Without new biomedical advances, the world will soon experience a "rising tide of blindness," according to HHMI investigator Val C. Sheffield. As people age, they are more likely to have vision problems. More than one-third of the population older than 75, for example, suffers from a heritable deterioration of the eye known as age-related macular degeneration (AMD). Yet fighting blindness in an aging population will require cracking what has become the single toughest problem in biomedical research: How do our genes and our experiences combine to produce common ailments that have complex origins, including cancer, heart disease, obesity, and vision problems?

Sheffield, his colleague and fellow HHMI investigator Edward M. Stone, and their collaborators have pursued this problem largely by looking for defects in single genes that produce marked deficiencies in a person's "phenotype," or the form and function of a person's body. "We've been going after the low-hanging fruit," says Sheffield. "We said, 'Let's look at the genes that have a big effect on phenotype. Maybe there are variants in those genes that cause common disorders.'"

Their greatest accomplishment to date has been the discovery of a genetic mutation that may account for 3 to 4 percent of all glaucoma cases, including a substantial portion of early-onset glaucoma. Still, most glaucoma patients have no mutations in the gene, meaning that other genetic defects or environmental factors also must be in play.

Other genes have given tantalizing signs that they may be involved in widespread vision problems, but the evidence has been hard to pin down. When defective, one such gene, ABCA4, produces a rare disorder known as Stargardt's disease. Some geneticists had suggested that mutations in the gene might be responsible for a substantial fraction of AMD cases. But when Sheffield, Stone, and colleagues investigated ABCA4 in patients with AMD, they found no evidence of a link.

Sheffield and other geneticists have also discovered unexpected complexities even in single-gene disorders. An example is Bardet-Biedl syndrome, a recessive disease that causes vision loss, obesity, extra fingers and toes, and other physical problems. Researchers have identified at least eight genes that when mutated can cause the syndrome. (The specific functions of seven of these genes remain unknown; the eighth seems to code for a chaperonin, a protein that helps other proteins fold into their correct shapes.) However, the same genetic mutation in different people can produce distinct phenotypes. "That difference gives us an opportunity to ask why," says Sheffield. "Is it the environment, or is there a genetic background effect, so that other genetic variations modify the phenotype?"

Answering these kinds of questions will be essential to understanding the biochemical mechanisms of disease, according to Sheffield. "What if the genome is more mutable than we think?" he says. "What if variants arise that create a susceptibility to disease, and these variants are due to genomic structure." For example, Sheffield has found the same DNA deletion in two people who have completely different patterns of surrounding genetic variation. That means the mutation must have occurred in two people independently. "If it arose twice, it probably arose more than twice," Sheffield says. If particular pieces of DNA are predisposed to break in the same way, common diseases could arise through mutations that are much more common than expected.

That's one reason why it's important to correlate individual genes with disorders in as many people as possible, says Sheffield. "Otherwise, we lose a tremendous opportunity to learn about the functions of those genes. We will overlook the science that those genes can teach us."

—Steve Olson

Download this story in Acrobat PDF format.
(requires Acrobat Reader)

Reprinted from the HHMI Bulletin,
December 2003, pages 24-27.
©2003 Howard Hughes Medical Institute