Does the HSD17B10 gene escape from X-inactivation?

We read with great interest the recent report by Garcia-Villoria et al¹ regarding the expression of the HSD17B10 gene from the inactive X chromosome that was published in the European Journal of Human Genetics (Advance online publication, 28 July 2010; doi:10.1038/ejhg.2010.118).

It had been reported previously that a cluster of six genes, including the HSD17B10 (formerly HADH2) gene in Xpl 1.2, escapes X-inactivation.² Subsequently, Carrel and Willard, in a more detailed study,³ showed that the escape of the HSD17B10 gene from X-inactivation is not complete. The expression of the HSD17B10 gene and the surrounding genes from the inactive X chromosome (Xi) is summarized in Figure 1 (adapted from Ref. Yang et al⁴).

Figure 1

Expression of transcripts of the HSD17B10 and surrounding genes from inactive X (Xi) hybrids. Samples scored as positive are expressed at least >10% of the Xa levels, and their number is shown as the numerator. The total number of hybrids tested is shown as the denominator. Genes with mutation(s)⁵ or copy number variation (CNV)⁶ causing mental retardation are marked with asterisk.

Two female patients heterozygous for HSD10 deficiency were the subjects of this present study¹ in which skin fibroblast cultures were examined to determine the inactivation ratio of the normal and mutated X chromosomes. It appears that these studies were performed on cultures originating from a single biopsy from each patient. Mosaicism due to lyonization results in relatively large patches of skin with the same inactivated X chromosome, commonly illustrated by the coloration of calico cats. Thus, an analysis of cells from a single biopsy is probably not adequate to determine the X inactivation ratio. Analysis of a blood sample might be more informative.

In addition, the standard deviation (SD) appears to be relatively large, that is, >15% of the mean value in most cases. This limitation makes it unlikely that partial (12%, Carrel and Willard³) escape of the HSD17B10 gene from X-inactivation in the first female patient would be detectable. Moreover, although monoallelic expression in one of the cell lines indicates that this gene is subject to inactivation, lack of data from other tissue samples makes the inference of widespread monoallelic expression of the HSD17B10 gene in the second female patient less than convincing. The statement that ‘as the girl was severely affected, a similar unfavorable X-inactivation in other tissues could be expected’¹ does not suffice for correcting the defect in data. The conclusion that ‘the HSD17B10 gene does not escape X-inactivation as has been reported previously’ is not adequately supported by the data included in this publication.