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


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













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








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






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


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


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


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)


a

a

a

a

a

a

a

a

0

50

100

150

200

250

Control 0 1 5 10 20 50 100


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


Primer P3

CCTGGCGGTA

100 bp

800 bp


Primer P4

CCGGCCTTAA

800 bp

200 bp


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


Primer P7

GTCCGGAGTG

800 bp

200 bp


Primer P8

GAGCACCAGG

200 bp

800 bp


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


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


Primer P7

GTCCGGAGTG

800 bp

200 bp


Primer P8

GAGCACCAGG

800 bp

200 bp


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.


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.


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.


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.


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


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


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.

profiles.uonbi.ac.ke · 2 2 DECLARATION BY THE CANDIDATE This thesis is my original work and has...

Documents

Transcript of profiles.uonbi.ac.ke · 2 2 DECLARATION BY THE CANDIDATE This thesis is my original work and has...