Abstract

The reproductive genes of fungi, like those of many other organisms, are thought to diversify rapidly. This phenomenon could be associated with the formation of reproductive barriers and speciation. Ascomycetes produce two classes of mating type–specific peptide pheromones. These are required for recognition between the mating types of heterothallic species. Little is known regarding the diversity or the extent of species specificity in pheromone peptides among these fungi. We compared the putative protein-coding DNA sequences of the 2 pheromone classes from 70 species of Ascomycetes. The data set included previously described pheromones and putative pheromones identified from genomic sequences. In addition, pheromone genes from 12 Fusarium species in the Gibberella fujikuroi complex were amplified and sequenced. Pheromones were largely conserved among species in this complex and, therefore, cannot alone account for the reproductive barriers observed between these species. In contrast, pheromone peptides were highly diverse among many other Ascomycetes, with evidence for both positive diversifying selection and relaxed selective constraint. Repeats of the α-factor–like pheromone, which occur in tandem arrays of variable copy number, were found to be conserved through purifying selection and not concerted evolution. This implies that sequence specificity may be important for pheromone reception and that interspecific differences may indeed be associated with functional divergence. Our findings also suggest that frequent duplication and loss causes the tandem repeats to experience “birth-and-death” evolution, which could in fact facilitate interspecific divergence of pheromone peptide sequences.

Introduction

Sexual reproduction in ascomycete fungi (phylum Ascomycota) commences with the interaction between compatible cells, which is facilitated by pheromones and receptors (Kurjan 1993; Kim and Borkovich 2004, 2006). In heterothallic Ascomycetes, sexual reproduction can occur only between two individuals of different “mating type.” Pheromones are essential for successful mating and are produced in a mating type–specific manner in heterothallic species (Bender and Sprague 1989; Bobrowicz et al. 2002; Kim and Borkovich 2004, 2006; Coppin et al. 2005; Mayrhofer et al. 2006). In homothallic species, which can reproduce sexually through self-fertilization, mating pheromones might be somewhat dispensable (Mayrhofer and Pöggeler 2005; Kim et al. 2008; Lee et al. 2008).

Ascomycete fungi produce two classes of mating pheromones. These were first comprehensively studied in the yeast Saccharomyces cerevisiae and characterized as mating factors a and α (or a-factor and α-factor) (Stötzler and Duntze 1976; Betz et al. 1977, 1987). Both pheromone classes are cleaved from larger precursor proteins (Kurjan and Herskowitz 1982; Singh et al. 1983; Brake et al. 1985). Homologous pheromones and precursor genes have been described from numerous yeasts and filamentous Ascomycetes (all references and GenBank accession numbers are provided as supplementary material, Supplementary Material online).

In S. cerevisiae, 2 genes (Mfa1 and Mfa2) each encode a Prepro a-factor precursor polypeptide, from which the 11 amino acid mature a-factor pheromone is cleaved. The precursor is characterized by the presence of a carboxyl (C)-terminal prenylation signal or so-called CaaX motif, typical of fungal pheromone precursors. Prenylation is required for transport of the precursor to the plasma membrane where cleavage occurs (Brake et al. 1985). There are also 2 genes (MFα1 and MFα2) that encode Prepro α-factor polypeptides, which, respectively, contain 4 and 2 repeats of the 13-residue mature α-factor (Singh et al. 1983). Each mature peptide repeat is bordered by signals for cleavage by Ste13p and Kex2p on its N and C-terminal borders, respectively (Julius et al. 1983, 1984). It is thought that all the copies of the mature α-factor are cleaved and released by the cell (Caplan and Kurjan 1991).

Observations in a wide range of organisms have demonstrated that reproductive proteins, particularly those under sex-biased expression such as gamete recognition proteins, evolve rapidly, often under positive, diversifying selection (reviewed by Swanson and Vacquier 2002; Clark et al. 2006). Whether ascomyete pheromones display this trend has not been ascertained, although this appears to be the case for fungal mating type (MAT) genes (Turgeon 1998; Brown and Casselton 2001; Wik et al. 2008). The MAT loci encode putative transcription factors that determine mating type and control sexual development (Coppin et al. 1997), including pheromone production (Bobrowicz et al. 2002). A number of processes have been proposed to explain the phenomenon of rapid evolution in reproductive proteins. These include reinforcement (Howard 1993) and sexual selection (Palumbi 1999; Galindo et al. 2003). Regardless of its source, a fundamental consequence of the rapid evolution of interacting reproductive proteins is the potential to generate reproductive isolation between populations or species (Clark et al. 2006; Palumbi 2008).

The importance of pheromone peptide sequence for successful mate recognition is unclear. The distinct pheromones of certain yeast species appear to offer some level of species specificity, although this may be weak when pheromone peptides are only slightly dissimilar, such as between members of the same genus (McCullough and Herskowitz 1979; Burke et al. 1980; Hisatomi et al. 1988). Naider and Becker (2004) have demonstrated that only certain residues in the α-factor are essential for receptor stimulation, whereas others might be less important. For the pheromones of Basidiomycetes, alteration of certain residues can drastically affect the success of pheromone reception (Olesnicky et al. 2000; Fowler et al. 2001). It is, therefore, conceivable that reproductive barriers between some ascomycete species could stem from differences in their pheromone peptide sequences.

The potential for reproductive isolation as a result of species-specific mating cues, including chemical signals, is well known (Coyne and Orr 2004, p. 214–215). However, in fungi, there has only recently been exploration into the potential role of the pheromone–receptor system in the development of species boundaries (e.g., Karlsson et al. 2008). In this study, we considered the diversity and evolution of ascomycete pheromones, with particular focus on the multiple tandem repeat α-factor–like pheromone, and whether these peptides could potentially play a role in the generation of reproductive isolation and speciation.

Nucleotide sequences of previously described pheromones, as well as putative pheromones identified from genomic sequences available in public domain databases, were used. In addition, we determined the pheromone gene sequences for 12 Fusarium spp. in the Gibberella fujikuroi complex. Species in this complex display varying levels of sexual intercompatibility, ranging from complete reproductive isolation to complete interfertility (Desjardins et al. 2000; Leslie et al. 2004; Leslie and Summerall 2006; Kvas et al. 2009). To understand diversity in ascomycete pheromones, the rate of pheromone evolution was compared with that of house-keeping (HK) genes, and tests for positive diversifying selection were performed. Furthermore, to test the hypothesis that concerted evolution could act to homogenize multiple tandem repeats of the α-factor–like pheromone, evolutionary relationships among repeats within and between species were examined. The structural organization of the α-class precursor gene was also compared among the 12 G. fujikuroi complex species.

Materials and Methods

Fungal Isolates and DNA Isolation

A total of 24 Fusarium isolates were included in this study (see supplementary material, Supplementary Material online). These isolates represent the standard MAT-1 and MAT-2 mating type tester strains for the nine biological species or reproductively isolated mating populations (MP-A to MP-I) of the G. fujikuroi complex (reviewed by Kvas et al. 2009). All but one of these mating populations corresponds to a single phylogenetic species (O’Donnell et al. 1998; Kvas et al. 2009). The exception is MP-E, which consists of two distinct phylogenetic species that are interfertile (referred to here as Fusarium subglutinans groups “1” and “2”) (Steenkamp et al. 2002). Mating populations C and D (Fusarium fujikuroi and Fusarium proliferatum, respectively) have been shown to share a moderate level of interfertility (Leslie et al. 2004). A single interspecific cross has also been observed between a pair of isolates from MP-H (Fusarium circinatum) and MP-E (F. subglutinans group 1) (Desjardins et al. 2000). We also included two isolates from each of two species (Fusarium mangiferae and Fusarium sterilihyphosum) for which no sexual stage is known (Britz et al. 2002). DNA was extracted from all isolates following a protocol based on that of Steenkamp et al. (1999).

Sequences and Organization of Pheromone Precursor Genes

Putative pheromone precursor genes were identified by BlastP and TBlastN using, as queries, all currently known homologs of these genes in Ascomycetes (see supplementary material for references and GenBank accession numbers, Supplementary Material online). In addition, a-class precursors were recognized by the presence of the characteristic C-terminal “CaaX” motif (Brake et al. 1985). Likewise, α-class precursors were recognized by the presence of multiple repeats of an α-factor–like peptide bordered by the characteristic cleavage signals for homologs of Ste13p (Julius et al. 1983; Pöggeler 2000; Bobrowicz et al. 2002) and Kex2p (Julius et al. 1984; Darby and Smyth 1990).

To amplify the putative a- and α-class precursor genes (designated ppg1 and ppg2, respectively) from species of the G. fujikuroi complex, Primer Designer 4.20 (Sci Ed Central, Cary, NC) was used to design primers in the conserved regions flanking the genes in Fusarium verticillioides and Fusarium oxysporum (GenBank accession numbers AAIM02000073 and AAXH01000548). To amplify ppg2, primers ppg2-2 (5′-TGTCTGGCAGCAACACCATC-3′) and ppg2-3 (5′-CCGTCACTCAGAGCCAGGTA-3′) were used. The ppg1 gene was amplified in three overlapping fragments with primer set ppg1-7c (5′-ATATCACCGACGTACTGTAA-3′) and ppg1-8c (5′-TACGAGTACCACTCACTT-3′), primer set ppg1-S1 (5′CTGCAACCTCGAYTAYAA-3′) and ppg1-T1 (5′-ACCAGTAGCACCRTCRTC-3′), and primer set ppg1-S1 and ppg1-3 (5′-ATGGAGCGCTTGGCCTTGTG-3′). Polymerase chain reaction (PCR) mixtures were 25 μl in volume and contained 4 ng/μl template DNA, 1.5 mM MgCl2, 0.4 μM of each primer, 1 mM deoxynucleotide triphosphates (0.25 mM of each), and 0.05 U/μl Super-Therm DNA Polymerase and reaction buffer (Southern Cross biotechnology [Pty.] Ltd., Cape Town, South Africa). The PCR cycling conditions consisted of an initial denaturation at 94 °C for 60 s; 30 cycles of denaturation at 94 °C for 30 s, annealing for 30 s and extension at 70 °C for 90 s; followed by a final extension step at 70 °C for 10 min. An annealing temperature of 60 °C was used for all primer combinations, except ppg1-7c and ppg1-8c, where an annealing temperature of 58 °C was used.

For sequencing, PCR products were precipitated overnight in ethanol containing 0.1 M sodium acetate (pH 3.8) at 4 °C followed by centrifugation at 16,000 rcf for 30 min at 4 °C. The pellets were washed in 70% ethanol, air dried following centrifugation (16,000 rcf; 10 min; 4 °C) and removal of the supernatant, and resuspended in 50 μl sterile distilled water. These purified products were then subjected to automated Sanger sequencing in both directions using the original PCR primers, the BigDye terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA) and an ABI PRISM 377 DNA sequencer (Applied Biosystems). Sequence assembly, annotation, and in silico translation was performed using Vector NTI 9.0 (Invitrogen Life Technologies, Carlsbad, CA) and CLC Bio Main Workbench (CLC Bio, Aarhus, Denmark).

PCR products that did not yield clean chromatograms when sequenced were purified by cloning in Escherichia coli using the pGEM-T Easy Vector cloning system (Promega Corporation, Madison, WI). Cloned inserts were then amplified directly from colonies using the primers SP-6 (5′-ATTTAGGTGACACTATAG-3′) and T-7 (5′-TAATACGACTCACTATAGGG-3′). This PCR reaction mixture was 25 μl and contained 1 mM deoxynucleotide triphosphates (0.25 mM of each), 0.4 μM of each primer, 0.05 U/μl FastStart Taq DNA Polymerase and FastStart reaction buffer with MgCl2 (Roche Diagnostics, Mannheim, Germany). The PCR reaction conditions were as follows: denaturation at 94 °C for 5 min; followed by 30 cycles of denaturation at 94 °C for 30 s, annealing at 50 °C for 30 s and extension for 90 s at 72 °C, followed by a final extension step at 72 °C for 10 min. Sequencing was performed as described above, using the SP-6 and T-7 primers.

Interspecific Variation and Tests for Positive Selection

For evolutionary analyses, only the regions of the precursor genes encoding the predicted mature peptides were considered. As these peptides were divergent over large phylogenetic distances, species were only compared with other members of the same fungal Class (Figs. 1–3). Coding DNA sequences were aligned using the codon-based algorithm implemented in SQUINT (Goode and Rodrigo 2007). As a measure of the rate of protein sequence evolution, pairwise dN/dS ratios were used, where dN is the number of nonsynonymous substitutions per nonsynonymous site and dS the number of synonymous substitutions per synonymous site. This is an effective measure of evolutionary rate as it controls for differential phylogenetic distances and differential mutation rates (Ina 1996). To calculate dN/dS, the Nei and Gojobori (1986) method was used as implemented by the PAML package, version 4.3 (Yang 1997). All sequence pairs that were inferred by the program to be saturated for synonymous differences were excluded. For the multiple-copy peptides, all possible pairs between two species were averaged to give a single pairwise value. Pairwise dN/dS values were also determined for HK genes using coding sequences for the β-tubulin and translation elongation factor 1-α genes (when available on GenBank).

FIG. 1.

a-Class pheromone precursor proteins and predicted mature peptides. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). CaaX and CpaX motifs are shaded yellow. Mature a-factor peptides from yeasts are underlined. The hypothetical mature peptide repeats of the Gibberella fujikuroi species are boxed. Multiple genes encoded in the same genome are indicated by numbers in brackets. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

FIG. 1.

a-Class pheromone precursor proteins and predicted mature peptides. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). CaaX and CpaX motifs are shaded yellow. Mature a-factor peptides from yeasts are underlined. The hypothetical mature peptide repeats of the Gibberella fujikuroi species are boxed. Multiple genes encoded in the same genome are indicated by numbers in brackets. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

FIG. 2.

α-Class precursor protein structures and mature pheromone sequences from filamentous Ascomycetes. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). Lines with blocks schematically represent the number of repeats in each precursor protein. Repeat colors indicate different amino acid sequence variants of the pheromone within a single species. For each species, the sequences of all putative pheromone repeats are displayed on the right, with dots indicating residues that are identical to those in the first sequence. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

FIG. 2.

α-Class precursor protein structures and mature pheromone sequences from filamentous Ascomycetes. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). Lines with blocks schematically represent the number of repeats in each precursor protein. Repeat colors indicate different amino acid sequence variants of the pheromone within a single species. For each species, the sequences of all putative pheromone repeats are displayed on the right, with dots indicating residues that are identical to those in the first sequence. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

FIG. 3.

α-Class precursor protein structures and mature pheromone sequences from yeasts. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). Lines with blocks schematically represent the number of repeats in each precursor protein. Multiple genes present within the same genome are displayed beneath one another and numbered. Repeat colors indicate different amino acid sequence “variants” of the pheromone within a single species. For each species, the sequences of all putative pheromone repeats are displayed on the right, with dots indicating residues that are identical to those in the first sequence. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

FIG. 3.

α-Class precursor protein structures and mature pheromone sequences from yeasts. See supplementary material for complete DNA sequences, GenBank accession numbers and references (Supplementary Material online). Lines with blocks schematically represent the number of repeats in each precursor protein. Multiple genes present within the same genome are displayed beneath one another and numbered. Repeat colors indicate different amino acid sequence “variants” of the pheromone within a single species. For each species, the sequences of all putative pheromone repeats are displayed on the right, with dots indicating residues that are identical to those in the first sequence. The default amino acid coloration of BioEdit (PAM 250) (Hall 1999) is used.

Statistical analyses were performed using MYSTAT (Wilkinson 1987) version 12. The data were nonnormally distributed according to the Shapiro–Wilk test (Shapiro and Wilk 1965). Because the data were paired (pheromone and HK value for each species pair), the Wilcoxon Signed Ranks test (Wilcoxon 1945) was used to determine whether dN/dS values differed significantly between pheromone and HK genes.

To test for positive diversifying selection acting at specific sites, two “site models” (Yang et al. 2000a) were tested using the maximum likelihood approach implemented by the CODEML program in the PAML package. Model M7 (the beta model) assumes a β distribution of ω with 0 ≤ ω ≤ 1, whereas model M8 (the beta&ω model) adds a class of codons with ω ≥ 1. The two models were compared with respect to their fit to the data using a likelihood ratio test (LRT), as described by Yang et al. (2000a). The Bayes empirical Bayes output of CODEML was used to identify specific codons likely to fall into codon class ω ≥ 1 in model M8. For the multiple-repeat α-factor pheromone, the first repeat in each module was used to represent each species. This was done to avoid the risk that purifying selection acting at the within-species level could mask diversifying selection at the between-species level.

Relationships among Tandemly Repeated α-Factor–like Pheromone Sequences

To test the hypothesis that concerted evolution acts to homogenize DNA sequences among tandemly repeated copies of the α-factor–like pheromone, the relationships among DNA sequences encoding the mature pheromone peptides were examined. All repeats from all species in the Saccharomycetes and Sordariomycetes data sets (see supplementary material for sequences, Supplementary Material online) were considered, as these two data sets contained a large number of species including both close and distant relatives. DNA sequences were aligned using ClustalW (Thompson et al. 1994). The proportion of nucleotide sites at which each pair of sequences differed (i.e., p distances; Nei and Kumar 2000) was used to construct Neighbor joining trees using MEGA version 4 (Tamura et al. 2007). This simple distance-based model was selected because the mode of evolution of repeats was unclear, and repeats experiencing concerted evolution would not necessarily follow any particular nucleotide-substitution model. Hence, the focus was not on evolutionary relationships per se but simply on whether the DNA sequence of each repeat shared more similarity with adjacent tandem repeats or with a corresponding repeat from a related species.

Another means, used by Swanson and Vacquier (1998), to distinguish between concerted evolution and purifying selection was applied by comparing synonymous distance (dS) with nonsynonymous distance (dN). Purifying selection should theoretically act only at the protein level, consequently maintaining a low dN while dS is allowed to increase. In contrast, concerted evolution acts at the DNA level, causing both dN and dS to be low among tandem repeats. These values were calculated using the Yang and Nielsen method, as implemented in the program yn00 in the PAML package. Pairwise dS and dN values were generated for every possible peptide pair and then averaged such that each species had a single within-species value and each species pair had a single between-species value. Pairs saturated for synonymous or nonsynonymous substitutions were discarded. When plotted, the gradient (m) of each point represents the dN/dS ratio. Because between-species and within-species samples were independent and different in size, the Mann–Whitney U test (Mann and Whitney 1947) was implemented to determine whether m differed significantly between the samples.

To investigate the structural events that give rise to gain or loss of repeats, the ppg1 gene sequence from an isolate of each of the 12 Fusarium species in the G. fujikuroi complex were used. The program GATA (graphic alignment tool for comparative sequence analysis; Nix and Eisen 2005) was used to visualize the DNA sequence relationships between each pair of species. Given two input sequences, this application uses the National Center for Biotechnology Information (NCBI) BlastN (Altschul et al. 1990) and Bl2seq (Tatusova and Madden 1999) algorithms to generate all possible local alignments between two sequences. In this way, the application is able to identify duplications, deletions and inversions. A word size of 20 was used, along with a lower cut-off bit score of 24.9 to ensure that only regions of high homology were considered.

Results

Sequences and Organization of Pheromone Precursor Genes

For the a-class precursors, published sequences were available for nine species (three yeast and six filamentous Ascomycetes). For the α-class precursors, published sequences were available for 17 species (9 yeast and 8 filamentous Ascomycetes; see supplementary material for references and GenBank accession numbers, Supplementary Material online). Putative a-class and α-class precursor genes in the genomes of another 20 and 42 species, respectively, including F. verticillioides, were identified. In addition, the putative a-class and α-class precursor genes (ppg2 and ppg1, respectively) from the 24 Fusarium isolates, representing 12 species of the G. fujikuroi complex, including F. verticillioides, were sequenced (see supplementary material isolate information, sequences and GenBank accession numbers, Supplementary Material online). Sequences were obtained for both genes from all 12 species, although one isolate of F. sterilihyphosum (MRC2802) failed to yield a fragment corresponding to ppg1. Thus, a-class precursor sequences were obtained from 40 species, covering one filamentous class: Sordariomycetes, and two yeast classes: Saccharomycetes and Taphrinomycetes (Hibbett et al. 2007) (Fig. 1). For the α-class precursor, sequences were obtained from 69 species, covering four filamentous classes Sordariomycetes, Eurotiomycetes, Leotiomycetes, and Dothidiomycetes (Fig. 2) and two yeast classes Saccharomycetes and Taphrinomycetes (Fig. 3) (Hibbett et al. 2007). Most of the yeasts carried two or more copies of one or both precursor genes, whereas among the filamentous species, only Cryphonectria parasitica carried two copies of the a-class gene (Mf2/1 and Mf2/2), as previously reported by Zhang et al. (1998) (Fig. 1). There was only one species, Ashbya gosypii, for which the a-class but no α-class precursor was obtained. In contrast, there were 31 species for which only the α-class precursor was obtained.

The putative a-class precursors were all small, ranging from 21 to 66 amino acids in length (Fig. 1). All carried a C-terminal CaaX motif, which contained a Cystein [C] residue, followed by two aliphatic (a) residues and terminated in an arbitrary [X] residue. However, in the Fusarium (Gibberella), Verticillium and Trichoderma (Hypocrea) species, the first aliphatic residue was replaced by a polar (p) residue (Serine, Threonine, or Asparagine), thus creating the pattern “CpaX,” as was recognized by Schmoll et al. (2010). In some of the species in these three genera, one or more additional CpaX motifs were present in a repeated nature. In most of these cases, each of the repeated CpaX motifs was usually preceded by a conserved peptide sequence. In all 12 species of the G. fujikuroi complex, the a-class gene, ppg2, carried 3 repeats, except for F. nygamai, which carried two (Fig. 1). Similar repeats were found in the a-class precursors of F. oxysporum and both Verticillium species studied. Multiple CpaX motifs were also identified in two of the three Trichoderma species, although these were not preceded by a conserved peptide motif. Gibberella zeae (Fusarium graminearum) and Trichoderma atroviridae both had a single C-terminal CpaX motif and therefore resembled the typical a-class precursors (Fig. 1).

Schmoll et al. (2010) have recently described the unusual precursor structure described above in several ascomycetes and demonstrated the necessity of the gene for male fertility in Trichoderma reesei. Given the similarities to both a-class and α-class precursors (CpaX motifs in a repeated arrangement), this precursor type was referred to as a “hybrid-type” (Schmoll et al. 2010). However, because not all Fusarium, Verticillium, and Trichoderma species carry more than one CpaX motif, we consider this gene structure to be a variation of the a-class precursor, as opposed to an entirely distinct precursor class.

Based on sequence similarity to described a-factor–like peptides (Brake et al. 1985; Davey 1992), it was possible to predicte the mature a-factor–like peptides from the remaining Saccharomycetes species (Fig. 1). Similar inferences could not be made for the 25 Sordariomycetes species examined because the exact sequence of the mature a-factor–like peptide has not been described from any filamentous Ascomycetes and the processing signals have not been identified. Therefore, only sequences from class Saccharomycetes were considered during further evolutionary analysis of mature a-factor–like pheromones.

In the case of a-class precursors carrying repeated CpaX motifs, the mechanism of processing is also unclear. In some species, including all 12 Fusarium species representing the G. fujikuroi complex, a conserved peptide sequence was located immediately upstream of each CpaX motif (Fig. 1). We hypothesize that these may represent repeats of the mature a-factor–like pheromone peptide; however, they are not bordered by Kex2p-like cleavage signals, like repeats of the mature α-factor–like peptide. Further analysis is therefore still necessary to determine the precise mechanism of processing. It is notable, however, that this hypothetical mature a-factor–like peptide repeat is identical among all 12 G. fujikuroi complex species, implying that they may all produce identical a-factor–like pheromones.

Putative α-class precursor proteins were all larger than the a-class, ranging from 102 to 711 amino acids in length. All α-class precursors carried copies of a motif bordered by the Ste13p and Kex2p protease-like cleavage signals: an N-terminal run of X-Alanine or X-Proline dipeptides (Julius et al. 1983; Pöggeler 2000; Bobrowicz et al. 2002) and a C-terminal KR (Lysine-Arginine) or RR dipeptide (Julius et al. 1984; Darby and Smyth 1990). The motifs occurred in a tandem arrangement of between 1 (e.g., Saccharomyces castellii) and 16 (e.g., F. subglutinans group 1) repeats, with an average of 5 (Figs. 2 and 3).

The predicted sequences of the mature α-factor–like peptides were inferred on the basis of sequence similarity to previously described homologs and the presence of the Ste13p and Kex2p protease-like cleavage signals. Some predicted mature peptides, such as the last peptide repeat in all of the Fusarium spp., had Kex2p protease-like cleavage sites at both ends of the peptide. In several species, one or more copies of the predicted mature peptide were not bordered by one or either of the signature cleavage motifs. For example, in 20% of the putative peptides among the 12 Fusarium species examined, the Kex2p protease-like cleavage site (usually KR or RR) was replaced by TR. However, these were still regarded as pheromones for evolutionary analyses, unless they were highly dissimilar from other repeats in the precursor. In 29 species, there was complete identity in amino acid sequence among all encoded repeats of the mature peptide, whereas the remaining species each encoded between two and six similar yet distinct “variants” of the pheromone (Figs. 2 and 3). The number of repeats in the precursor gene varied considerably among the species and sometimes between the two isolates of the same species, from a minimum of 1 to a maximum of 16.

In each of the G. fujikuroi complex isolates, the α-class gene, ppg1, encoded between 4 and 11 copies (average = 8) of a peptide with the sequence WCTWRGQPCW. Each isolate also encoded between two and five copies (average = 4) of a second peptide, in which the Threonine residue was substituted with Methionine. A single repeat of two additional variants of the peptide (one in which the Threonine was replaced by Leucine and one in which the third Tryptophan residue was replaced with Cysteine) was each found in both isolates of Fusarium konzum. A single copy of a fifth variant of the peptide in which the second Tryptophan residue is replaced with an Arginine residue occurred in one of the two F. circinatum isolates (Fig. 2).

Interspecific Variation and Tests for Positive Selection

Four data sets (a-factor–like pheromones of the Saccharomycetes and α-factor–like pheromones of the Saccharomycetes, Sordariomycetes, and Eurotiomycetes) included a sufficient number of species pairs for sound statistical analyses. For most species pairs, pheromones displayed a far greater evolutionary rate than did HK genes, although a few species had pheromones that were strongly conserved, with even slower evolutionary divergence than the HK genes. However, average pairwise dN/dS values were found to be significantly greater for pheromones than for HK genes in all four data sets (P ≤ 0.05) (Fig. 4).

FIG. 4.

Dot plots of pairwise dN/dS values for pheromones versus HK genes. Each point represents a pairwise dN/dS value between a pair of species. Data sets are displayed independently, with values for HK genes on the left and pheromones on the right. P values represent significance values from the Wilcoxon signed ranks test. P ≤ 0.05 indicates a significant difference between the two samples. Sample sizes for the four data sets are 156, 15, 145, and 55 species pairs, respectively.

FIG. 4.

Dot plots of pairwise dN/dS values for pheromones versus HK genes. Each point represents a pairwise dN/dS value between a pair of species. Data sets are displayed independently, with values for HK genes on the left and pheromones on the right. P values represent significance values from the Wilcoxon signed ranks test. P ≤ 0.05 indicates a significant difference between the two samples. Sample sizes for the four data sets are 156, 15, 145, and 55 species pairs, respectively.

Results of the CODEML analyses and LRTs are displayed in Table 1. In three of the four data sets, there was no significant difference in the likelihood values of each pair of models. Hence, the α-factor–like pheromone of Sordariomycetes and both the a- and α-factor–like pheromones of Saccharomycetes appear not to have diversified under positive selection. In contrast, the α-factor–like pheromone of the Eurotiomycetes showed evidence of having diversified under positive selection. Only the codon encoding the third residue of this nine-residue peptide was found to have a probability >0.95 of falling into the “positive selection” class of codons with ω > 1. Indeed, this residue is highly variable among the nine Eurotiomycetes examined (Fig. 2). Whether the identity of this amino acid affects pheromone function or specificity will require future investigations such as site-directed mutagenesis. The CODEML results did not appear to be skewed by the inclusion of homothallic and asexual species, as reanalysis of the Sordariomycetes data set containing only those species known to be sexual and heterothallic yielded similar results.

Table 1.

Results of Tests for Positive Selection.

Data Set Modela Parameter Estimatesb lnL2δ
Saccharomycetes α M7 P = 0.839, q = 2.764 −566.745 0.046 
M8 p0 = 0.964, p = 0.906, q = 3.339, ω = 1.000 −566.722  
Saccharomycetes a M7 P = 0.371, q = 3.837 −368.243 0.000 
M8 p0 = 1.000, p = 0.372, q = 3.837, ω = 1.000 −368.243 
Sordariomycetes α M7 p = 0.332, q = 22.689 −273.156 0.000 
M8 p0 = 0.999, p = 0.332, q = 22.689, ω = 1.000 −273.156 
Eurotiomycetes α M7 p = 0.164, q = 1.281 −148.346 6.230* 
M8 p0 = 0.933, p = 0.394, q = 11.128, ω = 9.923 −145.231 
Data Set Modela Parameter Estimatesb lnL2δ
Saccharomycetes α M7 P = 0.839, q = 2.764 −566.745 0.046 
M8 p0 = 0.964, p = 0.906, q = 3.339, ω = 1.000 −566.722  
Saccharomycetes a M7 P = 0.371, q = 3.837 −368.243 0.000 
M8 p0 = 1.000, p = 0.372, q = 3.837, ω = 1.000 −368.243 
Sordariomycetes α M7 p = 0.332, q = 22.689 −273.156 0.000 
M8 p0 = 0.999, p = 0.332, q = 22.689, ω = 1.000 −273.156 
Eurotiomycetes α M7 p = 0.164, q = 1.281 −148.346 6.230* 
M8 p0 = 0.933, p = 0.394, q = 11.128, ω = 9.923 −145.231 
a

“Site” models in CODEML (Yang et al. 2000a) M7, the beta model and M8 the beta&ω model.

b

Parameter estimates generated by CODEML.

c

ln Likelihood score for each model, as calculated by CODEML.

d

LRT statistic, δ is the differences between the ln likelihood values.

*

Significant difference at a value of P ≤ 0.05.

Relationships among Tandemly Repeated α-Factor–like Pheromone Sequences

Distance analysis of multiple tandem repeats of the α-factor–like pheromone yielded two conflicting patterns (Figs. 5–7). In some species, the multiple tandem repeats were more closely related in DNA sequence to one another than to repeats from other related species. In contrast, some closely related species, including the members of the G. fujikuroi complex and Saccharomyces sensu stricto clade, displayed a different pattern in which repeats from several different species grouped together. In these instances, there was some positional bias in the clustering, where clusters usually consisted of repeats from a similar region in the precursor protein. For example, repeat 5 of Fusarium sacchari clustered with repeat 5 of F. mangiferae, whereas repeats 13 and 14 of the same two species clustered together. However, this pattern only appeared in species that shared close relatives within the data set.

FIG. 5.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from the Saccharomycetes. Colors indicate repeats from the same species. Some Saccharomycetes carry multiple α-class precursor genes; therefore, taxon names are followed by two numbers. The first number represents the gene, the second represents position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50% are displayed on branches.

FIG. 5.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from the Saccharomycetes. Colors indicate repeats from the same species. Some Saccharomycetes carry multiple α-class precursor genes; therefore, taxon names are followed by two numbers. The first number represents the gene, the second represents position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50% are displayed on branches.

FIG. 6.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from the Sordariomycetes (excluding Fusarium). Colors indicate repeats from the same species. The number after each taxon names represent the position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50 are displayed on branches.

FIG. 6.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from the Sordariomycetes (excluding Fusarium). Colors indicate repeats from the same species. The number after each taxon names represent the position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50 are displayed on branches.

FIG. 7.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from Fusarium spp. Colors indicate repeats from the same species. The number after each taxon names represent the position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50 are displayed on branches.

FIG. 7.

Neighbor joining distance tree of DNA sequences encoding mature α-factor–like peptide repeats from Fusarium spp. Colors indicate repeats from the same species. The number after each taxon names represent the position of the particular repeat within the precursor, starting at the N-terminal. Bootstrap values >50 are displayed on branches.

Broken down into synonymous and nonsynonymous distances, the within-species comparisons among peptide repeats had on average a notably lower dN than the between-species comparisons. However, the dS values were not vastly different, and some within-species dS values were greater than many between-species values (Fig. 8). The gradient of the dN versus dS plots (Fig. 8), which corresponds to the dN/dS ratio, varied widely, but was significantly lower for within-species than between-species values. This implies stronger purifying selection acting at the within-species level.

FIG. 8.

Scatter plots of average pairwise dN values for α-factor–like pheromones plotted against the corresponding dS values. Each “between-species” point represents the averaged dN and dS values for pairwise comparisons between all possible pairs of repeats for a single pair of species. Each dark “within-specie” point represents averaged dN and dS values for pairwise comparisons between all possible pairs of tandem repeats from within the same precursor gene. Between-species points outnumber within-species points because the former represent all possible species pairs, whereas the latter each represent a single species. Shaded areas represent one standard deviation above and below the mean for the gradient (m), which corresponds to dN/dS. P values represent significance values for the difference in gradient between the samples from the Mann–Whitney U test, with P ≤ 0.05 taken as significant.

FIG. 8.

Scatter plots of average pairwise dN values for α-factor–like pheromones plotted against the corresponding dS values. Each “between-species” point represents the averaged dN and dS values for pairwise comparisons between all possible pairs of repeats for a single pair of species. Each dark “within-specie” point represents averaged dN and dS values for pairwise comparisons between all possible pairs of tandem repeats from within the same precursor gene. Between-species points outnumber within-species points because the former represent all possible species pairs, whereas the latter each represent a single species. Shaded areas represent one standard deviation above and below the mean for the gradient (m), which corresponds to dN/dS. P values represent significance values for the difference in gradient between the samples from the Mann–Whitney U test, with P ≤ 0.05 taken as significant.

Results of the GATA analysis were compiled (Fig. 9) so that all changes in repeat copy number in the G. fujikuroi complex could be considered simultaneously. The results demonstrated that multiple independent changes in repeat copy number have occurred within and among these species. In particular, it appeared that the more central repeats have been more frequently involved in these structural events.

FIG. 9.

GATA alignments of ppg1 among species in the Gibberella fujikuroi species complex. Gray boxes indicate positions of the predicted α-factor–like mature peptides (see Figs. 2 and 3 for amino acid sequences). Lines connect portions of sequence with high homology (lower cut-off bit score of 24.9) and the darker the shading the stronger the homology. The accompanying reference phylogeny represents a consensus that was inferred from the results of previous studies (O’Donnell et al. 1998; Kvas et al. 2009).

FIG. 9.

GATA alignments of ppg1 among species in the Gibberella fujikuroi species complex. Gray boxes indicate positions of the predicted α-factor–like mature peptides (see Figs. 2 and 3 for amino acid sequences). Lines connect portions of sequence with high homology (lower cut-off bit score of 24.9) and the darker the shading the stronger the homology. The accompanying reference phylogeny represents a consensus that was inferred from the results of previous studies (O’Donnell et al. 1998; Kvas et al. 2009).

Discussion

This study has substantially increased the number of predicted a- and α-class pheromone precursor genes that have been described, providing new insights into the distribution, sequence diversity and structural organization of these genes. Some acomycete pheromones are strongly conserved between species, but most are highly divergent. The rich diversity in peptide sequence has probably resulted from both adaptive and nonadaptive evolutionary forces. Both pheromone precursor classes can have a modular nature, a characteristic that could contribute to the interspecific divergence of the pheromone peptides. Much of this variation might also have functional relevance, implying a potential role of pheromones in species-specific mate recognition. The rapid evolution of pheromones could, therefore, contribute to speciation in Ascomycetes.

The α-class precursor was distributed across species from all four filamentous classes (Sordariomycetes, Eurotiomycetes, Dothidiomycetes, and Leotiomycetes) as well as both yeast classes (Saccharomycetes and Taphrinomycetes) included in this study. By contrast, the a-class precursor was found in far fewer species, spanning the two yeast classes and a single filamentous class, Sordariomycetes. Several previous studies have been unable to identify the a-class precursor in various genomes (Pöggeler 2002; Dyer et al. 2003; Hoff et al. 2008; Butler et al. 2009). This small gene might have been altered to a point where it is undetectable by Blast analysis in these fungi. If it is indeed absent, this would require that some other element has assumed the role of the a-factor–like pheromone, at least in heterothallic species. In some homothallic species, only one pheromone–receptor pair might be necessary, whereas the other could eventually be lost (Mayrhofer et al. 2006; Kim et al. 2008; Lee et al. 2008).

All the pheromone precursor genes that were found in this study appeared to have intact open reading frames. These include those in species for which no sexual stage is known such as F. mangiferae, F. sterilihyphosum, and F. oxysporum. Although this could be interpreted as evidence that these species might have a cryptic sexual cycle, these genes could be retained in asexual species if they perform additional functions outside of sexual reproduction. For example, work on Neurospora crassa pheromone precursor Mfa-1 has suggested a possible role in “conglutination,” the cementing of hyphae, which occurs, for example, during sclerotium formation (Kim et al. 2002). In Candida spp., the α-factor–like pheromone appears to be necessary for intercellular signaling during biofilm formation (Daniels et al. 2006; Sahni et al. 2010). Further work is, therefore, necessary to distinguish between the sexual and nonsexual functions of these pheromones.

This study is the first to scrutinize similarities and differences in the pheromones of a group of closely related filamentous Ascomycetes with varying degrees of sexual compatibility. Most species pairs in the G. fujikuroi complex are intersterile. However, we have found a lack of significant differences among the putative pheromone peptides in species of this complex. We hypothesize that the repeated eight-residue motif encoded by ppg2 represents the mature a-factor like pheromone. If so, there is complete identity in this pheromone among all 12 species. In ppg1, each species carried multiple copies of two or more distinct variants of the pheromone that differed slightly in amino acid sequence. Nevertheless, only one species (F. konzum) carried variants that were unique and potentially species specific. Unless these pheromones undergo differential posttranslational modifications (an avenue that has not been explored in ascomycete pheromones), our results imply that these species should recognize one another as potential mates. A mechanism further downstream must, therefore, be responsible for the observed reproductive barriers. Indeed, interspecific mating experiments in Fusarium can lead to the formation of barren perithecia (Leslie et al. 2004), suggesting that hyphal fusion could have occurred. Such postmating barriers are common in the genus Neurospora, which led Turner et al. (2010) to propose that pheromone divergence could be prevented by strong selective constraint.

On the whole, strong conservation of pheromones was not the norm; in fact, pheromones were significantly more divergent than HK genes in all four data sets considered. These findings are in agreement with the rapid evolution of reproductive proteins observed in the plant and animal kingdoms (as reviewed by Swanson and Vacquier 2002; Clark et al. 2006). This phenomenon has been proposed as a driving force in speciation, particularly when associated with mate-recognition proteins. A comparable example is the rapid evolution of interacting sperm and egg surface proteins in Abalone (Haliotis rufescens), which could cause reproductive isolation between populations (Yang et al. 2000b; Galindo et al. 2003). If pheromone peptide sequence is indeed crucial for successful reception, the observed interspecific differences and rapid evolution of pheromones in this study might have major implications for reproductive isolation and speciation in Ascomycetes. An understanding of the forces that drive pheromone diversification could shed light on the importance of these modifications for mate recognition. Alterations that come about through relaxed selective constraint are probably inconsequential, whereas those driven by positive selection are likely to have functional relevance.

Our analyses identified evidence for positive diversifying selection acting in the data set of α-factor–like pheromones from Eurotiomycetes but not in the larger data sets of the Sordariomycetes and Saccharomycetes. This implies that the variation among species in the latter groups might have accumulated over time through relaxed constraint. However, the test performed to detect positive selection in this study addresses each data set as a whole and is most sensitive when positive selection acts in all lineages. Yang and Nielsen (2002) demonstrated that selection acting only along certain lineages of the phylogeny might go undetected by the site models. It is plausible that pheromone peptides follow this sort of evolutionary pattern given the observation that the pheromones of some species pairs are highly divergent, whereas those of a few remain even more conserved than HK genes. The ability of Ascomycetes to reproduce efficiently through asexual means could further complicate the issue. In periods of preferentially asexual reproduction, sex-related genes become obsolete and may diverge under relaxed selective constraint. Therefore, the detection of relaxed constraint cannot necessarily be construed as evidence that peptide sequence is not important for successful pheromone reception.

In the α-class precursors described here and also in those previously described, amino acid sequences of the multiple tandem repeats of the mature peptide were usually identical or nearly identical. We found that this similarity was largely reflected at the DNA level, in that tandem repeats were often more closely related to one another in DNA sequence than to repeats from any other species. Such a pattern has been interpreted as the signature of concerted evolution (e.g., Swanson and Vacquier 1998). However, the repeats of more closely related species did not form independent clusters. For example, all repeats in each G. fujikuroi complex species had their closest relative in another species of this complex. This pattern has been termed transspecific polymorphism (Klein 1987) and occurs when duplication predates speciation. Thus, paralogs of a certain sequence are more similar to their corresponding copy in a related species than to the sister paralog in the same genome. The inclusion of many closely related species in this study has therefore demonstrated that concerted evolution in the conventional sense, which is thought to rapidly homogenize DNA sequences among tandem repeats (Elder and Turner 1995), appears not to occur in the α-class precursor.

The transspecific polymorphism was absent in all the species that did not have close relatives in our data set, indicating that it degrades over larger evolutionary distances. Rather than concerted evolution, this pattern is probably an inevitable result of continuous duplication and deletion of repeats, as evidenced by the huge variation in repeat copy number. Evolution through duplication and deletion has been termed “birth-and-death” evolution and is common in tandem repeat sequences (Nei and Rooney 2005). Characteristic examples of genes experiencing birth and death are the major histocompatibility complex and imunoglobin genes of mammals (Nei et al. 1997) as well as the ribosomal RNA genes in fungi (Rooney and Ward 2005). Concerted evolution has been rejected in these cases because, although some members of the multigene families are highly similar, others have their closest relatives in distantly related species. In Figure 10, we provide a model for how continuous duplications and deletions acting on a set of repeats could create such a phylogenetic scenario, starting with complete transspecific polymorphism and progressing, eventually, to exclusive monophyletic clustering.

FIG. 10.

Theoretical model showing birth and death evolution of tandem pheromone repeats. Blocks indicate repeats of the mature α-factor–like peptide. Fill patterns indicate distinct amino acid sequence variants. A constant rate of evolution is assumed, with one substitution (indicated by a small dot) occurring at each step. Accompanying phylogenies represent relationships among underlying DNA sequences at each step.

FIG. 10.

Theoretical model showing birth and death evolution of tandem pheromone repeats. Blocks indicate repeats of the mature α-factor–like peptide. Fill patterns indicate distinct amino acid sequence variants. A constant rate of evolution is assumed, with one substitution (indicated by a small dot) occurring at each step. Accompanying phylogenies represent relationships among underlying DNA sequences at each step.

Birth and death is not the only mechanism acting to maintain similarity among peptide repeats. We have shown that although repeats from the same species always have low nonsynonymous distances, some have high synonymous distances, and that dN/dS is significantly lower at the within-species level. This indicates that not all repeats are similar only due to recent duplication but that the sequences of the encoded peptides are also conserved by purifying selection (Ina 1996). We, therefore, propose that the similarity observed among repeats of the mature peptide is adaptive and reflects a requirement for sequence specificity in pheromone reception. Differences between species could therefore indeed be associated with functional divergence.

The unique structure and evolutionary system of the α-class precursor could further contribute to the rapid evolution of the α-factor–like pheromones. Random events of duplication and loss could cause the “fixation” of a certain pheromone variant in one species and the simultaneous fixation of a different variant in another species. This could be facilitated by positive selection but might even occur by chance in the absence of selection (Nei and Rooney 2005). The gradual process of succession by which a new repeat variant can spread through the precursor, replacing the old one, might also facilitate the coevolution of pheromones and receptors. A new compatible pheromone–receptor pair could spread in a population without a transition through an “adaptive valley.” Individuals carrying repeats of both the new and the wild-type pheromone maintain the ability to be recognized by the “wild-type” receptor and would therefore not suffer a dramatic fitness cost. Lastly, the ability to harbor multiple-repeat variants could increase the capacity to evolve by providing a constant source of variation, the raw material for natural selection.

Supplementary Material

The supplementary material presented in a single HTML file is available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/). It contains information on the Fusarium isolates used in this study as well as the structure of ppg1 and ppg2 in F. verticillioides, along with primer binding sites. Also provided are all nucleotide sequences of pheromone precursor genes used in this study, together with translation, GenBank accession numbers and references for previously described pheromones.

This work was funded by The National Research Foundation (NRF), the Department of Science and Technology (DST)/NRF Centre of Excellence in Tree Health Biotechnology, the University of Pretoria, and the Tree Protection Cooperative Program.

References

Altschul
SF
Gish
W
Miller
W
Myers
EW
Lipman
DJ
Basic local alignment search tool
J Mol Biol
 , 
1990
, vol. 
215
 (pg. 
403
-
410
)
Bender
A
Sprague
GF
Jr.
Pheromones and pheromone receptors are the primary determinants of mating specificity in the yeast Saccharomyces cerevisiae
Genetics
 , 
1989
, vol. 
121
 (pg. 
463
-
476
)
Betz
R
Crabb
JW
Meyer
HE
Wittig
R
Duntze
W
Amino acid sequences of a-factor mating peptides from Saccharomyces cerevisiae
J Biol Chem
 , 
1987
, vol. 
262
 (pg. 
546
-
548
)
Betz
R
MacKay
VL
Duntze
W
a-Factor from Saccharomyces cerevisiae: partial characterization of a mating hormone produced by cells of mating type a
J Bacteriol
 , 
1977
, vol. 
132
 (pg. 
462
-
472
)
Bobrowicz
P
Pawlak
R
Correa
A
Bell-Pedersen
D
Ebbole
DJ
The Neurospora crassa pheromone precursor genes are regulated by the mating type locus and the circadian clock
Mol Microbiol
 , 
2002
, vol. 
45
 (pg. 
795
-
804
)
Brake
AJ
Brenner
C
Najarian
R
Laybourn
P
Merrywheather
J
Gething
MJ
Structure of genes encoding precursors of the yeast peptide mating pheromone a-factor
Protein transport and secretion
 , 
1985
Cold Spring Harbor, NY
Cold Spring Harbor Laboratory Press
Britz
H
Steenkamp
ET
Coutinho
TA
Wingfield
BD
Marasas
WFO
Wingfield
MJ
Two new species of Fusarium section Liseola associated with mango malformation
Mycologia
 , 
2002
, vol. 
94
 (pg. 
722
-
730
)
Brown
AJ
Casselton
LA
Mating in mushrooms: increasing the chances but prolonging the affair
Trends Genet
 , 
2001
, vol. 
17
 (pg. 
393
-
400
)
Burke
D
Mendonça-Previato
L
Ballou
CE
Cell–cell recognition in yeast: purification of Hansenula wingei 21-cell sexual agglutination factor and comparison of the factors from three genera
Proc Natl Acad Sci USA
 , 
1980
, vol. 
77
 (pg. 
318
-
322
)
Butler
G
Rasmussen
MD
Lin
MF
, et al.  . 
(51 co-authors) Evolution of pathogenicity and sexual reproduction in eight Candida genomes
Nature
 , 
2009
, vol. 
459
 (pg. 
657
-
662
)
Caplan
S
Kurjan
J
Role of α-factor and the MFα1 α-factor precursor in mating in yeast
Genetics
 , 
1991
, vol. 
127
 (pg. 
299
-
307
)
Clark
NL
Aagaard
JE
Swanson
WJ
Evolution of reproductive proteins from animals and plants
Reproduction
 , 
2006
, vol. 
131
 (pg. 
11
-
22
)
Coppin
E
Debuchy
R
Arnaise
S
Picard
M
Mating types and sexual development in filamentous ascomycetes
Microbiol Mol Biol Rev
 , 
1997
, vol. 
61
 (pg. 
411
-
428
)
Coppin
E
de Renty
C
Debuchy
R
The function of the coding sequences for the putative pheromone precursors in Podospora anserina is restricted to fertilization
Eukaryot Cell
 , 
2005
, vol. 
4
 (pg. 
407
-
420
)
Coyne
JA
Orr
HA
Speciation
 , 
2004
Sunderland, MA
Sinauer
Daniels
KJ
Srikantha
T
Lockhart
SR
Pujol
C
Soll
DR
Opaque cells signal white cells to form biofilms in Candida albicans
EMBO J
 , 
2006
, vol. 
25
 (pg. 
2240
-
2252
)
Darby
NJ
Smyth
DG
Endopeptidases and prohormone processing
Biosci Rep
 , 
1990
, vol. 
10
 (pg. 
1
-
13
)
Davey
J
Mating pheromones of the fission yeast Schizosaccharomyces pombe: purification and structural characterization of M-factor and isolation and analysis of two genes encoding the pheromone
EMBO J
 , 
1992
, vol. 
11
 (pg. 
951
-
960
)
Desjardins
AE
Plattner
RD
Gordon
TR
Gibberella fujikuroi mating population A and Fusarium subglutinans from teosinte species and maize from Mexico and Central America
Mycol Res
 , 
2000
, vol. 
104
 (pg. 
865
-
872
)
Dyer
PS
Paoletti
M
Archer
DB
Genomics reveals sexual secrets of Aspergillus
Microbiology
 , 
2003
, vol. 
149
 (pg. 
2301
-
2303
)
Elder
JF
Jr
Turner
BJ
Concerted evolution of repetitive DNA sequences in eukaryotes
Q Rev of Biol
 , 
1995
, vol. 
70
 (pg. 
297
-
320
)
Fowler
TJ
Mitton
MF
Vaillancourt
LJ
Raper
CA
Changes in mate recognition through alterations of pheromones and receptors in the multisexual mushroom fungus Schizophyllum commune
Genetics
 , 
2001
, vol. 
158
 (pg. 
1491
-
1503
)
Galindo
BE
Vacquier
VD
Swanson
WJ
Positive selection in the egg receptor for abalone sperm lysin
Proc Natl Acad Sci USA
 , 
2003
, vol. 
100
 (pg. 
4639
-
4643
)
Goode
MG
Rodrigo
AG
SQUINT: a multiple alignment program and editor
Bioinformatics
 , 
2007
, vol. 
23
 (pg. 
1553
-
1555
)
Hall
TA
BioEdit: a user-friendly biological sequence alignment editor and analysis program for windows 95/98/NT
Nucleic Acids Symp Ser
 , 
1999
, vol. 
41
 (pg. 
95
-
98
)
Hibbett
DS
Binder
M
Bischoff
JF
, et al.  . 
(67 co-authors)
A higher-level phylogenetic classification of the Fungi
Mycol Res
 , 
2007
, vol. 
111
 (pg. 
509
-
547
)
Hisatomi
T
Yanagishima
N
Sakurai
A
Kobayashi
H
Interspecific actions of α mating pheromones on the a mating-type cells of three Saccharomyces yeasts
Curr Genet
 , 
1988
, vol. 
13
 (pg. 
25
-
27
)
Hoff
B
Pöggeler
S
Kuck
U
Eighty years after its discovery, Fleming's Penicillium strain discloses the secret of its sex
Eukaryot Cell
 , 
2008
, vol. 
7
 (pg. 
465
-
470
)
Howard
DJ
Harrison
RG
Reinforcement: origin, dynamics and fate of an evolutionary hypothesis
Hybrid zones and the evolutionary process
 , 
1993
Oxford
Oxford University Press
Ina
Y
Pattern of synonymous and nonsynonymous substitutions: an indicator of mechanisms of molecular evolution
J Genet
 , 
1996
, vol. 
75
 (pg. 
91
-
115
)
Julius
D
Blair
L
Brake
A
Sprague
G
Thorner
J
Yeast α factor is processed from a larger precursor polypeptide: the essential role of a membrane-bound dipeptidyl aminopeptidase
Cell
 , 
1983
, vol. 
32
 (pg. 
839
-
852
)
Julius
D
Schekman
R
Thorner
J
Glycosylation and processing of prepro-α-factor through the yeast secretory pathway
Cell
 , 
1984
, vol. 
36
 (pg. 
309
-
318
)
Karlsson
M
Nygren
K
Johannesson
H
The evolution of the pheromonal signal system and its potential role for reproductive isolation in heterothallic Neurospora
Mol Biol Evol
 , 
2008
, vol. 
25
 (pg. 
168
-
178
)
Kim
H
Borkovich
KA
A pheromone receptor gene, pre-1, is essential for mating type-specific directional growth and fusion of trichogynes and female fertility in Neurospora crassa
Mol Microbiol
 , 
2004
, vol. 
52
 (pg. 
1781
-
1798
)
Kim
H
Borkovich
KA
Pheromones are essential for male fertility and sufficient to direct chemotropic polarized growth of trichogynes during mating in Neurospora crassa
Eukaryot Cell
 , 
2006
, vol. 
5
 (pg. 
544
-
554
)
Kim
H
Metzenberg
RL
Nelson
MA
Multiple functions of mfa-1, a putative pheromone precursor gene of Neurospora crassa
Eukaryot Cell
 , 
2002
, vol. 
1
 (pg. 
987
-
999
)
Kim
HK
Lee
T
Yun
SH
A putative pheromone signaling pathway is dispensable for self-fertility in the homothallic ascomycete Gibberella zeae
Fungal Genet Biol
 , 
2008
, vol. 
45
 (pg. 
1188
-
1196
)
Klein
J
Origin of major histocompatibility complex polymorphism: the trans-species hypothesis
Hum Immunol
 , 
1987
, vol. 
19
 (pg. 
155
-
162
)
Kurjan
J
The pheromone response pathway in Saccharomyces cerevisiae
Annu Rev of Genet
 , 
1993
, vol. 
27
 (pg. 
147
-
179
)
Kurjan
J
Herskowitz
I
Structure of a yeast pheromone gene (MFα): a putative α-factor precursor contains four tandem copies of mature α-factor
Cell
 , 
1982
, vol. 
30
 (pg. 
933
-
943
)
Kvas
M
Marasas
WFO
Wingfield
BD
Wingfield
MJ
Steenkamp
ET
Diversity and evolution of Fusarium species in the Gibberella fujikuroi complex
Fungal Div
 , 
2009
, vol. 
34
 (pg. 
1
-
21
)
Lee
J
Leslie
JF
Bowden
RL
Expression and function of sex pheromones and receptors in the homothallic ascomycete Gibberella zeae
Eukaryot Cell
 , 
2008
, vol. 
7
 (pg. 
1211
-
1221
)
Leslie
J
Zeller
K
Wohler
M
Summerell
B
Interfertility of two mating populations in the Gibberella Fujikuroi species complex
Eur J Plant Pathol
 , 
2004
, vol. 
110
 (pg. 
611
-
618
)
Leslie
JF
Summerall
BA
The Fusarium laboratory manual
 , 
2006
Ames, IA
Blackwell Professional
Mann
HB
Whitney
DR
On a test of whether one of two random variables is stochastically larger than the other
Ann Math Stat
 , 
1947
, vol. 
18
 (pg. 
50
-
60
)
Mayrhofer
S
Pöggeler
S
Functional characterization of an α-factor-like Sordaria macrospora peptide pheromone and analysis of its interaction with its cognate receptor in Saccharomyces cerevisiae
Eukaryot Cell
 , 
2005
, vol. 
4
 (pg. 
661
-
672
)
Mayrhofer
S
Weber
JM
Pöggeler
S
Pheromones and pheromone receptors are required for proper sexual development in the homothallic ascomycete Sordaria macrospora
Genetics
 , 
2006
, vol. 
172
 (pg. 
1521
-
1533
)
McCullough
J
Herskowitz
I
Mating pheromones of Saccharomyces kluyveri: pheromone interactions between Saccharomyces kluyveri and Saccharomyces cerevisiae
J Bacteriol
 , 
1979
, vol. 
138
 (pg. 
146
-
154
)
Naider
F
Becker
JM
The α-factor mating pheromone of Saccharomyces cerevisiae: a model for studying the interaction of peptide hormones and G protein-coupled receptors
Peptides
 , 
2004
, vol. 
25
 (pg. 
1441
-
1463
)
Nei
M
Gojobori
T
Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions
Mol Biol Evol
 , 
1986
, vol. 
3
 (pg. 
418
-
426
)
Nei
M
Gu
X
Sitnikova
T
Evolution by the birth-and-death process in multigene families of the vertebrate immune-system
Proc Natl Acad Sci USA
 , 
1997
, vol. 
94
 (pg. 
7799
-
7806
)
Nei
M
Kumar
S
Molecular evolution and phylogenetics
 , 
2000
New York
Oxford University Press
Nei
M
Rooney
AP
Concerted and birth-and-death evolution of multigene families
Annu Rev of Genet
 , 
2005
, vol. 
39
 (pg. 
121
-
152
)
Nix
D
Eisen
M
GATA: a graphic alignment tool for comparative sequence analysis
BMC Bioinform
 , 
2005
, vol. 
6
 pg. 
9
 
O’Donnell
K
Cigelnik
E
Nirenberg
HI
Molecular systematic and phylogeography of the Gibberella fujikuroi species complex
Mycologia
 , 
1998
, vol. 
90
 (pg. 
465
-
493
)
Olesnicky
NS
Brown
AJ
Honda
Y
Dyos
SL
Dowell
SJ
Casselton
LA
Self-compatible B mutants in coprinus with altered pheromone-receptor specificities
Genetics
 , 
2000
, vol. 
156
 (pg. 
1025
-
1033
)
Palumbi
SR
All males are not created equal: fertility differences depend on gamete recognition polymorphisms in sea urchins
Proc Nat Acad Sci USA
 , 
1999
, vol. 
96
 (pg. 
12632
-
12637
)
Palumbi
SR
Speciation and the evolution of gamete recognition genes: pattern and process
Heredity
 , 
2008
, vol. 
102
 (pg. 
66
-
76
)
Pöggeler
S
Two pheromone precursor genes are transcriptionally expressed in the homothallic ascomycete Sordaria macrospora
Curr Genet
 , 
2000
, vol. 
37
 (pg. 
403
-
411
)
Pöggeler
S
Genomic evidence for mating abilities in the asexual pathogen Aspergillus fumigatus
Curr Genet
 , 
2002
, vol. 
42
 (pg. 
153
-
160
)
Rooney
AP
Ward
TJ
Evolution of a large ribosomal RNA multigene family in filamentous fungi: birth and death of a concerted evolution paradigm
Proc Natl Acad Sci USA
 , 
2005
, vol. 
102
 (pg. 
5084
-
5089
)
Sahni
N
Yi
S
Daniels
KJ
Huang
G
Srikantha
T
Soll
DR
Tec1 mediates the pheromone response of the white phenotype of Candida albicans: insights into the evolution of new signal transduction pathways
PLoS Biol
 , 
2010
, vol. 
8
 
5
pg. 
e1000363
  
doi:10.1371/journal.pbio.1000363
Schmoll
M
Seibel
C
Tisch
D
Dorrer
M
Kubicek
CP
A novel class of peptide pheromone precursors in ascomycetous fungi
Mol Microbiol
 , 
2010
, vol. 
77
 (pg. 
1483
-
1501
)
Shapiro
SS
Wilk
MB
An analysis of variance test for normality (complete samples)
Biometrika
 , 
1965
, vol. 
52
 (pg. 
591
-
611
)
Singh
A
Chen
EY
Lugovoy
JM
Chang
CN
Hitzeman
RA
Seeburg
PH
Saccharomyces cerevisiae contains two discrete genes coding for the α-factor pheromone
Nucleic Acids Res
 , 
1983
, vol. 
11
 (pg. 
4049
-
4063
)
Steenkamp
ET
Wingfield
BD
Coutinho
TA
Wingfield
MJ
Marasas
WFO
Differentiation of Fusarium subglutinans f. sp. pini by histone gene sequence data
Appl Environ Microbiol
 , 
1999
, vol. 
65
 (pg. 
3401
-
3406
)
Steenkamp
ET
Wingfield
BD
Desjardins
AE
Marasas
WFO
Wingfield
MJ
Cryptic speciation in Fusarium subglutinans
Mycologia
 , 
2002
, vol. 
94
 (pg. 
1032
-
1043
)
Stötzler
D
Duntze
W
Isolation and characterization of four related peptides exhibiting alpha factor activity from Saccharomyces cerevisiae
Eur J Biochem
 , 
1976
, vol. 
65
 (pg. 
257
-
262
)
Swanson
WJ
Vacquier
VD
Concerted evolution in an egg receptor for a rapidly evolving abalone sperm protein
Science
 , 
1998
, vol. 
281
 (pg. 
710
-
712
)
Swanson
WJ
Vacquier
VD
The rapid evolution of reproductive proteins
Nat Rev Genet
 , 
2002
, vol. 
3
 (pg. 
137
-
144
)
Tamura
K
Dudley
J
Nei
M
Kumar
S
MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0
Mol Biol Evol
 , 
2007
, vol. 
24
 (pg. 
1596
-
1599
)
Tatusova
TA
Madden
TL
BLAST 2 sequences, a new tool for comparing protein and nucleotide sequences
FEMS Microbiol Lett
 , 
1999
, vol. 
174
 (pg. 
247
-
250
)
Thompson
JD
Higgins
DG
Gibson
TJ
CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Res
 , 
1994
, vol. 
22
 (pg. 
4673
-
4680
)
Turgeon
BG
Application of mating type gene technology to problems in fungal biology
Ann Rev Phytopathol
 , 
1998
, vol. 
36
 (pg. 
115
-
137
)
Turner
E
Jacobson
DJ
Taylor
JW
Reinforced postmating reproductive isolation barriers in Neurospora, an Ascomycete microfungus
J Evol Biol
 , 
2010
, vol. 
23
 (pg. 
1642
-
1656
)
Wik
L
Karlsson
M
Johannesson
H
The evolutionary trajectory of the mating-type (mat) genes in Neurospora relates to reproductive behavior of taxa
BMC Evol Biol
 , 
2008
, vol. 
8
 pg. 
109
 
Wilcoxon
F
Individual comparisons by ranking methods
Biometrics
 , 
1945
, vol. 
1
 (pg. 
80
-
83
)
Wilkinson
L
MYSTAT: a personal version of SYSTAT
Am Stat
 , 
1987
, vol. 
41
 pg. 
334
 
Yang
Z
PAML: a program package for phylogenetic analysis by maximum likelihood
Comput Appl Biosci
 , 
1997
, vol. 
13
 (pg. 
555
-
556
)
Yang
Z
Nielsen
R
Codon-substitution models for detecting molecular adaptation at individual sites along specific lineages
Mol Biol and Evol
 , 
2002
, vol. 
19
 (pg. 
908
-
917
)
Yang
ZH
Nielsen
R
Goldman
N
Pedersen
AMK
Codon-substitution models for heterogeneous selection pressure at amino acid sites
Genetics
 , 
2000
, vol. 
155
 (pg. 
431
-
449
)
Yang
ZH
Swanson
WJ
Vacquier
VD
Maximum-likelihood analysis of molecular adaptation in abalone sperm lysin reveals variable selective pressures among lineages and sites
Mol Biol Evol
 , 
2000
, vol. 
17
 (pg. 
1446
-
1455
)
Zhang
L
Baasiri
RA
Van Alfen
NK
Viral repression of fungal pheromone precursor gene expression
Mol Cell Biol
 , 
1998
, vol. 
18
 (pg. 
953
-
959
)

Author notes

Associate editor: Helen Piontkivska