Fusarium graminearum Species Complex and Trichothecene Genotype

The fungal phytopathogen in Fusarium species can cause Fusarium head blight of wheat, barley, oats, and other small cereal grain crops worldwide. Most importantly, these fungi can produce different kinds of mycoxins, and they are harmful to humans and animal health. FAO reported that approximately 25% of the world’s grains were contaminated by mycotoxins annually. This chapter will focus on several topics as below: (1) composition of Fusarium graminearum species complex; (2) genotype determination of Fusarium graminearum species complex strains from different hosts and their population structure changes; (3) genetic approaches to genotype determination in type B-trichothecene producing Fusaria fungi; and (4) some newly identified trichothecene mycotoxins, their toxicity, and distribution of the producers.


Introduction
The fungal phytopathogen in Fusarium graminearum species complex (FGSC) are the primary etiological agent of Fusarium head blight (FHB) of wheat, barley, oats, and other small cereal grain crops worldwide. Besides, the Gibberella ear rot (GER) caused by FGSC and the related species F. verticillioides is one of the most devastating diseases on maize. FHB and GER are economically devastating plant disease that greatly limits grain yield and quality. Warm and humid weather conditions at the flowering stage are conducive to disease development. During the 1990s, economic losses in cereals (wheat and barley) caused by Fusarium were estimated at close to US $3 billion (US $2.5 billion in wheat and US $400 million in barley) and US $520 million (US $220 million in wheat and US $300 million in barley) in the United States and Canada, respectively [1]. It was reported that due to the changes in climatic conditions and in agricultural practices, outbreaks of FHB have occurred more frequent and serious in China. From 2008 to 2015, serious yield loss of wheat caused by FHB was occurred in more than 5 million ha each year.
In addition, infested grain is often contaminated with Fusarium toxins which are harmful to human and animal health and pose a serious threat to food or feed safety. FHB and GER are among the most destructive and economically important Fusarium graminearum Species Complex and Trichothecene Genotype DOI: http://dx.doi.org /10.5772/intechopen.89045 cosmopolitan species. Applying the genealogical concordance phylogenetic species recognition (GCPSR), FGSC was first divided into seven phylogenetic lineages in 2000 [3]. Phylogenetic analyses of multilocus genotyping (MLGT) of DNA sequences from portions of 13 housing keeping genes, combined with GCPSR and molecular marker technologies, it revealed that this morphospecies comprises at least 16 biogeographically structured, phylogenetically distinct species. After that the species designation Fusarium graminearum have been sensu stricto in some conditions. Up to now, 15 of the 16 species have been formally described, including F. acaciae-mearnsii, F. aethiopicum, F. asiaticum, F. austroamericanum, F. boothii, F. brasilicum, F. cortaderiae, F. gerlachii, F. graminearum sensu stricto, F. louisianense, F. meridionale, F. mesoamericanum, F. nepalense, F. ussurianum, F. vorosii, and one additional species was informally recognized based on genealogical exclusivity and conidial morphology on SNA [4].
Proper species identification is critical to research aimed at improving disease and mycoxins control programs. However, it is difficult to discriminate the FGSC strains accurately by morphological characters. A partial region of the translation elongation factor 1 alpha gene (TEF-1α) was widely used for molecular identification of Fusarium genus. Some specific databases were created for Fusarium DNA sequence alignment analysis. For example, similarity searches of the obtained sequences can be performed with the Pairwise DNA alignments network service of the Fusarium MLST database (http://www.westerdijkinstitute.nl/fusarium/), Basic Local Alignment Search Tool (BLAST) network service of the Fusarium ID database (http://www.fusariumdb.org/index.php), and NCBI nucleotide database.

Mycotoxins produced by FGSC
In addition to yield reduction, the FGSC fungi are also of concern because they can produce different kinds of mycotoxins, e.g. zearalenone (Figure 2) and trichothecenes (Figures 4 and 5) in infested grains. Mycotoxin contamination can occur in both unprocessed and processed grains, representing a risk for human and animal health. Deleterious health effects caused by different mycotoxins include nephropathy, infertility, cancer or death [5].
Up to now, more than 200 trichothecenes have been identified [6]. Due to the chemical structure diverse, trichothecenes are divided into four types, namely type A (have a single bond at carbon atom 8, C-8), e.g. T-2 toxin (Figure 4), type B (have a keto at C-8), type C (have an epoxide at C-7, 8), and type D (have a macrocyclic ring between C-4 and C-15). All trichothecenes share a common tricyclic 12, 13-epoxytrichothec-9-ene, and they are derived from the isoprenoid intermediate farnesyl pyrophosphate via a series of biochemical reactions in Fusarium.
Among these mycotoxins, type B trichothecenes ( Figure 5) are the most common detected in cereal grains and their related products. They are distinguished from type A by the presence of a keto function at C-8, and include deoxynivalenol (DON) and its acetylated forms 3-acetyldeoxynivalenol (3-ADON) and 15-acetyldeoxynivalenol (15-ADON), as well as nivalenol (NIV) and its acetylated form 4-acetylnivalenol (4-ANIV). DON is characterized by the absence of a hydroxyl function at C-4, whereas NIV is characterized by the presence of a hydroxyl function at C-4. 3-ADON and 15-ADON are the acetylated forms of DON at C-3 and C-15, respectively. Meanwhile, NIV and 4-ANIV can be differed by the absence (NIV) and presence (4-ANIV) of an acetyl function at C-4 ( Figure 5).
Type B trichothecenes are of the greatest concern in wheat and barley-growing regions worldwide, because they can represent a major threat to food and feed safety. These toxins are potent inhibitors of protein synthesis and are responsible for neurologic, gastrointestinal, immune function and other disorders. Although type B trichothecenes differ only slightly from each other in terms of the pattern and position of acetylation or hydroxylation, these changes can greatly affect the toxicity and the activity of these chemical compounds [7]. For example, DON is associated with feed refusal, vomiting and suppressed immune functions, and NIV is more toxic to humans and domestic animals than is DON. Determination of these trichothecene variations are important because the toxicity of DON and NIV may vary according to the eukaryotic organism affected. Minervini et al. [8] found that NIV was approximately four times more toxic than DON to human cells. Conversely, DON is 10 to 24 times more toxic to plant cells than NIV [9].
Type B trichothecenes are mainly produced by FGSC. Due to the ability of FGSC strains that producing different kinds of trichothecenes, three strain-specific trichothecene genotypes (chemotypes) were identified in FGSC: the 3-ADON genotype produces DON and 3-ADON, the 15-ADON genotype produces DON and 15-ADON, and the NIV genotype produces NIV and its acetylated derivatives 4-ANIV [10].

Mycotoxins act as virulence on certain hosts
Evidence is presented to show that trichothecene toxins act as virulence factors on certain hosts. Strains carrying a disrupted trichodiene synthase gene Tri5 do not produce trichothecenes or their biosynthetic intermediates [11]. Disruption of the trichodiene synthase which catalyzes the first step in trichothecene biosynthesis in a 15-ADON producer GZ3639 exhibited reduced virulence on seedlings and heads of wheat, but wild-type virulence on seedlings of maize. The results indicated that trichothecene production contributes to the virulence of FGSC on wheat [12]. Eudes et al. [13] confirmed earlier findings that trichothecenes are a principal determinant  of F. graminearum aggressiveness on wheat cultivars. More recently, Maier et al. [14] investigated the involvement of trichothecenes in the virulence of the pathogen by disrupting the Tri5 gene encoding the first committed enzyme trichodiene synthase in FGSC strains with 3-ADON, 15-ADON, and NIV genotype, respectively. The results demonstrated that disruption mutants can cause disease symptoms on the inoculated spikelet but the symptoms cannot spread into other spikelets on wheat. And on maize, mutants derived from the NIV genotype strain caused less disease than their progenitor strain, while no significant difference compared to the wild-type strains were observed on barley [14]. Trichothecenes are therefore also potent phytotoxins and act as virulence factors of pathogenic fungi thus facilitate tissue colonization on sensitive host plants, e.g. of DON produced by F. graminearum in wheat [15].
Host preference was identified among the FGSC on wheat, maize, barley, and rice in certain regions. Several studies suggested that maize played a significant role for the presence of the NIV genotype for FGSC. NIV producers were found to be more aggressive toward maize compared to DON producers [16]. This can be due to the fact that NIV is a virulence factor useful for maize colonization [14], and therefore the plant probably represents an ecological niche for hosting the NIV genotype strains. A high proportion of NIV producers of FGSC on maize were observed in China by our group based on a collection from 59 districts in 19 provinces throughout China, and NIV producers preferentially to maize [17].

Distribution and population changes of FGSC
The species composition and genotype prevalence of FGSC vary widely in different regions, which reflecting the level of risk factors in feed/food safety. Investigations on Fusarium species isolated from wheat, barley, and maize crops have been reported in the last two decades. Dynamic changes of species composition and chemotype proportion have been found in different agricultural ecosystems worldwide. Prior to 2000, strains from the United States and Canada were almost exclusively 15-ADON producers, while they have been increasingly replaced by the 3-ADON producers in some major wheat-growing regions, e.g. the frequency of the 3-ADON genotype in western Canada increased more than 14-fold between 1998 and 2004 [18]. Also Schmale et al. [19] analyzed the trichothecene genotypes of Gibberella zeae collected from winter wheat fields in the eastern parts of the US. They revealed an increasing gradient in 3-ADON distribution from south to north and closer to Canada. In some regions, the F. graminearum 15-ADON chemotype is being replaced by the 3-ADON chemotype. The epidemiology data indicated that 3-ADON chemotype dominates in northern Europe, while 15-ADON chemotype dominates in North America, central Europe and southern Russia and some parts of Asia.
The composition of FGSC population appears to be host and location dependent. The results by Zhang et al. [20] and Shen et al. [21] indicated that F. asiaticum was the predominated in wheat. Among the 97 FGSC assayed from rice (30 strains), maize (33 strains), and wheat (34 strains) by Qiu and Shi [22], 73 strains were identified as F. asiaticum and 63 of them were collected from rice or wheat. The remaining 24 strains belonged to F. graminearum sensu stricto and 23 of them were isolated from maize, only 1 strain was collected from wheat. Similarly, FGSC strains were isolated from GER samples in South Korea with F. graminearum sensu stricto to be the dominant species which account for 75% of the FGSC [23], while F. asiaticum was the dominant species (78.5%) on Korean rice and followed by F. graminearum sensu stricto [24].
The distribution of FGSC may correlate with annual temperature. Qu et al. [25] reported that temperature affected the geographic distribution of F. graminearum sensu stricto and F. asiaticum on wheat spikes in China. A comprehensive study on FGSC from wheat was conducted by Zhang et al. [20]. They found that the geographic distribution of FGSC associated with the annual average temperature. The cooler temperatures (annual average temperature ≤15°C) appear to favor F. graminearum sensu stricto, while the warmer regions (annual average temperature ≥15°C) appear to favor F. asiaticum. A hypothesis was made that the distribution of FGSC members are climate dependent [20].
F. graminearum sensu stricto with the 15-ADON genotype and F. asiaticum with either the NIV or the 3-ADON genotype were the dominant causal agents on wheat, and the two species dominated the northern and southern regions of China, respectively, which is consistent with earlier studies [20,26,27].
However, more recently the study by Zhang et al. [28] indicated that temperature may not be the only factor in the distribution of FGSC and that other, yet unknown factors affected their distribution. To explain genotype distribution in different geographic areas, hypotheses based on grain seed shipment, international trade, long-distance spore transportation, and environmental favorable conditions were proposed.

FGSC fitness vary
Phylogenetic analyses of trichothecene gene cluster demonstrated that genotype polymorphism is trans-specific and have been maintained by balancing selection on the ancestral pathogens, and genotype differences may have a significant impact on pathogen fitness [29].
The FGSC strains with different genotype showed different fitness to the ecological environment, such as the hosts, temperature, rotation, and so on. 3-ADON producer was more aggressive than 15-ADON population in susceptible wheat, and also the 3-ADON isolates exhibit a higher DON production than the 15-ADON isolates. Similar conclusions were made by Zhang et al. [28] that F. asiaticum strains with 3-ADON chemotype revealed significant advantages over the strains that produce NIV in pathogenicity, growth rate, trichothecene accumulation, etc. Their data also indicated that the growth of rice may be a key factor for the presence of F. asiaticum [28]. Liu et al. [30] compared the fitness of three chemotype Fusarium strains, and found that 15ADON producers had the advantage in perithecia formation and ascospore release, whereas more DON were produced by the 3-ADON chemotypes. Qiu and Shi [22] estimated the effect of rice or maize as former crops on mycotoxin accumulation in wheat grains, and they concluded that rice-wheat rotation favors DON accumulation.
Changes in DON chemotypes distribution were reported for FGSC from Canada, USA, and Northern Europe. Recently, Nicolli et al. [31] assessed a range of fitness-related traits (perithecia formation, mycelial growth, sporulation and germination, pathogenicity, and sensitivity to tebuconazole) with 30 strains representatives of 3ADON-, 15ADON-, and NIV-producers. The pathogenicity assay results indicated that strains with the DON chemotypes were generally more aggressive than the NIV ones [31].
Phenotypic analyses indicated that F. asiaticum with a 3-ADON genotype revealed significant advantages over F. asiaticum that produce NIV in pathogenicity, growth rate, and trichothecene mycotoxin accumulation. It shall be noted that a biased gene flow from 3-ADON to NIV producers was identified in F. asiaticum from wheat in China [28].
FGSC from wheat-maize rotation regions on wheat spikes and maize stalks in Henan province, China, was determined by Hao et al. [32], and significant Fusarium graminearum Species Complex and Trichothecene Genotype DOI: http://dx.doi.org/10.5772/intechopen.89045 differences were found in the frequencies of F. graminearum sensu stricto and F. asiaticum species within the hosts with F. graminearum sensu stricto to be the dominant. Genotype analysis revealed that 15-ADON producers represented 92.7 and 98.5% of isolates from wheat and maize, respectively. The three genotypes may affect species distribution or population ecology because these mycotoxins are differing in toxicity and bioactivity [7,29,33].

Genetic genotype determination of FGSC
Traditionally, chemotyping of FGSC strains has been carried out using gas chromatography/mass spectroscopy. This method can be time-consuming and expensive. The genome sequences of several FGSC strains have been published. The trichothecene core gene cluster nucleotide sequences of many strains representatives 3-ADON, 15-ADON, and NIV genotypes have also been deposited in the GenBank. The availability of this information makes it possible to reveal the structural features and allowed selection of several primer sets used successfully in PCR experiments for the molecular characterization of the various chemotypes. Molecular genetic assays allow for high throughput screening of large numbers of field isolates.
Lee et al. [34] sequenced the gene cluster for trichothecene biosynthesis from a 15-ADON producer (strain H-11) and a NIV producer (strain 88-1), and sequence polymorphisms within the Tri7 open reading frame was found between the two strains. Alignment analysis suggesting that the Tri7 gene of H-11 carried several mutations and an insertion compared to the Tri7 gene from 88-1, and based on the sequence difference a PCR-based diagnostic method for differentiating DON and NIV producers by polyacrylamide gel electrophoresis was developed.
Lee et al. [35] subsequently sequenced the Tri13 homolog from DON (strain H-11) and NIV producers (strain 88-1) and found that the gene differs drastically between the two producers, suggesting that the Tri13 gene could be used for genetic genotype distinction for DON and NIV producers [35,36]. They further confirmed the roles of the Tri7 and Tri13 genes in trichothecene production, and the results suggested that both the Tri7 and Tri13 genes are nonfunctional in DON producers [35].
The PCR assays to Tri7 and Tri13 genes developed by Lee et al. [34,35] allowed clear differentiation between DON and NIV genotypes. However, they could not be used to further classify the DON-producing isolates to 3-ADON or 15-ADON producer. Ward et al. [29] examined a 19-kb region of the trichothecene gene cluster that sequenced in 39 strains representing 3-ADON, 15-ADON, and NIV genotypes. They found that Tri-cluster haplotypes group according to genotype rather than by species indicated that 3-ADON, 15-ADON, and NIV genotypes each have a single evolutionary origin. Reciprocally monophyletic groups, corresponding to each of 3-ADON, 15-ADON, and NIV genotypes, were strongly supported in Tri3, Tri11, and Tri12 genes trees. Two sets of primers specific to the individual genotypes were designed from Tri3 and Tri12 genes. The genotype-specific PCR tests developed by Ward et al. [29] provide a rapid and direct genetic method for distinguishing among 3-ADON, 15-ADON, and NIV producer, this is the first report differentiated these three genotype strains by a PCR method.
The work by Lee et al. [34,35] and Brown et al. [37] indicated that the genes Tri13 and Tri7 from trichothecene biosynthetic cluster are responsible for conversion of DON to NIV (Tri13 gene) and the Tri7 gene product modifies NIV by acetylation of C-4 atom hydroxyl to produce 4-ANIV. Based on these results sets of positive-negative PCR assays to Tri7 and Tri13 genes for trichothecene determination of FGSC were developed by Chandler et al. [38], and the assays can accurately indicate a DON or NIV genotype in FGSC, F. culmorum and F. cerealis. The assays were successfully used to screen isolates from different countries and the genotypespecific assays were able to detect and characterize a wider range of species and haplotypes than previous methods.
By comparing the published sequences for Tri13 gene from known DON-and NIV-producers, Waalwijk et al. [39] designed a primer pair to discriminate the two genotypes which generated a 234 bp fragment in DON-producers and a fragment of 415 bp in NIV-producers. The Tri13 primer pair was capable and robust to determine the genotype of strains from F. culmorum.
Based on information reported and deposited by Ward et al. [29], three primer sets were designed to the Tri3 gene by Jennings et al. [40] to allow further differentiation of the DON genotype into either 3-ADON or 15-ADON. Each isolate produces a PCR product with only one of these primer sets but not the other two from F. culmorum and FGSC strains [40,41].
Li et al. [42] found that the intergenic sequences between Tri5 and Tri6 genes appear to be mycotoxin genotype-specific, and based on the sequence length polymorphism a generic PCR assay was developed to detect a 300 bp fragment of DON-genotype strains and a 360 bp fragment of NIV-genotypes from FGSC.
Based on the sequences of FGSC described by Lee et al. [34] and Ward et al. [29], a series of PCR assays have been designed to Tri3 and Tri7 by Quarta et al. [43], in order to permit specific detection of 3-ADON, 15-ADON, and NIV genotypes, respectively. These primers were subjected to a multiplex PCR assay for the identification of the different genotypes of Fusarium strains combined with the primer pair derived from the Tri5 gene by Bakan et al. [44]. The multiplex PCR was validated on FGSC, F. cerealis, F. culmorum strains from different European countries, and successfully used to identify the genotype of the Fusarium strain contaminating wheat kernels [43,45].
The possibility to distinguish by a singleplex PCR 3-ADON, 15-ADON, and NIV genotypes was not yet resolved until very recently. Wang et al. [46] developed a Tri13 based PCR assay and successfully identified the 3-ADON, 15-ADON, and NIV genotypes in FGSC from Asia, Europe, and America. Using the primer pair, specific amplification products of 644, 583, and 859 bp were obtained from isolates producing 3-ADON, 15-ADON, and NIV, respectively. All three types of PCR fragments had different molecular sizes with a smallest difference of 61 bp can be directly differentiated on an agarose gel. The method should be more reliable than other PCR-based assays that show the absence or presence of a PCR fragment since these assays may generate false-negative results. This is a rapid, reliable and cost-effective method for the determination of 3-ADON, 15-ADON, and NIV genotype strains in FGSC.
Recently Suzuki et al. [47] reported a multiplex PCR assay for simultaneous identification of the species and trichothecene genotypes for F. graminearum sensu stricto and F. asiaticum based on Tri3 and Tri6 genes. This approach proved successful for Japanese strains [47].
An alternative method based on Tri11 polymorphism was developed by Zhang et al. [48] to differentiate 3-ADON, 15-ADON, and NIV genotypes of FGSC strains. Similarly, we presented another multiplex assay based on the single nucleotide polymorphism of Tri11 gene between strains of different genotype [49]. The assay was also validated on plant material.
Recent work by Kulik [50] and Nielsen et al. [51] to detect and quantify FGSC genotypes in plants/grains were developed based on TaqMan probe set and SYBR green method with Tri12 gene, respectively.
Due to the toxicological differences between DON and NIV, it is important to monitor the population and determine the chemotypes of strains present in any given geographic region. Mycotoxin producing capability of a certain strain could be established both through biochemical and molecular techniques.  The biochemical approach involves the incubation and extraction of mycotoxins, the methods being complicated and time consuming. The molecular techniques are based on detection of specific gene by using specific primers. All these molecular methods developed for genotype analysis are based on nucleotide diversity of trichothecene synthesis genes. Chemotype characterization has been extensively used to characterize FGSC for their toxigenic potential [52]. The information about the genetic genotyping methods developed so far, such as targeted gene, primer name, primer sequence, and amplification fragment sizes are summarized in Table 1.
More effective and accuracy genetic methods are needed. We are doing genomic sequencing of FGSC strains with different trichothecene genotypes, and we believe some new molecular genetic methods will be developed based on the genomic data.

Newly identified trichothecene mycotoxins
In addition to the well characterized fungal mycotoxins, plant-derived mycotoxin metabolites, masked mycotoxins, have emerged as important co-contaminants in cereals [53,54]. The most commonly detected masked mycotoxin conjugates are β-linked glucose-conjugates of trichothecenes, such as DON-3-glucoside ( Figure 6). The possible hydrolysis of masked mycotoxins back to their toxic parents during mammalian digestion raises great concerns. Recently, a new series of type-A trichothecene, NX-toxins (Figure 7), produced by FGSC were characterized [5]. In vitro translation assays indicated that NX-3 can inhibit protein biosynthesis to almost the same extent as DON [5]. Comprehensive work on intestinal hydrolysis, absorption, metabolism, and toxicity of newly characterized mycotoxins need to be determined (Figure 7).

Conclusion
The knowledge about the mycotoxins chemotypes could contribute to a better management of fungal infections and breeding of resistance, in order to obtain grains of better quality. The results will also contribute to improve our understanding of the ecology and epidemiology of FGSC members, which may be of value for  improving models for assessing the risk or epidemics and mycotoxin production. Genetic genotyping has been proved to be a useful tool for predicting trichothecene type produced by FGSC, and future work on the more effective tools for genotype determination is needed. The discovery of novel toxic metabolites belonging to trichothecenes, such as NX-toxins is also suggesting that the prevalence, distribution, and genetic diversity of FGSC require continuous monitoring. Further research on the biosynthesis molecular mechanism of trichothecene, especially the novel mycotoxins is needed.