Resistant to Failure
In 2006 Duke University clinician Vance Fowler found the perfect animal model to investigate a question that had been bugging him ever since he started his residency at the university’s medical school in the mid-1990s: Why were some patients much better at fighting off bacterial infections than others?
Scanning research on more than 20 different inbred mouse strains, Fowler learned that they had a dramatic range of response to infection with the gram-positive bacterium Staphylococcus aureus. At one extreme was a strain that easily succumbed to infection: A/J (commonly used in cancer and immunology research because of its propensity to develop tumors). At the other was a highly resistant strain called C57BL/6J—the mouse strain most widely used in research.
As a clinician with no lab of his own, however, Fowler relied on longtime collaborator and then head of the National Institute of Environmental Health Sciences (NIEHS), David Schwartz, for bench space and experimental guidance. Shuttling back and forth between the institute and the university (only a 10-minute drive), Fowler and his postdoc, Hitesh Deshmukh (who worked full-time in Schwartz’ lab), hoped to pinpoint the genes involved in resistance or vulnerability to S. aureus infection using the A/J and C57BL/6J strains.
Within a few months, however, Fowler’s research came to a screeching halt. Following several misconduct investigations and complaints from NIEHS researchers, Schwartz, who had come to the agency from Duke, was asked to resign from his position.
As a result, Fowler and Deshmukh were forced to cut off all communication with Schwartz and join a dozen or so other Duke researchers in an exodus from the institute. “We were kind of like the Israelites, wandering around out there in the desert for 40 years,” Fowler says. “We had no lab space, we had no mice,” Deshmukh adds. “We were basically on the street.”
It was a shock to both researchers, who had to leave behind the comfortable accommodations of Schwartz’s NIEHS lab to share half a bench and a hood in a Duke lab that studied mycobacterial and fungal infections.
“I had a table next to a trash can, and right in front of me there used to be a stack of petri dishes growing very weird molds,” Deshmukh remembers.
It took Deshmukh a week to rid the hood of endotoxins produced by the fungi, which interfered with their bacterial cultures. But the duo pressed on, thanks in great part to what Fowler calls “a very lucky break.”
It so happened that a Maine-based genetics company had developed chromosomal substitution strains of the very mice Fowler and Deshmukh were interested in. At $250 per mouse, these animals were genetically C57BL/6J mice save for one chromosome pair per strain, which was substituted by one from an A/J mouse.
“Instead of looking at an entire genome,” Fowler says, “with these mice we could now ask the question: Is the susceptibility-associated gene found on A/J chromosome 1, or chromosome 2…?” Using these mice would save them months, if not years, of painstaking genetics work.
So in 2007, with a new R01 grant from the National Institutes of Health, Fowler acquired the entire panel of chromosomal substitution strains (21 total: one for each autosomal chromosome pair plus one for each sex chromosome), and Deshmukh monitored how each one responded to S. aureus infection when compared with the two parental stains.
By the summer of 2008, he had zeroed in on three A/J chromosomes—8, 11, and 18—that could render a resistant C57BL/6J mouse vulnerable to S. aureus.
Although a giant leap forward, this still left the team with 4,211 genes on the three chromosomes to sift through. Luckily, by that time Fowler’s team had expanded enough to occupy the entire lab, which Fowler eventually inherited from its former owner.
The new phase of the research—pinpointing the genes within those three chromosomes responsible for regulating an immune response to S. aureus infection—fell to Fowler’s new postdoc, Sun-Hee Ahn.
By combining mRNA expression data with quantitative trait loci analysis, Ahn was able to narrow the list of genetic elements likely involved in conferring S. aureus vulnerability down to 10 genes on chromosome 18—of which two were found to regulate the host’s cytokine response to the bacterium.
“Now that we’ve identified these genes in mice, the question is, do they matter in humans?” Fowler says. For the answer, he’s combing hundreds of human blood samples he’s been collecting from patients with S. aureus infections at Duke since the mid-1990s.
For Fowler, his promised land came into sight when he was finally able to publish his results in PLoS Pathogens in September 2010. “It was a long saga, but it had a happy ending,” he says.