Natural killer (NK) cells are part of the innate immune response and function in tumor and viral control . They are regulated by the integration of signals transmitted internally by activating and inhibitory surface receptors such as killer immunoglobulin-like receptors (KIRs) . A number of KIR molecules have been linked to infectious disease outcomes, including alleles encoded by the KIR3DS/L1 framework locus [3–6].
KIR3DL1 (hereafter 3DL1) alleles can be divided into three groups based on their expression levels on the NK cell surface [7–10]: the 3DL1*h (high expression) group, the 3DL1*l (low expression) group and 3DL1*004, which is not expressed at the cell surface. Using these categories, it is possible to classify 3DL1 homozygotes (3DL1 hmz) as having a 3DL1*h/*y genotype when they do not express a *l allele or as having a 3DL1*l/*x genotype when they express at least one *l allele. Given that the ligands for 3DL1 are major histocompatibility complex (MHC) class I-B alleles of the Bw4 serotype, Martin et al.  measured the effect of these 3DL1 genotypes in combination with different MHC class I molecules on HIV disease progression . The strongest protective effect was found for the 3DL1*h/*y-B*57 combination.
NK cell function [11–13] and KIR [13–15] have not only been linked to HIV disease progression, but also resistance to HIV infection in certain populations of exposed uninfected individuals who remain seronegative despite repeated exposure to HIV. Given that NK cells have the potential to mediate antiviral activity at the earliest phase of HIV infection and that there is genetic evidence for KIR involvement in resistance to HIV, we sought to determine whether coexpression of the subtypes of 3DL1 and their receptors that are associated with a slower disease progression would occur more frequently in exposed uninfected than in recently infected HIV susceptible individuals. To this end we compared the 3DL1 and human leukocyte antigen (HLA) genotype distribution in exposed uninfected versus that of individuals enrolled in an HIV primary infection cohort.
A total of 227 3DL1 hmz (41 exposed uninfected individuals and 186 individuals in primary infection) were included in this study. These individuals were a subset of previously described populations . Informed consent was obtained from all study participants, and the research conformed to all ethical guidelines of all the authors' institutions.
Peripheral blood mononuclear cells (PBMCs) were isolated from blood by density gradient centrifugation (Ficoll-Paque; Pharmacia, Uppsala, Sweden) and cryopreserved in 10% dimethyl sulfoxide (DMSO; Sigma-Aldrich, St Louis, Missouri, USA) with 90% fetal calf serum (FCS; Canadian Life Technologies, Burlington, Ontario, Canada).
Genomic DNA was extracted from PBMC or Epstein–Barr Virus (EBV)-transformed cells using a QIAamp DNA blood kit (QIAGEN, Inc., Mississauga, Ontario, Canada). 3DL1 hmz were 3DL1 subtyped at the allele level by gene sequencing as previously described . Single nucleotide polymorphisms (SNP) corresponding to the 3DL1*h subtypes (3DL1*001, 3DL1*002, 3DL1*008, 3DL1*009, 3DL1*015, 3DL1*020), the 3DL1*l subtypes (3DL1*005, 3DL1*007) and 3DL1*004 were identified by aligning the sequenced DNA to a reference consensus sequence consisting of a contig of 3DL1 cDNA sequences. All primer sequences and amplification conditions are available upon request. Participants were typed for MHC class I expression using the line probe assay (Innogenetics Inc, Alpharetta, Georgia, USA) and by sequencing (Atria Genetics, South San Francisco, California, USA) if high resolution typing was needed to resolve the assignment of B alleles to the Bw4 or Bw6 public specificities based on amino acids at positions 77 to 83.
Statistical analysis and graphical presentation were performed using GraphPad InStat 3.05 and GraphPad Prism 4.01 (GraphPad Software Inc, San Diego, California, USA). Fisher's exact test was used to compare proportions of selected genotypes between exposed uninfected individuals and individuals in primary infection. A P value less than 0.05 was considered significant. For comparisons of 3DL1 subtype frequency, a Bonferroni's correction was performed by multiplying the P value by the number of 3DL1 alleles tested (n = 9) to give the corrected P value, which was considered significant if less than 0.05.
There were no significant between-group differences in racial composition: 92.7 and 91.9% of exposed uninfected individuals and individuals in primary infection, respectively were whites, 2.4 and 2.7% were African, 0 and 0.5% were Asian, 0 and 3% were Hispanic, 0 and 0.5% were Native American, and 4.9 and 1% reported themselves of mixed race. All individuals sequenced for 3DL1 subtype were 3DL1 hmz in order to be able to classify them as 3DL1*h/*y (in which *y is *004 or *h) or 3DL1*l/*x (in which *x is *004, *l or *h) . The frequencies of each 3DL1 subtype for both populations are shown in Fig. 1. The only significant difference was in the frequency of *009, which was higher in exposed uninfected individuals (2.4%, 2/82 alleles) compared with individuals in primary infection (0%, 0/372 alleles) (P = 0.0323; odds ratio, 23.14; 95% confidence interval, 1.10–486.90; Fisher's exact test). However, this difference was attributable to a single exposed uninfected individual *009 hmz and was no longer significant after performing Bonferroni's correction (P = 0.2907). When alleles were grouped, based on expression, into *h and *l, we found no significant between-group differences in the distribution of these categories (*h: 59.8 and 57.3% and *l: 13.4 and 22.0% in exposed uninfected individuals and individuals in primary infection, respectively), although there was a nonsignificant trend (P = 0.096) toward lower frequency of *l alleles in the exposed uninfected population (Fig. 1 and data not shown).
The ligands for 3DL1 are alleles belonging to the HLA- Bw4 serotype. HLA-Bw4 alleles can be split into HLA-Bw4 80I and 80T alleles based on whether an Isoleucine (I) or Threonine (T) is present at amino acid 80 of the HLA heavy chain. In a recent publication, 3DL1 subtypes with their ligands conferred various degrees of protection in terms of time to AIDS and viral load, the strongest protection being associated with 3DL1*h/*y and HLA-B*57 coexpression . We thus compared the distribution of each of these alone and in combination in our study populations in order to determine whether they were differentially distributed. As shown in Fig. 2a, we found a higher proportion of 3DL1*h/*y-HLA-B*57 in exposed uninfected individuals than in individuals in primary infection [five of 41 (12.2%) versus five of 186 (2.7%) expressed this combined genotype, respectively; P = 0.019; odds ratio, 5.03; 95% confidence interval, 1.38–18.3; Fisher's exact test]. Considered on its own, expression of the 3DL1*h/*y haplotype was not significantly different between the two groups [68.3% (28/41) and 57.0% (106/186) for exposed uninfected individuals and individuals in primary infection, respectively; P = 0.221, Fisher's exact test] (Fig. 2b). Despite not being statistically significant, the percentage carriers of the HLA-B*57 allele tended to be higher in the exposed uninfected individuals than individuals in primary infection [five of 41 (12.2%) and eight of 186 (4.3%) were HLA-B*57 in these two groups, respectively; P = 0.0631, Fisher's exact test] (Fig. 2c). All other between-group comparisons of KIR3DL1 alleles or genotypes (3DL1*h/*y, 3DL*l/*x, 3DL1*004) and their receptors (HLA-Bw4, HLA-Bw4-80I, HLA-B*57 or HLA-B*27) taken individually or in combination were not significantly different (data not shown).
In this report, we assessed the distribution of 3DL1 subtypes and of their MHC class I ligands in HIV-resistant exposed uninfected individuals and individuals with HIV-primary infection. The latter were selected as a control population for these analyses because they were susceptible to HIV infection and because, in a recently infected population, any effect that disease progression would have on skewing the distribution of KIR or MHC class I allele expression has yet to occur. For example, as duration of untreated infection increases, individuals with alleles associated with rapid progression would be lost due to death, increasing the frequency of individuals expressing alleles associated with slow progression. The combined effect of high expression NK receptor 3DL1 subtypes in individuals expressing a 3DL1*h/*y genotype with its HLA-B*57 ligand, which has been shown to confer protection from disease progression and viral load control in HIV-infected individuals , is also associated with a reduced risk of infection as an increased proportion of 3DL1 hmz exposed uninfected individuals compared with individuals with 3DL1 hmz primary infection expresses this combined genotype. Limiting the study to homozygous individuals restricts the scope of our findings, but eliminates possible confounders associated with its allele, 3DS1, such as its effect on disease outcome, possible effects on 3DL1 expression levels or conflicting reports as to whether 3DS1 serves as a receptor for HLA-Bw4 80I or B*57 [3,15,17–21]. Although the frequency of 3DL1*h/*y was similar between groups, the difference in HLA-B*57 percentage carriers approached significance (P = 0.0631). Given the sample size of the exposed uninfected population and the lack of statistical significance of the observation, caution is warranted in concluding any effect of B*57 on its own. Thus, only coexpression of 3DL1*h/*y and HLA-B*57 was significantly linked to protection from HIV infection. However, it is possible that the combined effect of 3DL1*h/*y and B*57 is driven by the HLA allele. The small population size precluded making a definitive conclusion on this and confirmation of this observation in other exposed uninfected cohorts is warranted.
The mechanism underlying the synergistic protective effect of 3DL1*h/*y and HLA-B*57 in terms of HIV resistance may be related to the licensing process in NK cell development. During maturation, NK cells accumulate inhibitory receptors until they are able to quench autoreactivity upon encountering normal cells expressing self-MHC class I ligands [18–20,22–24]. NK cells from individuals with a 3DL1*h/*y genotype encode alleles expressed on the NK cell surface at high levels in terms of abundance per cell and the percentage of NK cells expressing the allele . In addition, the 3DL1*h alleles tested thus far have a higher affinity for HLA-Bw4 80I alleles, such as HLA-B*57, than HLA-Bw4 80T alleles [7,9]. Therefore, individuals expressing this combination of alleles would have NK cells that are potently inhibited under normal circumstances. As HIV target cells are exposed to the virus, MHC class I downregulation will lead to interruption of the inhibitory signal mediated by 3DL1 receptors, resulting in NK activation. Thus, in the case of the receptor ligand combination 3DL1*h/HLA-B*57 which mediates potent inhibitory signals, disruption of the interaction by viral infection has the potential to result in strong NK activation and antiviral activity that may play a role in preventing the establishment of an HIV infection. Functional studies evaluating the impact of different KIR-ligand combinations on NK cell activity are warranted to investigate the mechanisms underlying the association of HIV resistance with coexpression of KIR and MHC class I alleles, such as those reported here .
Whereas Martin et al.  were able to show a gradient of protection from HIV disease progression and viral load control for different 3DL1 and HLA combinations, we were only able to show a significant effect on protection from infection for the combination conferring the most potent protection vis-à-vis disease progression, that is, 3DL1*h/*y and B*57 in exposed uninfected individuals. Although other combinations influencing disease progression may also lower the risk of HIV infection, the sample size to which we had access was too small to detect the potentially more subtle effects of these interactions. It is also possible that 3DL1*h/*y and B*57 is the only 3DL1/HLA combination able to influence the risk of HIV infection.
We also evaluated the distribution of the 3DL1 subtypes in the two study populations. Although the 3DL1*009 allele had a higher frequency in the exposed uninfected population, a single 3DL1*009 hmz among the exposed uninfected individuals accounted for this between-group difference. Furthermore, following Bonferroni's correction for multiple comparisons, this difference was no longer significant such that 3DL1 allele frequencies did not differ between the two populations. The similar racial composition of the study populations likely was not responsible for differences in allele or combined genotype frequency between populations.
In conclusion, we have found that 3DL1*h/*y-HLA-B*57 individuals were more frequent in the exposed uninfected population when compared with a population in primary infection. This finding further supports a role of NK cells in the protection from HIV in the early stages of infection. However, this conclusion does not preclude a role for the full KIR genotype in HIV infection. Given the many possible combinations of KIR genes in a genotype and their variegated expression on NK cells, a complete picture of the influence of these molecules on disease outcome may not be defined by a single allele. Defining a role for KIR receptors in HIV susceptibility will contribute to our knowledge of immune defenses to this virus and may suggest new avenues for the design of HIV vaccines.
We would like to acknowledge Mr. Mario Legault and Mr. Martin Rioux for coordination of the Montreal primary infection and Injecting Drug Use (IDU) cohort, respectively, as well as the investigators of the Montreal primary infection cohort for subject recruitment and clinical follow-up. J.B., J.P.R., and C.M.T. are responsible for the cohorts from which clinical samples, behavioral and clinical information were obtained; M.K., J.K., P.K. S.S., and N.S. performed KIR and HLA genotyping and allotyping; S.B. analyzed results and prepared the article; N.F.B. designed the research and prepared the article.
Sponsorship: This study received support from the Canadian Institutes for Health Research #MOP-79515, the Fonds de la Recherche en Santé du Québec (FRSQ). S Boulet was supported by a PhD. scholarship from FRSQ. J.P.R. is a scientific scholar receiving support from FRSQ.
1. Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev Immunol 2001; 1:41–49.
2. Lanier LL. NK cell recognition. Annu Rev Immunol 2005; 23:225–274.
3. Martin MP, Gao X, Lee JH, Nelson GW, Detels R, Goedert JJ, et al
. Epistatic interaction between KIR3DS1 and HLA-B delays the progression to AIDS. Nat Genet 2002; 31:429–434.
4. Parham P. Immunogenetics of killer cell immunoglobulin-like receptors. Mol Immunol 2005; 42:459–462.
5. Carrington M, Martin MP. The impact of variation at the KIR gene cluster on human disease. Curr Top Microbiol Immunol 2006; 298:225–257.
6. Martin MP, Qi Y, Gao X, Yamada E, Martin JN, Pereyra F, et al
. Innate partnership of HLA-B and KIR3DL1 subtypes against HIV-1. Nat Genet 2007; 39:733–740.
7. Carr WH, Pando MJ, Parham P. KIR3DL1 polymorphisms that affect NK cell inhibition by HLA-Bw4 ligand. J Immunol 2005; 175:5222–5229.
8. Gardiner CM, Guethlein LA, Shilling HG, Pando M, Carr WH, Rajalingam R, et al
. Different NK cell surface phenotypes defined by the DX9 antibody are due to KIR3DL1 gene polymorphism. J Immunol 2001; 166:2992–3001.
9. Yawata M, Yawata N, Draghi M, Little AM, Partheniou F, Parham P. Roles for HLA and KIR polymorphisms in natural killer cell repertoire selection and modulation of effector function. J Exp Med 2006; 203:633–645.
10. Pascal V, Yamada E, Martin MP, Alter G, Altfeld M, Metcalf JA, et al
. Detection of KIR3DS1 on the cell surface of peripheral blood NK cells facilitates identification of a novel null allele and assessment of KIR3DS1 expression during HIV-1 infection. J Immunol 2007; 179:1625–1633.
11. Scott-Algara D, Truong LX, Versmisse P, David A, Luong TT, Nguyen NV, et al
. Cutting edge: increased NK cell activity in HIV-1-exposed but uninfected Vietnamese intravascular drug users. J Immunol 2003; 171:5663–5667.
12. Montoya CJ, Velilla PA, Chougnet C, Landay AL, Rugeles MT. Increased IFN-gamma production by NK and CD3+/CD56+ cells in sexually HIV-1-exposed but uninfected individuals. Clin Immunol 2006; 120:138–146.
13. Ballan WM, Vu BA, Long BR, Loo CP, Michaelsson J, Barbour JD, et al
. Natural killer cells in perinatally HIV-1-infected children exhibit less degranulation compared to HIV-1-exposed uninfected children and their expression of KIR2DL3, NKG2C, and NKp46 correlates with disease severity. J Immunol 2007; 179:3362–3370.
14. Jennes W, Verheyden S, Demanet C, Adje-Toure CA, Vuylsteke B, Nkengasong JN, et al
. Cutting edge: resistance to HIV-1 infection among African female sex workers is associated with inhibitory KIR in the absence of their HLA ligands. J Immunol 2006; 177:6588–6592.
15. Boulet S, Sharafi S, Simic N, Bruneau J, Routy J-P, Tsoukas C, et al
. Increased proportion of KIR3DS1 homozygotes in HIV-exposed uninfected individuals. AIDS 2008; 22:595–599.
16. Thananchai H, Gillespie G, Martin MP, Bashirova A, Yawata N, Yawata M, et al
. Cutting edge: allele-specific and peptide-dependent interactions between KIR3DL1 and HLA-A and HLA-B. J Immunol 2007; 178:33–37.
17. Barbour JD, Sriram U, Caillier SJ, Levy JA, Hecht FM, Oksenberg JR. Synergy or independence? Deciphering the interaction of HLA Class I and NK cell KIR alleles in early HIV-1 disease progression. PLoS Pathog 2007; 3:e43.
18. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, et al
. Licensing of natural killer cells by host major histocompatibility complex class I molecules. Nature 2005; 436:709–713.
19. Raulet DH, Vance RE. Self-tolerance of natural killer cells. Nat Rev Immunol 2006; 6:520–531.
20. Anfossi N, Andre P, Guia S, Falk CS, Roetynck S, Stewart CA, et al
. Human NK cell education by inhibitory receptors for MHC class I. Immunity 2006; 25:331–342.
21. Lopez-Vazquez A, Mina-Blanco A, Martinez-Borra J, Njobvu PD, Suarez-Alvarez B, Blanco-Gelaz MA, et al
. Interaction between KIR3DL1 and HLA-B*57 supertype alleles influences the progression of HIV-1 infection in a Zambian population. Hum Immunol 2005; 66:285–289.
22. Held W, Dorfman JR, Wu MF, Raulet DH. Major histocompatibility complex class I-dependent skewing of the natural killer cell Ly49 receptor repertoire. Eur J Immunol 1996; 26:2286–2292.
23. Dorfman JR, Raulet DH. Acquisition of Ly49 receptor expression by developing natural killer cells. J Exp Med 1998; 187:609–618.
24. Parham P. Taking license with natural killer cell maturation and repertoire development. Immunol Rev 2006; 214:155–160.
25. Alter G, Martin MP, Teigen N, Carr WH, Suscovich TJ, Schneidewind A, et al
. Differential natural killer cell mediated inhibition of HIV-1 replication based on distinct KIR/HLA subtypes. J Exp Med 2007; 204:3027–3036.