Esta página no está disponible en español.
THE WALL STREET JOURNAL
In Rural Texas, Scientists Seek A Genetic Cause for Diabetes
By SCOTT HENSLEY
February 21, 2002
A POPULATION AT RISK
Source: National Institute of Diabetes and Digestive and Kidney Diseases
Researcher Miriam Muiz talks with Estefana Castillo (center) and Elma Castillo in La Grulla, Texas. Elma Castillo is in a diabetes study.
(PHOTOS: David Pellerin)
RIO GRANDE CITY, Texas -- The blood of Ricardo Velasquez, a 54-year-old migrant farmworker, might hold the key that unlocks the mystery of diabetes. It's the job of Alda Clarke and Miriam Muñiz to find him and get a sample.
The two women, white-coated staffers for the University of Texas Health Science Center at Houston, drive a dusty white van around this impoverished border town looking for people to donate their time and blood for medical testing. Their goal is to understand the genetic risks for diabetes, a disease fast becoming an American epidemic. Here and in surrounding Starr County, diabetes is especially common and often fatal.
But locating Mr. Velasquez, a diabetic with a diabetic brother who already volunteered for the study, won't be easy. He lives in Las Lomas, a colonia, or unincorporated town, southeast of here. It's a hodgepodge of cinder-block and adobe dwellings about 10 minutes' drive from town, where telephones and names for the rutted dirt roads are in equally short supply. "We don't have a street address or know where to go," says Ms. Clarke, who does the driving.
The genetic map that the women are working to complete is about as sketchy as their mental map of Las Lomas. The publication a year ago of the draft set of hereditary instructions, or genome, of humans was a scientific landmark. But researchers still don't understand the normal function of most of the 30,000-odd human genes -- much less what happens if they go awry. So scientists around the world are looking for clues by studying populations that share a genetic heritage and a tendency to develop common maladies.
Here in Starr County, scientists have spent decades looking for diabetes genes. Already the seventh-leading cause of death in the U.S., diabetes strikes 800,000 new people each year and costs society nearly $100 billion a year, the American Diabetes Association estimates.
Starr County, a patch of scrubby lowlands along the Rio Grande River, is home to 54,000 people, many of whom trace their roots to a handful of families that settled the area 300 years ago. Nearly 98% of them are Mexican-American, the highest proportion of any U.S. county, and about half of all households have at least one family member suffering from Type-2 diabetes, which usually strikes in middle age.
The prevalence of diabetes here caught the eye of researchers at the University of Texas more than 20 years ago. In 1981, Craig Hanis, then 29 and a newly minted geneticist, and several colleagues set up shop in an apartment and its two-car garage downtown to collect blood samples, medical information and family trees from diabetics in the county.
During early house-to-house canvassing, Hilda Guerra, who now manages the project in Rio Grande City, came upon the patriarch of a prominent clan who had been diagnosed with diabetes. She asked him to help round up his numerous relatives, and he agreed to throw a party to bring them together.
At dawn one Saturday several weeks later, he assembled 42 members of his extended family for a battery of medical tests that took three hours per person. That evening the researchers stowed their clipboards and partied with the family in the patriarch's stately home until midnight. Dr. Hanis, who came to Texas from the University of Michigan, did his best to fit in: "It was the first and last time I had menudo," he says, referring to the aromatic, tripe-based Mexican soup.
News that one of the town's leading families had aided Dr. Hanis led to a surge in volunteers. A year later, Dr. Hanis helped establish that Mexican-Americans are nearly twice as likely as non-Hispanic whites to develop diabetes as adults. But he soon ran into a dry stretch as long as the Texas horizon. For nearly a decade, he searched in vain for flaws in diabetics' genes for insulin, the essential hormone that regulates blood sugar, or for the insulin receptor on the surface of cells. He found none, but he remained certain that genetic defects contribute to diabetes.
Taking a cue from research on other diseases, Dr. Hanis sifted through the genetic code of siblings afflicted with diabetes. On average, researchers say, if you compare DNA of one sibling with another, half of it will be identical. But if a common gene is contributing to illness, the chances of a match along the stretch of DNA where the disease gene lurks is better than 50-50.
In 1992, Dr. Hanis joined forces with Graeme Bell, a University of Chicago diabetes researcher. They figured it would take at least 300 pairs of diabetic brothers and sisters to pinpoint any diabetes genes. In less than a year, Dr. Hanis located 200 such pairs and soon thereafter received a $1.2 million grant from the National Institutes of Health to pursue a five-year study.
After years of painstaking lab work, they found a genetic fingerprint in 1996 in the blood donated by 346 pairs of diabetic siblings. A diabetes gene, or genes, lay near one end of chromosome 2, one of 23 pairs of human chromosomes.
But it would take two more years to isolate the culprit. Sibling studies, Dr. Hanis says, "tell us the ballpark," but not the "section, row and the seat that has the broken bolt in it." To draw the DNA dragnet tighter, the scientists compared the DNA sequence in the suspicious portion of chromosome 2 in 100 diabetics and 110 healthy volunteers from the county. In 1998, they found a previously unknown gene they named calpain 10. Every human has the gene, but nearly two-thirds of Starr County residents have a variant that, when inherited from both mother and father, raises the risk for diabetes by more than 14%.
The discovery opened a new chapter in diabetes research. Dr. Hanis says that experiments with animal cells inside test tubes confirm that calpains, the class of enzymes including calpain 10, are "centrally involved" in metabolism and diabetes, a role that wasn't previously suspected.
Nobody thinks calpain 10 is the diabetes gene. "It will be one of multiple genes that's involved additively to explain risk," says Alan Shuldiner, a researcher at University of Maryland in Baltimore. And there are environmental risk factors such as obesity and lifestyle.
Meantime, Ms. Clarke and Ms. Muñiz continue their door-to-door search for study subjects. Their search for Mr. Velasquez quickens after a social worker in the neighborhood gives them directions to his house, a low shed with a corrugated steel roof. Ms. Clarke knocks on the door, and at 10:16 a.m. she has her man. He agrees to an appointment at the clinic and takes a plastic cup for a urine sample. Mr. Velasquez is the first of four successes in the women's 11 attempts to find volunteers -- a pretty good day, they say.