The 3-ureidopropionase of Caenorhabditis elegans , an enzyme involved in pyrimidine degradation Tim Janowitz, Irene Ajonina, Markus Perbandt, Christian Woltersdorf, Patrick Hertel, Eva Liebau and Ulrike Gigengack Institut fu ¨ r Zoophysiologie, Westfa ¨ lische Wilhelms-Universita ¨ t, Mu ¨ nster, Germany Introduction Pyrimidine nucleotides, besides being constituents of nucleic acids, fulfil diverse important functions in the cell. Their cellular concentration is controlled by de novo synthesis, salvage of preformed molecules and degradation. The most common route to pyrimidine degradation is via the reductive pathway [1]. In addi- tion to this pathway, other routes to pyrimidine degra- dation exist [2,3]. The reductive route to pyrimidine degradation consists of three enzymatic steps. First, the pyrimidine molecule is reduced in a NADPH-dependent Keywords b-alanine; biochemical characterization; GFP fusion protein; nucleotides Correspondence T. Janowitz, Institut fu ¨ r Zoophysiologie, Westfa ¨ lische Wilhelms-Universita ¨ t, Hindenburgplatz 55, D-48143 Mu ¨ nster, Germany Fax: +49 251 8321766 Tel: +49 251 8321710 E-mail: tim.janowitz@rub.de Note To prevent confusion and ambiguity, in the present study, we use the terms ‘3-ureido- propionase’ instead of b-alanine synthase and ‘ureido’ to refer to a carbamoylamino- group. Other common nomenclatures are given in parenthesis where appropriate (Received 19 February 2010, revised 23 July 2010, accepted 3 August 2010) doi:10.1111/j.1742-4658.2010.07805.x Pyrimidines are important metabolites in all cells. Levels of cellular pyrimi- dines are controlled by multiple mechanisms, with one of these comprising the reductive degradation pathway. In the model invertebrate Caenorhabditis elegans, two of the three enzymes of reductive pyrimidine degradation have previously been characterized. The enzyme catalysing the final step of pyrimidine breakdown, 3-ureidopropionase (b-alanine synthase), had only been identified based on homology. We therefore cloned and functionally expressed the 3-ureidopropionase of C. elegans as hexahistidine fusion protein. The purified recombinant enzyme readily converted the two pyrimidine degradation products: 3-ureidopropionate and 2-methyl-3-urei- dopropionate. The enzyme showed a broad pH optimum between pH 7.0 and 8.0. Activity was highest at approximately 40 °C, although the half-life of activity was only 65 s at that temperature. The enzyme showed clear Michaelis–Menten kinetics, with a K m of 147 ± 26 lm and a V max of 1.1 ± 0.1 UÆmg protein )1 . The quaternary structure of the recombinant enzyme was shown to correspond to a dodecamer by ‘blue native’ gel elec- trophoresis and gel filtration. The organ specific and subcellular localiza- tion of the enzyme was determined using a translational fusion to green fluorescent protein and high expression was observed in striated muscle cells. With the characterization of the 3-ureidopropionase, the reductive pyrimidine degradation pathway in C. elegans has been functionally char- acterized. Structured digital abstract l MINT-7986015: 3-ureidopropionase (uniprotkb:Q19437) and 3-ureidopropionase (uniprotkb: Q19437) bind (MI:0407)byblue native page (MI:0276) Abbreviations GFP, green fluorescent protein; RNAi, RNA interference. 4100 FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS manner by dihydropyrimidine dehydrogenase (EC 1.3. 1.2) with a subsequent ring opening by dihydro- pyrimidinase (EC 3.5.2.2). In the last step, the formed ureido compound is hydrolyzed by 3-ureidopropionase (b-alanine synthase, N-carbamoyl-b-alanine amidohy- drolase; EC 3.5.1.6) to carbon dioxide, ammonia and a b-amino acid (Fig. S1). Almost all known 3-ureidopro- pionases, excluding yeast 3-ureidopropionase, belong to branch 5 of the so-called nitrilase superfamily of enzymes. Members of this superfamily all possess a conserved cysteine residue that is essential for enzy- matic activity [4–6]. Therefore, the addition of low amounts of reducing agents such as dithiothreitol have been reported to result in increased activity, presumably as a result of stabilization of the reduced state of the catalytic residue [7–9]. For 3-ureidopropionases purified from rat and maize, a dependence of enzymatic activity on Zn 2+ -ions has been reported [9,10]. The enzymes involved in pyrimidine catabolism uti- lize both uracil and thymine as substrates. Cytosine is not directly accepted as a substrate and must first be deaminated to uracil. The b-amino acid resulting from degradation can be channelled into energy metabolism via a semi-aldehyde intermediate [11]. A fraction of the resulting b-amino acid can also fulfil other functions in the cell. The degradation product of uracil, 3-amino- propionate (b-alanine), for example, can be condensed with histidine to form the dipeptide carnosine. Carno- sine and other similar dipeptides containing nonprotein- ogenic amino acids can be found in excitable tissues, brain and skeletal muscles. The physiological role of such dipeptides is not yet understood [12]. Defects in pyrimidine degradation are known to be the cause of several human disorders [13–15]. The clinical symp- toms of patients suffering form such disorders are very diverse, ranging from asymptomatic to severe symp- toms, with mental retardation and convulsive attacks being the most common. Even a normally asymptom- atic partial deficiency of dihydropyrimidine dehydro- genase can cause severe complications for patients receiving chemotherapy with pyrimidine analogs such as fluorouracil, as a result of a diminishing of the nor- mally high turnover rates and subsequent overdosing [16,17]. In the genome of the nematode Caenorhabditis ele- gans, only the reductive pathway (and none of the alternative routes) is present. The first two enzymes of reductive pyrimidine degradation in C. elegans, dihy- dropyrimidine dehydrogenase [18] and dihydropyrimi- dinase [19], have already been characterized by genetic and ⁄ or molecular methods. The enzyme catalyzing the last step, 3-ureidopropionase, has so far only been identified based on homology to 3-ureidopropionases of other organisms [4]. Because such predictions can be misleading [8], it is important to verify them at a functional level. Accordingly, we cloned the cDNA coding for the predicted 3-ureidopropionase of C. ele- gans and functionally expressed it as a hexahistidine fusion protein in Escherichia coli. The purified recom- binant protein was then functionally characterized in vitro. To analyze the expression pattern of the 3-ure- idopropionase, transgenic C. elegans were created expressing a green fluorescent protein (GFP) fusion protein under the control of the 3-ureidopropionase promoter. Results and Discussion Cloning and expression of recombinant 3-ureidopropionase The sole homolog of 3-ureidopropionase in C. elegans is encoded by the gene F13H8.7. The protein encoded by this gene clusters together with other 3-ureidopropi- onases of this family during phylogenetic analysis (Fig. 1). The 3-ureidopropionases used for phyloge- netic inferrence, similar to almost all 3-ureidopropion- ases identified so far, belong to the nitrilase superfamily of enzymes [4]. An exception is the enzyme from Saccharomyces (Lachancea) kluyveri. This enzyme appears to be phylogenetically unrelated to 3-ureido- propionases of other eukaryotes because it shows high structural similarity with dizinc-dependent exopepti- dases [20,21]. It has been proposed that this enzyme is prototypic of fungal 3-ureidopropionases. A homology model (Fig. S2) of the C. elegans enzyme based on the closely-related 3-ureidopropionase of Drosophila mela- nogaster (Dme3-UP; Fig. 1) was constructed. The cata- lytic triade Glu-Lys-Cys typical for enzymes of the nitrilase superfamily [4] could be observed in the model. For functional characterization, the cDNA of gene F13H8.7 was expressed as hexahistidine fusion protein in E. coli. The purified enzyme liberated ammonia upon incubation with 3-ureidopropionate (not shown). To verify that 3-aminopropionate (b-ala- nine) was produced, a sample from an activity assay was analyzed by MS. We were able to identify two substances with m ⁄ z ratios corresponding to 3-amino- propionate and 3-ureidopropionate. Furthermore, the MS ⁄ MS spectra of both ions corresponded to the MS ⁄ MS spectra of genuine 3-aminopropionate and 3-ureidopropionate respectively (data not shown). We therefore conclude that the recombinant protein shows a genuine 3-ureidopropinase activity, confirming that the gene product of F13H8.7 is the C. elegans 3-urei- dopropionase. T. Janowitz et al. 3-Ureidopropionase of C. elegans FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 4101 Biochemical characterization of recombinant 3-ureidopropionase On the basis of bioinformatics and the homology model (Fig. S2), a catalytically relevant cysteine was shown to be present in the C. elegans 3-ureidopropion- ase. To prevent oxidation of this residue common to all members of the nitrilase superfamily, low millimo- lar amounts of dithiothreitol were added to the activity assays. The activity of the enzyme was higher with 0.25 mm dithiothreitol added than without dithiothrei- tol or with higher concentrations. The enzyme did not show any dependence of activity on Zn 2+ -ions, which had been reported for 3-ureidopropionases from other species [9,10]. Incubation with either 1 mm EDTA or 1mm ZnCl 2 did not result in any change in activity compared to control reactions without additive. A reaction mechanism independent of divalent cations has been theoretically deduced for carbamylases, repre- senting another branch of ureido-group hydrolyzing enzymes [22]. Given that 3-ureidopropionases are clo- sely related to carbamylases (Fig. 1) and a recombi- nant 3-ureidopropionase from D. melanogaster also does not show any effect of Zn 2+ ions on enzymatic activity [23], the dependence of 3-ureidopropionase activity on Zn 2+ ions appears to be a peculiarity of some species. To test the substrate specificity of the recombinant 3-ureidopopionase, several ureido-com- pounds with different side chain architectures and also compounds with functional groups similar to the ureido- group (e.g. guanidino-group) were tested in activity assays. Besides uracil-derived 3-ureidopropionate, the recombinant enzyme also accepted thymine-derived 2-methyl-3-ureidopropionate (relative activity com- pared to 3-ureidopropionate: 82 ± 18%). 4-Amino-4- oxo-butanoic acid, which also appeared to show some conversion, proved not to be a reliable substrate; therefore, this result is not discussed further. Other substances tested were no substrate of the C. elegans 3-ureidopropionase (Table 1). The recombinant C. ele- gans enzyme did not show any activity with ureido- acetic acid and 2-ureidopropionic acid, as reported previously for the enzyme purified from rat liver [24]. Because the recombinant enzyme showed highest activ- ity with 3-ureidopropionate, this compound was used in all further activity measurements. The pH optimum of the enzymatic activity was determined to be quite broad, with an pH optimum between pH 7.0 and 8.0. The enzyme showed maximal activity at approximately 40 °C(Fig. 2). However, at this temperature, the C. elegans enzyme proved to be unstable. After preincubation at 40 °C and activity measurement at 30 °C (where activity was stable for ‡ 2 h), the half-life of enzymatic activity was only 65 s. The activity of the enzyme remained stable at a specific activity of 0.2 UÆmg protein )1 after 5 min of preincu- bation. The dependence of the activity on substrate concentration showed clear Michaelis–Menten kinetics. Nonlinear regression of the experimental data to the Michaelis–Menten equation yielded a K m of 147 ± 26 lm and a V max of 1.1 ± 0.1 UÆmg protein )1 (Fig. 3). In rat and humans, 3-ureidopropionase has Branch 3 Branch 5 Branch 11 Branch 6 Branch 1 Branch 2 Branch 10 Branch 8 Branch 7 Branch 9 Branch 4 Bootstrap support 0–50 50–90 90–100 0.1 changes/site Fig. 1. Phylogenetic tree of the nitrilase superfamily. For tree construction, protein sequences of known members of the nitrilase superfamily were aligned using CLUSTALX. For phylogenetic inference, the PHYML maximum likelihood method was used with 100 bootstrap trials. Definitions of the different branches and accession num- bers of proteins used are given in Table S1. Saccharomyces kluyveri 3-ureidopropionase has been excluded from the analysis because it belongs to a different phylo- genetic group. 3-Ureidopropionase of C. elegans T. Janowitz et al. 4102 FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS been reported to show positive co-operativity with 3-ureidopropionate as substrate [25,26]. Such kinetic behaviour is best described by the Hill equation. A nonlinear regression of our data to the Hill equation showed that the data fitted best for n  1 (when the Hill equation transforms into the Michaelis–Menten equation) (not shown). Other 3-ureidopropionases characterized so far from plants and animals all show a lower K m value (Table 2) than the C. elegans enzyme. Only the enzyme from S. kluyveri, which belongs to a different phylogenetic group, has an approximately 500-fold higher K m . Differences in the reaction mechanism as a result of the large phyloge- netic distance might be responsible for such a discrep- ancy. In plants, the 3-ureidopropionase is involved in the synthesis of the pantothenate moity of coenzyme A and 3-aminopropionate can serve as osmoprotectant [9]. Those different roles in metabolism might be responsible for the observed differences in K m values of 3-ureidopropionase of C. elegans and plants. It is quite unexpected that the K m of the D. melanogaster enzyme is approximately six-fold lower because both enzymes show a high degree of sequence identity. Fur- thermore, the homology model of the C. elegans enzyme corresponds well with the D. melanogaster template. To explain the difference in K m values, it might be valuable to solve the actual structure of the C. elegans enzyme. Determination of native protein mass To determine the native molecular mass of the recom- binant C. elegans 3-ureidopropionase, gel filtration Table 1. Substrate specificity of recombinant 3-ureidopropionase from Caenorhabditis elegans. Enzymatic activity was measured by quanti- fying the amount of ammonia released during reactions. Substrates were provided at a concentration of 3–10 m M. 100% activity equals 0.88 ± 0.10 UÆmg protein )1 . ND, not detectable. Substrate Activity (%) Linear formula 3-Ureidopropionic acid (N-carbamyl-b-alanine) 100 COOH(CH 2 ) 2 NHCONH 2 2-Methyl-3-ureidopropionic acid (N-Carbamyl-b-aminoisobutyric acid) 82 ± 18 a COOH(CHCH 3 )CH 2 NHCONH 2 4-Amino-4-oxo-butyric acid b (succinamic acid) 3 ± 1 COOH(CH 2 ) 2 CONH 2 2-Ureidoacetic acid (N-carbamylglycine) ND COOHCH 2 NHCONH 2 4-Ureidobutanoic acid (N-carbamyl-c-aminobutyric acid) ND a COOH(CH 2 ) 3 NHCONH 2 2-Ureidopropionic acid (N-carbamy-a-alanine) ND a COOH(CHNHCONH 2 )CH 3 2-Ureidobutanedioic acid (N-carbamylaspartic acid) ND COOH(CHNHCONH 2 )CH 2 COOH 3-Guanidinopropionic acid ND COOH(CH 2 ) 2 NHCNHNH 2 (S)-2-Amino-5-ureidopentanoic acid (L-citrulline) ND COOH(CHNH 2 )(CH 2 ) 3 NHCONH 2 1-Ureido-4-aminobutane (N-carbamylputrescine) ND NH 2 (CH 2 ) 4 NHCONH 2 N-Allylurea b ND CH 2 CHCH 2 NHCONH 2 a The substance was synthesized and crude synthesis solution used for the experiments. b The substance was not stable under the experi- mental conditions and influenced the ammonia determination. –0.2 0 –5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Temperature (°C) Specific activity (U·mg protein –1 ) 0.2 0.4 0.6 0.8 1 1.2 Fig. 2. Temperature dependence of the 3-ureidopropionase reac- tion. The enzyme showed maximal activity at approximately 40 °C. 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 500 1000 1500 2000 Substrate concentration (µ M) Specific activity (U·mg protein –1 ) 2500 3000 3500 Fig. 3. Dependence of specific activity on substrate concentration. Activity was measured with the indophenol blue method using 3- ureidopropionate as substrate. Data represent the means ± SD of n ‡ 3 experiments. Data points were used for nonlinear regression to the Michaelis–Menten equation, resulting in the solid curve shown. T. Janowitz et al. 3-Ureidopropionase of C. elegans FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 4103 separation and ‘blue native’ PAGE was used. Here, the apparent molecular mass was 472 ± 4 kDa, which, taking a monomer mass of 43.2 kDa into account, points to a dodecamer (Fig. 4). To support this find- ing, the oligomeric state of the enzyme, the recombi- nant protein was subjected to gel filtration using a calibrated Superdex S-200 column, followed by immunodetection of the His-tagged protein. The protein eluted as a single peak corresponding to a molecular mass of approximately 500 kDa (data not shown). Because temperatures of 40 °C had an influ- ence on activity (vide supra), we also preincubated samples at 40 °C before electrophoresis. The resulting protein pattern showed an additional signal at 404 ± 5 kDa (Fig. 4). Taking into consideration the decreased activity of enzyme that was preincubated at 40 °C, it can be speculated that the 404 kDa oligomer represents an inactive enzyme species diminished of its overall activity. 3-Ureidopropionases purified from rat also have been reported to have different enzymatic activities, depending on their oligomerization states. However, in these cases, the changes in oligomerization state were ligand-dependent [26]. We could not observe a similar behaviour for the C. elegans enzyme (data not shown). The recombinant D. melanogaster enzyme also does not to show ligand-dependent oligomeriza- tion [23]. Conclusive data on whether this is of any significance for the regulation of 3-ureidopropionase in different phylogenetic groups is presently unavailable. Characterization of 3-ureidopropionase in vivo The physiological relevance of 3-ureidopropionase for C. elegans is only poorly characterized. In one large- scale RNA interference (RNAi) experiment, maternal sterility has been reported [27], although this effect was not found in two other experiments [28,29]. When exposing worms to RNAi, we were also unable to observe any phenotypic abnormalities compared to wild-type animals. As described, low doses of 5-fluoro- uracil (5-FU) completely block F1 embryo hatching in wild-type worms [30]. To determine whether 3-ureido- propionase deficiency has any effect on worms exposed to 5-FU, RNAi- and wild-type worms were incubated with low doses of 5-FU. No significant difference was observed in the number of hatched F 1 embryos (Fig. S3). This is also observed in the D. melanogaster pyd mutants (loss-of function mutation of 3-ureidopro- pionase) that were fed dietary 5-FU. Whereas su(r) mutants (loss-of-function mutation of dihydropyrimi- dine dehydrogenase) are hypersensitive to 5-FU as a result of its accumulation, this is reduced in CRMP mutants (loss-of-function mutation dihydropyrimidinase) Table 2. Comparison of kinetic parameters of eucaryotic 3-ureidopropionases. V max and either K m or K ½ are shown depending on the cata- lytic mechanism. Only data reporting 3-ureidopropionate as substrate are included. Amino acid identities (aa identity) are given with refer- ence to the sequence of C. elegans 3-ureidopropionase. h, Hill coefficent; ND, not determined. Organism V max (UÆmg protein )1 ) K m (lM) K ½ (lM) h Amino acid identity (%) Purification Reference C. elegans 1.1 147 – – 100 Recombinant Present study Drosophila melanogaster 0.46 24.5 – – 66 Recombinant [21] Zea mays ND 11 – – 63 Partial [9] Arabodpsis thaliana 0.47 6 – – 64 Recombinant [21] Rat ND – 170 2.0 64 1045-fold [24] Rat a ND 8 ND 1.9 64 1096-fold [26] Homo sapiens ND – 100 2.0 65 Recombinant [25] Euglena gracilis ND 38 – – – b 95.5-fold [45] Saccharomyces kluyveri c 7.3 70900 – – 7 Recombinant [21] a Co-operativity only below 12 nM of substrate. b No sequence information is available. c Belongs to a different phylogenetic group. NH 480 720 242 M kDa Fig. 4. ‘Blue native’ gel electrophoresis of recombinant protein. In lane N, 2 lg of native protein were loaded onto the gel. In lane H, 5 lg of protein were incubated for 5 min at 40 °C before loading. Separation was performed on a 4–14% (w ⁄ v) polyacrylamide gradi- ent. Lane M shows the native protein standard. 3-Ureidopropionase of C. elegans T. Janowitz et al. 4104 FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS making them less sensitive to 5-FU. Because little or no 5-FU accumulates in the pyd mutants, they exibit essentially wild-type sensitivity towards the drug [31]. Transgenic worms were created that expressed a 3-ureidopropionase-GFP fusion protein under the con- trol of the 3-ureidopropionase promoter. GFP signals were detected during all stages of C. elegans develop- ment (larvae L1–4 and adult hermaphrodites). Strong GFP expression was observed in dense bodies of stri- ated body wall muscle cells (Fig. 5). C. elegans has striated and nonstriated muscles. The nonstriated mus- cles are the pharyngeal, intestinal, uterine, vulval and anal muscles, whereas the body wall muscles are stri- ated. In C. elegans, sarcomere attachment to the mus- cle membrane and the underlying basement membrane is performed by the dense body. This protein complex shares functional similarity with both the vertebrate Z-disk and the costamere. In addition to its structural role, the dense body also performs a signalling func- tion in muscle cells and communication between dense bodies and nuclei has been postulated [32]. However, colocalization experiments using 4¢,6¢-diamidino-2- phenylindole- and Hoechst staining indicated that 3-ureidopropionase is not localized in the nuclei of muscle cells (data not shown). In animals, catabolic processes are often associated with ‘liver-like’ organs. In humans, 3-ureidopropionase is expressed in the liver. In rat, it has been purified and cloned from liver tissue [10,33]. In C. elegans, the tissue that performs ‘liver-like’ function is the intestine. Therefore, the absence of 3-ureidopropionase expression is unex- pected. Presuming that there really is no expression of 3-ureidopropionase in the C. elegans intestine, this might indicate that its primary function does not lie in the degradation of pyrimidines. The localization in striated muscle cells might rather point to a role in providing b-amino acids for synthesis of dipeptides such as carnosine. Reporter gene analyses of dihydro- pyrimidinase in C. elegans also demonstrated expres- sion in body wall muscle cells [19]. Currently, the localization of dihydropyrimidine dehydrogenase has not been investigated. Therefore, at least two of the three enzymes of reductive pyrimidine degradation are known to occur in striated body wall muscle cells. Further studies will continue to investigate the func- tional role of reductive pyrimidine degradation and synthesis of carnosin-like dipeptides in striated muscle cells of C. elegans. Experimental procedures Organisms and growth conditions Caenorhabditis elegans strain N2 (Bristol variety) was used in the present study. Unless noted otherwise, nematodes were grown at 20 °C on nematode growth medium with E. coli OP50 provided as a food source ad libitum [34]. E. coli was cul- tivated in appropriate media under selection pressure at 37 °C. General molecular biological procedures Total RNA was prepared using TRIreagent (Segenetic, Borken, Germany) and checked for integrity by denaturing agarose gel electrophoresis. The first strand cDNA synthesis kit (Fermentas, St. Leon-Rot, Germany) was used for subsequent cDNA synthesis. Dephosphorylation of vector DNA was performed with calf intestine alkaline phospha- A B C Fig. 5. Analysis of the expression pattern of 3-ureidopropionase in Caenorhabditis elegans. GFP images of adult C. elegans carrying 3-ureidopropionase::GFP are shown. (A) Confocal differential interference contrast micrograph, (B) combined confocal differential interference contrast and fluorescence micrograph (C) fluorescence micrograph. Scale bar = 10 lm. T. Janowitz et al. 3-Ureidopropionase of C. elegans FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 4105 tase (Promega, Mannheim, Germany) in accordance with the manufacturer’s instructions. Cloning of 3-ureidopropionase The cDNA of 3-ureidopropionase (gene F13H8.7) was amplified from a cDNA library made of 1 lg of total RNA using primers 3-UP_F (5¢-CATATGTCTGCAGCTCCG GCT-3¢) and 3-UP_R (5¢-CTCGAGTTGCTCTCTTCT GATGTCTG-3¢). For amplification of promoter fragments, genomic DNA was used as a template with primers Prom.UP_F (5¢-AAGCTTAAGTCAATGTGGGCAAG-3¢) and Prom.UP_R (5¢-CATATGTTTTACCTGAATAAGAT- A-3¢). Primers used in the PCRs introduced restriction sites that were used for subsequent cloning. PCR-fragments were cloned into the pJET2.1 ⁄ blunt vector (Fermentas). Errors introduced during PCR were excluded by sequencing. For recombinant expression, cDNA was introduced NdeI ⁄ XhoI into pET22b(+) (excising the pelB leader sequence). To make translational GFP fusions, the 3-ureidopropionase cDNA was inserted NdeI ⁄ SalI into a pJET plasmid contain- ing a promoter fragment. The whole construct was excised with XhoI and then transferred into the SalI site of dephos- phorylated pPD95.77 (Addgene plasmid 1495). Correct ori- entation of the insert was checked by restriction analysis. Purification of heterologously expressed proteins Bacterial expression of recombinant protein was performed in E. coli BL21-CodonPlus Ò (DE3)-RIL cells (Stratagene, La Jolla, CA, USA). Expression cultures of 300 mL 2YT medium (16 gÆL )1 tryptone, 10 gÆL )1 yeast extract, 5gÆL )1 NaCl, pH 7.0–7.5) supplemented with ampicillin were inoculated at a dilution of 1 : 10 from a saturated over- night culture. After 1 h of incubation at 25 °C and agitation by rotary shaking at 170 r.p.m., expression was induced by addition of 1 mm isopropyl thio-b-d-galactoside. Cultures were subsequently incubated for 6 h at 25 °C with agitation by rotary shaking at 170 r.p.m. Cells were har- vested by centrifugation and resuspended into lysis buffer (50 mm NaH 2 PO 4 , pH 8.0, 300 mm NaCl) supplemented with 5 mm 2-mercaptoethanol, 0.6 gÆL )1 lysozyme. Cell lysis was performed by sonification (five bursts of 1 min) after 30 min of incubation on ice. Cell debris was removed by centrifugation. The crude protein supernatant was passed over a Ni 2+ -NTA-agarose column (Qiagen, Hilden, Ger- many; 2 mL matrix) equilibrated in lysis buffer with 10 mm imidazole. The column was washed with approximately ten volumes of lysis buffer with 30 mm imidazole. Bound pro- teins were eluted with 2.5 mL of lysis buffer with 250 mm imidazole and desalted over a PD10 column (GE Health- care, Mu ¨ nchen, Germany). Desalted protein was stored in 50 mm K-phosphate (pH 8.0), 1 mm dithiothreitol at )80 °C. The purity of the recombinat protein was ‡ 85% as judged by SDS ⁄ PAGE. Molecular size of the 3-ureidopropionase The molecular size and oligomeric state of the 3-ureidopro- pionase was assessed by subjecting the affinity-purified pro- tein to FPLC on a Superdex S-200 column (Amersham Biosciences, Piscataway, NJ, USA). The Superdex S-200 column, equilibrated with 100 mm sodium phosphate (pH 7.0) was calibrated using a gelfiltration standard (151–1901; Bio-Rad, Munich, Germany) containing thyroglobulin, a-globulin, ovalbumin, myoglobin and vitamin B 12 . ‘Blue native’ electrophoresis was performed as described previ- ously [35], using a 4–14% (w ⁄ v) polyacrylamide gradient and separating 2–5 lg of protein. Determination of enzymatic activity A standard reaction for determination of enzymatic activities was performed with 3.5 lgÆmL )1 protein in 100 mm K-phos- phate (pH 7.5), 0.25 mm dithiothreitol at 3 °C and substrates provided at a final concentration of 3–10 mm. Enzymatic activity was determined by measuring ammonia liberated during reactions with the indophenol blue method [36,37]. In brief, 100 lL of sample were mixed with 100 lL each of 0.33 m sodium phenolate, 0.02 m sodium hypochlorite and 0.01% (w ⁄ v) sodium pentacyanonitrosylferrate. After incubation at approximately 95 ° C for 2 min, samples were diluted with 600 lL of water and A 640 was measured. Ammonia concentrations were assessed using standard curves contructed with ammonium chloride. Background values of samples with inactive protein were subtracted. Chemical synthesis of ureido compounds Chemical synthesis of ureido compounds as substrates for activity assays was carried out as described previously [38]. Yield of synthesis was determined by TLC on a SIL ⁄ G matrix developed in chloroform ⁄ methanol ⁄ formic acid (65 : 18 : 1) [39]. Amines were stained by spraying with nin- hydrin [0.2% (w ⁄ v) in ethanol] and subsequent incubation at 80 °C. Ureido compounds were visualized by spraying with 4-(dimethylamino)-benzaldehyde [1% (w ⁄ v) in hydrochloric acid ⁄ methanol (1 : 1)]. Synthesis yield as judged by TLC was ‡ 80%. Synthesized substrates (2-methyl-3-ureidopropionic acid, 4-ureidobutyric acid and 2-ureidopropionic acid) were directly used for activity measurements without further puri- fication. Synthesized 3-ureidopropionic acid showed almost identical behaviour in activity measurements as the commer- cially available (Sigma, Steinheim, Germany) compound. MS determination of reaction products A standard reaction was performed as described above, except that 5 lgÆmL )1 enzyme were used in 5 m m K-phos- phate (pH 8.0). Samples were prepared for MS as described 3-Ureidopropionase of C. elegans T. Janowitz et al. 4106 FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS previously [8], with the modification that dry samples were resuspended in 1% (v ⁄ v) formic acid in methanol ⁄ water (1 : 1). MS was performed by Simone Ko ¨ nig of the core unit ‘Integrated Functional Genomics’ of the Interdisciplinary Center for Clinical Research Mu ¨ nster (Germany). Manual nanospray MS ⁄ MS using a modified stage [40] was carried out with Q-TOF Premier (Waters Corp., Manchester, UK). RNAi For RNAi experiments, double-stranded RNA was pro- duced in the E. coli strain HT115 transformed with the L4440 feeding vector pPD129.36 (L4440) that contained a cDNA fragment of the 3-ureidopropionase. Isopropyl thio- b-d-galactoside (1 mm) was added to induce transcription of the double-stranded RNA. To determine whether 3-urei- dopropionase deficiency has any effect on worms exposed to 5-FU, RNAi worms, as well as worms feeding on E. coli harbouring the L4440 vector only, were incubated with low doses of 5-FU. Because concentrations above 0.2 lgÆmL )1 resulted in a severe egg laying defect, 0.1 and 0.05 lgÆmL )1 were chosen. Progenies of 18 worms (3 · 6 plates) were counted and the experiments were performed twice. Transformation of worms using microinjection and fluorescence microscopy of GFP fusion proteins Transgenic C. elegans germline transformation was per- formed by coinjecting the vector construct 3-ureidopropion- ase::GFP with the pRF4 plasmid encoding the dominant marker gene rol-6 into the germline of young adults (Fig. 5A). To investigate the cell-specific, developmentally regulated transcription of 3-ureidopropionase, GFP expres- sion patterns were analyzed by fluorescence microscopy. Images were captured with a Zeiss axiovert 100 microscope (Carl Zeiss, Oberkochen, Germany) equipped with fluores- cein isothiocyanate ⁄ GFP filters. Hoechst-staining of nuclei was performed as described previously [41]. Construction of a phylogenetic tree and structural model Full length protein sequences were extracted from the NCBI protein database (http://www.ncbi.nlm.nih.gov). Sequences were trimmed to the same length and aligned using clustalx [42]. Alignment parameters were optimized until at least the three catalytic residues (Glu, Lys, Cys) [4] were aligned correctly. This data set was used for phyloge- netic inference with the phyml online platform [43] with 100 bootstrap trials. The resulting tree was visualized using treeview, version 1.6.6 (http://taxonomy.zoology.gla.ac.uk/ rod/treeview.html). Accession numbers of proteins used in phylogenetic analysis are presented in Table S1. For illustrative purposes, a 3D model was generated based on the crystal structure of the D. melanogaster 3-urei- dopropionase (Protein Data Bank code: 2VHH) (Fig. S2). The swiss-model workspace [44] was used with standard settings, and molecular visualization was conducted using pymol (The PyMOL Molecular Graphics System, version 1.2r3pre; Schro ¨ dinger, LLC, Mannheim, Germany). Acknowledgements Some of the organisms used in the present study were kindly provided by the Caenorhabditis Genetics Center (funded by the NIH National Center for Research Resources). Andrew Fire is acknowledged for the plas- mid pPD95.77. Some of the substrates tested were donated by Markus Piotrowski. Financial and material support for this project was generously provided by Ru ¨ diger J. Paul. References 1 Wasternack C (1980) Degradation of pyrimidines and pyrimidine analogs-pathways and mutual influences. Pharmacol Ther 8, 629–651. 2 Hayaishi O & Kronberg A (1952) Metabolism of cyto- sine, thymine, uracil, and barbituric acid by bacterial enzymes. J Biol Chem 197, 717–732. 3 Andersen G, Bjo ¨ rnberg O, Polakova S, Pynyaha Y, Rasmussen A, Møller K, Hofer A, Moritz T, Sandrini MPB, Merico A-M et al. (2008) A second pathway to degrade pyrimidine nucleic acid precursors in eukary- otes. J Mol Biol 380, 656–666. 4 Pace HC & Brenner C (2001) The nitrilase superfamily: classification, structure and function. Genome Biol 2, reviews0001.1–reviews0001.9. 5 Brenner C (2002) Catalysis in the nitrilase superfamily. Curr Opin Struct Biol 12, 775–782. 6 Bork P & Koonin EV (1994) A new family of carbon- nitrogen hydrolases. Protein Sci 3, 1344–1346. 7 Vorwerk S, Biernacki S, Hillebrand H, Janzik I, Mu ¨ ller A, Weiler EW & Piotrowski M (2001) Enzymatic characterization of the recombinant Arabidopsis thaliana nitrilase subfamily encoded by the NIT2 ⁄ NIT1 ⁄ NIT3- gene cluster. Planta 212, 508–516. 8 Piotrowski M, Janowitz T & Kneifel H (2003) Plant C-N hydrolases and the identification of a plant N- carbamoylputrescine amidohydrolase involved in polyamine biosynthesis. J Biol Chem 278, 1708–1712. 9 Walsh TA, Green SB, Larrinua IM & Schmitzer PR (2001) Characterization of plant b-ureidopropionase and functional overexpression in Escherichia coli. Plant Physiol 125, 1001–1011. 10 Kvalnes-Krick KL & Traut TW (1993) Cloning, sequencing, and expression of a cDNA encoding T. Janowitz et al. 3-Ureidopropionase of C. elegans FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 4107 b-alanine synthase from rat liver. J Biol Chem 268, 5686–5693. 11 Griffith OW (1986) b-Amino acids: mammalian metab- olism and utility as a -amino acid analogues. Annu Rev Biochem 55, 855–878. 12 Bauer K (2005) Carnosine and homocarnosine, the for- gotten, enigmatic peptides of the brain. Neurochem Res 30, 1339–1345. 13 van Gennip AH, Abeling NGGM, Vreken P & van Kuilenburg ABP (1997) Inborn errors of pyrimidine degradation: clinical, biochemical and molecular aspects. J Inherit Metab Dis 20, 203–213. 14 Nyhan WL (2005) Disorders of purine and pyrimidine metabolism. Mol Genet Metab 86, 25–33. 15 Kuilenburg ABP, Meinsma R, Beke E, Assmann B, Ribes A, Lorente I, Busch R, Mayatepek E, Abeling NGGM, van Cruchten A et al. (2004) b-Ureidopropion- ase deficiency: an inborn error of pyrimidine degrada- tion associated with neurological abnormalities. Hum Mol Genet 13, 2793–2801. 16 Heggie G, Sommadossi J, Cros D, Huster W & Diasio R (1987) Clinical pharmacokinetics of 5-fluorouracil in plasma, urine, and bile. Cancer Res 47, 2203–2206. 17 van Kuilenburg ABP (2004) Dihydropyrimidine dehy- drogenase and the efficacy and toxicity of 5-fluoroura- cil. Eur J Cancer 40, 939–950. 18 Kim S, Park D-H & Shim J (2008) Thymidylate syn- thase and dihydropyrimidine dehydrogenase levels are associated with response to 5-fluorouracil in Caenor- habditis elegans. Mol Cells 26, 344–349. 19 Takemoto T, Sasaki Y, Hamajima N, Goshima Y, Nanaka M & Kimura H (2000) Cloning and character- ization of the Caenorhabditis elegans CeCRMP ⁄ DHP-1 and -2; common ancestors of CRMP and dihydro- pyrimidinase. Gene 261, 259–267. 20 Beck H, Dobritzsch D & Pis ˇ kur J (2008) Saccharo- myces kluyveri as a model organism to study pyrimidine degradation. FEMS Yeast Res 8, 1209–1213. 21 Andersen B, Lundgren S, Dobritzsch D & Pis ˇ kur J (2008) A recruited protease is involved in catabolism of pyrimidines. J Mol Biol, 379, 243–250. 22 Nakai T, Hasegawa T, Yamashita E, Yamamoto M, Kumasaka T, Ueki T, Nanba H, Ikenaka Y, Takahashi S, Sato M et al. (2000) Crystal structure of N- carbamyl-D-amino acid amidohydrolase with a novel catalytic framework common to amidohydrolases. Structure 8, 729–737. 23 Lundgren S, Lohkamp B, Andersen B, Pis ˇ kur J & Dobritzsch D (2008) The crystal structure of b-alanine synthase from Drosophila melanogaster reveals a homo- octameric helical turn-like assembly. J Mol Biol 377, 1544–1559. 24 Tamaki N, Mizutani N, Kikugawa M, Fujimoto S & Mizota C (1987) Purification and properties of b-ureido- propionase from the rat liver. Eur J Biochem 169, 21–26. 25 Sakamoto T, Sakata SF, Matsuda K, Horikawa Y & Tamaki N (2001) Expression and properties of human liver beta-ureidopropionase. J Nutr Sci Vitaminol (Tokyo) 47, 132–138. 26 Matthews MM, Liao W, Kvalnes-Krick KL & Traut TW (1992) b-Alanine synthase: purification and alloste- ric properties. Arch Biochem Biophys 293, 254–263. 27 Rual JF, Ceron J, Koreth J, Hao T, Nicot AS, Hiroz- ane-Kishikawa T, Vandenhaute J, Orkin SH, Hill DE, van den Heuvel S et al. (2004) Toward improving Caenorhabditis elegans phenome mapping with an OR- Feome-based RNAi library. Genome Res 14, 2162–2168. 28 Kamath RS, Fraser AG, Dong Y, Poulin G, Durbin R, Gotta M, Kanapin A, Le Bot N, Moreno S, Sohrmann M et al. (2003) Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature 421, 231–337. 29 Sonnichsen B, Koski LB, Walsh A, Marschall P, Neu- mann B, Brehm M, Alleaume AM, Artelt J, Bettencourt P, Cassin E et al. (2005) Full-genome RNAi profiling of early embryogenesis in Caenorhabditis elegans. Nature 434, 462–469. 30 Kumar S, Aninat C, Michaux G & Morel F (2010) Anticancer drug 5-fluorouracil induces reproductive and developmental defects in Caenorhabditis elegans. Reprod Toxicol 29, 415–420. 31 Rawls JM Jr (2006) Analysis of pyrimidine catabolism in Drosophila melanogaster using epistatic interactions with mutations of pyrimidine biosynthesis and b-alanine metabolism. Genetics 172, 1665–1674. 32 Lecroisey C, Se ´ galat L & Gieseler K (2007) The C. ele- gans dense body: anchoring and signaling structure of the muscle. J Muscle Res Cell Motil 28, 79–87. 33 Vreken P, van Kuilenburg ABP, Hamajima N, Mei- nsma R, van Lenthe H, Go ¨ hlich-Ratmann G, Assmann BE, Wevers RA & van Gennip AH (1999) cDNA clon- ing, genomic structure and chromosomal localization of the human BUP-1 gene encoding b-ureidopropionase. Biochim Biophys Acta 1447, 251–257. 34 Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77, 71–94. 35 Scha ¨ gger H, Cramer W & von Jagow G (1994) Analy- sis of molecular masses and oligomeric states of pro- tein complexes by blue native electrophoresis and isolation of membrane protein complexes by two- dimensional native electrophoresis. Anal Biochem 217, 220–230. 36 Bolleter WT, Bushman CJ & Tidwell PW (1961) Spec- trophotometric detection of ammonia as indophenol. Anal Chem 33, 592–594. 37 Hiller A & van Slyke D (1933) Determination of ammo- nia in blood. J Biol Chem 102, 499. 38 Traut TW & Loechel S (1984) Pyrimidine catabolism: individual characterization of the three sequential enzymes with a new assay. Biochemistry 23, 2533–2539. 3-Ureidopropionase of C. elegans T. Janowitz et al. 4108 FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 39 Naguib FNM, el Kouni MH & Cha S (1985) Enzymes of uracil catabolism in normal and neoplastic human tissues. Cancer Res 45, 5405–5412. 40 Ko ¨ nig S, Haegele KD & Fales HM (1998) Comment on the cylindrical capacitor electrospray interface. Anal Chem 70, 4453–4455. 41 Wolf M, Nunes F & Paul RJ (2005) Coordinates, DNA content and heterogeneity of cell nuclei and segments of the Caenorhabditis elegans intestine. Histochem Cell Biol 124, 359–367. 42 Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F & Higgins DG (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25, 4876– 4882. 43 Guindon S & Gascuel O (2003) A simple, fast and accu- rate algorithm to estimate large phylogenies by maxi- mum likelihood. Syst Biol 52, 696–704. 44 Arnold K, Bordoli L, Kopp J & Schwede T (2006) The SWISS-MODEL Workspace: a web-based environment for protein structure homology modelling. Bio- informatics, 22, 195–201. 45 Wasternack C, Lippmann G & Reinbotte H (1979) Purification and properties of b-ureidopropionase of Euglena gracilis. Biochim Biophys Acta, 570, 341–351. Supporting information The following supplementary material is available: Fig. S1. Metabolism of 3-aminopropionate. Fig. S2. Homology model of the subunit of 3-ureido- propionase, based upon the crystal structure of the D. melanogaster 3-ureidopropionase (Protein Data Bank code: 2VHH). Fig. S3. To determine whether 3-ureidopropionase deficiency has any effect on worms exposed to 5-FU, RNAi- and control worms (carrying the feeding vector L4440) were incubated with low doses of 5-FU. Table S1. Proteins used in phylogenetic analysis. Enzyme classes and branches were assigned as described previously [4]. This supplementary material can be found in the online version of this article. Please note: As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. T. Janowitz et al. 3-Ureidopropionase of C. elegans FEBS Journal 277 (2010) 4100–4109 ª 2010 The Authors Journal compilation ª 2010 FEBS 4109 . The 3-ureidopropionase of Caenorhabditis elegans , an enzyme involved in pyrimidine degradation Tim Janowitz, Irene Ajonina, Markus Perbandt, Christian. 3-ureidopropionase is involved in the synthesis of the pantothenate moity of coenzyme A and 3-aminopropionate can serve as osmoprotectant [9]. Those different roles in metabolism