Genetic Diversity Of Eleven Polymorphic Loci Within HLA Gene Complex And Association With Nasopharyngeal Carcinoma

Nasopharyngeal carcinoma (NPC) is an epithelial malignancy with the highest incidence rate in populations of southern Chinese ancestry. Males are more prone to develop NPC. The human leukocyte antigen gene complex (HLA) encompasses a 4 Mb genomic segment on chromosome 6p21.3. Previous studies have identified the associations of HLA-B^*35, B^*38, B^*46 and B^*58 with NPC in Chinese populations.MHC class I chain-related gene A(MICA) maps 46 kb centromeric to HLA-B locus, MICA molecules activate natural killer (NK) cells,γδT cells and CD8⁺αβT cells to eliminate virally infected, transformed cells.In the first part of our study, 218 unrelated, consecutive patients newly diagnosed with pathologically confirmed NPC and 196 unrelated, consecutive normal controls recruited from Hunan province, southern China , were investigated for genetic polymorphism of (GCT)n short tandem repeat sequence of MICA exon 5(MICA-STR) and MICA gene deletion, using fluorescent polymerase chain reaction-size sequencing (PCR/Size-sequencing)and polymerase chain reaction- sequence specific priming (PCR/SSP)technology. MICA^*A9 was present at significantly increased frequency in the patient group (P_c=0.0001002, OR=2.528, 95% CI=1.636-3.907), whereas the frequency of MICA*A5.1 was significantly decreased (Pc=0.006, OR=0.594, 95% CI=0.437-0.806). Gender-based stratification revealed a significant increase of MICA*A9 frequency (Pc=0.000072, OR=3.255, 95% CI=1.855-5.709)and a significant decrease of MICA*A5.1 frequency (Pc=0.000737, OR=0.486, 95% CI=0.337-0.702) in male patients with NPC. A significant interaction between MICA*A9 and gender was observed (X_w²=41.58, P=0.0001). Heterogeneity of effects among MICA*A5.1, MICA*A9-bearing phenotypes and a dose-dependent effect of MICA*A5.1 and MICA*A9 on NPC risk was observed in males.We then investigated the haplotypic diversity and linkage disequilibrium (LD)between MICA-STR and HLA-B in this population. Fifty-eight randomly selected normal nuclear families with 183 members including 85 unrelated parental samples were analysed. HLA-B typing was performed by PCR/SSP, MICA-STR allelic variation and MICA gene deletion (MICA*Del) were detected by fluorescent polymerase chain reaction-size sequencing and PCR/SSP, respectively. Haplotype was determined through family segregation analysis. Statistical analysis was applied to the data of the 85 unrelated parental samples. The 170 parental HLA-B-MICA-STR haplotypes contained 30 kinds of HLA-B, MICA-STR haplotypic combinations, 9 of them have not been reported in the literature. MICA*A9 was linked to HLA-B*35, B*38, B*51 and B*58, and was in significant LD with HLA-B*38 and HLA-B*58 (A=0.0364, Pc=2.3558*10^-4 andΔ=0.0409, Pc=2.9830*10^-4, respectively). MICA*A5.1 was linked to HLA-B*07, B*08, B*13, B*27, B*37, B*39, B*48 and B60, and was in significant LD with HLA-B60 (A=0.1341, pc=1.6866*10^-11).In the third part of our work aimed to identify NPC-associated HLA haplotype(s), we conducted a case-control study which included 127 unrelated, consecutive patients with NPC and 112 unrelated,consecutive normal controls, both of which were used in the 1st part of our work. HLA-A, B, and DRB1 loci were genotyped by PCR/SSP. Our data showed that in addition to MICA*A9, HLA-B*58 and HLA-DR17(DRB*03) were significantly over-represented in the male patients with NPC than in the male controls (OR=9.975, 95%CI=2.637-37.73, Pc=0.0034 and OR= 12.391, 95%CI=2.05-75.01, Pc=0.0195, respectively). HLA-B60 was negatively associated with NPC in males with borderline significance (OR=0.416, 95%CI=0.22-0.787, Pc=0.055). In contrast, none of the HLA-A, B, and DRB1 genes displayed any frequency variation that was statistically significant between female case and female control subgroups. Multi-locus analysis revealed that HLA-B*58-MICA*A9-DR17(DRB*03) haplotypic segment was strongly linked to NPC in males(OR=12.391, 95%CI=2.05-75.01, P=0.00154). HLA-B*58-MICA*A9-non-DR17(DRB*03) haplotypic fragment displayed a much attenuated association with NPC in males (OR=5.734, 95%CI=0.54-60.88, P=0.06). we did not observe any significant frequency increase of HLA-A*02, B*46 in the NPC patients in the global comparison or after gender-based stratification.In the 4th part of our work, we investigated the associations of seven polymorphic markers, which reside in the telomeric side of HLA class I region and flank HLA-A locus, with NPC in this population, using the same sample panel as in the third part of our work. Fluorescent polymerase chain reaction-size sequencing was used to analyse the microsatellites D6S1624, D6S258, M6S211 and D6S510, polymerase chain reaction-amplification fragment length polymorphism (PCR/AFLP)technique was used to examine the insertion/deletion of AluyI-IF, AluyHG and AluyHJ. We did not find any frequency variation in the seven markers that was statistically different between NPC patients and normal controls in the overall or gender-based comparisons.Our study constitutes the first demonstration of gender-specific associations between MICA*A5.1, MICA*A9 and NPC in Hunan Han population. The exclusive linkage of multiple NPC-associated HLA-B genes with MICA*A9 indicate that MICA*A9 is probably a primary NPC-associated genetic variant for this population and could also be a genetic risk factor for NPC in other Chinese populations. Our data suggest that HLA-B*58-MICA*A9-DR17(DRB1*03) forms a male-specific haplotype strongly associated with NPC in this population and harbors additional genetic defects carried by other loci of this haplotype, which interact synergistically with MICA*A9. Further investigation of genetic defects in candidate loci on this haplotype, especially IKBL, 21-OHB loci residing at the central HLA classâ…¢region, will undoubtedly facilitate the understanding of interaction among NPC predisposing factors and shed new insights into the mechanisms underlying the uniquely high incidence and prevalence of NPC for males in southern Chinese populations. We did not observe any association between the genetic polymorphisms of seven markers located in the telomeric side of HLA class I region and NPC in this population, and four possibilities regarding the relationship between genetic variation in this subregion and NPC were proposed.