| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Previous Article | Next Article 
Eukaryotic Cell, April 2005, p. 661-672, Vol. 4, No. 4
1535-9778/05/$08.00+0 doi:10.1128/EC.4.4.661-672.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
-Factor-Like Sordaria macrospora Peptide Pheromone and Analysis of Its Interaction with Its Cognate Receptor in Saccharomyces cerevisiae
Lehrstuhl für Allgemeine und Molekulare Botanik, Ruhr-Universität, Bochum, Germany
Received 21 December 2004/ Accepted 12 February 2005
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-factor precursors and a-factor precursors of the yeast Saccharomyces cerevisiae. Likewise, sequence similarity has been found between the putative products of the pheromone receptor genes pre2 and pre1 and the S. cerevisiae Ste2p -factor receptor and Ste3p a-factor receptor, respectively. To investigate whether the -factor-like pheromone-receptor pair of S. macrospora is functional, a heterologous yeast assay was used. Our results show that the S. macrospora -factor-like pheromone precursor PPG1 is processed into an active pheromone by yeast MAT cells. The S. macrospora PRE2 protein was demonstrated to be a peptide pheromone receptor. In yeast MATa cells lacking the endogenous Ste2p receptor, the S. macrospora PRE2 receptor facilitated all aspects of the pheromone response. Using a synthetic peptide, we can now predict the sequence of one active form of the S. macrospora peptide pheromone. We proved that S. macrospora wild-type strains secrete an active pheromone into the culture medium and that disruption of the ppg1 gene in S. macrospora prevents pheromone production. However, loss of the ppg1 gene does not affect vegetative growth or fertility. Finally, we established the yeast assay as an easy and useful system for analyzing pheromone production in developmental mutants of S. macrospora. |
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pheromone response system of the heterothallic ascomycetous yeast Saccharomyces cerevisiae is a well-studied model of pheromone and pheromone receptor interaction. Binding of pheromones to their specific receptors triggers a G-protein-linked signal transduction pathway that induces the expression of several genes, facilitating the fusion of MATa and MAT cells to form diploid MATa/ cells (44). The peptide synthesized by MATa cells is derived from a precursor with a C-terminal CaaX (where C is cysteine, a is an aliphatic residue, and X is any amino acid residue) motif, a signal for carboxymethylation and farnesylation. The mature lipopeptide pheromone, called a-factor, is secreted via an ATP-binding cassette transporter (7). MAT cells produce -factor, which is secreted by the classical yeast secretory pathway (9, 35, 73). The -factor is a 13-amino-acid peptide derived from a precursor containing several copies of the 13-amino-acid sequence in tandem repeats. The first step of precursor processing is the removal of the signal sequence of the prepro--factor (79). Further processing includes the Kex2 endoprotease, which cleaves the pro-protein at characteristic KR dipeptides (36), the dipeptidyl-aminopeptidase Ste13p, which recognizes the dipeptide XA or XP (35), and the carboxypeptidase Kex1p, which removes C-terminal extensions (16).
The S. cerevisiae pheromone receptors for a-factor (Ste3p) and -factor (Ste2p) are members of the large family of G-protein-coupled receptors, which contain seven-transmembrane domains. In contrast to ascomycetes, only lipopeptide pheromones of the S. cerevisiae a-factor group and Ste3p-like lipopeptide receptors have been found in basidiomycetes (4, 12, 20, 43, 51, 70, 78). Interestingly, putative pheromone precursor genes encoding two different mating pheromones have been identified not only in heterothallic filamentous ascomycetes, such as Neurospora crassa, Cryphonectria parasitica, Magnaporthe grisea, and Podospora anserina, but also in the homothallic filamentous ascomycete Sordaria macrospora (3, 58, 69, 81).
The first step in the sexual reproduction of mycelial ascomycetes is to bring together two compatible nuclei in the same cell. In heterothallic species, a specialized hypha termed trichogyne is sent out from a female prefruiting body to grow towards a fertilizing male cell of the opposite mating type. A functional male cell may be a uninucleate spermatium or microconidium or a multinucleate macroconidium. Recently, it was demonstrated that male and female fertility of heterothallic mycelial ascomycetes depends on the interactions of pheromones with their specific receptors (17, 39, 40, 77). In the homothallic ascomycete S. macrospora, conidiospores, spermatia, and trichogynes are absent, suggesting that pheromones are not needed either for sensing a mating partner or for initializing fertilization events.
Nevertheless, two putative pheromone precursors genes, named ppg1 and ppg2, are transcriptionally expressed in S. macrospora. The ppg2 gene encodes a lipopeptide with a putative farnesylated and carboxymethylated C-terminal cysteine residue. The PPG1 precursor consists of a putative hydrophobic signal sequence and five repeats of the undecapeptide QWCRIHGQSCW. Each of the five repeated sequences is surrounded by maturation signals similar to those of the -factor pheromone precursors of S. cerevisiae (58) (Fig. 1). The presence of these maturation motifs suggests that the S. macrospora PPG1 precursor is processed to an undecapeptide pheromone, hereafter referred to as pheromone PHE1. In addition to pheromone genes, two pheromone receptor genes, termed pre1 and pre2, have also been identified and shown to be transcriptionally expressed in S. macrospora (59). The predicted products of both genes are proteins with seven transmembrane domains. The pre1 gene product shows extensive amino acid similarity to the a-factor receptor Ste3p of S. cerevisiae and to lipopeptide pheromone receptors of basidiomycetes. The S. macrospora pheromone receptor PRE2 exhibits significant sequence similarity to the S. cerevisiae Ste2p -factor receptor (59).
|
The fact that pheromone and pheromone receptor genes are transcribed in the homothallic ascomycete S. macrospora supports the idea that pheromones and receptors are functional in this fungus.
In order to examine the functionality of S. macrospora pheromone precursor PPG1 and pheromone receptor PRE2, the yeast S. cerevisiae was used as a heterologous assay system. A synthetic peptide was designed to analyze whether the undecapeptide predicted from the PPG1 sequence can act as an active peptide pheromone and trigger the pheromone response in the heterologous yeast system. Furthermore, we disrupted the pheromone precursor gene ppg1 of S. macrospora to investigate its involvement in fruiting body and ascospore development.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-mating pheromone was functioning as intended in the halo assay set-up, strain 786-11-1 was used as a control (32). Strains 7416-12-3 and 786-11-1 are both derived from yeast strain 381G (31). As appropriate, yeast strains were grown in YEPD or SD minimal medium lacking uracil or uracil and tryptophan (71). Transformation of S. cerevisiae was done by electroporation according to Becker and Lundblad (2) in a Multiporator (Eppendorf) at 1.5 kV. For each transformation, 100 ng of plasmid DNA was used.
|
Construction of plasmids. In order to determine whether a peptide pheromone encoded by the ppg1 gene is secreted by S. macrospora, we fused the signal sequence (SS) of the ppg1 gene in-frame to the egfp gene of plasmid p82.9 (61). The SS region of the ppg1 gene was reconstituted with two annealed complementary oligonucleotides (ppg1S1 and ppg1S2). The sequences of all oligonucleotides used in this work are listed in Table 2. The signal sequence was inserted into the NcoI site of p82.9 in the sense (pSppg1-1) and in the antisense (pSppg1-2) orientations. As a control, the signal sequence of an aspartic proteinase gene (etp) was fused either in sense in-frame with the egfp gene (pSetp-1) or antisense (pSetp-2) (61). The 1.4-kb EcoRI hph cassette of pCB1003 was inserted into all constructs to provide a selectable marker (8).
|
|
-factor (Trp-His-Trp-Leu-Gln-Leu-Lys-Pro-Gly-Gln-Pro-Met-Tyr) was obtained from Sigma-Aldrich (Munich, Germany; T6901). Pheromones were diluted in 2% dimethyl sulfoxide, and 1.5 nmol was used for halo and lacZ induction assays.
Determining the effects of S. macrospora pheromones on S. cerevisiae.
For halo assays, strain 7416-12-3 (MATa 
ste2 sst2-1) containing plasmid pPre2 or control plasmid pPGK was grown in SD medium lacking uracil to an optical density at 600 nm (OD600) of 1.0 to 1.3. One milliliter of cells was embedded in 5 ml of soft agar. To obtain a lawn, the cell-containing soft agar was plated onto SD plates lacking uracil and containing 0.1 M citric acid (pH 4.5). To test the interaction between the PRE2 receptor produced by the cells in the lawn and certain pheromones, 20 µl of the following solutions was applied to filter disks (6 mm diameter; Schleicher & Schuell) and placed onto the lawn, either yeast culture supernatants (MAT strain W303-1B containing pPpg1 or control plasmid pPGK) or culture medium of S. macrospora strains grown for 3 to 4 days in CM medium or 1.5 nmol of synthetic pheromones. The plates with the yeast lawns were grown for 24 to 48 h and analyzed for pheromone-induced cell cycle arrest by the halo assay. Measuring the distance from the center of the filter disks to the edge of the zone of growth inhibition determined the size of the halo. Plates were scanned electronically and recorded images were edited with Adobe Photoshop 6.0.
Analysis of shmoo formation was done according to Sprague (74). Yeast MATa cells were grown to a density of 5 x 106 to 107 cells/ml and were treated either with pheromone at a concentration of 5.0 µM or with dimethyl sulfoxide. Samples were removed at 0 h, 2 h, and 4 h and examined by differential interference contrast light microscopy on a Zeiss Axiophot microscope. Images were recorded with an MC80DX camera, and the Adobe Photoshop 6.0 software was used for further processing.
Yeast FUS1 and lacZ induction assays.
Yeast cell cultures of MAT transformants carrying plasmid pPpg1 or plasmid pPGK as a control were grown in SD medium lacking uracil. Supernatants were obtained by centrifugation when growth had reached an OD600 of 
1.0. Similarly, S. macrospora wild-type and mutant strains were grown in liquid CM medium, and cell-free supernatants were harvested after 3 to 4 days of growth.
Yeast MATa cells (7416-12-3) carrying pPre2 or the control vector pPGK, together with the pheromone-inducible reporter construct pTCFL1 (FUS1::lacZ) (11), were grown overnight to an OD600 of 1.0 in SD medium lacking uracil and tryptophan and harvested by centrifugation. Cell pellets were resuspended in 0.5 volume of fresh selective medium.
Equal volumes (10 ml) of receptor-containing MATa cells and supernatants of MAT S. cerevisiae or S. macrospora cultures containing the pheromones of interest were mixed and incubated in glass flasks at 30°C and 220 rpm for 2 h. In lacZ assays with synthetic pheromone, 1.5 nmol of synthetic PHE1 pheromone was added to 10 ml of MATa transformants expressing the S. macrospora pheromone receptor gene pre2. After the incubation period, the activity of ß-galactosidase was measured. This was done by determining o-nitrophenyl-ß-D-galactopyranoside cleavage by crude protein extracts (64). Crude protein extracts were obtained by destroying the yeast cells with glass beads (0.5-mm diameter) in the presence of the proteinase inhibitor phenylmethylsulfonyl fluoride. The method of Bradford (5) was used to determine the protein concentration. Enzyme activity was normalized to the protein concentration in the samples. Six to seven independent measurements of activity were taken from assays of two different transformants, and the mean activities were calculated.
Generation of ppg1-deficient S. macrospora mutants. A 1,622-bp EcoRI fragment containing the ppg1 coding sequence and its flanking regions was subcloned from cosmid H4 (58) into pBCKS(+) (Stratagene) to obtain plasmid p204-20. The hygromycin phosphotransferase gene (hph) within pCB1003 (8) was isolated as a 1.4-kb SalI fragment and cloned into the SalI cleavage site of p204-20 (see Table 3). This caused the desired disruption of the ORF of the ppg1 gene. The resulting plasmid construct was referred to as p210-1.
To insert the generated ppg1-hph construct into S. macrospora, a 3.4-kb fragment from p210-1 was amplified by PCR with primers pb41 and pb42 (Table 2). Transformation of S. macrospora was done according to Nowrousian et al. (54). ppg1 mutants were identified by Southern blot analysis. Genomic DNA was isolated as described previously (62). Successful homologous recombination was confirmed by PCR amplification with primers pb8 and tC1 for the 5'-flanking region and primers h3 and pb6 for the 3'-flanking region (see Table 2). Fungal transformants are often heterokaryotic, and mycelia carry transformed and nontransformed nuclei. A single spore isolate of a primary transformant carrying the disrupted ppg1 gene was therefore investigated further.
Immunodetection of secreted EGFP. EGFP secretion in S. macrospora was put under the control of the signal sequences from S. macrospora genes ppg1 and etp. To achieve this, the signal sequences were fused in the sense and antisense orientations in frame with the egfp gene. Immunodetection of secreted EGFP was performed as previously described (61). As a control, a pSM1 transformant expressing the egfp gene at high levels was used. In the pSM1 transformant, the EGFP protein is not secreted and is therefore located in the cytoplasm. After S. macrospora transformants had grown for 3 days in liquid CM medium, 5 ml of the cell-free medium was removed from the culture, and proteins secreted into the medium were precipitated by adding 10% trichloroacetic acid. Subsequently, precipitated proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis according to standard protocols (45). The proteins were transferred to a polyvinylidene difluoride Western blotting membrane (Biometra), using a semidry blotting system (Biometra). A polyclonal anti-GFP living color peptide antibody (BD Bioscience) (diluted 1:500) was used for the detection of EGFP. The chemiluminescence Western blotting kit (Roche) was used for visualization as described by the manufacturer.
Confocal laser microscopy. Confocal laser microscopy was carried out on a Zeiss LSM 510 META confocal system version 3.0 (excitation/emission 488 nm blue, 10% argon laser power, Ch2-1, BP505-550 filter, Ch2/72 µm pinhole, Zeiss Axiovert 100 M, Plan Apo 63/1.4 oil lens). Preparation of S. macrospora strains was done as described previously (61).
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-factor receptor Ste2p and the S. macrospora PRE2 receptor, we used strain 7416-12-3 (MATa ste2 sst2-1), lacking the S. cerevisiae STE2 gene, as a host for expression of the S. macrospora pre2 gene.
In an attempt to optimize the conditions for ppg1 expression and pheromone synthesis in an S. cerevisiae host, the MAT (W303-1B) strain was chosen because its MF1p and MF2p precursor proteins are structurally similar to the PPG1 prepro-protein of S. macrospora (58). To measure the activation of the heterologously expressed S. macrospora PRE2 receptor in MATa cells, an agar diffusion bioassay (halo assay) was employed. The assay is based on the fact that pheromone exposure triggers a cell division arrest in S. cerevisiae cells. In S. cerevisiae, binding of a pheromone to its receptor leads to the release of GDP from the G-protein -subunit Gpa1p, followed by the binding of GTP and the liberation of the G-protein ß
-subunits (Ste4p/Ste18p). The ß-dimer initiates the mitogen-activated protein kinase cascade, resulting in gene transcription, morphological changes, and a G1 cell cycle arrest (23, 25).
In addition to the ste2 mutation, the yeast MATa strain used in the halo assay also carried the sst2-1 mutation. The S. cerevisiae SST2 gene encodes a regulator of G-protein signaling (RGS) protein, which is crucial for recovery from pheromone-induced cell cycle arrest (22). Mutants lacking SST2 are approximately 100-fold more sensitive to pheromones (24). Supernatants of MAT transformants secreting an S. macrospora peptide pheromone were applied to filter disks and placed on a lawn of MATa cells expressing the S. macrospora pheromone receptor gene pre2 (Fig. 2A). As a control, the supernatant of ppg1-MAT transformants was applied to a MATa lawn containing only the mock vector pPGK. Alternatively, the supernatant of a MAT culture carrying pPGK was applied to a MATa pre2 lawn. Application of supernatants of S. cerevisiae cells secreting an S. macrospora peptide pheromone resulted in a bare zone immediately surrounding the filter disk. In the controls, no halo formation was observed (Fig. 2A). Hence, S. cerevisiae MAT cells expressing the S. macrospora ppg1 pheromone precursor gene must have been able to secrete an active pheromone which promotes cell cycle arrest in S. cerevisiae MATa cells expressing the S. macrospora pre2 gene.
|
cells seemed capable of processing the S. macrospora PPG1 precursor, the precise nature of the active pheromone remained unknown. So far, no pheromones have been purified from filamentous ascomycetes. However, since the overall structure of the PPG1 protein from S. macrospora resembles that of S. cerevisiae MF1p and MF2p with respect to signal sequence, cleavage sites for endopeptidases, dipeptidyl-aminopeptidases, and carboxypeptidases, we predicted that the active S. macrospora pheromone was an undecapeptide, QWCRIHGQSCW (Fig. 1).
To experimentally test the hypothesis that the undecapeptide PHE1 is one active form of an S. macrospora pheromone, the peptide QWCRIHGQSCW was synthesized in vitro; 1.5 nmol was added to an S. cerevisiae MATa ste2 sst2-1 lawn producing the S. macrospora receptor PRE2 (Fig. 2B). Halo formation occurred in response to the synthetic undecapeptide pheromone. However, no halo formation was seen when synthetic S. cerevisiae -factor or dimethyl sulfoxide was added in control experiments. As a positive control, we were able to induce halo formation when we applied the synthetic S. cerevisiae mating -factor to strain 786-11-1 (MATa sst2-1), which produces the S. cerevisiae -factor receptor Ste2p. Conversely, application of the synthetic PHE1 pheromone to MATa Ste2p cells resulted in no detectable halo (Fig. 2C). Thus, we were able to demonstrate that in S. cerevisiae cells expressing the S. macrospora pre2 gene, specific interaction occurs between the QWCRIHGQSCW undecapeptide PHE1 and the S. macrospora PRE2 receptor.
To quantify the extent of pheromone activation on gene expression, we used the pheromone-inducible FUS1-lacZ reporter gene (11, 76). S. cerevisiae cells expressing the S. macrospora receptor gene pre2 and the FUS1-lacZ reporter gene construct were treated with either culture supernatants of MAT cells containing pPpg1 or the control plasmid pPGK, synthetic pheromone, or 2% dimethyl sulfoxide (Fig. 3). The sixfold increase in ß-galactosidase activity in the reporter strains after addition of 2% dimethyl sulfoxide might be due to a nonspecific activation of the PRE2 receptor by the solvent. However, a 20-fold increase in ß-galactosidase activation in comparison to the negative controls was observed when the supernatant of MAT cells expressing the ppg1-encoded pheromone or the synthetic peptide QWCRIHGQSCW was added to the cells. This result suggests that the heterologous S. macrospora receptor PRE2 is capable of triggering the S. cerevisiae pheromone response pathway when activated by the synthetic peptide or the S. cerevisiae supernatant.
|
ste2 sst2-1 cells expressing the S. macrospora receptor PRE2 with synthetic pheromone for 2 h produced the characteristic morphological changes. The same extent of morphological changes was observed in the control experiments, where MATa sst2-1 cells (strain 786-11-1) carrying the endogenous Ste2p receptor were exposed to synthetic -factor (Fig. 4). In summary, we have shown that the S. macrospora pre2 gene, when expressed in an S. cerevisiae MATa ste2 mutant, can initiate the pheromone response pathway leading to cell cycle arrest, transcriptional activation of genes, and shmoo formation in the presence of S. macrospora peptide pheromone.
|
-factor-like pheromone is secreted by S. macrospora.
The secretion of diffusible peptide pheromones, which are recognized by pheromone receptors on the cell surface of a mating partner with an opposite mating type, initiates mating in heterothallic ascomycetes. In order to assess whether the homothallic fungus S. macrospora secretes a peptide pheromone into the extracellular medium and whether the ppg1-encoded pheromone was responsible for pheromone secretion, we disrupted the coding region of the ppg1 gene of S. macrospora by introducing a hygromycin resistance cassette (hph) by homologous recombination. The homologous recombination was confirmed by Southern blot and PCR analysis (Fig. 5). Phenotypic inspection of independent single-spore isolates revealed no differences from the wild type with respect to the timing of sexual development or the morphology and number of perithecia, asci, or ascospores.
|
|
On the basis of plasmid p82.9, we constructed plasmids pSppg1-1 and pSppg1-2. Plasmid pSppg1-1 contains the signal sequence in the sense orientation. As a control, it was introduced in the antisense orientation into pSppg1-2. Both plasmids were transformed into wild-type S. macrospora, and transformants were analyzed for EGFP fluorescence (Fig. 6B). Microscopic investigation revealed a different localization of EGFP in transformants containing pSppg1-1 than in transformants containing pSppg1-2. In transformants expressing the PPG1 signal sequence in the sense orientation, the GFP fluorescence appeared to be distributed in distinct patches in the fungal hyphae, suggesting localization within cytoplasmic vesicles. Transformants carrying pSetp1 were analyzed as a control. The plasmid pSetp1 contains the signal sequence of the S. macrospora etp gene fused to the egfp gene. Previously it has been shown that the signal sequence of the etp gene can mediate the secretion of EGFP in S. macrospora (61). As can be seen from Fig. 6B, transformants containing pSetp1 show the same localization of EGFP within cytoplasmic vesicles as pSppg1-1 transformants. In contrast, EGFP was distributed throughout the cytoplasm in transformants containing the PPG1 signal sequence in the antisense orientation (TpSppg1-2). A similar pattern was observed for pSetp2 transformants, which carry an antisense ETP signal sequence fused to EGFP and for pSM1 transformants expressing the egfp gene.
Extracellular culture broth of S. macrospora transformants was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting with a polyclonal antibody against GFP in order to prove the presence of EGFP in the culture medium (Fig. 6C). The 27-kDa EGFP polypeptide was detected in the culture medium of pSppg1-1 and pSetp1 transformants. No signal was detected in culture supernatants of wild-type, pSM1, pSppg1-2, and pSetp2 transformants. Thus, the signal sequence derived from PPG1 was shown to be sufficient to mediate secretion of EGFP.
S. cerevisiae as a powerful screening system for the identification of pheromone mutants of S. macrospora. In heterothallic species, mutations conferring male and/or female sterility can be detected directly because of their sterility effects in heterozygous crosses. However, recessive mutations that effect postfertilization fruiting-body development will remain undetected in heterothallic species until the mutant allele is available in both mating types, thus allowing homozygous crosses. In the homothallic S. macrospora, recessive mutations can be tested directly for defects in fruiting-body development. Therefore, S. macrospora was used to generate numerous mutants which are blocked at various stages of sexual development (28, 49, 50). These mutants were divided into four main groups: mutants displaying defects in ascogonium development, called asc mutants; mutants affected in early fruiting-body development, named pro (protoperithecia) mutants; mutants that form perithecia but no ascospores, termed per (perithecia) mutants; and mutants generating malformed and piled fruiting bodies, termed pile mutants (49). Using the halo assay, we have been able to demonstrate that the wild-type strain of S. macrospora does secrete a peptide pheromone.
Since pheromones are supposed to be involved in postfertilization events and fruiting-body development (18), we wanted to test if this assay may be suitable for analyzing pheromone production in the S. macrospora fruiting-body mutants. Three S. macrospora asc mutants, 37 pro mutants, 36 per mutants, and 5 pile mutants were screened by the halo assay. A lawn of S. cerevisiae strain 7416-12-3 (MATa ste2 sst2-1) expressing the S. macrospora pre2 pheromone receptor gene was used. For each of the 81 S. macrospora mutants tested, 20 µl of culture medium was applied to an individual S. cerevisiae lawn. The culture medium of 54 mutants caused halo formation, which was approximately the same size as the halo seen with the culture medium of wild-type S. macrospora (Table 4). However, a change in production or secretion of the peptide pheromone was seen in several mutants. Compared to the halo formed with culture medium from wild-type S. macrospora, the culture medium of 12 mutants caused larger halos. In 15 cases, smaller halos were observed (Table 4).
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-factor-like pheromone precursor PPG1 and its cognate receptor PRE2 in the yeast S. cerevisiae to provide experimental evidence that the pheromones and receptors of S. macrospora are truly functional.
This study demonstrated that the S. macrospora pheromone precursor gene ppg1 and the pheromone receptor gene pre2 are functionally homologous to the S. cerevisiae MF1 and MF2 pheromone genes and the STE2 receptor gene, respectively. Despite the fact that the amino acid sequence of the S. macrospora PRE2 receptor shows limited similarity to the S. cerevisiae Ste2p -factor receptor sequence (23.0% identity over 274 amino acids), PRE2 does act as a pheromone receptor in an ste2-deficient S. cerevisiae mutant (MATa ste2 sst2-1). Activation of the S. cerevisiae pheromone response pathway through heterologously expressed receptors has so far been demonstrated for a number of mammalian G-protein-coupled receptors (26, 41, 63, 66, 75) and for a-factor-like pheromone receptors from basidiomycetes (29, 33, 57).
-factor, which activates the Ste2p receptor in S. cerevisiae, is a 13-amino-acid peptide. It is processed from larger polypeptide precursors containing several copies of the active tridecapeptide. Similarly, an active S. macrospora peptide pheromone is thought to be an undecapeptide derived from the PPG1 precursor. This precursor contains five copies of the amino acid sequence QWCRIHGQSCW (58). In this work we demonstrated that a peptide pheromone secreted by S. cerevisiae MAT cells expressing the ppg1 gene interacts with the PRE2 receptor. So far, only a-factor-like pheromones of the basidiomycetes Coprinus cinereus and Schizophyllum commune have been heterologously expressed in S. cerevisiae (29, 57). Furthermore, we were able to show that the synthetic undecapeptide PHE1 (QWCRIHGQSCW) was biologically active when tested in the S. cerevisiae system, suggesting that the synthetic peptide might represent one active form of the S. macrospora peptide pheromone. The lowest concentration of synthetic PHE1 to cause visible halo formation was found to be 1.5 pmol.
The size of the halo formed when synthetic PHE1 was added to a MATa ste2-PRE2 lawn was compared to the halo seen in a MATa lawn treated with synthetic -factor. PHE1 seemed to have a more profound effect than the -factor (Fig. 2B and C). One possible explanation for this is the presence of the extracellular protease Bar1p in S. cerevisiae cells. The enzyme encoded by the BAR1 gene cleaves S. cerevisiae -factor between leucine 6 and lysine 7, rendering it nonfunctional. The enzymatic degradation of the -factor enables the target cells to recover from arrest in the G1 phase of the cell cycle (13, 48). The PHE1 pheromone does not contain leucine or lysine residues. This may protect it from enzymatic cleavage, making it more stable than S. cerevisiae -factor.
Activation of the heterologously expressed PRE2 receptor by synthetic pheromone PHE1 or an S. macrospora peptide pheromone secreted from MAT cells initiates the S. cerevisiae pheromone response pathway, resulting in characteristic changes. These include cell cycle arrest, gene activation, and the formation of one or more projections on the cell surface, commonly termed shmoo morphology (Fig. 2 to 4). It has previously been shown that pheromone receptors of basidiomycetes heterologously expressed in an S. cerevisiae host can initiate gene expression and cell cycle arrest in response to their cognate pheromone (29, 33, 57). However, formation of the shmoo morphology upon pheromone application has not been observed so far in experiments expressing basidiomycete receptors in S. cerevisiae (29). Dose-response curves have been determined for each of the three responses as a function of -factor concentration in S. cerevisiae. The analysis revealed that approximately 100-fold-higher concentrations of pheromones are required for the induction of shmoo formation than for transcriptional activation or cell cycle arrest (52). In contrast to the S. commune receptor, the S. macrospora PRE2 receptor is able to mediate all pheromone responses when expressed in S. cerevisiae (29).
Since, in S. cerevisiae, PRE2 was activated by extracellular pheromones and was able to trigger the pheromone response, it seems likely that PRE2 has been incorporated into the plasma membrane and might interact with the Gpa1p protein of S. cerevisiae. This hypothesis is supported by the fact that the MATa ste2 strain used in our analysis carried a mutation in the SST2 gene. Sst2p acts in desensitization to pheromone through negatively regulating the S. cerevisiae G-subunit Gpa1p. S. cerevisiae sst2-1 mutants are therefore at least 100-fold more sensitive to pheromones and fail to recover from pheromone-induced cell cycle arrest (10, 22, 24). Accordingly, increased sensitivity to pheromone was present in Sst2p-deficient MATa ste2 cells expressing PRE2, but PHE1 never caused cell cycle arrest in wild-type SST2 strains (our unpublished results). The third intracellular loop of the S. cerevisiae Ste2p receptor is important for G-protein activation (14). However, no homologies in the amino acid sequences of Ste2p and PRE2 were found in the region predicted to form the third intracellular loop (59). Thus, properties other than sequence homology have to account for the ability of PRE2 and the S. cerevisiae G-protein to interact.
The importance of heterotrimeric G-proteins during fertilization and sexual development has been demonstrated for the heterothallic fungus Neurospora crassa, which is closely related to S. macrospora (1, 34, 38, 80). The N. crassa a-factor-like pheromone receptor PRE1 is thought to couple to the G-protein -subunit GNA-1 of the heterotrimeric G-protein, thereby transferring the pheromone signal (39). A gene encoding a protein similar to the GNA-1 subunit of N. crassa has been identified in S. macrospora, and the PRE2 receptor may interact with this protein in a similar way as hypothesized for N. crassa. In addition to a gna-1 homologue, several genes encoding putative downstream components of the pheromone signal transduction cascade, including genes coding for subunits of G-proteins, RGS proteins, mitogen-activated protein kinase cascade components, and transcription factors have been isolated from S. macrospora (Table 5) (56). These findings suggest not only that pheromones and pheromone receptors are similar in structure and functionality, but also that the signal transduction pathways in S. macrospora, N. crassa, and S. cerevisiae are highly conserved.
|
The SignalP version 1.1 program (53) predicted a hydrophobic secretion signal in the N terminus of the S. macrospora PPG1 protein, which was expected to be cleaved between amino acids 16 and 17 (Fig. 1). The signal sequence of the S. cerevisiae -factor precursor MF1p has been used to manipulate secretion of heterologous proteins in S. cerevisiae for several years (6). The GFP fluorescence of transformants carrying the PPG1 signal sequence-egfp sense construct appeared as distinct patches distributed along the fungal hyphae, suggesting localization within cytoplasmic vesicles. Fluorescence signals in PPG1 signal sequence-egfp antisense transformants were distributed evenly throughout the cytoplasm (Fig. 6B). By means of Western blot analysis, we demonstrated that the PPG1 signal peptide carries sufficient information to enable EGFP to pass through the secretory pathway and to be released into the culture medium.
Using PCR, we identified the sequences of putative homologues of S. cerevisiae genes encoding the -factor precursor processing enzymes Kex2p and Kex1p in the S. macrospora genome (Table 5). Reverse transcription-PCR and Northern blot analysis confirmed that these genes are transcriptionally expressed in S. macrospora (data not shown). Thus, processing of the PPG1 precursor in S. macrospora may follow the same pathway as described in S. cerevisiae. According to our findings, we postulate that the undecapeptide QWCRIHGQSCW most likely represents one active form of an -factor-like pheromone in S. macrospora because it can activate the PRE2 receptor in the heterologous S. cerevisiae system. However, based on our data, we cannot exclude the possibility that longer or shorter forms of the peptide pheromone exert a pheromone function in S. macrospora or that the active peptide in the S. cerevisiae assay may have a different structure than the native, mature pheromone produced in S. macrospora.
Disruption of the S. macrospora ppg1 gene prevents secretion of the peptide pheromone but does not affect sexual development. After disrupting the ppg1 gene, we were no longer able to induce halo formation in an S. cerevisiae MATa lawn expressing the pre2 gene when we added culture medium from the S. macrospora ppg1 mutant. Thus, we proved that disruption of the ppg1 gene resulted in a defect of peptide pheromone production and secretion. However, no other phenotypic changes have been observed in the S. macrospora ppg1 mutant. Despite lacking a functional ppg1 gene, the S. macrospora ppg1 mutant was able to develop fruiting bodies, asci, and ascospores, and its vegetative growth was not affected.
The homothallic ascomycete S. macrospora expresses both pheromone genes throughout its development, whereas heterothallic ascomycetes express pheromone genes predominantly in conidia, and expression occurs in a mating type-specific manner (3, 17, 58, 69, 81). Deletion of pheromone genes in the heterothallic ascomycetes Podospora anserina and Cryphonectria parasitica revealed that pheromones play an essential role in promoting fertilization. Strains carrying these mutations are male sterile because their conidia are unable to attract and mate with cells of the opposite mating type. In P. anserina, deletion of pheromone genes affects male fertility without impairing vegetative growth or postfertilization events (17). The same observation was made in C. parasitica, where the Mf1-1 gene encoding the -factor-like pheromone was deleted by gene replacement (77). Our observations in S. macrospora were very similar, even though defects in male sterility could not be demonstrated because S. macrospora does not produce conidia. On the contrary, deletion of one of the two copies of the C. parasitica gene encoding the a-factor-like pheromone resulted in a pleiotropic phenotype. Mf2-2 mutants display reduced sexual reproduction. A cross between an Mf2-2 mutant (as the female) and a wild-type strain (as the male) produced only barren perithecia (82). It was therefore assumed that Mf2-2 of C. parasitica is required during a developmental phase after fertilization and that the CaaX-type pheromone acts in a dosage-specific manner in postfertilization events (77). Similarly, N. crassa mfa-1 mutants in which the critical cysteine residue of the open reading frame had been changed to the nonprenylatable residue tyrosine (YAAX mutants), as well as mutants with intact open reading frames but multiple mutations in the 3' noncoding region (CAAX mutants), were shown to display delayed and reduced vegetative growth and aberrant sexual development. Surprisingly, this pleiotropic phenotype was observed in both mating types. It was therefore postulated that the lipopeptide pheromone of N. crassa may have an additional role in cementing hyphae together to stabilize the structure of the perithecium (40).
Northern blotting and reverse transcription-PCR analysis indicated that, in N. crassa, in contrast to pheromone precursor genes, expression of the receptor genes does not occur in a mating type-specific manner (39, 59). Deletion of the N. crassa pre-1 gene encoding the a-factor-like receptor does not affect vegetative growth or male fertility. On the other hand, pre1 matA mutants are sterile females, because their trichogynes are unable to recognize and fuse with mata conidia (39). Deletion of two putative pheromone receptor genes, gprA and gprB, in the homothallic filamentous ascomycete Aspergillus nidulans resulted in the production of fewer and smaller fruiting bodies carrying a reduced number of ascospores. Fruiting-body formation in homothallic conditions was absent in the gprA gprB double mutant (68). This result implies that a pheromone-receptor system is essential for postfertilization events during the sexual development of homothallic ascomycetes. It has to be expected that the second active pheromone-receptor pair (PPG2 and PRE1) of S. macrospora is able to functionally complement the disruption of the ppg1 gene in S. macrospora. This would explain why no defects in sexual development were detected in the S. macrospora ppg1 mutant.
The S. cerevisiae system provides a tool to measure pheromone secretion in developmental mutants of S. macrospora. Expressing the functional S. macrospora receptor PRE2 in S. cerevisiae, we developed a powerful tool for studying pheromone secretion in developmental mutants of S. macrospora. Of 81 developmental mutants tested in this work, 27 secreted a different amount of peptide pheromone than the wild type (Table 4). The culture medium of these 27 strains led to the formation of either smaller or larger halos. The halo assay results for two mutants, per44 (larger halo) and per46 (no halo), were confirmed by testing their culture medium for activation of the pheromone-inducible FUS1-lacZ reporter (Fig. 7). Induction of the reporter gene was only slightly decreased with the supernatant of mutant per46. We detected a twofold increase in the induction of FUS1-lacZ with the supernatant of mutant per44. The FUS1-lacZ reporter assay was therefore more sensitive than the halo assay. Northern blot analysis showed that transcriptional expression of ppg1 was elevated in mutant per44, while in per46 the transcript level was only slightly decreased (our unpublished results). Currently we do not know which genes are defective in mutants per44 and per46.
However, four of the mutants screened (pro1, pro1, pro22, and per5) have previously been complemented by transformation with an indexed cosmid library. The pro1 gene encodes a transcription factor, and the pro22 gene codes for a putative membrane protein (50, 55). The PRO11 protein is a membrane-associated protein and thought to be a scaffolding protein in several signal transduction pathways (60). The defect in mutant per5 lies in the acl1 gene, encoding a subunit of an ATP-citrate lyase (54). The culture medium of per5 led to wild-type-like halo formation in the S. cerevisiae assay, whereas the culture medium of pro1, pro11, and pro22 caused increased halo sizes. In this context, it is of interest that interspecies microarray hybridization of N. crassa with targets of S. macrospora mutants pro1, pro11, and pro22 revealed that the pheromone precursor gene ppg1 is among the genes which are upregulated more than twofold in these mutants (55). These findings indicate that alterations in pheromone gene expression may be involved in the sexual development of homothallic ascomycetes.
The heterologous S. cerevisiae system offers an easy and valuable tool to identify mutants carrying a defect in ppg1 gene expression and pheromone secretion pathways. Further analysis of these mutants will help us to establish the way in which pheromones are processed in filamentous ascomycetes and how pheromones are involved in postfertilization events of filamentous ascomycetes.
|
ACKNOWLEDGMENTS |
|---|
This work was funded by the Deutsche Forschungsgemeinschaft SFB480 (Bonn, Germany) and the Ruhr-University of Bochum.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-factor pheromones. Mol. Cell. Biol. 2:11-20. subunits in the yeast pheromone response pathway. Mol. Cell. Biol. 14:1054-1065.
