Caenorhabditis elegans has two genes encoding functional D-aspartate oxidases Masumi Katane*, Yousuke Seida*, Masae Sekine, Takemitsu Furuchi and Hiroshi Homma Laboratory of Biomolecular Science, School of Pharmaceutical Sciences, Kitasato University, Tokyo, Japan As all amino acids except Gly have an asymmetric a-carbon atom to which both amino and carboxyl groups attach, they can exist as two kinds of stereo- isomers, namely, the l-forms and d-forms. It has long been believed that d-amino acids do not play a significant role in the physiology of most organisms, apart from bacteria, where they are essential as com- ponents of cell wall peptidoglycans and antibiotic peptides. However, advances in the methods used to separate chiral amino acids have revealed that var- ious living organisms contain several d-amino acids, either in their free form or as protein components. Studies to elucidate the physiological roles of such in vivo d-amino acids have focused particularly on free d-Ser and d-Asp. d-Ser was first found in the early 1960s in lower animals such as earthworms [1] and silkworms [2]. It was reported that the d-Ser concentrations in the blood of the silkworm increased at particular stages of metamorphosis [3], although how d-Ser acts in metamorphosis remains unclear. d-Ser was also found in the mammalian forebrain, where it persists over the lifetime of the animal at Keywords Caenorhabditis elegans; D-amino acid; D-amino acid oxidase; D-aspartate oxidase; flavoprotein Correspondence H. Homma, Laboratory of Biomolecular Science, School of Pharmaceutical Sciences, Kitasato University, 5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan Fax: +81 3 5791 6381 Tel: +81 3 5791 6229 E-mail: hommah@pharm.kitasato-u.ac.jp *These authors contributed equally to this work Note Nucleotide sequence data reported are avail- able in the DDBJ ⁄ EMBL ⁄ GenBank databas- es under the accession numbers AB275890, AB275891, AB275892, and AB275893 (Received 17 August 2006, revised 31 October, accepted 6 November 2006) doi:10.1111/j.1742-4658.2006.05571.x Four cDNA clones that were annotated in the database as encoding d-amino acid oxidase (DAAO) or d-aspartate oxidase (DASPO) were iso- lated by RT-PCR from Caenorhabditis elegans RNA. The proteins (Y69Ap, C47Ap, F18Ep, and F20Hp) encoded by the cloned cDNAs were expressed in Escherichia coli as recombinant proteins with an N-terminal His-tag. All proteins except F20Hp were recovered in the soluble fractions. The recombinant Y69Ap has functional DAAO activity, as it can deami- nate neutral and basic d-amino acids, whereas the recombinants C47Ap and F18Ep have functional DASPO activities, as they can deaminate acidic d-amino acids. Additional experiments using purified recombinant proteins revealed that Y69Ap deaminates d-Arg more efficiently than d-Ala and d- Met, and that C47Ap and F18Ep show distinct kinetic properties against d-Asp, d-Glu, and N-methyl-d-Asp. This is the first time that cDNA clo- ning of invertebrate DAAO and DASPO genes has been reported. In addi- tion, our study reveals for the first time that C. elegans has at least two genes encoding functional DASPOs and one gene encoding DAAO, although it had previously been thought that organisms only bear one copy each of these genes. The two C. elegans DASPOs differ in their substrate specificities and possibly also in their subcellular localization. Abbreviations C47Ap, the C47A10.5 gene product; DAAO, D-amino acid oxidase; DASPO, D-aspartate oxidase; F18Ep, the F18E3.7a gene product; F20Hp, the F20H11.5 gene product; NMDA, N-methyl- D-aspartate; PTS1, type 1 peroxisomal targeting signal; PTS2, type 2 peroxisomal targeting signal; Y69Ap, the Y69A2AR.5 gene product. FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS 137 high concentrations. d-Ser is also known to bind the Gly-binding site of the N-methyl-d-Asp (NMDA) subtype of the Glu receptor and to potentiate gluta- matergic neurotransmission [4,5]. Consequently, it has been proposed that d-Ser regulates the Glu-mediated activation of the receptor by acting as a co-agonist. With regard to free d-Asp, it differs from free d-Ser in that it appears only transiently in various tissues and cells of invertebrates and vertebrates. Several lines of evidence have suggested that d-Asp plays an important role in a wide variety of biological activities in the nervous and endocrine systems, inclu- ding hormonal secretion and steroidogenesis [6–17], although the target molecule(s) of d-Asp remain to be identified. d-Amino acid oxidase (DAAO, EC 1.4.3.3) and d-Asp oxidase (DASPO, EC 1.4.3.1) are FAD-containing flavoproteins that catalyze the oxidative deamination of d-amino acids with oxygen; this reaction produces hydrogen peroxide, ammonia, and the corresponding 2-oxo acid. The two enzymes are similar in molecular mass, primary structure, and the type of catalytic reac- tion, but differ in their substrate specificity. DAAO displays broad substrate specificity and generally pre- fers nonpolar neutral d-amino acids such as d-Ala and d-Met. In contrast, DASPO is highly specific for acidic d-amino acids such as d-Asp and d-Glu, which are not DAAO substrates. These enzymes have been found in various eukaryotes and are reported to be localized in the cellular peroxisomes [18–24]. Their biochemical properties have been thoroughly investigated in vitro by using proteins purified from fungus bodies [25–27] and animal tissues [28–30]. In addition, the DAAO-enco- ding genes of yeasts [31–33], fungus [34], fishes [35], and mammals [36–42], and the DASPO-encoding genes of yeast [43] and mammals [44–46] have been cloned and overexpressed in heterologous organisms (e.g. Escherichia coli and yeast) for functional characteriza- tion of the resulting recombinant proteins. Notably, neither DAAO nor DASPO activity has been reported in prokaryotic cells, which lack peroxisomes. However, recent genome analyses have revealed that eukaryotic DAAO and DASPO gene homologues exist in the genomes of eubacteria such as Streptomyces coelicolor [47], Mycobacterium tuberculosis [48,49], and Pseudo- monas aeruginosa [50]. The nematode Caenorhabditis elegans is a multicellu- lar model animal, the genome sequencing of which was completed in 1998 [51]. A large number of protein-cod- ing regions have been detected in its genome, and the functions of the protein products have been predicted by gene annotation techniques. Although it has been thought that organisms only bear one copy each of the DAAO and DASPO genes, one of the C. elegans databases, WormBase (http://www.wormbase.org/), has annotated four different genes (Y69A2AR.5, C47A10.5, F18E3.7a, and F20H11.5 genes) as encoding DAAO or DASPO. However, it has not been con- firmed experimentally that the proteins encoded by these genes (Y69Ap, C47Ap, F18Ep, and F20Hp, respectively) are functional DAAOs or DASPOs. In this report, we addressed this issue by first cloning the cDNAs of these C. elegans genes and expressing the recombinant proteins in E. coli. The enzymatic and kinetic properties of the proteins were then determined. This is the first time the cloning of invertebrate DAAO and DASPO cDNAs has been reported. Moreover, this study reveals that organisms can bear two genes encoding functional DASPOs. In C. elegans, the two DASPOs differ in their kinetic properties and perhaps also in their subcellular distribution. Results Isolation and sequence analysis of the C. elegans genes encoding DAAO and DASPO cDNA fragments corresponding to the sequences of each ORF of Y69A2AR.5, C47A10.5, F18E3.7a, and F20H11.5 genes were cloned as described in Experi- mental procedures. Sequence analysis of the cloned cDNAs revealed that their sequences were identical with the database sequences. Consequently, the Y69Ap, C47Ap, F18Ep, and F20Hp cDNAs were pre- dicted to encode proteins with 322, 334, 334, and 383 amino acids and calculated molecular masses of 36 117, 37 636, 37 607, and 42 501 Da, respectively. C47Ap, F18Ep, and F20Hp shared relatively high amino-acid sequence identity with each other (43.3– 69.0%) but were less homologous to Y69Ap (25.9– 30.2%). The four C. elegans proteins were only moder- ately homologous to the DAAOs and DASPOs of micro-organisms (19.8–28.7%) and vertebrates (27.0– 37.5%). Phylogenetic analysis also revealed that, whereas C47Ap, F18Ep, and F20Hp clustered together, the C. elegans proteins did not form a closed cluster with any of the DAAOs or DASPOs of other organisms (Fig. 1). These findings suggest that a pre- cursor of the C. elegans proteins evolved from a pre- cursor of the DAAOs and DASPOs found in other organisms early during evolution. Alignment of the deduced amino-acid sequences of Y69Ap, C47Ap, F18Ep, and F20Hp with those of pig DAAO [36], the yeast Rhodotorula gracilis DAAO [32], bovine DASPO [46], and mouse DASPO [44] revealed conservation of the FAD-binding consensus sequence Caenorhabditis elegans D-aspartate oxidase M. Katane et al. 138 FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS (GXGXXG) near the N-terminus (Fig. 2). Moreover, the amino-acid sequences of Y69Ap and C47Ap contain a C-terminal consensus sequence [(S ⁄ A ⁄ C)- (K ⁄ R ⁄ H)-L] that is a type 1 peroxisomal targeting signal (PTS1) [52]. The amino-acid sequences of F18Ep and F20Hp lack this sequence. The Tyr224, Tyr228, Arg283, and Gly313 residues in pig DAAO, and the Met213, Tyr223, Tyr238, Arg285, and Ser335 residues in R. gracilis DAAO have been identified as catalytically important residues by crystallographic analyses [53–56] and mutagenesis experiments [57–61]. The corresponding residues in bovine and mouse DASPOs have also been predicted to be catalytically important by modeling of the 3D structures of the DASPOs [44,61]. Moreover, a mutagenesis experiment has revealed that the Arg216 and Arg237 residues of mouse DASPO are catalytically important [44]. Our alignment analysis suggests that the Tyr215, Arg270, and Gly294 residues in Y69Ap, the Arg276 and Gly308 residues in C47Ap, the Arg279 and Ser311 resi- dues in F18Ep, and the Arg304 and Gly335 residues in F20Hp correspond to the above-mentioned residues of pig DAAO, R. gracilis DAAO, bovine DASPO, and mouse DASPO (Fig. 2). However, other enzymatically important residues mentioned are not conserved in the C. elegans proteins. Thus, the primary structures of the C. elegans proteins differ markedly from those of other reported DAAOs and DASPOs. It is possible that during evolution, the C. elegans proteins acquired dis- tinctive properties. Expression of the recombinant proteins in E. coli and characterization of their enzymatic properties To confirm that the cloned cDNAs encode functional DAAOs and DASPOs, expression plasmids for N-terminally His-tagged Y69Ap, C47Ap, F18Ep, and F20Hp were constructed. The molecular masses of the N-terminally His-tagged Y69Ap, C47Ap, F18Ep, and F20Hp were calculated to be 40 274, 41 794, 41 765, and 46 658 Da, respectively. E. coli strain BL21(DE3)- Fig. 1. Phylogenetic relationships of the C. elegans cDNA products with the DAAOs and DASPOs of other organisms. A data file for the phylogenetic tree was created with the CLUSTALW Multiple Sequence Alignment program (version 1.83) [75], and the phylo- genetic tree was generated by using the NJ PLOT software [76]. The fruit fly, mos- quito, and rat DASPOs are putative proteins. The bacterial DAAOs are also putative pro- teins and are used as an out-group. The scale bar indicates a distance of 0.05 substi- tutions per site. The UniProt accession num- bers are: P. aeruginosa DAAO, P33642; S. coelicolor DAAO, Q9X7P6; M. tuberculo- sis DAAO, O07727; R. gracilis DAAO, P80324; Trigonopsis variabilis DAAO, Q99042; Fusarium solani DAAO, P24552; Candida boidinii DAAO, Q9HGY3; Crypto- coccus humicola DASPO, Q75WF1; fruit fly DASPO, Q9VM80; mosquito DASPO, Q7Q7G4; rabbit DAAO, P22942; pig DAAO, P00371; guinea pig DAAO, Q9Z1M5; human DAAO, P14920; rat DAAO, O35078; mouse DAAO, P18894; hamster DAAO, Q9Z302; carp DAAO, Q6TGN2; rat DASPO, UPI000017E4D7; mouse DASPO, Q922Z0; human DASPO-1, Q99489; bovine DASPO, P31228. M. Katane et al. Caenorhabditis elegans D-aspartate oxidase FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS 139 Fig. 2. Comparison of the amino acid sequences of C. elegans cDNAs with those of the DAAOs and DASPOs of other organisms. The deduced amino-acid sequences of Y69Ap, C47Ap, F18Ep, F20Hp, pig DAAO [36], R. gracilis DAAO [32], bovine DASPO [46], and mouse DASPO [44] were aligned by using the CLUSTALW Multiple Sequence Alignment program (version 1.83) [75]. The amino-acid numbers of each sequence are indicated on the right. Asterisks indicate amino-acid residues that are identical in all sequences. Conserved amino-acid substi- tutions with low and high similarity are indicated by dots and double-dots, respectively. The FAD-binding motif (GXGXXG) is boxed. Amino- acid residues that were experimentally proven to be catalytically important are shown as white letters on a black background. Amino-acid residues presumed to be catalytically important are shaded in gray. Caenorhabditis elegans D-aspartate oxidase M. Katane et al. 140 FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS pLysS cells were transformed with each expression construct, then crude extracts and insoluble fractions were subjected to western blot analysis. Recombinant Y69Ap, C47Ap, and F18Ep were detected in the crude extracts (Fig. 3A), and their apparent molecular mas- ses were in good agreement with those calculated from their deduced amino-acid sequences (Fig. 2). The insol- uble fractions of the bacterial lysates exhibited intense bands that had the same mobility as the bands in the crude extracts (data not shown). Thus, the Y69Ap, C47Ap, and F18Ep recombinant proteins were expressed both as soluble and insoluble forms. In con- trast, recombinant F20Hp was only detected in the insoluble fraction (data not shown). We are now searching for the optimal conditions that would allow soluble recombinant F20Hp to be expressed. The crude extract and insoluble fraction of cells transformed with the parental plasmid did not have bands that were recognized by the antibody to His-tag (Fig. 3A and data not shown). We subsequently examined the enzymatic activity of the recombinant proteins against various amino acids. Crude extracts of the transformed cells served as the enzyme sources. The Y69Ap plasmid-transformed cell extracts reproducibly showed enzymatic activity against d-Ala. Three independent assays revealed that this activity was 216.7 ± 23.9 mUÆ(mg protein) )1 (mean ± SD). We repeatedly observed low levels of activity against l-Ala, which is an enantiomer of d-Ala (Table 1). This may be due to the fact that E. coli has Ala racemase, which catalyzes the direct interconver- sion between l-Ala and d-Ala [62] and thus may con- vert l-Ala into d-Ala. Of the other neutral d-amino acids examined, the Y69Ap extract was more active against d-Met than against d-Ala but showed low A B Fig. 3. Analysis of the expression of recom- binant proteins in E. coli and their purity. (A) Cellular expression of recombinant Y69Ap, C47Ap, and F18Ep was examined by west- ern blotting of the crude extracts (20 lg) using a His-tag antibody. (B) The proteins in crude extracts (20 lg) and the purified enzymes (0.5 lg) were separated on an SDS ⁄ 12% polyacrylamide gel and stained with Coomassie Brilliant Blue R-250. Crude extracts from the parental plasmid pRSET-B were also tested (Mock). MWM, molecular mass marker proteins. Table 1. Oxidase activities of the recombinant proteins against var- ious amino acids. Appropriate amounts of crude extracts (10, 40, and 15 lg of the Y69Ap, C47Ap, and F18Ep extracts, respectively) were used as enzyme. Percentage activities relative to that detec- ted with D-Ala are shown for Y69Ap. Similarly, percentage activities relative to those with D-Asp are shown for C47Ap and F18Ep. A 445 corresponding to 100% values of Y69Ap, C47Ap, and F18Ep was 0.745, 0.396, and 1.072, respectively. Each value shown is the mean ± SD from three independent assays. ND, Not determined; NMLA, N-methyl- L-aspartate. Substrate Relative activity (%) Y69Ap C47Ap F18Ep D-Ala 100 ± 11 4.5 ± 1.2 2.8 ± 0.2 D-Met 172 ± 19 3.8 ± 1.5 2.4 ± 1.1 D-Asn 19 ± 1.5 17 ± 3.2 6.1 ± 1.1 D-Arg 55 ± 6.4 < 0.1 < 0.1 D-Asp 2.2 ± 1.8 100 ± 2.1 100 ± 4.9 D-Glu < 0.1 247 ± 16 67 ± 2.0 NMDA < 0.1 313 ± 16 110 ± 2.6 L-Ala 34 ± 1.8 ND ND L-Met 1.9 ± 1.7 ND ND L-Arg < 0.1 ND ND L-Asp ND 2.6 ± 0.3 < 0.1 L-Glu ND 1.3 ± 0.9 0.3 ± 0.3 NMLA ND 6.7 ± 0.3 1.1 ± 1.0 M. Katane et al. Caenorhabditis elegans D-aspartate oxidase FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS 141 activities against d-Asn. The Y69Ap extract was also moderately active against the basic d-amino acid, d-Arg, but had very low or undetectable activity against the enantiomers, l-Met and l-Arg, and all acidic d-amino acids examined. Thus, the protein encoded by Y69Ap cDNA can catalyze the deamina- tion of neutral and basic d-amino acids. Three independent assays revealed that, unlike the Y69Ap extract, the C47Ap and F18Ep extracts had reproducible enzymatic activity against d-Asp [28.8 ± 0.6 and 208.1 ± 10.1 mUÆ(mg protein) )1 , respectively]. These assays also revealed that both extracts had activities against other acidic d-amino acids, namely, d-Glu and NMDA (Table 1). Only very low or undetectable activities were detected against the enantiomers l-Asp, l-Glu, and N-methyl-l-Asp and all neutral and basic d-amino acids examined. Thus, the proteins encoded by the cloned C47Ap and F18Ep cDNAs can catalyze the deamination of acidic d-amino acids. The crude extract of cells transformed with the parental plasmid lacked activity against every d-amino acid and l-amino acid examined (data not shown). Together, these observations confirm that the Y69Ap cDNA encodes a functional DAAO that is specific for a basic d-amino acid as well as for non- polar neutral d-amino acids, whereas the C47Ap and F18Ep cDNAs encode functional DASPOs that act on acidic d-amino acids. Purification of the recombinant proteins and their kinetic characterization To further characterize Y69Ap, C47Ap, and F18Ep, the recombinant enzymes were purified to near-homo- geneity by affinity chromatography using a chelating column (Fig. 3B). The specific activity of purified Y69Ap against d-Ala was 4.96 UÆ(mg protein) )1 , which is about 22.9 times higher than that of crude extract. As expected, purified Y69Ap lacked enzymatic activity against l-Ala (data not shown), which con- firms that the activity of Y69Ap against d-Ala is stere- ospecific. We then obtained Hanes–Woolf plots to determine the apparent kinetic parameters of the deamination of d-Ala, d-Met and d-Arg catalyzed by purified Y69Ap (Fig. 4A). The V max (maximal velocity) values were 5.41, 7.43 and 2.52 UÆ(mg protein) )1 for d-Ala, d-Met and d-Arg, respectively. The k cat (molecular activity) values (calculated from the V max values and the estimated molecular mass of N-termin- ally His-tagged Y69Ap) are listed in Table 2. Thus, the highest V max and k cat values for Y69Ap were against d-Met, followed by d-Ala and d-Arg. This matches the hierarchy of Y69Ap activities revealed by the experiments with the crude extract (Table 1). However, the K m (Michaelis constant) value against d-Arg was at least 10 times lower than those against d-Ala and d-Met (Table 2). Therefore, the catalytic efficiency (expressed as k cat ⁄ K m ) of Y69Ap against d-Arg was 7.3 and 3.8 times higher than against d-Ala and d-Met, respectively. This indicates that Y69Ap prob- ably prefers basic d-amino acid(s) to neutral d-amino acids as its substrate. The specific activities of purified C47Ap and F18Ep against d-Asp were 4.99 and 4.33 UÆ(mg protein) )1 , –5 0 5 10 15 20 25 30 35 40 15 10 5 [S] / V (mM·mg·U –1 ) [S] (mM) A D-Ala D-Met D-Arg –5 0 5 10 15 20 25 30 35 40 15 10 5 [S] / V (mM·mg·U –1 ) [S] (mM) B D-Asp D-Glu NMDA –5 0 5 10 15 20 25 30 35 40 15 10 5 [S] / V (mM·mg·U –1 ) [S] (mM) C D-Asp D-Glu NMDA Fig. 4. Hanes–Woolf plots of the oxidase activity of the purified enzymes. Enzymatic activities were assayed by using purified Y69Ap (A), C47Ap (B), and F18Ep (C). The substrates used were D-Ala (s), D-Met (h), and D-Arg (n) for Y69Ap, and D-Asp (d), D-Glu (j), and NMDA (m) for C47Ap and F18Ep. Caenorhabditis elegans D-aspartate oxidase M. Katane et al. 142 FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS which are 173 and 20.8 times higher than the specific activities of the crude extracts, respectively. Hanes– Woolf plots to determine the apparent kinetic parame- ters of the deamination of d-Asp, d-Glu, and NMDA catalyzed by these enzymes (Fig. 4B,C) revealed that the V max values of purified C47Ap were 6.16, 7.62, and 8.80 UÆ(mg protein) )1 , respectively, and the V max val- ues of purified F18Ep were 4.40, 3.03, and 4.31 UÆ(mg protein) )1 , respectively. The k cat , K m , and k cat ⁄ K m values of these enzymes against d-Asp, d-Glu, and NMDA are listed in Table 2 and show that the cata- lytic efficiency of C47Ap against d-Glu and NMDA was about 10.9 and 3.4 times higher than that against d-Asp, largely because of differences in the K m values (substrate affinity). In contrast, the K m value of F18Ep against NMDA was significantly higher than its K m value against d-Asp and d-Glu. Therefore, the cata- lytic efficiency of F18Ep against NMDA was lower than against d-Asp, whereas it was equally as efficient against d-Glu and d-Asp. F18Ep was 3.8 times more efficient against d -Asp than C47Ap, and C47Ap was 2.7 and 4.5 times more efficient against d-Glu and NMDA than F18Ep, respectively. Thus, although C47Ap and F18Ep both act on acidic d-amino acids, they differ in their substrate specificity profiles. Discussion The deduced C-terminal amino acids of Y69Ap and C47Ap are SKL (Fig. 2), which corresponds to the PTS1 consensus sequence [52]. These enzymes are thus predicted to localize to peroxisomes, like the DAAOs and DASPOs of other organisms [18–24]. In contrast, the three deduced C-terminal amino acids of F18Ep and F20Hp were LGL and LDD, respectively, which do not correspond to the PTS1 consensus sequence. The sequences of F18Ep and F20Hp also lacked the bipartite consensus sequence -(R ⁄ K)-(L ⁄ V ⁄ I)-X5- (H ⁄ Q)-(L ⁄ A)-, a type 2 peroxisomal targeting signal (PTS2) [52]. In C. elegans, the PTS2-dependent path- way was reported to be absent [63]. Representative sig- nal sequences that prompt the localization of proteins to organelles other than peroxisome were also not found in the F18Ep and F20Hp sequences. Hence, F18Ep and F20Hp probably localize to the cytoplasm. However, another possibility is that these proteins localize to peroxisomes via a PTS1-independent import pathway. The existence of such a novel pathway is sug- gested by the report that the peroxisomal importation of acyl-CoA oxidase of the yeast, Saccharomyces cerevisiae, was not disturbed in cells that lacked PTS1-dependent or PTS2-dependent importation [64]. Alternatively, it is possible that F18Ep and F20Hp are imported into peroxisomes by forming a complex(es) with PTS1-bearing proteins such as Y69Ap and C47Ap. This study demonstrates that the DAAO (Y69Ap) and DASPOs (C47Ap and F18Ep) in C. elegans can be expressed in E. coli as functional recombinant proteins. E. coli has often been used as a host organism to pre- pare recombinant DAAOs and DASPOs of various organisms. However, it can be difficult to overexpress these enzymes in active and soluble forms in E. coli for the following reasons. First, DAAO and DASPO cata- lyze the deamination of d-Ala and d-Glu, respectively, which are essential components of the peptidoglycans in bacterial cell walls. Secondly, the enzymatic reactions catalyzed by DAAO and DASPO produce hydrogen peroxide, which is highly toxic for E. coli. Conse- quently, overexpression of these enzymes may inhibit E. coli cell growth. Indeed, successful expression of mammalian DAAOs, apart from those from pigs, mice, and humans, in E. coli has not been reported. More- over, in the case of porcine DAAO, it was reported that only 25 mg purified enzyme was obtained from 40 g wet cell paste [59]. With regard to our own observa- tions, we found that recombinant Y69Ap was readily overexpressed in E. coli, recovered in the soluble frac- tion, and purified to near-homogeneity (Fig. 3). About 4 mg purified Y69Ap was obtained from 2 g wet cell paste. In contrast, recombinant F20Hp was not recov- ered in the soluble fraction. This may be related to the Table 2. Apparent steady-state kinetic parameters of the purified recombinant proteins against several D-amino acids. ND, Not determined. Substrate Y69Ap C47Ap F18Ep k cat (s )1 ) K m (mM) k cat ⁄ K m (s )1 ÆM )1 ) k cat (s )1 ) K m (mM) k cat ⁄ K m (s )1 ÆM )1 ) k cat (s )1 ) K m (mM) k cat ⁄ K m (s )1 ÆM )1 ) D-Ala 3.63 1.72 2113 ND ND ND ND ND ND D-Met 4.98 1.22 4082 ND ND ND ND ND ND D-Arg 1.69 0.11 15 394 ND ND ND ND ND ND D-Asp ND ND ND 4.29 2.02 2125 3.07 0.38 8066 D-Glu ND ND ND 5.31 0.23 23 066 2.11 0.25 8427 NMDA ND ND ND 6.13 0.84 7300 3.00 1.84 1629 M. Katane et al. Caenorhabditis elegans D-aspartate oxidase FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS 143 potentially adverse effects of overexpressing DAAOs and DASPOs in E. coli mentioned above. It will be necessary to improve the expression system and the purification procedure before we can functionally char- acterize recombinant F20Hp. A number of reports have characterized the enzymat- ic properties of the DAAOs of various organisms by analyzing recombinant proteins expressed in E. coli or yeast. These reports reveal that these DAAOs differ markedly in their activities and substrate specificities. For example, the k cat values of the porcine, human, R. gracilis, and carp DAAOs against d-Ala are repor- ted to be 6.4, 5.2, 350, and 190 s )1 , respectively [58,65– 67]. The k cat value of Y69Ap against d-Ala (3.63 s )1 )is similar to those of the pig and human DAAOs. More- over, like pig DAAO [59], Y69Ap was more active against d-Met than against d-Ala (Tables 1 and 2). In contrast, the R. gracilis and carp DAAOs are more act- ive against d-Ala than against any of the other amino- acid substrates examined [61,67]. This suggests that Y69Ap may be more similar in terms of its enzymatic properties to mammalian DAAOs than to microbial and fish DAAOs. However, Y69Ap was also active with d-Arg, which is a poor substrate for pig DAAO. This disparity may relate to structural difference(s) between the active sites of Y69Ap and pig DAAO. This is supported by the fact that, whereas the catalytically important Tyr228, Arg283, and Gly313 residues in pig DAAO [54,55,59] are all conserved in Y69Ap, Tyr224 is not (Fig. 2). Moreover, comparison of the experi- mentally determined crystal structure of pig DAAO [54] with the predicted 3D structure of Y69Ap generated with the SWISS-MODEL server [68] suggests that the Y69Ap residue Phe229 is in the same structural posi- tion as Tyr224 of pig DAAO (data not shown). Muta- genesis experiments and crystallographic analyses will be necessary to elucidate the role of Phe229 in the enzy- matic activity of Y69Ap against basic d-amino acid(s). In this study, we have demonstrated that, although C47Ap and F18Ep both act on acidic d-amino acids, they differ in their kinetics and preferences for partic- ular substrates. C47Ap functioned more efficiently against d-Glu and NMDA than against d-Asp, whereas F18Ep acted more efficiently against d-Asp and d-Glu than against NMDA (Table 2). Examina- tion of the reported kinetic properties of recombinant mammalian and micro-organism DASPOs suggests that C47Ap and F18Ep are unusual with regard to their specificity for d-Glu. The catalytic efficiency of C47Ap against d-Glu is 23 066 s )1 Æm )1 , which is 108 and 923 times higher than the respective catalytic effi- ciencies of bovine and Cryptococcus humicola DASPOs against d-Glu (213 and 25.0 s )1 Æm )1 , respectively) [43,69]. Similarly, the catalytic efficiency of F18Ep against d-Glu is 8427 s )1 Æm )1 , which is 39.6 and 337 times higher than the catalytic efficiencies of bovine and C. humicola DASPOs, respectively. As C. elegans lives in soil and eats micro-organisms that are prob- ably rich in d-Glu, it is possible that diet-derived d-Glu is incorporated into the body of C. elegans. Although little is currently understood about the amounts and physiological functions of d-amino acids in C. elegans, it was recently reported that injection of d-Glu into a silkworm, which is a multicellular model insect, induced muscle contraction [70]. It is possible that excess amounts of d-Glu are as toxic for C. ele- gans as they are for the silkworm, and that C. elegans may need C47Ap and F18Ep to deaminate d-Glu and thereby neutralize the toxicity of diet-derived d-Glu. To our knowledge, this is the first report of the clo- ning of invertebrate DAAO and DASPO cDNAs. In addition, we have demonstrated for the first time that an organism ( C. elegans) can have multiple active DASPO (C47Ap and F18Ep) genes. The tissue localization of C47Ap and F18Ep within the body of C. elegans remains to be elucidated. As C47Ap and F18Ep are encoded by distinct genes in different loci in the C. elegans genome, their transcriptional regula- tions are possibly independent. Thus, it is possible that these two enzymes are tissue-specific isoforms in C. elegans. Green fluorescent protein (GFP)-based or b-galactosidase-based gene expression and in situ hybridization analyses may reveal the localization of C47Ap and F18Ep within the whole body of C. elegans. If these two proteins are expressed in the same cell, they may localize to distinct regions within the cell. It is likely that C47Ap is localized in the per- oxisome in a PTS1-dependent manner, whereas F18Ep remains in the cytoplasm. However, the biological significance of the multiple DASPOs in C. elegans is currently unclear. That other organisms may also have more than one DASPO is suggested by a study show- ing that two proteins, DASPO-1 and DASPO-2, are translated from a single human DASPO mRNA by alternative splicing [45], although the function of DASPO-2 remains to be clarified. Further studies will be needed to determine the tissue and cellular distri- butions of C47Ap and F18Ep and their expression during the development of C. elegans. Experimental procedures Animals and chemicals Caenorhabditis elegans Bristol strain N2 and E. coli strain OP50 were kindly provided by Y. Nakagawa (Laboratory Caenorhabditis elegans D-aspartate oxidase M. Katane et al. 144 FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS of Hygienic Chemistry, School of Pharmaceutical Sciences, Kitasato University, Japan). The C. elegans worms were maintained at 20 °C on NGM agar plates seeded with E. coli as described by Brenner [71]. d- and l-amino acids, ampicillin, BSA, and Aspergil- lus niger catalase were purchased from Sigma-Aldrich Inc. (St Louis, MO, USA). FAD, isopropyl b-d-thiogalactopyr- anoside, and imidazole were purchased from Wako Pure Chemical Industries (Osaka, Japan). Other chemicals were of the highest grade available and purchased from commer- cial sources. Isolation of cDNA clones from C. elegans Cultured C. elegans were collected by the standard method [71]. Their total RNAs were extracted by using ISOGEN reagent (Nippon Gene, Tokyo, Japan), according to the manufacturer’s instructions. For first-strand cDNA synthesis, the RNA samples (5 lg) were reverse-transcribed for 50 min at 42 °C in a 20-l L reaction volume with 200 U SuperScript II Reverse Transcriptase and 0.5 lg oligo(dT) 12)18 primer (Invitrogen, Carlsbad, CA, USA). The cDNAs of the Y69A2AR.5 (WormBase gene ID: WBGene00022076; genomic location: 2 614 199–2 617 030 on chromosome IV), C47A10.5 (WormBase gene ID: WBGene00008127; genomic location: 17 777 009–17 774 090 on chromosome V), F18E3.7a (WormBase gene ID: WBGene00017565; genomic location: 7 433 290–7 434 835 on chromosome V), and F20H11.5 (WormBase gene ID: WBGene00017648; genomic location: 6 587 836–6 589 212 on chromosome III) genes were amplified by PCR using the first-strand cDNA as a template and the following primers: for Y69A2AR.5,5¢-AGATCTATGCCTAAA ATTGCTGTACTAGGCGCAGG-3¢ (forward primer) and 5¢-GAATTCTCACAACTTCGACTTTTTCATTTTCAGC- 3¢ (reverse primer); for C47A10.5 ,5¢-AGATCTATGACT CCAAAAATCGCAATAATCGGCG-3¢ (forward primer) and 5¢-GGTACCTCACAGTTTCGAAGAATTTAGAGC GG-3¢ (reverse primer); for F18E3.7a,5¢-AGATCTAT GGCAAACATAATTCCGAAGATTGC-3¢ (forward pri- mer) and 5¢-GAATTCTTATAATCCTAGTGCAGTCTT AACAAG-3¢ (reverse primer); and for F20H11.5,5¢-AG ATCTATGCTGTATGCTCTTCTTCTCCTC-3¢ (forward primer) and 5¢-GGTACCCTAATCATCAAGATATTTA ACCCATTCGG-3¢ (reverse primer). These primer sets were designed so that (a) additional BglII sites were created at the 5¢ ends of the forward primers for all genes, (b) addi- tional EcoRI sites were created at the 5¢ ends of the reverse primers for the Y69A2AR.5 and F18E3.7a genes, and (c) additional KpnI sites were created at the 5¢ ends of the reverse primers for the C47A10.5 and F20H11.5 genes. The PCR products were cloned into pT7Blue (Novagen, Madi- son, WI, USA), thus generating pT7-Y69Ap, pT7-C47Ap, pT7-F18Ep, and pT7-F20Hp, and were then sequenced. Construction of recombinant protein expression plasmids To construct the Y69Ap expression plasmid, the 1.0-kb BglII–EcoRI fragment containing the entire Y69Ap-coding sequence of pT7-Y69Ap was subcloned into pRSET-B (Invitrogen), resulting in the N-terminally His-tagged Y69Ap expression plasmid pRSET-His-Y69Ap. Similarly, the 1.0-kb BglII–KpnI fragment containing the entire C47Ap-coding sequence of pT7-C47Ap, the 1.0-kb BglII– EcoRI fragment containing the entire F18Ep-coding sequence of pT7-F18Ep, and the 1.2-kb BglII–KpnI frag- ment containing the entire F20Hp-coding sequence of pT7-F20Hp were subcloned into pRSET-B, resulting in the N-terminally His-tagged C47Ap, F18Ep, and F20Hp expression plasmids (pRSET-His-C47Ap, pRSET-His- F18Ep, and pRSET-His-F20Hp), respectively. Expression and purification of recombinant proteins Escherichia coli strain BL21(DE3)pLysS cells transformed with the expression plasmids were grown at 37 °C with sha- king in LB medium containing ampicillin (100 lgÆmL )1 ). When A 620 reached 0.5, isopropyl b-d-thiogalactopyrano- side was added to a final concentration of 0.01 mm, and the cells were grown at 30 °C for an additional 20 h. The cells were pelleted by centrifugation at 10 000 g for 10 min at 4 °C in a Kubota RA-200J rotor using a model 1920 Kubota centrifuge (Kubota corporation, Tokyo, Japan). The pellets were then resuspended in lysis buffer consisting of BugBuster Protein Extraction Reagent (Novagen), 50 lm FAD, and protease inhibitors (Roche Applied Science, Mannheim, Germany) (5 mL lysis buffer per g wet cell paste was used). The cell suspension was incubated for 20 min at room temperature with gentle shaking. The resulting lysates were centrifuged at 12 000 g for 20 min at 4 °C (Kubota model 1920 centrifuge with RA-200J rotor) to pellet the insoluble cell debris. The supernatant (crude extract) was filtered through a 0.45 lm membrane filter (Asahi Techno Glass Corporation, Tokyo, Japan) and used immediately for further purification or stored frozen at )80 °C until use. To prepare the insoluble fraction, the pel- leted cell debris was resuspended in 10 mm phosphate- buffered saline (pH 7.4), mixed with an equal volume of 4% SDS solution, and boiled immediately. The N-terminally His-tagged recombinant proteins were purified by affinity chromatography using a chelating col- umn. Crude extracts, prepared as described above, were applied to a HiTrap Chelating HP column (1 mL; Amer- sham Biosciences, Piscataway, NJ, USA) equilibrated with 20 mm sodium dihydrogen phosphate buffer (pH 7.4) con- taining 0.5 m NaCl and 10 mm imidazole. The column was washed with the same buffer, and the bound proteins were M. Katane et al. Caenorhabditis elegans D-aspartate oxidase FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS 145 eluted with a linear gradient of 10–500 mm imidazole. Each fraction (2 mL) containing the recombinant proteins was mixed with 50 lL2mm FAD and dialyzed for 3 h twice against 1 L 10 mm sodium pyrophosphate buffer (pH 8.3) containing 2 mm EDTA, 5 mm 2-mercaptoethanol, and 10% glycerol. The dialyzed fractions were pooled as the purified enzyme and used immediately for enzyme assays or stored frozen at )80 °C until use. Detection of recombinant proteins The protein concentrations in the crude extracts, insoluble fractions, and purified enzymes were determined by the method of Bradford [72] using BSA as a standard. Crude extracts (20 lg) and insoluble fractions (20 lg) were subjec- ted to SDS ⁄ PAGE (12% gel) and western blotting using anti-(His-tag) serum (His-probe; Santa Cruz Biotechnology, Santa Cruz, CA, USA) (1 : 1000 dilution) as the primary antibody and horseradish peroxidase-conjugated anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) (1 : 5000 dilution) as the secondary reagent. The protein bands were visualized with an enhanced chemilumi- nescence reagent (Amersham Biosciences) and by exposure to Lumi-Film Chemiluminescent Detection Film (Roche Applied Science). To analyze the protein purity, the pro- teins in the crude extracts (20 lg) and the purified enzymes (0.5 lg) were separated on an SDS ⁄ 12% polyacrylamide gel and stained with Coomassie Brilliant Blue R-250. Broad- range molecular mass standards (Bio-Rad, Hercules, CA, USA) served as molecular mass marker proteins. Assays of enzymatic activity Oxidase activities were determined by colorimetric measure- ment of the corresponding 2-oxo acids produced from the amino acids used, as previously described by Nagata et al. [73] with following modifications. Appropriate amounts of enzyme (10–40 and 0.5–1.0 lg crude extracts and purified enzymes, respectively) were mixed with a reaction mixture consisting of 40 mm sodium pyrophosphate buffer (pH 8.3), 50 lm FAD, and 20 mm amino acid in a final volume of 150 lL, and incubated at 37 °C. The incubation times with crude extracts and purified enzymes were 30 and 15 min, respectively. Subsequently, 10 lL 100% (w ⁄ v) tri- chloroacetic acid was added to stop the reactions, and pro- teins were pelleted by centrifugation at 20 000 g for 10 min at 4 °C (Kubota model 1920 centrifuge with RA-48J rotor). The supernatant (150 lL) was mixed with 100 lL 0.1% (w ⁄ v) 2,4-dinitrophenylhydrazine in 2 m HCl, and incuba- ted at 37 °C for 15 min, then 750 lL of 3.75 m NaOH was added, and the solution was cleared by centrifugation at 20 000 g for 10 min at 4 °C (Kubota model 1920 centrifuge with RA-48J rotor). The absorbance of the supernatant at 445 nm was measured against the blank prepared from a reaction mixture lacking the amino acid. One unit of enzyme activity was defined as the production of 1 lmol of the corresponding 2-oxo acid per min under the above assay conditions. For kinetic analyses, different final con- centrations (0, 0.5, 1, 2, 5, 10, 20, and 40 mm) of several d-amino acids were used as substrates. In some cases, A. niger catalase (5 l g) was added to the reaction mixture to prevent the decarboxylation of the 2-oxo acid by the hydrogen peroxide that was produced by the reaction [74]. Acknowledgements We thank Professor Y. Nakagawa (School of Pharma- ceutical Sciences, Kitasato University, Japan) for pro- viding C. elegans Bristol strain N2 and E. coli strain OP50. References 1 Rosenberg H & Ennor AH (1960) Occurrence of free d-serine in the earthworm. Nature 187, 617–618. 2 Srinivasan NG, Corrigan JJ & Meister A (1962) d-Serine in the blood of the silkworm Bombyx mori and other lepidoptera. J Biol Chem 237 , 3844–3845. 3 Corrigan JJ & Srinivasan NG (1966) The occurrence of certain d-amino acids in insects. Biochemistry 5, 1185–1190. 4 Matsui T, Sekiguchi M, Hashimoto A, Tomita U, Nishikawa T & Wada K (1995) Functional comparison of d-serine and glycine in rodents: the effect on cloned NMDA receptors and the extracellular concentration. J Neurochem 65, 454–458. 5 Mothet JP, Parent AT, Wolosker H, Brady RO Jr, Linden DJ, Ferris CD, Rogawski MA & Snyder SH (2000) d-Serine is an endogenous ligand for the glycine site of the N-methyl-d-aspartate receptor. Proc Natl Acad Sci USA 97, 4926–4931. 6 D’Aniello A, Di Cosmo A, Di Cristo C, Annunziato L, Petrucelli L & Fisher G (1996) Involvement of d-aspar- tic acid in the synthesis of testosterone in rat testes. Life Sci 59, 97–104. 7 D’Aniello G, Tolino A, D’Aniello A, Errico F, Fisher GH & Di Fiore MM (2000) The role of d-aspartic acid and N-methyl- d-aspartic acid in the regulation of prolactin release. Endocrinology 141, 3862–3870. 8 D’Aniello A, Spinelli P, De Simone A, D’Aniello S, Branno M, Aniello F, Fisher GH, Di Fiore MM & Rastogi RK (2003) Occurrence and neuroendocrine role of d-aspartic acid and N-methyl-d-aspartic acid in Ciona intestinalis. FEBS Lett 552, 193–198. 9 Lamanna C, Assisi L, Botte V & Di Fiore MM (2006) Involvement of D-Asp in P450 aromatase activity and estrogen receptors in boar testis. Amino Acids, doi:10.1007 ⁄ s00726-006-0351-9. Caenorhabditis elegans D-aspartate oxidase M. Katane et al. 146 FEBS Journal 274 (2007) 137–149 ª 2006 The Authors Journal compilation ª 2006 FEBS [...]... on d-aspartate Biochim Biophys Acta 1379, 399–405 21 Parveen Z, Large A, Grewal N, Lata N, Cancio I, Cajaraville MP, Perry CJ & Connock MJ (2001) D-Aspartate oxidase and d-amino acid oxidase are localised in the peroxisomes of terrestrial gastropods Eur J Cell Biol 80, 651–660 22 Van Veldhoven PP, Brees C & Mannaerts GP (1991) d-Aspartate oxidase, a peroxisomal enzyme in Caenorhabditis elegans D-aspartate. .. Katane M, Furuchi T, Sekine M & Homma H (2006) Molecular cloning of a cDNA encoding mouse d-aspartate oxidase and functional characterization of its recombinant proteins by site-directed mutagenesis Amino Acids, doi:10.1007 ⁄ s00726-006-0350-x Setoyama C & Miura R (1997) Structural and functional characterization of the human brain d-aspartate oxidase J Biochem (Tokyo) 121, 798–803 Simonic T, Duga S, Negri... Escherichia coli chromosome: location and cloning of the ubiA and alr genes Gene 129, 9–16 Motley AM, Hettema EH, Ketting R, Plasterk R & Tabak FH (2000) Caenorhabditis elegans has a single pathway to target matrix proteins to peroxisomes EMBO Rep 1, 40–46 Zhang JW, Han Y & Lazarow PB (1993) Novel peroxisome clustering mutants and peroxisome biogenesis mutants of Saccharomyces cerevisiae J Cell Biol 123, 1133–1147... homologymodeling server Nucleic Acids Res 31, 3381–3385 Negri A, Tedeschi G, Ceciliani F & Ronchi S (1999) Purification of beef kidney d-aspartate oxidase overexpressed in Escherichia coli and characterization of its Caenorhabditis elegans D-aspartate oxidase 70 71 72 73 74 75 76 redox potentials and oxidative activity towards agonists and antagonists of excitatory amino acid receptors Biochim Biophys Acta 1431,... tripeptide Ser-Asn-Leu (SNL) of human d-aspartate oxidase is a functional peroxisome-targeting signal Biochem J 336, 367–371 19 Beard ME (1990) d-Aspartate oxidation by rat and bovine renal peroxisomes: an electron microscopic cytochemical study J Histochem Cytochem 38, 1377– 1381 20 Kera Y, Niino A, Ikeda T, Okada H & Yamada R (1998) Peroxisomal localization of d-aspartate oxidase and development of... 2006 The Authors Journal compilation ª 2006 FEBS 147 Caenorhabditis elegans D-aspartate oxidase 36 37 38 39 40 41 42 43 44 45 46 47 48 M Katane et al carp Cyprinus carpio and its induction with exogenous free d-alanine Arch Biochem Biophys 420, 121–129 Fukui K, Watanabe F, Shibata T & Miyake Y (1987) Molecular cloning and sequence analysis of cDNAs encoding porcine kidney d-amino acid oxidase Biochemistry... and expression of a cDNA encoding mouse kidney d-amino acid oxidase Gene 90, 293–297 Tsuchiya M, Kurabayashi A & Konno R (2003) Hamster d-amino-acid oxidase cDNA: rodents lack the 27th amino acid residue in d-amino-acid oxidase Amino Acids 24, 223–226 Takahashi S, Takahashi T, Kera Y, Matsunaga R, Shibuya H & Yamada RH (2004) Cloning and expression in Escherichia coli of the d-aspartate oxidase gene... Pollegioni L, Langkau B, Tischer W, Ghisla S & Pilone MS (1993) Kinetic mechanism of d-amino acid oxidases from Rhodotorula gracilis and Trigonopsis variabilis J Biol Chem 268, 13850–13857 Yamada R, Ujiie H, Kera Y, Nakase T, Kitagawa K, Imasaka T, Arimoto K, Takahashi M & Matsumura Y (1996) Purification and properties of d-aspartate oxidase from Cryptococcus humicolus UJ1 Biochim Biophys Acta 1294, 153–158 Curti... Bombyx mori J Biochem (Tokyo) 137, 199– 203 Brenner S (1974) The genetics of Caenorhabditis elegans Genetics 77, 71–94 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding Anal Biochem 72, 248–254 Nagata Y, Shimojo T & Akino T (1988) Two spectrophotometric assays for d-amino acid oxidase: for the study of...M Katane et al 10 Long Z, Lee JA, Okamoto T, Nimura N, Imai K & Homma H (2000) d-Aspartate in a prolactin-secreting clonal strain of rat pituitary tumor cells (GH3) Biochem Biophys Res Commun 276, 1143–1147 11 Nagata Y, Homma H, Lee JA & Imai K (1999) d-Aspartate stimulation of testosterone synthesis in rat Leydig cells FEBS Lett 444, 160–164 12 Pampillo M, del Carmen . Caenorhabditis elegans has two genes encoding functional D-aspartate oxidases Masumi Katane*, Yousuke Seida*, Masae Sekine, Takemitsu. invertebrate DAAO and DASPO genes has been reported. In addi- tion, our study reveals for the first time that C. elegans has at least two genes encoding functional DASPOs and one gene encoding DAAO, although. invertebrate DAAO and DASPO cDNAs has been reported. Moreover, this study reveals that organisms can bear two genes encoding functional DASPOs. In C. elegans, the two DASPOs differ in their kinetic