Microtubule and Chromatin Dynamics during Fertilization and Early ...

BIOLOGY OF REPRODUCTION 55, 260-270 (1996)

Microtubule and Chromatin Dynamics during Fertilization and Early Development in Rhesus Monkeys, and Regulation by Intracellular Calcium Ions Gwo-Jang Wu,3 Calvin Simerly, Sara S. Zoran, L.R. Funte, and Gerald Schatten 2 Departments of Zoology and Obstetrics and Gynecology and Wisconsin Regional Primate Research Center, University of Wisconsin, Madison, Wisconsin 53706 fertilization, polyspermy, and parthenogenetic activation using oocytes from a nonhuman primate, the rhesus monkey. In vitro fertilization was performed with use of gametes from prime, fertile, breeding rhesus adults, and the microtubule and DNA configurations were explored with laserscanning confocal microscopy. Objectives included determining the pattern of centrosome inheritance in this primate, which was examined during monospermy, polyspermy (in which paternal contributions are multiplied), and parthenogenesis (in which the paternal contributions are absent), and exploring whether or not microtubules are essential for pronuclear apposition during primate fertilization. In the course of this research, methods for inducing parthenogenetic activation in rhesus oocytes were developed; and to better understand the mechanisms of these activation protocols, calcium imaging of fertilization and activation was performed. It is well known that an increase in cytosolic calcium occurs at fertilization in several species, including sea urchins, Xenopus, mice, hamsters, rabbits, cattle, and humans [3-6], suggesting that calcium plays a central role in fertilization. The role of calcium in triggering meiotic resumption and developmental progression in rhesus monkeys is not clearly defined. Our research characterized the pattern of calcium dynamics during in vitro insemination and parthenogenetic activation of rhesus oocytes to determine whether the rhesus may be an appropriate model for human fertilization. Results indicate that calcium dynamics during insemination in the rhesus are similar to those reported recently for the human [4-6].

ABSTRACTTo explore primate fertilization, oocytes and zygotes from fertile rhesus monkeys were imaged throughout fertilization, polyspermy, and artificial activation using confocal microscopy for microtubules and DNA, as well as ratiometric computer-enhanced video microscopy for intracellular calcium. Unfertilized oocytes displayed microtubules only in the radially oriented meiotic spindles. At insemination, a large calcium transient was followed by a series of smaller oscillations, and sperm astral microtubules had assembled from the sperm centrosome by 2.5 h after transient onset. This aster enlarged, and later duplicated, as the pronuclei converged near the cortex. Pronuclear apposition was prevented by microtubule inhibitors. At mitotic prophase, microtubules ensheathed both sets of condensing chromosomes. At metaphase, the spindle was barrelshaped and eccentrically positioned with two small asters at the pole with the sperm tail. Microtubules emanating from the telophase spindle interacted with the adjacent cortex and displaced the spindle toward the cell center as first cytokinesis ensued. During polyspermy, each sperm nucleated an aster, and the frequency of calcium oscillations increased. Activation resulted initially in disarrayed microtubules that eventually organized into functional mitotic spindles. These kinetic results demonstrate that rhesus monkeys accomplish fertilization in a fashion nearly identical to that of humans and are, therefore, ideal models in which to investigate cytoskeletal events during human reproduction. INTRODUCTION The deliberate creation of human zygotes and embryos for scientific research is unlikely to be ever completely free of ethical, political, financial, and practical complexities. Nevertheless, a detailed understanding of the cellular and molecular events during human fertilization is essential for the field of developmental biology as well as for clinical applications including infertility treatment, contraception, and the avoidance of developmental abnormalities. The motility and cytoskeletal rearrangements essential for successful union of the sperm and egg nuclei during fertilization [1] are poorly understood in primates, including humans. The vast bulk of the knowledge in mammals rests on murine models, and it is not clear whether or not this information can be extrapolated to fertilization in nonhuman and human primates [2]. In order to address this key step in primate fertilization and to avoid ethical, moral, and practical difficulties posed by work with fertilized human zygotes, this study explored microtubule organization during

MATERIALS AND METHODS Oocyte and Sperm Collection The protocols for hyperstimulation of fertile female rhesus monkeys and for collection of oocytes from ovaries obtained during scheduled autopsies, as well as for the collection and capacitation of spermatozoa, have been developed by Bavister et al. [7], Boatman [8], Wolf et al. [9], Morgan et al. [10], and VandeVoort and Tarantal [11], and were used in this research. Briefly, hyperstimulation of females (3-12 yr old) was accomplished by twice-daily injections with 30 IU Metrodin (Serono Laboratories, Randolph, MA) for 6 days followed by 3 days of 30 IU Pergonal (twice-daily; Serono Laboratories). Twelve hours after the last injection of Pergonal, 1000 IU of hCG was administered, and laparoscopic aspiration was performed 27 h later according to the methods of Bavister et al. [7]. Laparoscopic-derived oocytes were cultured in Tyrode's albumin lactate pyruvate (TALP) culture medium containing 20% heat-inactivated fetal calf serum (FCS) for 7-24 h so that they would achieve metaphase-II arrest before undergoing in vitro fertilization (maturation rate: 678 of 733 [92.5%]). Immature oocytes collected from ovaries of au-

Accepted March 29, 1996. Received October 23, 1995. 'We are pleased to acknowledge the support of the National Institutes of Health. WRPRC Publication No. 36-010. 2Correspondence: Dr. Gerald Schatten, 1117 West Johnson St., Madison, WI 53706. FAX: (608) 262-7319; e-mail: [email protected] 3 Current address: Department of Obstetrics and Gynecology, Triservice General Hospital, National Defense Medical University, No. 8, 3rd Section, Ting-Chow, Taipei, Taiwan 100.

260

MICROTUBULE AND CHROMATIN DYNAMICS DURING RHESUS IVF

261

topsied animals were matured in vitro for 48 h in CMRL medium (Gibco BRL, Gaithersburg, MD), containing 20% FCS, 10 jig/ml FSH-P, and 5 IU/ml hCG, to obtain metaphase-II-arrested oocytes as described by Morgan [10]. An in vitro maturation rate of 33.8% (364 of 1076) was achieved from necropsy-derived oocytes. Semen was collected by penile electroejaculation from rhesus males of proven fertility, and sperm capacitation was accomplished in TALP culture medium containing caffeine and dibutyryl cAMP (1 mM each), in accordance with the methods of Bavister et al. [7].

CA). DNA was detected by means of either 5 g/ml 4',6diamidino-2-phenylindole (DAPI; Sigma) or 5 jig/ml propidium iodide (Molecular Probes, Eugene, OR), which were added to the penultimate rinse. Coverslips were mounted in an anti-fade medium (SlowFade; Molecular Probes) to retard photobleaching. The methods of data collection and analysis, by both conventional immunofluorescence and laser-scanning confocal microscopy, have been described in a previous publication [14].

In Vitro Fertilization and Zygote Culture

Parthenogenetic activation of mature rhesus oocytes was accomplished according to the methods employed for bovine oocyte activation [14]. Oocytes were treated with 5 jM ionomycin for 5 min in TALP medium, rinsed in culture medium containing 20% FCS, and then incubated for 4 h in 1.9 mM 6-dimethylaminopurine (DMAP; Sigma), a nonspecific kinase inhibitor, in TALP At the end of the DMAP-activation period, oocytes were transferred into TALP, without DMAP, until the desired fixation time-point. Artificial activation rates of rhesus oocytes, by ionomycin/ DMAP treatment, were consistently high (76.9 + 7.2% activation rates; n = 13 trials). The formation of a single female pronucleus was observed by 3-4 h posttreatment, and >80% of these oocytes entered into first mitosis by 15-16 h postactivation.

Mature oocytes were fertilized in vitro with the use of capacitated, stimulated sperm diluted to a final concentration of 2 X 104 sperm/mi in TALP culture medium according to the methods described by Bavister et al. [7]. Successful sperm penetration was scored after removal of the oocyte from the sperm suspension. Oocytes were examined for second polar body extrusion and for the presence of pronuclei in the cytoplasm of living oocytes at 5-12 h postinsemination (p.i.) with use of differential interference contrast (DIC) optics. Oocytes were either processed for immunocytochemical detection of microtubules and DNA (see below) or transferred to freshly prepared TALP medium for additional culture to first mitosis (18-24 h p.i.). Culture of embryos beyond the 2-cell stage was performed with use of CMRL-1066 medium (Gibco BRL) containing 20% FCS [8]. Deliberate polyspermic inseminations were achieved with use of an elevated sperm concentration of 2-10 x 106 sperm/mi. The in vitro fertilization rate achieved in our lab using necropsy-derived oocytes did not differ significantly from the rate achieved with mature oocytes from hyperstimulated monkeys (74.2 7.0% [11 trials] vs. 81.9 8.8% [15 trials], respectively), nor did these two populations of oocytes differ in their microtubule and DNA configurations as determined by immunostaining. Therefore, the data from these two groups were combined in this study for all analysis purposes. Microtubule and DNA Imaging Microtubules and DNA were detected, in formaldehydefixed oocytes and embryos, by indirect immunofluorescent labeling techniques [12]. Before fixation, cumulus cells (if present) were removed using a combination of 2 mg/ml hyaluronidase and manual pipetting of the oocyte or zygote through a small-bore glass pipette. The zonae pellucidae were removed from oocytes or zygotes by a brief treatment (2-7 min) with 0.5% Pronase (Calbiochem, San Diego, CA). After a 30-min recovery at 37°C, zona-free oocytes were attached to polylysine-coated coverslips and fixed overnight in 2% formaldehyde/0.05% picric acid (pH 7.5) in Tyrode's lactate Hepes without added calcium or BSA protein. Fixed oocytes and embryos were then permeabilized with 0.1 M PBS containing 1% Triton X-100 detergent for 40 min, and next treated for 20 min in 0.1 M PBS containing 150 mM glycine and 3 mg/ml BSA (fraction V; Sigma, St. Louis, MO) in order to reduce free aldehydes. For microtubule observation, the oocytes were incubated for 1 h in a mouse monoclonal antibody to 3-tubulin (E-7) [13]. After the incubation, they were rinsed with PBS + 0.1% Triton X-100 detergent for 20 min. The primary antibody was detected by means of a fluorescein-labeled goat anti-mouse IgG antibody (1:40; Zymed, San Francisco,

Parthenogenetic Activation

Nocodazole and Colcemid Treatment Stock solutions of the microtubule-assembly inhibitors nocodazole and colcemid were prepared in anhydrous dimethylsulfoxide and diluted infTALP medium to final concentrations of 50 jiM and 20 jiM, respectively. Control oocytes were treated with 0.1% dimethylsulfoxide in TALP. Microtubule-assembly inhibitors were applied to inseminated oocytes (3-5 h p.i.) after completion of second meiosis, as determined by second polar body formation, but before pronuclear formation. Control and drug-treated oocytes were fixed at a time-point when the pronuclei were expected to have been fully decondensed and to be closely apposed to one another (approximately 14-16 h p.i.). Measurement of Intracellular Calcium Cytoplasmic free calcium levels were monitored with use of the ratiometric, fluorescence calcium probe fura-2 AM (Molecular Probes). Dye loading was accomplished by diluting a stock solution of fura-2 AM with TALP medium (final concentration, 10 jiM) and incubating the oocytes for 30 min at 37°C. Before imaging, 10-20 min were allowed for the de-esterification of the dye in TALP medium. Imaging of intracellular calcium changes was accomplished as described by Tombes et al. [15]. RESULTS Microtubule Patterns in Unfertilized and Fertilized Oocytes Microtubules were detected only in the barrel-shaped, anastral, meiotic spindle of the unfertilized rhesus monkey oocyte, which was arrested at second meiotic metaphase, as has also been reported by Johnson et al. [16]. The meiotic spindle was slightly focused at both poles and was oriented radially to the cell cortex (Fig. 1A, green; spindle length SD, 16.9 2.2 m). Shortly after sperm incorporation (Fig. 1B; blue, M), microtubules formed a small

262

WU ET AL. sperm aster, located at the junction between the head and tail of the sperm (Fig. B; green, M), as completion of meiosis was initiated (Fig. B; F). The sperm aster enlarged within the cytoplasm as the sperm chromatin decondensed (Fig. 1C; M). The meiotic midbody marks the location of the female pronucleus and the second polar body (Fig. 1C; F). By 6 h p.i., the sperm astral microtubules filled the zygote (Fig. 1D; green). At this time, the male and female pronuclei were tightly apposed near the cortex (Fig. 1D; blue). By the end of first interphase (18 h p.i.), the sperm aster developed a bipolar microtubule array (Fig. E; green) radiating from the apposed, eccentrically positioned pronuclei (Fig. E; blue). At the onset of prophase (18-20 h p.i.), the interphase microtubule array was replaced by a dense monaster of microtubules (Fig. 2A; green) as the chromosomes condensed and intermixed (inset, Fig. 2A). The incorporated sperm tail could often be distinguished in association with the developing spindle (Fig. 2A; arrow). By late prophase, a bipolar mitotic spindle assembled (Fig, 2B; green), while the chromosomes aligned at the equator (Fig. 2B; blue). The first mitotic apparatus was an anastral, barrel-shaped structure eccentrically positioned within the cytoplasm (Fig. 2C; average spindle length ± SD, 22.8 +_4.2 plm). Often, one or two small asters were associated with the sperm axoneme at one of the spindle poles (Fig. 2C; white bar marks sperm axoneme). By anaphase, the spindle pole asters enlarged (Fig. 2D; green) and the chromosomes separated, moving toward their respective poles (Fig. 2D; blue). During telophase, an increase in the assembly of microtubules between the spindle apparatus and the cortex occured (Fig. 2E; green). These microtubules are thought to participate in moving the spindle toward the cell center. After cleavage, an interphase-like pattern of microtubules occurred in both daughter blastomeres (Fig. 2F; green), and the nuclei reformed at the center of each cell (Fig. 2F; blue). Microtubule Patterns Formed during Polyspermic Inseminations and Parthenogenetic Activation

FIG. 1. Microtubule patterns in unfertilized and in vitro-fertilized rhesus monkey oocytes. A barrel-shaped, anastral, meiotic spindle (A: green), oriented radially to the cell cortex, is the only microtubule structure present in the mature, unfertilized rhesus oocyte. The chromosomes are aligned on the metaphase plate (A: blue). Between 3 and 5 h p.i., a small sperm aster (B: green, M) forms at the base of the sperm head (B: blue, M) shortly after sperm penetration. The female chromosomes begin to separate on the second meiotic spindle (B: blue, F). As development proceeds, the sperm aster microtubules lengthen in the cytoplasm (C: green, M) as the sperm DNA decondenses (C: blue, M). The meiotic midbody marks the site of the female chromosomes and the second polar body (C: blue, F). By 6-8 h p.i., the sperm aster enlarges to completely fill the zygote (D: green). Note that the adjacent male and female pronuclei are

In polyspermic zygotes, immunofluorescent microscopy demonstrated that each incorporated sperm organized a microtubule aster at the base of the sperm head (Fig. 3, A-D), reinforcing the observation that the sperm contributes the centrosome in this primate. In Figure 3C, a large sperm aster is seen to form at the base of a fully developed male pronucleus (Fig. 3C; bottom image), while a second sperm aster is organized by a condensed sperm head in another region of the cytoplasm (Fig. 3C; top image). As development proceeded, the sperm nuclei decondensed in the cytoplasm and the microtubule asters expanded (Fig. 3D). Interestingly, pronuclear apposition was not necessarily completed in polyspermic zygotes by the end of the first interphase, and often two or more paternal spindles would form at different cortical sites within these cells (Fig. 3, E-G). At anaphase (Fig. 3E), monastral arrays of microtubules formed at each condensing set of chromosomes (Fig. 3E, insets). At first metaphase in a dispermic zygote, as 4

eccentrically positioned at the cortex (D: blue; Pb: second polar body). By 18 h p.i., the apposed pronuclei remain decentralized in the cytoplasm, and the sperm aster develops a bipolar appearance emanating between the two pronuclei (E: arrowheads). All images are double-labeled for microtubules (green) and DNA (blue). M = male pronuclei; F = female pronuclei; Pb = second polar body. Bars = 20 Itm.


263

seen in Figure 3F, two separate spindles had formed within the common cytoplasm. Digital superimposition of confocal with DIC images demonstrates the presence of a sperm tail at one pole of each of the supernumerary spindles (Fig. 3F; arrowheads). A dispermic zygote is shown at anaphase (Fig. 3G). Astral development can be seen at a single pole of each spindle. Concurrently, the chromosomes began their migration to separate poles (Fig. 3G). Dispermic zygotes often cleaved into three or four equal-sized blastomeres at the time of the first division, and an extensive, interphase array of cytoplasmic microtubules formed within each daughter blastomere (Fig. 3H). The morphology of the DNA and microtubules was analyzed during the resumption of meiosis and the first cell cycle in ionomycin/DMAP parthenogenetically activated oocytes (Fig. 4). Polymerized, linear bundles of microtubules were visible in the cytoplasm 1 h after ionomycin/ DMAP treatment. In addition, microtubules were observed in the second meiotic spindle (Fig. 4A). Between 3 and 6 h after ionomycin/DMAP treatment, cytoplasmic microtubules grew to fill the entire cytoplasm in a disorganized fashion, and a single female pronucleus was found at the cortex (Fig. 4, B and C). In all instances, however, extrusion of the second polar body did not occur. By 16 h postactivation, ionomycin/DMAP-treated oocytes entered first mitosis, and astral microtubules organized around the condensing female chromatin after nuclear envelope breakdown had occurred (Fig. 4, D and E). By metaphase, the activated oocyte had an anastral, bipolar mitotic spindle at the cortex region, with well-aligned chromosomes at the spindle equator (Fig. 4F; spindle length + SD, 27.5 7.4 pIm). No small, astral microtubules were observed at the spindle poles in any of the mitotic parthenotes (compare Fig. 2C with Fig. 4F). During anaphase, the spindle apparatus lengthened and the chromosomes separated in a disorganized fashion (Fig. 4G)-actions that may lead to the partial or fragmented chromosome separation frequently seen in parthenotes. By cytokinesis, rhesus oocytes demonstrated equal cleavage, and a normal interphase cytoplasmic microtubule pattern was organized in each daughter blastomere (Fig. 4H). Intracellular Calcium Changes Elicited by lonomycin, 6-DMAP, and Sperm Penetration

FIG. 2. First mitosis in fertilized rhesus zygotes. At prophase, the interphase microtubule array is replaced by a dense monaster of microtubules (A: green) as the chromosomes condense within the cytoplasm (A: blue, inset). The incorporated sperm axoneme can be detected in the cytoplasm running from the cortex to the eccentrically positioned prophase array (A: green, arrow). The additional small asters are associated with the sperm axoneme. By prometaphase (18-20 h p.i.), a developing bipolar spindle emerges (B: green) from the monastral microtubule array as the two sets of chromosomes intermix on the forming spindle equator (B: blue). Mitotic metaphase is marked by an eccentrically positioned, barrel-shaped spindle, often with 1-2 small asters associated with a single pole (C: green). The sperm axoneme is always found associated with the spindle pole containing the asters (C: bar marks position of tail seen by DIC). At anaphase, microtubule assembly occurs at both mitotic poles (D: green)

We used the fluorescent, membrane-permeable calcium indicator fura-2 AM to monitor the intracellular calcium dynamics in rhesus oocytes during artificial activation and fertilization. The baseline calcium level in unactivated oocytes (approximately 150 nM) was monitored over a 4-h time period. The baseline remained constant for the duration of the recordings, demonstrating that organelle sequestration and photobleaching were not significant problems (results not shown). Unfertilized rhesus oocytes were exposed to 5 uM ionomycin (Fig. 5A; top graph) and monitored for changes in intracellular calcium levels. Ionomycin-treated oocytes

as the chromosomes separate (D: blue). During telophase (E: blue), the polar microtubules grow more extensive, and a prominent array of assembled tubulin appears between the spindle and the adjacent cortical region, possibly mediating the spindle displacement from the cortex and toward the cell center (E: green). First cleavage (F: blue) results in an equal cleavage as an interphase pattern of microtubules reassembles in both daughter blastomeres (F: green). All images are double-labeled for microtubules (green) and DNA (blue). Bars = 20 g±m.

264

WU ET AL.

FIG. 3. Sperm asters and first mitosis in polyspermic zygotes. In dispermic zygotes, each entering sperm organizes a microtubule aster from the base of the sperm head (A-C: green). Images A and B (green) are taken from the same zygote, showing a condensed sperm head with a small microtubule aster at one focal plane (A: green) and the corresponding prophase microtubule array, surrounding the adjacent male and female chromosomes condensing within the cytoplasm, at a separate focal plane (B: green). In C, a dispermic oocyte is shown at two different focal planes, contrasting the tightly focused sperm aster surrounding a decondensing sperm head (C: upper image) and a separate microtubule array emanating from a fully formed male pronucleus (C: lower image; inset: male pronucleus). During trispermic insemination, each incorporated sperm has fully decondensed at a cortical site within the cytoplasm (D: blue), and each pronucleus organizes an elaborate microtubule aster (D: green). The female pronucleus is located out of the focal plane in this image. In a dispermic prophase zygote, two separate spindles develop in the cytoplasm around each set of paternal chromosomes (E: blue, insets); the microtubules are organized in radial arrays about the condensing chromosomes (E: green). At first metaphase in a dispermic zygote, two separate, anastral, barrel-shaped spindles form within the cytoplasm (F). Digital superimposition of confocal and epifluorescence with a DIC image demonstrates the presence of a sperm tail at one pole of each of the supernumerary spindles (F: arrow heads). At anaphase, in a dispermic zygote, asters develop from the poles of each spindle as the chromosomes separate (G). Polyspermic zygotes often cleave into four equalsized blastomeres at the time of first division and reorganize an interphase array of cytoplasmic microtubules in each cell (H). All images are double-labeled for microtubules (green) and DNA (blue). F) Superimposed DIC image with fluorescent microscopy. Bars = 20 Lm.

(n = 16) exhibited an immediate increase (< 1 min) in the cytosolic calcium level, reaching a mean peak calcium value of 1650 + 150 nM (Fig. 5A; top graph). Immunofluorescent examination of these oocytes at 8 h after ionophore treatment demonstrated that all remained arrested at metaphase of second meiosis, although cytoplasmic bundles of microtubules were evident in the rhesus (Fig. 5B; inset, DNA). A representative time course of calcium dynamics before

and after the addition of DMAP, a nonspecific protein kinase inhibitor, is shown in a rhesus oocyte in Figure 5A (middle graph). The calcium level remained near the baseline for 10 min, at which time a slow rise in this value was observed. This rise in calcium continued until a new plateau of approximately 500 nM was reached. Normal activation of these rhesus oocytes was not observed after treatment with DMAP alone (Fig. 5C; n = 9; inset, DNA). The intracellular calcium dynamics in a representative


265

FIG. 4. Microtubule organization in parthenogenetically activated oocytes. A-C) Oocyte activation in 5 FLM ionomycin + 1.9 mM 6-DMAP for 4 h before processing for anti-tubulin and DNA staining. Artificial activation results in microtubule organization in the cytoplasm, initially as small linear bundles of microtubules (A: green) that progress into a large disorganized array (B-C: green). A single female pronucleus forms at the site of the meiotic spindle, without second polar body formation (A-C: blue). At the onset of first mitosis, astral microtubules (D and E: green) are organized surrounding the condensing female chromatin (D and E: blue). The first mitotic metaphase spindle formed at the cortex is an anastral, bipolar structure (F: green) with aligned chromosomes at the equator (F: blue). During anaphase, the spindle apparatus (G: green) lengthens as chromosomes separate in a highly disorganized fashion (G: blue). Artificially activated oocytes divide into two cells at cytokinesis as a normal interphase cytoplasmic microtubule pattern is reorganized in the daughter blastomeres (H: green; DNA: blue). All images are double-labeled for microtubules (green) and DNA (blue). Bars = 20 Ipm.

rhesus oocyte, activated by the ionophore/DMAP protocol, are shown in Figure 5A (lower graph). The initial calcium transient was followed by a persistent rise in the baseline calcium level, where it reached a new plateau level. Efficient activation of this oocyte was demonstrated by the organization of interphase microtubules and the formation of a single female pronucleus by 6 h after ionophore/DMAP treatment (Fig. 5D; n = 12; inset, DNA). These data suggest that both an increase in intracellular calcium and kinase inhibition are required for the activation of meiotically arrested rhesus oocytes. Intracellular calcium dynamics were also monitored during the in vitro insemination of mature rhesus oocytes, with 26 of 29 inseminated oocytes (89.6%) exhibiting calcium transients (Fig. 6). Figure 6A (top graph) shows a monospermically inseminated rhesus oocyte. Intracellular calcium rapidly increased to a peak of 1650 nM; this transient had a duration of about 4 min. This initial

calcium transient was followed by oscillations in the calcium level, with a frequency of one every 6 min (Fig. 6A; top graph). Microtubule and DNA imaging of this oocyte (2.5 h p.i.) demonstrated that a single sperm had been incorporated into the cytoplasm and had organized a small sperm aster at the base of the head (Fig. 6B, arrow; inset, DNA). A similar, initial calcium increase and repetitive oscillations were found to occur in oocytes penetrated by multiple sperm, except that the frequency of the successive transients was on the order of one every 1.5-2 min. Figure 6A (bottom graph) shows a representative time course of the calcium dynamics in a polyspermically inseminated oocyte. Multiple sperm head penetration was shown to have occurred, with a small microtubule array at the base of each of the sperm heads, by 2.5 h p.i. (Fig. 6C, arrows; inset, DNA). In both monospermic and polyspermic oocytes, successive calcium oscillations showed a characteristic de-

266

WU ET AL.

A .

mono\

eUUU

Rhesus 1000

0

40

2000*

Rhesus 10001

-

6 -~DMAP

Q)

E u

"o

40

0

80

2000

U

Rhesus

4-.

1000

6-DMAP

40

0 Ivi I

I

I

I

FIG. 5. Imaged calcium transients associated with ionomycin/DMAP artificial activation in primate oocytes. A) Calcium transients associated with ionomycin alone (top graph), 6-DMAP alone (middle graph), or sequential treatment with ionomycin/6-DMAP (lower graph) in rhesus oocytes. Intracellular calcium levels rise transiently in oocytes continually incubated in 10 M ionomycin. No secondary calcium releases were detected in the ionophore-treated oocytes. Despite the global intracellular calcium release, ionomycin-treated oocytes remain arrested at metaphase even after 8 h of culture in vitro, although wispy cytoplasmic microtubules are apparent (B: inset, DNA). DMAP treatment by itself induces a slow rise in the basal level of intracellular calcium (A: middle graph) but does not activate the rhesus oocyte. Often, the second meiotic spindle loses its bipolar organization after exposure to this kinase inhibitor (C: inset, DNA). lonomycin + DMAP triggers an intracellular calcium peak in rhesus oocytes, combined with a rise in the baseline calcium resting value (A: lower graph). By 6 h postactivation, a random array of interphase microtubules and a single female pronucleus form within the rhesus oocyte cytoplasm (D: inset, DNA). A) Graph of estimated intracellular calcium levels (nm; y-axis) vs. time (min; x-axis). B-D) Anti-tubulin images; insets, DAPI (DNA). Bars = 20 Rm.

crease in the peak level. This may suggest that calcium released from an internal calcium pool was depleted over time. Figure 7 depicts the characteristics of the calcium transients observed in the two rhesus oocytes shown in Figure 6. Calcium release first occurred in the cortex, then rapidly proprogated around the entire perimeter of the oocyte, and finally spread in a wave-like fashion across the central cytoplasm. A similar observation has been reported in human oocytes [6].

Effects of Microtubule Inhibitors on Unfertilized Oocytes and Pronuclear Apposition Following Sperm Incorporation A 2-h treatment with colcemid (50 FxM) or nocodazole (20 lpM) inhibited microtubule assembly and disrupted the meiotic spindle in > 80% of the mature, unfertilized rhesus oocytes (Fig. 8A). Disruption of the metaphase spindle did not cause the chromosomes to disperse along the cortex (Fig. 8B). This is dissimilar to the effects seen in mice [17,


267

A ---ZUOO 1000 _

75 IV

0

E

500

1000

1500

1000

1500

U. U

.:

----

/UUU

E 4.

1000

1000

0

500

Time (sec)

FIG. 6. Images of calcium dynamics and microtubule patterns observed after rhesus in vitro fertilization. A) Graphic analysis of the calcium transients initiated after monospermic (top graph) or polyspermic (bottom graph) inseminations. Traces of the calcium level in a single oocyte demonstrate that at the moment of sperm egg fusion, a large calcium transient is initiated within the cytoplasm for nearly 4.5 min before the ionic level returns to near baseline values. Periodic oscillations in the calcium level occur over the next 25 min, each with a decreasing amplitude in the amount of calcium released. When this oocyte was processed for anti-tubulin and DNA at 2.5 h p.i., it was found that a single sperm had penetrated the oocyte, organizing a small aster at the base of the decondensing sperm head (B: arrowhead; inset, DNA). Bottom graph: Calcium transient characteristics during polyspermic fertilization showing numerous calcium oscillations induced by the incorporating sperm. C) Microtubule and DNA imaging of this trispermic oocyte, demonstrating that each incorporated sperm organizes a small aster at the base of the head (C: arrows; inset, DNA). A) Graph of estimated intracellular calcium level (nm; y-axis) vs. time (sec; x-axis) for two oocytes. B-C) Anti-tubulin images; insets, DAPI (DNA). Bars = 20 pim.

18]. After pronuclear formation, both nocodazole (11 of 14) and colcemid (11 of 12) prevented pronuclear migration (Fig. 8C) but not the onset of chromosome condensation in zygotes fixed at prophase of first mitosis (18 h p.i.; Fig. 8D). DISCUSSION While it is essential to understand the events during human fertilization, investigations using inseminated human oocytes, zygotes, and embryos raise numerous concerns. These include the question of the normalcy of the scientific results, since the gametes are nearly always obtained as excess, discarded material from infertility clinics. Could the problem that leads the infertile patient to seek treatment manifest itself in the sperm and the oocytes? Furthermore, since the gametes are selected and the best, viable material is used for clinical therapy, questions remain regarding the reliability of results obtained with use of the discarded specimens. To address these concerns, and to determine whether the rhesus, as a nonhuman primate, is a reliable model system for examining the events during human fertilization, this investigation was undertaken. Microtubule and DNA patterns in both unfertilized and fertilized rhesus

oocytes were characterized. In addition, calcium dynamics during in vitro insemination and the role of calcium in artificial activation were examined. These results on rhesus fertilization, when compared with the recent characterization of microtubule dynamics during human fertilization [19], demonstrate that nearly identical events occur in the two primates. There are, however, three minor differences between human and rhesus fertilization. First, while the only microtubules found in unfertilized oocytes from either primate are those constituting the meiotic spindle, the spindle in human oocytes is asymmetric. Also, human oocytes establish bipolarity slightly earlier during first interphase (prophase vs. prometaphase in rhesus). Finally, human oocytes display a more central positioning of the mitotic spindle at first mitosis. The demonstration that microtubule inhibitors prevent pronuclear apposition indicates the essential role that the sperm aster plays during primate fertilization. By combining the techniques of single-cell calcium imaging and anti-tubulin immunocytochemistry, it was possible to determine when a particular oocyte was activated and also whether it was fertilized monospermically. With use of the calcium transient as a starting point, well-devel-

268

WU ET AL.

C >

o, r m0

u3 U

E

CL

--

0n Ec

C

CD

>

'T

U

&°

>- 0

0

8

L. oC)

o o

u

t0 -°

Co

.f

C