MOLECULAR REPRODUCTION A N D DEVELOPMENT 25:374-383 (1990)
Oogenesis: Chromatin and Microtubule Dynamics During Meiotic Prophase BRITTA A. MATTSON AND DAVID F. ALBERTINI Department of Anatomy and Cellular Biology, Tufts University Health Sciences Center, Boston, Massachusetts ABSTRACT Changes in the organization of germinal vesicle chromatin in mouse oocytes have been analyzed by fluorescence microscopy with respect to progressive stages of follicular development and the disposition of oocyte cytoplasmic microtubules. Four discrete patterns of chromatin organization exist in germinal vesicle (GV)stage oocytes isolated from the ovaries of 2125-day-old gonadotropin-primed mice. Analysis of ovarian cryosections stained with the DNA-binding fluorochrome Hoechst 33258 indicates that sequential changes in GV chromatin occur duing folliculogenesis that result in the formation of a continuous perinucleolar chromatin sheath at the time of antrum formation. Specific alterations in the cytoplasmic microtubule complex of GV-stage oocytes were observed that correlate with chromatin patterns. The extensive cytoplasmic microtubule complex seen in oocytes of preantral follicles initially localizes to perinuclear areas of the ooplasm. This is followed by a progressive reduction in cytoplasmic microtubules and the appearance of prominent microtubule-organizing centers at the nuclear periphery. Coordinated nuclear and microtubular alterations also occur under in vitro conditions prior to progression of meiosis to prometaphase-1. The results are discussed with respect to the ongoing differentiation of the oocyte nucleus and the microtubule cytoskeleton during folliculogenesis in preparation for the resumption of meiosis. Key Words: Ovarian follicle, Oocyte, Germinal vesicle, Chromosomes, Microtubules
INTRODUCTION The events that immediately precede the resumption of meiosis in mammalian oocytes are a subject of considerable interest with respect to the functional interrelationship of oogenesis and folliculogenesis. Although much is known about the growth and differentiation of the oocyte during follicular development as i t relates to meiotic competence, metabolism, and structural organization (Anderson et al., 1977; Wassarman et al., 1979; Moor et al., 1983; Tsafriri, 1985; Schultz, 1986; Thibault et al., 19871, little is
0 1990 WILEY-LISS, INC.
known about the nuclear and cytoplasmic transitions that are prerequisites in the preparation of the prophase-arrested oocyte for meiotic maturation in the preovulatory follicle. Correlative changes in nuclear structure and transcriptional activity have been shown to occur in mammalian oocytes a t specific stages of follicular development (Motlik and Fulka, 1986). Chouinard (1975) showed by both light and electron microscopy that heterochromatin and nucleoli undergo extensive reorganization in the germinal vesicles (GV) of oocytes in growing, preantral follicles of prepubertal mice. Chouinard (1975) noted little change in the nuclear morphology during subsequent growth of antral follicles, a time during which the follicle is profoundly influenced by gonadotropins. Moore et al. (1974) have shown that RNA synthesis and RNA polymerase activity in mouse oocytes are maximal prior to antrum formation and remain low in oocytes during subsequent follicular growth. In the human oocyte, however, heterochromatin and nucleolar rearrangements as well as a decrease in 3H-uridine incorporation have been noted to occur following antrum formation (Parfenov et al., 1989). Although similar nuclear changes have been noted in other mammalian species during oogenesis a t specific stages of follicular development (see Parfenov et al., 1989), it is not known whether ooplasmic structures are also modified in their organization prior to the resumption of meiosis. Moreover, if changes in cytoplasmic organization do occur, the extent to which these alterations can also be correlated with changes in nuclear structure and function andlor formation and development of antral follicles is not known. The objectives of this study were twofold: first, to determine if discrete patterns of chromatin organization can be identified during meiotic prophase in vivo and in vitro and whether these changes are sequentially expressed; second, to determine whether changes in chromatin organization can be correlated with remodeling of the
~
Received July 27, 1989; accepted November 14, 1989. Address reprint requests to Britta A. Mattson, Department of Anatomy and Cellular Biology, Tufts University Health Sciences Center, 136 Harrison Avenue, Boston, MA 02111.
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS microtubular component of the cytoskeleton prior to and following the resumption of meiosis.
MATERIALS AND METHODS Animals Virgin, female, 21-25-day-old CD-1 mice (Charles River Laboratories, Wilmington, MA) housed in a controlled environment (14 h r light/lO hr dark) were used for all experiments. Follicular development was stimulated by i.p. injection of 7.5 I.U. pregnant mare's serum gonadotropin (PMSG; Diosynth, Inc., Chicago, IL). The animals were killed by cervical dislocation 44-48 h r later, and the ovaries were excised and placed in a petri dish containing Eagle's MEM (GIBCO, Grand Island, NY) supplemented with 25 mM HEPES buffer, Hank's salts, 100 Uiml penicillin, 100 pg/ml streptomycin, and 0.3% Pentex BSA (Miles Laboratories, Kankanee, IL) a t 25°C. Ovarian Cryosections Ovaries were washed, fixed in 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for 4 hr, and infiltrated for 48 h r with 2.3 M sucrose and 0.1% paraformaldehyde in phosphate buffer. Five to eight micrometer sections were mounted on gelatin-coated slides. Ovarian cryosections were incubated in 10 pgiml Hoechst 33258 (Polysciences, Inc., Warrington, PA) for 30 min at 25"C, washed, and mounted. Microscope slides were stored a t 4°C until analysis. Vital Staining Excised ovaries were immediately transferred to collection medium containing 10 pg/ml Hoechst 33258. Oocytes were collected by follicular puncture, mechanically stripped of cumulus cells, and pipetted into a 50 pl drop of HEPES-buffered MEM on a 22 mm2 glass coverslip (No. 1 thickness). A gas-permeable plastic gasket was placed around the medium droplet and was covered with a glass coverslip to prevent evaporation. Oocytes were evaluated on a Zeiss IM-35 microscope, and differential interference contrast (DIC) and corresponding chromatin fluorescence images were photographed using a 63 x Zeiss Plan-Neofluar objective. Oocyte Collection and Culture Oocytes were collected by follicular puncture and immediately grouped into two classes based on cumulus integrity: oocytes in compact cumulus masses and cumulus-free oocytes 2 7 0 pm in diameter. Cumulus complexes/oocytes were washed through three drops of culture medium (see below) and either pipetted into culture medium or fixed immediately. The time between ovary removal and grouping of oocytes did not exceed 10 min. Cumulus-oocyte complexes or cumulus-free oocytes were cultured in Falcon 3037 organ culture dishes in 1 ml of phenol red-free Eagle's MEM supplemented with Earle's salts, 2 mM glutamine, 0.23 mM pyruvate, 100
375
Uiml penicillin, 100 pgiml streptomycin, and 0.3% Pentex BSA. Oocytes were cultured by cumulus class for specified periods of time in a humidified atmosphere of 5%C 0 2 at 37"C, with approximately 50 oocytesidish.
Oocyte Fixation Oocytes were fixed either immediately upon isolation or at designated time points (30, 60, 90, or 120 min) after in vitro culture. Stripped oocytes were simultaneously fixed and extracted in a microtubule-stabilizing buffer (0.1 M PIPES, 5 mM MgCl,, 2.5 mM EGTA, 0.01% aprotinin, 1mM dithiothreitol, 50% deuterium oxide, 1 p M taxol, 0.1% Triton X-100, and 3% formalin) for 20 min in a shaking water bath (37°C). Washed (thrice) oocytes were stored at 4°C in PBSazide until fluorescence staining. Fluorescence Staining All antibodies were prepared in PBS containing 0.02% sodium azide and 0.1%BSA, and all incubations were carried out in a shaking water bath (37°C). Oocytes were sequentially incubated in 1)YOL 34 antitubulin antibody, 1:50, 60 min (obtained from Dr. John Kilmartin, MRC, Cambridge, England); 2) fluorescein goat antimouse antibody, 1:50, 45 min (Tago Immunologicals, Burlingame, CA); and 3) Hoechst 33258,lO pg/ml, 5 min. Oocytes were mounted in 50% PBS-sodium azide/50% glycerol and gently compressed between a coverslip and microscope slide. Coverslips were sealed with clear nail polish, and microscope slides were stored at 4°C until analysis. Microscopic Analysis Stained oocytes and ovarian cryosections were viewed through a Zeiss IM-35 microscope using 25 x , OX, or 6 3 Plan-Neofluar ~ or 6 3 x Plan-Apo Zeiss objectives under epifluorescence illumination with a 50 W mercury arc lamp. Zeiss filter sets selective for fluorescein (487709) and Hoechst (487702) were used. DIC and fluorescence images were recorded on Tri-X film using exposures of 2-10 sec. The film was processed using full-strength Acufine developer. RESULTS Classification of Chromatin Staining Patterns in GV of Freshly Isolated Oocytes Oocytes isolated by follicular puncture from 2125-day-old mice primed with PMSG (n = 550) either were fixed immediately at the time of isolation and stained with Hoechst 33258 or were stained vitally with Hoechst 33258 and examined by fluorescence and DIC microscopy. These oocytes exhibited one of four discrete patterns of chromatin organization in both vitally stained (Fig. lB,E,H,K) and fixed (Fig. lC,F,I,L) oocytes. By DIC optics, however, intact GV with welldefined nucleoli were apparent in all freshly isolated oocytes examined (Fig. lA,D,G,J). The four fluorescence patterns of chromatin organization differ with
376
B.A. MATTSON AND D.F. ALBERTINI
Fig. 1. Differential interference contrast (DIC) and chromatin patterns in Hoechst 33258 stained GV-stage oocytes obtained from PMSstimulated 21-25-day-old prepubertal mouse ovaries. Left and middle panels show DIC (A,D,G,J) and correlative Hoechst patterns in
freshly isolated, living oocytes indicating stage I (B), I1 (E), I11 (H), and IV (K) GV. The right panels (C, F, I, L) illustrate the germinal vesicles of these stages a t higher magnification in fixed oocytes. Bars = 10 pm.
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
377
respect to the association of chromatin with the nucleolus. Although we recognize oogenesis to be a continuum, for simplicity we refer to these patterns of staining as stages I-IV GV based on the following criteria. Stage I GV contain six to ten chromatin foci and a homogeneous pattern of staining throughout the nucleoplasm that is evident under fluorescence optics (Fig. lB,C). Although a well-defined nucleolus is apparent under DIC optics, correlative fluorescence images do not conspicuously demarcate the boundaries of the nucleolus. In stage I1 GV, two or three chromatin foci are associated with the nucleolar periphery (Fig. lE,F). The most distinguishing feature of stage I11 GV is the appearance of a partial rim of chromatin staining at the nucleolar surface (Fig. 1HJ. Stage IV GV exhibit a complete layer of chromatin staining, which envelops the nucleolus (Fig. lK,L). In addition, stages I11 and IV GV exhibit a more reticular or fibrillar chromatin pattern of staining throughout the nucleoplasm. The encapsulation of the oocyte nucleolus with chromatin during the course of oogenesis has been shown in a number of mammals to occur a t various stages of follicular development (Motlik and Fulka, 1986). We therefore evaluated the organization of GV chromatin in oocytes contained within nonantral and antral follicles.
Fig. 2. Cryosections of formaldehyde-fixed, sucrose-impregnated ovaries from prepubertal mice stained with Hoechst 33258. A Cluster of primary follicles (arrowheads) surrounded by a single layer of granulosa cells; stage I GV are apparent. B: stage I1 GV resident within a preantral follicle of two to three granulosa cell layers. C: Stage IV GV from an antral follicle (a denotes antral cavity) with a surrounding compact layer of granulosa cells. Bar = 10 Km.
Classification of Oocyte Chromatin Patterns in Ovarian Sections The distribution of GV stages within the developing follicles of intact ovaries was studied by staining fixed cryosections with Hoeschst 33258 (Fig. 2). Analysis of ovarian sections from PMSG-primed mice (n = 6) revealed that the transitions from stages I-IV GV are correlated with the development of antral follicles. Unilaminar or multilaminar preantral follicles contained oocytes with stage I or I1 GV (Fig. 2A,B, Table 1).Early antral follicles were found to contain predominantly stage I11 GV (62%), although both stage I1 (16%)and stage IV (22%) GV were also observed in this follicle class (Fig. 2C, Table 1).In follicles with welldeveloped antral cavities, only stage IV GV were observed (Table 1). The data summarized in Table 1 indicate that oocytes contained within preantral follicles undergo a transition from stage I to stage I11 GV and that the transition from stage I11 to stage IV occurs a t or around the time of antrum formation. The same transition was Observed in ovarian cryosections Of adult, cycling females (unpublished data). We next asked whether oocytes liberated from follicles of PMSG-primed animals progress sequentially through the stages in GV morphology, and if so whether these oocytes initiate meiotic maturation in culture.
378
B.A. MATTSON AND D.F. ALBERTINI TABLE 1. In Situ Analysis of GV Stages in Ovarian Cryosections*
Follicle type Preantral (one layer; type 3a-3b) Preantral (two to five layers; type 4-5b) Early antral (type 6) Mid- to late antral (type 7-8)
n 58 97 45 7
I 84 6 0 0
Percent GV stage (%) I1 I11 16 0 85 9 16 62 0 0
IV 0 0 22 100
*Ovarian cryosections of six PMSG-primed 25-day-old mice were evaluated after Hoechst 33258 staining to classify GV chromatin stages as described in the text. Follicles were classified based on the criteria established bv Pedersen and Peters (1968) as indicated parenthetically in column 1.
Time Course of GV Stage Transitions During - the Resumption ofMeiosis In Vitro To determine whether oocytes progress sequentially through GV stages I through IV prior to the resumption of meiosis, both cumulus-enclosed and cumulusfree oocytes were cultured under conditions that support spontaneous meiotic maturation (Schroeder and Eppig, 1984). Oocytes measuring 2 7 0 Frn in diameter were fixed a t the time of isolation and a t 30 min intervals of culture for up to 2 h r and were analyzed after Hoechst 33258 staining. At time zero, cumulusenclosed oocytes were found to contain stage IV GV (96%, 500 oocytes from ten separate experiments), whereas oocytes lacking cumulus cells were composed of a mixture of GV stages, with stage I1 (31%)and stage IV (41%) being the major types observed (Fig. 3, top). Figure 3 shows that, over a 2 h r culture period, both oocyte classes progress from stage I to stage IV GV through to prometaphase of meiosis-1. By 60 min of culture, most cumulus-enclosed oocytes initiate GV breakdown (GVBD) a s evidenced by nucleolar dissolution and chromatin condensation (diakinesis; see Fig. 51, and by 90 min the majority of these oocytes have reached prometaphase. A similar but somewhat delayed pattern of meiotic progression is observed for cumulus-free oocytes between 60 and 90 min of culture which by 2 h r have similarly progressed to prometaphase. In cumulus-free oocytes cultures, it was consistently noted that approximately 20% of the oocytes remained a t stage I1 indicating that this subpopulation was meiotically incompetent a t least after a 2 h r culture period as shown in Figure 3. Prolonged culture for up to 18 h r resulted in the completion of meiotic maturation to metaphase-2 in both cumulus-enclosed oocytes (91%) and cumulus-free (78%) demonstrating that the oocytes were fully capable of maturing in vitro (unpublished observations). These observations show that GV chromatin is reorganized during the preantral to antral follicle transition in vivo and, further, that the sequential alterations in chromatin organization occur in vitro prior to overt (GVBD) and the spontaneous resumption of meiosis. To ascertain whether the changes in GV morphol-
ogy coincided with modifications in the organization of cytoplasmic microtubules, we next examined both cumulus-enclosed and cumulus-free oocytes a t the time of isolation and at 30 min intervals of culture for up to 2 h r using antitubulin antibodies and indirect immunofluorescence microscopy.
Patterns of Microtubule Reorganization in Association With GV Stage Transitions Cumulus-enclosed or cumulus-free oocytes were fixed within 10 min after follicular puncture and isolation under conditions that permit the simultaneous visualization of chromatin (Hoechst 33258) and microtubules (antitubulin staining). This protocol allows for the direct correlation of GV chromatin organization with ooplasmic microtubule disposition. Distinct microtubule arrays were found in each GV stage. Oocytes with stage I GV contain a n extensive cytoplasmic microtubule complex that is uniformly distributed throughout the ooplasm (Fig. 4A,B). In oocytes exhibiting stage I1 GV (Fig. 4C,D), cytoplasmic microtubules are most evident a t the margins of the GV and extend into the subcortical cytoplasm as revealed by throughfocus analysis of whole-mount preparations. The cytoplasmic microtubules in oocytes containing stage I11 GV appear to be diminished in both the length of individual microtubules and apparent number and are confined to multiple antitubulin stained foci located in a perinuclear position (Fig. 4E,F). The trend in overall reduction and localization of microtubules to a perinuclear position is clearly evident in oocytes harboring stage IV GV. These oocytes contain several well-defined antitubulin-stained foci that are intimately associated with the nucleolus, which has assumed a n eccentric position a t the edge of the GV (Fig. 4 G J ) . The transitions in microtubule organization described above seem to occur during the transition from stage I to stage IV GV because they are evident in a stagespecific fashion in freshly isolated oocytes and in cumulus-free oocytes cultured for 30, 60, or 90 min. The organization of chromatin and microtubules during the resumption of meiosis in vitro was studied next in cul-
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
I-
100
0
0 0 0
I
tured oocytes, both cumulus-enclosed and cumulusfree, as they progressed to prometaphase of meiosis-1. During the first 30-60 min of culture, both cumulusenclosed and cumulus-free oocytes progress to diakinesis, which is characterized by the appearance of condensing thread-like chromatin (Fig. 5A). Very few microtubules are seen in association with the chromatin in oocytes a t this stage where the chromatin remains confined to the circular borders of the GV (Fig. 5B). This is followed by a period during which nucleolar dissolution and continued chromatin condensation occur and when the appearance of a single microtubule organizing center (MTOC) is evident (Fig. 5C,D). By prometaphase of meiosis-1, many microtubules are found associated with the chromosomes as the spindle assembles, and frequently a single MTOC is observed at some distance from the spindle (Fig. 5E,F). Thus changes in oocyte microtubule organization are correlated with both the alterations in chromatin observed during meiotic prophase (stage I-IV) and the nuclear events associated with GVBD and meiotic maturation in vitro. The observation that limited microtubule assembly and restricted MTOC localization is maintained from stage IV up to prometaphase of meiosis-1 indicates that these transitions in microtubule stability and location are initiated in vivo in relation to the stage of follicular development.
60 minutes
801
40
I-
Z
w
2o
0 E
379
DISCUSSION
o
Discrete, sequentially expressed patterns of chromatin organization in the GV of living and fixed mouse oocytes have been identified in these studies by fluorescence microscopy. These transitions occur prior to and during the preantral to antral stages of follicular development. These studies have also shown that major rearrangements in the organization of cytoplasmic microtubules can be correlated with nuclear modifications.
Nuclear Transitions
I
It
Ill
IV
DKe
DKI
PM
GV STAGE Fig. 3. Bar graphs showing progressive changes in nuclear stages during the resumption of meiosis in vitro from 0 to 120 min following oocyte retrieval. Hatched bars indicate progression of cumulus-free oocytes, and solid bars refer to cumulus-enclosed oocytes. The data represent total percentages of oocytes in each GV or later stage (see text classification) taken from ten to twelve separate experiments for each time point. DKe, early diakinesis; DKl, late diakinesis; PM, prometaphase.
With respect to changes in GV morphology during oogenesis, i t appears that nucleolar-chromatin interactions in mammalian oocytes are temporally related to a reduction in transcriptional activity (Moore et al., 1974; Bachvarova, 1985; Motlik and Fulka, 1986; Parfenov et al., 1989). With the resolution afforded by fluorescence microscopy in these studies, we have been able to define a series of nuclear transitions that are correlated with specific stages of ovarian follicle development in the mouse. Based on these observations, four sequential stages in GV morphology have been defined, which we refer to as stage I to stage IV oocytes. Initially, chromatin foci appear at the nucleolar periphery (stages 1-11), and these foci eventually coalesce (stage 111) to envelop the nucleolus, giving rise to a stage IV oocyte. We consider the stage IV oocyte to be equivalent to “nucleolar compaction” (Motlik and Fulka, 1986) and to the “karyosphere” as defined by Parfenov
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
381
synthetic estrogens can modify microtubule organization lend further support to the notion that the observed changes in oocyte microtubules may be influenced directly by follicular steroids (Brinkley et al., 1985; Szego e t al., 1988). Coincident with the reduction in cytoplasmic microtubules is the emergence of perinuclear microtubule organizing centers that may signal the entry into the M phase of the meiotic cell cycle after a prolonged interphase-like state during which oocyte growth and remodeling occur. This idea receives support from the observation that modifications in the regulation of microtubule assembly are known to occur in somatic cells at the G,IM border to support spindle assembly during prophase (Kirschner and Mitchison, 1986). Centrosome Microtubule Organization duplication and separation are important steps in the Using GV staging criteria, discrete patterns of mi- formation of the mitotic spindle apparatus. Recently, crotubule organization were observed during the stage Rime et al. (1988) reported the appearance of perinuI to stage IV GV transitions. The results of correlative clear MTOC in sections of mouse oocytes that stained DNA and antitubulin staining patterns indicate that with both antitubulin antibodies and human autoimthere is a progressive reduction in immunodetectable mune serum containing anti-centrosomal antibodies, tubulin polymer that becomes topographically re- indicating that perinuclear foci are centrosomal strucstricted in this location a s oocytes progress from stage tures. Tubulin-containing foci, which correspond to the I to stage IV. A possible explanation for the microtu- structures observed by Rime et al. (19881, were consisbule modifications observed is suggested with respect tently observed in stage IV GV in the present study. to functional alterations during oogenesis. The growth Moreover, analysis of whole-mount preparations (Fig. phase of oogenesis is a biosynthetically active period 4)reveals the presence of little tubulin polymer, a t the characterized by increased rates of protein and RNA resolution level afforded by fluorescence microscopy, synthesis, endocytosis, formation of cell-surface mi- except at these perinuclear foci. Although our results crovilli and cortical granules, and elaboration of the do not address the question of how tubulin assembly zona pellucida (Anderson, 1972; Anderson et al., 1977; properties are modified during the course of meiotic Moor et al., 1983; Schultz, 1986; Thibault e t al., 1987). prophase, the data do indicate that the stage I1 to stage We suggest that these activities, which involve mem- IV transitions are characterized by the appearance of brane reorganization during the growth phase of oo- microtubule foci that may be required for spindle assembly during later stages of meiotic progression (Cagenesis, require a n intact microtubular cytoskeleton A gradual loss of microtubules was correlated with larco et al., 1972; Wassarman and Fujiwara, 1978; development of antral follicles. MTOC were apparent Longo and Chen, 1984). The observation that a popuin association with the GV of stage IV oocytes. It is lation of isolated stage I1 GV oocytes was unable to interesting to note that follicular fluid contains high resume meiosis in vitro (Fig. 3) suggests that the acconcentrations of steroid hormones and that the loss of quisition and the expression of meiotic competence in oocyte microtubules coincides temporally with the on- culture require the transition to stage I11 or IV GV. set of follicular estrogen biosynthesis a t the time of Further studies on the distribution and function of cenantrum formation (Richards et al., 1987). In this con- trosomes using specific immunological probes may help text, the recent demonstrations that both natural and clarify both the identity and role of MTOCs and their altered organization in the mouse oocyte prior to GVBD. In summary, we have described the coordinated Fig. 4. Correlative Hoechst 33258 and antitubulin staining pat- oocyte nuclear and cytoskeletal modifications that octerns for GV-stage oocytes fixed within 10 min of isolation. A: Stage cur at specific stages of follicle development during the I GV with multiple foci of heterochromatin (arrowheads).Correspond- course of oogenesis in the mouse. Whereas the stage I1 ing antitubulin (B) shows a n extensive cytoplasmic array of microtu- to stage IV transitions observed can occur in culture bules. C: Stage I1 GV with nucleolar-associated heterochromatin foci (arrowheads). Corresponding antitubulin staining (D) shows micro- during meiotic progression, our data suggest in additubules surrounding the GV and extending throughout the cytoplasm. tion that these alterations are closely coupled to speIn stage 111 GVs (E,F), partial rim staining at the nucleolar periphery cific stages of follicular development in vivo. Although is noted and microtubules are restricted to numerous foci near the care was taken to exclude oocytes that may have been nucleus. Stage IV GV are represented in G-J, and show top (G,H)and derived from atretic follicles, this possibility cannot be side (I,J)views. G and I illustrate fully rimmed nucleoli, with which several antitubulin stained foci are associated (H,J).Bar = 10 km (all ruled out completely. However, in view of the complex and changing hormonal milieu the oocyte is subjected micrographs are printed at the same magnification).
and colleagues (1989). Oocytes progressing from stage I to stage I11 GV represent developmental precursors of stage IV oocytes based on two lines of evidence. First, analysis of cryosections of intact mouse ovaries from both PMSG-primed and unprimed animals (data not shown) clearly shows that stage IV GV are found solely in antral follicles. Second, fully grown, cumulus-free oocytes obtained from primed animals steadily progress to stage IV prior to the resumption of meiosis in culture (Fig. 3).It should be noted that modifications in nucleolar structure, which appear to be a general characteristic of mammalian oogenesis, occur at different stages of follicle development in different species (Motlik and Fulka, 1986; Parfenov et al., 1989).
382
B.A. MATTSON AND D.F. ALBERTINI
Fig. 5. Correlative Hoechst 33258 and antitubulin staining patterns for oocytes that have resumed meiosis after a 60-90 min period of culture. A Chromatin staining pattern of an oocyte in early diakinesis with two astral microtubule arrays associated with condensing chromatin (B). C : Chromatin staining pattern of an oocyte in late diakinesis in which the antitubulin staining pattern illustrates a sin-
gle microtubule organizing center (MTOC) displaced from the condensing chromatin (D). An oocyte at prometaphase-1 is shown with condensed chromosomes (E) associated with microtubules of the forming meiotic spindle (F). A single MTOC is spatially dissociated from the spindle. Bar = 10 km (all micrographs are printed a t the same magnification).
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS to during folliculogenesis, it seems likely that the transitions observed are aspects of mammalian oogenesis important for the expression of viable and meiotically competent oocytes.
ACKNOWLEDGMENTS We thank Dr. John Kilmartin for his generous supplies of YOL 34 antitubulin antibodies, Dr. Matthew Suffness of the National Cancer Institute for providing the taxol, and Dr. Everett Anderson for his critical reading of the manuscript. This work was supported by NIH grant HD 20068 (D.F.A.) and a postdoctoral fellowship from the Pharmaceutical Manufacturers Association Foundation (B.A.M.).
REFERENCES Anderson E (1972): The localization of acid phosphatase and the uptake of horseradish peroxidase in the oocytes and the follicle cells of mammals. In J D Biggers, AW Schuetz (eds): “Oogenesis,” Baltimore: University Park Press, pp 87-117. Anderson E, Albertini DF, Wilkinson RF (1977): Cytological differentiation of the female gamete. In BR Brinkley, KR Porter (eds): “International Cell Biology,” New York: Rockefeller University Press, pp 561-568. Bachvarova R (1985): Gene expression during oogenesis and oocyte development in mammals. In LW Browder (ed): “Developmental Biology, A Comprehensive Synthesis, Oogenesis.” New York: Plenum Press, pp 453-524. Brinkley BR, Tousson A, Valdivia MM (1985): The kinetochore of mammalian chromosomes: Structure and function in normal mitosis and aneuploidy. In V Dellarco, PE Voytek, A Hollaender (eds): “Aneuploidy: Etiology and Mechanisms.” New York: Plenum Press, pp 243-256. Calarco PG, Donahue RP, Szollosi D (1972): Germinal vesicle breakdown in the mouse oocyte. J Cell Sci 10:369-385. Chouinard LA (1975): A light- and electron microcope study of the oocyte nucleus during development of the antral follicle in the prepubertal mouse. J Cell Sci 17589-615. Kirschner MW. Mitchison T (1986): Beyond self-assembly: from microtubules to’morphogenesis. Cell 45:329-342.
383
Longo FJ, Chen DY (1984):Development of cortical polarity in mouse eggs: Involvement of the meiotic apparatus. Dev Biol 107:382-394. Moor RM, Crosby IM, Osborn J C (1983): Growth and maturation of mammalian oocytes. In PG Crosignani, BL Rubin (eds): “In Vitro Fertilization and Embryo Transfer.” London: Academic Press, Inc, pp 39-63. Moore GPM, Lintern-Moore S, Peters H, Faber M (1974): RNA synthesis in the mouse oocyte. J Cell Biol 60:416-422. Motlik J , Fulka J (1986): Factors affecting meiotic competence in pig oocytes. Theriogenology 25:87-96. Parfenov V, Potchukalina G, Dudina L, Kostyuchek D, Gruzova M (19891: Human antral follicles: Oocyte nuclear and the karyosphere formation (electron microscopic and autoradiographic data). Gamete Res 22:219-231. Pedersen T, Peters H (1968): Proposal for a classification of oocytes and follicles in the mouse ovary. J Reprod Fertil 17:555-557. Richards JS, Jahnsen T, Hediu L, Lifia J, Ratoosh SL, Durica JM, Goldring NB (1987): Ovarian follicular development: From physiology to molecular biology. Rec Prog Hormone Res 43:231-276. Rime H, Jessus C, Ozon R (1988): Estramustine phosphate inhibits germinal vesicle breakdown and induces depolymerization of microtubules in mouse oocytes. Reprod Nutr Dev 28:319-334. Schroeder AC, Eppig JJ (1984):The developmental capacity of mouse oocytes that matured spontaneously in vitro is normal. Dev Biol 102:493-497. Schultz RM (1986): Molecular aspects of mammalian oocyte growth and maturation. In J Rossant, RA Pedersen (eds): “Experimental Approaches to Mammalian Embryonic Development.” Cambridge: Cambridge University Press, pp 195-237. Szego CM, Sjostrand BM, Seeler BJ, Baumer JM, Sjostrand FS (1988): Microtubule and plasmalemmal reorganization: acute response to estrogen. Am J Physiol 254LEndo-crinol Metab 17l:E775-E785. Thibault C, Szollosi D, Gerard M (1987): Mammalian oocyte maturation. Reprod Nutr Dev 27:865-896. Tsafriri A (1985):The control of meiotic maturation in mammals. Biol Fertil 1:221-252. Wassarman PM, Fujiwara K (1978): Immunofluorescent anti-tubulin staining of spindles during meiotic maturation of mouse oocytes in vitro. J Cell Sci 29:171-188. Wassarman PM, Schultz RM, Letorneau GE, Lemarca MI, Josefowicz WJ, Bleil J D (1979): Meiotic maturation of mouse oocytes in vitro. In CP Channing, JM Marsh, WA Sadler (eds1: “Ovarian Follicular and Corpus Luteum Function.” New York: Plenum Press, pp .. 251268.
Oogenesis: Chromatin and Microtubule Dynamics During Meiotic Prophase BRITTA A. MATTSON AND DAVID F. ALBERTINI Department of Anatomy and Cellular Biology, Tufts University Health Sciences Center, Boston, Massachusetts ABSTRACT Changes in the organization of germinal vesicle chromatin in mouse oocytes have been analyzed by fluorescence microscopy with respect to progressive stages of follicular development and the disposition of oocyte cytoplasmic microtubules. Four discrete patterns of chromatin organization exist in germinal vesicle (GV)stage oocytes isolated from the ovaries of 2125-day-old gonadotropin-primed mice. Analysis of ovarian cryosections stained with the DNA-binding fluorochrome Hoechst 33258 indicates that sequential changes in GV chromatin occur duing folliculogenesis that result in the formation of a continuous perinucleolar chromatin sheath at the time of antrum formation. Specific alterations in the cytoplasmic microtubule complex of GV-stage oocytes were observed that correlate with chromatin patterns. The extensive cytoplasmic microtubule complex seen in oocytes of preantral follicles initially localizes to perinuclear areas of the ooplasm. This is followed by a progressive reduction in cytoplasmic microtubules and the appearance of prominent microtubule-organizing centers at the nuclear periphery. Coordinated nuclear and microtubular alterations also occur under in vitro conditions prior to progression of meiosis to prometaphase-1. The results are discussed with respect to the ongoing differentiation of the oocyte nucleus and the microtubule cytoskeleton during folliculogenesis in preparation for the resumption of meiosis. Key Words: Ovarian follicle, Oocyte, Germinal vesicle, Chromosomes, Microtubules
INTRODUCTION The events that immediately precede the resumption of meiosis in mammalian oocytes are a subject of considerable interest with respect to the functional interrelationship of oogenesis and folliculogenesis. Although much is known about the growth and differentiation of the oocyte during follicular development as i t relates to meiotic competence, metabolism, and structural organization (Anderson et al., 1977; Wassarman et al., 1979; Moor et al., 1983; Tsafriri, 1985; Schultz, 1986; Thibault et al., 19871, little is
0 1990 WILEY-LISS, INC.
known about the nuclear and cytoplasmic transitions that are prerequisites in the preparation of the prophase-arrested oocyte for meiotic maturation in the preovulatory follicle. Correlative changes in nuclear structure and transcriptional activity have been shown to occur in mammalian oocytes a t specific stages of follicular development (Motlik and Fulka, 1986). Chouinard (1975) showed by both light and electron microscopy that heterochromatin and nucleoli undergo extensive reorganization in the germinal vesicles (GV) of oocytes in growing, preantral follicles of prepubertal mice. Chouinard (1975) noted little change in the nuclear morphology during subsequent growth of antral follicles, a time during which the follicle is profoundly influenced by gonadotropins. Moore et al. (1974) have shown that RNA synthesis and RNA polymerase activity in mouse oocytes are maximal prior to antrum formation and remain low in oocytes during subsequent follicular growth. In the human oocyte, however, heterochromatin and nucleolar rearrangements as well as a decrease in 3H-uridine incorporation have been noted to occur following antrum formation (Parfenov et al., 1989). Although similar nuclear changes have been noted in other mammalian species during oogenesis a t specific stages of follicular development (see Parfenov et al., 1989), it is not known whether ooplasmic structures are also modified in their organization prior to the resumption of meiosis. Moreover, if changes in cytoplasmic organization do occur, the extent to which these alterations can also be correlated with changes in nuclear structure and function andlor formation and development of antral follicles is not known. The objectives of this study were twofold: first, to determine if discrete patterns of chromatin organization can be identified during meiotic prophase in vivo and in vitro and whether these changes are sequentially expressed; second, to determine whether changes in chromatin organization can be correlated with remodeling of the
~
Received July 27, 1989; accepted November 14, 1989. Address reprint requests to Britta A. Mattson, Department of Anatomy and Cellular Biology, Tufts University Health Sciences Center, 136 Harrison Avenue, Boston, MA 02111.
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS microtubular component of the cytoskeleton prior to and following the resumption of meiosis.
MATERIALS AND METHODS Animals Virgin, female, 21-25-day-old CD-1 mice (Charles River Laboratories, Wilmington, MA) housed in a controlled environment (14 h r light/lO hr dark) were used for all experiments. Follicular development was stimulated by i.p. injection of 7.5 I.U. pregnant mare's serum gonadotropin (PMSG; Diosynth, Inc., Chicago, IL). The animals were killed by cervical dislocation 44-48 h r later, and the ovaries were excised and placed in a petri dish containing Eagle's MEM (GIBCO, Grand Island, NY) supplemented with 25 mM HEPES buffer, Hank's salts, 100 Uiml penicillin, 100 pg/ml streptomycin, and 0.3% Pentex BSA (Miles Laboratories, Kankanee, IL) a t 25°C. Ovarian Cryosections Ovaries were washed, fixed in 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for 4 hr, and infiltrated for 48 h r with 2.3 M sucrose and 0.1% paraformaldehyde in phosphate buffer. Five to eight micrometer sections were mounted on gelatin-coated slides. Ovarian cryosections were incubated in 10 pgiml Hoechst 33258 (Polysciences, Inc., Warrington, PA) for 30 min at 25"C, washed, and mounted. Microscope slides were stored a t 4°C until analysis. Vital Staining Excised ovaries were immediately transferred to collection medium containing 10 pg/ml Hoechst 33258. Oocytes were collected by follicular puncture, mechanically stripped of cumulus cells, and pipetted into a 50 pl drop of HEPES-buffered MEM on a 22 mm2 glass coverslip (No. 1 thickness). A gas-permeable plastic gasket was placed around the medium droplet and was covered with a glass coverslip to prevent evaporation. Oocytes were evaluated on a Zeiss IM-35 microscope, and differential interference contrast (DIC) and corresponding chromatin fluorescence images were photographed using a 63 x Zeiss Plan-Neofluar objective. Oocyte Collection and Culture Oocytes were collected by follicular puncture and immediately grouped into two classes based on cumulus integrity: oocytes in compact cumulus masses and cumulus-free oocytes 2 7 0 pm in diameter. Cumulus complexes/oocytes were washed through three drops of culture medium (see below) and either pipetted into culture medium or fixed immediately. The time between ovary removal and grouping of oocytes did not exceed 10 min. Cumulus-oocyte complexes or cumulus-free oocytes were cultured in Falcon 3037 organ culture dishes in 1 ml of phenol red-free Eagle's MEM supplemented with Earle's salts, 2 mM glutamine, 0.23 mM pyruvate, 100
375
Uiml penicillin, 100 pgiml streptomycin, and 0.3% Pentex BSA. Oocytes were cultured by cumulus class for specified periods of time in a humidified atmosphere of 5%C 0 2 at 37"C, with approximately 50 oocytesidish.
Oocyte Fixation Oocytes were fixed either immediately upon isolation or at designated time points (30, 60, 90, or 120 min) after in vitro culture. Stripped oocytes were simultaneously fixed and extracted in a microtubule-stabilizing buffer (0.1 M PIPES, 5 mM MgCl,, 2.5 mM EGTA, 0.01% aprotinin, 1mM dithiothreitol, 50% deuterium oxide, 1 p M taxol, 0.1% Triton X-100, and 3% formalin) for 20 min in a shaking water bath (37°C). Washed (thrice) oocytes were stored at 4°C in PBSazide until fluorescence staining. Fluorescence Staining All antibodies were prepared in PBS containing 0.02% sodium azide and 0.1%BSA, and all incubations were carried out in a shaking water bath (37°C). Oocytes were sequentially incubated in 1)YOL 34 antitubulin antibody, 1:50, 60 min (obtained from Dr. John Kilmartin, MRC, Cambridge, England); 2) fluorescein goat antimouse antibody, 1:50, 45 min (Tago Immunologicals, Burlingame, CA); and 3) Hoechst 33258,lO pg/ml, 5 min. Oocytes were mounted in 50% PBS-sodium azide/50% glycerol and gently compressed between a coverslip and microscope slide. Coverslips were sealed with clear nail polish, and microscope slides were stored at 4°C until analysis. Microscopic Analysis Stained oocytes and ovarian cryosections were viewed through a Zeiss IM-35 microscope using 25 x , OX, or 6 3 Plan-Neofluar ~ or 6 3 x Plan-Apo Zeiss objectives under epifluorescence illumination with a 50 W mercury arc lamp. Zeiss filter sets selective for fluorescein (487709) and Hoechst (487702) were used. DIC and fluorescence images were recorded on Tri-X film using exposures of 2-10 sec. The film was processed using full-strength Acufine developer. RESULTS Classification of Chromatin Staining Patterns in GV of Freshly Isolated Oocytes Oocytes isolated by follicular puncture from 2125-day-old mice primed with PMSG (n = 550) either were fixed immediately at the time of isolation and stained with Hoechst 33258 or were stained vitally with Hoechst 33258 and examined by fluorescence and DIC microscopy. These oocytes exhibited one of four discrete patterns of chromatin organization in both vitally stained (Fig. lB,E,H,K) and fixed (Fig. lC,F,I,L) oocytes. By DIC optics, however, intact GV with welldefined nucleoli were apparent in all freshly isolated oocytes examined (Fig. lA,D,G,J). The four fluorescence patterns of chromatin organization differ with
376
B.A. MATTSON AND D.F. ALBERTINI
Fig. 1. Differential interference contrast (DIC) and chromatin patterns in Hoechst 33258 stained GV-stage oocytes obtained from PMSstimulated 21-25-day-old prepubertal mouse ovaries. Left and middle panels show DIC (A,D,G,J) and correlative Hoechst patterns in
freshly isolated, living oocytes indicating stage I (B), I1 (E), I11 (H), and IV (K) GV. The right panels (C, F, I, L) illustrate the germinal vesicles of these stages a t higher magnification in fixed oocytes. Bars = 10 pm.
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
377
respect to the association of chromatin with the nucleolus. Although we recognize oogenesis to be a continuum, for simplicity we refer to these patterns of staining as stages I-IV GV based on the following criteria. Stage I GV contain six to ten chromatin foci and a homogeneous pattern of staining throughout the nucleoplasm that is evident under fluorescence optics (Fig. lB,C). Although a well-defined nucleolus is apparent under DIC optics, correlative fluorescence images do not conspicuously demarcate the boundaries of the nucleolus. In stage I1 GV, two or three chromatin foci are associated with the nucleolar periphery (Fig. lE,F). The most distinguishing feature of stage I11 GV is the appearance of a partial rim of chromatin staining at the nucleolar surface (Fig. 1HJ. Stage IV GV exhibit a complete layer of chromatin staining, which envelops the nucleolus (Fig. lK,L). In addition, stages I11 and IV GV exhibit a more reticular or fibrillar chromatin pattern of staining throughout the nucleoplasm. The encapsulation of the oocyte nucleolus with chromatin during the course of oogenesis has been shown in a number of mammals to occur a t various stages of follicular development (Motlik and Fulka, 1986). We therefore evaluated the organization of GV chromatin in oocytes contained within nonantral and antral follicles.
Fig. 2. Cryosections of formaldehyde-fixed, sucrose-impregnated ovaries from prepubertal mice stained with Hoechst 33258. A Cluster of primary follicles (arrowheads) surrounded by a single layer of granulosa cells; stage I GV are apparent. B: stage I1 GV resident within a preantral follicle of two to three granulosa cell layers. C: Stage IV GV from an antral follicle (a denotes antral cavity) with a surrounding compact layer of granulosa cells. Bar = 10 Km.
Classification of Oocyte Chromatin Patterns in Ovarian Sections The distribution of GV stages within the developing follicles of intact ovaries was studied by staining fixed cryosections with Hoeschst 33258 (Fig. 2). Analysis of ovarian sections from PMSG-primed mice (n = 6) revealed that the transitions from stages I-IV GV are correlated with the development of antral follicles. Unilaminar or multilaminar preantral follicles contained oocytes with stage I or I1 GV (Fig. 2A,B, Table 1).Early antral follicles were found to contain predominantly stage I11 GV (62%), although both stage I1 (16%)and stage IV (22%) GV were also observed in this follicle class (Fig. 2C, Table 1).In follicles with welldeveloped antral cavities, only stage IV GV were observed (Table 1). The data summarized in Table 1 indicate that oocytes contained within preantral follicles undergo a transition from stage I to stage I11 GV and that the transition from stage I11 to stage IV occurs a t or around the time of antrum formation. The same transition was Observed in ovarian cryosections Of adult, cycling females (unpublished data). We next asked whether oocytes liberated from follicles of PMSG-primed animals progress sequentially through the stages in GV morphology, and if so whether these oocytes initiate meiotic maturation in culture.
378
B.A. MATTSON AND D.F. ALBERTINI TABLE 1. In Situ Analysis of GV Stages in Ovarian Cryosections*
Follicle type Preantral (one layer; type 3a-3b) Preantral (two to five layers; type 4-5b) Early antral (type 6) Mid- to late antral (type 7-8)
n 58 97 45 7
I 84 6 0 0
Percent GV stage (%) I1 I11 16 0 85 9 16 62 0 0
IV 0 0 22 100
*Ovarian cryosections of six PMSG-primed 25-day-old mice were evaluated after Hoechst 33258 staining to classify GV chromatin stages as described in the text. Follicles were classified based on the criteria established bv Pedersen and Peters (1968) as indicated parenthetically in column 1.
Time Course of GV Stage Transitions During - the Resumption ofMeiosis In Vitro To determine whether oocytes progress sequentially through GV stages I through IV prior to the resumption of meiosis, both cumulus-enclosed and cumulusfree oocytes were cultured under conditions that support spontaneous meiotic maturation (Schroeder and Eppig, 1984). Oocytes measuring 2 7 0 Frn in diameter were fixed a t the time of isolation and a t 30 min intervals of culture for up to 2 h r and were analyzed after Hoechst 33258 staining. At time zero, cumulusenclosed oocytes were found to contain stage IV GV (96%, 500 oocytes from ten separate experiments), whereas oocytes lacking cumulus cells were composed of a mixture of GV stages, with stage I1 (31%)and stage IV (41%) being the major types observed (Fig. 3, top). Figure 3 shows that, over a 2 h r culture period, both oocyte classes progress from stage I to stage IV GV through to prometaphase of meiosis-1. By 60 min of culture, most cumulus-enclosed oocytes initiate GV breakdown (GVBD) a s evidenced by nucleolar dissolution and chromatin condensation (diakinesis; see Fig. 51, and by 90 min the majority of these oocytes have reached prometaphase. A similar but somewhat delayed pattern of meiotic progression is observed for cumulus-free oocytes between 60 and 90 min of culture which by 2 h r have similarly progressed to prometaphase. In cumulus-free oocytes cultures, it was consistently noted that approximately 20% of the oocytes remained a t stage I1 indicating that this subpopulation was meiotically incompetent a t least after a 2 h r culture period as shown in Figure 3. Prolonged culture for up to 18 h r resulted in the completion of meiotic maturation to metaphase-2 in both cumulus-enclosed oocytes (91%) and cumulus-free (78%) demonstrating that the oocytes were fully capable of maturing in vitro (unpublished observations). These observations show that GV chromatin is reorganized during the preantral to antral follicle transition in vivo and, further, that the sequential alterations in chromatin organization occur in vitro prior to overt (GVBD) and the spontaneous resumption of meiosis. To ascertain whether the changes in GV morphol-
ogy coincided with modifications in the organization of cytoplasmic microtubules, we next examined both cumulus-enclosed and cumulus-free oocytes a t the time of isolation and at 30 min intervals of culture for up to 2 h r using antitubulin antibodies and indirect immunofluorescence microscopy.
Patterns of Microtubule Reorganization in Association With GV Stage Transitions Cumulus-enclosed or cumulus-free oocytes were fixed within 10 min after follicular puncture and isolation under conditions that permit the simultaneous visualization of chromatin (Hoechst 33258) and microtubules (antitubulin staining). This protocol allows for the direct correlation of GV chromatin organization with ooplasmic microtubule disposition. Distinct microtubule arrays were found in each GV stage. Oocytes with stage I GV contain a n extensive cytoplasmic microtubule complex that is uniformly distributed throughout the ooplasm (Fig. 4A,B). In oocytes exhibiting stage I1 GV (Fig. 4C,D), cytoplasmic microtubules are most evident a t the margins of the GV and extend into the subcortical cytoplasm as revealed by throughfocus analysis of whole-mount preparations. The cytoplasmic microtubules in oocytes containing stage I11 GV appear to be diminished in both the length of individual microtubules and apparent number and are confined to multiple antitubulin stained foci located in a perinuclear position (Fig. 4E,F). The trend in overall reduction and localization of microtubules to a perinuclear position is clearly evident in oocytes harboring stage IV GV. These oocytes contain several well-defined antitubulin-stained foci that are intimately associated with the nucleolus, which has assumed a n eccentric position a t the edge of the GV (Fig. 4 G J ) . The transitions in microtubule organization described above seem to occur during the transition from stage I to stage IV GV because they are evident in a stagespecific fashion in freshly isolated oocytes and in cumulus-free oocytes cultured for 30, 60, or 90 min. The organization of chromatin and microtubules during the resumption of meiosis in vitro was studied next in cul-
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
I-
100
0
0 0 0
I
tured oocytes, both cumulus-enclosed and cumulusfree, as they progressed to prometaphase of meiosis-1. During the first 30-60 min of culture, both cumulusenclosed and cumulus-free oocytes progress to diakinesis, which is characterized by the appearance of condensing thread-like chromatin (Fig. 5A). Very few microtubules are seen in association with the chromatin in oocytes a t this stage where the chromatin remains confined to the circular borders of the GV (Fig. 5B). This is followed by a period during which nucleolar dissolution and continued chromatin condensation occur and when the appearance of a single microtubule organizing center (MTOC) is evident (Fig. 5C,D). By prometaphase of meiosis-1, many microtubules are found associated with the chromosomes as the spindle assembles, and frequently a single MTOC is observed at some distance from the spindle (Fig. 5E,F). Thus changes in oocyte microtubule organization are correlated with both the alterations in chromatin observed during meiotic prophase (stage I-IV) and the nuclear events associated with GVBD and meiotic maturation in vitro. The observation that limited microtubule assembly and restricted MTOC localization is maintained from stage IV up to prometaphase of meiosis-1 indicates that these transitions in microtubule stability and location are initiated in vivo in relation to the stage of follicular development.
60 minutes
801
40
I-
Z
w
2o
0 E
379
DISCUSSION
o
Discrete, sequentially expressed patterns of chromatin organization in the GV of living and fixed mouse oocytes have been identified in these studies by fluorescence microscopy. These transitions occur prior to and during the preantral to antral stages of follicular development. These studies have also shown that major rearrangements in the organization of cytoplasmic microtubules can be correlated with nuclear modifications.
Nuclear Transitions
I
It
Ill
IV
DKe
DKI
PM
GV STAGE Fig. 3. Bar graphs showing progressive changes in nuclear stages during the resumption of meiosis in vitro from 0 to 120 min following oocyte retrieval. Hatched bars indicate progression of cumulus-free oocytes, and solid bars refer to cumulus-enclosed oocytes. The data represent total percentages of oocytes in each GV or later stage (see text classification) taken from ten to twelve separate experiments for each time point. DKe, early diakinesis; DKl, late diakinesis; PM, prometaphase.
With respect to changes in GV morphology during oogenesis, i t appears that nucleolar-chromatin interactions in mammalian oocytes are temporally related to a reduction in transcriptional activity (Moore et al., 1974; Bachvarova, 1985; Motlik and Fulka, 1986; Parfenov et al., 1989). With the resolution afforded by fluorescence microscopy in these studies, we have been able to define a series of nuclear transitions that are correlated with specific stages of ovarian follicle development in the mouse. Based on these observations, four sequential stages in GV morphology have been defined, which we refer to as stage I to stage IV oocytes. Initially, chromatin foci appear at the nucleolar periphery (stages 1-11), and these foci eventually coalesce (stage 111) to envelop the nucleolus, giving rise to a stage IV oocyte. We consider the stage IV oocyte to be equivalent to “nucleolar compaction” (Motlik and Fulka, 1986) and to the “karyosphere” as defined by Parfenov
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS
381
synthetic estrogens can modify microtubule organization lend further support to the notion that the observed changes in oocyte microtubules may be influenced directly by follicular steroids (Brinkley et al., 1985; Szego e t al., 1988). Coincident with the reduction in cytoplasmic microtubules is the emergence of perinuclear microtubule organizing centers that may signal the entry into the M phase of the meiotic cell cycle after a prolonged interphase-like state during which oocyte growth and remodeling occur. This idea receives support from the observation that modifications in the regulation of microtubule assembly are known to occur in somatic cells at the G,IM border to support spindle assembly during prophase (Kirschner and Mitchison, 1986). Centrosome Microtubule Organization duplication and separation are important steps in the Using GV staging criteria, discrete patterns of mi- formation of the mitotic spindle apparatus. Recently, crotubule organization were observed during the stage Rime et al. (1988) reported the appearance of perinuI to stage IV GV transitions. The results of correlative clear MTOC in sections of mouse oocytes that stained DNA and antitubulin staining patterns indicate that with both antitubulin antibodies and human autoimthere is a progressive reduction in immunodetectable mune serum containing anti-centrosomal antibodies, tubulin polymer that becomes topographically re- indicating that perinuclear foci are centrosomal strucstricted in this location a s oocytes progress from stage tures. Tubulin-containing foci, which correspond to the I to stage IV. A possible explanation for the microtu- structures observed by Rime et al. (19881, were consisbule modifications observed is suggested with respect tently observed in stage IV GV in the present study. to functional alterations during oogenesis. The growth Moreover, analysis of whole-mount preparations (Fig. phase of oogenesis is a biosynthetically active period 4)reveals the presence of little tubulin polymer, a t the characterized by increased rates of protein and RNA resolution level afforded by fluorescence microscopy, synthesis, endocytosis, formation of cell-surface mi- except at these perinuclear foci. Although our results crovilli and cortical granules, and elaboration of the do not address the question of how tubulin assembly zona pellucida (Anderson, 1972; Anderson et al., 1977; properties are modified during the course of meiotic Moor et al., 1983; Schultz, 1986; Thibault e t al., 1987). prophase, the data do indicate that the stage I1 to stage We suggest that these activities, which involve mem- IV transitions are characterized by the appearance of brane reorganization during the growth phase of oo- microtubule foci that may be required for spindle assembly during later stages of meiotic progression (Cagenesis, require a n intact microtubular cytoskeleton A gradual loss of microtubules was correlated with larco et al., 1972; Wassarman and Fujiwara, 1978; development of antral follicles. MTOC were apparent Longo and Chen, 1984). The observation that a popuin association with the GV of stage IV oocytes. It is lation of isolated stage I1 GV oocytes was unable to interesting to note that follicular fluid contains high resume meiosis in vitro (Fig. 3) suggests that the acconcentrations of steroid hormones and that the loss of quisition and the expression of meiotic competence in oocyte microtubules coincides temporally with the on- culture require the transition to stage I11 or IV GV. set of follicular estrogen biosynthesis a t the time of Further studies on the distribution and function of cenantrum formation (Richards et al., 1987). In this con- trosomes using specific immunological probes may help text, the recent demonstrations that both natural and clarify both the identity and role of MTOCs and their altered organization in the mouse oocyte prior to GVBD. In summary, we have described the coordinated Fig. 4. Correlative Hoechst 33258 and antitubulin staining pat- oocyte nuclear and cytoskeletal modifications that octerns for GV-stage oocytes fixed within 10 min of isolation. A: Stage cur at specific stages of follicle development during the I GV with multiple foci of heterochromatin (arrowheads).Correspond- course of oogenesis in the mouse. Whereas the stage I1 ing antitubulin (B) shows a n extensive cytoplasmic array of microtu- to stage IV transitions observed can occur in culture bules. C: Stage I1 GV with nucleolar-associated heterochromatin foci (arrowheads). Corresponding antitubulin staining (D) shows micro- during meiotic progression, our data suggest in additubules surrounding the GV and extending throughout the cytoplasm. tion that these alterations are closely coupled to speIn stage 111 GVs (E,F), partial rim staining at the nucleolar periphery cific stages of follicular development in vivo. Although is noted and microtubules are restricted to numerous foci near the care was taken to exclude oocytes that may have been nucleus. Stage IV GV are represented in G-J, and show top (G,H)and derived from atretic follicles, this possibility cannot be side (I,J)views. G and I illustrate fully rimmed nucleoli, with which several antitubulin stained foci are associated (H,J).Bar = 10 km (all ruled out completely. However, in view of the complex and changing hormonal milieu the oocyte is subjected micrographs are printed at the same magnification).
and colleagues (1989). Oocytes progressing from stage I to stage I11 GV represent developmental precursors of stage IV oocytes based on two lines of evidence. First, analysis of cryosections of intact mouse ovaries from both PMSG-primed and unprimed animals (data not shown) clearly shows that stage IV GV are found solely in antral follicles. Second, fully grown, cumulus-free oocytes obtained from primed animals steadily progress to stage IV prior to the resumption of meiosis in culture (Fig. 3).It should be noted that modifications in nucleolar structure, which appear to be a general characteristic of mammalian oogenesis, occur at different stages of follicle development in different species (Motlik and Fulka, 1986; Parfenov et al., 1989).
382
B.A. MATTSON AND D.F. ALBERTINI
Fig. 5. Correlative Hoechst 33258 and antitubulin staining patterns for oocytes that have resumed meiosis after a 60-90 min period of culture. A Chromatin staining pattern of an oocyte in early diakinesis with two astral microtubule arrays associated with condensing chromatin (B). C : Chromatin staining pattern of an oocyte in late diakinesis in which the antitubulin staining pattern illustrates a sin-
gle microtubule organizing center (MTOC) displaced from the condensing chromatin (D). An oocyte at prometaphase-1 is shown with condensed chromosomes (E) associated with microtubules of the forming meiotic spindle (F). A single MTOC is spatially dissociated from the spindle. Bar = 10 km (all micrographs are printed a t the same magnification).
OOGENESIS: CHROMATIN/MICROTUBULE DYNAMICS to during folliculogenesis, it seems likely that the transitions observed are aspects of mammalian oogenesis important for the expression of viable and meiotically competent oocytes.
ACKNOWLEDGMENTS We thank Dr. John Kilmartin for his generous supplies of YOL 34 antitubulin antibodies, Dr. Matthew Suffness of the National Cancer Institute for providing the taxol, and Dr. Everett Anderson for his critical reading of the manuscript. This work was supported by NIH grant HD 20068 (D.F.A.) and a postdoctoral fellowship from the Pharmaceutical Manufacturers Association Foundation (B.A.M.).
REFERENCES Anderson E (1972): The localization of acid phosphatase and the uptake of horseradish peroxidase in the oocytes and the follicle cells of mammals. In J D Biggers, AW Schuetz (eds): “Oogenesis,” Baltimore: University Park Press, pp 87-117. Anderson E, Albertini DF, Wilkinson RF (1977): Cytological differentiation of the female gamete. In BR Brinkley, KR Porter (eds): “International Cell Biology,” New York: Rockefeller University Press, pp 561-568. Bachvarova R (1985): Gene expression during oogenesis and oocyte development in mammals. In LW Browder (ed): “Developmental Biology, A Comprehensive Synthesis, Oogenesis.” New York: Plenum Press, pp 453-524. Brinkley BR, Tousson A, Valdivia MM (1985): The kinetochore of mammalian chromosomes: Structure and function in normal mitosis and aneuploidy. In V Dellarco, PE Voytek, A Hollaender (eds): “Aneuploidy: Etiology and Mechanisms.” New York: Plenum Press, pp 243-256. Calarco PG, Donahue RP, Szollosi D (1972): Germinal vesicle breakdown in the mouse oocyte. J Cell Sci 10:369-385. Chouinard LA (1975): A light- and electron microcope study of the oocyte nucleus during development of the antral follicle in the prepubertal mouse. J Cell Sci 17589-615. Kirschner MW. Mitchison T (1986): Beyond self-assembly: from microtubules to’morphogenesis. Cell 45:329-342.
383
Longo FJ, Chen DY (1984):Development of cortical polarity in mouse eggs: Involvement of the meiotic apparatus. Dev Biol 107:382-394. Moor RM, Crosby IM, Osborn J C (1983): Growth and maturation of mammalian oocytes. In PG Crosignani, BL Rubin (eds): “In Vitro Fertilization and Embryo Transfer.” London: Academic Press, Inc, pp 39-63. Moore GPM, Lintern-Moore S, Peters H, Faber M (1974): RNA synthesis in the mouse oocyte. J Cell Biol 60:416-422. Motlik J , Fulka J (1986): Factors affecting meiotic competence in pig oocytes. Theriogenology 25:87-96. Parfenov V, Potchukalina G, Dudina L, Kostyuchek D, Gruzova M (19891: Human antral follicles: Oocyte nuclear and the karyosphere formation (electron microscopic and autoradiographic data). Gamete Res 22:219-231. Pedersen T, Peters H (1968): Proposal for a classification of oocytes and follicles in the mouse ovary. J Reprod Fertil 17:555-557. Richards JS, Jahnsen T, Hediu L, Lifia J, Ratoosh SL, Durica JM, Goldring NB (1987): Ovarian follicular development: From physiology to molecular biology. Rec Prog Hormone Res 43:231-276. Rime H, Jessus C, Ozon R (1988): Estramustine phosphate inhibits germinal vesicle breakdown and induces depolymerization of microtubules in mouse oocytes. Reprod Nutr Dev 28:319-334. Schroeder AC, Eppig JJ (1984):The developmental capacity of mouse oocytes that matured spontaneously in vitro is normal. Dev Biol 102:493-497. Schultz RM (1986): Molecular aspects of mammalian oocyte growth and maturation. In J Rossant, RA Pedersen (eds): “Experimental Approaches to Mammalian Embryonic Development.” Cambridge: Cambridge University Press, pp 195-237. Szego CM, Sjostrand BM, Seeler BJ, Baumer JM, Sjostrand FS (1988): Microtubule and plasmalemmal reorganization: acute response to estrogen. Am J Physiol 254LEndo-crinol Metab 17l:E775-E785. Thibault C, Szollosi D, Gerard M (1987): Mammalian oocyte maturation. Reprod Nutr Dev 27:865-896. Tsafriri A (1985):The control of meiotic maturation in mammals. Biol Fertil 1:221-252. Wassarman PM, Fujiwara K (1978): Immunofluorescent anti-tubulin staining of spindles during meiotic maturation of mouse oocytes in vitro. J Cell Sci 29:171-188. Wassarman PM, Schultz RM, Letorneau GE, Lemarca MI, Josefowicz WJ, Bleil J D (1979): Meiotic maturation of mouse oocytes in vitro. In CP Channing, JM Marsh, WA Sadler (eds1: “Ovarian Follicular and Corpus Luteum Function.” New York: Plenum Press, pp .. 251268.
