Competing/conflict of interest: No stated conflict of interest.
Screening glaucoma genes in adult glaucoma suggests a multiallelic contribution of CYP1B1 to open-angle glaucoma phenotypes
Version of Record online: 6 DEC 2011
© 2011 The Authors. Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
Clinical & Experimental Ophthalmology
Volume 40, Issue 4, pages e208–e217, May/June 2012
How to Cite
Patel, H. Y., Richards, A. J., De Karolyi, B., Best, S. J., Danesh-Meyer, H. V. and Vincent, A. L. (2012), Screening glaucoma genes in adult glaucoma suggests a multiallelic contribution of CYP1B1 to open-angle glaucoma phenotypes. Clinical & Experimental Ophthalmology, 40: e208–e217. doi: 10.1111/j.1442-9071.2011.02714.x
Funding sources: University of Auckland Staff Research Fund and Glaucoma New Zealand.
- Issue online: 14 JUN 2012
- Version of Record online: 6 DEC 2011
- Accepted manuscript online: 17 OCT 2011 08:46AM EST
- Received 15 April 2011; accepted 29 August 2011.
- genetic analysis;
- open-angle glaucoma
Background: Despite increasing knowledge of the genetic pathophysiology of glaucoma, mutations in known genes account for less than 15% of disease. Gene screening predominantly remains a research tool rather than an essential part of the clinical work-up. We aimed to determine the mutational spectrum and frequency in the genes implicated in glaucoma, in a range of glaucoma and ‘glaucoma suspect’ (GS) participants, with a positive family history.
Methods: Observational large case series. One hundred fifteen patients recruited from public hospital and private clinics had diagnoses of GS, ocular hypertension, pseudoexfoliative glaucoma (PXG) or primary open-angle glaucoma (POAG), and at least one affected family member. In a university laboratory, DNA samples were screened for mutations in all coding exons of MYOC and CYP1B1, and OPTN (exons 4, 5 and 16). WDR36 (exons 1, 4, 5, 8, 11, 13 and 17) was screened in those with CYP1B1 changes. LOXL1 risk variants were screened in PXG pedigrees. Cascade screening of family members was undertaken.
Results: Seven out of one hundred fifteen (6.1%) individuals had at least one pathogenic or hypomorphic CYP1B1 allele associated with GS, POAG (5) and PXG phenotypes, including two novel sequence variations (p.Ser6Gly, p.Val243Leu). No pathogenic MYOC change was detected. Five individuals (4.3%) carried an OPTN sequence variation. Three of the seven with CYP1B1 changes had polygenic changes.
Conclusions: Mutational analysis of known glaucoma genes in a mixed glaucoma population replicates the reported frequency of pathogenic CYP1B1 changes. Heterozygous CYP1B1 changes occurred at a greater frequency than other genes. Glaucoma pathogenesis in the clinic setting is genetically heterogeneous and may be polygenic.
Glaucoma describes a group of complex, genetically heterogeneous diseases, which all result in progressive loss of retinal nerve fibres causing irreversible optic disc excavation and visual field loss. It is one of the leading causes of blindness in the world, due to late diagnosis or resistance to current treatments.1 Primary open-angle glaucoma (POAG) is the most common form of glaucoma, accounting for approximately half of all cases.1 To date, three genes associated with POAG have been identified, Myocilin (MYOC), Optineurin (OPTN) and the WD repeat domain 36 gene (WDR36).2–5 In addition, a potential modifier gene, CYP1B1, has been identified.6,7 This gene codes for an autosomal recessive form of primary congenital glaucoma (PCG) and is a member of the cytochrome p450 gene superfamily.8CYP1B1 mutations have also been reported in families where congenital glaucoma and POAG coexist, and in sporadic cases of POAG in different populations.7–13
Despite an increasing knowledge of the genetic pathophysiology of glaucoma, mutations in known genes account for less than 12–15% of all disease. The majority of patients remain without molecular characterization, and gene screening predominantly remains a research tool rather than an essential part of the clinical work-up. Many previous studies have examined genetic variation in well-defined populations of specific glaucoma phenotypes – that is, juvenile open-angle glaucoma (JOAG), or POAG, or PCG. We aimed to determine the mutational spectrum and frequency in the genes implicated in glaucoma, in a range of participants, including glaucoma suspects (GS), from glaucoma subspecialty clinics – the only requirement being a positive family history. In particular, we aimed to investigate the role of CYP1B1 in non-congenital glaucoma phenotypes.
The project received approval from the Northern X Regional Ethics Committee (AKX/04/08/224), Ministry of Health, New Zealand. The study conformed to the principles of the Declaration of Helsinki, with written informed consent obtained from 115 unrelated probands recruited from glaucoma clinics (hospital and private). The inclusion criteria were the following: a clinical diagnosis of GS, ocular hypertension (OHT), POAG or pseudoexfoliative glaucoma (PXG); and a history of at least one family member also affected by POAG. For this study, a GS is defined as an individual who has one or two abnormal parameters of glaucomatous optic neuropathy, but evidence of visual field abnormality or progression has not yet been elucidated (i.e. preperimetric glaucoma), in addition to having at least one affected family member. The exclusion criteria were secondary glaucomas (uveitic, trauma) and patients unable to provide written informed consent. Where possible, affected family members were recruited for cascade screening.
Each proband underwent phenotyping with clinical examination, including history, visual acuity, intraocular pressure (IOP), slit-lamp examination, gonioscopy, automated Humphrey visual fields and optic nerve head characterization clinically, and with either optic disc photography, Heidelberg retinal tomography (Heidelberg Engineering, Dossenheim, Germany) or optical coherence tomography (Stratus OCT, Carl Zeiss Meditec AG, Jena, Deutschland). Clinical notes were reviewed to obtain each proband's presenting data (age at diagnosis, presenting IOP, cup-to-disc ratio and visual field parameters).
Following informed consent, biological samples (peripheral venous blood or saliva specimen) were collected for DNA extraction using the salt extraction method from blood,14 and according to the manufacturer's instructions for saliva kits (Oragene, DNA Genotek, Ottawa, ON, Canada). For controls, DNA samples were collected from randomly selected and ethnically matched individuals attending the Ophthalmology Department who do not exhibit any clinical evidence of glaucoma on routine examination, including normal IOP, open angle and optic nerve appearance.
Mutational analysis of genes
DNA samples were screened for mutations in all coding exons of CYP1B1 and MYOC, as well as OPTN (exons 4, 5 and 16). Those with CYP1B1 mutations also underwent mutational analysis of WDR36 (exons 1, 4, 5, 8, 11, 13 and 17). The pedigree with PXG underwent mutational analysis of LOXL1 single nucleotide polymorphisms (SNPs) rs1048661 and rs3825942. PCR amplification of all the coding exons of CYP1B1 (exons 2–3) was undertaken using previously described primers.7 Further details of primers and PCR conditions for OPTN, MYOC, WRD36, LOXL1 and CYP1B1 are available on request. Following column purification with High Pure PCR Purification Kit (Roche Diagnostic, Mannheim, Germany), the product was sequenced directly according to protocols accompanying the ABI BigDye Terminator kit v3.1 (Applied Biosystems Inc, Foster City, CA, USA). Bidirectional sequencing of amplicons was undertaken on an ABI 3700 prism genetic analyzer (Applied Biosystems Inc, Foster City, CA, USA). Nucleotide sequences were compared with the published CYP1B1 mRNA sequence (GenBank Accession No: NM_000104.3) and polymorphic variation data in electronic databases to determine pathogenicity. For any sequence change, relatives of the proband were recruited to determine segregation.
To determine the frequency of the CYP1B1 pathogenic changes p.Trp57X, p.Tyr81Asn and p.Glu229Lys, and the novel variants c.16A>G, and c.727G>T, in the general population, 100 control (glaucoma excluded on clinical examination) individuals (200 alleles) were screened. Control population demographics are listed in Table 1. Screening for the detected CYP1B1 sequence variants used high resolution melting analysis (HRMA) on the Rotor-Gene 6000 (Corbett Life Sciences, San Francisco, CA, USA), using high resolution melting master kit (Roche Diagnostic, Mannheim, Germany). Primers are available on request. Each reaction included a positive and negative control based on sequencing confirmation. Any sample on the melt curve that produced an equivocal reading was subject to further PCR and sequencing to confirm or exclude the presence of the sequence variation.
|Demographic||No. of study subjects (n = 115)||No. of controls (n = 100)|
|Male||48 (41.7%)||42 (42.0%)|
|Female||67 (58.3%)||58 (58.0%)|
|Caucasian||107 (93.0%)||75 (75.0%)|
|Indian||5 (4.4%)||2 (2.0%)|
|Asian (Chinese)||3 (2.6%)||12 (12.0%)|
|Mean ± SD||61 ± 13||40 ± 22|
For the novel sequence variants, homology and predicted destruction or creation of exonic splicing enhancers, or effects on splicing, was evaluated using a variety of publicly available software. PolyPhen analysis was used to predict the impact of p.Ser6Gly and p.Val243Leu missense mutations on protein structure and function. Protein modelling is discussed later. Potential pathogenicity was based on positive family segregation, an allele frequency of <1/100 chromosomes, homology and bioinformatic prediction of biological significance.
A total of 115 unrelated probands were screened (demographic details in Table 1). The mean age was 61 ± 13 years, and 58% were female. The majority of probands were of Caucasian ethnicity (93.1%) followed by Indian (4.3%). The most common diagnosis was POAG (49.6%), followed by GS (43.5%), OHT (5.2%) and PXG (1.7%).
The results of mutational analysis of the glaucoma genes are presented in Table 2. Eleven pathogenic or novel sequence variations in 13 probands (11.3%) were identified in the MYOC, CYP1B1 or OPTN gene analysis.
|Gene||Sequence variation||No. in study cohort (%) (n = 115)||Diagnosis||Previously published||Allele frequency CYP1B1|
|Study cohort (%) n = 230||Control cohort (%) (n = 200)|
|CYP1B1||Trp57X||1 (0.9%)||PXG||Yes||1 (0.45%)||0|
|Tyr81Asn||1 (0.9%)||POAG||Yes||1 (0.45%)||3 (1.5%)|
|Glu229Lys||4 (3.5%)||POAG (n = 4)||Yes||4 (1.75%)||1 (0.5%)|
|Ser6Gly||1 (0.9%)||POAG||No||1 (0.45%)||0|
|Val243Leu||1 (0.9%)||GS||No||1 (0.45%)||0|
|MYOC||Lys398Arg||2 (1.7%)||POAG (n = 1), GS (n = 1)||Yes||Total||Total|
|8/230 = 3.4%||4/200 = 2%|
|OPTN||Met98Lys||5 (4.3%)||POAG (n = 4), GS (n = 1)||Yes||–||–|
|His212Pro||2 (1.8%)||GS (n = 2)||Yes||–||–|
Mutational analysis of CYP1B1
Screening of CYP1B1 revealed seven probands (6.1%) with at least one pathogenic or novel CYP1B1 allele. The sequence variations and respective genotype–phenotype correlation for each case subject are summarized in Table 3. Six probands had sequence variations present in the heterozygous state, and one was a compound heterozygote (case subject 1: p.Tyr81Asn and p.Glu229Lys). Six of the sequence variations have previously been published, whereas two were novel: p.Ser6Gly and p.Val243Leu. These two novel sequence variations were not present in 200 ethnically matched, unaffected alleles.
|Case||CYP1B1 sequence change||Other sequence variations||Phenotype||Sex||Ethnicity||Features at diagnosis||Treatment required|
|Age||IOP (OD/OS)||CDR (OD/OS)||VF|
|MD (OD/OS)||PSD (OD/OS)|
|1||c.241T>A, p.Tyr 81Asn c.685G>A, p.Glu229Lys||–||POAG||M||Indian||67||NA||NA†||NA||Medical + Surgical (Trab OU)|
|2||c.171G>A, p.Trp57X||–||PXG||M||Caucasian||69||38/22||0.8/0.6||−1.6/−1.4||1.9/1.4||Medical only|
|3||c.685G>A, p.Glu229Lys||–||POAG||F||Caucasian||55||28/28||0.3/0.3||−1.4/+0.2||1.7/1.4||Medical + Surgical (Trab OD)|
|4||c.16A>G, p.Ser6Gly||OPTN Met98Lys||POAG||F||Caucasian||71||23/21||0.6/0.6||−2.2/−1.4||2.1/1.8||Medical + Surgical (Trab OD)|
|5||c.685G>A, p.Glu229Lys||OPTN Met98Lys||POAG||M||Indian||53||32/36||0.8/0.9||NA||Medical + Surgical (Trab OU)|
|WDR36 Asp 33Glu|
|6||c.685G>A, p.Glu229Lys||LOXL1||POAG||F||Caucasian||58||26/24||0.8/0.7||−9.7/+0.3||13.7/1.3||Medical and SLT|
|7||c.727G>T, p.Val243Leu||WDR36||GS†||F||Caucasian||5||23/21||0.7/0.4||−0.6/+0.2||1.7/1.7||Nil at present|
Homology modelling showed some conservation of the serine in position 6, but marked conservation of the 243 valine (Fig. 1). PolyPhen prediction of both these missense variants suggests they are likely to be benign, p.Ser6Gly PSIC score difference of 0.179, and p.Val243Leu PSIC score difference of 0.976.
The pedigrees of each proband with a CYP1B1 sequence variation are presented in Figure 2. Cascade screening of affected family members was possible in four cases (2, 3, 4 and 7) – the results included in their respective pedigrees (Fig. 2). Segregation of the CYP1B1 sequence variation was seen in the affected family members of Cases 2 and 3. Case 2 was diagnosed with PXG at age 73 years. His affected sister, diagnosed with POAG at age 69 years, also carried the p.Trp57X change. Both had the LOXL1 rs1048661 G/G and rs3825942 G/G genotype. The two sisters of Case 3 have a diagnosis of POAG and were both heterozygous for the p.Glu229Lys change. The OPTN sequence variation (Met98Lys) was identified to segregate with the disease phenotype in the family of Case 4.
The frequency of the pathogenic CYP1B1 variants p.Trp57X, p.Tyr81Asn and p.Glu229Lys in 200 control alleles determined by HRMA detected one allele (1/200) carrying the p.Glu229Lys change, an Indian individual, unaffected by glaucoma. The nucleotide change c.241T>A, p.Tyr81Asn was present as a heterozygous change in three clinically unaffected Caucasian individuals, and c.171G>A, p.Trp57X was not observed in any controls.
The frequency of the CYP1B1 SNP rs1056836 (c.1294C>G, p.Leu432Val) was analysed for a subset (n = 108) of the study group, and was identified in 61% (n = 68) of probands. Of these, 22 (20%) were homozygous (G/G) and 46 (43%) were heterozygous(C/G) for the sequence variation. Using reference genotype Caucasian data (HapMap), there was no statistically significant difference in the p.Leu432Val polymorphism frequency (P = 0.80, chi-square: 0.427, degrees of freedom: 2) between the glaucoma patients in this study and ethnically matched normal controls (Table 4).
|Val 432Leu SNP-Leucine variant||New Zealand population (%)||Controls† (%)|
By looking at the POAG cohort alone (n = 57) then, 5/57 (n = 8.77%) individuals carried six CYP1B1 probable pathogenic changes, or 6/114 alleles (5.2%).
No pathogenic MYOC sequence variations were observed. Five probands (4.3%) had a potentially significant sequence variation in OPTN– all had the previously published Met98Lys change. Of those with CYP1B1 mutations, two had possibly significant sequence variations in WDR36. Genotype/phenotype is listed in Table 2. Three probands with CYP1B1 pathogenic changes also had possible pathogenic changes in other glaucoma genes – subjects 4, 5 and 7. The phenotypic details are presented in Table 3.
This is the first study to look at the frequency of the known glaucoma genes in ‘all-comers’– a range of open-angle glaucomas and those that are GS – with a first degree relative affected with POAG. This study aimed to answer the question ‘What is the utility of screening specific glaucoma genes in the general glaucoma clinic setting?’– that is, are the type and frequency of changes in any one gene of sufficient significance render it an integral part of a genetic work-up? For example, if a microarray technology was applied, which genes and which gene changes should be included on that microarray?
Although the MYOC gene has been the most commonly implicated in POAG, reported in 2.6–4.4% of POAG populations,2 no pathogenic sequence variation was detected in this cohort. However, the major limitation in detection in this study is the sample size of our cohort.
The mutational analysis of other known or associated glaucoma genes in this mixed glaucoma population replicated the frequency of pathogenic or hypomorphic CYP1B1 changes (6.1%) demonstrated in previous reports (4.5–10.9%).7,11,13,15 Previous reports involved juvenile and/or adult-onset POAG only, in contrast to the mixed glaucoma population of this study. Within our POAG population alone, the number of individuals with a change was 8.77%. Heterozygous changes in CYP1B1 occurred at a greater frequency than any other of the implicated glaucoma genes, suggesting some contribution to the pathogenesis of a range of non-congenital glaucoma phenotypes including POAG. However, the CYP1B1 pathogenic allele frequency within our entire cohort, and within the subset of the POAG cohort, was not statistically significant when compared with the allele frequency in the control population (P = 0.178 and P = 0.287, respectively, Fisher's exact test, paired).
Despite initial reports, the role of the OPTN and WDR36 genes in glaucoma has not been fully validated. Only one sequence variation in OPTN was identified in this study (Met98Lys, n = 5, 4.7%). Initially considered a ‘risk associated’ change, subsequent studies have not confirmed this association.4,16,17WDR36 has also been reported to cause POAG, but these results are not well validated.3,18–21 As more recent evidence suggests that WDR36 may be a modifier gene for POAG rather than directly disease-causing,22 we only screened in those with a CYP1B1 variant. Our rationale for only sequencing selected exons was based on the significant cost of screening the whole coding sequences of WDR36 and OPTN, and published data suggesting a minimal contribution of these genes to glaucoma. Therefore, we concentrated on the alleles that were more suggestive of high risk and had been validated in more than one study. Two previously described changes in WDR36, p.Asp33Glu and p.His212Pro, were detected in two patients with CYP1B1 variants (Table 3).
Five different CYP1B1 sequence variations (excluding known polymorphisms) in seven (6.1%) of the 115 probands were identified. Three pathogenic changes have previously been published (p.Glu229Lys, p.Tyr81Asn, p.Try57X), and two are novel (p.Ser6Gly, p.Val243Leu). Six of the probands had sequence variations present in the heterozygous state, and one was a compound heterozygote.
The p.Glu229Lys change was the most common, present in 3.5% (n = 4) of probands, all who had POAG with the age of onset ranging from 53 to 67 years. Three required trabeculectomy and the fourth laser trabeculoplasty in addition to medical treatment possibly reflecting more severe disease. Cascade screening was possible for one proband, with both affected siblings heterozygous for p.Glu229Lys. (Fig. 2, Pedigree 3). Whether the p.Glu229Lys sequence variation is specifically associated with POAG is yet to be fully established. Some reports support the association between the p.Glu229Lys sequence variation in the heterozygous state and POAG,11,13 whereas others report a similar frequency among POAG patients and controls.23,24 There was significant phenotypic variability among the POAG patients in these reports with age of onset ranging from 35 to 74 years.
One individual was a compound heterozygote with p.Tyr81Asn, in addition to the p.Glu229Lys. This individual was diagnosed and treated overseas prior to being seen in New Zealand, but clinical examination excluded a diagnosis of PCG (i.e. no buphthalmos, Haab striae nor angle abnormality).
The frequency of the p.Glu229Lys allele in our control cohort was 0.5% (1/200). The p.Glu229Lys mutation was first reported in an Indian PCG pedigree, in a compound heterozygote,25 although the same change occurred in 12.8% of their 70 Indian controls in the heterozygote state. A subsequent paper from this same research group from India, Reddy et al.26 reported the frequency of this change in an Indian control population as 6.4% of 140 control chromosomes, which suggests this is the same reference population (12.8% of 70 controls, and 6.4% of 140 control chromosomes); however, in another eastern Indian population, this change was not observed in 100 controls15 nor in 200 controls from Hyderabad, India.27 It was not seen in 47 controls from France,13 and no control population is reported in the Colomb paper.10 It has been reported as occurring in both the heterozygous and compound heterozygous state in association with PCG. In terms of effect on protein and function, this change occurs in the carboxyl terminal of the F helix in the vicinity of the substrate binding region in the CYP1B1 protein.28 Substitution of glutamic acid to lysine affects the local charge distribution, which disturbs an important cluster of salt bridges, and has the potential to destabilize the other ionic interactions within the protein.29 Similarly, the enzymatic activity of the CYP1B1 protein with the p.Glu229Lys mutation shows 20–40% enzymatic activity compared with the wild type protein, rendering it ‘hypomorphic’.30,31 This mutant protein also showed reduced stability, compared with wild type, in transfected cells.30,31
The p.Tyr81Asn sequence variation was observed in a compound heterozygous state in association with p.Glu229Lys in Case 5. Cascade screening was not possible for this proband. A significant association between the heterozygous p.Tyr81Asn change and POAG has previously been reported, with considerable phenotypic variability observed.11,13,24 This change was also seen in 1.5% (3/200) of our unaffected control alleles, yet in other series, 0/200 Indian controls27 and 0/47 French controls.13 This p.Tyr81Asn change is also ‘hypomorphic’ with a reported significantly reduced enzymatic activity and reduced protein stability.30,31
The p.Trp57X sequence variation was identified in a Caucasian male with PXG diagnosed at 69 years of age, and was also present in the proband's affected sister (Fig. 2, Pedigree 2). Interestingly, rather than PXG, she had a diagnosis of POAG, which was diagnosed at aged 69 years and was well controlled on medical treatment. This sequence variation has previously been reported in a patient with Peters' anomaly and secondary congenital glaucoma but not in POAG or PXG.32 This change was not observed in the control population. LOXL1 genotyping for the rs1048661 (c.422G>T, p.Arg141Leu) and rs3825942 (c.458G>A, p.Gly153Asp) variants was undertaken for proband 2 and his sister.
Both had the rs1048861 G/G genotype and the rs3825942 G/G phenotype.
The first study demonstrating the association with LOXL1 SNPs and XFS/PXG identified one intronic and two non-synonymous coding SNPs (rs1048861 G/G and rs3825942 G/G) that had a significant association with disease in the Icelandic and Swedish populations.33 Other studies have replicated this association, yet in other ethnic groups, the association is actually reversed.34 Hewitt et al.35 also showed a significant association but comment that ‘our Caucasian population has a 9-fold lower lifetime incidence of pseudoexfoliation syndrome compared with Nordic populations despite having similar allelic architecture at the LOXL1 locus. This strongly suggests that as yet unidentified genetic or environmental factors independent of LOXL1 strongly influence the phenotypic expression of the syndrome’.
Tarkkanen also demonstrated in 1962 that the frequency of exfoliation was 7% among relatives of POAG over 60 years of age.36
Therefore, the presence or absence of LOXL1 high-risk alleles is not a guarantee of development of PXG, with other factors contributing to the manifestation of the disease. These factors maybe genetic and potentially include the presence of a modifying allele of CYP1B1, just as CYP1B1 has previously been demonstrated to modify the MYOC-associated glaucoma phenotype.7
The novel CYP1B1 p.Ser6Gly was present in the Pedigree 4 proband, who also had the OPTN Met98Lys sequence variation. The proband's affected sister (POAG, diagnosis age 77 years) did not carry the CYP1B1 sequence variation. PolyPhen prediction classifies this change as a benign variant, although it was not observed in 200 control alleles.
The other novel CYP1B1 sequence variation, p.Val243Leu, in association with a WDR36 His212Pro was seen in a GS who also had Meesman's corneal dystrophy and a strong family history of POAG (Case 7, Fig. 2, Pedigree 7) The proband also had two affected relatives available for screening – both were negative for the CYP1B1 Val243Leu and WDR36 His212Pro changes. Although the CYP1B1 valine residue is highly conserved and not observed in 200 control alleles, the PolyPhen prediction is that this is a benign variant. This does not necessarily exclude it as having functional activity, however.
The exact role CYP1B1 plays in the development of POAG remains to be investigated. It is well documented as a cause of autosomal recessive PCG, although in a number of PCG-affected consanguineous pedigrees and individuals, only one heterozygous change is documented, along with non-penetrance.6,9,10,12,28,30,37 It has previously been reported that CYP1B1 may be a modifier gene for the expression of MYOC in patients with JOAG.7 Inheritance of glaucoma may be multigenic in some cases, and the CYP1B1 sequence variations may only be one component contributing to the pathogenicity of POAG. It has been postulated that in the heterozygous state, the enzymatic activity of CYP1B1 is reduced to below a particular threshold. This consequently increases susceptibility to the development of POAG and/or the severity of the glaucoma phenotype expressed via its influence on the activity of other glaucoma-associated genes.7,11,24 In contrast, other reports have suggested there may, in fact, be a direct causative role of CYP1B1 in POAG, with a possible monogenic association seen in some patients.7,11,13,15 Using tag SNPs, Burdon et al. also demonstrated an association of the CYP1B1 gene with POAG.38 In the current study, three probands with CYP1B1 sequence variations had additional changes present in the known or suspected glaucoma genes screened.
Given the primary association of CYP1B1 as a causative gene for autosomal recessive PCG, it is expected that carriers will be present within the community. New Zealand is a relatively non-consanguineous society, although recent immigration may have slightly changed the frequency. Given that four carriers were detected in 100 control individuals (4 of 200 alleles), theoretically means the frequency of PCG in NZ would be 1 in 2500, which certainly is not the case. Although a rate for New Zealand is not known, it would be comparable to other western countries. A recent population-based study in Minnesota determined the birth prevalence of PCG during the 40-year period was one per 68 254 residents younger than 20 years, or 1.46 per 100 000.39
The functional implication of CYP1B1 polymorphisms also warrants further investigation. It has previously been demonstrated that the Val432Leu variant alters CYP1B1 metabolic activity such that there is depletion of the protective 17-β-estradiol and increased production of the harmful 4-hydroxyestradiol.40–42 It has been postulated that such polymorphisms may alter CYP1B1 function sufficiently for it to modify the effect of other genes involved in glaucoma.7 It has also been postulated that in an individual that is heterozygous for a CYP1B1 sequence variation, the polymorphism may additionally alter the metabolic activity of CYP1B1, thereby further predisposing to the development of glaucoma.7 In a recent study involving an Indian population, a significant association between POAG and the leucine variant of Val432Leu was reported for the first time, and supported by in vitro functional analysis.43 Although the leucine variant of this SNP was present in more than half of the predominantly Caucasian glaucoma cohort in this study, a statistically significant relationship between the genotype and POAG was not observed. Thus, the possible association between the leucine variant and increased risk of POAG was not supported by the results of this study.
This study is not, however, without limitations. There is the possibility of ascertainment bias as family history data were based on patient reporting only. The medical records of genetically screened relatives of the probands were not checked to confirm the reported glaucoma diagnosis, and unaffected family members were not clinically or genetically screened to confirm a clinical absence of disease, nor segregation of genotypes. This was largely not possible because of the wide geographic distribution of family members. Another possible limitation is the heterogeneous nature of the glaucoma population resulting in a smaller sample size of POAG participants for analysis. However, there are few previous reports in the literature that have included GS, OHT and PXG. The results of this study suggest that the contribution of CYP1B1 and other glaucoma genes should not be limited to its involvement in PCG, POAG or JOAG phenotypes. Although this study specifically involves a New Zealand population, which has its own unique ethnic make-up, further insights into the underlying genetic basis of glaucoma can still be provided, demonstrated by the identification of two previously unreported changes in CYP1B1 and the polyallelic nature of changes in this group of glaucoma genes.
In conclusion, the molecular screening of glaucoma genes in an adult glaucoma clinic setting suggests the pathogenesis is genetically heterogeneous and may be polygenic. CYP1B1 also contributes to the pathogenesis of a range of non-congenital glaucoma phenotypes including POAG, perhaps in a modifier or a multiallelic fashion.
The increased incidence of pathogenic CYP1B1 sequence variations was in concordance with previous reports. Heterozygous changes in CYP1B1 occurred at a greater frequency than other implicated genes highly suggestive of its probable contribution to the pathogenesis of a range of glaucoma phenotypes. Therefore, CYP1B1 should be considered an essential part of the mutational work-up of a glaucoma population. Further research is required to establish the exact role CYP1B1 plays in the development of glaucoma.
The authors would like to thank participating patients and family members, Gladys Ko, Dasha Nelidova, Lucia Rong and James Kan for their contribution during their University of Auckland summer studentships.
- 37Mutation screening of the CYP1B1 gene and phenotype-genotype correlation in primary congenital glaucoma cases from Brazil. Invest Ophthalmol Vis Sci 2001; 41: A2848., , et al.