Molecular Genetics and Metabolism 91 (2007) 157–164 www.elsevier.com/locate/ymgme
Clinical, biochemical and genetic findings in two siblings with a dihydropyrimidinase deficiency Andre´ B.P. van Kuilenburg a,b,c,*, Judith Meijer a,b,c, Doreen Dobritzsch d, Rutger Meinsma a,b,c, Marinus Duran a,b,c, Bernhard Lohkamp d, Lida Zoetekouw a,b,c, Nico G.G.M. Abeling a,b,c, Herman L.G. van Tinteren e, Annet M. Bosch a,b,c a
Academic Medical Center, University of Amsterdam, Emma Children’s Hospital, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands b Department of Clinical Chemistry, University of Amsterdam, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands c Department of Pediatrics, University of Amsterdam, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands d Karolinska Institute, Department of Medical Biochemistry and Biophysics, S-17177 Stockholm, Sweden e Tergooiziekenhuizen, Department of Pediatrics, Hilversum, The Netherlands Received 20 November 2006; received in revised form 9 February 2007; accepted 9 February 2007 Available online 26 March 2007
Abstract Dihydropyrimidinase (DHP) is the second enzyme of the pyrimidine degradation pathway and it catalyses the ring opening of 5,6dihydrouracil and 5,6-dihydrothymine to N-carbamyl-b-alanine and N-carbamyl-b-aminoisobutyric acid, respectively. To date, only nine individuals have been reported suffering from a complete DHP deficiency. We report two siblings presenting with strongly elevated levels of 5,6-dihydrouracil and 5,6-dihydrothymine in plasma, cerebrospinal fluid and urine. One of the siblings had a severe delay in speech development and white matter abnormalities, whereas the other one was free of symptoms. Analysis of the DHP gene (DPYS) showed that both patients were compound heterozygous for the missense mutation 1078T > C (W360R) in exon 6 and a novel missense mutation 1235G > T (R412M) in exon 7. Heterologous expression of the mutant enzymes in Escherichia coli showed that both missense mutations resulted in a mutant DHP enzyme without residual activity. Analysis of the crystal structure of eukaryotic DHP from the yeast Saccharomyces kluyveri and the slime mold Dictyostelium discoideum suggests that the W360R and R412M mutations lead to structural instability of the enzyme which could potentially impair the assembly of the tetramer. 2007 Elsevier Inc. All rights reserved. Keywords: Crystal structure; Dihydropyrimidinase; Dihydrouracil; Dihydrothymine; DPYS; Pyrimidines, 5-Fluorouracil
In humans, the pyrimidine bases uracil and thymine are degraded via a three step pathway to b-alanine and b-aminoisobutyric acid, respectively. Dihydropyrimidine dehydrogenase (DPD, EC 1.3.1.2) is the initial and rate-limiting enzyme, catalysing the reduction of uracil and thymine to 5,6-dihydrouracil and 5,6-dihydrothymine, respectively. The second step consists of a hydrolytic ring opening of the dihy*
Corresponding author. Present address: Academic Medical Center, Laboratory Genetic Metabolic diseases, F0-224, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. Fax: +31 206962596. E-mail address:
[email protected] (A.B.P. van Kuilenburg). 1096-7192/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2007.02.008
dropyrimidines which is catalysed by dihydropyrimidinase (DHP, EC 3.5.2.2). Finally, the resulting N-carbamyl-b-aminoisobutyric acid and N-carbamyl-b-alanine are converted in the third step to b-aminoisobutyric acid and b-alanine, ammonia and CO2 by b-ureidopropionase (EC 3.5.1.6). The enzymes of the pyrimidine degradation pathway are also involved in the degradation of the widely used antineoplastic agent 5-fluorouracil (5FU) [1]. Patients with a deficiency of DPD have an increased risk of developing severe toxicity after the administration of 5FU [1–3]. Recently, it has been shown that also patients with a DHP deficiency are prone to the development of severe 5FU-associated toxicity [4,5].
158
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
To date, only nine patients have been described with a complete DHP deficiency (MIM 222748). The clinical phenotype of these patients was highly variable, ranging from asymptomatic to seizures, mental retardation, growth retardation and dysmorphic features [4,6–13]. The underlying mechanism of the observed clinical abnormalities is still not known. In these patients, a large accumulation of dihydrouracil and dihydrothymine could be detected in urine, blood and cerebrospinal fluid, whereas no activity of DHP could be detected in liver tissue [10,14]. The DPYS gene has been mapped to chromosome 8q22 and consists of 10 exons spanning >80 kb of genomic DNA [13]. The cDNA coding for human DHP contains an open reading frame of 1560 nucleotides, corresponding to a protein of 519 amino acids with a calculated molecular weight of 56629 Da [13]. DHP has been purified and extensively characterised from a large number of eukaryotic species [15–18]. These studies showed that the native enzyme is composed of four identical subunits containing two zinc ions [18]. Sequence alignments suggest a close relationship between eukaryotic DHP and the hydantoinases in bacteria and several proteins involved in neuronal development, such as collapsin response mediator proteins (CRMPs) [19,20]. Recently, the crystal structure of eukaryotic DHP from the yeast Saccharomyces kluyveri (SkDHP) and the slime mold Dictyostelium discoideum (DdDHP) has been determined [18]. This allows for the first time to investigate mutations in a three dimensional framework. In this study, we report the clinical, biochemical and genetic findings of two newly identified patients with a complete DHP deficiency as well as the analysis of the mutations at the protein structure level.
The figures showing the point mutation sites and their environment in a model of the human DHP protein were generated using the program PyMOL (DeLano, W.L. The PyMOL Molecular Graphics System (2002) on World Wide Web http://www.pymol.org). The modelling of the human DHP structure and comparisons were based on the crystal structures of SkDHP (PDB accession code: 2fty), DdDHP (2ftw), mouse CRMP1 (1kcx) and of several bacterial hydantoinases (1nfg, 1kld, 1gkq, 1yny, 1gkr).
Materials and methods
Results
Analysis of pyrimidines in body fluids
Clinical evaluation
The concentrations of uracil, thymine, dihydrouracil, dihydrothymine, N-carbamyl-b-alanine and N-carbamyl-b-aminoisobutyric acid in urine, plasma and CSF were determined using reversed-phase HPLC combined with electrospray tandem mass spectrometry [21].
Patient A is the eldest son of non-consanguineous Moroccan parents. At the age of 4 years, he presented with a severe delay in speech development, whereas his motor development was normal. He was communicative, using sentences of 2–3 words. On physical examination, height (15th centile) and weight (50th centile) were normal and no dysmorphic features were noted. An extensive neurological examination demonstrated no abnormalities. His IQ was measured with an age-specific non-verbal intelligence test and was 93. MRI scan of the brain demonstrated slight widening of the lateral ventricles and thinning of the parieto-occipital white matter. Patient B is the 1-year younger brother of patient A. He was screened for dihydropyrimidinase deficiency after his brother had been diagnosed. His psychomotor development was normal and he attended a normal primary school. However, he was recently enrolled in a remedial teaching program because of insufficient mastery of the Dutch language which was attributed to the fact that his parents did not speak Dutch at home. Physical examination, including an extensive neurological examination,
PCR amplification of coding exons DNA was isolated from leukocytes by standard procedures. The PCR amplification of all nine coding exons and flanking intronic regions was carried out using the primer sets as described previously [13]. For exon 2–9, however, the sequence of the forward primers had a 5 0 -TGTAAAA CGACGGCCAGT-3 0 extension, whereas reverse primers had an 5 0 -CAGG AAACAGCTATGACC-3 0 extension at their 5 0 -ends. These sequences were complementary to the labeled, 21M13 and M13, reversed primers used in the dye-primer sequence reaction. PCR products were separated on 1% agarose gels, visualized with ethidium bromide and used for direct sequencing.
Expression of DHP in Escherichia coli To obtain an expression plasmid containing the wild-type human DHP cDNA (pSE420-DHP), the EcoRI–BamHI insert of a plasmid comprising the complete coding region of the human DHP gene was subcloned into the pSE420 vector [20]. The mutations 1078T > C (W360R) and 1235G > T (R412M) were introduced into the human DHP sequence using the megaprimer technique and the resulting mutant genes cloned
into the pSE420 vector (pSE420-DHP-W360R and pSE420-DHPR412M). The resulting clones were sequenced completely to verify the presence of the mutation and to exclude the presence of random mutations introduced by PCR artifacts. Expression plasmids were introduced into the Escherichia coli strain BL21 and wild-type DHP and mutant DHPs were expressed as described before [5]. The activity of DHP was determined using a radiochemical assay with subsequent separation of radiolabeled dihydrouracil from N-carbamyl-b-alanine with reversed-phase HPLC [22].
Western blot analysis Cell extracts (2.5 lg) were fractionated on a 7.5% (w/v) SDS–polyacrylamide gel and transferred to a nitrocellulose filter. Detection of the DHP proteins was performed using a polyclonal antibody against rat DHP [5]. Antigen–antibody complexes were visualized using anti-rabbit antibody conjugated to alkaline phosphatase.
Sequence analysis Sequence analysis of genomic fragments amplified by PCR and expression plasmids was carried out on an Applied Biosystems model 3100 automated DNA sequencer using the dye-terminator method (Perkin-Elmer Corp., Foster City, CA, USA).
Analysis of the crystal structure
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
demonstrated no abnormalities. His IQ, measured with an age specific non-verbal intelligence test, was 83. Pyrimidine bases and degradation products in body fluids The urine samples of the two patients showed normal to marginally elevated levels of uracil and thymine compared to controls. In contrast, the concentrations of dihydrouracil and dihydrothymine were strongly elevated, when compared to those observed in controls (Table 1). Furthermore, no N-carbamyl-b-alanine and N-carbamyl-b-aminoisobutyric acid could be detected. The observed dihydropyrimidinuria in the two siblings strongly suggested a deficiency of dihydropyrimidinase. Analysis of plasma and cerebrospinal fluid of patient A showed a similar pattern as observed in urine. The concentrations of dihydrouracil and dihydrothymine were strongly increased when compared to controls, whereas the levels of the pyrimidine bases were normal to slightly increased and the N-carbamyl-b-amino acids were undetectably low.
159
the 1235G > T mutation results in the amino acid substitution R412M. Analysis of DPYS of the parents showed that the father was heterozygous for the 1078T > C mutation and the mother was heterozygous for the 1235G > T mutation. To investigate the effect of the W360R and R412M mutation on the activity of DHP, the mutations were introduced into the pSE420-DHP vector by site-directed mutagenesis and expressed in E. coli. No endogenous DHP activity (<0.3 nmol mg-1 h-1) could be detected in the E. coli strain used for the expression of the constructs. Introduction of the wild-type DHP construct increased the DHP activity more than 600· above the background
Analysis of the DPYS mutations and expression of mutant DHP proteins Analysis of the genomic sequences of exons 1–9 of DPYS, including their flanking intronic sequences, showed that both patients were compound heterozygous for the missense mutation 1078T > C in exon 6 and the novel missense mutation 1235G > T in exon 7. The 1078T > C mutation results in the amino acid substitution W360R, whereas
Fig. 1. DHP activity and immunoblot analysis of wild-type and mutant DHPs expressed in E.coli. The activity of DHP was assayed in E. coli lysates. The results represent the mean of three to six independent measurements and the error bar represents 1 SD. For immunoblot analysis, equal amounts of protein (2.5 lg) were separated by 7.5% SDS– PAGE and analyzed on immunoblot with DHP-specific antibodies (insert).
Table 1 Pyrimidine degradation metabolites of patients with a DHP deficiency Patient A
Patient B
Controlsa (mean ± SD)
Urine (lmol/mmol creatinine) Uracil Thymine Dihydrouracil Dihydrothymine N-Carbamyl-b-alanine N-Carbamyl-b-aminoisobutyric acid
14 5 142 176 <0.5 <0.5
21 7 192 158 <0.4 <0.2
7.0 ± 5.4 0.1 ± 0.3 2.1 ± 1.6 1.0 ± 0.7 2.8 ± 2.0 0.1 ± 0.4
Plasma (lM) Uracil Thymine Dihydrouracil Dihydrothymine N-Carbamyl-b-alanine N-Carbamyl-b-aminoisobutyric acid
<1 0.24 9.5 27.4 <1 <0.5
1.4 0.7 25.8 25.9 <2 <1
0.1 ± 0.3 (n = 55) <0.1 (n = 55) 1.6 ± 1.0 (n = 55) 0.7 ± 0.2 (n = 55) 0.1 ± 0.2 (n = 55) 0.1 ± 0.2 (n = 55)
Cerebrospinal fluid (lM) Uracil Thymine Dihydrouracil Dihydrothymine N-Carbamyl-b-alanine N-Carbamyl-b-aminoisobutyric acid
<1 <0.5 11.1 24 <1 <0.5
n.a. n.a. n.a. n.a. n.a. n.a.
<0.4 (n = 16) <0.1 (n = 16) 2.2 ± 0.9 (n = 16) 0.8 ± 0.2 (n = 16) <0.5 (n = 16) <0.1 (n = 16)
(n = 111) (n = 111) (n = 111) (n = 111) (n = 111) (n = 111)
n.a., not available. a Age matched controls (>3 years). In case the concentration of a metabolite was below the detection limit, the value has been set at 0 for this sample.
160
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
level. Expression of the DHP constructs containing the W360R or R412M mutation yielded no detectable DHP activity (Fig. 1). To exclude the possibility that the lack of enzyme activity was the result of an inability to produce the mutant proteins in E. coli, the expression levels were analysed by
immunoblotting. Fig. 1 shows that the mutant proteins carrying the W360R or R412M mutation were readily expressed. Thus, the lack of DHP activity in E. coli transfected with the constructs containing the mutations was not due to rapid degradation of the mutant DHPs in the E. coli lysates.
Fig. 2. Structural homology between DdDHP and related proteins, and location of the mutation sites. (a) Shows the subunit structure of the DdDHP coloured according to its structural conservation to SkDHP, mouse CRMP1, D-specific hydantoinases from four different species, and to the L-stereospecific hydantoinase from Arthobacter aurescens. The degree of structural conservation decreases from blue (corresponding structures can be aligned for all homologues), via cyan, white, yellow, orange, magenta and red to purple (DdDHP structure not conserved in any of the other proteins). The approximate location of W360 and R412 in the human enzyme is indicated by space-fill representation of the corresponding DdDHP residues W359 and K411. The two zinc-ions of the di-metal centre are shown as black spheres. (b) Shows a model of the human DHP tetramer, which is based on the structure of DdDHP. Each subunit is shown in a different colour. The location of the mutation sites is indicated in one of the subunits by the space-fill representation of W360 and R412. The helix harbouring the W360R site is marked in pink. The figure on the right zooms in on the mutation sites, showing them in a slightly different orientation than in the figure on the left. The mutation sites are located close to an oligomerization interface within the DHP tetramer. The C-terminal tail sequence of one subunit (orange) interacts with a loop-region following the mutation site in another subunit (green), as well as with residues from a third monomer (blue). (For interpretation of the references in colour in this figure legend, the reader is referred to the web version of this article.)
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
Analysis of the structural effects of the DHP mutations By comparison of the available crystals structures for two DHPs [18], five hydantoinases [23–27] and one CRMP [28] structurally well-conserved areas could be identified. Within these areas, the structure of the human DHP can be modelled with relatively high confidence. Fig. 2a shows the subunit of DdDHP colored according to its structural homology to the other above mentioned proteins. Variable regions are almost exclusively limited to the N- and C-terminal sequences and to surface-exposed loops at the periphery of the molecule. Furthermore, the sequences of DdDHP and the human enzyme are 58% identical, and additional 15% of the amino acids are conservatively exchanged. Non-conservative exchanges mostly regard residues located on the protein surface whose side chains are oriented towards the bulk solvent, and those occurring in the subunit core are often accompanied by compensating exchanges of spatially close amino acids. It is, noteworthy that modelling of the human sequence onto the structure of DdDHP does not cause any serious clashes between amino acid residues which can not easily be avoided by selecting another rotamer of the corresponding or neighbouring side chain.
161
The monomer can be divided into two domains. The larger domain has a (b/a)8-barrel core embedding the catalytic di-zinc centre, whereas the characterizing feature of the smaller domain is a b-sandwich formed by a sevenstranded and a four-stranded mixed b-sheet. Both domains are contributing to intermolecular contacts within the physiologically relevant DHP tetramer (Fig. 2b). W360 belongs to the larger and R412 to the b-sandwich domain. Both are located spatially close to each other near the surface of the enzyme but rather distant from the active site (Fig. 2a and b), which suggests that the lack of residual activity for the point mutants is based on global effects of the exchanges on the protein structure. The tryptophan side chain (W360) is conserved between human and DdDHP and most other DHPs as well as in mouse CRMP1. In hydantoinases and SkDHP, the tryptophan is replaced by tyrosine or other hydrophobic residues (Fig. 3). The environment of W360 varies very little in structure and sequence between the homologues (Fig. 2a). Thus, it is likely that also in human DHP the side chain of W360 is surrounded by exclusively hydrophobic residues (Fig. 4). Its mutation to arginine would introduce a charged residue into a highly hydrophobic context and
Fig. 3. Alignment of amino acid sequences of diverse DHPs, Hyds and mouse CRMP to residues 332-431 of human DHP. Used abbreviations correspond to the following: Hs, Homo sapiens, Rn, Rattus norvegicus, Mm, Mus musculus, Dm, Drosophila melanogaster, Aa, Arabidopsis thaliana, Dd, Dictyostelium discoideum, Sk, Saccharomyces kluyveri, B9, Bacillus sp. AR9, Bs, Bacillus stearothermophilus, Bp, Burkholderia pickettii, Tsp, Thermus sp., Aa, Arthobacter aurescens. DHyd and LHyd are used to distinguish between hydantoinases with D- and L-stereospecificity. The described mutation sites W360 and R412 are highlighted by a box. Some of the residues interacting with K411 in DdDHP, which corresponds to R412 in human DHP, are marked by an arrow. Conservation of amino acids of human DHP in the other sequences is indicated by dark-grey shading, conservative exchanges by light-grey shading. Enzymes of known crystal structure are marked by an asterisks.
162
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
Fig. 4. The environment of the mutation sites in a model of human DHP. The model is based on the structure of DdDHP, whose amino acid sequence was adjusted according to the sequence of the human enzyme. When necessary, the side chain conformations of the exchanged and/or of neighbouring residues were optimized. Stick representations of side chains belonging to the model of human DHP are shown in green. Side chains of DdDHP are depicted in magenta, when they are not conserved between the two enzymes. The helix harbouring the W360R site is marked in pink. The left figure shows a close-up view of the W360R mutation site and its environment. The native W360 and the corresponding arginine introduced by the mutation are shown with thicker sticks and with carbons atoms in green and yellow, respectively. Nitrogen atoms of these two side chains are shown in blue. The mutation site and the closest surrounding residues are labelled. The figure on the right shows a close-up view of the R412M mutation site. The side chains of the native R412 and of potentially interacting residues are shown with thicker sticks, carbon atoms in green, nitrogen atoms in blue, and oxygen atoms in red. A possible hydrogen bond to the side chain of D354 is indicated by a dotted line. The stick models of the methionine side chain introduced by the mutation and of the lysine naturally found at this position in DdDHP are depicted with carbon atoms in yellow and magenta, respectively. The mutation site, interacting residues and the nearby side chain of W360 are labelled. (For interpretation of the references in colour in this figure legend, the reader is referred to the web version of this article.)
hence lead to structural instability of the enzyme in this region. In addition, the W360R mutation might also have more global effects on protein folding. For instance, the necessity to orient the charged side chain away from the hydrophobic surrounding towards the protein surface could cause a shift of the helix it belongs to. At least in Dd DHP, this helix flanks a loop interacting with the C-terminal tail of another subunit (Fig. 2b) which in turn, contributes significantly to the formation of monomer– monomer surfaces. Because the C-terminal regions are generally shorter in hydantoinases, differ in length between DHPs and are not structurally equivalent in SkDHP, Dd DHP and CRMP1, it is not possible to predict the interactions with the C-terminal tail in the human enzyme. However, if the C-terminal tail in human DHP has a similar structure as in DdDHP, the W360R mutation could potentially impair the assembly of the tetramer. The analysis of the effect of the R412M exchange is much more hypothetical due to the lower degree of structural and sequence conservation of the residue itself and its environment. Only in five homologous proteins of known structure the arginine is conserved or conservatively exchanged to lysine (Fig. 3). In four of them it interacts with acidic residues (E352, D353 in DdDHP) and other polar amino acids nearby (Fig. 4), while in SkDHP the insertion of eight additional amino acids right in front of K467 causes its re-orientation towards the bulk solvent. In all remaining homologues, neither the arginine/lysine nor the acidic side chains and other interaction partners are conserved (Fig. 3). Thus, we can speculate that the primary roles of R412 are compensation for the negative charge of the acidic residues and formation of a hydrogen
bonding network which e.g. stabilizes the N-terminus of the helix carrying W360, the mutation site discussed above. The exchange of R412 to a hydrophobic and uncharged methionine residue eliminates a crucial component of the hydrogen bonding network and is likely to cause a destabilization of the position of the helix. R412M may, therefore, have similar global effects as the W360R point mutation. Discussion In this study, we describe the clinical, biochemical and genetic findings of two newly identified patients with a complete DHP deficiency, an inborn error of the pyrimidine degradation pathway. DHP deficiency is an autosomal recessive disease characterised by dihydrothymine-dihydrouraciluria in homozygous deficient patients. To date, only nine individuals have been described with a DHP deficiency [4,7–13]. Five of these nine patients have been identified during a screening programme for inborn errors of pyrimidine degradation in Japan and the three children and two adults were healthy at the time of diagnosis [4,11–13]. In contrast, the clinical spectrum of the other four patients included seizures, mental retardation, growth retardation and dysmorphic features [6–10]. The clinical phenotype of the two patients described in this paper included a delay in speech development and white matter abnormalities in the eldest. Interestingly, white matter abnormalities were also observed in another DHP patient and there may be an association of white-matter abnormalities and DPD deficiency [9,29]. To date, the pathological mechanism underlying the various clinical abnormalities is still not known. The pyrimidine degradation pathway plays an important role in the
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
synthesis of b-alanine and b-aminoisobutyric acid [30]. b-Alanine is a structural analogue of c-aminobutyric acid and glycine, which are the major inhibitory neurotransmitters in the central nervous system. Furthermore, b-aminoisobutyric acid has been shown to be a partial agonist of the glycine receptor. Thus, the altered homeostasis of these b-amino acids, as observed in patients with a DPD deficiency, might underlie some of the clinical abnormalities encountered in the DHP patients [30]. However, the finding that some individuals with a complete DHP deficiency did not present with any clinical abnormalities suggest that additional factors may be involved determining the clinical outcome. Thus, a DHP deficiency is probably a necessary, but not a sole, prerequisite for the onset of a clinical phenotype. Strongly elevated levels of the dihydropyrimidines were detected in urine, blood and cerebrospinal fluid. In humans, DHP is expressed mainly in liver and kidney although very low residual DHP activities have been detected in other tissues [20,31]. However, no DHP activity has been detected in brain tissues [31]. Thus, the presence of the dihydropyrimidines in cerebrospinal fluid is most likely due to transport of these metabolites across the blood-cerebrospinal fluid barrier. Analysis of DPYS for the presence of mutations showed that both patients were compound heterozygous for the missense mutation 1078T > C (W360R) in exon 6 and a novel missense mutation 1235G > T (R412M) in exon 7. The W360R mutation has been identified before in a DHP patient presenting with severe neurological abnormalities [13]. Thus far, the crystal structure of human DHP is unknown. However, the coordinates of SkDHP and DdDHP which show 28% and 58% amino acid sequence identity to human DHP, respectively, have recently been determined [18]. In spite of a few loop-forming insertions found in Sk DHP, the subunit folds of these two enzymes are nearly identical. Furthermore, they are highly resembled by the structures of hydantoinases (Hyds), bacterial counterparts of DHP whose preferred substrates are 5 0 -monosubstituted hydantoins, and of DHP-like non-catalytic proteins involved in neural development. The comparison of the available crystal structures for two DHPs [18], five hydantoinases [23–27] and one CRMP [28] enabled the identification of structurally well-conserved areas, and modelling of the human DHP structure within these areas. The DHP monomer comprises two domains, a larger domain with a (b/a)8-barrel core harbouring the catalytic di-zinc centre and a b-sandwich domain. Both are contributing to the formation of intermolecular surfaces within the physiologically relevant DHP tetramer. W360 and R412 are located on the periphery of the enzyme rather distant from the catalytic centre, excluding a direct effect of the amino acid exchanges on active site architecture and catalysis. Instead, the lack of residual activity for the point mutants is likely based on global effects of the exchanges on the protein structure. It should be noted that the patient was compound heterozygous for two mutations and the
163
combined effect of the two amino acid exchanges in a heterotetramer of DHP remains hard to predict. The diagnosis of patients with a pyrimidine degradation defect is of paramount importance, not only to avoid severe toxicity during treatment of tumour patients with the pyrimidine analogue 5FU but also to gain further insight into the relationship between the biochemical abnormalities and the onset of a clinical phenotype. Acknowledgments We thank Dr. Koichi Matsuda and Prof. Dr. Nanaya Tamaki for the dihydropyrimidinase antibody. References [1] A.B.P. van Kuilenburg, Dihydropyrimidine dehydrogenase and the efficacy and toxicity of 5-fluorouracil, Eur. J. Cancer 40 (2004) 939–950. [2] M. Tuchman, J.S. Stoeckeler, D.T. Kiang, R.F. O’Dea, M.L. Ramnaraine, B.L. Mirkin, Familial pyrimidinemia and pyrimidinuria associated with severe fluorouracil toxicity, N. Engl. J. Med. 313 (1985) 245–249. [3] R.B. Diasio, T.L. Beavers, J.T. Carpenter, Familial deficiency of dihydropyrimidine dehydrogenase. Biochemical basis for familial pyrimidinemia and severe 5-fluorouracil-induced toxicity, J. Clin. Invest. 81 (1988) 47–51. [4] S. Sumi, M. Imaeda, K. Kidouchi, S. Ohba, N. Hamajima, K. Kodama, H. Togari, Y. Wada, Population and family studies of dihydropyrimidinuria: prevalence, inheritance mode, and risk of fluorouracil toxicity, Am. J. Med. Genet. 78 (1998) 336–340. [5] A.B.P. van Kuilenburg, J.R. Meinsma, B.A. Zonnenberg, L. Zoetekouw, F. Baas, K. Matsuda, N. Tamaki, A.H. van Gennip, Dihydropyrimidinase deficiency and severe 5-fluorouracil toxicity, Clin. Cancer Res. 9 (2003) 4363–4367. [6] M. Duran, P. Rovers, P.K. de Bree, C.H. Schreuder, H. Beukenhorst, L. Dorland, R. Berger, Dihydropyrimidinuria, Lancet 336 (1990) 817–818. [7] M. Duran, P. Rovers, P.K. de Bree, C.H. Schreuder, H. Beukenhorst, L. Dorland, R. Berger, Dihydropyrimidinuria: a new inborn error of pyrimidine metabolism, J. Inherit. Metab. Dis. 14 (1991) 367–370. [8] M.J. Henderson, K. Ward, H.A. Simmonds, J.A. Duley, P.M. Davies, Dihydropyrimidinase deficiency presenting in infancy with severe developmental delay, J. Inherit. Metab. Dis. 16 (1993) 574–576. [9] C.W. Putman, J.J. Rotteveel, R.A. Wevers, A.H. van Gennip, J.A. Bakkeren, R.A. De Abreu, Dihydropyrimidinase deficiency, a progressive neurological disorder? Neuropediatrics 28 (1997) 106–110. [10] B.E. Assmann, G.F. Hoffmann, L. Wagner, C. Brautigam, H.W. Seyberth, M. Duran, A.B.P. van Kuilenburg, R.A. Wevers, A.H. van Gennip, Dihydropyrimidinase deficiency and congenital microvillous atrophy: coincidence or genetic relation? J. Inherit. Metab. Dis. 20 (1997) 681–688. [11] S. Sumi, K. Kidouchi, K. Hayashi, S. Ohba, Y. Wada, Dihydropyrimidinuria without clinical symptoms, J. Inherit. Metab. Dis. 19 (1996) 701–702. [12] S. Ohba, K. Kidouchi, S. Sumi, M. Imaeda, N. Takeda, H. Yoshizumi, A. Tatematsu, K. Kodama, K. Yamanaka, M. Kobayashi, Y. Wada, Dihydropyrimidinuria: the first case in Japan, Adv. Exp. Med. Biol. 370 (1994) 383–386. [13] N. Hamajima, M. Kouwaki, P. Vreken, K. Matsuda, S. Sumi, M. Imaeda, S. Ohba, K. Kidouchi, M. Nonaka, M. Sasaki, N. Tamaki, Y. Endo, R.A. De Abreu, J. Rotteveel, A.B.P. van Kuilenburg, A.H. van Gennip, H. Togari, Y. Wada, Dihydropyrimidinase deficiency: structural organization, chromosomal localization, and mutation analysis of the human dihydropyrimidinase gene, Am. J. Hum. Genet. 63 (1998) 717–726.
164
A.B.P. van Kuilenburg et al. / Molecular Genetics and Metabolism 91 (2007) 157–164
[14] A.H. van Gennip, R.A. De Abreu, H. van Lenthe, J. Bakkeren, J. Rotteveel, P. Vreken, A.B.P. van Kuilenburg, Dihydropyrimidinase deficiency: confirmation of the enzyme defect in dihydropyrimidinuria, J. Inherit. Metab. Dis. 20 (1997) 339–342. [15] M. Kikugawa, M. Kaneko, S. Fujimoto-Sakata, M. Maeda, K. Kawasaki, T. Takagi, N. Tamaki, Purification, characterization and inhibition of dihydropyrimidinase from rat liver, Eur. J. Biochem. 219 (1994) 393–399. [16] J. Kautz, K.D. Schnackerz, Purification and properties of 5,6dihydropyrimidine amidohydrolase from calf liver, Eur. J. Biochem. 181 (1989) 431–435. [17] K. Jahnke, B. Podschun, K.D. Schnackerz, J. Kautz, P.F. Cook, Acid–base catalytic mechanism of dihydropyrimidinase from pH studies, Biochemistry 32 (1993) 5160–5166. [18] B. Lohkamp, B. Andersen, J. Pisˇkur, D. Dobritzsch, The crystal structures of dihydropyrimidinases reaffirm the close relationship between cyclic amidohydrolases and explain their substrate specificity, J. Biol. Chem. 281 (2006) 13762–13776. [19] Z. Gojkovic, L. Rislund, B. Andersen, M.P. Sandrini, P.F. Cook, K.D. Schnackerz, J. Pisˇkur, Dihydropyrimidine amidohydrolases and dihydroorotases share the same origin and several enzymatic properties, Nucleic Acids Res. 31 (2003) 1683–1692. [20] N. Hamajima, K. Matsuda, S. Sakata, N. Tamaki, M. Sasaki, M. Nonaka, A novel gene family defined by human dihydropyrimidinase and three related proteins with differential tissue distribution, Gene 180 (1996) 157–163. [21] H. van Lenthe, A.B.P. van Kuilenburg, T. Ito, A.H. Bootsma, A.G. van Cruchten, Y. Wada, A.H. van Gennip, Defects in pyrimidine degradation identified by HPLC-electrospray tandem mass spectrometry of urine specimens or urine-soaked filter paper strips, Clin. Chem. 46 (2000) 1916–1922. [22] A.B.P. van Kuilenburg, H. van Lenthe, A.H. van Gennip, Radiochemical assay for determination of dihydropyrimidinase activity using reversed-phase high-performance liquid chromatography, J. Chromatogr. B Biomed. Sci. Appl. 729 (1999) 307–314.
[23] K.V. Radha Kishan, R.M. Vohra, K. Ganesan, V. Agrawal, V.M. Sharma, R. Sharma, Molecular structure of D-hydantoinase from Bacillus sp. AR9: evidence for mercury inhibition, J. Mol. Biol. 347 (2005) 95–105. [24] J. Abendroth, K. Niefind, D. Schomburg, X-ray structure of a ˚ resolution, J. Mol. dihydropyrimidinase from Thermus sp. at 1.3 A Biol. 320 (2002) 143–156. [25] Y.H. Cheon, H.S. Kim, K.H. Han, J. Abendroth, K. Niefind, D. Schomburg, J. Wang, Y. Kim, Crystal structure of D -hydantoinase from Bacillus stearothermophilus: insight into the stereochemistry of enantioselectivity, Biochemistry 41 (2002) 9410–9417. [26] J. Abendroth, K. Niefind, O. May, M. Siemann, C. Syldatk, D. Schomburg, The structure of L-hydantoinase from Arthobacter aurescens leads to an understanding of dihydropyrimidinase substrate and enantio specificity, Biochemistry 41 (2002) 8589–8597. [27] Z. Xu, Y. Liu, Y. Yang, W. Jiang, E. Arnold, J. Ding, Crystal structure of D-hydantoinase from Burkholderia pickettii at a resolution of 2.7 Angstroms: insights into the molecular basis of enzyme thermostability, J. Bacteriol. 185 (2003) 4038–4049. [28] R.C. Deo, E.F. Schmidt, A. Elhabazi, H. Togashi, S.K. Burley, S.M. Strittmatter, Structural bases for CRMP function in plexin-dependent semaphorin3A signaling, EMBO J. 23 (2004) 9–22. [29] G.M. Enns, A.J. Barkovich, A.B.P. van Kuilenburg, M. Manning, T. Sanger, D.R. Witt, A.H. van Gennip, Head imaging abnormalities in dihydropyrimidine dehydrogenase deficiency, J. Inherit. Metab. Dis. 27 (2004) 513–522. [30] A.B.P. van Kuilenburg, A.E.M. Stroomer, H. van Lenthe, N.G.G.M. Abeling, A.H. van Gennip, New insights in dihydropyrimidine dehydrogenase deficiency: a pivotal role for beta-aminoisobutyric acid? Biochem. J. 379 (2004) 119–124. [31] A.B.P. Van Kuilenburg, H. van Lenthe, A.H. van Gennip, Activity of pyrimidine degradation enzymes in normal tissues, Nucleosides Nucleotides Nucleic Acids 25 (2006) 1211–1214.