icon-folder.gif   Conference Reports for NATAP  
  10th Conference on Retroviruses and Opportunistic Infections
Boston, Mass, Feb 10-14, 2003
Back grey_arrow_rt.gif
Impact of Host Genetics on Clinical Outcomes
Ronald Swanstrom, Ph.D.
UNC Center For AIDS Research, University of North Carolina at Chapel Hill
  One of the Tuesday afternoon symposia at the 10th Retrovirus Conference titled "Impact of Host Genetics on Clinical Outcomes" reviewed the role of human genetic variation in transmission and clinical progression of HIV-1. There is great variability in the rates of clinical progression. In a few cases slow clinical progression has been linked to transmission of an attenuated virus. However, human variation is likely responsible for the vast majority of the variability in clinical progression that typifies HIV-1infections.
In the first presentation Sunil Ahuja (UT San Antonio) ("Genetic Condoms of HIV-1 Infection: Using Host Genetics to Dissect HIV-1 Pathogenesis") described studies of genetic polymorphisms in a cohort of 1300 HIV-1+ adults and 850 perinatally exposed children. In one study the relationship between polymorphisms in the CCR5 gene and clinical progression CCR5 haplogroups defined as nucleotide polymorphisms in the cis-region upstream of the CCR5 coding domain was examined. These haplotypes include a polymorphism in the nearby CCR2 coding region (at codon 64 encoding either valine or isoleucine). The G*2 haplotype in the child either as homozygous or as heterozygous with the C haplotype was associated with reduced mother-to-child transmission,while the E haplotype either as homozygous or heterozygous with either the C or G*2 haplotype was associated with increased transmission rates. Similar associations were seen with disease progression, although in comparing different adult cohorts race-associated haplotypes must be considered. The extreme polymorphism known for CCR5 is a 32 base pair deletion that is protective from infection with an R5 virus (one that uses CCR5 as a coreceptor). It is not clear what the mechanism of protection or enhancement is for the haplotypes followed in these studies, although a possible explanation is that these polymorphisms affect the level of CCR5 expression. In a separate set of studies the potential role of MCP-1 (monocyte chemoattractant protein-1) in HIV-associated dementia (HAD) was examined. Previous studies have shown that increases of MCP-1 in the CSF precede neurological symptoms. A single nucleotide polymorphism (SNP) guanosine at position 2578 upstream of the MCP-1 coding domain was associated with an decreased risk of acquiring an HIV-1 infection but an increased risk of clinical progression and HAD. This genotype has been associated with increased level of gene expression and protein production, and with increased serum levels of MCP-1. One possible mechanism is that the increased levels of MCP-1 result in increased recruitment of monocytes into the brain potentiating HAD.
The second presentation was from Dr. Mina John, Royal Perth Hospital, (speaking in place of Dr. Simon Mallal) on "HIV Adaptation to Genetic Polymorphisms of the Host." Dr. John first reviewed an ongoing study into the genetic basis of abacavir hypersensitivity. ABC HS has been associated with the presence of a specific MHC HLA 57.1 ancestral haplotype (Mallal et al., Lancet 359:727, 2002). All people with the full haplotype had some reaction to ABC, while no patients without any part of the haplotype were HS. The region of the HLA locus that confers HS has now been mapped to a 14 kbp region within the HLA locus (poster 713). The next study presented is based on the hypothesis that HIV-1 sequence polymorphisms (here studied within the RT coding region) reflect selection on a population basis for HLA escape mutants within specific HLA epitopes (Moore et al., Science 296:1439, 2002). The analysis has been done with the Western Australian HIV Cohort involving 473 people with a total of 2303 person-years of observation. Not all CTL epitopes are capable of changing as some are constrained by requirements of protein function. A multivariate analysis using logistic regression was done correlating the positions of variability with HLA alleles in the population. This analysis revealed positions of variation that corresponded to known CTL epitopes and led to the suggestion that other sites of variation may represent unidentified CTL epitopes.
This study was extended to include sequencing of the entire viral genome to determine the level of adaptation to the HLA haplotypes present in a subject by analyzing the presence of nonconsensus amino acids at predicted CTL epitopes. This analysis revealed a relationship between the level of adaptation and the viral load with 40-60% adaptation corresponding to a mean viral load of 11,000, 60-80% adaptation to a mean viral load of 67,000, and 80-100% adaptation to 107,000. However, CTL escape mutants may also result in a fitness loss that may result in a reduced viral load or slow adaptation at these sites. HLA B57, HLA B27, and HLA-Bw4 were all seen to be associated with reduced viral load in a univariate analysis. HLA-B8 was seen to be associated with a rapid loss of CD4 T cell loss early in infection likely through the effect of a KIR allele.
Analysis of protease variability suggested a similar situation and in this case there was a suggestion of overlap between CTL selected variability and a position of drug resistance. In this example a mutation at position 82 of the protease could relieve selective pressure of selection associated with a B7 haplotype and also from a protease inhibitor like IDV. This is an example of a potential synergistic selective pressure while antagonistic effects can also by hypothesized. In considering the implications for vaccine design it was pointed out that in this cohort there was significant variability both within the viral sequences and within the HLA alleles. An optimal vaccine was proposed that would maximize the number of epitopes for the population while minimizing those epitopes that had a negative association.
The third presentation was from Mary Carrington from NCI, NIH, examining the potential contribution of KIR alleles in NK cell function. NK (natural killer) cells function to kill cells that lack surface HLA class 1 molecules. One mechanism of immune avoidance is the down regulation of class 1 molecules to prevent the presentation of foreign peptide epitopes as part of the adaptive immune response. NK cells are a type of T cells that are part of the innate immune response. HLA class 1 is the ligand for inhibitory KIR molecules on the surface of the NK cell blocking the killing of the target cell by the NK cell. Aberrant expression of class 1 can engage an activating receptor on the surface of the NK cell leading to NK cell-mediated cell killing. Each of us has multiple copies of genes encoding KIR proteins and the genes themselves are polymorphic within the human population. Also the number of KIR genes can vary, including the number of activating or inhbitory KIR genes which could affect the outcome of a disease process. One example of these pathways is the KIR3DL1/3DS1 protein interactions with HLA-V Bw4. Two copies of Bw4 slow progression to AIDS. This effect appears to be associated with Bw4Ile80 alleles. This effect was further enhanced with the presence of KIR3DS1 suggesting an interaction between the encoded proteins. Distributions of these alleles vary by race and these differences may affect disease progression. These studies indicate that polymorphisms in the KIR locus can have an effect on disease progression but as yet little is known about their effects.
In the final presentation Dr. Richard Kaslow, University of Alabama--Birmingham, reviewed a number of studies on the role of human genetics in HIV-1 transmission and disease progression. The most significance was given to studies in larger cohorts where the effects had been seen in more than one cohort. For some studies he noted that the reported patterns of involvement of host genetic polymorphism have varied. The most consistent effect for an HLA locus has been seen with the B locus, with an effect on disease progression and the suggestion of an effect on transmission. The reported effects of the HLA A locus have not been as consistent.