SEARCH

SEARCH BY CITATION

Keywords:

  • evolution;
  • fungi;
  • gene family;
  • phylogenomics;
  • serine proteases

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Using a phylogenomic approach with 10 fungi of very different virulence and habitat, we determined that there was substantial diversification of subtilase-type proteases early in ascomycete history (with subsequent loss in many lineages) but with no comparable diversification of trypsins. Patterns of intron loss and the degree of divergence between paralogues demonstrated that the proliferation of proteinase K subtilases and subtilisin type subtilases seen in pathogenic ascomycetes (Metarhizium anisopliae, Magnaporthe grisea, Fusarium graminearum) occurred after the basidiomycete/ascomycete split but predated radiation of ascomycete lineages. This suggests that the early ascomycetes had a lifestyle that selected for multiple proteases, whereas the current disparity in gene numbers between ascomycete lineages results from retention of genes in at least some pathogens that have been lost in other lineages (yeasts, Aspergillus nidulans, Neurospora crassa). A similar prevailing trend towards lineage specific gene loss of trypsins in saprophytes and some pathogens suggests that their phylogenetic breadth will have been much wider in early fungi than currently.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

The serine endoproteases are divided into two superfamilies that independently evolved similar catalytic mechanisms. The trypsin superfamily includes the trypsins and chymotrypsins that are ubiquitous in animals. The subtilase superfamily is similarly ubiquitous in bacteria and fungi. The classification of Siezen & Leunissen (1997) delineates three families of subtilases in fungi, the proteinase K family (fungal or class II) named after an enzyme found in the ascomycete Tritirachium album, the subtilisin family (also called bacterial or class 1) which until recently were only thought to be found in bacteria, and kexins (preprotein convertases) that are also found in animals.

Fungi depend for their life activities on their ability to harvest nutrients from living or dead plant and animal material. The ecological diversification of fungi is therefore profoundly affected by the array of enzymes they secrete. Thus, fungi pathogenic to plants and animals show adaptation in the range of enzymes produced to the polymers present in the integuments of their particular hosts; the ability to colonize these taxa among different fungal species being linked to enzyme evolution (St. Leger et al., 1997). Among many saprophytes, subtilases are the principal broad-spectrum proteases (Gunkle & Gassen, 1989). Their ubiquity among fungi suggests that they are unlikely to be specifically developed to implement pathogenicity. In fact, during the evolution of many plant pathogens the functions of a broad-spectrum protease are performed by trypsins (Murphy & Walton, 1996; St. Leger et al., 1997; Bidochka et al., 1999). Transcripts of two trypsin genes are also the most highly expressed by the insect pathogen Metarhizium anisopliae under some conditions (Freimoser et al., 2003a).

Most fungal genes encoding hydrolytic enzymes are members of gene families-genes of common origin that encode products of similar function (Walton, 1996). However, the contribution gene families make to ecological diversification and the nature of the evolutionary forces acting during this process remain poorly known. In part this is because genes directly involved in ecological attributes are hard to identify (Duda & Palumbi, 1999).

Fungal protease genes should provide a good model for studies of adaptive evolution of multigene families. The majority are secreted directly into the environment, and so can prima facie be defined as ecological traits, their production is repressed by readily utilized nutrients but induced by proteins and there are clear selectable differences in their substrate specificities (St. Leger et al., 1987; Rawlings & Barrett, 1995; Murphy & Walton, 1996; Siezen & Leunissen, 1997). Thus gene divergence and expression in genetically and ecologically distinct species can be more readily interpreted in terms of adaptation. For example, increases and decreases in protein family sizes in fungi could correlate with differences in function that are indicative of adaptations to environment and life strategies. Metarhizium anisopliae produces at least 11 subtilisins; to date the largest number reported from any fungus (Bagga et al., 2004) but only two trypsins (Freimoser et al., 2003a). Although it is not clear which specific protease is responsible for any given activity, either scavenging for nutrients or pathogenicity related, strong selective constraint against mutations in active site regions of each enzyme confirms that they are not redundant and each must have a role (Bagga et al., 2004).

Sequences of subtilisin and trypsin homologues continue to pour into sequence databases, including from genome projects. At the time this work was submitted there were nine virtually completed sequenced fungal genomes representing six ascomycete and three basidiomycete lineages [Goffeau et al., 1996; Wood et al., 2002; Galagan et al., 2003, Genome sequencing projects for Aspergillus nidulans, Coprinus cinereus, Cryptococcus neoformans, Fusarium graminearum, Magnaporthe grisea, Neurospora crassa (version 3) and Ustilago maydis, Center for Genome Research (http://www.broad.mit.edu)]. These fungi have very different virulence and habitat that provide a formidable new resource for determining the identity, origin and evolution of traits that contribute significantly to fitness.

In this study we used genomic sequences and expressed sequence tag (EST)/polymerase chain reaction (PCR) data from a project designed to identify the full spectrum of proteases in M. anisopliae (Freimoser et al., 2003a; Bagga et al., 2004), to identify likely evolutionary events such as patterns of gene duplications and gene loss in the history of the subtilase and trypsin superfamilies in fungi. We reasoned that these events would likely reflect specialization and adaptation. The phylogenomics approach, combining phylogenetic reconstruction and analysis of complete genomic sequences (Eisen & Fraser, 2003), allows estimates of the age of duplication events, which in turn can greatly aid in functional studies (i.e. recent duplications suggest the expansion of an activity in a species, old duplications likely reflect divergent functions) (Heidelberg et al., 2000; Jordan et al., 2001). In addition, in order to reconstruct trypsin evolution, we have related the presence or absence of trypsin genes to the phylogenetic relationship of 35 representative fungi. This is because available sequence data derives from too few fungal species to conduct a comparative analysis on the evolution of gene diversity. The origin of the trypsin superfamily in prokaryotes or eukaryotes is controversial (Rawlings & Barrett, 1995), as until this study trypsin homologues had only been found in actinomycete bacteria, five pathogenic ascomycetes, including M. anisopliae, and animals.

Our study demonstrated that a substantial diversification of subtilase-type proteases occurred early in ascomycete history (with subsequent loss in saprophytic lineages). However, the pathogens retained and occasionally expanded different gene families. Thus, M. grisea has 15 subtilisins and six proteinase K subtilases, whereas M. anisopliae and Fusarium graminearium each possess 11 proteinase K subtilases but three or fewer subtilisins. Trypsin genes are lacking in most saprophytes, but are present in a basidiomycete insect symbiont (Septobasidium canescens), most zygomycetes and many ascomycete plant and insect pathogens. The patchy distribution of trypsins suggests that their phylogenetic breadth will have been much wider in early fungi than currently.

Organisms

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Allomyces macrogynus (ATCC 38327), Mucor mucedo (ATCC 38694), S. canescens (ATCC 20021), Sporobolomyces roseus (ATCC 24257), C. cinereus (ATCC 20120) and Leptosphaeria taiwanensis (ATCC 38203) were obtained from the American type Culture Collection. Conidiobolus coronatus (ARSEF512), C. lamprauges (ARSEF), C. obscurus (ARSEF133), Neozygites parvispora (ARSEF320), Zoophthora radicans (ARSEF1341), Basidiobolus ranarum (ARSEF264), Paecilomyces fumosoroseus (ARSEF5540), Hirsutella thompsonii (ARSEF194) and M. anisopliae sf. anisopliae (ARSEF2575) were obtained from the US Department of Agriculture Entomopathogenic Fungus Collection in Ithaca, NY Rhizopus stolonifer (ER-15-6223), Penicillium chrysogenum (=Penicillium notatum) (ER-15-6157), Sphaerostilbella lutea (GJS 82-274), Hypomyces chrysospermus (GJS 97-173), Hypocrea gelatinosa (CBS 887-72), Diaporthe arctii (AR2831) and Bionectria ochroleuca (GJS90-167) were obtained from Dr Mary Rossman at the US Department of Agriculture Fungus Collection in Beltsville, MD, USA. Cultures were maintained on Sabouraud dextrose agar (DIFCO, Franklin Lakes, NJ, USA). Agaricus bisporus mushrooms were obtained from a local market.

Trypsin sequences from fungi

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Mushrooms (A. bisporus) were used directly as a source of genomic DNA. Otherwise, genomic DNA from 2-day-old Sabouraud dextrose broth cultures of each fungus was isolated as described (Screen & St. Leger, 2000). Degenerate primers based on an alignment of all available fungal trypsins were used as templates for PCR. They were forward GTISTIACYGCNGSYCAYTG and reverse AKIGGRCCICCRDWRTCDCC (where I = Inosine; S = C,G; Y = C,T; N = A,G,C,T; K = G,T; R = A,G; D = A,G,T; W = A,T). Standard cycling conditions following optimization were 94 °C for 4 min, one cycle; 94 °C for 1 min, 45–48 °C for 1 min, 72 °C for 1 min, 32 cycles; 72 °C for 7 min, one cycle. PCR reactions were analysed on gels, the bands of interest excised, purified with a Qiagen gel extraction kit (Qiagens Science, Germantown, MD, USA), and the PCR products sequenced (see Fig. 4 for accession numbers). To confirm the absence of trypsins, we probed Southern blots with a mixture of trypsin fragments from M. anisopliae, C. coronatus and S. canescens. Blots were performed at moderate stringency using standard procedures (St. Leger et al., 1992).

image

Figure 4. (a) A consensus tree showing the relationships among 35 fungi as inferred from 18S rRNA gene sequence data. Bootstrap support (>50%) is provided in the figure. Taxa name in bold means that we PCR amplified at least one trypsin. Asterisks denote validation of the presence of trypsins by genome sequencing. Taxa name underlined means that the genome of the fungus has been sequenced but contains no trypsin genes. Taxa name in uppercase means one or more bands cross-hybridizing with trypsin genes in Southern blot analysis although PCR reactions did not produce trypsin fragments (Fig. 5). (b) Neighbour-joining tree of trypsin proteins showing congruency with the 18s RNA sequences. Bootstrap support (>50%) and accession numbers are provided in the figure. Sequences from Fusarium oxysporum, Cochliobolus carbonum, Trichoderma harzianum and Phaeospaeria nodurum were obtained from public data bases. The accession numbers for the 18s RNA sequences used in (a) are: A. bisporus, AJ244527; A. macrogynus, U23936; A. nidulans, X78539; B. ranarum, AF113414; B. ochroleuca, AH007787; C. carbonum, U42479; C. coronatus, AF113417; C. lamprauges, AF296754; C. obscurus, AF368508; C. coprinus, M92991; C. neoformans, M55625; D. arctii, L36985; F. oxysporum, Z94126; H. thompsonii, U32406; H. gelatinosa, U32407; H. chrysospermus, M89993; L. taiwanesis, U43447; M. grisea, AF277124; M. anisopliae, AF280631; M. mucedo, X89434; N. parvispora, AF296760; N. crassa, AY046271; P. fumoso-roseus, AB032475; P. chrysogenum (=P. notatum), L76153; P. nodorum, U04236; R. stolonifer, AF113441; S. cerevisiae, AF331938; S. pombe, AY046272; S. canescens, AY123320; S. lutea, U32415; S. roseus, X60181; T. harzianum, AF548100; U. maydis, X62396; Z. radicans, D61381. The 18s RNA sequence for F. graminearum was obtained from the F. graminearum excluded reads database at Whitehead Institute web site (http://www-genome.wi.mit.edu/), sequence G578P61595RER (105–1114).

Download figure to PowerPoint

Phylogenetic analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

The sequences of previously characterized trypsins and subtilases were downloaded from the National Center for Biotechnology Information (NCBI) databases using M. anisopliae sequences as query sequences in blast, blast2 and PSI-blast searches (accession numbers or genomic locus ID numbers are given in Figs 1–4). Databases searched included genome sequences of three Basidiomycetes, C. neoformans (an opportunistic pathogen of immunocompromised individuals) C. cinereus (saprophyte), U. maydis (plant pathogen), two saprophytic ascomycetes (A. nidulans, N. crassa) and two plant pathogenic ascomycetes (M. grisea, F. graminearum) at the Whitehead Institute. Databases for Saccharomyces cerevisiae and Schizosaccharomyces pombe were accessed at Stanford University and the Sanger Institute, respectively. DNA and protein sequence alignments were generated with Clustal W (Thompson et al., 1994). Signal peptides were predicted using SignalP V2.0.b2 (Brunak & Von Heijne, 1999) at http://www.cbs.dtu.dk/services/SignalP-2.0. Exons were predicted by alignment with known subtilsins and using the GT/AG rule. Phylogenetic analyses on the aligned amino acid sequences were performed using the neighbour-joining method (Poisson correction) in the MEGA 2.1 (Kumar et al., 2001) program. Confidence in the topology of phylogenetic trees was evaluated by performing 1000 bootstrap replicates in the program. Comparisons of sequence pairs in the alignments for each gene group were made by calculating synonymous (dS) and nonsynonymous (dN) nucleotide substitutions with the Yang et al. (2000) model for codon change in the PAML program (Department of Biology, UCL, London, UK).

image

Figure 1. Neighbour-joining tree of proteinase K gene family protein sequences. Accession numbers for S. cerevisiae, and S. pombe, locus ID (A. nidulans , F. graminearium, M. grisea and N. crassa) or contig numbers (C. neoformans, C. cinereus and U. maydis) as available from the Whitehead Institute web site (http://www-genome.wi.mit.edu/) are provided in the figure. Accession numbers for M. anisopliaePr1A to Pr1K are from Bagga et al. (2004). Genes classifiable into subfamilies 1 (sf1), 2 (sf2) and 3 (sf3) and bootstrap support (>50%) are also shown. Full names of each species are given in the first instance in the figure.

Download figure to PowerPoint

image

Figure 2. Neighbour-joining tree of subtilisin class I gene family protein sequences. Bootstrap support (>50%), locus ID (F. graminearium, M. grisea, N. crassa) or contig numbers (C. cinereus, C. neoformans and U. maydis) (from the Whitehead Institute web site) and accession numbers (for M. anisopliae) are provided in the figure. The 11 intron positions shared by two or more genes are numbered consecutively from 5′ to 3′. Genes classifiable into subfamilies 1 (SF1) and 2 (SF2), and the subfamily of basidiomycete sequences are also shown.

Download figure to PowerPoint

image

Figure 3. Neighbour-joining tree of kexin-like subtilisin protein sequences. Bootstrap support (>50%), locus ID (A. nidulans, F. graminearium, M. grisea and N. crassa) or contig numbers (C. cinereus, C. neoformans and U. maydis) (from the Whitehead Institute web site), and accession numbers (S. cerevisiae, and S. pombe) are provided in the figure.

Download figure to PowerPoint

Phylogenetic analysis of the subtilase superfamily

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Three families of fungal subtilases (Siezen & Leunissen (1997), the proteinase K family (fungal or class II) (Fig. 1), the subtilisin family (bacterial or class 1) (Fig. 2), and kexins (preprotein convertases) (Fig. 3) were included in this study. We placed these families in separate phylogenies, as they are very divergent from each other. Throughout the paper the ascomycete subtilisin subfamilies are abbreviated as SF1 and SF2 to delineate them from the proteinase K subfamilies (sf1 and sf2).

Proteinase K family

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

A previous survey of published subtilases identified three distinct clusters within the proteinase K family in fungi, two subfamilies containing only secreted proteases (sfs1 and 2) and a subfamily that includes endocellular subtilisins (sf3) (Bagga et al., 2004). A similar topology was found in the gene sequences in genomic databases, albeit sf2 receiving weaker bootstrap support (60%) than sf1 (93%). The sf3 group was without bootstrap support when S. pombe sequences were included, but in their absence the S. cerevisiae sequences formed a cluster with the sf3 sequences that received 96% support (data not shown). The presence of at least one sf3 proteinase K like enzyme (Fig. 1) and a bacterial class 1 subtilisin (Fig. 2) in each of the basidiomycetes demonstrates that the last common ancestor before the basidiomycete/ascomycete split [ca. 500 Ma (Berbee & Taylor, 2001)] possessed proteinase K and subtilisin proteases. However, the four proteinase-K subtilases of the basidiomycete C. cinereus are basal to the sf1/sf2 split and cluster together (100% bootstrap support) suggesting that the sf1-sf2 duplication occurred in the Ascomycota. This presumably preceded the filamentous ascomycete/yeast split approximately 400 Ma (Berbee & Taylor, 2001) as S. cerevisiae has a single sf2 enzyme.

At first sight, by the measure of total number of sequences, gene duplication would seem to have occurred considerably more often in pathogenic ascomycetes than in the other groups. However, the majority of sf1 and sf2 subtilases from M. anisopliae, F. graminearium, M. grisea are dispersed throughout the assemblage of sequences indicating that they originated from a common ancestor and that many duplication events preceded speciation of ascomycete lineages. Divergence of these lineages may have been taking place 240 Ma (Berbee & Taylor, 2001). Other sequences clustered with sequences from the same species, suggesting more recent duplications. Consistent with these being recent duplications the lineage specific clusters consist of just two or three genes e.g. F. graminearium sequences 1467 and 1340, and M. anisopliaePr1A, Pr1B and Pr1I.

In a minority of cases it was possible to assign orthologues. This was most straightforward with the three genes from N. crassa. Thus N. crassa sequence 07159 with M. grisea sequence 08966; N. crassa sequence 06066 with F. graminearium sequence 1383 and N. crassa sequence 06949 with F. graminearium sequence 1441 (albeit the last one with low bootstrap support).

The class I subtilisin family

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Aspergillus nidulans, S. pombe and S. cerevisiae have lost subtilisins. Otherwise the genomes contain a single subtilisin gene except for F. graminearium and the rice blast fungus M. grisea that contain three and 15, respectively. All the sequences predict a signal peptide suggesting they are secreted. One of the M. grisea genes is truncated and therefore not considered in the phylogeny (Fig. 2).

The tree contains three clusters; a basidiomycete grouping and two subfamilies of ascomycete sequences (>96% bootstrap support), each containing some M. grisea genes (ascomycete subtilisin subfamilies are abbreviated as SF1 and SF2 to delineate them from the proteinase K subfamilies, sf1 and sf2). Subfamily 1 was less tightly clustered than SF2 and includes a subcluster of sequences from M. anisopliae, N. crassa and F. graminearium. The tree provides evidence for an independent duplication in Fusarium, as the two SF1 F. graminearium sequences, Fg11472 and Fg506572, cluster together with relatively high pairwise similarity (59%). Because SF2 also contains a F. graminearium sequence the tree indicates that the SF1-SF2 duplication preceded radiation of the ascomycete orders, Sordariales (M. grisea and N. crassa) and Hypocreales (F. graminearium and M. anisopliae). We can also infer from their absence the differential loss of SF2 genes in N. crassa and possibly M. anisopliae (inferred from failure to find an EST).

Nevertheless, as most basidiomycetes and ascomycetes possess just one subtilisin, the parsimonious hypothesis based on the data would be that the multigene family in M. grisea originated primarily through duplication at the origin of the genera. Consistent with this, most of the M. grisea sequences clustered with other M. grisea sequences. However, the average percent sequence identity (range) between M. grisea subtilisins in SF1 and SF2 are only 23% (18–33) and 36% (30–45), respectively, which suggests divergence early in ascomycete history. This conclusion is based on the lower level of sequence divergence between Fg11472 and Fg506572 in F. graminearium and the similarity of these sequences to their homologues in M. anisopliae (55%) and N. crassa (48%). In addition, although the six proteinase K M. grisea subtilases are dispersed among sf1, sf2 and sf3 sequences from other genera (Fig. 1) they show similar sequence divergence [36 (25–58)%] to the subtilisins suggesting they duplicated over a similar time scale. In contrast, the four proteinase K genes in C. cinereus (Fig. 1) appear to have diverged comparatively recently, and perhaps within the Coprinus genus itself, judging by the comparatively high sequence similarity between them [54 (48–61)%] and the presence of the two most similar sequences (Cc1277A and Cc1277B) 6000 base pairs apart on the same scaffold, consistent with tandem duplication. In contrast, the two closest M. grisea sequences were >20 000 base pairs apart. Likewise the three N. crassa proteinase K sequences have average identities of 46% (37–67) with their putative orthologues. Consistent with both fungi belonging to the same order, the N. crassa 07159/M. grisea 08966 sequences are 67% similar. Thus, sequences with <46% identity may have diverged before the origin of major ascomycete lineages. It is also likely that purifying selection for functionally important domains will have constrained divergence much below the 35% sequence similarity found in the subtilisins.

Recently duplicated genes could show accelerated evolution by being free to accumulate amino-acid changes toward functional divergence (Graur & Li, 2000). We calculated dN/dS ratios (the ratio of amino acid altering substitutions to silent substitutions), to determine, first, whether dN/dS varies between the subtilisin and proteinase K multigene families in M. grisea and, secondly, whether positive selection governs gene family diversification. The M. grisea subtilisin gene grouping shows an average pairwise dN/dS of 0.742/0.976 = 0.76 ± 0.07. This is similar to the diversification in the proteinase K genes (0.717/0.825 = 0.903 ± 0.08), and provides no evidence for increased selection in the subtilisins.

A further indication that the duplications are not recent is that the dS values for subtilisin genes are close to 1, i.e. divergence has been sufficient to allow saturation of nucleotide substitutions. At this level of divergence it may be difficult to detect positive selection among duplicated genes, simply because positive selection may only occur in a short evolutionary time after gene duplication during the functional shift of the protein, and its effect can be obscured by later substitutions (Zhang et al., 1998).

Structure of M. grisea subtilisin genes

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

Detection of intron losses is a useful tool for interpreting the origin, development, and divergence of multigene families, and for reconstruction of gene evolution (Frugoli et al., 1998). Consistent with M. grisea subtilisins not being recent duplications, their genetic structure was very different as shown by mapping intron insertion sites on the multiple protein sequence alignments. The identities of intron locations were considered stringently. Thus, Fig. 2 shows introns located at the same position in the alignment and with the same phase in the codon. We found only one pair of closely separated (four codons apart) intron sites among genes (indicated by a question mark in Fig. 2). Thus, if intron sliding occurs at all, it must be much rarer than loss or gain of introns.

Two introns in basidiomycete sequences had counterparts in some ascomycete genes. Intron 2 is common to F. graminearium sequences 11472 and 06572 (SF1), and to homologues in the basidiomycetes C. cinereus (Cc152) and U. maydis (Um1103). Intron 5 is common to SF2 genes Mg025313 and Fg06332, and the basidiomycete C. neoformans gene, Cn14. Given that the shared basidiomycete/ascomycete introns are in either SF1 and SF2, and the basidiomycete sequences cluster together (98% bootstrap support) rather than with SF1 or SF2, the simplest explanation for the SF1/SF2 split is that the duplication that gave rise to the progenitors of SF1 and SF2 predated the divergence of basidiomycetes and ascomycetes from a common ancestor. In which case, the ancestral ascomycete inherited two subtilisins that possessed introns 2 and 5, respectively. However, we can also infer from the discontinuity between basidiomycete and ascomycete sequences that most duplication events within SF1 and SF2 occurred after the basidiomycete/ascomycete split. Introns 1, 4 and 7 have counterparts in some sequences from both SF1 and SF2. As they are present in F. graminearium as well as M. grisea, they predate speciation within ascomycetes, confirming some duplication events within both SF1 and SF2 preceding radiation of ascomycete orders with subsequent differential loss of genes in different lineages.

There is extensive evidence for multiple intron loss in the phylogeny. Intron 2, being present only in F. graminearium and basidiomycete genes was presumably lost early in the lineage leading to M. grisea, before the divergence of most sequences, whereas the patchy distribution of introns 1, 4, 5 and 7 provide examples of multiple independent instances of intron loss. Inspite of losses of single introns such as intron 2 in Fg06572 and intron 10 in Mg084363, most genes with few shared introns (e.g. Fg11472, Fg06572, Nc002631, Mg104453, Mg00282, Mg073583, Cc152, Um1103) tend to have those introns at the 5′ termini. Also, in each case, the intron losses were exact without changes in the surrounding codon sequence. These circumstances favour a model for concerted loss of introns through gene conversion with intronless (cDNA) copies of the gene (Hartung et al., 2002). An exception, Mg047333, only retains introns 10 and 11 and so the recombination break point may have lain upstream of intron 10. This mechanism for concerted intron loss may have been more active in some clusters than other, as sf1 paralogues have fewer introns than sequences in sf2.

Following events of intron loss, the appearance of a new intron in a novel position is consistent with a mechanism of intron gain because one gain event is more likely than many independent intron loss events (Dibb & Newman, 1989; Hartung et al., 2002). Most paralogues had one or more unique introns that cannot be aligned clearly to any other gene. This is consistent with ancient gene duplication events, as sufficient time must have elapsed since duplication to allow for intron additions. The exclusive presence of intron 6 in the sub-cluster Mg084153 and Mg038703 also suggests recent gain. However, intron loss seems common and widespread in the subtilisins, which emphasizes the difficulty of reconstructing the evolution of gene structure because of a labile nature of intron presence or absence (Krzywinski & Besansky, 2002).

Kexins

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

The tree topology recovered from the subtilisin data likely reflects a combined effect of phylogenetic relationships, and uneven patterns of gene duplication and loss. A much simpler case is presented by the fungal kexins, probably because they appear to have diversified principally in animals. We searched the human genome for kexin-like sequences and identified 13 distinct kexin genes. In contrast, individual fungi contain three at the most (Fig. 3). All the sequences predicted an endocellular function consistent with their role as preprotein convertases (Siezen & Leunissen, 1997). There is evidence from degree of divergence for relatively recent duplication of the kexin gene in C. neoformans and more ancient duplication events in the lineage leading to C. cinerea. However, aside the anomalous basal position of yeast sequences the observed phylogenetic distribution of the kexin orthologues is in agreement with taxonomically accepted relationships with excellent statistical support at most nodes (Fig. 3).

Phylogenic analysis of the trypsin superfamily

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

In order to reconstruct trypsin evolution in fungi and investigate the possibility that gene retention may represent niche-specific traits (traits shared by organisms that occupy the same niche irrespective of their phylogenetic position), we screened for the presence or absence of trypsins in 35 representative fungi (Fig. 4). To confirm the absence of trypsins, we probed Southern blots with trypsin genes (Fig. 5). In 33 cases PCR and Southern analyses were consistent. However, bands were detected in DNA blots of the saprophytic zygomycete M. mucedo, the plant pathogenic ascomycete B. ochroleuca and the basidiomycete phylloplane/opportunistic pathogen S. roseus, although PCR reactions using their DNA as template did not produce trypsin fragments. One possibility is that these represent degenerating sequences not recognized by primers as we previously failed to detect trypsin activities in Mucor spp. (Freimoser et al., 2003b). Although failure to detect trypsins using PCR technology/Southern analysis does not prove loss, the observed cases are totally consistent with available genomic data. Thus, PCR successfully amplified the trypsin sequences from A. nidulans and F. graminearum although not detecting trypsins in the other sequenced fungi, and this interpretation was supported by Southern analysis.

image

Figure 5. Southern blot analysis of representative fungi probed with PCR amplified trypsin fragments from M. anisopliae, C. coronatus and S. canescens. Abbreviations are: N.c., N. crassa; C.c., C. coronatus; M.a., M. anisopliae; A.m., A. macrogynus; Z.r., Z. radicans; B.r., B. ranarum; M.m., M. mucedo; A.b., A. bisporus; Co., C. cinereus; S.r., S. roseus; P.f., P. fumosoroseus; H.t., H. thompsonii; A.n., A. nidulans.

Download figure to PowerPoint

Trypsin genes were not detected in five of the seven basidiomycetes screened. The exceptions, clustering together in the phylogenetic tree (Fig. 4) were S. roseus (Southern blot data only) and S. canescens, an obligate scale insect symbiont that penetrates (using proteases?) its hosts cuticle. Trypsins were also present in six of 11 ascomycete pathogens of plants and insects, and the mycopathogen T. harzianum. A. nidulans was the only one of eight saprophytic ascomycetes with a trypsin. Trypsin genes were detected in six of eight zygomycetes, all but one a pathogen, showing that their last common ancestor with the basidiomycete/ascomycete clade had this enzyme [about 850 Ma (Berbee & Taylor, 2001)]. We were also able to find a trypsin in B. ranarum, an increasingly important pathogen of animals and humans that may be a chytrid (Bruun et al., 1998). However, no trypsin was detected in the basal chytrid A. macrogynus.

The estimate of phylogeny obtained using trypsin sequences (Fig. 4b) is largely congruent with the unconstrained rDNA tree (Fig. 4a), which is consistent with current estimates of organismal phylogeny for the fungi (Berbee & Taylor, 2001). This suggests a pathway of evolution in which the divergence of trypsins largely reflects the speciation of fungal lineages and the absence of trypsins in specific fungi reflects gene loss. Although the bootstrap support for many groups in the trypsin tree is low, these low values simply emphasize the absence of conflict with the fungal species tree.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References

The present study has revealed evolutionary selection of one gene family of proteases over another in species of ascomycete fungi. This suggests that differences in the properties of the enzymes provide selective advantages in different habitats. Although we presently cannot link these observations to any organisms’ lifestyle in specific terms, the presence of large clusters of genes is likely to reflect selective pressure for their increased or varied coding capacity (Jordan et al., 2001). The multiple subtilases in pathogenic ascomycetes could play different roles in pathogenesis, increase adaptability and host range, or have different functions in survival in various ecological habitats outside the host. There are clear catalytic differences between subtilases and trypsins, and within the proteinase K family of subtilases in their substrate specificities (St. Leger et al., 1994, 1996) that could allow their selection. In the case of M. anisopliae proteinase K subtilases (Pr1 enzymes) this extends to probable differences in their stability, absorption properties to insoluble substrates and interactions with protease inhibitors (Bagga et al., 2004). The selectable properties of class I subtilisins as compared to proteinase K subtilases is unknown. It remains to be determined if diversity in function parallels the subtilisin gene family diversity in M. grisea that exceeds that found in other fungi. The unusual number of subtilisins implies some particular adaptive significance for this lineage. Possibly the retention by M. grisea of so many subtilisin paralogues simply results from selection to achieve high expression by gene dosage; an example of gene copy number selection (Gruhl et al., 1997). This requires experimental validation to determine if the isoenzymes encoded by proteinase K genes change more in response to environments than those encoded by subtilisin genes, suggesting their stronger regulation (expression).

We suggest that the divergence of subtilisins seen in M. grisea may be sufficiently ancient as to represent the ancestral condition in early ascomycetes. Our results, both from protein sequence data and intron profiling, are consistent with most duplication events occurring after the ascomycete/basidiomycete split but before radiation of major fungal lineages such as the Sordariales and Hypocreales. If this is the case, then the relative paucity of these genes in most lineages reflects gene loss as compared with M. grisea. Uncertainty surrounding the timing of duplication and loss events will remain problematic, pending improved resolution of basal lineages of ascomycetes and extended taxon sampling. Clearly however, the absence of subtilisins in three distantly related fungi (A. nidulans, S. pombe and S. cerevisiae) is consistent with a trend towards loss of this group in diverse lineages.

Every fungus in this study possessed proteinase K subtilase genes. That includes the protease-impoverished yeast species that lack subtilisins but contain at least one proteinase K gene. Comparing the yeast genomes to those of other fungi suggests gene loss has taken place at many different points during the evolution of the S. cerevisiae and S. pombe lineages (Braun et al., 2000; Braun, 2003; Krylov et al., 2003). However, one can imagine selection acting particularly strongly against the deletion of the last gene in a gene family. This suggests that the proteinase K subtilases are less expendable than subtilisins even in the yeasts carbohydrate rich, protein poor habitat. The high level of sequence polymorphism shown inside each of the subtilase sub-families suggest that like subtilisins these genes also diversified in ascomycetes after the ascomycete/basidiomycete split but before radiation of several major ascomycete lineages. Aside the four proteinase K genes in C. cinerea, which appear to have diverged relatively recently, there is no evidence for a parallel divergence of genes encoding secreted proteases in basidiomycetes. The earliest basidiomycetes seem to have been wood inhabiting (Lewis, 1987), and in sharp contrast to the ascomycetes extant basidiomycetes include very few pathogens of animals, including insects (Carlile et al., 2001). It is intriguing to speculate that extra proteolytic competence may have allowed the early ascomycetes to grow on a greater variety of living and nonliving proteinaceous substrates. This may have been a component allowing niche differentiation between the ascomycetes and the basidiomycetes that will have adapted the former to pathogenicity to animals or may itself have derived from adaptation to pathogenicity. In any event, the fact that two families of subtilases radiated in the early ascomycetes suggest that these fungi had a lifestyle that selected for multiple protease activities. That the majority of fungi have a single sf3 activity suggests that this selective pressure did not extend to endocellular housekeeping enzymes. Aspergillus nidulans as a Eurotiomycete (=plectomycetes) provides an interesting outgroup as these diverged about 400 Ma (Kasuga et al., 2002) from the Sordariomycete (=pyrenomycetes) Ascomycota lineage that includes the Sordariales and Hypocreales. Aside the sf3 gene (An02381), A. nidulans has a single proteinase K subtilase (An5581) in sf2. Likewise, all genes cloned so far from A. fumigatus and A. niger fall within sf2 (Bagga et al., 2004). However, related Penicillium spp (Eurotiomycetes) have sf1 genes (Bagga et al., 2004) demonstrating that their last common ancestor with Aspergillus spp had sf1 genes too. Thus, as with subtilisins there has been a general trend for the proteinase K gene family to decrease in size on average for many lineages following the early diversification. The most obvious explanation is that as the requirements for proteases varied over time in different lineages so did the gene family. Selection for some lifestyles, particularly pathogenicity, may favour retention of large gene families of proteases whereas saprophytes such as N. crassa adapted to live principally on readily utilized carbohydrates have lost genes. In fact, for the pathogens there is evidence from small lineage specific clusters for more recent expansions in subsets of subtilases indicating that the genes in a family do not necessarily behave alike. Supporting this, evidence has been presented for recent tandem duplication of subtilase Pr1F as M. anisopliae sf anisopliae and M. anisopliae sf. acridum diverged (Bagga et al., 2004). Clearly for the strongest and most general inferences, future comparisons should consist of matched pairs of closely related pathogens and nonpathogens replicated throughout the ascomycete tree.

As extended sampling, with more taxa representing basal fungal lineages is necessary to distinguish the contributions of gene loss and gain, we used this approach to reconstruct the evolution of trypsins. Overall, Fig. 4 suggests that independent loss of ancestral trypsins has occurred in many fungal lineages. Several species that have no trypsins are nested among species that do, indicating that the trypsin has been lost comparatively recently in the lineage leading to that species. For example, the trypsin in A. nidulans is without a counterpart in P. notatum. The presence of a trypsin in A. nidulans, implies that it is not redundant in this species which may be linked with it lacking a subtilisin and having just two proteinase K activities. The convergence of the trypsin gene tree with the rDNA tree indicated that trypsins, where present, have diverged in parallel with the organisms in which they are expressed and were not obtained via horizontal gene transfer. If horizontal gene transfer is negligible then the evolutionary significance of lineage specific gene loss will be greater in fungi than in prokaryotes (Krylov et al., 2003). The ecological implication of this is that enzymes lost by fungi are not regained making gene loss a one-way street that will reduce future adaptive options compared to genetically more diverse ancestors.

We did not find an orthologue of the previously described chymotrypsin gene of M. anisopliae (Screen & St. Leger, 2000) in any of the sequenced genomes. We had previously made a case for the chymotrypsin being derived from a streptomycete via horizontal gene transfer as its closest amino acid identity was to S. griseus protease C (55%) and it seemed highly unlikely that this gene had been lost in all the other divergently evolving taxa (Screen & St. Leger, 2000). However, because of similarities between structures and functions in the actinobacteria (which includes Streptomyces), and lower Eukaryota, Cavalier-Smith (2002) has postulated an actinobacterial ancestry for eukaryotes. A search of the NCBI database, including more than 60 bacterial genomes, revealed only distant homologues to trypsins and chymotrypsins in nonactinomycete bacteria, sustaining the actinobacterial origin hypothesis. In any event, the widespread, if patchy distribution of fungal trypsins and their correlation with the species tree, suggests that the functional spectrum and phylogenetic breadth of the trypsin superfamily will have been much wider in early fungi than currently. It is of interest that the kexins, which like the trypsins exist as one or two copies per genome, have not been similarly lost suggesting that their endocellular housekeeping functions as preprotein convertases are indispensable and highly conserved between phylogenetically distant lineages.

Like subtilases, the trypsins are inducible by environmental cues and secreted (St. Leger et al., 1996). Their interactions with the environment presumably confer considerable selective functions in those fungi that express them at high levels and possibly pre-adapt them to certain lifestyles such as pathogenesis. The fact that both animal and plant host-associated genera harbour trypsins suggest that in some manner plants and animals offer similar physiological selective pressures. Conversely, in fungi that lost them, trypsins may have been rendered nonessential by functionally analogous subtilase genes, they may have otherwise ceased to provide any benefit or they may have actually interfered with ecological functions. The species tree (Fig. 4) has many exceptions to the correlation of trypsins and pathogenic lifestyle. Thus, M. grisea and the insect pathogen H. thompsonii lack trypsins, whereas A. nidulans possesses one. Metarhizium anisopliae may be a transitional species in this regards as some strains express high levels of a broad spectrum trypsin whereas other strains produce lower levels of an enzyme with a narrow specificity for Phe-Val-Arg (St. Leger et al., 1987; Bidochka et al., 1999). In addition the trypsin gene appears to be silent in M. anisopliae sf. acridum strain 324 (Freimoser et al., 2003a). Thus, there are several mechanisms available for different strains to adapt enzyme activities to their specific needs on their particular hosts. A plausible explanation for M. anisopliae sf. acridum strain 324 not expressing trypsin is that it has a very narrow host range and does not require a gene that facilitates opportunistic host jumping. Likewise, the obligate mite pathogen N. parvispora also lacks a trypsin although it clusters among other less fastidious zygomycetes that do possess one (Fig. 4).

As the protease gene families studied here seem likely to contribute substantially to the genomic determinants of phenotypic differences between fungal lineages, the challenge now is to better establish why evolutionary change has taken place in different fungi to determine the cause as well as the function of changes in enzyme profiles. The availability of genome sequences will also facilitate investigations of other paralogous gene families that may give clues as to the potential adaptive significance of lineage-specific expansions in fungi with different lifestyles.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Organisms
  6. Trypsin sequences from fungi
  7. Phylogenetic analysis
  8. Results
  9. Phylogenetic analysis of the subtilase superfamily
  10. Proteinase K family
  11. The class I subtilisin family
  12. Structure of M. grisea subtilisin genes
  13. Kexins
  14. Phylogenic analysis of the trypsin superfamily
  15. Discussion
  16. Acknowledgments
  17. References
  • Bagga, S., Hu, G., Screen, S.E. & St. Leger, R.J. 2004. Reconstructing the diversification of subtilisins in the pathogenic fungus Metarhizium anisopliae. Gene 324: 159169.
  • Berbee, M.L. and Taylor, J.W. 2001. Fungal molecular evolution: gene trees and geologic time. In: The Mycota: A Comprehensive Treatise on Fungi as Experimental Systems for Basic and Applied Research Systematics and Evolution, Part B. (D. J.McLaughlin, E. G.McLaughlin & P. A.Lemke, eds), pp. 229245. Springer-Verlag, Germany.
  • Bidochka, M.J., St. Leger, R.J., Stuart, A. & Gowanlock, K. 1999. Nuclear rDNA phylogeny in the fungal genus Verticillium and the relationship to insect and plant virulence, extracellular proteases and carbohydrases. Microbiology 145: 955963.
  • Braun, E. L. 2003. Innovation from reduction: gene loss, domain loss and sequence divergence in genome evolution. Appl. Bioinform. 2: 1334.
  • Braun, E.L., Halpern, A.L., Nelson, M.A. & Natvig D.O. 2000. Large scale comparison of fungal sequence inovation in Neurospora crassa and gene loss in Saccharomyces cerevisiae. Genome Res. 10: 416430.
  • Brunak, H.N. & Von Heijne, G. 1999. Machine learning approaches to the prediction of signal peptides and other protein sorting signals. Protein. Eng. 12: 39.
  • Bruun, J.A., Andrea, G., Jorgen, E. & Soren, R. 1998. Relationships of the insect-pathogenic order entomophthorales (Zygomcota, fungi) based on phylogenetic analyses of nuclear subunit ribosomal DNA sequences (SSU rDNA). Fungal Gen. Biol. 24: 325334.
  • Carlile, M.J. Watkinson, S.C. & Gooday, G.W. 2001. The Fungi, 2nd edn. Academic press, San Diego.
  • Cavalier-Smith, T. 2002. The phagotrophic origin of eukaryotes and phylogenetic classification of protozoa. Int. J. Syst. Evol. Microbiol. 52: 297354.
  • Dibb, N.J. & Newman, A.J. 1989. Evidence that introns arose at protosplice sites. EMBO J. 8: 20152021.
  • Duda, T.F.J. & Palumbi, S.R. 1999. Developmental shifts and species selection in gastropods. Proc. Natl. Acad. Sci. USA 96: 1027210277.
  • Eisen, J.A. & Fraser, C.M. 2003. Phylogenomics: intersection of evolution and genomics. Science 300: 17061707.
  • Freimoser, F.M., Screen, S.E., Bagga, S., Hu, G. & St. Leger R.J. 2003a. EST analysis of two subspecies of Metarhizium anisopliae reveals a plethora of secreted proteins with potential activity in insect hosts. Microbiology 149: 239247.
  • Freimoser, F.M., Screen, S.E., Bagga, S., Hu, G. & St. Leger R.J. 2003b. EST analysis of genes expressed by the zygomycete pathogen Conidiobolus coronatus during growth on insect cuticle. Microbiology 149: 18931900.
  • Frugoli, J.A., PcPeek, M.A., Thomas, T.L. & McClung, C.R. 1998. Intron loss and gain during evolution of the catalase gene family in angiosperms. Genetics 149: 355365.
  • Galagan, J.E., Calvo, S.E., Borkovich, K.A., Selker, E.U., Read, N.D., Jaffe, D., FitzHugh, W., Ma, L.J., Smirnov, S., Purcell, S., Rehman, B., Elkins, T., Engels, R., Wang, S., Nielsen, C.B., Butler, J., Endrizzi, M., Qui, D., Ianakiev, P., Bell-Pedersen, D., Nelson, M.A., Werner-Washburne, M., Selitrennikoff, C.P., Kinsey, J.A., Braun, E.L., Zelter, A., Schulte, U., Kothe, G.O., Jedd, G., Mewes, W., Staben, C., Marcotte, E., Greenberg, D., Roy, A., Foley, K., Naylor, J., Stange-Thomann, N., Barrett, R., Gnerre, S., Kamal, M., Kamvysselis, M., Mauceli, E., Bielke, C., Rudd, S., Frishman, D., Krystofova, S., Rasmussen, C., Metzenberg, R.L., Perkin, D.D., Kroken, S., Cogoni, C., Macino, G., Catcheside, D., Li, W., Pratt, R.J., Osmani, S.A., DeSouza, C.P., Glass, L., Orbach, M.J., Berglund, J.A., Voelker, R., Yarden, O., Plamann, M., Seiler, S., Dunlap, J., Radford, A., Aramayo, R., Natvig, D.O., Alex, L.A., Mannhaupt G., Ebbole, D.J., Freitag, M., Paulsen, I., Sachs, M.S., Lander, E.S., Nusbaum, C., Birren, B. 2003. The genome sequence of the filamentous fungus Neurospora crassa. Nature 422: 859868.
  • Goffeau A., Barrell, B.G., Bussey, H., Davis, R.W., Dujon, B., Feldmann, H., Galibert, F., Hoheisel, J.D., Jacq, C., Johnston, M., Louis, E.J., Mewes, H.W., Murakami, Y., Philippsen, P., Tettelin, H. & Oliver S.G. 1996. Life with 6000 genes. Science 274: 563567.
  • Graur, D. & Li, W.H., 2000. Fundamentals of Molecular Evolution, 2nd edition. Sinauer Associates Inc., Sunderland, Massachusetts.
  • Gruhl, M., Kao, W.Y. & Bergtrom, G. 1997. Evolution of orthologous intronless and intron-beraing globin genes in two insect species. J. Mol. Evol. 45: 499508.
  • Gunkle, F.A. & Gassen, H.G. 1989. Proteinase K from Tritirachum album limber: characterization of the chromosomal gene and expression of the cDNA in Escherichia coli. Eur. J. Biochem. 179: 185194.
  • Hartung, F., Blattner, F.R. & Puchta, H. 2002. Intron gain and loss in the evolution of the conserved eukaryotic recombination machinery. Nucleic Acids Res. 30: 51755181.
  • Heidelberg, J.F, Eisen, J.A., Nelson W.C., Clayton, R.A., Gwinn, M.L., Dodson, R.J., Haft, D.H., Hickey, E.K., Peterson, J.D., Umayam, L., Gill, S.R., Nelson, K.E., Read, T.D., Tettelin, H., Richardson, D., Ermolaeva, M.D., Vamathevan, J., Bass, S., Qin, H., Dragoi, I., Sellers, P., McDonald, L., Utterback, T., Fleishmann, R.D., Nierman, W.C., White, O. 2000. DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature 406: 477483.
  • Jordan, I.K., Makarova, K.S., Spouge, J.L., Wolf, Y.I. & Koonin, E.V. 2001. Lineage-specific gene expansions in bacterial and arcaeal genomes. Genome Res. 11: 555565.
  • Kasuga, T., White, T.J. & Taylor, J.W. 2002. Estimation of nucleotide substitution rates in eurotiomycete fungi. Mol. Biol. Evol. 19: 23182324.
  • Krylov, D.M., Wolf, Y.I., Rogozin, I.B. & Koonin, E.V. 2003. Gene loss, protein sequence divergence, gene dispensability, expression level, and interactivity are correlated in eukaryotic evolution. Genome Res. 13: 22292235.
  • Krzywinski, J. & Besansky, N.J. 2002. Frequent intron loss in the white gene: a cautionary tale for phylogeneticists. Mol. Biol. Evol. 19: 362366.
  • Kumar, S., Tamura, K., Jakobsen I.B. & Nei, M. 2001. MEGA2: Molecular Evolutionary Genetics Analysis software. Arizona State University, Tempe, Arizona, USA.
  • Lewis, D.H. 1987. Evolutinary aspects of mutualistic associations between fungi and photosynthetic organisms. In: Evolutionary Biology of the Fungi (A. D. M.Rayner, C. M.Brasier & D.Moore, eds), pp. 161178. Cambridge University Press, Cambridge.
  • Murphy, J.M. & Walton, J.D. 1996. Three extracellular proteases from Cochliobolus carbonum: cloning and targeted disruption of ALP1. Mol. Plant. Microbe. Interact. 9: 290297.
  • Rawlings, N.D. & Barrett, A.J. 1995. Families of serine peptidases. Methods Enzymol. 244: 1960.
  • Screen, S.E. & St. Leger, R.J. 2000. Cloning, expression, and substrate specificity of a fungal chymotrypsin. Evidence for lateral gene transfer from an actinomycete bacterium. J. Biol. Chem. 275: 66896694.
  • Siezen, R.J. & Leunissen, J.A.M. 1997. Subtilases: the superfamily of subtilisn-like serine proteases. Protein. Sci. 6: 501523.
  • St. Leger, R.J., Cooper, R.M. & Charnley, A.K. 1987. Distribution of chymoelastases and trypsin-like enzymes in five species of entomopathogenic deuteromycetes. Arch. Biochem. Biophys. 258: 123131.
  • St. Leger, R.J., Frank, D.C. Roberts, D.W. & Staples, R.C. 1992. Molecular cloning and regulatory analyses of the cuticle-degrading structural gene from the entomopathogenic fungus Metarhizium anisopliae. Eur. J. Biochem. 204: 9911001.
  • St. Leger, R.J., Bidochka, M.J. & Roberts, D.W., 1994. Isoforms of the cuticle-degrading Pr1 proteinase and production of a metalloproteinase by Metarhizum anisopliae. Arch. Biochem. Biophys. 313: 17.
  • St. Leger, R.J., Joshi, L., Bidochka, M.J., Rizzo, N.W. & Roberts, D.W. 1996. Biochemical localization and ultrastructural localization of two extracellular trypsins produced by Metarhizium anisopliae in infected cuticles. Appl. Environ. Microbiol. 62: 12571264.
  • St. Leger, R.J., Joshi, L. & Roberts, D.W., 1997. Adaptation of proteases and carbohydrases of saprophytic, phytopathogenic and entomopathogenic fungi to the requirements of their ecological niches. Microbiology 143: 19831992.
  • Thompson, J., Higgins, D. & Gibson, T., 1994. Clustal W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 46734680.
  • Walton, J.D. 1996. Host selective toxins: agents of compatibility. Plant. Cell. 8: 17231733.
  • Wood V, Gwilliam, R., Rajandream, M.A., Lyne, M., Lyne, R., Stewart, A., Sgouros, J., Peat, N., Hayles, J., Baker, S., Basham, D., Bowman, S., Brooks, K., Brown, D., Brown, S., Chillingworth, T., Churcher, C., Collins, M., Connor, R., Cronin, A., Davis, P., Feltwell, T., Fraser, A., Gentles, S., Goble, A., Hamlin, N., Harris, D., Hidalgo, J., Hodgson, G., Holroyd, S., Hornsby, T., Howarth, S., Huckle, E.J., Hunt, S., Jagels, K., James, K., Jones, L., Jones, M., Leather, S., McDonald, S., McLean, J., Mooney, P., Moule, S., Mungall, K., Murphy, L., Niblett, D., Odell, C., Oliver, K., O'Neil, S., Pearson, D., Quail, M.A., Rabbinowitsch, E., Rutherford, K., Rutter, S., Saunders, D., Seeger, K., Sharp, S., Skelton, J., Simmonds, M., Squares, R., Squares, S., Stevens, K., Taylor, K., Taylor, R.G., Tivey, A., Walsh, S., Warren, T., Whitehead, S., Woodward, J., Volckaert, G., Aert, R., Robben, J., Grymonprez, B., Weltjens, I., Vanstreels, E., Rieger, M., Schafer, M., Muller-Auer, S., Gabel, C., Fuchs, M., Dusterhoft, A., Fritzc, C., Holzer, E., Moestl, D., Hilbert, H., Borzym, K., Langer, I., Beck, A., Lehrach, H., Reinhardt, R., Pohl, T.M., Eger, P., Zimmermann, W., Wedler, H., Wambutt, R., Purnelle, B., Goffeau, A., Cadieu, E., Dreano, S., Gloux, S., Lelaure, V., Mottier, S., Galibert, F., Aves, S.J., Xiang, Z., Hunt, C., Moore, K., Hurst, S.M., Lucas, M., Rochet, M., Gaillardin, C., Tallada, V.A., Garzon, A., Thode, G., Daga, R.R., Cruzado, L., Jimenez, J., Sanchez, M., del Rey, F., Benito, J., Dominguez, A., Revuelta, J.L., Moreno, S., Armstrong, J., Forsburg, S.L., Cerutti, L., Lowe, T., McCombie, W.R., Paulsen, I., Potashkin, J., Shpakovski, G.V., Ussery, D., Barrell, B.G., Nurse, P. & Cerrutti, L. 2002. The genome sequence of Schizosaccharomyces pombe. Nature 415: 87180.
  • Yang, Z., Nielsen, R., Goldman, N., & Pedersen, A.-M. K. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics 155: 431449.
  • Zhang, J., Rosenberg, H.F. & Nei, M. 1998. Positive Darwinian selection after gene duplication in primate riboniclease genes. Proc. Natl. Acad. Sci. USA 95: 37083713.