Effects of different carbon sources on trichothecene production and Tri gene expression by Fusarium graminearum in liquid culture

Authors


  • Present address: Feng Jiao, College of Animal Science and Technology, Northwest A&F University, Shaanxi 712100, China.

  • Editor: Michael Bidochka

Correspondence: Akira Kawakami, National Agricultural Research Center, 3-1-1 Kannondai, Tsukuba, Ibaraki 305-8666, Japan. Tel.: +81 29 838 8940; fax: +81 29 838 8837; e-mail: akirak@affrc.go.jp

Abstract

Fusarium head blight caused by Fusarium graminearum is a disease of cereal crops that not only reduces crop yield and quality but also results in contamination with trichothecenes such as nivalenol and deoxynivalenol (DON). To analyze the trichothecene induction mechanism, effects of 12 carbon sources on the production of DON and 3-acetyldexynivalenol (3ADON) were examined in liquid cultures incubated with nine strains of 3ADON-producing F. graminearum. Significantly high levels of trichothecene (DON and 3ADON) production by sucrose, 1-kestose and nystose were commonly observed among all of the strains tested. On the other hand, the levels of trichothecene biosynthesis induced by the other carbon sources were strain-specific. Tri4 and Tri5 expressions were up-regulated in the sucrose-containing medium but not in glucose. Trichothecene accumulation in the sucrose-containing medium was not repressed by the addition of glucose, indicating that trichothecene production was not regulated by carbon catabolite repression. These findings suggest that F. graminearum recognizes sucrose molecules, activates Tri gene expression and induces trichothecene biosynthesis.

Introduction

Fusarium graminearum Schwabe (teleomorph Gibberella zeae) is the most common causal agent of Fusarium head blight (FHB), which infects many crop plants such as wheat and maize in areas where warmth and humidity prevail during the flowering period (Goswami & Kistler, 2004). Fusarium graminearum decreases crop yield and quality (Champeil et al., 2004), and can contaminate grains with mycotoxins. The mycotoxins belong to type B group of trichothecenes including nivalenol (NIV), deoxynivalenol (DON) and their acetylated derivatives (4ANIV, 3ADON and 15ADON) (Miller et al., 1983; Miller & Greenhalgh, 1985). Fusarium graminearum that produces 15ADON is prevalent in the USA and UK, whereas the 3-acetyldexynivalenol (3ADON) producer is prevalent in Asia (Miller et al., 1991; Jennings et al., 2004). The acute toxicity of these trichothecenes has been demonstrated in both animals and humans (Champeil et al., 2004). Therefore, a better understanding of trichothecene biosynthesis and its regulation mechanisms is desirable for the development of new strategies to manage efficiently the risks posed by F. graminearum.

Significant progress has been made toward determining the trichothecene biosynthesis pathway (Kimura et al., 2003; Brown et al., 2004; Goswami & Kistler, 2004). The current knowledge about trichothecene biosynthesis by F. graminearum is derived from the core or main 25-kb region of the Tri-cluster. Among these clustered genes, Tri4, Tri5, Tri6 and Tri10 encode key factors for trichothecene biosynthesis pathway. Tri5 encodes trichodiene synthase, which catalyzes the first committed step in the pathway (Hohn & Beremand, 1989). Tri4 encodes a cytochrome P450 mono-oxygenase involved in four oxygenation steps of the pathway after trichodiene synthesis by Tri5 (Tokai et al., 2007). Tri genes are positively regulated by  Tri6 (Hohn et al., 1999) and Tri10 (Peplow et al., 2003).

Several reports have indicated that signals or substrates from plants regulate mycotoxin production in some fungi such as Aspergillus species (Maggio-Hall et al., 2005) and Fusarium verticillioides (Bluhm & Woloshuk, 2005). Induction and regulation mechanisms of trichothecene production for F. graminearum have not yet been established. A comparison of trichothecene production between Fusarium roseum (3ADON producer) and F. graminearum (15ADON producer) in liquid cultures containing various carbon sources revealed that the carbon sources resulting in the highest yield of trichothecene by the 3ADON producer were not consistent to those by the 15ADON producer (Miller & Greenhalgh, 1985). However, before the significance of these observations is fully apparent, the effects of different carbon sources on trichothecene production need to be elucidated to analyze DON productivity of multiple strains belonging to the same species. The F. graminearum complex has been classified into nine distinct species in recent years (O'Donnell et al., 2004). Among these, F. graminearum s. str. (lineage 7) and Fusarium asiaticum (lineage 6) were isolated from Japan, and their trichothecene productivity when cultured with different carbon sources is unknown.

There is a possibility that carbon catabolite repression (CCR) is a major influence on the carbon-source-mediated regulation of trichothecene biosynthesis (Ronne, 1995). CCR, also known as glucose repression, is a widespread phenomenon that regulates metabolism in many microorganisms. Via CCR, readily metabolizable carbohydrates repress the synthesis of enzymes related to catabolism of alternative carbon sources, which ensure preferential utilization of the most favored carbon source present (Ruijter & Visser, 1997). In Aspergillus nidulans, expression of isopenicillin-N-synthetase gene is repressed in glucose-containing medium and penicillin (second metabolite) biosynthesis is inhibited (Espeso & Penalva, 1992). CCR-mediated control of virulence has been reported for phytopathogenic fungi such as Fusarium oxysporum (Ospina-Giraldo et al., 2003) and Cochliobolus carbonum (Tonukari et al., 2000). Trichothecenes produced by F. graminearum are second metabolites, and have been suggested to be virulence factors in host–Fusarium interactions (Proctor et al., 1995; Bai et al., 2002). Therefore, it is of interest to ascertain whether trichothecene production is regulated by CCR.

Here we focused on the effects of carbon sources on trichothecene biosynthesis by F. graminearum. Nine strains of 3ADON-producing F. graminearum (lineage 7) were analyzed with respect to the effects of different carbon sources on trichothecene (DON and 3ADON) biosynthesis during culture. These effects on expression of Tri genes were confirmed using real-time PCR. We investigated whether trichothecene biosynthesis was subject to CCR.

Materials and methods

Fungal strains, media and culture conditions

Wheat ears displaying classic FHB symptoms were collected from Japan. Nine F. graminearum s. str (linkage 7) strains (H3 〈MAFF101551〉, AO-3, AO-4, AO-6, MY-4, FK-8, YM-4, IB-35 and HR-3) were isolated from the ear tissues. All strains were identified to the species level based on colonial characterization on potato dextrose agar, spore morphology on synthetic nutrient agar, and PCR-amplified fragment length polymorphism (Suga et al., 2008). 3ADON productivity of the strains was determined by GC/MS analysis of extracts from cultures inoculated on rice grains and PCR analysis using primer pairs for identifying 3ADON and 15ADON producers (Ward et al., 2002). Modified Czapek liquid medium (pH 7.7) consisted of 1 g K2HPO4, 0.5 g KCl, 0.5 g MgSO4·7H2O, 10 mg Fe-EDTA, 2 g l-glutamic acid and 10 g of a carbon source (glucose, fructose, sucrose, 1-kestose, nystose, fructan, maltose, amylose, amylopectin, cellobiose, xylose or galactose) per liter. The carbon sources selected comprise the main carbohydrates in plants and were purchased from Wako Chemical Industries (Osaka, Japan). Five milliliters of the medium was dispensed in a glass tube, inoculated with 2 × 103 conidia and incubated with shaking (190 r.p.m.) at 25 °C in the dark. All experiments were repeated in triplicate.

Determination of fungal growth and trichothecene concentration

Trichothecene (DON and 3ADON) concentration in the culture filtrate was determined using the Ridascreen Fast DON ELISA kit or the Ridascreen DON ELISA kit (R-Biopharm AG, Darmstadt, Germany) according to the manufacturer's instruction. For quantitative determination of fungal growth after incubation, mycelium harvested on filter paper was dried at 60 °C overnight before measurement of dry weight.

Extraction of total RNA and preparation of cDNA

After the F. graminearum H3 strain was incubated in modified Czapek liquid medim, the mycelim was collected by centrifugation and was ground for 20 s at 1500 r.p.m. several times with a Multi-Beads Shocker (Yasui Kikai, Osaka, Japan) in liquid nitrogen. Total RNA was extracted using RNAiso (Takara, Shiga, Japan) according to the manufacturer's instruction. The extracted RNA pellet was dissolved in 70 μL of RDD buffer (Qiagen, Valencia, CA), incubated with 10 μL DNAse I (Qiagen) for 30 min at 28 °C, and then purified using the RNeasy Plant Mini kit (Qiagen) according to the manufacturer's instruction for RNA cleanup. First-strand cDNA was synthesized using SuperScript III First-Strand Synthesis Reverse Transcriptase (Invitrogen, Carlsbad, CA) from 1 μg total RNA primed with oligo(dT)20.

Expression analysis by real-time PCR

Abundances of the transcripts of Tri4, Tri5, Tri6 and Tri10 were analyzed using the ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA) with 0.2 μL cDNA in a 20-μL PCR reaction, using a SYBR Premix Ex Taq kit (Takara), according to the manufacturer's instruction. The expression of β-tubulin was used as the endogenous control. PCR reaction was carried out at 95 °C for 30 s, followed by 40 cycles of 94 °C for 10 s, 58 °C for 15 s and 72 °C for 35 s. Results were obtained from three replicate experiments. Primer pairs for amplifying β-tubulin and Tri10 were designed as previously described (Ponts et al., 2007). The other three primer pairs were designed based on the sequences of F. graminearum H3 strain Tri4 (GenBank Accession no. EF685280), Tri5 (GenBank Accession no. EF661664) and Tri6 (GenBank Accession no. EF685281). The primers used for real-time PCR are listed in Table 1. All cDNA samples were checked for residual genomic DNA using the β-tubulin primer pair that was designed in two different exons. The absence of nonspecific PCR amplification or primer-dimer formation was checked by running melting curve and agarose gel analysis of the final PCR product.

Table 1.   Primer pairs used to amplify β-tub, Tri4, Tri5, Tri6 and Tri10 by Real-Time PCR
GeneSequence forwards (5′–3′)Sequence reverse (5′–3′)Tm (°C)
β-tubβtub-F: GGTAACCAAATCGGTGCTGCTTTCβtub-R: GATTGACCGAAAACGAAGTTG57
Tri4*Tri4-F: TAAACGCCCGCGAAGTTCACATri4-R: TGGTGATGGTTCGCTTCGAG57
Tri5Tri5-F: TGCAAGTTCTTTGAGCAGGCTri5-R: CTCCACTAGCTCAATTGAACTTAG56
Tri6Tri6-F: AGCGCCTTGCCCCTCTTTGTri6-R: TCAACACTTATGTATCCGCCTATA55
Tri10Tri10-F: TCTGAACAGGCGATGGTATGGATri10-R: CTGCGGCGAGTGAGTTTGACA58

The relative expression method was used in the analysis according to the ABI user Bulletin (Applied Biosystems) after performing a validation experiment to demonstrate that amplification efficiencies of the four target Tri genes and the reference β-tubulin gene were approximately equal. Fold change in gene expression was normalized to β-tubulin and relative to the expression on the third day after inoculation with the cultures containing glucose, using the inline image method (Livak & Schmittgen, 2001).

Results

Effects of carbon sources on mycelium growth and trichothecene yield

Mycelium growth and trichothecene (DON and 3ADON) yields of nine strains of F. graminearum were compared using a defined liquid medium whose formation differed only with respect to the carbon source (Table 2).

Table 2.   Effects of different carbon sources on mycelium growth and deoxynivalenol (DON) production of Fusarium graminearum 3ADON producers
StrainH3AO-3AO-4
Carbon sourceMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDW
Glucose22.6 ± 1.610.7 ± 5.60.520.8 ± 1.310.7 ± 5.50.518.7 ± 1.67.3 ± 7.30.4
Fructose16.0 ± 1.8ND20.8 ± 0.630.3 ± 30.31.520.2 ± 1.663.3 ± 47.93.1
Sucrose20.4 ± 2.02706.0 ± 851.3132.622.3 ± 0.2961.7 ± 164.643.124.0 ± 0.51238.3 ± 598.851.7
1-Kestose20.5 ± 1.41912.0 ± 703.393.320.4 ± 0.61144.3 ± 186.156.222.2 ± 2.34643.3 ± 1095.0209.2
Nystose21.1 ± 1.8363.7 ± 322.417.320.6 ± 0.62120 ± 1029.3102.718.9 ± 1.41660.3 ± 288.287.7
Fructan9.4 ± 0.8ND11.3 ± 0.6ND8.1 ± 0.8ND
Maltose19.2 ± 0.4ND14.8 ± 3.05.0 ± 5.00.315.6 ± 0.867.0 ± 67.04.3
Amylose20.5 ± 1.5ND21.5 ± 0.410.3 ± 5.30.522.6 ± 0.9ND
Amylopectin18.5 ± 1.4ND25.2 ± 0.236.7 ± 36.71.523.6 ± 0.48.3 ± 8.30.4
Cellobiose21.4 ± 1.4ND18.5 ± 2.7ND21.9 ± 0.237.3 ± 30.11.7
Xylose14.8 ± 2.36.0 ± 6.00.419.9 ± 0.625.0 ± 14.71.313.4 ± 1.0ND
Galactose13.8 ± 1.4ND16.5 ± 2.7ND16.5 ± 2.666.7 ± 58.84
StrainAO-6MY-4FK-8
Carbon sourceMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDW
Glucose19.9 ± 2.117.0 ± 10.10.923.6 ± 0.287.7 ± 31.63.720.1 ± 0.49.8 ± 9.80.5
Fructose16.9 ± 1.3211.7 ± 92.012.516.1 ± 1.825.2 ± 8.71.617.9 ± 0.521.5 ± 12.31.2
Sucrose24.5 ± 0.31385.0 ± 406.556.520.2 ± 1.91325.0 ± 471.465.520.5 ± 2.31073.3 ± 469.252.3
1-Kestose17.8 ± 1.05389.7 ± 1793.5302.819.0 ± 1.62335.0 ± 1272.4123.119.5 ± 1.5611.7 ± 206.831.4
Nystose19.9 ± 0.78312.0 ± 1991.3418.416.2 ± 0.3556.7 ± 147.334.317.8 ± 1.6482.3 ± 176.827.1
Fructan9.8 ± 0.5ND5.9 ± 0.7ND8.6 ± 0.6ND
Maltose20.2 ± 0.325.3 ± 13.41.315.8 ± 3.16.2 ± 6.20.417.1 ± 0.9ND
Amylose21.4 ± 0.642.5 ± 10.4218.2 ± 1.5289.7 ± 113.615.918.6 ± 3.617.8 ± 9.51
Amylopectin18.9 ± 1.356.2 ± 19.8318.7 ± 1.7265.0 ± 156.314.224.5 ± 0.7ND
Cellobiose17.8 ± 1.45.8 ± 5.80.317.4 ± 0.578.5 ± 11.44.519.9 ± 0.5107.7 ± 86.35.4
Xylose18.1 ± 1.031.0 ± 20.81.717.8 ± 2.0178.2 ± 64.01019.0 ± 0.314.2 ± 1.10.7
Galactose17.0 ± 2.022.5 ± 16.91.310.5 ± 2.2220.3 ± 204.821.116.7 ± 0.613.0 ± 0.80.8
StrainYM-4IB-35HR-3
Carbon sourceMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDWMDW (mg)DON (μg L−1)DON/MDW
  1. MDW, mycelium dry weight; ND, not detected.

Glucose23.1 ± 0.57.0 ± 7.00.320.2 ± 0.35.5 ± 5.50.315.3 ± 2.7ND
Fructose17.1 ± 2.515.3 ± 15.30.919.3 ± 0.8115.0 ± 11.85.912.2 ± 1.94.3 ± 4.30.4
Sucrose20.6 ± 3.11349.7 ± 579.365.621.5 ± 0.81890.7 ± 425.587.916.5 ± 2.9581.0 ± 227.135.2
1-Kestose17.6 ± 0.9710.0 ± 341.340.321.8 ± 0.51489.7 ± 182.068.319.5 ± 1.8300.3 ± 132.915.4
Nystose19.8 ± 0.31240.0 ± 486.262.718.9 ± 0.78720.7 ± 4259.2460.612.6 ± 0.6288.3 ± 128.422.8
Fructan13.8 ± 0.4ND7.0 ± 0.5ND10.3 ± 0.4ND
Maltose16.8 ± 2.4ND22.1 ± 0.5ND15.8 ± 1.64.2 ± 4.20.3
Amylose19.0 ± 2.4ND22.1 ± 0.516.5 ± 9.70.722.8 ± 0.9ND
Amylopectin23.4 ± 0.4ND23.4 ± 0.4ND23.7 ± 1.3ND
Cellobiose19.1 ± 3.74.3 ± 4.30.220.5 ± 1.251.5 ± 51.52.520.5 ± 0.33.7 ± 3.70.2
Xylose20.3 ± 0.14.7 ± 4.70.219.3 ± 0.794.7 ± 65.94.915.0 ± 2.28.3 ± 4.20.6
Galactose17.7 ± 1.93.7 ± 3.70.214.7 ± 2.0ND13.3 ± 2.5ND

To analyze the effects of the 12 carbon sources on mycelium growth of the nine strains, the mycelium dry weight in the medium containing different carbon sources was determined on the eighth day after inoculation. Glucose, sucrose, 1-kestose, nystose, amylose, amylopectin, cellobiose, fructose, maltose, xylose and galactose supported significant mycelial growth for all strains, although the growth varied somewhat from strain to strain. Fructan supported less mycelium growth than the other carbon sources.

Significantly higher yields of trichothecenes were evident in all of the strains tested when the medium was supplemented with sucrose, 1-kestose or nystose (P<0.05). The ratios of trichothecene yield to mycelium dry weight were also higher on incubation with sucrose, 1-kestose or nystose than with the other carbon sources in all strains tested. On the other hand, several strains including AO-6, MY-4, FK-8 and IB-35 produced trichothecens at levels of more than 100 μg L−1 by different carbon sources other than sucrose, 1-kesotose and nystose. In the medium containing 0.5% glucose and 0.5% fructose, F. graminearum did not produce trichothecenes (data not shown). The pH of every tested medium after incubation ranged from 7.9 to 8.4.

Time-dependent change of mycelium growth, trichothecene accumulation and expression of Tri genes in the presence of sucrose or glucose

Mycelium growth, trichothecene production and expression levels of Tri4, Tri5, Tri6 and Tri10 by F. graminearum H3 strain cultured in the presence of glucose or sucrose were compared from the second to the eighth day after inoculation. The levels of mycelium growth during incubation were almost same between the media containing glucose and sucrose (Fig. 1). Trichothecenes in the glucose-containing medium were not detected throughout the incubation period. On the other hand, trichothecenes in the sucrose-containing medium were initially detected on the third day after inoculation, the contents increased significantly by the fifth day and continued to increase, albeit slightly, until the eighth day. Tri4 and Tri5 transcripts in the glucose-containing medium were detected, but these abundances did not increase throughout the incubation period (Fig. 2). On the contrary, the expression levels of Tri4 and Tri5 in the sucrose-containing medium increased 10 and 79 times, respectively, from the second to the third day after inoculation. On the third day, the expression levels of Tri4 and Tri5 in the sucrose-containing medium increased seven and 32 times, respectively, than those in the glucose-containing medium. From the fourth to the eighth day after inoculation, the expression levels of Tri4 and Tri5 rapidly decreased. The abundance of Tri6 transcripts slightly increased from the third to the fourth day after inoculation, but the differences in the expression levels during incubation were not detected between the media containing glucose and sucrose. Tri10 expression remained almost constant during incubation and the expression level was similar between the media containing glucose and sucrose.

Figure 1.

 Deoxynivalenol (DON) accumulation and mycelium dry weight of Fusarium graminearum H3 strain incubated with the medium containing glucose (G) or sucrose (S). White and black bars represent DON amount and mycelium dry weight during incubation period, respectively. Values are means±SD of three samples.

Figure 2.

Tri gene expression of Fusarium graminearum H3 strain incubated with the medium containing glucose or sucrose. White and black bars represent relative expression levels of each gene (Tri4, Tri5, Tri6 and Tri10) during incubation with the medium containing glucose and sucrose, respectively. Values are means±SD of three samples.

CCR and trichothecene production

To examine whether glucose repressed trichothecene biosynthesis induced by the sucrose-containing medium, F. graminearum H3 strain was cultured in the medium containing different concentrations of glucose, sucrose, or both, to observe DON production (Fig. 3). The strain grew well in all medium formulations. Trichothecenes were detected in trace amounts in the glucose-containing medium, but in significant amounts in the other medium formations on the fourth and eighth days after inoculation. These data suggest that glucose supports fungal growth but not trichothecene production, while sucrose supports not only fungal growth but also significant trichothecene production regardless of the coexisting glucose concentration in the culture medium.

Figure 3.

 (a) Mycelium dry weight and (b) deoxynivalenol (DON) concentration of Fusarium graminearum H3 strain incubated with the medium containing different concentrations of glucose (G), sucrose (S) or both. White and black bars represented mycelium dry weight or DON concentration after 4 and 8 days of incubation, respectively. Values are means±SD of three samples.

Discussion

Trichothecene production in liquid cultures of Fusarium species are influenced by hydrogen peroxide (Ponts et al., 2007), fungicides (Covarelli et al., 2004) and carbon and nitrogen sources (Miller & Greenhalgh, 1985). In the present study, the effects of 12 carbon sources on trichothecene biosynthesis were examined using nine strains of F. graminearum 3ADON producer growing in modified Czapek liquid medim.

Significantly higher levels of trichothecene (DON and 3ADON) production by sucrose, 1-kestose and nystose compared with the other carbon sources were commonly observed among all of the strains tested (Table 2). In addition, high levels of trichothecene yield/mycelium dry weight ratio upon incubation in the presence of sucrose, 1-kestose and nystose indicate that large amounts of the trichothecenes are induced by specific carbon sources rather than by higher mycelium growth rate. On the other hand, several strains including AO-6, MY-4, FK-8 and IB-35 produced trichothecenes at levels of more than 100 μg L−1 by different carbon sources other than sucrose, 1-kesotose and nystose. The results suggest that trichothecene productions by sucrose, 1-kestose and nystose are commonly observed characteristics in 3ADON producers of F. graminearum and the productions by the other carbon sources were strain-specific characteristics. The results of a previous study that compared trichothecene production between F. roseum (3ADON producer) and F. graminearum (15ADON producer) in YEP medium (0.1% yeast extract, 0.1% peptone and 1% carbon source) suggested that carbon sources resulting in the high yields of trichothecenes were different between the two strains (Miller & Greenhalgh, 1985). However, a more detailed comparison of this previous study with the present one reveals that high levels of trichothecene biosynthesis induced by sucrose is a common characteristic among DON producers. Trichothecene production was also effectively induced by 1-kestose and nystose, although the previous study omitted these carbon sources (Miller & Greenhalgh, 1985). 1-Kestose and nystose are fructo-oligosaccharides that bind one and two molecules of β-(2,1)-linked fructosyl units to sucrose, respectively, and therefore have structures very similar to sucrose. Moreover, these fructo-oligosaccharides can be degraded to sucrose and fructose by exoinulinases secreted by fungi. In addition, mixtures of glucose and fructose, which were degradation products of sucrose, did not induce trichothecene biosynthesis. Fusarium verticillioides induces high amounts of fumonisin B1 by incubation with amylopectin and dextrin, and α-1,6-glucosides that comprise these carbon sources are suggested to be inducers for fumonisin B1 (Bluhm & Woloshuk, 2005). Therefore, the induction of DON biosynthesis by sucrose, 1-kestose and nystose raises the possibility that the α-glucoside structures included in these carbon sources might be key factors for the induction. In bacteria and plants, many α-glucoside-binding proteins have been identified. Further research is needed to determine the induction mechanism of these mycotoxins by specific carbon sources.

Trichothecenes promote the spread of F. graminearum in wheat spikes after infection but not the infection (Proctor et al., 1995; Jansen et al., 2005). Based on our data that sucrose, 1-kestose and nystose significantly induced trichothecene biosynthesis, we suggest that trichothecenes are not produced when F. graminearum germinates and penetrates the surface of wheat epidermal tissues, which contain low amounts of specific carbon sources needed to induce trichothecene production, and that trichothecene production is markedly elevated in the infected tissues during mycelium spreading via cells or intercellular spaces of wheat spike tissues, in which carbon sources such as sucrose and fructo-oligosaccharides are the major carbohydrates after anthesis compared with mature kernels (Takahashi et al., 2001). Trichothecene production by F. graminearum is low during the growth on mature wheat kernels but high during infection in the spike tissues after anthesis, indicating that the specific factors present only during infection in the spike tissues enhance trichothecene production and seem to be absent during colonization in mature kernels (Voigt et al., 2007). Our results indicate the possibility that the amounts of sucrose and fructo-oligosaccharides in the infected tissues affect trichothecene productivity. Our data provides the insight to develop new strategies for reducing trichothecene production in wheat tissues infected by F. graminearum.

Expression levels of Tri5 and Tri4 reached peaks on the third day and trichothecene amounts drastically increased on the fourth day after inoculation in the sucrose-containing medium (Fig. 2). These data demonstrate that sucrose strongly induces the expression of various Tri genes of F. graminearum, and trichothecene production induced by sucrose is regulated by transcription levels of Tri genes but not by posttranscription levels. However, expression levels of Tri6 and Tri10 did not change during incubation periods. These results agree with published results for F. graminearum (15ADON producer) (Ponts et al., 2007). Because Tri6 and Tri10 encode positive regulators for the expression of the other Tri genes, it is expected that expression levels of Tri6 and Tri10 increase before the increased expression of Tri5 and Tri4 encoding the crucial enzymes in trichothecene biosynthesis. Accordingly, our results indicate that Tri6 and Tri10 activate the other Tri gene expressions but cannot be detected in this study because the expression periods of Tri 6 and Tri10 are very short (Kimura et al., 2003), or the other mechanisms activate Tri gene expression. Further studies are necessary to elucidate the mechanisms by which sucrose induces Tri gene expression and regulates the trichothecene biosynthesis pathway.

Microorganisms can adjust their catabolism to prevailing conditions; CCR is one of these regulatory mechanisms. Many cell-wall-degrading enzymes secreted by phytopathogenic fungi such as F. oxysporum (Ospina-Giraldo et al., 2003) and C. carbonum (Tonukari et al., 2000) are regulated by CCR, and the mutants created by disruption of the genes required for expression of CCR exhibit much less virulence to plants. Thus, CCR seems to be directly linked to phytopathogenicity. We considered the possibility that trichothecene biosynthesis was regulated by CCR. However, the trichothecene production induced by sucrose was not inhibited by the addition of glucose (Fig. 3). These data indicate that the F. graminearum 3ADON producer recognizes α-glucoside structures in plant tissues containing various carbon sources, induces Tri gene expressions, and then produces DON.

In summary, the results of this study showed the effects of carbon sources on trichothecene production and suggested that sucrose increased Tri gene expression and trichothecene production. The results also suggested that trichothecene production by sucrose is not regulated by CCR. However, it appears that various factors other than carbon sources also influence trichothecene production. Further study is needed to identify those factors and elucidate the correlations among the factors.

Acknowledgement

We thank Dr Norio Shiomi, Graduate School of Dairy Science Research, Rakuno Gakuen University, Japan for providing 1-kestose.

Ancillary