Introduction

Behçet’s disease (BD) is a multi-system inflammatory disease characterized by recurrent oral and genital ulcers, uveitis, and skin lesions. Although the etiology and pathogenesis of BD are still uncertain, multiple genetic factors have been linked to BD1,2. Among them, HLA-B*51 appears to be the most strongly associated known genetic risk to BD in different ethnic groups2.

MICA (major histocompatibility complex class I chain related gene A), located only 46 kb centromeric of HLA-B, is a highly polymorphic gene. It normally expresses on the cell membrane, and functions in immune activation under cellular stress conditions, such as infections, tissue injury, pro-inflammatory signals, and malignant transformation2. The MICA transmembrane (TM) A6 allele and the MICA*009 allele were associated with BD in multiple previous reports3,4,5,6,7,8,9,10,11. According to updated IMGT/HLA database, there are 107 MICA alleles identified. The MICA*009 can be further subtyped into MICA*00901, MICA*0090201 and MICA*0090202. The only difference between the MICA*00901 and the MICA*049 is at codon 335 in exon 6 (https://www.ebi.ac.uk/ipd/imgt/hla/align.html). In the previous studies5,8,9,10,11, the ambiguity between the MICA*009 allele and the MICA*049 allele was not addressed, because exon 6 was not studied. Therefore, the MICA*009 allele maybe mixed with the MICA*049 allele. Here, we examined the association between MICA and BD in a Han Chinese cohort with MICA sequencing approach, along with a simple tetra-primer amplification refractory mutation system-polymerase chain reaction (T-ARMS-PCR) method to discriminate between the MICA*00901 allele and the MICA*049 allele.

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).

Table 1 Comparison of MICA alleles between BD patients and controls.
Full size table

The genotype frequencies of MICA*008(:01 or:04)/MICA*049, MICA*010:01/MICA*049 and MICA*049/MICA*049 were significantly higher in the patients (see Supplementary Table S1).

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).

Table 2 Phenotype frequencies of MICA alleles in BD patients and controls.
Full size table

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).

Table 3 Association of BD with HLA-B*51.
Full size table

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).

Table 4 Association of MICA*049 with BD stratified for the effect of HLA-B*51.
Full size table
Table 5 Association of HLA-B*51 with BD stratified for the effect of MICA*049.
Full size table

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 studies5,8,9,10,11, 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 population12. 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.

Methods

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 BD13. 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 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.

HLA-B*51 genotyping

For control samples, the HLA-B*51 genotyping was performed with sequence-based typing (SBT) method using secore kits (Life Technologies, USA)14. For patients, each sample was genotyped for HLA-B*51 positivity by ARMS PCR method15.

MICA genotyping

MICA was genotyped by PCR sequencing exon 2–5 regions using bidirectional Sanger sequencing methods16. For samples in patient group which were discriminated as MICA*009:01/*049, Sanger sequencing was used to distinguished the two alleles. Two primers (Forward primer: 5′-AGAGAAAGGGCGAATCTGGT-3′, Reverse primer: 5′-AAGAGGGAAA-GTGCTCGTGA-3′) were used to amplify 301 bp PCR products. The PCR was performed in a total volume of 20 μl containing 10 μl of 2 × Taq Master Mix (Jinan, Shanghai, China), 0.4 μM of each primer (Invitrogen, Shanghai, China) and 10 ng of genomic DNA. PCR was carried out on a Veriti Thermal cycler. The PCR thermal cycling condition was an initial denaturation at 94 °C for 3 min, followed by 35 cycles at 94 °C for 20 s, 60 °C for 20 s and 72 °C for 20 s, and a final extension at 72 °C for 5 min. For samples in control group which were detected as MICA*009:01/*049, T-ARMS-PCR was used to differentiate MICA*009:01 from MICA*049. The sequence of the four primers and concentration of each primer were listed in 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 Red Master Mix (PHENIX, CA, USA), 0.4–1.4 μM of each primer (IDT, Skokie, USA) and 10 ng of genomic DNA. The PCR program on the Veriti Thermal cycler was as follow: 95 °C for 10 min; 35 cycles of 20 s at 94 °C, 30 s at 62 °C and 25 s at 72 °C, followed by a final extension of 5 min at 72 °C. The results of T-ARMS-PCR were verified by DNA sequencing. Direct sequencing was done with the two outer primers. The PCR products were purified using DNA clean & concentrator kit (Irvine, CA, USA), then the purified PCR products were sent to company for sequencing (GENEWIZ, NJ, USA). The sequencing data were analyzed using chromas software.

Table 6 Primers for the T-ARMS-PCR to distinguish MICA*009:01 from MICA*049.
Full size table

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).