L. J. Montaner, The Wistar Institute, HIV-1 Immunopathogenesis Laboratory, 3601 Spruce Street, Room 480, Philadelphia, PA 19104, USA. E-mail: Montaner@wistar.org
The description of highly exposed individuals who remain seronegative (HESN) despite repeated exposure to human immunodeficiency virus (HIV)-1 has heightened interest in identifying potential mechanisms of HIV-1 resistance. HIV-specific humoral and T cell-mediated responses have been identified routinely in HESN subjects, although it remains unknown if these responses are a definitive cause of protection or merely a marker for exposure. Approximately half of HESN lack any detectible HIV-specific adaptive immune responses, suggesting that other mechanisms of protection from HIV-1 infection also probably exist. In support of the innate immune response as a mechanism of resistance, increased natural killer (NK) cell activity has been correlated with protection from infection in several high-risk cohorts of HESN subjects, including intravenous drug users, HIV-1 discordant couples and perinatally exposed infants. Inheritance of protective NK KIR3DL1high and KIR3DS1 receptor alleles have also been observed to be over-represented in a high-risk cohort of HESN intravenous drug users and HESN partners of HIV-1-infected subjects. Other intrinsic mechanisms of innate immune protection correlated with resistance in HESN subjects include heightened dendritic cell responses and increased secretion of anti-viral factors such as β-chemokines, small anti-viral factors and defensins. This review will highlight the most current evidence in HESN subjects supporting the role of epithelial microenvironment and the innate immune system in sustaining resistance against HIV-1 infection. We will argue that as a front-line defence the innate immune response determines the threshold of infectivity that HIV-1 must overcome to establish a productive infection.
From the earliest days of the human immunodeficiency virus (HIV)-1 epidemic, anecdotal evidence of high-risk HIV-exposed but persistently uninfected individuals generated hope that natural resistance to HIV-1 existed in some individuals. The description of persistently seronegative prostitutes in Nairobi, Kenya who maintained resistance to HIV-1 infection despite numerous years of high-risk activity confirmed that resistance to HIV-1, although rare, was possible . This early interest led to the recruitment of HIV-exposed but -seronegative individuals into geographically diverse cohorts of high-risk subjects based upon the route of exposure to HIV-1 (Table 1). Mucosal exposure to HIV-1 in the absence of infection was documented in numerous cohorts from across the globe, including commercial sex workers [1,2] and individuals practising unprotected heterosexual or homosexual sexual intercourse with an HIV-1-infected partner [3–7]. Importantly, the phenotype of vaginal  and rectal [8,9] mucosal resistance to infection in the absence of adaptive T cell responses has been recapitulated in low-dose simian immunodeficiency virus (SIV) rhesus macaque studies, where macaques remained uninfected even after multiple mucosal exposures to SIV, and yet could be infected if virus was given intravenously (i.v.). The absence of vertical transmission has been observed in children born to HIV-1-infected mothers and exposed to HIV-1 through natural birth and/or breast feeding [10–13]. Resistance to infection despite direct blood-borne exposures to HIV-1 were also seen among HIV-seronegative occupationally exposed health workers , haemophiliacs receiving tainted blood products [15,16] and i.v. drug users sharing needles [17–20]. The potential diversity of the exposure routes and varied epidemiological background of HIV-1 exposed, uninfected subjects initially complicated the creation of a unifying definition for these seemingly resistant individuals . Recently, at the Workshop of HIV-1 Exposed and Resistant Subjects held in Rockville, Maryland, USA, in July 2010, the mutually agreed-upon designation for high-risk HIV-exposed, uninfected individuals was changed to ‘highly exposed seronegative’ (HESN) subjects . HESNs were defined collectively as individuals lacking anti-HIV-1 IgG seropositivity or evidence of infection despite frequent exposure to HIV-1 and/or repeated high-risk behaviour in areas with high HIV-1 prevalence. The seronegative description addresses the possibility that some HESN subjects may have mucosal immunoglobulin (Ig)A responses to HIV-1, but by definition all HESN subjects must be anti-HIV-1 IgG seronegative and are often also tested for the presence of HIV-1 by ultra-sensitive polymerase chain reaction (PCR). In terms of documenting exposure to HIV-1, studies of HIV-1 discordant couples and haemophiliacs have had the advantage of known exposures to quantifiable amounts of HIV-1 . Nevertheless, studies of commercial sex workers and i.v. drug users have inferred exposure to HIV-1 based upon mathematical models of the frequency of high-risk activity and the prevalence of HIV-1 in the community being studied [1,18,22].
Table 1. Cellular mechanisms of immune control in resistance to human immunodeficiency virus (HIV)-1 infection.
Immune mediated mechanism(s) and references
Sexual exposure (male/female)
• Discordant couples • Sex workers
Individuals exposed to HIV-1 via unprotected vaginal intercourse with an HIV-infected person
Children exposed to HIV-1 through nursing. Sexually active adults
HIV-1-specific T cell responses (no data available) Heightened innate immune cell activity (no data available)
Throughout this review, we will compare and contrast the evidence for adaptive and innate responses as correlates of resistance in high-risk HESN subjects. We will also explore how mechanism(s) of innate resistance to HIV-1 in HESN subjects intersect or differ with mechanisms of control over HIV-1 replication during chronic infection.
HIV-1-specific T cell responses in HESN subjects
Since the first identification of HIV-specific T cell responses in HESN subjects , HIV-specific T cell responses have been identified in a number of high-risk uninfected individuals from multiple cohorts [3–5,14,24]. Subsequent reports confirmed the presence of antigen-specific T cell responses to HIV-1 in HESN subjects while characterizing the functional and proliferative capacity of HIV-specific T cells in these subjects [7,25–27]. Genetically, both major histocompatibility complex (MHC) class I  and human leucocyte antigen (HLA) class II  alleles have been associated with a reduced risk of infection with HIV-1. In terms of protection, the anti-viral mechanisms utilized by T cells against HIV-1 may come in the form of direct lysis of virally infected cells or through the secretion of anti-viral factors such as chemokines/cytokines or other CD8 non-cytolytic anti-viral factors (CNAR) . Together with the description of anti-HIV-specific responses in HIV-infected long-term non-progressor subjects controlling viral replication [31,32], these findings raised hope that the generation of antigen-specific T cell immune responses to HIV-1 following high-risk contact could result in protection from HIV-1 in subsequent exposures.
Despite the generation of strong CD4 and CD8 responses to HIV-1 antigens , the failure of the much-publicized STEP trial utilizing a Merck T cell-mediated vaccine to protect against HIV-1 infection [34,35] raised doubt about the role of HIV-1-specific T cells in mediating resistance against HIV-1 infection. Recently, a blinded study utilizing a highly sensitive in vitro expansion method of detecting CTL responses failed to identify HIV-specific T cell responses in the HESN partners among HIV-discordant couples from Zambia . Among HESN individuals with detectible T cell responses to HIV-1 antigens, the breadth and magnitude of the HIV-specific responses has often been significantly lower than comparable responses observed in HIV-1-infected individuals [25,37], due probably to the clear differences in antigen exposure between these subjects. Work from several groups showing that pre-existing CTL responses against HIV-1 do not ensure a sustained resistance against infection in some persistently exposed HESN subjects who later seroconvert [38–40] further dampened interest in the potential role of T cells in sterilizing immunity. Currently, the potential role of antigen-specific T cell responses to HIV-1 in natural resistance from infection remains debated, and it is currently unknown if HIV-1-specific T cell responses represent an active mechanism of protection or merely a marker of exposure to the virus, as suggested recently . The fact that 30–60% of HESN subjects lack detectable T cell responses to HIV-1 (reviewed elegantly by Piacentini et al. and Miyazawa et al. in complementary analyses of HESN studies to date [42,43]) suggests that the presence of adaptive anti-HIV T cell responses has not been a unifying functional attribute of HESNs. Rather, the collective evidence supports the notion that non-T cell-mediated immune responses may also be involved in protection from HIV-1 in a subset of HESN subjects.
Humoral responses to HIV-1 in HESN subjects
Similar to adaptive T cell responses, HIV-specific IgA responses have been identified in the mucosa and sera of high-risk HIV-exposed seronegative subjects from multiple HESN cohorts [5,44–48]. HIV-specific IgA responses have also been documented in the absence of infection following oral exposure to HIV-1 through unprotected oral sex [49,50] and breast feeding . Although there have been cohorts where no HIV-specific IgA has been evidenced , most HESN cohorts with documented mucosal exposure have evidenced detectable levels of HIV-specific IgA (see Table 2) [42,43]. Various reports have shown that HIV-specific IgA can neutralize HIV in ex-vivo assays [47,53], with most neutralizing epitopes found in gp41 and gp120 . HIV-specific IgA from HESN subjects has also been shown to inhibit transcytosis across epithelial barriers, suggesting a functional mechanism of action in protection against HIV-1 infection [54,55]. In addition to direct neutralization of viral particles, HIV-specific IgA responses may also trigger antibody-dependent cellular cytotoxicity (ADCC) of infected target cells in conjunction with innate immune cells bearing the IgA-specific Fc receptor, CD89 [56,57].
Table 2. Evidence for secreted factors in resistance to mucosal human immunodeficiency virus (HIV)-1 infection.
Immune-mediated mechanism(s) and references
Sexual exposure (male/female)
• Discordant couples • Sex workers
Individuals exposed to HIV-1 via unprotected vaginal intercourse with an HIV-infected person
Although these findings have renewed hope for a mucosal-based humoral HIV-vaccine, there is still an open debate as to whether these responses are truly protective. Some longitudinal studies have found a strong correlation between HIV resistance and IgA responses [48,58]. In contrast, a recent multi-laboratory blinded study  found that HIV-specific IgA responses were either absent or detected inconsistently in plasma or cervicovaginal lavage from many HESN sex workers from Tanzania. In the oral mucosa, research on HESN infants in Kenya showed that the frequency or titre of HIV-specific salivary IgA was similar between exposed, uninfected infants and infants who acquired HIV-1 . A larger study of Kenyan sex workers also found no correlation between HIV resistance and IgA responses . In summary, the presence of HIV-specific IgA responses at the site of infection may constitute one potential mechanism of resistance against HIV-1, but its relevance in protection of HESN subjects from HIV-1 transmission remains highly contested. Geographical sex work practice differences, such as the use of bleaching/drying douches in female sex workers from some African countries , may also greatly alter the risk of transmission and should be controlled for in order to establish more clearly the effectiveness of immune-mediated protective mechanism such as HIV-specific IgA.
Role of epithelial and secreted factors in preventing mucosal transmission of HIV-1
In addition to HIV-specific IgA mucosal responses many secreted factors have been associated with reducing mucosal transmission of HIV-1 infection, as summarized in several comprehensive reviews on the subject [62,63]. The CC (β)-chemokine family of chemokines in particular, including macrophage inflammatory protein (MIP)-1α, MIP-1β and regulated upon activation normal T cell expressed and secreted (RANTES), are presumed to play an important role in resistance to infection by competing with HIV-1 for use of the CCR5 co-receptor on target cells. Spontaneous and antigen-induced CC-chemokine production by peripheral blood mononuclear cells (PBMCs) from exposed but uninfected partners of HIV-1-infected individuals were observed in independent cohorts of HIV-discordant couples from North India  and France . HIV-1 exposed uninfected men who have sex with men have increased levels of several salivary CC-chemokines associated with the frequency of oral sexual behaviour . In addition to the oral mucosa, elevated RANTES expression was also observed in the genital mucosa of HIV-1-resistant Kenyan commercial sex workers . In the SHIV (virus combining parts of the HIV and SIV genomes) macaque model of repeated virus challenges, resistance to simian HIV infection was also associated with high plasma levels of RANTES as well as other soluble factors, including interleukin (IL)-8 and eotaxin . However, increased plasma levels of RANTES has also been observed in HIV-1 infection during primary infection and may constitute a marker for low-level viral replication .
Several additional small molecular weight proteins have been discovered in the mucosal secretions of HESN subjects from independent cohorts. The secretory leucocyte protease inhibitor, SLP1, is a human saliva protein exhibiting anti-HIV activity  that may constitute part of the mucosal barrier to HIV-1 resistance in the oral cavity. Lactoferrin, a component of breast milk and genital secretions, has also been shown to inhibit HIV-1 replication and transmission from dendritic cells (DCs) to T cells in vitro[70–72]. Nevertheless, cervicovaginal levels of lactoferrin, RANTES and SLP1 were tested in HIV-1 seronegative women at a high risk of heterosexual acquisition of HIV infection and were found to be associated with bacterial vaginosis and inflammation rather than exposure to HIV-1 . In contrast, elafin/trappin-2 was found to be elevated in the female genital tract of HESN Kenyan sex workers and was associated with protection against HIV-1 acquisition .
Defensins represent a family of small cationic peptides expressed in the mucosal epithelium with broad anti-microbial properties against HIV-1 and other sexually transmitted diseases relevant to HIV-1 transmission . Both α-defensins and β-defensins have been associated repeatedly with protection in several independent studies of HESN subjects [76–80]. This includes the description of alpha-defensins in the prevention of HIV transmission among breastfed infants  and the identification of elevated levels of both alpha and beta-defensins in sexually HIV-1 exposed but uninfected individuals [79,80]. Despite potent HIV inhibitory activity, however, cervicovaginal levels of α-defensins have also been associated with increased HIV acquisition due to their association with bacterial sexually transmitted infections . The role of α-defensins in HIV-1 vertical transmission remains contentious, with one study showing no association between α-defensin concentration in breast milk and risk of HIV-1 transmission  while another study showed the opposite . Overall, the varied secreted proteins identified in the mucosa of HESN subjects (summarized in Table 2) may represent true factors associated with reducing mucosal transmission of HIV-1 infection. Rather, they may reflect the innate immune response to genital tract inflammation due to ongoing bacterial infections or sexually transmitted diseases, which may be endemic in the case of sex worker cohorts. Taking the data as a whole, we interpret that soluble innate factors are likely to modulate the infectivity threshold for HIV-1 upon exposure. However, secreted anti-viral factors alone are unlikely to render a complete barrier to infection, and innate immune cells such as natural killer (NK) cells and DCs may also bolster the threshold to infection that HIV-1 must overcome.
Genetic evidence for NK cell control in HESN subjects
NK cells represent a critical component of the host innate immune response against viral infection and serve as a front-line defence against a diverse array of pathogens. Unlike antigen-specific CD8 ‘killer’ T cells, NK cells use the co-ordinated interaction of both inhibitory and activating receptors to recognize target cells that exhibit signs of stress and have down-regulated MHC class I (MHC-I) proteins to escape CD8 T cell recognition. Of the main types of NK inhibitory receptors, the killer inhibitory receptor (KIR) family exhibits a restricted pattern of expression and interact with only a limited subset of MHC class I ligands [83,84]. Nevertheless, inheritance of specific KIR alleles has profound implications for individual susceptibility to infectious diseases [85,86]. As shown in Table 3, the KIR3DL1/S1 locus has been associated with both slow progression to AIDS and resistance to HIV-1 infection. Inheritance of protective KIR3DL1high receptor alleles that lead to high cell surface expression and greater NK licensing were observed to be over-represented in a high-risk cohort of HESN i.v. drug users from Montreal compared to HIV-1-infected subjects from the same geographic area (68·3% compared to 57·0%, respectively) . KIR3DS1, an activating allele of the same KIR3DL1 locus, was also identified to be enriched in HESN subjects within the same Montreal cohort (13·8% compared to 5·3%, respectively) . A smaller study of high-risk HESN female sex workers from the Ivory Coast found no such association , although this latter finding is limited by the low frequency of the KIR3DS1 allele in African populations compared to Caucasians . In support of a functional link with these protective alleles, NK cells expressing KIR3DS1 have been shown to produce more interferon (IFN)-γ and mediate stronger inhibition of HIV-1 replication .
Table 3. Role of protective alleles in control of/and resistance to human immunodeficiency virus (HIV)-1 infection.
Outcome and references
AIDS, acquired immune deficiency syndrome; HLA, human leucocyte antigen; NK, natural killer; MHC, major histocompatibility complex.
Increased NK licensing, heightened NK cell responses
Additional evidence for the protective role of NK cells in resistance to HIV-1 stems from a genetic study linking variants in non-classical MHC class I HLA-E and HLA-G molecules with reduced susceptibility to heterosexual acquisition of HIV-1 . Among the NK inhibitory receptors, the CD94/NKG2A receptor complex is unique in that it interacts specifically with the non-classical MHC protein, HLA-E, which presents leader peptides from the other classical MHC class I HLA-A, B, C molecules [83,84]. Inheritance of the HLA-E*0103 genetic variant, which leads to increased surface expression of HLA-E proteins and heightened NK surveillance of virally infected cells that down-regulate MHC class 1 proteins, was associated with a decreased risk of human immunodeficiency virus 1 (HIV-1) infection in Zimbabwean women . Similarly, women carrying the HLA-G*0105N genotype, resulting in a null HLA-G inhibitory protein that cannot inhibit NK cells, also have a significantly decreased risk of HIV-1 infection . While these genetic data suggest that NK stimulatory alleles are associated with protection from infection in some HESN subjects, a good number of HESN subjects lack these protective alleles. We propose here that the repeated exposure of HIV-1 and other pathogens during high-risk activity could reflect an independent mechanism of resistance that induces changes in the tissue microenvironment leading to innate activation similar to those elicited in HESN subjects that inherit the target genotypes of resistance described above.
Functional evidence for heightened NK cell activity in HESN subjects
In 2003, Scott-Algara et al. were the first to investigate the functional role of the innate response in protection from HIV-1 in HESNs  by studying a well-described cohort of high-risk HESN i.v. drug users from Vietnam . Their findings showed convincingly that NK cells from HESN i.v. drug users exhibited a significant increase in their capacity to mediate cytotoxicity and secrete antiviral cytokines when compared to control uninfected donors or HIV-1-infected subjects . Functional modulation of NK responses has also been reported following mucosal exposure in a report from Montoya et al., showing that IFN-γ production by NK cells was elevated in a cohort of HESN individuals exposed to HIV-1 through sexual intercourse with a known HIV-1-infected partner .
Heightened NK activation marker (CD69) expression and increased NK cell degranulation (CD107a) are two NK cell surface changes associated with resistance to infection in several independent cohorts of HESN subjects, including perinatally exposed children born to HIV-1-infected mothers  and HESN i.v. drug users from Ho Chi Minh City, Vietnam . Recently, we also confirmed that NK cells from HESN subjects exhibited increased NK activation and degranulation as measured by CD69 and CD107a in a high-risk needle-sharing cohort of i.v. drug users from Philadelphia . While we did not observe a statistically significant increase in the cytotoxic function of NK cells from HESN subjects against K562 tumour targets, we confirmed that heightened NK activation was not associated with a loss in activity or any sign of functional exhaustion.
We have shown previously that CD107a degranulated NK cells retain the capacity to lyse multiple targets in succession without a loss in cytotoxicity or viability and that CD107a expression represents a stable indicator of NK cell degranulation over time . Based upon these findings, we speculate that the higher CD107a expression observed in HESN subjects from our cohort and others reflect the evidence of sustained cytotoxic activity in vivo, as captured by the staining with CD107a ex vivo. Together with genotypic data showing an enrichment of protective NK receptor alleles in HESN subjects as discussed above [17,28], these findings suggest that increased NK activity is associated with protection from HIV-1 during high-risk activity (summarized in Table 1). Further research will determine what the relationship is, if any, between increased NK activation and the presence of protective NK KIR receptor genotypes or whether repeated exposures to pathogens during high-risk activity can sustain innate activation through DC activation of NK cells as described below.
Role of accessory cells in NK control of viral replication
In addition to signalling by inhibitory and activating receptors, NK activity is influenced directly by soluble cytokines secreted by accessory cells that can augment NK lysis such as IFN-α, IL-2, IL-12, IL-15, IL-18 and IL-21 [93–95]. In particular, plasmacytoid DCs (PDC), through the secretion of IFN-α, have been shown to be essential for orchestrating early resistance mechanisms against acute viral infection [96–98]. PDCs recognize ssRNA and dsDNA pathogens through the use of their intracellular Toll-like receptors (TLR) TLR-7 and TLR-9, and comprise the main IFN-α secreting cell type in the blood. In vitro, PDC secretion of IFN-α has been shown to be necessary for NK-mediated lysis against several virally infected target cell types including herpesvirus-infected fibroblasts [99–103] and HIV-infected autologous CD4+ primary T cells . The secretion of IFN-α by PDC may also limit the spread of HIV-1 at the site of infection prior to NK cell recruitment through the direct or indirect anti-viral activity of type-1 IFNs and the induction of intracellular defences against lentiviruses such as APOBEC3G and tetherin [105–108]. Indeed, the uniform recruitment of PDC cells able to express IFN-α at the subepithelial layer of the endocervix following vaginal exposure to SIV raises the hypothesis for an antiviral role for this cellular subset in mucosal resistance to infection .
Recently, we confirmed previous reports of increased NK activation in HESN subjects and showed for the first time that increased PDC maturation is also a marker of the heightened innate immune activation state in a cohort of i.v. drug users from Philadelphia . Despite a state of persistent activation, both PDCs and NK cells from HESN i.v. drug users maintained strong effector cell function and did not exhibit signs of exhaustion. In a parallel study with commercial sex workers from Puerto Rico, we have also observed that heightened PDC maturation was increased in HESN subjects exposed through high-risk sexual contact (Shaheed and Montaner, unpublished findings), supporting a potential role for PDC activation/maturation in sustaining HESNs states. Recently, TLR stimulation and responses were studied in a cohort of high-risk HESN subjects practising unprotected sexual intercourse . The data from Biasin et al. suggested that stimulation through TLR-3, TLR-4 and TLR-7/-8 in HESN individuals resulted in a more robust release of immunological factors, including IL-1β, IL-6, TNF-α and CCL3 . If confirmed, heightened TLR stimulation in HESN individuals may maintain resistance to HIV-1 through the release of immunological factors that can influence the induction of stronger innate anti-viral mechanisms involving DC and macrophage subsets alike. Taken together, these data support the notion that DC-mediated innate immune activation may co-operate with DC-mediated T cell activation in lowering viral infectivity at the initial period between exposure and productive infection.
It should be noted, however, that a converse interpretation has also been proposed whereby the initial PDC recruitment and innate immune activation driven by HIV-1 exposure is a necessary step to recruit CD4+ target cells and establish a permissive site for viral replication [109,111]. Although more experimental data are needed to resolve this question, epidemiological data documenting resistance in HESN sex workers suggests that repeated mucosal exposure and the associated cell infiltrates do not result in higher infectivity, but rather a sustained resistance against HIV-1 infection . Further research utilizing animal models of SIV mucosal exposure would be helpful in elucidating if pathogen-induced DC activation at the site of exposure is associated with recruitment of NK cell activity and protection from HIV-1 infection in spite of the recruitment of CD4+ T cell targets.
NK activity as a correlate of resistance to HIV-1 infection as well as a correlate of control in limiting viral replication after infection
Most anti-viral mechanisms are expected to act both at preventing infection during exposure and in reducing viral replication after infection. However, adaptive T cell responses may be more effective at control of viral replication after infection, as memory responses are probably amplified as CD8 T cell effectors only after infection is established. In contrast, NK cells remain an immune cell type associated with both resistance to HIV-1 infection in HESN subjects and control over viral replication following infection.
The case for the anti-viral capacity of NK subsets during infection is suggested by its loss of function in chronic infection. Progressive HIV disease is associated clearly with increasingly impaired NK responses and the selective depletion of CD56dim NK cells during chronic HIV-1 infection [112–115]. The loss of CD56dim NK cells, the main circulating NK subset that mediates cytotoxicity, results in the enrichment of CD56null NK cells with decreased function [113,116–118]. HIV-1 replication also results in the altered expression of inhibitory and activating receptors on NK cells further impairing the lytic potential of the remaining NK pool [119–121]. Defects in the NK cell compartment have been hypothesized to be part of the profound immunodeficiency observed during chronic HIV-1 infection and host susceptibility to opportunistic infections . In contrast, NK frequency and IFN-γ production have been shown to be retained in HIV-1 long-term non-progressors . HIV-1-infected elite controllers that suppress viral replication in the absence of anti-retroviral therapy also exhibit NK activity that is comparable to uninfected control donors . Together, these results correlate an increasingly dysfunctional NK cell compartment after infection with loss in control over HIV-1 replication during chronic infection.
Genetic studies of the KIR3DL1 locus in disease progression studies indicate that inheritance of KIR3DS1 and KIR3DL1high receptor alleles in conjunction with their HLA ligands can delay disease progression [87,125]. These genotypes are the same as those observed to be over-represented in a high-risk cohort of HESN i.v. drug users and HESN partners of HIV-1-infected subjects [17,28]. As genotypes associated with higher CD8 T cell responses are also associated with increased NK activity (i.e. HLA-B*57, etc.), we interpret that NK cells can contribute to both resistance against infection and to viral control once infected (Table 3). Together with data illustrating increased activation [10,20,91] and function [6,19] of NK cells in HESNs, these results suggest that NK cells fit the model of a candidate cell type whose retained function and heightened activation status may contribute to both control over HIV-1 replication and resistance to HIV-1 in HESN subjects.
The identification of highly exposed but persistently uninfected individuals that maintain resistance to HIV-1 infection despite high-risk exposure has generated hope that mechanisms of natural resistance to HIV-1 may some day be translated into a sterilizing vaccine to prevent infection. The failure of T cell vaccine strategies [34,35] and pre-existing CTL responses in HESN subjects to HIV-1 to protect against HIV-1 infection [38–40] has dampened interest in the potential role of T cells in sterilizing immunity. Similarly, a recent study from Africa documenting an absence of consistent HIV-specific IgA responses in plasma or cervicovaginal lavage from HESN sex workers  is in agreement with previous findings indicating a lack of a direct correlation between HIV-resistance and IgA responses . Collectively, the presence of HIV-specific humoral or cellular responses has not been a unifying functional attribute among HESN subjects, thereby highlighting the potential role of non-adaptive mechanisms of immunity in protection from HIV-1. Genotypic and functional association between increased NK activity and resistance to HIV-1 infection in multiple cohorts of HESN subjects suggests that the innate immune response may play a greater role than proposed to date in maintaining natural resistance to infection in high-risk subjects. Alternatively, synergistic responses involving both the innate and adaptive immune compartments against HIV-1 may act in concert to resist infection with HIV-1. Examples of the co-operative response between the adaptive and innate immune system include the targeting of MHC class I highly expressing cells by CD8 T cells and the targeting of MHC-class I down-regulated cells by NK cells. Similarly, HIV-specific IgA antibodies may act alone in neutralizing HIV-1 (dimeric IgA), or may increase HIV-1 clearance by binding to macrophages or neutrophils via the monomeric IgA Fc receptor, CD89 [56,57]. During chronic infection, HIV-specific IgGs are known to mediate neutralization of viral particles while also complementing well with NK cells to trigger antibody-mediated antibody-dependent cytotoxicity of infected target cells. Moving forward, non-human primate studies modelling HESN resistance to infection will be critical in investigating the complementary role of innate and adaptive immunity in resistance to HIV-1 infection.
As shown in Fig. 1, we illustrate that along with physical and mucosal barriers to HIV-1 entry, innate immune cells may intercede as a front-line defence against the establishment of a productive infection and dissemination of HIV-1. As reflected by time-line of the uncharacterized ‘eclipse’ phase of acute infection (0–6-day period after mucosal exposure before any detectable viral RNA in circulation [126–129], HIV-1 needs to overcome many intrinsic and innate immune-mediated anti-viral mechanisms to establish a productive infection. As summarized elegantly in several recent review articles [41,43,62,63,130], secreted anti-viral factors are probably more effective early in infection (step 1) and at the site of infection rather than after viral dissemination. In contrast, intracellular barriers to infection such as APOBEC3G and Tetherin may limit viral production and egress at the later steps of infection (step 4). Innate immune cells, including NK cells and PDCs, are probably most powerful at the juncture of exposure (step 2) rather than after the virus has achieved systemic dissemination (step 5). During chronic infection, the NK response can contribute to viral control but it is expected that the CD8 T cell response will take over from the NK response in applying pressure to viral replication, although the multiple viral escape mechanisms HIV-1 employs will eventually render them both ineffective [131–134]. As the virus climbs towards productive infection, recruitment of activated CD4 cells and macrophages to the site of infection (step 3) may provide target cells to fuel viral replication. Ultimately, the virus needs to modulate infected targets against cell death while promoting activation and replication within activated T cells [135–138]. A local, occult or abortive infection may ensue during the eclipse phase, characterized by transient low-level viraemia and cell death. Localized pockets of viral replication probably trigger HIV-specific adaptive T cell responses in some HESN individuals in the absence of a systemic humoral IgG response. Nevertheless, HIV-1-specific T cell responses may only be able to limit viral replication at the juncture before dissemination (step 5), rather than at the earlier stages of viral entry. In the SIV/rhesus macaque model of intravaginal transmission, a strong virus-specific CD8 T cell response was documented in cervicovaginal tissues, but only several days after the peak of virus production . As a result, the authors describe the adaptive cellular immune response as ‘too late and too little’ to clear infection and prevent CD4+ T lymphocyte loss .
Taking all data together, we believe the evidence supports a major role for the epithelial microenvironment and the innate immune system in sustaining resistance against HIV-1 infection. NK cells and PDC cells, specifically, may represent candidate cell types whose retained function and heightened activation status may contribute to continued resistance to HIV-1 in some HESN subjects. Collectively, the innate immune response acts as a front-line defence that determines the threshold of infectivity (due in part to genotype) that HIV-1 must overcome to disseminate and establish a productive infection. HIV-1 may overcome these innate mechanisms of resistance in the case of high viral inoculum, mucosal trauma or co-infections that induce local infiltrates of activated T cells. Consequently, strategies aimed at augmenting innate resistance factors or NK cell activity may bolster natural barriers to HIV-1 infection regardless of genotype. Prophylactic approaches aimed at augmenting DC/NK cross-talk within sites of exposure or harnessing the ability of Fc-bearing immune cells to trigger ADCC as an innate/adaptive mechanism of protection warrant further investigation. The ultimate goal of such approaches is to understand how best to recruit innate and adaptive factors best suited to prevent infection before HIV-1 reaches its ultimate goal of dissemination and T cell activation/depletion. Once the onset of systemic HIV-1 replication in activated T cells starts in the gut/periphery during the post-eclipse phase of acute infection, it is probably too late to intercede with innate or adaptive immune-mediated mechanisms of resistance that are critical at the site of exposure.
This study was supported by grants from the National Institutes of Health (NIDA R01 DA028775, R01 AI073219, RO1 AI065279, Core grant P30 CA10815), the Philadelphia Foundation and funds from the Pennsylvania Commonwealth Universal Research Enhancement Program.
The authors do not have any conflicts of interest or any other disclosures.