WO2009112505A2 - Plants having improved resistance to pathogens - Google Patents

Abstract

The invention relates to methods for increasing pathogen resistance in plants and decrease the production of mycotoxin in infected plant products by modulation of the polyamine biosynthetic pathways and to plants obtained thereof.

Description

PLANTS HAVING IMPROVED RESISTANCE TO PATHOGENS

The invention relates to field of agriculture and to the resistance to disease and the safety of food and feed products of cultivated crops. Fusarium head blight (FHB) is one of the most economically important diseases of wheat world wide. Fusarium spp. that cause FHB on wheat are also capable of causing this disease on other plants, and in particular small grain cereals such as barley and many also cause ear rot in maize. Globally, members of the species complex containing F. graminearum (teleomorph Gibberella zeae) are the most common cause of FHB. Members of the F. graminearum species complex together with many other more distantly-related Fusarium species, such as F. pseudograminearum and F. culmorum, are all economically important causes of head blight disease in various regions in the world.

Many fungal pathogens in the genus Fusarium (teleomorph Gibberella) that infect cereals can produce trichothecene mycotoxins during infection.

It has been demonstrated that production of trichothecene mycotoxins such as deoxynivalenol (DON) are important for infection and virulence on cereals. Mutants of F. graminearum unable to produce trichothecene toxins show reduced aggressiveness on wheat (Bai et al., 2002, Mycopathologia. 153, 91-98; Jansen et al., 2005, PNAS. 102, 16892-16897; Proctor et al., 1995, MoI. Plant Microbe Interact. 8, 593-601 ). Infection by F. graminearum occurs at the time of anthesis and is thought to be through the anthers, followed by colonisation of other parts of the floret before entering the vasculature and pith of the rachis from where it spreads up and down the head (reviewed in Trail et al., 2005, Canadian Journal of Plant Pathology. 27, 486-498). Although trichothecene non-producing mutants can cause initial infection, they are unable to spread throughout the head, and their growth is restricted at the rachis node between the rachis and floret (Jansen et al., 2005, op. cit). Even if the mechanism of virulence is not fully clear, there is such genetic evidences that the DON production and accumulation in the plant is tightly correlated to the spread resistance and yield loss (Jianga et al., 2006, Crop Sci 46:2590-2597).

In addition to their role in fungal virulence, many trichothecenes, such as DON, nivalenol (NIV) & T2 toxin (produced by F. sporotrichioides, a pathogen of maize), are harmful to humans and livestock when consumed in food/feed products derived from infected plants. Thus, the presence of trichothecene toxins in grain is coming under increasing scrutiny in both Europe and the United States of America, where legally enforceable limits on contamination by trichothecenes such as DON in food products are now in place. In mammalian systems, DON has been shown to act as a potent translational inhibitor that lead to the alteration of the synthesis of the proteins and causes acute symptoms such as vomiting and feed refusal in livestock and also chronic symptoms such as growth retardation, reduced ovarian function and immunosuppression. Processes that can reduce mycotoxin production in grain products would have important utility in improving animal feed and human food safety.

This invention is based initially on the identification of compounds that can induce the expression of the TRI5 gene of F. graminearum. This gene encodes the trichodiene synthase enzyme that catalyses the conversion of farnesyl pyrophosphate to trichodiene and this is the first committed step from primary metabolism to the secondary metabolism pathway of all trichothecene mycotoxin biosynthesis. TRI5 is conserved in all trichothecene producing Fusarium spp. including F. graminearum, F. pseudograminearum, F. culmorum, F. sporotrichioides, F. crookwellense, F. poae, F. tricinctum, F.acuminatum, F. sambucinum, F. venenatum as well as Myrothecium roridum (Fekete et al., 1997, Mycopathologia. 138, 91-97; Hohn and Desjardins, 1992, MoI Plant Microbe Interact. 5, 249-56; Proctor et al., 1995, MoI. Plant Microbe Interact. 8, 593-601 ; Royer et al., 1999, Fungal Genetics and Biology. 28, 68-78; Trapp et al., 1998, Molecular and General Genetics. 257, 421-432). Its expression is regulated by the pathway specific transcription factor TRI6 (Proctor et al., 1995, Applied and Environmental Microbiology. 61 , 1923-1930) and regulatory protein TRHO (Peplow et al., 2003, Applied and Environmental Microbiology. 69, 2731-2736). The TRI6 binding sites are conserved upstream of all TRI5 homologous genes in Fusaria. To create an easy assay for the expression of this gene the regulatory promoter region of the TRI5 gene was fused to the gene encoding the jellyfish green fluorescent protein and this new chimerical gene introduced into F. graminearum as described in Figure 1. The transformed fungal strains showed fluorescence during the infection of wheat heads at times when the trichothecene mycotoxins are known to be produced. An example of this fluorescence is shown in Figure 2. Thus these new transgenic strains appear to act as 'reporter strains' for TRI5 gene expression and toxin production and were subsequently used to identify compounds that induce toxin production. It is to be noted that some Fusarium strains do not produce trichothecene, and that this may be due to absence of function in most genes that are important for toxin production, as was demonstrated by Kimura et al. in Fusarium oxysporum or F. moniliforme that are trichothecene non-producing Fusaria (Genetics. 2003 163,

677-684).

Prabhavathi and Rajam (Plant Biotechnology, 2007 24, 273-282) describe generation of transgenic eggplants expressing oat adc. These plants showed increased tolerance to F. oxysporum, which is not a trichothecene producing pathogen.

Wale and Rajam (Plant Science, 2003 164,727-734) describe transgenic tobacco transformed with human SAMDC. The plants are tolerant to Verticillium dahliae and F. oxysporum, which are not trichothecene producing pathogens. Di and Turner (Molecular Plant-Microbe Interactions, 2005, 18, 762-770) describe resistance to Fusarium DON in tobacco plants expressing a ribosomal protein.

Capell et al (Molecular and General Genetics, 2000, 264, 470-476) and Trung-

Nghia et al (PLANTA (BERLIN), 2003, 218,125-134) describe generation of transgenic rice with downregulated ADC through expression of the antisense oat ortholog. No information relating to resistance to trichothecene producing pathogens is given for these plants.

Rajam (Current Science (BANGALORE), 1993, 65, 461 -469) suggests the use of DFMO, a polyamine synthesis inhibitor, to inhibit the growth of Fusarium oxysporum (Table 3), which is not a trichothecene-producing pathogen. It is to be noted that the teachings of this article are conflicting with the data reported by the same author and mentioned above, in transgenic plants.

US 6,825,216 describes the use of an alkaloid to inhibit synthesis of aflatoxin and deoxynivalenol. This document does not mention polyamines.

The invention is based on the demonstration that the production of trichothecene mycotoxins including DON is stimulated in vitro by a range of metabolites, most of which contain either one or multiple amine groups.

The invention is thus based on the concept that presence of polyamines in plants is key for infection with a trichothecene producing pathogen, and for the production of the toxins. The invention is thus based on the concept that suppressing availability of these polyamines will control the infection and reduce the production of the mycotoxins, as demonstrated in the examples.

During in vitro growth of the GFP-expressing F. graminearum strain, induction of DON was observed when such identified inducers, including putrescine, arginine and cadaverine, were applied in the media, alone or combined with others. Moreover, during the infection of wheat heads by Fusarium graminearum, induction was observed for several of the identified inducers including putrescine, arginine, cadaverine and lysine. In addition, Fusarium infection of wheat heads led to the induction of wheat gene transcripts encoding enzymes responsible for the synthesis of arginine and polyamines which in turn are strong inducers of trichothecene synthesis in the fungus. Induced cereal gene sequences include those encoding arginine decarboxylase, ornithine decarboxylase and argininosuccinate synthase, in particular. The invention describes that the production of trichothecene toxins can be reduced by the selection of plants that accumulate lower amounts of the mycotoxin inducers, in particular in the parts of the plant where the production of mycotoxins takes place. Specifically, the reduction of inducing compounds in the host plant can be obtained via the manipulation of host genes, e.g., arginine decarboxylase, ornithine decarboxylase and argininosuccinate synthase, that are responsible for the production of amine/polyamine inducers of mycotoxins. These manipulations can lead to increased resistance to infection, reduced toxin level in plants, and thus to improved food and feed safety, and increased yield via plant resistance to Fusarium pathogens. Alternatively, it is possible to reduce the level of infection by Fusarium and/or the level of mycotoxins by reducing the availability of inducers to the fungi. These can be reduced by limiting their synthesis by the plant, increasing their degradation, or by using competitor chemicals that would saturate the Fusarium amine sensing or transport systems.

Several patent applications describe methods that result in detoxification of trichothecenes (WO200060061 , WO2005021740 or WO200020573), identify an insensitive molecular target site (US2002088022 or WO200105976) or make cells less sensitive to trichothecene toxins (WO2006070815 or US2003165991 ). The inventors have now identified potent stimulators of trichothecene production by Fusarium pathogens. The use of this information makes it possible to reduce the production of DON or other trichothecenes by the manipulation of plant gene expression in plant tissues.

US2005026953 indicates that it is possible to reduce the production of DON by the application of a chemical inhibitor alkenylpiperidine. The use of gene technology to reduce the induction of other mycotoxins by other pathogenic fungi such as the silencing of plant lipoxygenases to reduce induction of aflatoxins by Aspergillus spp. is described in WO9726364. However, the production of aflatoxins by Aspergilli and the production of trichothecenes by Fusaria do not appear to be related either in their biosynthetic pathways, their regulation, their effect on humans and their role in plant pathogenesis.

The production of trichothecenes by F. graminearum appears to be somewhat specific to host infection. Autoclaved grain and axenic cultures are often used for the analysis of trichothecene production. However, the levels of accumulated toxin observed in most in vitro culture conditions are orders of magnitude lower than those measured during the infection of live wheat heads and other tissues (Mudge et al., 2006, Physiological and Molecular Plant Pathology. 69, 73-85; Voigt et al., 2007, European Journal of Plant Pathology. 117, 1-12). In some instances, complex sequential nutritional regimes have been reported to partially induce DON production in culture (Miller and Blackwell, 1986, Canadian Journal of Botany. 64, 1-5), but factors responsible for high level induction of DON have not been identified. There is also evidence that H₂O₂, a product of the host oxidative burst sometimes elicited during plant infection, may enhance DON production in culture (Ponts et al., 2006, Ferns Microbiology Letters. 258, 102-107) but the levels of DON induced by H₂O₂ in vitro are still very low when compared to those that accumulate in infected plants.

To find in vitro conditions that allow trichothecene production to levels similar to in planta, the inventors have used novel approaches such as nutritional profiling. The use of large scale nutritional profiling had never been applied to study mycotoxin production. This approach was used to identify a series of molecules that can induce trichothecene production in vitro. Many of the identified inducing compounds were either precursors or products in the polyamine biosynthetic pathway of plants and this pathway appears indeed to be activated early on during the infection of wheat heads. The polyamine biosynthetic pathway is well known, as several patent applications propose to manipulate it in transgenic plants. One could cite JP2006020601 , JP2006067947, WO2006057306, US2006225154,

US2004086985, JP2003199582, WO2005018307, JP2006325554,

WO200109358, WO200218547, WO2001 11062, WO200223974, WO2003084314.

However, these patent applications relate, for most of them to abiotic stress tolerance and none of them cite nor suggest that such a strategy can be used to obtain a reduction in mycotoxin production or an increased resistance to trichothecene-producing Fusarium pathogens.

The invention thus relates to a method for inducing resistance and/or tolerance to a trichothecene-producing pathogen infection in a plant, comprising the step of inhibiting or downregulating expression of an endogenous polyamine synthesis gene in said plant, said inhibition or downregulation resulting in said plant to have increased resistance to said trichothecene-producing pathogen infection as compared to a nearly isogenic plant in which said endogenous polyamine synthesis gene is not inhibited or downregulated.

The invention also relates to a method for reducing trichothecene level in a plant infected by a trichothecene-producing pathogen, comprising the step of inhibiting or downregulating expression of an endogenous polyamine synthesis gene in said plant, said inhibition or downregulation resulting in said plant to have reduced trichothecene level when infected by trichothecene-producing pathogen as compared to a near isogenic plant in which said endogenous polyamine synthesis gene is not inhibited or downregulated.

The invention can also be performed by inhibiting or downregulating expression of an endogenous polyamine synthesis gene in a part of a plant including rachis, stem, spikelet or spike, thereby inducing resistance to a trichothecene-producing pathogen infection or reducing trichothecene level in said part of a plant.

In a preferred embodiment, said trichothecene is deoxynivalenol (DON). In a preferred embodiment, said trichothecene-producing pathogen is a Fusarium from Fusarium spp, and is in particular chosen among F. graminearum, F. pseudograminearum, F. sporotrichioides, F. poae, F. sambucinum, F. venenatum, F. culmorum, F. cerealis, and F acuminatum. In another embodiment, said trichothecene-producing pathogen is Myrothecium roridum. In the context of the present invention, nearly isogenic plants are plant whose genome is at least 90 % identical, more preferably at least 95 % identical. In a preferred embodiment, the genomes of near isogenic plants only differ at one or a few specific loci. In the context of the present invention, said loci are the loci corresponding to the genes involved in the polyamine biosynthetic pathway. Inhibition or down-regulation of these genes will make the plant more resistant to infection by the pathogen.

The present invention makes it possible to induce resistance and/or tolerance in a plant that does not have, per se, a level of resistance and/or tolerance to pathogen infection. It makes also it possible to increase resistance and/or tolerance to pathogen infection in plants which already have a basal level of resistance and/or tolerance.

As used herein, the terms "altering", "increasing", "increased", "reducing", "down-regulating", "reduced", "inhibited" or the like are considered relative terms, i.e. in comparison with the wild-type or unaltered state, preferably in nearly isogenic plants. The "level of a protein" refers to the amount of a particular protein, which may be measured by any means known in the art such as, for example, Western blot analysis or other immunological means. The "level of an enzyme activity" refers to the amount of a particular enzyme measured in an enzyme assay. It would be appreciated that the level of activity of an enzyme might be altered in a mutant but not the expression level (amount) of the protein itself. Conversely, the amount of protein might be altered but the activity would the same if a more or less active protein is produced. Reductions in both amount and activity are also possible such as, for example, when a gene encoding the enzyme is inactivated. In certain embodiments, the reduction in the level of protein or activity is by at least

40% or by at least 60% compared to the level of protein or activity in the unmodified plant, or by at least 75%, at least 90% or at least 95%.

Resistance and/or tolerance to pathogen infection represents the ability for the plant to limit the growth of the pathogen and/or to limit the production of trichothecenes by the pathogen during infection. Resistance and/or tolerance can be monitored by a reduction in necrosis, bleaching and spread of the fungus (growth), or measurement of lower amount of mycotoxins including DON and derivates by usual methods in the infected plant. As will be seen, these effects are usually observed. In the context of the invention, a polyamine synthesis gene (also mentioned as gene of the polyamine biosynthetic pathway) is a gene coding for an enzyme that is active during the synthesis of polyamines, and/or makes use or produces a polyamine, such as L-ornithine, putrescine, L-arginine, agmatine, citrulline, N- carbamoylputrescine, spermine and spermidine, including genes involved in the synthesis of precursors thereof. This term may also include, by extension, a gene coding for an enzyme that is active during the synthesis of, and/or makes use or produces cadaverine, such as lysine decarboxylase. Regulation of these latter genes is also part of this invention. Polyamines are small, ubiquitous, nitrogenous compounds that have been implicated in a variety of stress responses in plants (Bajaj et al., (1999) MoI. Breeding 5, 493-503). Their biosynthesis pathway is well known and described in the literature.

In a specific embodiment, said gene is chosen in the group consisting of arginine decarboxylase (ADC, EC 4.1.1.19), ornithine decarboxylase (ODC, EC 4.1.1.17), agmatinase (EC 3.5.3.1 1 ), argininosuccinate synthase (ASS, EC

6.4.3.1 ), and argininosuccinate lyase (EC 4.3.2.1 ). It can also be a lysine decarboxylase, producing cadaverine from lysine.

In a preferred embodiment, said plant is a monocotyledon. In the most preferred embodiment, said plant is a cereal, in particular chosen in the group consisting of wheat, maize, rice, rye, barley, triticale, sorghum and oat. In a specific embodiment, said cereal is wheat. In another embodiment, said cereal is maize. In yet another embodiment, said cereal is rice.

In the context of the present invention, a gene of the polyamine biosynthetic pathway is down-regulated or inhibited. Consequently, the production of the enzyme is decreased, as compared to a wild-type plant, and the eventual production of the polyamines is also decreased.

There are numerous ways, known in the art, to downregulate or inhibit a specific gene in a plant.

The inhibition of a gene involved in the biosynthesis pathway of polyamines can be obtained by any means known in the art. Thus, one can proceed with the insertional mutagenesis by inserting a transposable element or a T-DNA within the gene to be inhibited. One can also make a physical or chemical mutagenesis, in particular by the use of EMS, X-rays or ultraviolet.

The mutated plants are then screened for example by PCR, using primers located in the target gene. However, one can also use other methods of screening, such as Southern Blots or screening through the AIMS method described in WO9927085 (to detect insertions), using probes which are specific of the target genes. One can also use methods for detecting point mutations or small insertions / deletions with specific endonucleases (such as Ce/I, or Endo\) as described in WO2006010646. In another embodiment, the inhibition or down-regulation is achieved by transforming the plant with a vector containing a sense or antisense construct. Both methods are known to allow inhibition or down-regulation of a target gene. It also uses the method of RNA interference (RNAi), which is particularly effective for the inhibition of genes in plants. This method is well known to the skilled person and consists of the transformation of the plant with a construct producing after transcription, a double duplex RNA, one of the strands of which being complementary to the mRNA of the gene target.

Genetic engineering approaches to altering, in particular specifically reducing, gene activity in plants such as wheat are well known in the art. These methods include the introduction of gene constructs for expression of a suitable antisense molecule that is complementary to the RNA of the target gene and can hybridize with it. Antisense molecules are thought to interfere with the translation or processing or stability of the RNA of the target gene, thereby inactivating expression of the gene. Methods of devising antisense sequences are well known in the art and examples of these can be found in US5190131 , EP467349, EP223399 and EP240208, which are incorporated herein by reference. The use of antisense methods in plants has been reviewed by Bourque (1995, Plant Science 105,125-149) and Senior (1998, Biotechnology and Genetic Engineering Reviews 15,79-1 19). Bourque lists a large number of examples of gene inactivation using antisense sequences in plant systems. She also states that attaining 100% inhibition of an enzyme activity may not be necessary as partial inhibition will more than likely result in measurable change in the system. Senior states that antisense methods are now a very well established technique for manipulating gene expression in plants.

Antisense molecules for ODC, ADC or other polyamine biosynthesis or modification genes can be based on the sequences given therein or derived from homologous DNA or mRNA sequences obtained from other species, for example barley. The antisense sequences may correspond to all or part of the transcripts of any of these genes or for sequences that affect control over their expression, for example their splicing. The antisense sequence may correspond to the targeted coding region of the target gene, or the 5'-untranslated region (UTR) or the 3'-UTR or combination of these. It may be complementary in part to intron sequences, which may be spliced out during or after transcription, preferably only to exon sequences of the target gene. In view of the generally greater divergence of the UTRs, targeting these regions provides greater specificity of gene inhibition. In particular embodiments, the length of the antisense sequence is at least 19 contiguous nucleotides, at least 50, at least 100, at least 200, at least 500 or at least 1000 nucleotides corresponding to the complement of the gene RNA sequence. The full-length sequence complementary to the entire gene transcript may be used. In a particular embodiment, the length of the antisense sequence is 100-2000 nucleotides. In further embodiments, the degree of sequence identity of the antisense sequence to the complement of the targeted transcript is at least 85%, at least 90% or 95-100%. The antisense RNA molecule may of course comprise unrelated sequences which may function to stabilize the molecule.

Another molecular biological approach that may be used is co-suppression. The mechanism of co-suppression is not well understood but is thought to involve post- transcriptional gene silencing (PTGS) and in that regard may be very similar to many examples of antisense suppression. It involves introducing an extra copy of a gene or a fragment thereof into a plant in the sense orientation with respect to a promoter for its expression. The size of the sense fragment, its correspondence to target gene regions, and its degree of sequence identity to the target gene are as for the antisense sequences described above. In some instances the additional copy of the gene sequence interferes with the expression of the target plant gene. Reference is made to WO9720936 and EP465572 for methods of implementing co- suppression approaches.

A further method that might be employed to introduce genetic variation into the wheat plant is duplex or double stranded RNA mediated gene silencing (RNAi). In this method a DNA is introduced that directs the synthesis of an at least partly double stranded RNA product(s) with homology to the target gene to be inactivated. The DNA therefore comprises both sense and antisense sequences that, when transcribed into RNA, can hybridize to form the double-stranded RNA region. In a preferred embodiment, the sense and antisense sequences are separated by a spacer region that comprises an intron which, when transcribed into RNA, is spliced out. This arrangement has been shown to result in a higher efficiency of gene silencing. The double-stranded region may comprise one or two RNA molecules, transcribed from either one DNA region or two. The presence of the double stranded molecule triggers a response from an endogenous plant system that destroys both the double stranded RNA and also the homologous RNA transcript from the target plant gene, efficiently reducing or eliminating the activity of the target gene. Reference is made to AU99/29514 and WO9953050 for methods of implementing this technique. In particular embodiments, the length of the sense and antisense sequences that hybridise are at least 19 contiguous nucleotides, at least 30, at least 50, at least 100, at least 200, at least 500 or at least 1000 nucleotides. The full length sequence corresponding to the entire gene transcript may be used. In a particular embodiment, the lengths are in the range 100-2000 nucleotides. In further embodiments, the degree of sequence identity of the sense and antisense sequences to the targeted transcript is at least 85%, at least 90% or 95-100%. The RNA molecule may of course comprise unrelated sequences which may function to stabilize the molecule. The RNA molecule may be expressed under the control of a RNA polymerase Il or RNA polymerase III promoter.

The double-stranded RNA molecule may also comprise sequences from more than one gene, joined together, and thereby target multiple genes.

This RNAi method is particularly well suited in wheat. Indeed, due to the hexaploid nature of wheat, it may thus be difficult to identify mutants in all three genomes for a specific gene. The RNAi technology makes it possible to design a construct that would be directed against a region of the target gene, that is both specific of the target gene (for example ADC or ODC), and similar among the copies of the gene on each genome. Alternatively, as indicated above, a RNAi construct containing sequences from all three genes of the genomes can also be used, which makes it possible to inhibit the expression of these genes while transforming the plant with only one construct. The RNAi method is also well suited for other plants such as maize, barley in rice. Indeed, when multiple genes coding for an enzyme are present in the genome of these plants (for example on different chromosomes), or only an EST (expressed sequence tag) is available to the person skilled in the art, this person can easily design an RNAi vector that will inhibit said gene. Indeed, as indicated above, there is no need of the full sequence of a gene to be able to successfully inhibit it with the RNAi method. Furthermore, due to the proximity of some of the cereal genomes (rice and maize, wheat and barley, for example), it may also be possible to successfully inhibit a gene in an organism, while using a RNAi vector comprising a sequence from another organism.

Note that ESTs can routinely be obtained from a given organism by using sequence information from another organism, an alignment program such as BLAST and subprograms blastn or tblastx, and the information present in databases (such as GeneBank, EMBL or TIGR).

There is thus no need to have the full sequences of the genes for each organism to be able to practice the invention (inhibit the expression of the genes) in different organisms.

The genetic variation responsible for the desired inactivation of gene expression in wheat may comprise a nucleic acid molecule encoding one or more ribozymes. Ribozymes are RNA molecules with enzymatic or catalytic function that can cleave other RNA molecules at specific sites defined by one or often two hybridizing sequences. The cleavage of the RNA inactivates the expression of the target gene. The ribozymes may also act as an antisense molecule, which may contribute to the gene inactivation. The ribozymes contain one or more catalytic domains, preferably of the hammerhead or hairpin type, between the hybridizing sequences. Other ribozyme motifs may be used including RNAseP, Group I or Il introns, and hepatitis delta virus types. Reference is made to EP 321201 and US 6,221 ,661. The use of ribozymes to inactivate genes in transgenic plants has been demonstrated.

Another method that might be employed to alter in particular and specifically to reduce gene activity in plants such as wheat comprises use a transcription factor. The inhibition or down-regulation is achieved by transforming the plant with a vector containing a sense or antisense construct including a transcription factor that would repress the activity of the targeted genes.

The invention also provides isolated nucleic acid molecules comprising RNA or

DNA, preferably DNA, which encode the gene-inhibiting molecule. In certain embodiments, the nucleic acid molecules encode antisense, sense (co- suppression), double-stranded RNA or ribozyme molecules which target the polyamine biosynthetic gene sequence and which inactivate its expression in the plant. The invention also provides genetic constructs comprising or encoding the isolated nucleic acid molecule, comprising one or more regulatory elements such as promoters, enhancers and transcription termination or polyadenylation sequences. Such elements are well known in the art. The genetic constructs may also comprise intron sequences that aid expression of the transgene in plants, particularly in monocotyledonous plants such as wheat. The term "intron" is used in its normal sense as meaning a genetic segment that is transcribed but does not encode protein and which is spliced out of an RNA before translation. lntrons may be incorporated in a 5'-UTR or a coding region if the transgene encodes a translated product, or anywhere in the transcribed region if it does not.

The invention further provides vectors, for example plasmid vectors, comprising the genetic constructs. The term "vector" includes an expression vector, being capable of in vitro or in vivo expression, and a transformation vector, capable of being transferred from one cell or organism to another. The vectors comprise sequences that provide for replication in cells, for example in prokaryotic cells such as E.coli or Agrobacterium. In a particular embodiment, the vector is a binary vector comprising a T-DNA sequence, defined by at least one T-DNA border sequence that can be introduced into the plant cells.

The invention further provides cells comprising the vectors, for example Agrobacterium or plant cells which may be regenerable cells such as the cells of the scutellum of immature embryos. Alternatively, the cells may be transformed plant cells comprising the transgene.

In a specific embodiment, said genetic construct comprises a promoter that is induced by the pathogen infection. Using such a promoter would ensure that the construct inducing the downregulation / inhibition of the target gene would only be translated after infection by the pathogen, therefore allowing the plant to produce and use the polyamine during its normal growth. As demonstrated below, the genes of the polyamine biosynthetic pathway are induced upon Fusarium infection. Their native promoters are therefore good candidates for such a purpose. Alternatively, those derived from known pathogen inducible genes of cereals described elsewhere (Golkari et al., 2007, Plant Biotechnol J. 5(1 ):38-49; Kong et al., 2007, Genome 50: 1038-1048; Boddu et al., 2006, MoI. Plant-Microbe Interact. 19: 407-417; Pritsch et al., 2000, MoI. Plant-Microbe Interact. 13: 159-169; Pritsch et al., 2001 , Physiol. MoI. Plant Pathol. 58: 1-12; Bernardo et al., 2007, Functional & Integrative Genomics 7: 69-77) may be used. Other examples of pathogen- inducible promoters include those from pathogenesis-related proteins (PR proteins), which are induced following infection by a pathogen; e.g., PR proteins, SAR proteins, beta-1 ,3-glucanase, chitinase, and the like (Redolfi et al. (1983) Neth. J. Plant Pathol. 89:245-254; Uknes et al. (1992) Plant Cell 4:645-656; Van Loon (1985) Plant MoI. Virol. 4:11 1-116; WO9943819). Of particular interest is the inducible promoter for the maize An2 gene whose expression is induced by the pathogen Fusarium graminearum (see Harris et al., Plant Molecular Biology (2005) 59:881-894), the inducible promoter for the maize PR1 gene whose expression is induced by many pathogens including Fusarium sp (US20010025380 and Cordero et al., (1992) Physiological and Molecular Plant pathology 41 :189-200). One also can also cite the AoPRT-L promoter described in WO9966057.

Interesting promoters are also the ones that have capacity to be induced locally at or near the site of pathogen infection (Che et al., (1996) Plant Journal 10:955:966; Warner et al., (1993) Plant Journal 3:191-201 ; Marineau et al. (1987) Plant MoI. Biol. 9:335-342; Matton et al. (1989) Molecular Plant-Microbe Interactions 2:325-331 ; Somsisch et al. (1986) Proc. Natl. Acad. Sci. USA 83:2427- 2430; Somsisch et al. (1988) MoI. Gen. Genet. 2:93-98; Yang (1996) Proc. Natl. Acad. Sci. USA 93:14972-14977. Zhang et al. (1994) Proc. Natl. Acad. Sci. USA 91 :2507-251 1 ; Siebertz ef a/. (1989) Plant Cell 1 :961-968; US 5750386).

A wound-inducible promoter may also be considered in the context of the invention, as pathogens may find entry into plants through wounds or insect damage. Such wound-inducible promoters include promoters of the potato proteinase inhibitor pin Il gene (Ryan (1990) Ann. Rev. Phytopath. 28:425-449; Duan et al. (1996) Nature Biotechnology 14:494-498), wun1 and wun2 (US 5428148), win1 and win2 (Stanford et al. (1989) MoI. Gen. Genet. 215:200-208), systemin (McGurl et al. (1992) Science 225:1570-1573), WIP1 (Rohmeier et al. (1993) Plant MoI. Biol. 22:783-792; Eckelkamp et al. (1993) FEBS Letters 323:73- 76), and MPI gene (Corderok et al. (1994) Plant J. 6(2):141-150).

In another embodiment, said genetic construct comprises a promoter that is inducible by a chemical (for review, see Moore et al., 2006, Plant J. 45:651-83,

Padidam M. 2003, Curr Opin Plant Biol. Apr;6(2): 169-77, Wang et al., 2003,

Transgenic Res. 2003 Oct;12(5):529-40 and Zuo and Chua 2000, Curr Opin

Biotechnol. Apr; 1 1 (2): 146-51 ). Some examples of couples of chemically inducible systems and chemical inducer used in plants are, the alcA promoter from A. nidulans, inducible by ethanol (Roslan et al., 2001 , Plant J. Oct;28(2):225-35) or the ecdysone receptor from C. fumiferana, inducible by an ecdysone agonist (Koo et al., 2004, Plant J. Feb;37(3):439-48). In this case, said chemical shall be applied to the plant upon Fusarium infection. This would induce inhibition and/or downregulation of the targeted genes for the polyamine biosynthesis and hence, control of the pathogen infection. In another embodiment, the promoter is tissue specific and particularly induces expression in the rachis, spikelets and/or stem cavity.

In a preferred embodiment, said cereal is rice and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 1 , SEQ ID N° 2 (these two sequences being ADC), SEQ ID N° 14, SEQ ID N° 15, SEQ ID N° 16 (these three sequences being ODC), SEQ ID N° 23 and SEQ ID N° 24 (these two sequences being ASS).

In another preferred embodiment, said cereal is wheat and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 3, SEQ ID N° 4, SEQ ID N° 5, SEQ ID N° 6, SEQ ID N° 7, SEQ ID N° 8, SEQ ID N° 9, SEQ ID N° 10 (these eight sequences being ADC), SEQ ID N° 17, SEQ ID N° 18, SEQ ID N° 19, SEQ ID N° 20, SEQ ID N° 21 (these five sequences being ODC), SEQ ID N° 25, SEQ ID N° 26 and SEQ ID N° 27 (these three sequences being ASS). In another preferred embodiment, said cereal is barley and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 1 1 , SEQ ID N° 12, SEQ ID N° 13 (these three sequences being ADC), SEQ ID N° 22 (this sequence being ODC), and SEQ ID N° 28 (this sequence being ASS).

In another preferred embodiment, said cereal is maize and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 29, SEQ ID N° 55, SEQ ID N° 56, SEQ ID N° 57 (these four sequences being ODC), SEQ ID N° 58, SEQ ID N° 59, SEQ ID N° 60, SEQ ID N° 61 , SEQ ID N° 62, SEQ ID N° 63, SEQ ID N° 64, SEQ ID N° 65 (these eight sequences being ADC), SEQ ID N° 66, SEQ ID N° 67, SEQ ID N° 68, SEQ ID N° 69, SEQ ID N° 70, and SEQ ID N° 71 (these six sequences being ASS).

The invention also relates to a plant preferably a cereal presenting an inhibition or downregulation of the expression of an endogenous polyamine synthesis gene.

In a first embodiment, said cereal is rice and said endogenous polyamine synthesis gene is an ADC chosen among SEQ ID N° 1 and SEQ ID N° 2.In another embodiment, said cereal is rice and said endogenous polyamine synthesis gene is an ODC chosen among SEQ ID N° 14, SEQ ID N° 15 and SEQ ID N° 16.

In another embodiment, said cereal is rice and said endogenous polyamine synthesis gene is an ASS chosen among SEQ ID N° 23 and SEQ ID N° 24. In another embodiment, said cereal is wheat and said endogenous polyamine synthesis gene is an ADC chosen among EQ ID N° 3, SEQ ID N° 4 SEQ ID N° 5, SEQ ID N° 6, SEQ ID N° 7, SEQ ID N° 8, SEQ ID N° 9 and SEQ ID N° 10.

In another embodiment, said cereal is wheat and said endogenous polyamine synthesis gene is an ODC chosen among SEQ ID N° 17, SEQ ID N° 18, SEQ ID N° 19, SEQ ID N⁰ 20 and SEQ ID N⁰ 21.

In another embodiment, said cereal is wheat and said endogenous polyamine synthesis gene is an ASS chosen among SEQ ID N° 25, SEQ ID N° 26 and SEQ ID N° 27.

In a first embodiment, said cereal is barley and said endogenous polyamine synthesis gene is an ADC chosen among SEQ ID N° 11 , SEQ ID N° 12 and SEQ ID N° 13.

In another embodiment, said cereal is barley and said endogenous polyamine synthesis gene is an ODC and is represented by SEQ ID N° 22.

In another embodiment, said cereal is barley and said endogenous polyamine synthesis gene is an ASS represented by SEQ ID N° 28.

In a first embodiment, said cereal is maize and said endogenous polyamine synthesis gene is an ADC chosen among SEQ ID N° 58, SEQ ID N° 59, SEQ ID N° 60, SEQ ID N° 61 , SEQ ID N° 62, SEQ ID N° 63, SEQ ID N° 64 and SEQ ID N° 65.

In another embodiment, said cereal is maize and said endogenous polyamine synthesis gene is an ODC chosen among SEQ ID N° 29, SEQ ID N° 55, SEQ ID N° 56 and SEQ ID N⁰ 57.

In another embodiment, said cereal is maize and said endogenous polyamine synthesis gene is an ASS chosen among SEQ ID N° 66, SEQ ID N° 67, SEQ ID N° 68, SEQ ID N° 69, SEQ ID N° 70, and SEQ ID N° 71.

Inhibition or downregulation of an endogenous polyamine synthesis gene may be achieved by the methods described above. It is preferred when said plant is such that the inhibition or downregulation of the endogenous polyamine synthesis gene is not permanent but may be induced by an external stimulus such as a pathogen infection or application of a chemical. In particular, the invention relates to a transgenic plant containing a transgene comprising a genetic construct encoding a gene-inhibiting molecule such as described above (RNAi, antisense, co-suppression, ribozyme, and the like). This gene-inhibiting molecule is under the control of a promoter which may be inducible as described above.

In a specific embodiment, the plant is such that inhibition or downregulation of the endogenous polyamine synthesis gene is induced by the pathogen infection and in particular by Fusarium infection.

In another specific embodiment, said inhibition or downregulation is restricted to the rachis, the spikelets and/or the stem cavity. This may be achieved by use of specific promoters of said tissues of the plant.

The invention also relates to a cell of a plant, preferably a cell of a plant, or a part of a plant. Spikelets and seeds and a cell of a plant are among preferred part of plants according to the invention. Preferably, said plant is a cereal.

Transgenic plants are obtained by traditional methods for genetic transformation of plants and can be obtained by any one of the techniques known to one skilled in the art : methods of direct transfer of genes such as direct microinjection into plant embryoids, vacuum infiltration (Bechtold et al. 1993, C R Acad Sci III. 316(10) : 1 194-1 199) or electroporation (Chupeau et al., 1989, Biotechnology vol.7: pp. 503-508) or the bombardment by gun of particules covered with the plasmidic DNA of interest (Fromm et al., 1990, Nature, 319(6056)791-3 ; Finer et al., 1992, Plant Cell Report 11 : 323-328). Agrobacterium mediated transformation methods may also be used Agrobacterium tumefaciens, in particular according to the method described in the article by An et al., (1986, Plant Physiology, 81 :86-91 ), or Agrobacterium rhizogenes, in particular according to the method described in the article by Guerche et al., (1987, Biochimie. Jun-Jul;69(6-7):621-8). According to a preferred mode, it is possible to use the method described by lshida et al., (1996, Nature Biotechnol. 14, 745-50) for the transformation of maize. According to another method, the wheat is transformed according to the method described in WO 00/63398.

In another embodiment, the invention relates to a method for reducing trichothecene production by a trichothecene-producing pathogen, in particular a Fusarium, comprising the step of limiting concentration of polyamines and/or polyamine amino acid precursors accessible to said pathogen, in particular said Fusarium during its growth. This method is particularly useful in vivo, and can also be performed in vitro, by using a growth medium deficient in said polyamines and/or polyamine amino acid precursors. It is to be noted that limiting the concentration of polyamines and/or polyamine amino acid precursors accessible to said pathogen, in particular said Fusarium, during its growth will also limit the ability of said pathogen, in particular said Fusarium, to grow if no other source of both carbon and nitrogen is available.

Indeed, the inventors have demonstrated that the presence of these compounds, and in particular putrescine and L-arginine is higher in infected plants, and that these compounds induce trichothecene production by the pathogens. Other interesting compounds are guanine, citrulline, ornithine, agmatine, cadaverine, N-amylamine, ethanolamine, β-phenylethylamine, tyramine, L- methionine, L-lysine and D-glucosamine.

Alternatively, the invention also relates to a method for inducing trichothecene production by a DON-producing pathogen, in particular a Fusarium, comprising the step of making polyamines and/or polyamine amino acid precursors accessible to said pathogen, in particular said Fusarium, during its growth. This method is useful in vitro, by adding polyamines and/or polyamine amino acid precursors in the growth medium. This makes it possible to improve the yield of mycotoxin production by the fungi.

When performed in vivo, limitation of the concentration of polyamines and/or polyamine amino acid precursors accessible to said pathogen, in particular said Fusarium, during its growth may be performed by growing said pathogen, in particular said Fusarium, on a plant presenting an inhibition or downregulation of the expression of an endogenous polyamine synthesis gene. Said plant is described above, and both the spread of said pathogen, in particular said Fusarium, and its ability to produce mycotoxins will be reduced when on said plant, as compared to when the pathogen is on a nearly isogenic plant which does not present said inhibition or down-regulation.

In another embodiment, one can perform the method of the invention in planta by applying an inhibitor of an enzyme of the polyamine biosynthetic pathway to said cereal, in the presence of said pathogen, in particular said Fusarium. One could use an inhibitor as described in US 43761 16, an inhibitor of ornithine decarboxylase, such as α-monofluoromethylornithine (DFMO), or an S- adenosylmethionine decarboxylase (SAMDC) inhibitor (Matters et al, Clin, and Exp. Metastasis, 22(4), 2005 , 331-339(9), Mehrotra et al, Contraception, 57(1 ), 1998, 55-60).

Reduction of the level of the available polyamines and/or polyamine amino acid precursors may also be obtained by the use of another non-pathogenic microorganism that would use said compounds competitively with said pathogen, in particular said Fusarium.

In another embodiment, it is possible to reduce the production of trichothecene by a DON-producing pathogen, in particular a Fusarium, by adjunction of a compound that would abolish the effect of said polyamines and polyamine amino acid precursors to said pathogen, in particular said Fusarium (induction of trichothecene production and increased growth). Such compounds may in particular be chosen among nitrate and cysteine.

In another embodiment, an inhibitor of a polyamine/amino acid transporter of a Fusarium is used.

FIGURES Figure 1 : Insertion of TRI5-GFP at the TRI5 locus. (A) Vector, target locus and predicted resultant genomic locus. (B) Southern analysis of transformants for site specific integration of the construct. Genomic DNA was cut with Hind\\\. The position of the probe used is shown in A. All transformants except 05T0002 show homologous integration at the TRI5 locus. Figure 2: In planta expression of GFP driven by the TRI5 promoter (transformant 05T0020). A, bright field of florets. The upper shown floret was inoculated. Hyphae can be seen emerging from the spikelet rachis junction (bar = 5 mm). B, bright field close up (bar = 0.5 mm) of boxed region in A, showing hyphae emerging from the rachis. C, fluorescent image of B, showing strong GFP expression in the hyphae emerging from the rachis (bar = 0.5 mm). D, mycelia in the pith cavity of a stem cross section about 5 cm below stem head junction (bar = 0.5 mm). E, hyphae at the very front on the infection about 5 cm below the stem head junction as the fungus goes down the stem expressing GFP (bar = 0.5 mm). A, B, C cultivar INDIA44 4 dpi. D, E cultivar Kennedy 14 dpi. Figure 3: A, DON production in selected nitrogen sources from the phenotype microarray plates with the wild type isolate CS3005. Toxin was measured after seven days culture. Data are the mean of three biological replicates. Error bars are standard errors of the mean. B, Total area under the expression curve for the same nitrogen sources as in A. The correlation coefficient (R2) between the DON concentration and total area under the expression curve was 0.66. Figure 4: T-2 toxin production by F. sporotrichioides in defined media containing either glutamine or agmatine as the nitrogen source. Cultures were grown for 7 days at 28°C in the dark.

Figure 5: Fluorescence from the TRI5-GFP isolate in cultures grown for 2 days in glutamine followed by either glutamine, arginine or agmatine addition to the media to a final equivalent concentration of 5 mM. Values are the mean relative fluorescence units (RFU) of twelve biological replicates, with error bars representing the standard error of the mean.

Figure 6: TRI5 gene expression relative to fungal 18S rRNA in culture 4 days post- inoculation (dpi) in glutamine or agmatine as nitrogen sources and in planta. In planta expression is 4 dpi (the maximum level observed in a seven day time course - data not shown). Error bars represent the standard error of the mean of four biological replicates.

Figure 7: A, B and C are independent experiments conducted at different times. A, data for gene expression in whole heads at 5 days post-inoculation. B, data for gene expression in the rachis and the spikelets separately at 5 days post- inoculation. C, Time course induction of ODC and ADC2 gene expression during infection of wheat by F. graminearum. Data in A and B is the mean of eight biological reps for both mock and infected samples. Data in C is the mean of four biological reps for both mock and infected samples. Error bars are the standard error of the mean.

Figure 8: (A) Induction of free amino acid and polyamine inducers during Fusarium Head Blight disease of wheat (MET = methionine, ARG = arginine, LYS = lysine). (B) Increases in both deoxynivalenol (DON) and fungal biomass during Fusarium Head Blight disease of wheat in same experiment as described in (A). All data points consist of four biological replicates with error bars representing the standard error of the mean.

Figure 9: plasmid pBIOS 1577. a partial sequence of the wheat ADC gene (ADC2 partial) has been introduced in sense and antisense orientation in the vector, separated by the rice tubulin intron (TUBL int), under the control of the Actin promoter and intron (proActin and Actin int). Figure 10: plasmid pBIOS 1589. a partial sequence of the wheat ODC gene [TaODC partial) has been introduced in sense and antisense orientation in the vector, separated by the rice tubulin intron (OsTUBL int), under the control of the Actin promoter and intron (proActin and Actin int).

EXAMPLES

The invention is based initially on the identification of compounds that can induce the expression of the TRI5 gene of Fusarium graminearum. This gene encodes the trichodiene synthase enzyme that catalyses the conversion of farnesyl pyrophosphate to trichodiene and this is the first committed step from primary metabolism to the secondary metabolism pathway of trichothecene mycotoxin synthesis. TRI5 is conserved in all trichothecene producing Fusarium spp. including F. graminearum, F. sporotrichioides, F. poae, F. sambucinum, F. venenatum as well as Myrothecium roridum (Fekete et al., 1997, Mycopathologia. 138, 91-97; Hohn and Desjardins, 1992, MoI Plant Microbe Interact. 5, 249-56; Proctor et al., 1995, MoI. Plant Microbe Interact. 8, 593-601 ; Royer et al., 1999, Fungal Genetics and Biology. 28, 68-78; Trapp et al., 1998, Molecular and General Genetics. 257, 421-432). Its expression is regulated by the pathway specific transcription factor TRI6 (Proctor et al., 1995, Applied and Environmental Microbiology. 61 , 1923-1930) and regulatory protein TRHO (Peplow et al., 2003, Applied and Environmental Microbiology. 69, 2731-2736). The TRI6 binding sites are conserved upstream of all TRI5 homologous genes in Fusaria. To create an easy assay for the expression of this gene the regulatory promoter region of the TRI5 gene was fused to the gene encoding the jellyfish green fluorescent protein (GFP) and this new chimerical gene introduced into F. graminearum. The transformed fungal strains showed fluorescence during the infection of wheat heads at times when the trichothecene mycotoxins are known to be produced. Thus these new transgenic strains appear to act as 'reporter strains' for TRI5 gene expression and toxin production and were subsequently used to identify compounds that induce toxin production and characterize the mechanisms of action of production of these mycotoxins.

EXAMPLE 1 : experimental conditions Construction of reporter strains The strategy adopted was to first develop transgenic strains of F. graminearum (Fg) that contained the reporter GFP gene regulated by the TRI5 promoter (promoter of the gene encoding the Fg trichodiene synthase, said gene determining the first committed step in the biosynthesis of trichothecene mycotoxins) as tools to identify conditions that would induce gene expression necessary for toxin production in the fungus. Vector construction

The TRI5 promoter-GFP fusion vector was constructed by overlap PCR using primers to the TRI5 promoter (5'- GGCCTCGAGGCTCACGGGCTACAGTGAAT-S', SEQ ID N° 30 and 5'- TGCTCACCATGATGGCAAGGTTGTACTGGT-3', SEQ ID N° 31 ) and eGFP (5'- TTGCCATCATGGTGAGCAAGGGCGAGGAG-3', SEQ ID N° 32 and 5'- GGCCTCGAGCGCCTTAAGATACATTGATGAGTTT-3', SEQ ID N° 33). The PCR construct consisted of 1.5 kb of TRI5 promoter amplified from F. graminearum CS3005, followed by eGFP and the SV40 terminator amplified from peGFP-N1 (Clontech). Individual fragments were amplified with Phusion DNA polymerase (Finnzymes), gel purified, then fused using the same polymerase with the two external primers. The PCR product was digested with Xho\ and ligated into the Sal\ site of pPZPhygHindX containing the hygromycin phosphotransferase gene under the Aspergillus nidulans TrpC promoter and terminator (Elliott and Howlett, 2006, MoI Plant Microbe Interact. 19, 588-96).

Fungal transformation

The F. graminearum isolate CS3005 was transformed with the TRI5-GFP vector (Figure 1 ).

Fungal protoplasts were generated from approximately 1 x10⁷ conidia germinated overnight in 200 ml. of TB3 (0.3% yeast extract, 0.3% casamino acids, and 20% sucrose) media at 20⁰C with shaking at 150 rpm. Germlings were collected on miracloth and washed with 1 M sorbitol. Germlings were resuspended in 20 ml. of 1 M sorbitol containing 5 U chitinase (Sigma), 500 mg Driselase (Sigma) and 200 mg lysing enzymes (Sigma), and incubated 1-2 hours at 28°C to release protoplasts. Protoplasts were collected by centrifugation at 4°C, washed three times with ice cold STC (20% sucrose, 50 mM Tris/HCI pH 8, 50 mM CaCI2). To a 200 μL aliquot of STC containing 2x10⁷ protoplasts 10 μg circular plasmid DNA was added, incubated for 20 minutes at room temperature, then 1 mL of 40% PEG4000, 10 mM Tris HCI pH 8, 50 mM CaCI2 was added, incubated for 20 minutes at room temperature and then 5 ml. of TB3 media was added and the tubes incubated overnight at room temperature with gentle agitation. 3 ml. of the transformation mixture was embedded in 30 ml. regeneration media (0.2% yeast extract, 0.2% casamino acids, 0.55% low melt agarose, 27% sucrose) and set in 14 cm Petri dishes, incubated overnight and then overlaid with 30 ml. of regeneration media containing hygromycin at 400 μg ml.^"1. After 7-14 days of growth at room temperature in the dark, transformants were transferred to individual plates containing hygromycin (200 μg ml.^"1), allowed to grow, single spored then checked for insertion of the construct by PCR and Southern hybridisation. The hybridisation probe was amplified using primers binding in the promoter region used for the reporter construct (δ'-AGGTTTCAACCTCCGAGGAA- 3', SEQ ID N° 34, δ'-TATCTGGCATGCCTGTCATC-S', SEQ ID N° 35).

Plant growth and inoculation Seeds of the susceptible bread wheat cultivars Kennedy and INDIA 44 were grown in controlled environment rooms with photoperiod length of 14 h, 25°C and 50% relative humidity at 500 mmol m-2 s. Night-time temperature was set at 15°C, with 90% relative humidity. Three seeds were sown per pot (14 cm diameter, 17 cm height) containing steamed soil (50% peat and 50% sand by volume). Heads were spray inoculated at mid anthesis with a conidial suspension at 1 x10⁶ spores per ml_-1. Water was used for the mock inoculation. After spraying heads were covered with plastic bags containing a small volume of water to assist in humidification. After two days plastic bags were replaced with glassine paper bags.

Reporter characterisation

Transformants comprising the reporter strain containing the eGFP driven by the TRI5 promoter have been screened on wheat heads for GFP expression and a number of transformants showed excellent fluorescence in planta.

No fluorescence was observable in plate culture. Temporally and spatially restricted fluorescence was observed with transformants in grain cultures. In planta GFP was visible as earlier as two days post inoculation (dpi). Fluorescence was still visible 10 dpi but was greater at earlier time points (3-6 dpi). There also appears to be some spatial restriction to GFP expression, with the greatest signal observed from the area surrounding the spikelet node.

Toxin assays Deoxynivalenol (DON) was measured using the DON Plate Kit ELISA from Beacon Analytical Systems according to the manufacturer's instructions, except an additional low concentration standard was used (0.02 ppm). Culture filtrates were initially diluted 1 :25 giving a detection range of 0.5-12.5 ppm. Further sequential 1 :10 dilutions were made for samples outside of this range as required.

T-2 toxin was measured using the Veratox T-2 ELISA kit from Neogen (Lansing, Ml, USA) according to the manufacturer's instructions, except an additional low concentration standard was used (0.0025 ppm). Culture filtrates were initially diluted 1 :50 into 50% methanol to give a detection range of 0.125-2.5. Further dilutions were made as for DON detection.

Microscopy and fluorometry

Hand sections and whole tissue mounts were examined using a Leica

Microsystems MZ16FA fluorescent stereomicroscope (Wetzlar) fitted with GFP plus filter set. GFP fluorescence was quantified in microtitre plates using a Fluoroskan

Ascent FL microplate fluorometer (Thermo Scientific) using excitation at 485 nm and emission at 520 nm.

Gene expression analysis Plant gene expression was quantified using reverse transcriptase- quantitative polymerase chain reaction (RT-qPCR), as described by (Mudge et al., 2006, Physiological and Molecular Plant Pathology. 69, 73-85). Primers are listed in Table I.

Table I: Primers used for RT-qPCR and the accessions of genes used to design the primers. Accessions are from GenBank (PR5 and Peroxidase) or from the DFCI Wheat Gene lndicies database (http://compbio.dfci.harvard.edu/tgi/cgi- bin/tgi/gimain.pl?gudb=wheat)

Amino acid and polyamine measurements

Free amino acids were extracted from individual wheat heads by grinding the heads under liquid nitrogen and resuspending 500 mg of the powder in 2 ml. of methanol, incubated at -20⁰C overnight and then 8 ml. of water added and centrifuged to pellet solid matter. 500 μl_ of supernatant was dried down under vacuum. This extract was used with the Waters (Milford, MA, USA) AccQ-tag amino acid detection kit, with quantification by HPLC according to the manufacturer's instructions. For polyamine analysis wheat tissue is ground to fine powder in liquid nitrogen, extracted in perchloric acid, centrifuged and the supernatant mixed with a saturated solution of sodium carbonate and then dansyl chloride. Dansylated polyamines were extracted in toluene, the organic phase is collected, dried and re-dissolved in acetonitrile; 20 μl are injected into the HPLC as described by Marce et al. (Journal of Chromatography B 666 (1995) 329-335). Example 2: Construction and testing of the TRI5-GFP isolate of Fusarium graminearum

The F. graminearum isolate CS3005 was transformed with the TRI5-GFP vector as described above (Figure 1.A) and 7 putative transformants were selected for further analysis. Six of these putative transformants showed integration at the

TRI5 locus when tested by Southern blot analysis (Figure 1.B) and these were used in further experiments.

The TRI5-GFP transformants showed no GFP fluorescence during culture on PDA while in autoclaved wheat grain culture they showed diffuse GFP throughout the cultures and strong GFP expression in the hyphae surrounding a limited number of grains. Selected transformants were tested in point inoculations of wheat cv Kennedy heads. Transformants 05T0005, 05T0017 and 05T0020 were consistent in their expression of GFP during infection. GFP expression in planta was strongest at sites of re-emergence from the rachis and during colonisation of the pith cavity of the stem (Figure 2). Weak fluorescence could also be seen in the hyphae collaring the spikelet-rachis junction, although these were typically partially obscured by non-fluorescent hyphae. The induction of TRI5 expression during infection and the infection patterns of the TRI5-GFP transformants generated in this study were consistent with previously published observations on the spatial location of trichothecene action (Jansen et al., 2005, PNAS. 102, 16892-16897) indicating that the TRI5-GFP strains may be used as tools for high-throughput screening of inducers of TRI5 and subsequently DON biosynthesis.

Example 3: Nutrient profiling for the ability of F. graminearum to grow in various carbon and nitrogen sources

The ability of F. graminearum to grow in a variety of carbon and nitrogen sources was assessed using the phenotype microarray technology. Amongst the carbon sources D Fructose allowed the most growth but a number of other carbon sources, particularly the simple sugars, supported good growth. Similarly with nitrogen source profiling, F. graminearum was able to use a number of different compounds as sole nitrogen sources, including some that may be biochemically difficult to access in nature such as D enantiomers of amino acids. Both the Australian CS3005 and American Ph-1 strains of F. graminearum were assayed for their ability to access nitrogen sources, and although generally these isolates were similar in their growth across the plate, some differences were observed between these strains. Most notable was the considerably poorer ability of the Australian CS3005 isolate to grow on biuret and methylamine. Neither isolate could grow on D-lysine, Hydroxylamine, ethylenediamine, N-acetyl-D-galactosamine, N-acetyl-D- mannosamine, thymine, thymidine, uracil, uridine or alloxan as a sole nitrogen source.

Example 4: Nutrient profiling for induction of DON production in F. graminearum

Using the TRI5-GFP reporter strain (05T0017) and parent (CS3005) as a control, the role of carbon and nitrogen sources were assessed for their ability to induce TRI5 promoter activity and GFP expression. To test carbon sources, the

TRI5-GFP transformant 05T0017 was grown in the PM1 and PM2 carbon source plates but no fluorescence was observed with any of the 190 carbon sources tested in these plates. Growth of the reporter strain in PM3 nitrogen source test plates revealed a key role for several specific nitrogen sources in regulating TRI5 expression. Even though many nitrogen sources supported growth of the fungus, most did not induce TRI5 expression. The exceptions were agmatine, putrescine, arginine and guanine, which promoted strong GFP fluorescence and to a lesser extent citruline, ornithine, N-amylamine, ethanolamine, β-phenylethylamine, tyramine and D-glucosamine. Expression of GFP seemed to appear after most of the growth of the fungus had occurred. For example in agmatine, fluorescence was maximal at 4 dpi, but at 3 dpi the well showed near maximal optical density and microscopically was essentially full of hyphae.

The culture filtrate for 30 selected nitrogen sources was assayed for DON content at the end of the seven days incubation. As shown in Figure 3, the production of toxin (Figure 3.A) correlated with the occurrence of GFP fluorescence (Figure 3.B). With some nitrogen sources, e.g. arginine, ornithine, putrescine and agmatine, the DON concentrations that were measured in the media were extremely high and either equal to, or greater than 1000 ppm. It should be noted the basal media used in the PM3 plates is very similar to that used in the widely used two stage DON inducing media (Miller and Blackwell, 1986, Canadian Journal of Botany. 64, 1-5) except for the nitrogen source. Well A2 containing ammonia as the nitrogen source is equivalent to the second stage media of Miller and Blackwell (1986, op. cit), but relatively small amounts of DON were present in this media (44±9 ppm) compared to other wells. It should also be noted that the fluorescence read out for TRI5 GFP expression is in some circumstances dependent on the growth and biomass of the fungus and may not truly represent the activity of the fungus and its potential for DON production. This is exemplified by cultures using methylamine as a nitrogen source where growth was very limited and only a low fluorescence signal was obtained, however, analysis of media showed toxin was still produced (109±9 ppm).

The role of agmatine in inducing trichothecene production appears to be a general phenomenon, as the production of T-2 toxin by F. spropotrichioides was also higher in agmatine (134±14 ppm, n=3) compared to glutamine (1.3±0.3 ppm, n=3) containing media after 7 days growth in culture (Figure 4).

To investigate whether TRI5-GFP could be induced by either arginine or agmatine in mature hyphae pre-grown on other nitrogen sources, the induction was analysed after initial growth in the non-inducing nitrogen source glutamine, followed by addition of glutamine, arginine or agmatine. As shown in Figure 5, induction by arginine and agmatine occurred within 24 hours of their application, but did not reach maximal levels until 48-72 hours after addition of the inducing nitrogen source. The addition of additional glutamine did not have any effect on TRI5-GFP expression.

Example 5: TRI5 gene expression level in response to agmatine is similar to that observed in planta

Although DON production was strongly induced in vitro by the identified compounds it is difficult to compare directly the levels of DON production in liquid culture to that in planta where the majority of the tissue is of plant origin. However, on a fresh weight basis DON levels in excess of 500 ppm in heavily infected heads have been observed (Mudge et al., 2006, Physiological and Molecular Plant

Pathology. 69, 73-85). Analysis of fungal gene expression relative to fungal specific reference gene in both culture and in planta provides a better comparison of fungal mycotoxin biosynthetic potential. To address this, RT qPCR was carried on TRI5 gene transcripts, relative to the fungal 18S rRNA. As shown in Figure 6, the level of

TRI5 gene expression, when the fungus was grown in agmatine as the nitrogen source was very similar to that observed during infection. Example 6: Abolition and synergism of TRI5 induction by nitrogen source combinations

The role of nitrogen in regulating the production of secondary metabolites is well documented in many other fungi. We were interested to see if there is a hierarchy of TRI5 activation when multiple nitrogen sources are present. Interactions between nitrogen sources and their effects on TRI5 transcript levels were investigated in media where two nitrogen sources were present during growth (Table II). Nitrate and cysteine present in the media at the time of inoculation completely inhibited the induction of TRI5-GFP by nitrogen sources e.g. agmatine, putrescine, that when supplied alone were highly inductive. Synergistic effects were also observed, particularly between methionine and agmatine. It should be noted that there was some concentration dependence observed with the inducing nitrogen sources. For example the agmatine/control combination (where agmatine was at a concentration of 5 mM) showed considerably less fluorescence than the agmatine/agmatine combination (final concentration 10 mM).

Table II: Interaction of selected nitrogen sources in induction of TRI5-GFP. Cultures were grown in the presence of two nitrogen sources for four days then the fluorescence measured. Data are the mean of four replicate cultures relative to the maximal level observed (Methionine/Agmatine, 100%) with the standard error of the mean shown in brackets. The control is no nitrogen source.

C A N Ag P G M Cy

Control (C) 0(0)

Ammonium (A) 0(1 ) 0(1 )

Nitrate (N) 0(0) 0(0) 0(0)

Agmatine (Ag) 5(1 ) 15(9) 0(1 ) 26(23)

Putrescine (P) 0(1 ) 9(6) 0(0) 69(11 ) 61(0)

Glutamine (G) 0(0) 0(0) 0(0) 47(4) 37(11 ) 0(0)

Methionine (M) 18(5) 24(6) 0(0) 100(18) 79(7) 71(5) 30(3)

Cysteine (Cy) 0(0) 0(0) 0(0) 2(6) 5(4) 0(0) 3(3) 0(0)

Comments about the data

The induction of TRI5 gene expression and DON production induced in these experiments is notable for two important reasons. Firstly the levels of TRI5 mRNA measured in the fungus when induced by strong inducers like agmatine and arginine (Figure 6) were equivalent to those observed in the fungus when growing in infected wheat plants indicating that these inducers either mimic the inducing factors present in infected wheat heads or they are the actual inducers present in infected grains. Secondly, the concentrations of DON and T2 toxin produced in the presence of inducers like agmatine (Figure 4) are extremely high and equivalent or greater than those observed in infected plant material again demonstrating that these inducing compounds either mimic or reproduced the inducing conditions in infected plants.

Example 7: Analysis of the expression of genes of polyamine biosynthesis pathway in wheat during infection

Analysis of the expression of genes of the polyamine biosynthesis pathway in wheat during infection (Figure 7) showed that genes encoding ornithine decarboxylase (E. C. 4.1.1.17), arginine decarboxylase (E. C. 4.1.1.19) and argininosuccinate synthase (E. C. 6.3.4.5) were the most strongly induced in expression during infection (Figure 7). This indicates that the levels of these enzymes and the levels of synthesis of arginine and polyamines are likely to increase during the infection of wheat and that these may in turn induce trichothecene production and particularly DON production in the infecting Fusarium fungus. To verify that metabolites capable of inducing trichothecene toxins are also induced following challenge of wheat by F. graminearum infected heads were sampled to see if the strong inducers were increased. Following inoculation, there were substantial increases in the concentrations of the polyamines, putrescine and cadaverine as well as the amino acids arginine, methionine and lysine at times that coincided with DON induction and fungal biomass increases (Figure 8.A and B).

Example 8: Characterization of Wheat ADC2 and ODC sequences.

Putative function (text) searches of both rice and Arabidopsis genomes were performed to identify putative ADC genes. Arabidopsis contains two ADC sequences [AtADCI, AT2G16500 and AtADC2, AT4G34710) and three loci were identified in rice (Os04g01690, Os06g04070 and Os08g33620) with two of these known to be expressed (Os04g01690 and Os06g04070). These genomics sequences were used to find wheat orthologous sequences, by a blast to a wheat EST databank. Two groups of sequences were identified encoding for the wheat equivalents of ADCI and ADC2. Five wheat EST (EMBLCK1631 10 ; EMBLCD8420; EMBLCD930750 and two private sequences) with significant homologies to the rice ADC2 were identified and clustered on a unique sequence of 1351 nt (SEQ ID N⁰ 7).

This sequence can be transcribed to the amino acid sequence depicted in SEQ ID N° 52.

Two wheat EST (EMBLBF474072 and EMBLBI479919) were used to determine the ADC1 wheat sequence. The alignment between the wheat contigs for ADC1 and ADC2 shows only poor homology.

A 450 nt fragment of the AtADC2 contig was choosen for designing RNAi constructs (SEQ ID N° 53). This sequence is divergent enough to ADC1 to allow the specific inhibition of the wheat ADC2 gene on the three genomes by the RNAi construct.

Three rice ODC sequences were identified with putative function searches in the rice genome. These were located on chromosome 9 (Os09g37120.1 ), on chromosome 2 (Os02g28110.1 ) and on chromosome 4 (os04g04980.1 ).

These three sequences were used to find wheat homologous sequences. As most of the wheat sequences identified by these different alignments were similar, it was concluded that the ODC sequence is highly conserved in wheat. One of the sequences (EMBLCN012161 , SEQ ID N° 17) was identified to be homologous to the three rice sequences; this sequence is issued from a Fusarium infected library. It was chosen to define the ODC fragment sequence to be use for RNAi (SEQ ID N° 54). The RNAi construct including this fragment may inhibit the expression of the three wheat ODC paralogues.

Example 9: Wheat RNAi transformation constructs for ADC and ODC :

The synthetic fragment of ADC2 was cloned in a pUC57 vector (GenBank accession Y14837), leading to the pUC57 TaADC2 vector. The BamH\ I Sal\ fragment of 451 nt (TaADC2) from pUC57-ADC2 was then introduced in the pENTRIA vector (Invitrogen) linearised with BambW I Xho\, to create the pBIOS1561.

An LR clonase reaction between the pBIOS1561 and the pBIOS1560, allows the creation of pBIOS 1577 (Figure 9). pBIOS1560 is a vector used to create RNAi vectors. A PCR reaction on pBIOS1577 controlled that the tubulin intron is in the right orientation to be properly spliced. The binary plasmid pBIOS1577, was then introduced in the A. tumefaciens hypervirulent strain EHA105, and used for transformation experiments. The synthetic fragment of ODC was cloned in a pUC57 vector, leading to the pUC TaODC vector. The Xba\ I EcolCRI fragment of 450 nt from pUC57-ODC was introduced into the pENTRIA vector opened by Xba\ I Xmn\ to create the pBIOS1588. In a similar way an LR clonase reaction between pBIOS1588 and pBIOS1560, led to the binary vector pBIOS1589 (Figure 10) to be introduced in the EHA105 Agrobacterium strain for transformation experiments.

Example 10: Wheat transformation protocol The method is essentially similar to the one described in WO 00/63398. Wheat tillers, approximately 14 days post-anthesis (embryos approximately 1 mm in length) were harvested from glasshouse grown plants to include 50cm tiller stem, (22/15°C day/night temperature, with supplemented light to give a 16 hour day). All leaves were then removed except the flag leaf and the flag leaf was cleaned to remove contaminating fungal spores. The glumes of each spikelet and the lemma from the first two florets were then carefully removed to expose the immature seed. Only these two seed in each spikelet were generally uncovered. This procedure was carried out along the entire length of the inflorescence. The ears were then sprayed with 70% IMS as a brief surface sterilization. Agrobacterium tumefaciens strains containing the vector for transformation were grown on solidified YEP media with 20mg/l kanamycin sulphate at 27°C for 2 days. Bacteria were then collected and re-suspended in TSIM1 (MS media with 100mg/l myo-inositol, 10g/l glucose, 50mg/l MES buffer pH5.5) containing 400μM acetosyringone to an optical density of 2.4 at 650nm. Agrobacterium suspension (1 μl) was inoculated into the immature seed approximately at the position of the scutellum: endosperm interface, using a 10μl Hamilton, so that all exposed seed were inoculated. Tillers were then placed in water, covered with a translucent plastic bag to prevent seed dehydration, and placed in a lit incubator for 3 days at 23°C, 16hr day, 45μEm-2s-1 PAR. After 3 days of co-cultivation, inoculated immature seed were removed and surface sterilized (30 seconds in 70% ethanol, then 20 minutes in 20% Domestos, followed by thorough washing in sterile distilled water). Immature embryos were aseptically isolated and placed on W4 medium (MS with 20g/l sucrose, 2mg/l 2,4- D, 500mg/l Glutamine, 100mg/l Casein hydrolysate, 150mg/l Timentin, pH5.8, solidified with 6g/l agarose) and with the scutellum uppermost. Cultures were placed at 25°C in the light (16 hour day). After 12 days cultivation on W4, embryogenic call i were transferred to W425G media (W4 with 25mg/l Geneticin (G418)). CaIIi were maintained on this media for 2 weeks and then each callus was divided into 2mm pieces and re-plated onto W425G.

After a further 2 weeks culture, all tissue was assessed for development of embryogenic callus: any callus showing signs of continued development after 4 weeks on selection was transferred to regeneration media MRM 2K 25G (MS with

20g/l sucrose, 2mg/l Kinetin, 25mg/l Geneticin (G418), pH5.8, solidified with 6g/l agarose). Shoots were regenerated within 4 weeks on this media and then transferred to MS20 (MS with 20g/l sucrose, pH5.8, solidified with 7g/l agar) for shoot elongation and rooting.

The presence of the T-DNA, and the number of copies are quantified by Quantitative PCR.

For ODC RNAi, 41 transformants (TO) were obtained, regenerated and cultivated in greenhouse. For ADC2 RNAi, 55 transformants (TO) were obtained, regenerated and cultivated in greenhouse.

Example 11 : Analysis of wheat transformants in the presence of pathogenic fungus Inoculation of wheat transformants and control by Fusarium a- glasshouse and culture chamber

The artificial infection occurs using an inoculum of 10⁶ spores / ml in water for Fusarium culmorum or 5x10⁵ spores / ml for Fusarium graminearum. The inoculation is performed on the spike by Ecospray (LCF Labo chimie France, Aix en provence) on plants at a mid flowering stage. After inoculation the spike is covered by a transparent cellophane packet. 24h after the first inoculation a second inoculation is performed on these spikes to cover the flowering period. The cellophane packet is removed 7 days after the second inoculation.

Five spikes are inoculated by plant, 6 plants by repetition and 3 repetitions for each event. The first notation is made 15 days post inoculation, followed by notations every 7 days for a period of 4 weeks. The percentage of infected spikelets is noted, with a spikelet considered to be infected if at least two glumes are infected (white coloured).

b- in field The above described spore solution for inoculation is completed by addition of 0,1 % of tween 20. The inoculation is performed by a nitrogen spray (2bar) at the density of 100ml/m², at a mid flowering stage, at the end of the day to limit the evaporation. The notation is done on 10 plants by line, 2 and 3 weeks after inoculation.

The percentage of infected spikelets is noted as previously described.

Inoculation of wheat transformants

Transformed plants may also be growth and inoculated by the reporter strain TRI5-GFP as described in Example 1 , and the fungus growth may be followed by microscopy and fluorometry (Example 1 ).

Transformed plant may also be inoculated by wild type fungus, and a toxin quantification may be done by an ELISA test as described in Example 1 or by HPLC.

Claims

Claims

1. Method for inducing and/or increasing resistance to a trichothecene-producing pathogen infection in a plant comprising the step of inhibiting or downregulating expression of an endogenous polyamine synthesis gene in said plant, said inhibition or downregulation resulting in said plant to have increased resistance to said pathogen infection as compared to a nearly isogenic plant in which said endogenous polyamine synthesis gene is not inhibited or downregulated.

2. Method for reducing trichothecene level in a plant infected by a trichothecene- producing pathogen, comprising the step of inhibiting or downregulating expression of an endogenous polyamine synthesis gene in said plant, said inhibition or downregulation resulting in said plant to have reduced trichothecene level when infected by said pathogen as compared to a nearly isogenic plant in which said endogenous polyamine synthesis gene is not inhibited or downregulated.

3. Method according to claim 1 or 2, wherein said pathogen is Fusarium.

4. Method according to any of claim 1 to 3, wherein said plant is a cereal.

5. Method according to claim 4, wherein said cereal is chosen in the group consisting of wheat, maize, rice, rye, barley, sorghum and oat.

6. Method according to any of claims 1 to 5, wherein said gene is chosen in the group consisting of arginine decarboxylase (ADC, EC 4.1.1.19), ornithine decarboxylase (ODC, EC 4.1.1.17), agmatinase (EC 3.5.3.1 1 ), argininosuccinate synthase (EC 6.4.3.1 ), argininosuccinate lyase (EC 4.3.2.1 ).

7. Method according to any of claims 1 to 6 wherein said gene is inhibited or downregulated by mutation of said gene by insertion of a transposable or a T- DNA element or by physical mutagenesis.

8. Method according to any of claims 1 to 5, wherein said gene is inhibited or downregulated by transformation of said cereal with a genetic construct.

9. The method of claim 8, wherein said genetic construct comprises a promoter that is induced by Fusarium infection.

10. Method according to any of claims 1 to 9, wherein said plant is rice and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 1 , SEQ ID N° 2, SEQ ID N° 14, SEQ ID N° 15, SEQ ID N° 16, SEQ ID N° 23 and SEQ ID N⁰ 24.

1 1. Method according to any of claims 1 to 9, wherein said plant is wheat and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 3, SEQ ID N° 4 SEQ ID N° 5, SEQ ID N° 6, SEQ ID N° 7, SEQ ID N° 8, SEQ ID N° 9, SEQ ID N° 10, SEQ ID N° 17, SEQ ID N° 18, SEQ ID N° 19, SEQ ID N° 20, SEQ ID N° 21 , SEQ ID N° 25, SEQ ID N° 26 and SEQ ID N° 27.

12. Method according to any of claims 1 to 9, wherein said plant is barley and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 1 1 , SEQ ID N° 12, SEQ ID N° 13, SEQ ID N° 22 and SEQ ID N° 28.

13. Method according to any of claims 1 to 9, wherein said plant is maize and said gene is chosen among the genes having a sequence represented by any of SEQ ID N° 29, SEQ ID N° 55, SEQ ID N° 56, SEQ ID N° 57, SEQ ID N° 58, SEQ ID N° 59, SEQ ID N° 60, SEQ ID N° 61 , SEQ ID N° 62, SEQ ID N° 63,

SEQ ID N° 64, SEQ ID N° 65, SEQ ID N° 66, SEQ ID N° 67, SEQ ID N° 68, SEQ ID N° 69, SEQ ID N° 70, and SEQ ID N° 71.

14. A plant presenting an inhibition or downregulation of the expression of an endogenous polyamine synthesis gene, wherein said plant is resistant to a trichothecene-producing pathogen infection, and wherein said plant is not rice.

15. The plant of claim 14, wherein said inhibition or downregulation is induced by Fusarium infection.

16. The plant of claim 14, wherein said inhibition or downregulation is restricted to the rachis, the spikelets and/or the stem cavity.

17. The plant of any of claims 14 to 16, which is a cereal.

18. The plant of claim 17, which is chosen in the group consisting of wheat, maize, rice, rye, barley, sorghum and oat.

19. A part of the plant of any of claims 14 to 18, wherein said part is a cell or a seed.

20. Method for reducing trichothecene production by a trichothecene-producing pathogen, comprising the step of limiting concentration of polyamines and polyamine amino acid precursors available to said trichothecene-producing pathogen during its growth.

21. Method according to claim 20, wherein said limitation is performed in the growth medium in vitro.

22. Method according to claim 20, wherein said limitation is performed in planta, by growing said trichothecene-producing pathogen on a plant presenting an inhibition or downregulation of the expression of an endogenous polyamine synthesis gene.

23. Method according to claim 20, wherein said limitation is performed in planta by growing said trichothecene-producing pathogen on a plant and applying an inhibitor of an enzyme of the polyamines biosynthetic pathway to said cereal.

24. Method according to claim 20, wherein said limitation is performed by supplying a compound abolishing the effect of said polyamines and polyamine amino acid precursors to said trichothecene-producing pathogen.

25. Method according to any of claims 20 to 24, wherein said pathogen is Fusarium.

26. Method according to claim 25, wherein said limitation is performed by applying an inhibitor of a Fusarium polyamine/amino acid transporter to said Fusarium.

27. Method according to any of claims 20 to 26, wherein said polyamines and polyamine aminoacid precursors are chosen in the group consisting of ornithine, arginine, putrescine and agmatine.