Figures
Abstract
The aim of this study was to analyse the distribution of KIR haplotypes and the KIR2DL2/3 alleles in chronic HCV-infected patients in order to establish the influence on the response to pegylated interferon plus ribavirin classical treatment. The alleles study of previously associated KIR2DL2/3 showed that KIR2DL2*001 was more frequent in non-SVR (NSVR) (42.2% vs. 27.5%, p<0.05) and KIR2DL3*001 was associated with sustained viral response (SVR) (41.6% vs. 61.2%, p<0.005). The KIR2DL3*001-HLA-C1 association was also significant (24.5% vs. 45.7%, p<0.001). From the frequencies of KIR obtained, 35 genotypes were assigned on the basis of previous studies. The centromeric A/A genotype was more frequent in SVR (44.1% vs. 34.5%, p<0.005) and the centromeric B/B genotype was found to be significantly more frequent in NSVR (20.9% vs. 11.2%, p<0.001). The logic regression model showed the importance of KIR genes in predicting the response to combined treatment, since the positive predictive value (PPV) was improved (from 55.9% to 75.3%) when the analysis of KIR was included in addition to the IFNL3 rs12979860 polymorphism. The study of KIR receptors may be a powerful tool for predicting the combined treatment response in patients with chronic HCV infection in association with the determination of IFNL3 polymorphism.
Citation: Vidal-Castiñeira JR, López-Vázquez A, Martínez-Borra J, Martínez-Camblor P, Prieto J, López-Rodríguez R, et al. (2014) Diversity of Killer Cell Immunoglobulin-Like Receptor (KIR) Genotypes and KIR2DL2/3 Variants in HCV Treatment Outcome. PLoS ONE 9(6): e99426. https://doi.org/10.1371/journal.pone.0099426
Editor: Niklas K. Björkström, Karolinska Institutet, Sweden
Received: February 19, 2014; Accepted: May 14, 2014; Published: June 13, 2014
Copyright: © 2014 Vidal-Castióeira et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Supported by Spanish grant PI12-02587 from the Instituto de Salud Carlos III, RETICS REDiNREN (RD12/0021/0021) from the Instituto de Salud Carlos III and by European Union “Fondos FEDER ”. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Innate immunity is the first line of defence against pathogens, and operates non-specifically [1]. Natural killer (NK) cells are important effector lymphocytes that participate in this early immune response to pathogens, including virally infected cells, and tumoral cells through the production of cytokines and chemokines [2]. NK cell function is regulated by a network of activating and inhibitory receptors [3], including killer immunoglobulin-like receptors (KIRs) [4].
KIRs, which are members of the CD158 gene family, are clustered in a 160 kilobase (kb) length of the 19q13.4 chromosome region within the leukocyte receptor complex (LRC) [5]. This diverse family of activating and inhibitory receptors modulates the development and activity of NK cells and some subpopulations of CD8+ T cells by interacting with the class I major histocompatibility complex (MHC) [6]. Although this recognition is well established, their interactions have yet to be fully understood. Genetic analyses indicate that KIR variation, in conjunction with polymorphic MHC class I genes, plays a key role in immune defence [7]. Thus, the extensive polymorphisms of HLA and KIR genes and their independent segregation give rise to unusual expression features. KIR receptors for which there is no HLA ligand can be expressed, while conversely an HLA ligand can be expressed for which there is no KIR. Furthermore, the interactions are influenced by peptides that bind to HLA class I and are contacted by KIR [8].
To date, 14 KIR receptors and two pseudogenes have been identified, and on the basis of this variation in gene content, more than 50 KIR haplotypes have been identified [9].
Previous analyses of these haplotypes indicate that they are subdivided into two groups: haplotype A and haplotype B [10]. Group A haplotypes comprise seven genes (KIR2DL1, KIR2DL3, KIR2DL4, KIR3DL1, KIR3DL2, KIR3DL3 and KIR2DS4) and two pseudogenes (KIR2DP1 and KIR3DP1). The gene KIR2DS4 is potentially activating but is disabled by a 22-bp frameshift deletion in approximately 75% of A haplotypes and is only functional in a minority of individuals [11]. Moreover, KIR2DL4 encodes a receptor that has both inhibitory and activating functions [12]. In contrast, Group B haplotypes are composed of varying numbers of KIR genes, including at least one of the following KIRs: KIR2DL2, KIR2DL5, KIR3DS1, KIR2DS1, KIR2DS2, KIR2DS3 and KIR2DS5. Subsequently, these two groups were shown to have distinct associations with several diseases [13], [14]. Classic linkage disequilibrium (LD) studies identified two distinct regions in the KIR cluster around the KIR2DL4 gene: a centromeric (cen) region, which seems to be driven by the KIR2DL5 and KIR2DL2/3 loci [15], and a telomeric (tel) region driven by the KIR3DL1/S1 locus [10].
Several studies have shown that individuals vary in the number and type of KIR loci they contain [16]. This variability and biomedical relevance of KIRs make it important to study their organization. There is increasing evidence that receptor–ligand specificity between polymorphic KIRs and polymorphic MHC class I genes is associated with a wide range of infections, such as HIV and hepatitis C, in addition to autoimmune diseases [17], disorders in pregnancy [18], and in bone marrow and solid organ transplantation [19], [20], [21].
Hepatitis C virus (HCV) is a hepatotropic non-cytopathic positive-strand RNA virus that is a major cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma, affecting over 170 million people worldwide [22]. Pegylated interferon-alpha (Peg-IFN-α) plus Ribavirin has been the most effective therapy for chronic hepatitis C [23], until the emergence of new directly acting antiviral agents (DAA) like Boceprevir and Telaprevir. Several host and viral factors influence the treatment outcome [24], [25] and approximately 50% of patients with HCV genotype 1 achieve a sustained viral response (SVR) [26]. Recently, single nucleotide polymorphisms (SNPs) in the IFNL3 gene (Also known as IL28B) that influence the responsiveness of treatment of HCV infection and the clearance of HCV have been identified [27], [28]. The strongest association has been shown with the IFNL3 SNP rs12979860 C/C genotype, which has an SVR ratio over 70% in European-American and Hispanic patients with the HCV genotype 1 [27]. Other host genetic factors associated with the response outcome have been reported, such as KIR [29] and the PD-1.3 polymorphism in the PDCD1 gene, which has recently been described by our group [30].
The aim of this study was to investigate the role of KIR genes and their genotypes on the response to combined therapy in a well-characterised group of patients with chronic HCV infection. We also examined the effect of the different KIR2DL2/2DL3 alleles and KIR haplotypes on the response outcome. Finally, a mathematical model was developed to evaluate the combined effect of IFNL3 polymorphisms and KIR receptors.
Materials and Methods
Study population
A cohort of 811 unrelated HCV genotype 1 infected Caucasian patients was recruited from several Spanish Hospitals: Hospital Universitario Central de Asturias (HUCA), Hospital San Agustín (Aviles), Clínica Universitaria de Navarra (CUN) and Hospital La Princesa (Madrid).
Patients included in this study had received their first course of antiviral therapy. The duration of treatment with PEG-IFN-α-2a or α-2b and Ribavirin was established in 48 weeks on the basis of the HCV genotype, according to consensus clinical guidance. Those without HCV RNA in their sera by six months post-treatment were defined as sustained viral responders (SVR, n = 313). The other patients were defined as non-sustained viral responders (NSVR, n = 498) which included non-responders, relapsers and partial responders.
This study was granted ethical approval by the Regional Ethics Committee for Clinical Investigation of all hospitals (Regional Ethics Committee of Clinical Research of Principado de Asturias; Ethics Committee of Clinical Research of Navarra; Regional Ethics Committee of Clinical Research of Madrid). All participants provided written informed consent.
DNA extraction
Genomic DNA was extracted from peripheral blood with the Magtration-MagaZorb DNA Common Kit-200 N using the Magtration 12GC system (Precision System Science Co., Ltd., Woerrstadt, Germany) and the Maxwell 16 Blood Purification Kit using the Maxwell 16 Instrument (Promega Corporation, Madison, Wisconsin, USA).
IFNL3 genotyping
The rs12979860 SNP was genotyped by amplifying the region containing the polymorphic site and hybridisation with fluorescent-labelled probes in an RT-PCR based on the melting-curve analysis using the Light-Cycler system (Roche Diagnostics, Mannheim, Germany) [30].
HLA/KIR genotyping
The HLA-B, HLA-C, and KIR genes were typed by using Lifecodes HLA-SSO and KIR-SSO typing kits (Tepnel Lifecodes Corporation, Stamford, UK) based on the Luminex Multi-Analyte Profiling system (xMAP technology) (Luminex Corp., Austin, TX), following the manufacturer's instructions. Ambiguities in KIR typing were resolved by PCR-single specific primer (SSP), as defined previously [31]. HLA-Bw4, HLA-Bw6, HLA-C1 and HLA-C2 were assigned on the basis of the amino acid residues of HLA-B and HLA-C alleles [32]. The KIR genotypes were deduced from KIR profiles as previously described [10], [33]. KIR gene profiles were classified with respect to the centromeric and telomeric regions of the KIR A and B haplotypes: Cen-A driven by KIR2DL3 while Cen-B by KIR2DL2. On the other hand the Tel-A is driven by KIR3DL1 and Tel-B by KIR3DS1.
Finally, a PCR was also designed for KIR2DL5 subtyping to distinguish between expressed and non-expressed variants of this gene, as previously described [34].
KIR2DL2 and KIR2DL3 oligotyping
Based on the results of KIR genotyping, patients with KIR2DL2 and/or with KIR2DL3 genes were genotyped as described elsewhere [35], [36] to identify different alleles. 146 NSVR and 103 SVR patients were randomly selected for this study. Different regions of these KIR receptors were selectively amplified by a PCR using locus-specific primers. For SSOP hybridisation, amplified DNA was blotted onto nylon membranes and hybridised with a panel of 5 and 13 SSOPs designed to detect unique sequence motifs of known KIR2DL2 and KIR2DL3 alleles, respectively. Alleles were assigned with respect to the reaction patterns of the SSOP, based on the known KIR2DL2 and KIR2DL3 sequences. This method identifies the most common alleles for KIR2DL2 (*001, *002, *003, *004 and *005) and for KIR2DL3 (*001, *002, *003, *004, *005 and *006).
Statistical analysis
Continuous variables were summarised as the mean and standard deviation. Categorical variables were summarised as absolute frequencies and percentages. The Genetic Analysis Package (GAP) in R (www.r-project.org) was used to compute LD statistics for pairs of alleles. In particular, the D' statistic based on Cramer's V was used. The general bootstrap algorithm (gBA) [37] was used to compare global and particular LDs in responders and non-responders. Fisher’s exact test was used to examine the relationship between categorical variables. In addition, where the relationships were statistically significant (p < 0.05) the odds ratios (ORs) and the respective 95% confidence intervals were also reported. Finally, the R package LogicReg (available in the CRAN) was used to develop a logic regression (LR) model [38]. A stepwise criterion based on the Akaike interaction was included in this model. LR aims to identify predictors that are Boolean (logical) combinations of the original predictors. In particular, LR looks for an optimal model of the form: where Y is the response (Yes/No), β1,…, βp are the parameters, and L1,…, Lp are Boolean combinations (intersection (AND,
), unions (OR
) and complement (NOT, C) of the dichotomous variables considered. The Li's are also called logic trees. Our model considered the variables: IFNL3 rs12979860 genotype, KIR2DL2/3, KIR2DL2-HLA, KIR2DL3-HLA, KIR Genotype, KIR centromeric motif, KIR telomeric motif and KIR2DL2/3 alleles described in this study. We also included clinical parameters like viral load, BMI, sex, etc. The logic trees were selected by a greedy search based on the usual deviance as the measure of model fit. Finally, the ROC curve and the area under the ROC curve (AUC) were used to check the quality of the model obtained.
Results
Distribution of KIR receptors frequency
A total of 313 patients (38.4%) with SVR and 498 (61.4%) with NSVR were recruited for the study. All of them were genotyped for HLA-B, HLA-C and KIR receptors. As shown in table 1, the framework genes KIR2DL4, KIR3DL2, KIR3DL3 and the pseudogene KIR3DP1 were present in all individuals. Also, as previously described [29], [39], three KIR genes were associated with chronic hepatitis C treatment response: KIR2DL2 was more frequent in the NSVR group (OR = 1.51 95%, CI = 1.13–2.02; p<0.005) while KIR2DL3 was associated with good treatment outcome (OR = 0.48, 95% CI = 0.32–0.72; p<0.001). On the other hand, the activating KIR2DS2 gene was also associated with a poor response to combined therapy (OR = 1.55, 95% CI = 1.16–2.07; p<0.005). Moreover, considering KIR2DL2 and KIR2DL3 alleles of the same locus [15], the differences between the two groups of patients were statistically significant when they were in the homozygous state: KIR2DL2/KIR2DL2 with NSVR (20.7% vs. 11.2%, p<0.001) and KIR2DL3/KIR2DL3 with SVR (35.3% vs 45%, p < 0.01). Differences in other KIR receptors studied were not significant.
Next, we have genotyped for HLA-B and HLA-C alleles but we did not find statistical differences except for HLA-C*07. We found that this allele, which belonged to HLA-C1 group, was more frequent in patients with SVR (2n = 148, 23.6%) than in non responders (2n = 171, 17.2%) (p<0.005, Table S1).
When the distribution of KIR inhibitory receptors with their HLA ligands was analysed, we found, as previously described, that the homozygous KIR2DL3-HLA-C1 genotype was significantly more common in patients with SVR than in those with NSVR (OR = 0.34, 95% CI = 0.21–0.54; p<0.001). In contrast, KIR2DL2-HLA-C1C2 was more frequent in NSVR patients (OR = 2.85, 95% CI = 1.56–5.26; p<0.001). We also found that KIR2DL3-HLA-C*07 was significantly associated with a good treatment outcome (27.3% in NSVR vs 45% in SVR, p<0.001, table S1).
As expected, with regard to KIR2DS2, we observed an association of KIR2DS2-HLA-C1 with NSVR (51.2% vs. 40.9%; p<0.01). No other significant association was found in relation to KIR2DS2. Finally, we did not find statistical associations between KIR3DL1/S1+HLA-Bw480I/80T genotypes and treatment outcome. In fact, KIR3DS1, previously associated with hepatocellular carcinoma (HCC) progression [40], was also not significant.
Oligotyping of KIR2DL2 and KIR2DL3
Having found an association of KIR2DL2 and KIR2DL3 with treatment outcome, we decided to analyse the possible influence of different KIR2DL2/KIR2DL3 alleles in the response to HCV treatment (Table 2). For KIR2DL2, five alleles were analysed (KIR2DL2*001-005), but only KIR2DL2*001 and KIR2DL2*003 were identified in our cohort. In the case of KIR2DL3, six alleles were analysed (KIR2DL3*001-006), with only KIR2DL3*001 and KIR2DL3*002 being found in our population.
The study of the contribution of these alleles to the treatment response showed that the KIR2DL2*001 was significantly more frequent in NSVR (OR = 1.92, 95% CI = 1.11–3.31; p<0.05) and the KIR2DL3*001 was associated with SVR (OR = 0.44, 95% CI = 0.26–0.74; p<0.005). Moreover, KIR2DL3*001 in combination with HLA-C1 ligand was related to a good treatment outcome (OR = 0.37, 95% CI = 0.22–0.65; p<0.001). No association was found between KIR2DL3*002-C1, KIR2DL2*001-C1 and KIR2DL2*003-C1 (data not shown).
KIR haplotype frequency in patients with chronic HCV infection
In the current study, the KIR genotypes were analysed to determine the KIR haplotypes, based on those characterised in other studies [10], [33]. Thirty-five genotypes were identified in this cohort and were resolved into corresponding pairs of haplotypes (Figure 1).
The genotypes were deduced from KIR profiles as previously described [10], [33]. Note: 1ID assigned by the Allele Frequency Net Database (http://www.allelefrequencies.net)
The most common genotype was KIR3DL3-2DL3-2DP1-2DL1-3DP1-2DL4-3DL1-2DS4-3DL2 (Genotype N° 1), corresponding to homozygosity for the major subtype of the previously reported A haplotype (28.1% in SVR and 23.5% in NSVR). Most of these genotypes included the KIR2DL3*001 allele. The other KIR genotypes found included at least one activating locus in addition to KIR2DS4, corresponding to haplotype B. These genotypes are named Bx. In general, we observed that genotypes that included KIR2DL2, especially the *001 allele, were present at a higher frequency in NSVR patients. The most prevalent was KIR3DL3-KIR2DS2-KIR2DL2-2DL3-2DP1-2DL1-3DP1-2DL4-3DL1-2DS4-3DL2 (genotype N° 2), which includes the KIR2DS2 in strong LD with the KIR2DL2 gene. On the other hand, in SVR patients, the predominant genotypes were those that included KIR2DL3, like the aforementioned genotype 1.
The overall patterns of linkage disequilibrium of KIR receptors were similar in two groups of patients (Figure 2). As previously described, the measures showed two distinct regions in the KIR cluster around the KIR2DL4 gene. One was a centromeric region (KIR3DL3, KIR2DS2, KIR2DL2, KIR2DL3, KIR2DS3, KIR2DP1, KIR2DL1 and KIR3DP1) that appeared to be driven by the KIR2DL5 and KIR2DL2/3 locus; the other was a telomeric region (KIR2DL4, KIR3DL1, KIR3DS1, KIR2DL5, KIR2DS5, KIR2DS1, KIR2DS4 and KIR3DL2), driven by KIR3DL1/S1.
In this approach to assessing KIR LD, the KIR cluster genetic polymorphism is considered simply as the presence (POS) or absence (NEG) of KIR genes. Note: Several conditions were applied for this analysis: 1) KIR3DL3, KIR3DP1, KIR2DL4 and KIR3DL2 were not included because they were present in all patients (framework genes), 2) KIR2DL2 and KIR2DL3 were considered as alleles of the same locus, 3) KIR2DL5 was differentiated in KIR2DL5A and KIR2DL5B because they are located at different positions, and 4) KIR2DS3 was included in the centromeric region.
Subdividing the KIR cluster into the two regions identified by the LD study, simplifies the description of the haplotype structure of the KIR genes. When we compared the frequency of centromeric and telomeric motifs in the two groups, we found that the Cen-A haplotype in homozygous was significantly more common in SVR patients (OR = 0.63, 95% CI = 0.47–0.84; p<0.005) and the Cen-B haplotype segment in homozygous was more frequent in NSVR patients (OR = 2.04, 95% CI = 1.35–3.03; p<0.001) (Table 3). Other combinations were not significant. Finally, we analysed the association of the centromeric and telomeric KIR haplotypes with HLA-C and HLA-B. We found a statistically significant association of the Cen-A/A motif and HLA-C with good treatment outcome (OR = 0.62, 95% CI = 0.45–0.84; p<0.005). In contrast, the association of the Cen-B/B motif and HLA-C combination was significant (OR = 2.5, 95% CI = 1.5 –3.96; p<0.001) in relation with non-response. The Tel-HLA combinations were not found to be associated with treatment outcome.
Analysis of KIR genotyping and IFNL3 polymorphism combinations
We used logic regression to examine Boolean combinations (AND/OR/NOT) of binary covariates considered in this study, in order to derive a model for predicting chronic HCV treatment outcome. The final model obtained is summarised as follows. Combinations of KIR genotyping, IFNL3 and the intersections and unions of other factors can be expressed mathematically (figure 3).
The measure of the importance of a variable in the response to treatment is the frequency with which it appears independently in all models. From the summary, the contribution of each variable can be evaluated by the ratio of the number of times it was included in any of the models divided by the total number of models.
Logic regression revealed a strong association of the IFNL3 rs12979860 polymorphism in the prediction of response to combined treatment (PPV = 55.9%) (Table 4). It was improved when we incorporated different studies of KIR combinations in the logic model. The inclusion of the complete analysis of KIR genes and their combinations to the mathematical model yielded the best result (PPV = 75.3%). The ROC curve of different logic regression models was derived. In the best model, the area under the curve (AUC) was 0.729 (95% CI = 0.692–0.772) (Figure 4).
We also included the ROC curves of IFNL3 (dotted line) and KIR2DL2/3-HLA (dashed line) logic model determinations. It can be observed they do not overlap with the best logic regression model. Note: Area under the curve (AUC) of the best model: 0.729 (95% CI, 0.692–0.772).
In conclusion, KIR2DL3/KIR2DL3-HLA-C1/C1 and the KIR centromeric haplotype motif (A/A genotype) are good predictors of SVR. On the other hand, KIR2DL2/KIR2DL2-HLA-C1/C2 and the KIR centromeric haplotype motif (B/B genotype) are associated with a high percentage of NSVR in combination with IFNL3 T/T. Our statistical analysis did not show any association between KIR2DS2 and treatment outcome.
Discussion
Previous studies have described several biomarkers of treatment response in addition to IFNL3 polymorphisms [41], [42]. Other study by our group has described the influence of KIR genes on the progression of chronic HCV infection and treatment outcome [29]. As well as KIR receptors, other genetic markers related to immune response have been reported [42]. One such marker is PDCD1, which allowed us to develop a model for predicting the response to conventional therapy. Combining the analysis of this marker with IFNL3 genotyping considerably improved the predictability of response to standard treatment established using other parameters such as viral load or viral genotype [30].
The current models predict the treatment response of each patient only approximately, so the search for new genetic markers is essential if we are to be able to establish the most appropriate therapy for each patient from the beginning of treatment [30].
As previously reported, factors such as the KIR-HLA genotype were relevant to this response. The KIR2DL3/KIR2DL3-HLA-C1C1 genotype was more frequent in the SVR group, while the KIR2DL2/KIR2DL2-HLA-C1C2 genotype was more frequent in NSVR patients. In relation with this, we found that the HLA-C*07 was associated with a good treatment response, especially in combination with KIR2DL3. This data is interesting because this HLA-C allele was previously associated with the HCV persistence [43] and, more recently, with high viral load [44].
In order to determine specifically which allelic variant of these genes predominates in the two groups of patients, we carried out an oligotyping analysis of KIR2DL2 and KIR2DL3, as previously described [35], [36]. In relation to KIR2DL3, we found that the KIR2DL3*001 allele was most common in SVR while KIR2DL2*001 was more frequent in NSVR. In addition, the relevance of KIR2DL3*001 to good treatment outcome is notably greater when analysed in combination with HLA-C1. These findings suggest these alleles are important in the response to standard treatment, and may be associated with a weak interaction with its ligand, as described for other alleles from KIR2DL2/3 [45]. Our results confirm the relevance of the KIR2DL2/3 alleles and the interaction with HLA-C in the treatment outcome, which could also be involved in the progression and chronicity of HCV infection. With regard to KIR3DL1-HLA-Bw4, Nozawa et al showed a significant association of this combination with SVR to Peg-IFN/RBV in Japanese population [46]. Nonetheless, we did not find statistical differences in our cohort, in line with previous reports in Caucasian populations. [29], [47]. This data underscore the distinctive organisation of the human KIR locus that drives the generation of KIR gene-content diversity in different populations [48], .
The human KIR locus contains highly conserved genes that are situated in the middle (KIR3DP1 and KIR2DL4) and at the ends (KIR3DL3 and KIR3DL2) of the cluster, creating a framework around two regions of variability in which highly homologous KIR genes are packed very close together in a head-to-tail configuration, separated by short and highly conserved intergenic regions. In different human populations, there is a variable balance between A and B groups of KIR haplotypes, which seems to be maintained by balancing selection for inhibitory and activating functions [50]. This selection is mediated, in part, by the interaction of inhibitory KIR with their HLA class I ligands [51]. A relationship between KIR haplotypes and chronic HCV infection has been demonstrated. Thus, the centromeric KIR haplotype A was increased in patients with resolved HCV infection compared with those who developed a chronic infection [52].
To investigate the role of KIR genes underlying HCV action from a different perspective, in the present study we analysed the relationship between the KIR genotypes and haplotypes with the response to conventional treatment. We found 35 KIR genotypes in our cohort, all of them previously described. The biological significance of the A/B haplotype difference means that it is worth considering whether combinations of A and B haplotypes can influence chronic HCV treatment outcome. The present study has clearly confirmed that such a situation occurs in this series of patients. We found a high frequency of homozygosity for haplotype A in patients with SVR. When we divided this haplotype into the A-centromeric and A-telomeric motifs, we found that the homozygosity for Cen A/A was also more common in SVR patients. This result is controversial because of the high frequency of inhibitory receptors in patients with a good response. This association may have arisen because KIR2DL3 is present in the Cen A haplotype motif, while KIR2DL2 is part of Cen B haplotype motif. It has also been shown that KIR2DL2 and KIR2DL3 have different inhibitory capabilities arising from dissimilar interactions with their ligands [45]. KIR2DL1 has the strongest interaction with its ligands, the HLA-C2 alleles, while KIR2DL2 associates to HLA-C1 more intensely than KIR2DL3 [53]. Bearing these data in mind, we may surmise that the stronger interaction and therefore the stricter inhibition of NK cells would occur in individuals homozygous for KIR2DL2 who carry the HLA-C1/C2 genotype. Conversely, NK cells from patients who are homozygous for KIR2DL3 and for HLA-C1 would have a greater activation ability due to the lack of intense inhibition arising from the interaction between KIR2DL1 and the HLA-C2 ligand. However, it should be taken into consideration that KIR-activating receptors in the A haplotype are not reduced to KIR2DS4, because KIR2DL4 can sometimes act as an activating receptor [54], [55].
Suppiah et al [56], observed a significant association between HLA groups C1 and C2 with treatment outcome and established a logistic model to predict a failure to clear virus on therapy with PegIFN/RBV. We did not find any association with HLA-C groups and treatment except HLA-C*07. Nonetheless, we observed a clear association between KIR2DL2/3 and therapy response. Also, we studied all KIR genes to establish the different haplotypes of these receptors, and found a statistical association with SVR. These different results may be due to patient selection and different study population. To assess the significance of these results to treatment outcome, we developed a logic regression model [37], different from Suppiah. This showed that genotyping IFNL3 alone had a specificity of 74.3% and a positive predictive value (PPV) of 55.9%. However, this result was improved by the determination of KIR genes, increasing the specificity to 95.6% and the PPV to 75.3%. Although the sensitivity decreased from 51.7% to 21.4%, these results are further evidence of the important role that KIR receptors play in the response to combined therapy in HCV-infected patients.
In conclusion, based on our results we propose a novel relationship between KIR/HLA genotypes and KIR haplotypes (A/B) in the chronicity of HCV infection. These could be used as tools in combination with other parameters, such as the determination of IFNL3 rs12979860, to predict treatment outcome and in order to establish the best cost-effective treatment between the classical regimen and the new arrival therapies. Further research is required to determine the molecular mechanisms involved in this process.
Supporting Information
Table S1.
A) HLA-C alleles distribution in patients with HCV chronic infection. B) KIR2DL2 and KIR2DL3 association with HLA-C*07 allele.
https://doi.org/10.1371/journal.pone.0099426.s001
(DOC)
Author Contributions
Conceived and designed the experiments: JVC ALV CLL. Performed the experiments: JVC ALV JMB. Analyzed the data: JVC CLL PMC. Contributed reagents/materials/analysis tools: JP RLR PSC JdlV LR RPL RPÃ. Wrote the paper: JVC ALV RPÃ CLL.
References
- 1. Akira S, Uematsu S, Takeuchi O (2006) Pathogen recognition and innate immunity. Cell 124: 783–801.
- 2. Jost S, Altfeld M (2013) Control of human viral infections by natural killer cells. Annu Rev Immunol 31: 163–194.
- 3. Bryceson YT, Chiang SC, Darmanin S, Fauriat C, Schlums H, et al. (2011) Molecular mechanisms of natural killer cell activation. J Innate Immun 3: 216–226.
- 4. Middleton D, Gonzelez F (2010) The extensive polymorphism of KIR genes. Immunology 129: 8–19.
- 5. Hsu KC, Chida S, Geraghty DE, Dupont B (2002) The killer cell immunoglobulin-like receptor (KIR) genomic region: gene-order, haplotypes and allelic polymorphism. Immunol Rev 190: 40–52.
- 6. Bjorkstrom NK, Beziat V, Cichocki F, Liu LL, Levine J, et al. (2012) CD8 T cells express randomly selected KIRs with distinct specificities compared with NK cells. Blood 120: 3455–3465.
- 7. Passweg JR, Huard B, Tiercy JM, Roosnek E (2007) HLA and KIR polymorphisms affect NK-cell anti-tumor activity. Trends Immunol 28: 437–441.
- 8. Fadda L, Korner C, Kumar S, van Teijlingen NH, Piechocka-Trocha A, et al. (2012) HLA-Cw*0102-restricted HIV-1 p24 epitope variants can modulate the binding of the inhibitory KIR2DL2 receptor and primary NK cell function. PLoS Pathog 8: e1002805.
- 9. Gonzalez-Galarza FF, Christmas S, Middleton D, Jones AR (2011) Allele frequency net: a database and online repository for immune gene frequencies in worldwide populations. Nucleic Acids Res 39: D913–919.
- 10. Pyo CW, Guethlein LA, Vu Q, Wang R, Abi-Rached L, et al. (2010) Different patterns of evolution in the centromeric and telomeric regions of group A and B haplotypes of the human killer cell Ig-like receptor locus. PLoS One 5: e15115.
- 11. Middleton D, Gonzalez A, Gilmore PM (2007) Studies on the expression of the deleted KIR2DS4*003 gene product and distribution of KIR2DS4 deleted and nondeleted versions in different populations. Hum Immunol 68: 128–134.
- 12. Rajagopalan S, Long EO (2012) KIR2DL4 (CD158d): An activation receptor for HLA-G. Front Immunol 3: 258.
- 13. Lu Z, Zhang B, Chen S, Gai Z, Feng Z, et al. (2008) Association of KIR genotypes and haplotypes with susceptibility to chronic hepatitis B virus infection in Chinese Han population. Cell Mol Immunol 5: 457–463.
- 14. Hirayasu K, Ohashi J, Kashiwase K, Hananantachai H, Naka I, et al. (2012) Significant association of KIR2DL3-HLA-C1 combination with cerebral malaria and implications for co-evolution of KIR and HLA. PLoS Pathog 8: e1002565.
- 15. Uhrberg M, Parham P, Wernet P (2002) Definition of gene content for nine common group B haplotypes of the Caucasoid population: KIR haplotypes contain between seven and eleven KIR genes. Immunogenetics 54: 221–229.
- 16. Jiang W, Johnson C, Jayaraman J, Simecek N, Noble J, et al. (2012) Copy number variation leads to considerable diversity for B but not A haplotypes of the human KIR genes encoding NK cell receptors. Genome Res 22: 1845–1854.
- 17. Diaz-Pena R, Vidal-Castineira JR, Alonso-Arias R, Suarez-Alvarez B, Vicario JL, et al. (2010) Association of the KIR3DS1*013 and KIR3DL1*004 alleles with susceptibility to ankylosing spondylitis. Arthritis Rheum 62: 1000–1006.
- 18. Chazara O, Xiong S, Moffett A (2011) Maternal KIR and fetal HLA-C: a fine balance. J Leukoc Biol 90: 703–716.
- 19. Clausen J, Kircher B, Auberger J, Schumacher P, Grabmer C, et al. (2012) Bone marrow may be the preferable graft source in recipients homozygous for HLA-C group 2 ligands for inhibitory killer Ig-like receptors. Bone Marrow Transplant 47: 791–798.
- 20. Gallez-Hawkins GM, Franck AE, Li X, Thao L, Oki A, et al. (2011) Expression of activating KIR2DS2 and KIR2DS4 genes after hematopoietic cell transplantation: relevance to cytomegalovirus infection. Biol Blood Marrow Transplant 17: 1662–1672.
- 21. van Bergen J, Thompson A, Haasnoot GW, Roodnat JI, de Fijter JW, et al. (2011) KIR-ligand mismatches are associated with reduced long-term graft survival in HLA-compatible kidney transplantation. Am J Transplant 11: 1959–1964.
- 22. Lavanchy D (2009) The global burden of hepatitis C. . Liver Int 29 Suppl 174–81.
- 23. Ghany MG, Strader DB, Thomas DL, Seeff LB (2009) Diagnosis, management, and treatment of hepatitis C: an update. Hepatology 49: 1335–1374.
- 24. Piasecki BA, Lewis JD, Reddy KR, Bellamy SL, Porter SB, et al. (2004) Influence of alcohol use, race, and viral coinfections on spontaneous HCV clearance in a US veteran population. Hepatology 40: 892–899.
- 25. Chen L, Borozan I, Feld J, Sun J, Tannis LL, et al. (2005) Hepatic gene expression discriminates responders and nonresponders in treatment of chronic hepatitis C viral infection. Gastroenterology 128: 1437–1444.
- 26. Farnik H, Lange CM, Sarrazin C, Kronenberger B, Zeuzem S, et al. (2010) Meta-analysis shows extended therapy improves response of patients with chronic hepatitis C virus genotype 1 infection. Clin Gastroenterol Hepatol 8: 884–890.
- 27. Ge D, Fellay J, Thompson AJ, Simon JS, Shianna KV, et al. (2009) Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature 461: 399–401.
- 28. Thomas DL, Thio CL, Martin MP, Qi Y, Ge D, et al. (2009) Genetic variation in IL28B and spontaneous clearance of hepatitis C virus. Nature 461: 798–801.
- 29. Vidal-Castineira JR, Lopez-Vazquez A, Diaz-Pena R, Alonso-Arias R, Martinez-Borra J, et al. (2010) Effect of killer immunoglobulin-like receptors in the response to combined treatment in patients with chronic hepatitis C virus infection. J Virol 84: 475–481.
- 30. Vidal-Castineira JR, Lopez-Vazquez A, Alonso-Arias R, Moro-Garcia MA, Martinez-Camblor P, et al. (2012) A predictive model of treatment outcome in patients with chronic HCV infection using IL28B and PD-1 genotyping. J Hepatol 56: 1230–1238.
- 31. Gomez-Lozano N, Vilches C (2002) Genotyping of human killer-cell immunoglobulin-like receptor genes by polymerase chain reaction with sequence-specific primers: an update. Tissue Antigens 59: 184–193.
- 32. Deng L, Mariuzza RA (2006) Structural basis for recognition of MHC and MHC-like ligands by natural killer cell receptors. Semin Immunol 18: 159–166.
- 33. Hsu KC, Liu XR, Selvakumar A, Mickelson E, O'Reilly RJ, et al. (2002) Killer Ig-like receptor haplotype analysis by gene content: evidence for genomic diversity with a minimum of six basic framework haplotypes, each with multiple subsets. J Immunol 169: 5118–5129.
- 34. Gomez-Lozano N, Gardiner CM, Parham P, Vilches C (2002) Some human KIR haplotypes contain two KIR2DL5 genes: KIR2DL5A and KIR2DL5B. Immunogenetics 54: 314–319.
- 35. Gonzalez A, Meenagh A, Sleator C, Middleton D (2008) Investigation of killer cell immunoglobulin-like receptor (KIR) gene diversity: KIR2DL2, KIR2DL5 and KIR2DS5. Tissue Antigens 72: 11–20.
- 36. Keaney L, Williams F, Meenagh A, Sleator C, Middleton D (2004) Investigation of killer cell immunoglobulin-like receptor gene diversity III. KIR2DL3. Tissue Antigens 64: 188–194.
- 37. Martínez-Camblor P, Corral N (2012) A general bootstrap algorithm for hypothesis testing. Journal of Statistical Planning and Inference 142: 589–600.
- 38. Ruczinski I, Kooperberg C, LeBlanc M (2003) Logic Regression. Journal of Computational and Graphical Statistics 12: 475–511.
- 39. Askar M, Avery R, Corey R, Lopez R, Thomas D, et al. (2009) Lack of killer immunoglobulin-like receptor 2DS2 (KIR2DS2) and KIR2DL2 is associated with poor responses to therapy of recurrent hepatitis C virus in liver transplant recipients. Liver Transpl 15: 1557–1563.
- 40. Lopez-Vazquez A, Rodrigo L, Martinez-Borra J, Perez R, Rodriguez M, et al. (2005) Protective effect of the HLA-Bw4I80 epitope and the killer cell immunoglobulin-like receptor 3DS1 gene against the development of hepatocellular carcinoma in patients with hepatitis C virus infection. J Infect Dis 192: 162–165.
- 41.
Thompson AJ, Muir AJ, Sulkowski MS, Ge D, Fellay J, et al. (2010) Interleukin-28B polymorphism improves viral kinetics and is the strongest pretreatment predictor of sustained virologic response in genotype 1 hepatitis C virus. Gastroenterology 139: 120–129 e118.
- 42. Lopez-Rodriguez R, Trapero-Marugan M, Borque MJ, Roman M, Hernandez-Bartolome A, et al. (2011) Genetic variants of interferon-stimulated genes and IL-28B as host prognostic factors of response to combination treatment for chronic hepatitis C. . Clin Pharmacol Ther 90: 712–721.
- 43. McKiernan SM, Hagan R, Curry M, McDonald GS, Kelly A, et al. (2004) Distinct MHC class I and II alleles are associated with hepatitis C viral clearance, originating from a single source. Hepatology 40: 108–114.
- 44. Tseng KC, Tseng CW, Hsieh YH, Chang CK, Lai NS, et al. (2013) Effect of human leukocyte antigen class I and II alleles on hepatitis C viral load among chronic hepatitis C patients in Southern Taiwan. Hum Immunol 74: 978–982.
- 45. Moesta AK, Norman PJ, Yawata M, Yawata N, Gleimer M, et al. (2008) Synergistic polymorphism at two positions distal to the ligand-binding site makes KIR2DL2 a stronger receptor for HLA-C than KIR2DL3. J Immunol 180: 3969–3979.
- 46. Nozawa Y, Umemura T, Joshita S, Katsuyama Y, Shibata S, et al. (2013) KIR, HLA, and IL28B variant predict response to antiviral therapy in genotype 1 chronic hepatitis C patients in Japan. PLoS One 8: e83381.
- 47. Knapp S, Warshow U, Hegazy D, Brackenbury L, Guha IN, et al. (2010) Consistent beneficial effects of killer cell immunoglobulin-like receptor 2DL3 and group 1 human leukocyte antigen-C following exposure to hepatitis C virus. Hepatology 51: 1168–1175.
- 48. Vierra-Green C, Roe D, Hou L, Hurley CK, Rajalingam R, et al. (2012) Allele-level haplotype frequencies and pairwise linkage disequilibrium for 14 KIR loci in 506 European-American individuals. PLoS One 7: e47491.
- 49. Jobim M, Salim PH, Portela P, Wilson TJ, Fraportti J, et al. (2010) Killer cell immunoglobulin-like receptor gene diversity in a Caucasian population of southern Brazil. Int J Immunogenet 37: 83–89.
- 50. Rajalingam R, Du Z, Meenagh A, Luo L, Kavitha VJ, et al. (2008) Distinct diversity of KIR genes in three southern Indian populations: comparison with world populations revealed a link between KIR gene content and pre-historic human migrations. Immunogenetics 60: 207–217.
- 51. Parham P (2005) MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 5: 201–214.
- 52. Dring MM, Morrison MH, McSharry BP, Guinan KJ, Hagan R, et al. (2011) Innate immune genes synergize to predict increased risk of chronic disease in hepatitis C virus infection. Proc Natl Acad Sci U S A 108: 5736–5741.
- 53. Winter CC, Gumperz JE, Parham P, Long EO, Wagtmann N (1998) Direct binding and functional transfer of NK cell inhibitory receptors reveal novel patterns of HLA-C allotype recognition. J Immunol 161: 571–577.
- 54. Faure M, Long EO (2002) KIR2DL4 (CD158d), an NK cell-activating receptor with inhibitory potential. J Immunol 168: 6208–6214.
- 55. Kikuchi-Maki A, Catina TL, Campbell KS (2005) Cutting edge: KIR2DL4 transduces signals into human NK cells through association with the Fc receptor gamma protein. J Immunol 174: 3859–3863.
- 56. Suppiah V, Gaudieri S, Armstrong NJ, O'Connor KS, Berg T, et al. (2011) IL28B, HLA-C, and KIR variants additively predict response to therapy in chronic hepatitis C virus infection in a European Cohort: a cross-sectional study. PLoS Med 8: e1001092.