• KIR–HLA Combination Predicts Activation-Driven Risk in Chronic Myeloid Leukemia
  • Masoumeh Basirat,1 Maral Mokhtari,2 Hassan Ghader Abadi,3 Shirin Farjadian,4,*
    1. Department of Immunology, Shiraz Medical School, Shiraz University of Medical Sciences, Shiraz, Iran.
    2. Department of Pathology/Faghihi Hospital, Shiraz Medical School, Shiraz University of Medical Sciences, Shiraz, Iran
    3. Raz Molecular Pathobiology Laboratory, Shiraz, Iran
    4. Department of Immunology, Shiraz Medical School, Shiraz University of Medical Sciences, Shiraz, Iran.


  • Introduction: Background: Killer-cell immunoglobulin-like receptors (KIRs) regulate natural killer (NK) cell development and operate by interacting with their cognate ligands, human leukocyte antigen class I (HLA-I) molecules. Variations in KIR and HLA-I genes and their independent inheritance can affect NK cell action and impact predisposition to hematological malignancies. Aim: In this study, the correlation between KIR gene contents, HLA-I alleles, and their combined KIR/HLA-I patterns with the risk of developing CML was evaluated among patients from southwestern Iran. Methods: The gene content of KIR and HLA-I was examined using PCR-SSP in 100 patients with CML and 181 healthy individuals. The distribution of each gene and different KIR/HLA-I combinations were compared between the two groups. Results: The study disclosed a notably greater frequency of the activating KIR3DS1 gene, the KIR2DS4 fl/fl genotype, the C4T4 genotype, and the T4 gene cluster in CML patients compared with the control group. There was no notable difference between patients and controls in the various combinations of inhibitory (i)KIR genes and their corresponding HLA-I ligands. KIR3DL1(+)/HLABw4(+) combination in the presence of 3DS1 was more common in patients, while, the frequency of this combination in the absence of 3DS1 was more frequent among controls. Additionally, activating KIR/HLA-I combinations like KIR3DS1(+)/HLA-B Bw4Thr80(+) and KIR3DS1(+)/HLA-Bw4(+) were significantly associated with CML. Conversely, healthy individuals exhibited a higher frequency of HLA-B Bw4Thr80, HLA-A Bw4, and HLA-Bw4 ligands lacking their activating receptor (KIR3DS1). The frequency of aKIR=6 and aKIR/HLA-I≥3 was significantly higher in patients. Conclusion: Based on our results, it seems that there is a link between aKIRs and aKIR/HLA-I combinations with vulnerability to CML. Overall, NK cells expressing iKIRs matched to their HLA-I ligands, but lacking the corresponding activating receptors, are essential for eliciting an effective antitumor response.
  • Methods: 2. Material and Methods 2.1 Study population In this research, the study population consisted of 100 patients with CML, whose diagnosis was made based on FAB and WHO criteria and in collaboration with the Pathobiology and Molecular Laboratory of Raz in Shiraz and 181 healthy sex- and ethnicity-matched individuals who included from the same geographical region. The control group had no history of chronic diseases, autoimmune disorders, malignancies, or family history of hereditary diseases. The Ethics Committee of Shiraz University of Medical Sciences approved the study protocol (approval number IR.SUMS.MED.REC.1403.572). Patients were selected from those referred to the Pathobiology and Molecular Laboratory of Raz Shiraz for diagnosis or follow up between June 2023 to March 2024. Written informed consent was taken from each patient for the use of their remaining blood sample for the tests required in our research. DNA was extracted from blood samples using a commercial kit (Pishgaman Sanjesh, Tehran, Iran). Required DNA samples from healthy controls were obtained from samples available in the Biobank of the Immunogenetics laboratory of the Immunology Department. Patients who underwent high-dose chemotherapy, because of the low white blood cell counts and inability to extract DNA in the required amount, and those who received bone marrow transplantation due to chimerism were excluded from the study. 2.2. Determination of KIR Gene Content PCR-SSP was used for KIR genotyping, with a set of 11 primers specific to each KIR genes . To identify the full-length versus deleted forms of KIR2DS4, a supplementary primer pair was used . After KIR genotyping, individuals were classified into AA or Bx genotype according to the presence or absence of certain KIR genes. Accordingly, individuals possessing only iKIR genes 2DL1, 2DL3, 3DL1, along with 2DS4 as the sole activating gene, were classified as AA. Individuals carrying the full gene set of A haplotype along with at least one gene specific to B haplotype were assigned to AB group, while those lacking one or more of the four defining genes of A haplotype were considered BB homozygotes. Furthermore, individuals with a Bx genotype were subclassified according to the presence of specific genes within the centromeric and telomeric regions of the KIR gene cluster. The centromeric KIR gene cluster (C4) includes 2DS2, 2DL2, 2DS3, and 2DL5B, while the telomeric KIR cluster (T4) includes 2DS1, 3DS1, 2DS5A, and 2DL5. Individuals carrying all genes from both centromeric and telomeric clusters were designated as C4T4. If all genes from only one region were present while at least one gene was absent in the other, the individuals were categorized as C4Tx or CxT4, respectively. In cases where at least one gene was missing from each cluster, individuals were classified as CxTx. 2.3. Determination of HLA Class I alleles as cognate KIR ligands HLA class I genotyping was done for those alleles which serve as cognate ligands for KIRs by PCR-SSP using 9 previously designed primer pairs , including HLA-A23, A24, A32, A3, A11, HLA-C1, C2, HLA-Bw4Ile, Bw4Thr, and Bw6. Duplex PCR analysis was conducted on every sample, employing both specific primer pairs and a primer pair for the HLA-DRA gene’s non-polymorphic region to provide internal control. Positive controls consisted of DNA samples whose KIR and HLA genotypes had been determined using high-resolution methods. Simultaneously, a PCR reaction lacking template DNA was included to serve as a negative control. Reactions were prepared in 10 μl volumes. For each, 30 ng of DNA template was combined with 0.5 μM target primers, 0.1 μM internal control primers, and 5 μl of 2× Taq DNA polymerase master mix (Ampliqon A/S, Odense, Denmark). Cycling was performed using previously standardized conditions (26). The amplified HLA and KIR gene fragments were resolved using 2% agarose gels with Safe Stain dye (Pishgam, Tehran, Iran) and electrophoresed in Tris-acetate-EDTA buffer. Due to the small size difference between the KIR2DS4fl and 2DS4del alleles, a 3.7% agarose gel was used for their discrimination. Following KIR genotyping, individual genotypes were interpreted according to the allele frequency database (http://www.allelefrequencies.net/kir6001a.asp). 2.4. Statistical analysis Chi-square tests based on 2 × 2 contingency tables were applied to compare KIR gene, haplotype, genotype, HLA-I allele, and KIR/HLA-I combination frequencies between patients and controls. Fisher’s exact test was employed when one or more cells had counts under five. For significant findings, odds ratios (ORs) and 95% confidence intervals (CIs) were computed to ascertain their link with CML. Analyses were carried out in SPSS version 27, considering P < 0.05 as statistically significant.
  • Results: 3. Results 3.1. Demographics data of study population To investigate how KIR genes together with HLA class I ligands influence susceptibility to CML, a comparison was made between 100 patients and 181 healthy controls. The mean age of the patients was 43.09 ± 14.72 years (age range: 16 to 82 years). Of these, 49 were female with a mean age of 44.71 ± 14.45 years and 51 were male with a mean age of 41.53 ± 14.95 years. 3.2. Frequency of KIR genes and genotypes Table 1 summarizes the comparison of KIR gene frequencies between patients and controls. The 3DS1 gene was observed at a significantly higher frequency in patients (P = 0.011; OR = 1.92; 95% CI: 1.16–3.19). Likewise, the 2DS4 fl/fl genotype was more prevalent among patients than controls (P = 0.032; OR = 2.27; 95% CI: 1.06–4.88) (Supplementary Table S2). As shown in Table 2, the frequency of the number of aKIR=6 was higher in the patient group than in the control group (P=0.013, OR=2.73, 95% CI: 1.20-6.20). As revealed in Table 3, despite the higher frequency of AA genotype in the control group and Bx genotype in patients, these variations were not statistically significant. The frequency of C4T4 genotype and T4 cluster in the patient group was significantly higher than in the control group (P=0.007, OR=2.90, 95% CI: 1.31-6.45 and P=0.0035, OR=2.45, 95% CI: 1.31-4.42, respectively). Whereas, The CxTx genotype was notably more prevalent in the control group than among patients (P = 0.008; OR = 0.45; 95% CI: 0.25–0.82). As listed in Supplementary Table S1, 72 different KIR gene profiles were detected, with 8 specifics to patients, 50 specifics to controls, and 14 found in both populations. 3.3. Frequency of HLA class I alleles and genotypes Our data showed no noteworthy difference in the frequency of HLA class I alleles and genotypes between patients and controls (Supplementary Tables S3 and S4). 3.4. Frequency of KIR/cognate HLA I ligand combinations 3.4.1. Frequency of iKIR/cognate HLA-I ligand combinations Findings indicated that patients and controls exhibited comparable patterns of iKIR/HLA-I ligand combinations, with no statistically significant differences detected (Tables 4). 3.4.2. Frequency of iKIRs/cognate HLA-I ligands with their activating counterparts According to the results presented in Table 5, the frequency of co-existence of KIR3DL1(+)/HLA-B Bw4Thr80(+) combination and its activating counterpart (KIR3DS1) was notably augmented in patients with CML compared to the control group (P=0.021, OR=3.24, 95% CI: 1.22-9.13). In contrast, in the control group, the frequency of co-existence of this combination in the absence of activating counterpart was raised (P=0.032, OR=0.32, 95% CI: 0.12-0.92). Furthermore, co-existence of KIR3DL1(+)/3DS1(+)/HLA-Bw4(+) combination was more frequent in patients than in controls (P=0.012, OR=2.04, 95% CI: 1.14-3.50). Conversely, controls showed a significantly higher occurrence of the KIR3DL1(+)/3DS1(-)/HLA-Bw4(+) genotype (P = 0.038, OR = 0.59, 95% CI: 0.36–0.98). 3.4.3. Frequency of aKIR/cognate HLA-I ligand combinations According to the results presented in Table 6, in contrast to the patients with raised frequency of KIR3DS1(+)/HLA-B Bw4Thr80(+) combination (P=0.04, OR=2.85, 95% CI: 1.02–7.97), KIR3DS1(-)/HLA-B Bw4Thr80(+) combination was observed with a significant higher frequency in the controls (P=0.005, OR =0.22, 95% CI:0.07–0.68). Also, KIR3DS1(-)/HLA-B Bw4Ile80(+) combination was more frequent in the controls than in patients, although this difference was not statistically significant. Likewise, KIR3DS1(+)/HLA-A Bw4(-) combination was also showed a significant increased frequency in patients (P=0.009, OR=2.44, 95% CI:1.24–4.79). In contrast, the existence of the ligand in the absence of the receptor was more frequent in the control group (P=0.027, OR=0.46, 95% CI:0.23–0.92). The combination of Bw4-positive HLA-A and HLA-B alleles with KIR3DS1 was considerably more frequent in patients compared to controls (P = 0.029; OR = 1.87; 95% CI: 1.06–3.30). Controls exhibited a significantly greater prevalence of these ligands when KIR3DS1 was absent (P = 0.009; OR = 0.49; 95% CI: 0.28–0.84). Based on the results presented in Table 7, the frequency of aKIRs/cognate HLA-I lignads≥3 in patients was significantly higher than in controls (P=0.034, OR=2.03, 95% CI:1.05-3.93).
  • Conclusion: The survey demonstrated an increased frequency of the KIR3DS1 gene, the KIR2DS4 fl/fl genotype, the C4T4 genotype, and T4 gene cluster among patients. Between patients and controls we observed no difference in the various combinations of iKIR genes and their cognate HLA-I ligands. The concurrent presence of KIR3DL1(+)/3DS1(+)/HLA-Bw4(+) combination was more common in patients. In contrast, controls showed a higher prevalence of the KIR3DL1(+)/3DS1(-)/HLA-Bw4(+) combination. Analysis of activating KIRs and their corresponding ligands revealed that patients had a higher prevalence of KIR3DS1(+)/HLA-B Bw4Thr80(+) and KIR3DS1(+)/HLA-Bw4(+) combinations. In contrast, controls more frequently carried the HLA-B Bw4Thr80, HLA-A Bw4, and HLA-Bw4 ligands without the activating receptor KIR3DS1. Overall, the findings of this study may indicate a potential functional impairment in NK cells among patients—an impairment that likely reduces their ability to recognize and eliminate malignant cells, ultimately resulting in weakened immune responses and failure to effectively prevent disease progression.
  • Keywords: Chronic Myelogenous Leukemia (CML), Killer-cell Immunoglobulin-like Receptors (KIR), Human Leukocyte