Functional characterization and molecular modeling

0 downloads 0 Views 1MB Size Report
Apr 22, 2018 - in CYP21A2 gene from patients with Congenital Adrenal Hyperplasia. Ragini Khajuria a, Rama Walia b, Anil Bhansali b, Rajendra Prasad a, *.
Biochimie 149 (2018) 115e121

Contents lists available at ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Research paper

Functional characterization and molecular modeling of the mutations in CYP21A2 gene from patients with Congenital Adrenal Hyperplasia Ragini Khajuria a, Rama Walia b, Anil Bhansali b, Rajendra Prasad a, * a b

Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh, India Department of Endocrinology, Postgraduate Institute of Medical Education and Research, Chandigarh, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 August 2017 Accepted 16 April 2018 Available online 22 April 2018

Steroid 21-Hydroxylase deficiency is an inherited autosomal recessive metabolic disorder of the adrenal steroidogenesis caused due to mutations in the CYP21A2 gene in 95% of CAH cases. Notably, the de novo mutations arise at the rate of 3e5%, therefore the functional characterization is of utmost importance for categorization of the novel mutations. Herein, we have functionally characterized the CYP21A2 missense mutations viz., p. F306V and p. H365N. Notably, both the mutations were harbored by the patients exhibiting the non classical phenotype. We followed the approach of in vitro characterization of the mutant proteins expressed in COS7 cells. Of note, all the mutant constructs exhibited reduced residual enzyme activity fraternized with altered kinetic constants accompanied by higher requirement for the activation energy. Further, there was reduced protein expression in the mutant constructs to that of the wild-type. Molecular modeling suggested alteration in the structure-function relationship of the protein due to mutations. The evidence suggested by the in vitro and the in silico characterization of mutations directed us to conclude that both, p. F306V and p. H365N should be considered as non classical CAH causing mutations. Conceivably, the knowledge about the functional consequences of the mutations is the basis for improved genetic counseling with respect to prognosis and therapeutic implications. © 2018 Elsevier B.V. and Société Française de Biochimie et Biologie Moléculaire (SFBBM). All rights reserved.

Keywords: Congenital Adrenal Hyperplasia CYP21A2 gene Steroid 21-Hydroxylase deficiency Molecular modeling p.H365N p.F306V

1. Introduction Congenital Adrenal Hyperplasia (CAH) is an autosomal recessive inherited metabolic disorder of the adrenal steroid metabolism, due to mutations in the genes involved in the steroidogenic pathway [1,2]. It is characterized by deficiency of one of the five steroidogenic enzymes viz., 21-Hydroxylase (OMIM: 201910), 11bHydroxylase (OMIM: 202010), 3b-Hydroxysteroid dehydrogenase (OMIM: 201810), 17a-Hydroxylase (OMIM: 202110) and 20, 22Desmolase (OMIM: 201710). CAH due to 21-Hydroxylase deficiency is the common form which accounts for more than 90% of CAH cases [3]. CAH has wide range of clinical manifestations viz., severe or the classical form to

Abbreviations: CAH, Congenital adrenal hyperplasia; CYP21A2, 21-Hydroxylase gene; NC, non classical; COS, CV-1 (simian) in Origin, carrying the SV40 genetic material. * Corresponding author. Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh, 160012, India. E-mail address: [email protected] (R. Prasad).

the mild or the non classical (NC) form [4]. In most of the populations, the frequency of classical CAH is approximately 1 in 15,000 live births whereas NC CAH is estimated to be more common than the classical CAH, with prevalence of 1 in 1000 in the white population, and more frequent in certain ethnic groups such as Jews, Hispanic and Yugoslavs [5]. 21-Hydroxylase enzyme is encoded by the CYP21A2 gene which is located on the chromosome 6p21.3 adjacent to its pseudogene, CYP21A1P alternating with C4B and C4A genes encoding the fourth component of complement [6,7]. The major disease-causing mutations in the CYP21A2 are transferred from the CYP21A1P due to unequal crossing over during meiosis or apparent gene conversion events [8], macro or microconversion events [9]. More rarely, certain mutations arise independently to that of pseudo gene and these unique mutations are often found in the specific families [10]. The novel mutations were detected at the frequency of 3%e5%, when large cohorts were investigated [11]. The degree of the disease presentation depends on the underlying set of combination of mutations, therefore in vitro functional studies for each mutation need to be appraised before categorizing any

https://doi.org/10.1016/j.biochi.2018.04.012 0300-9084/© 2018 Elsevier B.V. and Société Française de Biochimie et Biologie Moléculaire (SFBBM). All rights reserved.

116

R. Khajuria et al. / Biochimie 149 (2018) 115e121

mutation in CAH to be disease-causing. These rare/familial/orphan mutations cannot provide sufficient information about the phenotype of the patient, consequently prognosis and genetic counseling is particularly difficult when fetuses or patient harbor these mutations. In the present study, we have functionally characterized two point mutations viz., p. F306V and p. H365N as well as a double mutant (p.F306V, F306 þ T). A cell culture based system was followed to investigate the downstream 21-Hydroxylase enzyme impairment due to mutations. Moreover, in silico analysis acted as a complement to in vitro characterization of the mutants. In conclusion, the functional characterization elucidated good agreement between the degree of enzymatic impairment investigated in vitro and in silico with the clinical disease presentation. 2. Materials and methods 2.1. Patients A total of 55 Indian CAH patients were subjected to molecular screening of CYP21A2 gene after general clinical and endocrine evaluation over the period of 5 years. Fifty-five healthy controls were also included in the present study. The controls and patients were from same ethnic origin. The study was approved by the institutional ethics committee with reference number e 8822/PG/ 10/TRG/11135. Information sheet was provided to all patients/ family members of the patients and written informed consent was taken from each patient. A deep review of the records and genotype analysis was carried out in the patients harboring the novel mutations as mentioned in Table 1. 2.2. Molecular analysis of CYP21A2 gene Genomic DNA was isolated from peripheral blood leukocytes of the affected subjects using protocol described by Daly et al. [12]. All the exons-intron boundaries were selectively amplified by PCR as described by Lee et al. [13]. Isolated DNA was subjected to the quantitative and the qualitative analysis, which was then stored at 20  C till further use. The PCR products were subjected to single stranded conformation polymorphism (SSCP) analysis for detection of novel mutations as reported earlier by Sharma et al. [14]. Simultaneously, amplified exons from healthy control DNA samples were also subjected to SSCP in order to compare it to patient's

samples exhibiting a relative shift. Samples screened from SSCP were subjected to automated DNA sequencing, using an ABI Prism Big Dye Terminator Sequencing Ready Reaction Kit (Perkin Elmer, USA) and a DNA sequencer ABI Prism (Model 3100, Perkin Elmer). All through the manuscript, we refer to the CYP21A2 reference sequence: GenBank M13936.1 and the DNA mutation numbering is based on the cDNA sequence. 2.3. Site-directed mutagenesis Human full length CYP21A2 cDNA cloned into the pCMV4 vector was kindly provided by Dr Rita Menassa [15]. Mutations viz., p. F306V, p. H365N and p. F306V, F306 þ T were introduced using the quik-Change II XL site directed mutagenesis kit (Stratagene, La Jolla, CA). Nucleotide exchanges and insertion were introduced individually by PCR amplification using the sense and anti sense primers (primer sequence available on request). Finally complete CYP21A2 cDNA was sequenced to verify the correct mutation incorporation and to exclude additional sequence aberrations. Empty pCMV4 plasmid was used as negative control while pCMV4-CYP21A2 with wildtype cDNA was considered as positive control. 2.4. Expression of CYP21A2 in vitro COS7 cells were transiently transfected using lipofectamine (Invitrogen, Karlsruhe, Germany) with mutants, negative and positive control constructs. According to manufacturer's instruction, 3  105 cells were plated and were co-transfected per well with 1 mg pCMV4 constructs and 1 mg b-galactosidase pSV-gal vector (Promega, Madison, USA) serving as transfection control. Cells recovery and protein expression was achieved by incubating cells under standard conditions (37  C, 5%CO2) for 48 h [15]. 2.5. Assay of enzyme activity and kinetic parameters 21-Hydroxylase enzyme activity and apparent kinetic parameters were measured using the method as described by Taboas et al., [5]. The permission for import of radioactive substrates was granted by Atomic energy regulatory board (AERB) with reference numberAERB/RSD/Res-NOC/HN-036/2014/2325. Each experiment was repeated thrice. Apparent kinetic parameters were calculated using Graph Pad Prism software (version 5, Graphpad software Inc. San Diego, CA).

Table 1 Genotype and phenotype of the patients harboring the novel mutations. Genotype Phenotype Age of Age/ onset Sex (years) (years)

Age of diagnosis (years)

Presenting complaint

Undue gain in weight

Undue gain in height

Consan- Origin 17OHP (ng/ml) guinity

17/F

F306V/ V281L

NC

8

16

Primary amenorrhea, Hirsuitism, straie, knuckle pigmentation, darkening of face and acne

No

Yes

No

34/M

F306V/ NC F306 þ T

e

34

Primary infertility, short stature

No

Yes

No

24/F

H365N/ P267L

NC

19

24

Excessive hair growth- over face and upper chin

No

No

No

19/F

H365N/ V281L

NC

17

19

Hirsuitism, PCOD

Yes

No

No

Medication

Indian 89.04 1 Dexona: (stimulated after 0.25 mg 250 mg ACTH) 2 Aldactone: 50 mg 3 Novelon: d5 e d25 (2 months) Indian 170 (stimulated 1 Dexona: after 250 mg 0.25 mg ACTH) Indian 40.62 1 Aldactone: (stimulated after 50 mg 250 mg ACTH) 2 Finarteide: 5 mg Indian 44.34 1 Dexona: (stimulated after 0.25 mg 250 mg ACTH)

R. Khajuria et al. / Biochimie 149 (2018) 115e121

2.6. Assay of activation energy (Ea) Different mutant constructs and wild-type construct were transfected in COS7 cells and expressed for 48 h which were then, incubated with 1 ml DMEM media supplemented with 10% FBS containing 0.5 mCi H3 progesterone or H3 17OHP with 2 mM unlabelled steroid substrates (17OHP or Progesterone) at different temperatures viz., 10  C, 20  C, 40  C and 50  C. Post-incubational treatment was performed as described by Taboas et al., [5]. Activation energy of mutants and wild-type were calculated using Arrhenius equation in cal/mol and plotted in Arrhenius plot. 2.7. Western blot analysis Western blot assay was performed using an antihuman-CYP21 rabbit polyclonal antiserum, kindly provided by Dr WL Miller in a standard protocol to determine the expression of CYP21A2 wildtype and mutant proteins. Protein levels from the CYP21A2 constructs were expressed as ratio between CYP21A2 and b-actin [15]. 2.8. Molecular modeling 2.8.1. Multiple sequence alignment The structurally conserved regions in several P450 cytochrome sequences were identified by multiple sequence alignment (MSA). Homologous CYP21A2 sequences were found during BLAST search in UNIPROT database (www.uniprot.org) and aligned in CLUSTAL W (http://www.ebi.ac.uk/clustwalw/) as described previously [15].

117

substitution of histidine (H) encoded by CAC to asparagine (N) encoded by AAC at codon 365, p. H365N [18]. None of the control alleles exhibited the presence of this point mutation. The accession number assigned to p. H365N mutation by NCBI GenBank database was KF447378 [20].

3.2. Functional characterization of novel mutations To elucidate the effect of the point mutations on 21-Hydroxylase activity, the mutations viz., F306V, H365N and the double mutant (F306V, F306 þ T) were introduced into pCMV4-CYP21A2 WT cDNA construct. F306 þ T/L307frameshift mutation is a known frameshift mutation which leads to severe phenotype of the disease. In one of the NC patient, it was present in combination with p. F306V mutation. Therefore, the double mutant was constructed, in order to determine the effect of severe mutation (F306 þ T) in combination with p. F306V. Each mutants ability to convert the two natural substrates i.e. 17OHP or progesterone to their products 11-deoxycortisol or 11deoxycorticosterone respectively was assayed relative to wildtype activity which was arbitrarily taken as 100%. As illustrated in Fig. 1, the p. F306V mutant construct demonstrated reduced activity of the enzyme to 63.23% ± 5.5 (SD) for the conversion of 17OHP to 11-deoxycortisol and to 64.17% ± 7.98 (SD) for the conversion of progesterone to 11-deoxycorticosterone at 2 mM concentration of the respective steroid. The p. H365N mutant construct illustrated reduced enzyme activity of 46.13% ± 4.8 (SD) for the conversion of

2.8.2. Homology modeling The template model for generation of human CYP21A2 was available at the Protein Data Bank with ID 2GEG [16]. The localization of helices, strands and coils were marked in 3-D structure of the protein. The net effect of the novel mutations were studied in silico with the help of Deep View program-Swiss PDB Viewer (http://www.expasy.org/spdbv) [17]. 2.9. Statistical analysis Different mutant construct groups were compared by one way ANOVA followed by post hoc test. All the statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS), GraphPad Prism version 5.0 for Windows (GraphPad Software, San Diego, CA, USA). 3. Results 3.1. Molecular characterization of mutations The amplified PCR product of exon 7 of the CYP21A2 gene showed the deletion of single stranded DNA band in patient sample as compared to control upon SSCP [18]. Subsequently upon DNA sequencing, a c916 T > G transversion was exhibited resulting in p. F306V, missense mutation. This mutation led to substitution of phenylalanine (F) to Valine (V) at codon 306 with codon change of TTT to GTT, p. F306V [18]. Two NC patients were identified to harbor heterozygous p. F306V mutation. Additionally, p. F306V mutation was not detected in the control alleles, so it is a rare point mutation that arises independently of the pseudogene. Moreover, F306V mutation had been described previously [19,20]. While screening the exon 8 of the CYP21A2 gene by SSCP, we found the deletion of single stranded DNA band in patient sample as compared to control upon SSCP in two NC CAH patients [18]. DNA sequencing revealed the transversion of (c1095C > A), nucleotide numbering according to White et al. (1986) leading to the

Fig. 1. Residual enzymatic activity of different mutants. Activity is expressed as percentage of the wild-type activity, which is arbitrarily defined as 100%. Conversion values are shown for the two natural substrates (A) 17OHP to 11-deoxycortisol, (B) Progesterone to 11-deoxycorticosterone at the substrate concentration of 2 mmol/l of unlabelled steroid. The bars represent the Mean ± SD of three independent experiments.

118

R. Khajuria et al. / Biochimie 149 (2018) 115e121

17OHP to 11-deoxycortisol and 57.77% ± 3.69 (SD) for the conversion of progesterone to 11-deoxycorticosterone (Fig. 1). The double mutant (p.F306V, F306 þ T) exhibited reduced activity of 42.93% ± 5.44 (SD) for the conversion of 17OHP to 11-deoxycortisol and to 52.83% ± 7.05 (SD) for the conversion of progesterone to 11deoxycorticosterone (Fig. 1). The relative enzymatic activity of mutants was significantly reduced in comparison to wild-type. Further, the kinetic parameters viz., maximum velocity (Vmax) and Michaelis -Menten constant (Km) were investigated which indicated that they followed Michaelis -Menten kinetics (Fig. 2). The maximal velocity was decreased in all the mutants as compared to wild-type towards the natural substrates, 17OHP and progesterone (Fig. 2). On the other hand, affinity constant was increased for all the variants indicating impaired substrate binding (Fig. 2). The temperature dependence of wild-type and mutant enzymes when 17OHP and progesterone act as substrates was expressed as Arrhenius plot (Log V versus 1/K *103) (Fig. 3). There was alteration in activation energy of mutants to that of wild-type, indicating alteration in physico-chemical state of the enzyme. In accordance with these findings, immunoblot for detection of protein levels in wild-type and mutant constructs clearly revealed lower CYP21A2 protein expression in comparison to the wild-type (Fig. 4).

4. Discussion 21-Hydroxylase is an adreno-cortical enzyme which has key role in biosynthesis of steroids. Clearly, mutations in the CYP21A2 gene are responsible for the 21-Hydroxylase deficiency. The functional characterization of new mutations in CYP21A2 gene had been performed earlier which show good agreement between the in vitro enzymatic activity and the degree of clinical disease phenotype presentation [21]. The present study investigated the potential detrimental functional consequences of the mutations characterized in the CYP21A2 gene from Indian patients with NC phenotype of CAH. The findings of the in vitro experiments were validated by the computational analysis via bioinformatics which provided valuable insights regarding the correlation between genotype and

phenotype in CAH. All together, the data provide the information about the mutation severity and its structure-function relationship. To put it into perspective, in vitro characterization with clear and extensive data is indispensable for good clinical counseling. In this study, p. F306V (c916T > G) and p. H365N (c1095C > A) were found in two patients with NC CAH (Table 1). Notably, it has been reported that the activity of 21-Hydroxylase enzyme in NC form is characterized by >20% enzyme activity whereas >95% enzyme activity is similar to wild-type protein [21]. Also, depending on the residual enzymatic activity, mutations are classified in to four groups (0, A, B and C) [20,22]. The group C includes mutations with the residual activity of 20e50% resulting in NC CAH. Other NC CAH mutations (p. K121Q, p. P453S, p. P482S) have been functionally characterized [23,24] and demonstrated enzymatic activity in range of 36e72%. Therefore, the in vitro relative enzymatic activity exhibited by p. F306V and p. H365N curtailed to the NC phenotype which is in accordance with the phenotype as assigned by clinicians. In this study, the double mutant construct (F306V, F306 þ T) exhibited activity of 42.93% and 52.83% for 17OHP and progesterone respectively. Even though the genotype revealed presence of severe F306 þ T mutation, the double mutant indicated NC phenotype. This suggests that the milder mutation, p. F306V determined the phenotype of the patient [25,26]. Furthermore, the findings of association of reduced relative enzyme activity of mutant constructs to that of kinetic parameters viz., maximal velocity and any alteration in tenacity of binding of substrate to enzyme indicate that there was reduction in maximal velocity toward both the substrates (17OHP and progesterone) in mutant constructs to that of wild-type construct (Fig. 3). It is noteworthy here that the Michaelis-Menten constant (Km) was increased in mutants enzyme in comparison to wild-type enzyme. The increased Michaelis-Menten constant (Km) in NC CAH has also been observed previously [22]. Indeed, decreased level of 21Hydroxylase enzyme as depicted by the western blotting fraternized with the reduced maximal velocity of the enzyme expressed by the mutants (Fig. 4). These findings are suggestive of either an impaired synthesis or an augmented degradation of 21Hydroxylase protein. The p. R483P exhibited diminished protein

Fig. 2. AeB Kinetic analysis of wild-type and mutant vectors F306V, H365N, F306V, F306 þ T. A-, Linear lineweaver-Burk plots of enzymatic activity between 1/V (velocity) against 1/S (substrate concentration) for conversion of 17OHP to 11-deoxycortisol and B-, Linear lineweaver-Burk plots of enzymatic activity between 1/V (velocity) against 1/S (substrate concentration) for conversion of progesterone to 11-deoxycorticosterone. Inset to A and B is the non linear plots of velocity (V) (picomoles/milligram total protein/per minute) against substrate concentration (S) (micromoles per liter).

R. Khajuria et al. / Biochimie 149 (2018) 115e121

119

Fig. 3. Arrhenius plot (log V (pmol/mg/min) & 1/T(K) *103) for determination of activation energy for WT, p. H365N mutant, p. F306V mutant & double mutant for 17OHP and progesterone as substrates for 21- Hydroxylase enzyme. Activation energy was calculated using Arrhenius equation. There was no change in the transition temperature of the wild type and mutant enzyme.

Fig. 4. Expression analysis of different mutant. Representative western blot showing expression levels of the Mock, wild-type CYP21A2 and the different mutant construct's assayed in COS7 cells collected after harvesting cells with lysis buffer directly from the well. Corresponding panel shows the b-actin expression in the constructs.

stability and half-life, p. D322G demonstrated reduction in the amount of protein which indicated reduced stability of protein [5]. Moreover, the activation energy required by mutants 21Hydroxylase enzyme was higher than the wild-type enzyme. The reasonable explanation is that the enzyme expressed by mutants is unstable due to which the protein expression is low. Taken together the outcome of the in vitro functional characterization indicates that the mutant enzyme's are inactive as well as less stable than the wild-type enzyme. A glance at the in silico computational approach and severity prediction by softwares viz., SIFT, PROVEAN, PolyPhen predicted p. F306V and p. H365N to be damaging to the protein [20]. Numerous studies have delineated the conservation of the amino acid in different cytochromes residues [3,15]. Of note, Multiple sequence alignment (MSA) revealed that the wild-type residue in p. F306V was not conserved at this position in homologous sequences (Fig. 5). Phe306 is present in I-Helix of the protein [21]. The helix L on the proximal side of the protein and helix I on the distal side of the protein, sandwiches the heme-binding site [26]. As

illustrated by the homology modeling by the 3D of the structure of CYP21A2 protein (Fig. 6a), the substitution of aromatic amino acid (phenylalanine) with the non-aromatic amino acid (valine) causes interruption in p-p stacking interaction of the interacting partners of the phe306 in the protein. Moreover, HOPE program illustrated that wild-type and mutant amino acids differ in size [18]. Hence, p. F306V proves to be damaging to the protein. Furthermore, specific size, charge, and hydrophobicity values are different for each amino acid and in case of the p. H365N, wildtype residue (histidine) and mutant residue (asparagine) often differ in these properties. Based on MSA, the wildtype residue was conserved so we strictly suggest that the mutation is probably damaging to the protein (Fig. 5). Histidine and asparagine are both polar amino acids but lack of imidazole ring in asparagine modifies the structure of the protein and enzyme activity [3]. 3-D computational analysis by Swiss-PDB viewer revealed that the wild-type residue has interactions with heme as ligand (Fig. 6b). The p. H365N introduces a smaller residue in the protein which was unable to make multimer contacts and form bond with the ligand, heme. Protein structure prediction algorithm, HOPE program also described that wild-type and mutant amino acids actually differ in size [18]. Therefore, in conclusion H365N mutation proves to be damaging to the protein as shown by the bioinformatic tools. In conclusion, present study confirms that the CYP21A2 mutations viz., p. F306V and p. H365N result in NC phenotype. In vitro characterization by analyzing relative 21-Hydroxylase activity, its kinetic parameters, activation energy requirement and 21Hydroxylase protein expression revealed the functional consequences of the novel/rare mutations. The findings were further validated by the computational analysis by bioinformatics which provided valuable insights regarding the correlation of the genotype and the phenotype in CAH. Conceivably, the knowledge about the functional consequences of the mutations is the basis for improved genetic counseling with respect to prognosis and therapeutic implications.

120

R. Khajuria et al. / Biochimie 149 (2018) 115e121

Fig. 5. Multiple sequence alignment of human CYP21A2 with other mammalian CYP21A2s, mammalian crystallized CYPs or human steroidogenic cytochromes. Sequence alignment was performed with CLUSTAL W. Yellow boxes represent conserved amino acid residues. Mutant F306V residue is indicated by red box and arrow head above. Phenylalanine residue is present in I-helix of 21-hydroxylase enzyme. Mutant H365N residue is indicated by red box and arrow head above. Histidine is conserved amino acid residue in this region.

and Engineering Research Board, New Delhi, India for funding the project (Sanction number: SR/SO/HS/0045/2010) and Indian council of medical research for ICMR-JRF and SRF (Reference number: 3/1/3/JRF-2011/HRD-3). We thank Atomic Energy Regulatory Board, India for granting permission for import of radioactive substrates (AERB/RSD/Res-NOC/HN-036/2014/2325).We are also thankful to all the patients included in this study. References

Fig. 6. Ribbon diagram of CYP21A2 a. (F306V) b. (H365N). a. Phe306 is present in the hydrophobic cavity of the protein. Localization of Phe306 in the region of the protein where it interacts with other amino acid residues viz. Trp302 and Phe404 by forming p-electron stacking interaction. Mutation at Phe306 to nonaromatic Val306 interrupts this interaction. b. Localization of His365 in the heme-interacting region of protein but upon substitution with asparagine it leads to Asn365 which interrupts the interaction with heme disturbing the protein function.

Acknowledgements We are grateful to Dr Rita Menassa (France) for providing pCMV4-CYP21A2 WT cDNA vector and to Dr WL Miller (University of San Francisco, CA) for providing the antihuman CYP21A2 rabbit polyclonal antiserum. We express our sincere gratitude to Science

[1] T. Huynh, I. McGown, D. Cowley, et al., The clinical and biochemical spectrum of congenital adrenal hyperplasia secondary to 21- hydroxylase deficiency, Clin. Biochem. Rev. 30 (2009) 75e86. [2] T. Sahakitrungruang, Clinical and molecular review of atypical congenital adrenal hyperplasia, Annals of Pediatric Endocrinology & Metabolism 20 (2015) 1e7. [3] P. Concolino, E. Mello, M.P. Patrosso, et al., p.H282N and p.Y191H: 2 novel CYP21A2 mutations in Italian congenital adrenal hyperplasia patients, Metab., Clin. Exp. 61 (2012) 519e524. [4] M. Barbaro, F.C. Soardi, L.J. Ostberg, et al., In vitro functional studies of rare CYP21A2 mutations and establishment of an activity gradient for nonclassic mutations improve phenotype predictions in congenital adrenal hyperplasia, Clin. Endocrinol. 82 (2015) 37e44. [5] M. Taboas, L.G. Acuna, M.F. Scaia, et al., Functional Studies of p.R132C, p.R149C, p.M283V, p.E431K, and a Novel c.652-2A.G Mutations of the CYP21A2 Gene, PLoS One 9 (3) (2014), e92181, https://doi.org/10.1371/ journal.pone.0092181. [6] A. Khattab, T. Yuen, S. Al-Malki, et al., A rare CYP21A2 mutation in a congenital adrenal hyperplasia kindred displaying genotype-phenotype nonconcordance, Ann. N. Y. Acad. Sci. 1 (2016) 5e10, https://doi.org/10.1111/ nyas.12864. [7] Y. Higashi, H. Yoshioka, M. Yamane, et al., Complete nucleotide sequence of two steroid 21-hydroxylase genes tandemly arranged in human chromosome: a pseudogene and a genuine gene, Proc. Natl. Acad. Sci. Unit. States Am. 83 (1986) 2841e2845. [8] K. Urabe, A. Kimura, F. Harada, et al., Gene conversion in steroid 21hydroxylase genes, Am. J. Hum. Genet. 46 (1990) 1178e1186. [9] P. Concolino, F. Vendittelli, E. Mello, et al., Two novel CYP21A2 missense mutations in Italian patients with 21-hydroxylase deficiency: identification and functional characterisation, IUBMB Life 6 (2009) 229e235.

R. Khajuria et al. / Biochimie 149 (2018) 115e121 [10] M. Barbaro, L. Baldazzi, A. Balsamo, et al., Functional studies of two novel and two rare mutations in the 21-hydroxylase gene, J. Mol. Med. 84 (2006) 521e528. €tzinger, et al., Functional characterization of two [11] N. Krone, F.G. Riepe, J. Gro novel point mutations in the CYP21 gene causing simple virilizing forms of congenital adrenal hyperplasia due to 21-hydroxylase deficiency, J. Clin. Endocrinol. Metab. 90 (2005) 445e454. [12] A.K. Daly, V.M. Steen, K.S. Fairbrother, et al., CYP2D6 multiallelism, in: E.F. Johnson, Wateman (Eds.), Methods in Enzymology, 1996, pp. 199e210. [13] H.H. Lee, H.T. Chao, Y.J. Lee, et al., Identification of four novel mutations in the CYP21 gene in congenital adrenal hyperplasia in the Chinese, Hum. Genet. 103 (1998) 304e310. [14] H. Sharma, R.S. Mavuduru, S.K. Singh, et al., Heterogenous spectrum of mutations in CFTR gene from Indian patients with congenital absence of the vas deferens and their association with cystic fibrosis genetic modifiers, Mol. Hum. Reprod. 20 (9) (2014) 827e835, https://doi.org/10.1093/molehr/ gau047. [15] R. Menassa, V. Tardy, F. Despert, et al., p.H62L, a rare mutation of the CYP21 gene identified in two forms of 21- hydroxylase deficiency, J. Clin. Endocrinol. Metab. 93 (2008) 1901e1908. [16] H.M. Berman, J. Westbrook, Z. Feng, et al., The protein Data Bank, Nucleic Acids Res. 28 (2000) 235e242. [17] N. Guex, M.C. Peitsch, SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling, Electrophoresis 18 (1997) 2714e2723. [18] R. Khajuria, R. Walia, A. Bhansali, et al., Data on the Functional Consequences of the Mutations Identified in 21-Hydroxylase Deficient CAH Patients, Data in

121

Brief, 2017 (in press). [19] S. Haider, B. Islam, V. D'Atri, et al., Structuree phenotype correlations of human CYP21A2 mutations in congenital adrenal hyperplasia, Proc. Natl. Acad. Sci. Unit. States Am. 110 (2013) 2605e2610. [20] R. Khajuria, R. Walia, A. Bhansali, et al., The spectrum of CYP21A2 mutations in congenital adrenal hyperplasia in an indian cohort, Clin. Chim. Acta 464 (2017) 189e194. [21] S. Lajic, T. Robins, N. Krone, et al., CYP21 mutations in simple virilizing congenital adrenal hyperplasia, J. Clin. Endocrinol. Metab. 79 (2001) 581e586. [22] P.W. Speiser, J. Dupont, D. Zhu, et al., Disease expression and molecular genotype in congenital adrenal hyperplasia due to 21- hydroxylase deficiency, J. Clin. Invest. 22 (1992) 584e595. [23] F.G. Riepe, O. Hiort, J. Grotzinger, et al., Functional and structural consequences of a novel point mutation in the CYP21A2 gene causing congenital adrenal hyperplasia: potential relevance of helix C for P450 oxidoreductase21-hydroxylase interaction, J. Clin. Endocrinol. Metab. 93 (2008) 2891e2895. [24] M. Barbaro, S. Lajic, L. Baldazzi, et al., Functional analysis of two recurrent amino acid substitutions in the CYP21 gene from Italian patients with congenital adrenal hyperplasia, J. Clin. Endocrinol. Metab. 89 (2004) 2402e2407. [25] P.C. White, P.W. Speiser, Congenital adrenal hyperplasia due to 21- hydroxylase deficiency, Endocr. Rev. 21 (2000) 245e291. [26] T. Robins, J. Carlsson, M. Sunnerhagen, et al., Molecular model of human CYP21 based on mammalian CYP2C5: structural features correlate with clinical severity of mutations causing congenital adrenal hyperplasia, Mol. Endocrinol. 20 (2006) 2946e2964.