Abstract
The polymorphisms in Toll-like receptor (TLR) 7 and 9 genes are shown to influence HIV-1 infection. We studied HIV-1-infected Indian individuals for presence and association of TLR7 and 9 gene polymorphism with different disease outcomes. Genomic DNA from 65 HIV-infected individuals (35 long-term nonprogressors and 30 progressors) and 89 uninfected healthy donors was isolated, amplified, and sequenced for the reported polymorphisms in TLR7 [Gln11Leu (A/T); rs179008] and TLR9 (1635A/G; rs352140) genes. Of these, only the reported TLR9 single-nucleotide polymorphism [SNP; p = .017, odds ratio (OR) = 0.20] and its allele A frequency (p = .038, OR = 0.41) were found to be associated with slow disease progression. Of the new SNPs observed (three TLR7 and two TLR9), the TLR7 rs2074109 G allele showed less likely association with HIV-1 acquisition (p = .019, OR = 0.27). These findings indicate that TLR7 SNP (rs2074109) could be one of the factors for predisposition to HIV-1 and TLR9 1635A/G genotype and allele might have a role in HIV-1 disease progression in Indian population.
Introduction
Host genetic factors have been shown to influence on acquisition of Human Immunodeficiency virus (HIV) and disease progression after HIV infection. 1,2 The host restriction factors like APOBEC3G, TRIM5a, SAMHD1, BST/Tetherin, HERC5, and MHC molecules have shown to play a role in HIV infection by blocking or restricting retroviral replication at various stages of HIV life cycle. 2,3 While some host factors are associated with the resistance to HIV acquisition, some are associated with slower/faster disease progression. 3,4 Recently, the innate immune response has been shown to influence the HIV disease progression. 5
The Toll-like receptor (TLR) genes encode a family of transmembrane proteins essential for the innate immune recognition of pathogens. 6 With the help of an extracellular domain, TLRs identify conserved molecular motifs from a variety of organisms, including bacteria, fungi, parasites, and viruses. 5,7 Through the intracellular domains, TLRs interact with several adaptor proteins to activate transcription factors, leading to the production of inflammatory cytokines and the activation of the adaptive immunity. 6 Polymorphisms in TLR genes are known to be associated with increased susceptibility or protection against several infections. 8,9 Studies on single-nucleotide polymorphisms (SNPs) in TLR7 [Gln11Leu (A/T); rs179008] and TLR9 (1635A/G; rs352140) have shown to be associated with HIV disease progression, 1,10 –14 while some studies have showed that there is no such association. 15 These observations are made in Caucasian populations. Indians differ ethnically from the Caucasians and also have number of environmental influences on the innate immune responses. 16
The TLR polymorphism has not been studied in Indian population in context with the HIV disease progression. Hence, a study was planned to determine the known reported polymorphisms in TLR7 [Gln11Leu (A/T); rs179008] and TLR9 (1635A/G; rs352140) in HIV-1-infected individuals with progressive and nonprogressive HIV-1 infection and HIV-1-uninfected healthy individuals. We used the HIV-1-infected Long-Term Nonprogressors (LTNPs) as a model of absence of disease progression.
Thirty five LTNPs [antiretroviral therapy (ART)-naive, asymptomatic HIV-1-infected individuals with stable CD4 count above 500 cells/mm 3 for 7 or more years without the ART] were enrolled from the ongoing cohort of LTNPs at the National AIDS Research Institute (NARI), Pune, along with 30 progressors (ART-naive HIV-1-infected individuals with CD4 < 500 cells/mm 3 and not fitting in LTNP criteria) and 89 healthy controls from the outpatient clinics of the institute. Demographic details are given in Table 1. The study was approved by the Institutional Ethics Committee, and prior written informed consent was obtained from all the participants. Peripheral blood mononuclear cells (PBMCs) were separated from 4 mL whole blood by density gradient centrifugation, cryopreserved, and revived and used in the assay.
Demographic Details of Study Participants and TLR7 and TLR9 Primers Sequence and Polymerase Chain Reaction Condition Details
LTNP, long-term nonprogressors; PCR, polymerase chain reaction; SD, standard deviation; TLR, Toll-like receptor.
Around 1 × 10 6 PBMCs were used for deoxyribonucleic acid (DNA) extraction by using a QIAamp DNA kit (QIAGEN, Inc., Valencia, CA) as per the manufacturer's instructions and stored at −20°C till further use. The target DNA was amplified for SNP TLR7 (Gln11Leu) and TLR9 (1635A/G) using specific primers by polymerase chain reaction (PCR). Primers were designed by Oligo Analyzer software and cross verified using National Centre for Biotechnology Information (NCBI) blast for TLR7 and TLR9 specific regions as shown in Table 1. All amplification reactions were performed in ABI Veriti 96-well Thermal cycler (Applied Biosystems, Foster City, CA). PCR products in 1% agarose gel were visualized for expected bands of template size as mentioned in Table 1, along with molecular weight markers. Two microliters of PCR amplicons was purified by using High Prep PCR clean up kit (MagBio Genomics, Inc.) to remove excess dNTPs and primers. Purified DNA was then used for sequencing by using a Big Dye terminator sequencing kit (Applied Biosystems) as per the manufacturer's instructions. PCR was again carried out at the following conditions: 96°C for 1 min (96°C for 10 s, 55°C for 5 s, and 60°C for 4 min), and the cycle was repeated for 30 times. Sequencing was performed on an automated sequence analyzer of 3730xl DNA analyzer (Applied Biosystems). About 10% of samples were randomly picked for repeat assays, and the final concordance rate was 100%. Sequenced data were acquired and analyzed by using SeqScape_ Software v2.6 data tracking, management, and storage system. The UCSC database was used as a reference gene database to analyze any polymorphic change if present in the TLR7/TLR9 gene at various regions.
The mean age variable and standard deviation were used for analysis. Deviations from Hardy-Weinberg equilibrium (HWE) in healthy controls were assessed by Chi-square (χ
2
) goodness-of-fit test. AsTLR7 gene is an X-linked gene, the HWE of TLR7 (Gln11Lue) was done only in females from healthy control groups. Chi-square test was also used to compare genotype frequencies in HIV-infected progressors, LTNPs, and healthy controls. SNPStats (
We genotyped 65 HIV-1-infected individuals (35 LTNPs and 30 progressors) and 84 HIV-1-uninfected healthy subjects for TLR7 and 56 HIV-1-infected individuals (30 LTNPs and 26 progressors) and 89 HIV-1-uninfected subjects for TLR9 gene polymorphism. Our main aim was to screen known TLR7 (Gln11Leu) and TLR9 (1635A/G) genotype/allele frequency distribution in the study groups. While doing this, we observed three new SNPs in TLR7gene (rs2074109 A/G, rs179009 T/C, and rs-Not allotted T/C) and two new TLR9 SNPs (rs17846009 G/T and rs35342983 G/A) in coding region of TLR gene.
All observed genotype polymorphisms have been searched for “rs” (Reference Sequence no.) with the help of PredictSNP online tool (
Association of TLR7 Gene Polymorphism with Acquisition of HIV-1 Infection and Disease Progression
The genotype and allele frequencies of TLR7 in HIV-1-infected individuals and healthy controls, as well as in LTNPs and progressors, are shown in Table 2. The frequency of already reported TLR7 (Gln11Leu A/T) genotype was slightly higher in HIV-1-infected individuals than healthy controls (6.2% vs. 3.6%, p = .503, OR = 1.69) and allele (A/T) was slightly lower in HIV-1-infected individuals than healthy controls (4.6% vs. 7.7%, p = .279, OR = 0.58), but both did not differ significantly, showing no association with HIV-1 infection. Of the three new SNPs observed in TLR7 gene, only the frequency of TLR7 rs2074109 G allele was significantly higher in healthy controls compared to the HIV-1-infected individuals (10.7% vs. 3.1%, p = .019, OR = 0.27) and showed less likely association with HIV-1 acquisition. None of the TLR7 SNPs showed association with disease progression (Table 2). The genotype distributions of both reported and newTLR7SNPs (rs179008 A/T, rs2074109 A/G, and rs-Not allotted T/C) followed the HWE (p = .05, .05, and .156, respectively), except rs179009 T/C is p < .02 not consistent with HWE.
Frequency Distribution of TLR7 Genotypes/Alleles in Healthy Control and HIV-1-Infected Individuals, and Progressors and Nonprogressors
Significant p values (0.05) and related OR (95% CI) are shown in bold.
OR and 95% CIs were derived from logistic regression models comparing the homozygous wild-type genotype/allele (rs179008 A/A genotype and allele A, rs2074109 A/A genotype and allele A, rs179009 T/T genotype and allele T, and rs-Not allotted T/T genotype and allele T for TLR7) with other genotypes/alleles. Zero percent genotype frequencies in either of variables are avoided in the analysis.
CI, confidence interval; n, total number of subject participants; NA, not significant; OR, odds ratio.
Association of TLR9 Gene Polymorphism with Acquisition of HIV-1 Infection and Disease Progression
The genotype and allele frequencies of TLR9 in HIV-1-infected individuals and healthy controls, and in LTNPs and progressors, are shown in Table 3. The frequency of reported TLR9 rs352140 G/A genotype was slightly higher in HIV-1-infected individuals than healthy controls (44.6% vs. 38.2%, OR = 1.24), but did not differ significantly (p = .573), showing no association with HIV-1 acquisition. Whereas this genotype (rs352140 G/A) and allele A were seen more frequently in nonprogressors group (56.7% vs. 30.8%, p = .017, OR = 0.20) and (45% vs. 30.8%, p = .038, OR = 0.41), respectively (Table 3), and hence showed less likely associated with HIV-1 disease progression. Both newly observed TLR9 SNPs do not show any association either with HIV-1 acquisition or with the disease progression (Table 3). The genotype distributions of TLR9 SNPs rs352140 G/A, rs-17846009 G/T, and rs35342983 G/A followed the HWE (p = .05, .70 and .88, respectively) and were consistent in healthy controls.
Frequency Distribution of TLR9 Genotypes/Alleles in Healthy Control and HIV-1-Infected Individuals, and Progressors and Nonprogressors
Significant p values (0.05) and related OR (95% CI) are shown in bold.
OR and 95% CIs were derived from logistic regression models comparing the homozygous wild-type genotype/allele (rs17846009 G/G genotype and allele G, rs352140 G/G genotype and allele G, and rs35342983 G/G genotype and allele G, for TLR9) with other genotypes/alleles. Zero percent genotype frequencies in either of variables are avoided in the analysis.
n, total number of subject participants.
Gene-Gene Interaction (Haplotype Analysis)
The haplotype frequencies of TLR7 (rs179008 A/T, rs2074109 A/G, rs179009 T/C, and rs-Not allotted T/C) and TLR9 (rs17846009 G/T, rs352140 G/A, and rs35342983 G/A) genes in study groups were analyzed, respectively. The LD values and comparison of LD values (Dij) between the study groups were calculated using SNPStats software, and did not show significant differences for all the genes (“D” values: Progressors vs. LTNPs: 0.7314, Progressors vs. Healthy: 0.3157, and LTNPs vs. Healthy: 0.3099). Haplotypes of TLR7 (data not shown) and 9 genes did not show association with HIV acquisition in any study group, whereas GAG haplotype of TLR9 gene showed less likely association with rapid disease progression (43.33% vs. 26.92%, p = .019, OR = 0.33) (Table 4).
Frequency Distribution of Haplotypes of TLR9 (rs17846009 G/T, rs352140 G/A, and rs35342983 G/A) in Healthy Controls and HIV-1-Infected Long-Term Nonprogressors and Progressors
Significant p values (.05) and related OR (95% CI) have been shown in bold.
Age-adjusted ORs and 95% CIs were derived from logistic regression models comparing the genotype CC+CC with other genotypes.
Frequencies fewer in the haplotypes are excluded from the data.
(%), frequency of subjects.
We reported presence of the already reported polymorphisms in TLR7 and TLR9 genes in Indian populations also. The known polymorphism of TLR7 (rs179008 A/T) was not found to be associated with either HIV-1 acquisition or disease progression, contradicting the previous observations by Oh et al. 11 It could be due to the differences in the ethnic background of the populations studied in these and our study. Also, we could carry out the polymorphism studies in well-characterized populations with nonprogressive disease such as LTNPs and progressive diseases (progressors), whereas in the previously reported studies, the association between CD4 count and viral load has been studied.
The known polymorphism in TLR9 (rs352140 G/A), both the genotype and the allele, was not associated with HIV-1 acquisition, but it was found to be less likely associated with rapid disease progression. Contradicting to our study, Bochud et al. 10 reported that TLR9 (rs352140) is more in HIV-1 rapid disease progression in Swiss HIV cohort study by correlating percentile decline rate of CD4 count, and Freguja et al. 14 noted that TLR9 1635A/G genotype was associated with rapid disease progression in prenatally HIV-1-infected children. Our results are consistent with the observations reported by Soriano-Sarabia et al., 1 who also found that the 1635G allele is associated with slow progression in a seroprevalent, dynamic therapeutic cohort, and Pine et al. 12 reported that the 1635G allele was associated with lower viral load set point and slower disease progression. We first time reported TLR9 1635A/G genotype and allele is associated with slow disease progression in a well-characterized population with nonprogressive HIV-1 infection.
In addition, we reported for the first time that new TLR7 rs2074109 G allele (minor allele frequency) was less likely associated in HIV-1 acquisition. Recently, the same allele was reported in HBV infection without any significance in Chinese population.
17
The changes in TLR7 SNP (rs2074109) nucleotide from A to G do not cause alteration in amino acids. However, on the basis of SNP function prediction analysis tool (
In the haplotype analysis, any of TLR7 gene haplotypes did not show association with HIV-1 acquisition or disease progression. However, we reported for the first time that TLR9 haplotype GAG was less likely associated with rapid disease progression.
Recently, TLR7 18 and TLR9 19 agonist have shown a role in activation of HIV reservoir and innate immunity; hence, polymorphism in these genes may influence the natural control/restriction of HIV-1 infection. Genetic variations in TLRs may affect host-virus interactions and impact HIV-1 disease progression. 14
Overall, our findings indicate the probable role of TLR7 (rs2074109) G genotype in predisposition to HIV-1 and of TLR9 1635A/G genotype and allele in HIV-1 disease progression in Indian population. With respect to the limited sample size, this finding requires verification by larger sample size studies in diverse ethnicities of Indian populations.
Footnotes
Acknowledgments
The authors would like to thank all study participants and the staff of NARI clinics, Immunology and Molecular Biology Laboratory of NARI, for their assistance. This work was supported through institutional funding. N.S. gratefully acknowledges the Indian Council of Medical Research (ICMR), New Delhi, for providing senior research fellowship through grant no. ICMR/SRF/80/933/15/ECD-1.
Author Disclosure Statement
The authors declare no conflict of interests.
