Abstract
Objective. Interleukin 23 (IL-23) stimulates the differentiation of T helper 17 (Th17) cells, which are involved in the pathogenesis of ankylosing spondylitis (AS). Binding of IL-23 to the IL-23 receptor complex activates Janus kinases 2 and tyrosine kinase 2, which phosphorylate IL-23R and subsequently promote the transcription of the IL-17 gene. IL-12B encodes a p40 subunit common to IL-12 and IL-23. We evaluated the effects of IL-12B and IL-23R genotype on the occurrence and clinical features of AS.
Methods. A total of 362 patients with AS and 362 healthy controls were enrolled in the study. Genotypes of IL-12B A1188C (rs3212227) and IL-23R C2370A (rs10889677) were identified by polymerase chain reaction/restriction fragment-length polymorphism. Disease activity and functional status were assessed by Bath AS indices.
Results. Subjects carrying IL-12B CC [matched relative risk (RRm) 1.93, 95% CI 1.23–3.03] and IL-12B AC (RRm 1.73, 95% CI 1.21–2.46) genotypes had a significantly greater risk of developing AS than subjects with the IL-12B AA genotype. Subjects carrying both IL-12B CC and IL-23R AA genotypes also had a significantly higher risk (RRm 2.98, 95% CI 1.51–5.89) of developing AS compared to those with IL-12B AA and IL-23R CC/CA genotypes, and this interaction between IL-12B and IL-23R was significant. Patients with AS who had IL-12B CC and IL-12B AC genotypes had an obviously increased Bath Ankylosing Spondylitis Disease Activity Index score compared to those who carried the IL-12B AA genotype (4.3 vs 3.7).
Conclusion. The IL-12B A1188C genotype was associated with the development and disease severity of AS.
- ANKYLOSING SPONDYLITIS
- GENETIC POLYMORPHISM
- INTERLEUKIN-12B
- BATH ANKYLOSING SPONDYLITIS DISEASE ACTIVITY INDEX
Ankylosing spondylitis (AS) is an autoimmune disease in which the main clinical symptoms are sacroiliitis and extraarticular manifestations1. However, the pathogenesis of AS is presently unclear. HLA-B27 gene is strongly associated with AS, but accounts for only 16% of the genetic variability seen in AS patients2,3.
It has been suggested that T helper 17 (Th17) cells are involved in AS pathogenesis4, and are also thought to be involved in psoriasis, Crohn’s disease, and inflammatory arthritis5,6,7. Increased expression of serum interleukin 17 (IL-17) and IL-23 was observed in patients with AS compared to healthy controls8. IL-23 is an IL-12-related cytokine, which might stimulate the differentiation and proliferation of Th17 cells9. However, the effect of IL-12 on Th17 is controversial. IL-12 might be involved in pathogenesis, as it affects the expansion of IL-23R+CD4+ T cells, and preferentially stimulates these cells to secrete predominately IL-1710. Binding of IL-23 to the IL-23 receptor complex, which is composed of an IL-23R and IL-12Rß111, leads to the activation of Janus kinases 2 (Jak2) and tyrosine kinase 2 (Tyk2), which can phosphorylate IL-23R, which increases binding of the STAT proteins. The STAT are then phosphorylated by the Jaks, resulting in their translocation to the nucleus and subsequently transcriptional activation of IL-1711,12.
Formation and secretion of IL-12 requires that the IL-12p35 (IL-12α) and IL-12p40 (IL-12ß) subunit be bound to one another13. The IL-12B gene is located in the region of chromosome 5q31–3314. IL-12B encodes a common p40 subunit to IL-12 and IL-2315. It was also noted that IL-12 promotes the development of naive T cells, while IL-23 mediates chronic inflammation16. Several intronic polymorphisms and a polymorphic TaqI site (+16974 A/C) in the 3′ untranslated regions (3′-UTR) have been identified; however, no polymorphisms were found that resulted in amino acid substitutions17. Importantly, a single-nucleotide polymorphism (SNP) at position 1188 in the IL-12B gene 3′-UTR (rs3212227) causes an A to C transition and results in increased IL-12 secretion18. Subjects with the IL-12B A1188C CC genotype also have significantly higher IL-12 expression from lipopolysaccharide-stimulated peripheral blood mononuclear cells19.
A genomewide association study in a population of white European ancestry found that the IL-23R gene may also be involved in the development of AS20. The IL-23R gene is located on chromosome 1p3111. Several SNP of IL-23R have been identified, but the biological influence of these variants on the expression and functionality on IL-23R is currently unknown. A population study in Hungarians also revealed that a polymorphism (IL-23R C2370A) in the 3′-UTR (rs10889677) in IL-23 might be associated with the risk of AS21. However, the influence of carrying both IL-12B A1188C and IL-23R C2370A polymorphisms on the development of AS has not been evaluated in other populations.
Genetic factors affect not only AS development but also contribute to the severity of the disease22,23. Using a case-control study, we investigated the relationship of the IL-12B and IL-23R genotypes with the development and clinical characteristics of AS in Taiwanese patients.
MATERIALS AND METHODS
Study subjects
Our study recruited 362 patients with AS from the arthritis clinic of Chung Shan Medical University Hospital (CSMU; Taichung, Taiwan). All patients were diagnosed using the modified New York criteria, by qualified rheumatologists24. Patients were ≥ 18 years of age and gave their informed consent to participate.
At the beginning of our study, physicians recorded detailed clinical histories, including age at initial symptoms, medication history, family history of AS, and extraspinal manifestations. Age at initial symptom was defined as the time when the first symptom developed, whether it was an axial symptom, peripheral arthritis, uveitis, or enthesitis. Delayed diagnosis was defined as the interval between the onset of first symptom and the correct diagnosis of AS. The use of nonsteroidal antiinflammatory drugs and disease-modifying antirheumatic drugs was defined as administration for > 3 months. Family history was defined as first-degree relative with AS, inflammatory bowel disease (IBD), reactive arthritis, or psoriatic arthritis. Peripheral arthritis was diagnosed if there was at least 1 swollen joint. IBD was defined as presence of the inflammatory condition of the colon and small intestine, including ulcerative colitis and Crohn’s disease. Uveitis was defined as the presence of inflammation of the middle layer of the eye involving unilateral, bilateral, or alternative patterns. The above symptoms were ascertained by a rheumatologist, ophthalmologist, and gastroenterologist.
Healthy controls were matched 1:1 with AS patients by age (± 5 years) and sex. A total of 362 potential controls were randomly selected from subjects with regular health examinations in the center, and they had no rheumatic or autoimmune symptoms. The study conformed to the Declaration of Helsinki, and the design of the work and final report were approved by the appropriate ethics committees of CSMU Hospital.
Bath AS indices
The Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), and Bath Ankylosing Spondylitis Global (BAS-G) were applied to evaluate the disease activity, physical function, and global well-being, respectively. The modified Chinese versions of BASDAI, BASFI, and BAS-G have good intraclass correlations and Cronbach’s alpha measures25.
Laboratory analyses
Peripheral blood was collected and centrifuged to separate serum and cells. Erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and IgA were measured. Carriage of HLA-B27 was assessed by flow cytometry26.
Genomic DNA was extracted from peripheral blood using the AxyPrep™ Blood Genomic DNA Miniprep Kit (Axygen Scientific, Union City, CA, USA). The IL-12B A1188C (rs3212227)27 and IL-23R C2370A (rs10889677)28 genotypes were identified by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP).
The following primers were used for the amplification of the IL-12B A1188C genotype: 5′-TTC TAT CTG ATT TGC TTT A-3′ and 5′-TGA AAC ATT CCA TAC ATC C-3′. DNA (0.5 μl) was added to a PCR buffer containing a 200 ng mix of primers, 1.5 mM MgCl2, 0.2 mM dNTPs, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), and 0.1% bovine serum albumin in a final volume of 50 μl. Amplification was carried out with the following PCR conditions: the denaturing step at 95°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 1 min. PCR products were digested with TaqI for 16 h at 65°C. Homozygous AA individuals exhibited a product fragment of 234 bp, whereas homozygous CC individuals resulted in 165 bp and 69 bp fragments, and heterozygous AC individuals had all 3 fragments. Primers for amplification of the IL-23R C2370A genotype were 5′-ATC GTG AAT GAG GAG TTG CC-3′ and 5′-TGT GCC TGT ATG TGT GAC CA-3′. Amplification occurred under the following PCR conditions: the denaturing step at 95°C for 30 s, annealing at 50°C for 30 s, and extension at 72°C for 1 min. PCR products were digested with MnlI for 16 h at 37°C. Homozygous AA showed 2 fragments of 286 bp and 185 bp, homozygous CC showed 3 fragments of 225 bp, 185 bp and 61 bp, and heterozygous CA individuals had all 4 fragments.
For each assay, a nontemplate control was also added to monitor reagent contamination. All genotypic analyses of case and control samples were performed blinded. To validate the results, about 20% of the samples in each genotype group were randomly selected for repeated PCR analysis.
Statistical analyses
Hardy-Weinberg equilibrium was performed to test IL-12B A1188C and IL-23R C2370A genotypes for goodness-of-fit. Demographic data and clinical features of AS patients and controls are presented by number (%) and mean ± SD. Both genotypic and allelic frequencies were compared between patients with AS and controls by chi-square tests. The matched relative risks (RRm) and 95% CI on development of AS were evaluated for the IL-12B A1188C and IL-23R C2370A genotypes using a conditional logistic regression model. Interaction was further assessed using the likelihood ratio test to calculate chi-square and p values. In the test for interaction, the conditional logistic regression model with only main effects was compared to that with both main effect terms and the interaction term. Interaction effect was defined as the difference of their deviance. The differences among the genotypes and alleles for continuous variables (BASDI, BASFI, BAS-G, ESR, and CRP) were calculated using Student t tests for 2 groups and analysis of variance for more than 2 groups. After adjustment for effects of potential confounding factors, association among BASDAI, BASFI, BAS-G, ESR, and CRP and IL-12B and IL-23R genotypes was further assessed using a general linear model. All p values were calculated using 2-tailed tests and a value < 0.05 was considered statistically significant. SAS 9.1 for Windows (SAS Inc., Cary, NC, USA) was used for all analysis.
RESULTS
Demographic characteristics and clinical features of patients with AS and healthy controls are shown in Table 1, and genotyping and allele frequencies of IL-12B A1188C and IL-23R C2370A polymorphisms are shown in Table 2. Among the controls, the frequencies of IL-12B CC, AC, and AA genotypes were 17.1%, 42.8%, and 40.1%, respectively; and IL-23R AA, CA, and CC genotypes were 53.3%, 37.9%, and 8.8%. IL-12B A1188C and IL-23R C2370A conformed to Hardy-Weinberg equilibrium (p > 0.05). Individuals with IL-12B CC (RRm 1.93, 95% CI 1.23–3.03, p = 0.0044) and AC (RRm 1.73, 95% CI 1.21–2.46, p = 0.0025) genotypes had significantly greater risk of AS development compared to those with the AA genotype. The IL-12B C allele had a 1.45-fold risk (95% CI 1.16–1.81, p = 0.0011) for AS development compared to the IL-12B A allele. In contrast, the IL-23R C2370A genotype was not significantly associated with development of AS.
We further evaluated the interaction of IL-12B A1188C and IL-23R C2370A genotypes on development of AS (Table 3). Because the number of IL-23R CC genotypes was small, IL-23R CC and CA genotypes were combined for subsequent analyses. A statistically significant interaction between IL-12B and IL-23R genotypes on AS development was observed (p = 0.0013). When subjects with IL-12B AA and IL-23R CC/CA genotypes were selected as the reference, those with IL-12B AC and IL-23R CC/CA genotypes had a significantly increased risk for AS development (RRm 1.81, 95% CI 1.05–3.13, p = 0.0329). Subjects with IL-12B AC and IL-23R AA genotypes also had a 1.83-fold risk (95% CI 1.07–3.12, p = 0.0278) for AS development compared to those with IL-12B AA and IL-23R CC/CA genotypes. Subjects with IL-12B CC and IL-23R AA genotypes had the highest risk for AS development (RRm 2.98, 95% CI 1.51–5.89, p = 0.0017). Individuals with IL-12B CC and IL-23R CC/CA genotypes and those with IL-12B AA and IL-23R AA genotypes also had elevated risks for AS development compared to those with IL-12B AA and IL-23R CC/CA genotypes, but these were not statistically significant.
Subjects with the IL-12B C allele might have higher expression of IL-1211, thus IL-12B CC and AC genotypes were combined in order to increase statistical power. As shown in Table 4, AS patients with IL-12B CC/AC genotypes had a significantly increased BASDAI score compared to those with IL-12B AA genotype (4.3 vs 3.7, respectively; p = 0.032). Patients with the IL-23R AA and CA genotypes also had increased BASDAI scores compared to those with the IL-23R CC genotype (3.9 and 4.5 vs 3.6; p = 0.055). After adjustment for effects of age, sex, disease duration, and medication history, differences in BASDAI score for patients with AS who had different IL-12B genotypes were marginally significant (p = 0.072). No significant differences in BASFI, BAS-G, ESR, or CRP levels were found in patients with AS who had different IL-12B and IL-23R genotypes. In addition, the proportions of peripheral arthritis, uveitis, and IBD were not different in the IL-12B and IL-23R genotypes (data not shown).
DISCUSSION
In our study, the IL-12B A1188C genotype was associated with the development of AS. Further, there was a significant interaction between IL-12B and IL-23R genes on development of AS. In patients with AS, the IL-12B A1188C genotype was also related to the disease activity indicated by the BASDAI score.
It has been suggested that Th17 cells are involved in the pathogenesis of AS4. IL-12, made up of the IL-12p35 (IL-12α) and IL-12p40 subunits, is a proinflammatory cytokine that induces expression of interferon-γ, leading to the differentiation and proliferation of Th113. Elevated expression of Th1 is known to be involved in the pathway of the spondyloarthropathies29,30. IL-12p40 knockout mice, which are defective in both the IL-12 and IL-23 pathways, are not susceptible to experimentally induced autoimmune encephalomyelitis31, suggesting that IL-12p40 might play a key role in the Th1 and Th17 pathways. However, the effect of IL-12 on Th17 cells is still controversial, and the role of IL-12 in the pathogenesis of AS remains unclear. An A to C polymorphism at position 1188 in the 3′-UTR of the IL-12B gene might result in the increased secretion of IL-1218. Similar to a recent study32, we observed that IL-12B genetic polymorphism confers susceptibility to AS in our population. This might be due to increased IL-12p40 protein secretion, resulting in enhanced differentiation of Th17 cells. Further, IBD and psoriasis frequently occur in patients with AS33, and spondyloarthritis is common in patients with IBD32. IL-12B has also been shown to be associated with IBD and psoriasis32,34. These findings suggest that differentiation and proliferation of Th17 cells might be a common etiopathogenic pathway for AS, IBD, and psoriasis, and also highlight the involvement of common risk variants across multiple diseases. However, IL-12B encodes the p40 subunit common to IL-12 and IL-2315. A previous study observed that IL-23 plays a more dominant role than IL-12 in psoriasis35. Again, the role of IL-12 in the pathogenesis of AS is rather unclear, thus the effect of the IL-12B genotype on increasing production of IL-12 or IL-23 requires careful consideration.
Genomewide association study has identified IL23R as having a strong association with AS20. But the IL-23R C2370A polymorphism (rs10889677) was not associated with AS in our study. This result is similar to previous reports36,37. The statistical power of the IL-23R C2370A genetic polymorphism in our study was only 6.7%. Thus, we could not observe any statistical association between IL-23R C2370A polymorphism and development of AS in our subjects. This SNP is likely to be truly associated with AS, although most likely through linkage disequilibrium with rs11209026 and rs11209032, which are thought to be the causative variants20.
IL-12 and IL-23 have been proposed to play critical roles in susceptibility to AS and rheumatoid arthritis38. Binding of IL-23 to the IL-23 receptor complex leads to overexpression of IL-1711,12. We observed a significant interaction of IL-12B A1188C and IL-23R C2370A genotypes in development of AS. We also found that subjects that carried both IL-12B AC and IL-23R CC/CA or IL-12B AC and IL-23R AA genotypes had moderately increased risks of AS development compared to those with both IL-12B AA and IL-23R CC/CA genotypes. Subjects with both of the IL-12B CC and IL-23R CC/CA genotypes also had an elevated risk of developing AS, although this was not significant, which may indicate that these polymorphisms have little influence on disease development, or that the study did not have sufficient power to detect a difference. Interestingly, subjects who carried the IL-12B CC and IL-23R AA genotypes were more likely to experience development of AS than those with IL-12B AA and IL-23R CC/CA genotypes. These findings suggest that the combination of specific “high-risk” polymorphisms creates a greater risk for AS than either polymorphism alone. IL-12 is mainly expressed by activated macrophages, and acts on T cells and natural killer cells39. IL-23, which shares the common p40 subunit with IL-1215, is an essential initiating cytokine in autoimmune diseases and has been linked to the development and maintenance of Th17 cells. A subunit of the receptor for IL-23A/IL-23 encoded by the IL-23R gene accompanies the receptor IL-12RB1 to form a receptor complex enabling the heterodimeric cytokine to confer immune responses associated with IL-23R40. Therefore, variants in both genes are very likely to influence signaling activities and to result in susceptibility to AS. Additional functional studies are required to test this hypothesis.
The disease severity of AS may also be associated with genetic characteristics21,22. We found obvious differences in the BASDAI in AS patients with IL-12B polymorphisms, although the results was marginally significant after adjustment for the effects of potential confounders. Our AS patients with IL-12B CC/AC genotypes had higher BASDAI score than those with the AA genotype. This might reflect that overexpression of IL-12p40 leads to increased expression of Th17.
The frequencies of IL-12B C (38.5%) and IL-23R A alleles (72.2%) in our healthy controls were similar to those reported previously for Han Chinese (IL-12B C allele: 41.5%; IL-23R A allele: 67.1%; Website: http://www.ncbi.nlm.nih.gov/SNP/) and conformed to Hardy-Weinberg equilibrium. These findings indicate the validity of the genotyping methodology used in this study.
One must be aware of several limitations when interpreting our results. Our study has no confirmation cohort, and the sample size was too small to be robust for primary associations of modest effects. In particular, our finding for gene-gene interaction might have a false-positive result. Larger cohort studies would be necessary to confirm evidence regarding our findings. In our study, only 28 patients with AS were HLA-B27-negative. Thus, it was difficult to evaluate the genetic effects of IL-12B A1188C and IL-23R C2370A in HLA-B27-negative subjects. Subjects were interviewed retrospectively, raising issues with regard to recall bias. However, the 3 questionnaires (BASDAI, BASFI, and BAS-G) were administered simultaneously. In addition, our patients and controls were matched by age and sex in an attempt to reduce for selection bias.
Our study suggests there is an association of IL-12B A1188C with susceptibility of a person for development of AS. In addition, an interaction between IL-12B A1188C and IL-23R C2370A genotypes appeared to influence development of AS, and indicated that the IL-12B A1188C genotype was related to the disease activity of AS.
Footnotes
-
Supported by a grant from the CSMU Hospital, Taiwan (CSH-2010-C-004) and the National Science Council, Taiwan (NSC-99-2314-B-040-019-MY3). The Instrument Center of CSMU is supported by the National Science Council and the Ministry of Education.
- Accepted for publication August 31, 2011.