profiles.uonbi.ac.ke · 2 2 DECLARATION BY THE CANDIDATE This thesis is my original work and has...
Transcript of profiles.uonbi.ac.ke · 2 2 DECLARATION BY THE CANDIDATE This thesis is my original work and has...
1
1
COMPARATIVE STUDIES ON VIRULENCE, GENETIC VARIABILITY
AND MYCOTOXIN PRODUCTION AMONG ISOLATES OF
FUSARIUM SPECIES INFECTING WHEAT
BY
James Wanjohi Muthomi
BSc. (Agric.), MSc. (Plant Pathology)
A thesis submitted in fulfilment of the requirements for the
Degree of
Doctor of Philosophy
In
CROP PROTECTION
Faculty of Agriculture
College of Agriculture and veterinary sciences
University of Nairobi
KENYA
2001
2
2
DECLARATION BY THE CANDIDATE
This thesis is my original work and has not been presented for a degree in any other
university.
Signed …………………………… Date……………………….
James Wanjohi Muthomi
DECLARATION BY THE SUPERVISORS
This thesis has been submitted for examination with our approval as university supervisors.
1st Supervisor
Signed…………………………..….. Date………………………………
Dr. E. W. Mutitu
Department of crop protection,
University of Nairobi,
Kenya.
2nd
supervisor
Signed……………………………… Date………………………………
Prof. Dr. H.-W. Dehne
Institute for Plant Diseases,
University of Bonn,
Germany.
3
3
Dedicated
To
My late mother, Mrs. Gladys Wangui Muthomi
and
My father, Mr. Daniel Muthomi
who sacrificed a lot to build a strong foundation in my life.
4
4
Table of contents List of tables………………………………………………………………………………… vii
List of figures……………………………………………………………………………… viii
List of plates………………………………………………………………………………….. x
List of appendices……………………………………………………………………………. x
Acknowledgements………………………………………………………………………….. xi
Abstract………………………………………………………………………...…………… xii
Chapter 1: Introduction…………………………………………………………………….. 1
Chapter 2: Literature review………………………………………………………………. 3
2.1 Fusarium head blight of wheat………………………………………………………… 3
2.1.1 Economic importance……………………………………………………………….. 3
2.1.2 Aetiology…………………………………………………………………………….. 3
2.1.3 Symptoms……………………………………………………………………………. 4
2.1.4 Epidemiology……………………………………………………………………….. 5
2.2 Fusarium mycotoxins…………………………………………………………………….. 6
2.2.1 Significance…………………………………………………………………………. 6
2.2.2 Mycotoxins produced by Fusarium spp. pathogenic to wheat……………………… 7
2.2.3 Analytical techniques for Fusarium mycotoxins……………………………………. 8
2.3 Polymerase chain reaction ……………………………………………………………….. 9
Chapter 3: Materials and methods……………………………………………………….. 11
3.1 Plant material………………………………………………………………………… 11
3.2 Pathogens……………………………………………………………………………….. 11
3.3 Plant cultivation………………………………………………………………………… 13
3.3.1 Green house pot experiments……………………………………………………… 13
3.3.1.1 Wheat seedling tests………………………………………………………….. 13
3.3.1.2 Wheat ear tests……………………………………………………………….. 13
3.3.2 Field experiment…………………………………………………………………… 14
3.4 Pathogen cultivation and inoculation…………………………………………………… 14
3.4.1 Culture media …………………………………………………………………..… 14
3.4.2 Isolation and maintenance…………………………………………………………. 16
3.4.3 Fusarium identification…………………………………………………………….. 16
3.4.4 Inoculum production and inoculation……………………………………………… 17
3.5 Disease assessment and effect on grain weight…………………………………………. 17
3.5.1 Disease assessment on seedlings…………………………………………………… 17
3.5.2 Assessment of Fusarium head blight………………………………………………. 17
3.6 Fermentation of Fusarium species for mycotoxin production and ergosterol formation.. 18
3.6.1 Liquid culture fermentation……………………………………………………….. 18
3.6.2 Solid culture fermentation…………………………………………………………. 19
5
5
3.7 Analysis of mycotoxins and ergosterol…………………………….…………………... 19
3.7.1 Standards and calibration………………………………………………………….. 20
3.7.2 Mycotoxin analysis………………………………………………………………… 20
3.7.2.1 Extraction and clean up………………………………………………………. 20
3.7.2.2 Mycotoxin detection and quantification by HPLC……………………………. 21
3.7.3 Ergosterol analysis…………………………………………………………………. 22
3.7.3.1 Extraction and clean up………………………………………………………. 22
3.7.3.2 Ergosterol detection and quantification by HPLC……………………………. 23
3.8 In vitro seedling mycotoxin bioassay……………………………………………………. 24
3.9 Polymerase chain reaction-based characterization of Fusarium isolates……………….. 25
3.9.1 Buffers…………………………………………………………………………….. 25
3.9.2 DNA extraction……………………………………………………………………. 25
3.9.3 Random amplified polymorphic DNA (RAPD)-PCR……………………………... 26
3.9.4 Species-specific PCR of Fusarium culmorum and F. graminearum………………. 26
3.9.5 Gel electrophoresis and banding pattern evaluation……………………………….. 26
3.10 Statistical data analysis………………………………………………………………… 27
Chapter 4: Results…………………………………………………………………………. 29
4.1 Fungi isolated from wheat samples …………………………………………… 29
4.2 Detection and quantification of mycotoxins and ergosterol by HPLC ……………… 34
4.3 Effect of substrate and incubation time on mycotoxin production and fungal
biomass formation by Fusarium culmorum and F. graminearum …………………….. 36
4.3.1 Effect of liquid media……………………………………………………………… 36
4.3.1.1 Mycotoxin production in different liquid media ……………………………… 36
4.3.1.2 Time course of mycelial biomass and ergosterol production in liquid media:…36
4.3.1.3 Time course of deoxynivalenol and mycelia dry weight production
in liquid culture……………………………………………………………….. 38
4.3.2 Effect of solid substrates on mycotoxin production and ergosterol formation……. 39
4.3.2.1 Mycotoxin production in different solid substrates…………………………… 40
4.3.2.2 Time course of mycotoxin production and ergosterol formation
in solid fermentation………………………………………………………….. 41
4.4 Effect of different Fusarium species on wheat seedling growth, head blight severity,
grain weight and grain mycotoxin content……………………………………………… 43
4.5 Variability in mycotoxin production and aggressiveness among isolates of
Fusarium culmorum……………………………………………………………………. 46
4.5.1 Variation in aggressiveness and mycotoxin production on wheat ears…………….. 46
4.5.2 Variation in mycotoxin production in vitro……………………………………….. 47
4.5.3 Relationship between mycotoxin production and aggressiveness………………….. 48
4.6 Variability in mycotoxin production and aggressiveness among isolates
of Fusarium graminearum……………………………………………………………. 49
4.6.1 Variation among isolates in aggressiveness on wheat ears………………………… 49
6
6
4.6.2 Differences among the isolates in aggressiveness as determined by
fungal biomass in wheat kernels…………………………………………………… 50
4.6.3 Variation among isolates in mycotoxin production on wheat ears……………….. 52
4.6.4 Variation among isolates in mycotoxin production in vitro……………………….. 53
4.6.5 Relationship among aggressiveness, fungal biomass and mycotoxin production…. 56
4.7 Differences among wheat varieties in susceptibility to head blight due
to Fusarium graminearum……………………………………………………………. 57
4.7.1 Differences among varieties in head blight susceptibility………………………… 57
4.7.2 Variety differences in grain mycotoxin content and fungal biomass………………..59
4.8 Effect of pure Fusarium mycotoxins on wheat seedlings in vitro………………………. 62
4.9 Genetic variability of Fusarium isolates………………………………………………… 65
4.9.1 Species-specific PCR identification of Fusarium culmorum and F. graminearum…65
4.9.2 Differentiation of Fusarium species by random amplified polymorphic
DNA (RAPD)-PCR analysis………………………………………………………. 67
4.9.3 Genetic variation among isolates of Fusarium culmorum as determined by RAPD-
PCR ……………………………………………………………………………….. 75
4.9.4 Genetic variation among isolates of Fusarium graminearum as determined by RAPD-
PCR………………………………………………………………………………... 79
Chapter 5: Discussion…………………………………………………………………….. 84
5.1 Fungal contamination of wheat samples……………………………………………….. 84
5.2 Fungal biomass and ergosterol production in vitro……………………………………. 85
5.3 Dynamics of mycotoxin production in vitro……………………………………………..86
5.4 Mycotoxin production by isolates of Fusarium culmorum and F. graminearum………. 88
5.5 Mycotoxin production on wheat ears………………………………………………….. 90
5.6 Aggressiveness of isolates on wheat ears………………………………………………. 91
5.7 Cultivar susceptibility to Fusarium head blight and mycotoxin content……………….. 95
5.8 RAPD-PCR analysis of Fusarium species……………………………………………… 96
5.9 Polymerase chain reaction (PCR) analysis of isolates of Fusarium graminearum
and F. culmorum………………………………………………………………………..97
Chapter 6: Conclusions ………………………………………………………………… 100
References…………………………………………………………………………………. 102
Appendices…………………………………………………………………………….….. 118
7
7
List of tables
2.1 Tolerance levels of Fusarium mycotoxins in wheat and wheat products.
2.2 Important mycotoxins produced by Fusarium species pathogenic to wheat.
3.1 Fusarium culmorum isolates used in the study.
3.2 Fusarium graminearum isolates used in the study.
3.3 Other Fusarium species used in the study.
3.4 Reference isolates used for comparison.
3.5 Solvent gradient profile for HPLC analysis of mycotoxin.
4.1 Effect of different liquid media on mycotoxin production.
4.2 Deoxynivalenol production by Fusarium culmorum and F. graminearum in solid
substrates.
4.3 Effect of oxygen supply on mycotoxin production and ergosterol formation by F.
culmorum and F. graminearum.
4.4 Mycotoxin production by isolates of different Fusarium species in autoclaved rice and
cracked corn cultures.
4.5 Correlation coefficients among time course of mycotoxin production and ergosterol
formation by F. culmorum.
4.6 Correlation coefficients among time course of mycotoxin production and ergosterol
formation by F. graminearum.
4.7 Effect of Fusarium isolates on wheat seedling growth, head blight severity, grain weight
and grain mycotoxin content. 4.8 Correlation coefficients among reduction in seedling growth, ear aggressiveness parameters and
deoxynivalenol content in kernels.
4.9 Correlation matrix for kernel weight reduction and mycotoxin production by F.
culmorum.
4.10 Correlation coefficients among aggressiveness, deoxynivalenol production and fungal
biomass (ergosterol) formation by isolates of F. graminearum batch 1.
4.11 Correlation coefficients among aggressiveness, deoxynivalenol production and fungal
biomass (ergosterol) formation by isolates of F. graminearum batch 2.
4.12 Correlation coefficients among head blight susceptibility parameters for wheat varieties
after inoculation with F. graminearum.
4.13 Correlation coefficients among head blight ratings, deoxynivalenol and ergosterol
contents in wheat kernels of varieties differing in susceptibility after inoculation with F.
graminearum.
4.14 Effect of wheat variety on head blight rating, pathogen re-isolation rate, deoxynivalenol
and ergosterol content parameters after inoculation with F. graminearum.
4.15 Scores of RAPD-PCR banding patterns for isolates of different Fusarium species.
4.16 Example of Jaccards similarity coefficient matrix among isolates of Fusarium species
calculated from RAPD-PCR banding pattern scores.
4.17 Similarity among 21 F. culmorum isolates determined by RAPD-PCR with 6 primers.
4.18 Similarity among isolates of F. graminearum determined by RAPD-PCR with different
primers.
8
8
List of figures
1. Major genera of fungi contaminating wheat samples collected from different localities in
Kenya.
2. Fusarium species isolated from wheat samples collected in different localities in Kenya.
3. HPLC chromatograms illustrating mycotoxin peaks detected in contaminated wheat and
culture extracts.
4. Ultraviolet absorbance spectra of nivalenol, deoxynivalenol, 3-acetyldeoxynivalenol, 15-
acetyldeoxynivalenol and zearalenone.
5. HPLC chromatogram and UV absorbance spectrum of ergosterol.
6. Time course of ergosterol formation and mycelial biomass production by Fusarium
culmorum in liquid medium.
7. Time course of ergosterol formation and mycelial biomass production by Fusarium
graminearum in liquid medium.
8. Time course of deoxynivalenol and mycelium dry weight production by F. culmorum in
liquid medium.
9. Time course of deoxynivalenol and mycelium dry weight production by F. graminearum
in liquid medium.
10. Time course of mycotoxin production and ergosterol formation by F. culmorum in
autoclaved cracked corn culture.
11. Time course of mycotoxin production and ergosterol formation by F. graminearum in
autoclaved cracked corn culture.
12. Variation in aggressiveness among F. culmorum isolates on wheat ears under field
conditions.
13. Variation in mycotoxin production among 27 isolates of F. culmorum in wheat kernels
after inoculation of ears under field conditions
14. Trichothecene mycotoxin production by 27 isolates of F. culmorum in autoclaved cracked
corn culture.
15. Zearalenone production by 27 isolates of F. culmorum in autoclaved cracked corn culture.
16. Variation in aggressiveness among isolates of F. graminearum, batch 1, as determined by
disease severity (% spikelets bleached) on inoculated wheat ears.
17. Variation in aggressiveness among isolates of F. graminearum, batch 2, as determined by
disease severity (% spikelets bleached) on inoculated wheat ears.
18. Effect of isolates of F. graminearum, batch 1, on ergosterol content in wheat kernels after
inoculation of ears under greenhouse conditions.
19. Effect of isolates of F. graminearum, batch 2, on ergosterol content in wheat kernels after
inoculation of ears under greenhouse conditions.
20. Deoxynivalenol and nivalenol content in wheat kernels from ears inoculated with isolates
of F. graminearum, batch 1, differing in aggressiveness.
21. Deoxynivalenol and nivalenol content in wheat kernels from ears inoculated with isolates
of F. graminearum, batch 2, differing in aggressiveness.
22. Trichothecene mycotoxin production by isolates of F. graminearum, batch 1, in
autoclaved cracked corn cultures.
23. Trichothecene mycotoxin production by isolates of F. graminearum, batch 1, in
autoclaved cracked corn cultures.
24. Zearalenone production by isolates of F. graminearum, batch 1, in autoclaved cracked
corn cultures.
9
9
25. Zearalenone production by isolates of F. graminearum, batch 2, in autoclaved cracked
corn cultures.
26. Differences in shape of disease progress curves of less susceptible and more susceptible
wheat varieties after inoculation with F. graminearum.
27. Differences among 15 wheat varieties in susceptibility to head blight, measured as the area
under disease progress curve (AUDPC), after inoculation with F. graminearum under
greenhouse conditions, 1999.
28. Variation in deoxynivalenol (DON) content in kernels of 15 wheat varieties that differ in
susceptibility to head blight after inoculation with F. graminearum.
29. Ergosterol content in wheat kernels of 15 varieties that differ in susceptibility to head
blight after inoculation of ears with F. graminearum.
30. Comparison between less susceptible and more susceptible wheat varieties in head blight
ratings, fungal biomass (ergosterol) and deoxynivalenol content in kernels after
inoculation with F. graminearum.
31. Effect of varying mycotoxin concentrations on wheat seedling dry weight.
32. Effect of different Fusarium mycotoxins on wheat seedling dry weight.
33. Dendrogram illustrating differentiation of isolates of Fusarium species by cluster analysis
of mean similarity coefficients of RAPD-PCR banding patterns.
34. Examples of dendrograms derived from cluster analysis of RAPD-PCR banding pattern
scores for different Fusarium species.
35. Dendrograms showing clustering of isolates of Fusarium head blight pathogens according
to their respective species after RAPD-PCR analysis.
36. Histogram showing distribution of randomly amplified DNA fragments after RAPD-PCR
with 6 primers and template DNA from isolates of F. culmorum and F. graminearum.
37. Dendrogram showing variation among isolates of F. culmorum after cluster analysis of
mean similarity coefficients of RAPD-PCR banding patterns
38. Examples of dendrograms illustrating differences among isolates of F. culmorum
determined by RAPD-PCR with different primers.
39. Dendrogram showing variation among isolates of F. graminearum as determined by
cluster analysis of mean similarity coefficients of RAPD-PCR banding patterns.
40. Examples of dendrograms illustrating differences among isolates of F. graminearum as
determined by RAPD-PCR with different primers.
10
10
List of plates
1. Cultural characteristics, conidia and fruiting structures of the major fungal genera
contaminating wheat samples in Kenya.
2. Microconidia and conidia of some of the Fusarium species isolated from wheat and used
in the study.
3. Cultural characteristics of Fusarium graminearum and fruiting structures of its perfect
stage, Giberrella zeae.
4. Symptoms of Fusarium infection on wheat seedlings.
5. Symptoms of Fusarium head blight due to F. graminearum and F. culmorum.
6. Growth retardation and stunting of wheat seedlings exposed to trichothecene mycotoxins
7. Gel electrophoresis bands showing identification of Fusarium culmorum isolates by
species specific PCR.
8. Gel electrophoresis bands illustrating identification of Fusarium graminearum isolates by
PCR with species specific primers.
9 and 10. RAPD-PCR banding patterns showing differences among Fusarium species.
11. RAPD-PCR banding patterns showing differences among the major wheat Fusarium head
blight pathogens Fusarium graminearum, F. culmorum, F. avenaceum and F. poae
compared with a non-pathogen, F. oxysporum
12 and 13. Differentiation of Fusarium culmorum isolates by RAPD-PCR banding patterns
obtained with different primers.
14 and 15. RAPD-PCR banding patterns obtained with different primers showing differences
and similarities among isolates of Fusarium graminearum.
List of appendices
1. Structure of trichothecene mycotoxins produced by Fusarium species.
2. Zadoks’ growth stages of small grain cereals.
3. Trichothecene biosynthetic pathway in Fusarium species.
11
11
Acknowledgements
I would like to express my heartfelt gratitude to the German Academic Exchange Service
(DAAD) for financial support in research and stay in the Federal Republic of Germany.
Thanks to Prof. Dr. H.-W. Dehne for recommending me for the scholarship and inviting me to
work in his department. He also introduced me to the interesting area of Fusarium and
mycotoxins in wheat and his strong support during the study is highly appreciated. I am also
grateful to Dr. Mutitu for the keen guidance and support from the formative stages to
completion of my PhD. study.
I am indebted to my loving wife (Pauline Wanjohi) and children (Gladys Wangui and Paul
Muthomi) for the great sacrifices they made. With love, hope and determination, they allowed
me to stay away for two and half years. Without their love and encouragement, it would have
been impossible to complete this work.
Sincere gratitude also goes to Dr. Oerke and Dr. Hindorf of the institute for plant Diseases,
university of Bonn, Germany, for their dedicated support during my stay in Bonn. Dr. Hindorf
initially introduced me to Prof. Dehne and he always made sure that I was comfortable in
Bonn. Thanks to Dr. Oerke for his keen academic support and inspiration. His help in
organizing the ideas during experimental stages and writing up of the thesis is sincerely
appreciated. I am also thankful to staff and colleagues at the institute for their assistance and
friendship.
I am thankful to the following for their valuable contribution to this study: Dr. Udo Bickers of
Aventis Crop Science for training me in HPLC analysis; Dr. E. M. Möller of university of
Hohenheim, Stuttgart, for introducing and training me in PCR technology; Dr. Klaus Pillen of
the institute for Plant Production, university of Bonn for his assistance in photography of PCR
gels; Dr. H. Nirenberg of the Federal Biological Research Centre for Agriculture and Forestry
(BBA), Berlin, for assisting in identification of some of the fungi; Dr. Anja Schade-Shütze for
providing some of the Fusarium isolates and assistance at the initial stages.
Thanks also to the following people who assisted a lot during research work in Kenya: Mrs.
Pauline Wanjohi of the Ministry of Agriculture for organizing collection of wheat samples
and moral support; Ms. Violet Gathara of the National Plant Breeding Centre for assistance in
acquiring certified wheat variety research material and collection of samples; Dr. S. M.
Githiri, Joyce Githinji and Keneth Mbutu for assistance in wheat sample collection; Mr
Francis Gathuma, Mr. Kariuki Mugane, Mr. Simon Muthiora, Mr. A. B. Musyimi and Mrs.
Grace Njenga for their help in carrying out laboratory and greenhouse experiments at Kabete.
Great appreciation also goes to my relatives for their encouragement and moral support.
12
12
Abstract
Isolates of Fusarium species isolated mainly from wheat in Germany and Kenya were
investigated for variation in mycotoxin production, aggressiveness on wheat and PCR-based
DNA characteristics. The isolates were tested for mycotoxin production on wheat ears and in
culture while differences in aggressiveness was tested on wheat ears and seedlings in pot and
field experiments. DNA-based differentiation among the isolates was by random amplified
polymorphic DNA (RAPD) and species-specific polymerase chain reaction (PCR).
Twenty seven isolates of Fusarium culmorum had been collected during a different study on
Fusarium head blight on wheat grown in organic and conventional farming systems in the
Rheinlands, Germany. Fusarium strains from Kenya were isolated from wheat samples
collected in five wheat-growing areas. The major Fusarium species isolated from the wheat
samples in Kenya according to decreasing frequency were F. poae (43%), F. graminearum
(39%), and F. avenaceum (8%). Other Fusarium species identified were F. equiseti, F.
oxysporum, F. camptoceras and F. chlamydosporum. The wheat samples collected in Kenya
were found to be highly contaminated with Epicoccum spp. and Alternaria spp., each being
isolated from 40% and 25% of the wheat kernels plated, respectively. The frequently isolated
Epicoccum sp. was identified as E. purpurascenes. Fusarium sp. were isolated from only 5%
of kernels plated.
Fusarium isolates belonging to different species could be identified by PCR-based DNA
analysis using random primers. RAPD-PCR showed that F. culmorum and F. graminearum
are genetically closely related. Species-specific PCR confirmed identity of some isolates
belonging to these two species. Fusarium culmorum and F. graminearum were more
aggressive on wheat and produced mycotoxins on wheat ears and in culture. Fusarium poae
and F. sporotrichioides caused only mild disease symptoms on wheat seedlings and on ears.
No mycotoxin was detected in cultures of F. poae isolates tested. The negative effect on
wheat seedling growth due to infection with Fusarium species was correlated to head blight
severity (r = 0.9), grain weight reduction (r = 0.92) and grain DON content (r = 0.78).
Mycotoxin analysis was by high pressure liquid chromatography. Trichothecene mycotoxins
nivalenol (NIV), deoxynivalenol (DON), 3-acetyldeoxynivalenol (3-acDON) and 15-
acetyldeoxynivalenol (15-acDON) could be detected together with zearalenone (ZEA) in one
analysis run. The detection limit was 1 µg/g and therefore the method was found to be
especially suitable for trichothecene quantification in grain samples from inoculation
experiments. The recovery rates for the mycotoxins were 55-65% (NIV), 70-85% (DON),
100% (3-acDON and 15-acDON) and 70% (ZEA).
Solid cultures were found to support high mycotoxin yields. Mycotoxin production in liquid
fermentation cultures was low and erratic. Fusarium culmorum isolates produced mycotoxins
NIV, DON, 3-acDON and ZEA while F. graminearum isolates produced NIV, DON, 15-
acDON and ZEA in fermentation cultures. In both F. culmorum and F. graminearum, the
acetylated derivatives of DON (3-acDON and 15-acDON) were produced during the active
growth phases of fungal mycelium while high levels of DON were produced during the
stationary growth phase. Reduced levels of 3-acDON and 15-acDON were detected in the late
stationary growth phases of the mycelium. This supported the proposition that DON is formed
from the acetylated derivatives and may, therefore, explain the low occurrence of 3-acDON
and 15-acDON in naturally contaminated grain. Most of the acetylated forms are most likely
13
13
converted to DON by the time grain is harvested. ZEA was mainly produced during the late
stationary growth phases of the fungal mycelium.
Investigations with twenty seven isolates of F. culmorum and forty two isolates of F.
graminearum in cracked corn cultures showed high variation among the isolates in regard to
amount and type of mycotoxin they produced. Sixteen out of the 27 isolates of F. culmorum
produced NIV (range 26.2 to 220.6 µg/g, mean 96.1 µg/g), 10 produced DON (range 27.2 to
180.8, mean 77.7 µg/g) and 3-acDON (range 64.7 to 548.7 µg/g, mean 230.7 µg/g). Only one
isolate produced all the three mycotoxins (NIV, DON and 3-acDON). Chemotaxonomy of F.
culmorum into NIV- and DON-chemotypes is, therefore, suggested. All but two of the 27
isolates produced detectable amounts of ZEA (range 0.23 to 68.4 µg/g). Thirty five out of the
42 F. graminearum isolates produced DON (range 3.8 to 575 µg/g, mean 149 µg/g), 27
produced 15-acDON (range 3.1 to 81.6 µg/g, mean 23.7 µg/g) and 3 isolates produced NIV
(range 147 to 338 µg/g, mean 247 µg/g). 15-acDON was only detected in cultures of DON-
producing isolates and no isolate was found to produce both NIV and DON. All the F.
graminearum isolates produced ZEA (5 to 637 µg/g, mean 100 µg/g). The production of DON
was correlated to that of the acetyl DON (r = 0.5 to 0.99) for both F. culmorum and F.
graminearum, but production of trichothecene was not correlated to that of ZEA. The results
showed that both NIV- and DON-producing isolates of F. culmorum are prevalent in the
Rheinlands, Germany and contamination of grain with both mycotoxins would be expected in
case of Fusarium head blight epidemic.
The same isolates of F. culmorum and F. graminearum were inoculated on to wheat ears and
the pattern of mycotoxin production in the kernels closely matched that observed in culture.
The isolates of both species varied greatly in aggressiveness to wheat ears, as determined by
head blight severity and grain weight reduction. Grain weight reduction varied from 14 to
61%. Invasion of the kernels by the fungal mycelium, determined as ergosterol content, also
varied among the isolates of F. graminearum (range 49 to 228 µg/g, mean 121.2 µg/g
ergosterol). The type of mycotoxin detected in the wheat kernels for each isolate was similar
to that produced in culture. The more aggressive isolates produced mainly DON while the less
aggressive isolates produced mainly NIV, therefore suggesting involvement of DON in
pathogenicity. The mean NIV and DON content in the kernels was 34.7 µg/g (range 0.95 to
55.2 µg/g) and 40.8 µg/g (range 2.9 to 74.3 µg/g), respectively, for F. culmorum. Both NIV
and DON were detected in kernel samples of three isolates, indicating capability of some F.
culmorum isolates to produce the two mycotoxins simultaneously. The mean levels of NIV
and DON were 8.1 µg/g (range 15.3 µg/g) and 9.6 µg/g (range 2 to 25 µg/g), respectively, for
F. graminearum. Acetyl DON was detected in only one sample for F. culmorum and in none
of the samples for F. graminearum. ZEA was detected for neither of the species. Mycotoxin
content of the kernels was correlated (r = 0.45 to 0.76) to respective production in culture for
F. culmorum isolates but no correlation was found for the F. graminearum isolates.
Exposure of wheat seedlings to varying concentrations of pure DON and its acetyl derivatives
(3-acDON and 15-acDON) in vitro caused 17 to 77% reduction in dry weight. Exposure to 10
ppm (3.3 x 10-5
M) and 20 ppm (6.8 x 10-5
M), resulted in 40% and 48% reduction in seedling
dry weight, respectively. Severe Fusarium head blight infections typically result in 5 to 30
ppm DON contamination of kernels. However, similar concentrations of NIV and ZEA had no
significant effect on the seedlings. Therefore, this evidence and the fact that aggressiveness of
both F. culmorum and F. graminearum was correlated to DON production suggests a
phytotoxic effect of DON on wheat, hence contributing to aggressiveness of a given isolate.
14
14
The high amount of variation among isolates of F. culmorum and F. graminearum was also
found on the RAPD-PCR profiles, indicating that it is genetically based. However, the
variation in RAPD patterns was not related to aggressiveness. There was also no clear
differentiation by the RAPD profiles between NIV- and DON-producing isolates of F.
culmorum but the two chemotypes could be differentiated on the RAPD-PCR profiles of F.
graminearum. Therefore, most of the F. culmorum isolates may be capable of producing both
NIV and DON, as was demonstrated by a few isolates of this species.
Fifteen wheat varieties, commonly grown in Kenya, were found to be susceptible to head
blight after inoculation with F. graminearum in greenhouse pot experiments conducted at
Kabete, Kenya. However, the varieties differed in the level of susceptibility and the resulting
mycotoxin contamination. Susceptibility was determined as head blight severity and amount
of fungal mycelium (ergosterol) in kernels. Head blight severity rating varied from 29% to
68% (mean 54%) of spiekelets bleached while the fungal biomass content varied from 67 µg/g
to 187 µg/g (mean 111 µg/g) of grain. The resulting grain weight reduction ranged from 23 to
57% (mean 44%). Mycotoxin (DON) content in the grain ranged from 5 µg/g to 31 µg/g
(mean 13.5 µg/g). Head blight severity was correlated to fungal biomass and mycotoxin
content of the kernels. Therefore, there is a potential risk of Fusarium head blight epidemic
with resultant yield loss and mycotoxin contamination in Kenya, given favourable weather
conditions. The F. graminearum strains used were isolated from wheat samples collected in
wheat growing areas of Kenya.
15
15
Chapter1: Introduction
Wheat accounts for about a third of the total world cereal production (Gibbon and Pain, 1985),
with a total production of 580 million metric tons in 1996/97 (USDA, 1998). About 2.6
million hectares were cultivated with wheat in Germany in 1996/97 and produced about 19
million metric tons (USDA, 1998). During the same period, 100,000 hectares were planted
with wheat in Kenya, with a total production of 200,000 metric tons. In terms of consumption,
wheat is the main staple cereal in Germany while in Kenya it is second to maize.
Fusaria are widespread in nature occurring both as facultative saprophytes and parasites. The
parasitic Fusarium species are economically important on most agricultural crops and they
affect all vegetative and reproductive organs, causing wilts, rots or blights (Nelson et al.
1981). Wheat can be infected during all growth stages but the most susceptible and
economically important developmental stage is at flowering. Fusarium head blight (scab) has
recently re-emerged as a devastating disease of wheat and barley throughout the world
(McMullen et al. 1997; Windels, 2000). The disease has been reported world-wide wherever
cereals are grown. Undesirable effects of the disease include reduction of grain yield and
quality, losses in livestock fed with contaminated cereals and mycotoxin carry-over to food
products resulting in potential toxicity to humans (Chelkowski, 1998). Many of the species
responsible for Fusarium head blight can also cause wheat seedling blight and brown foot rot
(Parry et al. 1995).
Deoxynivalenol (DON; vomitoxin) is probably one of the most widely distributed Fusarium
mycotoxins (Wood and Trucksess, 1998) but co-occurrence with nivalenol (NIV) and
zearalenone (ZEA) is common (Müller and Schwadorf , 1993; Müller et al. 1997; Mirocha et
al. 1994) and the frequency of mycotoxin producing isolates in natural populations seems to
be high (Gang et al. 1998; Ichinoe et al. 1983; Langseth et al. 1999; Mirocha et al. 1989;
Sugiura et al. 1990). Field surveys have recorded 5 to 80% Fusarium head blight incidence
and severity resulting in as high as 70% scabby kernels and 44 ppm vomitoxin contents in
grain (McMullen et al. 1997) and yield losses ranging from 30% to 70% (Martin and Johnson,
1982). There are a few effective, economical methods for decontamination of trichothecenes
in grains. Contaminated grains can be diverted to non-food use such as fuel ethanol
production, or their toxicity can be reduced by dilution with clean grain. However, sufficient
quantities of non-contaminated grain for blending are not always available.
Fungicide treatment and agricultural management practices only reduce the damage but they
cannot prevent yield and quality losses (Jones, 2000; Matthies and Buchenauer, 2000). In
addition, fungicide costs in relation to return per hectare is a limiting factor (McMullen et al.
1997). Most wheat cultivars are susceptible, none are immune and a few are moderately
resistant (Parry et al. 1995). Durum wheats are more susceptible than common wheats. Less
than 1% of lines tested have shown resistance to Fusarium head blight but this resistance has
to be incorporated into cultivars that are adopted and have the desired yield and quality
characteristics (McMullen et al. 1997).
Mycotoxins are toxic secondary metabolites produced by certain fungi growing on agricultural
commodities in the field and/or during storage (Wood and Trucksess, 1998). About 25% of
the world’s food crops are affected by mycotoxins each year with a substantial impact of
Fusarium species to food contamination (FAO, 1997). The level of contamination with a
particular mycotoxin varies with geographical location, agricultural practices and the
susceptibility of the plants to fungal invasion (Wood and Trucksess, 1998). Because weather
16
16
conditions play a major role in the occurrence of mycotoxins on the grain, the incidence of
contamination also varies from year to year. Mycotoxins can enter the human food chain by
direct contamination resulting from the growth of the fungi on the grain or indirect
contamination due to incorporation of contaminated ingredients into food. Human exposure to
direct contamination can be a significant problem in tropical areas and some underdeveloped
countries where the consumption of mouldy food may be unavoidable because of inadequate
storage and processing facilities or actual shortage of good quality foods. In developed
countries, mouldy food is usually discarded or fed to animals and therefore, indirect exposure
is more significant from the consumption of processed foods made from contaminated grains
or products from animals that have consumed contaminated feed.
Efforts to minimize mycotoxins include monitoring, managing, and controlling quality of
product from the farm to the market place, establishing regulatory limits or guidelines,
utilizing decontamination procedures, and implementing strategies to divert contaminated
grains to feed and other industrial uses (Wilson et al. 1998). Currently, there are no
international regulatory guidelines for mycotoxins in food (wood and Trucksess, 1998).
Chemical and immunological assays, which can routinely be used to rapidly analyze large
numbers of samples, are needed in all monitoring programs. Understanding the biochemistry
of trichothecene biosynthesis is essential in the future research to develop practical and
specific measures to control mycotoxin contamination in grains. Metabolic profiles of
toxigenic Fusarium species are increasingly being considered as important criteria in the
taxonomic system (Ichinoe et al. 1983; Langseth et al. 1999; Lauren et al. 1992; Miller et al.
1991; Sugiura et al. 1990) but a complete list of the possible metabolites of each Fusarium
species not yet available.
Much of the information on Fusarium head blight is on wheat grown in developed countries.
Information on the effects of the diseases on wheat production in Kenya and the species
involved is lacking. This study was therefore carried out with the objective of differentiating
isolates of Fusarium species by their aggressiveness to wheat, mycotoxin production and
polymerase chain reaction-based DNA analysis. The isolates used were mainly isolated from
wheat in Germany and Kenya, but comparison between the two countries was not part of the
main goals of this study.
The specific objectives of the research were:
1. To determine the Fusarium species infecting wheat in Kenya.
2.To differentiate Fusarium isolates on the basis of types and quantity of mycotoxins they
produce.
3. To determine the variability in aggressiveness to wheat among the Fusarium isolates.
4. To differentiate the isolate by polymerase chain reaction-based DNA analysis.
5. To determine susceptibility of wheat varieties grown in Kenya to Fusarium head blight and
mycotoxin accumulation.
17
17
Chapter 2: Literature review
2.1 Fusarium head blight of wheat
Head blight of wheat due to Fusarium spp. is a potential threat to wheat production wherever
the crop is grown. In some countries, it has been ranked as the worst plant disease in the last
ten years (Windels, 2000). The disease can greatly reduce grain yield and quality, and may
indirectly affect animal production due to presence of mycotoxins.
2.1.1 Economic importance
Direct yield reduction results from shrivelled grains, which may be light enough to be
expelled with chaff during harvesting (Bai and Shaner, 1994). Diseased kernels that are not
eliminated with the chaff reduce grain weight because they are light and shrivelled. Indirect
losses occur as mycotoxin contamination, reduced seed germination, seedling blight and poor
stand. Losses in quality result in difficulties in marketing, processing, and export (Windels,
2000). Intense cleaning of infected wheat grain does not guarantee that the wheat will meet
DON guidelines, which vary depending on the market and intended use (FAO, 1997). Grain
not meeting the minimum quality standards, which include the percentage of scabby kernels
and DON guideline, is reduced to lower grade and producer receives a lower price. This has
led to increasingly reduced revenue per harvested hectare (Windels, 2000). In severe
epidemics the producers have to take tough harvesting-marketing-end-use decisions of either
destroying the crop in the field or try to salvage some yield. Farmers adjust to losses of farm
income by applying less fertilizer and pesticides, reducing seeding rates, and skimping on
equipment maintenance. The amount of land planted to wheat is also reduced and replaced
with alternative crops.
Fusarium infection of wheat kernels destroys starch granules, storage proteins and cell walls.
(Dexter et al. 1996; Nightingale et al. 1999). The Fusarium damaged kernels show reduced
semolina yield and the semolina is duller and more red (Dexter et al. 1997). Gluten
functionality (strength) is also reduced because glutenin concentration is less. The flour made
from FDK show decreased dough consistency, resistance to extension and the loafs made
from such flour are substantially reduced in volume (Nightingale et al. 1999).
2.1.2 Aetiology of Fusarium head blight
Methods for identifying Fusarium species are based on morphological characteristics
observed on selective media (Gerlach and Nirenberg, 1982; Nelson et al. 1983; Nirenberg,
1981). The cultural and morphological appearance of Fusarium strains can be highly variable
depending on the culture conditions employed (Yoder and Christianson, 1997). Enzyme-
linked immunosorbent assay (ELISA)-based methods are also available (Abramson et al.
1998; Beyer et al. 1993; Gan et al. 1997). Formae speciales and races are defined by non-
morphological characters such as pathogenicity tests on a range of hosts, (Ouellet and Seifert,
1993). Other methods include determination of vegetative compatibility groups (VCG) by
pairing experiments (Bowden and Leslie, 1992; 1999), isozyme analysis (Yli-Mattila et al.
1996) and examination of the mycotoxin profile (Miller et al. 1991; Langseth et al. 1999;
Mule et al. 1997). In addition, PCR-based techniques have become reliable and highly
suitable tools for identifying Fusarium species and for assessing genetic variation within
collections and population (Burgess et al. 1996; Ouellet and Seifert, 1993; Parry and
Nicholson, 1996; Schilling et al. 1994, 1996a; Turner et al. 1998).
18
18
The common Fusarium species involved in wheat fusariosis are F. graminearum (Gibberella
zeae), F. culmorum, F. crookwellense, F. avenaceum (G. avenacea), F. sporotrichioides, F.
poae, and Microdochium nivale (Bilgrami and Choudhary, 1998; Parry et al. 1995). Fusarium
graminearum predominates in hotter regions of the world while F. culmorum, F. poae and M.
nivale are important in the cooler maritime regions of Northwest Europe (Parry et al. 1995).
Fusarium avenaceum has been isolated over a range of climatic zones. The distribution of F.
crookwellense has not been fully determined but has been found cause head blight in South
Africa (Miller and Trenholm, 1994).
Fusarium graminearum and F. culmorum are the most aggressive, causing severe blighting of
wheat ears (Manka et al. 1985; Stack and McMullen, 1985). F. graminearum produces
macroconidia and the perfect stage, G. Zeae. Gibberella Zeae can be found abundantly as blue
black perithecia developing on the affected host tissue early in the growing season (Sutton,
1982). Two distinct populations of F. graminearum designated as Groups 1 and 2 have been
observed (Burgess et al. 1975; Blany and Dodman, 1988; Francis and Burgess, 1977) with
different sexual behaviour. Perithecia of group 1 are found in the field on a few occasions
(Francis and Burgess, 1977). Perithecia have never been obtained from monoconidial cultures
of group 1 and only rarely in paired cultures. Members of group 1 are, therefore, presumed to
be heterothallic, infertile, or both. Group 2 is primarily airborne and causes scab of small
grains, stalk or ear rot of maize, and carnation crown rot. Members of this group are
homothalic and often produce abundant perithecia in the field and under laboratory conditions
on some media (Francis and Burgess, 1977; Bowden and Leslie, 1999). Sexual reproduction
in Gibberella zeae is reportedly regulated by zearalenone but Blany and Dodman (1988) and
Windels et al. (1989) found no correlation between ZEA production and perithecia formation
in cultures of group 1 and group 2 isolates. Two types within group 2 of G. zeae have been
observed (Cullen et al.1982).
Fusarium culmorum is an imperfect fungus with asexually-formed macroconidia. No sexual
stage is known. Morphological variability in culture is high for both F. graminearum and F.
culmorum (Puhalla, 1981). Sources of genetic variability are sexual recombination in F.
graminearum and mutation, somatic recombination by heterokaryosis in both species. Host
specialisation in both species is low (Miedaner, 1997).
2.1.3 Symptoms of Fusarium head blight
Symptoms caused by different species are almost the same (Bai and Shaner, 1994). Initial
symptoms appear as small, water-soaked brownish spots at the base or middle of the glume, or
on the rachis (Parry et al. 1995). Water soaking and discoloration then spread in all directions
from the point of infection. A salmon-pink to red fungal growth may be seen along the edge of
the glumes or at the base of the spikelet. Premature death or bleaching of spikelets is a
common symptom. Necrosis of most or all of the head probably occurs by invasion and death
of the rachis, rather than spikelet-to-spikelet spread (Stack and McMullen, 1985). The
infection of rachis results in bleaching of all the tissues (scab) above the infection point.
Bleached spikelets are usually sterile or contain partially bleached seed. In a severely infected
field, virtually every head show symptoms of scab.
Only F. graminearum and F. culmorum cause severe blighting of wheat heads resulting in
both visually damaged kernels and latently infected kernels. (Stack and McMullen, 1985;
Schipilova and Gagkaeva, 1997). F. avenaceum, F. acuminatum, F. sporotrichioides F.
19
19
equiseti, and F. poae do not blight the heads but sometimes cause damage to the spikelets
resulting in low disease severity and latent seed infection. Fusarium-infected kernels are
shrivelled, light in weight, and often have pink or red patches. Kernels infected at the time of
heading are chalky white. Kernels of this description are referred to as Fusarium damage
kernels (FDK) or tombstone kernels (Tkachuk et al. 1991). Infection at mass flowering results
in greatest grain weight reduction, highest number of visually infected seeds and least number
of seeds per head (Schipilova and Gagkaeva, 1997). The more mature the heads at
inoculation, the less the visible symptoms, the more the latently infected seeds and the number
of seeds per head may not be affected. Latently infected seeds do not germinate, or produce
non-viable seedlings.
2.1.4 Epidemiology of Fusarium head blight
The pathogen survives in old stalks and ears of maize, and on stubble and debris of wheat,
barley and other cereals (Parry et al. 1995; Sutton, 1982). Infected grain may give rise to
diseased seedlings when used as seed. Propagules that may constitute inoculum are
ascospores, conidia, chlamydospores, and hyphal fragments. Inoculum requires aerial
dispersal to infection sites, and therefore, ascospores and conidia are probably the most
important types inoculum. Inoculum is formed mainly under warm and moist conditions.
Critical weather and plant conditions that are likely to result in head blight epidemic and a
high mycotoxin risk include wheat crop at 50% heads emerged (GS 55) to end of flowering
(GS 69), mean temperature of >18° C for at least 24 h 5 mm rain at this actual day or the
preceding day (Obst, 1997). Incubation periods are shorter when temperatures are warm and
the duration of post-inoculation wetness is increased. Invasion of the rachis may result in
blighting of the entire portion of the spike above the zone of invasion. Rates of colonization of
spikes differ among cultivars. Disease is particularly severe in plants affected by powdery
mildew. Perithecia and conidia develop superficially on the spikelets and rachis in humid
weather. A head blight epidemic is usually completed in one infection cycle and therefore
require production and dispersal of inoculum and infection to coincide with host receptivity.
Spores deposited on wheat spikes germinate and grow initially on anthers and then grow into
the kernels, glumes, or other head parts (Bilgrami and Choudhary, 1998; McMullen et al.
1997; Ribichich et al. 2000). Choline and betaine in anthers have been reported to stimulate
fungal growth (Strange and Smith, 1978). Once the spikelets are infected, the pathogen grows
slowly both intra- and intercellulary throughout the entire kernels (Chelkowski, 1998). The
inter- and intracellular growth of the pathogen in the ovary, lemma and rachis of the infected
spike causes pronounced alterations of cell walls and middle lamella matrices (Kang and
Buchenauer, 2000). Vessels and sieve tubes are occluded (Ribichich et al. 2000). In severe
infections, development of the seed is halted by fungal invasion at about early milk to early
dough stage, depending on the time of initial infection. Fusarium poae is predominantly
infects the glumes rather than the grain and rachis unlike F. culmorum (Doohan et al. 1998).
Kernels originating from infected spikelets contain abundant mycelium well visible on the
surface and in cross section under microscope (Chelkowski, 1998). Moisture is required for
the fungus to sporulate on the heads and infects adjacent plants and late-flowering tillers
(Bilgrami and Choudhary, 1998).
Farming practices that encourage less tillage for control of soil erosion create an environment
rich in crop residues that is conducive to wheat head scab (Sutton 1982; Bai and Shaner, 1994;
Parry et al. 1995; Windels, 2000). Continuous wheat cropping or wheat-maize rotations
provide an increased source of Fusarium inoculum for the development of epidemic (Dill-
20
20
Macky and Jones, 2000; Obst, 1997). Grain maize is more conducive for Fusarium infections
than silage maize because it leaves a higher quantity of crop debris. The Fusarium risk for a
wheat crop following maize is at least four times higher than after a preceding wheat crop.
Rainfall may be needed for perithecia and ascospore formation and maturity on crop debris
(Paulitz, 1996). Perithecial drying during the day, followed by sharp increase in relative
humidity, may provide the stimulus for release of ascospores.
2.2 Fusarium mycotoxins
The Fusarium species causing head blight of wheat are able to produce mycotoxins such as
trichothecene, zearalenone, moniliformin and fusarin C as secondary metabolites (Marasas et
al. 1984). The mycotoxins accumulate in the affected ears as the crop matures in the field. The
mycotoxins are distributed unevenly in individual grain samples as well as in grain bulks
(Chelkowski, 1998).
2.2.1 Significance of Fusarium mycotoxins
Mycotoxins are subject to legislative limits in many countries (Table 2.1) and their presence
in foods, animal feeds, or raw materials, may have serious economic implications in
international trade (Bennett and Keller, 1997; Moss, 1996). Use is restricted when these
regulatory limits are exceeded.
Methods of removing Fusarium damaged kernels include density floatation and specific
gravity tables (Tkachuk et al. 1997). Milling, polishing and treatment of kernels with air
containing high concentrations of chlorine have also been used to reduce DON contamination
(Young et al. 1984). However, none of these methods has found widespread commercial
application. Cleaning of infected wheat grain does not guarantee that the wheat will meet
DON guidelines. This is because cleaned wheat virtually free from Fusarium damaged
kernels, still contain significant levels of DON (Dexter et al. 1997).
Complete removal of mycotoxins in naturally contaminated wheat by milling is impossible
because the mycotoxins are distributed within the kernel (Nightingale et al. 1999; Young et al.
1984). The levels of NIV, DON and ZEA are generally higher in the bran and lowest in the
flour (Lee et al. 1987,1992; Trigo-Stockli et al. 1996; Young et al. 1984). Polishing is
effective in removing DON and ZEA, but not NIV. Milling reduces mycotoxins NIV, DON
and ZEA only in flour fractions intended for human consumption, but increases the toxins in
bran intended for animal feed. Therefore, the bran fractions intended for animal consumption
can contain NIV, DON and ZEA in high concentrations and possible residues of these toxins
in animal products may cause a secondary health hazard to humans. DON is very stable during
baking, surviving temperatures of up to 350° C (Tkachuk et al. 1991). DON levels are
reduced in cooked pasta and noodles because of leaching into the cooking water. The toxin
levels may also be reduced during processing of alkaline products such as Chinese noodles
and tortillas.
Table 2.1 Tolerated levels of Fusarium mycotoxins in wheat and wheat products
Mycotoxin Level (ng/g) Countries
21
21
Deoxynivalenol (DON) 500 to5000 Austria, Canada, Russia, USA
Zearalenone (ZEA) 60 to 1000 Austria, France, Romania, Russia
HT-2 toxin (HT-2) 25 to 100 Canada
T-2 toxin (T-2) 100 Israel, Russia
Diacetoxyscirpenol (DAS) 1000 Israel Source: FAO, 1997.
Presence of DON and other trichothecenes in cereal grains has been reported to be the cause
of several human mycotoxicoses in some countries (Bilgrami and Choudhary, 1998;
Chelkowski, 1998). All animals appear to be sensitive to trichothecenes. The oral toxicity
(LD50) on mouse of the main Fusarium mycotoxins in mg/kg is 46 (DON), 34 (3-acDON and
15-acDON), 19.5 (NIV), 4.5 (Fusarenone-X), 5.2 (T-2), 7.2 (HT-2), 7.3 (diacetoxyscirpenol)
and >5000 (ZEA; Abramson, 1998). Fusarium mycotoxins can cause acute or chronic toxic
effects. Acute effects cause rapid, often fatal diseases while chronic effects may result in
immuno-suppression, cancer, reduced milk production, reduced feed intake and feed
conversion efficiency, vomiting, diarrhoea, weight loss and reproductive problems (Bennett
and Keller, 1997; Miller and Trenholm, 1994). Zearalenone has been associated with
premature puberty (early telarche) in children (Szuets et al. 1997). Poultry appear to be
relatively insensitive to DON and toxin present in feed is not transmitted into eggs or
tissues(El-Banna et al. 1983; Hamilton et al. 1985). Sheep have a high capacity to eliminate
DON by metabolism and urinary excretion (Prelusky et al. 1987).
2.2.2 Mycotoxins produced by Fusarium species pathogenic to wheat
Fusarium trichothecenes are relatively simple alcohols and short chain esters (Desjardins et
al. 1993). They share a tricyclic nucleus named trichothecene (12, 13-epoxytrichothec-9-ene)
and usually contain an epoxide at C-12 and C-13, which is essential for toxicity (Desjardins et
al. 1993; Hesketh et al. 1991). The chemical structure of the different trichothecenes vary in
position and the number of hydroxylations, as well as in the position, number and complexity
of esterifications (Appendix 1). Multiple loci are involved in the genetic regulation of
trichothecene biosynthesis (Desjardins and Beremend, 1987; Hohn et al. 1993). The following
is a schematic presentation of the proposed biosynthetic pathway of deoxynivalenol in F.
culmorum (Desjardins et al. 1993; Hesketh et al. 1991):
22
22
Farnesyl
pyrophosphate
11-hydroxytrichodiene trichodiene isotrichodiol
trichothecene sambucinol
trichodiol
isotrichotriol trichotriol
isotrichodermol 3-acDON
DON
Table 2.2 Important mycotoxins produced by Fusarium species pathogenic to wheat.
DON1 NIV
2 ZEA HT-2 T-2 MON DAS
F. graminearum ++ ++ ++ - - - -
F. culmorum ++ ++ ++ - - - +
F. avenaceum - - - - - + -
F. crookwellense - ++ + - - - +
F. poae - + + - + - +
F. sporotrichioides - - - ++ ++ - +
F. equiseti - + + - + - +
F. tricinctum - - - - - + -
F. acuminatum - - - - - + -
Source: Langseth et al. 1999; Lauren et al. 1992; Marasas et al. 1984; Sugiura et al. 1993.
+ toxin produced; - toxin not produced; 1 produce also as acetylated forms 3-acDON or 15-acDON; 2 can also be
found as acetylated form, fusarenone X.
The important mycotoxins produced by the major wheat head blight pathogens are shown in
table 2.2. Five Fusarium species are reported to be the main producers of trichothecene
mycotoxins (Lauren et al. 1987; Marasas et al. 1984). Fusarium sporotrichioides and F. poae
of the section Sporotrichiella produce mainly T-2 toxin and diacetoxyscirpenol (DAS). Three
species, F. crookwellense, F. culmorum, and F. graminearum of section Discolor produce
mainly deoxynivalenol (DON), nivalenol (NIV), and diacetoxyscirpenol (DAS). Field strains
differ widely in the amount of trichothecenes they produce and genetic crosses between such
strains indicate that multiple genes affect trichothecene yields (Desjardins et al. 1993). Type A
trichothecenes include T-2 toxin, HT-2 toxin, diacetoxyscirpenol (DAS) and neosolaniol
(NEO) while the type B include NIV, DON and their acetylated derivatives fusarenone X, 3-
acDON and 15-acDON (Romer, 1986). Other mycotoxins are moniliformin (MON),
beauvericin (BEA), fusarochromanone (FUCH) and fusarin C (FU-C).
The main toxin associated with Fusarium head blight is DON, although 3-acDON and 15-
acDON may also be present (Abramson, 1998). DON accumulation begins about three days
after infection (Evans et al. 2000; Mirocha et al. 1997), peak about six weeks later, and then
begins to decline and reaches a constant level before harvest (Miller et al. 1983b; Scott et al.
1984). Concentration of DON is highest in chaff (Perkowski et al. 1987), less in the bran, and
least in the endosperm (Lee et al. 1987, 1992; Young et al. 1984). The amount of DON and
other trichothecenes is about 10 times higher in the Fusarium-damaged kernels.
23
23
Plant factors that affect trichothecene biosynthesis include xanthotoxin and other
furanocoumarins induced by a wide variety of plants (Desjardins et al. 1988, 1993). Plant
enzymes have also been suggested to metabolise DON (Miller and Arnison, 1986; Miller et al.
1983b; Scott et al. 1984).
2.2.3 Analytical techniques for Fusarium mycotoxins
Analytical techniques are usually based on extraction into an organic solvent, followed by a
partial purification step (cleanup) and quantification. Extraction solvents are mixed with a
given ratio of a more polar solvent like water, dilute acid, and aqueous solution of salts to aid
the breaking of weak electrostatic bonds which bind some mycotoxins to other substrate
molecules like proteins (Coker et al. 1984). Considerations in analytical schemes include the
chemical nature of the mycotoxin of interest, the molecular weight, and the functional groups
(Wilson et al. 1998). These factors determine the volatility and solubility properties of the
mycotoxin and influence the analytical approach selected. The sample matrix to be extracted
for mycotoxin assay also influences the extraction and analytical method selected.
The clean-up procedures for trichothecenes consist of various steps of solid phase extraction
(SPE) with silica gel, florisil, ion exchange resins, C 18 cartridges, immuno-affinity column
and sometimes also liquid-liquid extraction is used (Krska et al. 1997; Weingaertner et al.
1997). The cleanup procedure can be reduced to a one-step extract purification by use of
multifunctional cleanup columns containing charcoal, celite, and alumina oxide (Romer,
1986; Trucksess et al. 1996; Weingaertner et al. 1997). However, use of this column shows
reduced recovery rate for NIV because the toxin is more polar and therefore strongly adsorbs
on polar sites of the packing material (Krska et al. 1997). Chromatographic quantification
methods for trichothecenes include thin-layer chromatography (TLC; Kamimura et al. 1981;
Romer, 1986; Trucksess et al. 1984), gas chromatography (GC; Lauren and Agnew, 1991;
Mossoba et al. 1996; Walker and Meier, 1998) with electron capture detection (ECD) or mass
spectrometry (MS) and high-performance liquid chromatography (HPLC; Lauren and Agnew,
1991; Stratton et al. 1993; Trucksess et al. 1996; Walker and Meier, 1998) with UV- or
fluorescence detection. Immunological methods such as enzyme linked immuno sorbent assay
(ELISA; Usleber et al. 1992; Wolf-Hall and Bullerman, 1996) or radio immuno assay (RIA)
are also used. High sensitivity is achieved with GC-ECD detection after derivatization
(Walker and Meier, 1998; Wilson et al. 1998). The recovery rates for DON range from 77 to
100% with detection limits of 0.2 to 1 µg/g (HPLC), 40 ng/g (TLC), <30 ng/g (GC), and 200
ppb (ELISA). Total trichothecene content can be analysed by converting all the trichothecenes
to their basic chemical skeleton (Lauren and Agnew, 1991).
2.3 Polymerase chain reaction
Polymersae chain reaction (PCR) is a technique for the in vitro amplification of specific DNA
sequences by the simultaneous primer extension of complementary strands of DNA (Weising
et al. 1995; McPherson et al, 1991). The requirements of PCR are: a buffer, usually
containing Tris-HCl, KCl, and MgCl2; a thermostable DNA polymerase which adds
nucleotides to the 3’ end of a primer annealed to single-stranded DNA; four deoxynucleotides
(dNTPs: dATP, dCTP, dGTP, dTTP) to provide both the energy and nucleosides for the
synthesis of DNA; primer(s) and template DNA. The deoxynucleotides and primers are
present in large excess, so the synthesis step can be repeated by heating the newly synthesized
24
24
DNA to separate the strands and cooling to allow the primers to anneal to their
complementary sequences.
The three essential steps in PCR include melting of the target DNA, annealing of two
oligonucleotide primers to the denatured DNA strands, and primer extension by a
thermostable DNA polymerase (Henson, 1993; Weising et al. 1995). Newly synthesized DNA
strands serve as targets for subsequent DNA synthesis as the three steps are repeated up to 30
to 50 times. The specificity of the method derives from the synthetic oligonucleotide primers,
which base-pair to and define each end of the target sequence to be amplified.
Single stranded oligonucleotides are routinely used to prime amplification reactions (Henson,
1993). DNA sequences within a few hundred bp are usually chosen as primer annealing sites
because shorter sequences on the order of 100 to 1000 base pairs (bp) in length are most
efficiently amplified and easily resolved by agarose gel electrophoresis. Species specific
primers are derived from sequences of the internal transcribed spacer regions of cloned DNA
(cDNA, rDNA or RNA) or sequence characterized amplified regions (SCARs) of the micro-
organism to be detected (Henson, 1993; Parry and Nicholson, 1996; Schilling et al. 1996b).
Specific oligonucleotide primers are typically 18-30 nucleotides in length, about 50% in G+C
content and without complimentary 3’ ends or inverted repeats.
Advantages of species specific PCR in diagnosis include:- organisms need not be cultured
prior to their detection by PCR; high sensitivity, with a theoretical potential to detect a single
target molecule in a complex mixture; and it is rapid and versatile (Henson, 1993; Oullet and
Seifert, 1993). It is also possible to assay for several pathogens simultaneously using multiple
sets of pathogen-specific primer pairs, or ”multiplex PCR”. The major disadvantage is risk of
contamination because it is possible for a single copy of contaminating target sequence to
produce a positive PCR result. In addition, the expense of Taq polymerase, primers, high
quality agarose, and thermalcyclers is prohibitive for routine diagnostic testing of some
phytopathogens, especially those easily cultured and identified.
Random amplified polymorphic DNA (RAPD)-PCR employs short random-nucleotide
sequence oligonucleotide primers (Henson, 1993; Schilling et al. 1994). It is used to
distinguish between organisms or even between different strains or isolates of the same
organism. Specific sequence information of the organism under investigation is not required
and amplification of genomic DNA is initiated at target sites which are distributed throughout
the genome. Nucleotide sequence differences (polymorphism) between individuals is
manifested in the number and length of products that are successfully amplified (Gerlach and
Stösser, 1997). DNA polymorphism results from changes in DNA sequence within the primer
binding region or between the two primer binding sites. Polymorphism is detected by agarose
gel electrophoresis without further manipulation and marker recording can be done by visual
scoring. Fragments amplified in at least one isolate but not in all isolates are considered to be
polymorphic and used as markers for isolate identification. Fragments on the same gel
position in a banding pattern are assumed to be homologous, when amplified through the
same primer (Gerlach and Stösser, 1997). However, this is only true for closely related
genomes, for example, different isolates from one species. Fragments of similar molecular
weight may not necessarily contain homologous sequences and, therefore, a single
amplification product (band) occasionally represents two or more DNA fragments that co-
migrate during electrophoresis. This causes false scoring of bands. By use of several primers,
events of false scoring can be compensated. Because the DNA of plants and other organisms
are also amplified with random primers, plant pathogens must first be purified or cultured
25
25
from their hosts to obtain fingerprints. Hence, RAPD analysis is not useful for direct detection
of plant pathogens in tissue samples, or for organisms that are difficult or impossible to
culture.
Factors affecting primer specificity include primer length, annealing temperature, magnesium
concentration, and secondary structure of target and primer sequences (Henson, 1993). At
high DNA concentrations, phenolics and other enzyme inhibitors present in the DNA
preparation can cause reduced amplification (Schilling et al. 1994; Turner et al. 1998).
Varying annealing temperature from 30 to 38° C has no obvious influence on the RAPD
patterns (Schilling et al. 1994). However, only a narrow range of MgCl2 concentrations (2 to 4
mM) results in reproducible bands. Reproducibility is also influenced by the source of the
polymerase enzyme (Schilling et al. 1994; Gerlach and Stösser, 1997). Use of a negative
control (PCR reagents without target DNA) helps to check purity (absence of contaminating
DNA) of the reaction mix.
26
26
Chapter 3: Materials and methods
3.1. Plant material
The main experimental plant was bread wheat (Triticum aestivum L.). Winter wheat cv.
‘Contra’ was used for studies on Fusarium culmorum aggressiveness while cv. ‘Mbuni’ was
used for F. graminearum isolate aggressiveness studies. Wheat cultivar susceptibility study
was conducted using the following varieties:
Early maturing varieties - ‘k.Fahari’, ‘Nungu’, ‘k.Paka’, and ‘Duma’.
Medium maturing varieties - ‘k.Chiriku’, ‘Ngamia’, ‘k.Popo’, ‘k.Kongoni’, ‘k.Tembo’,
‘Mbuni’ and ‘k.Nyangumi’.
Late maturing varieties - ‘Kwale’, ‘Pasa’, ‘Nata’, and ‘Mbega’.
3.2. Pathogens
The main pathogens used in the study were isolates of Fusarium culmorum (W.G. Smith)
Sacc. and of F. graminearum Schwabe teleomorph Gibberella zeae (Schw.) Petch. The F.
culmorum isolates were kindly supplied by Dr. Anja Shade-Schütze. They had been isolated
from wheat during a 3-year study on Fusarium head blight in the Rheinlands, Germany. The
F. graminearum isolates were isolated from wheat samples from Kenya. Other Fusarium
species included F. avenaceum (Fr.) Sacc. teleomorph G. avenacea R. J. Cook, F. poae
(Peck) Wollenw., F. oxysporum Schlecht. emend. Snyd. & Hans., F. moniliforme Sheldon
teleomorph Gibberella fujikuroi (Sawada) Wollenw. in addition to a few other species
(Tables 3.1, 3.2, 3.3 and 3.4). Reference isolates were kindly supplied by Dr. E. M. Möller,
University of Hohenheim, Germany.
Table 3.1 Fusarium culmorum isolates from wheat grown in the Rheinlands, Germany
Isolate code Test done* Isolate code Test done*
C5 EA, T, P C38 EA, T, P
C11 EA, T, P C31 EA, T, P
C33 EA, T, P C36 EA, T, P
C27 EA, T, P C32 EA, T, P
C26 EA, T, P C24 EA, T, P
C25 EA, T, P C40 EA, T, P
C34 EA, T, P C35 EA, T, P
C28 EA, T, P C39 EA, T, P
C42 EA, T, P C20 EA, T, P
C21 EA, T, P C29 EA, T, P
C8 EA, T, P C22 EA, T, P
C6 EA, T, P C23 EA, T, P
C37 EA, T, P C30 EA, T, P
C41 EA, T, P
* Isolate tested for: EA = ear aggressiveness, T = mycotoxin production, P = PCR
Table 3.2. Fusarium graminearum isolates from wheat samples collected in Kenya
27
27
Isolate code Test done* Isolate code Test done*
K1 EA, T, P K66 EA, T, P
K2 EA, T, P K68 EA, T, P
K3 EA, T, P K70 EA, T, P
K4 EA, T, P K71 EA, T, P
K5 EA, T, P K72 EA, T, P
K6 EA, T, P K74 EA, T, P
K8 EA, T, P K92 EA, T, P
K15 EA, T, P K93 EA, T, P
K17 EA, T, P K108 EA, T, P
K18 EA, T, P K109 EA, T, P
K31 EA, T, P K116 EA, T, P
K33 EA, T, P K118 EA, T, P
K42 EA, T, P K133 EA, T, P
K52 EA, T, P K134 EA, T, P
K53 EA, T, P K135 EA, T, P
K57 EA, T, P K138 EA, T, P
K58 EA, T, P K139 EA, T, P
K59 EA, T, P K140 EA, T, P
K60 EA, T, P K144 T, P
K61 EA, T, P K145 EA, T, P
K64 EA, T, P K151 EA, T, P
* Isolate tested for: EA = ear aggressiveness, T = mycotoxin production, P = PCR.
Table 3.3. Other Fusarium species use in the study
Isolates from wheat Isolates from maize kernels
Isolate code Species Test done Isolate code Species Test done*
K23 F. poae P BMM1 F. graminearum SA, EA, T, P
K63 F. avenaceum P B9J4 F. graminearum SA, EA, T, P
K79 F. avenaceum P B9F12 F. graminearum SA, EA, T, P
K97 F. poae P 2.2a Fusarium sp. SA, EA, T, P
K98 F. poae P 2.2b F. graminearum SA, EA, T, P
K105 F. graminearum P B7M1 Fusarium sp. SA, EA, T, P
K117 F. avenaceum P 6.4 F. sporotrichioides SA, EA, P
K122 F. avenaceum P B1M5 F. poae SA, EA, T
K128 F. avenaceum P B1M1 F. poae SA, EA,
K136 F. chlamydosporum P B11M4 F. poae SA, EA,
K141 Fusarium sp. P
K175 F. poae P
* Isolate tested for: SA = seedling aggressiveness, EA = ear aggressiveness, T = mycotoxin production, P = PCR
Table 3.4 Reference isolates used for comparison
Isolate code Species Test done*
28
28
Fc3.1 F. culmorum SA, EA, T, P
Fg7.1 F. graminearum SA, EA, T, P
FPIwt F. poae SA, EA, T, P
Foxy3 F. oxysporum P
F.mon. F. moniliforme P * Isolate tested for: SA = seedling aggressiveness, EA = ear aggressiveness, T = mycotoxin production, P = PCR
3.3 Plant cultivation
3.3.1 Green house pot experiments
3.3.1.1 Wheat seedling tests
The experiment was carried out with wheat and maize seedlings. Seedling aggressiveness tests
were carried out according to Chelkowski and Manka (1983) Manka et al. (1989) and
Mesterhazy (1978). Healthy seeds were surface sterilised in 5% sodium hypochlorite for 3
minutes and rinsed 3 times in sterile distilled water. The surface sterilised seeds were soaked
in the inoculum (105 spores/ml) of Fusarium species for 24 hours to initiate infection. Control
seeds were soaked in distilled water. Steam-sterilised peat was ¾ filled in 7.5 cm diameter
plastic pots and the surface was irrigated with 20 ml of the spore suspension. Control pots
were irrigated with distilled water. Ten seeds were evenly placed on the infested medium
surface and covered with 2 cm layer peat. The medium surface was dampened with 10 ml
distilled water. The pots were arranged in a complete randomised design on raised greenhouse
benches, each treatment replicated 4 times. Regular irrigation was done to ensure good
germination and growth of the seedlings
3.3.1.2 Wheat ear tests
Seeds were planted in forest soil/farm yard manure medium (2:1 v/v) in 10 l., 22 cm diameter,
pots (25 seeds/pot). The plants were allowed to grow outside the greenhouse until flowering
(GS 61, Appendix 2; Zadoks et al. 1974) to simulate natural conditions. Fertilisation was done
at different growth stages as follows: urea (46% N) 5 g/pot after germination (GS 10), N-P-K
(20-20-0) 5 g/pot at tillering (GS 22), and urea (46 % N) 5g/pot at booting stage (GS 41). Leaf
rust (Puccinia recondita) was controlled by spraying with Bayleton
(1 g/l) at tillering (GS
22) while leaf chewing insects and aphids were controlled with dimethoate
(1 ml/l). The
respective days to flowering for each variety was determined in a preliminary experiment to
enable synchronisation of the flowering dates for uniform treatment at inoculation. The
flowering dates were synchronised by early-planting of the late maturing varieties and late-
planting of the early maturing varieties. After inoculation, the plants were allowed to mature
inside greenhouse to ensure uniform environmental conditions and to avoid destruction by
birds. The experiment was conducted in 2 greenhouse cropping cycles for the isolate
aggressiveness and 3 cropping cycles for variety studies.
3.3.2 Field experiment
Wheat ears of cv. ‘Contra’ was spray-inoculated at mid flowering (GS 65) with isolates of F.
culmorum using spore suspension with 5 x 105 spores /ml. Two replicate mini plots ( 1 m)
were inoculated per isolate using a randomized design. The inoculated plots were separated by
29
29
at least 2 m of non-inoculated wheat to minimize plot-by-plot interference. High humidity for
infection was assured by covering the plants with plastic bags for 12 h. Percent head blight on
the ears was assessed visually at 14 days post inoculation (GS 73) and 28 days post
inoculation (GS 85). At ripening (GS 92), the ears were harvested and threshed using a mini-
plot thresher. Grain weight for each isolate treatment was determined by measuring the 1000
kernel weight five times per sample.
3.4 Pathogen cultivation and inoculation
3.4.1 Culture media
The stated recipes are per litre of distilled water. All the culture media were autoclaved at
121°C for 20 minutes and dispensed into 9cm diameter disposable petri dishes. Unless where
stated, the media was allowed to cure for 2 days before use.
Czapek-Dox-Iprodione-Dicloran agar-CZID (Abildgren et al, 1987)
Czapek Dox agar 35 g
CuSO4.5H2O 0.005 g
ZnSO4.7H2O 0.01 g
Chloramphenicol 0.05 g
Dicloran 1 ml of 0.02 g dissolved in 10 ml ethanol
Agar 10 g
After autoclaving and cooling to 50°C, the following were added:
Penicillin 50 mg
Aureomycin (tetracycline) 50 mg
Streptomycin 50 mg
Rovral (Iprodione) 1 ml of 0.06 g in 10 ml water
Synthetic Nutrient Agar-SNA (Nirenberg, 1981)
KH2PO4 1.0 g
KNO3 1.0 g
MgSO4.7H2O 0.5 g
KCl 0.5 g
Glucose 0.2 g
Sucrose 0.2 g
Agar 20.0 g
Low strength PDA amended with mineral salts
PDA 17 g
KH2PO4 1.0 g
KNO3 1.0 g
MgSO4.7H2O 0.5 g
KCl 0.5 g
Agar 10 g
Carnation leaf agar - CLA (Fisher et al., 1982; Stack & McMullen, 1985; Tschanz et al.,
1975):
Fresh young and healthy carnation (Dianthus caryophilus L.) leaves were cut into 5 mm2
pieces and dried in oven at 45 to 55°C for 2 h. The dried leaf pieces were soaked for 5 minutes
in 70% ethanol and then allowed to dry in lamina flow cabinet. Several leaf pieces were
placed in petri dishes and floated on 2% water agar cooled to 45°C. The media was left at
30
30
room temperature for 3-4 days before use to check for growth of possible contaminants from
the leaf pieces.
Water agar amended with wheat straw (Manka et al., 1989):
Straw was obtained from dry greenhouse-grown wheat plants, cut into 5 mm2 pieces and
soaked in 70% ethanol for 10 min. Several straw pieces were placed on 2% water agar cooled
to 50°C in disposable petri dishes. The media was left at room temperature for 3-4 days before
use to check for growth of possible contaminants from the straw pieces.
Autoclaved cereal media- course ground maize (Blany and Dodman, 1988), wheat kernels
(Miedaner, 1996) and rice (Abouzied & Pestka, 1986; Sugiura et al, 1990):
Fifteen grams substrate was placed in 100 ml Erlenmeyer flask, moistened with 7.5 ml
distilled water and allowed to absorb the water for 3 hours. Flasks were shaken to uniformly
mix the moist substrate, closed with cotton plugs and autoclaved. After autoclaving, the media
was allowed to stand at room temperature for 24 hours and then re-autoclaved.
Malt extract broth (Matthies and Buchenauer, 1995)
Malt extract 15g
Peptone 0.75g
Maltose 12.75g
Dextrin 2.75g
KH2PO4 0.75g
NH4Cl 1g
Na2HPO4 28.4g
Citric acid 21g
Glucose-Yeast-Extract-Peptone (GYEP) broth (Abouzied and Pestka, 1986; Beremand, 1987;
Miller et al, 1991):
Glucose 10 g
Yeast extract 1g
Peptone 1 g
Yeast extract-mineral salts broth (modified from SNA):
Sucrose 10 g
Yeast extract 1 g
KH2PO4 1 g
KNO3 1 g
MgSO4.7H2O 0.5 g
KCl 0.5 g
3.4.2 Isolation and maintenance
Fusarium species were isolated from samples collected in Germany and Kenya. Strains of
Fusarium culmorum had been isolated from wheat in an earlier a 3-year study of Fusarium
head blight in conventional and organic farming systems in the Rheinlands, Germany.
31
31
Isolation medium was Czapek-Dox-Iprodione-Dichloran agar (CZID) which is selective for
Fusarium. Sub-samples of wheat kernels from each sample were surface-sterilised for three
minutes in 5% sodium hypochlorite solution containing few drops of Tween 20, rinsed twice
with sterile distilled water and dried on sterile paper towels. In each petri dish, 10 kernels
were evenly plated on the molten agar medium cooled to 45°C. From each sub-sample 100
kernels were plated. The plates were incubated at room temperature (20°C) for 7 to 14 days,
after which the number of kernels showing fungal infection were recorded. Fungal colonies
were sub-cultured on potato dextrose agar (PDA) and identified to genus level following
descriptions of Barnet and Hunter (1972), Zillinsky (1983) based on cultural characteristics
and spore morphology. Fungal genera growing from the kernels were recorded. The isolation
frequency (Fq) of genera and species were calculated as follows (Gonzalez et al.1999).
Fq (% = No. of samples with occurrence of a genus or species x 100
Total No. of samples.
The Fusarium isolates were maintained on sterilized soil. Repeated sub-culturing on nutrient
rich media like PDA was minimized to reduce genetic changes of the isolates. Regeneration
from soil was done on SNA medium.
3.4.3 Fusarium identification
Each Fusarium colony growing from the kernels was sub-cultured on PDA and SNA. The
cultures were incubated under near UV black light for 14-21 days to induce sporulation. The
cultures were identified to species level according to Nelson et al. (1983). PDA was used for
gross morphological appearances and colony coloration. Cultural characteristics used for
identification included rate of growth, presence of aerial mycelium, colour of aerial mycelium
and colour of colony from below. Cultures grown on SNA were used for microscopic
identification based on conidia, conidiophore and chlamydospore observations. Microscopic
observations were made using distilled water as mounting medium for examination of
conidia. Identification features used were macroconidia morphology (size, shape of basal and
apical cells), microconidia (present or absent, whether produced in chains or false heads,
shape), type of conidiophores, and chlamydospores (present or absent,
arrangement).Photographs were taken on a Leitz research microscope with 40x objective and
10x ocular. The morphological identification of the Fusarium species was confirmed using
PCR-based markers
Perithecia production by F. graminearum isolates was tested by growing the isolates on
carnation leaf agar (CLA) and on water agar amended with wheat straw. The media was
inoculated with agar plugs cut from Fusarium graminearum cultures on SNA. Plates were
incubated for 3 to 4 weeks days under near UV black light. Production of perithecia was
determined by visual and low magnification.
3.4.4 Inoculum production and inoculation
Fusarium isolates were cultivated for 21 days under near UV light on low strength PDA
amended with mineral salts. Conidia suspension was harvested by washing the cultures with
distilled water and scrapping the mycelial growth to dislodge the conidia. Each isolate was
harvested separately. The suspension was passed through double layer cheese cloth to separate
32
32
mycelia from the conidia. Conidia concentration was determined by use of a haemocytomter
and adjusted to 5 x 105
spores/ml. For variety inoculation, conidia suspensions from 10
pathogenic isolates were mixed before conidia concentration adjustment. A few drops of
Tween20 were added to the adjusted conidia suspension to ensure uniform dispersion of the
conidia.
Inoculation was done at 50% flowering (GS 65). For isolate aggressiveness tests, each isolate
was inoculated separately while mixed isolates were used on each cultivar for the variety
experiments. The ears were uniformly spray-inoculated using a hand sprayer, ensuring that all
spikelets were exposed to inoculum. Ears of the control plants were treated similarly but
distilled water was used in place of spore suspension. The ears were then covered with
polythene bags for 48 h to provide 100% relative humidity for infection.
3.5 Disease assessment and effect on grain weight
3.5.1 Disease assessment on seedlings
Disease was assessed on the basis of percentage germination, seedling mortality, shoot length
and plant dry weight. Percentage of seedlings emerged was recorded after germination (GS
10) while percentage seedling mortality and shoot length were recorded 3 weeks after
emergence (GS 20). Shoot length was measured from stem base to the base of the top-most
fully expanded leaf. The potting medium was then washed off the roots, the plants separated
into roots and shoots and dried in oven at 110°C for 48 hours for dry weight determination. In
data analysis, a mean of 10 plants for each replicate was used and in cases where some plants
were dead, the recorded values of the surviving plants was divided by 10 (total number of
plants planted in each replicate). Percentage reduction in emergence, shoot length root and
shoot dry weights was calculated as follows:
X = (control mean - treatment mean) x 100
control mean
3.5.2 Assessment of Fusarium head blight
Head blight rating was initiated with the first appearance of symptoms. Ten average sized ears
per plot (40 ears per treatment) were tagged and the disease severity on these ears was
recorded every 5th day until yellow ripening (GS 92). Disease assessment was done by visual
assessment of the proportion of spikelets bleached based on a 0 - 9 scale (Miedaner et al,
1996) as follows:
1 = No symptoms
2 = <5% of spikelets bleached
3 = 5 to 15% of spikelets bleached
4 = 16 to 25% of spikelets bleached
5 = 26 to 45% of spikelets bleached
6 = 46 to 65% of spikelets bleached
7 = 66 to 85% of spikelets bleached
8 = 86 to 95% of spikelets bleached
9 = 96 to100% of spikelets bleached
33
33
Isolate aggressiveness was determined on the basis of percentage spikelets bleached and
reduction of kernel weight. For each isolate, a mean disease severity rating was calculated
from 5 single ratings recorded over the assessment period. Variety susceptibility was
measured as disease severity based on area under disease progress curve, in addition to disease
severity rating based on percentage spikelets bleached. The area under the disease progress
curve (AUDPC) was calculated for each replicate of each variety using the following formula
(Shaner and Finney, 1977):
n
AUDPC = (Yi+1 +Yi)/2 Ti+1 - Ti i = 1
where Yi = disease severity at the ith observation, Ti = time (days) at the ith observation, and
n = total number of observations.
At maturity (GS 92), the ears of each replicate for all the treatments were harvested separately
and threshed to determine the 1000 and 10-ear kernel weight kernels. Percentage kernel
weight reduction as compared to the controls was calculated for each treatment. Percentage re-
isolation of the pathogen from the kernels was determined by plating 100 kernels per
treatment, 10 kernels/plate for each treatment on CZID agar. The rest of the kernels was stored
at –20°C until mycotoxin and ergosterol analysis.
3.6 Fermentation of Fusarium species for mycotoxin production and ergosterol
formation
3.6.1 Liquid culture fermentation
Liquid culture fermentations were used for mycotoxin production, mycelia production for
fungal biomass studies and for DNA extraction. Twenty five millilitres of each media were
dispensed in 100 ml Erlenmeyer flasks, closed with cotton plugs and autoclaved at 121°C for
20 minutes. The media was cooled to room temperature and inoculated with 10 mm2 agar
discs cut from actively-growing Fusarium cultures on SNA. Duplicate flasks were inoculated
in each case.
In mycotoxin production experiments, one isolate of F. culmorum (C41) and one isolate of F.
graminearum (BMM1) were tested on glucose-yeast extract-peptone broth (GYEP), yeast
extract-mineral salts broth and malt extract broth. The cultures were incubated at room
temperature (20 + 5 °C) on a mechanical shaker (65 cycles/min.) for up to 25 days after which
mycotoxin content was analysed. The media showing highest mycotoxin yield was selected
for time course mycotoxin production studies. In the time course experiments, duplicate flasks
of the 2 isolates were analysed every 5 days up to 25 days post inoculation.
Time course ergosterol and mycelial biomass formation studies were done on malt extract
broth using F. culmorum isolate C41 and F. graminearum isolate BMM1. Cultures were
incubated at room temperature (20 + 5 °C) in stationary conditions. Ergosterol and mycelial
dry weight analyses were done for duplicate flasks of each isolate at day 5, 10, 15, 20, 25, 30,
and 35. Mycelia was separated from liquid medium by vacuum filtration and was washed with
distilled water. Mycelia pellet was freeze dried, weighed and stored at -20°C until ergosterol
analysis.
34
34
3.6.2 Solid culture fermentation
Solid state fermentations were used for mycotoxin and fungal biomass (ergosterol)
production. Solid substrates tested for mycotoxin production were autoclaved course ground
maize, wheat kernels and rice. Duplicate flasks were prepared for each substrate. After
cooling to room temperature, each flask was inoculated with 2 x10 mm diameter agar plugs
cut from actively growing cultures on SNA. The flasks were incubated at 25°C for 14 to 21
days in the dark, after which the culture clumps were broken up and dried in oven at 50°C for
12 h. The duplicate cultures were combined, ground into a fine powder and stored at -20°C
until mycotoxin extraction. The substrate showing the highest mycotoxin yield was adopted
for further experiments using different isolates of F. graminearum and F. culmorum and for
time course studies. A total of 27 isolates of F. culmorum and 42 isolates of F. graminearum
were tested for mycotoxin production in vitro on autoclaved cracked corn. A reference isolate
of F. culmorum (Fc3.1) and of F. graminearum (Fg7.1) was included as positive control.
Negative control was non-inoculated autoclaved corn which was similarly treated as the
inoculated treatments.
For time course mycotoxin and ergosterol formation, two mycotoxin-producing isolates, one
F. culmorum (C41) and one F. graminearum (BMM), were used. The cultures were incubated
at 25°C. After every 5 days, duplicate cultures for each species were combined and dried in
oven at 50°C for 12 h. The samples were ground into a fine powder, divided into 2 portions
and stored at -20°C until analysed. One portion of each sample was analysed for mycotoxin
while the other portion was reserved for ergosterol analysis. Analysis was done for cultures
terminated at 5, 10, 15, 20, 25, 30, 35, and 40 days after inoculation. Non-inoculated substrate
but treated similarly was used as control.
3.7 Analysis of mycotoxins and fungal biomass (ergosterol)
Samples used were from wheat ear inoculation experiments and from fermentation cultures.
For the ear inoculation samples, equal amounts of kernels were taken from each replicate of
each experiment and mixed to make the analytical sample. At least 20g were obtained for each
treatment. Samples from non-inoculated ears were analysed as negative controls while
positive controls comprised of clean wheat flour spiked with known amounts of standards.
The samples were ground into a fine powder using a laboratory mill and divided into 2
portions, one for mycotoxin analysis and the other for ergosterol analysis. The samples were
stored at -20°c until analysed.
Mycotoxin and ergosterol analysis was carried out using high pressure liquid chromatography
(Hewlett-Packard HP1090) supplied with UV diode-array detector. The column compartment
heater was set at 40°C. All instrument parameters were controlled by a HPChem software.
Solvents used were 99.8% HPLC grade acetonitrile (CH3CN), ethylacetate (CH3CO2C2H5),
methanol (CH3OH), ethanol (C2H6O) and n-hexane CH3(CH2)4CH3 from Sigma-Aldrich
Chemie (Steinheim, Germany). De-mineralised water was used where applicable and was
obtained using Millipore purification system. All extraction and cleanup procedures were
carried out in fume chamber.
3.7.1 Standards and calibration
Nivalenol (NIV), deoxynivalenol (DON), 3-acetyldeoxynivalenol (3acDON), 15-
acetyldeoxynivalenol (15acDON) and zearalenone (ZEA) certified standards were from
35
35
Sigma-Aldrich Chemie. Ergosterol analytical standard (>98% HPLC grade) was from Fluka
Chemie AG. Mycotoxin stock solutions were prepared as described in the official methods of
analysis (Scott, 1990). For each mycotoxin standard 5 mg was weighed into a 10 ml
volumetric flask and diluted to volume with acetonitrile to make a 0.5 mg/ml (500 ppm) stock
solution. A 20 µg/ml (20 ppm) NIV, DON, 3acDON, and 15acDON mixed working standard
solution was prepared by transferring 1ml of each of the respective 500 ppm individual
mycotoxin stock solution to a 25 ml volumetric flask and bringing to volume with acetonitrile.
A 10 µg/ml (10 ppm) ZEA working stock solution was made by transferring 1 ml of the 500
ppm ZEA stock solution to a 50 ml volumetric flask and bringing to volume with acetonitrile.
A 100 µg/g (100 ppm) ergosterol stock solution was made by dissolving 1 mg ergosterol
powder in 10 ml methanol. Working stock solution (10 µg/g) was prepared by diluting 1 part
of the 100 µg/g stock solution with 9 parts methanol (i.e. 1 ml+9 ml methanol). Fresh
ergosterol working standard stock solutions were prepared every 2 weeks. The standards were
stored at 4 °C.
Calibration data was generate by injecting 1, 4, 8, 10, and 20µl of the working standards. The
standard data (peak area, concentration, retention time) was fed into the software calibration
table to generate calibration curves and for automatic identification of peaks based on
retention time. UV absorbance spectra were also generated and the respective peak area ratio
of the two detection wavelengths calculated. Reproducibility of the calibration data was
checked daily by making at least 2 injections of each standard.
3.7.2 Mycotoxin analysis
3.7.2.1 Extraction and clean up
The liquid fermentation cultures were filtered with aid of vacuum pump and mycelia collected
on filter paper. The mycelia were freeze-dried for dry weight determination. The culture
filtrates from the duplicate flasks were combined in a separating funnel and extracted two
times by shaking vigorously for 1 minute with equal volumes of ethylacetate. Typically about
20 ml filtrate was recovered from each culture and filtrates from duplicate cultures were
combined to make 40 ml which was extracted with 40 ml ethylacetate. The layers were
allowed to separate and the lower water phase was drained off. The upper (ethylacetate) layer
was collected in a round bottomed flask. The water phase was extracted with another 40 ml
ethylacetate. The lower water phase was discarded. The combined ethylacetate extracts was
evaporated to dryness under vacuum at 40°C in a rotary evaporator and residue re-dissolved in
5 ml acetonitrile-water (80 : 20 v/v).
In the case of solid substrates, 5 g of the ground sample was placed in 250 ml screw cap bottle
and extracted with 50 ml of acetonitrile/water (3:1 v/v) by vigorously shaking on a rotary
mechanical shaker for 30 minutes. The extract was filtered through 125 mm diameter pre-
folded filter paper (Schleicher & Schuell, Germany) and the filtrate was evaporated to dryness
under vacuum in a rotary evaporator, at 50°C water bath temperature. The residue was re-
dissolved in 5 ml acetonitrile/water (80:20 v/v) for clean up.
Mycotoxin clean up was done using a charcoal alumina solid phase extraction (SPE) column
(Mycosep®, Romer labs, Inc.) according to Romer (1986), Trucksess et al. (1996), Walker and
Meier (1998), and Weingaetner et al. (1997). Five millilitres of the sample extract were
pipetted into culture tube of the column. The rubber flange end of the clean up column was
slowly pushed into the culture tube, creating a tight seal between rubber flange and glass wall
36
36
of the culture tube. Purified extract was collected in the column reservoir and 1 ml of the
purified extract was transferred into vial for HPLC analysis. Extracts were stored at 4°C until
analysis.
3.7.2.2 Mycotoxin detection and quantification by HPLC
The mobile phase was acetonitrile-water while the stationary phase was a reverse phase
Hilbar® pre-packed RT 125-4 Lichrosorb
® RP-18 (5µm) column (Merck Darmstadt,
Germany). Flow rate was set at 0.75 ml/min. The diode array detector was set at 220 nm and
245 nm for trichothecenes and 236 nm and 274 nm for zearalenone. At start of analysis, the
column was washed with acetonitrile (100%) and then preconditioned with several column
volumes of mobile phase by allowing the liquid chromatography to run for 30 minutes. One to
fifty microlitres of the sample extract were injected and subjected to elution gradient. An
elution gradient was generated by gradually increasing the acetonitrile/water ratio from 0 to 1
(Table 3.5). After elution of the mycotoxins, the column was washed with 100% acetonitrile
for 5 minutes before being re-equilibrated to the original conditions. Mycotoxin peaks were
identified by retention times as compared to the standards.
The sample extract was diluted in cases where its concentration was outside the response
range of the standard curve. The individual mycotoxin peaks in all sample extracts were
confirmed by respective UV-absorbance spectrum and peak area ratio of two
wavelengths(220/245 for trichothecenes and 236/274 for zearalenone). The sample peak area
ratio should agree within 5% with peak area ratio of the standard.
Table 3.5 Solvent gradient profile for HPLC analysis of mycotoxins
Time (min.) Acetonitrile (%) Water (%)
0 0 100
5 20 80
8 40 60
12 100 0
17 100 0
22 0 100
27 0 100
Concentration of individual mycotoxin in each sample was calculated using a modification of
methods described by Coker et al. (1984), Pons and Franz (1977), and Trucksess et al. (1996)
as follows:-
(i) Representative sample extract volume (Xv, ml) from the liquid media:-
X v = V1 x V3 e.g. 40 ml x 1 ml = 0.5 ml
V2 80 ml
37
37
Where, V1 = total volume of filtrate extracted; V2 = total volume of extracting ethylacetate;
V3= volume of acetonitrile-water (80:20 v/v) to dissolve dried residue; 1 ml = volume of the
cleaned up extract for analysis
(ii) Representative sample extract weight (Xw, g) from the solid substrate:-
Xw = 10 g x 1 ml = 0.2 g
50 ml
Where, 10 g = weight of sample extracted; 50 ml = volume of extracting solvent (acetonitrile-
water 80:20 v/v); 1 ml = volume of the cleaned up extract for analysis.
(iii) Amount of mycotoxin (Y, ng) in the test sample injected was determined from the
standard curves or by using the following equation:-
Y (ng) = Ax x Cs x Vs
As
where, Ax = sample extract peak area; Cs = standard concentration (ng/µl); Vs = volume of
standard injected (µl); As = standard peak area
(iv) Concentration of mycotoxin (TS [ng/g] or TS [ng/ml] in test sample:-
TS [ng/g] or (TS [ng/ml] = Y x F
Iv x X
Where, Y = ng mycotoxin in test extract injected; F = dilution factor of sample extract; Iv =
amount of test extract injected [µl]; X = weight [g] (or volume [ml] of test sample represented
by the extract.
For the liquid media, the mycotoxin yield per unit dry weight fungal mycelia (ng/mg mycelia)
was calculated as follows:-
Mycotoxin yield = Amount of mycotoxin [ng/ml] produced in the medium
[ng/mg mycelia] Weight mycelia [mg/ml] produced in the same media
3.7.3 Ergosterol analysis
3.7.3.1 Extraction and clean up
Ergosterol extraction was done by direct saponification according to Murkovic et al. (1997),
Newell et al. 1988, Schwadorf and Müller (1989) and Zill et al. (1988). Five gram sub-sample
was weighed into 250 ml screw cap bottle together with 35 ml methanol, 25 ml ethanol and 5
g potassium hydroxide (KOH) pellets. The mixture was vigorously shaken for 5 minutes on
rotary shaker and then refluxed for exactly 30 min. at 80 °C in a water bath. The mixture was
cooled to room temperature and 20 ml distilled water added. This was followed by filtration
using pre-folded filter paper into 250 ml separating funnel. The reflux bottle and filter were
38
38
rinsed with 10 ml methanol. Ergosterol in the filtrate was extracted by liquid-liquid
partitioning with 40 ml n-hexane by vigorously shaking the mixture for 1 minute. The mixture
was allowed to stand for 15 min. for complete separation of layers. The lower phase was
drained off together with a few drops of the upper phase. The upper n-hexane phase was dried
over anhydrous sodium sulphite (Na2SO3) into 250 ml round-bottomed flask. The lower phase
was re-extracted with additional 40 ml n-hexane as above and the two n-hexane fractions
were combined. The separating funnel and filter were rinsed with additional 10 ml n-hexane.
The lower phase was discarded. The combined n-hexane fractions were evaporated to dryness
under vacuum in a rotary evaporator at 40°C water bath temperature. The residue was
dissolved in 5 ml methanol with brief sonification. The sample was cleaned by passing
through 0.22 µm millipore filter and 0.5-1 ml collected in vial for HPLC analysis.
For the time course ergosterol formation, the freeze dried mycelia was crushed in 2-5 ml
ethanol by use of pestle and mortar. The solution was transferred into a 100 ml screw cap
bottle. The mortar and pestle were rinsed with 10 ml ethanol. To the mycelial solution, 35 ml
methanol was added and total volume made up to 60 ml with ethanol. The solution was
saponified by adding 5 g potassium hydroxide pellets and shaken vigorously for 10 minutes
on a shaker before refluxing and extracting as described above.
3.7.3.2 Ergosterol detection and quantification by HPLC
The stationary phase was a Lichrospher® Si-60 250 x 4mm, 5µm (E. Merck, Darmstadt,
Germany) column. The mobile phase was methanol/water (95/5 v/v). Flow rate was adjusted
to 1 ml/min. and wavelength of UV detector set at 282nm. The liquid chromatograph was
equilibrated with the mobile phase for 30 min. before starting the analysis. Five to fifty
microliters of sample extract were injected and ergosterol peak was identified by retention
time and UV-absorbance spectrum as compared to the standard. Peak purity was controlled by
acquiring spectra on up slope, Peak apex, and down slope of peak. Differences in curve shape
indicate presence of impurity.
Representative weight (W) of original sample in extract was calculated using the following
equation (Coker et al. 1984): -
W = X x V2 x V4
V1 x (V2 + V3)
where, X = weight (g) of sample extracted, V1 = total volume of the extracting solvents, V2 =
volume of extract filtrate, V3 = total volume of n-hexane used, and V4 = volume of the
combined n-hexane extract for drying.
Ergosterol (E [ng/g] content of sample was calculated as follows:-
E = Ax x Cs x Vs x Tv
As x Vx x W
where, Ax = peak area of sample extract, As = peak area of standard, Cs = concentration of the
standard, Vs = volume of standard injected, Vx = volume of sample extract injected, Tv =
volume of methanol used to dissolve extract residue, and W = representative weight of
original sample in extract
For the time course studies in liquid media,total ergosterol in mycelia was calculated as
follows:
39
39
Total ergosterol [µg] = Ergosterol yield [µg/g] x Mycelia dry weight [g].
Correlation coefficient between the amount ergosterol and respective mycelia weights were
then determined.
3.8 In vitro seedling mycotoxin bioassay
The experiments were carried out using authentic mycotoxin standards. Pathogen-free wheat
seeds of a susceptible variety, cv. ‘Mbuni’, were used in the bioassays. The seeds had been
obtained from greenhouse-grown plants and certified to be free of infection by plating on
PDA medium.
Mycotoxins NIV, DON, and ZEA were tested at 100, 50, 20, 10, 5 and 1 µg/ml concentration
levels, which correspond to the following:.
100 µg/ml 50 µg/ml 20 µg/ml 10 µg/ml 5 µg/ml 1 µg/ml
NIV (312
*) 3.2 x 10
-4 M 1.6 x 10
-4 M 6.4 x 10
-5 M 3.2 x 10
-5 M 1.6 x 10
-5 M 3.2 x 10
-6 M
DON (296*) 3.4 x 10
-4 M 1.7 x 10
-4 M 6.8 x 10
-5 M 3.3 x 10
-5 M 1.7 x 10
-5 M 3.3 x 10
-6 M
ZEA (318*) 3.1 x 10
-4 M 1.6 x 10
-4 M 6.3 x 10
-5 M 3.1 x 10
-5 M 1.6 x 10
-5 M 3.1 x 10
-6 M
* Molecular weight
The bioassay was conducted in 9 cm plastic petri dishes lined with double layer filter paper.
Mycotoxin stock solutions (500 µg/ml and 100 µg/ml) were prepared in acetonitrile. The 100
and 50 µg/ml concentration levels were made by dispensing 600 µl and 300 µl respectively, of
the 500 µg/ml stock solution onto filter paper surfaces while the 20, 10, 5, and 1 µg/ml levels
were made by dispensing 600 µl, 300 µl, 150 µl, and 30 µl respectively of the 100 µg/ml
stock solution. Duplicate plates were prepared for each concentration level. One set of control
consisted of 600 µl acetonitrile (0 µg/ml) while a second set consisted of 600 µl distilled
water blank. The acetonitrile was evaporated by allowing the opened plates dry under gentle
flow of air in fume chamber for 30 - 40 minutes. The different toxin concentration levels were
then reconstituted by adding 3 ml distilled water onto the filter paper surface in each petri
dish. Seeds were washed in distilled water, dried on paper towel and 20 seeds were evenly
placed in each petri dish. The plates were sealed with Parafilm® to prevent moisture loss.
Duplicate plates were prepared for each concentration level. Plates were incubated at 20°C
under white fluorescent light. On the 5th
day the seedlings were observed for germination,
growth rate and root mass. Shoot and roots were separated from the seed, weighed (fresh
weight) and then dried in oven at 60°C for 24 h. for dry weight determination. Other
observations made were germination, growth rate and root mass but only fresh and dry
weights were used for data analysis.
The acetylated derivatives of DON (3-acDON and 15-acDON) were tested at 10 µg/ml along
side similar concentrations of NIV, DON and ZEA for comparison.
3.9 Polymerase chain reaction-based characterisation of Fusarium isolates
3.9.1 Buffers
The following buffers were prepared according to Sambrook et al. (1989):
40
40
5X TBE Electrophoresis buffer
Tris base 54 g/l
Boric acid 27.5 g/l
EDTA 20 ml of 0.5 M solution adjusted to pH 8.0 with NaOH
The stock was diluted to 0.5X (0.045 M Tris-borate, 0.001 M EDTA) to make the working
solution.
6X Gel loading buffer
Bromophenol blue 0.25%
Sucrose 40% (w/v) in water
The buffers were autoclaved before use. The gel loading buffer was stored at 4°C, while the
TBE buffer was stored at room temperature.
3.9.2 DNA extraction
Fusarium isolates were cultivated for 4 to 6 days in malt extract broth. Mycelium was
harvested on sterile filter paper, washed with sterile distilled water and freeze dried. The
freeze dried mycelium was stored at 4 °C until
DNA extraction. DNA was isolated using DNeasy Plant Mini Kit (Qiagen, Hilden, Germany)
following the manufacture’s instructions. The buffers were coded and their composition
cannot be quoted here. Fifty milligrams of fungal mycelium was ground under liquid nitrogen
using mortar and pestle. The tissue powder was transferred to 1 ml Eppendorf tubes and liquid
nitrogen allowed to evaporate. Cell lysis was effected by addition of 400 µl buffer AP1,
together with 4 µl RNase A (100 mg/ml) and vortexed vigorously followed by incubation at
65°C for 10 min. Detergents, proteins and polysaccharides were precipitated by addition of
buffer AP2 and incubation for 5 minutes on ice. The precipitates and cell debris were removed
by centrifugation at 14000 rpm through QIAshredder spin column. The elute was mixed with
0.5 volume of buffer AP3, 1 volume of 96% ethanol and 650 µl of this mixture was applied
onto a DNeasy mini spin column and centrifuged at 8000 rpm for 1 minute. The flow through
was discarded and the column washed with 500 µl of buffer AW by centrifugation at 8000
rpm for 1 minute. One hundred microlitres of pre-heated (65°C) buffer AE was directly
pipetted onto the DNeasy column membrane and incubated for 5 minutes at room
temperature. DNA on the column was eluted by centrifugation at 8000 rpm for 1 minute.
DNA concentration and purity was determined by measuring the absorbance at 260 nm (A260)
and 280 (A280) in a spectrophotometer.
DNA concentration C [ng/µl] was calculated as follows:
C [ng/µl] A260 X dilution factor
20
Working DNA stock solutions was prepared by diluting the original stock to 25ng/µl. All
DNA solutions were stored at -20°C.
3.9.3 Random amplified polymorphic DNA (RAPD)-PCR
The RAPD-PCR reactions were carried out according to Schilling (1996) and Schilling et al.
(1996). Taq polymerase buffer, Taq polymerase and nucleotide mix were purchased from
Amersham Pharmacia biotech (Germany), while magnesium chloride was from Promega.
The following ten-base oligonucleotide primers were synthesized by Carl Roth GmbH
41
41
(Karlsruhe, Germany): 5’ - 3’ GGGGGTTAGG, GTGCTCGTGC, GATGCCAGAC,
GTCCGGAGTG, GGGCCACTCA, CCGGCCTTAG, CCTGGCGGTA, CCGGCCTTAA,
GAGCACCAGG. These primers have been used in amplification of Fusarium DNA
elsewhere (Chelkowski et al., 1999; Parry and Nicholson, 1996; Schilling et al., 1996a;
Turner et al., 1998; Yoder and Christiansen, 1998;). The PCR reactions were set up in master
mixes as follows: 10X Taq buffer with 1.5M MgCl2 (10mM Tris, 50 mM KCl, pH 7.5), 100
µM of each dNTP (dATP, dCTP, dGTP and dTTP), 1 mM MgCl2, 0.2 µM primer, 1U Taq
Polymerase per reaction. All the reagents were kept on ice during preparation of the PCR
reactions. The reaction mix was dispensed as 24 µl fractions into 0.2 ml Eppendorf tubes and
2 ng/µl of the test DNA was added per reaction. Negative controls comprising of all the PCR
components except DNA were used to check introduction of contamination during setting up
of the reactions. The samples were amplified through 1 cycle of 94 °C (2 min.), 35 °C (1
min.), 72 °C (2 min.) followed by 43 cycles of 94 °C (1 min.), 35 °C 1 min.), 72 °C (2 min.)
and 1 cycle of 94 °C (1 min.), 35 °C (1 min.), 72 °C (5 min.) in a programmable
thermocycler (Perkin Elmer GeneAmp PCR system 9600). After completion of PCR, the
samples were cooled to 4 °C until gel electrophoresis.
3.9.4 Species-specific PCR
Polymerase chain reaction using species-specific primers was carried out to confirm species
identity. The reactions were conducted according to Schilling (1996) and Schilling et al.
(1996b). Oligonucleotide primers pairs (20 bases in length) were: 5’ - 3’
GATGCCAGACCAAGACGAAG / GATGCCAGACGCACTAAGAT for F. culmorum and
GCAGGGTTTGAATCCGAGAC/ AGAATGGAGCTACCAACGGC for F. graminearum.
Master mixes were prepared with 1X Taq polymerase buffer (10mM Tris, 50 mM KCl, 1.5
mM MgCl2, pH 7.5), 1U Taq polymerase, 0.2 mM of each dNTP, 25 pM of each forward and
reverse primer. The reaction mix was dispensed as 24 µl fractions into 0.2 ml Eppendorf
tubes and 2 ng/µl test DNA was added per reaction. Negative control containing all the
components except DNA was included to check presence of contamination. Amplification
was performed with 1 cycle of 94 °C (2 min.), 55 °C (1 Min.), 72 °C (2 Min.) followed by 30
cycles of 94 °C (1 Min.), 55 °C (1 Min.), 72 °C (2 Min.) and 1 final cycle of 94 °C (1 Min.),
55 °C (1 Min.), 72 °C (5 Min.). After completion of PCR, the samples were cooled to 4 °C
until gel electrophoresis.
3.9.5 Gel electrophoresis and banding pattern evaluation
PCR amplification products were resolved on 20 cm x 10 cm horizontal electrophoresis
chamber (Biometra, Göttingen, Germany). Gels (0.5 cm thick) were prepared with 1.5%
agarose (Amersham Pharmacia biotech) in 0.5X TBE buffer (0.045M Tris-borate, 0.001M
EDTA). The agarose was heated in a microwave oven until boiling and completely dissolved.
It was allowed to cool to 60 °C and Ethidium bromide (10 mg/ml) was added to a final
concentration of 5 µg/ml. The gel was cast in a gel try and a 1 mm, 22-teeth well-forming
comb. The gel was allowed to cool and solidify for 30-40 min., after which the comb was
carefully removed. It was then placed in the electrophoresis chamber so that the wells were
closer to the cathode. The chamber was filled with 0.5X TBE buffer until the upper surface of
gel was submerged to about 1mm.
To each 25 µl PCR amplification product sample, a 5 µl aliquot of the 6X bromophenol blue
loading buffer was added and 25 µl sample was loaded to each well. A 100-bp ladder was
42
42
loaded to one of the wells as molecular weight standard. The PCR amplification product
fragments were separated by electrophoresis at 100 V for 90 minutes and banding pattern was
recorded by photography under UV-light using computerised gel image photography system
provided with molecular analyst software.
The banding patterns were scored visually. Fragments generated with a particular primer
appearing on the same gel position in banding patterns from different isolates were considered
homologous. Those fragments which have been amplified in at least one isolate but not in all
isolates were considered as polymorphic and used as markers for isolate identification. The
reproducibility of every marker was confirmed with at least two single amplification
reactions. Banding patterns were scored as ‘present’ or ‘absent’ and translated into a binary
code with 1 or 0, respectively. Jaccard’s coefficients of genetic similarity were then calculated
by pair-wise comparison of the isolates using the formula (Gerlach and Stösser, 1997):
Sab = nab / (nab + na + nb), where nab is the number of bands common to individuals a and b,
na and nb are the number of bands unique to individual a and b, respectively.
Cluster analysis of the binary code data was conducted using Statgraphics
Plus version 1.4
statistical software. Dendograms were derived using nearest neighbour (single linkage)
squared Euclidean method. Isolates clustering together were considered to be genetically
closely related. The isolates were evaluated in terms of genetic similarity, taking into account
the species and type of mycotoxin produced by correlating the PCR data with mycotoxin and
aggressiveness data.
3.10 Statistical data analysis
The amount of disease due to the effect of fungal invasion and amount of disease due to the
effect of DON production were deduced by calculating the ratio of disease severity to
ergosterol content and disease severity to DON content, respectively, for the different isolates.
Variety resistance based on ergosterol and DON synthesis (Miller et al. 1985; Perkowski et al.
1995) was inferred by calculating the ratios of ergosterol to DON and DON to ergosterol (rate
of DON synthesis by the mycelium). The in vitro mycotoxin data was log10 transformed in
order to attain approximately normally distributed data with homogenous variances (Steel and
Torrie, 1981). To take care of the zero and less than 10 values, 1 was added to each reading
prior to taking logarithms, i.e. log10 (Y + 1).
Data analysis was done with the help of SigmaStat
statistical software (SPSS inc.). Analysis
of variance was carried out to determine whether different isolates significantly differ in
aggressiveness and mycotoxin production. Varieties were also tested for significant
differences in susceptibility to Fusarium head blight and mycotoxin content. Pair wise
treatment mean differences was determined by Tukey test. Pearson product moment
correlation coefficients were derived to determine relationship among disease severity, kernel
weight reduction, mycotoxin production in ears, ergosterol content, mycotoxin production in
vitro, and re-isolation of the pathogen. The preliminary seedling and ear inoculation
experiment data was also subjected to correlation analysis to test whether seedling
aggressiveness was correlated to ear aggressiveness and mycotoxin production.
43
43
Chapter 4: Results
4.1 Fungi isolated from wheat samples
Wheat samples were collected from five wheat growing regions in Kenya with the objective
of determining prevalence of Fusarium species and other fungi in the kernels.
The major genera of fungi isolated from wheat kernels were Alternaria, Epicoccum,
Fusarium, Aspergillus and Penicillium (Fig.1; Plate 1A and B). The most frequently isolated
genera were Epicoccum (mean 40% kernels infected) and Alternaria (mean 25% kernels
infected). Only 5% of wheat kernels showed Fusarium infection. The strains of Epicoccum
frequently were identified as Epicoccum purpurascenes at the Federal Biological Research
Centre for Agriculture and Forestry (BBA), Berlin, Germany. The strains were found to be
different from those commonly isolated in Europe. Four strains are deposited at the BBA
culture collection (BBA 70981, BBA 70982, BBA 70983 and BBA 70984).
Epicoccum and Alternaria were frequently isolated in samples collected from Nakuru and
Laikipia (54% and 28% kernel infection, respectively). Epicoccum was least isolated in
samples from Laikipia (28%) while Alternaria was least frequently isolated in samples from
Nyandarua and Meru (both with 22% kernel infection). Localities with low Epicoccum and
Alternaria infection frequency, however, had the highest Fusarium infection (7% kernel
infection). Localities with very high Alternaria or Epicoccum infection (Nakuru and Laikipia)
showed low Fusarium infection. Storage fungi like Aspergillus and Penicillium were isolated
at very low frequencies and were mainly found in samples bought at the market.
0
10
20
30
40
50
60
70
N akuru
N yand arua
L aikip ia
M eru
N aro k
Rel
ativ
e fr
equen
cy (
%)
Fusarium Alternaria Epicoccum Penicillium Aspergillus Other Fig.1 Frequency of the major genera of fungi contaminating wheat samples collected from
different districts in Kenya.
44
44
Out of the different Fusarium species isolated 80 isolates were of F. graminearum, 89 isolates of F. poae, 17 isolates of F. avenaceum, 9 isolates of F. oxysporum, 3 isolates of F. equiseti and 8 isolates of other Fusarium species (Fig. 2; Plate 2). Two of the other Fusarium species were identified at BBA as F. camptoceras and F. chlamydosporum and are held in culture collection at BBA as BBA 71491 and BBA 71496, respectively. The overall relative frequency of the three major Fusarium species was 38.8% for F. graminearum, 43.2% for F. poae and 8.3% for F. avenaceum. On the basis of locality, the highest number of F. graminearum isolates were obtained in samples collected from Nyandarua (44 isolates) followed by Laikipia (21 isolates), Meru (10 isolates), Nakuru (5 isolates) and lowest in Narok (0). Fusarium poae was most isolated in samples collected at Meru (37 isolates) followed by Nakuru (31 isolates), Nyandarua (17 isolates), Laikipia (4 isolates) and least in Narok (0). F. avenaceum was most frequently isolated from Meru (10 isolates) Narok (3 isolates), Nyandarua (3 isolates), Nakuru (1 isolate) and none in Laikipia.
None of the 41 isolates of F. graminearum (Plate 3) from Kenya were found to form
perithecia on carnation leaf agar (CLA) and water agar amended with wheat straw. However,
the reference isolate (Fg7.1) formed abundant black perithecia on both media within 3 weeks
of incubation (Plate 4). Other F. graminearum isolates (BMM1, B9J4 and 2.2b) found to form
perithecia had been isolated from maize kernels from Germany.
Twenty seven Fusarium isolates were isolated in an earlier study on Fusarium species
infecting wheat in organic and conventional farming systems in the Rheinlands, Germany.
These isolates had been morphologically identified as F. culmorum.
0
5
10
15
20
25
30
35
40
45
Nakuru Nyand arua Laikip ia Meru Naro k
F .g ra m in ea ru m
F .p oa e
F .a ven a ceu m
F .oxysp oru m
F .eq u iseti
O th er
No
. o
f is
ola
tes
Fig.2. Frequency of Fusarium species isolated from wheat samples collected in different
localities in Kenya.
45
45
4.2 Detection and quantification of mycotoxins and ergosterol by High pressure liquid
chromatography (HPLC).
The aim of this experiment was to develop a high pressure liquid chromatographic (HPLC)
analytical method for simultaneous detection and quantification of Fusarium mycotoxins.
Mycotoxins investigated were trichothecenes (NIV, DON, 3-acDON, 15-acDON) and
zearalenone. In addition, ergosterol determinations were also carried out.
The mycotoxins nivalenol, deoxynivalenol, 3-acetyldeoxynivalenol, 15-acetyldeoxynivalenol
and zearalenone could simultaneously be detected in a single analysis run. The mycotoxin
extraction and quantification procedures were relatively fast and simplified. Over 15 samples
could be extracted and quantified by one person per day. The sample extracts were free from
most of the other grain matrices which usually result in multiple overlapping peaks, and
therefore interfering with detection and quantification (Fig. 3). The recovery rates for the
different mycotoxins were NIV 55-65%, DON 70-85%, 3-acDON and 15-acDON 100% and
ZEA 70%.
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
-5
0
5
10
15
5.7
02
6.8
28
9.6
11
DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D \STDN8-1.AIA\SIGNAL01.CDF (AIA imported)
15-acDON
DON
NIV
Standard mycotoxin mix
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
0
20
40
60
80
100 5.7
78
12
.94
2
DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D \K4CULT.AIA\SIGNAL01.CDF (AIA imported)
NIV
ZEANIV isolate
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
0
200
400
600
800
1000
1200
6.8
86
9.5
03
12
.94
4
18
.30
4 DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D \K60CULT.AIA\SIGNAL01.CDF (AIA imported)
DON
ZEADON isolate
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
0
20
40
60
80
100
120
140
5.7
15
6.8
90
9.4
96
12
.94
2
DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D \K59CULT.AIA\SIGNAL01.CDF (AIA imported)
NIV
DON
ZEA
NIV-DON isolate
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
-5
0
5
10
15
20
25
6.7
53
DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D \K15N9-1.AIA\SIGNAL01.CDF (AIA imported)
DON
Infected wheat sample
min2.5 5 7.5 10 12.5 15 17.5 20
mAU
0
20
40
60
80
100
120
140
160
18.0
73
DAD1 A, Sig=220,4 Ref=550,100, TT of MUTHOMI\NOV11-1\009-0901.D\K61N9-1.AIA\SIGNAL01.CDF (AIA imported)
Uninfected sample
Fig.3. HPLC chromatograms illustrating mycotoxin peaks detected in contaminated wheat and culture extracts after gradient elution with acetonitrile/water (Hilbar
RT 125-4 Lichrosorb
RP-18, 5 µm column), as compared to a mixed trichothecene standard and a clean wheat sample.
46
46
nm200 250 300 350 400 450 500 550
mAU
0
50
100
150
200
250
*DAD1, 6.861 (329 mAU, - ) Ref=3.515 & 17.008 of 010-1001.D
nm200 250 300 350 400 450 500 550
mAU
0
20
40
60
80
100
120
140
*DAD1, 7.775 (159 mAU,Apx) Ref=2.762 & 8.709 of 017-3801.D
nm200 250 300 350 400 450 500 550
mAU
-2
0
2
4
6
8
10
*DAD1, 9.350 (14.3 mAU,Apx) Ref=9.030 & 9.764 of 003-0301.D
NIV DON
3-acDON
nm200 250 300 350 400 450 500 550
mAU
0
2.5
5
7.5
10
12.5
15
*DAD1, 9.440 (18.8 mAU, - ) Ref=9.280 & 9.920 of 003-0301.D15-acDON
Wavelength(nm) Wavelength(nm)
Wavelength(nm) Wavelength(nm)
nm200 250 300 350 400 450 500 550
mAU
0
250
500
750
1000
1250
1500
*DAD1, 12.916 (1897 mAU, - ) Ref=4.103 & 18.010 of 002-0201.D
ZEA
Wavelength(nm) Fig. 4. Absorbance spectra of nivalenol (NIV), deoxynivalenol (DON), 3-acetyldeoxynivalenol (3-acDON), 15-acetyldeoxynivalenol (15-acDON), and zearalenone (ZEA) obtained from HPLC UV-diode array detector after elution with acetonitrile/water (Hilbar
RT 125-4 Lichrosorb
RP-18, 5 µm column).
nm200 250 300 350 400 450 500 550
mAU
0
100
200
300
400
DAD1, 2.280 (474 mAU,Apx) of 008-0801.D
Wavelength (nm)
UV-spectrum
min1 2 3 4 5 6 7 8
mAU
0
5
10
15
20
2.2
43
DAD1 A, Sig=282,4 Ref=550,100 of MUTHOMI\ERG.AIA\SIGNAL01.CDF (AIA imported)
Chromatogram
Wavelength (nm)
Fig. 5 HPLC chromatogram and UV-absorbance spectrum of ergosterol eluted in
methanol/water (95 + 5; Lichrospher® Si-60 250 x 4mm, 5µm column).
47
47
The respective retention times in minutes for each mycotoxin were 5.7 (NIV), 6.8 (DON), 9.8
(3-acDON), 9.6 (15-acDON) and 12.9 (ZEA). NIV was eluted with 23% acetonitrile, DON
30%, 3-acDON 63%, 15-acDON 61% and ZEA 100%. Each of the mycotoxins NIV, DON, 3-
acDON and 15-acDON had a single UV-absorbance maxima at 224 nm, 218 nm, 218 nm, and
218 nm respectively (Fig. 4). Zearalenone had three absorbance maxima at 236 nm, 274 nm
and 314 nm. The absorbance at 220 nm was approximately 2 times the absorbance at 245 nm
for the trichothecene mycotoxins while the absorbance at 236 nm for ZEA was also about 2
times the absorbance at 274 nm. The ratios of peak areas at 220/245 nm and peak area at
236/274 nm were found to be approximately 2.3 and were also used to confirm trichothecene
and zearalenone peaks, respectively. The major disadvantage of the mycotoxin method was
the reduced detection limit because some of the mycotoxins are retained in the clean up
column. The method was, therefore, found to be suitable for analysing samples containing
more than 1 ppm mycotoxin concentration, and hence ideal for analysing samples from
inoculation experiments.
The ergosterol extraction and purification procedures were, however, more demanding and on
average 8 samples could be extracted, purified and quantified by one person in one day.
Ergosterol had a retention time of 2.3 minutes (Fig. 5). Ergosterol had three absorbance
maxima at 271 nm, 282 nm and 294 nm and the recovery rate was about 95%.
4.3 Effect of substrate and incubation time on mycotoxin production and fungal biomass
formation by Fusarium culmorum and F. graminearum
Suitability of different substrates to support mycotoxin production and fungal growth was
tested. The information obtained was used to select the best medium for further
characterization of Fusarium isolates on the basis of mycotoxinproduction.
4.3.1 Effect of liquid media on mycotoxin production and ergosterol formation
4.3.1.1 Mycotoxin production in different liquid media
The liquid media tested were yeast-extract-mineral salts broth (YEM), glucose-yeast-extract
broth (GYEP), and malt extract broth (ME). Both Fusarium culmorum and F. graminearum
produced mycotoxin in liquid media, but only deoxynivalenol was detected in yeast-extract-
mineral salts broth (YEM) and glucose-yeast-extract broth (GYEP) (Table 4.1). No mycotoxin
was detected in malt extract broth (ME) and only the F. culmorum isolate produced DON in
GYEP. Mycotoxin production was low as compared to solid fermentation cultures.
Table 4.1 Effect of different liquid media on mycotoxin production (µg/ml) by Fusarium
culmorum, isolate C41 and of F. graminearum, isolate BMM1.
F. culmorum F. graminearum
Medium DON 3-acDON DON 15-acDON .
YEM 3.2 - 4.1 -
GYEP 0.2 - - -
ME - - - - . - = not detected; YEM = Yeast extract mineral salts broth; GYEP = Glucose-yeast extract-peptone broth;
ME = Malt extract broth
48
48
4.3.1 2 Time course of mycelial biomass and ergosterol production in liquid
fermentation
The objective of this experiment was determine the relationship between fungal biomass
(mycelium dry weight) and the respective ergosterol content. This would give an idea of
whether ergosterol could be used as a reliable estimate of fungal biomass in wheat kernels
from ear inoculation experiments.
0
500
1000
1500
2000
2500
3000
3500
4000
0 5 10 15 20 25 30 35
0
50
100
150
200
250
300
350
400
450
500
Ergosterol Mycelial biomass
Erg
ost
ero
l (µ
g)
Myce
lial
bio
mas
s (m
g)
Incubation time (days)
r = 0.88, p = 0.01
a
a
bb
b
b
b
Fig. 6. Time course of total ergosterol formation and mycelial biomass (dry weight)
production by Fusarium culmorum, isolate C41, in malt extract broth (ME) at 25°C (Levels
followed by different letters are significantly different).
0
500
1000
1500
2000
2500
3000
3500
4000
0 5 10 15 20 25 30 35
0
50
100
150
200
250
300
350
400
450
500
Ergosterol Mycelial biomass
Erg
ost
erol
(µg)
Myce
lial
bio
mas
s (m
g)
Incubation time (days)
r = 0.89, p = 0.01
a
a
b
c
bb
b
Fig.7 Time course of total ergosterol formation and mycelial biomass (dry weight) production
by Fusarium graminearum isolate BMM1 in malt extract broth (ME) at 25° C (Levels followed
by different letters are significantly different).
49
49
The pattern of time course of mycelium biomass and ergosterol production was basically
similar for Fusarium culmorum and F. graminearum, except for some minor differences. In
the case of F. culmorum isolate C41, mycelium dry weight increased at a high rate from 73
mg at day 5 to 380 mg at day 15, thereafter remaining relatively constant up to the end of the
experiment (Fig.6). Maximum ergosterol was measured at day 35. The amount of mycelium
produced after 5 days was significantly different (p<0.001) from that produced after 15 days
of incubation. However, there existed no significant differences among the amounts of
mycelium produced after 15, 20, 25, 30 and 35 days. Ergosterol production pattern generally
followed that of mycelium dry weight, the two being significantly correlated (r = 0.88,
p<0.005). Total ergosterol increased at a high rate from 680 µg at day 5 to 2000 µg at day 15.
Maximum total ergosterol of up to 2580 µg was measured at day 30. Lower amount of
ergosterol (2180 µg) was measured at day 35. However, no significant differences (p<0.1)
were noted among the amounts of ergosterol measured at 20, 25, 30, and 35 days. Ergosterol
measured at day 15 was significantly different from that measured at day 5 and 10.
For F. graminearum, isolate BMM1, the total mycelium dry weight steadily increased from
119 mg at day 5 to a maximum of up to 375 mg at the 20th
day (Fig. 7). The mycelium dry
weight then remained constant until the end of the experiment. There were no significant
differences (p<0.10) among the mycelia dry weights determined at the 5th
, 10th
and 15th
day of
incubation but the amount of mycelium determined at the 20th
day was significantly different
from that determined at the 15th
day. Amounts of mycelium produced after 20, 25, 30, and 35
days were not significantly different. The time changes in total ergosterol content of the
mycelium closely followed that of the mycelium dry weight and the two were significantly
correlated (r = 0.89, p<0.005). The ergosterol content increased at a high rate from 790 µg at
the 5th
day until 3490 µg at the 20th
day, then slightly decreased to 2470 µg at the 35th
day.
Total ergosterol produced after 20 days was significantly different (p<0.001) from that
produced before and after 20 days. No significant differences were noted among the amounts
of total ergosterol produced after 25, 30, and 35 days.
4.3.1.3 Time course of deoxynivalenol and mycelia dry weight production in liquid
culture.
Deoxynivalenol production rate and fungal growth over time were determined in yeast-
extract-mineral salts broth (YEM).
The two isolates tested, F. culmorum isolate C39 and F. graminearum isolate 2.2a, showed
different time course DON production patterns (Fig 8 and 9). However, the time course of
mycelial dry weight production by the 2 isolates were almost similar. Mycelium production by
F. culmorum, isolate C39, increased sharply during the first 12 days from 1.1 mg/ml at day 4
to 2.3 mg/ml at day 12. Production of mycelia then stabilised up to day 20. In contrast, the rate
of DON production increased very slowly in during the first 12 days from 90 ng/mg mycelia at
day 4 to a maximum (320 ng/mg) mycelia by day 12. The rate of DON production then
increased drastically to 2950 ng/mg mycelia by day 20.
In the case of F. graminearum, isolate 2.2a, the maximum mycelial dry weight yield was
recorded at day 8. The mycelia yield increased from 3.0 mg/ml at day 4 to 4.0 mg/ml at day 8.
Deoxynivalenol production by this isolate closely followed the production of mycelium.
Deoxynivalenol yield increased sharply from 60 ng/mg mycelia on the 4th
day to 120 ng/mg
50
50
mycelia at day 8 and then slowly increased to 130 ng/mg mycelia by day 16. Thereafter, DON
production declined gradually to 74 ng/mg mycelia by day 20.
0
5 0 0
1 0 0 0
1 5 0 0
2 0 0 0
2 5 0 0
3 0 0 0
0 4 8 1 2 1 6 2 0
0
1
2
3
4
5
6
D O N M y c e lia
DO
N y
ield
(n
g/m
g m
yce
lia)
My
celi
a yie
ld (
mg
/ml)
Time (Days) Fig. 8 Time course of deoxynivalenol and mycelium dry weight production by Fusarium
culmorum , isolate C39, in yeast-extract-mineral salts broth (YEM).
0
20
40
60
80
100
120
140
160
180
200
0 4 8 12 16 20
0
1
2
3
4
5
6
DO N Mycelia
DO
N y
ield
(n
g/m
g
my
celi
a)
My
celi
a y
ield
(m
g/m
l)
Time (Days)
Fig. 9 Time course of deoxynivalenol and mycelium dry weight production by Fusarium
graminearum, isolate 2.2a, yeast-extract-mineral salts broth (YEM).
51
51
4.3.2 Effect of Solid substrates on mycotoxin production and ergosterol formation
Solid substrates tested were cracked corn, rice and wheat grain. In addition to determining the
best solid substrate for mycotoxin production, time course studies were carried out to
determine the type and quantity of mycotoxin produced at different stages of fungal growth.
The information obtained would help understand mycotoxin biosynthesis and type of
mycotoxin most likely to be found in wheat Fusarium infected grain at harvest.
4.3.2.1 Mycotoxin production in different solid substrates
The two isolates, F. culmorum C41 and F. graminearum BMM1, produced mycotoxins in all
the three media tested (Table 4.2). Trichothecene mycotoxins detected were DON, 3acDON
and 15acDON. Fusarium culmorum produced 3acDON as the acetyl derivative of DON, while
F. graminearum produced 15acDON. Cracked corn supported the highest production of the
mycotoxins (up to 31 µg/g DON and 36 µg/g acetyl-DON) while least mycotoxin production
was found in wheat kernels (up to 4.5 µg/g). Unlike in liquid fermentation cultures,
mycotoxin production was consistent in solid fermentation. Therefore, these cultures were
adopted for further experiments using more isolates of different Fusarium species.
Trichothecene production was higher and more consistent in less aerated fermentation cultures
than in well aerated cultures. However, more aerated conditions were favourable for the
production of ZEA. Therefore, the fermentation cultures which were occasionally shaken and
the cotton plugs regularly opened in the lamina flow cabinet contained less trichothecene
mycotoxins but more ZEA than the cultures which were permanently closed with cotton plugs
and aluminium foil (Table 4.3).
Table 4.2 Deoxynivalenol production (µg/g) by Fusarium culmorum, isolate C41, and F.
graminearum, isolate BMM1, in autoclaved cracked corn, rice and wheat grain cultures.
Substrate F. culmorum, F. graminearum,
DON 3acDON DON 15acDON
Cracked corn 30.9 35.5 26.2 21.0
Rice 22.5 20.6 17.9 13.0
Wheat 4.5 3.0 1.9 -
Table 4.3 Effect of oxygen supply on mycotoxin production (µg/g) and ergosterol formation
by F. culmorum, isolate C41 and F. graminearum, isolate BMM1, in autoclaved cracked corn.
Isolate Well aerated fermentation Less aerated fermentation
DON acDON ZEA ERG DON acDON ZEA ERG
C41 136 42 1142 761 218 386 448 1556
BMM1 138 64 643 1114 259 93 716 1067
52
52
Table 4.4 Mycotoxin production (µg/g) by isolates of Fusarium species in autoclaved rice and
cracked corn cultures.
Isolate Rice fermentation . Cracked corn fermentation
NIV DON 3acDON 15acDON NIV DON 3acDON 15acDON
BMM1 - 14.6 - 6.6 - 49.5 - 36.1
2.2a - 252.0 - 176.9 - 85.4 - 60.1
2.2b - 52.5 - 25.3 - 416.7 - 102.7
B9F12 - 3.2 - 5.7 - 53.8 - 47.1
B9J4 - 5.7 - 5.1 - 46.2 - 39.5
Fg7.1
F. culmorum
176.0 - - - 60.6 - - -
C38 6.1 - - - 73.1 - - -
C39 - 22.6 36.9 - - 55.1 110.6 -
C40 - 36.7 26.8 - - 106.5 333.1 -
C41 - 23.9 21.0 - - 26.7 66.1 -
C42 42.3 - - - 154.0 - - -
Fc3.1
F. poae
- - - - - - - -
B1M5 - - - - NT NT NT NT
FPIwt . - - - - NT NT NT NT
- = not detected; NT = not tested.
Most of the isolates of F. culmorum and F. graminearum tested produced mycotoxins in rice
and cracked corn fermentation cultures (Table 4.4). Both substrates were found to support
high production of mycotoxins but cracked corn was found to be generally better for most of
the isolates tested. Out of 6 isolates of F. culmorum tested, 3 produced DON and 3acDON in
both rice and cracked corn, 2 produced NIV in both substrates but no mycotoxin was detected
for 1 isolate (Fc3.1). Only one isolate (Fg7.1) of F. graminearum produced NIV, while the
other 5 isolates produced DON and 15acDON. None of the 2 isolates of F. poae produced
detectable amounts of the mycotoxins analysed (NIV, DON, 15acDON and 3acDON) in rice
fermentation culture. Because cracked corn proved to be a superior substrate for the
production of the trichothecene mycotoxins, it was adopted for all further mycotoxin
characterisation tests using more isolates of only F. culmorum and F. graminearum.
4.3.2.2 Time course of mycotoxin production and ergosterol formation in solid
fermentation by Fusarium culmorum and F. graminearum
Toxigenesis and fungal growth was tested on autoclaved cracked fermentation cultures at
reduced oxygen supply. Fusarium culmorum, isolate C41, produced DON, 3acDON and ZEA
while F. graminearum, isolate BMM1, produced DON, 15acDON and ZEA.
Time course of mycotoxin and fungal biomass (ergosterol content) production by F.
culmorum are illustrated in Fig.10. Ergosterol content increased at a high rate from 450 µg/g
at day 5 to 1070 µg/g at day 15 and then increased slowly to a maximum at day 30. There
were significant differences among the ergosterol measurements at days 5, 10, and 15 but no
significant differences were found among ergosterol measurements at days 15, 25 and 35.
Deoxynivalenol was first detected at day 10 (31 µg/g) and then increased until day 20 (200
µg/g). Maximum DON content was detected at day 35. Unlike DON, 3acDON was detected at
F. graminearum
53
53
relatively high amounts at day 5 (20 µg/g) and could be detected in the fermentation cultures
over the whole analysis period. This metabolite sharply increased to a maximum (380 µg/g) at
day 20 when production stabilised until day 35. Reduced amount (46 µg/g) of 3acDON were
measured at day 40. ZEA was detected from day 5 (0.2 µg/g) but its levels remained low until
day 15 when production sharply increased from 20 µg/g to 200 µg/g by day 20 and to a
maximum at day 35 (450 µg/g). Lower amounts of ZEA (173 µg/g) were detected at the 40th
day. Production of DON over time was significantly (p<0.005) correlated to 3acDON, ZEA,
and ergosterol production (Table 4.5). 3acDON was correlated to ZEA formation but it was
not correlated to ergosterol production. ZEA was correlated to ergosterol production.
Time course production of mycotoxins and ergosterol by F. graminearum, isolate BMM1, is
shown in Fig.11. Fungal biomass (ergosterol) increased from 350 µg/g at day 5 to a maximum
of 1070 µg/g at day 15 and then remained relatively constant until the end of the experiment.
Deoxynivalenol (DON) was first detected at day 10 (85 µg/g) and it increased steadily to a
maximum of up to 260 µg/g at day 25. It then levelled off until day 40. Unlike in F.
culmorum, 15-acDON was detected in F. graminearum cultures. 15-acDON was detected
from day 5 (50 µg/g) and it slightly increased to a maximum at day 15 (90 µg/g). It then
gradually decreased to a minimum at 35 day (30 µg/g). The pattern of ZEA production by F.
graminearum was basically the similar to that of F. culmorum. The production was low
during the first 10 days (0.3 µg/g at day 5 and 15 µg/g at day 10) but it sharply increased up to
a maximum of 720 µg/g at day 25. Reduced amounts of ZEA were detected from day 30 to
day 40 (420 µg/g at day 30). DON production over time was significantly correlated (p<0.05)
to ZEA and ergosterol production but it was only negatively correlated to 15acDON
production (Table 4.6).
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
3 0 0
3 5 0
4 0 0
4 5 0
5 0 0
0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0
0
2 0 0
4 0 0
6 0 0
8 0 0
1 0 0 0
1 2 0 0
1 4 0 0
1 6 0 0
1 8 0 0
D O N 1 5 acD O N Z E A R E R G
Time (days)
DO
N /
3ac
DO
N (
µg/g
)
Erg
este
rol
/Zea
rale
none
(µg/g
)
Fig.10. Time course of mycotoxin production and ergosterol formation by Fusarium
culmorum, isolate C41, in autoclaved cracked corn culture at 25 °C.
54
54
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
3 0 0
3 5 0
4 0 0
4 5 0
5 0 0
0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0
0
2 0 0
4 0 0
6 0 0
8 0 0
1 0 0 0
1 2 0 0
1 4 0 0
1 6 0 0
1 8 0 0D O N 1 5 ac D O N Z E A R E R G
Time (days)
DO
N /
15ac
DO
N (
µg/g
)
Erg
ost
erol
/ Z
eara
lenone
(µg/g
)
Fig.11 Time course of mycotoxin and ergosterol formation by Fusarium graminearum, isolate
BMM1, in autoclaved cracked corn culture at 25 °C.
Table 4.5 Correlation coefficients among time course of mycotoxin production and ergosterol
formation by F. culmorum, isolate C41.
DON 3-acDON ZEA
3-acDON 0.769
ZEA 0.810 0.631
Ergosterol 0.678 0.297 0.542
Table 4.6. Correlation coefficients among time course of mycotoxin production and ergosterol
formation by F. graminearum, isolate BMM1.
DON 15acDON ZEA
15acDON -0.464
ZEA 0.903 -0.497
Ergosterol 0.669 0.026 0.408
55
55
4.4 Effect of different Fusarium species on wheat seedling growth, head blight severity,
grain weight and grain mycotoxin content
The objective of the experiment was to determine the aggressiveness of different Fusarium
species on wheat and to determine the relationship among seedling aggressiveness, ear
aggressiveness and grain mycotoxin content. The seedling aggressiveness parameters
determined were emergence, shoot length, shoot dry matter and root dry matter. Proportion of
spikelets bleached and grain weight reduction were for ear aggressiveness.
Fusarium infection on seedlings produced pre- and post-emergence damping off, stunting and
reduced vigour, resulting in reduced dry matter as compared to non-inoculated seedlings. The
infected seedling stem bases showed brown lesions, occasionally with fungal mycelia growth
(Plate 5). On seedlings inoculated with very aggressive isolates, the stem base lesions girdled
the stems and progressed up and down the stem resulting in post-emergence damping off.
Inoculation of ears resulted disease symptoms as early as 5 days later. Initial disease
symptoms were bleaching of single spikelets, but with time, the infection spread to
neighbouring spikelets, sometimes causing bleaching of the whole ears (Plate 6). The
Fusarium isolates tested significantly differed (p<0.05) in aggressiveness on both seedlings
and ears (Table 4.7).
Table 4.7 Effect of Fusarium isolates on wheat seedling growth, head blight severity, grain
weight and grain mycotoxin content. Except mycotoxin content, all the parameters are in
percent reduction as compared to non-inoculated controls.
Seedling infection tests . Ear infection tests .
Isolate
Emergence
(%)
Shoot length
(mm)
Root dry
matter (g)
Shoot dry
matter (g)
Disease
severity (%)
Grain
wt. (g)
DON
(µg/g)
BMM1 77.1d 86.3
de 93.3
f 86.7
f 41.9
bc 60.0
ab 25.3
B9F12 68.4cd
81.3d 94.7
f 85.0
f 60.9
c 70.6
a 55.4
B9J4 63.2cd
77.3d 87.4
e 74.9
e 31.5
ab 58.7
ab 11.1
2.2a 43.1c 58.9
c 71.6
d 55.3
d 55.0
c 61.2
ab 32.1
2.2b 85.9d 95.5
e 97.6
f 95.9
g 66.1
c 72.7
a 53.4
Fg7.1 21.6b 41.8
c 47.7
c 37.2
c 16.9
ab 21.2
c 0
6.4 F. poae
7.5a 10.8
a 13.4
b 10.8
ab 12.8
ab 29.3
c 0
B1M5 1.3a 2.1
a 1.7
a 7.6
a 0
a 21.0
c 0
B11M4 0.0
a 4.7
a 2.5
a 12.7
b 0
a 17.5
cd 0
B1M1 1.3a 6.0
a 4.4
a 4.8
a 0
a 28.6
c 0
Values followed by different letters are significantly different.
F. graminearum
F. sporotrichioides
56
56
Table 4.8 Correlation coefficients (p<0.05) among percentage reduction in seedling growth,
ear aggressiveness parameters and DON content in kernels.
Emergence Shoot length Root dry
matter
Shoot dry
matter
Disease
severity
Grain
weight
Shoot length 0.989 Root dry matter 0.962 0.971 Shoot dry matter 0.986 0.981 0.986 Disease severity 0.905 0.910 0.911 0.908 Grain weight 0.933 0.913 0.921 0.923 0.948 DON content 0.811 0.801 0.736 0.782 0.911 0.857
Five isolates of F. graminearum (BMM1, B9F12, B9J4, 2.2a and 2.2b) were the most
aggressive. Fusarium poae isolates were least aggressive, causing almost negligible disease on
both seedlings and ears. The most aggressive isolates also resulted in production of high
amounts of DON in the inoculated ears, but no DON was detected in kernels from ears
inoculated with species showing very low aggressiveness. Also one isolate of F. graminearum
(Fg7.1), which showed reduced aggressiveness as compared to the other F. graminearum
isolates, did not produce DON. The seedling and ear aggressiveness parameters showed high
positive correlation (Table 4.8). These parameters were also highly correlated to DON content
of the kernels.
4.5 Variability in mycotoxin production and aggressiveness among isolates of Fusarium
culmorum
The experiment was carried out to determine whether isolates of F. culmorum differ in
mycotoxin production and amount of head blight on wheat under field conditions. This
included differentiating the isolates on the basis of the type and amount of mycotoxin they
produce. Aggressiveness was measured as reduction in 1000 kernel weight. Twenty seven
strains of F. culmorum isolated from wheat in the Rheinlands, Germany, were used. The data
would give an idea on the prevalence of mycotoxin-producing isolates and the types of
mycotoxin that would be expected in grain under natural epidemic conditions.
4.5.1 Variation in aggressiveness and mycotoxin production on wheat ears
The twenty seven isolates of F. culmorum differed significantly in aggressiveness (Fig. 12).
Grain weight reduction varied from 10 to 35% for the less aggressive isolates and 50 to 60%
for the more aggressive isolates.
All the wheat samples from ear inoculation tests with twenty seven isolates contained either
NIV, DON or both. Out of the 27 isolates, 14 produced NIV, 9 produced DON, 3 isolates
produced both NIV and DON and 1 isolate produced both DON and 3-acDON (Fig.12). NIV
was produced in the range of 0.95 to 55 µg/g (mean 34.6 µg/g) while DON was produced in
the range of 2.9 to 74.3 µg/g (mean 40.8 µg/g). No ZEA was detected in any of the samples.
There were significant differences (p<0.001) among the isolates in terms of mycotoxin
production.
57
57
0
10
20
30
40
50
60
70
C5
C1
1
C3
3
C2
7
C2
6
C2
5
C3
4
C2
8
C4
2
C2
1
C8
C6
C3
7
C4
1
C3
8
C3
1
C3
6
C3
2
C2
4
C4
0
C3
5
C3
9
C2
0
C2
9
C2
2
C2
3
C3
0
Ker
nel
wei
ght
reduct
ion (
%)
Isolate Fig. 12. Variation in aggressiveness (reduction of 1000 kernel weight) among Fusarium
culmorum isolates on wheat ears under field conditions.
0
10
20
30
40
50
60
70
80
C5
C11
C33
C27
C26
C25
C34
C28
C42
C21
C8
C6
C37
C41
C38
C31
C36
C32
C24
C40
C35
C39
C20
C29
C22
C23
C30
NIV DON 3acDON
Isolate
Myco
toxin
conte
nt
(µg/g
)
Fig.13 Variation in mycotoxin production among 27 isolates of Fusarium culmorum in wheat
kernels after inoculation of ears under field conditions. The order of the isolates is from less
aggressive (left) to more aggressive (right).
The isolates could be distinctly divided into either DON- or NIV-producing isolates. The
NIV-producing isolates generally did not produce DON and vice versa. Almost all the DON-
producing isolates were more aggressiveness than the NIV-producing isolates. The few DON-
producing isolates that were less aggressive produced less amounts of DON compared to the
highly aggressive isolates.
58
58
4.5.2 Variation in mycotoxin production in vitro
Mycotoxin production by the twenty seven isolates of F. culmorum was determined on
autoclaved cracked corn cultures. All the 27 isolates produced either NIV or DON plus 3-
acDON. ZEA was detected in cultures of 25 isolates. The isolates which had been shown to be
highly aggressive on wheat ears produced DON while the less aggressive isolates produced
NIV (Fig. 14). Just like in the kernels, the isolates that produced DON generally did not
produce NIV.
0
1
2
3
4
5
6
C5
C11
C33
C27
C26
C25
C34
C28
C42
C21
C8
C6
C37
C41
C38
C31
C36
C32
C24
C40
C35
C39
C20
C29
C22
C23
C30
NIV DON 3acDON
Isolate
Lo
g10 M
yco
toxin
co
nte
nt
(µg
/g)
Fig. 14 Trichothecene mycotoxin production by 27 isolates of Fusarium culmorum in
autoclaved cracked corn culture. The ranking of the isolates from left to right is according to
increasing aggressiveness on wheat ears.
0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
1,8
2
C5
C1
1
C3
3
C2
7
C2
6
C2
5
C3
4
C2
8
C4
2
C2
1
C8
C6
C3
7
C4
1
C3
8
C3
1
C3
6
C3
2
C2
4
C4
0
C3
5
C3
9
C2
0
C2
9
C2
2
C2
3
C3
0
Isolate
Lo
g1
0 Z
eara
len
on
e co
nte
nt
(µg
/g)
Fig. 15 Zearalenone production by 27 isolates of Fusarium culmorum in autoclaved cracked
corn culture. The isolates are ranked from left to right according to increasing aggressiveness
on wheat ears .
59
59
Out of the 27 isolates, 16 produced NIV (range 26.2 to 220.6 µg/g, mean 96.1 µg/g) while 10
isolates produced both DON (range 27.2 to 180.8 µg/g, mean 77.7 µg/g) and 3-acDON (range
64.7 to 548.7 µg/g, mean 230.7 µg/g). Only one isolate produced all the three mycotoxins.
The amount of 3-acDON was about three times that of DON and the production of the two
mycotoxins was highly correlated (r = 0.99; Fig. 14). There were significant differences
among the isolates in total trichothecene mycotoxin production. All but two of the 27 isolates
produced detectable amounts of ZEA (range 0.23 to 68.3 µg/g, mean 7.9 µg/g). The ZEA
production differed significantly among the different isolates (Fig.15).
4.5.3 Relationship between mycotoxin production and aggressiveness
Analyses was performed to determine the association among aggressiveness on wheat ears,
mycotoxin production under field conditions and mycotoxin production in culture. The
correlation would indicate whether the amount and type of mycotoxins likely to be produced
in the field can be predicted from that produced in culture.
Inoculation of the ears with DON-producing isolates resulted in higher kernel weight
reduction than inoculation with NIV-producing isolates. The more aggressive isolates
produced DON in the ears and in solid fermentation while the less aggressive isolates
generally produced NIV. Kernel weight reduction was not correlated to NIV content of the
kernels but there was a strong significant correlation between kernel weight reduction and
DON content of the kernels (Table 4.9). Aggressiveness was also significantly correlated to
DON and 3-acDON production in vitro; however, aggressiveness was significantly negatively
correlated to NIV production but it was not correlated to ZEA production in vitro. Production
of mycotoxins DON and NIV in kernels were significantly correlated to their respective
production in vitro. Also DON content in the kernels was significantly correlated to DON and
3-acDON production in vitro and there was a high correlation (r = 0.995) between DON and
3acDON production in vitro. Trichothecene mycotoxin (NIV, DON, and 3acDON) production
in kernels and in vitro was not correlated to ZEA production in vitro.
Table 4.9 Correlation matrix (p<0.05) for kernel weight reduction and mycotoxin production
by F. culmorum isolates.
In vivo In vitro
TKW
reduction
NIV DON NIV DON 3acDON
NIV in vivo -0.304
DON in vivo 0.717 -0.806
NIV in vitro -0.670 0.476 -0.737
DON in vitro 0.603 -0.641 0.764 -0.607
3-acDON in vitro 0.603 -0.622 0.753 -0.589 0.995
ZEA in vitro -0.142 -0.165 0.012 -0.054 -0.139 -0.168
60
60
4.6 Variability in mycotoxin production and Aggressiveness among isolates of Fusarium
graminearum
The objective this experiment was to determine whether isolates of F. graminearum differ in
ability to cause head blight on wheat and to produce mycotoxins on the ears and in culture.
Aggressiveness of the isolates was measured as proportion spikelets bleached, reduction in
grain weight and fungal invasion (ergosterol content) in grain. Forty one strains of F.
graminearum isolated from wheat in Kenya and one reference isolate were used in the study.
The experiments were conducted under greenhouse conditions. The data would give a view
the prevalence of mycotoxin-producing isolates and the types of mycotoxin that would be
expected to be found in grain in Kenya.
4.6.1 Variation among isolates of F. graminearum in aggressiveness on wheat ears
All the forty two F. graminearum isolates tested were pathogenic on wheat ears but they
differed significantly (p<0.001) in the level of aggressiveness. Disease symptoms started
appearing 4 days after inoculation for the batch 1 isolate experiments which were conducted
during the humid months of March to May. Pink-coloured mycelium could be observed on the
bleached spikelets. However, symptom development was slow in the batch 2 isolate
experiments which were carried out during drier months of July to September and the initial
disease symptoms appeared only 8 days post inoculation.
0
10
20
30
40
50
60
70
K4
K6
K6
8
K7
0
Fg
7-1
K6
6
K5
8
K2
K1
5
K3
K8
K7
1
K6
4
K5
2
K5
9
K5
K1
K6
0
K1
7
Dis
ease
sev
erit
y (
%)
Isolate
a a
ab
abab
ab abab
abb
bcbc
bcbc
bc bcbc
bc
c
Fig.16 Variation in aggressiveness among 19 isolates of F. graminearum batch 1 as
determined by disease severity (% spikelets bleached) on inoculated wheat ears (Levels followed
by different letters are significantly different).
61
61
0
10
20
30
40
50
60
70
K6
1
K1
18
K1
8
K1
08
K1
35
K3
3
K7
4
K1
40
K1
51
K7
2
K1
39
K1
45
K4
2
K1
34
K3
1
K5
3
K1
09
K9
2
K1
33
K1
38
K9
3
K1
16
K5
7
Dis
ease
sev
erit
y %
Isolate
a
b
bcbc
cbcbcbc
bc
cdccc
c
cdcdcdcd
cdcd
d dd
Fig.17 Variation in aggressiveness among 19 isolates of F. graminearum batch 2 as
determined by disease severity (% spikelets bleached) on inoculated wheat ears (Levels followed
by different letters are significantly different).
In batch 1 of 19 isolates, the average disease severity ranged from 23.7% to 65.3% (mean
44.4%) spikelets bleached (Fig.16). That of the batch 2 isolates ranged from 0.45% to 38%
spikelets bleached (Fig.17). Kernels harvested from the ears infected with all the F.
graminearum isolates showed reduced weight as compared to kernels from non-inoculated
ears. Kernel weight reduction was significantly correlated to percentage spikelets bleached (r
= 0.79 for batch 1 isolates and r = 0.92 for batch 2 isolates, p<0.05). The batch 1 isolates
caused kernel weight reduction of 15.7% to 48.6% (mean 35.3%) while kernel weight
reduction due to batch 2 isolates ranged from 15% to 59.8% (mean 42.4%). Kernel weights
significantly (p<0.001) differed among the isolates. Re-isolation rates of F. graminearum
from the kernels were relatively high, with batch 1 isolate kernels having between 49% and
89% (mean 72%) re-isolation rates. Batch 2 kernels had re-isolation ranging from 11% to 81%
(mean 66%).
4.6.2 Differences among F. graminearum isolates in aggressiveness as determined by
fungal biomass (ergosterol) in infected wheat kernels
The objective was to estimate the amount of fungal mycelium in the infected kernels by
measuring ergosterol content. This would indicate the rate of kernel invasion by each isolate
and, therefore, a measure of aggressiveness of the isolates. The amount of ergosterol
determined was assumed to be due to F. graminearum.
All the samples contained substantial levels of fungal mycelium as indicated by the ergosterol
content detected (Fig. 18 and Fig. 19). The 19 batch 1 isolate samples contained 82.9 to 227.8
µg/g (mean 141.7 µg/g) ergosterol while the 23 batch 2 isolate samples had 48.9 to 141.1 µg/g
(mean 100.7 µg/g) ergosterol. Lower amounts of ergosterol (20-30 µg/g) were detected in
kernels from the controls ears. The kernel samples significantly differed (p<0.05) in the
amount of ergosterol. Kernel samples from ears with high disease severity generally contained
higher amounts of ergosterol than those from ears with low disease severity. For example,
isolates K17 and K57, from batch 1 and batch 2 respectively, resulted in high disease severity
62
62
and ergosterol contents; on the contrary, isolates K4 and K6 in batch 1 and K61 and K118 in
batch 2 caused low disease severity and ergosterol contents. An outstanding exception was a
sample from ears inoculated with isolate Fg7.1 (in batch 1). Infection of ears with this isolate
resulted in the highest amount of ergosterol but less disease severity.
0
50
100
150
200
250
K4
K6
K68
K70
Fg7-1
K66
K58
K2
K15
K3
K8
K71
K64
K52
K59
K5
K1
K60
K17
Isolate
Erg
ost
erol
(µg/g
)
Fig.18. Effect of isolates of Fusarium graminearum (batch 1) on ergosterol content in wheat
kernels after inoculation of ears under greenhouse conditions. The isolates are ranked from
left to right in the order of increasing aggressiveness on the basis of disease severity.
0
50
100
150
200
250
K61
K118
K18
K108
K135
K33
K74
K140
K151
K72
K139
K145
K42
K134
K31
K53
K109
K92
K133
K138
K93
K116
K57
Isolate
Erg
ost
erol
(µg/g
)
Fig.19 Effect of different Fusarium graminearum isolates (batch 2) on ergosterol content in
wheat kernels after inoculation of ears under greenhouse conditions. The isolates are ranked
from left to right in the order of increasing aggressiveness on the basis of disease severity.
63
63
4.6.3 Variation among F. graminearum isolates in mycotoxin production in wheat ears
The wheat grain samples from the greenhouse inoculation experiments were analysed for
mycotoxins by HPLC to determine whether the F. graminearum isolates differed in type and
amount of mycotoxin production.
0
5
10
15
20
25
K4
K6
K68
K70
Fg7-1
K66
K58
K2
K15
K3
K8
K71
K64
K52
K59
K5
K1
K60
K17
NIV DON
Myco
toxin
conte
nt
(µg/g
)
Isolate
Fig.20 Deoxynivalenol and nivalenol content in wheat kernels from ears inoculated with 19
isolates of F. graminearum (batch 1) differing in aggressiveness. The isolates are ranked from
left to right according to increasing aggressiveness on wheat ears.
0
5
10
15
20
25
K6
1
K1
18
K1
8
K1
08
K1
35
K3
3
K7
4
K1
40
K1
51
K7
2
K1
39
K1
45
K4
2
K1
34
K3
1
K5
3
K1
09
K9
2
K1
33
K1
38
K9
3
K1
16
K5
7DON NIV
Myco
toxin
conte
nt
(µg/g
)
Isolate Fig.21 Deoxynivalenol and nivalenol content in wheat kernels from ears inoculated with 23
isolates of F. graminearum (batch 2) differing in aggressiveness. The isolates are ranked from
left to right according to increasing aggressiveness on wheat ears.
64
64
Almost all the kernel samples from the inoculation experiments contained the trichothecene
mycotoxins DON or NIV. However, the acetyl derivatives of DON (3-acDON and 15-
acDON) and ZEA were not detected. In batch 1, 15 out of the 19 isolates produced DON in
the range of 2.15 to 24.7 µg/g (mean 12.0 µg/g), three isolates produced NIV in the range of
0.1 to 15.3 µg/g (mean 8.5 µg/g). No mycotoxin was detected in a sample from ears
inoculated with isolate K4, which had also been found to be least aggressive on ears (Fig.20).
Out of the 23 isolates in batch 2, 21 produced DON (range 3.5 to 22.6 µg/g, mean 7.3 µg/g).
Only 1 isolate (K109) produced NIV (7.7 µg/g; Fig 21). No mycotoxins were detected in
kernels from ears inoculated with isolate K61. None of the samples from the two batches was
found to simultaneously contain detectable levels of both NIV and DON. Kernel samples from
ears inoculated with the more aggressive isolates generally contained higher amounts of
mycotoxin than those from ears inoculated with less aggressive isolates. Kernels from ears
inoculated with isolate K17 in batch 1 contained the highest amount of DON (25 µg/g). This
isolate also caused the highest amount of disease on ears.
Also, all the three kernel samples in batch 1 found to contain NIV were from ears which had
been inoculated with less aggressive isolates (K70, Fg7.1 and K66). A similar pattern was
observed with the batch 2 kernel samples.
4.6.4 Variation among F. graminearum isolates in mycotoxin production in vitro
Mycotoxin production was determined on autoclaved cracked corn. As in the inoculated ears,
most of the isolates produced mycotoxins in solid fermentation, but in addition to DON,
15acDON and ZEA were detected. In batch 1, 15 out of the 19 isolates produced DON (17
µg/g to 575.6 µg/g, mean 165.7 µg/g) while 20 of the 23 batch 2 isolates produced DON in
the range of 3.8 µg/g to 507.4 µg/g (mean 132.1 µg/g). Most of the DON-producing isolates
also produced 15-acDON (Fig. 22 and 23). Out of the 19 isolates in batch 1, 13 produced 15-
acDON (3.1 µg/g to 81.6 µg/g, mean 26.4 µg/g) while 14 out of the 23 isolates in batch 2
produced the 15acDON (8.5 µg/g to 63.1 µg/g, mean 21 µg/g). There were significant
(p<0.05) differences among the isolates in the amounts of DON and 15acDON they produced.
The mycotoxins DON and 15-acDON were not detected for three isolates in batch1 and two
isolates in batch 2.
There was moderate to strong correlation between DON and 15acDON (r = 0.48 and 0.86,
p<0.05, for the batch 1 and batch 2 isolates respectively). Only three isolates produced NIV,
two in batch 1 (147.2 and 337.9 µg/g) and one isolate in batch 2 (256 µg/g).All the isolates
produced detectable amounts of ZEA, 6.85 µg/g to 423.2 µg/g (mean 93.6 µg/g) and 5.1 µg/g
to 637.1 µg/g (mean 106.4 µg/g) for batch 1 and 2 respectively. ZEA production was not
correlated to DON and 15acDON production, correlation coefficient ranging from 0.01 and
0.03. ZEA production significantly differed (p<0.001) among the isolates (Fig.24 and Fig. 25).
65
65
0
1
2
3
4
5
6
7
K4
K6
K6
8
K7
0
Fg
7.1
K6
6
K5
8
K2
K1
5
K3
K8
K7
1
K6
4
K5
2
K5
9
K5
K1
K6
0
K1
7
NIV DON 15-acDON
Lo
g10 M
yco
tox
in c
on
cen
trat
ion
(µ
g/g
)
Isolate
Fig.22 Trichothecene mycotoxin production by 19 isolates of Fusarium graminearum (batch
1) in autoclaved cracked corn cultures ranked from less (left) to more (right) aggressive.
0
1
2
3
4
5
6
7
K6
1
K1
18
K1
8
K1
08
K1
35
K3
3
K7
4
K1
40
K1
51
K7
2
K1
39
K1
45
K4
2
K1
34
K3
1
K5
3
K1
09
K9
2
K1
33
K1
38
K9
3
K1
16
K5
7
NIV DON 15-acDON
Log
10 M
yco
toxin
conce
ntr
atio
n (
µg/g
)
Isolate
Fig 23 Trichothecene mycotoxin production by 23 isolates of Fusarium graminearum (batch
2) in autoclaved cracked corn cultures ranked from less (left) to more (right) aggressive.
66
66
0
0,5
1
1,5
2
2,5
3
K4
K6
K6
8
K7
0
Fg
7.1
K6
6
K5
8
K2
K1
5
K3
K8
K7
1
K6
4
K5
2
K5
9
K5
K1
K6
0
K1
7
Log
10 Z
EA
conce
ntr
atio
n (
µg/g
)
Isolate
Fig.24 Zearalenone production by 19 isolates of Fusarium graminearum (batch 1) in
autoclaved cracked corn cultures ranked from less (left) to more aggressive (right).
0
0 ,5
1
1 ,5
2
2 ,5
3
K6
1
K1
18
K1
8
K1
08
K1
35
K3
3
K7
4
K1
40
K1
51
K7
2
K1
39
K1
45
K4
2
K1
34
K3
1
K5
3
K1
09
K9
2
K1
33
K1
38
K9
3
K1
16
K5
7
Log
10 Z
EA
conce
ntr
atio
n (
µg/g
)
Isolate
Fig.25 Zearalenone production by 23 isolates of Fusarium graminearum (batch 2) in
autoclaved cracked corn cultures ranked from less (left) to more (right) aggressive.
4.6.5 Relationship among aggressiveness, fungal biomass and mycotoxin production
The objective was to determine the correlation among aggressiveness, mycotoxin production
and fungal biomass for the different isolates of F. graminearum. This would indicate whether
mycotoxin production in wheat ears is dependent in the level of disease and amount of fungal
growth or whether the level of disease and fungal invasion is dependent on the type of
mycotoxin produced. DON biosynthesis rate was determined as the DON to ergosterol ratio;
67
67
amount disease per unit of DON was determined as disease severity to DON ratio; and
amount of disease per unit of ergosterol was determined as disease severity to ergosterol ratio.
The aim of using these ratios to determine the direct contribution of DON and fungal
mycelium to disease.
The different parameters for Fusarium graminearum batch 1 isolates is shown in table 4.10.
Disease severity (= % spikelets bleached) was positively correlated to kernel weight reduction,
DON content of the resulting kernels and to the amount of DON per unit of fungal biomass (=
ergosterol). Similarly, percentage kernel weight reduction was correlated to DON content, the
amount of DON per unit of fungal biomass and disease severity per unit of fungal biomass.
The rate of DON synthesis by the mycelium (DON to ergosterol ratio) was higher for the more
aggressive isolates. The amount of disease due to the presence of fungal mycelium (disease to
ergosterol ratio) was also generally higher for the more aggressive isolates, as indicated by the
significant correlation coefficient. However, there was a low correlation between disease
severity and DON synthesis in vitro. Therefore, the aggressive isolates did not necessarily
produce correspondingly high amounts of the mycotoxin in culture. This was also indicated by
the low correlation between DON content of kernels and DON synthesis in vitro. The same
was observed for kernel weight reduction and DON synthesis in vitro. However, amount of
disease was not correlated to fungal biomass and the amount of disease per unit of DON in
kernels. Production of DON by the isolates in the kernels was also not correlated to respective
production in vitro.
Similar results were obtained with the batch 2 isolates, except for some few differences (Table
4.11). Unlike in the batch 1 isolates, both disease severity and kernel weight reduction were
found to be significantly correlated to fungal biomass content but they were not correlated to
DON per unit of fungal biomass.
Table 4.10. Correlation coefficients among aggressiveness, DON production and fungal
biomass (ergosterol) formation by isolates of F. graminearum batch1.
Disease
severity
KW
reduction
DON
in vivo
DON
in vitro
Ergosterol DON/Erg Disease/DON
KW reduction 0.658 DON in vivo 0.522 0.713 DON in vitro 0.429 0.353 0.396 Ergosterol 0.257 0.323 0.677 -0.010 DON/Erg 0.631 0.801 0.916 0.527 0.377 Disease/DON 0.063 0.306 0.134 0.186 0.193 0.232 Disease/Erg 0.668 0.555 0.259 0.393 -0.362 0.519 -0.148 KW = kernel weight; DON/Erg = DON to ergosterol ratio; Disease/DON = disease severity to DON ratio;
Disease/Erg = disease severity to ergosterol ratio.
68
68
Table 4.11 Correlation coefficients among aggressiveness, DON production and fungal
biomass (ergosterol) formation by isolates of F. graminearum batch 2.
Disease
severity
KW
reduction
DON
in vivo
DON
in vitro
Ergosterol DON/Erg Disease/DON
KW reduction 0.924 DON in vivo 0.501 0.534 DON in vitro 0.494 0.382 0.436 Ergosterol 0.808 0.850 0.683 0.387 DON/Erg 0.387 0.433 0.950 0.384 0.526 Disease/DON 0.395 0.288 -0.117 0.414 0.122 -0.075 Disease/Erg 0.877 0.735 0.243 0.401 0.500 0.246 0.548 KW = kernel weight; DON/Erg = DON to ergosterol ratio; Disease/DON = disease severity to DON ratio;
Disease/Erg = disease severity to ergosterol ratio.
4.7 Differences among wheat varieties in susceptibility to F. graminearum head blight
and mycotoxin accumulation
Susceptibility of fifteen wheat varieties to Fusarium head blight was assessed by inoculation
with F. graminearum under greenhouse conditions. These varieties are commercially grown in
Kenya. The objective was to determine whether these varieties are susceptible to Fusarium
head blight and the associated mycotoxin accumulation.
4.7.1 Variation among varieties in Fusarium graminearum head blight susceptibility
All the 15 varieties were susceptible to Fusarium head blight but they significantly differed
(P< 0.01) in susceptibility (Fig. 26). Depending on the variety, symptoms appeared as early as
5 days post inoculation and increased at different rates in different varieties. The shape and
area under disease progress curve differed depending on the susceptibility of a given cultivars
(Fig. 27). The disease increased slowly in the less susceptible varieties and the least
susceptible cultivar (Nungu) showed longer incubation period.
Average disease severity measured as percent of spikelets bleached ranged from 29.4% for
‘Fahari’ to 68% for ‘Nata’ (mean 54.1%). In terms of area under disease progress curve
(AUDPC), disease severity ranged from 476 for cv. ‘Nungu’ to 1220 for cv.’Duma’ (mean
870). Percent spikelets bleached was significantly correlated (r = 0.68, p < 0.001) to area
under disease progress curve. There were significant differences (p<0.001) among the
varieties in head blight susceptibility. Mean ear yield reduction ranged from 23% to 57%
(mean 44%). Kernel weight reduction was significantly different (p<0.001) among the
varieties. Re-isolation rate of F. graminearum from the kernels ranged from 40% (‘Fahari’) to
60% (‘Ngamia’), with a mean of 52.8%.
69
69
0
200
400
600
800
1000
1200
1400
NunguNyangumi
FahariKware
ChirikuTembo
Popo
NgamiaKongoni
Mbega Paka
Pasa Nata
Mbuni
Duma
AU
DP
C
Variety
a
abab ab
abab
ab
ab
b bb
bb
bb
Fig.26 Differences among 15 wheat varieties in susceptibility to head blight, measured as the
area under disease progress curve (AUDPC), after inoculation with F. graminearum in
greenhouse pot experiments in 1999 (Levels followed by different letters are significantly different).
0
10
20
30
40
50
60
70
80
90
100
0 10 15 20 25 30
Nyangumi
AUDPC = 632
Time (days)
Sev
erit
y (
%)
0
10
20
30
40
50
60
70
80
90
100
0 10 15 20 25 30
Nungu
AUDPC = 476
Time (days)
Sev
erit
y (
%)
Less susceptible
0
10
20
30
40
50
60
70
80
90
100
0 10 15 20 25 30
Duma
AUDPC = 1225
Time (days)
Sev
erit
y (
%)
0
10
20
30
40
50
60
70
80
90
100
0 10 15 20 25 30
Mbuni
AUDPC = 1175
Time (days)
Sev
erit
y (
%)
More susceptible
Fig. 27 Differences in shape of disease progress curves of less susceptible and more
susceptible wheat varieties after inoculation with Fusarium graminearum in greenhouse pot
experiments.
70
70
Table 4.12 Correlation coefficients among the head blight susceptibility parameters for wheat
varieties after inoculation with Fusarium graminearum.
Disease severity AUDPC Kernel weight reduction
AUDPC 0.688 Kernel weight reduction 0.308 0.405 Re-isolation 0.310 0.082 0.360
Kernel weight reduction was not correlated with percent spikelets bleached (r = 0.31, p =
0.10) and to area under disease progress curve (r = 0.40, p = 0.10; Table 4.12). However, the
grain yield was significantly correlated to the 1000 kernel weight (r = 0.78, p<0.01). Re-
isolation of F. graminearum from the kernels was not correlated to the corresponding disease
severity (r = 0.31).
4.7.2 Variety differences in grain mycotoxin content and fungal biomass
The wheat kernels were analysed for mycotoxin and ergosterol contents to determine whether
the varieties differed in mycotoxin accumulation and fungal invasion, respectively. The
measured ergosterol was assumed to due to presence of F. graminearum mycelium.
The kernels of 15 wheat varieties contained DON in the range of 5.3 µg/g for cv. ‘Fahari’ to
31.6 µg/g, for cv. ‘Pasa’ (mean 13.5 µg/g). The acetyl derivatives of DON (3-acDON and 15-
acDON) and ZEA were not detected. The varieties differed significantly (p<0.05) in DON
content of kernels (Fig.28).The more susceptible varieties contained higher amounts of DON
than the less susceptible varieties. Susceptibility, based on the area under disease progress
curve, was positively correlated (r = 0.64, p<0.05) to DON content of the kernels (Table 4.13).
Therefore, the more susceptible varieties generally contained higher amounts of DON.
However, maximum DON content (31.6 µg/g) was found in cv ‘Pasa’ as compared to DON
content of the two most susceptible varieties, ‘Mbuni’ and ‘Duma’ with 17.7 µg/g and 9.5
µg/g, respectively.
Fungal mycelium was detected in kernels of all varieties, as indicated by ergosterol content.
The amount of ergosterol ranged from 67 µg/g for cv. ‘Popo’ to 187.8 µg/g for cv. ‘Pasa’
(mean 110.6 µg/g ). There was significant difference (p<0.05) among the varieties in
ergosterol content of the kernels. Ergosterol content pattern generally followed that of DON
(Fig.28 and 29), the two being highly positive correlated (r = 0.96, p<0.05; Table 4.13). The
less susceptible varieties contained low ergosterol content and the very susceptible varieties
had high ergosterol contents. Deoxynivalenol production rate by the pathogen, which was
calculated as the amount of DON per unit of ergosterol, was generally higher in the more
susceptible varieties as compared to the less susceptible varieties. This was illustrated by the
significant positive correlation (r = 0.61) between head blight rating (AUDPC) and
DON/ergosterol ratio. The amount of ergosterol per unit of DON (Ergosterol/DON ratio) was
found to be generally low for the more susceptible varieties, the two being negatively
correlated (r = -0.6).
71
71
0
5
10
15
20
25
30
35
Nungu
Nyangumi
Fahari
Kware
Chiriku
Tembo
Popo
Ngamia
Kongoni
Mbega
Paka
Pasa
Nata
Mbuni
Duma
Variety
DO
N c
on
ten
t (µ
g/g
)
Fig.28 Variation in deoxynivalenol (DON) content in kernels of 15 wheat varieties that differ
in susceptibility to head blight after inoculation with Fusarium graminearum. Varieties are
arranged from less susceptible (left) to more susceptible (right).
0
20
40
60
80
100
120
140
160
180
200
NunguNyangumi
Fahari
Kware
Chiriku
Tembo
Popo
Ngamia
Kongoni
Mbega
Paka
Pasa
Nata
Mbuni
Duma
Erg
ost
erol
conte
nt
(µg/g
)
Variety Fig. 29 Ergosterol content in wheat kernels of 15 varieties that differ in susceptibility to head
blight after inoculation of ears with Fusarium graminearum. The varieties are ranked from
left to right according to increasing level of susceptibility.
It was observed that some varieties showed similar disease rating but differed in fungal
biomass and mycotoxin content of the kernels (Fig. 30). For example cultivars ‘Pasa’, ‘Nata’,
‘Mbuni’ and ‘Nduma’ showed significantly the same disease severity rating but differed in
ergosterol and DON contents of the kernels. This is also shown by the ergosterol to DON
ratios (Table 4.14). Cultivars ‘Pasa’ and ‘Duma’ had similar disease ratings (64% and 66%,
respectively) but differed in ergosterol to DON ratio (5.3 and 9.8, respectively). Therefore, the
higher ratio for cultivar ‘Duma’ shows that though it was similarly susceptible to infection
like ‘Pasa’, it contained fungal mycelium without a corresponding amount of DON.
72
72
A Less susceptible varieties
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
3 0 0
3 5 0
N u n g u N y an g u m i F ah a ri K w are C h irik u T em b o
D is eas e s ev e rity (% ) K ern e l w e ig h t red u c tio n (% )
E rg o s te ro l (µ g /g ) D O N (1 0 X ; µ g /g )
Variety
B More susceptible varieties
0
50
100
150
200
250
300
350
M b ega P aka P as a Nata M b uni Nd um a
Dis eas e s everity (% ) Kernel weight red uc tio n (% )
Ergo s tero l (µ g/g) DO N (10X; µ g/g))
Variety
Fig. 30. Comparison between less susceptible and more susceptible wheat varieties in head
blight ratings, fungal biomass (ergosterol) and deoxynivalenol content in kernels after
inoculation with Fusarium graminearum.
73
73
Table 4.13 Correlation coefficients (p<0.005) among head blight ratings, DON and ergosterol
contents in wheat kernels of varieties differing in susceptibility after inoculation with F.
graminearum.
AUDPC KW reduction DON
Ergosterol Ergosterol/DON
KW reduction 0.403
DON 0.637 0.083
Ergosterol 0.564 0.088 0.916
Ergosterol/DON -0.597 -0.005 -0.788 -0.524
DON/Ergosterol 0.608 0.136 0.874 0.632 -0.959
AUDPC = area under disease progress curve; KW = kernel weight; Ergosterol/DON = ergosterol to DON ratio;
DON/Ergosterol = DON to ergosterol ratio.
Table 4.14 Effect of wheat variety on the head blight rating, pathogen re-isolation rate,
deoxynivalenol and ergosterol content parameters after inoculation with Fusarium
graminearum.
Variety Control
grain wt (g)
Spike
bleached (%)
Kernel wt.
reduction (%)
re-isolation
(%)
DON/Erg
ratio
Erg/DON
ratio
Nungu 8.0 51.1bc
38.8 54 0.10 9.88
Nyangumi 9.1 62.2c 56.7 58 0.06 16.57
Fahari 9.8 29.4a 41.0 40 0.07 14.95
Kware 9.9 54.5bc
22.9 49 0.10 9.82
Chiriku 9.7 31.7ab
40.8 46 0.08 12.43
Tembo 8.0 40.5ab
40.7 55 0.08 13.21
Popo 10.5 45.5b 38.7 59 0.09 11.14
Ngamia 12.9 43.1ab
52.0 60 0.15 6.89
Kongoni 8.7 63.9c 43.8 55 0.08 12.9
Mbega 13.4 60.7bc
36.0 54 0.15 6.45
Paka 9.3 64.9c 56.3 56 0.15 6.71
Pasa 10.5 64.0c 44.2 55 0.17 5.29
Nata 10.7 68.7c 43.8 45 0.14 7.20
Mbuni 10.6 64.8c 49.5 55 0.12 7.85
Duma 9.6 66.2c 52.1 51 0.10 9.81
74
74
4.8 Effect of pure Fusarium mycotoxins on wheat seedlings in vitro
Wheat inoculation studies and in vitro mycotoxin production experiments showed that DON
producing isolates caused more disease than the NIV producing isolates, suggesting a role of
DON in disease development. Therefore, this experiment was carried out to determine the
effect of pure mycotoxins in the absence of the pathogen. The result would give a hint on
which of the mycotoxins, NIV, ZEA, DON and acetylated derivatives, has an effect on wheat
tissues.
None of the mycotoxins affected seed germination, even at 100 µg/ml. The acetylated
derivatives of DON (3acDON and 15acDON) caused similar effect like DON (Fig.31). No
significant differences were noted among seedlings treated with DON, 3acDON and
15acDON. However, the weights of the seedlings treated with NIV and ZEA were not
significantly different from those of the controls. Exposure to DON, even at 1 µg/ml (3.37 x
10-6
M), caused reduction in shoot and root mass and overall stunting of the seedlings as
compared to the untreated controls (Fig. 32a; Plate 7). However, pronounced reduction in
shoot and root dry weight was observed at 5 µg/ml (1.69 x 10-5
M) DON. There was a gradual
increase in weight reduction at DON concentrations above 5 µg/ml. Exposure to 1 µg/ml
DON resulted in 21% fresh weight reduction, 5 µg/ml (52%), 10 µg/ml (58%), 20 µg/ml
(65%), 50 µg/ml (75%) and 100 µg/ml (86%). The respective dry weight reductions were
17%, 31%, 40%, 48%, 61%, and 77%. The water and the acetonitrile treatments were not
significantly different from the 1 µg/ml DON treatment but these three treatments were
significantly different from the 5 µg/ml DON treatment. However, the weight reductions due
to 10 µg/ml (3.37 x 10-5
M), 20 µg/ml 6.75 x 10-5
M), 50 µg/ml (1.69 x 10-4
M) and 100
µg/ml (3.37 x 10-4
M) treatments were not statistically different.
On the contrary, NIV and ZEAR produced no significant effect on the seedling weights
(Fig.31b and c). NIV caused only a slight reduction in fresh weight at 50 (1.6 x 10-4
M) and
100 µg/ml (3.2 x 10-4
M), but this was not significantly different from the untreated seedlings.
100 µg/ml NIV resulted in only 39% reduction in fresh weight and 29% reduction in dry
weight. All the ZEA treatments resulted in less than 25% fresh and dry weight reductions.
0
50
100
150
200
250
Control NIV ZEA DON 3-acDON 15-acDON
Dry
wei
ght
(mg
)
a
a
b b
b
a
Fig. 31 Effect of different Fusarium mycotoxins (10 µg/ml each) on wheat seedling dry
weight.
75
75
76
76
0
50
100
150
200
250
C o ntro l 0 1 5 10 20 50 100
Dry
wei
ght
(mg)
Concentration (µg/ml)
aa
a
bbc
bc
c
c
0
50
100
150
200
250
Control 0 1 5 10 20 50 100
Dry
wei
ght
(mg)
Concentration (µg/ml)
a
a
a
a
a
a
a
a
0
50
100
150
200
250
Control 0 1 5 10 20 50 100
Concentration (µg/ml)
Dry
wei
gh
t (m
g)
aa
aa
a
a
a
a
Fig. 32 Effect of varying mycotoxin concentration on wheat seedling dry weight.
4.9 Polymerase chain reaction-based characterization of Fusarium isolates
In the preceding experiments isolates of different Fusarium species were found to show
variation in aggressiveness and mycotoxin production on wheat ears. They also showed
variation in quantity and quality of mycotoxin produced in culture.. Therefore, polymerase
chain reaction (PCR) analysis was performed to determine whether these differences could
a
Deoxynivalenol
b
Nivalenol
c
Zearalenone
77
77
also be observed at the DNA level. This would give a hint to genetic contribution to
aggressiveness and mycotoxin production of a given Fusarium isolate.
4.9.1 Specific PCR identification of Fusarium culmorum and F. graminearum isolates
Polymerase chain reaction using 20-base pair species-specific primers was carried out to
confirm the morphological identification of isolates of F. culmorum and F. graminearum.
The primer pairs used were GATGCCAGACCAAGACGAAG/
GATGCCAGACGCACTAAGAT for F. culmorum and GCAGGGTTTGAATCCGAGAC /
AGAATGGAGCTACCAACGGC for F. graminearum. Twenty seven strains of F. culmorum
isolated from wheat were used. The F. graminearum strains used in this experiment included
5 isolates (BMM1, B9F12, B9J4, 2.2a and 2.2b) from maize kernels collected in Germany and
8 isolates (K31-10, K11-12, K11-13, K17-011, K31-1, K31-5, K31-8, K31-11 and K32-1)
from wheat kernels collected in Kenya and 1 reference isolate (Fg7.1).
All the twenty seven isolates of F. culmorum showed a common band of the amplified DNA
fragment of 450bp (Plate 7). PCR analysis of 14 isolates of F. graminearum, including the
reference isolate (Fg7.1), produced a common amplified DNA fragment band of 330 bp (Plate
8). One isolate (2.2a) of F. graminearum, however, did not show the characteristic species
specific band indicating misidentification of this isolate.
200 bp
300 bp
400 bp
500 bp
600 bp
700 bp
800 bp
1000 bp
1500 bp
2000 bp
M C27 C29 C25 C32 C31 C24 C36 C23 C21 C28 C30 C22 C34 C37 C26 C5 C6 C20 C35 C8 C38 C39 C40 C41 C42 C11 C33 C M
Plate 7. Agarose gel electrophoresis bands of DNA from 27 Fusarium culmorum isolates
amplified by species specific PCR using primer pair 5’GATGCCAGACCAAGACGAAG 3’/
5’GATGCCAGACGCACTAAGAT 3’. M = molecular weight marker; C = control (PCR mix
without DNA).
78
78
B
MM
1
B9
F1
2
B9
J4
2
.2a
2
.2b
Fg
7.1
K3
1-1
0
K
11
-12
K
11
-13
K1
7-0
11
K
31
-1
K3
1-5
K
31
-8
K3
1-1
1
K
32
-1
M
200 bp
300 bp
400 bp
500 bp
600 bp
700 bp
800 bp
1000 bp
1500 bp
Plate 8. Gel electrophoresis bands of DNA from 15 Fusarium graminearum isolates amplified
by species specific PCR using primer pair 5’GCAGGGTTTGAATCCGAGAC 3’ /
5’AGAATGGAGCTACCAACGGC 3’. M = 100 bp molecular weight marker
4.9.2 Differentiation of Fusarium species by RAPD-PCR analysis
The DNA of twenty isolates belonging to eight Fusarium species was subjected to polymerase
chain reaction with ten-base pair random primers with the objective of determining the
differences among isolates. The isolates tested belonged to F. culmorum (4 isolates), F.
graminearum (4), F. avenaceum (3), F. poae (3), F. moniliforme (1), F. chlamydosporum (1),
F. oxysporum (1), F. sporotrichioides (1) and other Fusarium spp. (2). The nine primers used
for the random amplified polymorphic DNA-PCR analysis were primer P1
(GGGGGTTAGG), P2 (GTGCTCGTGC), P3 (CCTGGCGGTA), P4 (CCGGCCTTAA), P5
(CCGGCCTTAG), P6 (GATGCCAGAC), P7 (GTCCGGAGTG), P8 (GAGCACCAGG), and
P9 (GGGCCACTCA).
RAPD-PCR analysis with 8 out of the 9 primers showed amplification of the Fusarium DNA
resulting in banding patterns after gel electrophoresis. These were primers P2, P3, P4, P5, P6,
P7, P8 and P9 (Plate 9 and 10). Reactions with primer P1 did not produce banding patterns.
The banding patterns of RAPD-PCR with different primers showed amplified DNA fragments
with varying degrees of polymormism. Reactions with primer P4 produced the most
polymorphic fragments and therefore showed no species specific bands among most isolates.
On the contrary, patterns with primer P7 showed low polymorphism of the amplified
fragments and isolates of F. graminearum, F. culmorum, F. avenaceum and F. poae had a
common band of 700 bp. Detailed differences among the isolates in different species was
revealed by dendrograms derived from cluster analysis of the banding pattern scores and
similarity coefficients. Table 4.15 and Table 4.16 illustrate examples of banding pattern scores
and the corresponding Jaccards similarity coefficients, respectively.
79
79
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P2
GTGCTCGTGC
200 bp
800 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P3
CCTGGCGGTA
100 bp
800 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P4
CCGGCCTTAA
800 bp
200 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P5
CCGGCCTTAG
200 bp
800 bp
Plate 9. PAPD-PCR banding patterns of different Fusarium species: Fg = F. graminearum, Fc
= F. culmorum, Fa = F. avenaceum, Fp = F. poae, Fm = F. moniliforme, Fcl = F.
chlamydosporum, F = Fusarium sp., Fs = F. sporotrichioides; M = 100 bp molecular weight
marker.
80
80
Fg Fg Fg Fg Fc Fc Fc Fc Fc Fa Fa Fa Fp Fp Fp Fm Fcl Fg Fo F Fs M
Primer P6
GATGCCAGAC
800 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P7
GTCCGGAGTG
800 bp
200 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P8
GAGCACCAGG
200 bp
800 bp
Fg Fg Fg Fg Fc Fc Fc Fc Fa Fa Fa M Fp Fp Fp Fm Fcl Fg Fo F Fs C
Primer P9
GGGCCACTCA
800 bp
200 bp
Plate 10. RAPD-PCR banding patterns of different Fusarium species: Fg = F. graminearum,
Fc = F. culmorum, Fa = F. avenaceum, Fp = F. poae, Fm = F. moniliforme, Fcl = F.
chlamydosporum, F = Fusarium sp., Fs = F. sporotrichioides; M = 100 bp molecular weight
marker. Banding patterns for the empty lanes were proved in replicate reactions using the
same order of the isolates.
81
81
Table 4.15 Scores of RAPD-PCR banding pattern obtained with primer P7 (GTCCGGAGTG)
and DNA from isolates of different Fusarium species. 1 = band present; 0 = band absent.
Isolate DNA fragment size (base pairs)
250 330 360 400 480 500 600 650 700 800 900 1000 1100 1200 1600 Total
Fg7.1 0 0 0 0 0 0 0 1 1 1 1 0 0 0 1 5
K8 0 0 1 0 0 0 0 1 1 1 1 0 0 1 1 7
K2 0 0 1 0 0 0 0 1 1 1 1 0 0 1 1 7
K66 1 0 0 0 0 0 0 1 1 1 0 0 1 0 0 5
FC3:1 0 0 1 0 0 0 0 1 1 1 1 0 0 0 1 6
C26 0 1 0 0 0 0 0 1 1 1 0 0 0 0 0 4
C11 0 1 0 0 0 0 0 1 1 1 0 0 0 1 1 6
C37 0 1 0 0 0 0 0 1 1 1 0 0 0 0 0 4
K122 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 3
K63 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 3
K159 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 3
FPIwt 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 3
K97 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 2
K23 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 2
F.mon 0 0 0 1 0 0 0 1 0 0 0 1 0 1 0 4
K136 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 2
K141 0 0 0 1 1 0 0 1 0 0 1 0 0 1 0 5
Foxy3 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 2
B7M1 0 1 0 0 0 0 0 0 1 0 1 0 0 0 0 3
6,4 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2
%similarity 10 20 15 10 5 5 15 50 80 45 60 5 5 35 25
Table 4.16 Example of Jaccards similarity coefficient (%) matrix among isolates of Fusarium
species calculated from the RAPD-PCR banding patterns scores presented in table 4.15 above.
K8 71
K2 71 100
K66 50 38 38
Fc31 67 71 71 40
C26 50 38 38 60 29
C11 43 50 50 29 43 50
C37 50 38 38 60 29 100 50
K122 33 25 25 17 33 17 14 17
K63 33 25 25 17 33 17 14 17 100
K159 33 25 25 17 33 17 14 17 100 100
FpIwt 29 22 22 33 29 33 29 33 40 40 40
K97 40 40 40 20 40 20 17 20 67 67 67 50
K23 33 25 25 17 33 40 33 40 50 50 50 75 67
Fmo. 13 22 22 14 13 14 29 14 0 0 0 0 0 0
K136 0 13 13 0 0 0 17 0 0 0 0 0 0 0 20
K141 25 33 33 13 25 13 25 13 14 14 14 13 17 17 50 17
Foxy 17 29 29 20 17 25 40 25 25 25 25 25 33 25 20 33 17
B7M 33 25 25 17 33 17 33 17 50 50 50 75 67 40 0 0 14 20
B64 17 13 13 50 0 25 0 25 0 0 0 20 0 0 0 0 0 0 0
Fg71 K8 K2 K66 Fc31 C26 C11 C37 K122 K63 K159 FpIwt K97 K23 Fmo. K136 K141 Foxy B7M1
82
82
Cluster analysis showed that RAPD-PCR with primers P2, P6 and P9 resulted close
association among isolates belonging to a particular species and therefore differentiated the
isolates according to respective species. However, better differentiation of the isolates was
attained with cluster analysis of the mean similarity coefficients calculated from scores of
banding patterns of all the 8 primers. Different clusters were formed by isolates of F.
graminearum, F. culmorum, F. avenaceum and F. poae, respectively (Fig 33). The only
exception was isolate Fc3.1 of F. culmorum clustered together with isolates of F.
graminearum. A single cluster was formed by isolates belonging to other Fusarium species.
Fig 34 illustrates the differentiation of the Fusarium species by RAPD-PCR with primers P2,
P6 and P9.
K8
Fg
7.1
K2
K6
6
Fc3
.1
C2
6
C1
1
C3
7
K1
22
K6
3
K1
59
FP
Iwt
K9
7
K2
3
Fm
on
K1
36
K1
41
Fo
xy
3
B7
M1
6.4
F. p
oae
F.a
vena
ceum
F. gra
min
earu
m
F. c
ulm
orum
F.mon
ilifo
rme
F. chl
amyd
ospo
rum
Fusar
ium
sp.
F. o
xysp
orum
Fus
ariu
m s
p.
F. s
poro
tric
hioi
des
Means for 8 primers
Fig. 33 Differentiation of Fusarium isolates into their respective species by cluster analysis of
mean similarity coefficients calculated from banding pattern scores of RAPD-PCR with 8
primers. Isolate Fc3.1 (F. culmorum) clusters together with isolates of F. graminearum.
83
83
F
G
F
G
FG
F
ox
y
FG
FC
FC
FC
FC
FP
FP
FP
F.s
p.
FA
FA
FA
FM
FS
FG
FC
L
Primer P2 (GTGCTCGTGC)
FG
FG
FG
F
ox
y
FG
F
C
F
C
F
C
FC
FP
FP
FP
F.s
p.
F
A
FA
F
A
F
M FS
FG
FC
L
F
C
Pimer P6 (GATGCCAGAC)
FG
F
G
FG
F
ox
y
FG
FC
FC
FC
FC
FP
FP
FP
F.s
p.
FA
FA
FA
F
M FS
FG
FC
L
Primer P9 (GGGCCACTCA)
Fig. 34. Differentiation of Fusarium species by cluster analysis of the RAPD-PCR banding
pattern score: FG = F. graminearum, FC = F. culmorum, FA = F. avenaceum, FM = F.
moniliforme, FP = F. poae, FS = F. sporotrichioides, FCL = F. chlamydosporum, Foxy = F.
oxysporum.
84
84
RAPD-PCR analysis with isolates of four major wheat head blight pathogens, F. culmorum,
F. graminearum, F. avenaceum, and F. poae , was compared to an isolate of F. oxysporum.
The banding patterns were characteristic for each species and the isolates could be identified
by their characteristic banding patterns (Plate 11). A major characteristic band was observed
at 700 bp for F. graminearum, 400 bp for F. culmorum, 600 bp for F. avenaceum and 1100 bp
for F. poae. The F. oxysporum isolate formed a banding pattern different from all the other
species. Cluster analysis of the banding pattern scores showed a clear differentiation among
the species, with isolates of each species forming a separate cluster (Fig.35). The isolate of F.
oxysporum was completely different from all the other four species. A test with isolates of F.
graminearum and F. culmorum only also showed clear differences between the two species
(Plate 11 and Fig.35).
Plate 11. RAPD-PCR banding patterns showing differences among the major wheat Fusarium
head blight fungi Fusarium graminearum (Fg), F. culmorum (Fc), F. avenaceum (Fa) and F.
poae (Fp) in comparison with F. oxysporum (Fo); M = 100 bp molecular weight marker; C =
control (PCR mix without DNA .
Fg Fg Fg Fg Fc Fc Fc Fc Fc Fa Fa Fa Fa Fa Fp Fp Fp Fp Fp Fo M
Primer P6
GATGCCAGAC
200
300
400
800
F. graminearum F. culmorum F. avenaceum F. poae
200 bp
800 bp
Fg Fg Fg Fg Fg Fg Fg Fg M Fc Fc Fc Fc Fc Fc Fc Fc Fc Fc
Primer P8
GAGCACCAGG
Fusarium graminearum F. culmorum
85
85
F. graminearum F. culmorum F. avenaceum F. poae
F
G
F
G
FG
F
ox
y
FG
FC
FC
FC
FC
F
P
FP
FP
FA
FA
FA
FA
FA
FP
FP
Fusarium spp.; primer P6 (GATGCCAGAC)
F
G
F
G
FG
FG
FC
FC
FC
FC
FG
FG
FG
FG
FG
FG
FC
FC
FC
FC
FC
FC
Fusarium graminearum and F. culmorum;
primer P8 (GAGCACCAGG)
F. graminearum F. culmorum
Fig. 35. Clustering of isolates of the major Fusarium head blight pathogens according to their
respective species after cluster analysis of RAPD-PCR banding patterns score. Foxy = F.
oxysporum as a comparison.
The amplified the DNA fragments were mainly in the region between 200 base pairs to 1000
base pairs. A skewed distribution of amplified DNA fragment sizes was observed, with low
amplification above 1400 bp (Fig. 36).
86
86
0
20
40
60
80
100
120
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2
F. culmorum F. graminearum
Fragment size (kbp)
No
. o
f b
and
s
Fig 36. Distribution of randomly amplified DNA fragment bands after RAPD-PCR with with
6 primers and template DNA from isolates of F. culmorum and F. graminearum.
4.9.3 Variation among isolates of Fusarium culmorum as determined by RAPD-PCR
analysis.
The DNA of F. culmorum isolates was characterized by polymerase chain reaction using eight
ten-base pair random primers. The random primers used were P4 (CCGGCCTTAA), P5
(CCGGCCTTAG), P6 (GATGCCAGAC), P7 (GTCCGGAGTG), P8 (GAGCACCAGG), and
P9 (GGG CCACTCA).
RAPD-PCR with the different primers resulted in different banding patterns and the amplified
DNA fragments showed varying degrees of polymorphism (Plate 12 and 13). RAPD reactions
with primer P4 showed a high number of polymorphic DNA bands while low polymorphism
was observed in reactions with primer P8and P9. Therefore, more differences among the
isolates could be observed on the banding patterns with primer P4. On the contrary, the bands
produced by reactions with primer P8 and P9 were common to most of the isolates. The mean
percentage similarity among the isolates varied from 35% for banding patterns with primer P4
to 67% with primer P8 (Table 4.17).
Prominent banding patterns could be observed for a number of isolates. For example, banding
patterns for isolates C28 and C40 were different from those of all the other isolates with most
primers. Similarly, isolates C11, C28, C32, C24, C29 and C22 showed two characteristic
bands at about 1300 bp and 1800 bp each, in the RAPD profile with primer P7. Common
bands to all the isolates were observed on RAPD profiles with primer P6 (450 bp), P7 (680
bp), P8 (200 and 230 bp), and primer P9 (250, and 640 bp).
87
87
C5 C11 C34 C28 C21 C8 C38 C31 C32 C24 C36 C25 C41 C40 C35 C39 C20 C29 C22 C23 C30 M
Primer P4
CCGGCCTTAA
800 bp
300 bp
C5 C11 C33 C27 C26 C34 C28 C42 C21 C25 C41 C29 C35 C39 C20 C8 C6 C37 C38 C31 C M
Primer P5
CCGGCCTTAG
800 bp
100 bp
C5 C11 C34 C28 C21 C8 C38 C31 C32 C24 C36 C25 C41 C40 C35 C39 C20 C29 C22 C23 C30 M
Primer P6
GATGCCAGAC
800 bp
300 bp
Plate 12. Differentiation of Fusarium culmorum isolates by RAPD-PCR banding patterns
obtained with different primers; M = 100 base pair(bp) molecular weight marker.
88
88
C5 C11 C34 C28 C21 C8 C38 C31 C32 C24 C36 C25 C41 C40 C35 C39 C20 C29 C22 C23 C30 M
Primer P7
GTCCGGAGTG
800 bp
200 bp
C5 C11 C34 C28 C21 C8 C38 C31 C32 C24 C36 C25 C41 C40 C35 C39 C20 C29 C22 C23 C30 M
Primer P8
GAGCACCAGG
800 bp
200 bp
C5 C11 C34 C28 C21 C8 C38 C31 C32 C24 C36 C25 C41 C40 C35 C39 C20 C29 C22 C23 C30 M
Primer P9
GGGCCACTCA
800 bp
500 bp
200 bp
Plate 13. Differentiation of Fusarium culmorum isolates by RAPD-PCR banding patterns
obtained with different primers; M = 100 base pair(bp) molecular weight marker.
89
89
Table 4.17 Similarity (%) among 21 F. culmorum isolates determined by RAPD-PCR with 6
different primers.
Primer Minimum Maximum Mean similarity
P3 5 90 44
P4 5 86 35
P6 5 100 53
P7 5 95 44
P8 5 100 67
P9 5 100 48
More information on the differences among the isolates was obtained by cluster analysis of
the Jaccards similarity coefficients calculated from the RAPD-PCR banding pattern scores for
each of the 6 primers. Dendrogram of the mean similarity coefficients of the banding patterns
with the six primers showed that each isolate differed from every other (Fig 37). Isolates C28
and C40, which had shown outstanding bands, formed a cluster different from the all the other
isolates. However, there was neither a complete differentiation between NIV- and DON-
producing isolates nor association between aggressiveness and the clustering of isolates
according to RAPD-PCR banding patterns. Two minor clusters containing only DON-
producing isolates were evident. One cluster comprised of isolates C35, C20 and C39 and the
other was made up of isolates C22, C23 and C30. Similarly, a cluster of isolates C11, C31,
C32, and C24, which were NIV-producing, could be observed. Examples of dendrograms
derived from cluster analysis of RAPD-PCR banding pattern scores are illustrated in fig 38 for
primer P4 and P7.
C3
2
C5
C1
1
C3
4
C2
8
C2
1
C8
C3
8
C3
1
C
23
C30
C22
C29
C20
C39
C35
C40
C41
C
25
C36
C24
21 isolates; mean for six primers
Fig. 37. Variation among isolates of Fusarium culmorum as determined by cluster analysis of
the mean similarity coefficients of RAPD-PCR banding patterns ( = NIV-producing isolates;
isolate C36 also produced DON).
90
90
C32
C5
C11
C34
C2
8
C21
C8
C38
C31
C23
C30
C22
C29
C20
C39
C35
C40
C41
C25
C36
C24
C37
C6
C42
C26
C27
C33
27 isolates; Primer P4 (CCGGCCTTAA)
C
32
C
5
C
11
C3
4
C
28
C2
1
C
8
C3
8
C
31
C
23
C
30
C2
2
C2
9
C2
0
C39
C3
5
C4
0
C4
1
C
25
C36
C2
4
21 isolates; primer P7 (GTCCGGAGTG)
Fig. 38. Variation among isolates of Fusarium culmorum as determined by cluster analysis of
RAPD-PCR banding pattern scores for primer P4 and P7 ( = NIV-producing isolates; isolates C33,
C27 and C36 also produced small amounts of DON in kernels from inoculated wheat ears).
4.9.4 Differentiation of Fusarium graminearum isolates by RAPD-PCR
Isolates of F. graminearum were analyzed by RAPD-PCR to determine the intraspecific
variation in relation to mycotoxin production and aggressiveness to wheat. The batch 1
isolates were analysed using six random primers while the batch 2 isolates were analysed with
three primers. The primers used were P3 (CCTGGCGGTA), P4 (CCGGCCTTAA), P6
(GATGCCAGAC), P7 (GTCCGGAGTG), P8 (GAGCACCAGG), and P9 (GGGCCACTCA).
The isolates differed in RAPD-PCR banding patterns with the different primers (Plate 14 and
15). The mean similarity among the isolates ranged from 28 to 63% for the batch 1 isolates
and 33 to 67% for the batch 2 isolates (Table 4.18). Like in the case of F. culmorum, more
differences among the isolates could be noted on RAPD-PCR banding patterns with primer P4
and low differences were observed on banding patterns with primer P8 and P9. The banding
patterns showed some differences between NIV- and DON-producing isolates, especially on
RAPD-PCR profiles with primer P4. All the DON-producing isolates , except isolate K57,
showed a common prominent band of 1200 bp which was not observed on the NIV-producing
isolate banding patterns.
91
91
Primer P3 batch 1
CCTGGCGGTA
K1 K2 K3 K4 K5 K6 K8 K15 K17 K52 K58 K59 K60 K64 K66 K68 K70 K71 Fg71 K61 K144 M
800 bp
300 bp
Primer P4 batch 1
CCGGCCTTAA
K1 K2 K3 K5 K6 K8 K15 K17 K52 K58 K59 K60 K64 K68 K71 K70 K66 Fg71 K61 K144 K4 M
800 bp
300 bp
200 bp
K1
39
K1
18
K1
40
K1
44
K1
08
K6
1
K1
33
K1
35
K1
8
K1
09
K5
3
K1
38
K3
1
K1
34
K9
3
K1
16
K4
2
K1
45
K9
2
MK3
3
K5
7
Primer P4 batch 2
CCGGCCTTAA
800 bp
300 bp
K1 K2 K3 K5 K6 K8 K15 K17 K52 K58 K59 K60 K64 K68 K71 K70 K66 Fg71 K61 K144 K4 M
Primer P6 batch 1
GATGCCAGAC
400 bp
800 bp
Plate 14. RAPD-PCR banding patterns obtained with different primers showing differences
and similarities among isolates of Fusarium graminearum. NIV-producing isolates are K70,
K66, Fg7.1, K61, K144, K4 and K109. M = 100 base pair (bp) molecular weight marker.
Banding patterns for the empty lanes were confirmed in repeat experiments.
92
92
K1 K2 K3 K5 K6 K8 K15 K17 K52 K58 K59 K60 K64 K68 K71 K70 K66 Fg71 K61 K144 K4 M P7 batch 1
GTC CGG AGT G
200 bp
800 bp
K1 K2 K3 K4 K5 K6 K8 K15 K17 K52 K58 K59 K60 K64 K66 K68 K70 K71 Fg71 K61 M K144
Pimer P8 batch 1
GAGCACCAGG
800 bp
100 bp
Primer P9 batch 1
GGGCCACTCA
K1 K2 K3 K4 K5 K6 K8 K15 K17 K52 K58 M K59 K60 K64 K66 K68 K70 K71 Fg71 K144
100 bp
800 bp
Primer P9 batch 2GGGCCACTCA
K7
4
K1
51
K7
2
K1
18
K6
1
KF
g71
K9
2
K1
16
K9
3
K1
38
K1
8
K1
35
K1
08
K3
3
K1
40
K1
45
K4
2
Co
ntr
ol
M
K1
39
K5
7
800 bp
200 bp
Plate 15. RAPD-PCR banding patterns obtained with different primers showing differences
and similarities among isolates of Fusarium graminearum. NIV-producing isolates are K70,
K66, Fg7.1, K61, K144 and K4. M = 100 base pair (bp) molecular weight marker. Banding
patterns for the empty lanes were confirmed in repeat experiments.
93
93
Examples of prominent polymorphic bands were like the 1400 bp for isolates K15, K66,
Fg7.1, K139, K140, K144K109 and K57 on the profiles of Primer P4 and the 1200 bp band
for the isolates K3, K70, K151, K72, K61, K108 and K33 on profiles of primer P9.
Characteristic bands common to all the isolates were observed on RAPD profiles of primers
P7 (680 bp), P8 (1100 bp and approximately 2000 bp) and P9 (500 bp and 650 bp).
Table 4.18 Similarity (%) among isolates F. graminearum determined by RAPD-PCR with
different primers.
Primer No. of isolates Minimum Maximum Mean similarity
P3- batch 1 21 5 100 54
P4- batch 1 21 5 71 28
P6- batch 1 21 5 100 54
P7- batch 1 21 5 100 53
P8- batch 1 21 10 100 61
P9- batch 1 21 10 100 63
P4 - Batct 2 25 4 100 33
P6- Batch 2 25 4 100 67
P9- Batch 2 24 4 100 61
Cluster analysis of the mean Jaccards similarity coefficients for the six primers showed a clear
differentiation between DON- and NIV-producing isolates, each forming a separate cluster
(Fig.39). Only isolate K61 formed a cluster alone. Most of the DON-producing isolates
showed close similarity within 8% Euclidean distance. However, within the DON or the NIV
cluster, no relationship was observed between the grouping of the isolates and their respective
aggressiveness to wheat ears. Examples of other dendrograms from cluster analysis of the
RAPD-PCR banding pattern scores are shown in Fig.40 for primer P3 and P6. The DON-
producing isolates showed high similarity and tended to cluster together for most primers.
K1
K5
K3
K2
K6
K8
K1
5
K1
7
K5
2
K5
8
K5
9
K6
0
K6
4
K6
8
K7
1
K7
0
K6
6
FG
71
K6
1
K1
44
K4
Batch 1 isolates; means for 6 primers
Fig. 39. Variation among isolates of Fusarium graminearum determined by cluster analysis of
the mean similarity coefficients of RAPD-PCR banding patterns with 6 primers (= NIV-
producing isolates).
94
94
K1
K5
K3
K
2
K6
K8
K1
5
K1
7
K5
2
K5
8
K5
9
K6
0
K6
4
K6
8
K7
1
K7
0
K6
6
Fg
71
K6
1
K1
44
K4
(%) Batch 1 isolates; prmer P3 (CCTGGCGGTA)
K1
K5
K3
K2
K6
K8
K1
5
K1
7
K5
2
K5
8
K5
9
K6
0
K6
4
K
68
K7
1
K7
0
K6
6
F
g7
1
K6
1
K1
44
K4
Batch 1 isolates; primer P6 (GATGCCAGAC)
Fig. 40 Dendrograms derived from RAPD-PCR banding pattern scores showing relationship
among isolates of Fusarium graminearum for reactions with primer P3 and P6 (= NIV-
producing isolates).
95
95
Chapter 5: Discussion
Isolates of Fusarium species were isolated from wheat kernels and identified by
morphological and species-specific PCR characteristics. The isolates were further
characterized to determine the differences among isolates on the basis of aggressiveness to
wheat, mycotoxin production and RAPD-PCR banding patterns.
5.1 Fungal contamination of wheat samples
Mycological examination of kernels of the wheat samples from different localities showed
high prevalence of Epicoccum spp. and Alternaria spp.. These fungi are mainly saprophytic
and widely distributed, reproducing on diseased, dead, or ripe plant material, especially during
periods of high humidity (Zillinsky, 1983). Infection of wheat heads with these fungi causes
grey or black discoloration on heads and seed resulting in symptoms referred to as sooty
moulds, black point, or smudge. This may reduce yields and grain quality. Moreover, some
Alternaria species, like A. alternata, are known to produce mycotoxins (e.g. alternariol) which
are possible food contaminants with a potential toxicological risk (Bottalico and Logrieco,
1996). The fungi reported to contaminate wheat grain in other areas include Alternaria spp.,
Epicoccum spp., Fusarium spp., Cladosporium spp. Nigrospora spp., Mucor spp., Penicillium
spp., Aspergillus spp. and Drechslera spp. (Dexter et al. 1997; Gonzalez et al. 1999; Stack
and McMullen, 1985; Vrabcheva et al. 1996). Storage fungi such as Aspergillus and
Penicillium were detected at very low levels and only in a few samples collected from the
markets. This shows that there is a potential risk of mycotoxins like, aflatoxin and ochratoxin,
produced by these fungi. This can be of great concern especially because majority of the small
scale farmers may store some of their crop for a number of months sometimes under
unfavourable conditions.
The Fusarium species isolated from wheat in Kenya in order of decreasing frequency were F.
poae, F. graminearum, F. avenaceum, F. oxysporum, F. equiseti, F. camptoceras and F.
chlamydosporum. All these Fusarium species are potential mycotoxin producers but F.
graminearum is the most aggressive on wheat and produces the more toxic type B
trichothecenes NIV and DON (Marasas et al. 1984; Langseth et al. 1999). Some strains of
Fusarium poae have also been reported to produce NIV (Langseth et al. 1999; Sugiura et al.
1993). Fusarium species occasionally isolated from wheat grown under temperate climates
include F. culmorum, F. graminearum, F. poae, F. sporotrichioides, F. equiseti, F.
tricinctum, and F. torulosum (Langseth et al. 1999; Schütze et al. 1997). The absence of F.
culmorum in Kenya is probably because this species is favoured by cooler maritime climates
(Parry et al. 1995). Interestingly, F. poae was found to be dominant in some regions in Kenya,
although it is reported to be a temperate species. Fusarium poae was most prevalent in
samples from Njoro (Nakuru) and Timau (Meru), which are cooler. On contrary, F.
graminearum was most prevalent in samples from warmer regions of Nyandarua and Laikipia.
Prevalence of Fusarium species was found to be quite low (2-7% contamination) indicating
that level of Fusarium head blight was also low. However, the observed contamination levels
may not reflect the actual infection level in the field because the severely infected kernels are
light enough to be expelled with chaff during harvesting (Bai and Shaner, 1994). These results
also indicate that F. graminearum is widespread in the wheat growing area in Kenya and this
inoculum is potentially capable of producing severe head blight epidemics and mycotoxin risk
given optimum weather conditions. Critical conditions include a wheat crop at flowering to
hard milk stage and temperatures higher than 18° C accompanied by more than 5 mm rainfall
96
96
for at least 24 h (Obst, 1997; Parry et al. 1995). In addition, the growing of maize, which is
the main staple grain, and continuous wheat cropping may increase this risk. Higher head
blight infection and mycotoxin load has been observed when wheat crop is preceded by maize
or wheat in rotation (Dill-Macky and Jones, 2000; Obst 1997). Maize and wheat stubble are
good reservoirs for F. graminearum on which a primary source of inoculum could develop
(Parry et al. 1995; Sutton, 1982).
5.2 Fungal biomass and ergosterol production in vitro
Ergosterol is a useful measure of fungi in grains because it detects viable and non-viable
fungal propagules whereas the plating technique detects only the viable fungi and often
requires long incubation periods (Gourama and Bullerman, 1995) Ergosterol is the
predominant sterol in fungal cell membranes and is specific to fungi, seldom present in
animals or plants (Schwadorf and Müller, 1989; Tothill et al. 1992). The main sterols of
plants are -sitesterol and sigmasterol. Some members of Graminae contain low
concentrations of cholesterol. The major disadvantage of the ergosterol assay is lack of
specificity. Ergosterol extracted from naturally infected grain may be from a variety of
different fungi, including saprophytes (Gordon and Webster, 1984; Padgett and Posey, 1993).
The primary advantage of the ergosterol assay relative to other methods is that results can be
obtained in about 48 h and it might prove useful in inoculation experiments and where a more
rapid screening procedure is required.
Growth of both Fusarium culmorum and F. graminearum was found to be associated with
amount of ergosterol. The exponential growth phase in liquid and solid media was observed to
last up to 15 to 20 days of incubation for both species. The active growth phase was evidenced
by the high increase in mycelia dry weight and ergosterol while the stationary growth phase
was indicated by levelling off of the two parameters. These observations are in agreement
with what has been reported elsewhere. In a study with F. graminearum in culture Padgett and
Posey (1993) and Zill et al. (1988) found that ergosterol levels increased rapidly during the
first two weeks of incubation, indicating rapid growth of mycelium, followed by a period of
minor production, which demonstrated the beginning of the stationary growth phase. In liquid
cultures, ergosterol content remained constant until 14 days when mycelia began to
senescence.
Miller et al. (1983a) observed that maximum protein content occurred at day 8, while
maximum fungal biomass occurred at day 10. Both values declined sharply after reaching
their maxima. Ergosterol concentrations continued to increase well after protein and fungal
biomass decreased, suggesting that the enzymes of secondary metabolism were still active.
The ergosterol content of a distinct fungus may be influenced by substrate composition, water
availability and in laboratory media, by the extent of aeration and the growth phase of the
mycelium (Bermingham et al. 1995; Padgett and Posey, 1993; Tothill et al, 1992; Zill et al.
1988). The physiological status of mycelium changes with age and this affects the ergosterol
content. Experiments with different media in this study showed that ergosterol content was
correlated to mycelium biomass indicating that it not a secondary metabolite but a constituent
of the mycelium.
Ergosterol yield in culture and inoculated ears has been found to be correlated to fungal
biomass or the viable count (Love and Seitz, 1987; Miller et al., 1983b; Murkovic et al. 1997;
Tothill et al. 1992). In corn inoculated with F. graminearum the colony forming units (CFU)
of F. graminearum slowly increased until 6 weeks after inoculation and then decreased
97
97
(Miller et al. 1983b). F. graminearum was the only fungus found in samples from week 2 to
week 7 and increase in ergosterol during the 50 days after inoculation roughly paralleled the
increase in the viable count of F. graminearum. The rate of increase then slowed to zero
towards harvest (week 9) as the viable count precipitously decreased. These observations
indicate that fungal biomass of F. culmorum and F. graminearum is related to ergosterol
content and the ergosterol assay can be confidently used as a measure of fungal invasion in
cereals infected by these pathogens.
5.3 Dynamics of mycotoxin production in vitro
In this study, solid fermentation cultures were found to be more suitable for mycotoxin
production than liquid fermentation. Mycotoxin yields were much higher and more reliable in
the solid fermentation. This is in line with the results of Miller et al. (1983a) who observed
that solid state fermentation is easy to duplicate and yields higher amounts of mycotoxins.
Therefore, sterilised moist grains are most commonly used to produce mycotoxins. Natural
substrates, such as grains, probably have the optimum nutrient composition for fungal growth
and mycotoxin production, as opposed to synthetic nutrient medium. Factors that affect the
production of mycotoxins in culture include incubation temperature, initial moisture content,
pH, size of the culture flask, oxygen and carbon dioxide concentrations and nutrient
composition (Greenhalgh et al. 1983; Miller and Greenhalgh, 1985; Miller et al. 1983a;
Pestka et al. 1985).
In solid fermentation, ZEA was mainly produced at 19° C and its concentration increased with
both moisture content and time, with highest production being after 40 days at the highest
moisture content. The optimum temperature for DON and acetylated DON production in
liquid and solid fermentation was found to be 28° C at 35% initial moisture content
(Greenhagh et al. 1983). The best yields of trichothecenes in liquid fermentation occurred at
an initial pH of 6.5 while peak ZEA occurred at an initial pH of 7.0. The pH of rice culture at
28° C and 35% initial moisture content varied from 6.3 at day 1 to 5.7 at day 28. The degree
of aeration determines the mixture of mycotoxins produced. Deoxynivalenol, ZEA and
pigment were only produced in the 500 ml flasks, and small amounts of acetyl DON were
produced only in the 1500 ml flask (Greenhalgh et al. 1983). No mycotoxins were detected in
the 2000 ml and 2800 ml flasks, although fungal growth was good in all sizes of flasks. In the
flasks where DON and ZEA were produced, high levels of carbon dioxide (12%) but trace
amounts of oxygen were found. All the bigger flasks were found to have oxygen levels similar
to external air and carbon dioxide levels of 0.02%.
Miller et al. (1983a) found that highest yields of trichothecene (DON and acetyl-DON) in
liquid fermentation were produced at a surface :volume ratio of 0.97 under still conditions.
Under shake conditions (increased oxygen) F. graminearum produced more T-2 toxin and
ZEA. Optimal sucrose concentration for DON production is between 1% and 3% (Pestka et
al. 1985; Miller et al. 1983a). Higher sucrose concentrations increase mycelial dry weight but
decrease the final DON yield. Corn steep liquor (CSL) and low levels of ammonium tartrate
were found to stimulate DON production rather than 15-acDON (Pestka et al. 1985).
Therefore, presence of CSL might enhance production or activity of the enzymes involved in
conversion of 15-acDON to DON and a single nutrient can greatly affect the DON yield in a
complex medium.
Media containing high carbon to nitrogen ratio have also been found to give highest
trichothecene yields (Desjardins et al. 1993).This explains the observed low levels of DON
98
98
and high levels of ZEA detected in well aerated cultures in the present study. Nutritional
composition can also be a possible explanation of the observed differences in the amount of
mycotoxins produced in the media tested. Barath et al. (1997) observed no significant
influence of substrate on mycotoxin recovery but toxin synthesis by individual strains
depended on the substrate. The finding of the present study that cracked corn was a better
substrate for mycotoxin production agrees with that of Barath et al. (1997) who found that
zearalenone, zearalenol and T-2 production was higher in corn than in rice for all the species
they tested.
Detectable levels of DON were observed at the 10th
day and maximum levels were observed
during the stationary growth phases of both F. culmorum and F. graminearum. However,
acetyl DON was detected early in the growth phases of both fungi and maximum levels
occurred just at the end of the exponential growth phase. Low levels of zearalenone were
initially detected towards the end of the exponential growth phase for both Fusarium species
but the levels did not increase until the late stationary phase. These observations indicate that
acetylated DON is formed very early during the active growth phase of the fungus and precede
DON, which is formed during the stationary phase and ZEA is formed during the stationary
and decline phases of growth. This agrees with the observations of Abouzied and Pestka
(1986), Greenhagh et al. (1983), Miller et al. (1983a), and Pestka et al. (1985) for F.
graminearum in culture and Miller et al. (1983b) in inoculated corn. In inoculated corn, Miller
et al. (1983b) found that F. graminearum viable count and ergosterol increased during 6 days
after inoculation and 15-acDON concentration rose rapidly in the first 2 weeks and then
slowly declined. Deoxynivalenol increased until week 7 while ZEA remained low (<1 ppm)
for the first half of the sampling period and increased to ca. 10 ppm in the last week.
A similar pattern of fungal growth and mycotoxin production was also observed for the NIV-
and DON-chemotypes of F. graminearum in culture (Abouzied and Pestka, 1986). The NIV-
chemotype produced fusarenone X (FX) while the DON-chemotype produced 15-acDON as
the acetylated forms. Fusarenone X and 15-acDON appeared early in the fermentation but
disappeared concomitantly with the accumulation of the deacetylated forms, NIV and DON,
respectively. In rice, both NIV- and DON-chemotypes initially accumulated NIV and FX, or
DON and 15-acDON, respectively. However as the fermentation progressed, FX and 15-
acDON levels declined with a concurrent increase in NIV and DON levels, respectively. The
observations that the acetylated forms (FX and 3-acDON or 15-acDON) preceded the de-
acetylated forms (NIV and DON), respectively supports the evidence that FX and 3-acDON or
15-acDON are biosynthetic precursors of NIV and DON, respectively (Hesketh et al. 1991).
The biosynthetic pathway sequence of trichothecenes has been established (Appendix 3).
Biosynthesis occurs through a series of enzymatic reactions beginning with the cyclization of
farnesyl pyrophosphate to trichodiene by the enzyme trichodiene synthase (Desjardins et al.
1993; Proctor et al. 1995a). The sequence of oxygenations, isomerizations, cyclizations, and
esterifications leading from trichodiene to the more complex trichothecene toxins, such as T-2
toxin, and 3-acetyldeoxynivalenol has been established through experiments with F.
sporotrichioides, Giberrella pulicans, F. culmorum and F. graminearum (Beremend, 1987;
Desjardins et al. 1986, 1987, 1993; Fekete et al. 1997; Hesketh et al. 1991; McCormic et al.
1990). The biosynthetic pathway genes have also been identified. Tri5 (Tox5) encodes
trichodiene synthase (Fekete et al. 1997; Hohn and Beremend, 1989b; Hohn and Desjardins,
1992). Tri3 (Tox3) encodes a transacetylase that converts 15-decalonectrin to calonectrin, and
Tri4 (Tox4) encodes a cytochrome P-450 monooxygenase that converts trichodiene to an
oxygenated product (Hohn et al. 1995; Proctor et al. 1995a). Tri6 gene acts as the positive
99
99
regulator of the other trichothecene biosynthetic genes and therefore if Tri5 is active, but Tri6
is not, toxic trichothecenes are not synthesised by the cell (Proctor et al. 1995a). The Tri5
nucleotide sequence has been characterized in F. sprotrichioides (Hohn and Beremend,
1989b), Gibberella pulicans (Hohn and Desjardins, 1992), Gibberella zeae (Proctor et al.
1995b). Enzyme activity and trichodiene synthase polypeptide have been detected and
preceded the initial detection of trichothecenes by about 3 hours in media containing a high
carbon to nitrogen ratio (Hohn and Beremend, 1989a).
The study also indicates that, like most secondary metabolites, mycotoxins are produced
during the stationary phase of growth. Secondary metabolite formation is often associated
with some form of morphological differentiation of the fungus (Moss, 1996).
5.4 Mycotoxin production by isolates of Fusarium culmorum and F. graminearum in
vitro
Different studies indicate that mycotoxin production is a common feature of Fusarium
culmorum and F. graminearum and almost all isolates investigated were capable of producing
mycotoxins (Ichinoe et al. 1983; Mirocha et al. 1989; Sugiura et al. 1990; Szecsi and Bartok,
1995; Gang et al. 1998; Miedaner et al. 2000; Muthomi et al. 2000). In the present study, all
the isolates of Fusarium culmorum and F. graminearum produced mycotoxins in autoclaved
cracked corn. The mycotoxins detected were NIV, DON and ZEA for both species. The
acetylated derivative of DON detected was 3-acDON for F. culmorum and 15-acDON for F.
graminearum. This indicates that these two species produce a similar spectrum of mycotoxins
and this observation has been reported elsewhere (Marasas et al. 1984; Atanassov et al. 1994;
Lauren et al. 1992; ).
The ability of F. culmorum to produce NIV has only been recently detected (Atanassov et
al.1994; Gang et al. 1998; Langseth et al. 1999; Mirocha et al. 1994; Muthomi et al. 2000).
The high percentage of mycotoxin producing isolates observed indicate that natural head
blight epidemics should result in DON and NIV accumulation in grain. This is supported by
the findings of Müller et al. (1997) that 69-96% of all randomly collected wheat samples from
farmers’ fields in a 4 year survey in Southwest Germany were contaminated with DON and
25-64% with NIV. Studies have also shown that NIV- and DON-producing isolates form the
respective mycotoxins under field and in vitro conditions (Gang et al. 1998; Muthomi et al.
2000).
Therefore, in vitro mycotoxin studies can be used to predict the type of mycotoxins produced
by individual isolates in the field. For both species, the DON production was correlated to the
acetylated form of DON. In the case of F. culmorum, 3-acDON was the major product, being
about 3 times the amount of DON for most isolates. However, 15-acDON was the minor
product for the isolates of F. graminearum. This difference can be explained by the
differences in the length of incubation period. The F. culmorum fermentations were incubated
for 14 days while those of F. graminearum were incubate for 21 days. Therefore, the F.
culmorum cultures were still in the late exponential growth phase when production of the
acetylated DON is highest. The longer incubation period for F. graminearum means that the
cultures were analysed during the stationary growth phase when most of the 15-acDON had
been deacetylated to DON.
The production of both the acetylated and the deacetylated form of DON show that the
biosynthesis of DON goes via the acetylated form. This is supported by the fact that the
100
100
deacetylated DON is rarely detected in naturally contaminated grain. In naturally
contaminated products like grain, more time is available for the enzymes of the fungi and
probably also enzymes of the grain, to perform the deacetylation (Miller et al. 1983b).
The isolates of both F. culmorum and F. graminearum also varied in the amount of
mycotoxins produced. Therefore, the type and amount of mycotoxins produced depends
highly on the isolate investigated (Marasas et al. 1984; Atanassov et al. 1994; Muthomi et al.
2000). Fusarium. graminearum and F. culmorum are the main producers of ZEA (Langseth et
al. 1999; Marasas et al. 1984). The current study showed that zearalenone was produced by all
the isolates of F. graminearum investigated and only 2 isolates of F. culmorum did not
produce detectable levels of the toxin. The lower amounts of ZEA and lack of detection for 2
isolates in F. culmorum is probably because the cultures were harvested in early stationary
phase, as opposed to late stationary phase in F. graminearum. Time course studies of
mycotoxin production had shown that zearalenone reaches maximum production during late
stationary phase. There was no correlation between DON and ZEA production, indicating that
high quantities of DON in cereals is not automatically followed by large concentrations of
ZEA, and vice versa.
The isolates of both species could be grouped into either DON- or NIV- producing. Miller et
al. (1991) first proposed chemotaxonomy of Fusarium species in the section Discolor into 2
chemotypes. These are chemotype I (DON producers) and chemotype II (NIV, 4-acNIV
producers). Chemotype I was divided into 2 groups based on the type of actylated DON:
chemotype IA (DON, 3-acDON producers) and chemotype IB (DON, 15-acDON producers).
The F. graminearum isolates were proposed to be of both chemotype I and II (NIV- and
DON-producers) while F. culmorum isolates were proposed to belong to chemotype II (DON-
producers) only. The DON-chemotypes have been found to produce small quantities of NIV
and vice versa (Blany and Dodman, 1988; Langseth et al. 1999; Sugiura et al. 1990; Szecsi
and Bartok, 1995). A few isolates of F. culmorum used in the present study were also found to
produce both NIV and DON. The two chemotypes of F. graminearum strains are extensively
reported (Blany and Dodman, 1988; Ichinoe et al. 1983; Leonov et al. 1990; Miller et al.
1991; Sugiura et al. 1990; Sydenham et al. 1991; Szecsi and Bartok, 1995). The results of this
study show that the mycotoxin profiles of F. culmorum and F. graminearum are very similar
and the isolates of F. culmorum can also be classified as either NIV- and DON-chemotypes.
This observation is supported by reports of other investigators who have also reported the
existence NIV-producing F. culmorum strains (Atanassov et al. 1994; Gang et al. 1998;
Langseth et al. 1999; Lauren et al. 1992; Mirocha et al. 1994).
All the DON-chemotypes of F. culmorum isolates from Germany were found to form 3-
acDON and therefore belong to chemotype IA. About 60% and 40% of the F. culmorum
isolates formed NIV and DON, respectively, suggesting that both mycotoxins are likely
contaminate naturally infected wheat in Germany. Naturally infected wheat in Germany has
been reported to be contaminated with both NIV and DON (Gleissenthal et al. 1989; Müller
and Schadorf, 1993; Müller et al. 1997). The results showed that the F. graminearum isolates
from Kenya are mainly DON-chemotypes, with low frequency of NIV-chemotypes. Therefore,
in case of a natural infection DON is most likely to be found contaminating wheat in Kenya.
NIV- and DON- chemotypes have been reported to infect wheat and corn in South Africa
(Sydenham et al. 1989, 1991). Most of these isolates also produced 15-acDON and therefore
belong to chemotype IB. None of the isolates formed perithecia in culture and therefore F.
graminearum isolates from Kenya also belong to group 1.
101
101
The differences in structure of the acetylated trichothecenes produced could be based on the
presence of specific enzymes such as 3- or 15-acetyldeoxynivalenol deacetylase and the
oxygenases that induce oxidation at C-3 and the C-4 positions (Miller et al. 1991; Ichinoe et
al. 1983 ). However, Sugiura et al. (1990) suggested that 3- and 15-acetylation of DON are
dependent on environmental temperature and therefore, the differences in the production of 3-
acDON or 15-acDON are not chemotaxonomic criteria. A regional differences in the
distribution of the two chemotypes has been observed (Miller et al. 1991; Mirocha et al.
1989). No exclusive relationship has been observed between ZEA production and the NIV- or
DON-chemotypes (Szecsi and Bartok, 1995). On the basis of pathogenicity and cultural
characteristics, Blany and Dodman (1988), observed no distinction between the NIV- and
DON-chemotypes of F. graminearum. The different chemotypes are not restricted to any
particular grain type (Lauren et al. 1992) but some authors have observed some relationship
between NIV- or DON-chemotypes and certain crops (Sydenham et al. 1991; Szecsi and
Bartok, 1995; Bottalico et al. 1989).
5.5 Mycotoxin production on wheat ears
The isolates produced the same type of mycotoxin as in culture, except that ZEA and 15-
acDON were not detected. 3-acDON was detected at low levels for only one isolate of F.
culmorum. The isolates showed great variation in the quantity of mycotoxins they produced.
Miedaner et al. (2000) observed a highly significant coefficient of correlation between
locations for mycotoxin production in rye inoculated with F. graminearum, therefore,
suggesting that the majority of variation for DON and NIV contents is caused by genetic
effects. This indicates that the isolates used in the present study have a big potential to form
mycotoxins under natural epidemic conditions and the frequency of mycotoxin-producing
isolates in natural populations is high. Since the isolates naturally occur in a mixture, the
results imply that infected grain would be simultaneously contaminated with NIV and DON.
Co-occurrence of NIV and DON in naturally infected crop has been reported in Korea (Kim et
al. 1993; Sohn et al. 1999), Germany (Gleissenthal et al. 1989; Müller and Schwadorf, 1993;
Müller et al. 1997), Poland (Perkowski et al. 1990; Visconti et al. 1986), the Netherlands
(Tanaka et al. 1990), and South Africa (Syndenham et al. 1989). Nivalenol has been detected
in wheat from regions where F. culmorum was the most likely pathogen (Mirocha, et al.
1994). However, because NIV is about 10 to 15 times more toxic than DON (Mirocha et al.
1989; Yoshizawa and Morooka, 1973), the average toxicity of the NIV samples should be
higher than those of the DON-contaminated samples. Therefore, to estimate the potential
health hazard of Fusarium contaminated grain for animal and human consumption, it is
insufficient to analyse feed and food only for DON. Lack of detection of the 3-acDON and 15-
acDON in the samples was most likely due to complete deacetylation to DON by the time the
crop matured. The samples most likely contained the acetylated DON in trace amounts below
the detection limit. Similar findings have been reported by Miedaner and Perkowski (1996),
Miller et al. (1983b), Snijders (1990) and Vrabcheva et al. (1996).
Deoxynivalenol production by isolates of F. culmorum was found to be correlated to DON
production in vitro but no significant correlation was found for the F. graminearum isolates.
Nivalenol content in kernels of ear inoculated with F. culmorum was weakly correlated to
NIV production in culture. Gang et al. (1998) found no correlation between the amount of
DON produced in field-grown grain and grain incubated in vitro. In addition, the two batches
of F. graminearum isolates differed in the mean mycotoxin detected in the kernels. Batch 1
kernels had higher mycotoxin content than the batch 2 kernels. This was caused by the
102
102
differences in the environmental conditions during disease development. Infection and disease
development for the batch 1 isolates occurred during cooler and more humid months, unlike
the warmer, drier months for the batch 2 isolates. Environmental conditions greatly influence
the quantity of mycotoxin produced (Gang et al. 1998; Miedaner and Perkowski, 1996). The
authors observed that the amount of DON or NIV produced across 2 field locations on one rye
genotype was not associated with the mycotoxin content produced in vitro on grain of the
same genotype.
5.6 Aggressiveness of isolates on wheat ears
According to Vanderplank (1984) aggressiveness designates the quantity of disease caused by
a pathogenic isolate on a susceptible host while the isolate does not interact differentially with
a host cultivar. Aggressiveness depends not only on the genotypes of the pathogen and host
but also on the environment in which it is assessed (Andrivon, 1993). Pathogen
aggressiveness and host resistance both show a continuos variation with no isolate being fully
nonaggressive and no host genotype being fully resistant (Miedaner et al. 1996).
Aggressiveness in F. culmorum is governed by several genes and/or an allelic series at one
locus with the alleles having a different contribution to aggressiveness (Miedaner et al. 1996).
Qualitative and quantitative differences in enzyme and mycotoxin production are possible
causes for the great variation in aggressiveness.
In the present study, the isolates of both F. culmorum and F. graminearum showed significant
variation in aggressiveness, as determined by kernel weight reduction, disease severity and
ergosterol content. In experiments with Fusarium graminearum, disease severity (proportion
of bleached spikelets) and ergosterol content of the kernels was also determined. Disease
severity was correlated (r = 0.66 to 0.92) with kernel weight reduction. This is in agreement
with findings of Miedaner et al. (1996), indicating that either of the parameters can be used as
a measure of aggressiveness. Ergosterol was used as a measure of fungal invasion in the
kernels (Seitz et al. 1977, 1979; Schwadorf and Müller, 1989). Liquid fermentation studies
with F. culmorum and F. graminearum confirmed that mycelium biomass highly correlated to
ergosterol content, therefore showing that the latter is a reliable indirect measure of fungal
invasion. The measured ergosterol content was assumed to be due to F. graminearum
infection and this was confirmed by the high re-isolation rates of the pathogen. Virtually no
other fungi were detected in the kernels and the low levels of ergosterol in controls show that
cross contamination was minimal. Variation in severity of Fusarium head blight due to
different isolates of F. graminearum and F. culmorum has also been reported in rye by Gang
et al. (1998) Miedaner (1997), Miedaner et al. (2000), Miedaner and Schilling (1996),
Miedaner et al. (1997), and in wheat by Snijders (1990), and Snijders and Perkowski (1990).
Miedaner et al. (1997) observed high heritability estimates of aggressiveness and DON
production in field experiments with 42 isolates of F. culmorum indicating that a substantial
proportion of the observed phenotypic variation is caused by genetic effects. A high level of
genetic diversity for vegetative compatibility groups (VCG) has been found among field
isolates of F. graminearum from wheat heads (Bowden and Leslie, 1994). Therefore, sexual
recombination probably occurs frequently because different VCGs could be found on the same
wheat head. The observed genetic diversity resulting in variation in aggressiveness could be a
result of mutation and recombination (Miedaner et al. 2000). Mutation should play a
significant role because the large amount of propagules generated during each infection cycle.
Outcrossing and sexual recombination in G. zeae has been demonstrated under laboratory
103
103
conditions (Bowden and Leslie, 1992, 1999). Therefore, heterozygous perithecia and sexual
recombination could also occur under field conditions.
Also, asexual recombination is assumed to occur in nature in both F. graminearum and F.
culmorum (Puhalla, 1981). The alternating life style of F. graminearum and F. culmorum with
parasitic and saprophytic phase in crop rotations with a high percentage of cereals should
maintain a high level of genetic diversity by balancing selection when different physiological
mechanisms are needed for both lifestyles (Miedaner et al. 2000). Under these conditions, less
aggressive isolates should be maintained in the population at a certain frequency because they
can survive as secondary colonisers attacking plants during ripening or even as nectrophs.
However, F. graminearum and F culmorum show a low level of pathogenic host
specialisation in regard to host species and/or host organs and a high climatic adaptiveness
(Miedaner and Schilling, 1996; Miedaner et al. 1996).
The results showed that DON producing isolates were more aggressive than the NIV
producing isolates. The less aggressive DON producing isolates generally showed less toxin
production than the highly aggressive isolates. In contrast, no correlation (r = -0.34) was found
between relative kernel weight reduction and NIV content. This indicates a possible role of
DON in pathogenesis. This proposition is supported by the correlation (r = 0.68) between
ergosterol and DON content in both batches of F. graminearum experiments. The means that
isolates generating high amounts of ergosterol also produced medium to high amounts of
DON. Therefore, the differences in DON production could be a consequence of the different
amounts of fungal biomass present in the kernels.
These observations are in agreement with the findings of Miedaner et al. (2000) in winter rye.
The fact that the NIV-producing isolates were among the least aggressive suggests that NIV is
less phytotoxic than DON. This was confirmed in in vitro experiments with pure mycotoxins
bioassays. Exposure of wheat seedlings to 1 ppm to 100 ppm DON resulted in increasing
reduction in dry weight, but similar concentrations of NIV and ZEA had do significant effect
on seedling growth. The acetylated forms (3-acDON and 15-acDON) were found to have
similar effects like DON. Similar effects of DON and /or other trichothecenes at low
concentration have been reported by Desjardins et al. (1993) and Cutler and Jarvis (1985) on
various plants; Bruins et al. (1993), Wang and Miller (1988), Snijders (1990) on wheat
seedlings, coleoptile segments, embryos and calli (Scholbrock et al. (1992) on wheat leaves
and Adams and Hart (1989) on maize ears and carnation. Deoxynivalenol and 15-acDON
induced lesions on wheat leaves and the severity of the lesions was positively correlated with
head blight susceptibility (Scholbrock et al. 1992). Cassale and Hart (1988) showed that low
levels of DON inhibit protein synthesis in plants, since DON blocks various parts of the peptidyl
transferase reaction. The effect of DON on protein synthesis possibly causes the reported
symptoms on plants exposed to the toxin.
Pathogenicity is the ability to cause disease, while aggressiveness is the amount of disease caused
(Yoder, 1980). A toxin is a pathogenicity factor when it is essential for a pathogen to cause
disease, while a toxin is a virulence factor when it increases the extent of disease (Mitchell,
1984). Generally virulence factor toxins are non-host-specific whereas pathogenicity factor toxins
are host-specific. Non-host-specific toxins may produce part or all of the symptoms of the disease
and may constitute only one small part of the disease process caused by the pathogen (Mitchell,
1984). Toxins can cause either typical or atypical symptoms depending on concentration and on
assay procedures (Yoder, 1980). Some toxins known to be involved in disease also produce
physiological changes found in diseased plants while other toxins may exert toxic effects without
necessarily reproducing the typical physiological symptoms. Presence of a toxin in diseased
104
104
tissues may only indicate that when the pathogen grows, the toxin is produced. Therefore, the
toxin may not be the cause of the disease, but may rather be a result of it (Yoder, 1980). Available
evidence indicates that DON may not be essential for pathogenesis but it may play a role in
aggressiveness. There is no evidence that toxin action occurs during the penetration process.
Whether germinating conidia release DON to the plant surface prior to penetration has not been
reported. Moreover, macroconidia of F. graminearum and conidiophores of F. culmorum were
found to contain no detectable DON (Evans et al. 2000; Mirocha et al. 1997; Snijders, 1990).
Of the Fusarium species pathogenic to wheat, only F. graminearum and F. culmorum have been
found to be the most aggressive and to produce high levels of DON (Manka et al. 1985; Stack and
McMullen, 1985). However, field isolates of F. graminearum and F. culmorum that were unable
to produce DON and 3-acDON in culture were pathogenic on wheat, rye, and triticale seedlings
(Manka et al. 1985). The main drawback of using quantitative correlation between quantity of
toxin produced in vitro and aggressiveness is that toxin production in vitro is influenced by the
composition of the medium and by the physical environment (Miller and Greenhalgh, 1985;
Yoder, 1980). In addition, trichothecenes tend to accumulate to higher concentrations in the
kernels of more susceptible wheat cultivars than the more resistant cultivars (Snijders and
Perkowski, 1990; Atanassov et al. 1994). Also, there is some evidence for a positive correlation
between head blight resistance and the ability of wheat to metabolize DON (Miller and Arnison,
1986). Evans et al. (2000) found a concurrent increase in disease symptom development with
increase in DON and 15-acDON levels in inoculated barley spikelets. Visible disease symptoms
appeared 48 h post inoculation, when the toxins were also first detected. DON increased
dramatically after 72 h while 15-acDON peaked at 72-120 h, during which time chlorosis and
necrosis of the spikelets appeared. This indicates that DON and 15-acDON may be involved in
colonization of host tissue during early stages of infection.
Studies involving trichothecene biosynthetic pathway gene disruption have produced more
convincing evidence that certain trichothecenes play an important role in disease
development. Fusarium sporotrichioides strains in which T-2 toxin production was blocked by
ultraviolet-induced mutation or specifically altered by gene disruption were of low virulence on
parsnip roots but T-2 producing wild-type and mutant strains were of high virulence.(Beremend,
1987; Desjardins et al. 1989; Platner et al. 1989). The trichodiene mutant strains incapable of
producing T-2 toxin complimented each other to restore T-2 toxin production and to partially
restore virulence. This suggested that production of T-2 toxin is required for high virulence of F.
sporotrichioides on parsnip roots. Similarly, Fusarium graminearum trichodiene mutants
incapable of producing DON in culture and wheat spikelets showed only mild symptoms of
infection but wild-type DON producing strains caused more severe symptoms of infection
(Mirocha et al. 1997). The severity of the lesions was correlated with DON synthesis. Disruption
of Tri5, the gene encoding trichodiene synthase, which catalyzes the first step in trichothecene
biosynthetic pathway, resulted in reduced virulence of G. zeae on wheat seedlings and ears
(Desjardins et al. 1996; Proctor et al. 1995b). Although the Tri5- strains colonized wheat heads ,
the infected heads showed less disease by several parameters, including head bleaching
symptoms, seed weight, seed viability, and trichothecene contamination.
Nicholson et al. (1997) used quantitative PCR to quantify colonization of wheat plant tissues
by head blight pathogens as measured by DNA content. Kernel samples inoculated with the
wild-type isolates and trichothecene-producing revertants contained 15-50 times more G. zeae
DNA as samples inoculated with the Tri5- trichothecene-producing mutants. Therefore,
colonization of grain is reduced where trichothecene production is inhibited. The above
evidences indicate that trichothecenes are virulence factors in wheat head scab. If DON is a
virulence factor inhibition of its biosynthesis could decrease virulence and therefore protect plants
from infection. DON biosynthesis may be inhibited by application of synthetic or naturally
105
105
occurring trichothecene inhibitors directly to plants or by incorporation of genes for such
inhibition into the plants. Moreover, the purified toxin may be useful in selection for resistance
using whole plants, cells, pollen, and protoplasts in the development of Fusarium head blight
resistant cultivars (Yoder, 1980).
In the current study, the isolates of F. graminearum also differed in DON biosynthetic rate
(DON/Erg ratio), amount of disease per unit of DON (disease severity/DON ratio) and amount of
disease per unit of ergosterol (disease severity/Erg ratio). These ratios were useful in separating
the effects of DON from those of fungal biomass. DON/ergosterol ratio was not always correlated
to aggressiveness parameters. Significant correlations (r = 0.63 and 0.80) were found between
DON/Erg ratio and disease severity and relative kernel weight reduction, respectively, for the
batch 1 isolates. However, no correlations (r = 0.39 and 0.43, respectively) were found for the
batch 2 isolates. This difference in the two isolate batches was probably due to environmental
effect. In cases where infection occurs early due to favourable weather conditions, the
pathogen will infect the vascular system and the sections above the infection site are bleached
due to water and nutrient depletion without being colonised (Miedaner et al. 1997). Therefore,
mycotoxin contamination is unlikely to occur in such parts of the head but disease rating and
kernel weight determination take such into consideration.
Isolates with the same DON/ERG ratio were differing in aggressiveness and vice versa. Gang
et al. (1998) and Miedaner et al. (2000) also found no association between aggressiveness and
DON/ERG ratio on rye, indicating that the differences in aggressiveness of the isolates cannot
be explained by different DON production rates alone. Other mechanisms such as
extracellular enzymes necessary to penetrate and invade the host cells may also play a role in
the actual aggressiveness of the isolates (Miedaner, 1997; Miedaner et al. 2000). Proctor et al.
(1995b) and Desjardins et al. (1996) also found that trichothecene-deficient F. graminearum
mutants retained some level of aggressiveness, although no DON could be found in the host.
Cell-wall degrading enzymes have been associated with Fusarium infection in wheat
(Miedaner et al. 1997; Kang and Buchenauer, 2000). In greenhouse experiments, shoot
samples contained amounts of cellulase and xylanase which also correlated with
aggressiveness, suggesting that cell-wall degrading enzymes have an impact on aggressiveness
in young plant stage (Miedaner et al. 1997).
Ultrastructural and cytochemical studies of wheat spikes infected with Fusarium culmorum
showed that inter- and intracellular growth of the pathogen in the ovary, lemma and rachis
caused pronounced alterations of cell walls, middle lamella matrices and cell wall components
(Kang and Buchenauer, 2000). The degradation of cellulose, xylan and pectin in the host cell
wall of the infected tissues indicated that F. culmorum may secrete corresponding cell-wall
degrading enzymes such as cellulases, xylanases and pectinases during infection of the wheat
spike. The alterations of the cell walls was noticed to occur in advance of hyphae. In addition,
Fusarium culmorum is reported to produce carboxymethylcellulase, xylanase and pectin-
degrading enzymes in vitro (Boothby and Magreola, 1984). Fusarium infection of wheat
kernels has also been observed to result in destruction of starch granules, storage proteins and
cell wall, indicating enzymatic action (Nightingale et al. 1999). Therefore, it appears that DON
act together with cell wall degrading enzymes during colonization of the host tissues, hence
determining aggressiveness of an isolate.
5.7 Cultivar susceptibility to Fusarium head blight and mycotoxin content
The 15 varieties significantly differed in susceptibility to Fusarium head blight, as indicated
by mean disease rating, area under disease progress curve (AUDPC) and ergosterol content of
106
106
kernels. Fusarium head blight was observed to develop slowly in the less susceptible cultivars
indicating that the more resistant the genotype, the longer is the incubation period (Snijders
and Perkowski (1990). Disease rating was correlated to AUDPC but not to kernel weight
reduction. Both AUDPC and disease rating were not correlated to re-isolation rate of F.
graminearum. The significant correlation between disease rating and AUDPC indicates that
either of the traits can be used for evaluation of resistance. This agrees with the findings of
Miedaner (1997) and Miedaner et al. (1996). Because Fusarium head blight is a monocyclic
disease, the AUDPC is also highly correlated to a single rating date (Snijders, 1990). The lack
of correlation between the disease rating and relative kernel weight reduction indicates that
visual assessment of Fusarium head blight is a more reliable and sensitive criteria for
evaluation of resistance than yield reduction (Mesterhazy, 1995; Miedaner et al. 1993;
Snijders, 1990). Relative kernel weight reduction also depends on the host genotype.
The kernels from all the varieties showed high fungal invasion, as indicated by ergosterol
content and DON contamination. The significant correlation between disease rating (AUDPC)
versus ergosterol content and AUDPC versus DON content indicates that the more susceptible
varieties were more colonized (more fungal mycelium) and they contained more DON,
respectively. This is also reflected in the significant correlation (r = 0.61) between AUDPC
and DON/Erg ratio which means that the more susceptible varieties contained higher amounts
of DON per unit of mycelium. The significant correlation (r = 0.92) between DON and
ergosterol suggests that the differences among varieties in DON content are most likely due to
differences in amounts of fungal biomass present in the kernels. Therefore, toxin production is
related to the extent of the fungal infection, which is governed by host-pathogen relationships
and environment (Miller et al. 1983b). The lower ergosterol content in the less susceptible
varieties indicates that they have higher resistance to hyphal invasion (Miller et al. 1985;
Wang and Miller, 1988) or type II resistance (Schroeder and Christensen, 1963).
This study showed a negative correlation (r = -0.60) between disease rating (AUDPC) and
ergosterol to DON ratio indicating that ergosterol to DON ratios in the more susceptible
cultivars are lower than in the less susceptible cultivars. This observation supports by similar
findings by Love and Seitz (1987), Miller et al. (1985) and Perkowski et al. (1995). Miller et
al. (1985) showed that the ergosterol to DON ratio approached 2.4 and ca. 12 in susceptible
and resistant cultivars, respectively. This means that appreciable concentrations of ergosterol
(mostly F. graminearum biomass) were found in the less susceptible cultivars without the
corresponding amount of DON being present. Miller et al. (1985) therefore suggested that the
ratio could be an index of susceptibility to DON synthesis, a high ratio indicating resistance of
cultivars and vice versa.
The resistant cultivars could be less suitable for DON production or possess mechanisms for
DON degradation. This theory was supported by the observations of Miller and Arnison
(1986) and Wang and Miller (1988) that resistant wheat cultivars tolerat higher DON
concentrations and also degrade DON in vitro compared to susceptible cultivars. This
suggests that the lower disease ratings in the less susceptible cultivars could be partly due to
their ability to degrade DON, that is they possess type III resistance (Miller et al. 1985).
Tolerance to DON might be based on a modification of the peptidyl transferase enzyme,
rendering it insensitive to DON inhibition (Miller, 1989). A cultivar may be susceptible but
show a high ergosterol to DON ratio, indicating that the cultivar posses the DON degradation
system but not resistance to initial penetration (type I) and resistance to colonization (type II).
However, this ratio is only appropriate if F. graminearum (or F. culmorum) is the only
predominant fungus in the grain. In naturally infected wheat, other fungi will also be found in
107
107
the grain and the ergosterol to DON ratio will depend on the growth of F. graminearum/F.
culmorum compared to the other fungi.
5.8 RAPD-PCR analysis of Fusarium species
Polymerase chain reaction using 10-base pair random primers was carried out to determine the
differences among Fusarium species.
Most of the primers tested showed varying degrees of polymorphism and some resulted in
species-specific bands. RAPD profiles exhibiting distinct fragments were found to be
informative because the fragments are polymorphic and could be clearly and unambiguously
scored (Schilling et al. 1994). Cluster analysis of the RAPD profiles revealed interspecific and
intraspecfific variation and more detailed information could be obtained with different
primers. Isolates belonging to a certain species generally tended to cluster together, indicating
a relationship between RAPD-PCR profiles and morphological identification of the Fusarium
isolates. Chelkowski et al. (1999) were also able to identify toxigenic Fusarium species F.
graminearum, F. culmorum, F. crookwellense and F. avenaceum using RAPD-PCR and the
isolates of the four Fusarium spp. were easily distinguished by species-specific bands. Similar
observations have been reported for F. graminearum and F. culmorum (Schilling et al. 1994;
Schilling et al. 1996b), various trichothecene-producing Fusarium species (Mule et al. 1997)
and different Fusarium species (Yoder and Christianson, 1998).
The RAPD banding pattern within Fusarium species has been found to be uniform
irrespective of the geographical origin of the isolates or the host/substrate from which they
were isolated (Yoder and Christianson, 1998) indicating the reliability of the method in
classification as opposed to morphological identification. The cultural and morphological
appearance of Fusarium strains can be highly variable depending on the culture conditions
employed (Yoder and Christianson, 1998). The cluster analysis results showed that although
F. culmorum and F. graminearum produced different RAPD profiles, they are very closely
related, since they clustered together with most primers. Clustering of isolate FC3.1 of F.
culmorum with isolates of F. graminearum with a number of primers may be due to
misidentification. Close relationship between F. graminearum and F. culmorum has also been
demonstrated by use of PCR-based finger prints (Schilling et al. 1996a; Yoder and
Christianson, 1997; Mule et al. 1997) and analysis of ribosomal proteins (Patridge, 1991).
This close relationship between the two species conforms with their respective mycotoxin
profiles, which were found to be similar in this study and elsewhere (Miller et al. 1991;
Langseth et al. 1999). Fusarium culmorum and F. graminearum belong to section Discolor of
the Fusarium species, according to the classification system of Nelson et al. (1983).
The other isolates that clustered together belong to F. avenaceum (section Roseum) and F.
poae (section Sporotrichiella). The use of metabolic profiles in Fusarium has been suggested
based on the fact that each species appear to produce a more or less characteristic pattern of
toxins (Langseth et al. 1999). Using ribosomal DNA sequences, Mule et al. (1997) found that
all trichothecene-producing strains clustered together. Strains belonging to F. acuminatum, F.
sambucinum, F. tumidum, F. Comptoceras, F. sprotrichioides, and F. avenatum which
produced type A trichothecenes clustered together. The second cluster consisted of strains of
F. crookwellense, F. culmorum, and F. graminearum, which produced type B trichothecenes.
The authors found that the phylogenetic placement of the species based on rDNA correlated
better with toxic secondary metabolite data rather than with the classification system based on
morphology. Although they belong to different sections, F. acuminatum, F. camptoceras, F.
108
108
compactum, F. sporotrichioides, F. sambucinum, F. tumidum, and F. avenatum were found to
be more closely related to each other than they are to other members of the same section that
cannot synthesize type A trichothecenes. A PCR assay capable of detecting potential
trichothecene-producing Fusarium species has been described (Niessen and Vogel, 1997). The
results of the assay indicated that Fusarium tricinctum is incapable of producing
trichothecenes unlike closely related species like F. sporotrichioides and F. poae. Therefore,
molecular and secondary metabolite profile studies, in conjunction with cultural and
morphological characteristics, may be useful in Fusarium taxonomy.
5.9 Polymerase chain reaction (PCR) analysis of isolates of F. graminearum and F.
culmorum
Isolates of Fusarium culmorum and F. graminearum were subjected to polymerase chain
reaction to confirm morphological identification and to determine differences among the
isolates. Identification was by use 20-base pair species specific primers while isolate
differentiation was by use of 10-base pair random primers.
With the aid of reference isolates specific PCR assays confirmed the morphological
identification of isolates of both F. culmorum and F. graminearum. The isolates of each
particular species showed a single characteristic band. The specific primers can be used to
detect the Fusarium head blight pathogens directly in the infected host plant tissues without
isolation in pure culture (Henson, 1993). Polymerase chain reaction-based assays have been
developed for species-specific detection of F. graminearum, F. culmorum and F. avenaceum
(Schilling et al. 1996b), F. poae (Parry and Nicholson, 1996), Microdochium nivale (Doohan
et al. 1998) and F. avenaceum (Turner et al. 1998). The assay can detect these species directly
in extracts of infected tissues at early stages of disease with barely visible symptoms.
It is also possible to assay for several pathogens simultaneously using multiple sets of
pathogen-specific primer pairs, or ”multiplex PCR”(Hensen, 1993). Quantitative PCR, based
on species-specific primers has also been used to quantify the level of colonization of plant
tissues by head blight pathogens as measured by DNA content (Doohan et al. 1998, 1999;
Nicholson et al., 1997). These studies indicated that the amount of DNA in infected tissues is
correlated with visual disease assessment. Therefore, the high sensitivity of the species-
specific PCR could facilitate the implementation of disease management programs early in the
growing season. In addition, it can be used to test presence of Fusarium spp. as an indicator of
mycotoxin hazard (Parry and Nicholson, 1996). A major drawback of using species-specific
PCR probing of plant materials is interference from Phenolics and enzyme inhibitors in the
DNA extracts (Schilling et al. 1994; Turner et al. 1998).
The different random primers produced different degrees of variation and polymorphism
among the isolates. This is agrees with the findings of Nicholson et al. (1996) and Schilling et
al. (1996a, and 1997). However, RAPD-PCR with most primers showed some species-
specific bands which confirmed identity of the isolates. Therefore, RAPD-PCR has been
found to be useful because it does not require genome sequence information or radiolabeling,
and it distinguishes between organisms and different strains of the same organism (Hensen,
1993). RAPD patterns have also been used in identifying wrongly classified isolates (Schilling
et al. 1994). Combined RAPD-PCR profiles of different primers gave more detailed
information for differentiation of the isolates, as opposed to results of a single primer. The
results indicate that the differences among the isolates in aggressiveness and mycotoxin
production is based on differences at the DNA level, hence genetically controlled.
109
109
The differences, however, were not associated with the level of aggressiveness, amount of
mycotoxin production, source of isolates and in F. culmorum, the type of mycotoxin
produced. Oullet and Seifert, (1993) could also distinguish strains of F. graminearum by
combined profiles of 5 RAPD and PCR primers or primer pairs. The strains produced patterns
specific to F. graminearum. The strains could, however, not be grouped according to host or
geographical origin and no correlations were detected between the banding patterns and the
mycotoxin profiles or titer. Similar observations have been reported by Chelkowski et al.
(1999) with F. culmorum and F. graminearum, Yli-Mattila et al. (1996) with F. avenaceum,
and Miedaner et al. (1997) with Fusarium culmorum. Miedaner et al. (1997) and Miedaner et
al. (2000) demonstrated the genetic basis of the observed phenotypic variation in
aggressiveness and DON production of 42 isolates of F. culmorum by high heritability
estimates of the two traits in different environments and by DNA markers. However, isolate
groupings by DNA markers were not associated with aggressiveness nor with the type of
mycotoxin produced. Also, isolates from similar geographic or host origin did not group
together. Therefore, the lack of correlation between the RAPD patterns and aggressiveness
and mycotoxin production may be because these traits are controlled by several genes and/or
an allelic series at one locus with the alleles having a different contribution (Miedaner et al.,
1996). It has been established that ten closely linked genes, localized to a 25-kb region of
chromosomal DNA, are involved in trichothecene biosynthesis in F. sporotrichioides
(Desjardins et al. 1996; Hohn et al. 1995).
Cluster analysis of the RAPD banding patterns of F. graminearum showed clustering of the
isolates according to either DON- or NIV-producing isolate clusters. The DON- and NIV-
producing isolates of F. culmorum could, however, not be differentiated. This suggests that
the F. culmorum isolates could produce both mycotoxins, with the NIV-chemotypes
producing large amounts of NIV and very low quantities of DON below the detection limit
and vice versa. This would explain the detection of both mycotoxins in culture and wheat
kernels for a few isolates of F. culmorum. This ability to produce both mycotoxins would,
therefore, mean that the isolates possess genes for both NIV and DON. On the contrary, the
DON-chemotypes of F. graminearum tested may not possess genes for NIV production and
vice versa. None of the isolates of F. graminearum was found to produce both DON and NIV
in culture and in kernels.
Lack of cross-production of NIV and DON by F. graminearum has been reported elsewhere
(Ichinoe et al. 1983; Lauren et al. 1992; Miller et al. 1991; Mirocha et al. 1989) but isolates
capable of producing both mycotoxins have also been reported (Blany and Dodman, 1988;
Sugiura et al. 1990; Szecsi and Bartok, 1995). Sydenham et al. (1991) and Blany and Dodman
(1988) observed that F. graminearum group I isolates were mainly DON-chemotype whereas
group II isolates were either NIV- or DON-chemotypes. These findings suggest that different
strains may possess different biosynthetic pathways, hence resulting in different mycotoxins.
Environment seems to have an effect on the expression of the pathway genes as has been
suggested by the prevalence of different chemotypes in different geographical regions. Sugiura
et al. (1990) suggested that the 3-and 15-acetylation of DON are dependent on environment.
110
110
Chapter 6: Conclusions
Fusarium graminearum, F. avenaceum and F. poae which are capable of causing head blight
of wheat were isolated from wheat samples collected in producing areas of Kenya. In addition,
all the isolates of F. graminearum tested were aggressive on wheat ears and produced
mycotoxins in the kernels and in vitro. The locally-grown wheat varieties were also found to
be moderate to highly susceptible to Fusarium head blight, after inoculation with F.
graminearum isolated locally. Therefore, a certain level of Fusarium head blight is prevalent
and mycotoxin-producing Fusarium strains are widely distributed in the wheat growing areas
of Kenya. A more detailed study is needed to determine the level of the disease in the farmers’
fields and in different cropping seasons. Testing of the varieties under field conditions and at
different localities is needed to help in choice of cultivars and for future breeding programs.
Since maize is also grown alongside wheat in many areas, a study on the Fusarium species
infecting the crop would be worthwhile because this data is lacking for Kenya and other
African countries. This information will also be essential for creating farmers’ awareness on
the disease and mycotoxin risk.
About 60% and 40% of the tested Fusarium culmorum isolates from the Rheinlands,
Germany, were found to produce nivalenol (NIV) and deoxynivalenol (DON), respectively, in
the field and in vitro. Therefore, the two types of strains seem to be widely distributed in this
region and contamination of wheat with both NIV and DON during scab epidemic years is
highly probable. Although the NIV-producing isolates were found to cause lower yield
reduction, they are of more toxicological significance because NIV is much more toxic than
DON. Grain containing substantial amounts of NIV is more likely to be approved for animal
feed as it may show low percentage of Fusarium damaged kernels. A detailed study on the
occurrence of the two mycotoxins in naturally contaminated grain in the Rheinlands is needed.
This would be useful for setting mycotoxin guidelines and highlighting the actual mycotoxin
risk in this region. Current guidelines in most countries only take DON into consideration.
Moreover, further in vivo toxigenesis studies involving NIV and DON producing isolates are
also needed. These may involve determining the time course production of the two toxins and
their biosynthesis when wheat ears are simultaneously inoculated with both NIV- and DON-
producing isolates. This would indicate the natural competitive abilities of the two
chemotypes.
The fact that the F. culmorum isolates produced either NIV or DON suggests that strains
belonging to this species can be classified into NIV- or DON-chemotypes on the basis of their
mycotoxin profiles. In the few cases where both toxins were produced, one was detected a
minor metabolite. Therefore, the NIV-chemotypes probably produce NIV and trace amounts
of DON and vice versa. This may explain the lack of clear differences between the NIV- and
DON-chemotypes on the RAPD-PCR banding patterns, suggesting presence of NIV and DON
genes in both chemotypes. The differentiation between NIV- and DON-chemotypes of F.
graminearum by RAPD-PCR analysis indicate that the NIV-chemotypes probably possess
only NIV genes and DON-chemotypes possess only DON genes. Further research on
Fusarium chemotaxonomy based on mycotoxin profiles would be essential. The variability
among the isolates of both F. culmorum and F. graminearum in RAPD-PCR banding patterns
indicates that the differences in aggressiveness and mycotoxin production is partly due to
genetic effects. Multiple genes affect trichothecene yields (Desjardins and Beremend, 1987;
Desjardins et al. 1993; Hohn et al. 1993) and aggressiveness (Miedaner et al. 1996), therefore
accounting for the lack of correlation between RAPD-PCR banding patterns and mycotoxin
production and aggressiveness. The study also demonstrated that RAPD-PCR can be useful in
111
111
the classification of mycotoxin-producing Fusarium species but further molecular studies
would be needed to confirm the relationship among the different species.
NIV and DON-producing isolates were pathogenic to wheat ears but showed varying levels of
aggressiveness. Aggressiveness was generally higher for isolates that also produced high
amounts of DON. In addition, pure DON was phytotoxic to wheat seedlings. Nivalenol and
zearalenone are not phytotoxic. Therefore, production of DON is not essential for infection
but its phytotoxic nature helps the pathogen to colonize the host tissues. However, DON is not
the only factor utilized by Fusarium to colonize wheat tissues because the NIV-producing
isolates also showed low to moderate level of aggressiveness. Cell wall degrading enzymes
have been suggested to be involved wheat tissue colonization by Fusarium (Kang and
Buchenauer, 2000; Miedaner, 1997). Therefore, the variation in aggressiveness within the
NIV-producing isolates of F. culmorum could be due to differences in enzyme-producing
capabilities while the additive effects of enzyme and DON production enhances
aggressiveness of the DON-producing isolates. Research on the role of enzymes in wheat head
blight infection process after inoculation with different chemotypes of F. culmorum and F.
graminearum is needed.
Finally, research on trichothecene biosynthesis inhibitors as a control measure of mycotoxin
contamination in wheat may be helpful. This would involve detailed studies on the
trichothecene of the biosynthetic pathway in order to identify suitable points in the
biosynthetic pathway to disrupt.
112
112
References
Abildgren, M. P., F. Lund, U. Thrane, S. Elmholt. 1987. Czapek-Dox agar containing
iprodione as a selective medium for the isolation of Fusarium species. Let. Appl.
Microbiol. 5: 83-86.
Abouzied, M. M. and J. J. Pestka. 1986. Fusarenone X and nivalenol production by
Gibbereella zeae in liquid and rice cultures. Mycopathologia 96: 19-24.
Abramson, D. 1998. Mycotoxin formation and environmental factors. In: Sinha, K. K. and D.
Bhatnagar (Eds.). Mycotoxins in agriculture and food safety. Marcel Dekker, Inc. pp255-
277.
Abramson, D., Z. Gan, R. M. Clear, J. Gilbert and R. R. Marquardt. 1998. Relationships
among deoxynivalenol, ergosterol and Fusarium exoantigens in Canadian hard and soft
wheat. Int. J. Food Microbiol. 45: 217-224.
Adams, G. C. and L. P. Hart. 1989. The role of deoxynivalenol and 15-acetyldeoxynivalenol
in pathogenesis by Gibberella zeae, as elucidated through protoplast fusion between
toxigenic and nontoxigenic strains. Phytopathology 79: 404-408.
Andrivon, D. 1993. Nomenclature for pathogenicity and virulence: The need for precision.
Phytopathology 83: 889-890.
Atanassov, Z., C. Nakamura, N. Mori, C. Kaneda, H. Kato, Y. Jin, T. Yoshizawa, and K.
Murai. 1994. Mycotoxin production and pathogenicity of Fusarium species and wheat
resistance to Fusarium head blight. Can. J. Bot. 72: 161-167.
Bai, G. and G. Shaner.1994. Scab of wheat: Prospects for control. Plant Disease, 78: 760-766.
Barath, A., J. Sawinsky, A. Halasz, and G. Borbiro. 1997. Substrate influence on mycotoxin
production of Fusarium species and their analytical detection. Cereal Research
Communications 25: 353-354.
Barnett, H. L. and B. B. Hunter. 1972. Illustrated genera of imperfect fungi, 3rd
Ed. Burgess
Publishing Company, Minnesota. 241 pp.
Bennett, J. W. and N.P. Keller. 1997. Mycotoxins and their prevention. In: Anke, T. (Ed).
Fungal biotechnology. Chapmann & Hall, Weinheim.
Beremand, M. N. 1987. Isolation and characterization of mutants blocked in T-2 toxin
biosynthesis. Appl. Environ. Microbiol. 53: 1855 - 1859.
Bermingham, S., L. Maltby and R. C. Cooke. 1995. A critical assessment of the validity of
ergosterol as an indicator of fungal biomass. Mycol. Res. 96: 965-970.
Beyer, W., H. Höxter, T. Miedaner, E. Sander and H. H. Geiger. 1993. Indirect ELISA for
quantitative assessment of Fusarium spp. in rye. J. Plant Dis. Prot. 100: 278-284.
113
113
Bilgrami, K. S. and A. K. Choudhary. 1998. Mycotoxins in pre-harvest contamination of
agricultural crops. In: Sinha, K. K. and D. Bhatnagar (Eds.). Mycotoxins in agriculture
and food safety. Marcel Dekker, Inc. pp1-43.
Blany, B. J. and R. L. Dodman. 1988. Production of the mycotoxins zearalenone, 4-
deoxynivalenol and nivalenol by isolates of Fusarium graminearum groups 1 and 2 from
cereals in Queensland. Aust. J. Agric. Res. 39: 21-29.
Boothby, D. and N. O. Mongreola. 1984. Production of polysaccharide degrading enzymes by
Cochliobolus sativus and Fusarium culmorum grown in liquid culture. Trans. Br. Mycol.
Soc. 83: 275-280.
Bottalico, A. and A. Logrieco. 1998. Toxigenic Alternaria species of economic importance.
In: Sinha, K. K. and D. Bhatnagar (Eds.). Mycotoxins in agriculture and food safety.
Marcel Dekker, Inc. pp65-108.
Bottalico, A., A. Logrieco and A. Visconti. 1989. Fusarium species and their mycotoxins in
infected corn in Italy. Mycopathologia 107: 85-92.
Bowden, J. L., and J. F. Leslie. 1992. Nitrate nonutilizing mutants of Gibberella zeae (F.
graminearum) and their use in determining vegetative compatibility. Exp. Mycol. 16:
308-315.
Bowden, J. L. and J. F. Leslie. 1994. Diversity of Gibberella zeae at small spatial scales.
Phytopathology 84: 1140.
Bowden, J. L., and J. F. Leslie. 1999. Sexual recombination in Gibberella zeae.
Phytopathology 89: 182-188.
Bruins, M. B. M., I. Karsai, J. Schepers, and C. H. A. Snijders. 1993. Phytotoxicity of
deoxynivalenol to wheat tissue with regard to in vitro selection for Fusarium head blight
resistance. Plant Sci. 94: 195-206.
Burgess, L. W., A. H. Wearing, and T. A. Toussoun. 1975. Survey of Fusaria associated with
crown rot of wheat in Eastern Australia. Austr. J. Agric. Res. 26: 791-799.
Burgess, L. W., Summerell, D. Backhouse, F. Benyon, and J. Levic. 1996. Biodiversity and
population studies in Fusarium. Sydowia 48: 1-11.
Cassale, W. L. and L. P. Hart. 1988. Inhibition of 3H-leucine incorporation by trichothecene
mycotoxins in maize and wheat tissue. Phytopathology 78: 1673-1677.
Chelkowski, J. 1998. Distribution of Fusarium species and their mycotoxins in cereal grains.
In: Sinha, K. K. and D. Bhatnagar (Eds.). Mycotoxins in agriculture and food safety.
Marcel Dekker, Inc. pp45-64.
Chelkowski, J. and M. Manka. 1983. The ability of Fusarium pathogenic to wheat and corn to
produce zearalenone. Pytopath. Z. 106: 354 - 359.
114
114
Chelkowski, J., G. L. Bateman and C. J. Mirocha. 1999. Identification of toxigenic Fusarium
species using PCR assays. J. Phytopathology 147: 307-311.
Coker, R. D., B. D. Jones, M. J. Nagler, G. A. Gilman, A. J. Wallbridge and S. Panigrahi.
1984. Mycotoxin training manual. Tropical Development and Research Institute,
Overseas Development Administration. 127 Clerkenwell Road, London.
Cullen, D., R. W. Caldwell, and E. B. Smalley. 1982. Cultural characteristics, pathogenicity,
and zearalenone production by strains of Gibberella zeae isolated from corn.
Phytopathology 72: 1415-1418.
Cutler, H. G. and B. B. Jarvis. 1985. Preliminary observations on the effects of macrocyclic
trichothecenes on plant growth. Environmental and experimental Botany, 25: 115-128.
Desjardins, A. E., R. D. Plattner and F. Vanmiddelsworth. 1986. Trichothecene biosynthesis
in Fusarium sprotrichioides: Origin of the oxygen atoms of T-2 toxin. Appl. Environ.
Microbiol. 51: 493-497.
Desjardins, A. E., and M. Beremend. 1987. A genetic system for trichothecene toxin
production in Gibberella pulicans (Fusarium sambucinum). Phytopathology 77: 678-683.
Desjardins, A. E., R. D. Plattner and M. N. Beremend. 1987. Ancymidol blocks trichothecene
biosynthesis and leads to accumulation of trichodiene in Fusarium sporotrichioides and
Gibberella pulicans. Appl. Environ. Microbiol. 53: 1860-1865.
Desjardins, A. E., R. D. Plattner and G. F. Spencer. 1988. Inhibition of trichothecene toxin
biosynthesis by naturally occurring skimatic aromatics. Phytochemistry 27: 767-771.
Desjardins, A. E., G. F. Spencer, R. D. Plattner and M. N. Beremend. 1989. Furanocoumarin
phytoalexins, trichothecene toxins, and infection of Pastinaca sativa by Fusarium
sporotrichioides. Phytopathology 79: 170-175.
Desjardins, A. E., T. M. Hohn and S. P. McCormick. 1993. Trichothecene biosynthesis in
Fusarium species: Chemistry, genetics and significance. Microbiol. reviews 57: 595-604.
Desjardins, A. E., R. H. Proctor, G. Bai, S.P. McCormic, G. Shaner, G. Buechley., and H.
Hohn. 1996. Reduced virulence of trichothecene-noneproducing mutants of Gibberella
zeae in wheat field tests. Mol. Plant-Microbe-Interact. 9: 775-781.
Dexter, J. E., R. M. Clear and K. R. Preston. 1996. Fusarium head blight: Effect on the
milling and baking of some canadian wheats. Cereal Chem. 73: 695-701.
Dexter, J. E., B. A. Marchylo, R. M. Clear, and J. M. Clarke. 1997. Effect of Fusarium head
blight on semolina milling and pasta-making quality of durum wheat. Cereal Chem. 74:
519-525.
Dill-Macky, R. and R. K. Jones 2000. The effect of previous crop residues and tillage on
Fusarium head blight of wheat. Plant Dis. 84: 71-76.
115
115
Doohan, F. M., D. W. Parry, P. Jenkinson and P. Nicholson. 1998. The use of species-specific
PCR-based assays to analyse Fusarium ear blight of wheat. Plant Pathology 47: 197-205.
Doohan, F. M., D. W. Parry and P. Nicholson. 1999. Fusarium ear blight: the use of
quantitative PCR and visual disease assessment in studies of disease control. Plant
Pathology 48: 209-217.
El-Banna, A. A., R. M. G. Hamilton, P. M. Scott, and H. L. Trenholm. 1983. Non-
transmission of deoxynivalenol (vomitoxin) to eggs and meat in chickens fed
deoxynivalenol-contaminated diets. J. Agric. Food Chem., 31: 1381-1384.
Evans, C. K., W. Xie, R. Dill-Macky and C. J. Mirocha. 2000. Biosynthesis of deoxynivalenol
in spikelets of barley inoculated with macroconidia of Fusarium graminearum. Plant
Disease 84: 654-660.
Fekete, C., A. Logrieco, G. Giczey and L. Hornok. 1997. Screening of fungi for the presence
of the trichodiene synthase encoding sequence by hybridization to the Tri5 gene cloned
from Fusarium poae. Mycopathologia 138: 91-97.
Fisher, N. L., L. W. Burgess, T. A. Toussoun and P. E. Nelson. 1982. Carnation leaves as a
substrate and for preserving cultures of Fusarium species. Phytopathology 72: 151-153.
Food and Agriculture Organization (FAO). 1997. Worldwide regulations for mycotoxins
1995: A compendium. FAO Food and Nutrition Paper 64
Francis, R. G., and L. W. Burgess. 1977. Characteristics of two populations of Fusarium
roseum ‘Graminearum’ in Eastern Australia. Trans. Br. Mycol. Soc. 68: 421-427.
Gan, Z., R. R. Marquardt, D. Abramson and R. M. Clear. 1997. The characterization of
chicken antibodies raised against Fusarium spp. by enzyme-linked immunosorbent assay
and immunoblotting. Int. J. Food Microbiol. 38: 191-200.
Gang, G., T. Miedaner, U. Schuhmachar, M. Schollenberger, and H. H. Geiger. 1998.
Deoxynivalenol and nivalenol production by Fusarium culmorum isolates differing in
aggressiveness towards winter rye. Phytopathology 88: 879-884
Gerlach, W. and H. Nirenberg. 1982. The genus Fusarium - a pictorial atlas. Mitt. Biol.
Bundesanst. Land- und Forstw. Berlin-Dahlem. 209.
Gerlach, H. K., and R. Stösser. 1997. Patterns of random amplified polymerase DNAs for
sweet cherry (Prunus avium L.) cultivar identification. Angew. Bot. 71: 212-218.
Gibbon, D. and A. Pain. 1985. Crops of the drier regions of the tropics. Longman, London.
157pp.
Gleissenthal, L. J., R. Dietrich, E. Märtlbauer, M. Schuster, A. Suß, and G. Terplan. 1989. A
survey on the occcurrrence of Fusarium mycotoxins in Bavarian cereals from the 1987
harvest. Z. Lebensm. Unters. Forsch., 188: 521-526.
116
116
Gonzalez, H. H. L., E. J. Martinez, A. Pacin and S. L. Resnik. 1999. Relationship between
Fusarium graminearum and Alternaria alternata contamination and deoxynivalenol
occurrence on Argentinian durum. Mycopathologia 144: 97-102.
Gordon, T. R. and R. K. Webster. 1984. Evaluation of ergosterol as an indicator of infestation
of barley seed by Drechslera graminea. Phytopath.74: 1125-1127.
Gourama, H. and L. B. Bullerman. 1995. Detection of moulds in foods and feeds: potential
rapid and selective methods. J. Food Protection 58: 1389-1394.
Greenhalgh, R., G. A. Neish and J. D. Miller. 1983. Deoxynivalenol, acetyl deoxynivalenol and
zearalenone formation by Canadian isolates of Fusarium graminearum on solid substrates.
Appl. and Environ. Microbiol. 46: 625-629.
Hamilton, R. M. G., H. L. Trenholm, B. K. Thompson, and R. Greenhalgh. 1985. The
tolerance of white leghorn and broiler chicks, and turkey poults to diets that contained
deoxynivalenol (vomitoxin)-contaminated wheat. Poultry Science, 64: 273-286.
Henson, J. M. 1993. The polymerase chain reaction and plant disease diagnosis. Ann. Rev.
Phytopathol. 31: 81-109.
Hesketh, A. R., L. Glenhill, D. C. Marsh, B. W. Bycroft, P. M. Dewick and J. Gilbert. 1991.
Biosynthesis of trichothecene mycotoxins: Identification of isotrichodiol as post-
trichodiene intermediate. Phytochemistry 30: 2237-2243.
Hohn, T. M. and M. N. Beremend. 1989a. Regulation of trichodiene synthase in Fusarium
sporotrichioides and Gibberella pulicans (Fusarium sambucinum). Appl. Environ.
Microbiol. 55: 1500-1503.
Hohn, T. M. and M. N. Beremend. 1989b. Isolation and nucleotide sequence of a
sesquiterpene cyclase gene from the trichothecene-producing fungus Fusarium
sporotrichioides. Gene 79: 131-138.
Hohn, T. M. and A. E. Desjardins. 1992. Isolation and gene disruption of the Tox5 gene
encoding trichodiene synthase in Gibberella pulicans Mol. Plant-Microbe Interact. 5:
249-256.
Hohn, T. M., A. E. Desjardins, and S. P. McCormic. 1993. Analysis of Tox5 gene expression
in Gibberella pulicans strains with different trichothecene phenotypes. Appl. Environ.
Microbiol. 59: 2359-2363.
Hohn, T. M., A. E. Desjardins, and S. P. McCormic. 1995. The Tri4 gene of Fusarium
sporotrichioides encodes a cytochrome P450 monooxygenase involved in trichothecene
biosynthesis. Mol. Gen. Genet., 248: 95-102.
Ichinoe, M., H. Kurata, Y. Sugiura and Y. Ueno. 1983. Chemotaxonomy of Gibberrella zea
with special reference to production of trichothecenes and zearalenone. Appl. Environ.
Microbiol., 46: 1364-1369.
117
117
Jones, R. K. 2000. Assessment of Fusarium head blight of wheat and barley response to
fungicide treatment. Plant Dis. 84: 1021-1030.
Kamimura, H., M. Nishijima, K. Yasuda, K. Saito, A. Ibe, T. Nagayama, H. Ushiyama, and Y.
Naoi. 1981. Simultaneous detection of several Fusarium mycotoxins in cereals, grains
and foodstuffs. J. Assoc. Off. Anal. Chem. 64: 1067-1073.
Kang, Z. and H. Buchenauer. 2000. Ultrastructural and cytochemical studies on cellulose,
xylan and pectin degradation in wheat spikes infected by Fusarium culmorum. J.
Phytopathology 148: 263-275.
Kim, J.-C., H.-J. Kang, D.-H. Lee, Y.-W. Lee and Y. Yoshizawa. 1993. Natural occurrence of
Fusarium mycotoxins (trichothecenes and zearalenone) in barley and corn in Korea.
Appl. Environ. Microbiol. 59: 3798-3802.
Krska, R., R. Schuhmacher, J. Weingärtner and M. Grasserbauer. 1997. Modern sample
preparation techniques for the analysis of Fusarium mycotoxins in cereals. Cereal
Research Communications 25: 327-329.
Langseth, W., A. Bernhoft, T. Rundberger, B. Kosiak and M. Gareis. 1999. Mycotoxin
production and cytotoxicity of Fusarium strains isolated from Norwegian cereals.
Mycopathologia 144: 103-113.
Lauren, D. R., R. A. Ashley, B. A. Blackwell, R. Greenhalgh, J. D. Miller and G. A. Neish.
1987. Trichothecenes produced by Fusarium crookwellense DAOM 193611. J. Agric.
Food Chem. 35: 884-889.
Lauren, D. R. and M. P. Agnew. 1991. Multitoxin screening method for Fusarium mycotoxins
in grains. J. Agric. Food Chem. 39: 502-507.
Lauren, D. R., S. T. Sayer, and M. E. di Menna. 1992. Trichothecene production by Fusarium
species isolated from grain and pasture throughout New Zealand. Mycopathologia 120:
167-176.
Lee, U-S., H-S. Jang, T. Tanaka, Y-J. Oh, C-M. Cho, and Y. Ueno. 1987. Effect of milling on
decontamination of Fusarium mycotoxins nivalenol, deoxynivalenol, and zearalenone in
Korean wheat. J. Agric. Food Chem., 35: 126-129.
Lee, U.-S., M-Y. Lee, W-Y. Park, and Y. Ueno. 1992. Decontamination of Fusarium
mycotoxins, nivalenol, deoxynivalenol, and zearalenone, in barley by the polishing
process. Mycotoxin Research, 8: 31-36.
Leonov, A. N., G. P. Kononenko, and N. A. Soboleva. 1990. Production of DON-related
trichothecenes by Fusarium graminearum Schw. from Krasnodarski krai of the USSR.
Mycotoxin Res. 6: 54-60.
Love, G. R. and L. M. Seitz. 1987. Effect of location and cultivar on Fusarium head blight
(scab) in wheat from Kansas in 1982 and 1983. Cereal Chem. 64: 124-128.
118
118
Manka, M., A. Visconti, J. Chelkowski and A. Bottalico. 1985. Pathogenicity of Fusarium
isolates from wheat, rye and triticale towards seedlings and their ability to produce
trichothecenes and zearalenone. Phytopath. Z., 113: 24-29.
Manka, M., J. Chelkowski, D. Brayford, A. Visconti, H. Kwasna and J. Perkowski. 1989.
Fusarium graminearum- cultural characteristics, pathogenicity towards cereal seedlings
and ability to produce mycotoxins. J. Phytopath. 124: 143-148.
Marasas, W. F. O., P. E. Nelson and T. A. Toussoun. 1984. Toxigenic Fusarium species:
Identity and mycotoxicology. The Pennsylvania State University Press, University Park,
Pa.
Martin, R. A., and H. W. Johnston. 1982. Effects and control of Fusarium diseases of cereal
grains in the Atlantic Province. Can. J. Plant Pathol. 4: 210-216.
Matthies, A. and H. Buchenauer. 1995. Investigations on the action of different active
ingredients on the biosynthesis of mycotoxins in Fusarium culmorum and Fusarium
graminearum. In: Rüssel, P. E., and H. D. Sisler (Eds). Modern fungicides and antifungal
compounds- proceedings of 11th
international symposium. 578pp.
Matthies, A. and H. Buchenauer. 2000. Effect of tebuconazole (Folicur
) and prochloraz
(Sportak
) treatments on Fusarium head scab development, yield and deoxynivalenol
(DON) content in grains of wheat following artificial inoculation with Fusarium
culmorum. Journal of plant Diseases and Protection. 107: 33-52.
McCormic, S. P., S. L. Taylor, R. D. Plattner and M. N. Beremend. 1990. Bioconversion of
possible T-2 toxin precursors by mutant strains of Fusarium sporotrichioides NRRL
3299. Appl. Environ. Microbiol. 56: 702-706.
McMullen, M. P., R. Jones, and D. Gallenberg. 1997. Scab of wheat and barley: A re-
emerging disease of devastating impact. Plant Dis. 81: 1340-1348.
McPherson, M. J., P. Quirke and G. R. Taylor. 1991. PCR: A practical approach. Oxford
University Press.
Mesterha'zy, A'. 1978. Comparative analysis of artificial inoculation methods with Fusarium
spp. on winter wheat varieties. Phytopathol. Z. 93: 12 - 25.
Mesterhazy, A. 1987. Selection of head blight resistant wheats through improved seedling
resistance. Plant Breeding 98: 25-36.
Mesterhazy, A. 1995. Types and components of resistance to Fusarium head blight of wheat.
Plant Breeding 114: 377-386.
Miedaner, T. 1997. Breeding wheat and rye for resistance to Fusarium diseases. Plant
Breeding 116: 201-220.
Miedaner, T., and J. Perkowski. 1996. Correlations among Fusarium culmorum head blight
resistance, fungal colonisation, and mycotoxin contents in winter rye. Plant Breeding 115:
347-351.
119
119
Miedaner, T., and A. G. Schilling. 1996. Genetic variation of aggressiveness in individual
field populations of Fusarium graminearum and F. culmorum tested on winter rye.
Europ. J. Plant Pathol. 102: 823-830.
Miedaner, T., D. C. Borchardt and H. H. Geiger. 1993. Genetic analysis of inbred lines and
their crosses for resistance to head blight (Fusarium culmorum, F. graminearum) in
winter rye. Euphytica 65: 125-133.
Miedaner, T., G. Gang, and H. H. Geiger. 1996. Quantitative-genetic basis of aggressiveness
of 42 isolates of Fusarium culmorum for winter rye head blight. Plant Dis.80: 500-504.
Miedaner, T., G. Gang, A. G. Schilling and H. H. Geiger. 1997. Aggressiveness and
mycotoxin production of populations of Fusarium culmorum and F. graminearum in
winter rye. Creal Reasearch Communications25: 471-475.
Miedaner, T., C. Reinbrecht and A. G. Schilling. 2000. Association among aggressiveness,
fungal colonization and mycotoxin production of 26 isolates of Fusarium graminearum
in winter rye head blight. J. Plant Dis. Prot. 107: 124-134.
Miller, J. D. 1989. Effects of Fusarium graminearum metabolites on wheat cells. In:
Phytototoxins and plant pathogegenesis. Ed.: A. Graniti., R. D. Durbin and A. Ballio.
NATO ASI series H27. 508 pp. pp.449-452.
Miller, J. D. and R. Greenhalgh. 1985. Nutrient effects on the biosynthesis of trichothecenes
and other metabolites by Fusarium graminearum. Mycologia 77: 130-136.
Miller, J. D. and H. L. Trenholm. 1994. Mycotoxins in grains: Compounds other than
aflatoxins. Eagan Press, St. Paul, Minnesota, USA.
Miller, J. D., A. Taylor and R. Greenhalgh. 1983a. Production of deoxynivalenol and related
compounds in liquid culture by Fusarium graminearum. Can. J. Microbiol. 29: 1171-
1178.
Miller, J. D. and P. G. Arnison. 1986. Degradation of deoxynivalenol by suspension cultures
of Fusarium head blight resistant wheat cultivar Fontana. Can. J. Plant Pathol.8: 147-150.
Miller, J. D., J. C. Young and H. L. Trenholm. 1983b. Time course of fungal growth and
production of deoxynivalenol and other mycotoxins in field corn. Can. J. Bot. 61: 3080-
3087.
Miller, J. D., J. C. Young and D. R. Sampson. 1985. Deoxynivalenol and Fusarium head
blight resistance in spring cereals. Phytopath. Z.,113: 359-367.
Miller, J. D., R. Greenhalgh, Y. Wang and M. Lu. 1991. Trichothecene chemotypes of three
Fusarium species. Mycologia 83: 121- 130.
Mirocha, C. J., H. K. Abbas, C. E. Windels and W. Xie. 1989. Variation in deoxynivalenol,
15-acetyldeoxynivalenol, 3-acetyldeoxynivalenol, and zearalenone production by
Fusarium graminearum isolates. Appl. and Environ. Microbiol. 55: 1315-1316.
120
120
Mirocha, C. J., W. Xie, Y. Xu, R. D. Wilcoxson, R. P. Woodward, R. H. Etebarian and G.
Behele. 1994. Production of trichothecene mycotoxins by Fusarium graminearum and
Fusarium culmorum on barley and wheat. Mycopathologia 128: 19-23.
Mirocha, J., Y. Hui, C. K. Evans, E. K. Kolaczkowski and R. Dill-Macky. 1997. Chemistry
and physiology of deoxynivalenol in pathogenesis. Cereal research Communications
25:309-313.
Mitchell, R. E. 1984. The relevance of non-host –specific toxins in the expression of virulence
by pathogens. Ann. Rev. Phytopathol. 22:215-245.
Moss, M. O. 1996. Mycotoxins- Centenary review. Mycol. Res. 100: 513-523.
Mossoba, M. M., S. Adams, J. A. G. Roach and M. W. Trucksess. 1996. Analysis of
trichothecene mycotoxins in contaminated grains by gas chromatography/matrix
isolation/fourier transform infrared spectroscopy and gas chromatography/mass
spectrometry. Food Chem. Cont. 79: 1116-1123.
Mulè, G., A. Logrieco, G. Stea, and A. Bottalico. 1997. Clustering of trichothecene-producing
Fusarium strains determined from 28S ribosomal DNA sequences. Appl. Environ.
Microbiol. 63: 1843-1846.
Müller, H.-M. and K. Schwadorf. 1993. A survey of the natural occurrence of Fusarium
toxins in wheat grown in a southwestern area of Germany. Mycopathologia 121: 155-121.
Müller, H.-M., J. Reimann, U. Schumacher, and K. Schwadorf. 1997. Fusarium toxins in
wheat harvested during six years in an area of Southwest Germany. Natural toxins 5: 24-
30.
Murkovic, M., M. Gailhofer, W. Steiner and W. Pfannhauser. 1997. Formation of zearalenone
on wheat contaminated with Fusarium graminearum under controlled conditions. Z.
Lebensm. Unters. Forsch.204: 39-42.
Muthomi, J. W., A. Schütze, H.-W. Dehne, E. W. Mutitu and E.-C. Oerke. 2000.
Characterization of Fusarium culmorum isolates by mycotoxin production and
aggressiveness to winter wheat. J. Plant Dis. Prot. 107: 113-123.
Nelson, P. E., T. A. Toussoun, and W. F. O. Marassas. 1983. Fusarium species: An illustrated
manual for identification. Pennsylvania State University Press, University park. 193pp.
Nelson, P. E., T. A. Toussoun, and R. J. Cook (Eds.). 1981. Fusarium: Diseases, biology, and
taxonomy. The pennsylvania State University Press, University Park and London. 457pp.
Newell, S. Y., T. L. Arsuffi, and R. D. Fallon. 1988. Fundamental procedures for determining
ergosterol content of decaying plant material by liquid chromatography. Appl. Environ.
Microbiol. 54: 1876-1879.
121
121
Nicholson, P., P. Jenkinson, H. N. Rezanoor, and D. W. Parry. 1993. Restriction fragment
length polymorphism analysis of variation in Fusarium species causing ear blight of
cereals. Plant Pathol. 42: 905-914.
Nicholson, P., F. Doohan, H. N. Rezanoor, D. Simpson, P. H. Smith, A. Turner and G.
Weston. 1997. Detection and quantification of individual fungal species in Fusarium
disease complexes of cereals by polymerase chain reaction (PCR). Cereal Research
communications 25: 477-482.
Niessen, M. L. and R. Vogel. 1997. A molecular approach to the detection of potential
trichothecene producing fungi. Cereal Research Communications 25: 245-249.
Nightingale, M. J., B. A. Marchylo, R. M. Clear, J. E. Dexter, and K. R. Preston. 1999.
Fusarium head blight: Effect of fungal proteases on wheat storage proteins. Cereal Chem.
76: 150-158.
Nirenberg, H. 1981. A simplified method for identifying Fusarium species occurring in wheat.
Can. J. Bot. 59: 1599 - 1609.
Obst, A., J. Lepschy and R. Beck. 1997. On the etiology of Fusarium head blight of wheat in
South Germany- preceding crops, weather conditions for inoculum and mycotoxin
production. Cereal Research Communications 25: 699-703.
Ouellet, T., and K. A. Seifert. 1993. Genetic characterization of Fusarium graminearum
strains using RAPD and PCR amplification. Phytopathology 83: 1003-1007.
Padgett, D. E. and M. H. Posey. 1993. An evaluation of the efficiencies of several ergosterol
extraction techniques. Mycol. Res. 97: 1476-1480.
Parry, D. W. and P. Nicholson. 1996. Development of a PCR assay to detect Fusarium poae
in wheat. Plant Pathology 45: 383-391.
Parry, D. W., P. Jenkinson, and L. McLeod. 1995. Fusarium ear blight (scab) in small grain
cereals- a review. Plant Pathol. 44: 207-238.
Patridge, J. E. 1991. Specific gene products of Fusarium: Analysis of ribosomal proteins.
Phytopathology 81: 1055-1057.
Paulitz, T. C. 1996. Diurnal release of ascospores by Gibberella zeae in inoculated wheat
plots. Plant Dis. 80: 674-678.
Perkowski, J., J. Chelkowski and W. Wakulinski. 1987. Mycotoxins in cereal grain. Part 13.
Deoxynivalenol and 3-acetyldeoxynivalenol in wheat kernels and chaff with head
fusariosis symptoms. Nahrung 31: 326-328.
Perkowski, J., R.D. Plattner, P. Golinski, R. F. Vesonder, and J. Chelkowski. 1990. Natural
occurrence of deoxynivalenol, 3-acetyl-deoxynivalenol, 15-acetyl-deoxynivalenol, nivalenol,
4,7-dideoxynivalenol, and zearalenone in polish wheat. Mycotoxin Research, 6: 7-12.
122
122
Perkowski, J., T. Miedaner, H. H. Geiger, H-M. Müller, and J. Chelkowski. 1995. Occurrence
of deoxynivalenol (DON), 3-acetyldeoxynivalenol, and ergosterol in winter rye inoculated
with Fusarium culmorum. Cereal Chem. 72: 205-209.
Pestka, J. J., A. El-Bahrawy and L. P. Hart. 1985. Deoxynivalenol and 15-
monoacetylnivalenol production by Fusarium graminearum R6576 in liquid media.
Mycopathologia, 91: 23-28.
Pineiro, M. S., P. M. Scott and S. R. Khanhere.1996. Mycotoxin producing potential of
Fusarium graminearum isolates from Uruguayan barley. Mycopathologia, 132: 167-172.
Plattner, R. D., L. W. Tjarks and M. N. Beremend. 1989. Trichothecene accumulated in liquid
cultures of a mutant of Fusarium sporotrichioides NRRL 3299. Appl. Environ.
Microbiol. 55: 2190-2194.
Pons, W. A. and A. O. Franz. 1977. High performance liquid chromatography of aflatoxins in
cotton seed products. J. Assoc. Anal. Chem. 60: 89-95.
Prelusky, D. B., D. M. Veira, H. L. Trenholm and B. C. Foster. 1987. Metabolic fate and
elimination in milk, urine and bile of deoxynivalenol following administration to lactating
sheep. Journal of Environmental Science and Health, B22: 125-148.
Proctor, R. H., J. M. Hohn, S. P. McCormic and A. E. Desjardins. 1995a. Tri6 encodes an
unusual zinc finger protein involved in regulation of trichothecene biosynthesis in
Fusarium sporotrichioides. Appl. Environ. Microbiol. 61: 1923-1930.
Proctor, R. H., T. M. Hohn and S. P. McCormic. 1995b. Reduced virulence of Giberrella zeae
caused by disruption of a trichothecene toxin biosynthesis gene. Mol. Plant-Microbe
Interact. 8: 593-601.
Puhalla, J. 1981. Genetic considerations of the genus Fusarium. In: P. E. Nelson, T. A.
Toussoun, and R. J. Cook (eds), Fusarium- Diseases, Biology, and Taxonomy, 291-305.
The Pennsylvania State University Press, University Park, PA.
Ribichich, K. F., S. E. Lopez, and A. C. Vegetti. 2000. Histopathological spikelet changes
produced by Fusarium graminearum in susceptible and resistant wheat cultivars. Plant
Dis. 84: 794-802.
Romer, T. R. 1986. Use of small charcoal/alumina cleanup columns in determination of
trichothecene mycotoxins in foods and feeds. J. Ass. Off. Anal. Chem. 69: 699-703.
Sambrook, J., E. F. Fritsch and T. Maniatis. 1989. Molecular cloning-A laboratory manual.
2nd
ed. Cold Spring Harbour Laboratory press, USA
Schilling, A. G. 1996. Characterization and differentiation of the cereal pathogens Fusarium
culmorum and F. graminearum by PCR-based molecular markers. PhD. Thesis, Univ.
Hohenheim, Germany.
Schilling, A. G., E. M. Möller, and H. H. Geiger. 1994. RAPDs of Fusarium culmorum and F.
graminearum: Application for genotyping and species identification. In: Schots, A., F. M.
123
123
Dewey and R. P. Oliver (Eds.) Modern assays for plant pathogenic fungi - identification,
detection and quantification. CAB international. pp 47-56.
Schilling, A. G., E. M. Möller and H. H. Geiger. 1996a. Molecular differentiation and
diagnosis of the cereal pathogenic Fusarium culmorum and F. graminearum. Sydowia 48:
71 - 82.
Schilling, A. G., E. M. Möller, and H. H. Geiger. 1996b. Polymerase chain reaction- based
assays for species-specific detection of Fusarium culmorum, F. graminearum and F.
avenaceum. Phytopathology 86: 515-522.
Schilling, A. G., T. Miedaner, and H. H. Geiger. 1997. Molecular variation and genetic
structure in field populations of Fusarium species causing head blight in wheat. Cereal
Research Communications 25: 549-554.
Schipilova, N. P. and T. Y. Gagkaeva. 1997. The forms of manifestation of Fusarium head
blight on the seeds and heads of cereal crops. Cereal research communications, 25: 815-
816.
Scholbrock, L. L., B. K. Fleener, and J. A. Berry. 1992. Comparison of wheat seedling leaf
reactions to deoxynivalenol in relation to wheat scab (Fusarium graminearum) resistance
classes. Phytopathology 82: 1167.
Schroeder, H. W. and J. J. Christensen. 1963. Factors affecting resistance of wheat to scab
caused by Gibberella zeae. Phytopathology 53: 831-838.
Schütze, A., E.-C. Oerke, and H.-W. Dehne. 1997. Isolation and differentiation of Fusarium
spp. and Microdochium nivale on winter wheat in western Germany. Cereal Research
Communications 25: 615-616.
Schwadorf, K. and H-M. Müller. 1989. Determination of ergosterol in cereals, mixed feed
components, and mixed feeds by liquid chromatography. J. Assoc. Off. Anal. Chem. 72:
457-462.
Scot, P. M. 1990. Natural poisons. In: Helrich, K. (Ed.). Official methods of analysis. 15th
Ed.
Association of official analytical chemist, Arlington, VA. Pp1184 - 1213.
Scott, P. M., K. Nelson, S. R. Kanhere, K. F. Karpinski, S. Hayward, G. A. Neish and A. H.
Teich. 1984. Decline in deoxynivalenol (vomitoxin) concentrations in 1983 Ontario
winter wheat before harvest. Appl. Environ. Microbiol. 48: 884-886.
Seitz, L. M., H. E. Mohr, R. Burroughs and D. B. Sauer. 1977. Ergosterol as an indicator of
fungal invasion in grains. Cereal Chemistry 54: 1207-1217.
Seitz, L.M., D. B. Sauer, R. Burroughs, H. E. Mohr and J. D. Hubbard. 1979. Ergosterol as a
measure of fungal growth. Phytopathology 69: 1202-1203.
Shaner, G. and R. E. Finney. 1977. The effect of nitrogen fertilization on the expression of
slow-mildewing resistance in knox wheat. Phytopathology 67: 1051-1056
124
124
Snijders, C. H. A. 1990. Aspects of resistance to Fusarium head blight caused by Fusarium
culmorum in wheat. Ph.D. thesis, Agricultural university, Wageningen.
Snijders, C. H. A. and J. Perkowski. 1990. Effects of head blight caused by Fusarium
culmorum on toxin content and weight of wheat kernels. Phytopathology 80: 566-570.
Sohn, H.B., J.-A. Seo and Y.-W. Lee. 1999. Co-occurrence of Fusarium mycotoxins in
mouldy and healthy corn from Korea. Food Add. Cont. 16: 153-158.
Stack, R. W. and M. P. McMullen. 1985. Head blighting potential of Fusarium species
associated with spring wheat heads. Can. J. Plant Pathol.7: 79-82.
Steel, R. G. D. and J. H. Torrie. 1981. Principles and procedures of statistics . A biometric
approach. 2nd
Ed. McGraw-Hill Book Company. 633p.
Strange, R. N. and H. Smith. 1978. Specificity of choline and betaine as stimulants of
Fusarium graminearum. Trans. Br. Mycol. Soc. 70: 187-192.
Stratton, G. W., A. R. Robinson, H. C. Smith, L. Kittilsen and M. Barbour. 1993. Levels of
five mycotoxins in grains harvested in Atlantic Canada as measured by high performance
liquid chromatography. Arch. Environ. Toxicol. 24: 399-409.
Stubs, R. W., J. M. Prescott, E. E. Saari, and H. J. Dubin. 1986. Cereal disease methodology
manual. CIMMYT, Mexico.
Sugiura, Y., Y. Watanabe, T. Tanaka, S. Yamamoto and Y. Ueno. 1990. Occurrence of
Gibberella zeae strains that produce both nivalenol and deoxynivalenol. Appl. Environ.
Microbiol. 56: 3044-3051.
Sugiura, Y. K. Fukasaku, T. Tanaka, Y. Matsui, and Y. Ueno. 1993. Fusarium Poae and
Fusarium crookwellense, fungi responsible for the natural occurrence of nivalenol in
Hokkaido. Appl. Environ. Microbiol. 59: 3334-3338.
Sutton, J. C. 1982. Epidemiology of wheat blight and maize ear rot caused by Fusarium
graminearum. Can. J. Pl. Pathol. 4: 195- 209.
Sydenham, E. W., P. G. Thiel, W. F. O. Marasas, and J. J. Nieuwenhuis. 1989. Occurrence of
deoxynivalenol and nivalenol in Fusarium graminearum infected under grade wheat in
South Africa. J. Agric. Food Chem. 37: 921-926.
Sydenham, E. W., W. F. O. Marasas, P. G. Thiel, G. S. Shephard and J. J. Nieuwenhuis. 1991.
Production of mycotoxins by selected Fusarium graminearum and F. crookwellense
isolates. Food Additives and Contaminants 18: 31-41.
Szecsi, A and T. Bartok. 1995. Trichothecene chemotypes of Fusarium graminearum isolated
from corn in Hungary. Myctoxin Res. 11: 85-92.
Szuets, P., A. Mesterhazy, G. Y. Falky and T. Bartok. 1997. Early telarche symptoms in
children and their relations to zearalenone contamination in foodstuffs. Cereal Research
Communications 25: 429-436.
125
125
Tanaka, T., S. Yamamoto, A. Hasegawa, N. Aoki, J. R. Besling, Y. Sugiura and Y. Ueno.
1990. A survey of the natural occurrence of Fusarium mycotoxins, deoxynivalenol,
nivalenol and zearalenone, in cereals harvested in the Netherlands. Mycopathologia 110:
19-22.
Tkachuk, R., J. E. Dexter, K. H. Tipples and T. W. Nowicki. 1991. Removal by specific
gravity table of tombstone kernels and associated trichothecenes from wheat infected with
Fusarium head blight. Cereal Chem. 68: 428-431.
Tothill, I. E., D. Harris and N. Magan. 1992. The relationship between fungal growth and
ergosterol content of wheat grain. Mycol. Res. 96: 965-970.
Trigo-Stockli, D. M., C. W. Deyoe, R. F. Satumbaga, and J. R. Pedersen. 1996. Distribution
of deoxynivalenol and zearalenone in milled fractions of wheat. Cereal Chem. 73: 388-
391.
Trucksess, M. W., S. Nesheim and R. M. Eppley. 1984. Thin-layer chromatographic
determination of deoxynivalenol in wheat and corn. J. Assoc. Anal. Chem.67: 40-49.
Trucksess, M. W., D-W. E. Ready, M. K. Pender, C. A. Ligmond, G. E. Wood and S. W.
Page. 1996. Determination and survey of deoxynivalenol in wheat flour, whole wheat
flour and bran. J. Ass. Off. Anal. Chem. 79: 883 - 887.
Tschanz, A. T., R. K. Horst and P. E. Nelson. 1975. A substrate for uniform production of
perithecia in Gibberella zeae. Mycologia 67: 1101-1108.
Turner, A. S., A. K. Lees, H. N. Rezanoor and P. Nicholson. 1998. Refinement of PCR-
detection of Fusarium avenaceun and evidence from DNA marker studies for phenetic
relatedness to Fusarium tricinctum. Plant Pathology 47: 278-288.
United States Department of Agriculture (USDA). 1998. Agricultural statistics. United States
Government Printing Office, Washington.
Usleber, E., V. Renz, E. Märtlbauer and G. Terplan. 1992. Studies on the application of
enzyme immunoassays for the Fusarium mycotoxins deoxynivalenol, 3-
acetyldeoxynivalenol, and zearalenone. J. Vet. Med. 39: 617-627.
Vanderplank, J. E. 1984. Disease resistance in plants, 2nd
Edition. Academic Press, New
York, 194pp.
Visconti, A., J. Chelkowski, and A. Bottalico. 1986. Deoxynivalenol and 3-
acetyldeoxynivalenol- mycotoxins associated with head fusariosis in Poland. Mycotoxin
Research 2: 59-64.
Vrabcheva, T., R. Gessler, E. Usleber and E. Märtlbauer. 1996. First survey on the natural
occurrence of Fusarium mycotoxins in Bulgarian wheat. Mycopathologia 136: 47-52.
Walker, F. and B. Meier. 1998. Determination of the Fusarium mycotoxins nivalenol,
deoxynivalenol, 3-acetyldeoxynivalenol, and 15-O-acetyl-4-deoxynivalenol in
126
126
contaminated whole wheat flour by liquid chromatography with diode array detection and
gas chromatography with electron capture detection. J. Ass. Off. Anal. Chem. Intern. 81:
741-748.
Wang, Y. Z. and J. D. Miller. 1988. Effects of Fusarium graminearum metabolites on wheat
tissue in relation to Fusarium head blight resistance. J. Phytopathol. 122: 118-125.
Weingaertner, J., R. Krska, W. Praznik, M. Grasserbauer and H. Lew. 1997. Use of mycosep
multifunctional clean-up columns for the determination trichothecenes in wheat by
electron-capture gas chromatography. Fresenius J. Anal. Chem. 357: 1206 - 1210.
Weising, K., H. Nybom, K. Wolff and W. Meyer. 1995. DNA fingerprinting in plants and
fungi. CRC Press.
Wilson, D. M., E. W. Sydenham, G. A. Lombaert, M. W. Trucksess, D. Abramson and G. A.
Bennett. 1998. Myctoxin analytical techniques. In: Sinha, K. K. and D. Bhatnagar (Eds.).
Mycotoxins in agriculture and food safety. Marcel Dekker, Inc. pp 135-182.
Windels, C. E. 2000. Economic and social impacts of Fusarium head blight: Changing farms
and rural communities in the Northern Great Plains. Phytopathology 90: 17-21.
Windels, C. E., C. J. Mirocha, H. K. Abbas, and W. Xie. 1989. Perithecium production in
Fusarium graminearum populations and lack of correlation with zearalenone production.
Mycologia 81: 272-277.
Wolf-Hall, C. E. and L. B. Bullerman. 1996. Comparison of thin-layer chromatography and an
enzyme-linked immunosorbent asssay for detection and quantification of deoxynivalenol
in corn and wheat. J. Food Protect. 59: 438-440
Wood, G. E. and M. W. Trucksess. 1998. Regulatory control programs for mycotoxin-
contaminated food. In: Sinha, K. K. and D. Bhatnagar (Eds.). Mycotoxins in agriculture
and food safety. Marcel Dekker, Inc. pp459-480.
Yli-Mattila, T., S. Paavanen, A. Hannukkala, P. Parikka, R. Tahvonen and R. Karjalainen.
1996. Isozyme and RAPD-PCR analyses of F. avenaceum strains from Finland. Plant
Pathology 45: 126-134.
Yoder, O. C. 1980. Toxins in pathogenesis. Ann. Rev. Phytopathol. 18: 103-129.
Yoder, W. T. and L. M. Christianson. 1998. Species-specific primers resolve members of
Fusarium section Fusarium; Taxonomic status of the edible ”Quorn” fungus re-
evaluated. Fungal Gen. Biol. 23: 68-80
Yoshizawa, T., and N. Morooka. 1973. Deoxynivalenol and its monoacetate: new mycotoxins
from Fusarium roseum and mouldy barley. Agric. Biol. Chem. 37: 2933-2934.
Young, J. C., R. G. Fulcher, J. H. Hayhoe, P. M. Scott, and J. E. Dexter. 1984. Effect of
milling and baking on deoxynivalenol (vomitoxin) content of Eastern Canadian wheats. J.
Agric. Food Chem.32: 659-664.
127
127
Zadoks, J. C., J. J. Chang and C. F. Kunzak. 1974. Decimal code for growth stages of cereals.
Weed Research 15: 415-421.
Zill, G., G. Engelhardt and P. R. Wallnöfer. 1988. Determination of ergosterol as a measure of
fungal growth using Si 60 HPLC. Z. Lebensm. Unters. Forsch. 187: 246-249
Zillinsky, F. J. 1983. Common diseases of small grain cereals: A guide to identification.
CIMMYT. 141pp.