Killer Cell Immunoglobulin-Like Receptor Alleles Alter HIV Disease in Children

Background HLA class I molecules are ligands for killer cell immunoglobin like receptors (KIR) that control the antiviral response of natural killer (NK) cells. However, the effects of KIR and HLA (KIR/HLA) alleles on HIV disease of children have not been studied. Methods 993 antiretroviral naïve children with symptomatic HIV infection from PACTG protocols P152 and P300 were genotyped for KIR and HLA alleles using the Luminex platform. Linear regression was used to test the association between genotypes and baseline pre-ART HIV RNA, CD4+ lymphocyte count, and cognitive score, adjusting for age, race/ethnicity and study. The interaction between genetic markers and age was investigated. To account for multiple testing the false discovery rate (FDR) was controlled at 0.05. Results Children with the KIR2DS4*ALL FULL LENGTH (KIR2DS4*AFL) allele had higher CD4+ lymphocyte counts. Among children ≤2 years of age, the KIR2DS4*AFL was associated with lower plasma HIV RNA and higher cognitive index scores. KIR Cent2DS3/5_1 had lower CD4+ lymphocyte counts in children ≤2 years of age, while the presence of Tel1, Tel2DS4_2, Tel2DS4_4, Tel8, Tel2DS4_6 had higher CD4+ lymphocyte counts in all children. Presence of Cent2, Cent4 and Cent8 was associated with increased HIV RNA load in children ≤2 years. Presence of KIR3DL1+Bw4 was associated with higher CD4+ lymphocyte counts in all children. Among children >2 years old, KIR3DS1+Bw4-80I was associated with higher plasma HIV RNA, and Bw6/Bw6 was associated with lower plasma HIV RNA compared to children with KIR3DS1+Bw4-80I. Conclusions Presented data show for the first time that specific KIR alleles independently or combined with HLA ligands are associated with HIV RNA and CD4+ lymphocyte counts in infected, antiretroviral naive children; and many of these effect estimates appear to be age dependent. These data support a role for specific KIR alleles in HIV pathogenesis in children.


Introduction
Natural killer (NK) cells are key components of the innate immune system that act as the first line of defense and regulate antiviral immune responses [1]. NK cells mediate cytotoxicity and cytokine release via a large panel of activating and inhibitory receptors [2,3]. Although human leukocyte antigen (HLA) gene products are fundamental to acquired immune responses, they are also important in innate immunity as ligands for the killer cell immunoglobulin-like receptors (KIRs) that modulate NK cell activity [4]. HLA class I molecules closely regulate KIR functions. Both the families of KIR and the HLA class I genes are extremely diverse suggesting that NK cell mediated innate immune responses are at least partly genetically predetermined [3;5]. KIRs are expressed on both T cells as well as NK cells and may inhibit or activate their function. HLA and KIR subtype combinations can mount unique innate immune responses against human immunodeficiency virus type-1 (HIV) infection [6]. HLA and KIR allele combinations can be both protective and deleterious against HIV-related disease progression [7;8] and can affect mother-to-child HIV transmission [9]. Independent and combined KIR and HLA (KIR/ HLA) genotypes and haplotypes with an activating profile (presence of activating KIRs or absence of inhibitory KIRs or their respective HLA ligands) have been associated with HIV disease [10][11][12][13][14][15][16][17]. Effects of KIR/HLA alleles on HIV disease of children have not been previously studied. In the analyses presented here, we estimated the effects of KIR and HLA genotypes on plasma HIV RNA, CD4 + lymphocyte count and cognitive index score using a unique cohort of antiretroviral naïve HIV-infected children.

Subjects and Methods Participants
Nine hundred and ninety three antiretroviral naïve children with symptomatic HIV infection from Pediatric AIDS Clinical Trial Group (PACTG) protocols P152 and P300 were included in the analyses [18;19]. P152 and P300 were multicenter, prospective, randomized, double blind, placebo controlled trials that assessed the efficacy of combination nucleoside reverse transcriptase inhibitor (NRTI) treatment regimens in symptomatic HIV-infected children in the United States prior to the availability of effective combination antiretroviral therapy. Important eligibility criteria included children of an age range of 3 months to 18 years with symptomatic HIV infection for P152 [18], and an age range of 42 days to 15 years with symptomatic HIV infection for P300 [19]. In these two protocols, CD4 + lymphocyte count, plasma HIV RNA and the cognitive score were measured at entry prior to initiation of therapy [18;19].
Scales of Intelligence-Revised (WPPSI-R, 36 months to 6 years) [22]; Wechsler Intelligence Scale for Children: Revised (WISC-R III, 6 years to 17 years) [23] and Wechsler Adult Intelligence Scale: Revised (WAIS-R, >17 years) [24] for P300. P152 used Bayley scales, McCarthy [25] scales, WISC-R and WAIS-R as age appropriate. All cognitive scores were standardized (mean = 100, SD = 16). Children having a cognitive score below 70 are typically considered impaired. Studies followed the human experimentation guidelines of the US Department of Health and Human Services. The University of California San Diego Institutional Review Board has approved this study. Parents or legal guardians provided written informed consent to participate in these studies. Written informed consent had to be signed prior to participation in the studies. Each participating site was required to have Institutional Review Board approval prior to initiating the studies at their site.

Statistical methods
Individual KIR genotypes and haplotypes [35] were analyzed for their association with HIV disease. Additionally, KIR and HLA alleles were analyzed independently and in combination (including previously reported KIR3DL1, KIR3DS1 with Bw4, Bw4-80I or Bw4-80T alleles) [6][7][8] for their effects on immunological, virological and neurocognitive outcomes. An indicator variable was created to indicate the joint presence of specific KIR and HLA alleles. In HLA-Bw6/Bw6 individuals, it does not matter which KIR3DL1 subtype is present because the ligand Bw4 is absent, and KIR3DL1 molecule is non-functional in these individuals. The Bw6/ Bw6 group was also used as a control group for the analysis of the KIR subtypes.
Multivariable linear regression was used to test the association between the KIR and HLA allelic variants, and three baseline outcome measures: HIV RNA load, CD4 + lymphocyte count, and cognitive score. In order to correct for heterogeneity of variance, the robust variance estimator was used [36]. As children 2 years of age often have more rapid disease progression than those older than 2 years and have more immature immune systems, it was hypothesized that the effects of host genetics on HIV disease may vary with age. Therefore, the interaction between each genetic marker and age group (age 2 years and >2 years) was investigated. For genetic markers with a marginally significant (p <0.1) genotype by age (age 2 years and >2 years) interaction, regression models were fit to each age group separately. Potential confounders that were included in the adjusted analyses were determined a-priori and included age, race/ethnicity and study (P152 vs. P300).
To control the false discovery rate (FDR), we used methods developed by Benjamini and Hochberg [37] and evaluated the results with the FDR value set at 0.1, as well as 0.05. All genetic associations with a p-value <0.05 in adjusted models were included in the summary tables along with the corresponding 95% confidence intervals (CI). Associations with an FDR <0.05 were considered to be statistically significant, while those with an FDR between 0.05 and 0.1 were considered to be marginally significant.

Baseline characteristics
Of the 993 antiretroviral naïve children with symptomatic HIV infection, 430 (43%) were from P152 and 563 (57%) from P300; 453 (46%) were male. The median age was 2.3 years with 605 (61%) identified as Black, 245 (25%) Hispanic, 127 (13%) White and 16 (2%) as 'Others' race/ ethnicity. Of the 986 subjects with baseline CD4 + lymphocyte counts, the median baseline CD4 + lymphocyte count was 778 count/mm 3 ; 825 subjects had baseline HIV RNA data with a median baseline log 10 RNA of 5.14. Of the 935 subjects with available baseline cognitive scores, the median baseline score was 83. Detailed characteristics of this population are provided in Table 1.
Distribution of KIR alleles, centromeric (Cent), telomeric (Tel) and combined Cent/Tel alleles and combined KIR/HLA alleles in the studied cohort are listed in Tables 2, 3 and 4.

Associations of KIR and HLA alleles with baseline CD4 + lymphocyte counts
Results of association of independent and combined KIR and HLA alleles with CD4 + lymphocyte counts are summarized in Table 5.

Associations of KIR and HLA alleles with baseline HIV RNA load
Results of association of independent and combined KIR and HLA alleles with HIV RNA are summarized in Table 6.
No statistically significant associations were observed for any of the combined KIR/HLA alleles on the baseline cognitive score. Furthermore, no combination of HLA-C1 or C2 alleles with KIR alleles were significantly associated with the baseline CD4 count, logRNA viral load or cognitive score.

Discussion
Natural killer cells modulate antiviral immune response [38] by mediating the KIR mediated lysis of the targeted infected cells [10][11][12][13][14][15][16][17]. Through the release of various cytokines, a strong adaptive immune response is activated that leads to T cell proliferation and a reduction in viral replication [39]. Functions of NK cells are regulated by the activating or inhibiting KIRs and their HLA class I ligands [4]. Thus, the presence of KIR alleles coding specific NK receptors and HLA Class I ligand alleles for NK cell function can alter the anti-HIV innate immune response in infected children. In the research presented, we tested the association of the specific KIR alleles independently or in combination with ligand HLA class I alleles with the HIVrelated disease status markers in children. First, we estimated the effects of independent KIR alleles on HIV disease. We found that the presence of the NK cell activating allele, KIR2DS4 Ã AFL was associated with a higher CD4 + lymphocyte count and lower viral RNA loads in children 2 years old, as well as an increase in baseline cognitive score. KIR2DS4 is an activating gene for NK function. Hence, the presence of KIR2DS4 is expected to be associated with a functional NK response and protective effects against HIV infection and disease. Contrary to our findings, a study conducted in antiretroviral naïve adults has recently shown that KIR2DS4 promotes HIV pathogenesis [40]. Reasons for the difference in our findings are unclear. It is possible that in adults, the adaptive immune response via HLA-HIV peptide-CD8 interactions predominates over the KIR mediated innate immune response observed in children.
The presence of inhibiting allele KIR2DL2 Ã 004 was associated with lower CD4 + counts compared to those without it. This effect of KIR2DL2 has been reported in adults where the presence of KIR2DL2 was also shown to be associated with a more rapid rate of CD4 + lymphocyte decline due to inhibition of NK cell function [17].
Similar to KIR2DL2 allele, the presence of the KIR2DS2 allele has been shown to be associated with a more rapid rate of CD4 + T lymphocyte decline [17]. Consistent with these findings, our study showed that the presence of KIR2DS2 allele was associated with higher plasma HIV RNA in children 2 years old.
A unique aspect of our study was the evaluation of the association of KIR alleles with HIVrelated central nervous system (CNS) impairment. Of note, although a few KIR alleles had a pvalue <0.05 for the cognitive score analyses, none retained significance after controlling for FDR. In a simian immunodeficiency virus (SIV) model of encephalitis in macaques, animals lacking strong NK cell responses developed more severe CNS lesions than those with robust responses [41]. In vivo, both macaque and human cells showed that NK cells mediated anti-SIV and anti-HIV cytolytic effects directed against the envelope protein. Hence, NK cells recognize and lyse cells expressing SIV and HIV antigens suggesting that NK cells affect HIV related CNS disease [41]. Thus, activating and inhibiting KIR alleles can modulate these CNS effects against HIV as observed in our studies. Second, we investigated the effects of KIR centromeric and telomeric alleles on the HIV related disease in children. Among these, the presence of centromeric allele 2DS3/2DS5/2DL1 [Cent2DS3/5(1)], 2DL1/2DL3/2DL2 (Cent3), 2DL1/2DL3/2DS2 (Cent4) or 2DL2/2DL3/2DS2 (Cent8) was associated with higher HIV RNA load in children 2 years old probably because of the predominant presence of inhibiting KIR molecules that inhibited NK cell function. Among the KIR telomeric alleles, the presence of 2DS4_AFL/3DL1 [Tel-2DS4(2)] or 2DS4AFL/ 3DL1 (Tel1)), 3DL1 [Tel2DS4 (4) or Tel8] and 2DS4AFL [Tel2DS4(6) or Tel 7] was associated with higher CD4 + lymphocytes count and lower viral RNA load in the whole cohort because of the preponderance of stimulating KIR molecules that activated NK cells for killing of HIV infected cells. Significant age and genotype interactions observed in our study may suggest an age dependent maturation of the adaptive immune response reflected in KIR/HLA mediated NK cell response.
Finally, we tested the presence of combined KIR and HLA alleles on HIV disease in children. KIR and HLA class I alleles have been shown to act both independently and synergistically to modify HIV disease progression in adults [17]. HLA-Bw4 molecules with isoleucine at position 80 (Bw4-80I) are ligands for inhibitory KIR3DL1 receptors and Bw4-80Ile. Combined with the activating KIR3DS1 allele, these have been found to be associated with delayed progression to AIDS [11]. In our study, the absence of KIR3DL1 and Bw4 (non-KIR3DL1+Bw4) was associated with a lower CD4 + lymphocyte count. These results are in concordance with those in adults where the presence of KIR3DL1 and Bw4 was associated with slower HIV disease progression [6]. NK cells kill their HIV-infected target cells in a receptor ligand-specific manner that involved activating KIR3DS1 and its putative ligand HLA-Bw4-80I [42]. However, in the current study, different to other studies in adults [10], the absence of KIR3DS1+Bw4-80I was associated with lower HIV RNA load compared to those with Bw6/Bw6 group. Reasons for these differences are not clear but may reflect a different NK cell mediated innate immune response in children compared to adults. This may also be explained by the unique nature of our cohort (approximately 61% African-American), wherein the frequencies and protective effects of HLA-B alleles on HIV disease progression differs from Caucasian cohorts.
In an earlier study in adults, the frequency of the KIR3DS1 (3DS1/3DL1)-Bw4 combination was significantly higher in highly exposed and persistently seronegative patients versus discordant couples [6;12]. Higher frequency of KIR3DS1/3DL1 heterozygotes and HLA-Bw4-80I has been associated with long-term non-progressors [10]. Consistent with these findings, our study showed that the presence of KIR3DS1 and Bw4-80I was associated with higher CD4 + lymphocyte counts and lower HIV RNA.
Due to the presence of different KIR and HLA alleles, the varied expression of KIRs on different NK cells and CD8 + T lymphocytes potentially generate selective antiviral responses. For example, HLA-B alleles encode a peptide epitope sequence that controls allele-specific interactions with the inhibitory KIR3DL1 allele. HIV infected cells that may avoid immunosurveillance by downregulating the HLA-B expression, are killed by NK cells upon the loss of inhibitory NK receptor signals such as KIR3DL1. Thus, NK receptors directly participate in the adaptive immune response due to the expression of KIRs on CD8 + T cells [43;44].
An association of KIR/HLA alleles has been reported with the risk for HIV mother-to-childtransmission in two studies [45;46]. In these studies, the presence of KIR2DS4 allelic variants had differential effects on in utero and intrapartum transmission. Additionally, a strong association has been observed with maternal HLA-B alleles independent of viral load. This finding implicates innate immune mechanisms via NK receptor KIR3DL1 that are triggered by a decrease in the expression of HLA class I molecules [47]. A decreased HLA-B expression on infected cells removes the inhibitory signal by KIR3DL1 and lowers the threshold for CD8 + Tcell activation by viral peptides enhancing the adaptive CD8 + T-cell anti-HIV response. Furthermore, a blockade of NK cell inhibiting KIR molecules can also improve the anti-HIV-1 activity of NK cells [48].
The presence of HLA-C1 or C2 alleles in combination with KIR2DL or KIR2DS alleles has been reported to be associated with HIV disease [49;50]; however we did not observe it in the children cohort. Reasons for these results are not clear but as noted above, these may reflect a difference in the maturity of NK cell mediated innate immune response in children compared to adults and the unique nature of our predominant African-American cohort.
In summary, our study has shown that KIR alleles are associated with altered HIV disease pathogenesis in children independently and in combination with HLA class I ligands. In general, effects of KIR alleles on HIV disease in children follow the pattern observed in adults. Additionally, there was an age dependent association of KIR alleles with HIV disease observed particularly in younger children suggesting an effect of maturation of innate and adaptive immune responses. These studies will help guide the development of KIR/HLA based therapeutic targets against HIV disease.