DEVELOPMENTAL

BIOLOGY

152, 62-74 (1992)

Centrosome Phosphorylation and the Developmental Expression of Meiotic Competence in Mouse Oocytes DINELI

WICKRAMASINGHE

AND DAVID

F. ALBERTINI’

Previous studies suggested that the transition from an incompetent to a competent meiotic state during the course of oogenesis in the mouse involved a G./M-like cell cycle transition (Wickramasinghe et trl.. 1991. Da: Bid. 143,162). The present studies tested the hypothesis that centrosome phosphorylation, an event normally induced by MPF, is required for this developmental transition and the expression of meiotic competence in cultured growing mouse oocytes. Multiple fluorescence labeling techniques were used to evaluate centrosome number, phosphorylation status, and microtubule nucleating capacity in competent and incompetent oocgtes. Experimental conditions were established for reversibly altering the phosphorylation status of the centrosomes and the effects of these treatments on meiotic resumption were examined. Phosphoyylated centrosomes nucleating short microtubules were observed in competent oocytes, whereas nonphosphorylated centrosomes and interphase microtubule arrays were found in incompetent oocytes. Upon recovery from nocodazole-induced microtubule depolymerization, short microtubules formed from ccntrosomes in competent oocytes, whereas long microtubules reappear in the cytoplasm of incompetent oocytes. Perturbation of the phosphorglation state of oocytes with activators of protein kinase A or protein kinase C resulted in the formation of long interphase microtuhules in competent oocytes while centrosome phosphorylation was maintained. Treatment of competent oocytes with the phosphorylation inhibitor 6dimethglaminopurine also led to formation of long microtubules, although under these conditions centrosomes were dephosphorplated. %‘hen competent oocytes were treated simultaneously with puromycin and the phosphodiesterase inhibitor isobutyl methylxanthine (IBMXl for 6 hr, centrosomes became dephosphorglated; centrosomes were rephosphorylated when competent oocptes were further cultured in IBMX without puromycin. Conditions that induced centrosome dcphosphorylation in competent oocytcs resulted in the loss of the ability to express meiotic competence in culture, whereas maintenance of centrosome phosphorylation in these oocytes was correlated with the ability to resume meiosis. These results suggest that the G,/M transition that occurs when mouse oocytes progress from an incompetent to a competent state iw ciw involves the phosphorylation of centrosomes and that the maintenance of centrosome phosphorylation is required for the it/ ,?troexpression of meioticcompetence. ~~a 19%Academic Press. Inc.

is first observed in the mouse in oocytes obtained from 15-day-old animals (Szybek, 1972; Sorensen and Wassarman, 1976; Bachvarova et al., 1980; Wickramasinghe et nl., 1991). At the time of meiotic competence acquisition, characteristic M-phase-like alterations appear in the nucleus and cytoplasm of competent oocytes (Wickramasinghe et ul., 1991). These G,/M-like alterations include chromatin condensation, microtubule reorganization, and the emergence of phosphorylated cytoplasmic foci. The phosphorylated foci were detected using the antibody MPM-2, which is known to recognize phosphorylated centrosomes that appear at the G,/M cell cycle transition in mitotic cells (Vandre ef al., 1984; Centonze and Borisy, 1990). Recent evidence indicates that the centrosome is an important structure for regulating M-phase events that are initiated during the G/M cell cycle transition. In XenoplLs oocyte extracts, the phosphorylation of centro-

INTRODUCTION

During the course of oogenesis in mammals, oocytes enter meiosis around the time of birth and remain arrested at prophase of meiosis I until the time of ovulation. Just prior to ovulation, oocytes resume meiosis, progress through meiosis I, and arrest at metaphase of meiosis II. Upon removal from the ovarian follicle mammalian oocytes resume meiosis spontaneously and undergo meiotic maturation under appropriate culture conditions (Pincus and Enzmann, 1935; Cho et al., 1974; Schultz and Wassarman, 1977). Meiotic competence, the term used to describe the ability of an oocyte to resume meiosis, is acquired during the course of oogenesis and

’ To whom correspondence should be addressed at Department of Anatomy and (:ellular Biology, Tufts IJniversity Schools of Medicine, 136, Harrison Ave., Boston, MA 02111. FAX: (6171 956-6536. 0012-1606/98 $5.00 Copyright Zll rights

c 1992 by Academic Press, Inc. of reproduction in any form reserved.

62

ated from follicles using an enzyme digestion protocol somes observed in the presence of active M-phase prodescribed previously (Wickramasinghe ef t/l., 1991). Germoting factor (MPF)” elicits M-phase-like microtubule minal vesicle (GV) intact oocytes were washed three dynamics at the centrosomes (J’erde et ul., 1990; Beltimes and cultured in loo-p1 drops of modified BMOC-2 mont et crl., 1990). In contrast, interphase-like microtubule arrays were observed at centrosomes that were de- medium (Brinster and Cross, 1972) under filtered heavy phosphorglated either due to the absence of MPF or in mineral oil (Fisher, Lot No. 875080) at 37°C in a 5%’ CO, 5% O,, and 90% N, atmosphere. response to treatment of extracts with the phosphorylation inhibitor 6-dimethylaminopurine (GDMAP) (Verde ef (11.. 1990). Components of MPF, notably ~34”~“” and cyclin, have been localized to the centrosome immunocyOocytes were fixed using a microtubule stabilizing tochemically (Riabowol et trl., 1989; Bailly et crl., 1989; buffer-extraction fix for 20 min at 37°C and were stored Alfa et trl., 1990; Rattner et crl., 1990; Pines and Hunter, from 1 to 3 days in a phosphate-buffered saline (PBS) 1991), suggesting further that centrosomes play a key blocking solution supplemented with 1% bovine serum role in entry into M-phase. However, the precise relaalbumin, 2% normal goat serum, 0.2% powdered milk, tionship among centrosome phosphorylation, microtubule dynamics, and the early functions of MPF have not 0.1 ,%fglycine, and O.Ol’% Triton X-100 (Wickramasinghe been fully resolved. Moreover, the role of centrosomes in et crl.. 1991). Oocytes from each experiment mere sepadevelopmentally regulated cell cycle transitions during rated into two groups and processed as described below early embryogenesis has been examined in several lower to compare microtubule and centrosome organization or animal forms (Sluder ef ol., 1989; Raff and Glover, 1988; to ascertain the phosphorylation state of the centroPicard eftrl., 1987) and the mouse (Callarco-Gillam et ~1.. somes. All oocytes were labeled with Hoechst 33258 to 1983; Schatten cut ul., 1986). However, no studies have evaluate chromatin organization. To evaluate centrosomes in relation to microtubule addressed the role of this structure during the cell cycle organization, oocytes were labeled with the human autransitions associated with oogenesis in mammals when toimmune serum 5051, which recognizes centrosomal the foundation for later developmental events is being material (Callarco-Gillam ct crl., 1983), at a 1:50 dilution laid down. in PBS for 1 hr at 37°C followed by three washes in The present experiments were designed to examine blocking solution. Oocytes were then incubated with the role of centrosome phosphorylation during the develTexas red goat anti-human IgG (TRGAH; Accurate opment and functional expression of meiotic compeChem. Co., Lot No. F2077) at a 1:lOO dilution for 1 hr at tence in mouse oocgtes. Various experimental conditions were established in which centrosome phosphory37°C and washed three times. These samples were sublation status was altered in incompetent and competent sequently incubated with a mouse IgG monoclonal antimouse oocgtes and the consequences of these manipulabody raised against fi-tubulin (Accurate Chem. Co., Lot tions mere evaluated with respect to microtubule organiNo. NU2088) at a 1:50 dilution for 1 hr at 37°C and, zation and the ability to resume meiosis i)l r,ifw. The following three lvashes, kvere further incubated with a results indicate that centrosome phosphorylation oc- fluoresceinated goat anti-mouse IgG (FGAM; Cooper curs i,l I+,YJ at the time of competence acquisition and Biomedical Inc., Lot No. 22378) for 1 hr at 37°C. To dithis alteration in the centrosome is related to the funcrectly evaluate centrosome phosphorylation, oocytes tional expression of the competent state ir/ I-ifm. Lvere colabeled with the human 5051 antibody and the mouse monoclonal antibody MPM-2, which recognizes phosphorylated centrosomes (Vandre ef cd., 1984). All MATERIALS AND METHODS antibodies were used at a 1:50 dilution for 1 hr at 37°C. Primary antibodies were probed with appropriate species-specific secondary antibodies as described above. Oocytes were isolated from 22- to 25-day-old female Oocytes were mounted on slides using a 50% glycerolCF-1 mice (Charles River Labs, Wilmington, MA and PBS solution containing 1 &ml of Hoechst 33258 (PolyHarlan Sprague-Dawley, Indianapolis, IN) and libersciences Inc.) and 25 mg/ml of sodium azide as an antifading reagent as described previously (Wickrama’ Abbreviations used: GV, germinal vesicle; GI’BD, germinal vesicle singhe et ~1.. 1991). Control labeling was carried out on breakdown; MPF, maturation-promoting factor; PK-A, protein kinase oocytes incubated with the secondary antibody alone for .I; PK-C, protein kinase-C; BMOC-2, Brinster’s modified oocyte cul1 hr at 37°C. In addition, alkaline phosphatase-treated ture medium; PBS, phosphate-~luffcred saline; dbcAMP. dibutyrgl cyoocytes were labeled with MPM-2 as a control to test for clic adenosine monophosphate; iiDMAP, ti-dimethylaminopurinc; specific labeling of phosphorylated epitopes. No labeling PMA. _ nhorbol 1%mvristate 13.acetate: DMSO. dimethvl sulfoside: IBMX, :I-isohutgl-I-methylsxnthinr. was observed under these conditions.

64

DEVELOPMENTAL BIOLOGY

Labeled oocytes were analyzed using a Zeiss IM-35 microscope equipped with fluorescein (Zeiss 487709), rhodamine (Zeiss 487714), and Hoechst (Zeiss 487702) selective filter sets and a 50-W mercury arc lamp. Epifluorescence and phase optics were employed for this analysis. Data were expressed as a pooled sample from each experiment inclusive of oocytes labeled with 5051/ anti-tubulin and 5051/MPM-2. Oocytes were photographed on Tri-X-pan film with exposure times ranging from 1 to 10 see using 63X or 25X neofluar objective lenses. The film was processed using full-strength Acufine developer for 5.25 min at 25°C.

In one set of experiments, oocytes were isolated using the same protocol mentioned previously and cultured for 50 min at 37°C in medium containing 10 p&ml of nocodazole (Aldrich Chemical Co.). These oocytes were washed three times in drug-free medium and allowed to recover and assemble their microtubules for 5,10, or 15 min prior to fixation. The oocytes were triple-labeled with the centrosome antibody, 5051, and anti-tubulin antibodies, followed by Hoechst 33258. In another set of experiments isolated oocytes were cultured, within 20 min of isolation, in medium containing dibutyryl adenosine cyclic monophosphate (dbcAMP, Sigma) at a concentration of 200 PM, phorbol12myristate 13-acetate (PMA, Sigma) at 10 &ml, or GDMAP (Sigma) at 5 m,%? These agents have been shown to maintain meiotic arrest in competent mouse oocytes through different mechanisms (Cho et al., 1974; Bornslaeger et ol., 1986a,b; Rime ef ad., 1989). Incubations in various drugs were carried out for 14 hr under filtered heavy mineral oil (Fisher, Lot No. 875080) at 37°C in a 5% CO,, 5% 0,, and 90% N, atmosphere. Control medium containing 0.01% of the vehicle, dimethyl sulfoxide (DMSO), for PMA and nocodazole did not perturb oocyte morphology or ill tlifro maturation ability when compared to oocytes maintained in medium alone. In a separate experiment, oocytes were isolated as before and fixed at the time of isolation (T,) or were cultured in either freshly prepared 3-isobutyl-l-methylxanthine (IBMX, 200 PM, Aldrich Chemical Co.), an inhibitor of phosphodiesterase which arrests meiotic maturation (Bornslaeger ef (xl., 1986a; Downs, 1990), or both IBMX (200 rev) and puromycin (10 &ml, Sigma) for 6 hr (T,). Half of the oocytes from each of these groups were fixed separately while the remaining oocytes were washed three times in control medium and cultured in medium containing puromycin alone (10 &ml) for a further 3 hr and then fixed. Oocytes were cultured in medium containing 0.24 ethanol as a control for IBMX

VOLUME lc52.1992

and did not differ from control oocytes cultured dium alone in their ability to resume meiosis.

in me-

Previous studies in the mouse established that incompetent oocytes could be distinguished from competent oocytes by discrete nuclear and cgtoplasmic characteristics (Wickramasinghe et ol., 1991). Based on GV chromatin patterns described previously, fixed oocytes were classified as incompetent or competent after various experimental manipulations were carried out in culture. The percentage of oocytes under each experimental condition was evaluated with respect to centrosome number (5051 foci number), phosphorylation state (5051/ MPM-2 colocalization), and microtubule organization (505l/anti-tubulin). RESULTS

In order to identify centrosomes and determine their phosphorylation status in relation to microtubule organization, competent and incompetent oocytes were fixed at the time of isolation and labeled with 5051 and MPM2 antibodies or with 5051 and anti-tubulin. Incompetent oocytes contain diffuse chromatin within the GV (Fig. 1A) and display cytoplasmic foci labeled with 5051 (Fig. 1B) but not MPM-2 (Fig. lC), indicating that the centrosomes are not phosphorglated. Anti-tubulin staining in incompetent oocytes reveals a network of long interphase-like microtubules (Fig. 1E) and randomly distributed centrosomes in the cytoplasm detected by 5051 labeling (Fig. 1D). In contrast, competent oocytes containing condensed chromatin (Fig. IF) display 5051 reactive foci (Fig. 1G) that react with the MPM-2 antibody (Fig. 1H). Characteristically, short microtubules emanate from centrosomes in competent oocytes (Figs. 11 and IJ). These observations indicate that centrosomes are detectable in both incompetent and competent mouse oocytes using the 5051 antibody but appear to be phosphorylated only in competent oocytes. Moreover, differences in cytoplasmic microtubule organization are also apparent in competent and incompetent oocytes. To further investigate whether the differences observed in microtubule organization in freshly isolated oocytes are related to centrosome phosphorylation, microtubule depolymerization was induced by treatment with nocodazole and patterns of microtubule regrowth were evaluated in both competent and incompetent OOcytes at various times following culture in control medium. Figures 2A-2C show a GV intact incompetent OOcyte exposed to 10 pg/ml of nocodazole for 50 min and illustrate complete depolymerization of all microtubules as shown by anti-tubulin labeling (Fig. 2C); cen-

trosomes remained intact after nocodazole treatment (Fig. 2B). In incompetent oocytes (Fig. ZD), long microtuhules of variable length were observed radiating from centrosomes following a 5-min recovery in control medium (Figs. 2E, 2F). In competent oocytes (Fig. 2G), an array of short microtubules (Fig. 21) emanated from centrosomes (Fig. 2H) after a 5-min recovery. Fifteen minutes after nocodazole washout, extensive arrays of interphase microtubules were apparent in incompetent oocytes (Figs. ZJ-21,) whereas competent oocytes exhibited microtubules at the centrosome following a similar recovery period (Figs. 2M-20). Thus, centrosomes from incompetent and competent oocytes nucleate microtubules following nocodazole removal. But centrosomes in competent oocytes nucleate microtubules in a more localized pattern compared to their incompetent counterparts and both types of oocytes restore their distinctive patterns of microtubule organization upon recovery from drug-induced disassembly. Since centrosomc phosphorylation and microtubule organization distinguish incompetent from competent oocytes, we next examined the effect of agents known to modify protein phosphorylation with respect to both cgtoskeletal organization and progression into M-phase. Oocytes were cultured for 13 hr in medium containing agents known to maintain meiotic arrest through different mechanisms (Cho r’f trl., 1971; Bornslaeger of (I/., 1986a,b, 1988; Schultz et t/l., 1983). These included dhcAMP, an activator of protein kinase A (PK-A, 200 VU); PMA, a protein kinase C (PK-C, 10 n&ml) activator: or 6DMAP (5 mM), an inhibitor of protein phosphorylation. Oocytes from each treatment group wore processed as described above and evaluated for centrosome phosphorylation and microtubule organization. A common response to all of these treatments was the formation of long microtubules in competent oocytes (Figs. 3E, 35). Centrosome phosphorylation was maintained in competent oocytes treated with dbcAMP (Figs. 3B and 3C) or PMA (not shown), whereas GDMAP treatment resulted in centrosome dephosphorglation (Figs. 3G and 3H). All oocytes treated with 6DMAP exhibited condensed chromatin around the nucleolus, resembling the GVs of competent oocytes (Fig. 3F). Moreover, patches of MPM-2 staining were observed routinely in the GV of oocytes treated with GDMAP (Fig. 3H), but this staining did not colocalize with centrosomes (compare with Fig. 3G ). These results are summarized in Table 1 from triplicate experiments for centrosome phosphorylation, microtubule organization and meiotic competence. Control incompetent oocytes fixed at the time of isolation consistently display dephosphorylated centrosomes and interphase microtubules (Table 1, line 1). In contrast. competent oocytes uniformly exhibit phos-

phorylated centrosomes and lack interphase microtubule arrays (Table 1, line 7 ). In addition, the mean number of centrosomes/cell is similar in incompetent (2.9) and competent (3.6) oocytes at the time of isolation. When cultured for 14 hr, incompetent oocytes retain GVs and interphase microtubule arrays (line 2), whereas the competent fraction of oocytes undergoes GVBD (line 8). Oocytcs treated with the phosphodiesterasc inhibitor IBMX for 11 hr (200 p&l) failed to undergo GVBD and this treatment causes the appearance of interphase microtubules in competent oocytes without affecting either centrosome phosphorylation status or ccntrosome number (line 9). Fourteen-hour trcatmerits of competent oocytes with either dhcAMP or PMA similarly resulted in the appearance of interphase microtubule aira!-s and centrosomes remained I)hosphorylated (lines 10, ll), whereas unphosphorylatcd centrosomes were typically found in all GDMAP-treated oocytes (line 6). The various experimental conditions tested did not alter microtubule organization in incompetent oocytes, although the mean number of centrosomes in incompetent oocytes increased after 14 hr of culture (2.9 to 5.X) and decreased after a 14hr IBMX treatment (2.9 to 1.6). Although the significance of these changes is not clear, two to three centrosomes wrtr present in most oocytes under the different culture conditions employed (Figs. lB, 3B, 3D, 3G, 31, 4B, 1F). Thus, competent oocytes exhibit an interphase microtubule response to all treatments examined, although centrosome phosphorylation status was altered only in the presence of 6DMAP. The nest experiments were designed to address the question of whether ccntrosome phosphorylation was required for meiotic resumption to occur itr vitro. Culture conditions were therefore sought that would allow for reversible centrosome phosphorylation in cultured competent oocptos. Since it was shown previously that maintaining meiotic arrest in rodent oocytes in the prcsence of puromycin reversibly inhibits their ability to resume meiosis (Eckholm and Magnusson, 1979; Downs, 1990), we first examined centrosome phosphorylation under these conditions. Competent oocytes cultured in IBMX alone for 6 hr retained phosphorylated ccntrosomes (Figs. 4A-4C) and displayed interphase microtubules (Fig. 4D). Notably, competent oocytes treated with IBMX and puromycin for 6 hr displayed nonphosphorylated centrosomes and interphase microtubules (Figs. 4E-4H). When oocgtes cultured in IBMX alone for 6 hr were transferred to medium containing puromycin alone for 3 hr meiotic resumption ensued as evidenced by GVBD, chromatin condensation, and increased levels of cgtoplasmic phosphorylation detected by MPM-2 staining (Figs. 31-3K). In contrast, competent oocytes exhibiting greatly diminished ccntrosomc dephosphor-

66

DEVELOPMENTALBIOLOGY ~0~~~~152,19!32

glation as a result of a 6-hr incubation in IBMX and puromgcin were unable to resume meiosis when transferred to puromycin-containing medium (Figs. 4L-40). In this treatment group, GVs were maintained intact with prominent nucleoli evident (Figs. 4L, 4N), although minimal chromatin condensation was detectable after labeling with Hoechst 33258 (Fig. 4M). The incompetent oocyte fraction in these preparations was unaffected with respect to centrosome phosphorglation or microtubule organization. These results indicate that competent oocytes containing phosphorylated centrosomes resume meiosis in the presence of puromycin while those harboring dephosphorylated centrosomes are unable to resume meiosis under identical conditions. These experiments are summarized in Table 2 (lines 3-6) for all competent oocytes evaluated. Oocytes maintained in IBMX for 6 hr retained phosphorylated centrosomes (line 3, 100% 1, whereas only 13% of oocytes cultured in IBMX and puromycin for 6 hr exhibited phosphorylated centrosomes (line 4). All oocytes treated with IBMX alone and released into purompcin undergo GVBD (line 5), whereas only 19% of the oocytes treated with IBMX and puromycin initially were able to undergo GVBD in the presence of puromytin. In this set of experiments, all control competent oocytes contained phosphorylated centrosomes at the time of isolation (line 1) and resumed meiosis when cultured for 9 hr in control medium (line 2). Additional control treatments established the reversibility of IBMX treatment alone since all oocytes so treated resumed meiosis during a subsequent 3-hr period (line 7). Furthermore, following the induction of centrosome dephosphorylation using the combined IBMX and puromycin treatment, these oocytes when transferred into control medium exhibited either rapid centrosome rcphosphorylation (22%~) or the immediate expression of meiotic competence during a 3-hr period (‘78%) line 8). Finally, the rephosphorylation of centrosomes is full\ reversible since 77% of competent oocytes cultured initially in IBMX and puromycin for 6 hr, then puromycin for 3 hr and subsequently transferred into medium containing IBMX alone, exhibit phosphorglated centrosomcs (line 9). These results demonstrate that centrosome phosphorylation in competent oocytes maintained in meiotic arrest requires ongoing protein synthesis. Furthermore, the results of the experimental manipulations described

above suggest that the functional expression of meiotic competence in/ &l*o is associated with the state of centrosome phosphorylation.

I)IS(‘IJSSION

The present experiments were designed to investigate the role of centrosome phosphorylation in the developmental acquisition and functional expression of meiotic competence, a major cell cycle transition state that occurs during oogenesis. The results show that competent oocytes display phosphorglated centrosomes, whereas incompetent oocytes contain nonphosphorylatcd centrosomes and exhibit an interphase network of cytoplasmic microtubules. Moreover, upon recovery from nocodazole treatment, the centrosomes in competent oocytes exhibit M-phase-like nucleation characteristics, whereas those in incompetent oocytes display an interphase pattern of microtubule growth. Finally, a functional and reversible GJM cell cycle transition is documented since centrosome phosphorylation is correlated experimentally with the ability to resume meiosis i,r t>itro.

These results confirm and extend our previous studies that suggested that the i?r I+~Y~developmental progression from an incompetent to a competent state in mouse oocytes involves a G,/M cell cycle transition (Wickramasinghe rjt ~1.. 1991). A key event in the GJM transition of somatic cells is the phosphorylation of centrosomes, which is thought to be responsible for alterations in centrosome microtubule nucleation capacity needed for spindle assembly during mitosis (Vandre c$ trl., 1984). The present experiments illustrate a difference in microtubule nucleation capacity of centrosomes in incompetent or competent oocytes. When nocodazoletreated mouse oocgtes are allowed to reform microtubules, the initial reassembly occurs at the centrosome in both incompetent and competent oocytes. However, short microtubules reassemble in the competent oocytes relative to their incompetent counterparts. This distinction in microtubulc regrowth is conspicuous upon complete recovery from nocodazole treatment in that incompetent oocytes reestablish long interphase arrays of microtubules, whereas short microtubules are evident at the centrosomes of competent oocytes. Our in /)i~o results are consistent with the observations of Verde rot

FIG. 1. (~kntrosornts :LI‘L’phosphorylatcti anti nurlcate short microtuhnles in compc~tcnt oocytcs (‘orrcxlati\e micrographs of’ reprtsentatirt incomprtent ilcft) and comptcnt oocytw (right) labeld with Hoechst 3325X (A, PI. 50.51 (B. GJ anti MPM-2 (C, H). or 5051 CD, 1, and anti-tuhulin (E. J) to c~vatnate chromatin configurations. ccntrosome J)hosl)hor~lation, and microtuhule organization. (‘onrprtent oocytrs colocaliw 5051 and MPM-2, indicating the prrscncc of I,hosphorglatcd wntrosomc~s (G, 11. :~rro\vhcwls) which nucleate short microtutmlrs (I. .J); incompratrwt oocytcs cxhibit tlephosJ)hor!-l~~t~~(l rrntrosomes (R, C, xrro\vhc;ds, and interphaw microtuhulr arrays (E). Scale bar I E) reprcwnts 20 pm; all microgr;~phs arr c~nlargetl to the same magnification.

68

DEVELOPMENTALBIOLOGY

VOLUME~V < , 192 ftrl., 1988), in the expression of mciotic competence. Additional experiments were undertaken to determine the functional significance of centrosome phosphorylation with respect to the actual expression of meiotic competence. When IBMX was used to maintain meiotic arrest, all competent oocytes examined exhibited phosphorylated centrosomes (Table 2). In contrast, incubation of competent oocytes in IBMX and puromycin for 6 hr resulted in the dephosphorylation of centrosomes. Oocytes containing phosphorylated centrosomes (IBMX alone) were able to resume meiosis upon release

FIG. 2. L)iffvrential microtuhulc nucleation capacity of centrosomes in incompetent anti comlwtchnt oocytes after recovery from nocotfaz~l~~mctiiatd microtubule depol~merization. Corrr‘lative micrographs of incompetent (A-(‘, II-F, J-L) anti competent oocytes (G-l, M-0) at 0 min (rl-C’). ,T nrin ID-I), or I5 min (J-0) after recovery from nocotiazole. Oocytes \vere laheleti Lvith Hoechst 33258 (left) for chromatin conficur:ltions, T,O.>l (middle) to identify centrosomes , and anti-tuhulin (right) to evaluate microtuhule growth. Arrowheads in E and F indicak a wntrosome nuclcatin~ long microtuhules in an incompetent oocyte compared to shorter microtul,ulcs (arron~heads H, I) in comlwtent oocytes lisctl 5 min a1’ter drug recovery. Scale bar (11) represents 20 pm; all micrographs arc cwlars& to th(, same final ma~nilication.

70

DEVELOPMENTALBIOLOGY

V01,~~~152,1992

WICKRAMASINGHEANDALBERTINI

71

cer1t roso,,IPs ill !~fo/rsP ckJC!&‘S

TABLE 1 CENTR~~OMEANDMICROTUBULECHARACTERISTICSOFCONTROLANDDRUG-TREATEDGROWINGMOUSEOOCYTES 5 Cells with Treatment group

No. oocvtes examined

Incompctcmt Lint 1 Lint 2 Line ;1 Line 1 Line 5 Lint 6

TO Tl, IBMX T,, tll,cllMP T,, PMA T,, 6DMAP T,,*

cid 34 30

(‘onipctc~nt I,ine 7 Lint X Line 9 T,ine 10 T,inc 11

To T 1:MX T,, dbrAMP T,, PMA T,,

‘ZGVBD

Interphase microtubules

Phosphorylated centrosomes

Mean No. centrosomes/cell

69 99

0 0 0 0 0 0

100 100 100 100 100 100

0 0 0 0 0 0

2.9 i 0.3 (96) 3.8 i 0.7 (29) 1.6 + 0.2 (40) N.D. N.D. N.D.

13 21 29 60 44

0 100 0 3 9

0 0 100 97 91

100 100 100 97 91

3.6 i 0.5 (88, N.D. :I.8 i- 0.1 (76) N.D. N.D.

.iO

(~1)

,V~:o/e.Oocytw were isolated as described under Materials and Methods and fixed at the time of isolation (T,), 14 hr after culture in control n~edium (T,, 1, or in medium containing dbcAMP, PMA, or GDMAP. Oocytes were evaluated as described under Materials and Methods with respect to mciotic competence expression (% GTTBD), microtubule display (Interphase microtubules), and centrosome phosphorylation. (~1 Numhrr of oocytes analyzed; N.D., not determined. (*) Distinction between incompetent and competent oocytes could not be made by evaluation of chromatin patttlrns as nucleolar rimming occurred in all oocytrs treated with GDMAP.

into medium containing puromycin, whereas oocytes containing dephosphorylated centrosomes (IBMX and puromycin) were unable to resume meiosis under similar conditions. IBMX- and puromgcin-treated oocytes, harboring dephosphorylated centrosomes, mere washed into medium containing puromycin alone for 3 hr followed by medium containing IBMX alone for 3 hr to further maintain meiotic arrest. Since all of the competent oocytes examined now exhibited phosphorylated centrosomes, the appropriate factors needed to reversibly alter the phosphorylation state of centrosomes must have been acquired during development and resynthesized from stored maternal mRNA. These results support the notion that the ability of competent mouse oocytes to resume meiosis irr tlifm is dependent on conditions that maintain centrosome phosphorylation. It may have been fortuitous that rodent oocytes were chosen to study this important developmental process. Rodent oocytes, unlike amphibian and other mammalian oocytes, resume meiosis in the presence of protein synthesis inhibitors (Wassarman et nl., 1976; Eckholm and Magnusson, 1979), suggesting that sufficient stores of

MPF exist to support the early events of M-phase entry. This property, which may be unique to rodent oocytes, is consistent with our earlier suggestion that MPF and/or factors regulating its activation are expressed at the time of competence acquisition but could be tempered by influences presumably derived from the follicle cells until meiotic maturation is triggered hormonally at the time of ovulation (Wickramasinghe et (II., 1991). Since prolonged inhibition of protein synthesis (>6 hr) in meiotically arrested rodent oocytes impairs their ability to spontaneously resume meiosis in culture (Downs, 1990; Eckholm and Magnusson, 19791, it is likely that under these conditions, MPF stores are being depleted. When coupled with the observation that the loss of ccntrosome phosphorylation coincides with the loss of the ability to resume meiosis, it would appear that the centrosome is a primary target for MPF expressed at the time of competence acquisition. In summary, the results derived from localization studies, pharmacological manipulations, and developmental patterns of expression with regard to centrosome dynamics in the growing mouse oocyte indicate

FIG. 3. Correlative micrographs of oocytes treated with dbcAMP (left) and GDMAP (right). Oocytes labeled with Hoechst 33258 (A, F) to cvaluatc competence status (based on chromatin staining pattern; see text), 5051 (8, G) and MPM-2 (C, H) to evaluate the phosphorylation state of ccntrosomr, or 5051 (D, I) and anti-tuhulin (E, J) to evaluate tnicrotubule organization. Centrosomes remain phosphorglated (arrowheads 8, (‘1 while interphase-like microtubules are present (D, E) in dbcAMP-treated competent oocytes and centrosomes are dephosphorylated (arrowhead G, II) and interyhaw-like microtubules are observed (I, J) in all 6DMAP-treated oocytes. Scale bar (El represents 20 Mm; all micrographs are enlarged to the same magnification.

72

DEVELOPMENTAL BIOLOGY

VOLUME 152, 1992

FIG. 4. Correlation of centrosome phosphorylation with the ability to resume meiosis in competent oocytes. Oocytes triple-labeled with Hoechst 33258 for chromatin configurations (A, E, I, M), 5051 (B, F, J, N, D, H) and MPM-2 (C, G, K, 0) for centrosome phosphorylation, or anti-tubulin (D, H) for microtubule organization and phase (L) for evaluation of the GV. Oocytes treated with IBMX (A-D; D inset above, 5051) or IBMX and puromycin (E-H; H inset above, 5051) for 6 hr are shown. Note the absence of centrosome phosphorylation in IBMX- and puromycin-treated oocytes (arrowheads F, G). Oocytes treated with IBMX alone (I-K) and IBMX and puromycin (L-O) for 6 hr were released into puromycin alone for 3 hr. Note GVBD in oocytes treated with IBMX alone followed by culture in puromycin, while oocytes in IBMX and puromycin released into puromycin maintain nucleoli (L, small open arrowhead), an intact GV (L, small closed arrowheads), and no centrosome phosphorylation (compare N with 0, arrowheads). Scale bar represents 20 pm. A-H, scale bar in E; I-O, scale bar in L.

TABLE 2 RELATIONSHIP OF CENTROSOME PHOSPHORYLATION AND IN VITROEXPRESSION OF MEIOTIC COMPETENCE

Treatment I,intx Line Line Lint,

1 2 3 4

Lint 5 Lincx 6 Lint 7 Linca X Lint 9

group

TO T, IMBX (T, j IBMX + PIJRO (T,) IBMX (T, I/ PIJRO (T,) IBMX + PURO CT,)/ PURO (T, i IBMX (T, )/ MEDIIJM (Ta) IBMX t PIJRO (T, ,/ MEDIIJM (TR 1 IBMX + PURO (T,)/ PIJRO (T, )/IBMX (T, )

No. oocytes examined

% G’V’BD

‘% of cells with phosphorylated centrosomes

31 47 27 23

0 100 0 0

100 100 100 13

70

100

N.D.

f,3

19

8

35

100

100

37

7x

22

22

23

77

Sots,. Oocytes were isolated as described under Materials and Methods and cultured under various experimental conditions as described under Matwials and Methods. Data are expressed as a percentage., of the total number uf competent oocytes that underwent GVBD (‘XGVBD) or displayed phosphorylated centrosomcs. N.D., not determined

collectively that this structure plays a crucial role in the regulation of the meiotic cell cycle. Further insights into the control of meiotic progression and competence acquisition in animal oocytes will require examination of the interactions among centrosomes, microtubules, and the components of MPF itself before a full appreciation of the role of centrosomes in early development can be established. Wv arc grateful to Drs. Potu Rao and Marc Kirschner for providing the MPM-2 and 5051 antibodies, respectively, and to Dr. Nancita R. Lomas of the National Cancer Institute for supplying taxol. We thank Drs. \:irginia Rider, Ann Allworth, and Susan Messinger for suggestions and advice and Mary Currier for help in preparing the manuscript. Portions of this work were submitted to the Sackler Graduate School of Biomedical Sciences in partial fulfillment of the requiremcnts for thr Ph.D. degree iD.1V.i. These studies \vcsre supported by NIII Grant IID 2006X.

REFEREN(:ES Alfa. C. E., Ducommun, B., Beach, D., and Hyams, J. S. (1990). Distinct nuclear and spindle pole body population of cyclin-cd&! in fission yeast. :L;ittiw 347, 680-6X2. Bachvarova, R., Baran, M. M., and Te,ihlum. A. (1980). Development of naked growing mouse oocytes in vitro. rJ. Exp %ool.211, 159-169. EGllv, E., Doree, M., Nurse, I’., and Bornens, M. (1989). P34cdc2 is located in both nucleus and CytOphSm; part is centrosomally associatcd at GX/M and enters vesicles at anaphasc. E,W,!30 ,/. 8, 3X35:1995. Belmont. 1,. D.. Hyman, A. A., Saa-in, K. E., and Mitchison, T. J. (1990). Rewl-time visualization of cell cycle-dependent changes in microtuhulc dynamics in cytoplasmic vxtraets. @II 62, 579-589. Bornslaegvr, E. A., Mattvi, P., and Schultz, R. &I. (1986a). Involrcment

of AMP-dependent protein kinasc and protein phosphorglatian in regulation of mouse oocyte maturation. Lkv. Rid. 114, 453-462. Bornslaeger, E. A., Pouch-mirou, TV. T., Mattei, P., and Schultz, R. M. (1986b). Effects of protein kinase C activators on germinal vesicle breakdown and polar body emission of mouse oocytcs. Exp Cc,//Rcs. 165,507917. Bornslacger, E. A., Mattei. P. M., and Schultz, R. M. (19%). Protein phosphorylation in meiotically competent and incompetent mouse oocytes. X01. Kt~)rorl. Ilev. 1, 1%2% Brinster, R. L., and Cross, P. C. (197’2). Effect of coppcsr on the preimplantation mouse embryo. Av/rt2r~~(~ 238, X%39!). Callarco-Gillam, P. D., Siebert, 11.. Hubble, R., Mitchison, T.. and Kirschner, M. (1983). Centrosome development in early mouse: embryos as defined by an autoantibody against pericentriolar matcrial. Cf4/ 35, 681-629. Centonze, V. E., and Borisy, G. G. (1990). Nucleation of microtubules from mitotic ccntrosomes is modulated by a phosphorylated cpitopc. J. (‘id sci. 95, 405-411. Cho. W. K., Stern. S., and Biggers, J. D. (1974). Inhibitory effect of dibutyryl clZMP on mouse oocgte maturation in vitro. .I. I;;,r~j. Zool. 1x7, 3x:3-:%33. Downs, S. M. 11990~. Protein synthesis inhibitors prevent both spontaneous and hormone-deyndent maturation of isolated mouse oocytes. Alo/. f?cjwotl. I)(,(,. 27, 235-243. Ekholm, C., and Magnusson, C. (1979). Rat oocyte maturation: Etfects of protein synthesis inhibitors. Rio/. R~protl. 21, 1287-121X1. Mattson, B. A., and Albertini, D. F. (1990). Oogenesis: (:hromatin and microtubule dynamics during meiotic prophase. :Iloi. R~prvtl. ,%,z*. 25, :374-x3:1. Picard, A., Karsenti, E., Dahauvalle, M. C., and Doree, M. (1987). Release of mature starfish oocgtes from interphase arrest t)y mieroinjcction of human ccntrosomes. Xot~cw 327, 170-172. Pincus, E., and Enzmann. E. V. (1935). The comparative behaviour of mammalian eggs in viva and in vitro. I. The activation of ovarian eggs. ,J. E,p :lLwl. 62, 665-675. Pines, J., and Hunter, T. (1991). Human cyclins A and B2 are ditl’erentially located in the cell and undergo cell cycle-dependent nuclear transport. ./ (‘fj// !?h~/. 11.5, l-17.

74

DEVELOPMENTALBIOLOGY

Raff, J. W., and Glover, D. M. (1988). Nuclear and cgtoplasmic mitotic cycles continue in Drosophila embryos in which DNA synthesis is inhibited with aphidicolin. J. Cell Biol. 107, 2009-2019. Rattner, J. B., Lew, J., and Wang, J. H. (1990). p34cdc2 kinase is localized to distinct domains within the mitotic apparatus. Cc/l Mofil.

('/Joske/. 17, 227-235. Riabowol, K., Draetta, G., Brizuela, L., Vandre, D., and Beach, D. (1989). The cdc2 kinase is a nuclear protein that is essential for mitosis in mammalian cells. CPU 57, 393-401. Rime, H., Neant, I., Guerrier, P., and Ozon, R. (1989). 6-Dimethylaminopurine (6-DMAP), a reversible inhibitor of the transition to metaphase during the first meiotic cell division of the mouse oocgte. De/:. Bid. 133, 169-179. Schatten, H., Schattcn, G., Mazia, D., Balzcou, R., and Simerly, C. (1986). Behavior of centrosomes during fertilization and cell division in mouse oocytes and in sea urchin eggs. Proc. Arofl. Amd. Sci. 1LS’A83, 105-109. Schultz, R. M., Montgomery, R. R., and Belanoff, J. R. (1983). Regulation of mouse oocyte meiotic maturation: Implication of a decrease in oocyte cAMP and protein dephosphorylation in commitment to resume meiosis. Dcr,. Rio2. 97, 264-273. Schultz, R. M., and Wassarman, P. M. (1977). Biochemical studies of

VOLUME1.52 1992

mammalian oogenesis: Protein synthesis during oocyte growth and meiotic maturation in the mouse. J. Cell Sei. 24, 167-194. Sluder, G., Miller, F. J., and Rieder, C. L. (1989). Reproductive capacity of sea urchin centrosomes without centrioles. Cell Motil. C~toskel.

13,264-27X Sorensen, R. A., and Wassarman, P. M. (1976). Relationship between growth and meiotic maturation of the mouse oocyte. i%e. Biol. 50, 531-536. Szybck, K. (1972). In vitro maturation of oocgtes from sexually immature mice. J. Edocri~rol. 54, 527-528. Vandre, D. D., Davis, F. M., Rao, P. N., and Borisy, G. G. (1984). Phosphoproteins are components of mitotic microtubule organizing centers. I’tnc. Nufl. Actrd. SC?. lLSA 81, 4439-4443. Verde, F., Labbe, J., Doree, M., and Karsenti, E. (1990). Regulation of microtubule dynamics by cdc2 protein kinase in cell-free extracts of Xenopus eggs. ,Virfnrc 343, 233-238. Wassarman, P. M., Josefowicz, W. J., and Letourneau, G. (1976). Meiotic maturation of mouse oocytes in vitro: Inhibition of maturation at specific stages of nuclear progression. J. Cdl Sfi. 22,431-445. Wickramasinghe, D., Ebert, K. M., and Albertini, D. F. (1991). Mciotic competence acquisition is associated with the appearance of Mphase characteristics in growing mouse oocytes. L)rr). Biol. 143,162172.