Article Microtubule turnover in ooplasm biopsy reflects ageing phenomena in the parent oocyte

Transcription

1 RBMOnline - Vol 11. No Reproductive BioMedicine Online; on web 1 June 2005 Article Microtubule turnover in ooplasm biopsy reflects ageing phenomena in the parent oocyte Dr Pravin T Goud Pravin T Goud, MD, PhD, has been active in reproductive research since After completing his medical degree and Ob/Gyn residency in Mumbai, India, Dr Goud completed his PhD with honours in Reproductive Biology from University of Ghent, Belgium. He has a longstanding interest in molecular studies on oocytes, embryos, and gamete/embryo cryopreservation. Dr Goud s field of research has included oocyte maturation and ageing, and his original contributions include studies on inositol 1,4,5- trisphosphate sensitive receptors in human oocytes and embryos, and studies on abnormal fertilizations after ICSI. Dr Goud is currently active in research at the Wayne State University. The current study has received the DF Richardson Memorial Award by the American College of Obstetrics and Gynecology. AP Goud 1,5, PT Goud 1, MP Diamond 1,2, P Van Oostveldt 3, MR Hughes 1,4 1 Department of Obstetrics and Gynecology, Wayne State University, Detroit, MI; 2 Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Wayne State University, Detroit, MI, USA; 3 Laboratory of Biochemistry and Molecular Cytology, Faculty of Agriculture, Ghent University, Ghent, Belgium; 4 Centre for Molecular Medicine and Genetics, Wayne State University Detroit, MI, USA 5 Correspondence: CS Mott Centre for Human Growth and Development, 275 E Hancock, Detroit, MI 48201, USA. Tel: ; Fax: ; agoud@med.wayne.edu Abstract Oviductal oocytes retrieved from superovulated B6D2F1 mice at 13.5, 16 and 19 h after human chorionic gonadotrophin (HCG) (groups A, B and C respectively, n = 382) were micromanipulated to obtain µm sized ooplasm biopsy fragments. Experiments were divided into three sets. Ooplasmic microtubule dynamics were studied in ooplasm biopsy specimens and parent oocytes (set 1) and ooplasm biopsy specimens (set 2), whilst zona pellucida dissolution time, cortical granule loss and spindle/chromatin morphology using confocal microscopy were also studied in parent oocytes (set 2). Oocytes withstood oocyte biopsy with a high survival rate (98.2%) and the biopsied oocytes underwent successful fertilization and development (set 3). An absolute one-to-one correlation was seen between the oocyte biopsy specimens and the parent oocytes in terms of ooplasmic microtubule dynamics (set 1), and increased ooplasmic microtubule dynamics in oocyte biopsy specimens paralleled ageing phenomena in the parent oocytes (set 2). Zona pellucida dissolution time was significantly lower in parent oocytes from group A versus groups B (P= 0.032), and C (P< 0.001). (Groups A, B, C include minimal, moderate, increased ooplasmic microtubule dynamics in oocyte biopsy specimens respectively.) Oocyte cortical granule loss and spindle/chromatin abnormalities were mainly seen in group C (P < 0.001). Oocyte biopsy can thus be applied to judge age-related changes in the parent oocytes. Keywords: cortical granules, ICSI, microtubule dynamics, oocyte, ooplasm biopsy, post-ovulatory ageing, spindle Introduction Achievement of success in obtaining a viable pregnancy while avoiding multiple order births is a major challenge in assisted reproduction treatment programmes today. Both of these goals could be achieved by transferring to the uterus a single embryo with known high implantation potential (Ozturk et al., 2001). However, there are limitations to optimal embryo selection, since there are no current methods that guarantee improved developmental potential. The contemporary approach to embryo selection is based on examining the morphology and rate of in-vitro development of the oocyte/embryo (Saith et al., 1998). Nevertheless, this method does not definitively predict the success of implantation or continual development. Hence, more than one embryo is generally transferred to the subjects undergoing assisted reproduction. As a result, there has been a significant increase in multiple gestations in addition to higher pregnancy rates in the SART registry (Reynolds et al., 1997). Research in the area has continued over the years, and the concept of oocyte and embryo quality has emerged. The term quality refers not only to the structural integrity, but also to the normality of chromosomes, metabolism, and developmental potential. Although direct and indirect 43

2 44 techniques such as embryo blastomere or polar body biopsy are available to assess chromosomal numbers, they have certain limitations, and their use has generally been confined either to women with advanced age or to couples with known genetic disorders (Verlinsky et al., 1998; Wilton et al., 2002). There are no established clinically useful methods to assess the oocyte/embryo metabolism and developmental potential so far. Oocyte pre-/post-ovulatory ageing is known to cause abnormal fertilization as well as developmental compromise (Chang et al., 1958; Gray et al., 1984; Wilcox et al., 1998; Goud et al., 1999). It is also possible that within a cohort of oocytes obtained at oocyte retrieval, there may be certain oocytes that have impaired fertilization and developmental potential, secondary to being aged or having a higher likelihood to undergo ageing. Furthermore, the cytopathology and to some extent, the pathophysiology of oocyte ageing are relatively well understood (Igarashi et al., 1997; Xu et al., 1997; Takahashi et al., 2003). This study therefore used oocyte ageing as a model abnormality, and applied a newly developed technique to obtain a viable ooplasm sample to assess oocyte ageing. Microtubules are components of cytoskeleton involved in various functions, including organelle trafficking and chromosome separation at cell division. Microtubules are made up of tubulin monomers that undergo dynamic changes of polymerization and depolymerization, and finally reach a steady state of microtubule turnover. This steady state is dependent on the cellular metabolism. The critical concentration for tubulin polymerization is affected by various factors, including the cell cycle stage (Alberts et al., 1994). Remarkably, ooplasmic microtubule turnover increases with post-ovulatory ageing (Pickering et al., 1988; Zernicka-Goetz et al., 1993; George et al., 1996). This increase can be made obvious by exposing oocytes to taxol, which enhances microtubules in the ooplasm of post-ovulatory old, but not young oocytes (Goud et al., 2004). Ooplasm biopsy could be representative of the parent oocyte, and increased microtubule dynamics in the ooplasm biopsy may reflect ageing related phenomena of the parent oocyte. This study therefore examined the response of ooplasm biopsies and parent oocytes to taxol and used taxol-enhanced microtubule dynamics in the ooplasm biopsy (Goud et al., 2004) as a marker to predict ageing related changes in the parent oocyte. Materials and methods Study design Approval for the current study was obtained from Wayne State University s Animal Investigation committee. Design of the study involved obtaining MII stage oocytes from superovulated mice at 13.5, 16 and 19 h after human chorionic gonadotrophin (HCG). In experiment set 1, normal appearing MII stage oocytes obtained from B6D2F1 mice were divided into treatment and control groups. The oocytes in the treatment group were subjected to the micromanipulation procedure of ooplasm biopsy. The biopsy fragment, and the parent and control oocytes were all treated with taxol. Microtubules were examined after α-tubulin fluorescence immunocytochemistry and confocal microscopy. In experiment set 2, oocytes were obtained and divided in the same way as set 1, and the oocytes in the treatment subgroup were subjected to ooplasm biopsy. Only the ooplasm biopsy fragments were treated with taxol prior to tubulin immunocytochemistry, whereas the parent oocytes were processed for cortical granule, tubulin, and chromatin staining after assessment of zona pellucida (ZP) dissolution time. The operator was blinded to oocyte age prior to ooplasm biopsy and assessment of ooplasmic microtubule dynamics. In experiment set 3, oocytes were subjected to ooplasm biopsy followed by intracytoplasmic sperm injection (ICSI), and followed through fertilization and development in culture. Superovulation and oocyte retrieval Four- to 6-week-old B6D2F1 mice were obtained from Jackson Laboratories (Bar Harbor, ME, USA), and were adjusted to a 14-h light -10-h dark cycle for at least 1 week prior to superovulation with 7.5 IU each of pregnant mare s serum gonadotrophin (PMSG) and HCG (Sigma, Saint Louis, MO, USA), administered IP h apart. Mice were killed h after HCG injection and oocytes were retrieved from their oviductal ampullae. The cumuli were treated with 0.1% hyaluronidase (w/v) in M2 medium (Sigma) for 2 3 min at 37 C to release oocytes, which were subsequently denuded to remove all cumulus corona cells with a narrow bore pulled glass Pasteur pipette. Oocytes were thoroughly rinsed in M2 medium, inspected to rule out abnormal morphology and were kept ready in M16 medium (Sigma) pre-equilibrated with 5% CO 2 in air at 37 C for subsequent procedure with or without micromanipulation. Micromanipulation set-up and procedure The set-up used for the micromanipulation procedure was similar to that used for embryo-biopsy and preimplantation diagnosis (Dozortsev and McGinnis, 2001), except for some differences in the dimensions of the micro-tools. The external and internal diameters of the holding pipettes were µm and µm respectively and those for the embryo biopsy (EB) micropipettes were µm and µm respectively. Micromanipulations were performed on a warm stage (37 C) of a Nikon Diaphot microscope equipped with coarse hydraulic micromanipulators (Narishige, Tokyo, Japan). The micromanipulation procedure for ooplasm biopsy was performed as described below (Figure 1A L). The oocytes were steadily held with the holding pipette and a partial zona dissection was performed using a sharp tipped partial zona dissection micropipette at a pole away from the polar body and the spindle area. Care was taken to maintain the same orientation in each oocyte. The oocyte position was then readjusted using the holding micropipette, and an EB micropipette was inserted into the perivitelline space gently abutting against the oolemma. The ooplasm was carefully aspirated without puncturing the oolemma. This was followed by a gentle withdrawal of the EB pipette, thereby stretching the ooplasm to a point where the ooplasm fragment broke from the parent oocyte. The oolemma enveloped ooplasm fragment that was then allowed to

3 Figure 1. Photomicrographs (A) through (L) depict the ooplasm biopsy micromanipulation procedure. An oocyte was held with a holding micropipette and partial zona dissection was performed using conventional technique (A D). An ooplasm biopsy micropipette was then introduced through the newly created slit in zona pellucida (arrowhead in D), and the ooplasm was aspirated into the biopsy micropipette (I), which was withdrawn gently to create a break in the oolemma. An intact parent oocyte (parent oocytes) after biopsy is seen in (J). The ooplasm biopsy fragment (oocyte biopsy specimens) underwent a spontaneous change in its initially elongated shape, finally assuming a rounded shape in about five minutes (J L). Initial magnification Scale bars represent 50 µm. recover for 4 5 min, during which it assumed a rounded shape (Figure 1L). Taxol treatment, zona removal and fixation Stock solution of 1 mmol/l taxol (Paclitaxel; Sigma) was prepared in dimethyl sulphoxide (DMSO) and stored at 20 C. The final concentration of taxol in the medium carrying the ooplasm biopsy/oocytes was 5 µmol/l, which was prepared in M2 medium containing 10% (v/v) fetal bovine serum (FBS; Invitrogen, Paisley, UK). The ZP of the parent and control oocytes was removed with acid Tyrode s solution (Sigma), at 37 C and in experiment set 2, the time required for complete dissolution of ZP was recorded for each oocyte in seconds under direct microscopic visualization. The biopsy fragments and parent/control oocytes were fixed in freshly prepared 4% paraformaldehyde at 37 C for 1 h and rinsed thoroughly with phosphate buffered saline containing 0.1% Triton X-100 (PBS TX) and 0.3% (w/v) bovine serum albumin (BSA, Sigma). Finally, the oocytes were stored in a blocking solution of PBS containing 3% BSA. Tubulin fluorescence immunocytochemistry The biopsy fragments and oocytes were then incubated with anti α-tubulin monoclonal antibody (mouse, antihuman; Sigma, 1:300, 1 h), rinsed thoroughly with PBS TX 0.3% BSA and reincubated with a fluorescein isothiocyanate (FITC)-conjugated secondary antibody (rat, anti-mouse, IgG; Sigma, 1:500). The oocyte/biopsy fragments were thoroughly rinsed once again with the PBS TX 0.3% BSA solution prior to mounting in Vectashield with DAPI (Vector Laboratories, Burlingame, CA, USA), that contained 4 6-diamidino-2-phenylindole (DAPI), a fluorescent chromatin stain. The oocytes/biopsy fragments were stored in the mounting medium in glass chambers at 4 C until processing with confocal microscopy (Carl Zeiss or Bio-Rad UV 1024). Image processing and threedimensional (3-D) reconstructions were done with Imaris and Huygens system-2 (SVI, Hilversum, Netherlands). In experiment set 2, staining for cortical granules was performed with a technique described earlier (Goud et al., 2002) using rhodamine-conjugated lens culinaris agglutinin (LCA; Vector). The cortical granules were stained 45

4 fluorescent red, which was distinct from the fluorescent green staining of the microtubules and fluorescent blue staining of chromosomes. Individual ooplasm biopsy fragments were closely examined for ooplasmic microtubules, whereas their parent and control oocytes were examined for spindle/ooplasmic microtubules, chromatin and cortical granule status. Comparisons were made utilizing individual parent oocyte/ooplasm biopsy pairs in set 1, taking into consideration the post-ovulatory age. In set 2, parent oocytes with minimal, moderate or increased ooplasmic microtubule dynamics in oocyte biopsy specimens were compared for ageing phenomena. Within the parent oocytes and controls, special attention was given to spindle morphology and orientation, condensation status of chromatin, location of chromosomes in relation to the metaphase plate and presence or exocytosis of cortical granules. Ooplasmic microtubule dynamics in response to taxol was evaluated and graded into the following three categories namely, minimal or negligible, moderate, and increased. The first category, minimal/negligible, included oocytes with microtubules restricted to the spindle and occasional microtubule organizing centre, no free microtubules in the cortex; the second category, moderate included oocytes displaying some free microtubules in the cortex in addition to the microtubule organizing centre, without formation of dense microtubule network. The third category of (markedly) increased microtubules included oocytes with extensive dense cortical microtubules extending into the rest of the ooplasm. Similarly, cortical granule (cortical granule) status in each oocyte was categorized as intact cortical granule, minor cortical granule loss and major cortical granule loss. Those with intact cortical granule had a rim of cortical granule visible all along the oolemma without aggregation of exocytosis in all optical sections examined. An exception was the cortical granule free domain in the vicinity of the spindle apparatus. Those categorized into the minimal loss category had <10% loss in cortical granule in one or more optical sections. Oocytes with >10% cortical granule loss were categorized into the major cortical granule loss group. The categorization of oocytes based on MT and cortical granule status was confirmed by two independent observers blinded to treatment group assignment, who used comprehensive evaluation of the individual optical sections and the 3-D reconstructed images. Exact concordance was noted between the scores provided by the two observers. In experiment set 3, oocytes were obtained after superovulation as above, subjected to cumulus removal, and ooplasm biopsy, and were further subjected to ICSI with epididymal spermatozoa using conventional technique (Kimura and Yanagimachi, 1995; Rybouchkin et al., 1996). Sibling control oocytes were subjected to ICSI without ooplasm biopsy. All the oocytes were subjected subsequently to embryo culture in M-16 medium (Sigma). Fertilization was checked the next morning, and further development was monitored during the following 72 h. Statistical analysis Statistical analysis was performed by the SPSS (Version 11.0; SPSS Inc., Chicago, IL, USA). In experiment set 2, the zona dissolution time between groups was compared using the two-way ANOVA. Oocyte numbers in individual subgroups were compared using the Fisher s exact test. Significance was defined as P < Data are expressed as means ± SE. Results The numbers of oocytes utilized in individual experiment sets as well as subgroups are presented in Table 1. Accordingly, 166 oocytes were biopsied in experiment sets 1 (58) and 2 (108) respectively. Only three oocytes were damaged during the procedure, giving a survival rate of 98.2%. Furthermore, all the oocytes damaged belonged to the 19-h subgroup and for the oocytes in the other subgroups, the survival rate of oocytes undergoing the biopsy procedure was 100%. Processing of the ooplasm biopsy was successful in all the oocytes subjected to ooplasm biopsy in both experiment sets. Control oocytes from experiment set 1 were exposed to taxol in the same way as the biopsied oocytes and ooplasm biopsy fragments. Taxol resulted in significant (P < 0.05) enlargement of the spindle apparatus with broadening of the poles in all subgroups of oocytes. However, there were negligible or only minimal microtubules seen in the ooplasm of oocytes from group A (Figure 2A). In the control oocytes from group B, some foci of the microtubule organizing centre were seen mainly in the cortical ooplasm, but there were none or minimal microtubules visualized in the rest of the ooplasm. Nonetheless, the overall Table 1. Summary of the study material including numbers of oocytes within each individual experiment set and subgroup. Experiment set 1 Experiment set 2 Group A Group B Group C Total Group A Group B Group C Total 46 Post-ovulatory age (h) No. oocytes Control oocytes Oocytes biopsied Oocytes survived

5 Figure 2. (A F) are fluorescent photomicrographs of optical sections obtained on a confocal laser scanning microscope. (G) and (H) are 3-D image reconstructions created from several 2 3 µm optical sections obtained using a two-photon confocal laser scanning microscope. In (A), an optical section obtained approximately at the equator shows the spindle apparatus with significant broadening (arrowhead) as a result of taxol effect. However, negligible ooplasmic microtubular activity is seen. In (B), minimal to moderate microtubule activity is seen in an oocyte obtained at 16 h after human chorionic gonadotrophin (HCG). The spindle in this oocyte is not seen in the current section. In (C), an optical section of an oocyte obtained at 19 h post-hcg is seen. Taxol in this oocyte resulted in significantly increased turnover in both spindle and ooplasmic microtubules (P < 0.05). (A), (B) and (C) were all control oocytes. (D), (E) and (F) are optical sections of biopsied oocytes and their biopsy fragments. The same pattern of progressively increasing ooplasmic microtubule turnover is seen in the parent oocytes and their respective biopsy fragment. (D) is an oocyte from the 13.5-h group, which shows negligible ooplasmic microtubules and a prominent spindle due to taxol effect. The ooplasm biopsy is difficult to visualize due to extremely low microtubule turnover (small vertical downward arrow in D). The ooplasmic microtubule turnover is noted to increase moderately in (E), where the oocyte belongs to the 16-h group. The same moderate microtubule turnover is also seen in the ooplasmic biopsy fragment (oblique small arrow to the right in E). In (F), the optical section of an oocyte from the 19-h group shows significantly increased ooplasmic microtubules in addition to spindle enhancement (P < 0.05). A marked increase is seen in the ooplasmic biopsy fragment as well (small arrowhead in F). (G) and (H) are 3-D image reconstructions of control and biopsied oocytes respectively from the 19-h group. A remarkable enhancement of spindle poles is prominently seen in three dimensions (oblique arrowhead in G). Furthermore, there are several microtubule organizing centres or asters (MTOC) seen in the ooplasm. In (H) a similarity is evident between an ooplasm biopsy fragment (oocyte biopsy specimens) and its parent oocyte (parent oocytes) in terms of the ooplasmic microtubule turnover. All images initially recorded at magnification. Image processing has resulted in different apparent sizes. All scale bars represent 50 µm. Green fluorescence represents the fluorescein isothiocyanate (FITC)-stained α-tubulin. Red and green colours in (H) are due to two-photon confocal microscopy. 47

6 microtubule activity in the ooplasm was relatively increased in the control oocytes from group B compared with those from group A (Figure 2B). In group C, there was a marked increase in the ooplasmic microtubules of taxol exposed control oocytes. The ooplasmic microtubules were numerous and were not restricted to the MTOC. In addition, the MTOC were more prominent in comparison to groups A and B (Figure 2C). Among the oocytes subjected to the biopsy procedure in experiment set 1, all oocytes and ooplasm biopsy fragments in group A (20/20, 13.5 h post-hcg) had minimal or negligible microtubules, whereas all ooplasm biopsy fragments and biopsied oocytes from groups B and C showed progressive increase in the ooplasmic microtubules (Figure 2D F). This phenomenon was similar to the control oocytes described above. Thus, the spindle and ooplasmic microtubule morphology of biopsied oocytes was the same as the corresponding control groups (Figure 2D F). Moreover, each individual oocyte ooplasm biopsy pair had identical morphology of ooplasmic microtubules, namely minimal in group A, moderately increased in group B and markedly increased in group C (Figure 2D F, arrows). In experiment set 2, only the biopsy fragments were exposed to taxol whereas the biopsied oocytes were subjected to zona removal, zona dissolution assessment, and cortical granule, spindle and chromatin staining. All the ooplasm biopsy fragments in group A had minimal microtubules. Whereas in group B, most ooplasm biopsies (86.1%) had moderately increased microtubules and the remaining biopsies (13.9%) had minimal microtubules (Figure 3J, Table 2). In group C, most ooplasm biopsies (82.9%) displayed markedly increased microtubules while the remaining ooplasm biopsies (17.1%) exhibited moderately increased microtubules. Furthermore, the 48 Figure 3. Fluorescent photomicrographs of oocytes depicting microtubules (green), cortical granules (red) and chromosomes (blue). The biopsy fragments but not the parent oocytes were exposed to taxol. Thus, the ooplasmic microtubule enhancement is seen only in the ooplasm biopsy fragments, mainly in 16-h and 19-h groups. In (A) and (C), an intact spindle and chromosome metaphase plate is visualized (arrowheads) in oocytes from and 16-h groups respectively. Cortical granules are peripheral and intact in (B) and (E), in the same oocytes as (A) and (C) respectively. Elongation of the spindle is seen in (D) (arrowhead). (F H) are old oocytes from 19-h group with their respective ooplasm biopsies. An increased microtubule turnover in the ooplasm biopsy is evident, and is paralleled by cortical granule exocytosis in (F) and (G), and spindle morphological abnormality in (H). (J) is an oocyte from the 16-h group with minimal microtubules in the ooplasm biopsy and normal spindle and metaphase plate. (I) is a control old oocyte with spindle orientation abnormality. In (K), a spindle abnormality with some disruption of chromosome metaphase plate is seen, and is paralleled by a marked increase in the microtubule turnover in the corresponding ooplasm biopsy fragment. (L) shows three intact chromosome metaphase plates of oocytes from and 16-h groups that were subjected to ooplasm biopsy, which exhibited minimal microtubule turnover. All images acquired initially at a magnification of Scale bars represent 50 µm. Slight alterations in image sizes may have resulted from image processing.

7 number of biopsies with markedly increased microtubule dynamics was significantly higher in group C (P < ) and those with moderately increased microtubule turnover were significantly increased in group B (group A versus group B, P = and group B versus group C, P = 0.025). These indicate a significant progressive increase in microtubule dynamics in ooplasm biopsies from groups A C (Table 2). The zona dissolution time was significantly (P < ) increased in oocytes from group B compared with group A and in oocytes from group C compared with groups A and B. A similar significant (P < ) progressive increase was also noted in the corresponding control oocytes (Table 3). Assessment of cortical granule status and chromatin as well as spindle morphology in oocytes from groups A and B revealed intact cortical granules, normal spindle morphology and orientation as well as intact chromosome metaphase plates in all oocytes studied (Figure 3A C, E and Figure 4). However, in group C, significantly (P < ) more oocytes showed abnormal spindle morphology (abnormal shape, disruption or disappearance, and centripetal migration of the spindle), disrupted alignment of chromosomes on the metaphase plates and metaphase to anaphase transition (Figure 3H, I, K and Figure 4B). There was no difference in ZP dissolution time, cortical granule status, or spindle/chromatin integrity between biopsied and control oocytes. The ooplasm biopsy fragments from all oocytes with abnormal spindle morphology and/or major cortical granule loss revealed increased ooplasmic microtubule dynamics. Similarly, all oocytes with minimal microtubules in ooplasm biopsy revealed intact cortical granules and normal morphology of the spindle and chromatin irrespective of oocyte post-ovulatory age. Thus, microtubule activity in the ooplasmic biopsy fragment was closely related to ageing phenomena in the parent oocyte, namely, ooplasmic microtubule dynamics, ZP hardening (increased time required for zona pellucida dissolution), cortical granule loss, as well as spindle and chromatin changes. In experiment set 3, of the 114 oocytes retrieved, 54 were subjected to ooplasm biopsy followed by ICSI, and the other 60 oocytes were subjected to ICSI without ooplasm biopsy. Survival after both procedures and successful fertilization (two pronuclei, two polar bodies) occurred in oocytes irrespective of ooplasm biopsy (69.2 versus 71.7% in oocytes with or without oocyte biopsy specimens respectively). Similarly, the rates of development of the pronuclear zygotes to the 2-cell stage, 2-cell stage to morulae, and morulae to blastocysts were similar in the two groups (2-cell 100% in both, morulae: 80.6 versus 81.4%, and blastocysts: 86.2 versus 85.7% respectively, in oocytes subjected to ICSI with or without ooplasm biopsy). Table 2. Ooplasm biopsy and microtubule dynamics in the biopsy fragments. Group A Group B Group C Total P-value Oocytes biopsied No. oocytes surviving biopsy NS Microtubules in biopsy-minimal (%)* 35 (100) 5 (13.9) 0 40 < a,b,c Microtubules moderately increased (%) 0 31 (86.1) 6 (17.1) d, e, f Microtubules markedly increased (%) (82.9) 29 < g,h *a Groups A versus B, b groups A versus C, c groups B versus C; d groups A versus B, e groups A versus C, f groups B versus C, g groups A versus C, h groups B versus C. NS = not significant Table 3. Zona pellucida (ZP) dissolution time, cortical granule status as well as spindle morphology in the biopsied parent and control oocytes. Group A Group B Group C Total P-value No. oocytes with ZP assessment Biopsied oocytes Control oocytes ZP dissolution time (s; mean ± SE) Biopsied oocytes 52.9 ± ± ± 4.1 < a c Control oocytes 54.2 ± ± ± 2.5 < d f Biopsied oocytes: a groups A versus B, b groups A versus C, c groups B versus C. Control oocytes: d groups A versus B, e groups A versus C, f groups B versus C. 49

8 A B Figure 4. Bar diagrams in (A) and (B) depict cortical granule status and spindle/chromatin morphology respectively. Oocyte numbers in group (C) and controls are significantly different compared with groups (A), (B) and corresponding controls, P< Spindle and chromatin abnormalities include: centripetal migration of the spindle, chromatin decondensation and metaphase anaphase transition. 50 Discussion Embryo selection forms the very epicentre of assisted reproduction, since the success of achieving pregnancy is dependent on the oocyte/embryo quality. Nonetheless, selecting the best embryos is very difficult due to lack of objective methods. As a result, there is a tendency to transfer more than one embryo with the expectation of improving chances of pregnancy. However, this also contributes to an increase in multiple gestations. Hence there is an urgent need to devise a method to objectively and definitively assess oocyte and embryo quality to predict developmental potential. This study describes a method to identify aged oocytes by performing an ooplasm biopsy. The technique of ooplasm biopsy was performed successfully without hampering the oocyte integrity and the biopsy fragments were found to be representative of the parent oocytes in terms of the microtubule dynamics. Moreover, increased microtubule dynamics in the ooplasm biopsy fragments were also predictive of ageing related changes in the parent oocytes. Oocyte pre- and post-ovulatory ageing could be one of the major contributors to diminution of assisted reproduction success. There is a definitive impact of oocyte ageing on fertilization, preimplantation development as well as implantation (Wilcox et al., 1998; Goud et al., 1999). It is also suggested that oocyte post-ovulatory ageing could contribute to defective chromosome segregation, eventually contributing to chromosome abnormalities in the offspring (Mailhes et al., 1998; Ma et al., 2004). Nonetheless, there is no definitive method to identify an aged oocyte other than oocyte morphology, and its ability to result in a normally fertilized embryo that is capable of undergoing cleavage divisions without fragmentation (Xia et al., 1997; Ezra et al., 2003). The marker used in this study to assess oocyte quality was taxol enhanced ooplasmic microtubule dynamics to identify aged oocytes (Goud et al., 2004). Microtubules are made up of tubulin monomers, which assemble to form oligomers and subsequently polymers, which in an MII stage oocyte are in the state of dynamic instability or a state of equilibrium between polymerization and depolymerization of tubulin (Alberts et al., 1994). The microtubule metabolism or dynamics are, however, also dependent on the physiological and cell cycle stage of the oocytes. Accordingly, microtubule dynamics are highly active in the spindle apparatus in an MII stage oocyte, while the ooplasmic microtubule dynamics are relatively low. This explains why under normal circumstances, in a freshly ovulated MII oocyte there are no ooplasmic microtubules seen, barring a few MTOC (Maro et al., 1985). This situation nonetheless changes as oocytes age. Ageing increases the microtubular dynamics in the ooplasm (Pickering et al., 1988; Zernika-Goetz et al., 1993; George et al., 1996). Ageing related increased ooplasmic microtubule dynamics thus forms the basis of the current study (Goud et al., 2004). Furthermore, to make the ooplasmic microtubules more prominent for the ease of detection, the microtubular enhancer taxol (Paclitaxel) was used. Taxol is known to decrease the critical concentration for tubular polymerization, thereby increasing polymerization and lengthening the microtubules (Alberts et al., 1994). Interestingly, taxol increases the

9 microtubules in the ooplasm of older but not young oocytes (Goud et al., 2004). The same principle was used to detect increased microtubule dynamics in the ooplasm biopsy in the current study. The ooplasm biopsy was minimally invasive, since the biopsied oocytes could easily survive this procedure. Thus, following an ooplasm biopsy, the parent oocyte could still retain its viability and possibly also the developmental potential, while the ooplasm biopsy could be subjected to tests to determine quality of the parent oocyte. In the first experiment set, the ooplasm biopsies were found to be representative of their respective parent oocytes in terms of the microtubule turnover. Thus, the ooplasmic microtubule turnover was minimal in young oocytes as well as their ooplasm biopsy fragments in the 13.5-h group. The ooplasmic microtubule turnover moderately increased in the 16-h group. Therefore, both, the ooplasm biopsy fragments and their parent oocytes exhibited a moderate increase in ooplasmic microtubules. Finally, in the 19-h group, the ooplasmic microtubules increased markedly in both, the ooplasm biopsy fragments as well as the parent oocytes. Thus, the ooplasmic microtubule activity increased progressively with postovulatory oocyte ageing and the ooplasmic biopsies were representative of their parent oocytes. Hence ooplasm biopsy most likely represents physiological processes in the parent oocyte. This is encouraging for other future applications of ooplasm biopsy. In experiment set 2, the ooplasm biopsies showed the same post-ovulatory age-related progressive increase in the microtubule turnover in the ooplasm from 13.5-h to 19-h groups. Moreover, the parent oocytes showed other findings such as significant progressive hardening of the zona pellucida of the parent and control oocytes. Hardening of ZP is an indicator of oocyte ageing and is possibly related to cortical granule exocytosis secondary to ageing related disturbance in Ca 2+ homeostasis (Igarashi et al., 1997; Xu et al., 1997). This possibility becomes even stronger in view of the finding of cortical granule exocytosis, which was prominently noted in the parent oocytes, whose biopsies revealed increased microtubular dynamics. It is likely that increased ooplasmic microtubule turnover, cortical granule exocytosis and hardening of ZP are all related to the same pathophysiological mechanism of oocyte ageing. The most likely explanation is dysregulation of Ca 2+ homeostasis secondary to failure of the Ca 2+ pumps in the membrane and endoplasmic reticulum in the oocytes, resulting in an increase in baseline cytosolic Ca 2+ concentrations (Igarashi et al., 1997; Takahashi et al., 2000). This can secondarily activate the oocyte mechanisms that advance the cell cycle stage (Xu et al., 1997; Kikuchi et al., 2000). This theory is further supported by other findings of influence of ageing on the spindle and chromatin morphology. Altered spindle morphology and orientation were seen, as well as disruption of the chromosome metaphase plate and metaphase anaphase transition in older oocytes, whose biopsies had increased ooplasmic microtubule turnover. On the other hand, there were no abnormalities found in the 13.5 h and 16-h groups with minimal or moderate ooplasmic microtubules in the biopsy fragment. Ooplasm biopsies were thus very useful in predicting the age-related phenomena, including spindle morphology in older oocytes. In addition, the biopsy procedure itself did not affect the incidence of ageing phenomena and spindle changes in all groups since the rate of these phenomena were similar between biopsied and control oocytes. The oocyte biopsy procedure thus did not affect the integrity of the spindles, chromatin or cortical granule in the oocytes. Young oocytes from both the 13.5-h and 16-h groups had a normal spindle apparatus. This indicates that the spindle physiology is normal when the ooplasmic microtubule turnover is minimal. Spindle morphology and orientation are closely related to embryonic developmental potential (Eichenlaub-Ritter et al., 2002; Moon et al., 2003; Smith and Silva, 2004; Stachecki et al., 2004). Therefore, minimal microtubule turnover in ooplasm biopsy may predict normal spindle structure and function. Furthermore, it may also predict the normality of chromosome numbers and developmental potential of the ensuing embryo. Finally, the results in experiment set 3 demonstrated that the oocytes subjected to ooplasm biopsy retain their ability to undergo normal fertilization and embryo development to the blastocyst stage. In conclusion, the ooplasm biopsy not only represented the parent oocytes in terms of ooplasmic microtubule turnover, but also reflected other ageing related changes in the parent oocytes, such as cortical granule exocytosis and ZP hardening. Particularly vital was the detection of abnormalities of spindle and chromosome metaphases in older oocytes, whose biopsies showed increased microtubule turnover. Furthermore, the ooplasm biopsy procedure per se did not affect the viability and spindle integrity or developmental ability of the oocytes. Ooplasm biopsy is thus a potentially useful technique for detection of aged oocytes. Further experiments will continue to work towards the application of ooplasm biopsy to predict fertilization and preimplantation development (Goud et al., unpublished data). Acknowledgements The authors wish to sincerely thank Drs K Moin and Riyaz Ul Haq, Department of Pharmacology, Wayne State University, Detroit, MI, for their technical support with confocal microscopy, and Michael Kruger, Biostatistician, Department of Obstetrics and Gynecology, for statistical support. References Alberts B, Bray D, Lewis J et al The cytoskeleton. In: Molecular Biology of the Cell 3rd edn. Garland Publishing, New York and London, pp Chang MC, Fernandez-Cano L 1958 Effects of delayed fertilization on the development of development of pronucleus and segmentation of hamster ova. Anatomical Research 132, Dozortsev D, McGinnis KT Micromanipulation in preimplantation genetic diagnosis Reproductive Science 12, Eichenlaub-Ritter U, Shen Y, Tinneberg HR 2002 Manipulation of the oocyte: possible damage to the spindle apparatus. Reproductive BioMedicine Online 5, Ezra Y, Simon A, Moon JH et al Visualization of the metaphase II meiotic spindle in living human oocytes using the 51

10 52 polscope enables the prediction of embryonic developmental competence after ICSI. Human Reproduction 18, George MA, Pickering SJ, Braude PR, Johnson MH 1996 The distribution of alpha- and gamma-tubulin in fresh and aged human and mouse oocytes exposed to cryoprotectant. Molecular Human Reproduction 2, Goud AP, Goud PT, Van Oostveldt P et al Dynamic changes in microtubular cytoskeleton of human postmature oocytes revert after ooplasm transfer. Fertility and Sterility 81, Goud P, Goud A, Van Oostveldt P et al Fertilization abnormalities and pronucleus size asynchrony after intracytoplasmic sperm injection are related to oocyte postmaturity. Fertility and Sterility 72, Goud PT, Goud AP, Leybaert L et al Inositol 1,4,5- trisphosphate receptor function in human oocytes: calcium responses and oocyte activation-related phenomena induced by photolytic release of InsP(3) are blocked by a specific antibody to the type I receptor. Molecular Human Reproduction 8, Gray RH 1984 Aged gametes, adverse pregnancy outcomes and natural family planning. An epidemiologic review. Contraception 30, Igarashi H, Takahashi E, Hiroi M, Doi K 1997 Aging related changes in calcium oscillations in fertilized mouse oocytes. Molecular Reproduction and Development 48, Kikuchi K, Naito K, Noguchi J et al Maturation/M-phase promoting factor: a regulator of aging in porcine oocytes. Biology of Reproduction 63, Kimura Y, Yanagimachi R 1995 Intracytoplasmic sperm injection in the mouse. Biology of Reproduction 52, Ma W, Zhang D, Hou Y et al Reduced expression of MAD2, Bcl-2 and MAP kinase activity in pig oocytes after in vitro ageing are associated with defects in sister chromatid segregation during meiosis II and embryo fragmentation after activation. Biology of Reproduction 72, Mailhes JB, Young D, London SN 1998 Postovulatory aging of mouse oocytes in vivo and premature centromere separation and aneuploidy. Biology of Reproduction 58, Maro B, Howlett SK, Webb M 1985 Non-spindle microtubule organizing centers in metaphase II-arrested mouse oocytes. Journal of Cell Biology 101 (5 Pt 1), Moon JH, Hyun CS, Lee SW et al Visualization of the metaphase II meiotic spindle in living human oocytes using the Polscope enables the prediction of embryonic developmental competence after ICSI. Human Reproduction 18, Ozturk O, Bhattacharya S, Templeton A 2001 Avoiding multiple pregnancies in assisted reproduction: evaluation and implementation of new strategies. Human Reproduction 16, Pickering SJ, Johnson MH, Braude PR, Houliston E 1988 Cytoskeletal organization in fresh, aged and spontaneously activated human oocytes. Human Reproduction 3, Reynolds MA, Schieve LA, Martin JA et al. Trends in multiple births using assisted reproductive technology, United States, Pediatrics 111, Rybouchkin A, Dozortsev D, Pelinck MJ et al Analysis of the oocyte activating capacity and chromosomal complement of round-headed human spermatozoa by their injection into mouse oocytes. Human Reproduction 11, Saith RR, Srinivasan A, Michie D, Sargent IL 1998 Relationships between the developmental potential of human in-vitro fertilization embryos and features describing the embryo, oocyte and follicle. Human Reproduction Update 4, Smith GD, Silva CASE 2004 Developmental consequences of cryopreservation of mammalian oocytes and embryos. Reproductive BioMedicine Online 9, Stachecki J, Munné S, Cohen J 2004 Spindle organization after cryopreservation of mouse, human and bovine oocytes. Reproductive BioMedicine Online 8, Takahashi T, Saito H, Hiroi M et al Effects of aging on inositol 1,4,5-triphosphate-induced Ca 2+ release in unfertilized mouse oocytes. Molecular Reproduction and Development 55, Takahashi T, Takahashi E, Igarashi H et al Impact of oxidative stress in aged mouse oocytes on calcium oscillations at fertilization. Molecular Reproduction and Development 66, Verlinsky Y, Cieslak J, Ivanhenko V et al Preimplantation diagnosis of common aneuploidies by the first- and second-polar body FISH analysis. Journal of Assisted Reproduction and Genetics 15, Wilcox AJ, Weinberg CR, Baird DD 1998 Postovulatory ageing of the human oocyte and embryo failure. Human Reproduction 13, Wilton L Preimplantation genetic diagnosis for aneuploidy screening in early human embryos: a review. Prenatal Diagnosis 22, Xia P 1997 Intracytoplasmic sperm injection: correlation of oocyte grade based on polar body, perivitelline space and cytoplasmic inclusions with fertilization rate and embryo quality. Human Reproduction 12, Xu Z, Abbott A, Kopf GS, Schultz RM, Ducibella T 1997 Spontaneous activation of ovulated mouse eggs: time dependent effects on M-phase exit, cortical granule exocytosis, maternal messenger RNA recruitment and inositol 1,4,5-trisphosphate sensitivity. Biology of Reproduction 57, Zernicka-Goetz M, Kubiak JZ, Antony C, Maro B 1993 Cytoskeletal organization of rat oocytes during metaphase II arrest and following abortive activation: a study by confocal laser scanning microscopy. Molecular Reproduction and Development 35, Received 21 February 2005; refereed 7 March 2005; accepted 25 April 2005.