Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision Received 26 November 2009 | Accepted 24 December 2009 | Published 30 December 2009 Identification of four novel cytochrome P4501B1 mutations (p.I94X, p.H279D, p.Q340H, and p.K433K) in primary congenital glaucoma patients Mukesh Tanwar,1 Tanuj Dada,2 Ramanjit Sihota,2 Rima Dada1 1Laboratory For Molecular Reproduction and Genetics, Department of Anatomy, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India; 2Dr. R.P. Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India Purpose: Primary congenital glaucoma (PCG) is an autosomal recessive eye disorder that is postulated to result from developmental defects in the anterior eye segment. Mutations in the cytochrome P4501B1 (CYP1B1) gene are a predominant cause of congenital glaucoma. In this study we identify CYP1B1 mutations in PCG patients. Methods: Twenty-three unrelated PCG patients and 50 healthy controls were enrolled in the study. CYP1B1 was screened for mutations by PCR and DNA sequencing. Results: DNA sequencing revealed a total of 15 mutations. Out of these, four (p.I94X, p.H279D, p.Q340H, and p.K433K) were novel mutations and five were known pathogenic mutations. Five coding single nucleotide polymorphisms and one intronic single nucleotide polymorphism (rs2617266) were also found. Truncating mutations (p.I94X and p.R355X) were associated with the most severe disease phenotype. It is possible that patients with two null alleles with no catalytic activity may present with a more severe phenotype of the disease compared to patients with one null allele (heterozygous). The disease phenotype of patients with CYP1B1 mutations was more severe compared with the clinical phenotype of patients negative for CYP1B1 mutations. Conclusion: Mutations in CYP1B1 are a major cause for PCG in our patients. Identifying mutations in subjects at risk of developing glaucoma, particularly among relatives of PCG patients, is of clinical significance. These developments may help in reducing the disease frequency in familial cases. Such studies will be of benefit in the identification of pathogenic mutations in different populations and will enable us to develop simple and rapid diagnostic tests for analyzing such cases. Primary congenital glaucoma (PCG; OMIM 231300) is births in Europe and 1 in 3,300 births in Andhra Pradesh, India an autosomal recessive disorder of the eye. In this disease the [2,3]. trabecular meshwork (TM) and anterior chamber of the eye Recently a putative PCG locus, GLC3A, was linked to are affected, leading to impairment in the aqueous drainage, markers on the short arm of chromosome 2 in 11 Turkish increased intraocular pressure (IOP), and optic nerve damage. families [4]. Six other families did not show linkage to this PCG occurs during the neonatal or early infantile period [1]. locus, suggesting locus heterogeneity for this disease. Another The term PCG is reserved for those cases in which the only PCG locus, GLC3B, was localized on chromosome 1p36 in anatomic defect observed is isolated trabeculodysgenesis. some families but did not show linkage to chromosome 2 This increased IOP results in ocular enlargement markers [5]. Other subsets of families that did not show (buphthalmos), corneal clouding, and rapid optic nerve linkage to these two loci provide evidence for at least a third cupping. Progressive degeneration of the retinal ganglion cells of the unmapped loci [5]. Recently Stoilov et al. [6] identified (RGCs) results in the characteristic optic nerve atrophy and three different mutations in the cytochrome P4501B1 visual field defects found in glaucoma. Most cases of PCG are (CYP1B1) gene in five unrelated Turkish families in which sporadic, but familial cases have also been reported. PCG is the disease had been linked to the 2p21 locus [6]. Even though the most common type of pediatric glaucoma and accounts for three different loci have been mapped for PCG, mutations in 55% of such cases. Its expression and penetrance vary from CYP1B1 (GLC3A) are the most predominant cause of disease 40–100%. Its incidence varies substantially from one and are reported in various ethnic backgrounds [6-15]. population to another. It is estimated to occur in 1 in 10,000 Further, it is estimated that all the known loci/genes of glaucoma account for a minority of the total cases of glaucoma [4,5], and hence many other genes remain to be identified. Correspondence to: Dr. Rima Dada, Associate Professor, Laboratory CYP1B1 is located on chromosome 2 and consists of three For Molecular Reproduction and Genetics, Department of Anatomy, exons and two introns. The coding region of CYP1B1 starts at All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India 110029; Phone: +91-11-26546716; FAX: +91-11-26588663; the 5′ end of exon 2 and ends within exon 3. It codes for a 543- email: [email protected] amino acid protein and is expressed in the ocular tissues, such 2926 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision TABLE 1. THE PRIMERS USED FOR PCR AMPLIFICATION. Primer sequence Product size (bp) 1F-5′-TCTCCAGAGAGTCAGCTCCG-3′ 786 1R-5′-GGGTCGTCGTGGCTGTAG-3′ 2F-5′-ATGGCTTTCGGCCACTACT-3′ 787 2R-5′-GATCTTGGTTTTGAGGGGTG-3′ 3F-5′-TCCCAGAAATATTAATTTAGTCACTG-3′ 885 3R-5′-TATGGAGCACACCTCACCTG-3′ as the anterior chamber, and in several nonocular tissues method. The entire coding region, including exon–intron [16]. CYP1B1 is a member of the cytochrome P450 boundaries of CYP1B1, from patients and controls was superfamily of drug-metabolizing enzymes. It catalyzes amplified and screened for mutations by using three sets of several oxidative reactions, some of which are biosynthetic, overlapping primers (Table 1) [7,21]. The primers used were producing necessary hormones or compounds of intermediary set I (1F–1R, 786 bp) [12], set II (2F–2R, 787 bp) [13], and metabolism in most living organisms and substrates, including set III (3F–3R, 885 bp) [12]. PCR amplifications for primer many xenobiotics, vitamins, and steroids [17]. CYP1B1 also sets I and II were performed in a 40 µl volume containing 1.0 metabolizes vitamin A in two steps to all-trans-retinal and all- µl of 20 µM stock solution for each primer, 100 ng of genomic trans-retinoic acid. The latter is a potent morphogen and DNA, 1 unit of Taq polymerase (Banglore Genei), 0.1 mM of regulates in utero development of tissue growth and each dNTP, 4 µl of 10X PCR buffer (with 15 mM MgCl) and 2 differentiation. CYP1B1 is involved in the metabolism of 4 µl of dimethyl sulphoxide (Sigma), by means of 35 cycles endogenous and exogenous substrates that take part in early of amplification, each consisting of 30 s denaturation at 94 ocular differentiation [18-20]. In the present study we °C, 30 s annealing at 56 °C and 1 min extension at 72 °C [12], screened all coding exons of CYP1B1 in 23 unrelated while conditions for set III were initial denaturation at 94 °C congenital glaucoma patients. for 3 min followed by 30 cycles each at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 1 min. METHODS Amplified PCR products were purified using a gel/PCR Clinical evaluation and patient selection: Primary congenital DNA fragments extraction kit (number DF100; Geneaid glaucoma cases presenting at the Dr. R. P. Centre for Biotech Ltd., Sijhih City, Taiwan). Purified PCR products Ophthalmic Sciences (AIIMS, New Delhi, India), were were sent for sequencing to MCLAB (Molecular Cloning enrolled for this study. Six patients were female and 17 were Laboratories, South San Francisco, CA). DNA sequences male. Mean age of presentation was 15.17 months (range 1.5 were analyzed against the CYP1B1 reference sequence – 132 months). After ethical approval of the Institutional ENSG00000138061 using ClustalW2 (multiple sequence Review Board (IRB00006862; All India Institute of Medical alignment program for DNA; European Molecular Biology Sciences, New Delhi, India), 23 PCG cases were enrolled in Laboratory (EMBL) – European Bioinformatics Institute this study. The diagnosis involved clinical ocular and systemic (EBI)). examination. Inclusion criteria of the patients were increased Computational assessment of missense mutations: Two corneal diameter (>12.0 mm) and raised IOP (>21 mmHg) homology-based programs PolyPhen (polymorphism with presence/absence of Haab’s striae and optic disc changes phenotyping; Division of Genetics, Department of Medicine, (where examination was possible). Symptoms of epiphora and Brigham and Women’s Hospital/Harvard Medical School, photophobia were the additional inclusion factors. The age of Boston, MA) and SIFT (sorting intolerant from tolerant; the onset ranged from birth to 3 years. Detailed family histories J. Craig Venter Institute Rockville, MD and La Jolla, CA) up to three generations were taken, and pedigree charts were were used to predict the functional impact of missense constructed. The history of ocular or other hereditary changes identified in this study. PolyPhen structurally disorders was recorded. Glaucoma cases other than PCG were analyzes an amino acid polymorphism and predicts whether excluded. Fifty ethnically matched normal individuals that amino acid change is likely to be deleterious to protein without any ocular disorders were enrolled as controls. function [22-24]. The prediction is based on the position- Peripheral blood samples were collected from patients and specific independent counts (PSIC) score derived from controls by venipuncture after informed consent. Blood multiple sequence alignments of observations. PolyPhen samples were collected in EDTA vaccutainer and stored in -80 scores of >2.0 indicate the polymorphism is probably °C until DNA isolation. damaging to protein function. Scores of 1.5–2.0 are possibly Mutation screening and sequence analysis: Genomic DNA damaging, and scores of <1.5 are likely benign. SIFT is a was isolated from peripheral blood by the phenol chloroform sequence homology-based tool that sorts intolerant from 2927 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision tolerant amino acid substitutions and predicts whether an number c.835. This resulted in a codon change from CAC to amino acid substitution in a protein will have a phenotypic GAC and an amino acid change from histidine to aspartic acid effect [25-28]. SIFT is based on the premise that protein (p.H279D), a nonsynonymous mutation in the CYP1B1 evolution is correlated with protein function. Positions protein. This mutation was identified in one patient (P61) and important for function should be conserved in an alignment was heterozygous. of the protein family, whereas unimportant positions should Glutamine340histidine (p.Q340H) mutation—In this appear diverse in an alignment. Positions with normalized mutation a single base G was replaced by thymine (T) (Figure probabilities <0.05 are predicted to be deleterious and those 3) at genomic position 38301512, coding nucleotide number ≥0.05 are predicted to be tolerated. c.1020. This resulted in a codon change from CAG to CAT and an amino acid change from glutamine to histidine RESULTS (p.Q340H), a nonsynonymous mutation in the CYP1B1 All cases were found to be sporadic in origin. A total of 15 protein. This mutation was identified in one patient (P56) and nucleotide changes were observed in this study. Out of these, was heterozygous and present with the p.R390H mutation in five were previously reported coding single nucleotide this patient. polymorphisms (SNPs) and one was already reported as an Lysine433lysine (p.K433K) mutation—In this intronic SNP; five were known pathogenic CYP1B1 mutation a single base G was replaced with adenine (A) mutations. Four novel nucleotide changes (two (Figure 4) at genomic position 38298198, coding nucleotide nonsynonymous, one frameshift, and one synonymous number c.1299. This resulted in a codon change from AAG mutation) were also found in this study. Details of all to AAA and resulted in no amino acid change (lysine). This nucleotide changes are presented below. was a neutral mutation (p.K433K) in patient P69. Identification of four novel mutations: All four novel mutations p.I94X, p.H279D, p.Q340H, Isoleucine94stop (p.I94X) mutation—In this mutation and p.K433K have been registered in GenBank with accession a single base guanine (G) deletion (Figure 1) was observed at numbers GQ925803, GQ925804, GQ925805, and genomic position 38302285, coding nucleotide number c.247. GQ925806, respectively. This caused a frameshift after codon 82 and introduced a stop Other previously reported pathogenic CYP1B1 mutations: codon (TAG) at position 94 in the protein. This mutation Glutamic acid229lysine (p.E229K) mutation—This produced a truncated CYP1B1 protein of 93 amino acids. This mutation resulted in G being replaced with A at genomic change was identified as a homozygous mutation in the patient position 38301847 (rs57865060), coding nucleotide number (P55). c.685. This resulted in a codon change from GAA to AAA and Histidine279aspartic acid (p.H279D) mutation—In an amino acid change from glutamic acid to lysine (p.E229K), this mutation a single base cytosine (C) was replaced by G a nonsynonymous mutation in the CYP1B1 protein. This (Figure 2) at genomic position 38301697, coding nucleotide change was found in one patient (P65) and was heterozygous. Figure 1. DNA sequence chromatogram of CYP1B1 exon 2 Figure 2. DNA sequence chromatogram of CYP1B1 exon 2 equivalent to codon 81–85. A: The reference sequence derived from equivalent to codon 277–280. A: The reference sequence derived control is shown. B: Sequence derived from congenital glaucoma from control is shown. B: Sequence derived from congenital patient P55 shows the homozygous deletion of G at c.247, which glaucoma patient P55 shows heterozygous c.835C>G, which caused a p.asp83thrfsX12 (p.I94X) mutation. predicts a codon change of CAC>GAC and a p.H279D mutation. 2928 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision Arginine355stop (p.A355X) mutation—In this Arginine368histidine (p.R368H) mutation—In this mutation a single base C was replaced by T (Figure 5) at mutation a single base G was replaced by A at genomic genomic position 38298434, coding nucleotide number c. position 38298394 (rs28936414), coding nucleotide number 1063. This resulted in a codon change from CGA to TGA c.1103. This resulted in a codon change from CGT to CAT (p.R355X), a nonsense mutation in the CYP1B1 protein. This and an amino acid change from arginine to histidine resulted in a truncated CYP1B1 protein of 355 amino acids. (p.R368H), a nonsynonymous mutation. This change was The p.R355X mutation was described only once in the homozygous in one patient (P68). literature [29]. This change was homozygous in one patient Arginine390cysteine (p.R390C) mutation— In this (P70). mutation a single base C was replaced by T at genomic position 38298329 (rs56010818), coding nucleotide number c.1168. This resulted in a codon change from CGC to TGC and an amino acid change from arginine to cysteine (p.R390C), a nonsynonymous mutation. This mutation was identified in one patient (P64) and was heterozygous. Arginine390histidine (p.R390H) mutation—In this mutation a single base G was replaced by A (Figure 6) at genomic position 38298328, coding nucleotide number c. 1169. This resulted in a codon change from CGC to CAC and an amino acid change from arginine to histidine (p.R390H), a nonsynonymous mutation. This mutation was identified in one patient (P56) and was heterozygous. Nonpathogenic CYP1B1 single nucleotide polymorphisms: In addition to these pathogenic mutations, six previously reported single nucleotide polymorphisms [8] were identified in a less conserved region of the CYP1B1 protein. Details of these polymorphisms are provided below. Figure 3. DNA sequence chromatogram of CYP1B1 exon 2 equivalent to codon 339-342. A: The reference sequence derived Cytisine (C) to thymine (T) change in intron 1—In this from control is shown. B: Sequence derived from congenital mutation, C was replaced by T at genomic position 38302544, glaucoma patient P56 shows heterozygous c.1020G>T change, nucleotide position 780 in CYP1B1 (rs2617266) in intron I. which predicts a codon change CAG>CAT and heterozygous This was observed in 13 patients but was absent in controls. p.Q340H mutation. Figure 4. DNA sequence chromatogram of CYP1B1 exon 3 equivalent to codon 431-434. A: The reference sequence derived Figure 5. DNA sequence chromatogram of CYP1B1 exon 3 from control is shown. B: Sequence derived from congenital equivalent to codon 353-356. A: The reference sequence derived glaucoma patient P69 shows heterozygous c.1294C>G and from control is shown. B: Sequence derived from congenital heterozygous c.1299G>A, which predicts codon change CTG>GTG glaucoma patient P70 shows homozygous c.1063C>T, which and AAG>AAA and heterozygous p.L432V and p.K433K predicts a codon change CGA>TGA and p.R355X nonsense mutations, respectively. mutation. 2929 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision Arginine48glycine (p.R48G)—In this mutation, C was Asparagine453serine (p.N453S)—In this mutation a replaced by guanine (G) at genomic position 38302390 single base A was replaced by G at genomic position (rs10012), coding nucleotide number c.142. This resulted in 38298139 (rs1800440), coding nucleotide number c.1358. a codon change from CGG to GGG and an amino acid change This resulted in a codon change from AAC to AGC and an from arginine to glycine (p.R48G), a nonsynonymous amino acid change from asparagine to serine (p.N453S), a mutation in the CYP1B1 protein. This change was also present nonsynonymous mutation in the CYP1B1 protein. The in controls. p.N453S mutation was present in two patients (P51 and P62) Alanine119serine (p.A119S)—In this mutation G was but absent in controls. replaced by T at genomic position 38302177 (rs1056827), The clinical manifestations of PCG patients have been coding nucleotide number c.355. This resulted in a codon tabulated (Table 2), and the CYP1B1 sequence variants change from GCC to TCC and an amino acid change from identified in the various studies to date have been summarized alanine to lysine (p.A119S), a nonsynonymous mutation in (Table 3). The clinical phenotype of the cases with pathogenic the CYP1B1 protein. This change was found in patients P55 CYP1B1 mutations was more severe compared to cases and P73 but absent in controls. without CYP1B1 mutations. The mean IOP of cases with Leucine432valine (p.L432V)—In this mutation a single pathogenic CYP1B1 mutations was 30.21 mmHg compared base C was replaced by G at genomic position 38298203 to 23.96 mmHg in mutation-negative cases; the difference is (rs1056836), coding nucleotide number c.1294. This resulted significant (p value <0.005). The mean corneal diameter in in a codon change from CTG to GTG and an amino acid patients without the CYP1B1 mutations was 12.625×12.181 change from leucine to valine (p.L432V), a nonsynonymous mm (left eye) and 12.406×12.781 mm (right eye), whereas it mutation in the CYP1B1 protein. This mutation was identified was 13.833×13.750 mm (left eye) and 13.416×15.50 mm in four patients; it was homozygous in three patients (P52, (right eye) in mutation-negative cases. Haab’s striae were P55, and P68) and heterozygous in one patient (P69) and was present in two cases (P56 and P61), which were positive for also present in controls. the CYP1B1 mutations. Aspartic acid449aspartic acid (p.D449D)—In this mutation a single base T was replaced by C at genomic DISCUSSION position 38298150 (rs1056837), nucleotide position 5174 in Structural/functional implications of mutations: the gene, coding nucleotide number c.1347. This resulted in p.I94X mutation—In the isoleucine94stop mutation (p. a codon change from GAT to GAC and no change in the amino 194X) mutation a truncated protein of 93 amino acids is acid (aspartic acid) (p.D449D), a synonymous mutation in the produced in which only the first 82 amino acids are the same CYP1B1 protein. This mutation was identified in 20 patients as the wild-type CYP1B1 protein (Figure 7). This truncated and was homozygous in all. This change was also present in protein lacks all functional domains of the CYP1B1 protein controls. and is a nonfunctional protein [6,21,29,30]. p.H279D mutation—This histidine residue lies in the carboxyl terminal of the G helix in the CYP1B1 protein. Replacement of an aromatic, weak basic, amino acid histidine whose charge state depends upon its protonation state with an aliphatic, strong acidic, and negatively charged aspartic acid at this locus. This in turn affects the local charge distribution, and hence the structure of the protein is disturbed. Histidine is conserved at this locus in the CYP1A1 protein from 12 different species (Figure 8) and in the CYP1B1 protein from seven different species (Figure 9) analyzed, suggesting that histidine performs some important functions at this locus. No other known pathogenic mutation was present in the patient (P61), and the PSIC score of this mutation was 2.628, indicating that this change is probably damaging to the protein function. The SIFT score of p.H279D was 0.00 and is predicted to be deleterious for the protein function. p.Q340H mutation—This glutamine residue lies in the carboxyl terminal of the I helix. Replacement of a polar Figure 6. DNA sequence chromatogram of CYP1B1 exon 3 uncharged amino acid (glutamine) with a weak basic amino equivalent to codon 388-391. A: The reference sequence derived acid (histidine) may or may not alter the structure/function of from control is shown. B: Sequence derived from congenital glaucoma patient P56 shows homozygous c.11169G>A, which the protein. Glutamine is not conserved at this locus in the predicts a codon change CGC>CAC and p.R390H mutation. CYP1A1 protein from 12 different species analyzed (Figure 2930 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ab/ ent; Treatments Medical and OU TrTrab+MMC; OUcataract surgeryMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OUTrTrab+MMCMedical and OD TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMCMedical and OU TrTrab+MMC ycin C treatm m Mutations — — — — p.I94X (H) p.Q340H (H)+ p.R390H (H)— — — — p.H279D (h) — — p.R390C (h) p.E229K (h) — — p.R368H (H) — p.R355X (H) — — — y and mito m o ct e Photo-phobia Yes Yes Yes No No No No No No Yes No Yes No Yes No no Yes no no no Yes Yes no becul a r 1 1 y t Last Cup Discratio OS/OD 0.8:1/0.9:1 Hazy media/0.5:1Hazy media No glow NA/0.9:1 Hazy media 0.7:1/0.7:1 total cupping/0.5:1Not available 0.4:1/0.6:1 Not visible/0.9: 0.6:1/0.6:1 0.4:1/0.4:1 Not visible/0.7: 0.7:1/0.7:1 0.4:1/0.4:1 Not visible 0.8:1/0.8:1 0.5:1/0.6:1 Not visible 0.4:1/0.4:1 No glow Hazy media beculotom a r d t .NTS Haabs’striae no no no no no OU +ve no no OS +ve no OU +ve no no no no no no no no Notvisibleno no no mbine G PATIE - co C C T 2. C PABLELINICALMANIFESTATIONSOF BuphthalmosIOP OS/OD(mmHg) Atpresentation OU;OD>OS22/28 OU20/20 OU; OD>OS40/23 OU26/26 OD; OS PhthisicNA/37eyeOU; OS>OD30/28 OU; OD>OS28/30 OU; OS>OD20/16 OU; OS>OD31/31 OU; OS>OD24/18 OU; OS>OD30/34 OU; OS>OD22/16 OU; OS>OD22/23 OU; OS>OD28/24 OU; OS>OD25/26 OU22/24 OU; OD>OS21/24 OU; OD>OS26/28 OU; OS>OD24/26 OU/; OS>OD30/40 OU22/23 OU; OS>OD20/24 OU; OS>OD22/22 us; X- times; Trab/Trab+MMd letters- novel mutations. Corneal Diameter(mm) OS/OD andclarity atdiagnosis11x11.5/13x13;OU mild edema 12.5x13/12.5x13;OS mild edema11.5x12/12x12;OU mild edema15x14.5/15x14.5;no edemaPhthisic eye/12x12;OU severe edema14.5x14.5/14x14;OU severe edema13x13/13.5x13.5;No edema14x14/12.5x12.5;No edema14x14.5/13.5x14;OS edema14x14/13x13.5; noedema15.5x15/14x14;OU mild edema11x11.5/10x11.5;OD edema12x12/11x12.5; Noedema12x12/11.5x11.5;OD severe edema13x13/12.5x13;OU edema12.5x12/12.5x12;No edema12x13/13x14; OUmild edema13x13/13.5x14;OU severe edema13x13/12x13; Noedema15x15/15x14.5;OU severe edema12x2/12x12; Noedema12x12/12x11.5;OU mild edema12.5x13/11x12;OU mild edema s; h-heterozygomutations in bol goues; Age at presentation/sampling 36 months 2 months 4 months 9 months 8 months 12 months 3 months 15 months 10 months 41 months 4 months 8 months 12 months 1 month 6 months 13 months 132 months 6 months 4 months 45 days 18 months 45 days 2 months male; H-homozyye; OU- both ey Sex F F F M M F M M F F M M M M M M M M M M M M M e; F- feS- left e alO Age ofonset ofdisease By birth By Birth By birth By birth By birth By birth By birth By birth By birth 7 months By birth By birth 3 months By birth By birth 3 months 11 months By birth By birth By birth 13 months By birth By month e: M- mht eye; notrig Pt.ID P51 P52 P53 P54 P55 P56 P57 P58 P59 P60 P61 P62 P63 P64 P65 P66 P67 P68 P69 P70 P71 P72 P73 FootOD- 2931 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision e; Origin (reference) India, Saudia Arabia, Oman, Brazil [7,13,35,36] Saudia Arabia, India, Japan [8,11,15] This study Saudia Arabia, Japan, India [7,11,15] France, India, Germany [7,29,31] This study This study Germany, India [28] this study Saudia Arabia, India. France [7,15,32] Ecuador, India [8,37] Pakistan, India, France [7,21,32] India, Japan, Turkey [7,11,29] This study Japan, India [7,11] France, India [7,12] X- stop codon; NA- not applicabl Observationalhistory ofmutations indifferent diseasesPCG PCG PCG PCG PCG, POAG PCG PCG PCG PCG, PA, POAG PCG PCG, POAG PCG, PA PCG PCG PCG FS-frameshift; .OUSSTUDIESTILLDATE Mutationidentified NA p.arg48 gly(p.R48G)p.asp83thrfsX12(p.I94X)p.ala119ser(p.A119S)p.glu229lys(p.E229K)p.his279asp(p.H279D)p.gln340his(p.Q340H)p.ala355stop(p.A355X)p.arg368his(p.R368H)p.arg390cys(p.R390C)p.arg390his(p.R390H)p.leu432lys(p.L432V)p.lys433lys(p.K433K)p.asp449asp(p.D449D)p.asp453ser(p.N453S) eter’s anomaly; RI P NTSIDENTIFIEDINVA Location inprotein NA 48 FS after 82 119 229 279 340 355 368 390 390 432 433 449 453 ucoma; PA- T 3. S VARIATHEABLEUMMARYOFSEQUENCE Codon changeType ofmutation NAIntronic CGG>GGGMissense FSFS GCC>TCCMissense GAA>AAAMissense CAC>GACMissense CAG>CATMissense GGA>TGANon-sense CGT>CATMissense CGC>TGCMissense CGC>CACMissense CTG>GTGMissense AAG>AAANeutral GAT>GACNeutral AAC>AGCMissense G- Primary open angle gla Nucleotidechange C>T c.142 C>G c.247del. G c.355G>T c.685G> A c.835C>G c.1020G>T c.1063C>T c.1103G>A c.1168C>T c.1169G>A c.1294C>G c.1299G>A c.1347T>C c.1358A>G ma; POA o ucns. Genomiclocation g.38302544 g.38302390 g.38302285 g.38302177 g.38301847 g.38301697 g.38301512 g.38298434 g.38298394 g.39298329 g.38298328 g.38298203 g.38298198 g.38298350 g.38298139 genital glael mutatio nv oo cn Patient number P51, P53, P55, P58-P65,P71, P72, P73P51, P53, P54, P57-P61,P63-P65, P71-P73P55 P55, P73 P65 P61 P56 P70 P68 P64 P56 P52, P55, P68, P69 P69 P51, P53-P67, P69-P71,P73P56, P67 e: PCG- Primary ns in bold letters- S. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Footnotmutatio 2932 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision Figure 7. Amino acid sequence of CYP1B1 protein. A: Wild-type CYP1B1 protein. B: Truncated CYP1B1 protein of 93 amino acids (black arrow shows the position after which frameshift takes place and red letters shows amino acids after frameshift). C: Truncated CYP1B1 protein of 354 amino acids. 8) but is conserved in the CYP1B1 protein from seven terminal of the J helix, carboyxl terminal of the J helix is different species analyzed (Figure 9). The PSIC score of this involved in the functionally important heme-binding domain. mutation was 0.276, indicating that this change is benign to This truncating mutation results in a loss of the heme-binding protein function. The SIFT score of p.Q340H was 0.05 and is domain and a functionally inactive protein [6,21,29,30]. predicted to be tolerated. The patient with the p.Q340H p.R368H mutation—This arginine residue lies between mutation also had a known pathogenic CYP1B1 mutation the J and K helix in an exposed loop [8,15]. In this mutation (p.R390H) and had a PSIC score of 2.799 and a SIFT score of the positively charged amino acid arginine is replaced by 0.00. The p.R390H mutation has previously been reported histidine whose charge state depends upon its protonation [21] to adversely affect or damage protein function. state. Consequences of this change are not immediately p.E229K mutation—The p.E229k mutation occurred in apparent. In the wild type, arginine at position 368 interacts the carboxyl terminal of the F helix in the vicinity of the with G-365, D-367, V-363, and D-374. Because of the R368H substrate-binding region in the CYP1B1 protein. Substitution mutation, interaction between D-367 and D-374 are of E to K leads to a change from a negatively charged residue weakened. The PSIC score of this mutation was 2.653, to a positively charged side chain, and this in turn affects the indicating that this change is probably damaging to protein local charge distribution. This disturbs an important cluster of function. The SIFT score of p.R368H was 0.00 and is salt bridges. In wild-type CYP1B1 protein, R-194::E-229, predicted to be deleterious for the protein function. How R-194::D-333, and D-333::K-512 form a triangle of ionic p.R368H affects the conformation and functionality of the bond interactions, holding the I helix with the F helix and β- protein is still not clear [31]. strand S3.2. As a result of this mutation, the R-194::E-229 p.R390H/C mutation—This arginine residue is located interaction is lost, which has the potential to destabilize the in the conserved α helix K [8]. It forms the consensus sequence other ionic interactions in the protein [30]. The SIFT score of GluXXArg, which is conserved among all members of the the p.E229K mutation was 0.01 and is predicted to be cytochrome P450 superfamily [21]. Arg390 and Glu387 are deleterious for the protein function. The CYP1B1 protein with one helical turn apart and are predicted to form a salt bridge. the p.E229K mutation shows 20–40% enzymatic activity The parallel orientation of their side chains is more transparent compared to the wild-type CYP1B1 protein [31]. in the three-dimensional model. Conservation of this motif p.R355X mutation—In the p.R355X mutation, a indicates that presence of arginine at this position is essential truncated protein of 354 amino acids is produced (Figure 7). for the normal function of the P450 molecule. The PSIC scores The arginine residue at position 355 lies in the carboxyl of p.R390C and p.R390H were 3.474 and 2.799, respectively, 2933 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision Figure 8. Multi sequence alignment of the human CYP1B1 protein with the CYP1A1 protein from different species. Red Underlined amino acids shows the conserved residues in human CYP1B1 and different CYP1A1 protein from different species (when mutated) causing primary congenital glaucoma phenotype. While Red letter shows amino acid conserved in different CYP1A1 protein from different species but not present in human CYP1B1 protein. Figure 9. Multisequence alignment of the human CYP1B1 protein with the CYP1B1 protein from different species. Underlined red amino acids show the conserved residues (when mutated) causing the primary congenital glaucoma phenotype. Red colored amino acid shows the non-conservation of glutamic acid at this locus in Zebrafish CYP1B1. Blue-colored amino acids show the less conserved residues in CYP1B1 protein from different species. indicating that both these changes are probably damaging to The patient (P55) with the p.I94X (homozygous) protein function. The SIFT score of p.R390H/C was 0.00 and mutation is a male child of a consanguineous marriage without is predicted to be deleterious for the protein function. any family history of glaucoma; he presented at 8 months of The PSIC scores of the nonpathogenic single nucleotide age. He was born at full term through a normal vaginal polymorphisms were <2 for p.R48G, p.A119S, p.N453S, and delivery. He had severe bilateral corneal edema at birth. At p.L432V, indicating that all these changes were benign to the age of 2 months he had congestion with discharge in the protein function. The SIFT scores of the nonpathogenic single left eye and was diagnosed to have a left corneal ulcer and was nucleotide polymorphisms were >0.05 for p.R48G, p.L432V, treated with antibiotics; the left eye consequently developed p.K433K, and p.D449D, indicating that all these changes were phthisis. The right eye dimensions increased, and he was tolerated in the protein. diagnosed as having buphthalmos at the age of 8 months. PCG is a clinically and genetically heterogeneous Combined trabeculotomy and trabeculectomy with disorder. More than 50 different mutations have been reported mitomycin C was performed in his right eye. He was in the entire coding region of CYP1B1 from various diagnosed as having 100% blindness at 8 months. His parents populations. We screened the entire coding region of were also screened for CYP1B1 mutations by DNA CYP1B1 in 23 congenital glaucoma patients by using primers sequencing but were found to be negative for any pathogenic described elsewhere [8]. Of all mutations identified herein, CYP1B1 mutations. the frameshift mutation (c.247delG) and nonsense mutation Patient P70 has a p.R355X (homozygous) mutation and (c.1063C>T) resulted in the most severe disease phenotype. is a male offspring of a non-consanguineous marriage; he presented at 45 days. At birth, he had bilateral congenital 2934 Molecular Vision 2009; 15:2926-2937 <http://www.molvis.org/molvis/v15/a310> © 2009 Molecular Vision glaucoma and had IOPs of 30 and 40 mmHg in his left and that the onset age in three patients (P60, P66, and P67) was 7, right eye, respectively. He had severe corneal clouding in both 3, and 11 months, while the rest of the patients presented at eyes, at birth, and therefore the fundus was not visualized. birth. In these 20 patients there is no significant difference in Combined trabeculotomy and trabeculectomy with the age of disease onset in CYP1B1 mutation-positive and mitomycin C was performed in both eyes. He had no light mutation-negative cases, although clinical phenotypes of perception and was visually blind since 45 days of age. His patients (P55, P56, P61, P68, and P70) with homozygous parents were also negative for the pathogenic CYP1B1 CYP1B1 mutations were more severe compared to patients mutations. The absence of mutations in the parents of P55 and (P64 and P65) who were heterozygous for the CYP1B1 P70 could be due to a parental germline mutation, which mutations (Table 1). It is possible that patients with two null cannot be tested by using peripheral leukocytes. alleles with no catalytic activity may present with a more Patient P56 is a female child of a non-consanguineous severe phenotype of the disease compared to patients with one marriage; she presented at the age of 1 year. She has p.Q340H null (heterozygote) allele. The disease phenotype of patients (heterozygous) and p. R390H (homozygous) mutations. She with homozygous/heterozygous CYP1B1 mutations was more had bilateral congenital glaucoma since birth. She had a severe compared to the clinical phenotype of patients negative corneal diameter of 14.0×14.0 mm (right eye) and for the CYP1B1 mutations. 14.5×14.5 mm (left eye) and IOPs of 28 and 30 mmHg in the We also observed a higher mean IOP in a group of right and left eye, respectively. She had Haab’s striae in both patients with CYP1B1 mutations. In accordance with the idea eyes, and the fundus was not visible. Combined of associating the severe phenotypes with the null CYP1B1 trabeculotomy and trabeculectomy with mitomycin C was allele, the percentage of severe phenotypes in at least one eye performed in both eyes. has been reported to be associated with various mutations Patient P61 is a male child of a non-consanguineous ranging 80-100% for a frameshift mutation (e.g., c.376insA) marriage; he presented at the age of 4 months and has a and truncating mutations [11]. Three different truncation p.H279D (heterozygous) mutation. He had bilateral mutations (p.C280X, p.E281X, and p.R355X) producing a congenital glaucoma at birth. At presentation corneal diameter truncated protein of 279, 280, and 354 amino acids, and IOPs of his left and right eye were 15.5×15.0 mm and respectively, have also been associated with more severe 14.0×14.0 mm and 30 mmHg and 34 mmHg, respectively. The disease phenotypes [11,21,29]. In patient P55 with a cup to disc ratio of the left eye was not visible due to the hazy homozygous p.I94X mutation, a truncated protein of 93 amino media and that of the right eye was 0.9:1. He had Haab’s striae acids is produced that has the first 82 amino acids similar to in both eyes. Combined trabeculotomy and trabeculectomy the wild-type CYP1B1 protein. The disease phenotype of this with mitomycin C was performed in both eyes. patient is severe with a left phthisic and a right buphthalmic An intriguing finding that apparently does not match a eye with a cup to disc ratio of 0.9:1. He is visually blind. typical recessive pattern of inheritance is the presence of a Another patient (P70) with a p.R355X mutation had bilateral heterozygous CYP1B1 mutation in PCG patients. This buphthalmos with severe corneal edema and a corneal situation has been previously reported [7,29]. A heterozygous diameter of 15.0×15.0 mm and 15.0×14.5 mm in the left and p.Y81N mutation has also been described in PCG patients right eye, respectively. He was blind at the age of 45 days. from Germany, and a heterozygous p.E229K mutation has Patient P61 with a p.H279D mutation had bilateral been identified in unrelated French and Indian patients [7, buphthalmos with mild edema in both eyes and a corneal 32]. Few heterozygous CYP1B1 mutations were associated diameter of 15.5×15.5 mm and 14.0×14.0 mm in the left and with the milder, primary, open-angle glaucoma phenotypes in right eye, respectively. He was blind at the age of 4 months. patients from Spain, France, and India. The presence of a The range of percentages of severe phenotypes in at least one heterozygous CYP1B1 mutation in PCG suggests the eye is 62–83% for different mutations, such as p.G61E, possibility of other loci, yet undetected, that may be involved p.E229K, p.R368H, and p.R390C [9]. in anterior chamber formation. Recently the presence of Membrane-bound cytochromes, such as CYP1B1, have double heterozygote variants CYP1B1 and FOXC1 has been a molecular structure containing a transmembrane domain described in two PCG cases, although the role of possible located at the N-terminal end of the molecule. This is followed digenic inheritance in disease causation is yet to be established by a proline-rich “hinge” region, which permits flexibility [33]. Defective variants of modifier genes and/or between the membrane-spanning domain and the cytoplasmic environmental factors have an additive effect with loss-of- portion of the protein molecule. The COOH-terminal ends are function CYP1B1 alleles to produce the disease phenotype. highly conserved among different members of the cytochrome However further work is required to understand this P450 superfamily [17]. This family contains a set of conserved mechanism. core structures responsible for the heme-binding region of Previous studies have reported that the age of disease these molecules. The heme-binding region is essential for the onset in PCG patients with CYP1B1 mutations is younger than normal function of every P450 molecule. Between the hinge in patients without CYP1B1 mutations [34]. Our data show region and the conserved core structure lies a less conserved 2935