golgi dynamics during meiosis are distinct from mitosis and are coupled to endoplasmic reticulum...

Golgi dynamics during meiosis are distinct from mitosis and arecoupled to endoplasmic reticulum dynamics until fertilization

Christopher Paynea,b and Gerald Schattenb,*a Program in Molecular and Cellular Biosciences, Department of Cell and Developmental Biology,

Oregon Health and Science University, Portland, OR 97201, USAb Pittsburgh Development Center, Magee-Womens Research Institute, Departments of Obstetrics, Gynecology and Reproductive Sciences,

and Cell Biology, and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA

Received for publication 20 March 2003, revised 1 August 2003, accepted 6 August 2003

Abstract

One current theory of the Golgi apparatus views its organization as containing both a matrix fraction of structural proteins and a reservoirof cycling enzymes. During mitosis, the putative matrix protein GM130 is phosphorylated and relocalized to spindle poles. When thesecretory pathway is inhibited during interphase, GM130 redistributes to regions adjacent to vesicle export sites on the endoplasmicreticulum (ER). Strikingly, meiotic maturation and fertilization in nonrodent mammalian eggs presents a unique experimental environmentfor the Golgi apparatus, because secretion is inhibited until after fertilization, and because the centrosome is absent until introduced by thesperm. Here, we test the hypothesis that phosphorylated GM130 associates not with meiotic spindle poles, but with ER clusters in the maturebovine oocyte. At the germinal vesicle stage, phosphorylated GM130 is observed as fragments dispersed throughout the cytoplasm. Duringmeiotic maturation, GM130 reorganizes into punctate foci that associate near the ER-resident protein calreticulin and is notably absent fromthe meiotic spindle. GM130 colocalizes with Sec23, a marker for ER vesicle export sites, but not with Lens culinaris agglutinin, a markerfor cortical granules. Because disruption of vesicle transport has been shown to block meiotic maturation and embryonic cleavage in somespecies, we also test the hypothesis that fertilization and cytokinesis are inhibited with membrane trafficking disruptor brefeldin A (BFA).Despite Golgi fragmentation after BFA treatment, pronuclei form and unite, and embryos cleave and develop through the eight-cell stage.We conclude that, while the meiotic phosphorylation cycle of GM130 mirrors that of mitosis, absence of a maternal centrosome precludesGolgi association with the meiotic spindle. Fertilization introduces the sperm centrosome that can reorganize Golgi proteins, but neitherfertilization nor cytokinesis prior to compaction requires a functional Golgi apparatus.© 2003 Elsevier Inc. All rights reserved.

Keywords: Golgi apparatus; Oocyte; Meiosis; Fertilization; Phosphorylation

Introduction

Upon resumption of meiosis, mammalian oocytes inprophase I proceed through germinal vesicle breakdown(GVBD), undergo genomic reduction through the extrusion ofthe first polar body, and arrest in metaphase II (Gosden et al.,1997). Sperm penetration activates the oocyte, leading to theemission of a second polar body and the formation, migration,and apposition of the two pronuclei (Aitken, 1997). Successfulgenomic union then completes the fertilization process. These

events are highly dynamic, involving the active reorganizationof the chromatin, cytoskeleton, membrane organelles, andother structural components within the oocyte (Albertini, 1992;Perreault, 1992). Genomic reduction during meiosis is accom-panied by centrosome reduction in most mammalian species,including bovine, rhesus monkey, and human (Schatten, 1994).Rodent oocytes, in contrast, retain their centrosomes through-out meiotic maturation (Schatten et al., 1986). Bovine is anideal model species for studying oocyte maturation and fertil-ization because, as in rhesus and human oocytes, the centro-some is not present until sperm entry (Sathananthan et al.,1991; Navara et al., 1994; Schatten, 1994; Sutovsky et al.,1996).

* Corresponding author. Fax: �1-412-641-2410.E-mail address: [email protected] (G. Schatten).

R

Available online at www.sciencedirect.com

Developmental Biology 264 (2003) 50–63 www.elsevier.com/locate/ydbio

0012-1606/$ – see front matter © 2003 Elsevier Inc. All rights reserved.doi:10.1016/j.ydbio.2003.08.004

The dynamics of Golgi membranes during meiosis andfertilization in mammals are not entirely known. Mouse,rhesus, and bovine GV oocytes contain dispersed fragmentsof Golgi that vesiculate following GVBD (Calarco et al.,1972; Wassarman and Josefowicz, 1978; Hyttel et al., 1986;McGaughey et al., 1990; Assey et al., 1994; Moreno et al.,2002), yet the distribution and modification of these vesiclesfollowing polar body extrusion and insemination have notbeen fully examined and warrant further study. Brefeldin A(BFA), a fungal metabolite that inhibits protein secretion bydisrupting ER-to-Golgi vesicle transport (Lippincott-Schwartz et al., 1989), blocks meiotic maturation in mouseoocytes and prevents cell division in C. elegans zygotes andembryos (Moreno et al., 2002; Skop et al., 2001). It isunclear, however, whether Golgi disruption with BFA in-hibits the fertilization process that occurs between meiosisand cytokinesis. While primarily dependent on the cytoskel-eton to unite the pronuclei and initiate cleavage, fertilizationand cell division both involve extensive membrane or-ganelle transport (Terasaki and Jaffe, 1991; Rouviere et al.,1994; Danilchik et al., 1998; Skop et al., 2001). Thesetrafficking events include the intermingling of pronuclearenvelopes with ER and Golgi membranes along microtu-bules during fertilization (Terasaki and Jaffe, 1991; Rou-viere et al., 1994), as well as the delivery of Golgi mem-brane-derived vesicles to cleavage furrows duringcytokinesis (Danilchik et al., 1998; Skop et al., 2001). BFAhas been shown to inhibit pronuclear movement in parthe-nogenetically activated oocytes (Clayton et al., 1995), rais-ing the possibility that BFA might also inhibit pronuclearmigration in zygotes during fertilization.

Golgi structure has been theorized to contain both matrixproteins and resident enzymes (Slusarewicz et al., 1994; Na-kamura et al., 1995; Lippincott-Schwartz et al., 2000). Putativematrix proteins, such as GM130, become phosphorylated andassociate with spindle poles during mitosis, while residentenzymes are reabsorbed into the ER (Lowe et al., 1998; Jeschand Linstedt, 1998; Zaal et al., 1999; Seemann et al., 2002).Phosphorylation of GM130 occurs during prophase and coin-cides with the fragmentation and partitioning of Golgi proteinsinto daughter cells (Lowe et al., 2000). Cyclin B–Cdc2 kinasephosphorylates GM130 on serine 25, while PP2A phosphatasedephosphorylates the residue during telophase (Lowe et al.,1998, 2000). Because mammalian oocytes contain large stock-piles of cyclin B1, cyclin B2, cdc2, and PP2A mRNA andprotein (Kubiak et al., 1993; Smith et al., 1998; Ledan et al.,2001), it is possible that these molecules act upon Golgi pro-teins during meiosis to partition the organelle.

In this report, we examine whether GM130 becomesphosphorylated during meiosis, and whether it associateswith spindle poles or ER vesicle export sites in maturebovine oocytes. Fertilized oocytes were then treated withBFA to examine whether fertilization and cytokinesis areinhibited when membrane trafficking is disrupted. Our find-ings show that phosphorylated GM130 localizes to ER ves-icle export sites and not to meiotic spindle poles, and that

neither fertilization nor cell division to the eight-cell stagerequires trafficking through the secretory pathway. Theseresults suggest that the Golgi apparatus is organized by amechanism independent of the centrosome and unique tomeiotic maturation, but that the structure and function of theGolgi are not essential for initial embryonic development.

Materials and methods

In vitro maturation and in vitro fertilization

Bovine in vitro maturation and in vitro fertilization wereperformed following standard protocols (Sirard et al., 1988),with bovine oocytes obtained either from a local abattoir orfrom BOMED, Inc. Briefly, immature oocytes were aspiratedfrom ovarian follicles and matured in 50-�l drops of TC199medium modified with 10% fetal bovine serum, 5 �g/ml fol-licle stimulating hormone, 25 �g/ml gentamycin, and 1 �g/mlestrogen for 24 h at 39°C in 5% CO2 under mineral oil. Matureoocytes were then placed into 50-�l drops of modified Ty-rode’s Albumin-Lactate-Pyruvate (TALP) culture medium(Bavister et al., 1983). Frozen bull semen (American BreedersService) was thawed to room temperature, layered over atwo-part 45%, 90% percoll gradient, and centrifuged at 700gfor 15 min to isolate live sperm from the semen. Sperm wereadded to the drops of culture medium containing the oocytes togive a final concentration of 1 � 106 sperm/ml. Oocytes andsperm were incubated at 39°C, 5% CO2 until the appropriatetime point in development.

Brefeldin A treatment

Some bovine zygotes and embryos were cultured in thepresence of the fungal metabolite brefeldin A (BFA; Sigma-Aldrich). Concentration of BFA in the culture mediumvaried from 1 �g/ml to 20 �g/ml (see Results). Zygotesincubated with BFA were either fixed 18–20 h postinsemi-nation or cultured with BFA for an additional 70 h. At 36 hpostinsemination, embryos were transferred into BFA-con-taining CR1 medium: 115 mM NaCl, 3 mM KCl, 26 mMNaHCO3, 0.4 mM pyruvate, 1 mM L-glutamine, 5 mM L(�) hemi-Ca2� lactate, 0.3% BSA, 1� BME, and MEMamino acids. Embryos were then transferred into fresh dropsof BFA-CR1 medium every 12 h and allowed to developuntil 90 h postinsemination. BODIPY FL-conjugated BFA(Molecular Probes) was used to verify that BFA entered thezygotes and embryos in culture.

Antibodies and immunocytochemistry

GM130 was identified by using mouse monoclonal an-tibody clone 35 (BD Transduction Labs) at 1:20; PS25 (giftfrom Martin Lowe), a rabbit polyclonal antibody that spe-cifically recognizes GM130 phosphorylated on serine resi-due 25, was used at 1:50. Anti-�-COP rabbit polyclonal

51C. Payne, G. Schatten / Developmental Biology 264 (2003) 50–63

antibody (1:400; Affinity BioReagents) and anti-giantinmouse monoclonal antibody (1:50; Calbiochem) were usedto identify Golgi components on COP I vesicles. Anti-mannosidase II mouse clone 53FC3 (1:1600; Covance/Babco) was used to recognize the Golgi enzyme foundwithin the cis/medial Golgi apparatus. Calreticulin, a Ca2�-binding resident protein of the endoplasmic reticulum, wasidentified by using a rabbit polyclonal antibody (1:1600;Affinity BioReagents). Sec23, a component of the COP IIvesicle coat that is a marker for export sites on the endo-plasmic reticulum, was identified by using a goat polyclonalantibody (1:50; Santa Cruz Biotechnology). Spindle micro-tubules were identified in some oocytes by using mousemonoclonal antibody clone E7 (1:5; Developmental StudiesHybridoma Bank). Control experiments were performed byusing preimmune mouse IgG antibodies at 1:20 (Chemi-con). Preincubation of antibodies for 1 h either with theircorresponding antigens or with human endothelial cell(HEC) lysates was performed as an additional control. HEClysates were provided by the cell culture core facility at thePittsburgh Development Center. AlexaFluor 488- and 568-conjugated secondary antibodies were obtained from Mo-lecular Probes and used at 1:200. To label the corticalgranules, FITC-conjugated Lens culinaris agglutinin (FITC-LCA; Sigma-Aldrich) was added to 10 mM PBS � 0.3%BSA to prepare a 10-�g/ml solution. Mature, zona pellu-cida-free oocytes were incubated in the FITC-LCA solutionfor 30 min, then washed three times in 10 mM PBS � 0.3%BSA and attached to coverslips for fixation and immuno-cytochemistry (ICC).

For ICC, the attached cumulus cells and zona pellucidaeof oocytes, zygotes, and embryos were removed with shortincubations in TALP culture medium containing 1 mg/mlhyaluronidase and 2 mg/ml pronase, respectively. Gametesand embryos were then gently attached to poly-L-lysine-coated coverslips in Ca2�-free medium, fixed for 40 min in2% formaldehyde, and permeabilized in 10 mM PBS �0.1% Triton X-100 for an additional 40 min. The sampleswere blocked in 10 mM PBS � 0.3% BSA prior to incu-bation with 1° and 2° antibodies, and DNA was labeled with10 �g/ml TOTO-3 (Molecular Probes). Fixed oocytes, zy-gotes, and embryos were imaged with a Leica TCS SP2spectral confocal microscope, using laser lines at 488-, 568-,and 633-nm wavelengths. Live embryos were visualized byusing a Nikon TE 300 inverted microscope equipped with

Hoffman modulation contrast optics, and imaged using aHamamatsu Orca CCD camera controlled by MetaMorphsoftware (Universal Imaging). Images were then processedby using Adobe Photoshop.

Results

Golgi protein GM130 disperses to clusters of endoplasmicreticulum and not to meiotic spindle poles during oocytematuration

To determine whether the putative Golgi matrix relocal-izes to spindle poles or to ER membranes during meioticmaturation, we examined the distribution of GM130 andER-resident protein calreticulin in both GV and metaphaseII-arrested (Met-2) oocytes. At the GV stage, the majority ofoocytes (81%; 116/143) show GM130 dispersed throughoutthe ooplasm as fragments (Fig. 1A). This pattern is similarto the distribution of Golgi markers giantin and BODIPY-ceramide recently characterized in mouse and rhesus GVoocytes (Moreno et al., 2002). Like these other markers, theGM130 staining concentrates around the GV surface in theinterior of the oocyte, but unlike those markers, it is reducednear the cortex and is absent from the oocyte surface. TheGM130 staining is distinct from calreticulin, which bothconcentrates around the GV surface in a punctate distribu-tion and localizes near the cortex in diffuse patches (Fig.1B). No colocalization is detected between the Golgi andthe ER in GV oocytes (Fig. 1C).

Following GVBD, completion of first meiosis, and arrestat Met-2, 86% of oocytes (134/155) show punctate foci ofGM130 localized in specific ooplasmic domains (Fig. 1D).Strikingly, these domains correspond to regions of concen-trated staining for calreticulin (Fig. 1E), showing a distri-bution of punctate Golgi proteins near ER clusters (Fig. 1F).This pattern resembles somatic cells treated with either BFAor dominant-negative Sar1p protein, in which putative ma-trix proteins appear to associate with vesicle export sites onthe ER (Ward et al., 2001; Miles et al., 2001). GM130 doesnot preferentially localize to meiotic spindle poles in Met-2oocytes (Fig. 1D, asterisks), unlike the observed distributionof GM130 to mitotic spindle poles in somatic cells (See-mann et al., 2002).

Fig. 1. Golgi and ER dynamics during oocyte in vitro maturation, in vitro fertilization, and early embryonic development. (A) GV oocytes show Golgi proteinGM130 (green) dispersed in fragments throughout the ooplasm; arrow denotes GV DNA (blue). (B) ER marker calreticulin (red) localizes in a reticulatedpattern around the GV and distributes in patches near the cortex. Distinct patterns of GM130 and calreticulin are detected when the channels are merged (C).Following in vitro maturation and Met-2 arrest (D), GM130 is localized to specific ooplasmic domains in the form of punctate foci; arrow denotes Met-2DNA and asterisks mark the meiotic spindle poles. (E) Calreticulin is now enriched in clusters throughout the cytoplasm; these clusters are detected in thesame regions as the GM130 foci when the channels are merged (F). In vitro fertilization induces oocyte activation and pronuclear (PN) formation, migration,and apposition (G), with GM130 dispersed once again as fragments that localize around the pronuclei. (H) Calreticulin distributes predominantly as clustersthroughout the cytoplasm. Distinct patterns of GM130 and calreticulin are detected when the channels are merged (I). Following the first mitotic divisionand formation of two-cell embryos (J), GM130 shows both juxtanuclear localization and cytoplasmic aggregation within the blastomeres. (K) Calreticulindistribution is enriched near the nuclei, with additional diffuse staining throughout the cytoplasm that is absent from much of the cortical region. (L) Mergeof the channels shows extensive colocalization between the Golgi and the ER.

52 C. Payne, G. Schatten / Developmental Biology 264 (2003) 50–63

Fig. 2. GM130 colocalizes with Sec23 at ER vesicle export sites but not with LCA at cortical granules. (A–C) Met-2 oocytes show GM130 (green) and Sec23(red), a marker for ER vesicle export sites, colocalizing as punctate foci throughout the cytoplasm when the channels are merged (C); arrow denotes Met-2DNA (blue) and asterisks mark the meiotic spindle poles. In pronucleate-stage (PN) zygotes (D–F), GM130 reorganizes as fragments that localize aroundthe pronuclei. An additional band of GM130 appears near the cortex (D). Sec23 continues to distribute in the cytoplasm as punctate foci (E); colocalizationbetween GM130 and Sec23 is now limited to the reticulated band near the cortex when the channels are merged (F). (G–I) Cortical sections of Met-2 oocytesshow distinct distribution patterns for GM130 and FITC-conjugated Lens culinaris agglutinin (LCA), a marker for cortical granules. No colocalization isobserved when the channels are merged (I). [Note: GM130 was originally detected in (G) with 568 nm excitation (red), while LCA was detected in (H) with488 nm excitation (green); colors were then reversed using Adobe Photoshop software.]


GM130 reorganizes around the pronuclei duringfertilization

Upon the insemination of oocytes and the formation ofpronucleate-stage (PN) zygotes, GM130 once again appearsfragmented and dispersed around the male and female pro-nuclei in the majority of oocytes (92%; 167/182; Fig. 1G).The staining pattern for calreticulin in zygotes resemblesthat observed in Met-2 oocytes, with clusters of ER distrib-uted throughout the cytoplasm (Fig. 1H). No preferentialaccumulation or distribution of ER proteins are detectedaround the two pronuclei. Like in GV oocytes, no colocal-ization is detected between GM130 and calreticulin in PNzygotes (Fig. 1I).

Following the first mitotic division, two-cell embryosshow a juxtanuclear distribution of GM130 within each ofthe blastomeres (Fig. 1J). The dispersed Golgi fragmentsobserved in GV oocytes and PN zygotes are no longer seen.Either a single Golgi or several Golgi clusters are detectedin 83% of embryos at the two-cell stage (81/98). Calreticu-lin, meanwhile, distributes more diffusely throughout thecytoplasm, with less concentrated staining observed nearthe cortex (Fig. 1K). Codistribution of the Golgi and ER isalso seen, with an increased localization of calreticulin de-tected in the regions enriched with GM130 (Fig. 1L). Thestaining patterns of the Golgi and ER in two-cell embryos,therefore, more closely resemble those observed in somaticcells.

COP II vesicle component Sec23 colocalizes with GM130during meiotic maturation

Golgi proteins were recently shown to redistribute tovesicle export sites on the ER following the inhibition of theER-to-Golgi transport pathway by BFA or mutant Arf1(Ward et al., 2001). A marker for these sites is Sec23, acomponent of the COP II coat that mediates vesicle buddingfrom regions of transitional smooth ER (Barlowe et al.,1994; Kuge et al., 1994; Paccaud et al., 1996). Because fociof GM130 localize near clusters of calreticulin in Met-2oocytes, we decided to examine the distribution of GM130with respect to Sec23 in both Met-2 oocytes and PN zy-gotes. The majority of Met-2 oocytes (95%; 72/76) show astriking colocalization between punctate foci of GM130 andSec23 dispersed throughout the cytoplasm (Fig. 2A–C).Neither protein distributes to the meiotic spindle poles (Fig.2A, asterisks). While Sec23 foci are also seen in PN zy-gotes, GM130 predominantly and distinctly reorganizesaround the pronuclei, and colocalization is now limited toregions of reticulated distribution near the cortex (Fig. 2D–F). This staining pattern is observed in 78% of zygotes(63/81). It appears that GM130 relocalizes during meioticmaturation to associate with Sec23 at ER export sites, thenredistributes again during fertilization to localize to theapposed pronuclei.

The distribution of GM130 near the oocyte cortex sug-

gests that Golgi proteins may also associate with corticalgranules, membrane-bound organelles which secrete a va-riety of enzymes during fertilization to block polyspermy(Wessel et al., 2001). Cortical granules release their con-tents upon sperm activation to modify the extracellularenvironment of the oocyte, and can be labeled in Met-2oocytes with FITC-conjugated Lens culinaris agglutinin(LCA), a lectin isolated from lentils (Cherr et al., 1988;Ducibella et al., 1988a; Wang et al., 1997). When Met-2oocytes are double labeled for GM130 and LCA, the Golgiproteins and cortical granules distribute as distinct foci (Fig.2G–I), with no colocalization detected in any of the oocytesexamined (68/68). Thus, while GM130 associates withSec23 and ER export sites during meiotic maturation, itdoes not associate with either meiotic spindle poles orcortical granules.

GM130 is phosphorylated during oocyte maturation anddephosphorylated during fertilization

The fragmentation, dispersion, and reorganization of theGolgi apparatus observed between GVBD and PN apposi-tion raise the possibility that GM130 might be phosphory-lated in the oocyte during this time. Therefore, to charac-terize the condition of this protein in GV and Met-2 arrestedoocytes, PN zygotes, and two-cell embryos, we used anantibody that specifically recognizes the phosphorylatedform of GM130 (Lowe et al., 2000).

At the GV and Met-2 stages, the majority of oocytes(90%; 114/127) contain almost all of their GM130 proteinin a phosphorylated state, revealed through double labelingwith anti-GM130 antibodies (Fig. 3A–F). Since GV oocytesare in prophase of first meiosis and Met-2 oocytes are inmetaphase of second meiosis, the phosphorylation ofGM130 appears to persist throughout meiotic maturation.Following insemination and pronuclear formation, all ob-served zygotes (92/92) contain GM130 in a dephosphory-lated state, since no signal is detected with anti-phospho-GM130 antibodies (Fig. 3G–I). When two-cell embryos areexamined shortly after cytokinesis, the majority of GM130is dephosphorylated and juxtanuclear (85%; 76/89). A nar-row band of GM130, however, is phosphorylated and re-sides near the cleavage furrow in some of the embryos (Fig.3J–L).

Brefeldin A disrupts a functionally active Golgi apparatusbut does not inhibit development during fertilization

By assessing the motility of energy-dependent vesiclescoated with �-coatomer (�-COP) and the sensitivity ofmeiotic maturation to brefeldin A (BFA), it was recentlyshown that Golgi fragments are functionally active in mouseGV oocytes (Moreno et al., 2002). COP I vesicles, whichare comprised of �-COP, participate in anterograde andretrograde membrane transport in the ER and Golgi (Loweand Kreis, 1998). While the Golgi fragments disperse dur-


ing meiotic maturation, they relocalize around the two pro-nuclei during fertilization. Therefore, we questionedwhether �-COP distribution and BFA sensitivity are alsoexhibited during the fertilization process.

Control-treated zygotes show �-COP protein dispersedaround the male and female pronuclei in a pattern thatresembles GM130 distribution (Fig. 4A). The majority ofzygotes treated with 5 �g/ml BFA, however, display asevere disruption in �-COP staining (89%; 39/44). Punctate�-COP redistributes throughout the cytoplasm under theseconditions (Fig. 4B). Despite the disruption and redistribu-tion of �-COP, however, pronuclear union appears unaf-fected. The surface-to-surface internuclear distances of thepronuclei do not differ greatly between control and BFAtreatment conditions (Table 1). All distances are �10 �m,the average diameter of a pronucleus. Successful pronuclearmigration and apposition occur despite the fragmentation ofthe Golgi apparatus, measured by examining the distribu-tion of GM130 (Fig. 4C and D), cis/medial Golgi enzymemannosidase II (Fig. 4E and F), and COP I vesicle proteingiantin (Fig. 4G and H). All of the BFA-treated zygotesstained for these Golgi markers contain pronuclei that are�10 �m apart, identical to controls (Table 1). An intactGolgi apparatus, therefore, does not appear to be necessaryfor the microtubule-dependent mechanism of pronuclearmigration in the zygote. We wondered, however, whetherBFA would affect subsequent mitosis and cytokinesis.

Brefeldin A does not inhibit cytokinesis in theprecompacted embryo

Embryos treated with 5 �g/ml BFA show no significantdifferences in developmental rates when compared withcontrol-treated embryos; approximately the same number ofprecompacted eight-cell embryos form (48.7% control vs.44.5% BFA-treated; Table 2). Similar numbers of embryosarrest at the 1-cell, 2-cell, and 3/4-cell stages under bothculture conditions. Higher concentrations of BFA inducecell cycle arrest in the zygote, causing death prior to firstmitosis (data not shown). When compared with controls,zygotes treated with BFA form 2-cell embryos that displaynormal outward appearance, showing well-formed cleavagefurrows and evenly distributed cytoplasm (Fig. 5A and B).Under BFA treatment, development to the 3/4-cell stage isalso achieved (Fig. 5D), although the cleavage furrows arenot as clearly defined between all blastomeres as they are incontrols (Fig. 5C). By the time 8-cell embryos form, thosetreated with BFA show an increase in misshapen blas-

tomeres and blebbing, in addition to poorly defined cleav-age furrows (Fig. 5F). Despite these anomalies, individualblastomeres can still be discerned. However, unlike thecontrol embryos, development arrests at this stage and theseembryos fail to compact (Fig. 5E; shown as precompactedembryos).

Discussion

We have examined the potential roles of a putative Golgimatrix protein during bovine oocyte maturation, fertiliza-tion, and cytokinesis. Our results provide new insights intoGolgi function during meiosis as compared with mitosis,and during the cell cycles of fertilization and early embry-onic development. The finding that Golgi protein GM130localizes to ER vesicle export sites during meiotic matura-tion suggests that meiosis may follow a unique set of re-quirements that could be facilitated by the absence of acentrosome. GM130 phosphorylation coincides with Golgifragmentation, and might ensure correct partitioning withinthe oocyte. While a functional secretory pathway is ulti-mately necessary for embryo compaction, it appears not tobe required for either fertilization or subsequent cell divi-sion.

ER-resident proteins redistribute during oocytematuration but do not reorganize during fertilization

Previous studies using mouse and hamster oocytes haveshown that a dramatic reorganization and concentration ofmembrane vesicles and inositol 1,4,5-triphosphate (IP3) re-ceptors occurs near the cortex during meiotic maturation(Ducibella et al., 1988b; Mehlmann et al., 1995; Shiraishi etal., 1995). IP3 receptors, which mediate Ca2� release fromthe ER during fertilization, become sensitized during thematuration process (Kline, 2000). It is thought that theformation of cortical ER clusters facilitates the generationof repetitive Ca2� waves upon oocyte activation (Kline etal., 1999). Our current observations of ER rearrangementsin bovine, shifting from irregular masses at the GV stage toordered clusters at Met-2 arrest, resemble the redistributionpreviously detected in hamster (Shiraishi et al., 1995). ERmembranes initially concentrated around the GV and inpatches near the cortex disperse to form clusters positionedat periodic intervals that are absent from the region sur-rounding the meiotic spindle.

When oocyte activation and sperm entry trigger Ca2�

Fig. 3. Phosphorylation cycle of GM130 during oocyte in vitro maturation, in vitro fertilization, and early embryonic development. (A–C) GV oocytes showGolgi protein GM130 (green; detected with antibody clone 35) in a phosphorylated state (red; detected with phospho-specific polyclonal antibody PS25) atprophase of meiosis I; arrow denotes GV DNA (blue). Nearly identical GM130 staining is observed when the channels are merged (C). In Met-2 arrestedoocytes (D–F), the punctate GM130 is phosphorylated, with similar distribution patterns for both antibodies detected when the channels are merged (F); arrowdenotes Met-2 DNA. During pronuclear (PN) migration in zygotes (G–I), GM130 is dephosphorylated, as no PS25 labeling is detected (H). In two-cellembryos (J–L), the majority of GM130 is dephosphorylated, except for a narrow band of protein at the cleavage furrow (arrows).


release during fertilization, sperm aster microtubules form,pronuclei develop, and Golgi membranes reorganize. Likethe observations made in activated mouse oocytes (Kline etal., 1999), we note here that the cortical ER clusters do notreorganize in bovine zygotes. This stable organization maybe required for the generation of Ca2� oscillations thatoccurs following oocyte activation in mammals (Kline,2000). This differs from events observed in other phyla: seaurchin zygotes contain ER membranes that reorganize andredistribute to the centrosome attached to the male pronu-cleus (Terasaki and Jaffe, 1991). Such rearrangements cor-respond to a single Ca2� transient (Eisen et al., 1984). ERdynamics at fertilization, therefore, may depend uponwhether Ca2� release is singular or oscillatory (Kline et al.,1999).

In the two-cell bovine embryo, cortical ER clusters areno longer detected. The redistribution of membrane to amore diffuse network surrounding the nucleus reflects anorganization similar to that found in somatic cells. Obser-vations in mouse embryos have shown that IP3 receptors aredownregulated after fertilization (Parrington et al., 1998),suggesting that ER reorganization may correspond to thesechanges. Our results indicate that structural alterations ofthe ER, including a loss of clusters, occur following thezygotic cell cycle. These changes may reflect a shift in ERstructure and function required by multicellular embryos.

Phosphorylated GM130 localizes to the ER but not themeiotic spindle in the mature oocyte

We have shown that, throughout oocyte meiosis, as insomatic cell mitosis, Golgi protein GM130 is in a phosphor-ylated state. Both the clustered Golgi fragments at the GVstage and the dispersed Golgi foci at Met-2 arrest are rec-ognized by the phospho-specific antibody PS25. GM130becomes dephosphorylated following sperm incorporation,with PP2A previously identified as the phosphatase that actsupon the serine residue (Lowe et al., 2000). Interestingly,

Fig. 4. Brefeldin A (BFA) disrupts the Golgi apparatus but does notinhibit pronuclear apposition in zygotes. (A) COP I vesicle coatomerprotein �-COP localizes to a region surrounding the two apposingpronuclei in control zygotes; arrows indicate the edge of the stainingregion. In BFA-treated zygotes (B), �-COP distributes diffusely withinthe cytoplasm, with additional punctate staining. Pronuclear appositionis not inhibited, however, despite nearly 20 h of exposure to BFA. Incontrol zygotes, Golgi proteins GM130 (C), mannosidase II (E), andgiantin (G) localize to regions surrounding the apposing pronuclei;arrows indicate edges of the staining regions. BFA-treatment alters thedistribution of these Golgi proteins, with punctate GM130 (D) dispersedthroughout the cytoplasm, and both mannosidase II (F) and giantin (H)enriched in clusters near the cortex and dispersed as punctate foci.Despite the disruption caused to Golgi membranes, BFA does notinhibit pronuclear apposition, with all pronuclei showing surface-to-surface internuclear distances of �10 �m. The average diameter of apronucleus is 10 �m.

Table 1Distance between pronuclei in one cell zygotes

Condition of zygote �10 �m apart �10 �m apart

control 0% (0/127) 100% (127/127)� brefeldin A 0% (0/131) 100% (131/131)

Table 2Effect of brefeldin A (BFA) on bovine embryonic development in vitro

Number of fertilized eggs at the 1-, 2-, 3/4-,8-cell stage 90 hours after sperm addition

% of embryos atthe 8-cell stage

control234

1-cell50

2-cell13

3/4-cell57

8-cell114

114/234 (48.7%)

BFA-treated218

1-cell48

2-cell24

3/4-cell49

8-cell97

97/218 (44.5%)


PP2A has been shown to localize to the Met-2 spindle (Luet al., 2002), a region to which GM130 does not redistribute.The majority of GM130 is dephosphorylated in two-cellembryos, except for a small band of phosphorylated proteinat the site of the cleavage furrow (Fig. 3G and H). Perhapsthis retention is important for the cytokinesis process in theearly embryo.

Immunofluorescence detection of the Golgi apparatus,fragmenting during meiotic maturation and reorganizingaround the pronuclei during fertilization and early embryo-genesis, is consistent with previous ultrastructural reports(Hyttel et al., 1986, 1988, 1989; Assey et al., 1994; Planteand King, 1994). Transmission electron microscopy hasshown that the Golgi changes from well-developed com-plexes at the GV stage to highly dispersed fragments atMet-2 arrest (Hyttel et al., 1986; Assey et al., 1994). In PNzygotes, the Golgi appears once again as defined complexes,or flattened stacks of lamellae, that often associate near thepronuclear membranes (Hyttel et al., 1988; P. Sutovsky,personal communication). The dynamics of GM130 ob-served here, together with the phosphorylated state ofGM130 detected in GV and Met-2 oocytes, suggests that thefragmentation and dispersion of GM130 may accompany itsphosphorylation.

Somatic cells treated with BFA show redistribution ofGM130 to structures adjacent to ER domains (Ward et al.,2001). This evidence supports the view of the Golgi appa-ratus as a dynamic structure, the maintenance of whichdepends upon membrane recycling to and from the ER(Lippincott-Schwartz et al., 2000). An alternative view isthat the Golgi is an autonomous organelle with stable com-ponents, capable of providing a template for its own growthand division (Shorter and Warren, 2002). Supporting evi-dence for this model is provided by mitotic cells, in whichER markers distribute as a fine reticulum that is excludedfrom areas near the spindle poles. Golgi markers—includ-ing GM130—localize under these conditions as tiny vesi-cles that are enriched at the centrosome regions of thespindle (Jesch and Linstedt, 1998; Seemann et al., 2002).This association of the Golgi with the mitotic spindle isthought to ensure accurate partitioning of Golgi proteinsinto the daughter cells. We find here that GM130 colocal-izes with the ER vesicle export site marker Sec23 and notwith the meiotic spindle in Met-2 oocytes. Perhaps thispreferential distribution ensures that the majority of Golgiprotein remains in the oocyte and does not get expelledalong with the chromosomes in either the first or secondpolar bodies. Because the polar bodies are destined for

degeneration, partitioning the Golgi into their cytoplasmwould not be developmentally advantageous.

Interestingly, GM130 localization to ER vesicle exportsites in bovine oocytes contrasts with giantin distribution tospindle poles observed in mouse oocytes (Moreno et al.,2002). This differential distribution of Golgi proteins couldbe attributed to the presence of a maternal centrosome inmouse oocytes and its absence in bovine (Schatten, 1994).Given that the mature bovine oocyte is acentrosomal, andbecause the secretory pathway is inhibited during meioticmaturation, the reorganization of Golgi during meiosismight naturally follow the mechanism of redistribution toER export sites after BFA or nocodazole treatment. Thiscondition would contrast with the distribution of Golgi tothe spindle poles during mitosis. Conversely, the centro-some in mouse oocytes might promote Golgi localization tothe meiotic spindle to partition Golgi proteins in a processdistinct to rodents. Further experiments will be needed toclarify these associations, and to address the role of thecentrosome in organizing the Golgi during meiotic matura-tion.

Pronuclear formation, migration, and apposition do notrequire a functional secretory pathway

Because the Golgi apparatus reorganizes around the pro-nuclei as they unite during fertilization, membrane traffick-ing along microtubules might establish a physical connec-tion between the Golgi and pronuclear membranes. Thiscould allow the Golgi to “drag” the female pronucleustoward the centrosome attached to the male pronucleus.Treatment of pronucleate-stage zygotes with BFA, how-ever, allowed us to determine that neither an intact Golgiapparatus nor a functional secretory pathway is necessaryfor successful pronuclear migration and apposition. Golgicomponents on COP I vesicles (�-COP, giantin), on puta-tive cis-matrix (GM130), and in the enzymatic pathway(mannosidase II) are all disrupted when exposed to BFA,yet the male and female pronuclei unite as in controls (Fig.4; Table 1). Specifically, the female pronucleus migratesalong microtubules toward the male pronucleus that is usu-ally positioned near the center of the egg (Schatten, 1994).These results contrast with previous observations using aCa2� ionophore to parthenogenetically activate mouse oo-cytes, in which BFA prevented female pronuclei from mi-grating to the center as in controls (Clayton et al., 1995). Weattribute this discrepancy to different activation methods

Fig. 5. Effect of Brefeldin A (BFA) on development in early embryos. In vitro fertilized oocytes were cultured for 90 h under control conditions or in thepresence of BFA. (A, B) Two-cell embryos develop from zygotes in both control (A) and BFA-treated groups (B), with the latter displaying normal outwardappearance and defined cleavage furrows. The formation of three-cell and four-cell embryos occurs under BFA conditions (D), though outward appearanceis slightly diminished and cleavage furrows are not as clearly defined as in controls (C). Eight-cell embryos develop normally in the control (E), but not theBFA-treated group (F), in which embryos display poorly shaped blastomeres, ill-defined cleavage furrows and increased blebbing. Compaction occurs in thecontrol embryos, but not in those treated with BFA.


and divergent centrosome inheritance strategies betweenmouse and bovine (Schatten, 1994).

Golgi membranes are organized by the centriole pair andpericentriolar material of the centrosome that focuses theGolgi to a juxtanuclear position in many somatic cells(Thyberg and Moskalewski, 1999). Bull sperm, like rhesusand human, introduces the centrosome to the egg cytoplasm,resulting in centrosome attachment to the male pronucleus.Movement of Golgi membranes along microtubules couldtherefore accompany the migration of the female pronucleustowards the centrosome. Indeed, the dynamics of giantinand �-COP, which redistribute in meiotic mouse oocytes,identifies the likelihood of vesicle trafficking. Curiously,while BFA treatment arrests in vitro maturation (Moreno etal., 2002), it does not inhibit in vitro fertilization, highlight-ing a significant difference between the two events.

Cytokinesis in precompacted embryos may proceed in theabsence of a functional Golgi apparatus

It has been shown in Xenopus and C. elegans embryosundergoing cytokinesis that new membrane is delivered toand inserted within the region of the cleavage furrow (Da-nilchik et al., 1998; Skop et al., 2001). This vesicle traffick-ing involves microtubules and is accompanied by the secre-tion of a variety of proteins (Straight and Field, 2000).When C. elegans embryos are treated with BFA, cytokinesisfails after an initial, well-formed cleavage furrow regresses(Skop et al., 2001). This failure is attributed to impairedsecretion and the absence of membrane accumulation.Mouse embryos, in contrast, do not show cytokinesis failurewhen exposed to BFA, since activated oocytes cleave to thetwo-cell stage despite an inhibited secretory pathway (Clay-ton et al., 1995). Our finding that bovine zygotes completecytokinesis and divide through the eight-cell stage in thepresence of BFA agrees with those observations in mouse,and suggests that mammalian embryos might not require afunctional secretory pathway for cytokinesis prior to com-paction.

The differential effects of BFA on cytokinesis observedin C. elegans, mouse, and bovine embryos might, however,be attributed to the varying concentrations of drug used inthe different studies. It was noted that 15 �g/ml BFA con-sistently induced cytokinesis failures in C. elegans withoutimpacting cell viability, while lower concentrations did notreproduce the phenotype (Skop et al., 2001). Studies onmouse, and here using bovine, expose embryos to 5 �g/mlBFA which, despite failing to inhibit cytokinesis, com-pletely disrupts the Golgi apparatus and prevents the surfaceexpression of the secretory protein E-cadherin (Fig. 4; Clay-ton et al., 1995). Cytokinesis failure might therefore requirea higher concentration of BFA than that which is sufficientto disrupt the Golgi and inhibit secretion. We noted, how-ever, that BFA concentrations higher than 5 �g/ml wereoften detrimental to bovine cell cycle progression.

Loss of E-cadherin from the surface of mouse embryos

impairs the cell–cell adhesion of blastomeres (Clayton etal., 1995). The process of compaction requires a significantincrease in intercellular adhesion and polarization, and re-lies upon the preservation and maintenance of secretion.Morphological observations of bovine embryonic develop-ment in the presence of BFA suggest that both the shape anddefinition of the blastomeres are affected by this exposure,and that the inhibition of compaction could be due to animpaired secretory pathway. These findings raise the possi-bility that, while the Golgi apparatus is necessary for theexpression of essential proteins on the cell surface andsecretion of factors into the extracellular milieu, perhaps itis not required for cytokinesis progression in early mamma-lian embryos. Further studies will be needed to addressthese important questions concerning the role of Golgi pro-teins during early embryonic development.

Conclusions

Meiotic maturation prepares the oocyte for fertilizationthrough a dramatic reorganization and partitioning of chro-matin and cytoplasmic components. Putative Golgi matrixproteins likely associate with clusters of ER and vesicleexport sites to ensure their retention for embryogenesis.Unlike mitosis, the outcome of meiosis is not an equaldistribution of cytoplasmic content into daughter cells. Theabsence of the Golgi on the meiotic spindle reduces thelikelihood of Golgi partitioning into polar bodies. Interest-ingly, once fertilization is underway, processes as distinct asnuclear trafficking and cytokinesis do not depend upon afunctional Golgi apparatus. These results suggest that thesecretory pathway may not be essential for embryonic de-velopment prior to compaction, and that both oocyte matu-ration and fertilization provide unique environments inwhich to study Golgi protein structure and function.

Acknowledgments

We thank Martin Lowe for the gift of PS25 anti-phos-pho-GM130 antibody, and Tanja Dominko, RicardoMoreno, Chris Navara, Joao Ramalho-Santos, VanesaRawe, Cal Simerly, and Peter Sutovsky for experimentalassistance and helpful discussions. We are also grateful toLeah Kauffman for her critical reading of the manuscript.This work was supported by NIH research grants to G.S.

References

Aitken, R.J., 1997. The cell biology of fertilization. Adv. Exp. Med. Biol.424, 291–299.

Albertini, D.F., 1992. Cytoplasmic microtubular dynamics and chromatinorganization during mammalian oogenesis and oocyte maturation. Mu-tat. Res. 296, 57–68.

Assey, R.J., Hyttel, P., Greve, T., Purwantara, B., 1994. Oocyte morphol-ogy in dominant and subordinate follicles. Mol. Reprod. Dev. 37,335–344.


Barlowe, C., Orci, L., Yeung, T., Hosobuchi, M., Hamamoto, S., Salama,N., Rexach, M.F., Ravazzola, M., Amherdt, M., Schekman, R., 1994.COPII: a membrane coat formed by Sec proteins that drive vesiclebudding from the endoplasmic reticulum. Cell 77, 895–907.

Bavister, B.D., Leibfried, M.L., Lieberman, G., 1983. Development ofpreimplantation embryos of the golden hamster in a defined culturemedium. Biol. Reprod. 28, 235–247.

Calarco, P.G., Donahue, R.P., Szollosi, D., 1972. Germinal vesicle break-down in the mouse oocyte. J. Cell Sci. 10, 369–385.

Cherr, G.N., Drobnis, E.Z., Katz, D.F., 1988. Localization of corticalgranule constituents before and after exocytosis in the hamster egg. J.Exp. Zool. 246, 81–93.

Clayton, L., McConnell, J.M.L., Johnson, M.H., 1995. Control of thesurface expression of vomorulin after activation of mouse oocytes.Zygote 3, 177–189.

Danilchik, M.V., Funk, W.C., Brown, E.E., Larkin, K., 1998. Requirementfor microtubules in new membrane formation during cytokinesis ofXenopus embryos. Dev. Biol. 194, 47–60.

Ducibella, T., Anderson, E., Albertini, D.F., Aalberg, J., Rangarajan, S.,1988a. Quantitative studies of changes in cortical granule number anddistribution in the mouse oocyte during meiotic maturation. Dev. Biol.130, 184–197.

Ducibella, T., Rangarajan, S., Anderson, E., 1988b. The development ofmouse oocyte cortical reaction competence is accompanied by majorchanges in cortical vesicles and not cortical granule depth. Dev. Biol.130, 789–792.

Eisen, A., Kiehart, D.P., Wieland, S.J., Reynolds, G.T., 1984. Temporalsequence and spatial distribution of early events of fertilization insingle sea urchin eggs. J. Cell Biol. 99, 1647–1654.

Gosden, R., Krapez, J., Briggs, D., 1997. Growth and development of themammalian oocyte. Bioessays 19, 875–882.

Hyttel, P., Callesen, H., Greve, T., 1986. Ultrastructural features of pre-ovulatory oocyte maturation in superovulated cattle. J. Reprod. Fertil.76, 645–656.

Hyttel, P., Greve, T., Callesen, H., 1988. Ultrastructure of in-vivo fertili-zation in superovulated cattle. J. Reprod. Fertil. 82, 1–13.

Hyttel, P., Greve, T., Callesen, H., 1989. Ultrastructural aspects of oocytematuration and fertilization in cattle. J. Reprod. Fertil. Suppl. 38,35–47.

Jesch, S.A., Linstedt, A.D., 1998. The Golgi and endoplasmic reticulumremain independent during mitosis in HeLa cells. Mol. Biol. Cell 9,623–635.

Kline, D., Mehlmann, L., Fox, C., Terasaki, M., 1999. The cortical endo-plasmic reticulum (ER) of the mouse egg: localization of ER clusters inrelation to the generation of repetitive calcium waves. Dev. Biol. 215,431–442.

Kline, D., 2000. Attributes and dynamics of the endoplasmic reticulum inmammalian eggs. Curr. Top. Dev. Biol. 50, 125–154.

Kubiak, J.Z., Weber, M., de Pennart, H., Winston, N.J., Maro, B., 1993.The metaphase II arrest in mouse oocytes is controlled through micro-tubule-dependent destruction of cyclin B in the presence of CSF.EMBO J. 12, 3773–3778.

Kuge, O., Dascher, C., Orci, L., Rowe, T., Amherdt, M., Plutner, H.,Ravazzola, M., Tanigawa, G., Rothman, J.E., Balch, W.E., 1994. Sarlpromotes vesicle budding from the endoplasmic reticulum but notGolgi compartments. J. Cell Biol. 125, 51–65.

Ledan, E., Polanski, Z., Terret, M.E., Maro, B., 2001. Meiotic maturationof the mouse oocyte requires an equilibrium between cyclin B synthesisand degradation. Dev. Biol. 232, 400–413.

Lippincott-Schwartz, J., Yuan, L.C., Bonifacino, J.S., Klausner, R.D.,1989. Rapid redistribution of Golgi proteins into the ER in cells treatedwith brefeldin A: evidence for membrane cycling from Golgi to ER.Cell 56, 801–813.

Lippincott-Schwartz, J., Roberts, T.H., Hirschberg, K., 2000. Secretoryprotein trafficking and organelle dynamics in living cells. Annu. Rev.Cell Dev. Biol. 16, 557–589.

Lowe, M., Kreis, T.E., 1998. Regulation of membrane traffic in animalcells by COPI. Biochim. Biophys. Acta 1404, 53–66.

Lowe, M., Rabouille, C., Nakamura, N., Watson, R., Jackman, M., Jamsa,E., Rahman, D., Pappin, D.J., Warren, G., 1998. Cdc2 kinase directlyphosphorylates the cis-Golgi matrix protein GM130 and is required forGolgi fragmentation in mitosis. Cell 94, 783–793.

Lowe, M., Gonatas, N.K., Warren, G., 2000. The mitotic phosphorylationcycle of the cis-Golgi matrix protein GM130. J. Cell Biol. 149, 341–356.

Lu, Q., Dunn, R.L., Angeles, R., Smith, G.D., 2002. Regulation of spindleformation by active mitogen-activated protein kinase and protein phos-phatase 2A during mouse oocyte meiosis. Biol. Reprod. 66, 29–37.

McGaughey, R.W., Racowsky, C., Rider, V., Baldwin, K., DeMarais,A.A., Webster, S.D., 1990. Ultrastructural correlates of meiotic matu-ration in mammalian oocytes. J. Electron Microsc. Tech. 16, 257–280.

Mehlmann, L.M., Terasaki, M., Jaffe, L.A., Kline, D., 1995. Reorganiza-tion of the endoplasmic reticulum during meiotic maturation of themouse oocyte. Dev. Biol. 170, 607–615.

Miles, S., McManus, H., Forsten, K.E., Storrie, B., 2001. Evidence that theentire Golgi apparatus cycles in interphase HeLa cells: sensitivity ofGolgi matrix proteins to an ER exit block. J. Cell Biol. 155, 543–555.

Moreno, R.D., Schatten, G., Ramalho-Santos, J., 2002. Golgi apparatusdynamics during mouse oocyte in vitro maturation: effect of the mem-brane trafficking inhibitor brefeldin A. Biol. Reprod. 66, 1259–1266.

Nakamura, N., Rabouille, C., Watson, R., Nilsson, T., Hui, N., Slusar-ewicz, P., Kreis, T.E., Warren, G., 1995. Characterization of a cis-Golgi matrix protein, GM130. J. Cell Biol. 131, 1715–1726.

Navara, C.S., First, N.L., Schatten, G., 1994. Microtubule organization inthe cow during fertilization, polyspermy, parthenogenesis, and nucleartransfer: the role of the sperm aster. Dev. Biol. 162, 29–40.

Paccaud, J.P., Reith, W., Carpentier, J.L., Ravazzola, M., Amherdt, M.,Schekman, R., Orci, L., 1996. Cloning and functional characterizationof mammalian homologues of the COPII component Sec23. Mol. Biol.Cell 7, 1535–1546.

Parrington, J., Brind, S., De Smedt, H., Gangeswaran, R., Lai, F.A.,Wojcikiewicz, R., Carroll, J., 1998. Expression of inositol 1,4,5-triphosphate receptors in mouse oocytes and early embryos: the type 1isoform is upregulated in oocytes and downregulated after fertilization.Dev. Biol. 203, 451–461.

Perreault, S.D., 1992. Chromatin remodeling in mammalian zygotes. Mu-tat. Res. 296, 43–55.

Plante, L., King, W.A., 1994. Light and electron microscopic analysis ofbovine embryos derived by in vitro and in vivo fertilization. J. Assist.Reprod. Genet. 11, 515–529.

Rouviere, C., Houliston, E., Carre, D., Chang, P., Sardet, C., 1994. Char-acteristics of pronuclear migration in Beroe ovata. Cell Motil. Cy-toskeleton 29, 301–311.

Sathananthan, A.H., Kola, I., Osborne, J., Trounson, A., Ng, S.C., Bongso,A., Ratnam, S.S., 1991. Centrioles in the beginning of human devel-opment. Proc. Natl. Acad. Sci. USA 88, 4806–4810.

Schatten, H., Schatten, G., Mazia, D., Balczon, R., Simerly, C., 1986.Behavior of centrosomes during fertilization and cell division in mouseoocytes and in sea urchin eggs. Proc. Natl. Acad. Sci. USA 83, 105–109.

Schatten, G., 1994. The centrosome and its mode of inheritance: thereduction of the centrosome during gametogenesis and its restorationduring fertilization. Dev. Biol. 165, 299–335.

Seemann, J., Pypaert, M., Taguchi, T., Malsam, J., Warren, G., 2002.Partitioning of the matrix fraction of the Golgi apparatus during mitosisin animal cells. Science 295, 848–851.

Shiraishi, K., Okada, A., Shirakawa, H., Nakanishi, S., Mikoshiba, K.,Miyazaki, S., 1995. Developmental changes in the distribution of theendoplasmic reticulum and inositol 1,4,5-triphosphate receptors and thespatial pattern of Ca2� release during maturation of hamster oocytes.Dev. Biol. 170, 594–606.

Shorter, J., Warren, G., 2002. Golgi architecture and inheritance. Annu.Rev. Cell Dev. Biol. 18, 379–420.


Sirard, M.A., Parrish, J.J., Ware, C.B., Leibfried-Rutledge, M.L., First,N.L., 1988. The culture of bovine oocytes to obtain developmentallycompetent embryos. Biol. Reprod. 39, 546–552.

Skop, A.R., Bergmann, D., Mohler, W.A., White, J.G., 2001. Completionof cytokinesis in C. elegans requires a brefeldin A-sensitive membraneaccumulation at the cleavage furrow apex. Curr. Biol. 11, 735–746.

Slusarewicz, P., Nilsson, T., Hui, N., Watson, R., Warren, G., 1994.Isolation of a matrix that binds medial Golgi enzymes. J. Cell Biol. 124,405–413.

Smith, G.D., Sadhu, A., Mathies, S., Wolf, D.P., 1998. Characterization ofprotein phosphatases in mouse oocytes. Dev. Biol. 204, 537–549.

Straight, A.F., Field, C.M., 2000. Microtubules, membranes and cytokine-sis. Curr. Biol. 10, R760–R770.

Sutovsky, P., Hewitson, L., Simerly, C.R., Tengowski, M.W., Navara,C.S., Haavisto, A., Schatten, G., 1996. Intracytoplasmic sperm injec-tion for Rhesus monkey fertilization results in unusual chromatin,cytoskeletal, and membrane events, but eventually leads to pronucleardevelopment and sperm aster assembly. Hum. Reprod. 11, 1703–1712.

Terasaki, M., Jaffe, L.A., 1991. Organization of the sea urchin egg endo-plasmic reticulum and its reorganization at fertilization. J. Cell Biol.114, 929–940.

Thyberg, J., Moskalewski, S., 1999. Role of microtubules in the organi-zation of the Golgi complex. Exp. Cell Res. 246, 263–279.

Wang, W., Hosoe, M., Li, R., Shioya, Y., 1997. Development of thecompetence of bovine oocytes to release cortical granules and blockpolyspermy after meiotic maturation. Dev. Growth Differ. 39, 607–615.

Ward, T.H., Polishchuk, R.S., Caplan, S., Hirschberg, K., Lippin-cott-Schwartz, J., 2001. Maintenance of Golgi structure and func-tion depends on the integrity of ER export. J. Cell Biol. 155,557–570.

Wassarman, P.M., Josefowicz, W.J., 1978. Oocyte development in themouse: an ultrastructural comparison of oocytes isolated at variousstages of growth and competence. J. Morphol. 156, 209–235.

Wessel, G.M., Brooks, J.M., Green, E., Haley, S., Voronina, E., Wong, J.,Zaydfudim, V., Conner, S., 2001. The biology of cortical granules. Int.Rev. Cytol. 209, 117–206.

Zaal, K.J., Smith, C.L., Polishchuk, R.S., Altan, N., Cole, N.B., Ellenberg,J., Hirschberg, K., Presley, J.F., Roberts, T.H., Siggia, E., Phair, R.D.,Lippincott-Schwartz, J., 1999. Golgi membranes are absorbed into andreemerge from the ER during mitosis. Cell 99, 589–601.


golgi dynamics during meiosis are distinct from mitosis and are coupled to endoplasmic reticulum...

Documents