Epigenetic what?

The other day I posted on my FaceBook profile that I better hurry up to finish my presentation on epigenetic inheritance. One of my friends commented: “I have no idea what that means, but good luck to you!” Ironically, that is, in part, the point of my presentation: understanding what it all means. Let me explain.

I am in Durham, NC, at the National Center for Evolutionary Synthesis, where — together with two former postdocs of mine — Christina Richards (soon at the University of South Florida) and Oliver Bossdorf (now at the University of Bern), I am running a brainstorming workshop on the meaning and potential importance of epigenetic inheritance.

Epigenetics is an old word, which traces back to Aristotle, and has a long and convoluted history in biology. Today it refers to a panoply of molecular phenomena ranging from methylation patterns of DNA sequences to prions, from so-called “interference RNA” to the tridimensional arrangement in the cell nucleus of a particular class of proteins called histones, which closely bond to and stabilize DNA. What all these things have in common is that in one way or the other they help direct the development of living organisms, turning certain genes on or off in particular cell lineages, and responding to signals from the external environment. The “epi-” in epigenetics stands for a general class of phenomena “beyond” the genes.

But why, you may ask, is an evolutionary biologist interested in epigenetics? Because a subset of epigenetic effects turns out to be heritable across generations. This means that there is something else other than classical genes (i.e., sequences of DNA) that both carries information and is passed from one generation to the next. This is big news for biologists (though the suspicion had been around for a while), because it suddenly broadens and complicates — possibly dramatically — our concept of inheritance, with a wide range of consequences for how we understand evolution. After all, the natural variation among organisms so crucial for natural selection to work had been assumed until recently to originate only from changes in gene sequences. Depending on how much epigenetic inheritance there turns out to be in the living world, the job of biologists will become much more complicated and interesting at the same time (biologists, by nature, like messy stuff, unlike, say, physicists, who always look for simple solutions to simple questions — oh boy, am I going to get in trouble for this one!).

What we are discussing here at Durham, among other things, is precisely how to find out whether epigenetic inheritance is a negligible curiosity or a widespread phenomenon and, if the latter, what consequences it might have for the way we look at evolution, genetics and development. The consensus answer that is already emerging (both from this workshop and from the recent literature) is that epigenetic inheritance as a whole is no fluke, but that different types of heritable epigenetic effects range all the way from very rare (e.g., structural inheritance of cell-surface properties, probably confined to some unicellular organisms) to ubiquitous (e.g., DNA methylation and RNA interference), with others being common in some groups of organisms but absent or rare in others (e.g., a phenomenon called “paramutation,” found in some species of plants).

Skeptics of epigenetic inheritance (of which there are a good number among professional biologists) point out that the empirical evidence is scarce and that the very concept of epigenesis is rather fuzzy. The first objection is becoming less and less tenable. One of the participants to our workshop, Eva Jablonka of Tel-Aviv University, has a huge review paper in press in the Quarterly Review of Biology, in which she details hundreds of known and published examples; and if some of the talks I’ve heard here so far are any indication, there will be much more hard data coming out soon.

As for conceptual fuzziness, my response during the introductory talk I gave at the workshop is that — contrary to what most biologists would acknowledge — the concept of gene itself is not exactly crystal clear either. This is not because geneticists don’t know what they are talking about, but because there are several legitimate uses of the word “gene” that can be deployed in different contexts, depending on one’s research agenda. And some of these uses are not entirely compatible either, and certainly not equivalent to each other.

Consider, for instance, the fact that some biologists refer to genes as whatever has causal effects on the formation of phenotypes and happens to be heritable. Well, by that definition both classical DNA-based “genes” and a variety of epigenetic phenomena qualify! In other cases, genes are defined simply as sequences of DNA that code for a particular protein. That not only excludes epigenetic effects, but also large swaths of DNA sequences that regulate development even though they do not produce proteins. You see what I mean?

Epigenetics is at the threshold of becoming an established discipline in the biological sciences, with implications for genetics, developmental biology, evolution and even medicine (many epigenetic effects are causally involved in a variety of diseases). P.B. and J.S. Medawar, in their classic Aristotle to Zoos (1983) famously said that “genetics proposes, epigenetics disposes,” meaning that the whole of epigenetic processes is what allows genes to produce phenotypes. If that is the case, and I don’t see any good reason to doubt it, epigenetics is poised to become a central discipline in 21st century biology.