KIR/HLA Gene Profile Implication in Systemic Sclerosis Patients from Mexico.

INTRODUCTION: Systemic Sclerosis (SSc) is an autoimmune, inflammatory, and multisystemic disease characterized by the presence of autoantibodies and fibrosis. The pathogenesis involves the interaction between immune system cells such as macrophages, NK cells, T cells, and B cells. Killer-cell Immunoglobulin-like Receptors (KIR) are expressed in NK cells and some T cell subsets that recognize HLA class I molecules as ligands and are involved in regulating the activation and inhibition of these cells. The KIR family consists of 14 genes and two pseudogenes; according to the gene content, the genotype could be AA and Bx. The aim of this study was to evaluate the association between KIR/HLA genes and genotypes with SSc and the clinical characteristics.
METHODS: We included 50 SSc patients and 90 Control Subjects (CS). Genotyping of KIR, HLA-C, -Bw4, and -A ∗ 03/ ∗ 11 was made by SSP-PCR.
RESULTS: In SSc patients, a higher frequency of KIR2DL2 (p = 0.0007, p' = 0.011), KIR2DS4del (p = 0.001, p' = 0.021), and HLA-C2 (p = 0.02, p' = 0.09) was found. This is the first study to evaluate the frequency of HLA-A ∗ 03/ ∗ 11 in SSc patients, of which a low frequency was found in both groups. Compound genotypes KIR2DL2+/HLA-C1+ or KIR2DL2+/HLA-C2+ have a higher frequency in SSc patients. The Bx genotype was the most frequent and was associated with risk to SSc (p = 0.007, OR = 3.1, 95% CI = 1.4-7.9, p' = 0.014). The genotypes with a higher iKIR number than aKIR (iKIR > aKIR) were found in all individuals; genotypes with 7-8 iKIR genes were increased in SSc patients. We do not find an association between the KIR genes with the clinical characteristics.
CONCLUSION: The results suggest that KIR2DL2 and 2DS4del could have a risk role in the development of SSc, but not with clinical manifestations.

1. Introduction

Systemic Sclerosis (SSc) is a chronic and autoimmune disease, characterized by vascular dysfunction and damage, alteration in angiogenesis, inflammation, cytokine release, and fibrosis. The damage in the endothelial cells can be mediated by autoantibodies, viruses, oxidative stress, or by granzymes and perforins. It mainly affects the skin, gastrointestinal tract, lungs, and kidneys. According to the extent of skin fibrosis, SSc can be classified as limited (lcSSc) or diffuse (dcSSc) [1]. The triggering agents are not entirely established; however, environmental (silica, asbestos, organic solvents, and industrial emissions), infectious (mainly Cytomegalovirus and Epstein Barr virus infections), and genetic factors (mainly HLA class I and II alleles) are commonly involved in the susceptibility and disease development [2, 3].

Natural Killer (NK) cells play an important role in the innate and adaptive immune response by the secretion of cytokines and chemokines or by cytotoxicity. NK cells, in addition to T cells, have been implicated in different autoimmune processes [4]. The functions of NK cells and some subpopulations of T cells are regulated by membrane receptors such as the Killer-cell Immunoglobulin-like Receptors (KIR), which recognize as ligands HLA (Human Leukocyte Antigen) class I molecules (-A, -B, -C, -F, -G). The KIR extracellular region can have two (KIR2D) or three (KIR3D) domains, and the intracytoplasmic tail can be long tailed (L; e.g., KIR2DL and KIR3DL) with an inhibitory function or short tailed (S; e.g., KIR2DS and KIR3DS) with an activating function. The KIR/HLA-I interaction occurs between the extracellular domains and the peptide residues of the HLA-I molecule. Its affinity is influenced by the amino acid positions [5]. The HLA-C KIR ligands are classified into the -C1 (position 77 has a serine and position 80 an asparagine) ligand of KIR2DL2, 2DL3, 2DS2, and 2DS4 and the -C2 (position 77 has an asparagine and position 80 a lysine) ligand of KIR2DL1, 2DL2, 2DL3, 2DS1, 2DS4, and 2DS5. HLA-B alleles are classified into -Bw4 and -Bw6 based on the amino acid positions 77-83. The products of KIR3DL1 gene variants are associated with HLA alleles with -Bw4 motifs (HLA-A, -Bw4, -Bw4). With regard to HLA-A, it has mainly been described that HLA-A∗03 and -A∗11 are ligands of KIR3DL2 and, in addition, HLA-A∗11 is a ligand of KIR2DS4 [5, 6].

KIR genes are located in the leukocyte cluster receptor, at chromosome 19q.13.4. This family includes genes encoding seven inhibitory receptors (KIR2DL1, 2DL2, 2DL3, 3DL1, 3DL2, 3DL3, and 2DL5), six activating receptors (2DS1, 2DS2, 2DS3, 2DS4, 2DS5, and 3DS1), one receptor with both functions (2DL4), and two pseudogenes (2DP1 and 3DP1). KIR3DL2, 3DL3, 2DL4, and 3DP1 are considered framework genes and usually are present in all individuals [5, 7]. The KIR haplotype is defined, depending on the gene content, as A or B. The B haplotype is characterized by the presence of at least one of its exclusive genes, KIR2DL5, 2DS1, 2DS2, 2DS3, 2DS5, or 3DS1. The A haplotype is identified by the absence of all exclusive KIR genes from the B haplotype. In addition, this characteristic is used to define the AA genotype in subjects. Because we cannot distinguish between AB and BB genotypes, any of these are called Bx [8].

Studies in different populations have reported associations of KIR genes and KIR/HLA-I genotypes with SSc susceptibility: KIR2DS2+/2DL2- in Germany [9]; KIR2DS1+/2DS2- and 2DS1+/HLA-C2+ in Canada [10]; 2DS3+, 2DS2+HLA/-C1+, and 2DL2+/HLA-C1+ in Turkey [11]; and 3DL1+/HLA-Bw4 in Iran [12]. On the other hand, other studies have shown the association of KIR genes and KIR/HLA-I genotypes with lower risk to SSc: 2DL2+ in Brazil [13] and 2DL3+ in Turkey [11]. In the Mexican population, there are only a few studies performed in SSc patients and the role of KIR in SSc has not been evaluated. Therefore, the aim of our study was to evaluate the association between KIR and HLA-I genes with SSc and the clinical manifestations in patients from Guerrero state (southern Mexico).

2. Material and Methods

2.1. Studied Populations

We included 50 patients with SSc and 90 Control Subjects (CS) without familiar autoimmune antecedents. A rheumatologist evaluated all patients and classified for SSc according to the 2013 American College of Rheumatology and The European League Against Rheumatism (ACR/EULAR) criteria. The patients were recruited by continuing inclusion during January 2016 to June 2017, in the Hospital General de Chilpancingo Dr. Raymundo Abarca Alarcón, Guerrero, México. The patients were matched with CS in a 1 : 2 ratio, considering age and gender. All subjects were from Guerrero state (southern Mexico), at least until the third generation. Due to the genetic heterogeneity of the Mexican population previously demonstrated in KIR and HLA genes [14, 15], we did not include patients from other regions to conserve the ethnic homogeneity and to avoid, as much as possible, this influence in our results. The study was part of a protocol about genetic associations in Mexicans, and informed consent was obtained from each individual. This project was conducted according to the Helsinki Declaration and approved by the Committee of Biosecurity and Ethics of the Centro Universitario de Ciencias de la Salud at the Universidad de Guadalajara and at the Universidad Autónoma de Guerrero, México.

2.2. KIR and HLA Genotyping

Genomic DNA was obtained from peripheral blood, using EDTA as an anticoagulant. For the DNA extraction, we utilized the modified technique of Miller et al. [16]. The DNA samples were quantified in a NanoDrop (Thermo Fisher Scientific, USA) at 260/280 nm wavelength, and work dilutions were settled at 100 ng/μL.

The identification of KIR genes (2DL1, 2DL2, 2DL3, 3DL1, 3DL2, 2DS1, 2DS2, 2DS3, 2DS4, 2DS5, 3DS1, 2DP1, 3DL3, 2DL4, 2DL5, and 3DP1) was made by Sequence Specific Primer Polymerase Chain Reaction (SSP-PCR). The primers and conditions were established according to the method described by Vilches et al. [17]. The unique KIR genotype profile was analyzed and confirmed in duplicate. For KIR2DS4, we identify the full gene and del variant, which was considered for the group assignment. HLA-C (-C1 and -C2) typing was made by SSP-PCR using the methodology established by Hiby et al., [18]. HLA-Bw4 (-A, -Bw4, and -Bw4) were identified by SSP-PCR using primers described by Tajik et al. [19] and the conditions described by Omar et al. [20]. We used 100 ng of DNA in each reaction. The electrophoresis of KIR and HLA-C PCR products were done in 3.0% agarose gel (Vivantis®) and the HLA-Bw4 PCR products in 2.0% agarose gel (Vivantis®). All gels were prepared with Ultrapure 0.5x TBE buffer (Life Technologies®) and stained with SYBR Green Safe (Invitrogen®). A Kodak® Gel Logic 112 imaging system was used for visualization and photography. Genotyping of HLA-A∗03 and -A∗11 was made by SSP-PCR, according to Kurz et. al., [21]. With this strategy, alleles of both allelic groups are coamplified from two PCR mixes and the products visualized in a 6% polyacrylamide gel stained with silver nitrate. Subsequently, to discriminate between HLA-A∗03 and -A∗11, samples positive for HLA-A∗03/-A∗11 were purified and genotyped by sequence-based typing (SBT) of a fragment containing exons 2 and 3 of the HLA-A gene. Sequencing was performed by using the Dye Terminator Cycle Sequencing (DTCS) Quick Start Kit (Beckman Coulter, Fullerton, CA) and, after dye terminator washing, samples were sequenced in a Genetic Analyzer Beckman Coulter CEQ8800 instrument according to the manufacturer's recommendations. For allele assignment, all sequences were analyzed in the Chromas Software Version 2.4.6 and online blasted with those reported in the IMGT-HLA database.

2.3. Statistical Analysis

The carrier frequency (CF) estimation of each KIR and HLA gene was done using direct counting. The KIR and HLA gene frequency (GF) was determined by Bernstein's formula (using CF) [22]. The differences in clinical parameters, KIR/HLA genes, and genotype frequencies were analyzed by Fisher's exact test and adjusted p′ with Holm-Bonferroni correction for multiple comparisons. Hardy-Weinberg equilibrium was evaluated in both study groups according to the KIR genotype profile. Comparisons between KIR and HLA genes with the clinical characteristics and clinical evaluation scales were done with multiple comparisons, and each group was analyzed by Fisher's exact test. GraphPad Prism Software Version 6 was used for all analyses and the statistical power was calculated using the program Power and Sample Size Calculation version 3.1.6 [23]. The results were considered significant when p < 0.05, and the statistical power was between 82 and 100%. The linkage disequilibrium (LD) in KIR genes was analyzed using Cramer's V statistic [22], which was calculated from the contingency table of presence/absence, and we refer to this statistic as . Wn∗ value was calculated only in the KIR genes associated with the disease. Chi square was used to test the association in pairs of genes, and p < 0.05 was considered statistically significant.

3. Results

3.1. Demographic and Clinical Features

A total of 50 SSc patients were included, 44 female (88%) and six male (12%). 14% of the patients have dcSSc (all women), whereas 86% of the patients have lcSSc. SSc patients have a median age of 48.7 years at the inclusion time (Table 1). Regarding the risk factor exposure, the majority of the SSc patients were exposed to wood smoke (78%; p < 0.0001, OR = 6.8, 95% CI = 3.0–14.9) mainly from cooking; they were also exposed to fertilizers (38%), organic solvents (12%), silica (8%), and asbestos (2%). The more frequent clinical manifestations are shown in Table 1. Concerning the clinical evaluation, at the inclusion time, the median of HAQ (Health Assessment Questionnaire disability index) was 0.4 and median of MRSS51 (Modified Rodnan Skin Score) was 7.3 (Table 1).

Table 1

Demographic and clinical characteristics of SSc patients and CS.

	SSc(%) (n = 50)	CS(%) (n = 90)	p
Age (years)†	48.7 (18-75)	48.3 (20-73)	0.9∗
Gender
Female	88 (44)	94.4 (85)	0.2∗∗
Male	12 (6)	5.5 (5)	0.2∗∗
Disease subtype
lcSSc	86 (43)	—
dcSSc	14 (7)	—
Disease duration (months)†	94.8 (3-420)	—
Clinical manifestations
Sclerodactyly	88 (44)	—
Musculoskeletal damage	82 (41)	—
Raynaud's phenomenon	70 (35)	—
Puffy fingers	68 (34)	—
Telangiectasia	54 (27)	—
Digital ulcers	40 (20)
Clinical evaluation
HAQ†	0.4 (0-1.29)	—
MRSS51†	7.3 (0-30)	—

†Data are shown in median and rank in parentheses. SSc: Systemic Sclerosis; CS: Control Subjects; HAQ: Health Assessment Questionnaire disability index; MRSS51: Modified Rodnan Skin Score. ∗Mann–Whitney U test; ∗∗Fisher's test.

3.2. KIR and HLA Genes

The distribution of KIR and HLA class I genes in SSc patients and CS is shown in Table 2. The KIR2DL2 gene frequency was higher in SSc patients compared with CS (p = 0.0007, OR = 3.6, 95% CI = 1.7–7.3, p′ = 0.011). According to the methodology used, we were able to distinguish the KIR2DS4 gene variants: 2DS4full and 2DS4del. Taking into consideration whether individuals have or do not have the full or del gene variant, four groups were defined: (a) the individuals with both variants (full/del), (b) those with only the full gene (full), (c) those with only the del variant (del), and (d) those who are negative for KIR2DS4 (neg). The KIR2DS4full frequency was lower in SSc patients than in CS (p = 0.02, OR = 0.3, 95% CI = 0.1–0.7, p′ = 0.09), whereas the 2DS4del frequency was higher in SSc patients than in CS (p = 0.001, OR = 7.4, 95% CI = 1.6–35.2, p′ = 0.021). Individuals with full/del and the KIR2DS4neg have similar frequencies in both groups (Table 2). Concerning HLA class I gene distribution, HLA-A and HLA-C2 were more frequent in SSc patients than in CS (Table 2) without significant differences. Regarding HLA-A, we found 7 SSc patients (14%) and 12 CS (13.3%) positive to -A∗03/∗11. In SSc patients, 8% are positive to -A∗03 and 6% to -A∗11. In CS, 11.1% are positive to -A∗03 and 1.1% to -A∗11 without significant difference (Table 2).

Table 2

KIR and HLA gene frequencies in SSc patients and CS.

	SSc (n = 50)			CS (n = 90)			p ∗	OR	95% CI	p′
	n CF	% CF	GF	n CF	% CF	GF
KIR2DL1	47	94	0.76	87	96.7	0.82
KIR2DL2	32	64	0.40	30	33.3	0.18	0.0007	3.6	1.7-7.3	0.011
KIR2DL3	46	92	0.72	87	96.7	0.82
KIR3DL1	44	88	0.65	81	90	0.68
KIR3DL2	50	100	1.00	88	97.8	0.85
KIR2DS1	23	46	0.27	42	46.7	0.27
KIR2DS2	18	36	0.20	28	31.1	0.17
KIR2DS3	8	16	0.08	14	15.6	0.08
KIR2DS4	41	82	0.58	80	88.9	0.67
2DS4full/del	12	24	0.13	25	27.8	0.15
2DS4full	18	36	0.20	51	56.7	0.34	0.02	0.4	0.2-0.9	0.09
2DS4del	12	24	0.13	4	4.4	0.02	0.001	6.8	2.1-22.4	0.021
2DS4neg	8	16	0.08	10	11.1	0.06
KIR2DS5	20	40	0.23	37	41.1	0.23
KIR3DS1	24	48	0.28	39	43.3	0.25
KIR2DP1	46	92	0.72	87	96.7	0.82
KIR3DL3	50	100	1.00	90	100	1.00
KIR2DL4	50	100	1.00	90	100	1.00
KIR2DL5	26	52	0.31	46	51	0.30
KIR3DP1	49	98	0.86	90	100	1.00
HLA-C1	41	82	0.58	82	91	0.70
HLA-C2	37	74	0.49	48	53	0.31	0.02	2.5	1.2-5.3	0.09
HLA-ABw4	48	96	0.80	75	83	0.59
HLA-Bw4Iso80	11	22	0.12	23	26	0.14
HLA-Bw4Thr80	5	10	0.05	7	8	0.04
HLA-A ∗ 03	4	8	0.04	10	11.1	0.06
HLA-A ∗ 11	3	6	0.03	1	1.1	0.005

CF: Carrier Frequencies; GF: Gene Frequencies; KIR: Killer-cell Immunoglobulin-like Receptors; HLA: Human Leukocyte Antigen; SSc: Systemic Sclerosis; CS: Control Subjects; CI: Confidence Interval. KIR2DS4full/del includes individuals with a copy of the full gene and del variant; 2DS4full includes individuals with one or two copies of the full gene; 2DS4del includes individuals with one or two copies of the del variant; and 2DS4neg includes individuals without KIR2DS4. p∗ based on Fisher's exact test and adjusted p′ based on Holm-Bonferroni correction for multiple comparisons.

3.3. Combinations of KIR Genes and KIR/HLA Genotypes

The KIR gene combinations involve the presence/absence of the genes previously associated with SSc in this study and in previous studies (2DL2, 2DS4, 2DS2, and 2DS1). Analysis shows an increased frequency of 2DL2+/2DS4del+ in SSc patients compared with CS. In SSc patients, a low frequency of 2DL2-/2DS4del- (p = 0.0005, OR = 0.2, 95% CI = 0.1–0.6, p′ = 0.005), 2DL2-/2DS4full+/2DS4del- (p = 0.0003, OR = 0.2, 95% CI = 0.08–0.5, p′ = 0.003), 2DL2+/2DS2- (p < 0.0001, OR = 17.1, 95% CI = 3.7–79.2, p′ = 0.0009), and 2DL2-/2DS2- (p = 0.0007, OR = 0.3, 95% CI = 0.1–0.6, p′ = 0.006) was found. The other KIR gene combinations do not show significant differences between both groups. Due to the differences observed in the frequencies of KIR2DL2 and KIR2DS2 in the SSc patients (Table 2) and the preferential association between these genes reported in other investigations [24], we calculated the LD of these genes. In CS, a strong LD is observed (Wn∗ = 0.9540, p < 0.0001), while in SSc patients a decreased LD is observed (Wn∗ = 0.5625, p < 0.0001).

Taking into consideration the presence of each of the KIR genes with their specific HLA-I ligand gene, the compound KIR/HLA genotypes were evaluated. The comparison of the compound KIR/HLA genotypes of SSc patients and CS with a statistically significant difference is shown in Table 3. In SSc patients, the frequency of the compound genotypes 2DL2+/HLA-C2+ was increased (p = 0.01, OR = 2.8, 95% CI = 1.3–5.8, p′ = 0.045). On the other hand, the frequency of 2DL2-/C1+ (p = 0.001, OR = 0.3, 95% CI = 0.1–0.6, p′ = 0.004), 2DL2-/HLA-C2- (p < 0.0001, OR = 0.1, 95% CI = 0.03–0.4, p′ = 0.0004), 2DS4full+/HLA-C1+ (p = 0.003, OR = 0.3, 95% CI = 0.2–0.7, p′ = 0.01), and 2DS4del-/HLA-C2- (p = 0.0005, OR = 0.1, 95% CI = 0.04–0.5, p′ = 0.002) was lower in SSc patients (Table 3). The other KIR/HLA gene combinations that involve the presence of HLA-A, -Bw4, -Bw4, and -A∗03/∗11 and their respective KIR did not show a significant difference between both groups.

Table 3

Combinations between the KIR gene and the HLA-I ligand.

KIR/HLA	SSc(%) (n = 50)	CS(%) (n = 90)	p ∗	OR	95% CI	p′
KIR2DL2+/HLA-C1+	54 (27)	32.2 (29)	0.02	2.5	1.2–5.0	0.08
KIR2DL2+/HLA-C2+	44 (44)	22.2 (20)	0.01	2.8	1.3–5.8	0.045
KIR2DS4full+/HLA-C1+	50 (25)	75.6 (68)	0.003	0.3	0.2–0.7	0.01
KIR2DL2-/HLA-C1+	28 (14)	58.9 (53)	0.001	0.3	0.1–0.6	0.004
KIR2DL2-/HLA-C2-	6 (3)	35.6 (32)	< 0.0001	0.1	0.03–0.4	0.0004
KIR2DS4del-/HLA-C2-	6 (3)	31.1 (28)	0.0005	0.1	0.04–0.5	0.002

KIR: Killer-cell Immunoglobulin-like Receptors; HLA: Human Leukocyte Antigen; SSc: Systemic Sclerosis; CS: Control Subjects. p∗ based on Fisher's exact test and adjusted p′ based on Holm-Bonferroni correction for multiple comparisons.

3.4. KIR Genotypes

In this study, 35 genotypes were found in all individuals, including 24 genotypes in SSc and 24 genotypes in CS. Eleven genotypes were exclusive to the SSc patients, 11 were exclusive to the CS, and 13 were shared between both groups (Table 4). A novel genotype (not reported in the Allele Frequency Database [8]) was found in an SSc patient. The AA genotype was less frequent in SSc patients (16%) than in CS (38.9%), whereas the Bx genotype was more frequent in SSc patients (84%) than in CS (61.1%) (p = 0.007, OR = 3.1, 95% CI = 1.4–7.9, p′ = 0.014). In particular, the frequency of the genotype with ID 1 (as reported in the Allele Frequency Database [8]) was lower in SSc patients (14%) than in CS (37.8%) (p = 0.03, OR = 0.3, 95% CI = 0.1–0.7, p′ = 0.02). The genotype with ID 19 (as reported in the Allele Frequency Database [8]) was found only in patients (14%) (Table 4).

Table 4

KIR genotypes.

Genotype ID		3DL1	2DL1	2DL3	2DS4	2DL2	2DL5	3DS1	2DS1	2DS2	2DS3	2DS5	2DL4	3DL2	3DL3	2DP1	3DP1	SSc (%)(n = 50)	CS (%)(n = 90)
AA	1	—	—	—	—								—	—	—	—	—	14.0	37.8∗
AA	195	—	—	—									—	—	—	—	—	0.0	1.1
AA	203	—	—	—	—								—	—	—		—	2.0	0.0
Bx	2	—	—	—	—		—	—	—			—	—	—	—	—	—	10.0	18.9
Bx	3	—	—	—	—	—	—	—	—	—		—	—	—	—	—	—	4.0	5.6
Bx	4	—	—	—	—	—				—			—	—	—	—	—	8.0	5.6
Bx	5	—	—	—	—	—	—			—	—		—	—	—	—	—	4.0	3.3
Bx	6	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	0.0	1.1
Bx	7	—	—	—	—	—	—	—	—	—	—		—	—	—	—	—	0.0	2.2
Bx	8	—	—	—	—		—	—	—		—		—	—	—	—	—	2.0	3.3
Bx	9	—	—	—	—	—	—		—			—	—	—	—	—	—	2.0	3.3
Bx	18	—	—	—	—	—	—	—	—			—	—	—	—	—	—	8.0	1.1
Bx	19	—	—	—	—	—							—	—	—	—	—	14.0	0.0
Bx	31	—	—	—	—	—	—			—			—	—	—	—	—	2.0	1.1
Bx	58	—	—	—	—	—	—	—	—		—	—	—	—	—	—	—	2.0	0.0
Bx	68		—	—		—	—	—	—	—		—	—	—	—	—	—	0.0	1.1
Bx	69		—	—			—	—	—			—	—	—	—	—	—	2.0	2.2
Bx	70		—	—		—	—	—	—	—	—	—	—	—	—	—	—	2.0	0.0
Bx	71	—	—		—	—	—			—	—		—	—	—	—	—	2.0	1.1
Bx	72	—			—	—				—			—	—	—		—	0.0	1.1
Bx	74					—	—	—	—	—		—	—	—	—		—	2.0	1.1
Bx	75		—	—			—	—	—		—	—	—	—	—	—	—	0.0	2.2
Bx	76	—			—	—	—	—	—	—		—	—	—	—		—	2.0	0.0
Bx	88		—	—		—	—	—	—	—		—	—	—	—	—	—	6.0	1.1
Bx	117		—	—			—	—	—		—		—	—	—	—	—	0.0	1.1
Bx	275	—	—	—	—	—					—		—	—	—	—	—	2.0	0.0
Bx	331	—	—	—			—	—	—		—		—	—	—	—	—	2.0	0.0
Bx	336	—	—	—	—	—		—					—	—	—	—	—	0.0	1.1
Bx	384	—	—	—	—	—	—	—	—				—	—	—	—	—	2.0	0.0
Bx	475	—	—	—				—					—	—	—	—	—	2.0	0.0
Bx	476	—		—	—			—					—	—	—	—	—	2.0	0.0
Bx	620	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—		0.0	1.1
Bx	651	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—		0.0	1.1
Bx	653	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—		0.0	1.1
Bx	No ID	—			—	—				—			—	—	—			2.0	0.0

Genotype ID was assigned according to the Allele Frequency Database [8]. Gene presence is indicated by a “—”, and its absence is indicated with empty spaces. The frequency of each genotype is presented in percentage. SSc: Systemic Sclerosis; CS: Control Subjects; AA: AA genotype; Bx: BB and AB genotypes. ∗ is based on Fisher's exact test (p = 0.03) and adjusted p′ based on Holm-Bonferroni correction for multiple comparisons (p′ = 0.02).

According to the genotype, the number of activating/inhibitory KIR genes was evaluated (Figure 1). The percentage of genotypes with 2, 3, and 5 activating KIR genes (aKIR) were higher in SSc patients than in CS, whereas genotypes with 1 and 4 aKIR genes have a lower percentage in SSc patients (Figure 1(a)), without showing statistical significance. In addition, the genotypes with 7 and 8 iKIR genes (iKIR) were more frequent in SSc patients (Figure 1(b)); however, it was not statistically significant. The percentage of individuals with 6 iKIR genes was lower in SSc patients than in CS (p = 0.002, OR = 0.3, 95% CI = 0.1–0.6, p′ = 0.008).

Figure 1

Distribution of activating/inhibitory KIR gene number between SSc patients and CS. (a) The number of activating KIR (aKIR) gene number percentage in SSc patients (gray circle) and CS (white triangles) is shown. (b) The number of inhibitory KIR (iKIR) gene number percentage in SSc patients (gray circle) and CS (white triangles) is shown. (c) The genotypes with higher iKIR than aKIR (iKIR > aKIR) is indicated according to the number of iKIR and aKIR genes present. SSc: Systemic Sclerosis; CS: Control Subjects; ∗p = 0.002 based on Fisher's exact test and p′ = 0.008, based on Holm-Bonferroni correction for multiple comparisons.

All the genotypes found in our studied groups have higher iKIR genes than aKIR genes (iKIR > aKIR). We established two groups according to the aKIR gene number: 0-3 aKIR and 4-6 aKIR. The frequency of individuals with 0-3 aKIR and 3-6 aKIR does not show a significant difference between SSc patients and CS (Figure 1(c)). We established two groups according to the iKIR gene number: 5-6 iKIR and 7-8 iKIR. The frequency of individuals with 7-8 iKIR genes was higher in SSc patients than in CS (p = 0.01; OR = 2.7; 95% CI = 1.3–5.8, p′ = 0.02), and the percentage of individuals with 5-6 iKIR genes was lower in SSc patients than in CS (p = 0.01; OR = 0.3; 95% CI = 0.2–0.8, p′ = 0.02).

3.5. KIR and HLA Genetic Associations with the Presence of Clinical Manifestations

The KIR genes associated with the disease (KIR2DL2 and 2DS4del) were analyzed with the following clinical manifestations: sclerodactyly, musculoskeletal damage, Raynaud's phenomenon, inflamed fingers, telangiectasia, digital ulcers, calcinosis, esophageal dysfunction, interstitial lung disease, and pulmonary arterial hypertension. In almost all clinical manifestations, we found a higher frequency of 2DL2 than the other genes (Table 5); however, there was no significant difference, probably due to the number of patients. In addition, our evaluation shows that the presence of HLA, KIR/HLA genotypes, and iKIR/aKIR number does not have an association with the clinical manifestations, clinical evaluation (HAQ levels and MRSS51 score), and exposure to environmental risk factors (silica, organic solvents, fertilizers or, wood smoke).

Table 5

Clinical manifestations and KIR genes.

	Sclerodactyly (%) (n = 44)	Musculoskeletal damage(%) (n = 41)	Raynaud's phenomenon(%) (n = 35)	Inflamed fingers(%) (n = 34)	Telangiectasias (%) (n = 27)	Digital ulcers(%) (n = 20)
2DL2	63.6 (28)	58.5 (24)	60.0 (21)	55.9 (19)	59.3 (16)	70.0 (14)
2DS4full	34.1 (15)	31.7 (13)	37.1 (13)	35.3 (12)	44.4 (12)	45.0 (9)
2DS4del	25.0 (11)	24.4 (10)	31.4 (11)	29.4 (10)	22.2 (6)	45.0 (4)
2DS4full/del	22.7 (10)	26.8 (11)	17.1 (6)	17.6 (6)	22.2 (6)	45.0 (4)
2DS4neg	18.2 (8)	17.1 (7)	14.3 (5)	17.6 (6)	11.1 (3)	15.0 (3)

KIR2DS4full/del includes individuals with a copy of the full gene and del variant; 2DS4full includes individuals with one or two copies of the full gene; 2DS4del includes individuals with one or two copies of the del variant; and 2DS4neg includes individuals without KIR2DS4.

4. Discussion

Systemic Sclerosis (SSc) is an autoimmune disease more frequent in women than in men, according to the EUSTAR (EULAR Scleroderma Trials and Research group) records which showed a 5-10 : 1 ratio depending on the population. A female predominance was found in our studied patients (7.3 : 1), which is within the range of the EUSTAR records. In patients from Mexico City, there was also a higher prevalence of women with SSc; however, there was a dissimilar proportion compared to that of our study (9 : 1) and the majority of our patients have lcSSc, as previously observed in patients from Mexico City [2, 25]. Concerning the environmental factors, silica exposure was possibly involved with construction-related occupations. The organic solvent was suspected in patients with occupational exposure to cleaning products and trichloroethylene derivatives or in those working in the paint industry [26].

In this study, most of the patients live in rural areas and exposure to wood smoke is a significant risk factor. The exposed subjects are 6.8 times more likely to develop the disease compared to those who are not exposed. It is currently estimated that people exposed to wood smoke are mainly from the rural areas. Until now, welding smoke and industrial emission particles have been associated with risk to SSc development; however, the role of wood smoke in SSc susceptibility has not been described [2, 3]. Previous studies have shown that wood smoke exposure can lead to NF-κB overactivation; increased serum levels of TNF-α, IFN-γ, IL-1β, IL-18, IL-6, IL-21, and CCL2; and higher numbers of neutrophils, monocytes, and lymphocytes; as well as lower expression of MMP-9 and MMP-12. This mechanism induces chronic systemic inflammation and fibrosis. In addition, wood smoke produces oxidative damage, which could affect proteins and lipids (lipid peroxidation) or induce DNA damage, favoring autoantibody production [26-31]. In this study, we did not evaluate the autoantibody positivity; however, future studies could evaluate the relationship between wood smoke and the presence of autoantibodies in SSc patients.

In SSc immunopathology, various immune cells are involved; with regard to the NK cell, there are discrepant reports concerning the number and functions, which are probably related to the disease stage (inflammatory, fibrotic, or atrophic) and different clinical complications (CREST syndrome, skin fibrosis extension, vascular involvement, and lung damage). The decrease of NK cell numbers in peripheral blood is attributed to the infiltration in affected tissues; nevertheless, others studies report an increased NK cell number in dsSSc. In any of the cases, NK cells from SSc patients exhibited phenotypes characterized by altered cytokine production, diminished natural cytotoxicity, and decreased granzyme B release [32, 33]. The NK cell functions are controlled by inhibitory and activating receptor-ligand interactions. The signals from inhibitory receptors and HLA ligands are considered a mechanism for NK cell self-tolerance in a process named NK cell education, which is essential for tolerance to self-components. Strikingly, the NK cell licensing is reversible and the responsiveness of mature NK cells may adapt to environmental changes; for this reason, diverse studies have analyzed KIR genes in various autoimmune pathologies, such as psoriasis vulgaris and rheumatoid arthritis (RA) [7, 34].

This is the first study to evaluate the association of KIR/HLA genes and genotypes with SSc in the Mexican population. In this study, the framework genes (KIR3DL2, 3DL3, 2DL4, and 3DP1) and pseudogenes (2DP1 and 3DP1) were found in almost all individuals. KIR gene frequencies in CS were similar to a previous report from the Guerrero population [14]. In the patients studied, KIR2DL2 presence was associated with risk to SSc; in contrast, this gene has been reported with a lower risk of SSc development in Brazilian patients. However, in western Mexico, 2DL2 has also been associated with the risk for RA and response to treatment; this could suggest that in the Mexican population, 2DL2 is a gene associated with susceptibility to autoimmunity [13, 35].

Different studies have postulated that the increased expression of KIR2DL2 induces lower NK cell activation (mainly in viral infection), and the blockade of this receptor can favor cytotoxicity and cytokine secretion. KIR2DL2 could inhibit the signals from activating KIR or other activating receptors [36]. Thus, the NK cells of individuals with KIR2DL2 could have affected the antifibrotic function of the NK cells, characterized by the secretion of cytotoxic substances that induce cell lysis and the release of cytokines such as IFN-γ. It has been reported that IFN-γ production by NK cells inhibits liver fibrosis; therefore, altered NK responses could contribute to fibrogenesis [4, 32, 34]. In autoimmune diseases and particularly in SSc, the role of KIR2DL2 gene products is not fully established; however, it is possible that it is overexpressed and has high affinity with their ligands (HLA-C1 and -C2). This could be influential in the loss of the NK cells' antifibrotic effect and eventually contribute to fibrosis progression in SSc.

In addition, a strong LD between KIR2DS2 and KIR2DL2 in CS was found, while in SSc patients the LD was decreased. A possibility is that the patients have a KIR2DS2 allele that is not reported and that is not amplified with the primers used (must be confirmed by sequencing techniques). Moreover, NK cell receptors can only diversify through meiotic mutation and recombination processes; therefore, an incident during these processes affects the lineage distinctions and even disturbs the segregation of these receptors, as is described in other KIR gene combinations [37].

The KIR2DS4 protein can be encoded by 33 alleles, of which some code for a complete protein (2DS4full) and others for a truncated protein (2DS4del), for having the loss of 22 pb (mutations in exon 5 or 7). Human 2DS4del was initially called KIR1D, and its function is still unknown. However, the changes in the protein structure can affect their activating function [38]. In SSc patients, the 2DS4del presence was associated with risk to SSc, while 2DS4full was associated with reduced risk. In autoimmune diseases, 2DS4full was associated with a lower chance of response to methotrexate in RA patients [39]. In other medical conditions like a kidney transplant, KIR2DS4 gene variants are associated with the risk of rejection to kidney transplantation [40]. KIR2DS4full was associated with high survival after one year in patients with hematopoietic cell transplantation. Furthermore, in this group of patients 2DS4del was associated with risk to graft-versus-host disease [41] and also with susceptibility to syphilis in China [42]. It has been believed that KIR2DS4del is a nonfunctional protein because it is not anchored on the surface of the cell membrane. Nevertheless, it can be secreted in a soluble form and bind to its ligands (HLA-A∗11 and -C1/-C2) avoiding the recognition for other receptors or acting as a ligand to other receptors, which can trigger the activation of other types of cells that participate in the disease immunopathology [43].

In HLA class I genes, HLA-A has been previously associated with risk to SSc in Iranian subjects [12]. In a previous study in Mexican patients, the genes -C∗12:03/-B18:01 and -B∗08:01 were reported as a genetic risk to SSc, whereas -C∗07:02, -C∗07:02+/-B∗39:05, and -B∗39:06 were reported with reduced risk to dcSSc development [25]. In Brazilian patients, alleles of HLA-I have been associated not only with SSc development (-A∗30, -A∗32, -B∗57 and -Cw14) but also with medical complications, such as pulmonary fibrosis (-Cw∗0602) and pulmonary arterial hypertension (-C∗04) [44]. Thus, the presence of HLA class I genes may have a role in SSc susceptibility and development. None of the HLA evaluated in this research are associated with the SSc; therefore, future studies should evaluate other HLA groups in the Mexican population.

The combined genotypes associated with risk (2DL2+/2DS2-) or decreased risk of SSc (2DL2-/2DS4del-, 2DL2-/2DS4full+/2DS4del-, and 2DL2-/2DS2-), have not been reported in any other investigation. Combined genotypes associated with risk to SSc are 2DS2+/2DL2- in Germany and Brazil and 2DS1+/2DS2- in Canada [9, 13]. The analysis of KIR/HLA compound genotypes demonstrates that KIR2DL2+/HLA-C1+ and 2DL2+/-C2 are associated with risk in SSc patients of our population. The compound genotype 2DL2+/HLA-C1+ had already been associated with SSc in patients from Turkey and 2DS1+/HLA-C2+ in Germany [9, 11]. The other combinations found have not been associated with autoimmune pathologies. And 2DS4del+/HLA-C1+ and -C2+ have been associated with high transmission of HIV from Black South African mothers to children during pregnancy.

Regarding KIR genotypes, the Bx genotype was more frequent in SSc patients; in particular, the Bx genotype with ID 19 (according to the Allele Frequency Database [8]) was found only in SSc patients, whereas the AA genotype with ID 1 (according to the Allele Frequency Database [8]) was associated with lower risk to SSc development. Several investigations have confirmed that the AA genotype (predominantly inhibitor) has a certain protective effect to autoimmune diseases, such as RA, while the Bx genotype (predominantly activator) represents a risk to autoimmunity. Also, the Bx genotype with ID 19 was associated with risk for SSc in patients from Iran and risk for RA in patients from western Mexico[7, 12, 36]; however, we do not get a significant difference, probably due to the number of patients. A particularity of the genotype with ID 19 is that it is only distinguished from the genotype AA with ID 1 by the presence of KIR2DL2. It is important to consider that in addition to the KIR2DL2 association with the SSc, we confirm that the development of some clinical characteristics, such as sclerodactyly, musculoskeletal damage, inflamed fingers, and digital ulcers are more frequent in patients with 2DL2, in which the presence or absence of 2DS4 or any of its genetic variants was not involved. The involvement of KIR2DL2 in the development of various clinical characteristics is not defined; however, if the antifibrotic activity of the NK cells is inhibited, many other mechanisms could stimulate fibrosis and disease progression in conjunction with the various clinical complications. It is important to consider that we could not obtain a significant OR due to the small number of patients; future studies could evaluate the involvement of KIR2DL2 with the clinical characteristics of the disease in a larger group of SSc patients.

The influence of the activating/inhibitory KIR gene number in SSc susceptibility was demonstrated. The genotype with 7-8 iKIR genes was increased in SSc patients. The role of iKIR and aKIR gene number in autoimmune diseases is not well established; however, the effect of KIR gene number variation on NK cell education and the ability to respond to infections (mainly viral infections) is widely known. NK cell education via KIR is a process dependent on the interaction with HLA-I molecules, and for this reason, they must be analyzed altogether to estimate their potential involvement [45]. It is not clear how NK education can affect SSc immunopathology. However, the NK cell action depends on the balance between the signals from inhibitory and activating KIR receptors for different ligand specificities and the difference in the iKIR and aKIR numbers could affect its cytotoxic activity or cytokine secretion capacity.

In summary, this study provides a comprehensive assessment of the KIR and HLA genetic susceptibility with SSc in a southern Mexican mestizo population. The differences found in the KIR and KIR/HLA frequencies in our SSc patients in comparison to the SSc patients from other populations can be produced by genetic admixture and adaptation processes influenced by infectious agents and the environmental factors of the different regions and continents [7].

One limitation of this study is the low number of patients that we included; however, SSc has a low incidence rate. In addition, we consider preserving the genetic homogeneity more valuable than including patients from other regions because we previously demonstrated the genetic diversity of KIR in various Mexican regions [14]. Future studies should be done in the regions of other countries. Moreover, alternative techniques should be used to determine the specific alleles associated with the disease, evaluate KIR implication in the clinical manifestation in a larger group of SSc patients, and evaluate the expression of KIR in infiltrating cells in biopsies from SSc patients.