Do Fathers Matter?: What Science Is Telling Us About the Parent We've Overlooked (7 page)

Genes that can lead to such serious illnesses would not have arisen in evolutionary history if there wasn’t a very important reason for them to be there. After my visits with Alexander and James, I looked for someone who could explain why imprinting occurs to begin with and what it might tell us about what fathers contribute to their children. That led me to David Haig at Harvard.

*   *   *

A few years after Surani’s discovery of imprinting, Haig, then a young biologist in Australia, was preparing for a career that would one day lead him to a fascinating explanation for why imprinting exists. Now a professor of evolutionary biology, Haig almost abandoned the field after his college graduation for a life of travel and adventure. Instead, he decided to work toward a Ph.D. in what seemed to be an obscure corner of evolutionary biology: conflict between parents and offspring in plants. But the work he did then would ultimately help him develop what he calls “the kinship theory” to explain why genetic imprinting occurs.

To get to Haig’s office in Harvard’s Museum of Comparative Zoology, visitors must walk through the museum exhibits and past the famous Blaschka glass flowers. Haig has no laboratory. His job is not to do experiments, but to try to explain the results of others’ lab work. His greatest experimental achievement, early in his career, involved counting “a quarter of a million bristles on fly bellies,” he said, and that was enough to drive him out of the laboratory for good.

As his research led him beyond the study of plants, Haig began to explore the curious competition that occurs between parents and their offspring. In 1993, Haig published a paper on one aspect of the competition—a conflict that occurs between the mother and the fetus during pregnancy. Fathers might have reasons to compete with mothers, too, but Haig showed that a fetus does so even while utterly dependent upon her for survival. “Pregnancy has commonly been viewed as a cooperative interaction between a mother and her fetus,” he wrote. But that’s not true. It’s warfare in which “fetal actions are opposed by maternal countermeasures.”

One rather remarkable example of this is the ability of the fetus to alter its mother’s arteries so they can’t constrict. The fetus can then harvest whatever nutrients it wants from the mother’s bloodstream, through the placenta. And she is powerless to resist. That control also means that the fetus can release hormones directly into the mother’s bloodstream. One such hormone alters the mother’s regulation of insulin. The fetus can make its mother’s blood sugar rise. As that sugar-rich blood circulates through the placenta, it delivers more sugar to the fetus. But if the process goes too far—if the mother loses too much control over her blood sugar—the mother can develop diabetes, as was the case with James’s mom, Barbara. According to Haig, gestational diabetes is just one possible outcome of the survival struggle between mother and fetus.

Other hormones are believed to increase the mother’s blood pressure, enhancing the flow of blood to the fetus. If these fetal hormones overwhelm the mother, she can develop dangerously high blood pressure. This, too, is seen in the clinic. The condition is called preeclampsia, and when it occurs, blood pressure can climb to a level at which it becomes fatal. Haig marvels at these delicate relationships. “Natural selection produces things on this planet that are much more complex than any nonliving part of the universe. I come out of evolutionary biology, and I’m wanting to address this question—why has this thing evolved?”

As he was trying to work that out, he began thinking about another kind of genetic conflict, not between parent and child, but between parent and parent. It was an outgrowth of the work of Surani and Solter and others. He knew that they had found there was something that differentiated mothers’ genes from fathers’ genes. Haig was trying to explain why that is the case.

The result was his kinship theory. In broad terms, it goes something like this: Fathers and mothers both have strong interests in seeing their offspring survive. But they want different things for their children, because their reproductive strategies differ. In most mammals, the male is unlikely to mate with a given female more than once. He mates, moves on, and mates again. He doesn’t care whether that leaves her too depleted to have further offspring. They’re not his. Females, on the other hand, pursue the opposite strategy. Each of her young ties up a female for a good portion of her reproductive life. She is unable to mate as often and have as many children as a male, so she has to make sure that all her offspring survive. It’s a case of quality over quantity. With any given pregnancy, her strategy is to give the embryo what it needs, but no more. That conserves her resources for her subsequent children. Expending too many resources on one offspring might leave her undersupplied for the next, risking its life or her own. The male, on the other hand, wants to extract as much of the mother’s energy and resources as he can for
his
offspring. Or, as Haig puts it, “Maternal genes have a substantial interest in the mother’s well-being and survival. Paternal genes favor greater allocation of maternal time and effort to their particular child.”

The stage is now set for competition. The male and female do whatever they can to advance their competing strategies. But how do they do that? How can a father manipulate the mother to extract maximum resources for the child? And how can a mother conserve her resources and make sure that doesn’t happen? Haig’s insight was that imprinted genes are the weapons that males and females use to pursue their competing strategies. The imprints that mothers and fathers put on the genes turn them off or on as needed to pursue these strategies.

The gene’s “imprint” is stamped on it by the parent, and that imprint affects the expression of the gene in the offspring. The genes that are expressed when inherited from fathers tend to encourage more growth in the fetus. They push the fetus to demand more resources from its mother while it’s in the womb—to suck up as many maternal resources as possible, pursuing the father’s competitive strategy. Genes that are expressed when inherited from mothers tend to slow that growth, so the mother can conserve resources for subsequent children.

*   *   *

Haig’s theory was meant to explain not only what Surani had found but also what others found as they began to explore the implications of Surani’s research. While Surani can claim credit for discovering the phenomenon of imprinting, he didn’t know which genes were imprinted or precisely what each of them did. It took about a decade for researchers to discover the first imprinted gene. That was the work of Elizabeth Robertson, then in the Department of Genetics and Development at Columbia University and now at the University of Oxford in England. She was investigating normal growth and development in mice by inactivating, or knocking out, certain genes to see what effect that would have on developing embryos. In a paper published in
Cell
in 1991, Robertson and her colleagues reported an unusual discovery regarding a gene called
Igf2
, which is responsible for the production of something called insulin-like growth factor II, or IGF-II. When the researchers knocked out the gene in mouse mothers, nothing happened. The offspring were normal. That meant the gene must have little or no role in the offspring’s development. But when Robertson knocked it out in mouse
fathers
, the embryos grew to only about 60 percent of the size of their normal counterparts. When the gene came from a father—and only when it came from a father—it was clearly essential for growth.

This fit nicely with Haig’s theory. The paternal gene enabled the fetus to draw more nutrients from its mother, a crucial part of the father’s mating strategy. This was critical experimental evidence of the competition between mothers and fathers. Other researchers quickly followed with the discovery of other imprinted genes, and the genes that came from the fathers encouraged fetal growth, as Haig theorized. Taken to the extreme, this strategy had a serious drawback: if the fetus extracted too much from its mother, she would die, and so would the fetus.

The new discoveries also began to reveal mothers’ powerful counterweapon: genes that are maternally stamped fight back against the male strategy, encouraging the fetus to draw only the nutrients it needs to survive, not as many as it can. Mothers put their stamp on a gene that counters the growth boost by
Igf2
. It’s called
Igf2r
, and it’s the gene for what’s called the IGF-II receptor. In order for the male’s IGF-II to work, it must plug into the IGF-II receptor. If the female controls the receptor, she can moderate the greedy nutrient-seeking of IGF-II. Denise P. Barlow and colleagues at the Research Institute of Molecular Pathology in Vienna discovered that Haig’s theory held once again.
Igf2r
, as expected, was also imprinted—but in the opposite direction. The receptor gene was active only when it came from the mother. When the
Igf2r
gene is knocked out in fathers, nothing happens. But when the
Igf2r
gene is knocked out in female mice, the offspring grow too big and die before birth.

Haig published a paper about this fascinating competition with a title worthy of Sherlock Holmes: “Genomic Imprinting and the Strange Case of the Insulin-like Growth Factor II Receptor.” In it, Haig writes, with appropriate Sherlockian flourish, “Surely, it is no coincidence that IGF-II and its type 2 receptor are oppositely imprinted.” For Haig, this was an exciting confirmation of his theory. The two genes enabled the parents to battle over the size of their offspring, each of them advancing his or her own evolutionary goals. To those who might be critical of his theory, Haig chose Holmes himself to respond: “It is an old maxim of mine that when you have excluded the impossible, whatever remains, however improbable, must be the truth.”

Humans have counterparts to these genes, and when this system goes awry, the consequences can be devastating. Suppose, for example, a mother’s and father’s copies of
IGF2
(scientists capitalize the names of human genes and use lowercase letters to indicate nonhuman genes) are both mistakenly turned on—the mother’s isn’t turned off as it should be. Or suppose the fertilized egg accidentally gets two copies of the father’s turned-on gene. The fetus then gets a double dose of growth genes. This leads to a condition called Beckwith-Wiedemann syndrome, in which children have birth weights more than 50 percent above normal. And the opposite mistake can occur. If both genes are silenced, the fetus doesn’t draw on its mother’s resources the way it should, and it is born below normal weight.

“It’s a tug-of-war,” Haig said. “You’ve got these two sides tugging on the rope. They’re not shifting much—it’s just a little bit one way or the other. And they come to depend on each other, on the other side holding the rope. If you get a mutation in an imprinted gene, you get a really pathological outcome. One side has let go of the rope.”

Until recently, genes subject to this gender division were thought to be rare, numbering perhaps a hundred or so out of the estimated 25,000 human genes. But Haig and his colleague Catherine Dulac, a Harvard molecular biologist, used a different method to find imprinted genes and concluded that there could be more than a thousand of them. Some critics have questioned this result, arguing that the study had flaws and that imprinted genes are not as common as Haig and Dulac claim. But whether or not that’s the case, it’s clear that this genomic battle of the sexes is not a rare phenomenon that plays out in isolated corners of the human genome—it’s far more widespread.

Whatever the actual number of imprinted genes turns out to be, it is already clear that many of them are expressed only in the brain, where they can affect behavior in many ways. Indeed, maternal and paternal genes battle for control in the brains of every one of us. As Catherine Dulac told me, “We know we get conflicting advice from mom and dad. Here it’s in the genome—it’s in your own brain! So you can’t escape mom and dad fighting over what you’re supposed to do.” But these imprinted genes are not expressed in all parts of the brain. That raises an interesting question. Research has shown that imprinting errors can affect a fetus’s growth and even threaten its life. Could errors in imprinted
brain
genes be linked to mental illness?

Christopher Badcock of the London School of Economics and Bernard Crespi of Simon Fraser University in British Columbia think so. They believe that disruptions of the tug-of-war between imprinted genes in the brain could help to explain the origins of some mental illnesses—from autism to schizophrenia. This theory could also help to solve a long-standing riddle about the genetics of mental illness. Many of these illnesses tend to run in families, but it’s not a question of simple inheritance, like eye color. Once again, the facts collide with Mendel’s laws. Many of the inheritance patterns of mental illness are complex and poorly understood. They don’t follow the usual rules. That suggests that imprinting errors might have something to do with these ailments. If so, an understanding of what’s going on could lead to new treatments.

According to Crespi, there is already a link. Children with Beckwith-Wiedemann syndrome, the disorder of excessive growth associated with the
IGF2
gene, have larger-than-normal brains and an increased risk of autism. Studies of people with autism—but without Beckwith-Wiedemann syndrome—have shown that they, too, can have larger-than-normal brains. “There is a good bit of evidence for overgrowth of the entire body and the brain in autism,” Crespi says. “And there is work that has linked that to
IGF2
.”

Crespi and Badcock then looked at the opposite situation—when a fetus lacks proper expression of the
IGF2
genes and is smaller than normal. Would it have a condition that was somehow the “opposite” of autism? People with autism are unable to appreciate what is going on in groups of people around them. They have difficulty understanding what others are thinking. Now imagine individuals with an
enhanced
sensitivity to social cues, even to the point that they seem to “read into” others’ behavior things that are not happening. Such people might hear voices that are not there—a hallmark of schizophrenia. Crespi and Badcock devised a spectrum of mental illnesses based on their possible connection with imprinting disorders. Autism is at one end of their spectrum, and schizophrenia, bipolar disorder, and depression are at the other end. Crespi and Badcock do not think that imprinting and their mental-illness spectrum explain everything about mental illness. But it is essential, they say, to find all the imprinted genes in the brain, discover what they do, and explore how variations in those genes might be related to psychiatric ailments.