RT-PCR analysis of Crg1 were done in series mouse embryos, some adultmouse tissues, mouse embryo stem cell and mouse embryo fibroblast.

Results: 27 ESTs (expressed sequence tags ) of genes expressed differ-ently between early 8-cell embryos and 8-cell compacted embryos havebeen isolated and cloned. 17 of those were novel ESTs and they werebanked into GenBank with accession numbers: BQ740249,BQ740250,BQ740251,BQ740252,BQ740253,BQ740254,BQ740255,BQ740256,BQ740257,BQ740258,BQ740259,BQ740260,BQ740261,BQ740262,BQ740263,BQ740264,BQ740265. Length of Crg1 is 810bp. It include onlyone exon and code a 150aa protein with a theoretical molecular weight of17670.34 Dalton. The protein is similar to a protein encoded by a knowngene, Stella. Crg1 was mapped to chromosome 14 by database analyses.RT-PCR analysis shows that Crg1 expressed in series mouse embryos (2-cembryos, 4-c embryos, 8-c embryos, compacted embryos and blastocyst),and expressed little higher in compacted embryos. It also expressed inmouse embryo stem cell but not expressed in mouse embryo fibroblast. Itexpressed in some adult mouse tissues (kidney, testis, epididymis, ovary,liver and lung) too but not expressed in other adult mouse tissues (brain,spleen, heart, skeletal muscle).

Conclusion: All 17 ESTs might be for novel genes related to compactionin compacted embryos. And longer ESTs may obtained by cloning accord-ing to the size. Crg1 may related to compaction in compacted embryos andmaintaining cell’s pluripotentiality.

Wednesday, October 15, 20034:00 P.M.

O-208

Fusion of human gametes requires a CD9 dependent clustering of a6b1integrin-CD151 complexes on the oocyte membrane. Jean-PhilippeWolf, Frederique Monier-Gavelle, Eric Rubinstein, Virginie Barraud, Mor-gane Bomsel, Claude Boucheix. UFR SMBH, Univ Paris 13, Bobigny,France; Inserm U 268, Institut Andre Lwolf, Villejuif, France; Inserm 567,Paris, France; Inserm U268, Villejuif, France.

Objective: The a6b1 integrin is the putative binding site of the spermfertilin in mouse oocytes (Almeida et al., 1995). Furthermore mouse thathave been deleted from the tetraspan CD9-/- gene are infertile because ofthe oocyte’s membrane inability to fuse with that of the spermatozoa (LeNaour et al., 2000). We investigated whether these molecules could also beinvolved in human gamete interaction or not.

Design: Unfertilized oocytes were given by patients undergoing ARTprocedures. Test of fusion with human gametes were performed after zonapellucida removal in order to comply with french Ethic laws.

Materials and Methods: The zona pellucida were removed mechanicallyin order to prevent any membrane protein degradation. Tests of fusion wererun in presence of specific antibodies. Immunofluorescence detection wereperformed by incubating oocytes with a primary antibody followed byfixation with PFA 4%. Oocytes were then incubated with the secondantibodies and detection was performed using an Axiophot immunofluores-cent microscope. Controls were performed by omission of the first antibod-ies.

Results: We have previously shown that an anti-b1 integrin mAb partiallyinhibits sperm-zona free egg fusion (Ji et al; 1998). a6 and b1 integrinsubunits colocalize on the zona free human oocytes suggesting the existenceof the a6b1 integrin heterodimers on the human egg. GoH3, a mAb blockingthe a6 integrin subunit, almost entirely inhibited sperm-egg fusion. A mAbto CD151, a tetraspanin tightly associated with the integrin a6b1 induced apartial dose dependent inhibition of fusion.

The distribution of CD9, CD151 and a6b1 integrin at the intact humanoocyte plasma membrane was homogenous. Mechanical removal of zonapellucida induced the co-patching of a6b1 integrin and CD151, but had noeffect on CD9 distribution. However, addition of anti-CD9 mAb prior tozona removal prevents a6b1/CD151 co-patching and gamete fusion whereasit has no effect when the patches are already formed in zona pellucida freeoocyte, suggesting that CD9 controls the formation of these patches. This isfurther supported by the inability of intact CD9-/- mouse oocytes to form a6patches when inseminated with spermatozoa, in contrast to wild typeoocytes.

Altogether, these data suggest that CD9 organizes a6b1/CD151 patcheson human and mouse oolemma, and that in the human, a6b1/CD151 patchesare a prerequisite for gamete fusion.

Conclusions: Data provide evidences that the fertilizing sperm induces afusiogenic patch at the human oolemma and that inhibition of its formationresults in fertilization failure.

Wednesday, October 15, 20034:15 P.M.

O-209

DNA methylation patterns in mouse and human zygotes. Yanwen Xu,Lewis Krey, Jamie Grifo, John Zhang. NYU Sch of Medicine, New York,NY.

Objective: Epigenetic modification during embryogenesis is character-ized by selective gene expression achieved through methylation of 5-meth-ylcytosine. In mammals dynamic reprogramming of DNA methylationoccurs during gametogenesis and embryogenesis. Changes in environment,e.g., in vivo vs in vitro culture, can alter the process of epigenetic modifi-cation. Staining of mouse zygotes with an antibody against 5-methylcy-tosine (Anti-MeC) has revealed that by 4-6 h post-fertilization the chroma-tids in the male pronucleus (PN) have been actively demethylated. Thisstudy examines whether a similar pattern of DNA methylation takes placein human zygotes.

Design: Normally fertilized (2PN) mouse and abnormally fertilized hu-man (3PN) zygotes were examined with Anti-MeC immunofluorescencestaining. Sex chromosome status, assessed by fluorescence in situ hybrid-ization (FISH), was used to identify the parental origin of each pronucleus.

Materials and Methods: Mouse zygotes were obtained from in vitrofertilization of oocytes harvested from oviducts 16 h post-hCG. Abnormallyfertilized zygotes (3PN) were obtained from patients undergoing conven-tional IVF or ICSI with patient consent (IBRA Protocol H 6092). Eachzygote was fixed with methanol:acetic acid (3:1). DNA methylation patternswere detected in mouse zygotes by immunofluorescence staining with anAnti-MeC antibody (Barton et al 2001, Human Molecular Genetics, 26,2983-2987). Human pronuclei were subjected to FISH to identify the X andY chromosomes prior to staining with the anti-MeC antibody.

Results: Staining patterns in mouse revealed active demethylation be-tween sperm head decondensation and male PN formation. In contrast to thedecline in paternal chromatid staining intensity, maternal chromatid stainingremained strong throughout PN formation. All 19 conventionally insemi-nated human zygotes displayed one strongly and two weakly stained PNs.All 21 PNs with a Y chromosome stained poorly as did 17 of 34 PNs withan X chromosome, most likely those of paternal origin. In contrast, 8zygotes from ICSI were digynic, displaying two strongly and one weaklystaining PNs. Interestingly, an ICSI zygote displayed one strongly and twoweakly fluorescing PNs, but it was diandric with 2 Y chromosomes. How-ever, 2 ICSI zygotes presented 3 brightly stained PNs with no differences inintensity; these zygotes were X/X/Y and XX/X/O.

Conclusion: Selective, active demethylation of the male PN occurs inboth mouse and human zygotes. However, abnormal methylation patternscan occur in some zygotes and may indicate a dysfunctional ooplasm. A fullinvestigation of the methylation patterns in donated normal embryos willprovide more insights into the epigenetic modifications that occur during invitro embryogenesis.

Wednesday, October 15, 20034:30 P.M.

O-210

Reproductive outcomes in patients with recurrent pregnancy loss as-sociated with a structural chromosome abnormality. Sony Sierra, SylvieLanglois, Mary D. Stephenson. Dept. of Obstetrics & Gynaecology, Univof Toronto, Toronto, ON, Canada; Dept of Medical Genetics, Univ ofBritish Columbia, BC Women’s Hosp & Health Ctr, Vancouver, BC,Canada; Dept of Obstetrics & Gynaecology, Univ of British Columbia, BCWomen’s Hosp & Health Ctr, Vancouver, BC, Canada.

Objectives: Recurrent pregnancy loss (RPL), defined as two or morepregnancy losses, affects up to 5% of couples trying to establish a family.A structural chromosome abnormality is found in association with RPL in

S80 Abstracts Vol. 80, Suppl. 3, September 2003