DNA is tightly repackaged during sperm formation, with the histones that normally bind DNA widely replaced by smaller proteins. But a small fraction of early developmental genes remain somehow bound not just to histones but to histones that bear marks for precisely timed transcriptional activation. This finding, published in Nature, suggests that some instructions for proper embryogenesis may already be laid down in the form of epigenetic marks on the genome of sperm1.

A team led by Bradley Cairns of the University of Utah in Salt Lake City collected human sperm and used antibodies to capture modified histones along with any DNA bound to them, a technique known as chromatin immunoprecipitation (ChIP). Among the few genes still associated with histones in mature sperm was a large proportion of transcription factors active early in development, including many associated with patterns of histone modifications known to regulate gene expression in the embryo. Cairns's team was particularly intrigued by the genes associated with bivalent histone marks. These marks are tied to both gene activation and repression, and they poise certain genes to be turned on at the right time during development.

Wolf Reik investigates epigenetic regulation in germ and stem cells at the Babraham Institute in Cambridge, UK. He stresses the need to find out “whether histone marks present in sperm survive widespread epigenetic reprogramming in the early embryo”. Cairns agrees, adding that he would also like to find out whether these marks could serve “as a template to confer the same status on newly deposited histones following [DNA] replication”.

“Histone modifications are so similar in sperm and embryonic stem cells,” says Cairns. “Yet they are also different in some interesting ways.” In mature sperm, for instance, several genes for transcription factors linked to stem cell function, such as Oct4 and Nanog, are not retained on histones and are silenced by DNA methylation. Yet many of the targets for these essential pluripotency regulators are demethylated. A stem cell state is established after fertilization so that the embryo can develop, and Cairns wonders if active demethylation of stem cell transcriptional regulators is all that is needed to accomplish this transition.

According to Reik, Cairns's team described an interesting concept that could lead to a host of functional studies in model organisms, like the mouse. Cairns recognizes that his team's work so far is largely descriptive, but they have already begun work with zebrafish for future studies. They plan to manipulate the levels and types of histone modifications in the sperm and then assess the consequences for embryo development. Cairns suspects that they “will continue to do so for many years, as these are challenging issues even in the fish”.

Related stories

A Nature podcast of Bradley Cairns discussing this paper is available here.