Clinical Chemistry 45: 625-629, 1999;
(Clinical Chemistry. 1999;45:625-629.)
© 1999 American Association for Clinical Chemistry, Inc.
Single-Nucleotide Polymorphisms in Intron 2 of CYP21P: Evidence for a Higher Rate of Mutation at CpG Dinucleotides in the Functional Steroid 21-Hydroxylase Gene and Application to Segregation Analysis in Congenital Adrenal Hyperplasia
Renée R. Jiddou1,
Wan-Li Wei1,
Kumud S. Sane2 and
Anthony A. Killeen1,a
1
Department of Pathology, University of Michigan, Ann Arbor, MI 48109-0602.
2
Department of Pediatrics, University of Minnesota,
Minneapolis, MN 55455.
a Address correspondence to this author at: Department of Pathology, University of Michigan, 1301 Catherine St., Ann Arbor, MI 48109-0602. Fax 734-936-2756; e-mail akilleen{at}umich.edu
 |
Abstract
|
|---|
Background: Intron 2 of CYP21, the functional
steroid 21-hydroxylase gene contains several single-nucleotide
polymorphisms (SNPs). We tested the hypothesis that intron 2 of the
pseudogene, CYP21P, might also be polymorphic and
provide markers for segregation analysis of this region of the genome,
including observable markers for segregation analysis of
CYP21 gene deletions. A comparison of SNPs in both genes
might provide insights into the rates of mutation in these duplicated
genes.
Methods: After amplification with PCR, we examined restriction
site polymorphisms in intron 2 of CYP21P in 24 members
of the parental generation of the Centre d'Étude du
Polymorphisme Humain families and selected offspring.
Results: Intron 2 of CYP21P contains frequent SNPs
around nucleotide 398 and nucleotide 509, which can be typed by
PCR/restriction enzyme digestion with HaeIII. Of the 48
CYP21P alleles examined, 44 could be characterized
unambiguously. Of these 44 alleles, 4 were deleted, and the frequencies
of restriction at the polymorphic HaeIII sites were 20
of 40 at nucleotide 398 and 30 of 40 at nucleotide 509. Both
polymorphisms result from C
T transitions that occur at CpG
dinucleotides. The frequencies of C at these nucleotides in
CYP21P are significantly higher than at the
corresponding nucleotides in CYP21 of the same
individuals (P <0.01).
Conclusion: These data suggest that these CpG dinucleotides are
more frequently mutated in CYP21 than in
CYP21P, and that several mutations at CpG dinucleotides
in the coding regions of CYP21 might result from CpG
instability rather than the more usually proposed mechanism of gene
conversion. These frequent SNPs provide useful markers for studying
both allelic segregation of CYP21, particularly for
chromosomes with known CYP21 deletions, and for
investigating the origin of these polymorphisms.© 1999 American
Association for Clinical Chemistry
|
Introduction
|
|---|
The gene encoding adrenal steroid 21-hydroxylase,
CYP21, is located on 6p in the MHC class III region. This
region usually includes two genes, the functional CYP21 gene
and a highly homologous pseudogene, CYP21P, which are
arranged in tandem with the complement 4 genes
(1)(2) (Fig. 1
). The CYP21 and CYP21P genes overlap the
tenascin-X and XA genes, and two untranslated
messages termed YA and YB have also been identified as transcripts of
this region (3). Mutations in CYP21 produce the
most common form of congenital adrenal hyperplasia
(CAH),1
an autosomal recessive disease with an incidence of ~1 in
13 000 newborns for the classic form, and an incidence of up to 1% in
some populations for the nonclassic form (4)(5).
View larger version (5K):
[in this window]
[in a new window]
|
Figure 1. Arrangement of the steroid 21-hydroxylase genes on 6p.
CYP21 is the functional steroid 21-hydroxylase gene;
CYP21P is a pseudogene. C4A and
C4B encode isoforms of the complement 4 proteins.
|
|
We have reported the existence of frequent single-nucleotide
polymorphisms (SNPs) in intron 2 of CYP21, including SNPs at
nucleotide (nt) 395 and nt453 (6). The former involves a
CT transition at a CpG dinucleotide. Considering the sequence
homology between CYP21 and CYP21P, and the
well-known tendency of CpG dinucleotides to undergo mutation, which
leads to polymorphism among alleles (7)(8), we
hypothesized that the nucleotides at the corresponding positions in
CYP21P might also be polymorphic. The presence of SNPs in
the pseudogene might provide markers for determining the segregation of
this region of the MHC in genetic studies, especially segregation of
the closely linked CYP21 in families with CAH. We have shown
that CYP21 gene deletions, which account for ~20–30% of
all mutations at that locus (9), can be demonstrated by an
apparent failure to transmit a parental allele to an offspring as
determined by SNP analysis of CYP21 in informative families
(6)(10). SNPs in CYP21P might provide
closely linked, observable markers for this analysis. Finally, a
comparison of SNPs in CYP21P and CYP21 might
provide insight into the relative rates of mutation at CpG
dinucleotides in these duplicated genes.
To test these hypotheses, we examined the frequencies of SNPs which
occur in CYP21P at positions homologous to those we have
reported previously in CYP21 in members of the parental
generation of the Centre d'Étude du Polymorphisme Humain (CEPH)
families, and compared the frequencies of these SNPs in
CYP21P with those at the homologous positions in
CYP21.
|
Materials and Methods
|
|---|
genotyping cyp21p in intron 2
DNA samples from patients with CAH and their relatives were
extracted from peripheral blood as described previously
(11). The study was approved by institutional review boards,
and informed consent was obtained from all subjects or their parents.
Intron 2 of CYP21P was amplified by 30 cycles of PCR using
the following reagent concentrations: 20 mmol/L Tris-HCl, pH 8.4, 50
mmol/L KCl, 1.5 mmol/L MgCl2, 200 µmol/L each
dNTP, 500 nmol/L each primer (335F and P731R; Table 1
), 50–100 ng of genomic DNA, and 2.5 U of Taq polymerase (Life
Technologies) in a reaction volume of 50 or 100 µL. For genomic DNAs
that had been stored for several years and showed some evidence of
degradation, a nested strategy was used to amplify intron 2 from
CYP21P. First-round PCR reactions were performed for 30
cycles as described above, but with 500 nmol/L each of primers 1F and
822R (Table 1
). For the second-round reactions, 1 µL of first-round
product was used as template and the primers used were 335F and P731R
(Table 1
). Primer P731R is a CYP21P-specific primer that
hybridizes to the region in exon 3 of CYP21P, where there is
a sequence difference from CYP21 because of a deletion of 8
bp from CYP21P (1)(2). The initial
denaturation of DNA was for 1 min at 94 °C. For nonnested reactions,
an annealing temperature of 60 °C was used for 30 s. For nested
reactions, annealing conditions were 60 °C for round 1 and 65 °C
for round 2 reactions, both for 30 s. Extension conditions were
72 °C for 90 s (round 1) or 1 min (round 2).
Restriction digestion was performed by incubating 10 µL of PCR
product with 5 U of HaeIII (10 U/µL; Life Technologies) in
a volume of 30 µL at 37 °C for 2–3 h. The restriction solution
contained the manufacturer's 1x restriction buffer. After restriction
digestion, samples were electrophoresed in 3% Metaphor agarose (FMC)
containing 1x Tris-borate-EDTA buffer (0.089 mol/L Tris-borate, pH
8.3, 2 mmol/L EDTA; National Diagnostics) and 1 mg/L ethidium bromide
(Sigma). Restriction products were visualized by transillumination with
ultraviolet light and photographed.
haplotype analysis
To establish the haplotype pattern of SNPs at CYP21P in
the CEPH families, reference was made to the available inheritance data
for the HLA-B and HLA-DR loci in these families.
These loci flank the MHC class III region (containing the
21-hydroxylase genes) on the telomeric and centromeric sides,
respectively. The HLA data were used to identify second-generation
siblings who have inherited different parental haplotypes in this
region of 6p. On the basis of the HLA data, none of the offspring
showed recombination events between the HLA-B and
HLA-DR loci. Study of the transmission of parental
CYP21P SNPs to these siblings enabled assignment of SNPs to
individual parental haplotypes. This analysis was used to demonstrate
CYP21P deletions, which occur in ~14% of individuals in
the French population, to which the CEPH families belong.
CYP21P deletions produce apparent failure to transmit a
parental allele to an offspring when intragenic polymorphisms are
studied, as we have shown previously at the CYP21 locus
(6)(10).
genotyping cyp21p of the hla-a3, -b47, -dr7
haplotype
The pattern of CYP21P intron 2 SNPs was studied in four
unrelated individuals with the HLA-A3, -B47, -DR7 haplotype, which is
associated with a deletion of CYP21, and which is frequently
seen in patients with classic CAH. DNA samples from parents who are
heterozygous for this haplotype, and their child with CAH who is
homozygous for the haplotype, were studied. In two other families, the
intron 2 SNPs could be determined in parents who are heterozygous for
this haplotype by segregation analysis in their offspring.
|
Results
|
|---|
cyp21p intron 2 snps
PCR amplification with primers 335F and P731R yields a product of
~400 bp. The pattern of restriction by HaeIII was studied
in DNA samples from 24 members of the parental generation of the CEPH
families (48 chromosomes). A diagram showing the HaeIII
restriction sites in CYP21P is shown in Fig. 2
. The amplified region includes three HaeIII sites in
intron 2: nt398, nt509, and nt623. These correspond to nt395, nt503,
and nt614 in CYP21 (2). The patterns and frequencies
of restriction at these sites are shown in Table 2
.
View larger version (6K):
[in this window]
[in a new window]
|
Figure 2. Restriction map of the HaeIII sites in
intron 2 of CYP21P.
The arrows show the locations of primers 335F and P731R
in exons 2 and 3, respectively. The thicker line
represents intron 2. The polymorphic HaeIII sites are
indicated by *.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Frequencies of CYP21P Intron 2
HaeIII polymorphisms and CYP21P deletions in 24
CEPH parental generation members (44
chromosomes).1
|
|
Of the 48 chromosomes, 4 were found to have deletions of
CYP21P, as evidenced by the apparent failure of transmission
of a parental allele to an offspring (Fig. 3
). In an additional four individuals showing apparent
homozygosity for markers at CYP21P, a gene deletion could
not be excluded by segregation analysis because of a lack of
informativeness from the family or the unavailability of offspring with
the necessary combination of parental haplotypes to exclude a gene
deletion. For calculations of frequencies of polymorphisms and
deletions, only one allele was counted from these individuals. The CEPH
study sample therefore consisted of 40 distinct CYP21P
pseudogenes and 4 chromosomes with demonstrated deletions of
CYP21P, a deletion rate of 4 of 44 (9.1%).
View larger version (35K):
[in this window]
[in a new window]
|
Figure 3. Demonstration of a CYP21P deletion in CEPH
family 13292.
The numbers in the first row below each subject in the pedigree diagram
are CEPH sample numbers. The numbers in the second row below each
subject are haplotype identifications for the HLA-B and
HLA-DR loci. The pattern of restriction around nt398 and
nt509 are shown below the gel. + indicates restriction, - indicates
absence of restriction, and ± indicates both an allele that shows
restriction and one that does not. Molecular weight markers
(HaeIII digest of  X174) are shown in lane
1. Subject 13292-01 (lane 2), who shows a
pattern of restriction fragments associated with the presence of the
HaeIII site at nt509 and the absence of the site at
nt398, has apparently failed to transmit an allele to 13292-03
(lane 3). This is consistent with an allele bearing a
CYP21P deletion. In contrast, subject 13292-02
(lane 5) has transmitted an allele that shows
restriction at both sites to both offspring.
|
|
The HaeIII restriction site around nt623 was cut in all CEPH
DNA samples studied, but the restriction sites at nt398 and nt509 were
found to be frequently polymorphic. The nt398 site was restricted in 20
of 40 nondeletional chromosomes, and the nt509 site was restricted in
30 of 40 nondeletional chromosomes. The polymorphism between
CYP21 and CYP21P at the HaeIII site
around nt503/nt509 previously has been shown to result from a GA
transition on the sense strand (1)(2); however,
the reason for loss of the HaeIII restriction site at nt398
in CYP21P has not been established. We confirmed by sequence
analysis of three alleles that the polymorphism that destroys the
latter site results from a CT transition on the sense strand as
occurs at nt395 in CYP21 (data not shown).
cyp21p snps in the hla-a3, -b47, -dr7 haplotype
The pattern of HaeIII restriction in intron 2 of
CYP21P in chromosomes bearing the HLA markers A3, B47, and
DR7 was examined in four unrelated individuals with this haplotype.
This haplotype is associated with a deletion of CYP21 in
families with CAH. The CYP21P pseudogene on this haplotype
was found to be characterized by the presence of a HaeIII
restriction site at nt398 and absence of the site at nt509 in all
subjects examined.
|
Discussion
|
|---|
SNPs are reported to be the most frequent form of polymorphism in
humans, occurring on average every 1 kb, and their identification
is one of the goals of the Human Genome Project
(12)(13). The presence of SNPs in the
21-hydroxylase genes is of interest for several reasons. First, they
can be used for analysis of segregation of these alleles in genetic
studies, notably of families with CAH. Second, they enable
identification of gene deletions in informative families, and third,
they offer insights into the relative rates of mutation in these genes.
We have demonstrated previously the presence of frequent SNPs in intron
2 of CYP21, which is located ~30 kb from
CYP21P, including SNPs at nt395 and nt453. The latter were
demonstrated by restriction with the enzyme BbvI. This
enzyme did not reveal polymorphisms in CYP21P in the samples
examined in this study; however, we demonstrated that there are two
polymorphic HaeIII restriction sites in this intron, one
located around nt398, and one located around nt509. Another
HaeIII site around nt623 was restricted in all samples
studied. Among nondeletional alleles, the frequency of restriction at
the nt398 and nt509 sites among the CEPH samples studied was 20 of 40
alleles and 30 of 40 alleles, respectively (Table 2
).
There is a significantly greater degree of polymorphism at nt398 of
CYP21P than at the corresponding nt395 of CYP21
of the same subjects, as reported previously (6). Of 48
CYP21 alleles, only 8 (16.7%) were restricted by
HaeIII, indicating the presence of C at nt395, whereas 20 of
40 (50%) were restricted at the homologous site among nondeletional
CYP21P alleles. At nt509 in CYP21P, the
HaeIII restriction site (present in 30 of 40 alleles)
results from the presence of G (C on the antisense strand) at this
position. In 48 CYP21 alleles from the CEPH families and 7
wild-type alleles from heterozygotes for CAH, the corresponding
position (nt502) is A (T on the antisense strand) (6); thus,
this polymorphism also results from a CT transition that occurs on
the antisense strand. The frequencies of these polymorphisms were
significantly different between CYP21 and CYP21P
at both sites (
2, P <0.01). At
both positions, the CT transitions occur at CpG dinucleotides that
are situated in the sense orientation at nt395 and in the antisense
orientation at nt502. CpG dinucleotides are well-recognized "hot
spots" for genetic mutation involving CT transitions (GA on the
complementary strand) (7)(8). The basis of
mutation at these sites is believed to be the spontaneous deamination
of 5-methylcytosine to form thymidine. On the basis of the observed
higher frequency of T at nt395 in CYP21 and the apparently
invariable presence of T on the antisense strand at nt502 of that gene,
we suggest that the CpG dinucleotides at these positions in
CYP21 are subject to a higher rate of mutation than are the
corresponding nucleotides in CYP21P. Because CT mutations
at CpG dinucleotides are believed to involve methylation of cytosine,
one possible explanation for the different frequencies of cytosine at
these CpG dinucleotides is that the functional CYP21 is more
highly methylated at these positions in the germ line than is
CYP21P.
This point may have relevance to the origin of several mutations in the
coding sequence of CYP21 that involve CT transitions
occurring at CpG dinucleotides. Among these are P30L (C88T), and R356W
(C2108T) (9). Because these mutations have been reported in
CYP21P, they frequently are considered to be evidence of
gene conversion events that transfer deleterious sequences from
CYP21P to CYP21. An alternative possible etiology
of these mutations, which is consistent with our observation of
frequent polymorphisms at CpG dinucleotides in intron 2, is that these
mutations result from an instability of CpG dinucleotides in
CYP21. This mechanism may also underlie the frequent
mutation at a TaqI restriction site used for distinguishing
CYP21 from CYP21P on Southern blot analysis
(14).
In this study, we observed a CYP21P deletion rate of 9.1%,
which is in reasonable agreement with the 14% estimate of such
deletions in the French population, from which the CEPH families are
derived (15). A possible mechanism for deletions at either
locus is the loss of genetic material arising from chromosomal
misalignment during meiosis and unequal crossing over. The difference
in the observed rates of gene deletion between CYP21 and
CYP21P may be attributable to the selection against
haplotypes bearing CYP21 deletions, which are associated
with the salt-wasting form of CAH. Deletions of CYP21P
appear to be silent with respect to phenotype, although a higher rate
of some autoimmune diseases is associated with deletions of the
neighboring C4A gene (16), which is often deleted
with CYP21P.
It was of interest to examine the pattern of SNPs in the
CYP21P gene in subjects having the extended MHC haplotype
HLA-A3, -B47, -DR7, which is associated with a deletion of
CYP21 and which frequently is present in patients with CAH.
In four subjects with this haplotype, the site around nt398, but not
the site around nt509, in CYP21P was restricted by
HaeIII. We previously have shown that the inheritance of a
deletion at CYP21 can be demonstrated by apparent failure to
transmit parental CYP21 intragenic polymorphisms to an
offspring (6)(10). The existence of
polymorphisms at the closely linked CYP21P locus also allows
for this analysis of allelic segregation in informative families, and
has the advantage of providing an observable marker in both parent and
offspring. The HLA-B47 antigen is known to be in strong linkage
disequilibrium with a deletion of CYP21 in families with
CAH. The finding of the same pattern of SNPs at the CYP21P
locus in unrelated individuals with the HLA-B47 antigen is consistent
with the idea that the HLA-A3, -B47, -DR7 haplotype represents a highly
conserved segment of DNA, and that the HLA-B47 antigen, which is
infrequent in the general population, is possibly of relatively recent
evolutionary origin (17), so that these SNPs are still
conserved among apparently unrelated individuals with this haplotype.
These polymorphic restriction sites also serve as useful markers for
rapid analysis of the segregation of the class III region of the MHC in
family studies, and most obviously, for following the segregation of
the nearby CYP21 alleles. These CYP21P markers
may be especially useful for determining the segregation of
haplotypes containing known CYP21 deletions, which usually
requires Southern blot analysis to assess gene dosage. We
previously have discussed the advantages and disadvantages of using
intragenic polymorphic markers for segregation analysis of CYP21
(6)(10). In comparison with CYP21,
CYP21P shows a higher frequency of deletions. Occasional
haplotypes can contain duplications of one or other gene
(18). These considerations must be borne in mind if these
markers are used for diagnostic purposes.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: CAH, congenital adrenal hyperplasia; SNP, single-nucleotide polymorphism; nt, nucleotide; and CEPH, Centre d'Étude du Polymorphisme Humain. 
|
References
|
|---|
-
White PC, New MI, Dupont B. Structure of the human steroid 21-hydroxylase genes. Proc Natl Acad Sci U S A 1986;83:5111-5115.
[Abstract/Free Full Text]
-
Higashi Y, Yoshioka H, Yamane M, Gotoh O. Complete nucleotide sequence of two steroid 21-hydroxylase genes tandemly arranged in human chromosome: a pseudogene and a genuine gene. Proc Natl Acad Sci U S A 1986;83:2841-2845.
[Abstract/Free Full Text]
-
Burch GH, Gong Y, Liu W, Dettman RW, Curry CJ, Smith L, et al. Tenascin-X deficiency is associated with Ehlers-Danlos syndrome. Nat Genet 1997;17:104-108.
[ISI][Medline]
[Order article via Infotrieve]
-
Pang S. Congenital adrenal hyperplasia. Endocrinol Metab Clin N Am 1997;26:853-891.
[ISI][Medline]
[Order article via Infotrieve]
-
Donohoue PA, Parker K, Migeon CJ. Congenital adrenal hyperplasia. Scriver CR Beaudet AL Sly WS Valle D eds. The metabolic and molecular basis of inherited disease 7th ed. 1995:2929-2966 McGraw-Hill New York. .
-
Killeen AA, Jiddou RR, Sane KS. Characterization of frequent polymorphisms in intron 2 of CYP21: application to segregation analysis of CYP21 alleles. Clin Chem 1998;12:2410-2415.
-
Cooper DN, Youssoufian H. The CpG dinucleotide and human disease. Hum Genet 1988;78:151-155.
[ISI][Medline]
[Order article via Infotrieve]
-
Cooper D, Krawczak M. The mutational spectrum of single base-pair substitutions causing human genetic disease: patterns and predictions. Hum Genet 1990;85:55-74.
[ISI][Medline]
[Order article via Infotrieve]
-
White PC, Tusie-Luna M-T, New MI, Speiser PW. Mutations in steroid 21-hydroxylase (CYP21). Hum Mutat 1994;3:373-378.
[ISI][Medline]
[Order article via Infotrieve]
-
Wei W-L, Killeen AA. Analysis of 4 common salt-wasting mutations in CYP21 (steroid 21-hydroxylase) by cleavase fragment length polymorphism analysis, and characterization of a frequent polymorphism in intron 6. Mol Diagn 1998;3:171-178.
[ISI][Medline]
[Order article via Infotrieve]
-
Killeen AA, Seelig S, Ulstrom RA, Orr HT. Diagnosis of classical steroid 21-hydroxylase deficiency using an HLA-B locus specific probe. Am J Med Genet 1988;29:703-712.
[ISI][Medline]
[Order article via Infotrieve]
-
Zhao LP, Aragaki C, Hsu L, Quiaoit F. Mapping of complex traits by single-nucleotide polymorphisms. Am J Hum Genet 1998;63:225-240.
[ISI][Medline]
[Order article via Infotrieve]
-
Collins FS, Patrinos A, Jordan E, Chakravarti A, Gesteland R, Walters L. New goals for the U.S. human genome project: 1998–2003. Science 1998;282:682-689.
[Abstract/Free Full Text]
-
White PC, Vitek A, Dupont B, New MI. Characterization of frequent deletions causing steroid 21-hydroxylase deficiency. Proc Natl Acad Sci U S A 1988;85:4436-4440.
[Abstract/Free Full Text]
-
Morel Y, Andre J, Uring-Lambert B, Hauptman G, Betuel H, Tosi M, et al. Rearrangements and point mutations of the p450c21 genes are distinguished by five restriction endonuclease haplotypes identified by a new probing strategy in 57 families with congenital adrenal hyperplasia. J Clin Investig 1989;83:527-536.
-
Fielder AHL, Walport MJ, Batchelor JR, Rynes RI, Black CM, Dodi IA, Hughes GRV. Family study of the major histocompatibility complex in patients with systemic lupus erythematosus: importance of null alleles of C4A and C4B in determining disease susceptibility. Br Med J 1983;286:425-428.
-
Killeen AA, Sane KS, Orr HT. Molecular and endocrine characterization of a mutation involving a recombination between the steroid 21-hydroxylase functional gene and pseudogene. J Steroid Biochem Mol Biol 1991;38:677-686.
[ISI][Medline]
[Order article via Infotrieve]
-
Ramsden SC, Sinnott PJ. Characterization of gene rearrangements and gene conversion events in the 21-hydroxylase gene. Elles R eds. Methods in molecular medicine: molecular diagnosis of genetic disease 1996:121-140 Humana Press Totowa, NJ. .
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
C. E. Keegan and A. A. Killeen
An Overview of Molecular Diagnosis of Steroid 21-Hydroxylase Deficiency
J. Mol. Diagn.,
May 1, 2001;
3(2):
49 - 54.
[Full Text]
|
|
|
|
|
|
|
P. C. White and P. W. Speiser
Congenital Adrenal Hyperplasia due to 21-Hydroxylase Deficiency
Endocr. Rev.,
June 1, 2000;
21(3):
245 - 291.
[Abstract]
[Full Text]
|
|
|
Top
Abstract
Introduction
