NEW YORK (GenomeWeb) – Common variants in the HLA and CCR gene regions explain the majority of patients' genetic contribution to variability in HIV-1 viral load, according to an international team of researchers.
After primary infection with HIV, patients' viral loads generally stabilize, but this viral load set point varies from patient to patient, and while viral genetics and environmental factors influence this level, so does host genetics.
Researchers led by Jacques Fellay of the École Polytechnique Federale de Lausanne conducted a genome-wide association study of more than 6,000 people with HIV to identify which host genes affect viral load. As they reported today in the Proceedings of the National Academy of Sciences, this implicated polymorphisms in the HLA class I region and the chemokine receptor gene cluster. The researchers estimated that common variations — mostly in these regions — explain about a quarter of the variability seen in viral load.
"We demonstrate that multiple independent signals exist at two genomic loci and implicate novel, potentially causal variants within these regions," Fellay and his colleagues wrote in their paper.
For their analyses, the researchers drew upon genotyping data on 7,468 individuals of European descent from eight independent genome-wide association studies. Viral load set point data was available for 6,315 people from these cohorts, and Fellay and his colleagues tested some 8 million common variants for a link with viral load set point in that group.
They uncovered two significant associations, one on chromosome 6 and one on chromosome 3.
The MHC region on chromosome 6, Fellay and his colleagues noted, harbored the strongest signal of association with viral load levels, and they mapped multiple independent common variants of large effect to this region.
The signal, they noted, centered on the class I HLA gene HLA-B, which has been implicated in viral load levels.
By imputing classical class I and class II HLA alleles, variable amino acid positions in HLA proteins, and additional SNVs, the researchers found that a number of classical HLA-A, HLA-B, and HLA-C alleles were associated with viral load set point — some alleles, they noted, strongly decreased set point, while others strongly increased it.
Further, through testing variable amino acid positions in HLA class I proteins, the researchers uncovered associations between HLA-B positions 97, 67, and 45, and positions 77 and 95 in HLA-A and viral load set point.
These five spots fall within the peptide binding grooves of the two proteins, and changes there, the researchers added, had varying effects on viral load, with some tweaks increasing it and others decreasing it.
"The location of these amino acids on the peptide-binding groove of the respective protein supports the hypothesis that the presentation of specific epitopes, directly dependent on the shape of the HLA peptide binding groove, is critical in determining the efficiency of the cytotoxic T-cell response," Fellay and his colleagues wrote.
All together, the researchers estimated that these five positions explain 12.3 percent of the variation in viral set load and account for most of the association signal at this locus.
The chromosome 3 signal, meanwhile, was centered over the CCR gene cluster, the researchers reported, noting that variations in CCR are known to influence HIV pathogenesis. They added that the strongest known causal variant in this region is CCR5Δ32, though the CCR5 promoter haplotype P1 has also been linked to AIDS progression.
In a conditional analysis of the 5,559 individuals for whom CCR5Δ32 genotype was known and whose HAP-P1 status could be determined, the researchers noted strong associations between CCR5Δ32 and reduced viral load set point, and between CCR5 Hap-P1 haplotype and increased viral load.
After conditioning on CCR5Δ32, the researchers uncovered 122 SNPs significantly associated with viral load. These SNPs, the researchers added, were also significant after conditioning for Hap-P1, with the strongest signal at rs1015164. After conditioning for both CCR5Δ32 and Hap-P1, the rs1015164 SNP remained significant.
These three variants explain some 2.2 percent of the variance in viral load set point, the researchers estimated.
Meantime, the common variants in the HLA and CCR5 regions explain some 14.5 percent of the variability in viral set point load, Fellay and his colleagues reported.
However, through a genome-wide complex trait analysis, they found that genome-wide variation explains about a quarter of the narrow-sense heritability. After controlling for genome-wide significant signals, they found that the remainder of the genome only explains about 5 percent of the variation.
"This limited residual heritability underscores the primary role of common variants of large effects in the MHC and CCR5 in HIV-1 control," the researchers wrote.
This, they added, "suggests that the identified common variants of large effect explain the majority of the host genetic component of HIV-1" viral load set point.