Glaucoma is a complex multifactorial disease with high heritability and ethnic-specific predisposition, suggesting the involvement of genetic components in its pathogenesis (Chan et al., 2016). Among the various forms of glaucoma, primary angle-closure glaucoma (PACG) is a far more common type than primary open-angle glaucoma (POAG) in Asian populations, including Saudi Arabia (Al Obeidan et al., 2011; Chan et al., 2016) and involves anatomical blockage of the aqueous flow pathway (Weinreb et al., 2014). Likewise, pseudoexfoliation glaucoma (PXG) is the most severe type of open-angle glaucoma highly prevalent among the elderly, characterized by the abnormal deposition of pseudoexfoliative material (fibrillar extracellular matrix) in the anterior segment of the eye causing the aqueous blockage and associated with worse prognosis (Vazquez and Lee, 2014). Several disease-causing mutations and multiple susceptibility loci have been associated with different types of glaucoma in various ethnicities (Zukerman et al., 2021). However, the precise role of these genetic components and their underlying molecular mechanisms in glaucoma pathogenesis remains unclear.

Overlapping clinical manifestations among glaucoma types, such as high intraocular pressure (IOP), optic nerve damage, and retinal ganglion cell (RGC) apoptosis, suggest the plausibility of common disease pathways and genetic mechanisms in glaucoma pathogenesis. The previous genetic association studies of MYOC (Dai et al., 2008), CYP1B1 (Chakrabarti et al., 2007), LOXL1 (Shiga et al., 2018; Eliseeva et al., 2021), ARHGEF12 (Aung et al., 2018), and ACVR1 (Kondkar et al., 2020) that have been common to different types of glaucoma and therefore lend further support to this hypothesis. Besides, most of the studies investigating glaucoma genetics were conducted in Asian or Caucasian populations. Depending on the population being assessed, the same variants may or may not show an association. Therefore, replication studies must be performed to confirm the effects of these single nucleotide variants in other regions or with other ethnicities. The genetic basis of PACG and PXG in middle-eastern patients of Saudi origin is still largely unknown (Abu-Amero et al., 2010; Abu-Amero et al., 2013, 2015; Kondkar et al., 2021) and therefore warrants further genetic investigation. Accordingly, we hypothesized that genetic polymorphisms rs2472493 located upstream of the ABCA1 gene and rs7636836 in FNDC3B were previously reported to be associated with POAG and elevated IOP (Hysi et al., 2014; Shiga et al., 2018), may also be associated with PACG and PXG in Saudi patients.

ABCA1 functions as a cholesterol efflux pump, and recent findings suggest its role in neuroinflammation, neurodegeneration, and RGC apoptosis (Howell et al., 2011; Awadalla et al., 2013). Whereas FNDC3B codes for an extracellular matrix protein involved in several signaling pathways including, transforming growth factor-beta (TGFβ) signaling (Prendes et al., 2013). Both genes are highly expressed in the human retina, optic nerve, trabecular meshwork, and RGC supporting its alleged role in glaucoma development and/or progression (Shiga et al., 2018). The genetic contribution of the variants at the ABCA1 and FNDC3B locus among Saudi PACG and PXG patients is unknown. To date, there has been no study to address the relationship between the polymorphisms in ABCA1 and FNDC3B in PACG and PXG in Saudi Arabs. Thus, in this exploratory study, we investigated whether single-nucleotide polymorphisms (SNPs) rs2472493 and rs7636836 upstream of ABCA1 and in FNDC3B genes, respectively, were associated with PACG and PXG in a Saudi cohort.

Glaucoma is a complex polygenic disorder affected by multiple genetic and environmental factors. Previously, several genome-wide association studies have identified numerous genetic polymorphisms and susceptibility loci in different populations worldwide, few of which have been unique to specific ethnic groups (Zukerman et al., 2021). However, the genetic etiology of glaucoma in middle-eastern population is still lacking and therefore warrants further studies. Given the overlapping clinical manifestations and common genetic variants reported among glaucoma types, we investigated the association of two POAG-related gene polymorphisms, rs2472493 in ABCA1 gene and rs7636836 near FNDC3B, in the PACG and PXG patient cohort of Saudi origin.

In our study, the MAF of rs2472493 in ABCA1 was 0.44 and 0.37 in PACG and PXG, respectively, and were comparable to those in controls (0.39) and thus non-significant. The allele frequencies were similar to those reported in the Han Chinese and Uygur Chinese population of PACG (0.45) and PXG (0.40) patients, respectively (Luo et al., 2015). Likewise, the MAFs were comparable to the Hispanics (0.37), non-Hispanics (0.42), and African American (0.35) POAG subjects (Choquet et al., 2018), but lower than the East-Asians (0.56) (Choquet et al., 2018) and the European (0.51) POAG patients (Gharahkhani et al., 2014) highlighting slight ethnic and glaucoma type variability.

Previously, SNP rs2472493 in ABCA1 was associated with POAG in the genome-wide meta-analysis of 18 population cohorts from the International Glaucoma Genetics Consortium (Hysi et al., 2014). Similar genome-wide findings were reported by Gharahkhani et al., in the Australian cohort (Gharahkhani et al., 2014) and by Chen et al., in POAG patients from Southern China (Chen et al., 2014). Thus, multiple population studies indicated that the ABCA1 gene might play a crucial role in POAG pathogenesis. In contrast, negative ABCA1 gene associations have been reported in the Han Chinese cohort of 1,311 healthy controls and 1122 PACG patients (Luo et al., 2015), and in the 52 Jordanian Arab glaucoma patients (POAG and congenital) and 96 healthy individuals investigated for the ABCA1 rs2472493 variant by Alkhatib et al. (2019). Although our study did not replicate the findings in the PXG and PACG patient cohort of Saudi origin, we observed a significant association of ABCA1 rs2472493 polymorphism among men in the PXG cohort. However, there are no published reports of this variant being examined in PXG.

ABCA1 belongs to a large superfamily of ABC transmembrane transporters and is best known for its role in cholesterol efflux to lipid-free apolipoprotein AI and apolipoprotein E (Wang and Smith, 2014). ABCA1 is expressed in all the major tissues of the eye, including the trabecular meshwork, Schlemm’s canal endothelial cells, RGCs, and optic nerve that are primarily involved in glaucoma (Chen et al., 2014). However, the exact mechanism(s) by which ABCA1 may be involved in glaucoma pathogenesis is still unclear. ABCA1 was demonstrated to inhibit ocular inflammation via activation of liver X-receptor in an experimental model of autoimmune uveitis (Yang et al., 2014). Using the glaucoma model, Li et al. demonstrated that ABCA1 was related to RGC death (Li et al., 2018). These studies suggest that ABCA1 may have a significant role in eye research. Likewise, using the Encyclopedia of DNA Elements (ENCODE) project data and Genevar database, Gharahkhani and colleagues, reported that the polymorphism rs2472493 located upstream of ABCA1 is an expression quantitative trait locus (eQTL) in lymphoblastoid cell lines that might alter the motif sequences for proteins such as FOXJ2 and SIX5 (Gharahkhani et al., 2014). Also, rs2472493 was found to be in high linkage disequilibrium with polymorphism rs2472494 near ABCA1 that alters the regulatory motif for binding of PAX6, a gene involved in eye development. The authors thereby predicted a possible regulatory role for this polymorphism in gene expression in a pathway similar to that of rs2472494 variant near ABCA1 gene (Gharahkhani et al., 2014). However, the exact mechanism(s) by which rs2472493 polymorphism in ABCA1 might increase PXG risk in men is unknown. The effect of gender, gene-gene, and gene-environment interactions on ABCA1 gene polymorphisms in lipid and lipoprotein metabolism have been well documented (Junyent et al., 2010; Coban et al., 2014; Shi et al., 2021). The ABCA1 polymorphism might plausibly modulate the risk of PXG among men in a similar manner.

FNDC3B (also known as FAD104) is a known regulator of adipogenesis (Tominaga et al., 2004). FNDC3B codes for an extracellular matrix protein that has a vital role in cell adhesion and growth signaling pathways, including TGFβ, and Wnt/β-catenin signaling (Nishizuka et al., 2009; Goto et al., 2017; Li et al., 2020) all of which have been strongly implicated in glaucoma pathogenesis (Prendes et al., 2013; Zhong et al., 2013; Webber et al., 2018). Furthermore, FNDC3B is expressed in all the primary eye tissues relevant to glaucoma (Li et al., 2015; Shiga et al., 2018).

The variant rs7636836 in FNDC3B was one of the novel loci associated with POAG in a two-stage genome-wide study consisting of 7,378 Japanese POAG cases and 36,385 controls conducted by Shiga et al. (2018). However, further validation in Singapore Chinese, European, and Africans did not show any significant association (Shiga et al., 2018). In our study cohort, the MAF of rs7636836 in FNDC3B was 0.05 and 0.07 in PACG and PXG, respectively. The frequency distribution was similar to European (0.06) but lower than the Japanese (0.40) and Singaporean Chinese (0.21) POAG subjects reported by Shiga et al. (2018). Our data showed no significant allelic and genotype association between rs7636836 in FNDC3B and PACG/PXG. However, in men, a modest allelic and genotype association of rs7636836 in FNDC3B was observed in PXG. The potential mechanism(s) by which rs7636836 in FNDC3B would modulate the risk of PXG in men is unclear but could possibly be hormonal (androgen) related via Wnt signaling which can be stimulated by FNDC3B (Li et al., 2020). There is evidence for crosstalk between androgen receptor and Wnt/ß-catenin signaling pathway (Mumford et al., 2018). The androgen receptors are expressed in ocular tissues (Tachibana et al., 2000) and Wnt signaling is strongly implicated in the maintenance of glaucomatous trabecular meshwork (Dhamodaran et al., 2020). To the best of our knowledge, there are no published reports of studies evaluating this polymorphism in PACG and PXG patients.

Gene-gene and gene-environmental interactions have been suggested to play an essential role in the pathogenesis of complex human diseases and highlight the possible contribution of genetic background and environmental triggers in disease development and progression (Marchini et al., 2005; Pan, 2008; Brossard et al., 2015). Glaucoma is also a complex polygenic disease with no clear inheritance pattern, and similar mechanisms may exist to influence the disease outcomes (Zakharov et al., 2013). Interestingly, a combined allelic and genotype analysis of rs2472493 in ABCA1 and rs7636836 in FNDC3B showed that G-T allele haplotype of ABCA1 and FNDC3B was associated with a significantly increased risk of PACG. Although it is difficult to ascertain whether the risk observed in our study is attributable to a real haplotype effect or probably reflects a linkage with other variant(s) not included in this study, the possible role of rs2472493 in ABCA1 and rs7636836 in FNDC3B in PACG cannot be completely ruled out and would need further research in a larger cohort.

Polymorphisms in ABCA1 and FNDC3B have also been shown to influence IOP (Hysi et al., 2014). A recent study showed that ABCA1 regulated IOP by modulating caveolin-1, nitric oxide/endothelial nitric oxide synthase signaling pathway (Hu et al., 2020). However, the analysis of polymorphisms effect on clinical endophenotypes of PACG and PXG, such as IOP and cup/disc ratio in our cohort, showed no significant association.

Thus, our results show that rs2472493 in ABCA1 and rs7636836 in FNDC3B may not have a major direct role in PACG and PXG in this ethnic group. However, the association of rs7636836 in FNDC3B and rs2472493 in ABCA1 observed among PXG men; and that of G-T haplotype in PACG suggests that they may have a significant indirect role (possibly via epistatic interaction(s)) in PXG and PACG. But since the study was performed in a relatively small sample size and does not provide any mechanistic evidence, the results would require a cautious interpretation. Based on the observed allele frequencies, the study had an estimated power of >0.9 per allele for the ABCA1 variant but was underpowered (0.6 per allele) for the FNDC3B variant to detect a relative risk of 2.0 with an alpha type I error of 0.05. Thus, the possibility of chance association cannot be ruled out and further emphasizes the need for replication in a large sample cohort potentially with age and gender-matched controls to confirm these findings.

In conclusion, our results suggest that the polymorphisms rs2472493 in ABCA1 and rs7636836 in FNDC3B genes may be associated with PXG among men, and a G-T allelic combination may confer an increased risk of PACG in the middle-eastern Saudi cohort. However, further investigations in larger population-based samples and different ethnicity are needed to draw definite conclusions and validate these findings. Moreover, considering the genetic heterogeneity of glaucoma per se, the plausible involvement of gene-gene and/or gene–environmental interactions must be important considerations for future research.