Outcome to HIV-1 infection is dependent on a complex assortment of viral and host genetic factors. The contribution of HLA class I loci, particularly HLA-B locus, in regulating HIV-1 disease progression is the strongest among a dozen or so host genetic effects identified to date. The genetic influence of HLA class I on AIDS reflects the wealth of functional data that indicates the importance of Major Histocompatibility Complex (MHC) class I-restricted cytotoxic T cell (CTL) activity in controlling the virus. However, levels of CTL activity do not completely account for the extent of resistance to HIV-1 and additional branches of the immune system most certainly contribute to viral restriction. Natural killer (NK) cells participate in an early innate immune response against tumor- and virally-infected cells through a complex set of receptors that regulate inhibition and activation of NK cell activity (see Project number Z01 BC 10305-03 LGD). One group of NK receptors, the killer immunoglobulin-like receptors (KIR), recognizes certain HLA class I molecules and thereby inhibit or activate NK cell responses. We hypothesized that HLA-B might play a role in the innate immune response by behaving as a ligand for the NK cell receptor KIR3DS1, which is thought to participate in activation of NK cells. We typed for the presence or absence of KIR3DS1 in over 1,000 HIV-1 seroconverters and tested for potential effects of this gene, the gene encoding its putative ligand (i.e. HLA-B), and a combination of both. The data indicate that one activating KIR allele, KIR3DS1, in combination with HLA-B alleles that encode molecules having isoleucine at position 80 (HLA-B Bw4-80I) resulted in delayed progression to AIDS. In the absence of KIR3DS1, HLA-B Bw4-80I alleles were not associated with any of the AIDS outcomes measured. Conversely, in the absence of HLA-B Bw4-80I alleles, KIR3DS1 is significantly associated with more rapid progression to AIDS. Thus, an additive effect of the two unlinked loci would not explain the observed protection in individuals who have both variants, and only a model involving an epistatic interaction between the two loci appropriately fits the observations. The effect was most apparent on progression to CD4 T cell depletion, suggesting that a protective response of NK cells involving KIR3DS1 and its HLA class I ligands begins early after HIV-1 infection. These data were recently published. We have also completed typing of KIR3DS1 in over 200 high risk seronegative individuals to determine if this gene plays a role in infection. Genotypic analysis of families, as well as sequence analyses of KIR haplotypes indicates that the inhibitory KIR3DL1 and the activating KIR3DS1 segregate as alleles of one another. At least 20 additional alleles (all encoding inhibitory molecules) of the KIR3DL1/3DS1 gene have been identified. Some of the variants are not expressed, and others are expressed at either low or high levels, which could affect potential interactions between these KIR molecules and their infected cell targets. It is also possible that the various KIR3DL1/3DS1 molecules might differ in their binding affinity for their HLA ligand, which may in turn influence AIDS progression. We hypothesize that this variation may play a role in AIDS pathogenesis and we have begun to subtype the KIR3DL1/3DS1 alleles in our AIDS cohorts. We have recently completed high resolution typing of class II HLA -DRB1, -DRB3, -DRB4, -DRB5, -DQA1, and -DQB1 for 1,265 seroconverters including 882 Caucasians and 320 African Americans. Survival analyses showed that the HLA class II polymorphism is involved in AIDS pathogenesis though it appears to have a more modest effect on disease progression than HLA class I. Two DR, DQ haplotypes showed significant associations with a more rapid progression to AIDS. The DRB1*1301-DQA1*0103-DQB1*0603 haplotype was found to be weakly associated with a more rapid progression to AIDS 1987 (RH=1.42, p=0.0394) and death (RH=1.46, p=0.0344) in Caucasian seroconverters. The DRB1*0401-DQA1*0301-DQB1*0301 haplotype was also found to be associated with more rapid AIDS progression (AIDS 1987) in both the Caucasian (RH=1.60, p=0.01) and combined racial groups (RH=1.73, p=0.0016). Interestingly, the observed HLA class II associations with AIDS progression tend to be haplotype-dependent. For example, the three alleles, DRB1*0401, DQA1*0301 and DQB1*0301, present on the AIDS associated DR4 haplotype did not show any effect on AIDS progression when analyzed individually, but rather an effect was observed only when they are present on the same haplotype. In fact, DRB1*0401 linked with DQB1*0302 even showed a protective effect on AIDS progression. Haplotype-dependent disease associations indicate the occurrence of either an interaction between the gene loci on the haplotype, or the involvement of some other polymorphic gene that is in linkage disequilibrium with HLA class II genes. Detection of these AIDS related HLA class II factors requires high resolution typing and haplotype based analyses and these associations have not been reported previously. The effect of class II HLA (DR and DQ) polymorphism on hepatitis B virus (HBV) clearance/persistence was examined in 684 patients including 136 African Americans and 548 non-African Americans. Two alleles were found to be significantly associated with viral clearance, including DRB1*1302 (non-African Americans: OR=0.39, p=0.021) and DQB1*0602 (whole cohort: OR=0.66, p=0.017; African Americans: OR=0.32, p=0.005). Two DQB1 alleles, DQB1*0201 (whole cohort: OR=1.75, p=0.0001) and DQB1*0301 (whole cohort: OR=1.66, p=0.0006), were associated with viral persistence. The DRB1*1302 association with viral clearance confirms previously reported studies. DC-SIGN, a member of the CD209 gene family, is a dendritic cell-specific ICAM-3 adhesion receptor that enhances HIV-1 infection of T cells in trans, indicating that it may play an important role in HIV-1 transmission from the site of entry to peripheral lymph nodes. However, these results are derived from in vitro analyses and there is no in vivo data to support a role for DC-SIGN in HIV pathogenesis. Given the functional activity of this molecule, we proposed that polymorphisms in the coding sequence or promoter region of this gene might possibly influence outcomes to HIV-1 infection. Screening of about 100 individuals for nucleotide variation indicated conservation of DC-SIGN coding sequences, but six common and ten rare variants were identified in the region 2 Kb upstream of the ATG start codon. Six haplotypes with frequencies of >1% were identified. Preliminary analysis indicates that promoter region alleles containing the variant -336C are associated with susceptibility to parenteral (OR=2.11, p=0.0002), but not mucosal (OR=0.8, ns) infection. More than 1000 samples have now been typed and the potential effects of these variants in both infection and survival analyses are presently being analyzed. These epidemiologic data may provide support for DC-SIGN-mediated effects in HIV pathogenesis in vivo. We have also characterized the CD209 gene family in non-human primates and shown that recent evolutionary alterations have occurred in these genes across primate species. All primate species tested have orthologues of CD209. In contrast, CD209L is missing in old world monkeys (OWM), but is present in apes. A third novel family member, that we have named CD209L2, was cloned from rhesus monkey cDNA and was subsequently identified in OWM and apes except for humans. Despite high sequence similarity with both human and rhesus CD209, rhesus CD209L2 was substantially less effective in binding ICAM-3 and poorly transmitted HIV-1 and SIV to target cells relative to CD209. Our data suggest that the CD209 gene family has undergone recent evolutionary processes involving duplications and deletions, the latter of which may be tolerated due to potentially redundant functional activities of the molecules encoded by these genes.
Showing the most recent 10 out of 18 publications
