Development 111, 813-820 (1991)Printed in Great Britain © The Company of Biologists Limited 1991

813

Tyrosine phosphorylation of p34cdc2and p42 during meiotic maturation of

Xenopus oocyte

Antagonistic action of okadaic acid and 6-DMAP

CATHERINE JESSUS*1, HELENE RIME1, OLIVIER HACCARD1, JOHAN VAN LINT2, JOZEF

GORIS2, WILFRIED MERLEVEDE2 and RENE OZON1

1Laboratoire de Physiologic de la Reproduction, UA-CNRS/INRA 555, University P. et M. Curie, 4 Place Jussieu, 75252 Paris Ctdex 05,France2Afdeling Biochemie, Fakulteit Geneeskunde, Katholieke Universiteit te Leuven, B-3000 Leuven, Belgium

* To whom correspondence should be addressed

Summary

The tyrosine phosphorylation/dephosphorylation ofp34cdc2 was estimated by immunoblotting with anti-phosphotyrosine antibody during meiotic maturation ofXenopus oocytes. At the time of germinal vesiclebreakdown (GVBD), p34cdc2 is tyrosine dephosphory-lated whereas a p42 protein, which might correspond toa MAP2 kinase, becomes tyrosine phosphorylated. Nomodification in the level of tyrosine phosphorylation ofeither proteins was noticed during the whole maturationprocess from GVBD until metaphase II. When added toprophase oocytes, 6-DMAP (6-dimethyl-aminopurine)blocks GVBD, M-phase-promoting factor (MPF) acti-vation and Hl-histone, kinase activation induced byeither progesterone, MPF transfer or okadaic acidmicroinjection. In each case, the tyrosine dephosphoryl-ation reaction of p34c<fc2 is inhibited. In meiosis I oocytes(just after the initiation of GVBD), 6-DMAP provokesthe rephosphorylatlon of p34c<fc2 on tyrosine residue(s),inactivation of MPF and Hl-histone kinase and re-entry

of the cell into an interphase-like state. These processesare reversible by simply removing the agent. In contrastto the observations in prophase oocytes, okadaic acid isable to reverse the inhibitory effect of 6-DMAP himeiosis I oocytes on MPF and Hl-histone kinaseactivities and to initiate dephosphorylation pf p34c<fc2 ontyrosyl residue(s) even in the presence of 6-DMAP.Altogether, our results show that 6-DMAP and okadaicacid antagonistically control in vivo the level of tyrosinephosphorylation of p34c<fc2.

Key words: Xenopus oocyte, meiotic maturation, p34eJc2,maturation promoting factor, tyrosine phosphorylation.

Abbreviations: AMD, ATP Mg2"1"-dependent; 6-DMAP, 6-dimethylaminopurine; GVBD, germinal vesicle breakdown;I50, the concentration giving half-maximum inhibition;MPF, M-phase promoting factor; OA, okadaic acid; PBS,phosphate-buffered saline; PCS, polycation-stimulation.

Introduction

MPF. the maturation-promoting factor, was firstdiscovered as the result of its ability to release Xenopusoocytes from meiotic block (Masui and Markert, 1971;Smith and Ecker, 1971). A similar activity wassubsequently shown to be present at the G2/Mtransition in oocytes from other species and in mitoticcells from yeast to man (Sunkara et al. 1979; Kishimotoet al. 1982; Adlakha et al. 1988; Hashimoto andKishimoto, 1988). Therefore, MPF is now considered tocorrespond to M-phase promoting factor.

Two years ago, the serine/threonine protein kinasep34cdc2 of Schizosaccharomyces pombe (for review, seeMacNeill and Nurse, 1989) has been shown to be the

catalytic subunit of MPF (Dunphy et al. 1988; Gautier etal. 1988). At least two mechanisms are involved in theregulation of the kinase activity of p34crfc2; one isformation of complexes with other proteins such ascyclins (Swenson etal. 1986; Murray etal. 1989; Gautieret al. 1990) and the other is phosphorylation/dephosphorylation (Dunphy and Newport, 1989; Gau-tier et al. 1989). Phosphoaminoacid analysis of p34cdc2

from interphase mouse 3T3 fibroblasts indicates that itis phosphorylated on tyrosine, threonine and serineresidues (Morla et al. 1989). The putative role of serineand threonine phosphorylation or dephosphorylation isnot yet elucidated. In contrast, Gould and Nurse (1989)demonstrated that residue tyrosine 15 located withinthe ATP-binding site must be dephosphorylated in

814 C. Jessus and others

order to activate the kinase. Therefore a tyrosinephosphatase is required for MPF activation during theG2/M transition. The identification of the MPF-specifictyrosine phosphatase and of the tyrosine kinase thatcatalyzes the reverse reaction, as well as the elucidationof the mechanisms that regulate their activities,represent a crucial step to understand further how MPFis controlled during the cell cycle.

Dunphy and Newport (1989) reported that tyrosinedephosphorylation parallels MPF activation duringmeiotic maturation in Xenopus oocytes. They furthershowed that addition of an ATP-regenerating system toa 0-33 % ammonium sulfate fraction from prophaseoocytes, generated active MPF, as previously reportedby Cyert and Kirshner (1988), but triggered also thetyrosine dephosphorylation and activation of p?Acdc2.On the other hand, microinjection of okadaic acid(OA), a potent inhibitor of the polycation-stimulated(PCS or type 2A) and ATP-magnesium-dependent(AMD or type 1) protein phosphatases (Bialojan andTakai, 1988; Goris et al. 1989a; Haystead et al. 1989),into starfish, Xenopus and mouse oocytes induces MPFactivation and meiotic maturation (Goris et al. 19896;Picard et al. 1989; Rime et al. 1990; Rime and Ozon,1990). These results indicate that MPF and probablytyrosine dephosphorylation of p34cdc2 are negativelyregulated by an okadaic-acid-sensitive protein phospha-tase. Similar results were also obtained after addition ofokadaic acid to a cell-free system prepared fromactivated Xenopus eggs (Felix et al. 1990). Altogetherthese observations, although indirect, suggest that thetyrosine phosphatase that catalyzes p34 phospho-tyrosine dephosphorylation is positively regulated by anATP-dependent reaction and repressed by an okadaic-acid-sensitive protein phosphatase, most probably aPCS protein phosphatase (F61ix et al. 1990).

Recently, 6-DMAP (6-dimethyl aminopurine), an invitro protein kinase inhibitor, was shown to inhibitreversibly meiotic maturation of Patella, starfish andmouse oocytes (N6ant and Guerrier, 1988a,b; Rime etal. 1989). Interestingly, this purine derivative appearsalso to induce reversibly the exit of maturing mouseoocytes from M-phase (Rime et al. 1989), indicatingthat 6-DMAP could act as an in vivo reversible inhibitorof MPF.

We now present evidence that 6-DMAP is indeed apowerful reversible inhibitor of MPF in Xenopusoocyte. In vivo, 6-DMAP inhibits phosphotyrosinedephosphorylation of p34cdc2 in prophase oocytesinduced to mature and leads to its rephosphorylation ontyrosine residue(s) in metaphase oocytes. Therefore, inthe whole cell, 6-DMAP appears to stimulate thetyrosine phosphorylation of p34cdc2, consequentlyblocking the MPF activity. This is in contrast to theeffect of okadaic acid, which stimulates a rapiddephosphorylation of p34cdc2 together with the acti-vation of MPF.

Materials and methods

MaterialsXenopus laevis adult females (Centre de Recherche deBiochimie Macromol6culaire du Centre National de laRecherche Scientifique, Montpellier, France) were bred andmaintained under laboratory conditions. Reagents, unlessotherwise specified, were from Sigma.

Oocyte maturation and cytochemical proceduresIsolated oocytes were prepared as described in Jessus et al.(1987). Fully grown oocytes (referred as 'prophase oocytes')were induced to mature by addition of 1 ̂ JM progesterone. Thematured oocytes were referred as 'metaphase oocytes'. Thecriterion for maturation was the appearance of a white spotsurrounded by a pigmented ring at the animal pole of theoocyte. To check the presence of MPF activity, animalhemisphere cytoplasm was taken from oocytes, eitherincubated in the presence of progesterone (1 JJM) or aftervarious treatments, and transferred into recipient prophaseoocytes at the equator level. Germinal vesicle breakdown(GVBD) was ascertained by the absence of the germinalvesicle determined by dissection of the oocyte after 10 min offixation in 10 % trichloroacetic acid. Micropipettes made fromDumont Microscaps, were calibrated to distribute 50 nl ineach oocyte, 10 times (Cormier et al. 1989). A stock solutionof 2.5X10"4M OA in 10% dimethylsulfoxide, diluted in10 mM Hepes, pH7.4 before microinjection, was used.Oocytes were stained for cytochemical studies as previouslydescribed by Huchon et al. (1985). Cryostat sections werestained with 2.5^1ml"1 Hoe'chst dye 33258 (Calbiochem) inPBS to visualize the chromosomes.

Extract preparationFor kinase assay, 5 oocytes were washed in 2 ml MPFextraction buffer EB (Cyert and Kirschner, 1988); (80 mM/3-glycerophosphate pH7.3, 20mM EGTA, 15mM MgCl2,lmM dithiothreitol, and protease inhibitors: 2UM phenylmethyl sulfoxide, 25/igml"1 leupeptin, 25/igml aprotinin,10/igml"1 pepstatin, O.lmgmT1 soybean trypsin inhibitorand 1 mM benzamidine), homogenized in 100 /d EB at 4°C andcentrifuged at 100 000 revs min"1 at 4°C for 5min in a TL-100Beckman centrifuge (TLA-100 rotor). The supernatant wasthen collected for pl3-binding and Hl-histone kinase assays.For western blot analysis, 15 oocytes were washed in 2 ml EB,homogenized at 4°C in 40 jA EB and centrifuged at90 000 revs min"1 at 4°C for lOmin in a TL-100 Beckmancentrifuge (TLA-100 rotor). 30ml of Laemmli buffer(Laemmli, 1970) was then added to the supernatant and theextract was boiled for 5 min before electrophoresis andimmunoblotting.

Preparation and use of pl3-harose beadspl3 was purified and conjugated to Sepharose as describedpreviously (Brizuela etal. 1987). Oocyte supernatant obtainedfrom 5 oocytes was incubated for 90 min under constantrotation at 4°C with 50^1 of pl3-sepharose in a final volumeof lml of EB. After lmin centrifugation at 1000 revs min"1,the pl3-sepharose pellets were washed once with lml of EBand twice with 1 ml of kinase buffer (50 mM Tris-HCl, pH 7.4,5mM EGTA, 10mM MgCl2> lmin dithiothreitol).

Hl-histone kinase assaypl3-sepharose pellets were resuspended in 50^1 of kinasebuffer containing 0.2 mg HI histoneml"1 (Boerhinger),0.1 mM ATP and 1/xCi of y^P-ATP (l lOTBqmmol,

Reversible tyrosine dephosphorylation of p34cdc2 in Xenopus oocyte 815

Amersham, PB 10168) and incubated at 30°C for 15 min. Thereaction was stopped by adding 30j/l of Laemmli samplebuffer (Laemmli, 1970) and boiling for 3 min. After 12%polyacrylamide gel electrophoresis and autoradiography, thebands of Hl-histone were excised from the gel and counted inwater by Cerenkov counting in a SL 3000 scintillation counter(Intertechnique).

ImmunoblottingAfter electrophoresis on a 12% SDS-PAGE gel (Laemmli,1970), the proteins were transferred to nitrocellulose filters(Schleicher and Schull) using a semi dry blotting system(Millipore). The filters were soaked for 120 min in 3 % bovineserum albumin in phosphate-buffered saline (PBS) sup-plemented with 0.2% Tween-20 and further incubatedovernight with the rabbit antibody raised against phospho-tyrosine diluted 1/100 in PBS containing 3% bovine serumalbumin. The transfer membranes were washed six times inNET gel (50 mM Tris-HCl pH 7.4,150 im NaCl, 5 HIM EDTA,0.25% gelatin, 0.05% NP-40), incubated for 60min with4/iCi [^IJprotein A (Amersham) in 40ml PBS, washed sixtimes in NET gel, dried and exposed at —80°C using RoyalX-Omat films (Kodak).

Anti-phosphotyrosine antibodiesAffinity-purified anti-phosphotyrosine antibodies were pre-pared as described by Kamps and Sefton (1988) using asantigen keyhole limpet hemocyanin polymerised with anequimolar mixture of L-phosphotyrosine, L-alanine and L-threonine using l-ethyl-3-(3-dimethylaminopropyl)carbodii-mide as coupling antigen. Specificity of the antibodies wastested towards conjugates of phosphotyrosine, phosphoserineor phosphothreonine with bovine serum albumin, or towardsreduced carboximidomethylated and maleylated lysozymeexclusively phosphorylated on tyrosine residues (Hendrix etal. 1989), phospho inhibitor-1 (phosphothreonine) or glyco-gen phosphorylase a (phosphoserine), both phosphorylated asdescribed (Waelkens et al. 1987). Specificity was also tested byinhibition of the immunoreaction by phosphotyrosine.

Results

6-DMAP inhibits MPF and Hl-histone kinaseactivation6-DMAP is a puromycin analog that does not inhibitprotein synthesis but rather inhibits protein kinases. Itis able to block Patella, starfish and mouse oocytematuration, probably by acting at the level of MPFactivation (N6ant and Guerrier, 1988a,b; Rime et al.1989). Similarly, progesterone-induced maturation ofXenopus oocyte is inhibited by 6-DMAP in a dose-dependent manner, with an /50 of 100 ̂ M (Fig. 1A). Inorder to know if this inhibitor might interfere directlywith MPF activation, prophase oocytes were incubatedwith various concentrations of 6-DMAP before micro-injection of MPF. As shown in Fig. 1A, MPF-inducedmaturation is also inhibited by 6-DMAP with an I^ of370/IM. It was verified that no transferable MPF activitywas present in 6-DMAP-blocked oocytes either incu-bated in progesterone or microinjected with MPF.Furthermore, no activation of the pl3-sepharose-bound Hl-histone kinase was observed in these6-DMAP-blocked oocytes. These observations indicate

that 6-DMAP is a potent inhibitor of the Xenopusoocyte meiotic maturation, probably by acting directlyat the level of MPF activation.

The reversibility of 6-DMAP action was tested byincubating prophase oocytes in the presence of both

0.25 0,75 16-DMAP (mM)

B%GVBO

100 T

50 •

12

Time (h)

Fig. 1. Inhibition of progesterone- and MPF-inducedmaturation by 6-DMAP. (A) 6-DMAP dose-reponsecurves. Oocytes were incubated in the presence of 1 mMprogesterone and various 6-DMAP concentrations; after12 h, germinal vesicle breakdown was ascertained asdescribed in the Materials and methods section ( • • ) .Oocytes were also preincubated for 3 h in variousconcentrations of 6-DMAP and then microinjected withcytoplasm taken from progesterone-matured metaphase IIoocytes (O O). GVBD frequency was scored 8h afterinjection. (B) Reversibility of 6-DMAP action. Oocyteswere incubated overnight in the presence of 1 mMprogesterone plus 0.3 mM 6-DMAP, then washedextensively (time 0) and incubated in control medium(O O); control oocytes were incubated with lmMprogesterone ( • • ) .

816 C. Jessus and others

0.3 HIM 6-DMAP and 1 ̂ M progesterone overnight. Theoocytes were then washed and placed in 6-DMAP- andprogesterone-free incubation medium. Under theseconditions, GVBD occurs despite the absence ofprogesterone in the incubation medium. Furthermore,the time course of maturation is shorter than in controloocytes induced to mature by progesterone (Fig. IB).In the presence of 6-DMAP, progesterone inducessome precocious intracellular events, which couldexplain the shorter maturation time course afterremoval of the drug.

Hl-histone kinase activity as well as MPF activationare correlated with tyrosine dephosphorylation ofp34cdc2 (Gautier et al. 1989; Dunphy and Newport,1989; Jessus et al. 1990). Since both MPF and Hl-histone kinase activations were blocked by 6-DMAP, itwas important to analyze the tyrosine phosphorylationlevel of p34cdc2. By immunoblotting the supernatant ofprophase oocytes with anti-phosphotyrosine antibody,a single 34X103 Mr band was detected (Fig. 2, lane 1).This band is bound to pl3-sepharose and is missingfrom the prophase fraction after depletion through pl3-sepharose. Its behavior and its relative molecular massare indistinguishable from those of the p34cdc2. It istherefore considered to correspond to p34ciic2 (Dunphyand Newport, 1989; Jessus et al. 1990). To avoidpossible phosphorylation/dephosphorylation reactionsduring the binding of p34c*c2 to pl3-sepharose, thetyrosine phosphorylation level was estimated in oocytesupernatant in all further experiments. The supernatantof metaphase II oocyte was also immunoblotted with

MrXK)"3

67-

43 -

30 -

Time 0(h)

0.5 1 1.5 2.5

20 -

1 2 3 4 5Fig. 2. Anti-phosphotyrosine immunoblots after 6-DMAPtreatment of prophase oocytes. Lane 1, prophase oocytes;lane 2, metaphase II oocytes; lane 3, oocytes incubatedovernight with 0.5 HIM 6-DMAP; lane 4, oocytes incubatedovernight in the presence of 1 um progesterone and 0.5 nm6-DMAP; lane 5, oocytes incubated overnight with 1/IMprogesterone plus 0.5 mM 6-DMAP, then washedextensively and incubated in control medium until GVBD.

ZGVBD 0 0 0 0 10 22

Fig. 3. Time course of p34cdc2 tyrosine dephosphorylationand p42 tyrosine phosphorylation during the progesterone-induced maturation. Maturation of prophase oocytes wasinitiated by progesterone. 15 oocytes were homogenizedevery 30 min, and immunoblotted with the anti-phosphotyrosine antibody as described under the Materialsand methods section. GVBD frequency was scored inparallel. After 3h in the presence of progesterone, nochanges in the immunoblot pattern were observed untilmetaphase II. 50% GVBD was reached at 4h and 100%at 7h.

the anti-phosphotyrosine antibody; as shown in Fig. 2(lane 2), p34cdc2 is dephosphorylated on tyrosine, aspreviously reported by several authors (Dunphy andNewport, 1989; Jessus et al. 1990). In contrast, a42 x 103 Mr protein becomes heavily phosphorylated ontyrosyl residue in these oocytes (Fig. 2, lane 2). Thetime course of the tyrosine phosphorylation of boththese proteins was analyzed from prophase up tometaphase II arrest (Fig. 3). It shows that p34cdc2

dephosphorylation occurs at the time or slightly earlierthan the first pigment redistribution at the animal pole(Huchon et al. 1981); the 42xlO3Mr protein isphosphorylated before the complete tyrosine dephos-phorylation of p34crfc2. When progesterone-inducedmaturation is blocked by 6-DMAP, p34cdc2 remainsphosphorylated on tyrosine and the 42xlO3A/r proteindoes not undergo tyrosine phosphorylation (Fig. 2, lane4). When 6-DMAP was washed out, GVBD occurs(Fig. IB), p34cdc2 becomes dephosphorylated on tyro-sine and the 42xlO3Afr protein is phosphorylated onthis residue (Fig. 2, lane 5), as in metaphase controloocytes. We therefore conclude that 6-DMAP inhibitsMPF activation and tyrosine dephosphorylation of thep34cdc2.

6-DMAP causes exit from M-phase during meiosis IThe p34cdc2 kinase activity from metaphase II oocytes,bound to pl3-sepharose, is inhibited in vitro by6-DMAP: 75% inhibition is obtained with 100 /m6-DMAP. At 5 mM 6-DMAP, only 85 % inhibition isobserved, indicating that some forms of p34cdc2 mightbe 6-DMAP resistant (not illustrated). On the otherhand, since 6-DMAP blocks MPF activation, we furthertested whether the drug promotes in vivo inactivation ofboth MPF and Hl-histone kinase activities in maturingoocytes. Oocytes were incubated in the presence of1 fiM progesterone, and 3 HIM 6-DMAP was added at thetime of GVBD. After a 3 h incubation, no MPF activitywas detected by cytoplasmic transfer. The threshold

Reversible tyrosine dephosphorylation of p34 in Xenopus oocyte 817

concentration of 6-DMAP leading to the disappearanceof MPF activity is higher (500 (IM) than the dosesinhibiting its activation by progesterone or MPFmicroinjection. Interestingly, the removal of 6-DMAPafter this 3h period allows for the spontaneousreappearance of MPF activity, indicating again areversible effect of 6-DMAP. Hl-histone kinase activityof the p34cdc2 bound to pl3-sepharose was measuredduring maturation and after 6-DMAP treatment ofmetaphase I oocytes. The Hl-histone kinase is acti-vated about 100-fold at the GVBD period (Fig. 4A,compare lane 1 and lane 2) and the activity remains highuntil metaphase II, 3 h after GVBD (Fig. 4A, lane 4). A75 % inhibition of the Hl-histone kinase is observedwhen GVBD oocytes were treated for 3h with6-DMAP (Fig. 4A, lane 3). 6-DMAP is, therefore, ableto inactivate in vivo both MPF and Hl-histone kinaseactivities of meiosis I oocytes. A preliminary cytologicalanalysis was performed on oocytes treated for 2 h with6-DMAP after GVBD; it shows that the chromosomesnow appear decondensed and included into nuclei-likestructures (data not illustrated). An immunochemicalstudy is now underway in our laboratory in order toknow if in Xenopus metaphase I oocyte, like in mousemetaphase I oocyte (Rime et al. 1989), 6-DMAPprovokes the repolymerization of lamin around thedecondensed chromosomes.

We further examined whether MPF inactivation by6-DMAP is correlated with tyrosine rephosphorylationof p34cdc2 at meiosis I. By immunoblotting with anti-phosphotyrosine antibody, p34cdc2 was demonstrated tobe rephosphorylated on tyrosine in these conditions(Fig. 4B, lane 4). In contrast, the 42xlO3Mr protein,which becomes normally phosphorylated on tyrosineduring maturation, is completely dephosphorylatedfollowing incubation with 6-DMAP. Taken together,these results indicate that 6-DMAP inhibits MPFactivity in vivo by acting at least on the tyrosinephosphorylation level of p34crfc2. They further showthat the p34cdc2-specific tyrosine kinase is present inmaturing oocytes and is not highly sensitive to 6-DMAPin these in vivo conditions, whereas the tyrosine kinasespecific for the 42xl03Afr protein is more sensitive tothis agent.

Okadaic acid reverses the inhibitory effect of6-DMAP on MPF activity

Okadaic acid (OA), a potent inhibitor of the PCS(type 2A) and AMD (type 1) protein phosphatases(Bialojan and Takai, 1988; Goris et al. 1989a; Haysteadet al. 1989), is able to induce GVBD in Xenopus oocytesas well as an astonishingly rapid MPF activation (Goriset al. 19896; Rime et al. 1990). It was also shown thatOA induces Hl-histone kinase activity in 30min, to asimilar level as in metaphase II oocytes (Fig. 4A,compare lane 4 and lane 6). When OA microinjection isperformed in the presence of cycloheximide, MPF andHl-histone kinase are also activated. OA probablyprevents a PCS phosphatase from inactivating a proteinnecessary for MPF activation (F61ix et al. 1990),perhaps p34cdc2 protein phosphatase itself (Hunt, 1989),

which is clearly an OA-insensitive enzyme. If thishypothesis is correct, OA should be able to reverse theeffect of 6-DMAP on p34cdc2 tyrosine phosphorylationby initiating the phosphatase cascade. To verify thishypothesis, GVBD stage oocytes were treated for 3hwith 3mM 6-DMAP. This treatment results in thedisappearance of MPF while Hl-histone kinase activitywas inhibited by 75 % (Fig. 4A, lane 3). OA was thenmicroinjected always in the presence of 6-DMAP.

cpm in H16000

4000

2000.

M r x10'3B

43-

30-fc

1 4 5Fig. 4. Effect of 6-DMAP at GVBD period on Hl-histonekinase activity and tyrosyl phosphorylation of p34crfc2.(A) Hl-histone kinase activity of 6-DMAP- and OA-treated oocytes. 5 oocytes are homogenized and assayedfor Hl-histone kinase activity as described in Materials andmethods. Lane 1, prophase oocytes; lane 2, oocytes at thetime of GVBD; lane 3, oocytes at GVBD are incubated inthe presence of 3mM 6-DMAP for 3h, lane 4, oocytes 3hafter GVBD (metaphase II); lane 5, oocytes at GVBD areincubated in the presence of 3 mM 6-DMAP for 3 h andthen injected with 2.5X10~5M OA. Oocytes arehomogenized 35min after injection; lane 6, prophaseoocytes injected with 2.5X10~5M OA, and thenhomogenized 35min after injection. Upper panel:autoradiography of Hl-histone, lower panel: ctsmin"1 in3.75/ig of Hl-histone. (B) Anti-phosphotyrosineimmunoblots after 6-DMAP and OA treatment. Lane 1,prophase oocytes; lane 2, metaphase II oocytes; lane 3,oocytes at GVBD; lane 4, oocytes at GVBD incubated inthe presence of 3mM 6-DMAP for 3h; lane 5, oocytes atGVBD were incubated in the presence of 3 mM 6-DMAPfor 3h and then microinjected with 2.5X10~5M OA.Oocytes were homogenized 35min after OA injection; lane6, prophase oocytes injected with 2.5X10~5M OA, andhomogenized 35 min after injection

818 C. Jessus and others

35min later, MPF activity was again detectable bycytoplasmic transfer, and Hl-histone kinase activityreached again a level similar to control GVBD oocytes(Fig. 4A, lanes 2 and 5). Microinjection of OA in6-DMAP-treated oocytes leads to tyrosine dephos-phorylation of the p34cdc2, which has been previouslyrephosphorylated following 6-DMAP treatment(Fig. 4B, lanes 4 and 5). Therefore, OA is able tocounteract the 6-DMAP action, and both agentsantagonistically regulate, in vivo, the phosphorylationlevel of p34c^2. Surprisingly, the 42xlO3Mr proteinappears only slightly phosphorylated on tyrosine in OA-microinjected oocytes (Fig. 4B, lanes 5 and 6) bycomparison with GVBD or metaphase II oocytes(Fig. 4B, lanes 2 and 3). The tyTosine phosphorylationof this 42xlO3A/r protein is therefore not essential forMPF and Hl-histone kinase activation.

Discussion

Tyrosine dephosphorylation of p34cdc2 occurs in Xeno-pus oocytes induced to mature by progesterone, MPFtransfer or OA microinjection; the time course ofdephosphorylation parallels, in the three experimentalconditions, the appearance of MPF activity and theactivation of Hl-histone kinase assayed after binding topl3-sepharose. The time course of tyrosine dephos-phorylation of p34cdc2 induced by OA is astonishinglyrapid since 30min after OA microinjection, phospho-tyrosine is undetectable at the level of p34 byimmunoblotting with anti-phosphotyrosine antibody(Fig. 4B); the reaction occurs even in the presence ofcycloheximide. Two conclusions can be drawn fromthese results; first, and in agreement with the in vitroanalysis of Dunphy and Newport (1989), the p34cdc2

tyrosine phosphatase is present in the resting oocyte in alatent form; second the tyrosine dephosphorylationreaction is negatively regulated by an OA-sensitivephosphatase (Fig. 5). Therefore, in Xenopus oocytes,tyrosine dephosphorylation of p34crfc2 appears to be arequirement for MPF activation.

When Xenopus oocytes were treated with 6-DMAP,MPF activation and tyrosine dephosphorylation ofp34cdc2 are blocked (Rime et al. 1989; Fig. 2), whateverthe trigger used to induce maturation: either progester-one, MPF or OA. One in vivo target for 6-DMAPaction is the tyrosine dephosphorylation reaction. Bywhich mechanism(s)? At this moment, the possibilitythat the aminopurine drug works on protein phospha-tases can be excluded since we previously showed that ithas no in vitro effect (either inhibitory or activatory) ondifferent serine/threonine phosphatases or tyrosinephosphatases, either on the activation mechanism ofthe AMD phosphatase by the kinase FA, or on theactivation mechanism of the tyrosine phosphataseactivity of the PCSL phosphatase (Rime et al. 1990).The only reported in vitro action of 6-DMAP is theinhibition of serine/threonine protein kinases (Meijerand Pondaven, 1988; Ne"ant and Guerrier, 198Sa,b); inparticular, it is a potent inhibitor of Xenopus p34cdc2

Q 6-DMAP

pre-MPF MPF

Fig. 5. Proposed model for p34cdrf activation. Thephosphorylation level of p34cdc2, resulting from theequilibrium between a tyrosine kinase and a tyrosinephosphatase, is negatively controlled by an OA-sensitivephosphatase, probably phosphatase 2A. The kinase activityof p34cdc2 controls its own dephosphorylation. 6-DMAPinhibits the whole autocatalytic process by blocking thekinase activity.

kinase activity bound to pl3-sepharose (I50 less than100 ptM). If 6-DMAP is also an in vivo inhibitor ofp34cdc2 in our experimental conditions, then a simpleexplanation can be proposed, based on the modelhypothetized by Hunt (1989). We suggest (Fig. 5) thatp34cdc2 is the kinase that directly controls its tyrosinedephosphorylation perhaps by phosphorylating directlythe tyrosine phosphatase. This would explain theautoamplification of MPF. By inhibiting the Hl-histonekinase activity of p34cdc2, 6-DMAP, as far as it ispresent, would prevent its dephosphorylation and MPFautoamplification. In these conditions, the blockage ismaintained even in the presence of OA, as reported byRime et al. (1990).

A striking observation is the finding that 6-DMAPprovokes the rephosphorylation of p34cdc2 once it isactivated in M-phase oocyte at the time of meiosis I, i.e.immediately after GVBD. The rephosphorylationreaction is correlated with the disappearance oftransferable MPF, partial inactivation of Hl-histonekinase (70-80 % inhibition after a two hours 6-DMAPtreatment) and reentry into interphase. It was furthershown that this effect is perfectly reversible since simpleremoval of 6-DMAP leads to the restoration of MPFactivity and dephosphorylation of p34cdc2, suggestingthe presence in meiosis I oocytes of all components ofthe phosphorylation reactions required to inactivatereversibly MPF activity. Another interesting result isthe fact that OA, in contrast with the6-DMAP-blocking effect in prophase oocytes, can nowrelease the 6-DMAP blockage. This apparent discrep-ancy may easily be explained by the assumption that, inthis condition, the presence of a residual Hl-histonekinase activity (20-30 % of the total activity measuredin meiosis I oocytes) becomes sufficient to activate itsdephosphorylation reaction when the negative regu-lation of tyrosine dephosphorylation reaction by the

Reversible tyrosine dephosphorylation of p34 in Xenopus oocyte 819

OA-sensitive phosphatase is switched off completely bythe presence of OA (Fig. 5).

In progesterone and MPF-induced maturation, a42xl(rMr protein is tyrosine phosphorylated whenp34cdc2 is tyrosine dephosphorylated. The p42 phospho-tyrosine protein might correspond to the X-MAPkinase (Haccard et al. 1990) a serine/threonine kinaseactivated during Xenopus oocyte maturation, anal-ogous to the mitogen activated MAP2 kinase in 3T3cells (Ray and Sturgill, 1987). In fact X-MAP kinase,purified to homogeneity from matured Xenopusoocytes is a 42xlO3Mr phosphotyrosine protein(Haccard et al. in preparation). The time course studyshown in Fig. 3 indicates that the phosphorylation ofp42 precedes the total dephosphorylation of p34cdc2 butdoes not make it possible to establish whether thesereactions are interdependent. However, the effects of6-DMAP on meiosis I oocytes suggest that both eventsare temporally linked and do not exclude the possibilitythat p34cdc2 controls one of the enzymes regulating thelevel of tyrosine phosphorylation of p42. In any case,these results indicate that, in the same cell and at thesame period, the level of tyrosine phosphorylation ofp42 is stimulated when the level of tyrosine phosphoryl-ation of p34cdc2 is decreased, suggesting that thetyrosine kinases and tyrosine phosphatases catalyzingthe phosphorylation reactions of each protein aredifferent. We verified that p34cdc2 and p42 are nottightly associated since p34^ is found to be bound topl3-sepharose whereas p42 remains in the solublefraction (not illustrated).

Interestingly, OA, which readily induces Hl-histonekinase activity and p34cdc2 dephosphorylation, poorlyactivated the tyrosine phosphorylation of p42, sugges-ting that an OA-sensitive phosphatase affects either thetyrosine kinase or the tyrosine phosphatase thatcontrols the phosphorylation equilibrium reaction ofp42. This experimental result also means that activationof p34cdc2, which occurs readily after OA microinjec-tion, is not sufficient by itself to induce the tyrosinephosphorylation of p42.

The mechanisms that link the initial site of action ofprogesterone and MPF activation are poorly under-stood. Inhibitors of protein synthesis and an elevatedlevel of cAMP block the progesterone-induced matu-ration but not the MPF-induced maturation (Marot etal. 1977; Masui and Clarke, 1979; for review, seeMailer, 1987). Therefore, it is believed that, first, thesynthesis of new protein(s) and, second, the dephos-phorylation of a substrate of the cAMP-dependentprotein kinase are both required for MPF activation byprogesterone. It is possible that, despite the presence ofcyclin proteins and c-mos*e proto-oncogene product inprophase oocytes (Roy et al. 1990), progesteronestimulates initially the translation of mRNA of cyclin orof c-mos", since mRNAs microinjection of bothproteins is able to induce maturation (Freeman et al.1989; Sagata et al. 1989; Swenson et al. 1986; fordiscussion, see Lewin, 1990). The results presented inFig. 1 indicate that 6-DMAP inhibits progesterone-induced maturation but does not prevent some early

effects of progesterone since oocytes previouslyblocked in the presence of 6-DMAP and progesteronematured perfectly after simply washing out 6-DMAPand progesterone, with a faster time course than thecontrol progesterone-treated oocytes: removal of6-DMAP is sufficient to induce the appearance of MPFin the absence of progesterone. Therefore, some of theearly effects of progesterone that are initially requiredfor preparing the conversion of pre-MPF into MPF,occur in the presence of 6-DMAP. This drug may be anexcellent tool to unravel these early and primaryresponse.s of the oocyte to progesterone.

We wish to thank David Beach for providing the pl3overexpressing strain of E. Coli. Okadaic acid was a kind giftfrom Dr D. Uemura (Shizuoka University, Japan). We thankDenise Huchon for comments on the manuscript. Thisresearch was supported by grants from INRA, CNRS,INSERM, MRT (France) and FGWO and ASLK KankerFonds (Belgium). Johan Van Lint is a Research Fellow fromthe 'Belgish Werk Tegen Kanker'.

References

ADLAKHA, R. C , SHIPLEY, G. L., ZHAO, J. Y., JONES, K. B.,WRIGHT, D. A., RAO, P. N. AND SAUER, H. W, (1988).Amphibian oocyte maturation induced by extracts of Physarumpolycephalum in mitosis. / . Cell Biol. 106, 1445-1452.

BIALOJAN, C. AND TAKAI, A. (1988). Inhibitory effect of marine-sponge toxin, okadaic acid, on protein phosphatases. Biochem.J. 256, 283-290.

BRIZUELA, L., DRAETTA, G. AND BEACH, D. (1987). p l3 ! u c l acts inthe fission yeast cell division cycle as a component of the p34cdc2

protein kinase. EMBO J. 6, 3507-3514.CORMIER, P., MULNER-LORILLON, O., OZON, R. AND BELLE, R.

(1989). Involvement of protein kinase A and casein kinase II inthe in vivo protein kinase activities in prophase arrestedXenopus laevis oocytes. Bioscience Rep. 9, 351-358.

CYERT, M. S. AND KIRSCHNER, M. W. (1988). Regulation of MPFactivity in vitro. Cell 53, 185-195.

DUNPHY, W. G., BRIZUELA, L., BEACH, D. AND NEWPORT, J.(1988). The Xenopus homolog of cdc2 is a component of MPF,a cytoplasmic regulator of mitosis. Cell 54, 423-431.

DUNPHY, W. G. AND NEWPORT, J. W. (1989). Fission yeast pl3blocks mitotic activation and tyrosine dephosphorylation of theXenopus cdc2 protein kinase. Cell 58, 181-191.

FELIX, M-A., COHEN, P. AND KARSENTI, E. (1990). Cdc2 HIkinase is negatively regulated by a type 2A phosphatase in theXenopus early embryonic cell cycle: evidence from the effects ofokadaic acid. EMBO J. 9, 675-683.

FREEMAN, R. S., PICHAM, K. M., KANKI, J. P., LEE, B. A., PENA,S. V. AND DONOGHUE, D. J. (1989). Xenopus homolog of themos protooncogene transforms mammalian fibroblasts andinduces maturation of Xenopus oocytes. Proc. natn. Acad. Sci.U.S.A 86, 5805-5809.

GAUTIER, J., MATSUKAWA, T., NURSE, P. AND MALLER, J. L.(1989). Dephosphorylation and activation of Xenopus p34 cda

protein kinase during the cell cycle. Nature 339, 626-629.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.

GAUTIER, J., NORBURY, C , LOHKA, M., NURSE, P. AND MALLER, J.L. (1988). Purified maturation-promoting factor contains theproduct of Xenopus homolog of the fission yeast cell cyclecontrol.gene cdc2. Cell 54, 433-439.

GORIS, J., HERMANN, J., HENDRIX, P., OZON, R. AND MERLEVEDE,W. (1989a). Okadaic acid, a non-TPA tumor promotor, inhibitsspecifically protein phosphatases, induces maturation and MPFformation in Xenopus laevis oocytes. In Advances in Protein

820 C. Jessus and others

Phosphatases, Vol. 5 (ed. W. Merlevede and J. Di Salvo), pp.579-592. Leuven University, Press.

GORIS, J., HERMANN, J., HENDRIX, P., OZON, R. AND MERLEVEDE,W. (1989Z>). Okadaic acid, a specific protein phosphataseinhibitor induces maturation and MPF formation in Xenopuslaevis oocytes. FEBS Lett. 245, 91-94.

GOULD, K. L. AND NURSE, P. (1989). Tyrosine phosphorylation ofthe fission yeast cdc2+ protein kinase regulates entry intomitosis. Nature 342, 39-45.

HACCARD, O., JESSUS, C , CALYA, X., GORIS, J., MERLEVEDE, W.AND OZON, R. (1990). In vivo activation of a microtubule-associated protein kinase during meiotic maturation of theXenopus oocyte. Eur. J. Biochem. 192, 633-642.

HASHIMOTO, N. AND KISHIMOTO, T. (1988). Regulation of meioticmetaphase by a cytoplasmic maturation-promoting factor duringmouse oocyte maturation. Devi Biol. 126, 242-252.

HAYSTEAD, T. A. J., SIM, A. T. R., CARUNG, D. C , HONNOR, R.Y. C , TSUKITAMI, Y., COHEN, P. AND HARDIE, D. G. (1989).Effects of the tumour promoter okadaic acid on intracellularprotein phosphorylation and metabolism. Nature 337, 78-81.

HENDRJX, P., GORIS, J. AND MERLEVEDE, W. (1989).Phosphoryrosyl phosphatases in the cytosol of Xenopus laevisoocytes. In Advances in Protein Phosphatases, Vol. 5 (ed. W.Merlevede and J. Di Salvo), pp. 593-600. Leuven UniversityPress.

HUCHON, D., CROZET, N., CANTENOT, N. AND OZON, R. (1981).Germinal vesicle breakdown in the Xenopus laevis oocyte: effectof calcium ionophore A-23187 and taxol. Reprod. Nutr.Develop. 21, 135-148.

HUCHON, D., JESSUS, C. AND OZON, R. (1985).Immunofluorescence distribution of microtubules during the firstmeiotic cell division of the Xenopus laevis oocyte. C. r. hebd.sianc. Acad. Set. Paris 12, 463-465.

HUNT, T. (1989). Under arrest in the cell cycle. Nature 342,483-484.

JESSUS, C , DUCOMMUN, B. AND BEACH, D. (1990). Directactivation of cdc2 with phosphatase: identification of pl3'u<:1-sensitive and insensitive steps. FEBS Lett. 266, 4-8.

JESSUS, C., THIBIER, C. AND OZON, R. (1987). Levels ofmicrotubules during the meiotic maturation of the Xenopusoocyte. /. Cell Sci. 87, 705-712.

KAMPS, M. P. AND SEFTON, B. M. (1988). Identification of multiplenovel polypeptide substrates of the v-src, v-yes, v-fps, v-ros andv-eri>-B oncogenic tyrosine protein kinases utilizing antiseraagainst phosphotyrosine. Oncogene 2, 305-315.

KISHIMOTO, T., KURIYAMA, R., KONDO, H. AND KANATANI, H.(1982). Generality of the action of various-promoting factors.Expl Cell Res. 137, 121-126.

LAEMMU, U. K. (1970). Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227', 680-685.

LEWIN, B. (1990). Driving the cell cycle: M phase kinase, itspartners, and substrates. Cell 61, 743-752.

MACNEILL, S. A. AND NURSE, P. (1989). Genetic interactions inthe control of mitosis in fission yeast. Curr. Genet. 16, 1-6.

MALLER, J. L. (1987). Mitogenic signalling and proteinphosphorylation in Xenopus oocytes. J. Cyclic Nucl. Prot.Phosphor. Res. 11, 543-555.

MAROT, J., BELLE, R. AND OZON, R. (1977). In vitro facilitation ofXenopus oocyte maturation by subthreshold doses ofprogesterone. Devi Biol. 59, 91-95.

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

MASUI, Y. AND MARKERT, C. L. (1971). Cytoplasmic control of

nuclear behavior during meiotic maturation of frog oocytes. J.exp. Zool. Ill, 129-146.

MEUER, L. AND PONDAVEN, P. (1988). Cyclic activation of histoneHI kinase during sea urchin egg mitotic divisions. Expl Cell Res.174, 116-129.

MORLA, A. O., DRAETTA, G., BEACH, D. AND WANG, J. Y. (1989).Reversible tyrosine phosphorylation of cdc2: dephosphorylationaccompanies activation during entry into mitosis. Cell 58,193-203.

MURRAY, A. W., SOLOMON, M. J. AND KIRSCHNER, M. W. (1989).The role of cyclin synthesis and degradation in the control ofMPF activity. Nature 339, 280-286.

NEANT, I. AND GUERRIER, P. (1988c). Meiosis reinitiation in themollusc Patella vulgata. Regulation of MPF, CSF andchromosome condensation activity by intracellular pH, proteinsynthesis and phosphorylation. Development 102, 505-516.

NEANT, I. AND GUERRIER, P. (19886). 6-dimethylaminopurineblocks starfish oocyte maturation by inhibiting a relevant proteinkinase activity. Expl Cell Res. 176, 68-79.

PICARD, A., CAPONY, J. P., BRAUTIGAN, D. L. AND DOREE, M.(1989). Involvement of protein phosphatases 1 and 2A in thecontrol of M phase-promoting factor activity in Starfish. J CellBiol. 109, 3347-3354.

RAY, B. L. AND STURGILL, T. W. (1987). Rapid stimulation byinsulin of a serine/threonine kinase in 3T3-L1 adipocytes thatphosphorylates microtubule-associated protein 2 in vitro. Proc.natn. Aca. Sci. U.S.A. 84, 1502-1506.

RJME, H., HUCHON, D., JESSUS, C , GORIS, J., MERLEVEDE, W.AND OZON, R. (1990). Characterization of MPF by okadaic acidin Xenopus oocyte. Cell Differ. Dev. 29, 47-58.

RJME, H., NEANT, I., GUERRIER, P. AND OZON, R. (1989). 6-dimethylaminopurine (6-DMAP), a reversible inhibitor of thetransition to metaphase during the first meiotic cell division ofthe mouse oocyte. Devi Biol. 133, 169-179.

RIME, H. AND OZON, R. (1990). Protein phosphatases are involvedin the in vivo activation of histone HI kinase in mouse oocyte.Devi Biol. 141, 115-123.

ROY, L. M., SINGH, B., GAUTIER, J., ARLINGHAUS, R. B.,NORDEEN, S. K. AND MALLER, J. L. (1990). The cyclin B2component of MPF is a substrate for the c-mos" proto-oncogene product. Cell 61, 825-831.

SAGATA, N., DAAR, I., OSKARSSON, M., SHOWALTER, S. D. ANDVANDE WOUDE, G. F. (1989). The product of the mos proto-oncogene as a candidate 'initiator' for oocyte maturation.Science 245, 643-646.

SMITH, L. D. AND ECKER, R. E. (1971). The interaction of steroidswith Rana pipiens oocytes in the induction of maturation. DeviBiol. 25, 233-247.

SUNKARA, P. S., WRIGHT, D. A. AND RAO, P. N. (1979). Mitoticfactors from mammalian cells induce germinal vesiclebreakdown and chromosome condensation in amphibianoocytes. Proc. natn Acad Sci. U.S.A 76, 2799-2802.

SWENSON, K. I., FARRELL, K. M. AND RUDERMAN, J. V. (1986).The clam embryo protein cyclin A induces entry into M phaseand the resumption of meiosis in Xenopus oocytes. Cell 47,861-870.

WAELKENS, E., GORIS, J. AND MERLEVEDE, W. (1987).Characterization of the catalytic subunits of the different typesof polycation stimulated phosphatases. J. biol. Chem. 262,1049-1059.

{Accepted 5 December 1990)