KIR Gene Content Diversity in a Zimbabwean Population: Does KIR2DL2 Have a Role in Protection Against Human Immunodeficiency Virus Infection?

Abstract

Killer cell immunoglobulin-like receptors (KIRs) mediate natural killer cell function through interaction with their cognate human leukocyte antigen ligands. Thus, KIR gene variants have been implicated in resistance or susceptibility to viral infections. However, research on the role of these variants remains contradictory and inconclusive. In the present study, we investigated KIR gene content diversity and its association with human immunodeficiency virus (HIV) infection in an adult Black Zimbabwean population. Presence or absence of 15 KIR genes was determined in 189 HIV-infected adults and 97 HIV-uninfected blood donors using sequence specific primer polymerase chain reaction. Frequencies of KIR genes, genotypes, and haplotypes were compared between the cases and controls to identify putative associations between KIR gene variants and HIV status. We report in this study the frequencies of 15 KIR genes and 43 KIR genotypes (40 known and 3 novel) among Zimbabweans. Importantly, the frequency of the inhibitory KIR2DL2 gene was significantly higher in the uninfected group (62%) compared to the HIV-infected group (47%) (OR = 0.55, 95% CI: 0.33–0.90, p = 0.019). KIR2DL2/2DL2 homozygosity was also significantly higher in the uninfected group (35%) compared to HIV-infected group (53%) (OR = 0.33, 95% CI: 0.16–0.72, p = 0.005) under a recessive model. We conclude that the KIR2DL2 gene may be involved in protection against HIV infection. It may be possible that inhibitory KIR genes may have an important role to play in HIV acquisition among populations of African origin in whom the activating KIR genes are less frequent compared to among Caucasians.

Introduction

Resistance to human immunodeficiency virus (HIV) infection despite repeated exposure among a minority of the world population is partially explained by variation in genes that encode host immune factors. Among the immune factors, variants of genes encoding chemokine receptors, the human leukocyte antigen (HLA), and KIR are strong correlates of resistance/susceptibility to HIV infection (Martin and Carrington, 2013; Singh and Spector, 2009). KIRs are a diverse family of transmembrane activating and inhibitory proteins expressed on natural killer (NK) cells and some T lymphocytes. Activating and inhibitory KIRs (aKIRs and iKIRs, respectively) regulate the function of NK cells through binding their cognate HLA class I ligands (Lanier, 2005).

NK cells form the front line of defense against HIV and other viruses through cytotoxic destruction of virus-infected cells and release of cytokines that regulate the adaptive immune response (Vivier et al., 2008). NK cells can directly recognize stress ligands expressed on HIV-infected host cells leading to autolysis of the infected cell. Furthermore, secretion of chemokines by NK cells in response to HIV infection can block new cycles of infection (Scully and Alter, 2016). The critical role of NK cells in HIV infection control is also evidenced by the expansion of the NK cell population and the change in distribution of NK cell subsets in acute HIV infection (Mavilio et al., 2005). The KIR and HLA loci are highly diverse, hence antiviral functions of NK cells are heterogeneous among individuals and populations. Reported research findings on the effect of KIR gene variants on HIV infection are contradictory and inconclusive.

The KIR locus on chromosome 19 comprised 14 functional genes, 6 encoding activating receptors (KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, and KIR3DS1), 7 encoding inhibitory receptors (KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL5, KIR3DL1, KIR3DL2, and KIR3DL3), 1 encoding receptor with both inhibitory and activation activity (KIR2DL4), and 2 pseudogenes (KIR2DP1 and KIR3DP1) (Takeshita et al., 2013). Activating receptors carry short cytoplasmic tails while inhibitory receptors carry long cytoplasmic tails, hence the S and L designations, respectively (Marsh et al., 2003).

Genetic variation in KIR occurs at three levels as follows: (1) presence or absence of any of the 12 nonframework genes, (2) combination of genes that make up KIR genotypes and haplotypes, and (3) single nucleotide polymorphisms in the individual KIR genes (Carrington and Norman, 2003). Currently greater than 600 alleles, 50 haplotypes, and 398 genotypes have been identified (González-Galarza et al., 2015).

KIR genes and KIR/HLA class I gene variants are linked to outcomes of HIV disease course largely, but there is little evidence on how they influence HIV acquisition (Alter et al., 2007; Hens et al., 2016; Naranbhai et al., 2016; Singh et al., 2016). The activation role of KIR3DS1 justifies its candidature as an agent for resistance to HIV infection (Carrington et al., 2008; Hens et al., 2016; Hong et al., 2013). However, the KIR3DS1 gene is infrequent in Africans (generally below 25%) relative to Caucasians (>50%) and Asians (>30%) (Hollenbach et al., 2012).

There is emerging evidence of the influence of iKIR genes in differential outcomes of viral acquisition (De Re et al., 2015; Jennes et al., 2006, 2013; Petitdemange et al., 2014). Whether iKIR genes have greater influence on HIV acquisition among populations with low KIR3DS1 frequency remains unclear. Heterogeneity in frequencies and distribution of KIR genes, genotypes, and haplotypes among populations may explain the failure to establish the role of KIR genes in HIV acquisition.

The current study describes the frequencies and distribution of KIR genes, genotypes, and haplotypes among HIV-infected and HIV-uninfected adult Black Zimbabweans. Findings indicate a potential role of an iKIR gene KIR2DL2 in protection against HIV infection. This information may help predict risk of HIV infection.

Materials and Methods

Study participants

HIV-infected, antiretroviral therapy naive Zimbabweans (of Bantu origin) aged ≥18 years were enrolled under the Immunological and Virological Investigations of HIV-infected individuals with CD4 counts above 350 cells/μL (IVIHIV) Study at the Opportunistic Infections Clinic at Parirenyatwa Groups of Hospitals in Harare, Zimbabwe. Written informed consent was obtained from each participant. The study was conducted from 2010 to 2013. At enrolment, a questionnaire was administered to collect demographic data and medical history, whole blood was collected in ethylene diaminetetraacetic acid (EDTA) anti coagulant, and peripheral blood mononuclear cells (PBMCs) and plasma isolated within 6 h of collection. The PBMCs and plasma were immediately stored at −80°C until further use. In the current study, DNA was isolated from the enrolment archived PBMCs of 189 randomly selected HIV-infected individuals. EDTA anti-coagulated blood and demographic information was also collected from 97 consenting unmatched HIV-uninfected Black Zimbabwean blood donors aged ≥18 years attending the National Blood Service Zimbabwe (NBSZ) in Harare. The study was approved by the Medical Research Council of Zimbabwe (MRCZ/A/2016).

KIR typing

Genomic DNA was isolated from PBMC or whole blood at the University of Zimbabwe College of Health Sciences, Harare, Zimbabwe using the Quick-gDNA™ MiniPrep Kit (Zymo Research Corporation, Irvine, CA, USA) according to the manufacturer's instructions. KIR genotyping was carried out at the University of Oxford, Oxford, United Kingdom using sequence specific primer polymerase chain reaction (SSP-PCR) according to a previously described method (Martin and Carrington, 2008). Briefly, 2 pairs of SSPs were used to amplify each of 14 functional KIR genes: 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 3DS1, 3DL1, 3DL2, and 3DL3 and the pseudogene 2DP1. The ethidium bromide stained amplicons were electrophoresed on 2% agarose gel and visualized under ultraviolet light for determination of presence/absence of each gene. The allelic variants of the KIR2DS4 gene were also determined by gel electrophoresis since the deleted version of the gene (KIR2DS4v) is 22 base pairs shorter in length than the full gene (KIR2DS4).

Assignment of KIR genotypes and haplotypes

The combination of KIR genes occurring in an individual was used to assign a KIR genotype identity according to the Allele Frequency Net Database (AFND) (González-Galarza et al., 2015). KIR haplotypes AA (A) and Bx were assigned according to a previously described method (González-Galarza et al., 2015). In summary, four genes, KIR3DL2, KIR3DP1, KIR2DL4, and KIR3DL3, are present in almost all individuals and, thus, form the framework genes. Individuals carrying KIR2DL1, KIR2DL3, KIR2DL4, KIR2DS4, and KIR3DL1 genes in addition to the framework genes were assigned the AA haplotype. Individuals carrying all AA haplotype genes and any one of the following genes: KIR2DL2, KIR2DL5, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS5, and KIR3DS1 were denoted as AB haplotype, while individuals lacking any of the following: KIR2DL1, KIR2DL3, KIR3DL1, and KIR2DS4 were BB haplotype. Due to difficulties in distinguishing AB and BB haplotypes, the two groups form the Bx haplotype. Centromeric and telomeric haplotypes were assigned according to a previously described method (De Re et al., 2015). Briefly, KIR3DL3 and KIR3DP1 delimit the centromeric part of the KIR locus, whereas KIR2DL4 and KIR3DL2 delimit the telomeric part. Thus, the centromeric haplotypes were inferred from combinations of KIR2DS2, KIR2DL2, KIR2DL3, and KIR2DL1, whereas combinations of KIR3DS1, KIR2DS1, KIR3DL1, and KIR2DS4 formed telomeric haplotypes. KIR2DL5, KIR2DS3, and KIR2DS5 genes may occur in either centromeric or telomeric regions.

Statistical analysis

Statistical analysis was done using Stata version 13.0 (StataCorp, College Station, TX, USA). The Student's t-test was used to compare differences between means of continuous variables between HIV-infected and HIV-uninfected groups. Frequencies of genes, alleles, genotypes, and haplotypes were determined using direct counting. Pair-wise linkage disequilibrium (LD) between KIR genes was calculated using Cramer's V statistic and Lewontin's D′ on the MASSKIR software (Pandey et al., 2015). Heterogeneity in frequencies of genes, alleles, genotypes, and haplotypes between HIV-infected and HIV-uninfected groups was determined using chi-squared or Fisher's exact test as may be appropriate. A value of p < 0.05 was considered statistically significant. p-Values were adjusted for multivariate comparisons using Bonferroni correction method. Univariate logistic regression was used to determine association between KIR gene content variants and HIV status. Gene variants showing significant difference between HIV-infected and HIV-uninfected groups in univariate analysis were adjusted for gender and weight in a multivariate regression.

Results

Demographic characteristics of participants

The mean weight of the HIV-infected participants [66.9 ± 12.7 kg standard deviation (SD)] was significantly lower compared with the HIV-uninfected participants (74.8 ± 9.4 kg SD), p < 0.001. Females were overrepresented in the HIV-infected group than in the HIV-uninfected group (72% vs. 49%, p = 0.003) because women are more likely than men to seek healthcare in most African settings. The mean age was not significantly different between the HIV-infected (35.2 ± 9.4 years SD) and HIV-uninfected (37.5 ± 11.4 years SD, p = 0.074) participants.

Frequencies of single KIR genes and their association with HIV infection

We report frequencies of 15 single KIR genes in an adult Zimbabwean population (Table 1). As expected, three framework genes analyzed, KIR3DL3, KIR3DL2, and KIR2DL4, were present in all 97 HIV-uninfected controls. Four other genes, two inhibitory genes (KIR2DL1 and KIR3DL1), a pseudogene (KIR2DP1), and an activation gene (KIR2DS4) were present in >97% of the controls. KIR3DS1, a gene previously implicated in HIV acquisition and disease progression [reviewed by Hens et al. (2016)] was the least prevalent at 17% followed by KIR2DS1, which occurred in 22% of the control population. Three pairs of KIR genes exist as allelic variants KIR3DS1/KIR3DL1, KIR2DL2/KIR2DL3, and KIR2DS3/KIR2DS5 and will be referred to as KIR3DS1L1, KIR2DL2L3, and KIR2DS3S5, respectively. Either one or both of allelic variants at KIR3DS1L1 and KIR2DL2L3 loci were present in all participants. However, 40% of the control population had neither KIR2DS3 nor KIR2DS5 at the KIR2DS3S5 locus.

Table 1.

Frequencies of KIR Genes and Their Association with Human Immunodeficiency Virus Status

	HIV⁺ (n = 189)	HIV⁻ (n = 97)
Gene	Present	Present	OR ^a	95% CI ^b	p
KIR2DL1	188 (0.99)	96 (0.99)	1.96	0.12–31.65	0.636
KIR2DL2	89 (0.47)	60 (0.62)	0.55	0.33–0.90	0.019
			0.52 ^b	0.30–0.88 ^b	0.017 ^b
2DL3/2DL3	100 (0.53)	37 (0.38)	1
2DL2/2DL3	22 (0.12)	9 (0.09)	1.28	0.56–2.91	0.544
2DL2/2DL2	67 (0.35)	51 (0.53)	0.49	0.30–0.81	0.006
			0.46 ^b	0.27–0.79	0.005
KIR2DL3	167 (0.88)	88 (0.91)	0.78	0.34–1.75	0.342
KIR2DL4	188 (0.99)	97 (1.00)	N/A	N/A	N/A
KIR2DL5	107 (0.57)	58 (0.60)	0.87	0.53–1.44	0.44
KIR2DS1	35 (0.19)	21 (0.22)	0.82	0.48–1.51	0.521
KIR2DS2	76 (0.40)	46 (0.47)	0.75	0.46–1.22	0.244
KIR2DS3	39 (0.21)	27 (0.28)	0.67	0.38–1.18	0.165
2DS3/2DS3	24 (0.13)	18 (0.19)	1
2DS3/2DS5	15 (0.08)	9 (0.09)	0.84	0.35–2.00	0.699
2DS5/2DS5	68 (0.36)	31 (0.32)	1.19	0.71–2.01	0.499
KIR2DS5	83 (0.44)	40 (0.41)	1.11	0.68–1.83	0.671
KIR2DS4	187 (0.99)	96 (0.99)	0.97	0.08–10.8	0.6
2DS4f	84/188 (0.45)	41/93 (0.44)	1.02	0.62–1.68	0.925
2DS4f/v	65/188 (0.35)	33 (0.35)	0.96	0.57–1.61	0.88
2DS4v	36/188 (0.19)	18/93 (0.19)	0.98	0.52–1.85	0.967
KIR3DL1	188 (0.99)	97 (1.00)	N/A	N/A	N/A
3DL1/3DL1	162 (0.86)	80 (0.82)	1.27	0.65–2.47	0.473
3DL1/3DS1	26 (0.14)	17 (0.18)	0.75	0.38–1.46	0.4
3DS1/3DS1	1 (0.005)	0	N/A	N/A	N/A
KIR3DS1	27 (0.14)	17 (0.18)	0.78	0.40–1.52	0.404
KIR3DL2	189 (1.00)	97 (1.00)	N/A	N/A	N/A
KIR3DL3	189 (1.00)	97 (1.00)	N/A	N/A	N/A
KIR2DP1	187 (0.99)	96 (0.99)	0.97	0.08–10.8	0.373

Univariate logistic regression.

Adjusted for gender, age, and other genes.

95% CI, 95% confidence interval; HIV, human immunodeficiency virus; HIV⁺, HIV infected; HIV⁻, HIV uninfected; KIR, killer immunoglobulin-like receptor; OR, odds ratio.

Bold values show statistically significant differences between HIV⁺ and HIV⁻ groups.

When frequencies of the KIR genes were compared between the HIV-infected and HIV-uninfected groups to determine association between presence of single KIR genes with HIV status (Table 1), we found that the KIR2DL2 gene was significantly more frequent in the HIV-uninfected controls than in the HIV-infected cases (62% vs. 47%), suggesting that KIR2DL2 may protect against acquisition of HIV (OR = 0.55, 95% CI: 0.33–0.90, p = 0.019). The association between KIR2DL2 and the HIV uninfected status remained significant after multivariate analysis (adjusted OR = 0.52, 95% CI: 0.30–0.56, p = 0.017), but was not statistically significant after correcting for multiple comparisons (p = 0.209). None of the other 14 genes were associated with HIV infection.

When allelic variants at the KIR3DS1L1, KIR2DL2L3, and KIR2DS3S5 loci were considered, frequency of the KIR2DL2/KIR2DL2 genotype was significantly higher in the HIV-uninfected controls (53%) than the HIV-infected group (35%) (OR = 0.49, 95% CI: 0.30–0.81, p = 0.006) under a recessive model. The association between KIR2DL2 homozygosity and the HIV-uninfected status remained significant after multivariate analysis (adjusted OR = 0.46, 95% CI: 0.27–0.79, p = 0.005) and showed borderline significance after correction for multiple comparisons (p = 0.055).

Pair-wise LD between 12 KIR genes (framework genes were excluded from analysis) was calculated in HIV-infected and HIV-uninfected groups separately to determine patterns of coinheritance in the KIR locus (Table 2). Three pairs of genes were in strong positive LD in both groups: KIR2DL2 and KIR2DS2 (0.86 and 0.75, respectively); KIR2DL5 and KIR2DS5 (0.78 and 0.69, respectively); and KIR2DL1 and KIR2DP1 (0.71 and 1.00). KIR3DL1 and KIR2DS4 (0.71) showed strong LD in HIV-infected population, while KIR3DS1 and KIR2DS1 (0.68) were likely to be coinherited in the HIV-uninfected population.

Table 2.

Pair-Wise Linkage Disequilibrium Between KIR Genes in the Human Immunodeficiency Virus Positive and Human Immunodeficiency Virus Negative Groups

Genes	KIR3DL1	KIR2DL1	KIR2DL3	KIR2DS4	KIR2DL2	KIR2DL5	KIR3DS1	KIR2DS1	KIR2DS2	KIR2DS3	KIR2DS5	KIR2DP1
KIR2DL1	−0.005		0.325	−0.011	−0.083	0.124	0.059	0.052	−0.115	0.066	0.083	1
	NS		<0.005	NS	NS	NS	NS	NS	NS	NS	NS	<0.001
KIR2DL3	0.200	0.200		−0.032	0.262	−0.206	−0.042	−0.093	−0.344	−0.286	−0.325	0.319
	<0.01	<0.01		NS	<0.025	NS	NS	NS	<0.001	<0.01	<0.005	<0.005
KIR2DS4	0.705	−0.007	0.123		−0.081	−0.094	−0.221	−0.197	−0.113	−0.167	−0.128	−0.010
	<0.001	NS	NS		NS	NS	<0.055	NS	NS	NS	NS	NS
KIR2DL2	−0.076	−0.076	−0.381	−0.108		0.497	0.204	0.275	0.755	0.367	0.316	−0.081
	NS	NS	<0.001	NS		<0.001	<0.058	<0.01	<0.001	<0.001	<0.005	NS
KIR2DL5	−0.064	−0.064	−0.318	−0.090	0.559		0.176	0.244	0.445	0.523	0.695	0.121
	NS	NS	<0.001	NS	<0.001		NS	<0.025	<0.001	<0.001	<0.001	NS
KIR3DS1	−0.178	0.029	0.053	−0.253	−0.023	0.297		0.685	0.224	0.204	0.185	0.047
	<0.025	NS	NS	<0.001	NS	<0.001		<0.001	<0.05	NS	NS	NS
KIR2DS1	−0.153	0.034	0.002	−0.217	0.093	0.336	0.545		0.213	0.123	0.285	0.053
	<0.05	NS	NS	<0.005	NS	<0.001	<0.001		<0.05	NS	<0.01	NS
KIR2DS2	−0.089	−0.089	−0.409	−0.126	0.860	0.567	−0.024	0.110		0.483	0.214	−0.109
	NS	NS	<0.001	NS	<0.001	<0.001	NS	NS		<0.001	<0.05	NS
KIR2DS3	−0.145	0.036	−0.353	−0.206	0.447	0.440	0.060	−0.067	0.531		−0.088	0.063
	<0.05	NS	<0.001	<0.005	<0.001	<0.001	NS	NS	<0.001		NS	NS
KIR2DS5	−0.082	−0.082	−0.311	−0.117	0.354	0.775	0.370	0.430	0.320	−0.068		0.083
	NS	NS	<0.001	NS	<0.001	<0.001	<0.001	<0.001	<0.001	NS		NS
KIR2DP1	−0.00	0.705	0.285	−0.010	−0.108	−0.090	0.041	−0.084	−0.126	0.051	−0.117
	NS	<0.001	<0.001	NS	NS	NS	NS	NS	NS	NS	NS

Pair-wise LD plot for HIV-positive group (unshaded) and for HIV-negative group (shaded) calculated using Cramer's V statistic. Significance of LD at 0.05 confidence interval using chi-square test without Yates Correction.

LD, linkage disequilibrium; NS, nonsignificant p-value.

Bold values show statistically significant LD.

KIR genotypes and haplotypes and their association with HIV infection

Single KIR genes were combined to form KIR genotypes (Table 3 and Supplementary Table 1). A total of 43 genotypes were observed in the whole study population, 40 previously reported in other populations and 3 unique to the Zimbabweans as follows: (1) KIR3DL1-KIR2DL1, KIR2DL2-KIR2DL3-KIR2DS4-KIR3DS1-KIR2DS1-KIR2DP1, (2) KIR3DL1-KIR2DL1-KIR2DL2-KIR2DL5-KIR2DS4-KIR2DS1-KIR2DS2-KIR2DS5, and (3) KIR3DL1-KIR2DL2-KIR2DL5-KIR2DS4-KIR2DS2-KIR2DS5. Twenty-six (60%) genotypes occurred in 2 or more study participants while the remaining 17 (40%) occurred once. Seventeen (40%) genotypes were common to both HIV-infected and HIV-uninfected groups; 15 (35%) restricted to HIV-infected participants, while 11 (26%) occurred in the HIV-uninfected control group only. None of the genotypes observed were associated with HIV status.

Table 3.

Frequencies of KIR Genotypes in Human Immunodeficiency Virus Positive and Human Immunodeficiency Virus Negative Groups

HID	GID	HIV⁺ (n = 189)	HIV⁻ (n = 97)
AA	1	65 (0.34)	26 (0.27)
Bx	5	17 (0.089)	12 (0.12)
Bx	21	14 (0.074)	7 (0.07)
Bx	32	13 (0.07)	6 (0.06)
Bx	2	10 (0.053)	2 (0.02)
Bx	4	7 (0.037)	5 (0.05)
Bx	9	7 (0.037)	3 (0.03)
Bx	112	7 (0.037)	3 (0.03)
Bx	19	6 (0.032)	3 (0.03)
Bx	20	5 (0.026)	7 (0.07)
Bx	228	5 (0.026)	1 (0.01)
Bx	3	4 (0.021)	2 (0.02)
Bx	30	2 (0.011)	2 (0.02)
Bx	6	1 (0.005)	2 (0.02)
Bx	91	1 (0.005)	1 (0.01)
Bx	93	1 (0.005)	1 (0.01)
Bx	401	1 (0.005)	1 (0.01)
Bx	15	3 (0.016)	0
Bx	71	3 (0.016)	0
Bx	14	2 (0.011)	0
Bx	17	2 (0.011)	0
Bx	18	2 (0.011)	0
Bx	22	2 (0.011)	0
Bx	35	2 (0.011)	0
Bx	328	0	2 (0.021)
Bx	N	0	2 (0.021)
Bx	N	1 (0.005)	0
Bx	N	1 (0.005)	0

Shaded box: Gene present, clear box: Gene absent. Only genotypes with frequencies >1% and unique genotypes are shown.

GID, genotype identification; HID, haplotype identification; HIV⁺, HIV positive; HIV⁻, HIV negative; N, novel genotype.

Genotype 1, which belongs to haplotype AA, comprising six inhibitory genes and a single activation gene, was the most prevalent genotype in both HIV-infected (34%) and HIV-uninfected (27%) groups. However, the Bx haplotype, which comprises many genotypes each containing at least two activation genes, was predominant in both HIV-infected (66%) and HIV-uninfected (73%) groups. There were no significant differences in frequencies of AA and Bx haplotypes between the HIV-infected and HIV-uninfected groups.

We compared frequencies of centromeric and telomeric haplotypes between the HIV-infected and HIV-uninfected groups to determine their possible association with HIV infection (Table 4). The frequency of centromeric haplotype 1 (Ce1), which lacks the KIR2DL2, but carries the KIR2DL3, was significantly higher in the HIV infected than the HIV uninfected (53% vs. 38%, p = 0.02) compared to non-Ce1 haplotypes. Thus, individuals carrying the Ce1 haplotype were 1.82 times more likely to be HIV infected than individuals carrying any other centromeric haplotype. In contrast, Ce3, which carries both the KIR2DL2 and KIR2DL3 allelic variants, was associated with an HIV-uninfected status (OR = 0.44, 95% CI: 0.17–0.90, p = 0.028) as the haplotype was significantly overrepresented in the HIV-uninfected group than HIV-infected group (14% vs. 7%, p = 0.028) compared to non-Ce haplotypes.

Table 4.

Association of KIR Haplotypes with Human Immunodeficiency Virus Status

Haplotype	2DS2	2DL2	2DL3	2DL1	HIV⁺	HIV⁻	OR ^a	95% CI ^a	p^a
Centromeric
Ce2					55 (0.29)	37 (0.38)	1
Ce1					100 (0.53)	37 (0.38)	1.82	1.10–3.00	0.019
Ce3					13 (0.07)	14 (0.14)	0.44	0.17–0.90	0.028
Ce5					1 (0.005)	0	NA	NA	NA
Ce6					20 (0.11)	8 (0.08)	1.39	0.52–3.5	0.44
Ce7					1 (0.005)	1 (0.01)	0.76	0.08–9.2	0.76
Ce/Te	2DL5	2DS3	2DS5
Ce/Te1					15 (0.08)	9 (0.09)	1
Ce/Te2					68 (0.36)	31 (0.32)	1.19	0.71–2.01	0.499
Ce/Te3					24 (0.13)	18 (0.19)	0.63	0.33–1.24	0.187
Ce/Te6					83 (0.44)	39 (0.40)	1.14	0.69–1.87	0.606
Telomeric	3DS1	2DS1	3DL1	2DS4
Te2					17 (0.09)	13 (0.13)	1
Te1					147 (0.77)	73 (0.75)	1.1	0.62–1.97	0.707
Te3					16 (0.08)	7 (0.07)	1.1	0.47–2.99	0.712
Te4					8 (0.04)	3 (0.03)	1.38	0.35–5.3	0.636
Te6					1 (0.005)	0	NA	NA	0.477
Te9					1 (0.005)	1 (0.01)	NA	NA	0.99
Te10					1 (0.005)	1 (0.01)	NA	NA	0.99

Shaded box: Gene present, clear box: Gene absent.

Chi-squared test (unadjusted p-value).

Ce, centromeric; Te, telomeric.

Bold values show statistically significant differences between HIV⁺ and HIV⁻ groups.

Discussion

The current study highlights a possible role of KIR gene content diversity in HIV infection. KIRs are critical in the regulation of NK cell activation, hence potentially influence HIV acquisition through NK cell-based innate anti HIV immune activity. The KIR2DL2 gene is associated with an HIV-uninfected status evidenced by overrepresentation of KIR2DL2 gene carriers, KIR2DL2/KIR2DL2 homozygotes, and the Ce3 haplotype carriers in the HIV-uninfected group compared to the HIV-infected group. Activating genes KIR2DS1 and KIR3DS1 were infrequent in the Zimbabwean study population. We observed 43 KIR genotype profiles, 40 previously reported and 3 unreported. However, we found no association between KIR genotypes and HIV status. Pair-wise LD analysis of KIR genes reproduced the coinheritance of KIR2DL2 and KIR2DS2 genes that has been previously reported (Moesta and Parham, 2012; Nakimuli et al., 2013). Patterns of KIR gene pair coinheritance were mostly similar between the HIV-infected and HIV-uninfected groups.

The antimicrobial activity of NK cells is a function of balance between stimulation of activating and inhibitory KIRs on NK cells by their MHC ligands on target cells (Lanier, 2005). However, KIRs are not restricted to inhibition or activation roles as previously thought. Resistance to HIV infection largely correlates to the presence of activating KIR profiles such as the KIR3DS1 and haplotype Bx (Bashirova, et al., 2011; Hens et al., 2016; Hong et al., 2013) because activating KIR profiles are expected to enhance NK cell function. Nevertheless, we found inhibitory gene, KIR2DL2, more frequent among HIV-uninfected controls than HIV-infected individuals suggesting a protective function against HIV infection.

“Licensing” or “education” of NK cells by KIRs during NK cell maturation provides the most credible mechanism of KIR2DL2 protection against HIV infection. “Licensing” allots functionality to NK cells that possess iKIRs, which recognize self-HLA ligands during maturation to prevent uncontrolled immune activation (Hens et al., 2016). Receptors of the KIR2DL motif KIR2DL1/2/3 are recognized by ligands of the HLA-C allotype. While KIR2DL1 binds to HLA-C2 specifically, KIR2DL2 and KIR2DL3 bind to HLA-C1 strongly, but also cross-react with HLA-C2 (Moesta and Parham, 2012). Thus, expression of KIR2DL2 and/or KIR2DL3 receptors on NK cells in presence of HLA-C1 on other nucleated immune cells during licensing lowers the threshold of NK cell activation by signals from virus-infected cells. This facilitates a strong and quick immune response in the face of HIV infection (Hens et al., 2016). KIR2DL2-associated polymorphisms in the HIV genomes from chronic HIV patients suggest another mechanism of anti-HIV activity where KIR2DL2 receptors on NK cells bind HIV peptides presented on HLA-C (Alter et al., 2011).

Our study findings are consistent with reports from earlier studies involving both vertical and horizontal exposure to HIV (Jennes et al., 2006, 2013; Paximadis et al., 2011). KIR2DL2/3 genotypes alone and/or in absence of their cognate HLA-C1 ligands were associated with protection against HIV acquisition among Senegalese HIV discordant couples (Jennes et al., 2013) and also among commercial sex workers in Abidjan Cote d'Ivoire (Jennes et al., 2006). Low NK cell activation threshold was a plausible explanation for the protective effect of the iKIRs 2DL2/3 in both studies. Increased function and frequency of KIR2DL1/2/3 positive NK cells in primary HIV infection (Korner et al., 2014) make a case for antiviral activity of iKIRs. It appears that iKIR/HLA-C mismatches are critical in prevention of both vertical (Hong et al., 2013) and horizontal transmission of HIV. However, our study demonstrates that the KIR2DL2 gene may influence HIV acquisition although the role of HLA-C was not investigated. Whether presence of iKIRs may overcloud the role of HLA-C alone remains unclear.

Despite growing evidence of the influence of KIR2DL2/3 genes in HIV acquisition, a study among South Africans dismisses association between KIR gene variation and HIV acquisition (Naranbhai et al., 2016). South Africa is geographically close to Zimbabwe hence the two populations are assumed to have similar KIR gene frequencies. However, the South African study involved HIV⁻ females who were followed up until they seroconverted, whereas our study involved both men and women with an already known HIV status.

In contrast, the KIR2DL2 gene increases risk of HIV acquisition, while the KIR2DL3 protects against HIV infection among Polish Caucasians (Zwolińska et al., 2016). Differences in distribution of KIR genes between Africans and Caucasians possibly account for the contrast between our findings and those from the Polish study. Higher prevalence of the KIR3DS1 gene and other gene variants such as CCR5 Δ32 deletion among Caucasians compared to Africans perhaps overshadows the antiviral effect of the KIR2DL2 gene among Caucasians. The CCR5 Δ32 deletion is perhaps the single most important genetic determinant of HIV acquisition, but it is rare among African populations with an allele frequency of 0.1% compared to up to 9.8% in Caucasians (Williamson et al., 2000).

The distribution of several other genes implicated in HIV acquisition and disease course has been queried to reflect the HIV map among African and world populations (Mhandire et al., 2014; Skelton et al., 2014). Beyond HIV acquisition, presence of KIR2DL2 gene is associated with slow HIV disease progression (Naranbhai et al., 2016). The antiviral properties of KIR2DL2/3 are not restricted to HIV as presence of KIR2DL2/3 genes is also associated with protection against human lymphotropic virus type 1 and hepatitis C virus (De Re et al., 2015; Seich Al Basatena et al., 2011). Paradoxically, KIR2DL2 gene may also predispose to chronic HCV infection (Kusnierczyk et al., 2015).

We speculate that the antiviral influence of KIR2DL2 genes may be pertinent to African populations. The distribution of KIR genes, genotypes, and haplotypes is broadly similar, with minor variation among African populations (Nakimuli et al., 2013; Yindom et al., 2010). The KIR3DS1 gene, which is largely implicated in resistance to HIV infection and slow HIV disease progression, is infrequent among Zimbabweans (17%) and African populations in general (Nakimuli et al., 2013). This infrequence of the KIR3DS1 gene could limit its antiviral function among Africans. Perhaps, this facilitates function of iKIRs such as KIR2DL2 in activation of NK cells in response to viral infection.

However, we found KIR2DL2 in strong LD with KIR2DS2 in both HIV-infected (0.86) and HIV-uninfected groups (0.75). This finding is consistent with earlier reports (De Re et al., 2015; Moesta and Parham, 2012; Naranbhai et al., 2016). The two genes are in close proximity, thus tend to be inherited together. Whether the KIR2DS2 gene influences NK cell activity independently or through epistatic interaction with KIR2DL2 remains unclear. There is evidence that the KIR2DS2 can activate NK cells through an unknown mechanism (Moesta and Parham, 2012). The KIR2DS2 gene was more frequent in the HIV-uninfected than the HIV-infected participants, but the difference was not statistically significant, perhaps downplaying the independent role of the KIR2DS2 gene in HIV infection.

Diversity of KIR genotypes presents a challenge in determination of association between KIR genotypes and outcomes of disease. This study observed 43 KIR genotypes in 286 individuals, 40% of which occurred only once. Other populations have shown more genotypes (Chavan et al., 2016; Zwolińska et al., 2016). With growing literature and advances in genotyping technology, the list of KIR genotypes continues to grow as more novel genotypes are reported. Genotype 1 consisting of AA haplotype is the most prevalent genotype among African populations ranging from 12% to greater than 50% (Williams and Middleton, 2004; Yindom et al., 2010). The A haplotype is predominantly inhibitory, but contrary to expectation is linked to both protection and susceptibility to viral infection. No association was observed between KIR genotypes or KIR haplotypes with HIV status, until centromeric and telomeric haplotypes were considered separately.

Our findings require careful interpretation as the study has a few limitations. First, the statistical power of the study may have been limited by the sample size. Determination of association between KIR genotypes and risk of HIV infection requires a large sample size due to the high diversity of KIR gene content and genotype profiles. Second, association between KIR2DL2 gene variants and the HIV-uninfected status was only observed before Bonferroni correction for multiple comparisons, hence further studies are required to substantiate our findings. However, it must be noted that Bonferroni correction is a very stringent test, which may lead to rejection of pertinent findings. Third, other host genetic and immunological factors that were not investigated in this study may have a background effect on individuals' susceptibility to HIV infection. Such factors include variation in chemokine, chemokine receptor, cytokine, and HLA encoding genes.

Although the current study was limited to KIR genotyping, the findings strongly argue that KIR gene diversity may have a role in HIV acquisition. The study demonstrates the diversity of the KIR gene loci in a Zimbabwean adult population. To the authors' knowledge, this is the first study to report KIR gene, genotype, and haplotype frequencies in this population. The KIR2DL2 gene may be associated with protection against HIV acquisition. Inhibitory KIRs may have a critical role in viral infection among populations of African origin in whom the KIR3DS1 gene is infrequent.

Footnotes

Acknowledgments

The authors thank the Letten Foundation Research House, Norway; HIV Research Trust, United Kingdom; Germany Academic Exchange Service (DAAD); and the European and Developing Countries Clinical Trials Partnership for funding. The authors are also indebted to the Parirenyatwa Groups of Hospital, University of Zimbabwe College of Health Sciences, and Professor Sarah Rowland-Jones' laboratory at the Nuffield Department of Medicine, University of Oxford for facilitating the research activities. None of this would have been possible without the voluntary participation of the IVIHIV study participants and the NBSZ blood donors for which the authors are thankful.

Author Disclosure Statement

The authors declare there are no conflicting financial interests.

Abbreviations Used

References

Alter

, Heckerman

, Schneidewind

, et al. (2011). HIV-1 adaptation to NK-cell-mediated immune pressure. Nature, 476, 96–100.

Alter

, Maureen

, Nickolas

, et al. (2007). Differential natural killer cell–mediated inhibition of HIV-1 replication based on distinct KIR/HLA subtypes. J Exp Med, 204, 3027–3036.

Bashirova

, Thomas

, and Carrington

. (2011). HLA/KIR restraint of HIV: Surviving the fittest. Annu Rev Immunol, 29, 295–317.

Carrington

, Martin

, and van Bergen

. (2008). KIR-HLA intercourse in HIV disease, Trends Microbiol, 16, 620–627.

Chavan

, Swati

, Zakiya

, et al. (2016). Diversity in KIR gene repertoire in HIV‐1 exposed infected and uninfected infants: A study from India. J Med Virol, 88, 417–425.

De Re

, Caggiari

, De Zorzi

, et al. (2015). Genetic diversity of the KIR/HLA system and susceptibility to hepatitis C virus-related diseases. PLoS One, 10, e0117420.

González-Galarza

, Takeshita

LYC

, Santos

EJM

, et al. (2015). Allele frequency net 2015 update: New features for HLA epitopes, KIR and disease and HLA adverse drug reaction associations. Nucleic Acids Res, 28, D784–D788.

Hens

, Jennes

, and Kestens

. (2016). The role of NK cells in HIV-1 protection: Autologous, allogeneic or both?. AIDS Res Ther, 13, 15.

Hollenbach

, Nocedal

, Ladner

, et al. (2012). Killer cell immunoglobulin-like receptor (KIR) gene content variation in the HGDP-CEPH populations. Immunogenetics, 64, 719–737.

10.

Hong

, Paximadis

, Gray

, et al. (2013). KIR2DS4 allelic variants: Differential effects on in utero and intrapartum HIV-1 mother-to-child transmission. Clin Immunol, 149, 498–508.

11.

Jennes

, Verheyden

, Demanet

, et al. (2006). 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, 177, 6588–6592.

12.

Jennes

, Verheyden

, Mertens

, et al. (2013). Inhibitory KIR/HLA incompatibility between sexual partners confers protection against HIV-1 transmission. Blood, 121, 1157–1164.

13.

Korner

, Granoff

, Amero

, et al. (2014). Increased frequency and function of KIR2DLI-3(+) NK cells in primary HIV-1 infection are determined by HLA-C group haplotypes. Eur J Immunol, 44, 2938–2948.

14.

Kusnierczyk

, Mozer-Lisewska

, Zwolinska

, et al. (2015). Contribution of genes for killer cell immunoglobulin-like receptors (KIR) to the susceptibility to chronic hepatitis C virus infection and to viremia. Hum Immunol, 76, 102–108.

15.

Lanier

. (2005). NK cell recognition. Annu Rev Immunol, 23, 225–274.

16.

Marsh

SGE

, Parham

, Dupont

, et al. (2003). Killer‐cell immunoglobulin‐like receptor (KIR) nomenclature report, 2002. Eur J Immunogenet, 30, 229–234.

17.

Martin

, and Carrington

. (2013). Immunogenetics of HIV disease. Immunol Rev, 254, 245–264.

18.

Martin

, and Carrington

. (2008). KIR genotyping and analysis. Innate Immun, 415, 49–64.

19.

Mavilio

, Lombardo

, Benjamin

, et al. (2005). Characterization of CD56–/CD16+ natural killer (NK) cells: A highly dysfunctional NK subset expanded in HIV-infected viremic individuals. Proc Natl Acad Sci USA, 102, 2886–2891.

20.

Mhandire

, Pharo

, Kandawasvika

, et al. (2014). How does mother-to-child transmission of HIV differ among African populations? Lessons from MBL2 genetic variation in Zimbabweans. OMICS, 18, 454–460.

21.

Moesta

, and Parham

. (2012). Diverse functionality among human NK cell receptors for the C1 epitope of HLA-C: KIR2DS2, KIR2DL2, and KIR2DL3. Front Immunol, 3, 336.

22.

Nakimuli

, Chazara

, Farrell

, et al. (2013). Killer cell immunoglobulin-like receptor (KIR) genes and their HLA-C ligands in a Ugandan population. Immunogenetics, 65, 765–775.

23.

Naranbhai

, de Assis Rosa

, Werner

, et al. (2016). Killer-cell Immunoglobulin-like Receptor (KIR) gene profiles modify HIV disease course, not HIV acquisition in South African women. BMC Infect Dis, 16, 27.

24.

Pandey

, Sarangi

, Prakash

, and Agrawal

. (2015). MASS KIR analyzer: An in silico approach for analyzing the killer immunoglobulin receptor gene content and its diversity. Curr Bioinform, 10, 292–298.

25.

Paximadis

, Minevich

, Winchester

, et al. (2011). KIR-HLA and maternal-infant HIV-1 transmission in sub-Saharan Africa. PLoS One, 6, e16541.

26.

Petitdemange

, Wauquier

, Jacquet

, et al. (2014). Association of HLA class-I and inhibitory KIR genotypes in Gabonese patients infected by Chikungunya or Dengue type-2 viruses. PLoS One, 9, e108798.

27.

Scully

, and Alter

. (2016). NK cells in HIV disease. Curr HIV/AIDS Rep, 13, 85–94.

28.

Seich Al Basatena

, Macnamara

, Vine

, et al. (2011). KIR2DL2 enhances protective and detrimental HLA class I-mediated immunity in chronic viral infection. PLoS Pathog, 7, e1002270.

29.

Singh

, Qin

, Brummel

, et al. (2016). Killer cell immunoglobulin-like receptor alleles alter HIV disease in children. PLoS One, 11, e0151364.

30.

Singh

, and Spector

. (2009). Host genetic determinants of HIV infection and disease progression in children. Pediatr Res, 65, 55R–63R.

31.

Skelton

, Kampira

, Wonkam

, et al. (2014). Frequency variation among sub-Saharan populations in virus restriction gene, BST-2 proximal promoter polymorphisms: Implications for HIV-1 prevalence differences among African countries. OMICS, 18, 461–471.

32.

Takeshita

LYC

, Gonzalez-Galarza

, dos Santos

EJM

, et al. (2013). A database for curating the associations between killer cell immunoglobulin-like receptors and diseases in worldwide populations. Database (Oxford), 2013, bat021.

33.

Vivier

, Tomasello

, Baratin

, et al. (2008). Functions of natural killer cells. Nat Immunol, 9, 503–510.

34.

Williams

, and Middleton

. (2004). KIR allele frequencies in a Xhosa population from South Africa. Hum Immunol, 65, 1084–1085.

35.

Williamson

, Loubser

, Brice

, et al. (2000). Allelic frequencies of host genetic variants influencing susceptibility to HIV-1 infection and disease in South African populations. AIDS, 14, 449–451.

36.

Yindom

, Leligdowicz

, Martin

. (2010). Influence of HLA class I and HLA-KIR compound genotypes on HIV-2 infection and markers of disease progression in a Manjako community in West Africa. J Virol, 84, 8202–8208.

37.

Zwolińska

, Błachowicz

, Tomczyk

, et al. (2016). The effects of killer cell immunoglobulin-like receptor (KIR) genes on susceptibility to HIV-1 infection in the Polish population. Immunogenetics, 68, 327–337.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB