Mol Gen Genet (1991) 227:452-457 0026892591001972 © Springer-Verlag 1991

CDC7 protein kinase activity is required for mitosis and meiosis in Saccharomyces cerevisiae Vicky Buck, Anne White and John Rosamond Molecular BiologyGroup, Department of Biochemistryand Molecular Biology,University of Manchester, Oxford Road, Manchester M13 9PT, UK ReceivedMay 16, 1990 Summary. The product of the CDC7 gene of Saccharomyces cerevisiae has multiple cellular functions, being needed for the initiation of D N A synthesis during mitosis as well as for synaptonemal complex formation and commitment to recombination during meiosis. The CDC7 protein has protein kinase activity and contains the conserved residues characteristic of the protein kinase catalytic domain. To determine which of the cellular functions of CDC7 require this protein kinase activity, we have mutated some of the conserved residues within the CDC7 catalytic domain and have examined the ability of the mutant proteins to support mitosis and meiosis. The results indicate that the protein kinase activity of the CDC7 gene product is essential for its function in both mitosis and meiosis and that this activity is potentially regulated by phosphorylation of the CDC7 protein. Key words: C D C 7 - Saceharomyces cerevisiae - Protein kinase - Cell cycle DNA synthesis

Introduction The CDC7 gene of Saccharomyces cerevisiae encodes a 58.2 kDa protein whose predicted amino acid sequence contains a region with significant homology to the functional catalytic domain of a number of serine-threonine protein kinases (Patterson etal. 1986; Hanks etal. 1988). This domain of the CDC7 protein contains all of the key residues within the several short stretches of amino acids that are highly conserved in protein kinases, although the organisation of these conserved residues differs markedly from that found in all other protein kinases (Patterson et al. 1986). Despite this, the CDC7 gene product is enzymically functional as a protein kinase and can transfer phosphate from ATP to Offprint requests to: J. Rosamond

histone (Bahman et al. 1988; Hollingsworth and Sclafani 1990). Understanding the cellular function of the CDC7 protein kinase is complicated by the multiple roles that the protein can play in yeast physiology. In addition to a defect in error-prone repair (Njagi and Kilbey 1982), cells carrying a thermosensitive cdc7 lesion arrest during mitotic cell division at the G1-S phase bounda'ry without initiating D N A synthesis, although on return to permissive conditions cells can complete a round of DNA replication without recourse to protein synthesis (Hartwell 1973; Hereford and Hartwell 1974). This contrasts with the behaviour of diploid, homozygous cdc7 cells undergoing meiosis, when DNA synthesis initiates normally but the cells fail to commit to recombination, and do not form a synaptonemal complex or mature ascospores (Simchen 1974; Schild and Byers 1978). This implies that the cdc7 mutation affects the mitotic and meiotic pathways in quite distinct ways. We have sought to determine whether the CDC7 protein functions differently in mitosis and meiosis and, in particular, whether both pathways require the protein kinase activity of the CDC7 gene product. We have constructed mutations in the cloned CDC7 gene at sites that should be important if protein kinase activity is needed for the biological functions of the protein. The results that we present here are consistent with the idea that the protein kinase activity is essential for the function of the CDC7 protein in both mitosis and meiosis, and that this activity might be regulated in part by phosphorylation of the CDC7 protein itself. Materials and methods Strains and media. The strains and genotypes of Eseherichia coli and S. eerevisiae used in this study are listed in Table 1. E. coli strain HW87 was used for the

routine maintenance and propagation of plasmids, while strains HB2151 and HB2154 were used for recovery of

453 Table 2. Plasmids used in this study

Table 1. Strains and phage used Strain

Genotype

Escherichia coli : HW87 A (araD139-leu) A lacX74 gal recA hsdR rpsL HB2151 ara A (&e-pro) thi/F' lacIq lacZ A M 15 proA +B + HB2154 HB2151 rnutL: :Tnl0 Saccharomyces cerevisiae : SB311 MATa/MATe ura3/ura3 trpl/trpl cde7-1/edc7-1 SB433 MATa his3 ura3 ade2 arg4 lysl leu2 trpl cdc7-1

Phage : M13mp18 amIV

M13 cloning vector with amber mutation in M13 gene IV

Source or reference

Patterson et al. 1986 Carter et al. 1985 Carter et al. 1985 Bahman et al. 1988 This study

Carter et al. 1985 YanischPerron et al. 1985

mutagenised template (see below). Bacterial cultures were grown in L-broth or supplemented M9 minimal medium (Miller 1972); where appropriate, ampicillin and tetracycline were added to final concentrations of 40 gg/ml and 20 lag/ml respectively. Yeast cells were routinely grown in either yeast extract-peptone-dextrose (YEPD) or supplemented synthetic minimal medium (Sherman et al. 1979). Transformation. E. coli was transformed by the method of Warren and Sherratt (1978). S. cerevisiae strains were transformed after spheroplasting exactly as described previously (Sherman et al. 1979). Sporulation. Diploid cells were grown at 23 ° C to early log phase in 0.67% yeast nitrogen base, 1% potassium acetate, 0.1% yeast extract, 0.05% glucose with any necessary auxotrophic supplements. Ceils were recovered by centrifuging at 3000 x g for 3 rain, washed twice in pre-warmed 1% potassium acetate, then resuspended at approximately 108 cells/ml in 1% potassium acetate with any necessary supplements. For temperature sensitivity experiments, these samples were split into two aliquots, one of which was incubated with vigorous aeration at 2 3 ° C (permissive conditions) and the other at 32.5° C (restrictive conditions). Mature ascospores were scored in both samples by microscopy after incubating for 4 days. D N A manipulations. Restriction enzymes, T4 D N A li-

gase, Klenow polymerase and polynucleotide kinase were purchased from BCL and used according to the manufacturer's recommendations. Plasmids. The plasmids used in this work are listed in

Table 2. Plasmid D N A was prepared from cultures of

Plasmid designation

Relevant properties"

Source or reference

pMPI01

Apr, TRP1, ARS1, with CDC7 in a 3.3 kb yeast genomic fragment Apr, Tetr, URA3,

Patterson et al. 1986

YCp50

CEN4, ARS1

pVB4

pVB441 pVB4871 pVB471 pVB4451 pVB4461 pVB4951 pVB4961

Apr; YCp50 with CDC7 in a 1.8 kb DraI fragment As pVB4 expect cde7.Met76 As pVB4 except edc7.Glu76 As pVB4 except cdc7.Gln77 As pVB4 except cdc7.Asn182 As pVB4 except cde7.Gln288 As pVB4 except cdc7.Ser281 As pVB4 except edc7.Ala281

Rose et al. 1987 This study

This study This study This study This study This study This study This study

Apr, ampicillin resistance; Yetr, tetracycline resistance; ARS, autonomously replicating sequence in yeast; CEN, centromere function in yeast E. coli by CsCl-ethidium bromide density gradient cen-

trifugation after detergent lysis (Humphreys et al. 1975). Mutagenesis. The template for mutagenesis was a 1.8 kb DraI fragment o f yeast genomic D N A derived from pMP101 and cloned into the Sinai site of M13mpl8-

amIV (Patterson et al. 1986; Carter etal. 1985). Point mutations were introduced into the C D C 7 gene within this fragment by the two-primer oligonucleotide-directed method of Carter et al. (1985). The following mutagenic oligonucleotides were used: 5'GCTTTGATGAAAATA3'; 5'GTTGCTTTGGAGAAAATATA3'; 5'TTGAAGCAAATATAC3'; 5'GCTTGTTAATTTTGGTC3' ; 5'GGGCACCACAAGTGTTA3'; 5'AGAGCAGGGTCTCGTGGATTT3'; 5'AGAGCAGGGGCTCGTGGATTT3'.

cdc7.Met76, cdc7.Glu76, cdc 7. Gln 77, cdc7.Asn182, cdc 7.Gin288, cdc7.Ser281, cdc7.Ala281,

In all cases the second primer was SELl (Carter et al. 1985) which has the sequence 5 ' A A G A G T C T G T C C A T CACY and which converts the M13amIV mutation to wild type in the newly synthesised strand. After mutagenesis, D N A was transformed into E. coli HB2154 and the cells plated out on a lawn o f E . coli HB2151. Mutants were identified by plaque hybridisation (Zoller and Smith 1984) under semi-stringent conditions using the appropriate mutagenic oligonucleotide as a probe after labelling with [,/-a2p]ATP and polynucleotide kinase (Maxam and Gilbert 1977). All mutations were confirmed by chain-terminator sequencing (Sanger et al. 1977). P l a s m i d constructions. D N A fragments carrying a mutated C D C 7 gene were recovered into the yeast shuttle vector YCp50 (Rose et al. 1987), which carries the U R A 3

454 selectable marker, the A R S 1 replication origin and CEN4 for single-copy maintenance in yeast (Tschumper and Carbon 1983). The CDC7 gene was recovered from M13mp18amIV by digesting with EcoRI and SalI, and the 1.84 kb fragment carrying CDC7 was cloned into YCp50 which had also been cut with EcoRI and SalI.

Results Construction of mutant alleles of CDC7 The template used for the oligonucleotide-directed construction of mutant alleles of CDC7 was a 1.8 kb DraI fragment derived from pMP101 (Fig. 1A; Patterson et al. 1986). This fragment carries the essential basal promoter element (Ham et al. 1989) and 59 bp of T-flanking sequence in addition to the complete CDC7 coding region and is sufficient to complement both mitotic and meiotic lesions in a cdc7-i strain. Within the CDC7 cod-

A -224 Dra I

+I

+1583

ATG

CDC70RF

TAG

I . . . . . . . . . . . . . . . . . . . . . . . . . . .

A T P - b i n d i n g site

Consensus

* * B - G - - 6 ~A

CDC7 +

GE GT F S ~

- - I

I

-

-

P h o s p h o a c c e p t o r site

* - K .... A [ K K ~ I

Mutations

Dra I

I

M

* H R B ~-

* * (21) * D F G ---~-~ A P E ---~-~

H R D~

D F G~

I

a

~

A P E ~..

{101)

E Residue

76 77

182

288

c

) 275

CDC7

RANRA

281 G

R 6 F RAP

295 EV LMK

C GA

M utotions A

Fig. 1A-C. Sites of in vitro mutagenesis in the CDC7 protein kinase. A The organisation of the CDC7 open reading frame within the 1.8 kb DraI fragment which was used as a template for mutagenesis after subcloning into M13mp18amIV. B Amino acid sequence of part of the CDC7 catalytic domain compared with a consensus for these regions derived from Hanks et al. (1988). Amino acids are shown according to the standard IUPAC single-letter code; hyphens indicate single, non-conserved residues while the symbol - indicates gaps of variable size between the conserved domains. The size of the gap is shown in parentheses for that interval in which CDC7 differs significantly from the consensus. Asterisks above the consensus sequence designate residues that are universally conserved within the protein kinase family. The locations and nature of the mutations introduced into CDC7 are shown below the CDC7 sequence. C The sequence of the CDC7 protein around the putative phosphorylation site, threonine-281, and the substitutions introduced at this site

ing region is a domain of about 300 amino acids which corresponds to the conserved catalytic domain of protein kinases, conventionally regarded as comprising an ATPbinding site and a phosphoacceptor site (Fig. 1 B). The ATP-binding domain contains a highly conserved glycine-elbow characteristic of nucleotide-binding sites (Wierenga and Hol 1983) as well as a universally conserved lysine residue (lysine-76 in CDCT) whose modification has been shown to lead to a loss of enzymic activity in a number of other protein kinases (Zoller et al. 1981; Snyder et al. 1985; Hannink and Donoghue 1985; Weinmaster etal. 1986). To examine the effects of mutation at this site in CDC7, we constructed the alleles cdc7.Met76 and cdc7.Glu76. We also constructed an allele in which the neighbouring, non-conserved lysine residue was mutated (cdc7.Gln77). The phosphoacceptor region of the catalytic domain is structurally more complex than the ATP-binding domain and contains a number of residues that are universally conserved amongst protein kinases, including an aspartate in the motif aspartate-phenylalanine-glycine and a glutamate within the sequence alanine-proline-glutamate (Fig. 1 B), both of which are likely to be important for enzymic activity. We have independently mutated the aspartate and glutamate residues within these motifs in CDC7 to cdc7.Asn182 and cdc7.Gln288 respectively, to examine the effect of changes at these sites on the cellular functions of CDC7. Both of these motifs occur close to a potential site of phosphorylation that is present in several protein kinases and whose modification in many instances acts to regulate protein kinase activity (Booher and Beach 1986; Hanks et al. 1988). In the CDC7 protein, this site is equivalent to threonine281 (Patterson et al. 1986) which we have mutated to cdc7.Ser281 and cdc7.Ala281 (Fig. 1C). The effect of each mutation on the cellular function of CDC7 was studied by monitoring the ability of the mutated gene to allow mitotic division or ascospore formation at the restrictive temperature in cells carrying the chromosomal cdc7-1 allele.

Effects of catalytic mutations on CDC7 function in meiosis To determine whether CDC7 protein kinase activity is needed for biological function in cells undergoing meiosis, we recovered the various cdc7 alleles from M I 3 m p 1 8 R F into YCp50 and used these to transform S. cerevisiae SB311 to uracil prototrophy. Transformants were grown under selective conditions in presporulation medium at 23 ° C, then transferred to sporulation medium and incubated either at 23 ° or 32.5 ° C for 4 days when ascospore formation was scored. The results obtained are shown in Fig. 2, in which sporulation efficiency is the ratio of the extent of ascospore formation at 32.5 ° to that at 23 ° C. In contrast to the effect of mutating the non-conserved lysine-77, which does not influence the biological function of the CDC7 protein, the data show clearly that mutating conserved residues in either the ATP-binding region

455 100 o~

975 ¢. Ico

.o_ cO

50

I

5

~25 cO

i

I

I

o Q. )-

cn 121_

U-,

-.~

~

~

U-.

cn

CO

rn

en

en

Q_

ID_

n

Cl_

o_

Plosmid

Fig. 2. Functional analysis of edc7 alleles during meiosis. The ability of cells to form ascospores at 23° and 32.5° C was measured using Saccharomyces cerevisiae SB311 transformed with cdc7 alleles carried on single-copy plasmids. Sporulation efficiency is the ratio of ascospore formation at 32.5° C relative to that at 23° C. For each plasmid, the results are the average (_+SEM) for eight independent transformants

Cells carrying plasmids with mutations in the C D C 7 phosphoacceptor region (pVB4451 and pVB4461) grew normally at 23 ° C but failed to f o r m colonies or grow in liquid media at 3 7 ° C when they arrested with the terminal m o r p h o l o g y characteristic of the cdc7-1 mutation. Similar results were obtained with mutations altering the conserved lysine-76 residue, since cells carrying cdc7.Met76 or cdc7.Glu76 (pVB441 and pVB4871 respectively) also failed to grow at 37 ° C, whereas mutating the adjacent, non-conserved lysine-77 (pVB471) had no effect on C D C 7 function. The behaviour of cells carrying the cdc7.Met76 allele was remarkably dependent on the temperature of incubation and " l e a k y " growth was observed with this allele at temperatures below 36 ° C. Nonetheless, these results are consistent with the idea that the protein kinase activity of the C D C 7 protein is needed for biological function during mitotic cell division.

E f f e c t o f mutations directed to a potential phosphorylation site

To investigate the requirements for C D C 7 function in mitosis, we examined the effect of the various alleles on the growth of S. cerevisiae SB311 and the haploid strain SB433 at 23 ° and 37 ° C. No significant straindependent variation was observed and the results with S. cerevisiae SB311 are summarised in Table 3.

In addition to the effect of mutations targeted to residues likely to be required directly for the catalytic action of the C D C 7 protein kinase, we have also examined the consequences of mutation at a potential site of phosphorylation in the C D C 7 protein. This residue in the C D C 7 gene product, threonine-281, is equivalent to threonine-167 in the cdc2 + protein kinase of Sehizosaccharom y c e s p o m b e and threonine-161 in the chick cdc2 + homologue (Patterson et al. 1986). In the fission yeast cdc2 + protein, threonine-167 is one of only two phosphorylated residues (K. Gould, personal communication) while modification of the equivalent residue in the chick homologue is cell-cycle dependent (Krek and Nigg 1991). We have mutated this residue in C D C 7 to form the alleles cdc7.Ser281 and cdc7.Ala281, and have examined the effect of each of these mutations on C D C 7 function in mitosis and meiosis. The results of this analysis are shown in Table 4. When the conservative substitution of serine is made for threonine-281 in C D C 7 , the resulting protein retains apparently normal biological activity in both mitosis and meiosis. In contrast, substituting alanine for threonine

Table 3. Properties of cdc7 alleles during mitosis

Table 4. Effects of mutation at threonine-281 on CDC7 function

cdc7 allele

Plasmid

(pVB441 and pVB4871) or the phosphoacceptor region (pVB4451 and pVB4461) results in a loss of C D C 7 biological actiyity during meiosis. This result clearly suggests that the C D C 7 protein kinase activity is essential for meiosis. Moreover, the effects of all of the constructed cdc7 alleles are apparently recessive since 5 0 % 65% of all cells formed ascospores at 23 ° C in 4 days regardless of the cdc7 allele present on the plasmid.

E f f e c t s o f catalytic mutations on C D C 7 f u n c t i o n in mitosis

None CDC7 + cdc7.Met76 cdc7.Gln77 cdc7.Glu76 cdc7.Asn182 cdc7. Gln288

Plasmid

YCp50 pVB4 pVB441 pVB471 pVB4871 pVB4451 pVB4461

a +, growth; - , no growth

Growth at a 23° C

37° C

+ + + + + + +

+ + -

Allele

Growth at a 23° C

YCp50 pVB4 pVB4951 pVB4961

Ascospore formation at b

37° C

None

+

-

CDC7 + cdc7.Ser281 cdc7.Ala281

+ + +

+ + -

23° C

33° C

60 % 62% 50% 63%

_<1% 46% 48% _<1%

_{_ growth; --, no growth b Ascospores were scored after incubating at the designated temperature for 4 days. The values given are the average of two experiments using two independent transformants a

456 at this position inactivates CDC7 function in both pathways. This suggests that, like the edc2 + protein kinase (K. Gould, personal communication), the presence of a phosphorylatable residue at position 281 in the CDC7 gene product is necessary for biological activity in mitosis and meiosis, and thus that phosphorylation of the CDC7 protein at this residue may be an important mechanism by which protein kinase activity is regulated.

Discussion

The CDC7 protein has significant primary sequence homology with a number of protein kinases and contains all of the conserved residues that have been shown to be essential for the function of vertebrate and fungal protein kinases (Hanks et al. 1988). We have constructed mutations at some of these conserved residues within the CDC7 gene product and have determined the ability of the different cdc7 mutant alleles to function in mitosis and meiosis. One highly conserved region within the protein kinase catalytic domain surrounds a site at which many of the protein kinases are themselves phosphorylated; included within this region are the two triplets alanine-proline-glutamate and aspartate-phenylalanineglycine (Hanks et al. 1988). Mutagenesis studies have shown that each residue in the alanine-proline-glutamate motif is needed for protein kinase and biological activity in p60 ..... , while the aspartate-phenylalanine-glycine motif constitutes the most highly conserved component of the protein kinase catalytic domain (Bryant and Parsons 1984; Hanks et al. 1988). We have constructed mutations in both of these motifs to form cdc7.Gln288, which corresponds to mutation at the essential glutamate-432 in p60 ..... , and cdc7.Asn182. Both of these mutations result in a loss of biological activity in mitosis and meiosis even when overexpressed (data not shown), suggesting that CDC7 protein kinase activity is needed for both of these cellular functions. The other conserved domain of protein kinases is an ATP-binding site which contains an invariant lysine corresponding to lysine-72 in bovine cAMP-dependent protein kinase and lysine-295 in p60 v-~rc. Covalent modification of this residue by the ATP analogue p-fluorosulphonyl 5'benzoyladenosine or alteration by mutation leads to loss of enzymic activity and biologial function in a number of protein kinases including p60 v.... and Sz. pombe cdc2 + (Kamps et al. 1984; Booher and Beach 1986). In CDC7, this invariant residue corresponds to lysine-76, which we have mutated to form the alleles cdc7.Met76 and cdc7.Glu76. Although the behaviour of the cdc7.Met76 allele is critically dependant on temperature, neither of the constructed alleles rescues the chromosomal cdc7-1 mutation during mitosis or meiosis at restrictive temperatures, again consistent with the notion that CDC7 protein kinase activity is necessary for its function in both mitosis and meiosis. In addition to the inactivation of the CDC7 protein kinase activity by mutations directed to functional residues within the catalytic domain, we have also examined the effect of mutations at a site within the CDC7 protein

which, by comparison with the behaviour of other protein kinases, is a possible phosphoacceptor residue. Mutating threonine-281 to serine, where the potential for phosphorylation is maintained, has no apparent effect on the function of the CDC7 protein in either mitosis or meiosis. However, when threonine-281 is replaced by alanine, when presumably the potential for modification at this site is lost, the mutated protein is unable to function in either the mitotic or meiotic pathways. Similar results have been obtained when serine and alanine are substituted for threonine-167 in p34 cac2 of Sz. pombe, with the former substitution resulting in phosphoserine rather than phosphothreonine as the principal phosphorylated residue in p34 cat2 (K. Gould, personal communication). Our results thus support the idea that at least part of the mechanism by which the activity of the CDC7 protein is regulated involves phosphorylation at threonine-281. This is particularly interesting in view of the location of threonine-281, immediately C-terminal to a large insertion in the catalytic domain which is both unique to CDC7 and essential for its function (Bahman et al. 1988). Differential phosphorylation of threonine281 might thus serve to regulate CDC7 activity by modulating specific interactions between this region of the CDC7 protein kinase and other proteins. Experiments to test this hypothesis are in progress. The demonstration that CDC7 protein kinase activity is required for mitotic division suggests that progress through G1 and ultimately the transition to S phase are critically dependent on protein phosphorylation. This is supported by the demonstration of phosphoproreins as components 'of the polymerase-primase (Lucchini et al. 1987) and replicative complexes (Jazwinski 1988). Perhaps more significant though is the observation that one of the subunits of the initiation factor RF-A is phosphorylated in a cell-cycle dependent manner, with maximal phosphorylation occurring at the G1S phase boundary (Brill and Stillman 1989). It is clearly important to determine which, if any, of these processes require the CDC7 protein kinase activity. In addition to the differential expression of the CDC7 gene during mitosis and meiosis (Sclafani et al. 1988), there is also an apparent difference in the physiological requirement for CDC7 function in the two pathways (Hartwell 1973; Simchen 1974). Nonetheless, the results that we have presented here clearly support the idea that the protein kinase activity of the CDC7 gene product is necessary for CDC7 function in both mitosis and meiosis. It may well be that the differences between CDC7 function in mitosis and meiosis are reflected in different functional interactions in the two pathways, involving in each case specific regions of the CDC7 protein outside the catalytic domain, a possibility that is currently being investigated.

Acknowledgements. We are particularly grateful to Paul Sims for

help with the mutagenesis; to Kathy Gould and Eric Nigg for communicating results prior to publication; and to the rest of our group for their help and advice. We gratefully acknowledge the financial support of the CRC and SERC.

457 References

Bahman AM, Buck V, White A, Rosamond J (1988) Characterisation of the CDC7 product of Saccharomyces cerevisiae as a protein kinase needed for the initiation of DNA synthesis. Biochim Biophys Acta 951 : 335-343 Booher R, Beach D (1986) Site-specific mutagenesis of cdc2 ÷, a cell cycle control gene of the fission yeast Schizosaccharornyces pombe. Mol Cell Biol 6:3523-3530 Brill SJ, Stillman B (1989) Yeast replication factor-A functions in the unwinding of the SV40 origin of DNA replication. Nature 342: 92-95 Bryant BL, Parsons JT (1984) Amino acid alterations within a highly conserved region of the Rous sarcoma virus sre gene product pp60src inactivate tyrosine protein kinase activity. Mol Cell Biol 4:862-866 Carter P, Bedouelle H, Winter G (1985) Improved oligonucleotideinduced mutagenesis using M13 vectors. Nucleic Acids Res 13 : 4431-4443 Ham J, Moore D, Rosamond J, Johnston IR (1989) Transcriptional analysis of the CDC7 protein kinase gene of Saccharornyces cerevisiae. Nucleic Acids Res 17:5781-5792 Hanks SK, Quinn AM, Hunter T (1988) The protein kinase family: conserved features and deduced phylogeny of the catalytic domain. Science 241:42-52 Hannink M, Donoghue DJ (1985) Lysine residue 121 in the proposed ATP-binding site of the v-mos protein is required for transformation. Proc Natl Acad Sci USA 82:7894-7898 Hartwell LH (1973) Three additional genes required for deoxyribonucleic acid synthesis in Saccharornyces cerevisiae. J Bacteriol 115 : 966-974 Hereford LM, Hartwell LH (1974) Sequential gene function in the initiation of S. cerevisiae DNA synthesis. J Mol Biol 84: 445-461 Hollingsworth RE, Sclafani RA (1990) DNA metabolism gene CDC7 from yeast encodes a serine (threonine) protein kinase. Proc Natl Acad Sci USA 87:6272-6276 Humphreys GO, Willshaw GA, Anderson ES (1975) A simple method for the preparation of large quantities of pure plasmid DNA. Biochim Biophys Acta 383:457-463 Jazwinski SM (1988) CDC7-dependent protein kinase activity in yeast replicative-complex preparations. Proc Natl Acad Sci USA 85:2101-2105 Kamps MP, Taylor SS, Sefton BM (1984) Direct evidence that oncogenic tyrosine kinases and cyclic AMP-dependent protein kinases have homologous ATP binding sites. Nature 310: 589592 Krek W, Nigg EA (1991) Differential phosphorylation of vertebrate p34cec2 kinase at the G1/S and G2/M transitions of the cell cycle. Identification of major phosphorylation sites. EMBO J 10:305-316 Lucchini G, Francesconi S, Foiani M, Badaracco G, Plevani P (1987) Yeast DNA polymerase-DNA primase complex: cloning of PRH, a single essential gene related to DNA primase activity. EMBO J 6 : 732742

Maxam AM, Gilbert W (1977) A new method for sequencing DNA. Proc Natl Acad Sci USA 74:560-564 Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York Njagi GDE, Kilbey BJ (1982) cdc7-1, a temperature-sensitive cellcycle mutant which interferes with induced mutagenesis in Saccharomyces eerevisiae. Mol Gen Genet 186:478 481 Patterson MN, Sclafani RA, Fangman WL, Rosamond J (1986) Molecular characterization of cell cycle gene CDC7 from Saccharomyces cerevisiae. Mol Cell Biol 6:1590-1598 Rose MD, Novick P, Thomas JH, Botstein D, Fink GR (1987) A Saccharomyces cerevisiae genomic plasmid bank based on a centromere-containingshuttle vector. Gene 60:237-243 Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74: 54635467 Schild D, Byers B (1987) Meiotic effects of DNA defective cell division cycle mutations of Saccharomyces cerevisiae. Chromosoma 70 : 109-130 Sclafani RA, Patterson M, Rosamond J, Fangman WL (1988) Differential regulation of the yeast CDC7 gene during mitosis and meiosis. Mol Cell Biol 8 :293-300 Sherman F, Fink GR, Hicks J (1979) Methods in yeast genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York Simchen G (1974) Are mitotic functions required in meiosis? Genetics 76 : 745-753 Synder MA, Bishop JM, McGrath JP, Levinson AD (1985) A mutation at the ATP-binding site of pp60 v-s'~ abolishes kinase activity, transformation and tumorigenicity. Mol Cell Biol 5 : 1772-1779 Tschumper G, Carbon J (1983) Copy number control by a yeast centromere. Gene 23 : 221-232 Warren G, Sherratt D (1978) Incompatibility and transforming efficiency of ColE1 and related plasmids. Mol Gen Genet 161:39-47 Weinmaster G, Zoller MJ, Pawson T (1986) A lysine in the ATPbinding site of p130 gap'fps is essential for protein-tyrosine kinase activity. EMBO J 5:69-76 Wierenga RK, Hol WGJ (1983) Predicted nucleotide-bindingproperties of p21 protein and its cancer-associated variant. Nature 302: 842-844 Yanisch-Perron C, Vieira J, Messing J (1985) Improved M 13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33:103-119 Zoller MJ, Nelson NC, Taylor SS (1981) Affinity labelling of cAMP-dependent protein kinase with p-fluorosulfonylbenzoyl adenosine. J Biol Chem 256:1083%10842 Zoller MJ, Smith M (1984) Oligonucleotide directed mutagenesis: a simple method using two oligonucleotide primers and a singlestranded DNA template. DNA 3:479-488

C o m m u n i c a t e d by. B.J. K i l b e y