Association of TGFB -509C>T promoter polymorphism with primary angle closure glaucoma in a North Indian Punjabi cohort

Purpose Transforming growth factor beta (TGFB) is an important candidate gene implicated in glaucoma pathogenesis because it affects retinal ganglionic cell survival. The present study assessed the genetic association of -509C > T variant in the TGFB promoter region with primary open angle glaucoma (POAG) and primary angle closure glaucoma (PACG) in a North Indian Punjabi population. Method A total of 867 subjects (307 POAG, 133 PACG cases and 427 controls) were recruited from the targeted population. Genotyping was done by PCR-RFLP method and the data was analyzed using PLINK software (v1.07). Logistic regression under different genetic models was applied and genotype phenotype correlation was assessed by one-way ANOVA. Result A statistically significant difference in the frequency of heterozygotes among PACG cases (53.16%) and controls (30.07%) (p = 0.0002) was observed. Genetic model analysis revealed that mutant “TT” genotype conferred 2-fold risk towards PACG development under recessive model (p = 0.0019) while dominant model and co-dominant model provided 0.62 and 0.37 fold protection against PACG (p = 0.025 and p = 0.0001, respectively). Data segregation based on sex revealed a strong protective effect of heterozygous ‘CT’ genotype against progression of PACG among females (p = 0.002, OR = 0.37, 95% CI = 0.19–0.70), but conferred 2.14-fold risk among female POAG subjects (p = 0.013). Conclusion The study revealed a strong genetic association of -509C > T variant in TGFB with PACG in females. There is a need to replicate the results in a larger PACG cohort in other populations and further assess the contribution of sex specific factors in modifying genetic susceptibility to PACG.


Introduction
Glaucoma is a group of optic neuropathies with multifactorial etiology [1]. Being the second leading cause of blindness worldwide, it is a serious health condition with high socio-economic impact. Its prevalence is expected to rise to 111 million by 2040 with primary open angle (POAG) and angle closure glaucoma (PACG) contributing to bulk of the cases [2]. Common risk factors for the disease include advanced age, elevated intraocular pressure (IOP), high refractive errors (± > 3.00D), decreased corneal thickness, race/ethnicity, and familial history [1,3]. The adverse visual outcome which is a major concern for glaucoma patients occurs due to gradual degeneration of retinal ganglion cells (RGCs). Sufficient evidence exist to suggest the involvement of chronic neuroinflammation due to altered levels of cytokines under glaucomatous stress conditions in contributing to RGCs death [4,5]. Among the various cytokines implicated in the pathogenesis of glaucoma, transforming growth factor beta (TGFB) is significant because of its multifaceted role in the ocular milieu [6] specifically in affecting IOP dynamics. Several ex-vivo and in-vitro studies have established that high TGFB level is one causative factor for elevated IOP and patients with increased levels of this cytokine in the aqueous humor (AH) are at a higher risk to develop elevated IOP [7][8][9]. In one such study by Bhattacharya et al., infusion of TGFB-2 (an isoform of TGFB) to monkey and pig organ-cultured anterior segments led to elevated IOP [10]. An increased IOP was also observed after adenoviral vector mediated active TGFB-2 transfer in rodents [7]. TGFB also mediate excess extracellular (ECM) deposition to induce fibrosis [11] and since in glaucoma extensive ECM remodeling occurs in trabecular meshwork (TM) and ocular nerve head (ONH), dysfunctional TGFB signaling has been associated with glaucoma pathology [8,12]. In cultured human TM cells, exogenous TGFB treatment may increase the production of proteoglycans and other ECM proteins in these cells, which further reduce the outflow facility of the anterior chamber and thus contribute to increased IOP in the eye of glaucoma patients [13]. TGFB is also involved in the formation of scar tissue in glaucoma via eliciting production of matrix metallopeptidases (MMP)-2 [14]. High levels of TGFβ1 and TGFβ2 isoforms have been reported in different eye fluids including AH and vitreous humor (VH) of glaucoma patients [6,15]. High intensity fluorescence was detected in immunohistochemistry assay of ONH samples for the receptors of TGFβ-1 isoform in glaucomatous eye [16]. Thus, there is a strong evidence for the possible involvement of TGFβ in glaucoma pathology. Additionally, the promoter and exonic SNPs appear to have an impact on the expression level of TGFB and may contribute to inter-individual differences in TGFB levels and hence modify genetic susceptibility to glaucoma [17,18]. -509C > T promoter variant in TGFB1 is one such variant which alters its expression [17,19]. The mutant allele 'T' at this site has been associated with increased TGFB-1 expression [17,20,21]. Therefore, in the present study, association of the TGFB-1 -509C > T promoter polymorphism (rs1800469) with primary glaucoma (POAG/ PACG) was investigated in a North Indian Punjabi cohort.

Sampling and DNA isolation
Study participants were recruited from Baba Deep Singh Eye hospital, Amritsar (Punjab), India after obtaining written informed consent. The study was approved by the Research Ethical Committee of Guru Nanak Dev University, Amritsar (Institutional Ethical certificate approval number: No. 702/HG/28-06-2019), and the study protocols were in accordance to the principles of Declaration of Helsinki. 440 primary glaucoma patients (POAG = 307 and PACG = 133) were enrolled. For POAG patients, the inclusion was based on following criteria: IOP of greater than 21 mmHg in either of the eyes tested using Goldmann Applanation Tonometry, glaucomatous ONH damage defined as a vertical cup disc ratio (VCDR) 0.7 or greater as adjudged clinically on slit lamp biomicroscopy using hand held + 90 D. This was confirmed using contrast enhanced fundus photograph on optical coherence tomography (OCT) as well as optic disc analysis or glaucomatous visual field defect as detected on automated perimeter using Humpherys Visual Field Analyser using Swedish Interactive Thresholding Algorithm (SITA) standard protocols. PACG cases were also recruited on the basis of above-described criteria along with the presence of at least 180 degrees of closed angle in which the TM is not visible on gonioscopy. The control data set consisted of 427 unrelated age and gender matched subjects without any family history of glaucoma. The controls were all examined, prior to cataract surgery, for IOP less than 21 mmHg, normal visual field, normal optic nerve heads with CDR of < 0.5. Individuals with known chronic systemic inflammatory diabetes (random blood sugar-140 mg/Dl; according to American Diabetes Association (ADA) guidelines), autoimmune or immunosuppressive disease as well as a preexisting ocular disease (diabetic retinopathy, age-related macular degeneration) were excluded from the study. Individuals having ocular hypertension (OHT) were also excluded from the study group. 3 ml of blood sample was collected by venipuncture by a sterile 5 ml of syringe, in a vacutainer containing 0.5 M of EDTA, from all study participants from a local eye hospital after a complete ophthalmic examination. After collection, the blood samples were transferred to the lab in an insulating container containing icepacks and storage was done at − 20°C for further use.
Genomic DNA was then extracted by using phenol chloroform method [22]. All samples were diluted to 50 ng/μl concentration. Quantification of extracted DNA was done using NanoDrop ND-2000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).

PCR amplification, Genotyping and Electrophoresis
The target promoter flanking region of TGFB -509C > T of 400 bp was amplified with the primer sequences FP-5′ CCAGCTAAGGCATGGCACCG 3′ and RP 5′ GCGGTGTGGGTCACCAGAGA 3′ designed by Primer 3 software. The PCR was performed in a final reaction volume of 15 μl containing 50 ng DNA template, 1X Taq Polymerase buffer reagent, 2.5 mmol of each deoxynucleotide triphosphate (dNTPs) (3B BlackBio Biotech India Ltd), 20 mmol MgCl 2 , 0.3 μmol/μl of each primer, and 0.09 units Taq DNA polymerase (3B BlackBio Biotech India Ltd.; 5 U/μl). The cycling conditions were: initial denaturation at 95°C for 10 min, followed by 35 cycles at 95°C for 45 s, 65°C for 45 s and 72°C for 45 s, with a final extension at 72°C for 10 min. After amplification, the PCR products were digested with 3 units of Bsu-36I (New England Biolabs) for 12 h at 37°C. The digested products were separated by electrophoresis. Briefly 2.5 g of agarose (Sigma-Aldrich) was dissolved in 100 ml of 1X TBE buffer (pH 8.0) by boiling the solution. 5 μl of ethidium bromide (10 mg/ml) (GeNei™) was added after allowing the solution to cool to 60°C. The gels were allowed to solidify and placed in the horizontal electrophoresis chamber containing 1X TBE buffer (pH 8.0). 3 μl of each PCR product and 2 μl of gel loading dye were mixed and loaded in each well. For RFLP, 10 μl of digested products mixed with 5 μl gel loading dye were loaded in each well. The appropriate molecular weight marker was loaded in parallel to compare the sizes of PCR products and RFLP digested amplicon. Electrophoresis was carried out at 100 V for 30 min and the products were visualized under UV transilluminator and photographed. The genotypes were scored on the basis of Restriction Fragment Length Polymorphism (RFLP) pattern as given in Fig. 1.

Statistical analysis
Descriptive statistical analysis for demographic and clinical characteristics for study participants was performed using GraphPad Prism 5.0. The results were tabulated as mean ± standard deviation (SD). Association analyses were performed using PLINK software (v1.07). Odds ratios (ORs) for various genetic models, dominant, recessive, additive and allelic were calculated. P-values of less than 0.05 were considered to be statistically significant.
To investigate the association of the variant with IOP and VCDR which are known endophenotypes of glaucoma, the values for right eye was chosen arbitrarily (as the mean for both eyes was same, Table 1) for ANOVA. Binary logistic regression was applied for correcting possible confounders by using SPSS Software Version 21 [23].

Demographic and clinical characteristics of the study participants
The frequency distribution of demographic factors, sex and age among patients and control subjects is shown in Table 1 and Fig. 2. The mean age among primary glaucoma cases was 59.92 years with standard deviation (SD) of ±12.81, while the mean age among controls was 59.46 with SD of ±11.56. Males had higher prevalence of POAG (65%) while females had a higher prevalence of PACG (64%) in accordance with several other epidemiological surveys where a higher prevalence of POAG has been reported in males while females are at a higher risk for PACG [24]. Total number of males and females in cases (both POAG/PACG) and controls did not show any significant difference (p = 0.558) as given in Fig. 2.
These results highlight the gender difference in glaucoma prevalence; women are known to be at a higher risk for PACG [24]. The mean CDR for right and left eye in cases was 0.72 and 0.73, respectively, which was higher with respect to control subjects (0.23 and 0.25)   Table 2).

Significant association of -509C > T polymorphism with POAG/PACG subgroups in females
The data was stratified based on sex and reanalyzed (

Association of IOP and CDR with TGFB -509C > T polymorphism
No significant difference in the mean values of CDR (p = 0.563) and IOP (p = 0.108) was observed among CC, CT and TT genotypes, although mean value of CDR was higher among heterozygotes (Table 4).

Discussion
In the current analysis, association of -509C > T promoter variant in TGFB1 with primary glaucoma was assessed. In the ocular microenvironment, TGF-B1 is mostly found in the AH, VH and tears [25][26][27] where it is involved in the production of proteoglycans like chondroitin sulfate. The latter is the main component of the ECM of the TM cells, which alters AH outflow from the eye. Therefore, any alteration in TGFB1 levels may affect Fig. 2 Frequency distribution of males and females among POAG and PACG cases with respect to control subjects the AH outflow facility, and thereby contributing to increase in IOP which is the main risk factor for glaucoma [28,29]. Studies conducted in the past 20 years have repeatedly demonstrated elevated levels of TGFB in the AH of POAG patients [30][31][32][33][34].
In the present study, a statistically significant difference was observed between the frequency of heterozygotes 'CT' among PACG subgroup of patients (53.16%) and controls (30.07%) (p = 0.0002). However, allele frequencies did not differ in combined (p = 0.804) as well as two glaucoma subtypes, POAG/PACG cases (p = 0.913; p = 0.710, respectively). On segregating the data into males and females, the study found a protective effect of heterozygous 'CT' genotype against progression Such sex-specific differences in disease susceptibility are also known in cardiovascular and autoimmune diseases and may be attributed to hormonal and epigenetic differences between the sexes [35]. In  CDR-Cup-to-disc ratio; IOP-Intraocular pressure p < 0.05 were considered to be statistically significant glaucoma, however the effect of sex in genetic association with SNPs has not been widely investigated and to the best of our knowledge, only a single study determined whether genetic association at the 9p21 locus is influenced by sex in POAG [36]. Further, no study till date has assessed genetic association with SNPs in PACG in sex specific context. The results for female specific association of TGFB in PACG in the present study may suggest a plausible crosstalk pathway between estradiol and cytokines signaling. There is sufficient evidence in favor of role of estradiol in pathogenesis of glaucoma and it may have therapeutic potential for POAG in future [37][38][39]. Among the pleiotropic effects of estrogen, one role has been in modulating expression of anti-inflammatory cytokines like TGFB [40] through which it may alter the cytokine balance in the ocular microenvironment and increase susceptibility to glaucoma. The negative results with POAG in current study are similar to another study conducted on South Indian population (POAG cases = 106 and controls = 104) where no association was reported between TGFB-1 -509C > T polymorphism and POAG or IOP and CDR [41]. In contrast to these reports, significant association of -509C > T polymorphism was reported with POAG among East Iranian population (POAG cases = 112 and controls = 112) with 'T' allele as the risk allele (p = 0.005; OR = 1.73). The frequency of heterozygous 'CT' genotype (45.6%) was higher among POAG patients as compared to controls (38.4%) (p = 0.020; OR = 1.93) [42]. Grainger et al. also observed that the allele 'T' at -509 position, confers risk towards POAG progression and led to higher plasma levels of TGFB as compared to those who carried 'C' allele at this position [17]. However, lack of inclusion of PACG subcategory, absence of sex-based stratification among POAG and PACG cases despite reports stating sex to be an important risk factor for glaucoma, and insufficient sample size were some limitations of the previous studies. With a robust sample size and sex-based segregation in POAG and PACG performed in the present study, it can be concluded that TGFB gene variant is associated with PACG in the North Indian Punjabi population. For − 509 > C variant in TGFB, a potential female specific association was observed with PACG risk. The present study provides a strong impetus to conduct additional analyses in other populations to specifically test for potential sex specific effects in the genetic architecture of POAG and PACG.