The Xist gene is a candidate for the X inactivation centre and we have recently shown that Xist transcription is necessary for X inactivation in the mouse. Xist transcription is repressed in XX diploid embryonic stem cells and both X chromosomes are active. Upon differentiation one Xist allele becomes expressed, preceding cis inactivation of that X chromosome, while the Xist allele on the active X chromosome remains silent. This is normally a random process, either allele having an equal

more »

... lity of being activated. To identify sequence elements involved with Xist regulation we have surveyed 10 kb of the Xist promoter for DNaseI hypersensitive sites and have assayed their properties using a reporter gene system. Functional elements identified thus far affect reporter gene expression in a complex manner but are not sufficient to permit appropriate expression. A detailed examination of the Xist minimal promoter has revealed several previously uncharacterized transcription factor binding sites. One of these, designated XPF1, is a potential regulatory switch since it overlaps a canonical CAAT box and an SP1 site, and can displace these factors in itro. Some implications of these data regarding the regulation of random monoallelic Xist expression will be discussed. Regulation of the Xist minimal promoter studied by microinjection in preimplantation embryos and in transgenic mice TETSUYA GOTO, ELISABETH CHRISTIANS and MARILYN MONK The mouse Xist gene is expressed only from an inactive X chromosome and plays a role in Xchromosome inactivation in female development and in male spermatogenesis. Differential expression of the paternal Xist in preimplantation development has also been implicated in imprinted paternal X-inactivation in extra-embryonic tissues. To identify important regulatory elements in the Xist promoter, we have made a construct ligating a 233 bp PCRamplified Xist promoter fragment (k220 to j13) to the firefly luciferase reporter gene and the SV40 enhancer. Following injection of the construct into 1cell embryos, luciferase activity is detected at the 2-cell stage. Methylation of the construct by site-specific methylases before injection results in a decrease in luciferase activity. We have also created transgenic lines harbouring the 2n5 kb linearized Xist-luciferase insert. In all six lines established, luciferase is highly expressed in the testis. In contrast, only background luciferase activity is observed in the liver, kidney, muscle, brain and ovary. A DNA methylation study shows that the transgene is undermethylated in the testis compared with other tissues examined. The regulatory sequences involved in promoter activation, imprinting and tissue-specificity are being investigated. Determination of the active chromatin domain of the expressed Xist allele in mouse