INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

The life cycle of the yeast Saccharomyces cerevisiae comprises a series of interconnected growth states and developmental programs that are under both genetic and environmental control. Meiotic development (sporulation) is induced when a diploid cell is starved for essential nutrients and a fermentable carbon source (21). Following induction, the cell withdraws from the mitotic cycle at G₁ and enters meiotic prophase, during which a single round of DNA replication, synaptonemal complex formation, and recombination occur. Meiotic prophase is followed by the meiosis I reductional and meiosis II equational divisions. Spore wall morphogenesis initiates with the outgrowth of a double membranous structure (the prospore wall) from the outer plaques of each meiosis II spindle pole body which envelops each haploid meiotic product. Two layers that appear similar to the vegetative cell wall and two protective spore-specific layers are subsequently assembled from within and around the prospore wall. The end product of sporulation is the differentiated ascus which contains four dormant haploid spores. The tightly regulated sequence of cell cycle and morphogenetic events that occur during sporulation provides a model system for study of the mechanisms that regulate and coordinate development.

During sporulation, specific sets of genes that can be classified as early, middle, mid-late, or late are sequentially expressed. Early genes are expressed at the onset of sporulation and are involved in events including formation of the synaptonemal complex and the meiotic divisions. The transcriptional regulation of the early genes has been extensively studied, and regulatory cis-acting DNA sequences and proteins that bind to many of these sites have been identified (21, 26). One element commonly found in early promoters is recognized by the Abf1p DNA-binding protein, a general transcriptional activator (16). URS1, a second cis-acting element found in most and perhaps all early promoters, is recognized by the Ume6p DNA-binding protein (26, 29, 35, 37). During vegetative growth, Ume6p interacts with a complex of proteins which includes Sin3p and the Rpd3p histone deacetylase, to repress transcription (19). During early sporulation, changes in the Ume6p-complex occur that lead to transient derepression as well as transcriptional activation (4, 35). Gene products required for the developmentally regulated conversion from a repression to an activation complex include the Rim11p protein kinase, as well as Ime1p, which has transcriptional activation properties and may interact directly with Ume6p (3, 30).

In comparison to the early meiotic genes, less is known about transcriptional regulation of the later temporal classes of sporulation-specific genes. The cis-acting promoter sequences responsible for middle-transcriptional regulation have been identified in the SPS4 and SPR3 genes (17, 27, 28). The SPR3 promoter contains a consensus Abf1p binding site which is required for transcriptional activation. An additional site found in both SPS4 and SPR3, as well as in several other middle promoters, is the MSE (middle sporulation element), which has been shown to be required for the activation of middle gene expression and to be capable of activating heterologous promoters during middle sporulation (17, 28). Recently, Ndt80p has been shown to specifically interact with MSE DNA in vitro and to activate middle sporulation-specific gene expression in vivo (7). NDT80 is expressed predominantly as a middle gene, and it has been proposed that completion of a meiotic checkpoint is required for its full activity (7, 17a, 43). Mid-late and late genes appear to require distinct cis-acting elements. A negative regulatory element (NRE^DIT) that requires the Ssn6p and Tup1p corepressor complex regulates the divergently transcribed mid-late DIT1 and DIT2 genes (14). NRE^DIT derepression during sporulation requires complex interactions with at least two additional cis-acting elements. Thus, the different temporal classes of sporulation-specific promoters require multiple transcriptional activators that are expressed at different times during sporulation, as well as distinct transcriptional-repression pathways.

SMK1 encodes a middle sporulation-specific mitogen-activated protein (MAP) kinase homolog that is required for spore wall morphogenesis. In an smk1 null homozygous mutant, multiple abnormal spore wall assembly patterns are observed within a single ascus, with spore wall layers inverted or missing (20). Different smk1 missense mutants stall at distinct stages in spore morphogenesis. Furthermore, modest increases in dosage of certain smk1 alleles can restore wild-type spore wall morphogenesis, and different spore resistance phenotypes require distinct smk1 allelic thresholds (41, 42). These results show that SMK1 plays a central role in coordinating multiple events that are required for spore formation and suggest that different morphogenetic events can have distinct MAP kinase threshold requirements.

The transcriptional regulation of SMK1 raises a number of interesting questions that are now amenable to experimental investigation. What are the cis-acting promoter elements that regulate SMK1 expression? What is the functional significance of the tight control of SMK1 transcription during spore development? Can promoter mutants cause distinct phenotypes similar to those seen with the amino acid substitution mutants? Analysis of SMK1 transcriptional regulation should also provide insights on how the expression of different temporal classes of genes is regulated during development.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Yeast strains and culture conditions. The genotypes and sources of yeast strains used in this study are shown in Table 1. Vegetative cultures were propagated in either YEPD (1% yeast extract, 2% peptone, 2% glucose), YEPA (1% yeast extract, 2% peptone, 2% potassium acetate), SD (0.67% Difco yeast nitrogen base without amino acids, 2% glucose) or SA (0.67% yeast nitrogen base without amino acids, 1% potassium acetate, 1% phthalic acid [pH 5.5]) supplemented with nutrients essential for auxotrophic strains at the levels specified by Sherman et al. (31). For synchronous sporulation of strains containing plasmids, cells were pregrown in SA-uracil, and for strains lacking plasmids, cells were pregrown in YEPD. Logarithmically growing cultures were used to inoculate YEPA, the cultures were expanded overnight at 30°C to a density of 1 × 10⁷ cells/ml (4 to 5 generations), and cells were collected by centrifugation, washed once in 2% potassium acetate, and resuspended at 4 × 10⁷ cells/ml in SM (2% potassium acetate, 10 µg of adenine/ml, 5 µg of histidine/ml, 30 µg of leucine/ml, 7.5 µg of lysine/ml, 10 µg of tryptophan/ml, 5 µg of uracil/ml) prewarmed to 30°C. Under these conditions, the strains used in this study completed spore formation within 12 h. For testing the effects of ume6, sin3, and rpd3 mutations on SMK1 expression, the FT5-derived strains described by Kadosh and Struhl were used (19). The effects of ssn6 and tup1 mutations were tested by using the strains described by Friesen et al. (14).

Plasmids. Plasmids, markers, and sources are detailed in Table 2. Mutations in the SMK1 promoter were introduced by using PCR-based strategies and confirmed by dideoxynucleotide sequence analysis. The SMK1-lacZ reporter plasmid pMDP83 was constructed by replacing the BglII-SalI fragment of pLAK42 (codons 183 to 388 of SMK1 and the 3' noncoding region) with a 3,072-bp BglII-SalI fragment containing lacZ generated by PCR using pMC1871 as a template. The 5' primer changed the first and last nucleotides of the BamHI site found in pMC1871 to A and T, respectively, to generate the BglII site, and the 3' primer included the SalI site in pMC1871. Thus, pMDP83 contains 219 bp of the SMK1 promoter and 546 bp of the SMK1 coding region fused in frame to lacZ. SMK1 promoter deletion constructs pMDP86, -89, -126, -95, and -92 were generated by replacing the KpnI-BglII fragment of pMDP83 with KpnI-BglII fragments generated by PCR that lacked the indicated base pairs. The pseudo-urs1^S mutation removed bp 84 to 90, and the mse^S mutation changed the sequence TTTG at positions 79 to 76 to CCCA. Both of these mutations were introduced by using an XhoI site that was introduced into the SMK1 promoter by a TC AG double substitution at positions 68 and 65, respectively (see Fig. 1). In control experiments the XhoI site reduced the level of expression twofold but had no effect on vegetative repression or the timing or pattern of SMK1 expression. To test for mutant smk1 promoter phenotypes, an integrating SMK1 plasmid was constructed by subcloning the KpnI-XhoI fragment of SMK1 (containing the entire SMK1 gene and its promoter) into these sites in pRS406 to generate pMDP199. Promoter mutations were introduced into pMDP199 by replacing the KpnI/BglII fragment of pMDP199 with the KpnI/BglII fragments from the lacZ expression plasmids. The SMK1 promoter in pLAK42 was replaced with a 200-bp PCR fragment of the HOP1 promoter (measured from the ATG) to generate pJS1 by using the unique BstXI site located at codon 13 of the SMK1 gene and the unique KpnI site at the plasmid/insert junction.

Assays for spore wall assembly. For light microscopy, cells were fixed in ethanol and stained with the DNA-specific dye 4',6-diamidino-2-phenylindole (DAPI) (31). Samples were viewed and photographed as a wet mount under phase-contrast oil immersion optics by using a Nikon Optiphot equipped for epifluorescence. The procedure for the fluorescence assay has been described previously (42). Spore viability after heat shock (40 min at 55°C) or treatment with glusulase (1 h at 26°C) was determined as described by Briza et al. (5). Sensitivity of cells to ether exposure (3 min with constant gentle rocking) was assayed according to the method of Dawes and Hardie (8).

EMSA. For electrophoretic mobility shift assays (EMSA), oligonucleotides were end labeled with [-³²P]ATP by using polynucleotide kinase and were purified by Nensorb columns (NEN). The oligonucleotides were made double-stranded by mixing with a threefold excess of the matching strand, incubating at 90°C for 20 min, and slowly cooling to 25°C overnight. Binding reactions were carried out in a solution containing 10 mM Tris-HCl (pH 7.5), 40 mM NaCl, 4 mM MgCl₂, 6% (wt/vol) glycerol, 10 µg of sonicated salmon sperm DNA/ml, and ³²P-labeled oligonucleotide (10,000 cpm) in a total volume of 20 µl at room temperature for 20 min. Crude extracts and partially purified Abf1p were prepared as previously described (16). Protein dilutions were made in a solution containing 20 mM Tris-HCl (pH 8), 50 mM NaCl, 1 mM EDTA, 1 mg of bovine serum albumin (BSA)/ml, 5 mM -mercaptoethanol, and 1 mM phenylmethylsulfonyl fluoride (PMSF). Samples were analyzed on a 6% polyacrylamide gel (run in 0.5× Tris-borate-EDTA [TBE] buffer for 60 min at 250 V). Gels were dried after electrophoresis, exposed to a phosphor screen, scanned, and quantitated on a Model 425E Molecular Dynamics PhosphorImager.

Miscellaneous methods. Preparation of total RNA and Northern blot hybridization analysis were carried out as previously described (20). DNA probes from the coding regions of the indicated genes were isolated from preparative agarose gels, radiolabeled with ³²P by random priming (2), and used in hybridization analysis at 10⁶ dpm/ml. The DNA hybridization probe used for SMK1 consisted of the 0.8-kb StyI fragment of its coding region. This probe does not detect any transcripts that may be produced at the smk1::LEU2 locus, since it is internal to the boundaries of this deletion/insertion mutation (20). The DNA probe used for SPO12 was the 0.4-kb EcoRI-BamHI fragment of its coding region, and that for HOP1 was the BamHI-SacI restriction fragment of its coding region. The PC4/2 control probe for RNA loading has been described previously (22, 38). For -galactosidase assays, cell pellets were frozen at 80°C and subsequently resuspended with an equal volume of glass beads and breaking buffer (100 mM Tris-HCl, 1 mM dithiothreitol, 20% glycerol, 1 mM PMSF). Cell lysis was achieved by vortexing for seven 1-min intervals separated by cooling on ice. Samples were centrifuged for 10 min at 13,000 × g and stored at 80°C. Protein concentrations were determined by the method of Bradford with BSA as a standard, and -galactosidase was assayed by using o-nitrophenyl--D-galactopyranoside (ONPG). Oligonucleotides were synthesized on an Applied Biosystems 392-5 DNA synthesizer, and when appropriate, purified by C₁₈ reverse-phase high-performance liquid chromatography (HPLC).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

The SMK1 promoter contains distinct activating and repressing transcriptional control elements. We have previously shown that SMK1 is expressed as a middle sporulation-specific gene (20). To assay the SMK1 promoter we constructed pMDP83, which contains 219 bp of the promoter and 546 bp of the coding region fused in frame to lacZ (Table 2). The SK1 strain background (LNY150; Table 1) was used for these studies because it can be induced to undergo sporulation in a relatively rapid and synchronous manner. pMDP83 diploid transformants were sporulated, cells were harvested at 2-h intervals, and -galactosidase activities were assayed. Figure 1A compares the SMK1 mRNA expression profile directly determined by Northern blot hybridization analysis (20) to the lacZ expression profile. The levels of -galactosidase and endogenous SMK1 mRNA were low in vegetative cells and remained low until around the time of completion of meiosis II (6 h). Both -galactosidase and SMK1 mRNA levels peaked around the time when the major steps of spore wall morphogenesis were occurring (8 h). -Galactosidase levels remained high and decayed relatively slowly thereafter. In contrast, SMK1 mRNA levels decayed rapidly. These results demonstrate that 219 bp of the SMK1 promoter are sufficient for the regulated expression of SMK1. In addition, these results show that SMK1-lacZ fusions can be used to assay the onset and magnitude of SMK1 transcription.

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Analysis of the SMK1 promoter using SMK1-lacZ expression plasmids. (A) Comparison of relative SMK1 mRNA levels and SMK1-lacZ enzyme activities during sporulation. -Galactosidase activity levels were measured at 2-h intervals postinduction by using strain LNY150 transformed with pMDP83. Activity is expressed as units of -galactosidase per milligram of total protein. SMK1 mRNA levels were quantitated by Northern blot hybridization (20). (B) Deletion analysis of the SMK1 promoter. The indicated deletions were generated from the 219 promoter construct pMDP83 (see Table 2). -Galactosidase activities were determined in vegetative cells (V) and 10 h postinduction (S) and are averages from at least two separate determinations. In each case the deletion point indicates the number of base pairs between the initiator ATG and the C that is the most 3' base of the KpnI restriction endonuclease site of the parental plasmid. (C) Nucleotide sequence of the SMK1 promoter. Deletion endpoints diagrammed in panel B are indicated (*), and the initiator ATG of SMK1 is boldfaced. Consensus elements referred to in the text and the KpnI restriction enzyme site used in the construction of pMDP83 are underlined.

Abf1p interacts with UAS^S. The Abf1p DNA-binding protein is required for the regulated expression of many diverse genes. It appears to play an important role in sporulation-specific gene expression, and Abf1p-binding sites are found in a high percentage of sporulation-specific promoters (16, 27, 28). High-affinity Abf1p binding in vitro has previously been shown to require the 13-bp sequence RTCRYBNNNNACG (where Y is pyrimidine, R is purine, B is G, C, or T, and N is any base) (10, 12, 13). Within the 28-bp region that is required for UAS^S activity, an element that conforms to this consensus is found starting at position 122 (Fig. 1C). We tested whether Abf1p would bind to this site in vitro. As a positive control, binding to the characterized Abf1p-binding site, which is required for transcriptional activation of the early sporulation HOP1 promoter (UAS^H), was also assayed. As shown in Fig. 2A, Abf1p specifically bound to a 13-bp oligonucleotide duplex from UAS^S. Furthermore, substitution of the CG base pair corresponding to position 118 with an AT base pair (referred to below as the 118A substitution) eliminated detectable Abf1p binding, consistent with previously described binding requirements. In contrast, substitution of the GC base pair at position 115, which is found in the degenerate core of the Abf1p-binding site, with an AT base pair had no effect on binding. Although Abf1p bound to the UAS^S site, it bound with lower affinity than to the UAS^H site. We were therefore interested to determine if the SMK1 Abf1p site would be functional in the context of the HOP1 promoter and vice versa. It has previously been shown that removal of UAS^H from the HOP1 promoter reduces the early sporulation-specific expression peak (Fig. 2B) (39). Introduction of the SMK1 Abf1p-binding site into this mutant HOP1 promoter restored activity. Furthermore, the early sporulation timing of the HOP1 promoter containing the SMK1 Abf1p site was indistinguishable from the timing of the HOP1 promoter reconstituted with the HOP1 Abf1p site. Similarly, replacement of the Abf1p-binding site in the SMK1 promoter with the Abf1p-binding site from HOP1 had no detectable effect on the timing or level of SMK1 expression (Fig. 2B). These results show that the Abf1p sites from these promoters are functionally interchangeable. Thus, both biochemical and functional assays indicate that Abf1p can activate the SMK1 promoter. These results also suggest that different Abf1p-binding sites do not play a direct role in setting the timing of these temporally distinct classes of promoters.

View larger version (25K): [in this window] [in a new window]   FIG. 2.   Comparison of UASS and UASH sites. (A) EMSA of Abf1p binding. Partially purified Abf1p was serially diluted fivefold and used in binding reactions to radiolabeled 13-bp oligonucleotide duplexes containing the HOP1 Abf1p-binding site (lanes 1 to 5), the wild-type SMK1 Abf1p-binding site (lanes 6 to 10), and the mutant SMK1 Abf1p sites containing the 115A (lanes 11 to 15) and 118A (lanes 16 to 20) substitutions. (B) Analysis of HOP1 and SMK1 promoters containing heterologous Abf1p-binding sites. The HOP1 promoter (upper panel) lacking its Abf1p-binding element (open circles), reconstituted with the HOP1 binding site (open squares), or reconstituted with the SMK1 Abf1 site (closed squares) was tested by using the HOP1-lacZ plasmid pAV130, pCC83, or pJX33, respectively. The SMK1 promoter (lower panel) lacking a functional Abf1p-binding site (118A mutant; open circles), containing its normal Abf1p-binding site (open squares), or reconstituted with the Abf1p-binding site from HOP1 (closed squares) was tested for expression of -galactosidase by using the SMK1-lacZ plasmid pMDP126-118A, pMDP126, or pMDP113, respectively, in yeast strain LNY150 as described in Materials and Methods. The experiment was performed independently three times with similar results.

View larger version (16K): [in this window] [in a new window]   FIG. 3.   Mutational analysis of UASS. (A) Effects of UASS mutations on expression of SMK1 and binding to Abf1p. Strains with the indicated mutations in the 124 promoter SMK1-lacZ plasmid pMDP126 were assayed for -galactosidase activity in vegetative (V) and sporulated cultures at 10 h postinduction (S). -Galactosidase activities are averages from at least two separate comparisons to the wild-type plasmid (pMDP126). Both vegetative and sporulation values are shown as percentages of the wild-type sporulation value (79 ± 8 U/mg of total protein). Relative complex formation between oligonucleotide duplexes containing the indicated mutation and partially purified Abf1p (Binding) was quantitated from a single titration curve performed as shown in Fig. 2. The mutations are referred to by the numbers at the left. Sequence requirements for Abf1p binding (Abf1 con) are shown above for comparison. The HOP1-Abf1 mutation contains a 6-bp substitution to generate the HOP1 Abf1p-binding site, which reads 5'-ATCACTTCACACG-3'. (B) Hybridization analysis of UASS promoter mutants. Indicated promoter mutations in the context of the wild-type SMK1 gene were integrated at the ura3 locus, and RNA was prepared from vegetative (V) or sporulated (S) cultures 10 h postinduction. Hybridization analysis was performed on total RNA by Northern blot analysis using an SMK1-specific probe. The same filter was subsequently hybridized with the middle sporulation SPO12-specific probe as a normalization control. The SMK1-specific hybridization signal in sporulating samples (normalized for SPO12 hybridization) was reduced to 51, 48, 23, and 17% of the wild-type signal in the 116A, 115A, 115A 106A, 117A, and 118A mutant strains, respectively.

The MSE is required for vegetative repression as well as sporulation-specific activation of the SMK1 promoter. Within the 96-to-71 deletion interval that derepressed vegetative SMK1 expression, there are two sequence elements of interest. The first element is similar to the previously described MSE. The MSEs of the SPS4 and SPR3 middle genes have been shown to activate middle expression in heterologous promoter constructs (17, 27). A consensus MSE sequence (GNCRCAAAA/T) (28) is found at positions 72 to 80 of the SMK1 promoter, and we refer to it hereafter as MSE^S. The second DNA element of interest is similar to the consensus URS1 element (TCGGCGGCT) and is found at positions 84 to 92 of the SMK1 promoter (Fig. 1). URS1 sites have been shown to interact with Ume6p, and in the context of early sporulation promoters, they function to repress expression during vegetative growth (29, 37, 39). URS1 sites have also been shown to function as meiosis-specific activator elements (3, 30). The transcriptional regulatory properties of the URS1 consensus in SMK1 differs in certain respects from the URS1s that have been characterized in early sporulation-specific genes (see below). We therefore refer to this element in SMK1 as pseudo-URS1^S (or -URS1^S) hereafter.

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Expression of -galactosidase by mseS and pseudo-urs1S SMK1 promoter plasmids during sporulation. Yeast strain LNY150 transformed with a plasmid expressing -galactosidase under the control of the wild-type (A), pseudo-urs1S (B), mseS (C), or pseudo-urs1S mseS (D) 139 promoter (pMDP89, pMDP119, pMDP174, or pMDP176, respectively) was synchronously sporulated, and -galactosidase levels (expressed as units per milligram of total protein) were determined at 2-h intervals. The experiment was performed at least twice for each promoter with similar results.

View larger version (61K): [in this window] [in a new window]   FIG. 5.   Expression of SMK1 mRNA by mseS and urs1S promoter mutants during sporulation. Diploid yeast containing a single integrated copy of wild-type SMK1 coding information under the control of the indicated mutant promoter was transferred to sporulation medium, and total RNA was prepared from cells harvested at the indicated times and assayed by Northern blot hybridization. The indicated smk1 promoter mutants were generated using the following integrating plasmids, from left to right: wild-type (pMDP199), pseudo-urs1S (pMDP183), mseS (pMDP185), and pseudo-urs1S mseS (pMDP187) (see Table 2). The same filter was probed with sequences specific for the SPO12 middle gene, the HOP1 early gene, and the pC4/2 constitutive expression control.

View larger version (15K): [in this window] [in a new window]   FIG. 6.   Reconstitution of the HOP1 promoter with MSES and URS1S. Yeast cells of strain LNY273 transformed with HOP1 promoter-lacZ plasmids containing the indicated cis-regulatory elements were synchronously sporulated, and -galactosidase levels were measured. pAV124 contains a 5-bp substitution that destroys the naturally occurring URS1H and introduces an XhoI site (open circles); pCC51 contains URS1H (closed circles); pJX42 contains pseudo-URSS (open triangles); and pJX43 contains MSES (closed squares) in the XhoI site of pAV124. The control is pMDP89 (open squares), which is the SMK1 promoter-lacZ plasmid. Data are averages from three determinations.

SMK1 function during spore morphogenesis is not contingent on its expression as a sporulation-specific middle gene. To assess the functional significance of the tight transcriptional regulation of the SMK1 MAP kinase, we examined the phenotypes of the collection of integrated SMK1 promoter mutants that directed the expression of wild-type Smk1p. smk1 promoter mutants were sporulated for 36 h, and the end stage products were assayed for the completion of meiosis by DAPI staining, for spore wall morphogenesis by phase-contrast microscopy, and for acquisition of resistance to glusulase, heat shock, and ether (Table 3). The 118A promoter mutant that expressed SMK1 mRNA as a middle gene, but to only 17% of the level seen in the wild-type promoter control (Fig. 3B), formed spores with near-wild-type efficiency, as judged by phase-contrast microscopy. Based on these results, and taking into account modest differences between the expression of SMK1 from its normal chromosomal locus and from the ura3::SMK1 locus (data not shown) and the fact that the smk1 promoter alleles in these experiments were present in only a single copy, we estimate that spore wall formation can occur when the level of SMK1 mRNA is reduced to 10% of the level seen in wild-type SMK1 homozygotes. However, while spore wall formation did occur in these backgrounds, the resistance phenotypes were not like the wild-type phenotypes. In other experiments, we have shown that SMK1 is required to complete multiple steps in spore morphogenesis (20, 41, 42). In addition, we have isolated a series of smk1 missense hypomorphic alleles that fail to complete distinct steps in sporulation (41). Some of these mutants are defective only in their resistance to heat shock, some are defective in heat shock and ether resistance, and others are defective in all of the resistance phenotypes tested. The 118A SMK1 promoter mutant was resistant to glusulase and ether treatments. However, this mutant is over 10³-fold more sensitive to heat shock than the control strain (Table 3). The 117A mutant strain, which expresses levels of SMK1 mRNA only slightly higher than those of the 118A mutant (Fig. 3B), is indistinguishable from the wild type for all resistance phenotypes (data not shown). These results demonstrate that it is possible to reduce the magnitude of SMK1 expression by roughly 90% and still generate wild-type spores, but when expression is reduced below this level, there are threshold requirements for the acquisition of different resistance phenotypes. These data suggest that acquisition of heat shock resistance requires higher levels of SMK1 expression than either spore morphogenesis or the acquisition of glusulase or ether resistance.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This study demonstrates that the transcriptional regulation of SMK1 as a middle sporulation-specific gene requires two cis-acting DNA elements, UAS^S and MSE^S. UAS^S functions to activate expression of SMK1. MSE^S plays a dual role in regulating SMK1 expression. It is required for activation of SMK1 as a middle sporulation gene and also for repression during vegetative growth and early sporulation. Promoter mutants that express SMK1 during vegetative growth or early in meiosis show no overt phenotypes and are still able to complete spore wall morphogenesis as long as sufficient levels of SMK1 mRNA are present during the middle sporulation period. Mutants that reduce the level of SMK1 expression during middle sporulation to different extents show distinct sporulation defects. The implications of these results for transcriptional control mechanisms and the regulation of development by MAP kinases are discussed below.

Abf1 and transcriptional activation by UAS^S. Our results demonstrate that UAS^S specifically interacts with the Abf1p transcription factor in vitro. It has previously been shown that an Abf1p-binding element activates expression of the early sporulation-specific HOP1 and middle sporulation-specific SPR3 promoters. Furthermore, many other sporulation-specific promoters contain Abf1p-binding sites (16, 27). We have shown that the Abf1p-binding sites from the early HOP1 and middle SMK1 promoters are functionally interchangeable. In addition, the early HOP1 promoter can be converted to a middle gene when its URS1 is replaced with MSE^S. These results indicate that Abf1p-binding sites have a similar role in early and middle sporulation promoters and that they function to regulate the magnitude of transcription and not the timing.

MSE-dependent transcriptional activation and repression. The MSE has previously been shown by others to be capable of activating the transcription of heterologous promoters during middle sporulation (7, 28). Our results show that MSE^S, in addition to activating transcription during middle sporulation, can also repress transcription in vegetative cells. Hepworth et al. previously tested whether the MSE found in the SPS4 middle sporulation promoter could repress vegetative transcription (17). Based on reconstitution experiments using the CYC1 promoter, it was concluded that although the SPS4 MSE specifically binds to a factor in vegetative cells, this element is not a direct repressor of transcription. Our results show that MSE^S can repress the vegetative expression of the CYC1 as well as the SMK1 and HOP1 promoters. Taken together, these data suggest that there are two different classes of MSE. The first class (the SPS4 MSE being the founding member) activates expression during middle sporulation but does not repress vegetative expression. The second class (the SMK1 MSE^S characterized here) activates middle sporulation-specific expression and also represses promoter activity in vegetative cells. Recently, Ndt80p has been shown to activate transcription of middle sporulation-specific genes in an MSE-dependent fashion and to specifically interact with MSE DNA in vitro (7). NDT80 is itself expressed at high levels as a middle sporulation-specific gene. As expected based on its expression pattern, an ndt80- mutation does not derepress SMK1 expression in vegetative cells. These results show that MSE^S repression does not require Ndt80p and indicate that there is a distinct factor expressed in vegetative cells with an overlapping DNA-binding specificity.

Context-dependent URS1 effects on transcription. The presence of an early regulatory URS1 consensus site in the middle SMK1 promoter initially presented a paradox. URS1 has been shown to only transiently derepress expression during early sporulation, and we have shown that MSE^S is required for transient derepression during middle sporulation. If both of these elements were fully functional in a single promoter, it would be inactive throughout the entire sporulation program. This paradox is resolved by the observation that pseudo-URS1^S only weakly represses the SMK1 promoter. This element has a stronger repression activity in the context of the early HOP1 promoter (Fig. 6), suggesting either that the HOP1 promoter contains sequences that are required for full pseudo-URS1^S repression or that the SMK1 promoter contains sequences that antagonize the repression activity. In this respect, we have shown that pseudo-URS1^S only weakly represses vegetative expression of a heterologous SMK1 promoter that contains the HOP1 Abf1p-binding site (data not shown). In contrast, the MSE is fully able to repress this promoter. Thus, the inability of pseudo-URS1^S to function efficiently as a vegetative repressor of SMK1 is not dependent on the Abf1p element.

Functional significance of SMK1 transcriptional regulation. Our previous results demonstrate that SMK1 is required for coordination of spore morphogenesis and that regulated increases in SMK1 activity during spore development play a role in sequentially activating distinct steps required for spore morphogenesis (20, 41, 42). The tightly regulated increase in SMK1 transcription that occurs during spore morphogenesis raised the question of whether SMK1 transcriptional regulation is required to coordinate the morphogenetic program. Our results show that the constitutive expression of SMK1 does not affect vegetative growth, nor does it interfere with meiotic development or spore morphogenesis. In addition, promoter mutations which advance the timing of SMK1 expression to early sporulation make spores which are indistinguishable from wild-type spores. Thus, the developmental function of SMK1 is not contingent on its expression as a sporulation-specific middle gene. These results imply that mechanisms in addition to transcription regulate SMK1 activity during sporulation.