The Molecular Genetic Polymorphism Of KIR System In Southern Chinese Han Population And The Association With Leukemia

Killer cell immunoglobulin-like receptors (KIRs) are expressed on both natural killer (NK) cells and a subset of T cells. The KIR family regulates NK cell activities through interaction with class I HLA ligands, which plays an important role in tumor immunity, transplantation and the resistance of disease. The KIR gene cluster located on chromosome19q13.4consists of14functional loci (KIR2DL1-5,2DS1-5,3DL1-3,3DS1) and two pseudogenes (KIR2DP1,3DP1). Up to October2013, a total of678KIR alleles have been identified and released in the IPD-KIR Database (Release2.5.0).The molecular genetic polymorphisms of KIR, HLA genes varied in different ethnicity or region. Previous KIR population genetic studies mainly focused on determination the presence or absence of KIR framework genes and investigated the frequencies of each KIR framework gene and KIR gene profile. However, little is known about the frequencies of KIR ligands and KIR-HLA compound gene profile. In the KIR-associated studies with leukemia, the impact of KIR ligands and KIR-HLA compound gene profile were also neglected. The outcomes of KIR-associated with leukemia have been controversial due to the multiple sources of patient samples, different ethnic background, sample size and the reagents utilized in KIR PCR-SSP testing.The function of KIR on NK cells is dependent on the normal expression of class I HLA ligands on the target cell. KIR framework genes especially inhibitory KIR framework genes are extremely polymorphic. Even for the same KIR framework gene, different allele showed differential surface expression on NK cells and mediated varied activating or inhibitory signal.In the present study, we examine the frequencies of KIR gene profiles, class I HLA ligands, matched KIR+HLA pairs and compound gene profiles of KIR-HLA in southern Chinese Han population. Based on the obtained KIR population genetic data, we explore the association of KIR polymorphism with the occurrence of leukemia, and then the potential leukemia-associated KIR framework gene were further subjected to sequence-based typing with an aim of investigating the association of KIR allelic polymorphism with leukemia.Firstly, we examined the molecular genetic polymorphism of KIR system in southern Chinese Han population by analyzing the frequencies of KIR framework genes and KIR gene profiles, class I HLA ligands, matched KIR+HLA pairs and KIR-HLA compound profiles.All16KIR genes were detected in the study population and the four structure genes KIR3DL2,3DL3,3DP1and2DL4were present in all individuals. Common KIR genes in the study population included3DL1,2DL1,2DL3,2DS4and the pseudogene2DP1with an observed frequency of over90%. Less common (<40%) KIR genes included2DL2,2DL5,2DS1,2DS2,2DS3,2DS5and3DS1. Thirty unique KIR gene profiles were found reflecting a rather limited number of KIR haplotypes in this population. KIRAA1was the most common profile observed in54.7%of the samples. Among the AA1individuals,15.6%were homozygous for the deleted KIR2DS4which implied they had no any activating KIR framework gene. KIR genes are inherited by the form of haplotype A or B. Haplotype A (74.8%) was more common than haplotype B. HLA-C1was a much more common ligand for2D KIRs than C2. Bw4-80I, Bw4-80T and the Bw4-bearing HLA-A alleles were detected at similar frequencies (about32%). The matched KIR+HLA pairs2DL2/3+C1 (98.1%),3DL1+Bw4(73.3%),3DL2+A3/11(60.0%) KIR2DL2/3+C1(98.1%) were the most common ones whereas3DS1+Bw4-80I was the least common (9.4%). A total of193unique compound profiles of KIR-HLA were identified in the study population,130of the profiles being detected only once. The polymorphism of KIR-HLA compound profiles in southern Chinese Han can serve as a good reference for KIR/HLA-associated studies on transplantation, leukemia and so on.Secondly, based on the obtained KIR population genetic data in southern Chinese Han population, the associations KIR genetic polymorphism with acute lymphoid leukemia (ALL), acute myeloid leukemia (AML), chronic myeloid leukemia (CML) were explored at four levels, which aimed to present potential new clue for the pathogenesis of leukemia and to provide theoretical basis for clinical individual immune therapy.In ALL patient group, the frequencies of KIR2DL3, HLA-C1ligands and2DL3+HLA-C1matched pair were all significantly lower than the healthy control group, which showed protection effect against ALL (KIR2DL3:OR=0.302,95%CI=0.115-0.794, P=0.010; HLA-C1:OR=0.288,95%CI=0.117-0.708, P=0.004;2DL3+HLA-C1:OR=0.463,95%CI=0.220-0.974, P=0.038). Of the three inhibitory KIR receptors that bind HLA-C, the inhibitory signal mediated by KIR2DL3+HLA-C1pair is the weakest, we speculated that KIR2DL3+NK cells from KIR2DL3+HLA-C1individuals are more easily activated and preferably develop immune elimination. However, the frequencies of linkage-inherited KIR2DS3-2DL5,3DS1+Bw4-80T pair were all significantly higher in ALL patient group compared with the healthy control group, which seemed to be susceptible factors to ALL (KIR2DS3:OR=1.467, CI=1.038-2.703, P=0.030; KIR2DL5:OR=1.548,95%CI=1.033-2.320, P=0.033;3DS1+Bw4-80T:OR=1.731,95%CI=1.046-2.865, P=0.031). KIR2DS3and KIR2DL5are in complete linkage disequilibrium, but the mechanism how they work as susceptible factors to ALL is subjected to further study due to their unknown ligands.In AML patient group, the frequencies of KIR framework genes and KIR gene profiles, class I HLA ligands, matched KIR+HLA pairs and KIR-HLA compound profiles showed no significant difference compared with the healthy control group.In CML patient group, the frequencies of HLA-C2ligand and2DL1+HLA-C2pair and HLA-B Bw4-80I were all significantly lower than the healthy control group, which showed a protective effect against CML (HLA-C2:OR=0.386,95%CI=0.240-0.620, P=0.000;2DL1+HLA-C2:OR=0.316,95%CI=0.191-0.525, P=0.000; HLA-B Bw4-80I:OR=0.576,95%CI=0.384-0.862, P=0.007). Of the three inhibitory KIR2DL1-3that bind HLA-C1or HLA-C2ligands, the inhibitory signal mediated by KIR2DL1+HLA-C2pair is the strongest. The frequency of KIR2DL1ligand-HLA-C2was significantly lower than the healthy control group, we speculated that the HLA-C2expression probability decreased in CML patient group, which led to the relative reduced inhibitory signal and the relative enhanced activating signal of KIR2DL1+NK cells and facilitated the clearance of leukemic cells by KIR2DL1+NK cells. Similarly, the KIR3DL1ligand-HLA-B Bw4-80I expression probability decreased in CML patient group, which result in the reduced inhibitory signal of NK cells, and the NK cells are activated to kill the leukemic cells. Noticeably, we found the frequency of KIR-HLA compound profiles ID2(KIR3DL1-2DL1-2DL3-2DS4-2DL4-3DL2-3DL3-2DP1-3DP1-HLA-C1/C1-Bw6/Bw6-A3/11) in CML patients significantly increased in the CML patient group as compared to the healthy control group, the KIR-HLA compound profiles ID2was a risk factor for CML (OR=2.163,95%CI=1.198-3.906, P=0.009). The reduced expression or loss of HLA class I alleles may provide leukemic cells with an escape mechanism only from cytotoxic T cells and NK cells. HLA-B Bw6is not a ligand for any KIR. HLA-B Bw6was more frequently downregulated than HLA-B Bw4. All of KIR-HLA compound profiles ID2individuals are homozygous for HLA-B Bw6and downregulation of HLA-Bw6may provide leukemic cells with an escape mechanism only from cytotoxic T cells. Considering KIR framework gene profiles and their known class I HLA ligands, both2DL3+HLA-C1and3DL2-HLA-A3/11pairs mediate inhibitory signal. As for the only activating KIR2DS4in KIR-HLA compound profiles ID2individuals, we cannot ensure the presence or absence of activating signal mediated by KIR2DS4-ligand because the ligand for KIR2DS4is unknown. Due to the complexity of NK cell activity regulation, it is hard to predict the NK cell activity in KIR-HLA compound profiles ID2individuals. Based on the conclusion that KIR-HLA compound profiles ID2was a risk factor for CML, we speculated that there were three potential mechanisms:firstly, the CTL effect was suppressed; secondly, NK cell activity was suppressed, which could be verified by NK cell function experiment; lastly, both CTL effect and NK cell activity were suppressed or other mechanism was also involved, but the mechanisms was subjected to further study.Finally, as KIR2DL3and2DL3+HLA-C1pair showed protective effect against ALL which was described in the above paragraph, we established KIR2DL3gene sequence-based typing (SBT) method for the first time and further analyzed the association of KIR2DL3allelic polymorphism with ALL leukemia.As a result, fifteen KIR2DL3alleles were identified in the healthy control group by SBT at the entire coding sequence of KIR2DL3gene. KIR2DL3*00101was the most common alleles with the frequency of71.8%, followed by KIR2DL3*00201(13.1%).8alleles were found in ALL patients. KIR2DL3*00101,*00201,*015and*023were found in both groups, and the frequencies of KIR2DL3*00101,*00201and*023were more than1%, respectively. In the study, a total of4KIR2DL3novel alleles have been officially assigned by WHO Nomenclature Committee for Factors of the HLA System-KIR Subcommittee, in which one of them was generated by a synonymous mutation within the coding sequence while the others were all generated by a non-synonymous mutation within the encoding sequence. The novel allele KIR2DL3*00109was closest to KIR2DL3*00101and differed from KIR2DL3*00101with one synonymous mutation at position CDS nt478C> T in exon5; KIR2DL3*030was closest to KIR2DL3*00101and differed from KIR2DL3*00101with one non-synonymous mutation at position CDS nt201G>C in exon4, which led to the amino acid substitution (Lys> Asn) at position139; KIR2DL3*031was closest to KIR2DL3*00101and differed from KIR2DL3*00101with one synonymous mutation at position CDS nt735T>C in exon7and one non-synonymous mutation at position CDS736G>A in exon7, which led to the amino acid substitution (Val>Ile) at position225; KIR2DL3*032was closest to KIR2DL3*00101and differed from KIR2DL3*00101with one non-synonymous mutation at position CDS nt754G> C in exon7, which led to the amino acid substitution (Val> Leu) at position754.Compared the frequencies of KIR2DL3alleles in the ALL patient group with the healthy control, the frequency of KIR2DL3*00201was significantly decreased (OR=0.606,95%CI=0.375-0.980, P=0.040). Among all the KIR2DL3alleles, it has been demonstrated that KIR2DL3*00201showed lower expression on the NK cells surface and mediated the weakest affinity as well as inhibitory capacity with its ligand HLA-C1, we speculated this would lead to the activation of KIR2DL3*00201+NK cells killing the tumor cells.In this paper, the KIR2DL3gene sequence-based typing method established by us is owned by the following characterizations:firstly, the PCR amplification and sequencing reaction can be run simultaneously under the same PCR parameters, PCR product purified by the paramagnetic beads is suitable for high-throughput sequencing. Secondly, the entire KIR2DL3coding sequence was sequenced by both directions, which greatly reduced the ambiguous KIR allele combinations. Thirdly, the KIR2DL3sequences obtained by using our SBT assay showed no background, the sequences imported into the Assign3.5SBT software, the assignment of KIR2DL3genotype can be accomplished easily. Finally, the PCR amplification results using our KIR2DL3-specific PCR primers are completely in accordance with the outcome identified by the imported KIR PCR-SSP commercial Kit, which indicated the KIR2DL3PCR primers designed by us can also be used for detecting the presence or absence of KIR2DL3framework gene. The2DL3SBT assay established by us showed a great value for the manufacture of KIR SBT commercial kit and abroad applications in clinical and immungenetic area.