A geneticist views two theories of X-chromosome inactivation in a broad context.

Female mammals have two X chromosomes, one of which is inactivated to ensure that females get the same dose of X-linked genes as males. Two long, non-coding RNA molecules mediate this process. One, Xist, initiates silencing of the chromosome to be inactivated, whereas its antisense partner, Tsix, blocks such silencing of the remaining X. The exact mechanism by which Xist and Tsix exert their functions is not known.

RNA interference (RNAi) describes a process by which RNA fragments 'interfere' with the creation of proteins from their RNA recipes. The recent discovery of RNAi in mammalian cells made it tempting to postulate that sense and antisense non-coding RNA partners, such as Xist and Tsix, are processed by the RNAi-associated enzyme DICER into small RNAs. Jeannie Lee and her team at Harvard Medical School have presented several lines of evidence in support of this, including data that demonstrate improper X inactivation in DICER's absence (Y. Ogawa et al. Science 320, 1336–1341; 2008).

By contrast, a study by David Livingston of Harvard's Dana-Farber Cancer Institute and his colleagues has shown that, in the absence of DICER, the inactive X chromosome in stem cells remains coated with Xist and several other repressive markers (C. Kanellopoulou et al. Proc. Natl Acad. Sci. USA 106, 1122–1127; 2009). This group suggests that the effects observed by Ogawa et al. may have been indirect because DICER is also involved in the processing of a class of small, non-coding RNAs known as microRNAs that can alter gene expression by fine-tuning protein production.

Despite their differences, these studies should provide an incentive to further investigate the potential role of the RNAi pathway in the nucleus of mammalian cells. This will shed light on both X inactivation and gene regulation in general.

Discuss this paper at http://blogs.nature.com/nature/journalclub