| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatrics, University of Bologna and S. Orsola-Malpighi Hospital (M.B., L.B., A.B., P.P., A.C., E.C.), 40139 Bologna, Italy; and Department of Molecular Medicine (S.L., A.W.), Center for Molecular Medicine, Karolinska Hospital, 17176 Stockholm, Sweden
Address all correspondence and requests for reprints to: Michela Barbaro, Department of Molecular Medicine, Center for Molecular Medicine, Karolinska Hospital, 171 76 Stockholm, Sweden. E-mail: michela.barbaro{at}cmm.ki.se; or Prof. Alessandro Cicognani, Department of Pediatrics, Via Massarenti 11, 40138 Bologna, Italy. E-mail: cicogna{at}alma.unibo.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The two amino acid substitutions were reconstructed by in vitro site-directed mutagenesis, the proteins were transiently expressed in COS-1 cells, and enzyme activity toward the two natural substrates (17-hydroxyprogesterone and progesterone) was determined. The A15T mutant exhibited no significant difference in activity compared with the wild-type protein, whereas the P482S mutation reduced enzyme activity to 70% of normal. This impairment of activity was confirmed in vivo by detection of heterozygote carriers by the ACTH test.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The 21-hydroxylase locus has a complicated structure where the active gene CYP21 and the inactive pseudogene CYP21P are adjacent to and alternating with the C4B and C4A genes encoding the fourth component of complement (3). One C4 gene and one CYP21 gene are part of a repeated module (RCCX) located in the HLA class III region on chromosome 6p21.3 (4).
The genetic lesions that account for the majority of all cases of 21-hydroxylase deficiency are deletions and various gene conversion-type mutations in which CYP21 has partially been converted to pseudogene-like sequences. These gene conversion events give rise to CYP21 genes that carry one or multiple mutations normally present in the pseudogene or a chimeric CYP21P/CYP21 gene (5, 6). Rarely, mutations arise independently of the pseudogene, and these unique mutations are thus found in a single family or one specific population (7, 8, 9, 10, 11, 12).
Here we describe two amino acid substitutions. One (A15T) was found in a large gene conversion in the CYP21P-like part of the CYP21 sequence and was recently reported also in two Slovenian patients (13). The other is a novel mutation (P482S) found in unrelated subjects with a very mild form of CAH or in heterozygous carriers.
To investigate whether P482S could be responsible for the phenotypes of the patients and in order to study the potential effect of A15T, the mutations were reconstructed by in vitro site-directed mutagenesis, and the functional properties of the mutant enzymes were analyzed and compared with the normal protein. Activities toward the two natural substrates, 17OHP and progesterone, as well as apparent kinetic constants were determined.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients were referred to our center for endocrine evaluation. Informed consent was obtained from parents or tutors of all subjects involved in this study.
Subject 1 was the first child of nonconsanguineous parents of south Italian origin. The pregnancy was uneventful with cesarean delivery at term due to podalic presentation. Birth weight (BW) and birth length (BL) were in the upper normal female range (3910 g, 53.5 cm). Diagnosis was suspected at 3 d of life due to genital virilization (Prader IV). Hormonal evaluation performed on the third day of life, showed normal Na (143 mmol/liter) and elevated levels of K (7.4 mmol/liter), 17OHP (212.0 nmol/liter) and 
4-androstenedione (4-A) (2376.9 nmol/liter). Therapy with hydrocortisone was started on the 11th day of life, and fluorohydrocortisone was started on d 16. No vomiting or other clinical signs of salt wasting were shown before onset of substitutive therapy.
Subject 2 was the second child of nonconsanguineous parents of northern Italian origin. The pregnancy was uneventful with a premature (36 wk) cesarean delivery. BW and BL were appropriate for gestational age (2750 g, 46 cm). The girl was referred to our hospital at 3.6 months of age due to ambiguous genitalia (Prader stage III) not noted before. The hormonal levels of ACTH, 17OHP, and 4-A were 145.3 pmol/liter, 518.4 nmol/liter, and 2328.2 nmol/liter, respectively. The child had PRA more than 25 µg/liter·h and normal electrolyte levels (Na, 142 mmol/liter; K, 4.6 mmol/liter; and Cl, 100 mmol/liter). Failure to thrive was present. Treatment was started firstly with hydrocortisone alone and thereafter also with fluorohydrocortisone (0.05 mg/d).
Subjects 3a and 3b were siblings of nonconsanguineous parents of south Italian origin. The pregnancy was characterized by menace of spontaneous abortion in the first trimester for subject 3a (GA, 40 wk; BW, 2600 g) and menace for premature delivery in the fifth month of pregnancy for case 3b (GA, 40 wk; BW, 3450 g). Subject 3a was admitted to hospital for the first time at 1.2 yr of age for very mild hypertrichosis. She was examined at 2.3 yr of age for mild monolateral transitory thelarche. Bone age (BA) matched chronological age, Height for Chronological Age Standard Deviation Score (HCASD) was +0.8 SD and basal adrenal and ovarian steroid levels were normal for age. At 5.1 yr of age, an ACTH test [Synachten (Ciba Geigy, Huninge, France) 0.25 mg, at 0–60 min] was performed, and the 17OHP response was in the range of NC-CAH (Table 1
). Treatment with hydrocortisone (13.6 mg/m2·d) and LHRH analog (3.75 mg every 28 d) was started at 8.2 yr of age when BA reached 10.0 yr and breast Tanner stage 2. The girl reached a final adult height of 156.7 cm (target height 158.5 cm). The brother (3b) did not show any hyperandrogenic signs, but an ACTH test performed at 4.5 yr of age showed similar biochemical characteristics as those of his sister (Table 1). At present, he is 13.2 yr old and has never been treated with substitutive glucocorticoid therapy. He has reached a final height of 166.4 cm (target height, 171.5 cm).
|
4-A were 56.0 mmol/liter, 25.9 nmol/liter and 189.0 nmol/liter, respectively. An ACTH test showed a 17OHP response in the lower range of NC-CAH (Table 1
). Treatment with hydrocortisone (8.5 mg/m2·d) was started at 8.0 yr of age when the girl showed a BA of 11.0 yr, pubic hair Tanner stage 3, and immodified thelarche. Menarche occurred at 10.8 yr of age, and she achieved an adult height of 164.0 cm (target height: 161.0 cm).
Subject 5 was the second child of consanguineous parents (first cousins) of Italian origin. The pregnancy was uneventful with an at term delivery of a normal female newborn (BW, 2500 g; BL, 53 cm). Her growth was regular (adult height, –0.7 SD; target height, –1.2 SD) with menarche occurring at 14.5 yr and followed by irregular menses. She was examined for the first time at 17 yr of age for menstrual irregularities with no signs of virilization. 17OHP response to ACTH was in the lower range of NC-CAH (Table 1).
Subjects 6a and 6b were two sisters of nonconsanguineous parents of Italian origin. The uneventful pregnancies (2.5 yr apart) lasted 38 wk and generated two normal female newborns with a BW of 3600 g in both cases. The two subjects showed muscular problems for the first time at 2.5 and 1.0 yr of age, respectively, and spinal amyotrophy (Cugelberg-Welander disease) was diagnosed. Subject 6a showed hirsutism (Ferriman and Gallway score = 12) at 13.0 yr (she had regular menses since 10.8 yr). An ACTH test showed moderately elevated 17OHP and 4 levels (32.2 nmol/liter and 1257 nnmol/liter, respectively) (Table 1). She is now a 21-yr-old female, still hirsute but with regular menstrual periods and without hormonal treatment. Subject 6b came to our observation at the age of 13.9 yr for a mild form of hypertrichosis (Ferriman and Gallway score = 6). She has had regular menses since the age of 12.0 yr, and an ACTH test showed 17OHP and 4-A values in the range of heterozygosity for 21-hydroxylase deficiency (24.0 nmol/liter and 861.0 nmol/liter, respectively) (Table 1). No hormonal therapy was given.
Subject 7 was the first of two children of nonconsanguineous parents of Italian origin. Pregnancy was characterized by uterine contractions from the fourth month on, and vasopressin was given. An at-term cesarean delivery was performed due to suspected fetal distress. BW and BL were 4000 g and 51.0 cm, respectively. The boy was examined for the first time at our hospital at 7.4 yr of age for precocious pubarche. BA was 8.5 yr, HCASD was +1.9 SD (Height for Bone Age Standard Deviation Score = +0.6 SD), and post-ACTH 17OHP and 4-A were 23.6 mmol/liter and 237.7 nmol/liter, respectively (Table 1). CYP21 analysis was performed as part of a study aimed to correlate the levels of 17OHP after ACTH stimulation to CYP21 genotype (14).
Hormonal analysis
Neonatal screening for 21OH deficiency was carried out by measuring the levels of 17OHP from blood samples spotted on filter paper using a fluoroimmunometric assay (DELFIA Wallac Oy, Turku, Finland) (15). Serum 17OHP was assayed using a direct RIA method kit (Diagnostic Products Corp., Los Angeles, CA). It should be noted that nonextracted samples assayed for 17OHP by RIA will sometimes provide falsely elevated results in newborn infants. The ACTH test (Synacthen 0.25 mg iv) was carried out by measuring 17OHP levels at 0 and 60 min.
DNA analysis
Genomic DNA from subjects and their parents and relatives was obtained from peripheral blood by phenol/chloroform extraction.
The CYP21 gene amplification was carried out using CYP21-specific primers that create three partially overlapping fragments (P1P2, P3P4, CD) as previously described (16). For the CD fragment amplification, Tth DNA polymerase (Epicentre, Madison, WI) was used with a buffer without KCl to avoid the described allele dropout at nucleotide position 656 (17). Purified PCR fragments were sequenced using the Big Dye Terminator Sequencing kit (Applied Biosystems, Warrington, UK) and primers to sequence the entire gene and the proximal promoter region until nucleotide –370. The presence of a large gene conversion and/or deletion was excluded by heterozygosity for an intragenic polymorphism or a mutation and by segregation analysis when possible.
The pseudogene was amplified in two fragments, P1-8R and 8F-P4, using the primers 8R (5'-AGAGCAGAGACCAACGACA-3') and 8F (5'-TGTCCTTGGTCTCTGCTCT-3') that anneal to the 8-bp deletion in exon 3. The P1-8R and 8F-P4 fragments were purified and sequenced using the CEQ-DTCS Quick Start Kit (Beckman Coulter, Fullerton, CA), the reverse primer BR (5'-CGTCACCTCCGCCCCCTCCTAT-3'), and the forward primer FF (5'-GTGAAAATGTGGTGGAGGCTGGTC-3'), respectively.
Numbering of nucleotides and amino acids follows the reference sequence of White et al. (18).
Construction of plasmids and introduction of mutations
Construction of the expression plasmid pCMV4-CYP21, containing the cDNA of CYP21, has previously been described (19) as well as introduction of mutations by in vitro site-directed mutagenesis using the pALTER system (20). BglII/KpnI fragments of pALTER-CYP21(A15T) and pALTER-CYP21(P482S) were transferred to pCMV4-CYP21, generating pCMV4-CYP21(A15T) and pCMV4-CYP21(P482S). The entire CYP21 cDNA was sequenced to verify the correct mutation incorporation and to exclude additional sequence alterations.
Expression of P450c21 in mammalian COS-1 cells and assay of enzyme activity
Transient COS-1 transfection was performed by lipofectin (FuGene, Amersham Pharmacia Biotech, Arlington Heights, IL): approximately 2 x 105 cells, seeded in 3.5-cm Petri dishes, were transfected with 1 µg of each of the pCMV4-CYP21 constructs and together with 0.25 µg of the ß-galactosidase vector pCH110 (Pharmacia, Uppsala, Sweden) using 3 µl of FuGene for 48 h. To determine the 21-hydroxylase activity in intact cells, 250 µCi of H3-labeled substrate (17OHP or progesterone) were added to the medium together with 2 µmol/liter of unlabeled steroid and 4 mmol/liter of reduced nicotinamide adenine dinucleotide phosphate. After incubation at 37 C for 20 min, 240 µl of medium were collected in duplicate samples and steroids were extracted with 300 µl of methylene chloride (Sigma, St. Louis, MO), evaporated to dryness, and dissolved in ethanol. The steroids were separated by thin-layer chromatography in chlorophorm:ethylacetate (80:20), and the radioactivity was measured by liquid scintillation spectrophotometry. Subsequently the cells were trypsinized, sonicated (1 x 10 sec) and subjected to measurements of protein content and ß-galactosidase activity.
Enzyme activities were expressed as a percentage of conversion grade, taking the activity of the wild-type P450c21 as 100%, after correction for total protein content. The ratio of ß-galactosidase activity/total protein content was measured in each experiment to verify the reproducibility of transformation efficiency.
To determine apparent kinetic constants, intact cells were incubated as described above together with 0.5, 1.0, 2.0, 3.0, 4.0, or 7.0 µmol/liter unlabeled substrate. After incubation at 37 C for 20 min, steroids were extracted and analyzed as described above. Apparent kinetic constants were calculated after linear regression of the data derived from determinations of enzymatic activity at each of the six different substrate concentrations.
To ascertain the amount of translated P450c21 in transfected cells, proteins from supernatants of homogenized cells were trichloroacetic acid-precipitated, size-separated in a 10% SDS-PAGE, and transferred to a nitrocellulose membrane. Immunoblot analyses were performed according to standard procedures, using polyclonal antibodies against human P450c21.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
By direct-sequencing of the complete coding sequence of the CYP21 gene and of the proximal promoter in subjects 1 and 2, a chimeric CYP21P/CYP21 gene was found. The breakpoint was downstream of the P30L mutation and upstream of the Intron 2 splice mutation usually present in the pseudogene sequence. In the CYP21P part of the sequence, a G to A transition causing an alanine to threonine substitution at codon 15 (A15T) was also found. On the other allele, a null mutation was present (Q318X) in subject 1, and a CYP21 deletion or large gene conversion was present in subject 2. The chromosomes with the chimeric genes shared the HLA haplotype B13 CW6 DR7 DQ2 (Table 1). We screened for the A15T substitution among the CYP21P genes of 50 randomly selected individuals and four individuals were found to carry this alteration.
Subjects 3–7 had one allele with a C to T transition that caused a proline to be replaced by a serine at codon 482 (P482S). Subjects 3a and 4 presented the nonclassical form of CAH, with a very mild phenotype, and on their other alleles they had the common nonclassical mutation V281L. The asymptomatic brother (subject 3b) of subject 3a presented the same genotype. The severe Intron 2 splice mutation was present in the other allele of subject 5. Subjects 6a–7 were suspected to be heterozygous carriers of CAH suggested by the ACTH test, and the only alteration they presented was the P482S mutation in one allele. For all subjects, except for subject 7, the parental DNA was analyzed and mutation segregation confirmed. Of the five alleles with the P482S mutation, four were analyzed for the HLA haplotype and all of them presented the DR4: three the B41 DR4 and two the A2 B41 DR4 haplotype (Table 1). The CYP21P genes of 50 randomly selected Italian individuals were screened for the P482S substitution, but no one carried this alteration.
Enzyme activity of mutants
To assess the influence of the missense mutations on P450c21 activity, the mutations were reconstructed by in vitro site-directed mutagenesis. Both mutant forms as well as the wild-type protein were transiently expressed in COS-1 cells, and enzyme activity toward the two natural substrates, 17OHP and progesterone, was assayed in intact cells. The activity of the mutant forms was compared with the normal protein, which was arbitrarily defined as 100%. As illustrated in Fig. 1, the P482S mutation reduced enzyme activity to 72% for 17OHP and to 70% for progesterone (significantly different from normal, Student’s t test; P = 0.008 and P = 0.003, respectively). The A15T exhibited 95% activity for 17OHP and 101% for progesterone. This was not significantly different from normal (Student’s t test; P > 0.05).
|
. None of the variants showed an increased Michaelis-Menten constant (Km), indicating impaired substrate binding, whereas maximum velocity (Vmax) was clearly reduced for P482S (Student’s t test; P = 0.002 for 17OHP; P = 0.02 for progesterone), but not for A15T (Student’s t test; P > 0.05).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A15T
The A15T variant was found in a chimeric gene originated by gene conversion carrying the CYP21P promoter and the P30L mutation. Two unrelated girls (subjects 1 and 2) with a classical form of CAH had this allele. Both showed significant virilization (Prader stage IV and III, respectively), but no evident salt loss.
The CYP21P promoter has been shown to reduce the transcription rate to 20% of normal (28, 29), and the P30L mutation results in an enzyme with 11–60% activity in vitro (30, 31). This latter mutation is usually associated with NC-CAH, but it has also been reported in some cases presenting the SV phenotype (32).
CYP21P/CYP21 hybrid genes containing the CYP21P promoter in combination with the P30L mutation have been described by L’Allemand et al. (33) in a subject with an early manifestation of NC-CAH with virilization onset at 12–16 wk of life and by Deneux et al. (34) in a girl with an especially pronounced NC-CAH phenotype.
We have observed another patient in our clinic, with a chimeric gene comprising the pseudogene promoter and P30L, with a classical form of CAH (data not shown).
In a recent Slovenian study (13), two patients were hemizygous for an allele with the CYP21P promoter and P30L mutation. One of these patients had SV-CAH, whereas the other was reported to have NC-CAH because she was diagnosed at 12.3 yr with clitoromegaly. In the same study, two patients were reported to be hemizygous for the same chimeric gene that we found. Both of these had the SW form of CAH, suggesting that the presence of the A15T mutation would lead to a further decrease in enzyme activity than the combination of the CYP21P promoter and P30L alone.
As A15T lies in the pseudogene-derived sequence, and since we found this alteration in four additional pseudogenes in unrelated control subjects, we decided to investigate the consequence of this substitution in a CYP21 gene without other mutations.
Ala 15 lies within the first hydrophobic domain of the protein that not only serves as a membrane targeting and anchoring domain, but is important for the in vivo stability (35, 36). The balance between the net charge of the flanking sequences of this hydrophobic segment and the length of the segment were demonstrated to determine the function of this topogenic sequence (37). The substitution of the nonpolar amino acid by an uncharged polar threonine, before a charged residue (Arg16), could be suspected to result in slightly reduced membrane incorporation and enzyme function compared with the wild-type form. On the other hand, alignment of the N-terminal with P450c21 and P450c17 from other species showed that the alanine 15 in the mouse P450c21 is substituted by a threonine and followed by an arginine, suggesting a minimal effect of this substitution.
We compared the activity of the A15T mutant and wild-type forms, and no significant difference could be identified in enzyme activity (substrate conversion grade, velocity, and substrate binding). Thus, our results do not support the concept that the presence of A15T causes a more severe phenotype than the pseudogene promoter and P30L alone. However, we cannot completely rule out the possibility of a slight functional difference between wt-P450c21 and A15T-P450c21 forms because these in vitro measurements of enzyme activity are crude. We have not either studied the possibility of synergism between A15T and P30L.
P482S
By CYP21 gene analysis, we identified the new mutation P482S in the C-terminal part of P450c21 in seven subjects analyzed for different reasons (NC-CAH, precocious pubarche, menstrual irregularities, or hypertrichosis). Four subjects (subjects 5–7) were suspected to be heterozygous carriers suggested by the ACTH test values, and one (subject 4) had a borderline value between that of heterozygous carriers and patients affected with NC-CAH. The P482S mutation was the only alteration from the normal sequence we identified in three of the heterozygous carriers, whereas the other one (subject 5) had on the other allele the Intron 2 splice mutation. Subject 4 and the two siblings with an ACTH test suggestive for a NC-form (subjects 3a and 3b) presented the V281L mutation on the other allele, but only the two females (subjects 3a and 4) presented CAH symptoms. Thus, P482S appears to be associated with a very mild form of NC-CAH.
We found that P482S impaired the enzyme activity to approximately 70% for both of the natural substrates, and determination of apparent kinetic constant revealed that the substrate-binding capacity was of the same magnitude for this mutant and wild-type protein, whereas the Vmax was decreased for the mutant. Another Pro to Ser substitution at codon 453, associated with a NC form of CAH, was previously reported to reduce the enzyme activity to 20–68% (19, 38). The apparent higher activity of the P482S mutant compared with P453S is in agreement with the phenotype shown by our patients, who displayed very slight signs of impaired adrenocortical function.
The topological structure of the prokaryotic cytochrome P450BM-3 demonstrated that the C-terminal part is a separate domain without contacts to the rest of the P450BM-3 protein, and little is known about its function in P450c21. P482 lies 12 amino acids from the C-terminal end and is located in a hydrophilic region of the enzyme conserved among 21-hydroxylases of different species, and a mutation affecting the adjacent residue 483 (R483P) has been reported to severely impair enzyme activity (39). The three-dimensional model study (40), based on prokaryotic P450s of known structure, did not consider the R483P mutation and the common N493S polymorphism. Studies of other P450c21-mutants lying in this region suggest that single-amino acid substitutions of the carboxy-terminal region change the three-dimensional structure and may affect enzyme stability (41).
Three common missense mutations associated with NC-CAH are described: V281L (42), P453S (19, 38), and P30L (with some exceptions) (29, 30) ranging from 11–68% in activity in vitro. Three other rare missense mutations have been reported to cause a mild reduction of enzyme activity (20–64%): P105L (19), R339H (38), and V304M (43). Three other mutations, R341W, L317M, and R435C, have been reported as responsible for NC-CAH, but no functional studies have been performed (34, 44).
HLA analysis revealed that the promCYP21P-A15T-P30L allele was associated with the haplotype B13 CW6 DR7 DQ2, suggesting a gene founder effect. The P482S alleles were associated with the DR4 and partially with the B41 and A2, suggesting a founder affect followed by recombination events in the HLA locus. Similar founder effects and mutation spreading were previously observed within the Brazilian population (45). Although microconversion events are the main cause of mutations in the CYP21 gene and rare or new mutations usually are reported in a single patient or family, mutations with a common origin can also be the cause of 21-hydroxylase deficiency and be spread in the population.
In conclusion, our findings extend the arsenal of molecular lesions known to cause CAH and further emphasize the widely varying degrees of enzyme impairment that can result from single amino acid substitutions in the 21-hydroxylase protein. The P482S mutation should be added to the list of mutations that are analyzed in cases where the mildest form of CAH are suspected, especially in subjects of Italian ancestry. The determination by functional studies of mutated enzyme activity is useful not only to better understand the function/structure relationship of the protein. The elucidation of functional consequences of new mutations that are spreading in the population is also helpful to enable individualized treatment, and make possible accurate genetic counseling to patients and their families.
|
Footnotes |
|---|
Abbreviations: 4-A, 4-Androstenedione; BA, bone age; BL, birth length; BW, birth weight; CAH, congenital adrenal hyperplasia; HCASD, Height for Chronological Age Standard Deviation Score; NC-CAH, nonclassical CAH; 17OHP, 17-hydroxyprogesterone; P450c21, 21-hydroxylase; SV-CAH, simple-virilizing CAH; Vmax, maximum velocity.
Received September 18, 2003.
Accepted January 27, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Balestrieri, L. Zirilli, B. Madeo, E. Pignatti, G. Rossi, C. Carani, and V. Rochira 21-Hydroxylase Deficiency and Klinefelter Syndrome in an Adult Man: Striking a Balance Between Androgen Excess and Insufficiency J Androl, November 1, 2008; 29(6): 605 - 609. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Robins, J. Carlsson, M. Sunnerhagen, A. Wedell, and B. Persson Molecular Model of Human CYP21 Based on Mammalian CYP2C5: Structural Features Correlate with Clinical Severity of Mutations Causing Congenital Adrenal Hyperplasia Mol. Endocrinol., November 1, 2006; 20(11): 2946 - 2964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V Dolz;an, J Solyom, G Fekete, J Kovacs, V Rakosnikova, F Votava, J Lebl, Z Pribilincova, S. Baumgartner-Parzer, S Riedl, et al. Mutational spectrum of steroid 21-hydroxylase and the genotype-phenotype association in Middle European patients with congenital adrenal hyperplasia Eur. J. Endocrinol., July 1, 2005; 153(1): 99 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Robins, M. Barbaro, S. Lajic, and A. Wedell Not All Amino Acid Substitutions of the Common Cluster E6 Mutation in CYP21 Cause Congenital Adrenal Hyperplasia J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2148 - 2153. [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 |
