July 2012
Volume 53, Issue 8
Free
Genetics  |   July 2012
Association of C2 and CFB Polymorphisms with Anterior Uveitis
Author Notes
  • From the Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, China. 
  • Corresponding author: Chi Pui Pang, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Eye Hospital, 147K Argyle Street, Kowloon, Hong Kong; cppang@cuhk.edu.hk
Investigative Ophthalmology & Visual Science July 2012, Vol.53, 4969-4974. doi:10.1167/iovs.12-9478
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ming-ming Yang, Timothy Y. Y. Lai, Pancy O. S. Tam, Sylvia W. Y. Chiang, Tsz Kin Ng, Ke Liu, Chi Pui Pang; Association of C2 and CFB Polymorphisms with Anterior Uveitis. Invest. Ophthalmol. Vis. Sci. 2012;53(8):4969-4974. doi: 10.1167/iovs.12-9478.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Association of rs800292 (I62V) in the complement factor H (CFH) gene with anterior uveitis (AU) was identified in our previous study. We proceeded to investigate whether polymorphisms of two tightly linked genes in the complement pathway, complement component 2 (C2) and complement factor B (CFB), are associated with AU.

Methods.: Five single-nucleotide polymorphisms (SNPs), rs1048709, rs537160, rs4151657, rs2072633 in CFB, and rs3020644 in C2, were examined using genotyping assays in 98 Chinese AU patients and 291 unrelated controls. Adjustments and stratifications were given for sex, clinical manifestations, and HLA-B27 status.

Results.: There were significant increases in the frequency of A allele and AA homozygosity for CFB-rs1048709 in AU patients compared with that of controls (P value after Bonferroni correction [P corr] = 2.67 × 10−4, P corr = 0.001, respectively). No association was found between AU and the other four SNPs after adjustment for multiple testing. Logistic regression analysis showed none of the 5 SNPs had significant interaction with sex. Stratified analyses showed that only rs1048709 was significantly associated with AU in HLA-B27-positive patients but not in HLA-B27-negative patients. No association was found in the 5 tested SNPs with clinical manifestations. A haplotype block across CFB (AATA) was significantly predisposed to AU with increased risk of 1.97 (P corr = 0.0005). Additive effect of CFB-rs1048709 and CFH-rs800292 was identified with an odds ratio of 7.48.

Conclusions.: Our results revealed an association between AU and CFB-rs1048709. The influence on AU might differ depending on HLA-B27 status. The joint effect in CFB and CFH strengthens the concept that the complement system plays an important role in the pathogenesis of AU.

Introduction
Uveitis is a general term for intraocular inflammatory disorders of the uveal tract; it has been described as anterior, intermediate, posterior, and panuveitis anatomically. Anterior uveitis (AU) is the most common form and accounts for approximately 75% of all cases. 1 AU may be idiopathic and associated with other systemic diseases such as ankylosing spondylitis, Behçet's disease, juvenile idiopathic arthritis, and inflammatory bowel diseases. Infection is also one important cause. AU may ultimately develop into vision loss secondary to cataract, glaucoma, cystoid macular edema, and epiretinal membrane. 2  
Although the exact pathogenesis of AU is unclear, it is an immune system-mediated disorder and is regulated by various endogenous immunological factors. 3 AU can be triggered by genetic predisposition coupled with environmental factors. 4 Until now, the most common and strongest association of AU has been identified with human leukocyte antigen (HLA)-B27, the major histocompatibility complex type I gene. 5 Meanwhile, many non-HLA gene variants, such as chemokines and antioxidant enzyme gene polymorphisms, also play important roles in the pathogenesis of AU. 6,7 Complement system is a key innate factor for immune defense and modulates various immune and inflammatory responses. Several studies have demonstrated that the activation of complement system is critical for the development of experimental autoimmune anterior uveitis (EAAU), and depletion of the host's complement system could result in complete inhibition of EAAU. 8,9 Recently, we identified the association between CFH (I62V) and AU and thus revealed the genetic association of complement system with uveitis. 10 The complement system can be divided into classical, lectin, and alternative pathways triggered by different stimuli. CFH is one of the key regulators in the alternative pathway; complement via this pathway is under strict control. Disruption of the balance of complement activation and regulation results in harmful effects and leads to immune system-related diseases. 11 CFH binds to complement component 3b (C3b) and reduces activation of the alternative pathway C3-convertase (C3bBb). 12,13 CFB is paralogous with C2; both are control factors in the complement system. Especially for CFB, an opponent of CFH in an alternative pathway for the same binding site in C3b regulates the activation of complement. A recent study indicated that EAAU was completely inhibited in rats injected with CFB antibody via suppressing the alternative pathway. 14 The present study aimed to investigate the association of polymorphisms in C2 and CFB genes with AU. 
Methods
Study Design and Subjects
The study protocol was approved by the Ethics Committee on Human Research, the Chinese University of Hong Kong. All procedures were conducted according to the tenets of the Declaration of Helsinki. Informed consent was obtained from all study subjects after explanation of the nature of the study. All patients underwent a detailed ocular examination and clinical information was collected, including visual acuity, intraocular pressure, clinical features, sex, medical history such as systemic diseases, age at onset, laterality, complications, and pattern of AU (acute, recurrent, or chronic). All patients were recruited during the active phase of uveitis and were followed for at least 3 months after recruitment. The definition of uveitis was based on the Standardization Uveitis Nomenclature (SUN) classification. 15 Acute AU was defined as AU that resolved completely within 3 months; chronic AU as AU not fully resolved within 3 months; and recurrent AU as the development of AU more than once. Nonrecurrent AU was defined as lack of second episode of AU after at least 3 months of follow-up. Patients with any of the following situations were excluded: (1) AU secondary to ocular or systemic infections; (2) AU secondary to specific syndromes (eg, Posner-Schlossman's syndrome, Fuchs' uveitis, Vogt-Koyanagi-Harada, or Behçet's disease); or (3) patients who were unable to cooperate during ocular examination and had chronic uveitis at the onset of the study. A total of 291 sex- and age-matched subjects 50 years or older with no ophthalmic eye disease except senile cataract and without any systemic immune-mediated diseases were recruited as control subjects. 
DNA Extraction and Genotyping
Venous blood was obtained from each study subject, and genomic DNA was extracted with a DNA blood kit (QIAamp; Qiagen, Hilden, Germany). Five single-nucleotide polymorphisms (SNPs), C2-rs3020644, CFB-rs1048709, rs537160, rs4151657, and rs2072633 across the C2-CFB region, including upstream and downstream, were selected by using HapMap (http://hapmap.ncbi.nlm.nih.gov/) for the Han Chinese population (minor allele frequency ≥15%). These SNPs were assessed by TaqMan genotyping assays with a Prism 7000 sequence detection system (Applied Biosystems, Foster City, CA) according to the manufacturer's protocol. All PCR amplifications were performed with the following thermal cycling conditions: 95°C for 10 minutes, followed by 40 cycles of 92°C for 15 seconds, and 62°C for 1.5 minutes. The HLA-B27 allele was detected by nested PCR as described by Konno et al. 16 Genotypes were read by using Prism 7000 SDS version 1.1 software (Applied Biosystems). 
Statistical Analysis
Hardy-Weinberg equilibrium was tested by χ 2 test for genotype frequencies of the SNPs in control group. Allelic and genotypic frequencies between cases and controls were compared by χ 2 test or Fisher's exact test. Dominant and recessive models were applied to investigate the disease association with regard to the minor allele (A for rs3020644, A for rs1048709, A for rs537160, C for rs4151657, and G for rs2072633). Stratified analyses based on clinical manifestations and HLA-B27 status were also performed. Logistic regression analysis was applied to adjust the association of these SNPs with sex. Pairwise linkage disequilibrium (LD [D′]) between polymorphisms and expectation maximization (EM)-based haplotype association analysis were performed with Haploview (ver. 4.2) software. Odds ratios (OR) and 95% confidence intervals (CI) were also calculated. A P value of <0.05 was considered statistically significant. P values were corrected by using a Bonferroni test for multiple comparisons (n = total number of SNPs). 
Results
Patient Demographics
A total of 98 AU patients were recruited, including 45 (45.9%) males and 53 (54.1%) females and 47 (48.0%) unilateral and 51 (52.0%) bilateral patients. The mean age ± standard deviation of the patients was 49.7 ± 16.0 years, range 11 to 87 years. Acute AU occurred in 92 (93.9%) patients, of whom 59 (60.2%) had recurrent episodes of AU and 6 (6.1%) developed chronic AU after acute episodes. Systemic diseases associated with AU patients included ankylosing spondylitis (n = 19, 19.4%), 1 case (1.0%) each of psoriasis, systemic lupus erythematosus, ulcerative colitis, and interstitial nephritis. 
Associations between SNPs and AU
All 5 SNPs followed Hardy-Weinberg equilibrium in controls. Significant association was detected at CFB-rs1048709, where there was a significant increase in the frequencies of A allele and AA homozygosity and decrease in the frequency of GG homozygosity in AU patients compared with that in controls (P corr = 2.67 × 10−4, OR = 1.99, 95% CI = 1.42–2.78; P corr = 0.001, OR = 3.44, 95% CI = 1.74–6.82; and P corr = 0.01, OR = 2.14, 95% CI = 1.32–3.45, respectively). There was a significant increase in the frequency of AA homozygosity for C2-rs3020644 and decrease in the frequency of GG homozygosity for CFB-rs2072633 in AU patients compared with controls (P = 0.013, OR = 1.86, 95% CI = 1.13–3.05; and P = 0.029, OR = 0.44, 95% CI = 0.21–0.94, respectively), but these two associations lost significance after adjustment for multiple testing (P corr = 0.065 and P corr = 0.15, respectively). No significant differences in genotypic or allelic frequencies were observed for either CFB-rs537160 or CFB-rs4151657 SNPs between AU patients and controls (Table 1). As sex-specific genetic differences were found between AU and CFH-rs800292 in our previous study, sex ratio adjustments were performed by logistic regression analysis. The association did not alter between AU and all 5 SNPs after adjusting for sex (data not shown). No differences in sex susceptibility were found between AU and C2-CFB polymorphisms. 
Table 1. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in Patients with AU and Control Subjects
Table 1. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in Patients with AU and Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 196) Control (n = 582) P Value (P corr) Odds Ratio (95% CI) AU (n = 98) Control (n = 291) P Value (P corr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 110(56.1) 280(48.1) 0.052§ (0.26) 1.38 (1.0–1.91) AA 35(35.7) 67(23.0)
Promoter G 86(43.9) 302(51.9) AG 40(40.8) 146(50.2) 0.52†§
GG 23(23.5) 78(26.8) 0.013‡§ (0.065) 1.86 (1.13–3.05)
CFB
rs1048709 G > A A 85(43.4) 162(27.8) 5.34×10−5§ (2.67×10−4) 1.99 (1.42–2.78) AA 19(19.4) 19(6.5)
Exon3(R150R) G 111(56.6) 420(72.2) AG 47(48.0) 124(42.6) 0.002†§ (0.01) 2.14 (1.32–3.45)
GG 32(32.7) 148(50.9) 2.09×10−4‡§ (0.001) 3.44 (1.74–6.82)
rs537160 G > A A 108(55.1) 280(48.1) 0.09§ 1.32 (0.96–1.83) AA 28(28.6) 63(21.6)
IVS7 G 88(44.9) 302(51.9) AG 52(53.1) 154(52.9) 0.16†§
GG 18(18.4) 74(25.4) 0.16‡§
rs4151657 T > C C 42(21.4) 146(25.1) 0.3§ 1.23 (0.83–1.81) CC 6(6.1) 16(5.5)
IVS10 T 154(78.6) 436(74.9) CT 30(30.6) 114(39.2) 0.17†§
TT 62(63.3) 161(55.3) 0.803‡*
rs2072633 A > G G 71(36.2) 254(43.6) 0.069§ 1.36 (0.98–1.90) GG 9(9.2) 54(18.6)
IVS17 A 125(63.8) 328(56.4) AG 53(54.1) 146(50.2) 0.32†§
AA 36(36.7) 91(31.3) 0.029‡§ (0.15) 0.44 (0.21–0.94)
Associations between SNPs and AU Stratified by HLA-B27 Status and Clinical Features
Among the 98 AU patients, 42 (42.9%) were HLA-B27-positive and 56 (57.1%) HLA-B27-negative. For the HLA-B27-positive AU patients, significantly higher proportions of A allele and AA homozygosity and a lower proportion of GG homozygosity in CFB-rs1048709 were found than those in controls (P corr = 7.30 × 10−5, OR = 2.72(95% CI, 1.71–4.33); P corr = 0.005, OR = 4.47(95% CI, 1.92–10.46); and P corr = 0.0018, OR = 3.80(95% CI, 1.75–8.21, respectively). Similar differences were also found in C2-rs3020644, CFB-rs537160, and CFB-rs2072633. These differences did not remain after adjustment for multiple testing. No difference was found in allelic or genotypic frequency for CFB-rs4151657 between AU patients and controls (Table 2). For the HLA-B27-negative AU patients, only one modest association was found in CFB-rs1048709 with higher proportions of A allele and AA homozygosity, but the differences lost significance after correction (P = 0.04, P corr = 0.2; and P = 0.028, P corr = 0.14, respectively (Table 3). No significant differences were found in allelic and genotypic frequencies among AU patients stratified by either laterality or recurrence status (data not shown). 
Table 2. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Positive AU Patients versus Control Subjects
Table 2. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Positive AU Patients versus Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 84) Control (n = 582) P Value (Pcorr) Odds Ratio (95% CI) AU (n = 42) Control (n = 291) P Value (Pcorr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 48(57.1) 280(48.1) 0.12§ 1.44 (0.91–2.28) AA 17(40.5) 67(23.0)
Promoter G 36(42.9) 302(51.9) AG 14(33.3) 146(50.2) 0.93†§
GG 11(26.2) 78(26.8) 0.015‡§ (0.075) 2.27 (1.16–4.46)
CFB
rs1048709 G > A A 43(51.2) 162(27.8) 1.46 ×10−5§ (7.30×10−5) 2.72 (1.71–4.33) AA 10(23.8) 19(6.5)
Exon3(R150R) G 41(48.8) 420(72.2) AG 23(54.8) 124(42.6) 3.55×10−4†§ (0.0018) 3.80 (1.75–8.21)
GG 9(21.4) 148(50.9) 0.001‡* (0.005) 4.47 (1.92–10.46)
rs537160 G > A A 51(60.7) 280(48.1) 0.031§ (0.16) 1.67 (1.05–2.66) AA 16(38.1) 63(21.6)
IVS7 G 33(39.3) 302(51.9) AG 19(45.2) 154(52.9) 0.22†§
GG 7(16.7) 74(25.4) 0.019‡§ (0.095) 2.23 (1.13–4.41)
rs4151657 T > C C 19(22.6) 146(25.1) 0.62§ 1.15 (0.67–1.98) CC 3(7.1) 16(5.5)
IVS10 T 65(77.4) 436(74.9) CT 13(31.0) 114(39.2) 0.42†§
TT 26(61.9) 161(55.3) 0.72‡*
rs2072633 A > G G 26(31.0) 254(43.6) 0.028§ (0.14) 1.73 (1.06–2.82) GG 4(9.5) 54(18.6)
IVS17 A 58(69.0) 328(56.4) AG 18(42.9) 146(50.2) 0.036†§ (0.18) 2.0 (1.04–3.84)
AA 20(47.6) 91(31.3) 0.15‡*
Table 3. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Negative AU Patients versus Control Subjects
Table 3. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Negative AU Patients versus Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 112) Control (n = 582) P Value (P corr) Odds Ratio (95% CI) AU (n = 56) Control (n = 291) P Value (P corr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 62(55.4) 280(48.1) 0.16§ 1.34 (0.89–2.01) AA 18(32.1) 67(23.0)
Promoter G 50(44.6) 302(51.9) AG 26(46.4) 146(50.2) 0.401†§
GG 12(21.4) 78(26.8) 0.15‡§
CFB
rs1048709 G > A A 42(37.5) 162(27.8) 0.04§ (0.2) 1.56 (1.02–2.38) AA 9(16.1) 19(6.5)
Exon3(R150R) G 70(62.5) 420(72.2) AG 24(42.9) 124(42.6) 0.18†§
GG 23(41.1) 148(50.9) 0.028‡* (0.14) 2.74 (1.17–6.42)
rs537160 G > A A 57(50.9) 280(48.1) 0.59§ 1.12 (0.75–1.68) AA 12(21.4) 63(21.6)
IVS7 G 55(49.1) 302(51.9) AG 33(58.9) 154(52.9) 0.36†§
GG 11(19.6) 74(25.4) 0.97‡§
rs4151657 T > C C 23(20.5) 146(25.1) 0.3§ 0.77 (0.47–1.27) CC 3(5.4) 16(5.5)
IVS10 T 89(79.5) 436(74.9) CT 17(30.4) 114(39.2) 0.22†§
TT 36(64.3) 161(55.3) 0.97‡*
rs2072633 A > G G 45(40.2) 254(43.6) 0.5§ 0.87 (0.58–1.31) GG 5(8.9) 54(18.6)
IVS17 A 67(59.8) 328(56.4) AG 35(62.5) 146(50.2) 0.69†§
AA 16(28.6) 91(31.3) 0.083‡*
Linkage Disequilibrium and Haplotype Association Analysis
Pairwise LD analysis revealed extensive LD throughout the CFB gene. One haplotype block was detected, including all 4 SNPs in CFB (rs1048709G > A, rs537160G > A, rs4151657T > C, and rs2072633A > G) (Fig. 1). Haplotype AATA conferred a 1.97-fold significantly increased risk of AU (P = 8.92 × 10−5, permutation P = 0.0005) (Table 4). 
Figure 1. 
 
Pairwise LD among 5 SNPs across C2-CFB genes. The LD spans the region of CFB with a distance of 4 kb. LD was measure by the D′ statistic, using data from all subjects. A D′ value of 100 indicates a complete LD between two markers, and a D′ value of 0 indicates a complete linkage equilibrium. Haplotype version 4.2 software was used.
Figure 1. 
 
Pairwise LD among 5 SNPs across C2-CFB genes. The LD spans the region of CFB with a distance of 4 kb. LD was measure by the D′ statistic, using data from all subjects. A D′ value of 100 indicates a complete LD between two markers, and a D′ value of 0 indicates a complete linkage equilibrium. Haplotype version 4.2 software was used.
Table 4. 
 
Haplotype Analysis of C2-CFB Polymorphisms between Cases and Controls
Table 4. 
 
Haplotype Analysis of C2-CFB Polymorphisms between Cases and Controls
Frequency Odds Ratio (95% CI)
Haplotype Frequency Case Control P P corr
AATA 0.309 0.421 0.272 8.92 × 10−5 0.0005 1.97(1.41–2.76)
GGCG 0.242 0.214 0.251 0.3009 NS -
GATA 0.188 0.130 0.208 0.0157 NS -
GGTG 0.168 0.141 0.177 0.2372 NS -
GGTA 0.083 0.081 0.084 0.9094 NS -
Interaction Analysis between CFH-rs800292 and CFB-rs1048709 in AU
Combined effects of CFH-rs800292 and CFB-rs1048709 were assessed, and corresponding ORs of AU for each possible combination of the genotypes of the two variants were estimated (Table 5a, b). ORs were compared with the baseline genotype of the two genes. The frequency of the homozygous risk genotypes at both loci was 5× lower in controls (2.1%) than in AU patients (11.2%, P = 0.005) (Table 5a). A joint disease OR of 7.48 in individuals with homozygous risk alleles at both loci was observed compared with the baseline nonrisk genotypes (Fig. 2). 
Figure 2. 
 
Two-locus (CFH and CFB) genotype-specific AU risk.
Figure 2. 
 
Two-locus (CFH and CFB) genotype-specific AU risk.
Table 5. 
 
Interaction Analysis between CFH-rs800292 and CFB-rs1048709 in AU
Table 5. 
 
Interaction Analysis between CFH-rs800292 and CFB-rs1048709 in AU
Genotype at CFB-rs1048709 a. Genotype Distribution b. Joint Odds Ratios and 95% Confidence
Genotype at CFH-rs800292 CFH-rs800292
Control (n = 240) AU (n = 98)
AA AG GG AA AG GG AA AG GG
GG 17(7.1%) 53(22.1%) 55(22.9%) 5(5.1%) 13(13.3%) 14(14.3%) 1.00(Ref) 0.83(0.26–2.68) 0.87(0.27–2.75)
AG 19(7.9%) 47(19.6%) 34(14.2%) 3(3.1%) 23(23.5%) 21(21.4%) 0.54(0.11–2.59) 1.66(0.55–5.07) 2.10(0.67–6.54)
AA 3(1.3%) 53(22.1%) 5(2.1%) 2(2.0%) 6(6.1%) 11(11.2%) 2.27(0.29–17.58) 2.91(0.67–12.77) 7.48(1.75–32.0)
Discussion
This is the first genetic study examining the association of variants in C2 and CFB genes with uveitis. Our results showed that CFB-rs1048709 is significantly associated with AU. CFB is paralogous with C2. These genes are located in tandem in the major histocompatibility complex class III region, a cluster on chromosome 6p21 with respect to infection and autoimmunity. 17 Moreover, C2 and CFB are involved in the activation of classical complement pathway and the alternative complement pathway, respectively. CFB levels were significantly up-regulated in sera of uveitis patients. Treatment with CFB antibodies resulted in suppression of EAAU in animals. 14,18 In addition, CFB gene polymorphisms are associated with multiple inflammatory diseases, such as age-related macular degeneration (AMD), lupus, and atypical hemolytic-uremic syndrome. 1921 Our results provide additional evidence for the involvement of complement system in ocular inflammatory disease-uveitis, especially CFB. Meanwhile, there are limitations in the present study: the relatively small sample size may lower statistical power. Some of the modest associations such as CFB-rs537160 and -rs2072633 could not remain statistically significant after adjustment for multiple testing. Additionally, these variants in CFB represent either synonymous substitutions (rs1048709 R150R) or intronic SNPs (rs537160, rs4151657, and rs2072633). Currently, there is no information on its biological functions. These polymorphisms may be linked with an undiscovered but biologically relevant structural variant in this region. Also, synonymous or intronic regulation could be involved in gene transcription or tissue specificity of gene expression. Thus, similar to the impaired CFH-mediated complement inhibition conferring AU risk, decreased complement activation by CFB might protect from AU. 22 Further investigations of more polymorphisms in C2-CFB genes, using a larger cohort and in other ethnic groups, could help to consolidate the findings and identify more genetic associations. Also, comprehensive evaluation of this region by extensive resequencing to uncover unknown variation is worthwhile. 
In addition to the association of CFB polymorphism with AU, our results also showed that AA homozygosity at C2-rs3020644 might be associated with increased risk of AU, although the association was modest and could not be obtained after adjustment for multiple testing (Table 1). C2-rs3020644 is located at the promoter that regulates gene expression and could be a susceptibility factor to AU. As a key component in classical complement pathway, the probable genetic association between C2 polymorphism and AU suggests the participation of each complement activation pathway in uveitis. 
Sex susceptibility of CFH (I62V) in AU was found in our previous study. 10 In this study, logistic regression analysis showed none of the 5 SNPs had significant interaction with sex, indicating that sex factor in C2 and CFB may not confer analogous effect as CFH. In addition, for HLA-B27-positive patients, significant association was identified with CFB-rs1048709, where there were significant increased frequencies of A allele and AA homozygosity in HLA-B27-positive patients compared with controls (Table 2), while for HLA-B27-negative patients, such association could not be detected after correction for multiple testing. Our findings suggested that the influence of C2 and CFB polymorphisms on AU differed depending on HLA-B27 status, inconsistent with results of our previous study of CFH. 10 The exact reason for this discrepancy is unclear, further indicating the complexity of immune regulation in uveitis. 6,10 Moreover, based on our previous study, 6 only 6% of control subjects are HLA-B27 positive, therefore we did not perform stratified analysis of the controls based on HLA-B27 positivity due to the small sample size. To further clarify the association between CFB-rs1048709 and AU under the condition of HLA-B27, formal hypothesis testing was also performed by comparing the allele and genotype frequency of rs1048709 in AU patients with or without HLA-B27. Our results showed that no significant difference was found between allele frequency and HLA-B27 positivity (P = 0.06). Also, no significant different was found between genotypic frequency and HLA-B27 positivity (P = 0.117). The nonsignificant results suggested no association between HLA-B27 and rs1048709 in our study cohort. While the P value was marginal, the results should be interpreted cautiously, and further investigations are still needed. Meanwhile, stratified analyses according to recurrence and laterality status of AU patients showed that clinical severity did not affect the association of these SNPs on AU. 
We identified a novel risk haplotype block across the CFB gene. This haplotype (AATA), defined by all the risk alleles of SNPs in CFB, conferred a 1.97-fold increased susceptibility to AU. In this block, 4 SNPs were in high LD with each other (D′ ≥ 0.94). Studies also showed that these 4 SNPs were shared in other large blocks in the C2-CFB region, including the functional variants (L9H and R32Q) in CFB for several immunologic or inflammatory diseases, such as AMD and Lupus. 21,23,24  
In conclusion, we revealed the fact that CFB-rs1048709, a haplotype AATA in CFB and a joint effect between CFH and CFB, conferred a significantly increased risk for anterior uveitis. A high joint effect of risk homozygosity (GG+AA) in CFH-rs800292 and CFB-rs1048709 with AU (joint OR = 7.48) was identified. Genetic influences on AU may be affected by HLA-B27 status. 
Acknowledgments
We thank all the participants in this study. Support from the donation from the Li Ka Shing Foundation is gratefully acknowledged. 
References
Chang JH Wakefield D. Uveitis: a global perspective. Ocul Immunol Inflamm . 2002;10 (4);263–279. [CrossRef] [PubMed]
Nussenblatt RB Gery I. Experimental autoimmune uveitis and its relationship to clinical ocular inflammatory disease. J Autoimmun . 1996;9 (5);575–585. [CrossRef] [PubMed]
Caspi RR. Immune mechanisms in uveitis. Springer Semin Immunopathol . 1999;21 (2);113–124. [CrossRef] [PubMed]
Martin TM Rosenbaum JT. Genetics in uveitis. Int Ophthalmol Clin . 2005;45 (2);15–30. [CrossRef] [PubMed]
Chang JH McCluskey PJ Wakefield D. Acute anterior uveitis and HLA-B27. Surv Ophthalmol . 2005;50 (4);364–388. [CrossRef] [PubMed]
Lan C Tam PO Chiang SW Manganese superoxide dismutase and chemokine genes polymorphisms in Chinese patients with anterior uveitis. Invest Ophthalmol Vis Sci . 2009;50 (12);5596–5600. [CrossRef] [PubMed]
Yeo TK Ahad MA Kuo NW Chemokine gene polymorphisms in idiopathic anterior uveitis. Cytokine . 2006;35 (1-2);29–35. [CrossRef] [PubMed]
Jha P Sohn JH Xu Q The complement system plays a critical role in the development of experimental autoimmune anterior uveitis. Invest Ophthalmol Vis Sci . 2006;47 (3);1030–1038. [CrossRef] [PubMed]
Read RW Szalai AJ Vogt SD McGwin G Barnum SR. Genetic deficiency of C3 as well as CNS-targeted expression of the complement inhibitor sCrry ameliorates experimental autoimmune uveoretinitis. Exp Eye Res . 2006;82 (3);389–394. [CrossRef] [PubMed]
Yang MM Lai TY Tam PO CFH 184G as a genetic risk marker for anterior uveitis in Chinese females. Mol Vis . 2011;17(2655–2664. [PubMed]
Zipfel PF Heinen S Jozsi M Skerka C. Complement and diseases: defective alternative pathway control results in kidney and eye diseases. Mol Immunol . 2006;43 (1-2);97–106. [CrossRef] [PubMed]
Zipfel PF Skerka C Hellwage J Factor H family proteins: on complement, microbes and human diseases. Biochem Soc Trans . 2002;30(pt 6);971–978.
Brosnan JT Brosnan ME. Branched-chain amino acids: enzyme and substrate regulation. J Nutr . 2006;136(suppl 1);207S–211S.
Manickam B Jha P Matta B Liu J Bora PS Bora NS. Inhibition of complement alternative pathway suppresses experimental autoimmune anterior uveitis by modulating T cell responses. J Biol Chem . 2011;286 (10);8472–8480. [CrossRef] [PubMed]
Jabs DA Nussenblatt RB Rosenbaum JT. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol . 2005;140 (3);509–516. [PubMed]
Konno Y Numaga J Tsuchiya N HLA-B27 subtypes and HLA class II alleles in Japanese patients with anterior uveitis. Invest Ophthalmol Vis Sci . 1999;40 (8);1838–1844. [PubMed]
Horton R Wilming L Rand V Gene map of the extended human MHC. Nat Rev Genet . 2004;5 (12);889–899. [CrossRef] [PubMed]
Zipplies JK Kirschfink M Amann B Hauck SM Stangassinger M Deeg CA. Complement factor B expression profile in a spontaneous uveitis model. Immunobiology . 2010;215 (12);949–955. [CrossRef] [PubMed]
Sanchez E Comeau ME Freedman BI Identification of novel genetic susceptibility loci in African American lupus patients in a candidate gene association study. Arthritis Rheum . 2011;63 (11);3493–3501. [CrossRef] [PubMed]
Tawadrous H Maga T Sharma J Kupferman J Smith RJ Schoeneman M. A novel mutation in the complement factor B gene (CFB) and atypical hemolytic uremic syndrome. Pediatr Nephrol . 2010;25 (5);947–951. [CrossRef] [PubMed]
Gold B Merriam JE Zernant J Variation in factor B (BF) and complement component 2 (C2) genes is associated with age-related macular degeneration. Nat Genet . 2006;38 (4);458–462. [CrossRef] [PubMed]
Jha P Bora PS Bora NS. The role of complement system in ocular diseases including uveitis and macular degeneration. Mol Immunol . 2007;44 (16);3901–3908. [CrossRef] [PubMed]
McKay GJ Silvestri G Patterson CC Hogg RE Chakravarthy U Hughes AE. Further assessment of the complement component 2 and factor B region associated with age-related macular degeneration. Invest Ophthalmol Vis Sci . 2009;50 (2);533–539. [CrossRef] [PubMed]
Fernando MM Stevens CR Sabeti PC Identification of two independent risk factors for lupus within the MHC in United Kingdom families. PLoS Genet . 2007;3 (11);e192.
Footnotes
 Supported by a donation from the Li Ka Shing Foundation.
Footnotes
 Disclosure: M-M. Yang, None; T.Y.Y. Lai, None; P.O.S. Tam, None; S.W.Y. Chiang, None; T.K. Ng, None; K. Liu, None; C.P. Pang, None
Figure 1. 
 
Pairwise LD among 5 SNPs across C2-CFB genes. The LD spans the region of CFB with a distance of 4 kb. LD was measure by the D′ statistic, using data from all subjects. A D′ value of 100 indicates a complete LD between two markers, and a D′ value of 0 indicates a complete linkage equilibrium. Haplotype version 4.2 software was used.
Figure 1. 
 
Pairwise LD among 5 SNPs across C2-CFB genes. The LD spans the region of CFB with a distance of 4 kb. LD was measure by the D′ statistic, using data from all subjects. A D′ value of 100 indicates a complete LD between two markers, and a D′ value of 0 indicates a complete linkage equilibrium. Haplotype version 4.2 software was used.
Figure 2. 
 
Two-locus (CFH and CFB) genotype-specific AU risk.
Figure 2. 
 
Two-locus (CFH and CFB) genotype-specific AU risk.
Table 1. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in Patients with AU and Control Subjects
Table 1. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in Patients with AU and Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 196) Control (n = 582) P Value (P corr) Odds Ratio (95% CI) AU (n = 98) Control (n = 291) P Value (P corr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 110(56.1) 280(48.1) 0.052§ (0.26) 1.38 (1.0–1.91) AA 35(35.7) 67(23.0)
Promoter G 86(43.9) 302(51.9) AG 40(40.8) 146(50.2) 0.52†§
GG 23(23.5) 78(26.8) 0.013‡§ (0.065) 1.86 (1.13–3.05)
CFB
rs1048709 G > A A 85(43.4) 162(27.8) 5.34×10−5§ (2.67×10−4) 1.99 (1.42–2.78) AA 19(19.4) 19(6.5)
Exon3(R150R) G 111(56.6) 420(72.2) AG 47(48.0) 124(42.6) 0.002†§ (0.01) 2.14 (1.32–3.45)
GG 32(32.7) 148(50.9) 2.09×10−4‡§ (0.001) 3.44 (1.74–6.82)
rs537160 G > A A 108(55.1) 280(48.1) 0.09§ 1.32 (0.96–1.83) AA 28(28.6) 63(21.6)
IVS7 G 88(44.9) 302(51.9) AG 52(53.1) 154(52.9) 0.16†§
GG 18(18.4) 74(25.4) 0.16‡§
rs4151657 T > C C 42(21.4) 146(25.1) 0.3§ 1.23 (0.83–1.81) CC 6(6.1) 16(5.5)
IVS10 T 154(78.6) 436(74.9) CT 30(30.6) 114(39.2) 0.17†§
TT 62(63.3) 161(55.3) 0.803‡*
rs2072633 A > G G 71(36.2) 254(43.6) 0.069§ 1.36 (0.98–1.90) GG 9(9.2) 54(18.6)
IVS17 A 125(63.8) 328(56.4) AG 53(54.1) 146(50.2) 0.32†§
AA 36(36.7) 91(31.3) 0.029‡§ (0.15) 0.44 (0.21–0.94)
Table 2. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Positive AU Patients versus Control Subjects
Table 2. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Positive AU Patients versus Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 84) Control (n = 582) P Value (Pcorr) Odds Ratio (95% CI) AU (n = 42) Control (n = 291) P Value (Pcorr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 48(57.1) 280(48.1) 0.12§ 1.44 (0.91–2.28) AA 17(40.5) 67(23.0)
Promoter G 36(42.9) 302(51.9) AG 14(33.3) 146(50.2) 0.93†§
GG 11(26.2) 78(26.8) 0.015‡§ (0.075) 2.27 (1.16–4.46)
CFB
rs1048709 G > A A 43(51.2) 162(27.8) 1.46 ×10−5§ (7.30×10−5) 2.72 (1.71–4.33) AA 10(23.8) 19(6.5)
Exon3(R150R) G 41(48.8) 420(72.2) AG 23(54.8) 124(42.6) 3.55×10−4†§ (0.0018) 3.80 (1.75–8.21)
GG 9(21.4) 148(50.9) 0.001‡* (0.005) 4.47 (1.92–10.46)
rs537160 G > A A 51(60.7) 280(48.1) 0.031§ (0.16) 1.67 (1.05–2.66) AA 16(38.1) 63(21.6)
IVS7 G 33(39.3) 302(51.9) AG 19(45.2) 154(52.9) 0.22†§
GG 7(16.7) 74(25.4) 0.019‡§ (0.095) 2.23 (1.13–4.41)
rs4151657 T > C C 19(22.6) 146(25.1) 0.62§ 1.15 (0.67–1.98) CC 3(7.1) 16(5.5)
IVS10 T 65(77.4) 436(74.9) CT 13(31.0) 114(39.2) 0.42†§
TT 26(61.9) 161(55.3) 0.72‡*
rs2072633 A > G G 26(31.0) 254(43.6) 0.028§ (0.14) 1.73 (1.06–2.82) GG 4(9.5) 54(18.6)
IVS17 A 58(69.0) 328(56.4) AG 18(42.9) 146(50.2) 0.036†§ (0.18) 2.0 (1.04–3.84)
AA 20(47.6) 91(31.3) 0.15‡*
Table 3. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Negative AU Patients versus Control Subjects
Table 3. 
 
Comparison of Genotype and Allele Frequencies of C2-CFB Polymorphisms in HLA-B27-Negative AU Patients versus Control Subjects
SNP ID Designation Allele Distribution (%) Genotype Distribution (%)
AU (n = 112) Control (n = 582) P Value (P corr) Odds Ratio (95% CI) AU (n = 56) Control (n = 291) P Value (P corr) Odds Ratio (95% CI)
C2
rs3020644 G > A A 62(55.4) 280(48.1) 0.16§ 1.34 (0.89–2.01) AA 18(32.1) 67(23.0)
Promoter G 50(44.6) 302(51.9) AG 26(46.4) 146(50.2) 0.401†§
GG 12(21.4) 78(26.8) 0.15‡§
CFB
rs1048709 G > A A 42(37.5) 162(27.8) 0.04§ (0.2) 1.56 (1.02–2.38) AA 9(16.1) 19(6.5)
Exon3(R150R) G 70(62.5) 420(72.2) AG 24(42.9) 124(42.6) 0.18†§
GG 23(41.1) 148(50.9) 0.028‡* (0.14) 2.74 (1.17–6.42)
rs537160 G > A A 57(50.9) 280(48.1) 0.59§ 1.12 (0.75–1.68) AA 12(21.4) 63(21.6)
IVS7 G 55(49.1) 302(51.9) AG 33(58.9) 154(52.9) 0.36†§
GG 11(19.6) 74(25.4) 0.97‡§
rs4151657 T > C C 23(20.5) 146(25.1) 0.3§ 0.77 (0.47–1.27) CC 3(5.4) 16(5.5)
IVS10 T 89(79.5) 436(74.9) CT 17(30.4) 114(39.2) 0.22†§
TT 36(64.3) 161(55.3) 0.97‡*
rs2072633 A > G G 45(40.2) 254(43.6) 0.5§ 0.87 (0.58–1.31) GG 5(8.9) 54(18.6)
IVS17 A 67(59.8) 328(56.4) AG 35(62.5) 146(50.2) 0.69†§
AA 16(28.6) 91(31.3) 0.083‡*
Table 4. 
 
Haplotype Analysis of C2-CFB Polymorphisms between Cases and Controls
Table 4. 
 
Haplotype Analysis of C2-CFB Polymorphisms between Cases and Controls
Frequency Odds Ratio (95% CI)
Haplotype Frequency Case Control P P corr
AATA 0.309 0.421 0.272 8.92 × 10−5 0.0005 1.97(1.41–2.76)
GGCG 0.242 0.214 0.251 0.3009 NS -
GATA 0.188 0.130 0.208 0.0157 NS -
GGTG 0.168 0.141 0.177 0.2372 NS -
GGTA 0.083 0.081 0.084 0.9094 NS -
Table 5. 
 
Interaction Analysis between CFH-rs800292 and CFB-rs1048709 in AU
Table 5. 
 
Interaction Analysis between CFH-rs800292 and CFB-rs1048709 in AU
Genotype at CFB-rs1048709 a. Genotype Distribution b. Joint Odds Ratios and 95% Confidence
Genotype at CFH-rs800292 CFH-rs800292
Control (n = 240) AU (n = 98)
AA AG GG AA AG GG AA AG GG
GG 17(7.1%) 53(22.1%) 55(22.9%) 5(5.1%) 13(13.3%) 14(14.3%) 1.00(Ref) 0.83(0.26–2.68) 0.87(0.27–2.75)
AG 19(7.9%) 47(19.6%) 34(14.2%) 3(3.1%) 23(23.5%) 21(21.4%) 0.54(0.11–2.59) 1.66(0.55–5.07) 2.10(0.67–6.54)
AA 3(1.3%) 53(22.1%) 5(2.1%) 2(2.0%) 6(6.1%) 11(11.2%) 2.27(0.29–17.58) 2.91(0.67–12.77) 7.48(1.75–32.0)
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×