The plasma HIV-1 RNA load at the “set point” (4–6, 27) as well as the associated HIV-1 DNA reservoir cells (28, 29) are lower in women compared to men. The most classically suggested hypothesis to explain this gender bias is the difference in the innate immunity observed between men and women (25). Indeed, the frequency of IFN-α-producing pDCs is higher in women compared to men after the stimulation with TLR7 ligands (8, 10, 24). One of the mechanisms explaining this enhanced activation threshold of pDCs in female is related to estrogens (24, 30). Interestingly, beside the female sex hormone effect, it was also shown that X-chromosome dosage contributed to this sex biased response of pDCs (31). TLR7 is encoded on the X-chromosome, and we recently showed that the TLR7 gene escapes X-chromosome inactivation in female immune cells (32). This mechanism is responsible for a significant increase in the amount of TLR7 protein in the leukocytes of women compared to men (32), and probably contribute to the enhanced capacity of pDCs to express higher levels of TLR7 mRNA and to transcribe higher levels of IFN-α/ß mRNA at a steady state and in response to TLR7 ligands (33). Thus, both estrogen-signaling and X-linked genetic factors contribute to the stronger functional response of female pDCs (24, 30, 31).

The female predominance in the TLR7-driven production of IFN-I by pDCs represents a robust phenotype that distinguish the innate immunity between men and women, then it is not surprising that genetic mutations have been selected during the evolution to counterbalance this effect. Among them, the NM_016562.3:c.32A>T (dbSNP rs179008) polymorphism of TLR7 is a missense SNP substituting a leucine (c.32A) for a glutamine (c.32T, p.Gln11Leu) in the leader sequence of the TLR7 protein. This SNP was previously identified by Oh and coll (34) as a risk factor for being infected by HIV-1 in c.32T allele carriers in women but not in men, whereas an accelerated deterioration of CD4⁺ T-cell counts was also reported in men carrying the T0 (c.32T) genotype ( Figure 1 ) (34). Not tested in women by Oh and coll (34), the functional impact of rs179008 c.32T on protein expression and in TLR7-driven pDC responses was recently examined in both sexes (26). Interestingly, the stronger TLR7-driven IFN-α responses of pDCs from women as compared to men (8, 23, 24) was further magnified when all carriers of the c.32T allele of TLR7 were excluded from the analysis (26). Mechanistically, c.32 T allele was identified as a sex-specific pQTL found to determine the lower TLR7 protein expression in female leukocytes, which mirrored its negative functional impact on the TLR7-driven production of IFN-α by pDCs from women but not from men ( Figure 1 ). Unusually, this nonsynonymous SNP affects the signal peptide which controls protein dosage at the translational level through the mRNA sequence itself, rather than through the alteration of the signal peptide function. Indeed, evidence was provided for a translation rate-limiting role of Leu₁₁ codon in allele T leader peptide. Leu11 corresponds to the first rare codon in the path of the ribosome consistent with the 5′ to 3′ direction of ribosome scanning, and we hypothesized that the ribosome occupying codon 11 may sterically hinder the recognition of the translation initiation codon by the next ribosome upstream, given the 20- to 30-nucleotide footprint of a ribosome (35) and the 32- to 39-nucleotide interval between consecutive ribosomes in tightly packed polysomes (36). Local translation efficiency depends on the abundance of aminoacyl-tRNAs with compatible anticodons, and this is positively correlated with the relative usage of synonymous codons (37). It is conceivable that the effect of the codon substitution on translation efficiency might be felt only in female cells. Differences in codon usage bias have been described between male-, female-, and non-sex-biased genes in insects (38), avians (39), and fish (40), possibly arising from a selective advantage linked to the translation efficiency in either sex. Whether the sex dependency observed in the TLR7 quantitative protein phenotype for rs179008 could be mediated by sex-differences in the regulatory mechanism of translation will deserve further investigations.

Because the T allele of rs179008 pQTL affects TLR7 protein expression and in turn TLR7-driven production of type I IFNs by pDCs in women, but not in men, its direct contribution on the clinical parameters of HIV-1 infection at the initial/acute phase of the disease was investigated in women ( Figure 1 ). Contrary to the study of Oh and coll (34), a meta-analytical combination of two separate case-control settings, restricted to Caucasian women, did not support a significant genetic association of rs179008 with HIV-1 acquisition through the sexual route in women (26). In a striking contrast, in HIV-1 primary infected women from the French ANRS PRIMO cohort, the carriage of the rs179008 T allele was associated in the initial/acute phase of the disease with a decrease in viremia, a reduction in HIV-1 DNA associated cells, and lower plasma concentrations of the interferon alpha/gamma-induced protein 10, IP-10 also known as CXCL10 ( Figure 1 ). Symptomatic presentations of acute HIV-1 infection were also less common in T/T homozygous women, where the RNA viral load is decreased by an average of eight to nine times (26). Patients with detectable plasma IFN-α exhibit elevated levels of IP-10, and plasma IP-10 can be used as a surrogate of the IFN-I signature to demonstrate a positive correlation between IFN-I with viral load (41–43), and with a rapid disease onset during primary HIV-1 infection (43, 44). Thus, the higher frequency of asymptomatic clinical presentations at diagnosis for the T/T homozygous women implies a trend for a delayed onset of the symptoms of acute infection in this subgroup (26). Others frequent TLR7 SNPs were also investigated, such as the frequent intronic tag SNP, rs179012, previously associated with HIV-1 set-point viral load, especially in females (45), and a frequent functional SNP in the 3′ UTR of TLR7, rs3853839 in genetic association with the risk of developing systemic lupus erythematosus in males (46). None of these SNPs alone were found to be associated with the clinical parameters of a primary HIV-1 infection. Together, the genetic association studies reinforced the conclusion that rs179008 was a functional polymorphism of TLR7 in its own right (26), and strongly suggested that effector functions associated with TLR7 activation might exert dominant deleterious effects on viral load during acute HIV-1 infection in women ( Figure 1 ).

HIV-1/SIV transmission primarily occurs in the female reproductive tract (FRT) in a small founder population of cells at mucosal sites followed by a period of intense replication (47) ( Figure 2 ). The local expansion is fueled by the influx of new target cells recruitment, including CD4⁺ T cells, through a gradient of chemokines produced by innate immune cells such as pDCs (20) and monocytes (48). These sequential events in early mucosal responses lead to CD4⁺ T cell recruitment ( Figure 2 ). This starts with the release of chemokines, such as CCL20, by the epithelial mucosa sensitized by SIV, followed by a massive recruitment of CCR6⁺ pDCs which becomes shortly activated to produce IFN-I, IP-10/CXCL10 and other CCR5⁺ T cell-attracting chemokines (20) ( Figure 2 ). The virus expands locally using the influx of target cells and then disseminate via lymphatic drainage to genital draining lymph nodes where more viruses are quickly produced in the T cell areas (47). SIV then spread through the thoracic duct to the lymphatic tissues resulting in an intense viral replication, resulting in high levels of plasma viremia and systemic dissemination of virus to lymphoid organs (47, 49).

Sex differences in the acquisition rates of HIV-1 infection have been observed with a two-fold higher risk for women compared to men following a serodiscordant encounter (50). This differences in transmission are likely a consequence of a unique anomatological characteristic of the FRT as compared to penile or rectal surfaces (51). Indeed, this anatomic difference is a critical domain of sex differences relevant to HIV infection with the paradox that an activated mucosal immunity at the cervicovaginal mucosa lowers the barrier to HIV infection (51). This was confirmed in female macaques exposed with TLR7 or TLR9 agonist prior SIV infection, with the aim to trigger anti-viral immunity, showing that an endosomal TLR engagement is effective to enhance infection and viremia (52). In favor of an immune quiescence model of protection, it was further reported that HIV-negative exposed sex workers exhibited a depressed mucosal immunity as compared to uninfected controls (53). Indeed, genital tract inflammation during early HIV-1 infection predicts an increased risk for HIV acquisition (54), and higher viral load at set point and CD4⁺ T cell depletion in women. Thus, as suggested by others (19, 20, 47), infection through the FRT may depend on IFN-I signaling for CD4⁺ T cell recruitment and virus propagation. By contrast, infection through the rectal mucosa, which contains many resident CD4⁺ T cells can be protected by the antiviral properties of IFN-I (19). Thus, depending on the sites of entry rectum versus FRT antiviral mediators such as IFN-I might be protective or deleterious, respectively (19).

Based on all these considerations, the author propose that an enhanced TLR7 protein expression in a subset of women (c.32A allele carriers), associated with an enhanced innate sensing of viral components through TLR7, characterized by a higher IFN-I production by female pDCs could contribute to promote HIV-1 mucosal transmission and/or replication in the female genital tract ( Figure 2 ). This may explain the strong protective effect we observed in all clinical parameters of HIV-1 acute infection in women carrying the rs179008 T/T alleles, which reduces TLR7 protein expression in all immune cells and negatively impact the production of IFN-α by pDCs in response to HIV-1 ssRNA (26) ( Figure 2 ). Thus, the higher functional responsiveness of woman pDCs appears deleterious in the general context of primary HIV-1 infection, suggesting that the better control of viral load and the HIV-1 reservoir in primary infected women compared to men is not due to the enhanced capacity of female pDCs to produce IFN-I during the early-stage of HIV-1 infection. This necessarily implies that the better control of viremia at setpoints in women is probably not a direct consequence of the anti-viral properties of IFN-I during the acute phase of infection as previously suggested (25) but could be due to a direct effect of estrogen-signaling on HIV-1 replication (55) or other unknown mechanisms ( Figure 2 ).

By contrast, the detrimental role of IFN-I in chronic infection is supported by many studies and probably explains the faster progression to AIDS in women than in men. In chronic HIV infection, women have a higher level of CD8⁺ T cells activation after adjustment for viral loads, that correlates with an enhanced functional response of pDCs, suggesting that the sex-bias in the TLR7-driven production of IFN-I by pDCs could contribute to the inflammatory mechanism responsible for the more rapid disease progression seen in women (8). A deleterious role of pDC-derived IFN-I production has been illustrated by the lymphocytic choriomeningitis virus (LCMV) murine model of persistent viral infection. Here, blockade of IFN-I signaling even during the acute phase of infection facilitated a long-term viral clearance (56, 57). In humanized mouse and SIV models, pDC-depletion or IFN-I receptor (IFNAR) blockade, respectively, rescued human T cell depletion and function despite elevated HIV-1 or SIV replication (18, 58). In these models, the upregulation of ISG, high level T cell activation, and T cell dysfunction persisted during antiretroviral therapy (ART), and blocking IFN-I signaling not only enhanced T cell recovery (59, 60) but also reduced the HIV-1 reservoir (59, 60), and delays HIV-1 rebound after ART discontinuation (59). Whether sex influences the therapeutic response to IFNAR blockade in chronic HIV will deserve further investigation.