest control tactic because of the large amounts of water required in citrus production
(Wellington 1960). Some studies have shown that soil moisture content in excess of 30%
moisture holding capacity is detrimental to B. bassiana longevity in the soil. This could be
attributed to reduced oxygen concentrations in wet soil as well as higher bacterial decomposition
of organic matter, which is known to occur in soils with water potential approaching zero (Ekesi
et al. 2003). Jaronski (2007) observed that the infectivity of four B. bassiana isolates against the
Southern Corn Rootworm, Diabrotica howardii, was inversely proportional to soil moisture
above 25% field capacity, but that the controls were unaffected, implying that soil moisture
affected the fungal infection process. Thus, it is crucial during bioassay procedure that
appropriate soil moisture levels are maintained, preferably using moisture levels which would be
indicative of a field scenario.
Consideration of all the abovementioned points will ensure, to a large degree that the correct
assumptions are made about environmental factors which affect fungal performance and
subsequent variation which may occur among replicates during bioassay procedures. It further
ensures that the physiological effects and host range of various indigenous fungal isolates are
correctly quantified for the control of subterranean life stages of citrus insect pests.
The aims of this chapter were 1.) To compare the pathogenicity of 21 indigenous isolates of EP
fungi against T. leucotreta, C. capitata and C. rosa final instar larvae using sand-conidial
suspension incorporation bioassays. These tests were essentially screening trials to determine the
most virulent fungal isolates, which had broad host ranges and were capable of infecting all three
insect test species; and 2.) Test the four most virulent isolates, and a commercially available
74
product Bb Plus® (BCP, South Africa) (which has Beauveria bassiana as the active ingredient)
against the three citrus pests using various concentrations of fungal conidia incorporated into
sand to ascertain the LC50 and LC90 values of each isolate; and 3.) Test the three most virulent
fungal isolates which were highly pathogenic to T. leucotreta in concentration-response trials
against final instar T. leucotreta larvae using four different exposure times, 12, 48, 72 and 96 hrs.
3.2 METHODS & MATERIALS
3.2.1 Fungal cultures
The pathogenicity of 15 Beauveria bassiana, five Metarhizium anisopliae and one Metarhizium
flavoviride isolates, obtained using the insect-bait method (Table 3.2.1; cf Chapter II: Figs 2.4.12.4.6), were tested against T. leucotreta, C. capitata and C. rosa final instar larvae. All isolates,
except one (RHO G B Ar 23 B3), are currently housed in the South African National Collection
of Fungi at the Plant Protection Research Institute (Agricultural Research Council) in Pretoria,
South Africa. Fungal isolates used in this study were grown on Sabouraud Dextrose Agar (SDA)
supplemented with 1 ml Dodine, 50 mg/L Chloramphenicol, 50 mg/L Ampicillin or 50 mg/L
Rifampicin in Petri dishes and maintained at an ambient temperature of 25°C. Fungal isolates are
labelled according to the bait insect species they were originally isolated from, followed by the
name of the farm and then the orchard or refugia soil sample number.
75
Table 3.2.1 Twenty one indigenous EP fungal isolates tested against Thaumatotibia leucotreta, Ceratitis capitata and C. rosa final instar larvae
Fungal Species Rhodes Code
Isolate No.
Host Species
Locality
Soil Type
Co-ordinates
Date of Isolation
B. bassiana
FCM 11 3 B2
PPRI 9592
G 14 2 B5
PPRI 9555
G Moss R12
PPRI 9593
FCM 10 13 L1
PPRI 9680
G Moss R1
PPRI 9774
G Moss R10
PPRI 9557
G Moss R7
PPRI 9554
G Ar 17 B3
PPRI 9679
G B Ar 23 B3 RHO Ar 23 B3
FF J&B R5
PPRI 9556
G J&B 14 L2
PPRI 9678
G OL R10
PPRI 9691
G OL R11
PPRI 9690
G Rose 20 L2
PPRI 9687
FCM Rose R9
PPRI 9685
T. leucotreta
Galleria
Galleria
T. leucotreta
Galleria
Galleria
Galleria
Galleria
Galleria
C. capitata
Galleria
Galleria
Galleria
Galleria
T. leucotreta
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Mosslands
Arundel
Arundel
J&B Citrus
J&B Citrus
Olifantskop
Olifantskop
Rosedale
Rosedale
Oakleaf Caledon
Oakleaf Richie
Sandy red soil
Oakleaf Caledon
Unknown
Sandy red soil
Loam
Loam
Loam
Unknown
Unknown
Unknown
Unknown
Oakleaf Limpopo
Unknown
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°30′57″S; 25°39′11″E
33°30′57″S; 25°39′11″E
32°45′56″S; 25°45′46″E
32°45′56″S; 25°45′46″E
33°37′14″S; 25°40′49″E
33°37′14″S; 25°40′49″E
33°32′21″S; 25°41′39″E
33°32′21″S; 25°41′39″E
April 2008
May 2008
May 2008
April 2008
June 2008
June 2008
May 2008
May 2008
May 2008
April 2008
June 2008
July 2008
July 2008
August 2008
August 2008
M. anisopliae
G 14 2 B3
PPRI 9562
G 11 3 L6
PPRI 9803
FCM AR 23 B3 PPRI 9561
G OL R3
PPRI 9800
G OL R8
PPRI 9801
Galleria
Galleria
T. leucotreta
Galleria
Galleria
Mosslands
Mosslands
Arundel
Olifantskop
Olifantskop
Oakleaf Richie
Oakleaf Caledon
Loam
Unknown
Unknown
33°23′54″S; 26°25′41″E
33°23′54″S; 26°25′41″E
33°30′57″S; 25°39′11″E
33°37′14″S; 25°40′49″E
33°37′14″S; 25°40′49″E
March 2008
April 2008
April 2008
July 2008
July 2008
M. flavoviride
G Moss R2
Galleria
Mosslands
Unknown
33°23′54″S; 26°25′41″E
June 2008
PPRI 9560
PPRI = Plant Protection Research Institute Fungal Collection accession no RHO = Rhodes University accession no.
FCM = False codling moth, Thaumatotibia leucotreta FF = Fruit fly, Ceratitis capitata Galleria - G. mellonella
76
3.2.2 Preparation of conidial suspensions
Fungal conidia were harvested from 2 to 3 week old surface cultures by scraping with a glass
rod. Spores were suspended in 20 ml sterile distilled water supplemented with 0.05% Triton X100 in sterile glass bottles containing 3 mm glass beads. Bottles were sealed and vortex mixed
for 2 minutes to produce a homogenous conidial suspension. Conidial concentrations were
determined using a Neubauer haemocytometer following serial dilution in sterile distilled water
and conidial suspensions were used within 3 hrs of enumeration. Viability of conidia was
determined by spread plating 0.1 ml of conidial suspension (titrated to 1 × 107 conidia per ml-1)
on three SDA plates. A sterile cover slip was placed on each plate and incubated in complete
darkness at ambient temperature of 25°C. Percentage germination was examined after 24 hrs
from 100-spore counts on each plate (Ekesi et al. 2002).
3.2.3 Inoculation of insects
Sterile, dried river sand was sieved through a 16-mesh screen and transferred into 90 mm
diameter Petri dishes. A standard concentration of 1 x 107 conidia ml-1 (5ml) was used to
inoculate 50 g of sand, which was then vigorously mixed with a spatula. Controls were treated
with sterilized distilled water containing 0.05% Triton X-100. Bioassays were also conducted at
a higher dose concentration of 1 x 108 conidia ml-1 and were initiated using four fungal isolates
which performed well at the lower concentration (1 x 107 conidia ml-1). Twenty T. leucotreta, C.
capitata and C. rosa final instar larvae, which would be ready to pupate within the next 24 hrs,
were then introduced seperately into each Petri dish. The dishes were incubated at ambient
temperature (25°C) under a photoperiod of L12: D12. Seven days after the introduction of
larvae, pupae were removed from Petri dishes of treated sand and transferred into new Petri
dishes containing sterilized, untreated sand. Sterilized plastic bottles were cut 15 cm below the
neck of the bottle and used as individual emergence chambers. The Petri dishes were placed in
trays filled with autoclaved Pearlite and covered with cut plastic bottles (Fig. 3.2.3A). For fruit
flies, a cotton bud soaked in water and a small lid (3 x 3 cm) containing a 4:1 mixture of sugar
and yeast hydrolysate, were introduced into the neck of the bottle and the emergence chamber
respectively to provide food. For T. leucotreta adults, an 8% sugar solution was provided. Test
insects were maintained at the same conditions as described above. The number of adult flies and
moths that emerged from treated and control sand were recorded every second day until 10 days
77
after the first emergence. Records were also kept of the number of pupae that failed to emerge.
Adult flies and moths that died during this period and pupae or larvae that failed to emerge were
surface sterilized in 70% ethanol and transferred onto SDA (Fig. 3.2.3B). The criteria for scoring
mycoses were (1) failure of pupae to emerge followed by fungal sporulation and (2) death of
adults followed by fungal sporulation. Each treatment was replicated four times with 20 insects
per replicate.
3.2.4 Concentration-response of citrus pests to EP fungi
In a similar set up as described above (Section 3.2.3), sand-conidial incorporation bioassays
using four fungal isolates that were highly virulent to all three test insect species in previous
bioassays were selected to assess the concentration-mortality relationship and to determine the
lethal concentration LC50 and LC90 values for each fungal isolate. The registered microbial
control product, Bb Plus® (BCP, South Africa) was also included as an additional isolate in the
concentration-mortality trials. This product was stored and used according to the manufacturer‟s
instructions. Conidial concentrations of the five isolates were determined using a Neubauer
haemocytometer following five-fold serial dilution in sterile distilled water supplemented with
0.05% Triton X-100. Based on previous bioassays, four concentrations of fungal inocula were
calibrated for bioassays with T. leucotreta with the following values: 1 x 104, 1 x 105, 1 x 106, 1
x 107 conidia ml-1. For C. capitata and C. rosa the following concentration inocula were used 1 x
107, 1 x 108, 1 x 109, 1 x 1010 conidia ml-1. It was determined that concentrations of 1 x 1011 and
1x1012 conidia ml-1 would most likely be needed for the fruit fly inoculums. However, due to the
difficulty in obtaining enough conidial suspension volume, only the concentrations stipulated
were used. Viability of conidia was determined by spread plating (as described in Section 3.2.2),
and percentage germination was examined after 24 hrs from 100-spore counts on each plate. The
criteria for scoring mycosis were undertaken as described in Section 3.2.3. Each treatment was
replicated four times with 20 insects per replicate.
78
Figure 3.2.3 (A) Cut plastic bottles used as emergence chambers placed on Pearlite during
bioassay experiments. Note the cotton wool bungs used to introduce water into the chambers.
Orange lids were filled with 1:4 yeast hydrolysate and sugar mixture to feed flies (B) During the
scoring of mycosis dead insects were placed onto SDA agar. (Photo Credits: T. Goble, Rhodes
University).
3.2.5 Exposure time-response trial of Thaumatotibia leucotreta to EP fungi
In a similar experimental set up as described in Section 3.2.3, the exposure time-mortality
relationship of three isolates which had the lowest LC50 values against T. leucotreta in the
concentration-mortality trials (Section 3.2.4), were selected to be tested against T. leucotreta
final instar larvae. Based on previous bioassays it did not seem meaningful to ascertain lethal
exposure time (LT) values for fruit flies because concentration values were already far too high
to be viable. Thaumatotibia leucotreta larvae were subjected to EP fungal isolates at four
different time periods: 12, 48, 72 and 96 hrs after insect inoculation, to evaluate the effect of
exposure time on the incidence of mycosis (mortality) using two concentrations of fungal
conidia, 1 x 104 and 1 x 107 conidia ml-1. Viability of conidia was determined by spread plating,
as described in Section 3.2.2 and percentage germination was examined after 24 hrs from 100spore counts on each plate. The criteria for scoring mycosis were undertaken as described in
Section 3.2.3. Each treatment was replicated four times with 20 insects per replicate.
3.2.6 Statistical analyses
Percentage mortality data (Section 3.2.3) were corrected for control mortality (Abbott 1925) and
the data were normalized through Arcsine transformation and then subjected to analysis of
79
variance (ANOVA) followed by mean separation by the Student-Newman-Keul's test (P=0.05).
To investigate whether fungal isolates from orchards or natural habitats showed grouping in
terms of pathogenicity, this mortality data was normalized through Arcsine transformation and
subjected to ANOVA. Regression analysis was used (Section 3.2.4) to determine the functional
relationship between log concentration of inoculum and probit of mortality (pupae and adults
mycosis combined) using PROBAN (Van Ark 1995). PROBAN utilises a chi-squared (χ2) test to
determine the fit of a probit line and ascertains whether deviations from this line are
homogenous. The program also calculates a G-value which is used to calculate the amount of
response variation during bioassay procedures and in the calculation of the fiducial limits
Percentage mortality data from preliminary exposure time-response tests (Section 3.2.5) were
normalized through Arcsine transformation and then subjected to ANOVA.
3.3 RESULTS
In fungal conidial viability tests, germination for all isolates ranged from 88 to 95%.
Thaumatotibia leucotreta final instar larvae treated with EP fungi did not always pupate
normally because larvae were sometimes found mycosed (Fig. 3.3.1 A). For the purpose of
analyses, those mycosed larvae were grouped with mycosed pupae and considered as one
grouped life stage. Few T. leucotreta adults were found mycosed (Fig. 3.3.1 B) and for this
reason only adult emergence and not adult mycosis was recorded. In both fruit fly species, C.
capitata and C. rosa, all final instar larvae treated with fungi pupated normally but infection
established in puparia and emerging adults. Total adult survival was recorded for fruit flies; these
were recorded as total adult emergence minus mycosed adults.
3.3.1 The effects of EP fungi on Thaumatotibia leucotreta adult emergence and pupal
mycosis
Thirteen EP fungal isolates had a significant effect in reducing T. leucotreta adult emergence to
below 20%. Percentage adult emergence in the control treatment was 100%, and varied from 5 to
60% in fungal treated sand depending on fungal isolate (F=33.295; df=21; P=0.0001), with five
Beauveria isolates showing a marked reduction in adult emergence (Fig. 3.3.2).
80
Figure 3.3.1 (A) A Thaumatotibia leucotreta wandering larva infected with Metarhzium
anisopliae var. anisopliae (B) An adult T. leucotreta moth infected with Beauveria bassiana (all
images were taken with a Leica EZ 4D dissecting microscope at ~ X16) (Photo Credits, T.
Goble, Rhodes University).
Figure 3.3.2 Pathogenicity of isolates of Beauveria bassiana and Metarhizium anisopliae to
Thaumatotibia leucotreta: Mean (% ± SE) of adult emergence after treatment with a
concentration of 1 x 107 conidia ml-1. Bars with the same letter do not differ significantly by
SNK test (P= 0.05).
81
The percentage of pupae with visible signs of mycosis ranged from 21 to 93% depending on the
fungal isolate (F=96.436; df=21; P=0.0001) (Fig. 3.3.3). Four Beauveria isolates, FCM 10 13
L1, G B Ar 23 B3, G Moss R10 and G 14 2 B5 showed the greatest pathogenicity towards T.
leucotreta as both a decrease in adult emergence and an increase in percentage pupal mycosis
was observed (Fig. 3.3.3). The effect of fungal isolate habitat (either orchards or refugia) on the
incidence of mycosis was investigated to determine if isolates obtained from natural habitats
were more virulent towards T. leucotreta. Most orchard isolates seemed to show a greater
incidence of mycosis. However, there were no significant differences observed in the incidence
of mycosis by various fungal isolates grouped according to habitat (F (1,19)=1.63; P=0.21).
Figure 3.3.3 Pathogenicity of Beauveria bassiana and Metarhizium anisopliae to Thaumatotibia
leucotreta: Mean (% ± SE) of pupal mycosis after treatment with a concentration of 1 x 107
conidia ml-1. Bars with the same letter do not differ significantly by SNK test (P= 0.05).
82
3.3.2 The effects of EP fungi on Ceratitis rosa and C. capitata adult survival
When treating C. rosa and C. capitata final instar larvae to a fungal concentration of 1x107
conidia ml-1, percentage adult emergence in the control treatments were 88% and 86%
respectively. Percentage adult survival in C. rosa ranged from 30 to 90% depending on the
fungal isolate, while the percentage of C. capitata adults which survived various fungal
treatments at this concentration ranged from 55 to 86% (Fig. 3.3.4). Two Beauveria isolates: G
OL R11 and FCM 10 13 L1 had a significant effect on adult fruit fly survival compared to other
isolates (F=10.70; df=12; P=0.0001). However, there were no significant differences observed in
fruit fly species (C. rosa versus C. capitata) response to various fungal isolates (F=0.18; df=1;
P=0.66) (Fig. 3.3.4).
Figure 3.3.4 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis rosa (black bars) and C. capitata (grey bars): Mean (% ± SE) of fruit fly
adult survival after treatment with a concentration of 1 x 107 conidia ml-1. Bars with the same
letter do not differ significantly by SNK test (P= 0.05).
83
Compared to T. leucotreta, there was little effect on the reduction of fruit fly survival by fungal
isolates in the previous trial, thus bioassays at a higher dose concentration of 1 x 108 conidia ml-1,
using fungal isolates (four isolates) which performed well at the lower concentration (1 x 107
conidia ml-1), and by the addition of new fungal isolates (seven isolates) were initiated.
Percentage adult emergence in the control treatments were 87 and 89% for C. rosa and C.
capitata respectively (Fig. 3.3.5). Percentage adult survival in C. rosa differed significantly and
ranged from 47 to 85% depending on the fungal isolate (F=6.46; df=11; P=0.0001), while the
percentage of C. capitata adults which survived various fungal treatments also differed
significantly (F=7.74; df=11; P=0.0001) and ranged from 34 to 84% depending on the fungal
isolate (Fig. 3.3.5). Four Beauveria isolates: G 14 2 B5, G Moss R10, G B Ar 23 B3 and FCM 10
13 L1 had a significant effect on adult fruit fly survival compared to other isolates (F=14.03;
df=11; P=0.0001). However, there were no significant differences observed in fruit fly species
(C. rosa versus C. capitata) response to various fungal isolates (F=0.17; df=1; P=0.67) as both
fruit fly species showed similar susceptibilities to the fungal isolates (Fig. 3.3.5).
In Fig. 3.3.5, there appeared to be a greater reduction in adult fruit fly survival at a higher
conidial concentration. However, with the addition of new fungal isolates, these observed
reductions were not due to the increase in concentration, but rather from the addition of more
virulent fungal isolates. This was confirmed through the four fungal isolates (G14 2 B3, FCM
Ar23 B3, G OL R11 and FCM 10 13 L1), which were used in both bioassay trials at dose
concentrations of 1 x 107 and 1 x 108 conidia ml-1 being analysed for significant interactions
between concentration, insect species and fungal isolate. There was a weakly significant
difference observed between the two fungal concentrations (F=5.30; df=1; P=0.02), with the
weaker conidial concentration showing a greater reduction in fruit fly survival. However, there
was again no significant difference observed in the adult survival response to fungal isolates by
both fruit fly species (F=0.004; df=1; P=0.94), illustrating that both fruit fly species, C. rosa and
C. capitata, responded similarly to the effects of EP fungal isolates. The fungal isolates had a
significant effect on adult fruit fly survival with both Beauveria isolates, G OL R11 and FCM 10
13 L1, showing a marked decrease in adult fruit fly survival compared to the two Metarhizium
isolates G14 2 B3 and FCM Ar23 B3 (F=17.41; df=4; P=0.0001) (Fig. 3.3.5).
84
Figure 3.3.5 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis rosa (black bars) and C. capitata (grey bars): Mean (% ± SE) of fruit fly
adult survival after treatment with a concentration of 1 x 108 conidia ml-1. Bars with the same
letter do not differ significantly by SNK test (P= 0.05).
3.3.3 The effects of EP fungi on Ceratitis rosa and C. capitata adult and pupal mycosis
When treating C. rosa final instar larvae with a fungal concentration of 1 x 107 conidia ml-1, the
percentage of puparia with visible signs of mycosis ranged from 1 to 14% depending on the
fungal isolate (F=4.37; df=12; P=0.0001) (Fig. 3.3.6 and Fig. 3.3.10). The Metarhizium isolate,
FCM Ar 23 B3, was the most pathogenic towards C. rosa puparia at this concentration. Deferred
mortality due to mycosis in C. rosa adult flies ranged from 1 to 58% depending on the fungal
isolate (F=27.01; df=12; P=0.0001) (Fig. 3.3.6), with two of the Beauveria isolates, G OL R11
and FCM 10 13 L1, showing the greatest pathogenicity towards C. rosa adults. Natural mortality
in the untreated controls never exceeded 10%. EP fungal isolates had a significantly greater
effect on the adults of C. rosa than they did on the puparia (F(1,12)=13.11; P=0.0001) (Fig. 3.3.6).
Beauveria isolates, G OL R11, FCM 10 13 L1, G Moss R12, FCM Rose R9, FF J&B R5 and
85
FCM 11 3 B2 showed a noticeable increase in deferred adult mortality compared to the
Metarhizium isolates, FCM Ar 23 B3, G Moss R2, G 11 3 L6, G OL R3 and G OL R8.
Figure 3.3.6 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis rosa: Mean (% ± SE) of Ceratitis rosa adult (black bars) and pupal (grey
bars) mycosis after treatment with a concentration of 1 x 107 conidia ml-1. Bars with the same
letter do not differ significantly by SNK test (P= 0.05). Note: Y-axis goes to 70%.
A subsequent bioassay, at a higher dose concentration of 1 x 108 conidia ml-1, was initiated using
four fungal isolates which performed well at the lower concentration (1 x 107 conidia ml-1) and
by the addition of seven new fungal isolates. The percentage of C. rosa puparia with visible
signs of mycosis ranged from 1 to 25% (F=11.04; df=11; P=0.0001) (Fig. 3.3.7). One Beauveria
isolate, G 14 2 B5, showed the greatest pathogenicity towards C. rosa puparia. Deferred
mortality in C. rosa adult flies ranged from 3 to 32% depending on the fungal isolate (F=13.53;
86
df=11; P=0.0001) (Fig. 3.3.7), with two Beauveria isolates, G B Ar 23 B23 and FCM 10 13 L1,
showing the greatest pathogenicity towards C. rosa adults. Natural mortality in the controls
never exceeded 10%. Fungal isolate had a significant effect on pupal and adult fruit fly mycosis
(F(1,11)=12.55; P=0.0001) with a greater percentage mycosis occurring in adults than puparia
(Fig. 3.3.7). Four Beauveria isolates, G14 2 B5, G B Ar23 B3, FCM 10 13 L1 and G Moss R10,
showed the greatest potential as possible biological control agents as a significantly marked
increase in adult and pupal mycosis was observed with these isolates.
Figure 3.3.7 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis rosa: Mean (% ± SE) of Ceratitis rosa adult (black bars) and pupal (grey
bars) mycosis after treatment with a concentration of 1 x 108 conidia ml-1. Bars with the same
letter do not differ significantly by SNK test (P= 0.05). Note: Y-axis goes to 40%.
87
When treating C. capitata final instar larvae with a fungal concentration of 1 x 107 conidia ml-1,
the percentage of puparia with visible signs of mycosis ranged from 1 to 11% depending on
fungal isolate (F=3.76; df=12; P=0.0005) (Fig. 3.3.8). The Metarhizium isolate, FCM Ar23 B3,
and the Beauveria isolate, G OL R11, were the most pathogenic to the puparia of C. capitata.
Deferred mortality in C. capitata adult flies ranged from 1 to 33% depending on fungal isolate
(F=10.95; df=12; P=0.0001) (Fig. 3.3.8), with the same two Beauveria isolates, G OL R11 and
FCM 10 13 L1, exhibiting the greatest pathogenicity towards C. capitata adults, as was seen with
C. rosa adults at the same concentration. Natural mortality in the untreated controls never
exceeded 10%. EP fungal isolates had a significantly greater effect on C. capitata adults than on
the puparia (F(1,12)=3.17; P=0.0005) (Fig. 3.3.8).
Figure 3.3.8 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis capitata: Mean (% ± SE) of Ceratitis capitata adult (black bars) and
pupal (grey bars) mycosis after treatment with a concentration of 1 x 107 conidia ml-1. Bars with
the same letter do not differ significantly by SNK test (P= 0.05).
88
In a subsequent bioassay using a higher concentration of 1 x 108 conidia ml-1, four fungal
isolates which performed well at the lower concentration (1 x 107 conidia ml-1) were included and
seven new fungal isolates were also incorporated. The percentage of C. capitata puparia with
visible signs of mycosis ranged from 1 to 10% (F=2.68; df=11; P=0.01) (Fig. 3.3.9). One
Beauveria isolate, G Moss R10, showed the greatest pathogenicity to C. capitata puparia in this
bioassay. Deferred mortality in C. capitata adults ranged from 4 to 41% (F=8.31; df=11;
P=0.0001) depending on the fungal isolate (Fig. 3.3.9), with two Beauveria isolates, G Moss R10
and G 14 2 B5, showing the greatest pathogenicity towards C. capitata adults. Natural mortality
in the untreated controls never exceeded 10%. When the interaction of life stage (pupal and adult
mortality) and isolate were analysed, there was a significant difference observed with more
adults than puparia exhibiting mycosis (F(1,11)=4.19; P=0.0008) (Fig. 3.3.9).
Figure 3.3.9 Pathogenicity of indigenous fungal isolates of Beauveria bassiana and Metarhizium
anisopliae to Ceratitis capitata: Mean (% ± SE) of Ceratitis capitata adult (black bars) and
pupal (grey bars) mycosis after treatment with a concentration of 1 x 108 conidia ml-1. Bars with
the same letter do not differ significantly by SNK test (P= 0.05).
89
The four fungal isolates which were used in both pupal and adult mycosis bioassay, against both
fruit fly species, at dose concentrations of 1 x 107 and 1 x 108 conidia ml-1 were analysed to see if
there were any significant interactions between dose concentrations, insect species and fungal
isolates. A weakly significant difference was observed between the two conidial concentrations
used (F=4.03; df=1; P=0.04), with the lower concentration showing a greater increase in adult
and pupal mycosis. The interaction of concentration, insect species and stage was not significant
(F=0.27; df=3; P=0.59) despite the fact that higher deferred adult mortality was observed than
pupal mortality. There were no significant differences observed between C. rosa and C. capitata
(F=0.003; df=1; P=0.95), illustrating again that both fruit fly species responded similarly to the
effects of EP fungi.
The effect of isolate habitat (either orchards or refugia) on the incidence of mycosis (both adult
and pupal) was investigated to determine if isolates obtained from natural habitats maybe more
pathogenic towards fruit flies. At a conidial concentration of 1 x 107 conidia ml-1, there were no
significant differences observed in the incidence of mycosis by various fungal isolates grouped
according to habitat (F(1, 20)=0.02; P=0.87). At a conidial concentration of 1 x 108 conidia ml-1,
there were also no significant differences observed in the incidence of mycosis by various fungal
isolates grouped according to habitat (F(1, 18)=0.07; P=0.78).
Figure 3.3.10 (A) Ceratitis capitata pupa infected with Metarhizium flavoviride (B) Ceratitis
rosa adult infected with Beauveria bassiana (C) C. capitata infected with M. anisopliae (all
images were taken with a Leica EZ 4D dissecting microscope at ~ X16) (Photos Credits: T.
Goble, Rhodes University).
90
3.3.4 The effect of fungal concentration on fruit fly mortality
In fungal viability tests, germination of conidia for Beauveria isolates, G 14 2 B5, G Moss R10,
G B Ar 23 B3, FCM 10 13 L1 and Bb Plus® ranged from 80 to 92%. The relative potencies of the
five isolates were compared against one another as log-probit regressions of mortality against C.
rosa and C. capitata (Fig. 3.3.11). For C. rosa, when the five fungal regression lines were
compared against one another a positive relationship between increased conidial concentration
and fruit fly mortality was observed (Fig. 3.3.11). When log-probit regression lines of mortality
were compared, Bartlett‟s test found residual variances to be homogenous (χ2=0.38; df=4;
P=0.98) and the slopes of the lines were therefore comparable. The slopes of the lines were
parallel (χ2=0.25; df=4; P=0.99) and their elevations were significantly different (F(4,14)=5.23;
P=0.009), but when multiple comparisons of the elevations and slopes were undertaken using
Bonferroni‟s method (P=0.005), no significant differences were observed between the five
fungal isolates (Fig. 3.3.11).
The information regarding each fungal isolate probit line and the estimated LC50 and LC90 values
for each fungal isolate tested against C. rosa are shown in Table 3.3.1. The estimated LC50
values of isolates G B Ar 23 B3 and G 14 2 B5 were lower than isolates G Moss R10; Bb Plus®
and FCM 10 13 L1, indicating greater potency of the prior isolates. However, the slopes of the
latter three isolates were steeper than both the former isolates indicating that target insects may
be more vulnerable over a given time period to increasing concentrations of conidia.
With C. capitata, residual variances of the five fungal regression lines (Fig. 3.3.11) were
homogenous (χ2=0.13; df=4; P=0.99) and the slopes of the lines were compared. Slopes were
parallel (χ2=0.85; df=4; P=0.93) and the elevations were compared. The elevations of the lines
were significantly different (F(4,14)=5.72; P=0.006). However, when multiple comparisons of the
elevations and slopes of the lines were compared (P=0.005), no significant differences were
observed between the five isolates (Fig. 3.3.11). Despite there being no significant differences
between fungal isolates, there were differences observed in the LC50 and LC90 values among the
five isolates (Table 3.3.1) with G B Ar 23 B3 and G 14 2 B5 showing lower estimated LC50
values than those of Bb Plus®; G Moss R10 and FCM 10 13 L1. Bb Plus® and G B Ar 23 B3 had
steeper slopes than the other isolates, indicating stronger concentration responses by the test
insects.
91
Figure 3.3.11 The log-probit regressions of mortality (pupal and adult combined) caused by five
isolates of Beauveria bassiana against Ceratitis rosa and C. capitata.
92
Table 3.3.1 The calculated probit data of each of the five fungal isolates tested against Ceratitis rosa, C. capitata and Thaumatotibia
leucotreta. Highlighted areas indicate isolates which were considered the most pathogenic by relative potency comparison tests using
PROBAN.
Insect spp
Isolate
Fit of the Line
χ2
C. rosa
C. capitata
G B Ar 23 B3
G 14 2 B5
G Moss R10
Bb Plus®
R- Coefficient
Fiducial
a
b
G
Lethal Concentration
LC50
LC50 (SE)
11
LC90
LC90 (SE)
15
FCM 10 13 L1
1.65
1.24
0.25
1.57
0.66
0.44
0.53
0.87
0.45
0.72
0.96
0.72
0.46
0.23
0.10
0.40
0.42
0.45
0.45
0.46
0.17
0.17
0.17
0.21
0.22
7.3 X 10
8.9 X 1011
1.0 X 1012
2.6 X 1012
2.6 X 1012
5.2 ±
6.3 ±
7.2 ±
2.4 ±
2.3 ±
1.0 X 10
8.6 X 1014
7.0 X 1014
1.7 X 1015
1.5 X 1015
2.1 ±
1.7 ±
1.3 ±
4.0 ±
3.2 ±
G B Ar 23 B3
0.38
0.82
0.49
0.46
0.14
5.5 X 1011
3.3 ±
3.3 X 1014
5.6 ±
12
15
G 14 2 B5
G Moss R10
Bb Plus®
T. leucotreta
P
Intercept
FCM 10 13 L1
0.40
0.50
1.01
0.70
0.81
0.78
0.60
0.70
1.16
0.78
0.03
0.20
0.37
0.41
0.48
0.45
0.21
0.20
0.18
0.22
1.3 X 10
1.5 X 1012
1.3 X 1012
2.8 X 1012
1.1 ±
1.2 ±
9.9 ±
2.6 ±
3.2 X 10
1.9 X 1015
5.7 X 1014
1.7 X 1015
8.4 ±
4.4 ±
1.0 ±
4.2 ±
G B Ar 23 B3
G 14 2 B5
0.73
1.86
0.69
0.39
3.26
3.17
0.33
0.46
0.18
0.11
1.6 X 107
8.8 X 105
9.4 ±
5.9 ±
1.1 X 1011
5.2 X 108
2.2 ±
3.8 ±
G Moss R10
Bb Plus®
FCM 10 13 L1
2.44
0.26
0.64
0.29
0.87
0.73
3.19
3.50
2.93
0.47
0.34
0.37
0.12
0.17
0.15
6.8 X 105
2.1 X 106
3.7 X 107
4.8 ±
1.6 ±
1.8 ±
3.6 X 108
1.1 X 1010
1.0 X 1011
2.5 ±
1.5 ±
1.8 ±
93
Both the LC50 and LC90 values of all five fungal isolates tested against C. rosa and C. capitata
however fall outside of the tested concentration range (which was 1 x 107 to 1 x 1010 conidia/ml1
), because the number of exposed insects at the highest conidial concentration only just
exceeded 50% of the total response by all insects. Furthermore, there were large G-values
observed for each isolate, indicating a wide variation in response by test insects (Table 3.3.1).
Bonferroni‟s test showed no significant differences among the five isolates tested against fruit
flies, indicating poor resolution between the relative potencies of these fungal isolates. LC50
values were higher than the highest concentration used in this study and the further use of
increased conidial concentrations is impractical and unaffordable when translated into field
required inoculums levels.
3.3.5 The effect of fungal concentration on Thaumatotibia leucotreta mortality
The concentration-mortality response of T. leucotreta to the five isolates was far greater than that
observed for fruit flies (indicating a more susceptible insect species to the effects of these
indigenous isolates of fungi). There was a positive relationship between increased conidial
concentration and T. leucotreta mortality. Fig. 3.3.12 illustrates the plots of the probit regression
estimates against concentration for the five fungal isolates tested against T. leucotreta. When the
regression lines were compared, residual variances were homogenous (χ2=0.568; df=4; P=0.96)
and the slopes of the lines were compared. The slopes of the lines were found to be parallel
(χ2=2.821; df=4; P=0.59) and the elevations were compared. The elevations of the lines were
significantly different (F(4,14)=28.11; P=0.0001) and when multiple comparisons of the elevation
and the slopes of the lines was undertaken using Bonferroni‟s test, mean separation (P=0.005)
showed that the isolates G Moss R10, G 14 2 B5 and Bb Plus® were significantly different from
G B Ar 23 B3 and FCM 10 13 L1. These differences were observed in the relative LC50 and LC90
values of each isolate are shown in Table 3.3.1. The slopes of the regression lines for isolates: G
Moss R10 and G 14 2 B5 were steep, indicating a stronger concentration response to these
isolates, than to Bb Plus®. A weak concentration response was observed against isolates: G B Ar
23 B3 and FCM 10 13 L1 (Fig. 3.3.12). The LC50 values of all isolates fell within the tested
concentration range (which was 1 x 104 to 1 x 107 conidia/ml-1) and the accuracy of these values
remains intact. Only the LC90 values fell outside of the observed concentration range.
94
Figure 3.3.12 The log-probit regressions of mortality (pupal and adult combined) caused by five
isolates of Beauveria bassiana against Thaumatotibia leucotreta.
3.3.6 Exposure time-mortality relationship study between EP fungi and Thaumatotibia
leucotreta
In fungal viability tests, germination of conidia for three Beauveria isolates, Bb Plus®, G Moss
R10 and G 14 2 B5 ranged from 82, 90 and 92% respectively. Exposure time-response trials of T.
leucotreta final instar larvae, exposed to two conidial concentrations, at four different exposure
times, showed a positive relationship. An increase in total percentage mycosis was observed with
an increase in exposure time (F=5.43; df=3; P=0.001) (Fig. 3.3.13). At a higher concentration (1
x 107conidia/ml-1) and shorter exposure time (12 hrs), percentage mycosis in T. leucotreta ranged
from 22.7 to 46.2%, while at the longest exposure time (96 hrs), mortality due to mycosis ranged
from 46 to 56.16% depending on the fungal isolate. At a lower concentration (1 x 104 conidia/ml1
) across all time exposures, percentage mycosis ranged from 12.53 to 27% (Fig. 3.3.13). There
was a significant difference observed in percentage mycosis between the two conidial
concentrations, with the higher concentration resulting in a greater percentage mycosis (F=74.13;
95
df=1; P=0.0001). A significant difference was observed between the isolates at a higher
concentration (F=3.51; df=2; P=0.03), with G Moss R10 and G 14 2 B5 showing greater
percentage mycosis of pupae over time compared to Bb Plus®. At the higher concentration,
isolates G Moss R10 and G 14 2 B5 elicited a response in 50% of test insects at 72 hrs (3 days)
exposure (Fig. 3.3.13). No significant differences were observed between isolates when they
were compared against one another at a lower concentration (F=1.33; df=2; P=0.27).
An outlier was observed in that a large percentage of pupae exposed to isolate G Moss R10 were
mycosed at 12 hrs. The interaction of fungal isolate over time was not significant because of the
positive trend observed in mycosis with increased exposure time for all isolates. A significant
increase in adult moth emergence was observed at 1 x 104 conidia/ml-1compared to the higher
concentration (F=49.38; df=1; P=0.0001), for which many more pupae were noted. These results
are considered preliminary and could be used as a guideline for future work regarding exposure
time-response in T. leucotreta using these fungal isolates.
Figure 3.3.13 Thaumatotibia leucotreta total adult and pupal mycosis after final instar larvae
were exposed to three fungal isolates at two concentrations, for 12, 48, 72 and 96 hrs time
periods.
96
3.4 DISCUSSION
3.4.1 The effects of EP fungi on fruit flies, Ceratitis rosa and C. capitata
Twenty-one indigenous isolates of EP fungi were screened at conidial concentrations of 1 x 107
and 1 x 108 conidia/ml-1. Five Beauveria bassiana isolates, G Moss R10, GB Ar 23 B3, FCM 10
13 L1, G14 2 B5 and G OL R11, were able to reduce C. capitata and C. rosa adult survival to
below 60%, however adult survival was generally high for most other EP isolates (greater than
80%). De la Rosa et al. (2002) also showed high adult emergence when Mexican fruit fly,
Anastrepha ludens larvae were dipped into B. bassiana conidial suspensions of 1.6 x 108
conidia/ml-1 and then allowed to pupae on sand, from which between 92 to 98% adult emergence
occurred. Similarly, the effects of Metarhizium anisopliae against the western cherry fruit fly,
Rhagoletis indifferens, were determined when adults were exposed to fungal spores mixed into
soil at concentrations ranging from 7.6 x 105 to 1.6 x 106 conidia/gram. The results revealed that
adult emergence was not reduced (Yee & Lacey 2005). During fruit fly adult survival trials using
1 x 108 conidia/ml-1 in the present study, there appeared to be a greater reduction in the rate of
survival. This was the result of the additional screening of more virulent isolates. The isolates, G
OL R11 and FCM 10 13 L1, which showed a marked decrease in adult fruit fly survival at 1 x
107 conidia/ml-1, did not show the same resilience during bioassays conducted at a higher
concentration (1 x 108 conidia/ml-1). This was confirmed when analysing the four fungal isolates,
G14 2 B3, FCM Ar23 B3, G OL R11 and FCM 10 13 L1, which were used in both bioassay trials
at the abovementioned conidial concentrations. A weakly significant difference was observed
between the two fungal concentrations, with the weaker conidial concentration showing a greater
reduction of adult fruit fly survival. This trend was further observed during pupal and adult
mycosis trials against C. capitata and C. rosa, when G OL R11 and FCM 10 13 L1 elicited a
greater amount of mycosis during bioassays at lower concentrations than subsequent assays at
higher conidial concentrations. These variations were as a result of the marked increase in
deferred adult mortality which subsequently affected the survival rate and percentage adult and
pupal mycosis rate of fruit flies against these particular fungal isolates. A higher incidence of
deferred adult mortality at lower concentrations may have been as a result of any of the three
fungal inoculum transfer modes which are discussed in more detail below.
Indigenous isolates were not able to induce high mortality in C. capitata and C. rosa, puparia
when they were exposed to sand treated with conidia suspensions relative to other studies
(Castillo et al. 2000; Ekesi et al. 2002). During the initial fungal virulence screening trials, pupal
97
mycosis of both fruit fly species never exceeded 25% by any given fungal isolate, even at
concentrations of 1 x 108 conidia/ml-1. Toledo et al. (2006), in an investigation of various
Beauveria bassiana and Metarhizium anisopliae isolates and formulated products against
Anastrepha ludens and C. capitata, showed that EP fungi had no effect on the larvae or pupae of
these fruit fly species but that all isolates were highly virulent to the adults. Similarly, De la Rosa
et al. (2002) reported that the mortality caused by eight EP fungal isolates at a concentration of
1.6 x 108 conidia/ml-1 applied by dipping the immature stages of Mexican fruit fly, A. ludens,
was low, 2 to 8% in larvae and 0% in pupae, however very high levels of adult mortality were
noted against three Beauveria isolates causing 98 to 100% mycosis. In the evaluation of
susceptibility of Delia antique (Anthomyiidae) to eleven isolates of hyphomycetes, pupae were
generally found not to be susceptible to mycosis (Poprawski et al. 1985).
The low incidence of mycosis observed on fruit fly puparia may be attributed in part to the
introduction of wet conidial suspensions which caused clumping of the sand and may have
reduced the even distribution of conidia in the soil. Thus, as third instar larvae burrow into the
soil they could have missed the fungal inoculum. Jaronski (2007) suggested that soil pore
structure is an important consideration during the application of EP fungi to soils because as soils
are hydrated, increasingly larger pores become filled with water. Insects burrowing down into
the soil will follow the path of least resistance and thus favour the larger pore spaces, whereas
fungal conidia are more commonly contained within the smaller pore sizes, thus migrating
insects may not have come into contact with enough conidia (Jaronski 2007). The fact that river
sand was used, which is finer in texture, limits these larger soil pore spaces. However, it does not
negate conidia from becoming buried within soil clumps. Some authors have reported that if
soils are too moist, this may negatively affect conidial efficacy by reducing sporulation because
of lower oxygen concentrations (Ekesi et al. 2003). However, because no measurement of soil
moisture was recorded these suggestions should be considered lightly. Further, because the
number of mycosed puparia in the soil was low and infected puparia represent a site of infection,
the ability to infect other surrounding puparia in the soil, by increasing inoculum levels, may
have been reduced. Limited mycosis in puparia was only noted with fruit flies and was not
apparent with T. leucotreta pre-pupating larvae; these observations could possibly be due to the
relatively small sizes of fruit fly larvae (7 to 9 mm in length) (Thomas et al. 2001), compared to
T. leucotreta larvae (12 to 15 mm in length) (Stibick 2008), which limits body contact with
conidia as insect larvae burrow into the sand. It could also mean that the immature stages of fruit
98
flies are simply less susceptible to EP fungi than the adults (Poprawski et al. 1985; De la Rosa et
al. 2002; Toledo et al. 2006).
Deferred adult mortality due to mycosis on the other hand, reached 58% during one screening
trial. Deferred adult mortality occurs when immature insect stages are not affected by EP fungi
but as they emerge as adults so the cuticle is soft and they become susceptible to fungal
infection. It was observed consistently that mycosis due to deferred adult mortality was
significantly higher for most fungal isolates than mycosis of fruit fly puparia. During the
evaluation of SDA plates at the end of the 10 day bioassay trial period, more adults were
observed with fungal infection than puparia. Yee & Lacey (2005) reported that during the
evaluation of M. anisopliae against R. indifferens, adult emergence was not reduced however
14.9 to 68% of emerging adults were infected at death. Poprawski et al. (1985) noted that
deferred mortality was observed in D. antique adults which emerged after treatment with various
EP fungal species. Ekesi et al. (2002) reported that 11 of their 15 Metarhizium and Beauveria
isolates showed greater deferred adult mortality in C. capitata (due to mycosis) than pupal
mycosis although pupal mycosis was also very high (10 of the isolates were able to induce over
60% pupal mycosis). A couple of reasons could explain this observation. Firstly, when third
instar fruit fly larvae enter the fungal treated sand they pupate within less than 12 hrs of entering
(personal observation) and the puparia may become sclerotised before the fungal conidia can
germinate and penetrate the pupal case. Ekesi et al. (2002) showed that mortality of puparia
decreased with an increase in pupal age, as it was found that older pupae (96 hrs old) had a lower
level of fungal infection due to the sclerotization of the pupal case. Thus, the spores remain on
the sclerotised fruit fly pupae until emergence begins. Upon emergence, adult fruit flies break
open the pupal case and adult flies contort their bodies to rid themselves of the exuviae. At this
time the cuticle of the newly emerged adults may come into contact with the fungal conidia,
which then germinate and cause insect death (Ekesi et al. 2002).
Secondly, when adult fruit flies were infected with fungi as a result of deferred adult mortality,
then died in the emergence chambers early on in the 10 day bioassay trial period, they became a
source of infection and caused healthy adults to become infected simply by contact. In a study by
Dimbi et al. (2003a), the age of the host, in this case, C. capitata, had a significant effect on
mortality, with younger flies (0 to 7 days old) succumbing to infection by EP fungi earlier than
older flies (7 to 14 days). If the chance incidence of young fruit flies with deferred fungal
infection is experienced early on in the bioassay trial period, the incidence of mycosis in adult
99
flies could become higher because fungal inocula levels are increased in the emergence
chambers. This may explain to some degree why adult mycosis was high for some Beauveria
isolates during the investigation of puparia and adult mycosis of C. capitata and C. rosa at lower
conidial concentrations, which showed a slight significant difference from higher conidial
concentrations. Perhaps future research should evaluate the use of plastic bottles as emergence
chambers to reduce confounding factors, such as limited space, which may have increased
mycosis scores during bioassay procedures. In the evaluation of EP fungi against fruit flies,
Ekesi et al. (2002) placed Petri dishes containing untreated sand and puparia in Plexi-glass cages
which were 15 x 15 x 20 cm in size and therefore not much bigger than the sterilised plastic
bottles which were used in this study. Despite this, it is still suggested that this aspect be
investigated to optimise bioassay procedures and reduce variation.
Another possible reason for the observed increase in adult mortality due to mycosis compared to
pupal mortality could be horizontal transmission. During the 10 day bioassay trial period, adult
fruit flies of both species were observed mating and eggs were also found on the cotton wool
plugs which were used to introduce water into the emergence chambers. Horizontal transmission
of pathogens within the same target species is called autodissemination (Quesada-Moraga et al.
2008) and occurs when fungal infected individual flies mate with healthy flies. Whilst evaluating
the SDA plates after the trials had terminated, it was found that both male and female C. rosa
and C. capitata adults had EP fungi growing on the tip of the abdomen and on the thoracic
sclerites (Fig. 3.4.1). Quesada-Moraga et al. (2008) reported that when uninfected C. capitata
females were paired with Metarhizium inoculated males for 24 hrs, these females showed a
higher density of conidia in the dorsal thoracic scelrites. Conidia were also observed on the
sclerites and inter-segmental regions of the dorsal, lateral and ventral areas of the abdomen.
Inoculated males exhibited conidia on the aedeagi and females had conidia on the ovipositor.
Quesada-Moraga et al. (2008) reported that in Spain, wild C. capitata females will first oviposit
four to seven days after emergence. Thomas et al. (2001) reported that C. capitata females were
able to oviposit as early as four to five days after emergence. This was also observed in this
study for both fruit fly species. Horizontal transmission is more effective when infected male
flies mate with healthy females because female C. capitata fruit flies are monogamous or
oligogamous (Quesada-Moraga et al. 2008). In addition to the possible transmission of EP fungi
between male and female fruit flies, calling male flies also show the tendency of numerous
attempts at homosexual mating (Ekesi et al. 2007). These male to male mounting attempts as
well as heterosexual fruit fly mating could almost certainly have increased horizontal
100
transmission of EP fungi in adults in the emergence chambers during bioassay procedures.
Horizontal transmission of EP fungi in the laboratory has also been noted in diamondback moth,
Plutella xylostella (Furlong & Pell 2001) and in the Malaria vector, Anopheles gambiae (Scholte
et al. 2004). This was not the focus of this study however, but would be worthwhile investigating
in greater detail in future research, particularly in light of recent literature which lists horizontal
transmission as an important mechanism for EP fungal inoculum transfer in nature (Ekesi et al.
2007; Dimbi et al. 2009).
Figure 3.4.1 (A) Female adult C. rosa fruit fly showing B. bassiana growing on the abdominal
segments (B) Female fly showing B. bassiana growth on the dorsal and lateral thoracic scelrites
(C) Male adult C. rosa fruit fly showing B. bassiana growing on the abdominal segments (all
images were taken with a Leica EZ 4D dissecting microscope at ~ X16) (Photo Credits: T.
Goble, Rhodes University).
A high level of sporulation was observed for some fungal isolates, particularly some of the
Beauveria isolates (personal observation) and sporulation levels increase the amount of conidia
which are produced from cadavers. This increases the chances of other individuals becoming
infected due to the higher level of inoculum present in the emergence chambers. The level of
sporulation is considered an important function for selected biological control agents. As
contaminated pupae or adults represent a concentrated source of infection and ensures that
inoculum levels and contact chances between conidia and healthy individuals are increased
which subsequently ensures disease transmission (Ekesi et al. 2002). Apart from two Beauveria
isolates, FCM 10 13 L1 and G OL R11, which induced 58% and 41% mycosis respectively in
adult fruit flies, other virulent isolates showing a greater increase in adult mycosis were G Moss
R10, G 14 2 B5 and G Ar 17 B3. It is concluded that although a limited amount of mycosis was
observed in puparia of both fruit fly species, deferred adult mortality due to mycosis was
sufficiently high enough to warrant further investigation, particularly of the abovementioned
101
isolates. The increased incidence of deferred adult mortality suggests that post emergence
mycosis in adult fruit flies may play a more significant role in field suppression than the control
of fruit flies at the pupal stage.
During the concentration-response regression analyses of fruit flies, there was very little
delineation among fungal isolates in terms of virulence, and isolates did not vary significantly in
their effect against C. capitata and C. rosa. The results of the regression analyses however
suggest that these indigenous fungal isolates, which were considered some of the most virulent,
require a large amount of inoculum (LC50 values range from 1 x 1011 to 1 x 1015conidia/ml -1) to
elicit a meaningful response in fruit flies. Both the LC50 and LC90 values of all five fungal
isolates tested against C. rosa and C. capitata fell outside of the tested concentration range
because the number of exposed insects at the highest conidial concentration only just exceeded
50% of the total response by all insects. This unfortunately was due to the difficulty in obtaining
large enough volumes of conidial suspension to use concentrations of 1 x 1011 and 1 x 1012
conidia/ml-1 in the laboratory. LC50 values were higher than the highest concentration used in this
study and the further use of increased conidial concentrations is impractical and unaffordable
when translated in field required inoculum levels. Ekesi et al. (2002) reported high mortality in
C. capitata puparia and a significant reduction in adult emergence using conidial sand
incorporation bioassays, and also found that the fungal isolates used to control fruit fly puparia
had median lethal concentration (LC50) values of between: 1.7 x 105 to 7.7 x 106 conidia/ml,
depending on which fruit fly species the isolates were tested against. Ekesi et al. (2007) also
reported that the pathogenicity of several isolates of EP fungi was demonstrated against
Bactrocera invadens final instar larvae when exposed to soil treated with 1 x 109 conidia/ml-1,
reducing adult emergence to below 8 and 10%. Other typical LC50 values of the most virulent
fungal isolates used against adult fruit flies include 4.9 x 105 to 2.0 x 106 conidia/ml-1. However,
these authors sprayed fruit flies with various conidial suspensions instead of applying the fungal
inoculum to the soil (Quesada-Moraga et al. 2006). These studies reiterate the poor pathogenicity
observed in the effects of our indigenous isolates of EP fungi against the immature life stages
fruit flies, C. capitata and C. rosa.
Both C. capitata and C. rosa responded in a similar way to the effects of indigenous isolates of
fungi as there were no significant differences observed between fruit fly species response when
adult survival, or adult and pupal mycosis were compared. There were also limited observed
differences in fruit fly response to various concentrations of selected isolates during regression
102
analyses. This is of great importance when considering a biological control agent against these
fruit fly species because they occupy similar ecological niches in the warmer areas of southern
Africa (de Meyer et al. 2008). Ekesi et al. (2002) suggested that an isolate with a broader range
of activity would be a more desirable trait than an isolate which was species-specific. This was
the concept which prompted the selection of fungal isolates for further analyses which may not
necessarily have been the most pathogenic towards T. leucotreta alone but were most virulent
towards all three insect species.
Indigenous EP fungal isolates were collected from orchard and natural habitats and it was of
interest to ascertain whether grouping isolates from these habitats would reveal a pattern in
pathogenicity. For example, isolates from natural habitats may prove to be more pathogenic
towards citrus pests because there may be a greater diversity and abundance of insect hosts
present in these natural habitats (Meyling et al. 2009). Isolates originating from these natural
habitats may therefore have broader host ranges and greater virulence. Meyling et al. (2009)
concluded that natural habitats offered structural diversity to groups of organisms (fungi and
their insect hosts) and were important reservoirs for fungal species diversity compared to
cultivated fields. Thus, an increase in species diversity may translate into a greater probability of
encountering virulent fungal isolates. Conversely, attributes of fungal isolates obtained from
orchard habitats may include pesticide and UV-B tolerances or exhibit resilience towards
agronomic practices. However, no significant differences were observed in the incidence of
mycosis by various fungal isolates grouped according to habitat. This result is valid since
increasing evidence suggests that habitat selection and not insect host selection drives the
population structures of EP fungi (Bidochka et al. 2001; Bidochka et al. 2002) (cf Chapter II)
and virulence is linked to insect host susceptibility. Fungal species with broad host ranges are
generally facultative pathogens and relatively less pathogenic compared to highly specific,
generally obligatory EP fungi (Goettel 1995).
Konstantopoulou & Mazomenos (2005) demonstrated that both Bactrocera oleae and Ceratitis
capitata adult fruit flies showed different degrees of susceptibility to two EP fungal species,
Beauveria bassiana and B. brongniartii. Dimbi et al. (2003b) reported considerable variation in
virulence among fungal isolates tested against C. capitata, C. rosa var. fasciventris and C.
cosyra. If it is possible for fruit flies species to vary so much in susceptibility to EP fungi and
fungi to exhibit variations in virulence, then it would stand to reason that a moth species, such as
T. leucotreta, may also vary in its susceptibility to specific fungal isolates. Despite fruit flies
103
exhibiting greater resistance to the effects of indigenous isolates of EP fungi, these isolates were
found to be highly pathogenic towards T. leucotreta.
3.4.2 The effects of EP fungi on false codling moth, Thaumatotibia leucotreta
Most of the indigenous isolates of EP fungi which were tested in this study were able to induce
high mortality in T. leucotreta final instar larvae and pupae when exposed to sand treated with
conidia suspension. The overall effect of this subterranean treatment led to a significant
reduction in adult moth emergence (less than 20% emergence was achieved with most EP fungal
isolates). Eleven of the initial twenty-one fungal isolates tested showed marked pathogenicity
towards the pupae (mycosis greater than 80%) of this insect species. Loc & Chi (2007) reported
that B. bassiana, M. anisopliae and two commercially available biopesticides (Crymax® 35 WP
and Atabron® 5EC) applied at concentrations of 1 x 107 conidia/ml-1 were able to induce 70 to
87.3% mortality in diamondback moth, Plutella xylostella, larvae seven days after exposure
when applied directly as sprays. Tefera & Pringle (2003) reported high mortality (98 to 100%) in
the stem borer, Chilo partellus, after exposing second instar larvae to Beauveria bassiana at
concentrations of 1.25 x 107 conidia/ml. The authors concluded that the development of B.
bassiana on cadavers of C. partellus was influenced by temperature, conidial concentrations and
exposure methods.
Concentration-response trials against T. leucotreta revealed a dose-dependant relationship and
two selected isolates, G 14 2 B5 and G Moss R10 could induce mycosis in 50% of the test
population (LC50) at concentrations of 8.8 x 105 and 6.8 x 105 conidia/ml-1 respectively. The
steeper slopes of these two isolates also indicated that target insects will be more vulnerable over
a given time period to increasing concentrations of conidia. Anand et al. (2009) reported that EP
fungi applied to Spodoptera litura pupae in the soil, as a drench (conidial suspensions at a titre of
108 conidia/g of soil) reduced adult emergence by 81.3% and 72.5% depending on the fungal
isolate. However, when premixing the conidial suspensions into the soil, fewer pupae were killed
by the fungi; this was attributed to the accumulation of conidia up to 10 cm from the soil surface
during drenching. The authors reported LC50 and LC90 values of M. anisopliae (the most virulent
isolate) as 1.2 x 107 and 4.3 x 108 conidia/g respectively. In the investigation of EP fungi against
Thaumetopoea pityocampa (Thaumatopoeidae) at concentrations of 1 x 105 conidia/ml-1, all
tested isolates resulted in various levels of mortality (16 to 100%) and when dose-mortality tests
104
were undertaken by spraying five conidial concentrations on larvae, the estimated LC50 value
was 3.4 x 106 conidia/ml-1 (Er et al. 2007).
The registered microbial control product, Bb Plus®, elicited a weaker concentration response in
T. leucotreta which was subsequently noted in the preliminary time-response trial when this
product competed poorly with two indigenous isolates. Further, G 14 2 B5 and G Moss R10 were
significantly more pathogenic towards T. leucotreta than G B Ar 23 B3 and FCM 10 13 L1. This
was interesting because G B Ar 23 B3 was the most pathogenic isolate towards fruit flies,
indicating variation in virulence of fungal isolates towards various insect hosts.
The preliminary time-response trials did not compliment the results of the concentrationresponse trial for T. leucotreta because a concentration of 1 x 105conidia/ml-1 (LC50 values of G
14 2 B5 and G Moss R10) was not used. The nearest concentration to that was 1 x 104
conidia/ml-1, which only induced 12.53 to 27% mycosis in T. leucotreta pupae at the longest
exposure time (96 hrs). At the higher concentration (1 x 107conidia/ml-1), isolates G Moss R10
and G 14 2 B5, elicited a response in 50% of test insects after 72 hrs (3 days) exposure. Based on
earlier trials, when pupae were left in fungal treated sand for up to seven days, over 80% mycosis
was observed for isolates G 14 2 B5 and G Moss R10; this suggests that T. leucotreta requires a
longer exposure time period than 72 hrs to elicit mortality in the region of 80%, at a
concentration of 1 x 107conidia/ml-1. This is valid since more pupae become infected over time
with an increase in time as fungal inoculum levels in the soil increase; subsequently more pupae
or emerging adults become infected. A time-response extrapolation at the lowest concentration
(1 x 104 conidia/ml-1) suggests that it requires a longer time period than 96 hrs to elicit mycosis
greater than 27% in the test insects. Unfortunately the exposure time-response results are
inconclusive and can only be considered as guideline information regarding future research with
these fungal isolates against T. leucotreta.
During the time-response trials an outlier was observed in that a large percentage of pupae
exposed to isolate G Moss R10 were mycosed at 12 hrs. This observation was as a result of
extensive bunching of pupae (larvae spin pupal cases adjacent to one another) leading to the
higher mycosis of pupae observed during the trial. Thaumatotibia leucotreta pupae are
sometimes grouped in the soil, possibly due to the confines of the Petri-dish, (personal
observation) and this may increase the incidence of mycosis because if infected pupae are
positioned next to healthy pupae fungal hyphae may grow onto the adjacent pupae (Fig. 3.4.2).
105
Krueger & Roberts (1997) noted that dry mycelial particles, as a form of inoculum, function in
the same way as fungal-killed insects. When water in the soil hydrates the fungal mycelia, new
mycelial growth is experienced and infectious conidia are produced on the new mycelial growth.
During fungal preparations, cultures were scraped with a glass rod; upon enumeration of fungal
concentrations, bits of hyphae were often observed. Shah & Pell (2003) suggested that infection
by Beauveria spp. in soil environments was improved by the radiation of modified hyphal
strands away from cadavers. Thus, it is valid that fungal mycelia can grow while bioassay
procedures are carried out, and further investigation is warranted.
Figure 3.4.2 (A) Thaumatotibia leucotreta larvae spin cocoons around themselves and may
become attached to one another in Petri-dishes. (B) Beauveria bassiana fungal mycelia seen
growing from two adjacent T. leucotreta pupae (Photo Credits: T. Goble, Rhodes University).
Some variation during bioassay procedures occurred in this study in that attenuation of the
isolate FCM 10 13 L1 was observed. This isolate exhibited a change in mycelia growth form
(Fig. 3.4.3) as described by Wang et al. (2005). In subsequently conducted germination tests, a
lower percentage of germination occurred prior to its use in concentration and time-response
trials. During initial fungal screening trials against C. capitata, C. rosa and T. leucotreta, this
isolate was particularly pathogenic and a significant effect was seen in its ability to reduce adult
emergence and elicit mycosis in all three test insect species. Subsequent trials revealed a less
virulent FCM 10 13 L1 isolate. It is empirically unknown why this isolate exhibited
morphologically altered growth in culture. Perhaps the isolate is inherently unstable, as
106
Brownbridge et al. (2000) suggested that the effects of repeated in vitro sub-culturing and
subsequent morphological characteristics vary considerably among isolates and fungal species. A
loss of pathogenicity has been noted in isolates of B. bassiana; whereas other isolates retain their
virulence after repeated sub-culturing under the same conditions (Brownbridge et al. 2000).
Furthermore, an isolate which is inherently unstable is undesirable as a biological control agent
because large commercial losses may be experienced if isolates lose their virulence (Wang et al.
2005).
Figure 3.4.3 (A & B) Beauveria bassiana isolate FCM 10 13 L1 showing in vitro culture
degeneration (fluffy-mycelia growth) (Photo Credits: T. Goble, Rhodes University).
As mentioned, selected isolates were chosen on the basis that they were able to induce mortality
in all three insect test species, thus there is wide scope for future research on T. leucotreta with
isolates which were not focused on in this study. Isolates of noteworthy mention are the three
Metarhizium isolates, G 14 2 B3, FCM Ar 23 B3 and G 11 3 L6, and four Beauveria isolates, G
Ar 17 B3, G OL R11, G Rose R9 and FF J&B R5. The increased incidence of pupal mortality, as
well as the relatively low concentrations of conidia required to elicit meaningful responses in T.
leucotreta pupae may suggest that pre-emergent or prophylactic control of T. leucotreta will play
a more significant role in field suppression than the control of adult life stages using indigenous
isolates of EP fungi.
107
IV
GENERAL DISCUSSION
4.1 Integrated Pest Management (IPM) strategies for the use of EP fungi in citrus
Citrus in southern Africa has moved towards an IPM approach in the last three decades and the
viability of IPM in citrus in southern Africa has been pioneered (Terblanche 1998). IPM has now
become an established management strategy, particularly in light of the large export earnings of
the citrus industry (Terblanche 1998). There is no universally agreed upon definition for IPM
and its initial conception was to rationalise the use of pesticides to delay the resurgences of pest
populations that had become resistant to chemicals and to protect natural enemies (Orr 2003).
Presently, there is much concern over the injudicious use of pesticides and IPM may be defined
as: “A flexible and holistic approach which views the agroecosystem as an interrelated whole
and utilises a variety of biological, cultural, genetic, physical and chemical techniques to hold
pests below economically damaging levels with a minimum amount of disruption to the cropping
ecosystem and the surrounding environment” (Orr 2003). Further, the establishment of an
International IPM Facility in 1995 with the support of the World Bank illustrated a global desire
to reassess plant protection procedures to enhance IPM and restrict the use of chemicals to those
which were absolutely necessary (Terblanche 1998).
Citrus has had a long history of cultivation in southern Africa and the literature on citrus pests
and their control is already extensive, particularly with work done on the biology of pests and
their control with chemicals (Bedford 1998). Citrus IPM strategies against Thaumatotibia
leucotreta, Ceratitis capitata and C. rosa include: pre-harvest monitoring, orchard sanitation,
sterile insect technique (SIT), attract and kill, mating disruption, male annihilation technique and
biological control using predators, parasitoids and microbes (viruses) (cf Chapter I). In contrast
the use of microbial biological control agents, such as EP fungi within citrus IPM in southern
Africa is still in its infancy. However, with the growing demands of overseas markets to limit
chemical pesticides, and reduce phytosanitary risk as well as the recent consumer interest in
108
organically produced foods, microbial biological control using EP fungi as an additional IPM
management tool holds promise.
Based on the biodiversity studies using bait insects presented in this study, it appears that some
EP fungal genera are a widespread component of the soil environments of both cultivated citrus
orchards and the natural areas surrounding orchards in the Eastern Cape Province in South
Africa. Fungal occurrence in natural areas is significantly higher than in cultivated citrus
orchards, due to the structurally complex nature of natural habitats which are free from pest
control practices and include diverse abundances of insect hosts (Meyling et al. 2009). The
present study also revealed that there were no significant differences observed in the occurrence
of EP fungi from conventional and organic citrus farms when cultivated orchards and refugia
were considered separately. This illustrates that both farming strategies will be amenable to EP
fungal application because the use of fungicides and insecticides did not exclude the occurrence
of these microbes in conventional citrus cropping systems. Despite these observations, it appears
that very few EP fungal genera are present in citrus soils and natural areas surrounding orchards.
This has been noted in other countries and different environments by Chandler et al. (1997),
Klingen et al. (2002) and Quesada-Moraga et al. (2007). Furthermore, EP fungal inocula levels
in citrus soils are too low to elicit any meaningful level of pest control by these microorganisms.
The application of EP fungi as biological control agents in citrus cropping systems will require
inundative application, as has been observed with most microbial control strategies using
entomopathogens to date (McCoy 2007). This strategy may be a better application tactic than
either classical biological control of insect pests with exotic fungal isolates or conservation
biological control which serves to preserve and promote exisiting EPF populations within
southern African citrus. The application of existing microbial control products seems useful
because the costly and time consuming research and testing of commercial products has already
been undertaken. However, their application is limited by existing formulations which may not
necessarily be practical or compatible in citrus cropping systems. The intentional introduction of
exotic pathogens into South Africa for classical biological control has generally been difficult
because of regulatory restrictions imposed by the National Environmental Management
Biodiversity Act (DEAT 2004). The potential for maladapted exotic fungal isolates is also high.
Beauveria bassiana and Metarhizium anisopliae are two facultative EP fungal species which
have been studied in great detail for the control of citrus pests and their formulation and use as
109
microbial control products in these cropping systems has been extensive (Lacey et al. 2001;
Lacey & Shapiro-Ilan 2003; Dolinski & Lacey 2007). In the present study, both B. bassiana and
M. anisopliae were the most commonly isolated EP fungi in citrus soils and soils from natural
areas. Their broad host ranges (Bidochka & Small 2005; Rehner 2005) and large genetic
variation within species (Bidochka et al. 2001; Meyling et al. 2009) suggests that microbial
control efforts should not be limited to the currently available mycoinsecticides and their
inundative releases. Rather, indigenous fungal isolates, which may be more virulent and better
adapted to local citrus cropping systems, should be utilised. The present study identified a
number of virulent EP fungal isolates which showed potential as biological control agents in the
laboratory, particularly against T. leucotreta. Further investigations in environmental
competence and virulence may greatly increase our understanding of the persistence and efficacy
of microbial control products, which are based on indigenous fungal isolates, in the field. When
EP fungi are capable of becoming established in a particular cropping system they are then able
to confer long-term regulation of pest populations, which is the ultimate goal of biological
control (Dolinski & Lacey 2007). Thus, the presence of both B. bassiana and M. anisopliae in
citrus cropping systems in the Eastern Cape illustrates their establishment and ability to persist in
these soil environments albeit at low densities. It is however unclear as to what extent these
indigenous isolates exert natural control pressures on citrus pests in the field, an area of research
which requires attention in southern African citrus.
The results of the present study revealed that indigenous fungal isolates were not able to induce
high mortality in C. capitata and C. rosa puparia when exposed to sand treated with conidia
suspensions. This illustrated the lack of susceptibility of the immature life stages of fruit flies to
the effects of EP fungi which has been observed by Poprawski et al. (1985), De la Rosa et al.
(2002) and Toledo et al. (2006). Deferred adult mortality, which resulted in a greater incidence
of adult fruit flies mycosed by EP fungi after treatment of soil, was sufficiently high to consider
the use of EP fungi as a subterranean treatment against emerging fruit fly adults. The observation
that increased incidences of deferred adult mortality due to mycosis, as a subsequent result of
subterranean treatment of immature life stages of fruit flies, has also been reported by Poprawski
et al. (1985), Ekesi et al. (2002) and Yee & Lacey (2005). Thus, the initial aim of the study,
which set out to investigate the potential application of EP fungi against the larvae and pupae of
fruit flies, did not show potential control of these immature life stages but did show potential
control of emerging adults at a subterranean level. Most of the indigenous isolates of EP fungi
tested in this study were able to induce high mortality in T. leucotreta final instar larvae and
110
pupae when exposed to sand treated with conidial suspension. The overall effect of this
subterranean treatment led to a significant reduction in adult moth emergence and eleven of the
initial twenty-one fungal isolates tested showed marked pathogenicity towards the pupae of this
insect species. Based on the results of the present study, it appears that effective control of T.
leucotreta immature life stages may occur if EP fungi are applied as a subterranean soil barrier
below citrus trees. During bioassays in the laboratory, low levels of fungal inocula were required
to elicit meaningful control of T. leucotreta; reiterating the potency of these indigenous fungal
isolates against this insect pest species and the practicality of utilising these isolates in the field.
Likewise, this subterranean control method may reduce the incidence of fruit fly adult survival
because adults become infected with fungi upon emergence. However, much higher levels of
fungal inocula were required to elicit meaningful control of fruit flies in the laboratory and these
levels may translate into uneconomical or impractical inocula levels in the field. Thus, the
utilisation of alternative EP fungal control strategies may need to be employed which would
lower the required fungal inocula levels by enhancing the efficacy of the dissemination of fungal
conidia, as has been observed with the use of autoinoculation devices, discussed below.
Augmentation as a biological control strategy, using identified indigenous EP fungal isolates in
citrus, has a relatively low implementation cost and could increase presently low fungal density
levels in the soil (McCoy & Tigano-Milani 1992). There are two approaches which may be
undertaken, inoculation and inundation. The inoculation approach using fungal isolates involves
applying small amounts of inocula early on in the crop cycle, with the premise that fungi will
cycle and establish epizootics in pest populations and spread over a period of time, maintaining
pest populations below economic thresholds (Shah & Pell 2003). The main impact of EP fungi
on pest populations however, may only occur once economic thresholds have been surpassed
(Inglis et al. 2001). The utilisation of this approach alone may be difficult to undertake against
fruit flies and T. leucotreta because the economic thresholds of these pests are usually very low
(Moore et al. 2008). Furthermore, the effects of EP fungi are density dependant and will require
increased pest population numbers to elicit epizootics in orchards to maintain fungal recycling
(Inglis et al. 2001). This may be difficult to maintain because various IPM strategies targeting
and controlling these insect pests may reduce pest populations to levels which are unable to
sustain fungal recycling.
The most common method of employing EP fungi for pest control is through inundation using
mycoinsecticides for short-term, rapid control of insect pests, much like the use of chemical
111
insecticides (Shah & Pell 2003). Fungi may be administered via herbicide applicators or other
existing equipment which makes this approach appealing (McCoy et al. 2007) and there is no
expectation that fungi will cause secondary infection or cycle in the environment (Shah & Pell
2003). If inundation is to be considered as a biological control option using indigenous EP fungal
isolates, which are applied as a soil barrier below citrus trees for the control of T. leucotreta and
emerging fruit fly adults, formulation of bare conidia will need to be undertaken to enhance the
persistence and infectivity of fungi in the soil. This is an important prerequisite for the successful
biological control of fruit flies and T. leucotreta pupae which are often sedentary or less
susceptible in immature life stages (cf Chapter III). Efficacy of application will be determined by
the larvae of these soil-dwelling pests contacting a sufficient number of conidia to become
infected (Jaronski et al. 2005). The infectivity of unformulated conidia for example, is greatly
affected by soil texture and moisture which is thought to be a result of pore structure (Jaronski et
al. 2007). As the soil hydrates, increasingly larger pore spaces become filled with water. Insects
moving in the soil tend to take the path of least resistance and thus move through these larger
pore spaces. Unformulated conidia tend to remain in the smaller soil pores and the result of this
is the ineffective or sublethal contact of insects and conidia (Jaronski et al. 2007). The effect of
soil texture on formulated granules however is far less and discussed further below.
Vänninen et al. (2000) investigated the persistence and penetration of unformulated fungal
conidia in various soil types over three years in Finland. The fungal species, M. anisopliae,
generally persisted better than B. bassiana in all soil types although persistence decreased with
soil depth and time. Beauveria bassiana, when applied as an aqueous soil suspension, showed an
85 to 95% reduction of viable conidia at 12 days post-application. However, three years postapplication there were still enough infectious M. anisopliae conidia in soils at all sites to infect
over 80% of Tenebrio molitor insects employed as baits (Vänninen et al. 2000). The authors
attributed the decrease in Beauveria inoculum levels in field soils to three processes:
biodegradation (considered the most important factor), physical weathering and percolation into
deeper soil layers. However, percolation into deeper soil layers is thought to be the least
important factor because most conidia are retained in the top 5 cm regardless of soil texture
(Ekesi et al. 2007). Pupating insects, such as Ceratitis fruit flies and T. leucotreta, never exceed
a depth of 7 cm in the soil, thus infection by EP fungi is maximised in this upper region of the
soil (Ekesi et al, 2007; Stibick, 2008). In terms of soil type, clay soils are considered to have the
highest carrying capacity of fungal conidia; clay can also function as a protectant against
biodeterioration, resulting in an increase in stability and longevity of conidia (Vänninen et al.
112
2000). Sandy soils allow for a faster percolation of fungal conidia (depending on the size of the
particles) and are generally considered warmer soils during the summer months because of the
increased abundances of quartz (a good conductor of heat), compared to clay.
Commercial interest in using EP fungi to control soil dwelling pests have resulted in
formulations which are based on liquid culture-produced mycelial pellets, encapsulated fungal
biomass or solid substrate-produced conidia, either in aqueous or oil suspensions or on granular
carriers (Jackson & Jaronski 2009). Recently, the discovery of microslerotia (compact hyphal
aggregates which are often melanised and highly resistant to desiccation), have shown to be
effective fungal application inocula (Jackson & Jaronski 2009). The ingredients selected for
formulation should not interfere with the fungal infection process in anyway and at best should
enhance viability, disease transmission and field persistence of EP fungi.
Nutritive or non-nutritive granules coated with EP fungal conidia, granules colonised by fungus
or granules composed of dry mycelium are examples of commercial formulations which have
been used against soil pests (Jackson & Jaronski 2009). The principal behind granular
formulations is that EP fungi will grow out onto the granular carrier and resporulate in the
presence of water (water activity greater than 0.98 or water potential of -3 MPa), much like an
insect cadaver (Jaronski 2007). Granules seem to show more resilience in the soil because fungal
sporulation occurs on the granules. However, increases in fungal titres (sporulation) in the soil
are rarely observed when application by aqueous suspensions or dry powders are made. Krueger
et al. (1992) cited by Jaronksi (2007) showed that M. anisopliae mycelial-based granules (0.250.5 mm diameter) applied to the soil for the control of corn rootworm, Diabrotica
undecimpunctata, at a rate of 120 granules/cm3 increased conidial titres in the soil by 1.1 x 104
colony forming units per granule. This author concluded that granules are readily able to support
EP fungal growth and that the inability of bare conidia to germinate in soil is due to the absence
of sufficient stimuli. Based on the above considerations, if application of indigenous EP fungal
isolates as a soil barrier is to be considered then granular formulation may be the most effective
way to formulate fungal conidia to enhance persistence and efficacy.
Jaronski et al. (2007) suggested that the use of EP fungal granules should be compatible with
farming equipment. The size and bulk density of the granules should not exceed 0.5-1.5 mm in
diameter and an upper practical limit of 22 kg granules per hectare, if they are to be properly
applied by existing equipment. Most conventional insecticide granules are applied at 11-17 kg/ha
113
(Jaronski et al. 2007). These authors mentioned that a critical concentration of granules is
required for high efficacy. In bioassays against third instar larvae of the sugarbeet root maggot,
Tetanops myopaeformis (Ulidid fly), in clay soil at optimal temperature and moisture for M.
anisopliae, four or more granules per gram (corn grit granules 0.5-1 mm diameter coated with M.
anisopliae and having a bulk density of 1400 granules per gram) were needed for greater than
90% efficacy. At anything less than that critical density, sharp decreases in infection rates were
observed in sugarbeet root maggot larvae (Jaronski 2007). At four granules per centimetre cubed
(cc) of soil, 158 kg granules per hectare would be required to apply a 15 cm band over a row of
sugarbeet incorporated to a depth of 2.5 cm. These application rates were uneconomical and
impractical. Jaronski et al. (2007) concluded that critical concentration (4 granules/cc of soil)
could be achieved at 11 kg/ha when granules were applied using in-furrow applications in
sugarbeet. Granule sizes of 0.5-1 mm in diameter create foci of greater than 1 x 107 conidia,
depending on the sporulation potential of the particular EP fungal isolate (Jaronski 2007).
However, if unformulated conidia are present in the soil, an insect would have to encounter
many unformulated conidia to obtain a lethal dose. With granular formulation, sporulation
occurs on the granule carrier and thus an insect coming into contact with just one granule (foci of
greater than 1x107 conidia) may obtain a lethal dose (Jaronski 2007). Granular applications are
most effective where insects
4.2 How can soil application of EP fungi work in citrus?
Soil application of EP fungi within citrus IPM is an exciting prospect and the timing of
application will depend on the type of citrus cultivar and the geographic region in which citrus is
grown. Unlike viruses, which are only pathogenic towards the larvae of T. leucotreta (Moore
2009), EP fungi are pathogenic towards T. leucotreta adults, larvae and pupae. Thus application
would not be limited to the control of only pupae in the soil, but can include the additional
control of adults (which is discussed below). Soil application of EP fungi against fruit flies in the
present study concluded that the immature life stages were less susceptible to EP fungi, but that
adults were more susceptible and adults emerging from the soil could become infected with EP
fungi. This deferred adult mortality due to fungi is recognised as an important mechanism for the
recycling of fungal inocula. Recycling occurs from the soil environment where pupae and
emerging adults reside to epigeal habitats where adults mate and live out their lifecycles (Ekesi
et al. 2007) and which may greatly enhance IPM using soil treatments.
114
Soil application of EP fungi, particularly as granules, offers the most effective formulation for
the application of EP fungi to the soil. This application method can serve as a „bi-catch‟ strategy
within citrus IPM and could possibly reduce the frequency of orchard sanitation, and at best
improve the control of T. leucotreta and fruit flies. Having EP fungi in citrus soils offers growers
a „safety blanket‟ for fruit which drop throughout the year as a result of the staggered maturation
of various citrus cultivars (Lee 2009). Fungi can be effective during the hotter months of the
citrus growing season, just prior to harvest, and as a post harvest control measure. The
application of EP fungi can further promote the use of parasitoids and be available to „catch‟
escaping insects when orchard sanitation cannot, as when stung fruit remains on trees and is not
spotted by orchard cleanup crews. The use of granular formuations is normally associated with
seed treatments in row crops, protecting plant roots where insects are expected to congregate to
feed (Jaronski et al. 2005). Thus, one limitation to granular application below citrus trees may be
the inadequate coverage for pests dropping from trees to pupate in the soil, in this case aqueous
suspensions may be a more economical approach to control citrus pests.
Orchard sanitation is usually recommended from as early as November in some areas because
small fruit (15-20 mm in diameter) can become infested with T. leucotreta or fruit flies and drop
off trees (Newton 1998). Raking up abscised, small fruit has been suggested to remove
infestation potential. However, this is very labour intensive and usually only about 30% of these
fruit are infested with T. leucotreta (Moore & Kirkman 2009). If EP fungi are present in the soils
at these times, the need to remove such fruit is reduced because the small percentage of escaping
larvae entering the soil to pupate can become infected with EP fungi, this in turn can reduce
labour and time costs. Soil treatment can be particularly useful in summer, as higher numbers of
T. leucotreta pupae have been recovered from citrus orchards from October to mid-December
than at any other time in the season (Moore & Kirkman 2009). This illustrates that T. leucotreta
activity and population numbers are highest during these months, an observation which is also
true of fruit flies (Du Toit 1998). Higher numbers of pest populations are required to elicit
epizootics as transmission is increased. The increased incidence of humidity during the summer
months can also greatly accelerate the fungal infection process (Inglis et al. 2001).
Orchard sanitation reduces citrus pest larval populations in orchards. However, it reduces the
number of larval parasitoids which themselves are employed as biological control agents,
particularly against T. leucotreta (Moore & Kirkman 2009). A reduction in orchard sanitation
has been suggested to promote the activity of T. leucotreta larval parasitoids (Keeton 2007).
115
Thus, the benefits of soil application of EP fungi will include promoting the use of existing T.
leucotreta parasitoids, Agathis bishopi and Apophua leucotreta, which require larvae to complete
their lifecycles. Fruit fly parasitoids are not mentioned here because their presence in citrus
orchards has a negligible effect on fruit fly control (Du Toit 1998). Competition between fungi
and parasitoids is reduced because fungi are applied to the soil. Thus, parasitoids will host-seek
wandering larvae from fruit on the ground or on trees and those that are missed by these natural
enemies will subsequently be colonised by fungi in the soil.
The requirement of humidity for fungi to germinate suggests that they may be more effective at
controlling pests in the northern, hotter and more humid citrus growing regions of South Africa
such as KwaZulu-Natal, Mpumalanga and the Northern provinces. Further, natural infection by
Beauveria bassiana has often been recorded in pupae of T. leucotreta amongst the leaf litter in
Zebediela Citrus Estate in the Northern Province (Van den Berg 2001). The developmental rates
of citrus pests are higher in these regions and the percentage of dropped fruit still infested with T.
leucotreta and fruit flies is lower at these times due to faster insect development rates (Moore &
Kirkman 2009). In the hotter months of the citrus growing season, the percentage of fruit which
was still infested with T. leucotreta when collected weekly, dropped from 75% to between 60
and 70% (Moore & Kirkman 2009). Larval movement out of fruit is much faster, due to
increased temperatures and subsequent developmental rates and the employment of orchard
sanitation is not as effective because larvae have already exited fruit to pupate in the ground
(Moore & Kirkman 2009). Current control measures employ the use of labour to remove
dropped fruit twice a week, a practice that is expensive and requires time. Thus, the presence of
EP fungi in the soils offers additional protection as escaping larvae, which then pupate in the
soil, could subsequently become infected with fungi. This could reduce the need for twice
weekly orchard sanitation to a weekly practice even during summer months. Fruit fly larvae
rarely leave the fruit before it has fallen off citrus trees (Moore & Kirkman 2009), thus orchard
sanitation still remains an important control tactic.
The argument that fungi require hosts in order to build up populations in these systems and elicit
epizootics is valid. The fact that every IPM strategy in citrus, in various ways, targets the same
insect pests which are hosts to EP fungi, suggests that not enough hosts will be available for
fungi to recycle. However, the concept of the inundation approach is that EP fungi are often
applied in large amounts (usually as mycoinsecticides), for short-term, rapid control of insect
pests and there is often no expectation that fungi will cause secondary infection or cycle in the
116
environment (Shah & Pell 2003). Further, by applying the conservation biological control
strategy, an ecological approach, which usually requires the modification of the physical
environment or the management of other crop production practices (McCoy & Tigano-Milani
1992). The persistence of EP fungi in the soil may be increased which may subsequently reduce
follow-up applications as fungi may then survive for longer.
There are many citrus crop production and protection practices which could be considered to
enhance the persistence of EP fungi. Irrigation strategies can increase available moisture, which
is an important aspect of fungal germination (McCoy et al. 2007). The timing of pesticides and
especially fungicide used in citrus orchards will affect the survival of fungal conidia (Shah &
Pell 2003). Mulching can increase the presence of EP fungi because an increased carbon load in
the soil is favourable for soil dwelling pests, which themselves are potential hosts for EP fungi
(Klingen et al. 2002). Citrus orchards are perennial habitats and management practices which
increase disturbance, such as tilling or ploughing, are thus negligible in these systems.
Furthermore, tree canopies provide continuous shade for microbes in the soil and reduce the
amount of UV-B irradiation which is detrimental to EP fungal germination (Braga et al. 2001).
These factors support the implementation of conservation biological control in citrus cropping
systems and the abovementioned considerations can greatly increase the presence of EP fungi in
these soils which then subsequently increase IPM potential. Inundation and conservation
biological control strategies will benefit greatly from being used together or in conjunction with
other cultural methods used in citrus IPM (Shah & Pell 2003). The use of mycoinsecticides, for
example, are applied as products and their use is for short-term pest control however, the
persistence of fungal inocula in cropping systems may be greatly increased by applying the
conservation biological control strategy in citrus cropping systems, particularly because of the
perennial nature of these systems.
The importance of removing infested fruit from citrus trees, as well as out-of-season fruit once
harvest has finished, is crucial for the control of subsequent T. leucotreta population build-up
and for the control of various post-harvest fungal diseases such as sour rot, blue/green moulds
(Penicillium spp.) and Alternaria. It can however be difficult to detect all fruit fly stung fruit on
trees and the removal of such fruit is labour intensive. Thus, fruit that is missed by orchard
cleanups, due to the difficulty in detecting all stung fruit, may subsequently be protected by the
presence of EP fungi in the soils. Further, in areas such as Mpumalanga, where the coffee bean
weevil (Araecerus coffeae) infests mummified fruits (Moore & Kirkman 2009), EP fungi may
117
also serve as a control measure against these secondary pests, although further study is required.
The presence of EP fungi as a post-harvest control tactic is also valid and fungi will continue to
self-replicate in these systems provided the correct environmental conditions persist and the
appropriate conservation control tactics are applied (i.e. timing of fungicides should not coincide
with the application of EP fungi).
The timing of EP fungal application should consider the biology of insect pests and climatic
conditions of a particular citrus area which dictate the number of generations per year. Both T.
leucotreta and fruit flies are generally present throughout the year but at various levels
depending on the time of year, different growing regions and various cultivars. Successful soil
application of EP fungi will coincide with periods when the highest percentage of the T.
leucotreta population are wandering larvae and pupae, which can be as early as October in some
regions (Moore & Kirkman 2009). Emerging fruit fly adults is the recommended life stage to
target these pests in citrus using soil applications. Ceratitis fruit flies become highly active
around the fruit colouring period, which may be as early as January in early cultivar varieties
such as, Satsuma mandarins and some early navel oranges (Lee 2009). Du Toit (1998) reported
that in the Sundays River Valley in the Eastern Cape, C. capitata populations reached their peak
as early as December over five consecutive years. Monitoring of pest populations will reflect the
best times at which to apply fungi and the targeted life stage of each pest must be considered. For
best results it is recommended that fungi be applied before insects numbers are too high, as a
prophylactic treatment so as to give fungal inoculum levels a chance to increase in the soil,
particularly if granular applications are made. October would be a good month to initiate EP
fungal application in the soil against T. leucotreta and December would be the recommended
month to initiate application against fruit flies in some areas. It might however be impractical to
stagger these applications by just two months. It is recommended that perhaps just one initial
application be made against both pests in November, provided that both pests occur and are
targeted in that area.
The use of granules may greatly increase the efficacy and persistence of EP fungi in soils
compared to liquid formulations. High levels of EP fungal conidia (4.9 to 9.5 x 104 cfu/g) were
still recovered from soil treated with a granular formulation 668 days post-treatment in a study
by Ekesi et al. (2005). These persistence levels suggest that perhaps only one application to the
soil per growing season may be required when granular formulation is considered. In the case of
aqueous suspensions, when unformulated conidia are used, no sporulation is observed in the soil
118
because of a lack of stimuli (Jaronski 2007). However, Ekesi et al. (2005) observed that the
infectivity of aqueous and oil/aqueous (50:50) formulations of Metarhizium anisopliae persisted
in the soil until one year post-application, at which point conidial levels of these two liquid
formulations decreased significantly. So it is possible that only one application per growing
season may be required in the case of aqueous suspensions too. McCoy et al. (2007) for
example, reported that EP fungi were successfully applied beneath citrus trees as aqueous
suspensions for the control of Diaprepes weevil using herbicide applicators and injection via low
volume irrigation systems. Using a tractor-mounted herbicide sprayer (boom size 1.5-2.5 m) a
fungal treated band was uniformly applied to the soil beneath citrus trees at the point between the
trunk and the canopy edge (McCoy et al. 2007). McCoy et al. (2007) further suggested that the
application of EP fungi via microsprinkler irrigation could be a more cost effective method for
applying these microbes beneath citrus trees and that sprinklers be positioned about 30 cm from
the tree trunk and be suitable to cover about 80% of the root area. However, because of the
competitive nature of soil and the complicated abiotic factors associated with this environment,
persistence is not straight forward and future studies will be needed to determine appropriate
application rates. Furthermore, the distribution of conidia in the soil may be greatly increased if
repeated application is made.
Timing may be predicted with existing pre-harvest monitoring traps, such as the Delta or PVC
pipe traps fitted with a Lorelei dispenser for T. leucotreta (Moore 2009). Thaumatotibia
leucotreta flight peaks usually occur as early as October, with a main flight peak occurring at the
end of November depending on the region; in the cooler climes of the Western Cape these peaks
are less clearly defined (Moore & Kirkman 2009). Subsequent moth peaks will be smaller from
January to March and then tail off as winter approaches. For fruit flies, Sensus® traps fitted with
a Questlure® or Capilure® attractant are often employed to monitor fruit fly populations (Ware
2002; Ware et al. 2003). Timing of EP fungal application against fruit flies will be difficult
because of the low economic threshold of this pest and traps are not there to indicate when
timing of an application programme should be initiated but rather when a fixed control
programme is inadequate (Moore et al. 2008). Trap surveys should be initiated during November
until harvest (Moore et al. 2008) and application of EP fungi should be made when trap catches
exceed their thresholds.
119
4.3 Autoinoculation devices for the use of EP fungi in citrus IPM
Strategy options, in terms of the implementation of EP fungi in the field as part of an IPM
programme against fruit flies, have been summarised by Ekesi et al. (2007). Soil inoculation has
been discussed and due to the low mortality rates in the immature life stages of fruit flies as well
as the large inoculum rates required to suppress fruit flies in the soil, the application of
alternative techniques is required. Horizontal transmission was a listed factor which may have
increased the rate of adult mortality during bioassay procedure in the present study, and was
sufficiently high to further investigate the potential use of control strategies which are based on
this factor, such as the use of autoinoculation devices.
The assisted autodissemination control option using devices in which fungal conidia are spread
by adult insects, has recently been proposed for insect pests which respond to visual and
olfactory cues (Ekesi et al. 2007). The strategy is to attract insect pests using visual, chemical
and food lures to the focal point of EP fungi. Insects then inadvertently pick up the fungal
inoculum and disseminate the infective conidia via horizontal transmission to other healthy
individuals of the pest population (Dimbi et al. 2003a). Autoinoculation devices have been
widely used for a number of insect and fungal species; some notable studies include their use
against fruit flies with M. anisopliae (Dimbi et al. 2003a; Ekesi et al. 2007) and diamond back
moth, Plutella xylostella with the fungus, Zoopthora radicans (Pell et al. 1993; Furlong et al.
1995; Furlong & Pell 2001). The technique has many advantages over the mass application of
fungi: (1) the strategy is highly efficient, particularly when a food bait or pheromone lure is
incorporated into the device; (2) it may be species-specific; (3) it is cost effective because the
required fungal inoculum levels are low; (4) the autoinoculation devices are easy to build and
maintain and (5) infected insects will return and transmit EP fungi to the pest population
(Baverstock et al. 2009). Autoinoculation devices could contain an insecticide, which is what the
„attract and kill‟ method of pest control is based on, for both fruit flies and T. leucotreta in citrus
(Ware et al. 2003; Kirkman 2007). However, by incorporating EP fungi, which themselves are
self-replicating, it is ensured that vectoring and infection is then spread throughout a pest
population and persistence is enhanced (Baverstock et al. 2009).
Autoinoculation devices are based on 750 ml plastic bottles that hold both the autoinoculators
and the food/pheromone attractants. Autoinoculators which have been investigated thus far
consist of a maize cob (Ekesi et al. 2007), cheesecloth (Dimbi et al. 2003a), non-woven fabric
120
(Baverstock et al. 2009) or a Petri dish (Ekesi et al. 2007). The plastic bottle is usually divided
into three compartments: (A) the upper compartment is used to introduce the autoinoculators; (B)
the middle compartment where the autoinoculators are located and where flies and moths
become contaminated with EP fungi and (C) the lower compartment which is separated from
with a mesh gauze (20 mm diameter) to prevent insects from drowning. The lower compartment
contains a food bait or pheromone to lure insects to the device. Entrance and exit holes are made
to allow free movement of insects from the device (Dimbi et al. 2003a).
Laboratory and field caged experiments using autoinoculation devices have shown that the
number of conidia picked up by individual flies is in the range of 4.2 x 10 5 and 1.0 x 106 conidia
per fly (Dimbi et al. 2003a). Castillo et al. (2000) reported that adult C. capitata fruit flies, when
inoculated with various EP fungal isolates on the abdomen, required from 1 x 102 to 1 x 106
conidia per fly to elicit mortality greater than 50% at 10 days post inoculation. During the
investigation of B. bassiana as a pathogen in the autoinoculation device used to infect P.
xylostella, male moths acquired a lethal dose of B. bassiana within 1 second of exposure to all
concentrations tested: 0.24 to 6.2 µg conidia/mm2. This suggested that the use of these devices to
inoculate moths provided an efficient method for transferring inoculum to the rest of the
population (Furlong & Pell 2001). These authors showed the overall rate of mortality was not
influenced by concentration of B. bassiana conidia on the surface of the inoculator but time until
insect death was affected by both exposure time and conidia concentration. As moths spent more
time walking around the inoculation chamber, they acquired increased levels of fungal
inoculums. Thus, by designing devices which maximise time spent by moths in these chambers,
one ensures that lethal doses are then transferred to the susceptible target population (Furlong &
Pell 2001).
Population suppression by autodissemination is a function of horizontal transmission between
healthy insects and fungal contaminated insects (Ekesi et al. 2007) and is greatly increased
during insect mating. Mating competitiveness in fruit flies is a function of predisposition and
compatibility (Ekesi et al. 2007). The ability of insects to find a receptive mate and successfully
copulate will be crucial to the functioning of horizontal transmission of EP fungi in the field.
Dimbi et al. (2009) showed that healthy male fruit flies belonging to three species (C. capitata,
C. cosyra and C. fasciventris) showed no mating preference for healthy female flies over fungus
infected females. It was concluded that fungus treated male fruit flies competed equally for
virgin females on days 0 to 2 post inoculation. However, at day three, untreated males had a
121
significantly higher percentage of pairing with females compared to fungal treated males (Dimbi
et al. 2009). These authors further reported a significant delay among fungus infected male flies
at the onset of calling and mating due to grooming efforts to remove fungal inoculum. Grooming
behaviour further resulted in the spread of fungal inoculum to areas of the body that were not
contaminated, particularly the mouthparts (Dimbi et al. 2009). Feeding flies, which are also
known to regurgitate food and feed one another, may thus also spread EP fungal inoculum to
fruit or other individuals where conidia may be picked up by healthy flies.
Female fruit flies are oligogamous (Quesada-Moraga et al. 2008) and choose mates based on
male pheromone; if male mating behaviour is altered and females preferentially chose healthy
males over fungus infected males the effectiveness of horizontal transmission may be lost (Ekesi
et al. 2007). However, when considering an attractant which is more female specific, as in the
Questlure® attractant, more females are attracted to the autoinoculation devices. Female-bias
mating will thus not affect horizontal transmission because female flies are then also
contaminated with EP fungi. Fruit flies form leks (which are groups of calling males) and this
aspect of their biology may be a useful point for the autodissemination of fungal inoculum
among healthy individuals because of the close proximity of flies at this time which increases
contact chances and pathogen transmission (Ekesi et al. 2007). In addition, calling male flies also
show the tendency of numerous attempts at homosexual mating (Ekesi et al. 2007). These male
to male mounting attempts as well as heterosexual fruit fly mating will almost certainly increase
horizontal transmission of EP fungi in the field.
In small scale field plot experiments undertaken in Kenya using Metarhizium anisopliae in
autoinoculation devices against mango fruit fly, Ceratitis cosyra, Ekesi et al. (2007)
demonstrated the efficacy of using these novel devices to significantly reduce fruit flies in this
crop. Application of autoinoculation devices has considerable advantage over mass application
using soil treatments because relatively low amounts of fungal inoculum levels are required for
field application. The case study revealed that dry conidia applied to the devices remained viable
for five weeks (68% germination) but then dropped to 27% after six weeks. Longer persistence
of fungal conidia was attributed to the nature of the device and placement (under tree canopies)
which significantly reduced UV radiation. Currently, the cost of production of M. anisopliae
used in the biopesticide, Green Muscle® (BCP, South Africa) for locust control in Africa is US$
200/kg of dry spores which makes application rates unaffordable in developing countries (Ekesi
122
et al. 2007). However, by using these autoinoculation devices about 0.8-1.0 g of conidia are
required per device, which translates into a weekly application of 80-100 g of conidia to treat one
hectare of a crop (100 trees/ha). However, fungal inoculum could persist for one month,
therefore weekly application may not have been necessary, and this could have further reduced
cost. Further, the manufacturing of the devices is inexpensive and maintenance simple (Ekesi et
al. 2007).
The implementation of autoinoculation devices in citrus IPM as an additional tool for the control
of fruit flies and T. leucotreta seems promising, particularly because the ratio of fungal inoculum
to hosts is greatly reduced compared to the mass application of fungi in soils. Further horizontal
transmission appears to be an important aspect of fungal dissemination within insect populations.
Based on the study by Ekesi et al. (2007) it appears that the implementation of autoinoculation
devices within citrus must coincide with insect flight peaks, because peaks determine when
mating is at its highest. Data from monitoring trap systems within specific citrus areas will
dictate population peaks and the deployment of autoinoculation devices. The biology and
occurrence of fruit flies within specific geographic regions may require greater understanding
because trap monitoring does not dictate when to apply a strategy but rather when a particular
strategy is not working (Moore et al. 2008). The ultimate aim of citrus pest management is to
reduce oviposition by gravid females (Ekesi et al. 2007). Apart from obvious death elicited by
EP fungi, Yee & Lacey (2005) reported that the western cherry fruit fly, Rhagoletis indifferens,
females infected with M. anisopliae laid fewer eggs between three and seven days post-infection.
Castillo et al. (2000) reported that a reduction in C. capitata fecundity was evident four days
post infection. In fruit flies, young gravid females are more susceptible to fungal infection than
older flies, and autoinoculation must target this age group which will reduce the level of
oviposition in field populations and contribute to overall insect management. Based on bioassay
experiments in the present study, it appears that the adults of T. leucotreta are also susceptible to
fungal infection, but it is unclear as yet as to whether younger moths will be more susceptible
than older moths and warrants further investigation if autoinoculation is to be considered against
T. leucotreta.
Fruit fly population numbers may peak in some citrus growing areas as early as December,
weather dependant (Du Toit 1998). Thaumatotibia leucotreta population peaks may be as early
as October in some areas (Moore & Kirkman 2009). Autoinoculation devices can be employed in
early October or December (depending on insect species) just before insect populations reach
123
their peak. Further, if devices are baited with pheromones or parapheromones just prior to insect
peaks, the device efficacy may be increased because insects seeking mates will be attracted to the
devices and become infected with EP fungi. Upon habituation of the pheromone within devices,
fungal infected insects will then seek out mates again and copulate. Mating increases horizontal
transmission of fungi and thus inoculum may be spread through mating insect populations faster,
thus ultimately decreasing egg laying. In the current control of T. leucotreta using traps, only
synthetic female pheromones have been discovered which attract males (Newton 1998), thus the
incidence of female fungal infection may be low. However, T. leucotreta exhibits multiple
mating in both males and females (Newton 1998), sometimes moths can mate up to several times
per day (Stibick 2008), and this aspect of their biology would greatly increase horizontal
transmission.
The replacement of devices, according to the persistence of EP fungal conidia, may need to be
done every month from October or December onwards, until pest population numbers decrease.
A one hectare citrus orchard with tree spacing of approximately 2 m may contain over 1000 trees
in the orchard (Knott, pers. comm.). However, the most common modern citrus planting distance
is 6 m x 3 m, which gives 555 trees/ha (Moore, pers. comm.). Based on Ekesi et al. (2007), it is
suggested that a citrus orchard of approximately one hectare may require a weekly application of
800-1000 g (1kg/ha) of conidia (1000 trees/ha), which is still well within the application rate of
most conventional insecticides which are applied at 11-17 kg/ha (Jaronski et al. 2007). However,
based on the study by Ekesi et al. (2007), mango trees were widely spaced (7 m x7 m) and
perhaps the suggested rates above are excessive. Further study within citrus orchards will
estimate these values more precisely. Depending on fungal persistence (germination percentage),
the weekly application rate may be reduced to a monthly application treatment, as Ekesi et al.
(2007) showed that M. anisopliae conidia could last for up to five weeks. Furthermore, most fruit
fly monitoring trap lures are replaced after 4 to 6 weeks anyway because the lure attractant fades
(Grout, pers. comm.). Thus, the compatibility of fungal conidia and attractants is valid and they
appear to have similar periods of residual efficacies. This reiterates the practicality of using these
devices within citrus orchards to disseminate fungal conidia effectively for the control of adult
fruit flies and T. leucotreta.
Fruit flies are present throughout the year but overwinter in the adult stage and food, water, and
shelter are important overwintering factors (Weems 2002). T. leucotreta is also present
throughout the year at all life stages however, over winter developmental rates are slowed and
124
larvae will either spend a substantial amount of time in fruit or pupate and spend up to several
weeks (29 to 40 days) in the soil (Newton 1998). The possibility of erecting autoinoculation
devices, baited with food attractants in winter to try and reduce overwintering pest populations
was considered, however in the case of fruit flies, they are poorly attracted to traps during winter
months (Botha et al. 2004). Thus, the dissemination of fungal inoculum in winter may be slow
and negligible but worth investigating in future studies as these pests are present all year round.
4.4 Conclusion
No single microbial control agent will provide sustainable control of an insect pest or complex of
pests. However, the prospect of applying EP fungi in citrus as an additional insect management
strategy holds promise, particularly as applications via existing equipment, irrigation and
monitoring systems are already available. Research will continue to identify the limitations on
efficacy and field persistence of EP fungi to improve their use within cropping systems,
particularly because there is much concern over the injudicous use of pesticides which greatly
affects the lucrative southern African citrus export industry. Integration of EP fungi within IPM
systems will require detailed compatibility studies and the development of effective guidelines
for their use (Inglis et al. 2001). EP fungal integration within citrus IPM will require further
monitoring to optimise the levels of control achieved with each application and the strategy used.
One of the major limitations in using EP fungi is the inconsistency of efficacy when useful
isolates reach the stage of large scale field testing (Whipps & Lumsden 2001). Fungal isolates
must survive application and then persist in the environments for which they are required. Thus,
it is crucial that selection, screening and further development processes adopt an ecological
approach, which takes into account the features of the required environment (Whipps &
Lumsden 2001). Subsequent steps in the development of EP fungi as mycoinsecticides are the
inoculum production, formulation and application technologies which need to be cost effective
and easy to implement (Butt et al. 2001). Further, cost/benefit analyses of the EP fungal
commercialisation process have often slowed development because of small niche markets
which require EP fungi as biological control agents (Whipps & Lumsden 2001). The cost of
producing microbial control products must be judged in terms of the value of the crop protected
by using the control agent and in comparison to the cost of chemical control options (Dolinski &
Lacey 2007). In Florida, for example, the use of EP nematodes is an integral part of citrus IPM
and this indicates that the cost/benefit relationship is positive (Dolinski & Lacey 2007).
125
Technology transfer is also crucial to the successful implementation of EP fungi within cropping
systems and a greater understanding of the ecological requirements of these biological control
agents must be met by growers and microbial product representatives. If all the above
considerations are adhered to then the utilisation of EP fungi within citrus cropping systems may
have a bright future.
4.5 Future work
The acquisition of EP fungal isolates from citrus soils and soils of natural areas surrounding
orchards in the Eastern Cape Province has offered some insight into their biodiversity and
occurrence. Implemented bioassays against target citrus pests have offered further insight into
the relative potencies (as shown by the LC50 and LC90 values) of these obtained indigenous
isolates against T. leucotreta and fruit flies in the laboratory. However, bioassays need to be
repeated, particularly because the LC50 and LC90 values against fruit flies were higher than the
tested concentration range. Future work should include the repetition of bioassays against T.
leucotreta with the existing concentration range: 1 x 104, 1 x 105, 1 x 106, 1 x 107, 1 x 108
conidia/ml-1. Repeated bioassays against fruit flies should be initiated at higher concentrations. It
is suggested that a starting concentration range against fruit flies will include: 1 x 108, 1 x 109, 1
x 1010, 1 x 1011, 1 x 1012, 1 x 1013 conidia/ml-1, to accurately determine the LC values against
these insect pests.
Furthermore, time-response trials against fruit flies and T. leucotreta should be undertaken to
ascertain LT50 and LT90 values for the most virulent isolates. It is suggested that Beauveria
isolates, G Moss R10, G 14 2 B5 and G B Ar 23 B3 be considered primarily in repeated
bioassays. However, the selection of isolates was based on their ability to induce mortality in all
three insect test species, thus there is more scope for future work on T. leucotreta with isolates
which were not focused on in this study. Isolates of noteworthy mention are: three Metarhizium
isolates, G 14 2 B3, T. leucotreta Ar 23 B3 and G 11 3 L6 and four Beauveria isolates G Ar 17
B3, G OL R11, G Rose R9 and FF J&B R5.
A much needed step following laboratory bioassays should include the characterisation of the
ecological constraints of the candidate isolates. Environmental competence is an important
prerequisite particularly as isolates need to persist to be effective against soil pests.
126
1.)
Environmental competence tests will include temperature or thermo-tolerances (Ouedraogo
et al. 1997; Ekesi et al. 2003; Devi et al. 2005; Rangel et al. 2005a; Fernandes et al. 2008),
UV-B tolerance and range (Braga et al. 2001; Rangel et al. 2005b; Fernandes et al. 2007)
and moisture or drought tolerance (Ekesi et al. 2003; Devi et al. 2005) of isolates in the
soil. The effect of these factors on fungal virulence must also be considered. It is suggested
that these environmental competence tests be carried out in conjunction with known
environmental parameters in various citrus growing regions.
2.)
Other environmental competence factors which should be investigated to ensure
environmental compatibility in citrus cropping systems include:
-
The effects of various soil types (clay, loam, sandy) and the carrying capacity and
persistence of conidia (Vänninen et al. 2000; Jaronski 2007) present in the soil.
-
Citrus pesticide and fungicide compatibility tests using candidate isolates may offer some
insight into application techniques. For example, the viral biopesticide, Cryptogran ® (River
Bioscience, South Africa) has been shown to be compatible in tank mixes with the
following chemicals mancozeb, benomyl, strobilurins, abamectin, methomyl, methidathion
and pyriproxyfen which has greatly increased the ease of application and acceptance for its
use in citrus (Moore 2009). Some noteworthy studies have been undertaken to evaluate the
effects of pesticides and fungicides on EP fungi by Filho et al. (2001), James & Elzen
(2001), Shapiro-Ilan et al. (2002), Sterk et al. (2002), de Oliveira et al. (2003), Saenz-deCabez Irigaray et al. (2003), de Oliveira & Neves (2004), Mochi et al. (2005, 2006),
Purwar & Sachan (2006) and Thompson & Brandenburg (2006).
3.) The effects of various growth media on the sporulation, growth and germination of
candidate isolates need to be investigated, particularly if mass production of fungal conidia
is to be considered (Campbell et al. 1978; Li & Holdom 1995; Ibrahim et al. 2002; Safavi
et al. 2007). Fungal radial growth, sporulation potential and germination speed have been
shown to change not only with fungal species and isolates, but also with the
carbon/nitrogen ratios of the media used for production. It is important to note that
formation of distinct morphological variants or sectors is often observed in fungal cultures
maintained on artificial media, thus the stability of isolates needs to be further investigated
127
in conjunction with the effects of nutrients if mass production is considered (Brownbridge
et al. 2001; Shah & Butt 2005; Wang et al. 2005).
4.) Genetic characterisation studies of candidate EP fungal isolates are also an important
prerequisite for the identification and tracking of strains in nature and the determination of
genetic diversity. Molecular techniques used to ascertain genetic diversity in EP fungi
include restriction fragment length polymorphism (RFLP), the use of the ribosomal internal
transcribed spacer region (ITS) and the nuclear gene, elongation factor- 1á (EF-1á) (Rehner
& Buckley 2005). Further, multilocus microsatellite genotyping mapping (Meyling et al.
2009) and universally primed (UP) PCR fingerprinting methods (Meyling & Eilenberg
2006a) are additional genetic methods employed.
5.) If autoinoculation devices are to be considered in citrus IPM, then the following biological
and physiological aspects need to be evaluated:
- The horizontal transmission potential of candidate isolates against citrus pest adults (Furlong
& Pell 2001; Quesada-Moraga et al. 2008).
- The investigation of the effects of candidate isolates on mating behaviour of citrus pests
(Dimbi et al. 2009).
- The effects of candidate isolates on the fecundity and egg fertility of citrus pests (Ekesi &
Maniania 2000).
128
V
REFERENCES
ABBOTT, W.S. 1925. A method of computing the effectiveness of an insecticide. Journal of
Economic Entomology 18: 265-267.
AEMPRAPA, S. 2007. Entomopathogenic fungi screening against fruit fly in Thailand. KMITL
Science & Technology Journal 7: 122-126.
AGRICULTURAL RESEARCH COUNCIL - Institute for Soil, Climate and Water. 2009. Soil
Classification-A Binomial System for South Africa. [Available: http://www.agis.agric.za/
agisweb] accessed: 03/05/2008.
AGRICULTURAL RESEARCH COUNCIL - Institute for Tropical and Subtropical Crops.
2003.
[Available:
http://www.nda.agric.za/docs/CultivatingCitrus.pdf]
accessed:
10/09/2009.
ALI-SHTAYEH, M., MARA‟I A.B. & JAMOUS, R.M. 2002. Distribution, occurrence and
characterization of entomopathogenic fungi in agricultural soil in the Palestinian area.
Mycopathologia 156: 235-244.
ALVES, S.B., TAMAI, M.A., ROSSI, L.S. & CASTIGLIONI, E. 2005. Beauveria bassiana
pathogenicity to the citrus rust mite Phyllocoptruta oleivora. Experimental and Applied
Acarology 37: 117-122.
ANAND, R., PRASAD, B. & TIWARY, B.N. 2009. Relative susceptibility of Spodoptera
litura pupae to selected entomopathogenic fungi. BioControl 54: 85-92.
ANNECKE, D.P. & MORAN, V.C. 1982. Insects and mites of cultivated plants in South
Africa. 29-62. Butterworth & Co, Durban, South Africa.
ANONYMOUS. 2001. Azoxystrobin. Pesticide News 51: 21.
ANSARI, M.A., SHAH, F.A., WHITTAKER, M., PRASAD, M. & BUTT, T.M. 2007. Control
of western flower thrips (Frankliniella occidentalis) pupae with Metarhizium anisopliae
in peat and peat alternative growing media. Biological Control 40: 293-297.
129
ASENSIO, L., CARBONELL, T., LOPEZ-JIMENEZ, J.A. & LOPEZ-LLORCA, L.V. 2003.
Entomopathogenic fungi in soils from Alicante province. Spanish Journal of Agricultural
Research 1(3): 37-45.
BARNETT, H. L. 1960. Illustrated Genera of Imperfect Fungi. 54-75, 154. Burgess Publishing
Company, Minneapolis, USA.
BARRON, G. 2001. War in the microworld - Beauveria bassiana. Department of
Environmental
Biology,
University
of
Guelph,
Canada.
[Available:
www.uoguelph.ca/~gbarron/ MISCELLANEOUS /nov01.htm] accessed: 07/05/2008.
BATEMAN, R.P. & ALVES, R.T. 2000. Delivery systems for mycoinsecticides using oilbased formulations. Aspects of Applied Biology 57: 163-170.
BAUER, S. 2005. United States Department of Agriculture (USDA), Animal and Plant Health
Inspection Service. [Available: http://www.aphis.usda.gov/ppq/manuals] accessed:
08/03/2008.
BAVERSTOCK, J., ROY, H.E. & PELL, J.K. 2009. Entomopathogenic fungi and insect
behaviour from unsuspecting hosts to targeted vectors. BioControl DOI 10.1007/s10526009-9238-5.
BEDFORD, E.C. 1998. Introduction (Part I), Methods of controlling citrus pests (Part II) and
Pesticides (Part III). In: E.C.G Bedford, M.A. Van den Berg & E.A. de Villiers (Eds.)
Citrus Pests in the Republic of South Africa. 1-22. Dynamic Ad: Nelspruit, South Africa.
BEGEMANN, G. 2008. The mortality of Thaumatotibia leucotreta (Meyrick) final instar
larvae exposed to the entomopathogenic fungus Beauveria bassiana (Balsamo)
Vuillemin. African Entomology 16(2): 306-308.
BIDOCHKA, M.J., KAMP, A., LAVENDER, T., DEKONING, J. & DE CROOS A.J.N. 2001.
Habitat association in two genetic groups of the insect-pathogenic fungus Metarhizium
anisopliae: uncovering cryptic species? Applied and Environmental Microbiology 67(3):
1335-1342.
BIDOCHKA, M.J., KASPERSKI, J.E. & WILD, G.A.M. 1998. Occurrence of the
entomopathogenic fungi, Metarhizium anisopliae and Beauveria bassiana in soils from
temperate and near-northern habitats. Canadian Journal of Botany 76: 1198-1204.
BIDOCHKA, M.J., MENZIES, F.V. & KAMP, A.M. 2002. Genetic groups of the insectpathogenic fungus Beauveria bassiana are associated with habitat and thermal growth
preferences. Archives of Microbiology 178: 531-537.
130
BIDOCHKA, M.J. & SMALL, C. 2005. Phylogeography of Metarhizium, an insect pathogenic
fungus. In: F.E. Vega & M. Blackwell (Eds.) Insect-Fungal Associations. 28-49. Oxford
University Press Inc., New York.
BING-DA, S. & XING-ZHONG, L. 2008. Occurrence and diversity of insect-associated fungi
in natural soils in China. Applied Soil Ecology 39: 100-108.
BOER, P. 2008. Observations of summit disease in Formica rufa, Linnaeus 1761
(Hymenoptera: Formicidae). Myrmecological News 11: 63-66.
BOTHA, J., HARDIE, D. & BARNES, B. 2004. Natal fruit fly Ceratitis (Pterandrus) rosa:
exotic threat to Western Australia. Department of Agriculture, Western Australia and
ARC Infruitec-Nietvoorbij, Stellenbosch, South Africa. Fact Sheet 6. ISSN 1443-7783.
[Available: http://www.agric.wa.gov.au/objtwr/imported_asssets] accessed: 28/07/2009
BRAGA, G., FLINT, S., MILLER, C., ANDERSON, A. & ROBERTS, D. 2001. Variability in
response to UV-B among species and strains of Metarhizium isolated from sites at
latitudes from 61°N to 54°S. Journal of Invertebrate Pathology 78: 98-108.
BROWNBRIDGE, M., COSTA, S. & JARONSKI, S.T. 2001. Effects of in vitro passage of
Beauveria bassiana on virulence to Bemisia argentifolii. Journal of Invertebrate
Pathology 77: 280-283.
BRUCK, D.J. 2004. Natural occurrence of entomopathogens in Pacific Northwest nursery soils
and their virulence to the black vine weevil, Otiorhynchus sulcatus (F.) (Coleoptera:
Curculionidae). Environmental Entomology 33: 1335-1343.
BRUCK, D. J. 2009. Fungal entomopathogens in the rhizosphere. BioControl DOI
10.1007/s10526-009-9236-7
BUTT, T.M., JACKSON, C.W. & MAGAN, N. 2001. Introduction - fungi as biological control
agents: progress, problems and potential. In: Butt, T.M., Jackson, C.W. & Magan, N.
(Eds.) Fungi as Biocontrol Agents: progress, problems and potential. 1-8. CABI
International, Wallingford, United Kingdom.
CALIFORNIA DEPARTMENT OF FOOD AND AGRICULTURE. 2009. Mediterranean fruit
fly pest profile. [Available: http://www.cdfa.ca.gov/phpps/PDEP/target_pest_disease]
accessed: 07/05/2009.
CAMPBELL, R.K., PERRING, T.M., BARNES, G.L., EIKENBARY, R.D. & GENTRY, C.R.
1978. Growth and sporulation of Beauveria bassiana and Metarrhizium anisopliae on
media containing various amino acids. Journal of Invertebrate Pathology 31: 289-295.
131
CASTILLO, M., MOYA, P., HERNANDEZ, E. & PRIMO-YUFERA, E. 2000. Susceptibility
of Ceratitis capitata Wiedemann (Diptera: Tephritidae) to entomopathogenic fungi and
their extracts. Biological Control 19: 274-282.
CASTRILLO, L.A., GRIGGS, M.H. & VANDENBERG, J.D. 2004. Vegetative compatibility
groups in indigenous and mass-released strains of the entomopathogenic fungus
Beauveria bassiana: likelihood of recombination in the field. Journal of Invertebrate
Pathology 86: 26-37.
CASTRILLO, L., ROBERTS, D. & VANDENBERG, J. 2005. The fungal past, present and
future: germination, ramification, and reproduction. Journal of Insect Pathology 89: 4656.
CHANDLER, D., HAY, D. & REID, A.P. 1997. Sampling and occurrence of
entomopathogenic fungi and nematodes in UK soils. Applied Soil Ecology 5: 133-141.
CHASE, A.R., OSBORNE, L.S. & FERGUSON, V.M. 1986. Selective isolation of the
entomopathogenic fungi Beauveria bassiana and Metarhizium anisopliae from an
artificial potting medium. The Florida Entomologist 69(2): 285-292.
CITRUS GROWERS ASSOCIATION OF SOUTH AFRICA. 2008/9. Annual Report.
[Available:
http://www.cga.co.za/site/files/5438/CGA%20Annual%20Report%202008-
9.pdf] accessed: 08/12/2008.
CITRUS RESEARCH INTERNATIONAL (CRI). 2009. Fruit fly identification chart.
Integrated Pest Management Programme, Nelspruit, South Africa. [Available:
http://www.cri.co.za] accessed: 07/11/2008.
CROSS, J., SOLOMON, M., CHANDLER, D., JARRETT, P., RICHARDSON, P.,
WINSTANLEY, D., BATHON, H., HUBER, J., KELLER, B., LANGENBRUCH, G. &
ZIMMERMANN, G. 1999. Biocontrol of pests of apples and pears in northern and
central Europe: microbial agents and nematodes. Biocontrol Science & Technology 9(2):
125-149.
DE LA ROSA, W., LOPEZ, F.L. & LIEDO, P. 2002. Beauveria bassiana as a pathogen of the
Mexican fruit fly (Diptera: Tephritidae) under laboratory conditions. Journal of
Economic Entomology 95: 36-43.
DE MEYER, M., ROBERTSON, M., PETERSON, A.T. & MANSELL, M.W. 2008.
Ecological niches and potential geographical distributions of Mediterranean fruit fly
(Ceratitis capitata) and Natal fruit fly (Ceratitis rosa). Journal of Biogeography
35(2): 270-281.
132
DE OLIVEIRA, C.N., NEVES, P.M. & KAWAZOE, L.S. 2003. Compatibility between the
entomopathogenic fungus Beauveria bassiana and insecticides used in coffee plantations.
Scientia Agricola 60(4): 663-667.
DE OLIVEIRA, R.C. & NEVES, P.M. 2004. Compatibility of Beauveria bassiana with
acaricides. Neotropical Entomology 33(3): 353-358.
DEPARTMENT OF ENVIRONMENTAL AFFAIRS AND TOURISM (DEAT). 2004.
National Environmental Management: Biodiversity Act, No. 10 of 2004. Cape Town,
South Africa. Government Gazette 467 (No. 26436): 1-84.
DEVI, K.U., SRIDEVI, V., MURALI MOHAN, C.H. & PADMAVATHI, J. 2005. Effect of
high temperature and water stress on in vitro germination and growth in isolates of the
entomopathogenic fungus Beauveria bassiana (Bals.) Vuillemin. Journal of Invertebrate
Pathology 88: 181-189.
DIMBI, S., MANIANIA, N.K. & EKESI, S. 2009. Effect of Metarhizium anisopliae
inoculation on the mating behaviour of three species of African tephritid fruit flies,
Ceratitis capitata, Ceratitis cosyra and Ceratitis fasciventris. Biological Control 50: 111116.
DIMBI, S., MANIANIA, N., LUX, S., EKESI, S. & MUEKE, J. 2003b. Pathogenicity of
Metarhizium anisopliae Sorokin and Beauveria bassiana Vuillemin, to three adult fruit
fly species: Ceratitis capitata, C. rosa var. fasciventris and C. cosyra (Diptera:
Tephritidae). Mycopathologia 156: 375-382.
DIMBI, S., MANIANIA, N.K., LUX, S.A. & MUEKE, J.M. 2003a. Host species, age and sex
as factors affecting the susceptibility of the African tephritid fruit fly species, Ceratitis
capitata, C. cosyra and C. fasciventris to infection by Metarhizium anisopliae. Anzeiger
für Schädlingskunde 76: 113-117.
DIMBI, S., MANIANIA, N., LUX, S. & MUEKE, J. 2004. Effects of constant temperatures on
germination, radial growth and virulence of Metarhizium anisopliae to three species of
African tephritid fruit flies. BioControl 49: 83-94.
DOLINSKI, C. & LACEY, L. 2007. Microbial control of arthropod pests of tropical trees.
Neotropical Entomology 36(2): 161-179.
DOMSCH, K.H., GAMS, W. & ANDERSON, T.H. 2007. Compendium of Soil Fungi. 106109; 266-268; 504-511. IHW-Verlag, Germany.
DU TOIT, W.J. 1998. Fruit Flies. In: E.C.G. Bedford, M.A. Van den Berg & E.A. de Villiers
(Eds.) Citrus Pests in the Republic of South Africa. 229-233. Dynamic Ad: Nelspruit,
South Africa.
133
EKESI, S., DIMBI, S. & MANIANIA, N.K. 2007. The role of entomopathogenic fungi in the
integrated management of fruit flies (Diptera: Tephritidae) with emphasis on species
occurring in Africa. In: S. Ekesi & N.K. Maniania (Eds.) Use of entomopathogenic fungi
in biological pest management. Research Signpost 37/661 (2): 239-274.
EKESI, S. & MANIANIA, N.K. 2000. Susceptibility of Megalurothrips sjostedti
developmental stages to Metarhizium anisopliae and the effects of infection on feeding,
adult fecundity, egg fertility and longevity. Entomologia Experimentalis et Applicata 94:
229-236.
EKESI, S., MANIANIA, N. & LUX, S. 2002. Mortality in three tephritid fruit fly puparia and
adults caused by the entomopathogenic fungi, Metarhizium anisopliae and Beauveria
bassiana. Biocontrol Science and Technology 12: 7-17.
EKESI, S., MANIANIA, N. & LUX, S. 2003. Effect of soil temperature and moisture on
survival and infectivity of Metarhizium anisopliae to four tephritid fruit fly puparia.
Journal of Invertebrate Pathology 83: 157-167.
EKESI, S., MANIANIA, N.K. & LWANDE, W. 2000. Susceptibility of the legume flower
thrips to Metarhizium anisopliae on different varieties of cowpea. BioControl 45: 79-95.
EKESI, S., MANIANIA, N.K., MOHAMED, S.A. & LUX, S.A. 2005. Effect of soil
application of different formulations of Metarhizium anisopliae on African tephritid fruit
flies and their associated endoparasitoids. Biological Control 35: 83-91.
ER, M.K., TUNAZ, H. & GOKÇE, A. 2007. Pathogenicity of entomopathogenic fungi to
Thaumetopoea pityocampa (Shiff.) (Lepidoptera: Thaumatopoeidae) larvae in laboratory
conditions. Journal of Pest Science 80: 235-239.
EVANS, H.C. & WHITEHEAD, P.F. 2005. Entomogenous fungi of arboreal Coleoptera from
Worcestershire England, including the new species Harposporium bredonense.
Mycological Press 4(2): 91-99.
FARIA, M. & WRAIGHT, S.P. 2007. Mycoinsecticides and Mycoacaricides: A comprehensive
list with worldwide coverage and international classification of formulation types.
Biological Control 43: 237-256.
FERNANDES, E.K., RANGEL, D.E., MORAES, A.M., BITTENCOURT, V.R. & ROBERTS,
D.W. 2007. Variability in tolerance to UV-B radiation among Beauveria spp. isolates.
Journal of Invertebrate Pathology 96: 237-243.
FERNANDES, E.K., RANGEL, D.E., MORAES, A.M., BITTENCOURT, V.R. & ROBERTS,
D.W. 2008. Cold activity of Beauveria and Metarhizium, and thermotolerance of
Beauveria. Journal of Invertebrate Pathology 98: 69-78.
134
FILHO, A.B., ALMEIDA, J.E. & LAMAS, C. 2001. Effect of thiamethoxam on
entomopathogenic microorganisms. Neotropical Entomology 30(3): 437-447.
FINNEY, D.J. 1979. Bioassay and the practice of statistical inference. International Statistical
Review 47: 1-12.
FISCHER, N., RUEF, C., EBNÖTHER, C. & BÄCHLI, E.B. 2008. Rhino facial Conidiobolus
coronatus infection presenting with nasal enlargement. Infection 36(6): 594-597.
FOOD AND AGRICULTURE ORGANISATION OF THE UNITED NATIONS (FAO). 2006.
Trades & Markets. Citrus Fruit Annual Statistics. [Available: http://www.fao.org/es/ESC
/en/15/238/ highlight_243.html] accessed: 09/09/2008.
FOOD AND AGRICULTURE ORGANISATION OF THE UNITED NATIONS (FAO). 2009.
Profenofos. [Available: http://www.fao.org/ag/AGP/AGPP/Pesticid/JMPR /Download
/95_eva/ profeno.pdf] accessed: 09/09/2008.
FORESTRY AND AGRICULTURAL BIOTECHNOLOGY INSTITUTE (FABI). 1998. Citrus
Research.
[Available:
http://www.up.ac.za/academic/fabi/citrus/dsc.html]
accessed:
27/07/2009.
FURLONG, M.J. & PELL, J.K. 2001. Horizontal transmission of entomopathogenic fungi by
the diamondback moth. Biological Control 22: 288-299.
FURLONG, M. J., PELL, J.K., ONG, P.C. & SYED, A.R. 1995. Field and laboratory
evaluation of a sex pheromone trap for the autodissemination of the fungal
entomopathogen Zoophthora radicans (Entomophthorales) by the diamondback moth,
Plutella xylostella (Lepidoptera: Yponomeutidae). Bulletin Entomological Research 85:
331-337.
GAN, Z., YANG, J., TAO, N., LIANG, L., MI, Q., LI, J. & ZHANG, K.Q. 2007. Cloning of
the gene Lecanicillium psalliotae chitinase Lpchi 1 and identification of its potential role
in the biocontrol of root-knot nematode Meloidogyne incognita. Applied Microbiology
and Biotechnology 76: 1309-1317.
GILBERT, L. 2003. Graduate Field Course in Rainforest Research. The University of Texas,
Austin. [Available: http://www.utexas.edu/.../zoo384l/sirena/species/fungi/] accessed:
12/01/2009.
GILLESPIE, J.P., BURNETT, C. & CHARNLEY, A.K. 2000. The immune response of the
desert locust Schistocerca gregaria during mycosis of the entomopathogenic fungus,
Metarhizium anisopliae var. acridum. Journal of Insect Physiology 46: 429-437.
135
GOETTEL, M. S. 1995. The utility of bioassays in the risk assessment of entomopathogenic
fungi. In: Proceedings of the Biotechnology Risk Assessment Symposium, Pensacola,
Florida. 6-8 June 1995. pp 2-7.
GOETTEL, M.S., KOIKE, M., KIM, J.J., AIUCHI, D., SHINYA, R. & BRODEUR, J. 2008.
Potential of Lecanicillium spp. for management of insects, nematodes and plant diseases.
Journal of Invertebrate Pathology 98: 256-261.
GOETTEL, M.S., ST LEGER, R.J., RIZZO, N.W., STAPLES, R.C. & ROBERTS, D.W. 1989.
Ultrastructural
localization of a cuticle-degrading protease
produced by the
entomopathogenic fungus Metarhizium anisopliae during penetration of host (Manduca
sexta) cuticle. Journal of General Microbiology 135: 2233-2239.
GROUT, T.G. & STEPHEN, P.R. 2005. Use of an inexpensive technique to compare systemic
insecticides applied through drip irrigation systems in citrus. African Entomology 13(2):
353-358.
HAJEK, A.E. & GOETTEL, M.S. 2007. Guidelines for evaluating effects of entomopathogens
on non-target organisms. In: L.A. Lacey & H.K. Kaya (Eds.) Field Manual of Techniques
in Invertebrate Pathology. 847-868. Springer, Netherlands.
HATTING, J., HAZIR, S., MACUCWA, G., JOOSTE, H. & JANKIELSOHN, A. 2004.
Isolation of entomopathogens from South African soils using the Galleria mellonella-bait
technique. In: Proceedings of the 37th Annual Meeting of the Society for Invertebrate
Pathology, Helsinki, Finland 1-6 August 2004. pp 95.
HATTING, J., HUMBER, R., POPRAWSKI, T. & MILLERS, R. 1999. A survey of fungal
pathogens of aphids from South Africa, with special reference to cereal aphids.
Biological Control 16: 1-12.
HATTING, J.L. & WRAIGHT, S.P. 2007. Optimising bioassay precision with special
reference to the Aphididae and Aleyrodidae. In: S. Ekesi. & N.K. Maniania (Eds.) Use of
entomopathogenic fungi in biological pest management. Research Signpost 37/661(2):
197-237.
HOFFMANN, J.A., KAFATOS, F.C., JANEWAY, C.A. JNR. & EZEKOWITZ, R.A. 1999.
Phylogenetic perspectives in innate immunity. Science 284(5418): 1313-1318.
HU, G. & ST LEGER, R.J. 2002. Field studies using a recombinant mycoinsecticide
(Metarhizium anisopliae) reveal that it is rhizosphere competent.
Applied and
Environmental Microbiology 68(12): 6383-6387.
IBARGUTXI, M.A., MUÑOZ, D., BERNAL, A., RUIZ DE ESCUDERO, I. & CABALLERO,
P. 2008. Effects of stilbene optical brighteners on the insecticidal activity of Bacillus
136
thuringiensis and a single nucleopolyhedrovirus on Helicoverpa armigera. Biological
Control 47: 322-327.
IBRAHIM, L., BUTT, T.M. & JENKINSON, P. 2002. Effect of artificial culture media on
germination, growth, virulence and surface properties of the entomopathogenic
hyphomycete, Metarhizium anisopliae. Mycological Research 106(6): 705-715.
IHARA, F. 2005. Metarhizium anisopliae. Apple Research Station at the National Institute of
Fruit Tree Science, Japan. [Available: http://www.fruit.affrc.go.jp/.../Metarh/ M_aniso.
htm] accessed: 01/05/2008.
INGLIS, G.D., GOETTEL, M., BUTT, T. & STRASSER, H. 2001. Use of hyphomycetous
fungi for managing insect pests. In: T.M. Butt, C.W. Jackson & N. Magan (Eds.) Fungi
as Biocontrol Agents: progress, problems and potential. 23-70. CABI Publishing,
Wallingford, United Kingdom.
INGLIS, G.D., IVIE, T.J., DUKE, G.M. & GOETTEL, M. 2000. Influence of rain and conidial
formulation on persistence of Beauveria bassiana on potato leaves and Colorado potato
beetle larvae. Biological Control 18(1): 55-64.
INGLIS, G.D., JOHNSON, D.L., CHENG, K.J. & GOETTEL, M.S. 1997. Use of pathogen
combinations to overcome the constraints of temperature on entomopathogenic
hyphomycetes against grasshoppers. Biological Control 8: 143-152.
JACKSON, M.A. & JARONSKI, S.T. 2009. Production of microsclerotia of the fungal
entomopathogen Metarhizium anisopliae and their potential for use as a biocontrol agent
for soil-inhabiting insects. Mycological Research 113: 842-850.
JACKSON, T.A. 1999. Factors in the success and failure of microbial control agents for soil
dwelling pests. Integrated Pest Management Reviews 4: 281-285.
JAMES, R.R. & ELZEN, G.W. 2001. Antagonism between Beauveria bassiana and
Imidacloprid when combined for Bemisia argentifolii (Homoptera: Aleyrodidae) control.
Journal of Economic Entomology 94(2): 357-361.
JARONSKI, S. 2007. Soil ecology of the entomopathogenic ascomycetes: a critical
examination of what we (think) we know. In: S. Ekesi & N.K. Maniania (Eds.) Use of
entomopathogenic fungi in biological pest management. Research Signpost 37/661(2):
91-144.
JARONSKI, S.T., FULLER-SCHAEFFER, C., JUNG, K., BOETEL, M. & MAJUMDAR, A.
2007. Challenges in using Metarhizium anisopliae for biocontrol of sugarbeet root
maggot, Tetanops myopaeformis. IOBC/wprs Bulletin 30(7): 119-124.
137
JARONSKI, S.T., GRACE, J.A. & SCHLOTHAUER, R. 2005. Metarhizium anisopliae for
biocontrol of sugarbeet root maggot: constraints and challenges. In: Proceedings of the
33rd Biennial Meeting, American Association of Sugar Beet Technologists, Palm Springs,
California, 2-5 March 2005. pp 185-187.
JAROS-SU, J., GRODEN, E. & ZHANG, J. 1999. Effects of selected fungicides and the timing
of fungicide application on Beauveria bassiana - induced mortality of the Colorado
potato beetle (Coleoptera: Chrysomelidae). Biological Control 15: 259-269.
JOHNSON, D. 2009. Microbial control of grasshoppers, under changing climate and
management: general summary of background and progress. University of Lethbridge,
Canada. [Available: people.uleth.ca/~dan.johnson/metar.htm] access: 02/06/2009.
KEETON, K. 2007. Agathis bishopi (Braconidae): its biology and usefulness as a biological
control agent for false codling moth (FCM), Cryptophlebia leucotreta, on citrus. MSc
Thesis, Rhodes University, Grahamstown, South Africa.
KELLER, S., KESSLER, P. & SCHWEIZER, C. 2003. Distribution of insect pathogenic soil
fungi in Switzerland with special reference to Beauveria brongniartii and Metarhizium
anisopliae. BioControl 48: 307-319.
KHALIL, S.K., SHAH, M.A. & NAEEM, M. 1985. Laboratory studies on the compatibility of
the entomopathogenic fungus Verticillium lecanii with certain pesticides. Agriculture,
Ecosystems & Environment 13: 329-334.
KIRKMAN, W. 2007. Understanding and improving the residual efficacy of the Cryptophlebia
leucotreta
granulovirus
(CRYPTOGRAN).
MSc
Thesis,
Rhodes
University,
Grahamstown, South Africa.
KIRKMAN, W. & MOORE, S.D. 2007. A study of alternative hosts for the false codling
moth, Thaumatotibia (=Cryptophlebia) leucotreta in the Eastern Cape. South African
Fruit Journal 6(2): 33-38
KLINGEN, I., EILENBERG, J. & MEADOW, R. 2002. Effects of farming system, field
margins and bait insect on the occurrence of insect pathogenic fungi in soils. Agriculture,
Ecosystems and Environment 91: 191-198.
KOMAI, F. 1999. A taxonomic revision of the genus Grapholita and allied genera
(Lepidoptera: Tortricidae) in the Palaearctic region. Entomologica Scandinavica 55: 226.
KONSTANTOPOULOU, M.A. & MAZOMENOS, B.E. 2005. Evaluation of Beauveria
bassiana and B. brongniartii strains and four wild-type fungal species against adults of
Bactrocera oleae and Ceratitis capitata. BioControl 50: 293-305.
138
KRUEGER, S.R. & ROBERTS, D.W. 1997. Soil treatment with entomopathogenic fungi for
corn rootworm (Diabrotica spp.) larval control. Biological Control 9: 67-74.
LACEY, L. & SHAPIRO-ILAN, D. 2003. The potential role for microbial control of orchard
insect pests in sustainable agriculture. Food, Agriculture & Environment 1(2): 326-331.
LACEY, L. & SHAPIRO-ILAN, D. 2008. Microbial control of insect pests in temperate
orchard systems: Potential for incorporation into IPM. Annual Review of Entomology 53:
121-144.
LACEY, L.A., FRUTOS, R., KAYA, H.K. & VAIL, P. 2001. Insect pathogens as biological
control agents: do they have a future? Biological Control 21: 230-248.
LAGNAOUI,
A.
&
RADCLIFFE,
E.B.
1998.
Potato
fungicides
interfere
with
entomopathogenic fungi impacting population dynamics of Green Peach Aphid.
American Journal of Potato Research 75: 19-25.
LEE, A.T.C. 2009. Cultivar maturity sheet. Citrus Research International, Nelspruit, South
Africa. [Available: http://www.cri.co.za]
LEE, R.F. 2009. Citrus IPM. In: E. Radcliffe, W. Hutchison & R. Cancelado (Eds.) Integrated
Pest Management: Concepts, Tactics, Strategies and Case Studies. 341-354. Cambridge
University Press, New York.
LI, D.P. & HOLDOM, D.G. 1995. Effects of nutrients on the colony formation, growth and
sporulation of Metarhizium anisopliae (Deuteromycotina: Hyphomycetes). Journal of
Invertebrate Pathology 65: 253-260.
LOC, N.T. & CHI, V.T. 2007. Biocontrol potential of Metarhizium anisopliae and Beauveria
bassiana against diamondback moth, Plutella xylostella. Omonrice 15: 86-93.
LORD, J.C. 2005. From Metchnikoff to Monsanto and beyond: the path of microbial control.
Journal of Invertebrate Pathology 89: 19-29.
MADIGAN, M.T., MARTINKO, J. & PARKER, J. 2005. Brock Biology of Microorganisms
(11th edition). 102-108; 135-165. Prentice Hall, Upper Saddle River, New Jersey.
MAJCHROWICZ, I. & POPRAWSKI, T.J. 1993. Effects in vitro of nine fungicides on growth
of entomopathogenic fungi. Biocontrol Science & Technology 3: 321-336.
MATHER, C. 1999. Agro-commodity chains, market power and territory: re-regulating South
African citrus exports in the 1990‟s. Geoforum 30: 61-70.
MATHER, C. 2003. Regulating South Africa‟s citrus export commodity chain(s) after
liberalisation. Trade and Industrial Policy Strategies (TIPS) Working Paper Series.
[Available: http://www.tips.org.za/research/papers/showpaper.asp? id=558] accessed:
12/01/2009.
139
MATHER, C. & GREENBERG, S. 2003. Market liberalisation in post-apartheid South Africa:
the restructuring of citrus exports after „deregulation‟. Journal of Southern African
Studies 29(2): 394-411.
MATHER, C. & ROWCROFT, P. 2004. Citrus, apartheid and the struggle to re(present)
Outspan oranges. In: A. Hughes & S. Reimer (Eds.) Geographies of Commodity Chains.
156-172. Taylor & Francis Group, London.
McCOY, C.W., STUART, R.J., DUNCAN, L.W. & SHAPIRO-ILAN, D. 2007. Application
and evaluation of entomopathogens for citrus pest control. In: L.A. Lacey & H.K. Kaya
(Eds.) Field Manual of Techniques in Invertebrate Pathology. 567-581. Springer,
Netherlands.
McCOY, C.W. & TIGANO-MILANI, M.S. 1992. Use of entomopathogenic fungi in biological
control: a world view. Pesquisa Agropecuária Brasileira 27: 87-93.
MEYLING, N. 2007. Methods for isolating entomopathogenic fungi from the soil environment.
Department of Ecology, Faculty of Life Sciences, University of Copenhagen, Denmark.
[Available: http://orgprints.org/11200/1/11200.pdf] accessed: 12/01/2008
MEYLING, N. & EILENBERG, J. 2006a. Isolation and characterisation of Beauveria bassiana
isolates from phylloplanes of hedgerow vegetation. Mycological Research 110: 188-195.
MEYLING, N. & EILENBERG, J. 2006b. Occurrence and distribution of soil-borne
entomopathogenic fungi within a single organic agroecosystem. Agriculture, Ecosystems
and Environment 113: 336-341.
MEYLING, N. & EILENBERG, J. 2007. Ecology of the entomopathogenic fungi Beauveria
bassiana and Metarhizium anisopliae in temperate agroecosystems: potential for
conservation biological control. Biological Control 43: 145-155.
MEYLING, N., LUBECK, M., BUCKLEY, E., EILENBERG, J. & REHNERS, S. 2009.
Community composition, host range and genetic structure of the fungal entomopathogen
Beauveria in adjoining agricultural and seminatural habitats. Molecular Ecology 18:
1282-1293.
MIETKIEWSKI, R.T., PELL, J.K. & CLARK, S.J. 1997. Influence of pesticide use on the
natural occurrence of entomopathogenic fungi in arable soils in the UK: field and
laboratory comparisons. Biocontrol Science and Technology 7: 565-576.
MOCHI, D.A., MONTEIRO, A.C. & BARBOSA, J.C. 2005. Action of pesticides to
Metarhizium anisopliae in soil. Neotropical Entomology 34(6): 961-971.
140
MOCHI, D.A., MONTEIRO, A.C., DE BORTOLI, S.A., DORIA, H. & BARBOSA, J.C. 2006.
Pathogenicity of Metarhizium anisopliae for Ceratitis capitata (Wied.) (Diptera:
Tephritidae) in soil with different pesticides. Neotropical Entomology 35(3): 382-389.
MOORE, S.D. 2002. The development and evaluation of Cryptophlebia leucotreta
granulovirus (CrleGV) as a biological control agent for the management of false codling
moth,
Cryptophlebia
leucotreta,
on
citrus.
PhD
Thesis,
Rhodes
University,
Grahamstown, South Africa.
MOORE, S. 2009. Virus sprays for false codling moth - compatibility, timing and molasses.
Cutting Edge, Research News from Citrus Research International 95: 1-2.
MOORE, S., GROUT, T., HATTINGH, V. & HOFMEYR, H. 2008. Thresholds and guidelines
for intervention against citrus pests. South African Fruit Journal Aug/Sept: 77-81.
MOORE, S. & KIRKMAN, W. 2009. Citrus orchard sanitation with emphasis on false codling
moth control. South African Fruit Journal Dec/Jan: 57-60.
NEWTON, P.J. 1998. False codling moth, Cryptophlebia leucotreta (Meyrick). In: E.C.G.
Bedford, M.A. Van den Berg & E.A. de Villiers (Eds.) Citrus Pests in the Republic of
South Africa. 192-200. Dynamic Ad: Nelspruit, South Africa.
NICOLOSI, E., DENG, Z.N., GENTILE, A., LA MALFA, S., CONTINELLA, G. &
TRIBULATO, E. 2000. Citrus phylogeny and genetic origin of important species as
investigated by molecular markers. Theoretical and Applied Genetics 100(8): 1155–1166.
ORR, A. 2003. Integrated pest management for resource-poor African farmers: is the emperor
naked? World Development 31(5): 831–845.
OUEDRAOGO, A., FARGUES, J., GOETTEL, M.S. & LOMER, C.J. 1997. Effect of
temperature on vegetative growth among isolates of Metarhizium anisopliae and M.
flavoviride. Mycopathologia 137: 37-43.
PELL, J.K., MACAULAY, E.D. & WILDING, N.A. 1993. A pheromone trap for dispersal of
the pathogen Zoopthora radicans Brefeld (Zygomycetes: Entomophthorales) amongst
populations
of
the
diamond
back
moth,
Plutella
xylostella
(Lepidoptera:
Yponomeutidae). Biocontrol Science & Technology 3(3): 15-32.
PIRALI-KHEIRABADI, K., HADDADZADEH, H., RAZZAGHI-ABYANEH, M., BOKAIE,
S., ZARE, R., GHAZAVI, M. & SHAMS-GHAHFAROKHI, M. 2006. Biological
control of Rhipicephalus (Boophilus) annulatus by different strains of Metarhizium
anisopliae, Beauveria bassiana and Lecanicillium psalliotae fungi.
Parasitology
Research 100(6): 1297-1302.
141
POPRAWSKI, T.J. & MAJCHROWICZ, I. 1995. Effects of herbicides on in vitro vegetative
growth and sporulation of entomopathogenic fungi. Crop Protection 14(1): 81-87.
POPRAWSKI, T.J., ROBERT, P.H., MAJCHROWICZ, I. & BOVIN, G. 1985. Susceptibility
of Delia antiqua (Diptera: Anthomyiidae) to eleven isolates of entomopathogenic
hyphomycetes. Environmental Entomology 14(5): 557-561
PURWAR, J.P. & SACHAN, G.C. 2006. Synergistic effect of entomogenous fungi on some
insecticides against Bihar hairy caterpillar Spilarctia obliqua (Lepidoptera: Arctiidae).
Microbiological Research 161: 38-42.
PUTERKA, G.J. 1999. Fungal pathogens for arthropod pest control in orchard systems:
mycoinsecticidal approach for pear psylla control. BioControl 44: 183-210.
QUESADA-MORAGA,
E.,
MARTIN-CARBALLO,
I.,
GARRIDO-JURADO,
I.
&
SANTIAGO-ÁLVAREZ, C. 2008. Horizontal transmission of Metarhizium anisopliae
among laboratory populations of Ceratitis capitata (Wiedemann) (Diptera: Tephritidae).
Biological Control 47: 115-124.
QUESADA-MORAGA, E., NAVAS-CORTES, J., MARANHAO, E., ORTIZ-URQUIZA, A.
& SANTIAGO-ALVAREZ, C. 2007. Factors affecting the occurrence and distribution of
entomopathogenic fungi in natural and cultivated soils. Mycological Research 111: 947966.
QUESADA-MORAGA, E., RUIZ-GARCIA, A. & SANTIAGO-ALVAREZ, C. 2006.
Laboratory evaluation of entomopathogenic fungi Beauveria bassiana and Metarhizium
anisopliae against the puparia and adults of Ceratitis capitata (Diptera: Tephritidae).
Journal of Economic Entomology 99: 1955–1966.
RANGEL, D.E., BRAGA, G.U., ANDERSON, A.J. & ROBERTS, D.W. 2005a. Variability in
conidial thermotolerance of Metarhizium anisopliae isolates from different geographic
origins. Journal of Invertebrate Pathology 88: 116-125.
RANGEL, D.E., BRAGA, G.U., ANDERSON, A.J. & ROBERTS, D.W. 2005b. Influence of
growth environment on tolerance to UV-B radiation, germination speed, and morphology
of Metarhizium anisopliae var. acridum conidia. Journal of Invertebrate Pathology 90:
55-58.
RAO, C.U., DEVI, K.U. & KHAN, P.A. 2006. Effect of combination treatment with
entomopathogenic fungi Beauveria bassiana and Nomuraea rileyi (Hypocreales) on
Spodoptera litura (Lepidoptera: Noctuidaeae). Biocontrol Science & Technology 16(3):
221-232.
142
REHNER, S.A. 2005. Phylogenetics of the insect pathogenic genus Beauveria. In: F. Vega &
M. Blackwell (Eds.) Insect-Fungal Associations. 3-27. Oxford University Press Inc. New
York.
REHNER, S. & BUCKLEY, E. 2005. A Beauveria phylogeny inferred from nuclear ITS and
EF1-a sequences: evidence for cryptic diversification and links to Cordyceps
teleomorphs. Mycologia 97(1): 84-98.
RIVER BIOSCIENCE. 2008. CRYPTOGRAN™ for biological control of the false codling
moth.[Available: http://www.riverbioscience.co.za/cycle.html] accessed: 02/03/2008.
RODRIGUES, S., PEVELING, R., NAGEL, P. & KELLER, S. 2005. The natural distribution
of the entomopathogenic soil fungus Metarhizium anisopliae in different regions and
habitat types in Switzerland. IOBC/wprs Bulletin 28(2): 185-188.
SAENZ-DE-CABEZ IRIGARAY, F.J., MARCO-MANCEB, V. & PEREZ-MORENO, I.
2003. The entomopathogenic fungus Beauveria bassiana and its compatibility with
triflumuron: effects on the two-spotted spider mite, Tetranychus urticae. Biological
Control 26: 168-173.
SAFAVI, S.A., SHAH, F.A., PAKDEL, A.K., RASOULIAN, G.R., BANDANI, A.R. &
BUTT, T.M. 2007. Effect of nutrition on growth and virulence of the entomopathogenic
fungus Beauveria bassiana. FEMS Microbiology Letters 270: 116-123.
SALFELDER, K. 2009. Laboratorio de Investigación en Patología, Facultad de Medicina,
Universidad de Los Andes, Venezuela. [Available: http://www.saber.ula.ve/.../figuras
/20D-0003-de.html] accessed: 05/09/2009.
SCHOLTE, E.J., KNOLS, B.G. & TAKKEN, W. 2004. Auto dissemination of the
entomopathogenic fungus Metarhizium anisopliae amongst adults of the malaria vector
Anopheles gambiae sensu strico. Malaria Journal 3:45.
SHAH, P.A. & PELL, J.K. 2003. Entomopathogenic fungi as biological control agents. Applied
Microbiology & Biotechnology 61: 413-423.
SHAH, F.A. & BUTT, T.M. 2005. Influence of nutrition on the production and physiology of
sectors produced by the insect pathogenic fungus Metarhizium anisopliae. FEMS
Microbiology Letters 250: 201-207.
SHAPIRO-ILAN, D.I., REILLY, C.C., HOTCHKISS, M.W. & WOOD, B.W. 2002. The
potential for enhanced fungicide resistance in Beauveria bassiana through strain
discovery and artificial selection. Journal of Invertebrate Pathology 81: 86-93.
SMITH, T.L. 1937. Genetical studies on the wax moth Galleria mellonella Linnaeus. Genetics
23: 115-137.
143
SMITH, D. & PEÑA, J.E. 2002. Tropical Citrus Pests. In: J.E. Peña, J.L. Sharp & M. Wysoki
(Eds.) Tropical Fruit Pests and Pollinators. Biology, Economic Importance, Natural
Enemies and Control. 57- 101. CABI Publishing, Wallingford, United Kingdom.
SOIL CLASSIFICATION WORKING GROUP. 1991. Soil Classification - A Taxonomic
System for South Africa. Department of Agricultural Development, Pretoria. South
Africa.
SOOKAR, P., BHAGWANT, S. & AWUOR OUNA, E. 2008. Isolation of entomopathogenic
fungi from the soil and their pathogenicity to two fruit fly species (Diptera: Tephritidae).
Journal of Applied Entomology 132: 778-788.
STERK, G., HEUTS, F., MERCK, N. & BOCK, J. 2002. Sensitivity of non-target arthropods
and beneficial fungal species to chemical and biological plant protection products: results
of laboratory and semi-field trials. In: Proceedings of the 1st International Symposium on
Biological Control of Arthropods. Hawaii, USA, 14-18 January 2002. pp 306-313.
STIBICK, J. 2008. New pest response guidelines: false codling moth Thaumatotibia leucotreta.
USDA–APHIS–PPQ–Emergency and Domestic Programs, Riverdale, Maryland.
[Available:http://www.aphis.usda.gov/import_export/plants/manuals/emergency/down
loads/nprg-fcm.pdf] accessed: 3/08/2008.
ST LEGER, R. J. 2008. Studies on adaptations of Metarhizium anisopliae to life in the soil.
Journal of Invertebrate Pathology 98: 271-276.
ST LEGER, R.J. & SCREEN, S. 2001. Prospects for strain improvement of fungal pathogens
of insects and weeds. In: T.M. Butt, C.W. Jackson & N. Magan (Eds.) Fungi as
Biocontrol Agents: progress, problems and potential. 219-238. CABI International,
Wallingford, United Kingdom.
TEFERA, T. & PRINGLE, K.L. 2003. Effect of exposure method to Beauveria bassiana and
conidia concentration on mortality, mycosis, and sporulation in cadavers of Chilo
partellus (Lepidoptera: Pyralidae). Journal of Invertebrate Pathology 84: 90-95.
TERBLANCHE, J.H. 1998. Preface. In: E.C. Bedford, M.A. Van den Berg & E.A. de Villiers
(Eds.) Citrus pests in the Republic of South Africa. Institute of Tropical and Subtropical
Crops and Outspan, Nelspruit, South Africa.
THOMAS, M.C., HEPPNER, J.B., WOODRUFF, R.E., WEEMS, H.V., STECK, G.J. &
FASULO, T.R. 2001. Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Insecta:
Diptera: Tephritidae). Entomology and Nematology Department, Florida Cooperative
Extension Service, Institute of Food and Agricultural Sciences, University of Florida.
Document EENY-214 (IN371). DPI Entomology Circulars 4: 230-273.
144
THOMAS, M.B. & READ, A.F. 2007. Can fungal biopesticides control malaria? Nature
Reviews Microbiology 5: 377–383. [Available: http://www.nature.com/.../fig_tab/
nrmicro1638_F1.html] accessed: 07/03/2008.
THOMPSON, S.R. & BRANDENBURG, R.L. 2006. Effect of combining imidacloprid and
diatomaceous earth with Beauveria bassiana on mole cricket (Orthoptera: Gryllotalpidae)
Mortality. Journal of Economic Entomology 99(6): 1948-1954.
THUNGRABEAB, M. & TONGMA, S. 2007. Effect of entomopathogenic fungi, Beauveria
bassiana (Balsam) and Metarhizium anisopliae (Metsch.) on non target insects. KMITL
Science & Technology Journal 7: 8-12.
TOLEDO, J., LIEDO, P., FLORES, S., CAMPOS, S.E., VILLASEÑOR, A. & MONTOYA, P.
2006. Use of Beauveria bassiana and Metarhizium anisopliae for fruit fly control: a
novel approach. In: Proceedings of the 7th International Symposium on Fruit flies of
Economic Importance. Salvador, Brazil. 10-15 September 2006. pp 127-132.
VAN ARK, H. 1995. Introduction to the analysis of quantal response data. Agrimetrics
Institute, Agricultural Research Council, Pretoria, South Africa.
VAN DEN BERG, M.A. 2001. Cryptophlebia leucotreta (Meyrick). In: M.A. Van den Berg,
E.A. de Villiers & P.H. Joubert (Eds.) Pests and Beneficial Arthropods of Tropical and
Non-citrus Subtropical Crops in South Africa. 320-325. ARC-Institute of Tropical and
Subtropical Crops, Nelspruit, South Africa.
VÄNNINEN, I., TYNI-JUSLIN, J. & HOKKANEN, H. 2000. Persistence of augmented
Metarhizium anisopliae and Beauveria bassiana in Finnish agricultural soils. BioControl
45: 201-222.
VARELA, A.M. 2005. International Centre of Insect Physiology & Ecology (ICIPE).
[Available: http://www.infonet-biovision.org/default/ct/205/crops] accessed: 01/02/2009.
VEEN, K.H. & FERRON, P. 1966. A selective medium for isolation of Beauveria tenella and
of Metarrhizium anisopliae. Journal of Invertebrate Pathology 8: 268-269.
VEGA, F.E., GOETTEL, M.S., BLACKWELL, M., CHANDLER, D., JACKSON, M.A.,
KELLER, S., KOIKE, M., MANIANIA, N.K., MONZON, A., OWNLEY, B.H., PELL,
J.K., RANGEL, D. & ROY, H.E. 2009. Fungal entomopathogens: new insights on their
ecology. Fungal Ecology 2: 149-159.
VELDMAN, F. 2001. Value forming attributes of a fruit farm. Citriculture Technical Services,
Nelspruit, South Africa. Technical document no: CK2001/048099/23 [Available:
http://www.cga.co.za] accessed: 10/09/2009.
145
VENETTE, R.C., DAVIS, E.E., DA COSTA, M., HEISLER, H. & LARSON, M. 2003. Mini
Risk Assessment: false codling moth, Thaumatotibia (=Cryptophlebia) leucotreta
(Meyrick) [Lepidoptera: Tortricidae]. University of Minnesota, Department of
Entomology, CAPS PRA. 1-30. [Available: http://www.aphis.usda.gov/ppq/ep/pest
detection / pra /tleucotretapra.pdf] accessed: 01/05/2009.
VESTERGAARD, S., BUTT, T.M., GILLESPIE, A.T., SCHREITER, G. & EILENBERG, J.
1995. Pathogenicity of the hyphomycete fungi Verticillium lecanii and Metarhizium
anisopliae on the western flower thrips, Frankliniella occidentalis. Biocontrol Science &
Technology 5: 185-192
VILCINSKAS, A., MATHA, V. & GOTZ, P. 1997. Effects of the entomopathogenic fungus
Metarhizium anisopliae and its secondary metabolites on the morphology and
cytoskeleton of plasmatocytes isolated from the greater wax moth, Galleria mellonella.
Journal of Invertebrate Pathology 43(12): 1149-1159.
WANG, C., HU, G. & ST LEGER, R.J. 2005. Differential gene expression by Metarhizium
anisopliae growing in root exudate and host (Manduca sexta) cuticle or hemolymph
reveals mechanisms of physiological adaptation. Fungal Genetics and Biology 42: 704718.
WARE, A. 2002. A comparison of fruit fly attractants used in Southern Africa. South African
Fruit Journal 1: 45-47.
WARE, T., RICHARDS, G. & DANEEL, J. 2003. The M3 bait station: a novel method for
fruit fly control. South African Fruit Journal 1: 44-47.
WEEMS, H.V. JNR. 2002. Natal fruit fly, Ceratitis rosa. Florida Department of Agriculture
and Consumer Services, Division of Plant Industry, University of Florida. Document no:
EENY-257. DPI Entomology Circular 51. [Available: http://entomology.ifas.ufl.edu/
creatures/fruit/tropical/natal_fruit_fly.htm] accessed: 28/07/2009.
WELLINGTON, J.H. 1960. Southern Africa - A Geographical Study. 29-50. Cambridge
University Press, London.
WHIPPS, J.M. & LUMSDEN, R.D. 2001. Commercial use of fungi as plant disease biological
control agents: status and prospects. In: T.M. Butt, C.W. Jackson & N. Magan (Eds.)
Fungi as Biocontrol Agents: progress, problems and potential. 9-23. CABI International,
Wallingford, United Kingdom.
WILSON, K., THOMAS, M.B., BLANFORD, S., DOGGETT, M., SIMPSON, S.J. &
MOORE, S.L. 2002. Coping with crowds: density-dependent disease resistance in desert
locusts. PNAS 99 (8): 5471-5475
146
WYSOKI, M., VAN DEN BERG, M.A., ISH-AM, G., GAZIT, S., PENA, J.E. & WAITE,
G.K. 2002. Pests and pollinators of avocado. In: J.E. Peña, J.L. Sharp & M. Wysoki
(Eds.) Tropical Fruit Pests and Pollinators. Biology, Economic Importance, Natural
Enemies and Control. 57- 101. CABI Publishing, Wallingford, United Kingdom.
YANAGAWA, A., YOKOHARI, F. & SHIMIZU, S. 2008. Defense mechanism of the termite,
Coptotermes formosanus Shiraki, to entomopathogenic fungi. Journal of Invertebrate
Pathology 97: 165-170.
YANAGAWA, A., YOKOHARI, F. & SHIMIZU, S. 2009. The role of antennae in removing
entomopathogenic fungi from cuticle of the termite, Coptotermes formosanus. Journal of
Insect Science 9(6): 1-9.
YEE, W. & LACEY, L.A. 2005. Mortality of different life stages of Rhagoletis indifferens
(Diptera: Tephritidae) exposed to the entomopathogenic fungus Metarhizium anisopliae.
Journal of Entomological Science 40: 167-177.
ZARE, R. & GAMS, W. 2008. A revision of the Verticillium fungicola species complex and its
affinity with the genus Lucanicillium. Mycological Research 112: 811-824.
ZEHNDER, G., GURR, G., KUHNE, S., WADE, M., WRATTEN, S. & WYSS, E. 2007.
Arthropod pest management in organic crops. Annual Review of Entomology 52: 57-80.
ZIMMERMANN, G. 1986. The Galleria bait method for detection of entomopathogenic fungi
in soil. Journal of Applied Entomology 102: 213-215.
ZIMMERMANN, G. 2007a. Review on the safety of the entomopathogenic fungus
Metarhizium anisopliae. Biocontrol Science & Technology 17(9): 879-920.
ZIMMERMANN, G. 2007b. Review on the safety of the entomopathogenic fungus Beauveria
bassiana and Beauveria brongniartii. Biocontrol Science & Technology 17(9): 1-44.
147