New Republic Online June 25, 2001

Have No Fear. Genes Aren't Everything
Heredity and Humanity

by Francis S. Collins, Lowell Weiss, and Kathy Hudson

Forty-eight years ago, James Watson and Francis Crick introduced DNA's elegant double helix to the world in the pages of Nature. With extravagant understatement, they began their report by noting that DNA's "structure has novel features which are of considerable biological interest." Four months ago, with the publication of the sequence and the analysis of the human genome, scientists offered further evidence of just how considerable. Researchers gained a wealth of fresh insights into the miracle of life, and uncovered new mysteries that will occupy biomedical researchers for years to come.

Unfortunately, the new focus on the genome has left some people with the impression that DNA's power is perhaps too considerable--that is, that genes are too great a factor in defining who we are. This fear is understandable. It seems that every morning we awake to a news story presenting yet another way in which our genes appear to be controlling us, like the proverbial tail that wags the dog: "Scientists Zero in on `Genius Gene'"; "Kennedy Tragedies Linked to `Risk-Taking Gene'"; "Diabetes Gene Poses Risk for Latinos"; "Scientists Say a Study of Brothers Proves Existence of a `Gay Gene.'"

With a torrent of headlines such as these, reasonable people have come to fear that the more we learn about the human genome, the more we will see that every aspect of the human condition--from illness to intelligence to fear itself--is just the inevitable product of an unyielding, unfeeling genetic code. For this reason, they worry that our new genomic knowledge represents not a giant leap for humankind but rather a giant demotion. Perhaps we are just marionettes being tugged along by the strands of our DNA. Perhaps our lives are nothing more than a formulaic drama, with a plot line that was finalized before our birth.

Fortunately, ten years of intensive study of the human genome have provided ample evidence that these fears of genetic determinism are unwarranted. It has shown us definitively that we human beings are far more than the sum of our genetic parts. Needless to say, our genes play a major, formative role in human development--and in many of the processes of human disease; but high-tech molecular studies as well as low-tech (but still eminently useful) studies of identical and fraternal twins make it perfectly evident that our genes are not all-determining factors in the human experience.

To put it starkly, we have seen nothing in recent studies to suggest that nature's role in development is larger, or nurture's role smaller, than we previously thought. This is certainly an exciting time in genetic research; but if nature were to take advantage of this klieg-lights moment and boldly declare that it is in charge, history would remember it the same way we remember Alexander Haig.

In large measure, the fear of genetic determinism stems from misconceptions of how genetics works. As high school students, our first exposure to genetics came through the story of Gregor Mendel's experiments with his garden peas. First, we learned that each parent pea plant contributed one copy of each of its genes to its offspring. Second, we learned that certain genes were "dominant" and others "recessive." (The gene for white flowers [W] is dominant, while the gene for purple flowers [w] is recessive.) And third, we learned that a plant would need to inherit two copies of a recessive gene to manifest a recessive trait, but only one copy of a dominant gene to manifest a dominant trait. Offspring--to use the example of flower color--would grow white flowers if they inherited any of the following gene patterns: Ww, wW, or WW. Purple flowers, in contrast, result only from one combination: ww.

All of the above is correct--for flowers and for pea plants. But when it comes to the study of complex human beings, we must take Mendel's peas with a giant shake of salt. Despite what your high school biology teacher told you, Mendelian rules do not apply even to eye color or hair color. Truth be told, they do not apply to most characteristics of peas or other plants, either. As Robin Marantz Henig documented in her wonderful book The Monk in the Garden, Mendel himself came to question the validity of his work on peas when he turned to the study of the hawkweed and got much more complex and confusing results.

This is not to say that deterministic Mendelian rules never apply to human traits and disorders. One classic case in which they certainly apply is sicklecell anemia, a painful and often life-threatening disease that is caused by the presence of an abnormal form of hemoglobin (hemoglobin S) and that disproportionately afflicts families of African descent. Like purple flowers in pea plants, sickle-cell anemia is a recessive trait; it manifests itself in those who inherit two copies of the hemoglobin S gene.

And yet even in this case, human genetics proves far more complicated, and far less deterministic, than Mendel's pea flowers. It turns out that every case of sickle-cell anemia is not created equal. Even when patients have the same two copies of the hemoglobin S gene, the disease may manifest itself in different ways. This is in part because a separate set of genes in the genome--genes that code for fetal hemoglobin--can counteract some of the ill effects of the adult hemoglobin S genes. In most people, fetal hemoglobin genes turn off a few months after birth and the adult hemoglobin genes take over; but sometimes the fetal hemoglobin genes are "leaky," and they continue to produce fetal hemoglobin even into adulthood. When people with two copies of the hemoglobin S gene also inherit leaky fetal hemoglobin genes, their sickle-cell symptoms are usually much less severe. So sickle-cell anemia, widely considered to be the classic single-gene Mendelian disease, is not so clear-cut after all.

Phenylketonuria (PKU), a rare disorder that can cause severe mental retardation, is an even better example of how the most deterministic of genes may not determine much in real life. Like sickle-cell anemia, PKU is a recessive trait. If a child inherits two copies of the PKU gene, then he will get the disease. And yet, thanks to the newborn screening program now in place in all fifty states, the child will never experience mental retardation or the other devastating effects of PKU. Since the illness results from an inability to metabolize the amino acid phenylalanine, if you simply remove foods with phenylalanine from the child's diet, he or she will live a normal and healthy life. PKU is one hundred percent hard-wired in the genes. Yet it can be effectively cured with a one hundred percent environmental intervention.

Keep in mind also that sickle-cell anemia and PKU are about the closest people come to following Mendel's rules. When we look at other human diseases, the picture is far more complicated, with many more genes involved and an even greater involvement of environmental factors. Consider the case of juvenile (or type I) diabetes. Despite what researchers and reporters may have projected to the public over the past two years, there is no single "gene for diabetes." Instead, there are fifteen or more genes that may team up in an array of combinations to produce diabetes. In one person, having variants in five of these genes might be enough to cause symptoms, while in another it might take nine variants.

These gene-gene interactions represent just one layer of complexity. Geneenvironment interactions represent an entirely different story. We now believe that type I cases require not only a series of gene variants but also an external environmental trigger--probably a childhood viral infection. If that is the case, it is entirely possible that in the near future researchers will identify the viral offender, produce a childhood vaccine against it for those who are genetically at risk, and ease the fears of parents all over the world.

It follows from all this that the common use of the shorthand term "the gene for illness X" by scientists and journalists is deeply misleading. If illness X is not one of the rare single-gene Mendelian diseases, then the so-called "gene for illness X" is more correctly described as "a gene variant that may, in combination with other genetic and environmental factors, increase the risk of developing illness X." Just think of how many times in recent years we have heard the inherently deterministic label "the gene for breast cancer" in reference to the genes BRCA1 and BRCA2. For starters, BRCA1 and BRCA2 are actually anti-cancer genes. It is when someone inherits an abnormality in these genes that she can develop breast cancer or ovarian cancer.

But the larger point is that not everyone who has abnormalities in the BRCA1 or BRCA2 genes develops breast cancer, and not everyone who develops breast cancer has BRCA1 or BRCA2 abnormalities. Calling them "the genes for breast cancer" hopelessly confuses a correlation with a cause. And recall also the case of PKU: despite the fact that it is a single-gene Mendelian disease, nurture (in the form of a change of diet) still can trump nature. So yes, gene variants can and do increase our risk of developing diseases. But only extremely rarely do they determine our fate.

Why, then, all the fuss about the genomic revolution in medicine? If disease susceptibility is not deterministic, will it be all that revealing to discover the glitches that all of us have within our DNA? It most certainly will. First, identifying our individual predispositions to future illness will allow individualized programs of preventive medicine, in which we modify lifestyle, diet, and medical surveillance to reduce the risk of illness. In most cases, the resulting treatments will not be the all-or-nothing scenario of PKU; they will have much more in common with the steps that many of us are already taking to reduce our serum cholesterol (for which the set point has strong genetic roots) in an effort to lower our risk of heart disease.

More importantly, perhaps, every disease-susceptibility gene that scientists identify will shine a bright light on the molecular pathway by which that illness comes about. The proper understanding of those pathways offers us the best opportunity ever to develop targeted therapies that work. Even if someone's case of heart disease or cancer has only weak genetic roots, the knowledge of the pathway involved, discerned by the study of genetics, can form the basis of a treatment that may cure his or her disease.

But what about non-disease-related traits, such as intelligence and violent behavior? When it comes to behavioral traits like these, after all, a little genetic determinism can go a long way. The discovery of a prevalent gene variant strongly correlated with violence could have a profound effect upon our millennia-old understanding of free will, and weigh down the scales of justice in two equally dangerous ways. If someone who commits a violent crime has the gene variant, his lawyer could use a DNA defense ("If it's in the gene, the man is clean!"), and the defendant could well be seen by a judge and jury as not responsible for his actions. Yet it is also possible to imagine a scenario in which someone who has never even contemplated a violent act is found to have the gene variant and then subjected to the presumption of guilt (or even sent away to a postmodern-day leper colony) for the rest of his life.

If genes truly controlled behavior, our justice system and its guiding principle of equal protection would not be the only casualties. How would our concept of equal opportunity survive? What about the idea of merit? Just think of the frightening "genetocracy" depicted in the movie Gattaca (and note the letters that make up its name), a world in which children are assigned to castes at birth, based on an assessment of their intellectual capacity and professional potential as inscribed in their DNA.

These fears, too, are unwarranted. To be sure, scientists will find many behavioral factors in the genes. Researchers have long known that there is one extremely common genetic factor that confers at least a ten-fold increase in the propensity to exhibit criminally violent behavior. It is called the Y chromosome. No one has suggested that all those who possess this genetic marker--that is, all males--ought to be seen as lacking free will or inherently possessing criminal intent. More to the point, the case of the Y chromosome is an almost absurd extreme. In the vast majority of cases, genetic factors exert a much smaller influence on patterns of behavior and capability.

In 1998, for example, a researcher reported the discovery of the first gene correlated with general cognitive ability. Reports in the press lauded it as the "genius gene." Given humankind's history of eugenics (the hard diabolical Hitler kind and the soft insidious Bell Curve kind), the discovery of a gene linked to intelligence was genuinely explosive stuff. In reality, however, the so-called "genius gene" was found to give a boost of exactly two points on IQ tests. That's right: two points. Valuable science, yes. Society-altering discovery, no.

New findings that flow from the completion of the human genome draft are likely to follow the same complicated and undeterministic pattern. According to the combined wisdom of twin studies and molecular studies, human behaviors appear to be like the most complex diseases: if a particular behavior has a heritable component at all, it involves the interaction of numerous genes and numerous environmental influences. Surely this should come as no surprise. After all, behavior is a product of the brain, which is by far our most complex organ, and one that continues to develop throughout a lifetime of living and learning.

To build on a metaphor offered by the biologist Johnjoe McFadden, looking for genes that encode our unique behaviors and the other products of our minds is like analyzing the strings of a violin or the keys of a piano in the hope of finding the Emperor Concerto. Indeed, the human genome can be thought of as the grandest of orchestras, with each of our approximately thirty thousand genes representing a unique instrument playing in the wondrous and massive concert that is molecular biology. Each instrument is essential, and each must be in tune to produce the proper (and highly sophisticated) musical sound. Likewise, genes are essential to the development of the brain, and must be "in tune" to produce functioning neurons and neurotransmitters. But this emphatically does not imply that genes make minds any more than a viola or a piccolo makes a sonata.

For many of us, there is still another powerful reason, wholly apart from the mechanics of science, to reject the notion that DNA is the core substance of our humanity. It is the belief that a higher power must also play some role in who we are and what we become. Of course, some scientists and writers dismiss this spiritual notion as pure superstition. Thus Richard Dawkins has observed that "we are machines built by DNA whose purpose is to make copies of the same DNA.... It is every living object's sole reason for living." Really? Is there nothing about being human that is different from being a bacterium or a slug?

Can the study of genetics and molecular biology really account for the universal intrinsic knowledge of right and wrong common to all human cultures in all eras (though all of us have trouble acting on this knowledge)? Can it account for the unselfish form of love that the Greeks called agape? Can it account for the experience of feeling called to sacrifice for others even when our own DNA may be placed at risk? While evolutionary biologists proffer various explanations for human behaviors that undermine the efficient propagation of our genes, there is something about those claims that rings hollow to us.

The notion that science alone holds all the secrets of our existence has become a religion of its own. The faith of Dawkins and others in biology seems even greater than the faith of the simple believer in God. Science is the proper way to understand the natural, of course; but science gives us no reason to deny that there are aspects of human identity that fall outside the sphere of nature, and hence outside the sphere of science. For most believers, God has no meaning unless God is more than nature. If God is more than nature, then studying the natural may never reveal the true mystery.

In the end, we must acknowledge that we human beings have only scratched the surface of self-understanding. The structure of DNA does hold considerable interest for this line of inquiry; but it would be the purest form of hubris to take our rudimentary knowledge of our genetic code, craft theories about it with our puny minds, and declare that nature has once and for all trumped nurture and toppled God. This is the kind of arrogance that humans alone seem to possess, and that genes alone could never explain.

File Date: 7.12.01