|
|
||||||||
1From the Kallam Anji Reddy Molecular Genetics Laboratory and the 2VST Centre for Glaucoma Care, L. V. Prasad Eye Institute, Hyderabad, India.
| Abstract |
|---|
|
|
|---|
METHODS. The entire coding region of CYP1B1 was screened by resequencing in 224 unrelated cases of POAG (n = 134) and PACG (n = 90) and 200 ethnically matched normal control subjects from Indian populations. Six intragenic single nucleotide polymorphisms (SNPs) in CYP1B1 (–13T>C, R48G, A119S, V432L, D449D, and N453S) were used to generate haplotype data for the cases and controls and linkage disequilibrium (LD) and haplotype analysis were performed with Haploview software, which uses the EM (expectation-maximization) algorithm.
RESULTS. The frequency of CYP1B1 mutations was higher among POAG (18.6%; 95% CI, 12.9–26.1) than PACG (11.1%; 95% CI, 6.1–19.3) cases. There was a marked allelic heterogeneity, and the Arg368His was the most prevalent mutation across both the phenotypes. The spectrum of CYP1B1 mutations was largely similar across different POAG populations. Haplotypes generated with intragenic SNPs indicated the C-C-G-G-T-A to be a risk haplotype associated with CYP1B1 mutations in POAG (P = 0.006) and PACG (P = 0.043), similar to that observed in cases of primary congenital glaucoma worldwide.
CONCLUSIONS. The results demonstrate an involvement of CYP1B1 in a proportion of POAG and PACG cases that should be explored further. The similar haplotype background of these mutations is indicative of their common origin across multiple glaucoma phenotypes.
Genetic heterogeneity is well documented in POAG, and 11 chromosomal loci (GLC1A-GLC1K) have been mapped by linkage analysis.9 Of these, only three genes harboring GLC1A (Myocilin; MYOC; OMIM 601652),10 GLC1E (Optineurin; OPTN; OMIM 602432),11 and GLC1G (WDR36; OMIM 609669)12 have been characterized. In addition, approximately 15 candidate genes have been identified by association studies that require a thorough replication in different populations.9 Glaucoma being a complex disease would be attributed to multiple gene variants with various magnitudes of effect.13
Although the human cytochrome P450 gene CYP1B1 (OMIM 601771) has been implicated in primary congenital glaucoma (PCG; OMIM 231300) worldwide,14 15 16 it has been relatively less explored in POAG and not at all in PACG. An initial study implicated the involvement of CYP1B1 with MYOC through a digenic mechanism in a family with juvenile-onset open angle glaucoma (JOAG) and suggested that CYP1B1 is a modifier of MYOC expression. It was also observed that affected subjects harboring a mutant CYP1B1 allele in this family had an earlier age at onset than those with only a mutant MYOC allele.17 These findings led to the screening of CYP1B1 as a candidate gene among the patients with POAG and largely among those with JOAG. The frequency of CYP1B1 mutations varied in patients from Canada (5.0%),17 France (4.6%),18 Spain (10.9%),19 Eastern India (4.5%),20 and Southern India (10.8%).21 The differences in mutation frequency could be partly explained by the definition of disease used in these studies. The Canadian patients had JOAG17 whereas the French patients had POAG, but elevated IOP was not an inclusion criterion,18 similar to studies from Eastern20 and Southern21 India. The results from these studies indicate a minor involvement of CYP1B1 among JOAG and late-onset POAG cases and suggest a possible role of this gene in glaucoma pathogenesis.
We have reported the extent of CYP1B1 mutations along with their structural properties in PCG.22 23 We have also demonstrated a global clustering of these mutations on specific haplotype backgrounds, irrespective of geographic location, that could be useful in predictive testing.16 Herein, we report an extensive screening of the CYP1B1 gene in a cohort of patients with POAG or PACG from India, to determine its mutation spectrum and understand the haplotype backgrounds of these mutations.
| Methods |
|---|
|
|
|---|
POAG (Including JOAG)
The diagnosis of POAG was based on open angles on gonioscopy, an IOP >21 mm Hg, and characteristic optic disc changes and corresponding visual field defects in patients >35 years of age. Visual field defects were considered to be glaucomatous if they were consistent with optic disc damage and met at least two of the criteria laid out by Anderson and Patella.24 The presence of a visual field defect required confirmation by a repeatable field performed within 2 weeks of the first reliable visual field result showing the defect. The field defects were further classified as mild, moderate, or severe.25 Such findings in patients between 5 and 35 years of age were labeled as JOAG. As the presence of visual field defects was one of the inclusion criteria, only patients older than 10 years were included in the study.
Primary Angle-Closure Glaucoma
PACG was defined as the presence of optic disc and visual field changes characteristic of glaucoma, along with appositional or synechial primary angle-closure (PAC) in patients older than 18 years. The visual field defects were as defined in POAG. PAC (appositional) was defined as increased IOP (>21 mm Hg) associated with nonvisibility of the filtering trabecular meshwork for more than 180°, in the absence of PAS, disc damage, or field changes. PAC (synechial) was defined as the presence of PAS with nonvisibility of the filtering trabecular meshwork for more than 180°, with or without increased IOP (>21 mm Hg), without disc damage or demonstrable field defects. The presence of even a single PAS in an angle with more than 180° of nonvisibility of trabecular meshwork was considered diagnostic of PAC. Other causes of synechiae were excluded.
Ocular hypertension, normal-tension glaucoma, lens-induced glaucoma, neovascular and pseudoexfoliation glaucoma, and secondary open-angle glaucoma were excluded. Other ocular diseases that can lead to secondary glaucoma were also excluded.
Normal adult individuals without any signs or symptoms of glaucoma and other systemic diseases served as control subjects. Their visual acuity ranged from 20/20 to 20/40, and their IOP was <21 mm Hg. Clinical examination on stereo biomicroscopy did not reveal any changes in the optic disc suggestive of glaucoma. The patients and controls were matched with respect to their ethnicity and geographical region of habitat.
Molecular Analysis
Peripheral blood samples (5–10 mL) were collected from each subject by venipuncture, with prior informed consent. DNA was extracted by standard protocols26 and the entire coding region of CYP1B1 was amplified using appropriate oligonucleotide primers and PCR protocols, as published earlier.27 The amplicons were purified (SigmaSpin columns; Sigma-Aldrich, St. Louis, MO) and bidirectionally sequenced using dye termination chemistry (BigDye on a 3100 DNA Analyzer; Applied Biosystems, Inc. [ABI], Foster City, CA), according to the manufacturers protocol. Sequencing analysis software was used to read the individual sequences. Six mutations (G61E, Y81N, Q144R, P193L, E229K, and R368H) were further confirmed by restriction digestion of the amplicon with appropriate restriction enzymes as published earlier,18 22 27 whereas the remaining five mutations were verified by resequencing. Multiple sequence alignment of the human CYP1B1 protein was performed along with other CYP1 protein across different families, to check for the conservation of the residues. The SIFT (sorting tolerant from intolerant) homology tool (http://blocks.fhcrc.org/sift/SIFT.html/ provided in the public domain by the Fred Hutchinson Cancer Research Center, Seattle, WA) was used to assess the effect of the substituted amino acid on the CYP1B1 protein, and a threshold score of less than 0.05 was considered to be deleterious to the protein.28
Statistical Analysis
The maximum-likelihood estimates of allele frequencies, Hardy-Weinberg equilibrium, and haplotype frequencies were estimated from the genotype data at six single-nucleotide polymorphism (SNP) loci using Haploview software, which uses the EM (expectation-maximization) algorithm.29 Pair-wise linkage disequilibrium (LD) between the individual SNPs was calculated using the LD-plot function of this software. The odds ratios were calculated, to assess the risk of the individual genotypes at all six SNP loci. Clinical parameters, such as IOP at presentation, cup-to-disc ratio, and visual field defects for the worst eye were considered when correlating the genotype with the phenotype. All calculations were performed with commercial software (SPSS ver. 14; SPSS, Chicago, IL).
| Results |
|---|
|
|
|---|
|
|
Homozygosity of the mutant allele was noted in a JOAG case with G61E and in a POAG case with P193L mutations. There was only a single JOAG case with a compound heterozygous mutation (G61E and R368H). All other mutations were observed in the heterozygous state in JOAG (5/7), POAG (17/18), and all PACG.
The CYP1B1 mutation frequencies were different across all the studies performed on POAG in Indian populations.20 21 Of interest, the investigators in the study from Southern India found a carrier rate of 6.4% and 0.7% for the E229K and the R368H mutations, respectively, in their control populations21 that was not observed in the cohorts from Eastern India20 or in the present study.
Table 2 provides a comparison of JOAG, POAG, and PACG cases. As is evident from the table, JOAG cases had a higher prevalence of CYP1B1 mutations than did POAG cases. There was no significant difference in age at onset among JOAG cases with (20.1 ± 8.78 years) and without (20.9 ± 8.31 years) CYP1B1 mutations (P = 0.781). JOAG cases had a significantly higher mean IOP at presentation than did POAG cases, with and without CYP1B1 mutations (P < 0.001). The mean IOPs were similar among the JOAG and PACG cases with and without mutations. CYP1B1 mutations did not seem to be associated with disc changes (P = 0.192) and severe visual field defects (P = 0.417) in any of these phenotypes.
|
Four different haplotypes (with frequency >5%) were generated with these six SNPs in cases and controls. There were no significant differences in the haplotype frequencies when all POAG and PACG cases were compared with the controls (Tables 3 4) . Reanalysis of the cases with respect to their mutation status indicated a significantly higher frequency of the C-C-G-G-T-A haplotype in both POAG (P = 0.006) and PACG (P = 0.043) cases with CYP1B1 mutations (CYP1B1+) than controls. However, there was no observable difference in frequencies of the other haplotypes among cases and controls. The significantly higher frequency of the C-C-G-G-T-A haplotype in POAG (P = 0.001) and PACG (P = 0.020) cases with CYP1B1 mutations was consistent, even when compared with cases without (CYP1B1–) mutations (data not shown).
|
|
| Discussion |
|---|
|
|
|---|
|
Although we observed a higher mutation frequency of CYP1B1 in POAG than in other populations, our results are not very different from those in a Spanish population,19 when we look at the confidence intervals in these two studies (Table 5) . The frequency differs, however, from those in French and other Indian populations. These differences may be partially attributable to the definitions of POAG used in these studies.18 20 21 In contrast to the French and other Indian studies, we used raised IOP (>21 mm Hg) in the definition of POAG and PACG, as it was our inclusion criterion. It is well known that CYP1B1 is a major candidate gene in PCG that is associated with increased IOP.14 15 16 Hence, this could partially explain the higher frequency of CYP1B1 mutations in our patient cohort. The report on the Spanish patients with POAG19 also included increased IOP (>21 mm Hg) as a major inclusion criterion, and, as just noted, their mutation rates are not very different from ours (Table 5) .
It is interesting to note that the prevalent mutation was different across all previously reported POAG populations (Table 5) . Also, the frequency of heterozygous mutations was similar across these studies. Although the R368H mutation was common in patients in both the Indian and Canadian studies, it was noted in only 2 of the 60 patients with JOAG in the Canadian report.17 One of the Canadian patients with the R368H mutation had an East Indian/Guyanese ancestry,17 but we were not able to determine whether this patient shared a common haplotype background with the Indian patient due to unavailability of data.
The median age at onset of the patients with POAG in the present cohort was similar to that of the French sample,18 but was significantly lower than that of the Spanish19 patients with POAG. The median age of the Canadian patients was significantly lower, as no cases older than 40 years were enrolled.17 Another study on patients with POAG from Eastern India20 reported a mutation frequency (4.5%) similar to that of the French population but a higher mean age (52.43 ± 19.33 years) than that of our cohort. Of interest, the prevalent mutation in the Eastern Indian (S515L)20 and Southern Indian (E229K)21 cohort was also different from that in the present study (R368H).
Another interesting observation was the presence of CYP1B1 mutations on specific haplotypes that was earlier observed in PCG.16 We noted that C-C-G-G-T-A was the risk haplotype in cases of POAG and PACG with CYP1B1 mutations. These results were consistent (even after reanalyzing the data set) based on a five-locus haplotype (i.e., C-G-G-T-A), similar to previous studies in different PCG populations worldwide.16 32 33 35 37 On the other hand, the G-T-C-C-A haplotype that was largely associated with the unaffected controls and PCG cases without CYP1B1 mutations16 was similar in frequency in the POAG and PACG cases with CYP1B1 mutations and the controls (Tables 3 4) . In tune with our previous study on PCG,16 most of the mutations observed in the POAG and PACG clustered on the C-G-G-T-A haplotype. The R368H mutation, which was the prevalent mutation in POAG and PACG in the present study, similar to PCG in India,16 22 was found on the background of the C-G-G-T-A haplotype across all these phenotypes. Of interest, this mutation was also found on the same haplotype in Saudi Arabian32 and Brazilian37 PCG patients. The G61E mutation in POAG was also found on the C-G-G-T-A haplotype, similar to that observed among the PCG patients from Ecuador,31 Saudi Arabia,32 and Morocco.33 The E229K mutation that was observed on the G-T-C-C-A haplotype among patients with PCG in India16 and Germany35 was also seen to harbor the same mutation in POAG and PACG cases. Another striking similarity was the presence of the Y81N mutation in a case of POAG on the G-T-C-C-A haplotype. This mutation was also found on the same haplotype among German patients with PCG.35
Similar to our earlier hypothesis on the evolution of CYP1B1 mutations, we confirm that there is a strong clustering of these mutations on specific haplotype backgrounds, irrespective of geographical location.16 A larger proportion of mutations were seen on the C-G-G-T-A haplotype and a smaller proportion on the G-T-C-C-A haplotype, further confirming the former to be an ancient haplotype and the latter to be a recent haplotype.16 A formal haplotype comparison among other POAG cases with CYP1B1 mutations was not possible due to the unavailability of data from other populations. Based on the present analysis we speculate that the presence of specific CYP1B1 mutations on specific haplotype backgrounds in PCG worldwide and in patients with POAG and PACG in the present cohort is indicative of common founders. The mutations on these haplotypes would have migrated across different geographical regions due to population movements as reported in PCG in our previous study.16 Thus, glaucoma-associated CYP1B1 mutations share a similar haplotype background across POAG, PACG, and PCG.
The role of CYP1B1, particularly in retinoic acid synthesis is pivotal during embryonic development. Recent studies on chick embryogenesis have demonstrated its importance in the dorsoventral patterning of the neural tube that is consistent with its endogenous expression.38 Several in vitro and in vivo studies in lower organisms have demonstrated the sites of expression of CYP1B1 at different stages of development in the anterior retina and anterior segment of the eye.38 39 40 41 Although these studies have provided convincing evidence of the possible role of CYP1B1, its actual molecular mechanism leading to glaucoma in humans has to be deciphered. Although the functions of CYP1B1 mutations leading to POAG and PACG remain to be characterized, it is nevertheless an important candidate gene that should be screened in patients with glaucoma worldwide, to establish its involvement in the diseases pathogenesis.
| Acknowledgements |
|---|
| Footnotes |
|---|
Submitted for publication May 28, 2007; revised August 21, 2007; accepted October 10, 2007.
Disclosure: S. Chakrabarti, None; K.R. Devi, None; S. Komatireddy, None; K. Kaur, None; R.S. Parikh, None; A.K. Mandal, None; G. Chandrasekhar, None; R. Thomas, None
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C.
1734 solely to indicate this fact.
Corresponding author: Subhabrata Chakrabarti, Brien Holden Eye Research Centre, L.V. Prasad Eye Institute, Road No. 2, Banjara Hills, Hyderabad 500034, India; subho{at}lvpei.org.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Pasutto, G. Chavarria-Soley, C. Y. Mardin, K. Michels-Rautenstrauss, M. Ingelman-Sundberg, L. Fernandez-Martinez, B. H. F. Weber, B. Rautenstrauss, and A. Reis Heterozygous Loss-of-Function Variants in CYP1B1 Predispose to Primary Open-Angle Glaucoma Invest. Ophthalmol. Vis. Sci., January 1, 2010; 51(1): 249 - 254. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. N. Rao, I. Kaur, and S. Chakrabarti Lack of association of three primary open-angle glaucoma-susceptible loci with primary glaucomas in an Indian population PNAS, November 3, 2009; 106(44): E125 - E126. [Full Text] [PDF] |
||||
![]() |
S. Chakrabarti, K. N. Rao, I. Kaur, R. S. Parikh, A. K. Mandal, G. Chandrasekhar, and R. Thomas The LOXL1 Gene Variations Are Not Associated with Primary Open-Angle and Primary Angle-Closure Glaucomas Invest. Ophthalmol. Vis. Sci., June 1, 2008; 49(6): 2343 - 2347. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |