November 04, 2010
Genetic Analysis Shows What it Takes to Control HIV Naturally
Through an international study of the human genome in persons who control HIV without the need for medications, scientists have identified a handful of discrete amino acids that seem to predict a person’s ability to successfully keep HIV in check. The amino acids help determine how the immune system recognizes HIV-infected cells.
The findings may eventually offer clues that will aid the development of vaccines to prevent AIDS or improved treatments for infected individuals, notes senior author Bruce Walker, a Howard Hughes Medical Institute researcher at Massachusetts General Hospital (MGH). He is also a professor of medicine at Harvard Medical School and director of the Ragon Institute of MGH, the Massachusetts Institute of Technology, and Harvard.
“In some ways we've thought of T cells as being ‘the more, the better,’ but in fact this result. . . indicates that there are really big differences in the quality and function of immune responses that are being generated to HIV.”
Bruce D. Walker
Walker and his colleague Florencia Pereyra are leading the HIV International Controllers Study, a major global research effort to understand how a small fraction of people who are infected with HIV but have never been treated can fight the virus with their immune system. These people are what Walker and his colleagues call elite or viremic virus controllers. Elite controllers keep their HIV viral load to below 50 virus particles per milliliter of blood; viremic controllers maintain a range between 50 and 2,000 particles. The average untreated patient has a viral load of more than a million particles at the time of acute infection.
The range of outcomes among those infected with the virus is striking: although 1 in 300 individuals can keep the virus at bay without drugs, some develop AIDS within six months of becoming infected. "Many patients had been asking their physicians for years, 'Why are my friends getting sick and I'm not?'" Walker says, adding that these individuals were motivated to participate in the HIV International Controllers Study to get an answer.
The first genome-wide association study of these individuals, published online November 4, 2010, in Science Express, is the culmination of hundreds of collaborations with clinics and institutions all over the world to identify controllers and collect and analyze samples of their blood.
In the mid 1990s, Walker and colleagues first became aware of HIV controllers, as it gradually became apparent that a small portion of infected persons were remaining healthy. In the early 2000s, the scientists realized that although HIV controllers were uncommon, there were probably enough of them in the world to recruit them for a gene study to determine how their immune system responded to the virus. In 2006, with an initial donation from philanthropists Mark and Lisa Schwartz, Walker's team began to apply results from the International Hapmap Project — an international effort to understand common patterns of genetic variation in humans — to analyze the genomes of HIV controllers and progressors (those who respond to the virus more typically) of European, African-American and Hispanic descent.
In the current study, Walker’s team performed a genome-wide association study comparing the DNA of 974 HIV controllers to that of 2,648 individuals whose disease had progressed in a more typical fashion. This analysis involved assessing over 1 million variants, called single nucleotide polymorphisms (SNPs), in each subject, and revealed over 300 SNPs linked to the control of HIV. All of those SNPs resided on chromosome 6. The region contains so-called 'HLA genes,' which produce important immune system molecules contained in most cells in the body. HLA molecules grab pieces of the proteins made by an invading virus and display their fragments on the cell's surface, so that other cells can recognize and kill the infected cells.
HLA gene variants had been linked to better or worse HIV outcomes before. More surprisingly, the group found no gene variations outside of the region that are linked to the ability to control the viruses. "The fact that [the hits are] no place else is helpful, because it really helps you focus on what's important," Walker says.
The SNP data told the investigators the region of the gene was involved in control, but not whether the SNPs identified were actually mediating this effect, or whether they were simply associated with a neighboring unmeasured SNP that was causal. Further definition would require in-depth sequencing of the human genome in this region. With the help of previously published HLA data and a computational model developed by a medical student on the team, the group was able to see exactly whether gene variants tracked with control or lack of control. "When we did that, it was like suddenly turning the microscope from low power to high power," Walker says. In particular, six different amino acid residues, five of which lined the groove where the HLA peptide binds with the viral pieces, seem to be important in HIV control.
Some of the amino acids are contained in one subclass of HLA — the HLA-B — and Walker suspects they are somehow influencing the way the peptide is being presented to T cells, which then affects the activation of the T cells that recognize and destroy the infected cells.
Walker's group plans to examine T-cell signaling in the context of different HLA conformations. "In some ways we've thought of T cells as being ‘the more, the better,’ but in fact this result, plus some other data that we've recently published, indicate that there are really big differences in the quality and function of immune responses that are being generated to HIV," he says.
The new results help reveal what an effective natural immune response is and how it's induced. While he cautions that scientists are still a long way from an effective vaccine for HIV, Walker's hope is to use that information to generate more predictably effective vaccine-induced immune responses to prevent or treat HIV.