Preliminary characterization of maturation-promoting factor from yeast

Saccharomyces cerevisiae*

KAZUNORI TACHIBANA'2, NAOHIKO YANAGISHIMA2 and TAKEO KISHIMOTO'f1 Department of Developmental Biology, National Institute for Basic Biology, Okazaki 444, Japan^Department of Biology, Faculty of Science, Nagoya University, Chikusa, Nagoya 464, Japan

•This article is dedicated to the late Professor Naohiko Yanagishima, who died on March 28, 1987•f Author for correspondence

Summary

It has been known for some time that maturation-promoting factor (MPF) appears in a wide varietyof eukaryotic cells at M phase and exerts equal M-phase-promoting activity in both meiotic cellsand mitotic cells in a non-specific manner. MPFwas extracted from cdc20 mutant cells of the yeastSaccharomyces cerevisiae synchronized at Mphase by incubation at the restrictive tempera-ture. When injected into immature oocytes ofXenopus laevis, yeast MPF caused meiosis re-initiation in a dose-dependent manner and evenin the presence of cycloheximide. Yeast MPFexerted its activity in starfish oocytes as well. MPFactivity was obtained only from cells in M phaseand not from G|, S or G2 phase cells, indicatingcyclical changes during the yeast mitotic cellcycle. Preliminary characterization of yeast MPFrevealed that its activity was associated with aheat-labile protein having a sedimentation coef-ficient value of 6S. In contrast to the current

assumption that MPF is a Ca-sensitive phospho-protein stabilized by phosphorylated small mol-ecules, such as ATP and Na-jS-glycerophosphate,the present study revealed that yeast MPF wasstill active even after treatment with either Caz+

or alkaline phosphatase. Furthermore, it wasfound that yeast MPF and these phosphorylatedsmall molecules were complementary in in-ducing reinitiation of meiosis, since the meiosis-reinitiating activity was detected only when bothwere present simultaneously and almost un-detectable when either of them was present alone.These facts suggest that yeast MPF need notnecessarily be in a phosphorylated form in orderto be active and that the phosphorylated smallmolecules have another effect, such as preventingthe activity of phosphatases that might be presentin recipient Xenopus oocytes rather than stabiliz-ing the MPF molecule itself.

Key words: yeast, S. cerevisiae, maturation-promotingfactor, M phase, oocyte maturation, phosphorylation.

Introduction

The regulatory mechanisms controlling the initiationand completion of mitosis have been studied in a widespectrum of eukaryotic cells. In the yeasts Saccharo-myces cerevisiae and Schizosaccharomyces pombe,extensive genetic analysis has been performed to inves-tigate the regulation of the cell cycle, resulting in theisolation of a large number of conditional lethal cellcycle mutants (Pringle & Hartwell, 1981). Study ofsuch cell division cycle (ede) mutants has identifiedmany genes that have a role in cell cycle control. These

Journal of Cell Science 88, 273-281 (1987)Printed in Great Britain © The Company of Biologists Limited 1987

classical genetic approaches have recently been sup-plemented with the techniques of molecular biology byisolating and characterizing such genes (Russell &Nurse, 1986; Hayles & Nurse, 1986). However, inyeast little is known about the biological function ofeach gene product. This may be due to the lack of abiological assay system for detecting its regulatoryactivity during the cell cycle.

In contrast to yeast, amphibian and starfish oocyteshave provided some clues toward elucidation of thebiochemical mechanism underlying the regulation of Mphase, although genetic approaches are lacking. In

273

these animals, fully grown ovarian oocytes have a largenucleus called the germinal vesicle. Such immatureoocytes are arrested at prophase of the first meioticdivision. The reinitiation of meiosis and subsequentprocesses are called oocyte maturation or meioticmaturation and involve breakdown of the germinalvesicle (GVBD), chromosome condensation and pro-gression through meiotic division. The onset of oocytematuration is triggered by maturation-inducing hor-mones, progesterone in amphibians (Masui, 1967;Masui & Clarke, 1979) and 1-methyladenine in starfish(Kanatani el al. 1969; Kanatani, 1985), released fromthe follicle cells around the oocytes. Once these hor-mones act on the oocyte surface (Kanatani & Hira-moto, 1970; Smith & Ecker, 1971) they induce theproduction of a cytoplasmic maturation-promotingfactor (MPF), which in turn brings about GVBD andsubsequent meiotic maturation (Masui & Markert,1971; Kishimoto & Kanatani, 1976).

While each maturation-inducing hormone is specificfor each kind of animal, MPF activity has been foundnot only in maturing oocytes of various animals (Kishi-moto et al. 19846), but also in mitotically dividing cellssuch as cleaving blastomeres of starfish (Kishimoto etal. 1982), and amphibian (Wasserman & Smith, 1978)and mammalian cultured cells synchronized at M phase(Sunkara et al. 1979a; Kishimoto et al. 1982). MPFderived from this wide variety of eukaryotic cells exertsequal maturation-inducing activity when injected intointact immature oocytes of starfish or amphibians.During the cell cycle, MPF activity oscillates with apeak at each mitotic M phase (Wasserman & Smith,1978; Sunkara et al. 1979a; Kishimoto et al. 1982;Gerhart el al. 1984) or at meiotic metaphase (Kishi-moto & Kanatani, 1976; Doree et al. 1983; Gerhart etal. 1984), while it is not detected at interphase.Furthermore, it has been shown that MPF inducesnuclear membrane breakdown and chromosome con-densation in amphibian blastomeres (Miake-Lye et al.1983). Thus, MPF acts in a non-species-specific man-ner across different phyla playing a key role in promot-ing M phase within the cytoplasm during both meiosisand mitosis (Kishimoto, 19866). MPF has been par-tially purified and characterized preliminarily fromstarfish oocytes (Kishimoto & Kondo, 1986), amphib-ian oocytes (Wasserman & Masui, 1976; Drury, 1978;Wu & Gerhart, 1980; Nguyen-Gia et al. 1986) andHeLa cells (Sunkara etal. 19796; Adlakhae/a/. 1985);MPF from various origins behaves as a protein with amolecular size of approximately 5 S.

Considering the ubiquitous presence of MPF, it isreasonable to suggest that yeast may have a factorsimilar to MPF, which regulates M phase during theyeast cell cycle. Indeed, Weintraub et al. (1982) haveshown that MPF-like activity is present in extracts ofsome cclc mutants of S. cerevisiae as assayed by

injection into immature amphibian oocytes. In thispaper, we further developed their study and present apreliminary characterization of the yeast MPF. Theresults show that yeast MPF is a heat-labile proteinwith a sedimentation coefficient value of 6 S and that itsactivity is insensitive to Ca2+ and alkaline phosphatase.The role of protein phosphorylation on the activity ofMPF is discussed.

Materials and methods

Yeast strain and cell synchronyTemperature-sensitive cdc20 mutant cells of the yeast Sac-charomyces cerevisiae (strain 127; MATa, adel, ade2, ural,his7, Iys2, tyrl, gall), a gift from Dr Oosumi of theUniversity of Tokyo, were cultured in YEPD completemedium (20gl~' glucose, 20 gI"1 polypeptone, 10 g I"1 Difcoyeast extract) with continuous agitation. These cdc20 cellswere grown at 24°C (the permissive temperature) for 16-20 huntil mid-log phase (cell concentration, 1 X 107cellsml~'),and then the incubation temperature was shifted up to 37°C(the restrictive temperature) to block the completion ofnuclear division. Since these cells have a cell cycle time of 4 h,almost 100 % of the cells were synchronized at 1VI phase after4-5 h incubation at the restrictive temperature.

Preparation of yeast extractsA 3-1 culture of cdc20 cells (3x 10lu cells) synchronized at Mphase was harvested, chilled on ice water, and washed threetimes with cold extraction buffer (buffer A) containingISmM-NaCI, 60mM-KCl, 10mM-MgCl2, SmM-EGTA,50mM-Na-/S-glycerophosphate, 1 mM-ATP, 1 mM-dithioth-reitol (DTT), 1 mM-phenylmethylsulphonyl fluoride(PMSF) and Olmgml" 1 leupeptin (Peptide Institute,Osaka) at a final pH of 7-0. All subsequent procedures wereperformed at 2°C. The cells collected after centrifugation at3000revsmin~' for 8min (cell pack, 3 ml) were suspendedwith 1-5 ml of buffer A in a 12ml polycarbonate centrifugetube, and then disrupted in the presence of 12g of glass beads(diameter, 0-5 mm) by continuous shaking with a vortexmixer for 20min. For small-scale extraction, 1X109 cdc20cells were disrupted in an Eppendorf micro test tube usingsimilar techniques. The homogenate was centrifuged at105 000 £ for 30min. The supernatant was recovered andstored at —80°C, or immediately processed for assay of itsmeiosis-reinitiating activity in immature oocytes of Xenopusor starfish. The protein concentration in such crude extractswas approx. 50mgml~'.

For preliminary characterization of the meiosis-reinitiatingactivity present in the yeast M phase extracts, the crudeextract was treated with Ca2+ as follows. The crude extractwas dialysed for 2 h (Spectrapor 2 dialysis membrane) againstbuffer B lacking EGTA and EDTA but supplemented with2mM-CaCl2, while complete buffer B contained 10 mM-Na2HPO4/NaH2PO4, 10mM-MgCl2, 2mM-EGTA, I mM-EDTA, 5mM-Na-/3-glycerophosphate, 1 mM-ATP, 1 mM-DTT, lmM-PMSF and 5% glycerol at a final pH of 6-5.After centrifugation at IOOOOJJ for 15min to remove insol-uble materials, the supernatant was again dialysed for 2 h

274 K. Tachibana et al.

against complete buffer B, followed by centrifugation. Thefinal supernatant is referred to as a Ca-treated extract in thetext. The second dialysis, which removed Ca2+ again fromthe yeast preparation, enabled the Ca-treated extract todisperse freely in recipient oocytes when its activity wasassayed by microinjection. The protein concentration of theCa-treated extract was approx. ISmgml"1.

To examine the effects of treatment with protease oralkaline phosphatase, the M phase extract was exposedto carboxymethylcellulose-bound protease (Sigma, l'Ounitml"1) or beaded agarose-bound alkaline phosphatase (Sigma,200unitsml~') for l h at 25°C with continuous shaking,followed by low-speed centrifugation to remove the particle-bound enzymes. The supernatants were assayed for meiosis-reinitiating activity. As a control, non-treated yeast extractswere incubated in a similar manner. The activity of theinsoluble alkaline phosphatase was verified by its ability todephosphorylate^-nitrophenyl phosphate. With 5 units ml"1

in buffer B lacking ATP and Na-/3-glycerophosphate, dephos-phorylation occurred almost completely at 25 CC within 1 minwhen measured as the increase in absorbance at 410 nm.

Protein content was determined by the method of Bradford(1976) using bovine serum albumin as a standard.

Preparation of immature oocytes from Xenopus or

starfish

Xenopus laevis, obtained from a dealer in Hamamatsu, wasmaintained in laboratory aquaria at 20cC. Ovaries wereremoved surgically after anaesthesia by hypothermia andtransferred into Ca-free OR-2 medium (Wallace et al. 1973).After rinsing, ovarian tissue was cut into small pieces andthen treated with 0'2% collagenase (Sigma, type I) in Ca-freeOR-2, while agitating for 1 h by hand. This treatment freedeach oocyte of surrounding follicle cells (Hirai et al. 1983).The isolated oocytes were washed thoroughly with normalOR-2. Stage VI oocytes (Dumont, 1972) were collected afterbeing allowed to recover for more than 3 h at 20°C.

The starfish, Asterina pectmifera, were collected duringthe breeding season at Ise (Mie). The animals were kept inlaboratory aquaria and supplied with circulating cold seawater (14°C). Fully grown immature oocytes with follicleswere obtained by tearing the isolated ovaries with fineforceps. After the removal of follicles in Ca-free sea water,immature oocytes were placed in modified van't Hoff'sartificial sea water (Kishimoto et al. 1984a).

Assay for meiosis-reinitiating activityYeast extracts were assayed for meiosis-reinitiating activity bymicroinjection of 50 nl into each immature Xenopus oocyte,or 800 pi into each immature starfish oocyte. Microinjectionwas performed according to the method of Hiramoto (Hira-moto, 1974; Kishimoto, 1986a). In Xenopus, GVBD wasdetected by a white spot appearing at the animal pole within6h after the injection. The presence or absence of thegerminal vesicle was verified by dissection of the oocytes afterboiling. In starfish, the recipient oocytes were inspected forGVBD, 1 h after the injection. Since Asterina oocytes aretransparent, intact germinal vesicles and GVBD were bothclearly observable under a microscope. In both Xenopus andstarfish, no GVBD was observed after control injection ofbuffer A or B.

Visualization of chromosomes in Xenopus oocytesAccording to the method of Gerhart et al. (1984), chromo-somes in Xenopus oocytes were observed by staining with afluorescent dye, 4',6-diamino-2-phenylindole (DAPI). Theoocytes injected with yeast extracts were fixed in 10%formaldehyde in 006M-Hepes, pH7-4 for 2h, and thenDAPI was added to a concentration of SOngml"1. Afterstaining for 1 h, the oocyte was placed in solution (buffer C)containing 50% glycerol, 0-1 M-Tris-HC1, pH7-S, 2% n-propyl gallate and 10% ethanol, and a patch containing thewhite spot was dissected with a razor. The patch wasmounted in a drop of the same medium on a glass slide with acoverslip. In the case of immature oocytes, the formaldehydeconcentration was lowered to 1 % and fixation was performedfor 20 s (Miake-Lye et al. 1983). The oocytes were immedi-ately transferred to buffer C supplemented with 50ngml~'DAPI and then lysed with fine forceps. After removingexcess yolk, the isolated germinal vesicles were mounted on aglass slide. The DAPI-stained chromosomes were visualizedby epifluorescent illumination.

Sucrose density gradient centrifugation

Linear gradients of 5 % to 20% sucrose (5 ml) were made inbuffer B. The Ca-treated yeast extract (16ml) was concen-trated to 04ml (protein concentration, 60mgml~') in anAmicon ultrafiltration cell equipped with a PM-10 filtermembrane. The concentrated preparation was centrifuged at10 000g for 15 min. The supernatant (0-25 ml) was layered onthe gradient and centrifuged at 186000 £ for 12h at 2°C(Kishimoto & Kondo, 1986). Fractions of 022 ml each werecollected from the bottom using a peristalic pump. Eachfraction was assayed for meiosis-reinitiating activity by inject-ing 50 nl into each of 10 Xenopus oocytes. The standardreference proteins used (Pharmacia) were catalase frombovine liver, aldolase from rabbit muscle and bovine serumalbumin.

Results

Meiosis reinitiation in Xenopus oocytes by injection ofyeast extracts

When the crude extract prepared from cdc20 cellssynchronized at M phase was injected into immatureXenopus oocytes, the recipient oocytes underwentGVBD as reported previously by Weintraub et al.(1982). The activity of the M phase extract was foundto be dependent on the amount of the protein injected(Fig. 1). At a protein concentration of approximately5 0 m g m r ' , almost 100% GVBD occurred within 2 h,while no GVBD was observed at protein concen-trations of approximately 6mgml~' . After GVBD,however, almost all of the oocytes became discolouredand showed cytolysis within 4h after the injection ofthe crude extract, making further studies of meioticmaturation impossible, while the control injection ofextraction buffer alone had no effect. Cytologicalexamination revealed that chromosome condensationoccurred in the recipient oocytes with GVBD following

Yeast maturation-promoting factor 275

100 r

Qm

O

50

I M \ Gl S \G2 \M\ Gl

0 25 50Protein concentration (mgrnl"1)

Fig. 1. Meiosis reinitiation in Xenopus oocytes induced byinjection of M phase extract from the yeast 5. cerevisiae.The yeast M phase extract was obtained from cdc20mutant cells synchronized at M phase after incubation at37°C for 4-5 h. The extracts, non-diluted and diluted to1/2, 1/4 and 1/8 with buffer A, were assayed, respectively,for meiosis-reinitiating activity by injecting 50 nl into eachof 10 Xenopus oocytes.

GO © ©> G D Q O ©100

Q

o

50

Fig. 2. Chromosome condensation in Xenopus oocytesinduced by injection of the M phase extract from the yeastS. cerevisiae. Chromosomes were stained with afluorescent dye, DAPI, as described in Materials andmethods. A. Control preparation of germinal vesicle inimmature oocyte injected with buffer A alone. Dispersedchromatin is observed in the germinal vesicle. Bar, 40 fim.B. Condensed chromosomes in an oocyte with GVBD 2hafter injection with the yeast M phase extract. Bar, Zflm.

the injection of crude yeast extract (Fig. 2). Thus, Mphase extract of cdc20 cells induced the reinitiation ofmeiosis in Xenopus oocytes, and the observed GVBDmight not be a simple side effect of the progresstowards cytolysis, although the M phase extract islikely to contain materials deleterious to oocytes.

Since it had been known that protein synthesisinhibition blocked GVBD in Xenopus oocytes exposedto progesterone but not in those injected with MPF(Wasserman & Masui, 1975; Mailer, 1985), we exam-ined whether the yeast M phase extract caused GVBDin Xenopus oocytes in the absence of protein synthesis.Immature Xenopus oocytes were preincubated for 2h

\-Cl-Cr--fy 100

50 £

1OQ

0 1 2 3 4 5Time after temperature shift-down (h)

Fig. 3. Cycling of the meiosis-reinitiating activity duringthe mitotic cycle of the yeast 5. cerevisiae. The cdc20mutant cells were synchronized at M phase by atemperature shift up to 37°C, followed by a shift down to24°C to start and advance the mitotic cycle synchronously.The changes in meiosis-reinitiating activity ( • ) weremonitored at 15-min or 30-min intervals. Cell cycle phase,which is indicated in the upper row, was monitored byobserving the shape of the bud in each cell (A) .

in OR-2 containing cycloheximide at a concentration of100/igml~ (known to inhibit protein synthesis inamphibian oocytes; Wasserman & Masui, 1975). Whenthese oocytes were injected with the crude yeast extractand further treated with the same concentration ofcycloheximide, GVBD was invariably induced (10 outof 10 recipients) as in non-treated oocytes. Thus,GVBD induction by the yeast M phase extract does notrequire synthesis of new protein.

Cycling of meiosis-reinitiating activity during yeastmitotic cycle

To monitor the changes of meiosis-reinitiating activitycontained in the yeast during its cell cycle, 5X10'°cdc20 cells were synchronized at M phase by shiftingthe incubation temperature up to 37°C for 4-5 hfollowed by a shift down to 24°C to start and advancethe mitotic cycle synchronously. At 15-min or 30-minintervals during the subsequent 5h, 1X109 cells wereharvested, respectively, for extraction and assay ofmeiosis-reinitiating activity. According to Pringle &Hartwell (1981), each period of cell harvest was corre-lated with the cell cycle phase by examining budformation under the light microscopy. As shown inFig. 3, meiosis-reinitiating activity, which had beenmaintained at elevated levels at the restrictive tempera-ture, dropped to undetectable levels within 30 min afterthe temperature shift-down. After this time, extractsobtained at Gi, S and G2 phases had no GVBD-inducing activity. The activity reappeared when the

276 K. Tacliibana et al.

Table 1. Effect of temperature on the meiosis-reinitiating activity contained in the M phase extract

of the yeast S. cerevisiae

Incubation

Temperature (°C) Duration (h) i GVBD in recipients*

0252537

414

0-5

100754320

(10/10)t(9/12)(6/14)(2/10)

* Each preparation was assayed for meiosis-reinitiating activityby injecting 50 nl into immature Xenopus oocytes.

•fGVBD/total number of oocytes.

cell cycle reached the next M phase. Thereafter, itdropped again at the next G\ phase. Thus, meiosis-reinitiating activity was present only in M phase cells,indicating cyclical changes in its level during the yeastmitotic cycle.

Characterization of meiosis-reinitiating activitycontained in yeast extractsA preliminary characterization was carried out on themeiosis-reinitiating activity contained in M phase ex-tracts of cdc20 cells. The activity in the crude extractwas stable for months when stored at — 80°C, and for atleast 4h at 0°C. However, it was unstable at highertemperatures (Table 1). At 25°C, the activity wasreduced by more than 50% in 4h, although a l hincubation had a slight effect. At 37°C, most of theactivity was lost within 30min. When the crude extractwas dialysed for 2 h against buffer B lacking PMSF andthen treated with insoluble protease, the activity wasalmost completely lost (10% GVBD, 1 out of 10recipients), whereas control preparations retained theactivity (90% GVBD, 9 out of 10 recipients). Thus,the meiosis-reinitiating activity in the yeast M phaseextract is associated with a heat-labile protein.

To examine the effect of Ca2+ on the meiosis-reinitiating activity, the crude extract was treated withCaCl2- As shown in Table 2, no significant loss ofactivity was observed even when the Ca-treated extractwas injected into immature Xenopus oocytes; thus,indicating that the activity contained in the yeast Mphase extract is not sensitive to Ca + . Furthermore,the Ca-treated extract induced GVBD upon injectioninto immature starfish oocytes as well (80% GVBD, 8out of 10 recipients). Since the injection of the crudeextract untreated with Ca2+ caused cytolysis beforethe occurrence of GVBD in starfish oocytes, anydeleterious materials might be removed during Ca2+

treatment. Thus, the yeast M phase extract exertsmeiosis-reinitiating activity in immature oocytes ofboth amphibian and starfish.

Table 2. Effect ofCa2+, phosphorylated smallmolecules and alkaline phosphatase on the meiosis-

reinitiating activity contained in the M phase extractof the yeast S. cerevisiae

Injected material

Crude extractCa-treated extract*Ca-trcated extract - (ATP+GP)fCa-treatcd extract- (ATP+GP)

+ (ATP+GP)|Ca-treated extract - (ATP+GP)

+APase§Ca-treated extract - (ATP+GP)

+ APase+(ATP+GP)HComplete buffer B

% GVBD

90701592

11

91

0

in recipients

(9/lO)||(7/10)(4/26)

(11/12)

(2/18)

(20/22)

(0/20)

*Thc crude extract obtained from cdc20 mutant cellssynchronized at M phase was dialysed against buffer B lackingEGTA and EDTA, but supplemented with 2mM-CaCl2, followedby further dialysis against complete buffer B.

I Ca-treated extract was dialysed against buffer B lacking ATPand Na-/3-glyccrophosphate (GP).

\ The •)• preparation was further dialysed against completebuffer B.

§ The f preparation was treated with insoluble alkalinephosphatase (APase).

V The § preparation was dialysed against complete buffer B.||GVBD/total number of Xenopus oocytes after the injection of

50 nl each.

Since it has been reported that phosphorylated smallmolecules such as ATP and Na-/J-glycerophosphatc,which may act to inhibit phosphoprotein phosphatases,stabilize the MPF activity from starfish oocytes (Kishi-moto & Kondo, 1986), amphibian oocytes (Drury,1978; Wu & Gerhart, 1980; Hermann et al. 1983) andCHO cells (Nelkin et al. 1980), the effect of thepresence of phosphorylated groups on the meiosis-reinitiating activity obtained from yeast was examined.The Ca-treated extract was dialysed against buffer Blacking Na-/J-glycerophosphate and ATP for 2h at 0°Cto remove these phosphorylated small molecules. Thispreparation was then treated with insoluble alkalinephosphatase. When both preparations before and aftertreatment with alkaline phosphatase were injected intoimmature Xenopus oocytes, the meiosis-reinitiatingactivity was almost undetectable (Table 2). However,it was found that the meiosis-reinitiating activity ofboth preparations was restored if the)1 were againdialysed against complete buffer B containing ATP andNa-/3-glycerophosphate, while buffer B alone had noGVBD-inducing activity (Table 2). Thus, it appearsthat the yeast M phase extract and phosphorylatedcompounds have a complementary effect on reinitiat-ing meiosis in Xenopus oocytes. Furthermore, the factthat treatment by alkaline phosphatase did not destroythe meiosis-reinitiating activity (Table 2) suggestsstrongly that the yeast protein associated with the

Yeast maturation-pivmoting factor 277

100 i—

80

O60

Q03

O 40

20

20

10

M 7

l1-

10Fraction number

20

meiosis-reinitiating activity need not necessarily be in aphosphorylated form. In this connection, however, thepossibility cannot be excluded that some kinases, whichmay be contained in the Ca-treated extract, are respon-sible for the rephosphorylation and reactivation of thealkaline phosphatase-treated yeast protein dialysedagainst buffer B.

To determine the molecular size of the protein thathas the meiosis-reinitiating activity, the Ca-treatedextract was concentrated and centnfuged on lineargradients of 5 % to 20% sucrose. As shown in Fig. 4,the activity was observed in a single distinct peak aftercentrifugation at 186 000g for 12 h. The sedimentationcoefficient (s) value of this fraction was estimated to beapproximately 6S, as compared with marker proteinscentrifuged at the same time.

Discussion

The present study clearly demonstrates that the extractobtained at M phase from cdc20 mutant cells of theyeast, S. cerevisiae, has an activity that inducesreinitiation of meiosis in immature oocytes of amphib-ian and starfish. Although meiotic maturation is notcompleted after GVBD, the activity contained in theyeast M phase extract qualifies as a maturation-promot-ing factor (MPF) on the following counts.

(1) Yeast M phase extract causes Xeuopus oocytes toundertake apparently normal meiotic pro-metaphaseevents including GVBD and chromosome conden-sation. Further, GVBD occurs precociously whencompared with progesterone-induced maturation,which requires more than 4h.

(2) Yeast M phase extract causes GVBD mXenopusoocytes that are exposed to cycloheximide.

4 EDO

3-1

20.

Fig. 4. Sucrose density gradientcentrifugation of the M phase extractfrom the yeast S. cerevisiae. The Ca-treated extract of the cdc20 cells wasconcentrated and layered on a 5-ml linearsucrose gradient (5% to 20%), followedby centrifugation at 186000gfor 12 h.Each fraction was assayed for meiosis-reinitiating activity by injecting intoXeuopus oocytes (O), and processed forthe determination of protein content ( • ) .The reference standard proteins (A)used were: 1, catalase, 11-3S; 2,aldolase, 7 '4S; and 3, albumin, 4 5 S.Although marker proteins and the yeastM phase extract were centrifuged onseparate gradients in this experiment, themeiosis-reinitiating activity migratedsimilarly even if they were co-centrifugedon a single gradient.

(3) The meiosis-reinitiating activity is detected onlyduring M phase of the mitotic cycle of yeast.

(4) The meiosis-reinitiating activity contained in theyeast is associated with a heat-labile protein.

(5) While only the cdc20 mutant was used in thepresent study, meiosis-reinitiating activity has beenshown also in extracts from cdd6 mutant cellssynchronized at M phase (Weintraubef al. 1982). Thisindicates that the activity is generally present in yeast atM phase. Furthermore, the present yeast M phaseextract causes GVBD in oocytes of starfish as well asamphibians. Such phylogenetically ubiquitous meiosis-reinitiating activity distinguishes MPF from othersubstances that have maturation-inducing activity.

The apparent molecular size of MPF, on the basis ofsucrose density gradient centrifugation of crude ex-tracts is 5 S in starfish oocytes (Kishimoto & Kondo,1986), 4S, 15 S and 32 S in amphibian oocytes (Was-serman & Masui, 1976), 4-5 S in HeLa cells (Sunkaraet al. 19796) and 6S in yeast (present study). Thus,approximately 5 S appears to be a common unit formfor various MPF molecules.

The present result demonstrates that the yeast MPFextracted from cdc20 cells is not inactivated even in thepresence of 2mM-CaCl2- In contrast, it has beenassumed that MPF is inactivated by Caz +, since it isrequired to chelate Ca2+ with EGTA for successfulextraction of MPF from various kind of cells, includingstarfish oocytes (Kishimoto & Kondo, 1986), amphib-ian oocytes (Wasserman & Masui, 1976; Drury, 1978;Wu & Gerhart, 1980; Nguyen-Gia el al. 1986) andmammalian cultured cells (Sunkara et al. \979o,b;Nelkin et al. 1980). However, when compared withcrude extracts, Ca2+ treatment of partially purifiedMPF preparations of amphibian oocytes has beenreported to have either a moderate but not a striking

278 K. Tachibana et al.

effect (Wu & Gerhart, 1980) or no effect (Hermann etal. 1983) on MPF activity, as if a Ca2+-dependentinhibitor of MPF is removed following fractionation.In addition, it has been shown recently that theelevated levels of MPF in starfish oocytes do notdiminish even after certain activation treatments, in-cluding addition of the Ca ionophore A23187, insemi-nation and Ca-EGTA microinjection (Capony et al.1986). Considering that such activation causes a releaseof free Ca2+ in the oocyte cytoplasm in vivo (Eisen &Reynolds, 1984), Ca2+ may not directly affect theactivity of MPF. Taken together, it is most likely inother cell types, as in yeast, that MPF is inactivated notdirectly by Ca2+ itself but by some unknown Ca2+-dependent inactivation system retained in the crudeMPF preparations.

During both meiotic maturation of starfish oocytes(Doreeet al. 1983) and the mitotic cycle in embryos ofstarfish (Picard et al. 1987) and amphibia (Capony etal. 1986; Karsenti et al. 1987), a good correlation hasbeen observed between changes in MPF activity andthe level of protein phosphorylation. On the otherhand, it has been known that MPF activity is retainedin the presence of phosphorylated small molecules suchas ATP and Na-£J-glycerophosphate, which may act toinhibit phosphatases (Drury, 1978; Nelkine/ al. 1980;Wu & Gerhart, 1980; Hermann et al. 1983; Kishimoto& Kondo, 1986; Nguyen-Gia et al. 1986). From thesefacts, it has been assumed that MPF is a phosphopro-tein and that the phosphorylated form is required forthe MPF activity. In contrast to these assumptions, thepresent result demonstrates that'the yeast MPF is stillactive even after treatment with alkaline phosphatase.Similarly, it has been reported that the MPF activityobtained from HeLa cells is not affected by treatmentwith alkaline phosphatase (Adlakha e< al. 1985). Theseresults suggest strongly that active MPF need notnecessarily be a phosphoprotein. In addition, thepresent study reveals that either the phosphorylatedsmall molecules alone or the yeast MPF preparationalone is not sufficient to exert meiosis-reinitiatingactivity, while the activity is obtained if both arepresent simultaneously. Thus, phosphorylated smallmolecules and yeast MPF are complementary inexhibiting meiosis-reinitiating activity. This factsuggests that phosphorylated small molecules haveanother effect in the recipient amphibian oocytes, evenif they have a stabilizing effect on the MPF activityitself. In this connection, low molecular weight phos-phoesters, such as 2-glycerophophate and a'-naphthyl-phosphate, have been reported to induce maturation inoocytes of Xenopus (Hermann et al. 1984; Belle et al.1986) and limited species of starfish (Pondaven &Meijer, 1986), although such an effect was not ob-served in the present study. Those previous reportssuggest that the inhibition of phosphatases present in

immature oocytes affects other proteins that must bephosphorylated for GVBD to occur. In accordancewith this suggestion, it has been shown that injection ofalkaline phosphatase prevents MPF-induced GVBD inamphibian oocytes (Hermann et al. 1984; Adlakha etal. 1985). Taken together, it is most likely that phos-phatase inhibitors potentiate the activity of MPF not bypreventing the dephosphorylation of the MPF mol-ecule itself but by counteracting the action of phospha-tases in the recipient oocytes, which may be antagon-istic to the action of MPF. Further work should bedirected towards the isolation and characterization ofMPF.

We thank Dr Yoshinori Oosumi of the University of Tokyofor supplying cdc20 mutant; Dr Tadashi Uemura of KyotoUniversity for valuable advice; Dr Robert Belle of UniversitePierre et Marie Curie and Dr Marino Buscaglia of INSERMU33 for valuable discussion; Dr Tetsuo Toraya of KyotoUniversity and Dr Yoshitaka Nagahama for kind support;and Dr Michael Hanna of Rensselaer Polytechnic Institutefor reading the manuscript. This work was supported in partby grants-in-aid from the Ministry of Education, Science andCulture of Japan to N.Y. and T.K.

References

ADLAKHA, R. C , WRIGHT, D. A., SAHASKABUDDHE, C.

G., DAVIS, F. M., PRASHAD, N., BIGO, H. & RAO, P.

N. (198S). Partial purification and characterization ofmitotic factors from HeLa cells, lixpl Cell Res. 160,471-482.

BELLE, R., MULNER-LORILLON, ()., MAROT, J. & OZON, R.

(1986). A possible role for Mg2+ ions in the induction ofmeiotic maturation of Xenopus oocyte. Cell Differ. 19,253-261.

BRADFORD, M. M. (1976). A rapid and sensitive methodfor the quantification of microgram quantities of proteinutilizing the principle of protein-dye binding. Analyt.Biochem. 11, 248-254.

CAPONY, J.-P., PICARD, A., PEAUCELLIER, G., LABBE, J.-

C. & DOREE, M. (1986). Changes in the activity of thematuration-promoting factor during meiotic maturationand following activation of amphibian and starfishoocytes: Their correlations with protein phosphorylation.Dei-I Biol. 117, 1-12.

DOREE, M., PEAUCELLIER, G. & PICARD, A. (1983).

Activity of the maturation-promoting factor and theextent of protein phosphorylation oscillatesimultaneously during meiotic maturation of starfishoocytes. Devi Biol. 99, 489-501.

DRURY, K. (1978). Method for the preparation of activematuration-promoting factor (MPF) from in vitromatured oocytes of Xenopus laevis. Differentiation 10,181-186.

DUMONT, J. N. (1972). Oogenesis in Xenopus laefis.I. Stages of oocyte development in laboratory maintainedanimals. J. Morph. 136, 153-180.

Yeast maturation-promoting factor 279

EISEN, A. & REYNOLDS, G. T. (1984). Calcium transientsduring early development in single starfish (Asteriasforbesi) oocytes. J. Cell Biol. 99, 1878-1882.

GERHART, J., WU, M. & KIRSCHNER, M. (1984). Cell cycledynamics of an M-phase-specific cytoplasmic factor inXenopns laevis oocytes and eggs. J . Cell Biol. 98,1247-1255.

HAYLES, J. & NURSE, P. (1986). Cell cycle regulation inyeast. J. Cell Sci. Suppl. 4, 155-170.

HERMANN, J., BELLE, R., TSO, J. & OZON, R. (1983).

Stabilization of the maturation-promoting factor (MPF)from Xenopns laei'is oocytes. Protection against calciumions. Cell Differ. 13, 143-148.

HERMANN, J., MULNER, O., BELLE, R., MAROT, J., Tso,

J. & OZON, R. (1984). //; vivo effects of microinjectedalkaline phosphatase and its low molecular weightsubstrates on the first meiotic cell division in Xenopnslaevis oocytes. Proc. natn. Acad. Sci. U.SA. 81,5150-5154.

IIIRAI, S., LEGOASCOGNE, C. & BAULIEU, E. E. (1983).

Induction of germinal vesicle breakdown in Xenopuslaevis oocytes: Response of denuded oocytes toprogesterone and insulin. Devi Biol. 100, 214—221.

HIRAMOTO, Y. (1974). A method of microinjection. ExplCell Res. 87, 403-406.

KANATANI, H. (1985). Oocyte growth and maturation instarfish. In Biology of Fertilisation, vol. 1 (ed. C. B.Mctz & A. Monroy), pp. 119-140. New York: AcademicPress.

KANATANI, H. & HIRAMOTO, Y. (1970). Site of action of 1-methyladeninc in inducing oocyte maturation in starfish.Fxpl Cell Res. 61, 280-284.

KANATANI, H., SHIRAI, H., NAKANISHI, K. & KUROKAWA,

T. (1969). Isolation and identification of meiosisinducing substance in starfish. Xature, Loud. 221,273-274.

KARSENTI, E., BRAVO, R. & KIRSCHNER, M. (1987).

Phosphorylation changes associated with the early cellcycle in Xenopns eggs. Devi Biol. 119, 442-453.

KlSHlMOTO, T. (1986a). Microinjection and cytoplasmictransfer in starfish oocytes. Methods in Cell Biology1, vol.27 (ed. T. E. Schroeder), pp. 379-394. New York:Academic Press.

KISHIMOTO, T. (19866). Maturation-promoting factor instarfish oocytes: Preliminary characterization and actionon oocyte-follicle cell junctions. In Advances inInvertebrate Reproduction, vol. 4 (ed. M. Porchet, J. C.Andries & A. Dhainaut), pp. 257-264. Amsterdam:Elsevier Science.

KISHIMOTO, T. & KANATANI, H. (1976). Cytoplasmicfactor responsible for germinal vesicle breakdown andmeiotic maturation in starfish oocyte. Xature, Loud. 260,321-322.

KISHIMOTO, T. & KONDO, H. (1986). Extraction andpreliminary characterization of maturation-promotingfactor from starfish oocytes. Expl Cell Res. 163, 445-452.

KISHIMOTO, T., KURIYAMA, R., KONDO, H. & KANATANI,

H. (1982). Generality of the action of variousmaturation-promoting factors. Expl Cell Res. 137,121-126.

KISHIMOTO, T., USUI, N. & KANATANI, H. (1984«).

Breakdown of starfish ovarian follicle induced bymaturation-promoting factor. Devi Biol. 101, 28-34.

KISHIMOTO, T., YAMAZAKI, K., KATO, Y., KOIDE, S. S. &

KANATANI, H. (19846). Induction of starfish oocytematuration by maturation-promoting factor of mouse andsurf clam oocytes. J. exp. Zool. 231, 293-295.

MALLER, J. L. (1985). Regulation of amphibian oocytematuration. Cell Differ. 16, 211-221.

MASUI, Y. (1967). Relative roles of the pituitary, folliclecells, and progesterone in the induction of oocytematuration in Rana pipiens. J. exp. Zool. 166, 365-376.

MASUI, Y. & CLARKE, H. J. (1979). Oocyte maturation.Int. Rev. Cytol. 57, 185-282.

MASUI, Y. & MARKERT, C. L. (1971). Cytoplasmic controlof nuclear behavior during meiotic maturation of frogoocytes. J . exp. Zool. 177, 129-146.

MIAKE-LYE, R., NEWPORT, J. & KIRSCHNER, M. (1983).

Maturation-promoting factor induces nuclear envelopebreakdown in cycloheximide-arrested embryos ofXenopus laevis. J. Cell Biol. 97, 81-91.

NELKIN, B., NICHOLS, C. & VOGELSTEIN, B. (1980).

Protein factor(s) from mitotic CHO cells induce meioticmaturation in Xenopns laevis oocytes. FEBS Lett. 109,233-238.

NGUYEN-GIA, P., BOMSEL, M., LABROUSSE, J. P.,

GALLIEN, C. L. & WEINTRAUB, H. (1986). Partial

purification of the maturation-promoting factor MPFfrom unfertilized eggs of Xenopns laevis. Enr. J.Biochem. 161, 771-777.

PICARD, A., LABBE, J . -C , PEAUCELLIER, G., LEBOUFFANT,

F., LEPEUCH, C. & DOREE, M. (1987). Changes in the

activity of the maturation-promoting factor are correlatedwith those of a major cyclic AMP and calcium-independent protein kinase during the first mitotic cellcycles in the early starfish embryo. Dev. Growth Differ.29, 93-103.

PONDAVEN, P. & MEUER, L. (1986). Protein

phosphorylation and oocyte maturation. I. Induction ofstarfish oocyte maturation by intracellular microinjectionof a phosphatase inhibitor, <r-naphthylphosphate. ExplCell Res. 163, 477-488.

PRINGLE, J. R. & HARTWELL, L. H. (1981). The

Saccharomyces cerevisiae cell cycle. In MolecularBiology of the Yeast Saccharomyces: Life Cycle andInheritance (ed. J. N. Strathern, E. W. Jones & J. R.Broach), pp. 97-142. Cold Spring Harbor, New York:Cold Spring Harbor Laboratory Press.

RUSSELL, P. & NURSE, P. (1986). Schizosaccharomyces

pombe and Saccharomvces cerevisiae: A look at yeastsdivided. Cell AS, 781-782.

SMITH, L. D. & ECKER, R. E. (1971). The interaction of

steroids with Rana pipiens oocytes in the induction ofmaturation. Devi Biol. 25, 232-247.

SUNKARA, P. S., WRIGHT, D. A. & RAO, P. N. (1979a).

Mitotic factors from mammalian cells induce germinalvesicle breakdown and chromosome codensation inamphibian oocytes. Proc. natn. Acad. Sci. U.SA. 76,2799-2802.

280 K. Tachibarta el al.

SUNKARA, P. S., WRIGHT, D. A. & RAO, P. N. (197%).

Mitotic factors from mammalian cells: a preliminarycharacterization. .7. supramolec. Struc. 11, 189-195.

WALLACE, R. A., JARED, D. W., DUMONT, J. N. & SEGA,

M. W. (1973). Protein incorporation by isolatedamphibian oocytes. I l l : Optimum incubation conditions.7. exp. Zool. 184, 321-334.

WASSERMAN, W. J. & MASUI, Y. (1975). Effect ofcycloheximide on a cytoplasmic factor initiating meioticmaturation in Xenopus oocytes. Expl Cell Res. 91,381-388.

WASSERMAN, W. J. & MASUI, Y. (1976). A cytoplasmicfactor promoting oocyte maturation: Its extraction andpreliminary characterization. Science 191, 1266—1268.

WASSERMAN, W. J. & SMITH, L. D. (1978). The cyclic

behavior of a cytoplasmic factor controlling nuclearmembrane breakdown. J . Cell Biol. 78, R15-R22.

WEINTRAUB, H., BUSCAGLIA, M., FERREZ, M., WEILLER,

S., BOULET, A., FABRE, F. & BAULIEU, E. E. (1982).

Mise en evidence d'une activite "MPF" chczSacchawmyces cerevisiae. C. r. Acad. Sci., Paris, set: 3295, 787-790.

Wu, M. & GERHART, J. C. (1980). Partial purification andcharacterization of the maturation-promotinji; factor fromeggs of Xenopus laevis. Dezl Biol. 79, 465-477.

{Received 20 May 1987 - Accepted 6 July 1987)

Yeast maturation-promoting factor 281