Secondary Logo

Journal Logo

Translational Research

Brief Report: Toll-like Receptor 9-1635A/G Polymorphism Is Associated With HIV-1 Rebound After Four Weeks of Interruption of Antiretroviral Therapy

Vallejo, Alejandro PhDa; Molina-Pinelo, Sonia PhDb,c; de Felipe, Beatriz PhDb,d; Abad-Fernández, María PhDe; González-Escribano, María Francisca PhDb,f; Leal, Manuel MD, PhDb,g; Soriano-Sarabia, Natalia PhDh

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: October 1, 2020 - Volume 85 - Issue 2 - p 252-256
doi: 10.1097/QAI.0000000000002437

Abstract

INTRODUCTION

Over a decade ago, antiretroviral treatment (ART) interruptions were assayed in an attempt to reduce the negative effects associated with the prolonged exposure to ART and also to induce stronger cellular and humoral immune responses.1,2 Unfortunately, after interruption of therapy, plasma HIV-1 RNA often reached levels similar to those found during the pretreatment period.3,4 This practice became inadvisable because of the evidence demonstrating an association of the interruption with an increased risk of clinical progression.5,6 However, treatment discontinuation has regained attention because of current approaches aimed at reducing the persistent viral reservoir as part of the HIV-1 cure agenda.7

After ART discontinuation, plasma HIV-1 load rebound and consequent set-point viremia levels are shown to be highly variable among HIV-infected individuals.6,9,10 Mechanisms involved in heterogenic viral kinetics include, but may not be limited to, intact HIV-1 proviruses in circulating CD4 T cells,11 potent HIV-1-specific or innate immune responses,12–14 greater thymic volume and function,9 CCR5 coreceptor expression,15 or established HIV-1 reservoir size.16

Toll-like receptors (TLRs) are critical proteins involved in the establishment of effective immune responses.18,19 TLR2 and TLR7 single nucleotide polymorphisms (SNPs) have been associated with HIV-1 set-point, explaining up to 6% of the variation.13 We and others also reported the association of TLR2 (1892A/C and 2258G/A) and TLR4 (899A/G and 1196C/T) with HIV-1 RNA load, CD4 count, and disease progression.20–22 In addition, the combination of TLR7 (AA) and TLR9 (GG) genotypes might be linked to higher CD4 counts during the viremic period.23 Finally, we reported that the AA genotype of the TLR9-1635A/G SNP was associated with higher HIV-1 RNA load in ART-naïve chronically HIV-1-infected individuals.20

Because the knowledge about the influence of SNPs in TLRs on the viral rebound and/or the subsequent set-point achieved after ART interruption is limited, we hypothesized that the plasma HIV-1 RNA rebound variability after ART interruption may be influenced by the TLR9-1635A/G SNP. We also investigated the potential implication of other SNPs in TLR2 and TLR4, the CCR5-Δ32 deletion, and the presence of HLA-A and HLA-B alleles on the viral rebound and set-point on ART interruption.

MATERIALS AND METHODS

This was a retrospective longitudinal study performed from July 2003 to March 2008 aimed to analyze the potential association between common genetic variants in TLR2, TLR4, and TLR9 and the HIV-1 RNA load rebound and set-point on ART interruption, and the establishment of a CD4 count set-point. This study included 57 White HIV-1-infected individuals who had undetectable plasma viremia (<50 copies/mL) for at least 12 months before ART interruption and CD4 count above 500 cells/mm3. Clinical and immunovirological determinations were recorded at baseline and monthly subsequently. HIV-1 RNA load rebound was considered at week 4 on interruption.9,15 All individuals provided written informed consent, and the Ethical Committee of the Virgen del Rocio Hospital approved the study which complied with the stipulations of the Declaration of Helsinki.

The total CD4 count was obtained from the individual's routine laboratory follow-up visits provided by the Hospital. Plasma HIV-1 RNA was measured by quantitative real-time polymerase chain reaction (RT-PCR) (HIV Monitor Test Kit; Roche Molecular System, Basel, Switzerland), according to the manufacturer's instructions (with detection limit of 50 HIV-1 RNA copies/mL). HCV infection was diagnosed by the detection of anti-HCV antibodies by enzyme-linked immunosorbent assay (HCV-specific enzyme-linked immunosorbent assay; Siemens Healthcare Diagnosis, Malvern, PA), and the presence of plasma HCV RNA by RT-PCR (COBAS Amplicor; Roche Diagnostics, Barcelona, Spain).

Genomic DNA was extracted from PBMC using the QIAmp DNA Mini Kit (Qiagen, Barcelona, Spain) and stored at −20°C. The purity of the DNA was determined using NanoDrop spectrophotometer (Thermo Fisher Scientific, Wilmington, DE). TLR9-1635A/G, TLR2-1892A/C and 2258G/A, and TLR4-899A/G and 1196C/T SNPs were determined by RT-PCR and melting curve technology, as previously reported.19,21 In brief, PCR amplification was performed with initial denaturation of 95°C/5 minutes, 45 cycles of denaturation (95°C/30 seconds), annealing (55°C/30 seconds), and elongation (72°C/30 seconds) using the LightCycler 480 System (Roche Diagnostics, Mannheim, Germany). Melting curves analyses were performed using 0.2 µM of each detection probe. The LightCycler 480 control kit to detect the SNPs was run in parallel. After initial denaturation at 95°C/2 minutes at the ramp rate of 4.4°C/s, the temperature was dropped to 45°C at the ramp rate of 1°C/s and finally led to 80°C with 1 acquisition per degree centigrade. CCR5 alleles were amplified using primers flanking the 32-bp deletion in the CCR5 gene (Δ32F, ggaatcatctttaccagatctcaaaaa-3′ and Δ32R catgatggtgaagataagcctcaca-3′). PCR consisted in 35 cycles, including denaturation step at 94°C/20 seconds, hybridization step at 56°C/30 seconds, and extension step at 72°C/30 seconds. The fragment products were run and analyzed in 2% agarose gel. HLA-A and HLA-B low resolution typing were performed by PCR-SSOP Luminex method using LABType SSO using group-specific primers (One Lambda, Inc., Canoga Park, CA). The biotinylated PCR products were denatured and hybridized with specific probes bound to colored-coded microspheres. Reactions were analyzed through LABScanTM 100 flow analyzer and typed with HLA Fusion 2.0 software (One Lambda, Inc.).

Statistical Analysis

Continuous variables are expressed as median and interquartile range (IQ25–75) and categorical as number and percentages. The Mann–Whitney U test was used to analyze differences between continuous variables with 2 different levels. The 2-tailed Spearman correlation test was used to associate quantitative variables, and the χ2 test for comparison of qualitative variables. Models for the likelihood were tested to analyze differences according to TLR's SNPs and recessive model for the wild type allele was used for the comparison between the different genotypes in both cohorts.20,22 Variance inflation factor was used to detect multicollinearity. Variables with P < 0.1 in the univariate analysis were introduced in the forward stepwise multivariate linear regression analysis to assess independent factors associated with HIV rebound and set-point and the CD4 T-count set-point. All statistical analyses were conducted using SPSS v.24, and graphs were generated using GraphPad Prism v.8.

RESULTS

The characteristics of the 57 HIV-1-infected individuals are summarized in Supplemental Digital Content 1, https://links.lww.com/QAI/B506. TLR9-1635A/G-AA genotype was present in 13 individuals (23%), AG genotype in 35 (61%), and GG genotype in 9 (16%) (AG + GG genotypes in 44 individuals, 79%), as shown in Supplemental Digital Content 2, https://links.lww.com/QAI/B506, showing a similar distribution to that found in our chronically naïve HIV-1-infected population, as previously reported.20 All frequencies were in accordance with the Hardy–Weinberg equilibrium (data not shown). Frequencies of SNPs in TLR2-2258G/A were lower in this study than those previously reported for our general HIV-1-infected population.12 Three individuals (6%) harbored the TLR2-1892A/C SNP, whereas only 1 individual (2%) harbored the TLR2-2258G/A SNP. Finally, 13 individuals (25%) harbored the TLR4-896 A/G SNP, whereas 14 individuals (26%) harbored the TLR4-1296C/T SNP. In line with previous studies,24 4 individuals (7%) were heterozygous for the CCR5-Δ32 deletion, whereas homozygosity was not detected (see Supplemental Digital Content 2, https://links.lww.com/QAI/B506). HLA-A and HLA-B allele frequencies in our study population are shown in Supplemental Digital Content 3A and B, https://links.lww.com/QAI/B506.

HIV-1 RNA rebound positively correlated with the pre-ART viral load (R = 0.34, P = 0.025, Fig. 1A). Consistent with a previous report,5 women had lower HIV-1 RNA rebound compared with men [2.6 (1.7–3.8) and 4.1 (2.5–5.2) log10 HIV-1 RNA copies/mL, respectively, Fig. 1B]. Interestingly, individuals with the TLR9-1635AA genotype had higher levels of HIV-1 RNA rebound compared with those with AG + GG genotype (P = 0.007, Fig. 1C). Finally, individuals harboring the HLA-A26 allele had lower HIV-1 RNA rebound (P = 0.026, Fig. 1D). The analysis of HIV-1 RNA levels during the 72 weeks of follow-up showed that women consistently had a lower viral load compared with men (P < 0.05) in all time-points, Supplemental Digital Content 7, see https://links.lww.com/QAI/B506. However, TLR9-1635A/G SNP and HLA-A26 allele were not associated with HIV-1 RNA levels during the follow-up (see Supplemental Digital Content 7B and C, https://links.lww.com/QAI/B506, respectively).

FIGURE 1.
FIGURE 1.:
Variables associated with the HIV-1 RNA rebound after 4 weeks of ART interruption. A, Pre-ART HIV-1 plasma load positively correlated with postinterruption HIV-1 load. Pre-ART viral load was available in 43 individuals. Women are represented with open triangles and men with filled circles. B, Women had lower viral rebound compared with men. C, Individuals harboring the TLR9 1635AA genotype had lower HIV-1 rebound after ART interruption. D, Individuals with the HLA-A26 allele showed lower viral rebound compared with individuals who did not harbor the allele.

To determine factors involved in the HIV-1 RNA rebound, a stepwise linear regression analysis was performed. Because HIV-1 RNA load was lower in women both at rebound and during the 72 weeks of follow-up (see Fig. 1B and Supplemental Digital Content 7A, https://links.lww.com/QAI/B506), the statistical analysis was restricted to male individuals. Men and women had comparable baseline characteristics (P > 0.05, not shown). HLA-A and B allelic distribution in men is shown in Supplemental Digital Content 5, see https://links.lww.com/QAI/B506. We initially assessed that variables were not collinear (variance inflation factor >1 in all cases). Variables associated with HIV-1 RNA rebound included TLR9-1635AA, HLA-A26, and HLA-B18 and were included in a stepwise multivariate analysis. Only the TLR9-1635AA genotype was independently associated with higher HIV-1 RNA rebound (P = 0.004, Table 1).

TABLE 1. - Univariate and Multivariate Linear Regression Analyses to Assess Factors Independently Associated With the HIV-1 RNA Load Rebound After ART Interruption
Univariate Multivariate
P Regression Coefficient (95% Confidence Interval) P Regression Coefficient (95% Confidence Interval)
TLR9 1635AA 0.004 1.49 (0.51 to 2.47) 0.004 1.49 (0.51−2.47)
HLA-A26 0.056 −1.27 (−2.58 to 0.03) HLA-A26, P = 0.083
HLA-B18 0.057 1.37 (−0.04 to 2.78) HLA-B18, P = 0.571
Ns, not significant.

Both HIV-1 RNA load and CD4 count set-points were estimated as the median value of all available quantifications from week 8 to week 72 of follow-up after treatment interruption [4.4 (3.9–4.9) log10 copies/mL and 577 (425–650) cells/mm3, respectively]. As expected, the HIV-1 RNA load set-point after ART interruption positively correlated with pre-ART HIV-1 RNA load (R = 0.69, P < 0.001, 8A, https://links.lww.com/QAI/B506) and with the HIV-1 RNA rebound (R = 0.52, P < 0.001, 8B, https://links.lww.com/QAI/B506). In addition, CD4 count set-point correlated positively with nadir CD4 count (R = 0.52, P < 0.001, see Supplemental Digital Content 8C, https://links.lww.com/QAI/B506). Although pre-ART HIV-1 RNA load was similar according to sex, men had higher HIV-1 load set-point (see Supplemental Digital Content 9A, https://links.lww.com/QAI/B506). However, both CD4 nadir and CD4 count set-points were similar between men and women (see Supplemental Digital Content 9B, https://links.lww.com/QAI/B506).

Stepwise linear regression analysis showed that both pre-ART HIV-1 RNA load and time with undetectable HIV-1 RNA load were independently associated with the establishment of the HIV-1 RNA load set-point (P = 0.001 and P = 0.002, respectively, see Supplemental Digital Content 6, https://links.lww.com/QAI/B506). In addition, HLA-01 allele and nadir CD4 count was independently associated with the establishment of the CD4 count set-point (P = 0.027 and P = 0.001, respectively, see Supplemental Digital Content 6, https://links.lww.com/QAI/B506.

DISCUSSION

Our results show an association between the TLR9-1635A/G SNP and the HIV-1 RNA load rebound after interruption of ART. Specifically, patients with the TLR9-1635AA genotype showed a greater risk of higher viral load at rebound. We previously reported that chronically HIV-1-infected individuals, naïve for ART who harbored the TLR9-1635AA genotype were more likely to have both lower CD4 count and higher HIV-1 RNA load, and a higher risk of clinical disease progression.20 We hypothesized that in the setting of ART interruption, the TLR9-1635A/G SNP could be involved in both the magnitude of the viral load rebound and/or the subsequent establishment of the HIV-1 RNA load set-point.

TLR9 contributes to HIV-induced immune activation, and TLR9 SNPs are likely to contribute to the individual variability in the clinical course of HIV disease.19 Despite the limited size, this valuable interruption cohort allowed us to find that the TLR9 SNP was associated with the viral rebound on treatment discontinuation. In this study, individuals with the TLR9-1635AA genotype showed higher HIV-1 RNA rebound after ART interruption, which was in turn associated with the establishment of the viral set-point. These results may help understand the individual variation response to current TLR agonist's studies aimed to reactivate the latent virus as part of the effort to cure HIV-1 infection.25–28 TLR9 expression is directly correlated with the rate of immune activation, but in chronic infections shows reduced upregulation and diminished responsiveness after CpG-DNA stimulation.29 In this study, plasma levels of β-2 microglobulin and TNF-α were not associated with HIV-1 plasma load rebound or HIV-1 load and CD4 count set-points (data not shown). However, we cannot exclude that the higher rebound in patients with the AA genotype was the consequence of higher immune activation.

After the initial viral rebound, both plasma HIV-1 RNA levels and CD4 count reached set-points that were maintained through the follow-up. Interestingly, pre-ART viral load and time with undetectable plasma HIV-1 RNA load were associated with the viral load set-point. Our results also show a marginal effect of the presence of the HLA-B27/B57 protective alleles.30,31 However, we did not find a protective role of either of them separately likely because of the limited sample size. Although the presence CCR5-Δ32 deletion in heterozygosity has been associated with lower viral loads,32,33 in treatment interruption, it may not be involved in the viral or the CD4 count set-point, in line with a previous report.34 Finally, our results showed that the presence of the HLA-01 allele was associated with a higher CD4 count set-point, confirming a previous report that identified the presence of this class I allele as protective.35

Our work highlights the importance of analyzing the presence of common SNPs on TLR9 and potentially on others, such as TLR7, when performing studies aimed to interrupt treatment in the clinical setting of HIV-1 eradication. These common SNPs may help explain variations and may constitute a tool to predict time to rebound after treatment discontinuation.

ACKNOWLEDGMENTS

Authors thank the participants who made the study possible. Statistical analysis was supervised by Eloisa Rubio-Perez, from the Andalusian Public Foundation for the Management of Health Research in Seville (FISEVI), Seville, Spain. N.S.-S. designed the study; A.V., S.M.-P, M.L., and N.S.-S. analyzed results and wrote the article; B.d.F., M.A.-F., and M.F.G.-E. performed experiments and wrote the article.

REFERENCES

1. Heath KV, Hogg RS, Chan KJ, et al. Lipodystrophy-associated morphological, cholesterol and triglyceride abnormalities in a population-based HIV/AIDS treatment database. AIDS. 2001;15:231–239.
2. Pomerantz RJ, Horn DL. Twenty years of therapy for HIV-1 infection. Nat Med. 2003;9:867–873.
3. Hatano H, Vogel S, Yoder C, et al. Pre-HAART HIV burden approximates post-HAART viral levels following interruption of therapy in patients with sustained viral suppression. AIDS. 2000;14:1357–1363.
4. Lewin SR, Murray JM, Solomon A, et al. Virologic determinants of success after structured treatment interruptions of antiretrovirals in acute HIV-1 infection. J Acquir Immune Defic Syndr. 2008;47:140–147.
5. El-Sadr WM, Lundgren J, Neaton JD, et al. CD4+ count-guided interruption of antiretroviral treatment. N Engl J Med. 2006;355:2283–2296.
6. Van Lunzen J, Hoffmann C. Virological rebound and its consequences during treatment interruption. Curr Opin HIV AIDS. 2007;2:1–5.
7. Kim Y, Anderson JL, Lewin SR. Getting the “kill” into “shock and kill”: strategies to eliminate latent HIV. Cell Host Microbe. 2018;23:14–26.
8. Ndung'u T, McCune JM, Deeks SG. Why and where an HIV cure is needed and how it might be achieved. Nature. 2019;576:397–405.
9. Vallejo A, Valladares A, De Felipe B, et al. High thymic volume is associated with viral replication and immunologic impairment only early after HAART interruption in chronic HIV infection. Viral Immunol. 2005;18:740–746.
10. Tarwater PM, Parish M, Gallant JE. Prolonged treatment interruption after immunologic response to highly active antiretroviral therapy. Clin Infect Dis. 2003;37:1541–1548.
11. Lu CL, Pai JA, Nogueira L, et al. Relationship between intact HIV-1 proviruses in circulating CD4(+) T cells and rebound viruses emerging during treatment interruption. Proc Natl Acad Sci U S A. 2018;115:E11341–E11348.
12. Ortiz GM, Nixon DF, Trkola A, et al. HIV-1-specific immune responses in subjects who temporarily contain virus replication after discontinuation of highly active antiretroviral therapy. J Clin Invest. 1999;104:R13–R18.
13. Mackelprang RD, Carrington M, Thomas KK, et al. Host genetic and viral determinants of HIV-1 RNA set point among HIV-1 seroconverters from sub-Saharan Africa. J Virol. 2015;89:2104–2111.
14. Hurst J, Hoffmann M, Pace M, et al. Immunological biomarkers predict HIV-1 viral rebound after treatment interruption. Nat Comm. 2015;6:8495.
15. Soriano-Sarabia N, Vallejo A, Fernández G, et al. Control of HIV-1 RNA load after HAART interruption: relationship with CCR5 co-receptor density and proviral DNA load in HIV-infected patients. J Clin Virol. 2007;40:64–67.
16. Sáez-Cirión A, Bacchus C, Hocqueloux L, et al. Post-treatment HIV-1 controllers with a long-term virological remission after the interruption of early initiated antiretroviral therapy ANRS VISCONTI Study. PLoS Pathog. 2013;9:e1003211.
17. Hill AL, Rosenbloom DI, Goldstein E, et al. Real-time predictions of reservoir size and rebound time during antiretroviral therapy interruption trials for HIV. PLoS Pathog. 2016;12:e1005535.
18. Gordon S. Pattern recognition receptors: doubling up for the innate immune response. Cell. 2002;111:927–930.
19. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol. 2001;2:675–680.
20. Soriano-Sarabia N, Vallejo A, Ramírez-Lorca R, et al. Influence of the toll-like receptor 9 1635A/G polymorphism on the CD4 count, HIV viral load, and clinical progression. J Acquir Immune Defic Syndr. 2008;49:128–135.
21. Pine SO, McElrath MJ, Bochud PY. Polymorphisms in toll-like receptor 4 and toll-like receptor 9 influence viral load in a seroincident cohort of HIV-1-infected individuals. AIDS. 2009;23:2387–2395.
22. Pulido I, Leal M, Genebat M, et al. The TLR4 ASP299GLY polymorphism is a risk factor for active tuberculosis in Caucasian HIV-infected patients. Curr HIV Res. 2010;8:253–258.
23. Said EA, Al-Yafei F, Zadjali F, et al. Association of single-nucleotide polymorphisms in TLR7 (Gln11Leu) and TLR9 (1635A/G) with a higher CD4 T cell count during HIV infection. Immunol Lett. 2014;160:58–64.
24. Novembre J, Galvani AP, Slatkin M. The geographic spread of the CCR5 Delta32 HIV-resistance allele. PLoS Biol. 2005;3:e339.
25. Joshi A, Punke EB, Mehmetoglu-Gurbuz T, et al. TLR9 polymorphism correlates with immune activation, CD4 decline and plasma IP10 levels in HIV patients. BMC Infect Dis. 2019;19:56.
26. Offersen R, Nissen SK, Rasmussen TA, et al. A novel toll-like receptor 9 agonist, MGN1703, enhances HIV-1 transcription and NK cell-mediated inhibition of HIV-1-infected autologous CD4+ T cells. J Virol. 2016;90:4441–4453.
27. Vibholm L, Schleimann MH, Højen JF, et al. Short-course toll-like receptor 9 agonist treatment impacts innate immunity and plasma viremia in individuals with human immunodeficiency virus infection. Clin Infect Dis. 2017;64:1686–1695.
28. Vibholm LK, Konrad CV, Schleimann MH, et al. Effects of 24 week toll-like receptor 9 agonist treatment in HIV-1+ individuals: a single-arm, phase 1B/2A trial. AIDS. 2019;33:1315–1325.
29. Scheller C, Ullrich A, Lamla S, et al. CpG oligodeoxynucleotides activate HIV replication in latently infected human T cells. J Biol Chem. 2004;279:21897–21902.
30. Carrington M, O'Brien SJ. The influence of HLA genotype on AIDS. Ann Rev Med. 2003;54:535–551.
31. Altfeld M, Addo MM, Rosenberg ES, et al. Influence of HLA-B57 on clinical presentation and viral control during acute HIV-1 infection. AIDS. 2003;17:2581–2591.
32. Michael NL, Moore JP. Viral phenotype and CCR5 genotype. Nat Med. 1999;5:1330.
33. Samson M, Libert F, Doranz BJ, et al. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature. 1996;382:722–725.
34. Ruiz L, Martinez-Picado J, Romeu J, et al. Structured treatment interruption in chronically HIV-1 infected patients after long-term viral suppression. AIDS. 2000;14:397–403.
35. Peterson TA, Kimani J, Wachihi C, et al. HLA class I associations with rates of HIV-1 seroconversion and disease progression in the Pumwani sex worker cohort. Tissue Antigens. 2013;81:93–107.
Keywords:

TLR9 polymorphism; SNP; HIV; HIV latency; viral rebound; interruption of antiretroviral therapy

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.