doi: 10.1111/ahg.

12083

The Frequency and the Effects of 21-Hydroxylase Gene

Defects in Congenital Adrenal Hyperplasia Patients
Deniz Kirac1 , Ahmet Ilter Guney2 , Teoman Akcay3 , Tulay Guran4 , Korkut Ulucan5 , Serap Turan4 ,
Deniz Ergec2 , Gulsah Koc2 , Fatih Eren6 , Elif Cigdem Kaspar7 and Abdullah Bereket4
1
Yeditepe University, Faculty of Medicine, Department of Medical Biology, Istanbul, Turkey
2
Marmara University, Faculty of Medicine, Department of Medical Genetics, Istanbul, Turkey
3
Bakrkoy Dr. Sadi Konuk Hospital, Department of Pediatric Endocrinology, Istanbul, Turkey
4
Marmara University, Faculty of Medicine, Department of Pediatric Endocrinology, Istanbul, Turkey
5
Marmara University, Faculty of Dentistry, Department of Medical Biology and Genetics, Istanbul, Turkey
6
Marmara University, Faculty of Medicine, Department of Medical Biology, Istanbul, Turkey
7
Yeditepe University, Faculty of Medicine, Department of Biostatistics, Istanbul, Turkey

Summary
Congenital adrenal hyperplasia (CAH) is a group of genetic endocrine disorders, caused by enzyme deficiencies in the
conversion of cholesterol to cortisol. More than 90% of the cases have 21-hydroxylase deficiency (21-OHD). The clinical
phenotype of the disease is classified as classic, the severe form, and nonclassic, the mild form. In this study, it was
planned to characterize the mutations that cause 21-OHD in Turkish CAH patients by direct sequencing and multiplex
ligation-dependent probe amplification (MLPA) analysis and to investigate the type of CAH (classic or nonclassic type)
that these mutations cause. A total of 124 CAH patients with 21-OHD and 100 healthy volunteers were recruited to
the study. Most of the mutations were detected by direct sequencing. Large gene deletions/duplications/conversions
were investigated with MLPA analysis. Results were evaluated statistically. At the end of our study, 66 different variations
were detected including SNPs and deletions/duplications/conversions. Of these variations, 18 are novel, of which three
cause amino acid substitutions. In addition, 15 SNPs which cause amino acid changes were identified among these
variations. If similar results are obtained in different populations, these mutations, in particular the novel mutation 711
G>A, may be used as markers for prenatal diagnosis.
Keywords: CYP21A2, CAH, 21-OHD, direct sequencing, MLPA analysis

Introduction
Congenital adrenal hyperplasia (CAH) is a genetic, endocrine disorder in steroid biosynthesis, mainly caused by
enzyme deficiencies in the conversion of cholesterol to cortisol (Nimkarn & New, 2007). More than 90% of the cases
have 21-hydroxylase enzyme deficiency (21-OHD). This
enzyme is a cytochrome P450 enzyme that converts 17hydroxyprogesterone to 11-deoxycortisol and progesterone

Corresponding author: Deniz Kirac, Yeditepe University, Faculty of Medicine, Department of Medical Biology, 6th Floor,
Room Number: 1030, 34755, Kayisdagi-Atasehir/Istanbul, Turkey.
Tel: 00902165780568/00905424855292; Fax: 00902165780575;
E-mail:dyat@yeditepe.edu.tr


C

2014 John Wiley & Sons Ltd/University College London

to deoxycorticosterone (White, 2001). In 21-hydroxylase deficiency, the aldosterone and cortisol pathways are blocked
and the androgen pathway, which does not involve 21hydroxylation, is overstimulated (New, 2004).
The clinical phenotype of the disease is classified as classic,
the severe form, and nonclassic, the mild form. Classic CAH
is subclassified as salt-wasting (SW) and simple virilizing (SV),
due to the degree of aldosterone deficiency (Merke & Bornstein, 2005). The most severe form of the disease is the SW
type. In this type, most patients cannot synthesize sufficient
aldosterone to maintain sodium balance and may develop fatal salt wasting crises. In these patients, cortisol synthesis is
also insufficient, resulting in stimulation of corticotropin-releasing
hormone (CRH) and adrenocorticotropic hormone (ACTH).
As a result, the adrenal glands become hyperplastic. Excess

Annals of Human Genetics (2014) 78,399409

399

D. Kirac et al.

production of adrenal androgens cause ambiguous genitalia in

females (White & Speiser, 2000; Nimkarn & New, 2007). Affected males generally appear normal at birth (Keegan, 2001).
In SV type, patients have sufficient aldosterone production,
have signs of prenatal virilization, and do not have SW. In
the nonclassic (NC) type, patients do not have cortisol deficiency, but generally later in childhood or in early adulthood they have manifestation of hyperandrogenism (Merke
& Bornstein, 2005). The worldwide incidence of classic 21OHD is about 1 in 15,00020,000 live births, and the allelic
variant of nonclassic 21-OHD is about 1 in 100 persons based
on population genetic studies in the general Caucasian population (Sugino et al., 2006).
CAH due to 21-OHD is a monogenic autosomal recessive disorder (Wilson et al., 2007). The structural gene
encoding 21-hydroxylase, human CYP21A2, and a pseudogene, CYP21A1P, are located in the human leukocyte
antigen (HLA) major histocompatibility complex (MHC)
on Chromosome 6p21.3. Human CYP21A2 has a molecular mass of approximately 52 kDa and contains 494 amino
acid residues. CYP21A2 and CYP21A1P each contain 10
exons spaced over 3.1 kb. Their nucleotide sequences are
approximately 98% identical in exons and 96% identical in
introns (White & Speiser, 2000). These genes are included
in the RCCX module, which is also composed of a gene
encoding a serine/threonine kinase (RP), the gene encoding
serum complement 4 (C4), and tenascin X (TNX) genes, all
arranged tandemly as RP1-C4A-CYP21A1P-TNXA-RP2C4B-CYP21A2-TNXB (Friaes et al., 2006; Robins et al.,
2007). This region of Chromosome 6 (p21.3) is highly variable and is characterized by mutations and polymorphisms, as
well as variation in gene size and gene copy number (Robins
et al., 2007). The high degree of sequence similarity (96%
98%) between CYP21A2 and CYP21A1P allows two types of
recombination events: (1) unequal crossing-over during meiosis, which results in large deletions/duplications of CYP21A2
and the possible transmission of a null allele; (2) gene conversion events that transfer deleterious mutations present in
the pseudogene to CYP21A2 (Wilson et al., 2007). Although
hormonal parameters are important in classifying the types
of 21-OHD, molecular genetic techniques are best for making a secure diagnosis, as 90%95% of the allelic mutations
are detected by these techniques. More than 100 mutations
have been described, including point mutations, large and
small deletions, small insertions, and complex rearrangements
of the gene (Nimkarn & New, 2007). About 65%75% of
21-OHD patients are compound heterozygous for diseasecausing mutations (Krone et al., 2000).
In individuals with clinical symptoms or biochemical values indicative of CAH, and their families, genetic counseling
is necessary. Complete hormonal and molecular genetic analysis must be made in the index case and in the parents to

400

Annals of Human Genetics (2014) 78,399409

exclude or confirm a genetic change related to 21-OHD

(Trakakis et al., 2009). Also, the availability of prenatal diagnosis and treatment for 21-OHD may encourage parents with
severe mutations to request genetic counseling and carrier detection. Basic molecular screening of common mutations is
useful in the diagnosis of 21-OHD (Ezquieta et al., 2006).
Prenatal diagnosis of 21-OHD has been utilized for over 10
years, in particular by screening nine known mutations in the
CYP21A2 gene. However, the frequency of these mutations
may vary in different populations, and furthermore there may
be some other common mutations which cause 21-OHD
(Mao et al., 2002). In the present study, we aimed to detect
the mutations that cause 21-OHD in Turkish CAH patients
by direct sequencing and multiplex ligation-dependent probe
amplification (MLPA) analysis and to investigate the type of
CAH (SW, SV, or NC type) that these mutations cause.

Materials and Methods

Subjects and DNA Sample Preparation
A total of 124 patients who attended with several complaints
to Marmara University Pediatric Endocrinology Department
and Bakrkoy Mother and Child Health Education Center
and were diagnosed as 21-hydroxylase deficiency, between
2007 and 2012, were included in the study. One hundred
healthy volunteers were studied as a control group. The experimental protocol was approved by the Ethical committee
of Marmara University (MAR-YC-2007-0251).
The diagnosis was established after clinical and endocrinological evaluation. The SW form was defined as virilized
external genitalia in females, adrenal crisis (dehydration and
hyponatremia), high levels of serum 17-hydroxyprogesterone
(17OHP) and plasma renin activity, and increased urinary
sodium. The SV form was defined as variable degree of
ambiguous external genitalia in females, without SW, acceleration of growth and bone age, and elevated serum concentration of 17OHP before and after stimulation with intravenously administered ACTH. The NC form was defined by
normal genitalia, menstrual disturbance, infertility, hirsutism,
and mildly elevated levels of serum 17OHP before and after
stimulation with ACTH. In our study, 98 patients presented
the classic forms of the disease (75 SW and 23 SV) and 26
the NC form. Sixty-seven of these patients were females and
57 were males. Informed consent for mutation analysis was
obtained from all patients.
Total genomic DNA was extracted from peripheral blood
leukocytes collected from each subject, into EDTA-tubes, using the High Pure PCR Template Preparation Kit (Roche,
Basle, Switzerland), according to manufacturers instructions
(Friaes et al., 2006). Mutations were numbered 118 bp


C

2014 John Wiley & Sons Ltd/University College London

CYP21A2 Gene Defects in CAH Patients

before the start codon as the beginning for mutation numbering, as well as with cDNA numbering by using Human Genome Variation nomenclature (http://www.hgvs.
org/mutnomen/).

PCR Amplification
CYP21A2 was amplified by polymerase chain reaction (PCR)
using 50100 ng of total DNA. Specific primers were selected
for the amplification of the functional gene (CYP21A2) as
there is a pseudogene (CYP21A1P) with 98% homology.
Table S1 lists the sequences of primers specific for human
CYP21A2. PCR amplifications were performed in a total
volume of 50 l containing 50100 ng DNA template in 10
mM TrisHCl (pH 8.0), 50 mM KCl, 1.5 mM MgCl2 , 100
mM each of dNTPs, 1.0U Taq DNA polymerase, and 1.0
mM of each primer. The conditions of PCR amplification
were as follows: a denaturation step at 95 C for 3 min followed by 35 cycles at 95 C for 1 min, 59 C for 1 min, 72 C
for 1 min, a final extension at 72 C for 5 min, and a stop at
4 C. All PCR products were fractionated by electrophoresis
on a 2% agarose gel. The primers used for the amplification
of CYP21A2 and fragment sizes of the amplified gene are
shown in Table S1.

Purification of PCR Products and Direct

Sequencing of CYP21A2
All PCR products were purified by using the High Pure PCR
Product Purification Kit (Roche), according to the manufacturers instructions before direct sequencing. Then, purified PCR products were sequenced by using the DYEnamic
ET Terminator Cycle Sequencing Kit (Amersham, Buckinghamshire, UK) in ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA).

MLPA Analysis
To date, large CYP21A2 rearrangements have been mainly
detected by Southern blot analysis (White et al., 1988; Lobato et al., 1998) although PCR-based amplification analyses
have been developed and described (Lee et al., 2005). In this
paper, we report the results of an optimized protocol for an
MLPA assay, capable of obtaining easy and rapid detection of
deletions/duplications/conversions in the CYP21A2 gene.
The MLPA CYP21A2 kit is commercially available from
MRC Holland, Amsterdam, The Netherlands (SALSA MLPA
probemix P050-B3 CAH). The P050 CAH probemix is designed to detect large deletions, duplications, and conversions
in the CYP21A2, C4, and TNXB genes on 6p21.3. This


C

2014 John Wiley & Sons Ltd/University College London

P050-B3 CAH probemix contains five probes for CYP21A2

(exons 1, 3, 4, 6, and 8); among these are wild-type probes for
the 8 bp deletion, I172N, Cluster E6, and Q318X mutation.
Furthermore, this P050-B3 CAH probemix contains three
CYP21A1P-specific probes, three TNXB probes, one C4A
probe, one C4B probe, and one probe for the ATF6B gene located on q-telomeric of TNXB. In addition, two other probes
located on chromosome 6p21.3, one Y-chromosome specific
probe (UTY gene), and 16 reference probes are included. In
MLPA analysis, 50200 ng DNA was denatured (5 min at
98 C) and hybridized overnight at 60 C with the SALSA
probemix. The samples were then treated with Ligase-65 enzyme for 15 min at 54 C, and the reactions were stopped
by incubation at 98 C for 5 min. Finally, PCR amplification
was carried out with the specific SALSA PCR primers for
35 cycles (95 C for 30 s; 60 C for 30 s; 72 C for 1 min).
Amplification products were run on an SEQ8000 Genetic
Analyzer (Beckman Coulter Inc., Fullerton, CA, USA). The
setting of the capillary electrophoresis was: capillary temperature: 50 C; denaturation: 90 C for 120 s; injection time: 2.0
kV for 30 s and runtime: 60 min at 4.8 kV. The raw data were
analyzed by using Coffalyser 7.0 Software (MRC Holland,
Amsterdam, The Netherlands) (Concolino et al., 2009). The
peaks obtained after capillary electrophoresis could easily be
identified and assigned to specific probes on the basis of their
different lengths. All peaks matched with the expected sizes,
with a deviation up to 3 bases. Each result was repeated and
confirmed by three independent experiments.

Statistical Analysis
SPSS 21.0 was performed for statistical analysis. The 2 and
Fishers exact test were used for comparing nucleotide variations between patient and control groups. Correlations were
analyzed with the Spearman correlation coefficient test. The
genotype differences of disease types were analyzed with the
2 test. In addition, allele frequencies were compared between patient and control groups as well as between disease
types with 2 test. P-values less than 0.05 (P < 0.05) were
considered to be statistically significant.

Results
The results of direct sequencing and MLPA analysis
demonstrated many point mutations and large gene deletions/duplications/conversions in the CYP21A2 gene. At the
end of the study, 66 different variations were detected including SNPs and deletions/duplications/conversions. Of these
variations, 18 are novel, of which three cause amino acid substitutions. In addition to these, 15 SNPs which cause amino
acid change were identified in these variations. Mutations

Annals of Human Genetics (2014) 78,399409

401

402

Annals of Human Genetics (2014) 78,399409


C

Promoter
Exon 1
Exon 1
Exon 1
Exon 1
Exon 1
Exon 1
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Intron 2
Exon 3
Exon 3

115 C>T (c.-4C>T)

120 T>C (c.2T>C)

133 C-del (c.15delC)

135 T-del (c.17delT)
136 G-del (c.18delG)
236 T>C (c.118T>C)
256 A>C(c.138A>C)
516 T>C (c.386+9T>C)
540 A>C (c.386+33A>C)
574 C>T (c.386+67C>T)
641 C>T (c.386+134C>T)
681 G>C (c.669109G>C)
686 G-del (c.669104delG)
710 G>A (c.66980G>A)
711 G>A (c.66979G>A)
716 G>A (c.66974G>A)
723 C>G (c.66967C>G)
723 C>A (c.66967C>A)
742 A>G (c.66948A>G)
746 G>T (c.66944G>T)
751 C>G (c.66939C>G)
752 A>G (c.66938A>G)
777 A/C>G (c.66913A/C>G)
783 C>G (c.6697C>G)
786 G>A (c.6694G>A)
805 G>A (c.687G>A)
829836 (8 bp del)
(c.711_718delGAGACTAC)
982 C>T (c.826+38C>T)
1122 T>A (c.1004T>A)
1227 C>A (c.112415 C>A)
1234 T>C (c.11248 T>C)
1244 C>G (c.1126C>G)
1326 G>A (c.1208G>A)
1345 T>G (c.1225+2T>G)
1348 G>C (c.1225+5G>C)
Intron 3
Exon 4
Intron 4
Intron 4
Exon 5
Exon 5
Intron 5
Intron 5

Region

Nucleotide
positions and
variations (cDNA
numbering)

Silent
Silent

Affects splicing

K102R
Generates a stop
codon

I172N

D183E
V211M

Start codon
location changes
Leu deletion

Causes of
variations

Table 1 Localization, types, and P-values of variations in both of the groups.

0.769
1
0.131
0.197
0.197

0.02
0.443
1

0.89
0.89
0.89

0.029

0.006
0.516
0.368
0.202
0.504
0.226
0.480
1

0.003
0.149
0.206
0.583
0.398
0.248
0.360
0.398
n.c.
1
1
0.162

0.001

0.13
1

Comparison
of patient
and control
groups

0.856
1
0.378
0.468
0.468

0.023
0.257
1

0.89
0.89
0.89

0.001
0.329
0.625
0.148

0.014
0.504
0.056
0.408
1

2.1107

0.025
0.138
0.209
0.081

0.048
0.133
0.081

5.311021
1
1

0.028

0.001

0.13
1

Comparison of
patient and control
groups according to
the allele frequencies

0.018
0.15

0.002
n.c.

0.009

0.016
0.857
0.719

0.753
0.753
0.753
0.358
0.883
0.066
0.541
0.547
0.515

0.002
0.208
0.719
0.092
0.15
0.066
0.381

0.013

0.013

0.013

0.013

0.005
0.15
0.112
0.948
n.c.

0.271
0.109

Comparison
of disease
types

P-values

1.13108
0.151
0.081
0.065
0.065

0.022
0.363
0.72

0.753
0.753
0.753
0.358
0.883
0.559
0.632
0.682
0.515

2.26107
0.208
0.719
0.021
0.151

0.001
0.098

0.001

0.001

0.001

0.001

1.11107
0.151
0.112
0.948
n.c.

0.271
0.109

Comparison of
disease types
according to the
allele frequencies

2014 John Wiley & Sons Ltd/University College London

(Continued)

rs76565726
rs6475
rs58130573
rs145056075
rs1040310
novel
novel
novel

novel
novel
novel
rs6468
rs6464
rs6462
rs147805074
rs6449
rs144421484
novel
rs41315224
novel
novel
rs6450
rs6451
rs6451
rs59064806
rs6453
rs35147842
rs58256870
rs6467
novel
novel
rs6474
known

novel
novel

Reported
in pubmed
and other
databases

D. Kirac et al.


C

2014 John Wiley & Sons Ltd/University College London

Intron 5
Intron 5
Intron 5
Exon 6
Exon 6
Exon 6
Exon 6
Exon 6
Intron 6
Intron 6
Intron 6
Exon 7
Exon 7
Exon 7
Exon 7
Exon 7
Exon 7
Intron 7
Exon 8
Exon 8
Exon 9
Intron 9
Intron 9
Intron 9
Exon 10
Exon 10
Exon 10
Exon 10
Exon 10
3UTR

Region

Silent
I236N
V237E
E238K
M239K

Silent
P267L
S268T
Silent
V281L
L307F (F306+1nt.)

R339H
R356W
Silent

P453S
R483P
Silent
S493N
G494P

Complete or partial
enzyme
deficiency

Causes of
variations
0.394
0.599
0.134
0.061

0.029
0.502
0.727

0.001
1
0.465
0.117

3.7106
0.846
0.514
0.571
0.086
0.667

0.005
1

4.7105
0.667
1
0.655
0.306

0.034
0.465

0.025

0.026
1
0.255

n.c.

Comparison
of patient
and control
groups
0.401
0.745
0.137
0.113
0.066
0.556
0.73

0.02
1
0.465
0.122
0.935
0.753
0.44
0.498
0.299
0.667
0.162
1

0.049
0.775
1
0.661
0.272

0.036
0.468
0.991
0.642
1
0.257

n.c.

Comparison of
patient and control
groups according to
the allele frequencies
0.095

0.006
n.c.

0.019

0.013

0.004
0.523

0.002
0.996
0.996

0.038

0.004
0.082
0.301

0.007
n.c.
0.336

0.001
0.109
0.428
0.940
0.54
0.103

0.019
0.054
0.078

0.003

0.009
0.719
0.366

0.001

Comparison
of disease
types

P-values

0.104
0.129

0.001
0.159
0.128
0.083
0.53
0.063
0.996
0.996

0.042

0.006
0.1
0.096

0.008

0.025
0.336
0.113
0.11
0.693
0.987
0.543
0.111

0.023
0.058
0.082

0.035
0.105
0.719
0.371

0.001

Comparison of
disease types
according to the
allele frequencies

novel
rs12525076
novel
rs72552752
rs151344502
rs12530380
novel
rs6476
rs6458
rs6459
rs6465
rs6477
rs61732108
rs6472
rs11970671
rs6471
rs397515532
rs142027727
rs72552754
rs7769409
rs6469
rs182037914
rs77390388
novel
rs6445
rs200005406
rs6446
rs6473
novel
rs150697472
known

Reported
in pubmed
and other
databases

P < 0.05, P < 0.01, P < 0.001, del: deletion, ins: insertion, dup: duplication, con: conversion, UTR: untranslated region, n.c: no statistics were computed because one
group is constant (it was evaluated as P < 0.001).

1373 G>A (c.1225+30G>A)

1378 A>G (c.1225+35A>G)
1440 C>T (c.13275C>T)
1498 T>C (c.1380T>C)
1503 T>A (c.1385T>A)
1506 T>A (c.1388T>A)
1508 G>A (c.1390G>A)
1512 T>A (c.1394T>A)
1543 A>G (c.1413+12A>G)
1544 C>T (c.1413+13C>T)
1680 C>T (c.158321C>T)
1709 C>G (c.1591C>G)
1765 C>T (c.1647C>T)
1768 G>C (c.1650G>C)
1784 T>C (c.1666T>C)
1806 G>T (c.1688G>T)
1884 T-ins (c.1766_1767insT)
1912 G>C (c.1783+11G>C)
2181 G>A (c.2063G>A)
2231 C>T (c.2113C>T)
2370 C>T (c.2252C>T)
2489 T>C (c.2349+22T>C)
2493 G>A (c.2348+26G>A)
2544 C>T (c.244721C>T)
2702 C>T (c.2584C>T)
2793 G>C (c.2675G>C)
2815 G>A (c.2697G>A)
2823 G>A (c.2705G>A)
2825 C-ins (c.2707_2708insC)

2842 C>T (c. 2720C>T)

Large gene del/dup/con.

Nucleotide
positions and
variations (cDNA
numbering)

Table 1 Continued.

CYP21A2 Gene Defects in CAH Patients

Annals of Human Genetics (2014) 78,399409

403

D. Kirac et al.

were numbered 118 bp before the start codon, as well as

with cDNA numbering by using Human Genome Variation
nomenclature (http://www.hgvs.org/mutnomen/). Table 1
and Table S2 summarize all of the information about mutations/polymorphisms which were found with direct sequencing and MLPA analysis.
Accession numbers of nine novel mutations were
previously received from GenBank (http://www.ncbi.
nlm.nih.gov/nuccore/?term = kirac+d) on July 6,
2009. These numbers are: GQ166779.1-GI:242917609,
GQ166780.1-GI:242917611, GQ166783.1-GI:242917612,
GQ166784.1-GI:242917614, GQ166785.1-GI:242917616,
GQ166786.1-GI:242917618, GQ166787.1-GI:242917620,
GQ166788.1-GI:242917622.
Representative DNA sequencing chromatograms of novel
polymorphisms in the CYP21A2 gene are shown in Figure 1.
Representative relative peak ratios (RPR) of probes in
MLPA analysis are shown in Figure 2. Reference relative peak
ratios of probes are shown in Figure 2A. Relative peak ratios
of probes of a patient who has a homozygous loss of pseudogene exon 10 to CYP21A2 exon 5 are shown in Figure 2B.
Relative peak ratios of probes of a patient who has a homozygous loss of CYP21A2 exon 1 to 3 and homozygous gain of
four copies of CYP21A1P are shown in Figure 2C.

Comparison of Mutations/Polymorphisms
between Patient and Control Groups, and Allele
Frequencies in Both of the Groups
Mutations/polymorphisms at positions 711, 777, 829836
(8 bp del), 1326, 1512, 2231, 2702, and large gene deletions/duplications/conversions were found to be statistically
significant in patients (P < 0.05, 0.01, or 0.001) when patient
and control groups were compared with each other according
to the genotypes as well as according to the allele frequencies.
In addition, 1503, 1709, 2815, 2823 were only found to be
statistically significant in patients when patients were compared with controls. Other mutations in Table 1 which have
P-values less than 0.05 were mostly found in controls.

Association between Genotype and Disease

Types
In 21-OHD CAH, genotypes can be used to predict the
disease severity.

(a) Comparison of mutations/polymorphisms between disease types and the allele frequencies in both of the types
of the disease

404

Annals of Human Genetics (2014) 78,399409

Mutations/polymorphisms at position 777, 829836 (8

bp del), 982, 2544 and large gene deletions/duplications/
conversions were found to be statistically significant in the
SW type whereas mutations/polymorphisms at position 681,
1326, and 1440 were found to be statistically significant in the
NC type when disease types were compared with each other
according to the genotypes as well as according to the allele
frequencies (P < 0.05, 0.01, or 0.001). In addition, 723 was
only found to be statistically significant in the SW type when
disease types were compared between each other according
to the allele frequencies. Other mutations in Table 1 which
have P-values less than 0.05 were found to be statistically significant in both SV and NC types. However, mutations at
position 742, 746, 751, and 752 were found to be statistically
significant only in the SV type. Because of low frequency,
these values were not evaluated.

Correlation Results
If the P-value is smaller than 0.05 (P < 0.05) and the Spearman correlation coefficient value is 1 (r = 1), it means that
mutations/polymorphisms are found together in the same patient (100% correlation). Five 100% correlations were found
between 133 C-del, 135 T-del, and 136 G-del, as well as between 742 A>G and 752 A>G, 1234 T>C and 1244 C>G,
1515 A>T and 1517 C>A and between 1543 A>C and 1544
C>G (r = 1).

Discussion
The genetic diagnosis is more complicated for steroid 21hydroxylase deficiency than for many monogenic disorders
due to high variability of the locus. This includes coexistence of mutations/polymorphisms within an allele as well
as the presence of chromosomes containing more than one
CYP21A2 sequence (Wedell et al., 1994). The CYP21A2
and CYP21A1P genes consist of 10 exons and nine introns
and show high homology with a nucleotide identity of 98%
and 96% in their exon and intron sequences, respectively.
The proximity and the high degree of homology between
the CYP21A2 and CYP21A1P genes are believed to be the
main reason for unequal crossover and gene conversion-like
events, which give rise to mutations in CYP21A2. Approximately 95% of all disease-causing mutations in CYP21A2 are
either deletion/duplication/conversion or any of nine-point
mutations that have been transferred from CYP21A1P into
the active CYP21A2. The other 5% of the mutations are rare
and unique for single families or are considered as population
specific. These uncommon mutations do not originate from
the pseudogene. Large deletions are characterized by a single nonfunctional chimeric gene (CYP21A1P / CYP21A2)


C

2014 John Wiley & Sons Ltd/University College London

CYP21A2 Gene Defects in CAH Patients

Figure 1 DNA sequencing chromatograms showing novel polymorphisms of CYP21A2 in patients.


C

2014 John Wiley & Sons Ltd/University College London

Annals of Human Genetics (2014) 78,399409

405

D. Kirac et al.

Figure 2 (A) Reference relative peak ratios of probes of CYP21A2 and other genes near
CYP21A2. (B) Relative peak ratios of probes of a patient who has a homozygous loss of
pseudogene exon 10 to CYP21A2 exon 5. (C) Relative peak ratios of probes of a patient who
has a homozygous loss of CYP21A2 exon 1 to 3 and homozygous gain of four copies of
CYP21A1P.
406

Annals of Human Genetics (2014) 78,399409


C

2014 John Wiley & Sons Ltd/University College London

CYP21A2 Gene Defects in CAH Patients

having CYP21A1P sequences at the 5end and CYP21A2

sequences at the 3end (Kharrat et al., 2004).
A wide range of mutations causing 21-OHD have been
described so far. Complete gene deletions, single point mutations, large gene conversions, and a small 8bp deletion have
been reported (Krone et al., 1998). Common mutations are:
P30L, a C to T base change in exon 1; IVS2A/C>G, an A/C
to G base change in intron 2 causing abnormal RNA splicing
generating a stop codon downstream; an 8bp deletion in exon
3 that generates a stop codon; I172N, a missense mutation at
residue 172 in exon 4; missense mutations at residues 236,
237 and 239 in exon 6 (exon 6 cluster mutations), V281L, a
G to C base change in exon 7; Q318X, a base substitution
determining a stop codon in exon 8, and R356W, a missense
mutation in exon 8 (Torres et al, 2003).
In general, there is a good association between severe and
mild phenotypes of 21-OHD and the respective genotypes
(Parzer et al., 2001). The severity of the phenotype generally depends on the mutations present on both alleles, but
the phenotype is not entirely consistent among unrelated patients with the same mutations, or even among siblings of the
same genotype. Generally gene deletions, the 8-bp deletion
in exon 3, the exon 6 cluster of mutations, and R356W in
exon 8 are associated with SW type. The intron 2 mutation
(IVS2A/C>G also known as 656 A/C>G) can be found in
either SW or SV type. I172N is associated with SV, whereas
P30L and V281L are commonly seen in patients with NC
type. Based on consideration of the expected enzyme activity
of the less severely impaired allele, it is possible to predict
the phenotype with reasonable, but not complete, certainty
(Ferenczi et al., 1999; Keegan, 2001; White, 2001).
This study is the first to report distribution of
mutations causing 21-OHD in a Turkish population with
direct sequencing and MLPA analysis. To date, point mutations and small deletions in CAH patients have been investigated in particular by allele specific PCR, restriction fragment
length polymorphism, single stranded conformational polymorphism and many other techniques (Tukel et al., 2003).
However, these techniques detect only known mutations and
are unable to discover novel mutations. In our study, the
CYP21A2 gene was analyzed with direct sequencing. This
technique revealed 18 different novel mutations which have
not been described before. Also rather than Southern blot
analysis, MLPA analysis was used to detect large gene deletions
and conversions. This is a new, useful, and reliable technique
for detecting these types of mutations.
In this study, 124 patients with classical and nonclassical
forms of CAH and 100 healthy subjects were genotyped.
Most patients with the SW or SV forms were homozygous or
compound-heterozygous for mutations that completely abolish or severely reduce 21-OH activity. The most frequent
homozygous mutations are IVS2A/C>G at position 777


C

2014 John Wiley & Sons Ltd/University College London

(also known as 656 A/C>G in literature) (32%), 1709C>G

(16%), large gene deletions/duplications/conversions (15%),
711 G>A (11%) (novel mutation), and R356W (11%). These
mutations were found to be statistically significant in patients
when patient and control groups were compared with each
other (P < 0.05). These results are similar to those obtained
in previous reports, in which these mutations in particular
were the most common, with varying frequencies (White
& Speiser, 2000; Torres et al., 2003; Kharrat et al., 2004).
Our findings are also consistent with some other studies in
our country (Tukel et al., 2003; Sadeghi et al., 2008). In our
study, P30L mutation was not found in any of the cases. This
result is also similar to that observed in other studies (Dain
et al., 2002; Tukel et al., 2003). However, the nature and
the frequencies of the mutations in CYP21A2 vary among
different populations.
In our study, IVS2A/C>G, 8 bp deletion, and large gene
deletions/duplications/conversions were the most frequent
homozygous mutations in the SW form and 681G>C was
the most frequent mutation in the NC type, showing a significant association between these mutations and the clinical
forms of the disease (P < 0.05). However, 681G>C is a
novel mutation/polymorphism, and as it was also found in
the control group, we could not consider that this mutation
is specific for the NC type. In 21-OHD, patients are usually
compound heterozygotes for mutations which result in different categories of enzymatic activity. Therefore, there is a
high degree of overlap in moderate and mild forms of CAH,
reflected by the wide and heterogeneous spectrum of clinical
manifestation (Parzer et al., 2001). In previous reports, about
65%75% of CAH patients have been found to be in the
compound heterozygous state, having different mutations of
the CYP21A2 gene on each allele (Krone et al., 2000; Pinto
et al, 2003). In our study, this value is about 80%. Therefore,
our findings are consistent with other studies.
Additionally, in our study, five different 100% correlations
were found between 133 C-del, 135 T-del and 136 G-del,
as well as between 742 A>G and 752 A>G, between 1234
T>C and 1244 C>G, between 1515 A>T and 1517 C>A
and between 1543 A>C and 1544 C>G (P < 0.05, r = 1).
In other words, they were found to co-segregate in the same
patient.
Prenatal diagnosis, genetic counselling, and carrier detection are important for 21OH deficiency and have been carried
out for over 10 years by screening nine known mutations in
the CYP21A2 gene (Mao et al., 2002). Those carrying out
prenatal, and now more importantly, preimplantation diagnosis should always carefully investigate these complex loci
(Ezquieta et al., 2006). The few differences observed may reflect subtle sample variation in the ethnic background of the
studied population. This study may suggest the presence of
unknown mutations like 711 G>A in Turkish patients which

Annals of Human Genetics (2014) 78,399409

407

D. Kirac et al.

are associated with classic forms of 21-OHD. If similar results

are obtained in different populations, the novel 711 G>A
mutation in particular may be used as a marker for prenatal diagnosis. Therefore, the prevalence of genital ambiguity
and SW may be reduced in CAH patients by using prenatal
approaches.

Acknowledgements
This research was supported by the Scientific and Technological Research Council of Turkey (TUBITAK) 1002 grant and
Marmara University Scientific Research Project Coordination Unit (BAPKO) grant. All of the authors have no conflict
of interest to declare.

References
Concolino, P., Mello, E., Toscano, V., Ameglio, F., Zuppi, C. &
Capoluongo, E. (2009) Multiplex ligation-dependent probe amplification (MLPA) assay for the detection of CYP21A2 gene deletions/duplications in congenital adrenal hyperplasia: First technical report. Clinica Chimica Acta 402, 164170.
Dain, L. B., Buzzalino, N. D., Oneto, A., Belli, S., Stivel, M.,
Paqualini, T., Minutolo, C., Charreaus, E. H. & Alba, L. G. (2002)
Classical and nonclassical 21-hydroxylase deficiency: a molecular
study of Argentine patients. Clin Endocrinol 56, 239245.
Ezquieta, B., Beneyto, M., Pacheco, R. M., Barrio, R., Oyarzabal,
M., Lechugo, J. L., Luzuriaga, C., Hermoso, F., Quinteiro, S.
& Martinez, S. (2006) Gene duplications in 21-hydroxylase deficiency: The importance of accurate molecular diagnosis in carrier
detection and prenatal diagnosis. Prenat Diagn 26, 11721178.
Ferenczi, A., Garami, M., Kss, E., Pek, M., Szekely, M. S., Barta, C.,
Staub, M., Solyom, J. & Fekete, G. (1999) Screening for mutations
of 21-hydroxylase gene in Hungarian patients with congenital
adrenal hyperplasia. J Clin Endocrinol Metab 84, 23692372.
Friaes, A., Rego, A. T., Aragues, J. M., Moura, L. F., Mirante, A.,
Mascarenhas, M. R., Kay, T. T., Lopes, L. A., Rodrigues, J. C.,
Guerra, S., Dias, T., Teles, A. G. & Goncalves, J. (2006) CYP21A2
mutations in Portuguese patients with congenital adrenal hyperplasia: Identification of two novel mutations and characterization
of four different partial gene conversions. Mol Genet Metab 88,
5865.
Keegan, C. E. (2001) An overview of molecular diagnosis of steroid
21-hydroxylase deficiency. J Mol Diagn 3, 4954.
Kharrat, M., Tardy, V., Mrad, R., Maazoul, F., Jemaa, L. B., 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 prevalance
of Q318X mutation. J Clin Endocr Metab 89, 368374.
Krone, N., Braun, A., Roscher, A.A., Knorr, D. & Schwarts, H. P.
(2000) Predicting phenotype in steroid 21-hydroxylase deficiency.
Comprehensive genotyping in 155 unrelated, well defined patients
from southern Germany. J Clin Endocrinol Metab 85, 10591065.
Krone, N., Roscher, A. A., Schwartz, H. P. & Braun, A. (1998)
Comprehensive analytical strategy for mutation screening in 21hydroxylase deficiency. Clin Chem 44, 20752082.
Lee, H. H., Lee, Y. J., Chan, P. & Lin, C. Y. (2005) Use of PCRbased amplification analysis as asubstitute for the Southern blot

408

Annals of Human Genetics (2014) 78,399409

method for CYP21 deletion detection in congenital adrenal hyperplasia. Clin Chem 51, 480.
Lobato, M. N., Aledo, R. & Meseguer, A. (1998) High variability of CYP21 gene rearrangements in Spanish patients with
classic form of congenital adrenal hyperplasia. Hum Hered 48,
216225.
Mao, R., Nelson, L., Kates, R., Miller, C. E., Donaldson, D. L.,
Tang, W. & Ward, K. (2002) Prenatal diagnosis of 21-hydroxylase
deficiency caused by gene conversion and rearrangements: Pitfalls
and molecular diagnostic solutions. Prenat Diagn 22, 11711176.
Merke, D. & Bornstein, S. R. (2005) Congenital adrenal hyperplasia.
Lancet 365, 21252136.
New, M. I. (2004) An update of congenital adrenal hyperplasia. Ann
N Y Acad Sci 1038, 1443.
Nimkarn, S. & New, M. I. (2007) Prenatal diagnosis and treatment
of congenital adrenal hyperplasia. Hor Res 67, 5360.
Parzer, S. M. B., Schulze, E., Waldhausl, W., Pauschenwein, S. P.,
Rondot, S., Npwptny, 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, 47714775.
Pinto, G., Tardy, V., Trivin, C., Thalassinos, C., Jacob, S. L., Fekete,
C. N., Morel, Y. & Brauner, R. (2003) Follow-up of 68 children
with congenital adrenal hyperplasia due to 21-hydroxylase deficiency: Relevance of genotype for management. J Clin Endocrinol
Metab 88, 26242633.
Robins, T., Chantelot, C. B., Barbaro, M., Cabrol, S., Wedell, A.
& Lajic, S. (2007) Characterization of novel missense mutations
in CYP21 causing congenital adrenal hyperplasia. J Mol Med 85,
247255.
Sadeghi, F., Kutlay, N. Y., Berberoglu, M., C
etinkaya, E., Aycan,
Z., Kara, C., Ruhi, H. I., Ocal, G., Sklar, Z., Elhan, A. &
Tukun, A. (2008) Identification of frequency of distribution of the
nine most frequent mutations among patients with 21-hydroxylase
deficiency in Turkey. J Pediatr Endocr Met 21, 781787.
Sugino, Y., Usui, T., Okubo K., Nagahama, K., Takahashi, T.,
Okuno, H., Hatayama, H., Ogawa, O., Shimatsu, A. & H.
Nishiyama. (2006) Genotyping of congenital adrenal hyperplasia due to 21-hydroxylase deficiency presenting as male infertility:
Case report and literature review. J Assist Reprod Genet 23, 377
380.
Torres, N., Mello, M. P., Germano, C. M. R., Elias, L. L. K.,
Moreira, A. C. & Castro, M. (2003) Phenotype and genotype
correlation of the microconversion from the CYP21A1P to the
CYP21A2 gene in congenital adrenal hyperplasia. Braz J Med Res
36, 13111318.
Trakakis, E., Loghis, C. & Kassanos, D. (2009) Congenital adrenal
hyperplasia because of 21-hydroxylase deficiency. A genetic disorder of interest to obstetricans and gynecologist. Obstet Gynecol
Surv 64, 177189.
Tukel, T., Uyguner, O., Wel, J. Q., Apak, M. Y., Saka, N., Song,
D. X., Kayserili, H., Bas, F., Gunoz, H., Wilson, R. C., New,
M. I. & Wollnik, B. (2003) A novel semiquantitative polymerase
chain reaction/enzyme digestion-based method for detection of
large scale deletions/conversions of the CYP21 gene and mutation
screening in Turkish families with 21-hydroxylase deficiency. J
Clin Endocrinol Metab 88, 58935897.
Wedell, A., Thilen, A., Ritzen, E. M., Stengler, S. & 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.


C

2014 John Wiley & Sons Ltd/University College London

CYP21A2 Gene Defects in CAH Patients

White, P. C. (2001) Congenital adrenal hyperplasias. Best Pract Res

Clin Endocrinol Metab 15, 1741.
White, P. C. & Speiser, P. W. (2000) Congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Endocr Rev 21, 245
291.
White, P. C., Vitek, A., Dupont, B. & New, M. I. (1988)
Characterization of frequent deletions causing steroid 21hydroxylase deficiency. Proc Natl Acad Sci U S A 85, 4436
4440.
Wilson, R. C., Nimkarn, S., Dumic, M., Obeid, J., Azar, M., Najmabadi, H., Saffari, F. & New, M. I. (2007) Ethnic-specific distribution of mutations in 716 patients with congenital adrenal
hyperplasia owing to 21-hydroxylase deficiency. Mol Gen Metab
90, 414421.


C

2014 John Wiley & Sons Ltd/University College London

Supporting Information
Additional Supporting Information may be found in the online version of this article:
Table S1. PCR primers used for the amplification of
CYP21A2.
Table S2. Frequencies of variations in both of the groups.

Received: 8 May 2014

Accepted: 28 July 2014

Annals of Human Genetics (2014) 78,399409

409