It Takes a Genome: How a Clash Between Our Genes and Modern Life is Making Us Sick

by Greg Gibson

  Then there is the melanocortin receptor, MC1R, which has two variants that pretty much guarantee red hair. Both are rare, being found only in around ten percent of Icelanders (one would suspect that they are considerably more frequent in Scotsmen), which is just as well because they are also major susceptibility factors for skin cancer. Similar mutations are responsible for red hair in all sorts of other mammals, including cats and dogs, and even more rare variants are responsible for albinism in humans. Furthermore, it turns out that at least some Neanderthal Men had a mutation in this gene, leading to speculation that our ancient cave brethren were fair-skinned with red hair and freckles.

  If you were to scan for blonde hair color in Americans, I suspect you’d come up with genes for vanity, since so many of our blondes are artificially so. But in Europe, variation in all four of these genes plus another one called KITLG combines to allow fairly good prediction of whether a person is blonde or brunette. They’re better at excluding the latter than establishing the former.

  This type of result has fascinating, and perhaps troubling, implications for forensics. It raises the possibility that in the not too distant future police may begin to profile their suspects on the basis of features such as hair or eye color as implied by the contents of a blood drop. General facial features are likely to be too genetically complex to allow forensic scientists to sketch a face without a witness, but features such as detached earlobes might not be far off either.

  We will certainly be looking at racial profiling in a whole new light, since a substantial fraction of the variation in skin pigmentation is attributable to these same genes mentioned previously, working together with a few other genes. Looking at the patterns of variation in the DNA it is pretty clear that lighter skin has been under selection in northern latitudes for thousands of years. The general consensus is that dark skin inhibits the absorption of ultraviolet radiation needed for the manufacture of vitamin D3, but this disadvantage is offset by its protective role against sunburn and skin cancer, now that we’re a largely hairless species. More recent speculation has it that the great pigmentation shift actually did not occur coincidentally with human migration into northern Europe, but rather awaited the transition to ways of farming that also shifted our vitamin needs. In Australia, where kids have been encouraged to slip on a shirt, slop on some sunscreen, and slap on a hat to ward off skin cancer, osteoporosis due to vitamin D deficiency is on an alarming rise.

  The God Gene

  Not to belittle their powers, but crystal balls, tea leaves, and palm reading leave a lot to be desired when it comes to accurate prediction of the future. The question before us now is whether personal genotyping will prove to be any more accurate. It is one thing to fork out $20 for a séance on the Las Vegas strip, another to find $1,000 to send off to “23andMe” or “DecodeMe” for a saliva kit that gets your own genome profiled using the same methods that underlie much of the research in this book. I’m toying with the idea myself, even though I know the money could be better spent paying down the mortgage, or even better, providing heifers for Haitian villagers.

  To be fair, the objective of these new personal genome ventures is to provide people with an accurate read out of their ancestry. But our susceptibility to disease, along with some portion of our predilections for whom we become, is inscribed in the very same code. As the next few years unfold, companies will come online offering to interpret your personal genomes, providing a glimpse of the future for you and your kids. Will they be short or tall, extroverted, disagreeable, or emotionally unstable, athletic or beautiful? Well, of course they will be beautiful and smart, but you get my gist.

  The first person to head in this direction is the first person to have had his entire genome sequenced, none other than Craig Venter, the founder of Celera Genomics. He published a paper describing the millions of genetic variants in his DNA in late 2007, and then incorporated two dozen snippets of this information as boxes within his fascinating autobiography, A Life Decoded. If you read these closely, you will get a sense of how impossibly silly this genetic prediction business is for now. For example, he sees some agreement in the sequence of his GSTM1 and CYP1A2 genes and his sensitivity to asthma and ability to thrive on caffeine, respectively, but also notes inconsistency between his PER3 and DRD4 alleles and his tendency to be a night owl and to embark on risky ventures. Regarding the diseases we’ve discussed in this book, his genome is a perfectly average mixed blessing: Some hint of Alzheimer’s susceptibility here is offset by a protective variant there; his outlook for depression and coronary heart disease are about normal; and diabetes yields a completely unconvincing read out as well.

  In each case, Venter wryly remarks that, well, a single gene never tells the whole story, and more research is needed. If disingenuous, at least it is honest, more so to my mind than the approach of the second person to have his genome sequenced, James D. Watson of double helix fame. A rival company with a new technology, 454 Life Sciences, has compiled his famous genome, but certain parts will be masked from public release, ostensibly to protect the privacy of his children. Oh the painful decisions those future celebrities will face as a hungry public clamors for information about what makes them special: Will People magazine soon be publishing genome updates on Paris Hilton’s addictions and Tom Cruise’s fertility?

  From memory to conscientiousness, psychologists now agree that the genome has some role to play in human behavior. We can be sure that genome scans for specific behaviors will soon be published, and equally sure that they will identify no more than a handful of genes that explain no more than a few percent of the variability that we experience. Further, we will see a rash of popular articles arguing just why this or that particular variant has been selected by evolution because it is good to be aggressive or altruistic, zany or zealous.

  But behavior is more complex even than the diseases we’ve considered. While I have no doubt that bits of DNA are associated with all these behaviors, I am equally certain that we have little hope of unraveling just what balance of forces shape their prevalence in the human gene pool. Beware then of simplistic explanations for one person’s foibles and be skeptical of rationalizations on the basis of some specific ancestral human need.

  It would be equally unwise to completely dismiss the idea that attributes as personal as spirituality are somehow embedded in our genomes. Dean Hamer has written the first book on this topic. Provocatively titled The God Gene: How Faith Is Hardwired into Our Genes, it lays out research conducted in his own lab that purports to find a gene that influences how religious a person is. Read the last chapter, though: If anything, the wiring is extremely soft, the gene implicated has just a small effect, and it is definitely neither a gene for which religion a person follows, nor even for whether they are religious, actively praying to a public God. Rather, the book describes an allele that correlates with a psychological dimension that has something to do with how spiritual a person is, spirituality being more of a Dalai Lama trait than a Jerry Falwell one.

  Here’s how the strategy works, pretty much following how geneticists go about finding disease-associated genes, except that some measure of godliness is substituted for disease. You find a large number of unrelated people and ask them a bunch of questions designed to reveal their inner self on a five-point scale. For example, if you answer “I would prefer to be at the office rather than home with my new baby” with the “Strongly Agree” option, you’re probably disagreeable, while answering “Mildly Disagree” might knock points off your conscientiousness score. It turns out that a particular set of 30 or so such questions can be used to devise a single score that after the fact has been claimed to measure spirituality. The score also shows a high correlation between parents and their kids, showing that whatever the score is measuring is heritable. Then you just comb through suspect places in the genome for sites that have an A instead of a G, for example, in people with different spiritual values. Lo and behold a gene called VMAT2 has a variant that is a little bi
t enriched in your basic Dalai Lama types.

  I find it ironic that Hamer used basically the same approach to argue that he had found a gene for homosexuality a few years back as well, notably with equally controversial consequences. And equally nebulous association: Neither of these results has been extensively replicated. And although one or both may be true, it seems likely that just like the genetic variants that influence height, these do no more than gently nudge a person in a certain direction.

  There are in fact literally thousands of similar results in the behavioral genetics literature. Perhaps the most famous one is a change in front of one of the dopamine receptor genes, which has been found to be enriched in people who engage in all manner of risky behaviors from jumping out of planes to participating in orgies or exploring drugs. The thing about all these studies is that they are performed out of context of the rest of the genome, and now that we have the tools to study every gene all at once, their effects will be seen to be a tiny part of the complicated whole. They are like the corner pieces of a jigsaw puzzle, convenient anchor points, but alone they don’t even tell you what the picture is about.

  A Few Words About IQ

  Inevitably, pundits are going to start noticing that some of the variants supposedly associated with human differences are at different frequencies in different racial groups. Nowhere is this more worrying than in relation to inference about human intelligence. In fact, it has already started. Microcephalin, for example, is a gene required for brain development in mice, and rare mutations also cause human children to be born with tiny brains. It has been argued that it is one of the most strongly selected genes in humans, with particular alleles being prominent in Caucasians but absent in Africans. Ergo, in the minds of those quick to jump to flimsy conclusions, Africans are said to be genetically inferior intellectually. In fact, when others went and actually looked, they found absolutely no association between the selected allele and IQ.

  There is a long history of attempts to establish that African Americans have a genetically half full cupboard, but intriguingly little attempt to similarly argue the other way for Asians. The basic argument is that IQ testing has repeatedly demonstrated significant differences in IQ between ethnic groups, while testing of twins has pretty convincingly shown that there is a strong genetic component to IQ. Ergo, the difference between the races must be genetic—but for a simple error of logic. Just because the variation within a group has one cause does not imply that the difference between groups has the same cause. No one has a problem understanding this when the discussion is about the height of immigrants who move to a developed country: The average height of their children typically jumps a couple of inches. Obviously the genes didn’t change, so the difference between the generations must be dietary. So why are many people so quick to attribute racial differences to purely hypothetical genes?

  Now consider some of the human traits that are most commonly transmitted across generations: religious observance, allegiance to sporting teams, and—believe it or not—a love of opera. I would imagine that most of us have a hard time believing that genes are the least bit involved in any of these behaviors. Maybe they have something to do with musical appreciation, but clearly these examples of transmission can be explained by cultural and geographic factors that just happen to be correlated with familial relationships. Inheritance, in a broad and colloquial sense, refers to the transfer of something, such as property or assets, from a parent or guardian to a child. Since genes are responsible for the transmission of biological attributes, it has come to be assumed that if there is resemblance between parents and offspring, genes must be involved. But Muslims tend to beget Muslims and Catholics beget Catholics, and yet none of us would rightly conclude that there are genes for Islam and Christianity. The point is that the combination of education and the environment is much more likely than genetics to account for the differences. We need more data of a different sort to establish that genes are responsible.

  It is thus quite possible for any trait to be highly heritable, and to differ greatly between two groups, but for this difference to have nothing to do with the genes. Anyone who follows the syllogism above, that genetic variation within a group implies genetic divergence between them, is selling you a bill of goods. So long as the environment can bring about a difference between peoples, it is risky to invoke genetic differences and appropriate to demand an exceptional standard of proof when people’s lives are at stake.

  Nevertheless, bloggers are starting to notice that there are genetic variants that appear to be associated with intelligence or other behaviors and appear to differ in frequency between races. How should we react to reports that scientists have found some genetic variant in 70 percent of Whites that is associated with IQ score at a probability of 0.000001, and is present in just a minority of Blacks?

  First, recognize that significance values alone don’t say anything about the magnitude of an effect. Throughout this book we have discussed genetic associations that are significant at probabilities thousands of times more convincing than this, yet only explain a few percent of susceptibility to a disease. These results need to be replicated multiple times in many tens of thousands of people before they become convincing. Typically, when this is done, the estimated magnitude of the effect becomes less and less. For now it seems highly unlikely that any single difference will account for more than an IQ point or two, whatever that means. Miniscule in relation to all the incredible potential bottled up in all our genomes.

  Second, ask whether the experiment has been done in the reverse direction. Have they looked for associations with IQ in other races and then asked whether the alleles are at a low frequency in Whites? The trouble is that there is an enormous ascertainment bias in the way genetics is done predominantly on northern Europeans and affluent Americans. You might find dozens of associations that involve alleles that are enriched in these populations, but they say nothing about what variants might be present in the other racial groups and are lacking in Caucasians. The work has to be balanced. As it happens, Africans are on the whole considerably more genetically diverse than any other human group, and this fact alone suggests that a certain amount of genetic potential has been lost in the course of migration around the globe.

  Third, put the research in context. Another theme of this book has been that the environment modulates genetic effects, often swamping them. A case in point here is a just-published study from the group that brought us the interaction between stress and a serotonin transporter variant that influences suicidal tendencies. Breast-feeding, it is suggested, improves childhood IQ measurement, particularly in children who can digest particular fatty acids because they have a particular flavor of the FADS2 gene. Breast-feeding? Colored balls above the crib I can believe, but how many other subtle and not-so-subtle parental behaviors must affect how we develop?

  Finally, keep in mind that average values are meaningless when it comes to appreciating the worth of individuals. Even if there were a genetically based difference of a few points in IQ between ethnic, cultural, or geographic groups of humans, the brightest tier of the group with the lower average would still score higher than almost everyone in the other one, and almost half of the first group would be genetically as well if not better endowed than every other member of the second. How many of us know our own IQ within 10 points, let alone those of our friends and colleagues?

  Genes play their part, but let’s attend to the things we can control. It is perverse to think that a society would focus on the relatively small component that may be genetic, rather than doing everything in its power to maximize the potential of every individual by doing something about the much larger component that it is within our power to address, namely education and public health.

  On Being Human

  What is it that makes us humans human? What are the genetic changes that have made us a species that looks up at the stars and sees future colonies instead of pagan gods, that spends its free time picking out friend
s on Facebook instead of picking at gnats on friends, and that thrives in the jungles of Manhattan as ably as it roams the African savannah?

  Surely we should have some answers by now. If we compare the blueprints of a Porsche Boxster and a Honda Civic side by side, it is immediately apparent how they come to differ. Similarly, you might think that if we align the genomes of a human and a chimpanzee, it ought to be straightforward to stitch together the evolutionary path that leads to Homo sapiens sapiens.

  Superficially, it is. Our capacity for smell is different from other primates, since our repertoire of olfactory receptors, the molecules that sense perfumes and rotten eggs alike, looks more unlike that of our nearest relatives than do most genes. Similarly, the sequences of our immune system receptors are in general quite diverged, much as the bumper bars on a Porsche and a Honda are barely recognizable variations on a theme. Neither of these changes is remarkable, and both are entirely predictable given a moment’s thought.

  There are actually a half dozen subtly different ways to compare primate genomes, but they all tell pretty much the same story. Namely that hundreds if not thousands of genes are suspects of interest in the making of modern humans. Some have molecular roles in causing neurons to divide (so that the human brain got bigger), some look like they digest toxic compounds (so our livers can tolerate wild herbs in the new world), some are involved in deposition of calcium in the bones, in hearing, in pigmentation, and...you name it.