Molecular Biology of the Cell Vol. 3, 805-818, July 1992

Characterization of Four B-Type Cyclin Genes of the Budding Yeast Saccharomyces cerevisiae Ian Fitch,* Christian Dahmann,*t Uttam Surana,* Angelika Amon,* Kim Nasmyth,* Loretta Goetsch,§ Breck Byers,§ and Bruce Futcher* *Cold Spring Harbor Laboratory, Cold Spring Harbor, New York 11724; tResearch Institute of Molecular Pathology, 1030 Vienna, Austria; and §Department of Genetics, University of

Washington, Seattle, Washington 98195 Submitted March 23, 1992; Accepted June 3, 1992

The previously described CLB1 and CLB2 genes encode a closely related pair of B-type cyclins. Here we present the sequences of another related pair of B-type cyclin genes, which we term CLB3 and CLB4. Although CLB1 and CLB2 mRNAs rise in abundance at the time of nuclear division, CLB3 and CLB4 are turned on earlier, rising early in S phase and declining near the end of nuclear division. When all possible single and multiple deletion mutants were constructed, some multiple mutations were lethal, whereas all single mutants were viable. All lethal combinations included the clb2 deletion, whereas the clbl clb3 clb4 triple mutant was viable, suggesting a key role for CLB2. The inviable multiple clb mutants appeared to have a defect in mitosis. Conditional clb mutants arrested as large budded cells with a G2 DNA content but without any mitotic spindle. Electron microscopy showed that the spindle pole bodies had duplicated but not separated, and no spindle had formed. This suggests that the Clb/Cdc28 kinase may have a relatively direct role in spindle formation. The two groups of Clbs may have distinct roles in spindle formation and elongation. INTRODUCTION In the budding yeast Saccharomyces cerevisiae and the fission yeast Schizosaccharomyces pombe, there are two prominent cell cycle control events: Start in Gl phase and mitosis. Both events require the activity of a protein kinase that is encoded by CDC28 in S. cerevisiae and cdc2+ in S. pombe (Nurse and Bissett, 1981; Piggott et al., 1982; Reed and Wittenberg, 1990; Surana et al., 1991). Highly conserved Cdc28/Cdc2 homologs have been found in other eukaryotic species as diverse as Drosophila (Jimenez et al., 1990; Lehner and O'Farrell, 1990a), corn (Colasanti et al., 1991), and humans (Lee and Nurse, 1987). In a number of cases, these homologs are able to function in one or both of the yeasts (e.g., Lee and Nurse, 1987). The Cdc28/Cdc2 protein kinase activity resides within a complex containing at least one other protein, a cycin (Evans et al., 1983; Swenson et al., 1986; Draetta and Beach, 1988; Goebl and Byers, 1988; Hagen et al., 1988; Solomon et al., 1988; Booher et al., 1989; Draetta f Present address: Friedrich-Miescher-Laboratory, Spemannstr. 3739, D-7400 Tubingen, Germany. © 1992 by The American Society for Cell Biology

et al., 1989; Gautier et al., 1990). The cyclin may be thought of as a positive regulatory subunit of Cdc28/ Cdc2 and perhaps also of other protein kinase catalytic subunits. Six different sequence classes of cyclins have been described: A, B, C, D, E, and Cln (reviewed by Hunter and Pines, 1991). Functionally, only two roles have been defined: "mitotic" cyclins bind to and regulate Cdc28/Cdc2 in mitosis, whereas "Gl" cyclins bind to and regulate Cdc28 at Start. Most cyclins of the Btype sequence class are mitotic cyclins, although there appears to be at least one B-type cyclin that functions in Gl phase (Bueno et al., 1991). The three Cln cyclins of S. cerevisiae are Gl cyclins (Nash et al., 1988; Hadwiger et al., 1989; Richardson et al., 1989). The functional roles of the other sequence classes have not been defined precisely. The activity of the protein kinase complex at mitosis is regulated partly by availability of the cyclins. In some systems, notably the embryos of various marine invertebrates, cyclin synthesis is constant through the cell cycle, but cyclin abundance oscillates because of its proteolytic degradation in each cycle at the time of mitosis. In HeLa cells, not only is cyclin half-life regulated 805

I. Fitch et al.

with respect to the cell cycle, but in addition, the abundance of its mRNA also oscillates, peaking in G2 or M (Pines ond Hunter, 1989). In S. cerevisiae, the CLB1/ SCB1 and CLB2 mRNAs are strongly periodic (Ghiara et al., 1991; Surana et al., 1991). S. cerevisiae has at least four distinct B-type cyclins, in addition to the three known Gl cyclins Clnl, Cln2, and Cln3. It is not clear why there should be so many. We have characterized the sequence, expression, and mutant phenotypes of these four B-type cyclin genes in an effort to see if they function at mitosis and to distinguish any individualized roles they may have.

directly on both strands using the dideoxy chain termination method (Sanger et al., 1977). Primers were designed progressively to sequence out in both directions from the PCR amplified segment.

Cloning of the CLB3 and CLB4 Genes

Most other strains were constructed by transformation of one of the wild-type parents followed by crosses. A strain carrying a cdc15-2 mutation was constructed by crossing cdc15-2 to the W303 background, followed by multiple backcrosses to W303. For the a-factor block and release experiment, the a-factor sensitive strain RC629 (MATa sstl-2 ade2 adex ural his6 metl canl cyh2) was used. The sstl mutation is also known as barl.

A 387-bp fragment of CLB3 was amplified using 5' primer #310 (above) and 3' primer #103 (5'-AZYTCCATZAXXTAYTTZGC-3'), corresponding to amino acid sequence AKY(LF)ME(LVIFM). A 228-bp fragment of CLB4 was amplified using 5' primer #101 (5'ATZGCZVCZAAXTAYGAXGA-3', where V = A/T), corresponding to amino acid sequence (IM)A(ST)KYE(ED), and 3' primer #103. PCR conditions were as indicated above, with each primer added to 10 ,uM. One microgram of DNA cut with Sph I or EcoRV was used as template. The digestion prevented amplification of CLB1 gene fragments, because the relevant regions of CLBI were known to contain Sph I and EcoRV sites. The CLB3 and CLB4 fragments were gel purified and cloned into pUC119 as described for CLB1. Plasmid clones hybridizing to the CLB3 and CLB4 PCR fragments were isolated from an S. cerevisiae genomic DNA library of partially Sau3A-digested fragments cloned into YCp5O (obtained from M. Rose, Princeton). The CLB3 PCR fragment hybridized to a 5-kilobase (kb) EcoRI fragment; Southern analysis of genomic DNA identified the corresponding genomic fragment. Sequencing identified a complete open reading frame within this fragment. However, in the case of CLB4, the initial clone included only the 3' two-thirds of the gene. This partial clone was used to make the CLB4 disrupting plasmid pCD4 (see Figure 3). An overlapping fragment containing the 5' end of CLB4 was found in a clone from the YEp213 library (see above). The sequence deduced for CLB4 (see Figure 2) includes information from both clones.

Media

Plasmids

YEPD was 1% yeast extract, 2% peptone, 2% glucose. YEP gal + raff was 1% yeast extract, 2% peptone, 1% raffinose, and 1% galactose (crystalline grade, G 0750, Sigma Chemical, St. Louis, MO). Agar was added to 2% for plates. The synthetic media were YNB drop-out media as described by Rose et al. (1990).

Plasmid pCD3 (see Figure 3), used for disruption/deletion of CLB3, was constructed as follows. A 5-kb genomic EcoRI fragment containing CLB3 was cloned into pUC1 19. The plasmid was cut at a single EcoRV site by incubating with EcoRV in the presence of 25 Ag/ml ethidium bromide. After phenol/chloroform extraction and ethanol precipitation, the DNA was cut with Xho I, and a 7.6-kb fragment (consisting of the parental vector, less the 0.6-kb EcoRV to Xho I fragment) was purified from an agarose gel. A 0.95-kb Sma I/Sal I fragment carrying the TRPI gene was excised from plasmid pJJ280 (Jones and Prakash, 1991) and ligated to the compatible EcoRV/Xho I ends of the CLB3/ vector fragment to give plasmid pCD3. Plasmid pCD3 was digested with Pvu II (which cuts on either side of the polylinker in pUC119 and also in CLB3 0.47-kb upstream of TRP1) to give the fragment used for disruption/deletion of CLB3. Plasmid pCD4 (see Figure 3) was constructed by subcloning a 4kb Pst I fragment from the original YCp5O library clone into pUCl 19. One end of the Pst I fragment consisted of the 1 100-bp Pst I to BamHI region of YCp5O (originally derived from pBR322), whereas the rest of the fragment consisted of the 3' two-thirds of the CLB4 gene and additional yeast genomic DNA. The resulting plasmid was digested with Stu I and Spe I, and a 6.8-kb fragment was purified from an agarose gel. A 1.8-kb Sma I/Xba I fragment carrying the HIS3 gene was obtained from plasmid pJJ215 Uones and Prakash, 1991) and ligated to the compatible Stu I/Spe I ends of the CLB4/vector fragment to give plasmid pCD4. Plasmid pCD4 was digested with Pvu II (which cuts on either side of the polylinker in pUC1 19) to give the fragment used for disruption/deletion of CLB4. Note that the fragment liberated by Pvu II has 1200 bp of vector sequences at one end; furthermore, at this end of the HIS3 gene, the amount of homology to the CLB4 gene is only -.260 bp. Nevertheless, clb4:: HIS3 disruptants are obtained at a high frequency.

MATERIALS AND METHODS Yeast Strains The wild-type yeast strains were an isogenic pair. W303 1A: MATa ade2-1 trpl-l leu2-3,112 his3-11,15 ura3 canl-100 [psi'] W303 1B: MATa ade2-1 trpl-l leu2-3,112 his3-11,15 ura3 canl-100 [psi']

Cloning of the CLB1 Gene CLB1 was cloned using a two-step polymerase chain reaction (PCR) approach. First, a 387-base pair (bp) fragment was amplified using 5' primer #310 (5'-ATGWGZGCZATYYTZXTZGAYTGG-3', where W = A/C, X = A/G, Y = T/C, Z = A/T/C/G), corresponding to the amino acid sequence M(RS)AI(LF)(VIM)DW, and 3' primer #313 (5'-

YTCCATZAXXTAYTTZGCZAXZGT-3'), corresponding to amino acid sequence T(LF)AKY(LF)ME. One hundred-microliter reactions contained 10 mM tris(hydroxymethyl)aminomethane (Tris)-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 0.001% gelatin, 0.2 mM each of dATP, dTTP, dCTP, and dGTP, 10 AM of each primer, 1 Ag of S. cerevisiae genomic DNA template from strain LL20, and 2.5 U of Taq polymerase (Perkin Elmer-Cetus, Norwalk, CT). PCRs were carried out at 94°C for 5 min, then 30 cycles at 94°C for 1 min, 48°C for 2 min 30 s, 72°C for 1 min, followed by 72°C for 10 min. A 1/500 dilution of the resulting amplified material was used as the template in a second PCR. This time, an intemal 5' primer, #312 (5'-ATYGCZACZAAXTAYGAXGAXATGTA-3'), corresponding to the amino acid sequence IATKYEEMY, was used instead of primer #310. PCR conditions were as above, except primer concentrations were reduced to 1 AM. The resulting amplified 227-bp fragment was gel purified from a 5% polyacrylamide gel and blunt-end ligated into pUC119. Before the PCR, all primers were phosphorylated at their 5' ends using polynucleotide kinase to allow ligation of the amplified fragment to vector ends. The complete CLBI gene was isolated from an S. cerevisiae genomic DNA library of partially Hindlll-digested fragments in YEp213 (obtained from Dr. Scott Cameron, Cold Spring Harbor Laboratory) by screening _104 colonies with the radiolabeled PCR fragment. Plasmid was isolated from several positive clones, and the gene was sequenced

806

-

Synchronization with a-Factor A 500-ml culture of RC629 (sstl-2 = barl) was grown at 30°C to a density of 6 X 106 cells/ml in YEPD buffered to pH 5.3 with succinate.

Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

a-factor (synthesized at Cold Spring Harbor Laboratory) was added to a final concentration of 1 MM, and the culture was incubated for 2.5 h. More than 95% of the cells were arrested as unbudded "shmoos." The final cell density was 1 X 107 cells/ml. Cells were harvested, washed with 500 ml of YEPD, and resuspended in 400 ml of YEPD. The culture was incubated at 30°C, and samples were taken every 10 min.

1.28 kb

CLB3 ORF

(a)

I%

P"

I -1- 1

I 1- 1

AI

I

rij

I

Il

I

0.6 kb

Synchronization with cdc15 A 500-ml YEPD culture of a cdc15-2 strain was grown at 25°C to 1.5 X 106 cells/ml in a 2-1 flask. The culture was then shifted to a shaking platform in an air incubator at 37°C for 3 h, after which the cell density was 4 X 106 cells/ml. More than 95% of the cells in the culture had large buds. The cells were released from the arrest by shifting the culture to a shaking platform in a 25°C water bath. Samples were taken every 15 min.

(b)

Il

I

I I

I

0.45 kb

Extraction and Analysis of RNA Samples (10 ml) were chilled on ice, harvested by centrifugation, washed with ice-cold water, harvested, and resuspended in 250 IAl LETS buffer (0.1 M LiCl, 0.01 M EDTA, 0.01 M Tris-HCl, pH 7.4, 0.2% sodium dodecyl sulfate). The suspended cells were transferred to ice-cold 1.5-ml microcentrifuge tubes containing 300 Al phenol and 300,l acid washed glass beads (425-600,m, Sigma). Each tube was then vortexed three times for 15 s each at top speed. All tubes were stored on ice until all samples had been processed. The tubes were then spun briefly in a microcentrifuge, an additional 250 ,l LETS buffer was added, and the tubes were vortexed again for 15 s. The tubes were then spun at 14 000 rpm for 5 min. The supematant was taken and extracted twice with phenol:chloroform:iso-amyl alcohol (24:24:1). The RNA was precipitated with 1/10 volume 5 M LiCl and 2.5 volumes of ethanol. The pellet was resuspended in 360 Ml of diethyl pyrocarbonate (DEPC)-treated water. The RNA was reprecipitated using 1/10 volume of 3 M NaAcetate and 2.5 volumes of ethanol. The final pellet of RNA was redissolved in 20 Ml of DEPC-treated water and stored at -70°C. Approximately 5 ml was loaded per gel lane. The RNA was separated by formaldehyde-agarose gel electrophoresis (Rose et al., 1990) and transferred by blotting to a Nytran membrane. The RNA was cross-linked to the membrane using an ultraviolet cross-linker (Stratagene, La Jolla, CA), and then the membrane was baked for 2 h at 80°C in a vacuum oven. The desired RNAs were visualized using DNA probes labeled with 32P by random-priming and using standard hybridization techniques (Church and Gilbert, 1984).

1.38 kb

CLB4 ORF

Figure 1. Restriction maps of CLB3 (a) and CLB4 (b). The location of each open reading frame (ORF) is indicated; so too are the fragments replaced to disrupt each gene, respectively.

plates. A segregant that was leu+ trp+ his' and also nonviable on glucose was selected. This strain, called #245, was W303 MATa GAL10CLB2 clbl: :URA3 clb2: :LEU2 clb3: :TRPl clb4: :HIS3.

GAL10-CLB2 Depletion Strain #245 was grown to early exponential phase in YEP gal + raff at 30°C. The culture was treated with 1 Mg/ml a-factor for 1 h. Cells were collected by filtration, resuspended in an equal volume of YEPD containing 1 Mg/ml a-factor, and incubated for an additional hour, after which most cells were unbudded "shmoos." The culture was split in half, and cells in each half were harvested by filtration and washed with two volumes of YEP. One part was then resuspended in YEPD, whereas the control was resuspended in YEP gal + raff. Samples for kinase assays and tubulin and 4'6-diamidino-2-phenylindole* 2HCl (DAPI) staining were taken every 20 min for 3 h. An identical control experiment was done with an isogenic wild-type strain. Staining with anti-tubulin antibody and with DAPI was as described by Surana et al. (1991), and Clb2-associated kinase activity was assayed as described by Amon et al. (1992).

Electron Microscopy Construction of a GAL10-CLB2 clbl clb2 clb3 clb4 strain A 2.8-kb Xho I/EcoRI fragment containing CLB2 was cloned into pIC19H. Using PCR, the EcoRI site was destroyed, and a new EcoRI site was created just upstream of the CLB2 ATG start codon. The 2.2kb EcoRI/BamHI CLB2 fragment from this construct (BamHI is in the polylinker of pIC19H) and a 0.69-kb BamHI/EcoRI GALl-GALIO fragment Johnston and Davis, 1984) were cloned into the BamHI site of pNC161 (a pUC118 derivative) by triple ligation. This generated a GAL10-CLB2 construct on a CEN-based-TRP1-selectable pNC vector. The BamHI fragment containing the GALlO-CLB2 in pNC161 was excised and cloned into the BamHI site of the URA3-based integrative vector, YIplac2l 1. This construct was used to integrate GAL1O-CLB2 at the URA3 locus of a strain canrying the clb2: :LEU2 disruption (strain K1890). Integration was confirmed by Southern analysis. The resulting strain was crossed to strain K1886 (carrying clbi: :URA3). The diploid was sporulated, and tetrads were dissected on YEP + galactose plates. A leu+ segregant that was nonviable on glucose was selected. Such a segregant was GAL1O-CLB2 clbl::URA3 clb2::LEU2. This strain was then crossed to strain K2652 (clbl::LURA3 clb3::TRPI clb4::HIS3), the diploid was sporulated, and tetrads were dissected on YEP + galactose

Vol. 3, July 1992

Cultures were fixed and prepared for electron microscopy by methods described previously (Byers and Goetsch, 1991). Serial sections were mounted on formvar films covering single-hole grids for viewing in an electron microscope (model EM 300, Philips Electronic Instruments, Mahwah, NJ). All data reported represent sufficient serial sections of the cells to establish the spindle phenotype unambiguously (see Table 3).

RESULTS

Cloning of CLB3 and CLB4 Degenerate oligonucleotides corresponding to conserved regions of other cyclins were synthesized and used in PCRs as described in Materials and Methods. The fragments originally amplified were from the CLB1 gene, which was cloned and sequenced (Materials and Methods) (Surana et al., 1991). Because we felt there might be multiple B-type cyclins, we repeated the PCRs using genomic yeast DNA that had been digested with either 807

I. Fitch et al. (b)

(a)

Xbal

lbal Ban=8 -300

-300

-240 GCAGCAGGC

-240 GAAGATrACAAGA

GGC

-120 -60 1

-12G0

TAAMCA

IAC

AIC CAT CAT MC ICA CAG [CT TIC MC 'TCT GGA CAC ATC AGG AGC S G I R S S N S L H S Q H M H

91

ACT CGA [CC CI AGT CAT ATT TCA [C! GCG CAC CCG AGG GTC GCA P R A S I V S L S S A H H T G

136

CIT AGC GAC GIT ACC AAT ATA GTT GCG ACA AAC CT AGC AAC AAC T S S V A N N N N D V T I L S

181 AMC ATA AGT MG CIA AAA GC GCC CCA AT! AAA GAA AGA TIC GAT I K E R P K L D K A P S I S V PVuII 226 'CA GCT GCG ATA ATT GAG GAGM AGG CTG GAT GCG AAT AGT GCT S

A

A

I

I

E

E

E

R

-60 Tr_GGAAAC _cACTATACCI;GATACTAGGC'GCCCCAMACAAGGAAAIGACAG

TTAACITATTATM

CCC GAG GAT GAA AAT GCC GCA CCT ATA CGT AAT CIT AAA CAC AGC P I G N L K R V A H D E N P E

L

D

A

N

S

V

GCA CAG AGA AAA GAA GCT GAT CAT AAC GAT TIC TrA ACG GAC AGG L T D R K A D L A Q R E D H N

316

GAA CAA GAG GAA CCC GTT GAA GAC GAC GGA GAA AMC GAA GAG GAT S E D P V D D G E E Q E E E E

361

GAA GAA GAA GAC CAG GAG CIT CTA CCI TIG CAA CAT TAT GCT ACT S Q Y A D P L H L E E E Q E L D

T

L

V

W

E

H

A

F

R

T

Y

Y

R

T

ACA TrA GAT CCC AAT GAT GAT GAC GCC N D D D P D V T L

496

GCC GAA TTA '[! AAT GAG ATA TI L E I F S N A E

541

GAC CCI TAT AM CCC AAC CCG TAC TAC ATG GAT AM CAA CCA GAG Q P Y Y M P E Y K P D K D L N

586

[TA AGA CIG IC L S R W

TAC GAT CCI GCC A& GCIT Y D V V M V

GAG TAT AM AGG AM TIC GM K Y E M R L E

TrT CGA AGC ACA CTG ATT GAT [CI ATC GTC CAA F R S T L I D W I V Q

631 GTA CAT GM AAA TIT CAA CT TTA Q F V H E K L L UooRV 676 ATT AAT ATA ATA GAC AGA TAT C& I N I D R Y L I

CCT GAA ACT CTA TAT CIC IC T C L Y L P E

30

30

45

91 AAG MC GTA CIT TAC CMA GGA GIT CAA AAG GGT ATA AM AGG CTA G V Q K G I K R L L Y Q K N V

45

136 GM AAA AGA CM AGG AGG GTr GCA [TA GGT GAT GTA ACC tTC CAA

K

E

V

V

P

V

K

F

Q

L

V

G

A

A

S

L

F

I

A

A

K M

Y

E

E

'CA GM MC S

E

N

I

N

C

P

T

I

K

T

I

L

F

V

Y

N

G

L

E

F

E

L

G

W

P

G

[CA TIT TrA CGA AGA ATC AGT AAG GCA GAC GAT TAC GAG

P

S

M

F

L

R

R

I

S

K

A

D

D

Y

E

946 CAT CAT ACI AGA ACA CI Gcc AMA TAT CTA TI GM ICC ACA ATA H D T R T S T I K Y L L E L A

1036

A& GCAC CAT CGA CCI GTT TCC GCT CAA CCT ACT [CI TrA GCT GCI M D L A A A Q P S W R L V S H

GCT OCA TAC TIT G

1081

A

Y

F

CTA AGT MG ATr ATr CTG GGC CM MT CAG T&G Q I I L S K L G N Q W

ITr CCI GCG CAC GCI TAC TAT TCC MT TAT ACA CM GAA CM ATr S

L

A

H

V

Y

Y

S

N

Y

T

Q

E

Q

I

R

R

V

A

L

G

D

V

T

S

Q

90

90

105

271 GTA AAA GAA CAA CAA CGA GAC GTA AGG CAT GM GAT AGC GAC TAT V K E Q Q R D V R H E D S D Y

105

316 TIT TrA ATr GAT AGT 'CT GM GGC TCT TCT ACT GAT GAC GM CAA F

120

361

V

135

L

I

D

S

S

E

G

S

S

T

D

D

E

Q

N

E

D

A

I

D

D

L

S

L

R

R

V

N

150

IT GM GAG GAT GTT GAT AGT CM ATr GAA D V D S Q I I E E E

165

180

496 CCA CTA [CA CCA ATA MC AAC GAT GA ATT CAG ACT GAG CCI GAC T E L D I Q L S P I N N D E P

180

195

541 AGG GCI[TIT GM AAA TAT TTr CIG lCG GTT OX AAT CCG CTG GAT V P D R S P N L F R A F E K Y

150

451 GAA MG CAA GAT CIC E

165

M

Q

D

V

Sup!

225

240

676 CCC TAC TAT ATG CAM MT CM GTA GAG CTr ACA 'MG CCG TC AGA R 240 L T W P F P Y Y M Q N Q V E

210

GAC D

721 CGA ACT AIM ATA GAT TUG CTA GCT CAA CCI CAT mT AGA TT CAA R

255

T

M

I

D

W

L

V

Q

L

H

F

R

F

Q

L

270 811

L

P

E

T

L

Y

L

T

I

N

I

V

D

R

TTT CTG 'CA AAG AAG ACC GTT ACT TC AAC AGG TmT CM TIC GCT L S K K T V T L N R F Q L V F

lCG

GCT 'TA TI A L F

R

V

E

R

S

N

1373

CACACAAAAAC CAAGAAAAAGI I'ITMCA CAGIACTMT [88. CCA ACATA[rAAcMC AITMACCTMAACATACTMAAC= T [TMM

1433

CA

1313

285 300

GAT GAT CTA GCT TAC ATG CMI GAA AAT ACA TAC N T Y E D D Y M L L V

315

330

946 ACT AGA GAT GAC ATT ATT AGA GCG GAA CAG TAT ATI ATA GAT ACT I D T T R D D I I R A E-Q Y M

330

345

991 CTG GM TT GM ATA CGT TGG CCA CGA CCC A¶ E F P G P E I G W M L

300

G

V

S

901 TGC CCI ACT TI 315

360 375

C

P

T

L

CCA mI TrA AGA P L 345 F R.

1036 AGG ATA AGT AAA GCA GAT GAC TAT GAC TIC GM CCA AGA ACA TTA I D 360 R S K A P R T D Y D F E L a! s 1081 GCA AAG TAC TTA TIC GAA ACT ACA ATA GTA GM CCC AM CTA GTG 375 A K Y P K L V I V L L E T T E

TCI H

270

ATT GCT GCC AAG TIT GAA GAG ATT MC I A A K F I N E E

856 GGT GTA

1216

Xbhol

255

766 CIT TrA CIA GM ACG CTA TAC CCI ACG AT MT ATA GCC GAT AGA

1171 AGA ACA ATT CIT GGT 'CA AAT GAT TGG TCr [TA AAA CAT GTA TIC

AAAGCCCAGCICCAGACA[GGCACMAA

210

MT N 225

1171

B3gI

195

586 GAT GAT ACC CAT GAT GCT GIG A'[ GTT GIG GAG TAC GCr IC Y A S V E M V V D D T H D V stul 631 ATA TIC TAT TAC TC AGA GAA CIT GAA GIG AM TAT AGG CCT K Y R P E F Y Y L E V I L R

1126 GCT GCI GCA CCA AMC TGG TTA GCT GCT GGC GCI TAT TTr CG AGC A A A P S W Y F L S L A A G A

L

135

406 GAT CAG CAG MT CAA OCC GAT GAA GTG TAT GM GAT TIC GAT GGA Y E D F D G I D E V Q A D Q Q

888. AM CGT CAT MC GCC ATA lOG AGA AM TAT TCT 7CA CGT CGT TAT K R S R R Y H N A R K Y S I W

1261 CAC AGA GTA GM AGA 'CT AAC TM

120

GIT MT GAA GAT GCT ATT GAT GAT TIG TTA AGT CGA AGA GTA AAT

1126 CIT CCG TIC GCC ACC ATT AKT TrA GAA MT CC AGA TAT GCC 'CT L 390 Y A S L E N C R P I I L A T

CAT TCT [CA CAG AIC GTA GCIG AAG ICG ATA GCA TrA GCT GAA Q H S S A L A A K W I I V E

60

75

GGT [CG CCT GGT

CCG ATK

Q

226 ACI AAG AAC AAT T}T GAA ATA GAG MC ATA CIC [CA TCG GCC TG L R S A I S E E N I T K N N F

IGC TAC [CA AGG AAC GAC CIG CCI GAC GCA GAA C 285 D A E Y S R N D L L

856 AGA ACT AT! TIC AAC GGC [TA GAA TIT GAA T R

D

R

75

766 AAA TAT GAG GM ATC AAC T'T CCT ACA ATC AAG GAT TIC GTA TAC

811 ATG

K

181 AAG GCA AAC AM ATA CAC MT GCT ATA CAT MT MA TIC CAT CAG I H N K P H Q H N A K A N K I

721 MT AAG TIC CM CIT GIGCT GCA GCC [A CTC TIC AKT GCT GCT N

E

60

ICC AAA GM CIT GCT CCT GTA C

15

46 ATA GGT GAC ATr GM ATT CAG GAC GAA AAT OCA MC CAA GAA GTr Q D E N A N Q E V I E I I G D

SphI 406 GAT ACA TIC GTC l&G GAG CAT GCA TIT AGA ACT TAC TAT AGA ACT 451

AMG MG CIT GM GGG TAT ACG GTA CAA CCT CCA CAG TCT ACT TI Q S T L P P E G Y T V Q M L

M

15

M~ai

271

991

A

GAMTATAAAGTITAAATTAATAc ICGTATT

-120

C[ICTGMAMAGAATCA

ICICIICCICGI

46

901

AA

-180 TItCTr'CI'ICCICN3:TrITAACIAAA£CAAAcAAAACAKIGMC2V~

-180

405 420

R

T

I

L

G

S

N

D

W

S

L

K

H

V

F

1216 TAC IT GGC TAT ACA ICC AGC CM ATA AKT C'CT TrA GCA 'A CCI Y S G L A S L I I P Y T S S Q

390 405 420

1261 ATA TIC GAG AAT IC AAG AAC GCA [CT CGA CIC CAT CAT CA ATT I

L

E

N

C

K

N

A

S

R

R

1306 TGG AM AAA TAC TIT GAC CAA AAG CAT TAC W

K

K

Y

F

D

Q

K

H

Y

H

CIC lIT R

1351 GTA GM GM 'CG ATT GIT lCG ACA GM GCI TAA V E E W T E A I V S

C

H

S

I

TT CAA ATr S Q I

435 450

CTrACICAICAACA

1400

CCATACAT[TrACACITACACACTCACACATCATACTACATATACTCACAAGMGATAT

1460

TATPAAAAAAAAAACACTGCTTACACATATATATATAT AT:ICA

1520 ATMATATATATAT

of CLB3 and CLB4. The nucleotide sequences of CLB3 (a) and CLB4 (b) are shown, together with predicted amino acid sequences. Underlined DNA sequences represent upstream motifs common to both genes. Putative TATA elements are highlighted with lines above the respective sequences. Underlined amino acids represent a sequence common to all four Clbs that encompasses the FLRRSK motif characteristic of B-type cyclins.

Figure 2. Nucleotide

808

sequences

Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

(a) tI BamHI

IPVuIIHpaI S

PvuII

~~~~~clb3::TRPI

^

EcoRI

h I~~-Sh

p

NdeI

/

pCD3

(ca. 8.55 kb)

\pUCllg9

Pun (b)

HindII EcoRV

ECRIE SP

nun

clb4:.HIS3 pCD4 (ca. 8.55 kb)

ipUCll9§

Figure 3. CLB3 and CLB4 disrupting plasmids. pCD3 (a) and pCD4 (b), containing clb3::TRPI and clb4::HIS3, respectively, were constructed as described in Materials and Methods. Black shaded regions represent CLB3/CLB4 open reading frame. Arrows indicate orientation of open reading frames. Grey shaded regions represent flanking yeast genomic DNA.

Sph I or EcoRV. Because there is a Sph I site and an EcoRV site in CLB1 between the primer binding sites, either of these digestions would prevent the CLB1 fragment from being amplified but might allow amplificaVol. 3, July 1992

tion of other cyclin genes. Indeed, these new PCRs yielded new amplified fragments. On sequencing of the fragments, it was clear that they represented two new genes encoding B-type cyclins, which we now call CLB3 and CLB4. The intact genes were cloned by hybridization from a yeast genomic library and sequenced in their entirety (Figures 1 and 2). In an effort to find any additional B-type cyclin genes, we used each CLB gene as a probe in moderately high stringency Southern hybridizations. CLB1 and CLB2 cross-hybridized, but no other unexplained hybridizing bands were seen. PCR primers corresponding to two regions conserved between Clb3 and Clb4 (Y(K/ R)PNPYYM and GWPGPM(S/P)F) were synthesized and used in PCRs with DNA from clb3 clb4 deletion strains (see below). No additional CLB fragments were found. However, these negative results do not rule out the existence of additional CLB genes. A comparison of the complete amino acid sequences of the four known B-type cyclin genes of S. cerevisiae is shown in Figure 4. All four proteins are clearly Btype cyclins, as opposed to A-type, Cln-type, etc., because they contain the motif FLRR-SK. All four proteins contain a nine amino acid "destruction box" in their Nterminal region; this motif (B-type consensus RxALGyIxN, where x is no consensus, and y is N, D, or E) is thought to target mitotic cyclins for destruction via ubiquitination at mitosis (Glotzer et al., 1991). The only known B-type cyclin lacking a destruction box is cigl+ of S. pombe (Bueno et al., 1991), and this cyclin is thought to act in Gl or S phase; the presence of a mitotic destruction box in all four Clbs suggests they are mitotic cyclins. The four destruction boxes are not identical to each other nor are any of them identical to the consensus, so there may be differences in the efficiency or timing of destruction of individual Clbs. Notably, the four proteins form two homology groups, Clbl and Clb2 constituting one pair and Clb3 and Clb4 the other (Table 1). Furthermore, the Clbl/ Clb2 pair is on a different evolutionary branch of the cyclin tree from the Clb3/Clb4 pair (Figure 5). The two groups of genes are expressed at different times (see below), reinforcing the idea that they have diverged. Interestingly, Clb3 and Clb4 are most closely related to cigl+, which, like Clb3 and Clb4 (see below), appears to have a function relatively early in the cell cycle (Bueno et al., 1991). Map Position of the CLB Genes The four CLB genes were located on the physical map of the yeast genome by hybridizing each gene in turn to a CHEF gel chromoblot, to identify the chromosome containing each gene, and also to a set of lambda clones encompassing most of the yeast genome. Nitrocellulose filters carrying the set of lambda clones were provided by M. Olson and L. Riles (Washington University, St. Louis), and the hybridization data was interpreted by 809

-~

I. Fitch et al.

M. Olson and L. Riles. CLB1 resides on the right arm of chromosome 7, -200 kilobase pairs (kbp) distal to the centromere and just proximal to SPT6. CLB2 resides on the right arm of chromosome 16, -320 kbp from the centromere and distal to SUPI 6. CLB3 is on the left arm of chromosome 4, -90 kbp from the centromere and just proximal to PPH2. CLB4 is on the right arm of chromosome 12, '400 kbp distal to the centromere and proximal to CDC42. Meiotic mapping has not yet been done. To the best of our knowledge, none of the CLB genes corresponds to any previously mapped gene.

Time of Expression of the CLB Genes To examine the expression of the CLB genes, we synchronized cells by imposing cell cycle arrest and release. Because such arrests can introduce artifacts, two distinct methods of arrest were used: a-factor arrest in G 1 phase

Table 1. Percent identity among the Clb proteins Clbl

Clb2

Clb3

35 37 76

35 36

57

Clb4 Clb3 Clb2 Clbl

Numbers represent percentage of identical residues between pairs of aligned Clb proteins in a 315 amino acid region encompassing the cyclin boxes. This region corresponds to that beginning at residue 118 of Clb3 (EEDEE .) and ending at the C-terminus of the alignment shown in Figure 4.

(Figure 6A) and restrictive temperature arrest of a cdc15-2 mutant in anaphase (Figure 6B). The two experiments each gave essentially identical results with

Alignment of the Four Clb Proteins

CLB1 CLB2 CLB3 CLB4

MSRSLLV-ENSRTINSNEEKGVNESQYILQKRNVPRtTXZINVTMANILQEISMNRKIGMKNFSK

CLB1 CLB2

--------------------LNNFFPLKDDVSRADDFTSSFND_RQGVKQEVLNNKENIPEYGYS --FTRESVSRSTAAQEEKRTLKENGIQLPKNNLLDDKENQDPS_QQFGALTSIKEGRAELPANIS

CLB3 CLB4

------------______________________________

CLB1 CLB2

EQEKQQCSNDDSFHTN-----STALSCNRLIYSENKSIST2MEWQKKIMREDSKKKRPISTLVEQ

CLB3 CLB4

--KVAPIKERLDSAAIIEEERLDANSVAQRKEADHNDLLTDREQEEPVEDDGESEEDEEEDQEPL

CLB1 CLB2 CLB3 CLB4

MSNPIENTENSQNTSSSRFLRNVQRLAZJWV2TTFQKSNANNPALTNFKSTLNSVKKEGSRIPQ MHHNSQSLSSGHIRSPEDENVAPIGNLKHRTGSLSHISSAHPRVALSDVWIVATNSSNNSISKP I4LEGYTVQPPQSTLIGDIEIQDENANQEVKNVLYQGVQKGIKRLEKRQRtVAL6DVTSQKANKI

----------------------------HNAIHNKFHQTKNNFEIENIRSSALVKEQQRDVRHED

LQESSSAKEIIQHDPLKGVGSSTEVVHNSVENEKLHPARSQLQVRNTESETDSGKKRPISTIVEO SDYFLIDSSEGSSTDDEQVNEDAIDDLLSRRVNDQQIQADEVYEDFDG_EMQDVIEEDVDSQIEPL a

&M _ja& _"&=zu .AM ,MI X j6-

ELPKKFKVCDENGK-----EEYEWEDLDAEDVNDPFIWSZYVNDIFEYLjQLZVITLPPKKEDLYQH LLQHYASDTLVWEHAFRTYYRTTLDPNDDDVYDVV

wAZLSNEXrEYMRKxLDLYKPNPYYMDKO

SPINNDEIQTELDRAFEKYFRSVPNPLDDDTHDVVSVVIYASDIFYYLRELZVKYRPNPYYMQNQ

CLB1 CLB2

KNIKN-RDILVNWIIKIHNKGLLP3TLYhANMMDhICEZVV TNRQLVGTSCLU-

CLB3 CLB4

PELRWSFRSTLIDWIV2VHEVQZTLYLCINTLIDRYNI LVVPVNPpVGAQLVGMSSwzYZ VELTWPFRT

CLB1 CLB2

ZIYSPS_H_YETDGACSVEDIKEGZR_ILEKfiDQIOSFAN DDRRSKDDXISRTIA ZVYSPSXXHFASETDGACTEDEIKEGZKFILKTLKFNLNYPNPNNFIRR18 N DDYD SSRLA

CLB3 CLB4 CLB1 CLB2

CLB3 CLB4 CLB1 CLB2 CLB3 CLB4

810

Clb4

RNIHQN-RDILVNW LI

_LTLYLAN

RLVQLDKVQL_S£w_

R ZTLYLB TNXTLRTSKVTXSR LMID

TQLVG A

KFZ

ZINCPTIK_rVYMSENCYSRNDLLDAZRTILNGLEELGWPGPXSrLRR1ISADDYEHDTRTLAR

ZINCPTLDDLVYMLENTYTRSDIIRAOQYMTDTLETEIGWpGvyPmnlwwDDYDrwPTuhpLa

FLMZISXVDFKFIGILP8LCASAAMFL8RXMGK~GTIDGN.LIHY81GGYTKAX-YIPVCQLLMDYL-

FLLZISLVDFRFIGILPSLCAAAAMrMS4M.JGKGKNDGNLIHY8GNgIYEELAPVCHMXMDYLYLLUSTIMDHRLVSAQPSWLA&GAYFLSKI LGQNQWSLAHVYYSN-YTQEgJLPTLATIILENCR YLLKTTIVEPKLVAAAPSWLAAGAYFLSRTILGSNDWSLKHVFYSG-YSGSSIPLASLILENCK VG8TIHDEFLKCYQ8RRFLKASIISIEWALKVRKNGYDIMTLHE

VSPIVWEFHRKYSRRFMKASIISVQWALKVRKNGYDIMTLHE YA8KRHNAIWRKYSSRRYLHSS8QXVAKIALAEHRVERSN

Figure 4. Alignment of the Four Clb Proteins. The predicted amino acid sequences of Clbl, Clb2, Clb3, and Clb4 are aligned. Residues shown in bold type are identical in at least three of the four cyclins. Underlined residues are those conserved only between Clbl and Clb2, or between Clb3 and Clb4. "Destruction box" motifs are shown in bold italics, and this alignment highlights their relative distances from each respective N-terminus. Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

Cyclin B Gene Family CycB2-XI CycB2-Hs CycB2-Gg

CycB-Asp

they dropped at about the same time as the CLB1 and CLB2 transcripts. Although all four transcripts were periodic, each remained detectable throughout the cycle. This may indicate either that cell synchrony was imperfect or that all have a low, constitutive level of expression.

CycB-Arp

CycB-Ss

CycB-Pv CycBl-Hs CycBl-XI -

CycB-Dm

CycB-Gm CLBI/SCBI-Sc CLB2-Sc CDC13-Sp CIGI-Sp CLB3-Sc

CLB4-Sc

Figure 5. The cyclin B gene family. Evolutionary tree of the cyclin family constructed using the neighbor-joining method (Saitou and Nei, 1987), as described by Xiong et al. (1991). Lengths of the horizontal lines reflect divergence. This tree was constructed using only 106 of the central residues in the cyclins shown, a region that is somewhat homologous between all cyclins, and so allows comparisons between very divergent cyclins to be made. For Clb3, this region begins at residue 200 (FRSTL .) and ends at residue 305 (. PMSFL). Because many of the distinguishing residues for the Clb3/4 pair are outside this region, this tree underestimates the distance between the Clb3/4 pair and the Clbl/2 pair. In this diagram, cdc13+ of S. pombe is in the Clb3/4 branch of the tree. However, when the tree is rooted differently, or when different residues are used, cdc13+ sometimes moves to the Clbl/2 branch. The cyclins used are as follows: Arp, Arbacia punctulata (sea urchin) (Pines and Hunter, 1987); Asp, Asterina pectinifera (starfish) (Tachibana et al., 1990); Gg, Gallus gallus (chicken) (Gallant and Nigg, 1992); Dm, Drosophila melanogaster (Lehner and O'Farrell, 1990b); Gm, Glycine max (soybean) (Hata et al., 1991); Hs, Homo sapien (human Bi, Pines and Hunter, 1989; B2, Xiong, Connolly, Caligiuri, Futcher, and Beach, personal communication); Pv, Patella vulgata (gastropod) (van Loon et al., 1991); Sc, S. cerevisiae (Ghiara et al., 1991; Surana et al., 1991; this report); Sp, S. pombe (Booher and Beach, 1988); Ss, Spisula solidissima (clam) (Swenson et al., 1986); and XI, Xenopus laevis (frog) (Minshull et al., 1989).

all genes tested, suggesting that the results accurately reflected the events occurring in an unperturbed cell cycle. As previously described by Surana et al. (1991) for CLB1 and CLB2 and by Ghiara et al. (1991) for CLB1 (as SCB1), the CLB1 and CLB2 mRNAs are strongly periodic. The mRNAs appear late in the cycle, probably after S phase, peak 10 min before anaphase, and then dropped late in anaphase. Because they are present only at very low levels at the cdc15-2 block, we presume they dropped before the completion of anaphase. CLB3 and CLB4 had a different pattern of expression. Their mRNA levels were also strongly periodic but rose much earlier, at about the beginning of S phase, just after the peak in CLN1 and CLN2 mRNAs (Figures 6 and 7). The levels stayed high until late anaphase, when Vol. 3, July 1992

Phenotypes of cib Disruptions Each CLB gene was disrupted and partially deleted in a diploid as described in Materials and Methods and by Surana et al. (1991). For each CLB, the part of the gene deleted included sequences encoding highly conserved cyclin motifs. The diploids were sporulated, and tetrads were dissected. For each of the four single mutants, all four spores were viable, and the marker for the disruption segregated 2:2, showing that none of the CLB genes was individually essential. The complete absence of part of the CLB reading frame in two of the four spore clones was confirmed by Southern analysis for several tetrads of each type. The clb2 disruption had a mild phenotype: the cells were somewhat larger than normal, with a higher than normal percentage of budded cells, and a higher than normal proportion of cells with a G2 DNA content (Surana et al., 1991). These phenotypes suggested that the clb2 mutant was delayed at mitosis. The other three single disruptions had no obvious phenotype. The four single disruption strains were crossed with one another to make all possible combinations of double mutants. Spore clones predicted (from the markers present in the viable spore clones of the tetrad) to be clbl clb2 double mutants or clb2 clb3 double mutants were extremely abnormal and were very similar to each other. About two-thirds (roughly 20 of 30 for each type) of the dissected spores predicted to be double mutants died on the dissection plates as single large budded cells. The other third went through several cell divisions, and on prolonged incubation sometimes produced microcolonies visible to the naked eye. The other four possible double mutants were viable and healthy, although the clbl clb3, clb3 clb4, and clb2 clb4 double mutants had slight phenotypes (elongated enlarged cells or increased proportion of budded cells). To examine the phenotypes of the triple mutants, we constructed each of the four possible triple heterozygotes, sporulated them, and dissected tetrads. We examined all inviable spores microscopically to determine the state of arrest of the spore or spore clone. In some tetrads, the markers in the viable spore clones allowed us to predict the genotype of the inviable spore or spores. This analysis showed that four of eight predicted clbl clb2 clb4 triple mutant spores died as single cells with a large bud, whereas the remaining four went through several cell divisions before dying. The clbl clb2 clb3 and the clb2 clb3 clb4 triple mutants had a more severe phenotype: 10/10 and 11/11, respectively, germinated and budded but died as large budded cells without going 811

I. Fitch et al.

40

20

0

A

CLN1

a'

.4w.

CLB4

w

*

160

140

120

...

.

OO*

tsasF

.- _ ..e gqu 'g4

-

*.. :l

-

CLB2

B

100

*4*4it^u4O

CLB3

ACT1

80

60

a a 4aSS 0

30

00 0 000

a

60

90

120

0,041. is..4o

180

150 _m

CLN1

*

....

240

210

..

CLB3 ,w1-~ SWO

~

_o w

_ " _ .-W

W W

.....

WmwWWW!W.

...$

dli, W

CLB4

s.,,...,

CLB2

ACT I

*

*

*

*

Figure 6. Pattern of CLB3 and CLB4 expression during the cell cycle. (A) CLB3 and CLB4 mRNA levels in cells released from a-factor arrest. The filter was hybridized sequentially to CLN1, CLB3, CLB4, and CLB2-specific probes, and also ACT) as an RNA loading control. (B) CLB3 and CLB4 mRNA levels in cells released from cdc15 arrest. The filter was probed as in A. Arrows indicate the timing of bud emergence.

through any cell divisions. The clbl clb3 clb4 triple mutant was viable, displaying a slight phenotype similar to that of the clb2 single mutant. 812

Finally, we constructed the quadruple heterozygote, and from 40 tetrads obtained four spores predicted to be clbl clb2 clb3 clb4 quadruple mutants. These four Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

CLB3,4

CLN1,2

S

Gl SPB

SPBD

SPBS

CLB1,2

M

G2

I

SE

(Short Mitotic Spindle Formation)

Figure 7. Cell cycle timing of cyclin mRNAs in Saccharomyces cerevisiae. A schematic diagram indicating the relative timing of CLN1, CLN2, CLB1, CLB2, CLB3, and CLB4 mRNAs throughout the cell cycle (based on Northern analysis). (The filters shown in Figure 6 were also probed with CLB1. Diagram reflects timing only; no attempt was made to quantitate actual mRNA levels.) Also represented is the cell cycle timing of mitotic spindle formation. SPB, spindle-pole-body; SPBD, spindle-pole-body duplication; SPBS, spindle-pole-body separation; SE, spindle elongation.

germinated and budded but failed to go through even one cell division. These results are summarized in Table 2. Terminal Arrest Phenotype of a GAL10-CLB2 clbl clb2 clb3 clb4 Strain To characterize the lethal defect of a quadruple clb mutant, we constructed a strain that was disrupted/deleted in all four CLB genes but was kept alive by a GAL10CLB2 fusion stably integrated into the genome (see Materials and Methods). This strain grew well in the presence of galactose but was inviable when transferred to glucose medium. Some of the cells grown on galactose had various aberrant morphologies. This may have been due to the expression of CLB2 at inappropriate times of the cell cycle or at inappropriate levels. When an asynchronous culture of GAL1O-CLB2 clbl clb2 clb3 clb4 cells was switched from a galactose medium to a glucose medium, cell division ceased. A large majority of the cells arrested as large budded cells with a G2 DNA content. Some cells, however, had aberrant morphologies, and some cells had a Gi DNA content. In the hope of achieving a more homogeneous terminal arrest, we synchronized GAL1O-CLB2 clbl clb2 clb3 clb4 cells by arresting them in galactose medium with a-factor. One hour after addition of a-factor, glucose was added to repress the galactose promoter, and the cells were held in glucose at the Gi arrest point for a second hour before the a-factor was removed. (This protocol was developed with the idea that Clb2 protein might be somewhat unstable even in Gi phase and so might decay during incubation in glucose medium.) On Vol. 3, July 1992

removal of a-factor, essentially all of the cells budded and arrested as large cells with a single elongated bud (Figure 8Aa). Little or no cell division occurred. DAPI staining of DNA showed that each cell had a single nucleus (Figure 8Ab) that by FACS analysis was in G2 (Figure 8B). Staining of microtubules with an anti-tubulin antibody showed that almost all cells lacked a mitotic spindle, although cytoplasmic microtubules were observed (Figure 8Ac). Electron microscopy showed that the spindle pole bodies in similar cells from an independent experiment had become duplicated but remained in the side-by-side configuration (still connected by a bridge) that precedes mitotic spindle formation (Figure 9). In this independent experiment, a few cells did possess spindles (Table 3). The samples were also tested for Clb2-associated histone Hi kinase activity after precipitation with an antiClb2 antibody. Substantial amounts of Clb2-associated histone Hi kinase activity were detected at the terminal arrest point, although the amounts were about eightfold lower than observed when the cells were growing in galactose medium. Our interpretation of these results is that Clb2 was somewhat unstable even at the a-factor arrest point and that most (but not all) of the Clb2 was degraded during the incubation in glucose medium. On release from the a-factor block, DNA synthesis occurred but no spindle formed, showing that spindle formation and entry into mitosis is Clb-dependent. DISCUSSION We have found four B-type cyclins in S. cerevisiae, giving this yeast the largest repertoire of B-type cyclins of any

Table 2. Viability of clb disruption mutants

Phenotype Viable

Genotype

clbl

clb2 clb3 clb4

clb3

clbl clbl clb2

Nearly inviable

clbl clbl

Inviable

clbl clbl clbl

clb3 clb3

clb2 clb2 clb2 clb2 clb2 clb2

clb4 clb4 clb4 clb4

clb3 clb4

clb3 clb3 clb3

clb4 clb4

Some spores classified as being nearly inviable are able to undergo several rounds of division before dying. Some are eventually able to form microcolonies.

813

I. Fitch et al. A

GALIO-CL82 c/b! c/b2 c/b3 c/b4

Wild-type S

rL

(

r

t

;

&

(a)

.;

.:,

(b)

(c)

Figure 8. (Continued)

814

Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

B

IN

(b) YEPD

(a) YEP+Gal+Raff 2N

l I. I

Figure 8. Terminal arrest phenotype of a GALIO-CLB2 clbl clb2 clb3 clb4 strain. (A) Cultures of wild-type cells and GAL1O-CLB2 clbl clb2 clb3 clb4 cells (strain #245) were released from a-factor arrest into glucose medium. Differential interference contrast (Nomarski) micrographs of cells (a), corresponding nuclei stained with DAPI (b), and anti-tubulin antibody staining (c) of samples taken at T = 3 h are shown. (B) A culture of GAL1O-CLB2 clbl clb2 clb3 clb4 cells was released from a-factor arrest, half of the culture into a medium containing galactose (a) and the other half into a medium containing glucose (b). At 40-min intervals, samples were taken and stained with propidium iodide for FACS analysis. The T = 0 min, T = 40 min, and T = 200 min samples are shown.

I

-

--b-

T =O0min

T =0 min

T =40 min

T =40min

.1Jh~~~~~~~ .. A..... *.--.^......~--.

T = 200 min

organism to date. Because there is no reason to believe our search was exhaustive, additional B-cyclins may yet be found in S. cerevisiae. By amino acid sequence, the four cyclins fall into two distinct structural groups. The two genes in each structural group share similar patterns of expression. Levels of CLB1 and CLB2 mRNA peak around the time of nuclear division (Surana et al., 1991). Levels of CLB3 and CLB4 rise much earlier, at about the beginning of S phase, and fall at about the end of anaphase. The mechanism effecting this cell-cycle regulated oscillation in mRNA abundance remains to be explored. However, we note some sequence motifs shared by the upstream regions of CLB3 and CLB4 (Figure 2). Although none of the single clb disruptions was lethal, various multiple clb disruptions did result in lethality. Although CLB3 and CLB4 are quite similar, the clb3 clb4 double disruption had only a slight phenotype. Perhaps the functions of these two genes can be fully carried out by CLB1 and CLB2. Alternatively, perhaps there is one or more additional cyclin of the Clb3/Clb4 type that has not yet been found. Functionally, the CLB2 gene seems to be especially important. Among the single disruptions, only clb2 has a discernable phenotype, and all the lethal combinations of disruptions involved clb2. Furthermore, the clbl clb3 clb4 triple mutant is quite healthy. We cannot conclude that CLB2 alone is sufficient for viability, because there may be other CLB genes that remain undetected. CLB4 appears to be the least important CLB; its disruption had no phenotype on its own and only a very slight phenotype in combination with other clb disruptions. Vol. 3, July 1992

1

-. . .~. -

a.I

T = 200 min

However, CLB4 does play an important role in meiosis (Dahmann and Futcher, unpublished results). Our evidence shows that the Clbs are needed for mitosis and are perhaps very directly involved in spindle formation. However, because the arrested Clb-deficient cells still contained some Clb2 activity, we cannot rule out the possibility that the Clbs are also needed for some other processes, such as DNA replication, and that the residual Clb2 activity was sufficient for these processes.

Spindle Formation, Nuclear Division, and the Clbs In most eukaryotes, the mitotic spindle forms only after the completion of DNA replication (Hartwell and Weinert, 1989). Spindle formation may require the activity of the Cdc2 protein kinase, which, at least in other organisms, is dependent on the completion of DNA replication. This dependence is accomplished at least partly by the regulated phosphorylation and dephosphorylation of a tyrosine in the Cdc2 ATP binding site (Gould and Nurse, 1989; Enoch and Nurse, 1990, 1991; Nurse, 1990; Smythe and Newport, 1992). In S. cerevisiae, the sequence of events appears to differ. Inhibition of DNA replication by chemical inhibitors or by cdc mutations clearly demonstrates that formation of a short mitotic spindle is not dependent on completion of DNA replication (Byers and Goetsch, 1974; Byers, 1981; Pringle and Hartwell, 1981). Activation of Cdc28 protein kinase activity is also not dependent on completion of replication (Sorger and Murray, 1992; Amon et al., 1992). On the other hand, spindle elongation and nuclear di815

I. Fitch et al.

4.. 4.

&r, '½!

K

4 4.

,t..

*

t

Figure 9. Electron microscopy of terminally arrested Clb-deficient cells. A culture of GALIOCLB2 clbl clb2 clb3 clb4 cells was released from a-factor arrest into glucose medium. At 3 h 50 min after release, cells were serially thin sectioned and analyzed by electron microscopy. Most cells were arrested with spindle pole bodies in the side-by-side configuration, which are depicted here.

vision are dependent on completion of DNA replication. Thus, the process of mitosis in S. cerevisiae can be divided into at least two series of events: short spindle formation, which occurs relatively early in the cycle and is not dependent on S phase, and nuclear division, which occurs relatively late and is dependent on S phase. The CLB3 and CLB4 transcripts appear at '-20 min after release from a-factor, whereas the CLB1 and CLB2 816

transcripts appear -40 min after the release (Figure 6) (Surana et al., 1991). Interestingly, in the same strain under the same experimental conditions, p13-associated histone Hi kinase activity appears at -30 min after release from a-factor (Surana et al., 1991). Although we do not know the molecular identities of all the p13associated kinase activities, a straightforward interpretation is that a Clb3/Clb4 associated kinase activity appears very shortly after their transcripts appear and 10 Molecular Biology of the Cell

Four B-Type Cyclin Genes of S. cerevisiae

Table 3. Developmental state of spindles in mitosis-defective yeast

GALIO-CLB2 clbl clb2 clb3 clb4 Time after afactor removal

Single SPB Satellite-bearing SPB Double SPB (side-by-side) Complete spindle Complete spindle with double SPB at one pole

cdc28-IN Time after switch to restrictive temperature

1h

3h50min

3h

1 1

22 0

0 0 30 8

0 0 0 5

0

2

0

GAL1O-CLB2 clbl clb2 clb3 clb4 cells were released from a-factor into glucose medium. At 1 h and 3 h 50 min after release, cells were serially thin sectioned and analyzed by electron microscopy. A culture of cdc28-lN cells was grown to mid-log-phase at 25°C (permissive temperature) and then shifted to 36°C (restrictive temperature). After 3 h, cells were sectioned for electron microscopy.

min before the CLB1 and CLB2 transcripts. We have not measured the time at which the short spindle forms in these strains, but based on previous work (reviewed by Byers, 1981), it seems likely that it occurs before there is significant expression of CLB1 and CLB2 (Figure 7). Thus, a simple interpretation of our data would be that the Clb3/Clb4-associated Cdc28 kinase activity is normally responsible for formation of the short spindle and Clbl and Clb2 are normally responsible for some aspect of spindle elongation. When Clb3 and Clb4 are missing, Clb2 can provide the missing function for spindle formation. The fact that cdc28-IN cells arrest with a spindle (Surana et al., 1991) (Table 3) and that cdc28-IN can be suppressed by CLB1, 2, and 4 overexpression and has a specific lethal genetic interaction with clb2 (Surana et al., 1991), is consistent with this; that is, the Clb-deficient arrest with no spindle and the cdc28-IN arrest with a short spindle may identify two of the major mitotic events orchestrated by the Clbs. If this is so, then perhaps spindle formation is delayed in clb3 clb4 double mutants until Clbl and Clb2 expression rise. Thus, the division of the S. cerevisiae B-type cyclins into two subclasses, one expressed early and one expressed late, may help explain how S. cerevisiae manages to regulate spindle formation separately from nuclear division. The separate regulation of these events may require histone Hi kinase activity before replication, explaining why it is not dependent on replication in this organism (Amon et al., 1992; Sorger and Murray, 1992).

ACKNOWLEDGMENTS We thank M. Olson and L. Riles for help in mapping the CLBs. We are very grateful to Yue Xiong for nourishing a cyclin tree and providing the B-branch carrying our pairs (Figure 5). For financial support, we thank the NIH (grant GM-45410 to B.F. and GM-18541 to B.B.),

Vol. 3, July 1992

and EMBO for a postdoctoral fellowship to I.F. The GenBank accession numbers for the CLB gene nucleotide sequences are as follows: CLB1 M65069, CLB2 M65070, CLB3 M80302, CLB4 M80303.

REFERENCES Amon, A., Surana, U., Muroff, I., and Nasmyth, K. (1992). Regulation of p34cDc28 tyrosine phosphorylation is not required for entry into mitosis in S. cerevisiae. Nature 355, 368-371. Booher, R.N., Alfa, C.E., Hyams, J.S., and Beach, D.H. (1989). The fission yeast cdc2/cdc13/sucl protein kinase: regulation of catalytic activity and nuclear localization. Cell 58, 485-497. Booher, R., and Beach, D. (1988). Involvement of cdc13+ in mitotic control in Schizosaccharomyces pombe: possible interaction of the gene product with microtubules. EMBO J. 7, 2321-2327. Bueno, A., Richardson, H., Reed, S.I., and Russell, P. (1991). A fission yeast B-type cycin functioning early in the cell cycle. Cell 66, 149159. Byers, B. (1981). Yeast cytology. In: The Molecular Biology of the Yeast Saccharomyces. Life Cycle and Inheritance, eds. J.N. Strathern, E.W. Jones, and J. Broad, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 97-142. Byers, B., and Goetsch, L. (1974). Duplication of spindle plaques and integration of the yeast cell cycle. Cold Spring Harbor Symp. Quant. Biol. 38, 123-131. Byers, B., and Goetsch, L. (1991). Preparation of yeast cells for thinsection electron microscopy. Methods Enzym. 194, 602-608. Church, G.M., and Gilbert, W. (1984). Genomic sequencing. Proc. Natl. Acad. Sci. USA 81, 1991-1995. Colasanti, J., Tyers, M., and Sundaresan, V. (1991). Isolation and characterization of cDNA clones encoding a functional p34cdc2 homologue from Zea mays. Proc. Natl. Acad. Sci. USA 88, 3377-3381. Draetta, G., and Beach, D. (1988). Activation of cdc2 protein kinase during mitosis in human cells: cell cycle-dependent phosphorylation and subunit rearrangement. Cell 54, 17-26. Draetta, G., Luca, F., Westemdorf, J., Brizuela, L., Ruderman, J., and Beach, D. (1989). cdc2 protein kinase is complexed with both cyclin A and B: evidence for proteolytic interaction of MPF. Cell 56, 829838. Enoch, T., and Nurse, P. (1990). Mutation in fission yeast cell cycle genes abolishes dependence of mitosis on DNA replication. Cell 60, 665-673. Enoch, T., and Nurse, P. (1991). Coupling M-phase and S-phase: controls maintaining dependence of mitosis on chromosome replication. Cell 65, 921-923. Evans, T., Rosenthal, E.T., Youngblom, J., Distel, D., and Hunt, T. (1983). Cyclin: a protein specified by maternal mRNA in sea urchin eggs that is destroyed at each cleavage division. Cell 33, 389-396. Gallant, P., and Nigg, E.A. (1992). Cyclin B2 undergoes cell cycledependent nuclear translocation and, when expressed as a non-destructible mutant, causes mitotic arrest in HeLa cells. J. Cell Biol. 117, 213-224. Gautier, J., Minshull, J., Lohka, M., Glotzer, M., Hunt, T., and Maller, J.L. (1990). Cyclin is a component of maturation-promoting factor from Xenopus. Cell 60, 487-494.

Ghiara, J.B., Richardson, H.E., Sugimoto, K., Henze, M., Lew, D.J., Wittenberg, C., and Reed, S.I. (1991). A cyclin B homolog in S. cerevisiae: chronic activation of the Cdc28 protein kinase by cyclin prevents exit from mitosis. Cell 65, 163-174. Glotzer, M., Murray, A.W., and Kirschner, M.W. (1991). Cyclin is degraded by the ubiquitin pathway. Nature 349, 132-138.

817

I. Fitch et al. Goebl, M., and Byers, B. (1988). Cyclin in fission yeast. Cell 54, 739740. Gould, K.L., and Nurse, P. (1989). Tyrosine phosphorylation of the fission yeast cdc2+ protein kinase regulates entry into mitosis. Nature 342, 39-45. Hadwiger, J.A., Wittenberg, C., Richardson, H.E., de Barros Lopes, M., and Reed, S.I. (1989). A family of cyclin homologs that control the Gl phase in yeast. Proc. Natl. Acad. Sci. USA 86, 6255-6259. Hagen, I.M., Hayle, J., and Nurse, P. (1988). Cloning and sequencing of the cyclin related cdc13+ gene and a cytological study of its role in fission yeast mitosis. J. Cell Sci. 91, 587-595. Hartwell, L.H., and Weinert, T.A. (1989). Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629-634. Hata, S., Kouchi, H., Suzuka, I., and Ishii, T. (1991). Isolation and characterization of cDNA clones for plant cyclins. EMBO J. 10, 26812688. Hunter, T., and Pines, J. (1991). Cyclins and cancer. Cell 66, 10711074. Jimenez, J., Alphey, L., Nurse, P., and Glover, D.M. (1990). Complementation of fission yeast cdc2' and cdc25' mutants identifies two cell cycle genes from Drosophila: a cdc2 homologue and string. EMBO J. 9, 3565-3571. Johnston, M., and Davis, R.W. (1984). Sequences that regulate the divergent GALI-GAL1O promoter in Saccharomyces cerevisiae. Mol. Cell. Biol. 4, 1440-1448. Jones, J.S., and Prakash, L. (1991). Yeast Saccharomyces cerevisiae selectable markers in pUC18 polylinkers. Yeast 6, 363-366. Lee, M.G., and Nurse, P. (1987). Complementation used to clone a human homologue of the fission yeast cell cycle control gene cdc2. Nature 327, 31-35. Lehner, C.F., and O'Farrell, P.H. (1990a). Drosophila cdc2 homologs: a functional homolog is coexpressed with a cognate variant. EMBO J. 9, 3573-3581. Lehner, C.F., and O'Farrell, P.H. (1990b). The roles of Drosophila cyclins A and B in mitotic control. Cell 61, 535-547. Minshull, J., Blow, J.J., and Hunt, T. (1989). Translation of cyclin mRNA is necessary for extracts of activated Xenopus eggs to enter mitosis. Cell 56, 947-956. Nash, R., Tokiwa, G., Anand, S., Erickson, K., and Futcher, A.B. (1988). The WHII gene of Saccharomyces cerevisiae tethers cell division to cell size and is a cyclin homolog. EMBO J. 7, 4335-4346. Nurse, P. (1990). Universal control mechanism regulating onset of M-phase. Nature 344, 503-508. Nurse, P., and Bissett, Y. (1981). Gene required for commitment to cell cycle and in G2 for control of mitosis in fission yeast. Nature 292, 448-460.

818

Piggott, J.R., Rai, R., and Carter, B.L.A. (1982). A bifunctional gene product involved in two phases of the yeast cell cycle. Nature 298, 391-393. Pines, J., and Hunter, T. (1987). Molecular cloning and characterization of the mRNA for cyclin from sea urchin eggs. EMBO J. 6, 2987-2995. Pines, J., and Hunter, T. (1989). Isolation of a human cyclin cDNA: evidence for cyclin mRNA and protein regulation in the cell cycle and interaction with p34CdC2. Cell 58, 833-846. Pringle, J.R., and Hartwell, L.H. (1981). The Saccharomyces cerevisiae cell cycle. In: The Molecular Biology of the Yeast Saccharomyces. Life Cycle and Inheritance, eds. J.N. Strathem, E.W. Jones, and J. Broad, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 97142. Reed, S.I., and Wittenberg, C. (1990). Mitotic role for the Cdc28 protein kinase of S. cerevisiae. Proc. Natl. Acad. Sci. USA 87, 5697-5701. Richardson, H.E., Wittenberg, C., Cross, F., and Reed, S.I. (1989). An essential Gl function for cyclin-like proteins in yeast. Cell 59, 11271133. Rose, M.D., Winston, F., and Hieter, P. (1990). Methods in Yeast Genetics: A Laboratory Course Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Saitou, N., and Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406425. Sanger, F., Nicklen, S., and Coulson, A.R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 54635467. Smythe, C., and Newport, J.W. (1992). Coupling of mitosis to the completion of S-phase in Xenopus occurs via modulation of the tyrosine kinase that phosphorylates p34cdc2. Cell 68, 787-797. Solomon, M., Booher, R., Kirschner, M., and Beach, D. (1988). Cyclin in fission yeast. Cell 54, 738-739. Sorger, P.K., and Murray, A.W. (1992). S-phase feedback control in budding yeast independent of tyrosine phosphorylation of p34cdc28. Nature 355, 365-368. Surana, U., Robitsch, H., Price, C., Schuster, T., Fitch, I., Futcher, A.B., and Nasmyth, K. (1991). The role of CDC28 and cyclins during mitosis in the budding yeast S. cerevisiae. Cell 65, 145-161. Swenson, K.I., Farrell, K.M., and Ruderman, J.V. (1986). The clam embryo protein cyclin A induces entry into M phase and the resumption of meiosis in Xenopus oocytes. Cell 47, 861-870. Tachibana, K., Ishiura, M., Uchida, T., and Kishimoto, T. (1990). The starfish egg mRNA responsible for meiosis reinitiation encodes cyclin. Dev. Biol. 140, 241-252. van Loon, A.E., Colas, P., Goedemans, H.J., Neant, I., Dalbon, P., and Guerrier, P. (1991). The role of cyclins in the maturation of Patella vulgata. EMBO J. 10, 3343-3349. Xiong, Y., Connolly, T., Futcher, B., and Beach, D. (1991). Human Dtype cyclin. Cell 65, 691-699.

Molecular Biology of the Cell