Genetic Variations of NLR family genes in Behcet’s Disease

This study aimed to investigate whether single nucleotide polymorphisms (SNPs) of five NLR family genes (NOD1, NOD2, NLRP1, NLRP3 and CIITA) are associated with Behcet’s disease (BD) in a Chinese Han population. The study was carried out in 950 BD patients and 1440 controls for 19 SNPs in the selected NLR genes. In the first-stage study, significantly decreased frequencies of the CIITA//rs12932187 C allele (Pc = 1.668E-02) and NOD1//rs2075818 G allele (Pc = 4.694E-02) were found in BD patients as compared to controls . After performing a second stage validation study and combination of data we confirmed the association of CIITA//rs12932187 and NOD1//rs2075818 with BD. In CIITA//rs12932187, the frequencies of the CC genotype and C allele were significantly lower in BD than in controls (Pc = 3.331E-06; Pc = 6.004E-07, respectively). In NOD1//rs2075818, the GG genotype and G allele showed significantly decreased frequencies in BD patients when compared to controls (Pc = 1.022E-02; Pc = 6.811E-05, respectively). Functional experiments showed that carriers with the CC genotype in CIITA//rs12932187 had a lower CIITA mRNA expression level and an enhanced IL-10 secretion as compared to GG and CG carriers. This study provides evidence that the CIITA and NOD1 gene are involved in the susceptibility to Behcet’s disease.

Genotype Results. In first stage study, the frequency of the CIITA//rs12932187 C allele (Pc = 1.668 × 10 −2 , OR = 0.713, 95% CI = 0.591-0.861) and NOD1//rs2075818 G allele (Pc = 4.694E-02, OR = 0.698, 95% CI = 0.562-0.868) were decreased in BD patients compared to controls ( Table 2). The other seventeen SNPs did not show a significant association with BD (Supplementary Table S1). In the second stage, we tested another set of 566 BD patients and 870 healthy controls to confirm the result of the first stage study. After combination of the data, the frequencies of CC genotype and C allele in CIITA//rs12932187, were significantly decreased in the BD patients (Pc = 3.331 × 10 −6 , OR = 0.617, 95% CI = 0.519-0.735; Pc = 6.004 × 10 −7 , OR = 0.709, 95% CI = 0.629-0.799, respectively). In NOD1//rs2075818, the frequencies of the GG genotype and G allele were also decreased in the BD patients (Pc = 1.022E-02, OR = 0.536, 95% CI = 0.386-0.745; Pc = 6.811 × 10 −5 , OR = 0.720, 95% CI = 0.629-0.824, respectively). mRNA level and downstream inflammatory factors. Because of the significant association of CIITA// rs12932187 and NOD1//rs2075818 with BD, we tested the expression of NOD1 and CIITA in PBMCs obtained from healthy individuals with known genotypes of the two SNPs. Real-time PCR did not show a detectable association between the various genotypes and the expression of NOD1 and CIITA when testing unstimulated PBMCs (Supplementary Fig. S1 and Supplementary Fig. S3). Following stimulation by LPS, carriers with the CC genotype in CIITA//rs12932187 had a lower mRNA expression of CIITA compared with the GG or CG genotype carriers (P = 0.004, Fig. 1). anti-CD3/anti-CD28 stimulation did not affect CIITA expression ( Supplementary Fig. S2) and no effect on NOD1 mRNA expression was observed for the various rs2075818 genotypes by either normal or stimulated PBMCs (Supplementary Fig. S4 and Supplementary Fig. S5).
Since the different genotypes of CIITA//rs12932187 had an effect on CIITA mRNA expression, we decided to investigate whether the different genotypes influenced the cytokine response of PBMCs following LPS stimulation. We measured the PBMC expression levels of IL-6, IL-8, IL-10, IL-1β , TNF-α and MCP-1 by ELISA. These cytokines have all been shown to play a role in the development of BD as shown by earlier studies 22,23 . Carriers of the CC genotype had a higher secretion level of IL-10 as compared to GG and CG carriers ( P = 0.017, Fig. 2). No significant effect on secretion levels of IL-6, IL-8, IL-1β , TNF-α and MCP-1 was found ( Supplementary  Fig. S6-S10).

Discussion
In the present study, we investigated the associations of 19 SNPs in NOD1, NOD2, NLRP1, NLRP3 and CIITA with BD in a Chinese Han population. Two SNPs, rs12932187 of CIITA and rs2075818 of NOD1 contributed to the genetic susceptibility of BD. Functional studies showed that carriers of the CC genotype of CIITA// rs12932187 had a lower CIITA mRNA expression level and an increased IL-10 secretion by PBMCs as compared to GG and CG carriers.
CIITA acts as a transcriptional coactivator and has been associated with various inflammatory and autoimmune diseases 24,25 . CIITA mediates activated immune responses and its deficiency has been shown to cause Type II Bare lymphocyte syndrome (BLS) 26 . Variability in the CIITA gene has also been reported to be associated with several autoimmune and inflammatory diseases such as myocardial infarction (MI), rheumatoid arthritis (RA), type I diabetes (T1D) and multiple sclerosis (MS) 23,24 . A case-control study was performed in 1320 MS cases and 1363 independent healthy controls and the results showed that CIITA//rs4774 was associated with MS, particularly in DRB1*1501(+ ) cases 27 . Another study showed that the two SNPs rs4774 and rs6498122 of CIITA were associated with oral lichen planus (OLP) 28 . CIITA//rs8048002 was found to be associated with RA in a Swedish cohort 29 . CIITA//rs12932187 and rs11074938 were found to be susceptibility markers of nasal passages inflammation in asthma patients in a Japanese population 30 . In our study, only the CIITA//rs12932187 G allele and GG genotype were associated with BD risk. CIITA has been shown to function not only as a transcriptional regulator of MHC genes, but is also a transcriptional regulator of over 60 immunologically important genes, including IL-4, IL-10 and a number of thyroid-specific genes 24,25 . A study in CIITA-deficient (CIITA(−/−)) mice showed that CIITA negatively regulates the expression of IL-10 by DCs, which supports the findings in humans as presented here 31 . IL-10 is considered an immune regulatory cytokine which controls innate and adaptive immune responses 32 . Low IL-10 serum levels have been reported in Asian patients with BD 33 . The functional tests we performed showed that carriers with the CC genotype of CIITA//rs12932187 had a lower CIITA mRNA expression level and an enhanced IL-10 secretion as compared to GG and CG carriers. The protective effect of the C allele and CC genotype concerning BD development could thus be explained by the fact that these individuals produce more anti-inflammatory IL-10 in response to a microbial stimulus than carriers of the G allele. Further studies are needed to support this hypothesis. NOD1 has been characterized as a critical regulator of innate immunity. Various studies have reported the association between NOD1 gene variants and autoimmune disease 10 . The NOD1//rs2075818 G allele was found to decrease the risk of CD and rs2907748 AA and AG genotypes showed a decreased frequency in UC 13 . These findings are in agreement with our study and could be due to the fact that BD as well as these inflammatory bowel disease are considered as an autoinflammatory disease caused by an aberrant response to a microbial agent. We were not able to detect a functional explanation for the association with NOD1//rs2075818. NLRP1 and NLRP3 have been shown to play an important role in the processing of pivotal pro-inflammatory cytokines such as IL-1β and IL-18 14,16 . Gene variants of NLRP3 have been shown to be associated with Psoriatic Juvenile Idiopathic Arthritis in a Caucasian population 34 . Moreover, genetic variants of NLRP1 were observed to confer risk for the development of vitiligo 35 . Nevertheless, our study did not find an association between NLRP1 and NLRP3 and BD in a Chinese Han population. NOD2, that was already identified as a CD-susceptibility gene 11 , was not associated with BD.
Our study has some limitations. Since the subjects in our study were all Chinese Han, the conclusions of the study are only valid to the Chinese Han population and should be studied and replicated in other ethnic groups. Furthermore, all the BD patients in this study were recruited from ophthalmology departments and a selection bias in our patient population may be present. Whether our findings can be generalized to other uveitis entities is not known and deserves further study. We did test the SNPs described in this study on uveitis patients with Vogt Koyanagi Harada syndrome but did not observe statistically significant associations (data not shown).
In conclusion, this study for the first time reports an association of CIITA//rs12932187 and NOD1//rs2075818 with susceptibility to BD in a Chinese Han population. A functional variant of CIITA//12932187 was shown to regulate CIITA expression and IL-10 production.

Materials and Methods
Study population. In the first stage of this study, a total of 384 BD patients and 576 controls were enrolled to identify disease susceptibility loci in the family of NLR genes. In the second (confirmatory) stage, another set of 566 BD patients and 864 controls were added to replicate the susceptible SNPs found in the first stage study.
DNA extraction and genotyping. Genomic DNA was extracted from the blood of patients and healthy individuals using the QIAmp DNA Blood Mini Kit (Qiagen Inc., Valencia, CA, USA), all the samples were stored at − 80 °C until used. All SNPs except NLRP3//rs10925019 were genotyped by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). All the primers were designed using primer software 5.0 (Premier Biosoft International, Palo Alto, CA, USA). Primers and restriction enzymes are shown in Table 3. NLRP3//rs10925019 (TaqMan assay ID: C_26646027_10) genotyping was performed on the Applied Biosystems 7500 Real-Time PCR system using TaqMan ® SNP assay (Applied Biosystems, CA, USA). The analysis was performed by TaqMan Genotyper Software. To verify the accuracy of genotyping, direct sequencing was carried out (Beijing Biomed Co. Ltd. China) using randomly selected samples (10% of all samples). The genotyping success rate was above 95%.
Real-time PCR. In this study, peripheral blood mononuclear cells (PBMCs) were obtained from healthy controls by Ficoll-Hypaque density-gradient centrifugation. Cells were stimulated with or without anti-CD3 (0.5ug/ml) and anti-CD28 antibodies (0.1ug/ml, eBioscience, San Diego, CA, USA) to analog antigen presentation or lipopolysaccharide (LPS, 5ug/ml, Fluka, Buchs, Switzerland) to analog an inflammatory signal for 72 hours at a density of 1 × 10 6 cells/ml. RNA was acquired from the cultured cells by TRIzol (Invitrogen), after reserve transcription (transcription kit, Takara Biotechnology Co. Ltd., Dalian, China.), mRNA expression of NOD1 gene (forward: 5′ TTGACCACCCTGAGTCTTGC 3′ , reserve: 5′ TCATTTTGGGTCAGCCACAG 3′ ) and CIITA gene (forward: 5′ TGAGGCTGTGTGCTTCTGAG 3′ , reserve: 5′ ACACTGTGAGCTGCCTTGG 3′ ) was measured by using real-time PCR equipment with a commercial dye kit (Applied Biosystems), β -Actin was selected as the internal reference gene and its expression was detected by the following primers: forward 5′-GGATGCAGAAGGAGATCACTG -3′ and reverse 5′-CGATCCACACGGAGTACTT-3′ . Data were normalized to mRNA beta-actin and expression levels were calculated by the 2 −△△ method. Statistical analysis. Differences in alleles and genotypes of all SNP variations were evaluated by the Fisher's exact test or X 2 test using SPSS (version 17.0; SPSS Inc, Chicago, IL). The p values were corrected with the Bonferroni correction method and a Pc < 0.05 was considered to be significant. The X 2 test was used to determine the Hardy-Weinberg equilibrium (HWE). The independent samples t test or nonparametric Mann-Whitney U test was used to compare CIITA, NOD1 and cytokine (IL-6, IL-8, IL-10, TNF-α , IL-1β and MCP-1) expression levels among three genotype groups.