The subtitle of this site (and of the book on which it is based) is How genes set the balance between autism and psychosis, and clearly, this is a crucial issue for the imprinted brain theory, claiming as it does that imbalances in gene expression ultimately explain mental illness and normality alike. This week a landmark development in genetics has opened up some remarkable new possibilities regarding exactly how genes—or at least, DNA—might do this.
The Encyclopedia of DNA Elements, or ENCODE, published this week in Nature and a number of other journals, sensationally contradicts the belief that human DNA is 99% junk by suggesting that up to 80% of the genome may have some function, and that up to 75% is actually transcribed in some cells. Indeed, the DNA of this epigenome, if we may call it that, appears to vastly outweigh that of the 20,000 odd true genes, promoting gene-expression in 70,000 regions, and enhancing it at another 400,000. Furthermore—and crucially where the imprinted brain theory is concerned—the researchers report that many DNA variants previously correlated with specific diseases lie within or very near these epigenomic regions, and others have been found to exert their effect from parts of it much further away, thereby providing new leads for linking genetic variation with mental illnesses like autism and schizophrenia.
As I pointed out in a previous post, the imprinted brain theory could well prove to be the critical test of the so-called “selfish gene” version of modern Darwinism. This simply holds that, from the ultimate perspective of evolution, organisms evolved to copy their DNA, not DNA to copy the organism, as you might think. Today no educated person could believe that genes evolved to copy the organism, because we now know that the genes people pass on to their children are the very same ones—occasional random mutations apart—that they inherited from their parents. This is especially so if you are female, because a woman’s last will and genetic testament is written in DNA and secreted in her egg cells long before she is even born. Males, admittedly, produce sperm by the million daily, but there seemed no way in which they could edit their genes as they went along, so to speak.
But then the discovery of epigenetic factors like imprinting (which silences a gene depending on its parent of origin) seemed to re-open the question and suggested to some that Lamarckian inheritance of acquired characteristics might be possible after all—especially when it was discovered that some epigenetic effects could be transmitted across the generations. Indeed, I invoked one such trans-generational mechanism myself (what I called “lingering Lyonization”) as a hypothetical explanation of Asperger’s syndrome and its strange sex ratio.
But, of course, I never for one moment thought that this vindicated the DNA-copies-the-organism nonsense of the neo-Lamarckians and latter-day Lysenkoists, and ENCODE now gives us some insight, not only into seeming “junk-genes,” but into how and why evolution has managed to give us no more genes than toads or mice, half the number of corn, only 20% that of newts, and a mere half a per cent of that of a single-celled amoeba (A. dubia)!
As one of the Nature authors puts it (p. 71), the ENCODE findings raise "the possibility that more information in the human genome may be important for gene regulation than for biochemical function." And there could be several good reasons for this.
The first is that because we share so many of our true genes with chimps, the critical differences between our species—which are mainly ones of brain development and behavior—may lie in parts of our genome now being catalogued by ENCODE, and almost certainly are epigenetic elements relating to differences in gene expression: epigenes, if you like. Indeed, given that adult humans resemble nothing more closely than immature chimps, even our physical differences probably lie in the pattern—and especially the timing—of gene expression more than they do in the genes themselves.
Another point is that DNA evolves continuously, but is seldom completely deleted from the genome. This definitely applies to genes, and the best example is the faulty one for synthesizing vitamin C we carry and copy but can’t use to avoid diseases like scurvy. At some stage in our early evolution this gene became non-functional but was not selected out because our diet provided all the vitamin C we needed to get by. The same is almost certainly true of epigenes—but probably more so. Indeed, such discarded or non-functioning DNA sequences could open a window onto evolutionary history perhaps even wider and more panoramic than that opened by the discovery of the rusting wrecks of true genes, like vitamin C.
A final reason why there might be so many more epigenes than true genes probably relates to sex. Because sexually-reproducing organisms have two parents, rather than one, evolutionary forces like those which produced imprinting in mammals might lead to each parent’s genes favoring a different pattern of expression: just as we find in imprinting. The result might be many different epigenes effectively contradicting one another—and certainly arguing over the outcome where growth, development, and behavior is concerned. We know that this happens in a classical growth-factor gene like paternally-expressed Igf2 in mice, which is contradicted by a maternally-expressed anti-Igf2 gene, Igf2r. Mice who lack both Igf2 and Igf2r develop normally, underlining the fact that these genes are expressions of conflict between the parents, and there might to be human epigenes that are similar.
Historically, epigenesis emerged as the alternative to pre-formationism: the belief that the organism developed from a pre-existing homunculus or miniature model. Today, no one believes this in a crude, homuncular form, but those who deny modern, selfish-gene genetics are latter-day pre-formationists to the extent that they cling to the belief that DNA copies the organism rather than the contrary. The term gene was coined after genetics, and perhaps this sets a precedent for the adoption of epigene as the term for non-protein coding but functional DNA of the kind now being discovered and mapped by ENCODE. The era of the selfish epigene may have begun, and its ultimate vindication may well be the imprinted brain theory. Time—and advances like ENCODE—will tell.