Association of MICA and MICB polymorphisms with the susceptibility of leukemia in Korean patients

Hematologic malignancies including acute lymphocytic (lymphoblastic) leukemia (ALL), acute myeloid leukemia (AML), and myelodysplastic syndrome (MDS) are clonal diseases, which commonly manifest as a result of dysregulation of hematopoiesis due to genetic damage. The MHC class I chain-related genes A and B (MICA and MICB), which are located on the human chromosome 6, encode proteins with 3 extracellular domains, a transmembrane segment, and a carboxy-terminal cytoplasmic tail. The extracellular domains consist of α1, α2, and α3 chains that flexibly function as a ligand for Natural killer group 2D (NKG2D), γδ T cells, and αβ CD8 T cells. MICA polymorphisms have been identified in over 100 alleles. Over 40 alleles have been sequenced for exons 2–6 of MICB (IMGT/HLA database; http://www.ebi.ac.uk/ imgt/hla). The role of MICA and MICB in human disease is mainly defined by its allele associations with cancer, autoimmune disorders, and success rates of hematopoietic stem cell transplantation (HSCT). In this study, we analyzed the possible association of MICA and MICB polymorphisms with ALL, AML, and MDS. Our results also reveal a correlation between amino acid variants in MICA external domains and ALL, AML, and MDS. We randomly selected a total of 324 patients diagnosed with ALL, AML, and MDS who were treated at the Department of Internal Medicine, Catholic Blood and Marrow Transplantation Center at Seoul St. Mary’s Hospital, between November 2010 and April 2013, and enrolled 324 patients with ALL, AML, and MDS (53.7% male; age [mean ± SD] 43 ± 13 years) included 85 ALL (58.8% male; age 39 ± 13 years), 172 AML (47.7% male; age 43 ± 13 years), and 67 MDS (62.7% male; age 47 ± 13 years). Diagnoses were based on the criteria specified by World Health Organization. The study was performed in accordance with the Declaration of Helsinki. The patients with ALL, AML, and MDS and 200 ethnically matched non-leukemic controls signed appropriate informed consent to participate. The use of the material was reviewed and approved by the local institutional review board (IRB) of The Catholic University of Korea, with written informed consent obtained for all samples collected (IRB Number: SCMC07BR131). DNA was isolated from peripheral blood. We applied sequence-based typing (PCRSBT) as an authentication method to amplify the exons 2–5 of MICA. Genotyping was performed by PCR-SBT for MICA-129 alleles (A to G) in exon 3 at position 454 of the MICA gene. MICB primers were designed based on exons 2–4 for genotyping of MICB-sequence alleles. The χ test and Fisher’s exact test were used to determine statistical significance. P-values were corrected (Pc) using the Bonferroni’s method. The odds ratio (OR) was calculated using Haldane’s modification of Woolf’s method. Hardy-Weinberg equilibrium in controls was analyzed for each SNP using SNPStats on the website (http://bioinfo. iconcologia.net/snpstats/start.htm) (P > 0.05). In total patients, the frequencies of MICA*002:01 alleles and A9 alleles were higher than in controls (112 of 324 [34.6%] versus 42 of 200 [21.0%]; Pc= 0.012, OR= 2.0, 95% confidence interval [95% CI] 1.3–3.0, 118 of 324 [36.4%] versus 44 of 200 [22.0%]; Pc= 0.003, OR= 2.0, 95% CI 1.3–2.9). However, the A5.1 (67 of 324 [21.0%] versus 76 of 200 [38.0%]; Pc < 0.001, OR= 0.4, 95% CI

enrolled 324 patients with ALL, AML, and MDS (53.7% male; age [mean ± SD] 43 ± 13 years) included 85 ALL (58.8% male; age 39 ± 13 years), 172 AML (47.7% male; age 43 ± 13 years), and 67 MDS (62.7% male; age 47 ± 13 years). Diagnoses were based on the criteria specified by World Health Organization 6 . The study was performed in accordance with the Declaration of Helsinki. The patients with ALL, AML, and MDS and 200 ethnically matched non-leukemic controls signed appropriate informed consent to participate. The use of the material was reviewed and approved by the local institutional review board (IRB) of The Catholic University of Korea, with written informed consent obtained for all samples collected (IRB Number: SCMC07BR131). DNA was isolated from peripheral blood. We applied sequence-based typing (PCR-SBT) as an authentication method to amplify the exons 2-5 of MICA 7 . Genotyping was performed by PCR-SBT 4 for MICA-129 alleles (A to G) in exon 3 at position 454 of the MICA gene. MICB primers were designed based on exons 2-4 for genotyping of MICB-sequence alleles 8 . The χ 2 test and Fisher's exact test were used to determine statistical significance. P-values were corrected (P c ) using the Bonferroni's method. The odds ratio (OR) was calculated using Haldane's modification of Woolf's method. Hardy-Weinberg equilibrium in controls was analyzed for each SNP using SNPStats on the website (http://bioinfo. iconcologia.net/snpstats/start.htm) (P > 0.05).
In total patients, the frequencies of MICA*002:01 alleles and A9 alleles were higher than in controls (112 of (Table 1).
As an important ligand of human NKG2D, an activating cell surface receptor expressed on NK and T cells, the MICA protein is expressed on several tumors, especially epithelium-derived cancer cells 5 . Ligand binding of MICA to the NKG2D receptor activates the innate immune response by stimulating NK and γδT cells. Tumor cells expressing high levels of membrane MICA and other NKG2D ligands are rejected by NK and CD8αβT cells, thereby stimulating antitumor activity, while reduced expression of membrane MICA as well as increased soluble MICA in serum leads to inactivation of the NKG2D-mediated antitumor response 9 . Recently, it has been shown that soluble NKG2D ligands can activate NKG2D-dependent signaling, rather than lead to  inactivation 10 . It is assumed that the reduction of membrane MICA encoded by the MICA*002:01 and A9 alleles, by release or shedding from the cell surface, potentially reduces the immunogenic signals of tumor cells, resulting in these tumor cells becoming less detectable by NK and T cells.
NK cytotoxicity experiments have shown that this downregulation-resisting allele, MICA*008:01, is functionally relevant and may aid in the elimination of virusinfected cells. MICA*008:01 and A5.1 are the most common alleles in population studies 2 . We can infer that, at least to some extent, the allelic repertoire of MICA represents an evolutionary record of past pathogen-driven selection and that the protective role of MICA*008:01 and A5.1 casts a 'selective' effect on the distribution of this allele.
Although the MICA-129 val/val genotype and elevated sMICA serum levels post-HSCT are independently associated with the incidence of cGVHD (P = 0.002 and 0.001), regardless of the history of acute GVHD, the presence of MICA Abs before transplantation confers protection against cGVHD (P = 0.04) 4 . There is an inverse relationship between UC or cGVHD and total patients including ALL, suggesting that these genetic characteristics of MICA influence disease incidence.
With the typing of MICA*002:01, MICA*008:01, and MICB*005:03 alleles associated with ALL, AML, and MDS using previously developed microarrays 11,12 , it will be possible to predict within more short time whether they are at risk or not. Furthermore, the investigation of MIC alleles in individual tumor samples, at the level of protein expression, should be of further interest. Our results illustrate that MICA-NKG2D played a role in disease pathogenesis in the majority of patients in our cohort of cases of ALL, AML, and MDS, and further investigation into this signaling axis may reveal potent therapeutic targets.