The demographic characteristics of the HIV-infected subjects are shown in . The cohort included 201 participants, of whom 196 were males (98%), 193 reported a repeated history of men having sex with men as their HIV-1 risk factor (96%), and the median age was 33 (interquartile range 9). Data was available to assess peak viral load, setpoint viral load and CD4+ T-cell depletion slopes in 149, 122 and 77 subjects, respectively (). The demographics of subjects in the comparison groups in each arm of the study (high versus low/moderate peak and setpoint viral load and rapid versus slow/moderate CD4+ T-cell decline) were not different, except for age in the peak viral load comparison (P=0.03).
Demographic Characteristics of HIV-1 Infected Subjects
The characteristics of SNPs in the four TLRs (2, 3, 4, and 9) detected in the study population are shown in . All tested SNPs except one (TLR4
−2604 A/G) were at Hardy-Weinberg equilibrium. The primers used to genotype TLR4
−2604 A/G were used successfully to genotype other cohorts during the same study period, suggesting that the absence of equilibrium was not due to a technical issue, but rather to a random effect, and/or the small sample size. One SNP was present at a frequency of less than 5%, and therefore was not included in the analyses (TLR2
2258 G/A, rs5743708, minor allele frequency 4%). As expected from previous observations, two SNP pairs were in strong LD (TLR4
1063 A/G and 1363 C/T, R2
=1; and TLR9
+1174 G/A and 1635A/G, R2
TLR Single Nucleotide Polymorphisms in HIV-1 Infected Subjects
The risk of having a high viral load by TLR polymorphisms was assessed in logistic regression models, after adjustment for age and sex (). We observed that 2 TLR4 SNPs in strong LD (1063 A/G [D299G] and 1363 C/T [T399I]) were associated with the high peak viral load phenotype; 74%, 26% and 0% of subjects with high peak viral load had 0, 1 and 2 copies of the minor allele, respectively, versus 95%, 4% and 1% of low/moderate peak viral load subjects (OR=4.70, 95% confidence interval [CI] 1.65–13.36, adjusted P=0.004 under the additive model; OR=6.65, 95% CI 2.16–20.46, adjusted P<0.001, under the dominant model, and ). This difference was still significant after correction for multiple testing (P=0.002, dominant model). We also observed that two TLR3 SNP (−8921 A/T and 1234 C/T) tended to be associated with peak viral load, but these associations were not significant after correction for multiple testing (P=0.8 and P=1.0, respectively).
Frequencies of TLR Alleles in Subjects with High versus Low/Moderate Viral Load/Setpoint and Rapid versus Slow/Moderate Progression by CD4+ T-Cell Decline
Comparison of TLR Genotype Frequencies in Subjects with High versus Low/Moderate Viral Load/Setpoint and Rapid versus Slow/Moderate Progression by CD4+ T-Cell Decline
The SNPs associated with high peak viral load were not associated with high viral load setpoint or rapid HIV-1 disease progression as measured by CD4 slope. However, we observed that two SNPs in TLR9 were associated with these endpoints ( and ). A TLR9 SNP (1635 A/G) was protective against high viral load setpoint (OR=0.29, 95% CI 0.13–0.65, P=0.003, dominant model) and rapid progression (OR=0.45, 95% CI 0.24–0.97, P=0.04, additive model); this difference remained significant after correction for multiple testing for the viral load setpoint (P=0.04), but not the disease progression phenotype (P=0.4). Another TLR9 SNP (−1237 T/C) tended to be associated with high viral load setpoint (OR=2.12, 95% CI 0.99–4.54, P=0.05) and rapid progression (OR=3.21, 95% CI 1.16–8.90, P=0.03), but neither of these associations was significant after correction for multiple testing (P=0.8 and P=0.4, respectively). We also observed that a TLR2 SNP (−16934 T/A) was associated with protection against rapid progression (OR= 0.43, 9%% CI 0.20–0.92, P=0.03, under the additive model), but again, the association was not significant after adjustment for multiple testing (P=0.5).
In the analyses by haplotype, we observed that TLR4 haplotype 7, present in TLR4 1063G and 1363T carriers, was associated with an increased risk of having a high viral load compared to the reference haplotype (OR=5.71, 95% CI 1.34–24.45, adjusted P=0.02, ), but the association was lost after correction for multiple testing (P=0.08). We also found that TLR9 haplotype 3, present in TLR9 −1237C and 1635G carriers, was associated with an increased risk of having a high viral load setpoint (OR=2.54, 95%CI 1.12–5.78) as well as rapid progression measured by CD4+ T-cell decline (OR=3.56, 95% CI 1.21–10.49, adjusted P=0.02); however, this difference was not significant after correction for multiple testing (P=0.1 and P=0.08, respectively).
Association of TLR Haplotypes with Viral Load/Setpoint and Rapid Progression by CD4+ T-Cell Decline in HIV+ Subjects.