Since the release of the human haplotype map, or HapMap, genomewide association studies have uncovered a plethora of susceptibility loci for diseases from asthma to diabetes. But as noted by Mark Daly, HMS assistant professor of medicine at Massachusetts General Hospital, when it comes to association studies, “We don’t get biological insight just from a highly significant p value.”

“The key steps after discovering an association,” Daly said, “are to identify the precise causal variant, identify how it influences the action of a gene, and then connect that molecular function to a biological process that influences disease.”

Aiming to take these key steps and identify how a particular susceptibility locus affects disease risk, Daly began collaborating with gastroenterologist Ramnik Xavier, also an HMS assistant professor of medicine at MGH, to study the genetic component of Crohn’s disease. Characterized by inflammation of the digestive tract, Crohn’s is known to have a substantial genetic component and more than 30 loci have been linked to it. Despite these genetic clues, the molecules and pathways underlying disease pathogenesis have proven difficult to pin down.

Reporting in the September issue of Nature Genetics, co–senior authors Xavier and Daly, who is also affiliated with the Broad Institute, and colleagues describe a polymorphism upstream of a gene called IRGM that is the likely causal variant behind a previously identified SNP association. Variation within the coding region of IRGM, a poorly understood autophagy-related gene, had been excluded by previous studies, so the researchers had to look beyond the gene itself.

Using a new genotyping method that more finely parses the DNA, Steven McCarroll, a postdoctoral fellow working with Daly, identified a 20kb deletion that was highly correlated with the known SNP and with Crohn’s disease risk. Previously, these smaller copy number variants were nearly impossible to detect, but McCarroll spearheaded the development of a hybrid array that simultaneously analyzes SNPs and copy number variants; this technology accelerated his search for polymorphisms. The development of this array technology, published separately, appeared online Sept. 7, also in Nature Genetics.

As one might expect given its location upstream of the IRGM gene, the deletion polymorphism affects IRGM expression. But the mutation did not simply produce a “crippled promoter and reduced expression,” according to McCarroll, lead author of the study. Instead the researchers saw that the deletion actually stimulated IRGM expression in some cell types, such as primary smooth muscle cells, while suppressing it in others, including lymphoblastoid cells. These results suggest that the overall pattern of IRGM expression in individuals carrying the deletion, who are at higher risk of developing Crohn’s, is different from the pattern in individuals without the deletion, even though the IRGM protein is identical in both populations.

To determine how IRGM expression levels affect cellular function, the researchers reduced IRGM expression in cells using short interfering RNA. They saw that IRGM-deficient cells were impaired in their ability to destroy via autophagy potentially pathogenic bacteria. Specifically, when challenged with Salmonella, IRGM-deficient cells failed to efficiently encapsulate the bacteria in autophagic vesicles, the first step toward digesting invaders in the lysosome. Conversely, the bacterial load within autophagic vesicles increased with increasing levels of exogenous IRGM.

The possibility that deficits in autophagy might contribute to the pathogenesis of Crohn’s disease was first recognized when susceptibility loci were identified in or near two different autophagy-related genes, IRGM and ATG16L1. “Now,” said Daly, “genetics has given us hard evidence that the inherited susceptibility … of the disease to some degree resides in genes … that directly influence the front-line response to bacteria.”

Conflict Disclosure: The authors declare no conflicts of interest.

Funding Sources: The National Institute of Allergy and Infectious Diseases; Center for Computational and Integrative Biology, MGH; Lilly; the National Institute of Diabetes and Digestive and Kidney Diseases; the Burroughs Wellcome Fund; and the Crohn’s and Colitis Foundation of America