Mitogen‐activated protein (MAP) kinase pathways are evolutionarily conserved kinase cascades that are required for the response of eukaryotic cells to a wide variety of environmental stimuli. MAP kinase pathways are also required for the execution of developmental and differentiative programs in a variety of cell and tissue types. SMK1 encodes a developmentally regulated MAP kinase in yeast that is required for spore wall morphogenesis. Cyclin‐dependent kinase‐activating kinases (CAKs) phosphorylate a conserved threonine residue in the activating loop of cyclin‐dependent kinases. CAK1 encodes the major CAK activity in yeast and is required for cell cycle progression. The work presented here demonstrates that CAK1 functions positively in the spore wall morphogenesis pathway. First, CAK1 has been isolated as a dosage suppressor of a conditional smk1 mutant that is defective for spore wall morphogenesis. Second, CAK1 mRNA accumulates during spore development contemporaneously with SMK1 mRNA. Third, cak1 mutant strains have been isolated that are able to complete meiosis I and II but are specifically defective in assembly of the spore wall. These results show that cell cycle progression and morphogenetic pathways can be regulated by a single gene product and suggest mechanisms for coordinating these processes during development.
Mitogen‐activated protein (MAP) kinases comprise a family of signal transducing enzymes that phosphorylate regulatory molecules in response to a broad array of stimuli (Blenis, 1993; Blumer and Johnson, 1994; Marshall, 1994; Waskiewicz and Cooper, 1995). MAP kinases are activated by dual specificity MAP kinase kinases (MEKs) that phosphorylate both a tyrosine and threonine residue in the active site lip of the enzyme. MEKs are activated via serine and threonine phosphorylation by upstream MEK kinases. This sequential activation of MAP kinase by MEK through MEK kinase constitutes what has been defined as the MAP kinase module (Neiman et al., 1993). These modules have been evolutionarily conserved throughout eukaryotes in the biochemical mechanism and amino acid sequence of the individual kinase members. Activated MAP kinases phosphorylate regulatory molecules, such as transcription factors and components of the cell cycle machinery, which effect cellular responses appropriate to the input signal. These cellular responses can include adaptation to environmental stress, changes in proliferative activity, and induction and execution of developmental programs. The roles of MAP kinase pathways in development have been studied in a wide range of cell and tissue types and a diverse set of organisms. MAP kinase pathways have been shown to be required for neural differentiation in mammalian PC12 cells (Marshall, 1994), mesoderm induction in Xenopus (Gotoh et al., 1995; LaBonne et al., 1995; Umbhauer et al., 1995), eye development in Drosophila (Perrimon, 1994), vulval development in Caenorhabditis elegans (Eisenmann and Kim, 1994) and cell fate decisions in Dictyostelium (Firtel, 1995).
In the yeast Saccharomyces cerevisiae, there are a number of MAP kinase pathways that elicit appropriate responses to particular stimuli (Ammerer, 1994; Herskowitz, 1995; Kron and Gow, 1995; Levin and Errede, 1995). The mating pheromone response pathway is the most thoroughly understood and represents a paradigm for the molecular mechanisms that mediate MAP kinase signaling. Additional MAP kinase pathways in yeast include the osmoregulatory pathway, the protein kinase C/cell wall integrity pathway, the pseudohyphal differentiation pathway and the SMK1 spore wall morphogenesis pathway. SMK1 (sporulation MAP kinase 1) encodes a developmentally regulated MAP kinase in yeast that is essential for the completion of sporulation (Krisak et al., 1994).
Sporulation in yeast provides a model system in which to study developmental processes. Similarly to differentiation programs in higher eukaryotic cells, induction of sporulation is controlled by a combination of cell type and environmental signals (Esposito and Klapholz, 1981). Yeast spore development is restricted to the a/α diploid and occurs in response to starvation for nitrogen and a fermentable carbon source. Upon initiation, cells will exit the mitotic cell cycle and enter meiosis. The four main phases or landmark events of sporulation are: (i) meiotic prophase during which DNA synthesis, meiotic recombination and formation of synaptonemal complexes occur; (ii) meiosis I or the reductional division where homologous chromosomes segregate from each other; (iii) meiosis II or the equational division when sister chromatids segregate; and (iv) spore wall morphogenesis and spore maturation. The end‐product of development is an ascospore which contains four haploid spores, two a and two α, which are distinct cell types from the starting diploid cell. Spore development is tightly regulated at the level of transcription (Mitchell, 1994). Upon initiation, a cascade of sporulation‐specific gene expression ensues with genes being classified as early genes expressed at the onset of meiotic prophase, middle genes expressed during the meiotic divisions and late genes expressed during spore wall morphogenesis. Thus, sporulation in yeast is an example of a developmental process that is intimately coupled to a transcriptional program.
SMK1 is a tightly regulated middle sporulation gene required for morphogenesis of the spore wall and completion of the sporulation program (Krisak et al., 1994). Spore wall morphogenesis is characterized by growth of a bimembranous prospore wall which nucleates at each meiosis II spindle pole body and eventually encloses each meiotic product. Subsequently, the four‐layered spore wall is assembled from within/around the prospore wall (Byers, 1981; Esposito and Klapholz, 1981). The two innermost layers are made primarily of glucan; the next layer contains chitin and chitosan; and the outermost layer is rich in dityrosine (Briza et al., 1986, 1988, 1990b). These layers are vital to the integrity of the spore as they render it resistant to environmental stresses. smk1 null mutants initiate sporulation and progress through meiosis I and II normally. However, functional and ultrastructural assays show that subsequent developmental events are defective. Electron microscopy of smk1 null homozygous asci reveals a variety of aberrant spore wall assembly patterns. Even within a single smk1 mutant ascospore, varied aberrant patterns are observed. Spore wall layers are either missing, extranumerary or improperly ordered. These data indicate that SMK1 is required for the coordination of spore wall morphogenesis. Additionally, transcription of late but not early or middle sporulation genes is significantly reduced in smk1 null mutants, suggesting that SMK1 is also required for subsequent steps in spore development and maturation. SPS1 encodes a protein kinase homolog which is also required for spore wall assembly (Friesen et al., 1994). SPS1, like SMK1, is expressed as a middle sporulation gene. Also, sps1 null homozygous asci exhibit analogous defects in spore wall morphogenesis and transcription of late sporulation‐specific genes. These data and the homology of Sps1p to the Ste20p/Pak family of MAP kinase cascade activators have led to the proposal that SPS1 is an upstream kinase in the SMK1 spore wall morphogenesis pathway. The SMK1/SPS1 pathway and sporulation affords an excellent opportunity to study how a MAP kinase signaling pathway can coordinate complex processes during development.
Here we describe the isolation of a collection of temperature‐sensitive smk1 alleles. One of these alleles, smk1‐2, has been characterized in detail and is defective for spore wall morphogenesis and subsequent developmental events. A dosage suppressor of the smk1‐2 mutant phenotype has been identified. This dosage suppressor is allelic to CAK1 (cyclin‐dependent kinase‐activating kinase 1) which encodes an essential protein kinase that has been shown by others to activate Cdc28p (Espinoza et al., 1996; Kaldis et al., 1996; Thuret et al., 1996). Activated Cdc28p is required for and regulates mitotic as well as meiotic cell divisions. An allele of cak1 which allows initiation of sporulation and completion of meiosis I and II, but not spore wall morphogenesis, has been isolated. Thus, CAK1 plays a positive regulatory role in and is required for spore wall morphogenesis. These results suggest that CAK1 may play a role in coordinating meiotic cell divisions with spore wall morphogenesis during spore development.
Isolation of smk1 conditional alleles
The spore wall contains high levels of insoluble dityrosine, which fluoresces naturally in the visible spectrum when excited by UV light. The fluorescence of sporulated colonies, hereafter referred to as the fluorescence assay, provides a sensitive method by which spore wall assembly may be monitored (see Materials and methods). When subjected to this assay, sporulated wild‐type diploids fluoresce while smk1‐Δ, sps1‐Δ and smk1‐Δsps1‐Δ homozygous diploids do not. In order to isolate a collection of smk1 conditional alleles, an smk1‐Δ homozygous diploid was transformed with an smk1 mutant plasmid library which had been generated in vitro by hydroxylamine mutagenesis. Transformants were sporulated at the permissive (26°C) and non‐permissive (34°C) temperatures and scored by the fluorescence assay. Twenty smk1 temperature‐sensitive alleles have been identified in this manner, and smk1 mutant diploid strains generated by replacing the chromosomal SMK1 locus via standard homologous recombination techniques. All alleles isolated are recessive with respect to the fluorescence phenotype. Further characterization of one of these alleles, smk1‐2, is described below. Analyses and detailed phenotypic characterization of other smk1 conditional mutants will be described elsewhere.
Characterization of the smk1‐2 allele
The fluorescence phenotype of a sporulated smk1‐2/smk1‐Δ diploid is shown in Figure 1A. At the permissive temperature, the smk1‐2 strain fluorescence is comparable with that of the wild‐type, while at the non‐permissive temperature smk1‐2 fluorescence is comparable with that of the smk1 null mutant. smk1 null mutant spores are known to be hypersensitive to heat shock and exposure to glusulase or ether. Additionally, smk1‐Δ null mutants are deficient in transcription of late sporulation genes, such as SPS100. The resistances of smk1‐2, smk1‐Δ and wild‐type spores formed at 26°C or 34°C to heat shock, glusulase or ether treatments were compared. Additionally, comparisons of SPS100–lacZ reporter gene expression in these sporulated strains were made (see Materials and methods). The smk1‐2 spores exhibited temperature‐senstive defects in all of these phenotypes (Figure 1B). However, the resistance to the environmental stresses was not completely wild‐type at the permissive temperature. This suggests that, despite the ability to assemble birefringent spore walls (see below) and to fluoresce to wild‐type levels, smk1‐2 spores formed at the permissive temperature may be deficient in subtle aspects of spore wall morphogenesis or maturation. Interestingly, smk1‐2 spores formed at the non‐permissive temperature are still less sensitive to heat shock, glusulase or ether treatments than are smk1‐Δ spores. This suggests that certain developmental events that lead to these resistance phenotypes may be executed in the smk1‐2 mutant under these conditions.
Phase‐contrast and fluorescence microscopy of DAPI‐stained cultures sporulated at the permissive temperature revealed that smk1‐2 ascospores are indistinguishable from wild‐type (Figure 2). In both wild‐type and smk1‐2 mutant cultures there were four distinct nuclei, each encapsulated by a refractile spore wall, in >85% of the ascospores. smk1‐2 cultures sporulated at the non‐permissive temperature initiated sporulation and completed meiosis I and II with the same efficiency as wild‐type, as evidenced by the presence of four distinct DAPI‐staining foci. However, in contrast to both the smk1‐2 spores formed at 26°C and the wild‐type spores formed at 34°C, <1% of smk1‐2 spores formed at the non‐permissive temperature were surrounded by birefringent spore walls.
Nucleotide sequence analysis of the smk1‐2 gene revealed that it contains a single missense mutation in codon 169 which results in a serine for proline substitution. This altered residue lies in subdomain VI, which is highly conserved in all protein kinases and is 14 residues amino‐terminal to the catalytic aspartate in the presumed activation loop of Smk1p (Hanks et al., 1988). The analogous amino acid substitution in the Schizosaccharomyces pombe Cdc2, Drosophila MEK and Dictyostelium Erk2 kinases also confers a temperature‐sensitive phenotype (Carr et al., 1989; Hsu and Perrimon, 1994; Gaskins et al., 1996). Western analysis of an epitope‐tagged Smk1‐2 protein indicates that the P169S substitution does not affect protein stability, and thus the mutation must hinder either the ability of Smk1‐2p to be activated by upstream signaling components or its ability to interact with or phosphorylate downstream effectors required for spore development.
Isolation of CAK1 as a dosage suppressor of the smk1‐2 fluorescence defect
A genomic library of yeast DNA contained on a high‐copy (2μ‐based) vector was used to identify sequences able to suppress the smk1‐2 fluorescence defect at the non‐permissive temperature. Three classes of plasmid were recovered. Class I transformants fluoresced to levels indistinguishable from that seen in the wild‐type control strain. Diagnostic restriction enzyme analysis showed that this class of plasmid contains the SMK1 gene. Class II transformants fluoresced to an intermediate level at the non‐permissive temperature, and all class II plasmids share a common set of restriction enzyme fragments. Characterization of the class II suppressor is described in detail below. Class III suppressors, which yielded a relatively weak but reproducible increase in fluorescence, will be described elsewhere.
Plasmid pHCS12‐5 is the class II plasmid that contains the smallest DNA fragment capable of suppressing the smk1‐2 fluorescence defect. Nucleotide sequence analysis of the pHCS12‐5 insert revealed that it contains a 3.4 kb genomic fragment from chromosome VI with two complete putative open reading frames (YFL029c and YFL030w). Fluorescence assays of transformants harboring subcloned derivatives of pHCS12‐5 revealed that the YFL029c open reading frame is necessary and sufficient for dosage suppression of the smk1‐2 fluorescence defect (Figure 3). YFL029c did not increase fluorescence of the smk1‐Δ null mutant or the wild‐type control strain. The YFL029c gene product is a 368 residue protein kinase homolog most similar to the cyclin‐dependent kinase family and 54% similar and 30% identical to Cdc28p. The YFL029c protein contains most of the amino acid residues conserved in practically all serine/threonine protein kinases, with the noted exception of the glycine‐rich nucleotide‐binding fold which is typically located in subdomain I (Hanks et al., 1988; Hanks and Quinn, 1991). During the preparation of this manuscript it was reported that YFL029c encodes CAK1 (CIV1), the major CAK in yeast (Espinoza et al., 1996; Kaldis et al., 1996; Thuret et al., 1996). YFL029c is hereafter referred to as CAK1.
Suppression of other smk1‐2 phenotypes by multicopy CAK1
In order to characterize the dosage suppression phenotype further, smk1‐2 strains harboring the high‐copy control vector, YEp352 or YEp352+CAK1 were assayed for the ability to form birefringent spore walls as viewed by phase‐contrast microscopy. At the non‐permissive temperature, both strains underwent meiosis with >80% efficiency (Table I). Of those cells which completed meiosis, 68% of the asci harboring the CAK1‐containing plasmid formed birefringent spore walls, in contrast to <0.5% for the negative control plasmid. The smk1‐2 diploid harboring this dosage suppressor yielded asci indistinguishable from wild‐type at this microscopic level (Figure 4) and at the electron microscopic level (data not shown). Multicopy CAK1 had no discernible effect on spore formation in the smk1 null mutant.
Spores of the smk1‐2 strain carrying the CAK1‐multicopy plasmid or the negative control plasmid formed at either 26 or 34°C were tested for sensitivity to heat shock, glusulase or ether treatment. In all cases, CAK1 dosage suppressed the resistance defect of the smk1‐2 mutant (Figure 5). A large contribution to heat shock resistance occurs relatively early in sporulation and does not require assembly of the outer spore wall layer (M.Wagner, unpublished results); glusulase resistance presumably reflects assembly of the spore‐specific wall layers, while ether resistance is established later during sporulation in a period termed maturation (Law and Segall, 1988). Thus the overexpression of CAK1 can suppress multiple smk1 defects. CAK1 was also able to dosage suppress mutant phenotypes of other smk1 conditional alleles which contain missense mutations in different subdomains of the Smk1p protein kinase (M.Wagner, unpublished results).
Expression of CAK1 during sporulation
Northern blot hybridization analysis of CAK1 expression showed that it is transcribed in vegetatively growing cells (Figure 6, lane 0). Interestingly, Northern blot hybridization analysis of RNA isolated during a sporulation time‐course revealed that there is a sporulation‐specific peak of CAK1 mRNA accumulation. This peak of expression occurs concomitantly with the peaks of both SPS1 and SMK1 transcript accumulation which are known to occur shortly after meiosis II (Figure 6). Consistent with these observations, the CAK1 promoter region contains two T4C consensus sequences (both have six out of seven matches to the consensus) centered at positions −292 and −128 relative to the initiator ATG. T4C promoter elements have been implicated in positive regulation of sporulation‐specific gene expression (Mitchell, 1994). CAK1 mRNA levels during a sporulation time‐course are not altered in a smk1‐Δ mutant background, indicating that SMK1 is not required for transcriptional regulation of CAK1 during sporulation.
CAK1 protein kinase activity is required for vegetative growth and dosage suppression of smk1‐2
A TRP1‐disrupted null allele of CAK1 was introduced into a diploid by standard gene replacement techniques. CAK1 wild‐type/null heterozygotes were sporulated, tetrads microdissected, and individual spores germinated. In 15 out of 15 tetrads analyzed, 2 to 2 segregation for growth was seen. The growth of wild‐type and null haploids was indistinguishable for the first several generations. However, longer incubation gave normal‐sized colonies for the wild‐type and microcolonies (∼50 cells) which grew no further for the null mutant. Cells in these microcolonies exhibited elongated morphologies, reminiscent of pseudohyphal filamentous growth (Gimeno et al., 1992).
Several mutant alleles of the CAK1 gene were constructed and their vegetative and sporulation phenotypes assessed. Three of these cak1 alleles were made by site‐directed mutagenesis. The aspartate in the DFG peptide motif of protein kinase subdomain VII is absolutely required for enzymatic activity in protein kinases tested and has been proposed to play a crucial role in phosphotransfer (Hanks et al., 1988). The codon for the analogous aspartate residue (D179) in CAK1 was mutated to encode arginine. This substitution has been shown to ablate in vivo and in vitro kinase activity in all protein kinases examined. The cak1‐D179R mutant allele does not support vegetative growth and also does not suppress the sporulation defect of smk1‐2 strains when present in high (2μ) copy. These results indicate that Cak1p protein kinase activity is essential for vegetative growth, and for the suppression of the smk1‐2 sporulation defect. The amino‐terminal region of protein kinases, including subdomains I and II, typically contains a conserved glycine‐rich nucleotide‐binding fold and phosphotransfer lysine (Hanks et al., 1988). This is the region in Cak1p which is most divergent from typical protein kinases. A deletion mutation, which removes the 31 amino‐terminal codons 8–38, was constructed. This cak1‐Δ31 mutant allele did not support mitotic growth or suppress the smk1‐2 sporulation defect. Additionally, the residue predicted to be the phosphotransfer lysine residue by spatial alignment, K31, was mutated to arginine. This conservative substitution is known to cripple kinase activity severely in almost all protein kinases tested. The cak1‐K31R mutant allele does support vegetative growth and does suppress the smk1‐2 sporulation defect, indicating that this lysine residue is not required for the in vivo function of Cak1p. The vegetative and sporulation phenotypes of these three mutant alleles indicate that Cak1p is indeed a kinase, but with some unusual properties.
A number of cak1 temperature‐sensitive mutant alleles were selected from a library of randomly mutagenized CAK1 alleles. The cak1 library in a URA3–centromere‐based plasmid was generated in vitro by hydroxylamine mutagenesis. The library was transformed into a CAK1/cak1::TRP1 heterozygote. Transformants were pooled in liquid culture, sporulated en masse, the ascal sacs digested and the spores separated. The resulting haploid spores were plated on media lacking uracil and tryptophan to select for both the cak1::TRP1 null allele and the plasmid‐borne mutagenized CAK1. Isolates containing temperature‐sensitive alleles of CAK1 were identified as those which grew at the permissive (26°C) but not the non‐permissive (34°C) temperature. At both 26°C and 34°C, all of the cak1 mutant isolates examined exhibited highly elongated cells which remained attached to each other after budding, as viewed by phase‐contrast microscopy (see CEN/cak1‐17 cells in Figure 7). These vegetative phenotypes are consistent with observations of cak1‐22 mutant background described by Kaldis et al. (1996). The position of bud scars revealed by calcofluor staining showed that the cak1 mutant haploids manifest a bipolar budding pattern at both 26°C and 34°C. Actin staining with rhodamine phalloidin showed punctate foci at the cell periphery often concentrated at the end of elongated cells at both temperatures, suggesting the occurrence of polarized growth. DAPI staining showed that some cells within the mutant cak1 population lacked nuclear material, suggesting defects in fidelity of nuclear segregation.
One of the cak1 temperature‐sensitive mutant alleles, cak1‐17, was shown to confer wild‐type vegetative growth morphology at the permissive temperature while retaining a temperature‐sensitive growth defect when expressed on a 2μ‐based plasmid (Figure 7). At the non‐permissive temperature this cak1‐17 genetic background caused cells to arrest as large unbudded cells, similar to what has been seen in another cak1 (civ1‐4) mutant strain (Thuret et al., 1996). That different terminal phenotypes are exhibited when the cak1‐17 allele is expressed in low (centromere) versus high (2μ) copy number suggests that different mitotic requirements for CAK1 may be met by different threshold levels of CAK1 activity.
CAK1 is required for spore wall morphogenesis
cak1::TRP1 homozygous diploids harboring either the 2μ‐CAK1 or 2μ‐cak1‐17 plasmid were sporulated at 26°C or 34°C and scored by the fluorescence assay. The cak1‐17 mutant strain showed a temperature‐sensitive defect for fluorescence at the non‐permissive temperature, indicating that CAK1 is required for completion of sporulation (Figure 8). These same CAK1 and cak1‐17 strains were sporulated in liquid at the permissive or non‐permissive temperature and the morphology of the ascospores examined by phase contrast and fluorescence microscopy (Figure 9). DAPI staining for nuclei indicates that both strains initiate meiosis and complete meiosis I and II with the same efficiency, ∼80%, at both temperatures (Table II). However, at the non‐permissive temperature, only 9% of cak1‐17 ascospores form refractile spore walls as compared with 87% in the CAK1 control. That the cak1‐17 allele causes temperature‐sensitive spore wall defects is corroborated by the observation that while the heat shock resistance of the CAK1 and cak1‐17 spores formed at 26°C and the CAK1 spores formed at the 34°C are all comparable, the cak1‐17 strain sporulated at 34°C is >10‐fold more sensitive to heat shock.
These ascospores were also examined by electron microscopy (Figure 10). The spore walls found in wild‐type asci formed at 26°C and 34°C were indistinguishable and characterized by the two inner electron‐lucent glucan layers (see bracket in Figure 10C), surrounded by the more diffuse spore‐specific chitosan‐containing layer, which is in turn surrounded by the electron‐dense coat (see arrow in Figure 10C). The ultrastructure of the cak1‐17 mutant spore walls formed at 26°C were indistinguishable from those seen in the wild‐type (Figure 10D). In striking contrast, a variety of ultrastructural defects are observed in the cak1‐17 mutant asci formed at the non‐permissive temperature (Figure 10E–I). Of 200 asci scored, 10% appeared similar to wild‐type when superficially examined at low magnification. However, careful examination of these asci whose spore wall assembly patterns most closely resembled wild‐type (Figure 10E and F) revealed multiple interruptions in the outer electron‐dense coat (Figure 10E) and under high magnification the chitosan layer appeared to be diminished in thickness or even missing (Figure 10F). The spore walls in the remaining 90% of the cak1‐17 mutant asci (80% of which completed meiosis II as determined by DAPI staining of the culture) were heterogeneous (Figure 10G–I). Each spore within the mutant asci appeared to have characteristic defects in which layers were missing. In addition, structures that appeared similar to prospore wall structures were seen in 10–20% of the asci. The phenotype of the cak1‐17 asci formed at the non‐permissive temperature thus appears similar to that seen in a smk1 or in an sps1 mutant.
This work demonstrates that CAK1, which encodes the major CAK in yeast, plays a positive role in the spore wall morphogenesis pathway. The evidence in support of this conclusion is 3‐fold. First, CAK1 was isolated in a dosage suppression screen of the fluorescence defect of a smk1‐2 conditional strain. High‐copy (2μ) expression of CAK1 suppresses both multiple smk1 developmental defects and multiple smk1 mutant alleles. Second, CAK1 mRNA accumulates during sporulation concomitant with SMK1 and SPS1 mRNA accumulation. Third, conditional mutants in CAK1 have been identified that complete meiosis I and II, but that are defective for spore wall assembly. Thus, an essential regulator of cell cycle progression is also required for spore wall morphogenesis.
How does CAK1 regulate spore wall morphogenesis? The genetic interaction demonstrated between SMK1 and CAK1 and the similarity in sporulation phenotypes of smk1 and cak1 mutants suggest that CAK1 functions in the SMK1 MAP kinase pathway. CAK1 may exert its effects upstream or downstream of SMK1. The ability of CAK1 to dosage suppress multiple smk1 MAP kinase defects suggests that the mechanism of dosage suppression may be a general increase in pathway activity, and thus that CAK1 functions upstream of SMK1 in the spore morphogenesis pathway. Cak1p phosphorylates Thr169 in the activating loop of Cdc28p, a modification essential for cell cycle progression. One possibility is that CAK1 positively regulates the SMK1 pathway through CDC28. Shuster and Byers (1989), using temperature‐sensitive cdc28 mutant backgrounds, have demonstrated that CDC28 is required for meiosis I and II. In addition, it has been shown that the CLB genes, whose products are involved in CDC28 activation, are required for the execution of meiotic events and that different CLBs play specialized roles in this process (Grandin and Reed, 1993; Dahmann and Futcher, 1995). Nevertheless, it appears that CDC28 activity is not required for spore wall assembly since cdc28 temperature‐sensitive strains are fully competent to assemble spore walls at the non‐permissive temperature (Shuster and Byers, 1989). These results may suggest that CAK1 functions in the SMK1 pathway by activating substrates other than Cdc28p. In vertebrates, CAK has been shown to activate multiple substrates, including distinct CDKs involved in diverse cellular functions. In addition, many protein kinases exist which contain a threonine in the presumed activating loop and thus could be substrates for positive regulation by CAK (Johnson et al., 1996). At present, two kinases have been described that function directly in the spore wall morphogenesis pathway: SPS1, a sporulation‐specific STE20 (PAK1) homolog, and SMK1. Each of these gene products contains a threonine in its activating loop that could serve as a phosphorylation site for positive regulation. Thus, in principle, either SPS1, SMK1 or as yet unidentified kinases that function in this signaling pathway could be direct targets of CAK1. Epistasis experiments using partial function mutants in other members of this pathway and direct biochemical approaches may resolve this issue.
It should be pointed out that our experiments do not address the requirement of CAK1 for the completion of meiosis I and II. We suspect that CAK1 is required for the completion of meiosis since CDC28 is required for meiosis and since CAK activity appears to be absolutely required for mitotic CDC28 function. The isolation of a cak1 mutant background that is able to complete meiosis with wild‐type efficiency, but that is defective for spore wall morphogenesis, may suggest either that there is a higher critical threshold requirement for the function of CAK1 in spore wall morphogenesis than in meiosis, or that pre‐existing pools of CAK1‐phosphorylated Cdc28p are able to support the meiotic function but not the spore wall morphogenesis function. In support of this latter possibility, we note that all of the eight independently derived cak1 temperature‐sensitive strains isolated in the course of this work continue to undergo normal mitosis for at least three or four divisions after they are shifted to the non‐permissive temperature. These results are consistent with the idea that a pool of Cdc28p may exist that is phosphorylated on Thr169 and that this pool can interact with the appropriate regulatory subunits (Clns and Clbs) to achieve progression through the cell cycle. If meiosis can be completed using such a pre‐existing pool of activated Cdc28p, this would suggest that regulated changes in CAK1 activity are not required for progression through meiosis. However, the possibility that Cdc28p does not require CAK activation for its role in meiosis or that there exists a CAK activity that is specialized for meiotic progression cannot be ruled out at this time.
What is the signal which leads to activation of the SMK1 MAP kinase pathway? The initiating signal is not the completion of meiosis I or II, per se (Klapholz and Esposito, 1980; Shuster and Byers, 1989; McCarroll and Esposito, 1994). Nor is the mere duplication and separation of the spindle pole bodies a prerequisite to spore wall morphogenesis (Schild and Byers, 1980; Shuster and Byers, 1989). Given that smk1‐Δ strains do progress through meiosis II with wild‐type efficiency, the initiating signal may require execution of a specific event that normally occurs after meiosis II but before assembly of the spore wall layers. Furthermore, the tight transcriptional regulation of this MAP kinase may generate a defined temporal window during which the developing ascospore is competent to respond to a SMK1 pathway signaling event. The ultrastructural events that normally occur between the completion of meiosis II and spore wall assembly include a thickening of the outer plaque of the spindle pole bodies, outgrowth of the prospore wall from these plaques, and subsequent deposition of spore wall material from within and/or around the prospore wall (Moens, 1971; Moens and Rapport, 1971; Byers, 1981). It is tempting to speculate that the spindle pole body and its associated morphological modification or perhaps the cellularization event that occurs as the prospore wall encompasses each of the haploid spores participates in the SMK1 signal. It is possible that changes in the enzymatic activities of particular gene products associated with the execution or completion of such upstream events could directly activate the spore wall morphogenesis pathway.
What are the implications of CAK1's role in spore wall morphogenesis? Once sporulation has been induced, a sequence of developmental events ensues in a pre‐programmed and highly coordinated fashion (Esposito and Klapholz, 1981; Honigberg et al., 1992; Honigberg and Esposito, 1994). The SMK1 MAP kinase pathway may provide a developmental checkpoint to ensure that early events (meiosis) are completed before the onset of later events (spore wall morphogenesis). The requirement of CAK1 for both cell cycle progression and spore wall morphogenesis suggests that Cak1p may serve as a regulatory nexus which coordinates these processes. It is possible that feedback regulatory interactions between Cdc28p and Cak1p may provide a mechanistic basis for activation of the SMK1 morphogenesis pathway.
Materials and methods
Strains and culture conditions
The genotypes and sources of strains utilized in this study are shown in Table III. Vegetative cultures were propagated in either YEPD (1% yeast extract, 2% peptone, 2% glucose), 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 acids, pH 5.5) supplemented with nutrients essential for auxotrophic strains at the levels specified by Sherman et al. (1986). Synchronous sporulation of diploids in liquid culture was achieved by inoculating logarithmic cells into YEPA (1% yeast extract, 2% peptone, 2% potassium acetate), expanding the culture for at least 7 h and to a density of 1×107 cells/ml, collecting cells by centrifugation, washing with 2% potassium acetate, and resuspending the cell pellet at 4×107 cells/ml in SM (2% potassium acetate, 10 μg/ml adenine, 5 μg/ml histidine, 30 μg/ml leucine, 7.5 μg/ml lysine, 10 μg/ml tryptophan, 5 μg/ml uracil). Sporulating cultures were maintained with vigorous aeration for 30–36 h. Sporulation of diploids on solid media was performed by patching or replica plating colonies to YEPD, allowing 12–18 h pregrowth, and then replica plating either directly to an SM plate (liquid SM with 2% agar, 0.1% yeast extract, 0.05% glucose) or a nitrocellulose filter which was then placed on an SM plate colony‐side facing upward. Sporulation was allowed to proceed at the appropriate temperature for 48–72 h.
Plasmid constructs and libraries and genetic screens
Plasmid names, markers and sources are detailed in Table IV. The mutagenized plasmid libraries for SMK1 and CAK1 were made in the pLAK40 (URA3/CEN/SMK1) and pMWB105 (URA3/CEN/CAK1) yeast shuttle vectors, respectively. In either case, the vector was hydroxylamine‐mutagenized in vitro for increasing times (Busby et al., 1982). The mutation frequency of the mutagenized plasmid pools was assessed by scoring transformants of Escherichia coli that are mutant at pyrF (the functional counterpart of the yeast URA3 gene) for the inactivation of the plasmid‐linked URA3 (Sikorski and Boeke, 1991). Using this approach, a plasmid library representing in excess of 500 000 independent primary E.coli transformants with an average URA3 inactivation frequency of 3.5% was generated. A high‐copy yeast genomic library contained in the YEp352 yeast shuttle vector was kindly provided by Dr Shelly Berger.
For isolation of smk1 conditional alleles, the mutant SMK1 library in pLAK40 was transformed into yeast strain LAKY70 (smk1‐Δ/smk1‐Δ). Approximately 500 000 independent transformants were pooled and frozen in multiple aliquots for further analysis. Transformants were plated onto selective SD medium at a density of 100–200 colonies per 100 mm diameter Petri plate. Colonies were sporulated at the permissive (26°C) and non‐permissive (34°C) temperatures and scored by the fluorescence assay as described below. The sequence of the entire open reading frame and 200 bp of promoter of smk1‐2 allele in pLAK40 was determined by standard dideoxy‐chain termination methods (Ausubel et al., 1987). smk1‐2 diploid strains were made by replacing SMK1 in MATa and MATα haploids and mating the two conditional smk1 haploids to each other or to a smk1Δ or SMK1 strain of opposite mating type. For smk1‐2 integrations, the KpnI–XhoI smk1 fragment of pLAK40 was subcloned into pRS406 to create the YIPsmk1‐2 construct, which was then linearized with BglII, and smk1 conditional strains selected by standard gene replacement techniques (Rothstein, 1991).
For the isolation of dosage suppressors of smk1‐2, the yeast genomic library contained in YEp352 was transformed into yeast strain MWY12. Over 50 000 independent transformants were sporulated at 34°C and scored for fluorescence. Three classes of plasmid were isolated which suppressed the smk1‐2 fluorescence defect at 34°C: class I (SMK1‐containing) plasmids were isolated nine times; class II (CAK1‐containing) plasmids were isolated three times; class III plasmids were isolated twice (to be described later). Of the class II dosage suppressors, the pHCS12‐5 plasmid contained the smallest genomic insert (chromosome VI, 3.4 kb Sau3AI fragment, bp designation 75 816–79 247). The two complete open reading frames contained in the insert (YFL030w and YFL029c) were separately amplified from pHCS12‐5 by PCR and subcloned into YEp352 (pMWB76 and pMWB77). In order to obtain the entire presumed promoter region of YFL029c, the YFL029c open reading frame including 780 bp of 5′ non‐coding sequence was amplified from genomic DNA (bp designation 77 917–79 940) and subcloned into YEp352 (pMWB106) or pRS316 (pMWB105). The dosage suppression phenotype of pMWB106 was indistinguishable from that of pMWB77 and the pMWB106/pMWB105 CAK1 constructs were used in all subsequent studies. A CAK1 disruption allele was made by replacing the 0.7 kb NspV fragment contained in the CAK1 coding region with the TRP1 open reading frame (0.8 kb BamHI fragment of YDp‐W) via Klenow/blunt‐ended DNA ligation. The cak1::TRP1 allele was excised from the plasmid backbone and used for one‐step gene replacement in a diploid CAK1 strain by standard methods (Rothstein, 1991). Integrations and subsequent tetrad analyses were done in both the SK1 (MWY30) and W303 (MWY45) genetic backgrounds.
The cak1 mutant alleles ‐K31R, ‐D169R, and ‐Δ31, were made using PCR techniques and confirmed by DNA sequencing. The ability of these cak1 alleles to support vegetative growth was determined by transforming pRS316 (CEN URA3) vector alone, or pRS316 containing a wild‐type or mutant CAK1 allele into yeast strain MWY30 (cak1::TRP1/CAK1). Single transformants were sporulated in liquid, the spores separated enzymatically and physically as described above, and spore density of each culture quantitated in a hemacytometer. An equal number of spores for each sample was plated on SD lacking tryptophan and uracil to select for cak1::TRP1 haploid transformants. The number of Trp+Ura+ haploid survivors was determined for the mutants and compared with that of vector alone (0% survival) and wild‐type (taken as 100% survival). The ability of the cak1 mutant alleles contained in YEp352 to dosage suppress the smk1‐2 mutant sporulation phenotype was tested by fluorescence assay as previously described.
For the isolation of cak1 conditional alleles, the mutant CAK1 library in pMWB105 was transformed into yeast strain MWY30 (cak1::TRP1/CAK1). Approximately 5000 independent transformants were pooled as above. Pooled transformants were revived in SD media, and then sporulated en masse in liquid at 30°C. Glusulase (500 μl) and glass beads (2 ml) were added directly to a 5 ml sporulated culture which was incubated on a rolling drum until the ascal wall was digested and the spores physically separated as determined by microscopy. Spores were plated on SD lacking uracil and tryptophan to select for cak1::TRP1 spores harboring pMWB105 with a mutagenized cak1 allele and incubated at 26°C. The resulting colonies were replica plated to a fresh SD plate and incubated at 34°C. Those colonies which failed to grow at 34°C were chosen for further analysis.
For light microscopy, cells were fixed in ethanol and stained with DAPI (Sherman et al., 1986). Fixation and staining with rhodamine phalloidin and calcofluor were carried out as described by Sherman et al. (1986). Samples were viewed and photographed as a wet mount under phase‐contrast oil immersion optics using a Nikon Optiphot equipped for epifluorescence. For electron microscopy, cells were pelleted by centrifugation and fixed in 2.5% glutaraldehyde in 0.13 M cacodylate buffer, pH 7.4. The specimens were post‐fixed in 1% osmium tetroxide for 1.5 h, dehydrated through a graded series of ethanol, and embedded in Spurr low viscosity resin. Ultrathin sections of 600 Å thickness were cut, mounted on copper grids, and stained with saturated aqueous uranyl acetate and Reynold's lead citrate. Sections were viewed and photographed using a JEOL 100B transmission electron microscope at 60 or 80 kV.
Assays for spore wall assembly
The procedure for the fluorescence assay is modified from the method of Esposito et al. (1991). Nitrocellulose filters with sporulated colonies or patches were placed colony‐side facing upward in a Petri plate containing ascal wall lysis buffer [350 μl 0.1 M Na‐citrate, 0.01 M EDTA, pH 5.8; 70 μl glusulase (Dupont NEE‐154, crude solution); 15 μl β‐mercaptoethanol], incubated at 37°C for 4 h, briefly blotted on 3MM Whatman paper to remove excess liquid, and then placed in a Petri plate containing 300 μl concentrated ammonium hydroxide. Fluorescence was viewed under a 304 nm UV light source and photography performed using a blue filter (Kodak, Wratten filter, #98).
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. (1990a). Sensitivity of cells to ether exposure (3 min with constant gentle rocking) was assayed according to the method of Dawes and Hardie (1974). The level of SPS100 expression was assessed as β‐galactosidase activity in sporulating yeast strains harboring plasmid p152‐SPS100TB (the gift of J.Segall) which contains the entire SPS100 promoter and a portion of the coding sequence fused in‐frame to the lacZ reporter gene. The wild‐type diploid (LNY150) harboring p152‐SPS100TB expressed β‐galactosidase activity starting at 12 h after transfer to SM, maximal activity levels were reached by 20 h, and these levels remained constant for at least the next 48 h. β‐Galactosidase activity levels in smk1 mutant strains were determined at 24 h after transfer to SM. Preparation of cell lysates and β‐galactosidase assays were carried out as described by Rose et al. (1990).
Northern blot hybridization
A single filter was used for all Northern analyses shown in Figure 6. These RNA samples were previously described and characterized in Krisak et al. (1994). DNA probes from the coding regions of the indicated genes were isolated from preparative agarose gels, 32P‐radiolabeled by random priming (Ausubel et al., 1987), and used in hybridization analysis at 106 d.p.m./ml. DNA probes used were as follows: CAK1, 1.5 kb BamHI–SalI fragment of pMWB106; SMK1, 0.8 kb StyI fragment of pLAK40; SPS1, BglII–EcoRI SPS1‐specific fragment from pSPS1‐URA3.
We thank Helena Friesen, Perry Hall, Jacqueline Segall and Randy Strich for helpful comments and for critically reading the manuscript, and Albert Sedar for assistance with the electron microscopy. This work has been supported by a grant (GM45772) from the National Institutes of Health and a grant (MCB‐9630656) from the National Science Foundation.
- Copyright © 1997 European Molecular Biology Organization