|
|
||||||||
Regular Article |
a Department of Obstetrics and Gynecology, Women and Infants Hospital, Brown University, Providence, Rhode Island 02905
b Laboratory for Reproductive Medicine, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
| ABSTRACT |
|---|
|
|
|---|
apoptosis, developmental biology, early development, embryo, ovum
| INTRODUCTION |
|---|
|
|
|---|
In mammalian species, parthenogenetic development has been achieved by a variety of methods that induce Ca2+ transients [20, 21]. Although single Ca2+ elevations induced by ethanol, calcium ionophore, or an electrical pulse are not sufficient to promote development, multiple Ca2+ elevations, e.g., after strontium (Sr2+) treatment, significantly enhance development of oocytes [2226] by mimicking sperm-induced Ca2+ oscillations [22, 23, 2729]. Sr2+ itself does not cause chromosome abnormalities in activated oocytes [30]. Sr2+ used to activate cytoplasts during nuclear transfer produced cloned mice, demonstrating the suitability of Sr2+ in promoting full-term development [31, 32]. However, if oocytes activated by Sr2+ are allowed to extrude a second polar body, an aneuploid, specifically haploid, parthenote is induced. Exposure of activated oocytes to cytoskeletal inhibitors to inhibit polar body extrusion produce diploid parthenotes to control for the effects of Sr2+ and parthenogenetic activation on development and apoptosis.
Parthenogenetically developed embryos exhibit delayed development, reduced total cell number, and fewer cells in the inner cell mass of blastocysts compared with fertilized embryos [30, 33]. The developmental potential of human parthenogenetic embryos is also reduced [34]. Moreover, haploid parthenogenetic embryos are developmentally delayed compared with diploid parthenogenetic embryos in the mouse, pig, and cow [20, 3539]. These developmental defects have been hypothesized to result from insufficient parthenogenetic activation, suboptimal in vitro culture conditions, or genomic imprinting. To date, the possibility that apoptosis may underlie this developmental abnormality has not been examined.
To investigate whether haploidy initiates apoptosis during preimplantation development, we took advantage of the effectiveness of Sr2+ for activating mouse oocytes to produce parthenogenetic haploids and also generated diploid embryos (parthenote controls) and IVF control embryos and compared the rates of development and frequency of apoptosis in these groups. The results of these experiments indicate that parthenogenetic activation and parthenogenesis themselves do not cause apoptosis, but haploidy increases the incidence of apoptosis in preimplantation embryos.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Unless otherwise specified, all reagents were obtained from Sigma Chemical Co. (St. Louis, MO). B6C3F1 female mice at 610 wk of age (Charles River Laboratory, Raleigh, NC) were superovulated with 5 IU eCG (Calbiochem, La Jolla, CA) followed by 5 IU hCG 4648 h later. Oocytes enclosed in cumulus masses were collected from oviduct ampullae 14 h after hCG injection. Cumulus cells were removed by pipetting after brief incubation in 0.03% hyaluronidase prepared in potassium simplex optimized medium (KSOM) containing 14 mM Hepes and 4 mM sodium bicarbonate (HKSOM) [40, 41]. Oocytes were then washed and incubated in 50-µl droplets of preequilibrated KSOM [40] supplemented with nonessential amino acids and 2.5 mM Hepes [42] and covered with mineral oil pending further treatments. Oocyte culture and all subsequent cultures were carried out at 37°C in a humidified atmosphere of 7% CO2 in air. All manipulations were performed at 3637°C on heated stages, in chambers, or in incubators.
IVF and Parthenogenetic Activation of Oocytes
Sperm were expelled from the cauda epididymis of male B6C3F1 mice into 400 µl Toyoda Yokoyama Hoshi (TYH) medium containing 0.4% BSA and incubated under mineral oil for 12 h at 37°C to capacitate [43]. A sperm suspension at a concentration of 67 x 105 sperm/ml was used to inseminate oocytes in a 400-µl drop of TYH medium supplemented with 0.4% BSA under mineral oil. After coincubation with sperm for 6 h, the inseminated oocytes were washed and cultured in 50-µl droplets of KSOM under mineral oil. Cleavage and embryo development were examined every 24 h.
B6C3F1 mouse oocytes were reliably activated by a 3- to 4-h treatment with 10 mM SrCl2 (Sr2+) [26, 31, 44] prepared in Ca2+-free KSOM. Sr2+ failed to induce changes in Ca2+ in oocytes incubated in the Ca2+-containing medium. Cytochalasin D (CCD), an inhibitor of actin filament polymerization, was used to suppress the second polar body extrusion to generate diploid parthenotes. Activated oocytes with 2 pronuclei, visualized under a differential interference contrast microscope, were defined as diploid parthenotes, and oocytes with 1 pronucleus were defined as haploid parthenotes [30, 33, 35]. Activated oocytes were washed extensively and cultured in vitro under the same conditions as were IVF zygotes.
Chromosomal Analysis
Twenty to 30 oocytes or embryos from 2 replicates of each treatment group were prepared for chromosome complement analysis as described previously [45]. One-cell parthenotes and fertilized embryos were incubated with 2 µg/ml nocodazole overnight and arrested at metaphase. Embryos at later stages were arrested at mitosis by 45 h of incubation with nocodazole. Oocytes or embryos were transferred to 1% sodium citrate for 20 min and fixed in methanol:glacial acetic acid (3:1) on a glass slide. The chromosome spreads were air dried, stained with 50 µg/ml propidium iodide (Molecular Probes, Eugene, OR) in Dulbecco PBS for 30 min, washed, and mounted in Vectashield mounting medium (Vector Laboratories, Burlingame, CA) for observation with a rhodamine filter set using a Zeiss Axiovert 100TV inverted fluorescence microscope (Thornwood, NY).
Detection of Apoptosis by TUNEL Assay
Embryos at different developmental stages were fixed in 3.7% paraformaldehyde prepared in Dulbecco PBS supplemented with 0.1% polyvinylpyrrolidone. Nuclear DNA fragmentation in embryos was detected by the TUNEL assay [12, 1417, 41], using an in situ cell death detection kit (Boehringer Mannheim, Indianapolis, IN) according to the manufacturer's instructions, and the cell nuclei were counterstained with 50 µg/ml propidium iodide. Embryos were washed, mounted onto a slide under a coverslip in the Vectashield mounting medium, and sealed with nail polish. The number of total nuclei and nuclear morphology were evaluated using a rhodamine filter, and the DNA fragmentation was assessed using a fluorescein filter with an inverted Zeiss microscope equipped with epifluorescent optics and a Zeiss LSM 510 laser scanning confocal microscope. Three-dimensional images were reconstructed based on Z-sections from the confocal microscope and analyzed by Zeiss LSM image software. The total number of cells reported includes the mitotic cells. The percentage of apoptotic cells per embryo is expressed as apoptotic nuclei/total number of nuclei x 100.
Measurement of Cytosolic Ca2+
Metaphase II oocytes were incubated with 2.5 µM Fura-2/AM (Molecular Probes) in KSOM for 30 min at 37°C. Fura-loaded oocytes were washed extensively with KSOM and transferred to poly-D-lysine-coated glass-bottom dishes (MatTek Corp., Ashland, MA). Ca2+ imaging was conducted at excitations of 334/380 nm and measured at an emission of 520 nm using an Attofluor RatioVision digital fluorescent imaging system (Atto Instruments, Rockville, MD). The fluorescence signal is displayed as the ratio of fluorescence intensity for the 334/380-nm excitation wavelengths after subtraction of background. In each experiment, data acquired every 10 sec were analyzed using available Attofluor RatioVision imaging software. The fluorescence ratios were simultaneously converted to the Ca2+ concentration using the equation [46] included in the Attofluor imaging software. The standard curve was obtained by imaging fluorescence intensity of the low and high Ca2+ standard solutions mixed with Fura-2 penta K+ salt.
Statistical Analysis
Each experiment was repeated at least 3 times. Percentages were transformed using an arcsine transformation. Comparison of treatment means was carried out by ANOVA and Fisher PLSD using StatView software (SAS Institute, Cary, NC). Significant difference was defined as P < 0.05.
| RESULTS |
|---|
|
|
|---|
Mouse oocytes activated with Sr2+ in Ca2+-free medium exhibited cytosolic Ca2+ oscillations (Fig. 1A), with a long-lasting increase of the first transient followed by subsequent oscillations at an interval of 13.1 ± 3.8 min (n = 18, range 1018 min). Oocytes treated with Sr2+ in medium containing Ca2+ did not show changes in cytosolic Ca2+ (Fig. 1B). These results suggest that Sr2+ induces Ca2+ release from intracellular stores. Mouse oocytes activated by Sr2+ in combination with 2 µg/ml CCD in Ca2+-free medium exhibited cytosolic Ca2+ oscillations (Fig. 1C) similar to those induced by Sr2+ alone, indicating that CCD does not change Ca2+ profiles. During IVF, sperm triggered typical Ca2+ oscillations (Fig. 1D) at an interval ranging from 8 to 20 min. The first Ca2+ transient lasted longer than subsequent transients in both Sr2+ and sperm-activated oocytes. The peak duration (34 min) of subsequent Ca2+ transients triggered by Sr2+ was longer than that of sperm (3060 sec).
|
B6C3F1 mouse oocytes (Fig. 2A) activated with Sr2+ in Ca2+-free medium formed 1 pronucleus and extruded a second polar body (Fig. 2B), thereby producing haploid parthenotes. Oocytes treated with Sr2+ in the presence of Ca2+ failed to form a pronucleus and did not extrude a second polar body. Oocytes activated by Sr2+ plus CCD formed 2 pronuclei and developed as diploid parthenotes (Fig. 2C) [33, 38, 47], and the chromosome complements were further confirmed by karyotype analysis. Parthenotes with 1 pronucleus were haploid, and parthenotes with 2 pronuclei were diploid (Fig. 2, DF). Fertilized zygotes exhibited a diploid chromosome constitution (Fig. 2G).
|
Rates of Cleavage and Preimplantation Development of Parthenotes and IVF Embryos
Twenty-four hours after activation or IVF, the percentage of embryos that cleaved to the 2-cell stage did not differ between diploid parthenotes (99%) and IVF embryos (95%) but was significantly lower in haploid parthenotes (90%) (Table 1). The first mitosis was delayed 34 h for haploid parthenotes compared with diploid parthenotes (unpublished observations). The rate of cleavage by unactivated oocytes following Sr2+ treatment in KSOM containing Ca2+ (4%) was as low as that of nontreated control oocytes (2%) cultured in KSOM. At 48 h after activation or fertilization, diploid parthenotes and IVF embryos had developed to 8- to 16-cell compact morulae, whereas haploid parthenotes cleaved to only the 4- to 8-cell stage and did not show compaction (data not shown).
|
At 72 h after activation or fertilization, development of haploid parthenotes was significantly delayed compared with that of diploid parthenotes and IVF-produced embryos (Table 1 and Fig. 3). Most (78%) haploid parthenotes were at the morula stage (Fig. 3A), and only 5% had reached at the blastocyst stage, in contrast to a significantly (P < 0.01) higher rate of development to blastocysts of diploid parthenotes (87%, Fig. 3B), a rate comparable to that of IVF-produced embryos (88%, Fig. 3C). Nonactivated oocytes cultured in vitro remained a single cell and exhibited shrinkage or some degree of cytofragmentation. Karyotype analysis confirmed haploid chromosome complements in haploid parthenotes and diploidy in diploid parthenotes and IVF blastocysts (Fig. 3, A'C').
|
After 96 h of culture in vitro, haploid morulae arrested, and most did not form blastocysts but rather underwent degeneration (Fig. 3D), whereas diploid parthenogenetic blastocysts (Fig. 3E) and IVF blastocysts (Fig. 3F) had expanded or were hatching. These results indicate that haploid parthenotes were developmentally compromised prior to blastocyst stages in contrast to diploid parthenotes. However, the developmental rate of diploid parthenotes was indistinguishable from that of IVF embryos.
Comparison of Cell Number and Incidence of Apoptosis in Parthenotes and IVF Embryos
The developed embryos were evaluated for both total cell number and percentage of apoptotic cells. Morphological and biochemical features of apoptosis include fragmentation of both nuclei and DNA [48]. Both features were examined in mouse embryos at 72 and 96 h of in vitro culture. Cell nuclei stained with propidium iodide appeared red, and TUNEL-labeled apoptotic nuclei appeared green. Condensed nuclei or relatively larger condensed fragmented nuclei surrounded by small fragmented nuclei labeled by TUNEL were counted as single apoptotic nuclei (Fig. 4, A and B), and the cell region was also confirmed by the green background. When many fragmented nuclei labeled by TUNEL were sparsely distributed in an embryo and were marginally distinguishable by conventional fluorescence microscope, z-section imaging was performed using the confocal microscope. A 3-dimensional image then was reconstructed, allowing accurate registrations of TUNEL-positive nuclei as individual nuclei (Fig. 4B). At 72 h postactivation, the total cell number in haploid parthenotes (27 ± 7), mostly morulae, was significantly less (P < 0.001) than that in diploid parthenotes (39 ± 8) and IVF embryos (36 ± 5), which were mostly (
90%) blastocysts (Table 2). Moreover, the incidence of apoptosis in the haploid parthenotes (7%) was significantly higher (P < 0.05) than that of diploid parthenotes (3%) and IVF blastocysts (4%) (Table 2). In addition, haploid parthenotes that arrested at cleavage stages prior to the morula stage exhibited TUNEL-positive staining in some nuclei (data not shown).
|
|
At 96 h postactivation, the total cell number in degenerative haploid parthenotes (47 ± 21, Fig. 3D), including 7 blastocyst-like embryos, was considerably less (P < 0.001) than that in diploid blastocysts (104 ± 17) and IVF blastocysts (102 ± 17). Moreover, the incidence of apoptotic cells (17%) in haploid parthenotes was significantly higher (P < 0.01) than that in diploid parthenotes (4%) and IVF blastocysts (6%) (Table 2). At the stages examined, there were no differences in the cell number and incidence of apoptosis between diploid parthenotes and IVF blastocysts.
We further tested postimplantation development of diploid parthenogenetic embryos by transferring 213 diploid parthenotes (179 blastocysts and 34 morulae) into 8 pseudopregnant CD-1 recipient female mice. Although signs of pregnancy (e.g., swollen abdomen, weight gain) were seen 1012 days after embryo transfer, no diploid parthenotes developed into live pups. In contrast, a birth rate of 7075% has been routinely obtained for the transfer of IVF-produced blastocysts [49].
| DISCUSSION |
|---|
|
|
|---|
Transcripts of genes and proteins that regulate apoptosis have been found in both mouse and human preimplantation embryos, indicating the existence of apoptotic machinery throughout preimplantation development [1012, 18, 50]. A low incidence of apoptosis exists in blastocysts developed from fertilized zygotes, suggesting a role for apoptosis in normal development [17]. We found that apoptosis does not require the paternal genome during preimplantation development. Instead, the machinery necessary for apoptosis is inherited from the oocyte. The oocyte contributes not only stored mRNA and proteins for early development but also maternally inherited components such as mitochondria for mediating apoptosis [41]. It also seems likely that the expression of the genes involved in apoptosis from the maternal genome alone is sufficient for initiation of apoptosis during early development.
Haploid parthenotes were developmentally retarded (most arrested at the morula stage), and haploid embryos had fewer cells than did diploid embryos, consistent with previous reports [20, 51]. Developmentally retarded or arrested haploid parthenotes exhibited an increased incidence of apoptosis. Cell cycle arrest and apoptosis are closely interconnected in many other somatic cell types [52, 53] and also may be interrelated in preimplantation embryos.
We noticed that most nuclei positively labeled with TUNEL appeared either condensed or fragmented in mouse embryos. Caspase activation plays a central role in the execution of apoptosis and is involved in chromatin condensation and DNA fragmentation [54, 55], and nuclear DNA fragmentation is correlated with caspase activation during apoptosis in early embryos [12, 14, 41].
Consistent with previous reports [23, 26, 44], Sr2+ effectively induced Ca2+ oscillations at a frequency similar to that of sperm-induced Ca2+ oscillations. Sr2+ induces Ca2+ release from intracellular Ca2+ stores in other cell types [56, 57]. In the present study, Sr2+ failed to trigger Ca2+ oscillations and to activate oocytes in Ca2+-containing medium, presumably because Ca2+ inhibits Sr2+ entry into the cells through Ca2+ channels [58]. Using Sr2+ in combination with CCD to activate mouse oocytes in Ca2+-free medium, we achieved rates of diploid parthenogenetic development in KSOM that are indistinguishable from those of IVF embryos in the present study and those of in vivo preimplantation development of parthenotes in a previous study [35]. Moreover, the total cell number and the percentage of apoptotic nuclei were similar between diploid parthenotes and IVF-produced embryos. Diploid parthenogenetic mouse embryos developed in vitro exhibit reduced rates of development, fewer cells in the blastocyst, and defects in cell locations compared with IVF embryos [30, 33, 59]. The improved rate of development and enhanced cell number we observed in diploid parthenotes might be attributed to the method used for oocyte activation, culture conditions, or the mouse strain employed. Oocyte activation is a critical step in successful cloning by nuclear transfer [31, 60, 61]. Our results demonstrate that oocyte activation itself does not lead to apoptosis during preimplantation stages. The frequency of apoptosis is significantly increased, especially in the inner cell masses of cloned embryos obtained by splitting, which does not involve a step of parthenogenetic activation and may contribute to reduced fetal viability [62].
Although diploid parthenotes derived from activation with Sr2+ and CCD develop at a rate and display cell numbers and apoptotic cell numbers similar to those of embryos produced from IVF, the parthenotes failed to develop to term, implying that genomic imprinting is essential for later development [63, 64]. Previous experiments also demonstrated limited postimplantation development derived from diploid parthenogenetic mouse embryos [20, 65]. It is unclear whether the failure of blastocyst parthenotes to develop to term involves apoptosis in addition to genomic imprinting.
The findings that haploiy leads to apoptosis may have clinical implications. Haploidy and hypoploidy are not uncommon in human preimplantation embryos [24] and are probably linked to abnormal embryo development or developmental arrest [1, 5]. Haploidy can result from parthenogenetic activation of oocytes or failure of transformation of sperm nuclei into zygotic male pronuclei after IVF or intracytoplasmic sperm injection [6668]. Haploid embryos can cleave and develop to certain stages such as the morula stage, whereas normal diploid embryos reach the blastocyst stage. If embryo transfers are performed before the morula stage, some haploid embryos could be transferred, which would compromise the pregnancy rate. Because haploid and probably hypoploid embryos fail to reach the blastocyst stage, perhaps the blastocyst stage is a critical point for selection of healthy embryos with low incidence of apoptosis that are more competent for full-term development. The transfer of blastocysts or hatching blastocysts has improved pregnancy rates [6972].
The extent to which findings derived from haploidy-induced apoptosis can be generalized to other forms of aneuploidy is unclear at present. Haploidy but not parthenogenetic activation is associated with apoptosis during preimplantation development. The mechanism underlying the haploidy-associated apoptosis warrants further investigation.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 This work was supported by grants from the National Institutes of Health (K081099) and by the Women and Infants Hospital/Brown Faculty Research Fund. ![]()
2 Correspondence: David L. Keefe, Department of Obstetrics and Gynecology, Brown University and Women & Infants Hospital, 101 Dudley Street, Providence, RI 02905. FAX: 401 453 7599; dkeefe{at}wihri.org ![]()
Accepted: August 21, 2001.
Received: June 29, 2001.
| REFERENCES |
|---|
|
|
|---|
. Biol Reprod 1997; 56:1088-1096[Abstract]This article has been cited by other articles:
![]() |
T. Somfai, M. Ozawa, J. Noguchi, H. Kaneko, K. Ohnuma, N. W. K. Karja, M. Fahrudin, N. Maedomari, A. Dinnyes, T. Nagai, et al. Diploid porcine parthenotes produced by inhibition of first polar body extrusion during in vitro maturation of follicular oocytes. Reproduction, October 1, 2006; 132(4): 559 - 570. [Abstract] [Full Text] [PDF] |
||||
![]() |
N T Rogers, G Halet, Y Piao, J Carroll, M S H Ko, and K Swann The absence of a Ca2+ signal during mouse egg activation can affect parthenogenetic preimplantation development, gene expression patterns, and blastocyst quality Reproduction, July 1, 2006; 132(1): 45 - 57. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Suganuma, C. M. Walden, T. D. Butters, F. M. Platt, R. A. Dwek, R. Yanagimachi, and A. C. van der Spoel Alkylated Imino Sugars, Reversible Male Infertility-Inducing Agents, Do Not Affect the Genetic Integrity of Male Mouse Germ Cells During Short-Term Treatment Despite Induction of Sperm Deformities Biol Reprod, April 1, 2005; 72(4): 805 - 813. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Han, Y. G. Chung, S. Gao, and K. E. Latham Maternal Factors Controlling Blastomere Fragmentation in Early Mouse Embryos Biol Reprod, March 1, 2005; 72(3): 612 - 618. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Manandhar, H. Schatten, and P. Sutovsky Centrosome Reduction During Gametogenesis and Its Significance Biol Reprod, January 1, 2005; 72(1): 2 - 13. [Abstract] [Full Text] [PDF] |
||||
![]() |
N T Rogers, E Hobson, S Pickering, F A Lai, P Braude, and K Swann Phospholipase C{zeta} causes Ca2+ oscillations and parthenogenetic activation of human oocytes Reproduction, December 1, 2004; 128(6): 697 - 702. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Hao, L. Lai, J. Mao, G.-S. Im, A. Bonk, and R. S. Prather Apoptosis in Parthenogenetic Preimplantation Porcine Embryos Biol Reprod, June 1, 2004; 70(6): 1644 - 1649. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Lattanzi, C. B. Santos, M. D. Mudry, and J. L. Baranao Exposure of Bovine Oocytes to the Endogenous Metabolite 2-Methoxyestradiol During In Vitro Maturation Inhibits Early Embryonic Development Biol Reprod, December 1, 2003; 69(6): 1793 - 1800. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Lacham-Kaplan, J. Shaw, L. G. Sanchez-Partida, and A. Trounson Oocyte Activation after Intracytoplasmic Injection with Sperm Frozen Without Cryoprotectants Results in Live Offspring from Inbred and Hybrid Mouse Strains Biol Reprod, November 1, 2003; 69(5): 1683 - 1689. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Hao, L. Lai, J. Mao, G.-S. Im, A. Bonk, and R. S. Prather Apoptosis and In Vitro Development of Preimplantation Porcine Embryos Derived In Vitro or by Nuclear Transfer Biol Reprod, August 1, 2003; 69(2): 501 - 507. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Krivokharchenko, E. Popova, I. Zaitseva, L. Vil'ianovich, D. Ganten, and M. Bader Development of Parthenogenetic Rat Embryos Biol Reprod, March 1, 2003; 68(3): 829 - 836. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||