Mol Gen Genet (1992) 232:344-350

© Springer-Verlag 1992

Cold-sensitive mutants of p34 cdc2 that suppress a mitotic catastrophe phenotype in fission yeast K. Ayscough*, J. Hayles, S.A. MacNeill** and P. Nurse

ICRF Cell Cycle Group, Microbiology Unit, Department of Biochemistry, University of Oxford, South Parks Road, Oxford OX1 3QU, UK

Received October 10, 1991

Summary. The p34 ~a~2 protein kinase plays a central role in the regulation of the eukaryotic cell cycle, being re- quired both in late GI for the commitment to S-phase and in late G2 for the initiation of mitosis, p34 ~d¢2 also determines the precise timing of entry into mitosis in fission yeast, where a number of gene produts that regu- late p34 ca~2 activity have been identified and character- ised. To investigate further the mitotic role of p34 ~d°2 in this organism we have isolated new cold-sensitive p34 °d°2 mutants. These are defective only in their G2 function and are extragenic suppressors of the lethal pre- mature entry into mitosis brought about by mutating the mitotic inhibitor p107 w°° ~ and overproducing the mi- totic activator pS0 °ac25. One of the mutant proteins p34 °a~z-Es is only functional in the absence of p107 we~, and all the mutant strains have reduced histone H1 ki- nase activity in vitro. Each mutant allele has been cloned and sequenced, and the lesions responsible for the cold- sensitive phenotypes identified. All the mutations were found to map to regions that are conserved between the fission yeast p34 ~a~2 and functional homologues from higher eukaryotes.

Key words: p34 ~a~2 - Extragenic suppressors - Cell cycle Mitotic control - Protein kinases

Introduction

Several of the gene functions involved in the regulation of entry into mitosis have been functionally conserved through evolution from yeast to humans (reviewed in Nurse 1990). Genetic analysis of the fission yeast Schizo- s a c c h a r o m y c e s p o m b e has shown that a key element in

* Present address: Imperial Cancer Research Fund, P.O. Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK ** Present address: Institute of Cell and Molecular Biology, Uni- versity of Edinburgh, Mayfield Road, Edinburgh EH9 3JR, UK

Offprint requests to: J. Hayles

this control is the protein serine-threonine kinase p34 cat2, and functional homologues of p34 cac2 have now been identified in a range of higher eukaryotic cell types (Nurse 1990). p34 ~dc2 function is required twice within the cell cycle, both in late G1 at Start for commitment to S-phase and in late G2 for the initiation of mitosis (Nurse et al. 1976; Nurse and Bissett 1981). At the G2/M boundary, p34 ~d~/is complexed with the mitotic B-type cyclin p56 °a~13 (reviewed in MacNeill et al. 1991 a). Acti- vation of the p34ca~Z-p56~a¢13 complex, and subsequent entry into mitosis, requires dephosphorylation of p34 cd~2 on tyrosine 15, a residue located in the ATP binding site of the p34 ~ac/ protein kinase (Gould and Nurse 1989). A similar mechanism is likely to be important for p34 ~d~/activation in vertebrate cells (Norbury et al. 1991).

In addition to p56 ~a~3, several gene products have been identified that interact with p34 ¢dc2 at the G2/M transition in fission yeast. Two of these, p107 weea and pS0 Cae/s, act antagonistically to regulate the timing of p34 ~a~z activation and entry into mitosis, pS0 cde25 and p107 w~¢~ function as an activator and an inhibitor of mitosis respectively. Overproduction of p107 w~ or loss of p80 cde25 activity results in a failure to enter mitosis, whereas loss of p107 we~l activity or overproduction of p80 ~a~25 causes cells to be advanced into mitosis at a small cell size (Nurse 1975; Nurse and Thuriaux 1980; Russell and Nurse 1986). Furthermore, if p80 ~a~2s is overproduced in cells lacking p107 w¢¢1 activity, then the effect is additive and the cells enter mitosis at a very small cell size and undergo mitotic catastrophe (mc) as a result of entering mitosis in the absence of complete DNA replication (Russell and Nurse 1986; Enoch and Nurse 1990). The biochemical functions of the p107 *~el and p80 cac25 proteins have recently been the subject of much analysis, p107 wee~ is a protein kinase that has ser- ine-tyrosine kinase activity against exogenous substrates in vitro (Featherstone and Russell 1991), and may be responsible for tyrosine phosphorylation of p34 °a¢2 on tyrosine 15 in vivo, although this remains to be proven. In contrast, p80 ~d°2s is a putative threonine-tyrosine

phosphoprotein phosphatase (Moreno and Nurse 1991), and is thought likely to be responsible for tyrosine de- phosphorlyation of p34 ca°2 on tyrosine 15 (Kumagai and Dunphy 1991 ; Strausfeld et al. 1991). p34 °a°/activation is determined at least in part by the balance of p107 weea and pS0 °a°25 activities, p34 °a°2 is also subject to negative regulation by the protein kinase encoded by the mik l + gene, the function of which overlaps with p107 we°l (Lundgren et al. 1991).

The strain adh-cdc25 + weel-50 leul-32 ura4-294 con- stitutively overexpresses the cdc25 + gene and contains a temperature-sensitive p107 w°~l-s° protein (Russell and Nurse 1986). These cells are viable at 25 ° C, the permis- sive temperature for p107 w~el-5°, but when shifted to 35 ° C, p107 w°°~-5° becomes inactive and the cells under- go mitotic catastrophe. We have isolated suppressing mutations in this strain that allow growth at 35 ° C but which are also cold-sensitive (cs), that is, which are now incapable of growth at 20 ° C. Cs mutations are frequent- ly found in proteins that are involved in complex forma- tion, perhaps because at low temperature they lack the molecular flexibility required to form complexes with other proteins (discussed in Hartmann and Roth 1973). We concentrated specifically on identifying new mutant p34 °a°z proteins, reasoning that owing to the nature of our screen, these mutants were more likely to be defec- tive specifically at the G2/M transition, rather than at both G1/S and G2/M. As mentioned above, almost all the previously isolated p34 ¢ac2 mutants are defective at both points in the cycle (Nurse and Bissett 1981), so there is considerable interest in obtaining mutants that are defective at one point or the other. Such mutant proteins might, for example, be defective in their interac- tion with one of the well-characterised elements involved in the mitotic control in fission yeast such as p56 °d°t3 (Booher and Beach 1987; Hagan etal. 1988; Moreno et al. 1989), or perhaps with one or several (unknown) mitotic substrates, and yet still be perfectly functional at G1/S. Here we describe the isolation of 13 cold-sensi- tive mitotic catastrophe suppressors (mcs), five of which are the result of mutation of the p34 °d°2 protein kinase. We have characterised these mutants and show that, as predicted, they are defective solely for their G2/M func- tion and not for the passage of start. We have also cloned and sequenced all five mutants alleles with a view to identifying regions within the p34 °a°2 protein that may be important for its interaction with other proteins at mitosis (MacNeill et al. 1991 b).

Materials and methods

Yeast strains and media. All the strains used in this study are derived from 972h- and 975h + (Leupold 1970). The genetic nomenclature used throughout is that of Kohli (1987). The adh-edc25 + weel-50 leul-32 ura4-294 strain contains an integrated copy of pUSX179 (Russell and Nurse 1987). This plasmid contains the cdc25 + gene under the control of the fission yeast alcohol dehydroge- nase (adh) promoter resulting in high-level expression

345

(Moreno et al. 1990). Yeast extract, minimal, minimal low glucose and minimal minus nitrogen media are de- scribed elsewhere (Moreno et al. 1991). Histidine, leucine and uracil supplements were added as required at 250 mg/1. Phloxin B (Sigma) at 10 rag/1 was used during the suppressor screen to distinguish between colonies containing viable growing cells (which were pale pink) and colonies containing cells undergoing residual divi- sions with mitotic catastrophe (which stained dark red).

Mutagenesis. Cells were grown to 5 x 106 cells/ml in min- imal medium plus supplements at 25 ° C. After washing in 0.2 M sodium acetate buffer, pH 5.0, cells were resus- pended at 108 cells/ml in 0.2 M sodium acetate buffer, pH 5.0, with 2 mg/ml N-methyl-N-nitro-N-nitrosoguan- idine and incubated at the permissive temperature for 30 min to give 50% survival. The cells were then washed 3 times in 0.9% saline and resuspended at 5 x 106 cells/ ml in yeast extract medium at 25°C for 3 h. A total of 107 cells were then plated out at 105 cells/plate on yeast extract medium containing phloxin B and incubat- ed at 35 ° C. The plates were screened for suppressors after 3 days.

Yeast genetics, f low cytometry and transition point deter- mination. Random spore analysis, tetrad analysis and yeast transformation using the protoplast method were all carried out as described by Moreno et al. (1991). Flow cytometric analysis of fixed, propidium iodide- stained fission yeast cells was carried out according to the protocol of Sazer and Sherwood (1990) using a Bec- ton-Dickinson FACScan. For determination of transi- tion points, exponentially growing cultures were shifted to the restrictive temperature, sampled and processed for cell number counts using a Coulter counter. The transition points were calculated as described in Nurse et al. (1976).

Histone H I kinase assays. The histone H1 kinase activity of the mutant p34 cac2 proteins was determined in vitro by the method of Moreno et al. (1989).

Cloning and sequence analysis o f mutant alleles. The five cold-sensitive mutant alleles cdc2-A21, cdc2-B14, cdc2- D20, cdc2-E8 and cdc2-E9 were cloned and sequenced exactly as described in MacNeill et al. (1991 b).

Results

Isolation o f extragenic suppressors o f adh-cdc25 + wee1-50 leul-32 ura4-294

The adh-cdc25 + weel-50 leul-32 ura4-294 strain was mu- tagenised (see the Materials and methods) and 130 inde- pendent colonies were selected that could grow at 35 ° C and were pale pink on plates containing the dye phloxin B. Of these 130 strains, 13 were unable to form colonies at 20 ° C, 5 of these showing a classical cell division cycle (cdc) arrest. The relative abilities of these strains to form colonies at various temperatures are indicated in Table

346

Table 1. The ability of the cold sensitive mitotic catastrophe sup- pressors to form colonies at different temperatures

Strain 20 ° C 25 ° C 35 ° C Notes

A1 - + + + + + + A6 + + + + Sterile AI0 - + + + + + + A21 - + + + cdc- at 20 ° C B I 1 - + + + +

B14 - + + + + + cdc- at 20 ° C B15 -- + + + + + C1 - + + + + + C9 - + + + + D20 -- + + + cdc at 20 °C E8 -- + + + cdc- at 20 ° C E9 - + + + cdc- at 20 ° C E20 - + + + cdc- at 20 ° C, sterile

The ability to form colonies is given in relation to wild type (+ + +) and adh cdc25 ÷, weel-50 ( + / - ) which undergoes mitotic catas- trophe with some residual division at 35 ° C

1. In order to test whether the observed cold-sensitivity was linked to the mcs phenotype, each adh-cd¢25 + weel- 50 leul-32 ura4-294 suppressor strain was crossed to adh- edc25 + weet-50 his3-27 leul-32 ura4-294 and the meiotic products analysed by r andom spore analysis (see the Materials and methods). I f the two phenotypes (cold- sensitivity and mitotic catastrophe suppression) are linked, then all the suppressed colonies will also be cold- sensitive. In addition, in order to investigate whether any of the suppressors lay within the cdc2 gene, we made use of the fact that the his3 gene lies within I cM of the cdc2 gene and so can be used as a marker for the cde2 locus (Kohli et al. 1977). Two of the 13 mutants (A6 and E20) were sterile and therefore not analysed further. For the 11 other cs suppressors, about 100 spores were analysed f rom each cross. The results showed, that in all cases, the mitotic catastrophe sup- pression (mcs) phenotype and cs phenotypes were linked to within at least 2 cM. In five of the strains (A21, B14, D20, E8 and E9), the cs/mcs phenotype was also closely linked to his3, demonstrat ing that these suppressor mu- tations are likely to lie within or close to the cde2 gene. All of these strains were cdc- at 20°C (Table 1). To ascertain whether the remaining six suppressors lay with- in the same or different genes, all six were crossed to one another. In all crosses, a mcs phenotype was generat- ed indicating that the suppressor mutat ions lay within different complementat ion groups. This finding suggests that mutat ions in a large number of other genes can suppress this phenotype and as a consequence no further analysis of these mutants has been carried out.

Analysis o f strains carrying cs mutations linked to cdc2

To investigate the effect o f the suppressor muta t ion when crossed away f rom weel-50 and adh-cdc25 +, each of the five suppressor strains (cdc2 c~ adh-cdc25 weel-50 leul-32 ura4-294) was crossed to a his3-27 ura4-D18 leul-32 strain and the products analysed by r andom spore analy-

sis. Several colonies were selected as presumptive triple mutants leul-32 ura4-D18, together with one of the edc2 cs alleles A21, B14, D20, E8 or E9. These strains were backcrossed to a his3-27 ura4-D18 leul-32 strain and analysed by r andom spore analysis to determine their genotype. The edc2, wee! and ede25 genes are not linked and will recombine freely. Three of the suppres- sors cdc2-A21 leul-32 ura4-D18, cdc2-B14 leul-32 ura4- D18 and cdc2-D20 leul-32 ura4-D18 were found to be in the required weel + cdc25 + background; they all re- tained the cs cdc - phenotype. The presumptive cdc2-E8 leul-32 ura 4-D18 still retained the weel-50 mutation, suggesting that it could not form colonies in the presence of wee1 + and cdc2-E9 could not be analysed because of the high level of diploids generated in the cross. The results of the backcrosses of cdc2 cs alleles A21, B14, D20 and cdc2-E8 weel-50 to his3-27 ura4-D18 leul-32 were confirmed by tetrad analysis. For the crosses involving cdc2-A21, cdc2-B14 and cdc2-D20, only parental ditypes were generated, showing that there was no weel-50 or adh-cdc25 + present. For the cdc2-E8 wee 1-50 strain, one product in each of the 13 tetratypes obtained was cdc- at both 20°C and 35 ° C. This spore, which was deduced to carry the edc2-E8 mutat ion alone, germi- nated to form a highly elongated cell. We conclude that cdc2-E8 only remains viable in the absence of the wild- type p107 weet protein and that its apparent cold-sensitiv- ity is in fact due to the temperature-sensitive (ts) pheno- type of p107 weel-5°.

Cloning and sequence analysis of the mutant alleles

Each of the five cold-sensitive mutant alleles cdc2-A21, cdc2-B14, cdc2-D20, cde2-E8 and cdc2-E9 was cloned and sequenced, and, in each case, a single nucleotide deviation f rom the wild-type D N A sequence, sufficient to alter the sequence of the encoded (mutant) protein, was detected (see Fig. 1 and the Materials and methods). Three of the mutants cdc2-A21, edc2-B14 and cdc2-D20 were the result of the same nucleotide change at amino acid 90 (nucleotide 549) (Hindley and Phear 1984); the muta t ion is subsequently referred to as edc2-A21. The cdc2-E8 and cdc2-E9 mutat ions were the result of changes at amino acids 213 and 242 (nucleotides 1019 and 1106) respectively. The observed mutat ions were all G to A changes, and each altered an aspartate residue in the wild-type protein to an asparagine. The locations of the mutat ions within the p34 °a¢2 protein are discussed below (see the Discussion).

p34 ca°z kinase activity of cold-sensitive mutants

Kinase activity of the strains carrying isoalleles p34 edc2-n21 and p34 cac2-°2° and of p3& de2"~8 was as- sayed using in vitro extracts with histone H1 as the sub- strate (Moreno et al. 1989). The wild-type strain showed high kinase activity where as mutan t strains carrying p34 cd°2-Azl, p34 Ca~z-D2° and p34 ~ac2-r8 showed low H1 kinase activity when grown at the permissive tempera-

A

B

210 45

42 146 212

N22 l w / 2 w 183 - - 1 7 / 2 2

Ty~ 3 6 7 W ~ ~ Thi6@1 3 O--~ i ~

A21/B 1 4 /D20 I 177 90 1 M63

227

cdc2-A21/B 14/D20 I

90

79 80 N 100 t07

K L~---~V~--~F~D M~-~K Y M D R I SE T G AT S L D

V

C

cdc2-1 7/22 1 ~- cdc2-L 7 cdc2-E8 cdc2-45 -] t

208 210 212 213

191 200 S L S N 210

• O0 tO

IX

[ - - cdc2-E9 242

228 230 240 N 250 253 L

X Fig. l A-D. Location of cdc2 c" mutations. A Positions of all the mapped mutant alleles of cdc2. Cs mutants isolated in this study are boxed; the sites of phosphorylation on tyrosine 15 and threo- nine 167 are indicated. B--D Region of p34 cd°/containing the amino acid change of p34 cacz-Azl/m4/D2° p34 ~a°-zE8 and p34 cdcz-E9 respec- tively. Boxed residues are conserved between all functional homo- logues. Residues found in all protein kinases that are invariant or nearly so (e) and those that are well conserved (©) are indicated, as are protein kinase subdomains V, IX and X (Hanks et al. 1988). Sp, Sehizosaceharomyces pombe

ture (29 ° C), regardless of whether the tempera ture of the in vitro assay was 29 ° C or 20 ° C (Fig. 2 A and C). There was no further reduct ion in the kinase activity after i ncuba t ion of the cells at the restrictive tempera ture (20 ° C) for 5 h (see Fig. 2 B and D).

FA C S analysis o f arrested cells

Almos t all the previously identif ied temperature-sensi- tive p34 cat2 m u t a n t prote ins are defective for bo th their G1/S and G 2 / M funct ions (Nurse and Bissett 1981), al- though two temperature-sensi t ive p34 coo2 mu tan t s defec- tive only at G 2 / M have recently been reported (MacNeil l

A B 1 2 3 4 5 1 2 3 4 5

347

45kD

32kD

C D 1 2 3 4 5 1 2 3 4 5 ~ ~ . 45kD

32kD

Fig. 2A-D. Histone H1 kinase activity of cdc2 c+ mutants. A, B Strains grown at 29°C and assayed at 29°C after A 0 h and B 5 h at 29 ° C. C, D Strains grown at 29°C shifted to 20°C and assayed at 20°C after C 0 h and D 5 h at 20 ° C. Lane 1, wild type; lane 2, weel-50; lane 3, cdc2-A21; lane 4, cdc2-D20; lane 5, cdc2-E8 weel-50

PT RT

wt

cdc2-33

• • - I ~ cdc2-A21

I L . . . . i00 iG0:2;~o . . . . . . . . . . so +'O +oo +so 2oo

Fig. 3. Flow cytometry of cs cde2 mutants. Cells were analysed at the restrictive temperature (RT) after incubation at 20°C for 14 h for cde2-A21 and cde2-E8 weel-50 and after 5 h at 36°C for ede2-33. Cells analysed at the permissive temperature (PT) were grown to mid-log phase at 29 ° C. All strains growing exponentially at 29°C had a 2C DNA content. Upon shift to the restrictive temperature edc2-A21 and edc2-E8 weel-50 became arrested with a 2C DNA content only; cde2-33 arrested with both a 1C and 2C DNA content, indicating that it is defective at both the G1 and G2 control points. In this strain grown at 36 ° C there is in- creased staining of both G1 and G2 cells due to increased amounts of mitochondrial DNA in these elongated cells. The scale is orbi- tary and indicates DNA content

et al. 1991 b). F low cytometry was used in order to deter- mine where, wi thin the cell cycle, cells carrying the cdc2- A21 and cdc2-E8 wee l -50 muta t ions were arrested. These two cs m u t a n t s were grown in yeast extract medi- u m at 29 ° C to mid- log phase, shifted to the restrictive

348

temperature of 20°C and allowed to proceed through the cycle to their block point. Samples were taken at various time points and processed for flow cytometry (see the Materials and methods). The results obtained are shown in Fig. 3. Both mutants became blocked with a 2C DNA content, that is, in G2 only, indicating that these mutant proteins are not defective in the GI/S func- tion.

We also determined the transition point of the cdc2- E8 weeI-50 and cdc2-A21 mutants (see the Materials and methods). The transition point of a conditional cell cycle mutant indicates the execution point of the en- coded protein after which that particular mutant func- tion is no longer required for cell cycle progression. The transition point of cdc2-E8 was found to be 0.7, close to the onset of mitosis and similar to the transition points for other late G2 mutants (Nurse et al. 1976). The transition point of cdc2-A2I indicated that it under- went a complete cell cycle before becoming arrested.

Overproduction of p l3 ~"cl and p56Cdc13 fails to rescue cdc2- A21 or cdc2-E8

One possible explanation for the cold-sensitive pheno- type of p34 °a¢2-A2t protein is that its ability to bind to either pl 3 ~°a and p56 °a°~a may be reduced. We therefore tested whether overproduction of p13 ~°1 or p56 cdc13 could rescue the cs defect of p34 ~a°2-A21. If the mutant protein does have reduced ability to bind to either p13 s~°a or p56 ~a~3 at 20 ° C, then their overproduction might compensate for this and rescue the cold-sensitive defect. A cdc2-A21 leuI-32 strain was therefore trans- formed with plasmids carrying either the sucl ÷ or the cdci3 ÷ gene. The cdc2 ÷ gene was also transformed as a control. Transformant colonies obtained at the permis- sive temperature of 29 ° C, were then tested for growth at the restrictive temperature of 20 ° C. Only the cdc2 ÷ transformants grew at 20 ° C. These results suggest that p34 ~ac2-Aza does not have a reduced ability to bind to p13 ~u¢1 or p56 ¢a¢13 which can be rescued by overproduc- tion of these gene products, and also indicates that the cdc2-A21 allele is recessive to wild-type when the latter is in excess. We also tested whether cells carrying the cdc2-E8 allele could be rescued by overproduction of p34 ~dc2, p13 su~ or p56 ~a~3. As before, only transfor- mants overproducing p34 ~ac2 were able to grow at 20 ° C, suggesting that there is no reduction in the ability of the p34 ~d¢2-Es protein to bind either p13 Su°a or p56 ~d°a 3.

Discussion

We have identified extragenic suppressors of the mitotic catastrophe phenotype of the adh-cdc25 ÷ weel-50 leui- 32 ura4-294 strain, which can grow at 35°C but are unable to grow at 20 ° C. We have found that at least six different genes can rescue this phenotype. It may be that many genes can be mutated to cause a non- specific delay of entry into mitosis and thus suppress a mitotic catastrophe. We have therefore confined our

studies to the cdc2 gene whose function is required for entry in to mitosis. Our work parallels that of Booher and Beach (1987) who used a cdc2-3w weel-50 mitotic catastrophe strain to identify both intragenic and extra- genic mitotic catastrophe suppressing mutations (Booher and Beach 1987; Molz et al. 1989), including the cold-sensitive cdc2-59 allele. We do not know if the mutation in p34 °e°2-59 is the same as any of those seen in the protein characterised in this study.

Five new, independently isolated alleles of cdc2 were identified, although sequence analysis revealed that three of these were isoallelic. Each of these mutant proteins was defective only in its G2/M function and each strain showed reduced histone HI kinase activity, measured in vitro, at both 29 ° C and at 20 ° C. The reduction in kinase activity may delay entry into mitosis and thus suppress the mitotic catastrophe phenotype, perhaps by allowing more time for DNA replication to be complet- ed; it has previously been shown that attempting to enter mitosis in the absence of complete DNA replication is likely to be the cause of the lethal mitotic catastrophe phenotype (Enoch and Nurse 1990). The p34 °d°;-h21 and p34 cac2-Es proteins have very different properties, how- ever. The p34 cd°a-aal protein appears to be genuinely cold-sensitive at 20 ° C, whereas the cold-sensitivity of p34 ~a~2-~s is likely to be due to the fact that the p34 ~ac2-~s protein is hypersensitive to the action of the p107 wee1 protein kinase such that strains expressing the p34 ~a°2-Es protein are only viable in the absence of p107 we¢l activity. Alternatively p34 ~a°2-Es may have re- duced ability to respond to activation via the p80 .dc25 phosphatase (Moreno and Nurse 1991).

At the G2/M transition, p34 ~a~2 forms a complex with p56 °a~xa (cyclin B), which is essential for its activation. p34 ce~2 has also been shown to bind to p13 su°l in vivo (Brizuela et al. 1987) and although the functional signifi- cance of this interaction is not yet clear, overproduction of pl 3 su°~ is able to suppress the mutant defect of several temperature-sensitive mutant p34 ~a°2 proteins (Hayles et al. 1986a, b). However, we find that overproduction of either p56 ~a°la or p13 S"~a fails to suppress the cold- sensitive defect of cells carrying either p34 ~dc2-A21 or p34 ~d°2-Es. The mutation in the p34 °a~z-A2~ protein is an aspartate to asparagine change at amino acid 90 (Fig. 1). This residue lies some distance from the nearest temperature-sensitive mutations, at amino acids 43 and 137 and is within the well-conserved motif YL-FEFL- DLK found in all the p34 cd~2 functional homologue pro- teins identified to date and the Xenopus p34~a~2-related protein kinase Eg-1 (Lorincz and Reed 1984; Hindley and Phear 1984; Lee and Nurse 1987; Krek and Nigg 1989; Lehner and O'Farrell 1990; Jimenez et al. 1990; Hirt et al. 1991 ; Paris et al. 1991). Although this region is well-conserved in p34 ¢ac2 homologues, this is the first conditional lethal mutation to be found in this/egion. As mentioned earlier, the majority of temperature-sensi- tive mutations in p34 cd*2 are heat-sensitive. Thus this region may be more susceptible to cold-sensitive muta- tions and less susceptible to heat-sensitive mutations. It has been suggested that certain regions of proteins may behave like this, especially those involved in hydro-

349

phobic prote in-protein interactions (Moir et al. 1982). Interestingly the region sur rounding Asp 90 is relatively hydrophobic .

The cdc2-E8 muta t ion is also an aspar ta te to aspara- gine change, at amino acid 213 within the G D S E I D mo- tif (residues 212-217). This mo t i f is found in all p3& dc2 funct ional homologue proteins identified to date, and also in the p34CdC2-related prote in kinases Eg-1 and Dm- A (Lehner and O'Farre l l 1990). Previously three temper- ature-sensitive muta t ions have been identified in the vi- cinity o f this motif , including the G2-specific cdc2-17/22 muta t ion which muta tes glycine 212 to a serine (Mac- Neill et al. 1991 b). The G D S E I D region is highly nega- tively charged, suggesting that it m a y lie on the surface o f the p34 cd°2 prote in where it m a y interact with other proteins. The fact tha t the p34 cd°z-z8 mu tan t protein is either hypersensitive to inhibit ion by p107 we~ or defec- tive in its act ivat ion by p80 ca~25 m a y implicate G D S E I D in the interact ion between p34 ~d~z and one o f these pro- teins.

The third muta t ion identified in this s tudy was that in the p34 edcz-E9 protein. The behaviour o f strains carry- ing the cdc2-E9 allele could no t be analysed genetically because o f their tendency to diploidise at high frequency. Other alleles o f cdc2 have already been isolated, which can diploidise by re-replicating their D N A wi thout un- dergoing an intervening mitosis (Broek et al. 1991), and it m a y be that cdc2-E9 is similarly defective. The muta- t ion in the p3& a~zm9 protein maps to the C-terminal region o f p34 ~d~2 at posi t ion 242, an aspartate residue that is conserved in all p34 ~a~2 funct ional homologues but no t in other p34~a~2-related proteins. The extent o f diploidisation seen with different cdc2 alleles m a y reflect the instability o f the m u t a n t p34 ~d°a proteins that they encode (Broek et al. 1991). Thus the effect o f the muta - t ion in the p34 ~a~a-E9 protein m a y be to reduce its stabili- ty. Our results extend the number o f independent mu tan t p34 ~a*2 proteins to sixteen (Carr et al. 1989; Booher and Beach 1987; MacNeil l et al. 1991b, this p a p e r ) a n d for the first t ime add to this collection a putat ive p107 we~- sensitive protein, p34 cd~z-ES. Fur ther biochemical and genetical analysis o f these mutan t s should extend our unders tanding o f the regulat ion o f the eukaryot ic cell cycle and o f p34 °d°2 as a model protein kinase.

Acknowledgements. We would like to thank Chris Norbury and Tamar Enoch for reading the manuscript, Sergio Moreno for help with the kinase assays and our colleagues in the Cell Cycle Group for useful discussions. We thank the Imperial Cancer Research Fund and MRC for financial support.

References

Booher R, Beach D (1987) Interaction between cdcl3 + and cdc2 + in the control of mitosis in fission yeast; dissociation of the G1 and G2 roles of the cdc2 + protein kinase. EMBO J 6:3441- 3447

Brizuela L, Draetta G, Beach D (1987) p13 sue1 acts in the fission yeast cell division cycle as a component of the p34 °at2 protein kinase. EMBO J 6:3507-3514

Broek D, Bartlett R, Crawford K, Nurse P (1991) Involvement of p34 cd~2 in establishing the dependency of S phase on mitosis. Nature 349: 388-393

Carr AM, MacNeill SA, Hayles J, Nurse P (1989) Molecular clon- ing and sequence analysis of mutant alleles of the fission yeast cdc2 protein kinase gene: implications for ode2 + protein struc- ture and function. Mol Gen Genet 218:41-49

Enoch T, Nurse P (1990) Mutation of the fission yeast cell cycle control genes abolishes dependence of mitosis on DNA replica- tion. Cell 60:665-673

Featherstone C, Russell P (1991) Fission yeast p107 w°°l mitotic inhibitor is a protein tyrosine/serine kinase. Nature 349:808 811

Gould KL, Nurse P (1989) Tyrosine phosphorylation of the fission yeast cdc2 + protein kinase regulates entry into mitosis. Nature 343 : 3945

Hanks SK, Quinn AM, Hunter T (1988) The protein kinase family: conserved features and deduced phylogeny of the catalytic do- mains. Science 241:42-52

Hagan IH, Hayles J, Nurse P (1988) Cloning and sequencing of the cyclin related cdcl3 + gene and a cytological study of its role in fission yeast mitosis. J Cell Sci 91 : 587-595

Hartmann PE, Roth JR (1973) Mechanisms of suppression. Adv Genet 17:1-106

Hayles J, Beach D, Durkacz B, Nurse P (1986a) The fission yeast cell cycle control gene cdc2: Isolation of a sequence sucl that suppresses cdc2 mutant function. Mol Gen Genet 202:291-293

Hayles J, Aves S, Nurse P (1986b) sucl is an essential gene involved in both the cell cycle and growth in fission yeast. EMBO J 5:3373-3379

Hindley J, Phear G (1984) Sequence of the cell division gene CDC2 from Schizosaccharomyces pombe; patterns of splicing and ho- mology to protein kinases. Gene 31 : 129 134

Hirt H, Pay A, Gyorgyey J, Bako L, Nemeth K, Bogre L, Schweyen R, Heberle-Bors E, Dudits D (1991) Complementation of a fission yeast cell cycle mutant by an alfalfa cDNA encoding a protein kinase homologous to p34 cue2. Proc Natl Acad Sci USA 88:1636-1640

Jimenez J, Alphey L, Nurse P, Glover D (1990) Complementation of fission yeast cdc2 ~s and cdc25 ts mutants identifies two cell cycle control genes from Drosophila: a cdc2 homologue and string. EMBO J 11:3565 3571

Kohli J (1987) Genetic nomenclature and gene list of the fission yeast Schizosaccharomyces pombe. Curr Genet 11 : 575-589

Kohli J, Hottinger H, Munz P, Strauss A, Thuriaux P (1977) Genet- ic mapping in Schizosaccharomyces pombe by mitotic and meiotic analysis and induced haploidisation Genetics 87:471 489

Krek W, Nigg EA (1989) Structure and developmental expression of the chicken CDC2 kinase. EMBO J 8 : 3071-3077

Kumagai A, Dunphy WG (1991) The cdc25 protein controls tyro- sine dephosphorylation of the cdc2 protein in a cell free system. Cell 64:903-914

Lee M, Nurse P (1987) Complementation used to clone a human homologue of the fission yeast cell cycle control gene cdc2. Nature 327:31-35

Lehner CF, O'Farrell PH (1990) Drosophila cdc2 homologs: a func- tional homolog is coexpressed with a cognate variant. EMBO J 11:3573-3581

Leupold U (1970) Genetical methods for Schizosaccharomyces pombe. Methods Cell Physiol 4:16%177

Lorincz AT, Reed SI (1984) Primary structure homology between the product of the yeast cell division control gene CDC28 and vertebrate oncogenes. Nature 307:183 185

Lundgren K, Walworth N, Booher R, Dembski M, Kirschner M, Beach D (1991) mikl and wee1 cooperate in the inhibitory tyro- sine phosphorylation of cdc2. Cell 64: a 115-1122

MacNeill SA, Warbrick E, Fantes PA (1991a). Controlling cell cycle progress in the fission yeast Schizosaccharomyces pombe. Curt Op Genet Dev, in press

MacNeill SA, Creanor J, Nurse P (1991 b) Isolation, characterisa- tion and molecular cloning of new mutant alleles of the fission yeast p34 cdc2 protein kinase gene: Identification of temperature- sensitive Gz-arresting alleles. Mol Gen Genet 229:109-118

350

Moir D, Stewart SE, Osmond BC, Botstein D (1982) Cold sensitive cell division cycle mutants of yeast: isolation, properties and pseudoreversion studies. Genetics 100:547-563

Molz L, Booher R, Young P, Beach D (1989) cdc2 and the regula- tion of mitosis: six interacting mcs genes. Genetics 122:773-782

Moreno S, Nurse P (1991) Clues to action of cdc25 protein. Nature 351 : 194

Moreno S, Hayles J, Nurse P (1989) Regulation of p34 cdc2 protein kinase during mitosis. Cell 58: 361-372

Moreno S, Nurse P, Russell P (1990) Regulation of mitosis by cyclic accumulation of p80 cac25+ mitotic inducer in fission yeast. Nature 344 : 549-552

Moreno S, Klar A, Nurse P (1991) Molecular analysis of fission yeast Sehizosaecharornyces pombe. Methods Enzymol 194: 795- 823

Norbury C J, Blow JJ, Nurse P (1991) Regulatory phosphorylation of the p34 cdc2 protein kinase in vertebrates. EMBO J 10:3321- 3329

Nurse P (1975) Genetic control of cell size at cell division in yeast. Nature 256: 547-551

Nurse P (1990) Universal control mechanism regulating onset of M-phase. Nature 344: 503-508

Nurse P, Bissett Y (1981) Gene required in G1 for commitment to cell cycle and in Gz for control of mitosis in fission yeast. Nature 292: 558-560

Nurse P, Thuriaux P (1980) Regulatory genes controlling mitosis in the fission yeast Schizosaccharomyces pombe. Genetics 96: 627-637

Nurse P, Thuriaux P, Nasmyth KA (1976) Genetic control of the cell division cycle in the fission yeast Schizosaccharomyces pombe. Mol Gen Genet 146:167-178

Paris J, LeGuellic R, Couturier A, LeGuellic K, Omilli F, Camonis J, MacNeill SA, Philippe M (1991) Cloning by differential screening and characterisation of a Xenopus cDNA coding for a protein highly homologous to cdc2. Proc Natl Acad Sci USA 88 : 1039-1043

Russell P, Nurse P (1986) cdc25 ÷ functions as an inducer in the mitotic control of fission yeast. Cell 45:145-153

Russell P, Nurse P (1987) Negative regulation of mitosis by weei +, a gene encoding a protein kinase homolog. Cell 49:559-567

Sazer S, Sherwood SW (1990) Mitochondrial growth and DNA synthesis occur in the absence of nuclear DNA replication in fission yeast. J Cell Sci 97:509-516

Strausfield U, Labbe JC, Fesquet D, Cavadore JC, Picard A, Sadhu K, Russell P, Doree M (1991) Dephosphorylation and activa- tion of a p34cdcZ/cyclin B complex in vitro by human CDC25 protein. Nature 351:242-245

C o m m u n i c a t e d by B.J. Ki lbey