One and the Same (27 page)

Mack has also studied lupus, childhood diabetes, breast cancer, and melanoma. He said the hardest part of his research is recruiting enough twins. He built his twin roster one twin pair at a time. “Years ago, I put an ad in the
Los Angeles Times
, asking for twins with chronic disease, and eighty pairs responded. I put an ad in the
Seattle Post-Intelligencer
and the
Des Moines Register
and I got exactly the same response. I received several grants and began putting ads in papers all over America. Over the long term—over the course of ten years, I got seventeen thousand pairs with chronic disease. If we could just get enough twins with a given disease, we could characterize their genome, and then look within pairs to see what the environmental exposures are by controlling the genome.” In other words, look at two people who have the same genetic code and do detective work to track where their environments or exposures varied. That variance may tell you what caused the disease in one but not the other.

“But that's hard,” Mack admits, “because it takes a lot of twins.”

So is Mack saying that environment, not DNA, is ultimately the determinant of whether one gets sick?

“It's both,” Mack states. “It's always both. And twins are the best single example of why that is necessarily true. There are going to be diseases—like some of the metabolic diseases—that are due to one single gene abnormality. For example, Tay-Sachs—the Ashkenazi gene: You will find twins who are concordant when that gene is present. But for almost all important diseases, twins are
not
concordant. For example, in Hodgkin's disease, twins are fifteen times as likely to be concordant—both affected—if they're identical. But that meant that most twins
weren't—
just ten pairs out of 350 twins were affected. So something else has to be responsible for all that other difference. And that's always the case.”

The new frontier for understanding the mechanism behind discordance in twins is called epigenetics. The field is considered groundbreaking because its premise defies the conventional wisdom
that genes dictate our destiny. Epigenetics tells us genes can actually be changed by environmental factors over the course of a lifetime. They can be turned on or off—expressed or repressed—based on our behaviors or exposures.
Epigenetics
literally means “on top of genetics”: It's now widely believed that there is some force “above the genes”—chemical modifiers or “methyl groups” of hydrogen and carbon—that attaches to a chromosome and renders it dormant. Methylation can be instigated by environmental exposures, some of which include tobacco smoke, diesel exhaust, radioactivity, pesticides, bacteria, basic nutrients, and certain viruses.

“What it means,” Mack explains, “is that the methyl group gets in the way of the function of the gene; it's like putting a cap on the gene. Tissues develop in different ways because some genes are turned off.”

And environment is one force that can turn them off?

“It must be,” Mack replies. “Because we know that differences between identical twins become greater as they get older.” In other words, the longer Robin and I live, and the more our habits, vices, behaviors, or environments diverge, the greater the differential impact on our genes. Mack makes sure I understand how the term
environment
is used by scientists: “When we say, ‘environment,' we mean ‘everything
but
genetics.'”

Identical mice have been the involuntary pioneers for this science. In one 2006 experiment led by Eric J. Nestler at the University of Texas Southwestern Medical Center in Dallas, researchers put small, genetically identical mice in a confined space alongside larger, aggressive, mean mice—bullies—and watched how the smaller mice reacted. (The model is actually—hilariously, I think—called “social defeat.”) While some of the identical small mice became cowering, anxious, and depressed, interestingly, some of the small mice stood their ground. To oversimplify: For those mice who became depressed, the intimidation had a chemical effect, deactivating the genes that normally make a mouse resilient. So theoretically, if one could prevent that chemical change, or counter it once it has occurred, those “defeated” mice would not have been so flustered by the bullies. That
is what conventional antidepressants are supposed to do: counter dejection. Though drugs may not prevent it—”social defeat” still takes root—scientists can counteract it, emboldening the mouse.

Even more intriguing is that the effects of the intimidation on the relevant genes—even if the behavior is mitigated successfully by a drug—may still be passed on to the next generation. The genes may have been altered in such a way that the behavioral traits get inherited. A monthly journal on how the environment affects human health,
Environmental Health Perspectives
, called a similar 2005 report “startling” because it suggested “that epigenetic changes may endure in at least four subsequent generations of organisms.”

Psychiatrist Peter Kramer, who wrote
Listening to Prozac
and writes a regular blog on
Psychology Today's
Web site, said the bully mouse study shows “how adversity might reach inside the brain and scar the gene within the nerve cell. The research also points toward a medically exciting, if ethically complex, future in which traumatized people might be restored to the neurobiological state of their resilient twins.”

One rat study in 2004 demonstrated that it's not just chemicals or foods that influence a gene's expression; affection, or the lack of it, may also have an impact. The experiment, conducted at McGill University, in Montreal, found that rats who were not licked and groomed by their mother as often as their siblings went on to exhibit more stress. Dr. Moshe Szyf and his colleagues discovered that the mom's neglect had the effect of
turning off
the rats' stress-mitigation response, thus spiking their anxiety levels. Rats that weren't licked as babies ended up with methylation on a particular gene that normally produces a coping brain receptor. “The offspring of the high-licking moms exhibited better response to fear,” explains Dr. Szyf, who headed the research.

The trauma from maternal disregard was not irreversible: Drugs could bump off the methyl group, change the gene's activity, and reduce a rat's stress. In an e-mail to me, Szyf explained that derailing
the methylation—which had muzzled the gene in the first place—allowed the rat to get a grip and cope.

In the summer of 2005, a research group led by Manel Esteller at the Spanish National Cancer Center in Madrid, found that out of eighty sets of identical twins, 35 percent of the pairs differed from each other epigenetically, and the older the twins were, the less identical. Those who had spent the most years living in different places with different lifestyles showed the greatest DNA differences, more proof that epigenetic differences can account for why one identical twin gets a disease and the other doesn't. For example, both twins may start with the same genes that normally battle tumors, but one twin's tumor-fighting DNA is rendered impotent by an epigenetic change, and that twin ultimately gets cancer.

I asked Dr. Mack how he would counsel my sister, for instance, if I were diagnosed with breast cancer. “I can actually reassure her,” Mack replies. “I would tell you that it would be more likely than not that she would never get breast cancer, but she would have a substantially higher risk than an ordinary person. It still would be less than fifty percent. Fifty percent refers to lifetime risk. But if most women have a twelve percent chance, she might have as high as a forty percent chance. So it's substantially higher than the average person but still less than fifty percent.”

Mack urges me to put in perspective the studies that say identical twins prove certain diseases are inheritable just because they're found to be more alike for disease than fraternal twins. “Almost always conveniently, these studies omit the fact that identical twins tend to consciously adopt similar behavior,” Mack cautions, “and therefore have the same exposures not because of specific parts of the genome but because commonality of the entire genome makes them identify with and therefore copy each other.”

I tell Mack that the following day I'm scheduled to meet his compatriot, biologist Dr. Eric Vilain, a professor of human genetics based at UCLA's David Geffen School of Medicine, who is applying epigenetics
to the question of homosexuality—what makes one identical twin gay and the other straight? He and Mack are on similar quests, except where Mack is deconstructing predictions of disease, Vilain is trying to crack sexual orientation. “To me, he's trying too many possibilities,” Mack says bluntly. “I think it's a long shot. If it really works, in the sense that I mean it, it's going to be a revolution—because we will know the biologic basis for homosexuality, or one of them. And if it's epigenetic, well, we know we can influence epigenetics, so in theory, we could influence homosexuality. For example, if you eat a lot of broccoli when you're little, and you get a lot of folic acid, you may have different epigenetic patterns than if you don't. Well, so in theory, if you think a kid is susceptible to homosexuality, you might change eating habits to impact the genes. Now that's pie in the sky. It's predicated on Eric being right—substantially right—and everything falling into order, so I don't think that will really happen. But let's give it a shot. Because I think a lot of gay people really want to know why.”

Dr. Vilain is a youthful forty-one-year-old Parisian with wide-set eyes and the pale skin of a scientist who doesn't leave the laboratory much. We talk in his sunny, cramped office, sitting side by side next to his blondwood desk and built-in bookshelf; glancing at the wall behind him, I notice indecipherable equations scrawled on a dry-erase board. He has a thick French accent and speaks in a hurry. “I only became interested in sexual orientation a few years ago. I'm not a twins researcher. We've just started using twins recently because we think it's a great hypothesis to test the idea that epigenetic/environmental influences are important in behavior. … We're asking the question, ‘Do we see a difference in genes being turned on or off between co-twins who have opposite sexual orientation?'” In other words, is something activating or deactivating certain genes that results in homosexuality?

“People will say, ‘Are you studying the gay gene?'” Vilain asks the question for me. “No. We're studying the gene that makes people
attracted to either males or females. So in many respects, we're studying the
straight
gene; we're just using gay individuals as a model.”

Does he think it's ultimately going to turn out to be one gene or a confluence of genes that is responsible for sexual preference?

“We don't know,” Vilain replies frankly. “It's probably a confluence, and it becomes more complicated because of the environment influences. Right now, it's still a very poorly studied area. There are several levels of evidence that homosexuality has some genetic influence. The first evidence was actual twins studies—not
genetic
twins studies, just
traditional
twins studies.” He means the studies in the early 1990s by Dr. J. Michael Bailey and Dr. Richard Pillard, who looked at identical and fraternal twins and tested whether homosexuality is genetically determined. If it is, then both identical twins should be gay if one twin is, while fraternal twins should differ more often because they share only half their genes. They found that if one identical twin is gay, the other has a 50 percent chance of also being gay; if one twin is fraternal, the other twin has a 20 percent chance of also being gay.

Vilain explains how these results argue both for and against the notion that homosexuality is genetic: “You could say, ‘Well it's
only
fifty percent; if one twin is gay, the other twin is going to be gay in only fifty percent of the cases. That means it's
not
genetic.' But of course you have to hold this thought a moment, because you also have to think, Wait a second; if it was all
environmental
, then it wouldn't make any difference if it was a fraternal twin, and the fraternal twins would
also
be at fifty percent concordance. But they're not. They're systematically
less.”

Months earlier I'd interviewed psychologist Dr. J. Michael Bailey—the researcher whom Vilain cites as pioneering the first gay studies—in his Chicago apartment, a small walk-up with cloudy windows, an olive green rug, and the bachelor's requisite leather couch. Bailey, in his mid-fifties, was wearing jeans, loafers without socks, and wirerimmed glasses. When I inquired as to whether he is a twin or gay
himself, he said he is neither. Divorced and the father of two children, he was formerly chairman of the psychology department at Northwestern University, and obviously bruised from an onslaught of criticism of his 2003 book on transgender people,
The Man Who Would Be Queen: The Science of Gender-Bending and Transsexualism
, which was unrelated to twins.

In the two landmark twin studies Bailey shepherded—one that recruited gay twins through newspaper advertisements, one that made use of the Australian database of twins—the concordance rate (meaning both identical twins were gay) was, as mentioned, 50 percent in the first study, between 20 and 25 percent in the second. Bailey puts more stock in the lower concordance rate: “Most of the time a gay identical twin is going to have a straight twin,” he asserted. “At least seventy-five percent of the time, maybe more. It's even higher in fraternal twins. So basically, if you meet an identical twin who is gay, the chances are his twin is straight. And that is surprising to a lot of people.”

Of those pairs where one twin was gay, one not, Bailey said, “The twins both recalled the differences emerging in childhood. The gay twin was more gender nonconforming than the straight twin. So if it's a male pair, then the gay twin is more feminine; the gay twin is closer to the mom, the straight twin is a little closer to the dad—that kind of thing.”