Systematic analyses of transcripts in early embryogenesis by the group of Dr. M. Ko in our Laboratory have uncovered a relative dearth of X-linked genes. Nevertheless, some X-linked genes show prominent effects in early development. Three of these have been chosen for special study. One, Bex-3, had not been previously studied. The transcribed sequence is 789 bp in length and encodes a protein of 124 residues with a CAAX motif at its C-terminal end. It appears in a single cDNA and Northern species of 1.1 kb. Expression declines after the fetal period, but adult mouse and human show sustained expression in several organs, including brain, testis and ovary. We find Bex3 protein selectively associated with the subset of mitochondria that are actively replicating their DNA; Bex3 is the first marker for a stage in the mitochondria cycle. Its association is consequential, because siRNA experiments show that Bex3 is required for continued growth of target cells. A second gene, GPC3, causes the overgrowth Simpson-Golabi-Behmel Syndrome when it is interrupted; and we have shown that a mouse defective for the gene shows comparable effects on growth. A third gene, PLAC1, has important features that include: 1) expression with high specificity in placenta from early on in development; and 2) X-linkage very near the HPRT gene, at a location that makes it a candidate for involvement in placental dysplasia and placentomegaly when it is under- or overexpressed. It is one of very few genes that show very greatly increased activity in cloned animals, associated with placentomegaly. We have analyzed the structure of the cDNA and genomic form of PLAC1, and have shown that it is localized in giant cells and labyrinthine trophoblasts in the developing placenta, between 7.5 and 14.5 days post coitum. In additional studies with collaborators, we have found that the gene is comparably expressed in human placenta, but strong expression lasts longer, all the way to term. Currently we have started to analyze the basis for tissue-specific expression of the gene. Extensive transfections with constructs containing increasing amounts of DNA upstream of the transcription unit, fused to luciferase have thus far shown that neither 3.3 nor 7.5 kb of upstream DNA is sufficient to support the transcription of the gene in choriocarcinoma cells that express the endogenous PLAC1 gene. We interpret our results to mean that the specific expression of the gene involves more distant upstream (or intragenic!) elements, and are continuing to study increasingly large genomic regions to define regulatory elements more precisely; that work is being continued in another project.
