Somatic Cell and Molecular Genetics, Vol. 16, No. 1, 1990, pp. 85-90

Molecular Genetics of PKU in Eastern Europe: A Nonsense Mutation Associated with Haplotype 4 of the Pheny|alanine Hydyoxylase Gene Tao Wang, I Yoshiyuki Okano, I Randy C. Eisensmith, 1 Gyorgy Fekete, 2 Dezso Sehuler, 2 Gyyrgy Berencsi, 2 Istvan Nasz, 2 and Savio L.C. W o o 1 tHoward Hughes Medical Institute, Department of Cell Biology and Institute for Molecular Genetics, BayIor College of Medicine, Houston, Texas 77030; and 2Departments of Microbiology and Pediatrics II, Semmelweis University of Medicine, Budapest, Hungary Received 2 October 1989

Abstract--Phenylketonuria (PKU) & a genetic disorder secondary to a deficiency of hepatic phenyalanine hydroxylase (PAH). Several mutations in the PAH gene have recently been reported, and linkage disequilibrium was observed between RFLP haplotypes and specific mutations. A new molecular lesion has been identified in exon 7 of the PAH gene in a Hungarian PKU patient by direct sequencing of PCR-amplified DNA. The C-to-T transition causes the substitution of Arg 243 to a termination codon, and the mutant allele is associated with haplotype 4 of the PAH gene. The mutation is present in two of nine mutant haptotype 4 alleles among Eastern Europeans and is not present among Western Europeans and Asians. The rarity of this mutant allele and its restricted geographic distribution suggest that the mutational event occurred recently on a normal haplotype 4 background in Eastern Europe.

types (1, 2, 3, and 4) comprised 80% of all mutant alleles in the Northern European population (11) and were shown to be prevalent throughout the European continent as well (9). Further analyses of the PAH gone in PKU patients demonstrate that PKU is caused by multiple mutations, which contribute to the heterogenous clinical course. Interestingly, all PAH mutations characterized in Caucasians so far are point mutations and are in obvious linkage disequlibrium with the respective RFLP haplotypes (12-16). Although mutant haplotypes 1, 2, 3, and 4 are prevalent throughout the European continent, mutant haplotypes 1 and 3 are much less common in Eastern Europe than in Western Europe. As the result, haplotypes 2 and 4 become the predominant mutant haplo-

INTRODUCTION Classical phenylketonuria (PKU), the most common of inherited metabolic disorders, is caused by a deficiency in hepatic phenylalanine hydroxylase (PAH). The decrease in PAH activity results in the accumulation of phenylalanine in serum and irreversible mental retardation in untreated PKU patients (1). The disease is prevalent among Cauc, asians and 1 in 50 individuals is a carrier of the trait (2, 3). The human PAH locus has been studied extensively since the isolation of a full-length human PAH cDNA (4-7). RFLP haplotype analysis at the PAH locus revealed the presence of at least 47 haplotypes among different ethnic groups (8-10). Four haplo85

0740-7750/90/0100-0085506.00/0 © 1990 Plenum Publishing Corporation

86

Wang et al.

and dTTP; 10 mM Tris HC1 (pH 8.3); 50 mM KC1; 1.5 mM MgC12, 0.01% (w/v) gelatin; and 0.5 ~M each of an amplification primer pair. Samples were denatured at 97°C for 7 min and annealed at 50°C for 1 rain. Two units of Taq DNA polymerase (Perkin Elmer Cetus) were added and samples were incubated at 72°C for 1 min. The following cycles consisted of DNA denaturation at 92°C for 10 sec, annealing at 50°C for 10 see, and primer extension at 720C for 40 sec. A total of 35 cycles were performed using the Perkin Elmer Thermal Cycler with final extension at 720C for 7 min. PCR products were purified using 4% Nusieve agarose gels and recovered using Gene Clean Kit (Bio-101) according to manufacturer's instruction. Direct Sequencing of PCR-Amplified Products. Direct sequencing was performed using PCR-amplified DNA according to a modified procedure of that reported by Winship (18): 150 ng of purified template (200300 bp in length) was mixed with 150 ng of sequencing primer (17 mer) in 6 #1 of 40 mM Tris HC1 (pH 7.5), 25 mM MgC12, 50 mM NaC1, 10% DMSO. The samples were denatured at 97°C for 3 rain and put on dry ice immediately. Then 4 ~1 of labelling mixture containing 50 mM DTT, 10 #Ci [35S]dATP M A T E R I A L S A N D METHODS (>1000 Ci/mmol, NEN), and 2 units of Sequenase (USB) was added. The resulting Patients. The pedigrees selected for pop10-/zl mixture was divided equally into four ulation studies are nuclear families with at tubes, each containing 2 #1 of 80 ttM dATP, least one PKU child. All patients were dTTP, dCTP; 50 mM NaC1; 10% DMSO; and diagnosed and evaluated at their institutions with standarded clinical criteria (3, 9, 17). 8 t~M ddGTP, 0.08 #M ddATP, 8 #M ddTTP, The proband for gene analysis was a classical or 8 izM ddCTP. The samples were incubated PKU child from Hungary who is a haplotype at 37°C for 5 min, 2 #1 of chase solution containing 50 mM NaC1, 10% DMSO, 0.25 4 and 8 compound heterozygote (9). PCR Amplification of Exonic Regions mM dGTP, dATP, dTTP, and dCTP were in PAH Gene of the Patient. Primers were then added and the reaction was kept at 37°C designed for PCR amplification of exonic for another 5 min. Then 4 Fzl of stop solution regions of the human PAH gene and were was added to each tube. The samples were synthesized by Genetic Designs Inc., Houston, heated at 75°C for 2 rain immediately before Texas. Individual exons with flanking intronic electrophoresis. Allele-Specific Oligonucleotide ttybridsequence of the PAH gene were amplified independently as the following: a 100-~1 ization and Population Screening. A pair of reaction volume containing 0.5 Izg of genomic oligonucleotides (17-mers) was synthesized DNA; 0.2 mM each of dATP, dGTP, dCTP according to the determined sequence for the

types in this area, accounting for 62% and ! 7% of the total mutant alleles, respectively (9). This haplotype distribution suggests that there may be some differences in the genetic background of PKU between the two populations. The mutation associated with haplotype 2 has already been identified, and linkage disequlibrium between mutation and haplotype is almost absolute. This mutation, an amino acid substitution (Arg4°8-Try 4°8) of PAH, produces no PAH activity and therefore a severe clinical course (13). Another mutation, identified recently in Caucasians as a single amino acid substitution (Arg 158Gin158), produces less than 10% of normal PAH activity and may be responsible for a mild PKU phenotype (16). Interestingly, the mutation shows an obvious exclusive but not inclusive linkage disequilibrium with mutant haplotype 4. This finding suggests that there must be other mutations on haplotype 4 background of the PAH gene that are responsible for PKU in this population. In this paper, we report the identification of a nonsense mutation in the PAH gene that is in linkage disequilibrium with mutant haplotype 4 alleles in Eastern Europe.

PKU in Eastern Europe

87'

normal and mutant alleles. Probes were end-labeled with [32P]ATP (6000 Ci/mmol, N E N ) using polynucleotide kinase (Biolabs). Genomic D N A samples were individually amplified by PCR as described, and 5 ~1 of amplified material was denatured in 0.2 M N a O H and dot-blotted on to zeta-probe membrane (Bio-Rad) in t M NH4OAC. Hybridization and washing was carried out as described previously (19). RESULTS Identification of Nonsense Mutation in Exon 7 o f PAH Gene. Amplification of a 291-bp fragment containing exon 7 plus flanking intronic sequences of the human P A H gene has previously been reported (16). This region of a Hungarian PKU patient's DNA was amplified by PCR, purified from a Nusieve agarose gel, and sequenced directly using a nested primer. A C-to-T transition is present at the first base of codon 243, resulting in the substitution of an Arg (CGA) to a stop

codon (TGA) (Fig. 1). Both C and T bands are present in the patient's sample with reduced intensity as compared to the C band in the normal allele, indicating only one of the: two PAH alleles in this patient bears the termination mutation. Mendelian Transmission of Mutant Allele. In order to verify the authenticity of' the nonsense mutation in the patient's genomic D N A and its Mendelian transmission, we analyzed genomic D N A samples from the proband and both parents by dot-blot hybridization of independently amplified samples using allele-specific oligonucteotide probes. The normal probe hybridized with all three members of the family, indicating successful PCR amplification (Fig. 2, panel A). The mutant probe hybridized with the proband in this family, demonstrating the authenticity of the mutation in the patient's genomic D N A (Fig. 2, panel A). The mutant probe also hybridized with the maternal sample but not with paternal sample, suggesting that mutant allele is transmitted from the mother to the

Arg

243

NORMAL

SEQUENCE

5'-TTCCGCCTCCGACCTGT-3'

MUTANT

SEQUENCE

5 '

T~

Ter

3 ' 243

Fig. 1. Identification of a nonsense mutation in exon 7 of the human PAH gene. Direct sequencing of the exon 7-containing regionsof PAH gene using a nested sequencingprimer (5'-AGATGACGCTCAGTGTG-3').Left part of the figureshows the correspondingportionof the normal exon7 sequenceand right part showsthe mutation-containing region in exon 7 of the patient's PAH gene. The C-to-T transition results in the substitution of the A r g 243 codonto a termination codon. Equal but reduced intensity of the C and T bands at the identical position of the sequence gel suggests the presenceof a normaland a mutant allelesin the patient's DNA (indicatedby line).

Wang et al.

88

Fig. 2. Mendelian transmission of the nonsense mutation in two PKU kindred. Exon 7-containing regions of PAHgene were PCR-amplified from genomic DNA of different family members using a pair of PCR primers (16). Amplified products (5 #1) were dot-blotted on to z-probe membrane and hybridized using normal and mutant oligonucleotide probes separately (19). The probes used for the detection were as follows: the normal probe (5'TTCCGCCTCCGACCTGT-3') is the sense strand and the mutant probe (5'-ACAGGTCAGAGGCGGAA-Y) is the antisense strand sequence covering the mutation. The proband from Hungary in family A was the one characterized by molecular analysis. Family B was characterized by population studies from a Czechoslovakia PKU family. Solid symbols in pedigrees represent the Arg 243 to Ter 243 mutant allele, hatched symbtes for uncharacterized P A H mutant allele and open symbles for normal P A H alleles. The appropriate haplotypes of each family members are indicated at the bottom of pedigrees.

patient and is present on a haplotype 4 background. The paternal mutant allele associated with haplotype 8 did not show hybridization, suggesting that it is a different mutant allele, and the patient must therefore be an allelic compound. Population Genetics of Mutant Allele among Europeans and Orientals. To determine the distribution of the mutant allele in different populations and if there is linkage disequilibrium with the specific haplotype, we analyzed a total of 178 mutant and 147 normal alleles representing various regions in Europe and Asia. Using allele-specific oligonucleotide probes, we found another PKU family bearing the same mutation in Czechoslavakia. In this family, the mutant allele is paternal in origin and occurred also on haplotype 4

background (Fig. 2, panel B). Population screening indicated that two of nine mutant haplotype 4 alleles in the Eastern European countries bear this mutation (Table 1). This mutation is not present in mutant alleles of other haplotypes and is absent in Western Europeans and Asians. DISCUSSION We identified a new molecular lesion in the PAH gene of a Hungarian PKU patient. A nucleotide substitution (C to T) in exon 7 of the PAH gene results in the change of an arginine codon CGA z43 to the termination codon TGA. The mutation causes the loss of the carboxyl half of the PAH protein molecule, which constitutes the enzymatic core.

PKU in Eastern Europe

89

Table 1. PopulationGeneticsof Arg 243 to Ter243Mutation in Europe and Asia

Western EuropC

Eastern Europe"

Asian

Haplotypes

Normal

Mutantb

Normal

Mutantb

Normal

Mutantb

1 2 3 4 5-23

0/27 0/3 0/1 0/26 0/22

0/41 0/15 0/23 0/20 0/10

0/13 0/3 0/1 0/6 0/tl

0/5 0/15 0/1 2/9 0/5

0/0 0/1 0/3 0/23 0/7

0/1 0/1 0/1 0/24 0/7

~Western Europe:Switzerlandand Denmark; Eastern Europe: Hungary and Czechoslovakia;Asia: China and Japan. bNumber of hybridizationsignals/number of alleles analyzed.

To date 47 R F L P haplotypes in the human P A H locus have been reported (8, 9), and haplotypes 1-4 account for the majority of PKU-bearing chromosomes in Caucasians of northern European ancestry. The mutations associated with haplotypes 2 and 3 have been characterized (t2, 13), and linkage disequilibrium between mutation and hapIotype is maintained throughout the European continent (20). Both mutations produce no PAH activity in the cell and are responsible for severe clinical phenotypes of the patients. However, PKU associated with R F L P haplotypes 1 and 4 varies significantly in the biochemical and clinical phenotypes, ranging from mild to very severe forms (3). A PKU mutation characterized recently in a Swiss PKU patient was found to cause mild PKU and is tightly linked with mutant haplotype 4 throughout Europe (16). The identification of a distinct mutation on a haplotype 4 background reported in this study provides unambiguous evidence for the hypothesis that multiple PKU mutations have occurred on the same haplotype background. Population genetic analysis indicated that this mutation is rare, but accounts for about 20% of all haplotype 4 mutant alleles in Eastern Europe. Interestingly, this mutation is absent in Western Europeans and Asians. Thus the mutation must have occurred after racial divergence. Furthermore, in contrast to the two well-characterized mutations associated with haplotypes 2 and 3 of the P A H gene that are distributed throughout Europe, the rarity

and restricted geographic distribution of this mutant allele suggest that it may be the result of a very recent mutational event on a haplotype 4 allele in Eastern Europe. ACKNOWLEDGMENTS This work was supported in part by N I H grant HD-17711 to S.L.C. Woo, who is also an Investigator with the Howard Hughes Medical Institute. L I T E R A T U R E CITED 1. Foiling,A. (1934). Z. Physiol. Chem. 227:169-176. 2. Scriver, C.R., Kaufman, S., and Woo, S.L.C. (1988). Annu. Rev. Genet. 14:179-202. 3. Guttler, F. (1980). Acta Pediatro Seand. SuppL 280:77-80. 4. Woo,S.L.C., Lidsky,A.S., Guttler, F., Chandra, T., and Robson,K.J.H. (1983). Nature 306:151-155. 5. Kwok, S.C.M., Ledley, F.D., DiLella, A.G., Robson, D.J.H., and Woo,S.L.C. (1985). Biochemistry 24:556-56I. 6. Ledley,F.D., Grenett, H.E., DiLella, A.G., Kwok, S.C.M., and Woo, S.L.C. (1985). Science 228:7% 79. 7. DiLella,A.G, Kwok,S.C.M., Ledley,F.D., MarviL J., and Woo, S.L.C. (1986). Biochemistry 25:743749. 8. Woo,S.L.C. (1988). Am. J. Hum. Genet. 43:781783. 9. Daiger,S.P., Chakraborty,R., Reed, L., Fekete,G., Schuler, D., Berenssi, G., Nasz, I., Brdicka, R., Kamaryt, J., Pijackove,A., Moore,S., Sullivan, S., and Woo,S.L.C., et aL (1989).Am. Z Hum. Genet. 45:310--318. 10. Daiger, S.P., Reed, L., Huang, S.Z., Zeng, Y.T., Wang, T, Lo, W.H.Y., Okano, Y., Hase, Y., Fukuda, Y., Oura, T., Tada, K., and Woo, S.L.C. (1989). Am. J. Hum. Genet. 45:319-324.

90

11. Chakraborty, R., Lidsky, A.S., Daiger, S.P., Guttier, F., Sullivan, S., DiLella, A.G., and Woo, S.L.C. (1987). Hum. Genet. 76:40-46. 12. DiLella, A.G., Marvit, J., Lidsky, A.S, Guttler, F., and Woo, S.L.C. (1986). Nature 322:799-803. 13. DiLetla, A.G., Marvit, J., Brayton, K., and Woo, S.L.C. (1987). Nature 327:333-336. 14. Lichter, U., Konecki, D.S., DiLella, A.G., Brayton, K., Marvit, J., Hahn, T.M., Trefz, F,K., and Woo, S.L.C. (1988). Biochemistry 27:2881-2885. 15. Lyonnet, S., Caillaud, C., Rey, F., Berthelon, M.,

Wang et al.

Frezal, J., Rey, J., and Munnich, A. (1989). Am. J. Hum. Genet. 44:511-517.

16. Okano, Y,, Wang, T., Eisensmith, R.C., Gitzelmann, R., and Woo, S.L.C. (1990). Am. J. Hum. Genet. (in press January). 17. Winship, P.R. (1989). Nucleic Acid Res. 17:1266. 18. Guthrie, R , and Susi, A. (1963). Pediatrics 32:338-343, 19. DiLella, A.G., Huang, W.M., and Woo, S.L.C. (1988). Lancet 1:497-500. 20. Woo, S.L.C. (1989). Biochemistry 28:1-7.