TR: How do scientists uncover all the genes involved in complex diseases, let alone determine the role of each gene?VMcK: Investigators still have to find each gene one at a time, using the same process they employ for a single-gene disorder. But mathematicians and computer scientists have now devised methods for determining whether additional genes beyond those already identified are likely to be related to a complex-gene disorder. For example, Neil Risch, a distinguished biostatistician and professor of genetics at Stanford University, has proposed several approaches.
One entails working with genetic “markers”-known bits of DNA whose location on the genome has already been found. Risch has extended a long-understood notion: that if the same marker is found in many affected individuals, researchers can deduce that a gene that plays a role in the disease lies near the marker. Risch has determined that the frequency with which pairs of siblings with the same complex-gene disease have the same marker indicates how much of a role the nearby gene plays in the disease. The statistical findings tell investigators whether they have to search for other genes elsewhere. The technique is a little mind-boggling to someone who’s not a biostatistician but it works.
As you can see, medical genetics relies on contributions from many fields, and it especially benefits from people who combine disciplines, for instance, those who keep one foot in computer science and the other foot in biology.
TR: And the third trend?
VMcK: The exciting effort to compare the genomes, or entire genetic complexes, of different species. Right now researchers are determining the genetic makeup of the genomes of the mouse, the fruit fly, a very simple roundworm called Caenorhabditis elegans, and a variety of bacteria. Several bacteria have already been completely sequenced, as has the yeast, a relatively advanced organism in that each yeast cell has a true nucleus. Because computer databases now contain genome information of various species, including the functions of identified genes, scientists can compare a particular human DNA sequence with similar sequences of other creatures. Researchers do this when they think they have found, for instance, a human DNA sequence with the characteristics of a gene, but don’t have the foggiest idea about its nature. Using the databases, they can see if the human sequence closely resembles, say, a yeast gene with a known function.
TR: Why would our genes function like those in yeast cells?
VMcK: Humans obviously need many more genes than yeast-probably at least 12 times as many-but the fundamental program for cells is the same in both species. Consider, for example, what happens in a form of colon cancer known as hereditary nonpolyposis colon cancer (HNPCC). Investigators have figured out that this disease occurs because of defects in a class of genes called mismatch-repair genes. Those genes normally survey the genome and repair bits of DNA upon finding so-called mismatches between the two sides of the coiling DNA ladder. But a mutation in the mismatch-repair genes themselves can cause that check-and-balance system to go awry. When the resulting unrepaired mismatch is in a gene that normally suppresses tumors, such as of the colon, the result is the development of a cancerous tumor. For quite some time researchers knew only that mismatch-repair genes existed in bacteria and yeast. Then Bert Vogelstein, a professor of oncology at Johns Hopkins University School of Medicine, used a database to check a genetic sequence he had isolated from people with HNPCC against sequences of other species. He found that the sequence he had isolated was a mismatch-repair gene, and went on to show that mutations in this and related genes could lead to HNPCC.
TR: Given all these new directions, geneticists would seem to have their hands full for quite some time.
VMcK: Oh, but there’s more. Once the human genome project nears completion, we will have to face the fact that we do not know the distribution of various forms of genes around the world. That’s important for determining what diseases, both infectious and degenerative, particular populations are prone to. Experts have therefore been discussing the idea of a human genome diversity project, which would look at gene variability.
Some human-rights groups have reacted to the idea with alarm, because early advocates emphasized the notion of sampling the DNA of small, isolated populations in the Amazon and other out-of-the-way parts of the world before the groups disappear. Advocates for these groups worried that the populations would be exploited or stigmatized by the studies.
But the scientists promoting the human genome diversity project have started to recognize that they should instead focus on the rest of us. After all, the main pay dirt won’t be found by studying small populations but by understanding the larger groups around the globe. That information can help researchers to determine future health prospects of large populations-groups with millions of members.
TR: Don’t researchers already have a sense of how different diseases, such as stomach cancer, vary across the globe?
VMcK: To some extent, yes. For instance, we have a good deal of information based on studies from blood. But to get to the roots of disorders we need to examine the genes. After all, the variation of genes is related to the frequency of complex traits. Different populations can have different variations of those polymorphic genes I mentioned before-the ones that can have multiple forms-with the result being varying frequencies of specific complex disorders.
Of course, as I suggested before, environmental conditions also play a role in disease distribution. Diabetes is a good example. One of its two main forms, which usually occurs in people once they are adult, is often related to obesity. Increasingly sedentary populations are more apt to experience this kind of diabetes.