By David N. Leff

Scientists have found the genes that perpetrate cystic fibrosis, sickle cell anemia and Huntington's chorea, for example, but not those that cause Type 2 diabetes, essential hypertension, obesity, coronary heart disease and schizophrenia. Yet the latter five ailments are vastly more common and life-threatening in society than the first three.

So far, biotech research has focused on maladies falling under the "one gene, one disease" umbrella. Genetic linkage analysis technology works for such monogenic targets. The sequence-mapping tools to identify disorders caused by two or more genes acting in concert weren't there. Now they are.

DNA microchip arrays are beginning to take on the complex, polygenic diseases.

A paper in Nature Genetics for January 1999 reports what its lead author describes as "the first genes for a complex trait that have been identified in a linkage study." That author is molecular diabetologist Timothy Aitman, a senior lecturer at the Imperial College School of Medicine in London. His principal co-author is Lawrence Stanton, group leader for genomics technology at Scios Inc. in Sunnyvale, Calif.

Their article bears the highly explanatory title: "Identification of Cd36 (Fat) as an insulin-resistance gene causing defective fatty acid and glucose metabolism in hypertensive rats."

In fact, when the co-authors went hunting with a shotgun for a polygenic nestful of defective metabolic genes, it turned out that a single-shot rifle was more in order.

"The actual fatty acid defect that we mapped," Aitman told BioWorld Today, "turns out to cause a single-gene disorder. Its gene, Cd36 on human chromosome 7, encodes a fatty acid transporter protein. We had no idea about that at the time."

Aitman went on: "The insulin resistance phenomenon of Type 2 diabetes, however, is definitely polygenic. In 1997, we identified three insulin-resistance genes, on rat chromosomes 4, 12, and 16. The identities of those genes are not yet known."

He pointed out that the hypertensive rat model is not fat. "It was developed in Japan over 20 years ago and studied very intensively from the point of view of hypertension."

This idiopathic high blood pressure disorder remains an etiological mystery. "Genes that cause hypertension," Aitman noted, "have not been identified. Insulin resistance may be related to hypertension, or it could be an innocent bystander; we don't know at the moment which.

"If Cd36 is related to hypertension," he went on, "then that could be why the spontaneous hypertensive rat is insulin resistant, because insulin resistance also causes hypertension. There is a strong correlation between high blood pressure in humans and insulin resistance, which is why we decided to study this animal."

Aitman and his co-authors elected to pursue a differential expression approach. "Larry Stanton of Scios," he recounted, "was instrumental in allowing us access to microarray technology, in order to identify differentially expressed genes. I think it's a generally applicable approach, particularly for finding disease genes in the metabolic pathways on which they act.

How Microarray Technology Won The Gene

Scios' director of molecular biology, Tyler White, told BioWorld Today how that technology pinned down the Cd36 insulin resistance gene:

"When we print things on a microarray," he recounted, "there's a collection of known and unknown genes. We prepared a rat heart cDNA library and randomly picked 10,000 clones from it. Each clone was treated individually in an isolated format, and PCR used to generate DNA copies of each. Then those 10,000 DNAs were individually printed robotically onto glass microscope slides.

"Stanton's experiment," White went on, "was to take RNA from normal rats and from rats that were spontaneously hypertensive, then ask the question of our microarray of about 10,000 rat genes what sequences were differentially expressed between the two.

"That is," he explained, "seeing what things were up-regulated, what things were down-regulated, and comparing them in the situation of the normal rat and the spontaneously hypertensive rodent."

The basic answer to that basic question was, "Obviously there are lots of differences," White said. "Most things didn't change at all, or not to the level of the system's detectability. So those things didn't really go up or down. But a subset of them did."

White came to the point: "One of the things that was very much differentially regulated between those two situations was the Cd36 gene. Its protein's exact function is not totally nailed down."

Aitman observed, "We are trying now to work out the precise genetic details as to why this gene is defective in this model, and the precise biochemical changes that are occurring. We're looking to see whether human Cd36 protein deficiency or malfunction causes human disease, particularly diabetes, hypertension, hyperlipidemia and coronary heart disease.

"We are really chasing the human angle as hard as we can. We think there's quite a good chance that Cd36 will be involved in human disease, at least in some populations. The Japanese, as well as the Afro-Caribbean and some Far-Eastern peoples, do have Cd36 deficiency, and it will be intriguing to see what sort of diseases those individuals get.

"I'm sure the data will be applied clinically," Aitman continued, "because if the gene causes disease in humans it will be of diagnostic use. Even if it doesn't," he concluded, "I think there may be a desire to target therapeutic drugs against Cd36, or other genes on the same metabolic pathway." n