In mammals, dosage compensation at X-linked loci is achieved by the process of X-chromosome inactivation in the homogametic sex. Since this process is usually random, in a given 46, XX cell each X chromosome has a 50% chance of becoming inactive. Thus, in a normal population the relative proportion of inactive Xm (maternal X chromosome) versus inactive Xp (paternal X chromosome) tends to follow a Gaussian distribution, where most females exhibit a 50: 50 to 80: 20 pattern. In mice, skewed X-inactivation has been associated with the X-controlling element (Xce) locus [Cattanach et al., 1969]. Xce loci of different “strength” have been described in mice. Animals heterozygous for this locus will preferentially inactivate the X chromosome bearing the “weaker” Xce allele. The best known genetic element involved in the initial events of X inactivation is the X-inactivation center (Xic). The XIST gene, which maps to this region, is expressed exclusively from the inactive X chromosome [Brown et al., 1991]. Studies in mice indicate that this gene is necessary and sufficient for X-inactivation and that it has a role in the process of X chromosome counting and choosing [Penny et al., 1996; Herzing et al., 1997; Marahrens et al., 1997]. In humans, the proportion of normal females with skewed X-inactivation reported in different population studies range from 1% to 33%[Puck et al., 1992; Fey et al., 1992; Gale et al., 1991; Gale et al., 1992; Sumita et al., 1998]. Some rare families where several females have a highly skewed X-inactivation pattern (95: 5) have been reported [Pegoraro et al., 1995]. Women in these families present the same X chromosome preferentially inactive, suggesting the influence of a genetic factor in the choosing process of X-inactivation. A citosine to guanine (C43G) mutation in the XIST minimal promoter region has been associated with skewed X inactivation in two unrelated families [Plenge et al., 1997]. X chromosomes bearing this mutation were preferentially inactive in female carriers. In an attempt to confirm this association, we have analyzed the C-43G mutation in the XIST gene in a sample of women, all with skewed X-inactivation patterns in blood ranging from 80: 20 to 100: 0. These women, 32 Duchenne/Becker (DMD/BMD) muscular dystrophy carriers and 34 normal women controls, were selected from a larger sample assessed recently for X-inactivation status [Sumita et al., 1998]. Among the DMD/BMD carriers, two had clinical signs of muscular dystrophy. X-inactivation patterns were determined in blood DNA by methylation analysis of the androgen receptor (AR) locus as described previously [Edwards et al., 1992; Allen et al., 1992]. Analysis of the C43G mutation was performed as described by Plenge et al.[1997]. None of the 66 analyzed females had this mutation.

Plenge et al.[1997] suggested that the C43G mutation may be rare in the general population. However, since in the present sample all females had a skewed pattern in blood one would expect to find a higher frequency of genetic elements influencing non-random X-chromosome inactivation. Moreover, our sample included two manifesting DMD carriers, indicating that, in these individuals, other tissues such as skeletal muscle also had a skewed pattern of X-inactivation. Thus, our results suggest that, if the C43G mutation in the promoter region of the XIST gene is associated with X-chromosome inactivation, it is a rare cause of skewing, at least in the haematopoietic system. In that case, in most skewed females other genetic factors may be responsible for non-random X-inactivation. Alternatively, it is also possible that the C43G mutation in the promoter region of the XIST …