Complex genetic mechanisms in glaucoma: An overview

Glaucomas comprise a group of hereditary optic neuropathies characterized by progressive and irreversible visual field loss and damage to the optic nerve head. It is a complex disease with multiple molecular mechanisms underlying its pathogenesis. Genetic heterogeneity is the hallmark of all glaucomas and multiple chromosomal loci have been linked to the disease, but only a few genes have been characterized, viz. myocilin (MYOC), optineurin (OPTN), WDR36 and neurotrophin-4 (NTF4) in primary open angle glaucoma (POAG) and CYP1B1 and LTBP2 in congenital and developmental glaucomas. Case-control-based association studies on candidate genes involved in different stages of glaucoma pathophysiology have indicated a very limited involvement. The complex mechanisms leading to glaucoma pathogenesis indicate that it could be attributed to multiple genes with varying magnitudes of effect. In this review, we provide an appraisal of the various efforts in unraveling the molecular mystery in glaucoma and also some future directions based on the available scientific knowledge and technological developments.

Glaucomas are categorized into primary and secondary based on their etiology and aqueous humor dynamics. [9] Based on gonioscopy, primary glaucomas are further classified as primary open angle glaucoma (POAG) and primary angle closure glaucoma (PACG). POAG may be associated with or without an elevated IOP and has an adult onset (usually >35 years) or juvenile onset (usually <35 years). [9] Secondary glaucomas are characterized by the involvement of predisposing ocular or systemic diseases such as uveitis, trauma, and diabetes thereby resulting in an alteration of aqueous humor dynamics. These include pseudoexfoliation glaucoma (XFG) and pigmentary glaucoma (PG). [9] The mode of inheritance in adult-onset POAG and PACG is complex in nature. This has limited the identification of large affected families for gene mapping by linkage analysis. On the other hand, the hereditary component in juvenile-onset POAG has facilitated mapping of some candidate loci. [10] Developmental glaucomas include primary congenital glaucoma (PCG) and glaucoma-associated syndromes (Aniridia and Axenfeld Rieger syndrome). [9] These glaucomas largely follow a Mendelian pattern through autosomal dominant and autosomal recessive modes of inheritance. Thus, co-segregation of candidate gene mutations or the disease-susceptible alleles among the affected subjects in such families are relatively easy to determine. [11] Among these glaucoma subtypes, primary glaucomas (largely POAG) are the most common form and are attributed to multiple genes with varying magnitudes of effect. [12] Gene mapping in these disorders pose a major challenge but the success of new-generation technologies and highthroughput screening platforms have provided some hope in understanding their underlying molecular mechanisms. Herein, we present an update on the molecular genetics of primary glaucomas, particularly POAG.

Chromosomal Loci Mapped in POAG
Glaucoma being a complex disorder is attributed to several genes, the majority of which are yet unidentified. Until now, linkage analysis in large affected families has yielded 25 chromosomal loci linked to POAG. Of these, 15 have been designated as GLC1A to GLC1O by the HUGO genome nomenclature committee (www.gene.ucl.ac.uk/nomenclature) and mutations in only four genes have been identified in POAG. These include myocilin (MYOC, GLC1A), optineurin (OPTN; GLC1E), WD repeat domain 36 (WDR36, GLC1G) and neurotrophin-4 (NTF4, GLC1O). Further details are provided in Table 1. Although these genes harbor POAG-associated mutations, they exhibit a high degree of allelic heterogeneity in different populations. The mutation spectra in these genes do not indicate their overall involvement in the disease pathogenesis. The problem of understanding the underlying molecular mechanism is additionally compounded by the variable penetrance and expressivity of these gene mutations. The other loci that were mapped in POAG are yet to be characterized for the disease-association mutations.

Candidate Genes in POAG a) Myocilin (MYOC)
The first evidence of linkage in a large German family with juvenile open angle glaucoma (JOAG) was demonstrated by Sheffield et al, [10] on Chromosome lq21-q31 (GLC1A). Later Myocilin/TIGR (accession numbers: Nucleotide AH006047, Protein NP_000252) on this locus was characterized to be causative for POAG. [33] This gene is also termed as trabecular meshwork-induced glucocorticoid receptor (TIGR) as it overexpresses due to the induction of glucocorticoid to the cells.
It consists of three exons that code for an mRNA of about 2.5 kb and encompasses a 57-kDa glycoprotein of 504 amino acids. [34] Mutations in MYOC have been reported from different populations and account for 2-5% of POAG patients worldwide. [35] Around ~72 mutations have been reported till date and the Gln368Stop mutation is the most common mutation observed across multiple populations except the Japanese. [36] In the Indian scenario, the Gln48His was the predominant mutation observed in cases of JOAG, POAG and PCG. [37] Along with the coding region mutations, a few polymorphisms have been identified, of which the -1000C>G was referred to very frequently with variable degrees of association. [38] The probable disease-causing and benign variations in MYOC are available in the Myocilin allele-specific phenotype database (http://www. myocilin.com/variants.php). [39] Genotype-phenotype correlation has been demonstrated with some MYOC mutations. It has been observed that individuals with the T377M mutation usually have an onset in their fourth decade whereas those with P370L and Y437H mutations were diagnosed in the first and second decades with severe clinical presentations. The predominant mutation Q368X had an average onset in the fifth and sixth decades [40] and the C433R mutation exhibited an onset between 17 and 58 years and was largely associated with a higher IOP and vertical  [10] 2cen-q13 Adult onset GLC1B Stoilova et al. [13] 3q21-q24 Adult onset GLC1C Wirtz et al. [14] 8q23 Adult onset GLC1D Trifan et al. [15] 10p15-p14 Adult onset, NTG GLC1E OPTN Sarfarazi et al. [16] Rezaie et al. [17] 7q35-q36 Adult onset GLC1F Wirtz et al. [18] 5q22.1 Adult onset GLC1G WDR36 Monemi et al. [19] 2p16.3-p15 JOAG, Adult onset GLC1H Lin et al. [20] Suriyapperuma et al, [21] 15q11-q13 Adult-onset POAG GLC11 Allingham et al. [22] 9q22 JOAG GLC1J Wiggs et al. [23] 20p12 JOAG GLC1K Wiggs et al. [23] 3p21-p22 Adult onset GLC1L Baird et al. [24] 5q22.1-q32 JOAG GLC1M Pang et al. [25] 15q22-q24 JOAG GLC1N Wang et al. [26] 19q13.3 POAG and NTG GLC1O NTF4 Ip et al. [27] Pasutto et al. [28] 2q33.1-q33.3 POAG Nemesure et al. [29] 10p12.33-p13.3 POAG Nemesure et al. [29] 2p14 POAG Wiggs et al. [30] 14q11 POAG Wiggs et al. [30] 14q21-q22 POAG Wiggs et al. [30] 17p13 POAG Wiggs et al. [30] 17q25 POAG Wiggs et al. [30] 19q12-14 POAG Wiggs et al. [30] 1p 32 POAG Charlesworth et al. [31] 10q 22 POAG Charlesworth et al. [31] 2p14 Elevated IOP Duggal et al. [32] 19q12-q14 Elevated IOP Duggal et al. [32] IOP  [40] In humans the MYOC mRNA is expressed in a number of tissues which include both non-ocular and ocular tissues including the trabecular meshwork (TM) that exhibits the highest level of expression, followed by the sclera, ciliary body, choroid, cornea, iris, lamina cribosa, retina, and optic nerve. The non-ocular tissues include mammary gland, small intestine, thymus, prostate, testis, colon, stomach, thyroid, trachea, bone marrow, and brain. [42] Function of MYOC protein The normal physiological function of MYOC is still unclear but insights from the knockout animal models indicate that the disease-causing MYOC in humans may act by gain of function. In vitro studies have shown that MYOC is involved in the cytoskeletal organization and extracellular matrix (ECM) remodulation. [43] Mutations in MYOC may not directly affect its expression but it may interfere with protein folding or stability of the folded protein. The misfolded protein may not secrete and accumulate as soluble and insoluble aggregates but may associate with resident proteins of the endoplasmic reticulum (ER). This may lead to the activation of the unfolded protein response finally leading to apoptotic cell death. [44] The higher expression of MYOC in TM cells results in the intracellular accumulation of MYOC aggregates, which is deleterious to TM cells, thereby resulting in deterioration of their function and subsequent elevation of IOP. [44] Studies on the expression of normal and mutant MYOC in cultured ocular and non-ocular cells have suggested that while normal MYOC is secreted from the cultured cells, very little MYOC is secreted from cells expressing different mutations. Thus it can be surmised that glaucoma results either due to insufficient levels of secreted MYOC or compromised TM cell function caused by congestion of the TM secretory pathway. [45] MYOC is also known to be associated with the mitochondrial pathway and it has been shown that overexpression of MYOC carrying P370L mutation results in higher endogenous ROS (reactive oxygen species) production. This further suggests that mutant MYOC may cause mitochondrial defects which may lead to TM cell dysfunction and cell death. [46] b) Optineurin (OPTN) Sarfarazi et al., [16] identified the second gene in the GLC1E region (10p15-p14) called optic neuropathy-inducing protein or Optineurin (OPTN). Rezaie et al., [17] initially reported mutations in 16.7% families with hereditary POAG of which most of them had low or normal tension glaucoma (NTG). Subsequently, OPTN was screened across different populations that revealed few mutations implicated in NTG and POAG [ Table 2]. Among these mutations, the E50K located in the putative bZIP motif appears to be most strongly associated with NTG.
Structure and expression of OPTN OPTN contains three non-coding exons in 5'-untranslated region (UTR) and 13 exons that code for 577 amino acids. Alternative splicing in 5'-UTR generates at least three different isoforms, but all have same reading frame (gene accession no: AF420371 to AF420373). The structure of OPTN reveals several motifs, including one bZIP motif, two leucine zippers, coiledcoil motifs and a C-terminal C 2 H 2 -type zinc finger domain. OPTN is expressed in both non-ocular and ocular tissues that include the TM, non-pigmented ciliary epithelium, heart, brain, placenta, skeletal muscle, and kidney. [17] Functions of OPTN OPTN serves many cellular functions based on its interaction with a variety of proteins such as Rab8, Huntingtin, Myosin VI, RIP, transcription factor IIIA, metabotropic glutamate receptor and TBK1. Based on its interaction with Myosin VI, its role has been proposed in vesicular trafficking between the golgi and plasma membrane. [47] Overexpression of OPTN protects the cells from H 2 O 2 -induced cell death by inhibiting release of Cytochrome C from mitochondria. The common mutation E50K selectively induces the death of retinal ganglion cells (RGCs) due to TNFα-induced death of RGC. [48] Recently, it has been shown that OPTN negatively regulates TNFα-induced NFκB activation although the exact mechanisms by which these cytokines activate OPTN gene expression are yet unknown. [49] OPTN also plays an important role in the regulation of many genes which include MYOC although the mechanism involved is yet to be elucidated. [50] c) WDR36 Based on a genomewide scan, Monemi et al., (2005) characterized the WDR36 located on GLC1G locus (5q22.1) to be involved in POAG. They identified four mutations in WDR36 among 17 unrelated POAG subjects, 11 with high-pressure and six with low-pressure glaucoma. [19] The mutations were absent in 200 normal control chromosomes and their residues were conserved between WDR36 orthologs in mouse, rat, dog, chimpanzee and humans. Specific ocular expressions and the observed mutations were consistent with a role for WDR36 in the etiology of both high and low-pressure glaucoma. Analysis of WDR36 sequence revealed that several sequence alterations were exclusive to POAG patients and encoded predicted amino acid substitutions in conserved residues. [19] However, subsequent reports have revealed that the WDR36 may not be directly involved with POAG and may simply act as a modifier gene [ Table 3].

Structure and expression of WDR36
The gene spans about 38.3-kb genomic region and contains 23 exons expressed predominantly as two transcripts (5.

Functions of WDR36
Skarie et al., (2008) studied the functional role of WDR36 in zebra fish and found that it is essential for nucleolar processing of 18s rRNA and is thus required for the biogenesis of ribosomes. [65] In zebrafish, loss of WDR36 function resulted in reduced levels of 18s rRNA and also in ocular dysmorphology leading to activation of p53 stress response pathway. This may indicate the possible role of WDR36 sequence variants in POAG pathogenesis. Since variations in p53 are also known to be involved in POAG, co-inheritance of both P53 and WDR36 variations may thus be involved in the disease progression. [65] d) CYP1B1 The first genetic locus (GLC3A) for PCG was mapped by Sarfarazi and co-workers in 1995 based on their study of 17 Turkish families comprising 113 individuals that included 79 offsprings, of which 40 were affected with PCG. [11] Stoilov et al., (1997) identified CYP1B1 as a candidate gene at this locus for PCG [66] and subsequently pathogenic mutations in CYP1B1 have been identified in PCG with varying frequencies from 20-100% across different populations. [67][68][69][70][71][72] CYP1B1 has also been implicated in juvenile and adult onset forms of glaucoma, in various ethnic groups worldwide. Initially, Vincent et al., (2002) showed the involvement of CYP1B1 and MYOC in POAG through a digenic mechanism in a family of East Indian (Guyanese) origin. Based on these observations, it was suggested that PCG and JOAG are allelic variants of CYP1B1. It was also hypothesized that CYP1B1 and MYOC might act through a common biochemical pathway with CYP1B1 acting as a modifier for MYOC. [73] Later the association of CYP1B1 was reported across different populations ranging from 2.2%-23.3% with JOAG, POAG and PACG [ Table 4].

Structure and expression of CYP1B1
CYP1B1 encodes a 543 amino acid dioxin inducible member of the cytochrome p450 gene superfamily and consists of three exons that span 8.5 kb of genomic DNA. [73] CYP1B1 is known to express both in ocular and non-ocular tissues. Animal models have shown that CYP1B1 deficiency leads to abnormality in the ocular drainage structures and TM that are similar to those observed in human PCG.

Genome-Wide Association Studies
Glaucoma being a complex disease, linkage studies were not very successful in identifying the genes responsible for raised IOP and RGC death. This was further compounded by the late onset of symptoms and the unavailability of large affected families. Genome-wide association studies (GWAS) represent a powerful approach for gene mapping in large cohorts using high-density markers like single nucleotide polymorphisms (SNPs) on microarray platforms. These have revealed several loci associated with several complex diseases like diabetes, rheumatoid arthritis and age-related macular degeneration (AMD). [86,87] Based on this approach, Thorleifsson et al., (2007) identified three SNPs (rs1048661, rs3825942 and rs2165241) in lysyl oxidaselike protein 1 (LOXL1) gene on Chromosome 15q22 that were significantly associated with XFS/XFG. [88] Due to the association of POAG and XFS, further analysis of these SNPs in POAG in other populations revealed no significant association in POAG, PACG and PG indicating their exclusive involvement with XFS/XFG only. [89] A recent study by Nakano et al., (2009) demonstrated significant association of six SNPs located on three different chromosomal loci, viz. 1 (ZP4), 10 (PLXDC2) and 12 (DKFZp762A217) in a Japanese population with both high and low-pressure glaucomas. All these SNPs exhibited significant association with combined P values ranging from 1.0X10 -5 to 9.0X 10-5 along with an odds ratio (OR) of 1.33 -1.49. [90] Replication of these SNPs in an Indian cohort with POAG and PACG indicated that these were not associated with IOP-related glaucomas. [91] The failure of replication may be due to inclusion of cases with high IOP unlike mixture of both high and lowpressure glaucomas in the Japanese cohort.
By using linkage and SNP mapping, Jiao et al., (2009) have identified a locus on Chromosome 2 associated with POAG. [92] The region was characterized on Chromosome 2p by performing linkage analyses in 146 multiplex families from Barbados Family Study of Glaucoma (BFSG). Case-control analysis on independent groups from BFSG participants identified a strong association with rs12994401 and POAG. This region overlapped with previous linkage studies in Chinese and African families, indicating that this locus could be a significant cause of glaucoma in the Chinese and Europeans. [20,21]

Candidate Gene-Based Association Studies
Association studies have suggested many genes in single studies while a few of them have been investigated in multiple studies with conflicting results. The variations in association could be due to racial differences, sample size, poorly characterized controls, and clinical heterogeneity between different populations. The POAG-associated genes include ANP, APOE, OPA1, P53, GST, Interleukins, and TNFα. The role of these genes in the etiology of POAG is still controversial [ Table 5].

Interaction of Genes
Glaucoma is a complex disorder in which a single gene may not contribute to disease progression. It has been demonstrated that Apolipoprotein E promoter SNPs previously associated with Alzheimer's disease may also modify POAG phenotype. APOE (-219G) is associated with increased optic nerve damage. The interaction between APOE (−491T) and an SNP in the MYOC promoter, MYOC (−1000G) is associated with increased IOP and with limited effectiveness of IOP-lowering treatments indicating that APOE is modifier gene for the MYOC. [135] The interaction was also observed between TNF-α -863A/C and OPTN 603A/T (or met98Lys) and the carriers of TNF-α /-863A with OPTN /603A (or Lys98) had significantly worse (P =  [73] France POAG 11 (4.6)) A443G (1.3) Melki et al. [74] Germany POAG JOAG NTG 3 (2.2) 3(8.5) 1(4.1)

Gene Expression Studies
The TM tissue plays a key role in the regulation of IOP, and altered TM morphology and functions have been observed in POAG. The mechanisms of action in the TM as well as RGC and astrocytes are dependent on the expression of several genes. Gene expression analysis provides the identification of mechanisms which could be involved in the pathophysiology of POAG. Over the past few years several groups across the world have performed expression studies on TM, RGC and astrocytes that revealed several genes in these tissues. These findings support the existence of numerous regulatory mechanisms in the TM as well as in the optic nerve head (ONH). These genes are involved in the expression of ECM and its remodeling cellular metabolism, cell transport and cell defense. [2,143]

Factors Involved in The Elevation of IOP
Elevated IOP is the most important risk factor in POAG and all current treatments for glaucoma involve the lowering of IOP. Despite its importance in clinical practice, knowledge of IOP regulation in the human eye is limited. Cells which are present in the outflow pathway play an important role in IOP regulation. Recent reports have shown that there are several factors which are involved in maintaining the TM physiology thereby regulating IOP. Some of the important factors involved in raised IOP are age, abnormal function of genes, alterations of ECM and oxidative stress. Age plays an important role in the regulation of IOP because decreased TM cellularity with age could alter the synthetic and catabolic control of the extracellular environment. As the TM cellularity decreases with age their phagocytic activity is lost and leads to accumulation of several toxic molecules within the drainage channels thereby obstructing outflow. [144] Studies have also shown that genetic factors play an important role in the regulation of IOP. Several in vitro and in vivo studies have shown that mutations in the MYOC are a known cause of the abnormal function of TM causing elevation of IOP. [44] Animal models have shown that BAX, SPARC, bestrophin-2 (Best2) and aquaporin deficiency limits elevation of IOP and these results indicate that multiple pathways regulate IOP. [145][146][147][148] Several other molecules like endothelial leucocyte adhesion molecule-1 (ELAM), endothelins (ET), nitric oxide (NO) and modification of ECM molecules have been altered under oxidative stress and these are involved in raising of IOP. [149] ECM components of the TM play an important role in the regulation of IOP and several ECM molecules have been upregulated in TM of POAG patients. Along with oxidative stress, transforming growth factor beta (TGFβ) plays an important role in modulation of ECM molecules in TM. TGFβ2 is the predominant isoform in the eye and several groups have reported higher levels in the aqueous of POAG patients. Recently, adenoviral gene transfer of active human TGFβ2 has been shown to elevate IOP and reduces outflow facility in rodent eyes, indicating that increased levels of TGFβ2 play a major role in the elevation of IOP in POAG. [150]

Mechanism of RGC Death
Progressive loss of optic nerve axons and RGCs result in characteristic optic nerve atrophy and visual field defects in glaucoma patients. A number of hypotheses have been proposed to trigger ganglion cell injury and death in glaucoma. These include compromise to blood flow at the optic nerve, mechanical compression due to raised IOP, loss of neurotrophic factors, autoimmune mechanisms, nitric oxide-induced injuries to the optic nerve and glutamate excitotoxicity. A combination of these factors may be involved in causing glaucomatous RGC loss. Increased IOP is a major risk factor and is involved in the formation of mechanical stress on the ONH thereby inducing glaucomatous cell death. The ONH consists of axons which are projected from RGC, which exit the eye through the lamina cribrosa (LC), a collagenous structure with sieve-like openings. Elevated IOP causes mechanical stress on the LC which leads to decreased axonal transport which causes deprivation of neurotrophic factors. Animal models have shown that acute IOP elevation causes blockage of brain-derived neurotrophic factor (BDNF) transport and may contribute to neuronal death. Supplementation of neurotrophins transiently protect the retina from pressure-induced ischemic injury indicating that neurotrophin deprivation is involved in RGC death. [83] Tissue hypoxia in the ONH and/or retina is thought to develop secondary to or independent from the elevated IOP in glaucomatous eyes and has been proposed to be associated with pathogenic mechanisms underlying optic nerve degeneration in glaucoma. Considerable evidence suggests that tissue hypoxia in the retina may adversely affect the survival of retinal ganglion cells by inducing apoptosis. [151] Brief preconditioning hypoxia induces HIF-1α expression in the retina, which accompanies the expression of adaptive proteins and provides resistance to cell death; however, exposure to hypoxia for a longer period initiates the cell death program. [151] The eye is an organ that is predisposed to great levels of oxidative stress. [152] Oxidative stress evident in glaucomatous tissues is an important factor for the loss of neurons during glaucomatous neurodegeneration. RGC death induced by glaucomatous stimuli involves receptor-mediated caspase cascade, and mitochondria-mediated caspase-dependent and caspase-independent components of cell death cascade. TNF-α and hypoxia are two different stimuli known to preferentially trigger the receptor-mediated or mitochondria-mediated cell death pathways, respectively. Following oxidative modifications of retinal proteins in glaucomatous eyes, reduced ability of cells to cope with the glaucomatous tissue stress may result in impaired cellular homeostasis eventually contributing to neurodegeneration. [153][154][155][156] Nitric oxide (NO) is another important mediator of glial cell-mediated apoptosis in neuronal cells. Under normal physiological conditions NO plays an important role in performing several physiological functions, including regulation of vascular tone, neurotransmitter release and synaptic plasticity. NO can be synthesized by three enzymes (NOS-1, NOS-2, and NOS-3) and elevated levels of NO have been observed in the AqH and genetic association of iNOS polymorphisms have been reported in glaucoma patients. [157] Neurotrophins, particularly BDNF, are known to influence RGC survival in vitro, both during retinal development and

S38
Indian Journal of Ophthalmology Vol. 59 Suppl. 1 after lesioning. [158] Loss of physiological neurotrophin levels, particularly BDNF is consistent with known events in the clinical and pathological aspects of glaucoma. Overexpression of BDNF delays RGC death in experimental glaucoma.

Immune Response
Increasing evidence from both in vivo and in vitro studies over the past few years strongly supports the presence of the immune system in glaucoma pathogenesis. Recent studies have shown increased expression of several autoantibodies against many optic nerve and retinal proteins. These antibodies include hsp60, hsp27, alpha crystallins, vimentin and HSP70. The direct application of Abs against these proteins at similar concentration found to induce RGC death in in vivo and in vitro conditions. Recent reports have also shown that complement cascades have been implicated in glaucomatous neurodegeneration. IOP also modulates the immune system by inducing several complement components like C3, C1q and C3r. Interestingly, it has been shown that complement factor H, a common regulator of the complement system is down regulated, while several other complements components are up-regulated, indicating an abnormal activation of complement system. [159,160] The Future Glaucoma is a complex disease attributed to multiple genes with varying magnitudes of effect. Newer methods of gene mapping involving GWAS have revealed some interesting results over the last few years. However, many of these studies could not be replicated due to differences in phenotyping, other screening modalities and improper study designs. Several genes implicated through functional studies could not be associated in the genetic studies. Since the effect sizes of these genes would vary, there would be a need for precise characterization at the molecular level using multiple approaches. Whole genome sequencing would be an ideal choice for understanding the unknown genes involved in glaucoma pathogenesis. This would need to be supplemented with functional studies and further validation through proteomic approaches, animal models and replications in different ethnic populations. A collective effort including a multi-disciplinary approach involving the expertise of other branches of science is a must for unraveling the mystery and the underlying molecular mechanisms in glaucoma.