| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Studies |
Endocrine Unit, Choremis Research Laboratory, A’ Pediatric Department, Athens University Medical School, Aghia Sophia Children’s Hospital, Athens 11527, Greece
Address all correspondence and requests for reprints to: Dr. M. Dracopoulou-Vabouli, Endocrine Unit, Choremis Research Laboratory, A’ Pediatric Department, Athens University Medical School, Aghia Sophia Children’s Hospital, Athens 11527, Greece.
Abstract
Defective steroid synthesis due to 21-hydroxylase deficiency is the most common form of congenital adrenal hyperplasia. Knowledge of the molecular defects causing 21-hydroxylase deficiency in different populations is of both theoretical and practical interest. The types and the relative frequencies of molecular defects and the correlation between the genotype and the phenotype were examined in the Hellenic population. We searched for deletions, conversions, and 11 of the most frequent mutations of the CYP21 gene by Southern blot and allele-specific PCR in 222 chromosomes from 111 unrelated subjects and their parents. The most frequent molecular defects were 1) in the salt wasting form, I2 splice (42.9%), deletions and conversions (24.5%), and Q318stop (14.3%); 2) in the simple virilizing form, I172N (35.3%), I2 splice (29.4%), and P30L (19.1%); and 3) in the nonclassical form, V281L (41.1%), P30L (21.4%), and P453S (14.3%). Compared with other populations, Greek patients had a higher frequency of Q318stop in the salt-wasting form, of P30L in both simple virilizing and nonclassical forms and of P453S in the nonclassical form. The concordance of genotype to phenotype in the total sample was 87%. However, the concordance rate was different in the three forms of the disease. Thus, complete concordance was detected in the genotypes predicting the salt-wasting phenotype, a slightly lower concordance (95.2%) was detected in the genotypes predicting the simple virilizing phenotype, and the lowest concordance (67.6%) was observed in genotypes predicting the nonclassical phenotype. In conclusion, the concordance between genotype and phenotype decreases as the severity of the disease diminishes. This should be taken into consideration in genetic counseling and antenatal intervention.
DEFECTIVE STEROID synthesis due to derangements of the 21-hydroxylase gene (CYP21), constitutes the most frequent cause of congenital adrenal hyperplasia (CAH) and is one of the most frequent inborn errors of metabolism. The currently accepted classification of the different forms of CAH based on the clinical expression of the disease is the following: 1) the salt-wasting form (SW), 2) the simple virilizing form (SV), and 3) the nonclassical form (NC) (1, 2, 3). The CYP21 gene is located on the short arm of chromosome 6, within the region of the major histocompatibility complex, in proximity to the complement genes, and at a distance of 30 kb from a highly homologous (>95%) pseudogene, designated CYP21P. Both CYP21 and CYP21P contain 10 exons and 9 introns, covering a distance of 3 kb each. Despite their high homology, CYP21P is totally inactive (4, 5). The molecular defects of CYP21 may result from 2 types of recombinations between the CYP21 and the CYP21P pseudogene: unequal crossing-over during meiosis leading to deletion of CYP21 and conversions that result in transfer of altered sequences from CYP21P to CYP21, where they become detrimental (1, 4, 5, 6). The known defects of the CYP21 gene that result in the different forms of 21-hydroxylase deficiency are deletions, conversions, and more than 15 missense, nonsense, and frameshift mutations (4, 5, 6, 7).
The delineation of the types and frequency of the molecular defects of the CYP21 gene in different populations is both of theoretical and practical interest. Knowing the defects and the phenotypes produced is of help in detecting heterozygote carriers, in antenatal diagnosis, and in genetic counseling. We report the types and frequency of molecular defects in Greek patients with 21-hydroxylase deficiency as well as the relation of genotype to phenotype in the different forms of the disease.
Subjects and Methods
Subjects
DNA analysis was carried out in 222 chromosomes from 111 unrelated subjects of Hellenic origin with different forms of 21-hydroxylase deficiency. Informed consent was obtained from the parents. In 94% of the cases, both parents were also studied. Forty-nine of the index cases had the salt-wasting form, 34 had the simple virilizing form, and 28 had the nonclassical form of the disorder. The criteria for categorization were as follows: 1) salt-wasting form: initiation of signs and symptoms (virilization, failure to thrive, hyponatremia, hyperkalemia, acidosis, high renin values, and 17-hydroxyprogesterone (17-OHP) values higher than 90 nmol/L) in the neonatal period; 2) simple virilizing form: virilization of any degree without clinical evidence of salt loss, 17-OHP values higher than 90 nmol/L, normal renin values, and normal electrolytes at presentation or at follow-up. The mean age at presentation was 3.9 ± 2.1 yr; and 3) nonclassical form: premature appearance of pubic hair, severe acne or hypertrichosis at puberty, with or without menstrual disorders, and complete lack of virilization. The mean age at presentation was 8.2 ± 4.06 yr. The subjects with the nonclassical form had basal 17-OHP values greater than 15 nmol/L and/or 17-OHP values, after iv administration of 250 µg of ACTH-(1–24), greater than 36 nmol/L. Each of the subjects studied was clinically examined by one of the authors, and about 98% of the cases have been under our care.
Molecular analysis of the CYP21 gene
DNA was extracted from peripheral blood leukocytes using the phenol-chloroform method or the QIAGEN extraction kit (QIAGEN GmbH, Hilden, Germany). Southern blot analysis was employed for the detection of large deletions and conversions of the CYP21 gene. Genomic DNA was digested by the TaqI, KpnI, and EcoRI+BglII restriction enzymes. Digests were electrophoresed on agarose gels, transferred to nylon membranes, and hybridized with a radioactively labeled complementary DNA probe for 21-hydroxylase ([32P]complementary DNA pC21/3c) from American Type Culture Collection (Manassas, VA). After autoradiography, the density of DNA bands was determined using an automated densitometer. Allele-specific PCR was employed, as previously described by A. Wedell and H. Luthman (8), to identify the following molecular defects of the CYP21 gene: P30L (exon 1); I2 splice (intron 2); 8bpdE3 (exon 3); I172N (exon 4); clusterE6 (exon 6); V281L, G291S, and F306+T (exon 7); Q318stop and R356W (exon 8); and P453S (exon 10). For identification of 8bpdE3 in heterozygosity, the PCR primers proposed by Wilson et al. (9) were used. To detect more than one mutation, concomitantly present in the same allele, each DNA sample was tested for all herein described mutations.
Results
The type and frequency of the molecular defects detected in our
patients are depicted in Table 1
. In 14
of the 222 chromosomes (6.3%) delineation of the molecular defect was
not feasible, and these cases are designated undetected. In the
salt-wasting form 81.7% of the molecular defects detected were
distributed as follows: I2 splice mutation in
intron 2 (42.9%), Q318stop mutation in exon 8 (14.3%), and deletions
and large conversions (24.5%). In the simple virilizing form the most
frequent mutations were I172N in exon 4 (35.3%),
I2 splice in intron 2 (29.4%), and P30L in exon
1 (19.1%). These molecular abnormalities comprised 83.8% of the
derangements disclosed. In the nonclassical form the V281L mutation in
exon 7, the P30L mutation in exon 1, and the P453S mutation in exon 10
constituted 76.8% of the molecular defects (41.1%, 21.4%, and
14.3%, respectively).
|
. Genotypes 1–8, 12–14, and 21–28
were exclusively detected in the salt-wasting, simple virilizing, and
nonclassical forms, respectively. Genotypes 9–11 were found in both
salt-wasting and simple virilizing forms. Genotypes 15–20 were
detected in the expected form of the disease, namely nonclassical, but
also in the simple virilizing form, which was not expected. The
genotype to phenotype concordance rate in the total group (all forms
collectively) was quite high (87%). However, variable concordance was
found when analysis was made separately for each form of the disease
(salt-wasting, 100%; simple virilizing, 95.2%; nonclassical, 67.6%).
Twelve patients whose genotypes were not completely characterized were
not included in the estimation of the genotype to phenotype concordance
rate. Their phenotypes and incomplete genotypes are listed in Table 3.
|
|
Relative frequencies of molecular defects in the different clinical forms
Before comparing our data to those from other populations, it should be mentioned that in similar studies of other ethnic groups the proportions of the different forms of the disease (salt-wasting, simple virilizing, and nonclassical) varied in the total sample. This could explain some of the differences in the relative frequencies of molecular defects reported by different researchers. It appears that the separate calculation of the frequencies of the molecular defects for each clinical form of the disease is more useful than the analysis of an all-encompassing sample.
In our patient population the most frequent molecular defects in the salt-wasting form were the I2 splice mutation (42.9%) and large deletions and conversions of the CYP21 gene (24.5%), as reported previously in other ethnic groups (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). The mutation Q318stop (14.3%) had a higher frequency in our patients than in most other populations, but a rate similar to that observed in Italian, Argentinian, and Chilean samples (14, 16, 17). In the simple virilizing form, the dominant mutations were the I172N (35.3%) and the I2 splice (29.4%), as previously reported in other populations (14, 18), whereas the mutation P30L was detected in 19.1% of the simple virilizing chromosomes and in none of 50 control subjects of Hellenic origin (21). This high frequency of P30L in the simple virilizing form has not been previously reported. In the nonclassical form the prevalent mutation was the V281L (41.1%), as in all the populations studied (10, 12, 13, 14, 18, 22). However, the mutations P30L (21.4%) and P453S (14.3%) were detected in a remarkably higher frequency in our patients with the nonclassical form than in other ethnic groups. It is of interest that the mutations G291S, F306+T in exon 7 and the clusterE6 of three diverse mutations in exon 6, described in other populations, were not detected in any of our patients.
In our population study the percentage of undetectable mutations was 6.3%. In other studies it has ranged from 5–23% (9, 10, 13, 14, 17, 18, 22). Direct sequencing of the entire CYP21 gene as well as of the promoter and the 3'-untranslated regions might reveal novel disease-causing mutations.
Relation of genotype to phenotype
The clinical manifestation of 21-hydroxylase deficiency is determined by the remaining enzymatic activity, which is primarily, but not exclusively, related to the type of the molecular defect. In general, however, the genotypic abnormality of any genetic disorder does not always match the phenotypic expression. Some of the contributing factors in the variability of clinical expression of CYP21 defects may be the different abundance of extraadrenal 21-hydroxylase activity exerted by enzymes other than P450c21 or varying activities of the CYP21 promoter as well as factors that determine the sensitivity of target tissues to androgens (20, 23).
In our population when the data of the entire group were analyzed, the rate of concordance of genotype to phenotype was similar to that reported for other patient populations (10, 12, 13, 14, 15, 18, 20). Nevertheless, a variable concordance rate between genotype and phenotype emerged when a separate analysis for each of the three forms of the disease was carried out. More specifically, complete concordance was detected in the genotypes predicting the salt-wasting phenotype, whereas in the genotypes predicting the simple virilizing phenotype the concordance was 95.2%. In the latter case the rate of concordance was even lower if the genotypes I2 splice/I2 splice and I2 splice/deletion or conversion were considered discordant to the simple virilizing phenotype. The variability of phenotype (salt-wasting or simple virilizing) resulting from genotypes with the I2 splice mutation in homozygosity or in heterozygosity in combination with an inactivating mutation could be explained by differences in the expression of I2 splice, which has been reported by Higashi et al. (24). In genotypes predicting nonclassical expression of the disease, the concordance rate was much lower (67.6%); the phenotype was simple virilizing rather than nonclassical in 32.4% of the cases. Analogous data were recently reported by Krone et al. (25).
The lower concordance rate observed in genotypes predicting the nonclassical form in our study compared with other studies could result from differences in the clinical criteria of categorization. In our study the female subjects classified as having nonclassical CAH did not have any clitoral enlargement, which has not been the case in other studies (10, 12, 13, 15, 18, 22). The discrepancy might also be due to the presence of a second, unidentified, rare mutation in the same gene. It must be underlined that in 5 of 6 nonclassical genotypes (11 patients), associated with the simple virilizing phenotype, the mutation detected in at least 1 of the 2 chromosomes was P30L. This finding was also recently reported by Krone et al. (25). Thus, the postulation that the determination of 21-hydroxylase capacity by genotyping is more reliable than clinical classification (12) is proven valid for salt-wasting genotypes, but not for genotypes predicting simple virilizing and nonclassical CAH.
In conclusion, the variable concordance of genotype to phenotype in the different forms of 21-hydroxylase deficiency and specifically the high discordance rate in genotypes predicting the nonclassical form should be taken into consideration in genetic counseling and especially in a prospective or anticipated pregnancy.
Footnotes
1 This work was supported in part by a grant from the Central Council
of Health, Hellenic Ministry of Health. 
Received October 17, 2000.
Revised February 15, 2001.
Revised March 12, 2001.
Accepted March 18, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Grigorescu Sido, M. M. Weber, P. Grigorescu Sido, S. Clausmeyer, U. Heinrich, and E. Schulze 21-Hydroxylase and 11{beta}-Hydroxylase Mutations in Romanian Patients with Classic Congenital Adrenal Hyperplasia J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5769 - 5773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Pinto, V. Tardy, C. Trivin, C. Thalassinos, S. Lortat-Jacob, C. Nihoul-Fekete, Y. Morel, and R. Brauner Follow-Up of 68 Children with Congenital Adrenal Hyperplasia due to 21-Hydroxylase Deficiency: Relevance of Genotype for Management J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2624 - 2633. [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 |
