thickness was measured at cycle day 8, hCG day, transfer day, and seven days after embryo transfer. Progesterone and estradiol levels were taken on the day of hCG administration, and the size of the leading follicle was measured. The number of thawed embryos and transferred embryos, pa- tients’ demographics and the pregnancy rate were recorded. Institutional review board approval was obtained. RESULTS: Age and number of previous unsuccessful IVF cycles were similar between the two groups. Endometrial thickness was significantly higher in the letrozole group compared to the CC group at the hCG day of (9.13.6mm vs. 6.91.2mm, p=0.01) and at the embryo transfer day (101.7mm vs.7.61.4mm, p=0.01). Endometrial thickness at day 8 of the cycle and one week after embryo transfer was not significantly different, although there was a trend toward a thicker endometrium in the letrozole group. Estradiol levels in the letrozole group were significantly lower than in the CC group at the day of hCG (231132 pg/l vs. 515363pg/l, p=0.05). Progesterone levels at the hCG day and the numbers of embryos thawed and transferred were not different between the groups. Pregnancy was not achieved in either group, and the study was closed after publication of the manufacturer’s warning cautioning of letrozole usage for infertility treatments. CONCLUSION: Endometrial thickness is increased by letrozole com- pared to CC during frozen-thawed embryo transfer cycles. Whether letro- zole treatment has an effect on pregnancy rate awaits further investigation pending drug approval for infertility treatment. Supported by: None P-392 THE ENDOMETRIAL EXPLANT CULTURE SYSTEM: A PROM- ISING MODEL FOR THE EVALUATION OF ENDOMETRIAL RE- CEPTIVITY. R. H. Fogle, A. Li, J. K. Jain, R. J. Paulson. Univ. of Southern California Keck School of Medicine, Los Angeles, CA. OBJECTIVE: In vitro evaluation of endometrial receptivity remains elusive as cell culture systems lack the physiologic interactions between stroma, glands, and epithelium. In contrast, an explant model maintains these relationships and has the potential to encompass both paracrine and endocrine responses of the endometrium. Our aim was to establish a viable endometrial explant culture model for the study of endometrial receptivity. To this end, we subjected our system to varying levels of human chorionic gonadotropin (hCG) in an effort to demonstrate a vascular endothelial growth factor (VEGF) response in vitro since it has previously been dem- onstrated that hCG increases VEGF secretion in vivo. DESIGN: In vitro endometrial explant culture. MATERIALS AND METHODS: Endometrium was obtained by biopsy from two groups of subjects. Group A (42 8 yrs, n=10) consisted of prospective recipients of oocyte donation who received oral estradiol and vaginal progesterone in cyclical fashion, with an endometrial biopsy per- formed on the 7 th day of progesterone. Group B (26 5 yrs, n=4) consisted of oocyte donors, who underwent endometrial biopsy during follicle aspi- ration after standard controlled ovarian hyperstimulation and hCG admin- istration. Endometrial samples were cut into 1 mm 3 pieces and cultured in D-MEM/F-12 medium supplemented with estradiol (10nM) and progester- one (100nM), without (control) or with various dosages of hCG (10, 100, or 1000 ng/mL), on Millicell-CM culture inserts for 24 hours. Immunohisto- chemistry (IHC) was used to assess explant viability and evaluate VEGF localization in response to hCG (n=5). VEGF concentration in the culture supernatant was measured using ELISA (R&D Systems, MN). Student’s t-test was used for statistical analysis. RESULTS: Explant viability at 24 hours was confirmed histologically, as well as by staining with Ki-67, a nuclear antigen present only in proliferating cells. VEGF staining by IHC was noted in glandular epithelial cells in tissue examined before and after culture. VEGF staining intensity increased in both glands and stroma in hCG treated explants from 2 of 5 patients. Mean VEGF concentrations in the supernatant were increased after culture with 1000 ng/mL hCG for group A (416 231 vs 588 386 pg/mL, p0.05). A similar trend was noted for group B (322 196 vs 379 162 pg/mL, p=NS). Mean VEGF levels from tissue cultured at 10 ng/mL (507 558 pg/mL) and 100 ng/mL (412 520 pg/mL) did not differ significantly from each other or from the control and 1000 ng/mL groups; however the sample sizes were small (n=5). There was no difference in mean VEGF concentration between groups A and B for any hCG concentration (p=NS). CONCLUSION: 1) The endometrial explant system is a promising short- term model for the in vitro evaluation of endometrial receptivity. 2) The explant culture system, by retaining the potential for interactions between stroma, glands, and epithelium, may prove superior to cell culture. 3) Our findings of increased VEGF secretion in response to hCG are consistent with in vivo data, substantiating the ability of the system to model physi- ologic response at the time of implantation in an intact uterus. Supported by: None P-393 C-JUN N-TERMINAL KINASE (JNK) SIGNALING REGULATES CELL PROLIFERATION AND APOPTOSIS IN ENDOMETRIAL CELLS. G. Kizilay, M. Basar, H. Cakmak, C. Atabekoglu, U. Kayisli, A. Arici. Dept. of Obstetrics, Gynecology & Reproductive Sciences, Yale Univ. School of Medicine, New Haven, CT; Dept. of Histology and Em- bryology, Trakya Univ. Medical Faculty, Edirne, Turkey. OBJECTIVE: Mitogen-activated protein kinases (MAPK)s regulate many cellular and molecular activities ranging from cytokine expression, cell proliferation to apoptosis. c-Jun N-terminal kinase (JNK) is one of the main subfamilies of MAPK superfamily. JNK is activated in response to stress and modulate inflammation and apoptosis. Our hypothesis is that temporal and spatial changes in JNK activity regulate cell survival and inflammation throughout menstrual cycle and early pregnancy. Therefore, we studied to determine total- and active- (phosphorylated-) JNK expression in endometrial tissues in vivo and its impact on endometrial cell prolifera- tion and apoptosis, in vitro. DESIGN: A prospective in vitro and in vivo study assessing activity and function of JNK in endometrium. MATERIALS AND METHODS: Serial sections from endometrial (n=30) tissues were stained with phosphorylated (p-) and total (t-) JNK and evaluated semi-quantitatively using HSCORE system. Endometrial stromal and glandular cells were isolated from endometrial biopsies obtained from women undergoing surgery for benign gynecologic conditions. Endometrial stromal and epithelial cells were cultured in DMEM containing 10% FBS, and grown to confluence. Experiments were performed in 96-well micro- plates and 4-well chamber slides after incubating cells in serum-free, phenol red-free DMEM for 24 h. Thereafter, cells were incubated with the JNK inhibitor [SP600125; 10 M] alone or with estradiol (E2; 10 -8 M) for 24-72 h and evaluated with BrdU cell proliferation and TUNEL apoptosis assays. Statistical analysis of the data was performed using ANOVA. RESULTS: Human endometrial tissues were grouped according to the menstrual cycle phase and examined by immunohistochemistry for t-JNK and p-JNK. Both stromal and glandular cells revealed moderate cytoplasmic and nuclear t-JNK immunoreactivity without significant changes throughout the menstrual cycle. On the other hand, mostly nuclear immunoreactivity was detected in endometrial cells for p-JNK. The highest p-JNK immuno- reactivity was detected in late secretory endometrial epithelium and stroma (p0.03) compared to other cycle phases. Interestingly, the lowest p-JNK level was detected in mid-secretory phase endometrial samples. In culture, SP600125-tretead endometrial stromal cells showed a 3-fold higher apopto- sis ratio after 72 h of incubation (p0.05). Moreover, SP600125 treatment significantly lowered the DNA synthesis in both endometrial glandular and stromal cells (60% and 80% less than control cells, respectively) as assessed by BrdU-incorporation assay. When combined with E 2 , SP600125 signifi- cantly reduced the E 2 -stimulated BrdU-incorporation (p0.05). CONCLUSION: Increased expression of p-JNK during the late-secretory phase suggests that JNK signaling may be involved in regulation of inflam- matory processes in endometrium. Moreover, our in vitro results also support that JNK signaling may directly regulate endometrial cell survival by affecting proliferation/apoptosis ratio. Supported by: None P-394 RETINOIC ACID MODULATES MONOCYTE CHEMOTACTIC PROTEIN-1 AND LEUKEMIA INHIBITORY FACTOR BUT NOT INTERLEUKIN 1-BETA OR INTERLEUKIN 6 IN ENDOMETRIAL CELL CULTURE. B. Ozsait, S. Bulgurcuoglu, S. Bilgic, P. Balcik-Ercin, H. Serdaroglu, E. Attar. Istanbul Univ., Istanbul, Turkey. OBJECTIVE: To evaluate the in vitro effects of all-trans retinoic acid (ATRA) on endometrial cell cytokine production. FERTILITY & STERILITY S281