MICA*049, not MICA*009, is associated with Behçet’s disease in a Chinese population

Behçet’s disease (BD) is a multi-systemic inflammatory disease. Previous reports indicated that MICA*009 confers susceptibility to BD. MICA*049 differs from MICA*009:01, a major MICA*009 subtype, only at codon 335 in exon 6. However, the potential association of MICA*049 with BD has not been addressed. In this study, we differentiated association among MICA*049, MICA*009 and HLA-B*51 with BD. A Han Chinese cohort consisting of 41 BD patients and 197 ethnically matched controls were examined with sequencing and T-ARMS-PCR for genotyping of MICA, and ARMS-PCR for HLA-B*51. The phenotype frequency of MICA*049 (41.5% versus 8.1%, OR = 8.01, P = 1.91 × 10−8) and HLA-B*51 (46.3% versus 15.7%, OR = 4.62, P = 1.21 × 10−5) were significantly higher in BD patients than those in controls, whereas MICA*009 showed no significant difference between the two groups (17.1% versus 13.2%, OR = 1.35, P = 0.51). After stratification for the effect of HLA-B*51, MICA*049 was still associated with BD in HLA-B*51 negative patients (OR = 40.61, P = 0.02). Our results indicate that MICA*049, not MICA*009, is a risk factor to BD, and that is independent from HLA-B*51 in the Han Chinese cohort.


Results
The frequencies of MICA alleles in the 41 BD patients and 197 healthy controls were shown in Table 1. There were 8 different MICA alleles in patients and 16 in controls. The frequency of MICA*049 was significantly higher in the patient group (24.4% in BD versus 4.3% in control, OR = 38. 16, P = 6.52 × 10 −10 ). However, the frequency of MICA*009 (including MICA*009:01 and MICA*009:02) was similar between the two groups (8.5% versus 6.6%, OR = 1.32, P = 0.53).
The MICA allele phenotype frequencies in BD patients and controls were shown in Table 2. The MICA*049 was significantly increased in BD patients compared to that in controls (41.5% versus 8.1%, OR = 8.01, P = 1.91 × 10 −8 ). The difference of the MICA*009 frequency between patients and controls was not significant (17.1% in BD versus 13.2% in control, OR = 1.35, P = 0.51). The allele frequency of the MICA*A6 was significantly higher in BD patients than that in controls (32.9% versus 11.7%, OR = 3.71, P = 1.18 × 10 −6 ). The result of phenotype frequency was consistent with that of allele frequency (53.7% versus 21.8%, OR = 4.15, P = 3.16 × 10 −5 ).
The presence of HLA-B*51 in BD patients and controls were 46.3% and 15.7% (OR = 4.62, P = 1.21 × 10 −5 ), respectively (Table 3).  www.nature.com/scientificreports www.nature.com/scientificreports/ To examine whether the observed BD association of MICA*049 and HLA-B*51 are independent from each other, we performed subclonal analysis in HLA-B*51 negative subjects for MICA*049, and in MICA*049 negative subjects for HLA-B*51. As shown in Table 4, the MICA*049 remained significantly associated with BD (OR = 40.61, P = 0.02) in HLA-B*51 negative BD patients, but the association of HLA-B*51 with BD appeared lost in MICA*049 negative patients (Table 5).

Discussion
Previously, MICA*009 and MICA*A6 were suggested as susceptibility alleles for BD. The MICA*A6 is a polymorphism with 6 tendent repeats of GCT in exon 5 of MICA gene. This polymorphism is included in the MICA*009, and shared by MICA*049 and a number of other MICA alleles. In the previous studies 5, [8][9][10][11] , the MICA alleles were identified by PCR-SSP or PCR-SBT based on sequences of exon 2 to exon 5. However, the MICA*00901 and the MICA*049 differ by only one nucleotide at codon 335 of exon 6. Therefore, the ambiguity between these two alleles could not be addressed, and the MICA*009 allele reported in the previous studies may be mixed with MICA*049. According to allelic functional analysis using SIFT program (http://sift.bii.a-star.edu.sg/), the change at codon 335 may impact MICA function.
In the present study, we developed a rapid and cost-efficient T-ARMS-PCR to discriminate the MICA*009 from the MICA*049. Comparison analysis between BD patients and controls showed that the MICA*049, not *009, was strongly associated with BD. As we expected, the MICA*A6 showed a consistent BD association with previous reports as it is within the MICA*049 polymorphism. It is worth noting that the allele frequency of the MICA*009 and *049 in controls were consistent with the previous report of MICA alleles in a Chinese population 12 . Considering MICA and HLA-B genes are located next to each other, and strong linkage disequilibrium (LD) exists between alleles of these two genes, it is necessary to determine whether the observed association is due to LD effect from HLA-B*51. According to the clonal analysis, the MICA*049 was independently associated with BD in the Chinese cohort.
In conclusion, we investigated MICA polymorphisms in patients with BD of Chinese Han. It is the first report of MICA*049 in association with BD, and which appeared independent from HLA-B*51. Although the sample size is relatively small in the study, the association achieved significant p value with strong odd ratio. However, it still warrants further validation studies in a larger Chinese cohort and/or other ethnic populations. It may not rule out this observed association is ethnic specific for Chinese Han population.

Participants.
A total of 41 Patients (34 male, 7 female) were enrolled between March 2010 and September 2017 from the Eye Hospital of Wenzhou Medical University. The diagnosis of BD was followed the criteria of the International Study Group of BD 13 . The mean age of the patients was 37.8 years (range between 27-50 years) and the mean duration of the disease was 6.4 years (range between 1-18 years). A total of 197 unrelated healthy     www.nature.com/scientificreports www.nature.com/scientificreports/ individuals were recruited in the same geography. All of patients and controls were Chinese Han. The study was approved by the Ethics Committee of the Eye Hospital of Wenzhou Medical University and was conducted according to the Declaration of Helsinki Principles. Written inform consent was obtained from all participants.
Genomic DNA extraction. Genomic DNA was extracted from peripheral blood cells of all subjects using Bioteke DNA isolation kit (Beijing, China). After detecting DNA concentration by a Nanodrop 2000 spectrophotometer, a part of DNA of each subject was diluted to 10 ng/μl for genotyping assays.  Table 6. Product sizes were 246 bp for T allele, 182 bp for C allele, and 382 bp for the forward outer primer and reverse outer primer. The PCR was performed in a final volume of 10 μl containing 5 μl of 2 × Hot-start Taq  Statistical analysis. HLA-B*51 and MICA allelic frequencies were calculated by direct counting. The significance of the distribution of alleles between the patient group and the control group was calculated by Chi-square or Fisher's exact test using SPSS22.0 or Epi info software. If the cell frequency as zero, the odds ratio (OR) was calculated using MedCalc software (https://www.medcalc.org/calc/odds_ratio.php).

Data Availability
The data generated and/or analyzed in the current study are available from the corresponding authors on reasonable request.