"Walking" toward the CF gene was too slow. For geneticists, walking along a chromosome means using partly overlapping DNA fragments to move toward the target gene, one step at a time, checking each fragment to see whether it is inherited with the disease. It is an arduous task. The path is strewn with roadblocks—repetitive sequences of DNA or other stretches difficult to cross.

Tsui estimates that walking along the stretch between those two markers would take the average lab about 18 years.

They needed a shortcut. It was a time for imagination and ingenuity—as well as fierce competition. All thoughts of further collaboration vanished overnight as the seven research groups scrambled to be the first to find the CF gene.

Tsui's strategy was to bombard chromosome 7 with a large number of additional markers, which he created by cutting up and analyzing thousands of fragments of normal DNA from chromosome 7 "libraries." If he used enough markers, he reasoned, one of them was bound to be quite close to the CF gene. Eventually he hit on one that was. But he still needed to get from this marker to the gene.

So he contacted Francis Collins of the University of Michigan, who had devised an imaginative technique for "jumping" along chromosomes.

Though this technique is no longer necessary, now that the entire human genome has been decoded, jumping is five to ten times faster than walking. It allows researchers to cover 100,000 to 200,000 DNA bases at one time and simply leap over areas that might otherwise be difficult to cross. The technique involves snipping out a long segment of the DNA under study, labeling it at one end, and letting it curl into a circle. This brings the labeled end next to a sequence of DNA that would otherwise be thousands of bases away. The circular segment of DNA is opened up and used as a bridge over long stretches of DNA. Its far end is labeled as the starting point for another jump.

The researchers combined Collins' jumping technique with Tsui's marker. They walked, they jumped, and they walked some more.

Each DNA fragment they used to walk or jump with was compared to DNA from animal species. They found a match with a sequence from a gene of chickens, mice, and cows. This implied that the fragment was important because it had been "conserved" in different species in the course of evolution. But they still had no evidence that this had anything to do with the CF gene.

Finally John Riordan, a biochemist at Toronto's Hospital for Sick Children, found a match between a small part of the conserved fragment and a genetic message in sweat-gland cells—cells the scientists knew were involved in CF. The researchers established that a gene containing this snippet of DNA was also expressed in other tissues that are specifically affected by CF. Then the labs "began to sequence like crazy," Tsui says, "looking for a difference between DNA from normal and CF cells."

— Maya Pines

< Previous  |  Top of page  |  Next >

Francis Collins explains how a normal CF gene was pieced together in the lab out of smaller, overlapping fragments of DNA. Part of the gene's sequence is shown on the blackboard.

Photo: Burt Glinn/Magnum

© 2008 Howard Hughes Medical Institute. A philanthropy serving society through biomedical research and science education.
4000 Jones Bridge Road Chevy Chase, MD 20815-6789 | (301) 215-8500 | e-mail: webmaster@hhmi.org