Estrogen receptor β is expressed in human endometrium and levels of expression vary during the menstrual cycle.

many; Inst of Morphology, Bulgarian Acad of Science, Sofia, Bulgaria; Dept of Gynecology and Obstetrics, H-Heine Univ of Duesseldorf, Duesseldorf, Germany. Objective: Endometrial homeostasis, the balance between cell proliferation and programmed cell death (apoptosis) in the cyclic endometrium, is regulated by sexual steroid hormones and growth factors. Disregulation of this balance probably plays a determinant role in endometrial neoplastic transformation. Directed induction of endometrial apoptosis by steroid hormones, anti-hormones and Selective Estrogen Receptor Modulators (SERMs) may offer therapeutic benefits. Clarifying the mechanism of action of SERMs like raloxifene, a promising drug for postmenopausal hormone therapy, on the endometrium is therefore of particular interest. Design: Electron microscopical (transmission and scanning electron microscopy, TEM and SEM) and immunohistochemical survey of proliferation, differentiation and apoptosis markers to characterize the effects of raloxifene on cell morphology, proliferation and programmed cell death of human endometrial carcinoma cell line RL95-2. Materials/Methods: RL95-2 continuous cell line of human endometrial adenocarcinoma (39 passage) was studied in the presence of either estradiol17␤ (E2), raloxifene (RLX) or vehicle alone (ethanol: control cells) for 1, 2, 4 and 7 days. The expression of PCNA, Ki67; ER␣, PR, lactoferrin; Caspase3, M30Cytodeath, Bcl-2 and Bak was immunohistochemically detected applying monoclonal antibodies and BioStrept-AlPase detection system on RL95-2 cells, grown on microscopic slides. Semithin and ultrathin sections of cells embedded in Spurr’s resin were analyzed light and electron microscopically (TEM and SEM). To calculate the percentage of immunopositive cells by light microscopy, immunolabelled cells were counted out of 300 randomly selected RL95-2 cells from two microscopic slides at 400⫻ magnification to calculate the percentage (rate) of immunoexpression. Results: Raloxifene exposition reduces the proliferation rate (PCNA, Ki67) in endometrial adenocarcinoma cell line RL95-2 after 2 and 4 days and the expression of anti-apoptotic Bcl-2 (under 10%), whereas DNA fragmentation (M30Cytodeath, Caspase3) and expression of pro-apoptotic Bak (over 56%) are enhanced. The expressions of differentiation markers ER-␣, PR and lactoferrin are up-regulated and even higher than under E2. Randomly distributed apoptotic cells with condensed cytoplasm and darkly stained, fragmented nuclei are found only single in the controls (5– 6%, and no more than 9% after 7 days), whereas they are more numerous (12%) after 2-day and even more abundant (23%) after 7-day RLX-exposition. Electron microscopically, early and late apoptotic bodies (containing condensed nuclear fragments and either recognizable or no longer recognizable organelles) are seen inside as well as outside apoptotic cells. Conclusions: Our findings demonstrate complex effects of raloxifene on the ER␣-positive human endometrial carcinoma cell line RL95-2. Whereas the proliferation rate is reduced, the expression of all three differentiation markers is very high. Furthermore, apoptosis rate is significantly enhanced and is paralleled by up-regulated expression of pro-apoptotic Bak and down-regulated expression of anti-apoptotic Bcl-2 proteins. In conclusion, our findings support the view that raloxifene is a competitive antagonist of estrogen action in the endometrium.

P-305 Expression of ganglioside GD2/GM2 synthase mRNA in mouse oocytes and early embryos. S. Kawachiya, T. Kaneko, T. Takahashi, T. Seino, H. Saito, H. Kurachi. Yamagata Univ Sch of Medicine, Yamagata-shi, Japan. Objective: Gangliosides are the glycosphingolipids that have sialic acid in their structure. Their biological functions have been considered as cell-cell recognition, differentiation and proliferation. In reproductive organs, gangliosides were also detected by thin layer chromatography or immunohistochemical staining with monoclonal antibodies. The expression and the function of the gangliosides in the female mouse have not yet been clarified. In our previous study, expression of ganglioside GD2 was superior to other gangliosides in mouse ovarium. We examined the expression of GD2/GM2 synthase mRNA in mouse oocytes and early embryos. Design: RT-PCR using GD2/GM2 synthase specific primer was performed. Materials/Methods: Oocytes and early embryos were obtained from 6 week old female B6C3F1 mice stimulated by PMSG and hCG with an interval of 48 hrs. Fourteen hours after hCG injection, oviducts were obtained for unfertilized oocytes retrieval. Superovulated female mice were

S212

Abstracts

mated with the male mice to obtain the early embryos, and the oviducts were extracted from 24 hours to 72 hours after hCG injection. Total RNAs were extracted from oocytes and early embryos by a modified guanidine method. Single-strand cDNAs were synthesized using the total RNA with the oligo (dT) primer. PCR was performed with mouse ␤ 1,4 N-acetylgalactosaminyl-transferase (GD2/GM2 synthase) specific primer. Results: Eight cell stage embryos expressed GD2/GM2 synthase mRNA, but unfertilized oocytes, 2 cell- and 4 cell-stage embryos did not express GD2/GM2 synthase mRNA. Conclusions: We detected the expression of ganglioside GD2/GM2 synthase mRNA in mouse 8 cell embryos. Ganglioside GD2 might contribute to the development of early stage embryo in mouse.

P-306 Expression of ER ␣ and ER ␤ in human granulosa-lutein cells. K. Pagidas. Women & Infants Hosp, Brown Univ Sch of Medicine, Providence, RI. Objective: The autocrine role of intra ovarian estrogen, especially 17 ␤ estradiol in human folliculogenesis remains controversial. It is presumed to act via a receptor-mediated pathway. The role of the classic estrogen receptor, ER ␣ in human follicular development remains questionable. The discovery of a second estrogen receptor subtype, ER ␤ has raised great interest regarding the physiological roles of these two receptors in human follicular development. The purpose of our study was to determine the pattern of expression of ER ␣ and ER ␤ in human granulosa-lutein cells. Design: A prospective study of human granulosa cells obtained from pre-ovulatory follicles of women undergoing IVF. Materials/Methods: Granulosa-lutein cells were isolated and purified from follicular aspirates using Ficoll density and gravitational sedimentation techniques. Oocyte-cumulus complexes were isolated using a dissecting microscope. The granulosa cells were kept in culture with frequent media change for 48 –72 hours. The modulation of the expression of ER ␣ and ER ␤ by FSH, LH and ICI 164,384 was assessed. All experiments were done in duplicate, both different time order and dilutions were carried out. Two different clones of ER ␣ and ER ␤ were used, ER88 (BioGenex), ID5 (Immunotech), ER ␣ (ABR) and ER ␤ (ABR). The ER subtypes were characterized using immunocytochemical and Western blot analysis. Grading of positive nuclear staining was done using a scale of 0 –5. Results: Localization of ER ␤ was noted in the granulosa cell nuclei at the time of harvest (time 0). ER ␤ immunostaining was noted to be low, grade 2. A dose dependent increase in immunostaining was noted with increasing doses of antibody used, grade 3– 4. Using Western blot analysis a positive band was noted at the 55 KD region suggestive of a weakly positive ER ␤ signal. ER ␣ also localized to the nuclei of the granulosa-lutein cells at even lower concentration than ER ␤. A single protein band at 61 KD region was noted using Western blot suggestive of an extremely weak signal for ER ␣. No difference in the pattern of expression of ER ␣ or ER ␤ was noted between the mural and cumulus granulosa cells isolated. Pre-incubations with FSH for 24 and 48 hours increased the ER ␤ expression,LH or ICI 164,384 resulted in disappearance of the ER signals. Conclusions: Our results demonstrate that ER ␣ and ER ␤ are present in human granulosa-lutein cells at time 0, yet at very low levels. This very low level of expression of the ER is likely secondary to a very high concentration of estrogen present at the time of harvest and an LH priming effect in the down regulation of the ER at time 0. The modulation of the ER ␣ and ER ␤, in human granulosa-lutein cells in culture, by both estrogen and anti-estrogen is currently under study. Supported by: Departmental Research Funds.

P-307 Estrogen receptor ␤ is expressed in human endometrium and levels of expression vary during the menstrual cycle. R. M. Hearns-Stokes, C. Mayers, J. H. Segars, P. Stratton, J. Gustafsson, L. Nieman. Combined Fed Program in Reproductive Endocrinology and the Pediatric & Reproductive Endocrinology Branch/NIH, Reisterstown, MD; Pediatric and Reproductive Endocrinology Branch, NICHD/NIH, Bethesda, MD; Combined Fed Program in Reproductive Endocrinology and the Pediatric & Reproductive Endocrinology Branch/NIH, Bethesda, MD; Karolinska Institute, Huddinge, Sweden.

Vol. 76, No. 3, Suppl. 1, September 2001

Objective: The recent discovery of a second distinct estrogen receptor (ER) ␤, in addition to the previously studied ER ␣, suggests the possibility of dual regulation of estrogen action in steroid-dependent tissues and the possibility for targeted therapeutics. Currently, there are conflicting data in the literature regarding the expression of ER ␤ protein during the human menstrual cycle. The purpose of this study was to characterize the temporal and spatial localization of ER ␣ and ER ␤ in the endometrium of normally cycling women throughout the menstrual cycle. Design: Prospective cohort analysis with immunohistochemical analysis. Materials/Methods: Healthy ovulatory women (n ⫽ 15) gave informed consent for an endometrial biopsy. The paraffin-fixed tissue was stained with hematoxylin and eosin and dated using the Noyes’ criteria. Samples underwent immunohistochemical staining for ER if this histological date was within two days of a chronological date, assigned as the proliferative phase day from menses or the luteal phase day from the LH surge. Antibodies used were monoclonal antibody NCL-ER-6F11 (Vector Laboratories, Burlingame, California) directed against ER ␣ (1:50) and a polyclonal antibody directed against ER ␤ (1:250) previously described by Dr. J.-A. Gustafsson. Results: As noted in previous studies, ER ␣ had a nuclear localization which was greatest in the epithelial cells during the proliferative phase, with a gradual decrease in staining during the luteal phase, to a minimum around days 26 –28. ER ␤ showed a similar nuclear and epithelial cell preference, with consistently less stromal cell staining. However, both epithelial and stromal cells showed cytoplasmic staining, except from days 19 –22, when cytoplasmic staining was virtually absent. Additionally, stromal cells tended to show an increase in nuclear ER ␤ at days 19 –22, while epithelial cells tended to have consistent ER ␤ staining at all stages in the cycle. Conclusions: ER ␤ and ER ␣ show different patterns of staining during the menstrual cycle, with nuclear down-regulation of ER ␣, but not ER ␤ during the mid-luteal phase, and down-regulation of cytoplasmic ER ␤ at that time. This differential expression may reflect differential roles for the two receptors.

REPRODUCTIVE LABORATORY TECHNOLOGY P-308 Cryopreservation of ooplasmic fragments increases the amount and availability of a material for an ooplasm donation. D. I. Dozortsev, P. T. Goud, A. P. Goud, K. T. McGinnis, M. P. Diamond. Dept of Obstetrics and Gynecology, Wayne State Univ, Detroit, MI. Objective: There is evidence that human eggs differ in their ability to result in term pregnancy following fertilization with spermatozoa; This difference is responsible for pregnancy failure in some cases. At the same time it has been shown that developmental potential of a low quality egg can be increased by the injection of a small amount of cytoplasm from a good quality egg. The major limitation of this promising approach is that it is necessary to synchronize the menstrual cycle of the donor of the egg cytoplasm and the recipient, and also, one egg donor can only help one egg recipient. Another possibility is freezing of the donor eggs. This would avoid the necessity for synchronization, but still one egg donor would only benefit one egg recipient. In addition, an egg survival after freezing-thawing is still problematic. In this study we examined the hypothesis that splitting an egg into fragments will result in particles that, due to the smaller size, will survive freezing better than an entire egg and that this approach might allow the use of one eggs for multiple recipients without synchronization. Design: Eggs were divided into fragments of varying size using micromanipulation techniques. Fragments were cryopreserved, thawed and scored for survival using morphological criteria. Materials/Methods: Eggs were obtained from B6D2/F1 mice following superovulation. Using micromanipulation a slit was made in the zona pellucida. Following this, using a blunted pipet one or several ooplasts 10 –15 m in diameter were removed and placed back under the zona pellucida, so that the small and the large fragments (the bulk of an egg) were formed and served as a control for each other. Cryopreservation/thawing was achieved using slow freeze-rapid thaw protocol in M2 media (Biol Reprod 1993; 48:606 –12). Survival was assessed using morphological criteria 2 hours after thawing. Results: A total of 102 large and 114 small fragments were created. Of these, 87 large and 87 small fragments were recovered. Out of those recovered, 57 large fragments and 68 small fragments survived.

FERTILITY & STERILITY威

Conclusions: Although, it appears that smaller fragment survive better, the difference in these experiments did not reach statistical significance. Remarkably, the nearly 20 fold volume reduction just marginally improved the freezing results. At the same time, because at least 20 fragments or (for human eggs probably even more) can be obtained from one egg, the cumulative survival rates for all small fragments can be as much as 20 times or even higher than for 1 entire egg. Thus, this approach might be a viable option in ooplasm donation, particularly, because our preliminary data show, that micro-ooplasts can be successfully fused back at a very high rate. Supported by: Department of Obstetrics and Gynecology, Wayne State University, Detroit, Michigan.

P-309 Comparison of two sperm counting chambers: microcell and standard count. H. Kobayashi, P. Ranganathan, N. Park, J. Chae, A. M. Mahran, A. Agarwal. The Cleveland Clinic Foundation, Cleveland, OH; Pusan National Univ, Pusan, Korea. Objective: Semen analysis remains an essential test for evaluation of male infertility. The main components of semen analysis are sperm count and motility. MicroCell counting chambers are proven devices for accurate measurement of sperm characteristics, while Standard Count was introduced recently. The purpose of this study was to compare the accuracy of MicroCell and Standard Count chambers in measuring sperm count and percent motility. Design: Laboratory experiment comparing MicroCell and Standard count counting chambers for semen analysis. Materials/Methods: Semen specimens from 7 normal donors and 14 infertile patients were analyzed after 30 minutes of liquefaction on MicroCell (Conception Technologies, San Diego, CA) and Standard Count chamber (Mid Atlantic Diagnostics, Inc., Medford, NJ). Both counting chambers are disposable and have a fixed preparation depth of 20 ␮m. Five microliters of semen were loaded on the chambers and analyzed both manually and by a computer assisted semen analyzer (CASA, Motion Analysis Corporation, Model VP 110, Santa Rosa, CA). The two methods were compared with paired t-tests and Lin’s coefficient of concordance. Results: CASA results for Standard Count were significantly lower than MicroCell (mean difference 3.6 ⫻ 106/mL; p ⫽ 0.03. Using Standard Count, manual counts were on average 3.8 ⫻ 106/mL higher than CASA (p ⫽ 0.04). The manual counts between the two chambers were not significantly different (p ⫽ 0.07), however, Standard Count averaged 3.6 ⫻ 106/mL lower than MicroCell. In contrast, motility estimates were quite comparable between the two, with concordance of ⫹0.95 using manual methods. Conclusions: MicroCell counting chambers and Standard Count analysis chambers are precise in their measurement of sperm concentration and motility. Standard Count can also be used successfully in an Andrology Laboratory for semen analysis purposes. Supported by: This study was supported by a grant from the Cleveland Clinic Foundation.

P-310 Does the choice of catheter for preembryo transfer influence pregnancy rate? Z. Ye, N. Zaninovic, K. Xu, L. L. Veeck. Weill Medical Coll of Cornell Univ, New York, NY. Objective: Appropriate delivery of preembryos to the uterine cavity is crucial to successful in-vitro fertilization (IVF). Although the type of catheter used for transfer may impact pregnancy rate, the choice is generally program-dependent and based on the previous experience of the operator. This study compares cycle outcomes using Wallace and Tomcat catheters, and examines whether assisted hatching (AHA) is associated with negative findings in either group. Design: Retrospective analysis of patients in a university, hospital-based IVF clinic from May 1995 to December 2000. Materials/Methods: In 2,426 cycles, we compared the Edwards-Wallace catheter (Sims Portex Ltd., Kent, UK) and Tomcat catheter (Sherwood Medical, St. Louis, USA). To control for patient variation, all women included in the study were less than 38 years old, possessed more than three fertilized oocytes, and had two to four conceptuses available for transfer, at

S213