| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
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. Dacou-Voutetakis, Endocrine Unit, Choremis Research Laboratory, A' Pediatric Department, Athens University Medical School, Aghia Sophia Children’s Hospital, Athens 11527, Greece.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
4-androstenedione, dehydroepiandrosterone
sulfate, and 11-deoxycortisol were also determined. Bone age was
evaluated using the Greulich and Pyle atlas. Abnormal genotype was
detected in 45.8% of the studied subjects; 8.3% were homozygotes,
with genotypes concordant with the nonclassical phenotype of 21
hydroxylase deficiency, and 37.5% were heterozygotes for 9 different
molecular defects of the CYP21 gene. The children with no detectable
molecular defect were designated normal. The 60 min post-Synachten
values in homozygotes (17.9 ± 7.1 ng/mL) and heterozygotes
(7.1 ± 3.6 ng/mL) were significantly higher than that in normal
subjects (3.3 ± 1.5 ng/mL), but with significant overlapping of
values. The mean difference between bone age and chronological age
differed in the three groups with overlapping values. The basal
4-androstenedione level was lower in the normal subjects
(0.65 ± 0.3 ng/mL) than in those with abnormal genotype (1.1
± 0.8 ng/mL). The data indicate that the incidence of molecular
defects in PA is quite high. The CYP21 heterozygocity is clinically
expressed in some subjects prepubertally. In a significant number of
cases the genotype cannot be predicted by the age of onset of PA, the
mean difference between bone age and chronological age, or the results
of a Synachten test. Follow-up of these children through puberty is
imperative and may reveal the clinical significance of the molecular
defect, namely more hypertrichosis, intense acne, early puberty,
possible abnormal menses, and/or fertility problems in the affected. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the present study we searched for the presence of deletions, conversions, and 10-point mutations of the CYP21 gene in children presented with premature appearance of pubic hair in an effort to determine 1) their incidence in patients with PA and 2) the relation of the molecular defect to the results of ACTH stimulation test as well as pertinent clinical parameters.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study group was comprised of 48 unrelated children of Hellenic origin with premature adrenarche, 40 females and 8 males, who were presented to the endocrine clinic for early growth of pubic hair. They represented sequential cases with no selection related to the density of pubic hair or bone age (BA) advancement. The mean age of pubic hair development was 5.6 ± 1.3 yr. None of the girls had signs of virilization or breast development, and none of the boys had testicular volume greater than 3 mL. The BA was evaluated using the standards of Greulich and Pyle (11). Height SD score was calculated as follows: actual height minus mean height for chronological age divided by 1 SD of the height for age (12), using Tanner’s growth data.
Molecular analysis of the CYP21 gene
DNA was extracted from peripheral blood leukocytes using the Qiagen extraction kit (Qiagen, Chatsworth, CA).
Ninety-six chromosomes from the 48 subjects with PA were analyzed for deletions, conversions, and the following 10 of the most common mutations known to cause 21-hydroxylase deficiency, which also represent the most frequent mutations in our population: P30L (exon 1), I2 splice (intron 2), 8bpdelE3 (exon 3), I172N (exon 4), cluster E6 (exon 6), V281L (exon 7), F306+T (exon 7), Q318st (exon 8), R356W (exon 8), and P453S (exon 10). Our studies have shown that these 12 molecular defects cover 90% of the known mutations in the Hellenic population (data not shown). One hundred chromosomes from 50 blood donors were also analyzed for the 3 dominant mutations of the NC form of the disease in general and in the Hellenic population in particular: V281L, P30L, and P453S. The Southern blot technique 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 gel, transferred on nylon membranes, and hybridized with radioactively labeled complementary DNA probe for 21-hydroxylase (32 p-cDNA pC21/3c) from American Type Culture Collection (Manassas, VA). After autoradiography, the densities of DNA bands were determined using an automated densitometer.
Allele-specific PCR, after initial selection against pseudogenes, was used to identify smaller molecular defects of the CYP21 gene. The method proposed by Wedel and Luthman was followed with slight modifications (13).
Hormonal studies
The Synachten test was carried out by the iv
administration of 0.25 mg Synachten Ciba Laboratories, West Sussex,
United Kingdom. 17-Hydroxyprogesterone (17OHP) and cortisol were
determined at 0 and 60 min post-Synachten administration, whereas
4-androstenedione (4), 11-deoxycortisol
(compound S), and dehydroepiandrosterone sulfate (DHEAS) were measured
only at baseline.
17OHP was determined by RIA using commercially available reagents from ICN Pharmaceuticals, Inc. (intraassay variation, 9.1%; interassay variation, 13.6%).
Cortisol was determined by RIA, using commercially available reagents from Diagnostic Systems Laboratories, Inc. (intraassay variation, 8.3%; interassay variation, 9.8%).
4, DHEAS, and compound S were also determined using
commercially available reagents from Diagnostic Systems Laboratories, Inc.
Statistical analysis
For statistical analysis, the program STATISTICA for Windows, version 5.1 (1997) was used throughout together with a specially written module for the receiver operating curve (ROC) analysis (14, 15).
For the initial evaluation of the data, descriptive statistics were employed, and the means for the three groups were compared. Subsequent multivariate ANOVA was carried out to assess the statistical significance of the differences between normal and heterozygote subjects in the measurements of basal 17OHP, 60 min 17OHP, and the sum of these two values (16).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). Mutations V281L
and P453S were detected in 1 of 100 chromosomes studied in unselected
general population each, whereas mutation P30L was not detected in any
of the chromosomes studied (Table 2).
Children with PA and no detectable molecular defect were designated
normal, whereas normal children (n = 16) of comparable age
(5.9 ± 2.3 yr) without PA were designated controls. The basal
values of 17OHP (Table 3) in normals, heterozygotes, homozygotes for
21-hydroxylase deficiency, and controls were 0.79 ± 0.48,
1.01 ± 0.56, 3.5 ± 1.7, (F = 29.86; df 2,45;
P = 0.000), and 0.45 ± 0.2 ng/mL, respectively.
The 17OHP values 60 min post-Synachten administration (Table 3) in homozygotes (17.9 ± 7.1
ng/mL) and heterozygotes (7.1 ± 3.6 ng/mL) were significantly
higher than those in normal subjects (3.3 ± 1.5 ng/mL; F =
18.8; df 2,39; P = 0.000). It must be noted, however,
that there was significant overlapping of values (Fig. 1).
|
|
|
|
BA-CA) in the
three groups of children with PA differed significantly (F = 3.6;
df 2,42; P = 0.035), but overlapping values were also
observed (normals, 1.03 ± 1.1 yr; heterozygotes, 1.4 ± 1.15
yr; homozygotes, 3.2 ± 0.6 yr; Fig. 2).
|
).
The basal values of 4 in normal subjects (0.64 ± 0.3 ng/mL)
were lower than those in heterozygotes (0.97 ± 0.6 ng/mL) and
homozygotes (1.9 ± 1.4 ng/mL; F = 4.6; df 2,33;
P = 0.017), but there were high percentage of
overlapping values (Table 3). The basal values of compound S did not
differ in the three groups (Table 3). At the time of presentation pubic
hair development ranged from Tanner stage II to III (Table 3). The
height was above the mean for our population but did not differ in the
three groups (height SD score, 1.05 ± 0.99 in
normals, 0.8 ± 1.1 in heterozygotes, and 1.2 ± 0.8 in
homozygotes; Table 3).
We examined the probability of predicting heterozygocity by means of
clinical and biochemical indicators, namely age of presentation,
BA-CA, basal steroid values, and 60 min and sum values of 17OHP in
the Synachten test. Multiple ANOVA was used to assess statistical
significance between normals and heterozygotes. This process identified
the basal and 60 min values as well as the sum of basal plus 60 min
values of 17OHP as the most likely predictors. The overall index was
significant (Wilk’s 
= 0.158; df 3,34; P = 0.001).
In the light of the above results a ROC curve was drawn for basal, 60
min, and sum of basal plus 60 min 17OHP values to further investigate
the relative strength of each measurement in correctly identifying
heterozygotes in a mixed population. As it can be seen from the graph
(Fig. 3), the sum of basal plus 60 min
17OHP values is the most potent of the three. Thus, for the sum of
basal plus 60 min values, the cut-off point is 4.9 ng/mL, over which
there is a 76.5% certainty of heterozygocity.
|
), the sum value gave
the lowest probability of a false positive prediction.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The incidence of molecular defects detected in the present study is quite impressive; 45.8% of the subjects had abnormal genotypes. Of these, 8.3% were homozygotes for gene defects concordant with the NC form of 21-hydroxylase deficiency, whereas 37.5% were heterozygotes for various CYP21 gene abnormalities. It must be underlined that 50% of abnormal alleles in the heterozygotes are represented by the V281L mutation, which is the most frequent molecular defect detected in the NC form of 21-hydroxylase deficiency in many populations.
The frequency of carriers for 21-hydroxylase deficiency in the non-Jewish Caucasian population is estimated to be 7% (17, 18), whereas the frequency of carriers in our children with PA was 37.5%. The frequency of carriers for 21-hydroxylase deficiency in Hellenic population has not been estimated to date. Nevertheless, the frequency of carriers for the three mutations most frequently encountered in NCCAH (V281L, P30L, and P453S) was much lower in the general population than in our children with PA.
The present findings are in agreement with certain published reports, which were based on ACTH studies and have indicated a high incidence of heterozygocity of CAH in children with PA (1, 2, 7). Hence, the present findings prove beyond any doubt that alterations of the CYP21 gene in subjects with PA are quite high.
The most important questions raised by our results are the following. 1) Is the molecular defect detected related to the clinical phenotype? 2) Can the molecular defect be predicted by clinical and biochemical findings? 3) Has this finding any clinical relevance? 4) Would this information be useful to the individual?
Concerning the first question, our inclination will be that the molecular defect is related to the clinical expression of PA, but no proof can be provided. On the second question of anticipating the defect based on biochemical and/or clinical findings the answer is positive for most, but not all, cases.
The proposed cut-off point of 10 ng/mL after Synachten administration
was also useful in separating the three homozygous NC cases from the
heterozygotes and the normal subjects. As a group, the normal subjects
differ from the affected subjects in parameters such as BA-CA, basal
4, basal 17OHP, 60 min post-Synachten 17OHP, and the sum
of basal plus 60 min 17OHP values. Nevertheless, in a number of cases
overlapping values made the distinction doubtful. Such discordance
between molecular and hormonal diagnosis of molecular deviation of the
CYP21 gene have been noted by others (19).
On the basis of our data, slightly increased basal 4values and BA-CA greater than 1 are findings highly
suggestive of the presence of a molecular defect in the CYP21 gene. The
application of a ROC curve revealed that the sum of basal plus 60 min
17OHP values is the most potent indicator of heterozygocity, with a
cut-off point of 4.9 ng/mL over which there is 76.5% certainty of
heterozygocity.
By using the nomogram of 17OHP baseline and 60 min ACTH stimulation tests (20), the values of heterozygotes fell in the expected area, but the majority of normal values also fell into the same area.
The third question is more difficult to answer, namely the clinical relevance of the detected molecular defect. More specifically, do these girls have a higher probability of having earlier onset of gonadarche, hirsutism or severe acne at puberty, menstrual disturbances, fertility problems, or compromised final height (19)?
The available data on obligate heterozygotes, retrospectively collected, are not sufficient, and it seems that only long term prospective data will definitely answer these questions. Thus, the study of 38 obligate heterozygote females by Knochenhauer et al. (21) showed no increased incidence of clinically evident hyperandrogenism in these women. It must be stressed, however, that one third of these women were premenopausal and/or receiving hormonal therapy. They also had a higher mean total and free testosterone levels and lower circulating levels of sex hormone binding globulin.
Ibanez et al. (22) evaluated pubertal development and final height in girls with PA. They found no difference in final height or menarche. Nevertheless, this was not a random selection, as NC CAH due to 21-hydroxylase deficiency had been ruled out in all subjects by means of an ACTH test.
Obviously, a prospective study of girls studied biochemically and by
DNA analysis will give a definitive answer to the very important
question of the evolution of cases with PA and CYP21 heterozygocity. It
is quite possible that a number of heterozygotes have a certain
clinical expression of the genetic defect. Until more data are
available, we believe that girls with PA who have normal basal values
of adrenal androgens and increased BA-CA should be followed and
reevaluated at puberty. The value of such a finding for genetic
counselling is quite obvious, although the validity of delivering such
information at such an early stage is questionable.
|
Acknowledgments |
|---|
Received August 5, 1998.
Revised February 4, 1999.
Accepted February 8, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. Utriainen, R. Voutilainen, and J. Jaaskelainen Continuum of phenotypes and sympathoadrenal function in premature adrenarche Eur. J. Endocrinol., April 1, 2009; 160(4): 657 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. I. New Nonclassical 21-Hydroxylase Deficiency J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4205 - 4214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez, K. K. Ong, N. Mongan, J. Jaaskelainen, M. V. Marcos, I. A. Hughes, F. de Zegher, and D. B. Dunger Androgen Receptor Gene CAG Repeat Polymorphism in the Development of Ovarian Hyperandrogenism J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3333 - 3338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Lajic, S. Clauin, T. Robins, P. Vexiau, H. Blanche, C. Bellanne-Chantelot, and A. Wedell Novel Mutations in CYP21 Detected in Individuals with Hyperandrogenism J. Clin. Endocrinol. Metab., June 1, 2002; 87(6): 2824 - 2829. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. P. Merke, S. R. Bornstein, N. A. Avila, and G. P. Chrousos Future Directions in the Study and Management of Congenital Adrenal Hyperplasia due to 21-Hydroxylase Deficiency Ann Intern Med, February 19, 2002; 136(4): 320 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dracopoulou-Vabouli, M. Maniati-Christidi, and C. Dacou-Voutetakis The Spectrum of Molecular Defects of the CYP21 Gene in the Hellenic Population: Variable Concordance between Genotype and Phenotype in the Different Forms of Congenital Adrenal Hyperplasia J. Clin. Endocrinol. Metab., June 1, 2001; 86(6): 2845 - 2848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Adrenal Hyperandrogenism in Children J. Clin. Endocrinol. Metab., December 1, 1999; 84(12): 4431 - 4435. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
