ABSTRACT: In most mammalian species, a critical step of placental development is the fusion of trophoblast cell into a multinucleated syncytiotrophoblast layer, which separates the fetal and maternal blood circulations. Advancement in understanding this process came from the identification of two pairs of envelope genes of retroviral origin, independently acquired by the human (syncytin-1 and M-^V2) and mouse (syncytin-A and M-^VB) genomes, specifically expressed in the placenta and with in vitro cell-cell fusion activity. We previously showed that syncytin-A is involved in the formation of one of the two syncytiotrophoblast layer of the mouse placenta -ST-I , facing the maternal lacuna- with syncytin-A null embryos dying at mid-gestation, and their placenta disclosing impaired formation of the ST-I layer. Here by generating syncytin-B KO mice, we demonstrated that syncytin-B null placenta have defects in formation of the syncytiotrophoblast layer-II facing the fetal blood vessels, with refined electron microscopy analyses showing evidence of unfused apposed cells. Lack of trophoblast fusion is associated with signs of trophoblast degeneration and with formation of huge maternal blood lacuna, ultimately disrupting the architecture of the labyrinth layer. These alterations did not result in complete abortion of syncytin-B null embryos, at variance with the syncytin-A KO phenotype, but in a reduced proportion of homozygous syncytin-B knockout individuals at birth, with evidence of late-onset embryonic growth retardation. Double knockout mice experiments demonstrate a premature death of syncytin-A null embryos if syncytin-B is deleted, thus strongly suggesting cooperative involvement of both syncytiotrophoblast layer-I and layer-II for the structural and functional integrity of the maternofetal interface and exchanges. Finally, a microarray analysis of wild-type and syncytin-B null placenta transcripts disclosed alterations in gene expression level for only a very limited number of genes, with the largest change (a 7-fold induction in the syncytin-B null placenta) observed for the connexin-30 gene. Immunohistochemistry further localized the connexin-30 protein in the labyrinth zone of mutant placenta, at the level of the fetomaternal interface. It is proposed that this might correspond to a compensatory process whereby disruption of syncytialization in syncytin-B knockout placenta is counteracted by an enhanced, gap junction mediated, cell-cell communication. These data demonstrate that syncytin-B is required for ST-II syncytial differentiation and the functional integrity of the labyrinth. Altogether, these findings definitively demonstrate that the two endogenous retroviral syncytin-A and -B Env genes contribute independently to the formation of the two syncytiotrophoblast layers during placenta formation, and provide additional insights into the subtle mechanisms of syncytiotrophoblast differentiation in the mouse. Experimental design: Placenta RNA of embryos homozygous for the syncytin-B deletion (SynB-/-) were compared to placental RNA of wild-type embryos (SynB+/+) of the same litter. Two stages of gestation, day 12 and d14, were analyzed. At day 12, 3 litters were obtained; for two of them, the RNAs were analyzed individually, and for the third one, placenta RNAs of the wild-type genotype were pooled. At day 14, one litter was obtained; RNA samples with the same genotype were pooled
