• Glutathione transferase;
  • Detoxification;
  • Evolution;
  • Oxidative stress;
  • genomics


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

Most fungal glutathione transferases (GSTs) do not fit easily into any of the previously characterised classes by immunological, sequence or catalytic criteria. In contrast to the paucity of studies on GSTs cloned or isolated from fungal sources, a screen of databases revealed 67 GST-like sequences from 21 fungal species. Comparison by multiple sequence alignment generated a dendrogram revealing five clusters of GST-like proteins designated clusters 1, 2, EFIBγ, Ure2p and MAK16, the last three of which have previously been related to the GST superfamily. Surprisingly, a relatively small number of fungal GSTs belong to mainstream classes and the previously-described fungal Gamma class is not widespread in the 21 species studied. Representative crystal structures are available for the EFIBγ and Ure2p classes and the domain structures of representative sequences are compared with these. In addition, there are some “orphan” sequences that do not fit into any previously-described class, but show similarity to genes implicated in fungal biosynthetic gene clusters. We suggest that GST-like sequences are widespread in fungi, participating in a wide range of functions. They probably evolved by a process similar to domain “shuffling”.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

Studies of how enzymic detoxification removes xenobiotics and their metabolites unidirectionally from the cell have provided important insights into processes such as defence against oxidative stress, gene regulation and mechanisms of drug clearance from cells [1,2]. Quantitatively, one of the most important detoxification enzyme activities is that of the glutathione transferase (GST) superfamily, which catalyses conjugation of xenobiotics or their metabolites to glutathione (GSH) [3]. GSH conjugates are actively excreted from the cell by ATP-ase pumps, members of the ABC transporter protein class, such as the GSH conjugate pump, P-glycoprotein [see references cited in [3]], and similar bacterial proteins [1,2]. As the glutathionylate moiety is hydrophilic, the conjugate cannot simply re-diffuse back into the cell. In plants and yeasts, GSH conjugates are pumped into intracellular vacuoles by ABC transporters thus removing xenobiotic conjugates from the intracellular space [4].

Based on sequence, substrate specificity, three-dimensional structure and immunological properties, GSTs have been grouped into at least eight structurally distinct families [reviewed in reference [3]]. In addition, there is a separate microsomal group of activities catalysed by non-cytosolic enzymes grouped as membrane associated proteins involved in eicosanoid and glutathione metabolism (MAPEG). This includes a membrane-bound and trimeric microsomal GST and prostaglandin E-synthase. In contrast with mammals, plants and even bacteria, comparatively little is known about GSTs of yeasts and fungi generally but fungal GSTs seem to be especially diverse both structurally and functionally. A number of bioinformatic studies have been performed on GSTs [cited in [3]]. One of the largest and most comprehensive of these suggested that as many as 44 distinct GST classes may exist even though only eight mainstream classes (Alpha, Mu, Pi, Sigma, Zeta, Kappa, Omega & MAPEG) plus one bacterial-specific, two plant-specific and three insect-specific [5] classes (i.e. a total of 14 classes) have so far been described [6]. The usual yardstick for including two GSTs in one class is a minimum of 40% sequence identity and, when less than 30% identical, they are allocated to different classes. The relationship between fungal GSTs and the mainstream classification system remains unclear. The best-understood fungal enzymes (e.g. those from Issatchenkia orientalis, Saccharomyces cerevisiae, Cunninghamella elegans) do not fit conveniently into this classification. We now report genomic analyses of fungal GSTs and GST-like proteins and assess the implications of new fungal genome knowledge for the GST superfamily.

2Materials and methods

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

Complete genomes are available at NCBI for S. cerevisiae, Schizosaccharomyces pombe, Encephalitozoon cuniculi and Eremothecium gossypii (;organism=fungi). Partial genomes are searchable at this site for Aspergillus nidulans, Gibberella zeae, Magnaporthe grisea, Neurospora crassa, Candida albicans and Ustilago maydis. Preliminary open reading frame (ORF) product data for four new genomes –Yarrowia lipolytica, Candida glabrata, Kluyveromyces lactis and Debaromyces hansenii– are available for BLAST search at the Genolevures site (, which compares the S. cerevisiae genome to 14 other hemiascomycetous yeasts (Dr. David Sherman, personal communication; see literature cited in reference [7]). Sequences can be found by entering the protein identifier given in Table 1 into the search field of the Genolevures URL. Candidate ORFs for C. albicans were searched at the Candida database server ( Some unpublished genes and ORFs showing similarity to GSTs were also included in our analysis.

Table 1. 
No.Protein accession number (NCBI) or Protein identifier*/**OrganismProtein name/description
  1. *Genolevures site (See Material and Methods).

  2. **Candida database (See Materials and Methods).

1P30102Issatchenkia orientalisGSTY-2
2BAA77459Issatchenkia orientalisGSTY-1
3NP_013040Saccharomyces cerevisiaeGTT2
4NP_012304Saccharomyces cerevisiaeGTT1
5AAR98813Alternia alternataGST cDNA
6NP_588171Schizosaccharomyces pombeProtein with GST domain
7NP_586198Encephalitozoon cuniculiNuclear protein of GST family
8EAA32170Neurospora crassaORF, related to microsomal GST 3
9Q9Y7Q2Schizosaccharomyce pombeGST-1
10O59827Schizosaccharomyces pombeGST-2
11Q9P6M1Schizosaccharomyces pombeGST-3
12AAL02368Cunninghamella elegansClass Gamma GST(GST1)
13AAL02369Cunninghamella elegansClass Gamma GST(GST2)
14AAM48104Aspergillus nidulansGST A, Theta class
15BAB68404Gibberella fujikuroiPutative GST
16AAG43132Botryotinia fuckelianaGST 1, Theta class
17CAC16080Pichia augustaGST
18CAB66903Polymyxa betaeGST
19XP_367838Magnaporthe griseaORF
20XP_364293Magnaporthe griseaORF
21O43123Aspergillus nidulansMaleylacetoacetate isomerase (MAAI)
22Q00717Aspergillus nidulansPutative sterigmatocystin biosynthesis protein (STCT)
23EAA47645Magnaporthe griseaORF
24AAA16892Saccharomyces cerevisiaeEF1Bγ, gene TEF3
25NP_012842Saccharomyces cerevisiaeEF1Bγ homologue of TEF4
26P40921Schizosaccharomyces pombeEF1Bγ, gene TEF3
27NP_984243Eremothecium gossypiiADR147Cp, similar to TEF4
28EAA77260Gibberella zeaeORF
29XP_323127Neurospora crassaORF
30EAA55279Magnaporthe griseaORF
31EAA57903Aspergillus nidulansORF
32NP_587885Schizosaccharomyces pombePutative glutathione synthase
33EAA64302Aspergillus nidulansORF
34EAA66371Aspergillus nidulansORF
35NP_011717Saccharomyces cerevisiaeORF
36XP_391216Gibberella zeaeORF
37CAGL0L05148g*Candida glabrataORF, similar to S. cerevisiae TEF4
38YALI0C24420g*Yarrowia lipolyticaORF, similar to S. cerevisiae TEF3
39YALI0B12562g*Yarrowia lipolyticaORF, similar to S. cerevisiae TEF4
40DEHA0D17369g*Debaryomyces hanseniiORF, similar to C. albicans CaCAM1.exon2
41CA1723**Candida albicansTEF4
42KLLA0F26092g*Kluyveromyces lactisORF some similarities to C. albicans TEF4
43KLLA0D11594g*Kluyveromyces lactisORF, similar to S. cerevisiae TEF3
44CA5680**Candida albicansC. albicans CAM1.exon2, similar to TEF3. exon2
45CA3260**Candida albicansORF, IPF7968
46YALI0A06743g*Yarrowia lipolyticaORF, similar to A. nidulans GST
47YALI0C21021g*Yarrowia lipolyticaORF, similar to A. nidulans GST
48YALI0F25575g*Yarrowia lipolyticaORF, similar to A. nidulans GST
49KLLA0A00264g*Kluyveromyces lactisORF, similar to S. cerevisiae GTT1
50DEHA0D17699g*Debaryomyces hanseniiORF, similar to C. albicans GTT2
51DEHA0D17677g*Debaryomyces hanseniiORF, similar to C. albicans GTT1.3
52DEHA0D20295g*Debaryomyces hanseniiORF, similar to C. albicans IPF7968, unknown function
53CA5044**Candida albicansGTT2
54CA4712**Candida albicansGTT1.3, GST 3 prime end
55CA4392**Candida albicansGST
56YALI0C08052g*Yarrowia lipolyticaORF, similar to S. cerevisiae MAK16
57CAGL0G06248g*Candida glabrataORF, similar to S. cerevisiae MAK 16
58KLLA0A04037g*Kluyveromyces lactisORF, similar to S. cerevisiae MAK 16
59DEHA0F02112g*Debaryomyces hanseniiORF, similar to C. albicans MAK 16
60CA2670**Candida albicansMAK 16
61NP_009377Saccharomyces cerevisiaeMAK 16
62YALI0C03069g*Yarrowia lipolyticaORF, similar to C. albicans URE2p
63CAGL0J07392g*Candida glabrataORF, similar to S. cerevisiae URE2p
64KLLA0D19624g*Kluyveromyces lactisURE2p
65DEHA0F08635g*Debaryomyces hanseniiORF, similar to C. maltosa URE2p
66CA0233**Candida albicansURE2p
67NP_014170*Saccharomyces cerevisiaeURE2p
68CAD29476Triticum aestivumGSTF3, Phi class
69Q96266Arabidopsis thalianaGST6, Phi class
70CAA68993Petuniax hybridaGST, Phi class
71P46440Nicotiana tabacumGST APIC, Phi class
72AAB60886Triticum aestivumGST, Zeta class
73AAB96392Homo sapiensGSTZ1, Zeta class
74AAP99176Prochlorococcus marinusGST, Zeta class
75AAA33277Dianthus caryophyllusGST, Zeta class
76NP_509962Caenorhabditis elegansGST-42, Zeta class
771GWCBTriticum tauschiiGST, Tau class
78AAQ02687Oryza sativaGST, Tau class
79P30568Pleuronectes platessaGST A, Theta class
80AAC13317Homo sapiensGSTT2, Theta class
81AAB03534Mus musculusGSTT2, Theta class
82P30105Drosophilia mauritianaGST 1-1, Theta class
83AAA60963Homo sapiensGST, Mu class
84CAA41202Gallus gallusGSTCL2, Mu class
85AAA37747Mus musculusGST, Mu class
86AAA37751Mus musculusGST, Alpha class
87AAA16572Gallus gallusGST, Alpha class
88AAA70226Homo sapiensGST2, Alpha class
89P30113Schistosoma bovisGST, Alpha class
90AAL08414Takifugu rubripesGST, Omega class
91NP_899062Homo sapiensGSTO2, Omega class
92Q9N1F5Sus scrofaGSTO1-1, Omega class
931PN9AAnopheles gambiaeGST, Delta class
94AAA29294Musca domesticaGST1, Delta class
95CAA32449Drosophilia melanogasterGST1-1, Delta class
96AAL59658Anopheles gambiaeGSTE1, Epsilon class
97P41043Drosophilia melanogasterGSTS1, Sigma class
98AAA29405Ommmastrephes sloaniGST, Sigma class
99AAA63411Ommmastrephes sloaniGST, Sigma class
100P46436Ascaris suumGST, Sigma class
101NP_665735Homo sapiensGST-1, MAPEG class
102P08011Rattus norvegicusGST-1, MAPEG class
103CAA33508Homo sapiensGST, Pi class
104CAA26664Rattus norvegicusGST, Pi class
105NP_852036Rattus norvegicusGST 13-13, Kappa class
106Q9Y2Q3Homo sapiensGST 13-13, Kappa class
107AAC44362Proteus mirabilisGST, Beta class
108CAA76728Ochrobactrum anthropiGST, Beta class
109BAA07509Escherichia coliGST, Beta class
110CAA54033Pseudomonas sp. LB400GST, Beta class
111AAO15607Scacoptes scabieiGST


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

A total of 18 search sequences encoding fungal GSTs were initially obtained from the Entrez protein database at NCBI. These were used to identify 67 GST-like sequences deduced from individual genes or ORFs in the databases listed above covering 21 fungal species. Table 1 lists these “hits” along with 44 representative sequences of mainstream GST classes from a range of biological sources. All “hits” were well above the PSI-BLAST inclusion threshold of 0.005. Multiple alignment was used to generate a dendrogram shown in Fig. 1[8]. The pattern of divergence among mainstream GST classes in this tree is similar to that found in previous studies [3,6] and some fungal sequences cluster to well-described GST classes in this analysis. This conclusion is supported by generally high local bootstrap values in most cases (Fig. 1). The gene for A. nidulans maleylacetoacetate isomerase clearly belongs to Zeta class as previously described [9,10] and an N. crassa hypothetical protein containing a MAPEG domain shows homology to microsomal GST 3. Similarly, a putative gene from G. zeae shows 49% sequence identity to a Phi class GST from Arabidopsis thaliana and the gene for a GST of Polymyxa betae is 43% identical to a Beta class gene from Pseudomonas. However, the bulk of the fungal sequences are distributed in five “clusters” clearly outside the mainstream classification. These appear to be widespread in the genomes investigated and are discussed individually below. Interestingly, the C. elegans GSTs allocated to the fungal Gamma class are clearly distinct both from mainstream classes and the five clusters but do not appear to be widespread in the fungi studied [11]. [Note: sequences corresponding to residues 1–230 (i.e. the N-terminal domain) of EF1Bγ were used in this comparison see Section 3.2 below].


Figure 1. Phylogenetic relationship between GST-like protein sequences in 21 fungal species. An unrooted bootstrapped tree (n= 1000) was constructed with 111 individual GST-like sequences (Table 1). Multiple alignment and neighbour joining methods [8] were performed with full-length sequences (excepting EFIBγ for which only the N-terminal GST-like domain of 230 residues was included). Bootstrap values are presented as percentage at nodes and only probabilities greater than 50% are reported as the nodes in question were found in only half of the bootstrap replicates. Values greater than 70% are generally regarded as reflecting reliable groupings. Scale bar (0.1) gives a representation of 10% sequence divergence.

Download figure to PowerPoint

3.1Clusters 1 and 2

Cluster 1 includes; S. pombe GST3, S. cerevisiae GTT1, K. lactis GST, two Debaromyces hanseii GST ORFs and two sequences from C. albicans (IPF 7968 and GTT1, 3). Cluster 2 includes: three Y. lipoytica GSTs, Botyrotinia fuckelinia GST1, A. nidulans GST A, and an ORF from D. hanseii. It is noteworthy that the sequences of B. fuckelinia GST1 and A. nidulans GSTA were both originally allocated to Theta class which was previously a much more all-encompassing grouping [3]. It is also remarkable that several sequences such as those of Alternia alternata and Pichia augusta show almost no similarity to any GST classes. As new genomes are sequenced and as more is discovered about relationships amongst GSTs, it is likely that further genes similar to these “orphan” sequences will be discovered. A recent study of the sirodesmin biosynthetic gene cluster of the plant pathogenic fungus Leptosphaeria maculans revealed that sirG shows 59% sequence identity to the A. alternata GST (5 in Table 1) [12]. Another gene in this cluster, sirM, encodes an Omethyltransferase domain similar to that of an M. grisea GST-like sequence (19 in Table 1) while three genes (sirB, sirC and sirE) encode cytochrome P-450 domains. Interestingly, another gene cluster responsible for synthesising the highly mutagenic polyketide sterigmatocystin in A. nidulans also contains a gene with O-methyl transferase activity (stcP), a second gene encoding EFIBγ (stcT) (22 in Table 1) and four genes (stcB, stcF, stcL and stcS) encoding cytochrome P-450 domains [13]. GST-like proteins (e.g. encoded by sirG and stcT) might protect fungal cells against toxicity generated during biosynthetic pathways [12,13].

3.2Elongation factors EF1Bγ

At least three proteins form eukaryotic elongation factor 1 (eEF1) responsible for the elongation step of protein synthesis in yeast [see literature cited in reference [14]]. EF-1A binds aminoacyl-tRNAs to the 80S ribosome while EF-1B contains two distinct subunits; EF1Bα and EF1Bγ. EFIBγ seems to catalyse the GTP-GDP nucleotide exchange role of EF1Bα and also physically to attach the complex to membranes of the endoplasmic reticulum or tubulin. EF1Bγ is a dimer of approximately 50 kDa subunits which consist of distinct N- and C-terminal domains. In S. cerevisiae these domains are separable by mild tryptic digestion. Sequence alignment, motif searching and homology modelling had previously revealed that the N-terminal domain of fungal EF1Bγ is closely related to GSTs [14]. While all fungal EF1Bγ proteins seem to contain this N-terminal domain, the C-terminal domain is of differing length in various fungal species. In A. nidulans, it is not present at all and in this species the entire EF1Bγ protein consists of the N-terminal GST-like domain only (Fig. 2) [14]. This suggests that the functionality of the EFIBγ may reside in the GST-like domain. Despite this sequence homology, it was unclear whether the EF1Bγ protein retained GST-type catalytic activities even though this was predicted from motif searching. Recombinant EF1Bγ from Oryza sativa has been shown to catalyse GSH conjugation to 1-chloro-2, 4-dinitrobenzene [19]. Separately, we isolated a catalytically-active 100 kDa GST from the non-saccharomyces yeast Y. lipolytica and found it to possess catalytic properties similar to smaller GSTs [20]. However, this protein did not bind to GSH affinity resins and did not immunoblot with mainstream GSTs. More recently, a crystal structure for the recombinant 219 residue N-terminal GST-like domain of EF1Bγ from S. cerevisiae encoded by the TEF3 gene has been determined (Fig. 2) [16]. This showed unambiguously that this protein is a member of the GST structural superfamily and the authors allocated it to the Theta class. However the recombinant protein displayed no GST-like catalytic activity and GSH was absent from the putative catalytic site. A homologue from Y. lipolytica (TEF4: 38 in Table 1) gave 100% identity with the 30-residue N-terminal sequence determined from the GST which we had previously purified from this source [20]. This GST is, therefore, an EFIBγ containing a catalytically active GST-like N-terminal domain. In Fig. 1 it is clear that the N-terminal GST-like domains of a group of genes denoted TEF3/TEF4 from a wide range of yeast and fungal species cluster together and are distinct from Theta and all other mainstream GST classes. Interestingly, the group appears to be closely related to the GSTs of I. orientalis (originally allocated to Theta class) and includes the GSH synthase gene from S. pombe while Beta class bacterial GSTs are clearly distinct from the EF1Bγ class (Fig. 1). We propose that this group representing 23 of the 67 fungal sequences should be regarded as a distinct class widespread in yeasts and fungi that was recruited into the elongation factor machinery of eukaryotic cells at an early stage of evolution. Similar genes are present in higher organisms. The protein encoded by TEF3 has been identified in a complex binding to the msrA promoter, suggesting a possible function in regulation of expression of methione sulfoxide reductase and thus a role in the oxidative stress response [21]. Protecting the elongation step of protein synthesis against oxidative stress is a possible explanation for recruitment of GSTs.


Figure 2. Domain structure of selected fungal GST-like proteins. GST subunits have distinct N- and C-terminal domains [3] but these can apparently be combined with other functional units. Domains of selected GST-like proteins were identified with PFam ( [15] and are presented as follows: GST N-terminal domain (blue), GST C-terminal domain (red), P-kinase domain (grey), EF1Bγ C-terminal domain (green). Numbers in parentheses refer to Table 1. Putative N-terminal prion domains of Ure2p were separately identified by multiple sequence alignment [8] with that of S. cerevisiae (black). The domain structure of the S. cerevisiae EFIBγ sequence (24 in Table 1) is essentially identical to that for sequences 25–27, 29–31, 38, 39 and 41 (Table 1). Three-dimensional structures were viewed with Protein Explorer ( using PDB files 1NHY and 1G6W for: (a) EFIBγ[16]; (b) Ure2p [17], respectively. N-terminal domains (blue) are 4–72 (EFIBγ) and 95–196 (Ure2p) while C-terminal domains (red) are 93–198 (EFIBγ) and 206–354 (Ure2p), respectively, giving good agreement with Pfam. The flexible loop of Ure2p is represented in magenta; (c) Three dimensional structure of human Theta class hGST2-2 (PDB file 1LJR; 80 in Table 1) (N-terminal domain: 1–78; C-terminal domain: 89–244) [18]. Putative active site residues (a) Arg-13 and (c) Ser-11 are shown in black spacefill.

Download figure to PowerPoint


Ure2p is a dimeric protein of approx. 50 kDa subunits that negatively regulates GATA factor-mediated transcription in S. cerevisiae and thus plays a key role in nitrogen catabolite repression [22]. This protein possesses prion-like characteristics and has similar properties to the vertebrate PrP and Sup35p, another yeast prion, including non-Mendelian inheritance. The N-terminus encodes a functional prion domain (residues 1–93) rich in Asn and Gln while the C-terminal domain (residues 94–354) is responsible for suppression of GATA-factor-mediated transcription (Fig. 2). This latter domain shows a low level (11–20%) of sequence identity to GST but a crystal structure of this domain from S. cerevisiae shows unmistakeable similarity to GST structures (Fig. 2) [17]. Interestingly, residues implicated in substrate binding and catalysis in GSTs seem to have mutated to non-catalytic residues in Ure2p which might explain why it does not catalyse GST-like conjugations or bind GSH. However, S. cerevisiae Ure2p mutants have been found to show greater sensitivity than isogenic wild type strains to a range of GST substrates and compounds generating oxidative stress, so the functional significance of the GST-like domain remains obscure [23]. A study of synthetic peptides of Pi class GST α-helices revealed a common N-capping box motif at the N-terminus of the α-6 helix that is strictly conserved in all GSTs and is also present in both Ure2p and EF1Bγ, suggesting a measure of structural similarity between these otherwise disparate protein classes [24]. The dendrogram in Fig. 1 shows a cluster of proteins encoded by URE2 genes close to cluster 2, a widespread group of fungal GST genes, suggesting a significant measure of sequence identity. A previous bioinformatic analysis revealed that Ure2p is widespread in fungi and suggested that the prion-like domain and GST-like domains have diverged separately [25].

3.4MAK 16

MAK16 is a nuclear protein that plays a role in both cell cycle progression and biogenesis of 60S ribosomal subunits. The MAK16 gene from the parasitic worm Schistosoma mansoni was cloned by chance when a cDNA library was probed with antiserum to affinity-purified S. mansoni GSTs [see references cited in [26]. Subsequently, the corresponding gene from S. cerevisiae was found to have 43% identity and 66% similarity with the S. mansoni MAK16 [27]. The dendrogram in Fig. 1 reveals a group of proteins with 39–47% identity with the MAK 16 sequence of E. cuniculi. All of these proteins would possess a conserved MAK16 domain. The E. cuniculi sequence does not align well with other GSTs although it shows over 30% local alignment identity with human microsomal GST. Again, this suggests a distinct grouping of GST-like proteins in yeast and fungal genomes. The precise functional significance of this group remains unclear but the dendrogram suggests they are clearly distinct from the previous four groupings of GST-like proteins and may be related to Kappa and MAPEG GSTs, thought to have diverged from cytosolic enzymes early in evolution [3,28,29].

3.5Identification and significance of conserved sequence motifs

Conserved sequence motifs identified with the BioEdit sequence editor for each of the five classes of fungal GST-like proteins are shown in Fig. 3. These reveal important points of similarity with the mainstream GSTs and the thioredoxin suprafamily as well as class-specific features unique to each of the classes. The N-terminus contributes to the active site architecture of GSTs and this tends to be the most conserved region in mainstream GST classes [3]. In most cases a Tyr near the N-terminus is conserved in a motif that is characteristic for each class. In the case of the Ure2p group, known to be catalytically inactive, the position of a putative Tyr is changed to Ser-188 while Gly-125 aligns with a Ser responsible for catalysis in the Theta class [18]. A SNAIL/TRAIL motif found in all GSTs except the Kappa class is conserved in three of the five fungal classes (it is absent in cluster 1 and MAK16). A Pro homologous to a cis Pro characteristic of all thioredoxin suprafamily members is present in all fungal classes except MAK16. As mentioned above, an N-cap motif S/TXXD stabilises α-Helix 6 in all GSTs [30] and in our analysis this motif is conserved in all classes except MAK16 (Fig. 3). An associated hydrophobic staple is formed by Leu preceding and Ile succeeding the N-cap motif. This arrangement is known to be especially important for stabilising the structure of GSTs and is clearly evident in Cluster 1. In cluster 2 and EF1Bγ, however, the preceding Leu is replaced by Ile whilst in Ure2p only three sequences contain a preceding Leu/Ile. These similarities and differences support the conclusion that the five classes of GST-like proteins encoded in fungal genomes retain key structural similarities to mainstream enzymes while having clearly deviated in other ways. Differences noted with respect to the MAK16 grouping are not surprising since MAPEG GSTs deviate from mainstream classes in the same way as observed here for MAK16.


Figure 3. Identification of consensus sequences in five classes of fungal GST-like proteins. Conserved residues were identified by alignments in both Clustal W and using the BioEdit sequence editor ( Comparisons of N-terminal consensuses are shown in red while SNAIL/TRAIL and N-capping motifs (where present) are shown in green and blue, respectively. A strongly-conserved motif unique to cluster 2 is shown in purple. The thioredoxin-specific cis Pro is underlined and bold.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

The aim of this study was to explore fungal genomes as a new and growing resource for clarifying our picture of evolutionary and functional relationships among GSTs and related protein families. Our analysis defines five new groups of sequences with similarity to GSTs, three of which have previously been related to GSTs based on sequence alignments and/or biochemical properties. For two of these three classes representative crystal structures are available, which confirm tertiary as well as primary sequence similarity to the GST superfamily.

4.1GSTs: a multifunctional enzyme superfamily

GSTs were originally divided into three main classes: Alpha, Mu and Pi [31]. Later work led to identification of Theta [32], Kappa, Omega, and Zeta [reviewed in [3]] which are widespread in higher organisms. The Sigma class was originally described in squid and showed sequence similarity with cephalopod S-crystallins but later included a helminth GSH-dependent prostaglandin H E-isomerase and a human GSH-dependent prostaglandin synthetase suggesting that this class is biologically widespread. Other classes have been described in non-mammalian sources including bacterial Beta class, plant-specific Phi and Tau classes and insect-specific Class I and III [see references cited in [3]].

In several classes a conserved Tyr residue near the N-terminus is catalytically essential but this is replaced by a Ser in the Theta class and by a Cys forming a mixed disulphide with GSH in the Beta and Omega classes. Whilst in mammals there is substrate specificity within a class, this does not seem to hold up well as a classification criterion in non-mammalian sources. Immunological cross-reaction is observed within classes such that, for example, antisera to rat Alpha class enzymes cross-react with Alpha class GSTs from other species but not with rat Mu or Pi class enzymes. This cross-reaction seems to persist across a surprisingly wide range of species suggesting that overall topology is generally conserved within classes. There have been examples of antisera to GSTs recognising non-GST stress-response proteins, which may suggest an even wider measure of conserved architecture. For example the migration inhibition factor (MIF), which also binds to GSH-agarose, is recognised by Theta class antibodies [original references cited in [3]]. Another classification criterion is formation of stable dimers by subunits within a class but an inability to dimerize with subunits across classes. Representative crystal structures are available for all of the main classes and these explain this curious finding since there appears to be class-specific architecture at the inter-subunit interface. These structures also reveal class-specific features of active sites, which contribute to understanding substrate specificity. Recent publications including a crystal structure and model for the mitochondrial Kappa class suggests that this is quite an ancient class closely related to the bacterial Beta class and to a common thioredoxin/glutaredoxin ancestor [28,29].

As well as catalysing conjugation to GSH, GSTs possess extensive ligand binding properties with non-substrate ligands (e.g. haem, bilirubin, bile salts, dyes, carcinogens, antibiotics) [33]. Frequently this is associated with inhibition of catalytic activity but remarkably little is known about ligand binding sites in most GSTs. A third set of functions involves non-conjugation reactions mediated by GSH including GSH peroxidase, steroid isomerase and isomerization of 13-cis - to all-trans-retinoic acid [reviewed in reference [3]].

4.2Role of GSTs in cell redox status

Redox status is tightly controlled in fungi and several studies have focused on GSTs and related proteins. In an investigation of differential regulation of GST expression in S. pombe, it was found that GST II was induced by both menadione and mercuric chloride, suggesting an involvement of this enzyme in defence against oxidative stress [34]. Several proteins contribute to maintenance of redox balance in fungi. Thioredoxins (Trx1 and Trx2) use NADPH to reduce proteins oxidised by reactive oxygen species (ROS) and also possess ribonucleotide reductase activity while glutaredoxins (Grx1-Grx5) use GSH. Inactivation of either thioredoxin gene does not affect yeast growth but double-mutants are not viable [35]. Interestingly, the N-terminal domain of GSTs is known to follow a thioredoxin-like fold [3]. While there are at least five distinct glutaredoxins in S. cerevisiae these can be divided into two distinct subclasses depending on whether they possess one or two cysteines as part of their active site motif [36]. Both Grx1 and Grx2 (comprising glutaredoxin subclass 1) catalyse conjugation of GSH to 1-chloro-2, 4-dinitrobenzene [37]. Knockout mutations have demonstrated in vivo roles for various glutaredoxins in protection against oxidative stress [36]. Currently, the structural, functional and evolutionary relationships between thioredoxins, glutaredoxins and GSTs remain unclear although all three protein families are thought to function in protection against oxidative stress, sometimes by displaying overlapping activities, and all are members of the thioredoxin “suprafamily” of protein folds.

4.3Evolutionary implications of fungal GST domain structure

Several lines of evidence (e.g. immunoblotting, primary structure analysis) suggest that fungal GSTs are often distinct from mainstream classes found to be widespread in other biological groupings. The phylogenic analysis of recent genome and other sequences presented here (Fig. 1) reveals that several classes of proteins belonging structurally to the GST superfamily are widespread in fungi. In some of these classes there is evidence for combination of GST domains with other structural domains (Fig. 2). Domain “shuffling” is thought to underlie the evolution of many protein families [38]. It is possible that such a mechanism facilitated GSTs becoming involved in a wide range of other tasks at an early stage in the evolution of eukaryotes: the animal-fungus-plant split from a common ancestor is though to have occurred 1070 million years ago. Relatively little is known in detail about the in vivo functions of MAK16 and Ure2p families of GSTs. However it is tempting to speculate that contribution of GSTs to protection against oxidative stress might have led to their recruitment into other stress-response scenaria. It is not yet clear whether these protein families have retained catalytic activity in all cases although the EFIBγ proteins of Y. lipolytica and O. sativa are catalytically active [19,20] while the Ure2p class is not [22]. These protein families need to be investigated further to elucidate their catalytic properties, which may shed light on their current roles and help explain why GST-like domains were needed to carry them out. Lastly, there is a need to determine representative three-dimensional structures especially from clusters 1 and 2 and the MAK16 family. With the availability of crystal structures for fungal GST-like proteins [16,17] and an increase in genome data amenable to bioinformatic analysis, the GST superfamily of fungi presents an excellent target for functional genomics [39]. This could have implications for deeper understanding of fundamental cellular processes such as protein synthesis, defence against oxidative stress and protection against xenobiotics and endogenous toxins in fungi.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

Work in our laboratory is funded by the Higher Education Authority of Ireland Programme for Research in Third Level Institutions. We gratefully acknowledge access to the sequences in the Genoleveures site (Dr. David Sherman, personal communication) [7]. We are also grateful to Prof. Tommie McCarthy, Department of Biochemistry, UCC, for critically reviewing the manuscript.

Appendix A. Supplementary data

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References

Supplementary data associated with this article can be found, in the online version at doi:10.1016/j.femsle.2004.10.033.

Figure 1 Supplementary figures.

Figure 2 Supplementary figures.

Figure 3 Supplementary figures.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgments
  8. Appendix A. Supplementary data
  9. References
  • [1]
    Lage, H. (2003) ABC-transporters: implications on drug resistance from microorganisms to human cancers. Int. J. Antimicrob. Agent 22, 188199.
  • [2]
    Bolhuis, H., van Veen, H.W., Poolman, B., Driessen, A.J.M., Konings, W.N. (1997) Mechanisms of multidrug transporters. FEMS Microbiol Rev. 21, 5584.
  • [3]
    Sheehan, D., Meade, G., Foley, V., Dowd, C.A. (2001) Structure, function and evolution of glutathione transferases: implications for classification of non-mammalian members of an ancient enzyme superfamily. Biochem. J. 360, 116.
  • [4]
    Klein, M., Mamnun, Y.M., Eggmann, T., Schüller, C., Wolfger, H., Enrico Martinoia, E., Kuchler, K. (2002) The ATP-binding cassette (ABC) transporter Bpt1p mediates vacuolar sequestration of glutathione conjugates in yeast. FEBS Lett. 520, 6367.
  • [5]
    Ranson, H., Rossiter, L., Ortelli, F., Jensen, B., Wang, X.L., Roth, C.W., Collins, F.H., Hemingway, J. (2001) Identification of a novel class of insect glutathione S-transferases involved in resistance to DDT in the malaria vector Anopheles gambiae. Biochem. J. 359, 295304.
  • [6]
    Snyder, M.J., Maddison, D.R. (1997) Molecular phylogeny of glutathione S-transferases. DNA Cell Biol. 16, 13731384.
  • [7]
    Sherman, D., Durrens, P., Beyne, E., Nikolski, M., Souciet, J.-L. (2004) Genolevures: comparative genomics and molecular evolution of hemiascomycetous yeasts. Nucleic Acids Res. 32, D315D318.
  • [8]
    Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G. (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 48764882.
  • [9]
    Blackburn, A.C., Woolatt, E., Sutherland, G.R., Board, P.G. (1998) Characterization and chromosomal location of the gene GSTZ1 encoding the human Zeta class glutathione transferase and maleylacetoacetate isomerase. Cytogen. Cell Genet. 83, 109114.
  • [10]
    Fernandez-Canon, J.M., Penalva, M.A. (1998) Characterization of a fungal maleylacetoacetate isomerase gene and identification of its human homologue. J. Biol. Chem. 273, 329337.
  • [11]
    Cha, C.-J., Kim, S.-J., Kim, Y.-H., Stingley, R., Cerniglia, C.E. (2002) Molecular cloning, expression and characterization of a novel class glutathione S-transferase from the fungus Cunninghamella elegans. Biochem. J. 368, 589595.
  • [12]
    Gardiner, D.M., Cozijnsen, A.J., Wilson, L.M., Soledade, M., Pedra, C., Howlett, B.J. (2004) The sirodesmin biosynthetic gene cluster of the plant pathogenic fungus Leptosphaeria maculans. Mol. Microbiol. 53, 13071318.
  • [13]
    Brown, D.W., Yu, J.-H., Kelkar, H.S., Fernandes, M., Nesbitt, T.C., Keller, N.P., Adamis, T.H., Leonard, T.J. (1996) Twenty-five coregulated transcripts define a sterigmatocystin gene cluster in Aspergillus nidulans. Proc. Natl. Acad. Sci. USA 93, 14181422.
  • [14]
    Koonin, E.V., Mushgian, A.R., Tatusov, R.L., Altschuk, S.F., Bryant, S.H., Bork, P., Valencia, A. (1994) Eukaryotic translation elongation factor 1 gamma contains a glutathione transferase domain – study of a diverse, ancient protein superfamily using motif search and structural modelling. Protein Sci. 3, 20452054.
  • [15]
    Bateman, A., Coin, L., Durbin, R., Finn, R.D., Hollich, V., Griffiths-Jones, S., Khanna, A., Marshall, M., Moxon, S., Sonnhammer, E.L.L., Studholme, D.J., Yeats, C., Eddy, S.R. (2004) The Pfam protein families. Nucleic Acids Res. 32, D138D141.
  • [16]
    Jeppesen, M.G., Ortiz, P., Shepard, W., Kinzy, T.G., Nyborg, J., Andersen, G.R. (2003) The crystal structure of the glutathione S-transferase-like domain of elongation factor 1Bγ from Saccharomyces cerevisiae. J. Biol. Chem. 278, 4719047198.
  • [17]
    Bousset, L., Belrhali, H., Janin, J., Melki, R., Morera, S. (2001) Structure of the globular region of the prion protein Ure2 from the yeast Saccharomyces cerevisiae. Structure 9, 3946.
  • [18]
    Rossjohn, J., McKinstry, W.J., Oakley, A.J., Verger, D., Flanagan, J., Chelvanayagam, G., Tan, K.-L., Board, P.G., Parker, M.W. (1998) Human theta class glutathione transferase: The crystal structure reveals a sulphate-binding pocket within a buried active site. Structure 6, 309322.
  • [19]
    Kobayashi, S., Kidou, S., Ejiri, S. (2001) Detection and characterization of glutathione S-transferase activity in rice EF-1βγ and EF-1γ expressed in Escherichia coli. Biochem. Biophys. Res. Commun. 288, 509514.
  • [20]
    Foley, V., Sheehan, D. (1998) Glutathione S-transferases of the yeast Yarrowia lipolytica have unusually large molecular mass. Biochem. J. 333, 839845.
  • [21]
    Hanbauer, I., Boja, E.S., Moskovitz, J. (2003) A homologue of elongation factor 1 gamma regulates methionine sulfoxide reductase A gene expression in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 100, 81998204.
  • [22]
    Cooper, T.G. (2002) Transmitting the signal of excess nitrogen in Saccharomyces cerevisiae from the TOR proteins to the GATA factors: Connecting the dots. FEMS Microbiol. Rev. 26, 223238.
  • [23]
    Rai, R., Tate, J.J., Cooper, T.G. (2003) Ure2, a prion precursor with homology to glutathione S-transferase, protects Saccharomyces cerevisiae cells from heavy metal ion and oxidant toxicity. J. Biol. Chem. 278, 1282612833.
  • [24]
    Dragani, B., Cocco, R., Principe, D.R., Paludi, D., Aceto, A. (2001) Conformational properties of five peptides corresponding to the entire sequence of glutathione transferase domain II. Arch. Biochem. Biophys. 389, 1521.
  • [25]
    Baudin-Baillieu, A., Fernandez-Bellot, E., Reine, F., Coissac, E, Cullin, C. (2003) Conservation of the prion properties of Ure2p through evolution. Mol. Cell Biol. 14, 34493458.
  • [26]
    Milhon, J.L., Albert, T.J., Vande, W.E.A., O'Leary, K.A., Jackson, R.N., Kessler, M.A., Schuler, L.A., Tracy, J.W. (2000) SmMAK16, the Schistosoma mansoni homologue of MAK16 from yeast, targets protein transport to the nucleolus. Mol. Biochem. Parasitol. 108, 225236.
  • [27]
    Wickner, R.B., Koh, T.J., Crowley, J.C., O'Neil, J., Kaback, D.B. (1987) Molecular-cloning of chromosome-I DNA from Saccharomyces-cerevisiae– isolation of the mak16 gene and analysis of an adjacent gene essential for growth at low-temperatures. Yeast 3, 5157.
  • [28]
    Robinson, A., Huttley, G.A., Booth, H.S., Board, P.G. (2004) Modelling and bioinformatics studies of the human kappa-class glutathione transferase predict a novel third glutathione transferase family with similarity to prokaryotic 2-hydroxychromene-2-carboxylate isomerases. Biochem. J. 379, 541552.
  • [29]
    Ladner, J.E., Parsons, J.F., Rife, C.L., Gilliland, G.L., Armstrong, R.N. (2004) Parallel evolutionary pathways for glutathione transferases: Structure and mechanism of the mitochondrial class kappa enzyme rGSTK1-1. Biochemistry 43, 352361.
  • [30]
    Cocco, R., Stenberg, G., Dragani, B., Rossi Principe, D., Paludi, D., Mannervik, B., Aceto, A. (2001) The folding and stability of Human alpha class glutathione transferase A1-1 depend on distinct roles of a conserved N-capping box and hydrophobic staple motif. J. Biol. Chem. 276, 3217732183.
  • [31]
    Mannervik, B., Alin, P., Guthenberg, C., Jensson, H., Tahir, M.K., Warholm, M., Jornvall, H. (1985) Identification of three classes of cytosolic glutathione transferases common to several mammalian species: correlation between structural data and enzymatic properties. Proc. Natl. Acad. Sci. USA 82, 72027206.
  • [32]
    Meyer, D.J., Coles, B., Pemble, S.E., Gilmore, K.S., Fraser, G.M., Ketterer, B. (1991) Theta, a novel class of glutathione S-transferases purified from rat and man. Biochem. J. 274, 409414.
  • [33]
    Listowsky, I., Abramowitz, M., Hanuna, H., Nitu, Y. (1988) Intracellular binding and transport of hormones and xenobiotics by glutathione S-transferase. Drug Metab. Rev. 19, 305318.
  • [34]
    Cho, Y.-W., Park, E.-H., Fuchs, J.A., Lim, C.-J. (2002) A second stress-inducible glutathione S-transferase gene from Schizosacharomyces pombe. Biochim. Biophys. Acta 1574, 399402.
  • [35]
    Muller, E.G.D. (1995) A redox-dependent function of thioredoxin is necessary to sustain a rapid rate of DNA synthesis in yeast. Arch. Biochem. Biophys. 318, 356361.
  • [36]
    Rodriguez-Manzanesque, M.T., Ros, J., Cabiscol, E., Sorribas, A., Herrero, E. (1999) Grx5 glutaredoxin plays a central role in protection against protein oxidative damage in Saccharomyces cerevisiae. Mol. Cell Biol. 19, 81808190.
  • [37]
    Collinson, E.J., Grant, C.M. (2003) Role of yeast glutaredoxins as glutathione S-transferases. J. Biol. Chem. 278, 2249222497.
  • [38]
    Abascal, F., Valencia, A. (2003) Automatic annotation of protein function based on family identification. Proteins 53, 683692.
  • [39]
    Burley, S.K., Bonanno, J.B. (2002) Structuring the universe of proteins. Ann. Rev. Genomics Hum. Genet. 3, 243262.