Low Proportion of Whole Exon Deletions Causing Phenylketonuria in Denmark and Germany

Transcription

1 HUMAN MUTATION Mutation in Brief #952 (2007) Online MUTATION IN BRIEF Low Proportion of Whole Exon Deletions Causing Phenylketonuria in Denmark and Germany Lisbeth Birk Møller 1 *, Anders O.H. Nygren 2, Patrick Scott 3, Pia Hougaard 1, Jytte Bieber Nielsen 1, Caroline Hartmann 4, Flemming Güttler 1, Linda Tyfield 5, and Johannes Zschocke 4 1 Kennedy Institute-National Eye Clinic, Glostrup, Denmark; 2 MRC-Holland bv, Hudsonstraat 68, 1057SN Amsterdam, Holland; 3 Molecular Genetics Laboratory, Montreal Children s Hospital McGill University Health Centre/Human Genetics and Pediatrics, McGill University, Montreal, Quebec, Canada; 4 Institute of Human Genetics, University of Heidelberg,Heidelberg, Germany; 5 Department of Molecular Genetics, Southmead Hospital, Bristol, United Kingdom *Correspondence to: Lisbeth Birk Møller, Kennedy Institute-National Eye Clinic, Gl. Landevej 7, 2600 Glostrup Denmark; lbm@kennedy.dk Communicated by Graham Taylor Phenylketonuria (PKU) is an autosomal recessive genetic disorder caused by mutations of the gene encoding phenylalanine hydroxylase (PAH). More than 500 different PAH mutations have been identified and about 90% of these are single base mutations. Although the identification rate of the PAH mutations is generally very high, some variants remain unidentified. A fraction of these mutations are the result of genomic deletions or duplications, which are not recognized with standard PCR-based methods. Here we present the results of exon deletion or duplication analysis in a total of 34 families, in which two mutations had not been identified using conventional diagnostic screening techniques. Using multiplex ligation-dependent probe amplification (MLPA), we found a deletion covering exon 1 and exon 2 (c.1-?_168+?del) in one patient, a deletion of exon 3 (c.169-?_352+?del) in four patients, and a deletion of exon 5 (c.442-?_509+?del) in two patients. A deletion was thus detected in about 20% (7/34) of the families tested. Out of a combined cohort of 570 independent PKU patients from Denmark and Germany, exon deletions were identified in a total of four patients. The estimated allelic frequency of exon deletions in PKU patients in these two populations is therefore below 0.5% Wiley-Liss, Inc. KEY WORDS: phenylalanine hydroxylase; PAH; phenylketonuria; PKU; MLPA; comparative multiplex dosage analysis; CMDA INTRODUCTION Phenylketonuria (PKU; MIM# ) is the most common inborn error of amino acid metabolism in Europeans. It is caused by mutations in the gene encoding phenylalanine hydroxylase (PAH), and characterized by accumulation of phenylalanine in the serum due to impairment of hepatic hydroxylation to tyrosine. The disorder is transmitted in an autosomal recessive pattern with an average incidence among Caucasians of 1/10000 (Mathias and Bickel et al., 1986), corresponding to a carrier frequency of about 1 in 50. More than 500 different PAH Received 30 June 2006; accepted revised manuscript 10 November WILEY-LISS, INC. DOI: /humu.9481

2 2 Møller et al. mutations have been identified, and about 90% of these are single base mutations. The rate of unidentified mutations is low. Comprehensive mutation screening using DGGE and sequencing failed to detect mutations on 3.3 % (20/594) of PKU alleles in Danish patients born between 1961 and 2006 (unpublished data by LB Møller), and on 1.6 % (9/546) of PKU alleles in a study from Germany (Zschocke and Hoffmann, 1999; unpublished data by J. Zschocke). However, standard mutation scanning methods such as DGGE do not identify deletions encompassing one or more whole exons, except in a homozygous state, due to the masking of the non-deleted allele. In order to test if whole exon deletions or duplications are present on some of the alleles with an unidentified mutation, we analyzed 34 PKU patients from Denmark and Germany with at most one mutation identified, using Multiplex Ligation-dependent Probe Amplification (MLPA) (Schouten et al., 2002). This technique is new and fast and has already proven useful in quantification of the exon numbers in several diseasecausing genes (Hartmann et al., 2004; Janssen et al., 2005). MATERIALS AND METHODS Samples Thirty-four PKU patients, in whom only one or no mutation was identified in the PAH gene (GenBank: NM_ ), were included in this study. All patients had a clinical diagnosis of phenylketonuria as per validated clinical and biochemical criteria (Güttler 1980) and had previously been scanned for DNA alterations by DGGE (Guldberg et al., 1993; Zschocke and Hoffmann, 1999). Fourteen of the patients were from a cohort of 297 independent PKU patients, born in Denmark between 1961 and 2006, comprising almost all Danish PKU patients for that period. In six of the fourteen patients no mutations had been detected while only one mutation had been identified in the remaining eight. Twenty patients were from Germany. Of these, eight were from the previously published cohort of 273 independent PKU patients living in Germany (Zschocke and Hoffmann, 1999). In this cohort one allele harboured a deletion of exons 9 to11 previously identified through apparent non-paternity (Zschocke et al., 1999). There were originally eleven independent patients in this cohort, in whom only one mutation had been detected; sufficient DNA was not available in one of these patients whilst two of them were later shown to have a mutation at the splice donor site of intron 4 (missed by previously used DGGE primers, unpublished data by J Zschocke). The remaining eight patients were included in this study. Twelve additional PKU patients from Germany with one mutation missing were also investigated. MLPA Genomic DNA was extracted from leucocytes using standard methods. The MLPA PAH kit (SALSA P055) was obtained from MRC-Holland (Amsterdam, The Netherlands). The kit contains probes for all of the 13 encoding exons of the PAH gene. In addition two probes for the neighbouring genes ASCL1 (upstream PAH) and IGF1 (downstream PAH) and 10 control probes located on other chromosomes are included in the kit. Details on probe sequences can be found at the company s website ( MLPA was performed according to the manufacturer s protocol using 50 ng of genomic DNA per reaction. All runs included DNA from 2 unrelated unaffected control persons. Reaction products were detected on an ABI model 310 capillary sequencer (Applied Biosystems, Denmark) with the GeneScan 3.1 software to size the PCR products and to obtain peak areas. The results were exported into a Microsoft Excel spreadsheet (Microsoft, htpp:// or a custom MLPA analysis programme (SeqPilot, JSI Medisys, Kippenheim, Germany). The relative peak area (RPA) of each probe was determined by dividing the peak area of each individual probe by the sum of all 23 peak areas obtained on that sample. The RPA from each probe was then compared to that of a control sample by dividing the relative peak area with the relative peak area for the same probe obtained from a control sample, resulting in an RPA ratio. An RPA ratio of approximately 100 % is normal whilst a reduction to 50 % indicates a deletion and an increase to 150 % a duplication of one allele (Hartmann et al., 2004, Janssen et al., 2005). MLPA was performed twice with comparable results.

3 Exon Deletions in PAH 3 DNA Sequencing Sequencing of the exon 13 PCR amplified product was performed directly with a 32 P-labeled primer and ThermoSequenase (Unites states Biochemical). RESULTS In this study thirty-four PKU patients, in whom only one or no mutation was identified in the PAH gene, were analyzed for deletions and duplications by MLPA. Twenty-two of the patients were from a combined cohort of 570 independent PKU patients from Denmark and Germany. The remaining twelve were additional PKU patients from Germany with one mutation missing. In order to test the sensitivity and accuracy of the MLPA method we initially examined DNA samples with a proven deletion. We have previously identified deletions in the PAH gene of several patients using a combination of the comparative multiplex dosage analysis (CMDA) and Southern blot (Gable et al., 2003). We re-examined three of these samples (a-c, Table 1) with a known exon 5 deletion (c.442-?_509+?del) and two samples (d-e, Table 1) with a known exon 6 deletion (c.510?_706+?del) by MLPA. These experiments were performed blindly. As illustrated in Table 1, all the deletions were also detected by MLPA. Also exon 6 duplications were suspected with CMDA in two patients (Gable et al., 2003). The value of this finding had remained unclear, as the duplication could not be verified by Southern blot (Gable et al., 2003). Likewise, MLPA analyses failed to confirm the putative exon 6 duplications (g-h, Table 1). Table 1. RPA Ratios for the PAH Located MLPA Probes in Seven Patients Previously Investigated by CMDA PAH exon /patient a b c d e f g Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex The identity of seven samples are DB (pt a), MW (pt b), LBR (pt c), JPH (pt d), JML (pt e), MO (pt f), and CR (pt g). GenBank sequence NM_ Considering that MLPA is well suited for the detection of duplications (Janssen et al., 2005; unpublished results by LB Møller), and the RPA of exon 6 in the present study was close to 1, indistinguishable from the unaffected control samples (not shown), we assume that the reported duplication may be an artefact of the CMDA analysis. Abnormal RPA ratios were observed in seven out of 34 patients. Ratios for these patients are given in Table 2. Segregating deletions were suspected in a total of 7 families. A deletion of exon 3 (c.169-?_352+?del), a deletion of exon 5 (c.442-?_509+?del) and a deletion of exons 1+2 (c.1-?_168+?del) was found in four, two and one family respectively. Duplications were not observed. To test if the low signal obtained by MLPA is indeed due to deletion of an exon and not due to a sequence variant in the probes target sequences we screened all the suspected exons for polymorphisms by DGGE. No DNA variants could be identified in these patients. Thus a deletion could be identified in about 20% (7/34) of the analyzed families. The identified mutations are summarized in Table 3.

4 4 Møller et al. Table 2. Abnormal RPA Ratios for PAH Located MLPA Probes in Seven of 34 Investigated Patients PAH exon /patient Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex Ex The identity of the seven samples are: pt 1 belongs to the group of 297 Danish PKU patients, pts. 2 and 6 belong to the previously analyzed group of 273 PKU patients living in Germany (Zschocke and Hoffmann 1999). The remaining patients 3, 4, 5, and 7 belong to the group of recently analyzed German PKU patients. GenBank sequence NM_ Table 3. Summary of Exon Deletions Identified by MLPA Region Mutation Predicted effect Exon 1+2 c.1-?_168+?del Activation of alternative initiation ATG codon Exon 3 c.169-?_352+?del Frameshift Exon 5 c.442-?_509+?del Frameshift DISCUSSION In this study, MLPA was used to test for the presence of deletions or duplications in the PAH gene in a cohort of PKU patients in whom at least one mutation had remained undetected using standard diagnostic tools. Dosage imbalances were detected using this approach and allowed calculation of the frequency of exon deletions in the PAH gene as the cause of PKU. Of the combined cohort of 570 independent PKU patients we identified exon deletions in a total of four patients including the previously reported patient with an exon 9-11 deletion (Zschocke et al., 1999). This corresponds to an allelic frequency of less than 0.5 % for deletions. No duplications were identified. Large genomic deletions thus are rare causes of PKU in patients from Denmark or Germany. The prevalence of genomic deletions may be different in other populations. For example, the sole reported defect causing PKU in Yemenite Jews is a deletion of exon 3 (Avigad et al., 1990). Few other examples of exon deletions have been reported. Large deletions covering exons 1-5 and exons 9-13 in a homozygous form or exon 9-11 in a heterozygous form were identified in patients from India or Germany respectively (Guldberg et al., 1997; Zschocke et al., 1999). A deletion of exon 2-3 has been identified in a Scottish patient (Sullivan et al., 1989) a

5 Exon Deletions in PAH 5 deletion of exon 3 (different from the Yemenite deletion) has been identified in a patient from Sicily (Bosco et al., 1996) and a deletion of exons 5-6 in patients from Japan (Okano et al., 1994). CMDA relies on multiplex amplification with a large number of different primers, whilst MLPA uses a single universal primer set for quantification, which makes the exon quantification much easier. MLPA identified the deletions of exons 5 and 6 previously detected by CMDA (Gable et al., 2003). However, the putative exon 6 duplication was not confirmed, and we assume that the reported duplication is the result of an artefact of CMDA analysis. During the course of the study we identified a novel point mutation that had been previously missed in one family. MLPA analysis in a patient with one unidentified PAH allele showed an apparent deletion of exon 13. Reexamination of this exon by DGGE and sequencing revealed a single base (A) insertion in the A-repeat that constitutes part of the stop codon. This variant, denoted c.1360_1361insa, is located within the ligation site for the exon 13 specific MLPA probes. It changes the original sequence TAA AGC into TAA AAG C (stop codon in italics) and is not predicted to change the amino acid constitution of the PAH protein; however, it has not yet been previously identified by us or reported in the literature and a functional relevance cannot be ruled out. This patient was subsequently removed from the MLPA study and is not included in the 34 patients reported here. The case highlights the need to confirm absence of DNA sequence variants in apparent single exon deletions recognized by MLPA (Janssen et al., 2005). In the present study, exon deletions were the causative mutations in 7 of the 34 patients investigated. The molecular basis of PKU in the remaining 27 patients with at least one uncharacterized PKU allele remains to be identified. Future investigations will reveal if these mutations are located outside the exon regions investigated, i.e. in the intronic sequences, in the promoter region or at the polyadenylation sites. Identification of disease causing deletions in the large introns of the PAH gene affecting the resulting encoding transcript is difficult, as it would require cdna analysis, which is not easily possible since normal expression of PAH is restricted to the liver. REFERENCES Avigad S, Cohen BE, Bauer S, Schwartz G, Frydman M, Woo SL, Niny Y, Shiloh Y A single origin of phenylketonuria in Yemenite Jews. Nature 344: Bosco P, Ceratto N, Cali F, Goltsov AA, Eisensmith RC, Novelli G, Dalla Piccola B, Romano V RFLP discordance in a PKU family due to a deletion in the PAH gene.turk J Pediatr 38: Gable M, Williams M, Stephenson A, Okano Y, Ring S, Hurtubise M, Tyfield L Comparative multiplex dosage analysis detects whole exon deletions at the phenylalanine hydroxylase locus. Hum Mutat 21: Guldberg P, Henriksen KF, Güttler F Molecular analysis of phenylketonuria in Denmark: 99% of the mutations detected by denaturing gradient gel electrophoresis. Genomics 17: Guldberg P, Güttler F Broad-range DGGE for single-step mutation scanning of entire genes: application to human phenylalanine hydroxylase gene. Nucleic Acids Res 22: Guldberg P, Henriksen KF, Mammen KC, Levy HL, Güttler F Large deletions in the phenylalanine hydroxylase gene as a cause of phenylketonuria in India. J Inherit Metab Dis 20: Güttler F. (1980). Hyperphenylalaninemia: diagnosis and classification of the various types of phenylalanine hydroxylase deficiency in childhood. Acta Paediatr. Scand. 280 (Suppl.):1-80. Hartmann C, John AL, Klaes R, Hofmann W, Bielen R, Koehler R, Janssen B, Bartram CR, Arnold N, Zschocke J (2004) Large BRCA1 gene deletions are found in 3% of German high-risk breast cancer families. Hum Mutat 24:534 Janssen B, Hartmann C, Scholz V, Jauch A, Zschocke J (2005) MLPA analysis for the detection of deletions, duplications and complex rearrangements in the dystrophin gene: potential and pitfalls. Neurogenetics 6:29-35 Mathias D, Bickel H Follow-up study of 16 years neonatal screening for inborn errors of metabolism in West Germany. Eur J Pediatr 145: Okano Y, Asada M, Kang Y, Nishi Y, Hase Y, Oura T, Isshiki G Molecular characterization of phenylketonuria in Japanese patients. Hum Genet 103:

6 6 Møller et al. Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepyens F, Pals G Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 30:e57. Sullivan SE, Moore SD, Connor JM, King M, Cockburn F, Steinmann B, Gitzelmann R, Daiger SP, Woo SL Haplotype distribution of the human phenylalanine hydroxylase locus in Scotland and Switzerland. Am J Hum Genet 44: Zschocke J, Hoffmann GF Phenylketonuria mutations in Germany. Hum Genet 104: Zschocke J, Ouak E, Knauer A, Fritz B, Aslan M, Hoffmann GF Large heterozygous deletion masquerading as homozygous missense mutation: a pitfall in diagnostic mutation analysis. J Inherit Metab Dis 22: