Searching for the genie in the genome

HIV human genomics field evolving at warp speed

By Joseph F. John, Jr., MD, Associate Chief of Staff for Education, Ralph H. Johnson Veterans Administration Medical Center; Professor of Medicine, Medical University of South Carolina, Charleston. Dr. John reports no financial relationships relevant to this field of study.

Sources: Pelak K, et al. Host determinants of HIV-1 control in African Americans. J Infect Dis. 2010;201:1141-1149; Motsinger-Reif AA. Discussion of a genome-wide association approach to determine HIV-1 set point in African Americans. J Infect Dis. 2010;201:1118-1120.

Health care providers have been constantly amazed at the variation in response to infection with HIV-1. In our clinic, we now routinely see elderly patients who have been on various antiretroviral regimens over the years but still persist with high CD4 counts and low viral loads. They are truly long-term non-progressors. Work with whole human genome analysis to discover host determinants of HIV-1 control has gained steam in the middle of this current decade, and is continuing. The current study from the Center for Human Genome Variation at Duke and several government agencies is part of a DoD HIV NHS natural history study started in 1985 (http://www.idcrp.org/hiv-natural-history-study.html). The patients in this study continue to be seen every six months. Other cohorts like the MACS cohort are also included in the study. Using the HumanHap 1MDuo or Illumina HumanHap Bead Chips, 1,212,217 SNPs were analyzed. Complex statistical methods termed EIGENSTRAT were used to analyze the genetics of the human populations.

Study patients were from the U.S. Department of Defense Human Immunodeficiency Virus Natural History Study (DoD HIV NHS) or the Multicenter AIDS Cohort Study (MACS). Of 487 soldiers meeting criteria, 471 were successfully genotyped; 145 participants from the MACS cohort were also studied. No single, genome-wide SNP had an association with the set point, he viral load of which patients more or less stabilize after the burst of viral replication during primary infection. The most associated SNP in the African-American cohort was rs2523608, which is located on chromosome 6, in HLA-B gene, HLA-B•5703. Note the same association is true for set-point regulation in patients of European ancestry. This allele, therefore, is the most important variant for the entire group in this study as the genetic locus that would affect viral load set point and could account for about 10% of the variation in viral load set point. Likewise, 10% of the patients in these studies progressed during the course of the studies to CD4+ cell counts of < 200/mL within two years of seroconversion.

Other SNPs reported to have an association with viral-load set points were analyzed. For example, rs9261174, located near the ZNRD1 gene in MHC region, and CCR5- Δ32, a deletion in the CCR5 gene (rs33) that is rare in non-Europeans, showed no association in this African-American cohort with viral-load set point. So, mutations in alleles that are considered "rare variants," as shown from this study, may have significant effects on an increased burden of disease in some populations.

Commentary

This magnificent work aimed to define genome-wide variables that explain HIV progression has been made possible over many years, involving many willing and highly cooperative patients, and spending millions of dollars. We see how the genie of the genome can now open many secrets about HIV progression. In this study, it was surprising how few genetic variables were uncovered related to progression; in fact, it seems a rather obscure locus within HLA-B•5703. Nevertheless, there will be other regions recognized in the future. Our challenge will remain to find ways to exploit, in the future, the locus under question (and newer loci), and mechanisms that would silence or alter these alleles to allow better immune modulation of HIV infection.

There will be surprises along the way, like the one in the present study. A surprise, indeed, that the only uncovered genetic locus that affects viral-load variation in 515 African-Americans was the same one that is known to affect HIV+ patients of European ancestry. Along with the findings of genetic overlap among so-called races come implications that small segments of shared DNA among HIV-infected patients, regardless of racial background, are new targets for antiretroviral therapy or genetic immunomodulation. In an accompanying editorial, Dr. Allison Motsinger-Reif beautifully discusses the study that she thinks offers a "unique resource to study an important, understudied population — patients of African descent." She also makes a good point that using gene chips that are very generic will not uncover rare alleles.

Finally, Dr. Motsinger-Reif comments, and I would agree, that the ultimate regulation of viral load involves complex phenotypes. Clinicians would do well (and I keep telling myself!) to bone up on the evolving field of human genomics. Studies like this one, by Pelak et al, remind us that the field is evolving at warp speed.