Toll-like receptor 9 polymorphisms influence mother-to-child transmission of human immunodeficiency virus type 1

Mother-to-child transmission (MTCT) is the main source of pediatric HIV-1 infection. MTCT of HIV-1 is multifactorial, with plasma viral load and mode of delivery being important maternal factors [1]. Innate immunity may contribute to host-viral interactions and impact the risk of MTCT of HIV-1 that occurs when the adaptive immune system is still under development.

Toll-like receptors (TLRs) are type 1 transmembrane proteins differentially expressed among immune cells. They recognize and bind to conserved pathogen-associated molecular patterns shared by large groups of microorganisms, and trigger the activation of signal transduction pathways, that in turn induce dendritic cell maturation and cytokine production [2]. These receptors play a central role in the activation of innate immunity [3]. A growing body of data supports a role of specific single nucleotide polymorphisms (SNPs) in several TLR genes in modulating the risk of bacterial and viral infections. Few studies have analyzed the role of TLR SNPs in clinical HIV-1 infection. An association between a TLR4 SNP and a higher susceptibility to tuberculosis in HIV-1-infected patients in Tanzania has been reported [4], and a functional TLR8 variant has been found to be associated with HIV-1 clinical outcome [5]. Recently, two SNPs in the TLR9 gene, the c.4-44G > A (rs352139) and the c.1635A > G (rs352140), were linked to progression of HIV-1 disease and viral load in adult patients [6‐8]. TLR9 recognizes unmethylated cytidine-phosphate-guanosine (CpG) DNA motifs in bacteria and viruses [9], and is expressed in a wide variety of human cells, including plasmacytoid dendritic cells.

To date, the relationship between TLR gene polymorphisms and perinatal HIV-1 infection has not been investigated. The present study focuses on the role of c.4-44G > A and c.1635A > G SNPs in the TLR9 gene in MTCT of HIV-1 by analyzing a large cohort of HIV-1-infected and HIV-1-uninfected infants, all born to HIV-1-infected mothers.

To analyze the impact of c.4-44G > A and c.1635A > G SNPs on HIV-1 perinatal infection, a cohort of HIV-1-infected (n = 118) and exposed HIV-1-uninfected (n = 182) children was analyzed with different genetic models. Genotype and allele frequencies for the two SNPs were in Hardy-Weinberg equilibrium in both groups and were in agreement with frequencies reported in the NCBI database on the Caucasian population. c.4-44GA was the most frequent genotype in both HIV-1-uninfected and HIV-1-infected children, while c.1635AG was the most frequent genotype in HIV-1-uninfected children. c.4-44GG and c.1635AA genotypes had a higher prevalence in HIV-1-infected infants. However, the two SNPs showed no significant association with the risk of HIV-1 infection in the codominant model (Table 1), or in other genetic models considered (data not shown).

Linkage disequilibrium analysis confirmed a strong disequilibrium level between the two SNPs with D' = 0.87 and r² = 0.74. To investigate their combined effect, given their close proximity, haplotype frequencies were estimated in both groups of children. [G;A] was the most frequent haplotype in both HIV-1-uninfected and HIV-1-infected children, while the [G;G] and [A;A] haplotypes had a low frequency, but were more prevalent in HIV-1-infected infants, leading to a significantly increased risk of MTCT of HIV-1 compared to the most frequent [G;A] and [A;G] haplotypes (Table 1).

Values of maternal viral load at delivery were available in a subgroup of 109 infants (median 8790 HIV-1 RNA copies/ml of plasma, range 40-2106650 HIV-1 RNA copies/ml of plasma). Analyses of allele and genotype frequencies and the Hardy-Weinberg equilibrium in this subgroup were in agreement with those in the entire cohort (data not shown). For prevention of MTCT of HIV-1, highly active antiretroviral therapy was recently recommended for all women with HIV-1 RNA levels of ≥ 1000 copies/ml of plasma [13]. A further analysis was thus performed categorizing mothers according to low (< 1000 HIV-1 RNA copies/ml of plasma) and high (> 1000 HIV-1 RNA copies/ml of plasma) viral load. As for the larger cohort, while neither of the two SNPs was significantly associated with a risk of HIV-1 infection, the [G;G] haplotype remained associated with a higher risk of MTCT of HIV-1 after adjustment for maternal viral load (OR = 10.04, 95% CI 1.70-59.37, p = 0.012; p = 0.017 significance level after Bonferroni correction).

TLRs play a central role in innate immunity and a number of genetic association studies suggest that some TLR polymorphisms may be associated with susceptibility to different diseases. There is convincing evidence that common TLR SNPs modulate immune function by modifying the control of inflammatory cascades, elaboration of effector molecules, pathogen killing, and interactions with the adaptive immune response [14]. TLR9 plays a pivotal role in the induction of first-line defense mechanisms of the innate immune system and triggers effective adaptive immune responses to different bacterial and viral pathogens [15, 16]. A few genetic polymorphisms within the TLR9 gene have been reported to be associated with a variety of inflammatory and infectious diseases [6‐8, 17, 18].

Our study investigates for the first time the relationship between c.4-44G > A and c.1635A > G SNPs of TLR9 and MTCT of HIV-1, revealing that the TLR9 [A;A] and [G;G] haplotypes are associated with a significantly increased risk of MTCT of HIV-1. Genetic variations of TLR9, as a key gene of innate immunity, could impact on immunological downstream responses that are critically important for host defenses. The function of these SNPs and haplotypes is still largely unknown. Neither of the two SNPs induces amino acid change, but several findings suggest that they may affect TLR9 expression [7, 8, 17]. Functional studies demonstrated a critical role for the G allele of the c.4-44G > A SNP on TLR9 expression [17], and the c.4-44GG genotype was found to be associated with slow disease progression in HIV-1 infected adults [6]. The c.1635A > G SNP was also found to influence the progression of HIV-1 disease [6‐8]. Furthermore, association of the c.1635A > G with CD4 cell count and viral load has suggested a role of this SNP in TLR9-mediated immune activation [7, 8]. It has been demonstrated that sooty mangabeys have reduced levels of innate immune activation during apathogenic simian immunodeficiency virus infection and that plasmocytoid dendritic cells from these animals produce markedly low levels of interferon α in response to TLR9 ligands [19]. Specific haplotypes in the TLR9 gene might influence the functional ability of TLR9 to elicit a defense mechanism by affecting susceptibility or resistance to infections.

In conclusion, our results demonstrate an important role of genetic variants of the TLR9 gene in modulating the risk of MTCT of HIV-1, thus confirming the relevance of innate immunity in perinatal HIV-1 infection. This knowledge may be valuable in the development of new therapeutic strategies including the use the specific adjuvants. More studies are needed to evaluate if strategies aimed at modulating innate immunity might be useful for future treatment of pediatric HIV-1 infection and AIDS.