Article Text

21-hydroxylase deficiency in Italy: a distinct distribution pattern ofCYP21 mutations in a sample from southern Italy
  1. A BOBBA,
  2. E MARRA,
  3. S GIANNATTASIO
  1. CNR, Centro di Studio sui Mitocondri e Metabolismo Energetico, Via Amendola 165/A, 70126 Bari, Italy
  2. Dipartimento di Biomedicina dell’Età Evolutiva, Università di Bari, Italy
  3. Dipartimento di Pediatria, Università Federico II, Napoli, Italy
    1. A IOLASCON,
    2. F MONNO
    1. CNR, Centro di Studio sui Mitocondri e Metabolismo Energetico, Via Amendola 165/A, 70126 Bari, Italy
    2. Dipartimento di Biomedicina dell’Età Evolutiva, Università di Bari, Italy
    3. Dipartimento di Pediatria, Università Federico II, Napoli, Italy
      1. S DI MAIO
      1. CNR, Centro di Studio sui Mitocondri e Metabolismo Energetico, Via Amendola 165/A, 70126 Bari, Italy
      2. Dipartimento di Biomedicina dell’Età Evolutiva, Università di Bari, Italy
      3. Dipartimento di Pediatria, Università Federico II, Napoli, Italy

        Statistics from Altmetric.com

        Editor—Congenital adrenal hyperplasia (CAH) is an autosomal recessive genetic disease, the main cause of which is steroid 21-hydroxylase (21OH) deficiency.1 The classical phenotype of the disease includes a severe salt wasting (SW) form in which both cortisol and aldosterone synthesis is impaired, and a mild form with normal aldosterone synthesis, commonly known as simple virilising (SV). In most white populations, an incidence of 1:5000 to 1:15 000 has been reported for the severe classical forms.2 The late onset form (LO), also called non-classical (NC), is characterised by the appareance of symptoms of excessive androgen activity late in life with an incidence of 1:100 to 1:1000 persons.3

        The steroid 21-hydroxylase locus has a complex structure with two genes, CYP21P andCYP21, located on the short arm of chromosome 6 within the HLA class III gene region dowstream of each of the two genes encoding the fourth component of complement,C4A and C4B, respectively. Small exchanges of sequences, a process known as gene conversion, could create many of the disease causing mutated alleles by transferring some of the deleterious mutations from the pseudogene to the CYP21 gene.4 5 Owing to the complexity of the locus, complete characterisation of mutantCYP21 alleles in patients with 21OH deficiency could be performed either by direct sequencing or by a mutation scanning method, such as non-radioactive DNA single strand conformation polymorphism analysis of all exons and exon/intron junctions.6 To date, deletion and several sequence aberrations of the CYP21 gene have been reported to cause steroid 21-hydroxylase deficiency in different populations including Italians, with 10 mutations being the most frequent.7-11

        In cases of other well characterised recessive diseases, such as phenylketonuria and cystic fibrosis, clear heterogeneity of distribution of the relative frequencies of the disease mutation has been found in the Italian population.12 13 With the aim of determining the CYP21 mutation profile in a selected Italian sample, the screening of the 10 most frequentCYP21 gene mutations was performed on a cohort of 21OH deficient patients, originating from southern Italy. The results of this study may also prove to be of practical use for the implementation of regional programmes of carrier screening or prenatal diagnosis using methods to screen simultaneously for the most frequent mutations.

        Twenty eight patients from southern Italy with a total of 50 independent CAH chromosomes were enrolled in this study. Southern Italian origin was assessed on the basis of the birth place of the grandparents. Patients were diagnosed on the basis of increased plasma 17α-hydroxyprogesterone (17OHP) and the ACTH provocation test. Each patient examined had high values of both baseline and ACTH stimulated 17OHP serum concentration, aggregating in the upper part of the nomogram reported in Wilson et al,14 and were classified as having one of the three clinical forms of CAH.3 Seventeen patients suffered from the SW form, eight from the SV form, and three from the LO form. Where available, parents were also enrolled in the study.

        Genomic DNA was prepared from peripheral blood leucocytes by standard procedures. Oligonucleotides were purchased from Pharmacia (Sweden) and their sequences and names have been reported elsewhere.6 15 PCR amplifications were performed using approximately 100 ng of genomic DNA in the presence of 200 μmol/l of each dNTP, 1 mmol/l MgCl2, 10 mmol/l Tris-HCl (pH 8.3), 50 mmol/l KCl, 50 pmol of each primer, and 2.0 UTaq polymerase (Promega) in a volume of 100 μl. Selective amplification of the CYP21gene in two overlapping fragments was carried out as previously reported.6 Genotyping was performed in a second, allele specific round of PCR essentially as described by Wedell and Luthman15 using 1 μl of the initial reaction. The following primers were used: for P31L: P92T, P92C, and P48; for I2 splice: P659G, P659A, P659C, and P48; for I173N: P1004A, P1004T, and Zfor; for cluster E6: P1388A, P1388T, and Zfor; for V281L: P1688T, P1688G, and Ofor; for 306insT: P1678T, P1678G, and Ofor; for Q318X: P1999T, P1999C, and Ofor; for R356W: P2113T, P2113C, and Ofor; for P453S: P2584T, P2584C, and Ufor. The I172N, V281L, and R356W mutations were analysed using an allele specific PCR protocol involving the use of a competitive allele non-specific primer placed outside the allele specific primer, that is, Lrev, Prev, and Vrev, respectively.6 All PCR products, except those obtained from the screening of I2 splice, were analysed by agarose gel electrophoresis and ethidium bromide staining. In the case of I2 splice, PCR products were visualised after silver staining of 13% polyacrylamide minigels.

        To determine the presence of the 8 bp deletion in exon 3 (del8bp), a fragment of 616 bp was amplified with the gene specific forward primer Efor and Hrev.6 This fragment was used as a template for a second PCR reaction using the primer pair del8for/del8rev16 with amplification products analysed by electrophoresis on 12% polyacrylamide minigels followed by ethidium bromide staining. For those subjects carrying the del8bp mutation, specific amplification of the full lengthCYP21 gene was directly performed from genomic DNA with Yfor/Zrev primers after removing the pseudogene byTaqI cleavage, as previously described.17 In order to confirm homozygous point mutations, Southern blotting analysis was carried out whenever parents were not available; 10 μg of DNA was digested with the restriction endonuclease TaqI, and the products were analysed by electrophoresis on 0.8% agarose gels followed by Southern transfer to nylon membranes. The probe was a 2209 bp fragment of theCYP21 gene, radiolabelled with [32P]dCTP by using the Megaprime Kit (Amersham). Hybridisation conditions were essentially those reported by Haglund-Stengler et al.18 Autoradiography was carried out by exposure of filters at −80°C for 96 hours. The relative intensity of the 3.7 kb compared to the 3.2 kb fragment was determined using the GS-700 Imaging Densitometer implemented with the Molecular Analyst Software (Bio-Rad Laboratories).

        Statistical tests for heterogeneity of distribution ofCYP21 mutations were performed using 2 × 2 contingency tables and Fisher’s exact test.

        A total of 98% of the chromosomes analysed were genotyped by the approach used in this study, showing the highest rate of mutation detection so far achieved in the characterisation ofCYP21 mutations in 21OH deficiency. TheCYP21 gene mutation profile is reported in table 1. I2 splice was present in more than half (56%) of the southern Italian CAH chromosomes analysed. This mutation occurred de novo on one chromosome analysed. Cluster E6 was identified in the Italian population for the first time, whereas P453S was not found in our patients. No subject analysed was found to be homozygous for aCYP21 deletion since in every case a successful amplification of the CYP21 gene was obtained. Three patients were hemizygous for gene deletion, accounting for 8% of the chromosomes genotyped. Mutated alleles carrying only one mutation represented 78% of CAH chromosomes analysed. Six different complex alleles, as deduced from the detection of more than one mutation on the same chromosome, were identified with a total relative frequency of 12%. For all patients except one, the twoCYP21 alleles were fully characterised: 18 (64.3%) are compound heterozygotes for different mutations and 10 (35.7%) are homozygotes.

        Table 1

        Relative frequencies of CYP21 alleles in 50 southern Italian CAH chromosomes

        As for I2 splice frequency, this study provides the highest relative frequency found for this mutation both in Italian9 and other European populations.8 10 11 A statistically significant difference (p<0.001) was found between this relative frequency and that measured in the Italian population by Carreraet al 9 (20%) and in the Spanish by Ezquieta et al 10(22%).

        In comparison with previously characterised Italian patients,9 Q318X was found with a lower relative frequency. On the other hand, R356W was found in our cohort as a single mutation allele with a relative frequency of 4%, whereas Carreraet al 9 found it in only one complex allele together with five other mutations. The mutation cluster in exon 6 was found for the first time in the Italian population and was identified as part of a complex allele in three patients. However, these differences concern relatively rare mutations and are either not significant or only slightly so. Even if the size of our sample may not be considered large enough, it can be speculated that the statistically different frequency of I2 splice, as well as the other peculiar mutation frequencies found in our sample, could be secondary to ethnic differences. Recent evidence has been obtained concerning the genetic heterogeneity of the Italian population which is accounted for by prehistorical colonisation events.12 13 Since no information is given about the origin of the Italian patients genotyped by Carrera et al,9 the heterogeneity of distribution of relative frequencies of Italian CAH alleles will need further investigation.

        Genotype-phenotype correlation was assessed for all subjects and the results are summarised in table 2. Only five mutations, I2 splice,CYP21 deletion, I172N, Q318X, and R356W, represent 86% of CAH chromosomes associated with the classical forms of the disease. This finding could prompt the development of molecular diagnosis for phenotype prediction. Since the severity of the disease is thought to be determined by the activity of the less severely affected allele, both alleles of a 21OH deficient patient must be considered correctly to predict the phenotype, at least in the case of I172N and in other previously established severe mutations, such as Q318X and R356W.

        Table 2

        Genotype-phenotype correlation in 28 21-hydroxylase deficient patients

        However, the phenotype of patients does not always correlate with the genotype especially when the splicing mutation at nucleotide position 655 is observed. In vitro expression studies of theCYP21 gene containing the intron 2 mutation led it to be considered as a severe mutation since aberrant mRNA splicing, associated with lack of enzymatic activity, was detected.4 However, the in vitro studies might not reflect the in vivo situation in the adrenal cortex so, although the majority of patients homozygous for this mutation show a strong association between I2 splice and the SW form, a SV phenotype was observed in three out of 10 patients according to previous reports.8 11 14

        In view of these observations, the fact that I2 splicing mutation represents 61% of southern Italian CAH alleles associated with the classical forms of the disease (table 1) precludes the use of genotyping CYP21 alleles for correct 21OH deficiency phenotype prediction aimed at the implementation of effective therapy. In any case, the possibility of performing prenatal diagnosis and carrier detection through the screening of the most frequent mutations still retains its validity.

        References

        View Abstract

        Request permissions

        If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.