Systematic review and meta-analytic findings on the association between killer-cell immunoglobulin-like receptor genes and susceptibility to pulmonary tuberculosis

ABSTRACT

Several studies have evaluated the association between killer-cell immunoglobulin-like receptors (KIR) genes and susceptibility risk to tuberculosis (TB) infection. Nonetheless, their outcomes have not been conclusive and consistent. Here we implemented a systematic review and meta-analysis of KIR genes association to susceptibility risk of pulmonary TB (PTB) infection to attain a clear understanding of the involvement of these genes in susceptibility to PTB infection. A systematic search was conducted in the MEDLINE/PubMed and Scopus databases to find case-control studies published before November 2020. Pooled odds ratio (OR) and 95% confidence interval (95% CI) were calculated to determine the association between KIR genes and risk of PTB infection. After comprehensive searching and implementing the inclusion and exclusion criteria, 10 case-control studies were included in the meta-analysis. Four KIR genes were found to have significant positive association with PTB susceptibility risk of infection, including 2DL3 (OR = 1.454, 95% CI = 1.157–1.827; P = 0.001), 2DS1 (OR = 1.481, 95% CI = 1.334–1.837; P < 0.001), 2DS4 (OR = 1.782, 95% CI = 1.273–2.495; P = 0.001) and 3DL1 (OR = 1.726, 95% CI = 1.277–2.333; P < 0.001). However, the results showed that the remaining KIR genes (2DS2-4, 2DL1, 2, 4, 3DL1-2) and two pseudogenes (2DP1 and 3DP1) did not have significant associations with risk of PTB infection. This meta-analysis provides reliable evidence that the KIR genes 2DL3, 2DS1, 2DS4, and 3DL1 may be associated with an increased risk of PTB infection.

Introduction

For the past 25 years, tuberculosis (TB), which is caused by Mycobacterium tuberculosis (M. tuberculosis) infection, has been one of the global public health emergencies. According to the World Health Organization (WHO) reports, about 10 million new cases of TB occur in 2019 [1]. About 87% (8.7 million) of these individuals are from the WHO 30 high-burden countries [2,3]. Today, prevention of TB occurs in several ways, such as screening of persons at high risk, early detection and treatment of infected individuals, and vaccination with bacillus Calmette Guérin (BCG) vaccine of newborns [4,5]. One of the big challenges for the treatment of TB is the emergence of multidrug-resistant tuberculosis (i.e. MDR-TB) and extensively drug-resistant tuberculosis (i.e. XDR-TB). Moreover, strains resistant to the majority of drugs available have been seen globally [6,7]. Co-infection with the human immunodeficiency virus (HIV) increases the risk for pulmonary TB (PTB) and extra-pulmonary TB [8,9].

Natural killer (NK) cells and macrophages are two main cells of the innate immune cells that synergistically play an important role in defense against M. tuberculosis [10]. NK cells exert their functions by several mechanisms, such as secretion of different mediators like granzymes, perforin, and granulysin. Moreover, there are several membrane-bound death receptors that cause a direct lysis of target cells [11–13]. Both activating and inhibitory cell-surface receptors like killer-cell immunoglobulin-like receptors (KIR) are involved in regulating the cytolytic functions of NK cells. NK cells and a minority of T cell population express the KIRs (CD 158 families) that are linked to human leukocyte antigen (HLA) class I molecules [14]. The KIR gene cluster is present by the leukocyte receptor complex (LRC) on chromosome 19q13.4 and its size is about 150 kb [15,16]. The nomenclature of KIRs is based on two factors; the first one is defined according to the number of extracellular immunoglobulin-like domains (2D or 3D) and the second one is based on the length of cytoplasmic tail [long (L), short (S)]. There are also two pseudogenes (P), which do not encode functional receptors [17,18]. Totally, there are 17 highly homologous KIR genes, including 6 activating genes, 9 inhibitory genes, and 2 pseudogenes [19,20]. KIRs containing short or long tails are considered as inhibitory or activating KIRs, respectively. Most of them are inhibitory KIRs, indicating that the interaction of this group with major histocompatibility complex (MHC) molecules results in the suppression of the cytotoxic activity of NK cells. Inhibitory KIRs have an immunoreceptor tyrosine-based inhibitory motifs (ITIM) in their intracellular tail, but activating KIRs have an immunoreceptor tyrosine-based activating motifs (ITAM) in their intracellular tail [21,22].

The KIR genes show high genomic diversity, and several studies have demonstrated that such diversity may cause susceptibility to some infections and autoimmune disorders. In fact, various studies investigated the association between KIR genes and susceptibility to risk of PTB infection [23–25]. Nonetheless, there is some inconsistency between the results reported by these studies. Therefore, we performed a systematic review and meta-analysis, for the first time, of case-control studies to establish an association between KIR genes and susceptibility to PTB infection.

Methods

This study was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [26] statement, including review of literature, study selection, inclusion and exclusion criteria, data extraction and quality assessment, and statistical analysis.

Search strategy and study selection criteria

A search for published articles in the well-known databases MEDLINE/PubMed and Scopus were carried out until November 2020. For searching, following keywords were used: ‘killer-cell immunoglobulin-like receptors’ OR ‘KIR’ AND ‘Polymorphism’ AND ‘Tuberculosis’ OR ‘Pulmonary tuberculosis’ OR ‘Mycobacterium tuberculosis Infections’ OR ‘TB’ OR ‘PTB’. After retrieving the identified articles, the references of the eligible papers as well as review articles were also evaluated to find any relevant studies. Only case-control human studies written in the English language were considered in the final selection list.

Inclusion and exclusion criteria

Studies that met the following criteria were included in the final analysis: 1) All studies with cohort and retrospective case-control design that investigated the association between KIR genes and risk of PTB infection”. 2) Case-control studies evaluating the presence/absence of certain KIR genes in the studied population. 3) Studies involving cases with the phenotype of ‘pulmonary tuberculosis’ and the healthy individuals as the control group. 4) Articles containing adequate information for measuring the odds ratios (ORs) and 95% confidence intervals (CIs). 5) Articles possessing enough information demonstrating the prevalence of KIR genes in the PTB patients and healthy cases. On the contrary, studies with the following criteria were excluded from the final selected list: 1) Reported duplicate publications and overlapping data. 2) Studies not providing enough data about the frequency of KIR genes. 3) Other types of publications like case reports, reviews, comment, and letters.

Extraction of data and quality check

Data that were extracted from the eligible articles included name of the first author, year of publication, country and ethnicity of the study subjects, genotyping technique, the number of case and control subjects, and KIR genes frequency. The fine outcome of two independent authors was evaluated and conflicts were settled through consensus. We used Newcastle-Ottawa Scale (NOS) for quality evaluation of the involved studies. The range of NOS is between 0 and 9 and studies with 0–3, 4–6, and 7–9 scores were assumed as low, moderate, and high-quality, respectively.

Statistical methods

The pooled odds ratio (OR) and the corresponding 95% confidence interval (CI) were measured for each KIR gene to evaluate the effect size of the association of their absence/presence with risk of PTB. In order to calculate the phenotypic frequency (pf) of the KIR genes, the percentage of positive numbers among all of the samples was determined. To calculate the genotypic frequency (gf) among the subjects, the gf = 1 – (1 – pf) ½ equation was used. The heterogeneity among the included studies was computed employing the Q test, and the extent of variation (true heterogeneity) in pooled samples was determined by I-squared (I²) test. The significance level of the I² results was set at P < 0.1 [27]. If heterogeneity between the individual studies was established though a statistically significant values for I² test, random effects model would be applied. Otherwise, the fixed effects model would be exerted. To evaluate the predefined sources of heterogeneity among the included studies, the meta-regression analysis was conducted based on the continent of the study population and year of population. To conduct sensitivity analysis, a successively omission of a given study with the highest impact on the heterogeneity test was carried out. The publication bias among the studies was determined by Egger’s test and Begg’s test (P < 0.05 level was set as a statistically significant level) [28]. To implement statistical analyses and meta-regression analysis, MedCalc (version 19.0.5, Seoul, Republic of Korea) and STATA (version 11.0; Stata Corporation, College Station, TX, USA) were employed, respectively.

Results

Characteristics of the eligible studies

According to the inclusion and exclusion criteria mentioned earlier, finally 10 case-control studies were opted for performing meta-analysis that had reported the association between KIR gene and risk of PTB infection in each population (Figure 1). Total number of PTB patients and healthy controls in these 10 published studies were 1357 and 1357, respectively, which were from various ethnicities, including Canadian, Asian, American and African countries. The publication date of the selected studies was from 2006 to 2015. The precise information about the included studies and frequency of the KIR genes in this meta-analysis is available in Table 1.

Figure 1.

Flow diagram of the literature search and study selection

Table 1.

Characteristics of the included studies in the meta-analysis

Author (Ref)	Published Year	Country/Race	Detection Technique	PTB Patients	Controls	KIR Genes
Mendez [29]	2006	Mexico/Latin American	PCR-SSOP	97	51	2DL1, 2DL2, 2DL3, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DS1
Tajik [23]	2011	Iran/Asian	PCR-SSP	107	100	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1, 2DP1, 3DP1
Mahfouz [30]	2011	Lebanon/Asian	PCR-SSP	103	38	2DL4, 3DL2, 3DL3
Lu [31]	2012	China/Asian	PCR-SSP	200	200	2DL1, 2DL2, 2DL3, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DS1
Lu [24]	2012	China/Chinese Han	PCR-SSP	109	110	2DL1, 2DL2, 2DL3, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DS1
Farhad Shahsavar [32]	2012	Iran/Asian	PCR-SSP	50	200	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1, 2DP1, 3DP1
Kali Braun [33]	2013	Canada/Canadian First Nations (Dene, Cree, and Ojibwa) and Caucasian controls	PCR-SSP	34	59	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1, 2DP1, 3DP1
Satya Sudheer Pydi [34]	2013	India/Asian	PCR-SSP	144	144	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1
Kali Braun [35]	2015	Canada/American	PCR-SSP	105	104	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1, 2DP1, 3DP1
Muneeb Salie [25]	2015	South Africa/South African Colored	PCR-SSP	408	351	2DL1, 2DL2, 2DL3, 2DL4, 2DL5, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DL1, 3DL2, 3DL3, 3DS1, 2DP1,

PCR; Polymerase Chain Reaction, PCR-SSP; Polymerase Chain Reaction-Sequence Specific Primer, PCR-SSOP; Polymerase Chain Reaction-Sequence Specific Oligonucleotide Probes; PTB, Pulmonary tuberculosis

Main results

Table 2 shows the summary of meta-analysis results regarding the association between KIR genes and PTB infection risk. Meta-analysis revealed significant positive associations of four KIR genes with PTB susceptibility risk of infection (Figure 2), including 2DL3 (OR = 1.454, 95% CI = 1.157–1.827; P = 0.001), 3DL1 (OR = 1.726, 95% CI = 1.277–2.333; P < 0.001), 2DS1 (OR = 1.481, 95% CI = 1.334–1.837; P < 0.001), and 2DS4 (OR = 1.782, 95% CI = 1.273–2.495; P = 0.001). However, the results indicated that the resting KIR genes, including 2DS2-4, 2DL1, 2, 4, 3DL1-2 and two pseudogenes (2DP1 and 3DP1) did not have statistically significant association with PTB infection susceptibility.

Table 2.

Meta-analysis of the pooled association between KIR genes and risk of TB infection

Gene	No. of studies	PTB case n/N	Control n/N	P-value	Pooled OR (95% C.I)	Heterogeneity Test (Q, I2; P-value)	Publication Bias (Begg’s test P-value; Egger’s test P-value)	Effect Model
2DL1	9	1090/1254	1152/1319	0.865	1.052 (0.586–1.888)	(11.42, 29.95%; P = 0.179)	(Begg’s test P = 0.48; Egger’s test P = 0.62)	Fixed
2DL2	9	810/1254	842/1319	0.553	0.901 (0.648–1.252)	(22.01, 63.66%; P = 0.0049)	(Begg’s test P = 0.46; Egger’s test P = 0.33)	Random
2DL3	9	1071/1254	1074/1319	0.001	1.454 (1.157–1.827)	(25.12, 68.16%; P = 0.0015)	(Begg’s test P = 0.58; Egger’s test P = 0.33)	Random
2DL4	8	1040/1048	1041/1047	0.594	0.749 (0.259–2.167)	(0.48, 29.8%; P = 0.87)	(Begg’s test P = 0.11; Egger’s test P = 0.021)	Fixed
2DL5	9	812/1254	836/1319	0.661	1.038 (0.879–1.225)	(9.8, 18.70%; P = 0.276)	(Begg’s test P = 0.97; Egger’s test P = 0.38)	Fixed
2DS1	9	591/1254	487/1319	<0.001	1.481 (1.334–1.837)	(73.01, 89.04%; P < 0.0001)	(Begg’s test P = 0.04; Egger’s test P = 0.14)	Random
2DS2	9	652/1254	684/1319	0.588	0.922 (0.687–1.238)	(21.65, 63.05%; P = 0.0056)	(Begg’s test P = 0.48; Egger’s test P = 0.29)	Random
2DS3	9	484/1254	438/1319	0.552	1.184 (0.679–2.063)	(54.85, 85.42%; P < 0.0001)	(Begg’s test P = 0.83; Egger’s test P = 0.58)	Random
2DS4	9	1193/1254	1213/1319	0.001	1.782 (1.273–2.495)	(10.18, 21.47%; P = 0.252)	(Begg’s test P = 0.04; Egger’s test P = 0.28)	Fixed
2DS5	9	572/1254	554/1319	0.685	1.078 (0.751–1.545)	(33.56, 76.16%; P < 0.0001)	(Begg’s test P = 0.58; Egger’s test P = 0.66)	Random
3DL1	9	1147/1254	1159/1319	<0.001	1.726 (1.277–2.333)	(11.82, 32.32%; P = 0.159)	(Begg’s test P = 0.03; Egger’s test P = 0.041)	Fixed
3DL2	8	1042/1048	1043/1047	0.523	0.666 (0.187–2.367)	(3.244, 0.0%; P = 0.99)	(Begg’s test P = 0.48; Egger’s test P = 0.11)	Fixed
3DL3	8	1042/1048	1043/1047	0.523	0.666 (0.187–2.367)	(3.244, 0.0%; P = 0.99)	(Begg’s test P = 0.48; Egger’s test P = 0.011)	Fixed
3DS1	9	634/1254	594/1319	0.112	1.364 (0.930–2.000)	(30.60, 73.86%; P = 0.0002)	(Begg’s test P = 0.78; Egger’s test P = 0.58)	Random
2DP1	5	678/704	789/814	0.604	0.857 (0.479–1.535)	(3.52, 43.30%; P = 0.171)	(Begg’s test P = 0.47; Egger’s test P = 0.33)	Fixed
3DP1	4	296/296	463/463	0.823	0.639 (0.012–32.32)	NA	NA	NA

NA, Not applicable;

Figure 2.

Forest plot. The plot illustrates the results of pooled OR for 2DL3, 2DS1, 2DS4, and 3DL1 KIR genes

Meta-regression analysis

The meta-regression analyses were conducted to find potential sources of heterogeneity among the included studies (Table 3). It was observed that publication year was a source of heterogeneity for both KIR2DL1 and KIR2DL3 genes (Figure 3).

Table 3.

Meta-regression analyses of the KIR genes based on the continent and publication year

KIR gene	Heterogenicity Factor	Co-efficient	SE	T	P value (95% CI)
2DL1	Continent	1.87	2.85	0.66	0.53 (−4.86, 8.62)
	Year of Publication	−1.77	0.43	−4.03	0.005 (−2.08, −0.73)
2DL2	Continent	−0.18	0.21	−0.85	0.42 (−0.68, 0.32)
	Year of Publication	0.01	0.06	0.27	0.79 (−0.13, 0.16)
2DL3	Continent	−3.11	4.21	−0.74	0.48 (−13.08, 6.85)
	Year of Publication	−2.74	0.58	−4.65	0.002 (−4.13, −1.34)
2DL4	Continent	−0.15	0.38	−0.39	0.71 (−1.09, 0.79)
	Year of Publication	−0.11	0.17	−0.64	0.54 (−0.55, 0.32)
2DL5	Continent	0.01	0.13	0.13	0.90 (−0.3, 0.33)
	Year of Publication	0.05	0.04	1.16	0.28 (−0.05, 0.15)
2DS1	Continent	−1.08	0.73	−1.47	0.18 (−2.82, 0.65)
	Year of Publication	0.05	0.23	0.24	0.81 (−0.48, 0.59)
2DS2	Continent	−0.20	0.19	−1.07	0.32 (−0.66, 0.25)
	Year of Publication	0.02	0.06	0.45	0.66 (−0.11, 0.17)
2DS3	Continent	−0.90	0.71	−1.27	0.24 (−2.59, 0.77)
	Year of Publication	−0.01	0.22	−0.05	0.95 (−0.53, 0.50)
2DS4	Continent	−2.01	1.74	−1.16	0.28 (−6.14, 2.10)
	Year of Publication	0.29	0.50	0.58	0.57 (−0.90, 1.50)
2DS5	Continent	−0.22	0.49	−0.44	0.67 (−1.39, 0.95)
	Year of Publication	0.08	0.13	0.65	0.53 (−0.23, 0.41)
3DL1	Continent	−1.72	1.02	−1.67	0.13 (−4.15, 0.71)
	Year of Publication	−0.20	0.33	−0.62	0.55 (−0.99, 0.58)
3DL2	Continent	−0.25	0.59	−0.44	0.67 (−1.70, 1.19)
	Year of Publication	−0.13	0.20	−0.66	0.53 (−0.64, 0.37)
3DL3	Continent	−0.25	0.59	−0.44	0.67 (−1.71, 1.19)
	Year of Publication	−0.13	0.20	−0.66	0.53 (−0.64, 0.37)
3DS1	Continent	−0.72	0.45	−1.61	0.15 (−1.79, 0.34)
	Year of Publication	0.04	0.18	0.25	0.80 (−0.39, 0.48)

KIR, Killer cell immunoglobulin-like receptor; CI, Confidence interval

Figure 3.

Funnel plot. The plot depicts the publication bias and heterogeneity between studies for 2DL3, 2DS1, 2DS4, and 3DL1 KIR genes

Heterogeneity and publication bias

Statistically significant inter-study heterogeneity (I² > 50%; P < 0.10) was detected in 2DL2 (I² = 63.66%; P = 0.0049), 2DL3 (I² = 68.16%; P = 0.0015), 2DS1 (I² = 89.04%; P < 0.0001), 2DS2 (I² = 63.05%; P = 0.0056), 2DS3 (I² = 85.42%; P < 0.0001), 2DS5 (I² = 76.16%; P < 0.0001), and 3DS1 (I² = 73.86%; P = 0.0002) genes. As a consequence, the random-effects model was exerted to survey the associations. However, the remaining KIR genes demonstrated no significant heterogeneity and, hence, the fixed-effects model was employed in pooled findings. We detected publication bias in 2DL4 (Egger’s test P = 0.021), 2DS1 (Begg’s test P = 0.04), 2DS4 (Begg’s test P = 0.04), and 3DL1 (Begg’s test P = 0.03; Table 2, Figure 4).

Figure 4.

Meta-regression results of the 2DL1 and 2DL3 KIR genes based on the publication year

Discussion

Digging through the literature, we identified 10 case-control association studies that evaluated the role of KIR genes in susceptibility to PTB infection. That notwithstanding, considering heterogeneous ethnicity, relatively small sample sizes, and weak statistical power in each individual study, there are inconsistent and inconclusive results about the role of KIR genes in association with susceptibility to PTB infection risk. Herein, we carried out the first systematic review and meta-analysis on 1357 cases and 1357 control subjects to conclude a clear and consistent approximation of the associations between KIR genes and susceptibility to PTB infection. Our analyses led to identification of a significant association between both activating and inhibitory KIRs and increased susceptibility to PTB infection.

KIR genes are present or absent in each individual, resulting in remarkable variation in the gene content among the individuals or populations. In addition, the KIR genes might be deleted and/or duplicated, conferring additional variation and developing a variable number of KIR gene copies [36]. On the other side, each KIR gene might harbor allelic single nucleotide polymorphisms (SNPs). The combination of presence/absence variation and allelic SNPs confers a high level of KIR heterogeneity among populations as well as individuals [37]. Linkage disequilibrium (LD) also occurs in the KIR genes. Since KIR genes are structured in tandem across a 150 kb region [38], LD is frequently occurred at both allelic variation as well as gene content [39]. Such complicated LD patterns might result in misleading interpretation of the association studies, since non-independent associations between KIR alleles or genes may lead to potential modulations in the gene function.

NK cells are large lymphocytes involved in the first line of defense against infections, such as TB [40]. These cells play a role in delivering cytotoxic granules, such as perforin and granzyme, to the cells infected with M. tuberculosis, resulting in death of the bacterium. In addition, in vitro experiments have reported that NK cells can kill monocytes infected with M. tuberculosis [41,42]. During infection of cells with M. tuberculosis, NK cells are able to trigger specific CD8⁺ T cell responses through either killing infected macrophages [43] or regulatory T (Treg) cells [44], and through promoting a cross-talk with dendritic cells (DCs) [40]. This implies to the critical role of NK cells in the stimulation of adaptive immune responses against TB infection. NK cell function is modulated via signaling of activating or inhibitory receptors in response to interactions with HLA ligands [45]. Basically, KIR-HLA interaction does not determine the fate of target cell to be killed or survived. Instead, an orchestrated response develops the lysis or inhibition of lysis of target cells. In some specific conditions, inhibitory signals are able to overcome the activating signals, since the inhibitory receptors bind with higher affinity, providing survival signals [46,47].

There are two haplotypes of KIR genes, including A and B that are distinguished based on variable presence of a 24 Kbp HindIII band, which later was indicated to originate from the KIR2DL5 gene [38,48]. In the haplotype A, there are KIR2DP1, KIR3DL3, KIR2DL3, KIR2DL1, KIR3DP1, KIR2DL4, KIR3DL1, KIR2DS4, and KIR3DL2 genes. The frequency of A haplotype has been estimated to be 47–59% among the Caucasians. The haplotype B contains more activating KIR genes, including KIR2DS1, KIR2DS2, KIR2DS3, and KIR2DS5, that has been reported to be highly varied. Both haplotypes contain three KIR genes, including KIR2DL4, KIR3DL2, and KIR3DL3, which are collectively called as framework loci due to pretense in all haplotypes [49]. In all haplotypes, the framework KIR3DL3 gene is located at the centromeric end, while KIR3DL2 is found at the telomeric end and KIR2DL4 is placed in the middle [50]. Several studies have implied to the role of KIR haplotypes, instead of single KIR genes, in association with diseases [51,52]. In the current meta-analysis, case-control studies evaluating the presence or absence of single specific KIR gene (regardless of the variations occurred in each gene) were included. We detected the separate association of KIR2DL3, KIR2DS1, KIR2DS4, and KIR3DL1 with PTB infection risk. The KIR2DL3, KIR2DS4, and KIR3DL1 gens are structured in the haplotype A and show LD with each other. Hence, interpretation of involvement of these KIR genes in association with PTB risk requires further analysis, in case of sufficient data available.

A bulk of studies has assessed the contribution of KIR genes to the pathogenesis of several autoimmune diseases [53–56]. Nonetheless, there is paucity of reports with respect to the involvement of KIR genes in susceptibility to infectious diseases. A failure in clearing the bacteria in PTB patients stems probably from reduced NK cells activity, which may be due to genetically determined KIR/HLA ligand gene content that results in a predominant inhibitory signals over activating ones [57].

The meta-analysis performed here on 1357 cases and 1357 control subjects led to identification of four significantly associated KIR genes, including inhibitory 2DL3 and 3DL1 and activating 2DS1 and 2DS4, with susceptibility to risk of PTB infection. An imbalance in the distribution of activating and inhibitory KIRs may impress the activation of NK cells, hence conferring an increased susceptibility to risk of PTB infection. However, we identified two activating and two inhibitory KIR genes in association with risk of PTB infection. It seems that a complicated network of KIR gene signaling as well as the expression rate of these molecules on NK cells may determine the final fate of NK cell activation or inhibition. Furthermore, the meta-regression analysis revealed that publication year might be source of heterogeneity for KIR2DL3. Hence the association of 2DL3 should be interpreted cautiously.

Here we identified some KIR genes with significantly different distribution between PTB infected cases and healthy controls. That notwithstanding, this study does have some limitations and caveats because of the low number of publications having inclusion criteria. This issue may interrupt obtaining a conclusive, comprehensive, consistent, and decisive understanding of the KIR gene association with susceptibility to risk of PTB infection.

In consideration of all, this was the first meta-analysis of 10 case-control association studies (comprising 1357 cases and 1357 control subjects) that assessed the role of KIR genes in susceptibility to risk of PTB infection. Our analysis demonstrated significant associations between inhibitory 2DL3 and 3DL1 and activating 2DS1 and 2DS4 KIR genes in susceptibility to risk of TB infection. As there were limited number of studies to include in our analysis at the moment, we recommend to perform meta-analysis in the future after approaching with more original data in different populations. Such analysis in the future may contribute to clear understanding of the KIR genes roles in susceptibility to risk of PTB infection.

Funding Statement

The authors have no funding to report.

Ethics approval and consent to participate

Not applicable.

Consent to publish

All authors read the manuscript and consent for its publication.

Availability of data and materials

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Disclosure statement

No potential conflict of interest was reported by the authors.