| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Engelhardt Institute of Molecular Biology (Y.G., P.R., S.B., V.Pr.), Russian Academy of Sciences, Moscow 119991, Russian Federation; Division of Pediatric Endocrinology, Department of Pediatrics (F.G.R., W.G.S., N.K.), Christian-Albrechts-Universität zu Kiel, Universitätsklinikum Schleswig-Holstein (Campus Kiel), D-24105 Kiel, Germany; Biochemisches Institut (J.G.), Christian-Albrechts-Universität zu Kiel, D-24098 Kiel, Germany; and Institute of Pediatric Endocrinology, Endocrinological Research Center (N.K., T.S., V.Pe., A.T.), Moscow 117036, Russian Federation
Address all correspondence and requests for reprints to: Nils Krone, M.D., Division of Pediatric Endocrinology and Diabetes, Department of Pediatrics, Christian-Albrechts-Universität zu Kiel, Universitätsklinikum Schleswig-Holstein (Campus Kiel), Schwanenweg 20, D-24105 Kiel, Germany. E-mail: krone{at}pediatrics.uni-kiel.de.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Objective: We analyzed the functional and structural consequences of the four CYP21A2 missense mutations (C169R, G178R, W302R, and R426C) to prove their clinical relevance and study their impact on CYP21 function.
Results: Analyzing the mutations in vitro revealed an almost absent or negligible CYP21 activity for the conversion of 17-hydroxyprogesterone to 11-deoxycortisol and progesterone to deoxycorticosterone. Protein translation and intracellular localization were not affected by the mutants, as could be demonstrated by Western blotting and immunofluorescence studies. Analysis of these mutants in a three-dimensional model structure of the CYP21 protein explained the observed in vitro effects because all the mutations severely interfere either directly or indirectly with important structures of the 21-hydroxylase protein.
Conclusion: The in vitro expression analysis of residual enzyme function is a complementary method to genotyping and an important tool for improving the understanding of the clinical phenotype of 21-hydroxylase deficiency. This forms the foundation for accurate clinical and genetic counseling and for prenatal diagnosis and treatment. Moreover, this report demonstrates that the combination of in vitro enzyme analysis and molecular modeling can yield novel insights into CYP450 structure-functional relationships.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The 21-hydroxylase gene (CYP21A2) and its nonfunctional pseudogene (CYP21A1P) are located on chromosome 6 (6p21.3), sharing a high homology of about 98%. CYP21A2-inactivating mutations occur as complete gene deletions, large gene conversions, and pseudogene-derived mutations (1). Rare pseudogene-independent mutations are comprehensively listed by the Human Cytochrome P450 (CYP) Allele Nomenclature Committee (http://www.cypalleles.ki.se/cyp21.htm), and a strong genotype-phenotype correlation exists (3, 4, 5).
Here, we analyze the functional and structural consequences of four novel CYP21A2 missense mutations (C169R, G178R, W302R, R426C) identified in unrelated Russian CAH patients.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Four novel CYP21A2 mutations were identified analyzing a Russian cohort of more than 100 unrelated CAH patients with 21OHD, three female patients (cases V, K, and F) with simple virilizing CAH and one male patient (case P) with salt-wasting CAH.
In patient K (46,XX), virilization of the external genitalia (Prader stage II) was firstly noted at the age of 2.5 months. After biochemical confirmation of 21OHD, glucocorticoid replacement was started. Salt loss was never recorded, and plasma renin was normal.
Patient F (46,XX) was diagnosed to have 21OHD at the age of 6 yr. External genitalia was mildly virilized, not requiring surgical correction. She was referred for follow-up at the age of 18 yr because of hirsutism and amenorrhea. She has never suffered from salt loss.
Patient P (46,XY) suffered from severe salt wasting during the second week of life. 21OHD was established, and treatment with hydrocortisone and fludrocortisone was initiated.
Patient V (46,XX) was born with severely virilized external genitalia (Prader stage IV) and registered as male. The diagnosis of 21OHD was made at 6 yr of age when the child was referred to a pediatric endocrinologist with signs of precocious pseudopuberty (pubarche, bone age 13 yr).
Mutation analysis
Screening for the most frequent CYP21A2-inactivating mutations and screening analysis for novel mutations by direct DNA sequencing after allele-specific PCR amplification of the CYP21A2 gene was performed as described previously (4). CYP21A2 gene deletions and inactivating rearrangements were analyzed by a multiplex ligation-dependent probe amplification (MLPA) strategy according to the manufacturer’s protocol (MRC-Holland, Amsterdam, The Netherlands).
Construction of plasmids and site-directed mutagenesis
The human full-length CYP21A2 cDNA was synthesized from an adrenocortical adenoma mRNA and cloned into the pBluescript-II-KS(+) vector (Stratagene, La Jolla, CA). Mutations were introduced by site-directed mutagenesis using the corresponding mutagenesis primers. Wild-type and mutated CYP21A2 cDNAs were subsequently cloned into the BamHI-XhoI restriction sites of the pcDNA3.1(+) vector (Invitrogen, Karlsruhe, Germany). The introduction of mutations and the integrity of the insert were checked by DNA sequencing.
In vitro expression, assays of enzyme activity, and immunofluorescence
The in vitro expression experiments were performed as described (6), with a few modifications. For transient transfection of COS-7 cells, a pcDNA3.1(+)-CYP21A2 construct and Renilla luciferase pRL-vector (Promega, Mannheim, Germany) were used. The CYP21 activity in intact COS-7 cells was measured 24 h after transfection. The cells were incubated for 90 min at 37 C with 500 µl DMEM containing 0.2 µCi 3H-labeled substrate [17-hydroxyprogesterone (17OHP) or progesterone], 2 µmol/liter unlabeled steroid, and 8 mmol/liter nicotinamide adenine dinucleotide phosphate. Steroid extraction, thin-layer chromatography, data analysis, and Western blot analysis were performed as described (6).
The immunofluorescence was performed using a standard protocol. An antihuman-CYP21 polyclonal rabbit antiserum and a mouse anti-KDEL antibody (BIOMOL, Hamburg, Germany), each in 1:250 dilution, were used as primary antibodies. Secondary antibodies antirabbit-ALEXA Fluor 488 and antimouse-ALEXA Fluor 594 (Molecular Probes, Leiden, The Netherlands) were used in 1:500 dilutions.
Molecular modeling
The detailed generation of the human CYP21 three-dimensional structure model using the x-ray structure of the mammalian cytochrome CYP2C5 (Pdb code 1DT6) as a template has been described previously (6). The structural representations were generated using the programs Deep View/Swiss-Pdb Viewer (http://www.expasy.org/spdbv/) and POV-Ray (http://www.povray.org).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Direct sequencing of the entire CYP21A2 gene and MLPA analysis in the four patients revealed four novel CYP21A2-missense mutations.
Patient K carried the g.989T>C mutation (C169R; paternal allele) and a deletion of CYP21A2 exons 1–6 (maternal allele). Screening for the most common CYP21A2-inactivating mutations revealed a heterozygous I172N mutation in patient F. She was found to be compound heterozygous for the novel g.1016G>A (G178R) mutation (maternal allele). The MLPA analysis showed neither a gene deletion nor a rearrangement. Because paternal material was unavailable for analysis and the I172N mutation was not detected on the maternal allele, the patient’s genotype was assumed to be I172N (paternal allele)/G178R (maternal allele). Patient P carried the g.1746T>C mutation (W302R) (paternal allele) and a deletion of exons 1–6 (maternal allele). Patient V was compound heterozygous for the common I172N mutation and the g.2497C>T mutation (R426C) (maternal allele). The MLPA analysis showed a wild-type pattern. Because paternal DNA was unavailable, her genotype was categorized as I172N (paternal allele)/R426C (maternal allele).
Functional analysis
Transient expression of the novel mutants in COS-7 cells revealed complete or almost absent CYP21 enzymatic activity, both for the conversion of 17OHP and progesterone (Table 1
). This proved their pathogenicity, allowing for their categorization into the group of CYP21A2-null mutations. The previously described R426H mutation (7) was analyzed alongside the novel mutants, revealing nearly complete loss of activity (Table 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The functional analysis showed complete loss of enzymatic function of the 21-hydroxylase mutants neither affecting protein synthesis nor proper intracellular localization. The G178R and R426C mutations were each detected in a compound heterozygous state on one allele in CAH patients suffering from simple virilizing (SV)-CAH together with the I172N mutation. Hence, the expressed SV phenotype is consistent with the genotypes considering the I172N to be the milder mutation (1).
From the in vitro data, the W302R mutation is a null mutation. Thus, the expected genotype-phenotype correlation was found in the patient compound heterozygous for the W302R mutation and a partial CYP21A2 gene deletion.
The hemizygous C169R mutation was detected in a patient who had never suffered from salt wasting. From this manifestation, the C169R mutant was expected to have an activity of approximately 1–5%, typical for mutations associated with SV-CAH. However, according to the in vitro expression data, we would recommend classifying the C169R mutation as a severe null mutation. This view is also supported by the putative effect on the CYP21 structure.
The C169 residue is conserved in CYP21 in many different species (Fig. 1C). It is not necessarily found in other CYP450 enzymes (Fig. 1A). However, the residue’s hydrophobicity is conserved in different human steroidogenic enzymes (Fig. 1B). The C169 is localized in helix E of CYP21 and found in its hydrophobic core (Fig. 2, A and B). The substitution of the polar, long side-chained R169 for C169 will mainly interfere with Y191 located in helix F (Fig. 2B). The F helix is part of the CYP21 surface forming a water exit channel with helix I in CYP2C5. This channel can be closed by moving helix F relative to helix I (9). It has been suggested that the closure of this channel plays a role in controlling protonation of reduced oxygen intermediates bound to the heme iron during catalysis. Furthermore, the C169R mutation can change the E helix position relative to helix I. The importance of the positioning of helix E relative to helix I is illustrated by the common I172N mutation resulting in an impaired protein folding (10).
|
|
). The mild G178A mutation resulted in an enzyme activity of less than 20% (11). In CYP21, the G178 residue is located on the surface, in the E–F connecting loop (Fig. 2, A and C). The change to arginine will interfere with the residues I131, R132, and D133 located in the C–D connecting loop. This can influence the D helix position and thereby indirectly the C helix position. Because the N-terminal part of the C helix is involved in heme binding (12), this will indirectly influence heme positioning and heme binding. Secondly, the C helix is anchoring the B' helix, which together with the B-B'-C helices, facilitates substrate access and product release (9). Furthermore, the interference with R254, located N terminal from helix H, influences the positioning relative to helix I.
The W302 residue is localized in the highly conserved I helix (Fig. 2, A and D). This helix crosses the whole molecule and is involved in heme binding and substrate recognition. The substitution of the polar amino acid arginine for aromatic tryptophan in the protein core will displace the I helix relative to the heme, indirectly interfering with heme binding and consequently abolishing 21-hydroxylase activity. Two mutations (L300F, V304M) located in close vicinity to the W302 residue resulted in a milder decrease of enzyme function and were found in patients with SV and nonclassic CAH (13, 14). However, our patient’s salt-wasting phenotype is consistent with the in vitro-measured residual activity and the putative change of the CYP21 protein.
The R426 is located in the highly conserved Cys-pocket motif. The corresponding arginine residue is found in mammalian CYP450s (Fig. 1). It is one of four residues coordinating the heme propionate (12, 15) (Fig. 2E). Therefore, a substitution for cysteine at this key position will lead to a severe loss of enzyme function. However, even the conservative R426H mutation leads to almost absent 21-hydroxylase activity. Earlier functional analysis of the R426H mutation showed a higher enzymatic activity, but these data were measured under saturated conditions and hence overestimated the residual activity (7). Only recently similar data compared with our assay on the R426H activity were published (16). Substitutions of either a histidine or cysteine for arginine at the corresponding residue in naturally occurring mutations of CYP11B1 [R448H (17), R448C (18)] and CYP17A1 [R440H (19)] also resulted in absent enzyme activity.
The functional and structural analysis of all four novel CYP21A2 mutations showed a major impact on 21-hydroxylase activity. We conclude that the combination of in vitro and in silico analysis is an important tool for assessing the patients’ phenotype and valuable for counseling.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online September 19, 2006
Abbreviations: CAH, Congenital adrenal hyperplasia; CYP, cytochrome P450; MLPA, multiplex ligation-dependent probe amplification; 21OHD, 21-hydroxylase deficiency; 17OHP, 17-hydroxyprogesterone; SV, simple virilizing.
Received April 10, 2006.
Accepted September 8, 2006.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-hydroxylase deficiency. J Clin Endocrinol Metab 79:160–164[Abstract]This article has been cited by other articles:
 |
|
 |
|
  M. Welzel, N. Wustemann, G. Simic-Schleicher, H. G. Dorr, E. Schulze, G. Shaikh, P. Clayton, J. Grotzinger, P.-M. Holterhus, and F. G. Riepe Carboxyl-Terminal Mutations in 3{beta}-Hydroxysteroid Dehydrogenase Type II Cause Severe Salt-Wasting Congenital Adrenal Hyperplasia J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1418 - 1425. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
