OUP user menu

Molecular cloning of AbGst1 encoding a glutathione transferase differentially expressed during exposure of Alternaria brassicicola to isothiocyanates

Adnane Sellam, Pascal Poupard, Philippe Simoneau
DOI: http://dx.doi.org/10.1111/j.1574-6968.2006.00223.x 241-249 First published online: 1 May 2006


The AbGst1 gene encoding a glutathione transferase from the necrotrophic pathogen Alternaria brassicicola was cloned from a benzyl isothiocyanate-treated conidial culture using differential display reverse transcription. The deduced amino-acid sequence of AbGst1p showed a significant degree of similarity to glutathione transferase-I from Saccharomyces cerevisiae and glutathione transferase-III from Schizosaccharomyces pombe. The transcription of AbGst1 was significantly enhanced by isothiocyanates, heavy metals and 1-chloro-2,4-dinitrobenzene. However, no significant transcript response was obtained with superoxide-generating menadione and paraquat. Recombinant AbGst1p expressed in Escherichia coli exhibited high transferase activity with allyl and benzyl isothiocyanates as substrate compared with 1-chloro-2,4-dinitrobenzene, but no peroxidase activity was detected. AbGst1 was upregulated in planta during the first day postinfection, suggesting the potential involvement of this enzyme in isothiocyanate detoxification mechanisms during host plant infection.

  • Alternaria brassicicola
  • glutathione transferase
  • phytoanticipin
  • detoxification
  • pathogenicity determinant


Alternaria brassicicola is the causative agent of black spot disease in a wide range of Brassica crops. This necrotrophic fungus causes substantial losses in the field and in storage and seed production (Maude & Humpherson-Jones, 1980). During colonization of crucifer host tissues, Alternaria brassicicola is exposed to several antimicrobial plant defence compounds such as phytoanticipins and phytoalexins (Thomma et al., 1999; Tierens et al., 2001). Crucifer phytoanticipins are represented by glucosinolates, i.e. a group of thioglucoside compounds that are hydrolyzed when the plant tissues are damaged. These degradation compounds, which are mainly thiocyanates, nitriles and isothiocyanates (ITCs), exhibit a toxic effect towards bacteria, fungi and insects (Brabban & Edwards, 1995; Wittstock et al., 2004). Isothiocyanates, the major breakdown compounds (Fenwick et al., 1983), have frequently been shown to be highly toxic to fungi (Mari et al., 1996; Smolinska et al., 1997). Alternaria brassicicola have very likely evolved a broad range of detoxification mechanisms to circumvent this chemical barrier. The ability of several pathogenic fungi to detoxify plant chemical defence agents was demonstrated to be a potential pathogenicity determinant. For instance, the capacity of Botrytis cinerea to detoxify the tomato saponin α-tomatine and the phytoalexin resveratrol has been correlated with the aggressivity of this fungus (Quidde et al., 1998; Schoonbeek et al., 2001). Similarly, Gaeumannomyces graminis knockout mutants of avenacinase, an enzyme degradating the oat saponin avenacin A-1, were no longer able to infect host saponin-containing oats (Bowyer et al., 1995).

In animals, many studies have shown that phase II detoxification enzyme glutathione transferase (GST; E.C. plays a key role in the avoidance of isothiocyanate toxicity (Kolm et al., 1995; Munday & Munday, 2004). GSTs are a family of multifunctional enzymes which play an important role in cellular detoxification and excretion of a wide variety of xenobiotic substances. GSTs catalyse the S-conjugation between the thiol group of GSH and an electrophilic moiety in hydrophobic toxicants. Mammalian and plant GSTs have been well characterized, but little is known about fungal GSTs. However, some GSTs have been characterized from fungi such as Saccharomyces cerevisiae (Choi et al., 1998), Schizosaccharomyces pombe (Veal et al., 2002), Phanerochaete chrysosporium (Dowd et al., 1997), Fusarium oxysporum (Cohen et al., 1986), B. cinerea (Prins et al., 2000) and recently from Aspergillus fumigatus (Burns et al., 2005). Studies undertaken on these fungal GSTs revealed that they are potentially involved in protecting the cell against damage attributed to oxidative stress, xenobiotics, heavy metals and antifungal compounds, thus highlighting the functional diversity of these enzymes.

In order to investigate the mechanisms involved in Alternaria brassicicola isothiocyanate detoxification, we carried out differential-display reverse transcription PCR (DDRT-PCR) of mRNA and thus managed to isolate the first GST gene from this necrotrophic fungus. In the present work, we identified and characterized a GST gene that was differentially expressed in the presence of allyl isothiocyanate (AlITC), benzyl isothiocyanate (BzITC), 1-chloro-2,4-dinitrobenzene (CDNB) and heavy metals. We demonstrated that AbGst1 was also upregulated during interaction with Arabidopsis thaliana ecotype Columbia (Co). AbGst1 cDNA was expressed in Escherichia coli and the transferase and peroxidase activities of the recombinant protein were evaluated.

Materials and methods


AlITC, BzITC, CDNB and cupric sulfate (CuSO4) were purchased from Aldrich Chemical Co. (Milwaukee, WI). Camalexin was synthesized by Pr Pietrick Hudomme (Departement de Chimie, Université d'Angers) according to Fürstner & Ernst (1995). All other chemicals were from Sigma (Saint Quentin Fallavier, France).

Organisms, culture conditions, plant infections and pathogenicity assay

The Alternaria brassicicola wild-type strain Abra43 used in this study was isolated from Raphanus sativus seeds. Its taxonomic status was already confirmed, using both morphological and molecular criteria (Iacomi-Vasilescu et al., 2002). Alternaria brassicicola was cultured and maintained on potato dextrose agar. Stress conditions were applied by exposure to the following stress inducers: 10 mM paraquat dichloride, 2 mM manadione (2-methyl-1,4 naphtoquinone), 2.5 mM AlITC, 300 μM BzITC, 125 μM camalexin, 10 mM CuSO4, 10 mM NiCl2 and 200 μM CDNB. Media were prepared by adding stress inducers to 2-day-old germinating conidia grown in potato dextrose broth (PDB). All cultures were performed at 24°C.

Arabidopsis thaliana Co plants were grown to the 8- to 12-leaf stage in controlled environment rooms (21–19°C day and night temperature) and a 8 h light photoperiod. For inoculations, 5 μL drops of a freshly prepared Alternaria brassicicola spore suspension (5 × 106 spores per mL in water) were pipetted onto two to six leaves per plant (two drops per leaf). The plants were then maintained under saturating humidity (100% relative humidity). Control plants were not inoculated, but were otherwise treated in the same way.

To assess fungal development and disease severity during the interaction, the pathogenicity assay described by Gachon & Saindrenan (2004) was carried out. This assay is based on the relative quantification of fungal DNA in planta using real-time quantitative PCR. In this experiment we targeted the Alternaria brassicicola histidine kinase AbNIK1 gene (AY700092) using the following primers: 5′-TGG TGT TGA GGG TAC CTG GAA-3′ and 5′-TCC TCT TGC GAC GGC TGT A-3′, and Arabidopsis thaliana 5.8S ribosomal RNA gene sequence (X52320) using 5′-GCG TTG CTT CCG GAT ATC AC-3′ and 5′-GCC GTT CGT TTG CAT GTT C-3′ as primers, to assess fungal and plant DNA, respectively. AbNIK1 was quantified in infected leaves and normalized with plant 5.8S ribosomal RNA gene sequence.

Nucleic acid isolation and cloning

Whole-cell RNA from Alternaria brassicicola was isolated as described by Guillemette (2004). RT-PCR was performed by adding 5 μg of whole-cell RNA to the reverse transcription system (50 mM Tris-HCl, 75 mM KCl, 10 mM dithiothreitol, 3 mM MgCl2, 400 nM oligo(dT)15, 1 μM random hexamers, 0.5 mM dNTP, 200 U M-MLV reverse transcriptase; Promega, Charbonnière-les-Bains, France). The total volume was adjusted to 30 μL and the mixture was then incubated for 60 min at 42°C. Aliquots of the resulting first-strand cDNA were used for standard or real-time PCR amplification experiments.

The PCR-walking procedure described by Siebert (1995) was used to obtain the 5′ sequence of the AbGst1. PCR-amplified fragments generated by differential display of mRNA or PCR-walking were cloned into pGEMT (Promega). Positive clones were screened by enzymatic digestion. Plasmids were prepared with a plasmid purification kit (Macherey-Nagel, Hoerdt, France) and sequenced.

Differential display of mRNA

Nonradioactive differential display was performed to screen genes differentially expressed in liquid culture treated with BzITC, as described previously by Liang & Pardee (1992). PCR-amplified cDNA fragments were separated on a 6% polyacrylamide/6 M urea gel. The gel was visualized by silver staining according to the protocol of Creste (2001). Bands showing differential expression were excised and reamplified by PCR using the same primer set and the same PCR program used to generate them.

Expression analysis by real-time quantitative PCR

Real-time PCRs were performed with the ABI Prism 7000 sequence detection system (Applied Biosystems, Foster City, CA) using the SYBR green PCR master mix, according to the manufacturer's instructions. After 10 min denaturation at 95°C, the reactions were cycled 40 times at 95°C for 15 s and 60°C for 1 min. To verify that only the specific product was amplified, a melting point analysis was performed after the last cycle by cooling samples to 55°C and then increasing the temperature to 95°C at 0.2°C s−1. A single product at a specific melting temperature was found for each target. All samples were tested in triplicate and the mean was determined for further calculations. Each run included a no-template control to test for assay reagent contamination.

To evaluate the gene expression level, the results were normalized using Ct values obtained from β-tubulin RNA amplifications run on the same plate. Ct represents the first amplification cycle in which fluorescence (indicating the presence of PCR products) was detectable over the basal background fluorescence level. The relative quantification analysis was performed using the comparative Ct method as described by Guillemette (2004). Primers used in real-time quantitative PCR were: β-tubulin, 5′-TTC AAC GAA GCC TCC AAC AAC-3′ and 5′-GTG CCG GGC TCG AGA T-3′; and AbGst1, 5′-GCC CTA GTC ACC GGA AAC ATC-3′ and 5′-CGA CGC CGT TGG TAA TGG-3′. These primers were designated to overlap on two successive exons in order to avoid amplification of contaminant genomic DNA.

Expression and purification of recombinant AbGst1p

The ORF (lacking a stop codon) of the cDNA encoding the AbGst1p was amplified with the upstream primer 5′-ATG TCG AAC CAG CAA GGA GCA AAG AT-3′ and the downstream primer 5′-CAA CTT TGC GTC ACT AAA CGG AAC GTA-3′. The PCR product was cloned into the pTrcHis2TOPO expression vector (Invitrogen, Cergy-Pontoise, France) to express the recombinant protein with a C-terminal Myc-His tag. The identity of the insert was verified by sequencing, and the plasmid was designated pTrcHis2∷AbGst1. Escherichia coli TOPO10 One Shot cells were transformed with the expression vector and induced to express recombinant protein with 1 mM isopropyl-1-thio-β-d-galactopyranoside (IPTG) for 5 h at 30°C. The cells were harvested by centrifugation at 5000 g for 15 min, resuspended in TNG buffer (20 mM Tris-HCl, pH 8, 100 mM NaCl, and 10% glycerol) and disrupted by sonication and three freezing–thawing cycles.

The protein assay reagent MicroBCA™ protein assay kit (PIERCE, Rockford, IL) was used for determination of the total protein content in E. coli extract.

Enzyme assay

The GST activity of E. coli TOPO10 One Shot cell extract containing recombinant AbGst1p was analysed using the GST substrate, CDNB as described by Habig & Jakoby (1981). The assay of GST activity with AlITC and BzITC was carried out at 274 nm according to Kolm (1995). Glutathione peroxidase activity was evaluated as described by Veal (2002). GST transferase and peroxidase activities were calculated by subtracting the basal GST activity of E. coli harbouring empty pTrcHis2 from the activity of E. coli expressing recombinant AbGst1p.


Isolation, cloning and characterization of the AbGst1 gene

Several candidate bands were obtained following differential display analysis of total RNA derived from germinating conidia exposed to 300 μM BzITC for 30 min. One of them was isolated, cloned into pGEM-T and sequenced. The 541 bp cDNA fragment sequence obtained was used to search the GenBank nonredundant database and the Consortium for the Functional Genomics of Microbial Eukaryotes (COGEME) expressed sequence tag (EST) database using the BLASTX algorithm. Significant matches were obtained with ESTs coding for GST homologues from several filamentous fungi like Mycosphaerella graminicola, Gibberella zeae, Magnaporthe grisea, Aspergillus flavus and Ustilago maydis.

A PCR-walking procedure (Siebert et al., 1995) was adopted to obtain the 5′ sequence of AbGst1. The assembled DNA sequence contains an ORF of 259 amino acids interrupted by three introns, whose position was confirmed by comparing the AbGst1 cDNA and the corresponding genomic sequence (Fig. 1).

Figure 1

Nucleotide sequence of Alternaria brassicicola AbGst1 gene (Genbank accession number AY987487). The deduced amino-acid sequence is shown below the ORF. Boxed nucleotide sequences indicates putative TATA (325–331). A potential stress response element (ttaggggga, position 35–43) is underlined. Potential nitrogen-responsive box (inverted, position 123–129) is underlined and indicated in bold.

The predicted AbGst1p protein showed significant identity to GSTs proteins from other filamentous fungi. ClustalW alignment of AbGst1p with four other fungal GST proteins revealed that the most conserved residues were present in the N-terminal domain, where GST active sites are generally located (Wilce et al., 1995) (Fig. 2).

Figure 2

Alignment of the predicted amino-acid sequences of AbGst1p and those of other fungi. Boxes indicate consensus residues from Cluster 1 glutathione transferases (GSTs) as indicated by McGoldrick (2005). The alignment was performed with Clustal W (version 1.83). *, residue identical in all sequences; :, conserved substitutions; et al., semiconserved substitutions. A dash indicates a gap in the sequence; Nc (predicted GST protein, XP_325561), Neurospora crassa; Sc (GST-I, NP_012304), Saccharomyces cerevisiae; Ca (predicted GST, EAK95657), Candida albicans; Sp (GST-III, O59827), Schizosacchromyces pombe.

Phylogenetic analysis by ClustalX revealed that AbGst1p was very similar to GSTs grouped in cluster 1, as defined by McGoldrick (2005). AbGst1p was indeed similar to uncharacterized GSTs from the filamentous fungi Stagonospora nodorum (SNU01866.1), M. grisea (MG05677.4), Neurospora crassa (XP325561.1) and Aspergillus nidulans (AN0629.2). AbGst1p also displayed significant relatedness with the well-characterized GSTIII of Schizosaccharomyces pombe and the endoplasmic reticulum associated GSTI from Saccharomyces cerevisiae (see Fig. 3).

Figure 3

Phylogenetic analysis of some characterized and uncharacterized fungal glutathione transferases (GSTs) based on deduced amino-acid sequences. The phylogenetic analysis were carried out using Clustal X multiple alignment program (Thompson et al., 1997). Branch numbers indicate bootstrap support (1000 replicates). The bar represents 0.05 substitutions per site. Sequence GenBank accession numbers are as follows: Cunninhamella elegans GST1 (AAL02368), Cu. elegans GST2 (AAL02369), Aspergillus fumigatus GstA (AAX07320), A. fumigatus GstB (AAX07318), A. fumigatus GstC (AAX07319), Aspergillus nidulans GstA (AAM48104), A. nidulans predicted GST (AN0629.2), Botrytis cinerea Bcgst1 (AF061253), B. cinerea Bcgst2 (CAE55152), Schizosaccharomyces pombe GST1 (AAK77864), S. pombe GST2 (AAF21054), S. pombe GST3 (AAK59430), Saccharomyces cerevisiae GSTI (P40582), S. cerevisiae GSTII (Q12390), Neurospora crassa predicted GST (XP_325561.1), Magnaporthe grisea predicted GST (MG05677.4), Candida albicans ER-associated GST (EAK95657), Issatchenkia orientalis GSTY1 (BAA77459), I. orientalis GSTY2 (S16178) and Stagonospora nodorum predicted GSTs SNU01866.1, SNU00150.1 and SNU11211.1. Cluster 1, Cluster 2 and GAMMA cluster show GST groups as previously defined by McGoldrick (2005).

Sequence analysis of the 5-flanking region of AbGst1 revealed potential regulatory elements, such as the stress response element (STRE) site (Treger et al., 1998) and a consensus nitrogen-responsive transcription factor 2 (NIT2)-binding region (Fu & Marzluf, 1990) (Fig. 1).

AbGst1 is induced in response to several stress conditions

In order to verify the difference in gene expression of AbGst1 identified by DDRT-PCR, quantitative real-time RT-PCR analysis was used to examine transcript levels in germinating conidia cultures treated with BzITC (we performed RT-PCR with the RNA samples originally used for DDRT-PCR). The effect of AlITC, another breakdown product of glucosinolate sinigrin, on the AbGst1 gene expression level was also tested. Table 1 shows that the real-time RT-PCR data confirmed the results of the DDRT-PCR experiment. AbGst1 expression was indeed induced five times 0.5 h after exposure to BzITC. A modest induction was recorded after 1 h of AlITC treatment. AbGst1 expression was also evaluated in the presence of the major Arabidopsis thaliana phytoalexin camalexin, but no increase in transcript level was observed.

View this table:
Table 1

Time-course of expression of AbGst1 in Alternaria brassicicola germinating conidia after treatment with isothiocyanates (AlITC and BzITC), the phytoalexin camalexin, superoxide-generating compounds (paraquat and menadione) and metals (NiCl2 and CuSO4) measured by quantitative real-time reverse transcription-PCR

Fold change Δ(h) (±SD)
Stress inducers0.5124
Cupric sulfate11.8±0.91±0.11
Nickel chloride3.7±0.20.7±0.03
  • * The measured transcript quantities were normalized in each sample using the Ct obtained for the β-tubulin housekeeping gene. The relative quantification analysis was performed using the comparative Ct method. The values (fold change) represent the number of times the gene is expressed compared with that of free treatment fungal culture. Bold numbers indicate that the ΔCt values measured after treatment were significantly different (P<0.05, two-tailed t-test) from that obtained for controls.

  • Results are the mean of three repetitions.

  • AlITC, allyl isothiocyanates; BzITC, benzyl isothiocyanates; CDNB, 1-chloro-2,4-dinitrobenzene.

Previous studies have suggested that GST genes have differential expression during oxidative stress and after exposure to heavy metals and xenobiotics. To determine if AbGst1 is also affected by these stress inducers, germinating conidia were treated with the xenobiotic CDNB, organic superoxide-generating compounds menadione and paraquat dichloride, and heavy metals CuSO4 and NiCl2. The results summarized in Table 1 show that the AbGst1 transcript level increased markedly following exposure to heavy metals, particularly CuSO4. Treatment with CDNB produced a moderate increase in the AbGst1 transcript level. Moreover, no apparent increase in transcript levels was observed after treatment with oxidants.

These results illustrate that the AbGst1 gene is poised to respond to diverse environmental stresses.

Induction of AbGst1 expression in planta

AbGst1 expression was studied during an interaction with Arabidopsis thaliana Co which shows resistance against Alternaria brassicicola. Total RNA was isolated at different times after inoculation as well as from in vitro cultures of Alternaria brassicicola. Real-time RT-PCR was performed to quantify the relative amounts of AbGst1 transcripts. At the same time, we performed the pathogenicity assay as described by Gachon & Saindrenan (2004) in order to assess fungal development and disease severity during the interaction.

During the interaction with Arabidopsis thaliana, AbGSt1 showed a transient expression pattern in the early stages of infection (day 1 after inoculation) (Fig. 4). This induction overlapped with both the beginning of conidial germination, leaf tissues penetration and the peak of fungal DNA, thus suggesting that Alternaria brassicicola germinating conidia were attempting to set up infection mechanisms.

Figure 4

In planta expression of AbGst1 (histogram) and fungal growth quantification (curve) over an infection time course. The measured transcript quantities were normalized in each sample using the Ct obtained for the β-tubulin housekeeping gene. The relative quantification analysis was performed using the comparative Ct method. The values (fold increase) represent the number of times the gene is expressed compared with that of free-living fungal culture. The abundance of AbNIK1 was quantified in infected leaves, normalized with plant 5.8S ribosomal DNA and expressed by the ratio of the fungal AbNIK1 DNA amount to plant 5.8S ribosomal DNA (NIK1/ITS 5.8S). The data are the mean the standard deviation of triplicates. The star indicates the beginning of conidia germination and leaf tissues penetration.

Enzyme activity of recombinant AbGst1p

For the analysis of GST activity, recombinant AbGst1p was produced in E. coli. After induction with IPTG, the bacterial extract showed an extra 33 kDa band compared with that of the uninduced bacterial culture (data not shown).

The activity of recombinant AbGst1p was assayed with various potential substrates, including the GST substrate CDNB, cumene hydroperoxide and the plant antimicrobial compounds BzITC and AlITC, and the results are summarized in Table 2.

View this table:
Table 2

Glutathione transferase activities with four different substrates, CDNB, AlITC, BzITC and cumene hydroperoxide, of bacterial extracts expressing pTrcHis2∷AbGst1

SubstrateGST activity (μmol min−1 mg1) (±SD)
Cumene hydroperoxideND
  • * GST activity was calculated by subtracting basal activity of Escherichia coli harbouring empty pTrcHis2 from the activity of bacteria expressing recombinant AbGst1p.

  • GST activities are the average of three replicates and were calculated as micromoles of product formed min−1mg−1 of extract protein.

  • AlITC, allyl isothiocyanates; BzITC, benzyl isothiocyanates; CDNB, 1-chloro-2,4-dinitrobenzene; GST, glutathione transferase; ND, no detectable activity.

Compared with extracts from bacteria expressing empty pTrcHis2, bacterial extracts expressing pTrcHis2∷AbGst1 showed enhanced activity with CDNB, AlITC and BzITC. However, GST activity was higher with the two isothiocyanates than with CDNB. No GSH peroxidase activity was detected with cumene hydroperoxide.


Alternaria brassicicola, the causal agent of the black spot disease, is exposed to high levels of indolic phytoalexins and cyanogenic glucosides during host infection. It has been previously shown that some fungi are able to tolerate cyanogenic products by cyanide-resistant respiration (Periera et al., 1997) or by enzymatic detoxification involving cyanide hydratase (Fry & Myers, 1981). Recently, Cramer & Lawrence (2004) have shown that CyhAB, a cyanide hydratase encoding an Alternaria brassicicola gene, was upregulated during interaction with Arabidopsis thaliana, suggesting that it may be candidate for glucosinolate-breakdown product detoxification. We carried out DDRT-PCR in an attempt to isolate other Alternaria brassicicola genes potentially involved in cyanogenic isothiocyanate detoxification. This allowed us to clone a gene (AbGst1) encoding a GST homologue which, to our knowledge, is the first fungal GST gene differentially expressed after exposure to isothiocyanates.

Phylogenetic analysis of the deduced amino-acid sequence compared with 22 other fungal GSTs revealed that AbGst1p belongs to cluster 1 GSTs (McGoldrick et al., 2005). Other GSTs from filamentous fungi have recently been described in Aspergillus fumigatus (Burns et al., 2005) and Botrytis cinerea (Schulze-Gronover et al., 2005), but all belong to cluster 2 GSTs. Extensive sequence conservations within the N-terminal domain were identified upon alignment of AbGst1p with other cluster 1 GST sequences. In particular, the cluster 1 GST consensus motif defined by McGoldrick (2005) was strongly conserved in AbGst1p. The N-capping box ((S/T)XXD) present in all soluble GSTs (Cocco et al., 2001) and playing a key role in GST folding and stability was also present at position 196–199. The existence of these consensus motifs in the AbGst1p sequence indicates conservation of the two GST characteristics, i.e. recognition of GSH and the ability to dimerize (Wilce et al., 1995; Cocco et al., 2001), while it has been suggested that variability of the C-terminal domain is linked to the diversity of electrophilic compounds recognized as GST substrates (Marrs et al., 1995).

Glutathione transferases are proposed to be involved in the response against oxidative stress and protection against xenobiotics and endogenous toxins. For instance, in fungi, an antifungal drug detoxification function of the GST enzyme was demonstrated in the yeast Schizosaccharomyces pombe (Veal et al., 2002). In the present work, AbGst1 expression was found to be induced by metals and xenobiotic CDNB, but no significant change in expression was observed when the fungus was challenged by oxidative stress agents. The promoter region of AbGst1 contains several putative cis-regulatory elements that are potentially activated by environmental stimuli to regulate transcription (STRE and NIT2). Induction of AbGst1 transcription in the presence of different stress inducers supports the functionality of the STRE motif. However, no match was found when AbGst1 cDNA was used to search an EST database of Alternaria brassicicola mycelial culture grown under nitrogen starvation conditions (GenBank accession DN475719 to DN477106), which questions the functionality of the putative NIT2-binding site. Previous studies have shown that the expression of Scgst1 and Spgst3, two other cluster 1 GSTs from Saccharomyces cerevisiae and Schizosaccharomyces pombe, respectively, is also induced by metals and CDNB, but not by oxidant agents (Choi et al., 1998; Shin et al., 2002). This suggests that the GST function is conserved in these fungi, in addition to their structural relatedness.

The induction of AbGst1 by plant antimicrobial defence compound isothiocyanates and the early transient expression of AbGst1 in planta suggest that this protein may be involved in isothiocyanate detoxification during plant infection. This hypothesis is supported by the higher conjugation activity exhibited by recombinant AbGst1p towards AlITC and BzITC, compared with CDNB. Several studies previously showed that phase II detoxification mechanisms that engage the GST enzyme in phytophagous insects and humans were strongly affected by isothiocyanates (Kolm et al., 1995; Munday & Munday, 2004; Francis et al., 2005). It therefore appears that the isothiocyanate detoxification mechanism engaging GST is also remarkably conserved in fungi. By contrast, AbGst1p involvement in the neutralization of reactive oxygen species (ROS) generated by plant hosts during the interaction is not supported by its expression profile in the presence of organic superoxide-generating compounds. Furthermore, heterologously expressed AbGst1p showed no peroxidase activity, which is also in accordance with the fact that AbGst1 expression was not activated by oxidant agents.

Only a single report on the involvement of fungal GST during plant infection is currently available. Prins (2000) isolated the first GST gene from a filamentous fungus B. cinerea using RNA subtractive hybridization during an interaction with tomato leaves. Isolation of this gene was an artifact, as the Bcgst1 expression evaluated afterwards by Northern blot was not found to be induced during the interaction. Furthermore, disruption mutants entirely kept their ability to infect the plant host. The authors concluded that either Bcgst1 is not essential for pathogenicity on tomato or it is functionally complemented by other GST genes. The recent description of two additional cluster 2 GSTs in B. cinerea (Schulze-Gronover et al., 2005) may support the latter hypothesis. Although the complete genome sequence of Alternaria brassicicola is not yet available, it is likely that this fungus has more than one cluster 1 GST gene. Indeed, BLAST search in the Stagonospora nodorum genome database, i.e. the first dothideomycete genome sequence to be publicly released, revealed three predicted proteins with typical cluster 1 GST features. Owing to the probable functional redundancy of structurally similar GSTs and since gene disruptions and replacements are not yet routine in Alternaria brassicicola, we did not attempt to obtain and study the phenotype of AbGst1 disruptants. Nevertheless, our results showing upregulation by BzITC and high activity of AbGst1p in the presence of cyanogenic glucosinolate-breakdown products as substrates strongly suggest that AbGst1 plays an active role during the Alternaria brassicicola infection cycle by enabling the fungus to tolerate some plant defence compounds.


This work was supported by the Contrat de Plan Etat-Région Pays de Loire (2000–2006). We wish to thank The Conseil Général de Maine et Loire for providing Adnane SELLAM with a PhD fellowship.


View Abstract