Determination of deoxynivalenol and nivalenol chemotypes of Fusarium culmorum isolates from England and Wales by PCR assay
Version of Record online: 28 APR 2004
Volume 53, Issue 2, pages 182–190, April 2004
How to Cite
Jennings, P., Coates, M. E., Turner, J. A., Chandler, E. A. and Nicholson, P. (2004), Determination of deoxynivalenol and nivalenol chemotypes of Fusarium culmorum isolates from England and Wales by PCR assay. Plant Pathology, 53: 182–190. doi: 10.1111/j.0032-0862.2004.00985.x
- Issue online: 28 APR 2004
- Version of Record online: 28 APR 2004
- Accepted 26 November 2003
- Fusarium culmorum;
Functioning Tri13 and Tri7 genes are required for the production of nivalenol and 4-acetyl nivalenol, respectively, in Fusarium species producing type B trichothecenes. Mutations have been identified in isolates which are able to produce deoxynivalenol (DON) but unable to convert this to nivalenol (NIV). In such isolates of Fusarium culmorum, the Tri7 gene is deleted entirely. PCR assays specific for functional and nonfunctional/deleted versions of Tri7 and Tri13 were used to determine the ability of 153 single spore isolates of F. culmorum to produce the 8-ketotrichothecenes deoxynivalenol and nivalenol. The isolates were collected from 76 different locations across England and Wales between 1994 and 2002. Four isolates were also obtained from one field in Scotland. Both DON and NIV chemotypes of F. culmorum were identified, with DON chemotypes predominating overall. In addition, all DON chemotypes were shown to produce 3-acetyl DON using primer sets developed to Tri3. From fields where more than one F. culmorum isolate was obtained, isolates were not exclusively of a single chemotype. Differences in the distribution of DON and NIV chemotypes were identified, with a greater proportion of NIV chemotypes present in the south and west of England and Wales, whereas a greater proportion of DON chemotypes were found in the north and east of England. Seasonal differences in the ratio of DON:NIV chemotypes were indicated. However, these were related to seasonal variation in the distribution of F. culmorum.
In the UK four fusarium head blight (FHB) pathogens are commonly isolated from wheat crops: F. culmorum, F. poae, F. avenaceum and Microdochium nivale. Less abundant species include F. graminearum, F. cerealis, F. tricinctum and F. sporotrichioides. Worldwide, F. graminearum is potentially the most damaging of the FHB pathogens, in terms of both yield loss and mycotoxin production. However, under current conditions, F. culmorum is responsible for the greatest losses in the UK (Jennings & Turner, 1996; Turner & Jennings, 1997).
The occurrence of mycotoxins in grain is of particular importance, as they have potentially harmful effects on humans and livestock. The principal toxins produced by F. culmorum and F. graminearum are the type B trichothecenes deoxynivalenol (DON) and nivalenol (NIV). NIV is generally regarded as more toxic to humans and animals than is DON (Ryu et al., 1988), although DON may be more phytotoxic than NIV (Eudes et al., 2000). Trichothecene biosynthesis is a complex process, involving a series of oxygenation, isomerization and esterification steps. In F. graminearum, F. culmorum and F. sporotrichioides, many of the trichothecene biosynthesis genes are located in a gene cluster (Hohn et al., 1993; Brown et al., 2002; Lee et al., 2002; Kimura et al., 2003) comprising at least 10 genes.
Based on the production of different trichothecenes, Ichinoe et al. (1983) identified two chemotypes among Japanese isolates of F. graminearum: those producing deoxynivalenol and those producing nivalenol. These chemotypes were further divided depending on the position at which the toxins were acetylated (Miller et al., 1991). DON and NIV chemotypes of F. culmorum have been detected in several countries, including Germany (Muthomi et al., 2000), the Netherlands and Italy (Gang et al., 1998), Norway (Langseth et al., 1999), France (Bakan et al., 2001) and the United States (Mirocha et al., 1994), whereas only the DON chemotype was detected in western Canada (Abramson et al., 2001).
Traditionally, chemotyping of Fusarium isolates has been carried out using gas chromatography/mass spectroscopy. This method can be time-consuming and expensive. Recently Lee et al. (2001) reported a novel method for chemotyping F. graminearum isolates using PCR analysis. The analysis exploited differences in Tri7 which exist between DON and NIV chemotypes of F. graminearum. In F. culmorum the Tri7 gene is missing from the trichothecene biosynthetic and regulatory gene cluster (Chandler et al., 2003), rendering the Lee assay unsuitable for chemotyping of F. culmorum. More recently, a series of PCR assays have been designed to Tri7 and Tri13, in order to permit specific detection of functional and nonfunctional/deleted versions of these genes in F. culmorum and lineages of F. graminearum (Chandler et al., 2003). Furthermore, phylogenetic analysis of the trichothecene cluster has revealed that isolates of F. graminearum that produce NIV, 3-acetyl DON or 15-acetyl DON differ in the sequence of several genes within the cluster (Ward et al., 2002).
Several surveys of UK cereal crops have shown that both DON and NIV were present in grain (Tanaka et al., 1986; Polley et al., 1991; Turner et al., 1999; Prickett et al., 2000). Currently F. culmorum is the dominant type B trichothecene producer in the UK, suggesting that potentially both DON- and NIV-producing chemotypes of F. culmorum exist. The objective of this study was to chemotype UK isolates of F. culmorum, through PCR analysis of Tri7 and Tri13, in order to establish the presence and proportion of DON and NIV chemotypes in the UK. As isolates were collected across the UK between 1994 and 2002, this also allowed regional and seasonal differences in chemotype distribution to be determined (Fig. 1).
Materials and methods
Isolation and identification
Wheat ears showing classic symptoms of fusarium head blight were collected between 1994 and 2002. In 1994 and 1997, samples were collected from random samples which arrived at the laboratory. However, from 1998 onwards, infected ears were collected from the annual Defra survey of winter wheat diseases. The site of primary infection, normally indicated by an orange spore mass at the base of a spikelet, was cut from the affected ear, surface-sterilized for 10 min in 10% sodium hypochlorite, rinsed twice in sterile distilled water and blot dried between sterile filter paper. The surface-sterilized spikelets were placed onto potato dextrose agar (PDA; Oxoid Ltd, Basingstoke, UK) amended with streptomycin sulphate (120 mg L−1) and incubated for 5 days at 25°C. Fusarium isolates were purified by single spore isolation.
Single spore isolates were cultured onto both PDA and sucrose nutrient agar (SNA) (Nirenberg, 1976) plates and incubated at 25°C in a 12-h light/dark cycle for 10 days. Isolates were identified to species level based on colony characteristics on PDA and spore morphology on SNA. A total of 151 isolates of F. culmorum from 76 different locations across England and Wales (Table 1) were used throughout this study. In 1994, four isolates were also obtained from a field in Scotland.
|Culture number||Other ref.a||F. culmorum specific PCRb||F. culmorum chemotype using|
|Tri7 primers||Tri13 primers||Tri3 primers|
|DONc||NIVd||DONe||NIVf||3-acetyl DONg||15-acetyl DONh||NIVI|
Isolates of F. culmorum were grown on PDA and DNA was extracted using a method adapted from Chang et al. (1993).
F. culmorum species-specific PCR and determination of chemotype
Fusarium culmorum-specific PCR was performed using primer pair Fc01F/Fc01R to confirm visual identifications. This assay produces a PCR product of 570 bp only with DNA from F. culmorum, but produces no products with DNA from any other fungi tested (Nicholson et al., 1998). Amplification reactions were carried out in 20 µL volumes containing 5–20 ng fungal DNA. The reaction mixture consisted of 80 µm each of dATP, dCTP, dGTP and dTTP, 0·5 µm forward and reverse primers, and 0·6 units Taq DNA polymerase in 10 mm Tris-HCl (pH 9·0), 1·5 mm MgCl2, 50 mm KCl, 0·1% Triton X100. PCR was carried out in a Biometra T3 thermocycler. The PCR reaction was as Nicholson et al. (1998) with an annealing temperature of 66°C for the first five cycles, 64°C for the next five and 62°C for the next 15 cycles. The cycling conditions consisted of denaturation (95°C) for 30 s, annealing (as described previously) for 20 s, and extension (72°C) for 45 s followed by a final extension of 72°C for 5 min. PCR products were separated by electrophoresis in 2% agarose gels and DNA visualized with 1 µg mL−1 ethidium bromide solution.
Determination of F. culmorum chemotype
PCR assays developed to the Tri7 and Tri13 gene sequences (Chandler et al., 2003) were used to determine the chemotype of individual F. culmorum isolates. Chandler et al. (2003) determined that the Tri7 gene was absent in all isolates of F. culmorum that were unable to convert DON to NIV (even though the gene was not essential for this transformation). Two primer sets (MinusTri7F/MinusTri7R and Tri7Fgc/Tri7NIV) were used to characterize the Tri7 gene. Isolates produced a PCR product from either primer set MinusTri7F/MinusTri7R (1078 bp) or Tri7F/Tri7NIV (465 bp) but not from both (Fig. 2). A further two primer sets (Tri13F/Tri13DONR and Tri13NIVF/Tri13R) were used to characterize the Tri13 gene. Isolates produced a PCR product from either primer set Tri13F/Tri13DONR (282 bp) or Tri13NIVF/Tri13R (312 bp) but not from both. The cycling protocol for all primer sets consisted of denaturation at 94°C for 2 min followed by 30 cycles of 94°C for 30 s, 60°C for 1 min and 72°C for 2 min, with a final extension step of 72°C for 10 min. PCR products were analysed as described above. Isolates that produced PCR products following PCR with both Tri7Fgc/Tri7NIV and Tri13NIVF/Tri13R indicated a NIV producer, while isolates that amplified a product with both MinusTri7F/MinusTri7R and Tri13F/Tri13DONR indicated a DON producer.
A PCR assay was developed to the Tri3 gene to allow further differentiation of the DON chemotype into either 3- or 15-acetyl DON. Primers for the Tri3 gene were developed based on information reported within Ward et al. (2002) and the sequences of the Tri3 genes of isolates of F. graminearum, F. culmorum, F. pseudograminearum, F. cerealis and F. lunulosporum referred to in Ward et al. (2002) and deposited in GenBank. Three primer sets [Tri303F/Tri303R (586 bp), Tri315F/Tri315R (864 bp) and Tri3NIVF/Tri3NIVR (549 bp)] were used to characterize the Tri3 gene (Table 2). The three primer sets were validated using isolates of F. graminearum, F. culmorum and F. cerealis reported in Chandler et al. (2003). Each isolate produces a PCR product with only one of these primer sets but not the other two. In each case the result was in accordance with the mycotoxin profile of each isolate (results not shown).
Statistical analysis of chemotype distribution
The distribution of DON and NIV chemotypes was analysed by splitting England and Wales into quarters of approximately equal area, roughly based on county boundaries (Fig. 1). Chi-square analysis was then carried out to test the proportion of DON and NIV chemotypes in each area.
A total of 157 isolates from 77 different field locations were identified as F. culmorum using traditional identification techniques. All of these were confirmed as F. culmorum by PCR analysis (Table 1).
F. culmorum chemotype
Two PCR assays were used to determine the chemotype of each F. culmorum isolate, one targeted at the Tri7 gene and the other at Tri13. Both assays indicated that DON and NIV chemotypes of F. culmorum were present among the isolates tested (Table 1), with 59% being DON and 41% NIV chemotypes. No contradictory chemotyping was evident from the two assays, i.e. isolates identified as one chemotype using the Tri7 primer pairs were identified as the same chemotype with the Tri13 primer pairs. Further analysis of the isolates identified as DON chemotype using primers designed to Tri3 showed that all the F. culmorum DON chemotypes were 3-acetyl DON (Table 1). The use of Tri3NIV primers again confirmed the NIV-producing chemotypes of F. culmorum.
Overall, more fields contained the DON chemotype of F. culmorum (69%) than the NIV chemotype (43%). At 19 of the 77 field locations, more than one F. culmorum isolate was collected and chemotyped. Of these 19 fields, six contained only DON chemotypes and four only NIV chemotypes, while the remaining nine fields contained a mixed population of DON and NIV chemotypes (Fig. 2). In general, the predominant chemotype in the mixed population was related to the field location; where fields were located around the south-west of England and Wales, the NIV chemotype predominated, and elsewhere it was the DON chemotype.
Geographical and seasonal variations in F. culmorum chemotype
Analysis of field location and F. culmorum chemotype indicated differences in the distribution of the two chemotypes. Significantly more NIV chemotypes were found in the south and west (P = 0·01), whereas significantly more DON chemotypes were found in the north and east (P = 0·01) (Fig. 1). Fields containing a mixed chemotype population were predominantly, although not exclusively, found where the NIV chemotype predominated. The single Scottish field, sampled in 1994, contained a mixed-chemotype F. culmorum population (3 DON:1 NIV).
In most years, the number of fields containing DON chemotypes of F. culmorum outnumbered those with the NIV chemotype (Table 3). The only year where NIV chemotypes outnumbered DON chemotypes was 1998. A year-on-year breakdown of F. culmorum by region of isolation (Table 4) showed that 1998 was the only year where F. culmorum was predominantly isolated from the south-west.
|F. culmorum chemotype|
|Year||Number of fields||DON||NIV|
Chemotyping of 157 isolates of F. culmorum from 77 different field locations has shown that both DON and NIV chemotypes are present in the UK. This is consistent with the limited number of mycotoxin surveys carried out on UK grain which have detected the presence of DON and NIV in grain samples (Tanaka et al., 1986; Polley et al., 1991; Turner et al., 1999; Prickett et al., 2000). DON and NIV chemotypes of F. culmorum have been described from other European countries, including Germany, the Netherlands, Italy, Norway and France (Gang et al., 1998; Langseth et al., 1999; Muthomi et al., 2000; Bakan et al., 2001). However, this is the first report indicating the existence of both chemotypes in the UK F. culmorum population.
Overall, the DON chemotype predominated, representing 59% of isolates and being present in 69% of fields compared with 41% of isolates and 43% of fields for the NIV chemotype. The high proportion of NIV chemotypes within the F. culmorum population is consistent with the toxin survey results of Turner et al. (1999), who found NIV in 89% of samples tested. The discrepancy in the percentage of NIV in grain and the proportion of NIV-producing isolates may reflect contamination by other NIV-producing species such as F. cerealis or F. poae. Such high incidence of the NIV chemotype should be of concern, as NIV is potentially more toxic towards humans and animals than is DON (Ryu et al., 1988).
The use of the Tri7 and Tri13 primer pairs to chemotype F. culmorum isolates allowed clear differentiation between DON and NIV chemotypes. The PCR assays to these genes could not be used to further classify the DON-producing isolates based on the production of 3- or 15-acetyl DON described by Miller et al. (1991). These workers termed the former IA and the latter IB. The production of PCR primers designed to the Tri3 region enabled this distinction to be made. Surveys which have carried out full trichothecene analysis on UK grain samples have generally only indicated 3-acetyl DON (Polley et al., 1991; Turner et al., 1999). However, more recently 15-acetyl DON (Edwards, personal communication, 2001) has also been detected. This may indicate that both the IA and IB DON chemotypes exist in the UK F. culmorum population. However, results presented here indicate that all the DON-producing isolates tested were characterized as 3-acetyl DON (IA) producers. This suggests that 15-acetyl DON found in some UK grain samples was produced by another Fusarium species. As 15-acetyl DON production has been reported in F. graminearum (Miller et al., 1991), it is possible that the UK F. graminearum population is the source of the contamination. Results from Norway and Germany (Langseth et al., 1999; Muthomi et al., 2000) also indicated that only 3-acetyl DON was produced by isolates of F. culmorum, although both 3- and 15-acetyl DON types have been reported in France (Bakan et al., 2001).
Unlike F. graminearum, no geographical differences in the worldwide distribution of F. culmorum chemotypes have been identified to date. Results presented here seem to suggest that DON and NIV chemotypes differ in their distribution within the F. culmorum population in the UK. A significantly higher proportion of NIV chemotypes were found in the south and west, whereas a significantly higher proportion of DON chemotypes were found in the north and east. Differences in the geographical distribution of DON- and NIV-producing species in the UK have also been suggested by Tanaka et al. (1986), who observed that NIV contamination predominated in the south-east and DON contamination in the north. The differences in the distribution of NIV and DON chemotypes of F. culmorum observed in the present work provide at least a partial explanation for the findings of Tanaka et al. (1986). The reason for the differences in the distribution of the two chemotypes is unknown and it is possible that differences in the distribution of alternative hosts, soil type, cultivar, cropping practice or temperature may all play a part. Further study would be required to determine this.
Knowledge of F. culmorum chemotype distribution may help tailor forecasting schemes for disease development and mycotoxin contamination on a regional basis. However, for this to be fully effective, it is important to have information on potential differences in chemotype and distribution in the UK of other type B trichothecene-producing species such as F. graminearum. Further work is also required to determine potential differences in aggressiveness between, and fungicide activity on, the two chemotypes.
The level of F. culmorum, and therefore NIV chemotypes, isolated from the south-west has declined since 1998. This decrease in F. culmorum has been accompanied by an increase in levels of F. graminearum in this region (Turner et al., unpublished data). The replacement of F. culmorum by F. graminearum as the predominant FHB pathogen has also been reported in Bavaria (Obst et al., 1997) where the change was linked with increased maize production. Even though there has been increased production of fodder maize, particularly in southern England, to date there is no direct evidence to link the increase in F. graminearum with the increased prevalence of maize in the crop rotation. As there is no chemotaxonomic data currently available for UK isolates of F. graminearum, it is not known how the change in prevalent species may affect the future toxin profile of contaminated grain in the UK.
The authors would like to thank Moray Taylor, Jackie Stonehouse, Jean Slough and Tony Prickett for all their help in the production of this paper. PN gratefully acknowledges the support of Defra in the study of facultative pathogens of cereals within the JIC. Development of the NIV/DON chemotype primers was supported by EU project Detox-Fungi (QLK1-CT-1999-01380).
- 2001. Trichothecene and moniliformin production by Fusarium species from Western Canadian wheat. Journal of Food Protection 64, 1220–5. , , , , , ,
- 2001. Toxigenic potential of Fusarium culmorum strains isolated from French wheat. Food Additives and Contaminants 18, 998–1003. , , , , , ,
- 2002. Inactivation of a cytochrome P-450 is a determinant of trichothecene diversity in Fusarium species. Fungal Genetics and Biology 36, 224–33. , , , , ,
- 2003. Development of PCR assays to Tri7 and Tri13 and characterisation of chemotypes of Fusarium graminearum, Fusarium culmorum and Fusarium cerealis. Physiological and Molecular Plant Pathology 62, 355–67. , , , ,
- 1993. A simple and efficient method for isolating RNA from pine trees. Plant and Molecular Biology Reports 11, 113–6. , , ,
- 2000. Phytotoxicité de huit mycotoxines associées à la fusariose de l’épi chez le blé. Canadian Journal of Plant Pathology 22, 286–92. , , , ,
- 1998. Deoxynivalenol and nivalenol production by Fusarium culmorum isolates differing in aggressiveness toward winter rye. Phytopathology 88, 879–84. , , , , ,
- 1993. The Tri4 of Fusarium sporotrichioides encodes a cytochrome P450 monooxygenase involved in trichothecene biosynthesis. Molecular and General Genetics 248, 95–102. , , ,
- 1983. Chemotaxonomy of Gibberella zeae with special reference to production of trichothecenes and zearelenone. Applied and Environmental Microbiology 46, 1364–9. , , , ,
- 1996. Towards the prediction of fusarium ear blight epidemics in the UK – the role of humidity in disease development. In: Proceedings of the Brighton Crop Protection Conference: Pests and Diseases. Farnham, UK: BCPC Publications, 233–8. , ,
- 2003. The trichothecene biosynthesis gene cluster of Fusarium graminearum F15 contains a limited number of essential pathway genes and expressed non-essential genes. FEBS Letters 539, 105–10. , , , , , , , ,
- 1999. Mycotoxin production and cytotoxicity of Fusarium strains isolated from Norwegian cereals. Mycopathologia 144, 103–13. , , , , ,
- 2002. Tri13 and Tri7 determine deoxynivalenol- and nivalenol-producing chemotypes of Gibberella zeae. Applied and Environmental Microbiology 68, 2148–54. , , , , ,
- 2001. Identification of deoxynivalenol- and nivalenol-producing chemotypes of Gibberella zeae by using PCR. Applied and Environmental Microbiology 67, 2966–72. , , , , , , ,
- 1991. Trichothecene chemotypes of three Fusarium species. Mycologia 83, 121–30. , , , ,
- 1994. Production of trichothecene mycotoxins by Fusarium graminearum and Fusarium culmorum on barley and wheat. Mycopathologia 128, 19–23. , , , , , , ,
- 2000. Characterisation of Fusarium culmorum isolates by mycotoxin production and aggressiveness to winter wheat. Journal of Plant Diseases and Protection 107, 113–23. , , , , ,
- 1998. Detection and quantification of Fusarium culmorum and Fusarium graminearum in cereals using PCR assays. Physiological and Molecular Plant Pathology 53, 17–37. , , , , , , ,
- 1976. Untersuchungen über die morphologische und biologische differenzierung in der Fusarium-Sektion Liseola. Mitteilungen aus der Biologischen Bundesanstalt für Landund Forstwirtschaft, Berlin-Dahlem 169, 1–117. ,
- 1997. On the etiology of Fusarium head blight of wheat in south Germany – preceding crops, weather conditions for inoculum production and head infection, proneness of the crop to infection and mycotoxin production. Cereals Research Communications 25, 699–703. , , ,
- 1991. Survey of Fusarium Species Infecting Winter Wheat in England, Wales and Scotland, 1989 and 1990. Report No. 39. London, UK: Home Grown Cereals Authority. , , , , , , ,
- 2000. Survey of Mycotoxins in Stored Grain from the 1999 Harvest in the UK. Project Report No. 201E. London, UK: Home Grown Cereals Authority. , , ,
- 1988. The acute and chronic toxicities of nivalenol in mice. Fundamental and Applied Toxicology 11, 38–47. , , , , , , ,
- 1986. A limited survey of Fusarium mycotoxins nivalenol, deoxynivalenol and zearalenone in 1984 UK harvested wheat and barley. Food Additives and Contaminants 3, 247–52. , , , , ,
- 1997. The effect of increasing humidity on Fusarium ear blight and grain quality. Proceedings of the Fifth European Fusarium Seminar, Szeged. Cereals Research Communications 25, 825–6. , ,
- 1999. Investigation of Fusarium Infection and Mycotoxin Levels in Harvested Wheat Grain (1998). Project Report no. 207. London, UK: Home Grown Cereals Authority. , , ,
- 2002. Ancestral polymorphism and adaptive evolution in the trichothecene mycotoxin gene cluster of phytopathogenic Fusarium. Proceedings of the National Academy of Science USA 99, 9278–83. , , , , ,