Cloning of the phal Gene Encoding the Catalytic Subunit of the CAMP-dependent Protein Kinase in Schizosaccharomyces pombe*

We have isolated Schizosaccharomyces pombe genes that confer sterility to the fission yeast cell when expressed from a multicopy plasmid. One of these genes strongly hybridized to a probe carrying the open reading frame of Saccharomyces cerevisiae TPKl, which en- codes a catalytic subunit of the CAMP-dependent protein kinase (protein kinase A). This s. pombe gene, named pkal, has a coding potential of 512 amino acids, and the deduced gene product is 60% identical with the S. cerevisiae Tpkl protein in the C-terminal 320 amino acids. Disruption of pkal slows cell growth but is not lethal. The resultant cells, however, are highly derepressed for sexual development, readily undergoing conjugation and sporulation in the absence of nitrogen starvation. They are, thus, phenotypically indistinguish-able from the adenylyl cyclase-defective (cyrl-) cells pre- viously characterized, except that the pkal- spores are retarded in germination, whereas the cyrl- spores are not. Disruption of pkal is epistatic to a defect in cgsl, which encodes the regulatory subunit of protein kinase A. These results strongly suggest that the product of pkal is a catalytic subunit of protein kinase A and, fur-thermore, that s. pombe has only one gene encoding it. This situation contrasts with the case of S. cerevisiae, phosphate dehydrogenase promoter and terminator of pKT10. The resultant plasmid named

to the GenBankTMIEMBL Data Bank with accession numberis) 023667.
The nucleotide sequence(s1 reported in this paper has been submitted MA 02115. 8 To whom correspondence should be addressed. Tel.: 81-3-3814-9620; late. It is, thus, highly likely that the inhibitory effect of CAMP on S. pombe sexual development is exerted through protein kinase A. This is consistent with the widely accepted notion that cAMP causes its effect primarily by activating protein kinase A in eukaryotic cells (9).
The above results lead to a conclusion that activation of protein kinase A is inhibitory, whereas inactivation of it is stimulatory for the initiation of sexual development in S.
pombe. Consistent with this, the level of intracellular CAMP decreases when S. pombe cells initiate mating and meiosis physiologically under poor nutrition (4). Based on these observations, we proposed that the level of intracellular CAMP is a key determinant of the commencement of sexual development in the fission yeast (4, 7 ) .
The CAMP cascade thus appears to play a central role in transmission of the nutritional signal to gene expression when S. pombe cells commit themselves to sexual development. With respect to the regulatory mechanism of adenylyl cyclase activity in S. pombe, we showed t h a t a G-protein subunit is likely to control it according to the nutritional conditions (10). We also showed that a decrease in the intracellular cAMP level, and hence in the protein kinaseA activity, results in induction of the stell gene, the product of which is a transcription factor that activates genes necessary for sexual development (11). These findings suggest that S. pombe will provide a good experimental system to work out the signaling path involving the cAMP cascade.
The observations described above unambiguously predict the existence of the catalytic subunit of protein kinase A in S. pombe cells. However, the gene encoding i t has not been identified yet. We rationalized that the gene may be isolated as one that inhibits sexual development if overexpressed. There are already two examples of S. pombe genes that exhibit this characteristic, the pacl gene encoding a double-stranded RNase (12) and the pac2 gene with unknown function. ' We therefore screened for S. pombe genes (carried on a multicopy plasmid) that are able to convert the host cell to sterility. One of the genes thus isolated, which we named pkal, indeed encodes a protein highly homologous to the protein kinase A catalytic subunits of various organisms. Phenotypes of the pkal-disrupted strain strongly suggest that pkal is the sole gene that encodes the catalytic subunit of protein kinase A in S. pombe. EXPERIMENTAL PROCEDURES Yeast Strains and Media-Yeast strains used in this study are listed in Tables I and 11. General genetic procedures for S. pombe were according to Gutz et al. (13), and those for Saccharomyces cereuisiae were according to Sherman et al. (14). Transformation of S. cereuisiae was performed by the lithium method (151, and an improved lithium method (16) was used for transformation of S. pombe cells. Complete medium YPD and minimal medium SD (14) were used for routine cultivation of both yeast species. Synthetic medium PM and its derivative PM-N, which lacks a nitrogen source (NH,Cl), were used in physiological ex-H. Kunitomo, A cereuisine TPKl open reading frame (ORF) sequence using the polymerase chain reaction (PCR). Each primer consisted of 27 nucleotides. One had 18 nucleotides corresponding to the N-terminal6 amino acids of Tpkl, whereas the other had the same number of nucleotides corresponding to the C-terminal 6 amino acids. Both primers had an additional nine nucleotides at their 5' end, which provided an EcoRI cutting site. PCR was performed according to a standard method (19,21), using S. cereuisiae genomic DNA as a template. DNA carrying the entire TPKl ORF was thus amplified and recovered from an agarose gel after separation by electrophoresis. The DNA was labeled isotopically and used as a probe in Southern analysis. Essentially the same procedure was followed to prepare the probe carrying part of thephal ORF (Ala""-Ala'"). Blotting Analysis-Blotting analysis of DNA was performed according to Southern (22). The stringency conditions employed in this study are as follows. Either Hybond-N or Hybond-N+ (Amersham Corp.) was used as the membrane to accept DNA. Hybridization was done in 5 x SSC (0.75 LI NaCI, 0.075 11 sodium citrate) containing 0.1% N-lauroylsarcosinate, 0.02% sodium lauryl sulfate, and 3% blocking reagent (Boehringer Mannheim) a t 55 "C for 15 h. Wash was carried out in 2 x SSC (0.3 M NaCI, 0.03 M sodium citrate) containing 0.1% sodium lauryl sulfate. Wash a t room temperature for 5 min each was repeated four times, and wash at 50°C for 30 min each was repeated four times. Blotting analysis of RNA (Northern blotting) was done as previously described (17,23).
DNA Sequence Determination-The nucleotide sequence of two adjacent HindIII fragments that in combination cover the entire pkal gene was determined by the dideoxy chain termination method (24) using an automated DNA sequenator (Applied Biosystems). Subclones for sequencing were generated by progressive deletion with exonuclease I11 and S1 nuclease (Takara Shuzo Co., Kyoto), according to Henikoff (25). To rule out the possibility that a small HindIII fragment is located between the two fragments, sequencing across the joint was carried out using the original clone pAKl and two synthesized sequencing primers. All parts of the sequence (see Fig. 2) were determined in both directions at least once.
Disruption of phal-One-step gene disruption (26) of pkal was carried out as follows. A0.9-kb PuuII fragment within the clonedphal ORF was replaced by a 1.8-kb S. pomhe rtrn4' cassette (27). A 2.9-kb HindIII fragment carrying this disrupted pknl gene was used to transform both a haploid strain, JY742, and a diploid strain, JY765. Stable Ura' transformants were selected in each case. A successful replacement of the wild-type pkal allele by the disrupted gene was confirmed by PCR and Southern blot analysis.  Expression of pknl in S. cereuisiae-To express S. pornhe pkol in S. cereuisiae, we used an S. cereuisiae expression vector pKTlO, which carries the ARS sequence of 2-pm DNA, URA3, and the promoter and terminator regions of the glyceraldehyde-phosphate dehydrogenase gene (28). The ORF sequence ofpknl was amplified by PCR using a pair of primers carrying an EcoRI cutting site at one end, and the obtained fragment was inserted into the EcoRI site between the glyceraldehydephosphate dehydrogenase promoter and terminator of pKT10. The resultant plasmid was named pAKS1.

RESULTS
Isolation of Plasmids That Confer Sterility to the Wild-l!vpe S. pombe Cells-An S. pombe genomic library based on a multicopy vector pDB248' (20) was introduced into mating-proficient homothallic strains JY450 and JY476. Altogether 50,000 transformants were tested by iodine staining of the colony for their ability to mate and sporulate (13). Nearly 300 transformants were judged to have lost the ability, and plasmid-segregation analysis suggested that one-third of them were sterile due to the plasmids they retained. Plasmids were recovered from them, and those which hybridized to either pacl or pac2, as mentioned in the Introduction, were eliminated from the following analysis. We obtained 61 novel plasmids that are able to endow sterility to the host cell upon transformation. Preliminary restriction mapping of these plasmids suggested that at least four different genes are responsible for the sterility (data not shown).
Identification of the p k a l Gene That Encodes a Putative Catalytic Subunit of Protein Kinase A-The 61 plasmids obtained in the above screening were tested for their ability to hybridize to a probe carrying the S. cerevisiae TPKl ORF, which encodes a catalytic subunit of protein kinase A. One of the plasmids, which we hereafter call pAK1, hybridized strongly to the probe (data not shown). Cells transformed with pAKl were defective in mating and assumed a long rodlike shape under nitrogen starvation (Fig. l), like cells that have a high level of intracellular CAMP or protein kinase A activity (2, 3,7,8).
The genomic DNA of S. pombe inserted in pAKl was about 5.5 kb in length. Subcloning of this insert suggested that a HindIII site is within the region essential for expression of the inhibitory activity (data not shown). We, therefore, sequenced the two HindIII fragments flanking this site, which were 2.0 and 0.9 kb in length, respectively. An ORF encompassing the essential region was carried by these two fragments, and we call this gene pkal hereafter.  that in combination carry the entire p k a l gene is shown. The contiguity of the two fragments has been confirmed by sequencing across the HindIII site between them, which is underlined. Numbering of the amino acid residues starts with the first methionine codon of the possible ORF.

ACGmGTmGTmAAGATACTCGAATGCTTATTGATTCATTGTACTATTGCCTTAAAGCGTTGAGCAACGAATGCCTA~GA~AGTGCAATGCAAGTGACG~GTAQCACTQA T m G C A A T T C A C C A A T A T T G G C T T C A G~G T G C T G G T G T A A T G G A C C G A C m A T G m G G A T G G G A T A T G A T G A T C T C A G T T G A A G C T A m A~G A T G~C C A G Q C M T T A G C C T C C T C T G A T G T A m C C A C T C C A T m A m A T A T A T T C T G~A T G G T T A T G T T A A A C C C C A A G T C A m C T T C T T C A T G T T G A A C A G A C~G A G C A T T G m G T T
The nucleotide sequence of the pkal gene and its deduced amino acid sequence are shown in Fig. 2. There is no evidence of introns in the gene structure, and the ORF can encode 512 amino acids. The deduced P k a l protein has unambiguous homology to catalytic subunits of protein kinase A. Fig. 3 compares the amino acid sequence of P k a l with S. cereuisiae, Drosophila melanogaster, and bovine counterparts. Pkal is 60% identical with the S. cerevisiae TPKl gene product in the Cterminal 320 amino acids. It is approximately 50% identical with both the Drosophila and the bovine proteins in the same 320 amino acids. Pkal is most similar to the S. cerevisiae TPK2 gene product among known proteins (63% identity in the Cterminal 320 amino acids), although their comparison is not displayed here. S. pombe Pkal has a long N-terminal region that is missing in the other proteins. The meaning of this region is unclear, but the N terminus of P k a l again shows considerable homology to that of the bovine protein (6 out of 11 amino acids are identical; see Fig. 3).
Is pkal a Member of a Gene Family?-Because three genes, namely TPKl, TPK2, and TPK3, have been shown to encode the protein kinase A catalytic subunits in S. cerevisiae (291, we examined whether pkal is part of a gene family. S. pombe genomic DNA was digested with either EcoRI, HindIII, or BamHI and analyzed by Southern blotting under low strin-gency conditions using part of the pkal ORF sequence as a probe. This probe intensively hybridized only with a band representing pkal itself (Fig. 4A). Although an additional weak hybridization band was visible in the HindIII digest (Fig. 4A,

lane H ) , this hybridization signal appeared to be insignificant compared with the clear cross-hybridization between TPK family members reported by Toda et al. (29).
S. pombe genomic DNA was also probed by the TPKl sequence. The major band detected by this probe was obviously the same as that detected by the pkal probe, although the intensity of hybridization was considerably reduced (Fig. 4B). This indicates that the S. pombe gene that has the best homology to S. cerevisiae TPKl is p k a l . However, pkal is more similar to TPK2 than TPKl in both nucleotide and amino acid sequences (data not shown). Thus, it is unlikely that S. pombe has a gene family that precisely mimics the TPK family of S. cereuisiae. The TPKl probe generated several faint hybridization bands in a long exposure, which may represent only distantly related kinase genes (Fig. 4B'). All of these observations lead to the notion that pkal is the only gene encoding the catalytic subunit of protein kinase A in S. pombe. Genetic evidence described below confirms this notion. Phenotypes Caused by Disruption of the pkal Gene-S. pombe strains carrying disrupted pkal were constructed as described under "Experimental Procedures." Either a haploid strain ST742 (h' ade6-M226 leu2 ura4-DI8) or a diploid strain JY765 (h'lh-ade6-M210/ade6-M226 1eullleu.l ura4-Dl8Iura4-0 1 8 ) was used as the parent. 52633, derived from JY742, was proven to be a proper disruptant by PCR and Southern blot analysis of the chromosome structure (data not shown). For an unclear reason, the mating type of 52633 was changed to h" (homothallism).
Several diploid strains having one pkal allele properly disrupted were obtained from JY765. One of them, named 52634, was subjected to tetrad analysis. Dissection of ascospores produced by 52634 gave pkal-disrupted haploid strains. Because these pkal disruptants exhibited essentially the same phenotypes as 52633, we used 52633 as a representative in the following analysis. 52633 forms colonies very slowly even on the complete medium YF'D, and many zygotes and asci can be seen in these colonies. When 52633 is cultured in liquid medium, cells that retain the proper genotype are readily overgrown by apparent variants. These characteristics are quite similar to those of the cyrl-disrupted strain, which has no detectable adenylyl cyclase activity and is highly derepressed for sexual development (7,8). This suggests that the function of P k a l is indeed closely related to CAMP. Furthermore, because disruption of pkal alone caused these obvious phenotypes, S. pombe apparently has no functional homolog of Pkal. This is consistent with our assumption that pkal is a unique gene.
Asci produced by self-conjugation of 52633 were dissected. The average viability of the progeny spores was 50%. They formed colonies slowly, and the colony size varied greatly from spore to spore, although every spore must have the same genotype. The difference in the colony size was not due to different growth rate of the cells but due to differential timing of germination of the pkal-spores (data not shown).
In summary, disruption of pkal slows cell growth but is not lethal. The pkal-disrupted cells are derepressed for sexual development in the presence of rich nutrition. The pkal-spores are apparently impaired in germination and often fail to resume vegetative growth. The last observation may suggest a small phenotypic difference between the pkal-and the cyrlmutants because we have never recognized germination defi-ciency in cyrl-mutants, which otherwise behave quite similarly t o pkal- (7).
Expression of pknl-Expression of pkal mRNA in various strains under either nitrogen-rich or nitrogen-depleted conditions was measured by Northern blot analysis. A single RNA species of 2.9 kb in length was detected as the pkal transcript in all conditions examined. Neither the mating type of the cell, the abundance of a nitrogen source, nor the status of the cyrl allele greatly affected the level of expression of the pkal gene (data not shown).
Genetic Interaction of pkal with S. pornbe cgsl and patl-The cgsl gene of S. pombe encodes the regulatory subunit of protein kinase A (3). Mutants defective in cgsl are sterile because of the constitutive activation of protein kinase A. We constructed a strain in which both cgsl andpkal are disrupted. The phenotype of this strain is exactly the same as the pkal disruptant (data not shown). Existence of the null cgsl allele in this strain was proven by genetic and Southern analysis. Thus, pkal is epistatic to cgsl, again confirming that it encodes the catalytic subunit of protein kinase A.
Loss of function of the patl (also called ran1 1 gene derepresses sexual development ectopically in S. pombe (30-32). An increase in the level of intracellular CAMP can suppress this uncontrolled sexual development by repressing the expression of key genes essential for sexual development (2, 11, 17). Because pkal has not been isolated as a multicopy suppressor of patl, whilepacl andpac2 have (12, 17), we examined whether overexpression of pkal could inhibit ectopic sexual development driven by loss ofpatl function. To do this, apatl'* strain 52409 was transformed with either pAKl or the vector pDB248'. At 25 "C, both transformants could grow vegetatively. The transformant carrying pDB248' failed t o grow at 32 or 37 "C, being committed to ectopic sexual development due to loss of p a t l function. However, the transformant carrying pAKl could grow at the restrictive temperature (Table 111). Thus, overexpression of pkal has an ability to suppress p a t l .
Effects of Expression of pkal in S. cereuisiae-The activity of protein kinase A is essential for cell growth in S. cereuisiae (29, 33). The level of CAMP is regulated by Ras proteins in this yeast (34). The product of S. cereuisiae CDC25 is a positive regulator of Ras (35)(36)(37). Hence, the activity of CDC25 is essential for cell " Five independent transformants were replica-plated and tested for their growth at the respective temperature indicated. Results were scored after 3 days of incubation. The number of grown transformants/ total is given. growth, and a cdc25'" strain KMY208-3C fails to grow at the restrictive temperature because the cells do not have enough cAMP and protein kinase A activity to support growth. To examine whether pkal can generate protein kinase A activity in S. cereuisiae, we expressed it in KMY208-3C cells from a multicopy plasmid pAKS1. This plasmid carries the coding region of phal under the transcriptional control of the S. cereuisiae glyceraldehyde-phosphate dehydrogenase promoter. The KMY208-3C cells were converted to Ts' by pAKSl but not by the parental vector pKTl0 (Fig. 5). This again strongly sug- gests that S. pombe P k a l has protein kinase A activity.
S. cerevisiae cells become heat shock sensitive if they have a high level of either CAMP or protein kinase A activity. We, therefore, examined heat shock sensitivity of a wild-type S. cerevisiae strain (FtAY3A-1) transformed with pAKS1. The parental and the transformed strains showed no significant difference in the sensitivity to heat shock (data not shown). This result may suggest that the expression of pkal from pAKSl was sufficient to suppress cdc25 but not high enough to make S. cerevisiae cells sensitive to heat shock. Alternatively,

S.
pombe protein kinase A and S. cerevisiae protein kinase A may have somewhat different substrate specificities. DISCUSSION This study has shown that S. pombe pkal most likely encodes the catalytic subunit of CAMP-dependent protein kinase and, furthermore, that pkal appears to be the only gene encoding it in the fission yeast. This situation contrasts with that in S. cerevisiae, where three genes encode the subunit. The redundancy of the genes in s. cereuisiae may reflect that the activity of protein kinase A is indispensable for cell growth in this organism (29). In S. pombe, however, although protein kinase A has a pivotal role in the developmental choice, its loss does not result in growth arrest, as shown in this study.
Including the results of this study, four major components of the CAMP cascade are cloned in S. pombe, namely adenylyl cyclase (5, 6), CAMP phosphodiesterase (3, 4), the regulatory subunit of protein kinase A (3), and the catalytic subunit of it. Disruption and overexpression of these genes have given consistent results. Any genetic manipulation that increases the level of intracellular CAMP or the activity of protein kinase A makes S. pombe cells incapable of sexual development, whereas any manipulation that decreases the cAMP level or the protein kinase A activity propels cells toward ectopic sexual development (Refs. 3,4,7,8, and this study). This agrees very well with the formula we proposed previously that cAMP serves as the second messenger in the signaling pathway that controls gene expression for sexual development in response to the nutritional conditions (4, 7).
The CAMP cascade in S. pombe is involved also in the regulation of expression of the glucose-repressible gene fbpl, which encodes fructose-1,6-bisphosphatase (38). This gene is expressed in the absence of glucose, where the level of intracellular cAMP is decreased. Hoffman and Winston (39) isolated mutants in which transcription of fbpl is constitutive and named them git (glucose-insensitive transcription). These mutations identified 10 genes, and one of them, git2, has been shown to be the same as cyrl, the gene for adenylyl cyclase (38). T h e p k a l a n d g p a 2 genes, the latter of which possibly regulates the activity of adenylyl cyclase (lo), are also found among the git genes3 However, fbpl is not regulated by &ell4 (see below). Thus, it is an interesting question how the two pathways, one for sexual development and the other for fbpl expression, are differentially regulated by protein kinase A. This study predicts that the Pkal protein bears an extra N-terminal region not found in other protein kinase A. This region is not essential for the catalytic activity of the enzyme, because a domain carrying the kinase consensus alone apparently has the activity (data not shown). Whether this N-terminal region is involved in regulation of the enzyme activity unique to S. pombe remains unsolved.
Although the phenotypes of the pkal-defective and the cyrldefective mutants are quite similar, only pkal spores display apparent germination disability. This may mean that a low activity of protein kinase A in the cyrl mutant, due either to leakiness of the enzyme activity in the absence of cAMP or to a trace of cAMP in the medium, can accelerate the germination process.
The major substrate(s) of protein kinase A in S. pombe is yet unclear, as is true with many other eukaryotes. The most significant physiological function of S. pombe protein kinase A appears to be in repression of stell, which encodes a key transcription factor for sexual development (11). We pointed out a possibility t h a t t h e S t e l l protein autoregulates its expression, with its activity being down-regulated by phosphorylation by protein kinase A (11). Alternatively stell may be regulated by another transcription factor that is a substrate of protein kinase A. Differentiation of these two possibilities will be required to identify the major substrate of protein kinase A in S. pombe. and Akio Toh-e for S. cerevisiae strains and the vector plasmid pKT10.