Protein Phosphatase Type 1-Interacting Protein Ysw1 Is Involved in Proper Septin Organization and Prospore Membrane Formation during Sporulation

Strains and growth media.Unless otherwise noted, standard media and genetic techniques were used (1, 20). The strains used in the present study are listed in Table 1. All strains are in the fast-sporulating SK-1 background except for AH109 (Clontech, Mountain View, CA) used for two-hybrid analysis. The primers used in the present study are listed in Table 2. To construct the YSW1-GFP haploid, NY19, HT21, and HT22 were used as primers to amplify the green fluorescent protein (GFP)-tagging cassette in pFA6a-yEGFP-HIS3MX6 (38), and the product was used to transform strain AN117-4B (36). The YSW1-HA haploid, NY21, was created in the same way using pFA6a-HA-His3MX6 (26) as a template. The SPR28-RFP diploid strain, TC522, was constructed by transformation of AN117-4B and AN117-16D (36) by a red fluorescent protein (RFP)-tagging cassette, amplified with oligonucleotides HT322 and HT323 from pFA6a-mRFP5-His3MX6 (40), followed by mating of resulting haploids. NY703 was constructed by PCR-mediated knockout of SPR28 in AN117-4B and AN117-16D using pFA6a-kanMX6 (26) as a template for PCR and oligonucleotides HT39 and HT40, and mating of the resulting haploids. Deletion of YSW1 in AN117-4B and AN117-16D was performed in the same way, using pFA6a-His3MX6 (26) as a template for PCR and oligonucleotides HT319 and HT22 to create TC7 and TC8. These strains were mated to create TC504. TC7 and TC8 were transformed with pRS304 (43) digested with Eco81I, pRS304-YSW1 digested with HindIII, and pRS304-YSW1*** digested with HindIII, followed by mating of the resulting haploids to generate TC551, TC552, and TC553, respectively. All deletions and fusions were confirmed by genomic PCR.

Plasmids.The plasmids used in the present study are listed in Table 3. To construct pRS306-GIP1, the GIP1 gene was amplified from genomic DNA of AN120 (36) with MAC1 and HT306 as primers and cloned into Asp718 and SacII sites of pRS306 (43). pRS306ΔPst was constructed by PstI digestion of pRS306, followed by self-ligation to delete part of URA3 coding sequence. To recover and sequence the gip1-7 allele, PstI digests of genomic DNA were first prepared from the gip1-7 mutant, MIY101. These digests were self-ligated and used to transform Escherichia coli DH5α to obtain an ampicillin-resistant clone, pRS306-gip1-7. pRS424-YSW1 and pRS304-YSW1 were constructed by cloning the BamHI-Asp718 fragment from one of the suppressor candidate clones into pRS424 (7) and pRS304. YBR147w was amplified by PCR using the same clone as a template and MAC6 and MAC7 as primers and cloned into pRS424 to generate pRS424-YBR147w. The YSW1-GFP fusion was amplified by PCR with NY19 genomic DNA as a template and HT32 and HT66 as primers. The fragment was cloned into XhoI and SmaI site of pRS314 (43) and pRS424 to create pRS314-YSW1-GFP and pRS424-YSW1-GFP, respectively. pRS304-YSW1*** was generated by using a site-directed mutagenesis kit (Stratagene, La Jolla, CA) and the primers YSO177 and YSO178. A BamHI-Asp718 fragment from pRS304-YSW1*** was cloned into pRS424 to create pRS424-YSW1***. The 3′ region of YSW1-HA was amplified by PCR using HT259 and HT66K as primers and NY21 genomic DNA as a template. The amplified DNA was digested with ApaI and KpnI, and ligated to pRS424-YSW1 and pRS424-YSW1*** digested with the same enzymes to create pRS424-YSW1-HA and pRS424-YSW1***-HA.

The constructs used for two-hybrid analysis were made as follows. YSW1 residues were amplified from genomic DNA of AN120 for wild-type YSW1, or pRS304-YSW1*** for ysw1***, by using following pairs of oligonucleotides; MAC15 and MAC16 (YSW1_1-609 or YSW1***_1-609), MAC15 and YSO190 (YSW1_1-200), YSO155 and MAC21 (YSW1_201-400 or YSW1***_201-400), YSO191 and MAC16 (YSW1_401-609), MAC19 and MAC21 (YSW1_298-400), YSO155 and YSO171 (YSW1_201-350), YSO155 and MAC22 (YSW1_201-297), MAC18 and MAC16 (YSW1_474-609), and YSO191 and MAC20 (YSW1_401-473). All of the PCR products were digested with BamHI and XhoI and cloned into BamHI-XhoI-digested pGAD-T7 (Clontech). To generate pGBD-GIP1, the GIP1 open reading frame was amplified from genomic DNA of AN120 with the primers YS319 and HT315 and cloned into XhoI site of pGBK-T7 (Clontech), which was digested with NcoI, filled in, and self-ligated to place the fusion in-frame.

Screening for temperature-sensitive sporulation-defective mutants.A GIP1 mutation library was constructed as follows. Error-prone PCR (6) was performed to amplify a DNA fragment containing GIP1 and a part of URA3 using pRS306-GIP1 as a template and MAC1 and MAC2 as primers. The fragment was digested with Asp718 and PstI and ligated to pRS306ΔPst pretreated with the same enzymes. The resulting GIP1 mutation library was digested with PstI and used to transform NY501. Colonies were patched and subjected to sporulation at 22 and 30°C. Cells were replica plated onto yeast extract-peptone-dextrose (YPD) plates and exposed to ether vapor for 20 min by inversion over an ether-soaked paper filter. After incubation at 30°C for 1 day, colonies that sporulated on plates incubated at 22°C, but not those incubated at 30°C, were selected as candidates. One of them, which showed no survival when sporulated at 30°C, was named gip1-7.

Multicopy suppressor screening.The gip1-7 cells were transformed with a DNA library constructed in the pTV3 vector (20) with genomic DNA from NY501 (46). Colonies on each plate were collected in bulk by scraping of the plate, sporulated at 34°C, and subjected to ethanol treatment. For this treatment, ∼10⁶ cells from each plate were incubated in 540 μl of 28% ethanol for 40 min, and plated on synthetic dextrose minimal (SD) plates. Colonies formed on each plate were collected, and plasmids were isolated. These plasmids were used to transform the gip1-7 cells and subjected to another round of sporulation and ethanol treatment. Plasmids were isolated from each of the surviving colonies, and suppression of gip1-7 was confirmed by retransformation and sporulation, followed by observation under microscope.

Sporulation assays.Cells were sporulated in liquid medium as described previously (34). Briefly, strains were grown at 30°C overnight in YPD or in SD medium when required. Cells were then precultured at 30°C overnight in yeast extract-peptone-acetate (YPA) medium. Cells were collected and resuspended in sporulation medium (2% potassium acetate) at an optical density of 1.5 at 600 nm, and these cultures were incubated at 30°C. For spore number analysis, more than 200 cells were counted.

Two-hybrid analysis.Two-hybrid analysis was performed as described by the manufacturers (Clontech). AH109 was transformed with pGBD-GIP1 and pGAD-YSW1 fusions. Transformants were grown overnight on SD medium without tryptophan and leucine, and 10-fold dilutions were spotted onto SD plates without tryptophan, leucine, adenine, and histidine.

Immunoblotting.For the Western blot analysis of Ysw1-HA, cells were induced for sporulation for 7.5 h, and total protein was prepared by using glass beads as described previously (1). Proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, followed by Western blotting with anti-HA antibody (12CA5). Bands were visualized by horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G antibody (Biosource International, Camarillo, CA).

Microscopy.Differential interference contrast (DIC) images were obtained by using a BX70 microscope and DP Controller soft ware (Olympus, Tokyo, Japan). Fluorescence and immunofluorescence microscopy was performed as described previously (46) using a BX70 or a Zeiss Axioplan2 microscope (Carl Zeiss, Thornwood, NY) with a Zeiss mRM Axiocam and processed using Zeiss Axiovision 4.7 software. Cells were fixed with 3.7% formaldehyde when required. Chromatin was stained with DAPI (4′,6′-diamidino-2-phenylindole). Time-lapse imaging was done as follows. Sporulation medium containing 1.5% Agarose-S was dropped on the glass surface of a glass-bottom dish. Solidified medium was removed from the dish, and cells were spotted onto a flat surface of the medium and put again on a dish to sandwich cells between the glass and the medium. Images were captured on a Zeiss Axiovert 100 microscope at 2-min intervals using IPLab 3.6.5a software (Scanalytics, Rockville, MD). Temperature was held at 28°C during image collection. Deconvolution was performed using an EPR system (Scanalytics) and 3D stacks using IPLab 3.6.5a.

Isolation of a temperature-sensitive gip1 mutant.Gip1 is a sporulation-specific targeting subunit of Glc7, the yeast PP1 catalytic subunit, and is required for sporulation (46). However, targets and cofactors of this Gip1-Glc7 phosphatase complex are not known. To isolate genes that have a functional relationship to GIP1, we took a genetic approach. Temperature-sensitive sporulation-defective mutants of gip1 were isolated by introduction of randomly mutagenized GIP1 gene into the gip1 deletion mutant, followed by screening of transformants for those that formed ether-resistant spores at 22°C and not at 30°C. Of 1,000 colonies tested, four colonies showed temperature sensitivity. Among those, gip1-7 showed a complete sporulation defect at 30°C (Fig. 1A and B), which was ideal for suppressor screening. The gip1-7 cells subjected to sporulation at 22°C sporulated well, although the asci formed are predominantly monads and dyads (Fig. 1A and B). In contrast, at 30°C, the gip1-7 cells were indistinguishable to those in gip1Δ cells (Fig. 1A). In addition, spores formed at 22°C were aberrant in their shape; oval or football-like spores were observed (Fig. 1A). DNA sequence analysis of gip1-7 revealed that it has three mutations that affect the amino acid sequence of the encoded protein (K399R, D481G, and L500P). The interaction of gip1-7 with GLC7 was assessed by two-hybrid analysis and was comparable to that of wild-type GIP1 (data not shown).

• Open in new tab
• Download powerpoint

FIG. 1.

Temperature-sensitive sporulation defect of the gip1-7 mutant and partial suppression of it by YSW1. (A) MIY101 (gip1-7) was sporulated at 22 and 30°C and analyzed by DIC microscopy (lower panels). As controls, AN120 (wild-type) and NY501 (gip1Δ) sporulated at 30°C are shown (upper panels). (B) The number of spores in asci of MIY101 (gip1-7) and AN120 (wild-type) sporulated at 22 and 30°C were counted. (C) MIY101 (gip1-7) harboring pRS424-YSW1 (lower panel), or pRS424 (upper panel) as a control, was sporulated at 30°C and analyzed by DIC microscopy. (D) MIY101 (gip1-7) carrying pRS424, pRS424-YBR147w, pRS424-YSW1, and pRS424-YSW1-GFP, respectively, were sporulated. The distribution of ascal types is shown. Bars in panels A and C, 5 μm.

Isolation of YSW1 as a multicopy suppressor of a temperature-sensitive gip1 mutant.A screen was performed to isolate multicopy suppressor genes of the gip1-7 temperature-sensitive sporulation defect. The gip1-7 strain was transformed with a genomic library constructed using gip1Δ genomic DNA, sporulated at 34°C, and clones that formed ethanol-resistant spores were isolated. A total of 2 × 10⁴ colonies were subjected to screening and four plasmids were isolated as multicopy suppressors. All four plasmids contained a genomic region, which encompasses YBR147w and YSW1 (YBR148w). Therefore, YBR147w and YSW1 were cloned separately into multicopy vectors, introduced into the gip1-7 mutant, and sporulation was examined at restrictive temperature. YSW1, not YBR147w, restored sporulation, although sporulation efficiency was low and the asci formed were mostly monads and dyads (Fig. 1C and D). Expression of Ysw1 from low copy vector suppressed gip1-7 at a level comparative to that of Ysw1 expression from multicopy plasmid (data not shown). Suppression was not observed when YSW1 was introduced into gip1Δ cells (data not shown). These results indicate that overexpression of YSW1 partially suppresses the temperature-sensitive sporulation-defective phenotype of the gip1-7 mutant and that YSW1 has a genetic interaction with GIP1.

A conserved region of Ysw1 is required for interaction with Gip1 and suppression of gip1-7. YSW1 encodes a predicted protein of 609 amino acids (aa) that contains two coiled-coil domains. It was originally isolated in a large-scale analysis of gene expression and protein localization as a protein induced during meiosis and sporulation (4). A truncated form of Ysw1 fused to lacZ localized to the prospore membrane. The genetic interaction between YSW1 and GIP1 led us to examine whether Ysw1 can physically interact with Gip1. A two-hybrid analysis was performed, and Ysw1 was found to interact with Gip1 (Fig. 2A). To identify the region of Ysw1 involved in the interaction with Gip1, we constructed deletion series of Ysw1 and tested for interaction with full-length Gip1. Surprisingly, we found that two separate regions of Ysw1 interact with Gip1 (Fig. 2A). The first region includes aa 201 to 297, and the second region spans aa 401 to 473, which encompasses the second coiled-coil domain of Ysw1.

• Open in new tab
• Download powerpoint

FIG. 2.

Ysw1 interacts with Gip1 and a conserved region of Ysw1 is important for this interaction. (A) Ysw1 residues and Ysw1*** tested for two-hybrid analysis with GBD-Gip1 are indicated at the left. The various pGAD-YSW1 plasmids were introduced into AH109 with pGBD-GIP1 and strains were assayed for growth by plating 10-fold dilutions on SD His⁻, Ade⁻, Trp⁻, Leu⁻ plates. (B) Multiple alignment of conserved region in Ysw1 is shown. (C) Distribution of ascal types in MIY101 (gip1-7) harboring pRS424 or pRS424-YSW1-HA or pRS424-YSW1***-HA, respectively, is shown. (D) A Western blot analysis of strains in panel C sporulated for 7.5 h was performed with anti-HA antibody.

Comparative genomic studies have revealed that a whole-genome duplication occurred in the evolution of hemiascomycetes, S. cerevisiae being one of the postduplication species (52). Duplicated genes are referred to as ohnologs. It is reported that YSW1 is the ohnolog of SPO21, which encodes a meiosis-specific component of the SPB, although they have only 13% identity and do not hit each other in the basic local alignment tool for protein sequences (BLASTP) (51). Based on this report, we compared the YSW1 sequence with the SPO21/YSW1 genes of preduplication yeast species and found a small region (aa 230 to 240) of Ysw1, amino terminal to the first coiled-coil domain, that is conserved at the analogous position in the Spo21/Ysw1 orthologs of Kluyveromyces lactis, Ashbya gossypii, and Kluyveromyces waltii (Fig. 2B). This region is not conserved in S. cerevisiae Spo21. The motif resides in the first Gip1-interacting region of Ysw1 defined in our deletion studies. Therefore, to examine whether the motif was important for the interaction, we introduced the three point mutations Y233A, F235A, and F237A into conserved residues of Ysw1 (hereafter referred to as Ysw1***; Fig. 2A). Neither the full-length Ysw1*** nor aa 201 to 400 of Ysw1*** alone interacted with Gip1 in two-hybrid assay, indicating the importance of this conserved motif for binding of Ysw1 to Gip1. To examine whether Ysw1*** is functional or not, we expressed Ysw1***-HA in the gip1-7 mutant. The gip1-7 cells expressing Ysw1-HA sporulated to ca. 20% at a nonpermissive temperature (Fig. 2C). However, at 30°C the gip1-7 cells expressing Ysw1***-HA were completely defective in sporulation, as was the case with gip1-7 cells harboring vector alone (Fig. 2C). A Western blot with anti-HA antibody confirmed that Ysw1***-HA is expressed at a level similar to that of Ysw1-HA during sporulation (Fig. 2D). These results strongly indicate that interaction of Ysw1 with Gip1 through this conserved region is required for suppression of gip1-7 by YSW1.

Ysw1 colocalizes with septins and Gip1 along the extending prospore membrane.To investigate the localization of full-length Ysw1 protein during sporulation, we tagged the Ysw1 protein with GFP at its C terminus and examined the protein by fluorescence microscopy. Ysw1-GFP was at least partially functional, because overexpression of YSW1-GFP allowed the gip1-7 strain to sporulate at only a slightly lower level compared to the same strain overexpressing YSW1 (Fig. 1D). Ysw1-GFP was seen as four small rings or four pairs of short bars around the nucleus in early meiosis II, and four pairs of long bars along prospore membrane in late meiosis II (Fig. 3A). The Ysw1-GFP signal disappeared in postmeiotic cells, and no specific localization pattern was observed (data not shown). The localization pattern of Ysw1-GFP during meiosis II was similar to that reported for septins and Gip1 (46), although the absence of Ysw1 in postmeiotic cells was quite different from the spore periphery pattern seen for septins and Gip1 at this later stage (15, 46).

• Open in new tab
• Download powerpoint

FIG. 3.

Ysw1 colocalizes with septins and Gip1 during sporulation. (A) AN120 (wild-type) harboring pRS314-YSW1-GFP was sporulated for 7.5 h and analyzed by fluorescence microscopy. Nuclei were visualized with DAPI. (B) TC522 (SPR28-RFP) harboring pRS314-YSW1-GFP was sporulated and analyzed. (C) AN120 (wild-type) carrying pRS314-YSW1-GFP and pSB6 (HA-GIP1) was sporulated and analyzed by immunofluorescence with anti-HA antibody. (D) pRS314-YSW1-GFP was introduced into AN120 (wild-type), NY528 (spr3Δ), and NY703 (spr28Δ), and the resulting strains were sporulated and analyzed. Scale bars, 5 μm.

To confirm that Ysw1 colocalizes with septins and Gip1 in meiosis II, a YSW1-GFP plasmid was introduced into strains expressing Spr28-RFP and HA-Gip1, respectively. Colocalization of Ysw1-GFP with Spr28-RFP and HA-Gip1 was observed in meiosis II cells (Fig. 3B and C). These results suggest that Ysw1, Gip1, and Glc7 colocalize on septin structure along the extending prospore membranes.

Given that septins form filaments and function as a scaffold for many proteins at the bud neck in vegetatively growing cells, we reasoned that the Ysw1 localization pattern would be septin dependent. Therefore, we analyzed Ysw1-GFP localization in septin mutants. Deletion of either of the sporulation-specific septins SPR3 or SPR28 disrupts the organization of the remaining septin proteins (41). Loss of SPR28 causes the remaining septins to localize uniformly around the prospore membrane rather than in bars, while the loss of SPR3 leads to a failure of the remaining septins to associate with the prospore membrane at all. In the case of Ysw1-GFP, deletion of either SPR3 or SPR28 caused the same phenotype, an even distribution around the prospore membrane rather than a bar-like pattern (Fig. 3D). These observations indicate that localization of Ysw1 into bars, although not its association with the prospore membrane, is septin dependent.

Septins structures are aberrant in the ysw1Δ mutant.To examine the role of Ysw1 during sporulation, a ysw1Δ mutant was constructed and analyzed for sporulation. As reported in genomewide analysis (28), the ysw1Δ mutant asci had a reduced number of spores, and many dyads and triads were observed (Fig. 4A and B) . Using DAPI staining, the meiotic progression of the ysw1Δ mutant was monitored and found to be wild type (data not shown). These observations suggest that Ysw1 functions in spore formation rather than in meiosis. Expression of Ysw1*** did not rescue the sporulation defect of the ysw1Δ mutant (Fig. 4B), suggesting that interaction between Ysw1 and Gip1 is required for the function of Ysw1.

• Open in new tab
• Download powerpoint

FIG. 4.

Sporulation is aberrant in the ysw1Δ mutant. (A) TC504 (ysw1Δ) was sporulated and analyzed by DIC microscopy. (B) AN120 (wild-type), TC551 (ysw1Δ pRS304), TC552 (ysw1Δ, pRS304-YSW1), and TC553 (ysw1Δ, pRS304-ysw1***) were sporulated, and the distribution of the spore number is represented. (C) AN120 (wild-type) and TC504 (ysw1Δ) harboring pSB7 (SPR28-GFP) were analyzed by fluorescence microscopy. (D) AN120 (wild-type) and TC504 (ysw1Δ) harboring pSB5 (HA-GIP1) were analyzed by immunofluorescence with anti-HA antibody. Arrowheads in panels C and D indicate cells showing aberrant pattern. Scale bars (A, C, and D), 5 μm.

The involvement of Ysw1 in spore formation, together with its relationship to septins and Gip1, prompted us to examine the localization of septins and Gip1 in the ysw1Δ mutant. Spr28-GFP and HA-Gip1 were expressed in the mutant, and their localization during sporulation was determined by fluorescence microscopy. Septin structures were formed during sporulation in the ysw1Δ mutant (Fig. 4C). Both bars in meiosis II cells and distribution of proteins around the spore periphery in postmeiotic cells were observed. However, the bars in the ysw1Δ cells were abnormal and appeared excessively bent or twisted. A similar localization defect of Gip1 was observed in ysw1Δ cells (Fig. 4D), indicating that HA-Gip1 can localize to the septin structure in the ysw1Δ cells. These results suggest that Ysw1 is required for proper septin organization.

Ysw1 is required for proper prospore membrane formation.The bar-like structures containing Gip1, Ysw1, and septins form along the prospore membrane (15, 46). Therefore, we examined whether prospore membranes are properly formed in the ysw1Δ mutant. A fragment of the Spo20 protein fused to GFP (32) was used to visualize prospore membranes. In wild-type cells, as the cells go through meiosis II, prospore membranes progressively display horseshoe-like, tubular, and round morphologies (12) (Fig. 5A). In addition, all four prospore membranes forming within a wild-type cell appear to grow at similar rates. In the ysw1Δ cells, prospore membranes were also formed but appeared somewhat uncoordinated in their size (Fig. 5A). Strikingly, counting of prospore membranes in early- and post-meiosis II cells revealed that number of mature round prospore membrane per ascus was reduced, about half of the asci contained only two prospore membranes (Fig. 5B). This correlated well with distribution of spore number in mature asci. Very small prospore membranes and/or remnants of them were frequently observed in the ysw1Δ asci. These results suggest that the ysw1Δ cells are defective in prospore membrane growth.

• Open in new tab
• Download powerpoint

FIG. 5.

Prospore membrane growth is defective in ysw1Δ mutant. (A) AN120 (wild-type) and TC504 (ysw1Δ) were transformed with 424-G20 (GFP-SPO20^51-91), and prospore membranes were visualized during sporulation. (B) The numbers of small prospore membranes per cell (left) and mature prospore membranes per cell (middle) or spores per ascus (right) were counted. These categories correspond to cells in early meiosis II, late meiosis II, and postmeiosis, respectively. (C) AN120 (wild-type) and TC504 (ysw1Δ) carrying 424-G20 (GFP-SPO20^51-91) were sporulated and analyzed by time-lapse fluorescence microscopy. Each image is a projection through a deconvolved image stack. The numbers indicate the time elapsed, in minutes. Scale bars (A and C), 5 μm.

To further analyze the defect of ysw1Δ cells in prospore membrane formation, we performed time-lapse imaging. In wild-type cells, the extension of the four prospore membranes appeared coordinated (Fig. 5C and see Movie S1 in the supplemental material). In contrast, prospore membrane extension was slower and looked less coordinated in ysw1Δ cells (Fig. 5C and see Movies S2 to S5 in the supplemental material). Fewer than four prospore membranes extended in many asci, with the remaining prospore membranes stopped at the small round phase or earlier. These results indicate that Ysw1 is required for proper prospore membrane growth.