• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. May 13, 2003; 100(10): 5980–5985.
Published online May 2, 2003. doi:  10.1073/pnas.0931375100
PMCID: PMC156312

A novel class of gene controlling virulence in plant pathogenic ascomycete fungi


Insertional mutants of the fungal maize pathogen Cochliobolus heterostrophus were screened for altered virulence. One mutant had 60% reduction in lesion size relative to WT but no other detectable change in phenotype. Analysis of sequence at the insertion site revealed a gene (CPS1) encoding a protein with two AMP-binding domains. CPS1 orthologs were detected in all Cochliobolus spp. examined, in several other classes of ascomycete fungi, and in animals but not in basidiomycete fungi, bacteria, or plants. Phylogenetic analysis suggested that CPS1 represents a previously undescribed subset of adenylate-forming enzymes that have diverged from certain acyl-CoA ligases, which in bacteria are involved in biosynthesis of nonribosomal peptides or polyketide/peptide hybrids. Disruption of CPS1 caused reduced virulence of both race T and race O of C. heterostrophus on maize, of Cochliobolus victoriae on oats, and of Gibberella zeae on wheat. These results suggest that CPS1 functions as a general fungal virulence factor in plant pathogenic ascomycetes.

Fungi cause more plant disease than any other group of microbes; among fungi, the ascomycetes constitute the largest group of pathogens. Nevertheless, among all ascomycetes, plant pathogenicity is rare. Our goal is to understand the molecular distinction between pathogens and saprobes. One hypothesis is that there must be a common set of pathogenicity/virulence determinants, as well as a set of unique factors that, in sum, contribute to pathogenic ability. There are now plenty of examples of the latter, but few of the former beyond the obvious, such as signal transduction pathway components and cell-wall-degrading enzymes. Here, we describe a virulence factor from the maize pathogen, Cochliobolus heterostrophus, and show that it is common among ascomycetes.

The genus Cochliobolus contains more than 30 species, some of which are saprobes, whereas others are pathogens, usually of grasses (Poaceae; ref. 1). Phylogenetic analysis has revealed that Cochliobolus spp. fall into two distinct groups (2). Group 2 contains 18 species, none of which is notable as a pathogen. In contrast, all members of the genus that are known to cause serious crop diseases, e.g., C. heterostrophus, Cochliobolus carbonum, Cochliobolus victoriae, Cochliobolus sativus, and Cochliobolus miyabeanus are found among the 13 species in group 1. Low levels of nucleotide substitution in internal transcribed spacer (ITS) and GPD1 sequences among members of group 1 indicate that the highly virulent species underwent sudden radiation (2). This observation suggests that a progenitor within the genus Cochliobolus gave rise, over a relatively short period, to a series of distinct biotypes, each distinguished by having unique pathogenic capability to a particular type of plant. This radiation appears to be associated with acquisition of the ability to produce host-specific toxins required for both pathogenesis and host-range determination (3).

Each host-specific toxin is necessary for the development of a particular disease, e.g., T-toxin, a linear polyketide, is required by C. heterostrophus for virulence to maize, and victorin, a cyclized pentapeptide, is required by C. victoriae for pathogenicity to oats. However, ability to produce toxin is not sufficient to explain pathogenic ability entirely. C. heterostrophus, for example, is found in the field as either of two forms: race T, which produces T-toxin and is highly virulent on maize with Texas male sterile (T) cytoplasm, and race O, which does not produce T-toxin but is pathogenic (albeit weakly) on maize with either T cytoplasm or normal (N) cytoplasm (1). On the basis of this observation, it can be hypothesized that the Cochliobolus group 1 progenitor was a weakly virulent pathogen possessing genes for general virulence and pathogenicity.

While seeking general pathogenesis factors by random mutagenesis, we identified a C. heterostrophus gene (CPS1) that is conserved among ascomycete fungi and animals but does not appear to be in basidiomycete fungi, bacteria, or plants. In this paper, we provide genetic evidence that CPS1 is necessary for normal virulence of C. heterostrophus races T and O on maize, C. victoriae on oats, and Gibberella zeae (Fusarium graminearum), a distantly related ascomycete, on wheat; the latter causes wheat head scab, one of the most destructive cereal diseases in North America (4, 5). We also provide phylogenetic evidence that CPS1 encodes an enzyme belonging to the AMP-binding superfamily and is phylogenetically most closely related to a group of acyl-CoA-ligases in bacteria that are at the N-terminal end of nonribosomal peptide synthetase (NRPS) or polyketide synthase (PKS) genes, whose products are nonribosomal peptides, polyketides, or polyketide/peptide hybrids. This evolutionary insight is the starting point for predicting function and for biochemical and structural investigation.‡‡

Materials and Methods

Strains, Media, Crosses, and Transformation.

Race T strain C4 (Tox1+;MAT1-2) and race O strain C5 (Tox1;MAT1-1) are near-isogenic lines of C. heterostrophus (6). R.C4.2696 (cps1;Tox1+;MAT1-2;hygBR) is a C4-derived mutant generated by using restriction enzyme-mediated integration (REMI). Strains 1301-R-45 (cps1;Tox1;MAT1-1;hygBR) and 1301-R-26 (cps1;Tox1+;MAT1-2;hygBR) are progeny of the cross R.C4.2696 × C5. C. victoriae victorin-producing isolate HvW is from a field collection (7). G. zeae strain Z3639 (8) is a field isolate from Kansas. Other fungal isolates used in this report are listed in Table 4, which is published as supporting information on the PNAS web site, www.pnas.org. C. heterostrophus transformation procedures, described previously (9), were also used for C. victoriae and G. zeae. Crossing and culture conditions, methods for testing carbon, and nitrogen utilization are described in Table 5, which is published as supporting information on the PNAS web site.

Recovery of DNA from the REMI Insertion Site.

Plasmids and a detailed description of procedures used for cloning are described in Table 6 and Supporting Materials and Methods, which are published as supporting information on the PNAS web site. Cloning was initiated by recovering REMI vector pUCATPH plus flanking DNA from the C4 (race T) KpnI insertion site, by using plasmid rescue (10). Recovered clone p214B7 was used to disrupt CPS1 in the race T genome to prove that the DNA flanking the REMI insertion site was responsible for the reduced virulence phenotype. To extend sequence to the right of the REMI KpnI insertion site, two successive targeted gene disruption experiments were done in near-isogenic strain C5 (race O) to eliminate possible effects of T-toxin, a known polyketide virulence factor produced by race T (1).

Disruption of CvCPS1 in C. victoriae.

Plasmid p118B14 (Table 6) was linearized with BglII, transformed into victorin-producing isolate HvW, and transformants selected on 80 μg of hygromycin B/ml. Targeted integration of the plasmid into the C. victoriae genome was confirmed by DNA gel blot analysis. Victorin production was examined by HPLC (details available on request from B.G.T.).

Cloning of GzCPS1 from G. zeae and Targeted Gene Disruption.

The G. zeae ortholog of ChCPS1 was cloned by PCR by using procedures described in Supporting Materials and Methods and Table 6. pFgC8-hygB (Table 6) was transformed into Z3936 protoplasts. Targeted disruption of CPS1 was confirmed by DNA gel blot analysis.

DNA and RNA Manipulations.

Fungal genomic DNAs were prepared according to a previously described procedure (11). Plasmid DNA preparations, restriction enzyme digestions, and preparation of DNA gel blots were performed following standard protocols. Methods for DNA–DNA hybridization, DNA sequencing, and cDNA cloning are described in Supporting Materials and Methods.


The microbial assay for T-toxin was described (12). Pathogenicity assays for C. heterostrophus on corn, C. victoriae on oats, and G. zeae on wheat are described in Supporting Materials and Methods.

Comparative Molecular Analyses.

The amino acid sequence of CPS1 was used to query publicly available and Syngenta proprietary (G. zeae, Gibberella moniliformis, and Botrytis cinerea) databases by tblastn and blastp (13). blast hits of e-10 or lower were aligned and then adjusted manually. Conserved functional domains were identified by rps-blast searches (13) by using adenylation (AMP-binding, pfam00501), phosphopantetheine attachment site (pfam00550), and condensation (pfam00668) consensus sequences. clustalw (14) and T-coffee (15) were used to generate amino acid sequence alignments of the AMP binding domains of the CPS1 orthologs and other representatives of the adenylation domain superfamily (16). The two alignments were trimmed at the N- and C-terminal ends to include only well-aligned blocks of amino acids. Trimmed alignments were compared for phylogenetic signal by using paup 4.0b8 (17) with maximum parsimony (100 replicates with random additions of sequences) and bootstrap analysis (334 pseudoreplicates). Gaps were treated as a “21st amino acid.” The T-coffee alignment resulted in a longer most-parsimonious tree with more steps than the tree that resulted from the clustalw alignment. Additionally, the T-coffee-based tree had a higher consistency index and stronger bootstrap support and was therefore used to estimate phylogenetic relationships. The placement of CPS1 in the phylogeny was used to predict its possible function (18).


Characterization of a Weakly Virulent Mutant.

REMI mutant R.C4.2696 caused lesions about 60% smaller than those of WT on both N- and T-cytoplasm maize (Fig. (Fig.11 A and B; for quantitative data, see Fig. 8, which is published as supporting information on the PNAS web site). This phenotype was not associated with the Tox1 locus (Fig. (Fig.11C) or a reduced fitness trait linked to Tox1 (19). Note that high virulence of race T on T-cytoplasm maize, which requires production of T-toxin, also depends on the factor altered in mutant R.C4.2696 (Fig. (Fig.11 B and C), i.e., mutant strains produced WT levels of T-toxin but were less virulent than WT. Mutant and WT were identical in growth rate, pigmentation, spore germination, appressorium formation (Fig. (Fig.22 A and B), and conidiation (not shown). Growth tests on various carbon or nitrogen sources showed no differences between the mutant and WT (Table 5). Progeny of crosses between mutant and WT race T or race O segregated 1:1 for parental types only: all hygromycin B-resistant progeny produced small lesions similar to those of the mutant parent; all hygromycin B-sensitive progeny produced WT lesions, indicating the mutant phenotype was caused by REMI vector insertion at a single site. This result was confirmed by targeted gene disruption by using plasmid p214B7 (Table 6); homologous integration resulted in the original REMI mutant phenotype (Fig. (Fig.3).3).

Figure 1
REMI mutant R.C4.2696 has reduced virulence on corn. (A) Virulence test on N-cytoplasm corn. The mutant R.C4.2696 (Right) produced smaller lesions (compare those at arrows) than WT strain C4 (Left). (B) T-cytoplasm corn inoculated with (Left to Right ...
Figure 2
Mutant R.C4.2696 has normal growth (A) and appressorium formation (B). (A) Plates containing complete medium with xylose (Upper) or minimal medium (Lower) were inoculated with mutant (Right) or WT (Left) and incubated in the light at 22°C for ...
Figure 3
Targeted disruption of CPS1 in WT race T restored the original REMI mutant phenotype. (Upper) Corn plants (N cytoplasm) were inoculated with (Left to Right): REMI mutant R.C4.2696; WT race T (C4); WT race O (C5), and five race T disruptants obtained using ...

Analysis of DNA at the REMI Vector Insertion Site.

DNA (11.3 kb) surrounding the REMI insertion site contained a 5.4-kb ORF designated CPS1 (Fig. (Fig.44A). The ATG at nucleotide position 826 (GenBank accession no. AF332878) (CACCATGCT) initiates an ORF that starts 521 bp upstream of the KpnI REMI vector insertion site (arrowhead) and ends at nucleotide position 6209. cDNA sequencing confirmed the presence of three introns.

Figure 4
Identification of CPS1. (A) Genomic DNA on both flanks of the REMI vector insertion site (solid arrowhead at KpnI site) in the mutant R.C4.2696 genome was recovered by two successive cycles of plasmid rescue. A partial restriction enzyme map of the 11.3-kb ...

Comparative Analysis of CPS1.

blastp and tblastn searches using the deduced CPS1 protein sequence as query revealed high similarities to hypothetical proteins from fungi, i.e., the saprophytic pyrenomycete Neurospora crassa, the human pathogenic plectomycete Aspergillus fumigatus, the hemiascomycetes Candida albicans, and Saccharomyces cerevisiae, and two proteins from the archaeascomycete Schizosaccharomyces pombe and from animals, i.e., Drosophila melanogaster, Caenorhabditis elegans, Mus musculus, and Homo sapiens (Table (Table1;1; only CPS1 homologues in the GenBank database are shown).

Table 1
Sequence homologues of the predicted ORF1 protein

blast searches found weaker similarities to the N-terminal acyl Co-A ligase-encoding modules of bacterial NRPSs, including bleomycin synthetase BlmVI from Streptomyces verticillus and saframycin Mx1 synthetase SafB from Myxococcus xanthus (Table (Table1).1). Similarly, N-terminal acyl Co-A ligase-encoding modules were found in bacterial PKSs and NRPS/PKS hybrid genes (data not shown).

The fungal and animal blast hits are all proteins of unknown function, although the fungal proteins SpYD22, SpYA84, and ScYor093C were annotated as probable membrane proteins, based on the presence of putative transmembrane domains. In our analysis, however, computational predictions (PredictProtein, Columbia University, http://cubic.bioc.columbia.edu/predictprotein/submit_def.html) and TMHMM (prediction of transmembrane helices in proteins, Technical University of Denmark, http://www.cbs.dtu.dk/services/TMHMM/) did not predict transmembrane helices for CPS1 or any of the three yeast proteins.

Protein domain searches against the Pfam database revealed that CPS1 and all significant blast hits contain AMP-binding domains (Pfam00501; Fig. Fig.44B and Fig. 9, which is published as supporting information on the PNAS web site) characteristic of adenylate-forming enzymes that include NRPSs and various CoA ligase/synthetases such as acyl CoA ligase. Two AMP-binding domains (designated CPS1A and CPS1B) are recognizable in CPS1 and in fungal and animal homologs. This modular organization is similar to those of SafB, BlmVI, and other NRPSs (Fig. (Fig.44B).

Amino acid alignment of the putative AMP-binding domains of CPS1 and its fungal and animal orthologs confirmed the presence of conserved core sequences (cores 1–5) found in AMP domain-containing enzymes (2022) (Fig. 9). AMP domain cores 3 and 4 are well conserved in CPS1 except that, in core 4 of domain A, there is a leucine instead of an aspartic acid, and in core 4 of domain B, there is a tyrosine instead of an aspartic acid that is present in the animal orthologs. Cores 1, 2, and 5 are weakly conserved compared with the adenylation domain consensus (Fig. 9).

Although database searches did not reveal other functional domains, manual alignment of amino acids between CPS1A and CPS1B suggested that CPS1 has a thiolation (T, also called peptidyl carrier protein) domain that is also found in all NRPSs 3′ of each AMP domain (21, 22) (Fig. 9). This domain is found only after CPS1A, not after CPS1B.

Phylogenetic analysis of AMP-binding domains indicated that CPS1 and the strong blast hits form a previously undescribed clade of eukaryotic proteins so far known only in ascomycete fungi and animals (Fig. (Fig.55A and Fig. 10, which is published as supporting information on the PNAS web site). There are two subclades, one with the CPS1A modules from all orthologous ORFs, and the other with the CPS1B modules, suggesting a duplication before the common ancestor of fungi and animals (Fig. (Fig.55B). The CPS clade has a common origin with the weaker blast hits (Fig. 10), i.e., the N-terminal modules of several proteins involved in biosynthesis of bacterial NRPSs (e.g., BlmVI and SafB), polyketides (e.g., Streptomyces coelicolor, GenBank accession no. CAA16183), and polyketide/peptide hybrids (e.g., Bacillus subtilis mycA, GenBank accession no. AAF08795). These N-terminal modules encode acyl CoA ligases. CPS1 in C. heterostrophus and other filamentous ascomycetes is extended at the 3′ end relative to orthologs in other organisms (Fig. (Fig.44B). There is no conservation in this extension except for a high percentage of polar amino acids. The NRPS/PKS-associated CoA ligases and closely related members of the CPS clade form a superclade that is distantly related to previously described members of the AMP-binding domain superfamily (Figs. (Figs.55A and 10). This latter group divides into two clades identified previously (23); one clade includes the NRPSs and closely related α-aminoadipic acid reductases, and the other includes several groups of CoA ligase/synthetases including acetyl-, phenylacetyl-, long chain fatty acid-, and 4-coumarate-CoA ligases. The sequence of CPS1 has been deposited in the GenBank database (accession no. AF332878).

Figure 5
Phylogenetic analysis of CPS1. (A) CPS belongs to the adenylate-forming enzyme superfamily, as shown in this simplified rendering, which highlights the relationships of the major enzyme families. Numbers are percentages of bootstrap support for each clade. ...

Phylogenetic analysis also suggests that CPS gene duplication and divergence occurred early in ascomycete evolution, forming CPS1 and CPS2 clades (Figs. (Figs.55B and 10). Only CPS1 is retained in the filamentous ascomycetes, only CPS2 is retained in the hemiascomycetes, and both have been retained in the archaeascomycetes, represented here by S. pombe. CPS1 is more highly conserved among ascomycetes than is CPS2. For example, SpCPS1 (YD22, deduced protein sequence of 1,517 amino acids) has 47% identity and 63% similarity to the ChCPS1 protein for 1,430 alignable amino acids. In contrast, SpCPS2 (YA84, deduced protein sequence of 1,428 amino acids) has only 27% identity and 46% similarity to the ChCPS1 protein for 1,403 alignable amino acids (Table (Table1).1). Overall, these data indicate that CPS1 is highly conserved between both saprobic and pathogenic ascomycetous fungi and is likely involved in production of a similar molecule in all these fungi. On the basis of these data, we conclude that CPS1 represents a previously undescribed group of eukaryotic adenylate-forming enzymes.

Conservation of CPS1 Among Fungi.

Genomic DNAs from 31 fungal species, representing 13 ascomycete genera in the loculoascomycetes and the pyrenomycetes, hybridized when probed with CPS1 (Table 4 and Fig. 11, which is published as supporting information on the PNAS web site). All 16 Cochliobolus species, including apparent saprophytes (e.g., Cochliobolus homomorphus and Cochliobolus peregianensis), known plant pathogens (e.g., C. carbonum, C. victoriae, C. miyabeanus, and C. sativus), and representatives of five loculoascomycete genera closely related to Cochliobolus (i.e., Pyrenophora, Setosphaeria, Bipolaris, Stemphylium, and Alternaria) showed hybridization intensities comparable to that seen with C. heterostrophus itself (Fig. 11). DNAs of five distantly related pyrenomycete genera (e.g., Magnaporthe grisea and G. zeae) hybridized weakly to CPS1 (Fig. 11B), whereas no signal was detected in DNA of the basidiomycete Ustilago maydis (data not shown). In addition, DNA gel blots (not shown) of the saprobic fungus N. crassa hybridized with a CPS1 probe. blast searches confirmed the presence of CPS1 orthologs in the genomes of M. grisea, G. zeae, N. crassa, G. moniliformis, B. cinerea, and A. fumigatus and absence in the genomes of the basidiomycetes Phanaerochaete chrysosporium and Cryptococcus neoformans, suggesting that Basidiomycota have lost CPS1.

Targeted Disruption of C. victoriae CPS1.

CvCPS1 was disrupted by transformation of a victorin-producing isolate (HvW) with p118B14 (Table 6). Gel blot analysis showed that transformants Tx2, Tx5, Tx7, and Tx8 sustained disruptions of CvCPS1. Ectopic integration of the vector occurred in transformants Tx4 and Tx9. Cvcps1 disruptants (but not transformants with ectopic integrations) had reduced virulence on susceptible oats (Fig. (Fig.66 and Table Table2),2), although they produced WT levels of victorin in culture as determined by HPLC analysis (Fig. 12, which is published as supporting information on the PNAS web site). This result is reminiscent of the effect of the cps1 mutation on virulence of T-toxin-producing C. heterostrophus (Fig. (Fig.11 B and C), i.e., victorin production alone in C. victoriae is not sufficient for normal pathogenesis on oats; CPS1 function is needed as well. The same fungal growth tests were done for Cvcps1 mutants as for Chcps1 mutants (Fig. (Fig.2,2, Table 5), and no differences from WT growth rate were observed (data not shown).

Figure 6
Disruption of C. victoriae CPS1 causes reduced virulence. Inoculations: (Left to Right) WT (HvW), Tx7, Tx2, Tx9, and water. Oat seedlings inoculated with WT or Tx9 (an ectopic transformant) were completely killed; those inoculated with Tx7 and Tx2 (disruptants) ...
Table 2
Virulence of C. victoriae isolates to oats

Targeted Disruption of G. zeae CPS1.

The GzCPS1 deduced protein has 68% identity and 80% similarity to ChCPS1 in a 1,472 alignable amino acid overlap, as well as the same motif sequences and domain organization as ChCPS1. The nucleotide sequence of GzCPS1 has been deposited in the GenBank database (accession no. AY267334). Gel blot analysis confirmed that CPS1 disruption occurred in transformants Tx4, Tx10, and Tx11 and ectopic integration occurred in transformants TxB1 and TxH1 (data not shown). Disruption of G. zeae CPS1 resulted in reduced virulence of the Gzcps1 mutants (Fig. (Fig.77 and Table Table3).3). Carbon and nitrogen utilization was the same for WT and mutant, except that the mutant grew more slowly than WT on nitrate-containing medium (not shown).

Figure 7
Disruption of G. zeae CPS1 causes reduced virulence. Left to Right: WT, Tx4, Tx10, TxH1, Tx11, and water. Wheat heads inoculated with WT or TxH1 (ectopic transformant) were completely bleached, whereas those inoculated with Tx4, Tx10, and Tx11 (disruptants) ...
Table 3
Virulence of G. zeae isolates to wheat


The Role of CPS1 in Pathogenesis.

Fungal plant pathogenesis is a complex biological process requiring diverse pathogenicity and virulence factors. These factors tend to fall into two categories: specific and general. Specific factors, e.g., host-specific toxins (1), detoxification enzymes (24), and avirulence proteins (25, 26), are usually restricted to certain species or even to a particular race. Genes controlling these factors are not broadly conserved. In contrast, genes for general factors, e.g., signal transduction pathway components (27), cell-wall-degrading enzymes (2830), and melanin (31) are widely conserved among species and among large taxonomic groupings. The extensive conservation of CPS1 and results of targeted gene disruption experiments conducted in both Cochliobolus spp. and G. zeae argue that the CPS1-controlled product may be a general virulence factor.

Unlike the general factors mentioned above, however, the role of CPS1 in pathogenesis is not obvious. None of the CPS1 homologs previously deposited in GenBank has a known function. Expression of S. pombe YD22 (SPAC56F8.02, an ortholog of CPS1) was up-regulated by heat shock and down-regulated by hydrogen peroxide treatment and during meiosis (www.sanger.ac.uk/perl/SPGE/geexview?group = 2&q = SPAC56F8.02). The S. pombe YA84 gene (SPAC22F3.04, a paralog of CPS1) was up-regulated during meiosis but not affected by stress conditions (www.sanger.ac.uk/perl/SPGE/geexview?q = SPAC22F3.04). S. cerevisiae YOR093C (a paralog of CPS1) expression was up-regulated by stress conditions such as heat shock or stationary growth (32), but functional analysis showed that the yor093c null mutant is viable and did not show an altered phenotype (33). The human Y934 gene is expressed at low to moderate levels and shows tissue specificity, e.g., moderate in heart and brain and low in lung and pancreas (34). The C. elegans F28B3.1 gene is developmentally regulated and abundant only in adults (35). These data suggest that CPS1 and CPS1-like proteins may function in response to various environmental or developmental cues but actual function depends on the organism involved. For the fungi in our study, CPS1 function is required for normal virulence to plants.

Null mutations in ascomycete CPS1 genes cause no discernable growth phenotype, indicating this gene is not required for primary metabolism. The yeast gene expression data mentioned above raised the possibility that this group of proteins might play a role in fungal survival under stress conditions. For fungal pathogens, coping with various stresses, e.g., oxidative burst, iron sequestration, and antimicrobial activity imposed by host defense mechanisms, is essential for pathogen ingress (36). It is notable that cps1 mutants of host-specific toxin-producing fungi are reduced in virulence despite their ability to continue producing WT levels of toxin. This observation suggests that these toxins are not sufficient to combat host-generated stresses that are inhibitory to fungal growth in planta, and that CPS1 function is required to overcome host defense. The conservation of CPS1 in both pathogens and saprobes might reflect a general need of all aerobes to tolerate stresses present in natural environments. This would explain why saprobes like S. pombe and N. crassa maintain copies of CPS1.

CPS1 Belongs to a Superfamily of Adenylate-Forming Enzymes.

The most prominent feature of the CPS1-encoded protein is the presence of AMP-binding domains, diagnostic of a large group of proteins collectively called adenylate-forming enzymes. These enzymes activate carboxylic substrates via an ATP-dependent covalent binding of AMP to form acyl adenylates required for various biochemical reactions (20, 21). Some adenylate-forming enzymes function alone, e.g., insect luciferase, acyl-CoA or aryl-CoA ligases, 4-coumarate-CoA ligase, long chain fatty acid-CoA ligase, acetyl-CoA synthetase and other closely related synthetases, and α-amino adipate reductase. These autonomous proteins are relatively small (400–700 amino acids) and have one module with one AMP-binding domain. In contrast, other adenylate-forming enzymes, e.g., NRPSs, are large multimodular complexes (2022, 37, 38).

CPS1 has two AMP-binding domains, suggesting similarity to NRPSs. However, phylogenetic analyses distinguish CPS1 from NRPSs, or any other known adenylate-forming enzyme. CPS1 lacks the conserved thiolation and condensation domains characteristic of typical NRPSs (21, 22). Phylogenetic analyses indicate that CPS1 is a member of a previously uncharacterized clade of eukaryotic genes whose closest relatives are a clade of bacterial acyl CoA ligases found in the N-terminal position of NRPSs, PKSs, and hybrid genes. Establishing the phylogenetic placement of CPS1 gives us an advantage in understanding possible biochemical function, an ongoing project. Because the enzymatic activities and substrate specificities of the bacterial acyl-CoA ligases have not yet been well characterized, we expect discovery of the biochemical function of CPS1 will be a significant challenge.

Nevertheless, the value of the combined molecular genetic and comparative phylogenomic approach used here for gene annotation cannot be overemphasized. Without the latter, the discovery that CPS1 and CPS1-like proteins comprise a family of previously uncharacterized adenylate-forming enzymes would not have been possible. Although the enzymatic function of CPS1 remains to be identified, genetic tests indicate that CPS1 is important for virulence in at least two independently evolved plant pathogenic fungi (i.e., Cochliobolus spp. and G. zeae). However, the broad distribution of CPS1 in all filamentous ascomycetes also suggests that this gene plays a role in some fundamental aspect (e.g., adaptation to stress conditions) of all filamentous ascomycetes, a role that is also necessary as a virulence factor.

Supplementary Material

Supporting Information:


We thank colleagues indicated in Table 4 for providing fungal isolates or DNA samples. This work was supported by grants to O.C.Y. and B.G.T. from the U.S. Department of Agriculture. S.K. was supported by Torrey Mesa Research Institute funds to Louise Glass and John Taylor (Plant and Microbial Biology Department, University of California, Berkeley).


restriction enzyme-mediated integration
nonribosomal peptide synthetase
polyketide synthase


Data deposition: The sequences reported in this paper have been deposited in the GenBank database [accession nos. AF332878 (ChCPS1) and AY267334 (G2CPS1)].

‡‡Sequence data may be found as follows: for C. albicans at www-sequence.stanford.edu/group/candida; for N. crassa and M. grisea at www.genome.wi.mit.edu/annotation/fungi/neurospora and www-genome.wi.mit.edu/annotation/fungi/magnaporthe; for A. fumigatus at www.tigr.org; for C. neoformans at http://valefor.stanford.edu/group/Cneoformans/overview.html; and for P. chrysosporium at www.jgi.doe.gov.


1. Yoder O C, Macko V, Wolpert T J, Turgeon B G. In: The Mycota. Carroll G, Tudzynski P, editors. Vol. 5. Berlin: Springer; 1997. pp. 145–166.
2. Berbee M L, Pirseyedi M, Hubbard S. Mycologia. 1999;91:964–977.
3. Turgeon B G, Lu S-W. In: Fungal Pathology. Kronstad J W, editor. Dordrecht, The Netherlands: Kluwer; 2000. pp. 93–126.
4. Booth C. In: Fusarium: Disease, Biology, and Taxonomy. Nelson P E, Toussoun A T, Cook R J, editors. University Park, PA: Pennsylvania State Univ. Press; 1981. pp. 446–452.
5. Sutton J C. Can J Plant Pathol. 1982;4:195–209.
6. Leach J, Lang B R, Yoder O C. J Gen Microbiol. 1982;128:1719–1729.
7. Christiansen S K, Wirsel S, Yun S-H, Yoder O C, Turgeon B G. Mycol Res. 1998;102:919–929.
8. Bowden R L, Leslie J F. Exp Mycol. 1992;16:308–315.
9. Turgeon B G, Bohlmann H, Ciuffetti L M, Christiansen S K, Yang G, Schafer W, Yoder O C. Mol Gen Genet. 1993;238:270–284. [PubMed]
10. Yang G, Rose M S, Turgeon B G, Yoder O C. The Plant Cell. 1996;8:2139–2150. [PMC free article] [PubMed]
11. Yoder O C. In: Genetics of Plant Pathogenic Fungi. Sidhu G S, editor. Vol. 6. San Diego: Academic; 1988. pp. 93–112.
12. Ciuffetti L M, Yoder O C, Turgeon B G. Fungal Genet Newslett. 1992;39:18–19.
13. Altschul S F, Madden T L, Schaeffer A A, Zhang J, Zhang Z, Miller W, Lipman D J. Nucleic Acids Res. 1997;25:3389–3402. [PMC free article] [PubMed]
14. Thompson J D, Higgins D G, Gibson T J. Nucleic Acids Res. 1994;22:4673–4680. [PMC free article] [PubMed]
15. Notredame C, Higgins D G, Heringa J. J Mol Biol. 2000;302:205–217. [PubMed]
16. Turgay K, Marahiel M A. Pept Res. 1994;7:238–241. [PubMed]
17. Swofford D L, Waddell P J, Huelsenbeck J P, Foster P G, Lewis P O, Rogers J S. Syst Biol. 2001;50:525–539. [PubMed]
18. Thornton J W, DeSalle R. Annu Rev Genomics Hum Genet. 2000;1:41–73. [PubMed]
19. Klittich C R J, Bronson C R. Phytopathology. 1986;76:1294–1298.
20. Stachelhaus T, Marahiel M A. J Biol Chem. 1995;270:6163–6169. [PubMed]
21. von Döhren H, Keller U, Vater J, Zocher R. Chem Rev. 1997;97:2675–2705. [PubMed]
22. Marahiel M A. Chem Biol. 1997;4:561–567. [PubMed]
23. Turgay K, Krause M, Marahiel M A. Mol Microbiol. 1992;6:529–546. [PubMed]
24. Bouarab K, Melton R, Peart J, Baulcombe D, Osbourn A. Nature. 2002;418:889–892. [PubMed]
25. Lauge R, DeWit P J G M. Fungal Genet Biol. 1998;24:285–297. [PubMed]
26. Jia Y, McAdams S A, Bryan G T, Hershey H P, Valent B. EMBO J. 2000;19:4004–4014. [PMC free article] [PubMed]
27. Xu J-R. Fungal Genet Biol. 2000;31:137–152. [PubMed]
28. ten Have A, Mulder W, Visser J, Van K J A L. Mol Plant–Microbe Interact. 1998;11:1009–1016. [PubMed]
29. Reignault P, Kunz C, Delage N, Moreau E, Vedel R, Hamada W, Bompeix G, Boccara M. Mycol Res. 2000;104:421–428.
30. Isshiki A, Akimitsu K, Yamamoto M, Yamamoto H. Mol–Plant Microbe Interact. 2001;14:749–757. [PubMed]
31. Henson J M, Butler M J, Day A W. Annu Rev Phytopathol. 1999;37:447–471. [PubMed]
32. Gasch A P, Spellman P T, Kao C M, Carmel-Harel O, Eisen M B, Storz G, Botstein D, Brown P O. Mol Biol Cell. 2000;11:4241–4257. [PMC free article] [PubMed]
33. Winzeler E A, Shoemaker D D, Astromoff A, Liang H, Anderson K, Andre B, Bangham R, Benito R, Boeke J D, Bussey H, et al. Science. 1999;285:901–906. [PubMed]
34. Nagase T, Ishikawa K, Suyama M, Kikuno R, Hirosawa M, Miyajima N, Tanaka A, Kotani H, Nomura N, Ohara O. DNA Res. 1999;6:63–70. [PubMed]
35. Jiang M, Ryu J, Kiraly M, Duke K, Reinke V, Kim S K. Proc Natl Acad Sci USA. 2001;98:218–223. [PMC free article] [PubMed]
36. Wolpert T J, Dunkle L D, Ciuffetti L M. Annu Rev Phytopathol. 2002;40:251–285. [PubMed]
37. Cane D E, Walsh C T, Khosla C. Science. 1998;282:63–68. [PubMed]
38. Walsh C T, Chen H, Keating T A, Hubbard B K, Losey H C, Luo L, Marshall C G, Miller D A, Patel H M. Curr Opin Chem Biol. 2001;5:525–534. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • EST
    Published EST sequences
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...