Genome Scanning Technique Spots Disease Risk Through Sorting Ancestry Mix

Admixture Mapping Links Chromosomal Region for First Time with Multiple Sclerosis

Photo by HMS Public Affairs staff Comparing genomes of people with recently mixed ancestry can unveil common genes that contribute to prevalent diseases, according to the first successful demonstration of the method reported by (clockwise from left) David Reich, Philip De Jager, Arti Tandon, and their co-authors.

A certain genetic inheritance from European ancestors may put African Americans at higher risk of multiple sclerosis, report researchers who used a powerful new method to find a previously undetected chromosome region associated with the disease.

The study may be the first successful demonstration of a new genetic method that is effective only in populations of mixed ancestry, said David Reich, HMS assistant professor of genetics, who led the study.

The newly identified region on chromosome 1 contains about 70 genes. Reich and his colleagues are trying to narrow down their list of suspects in hopes of identifying a new target for a drug or ferreting out a pathway important in the disease.

The disease has a proven genetic component, but the culprit is likely to be a risky combination of many ordinary gene variants, plus environmental factors such as common microbial infections, rather than any single causative gene, said senior author David Hafler, the Jack, Sadie and David Breakstone professor of neurology at HMS and Brigham and Women’s Hospital. Pinpointing multiple genes with small effects may provide a roadmap for developing effective therapies that may only need to make slight modifications in disease-related pathways.

But first, researchers must track down these common allelic variants, a search that has proven mostly futile so far. Except for the huge HLA region on chromosome 6, which codes for the antigen-presenting molecules on cells, no underlying MS disease risk genes have been found. Researchers looking for the genetic contributions to other prevalent complex diseases, such as diabetes, have run into similar dead ends.

Shared Risk
For their study, reported in the October Nature Genetics, Reich, Hafler, and their colleagues used a novel twist on the genetic association study, one of the two main ways to scan the genome for disease genes. Association studies look for genetic variations that occur more frequently than would be expected due to chance in people with a disease.

The new technique, called admixture mapping, can only be done in populations of recently mixed ancestry, such as African Americans or Hispanic Americans. The method is most likely to work for cases in which the disease incidence—and presumably the risk variants—vary dramatically across human populations.

Interestingly, MS affects about twice as many people of European ancestry as it does African Americans, and the disease is virtually unheard of in Africa, said Hafler, who is a scientific leader of the International Multiple Sclerosis Genetics Consortium, which is also conducting a comprehensive and detailed scan for risk genes of European ancestry. Human populations share extremely similar genes, but scientists have reasoned that the few genetic variations that have risen to high frequency in Europe may hold a key to understanding and treating MS.

In America, Reich said, the two genomes began mingling about six generations ago, leaving people who now identify themselves as African American with genomes composed of 20 percent European ancestry and 80 percent African ancestry on average.

“Our genomes are a mosaic of our ancestors,” said Reich, who with Hafler is also an associate member at the Broad Institute of Harvard and MIT. “We are asking, if you trace a segment of DNA back six generations, where did it live, in West Africa or Europe?”

Since those great-great-great-great grandparents first mated and mixed their genomes, one or two recombinations per chromosome occurred in each generation, leaving people who call themselves African American with enormous blocks of African ancestry interspersed with chunks of European origin. The large stretches mean that fewer genetic signposts are needed to identify each segment of contiguous European or African ancestry, which translates into a fast and cheap way to do a genome scan.

Ancestral DNA
Working closely with Broad statistician Nick Patterson, Reich identified and validated the few signposts along the human genome that were very common in one ancestry and uncommon in the other. This map and the new statistical methods the researchers developed were published in May 2004 in The American Journal of Human Genetics. Typically, the markers, known as single nucleotide polymorphisms (SNPs), were in regions of the genome without genes. Among more than 3,000 SNPs, only one was found with 100 percent frequency in one population and zero in the other, and it involved the gene that confers resistance to malaria in West Africans.

“We are asking, if you trace a segment of DNA back six generations, where did it live, in West Africa or Europe?”

Next, Reich and his colleagues applied their new map and methods to the genomes of 605 African Americans with multiple sclerosis, many of whom volunteered to participate in genetic research in response to a national appeal by African-American talk show host Montel Williams, who has been diagnosed with MS. Among African Americans, the strongest correlation between the disease and European ancestry came along a region of chromosome 1.

The study did not pick up any correlation on chromosome 6, which hosts the HLA region, the only consistently replicated genetic association with MS. But this is unsurprising because there are no significant genetic differences between the European and African ancestry populations there.
For now, the researchers are treating their results as promising evidence that admixture mapping may be a valuable additional tool that can uniquely locate risk genes for common and complex diseases. No single genome scanning technique can do it all, and it is too soon even to say that admixture mapping works. “We haven’t found a gene yet,” Reich said.

Reich is working with other collaborators to apply the same method to other diseases, such as type 2 diabetes (more common in African Americans), osteoporosis (more common in Europeans), prostate cancer (more common in African Americans), ovarian cancer, (more common in Europeans), lupus (more common in African Americans), hypertension (more common in African Americans), and end-stage renal disease (more common in African Americans).

Meanwhile, the investigators continue to recruit more research participants. “We’re working hard to engage African Americans in this and other studies,” Reich said. “Like all clinical studies, the findings will be most relevant to the population studied, African Americans, but it will also be relevant to other populations.” Those interested in participating in the study may contact Scott Fruhan, the MS genetics susceptibility study coordinator, BWH Center for Neurologic Diseases, at 617-525-5314, sfruhan@rics.bwh.harvard.edu, or http://haflerlab.bwh.harvard.edu. People may also get in touch with the MS Genetics Group clinical coordinator at the University of California, San Francisco, neurology department, at 866-674-3637, msdb@itsa.ucsf.edu, or www.ucsf.edu/msdb.

—Carol Cruzan Morton

top