Association of T-Bet, GATA-3, RORC, and FOXP3 Copy Number Variations With Acute Anterior Uveitis With or Without Ankylosing Spondylitis in Chinese Han

Abstract

Purpose: Acute anterior uveitis (AAU) is the most common form of uveitis and is a frequent ocular manifestation of ankylosing spondylitis (AS). Thymocyte CD4⁺ cells have been reported to play an important role in the pathogenesis of both AAU and AS. To test whether the copy number variations (CNVs) of CD4⁺ T cell transcription factor genes including T-bet, GATA binding protein 3 (GATA)-3, related orphan receptor C (RORC) and forkhead box protein 3 (FOXP3) are associated with acute anterior uveitis either in the presence or absence of ankylosing spondylitis (AAU⁺AS⁺; AAU⁺AS⁻).

Methods: The study included 676 patients with AAU, including 298 patients with AAU⁺AS⁺, 378 patients with AAU⁺AS⁻, and 596 unrelated healthy controls in a Chinese Han population. Copy number variations were examined by real-time PCR.

Results: The frequency of a high copy number (CN) of T-bet was increased in AAU⁺AS⁺ as well as AAU⁺AS⁻ patients when compared with controls (P value after Bonferroni correction [P_corr] = 4.3 × 10⁻⁵; odds ratio [OR] = 2.0 and P_corr = 1.2 × 10⁻⁸; OR = 2.3, respectively). The frequency of a high CN of GATA-3 was significantly higher in AAU⁺AS⁺ patients than in controls (P_corr = 1.8 × 10⁻⁷; OR = 4.9). A higher frequency of CN of FOXP3 was found in female AAU⁺AS⁺ patients and female AAU⁺AS⁻ patients (P_corr = 0.005, OR = 5.9 and P_corr = 0.004, OR = 4.9, respectively ). No association was found between CNVs of RORC and AAU⁺AS⁻ or AAU⁺AS⁺ patients.

Conclusions: A high copy number of T-bet and GATA-3 confers susceptibility to AAU and AS, and a high copy number of FOXP3 confers susceptibility to female patients with AAU either with or without AS.

Uveitis is among the leading causes of visual impairment and blindness.¹ Acute anterior uveitis (AAU), which is characterized as an anterior uveitis with a clinical course lasting less than 3 months, is the most common form of uveitis.² It is a disease that often recurs and forms a treatment challenge for the ophthalmologist dealing with these cases.³ 

Acute anterior uveitis can present as an isolated ocular disease but may also be associated with systemic diseases such as ankylosing spondylitis (AS).⁴ The gene HLA-B27 shows a very strong association with AAU, and HLA-B27–positive AAU is the most common type of anterior uveitis. Early data showed that approximately 50% of patients with HLA-B27 AAU develop seronegative arthritis (SNA), while approximately 25% of the patients initially diagnosed with HLA-B27 SNA develop AAU.^5,6 Previous studies have shown that CD4⁺ T helper cells may play a central role in the development of certain AAU types.⁷ 

Thymocyte helper CD4⁺ cells are currently divided in a number of subtypes such as the T helper type 1 (Th1), T helper type 2 (Th2), T helper type 17 (Th17), and the CD4⁺ T regulatory cell (T_Reg). The various T helper subtypes can be differentiated according to a number of specific transcription factors that are necessary in the differentiation and function-development process. The so-called “T-box” expressed in T cells (T-bet), GATA binding protein 3 (GATA)-3, retinoid-related orphan receptor γt (RORγt) and transcription factor forkhead box protein 3 (FOXP3) are examples of transcription factors that are specific for Th1, Th2, Th17, and T_Reg cells, respectively.^8,9 

Expression of T-bet is upregulated in uveitis patients with Behçet disease (BD) and Vogt Koyanagi Harada (VKH) syndrome and its expression was shown to parallel inflammation during experimental endotoxin uveitis in rats.^10–12 Expression of GATA-3 has not yet been studied in clinical uveitis, but was shown to be increased during experimental autoimmune uveitis in mice.¹³ Expression of RORC has been studied in Behçet disease but was not shown to be different compared with healthy controls¹⁴ Expression of FOXP3 has been shown to be associated with the control of both experimental as well as clinical uveitis.^7,15,16 

Genetic polymorphisms of the transcription factors delineated above may alter their expression and function and an association between certain genetic variants of these factors and immune-related disorders have been reported.^17–19 Genetic variation of genes has mainly been addressed by investigating single nucleotide polymorphisms. Recently this has been expanded by analyzing the role of gene copy repeats in the predisposition to certain diseases.²⁰ The thymocyte helper CD4⁺ transcription factors mentioned above show a variable gene copy number in the population, which may affect their role in immune-related disorders.²¹ Earlier studies from our group showed that a high RORC-CNV was associated with an increased susceptibility to BD and that a low FOXP3-CNV predisposed to BD in female patients.²² Functional studies showed that the relative mRNA expression levels of RORC was increased in individuals with a high RORC gene copy number, but this could not be shown for FOXP3.²² The effect of CNVs on the expression of GATA-3 has, to our knowledge, not yet been reported for humans, although experiments in transgenic and mutant mice have shown that the GATA-3 gene copy number affects its expression.²³ Expression of T-bet mRNA was also shown to be increased in individuals carrying higher gene copy numbers.²¹ 

As yet the role of these CNVs has not yet been addressed in AAU and this was therefore the subject of the study reported here. This was performed by testing CNVs of transcription factors genes in a large cohort of Chinese AAU patients with or without AS and comparing the frequencies with those found in healthy controls. 

Materials and Methods

Ethics Approval

This study was approved by the ethics committee of the First Affiliated Hospital of Chongqing Medical University (Chongqing, China) with permit number 2009-201008. All participants signed a written informed consent and the project was carried out in accordance with the tenets of the Declaration of Helsinki. 

Study Participants

In the present study, a total of 676 AAU patients, including 298 AAU patients with AS (AAU⁺AS⁺), 378 AAU patients without AS (AAU⁺AS⁻), and 596 unrelated healthy controls were enrolled. All participants were Chinese Han and were recruited at the Ophthalmic Center of the Sun Yat-sen University (Guangzhou, China) and the First Affiliated Hospital of Chongqing Medical University (Chongqing, China) from 2005 to 2013. Diagnosis of acute anterior uveitis was primarily made through symptoms (red eyes, ocular pain, and vision loss), clinical examination (ciliary congestion, keratic precipitates, anterior chamber cells). Instrumental exams, such as anterior segment image and ultrasound biomicroscopy, were helpful for diagnosis. All patients with AS strictly fulfilled the modified New York criteria 1984 for AS diagnosis.²⁴ 

Genomic DNA Extraction and Measurement of CNVs

Genomic DNA was extracted (with Qiagen QIAamp DNA kit; QIAGEN, Inc., Hilden, Germany) from peripheral blood samples according to the manufacturer's instructions. The concentration of genomic DNA was detected by a spectrophotometer (NanoDrop 2000; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and template DNA was diluted at 200 μg/mL stock and stored at −20°C for later use. 

The copy number variations of the candidate genes in this study were determined using real-time quantitative PCR (RT-qPCR, TaqMan; Applied Biosystems, Foster City, CA, USA) method and was performed on a real-time PCR system (model 7500; Applied Biosystems).²⁵ Assays (labeled with FAM; Applied Biosystems) and commercial primers (Table 1) were used to detect the CNVs. The assay RNase P (labeled with VIC; Applied Biosystems) is used as a standard reference assay for the CNV as recommended by the protocol of the manufacturer. Real-time PCR data were analyzed using commercial software (model 7500, version 2.0.6; Applied Biosystems) and the copy numbers were calculated using the comparative Ct(2^–ΔΔCt) method by PCR software (CopyCaller, version 2.0; Applied Biosystems). 

Table 1

 

CNV Assays Used in This Study

 

Data Analysis

The copy number (CN) of T-bet, GATA-3, and RORC was classified into three categories, which were high CN (>2), normal CN ( = 2), and low CN (<2). Since FOXP3 is located on chromosome X, the CN of FOXP3 was analyzed in female and male participants separately. In female participants, categories were similar as for the other three genes. However, in male participants, the cutoff point was CN = 1 to classify the categories, that was high CN (>1), normal CN ( = 1), and low CN (<1). Based on the three categories of CN, the difference of the four candidate genes was estimated by χ² testing and the risk associated with CNVs of candidate genes was estimated by comparing cases and control cohorts. All data were analyzed using predictive analytics software (SPSS, version 17.0; SPSS, Inc., Chicago, IL, USA). Bonferroni correction was applied for multiple testing (n = 14). 

Results

The demographics and clinical features of all participants are summarized in Table 2. In our study, CNVs of T-bet, GATA-3, RORC, and FOXP3 genes were examined and compared in AAU patients and healthy controls. We then performed the comparisons between AAU⁺AS⁺ patients and AAU⁺AS⁻ patients versus healthy controls, respectively. 

Table 2

 

Clinical Features of Participants Recruited in This Study

 

CNV Distribution of T-bet, GATA-3, RORC, and FOXP3 in AAU Patients

Copy number variations of four CD4⁺ helper T-cell–related transcription factor genes were compared between AAU patients (cases) and healthy controls. An increased CN frequency of T-bet was found in AAU patients (P_corr = 1.0 × 10⁻⁹, OR = 2.2, 95% confidence interval [CI] 1.71–2.71; Table 3). Additionally, we also found that the CN frequency of GATA-3 was significantly increased in AAU patients (P = 4.2 × 10⁻⁵, OR = 3.5, 95% CI 1.99–5.97). Since FOXP3 is located on chromosome X, we analyzed CN of FOXP3 in male and female patients separately. A high CN frequency of FOXP3 was observed in female AAU patients (P_corr = 0.001, OR = 5.2, 95% CI 2.10–13.03; Table 2). No association with FOXP3 CN was found in male AAU patients. The frequency distribution of RORC-CN frequencies was not different between the patient groups and healthy controls. 

Table 3

 

CNV Frequencies of T-bet, GATA-3, RORC, and FOXP3 Genes in AAU⁺AS⁺ and AAU⁺AS^– Patients and Controls

 

CNV Stratification Analysis of T-bet, GATA-3, RORγt, and FOXP3 in AAU⁺AS⁺ and AAU⁺AS^– Patients

The frequency of a high T-bet CN was increased in AAU⁺AS⁺ as well as in the AAU⁺AS⁻ group (P_corr = 4.3 × 10⁻⁵, OR 2.0, 95% CI 1.48-2.64 and P_corr = 1.2 × 10⁻⁸, OR = 2.3, 95% CI 1.76–3.01, respectively). A significantly higher GATA-3 CN frequency was found in AAU⁺AS⁺ patients (P_corr = 1.8 × 10⁻⁷, OR = 4.9, 95% CI 2.69–8.80), but the significance was lost after Bonferroni correction in the AAU⁺AS⁻ group. An increased frequency of a high FOXP3 CN was found in both female AAU⁺AS⁺ and female AAU⁺AS⁻ patients (P_corr = 0.005, OR = 5.9, 95% CI 1.99–17.63 and P_corr = 0.004, OR = 4.9, 95% CI 1.75–11.94, respectively ), but there was no difference in male patients versus controls. No association of CNVs of RORC was observed between AAU⁺AS⁺ or AAU⁺AS⁻ patients and controls (Table 4). 

Table 4

 

CNV Stratification Analysis of T-bet, GATA-3, RORC, and FOXP3 in AAU⁺AS⁺ and AAU⁺AS^– Patients

 

Discussion

In this study, we show that a high copy number of the CD4⁺T cell transcription factors genes, T-bet and GATA-3, confer susceptibility to AAU and that a high copy number of the FOXP3 gene is associated with AAU in females. 

Copy number variation of DNA, defined as segment repeats which display copy number difference, are associated with segmental duplications.²⁶ In the present study, we investigated intragenic variation of the four target genes. To avoid influence of ends of the gene, the CNVs we chose were located in the middle of the gene and have a comparatively longer amplicon length. 

The first immunogenetic studies in AAU led to the discovery of the strong association of HLA-B27 with AAU.²⁷ These older studies were based on serological methods and have now been replaced by modern molecular biological techniques. Besides the association with HLA-B27 (class I major histocompatibility complex [MHC]), evidence for an association of AAU with class II MHC genes has also been reported.²⁸ Recently, a high throughput genotyping method was used in a multicenter study²⁹ to investigate the genetic associations of AAU. The results showed that single nucleotide polymorphisms (SNPs) located in or around MHC class I or class II are associated with the susceptibility to AAU. Furthermore, the study found that SNPs of interleukin (IL)-6 and IL-10, the marker cytokines of the Th2 pathway, are also associated with AAU. Earlier observations described that cytokines of the Th1 pathway such as TNF³⁰ were associated with AAU. The above evidence provided support for a role of the CD4⁺ T cell pathway in the etiology of AAU. In the present study, we found that an increased frequency of a high T-bet (Th1) and GATA-3 (Th2) CN were associated with AAU, supporting the hypothesis that a dysregulation of T-cell–mediated immune responses may contribute to the etiology of AAU. A high CN of T-bet has also been shown to be associated with systemic lupus erythematosus (SLE) in Chinese patients.²¹ Although T-bet was classically considered as a TH1 transcription factor it is now clear that it also has many other functions and that it bridges innate and adaptive immune systems.³¹ Which of these various functions play the most important role in AAU pathogenesis remains to be clarified. Triggers of GATA-3 the secretion of IL-4, IL-5, and IL-13 from Th2 cells.³² Furthermore, it induces the differentiation of Th0 cells into Th2 cells and suppresses Th1 cell differentiation. Why an increased number of CNV in the GATA-3 gene would confer a risk for AAU when this is considered to downregulate the pathogenic Th1 response seems irrational at first sight. Earlier studies in psoriasis for instance showed that a decrease in mRNA GATA-3 is associated with inflammation of the skin.³³ On the other hand, it was recently shown that GATA3 plays an important role in the development and function of intestinal group 3 innate lymphoid cells, which include RORγt and IL-17A expressing CD4+ lymphocytes that have been implicated in mucosal immunity.³⁴ Acute anterior uveitis and AS are thought to be triggered by the intestinal flora³⁵ and changes in the regulation of protective mucosal immunity caused by an altered expression of GATA-3 may provide an explanation for our findings. Although the effect of gene copy number on GATA-3 expression has not yet been shown in humans, experiments with mutant and transgenic mice have shown that GATA-3 levels can be altered by influencing the GATA-3 gene copy number. Similarly, a high CN of T-bet has been shown to significantly affect its mRNA expression.²¹ 

An association with FOXP3 gene CN was only observed in female AAU patients. Forkhead box protein 3 is a transcription factor that plays a critical role in the development and function of regulatory T cells.³⁶ A dysregulation of regulatory T-cell activity may play an important role in the pathogenesis of autoimmune diseases such as uveitis.³⁷ Our previous study has shown that the polymorphisms of certain genes which are related to the function of regulatory T cells contribute to the genetic susceptibility of AAU.³⁸ Why the FOXP3 gene CN was only associated with female AAU and not with male AAU is not clear and deserves further investigation. Earlier studies have shown sex differences in the occurrence of HLA-B27 associated AAU, which might provide an explanation for our findings. It is also intriguing that a factor that is involved in the function of regulatory cells would be increased in inflammatory disease. However, it should be noted that the number of individuals with a high FOXP3 gene copy number is low and further confirmation of these findings is necessary. As yet it is not known how a high FOXP3-CN affects its biological function, although earlier studies from our group did not show an effect of FOXP3 gene CN on its mRNA expression.²² 

We also tested CN of the RORC gene but did not find a significant association. Retinoid-related orphan receptor γt controls the differentiation of Th17 cells, which produce IL-17 as its characteristic cytokine.³⁹ Several studies have reported the association of Th17 or IL-17 with immune diseases, such as autoimmune arthritis⁴⁰ and SLE.⁴¹ A higher level of Th17 cells has been observed in peripheral blood of patients with AS in different ethnic groups including Chinese,^42–44 implying Th17 cells in the pathogenesis of AS.⁴⁵ The absence of an association with RORC-CN in our AS patients suggests that the copy number of this gene may not have a large effect on the function of this gene but functional experiments need to be performed to support this hypothesis. 

Gene copy number variations have been shown earlier to affect the predisposition to a variety of diseases and our study expands earlier observations concerning the association of CNVs with uveitis including IL-17F, IL-23A, and complement component C4.⁴⁶ Interestingly, an earlier study on the association of CNVs with other types of uveitis, such as VKH syndrome and BD, observed that high CNVs of RORC predispose to BD.⁴⁴ This discrepancy with AAU, where no effect of RORC could be shown, may be due to the different mechanisms involved in the pathogenesis of these uveitis entities. On the other hand, it is possible that RORC gene copy number might be correlated with the clinical outcome of AAU. Further longitudinal studies may shed more light on this matter. 

Gene HLA-B27 has been reported to have a strong association with AAU, and is positive in 45% to 71% of AAU patients.⁵ Gene HLA-B27 as a class I HLA molecular, plays a major function in presenting antigen via the T-cell receptor of class I–restricted CD8⁺ T cells.⁵ Whether the association with HLA-B27 has an additional effect on the observed associations with the gene copy numbers of the investigated transcription factors was not investigated in our study, because the HLA-B27 status of our AAU patients was incomplete and since we did not type the healthy controls for HLA-B27. Further studies which should also incorporate the HLA-B27 subtypes such as B*2704 and B*2706 are needed to adequately address this issue and may be included in our future research. 

Limitations of our study include the fact that we were not yet able to assess the exact role of HLA-B27 and the absence of functional data to support the observed associations with GATA3 and FOXP3. Another limitation is the fact that we only recruited Chinese Han subjects and confirmation of our findings in other ethnic populations is needed. Furthermore, we only recruited patients visiting an ophthalmology department and therefore only studied uveitis patients with AS and a separate control study should be performed in the future to study AS patients without uveitis. This may answer the question whether our findings are specific for ocular inflammation or may also be involved in autoinflammatory joint disease. 

In conclusion, our study showed that copy number variations of T-bet and GATA-3 genes were associated with acute anterior uveitis and ankylosing spondylitis with acute anterior uveitis. Copy numbers of the FOXP3 gene are associated with female AAU patients with or without AS. The results provide further support in our understanding of the genetic basis for acute anterior uveitis and ankylosing spondylitis. 

Acknowledgments

The authors thank all the participants who joined in this study. 

Supported by a Natural Science Foundation Major International (regional) Joint Research Project (81320108009); Key Project of Natural Science Foundation (81130019); National Natural Science Foundation Project (31370893); Research Fund for the Doctoral Program of Higher Education of China (20115503110002); basic research program of Chongqing (cstc2013jcyjC10001); Chongqing Key Laboratory of Ophthalmology (CSTC, 2008CA5003), Key Project of Health Bureau of Chongqing (2012-1-003); Chongqing Science & Technology Platform and Base Construction Program (cstc2014pt-sy10002); National Natural Science Foundation Project (81522013, 81270990), Chongqing Outstanding Youth Grant (cstc2014jcyjjq10005); and National Key Clinical Specialties Construction Program of China. 

Disclosure: L. Bai, None; Y. Liu, None; S. Hou, None; D. Liao, None; A. Kijlstra, None; P. Yang, None