Toll-like receptors (TLRs) recognize conserved antigenic motifs, enabling broad nonspecific pathogen recognition. Single nucleotide polymorphisms (SNPs) in TLR genes may impact the natural history of viral infections, including HIV [1–8] and human herpesviruses . The TLR9 1635 (rs352140) locus is of interest because having one or more copies of 1635A is associated with increased risk of HIV acquisition in discordant couples  and infants . Paradoxically, 1635AA was associated with lower HIV viral loads or slower CD4+ decline/HIV progression compared with 1635GG in some studies [2,6], including the infant cohort in which it was found to be associated with increased HIV acquisition risk . However, other studies found 1635AA to be associated with lower CD4+ cell count or higher HIV viral loads, suggesting differences in genetically distinct populations [5,7].
Cytomegalovirus (CMV) and Epstein–Barr virus (EBV) are often acquired in infancy [10–12], establish persistent infection with prolonged shedding, and are opportunistic infections in HIV [10,13–17]. To date, there are few studies of TLR polymorphisms in herpesvirus infections. Mutations in TLR3 increase susceptibility to herpes simplex virus encephalitis in children after primary infection . SNPs in TLR2, TLR4, and TLR9 have been examined in the context of congenital CMV; TLR9 1635A was associated with risk in Polish [19–21], but not Japanese infants . We are aware of no published studies examining TLR9 polymorphisms and primary EBV infection.
In a Kenyan infant cohort, having one or more copies of the TLR9 1635A allele was found to be associated with increased HIV acquisition risk and lower HIV viral load . In this secondary analysis, we evaluate whether 1635A is also associated with EBV and CMV risk or viremia. HIV-infected mother–infant pairs participated in an observational study in Nairobi, Kenya from 1999 to 2003 [23,24]. HIV, CMV, and EBV DNA levels were assessed from birth to 1 year in HIV-exposed uninfected (HEU) infants and HIV-infected (HIV+) infants, with sampling at birth, 1, 3, 6, 9, and 12 months of age; HIV+ children were followed for an additional year at quarterly intervals. Follow-up [23,25] and CMV  and EBV  assays are detailed elsewhere. For host genetic studies, infant DNA was extracted from cryopreserved blood and genotyped for TLR9 1635 (rs352140) and a set of genetic ancestry informative markers (AIMS) using an Illumina Custom Oligo Pooled Assay microarray platform (San Diego, California, USA) .
All analyses were conducted using StataSE 14.0 (College Station, Texas, USA), using two-tailed tests with alpha = 0.05. Based on previous associations between TLR9 and HIV, initial analyses were conducted assuming a dominant model of inheritance; we confirmed this was also the most appropriate inheritance model for CMV and EBV. The Mann–Whitney U test was used to compare the baseline (first detectable) viral DNA level in plasma between genotype groups, because this time-point had the most data and was the peak level for most infants. The analysis for viral acquisition differed for EBV and CMV because the pattern of acquisition differed by virus and HIV status. As EBV transmissions were dispersed over the first 2 years , Cox proportional hazards regression was performed to assess the risk of EBV infection, Kaplan–Meier was used to estimate probabilities of infection at 12 months, and the z test was used to compare probabilities between genotypes. As nearly all infants were CMV positive by 3 months of age, there were not adequate time-points for a time-to-event analysis ; Pearson's Chi-squared test was used to compare the proportion of infants with CMV detected by 1 month between TLR genotypes. AIMS were homogenous in this population and did not affect point estimates of association; given the small sample size, these were excluded from final analyses (data not shown).
The study cohort of 368 infants with host genetic data has been described elsewhere . Prevalence of the 1635A allele was 49% (180/368) in the overall cohort, with 188 GG (51%), 136 GA (37%), and 44 AA (12%).
CMV and TLR9 genotype data were available in a subset 37 HIV+ infants and 19 HEU. At 1 month of age, 42% of HIV+ infants with one or more copies of the 1635A allele had acquired CMV, compared with 11% in infants with the 1635GG genotype (P = 0.03; Table 1). We did not detect a difference in baseline CMV viral load by TRL9 genotype in either the HIV+ or HEU infants.
EBV and TLR9 genotype data were available in 62 HIV+ and 42 HEU infants. The probability of EBV infection by 12 months of age was 0.82 in HIV+ infants with one or more copies of 1635A, and 0.67 in infants with two copies of the 1635G allele (P = 0.2). The 1635A variant was not associated with EBV risk, univariately [hazard ratio (HR) = 1.5, 95% confidence interval (CI) = 0.74–2.9; P = 0.3], or adjusting for maternal CD4+ percentage at 32 weeks gestation [adjusted (a)HR = 1.4, 95% CI = 0.71–2.7; P = 0.3]. However, infants with one or more copies of the 1635A allele had a median baseline EBV viral load that was 0.4 log10 copies/ml lower than infants with two copies of the 1635G allele (P = 0.02).
In HEU infants, having one or more copies of the 1635A allele was associated with a 4.2-fold increased risk of EBV acquisition univariately (HR = 4.2, 95% CI = 1.2–15; P = 0.02), and adjusting for maternal CD4+ percentage (aHR = 4.2, 95% CI = 1.2–15; P = 0.03). The probability of EBV infection by 12 months of age was 0.67 in infants with one or more copies of 1635A and 0.21 in infants with two copies of 1635G (P = 0.003).
This small study suggests polymorphisms in TLR9 1635 may alter defenses against herpesvirus infections during infancy. Importantly, these effects appear to be modified by HIV infection, which has global and profound impacts on host immunity. Immunosuppression or unmeasured maternal factors may explain some of the differences observed between HIV+ and HEU infants. The A/G substitution is synonymous and occurs in exon 2 of the TLR9 gene, and it is unknown whether this SNP affects TLR9 expression or reflects changes in another gene in linkage disequilibrium. Studies to determine the effect of 1635A on TLR9 expression and function are needed to understand its impact on acquisition and viral control.
We would like to acknowledge the CTL study research personnel, laboratory staff, and data management teams in Nairobi, Kenya, and Seattle, Washington; the Department of Paediatrics and Child Health at Kenyatta National Hospital for providing facilities for laboratory and data analysis; Julie Overbaugh and Sandy Emery for provision of HIV diagnostics and viral studies; Meei-Li Huang for the EBV assays, and the University of Washington Northwest Genomics Center and the Center for Clinical Genomics at the University of Washington for technical assistance. We thank the Kizazi Working Group (UW Global WACh) for comments and insights provided during article development. Most of all, we thank the women and children who participated in the study.
The publication was made possible with support from the National Institutes of Health (NIH) awards K01AI087369 (NIAID, PI J.A.S.), R21AI073115 (NIAID, PI J.R.L.), R01HD023412 (G.J.-S.), U54 CA190146 (C.C.), and K24HD054314 (NICHD, PI G.J.-S.). K.B.-S. was supported by the University of Washington (UW) Institute of Translational Health Sciences Multidisciplinary Clinical Research Training Program (TL1 TR 000422) and the Fogarty International Clinical Research Scholars Program (grant number 5 R24 TW007988) from National Institutes of Health. This research was also supported by the UW Center for AIDS Research (CFAR) (new investigator and HIV-associated malignancy awards), an NIH funded program (P30AI027757) and the UW Global Center for Integrated Health of Women, Adolescents and Children (Global WACh) and the Canadian Institutes for Health Research (grant no. 136825; S.G.). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the funders.
Role of the funding source: The funding sources were not involved in the analyses or interpretation of data.
Conference presentation: The work was presented in part at the 2016 Conference on Retroviruses and Opportunistic Infections (CROI), poster no. 780.
Authors’ contributions: K.B.-S., R.M., and J.R.L. conducted the genomics studies and analysis, J.A.S. conducted the herpesvirus studies and analyses, G.J.-S./D.W. and E.M.-O. designed the cohort study and provided the clinical data and specimens, S.G. and C.C. provided expertise in the interpretation of the herpes data and manuscript preparation, J.A.S. and K.B.-S. wrote the article together.
Conflicts of interest
There are no conflicts of interest.
1. Beima-Sofie KM, Bigham AW, Lingappa JR, Wamalwa D, Mackelprang RD, Bamshad MJ, et al. Toll-like receptor variants are associated with infant HIV-1 acquisition and peak plasma HIV-1 RNA level
2. Bochud PY, Hersberger M, Taffe P, Bochud M, Stein CM, Rodrigues SD, et al. Polymorphisms in Toll-like receptor 9 influence the clinical course of HIV-1 infection
3. Freguja R, Gianesin K, Del Bianco P, Malacrida S, Rampon O, Zanchetta M, et al. Polymorphisms of innate immunity genes influence disease progression in HIV-1-infected children
4. Mackelprang RD, Bigham AW, Celum C, de Bruyn G, Beima-Sofie K, John-Stewart G, et al. Toll-like receptor polymorphism associations with HIV-1 outcomes among sub-Saharan Africans
. J Infect Dis
5. 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
6. Said EA, Al-Yafei F, Zadjali F, Hasson SS, Al-Balushi MS, Al-Mahruqi S, et al. Association of single-nucleotide polymorphisms in TLR7 (Gln11Leu) and TLR9 (1635A/G) with a higher CD4T cell count during HIV infection
. Immunol Lett
7. Soriano-Sarabia N, Vallejo A, Ramirez-Lorca R, Rodriguez Mdel M, Salinas A, Pulido I, 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
8. Valverde-Villegas JM, Dos Santos BP, de Medeiros RM, Mattevi VS, Lazzaretti RK, Sprinz E, et al. Endosomal Toll-like receptor gene polymorphisms and susceptibility to HIV and HCV co-infection – Differential influence in individuals with distinct ethnic background
. Hum Immunol
9. Netea MG, Wijmenga C, O’Neill LA. Genetic variation in Toll-like receptors and disease susceptibility
. Nat Immunol
10. Slyker JA, Lohman-Payne BL, John-Stewart GC, Maleche-Obimbo E, Emery S, Richardson B, et al. Acute Cytomegalovirus infection in Kenyan HIV-infected infants
11. Slyker JA, Casper C, Tapia K, Richardson B, Bunts L, Huang ML, et al. Clinical and virologic manifestations of primary Epstein–Barr virus (EBV) infection in Kenyan infants born to HIV-infected women
. J Infect Dis
12. Gantt S, Orem J, Krantz EM, Morrow RA, Selke S, Huang ML, et al. Prospective characterization of the risk factors for transmission and symptoms of primary human herpesvirus infections among Ugandan infants
. J Infect Dis
13. Kovacs A, Schluchter M, Easley K, Demmler G, Shearer W, La Russa P, et al. Cytomegalovirus infection and HIV-1 disease progression in infants born to HIV-1-infected women. Pediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted HIV Infection Study Group
. N Engl J Med
14. Chandwani S, Kaul A, Bebenroth D, Kim M, John DD, Fidelia A, et al. Cytomegalovirus infection in human immunodeficiency virus type 1-infected children
. Pediatr Infect Dis J
15. Nigro G, Krzysztofiak A, Gattinara GC, Mango T, Mazzocco M, Porcaro MA, et al. Rapid progression of HIV disease in children with Cytomegalovirus DNAemia
16. da Silva SR, de Oliveira DE. HIV, EBV and KSHV: viral cooperation in the pathogenesis of human malignancies
. Cancer Lett
17. Kutok JL, Wang F. Spectrum of Epstein–Barr virus-associated diseases
. Annu Rev Pathol
18. Zhang SY, Jouanguy E, Ugolini S, Smahi A, Elain G, Romero P, et al. TLR3 deficiency in patients with herpes simplex encephalitis
19. Wujcicka W, Paradowska E, Studzinska M, Gaj Z, Wilczynski J, Lesnikowski Z, et al. TLR9 2848 GA heterozygotic status possibly predisposes fetuses and newborns to congenital infection with human Cytomegalovirus
. PLoS One
20. Paradowska E, Jablonska A, Studzinska M, Skowronska K, Suski P, Wisniewska-Ligier M, et al. TLR9-1486T/C and 2848C/T SNPs are associated with human Cytomegalovirus infection in infants
. PLoS One
21. Wujcicka W, Paradowska E, Studzinska M, Wilczynski J, Nowakowska D. TLR2 2258 G>A single nucleotide polymorphism and the risk of congenital infection with human Cytomegalovirus
. Virol J
22. Taniguchi R, Koyano S, Suzutani T, Goishi K, Ito Y, Morioka I, et al. Polymorphisms in TLR-2 are associated with congenital Cytomegalovirus (CMV) infection but not with congenital CMV disease
. Int J Infect Dis
23. John-Stewart GC, Mbori-Ngacha D, Payne BL, Farquhar C, Richardson BA, Emery S, et al. HIV-1-specific cytotoxic T lymphocytes and breast milk HIV-1 transmission
. J Infect Dis
24. Shaffer N, Chuachoowong R, Mock PA, Bhadrakom C, Siriwasin W, Young NL, et al. Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial
25. Obimbo EM, Mbori-Ngacha DA, Ochieng JO, Richardson BA, Otieno PA, Bosire R, et al. Predictors of early mortality in a cohort of human immunodeficiency virus type 1-infected African children
. Pediatr Infect Dis J