Spatiotemporal Localization of d-Amino Acid Oxidase and d-Aspartate Oxidases during Development in Caenorhabditis elegans

Abstract

Recent investigations have shown that a variety of d-amino acids are present in living organisms and that they possibly play important roles in physiological functions in the body. d-Amino acid oxidase (DAO) and d-aspartate oxidase (DDO) are degradative enzymes stereospecific for d-amino acids. They have been identified in various organisms, including mammals and the nematode Caenorhabditis elegans, although the significance of these enzymes and the relevant functions of d-amino acids remain to be elucidated. In this study, we investigated the spatiotemporal localization of C. elegans DAO and DDOs (DDO-1, DDO-2, and DDO-3) and measured the levels of several d- and l-amino acids in wild-type C. elegans and four mutants in which each gene for DAO and the DDOs was partially deleted and thereby inactivated. Furthermore, several phenotypes of these mutant strains were characterized. The results reported in this study indicate that C. elegans DAO and DDOs are involved in egg-laying events and the early development of C. elegans. In particular, DDOs appear to play important roles in the development and maturation of germ cells. This work provides novel and useful insights into the physiological functions of these enzymes and d-amino acids in multicellular organisms.

INTRODUCTION

Among the free d-amino acids that are present in living organisms, d-serine (d-Ser) and d-aspartate (d-Asp) have been studied the most intensively. d-Ser was first identified in the early 1960s, in lower animals such as earthworms and silkworms (58, 61). Its concentration in the blood of silkworms was reported to increase during particular stages of metamorphosis, although the physiological role of d-Ser in metamorphosis remains unclear (11). It is also present in the mammalian forebrain, where it persists at high concentrations throughout the life of the animal. d-Ser is now considered a neuromodulator that binds to the glycine-binding site of the N-methyl-d-Asp (NMDA) receptor, a subtype of the l-glutamate (l-Glu) receptor, and potentiates glutamatergic neurotransmission in the central nervous system (51, 60, 68). In fact, astroglia-derived d-Ser has been shown to regulate NMDA receptor-dependent long-term potentiation and/or long-term depression, which are basic processes of learning and memory, in the hypothalamic and hippocampal excitatory synapses (25, 54). In addition, d-Ser is also found in the cerebellum during the early postnatal period, and it was recently reported that d-Ser derived from the Bergmann glia serves as an endogenous ligand for the δ2 l-Glu receptor to regulate long-term depression at synapses between parallel fibers and Purkinje cells in the immature cerebellum but not the mature one (31). These lines of evidence suggest the physiological significance of d-Ser in the regulation of higher brain functions through the l-Glu receptors, and indeed, perturbation of d-Ser levels in the central nervous system has now been implicated in the pathophysiology of various neuropsychiatric disorders, including schizophrenia (4, 23, 24, 71), Alzheimer's disease (28, 70), and amyotrophic lateral sclerosis (59).

Unlike the tissue-specific expression of d-Ser, substantial amounts of free d-Asp are present in a wide variety of tissues and cells in invertebrates and vertebrates, particularly in the central nervous, neuroendocrine, and endocrine systems. d-Asp is proposed to play an important role in regulating developmental processes, hormone secretion, and steroidogenesis (12, 26, 35). For instance, it has been reported that in male lizards, intraperitoneally administered d-Asp is taken up rapidly by the testis, which induces a significant increase in testosterone levels and a significant decrease in 17β-estradiol levels in the testis and plasma (57). Interestingly, a reverse relationship is found between males and females; specifically, in female lizards, exogenous d-Asp induces a significant decrease in testosterone levels in the ovary and plasma, while it enhances follicular production of 17β-estradiol by upregulating the local aromatase activity (2). Similar findings have also been reported in studies of the green frog (16, 56). These observations suggest that d-Asp is an important regulatory molecule of gonads. In addition, it was recently shown that the amount of d-Asp in human seminal plasma and spermatozoa is significantly reduced in oligoasthenoteratospermic and/or azoospermic donors compared with normospermic donors (14). Moreover, in human female patients undergoing in vitro fertilization, the d-Asp content of the preovulatory follicular fluid is lower in older patients than in younger ones (15). This decrease in d-Asp content appears to reflect a decrease in oocyte quality and fertilization competence. Thus, the current data suggest that d-Asp may be involved in the pathophysiology of infertility.

d-Amino acid oxidase (DAO) (also abbreviated DAAO; EC 1.4.3.3) and d-Asp oxidase (DDO) (also abbreviated DASPO; EC 1.4.3.1) are flavin adenine dinucleotide (FAD)-containing flavoproteins that catalyze the oxidative deamination of d-amino acids with oxygen to generate the corresponding 2-oxo acids along with hydrogen peroxide and ammonia (40, 63). DAO displays broad substrate specificity and acts on several neutral and basic d-amino acids, such as d-Ser and d-Ala. Meanwhile, DDO is highly specific for acidic d-amino acids, such as d-Asp and d-Glu, none of which are substrates of DAO. DAO and DDO have been identified in various organisms, and their physiological roles in vivo are being investigated extensively. Mammalian DAO and DDO are believed to regulate endogenous d-Ser and d-Asp levels, respectively, and to mediate the elimination of accumulated exogenous d-amino acids in various organs (34, 55). However, the significance of these enzymes and the relevant functions of d-amino acids are still being elucidated.

The nematode Caenorhabditis elegans is a multicellular model animal whose genome sequence was completed in 1998 (10). It is a soil nematode that is small, simple, and rapidly growing and can easily be raised in the laboratory by using the bacterium Escherichia coli as a food source. Because C. elegans worms are self-fertilizing hermaphrodites and rare males, it is possible to readily grow large quantities of the organism (7). In addition, this organism is transparent at every stage of its life cycle, which allows the use of fluorescent reporter genes, such as green fluorescent protein (GFP), to mark tissues and cells. Altogether, these distinctive advantages of C. elegans make it a useful experimental system for studying the biology of multicellular organisms.

In previous reports, we demonstrated that the C. elegans daao-1 gene encodes a functional DAO, while the ddo-1, ddo-2, and ddo-3 genes encode functional DDOs (DDO-1, DDO-2, and DDO-3, respectively), and that these enzymes have different and distinctive properties (36, 37). Thus, although most organisms are believed to carry only one copy each of DAO and DDO genes, C. elegans contains at least three genes encoding functional DDOs. At present, the significance of DAO and multiple DDOs is unclear, and the relevant functions of d-amino acids in C. elegans remain to be elucidated. To investigate the physiological functions of these enzymes and d-amino acids, we first investigated the localization of DAO and DDOs in C. elegans by using GFP-based gene expression analysis. We then determined the levels of several d- and l-amino acids within the body of wild-type C. elegans worms. In addition, to examine whether C. elegans DAO and DDOs are involved in the metabolism of d-amino acids in vivo, we also determined the levels of several d- and l-amino acids within the body for four mutants in which each of the genes for the DAO and DDOs was partially deleted and thereby inactivated. Furthermore, we also characterized several phenotypes of these mutants and of double and triple mutant strains in which two or three DDO genes were inactivated. The results described in this study show that C. elegans DAO and DDOs (and possibly d-amino acids) are involved in egg-laying events and the early development of C. elegans. In particular, C. elegans DDOs appear to play important roles in the development and maturation of germ cells. This work provides novel and useful insights into the physiological functions of these enzymes and d-amino acids in multicellular organisms.

MATERIALS AND METHODS

Chemicals and animals.

d-Amino acids, proteinogenic l-amino acids, catalase from Aspergillus niger, tryptophanase from Escherichia coli, and bovine serum albumin were purchased from Sigma-Aldrich (St. Louis, MO). o-Phthalaldehyde (OPA), N-acetyl-l-cysteine (NAC), 4-fluoro-7-nitro-2,1,3-benzoxadiazole (NBD-F), FAD, pyridoxal phosphate, reduced glutathione, and 5-fluoro-2′-deoxyuridine were purchased from Wako Pure Chemical Industries (Osaka, Japan). All other chemicals were of the highest grade available and were purchased from commercial sources.

The Bristol strain N2 was obtained from the Caenorhabditis Genetic Center (University of Minnesota, Minneapolis, MN) and was used as the standard wild-type strain. C. elegans daao-1, ddo-1, ddo-2, and ddo-3 deletion alleles (tm3673, tm2112, tm1989, and tm2028, respectively) were kindly provided by the National Bioresource Project for the Experimental Animal Nematode C. elegans (S. Mitani at Tokyo Women's Medical University, Tokyo, Japan) (20). These mutant strains were backcrossed six times before further analysis. The life cycle of C. elegans is comprised of the embryonic stage, four larval stages (L1, L2, L3, and L4), and adulthood. Since the mating efficiencies of hermaphrodites and males are high during the young adult stage, the backcrossing was performed with C. elegans worms at this stage. Unless otherwise noted, C. elegans worms were maintained at 20°C on nematode growth medium (NGM) agar plates seeded with E. coli strain OP50, and double and triple mutant strains were generated using standard genetic techniques (8). The genotypes of the mutant worms were confirmed by PCR, using their genomic DNAs as templates. The primers used were as follows: to check for the daao-1 deletion allele, 5′-CGC GAA TTC ATT AGG GGT AC-3′ (forward primer) and 5′-CTA CGA AAA CGC TGG ATT AC-3′ (reverse primer); to check for the ddo-1 deletion allele, 5′-CCA AGT TGA GAT GCA GTT TCG-3′ (forward primer) and 5′-CGA GAA ATT CTG GAT TTG TTC CCC-3′ (reverse primer); to check for the ddo-2 deletion allele, 5′-TGT TCC CCA ACC CAA CGT GA-3′ (forward primer) and 5′-TAT TCA CTC CCG TTC AGC CA-3′ (reverse primer); and to check for the ddo-3 deletion allele, 5′-CTA AGT GCA TTC CTA TCC CA-3′ (forward primer) and 5′-TGG GAC AAG TCC GTG ATA AG-3′ (reverse primer). These primers were designed to anneal outside the respective deletion sites in the daao-1(tm3673), ddo-1(tm2112), ddo-2(tm1989), and ddo-3(tm2028) mutants, and the gene deletions in the mutants were confirmed by analyzing the sizes of the PCR products on agarose gel electrophoresis. The sizes of the PCR products were 469 bp for the daao-1 deletion allele (versus 656 bp for the wild type), 337 bp for the ddo-1 deletion allele (versus 1,212 bp for the wild type), 350 bp for the ddo-2 deletion allele (versus 1,772 bp for the wild type), and 852 bp for the ddo-3 deletion allele (versus 1,427 bp for the wild type).

Construction of reporter plasmids.

A YAC clone, Y69A2AR, and a cosmid clone, C47A10, were kindly provided by A. Fraser (the Wellcome Trust Sanger Institute, Hinxton, Cambridge, United Kingdom). The promoterless GFP reporter vector pPD95.67 was kindly provided by A. Fire (Stanford University, Stanford, CA).

To generate daao-1 promoter-, ddo-1 promoter-, ddo-2 promoter-, and ddo-3 promoter-GFP fusion constructs (Pdaao-1::GFP, Pddo-1::GFP, Pddo-2::GFP, and Pddo-3::GFP, respectively), approximately 2.0-, 3.3-, 2.6-, and 2.4-kb genomic regions upstream of the daao-1, ddo-1, ddo-2, and ddo-3 initiation codons, respectively, and the adjacent first exons were amplified by PCR. The templates used were as follows: Y69A2AR for Pdaao-1::GFP, C47A10 for Pddo-1::GFP, and C. elegans genomic DNA for Pddo-2::GFP and Pddo-3::GFP. The primers used were as follows: for Pdaao-1::GFP, 5′-CTG CAG GAT CTT CCG TAT CCG-3′ (forward primer) and 5′-GAG CTC GGC ATT TCT GAA AAA TAT AG-3′ (reverse primer); for Pddo-1::GFP, 5′-CTG CAG CAA CAT CAT CTG ATC-3′ (forward primer) and 5′-GAG CTC GTC ATT TTT GAG GAA GAA C-3′ (reverse primer); for Pddo-2::GFP, 5′-CTG CAG CAA GAA GAA GAA GAA CGG AG-3′ (forward primer) and 5′-GAG CTC GCC ATT TTA TTC TG-3′ (reverse primer); and for Pddo-3::GFP, 5′-CTG CAG CCT TGA ACT ATT G-3′ (forward primer) and 5′-GAG CTC AGC ATA CCA CCT GCG-3′ (reverse primer). These primers were designed to contain additional PstI and SacI sites at the 5′ ends of the forward and reverse primers, respectively. In the case of the reaction for Pddo-2::GFP, the nested PCR method was employed, using the following first pair of primers: 5′-ATT GAG GAG TGC ATG TGT GCT CCA C-3′ (forward primer) and 5′-AAT CCT GCT GGT CCT GCA CTG CAC G-3′ (reverse primer). The PCR products were cloned into the pT7Blue vector (Novagen, Madison, WI), which generated pT7-Pdaao-1, pT7-Pddo-1, pT7-Pddo-2, and pT7-Pddo-3, and the sequences were confirmed. Subsequently, the PstI-SacI fragments containing the daao-1, ddo-1, ddo-2, and ddo-3 promoter regions of pT7-Pdaao-1, pT7-Pddo-1, pT7-Pddo-2, and pT7-Pddo-3, respectively, were subcloned into pPD95.67, resulting in the Pdaao-1::GFP, Pddo-1::GFP, Pddo-2::GFP, and Pddo-3::GFP reporter plasmids (pPD-Pdaao-1-GFP, pPD-Pddo-1-GFP, pPD-Pddo-2-GFP, and pPD-Pddo-3-GFP), respectively.

Preparation of transgenic worms.

Transgenic worms were established as described previously (29), with the following modifications. Briefly, pPD-Pdaao-1-GFP, pPD-Pddo-1-GFP, pPD-Pddo-2-GFP, and pPD-Pddo-3-GFP were injected at 50 ng/μl, along with 50 ng/μl pRL4 [rol-6(su1006)] as the injection marker, into the gonads of young adult hermaphrodites of the wild-type strain, according to the method of Mello et al. (47). Transgenic worms were identified by roller phenotype, and stable lines were isolated. The proportions of F1 and F2 transformants that produced heritably transformed lines were approximately 15% and 95%, respectively, and experiments were performed with descendants from the F3 generation that exhibited the roller phenotype. Patterns of GFP expression in transgenic worms were observed with an LSM510 confocal microscope or an Axio Imager.M1 fluorescence microscope (Zeiss, Jena, Germany).

Preparation of DNA and transgenic worms for rescue experiments.

To construct the DDO-2 expression plasmid for rescue experiments, DDO-2 cDNA was amplified by a PCR using the plasmid pT7-F18Ep, in which a cDNA fragment corresponding to the entire DDO-2 coding sequence was cloned into the pT7Blue vector, as a template (37). The primers used were as follows: 5′-TCT AGA GCA AAC ATA ATT CCG AAG ATT GC-3′ (forward primer) and 5′-GTC GAC TTA TAA TCC TAG TGC AGT CTT AAC-3′ (reverse primer). These primers were designed to contain additional XbaI and SalI sites at the 5′ ends of the forward and reverse primers, respectively. The PCR product was cloned into the pT7Blue vector (pT7-F18Ep-2), and its sequence was confirmed. Subsequently, a 1.0-kb XbaI-SalI fragment containing the entire DDO-2 coding sequence was subcloned into pT7-Pddo-2, resulting in a DDO-2 expression plasmid under the control of the native promoter (pT7-Pddo-2-DDO-2).

To construct the DDO-3 expression plasmid for rescue experiments, DDO-3 cDNA was amplified by a PCR using the plasmid pT7-F20Hp, in which a cDNA fragment corresponding to the entire DDO-3 coding sequence was cloned into the pT7Blue vector, as a template (37). The primers used were as follows: 5′-GGT ACC CTG TAT GCT CTT CTT CTC CTC-3′ (forward primer) and 5′-GGT ACC CTA ATC ATC AAG ATA TTT AAC CC-3′ (reverse primer). These primers were designed to contain additional KpnI sites at the 5′ ends. The PCR product was cloned into the pT7Blue vector (pT7-F20Hp-2), and its sequence was confirmed. Subsequently, a 1.2-kb KpnI-KpnI fragment containing the entire DDO-3 coding sequence was subcloned into pT7-Pddo-3, resulting in a DDO-3 expression plasmid under the control of the native promoter (pT7-Pddo-3-DDO-3).

For rescue experiments, an approximately 6.7-kb genomic region including the daao-1 gene and its promoter was amplified by PCR using C. elegans genomic DNA as a template. The primers used were as follows: 5′-CTG CAG GAT CTT CCG TAT CCG-3′ (forward primer) and 5′-CCA ATT GGC TCC GCC TGT AA-3′ (reverse primer). The PCR product was separated by 0.8% agarose gel electrophoresis and purified with a QIAquick gel extraction kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The purified DNA was injected at 1 to 10 ng/μl, along with 90 to 99 ng/μl pPD-Pdaao-1-GFP as an injection marker, into the gonads of young adult daao-1(tm3673) mutant hermaphrodites. Similarly, C47A10, pT7-Pddo-2-DDO-2, and pT7-Pddo-3-DDO-3 were injected at 10 to 50 ng/μl, along with 50 to 90 ng/μl pPD-Pddo-1-GFP, pPD-Pddo-2-GFP, and pPD-Pddo-3-GFP as injection markers, into the gonads of young adult ddo-1(tm2112), ddo-2(tm1989), and ddo-3(tm2028) mutant hermaphrodites, respectively. Transgenic worms were identified by the expression of GFP, and stable lines were isolated.

Determination of amino acid contents.

Synchronized worms were collected at specific stages according to standard methods (41). The collected worms were sonicated in 10 volumes of 10 mM phosphate-buffered saline (pH 7.4) containing protease inhibitors (Roche Applied Science, Mannheim, Germany) by use of a model 250 Sonifier (Branson Ultrasonics Co., Danbury, CT). The lysates were centrifuged twice at 20,000 × g for 5 min at 4°C. The supernatant was stored at −80°C until the analysis of amino acids was performed.

The concentrations of several d-amino acids (d-Ser, d-Ala, d-His, d-Thr, d-Arg, d-Tyr, d-Val, d-Trp, d-Phe, and d-Leu) and l-amino acids (l-Ser, l-Ala, l-His, l-Arg, l-Tyr, l-Met, l-Trp, l-Phe, l-Ile, and l-Leu) in worm samples were determined by high-performance liquid chromatography (HPLC) using the OPA precolumn derivatization technique as described by Nimura and Kinoshita (50). To an aliquot (150 μl) of worm lysate prepared as described above, 10 μl of 100% (wt/vol) trichloroacetic acid was added, and the mixture was centrifuged at 20,000 × g for 10 min at 4°C to remove precipitated proteins. The supernatant (140 μl) was then mixed with 50 μl of 1 M NaOH and filtered through a 0.45-μm filter (Millex-LH; Millipore, Bedford, MA). Subsequently, 10 μl of the filtered solution was mixed with 30 μl of 100 mM borate buffer (pH 10.2) and 20 μl of OPA-NAC reagent (prepared by mixing 8 mg of OPA with 10 mg of NAC in 1 ml of 100% methanol) to fluorescently derivatize the amino acids in the mixture. After a 2-min incubation at room temperature, an aliquot (10 to 20 μl) of the sample was injected into a Jasco chromatographic system consisting of a model PU-2089 pump, a model FP-2025 fluorescence detector, and a model 807-IT integrator (Jasco Corp., Tokyo, Japan). The sample was separated in a TSK gel ODS-100Z column (250 mm × 4.6-mm internal diameter [ID]; Tosoh Co., Tokyo, Japan) at a flow rate of 1 ml/min, with a gradient consisting of solution A (50 mM sodium acetate buffer, pH 5.3) and solution B (100% methanol) as follows: 0 to 16 min, 0 to 20% solution B; 16 to 22 min, 20% solution B; 22 to 40 min, 20 to 40% solution B; 40 to 46 min, 40% solution B; and 46 to 60 min, 60% solution B. Fluorescence was detected at an excitation wavelength of 350 nm and an emission wavelength of 445 nm. Under these conditions, d-Ser, d-Ala, and l-Trp were eluted with retention times of 21.2, 39.3, and 54.4 min, respectively (Fig. 1A). To verify that the peaks at 21.2 and 39.3 min were for d-amino acids, recombinant C. elegans DAO was used as follows. The recombinant DAO was expressed in E. coli cells and purified to near homogeneity as described previously (37). An aliquot (30 μl) of the worm lysate was mixed with 60 μl of 100 mM sodium pyrophosphate buffer (pH 8.3), 9 μl of 1 mM FAD, 1 μl (5 μg) of A. niger catalase, and 50 μl (10 μg) of purified recombinant C. elegans DAO, and the mixture was incubated at 37°C for 6 h. Subsequently, 10 μl of 100% (wt/vol) trichloroacetic acid was added, and the mixture was centrifuged at 20,000 × g for 10 min at 4°C. The supernatant (140 μl) was then mixed with 50 μl of 1 M NaOH and filtered through a 0.45-μm filter. The filtered solution was used for HPLC analysis following the OPA-NAC derivatization method described above. This treatment resulted in an almost complete disappearance of the peaks at 21.2 and 39.3 min, confirming the identity of d-Ser and d-Ala in the samples (Fig. 1B). Similarly, E. coli tryptophanase (EC 4.1.99.1) was used to verify that the peak at 54.4 min was in fact l-Trp. The following procedure for the assay of tryptophanase is based on the method of Newton and Snell (49). Specifically, 100 μl (1 mg) of E. coli tryptophanase was mixed with 40 μl of 1 M potassium phosphate buffer (pH 7.8), 20 μl of 1 mM pyridoxal phosphate, 10 μl of 5 mM reduced glutathione, 25 μl (250 μg) of bovine serum albumin, and 105 μl of H₂O, and the mixture was layered with 800 μl of toluene. After a 10-min preincubation at 37°C, 100 μl of the worm lysate was added, followed by incubation at 37°C for 6 h. Subsequently, 150 μl of the aqueous layer was mixed with 10 μl of 100% (wt/vol) trichloroacetic acid, and the mixture was centrifuged at 20,000 × g for 10 min at 4°C. The supernatant (120 μl) was then mixed with 50 μl of 1 M NaOH and filtered through a 0.45-μm filter. The filtered solution was used for HPLC analysis following the OPA-NAC derivatization method described above. This treatment also resulted in an almost complete disappearance of the peak at 54.4 min, confirming the presence of l-Trp in the samples (Fig. 1C). The amounts of d- and l-enantiomers of each amino acid were determined based on the peak areas in the chromatograms.

Fig 1.

Typical example of measurement of several amino acids in worm samples by HPLC using the OPA precolumn derivatization technique. (A) Chromatogram of the derivatives formed from standard d- and l-amino acids and OPA-NAC. Thirty picomoles of each of the derivatives was injected. Peaks: 1, d-Asp; 2, l-Asp; 3, d-Asn + l-Asn; 4, d-Ser; 5, l-Ser; 6, d-Gln + l-Gln; 7, d-Glu + l-Glu; 8, d-His; 9, l-His; 10, d-Thr; 11, l-Thr + Gly; 12, l-Arg; 13, d-Arg; 14, d-Ala; 15, l-Ala; 16, l-Tyr; 17, d-Tyr; 18, d-Met + l-Val; 19, l-Met; 20, l-Trp; 21, d-Val; 22, d-Trp; 23, d-Phe; 24, l-Phe; 25, l-Ile; 26, d-Ile + d-Lys + l-Lys; 27, d-Leu; 28, l-Leu. For details of the analytical conditions, see Materials and Methods. (B and C) Chromatograms of worm samples prepared from wild-type C. elegans at the young adult stage. The analytical conditions and peak numbers are the same as in panel A. The peaks corresponding to d-Ser and d-Ala were detected (B, left panel). In contrast, when the same sample was treated with DAO, these peaks almost completely disappeared (B, right panel), confirming the identities of d-Ser and d-Ala in the sample. Likewise, the peak corresponding to l-Trp was detected (C, left panel), and this peak almost completely disappeared when the same sample was treated with tryptophanase (C, right panel), confirming the presence of l-Trp in the sample.

For the determination of d-Asp, l-Asp, d-Glu, and l-Glu contents in worm samples, an aliquot (150 μl) of the worm lysate was mixed with 10 μl of 100% (wt/vol) trichloroacetic acid, and the mixture was centrifuged at 20,000 × g for 10 min at 4°C. The supernatant (140 μl) was then mixed with 50 μl of 1 M NaOH, and an appropriate amount (25 to 50 μl) of the mixture was added to a solution consisting of 100 mM borate buffer (pH 8.5) in a final volume of 200 μl. Subsequently, the amino acids in the mixture (60 μl) were fluorescently derivatized by the addition of 50 μl of 10 mM NBD-F in dry acetonitrile, followed by incubation for 5 min at 60°C. The reaction was terminated by the addition of 390 μl of 2% (wt/vol) trifluoroacetic acid. The sample was filtered through a 0.45-μm filter and applied to a column-switching HPLC system for the determination of the d-Asp and l-Asp contents as described previously (43). d- and l-Glu contents were determined similarly by adjusting the column-switching time of this system. d-Asp was eluted with a retention time of 14.1 min by enantioseparation with a Sumichiral OA-3200 chiral stationary-phase column (250 mm × 4.6-mm ID; Sumika Chemical Analysis Service, Osaka, Japan), and d-Glu was eluted at 26.5 min with an OA-3100 column (250 mm × 4.6-mm ID; Sumika Chemical Analysis Service). To verify that these peaks were indeed d-amino acids, recombinant C. elegans DDO-1 was used as follows. Recombinant DDO-1 was expressed in E. coli cells and purified to near homogeneity as described previously (36). An aliquot (30 μl) of worm lysate was mixed with 60 μl of 100 mM sodium pyrophosphate buffer (pH 8.3), 9 μl of 1 mM FAD, 1 μl (5 μg) of A. niger catalase, and 50 μl (19 μg) of purified recombinant C. elegans DDO-1, and the mixture was incubated at 37°C for 6 h. Subsequently, 10 μl of 100% (wt/vol) trichloroacetic acid was added, and the mixture was centrifuged at 20,000 × g for 10 min at 4°C. The supernatant (140 μl) was then mixed with 50 μl of 1 M NaOH, and an appropriate amount (25 to 50 μl) of the mixture was added to a solution consisting of 100 mM borate buffer (pH 8.5) in a final volume of 200 μl. An aliquot (60 μl) of the mixture was used for HPLC analysis following the NBD derivatization method described above. This treatment resulted in the almost complete disappearance of the peaks at 14.1 and 26.5 min, confirming the identity of d-Asp and d-Glu in the samples. The amounts of d- and l-enantiomers of each amino acid were determined based on the peak areas in the chromatograms.

For determination of the amino acid levels in E. coli strain OP50, bacterial cells cultured in Luria-Bertani medium containing streptomycin (200 μg/ml) were pelleted by centrifugation at 10,000 × g for 10 min at 4°C. The collected cells were washed twice with 10 mM phosphate-buffered saline (pH 7.4) and sonicated in the same buffer containing protease inhibitors by use of a model 250 Sonifier. Subsequently, the lysates were centrifuged twice at 20,000 × g for 5 min at 4°C, and the supernatant was stored at −80°C until analysis of the amino acid contents. The analytical procedures were the same as those used for the worm samples.

Phenotypic analyses.

C. elegans worms were grown at 20°C under ideal, nonstarving conditions, and their body lengths, life spans, growth rates, brood sizes, hatching rates, and fertilization rates were measured as follows. In all of the phenotypic analyses, lost and bagged worms (approximately 0 to 21%) were excluded from analyses. For measurement of body length, the fertilized eggs were incubated on a fresh 35-mm plate at 20°C for 4 days, during which all embryos grew to adult worms. Subsequently, the plate was incubated for 30 min at 4°C in order to paralyze the worms, and micrographs were taken on a stereoscopic microscope equipped with a cooled charge-coupled device (CCD) camera (Nikon, Tokyo, Japan) at a magnification of ×110. The body lengths of the worms were determined using ImageJ 1.4 software (http://rsb.info.nih.gov/ij/).

For measurement of the life span, 10 adult worms were transferred onto a fresh 35-mm plate and incubated at 20°C. After 2 to 6 h of incubation, the adult worms were removed from the plate, on which they had laid fertilized eggs. This time point was regarded as time zero. The plate was incubated at 20°C for 3 days, after which 50 μl of 0.5-mg/ml 5-fluoro-2′-deoxyuridine was added to the plate to prevent self-fertilization. Subsequently, the plate was incubated for an additional day at 20°C, and 40 to 100 adult worms were transferred onto fresh plates containing 5-fluoro-2′-deoxyuridine (20 worms per plate). The plates were incubated at 20°C and checked every day or two. The fraction of live worms was scored on the basis of body movement in response to a touch stimulus.

For measurement of the growth rate, 10 adult worms were transferred onto a fresh 35-mm plate and incubated at 20°C. After a 30-min incubation, the adult worms were removed from the plate, on which they had laid approximately 20 fertilized eggs. This time point was regarded as time zero. The plate was continually incubated at 20°C and checked every hour, and the incubation period required for the embryos to hatch was determined. Subsequently, the progeny worms were transferred onto fresh 35-mm plates (one worm per plate) when they reached the L3 or L4 stage. The plates were incubated at 20°C and checked every hour, and the incubation period required for the progeny worms to grow to gravid adults was determined. (The worms were defined as gravid adults when they laid more than one fertilized egg.)

For measurement of brood size and hatching rate, 10 to 30 late L4 larvae were identified by a white crescent with a tiny black dot in the presumptive vulval region and transferred onto fresh 35-mm plates (one worm per plate). Subsequently, the plates were incubated at the designated temperature (20°C or 25°C). The worms were repeatedly transferred to fresh plates every 12 to 24 h until they stopped laying fertilized eggs. The progeny worms were counted when they reached L3 larvae to young adults. The numbers of eggs remaining on the plates were also counted, and these eggs were defined as having an embryonic lethal phenotype; unfertilized oocytes were not counted. The sum of the number of progeny worms and eggs was defined as the brood size. The proportion of progeny worms in the brood size was defined as the hatching rate.

For measurement of the fertilization rate, 10 to 30 late L4 larvae were transferred onto fresh 35-mm plates (one worm per plate), and the plates were incubated at 25°C for 24 h, during which all the late L4 larvae grew to adult worms. Subsequently, the worms were transferred onto fresh 35-mm plates and incubated at 25°C. After 6 h of incubation, the adult worms were removed from the plate, on which they had laid fertilized eggs and/or unfertilized oocytes. The numbers of eggs and oocytes remaining on the plates were counted. The proportion of eggs in the sum of the numbers of eggs and oocytes was defined as the fertilization rate.

RESULTS

Spatiotemporal distribution of C. elegans DAO and DDOs.

To determine the expression and localization patterns of C. elegans DAO and DDOs during development, expression analyses with hermaphroditic transgenic worms were performed. These transgenic worms were created with transcriptional fusion genes in which the daao-1, ddo-1, ddo-2, and ddo-3 promoter regions were fused upstream of the GFP reporter gene (Pdaao-1::GFP, Pddo-1::GFP, Pddo-2::GFP, and Pddo-3::GFP, respectively). Experiments were performed with multiple transgenic lines for each fusion construct. Independent transgenic lines carrying the same construct yielded qualitatively identical expression patterns, and typical results are presented below.

In transgenic worms expressing the Pdaao-1::GFP construct, GFP expression was first observed in the intestinal cells of gastrula-stage embryos and continued throughout all larval and adult stages (Fig. 2). In worms that expressed the Pddo-1::GFP construct, GFP expression was first observed in the intestinal cells of the 3-fold-stage embryo and continued throughout all larval and adult stages (Fig. 3). GFP expression was also observed in the hypodermis and in unidentified cells in the head at all larval and adult stages. In transgenic worms expressing the Pddo-2::GFP construct, GFP expression was first apparent in the intestinal cells of comma-stage embryos and continued throughout all larval and adult stages (Fig. 4). GFP expression was also observed in the pharyngeal muscles and body wall muscles at all larval and adult stages. In contrast to these transgenic worms, GFP expression was not observed in the intestines of transgenic worms expressing the Pddo-3::GFP construct. In these worms, GFP expression appeared in probable head mesodermal cells and unidentified cells in the head during the L3 to adult stages, in vulval muscles during the L4 to adult stages, and in the hypodermis and gonad sheath cells only during the adult stage (Fig. 5). In the gonad sheath cells, the GFP signal in the U-shaped region was much brighter than that in the other regions. This GFP expression profile in transgenic worms expressing the Pddo-3::GFP construct is in agreement with previously reported results of GFP-based high-throughput expression analyses (27, 46).

Fig 2.

Spatiotemporal distribution of C. elegans DAO in hermaphroditic worms. Differential interference contrast (DIC) (A, D, G, and J), GFP (B, E, H, and K), and merged (C, F, I, and L) images of transgenic worms expressing the Pdaao-1::GFP construct at the embryo (A to C) and adult (D to L) stages are shown. For details, see the text. Bars, 20 μm and 100 μm for panels A and D, respectively.

Fig 3.

Spatiotemporal distribution of C. elegans DDO-1 in hermaphroditic worms. DIC (A, D, G, and J), GFP (B, E, H, and K), and merged (C, F, I, and L) images of transgenic worms expressing the Pddo-1::GFP construct at the embryo (A to C) and adult (D to L) stages are shown. For details, see the text. Bars, 20 μm and 100 μm for panels A and D, respectively.

Fig 4.

Spatiotemporal distribution of C. elegans DDO-2 in hermaphroditic worms. DIC (A, D, G, and J), GFP (B, E, H, and K), and merged (C, F, I, and L) images of transgenic worms expressing the Pddo-2::GFP construct at the embryo (A to C) and adult (D to L) stages are shown. For details, see the text. Bars, 20 μm and 100 μm for panels A and D, respectively.

Fig 5.

Spatiotemporal distribution of C. elegans DDO-3 in hermaphroditic worms. DIC (A, D, G, J, and M), GFP (B, E, H, K, and N), and merged (C, F, I, L, and O) images of transgenic worms expressing the Pddo-3::GFP construct at the L3 larva (A to C), L4 larva (D to F), and adult (G to O) stages are shown. For details, see the text. Bar, 100 μm.

Spatiotemporal expression analyses were also carried out with male transgenic worms. In transgenic male worms expressing the Pdaao-1::GFP, Pddo-1::GFP, or Pddo-2::GFP construct, no significant difference in expression or localization pattern was apparent compared to that in hermaphrodites (Fig. 6A to R). On the other hand, in male worms expressing the Pddo-3::GFP construct, GFP expression was observed in probable head mesodermal cells and unidentified cells in the head during the L3 to adult stages and in hypodermis, seminal vesicle, and tail cells during the adult stage only (Fig. 6S to X). Thus, a sex-specific difference in DDO-3 expression and localization patterns was observed. Taken together, the spatiotemporal distribution patterns of C. elegans DAO and DDOs in the body of C. elegans were different from one another, suggesting that these enzymes play distinct roles in the tissues and cells where they are expressed. In addition, these results suggest that the three C. elegans DDOs (DDO-1, DDO-2, and DDO-3) are tissue- and/or sex-specific isoforms that have independent functions in C. elegans.

Fig 6.

Spatiotemporal distribution of C. elegans DAO, DDO-1, DDO-2, and DDO-3 in male worms. DIC (A, D, G, J, M, P, S, and V), GFP (B, E, H, K, N, Q, T, and W), and merged (C, F, I, L, O, R, U, and X) images of transgenic worms expressing the Pdaao-1::GFP construct (A to F), the Pddo-1::GFP construct (G to L), the Pddo-2::GFP construct (M to R), or the Pddo-3::GFP construct (S to X) at the adult stage are shown. For details, see the text. Bar, 100 μm.

Amino acid contents in C. elegans.

The concentrations of several amino acids in whole-worm homogenates were determined by HPLC as described in Materials and Methods, and some example profiles of chromatograms are shown in Fig. 1. d-Enantiomers of Ser, Ala, Asp, and Glu were detected in the homogenates, and their levels at specific stages (egg, L1 larva, L3 larva, and young adult) are listed in Table 1. In contrast to these d-amino acids, which were detected in the homogenates at all developmental stages tested, other d-amino acids (d-His, d-Thr, d-Arg, d-Tyr, d-Val, d-Trp, d-Phe, and d-Leu) were not detected. d-Ser contents were relatively low at the egg and L1 stages and markedly increased at later stages (L3 larva and young adult). The proportion of d-Ser in total Ser {d% = [d-amino acid level/(d- + l-amino acid levels)] × 100%} was 1.8%, 1.1%, 1.2%, and 2.9% in the egg, L1 larva, L3 larva, and young adult stages, respectively (Table 1). The similar percentages at these four developmental stages were due to concomitant increases in l-Ser levels at the same time points (the contents of several l-amino acids are listed in Table 2). Similar to the d-Ser contents, d-Ala concentrations were relatively low in the egg and L1 stages and increased thereafter (Table 1). Similar ontogenic changes were also observed in d-Asp levels (Table 1). The d% for d-Ala and d-Asp showed negligible changes throughout all developmental stages tested. On the other hand, d-Glu levels were relatively low in the egg through L3 stages and then markedly increased during the young adult stage (Table 1). In addition, the d% of d-Glu was also relatively higher during the young adult stage than during other stages.

Table 1.

d-Amino acid levels in wild-type C. elegans

Amino acid	Level (fmol/worm)a
	Egg	L1 larva	L3 larva	Young adult
d-Ser	1.13 ± 0.46 (1.8)	1.18 ± 0.59 (1.1)	12.0 ± 5.3 (1.2)	212 ± 92 (2.9)
d-Ala	0.66 ± 0.09 (0.28)	1.06 ± 0.33 (0.22)	22.4 ± 7.3 (0.45)	270 ± 42 (0.36)
d-Asp	0.07 ± 0.01 (0.79)	0.20 ± 0.11 (0.69)	1.73 ± 0.78 (0.63)	14.4 ± 5.1 (0.86)
d-Glu	0.36 ± 0.12 (0.12)	0.13 ± 0.04 (0.08)	0.54 ± 0.09 (0.05)	58.6 ± 9.0 (0.42)

^aData are shown as means ± SD for three independent assays. Values in parentheses are the proportions of d-amino acids [d-amino acid level/(d- + l-amino acid levels) × 100%].

Table 2.

l-Amino acid levels in wild-type C. elegans

Amino acid	Level (pmol/worm)a
	Egg	L1 larva	L3 larva	Young adult
l-Ser	0.062 ± 0.020	0.11 ± 0.06	0.99 ± 0.31	8.02 ± 2.27
l-Ala	0.26 ± 0.08	0.48 ± 0.16	5.42 ± 2.68	74.6 ± 16.2
l-Asp	0.008 ± 0.001	0.032 ± 0.012	0.27 ± 0.05	1.69 ± 0.26
l-Glu	0.29 ± 0.10	0.18 ± 0.06	1.10 ± 0.34	14.3 ± 3.7
l-His	0.066 ± 0.014	0.22 ± 0.04	1.40 ± 0.48	20.1 ± 3.1
l-Arg	0.096 ± 0.026	0.19 ± 0.03	1.82 ± 0.67	18.3 ± 3.9
l-Tyr	0.021 ± 0.004	0.051 ± 0.013	1.16 ± 0.30	7.63 ± 1.67
l-Met	0.004 ± 0.001	0.011 ± 0.002	0.56 ± 0.19	2.43 ± 0.94
l-Trp	0.011 ± 0.001	0.009 ± 0.002	0.22 ± 0.05	1.21 ± 0.24
l-Phe	0.033 ± 0.006	0.042 ± 0.011	1.31 ± 0.37	8.01 ± 0.84
l-Ile	0.030 ± 0.006	0.027 ± 0.006	2.21 ± 0.65	12.8 ± 3.6
l-Leu	0.035 ± 0.008	0.036 ± 0.008	2.36 ± 0.63	11.8 ± 2.6

^aData are shown as means ± SD for three independent assays.

Subsequently, the concentrations of several d- and l-amino acids were determined in daao-1, ddo-1, ddo-2, and ddo-3 deletion mutants (tm3673, tm2112, tm1989, and tm2028, respectively) at specific stages (egg and young adult) to determine whether C. elegans DAO and DDOs are involved in the metabolism of d-amino acids in vivo. These mutant strains were isolated by PCR-based deletion screening of trimethylpsoralen-UV-mutagenized libraries (20), and the position and extent of the deletion and/or insertion sites are depicted in Fig. 7. d-Ser contents in all mutant worms were approximately equal to those in wild-type C. elegans at both the egg and young adult stages (Fig. 8A). On the other hand, the d-Ala level in the daao-1 mutant was slightly but significantly higher than that in the wild type during the egg stage (Fig. 8B). The d-Ala content in the daao-1 mutant at the young adult stage also appeared to be higher than that in the wild type, but this difference was not significant. Moreover, d-Asp and d-Glu contents in the ddo-1 and ddo-3 mutants, respectively, were markedly higher than those in wild-type C. elegans at both the egg and young adult stages (Fig. 8C and D). The levels of l-Ser, l-Ala, l-Asp, and l-Glu in all mutant worms were similar to those in the wild type at both the egg and young adult stages (Fig. 9A to D). Collectively, these results strongly suggest that DAO, DDO-1, and DDO-3 are responsible for the metabolism of d-Ala, d-Asp, and d-Glu, respectively, in C. elegans. No d-amino acids other than d-Ser, d-Ala, d-Asp, and d-Glu were detected in any of the mutant worms. Interestingly, the l-Trp level in the ddo-3 mutant was significantly higher than that in wild-type C. elegans at the young adult stage (Fig. 9I). The l-Trp content in the ddo-3 mutant at the egg stage also appeared to be higher than that in the wild type, but this difference was not significant. The levels of other l-amino acids in all mutants were similar to those in wild-type C. elegans at both the egg and young adult stages (Fig. 9E to H and J to L).

Fig 7.

Schematic structures of daao-1, ddo-1, ddo-2, and ddo-3 genes. Dark gray boxes indicate coding exons, while open boxes indicate untranslated regions. The positions of the initiation (ATG) codons and the stop (TGA, TAA, or TAG) codons are shown. The positions and extents of the deletion and/or insertion sites in daao-1(tm3673), ddo-1(tm2112), ddo-2(tm1989), and ddo-3(tm2028) are indicated by horizontal solid lines.

Fig 8.

d-Amino acid levels in whole-worm homogenates. The levels of d-Ser (A), d-Ala (B), d-Asp (C), and d-Glu (D) in the wild-type, daao-1 mutant, ddo-1 mutant, ddo-2 mutant, and ddo-3 mutant strains were determined in the egg and young adult stages. Data are shown as means ± standard deviations (SD) (n = 3). The single and double asterisks indicate significant differences (P < 0.05 and P < 0.01, respectively) compared to the wild type, based on the Dunnett multiple comparison test.

Fig 9.

l-Amino acid levels in whole-worm homogenates. The levels of l-Ser (A), l-Ala (B), l-Asp (C), l-Glu (D), l-His (E), l-Arg (F), l-Tyr (G), l-Met (H), l-Trp (I), l-Phe (J), l-Ile (K), and l-Leu (L) in the wild-type, daao-1 mutant, ddo-1 mutant, ddo-2 mutant, and ddo-3 mutant strains were determined in the egg and young adult stages. Data are shown as means ± SD (n = 3). The double asterisk indicates a significant difference (P < 0.01) compared to the wild type, based on the Dunnett multiple comparison test.

Phenotypes of mutant worms.

The daao-1, ddo-1, ddo-2, and ddo-3 mutant worms did not exhibit any changes in physical appearance under a stereomicroscope, and their body length was equal to that of wild-type C. elegans (approximately 993 to 1,025 μm). To further characterize their phenotypes, the life span, growth rate, brood size, and hatching rate of the mutant worms were assessed and compared with those of wild-type C. elegans. The daao-1, ddo-1, and ddo-2 mutant worms did not exhibit any change in life span; the maximum life spans of these mutants and of wild-type C. elegans were in the range of 20 to 27 days, and the mean life span of these mutants was equal to that of wild-type C. elegans (approximately 15.7 to 16.0 days). In contrast, the ddo-3 mutant had a substantially extended life span, with a maximum life span of 39 days (Fig. 10A), and the mean life span of the ddo-3 mutant was significantly longer than that of wild-type C. elegans (21.3 ± 7.0 days [n = 69] versus 16.0 ± 4.1 days [n = 79]; Student's t test P value of <0.01). The extended life span of the ddo-3 mutant was fully reverted by the expression of the ddo-3 transgene (Fig. 10A).

Fig 10.

Effects of the inactivation of genes for DAO and DDOs on the life span and growth rate of C. elegans. (A) Survival curves for the wild type, the ddo-3 mutant, and the ddo-3 mutant injected with the ddo-3 transgene are shown (n = 69 to 87). (B) Incubation periods required for embryos to hatch were determined for wild-type and mutant strains. Data are shown as means ± SD (n = 20 to 24). The double asterisk indicates a significant difference (P < 0.01) compared to the wild type, based on the Dunnett multiple comparison test. (C) Incubation periods required for progeny to grow to gravid adults were determined for the wild-type and mutant strains. Data are shown as means ± SD (n = 18 to 22). The double asterisk indicates a significant difference (P < 0.01) compared to the wild type, based on the Dunnett multiple comparison test.

As shown in Fig. 10B, the incubation periods required for daao-1, ddo-1, and ddo-2 mutant embryos to hatch were relatively equal to that required for wild-type worms. However, the incubation period required for ddo-3 mutant embryos to hatch was significantly longer. Likewise, a significant prolongation of the incubation period required for the progeny worms to grow to gravid adults was also evident for the ddo-3 mutant but not the other mutants (Fig. 10C). The time difference observed between the wild-type worms and the ddo-3 mutant was much longer with respect to growth to adulthood (approximately 4.0 h) than with respect to time to hatch (approximately 1.8 h). These observations suggest that the growth rate of the ddo-3 mutant is delayed throughout the life cycle, resulting in the extended life span described above.

When the brood sizes and hatching rates were examined at 20°C, the daao-1, ddo-1, ddo-2, and ddo-3 mutants all had a significantly decreased egg-laying capacity compared with wild-type C. elegans (Table 3). On the other hand, all worms exhibited similar hatching rates (Table 3). The brood sizes and hatching rates were also analyzed at 25°C. As expected, the brood sizes of all of the mutants were significantly smaller than that of wild-type worms (Table 4). In addition, the hatching rates of all of the mutant worms were significantly decreased at 25°C compared with that of wild-type C. elegans (Table 4), while similar hatching rates were observed at 20°C (Table 3). These phenotypes observed for the ddo-1, ddo-2, and ddo-3 mutants at 25°C were partially or almost completely rescued by the expression of ddo-1, ddo-2, and ddo-3 transgenes, respectively (Fig. 11A and B); however, the rescue experiment with the daao-1 mutant could not be performed because injection of the daao-1 transgene into the daao-1 mutant resulted in unexplained embryonic lethality. Taken together, these results suggest that the C. elegans DAO and DDOs are closely linked to egg-laying events and the early development of C. elegans.

Table 3.

Phenotypic analyses of mutant strains at 20°C^a

Strain	Brood size	Hatching rate (%)
Wild type	337 ± 38	99.5 ± 0.5
daao-1 mutant	276 ± 38b	99.7 ± 0.5
ddo-1 mutant	290 ± 20b	99.2 ± 0.8
ddo-2 mutant	282 ± 24b	99.7 ± 0.4
ddo-3 mutant	213 ± 24b	99.7 ± 0.4
ddo-1; ddo-2 mutant	292 ± 18b	98.9 ± 1.7
ddo-1; ddo-3 mutant	216 ± 29b	99.6 ± 0.3
ddo-2; ddo-3 mutant	203 ± 36b	99.7 ± 0.6
ddo-1; ddo-2; ddo-3 mutant	226 ± 26b	99.7 ± 0.4

^aData are shown as means ± SD (n = 8 to 18).

^bP < 0.01 (Dunnett multiple comparison test) compared to the wild type.

Table 4.

Phenotypic analyses of mutant strains at 25°C^a

Strain	Brood size	Hatching rate (%)	Fertilization rate (%)
Wild type	216 ± 32	96.3 ± 3.5	99.8 ± 0.6
daao-1 mutant	155 ± 28b	81.8 ± 11.4b	97.5 ± 6.4
ddo-1 mutant	121 ± 50b	72.6 ± 14.3b	68.6 ± 19.7b
ddo-2 mutant	147 ± 53b	79.8 ± 14.1b	77.6 ± 21.2b
ddo-3 mutant	106 ± 59b	85.0 ± 9.1c	99.0 ± 2.7
ddo-1; ddo-2 mutant	163 ± 54b	76.2 ± 15.8b	73.1 ± 19.3b
ddo-1; ddo-3 mutant	122 ± 55b	73.4 ± 17.8b	99.8 ± 0.8
ddo-2; ddo-3 mutant	115 ± 35b	82.6 ± 10.1b	99.6 ± 1.1
ddo-1; ddo-2; ddo-3 mutant	95 ± 46b	77.4 ± 12.7b	99.8 ± 0.9

^aData are shown as means ± SD (n = 13 to 29).

^bP < 0.01 (Dunnett multiple comparison test) compared to the wild type.

^cP < 0.05 (Dunnett multiple comparison test) compared to the wild type.

Fig 11.

Rescue experiments for decreased egg-laying capacity, hatching rate, and/or fertilization rate phenotype observed in the ddo-1, ddo-2, or ddo-3 mutant at 25°C. (A) Brood sizes of the indicated strains were measured at 25°C. Data are shown as means ± SD (n = 15 to 22). The single and double asterisks indicate significant differences (P < 0.05 and P < 0.01, respectively) among the strains, based on the Tukey-Kramer multiple comparison test. (B) Hatching rates of the indicated strains were measured at 25°C. Data are shown as means ± SD (n = 15 to 22). The double asterisks indicate significant differences (P < 0.01) among the strains, based on the Tukey-Kramer multiple comparison test, while “NS” indicates that the difference is not significant (P > 0.05). (C) Fertilization rates of the indicated strains were measured at 25°C. Data are shown as means ± SD (n = 8 to 13). The single and double asterisks indicate significant differences (P < 0.05 and P < 0.01, respectively) among the strains, based on the Tukey-Kramer multiple comparison test, while “NS” indicates that the difference is not significant (P > 0.05).

Since the brood sizes of the daao-1, ddo-1, ddo-2, and ddo-3 mutant worms were decreased compared with that of wild-type C. elegans, we addressed the possibility that these mutants might be defective in the capacitation of germ cells. Fertilization rates of the mutant worms were measured at 25°C and compared with that of the wild type. The daao-1 and ddo-3 mutants and wild-type C. elegans exhibited similar fertilization rates (Table 4). In contrast, fertilization rates of the ddo-1 and ddo-2 mutants were significantly decreased compared with that of wild-type C. elegans. These phenotypes were fully rescued by the expression of ddo-1 and ddo-2 transgenes, respectively (Fig. 11C). These lines of evidence suggest that DDO-1 and DDO-2 play an important role in the development and maturation of germ cells in C. elegans.

Phenotypes of double and triple mutant strains.

To further investigate the physiological significance of DDOs in C. elegans, ddo-1; ddo-2, ddo-1; ddo-3, and ddo-2; ddo-3 double mutants and a ddo-1; ddo-2; ddo-3 triple mutant were generated, and their brood sizes and hatching rates were measured. When the phenotypes were examined at 20°C, all double and triple mutant strains had a significantly decreased egg-laying capacity compared with wild-type C. elegans, while the hatching rates were similar between the double and triple mutant strains and wild-type C. elegans (Table 3). At 25°C, the brood sizes of all mutants were significantly smaller than that of wild-type C. elegans, and moreover, the hatching rates of all of the mutants were significantly decreased compared to that of wild-type C. elegans (Table 4). Note that the magnitudes of the decreases in brood sizes and hatching rates of the double and triple mutant strains were similar to those of the single mutant worms. In fact, the differences in brood sizes were not significant among the ddo-1; ddo-2, ddo-1, and ddo-2 mutants or among the ddo-1; ddo-2; ddo-3, ddo-1; ddo-3, ddo-2; ddo-3, and ddo-3 mutants (Tukey-Kramer multiple comparison test; P > 0.05), and the differences in hatching rates were not significant among any of the mutants (Tukey-Kramer multiple comparison test; P > 0.05). Thus, the inactivation of a second and/or third DDO gene did not elicit an additive or synergistic effect on these phenotypes. These results suggest that all three C. elegans DDOs are involved in egg-laying events and the early development of C. elegans, through similar molecular pathways.

When the fertilization rates of the double and triple mutant worms were measured at 25°C, the fertilization rate of the ddo-1; ddo-2 mutant was significantly decreased compared to that of wild-type C. elegans (Table 4). However, the magnitude of this decrease was similar to that observed for the ddo-1 and ddo-2 single mutants. In fact, the differences in fertilization rates were not significant among the ddo-1; ddo-2, ddo-1, and ddo-2 mutants (Tukey-Kramer multiple comparison test; P > 0.05), consistent with the idea that C. elegans DDO-1 and DDO-2 are involved in egg-laying events via similar molecular pathways. Notably, no decrease in the fertilization rate was observed in the ddo-1; ddo-3, ddo-2; ddo-3, and ddo-1; ddo-2; ddo-3 mutant worms (Table 4). Thus, the decrease in fertilization rate caused by the inactivation of the ddo-1 and/or ddo-2 gene was apparently rescued by the inactivation of the ddo-3 gene. These lines of evidence suggest that C. elegans DDO-1, DDO-2, and DDO-3 play important roles in the development and maturation of germ cells; however, the molecular details underlying these phenotypic effects remain to be elucidated.

DISCUSSION

Origin of d-amino acids in C. elegans.

Several d-amino acids are known to be essential components of cell wall peptidoglycans and antibiotic peptides in bacteria (6, 52). Indeed, quantitative analysis of the E. coli strain OP50 that is used as a diet revealed the presence of d-Ala, d-Asp, and d-Glu (approximately 3.5, 4.5, and 2.8 μmol/g protein, respectively), with the d% reaching approximately 22%, 67%, and 57%, respectively. Therefore, among the d-amino acids detected in C. elegans (d-Ser, d-Ala, d-Asp, and d-Glu), d-Ala, d-Asp, and d-Glu can be derived partially from the diet. On the other hand, d-Ser was not detected in E. coli, suggesting that it is biosynthesized in C. elegans. In fact, Ser racemase (EC 5.1.1.16), a synthetic enzyme that produces d-Ser from l-Ser, has been found in various organisms, including yeast Schizosaccharomyces pombe, plants, and mammals (19, 69, 73). Moreover, in the C. elegans database WormBase (http://www.wormbase.org/), the gene T01H8.2 is annotated as encoding a putative Ser racemase. However, the deduced amino acid sequence of the protein encoded by T01H8.2 is more homologous to an Asp-specific amino acid racemase (EC 5.1.1.13) identified in bivalves (Scapharca broughtonii), which converts l-Asp to d-Asp (1), than to Ser racemases. We are currently investigating the enzymatic properties of the protein encoded by the T01H8.2 gene.

C. elegans DAO.

In mammals, DAO and DDO activities are highest in the kidney, followed by the liver and brain, and are low in other peripheral tissues (3, 33, 67, 72). Exogenous d-amino acids derived from enterobacteria and/or the diet can be absorbed by the intestine and transferred via the circulatory system to the kidney and liver, where they are metabolized by DAO or DDO (13). In transgenic worms that expressed the Pdaao-1::GFP construct, GFP expression was observed exclusively in the intestine (Fig. 2). Moreover, d-Ala levels were increased in the daao-1 mutant (Fig. 8B). Therefore, C. elegans DAO appears to be responsible for the degradation of diet-derived d-Ala in the intestine, similar to mammalian DAO in the kidney and liver. Diet-derived d-Ala may be used as an energy source in C. elegans, since it has been reported that yeast Candida boidinii can catabolize d-Ala as a carbon source with DAO (74). DAO-mediated d-amino acid catabolism for energy production was also suggested in a study of a mouse mutant (ddY/DAO⁻) that lacks DAO activity (38). In the C. elegans daao-1 mutant, energy production may be perturbed due to its inability to degrade diet-derived d-Ala, thereby leading to the phenotypes of decreased egg-laying capacity and a decreased hatching rate (Tables 3 and 4), although the molecular details of these phenotypic changes remain to be elucidated.

Mammalian DAO is expressed in several regions of the brain in addition to the kidney and liver, while in C. elegans DAO is expressed exclusively in the intestine. In the mammalian central nervous system, the probable physiological substrate for DAO is d-Ser, which has been proposed to be a neuromodulator that binds to the NMDA receptor as a coagonist and potentiates glutamatergic neurotransmission (51, 60, 68). Among the NMDA receptor subunits identified in mammals, NR1 and NR2A are conserved in C. elegans (NMR-1 and NMR-2, respectively), and these C. elegans orthologues are expressed exclusively in the command interneurons (9, 32). Since d-Ser binds to the NR1 subunit in mammals, it likely binds to the NMR-1 subunit in C. elegans. However, in transgenic worms that expressed the Pdaao-1::GFP construct, GFP expression was not observed in the nervous system (Fig. 2). In addition, no increase in endogenous d-Ser was detected in the daao-1 mutant (Fig. 8A). These results suggest that d-Ser has a unique function in C. elegans and that it is metabolized by an enzyme(s) other than DAO. In fact, d-Ser dehydratases (EC 4.3.1.18) were recently identified in yeast Saccharomyces cerevisiae and chickens (30, 66).

C. elegans DDO-1 and DDO-2.

In transgenic worms that expressed the Pddo-1::GFP construct, GFP expression was observed in the intestine throughout the life of the worm (Fig. 3). Moreover, d-Asp levels were significantly increased in the ddo-1 mutant (Fig. 8C). Thus, C. elegans DDO-1 is presumed to be responsible for the degradation of diet-derived d-Asp in the intestine, similar to mammalian DDO in the kidney and liver. Diet-derived d-Asp is conceivably catabolized as an energy source, similar to d-Ala in C. elegans. In fact, DDO has been reported to play an essential role in the assimilation of d-Asp in yeast Cryptococcus humicola (65). In the C. elegans ddo-1 mutant, energy production may be perturbed due to its inability to degrade diet-derived d-Asp, as in the daao-1 mutant described above. This energy perturbation possibly leads to the phenotypes of decreased egg-laying capacity and a decreased hatching rate (Tables 3 and 4), although the molecular details of these phenotypic changes also remain to be elucidated. It is interesting that in transgenic worms with the Pddo-2::GFP construct, GFP expression was also observed in the intestine throughout the life of the worm (Fig. 4). Nevertheless, d-Asp content was not increased in the ddo-2 mutant (Fig. 8C). Collectively, these findings suggest that the primary enzyme for the degradation of exogenous d-Asp in C. elegans is probably DDO-1 rather than DDO-2. In fact, the recombinant form of C. elegans DDO-1 is more efficient for the deamination of d-Asp than that of DDO-2 (36).

In C. elegans, int5 cells of the intestine interact intimately with the primordial germ line, and the intestine is closely involved in reproduction (45, 48). d-Asp has been detected in human seminal plasma, spermatozoa, and preovulatory follicular fluid (14, 15), and several lines of evidence suggest that d-Asp is closely involved in the quality control of germ cells in mammals (12, 26, 35). The C. elegans ddo-1 mutant exhibited a decreased fertilization rate (Table 4) concomitant with an increase in d-Asp content (Fig. 8C). Thus, it is likely that d-Asp is also involved in the quality control of germ cells in C. elegans. Note that the ddo-2 mutant also exhibited a decreased fertilization rate (Table 4). Thus, C. elegans DDO-2, as well as DDO-1, is likely to be involved in the quality control of germ cells. However, in contrast to the ddo-1 mutant, no significant increase was observed in the d-Asp content of the ddo-2 mutant (Fig. 8C). The most plausible explanation for this discrepancy is that DDO-1 is more active and efficient than DDO-2 for the degradation of d-Asp, as described above (36). The decreased fertilization rates in the ddo-1 and ddo-2 mutants were presumably due to locally increased d-Asp, while the increased concentration of d-Asp in whole-worm homogenates was not appreciable in the ddo-2 mutant, in which the more active enzyme DDO-1 was still intact.

In contrast to the case for the ddo-1 mutant, no significant increase was observed in the d-Asp content of the ddo-2 mutant (Fig. 8C), as described above. Nevertheless, the ddo-2 mutant exhibited phenotypes similar to those of the ddo-1 mutant, namely, both mutants exhibited a decreased egg-laying capacity, hatching rate, and fertilization rate (Tables 3 and 4). Thus, C. elegans DDO-2, as well as DDO-1, is likely to be involved in egg-laying events and the early development of the worm, in addition to the quality control of germ cells. Furthermore, these phenotypic changes appear to be elicited via similar molecular pathways in the ddo-1 and ddo-2 mutants, since no additive or synergistic effect was observed in the phenotypes of the ddo-1; ddo-2 double mutant (Tables 3 and 4). Note, however, that GFP expression in transgenic worms that expressed the Pddo-2::GFP construct was observed in the pharyngeal muscle and body wall muscle as well as in the intestine (Fig. 4). This suggests that C. elegans DDO-2 may be involved in neurotransmission in the peripheral nervous system in the worm. On the other hand, unlike the case for transgenic worms that expressed the Pddo-2::GFP construct, GFP expression in transgenic worms that expressed the Pddo-1::GFP construct was observed in the hypodermis as well as in the intestine (Fig. 3). In C. elegans, the hypodermis surrounds the body of the worm and underlies the collagenous extracellular cuticle (exoskeleton) that is important for maintenance of body morphology and integrity, protection from the external environment, and motility (39, 53). Since the cuticle is synthesized and secreted by the hypodermis, C. elegans DDO-1 may be involved in this process. Determining the localization of d-amino acids in the body of C. elegans will aid in the identification of the physiological substrates of DDOs and the functions of d-amino acids in the worm.

C. elegans DDO-3.

d-Glu content was increased in the ddo-3 mutant (Fig. 8D); therefore, it appears that d-Glu is metabolized primarily by DDO-3 in C. elegans. In fact, it has been shown that recombinant DDO-3, as well as DDO-1 and DDO-2, can catalyze the deamination of d-Glu (36). d-Glu is most likely derived in part from a diet of E. coli, as described above. It is currently unclear whether the diet-derived d-Glu absorbed by the intestine escapes degradation by intestinal DDO-1 and DDO-2. DDO-3 does not appear to be expressed in the intestine, since GFP was not observed in the intestines of transgenic worms that expressed the Pddo-3::GFP construct (Fig. 5 and 6). It remains to be determined whether d-Glu is synthesized in the body of C. elegans, although there is no known gene in the C. elegans genome that is orthologous to a Glu racemase gene (EC 5.1.1.3).

The ddo-3 mutant exhibited a decreased egg-laying capacity and a decreased hatching rate (Tables 3 and 4), similar to the ddo-1 and ddo-2 mutants. Thus, C. elegans DDO-3, in addition to DDO-1 and DDO-2, is likely to be involved in egg-laying events and early development of the worm. Furthermore, these phenotypic changes appear to be elicited via similar molecular pathways in the ddo-1, ddo-2, and ddo-3 mutants, since no additive or synergistic effect was observed in the phenotypes of any of the multiple mutants (Tables 3 and 4). These lines of evidence were somewhat surprising, as the spatiotemporal distributions of these three DDOs in the body of C. elegans were different from one another (Fig. 3 to 6). Moreover, d-amino acid levels also differed among the ddo-1, ddo-2, and ddo-3 mutants (Fig. 8). The molecular pathway through which the inactivation of the ddo-1, ddo-2, and ddo-3 genes results in decreased egg-laying capacity and a decreased hatching rate is currently enigmatic. Identification of this pathway will aid in understanding the molecular details underlying these phenotypic changes. In addition, it will be interesting to examine whether the modification of another gene(s) involved in the identified pathway has an effect on the phenotypes of the ddo-1, ddo-2, and ddo-3 mutants.

In contrast to the ddo-1 and ddo-2 mutants, the fertilization rate was not decreased in the ddo-3 mutant (Table 4). Interestingly, the fertilization rate was also not decreased in the ddo-1; ddo-3, ddo-2; ddo-3, and ddo-1; ddo-2; ddo-3 mutants. Therefore, it appears that the additional inactivation of ddo-3 can rescue the decreased fertilization rate phenotype caused by the inactivation of the ddo-1 and ddo-2 genes. It is noteworthy that DDO-3 is thought to be expressed in sex-specific tissues and cells, as observed in transgenic worms with the Pddo-3::GFP construct. Specifically, GFP signals due to DDO-3 expression were observed in the vulval muscles and gonad sheath cells of hermaphrodites (Fig. 5), while they were observed in the seminal vesicle and tail cells of males (Fig. 6). The gonad sheath cells, which are specialized smooth muscle-like cells that surround oocytes, form gap junctions with oocytes and regulate meiotic maturation and ovulation (21). Importantly, apoptotic cell death of germ cells occurs near the gonad U-shaped region, which is a conserved feature of oogenesis. In C. elegans, developmental germ cell apoptosis is referred to as a physiological process because the sacrificed nuclei do not seem to be of poor quality and because their associated cytoplasm is provided to their surviving sisters (22). Interestingly, it was recently reported that C. elegans CAR-1 is a germ line RNA-binding protein that associates with the RNA helicase CGH-1 within a conserved RNA-protein complex, and its depletion by RNA interference results in an increase in physiological apoptosis, along with a modest decrease in brood size (5). More importantly, this increase in physiological apoptosis partially compensates for an oogenesis defect that otherwise leads rapidly to gonad failure. A similar effect of elevated germ cell apoptosis on an oogenesis defect was also observed in worms in which CPB-3 was depleted by RNA interference (5). CPB-3 is an orthologue of the cytoplasmic polyadenylation element-binding protein and is expressed primarily during oogenesis (44). It was consequently proposed that the physiological germ cell apoptosis pathway facilitates the formation of functional oocytes and may ameliorate particular defects in oogenesis (5). Note that in the gonad sheath cells of hermaphroditic transgenic worms that expressed the Pddo-3::GFP construct, the GFP signal in the U-shaped region was much brighter than that in the other regions (Fig. 5). Thus, C. elegans DDO-3 appears to be involved in oocyte development through the physiological germ cell apoptosis pathway in C. elegans. It is likely that the inactivation of the ddo-3 gene causes an increase in physiological germ cell death, thereby leading to amelioration of the decreased fertilization rate phenotype caused by the inactivation of the ddo-1 and ddo-2 genes, although further studies to examine the frequency of physiological germ cell apoptosis in the ddo-3 mutant are certainly necessary.

DDO-3 is also possibly involved in spermatogenesis in male C. elegans. Significant differences exist between spermatogenesis in hermaphroditic and male worms. For instance, in hermaphrodites, spermatogenesis occurs during a transient period in the L3 stage, followed by the self-fertilization of oocytes produced by switched oogenesis in advanced stages. In male worms, however, spermatogenesis begins in the L3/L4 stage and continues through adulthood. Moreover, spermatids produced in males are significantly larger than those produced in hermaphrodites (42). Notably, GFP expression was observed in the seminal vesicle in male transgenic worms expressing the Pddo-3::GFP construct (Fig. 6). In the seminal vesicle in males, spermatogenesis proceeds and the spermatids produced are stored until ejaculation. In contrast, in hermaphroditic transgenic worms with the Pddo-3::GFP construct, GFP expression was not observed in the spermatheca (Fig. 5), where produced spermatids are stored. Further studies are thus necessary to determine whether C. elegans DDO-3 is a sex-specific isoform that plays unique roles in male spermatogenesis. Mating of ddo-3 mutant males with wild-type hermaphrodites and/or wild-type males with ddo-3 mutant hermaphrodites will aid in understanding the DDO-3 function in C. elegans.

In transgenic worms with the Pddo-3::GFP construct, GFP signals due to DDO-3 expression were observed in the vulval muscles of hermaphrodites and in the tail cells of males, as described above (Fig. 5 and 6). Note that the former expression was observed during the L4 to adult stages, and the latter was observed only during the adult stage. In C. elegans, the vulva and tail are the reproductive organs of hermaphrodites and males, respectively, and their morphological and functional maturation takes place during the late L4 to adult stages (18, 62). Therefore, the above-mentioned temporal and tissue-specific patterns of C. elegans DDO-3 distribution suggest that it is involved in the maturation processes of these reproductive organs in C. elegans. On the other hand, GFP expression was also observed in the hypodermis in both hermaphroditic and male transgenic worms expressing the Pddo-3::GFP construct (Fig. 5 and 6). Thus, C. elegans DDO-3, as well as DDO-1, is likely to be involved in the production of cuticle by the hypodermis. It is noteworthy, however, that the hypodermal GFP expression was observed in the worms with the Pddo-3::GFP construct only during the adult stage, while it was observed in transgenic worms expressing the Pddo-1::GFP construct at all larval and adult stages (Fig. 3). These observations suggest that C. elegans DDO-3 plays a stage-specific role(s) in the hypodermis. In C. elegans, the hypodermis is connected to the cuticle via hemidesmosomes and is also connected to filaments extending from the body wall muscles through the basement membrane (39, 53). It is through these connections that the force of muscle contraction is transmitted to the cuticle, and therefore the hypodermis is critical for the motility of C. elegans. Motor activity assays with the ddo-1 and ddo-3 mutants will aid in understanding the roles of DDO-1 and DDO-3 in the motility of C. elegans.

l-Trp content was significantly increased in the ddo-3 mutant (Fig. 9I). l-Trp is a precursor of serotonin, a neurotransmitter found in many animals, including C. elegans. Tryptophan hydroxylase (EC 1.14.16.4) catalyzes the rate-limiting first step in the serotonin synthetic pathway by which l-Trp is converted to 5-hydroxy-l-Trp and then matured to serotonin by aromatic l-amino acid decarboxylase (EC 4.1.1.28). In addition, l-Trp can also be metabolized to N-formyl-kynurenine in C. elegans, since tryptophan 2,3-dioxygenase (EC 1.13.11.11), which catalyzes the oxidative cleavage of the pyrrole ring of l-Trp to generate N-formyl-kynurenine, is conserved in C. elegans. It is likely that these metabolic pathways are affected in the ddo-3 mutant, thereby leading to the accumulation of l-Trp. It has been reported that the C. elegans tph-1(mg280) mutant, which has a partially deleted and inactivated gene for tryptophan hydroxylase, exhibits several phenotypes, such as slower development and decreased egg-laying capacity, together with a decreased serotonin level (17, 64). Immunostaining of the ddo-3 mutant with an anti-serotonin antibody is now under way in our laboratory to clarify the involvement of DDO-3 in serotonergic neurotransmission in C. elegans.