Functional Characterization of Two Novel Point Mutations in the ...

0021-972X/05/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 90(1):445– 454 Copyright © 2005 by The Endocrine Society doi: 10.1210/jc.2004-0813

Functional Characterization of Two Novel Point Mutations in the CYP21 Gene Causing Simple Virilizing Forms of Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency Nils Krone, Felix G. Riepe, Joachim Gro¨tzinger, Carl-Joachim Partsch, and Wolfgang G. Sippell Division of Pediatric Endocrinology (N.K., F.G.R., C.-J.P., W.G.S.), Department of Pediatrics, Christian-AlbrechtsUniversita¨t zu Kiel, Universita¨tsklinikum Schleswig-Holstein (Campus Kiel), D-24105 Kiel, Germany; and Biochemisches Institut (J.G.), Christian-Albrechts-Universita¨t zu Kiel, D-24098 Kiel, Germany Congenital adrenal hyperplasia is a group of autosomal recessive disorders most often caused by deficiency of steroid 21-hydroxylase due to mutations in the CYP21 gene. We studied the functional and structural consequences of two novel missense mutations in the CYP21 gene, detected in two simple virilizing congenital adrenal hyperplasia patients. Both the male and female patient were compound heterozygous for the novel I77T and A434V point mutations, respectively. The in vitro expression analysis in COS-7 cells revealed a reduced 21-hydroxylase activity in the I77T mutant of 3 ⴞ 2% (SD) for the conversion of 17-hydroxyprogesterone to 11-deoxycortisol and of 5 ⴞ 3% for the conversion of progesterone to 11-deoxycorticosterone. The A434V mutant had a residual enzyme activity of 14 ⴞ 2% for 17-hydroxyprogesterone and 12 ⴞ 6% for

progesterone. Substrate affinity was similar in the mutants as in the CYP21 wild-type protein, whereas reaction velocity was markedly decreased in both mutants. These effects could be readily explained by structural changes induced by the mutations, which were rationalized by a three-dimensionalmodel structure of the CYP21 protein. We hypothesize that the I77T mutation markedly decreases the reaction product release and/or substrate entrance to the enzyme’s active site, whereas the A434V mutant reduces both the catalytic capacity and reaction velocity. Studying the enzyme function in vitro helps to understand the phenotypical expression and disease severity of 21-hydroxylase deficiency and also provides new insights into cytochrome P450 structure-function relationships. (J Clin Endocrinol Metab 90: 445– 454, 2005)

C

reported to occur at a frequency of up to 1 in 1000 to 1 in 500 in various Caucasian populations. Patients are asymptomatic at birth and may manifest in later childhood or adolescence with hirsutism and decreased fertility or precocious pseudopuberty (3). The 21-hydroxylase gene (CYP21) and a nonfunctional pseudogene (CYP21P) are located on the short arm of chromosome 6 (6p21.3). The CYP21 and CYP21P genes consist of 10 exons and show a high homology, with a nucleotide identity of 98% in their exon and 96% in their intron sequences (4 – 6). Due to the genomic localization in a region with a high frequency of genomic recombinations, the most frequent 21OHD-causing mutations are generated by apparent gene conversion events (7, 8). Complete gene deletions, large gene conversions, single-point mutations and an 8-bp deletion have been described (1, 9, 10). Commonly the eight most frequent CYP21-inactivating point mutations and the 8-bp deletion in exon 3 are transferred by microconversions from the CYP21P to the CYP21 gene (7, 8). To date 75 different mutations are listed in the Human Gene Mutation Database (11), 46 of which are nucleotide substitutions (nonsense or missense mutations). The vast majority of these mutations has been identified in single families or small populations. These novel mutations were detected in a frequency of about 3–5% when large cohorts were investigated, and it is estimated that about 1% of CYP21-inactivating mutations arise de novo (12–14). About 65–75% of the CAH patients are compound het-

ONGENITAL ADRENAL HYPERPLASIA (CAH) is one of the most frequent inborn errors of metabolism, inherited in an autosomal recessive trait. It is caused by the loss or severe decrease in activity of one of the five steroidogenic enzymes involved in cortisol biosynthesis. Steroid 21-hydroxylase deficiency (21OHD) is the result of mutations in the 21-hydroxylase gene (CYP21), and accounts for 90 – 95% of all cases. In addition to decreased cortisol production, aldosterone biosynthesis may be impaired. The disease can be divided into classical and nonclassical forms. In most populations classic CAH occurs in about 1 in 7,000 to 1 in 15,000 live births and presents clinically as the simple virilizing or salt-wasting form. Phenotypical expression of CAH is highly variable (1–3). The simple virilizing form is characterized by virilization of the external genitalia in newborn females and by hypocortisolism and precocious pseudopuberty due to reactive androgen overproduction in both sexes. In addition, the salt-wasting form presents with severe renal salt loss as a consequence of aldosterone deficiency. The milder nonclassical or late-onset form is

First Published Online October 13, 2004 Abbreviations: BA, Bone age; CA, chronological age; CAH, congenital adrenal hyperplasia; G, genital stage; HC, hydrocortisone; KM, Michaelis constant; 21OHD, 21-hydroxylase deficiency; 17OHP, 17-hydroxyprogesterone; PH, pubic hair stage; SDS, sd score; Vmax, maximal velocity. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

445

446

J Clin Endocrinol Metab, January 2005, 90(1):445– 454

erozygous for disease-causing mutations. It is well established that the clinical phenotype of CAH correlates with the less severely mutated allele and, consequently, with the residual activity of 21-hydroxylase. Several studies have addressed the correlation between the CYP21 genotype and clinical phenotype (12–18). Although divergence between genotype and phenotype occurs (19), this correlation appears to be rather high and can have implications for the clinical course of the patient because of the possibility of estimating the risk of disease severity when the residual 21-hydroxylase activity has been ascertained by in vitro analysis. In this study, we describe two CAH patients, in both of whom novel point mutations of the CYP21 gene were detected. The resulting CYP21 mutants were analyzed by in vitro expression experiments for their residual enzymatic activity, which enabled us to classify these novel mutations within the CYP21 gene mutation groups (13, 14, 20). The in vitro measured alterations of the 21-hydroxylase mutants’ enzymatic function could be rationalized with the help of a three-dimensional molecular model, which indicated a putative structural change in the protein structure and provided further insights into the function of cytochrome P450 enzymes.

Krone et al. • Functional Analysis of CYP21 Point Mutations

Patients and Methods Patients, clinical presentation, and hormonal analyses Case OK. The male patient presented at the age of 6 yr with the clinical symptoms of precocious pseudopuberty. Physical examination revealed pubic hair stage (PH) 3, genital stage (G) 3, and testicular volumes of 2 ml. His height was 135.0 cm [⫹2.8 sd score (SDS)] with a markedly accelerated bone age of 13 yr (Fig. 1A). Plasma testosterone was increased to a pubertal level of 4.15 nmol/liter (normal range for age 0.07– 0.40 nmol/liter) and androstenedione of 16.8 nmol/liter (normal range 0.10 –1.54 nmol/liter). A GnRH test showed prepubertal or decreased gonadotropin responses (LH ⬍ 0.5 3 2,5 IU/ml, FSH ⬍ 0.5 3 1,3 IU/ml), indicating gonadotropin-independent (peripheral) precocious puberty. Simple virilizing CAH was diagnosed by increased plasma 17-hydroxyprogesterone (24.2 nmol/liter; normal range 0.27– 2.33 nmol/liter) and normal plasma renin activity. After 3 yr of adequate hydrocortisone treatment, a decrease of height-SDS for chronological age (CA) and increasing height-SDS for bone age (BA), the patient showed progressing pubertal development with PH3, genital G3, and testes of 6 ml. A second GnRH test (LH 0.5 3 10.7 IU/liter, FSH 1.6 3 2.7 IU/liter; LHstim/FSHstim ⫽ 4) led to the diagnosis of secondary central precocious puberty. Plasma testosterone was elevated (1.32 nmol/liter), whereas androstenedione (1.78 nmol/liter) and 17-hydroxyprogesterone (17OHP) (0.39 nmol/liter) were normal. Treatment with the depot GnRH agonist Leuprorelin resulted in complete hormonal suppression (LH ⬍ 0.5 3 ⬍ 0.5 IU/ml, FSH ⬍ 0.5 3 ⬍ 0.5 IU/ml) and arrest of pubertal development (PH3, G3, testes 4 ml). Patient’s actual height is 161.5 cm (⫹2.46 SDSCA for CA 10.9 yr) with a BA of 13.75 yr (⫺0.23 SDSBA).

FIG. 1. Growth charts of the two CAH patients. TH, Target height (41); BA, f. Height is given as dots (F). The inset shows the contrasting of the development of height-SDSCA and height-SDSBA. PAH (⽧), Predicted adult height (42). A, Follow-up of the male patient. Dx, Time of diagnosis at the age of 6 yr. At this time, HC treatment was initiated. Three years later the patient developed secondary central precocious puberty (sCPP), which was treated with a GnRH agonist (GnRHa). Note that a marked increase of height SDSBA and PAH occurred during HC treatment. B, The female patient was diagnosed at the age 4.6 yr (Dx), and HC treatment was started. For the first 1.5 yr, treatment was insufficient to normalize growth and bone maturation. Thus, height-SDSBA and PAH showed an initial decrease. After increasing the HC daily dose to 17.5 mg/m2, bone maturation slowed down and PAH stabilized.


Case MM. The female patient was first seen by a pediatric endocrinologist at the age of 4.3 yr because of premature pubarche with PH3. She had an uneventful history. She was born in the 41st wk of gestation, with a birth weight of 3420 g and length of 52 cm. Clinical examination showed only mild clitoromegaly but no labial fusion, despite the accelerated bone age of 7 yr. Her height was 108.6 cm (⫹0.5 SDS) and weight 20.1 kg (body mass index 17 kg/m2, ⫹0.9 SDS) (Fig. 1B). Hormonal determinations revealed elevated 17OHP (11.29 nmol/liter; normal range 0.39 –2.90 nmol/liter), testosterone (3.16 nmol/liter; normal range 0.07– 0.29 nmol/liter), androstenedione (8.34 nmol/liter; normal range 0.14 –1.64 nmol/liter), and plasma dehydroepiandrosterone sulfate (4812 nmol/liter; normal range 35.4 –167.56 nmol/liter). The 17OHP to deoxycorticosterone ratio of 5850 after ACTH stimulation was highly elevated, as evidence for a homozygous form of 21OHD (21). All other parameters, especially plasma renin activity and aldosterone, were normal. At age 4.6 yr, the patient was diagnosed as mild simple virilizing CAH and put on hydrocortisone (HC) in a relatively low daily dose of 10 mg/m2. Despite this, urinary pregnanetriol and pregnanetriolone remained elevated, and growth acceleration with an increasing heightSDSCA and decreasing height-SDSBA persisted. HC dose had to be increased to 17.5 mg/m2 until hormonal parameters normalized and BA acceleration slowed down. At present she is PH3, breast stage B1, height 127.8 cm (⫹0.7 SDS), and weight 35.9 kg (body mass index 22 kg/m2, ⫹2.49 SDS).

Mutation analysis Screening for the most frequent CYP21 inactivating mutations was performed by multiplex-minisequencing (22) and showed none of the 10 common mutations. This was followed by an extensive mutation analysis, which was performed by Southern blot analysis and direct DNA sequencing, as previously described (23). Confirmation of the novel point mutations was performed in a singleplex minisequencing reaction with specific primers, binding one nucleotide upstream (5⬘) to the position of interest. The singleplex minisequencing reaction was carried out in a total volume of 5 ␮l containing 2.5 ␮l of SNaPshot multiplex ready reaction reagent (Applied Biosystems, Foster City, CA), 5 pmol of each primer (I77T-F-SPM: 5⬘-GCTGAACTCCAAGAGGACCA, A434VF-SPM: 5⬘-CGTGTGCCTGGGCGAGCCGCTGG), and 1 ␮l purified PCR product. The singleplex minisequencing was performed as follows: 25 cycles at 96 C for 10 sec, 50 C for 5 sec, and 60 C for 30 sec. After extension, the samples were treated with shrimp alkaline phosphatase according to the manufacturer’s protocol. The samples were electrophoresed on an automated ABI PRISM 310 sequencer and analyzed with the ABI GeneScan 3.1 software (22).


447

after a standard protocol (Invitrogen), using DMEM supplemented with glutamine, antibiotics, and 10% fetal calf serum. The activity of 21-hydroxylase in intact COS-7 cells was determined 48 h after transfection. The cells were incubated for 1 h at 37 C with 500 ␮l DMEM medium containing 0.2 ␮Ci 3H-labeled substrate (17OHP or progesterone), 2 ␮mol/liter unlabeled steroid, and 8 mmol/liter nicotinamide adenine dinucleotide phosphate reduced. After incubation, steroids were extracted from the culture medium with isooctane/ethylacetate (1:1, vol/vol), evaporated to dryness, and dissolved in ethanol. The steroids were separated by TLC using chloroform/acetone (70:30, vol/vol). The radioactivity was measured directly from the TLC plates using the Rita Star TLC-scanner (Ray Test, Straubenhardt, Germany) and analyzed using the version 1.97 software. Cells were trypsinized and lysed in reporter lysis buffer (Promega), followed by determination of protein content and measurement of ␤-galactosidase activity according to the manufacturer’s protocol. For the determination of apparent kinetic constants, intact COS-7 cells were incubated for 1 h as described above with 0.5, 1.0, 2.0, 3.0, 4.0, or 6.0 ␮mol/liter unlabeled steroid. Postincubational treatment and analysis were performed as described above. 21-Hydroxylase activity was expressed as a percentage of substrate conversion in picomoles per milligram⫺1 ⫻ min⫺1, taking the activity of cells expressing the CYP21 wild-type protein as 100% percent after correction for total protein and for the activity of cells transfected with the empty pcDNA3 plasmid. The apparent kinetic constants were calculated from the measurements of 21-hydroxylase activity in intact COS-7 cells at each of the different substrate concentrations. Calculation of enzymatic activities and kinetic constants was performed using the Graph Pad Prism software (version 4.0, GraphPad Software Inc., San Diego, CA). Western blot analysis was performed using an antihuman-CYP21 rabbit polyclonal antiserum, kindly provided by Dr. W. L. Miller, in a standard protocol to ensure the expression and translation of the intact CYP21 wild-type and CYP21 mutant proteins.

Immunofluorescence The immunofluorescence was performed using a standard protocol. The same antihuman-CYP21 rabbit polyclonal antiserum used for the Western blot in combination with a mouse anti-KDEL antibody (BIOMOL, Hamburg, Germany), as a marker for the smooth endoplasmatic reticulum, each in a 1:200 dilution was used as primary antibody. The antirabbit-ALEXA594 antibody (Molecular Probes, Leiden, The Netherlands) and the antimouse-fluorescein isothiocyanate antibody (Dianova, Hamburg, Germany) in 1:500 and 1:50 dilutions were used as secondary antibodies.

Construction of plasmids and site-directed mutagenesis

Molecular modeling

The human full-length CYP21 cDNA cloned into the pGEM3Z vector was kindly provided by Dr. W. L. Miller (Department of Pediatrics, University of California, San Francisco). From the pGEM3Z-CYP21 construct, a CYP21-cDNA fragment was cloned into the HindIII/ KpnI site of the pcDNA3 expression vector, resulting in the pcDNA3-CYP21-WT construct. The mutagenesis was performed from the pGEM3Z-CYP21, using the QuikChange XL site-directed mutagenesis kit (Stratagene, Germany) with the following primers: for the I77T mutation, I77T-F 5⬘-CTCCAAGAGGACCAcTGAGGAAGCCATGG and I77T-R 5⬘-CCATGGCTTCCTCAgTGGTCCTCTTGGAG and the A434V mutation, A434V-F 5⬘-GGCGAGCCGCTGGtGCGCCTGGAGCTC, and A434V-R 5⬘-GAGCTCCAGGCGCaCCAGCGGCTCGCC (positions of nucleotide substitutions are given in lower case, boldface letters). The introduction of the I77T and A434V mutations was verified by sequencing the entire construct. The transfer into the pcDNA3 vector was performed as described above for pcDNA3-CYP21WT. The plasmids were named according to the designed mutations pcDNA3-CYP21-I77T and pcDNA3-CYP21-A434V.

The human CYP21 sequence was compared with the corresponding sequences of the mouse, rat, cow, and pig proteins. To substantiate the relatedness of the human CYP21 to the overall fold of the cytochrome P450 family revealed by homology analyses, we used a fold recognition algorithm to show that the human CYP21 sequence is compatible with the architecture of this family (ProHit package, ProCeryon Biosciences GmbH, Salzburg, Austria). The template structure with the highest score of the pair potential was the x-ray structure of the mammalian cytochrome CYP2C5 (PDB accession code 1DT6), which served as the template for the three-dimensional model of CYP21. According to the alignment (Fig. 2) obtained by the fold recognition procedure, amino acid residues were exchanged in the template. Insertions and deletions in CYP21 were modeled using a database search approach included in the software package WHATIF (24). Finally, these model structures were energy minimized using the steepest descent. The structural representations were generated with the Ribbons 2.0 program. All programs were run on a Silicon Graphics Indigo2 workstation (Silicon Graphics GmbH, Grasbrunn, Germany).

In vitro expression and assays of enzyme activity Approximately 1 ⫻ 106 COS-7 cells were transiently transfected using lipofectamine (Invitrogen, Karlsruhe, Germany) with 1 ␮g each of pcDNA3-CYP21 construct and ␤-galactosidase pSV-Gal vector (Promega, Mannheim, Germany). Posttransfection treatment was performed

Results Mutation analysis

Patient OK. Screening for the 10 most frequent CYP21-inactivating mutations by multiplex minisequencing did not re-

448



FIG. 2. Alignment of the CYP21 and CYP2C5 cytochrome P450 enzyme obtained by the fold recognition procedure. This alignment is the basis of the three-dimensional model of CYP21.

veal a mutation. Southern blot analysis revealed a hemizygous gene deletion inherited from the mother. Complete DNA sequencing of the CYP21 gene elucidated a T to C mutation at position bp 327 [nucleotide numbering according to Higashi et al. (6)] in exon 2 of the CYP21 gene in hemizygous state in the patient and in heterozygous state in the father, leading to the substitution of isoleucine to threonine at amino acid position 77 of the CYP21 protein. The results were confirmed by singleplex minisequencing analysis (Fig. 3A). Therefore, the patient was found to be compound heterozygous for a large gene deletion of the CYP21 gene (maternal allele) and the I77T mutation (paternal allele). Patient MM. Southern blot analysis showed that the patient carried two CYP21 genes. The classical known intron 2 splice site mutation (I2 G) and the V281L mutation in exon 7 were detected in the mutation screening procedure, whereas segregation analysis showed that these mutations were present on the maternal allele. Because in this case the second disease-causing defect on the paternal allele was still undetected, complete sequencing of the CYP21 gene was performed, revealing a C to T point mutation at position bp 2522 in exon 10 in heterozygous state in both the female patient and her father. Mutation confirmation was conducted by singleplex minisequencing (Fig. 3B). This point mutation results in the substitution of alanine by valine at amino acid position 434 (A434V). The complete genotype (regarding the CYP21 gene) of the female patient is A434V (paternal allele)/I2 G ⫹ V281L (maternal allele). Functional analysis of enzyme activity and enzyme kinetics

The in vitro expression studies demonstrated that the I77T mutation reduced the 21-hydroxylase activity to 3 ⫾ 2% (sd) for the conversion of 17OHP to 11-deoxycortisol and 5 ⫾ 3% for the conversion of progesterone to 11-deoxycorticosterone. Expression of the CYP21 A434V mutant showed a

reduction of enzymatic activity to 14 ⫾ 2% for 17OHP and 12 ⫾ 6% for progesterone (Fig. 4). Determination of the apparent kinetic constants revealed that both wild-type and mutants achieved saturation under experimental conditions, and the apparent MichaelisMenten constant (KM) values were in the same range for both substrates (Table 1). The I77T mutation showed a remarkably decreased maximum velocity for the synthesis of 11-deoxycorticosterone and 11-deoxycortisol; however, the affinity for both substrates was not markedly changed. The A434V mutation also showed apparent KM values of similar magnitude to the wild-type 21-hydroxylase, whereas the reaction velocity for 17OHP and progesterone was noticeably decreased. The A434V mutant had a higher maximal velocity (Vmax) and Vmax to KM ratio than the I77T mutant, indicating a higher synthesis rate of the reaction products despite a lower substrate affinity. The Lineweaver-Burk plots are depicted in Fig. 5. Western blot analysis of the wild-type and mutant proteins expressed in vitro demonstrated that both mutations did not affect the translation efficiency (data not shown). The expression and correct intracellular localization of the wild type and mutants in the smooth endoplasmatic reticulum was also shown by immunofluorescence studies of transiently transfected intact COS-7 cells (Fig. 6). Limited proteinase K digestion did not elucidate the altered confirmation of the I77T and A434V mutant proteins in comparison with the CYP21 wild-type protein. No differences were detected in either the pattern of intermediate fragments or sensitivity to proteinase K leading to complete digestion (data not shown). Discussion

The adrenocortical 21-hydroxylase is one of the key enzymes in glucocorticoid and mineralocorticoid biosynthesis.



449

FIG. 3. Mutation analysis by direct DNA sequencing and singleplex minisequencing. A, The base change from T to C at position bp 327 (327T⬎C) leads to the substitution from isoleucine to threonine at position 33. The results of the direct sequencing were confirmed by singleplex minisequencing (black peak represents the mutant C allele, red peak the wild-type T allele at bp 327). B, The C at position bp 2522 is changed to T (2522C⬎T) leading to an amino acid substitution from alanine to valine at position 434 (A434V). The confirmation analysis resulted in a black (C allele, wild-type) and red peak (T allele, mutant) in the patient and the father, whereas the mother carried the wild-type C allele (black peak) in homozygous state.

Mutations in the CYP21 gene cause CAH due to 21OHD, and most mutations identified in CAH patients occur within coding regions, changing one or more amino acids in the translated protein. Investigation of the genetic mutations that cause enzymatic defects of steroid biosynthesis can help to

further define the clinical phenotype of subtypes of steroidogenic disorders. Naturally occurring mutants in CAH patients can provide insight into the structure-function relationship of CYP21 and other cytochrome P450 enzymes. Unique novel mutations were detected in a frequency of

450



FIG. 4. Residual 21-hydroxylase activity of the CYP21 mutants in transiently transfected intact COS-7 cells. The activities of the mutants are expressed in percent of wild-type (WT) activity, which is defined as 100%. Values are depicted for the conversion of 17OHP to 11-deoxycortisol and progesterone to 11-deoxycorticosterone at a substrate concentration of 2 ␮mol/liter of unlabeled steroid (time of incubation: 60 min). The bars represent the mean ⫾ 1 SD for a number (n) of determinations from different transfections.

about 3–5% in large cohort studies, and about 1% of mutations arise de novo (12–14). Both of the investigated novel mutants in the present report are rare or unique mutations, which have not been detected before in the CYP21 mutation analysis of several hundred alleles in our laboratory or the estimated several thousands worldwide. In the present report, we characterized two novel missense mutations of the CYP21 gene, resulting in an impaired 21hydroxylase function. The in vitro expression studies enabled us to categorize the patients into mutation groups. This is of great importance for subsequent genetic counseling in these patients. The necessity of segregation analysis is emphasized by the example of case MM because the detection of the I2 G and the V281L mutations in a heterozygous state, without having analyzed the parents, would have led to a wrong diagnosis. This would result in an incorrect classification into mutation group C and an underestimation of disease severity because the V281L mutation is highly associated with the nonclassical form of CAH.

FIG. 5. Apparent kinetics of the wild-type (WT, F), I77T (Œ), and A434V (f) human 21-hydroxylase. The graphs show LineweaverBurk plots of enzymatic activity measured in intact COS-7 cells expressing the CYP21 enzyme. A, Conversion of 17OHP to 11-deoxycortisol. B, Conversion of progesterone to 11-deoxycorticosterone.

Genotype-phenotype relationship TABLE 1. Apparent kinetic constants for the CYP21 wild-type (WT) and mutant proteins

17-Hydroxyprogesterone KM (␮M) Vmax (pmol/min⫺1䡠mg⫺1) Vmax/KM Progesterone KM (␮M) Vmax (pmol/min⫺1䡠mg⫺1) Vmax/KM

WT

I77T

A434V

1.6 ⫾ 0.3 223.3 ⫾ 14.4 137.9

5.8 ⫾ 3.2 17.0 ⫾ 5.6 2.9

3.5 ⫾ 1.9 77.5 ⫾ 21.2 22.3

1.1 ⫾ 0.2 91.5 ⫾ 6.0 80.6

1.2 ⫾ 0.5 7.1 ⫾ 0.9 5.9

2.6 ⫾ 1.0 35.3 ⫾ 6.18 13.4

Since the I77T mutation is the milder mutation in patient OK, it is responsible for the phenotypic expression of the disease severity. Because the residual enzymatic activity for the conversion of 17OHP to 11-deoxycortisol and progesterone to 11-deoxycorticosterone is 3 and 5% of wild-type activity, respectively, this mutation is comparable with the classical, known I172N mutation, which results in about 75% of cases in a simple virilizing form of CAH. This mutation is categorized in mutation group B in genotype-phenotype correlation studies (12–14). Although patient OK was diagnosed



451

FIG. 6. The intracellular colocalization of the CYP21 protein and KDEL protein in COS-7 cells is depicted in the upper panel. The middle panel shows the endoplasmatic localization of the human 21-hydroxylase by using antihumanCYP21 rabbit polyclonal antiserum as primary and the antirabbit-ALEXA594 antibody as secondary antibody. In the lower panel, the endoplasmatic KDEL protein was marked by using a mouse anti-KDEL antibody as primary and stained by the antimouse-fluorescein isothiocyanate antibody as secondary antibody.

at the age of 6 yr as suffering from CAH with the clinical manifestation of precocious pseudopuberty, he clinically suffered from a simple virilizing form of CAH rather than a nonclassical form. Clinical categorization of the female patient MM is somewhat more difficult. There was no sexual ambiguity of the external genitalia, except for a mild clitoromegaly, which was observed only at the time of diagnosis. Nevertheless, relatively high glucocorticoid substitution doses were necessary to suppress androgen production and prevent growth and BA acceleration. Because of its relatively high in vitro enzyme activity (14 and 12% of wild-type for 17OHP and progesterone, respectively), the A434V mutant can be characterized as a mutation of intermediate severity. It can be hypothesized that this mutation is a kind of intermediate mutation of a severity between the classical simple virilizing form and the late-onset form and has a special status similar to that described for the P30L mutation (14). Furthermore, this underlines the broad disease continuum in the manifestation of CAH, which results in a generally good correlation between genotype and phenotype (1, 2, 25). Most of the CYP21-inactivating mutations result in a total absence of 21-hydroxylase activity, although some result in only a partial loss of 21-hydroxylase activity. To date, four mutations have been shown to cause the nonclassical form of CAH (P30L, V281L, V304M, P453S) (26 –29), whereas the P30L can sometimes be found in mild simple virilizing forms.

Mild reduction of enzyme activity has also been reported for the P105L (30) and R339H (31) mutations. The decrease of residual enzyme activity (⬍10%) leads to an increasing disease severity as reported for the L300F (32), V281G (32), R483P (33), and I172N (34) mutations. All mutants associated with nonclassical CAH have an apparent Michaelis-Menten constant of the same order of magnitude as the wild-type 21-hydroxylase (26 –28, 30). This was also found, with a few exceptions in which no saturation of the enzyme could be achieved under experimental conditions (28, 32), for mutations associated with the simple virilizing form. The results of the functional analysis of the novel I77T and A434V mutations are comparable with these previous findings, whereas the apparent KM values of 17OHP are more affected than those of progesterone. Accordingly, as has been shown for the I172N mutation, as little as 0.6% of wild-type activity is sufficient to prevent significant salt wasting (27). This readily explains why our patients never suffered from salt wasting nor showed elevated plasma renin activity. The major effect of the I77T and A434V mutations was on the velocity of the enzyme reaction, illustrated by markedly decreased Vmax values. The percentage decrease of Vmax/KM is in the range of mutations associated with the simple virilizing form for the I77T mutation. For the A434V mutation, Vmax/KM indicates an intermediate value between mutations associated with the simple virilizing and nonclassical form of CAH (26 –28, 30).

452


Both case reports clearly exemplify the limited utility of the rather arbitrary descriptive terminology of 21OHD forms (i.e. classical vs. nonclassical, simple virilizing vs. salt wasting, etc.) and thus should be used with caution. Putative structure-function relationship

The comparative analysis of the three crystal structures of the bacterial cytochrome P450 enzymes P450terp, P450cam, P450BM3 (35), the computerized three-dimensional model of the CYP21 protein (36), and the three-dimensional model of CYP21 built by using the x-ray structure of the mammalian CYP2C5 as a template (Fig. 7) showed that the isoleucine 77 is localized in helix B. This helix is located at the N-terminal part of the protein and has been reported to show a much lower homology between different P450 enzymes than that detected in the C-terminal part. This is reflected by multiple alignments of cytochrome P450 enzymes, in which the isoleucine is not conserved at this position (Fig. 8). However, hydrophobic amino acids are present here because there are isoleucines in P450bm3 and P450cam, valines in P450terp and P450nor, and an alanine in P450eryf (36). This residue shows the same differences when


aligning different steroidogenic enzymes as CYP17 (GenBank accession no. AAH63388, corresponding residue: alanine), CYP11B1 (GenBank accession no. P15538, corresponding residue: valine), CYP11B2 (GenBank accession no. NP_000489 corresponding residue: valine), CYP19 (GenBank accession no. P11511, corresponding residue: methionine), although the hydrophobicity of amino acids is highly conserved (Fig. 8B). The I77 is conserved in a number of CYP21 enzymes of different species (Fig. 8C): human, mouse (GenBank accession no. BAA31153), rat (GenBank accession no. NP_476442), cow (GenBank accession no. NP_777064), and pig (GenBank accession no. A32525). The predicted B helix borders the B⬘ helix from the N terminus, which appears to participate in the substrate recognition site 1. The C helix is located C terminal from the B⬘ helix. The substrate recognition site 1 is one of the six substrate recognition sites postulated in cytochrome P450 enzymes (37). The crystal structures of the prokaryotic P450 enzymes (PDB codes 2BMH, 1PHB, 1ROM, 1OXA, 1CPT) and the recently published structures of CYP2C5 (PDB code 1DT6) (38) and CYP2C8 (PDB code 1PQ2) (39) illustrate that the B helix is located on the surface of the proteins and is not

FIG. 7. Three-dimensional molecular model of CYP21. A, Total view of the three-dimensional model structure of CYP21. The L helix is shown in dark blue. B, The I77 in helix B is depicted in dark blue. The part between G90 and G110 containing the B⬘-helix, marked in light blue, forms the putative lid of the active site of the enzyme. The flexibility of this part is necessary for substrate access to and product release from the active center of the CYP21 protein. We hypothesize that the introduction of a threonine at amino acid position 77 decreases the flexibility of the lid, resulting in a reduced access to and product release from the active center. C, The mutant V434 (orange) is located in the L helix (dark blue). The change from alanine to valine at this position introduces a longer apolar side chain near the apolar part of the heme group. This disturbs the spatial arrangement of the heme group relative to the enzyme resulting in a reduced 21-hydroxylase activity of the A434V mutant.



453

FIG. 8. Alignment of CYP21 at the helix B and L regions. The left panel shows the helix B region and the bordering parts of the proteins, the right panel the L helix and neighboring residues. Amino acid positions of each sequence are given in italics. The I77 and A434 residues of CYP21 and corresponding animo acids of the aligned cytochrome P450 enzymes are given in bold type. A, Alignment of human CYP21 with the mammalian CYP2C5 and CYP2C8 proteins. B, Alignment of different human steroidogenic enzymes. C, Alignment of mammalian CYP21 enzymes.

close to the heme (Fig. 7, A and B). The flexibility of the B-C loop is consistent with a conformational change required for substrate access when 4-methyl-N-methyl-N-(2-phenyl-2Hpyrazol-3-yl)benzenesulfonamide is bound to P4502C5. The amino acid residues G90 and G110 of CYP21 act as the two hinges, which are necessary for opening the protein. This gives rise to the speculation that changing the hydrophobic isoleucine to the uncharged polar amino acid threonine interferes with the hinge function of G90 and G110. In this case, the I77T mutation in the 21-hydroxylase enzyme disturbs a conformational change of the 21-hydroxylase required for substrate access and product release. This hypothesis could explain why Vmax is dramatically decreased and the apparent KM is of a similar magnitude as in the wild-type enzyme. The A434V mutation is localized in helix L (Fig. 7C), in close proximity to the heme, in which a conservation of nonpolar residues is observed. These nonpolar residues are supposed to build a hydrophobic environment for the heme (35). The A434 residue is highly conserved in prokaryotic P450 enzymes (P450bm3, P450terp, P450cam, P450eryf, and P450nor). It is also a conserved amino acid in the CYP21 enzyme in different species (Fig. 8B) (mouse, rat, cow, and pig) and can be found in the human CYP17, CYP11B1, CYP11B2, and CYP19 enzymes (Fig. 8C). The A434V is the second mutation detected in the L helix. The R435C mutation, not characterized in vitro to date, also seems to have a residual enzymatic function because the reported patient suffered from a mild form of CAH and was compound heterozygous for the Q318X mutation on the other allele (40). It is an interesting fact that amino acid changes in such a highly conserved region can result in 21-hydroxylase mutants with a residual activity that prevents the more severe forms of CAH, although in our case the change from alanine to valine is a change to one of the most closely related amino acids. The introduction of the longer side chain of valine appears to interfere in steric means with the apolar part of the heme group (Fig. 7C), resulting in a reduced 21-hydroxylase activity of the A434V mutant.

In vitro expression analysis of CYP21 gene mutations in CAH patients thus serves as a valuable tool for assessing the phenotypical expression and disease severity of 21OHD. Moreover, the combination of in vitro enzyme function and computerized protein analysis provides new insights in the understanding of cytochrome P450 structural-functional relationships. Acknowledgments The authors are grateful to Dr. W. L. Miller for providing the CYP21 cDNA and the antihuman-CYP21 rabbit polyclonal antiserum. We appreciate the expert technical assistance of Gisela Hohmann and Brigitte Andresen and thank Joanna Voerste for linguistic help with the manuscript. Received April 30, 2004. Accepted September 21, 2004. Address all correspondence and requests for reprints to: Wolfgang G. Sippell, M.D., Professor of Pediatrics, Division of Pediatric Endocrinology, Department of Pediatrics, Christian-Albrechts-Universita¨t zu Kiel, Universita¨tskinderklinik, Schwanenweg 20, D-24105 Kiel, Germany. Email: [email protected].

References 1. White PC, Speiser PW 2000 Congenital adrenal hyperplasia due to 21hydroxylase deficiency. Endocr Rev 21:245–291 2. Speiser PW, White PC 2003 Congenital adrenal hyperplasia. N Engl J Med 349:776 –788 3. Speiser PW, Dupont B, Rubinstein P, Piazza A, Kastelan A, New MI 1985 High frequency of nonclassical steroid 21-hydroxylase deficiency. Am J Hum Genet 37:650 – 667 4. White PC, Grossberger D, Onufer BJ, Chaplin DD, New MI, Dupont B, Strominger JL 1985 Two genes encoding steroid 21-hydroxylase are located near the genes encoding the fourth component of complement in man. Proc Natl Acad Sci USA 82:1089 –1093 5. White PC, New MI, Dupont B 1986 Structure of human steroid 21-hydroxylase genes. Proc Natl Acad Sci USA 83:5111–5115 6. Higashi Y, Yoshioka H, Yamane M, Gotoh O, Fujii-Kuriyama Y 1986 Complete nucleotide sequence of two steroid 21-hydroxylase genes tandemly arranged in human chromosome: a pseudogene and a genuine gene. Proc Natl Acad Sci USA 83:2841–2845 7. White PC, Vitek A, Dupont B, New MI 1988 Characterization of frequent deletions causing steroid 21-hydroxylase deficiency. Proc Natl Acad Sci USA 85:4436 – 4440 8. Higashi Y, Tanae A, Inoue H, Fujii-Kuriyama Y 1988 Evidence for frequent

454

9. 10. 11. 12.

13.

14.

15. 16.

17.

18.

19. 20. 21. 22. 23. 24. 25.


gene conversion in the steroid 21-hydroxylase P-450(C21) gene: implications for steroid 21-hydroxylase deficiency. Am J Hum Genet 42:17–25 White PC, Tusie-Luna MT, New MI, Speiser PW 1994 Mutations in steroid 21-hydroxylase (CYP21). Hum Mutat 3:373–378 Wedell A 1998 Molecular genetics of congenital adrenal hyperplasia (21hydroxylase deficiency): implications for diagnosis, prognosis and treatment. Acta Paediatr 87:159 –164 The Human Gene Mutation Database Available at: http://uwcmml1s. uwcm.ac.uk/uwcm/mg/search/120605.html Speiser PW, Dupont J, Zhu D, Serrat J, Buegeleisen M, Tusie-Luna MT, Lesser M, New MI, White PC 1992 Disease expression and molecular genotype in congenital adrenal hyperplasia due to 21-hydroxylase deficiency. J Clin Invest 90:584 –595 Wedell A, Thilen A, Ritzen EM, Stengler B, Luthman H 1994 Mutational spectrum of the steroid 21-hydroxylase gene in Sweden: implications for genetic diagnosis and association with disease manifestation. J Clin Endocrinol Metab 78:1145–1152 Krone N, Braun A, Roscher AA, Knorr D, Schwarz HP 2000 Predicting phenotype in steroid 21-hydroxylase deficiency? Comprehensive genotyping in 155 unrelated, well defined patients from southern Germany. J Clin Endocrinol Metab 85:1059 –1065 Ezquieta B, Oliver A, Gracia R, Gancedo PG 1995 Analysis of steroid 21hydroxylase gene mutations in the Spanish population. Hum Genet 96:198 –204 Jaaskelainen J, Levo A, Voutilainen R, Partanen J 1997 Population-wide evaluation of disease manifestation in relation to molecular genotype in steroid 21-hydroxylase (CYP21) deficiency: good correlation in a well defined population. J Clin Endocrinol Metab 82:3293–3297 Baumgartner-Parzer SM, Schulze E, Waldhausl W, Pauschenwein S, Rondot S, Nowotny P, Meyer K, Frisch H, Waldhauser F, Vierhapper H 2001 Mutational spectrum of the steroid 21-hydroxylase gene in Austria: identification of a novel missense mutation. J Clin Endocrinol Metab 86:4771– 4775 Kharrat M, Tardy V, M’Rad R, Maazoul F, Jemaa LB, Refai M, Morel Y, Chaabouni H 2004 Molecular genetic analysis of Tunisian patients with a classic form of 21-hydroxylase deficiency: identification of four novel mutations and high prevalence of Q318X mutation. J Clin Endocrinol Metab 89: 368 –374 Wilson RC, Mercado AB, Cheng KC, New MI 1995 Steroid 21-hydroxylase deficiency: genotype may not predict phenotype. J Clin Endocrinol Metab 80:2322–2329 Speiser PW, New MI, Tannin GM, Pickering D, Yang SY, White PC 1992 Genotype of Yupik Eskimos with congenital adrenal hyperplasia due to 21hydroxylase deficiency. Hum Genet 88:647– 648 Peter M, Sippell WG, Lorenzen F, Willig RP, Westphal E, Grosse-Wilde H 1990 Improved test to identify heterozygotes for congenital adrenal hyperplasia without index case examination. Lancet 335:1296 –1299 Krone N, Braun A, Weinert S, Peter M, Roscher AA, Partsch CJ, Sippell WG 2002 Multiplex minisequencing of the 21-hydroxylase gene as a rapid strategy to confirm congenital adrenal hyperplasia. Clin Chem 48:818 – 825 Krone N, Roscher AA, Schwarz HP, Braun A 1998 Comprehensive analytical strategy for mutation screening in 21-hydroxylase deficiency. Clin Chem 44: 2075–2082 Vriend G 1990 WHAT IF: a molecular modeling and drug design program. J Mol Graph 8:52–56, 29 Wedell A 1998 An update on the molecular genetics of congenital adrenal hyperplasia: diagnostic and therapeutic aspects. J Pediatr Endocrinol Metab 11:581–589


26. Tusie-Luna MT, Speiser PW, Dumic M, New MI, White PC 1991 A mutation (Pro-30 to Leu) in CYP21 represents a potential nonclassic steroid 21-hydroxylase deficiency allele. Mol Endocrinol 5:685– 692 27. Tusie-Luna MT, Traktman P, White PC 1990 Determination of functional effects of mutations in the steroid 21-hydroxylase gene (CYP21) using recombinant vaccinia virus. J Biol Chem 265:20916 –20922 28. Lajic S, Clauin S, Robins T, Vexiau P, Blanche H, Bellanne-Chantelot C, Wedell A 2002 Novel mutations in CYP21 detected in individuals with hyperandrogenism. J Clin Endocrinol Metab 87:2824 –2829 29. Owerbach D, Sherman L, Ballard AL, Azziz R 1992 Pro-453 to Ser mutation in CYP21 is associated with nonclassic steroid 21-hydroxylase deficiency. Mol Endocrinol 6:1211–1215 30. Nikoshkov A, Lajic S, Holst M, Wedell A, Luthman H 1997 Synergistic effect of partially inactivating mutations in steroid 21-hydroxylase deficiency. J Clin Endocrinol Metab 82:194 –199 31. Helmberg A, Tusie-Luna MT, Tabarelli M, Kofler R, White PC 1992 R339H and P453S: CYP21 mutations associated with nonclassic steroid 21-hydroxylase deficiency that are not apparent gene conversions. Mol Endocrinol 6:1318 – 1322 32. Lajic S, Robins T, Krone N, Schwarz HP, Wedell A 2001 CYP21 mutations in simple virilizing congenital adrenal hyperplasia. J Mol Med 79:581–586 33. Nikoshkov A, Lajic S, Vlamis-Gardikas A, Tranebjaerg L, Holst M, Wedell A, Luthman H 1998 Naturally occurring mutants of human steroid 21-hydroxylase (P450c21) pinpoint residues important for enzyme activity and stability. J Biol Chem 273:6163– 6165 34. Higashi Y, Hiromasa T, Tanae A, Miki T, Nakura J, Kondo T, Ohura T, Ogawa E, Nakayama K, Fujii-Kuriyama Y 1991 Effects of individual mutations in the P-450(C21) pseudogene on the P-450(C21) activity and their distribution in the patient genomes of congenital steroid 21-hydroxylase deficiency. J Biochem (Tokyo) 109:638 – 644 35. Hasemann CA, Kurumbail RG, Boddupalli SS, Peterson JA, Deisenhofer J 1995 Structure and function of cytochromes P450: a comparative analysis of three crystal structures. Structure 3:41– 62 36. Mornet E, Gibrat JF 2000 A 3D model of human P450c21: study of the putative effects of steroid 21-hydroxylase gene mutations. Hum Genet 106:330 –339 37. Gotoh O 1992 Substrate recognition sites in cytochrome P450 family 2 (CYP2) proteins inferred from comparative analyses of amino acid and coding nucleotide sequences. J Biol Chem 267:83–90 38. Wester MR, Johnson EF, Marques-Soares C, Dansette PM, Mansuy D, Stout CD 2003 Structure of a substrate complex of mammalian cytochrome P450 2C5 at 2.3 A resolution: evidence for multiple substrate binding modes. Biochemistry 42:6370 – 6379 39. Schoch GA, Yano JK, Wester MR, Griffin KJ, Stout CD, Johnson EF, Marques-Soares C, Dansette PM, Mansuy D 2003 Structure of human microsomal cytochrome P450 2C8: evidence for a peripheral fatty acid binding site. J Biol Chem 15:15 40. Deneux C, Tardy V, Dib A, Mornet E, Billaud L, Charron D, Morel Y, Kuttenn F 2001 Phenotype-genotype correlation in 56 women with nonclassical congenital adrenal hyperplasia due to 21-hydroxylase deficiency. J Clin Endocrinol Metab 86:207–213 41. Tanner JM, Goldstein H, Whitehouse RH 1970 Standards for children’s height at ages 2–9 years allowing for heights of parents. Arch Dis Child 45:755–762 42. Bayley N, Pinneau SR 1952 Tables for predicting adult height from skeletal age: revised for use with the Greulich-Pyle hand standards. J Pediatr 40:423– 441 (erratum corrected J Pediatr 41:371)

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.