D-Glutamate is metabolized in the heart mitochondria

D-Amino acids are enantiomers of L-amino acids and have recently been recognized as biomarkers and bioactive substances in mammals, including humans. In the present study, we investigated functions of the novel mammalian mitochondrial protein 9030617O03Rik and showed decreased expression under conditions of heart failure. Genomic sequence analyses showed partial homology with a bacterial aspartate/glutamate/hydantoin racemase. Subsequent determinations of all free amino acid concentrations in 9030617O03Rik-deficient mice showed high accumulations of D-glutamate in heart tissues. This is the first time that a significant amount of D-glutamate was detected in mammalian tissue. Further analysis of D-glutamate metabolism indicated that 9030617O03Rik is a D-glutamate cyclase that converts D-glutamate to 5-oxo-D-proline. Hence, this protein is the first identified enzyme responsible for mammalian D-glutamate metabolism, as confirmed in cloning analyses. These findings suggest that D-glutamate and 5-oxo-D-proline have bioactivities in mammals through the metabolism by D-glutamate cyclase.


SI Materials and Methods
Generation of 9030617O03Rik knockout mice 9030617O03Rik knockout mice were created primarily with Knockout Mouse Project (KOMP) targeted ES cells derived from the C57BL/6 mouse strain. The targeting strategy and cell lines produced were described previously 1 , and the specific alleles manufacturer's instructions were described in the KOMP Repository (www.komp.org) and the project webpage (www.kompphenotype.org). Using this ES cell (D-10), we mated the mice with CAG-Cre mice to make 9030617O03Rik knockout mice.

Construction of 9030617O03Rik expression plasmids
Mouse 9030617O03Rik was amplified using PCR with the forward and reverse primers 5′-GCCTGGTGCCGCGCGGCAGCCATATGAACACTTCCAGCATGACGGA-3′ and 5′-TCGGGCTTTGTTAGCAGCCGGATCCTCACATGTGCACTGTGGTGG-3′, respectively, to replace the ATG codon of 9030617O03Rik with a NdeI restriction site and to add a BamHI restriction site downstream of the Stop codon of 9030617O03Rik. Subsequently, the 9030617O03Rik fragment was reinserted between NdeI and BamHI sites of pET-15b (Novagen) using In-Fusion HD cloning Kits (Clontech).
Expression and purification of recombinant protein 3 Escherichia coli BL21 (DE3) pLysS cells were transformed with the expression plasmid and were cultured at 37°C with shaking in Luria-Bertani medium containing ampicillin (100 µg/mL). Cultures were grown to A620 = 0.5 and were then incubated for an additional 30 min at 30ºC. After addition of 0.1 mM isopropyl-β-D-thiogalactopyranoside, cultures were incubated at 30°C for a further 16 h, and cells were centrifuged at 10,000 × g for 10 min at 4°C. Subsequently, crude extracts were prepared using BugBuster Protein Extraction Reagent and Lysonase Bioprocessing Reagent (Novagen, Madison, WI, USA) in the presence of protease inhibitors (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions.
Recombinant protein was purified using affinity chromatography with a chelating column. Specifically, crude extracts (prepared as described above) were mixed with 1/49 volumes of 20 mM sodium phosphate buffer (pH 7.4) containing 0.5 M NaCl and 500 mM imidazole and were applied to a His GraviTrap column (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA) that had been equilibrated with 20 mM sodium phosphate buffer (pH 7.4) containing 0.5 M NaCl and 10 mM imidazole. The column was then washed with the same buffer, and bound proteins were eluted using 20 mM sodium phosphate buffer (pH 7.4) containing 0.5 M NaCl and 500 mM imidazole. The eluted fraction (6 mL) containing recombinant protein was dialyzed at 4°C for 1 day against 1 4 L of 50 mM Tris-HCl buffer (pH 8.0) containing 5 mM 2-mercaptoethanol and 10% (v/v) glycerol. The dialyzed fraction was mixed with a solution containing 10 mM phosphate buffered saline (pH 7.4) and 10 mM imidazole in a final volume of 50 mL. Subsequently, the mixture was applied to a His GraviTrap column that had been equilibrated as described above. The column was then washed with 20 mM sodium phosphate buffer (pH 7.4) containing 0.5 M NaCl and 50 mM imidazole, and bound proteins were eluted using a stepwise gradient of 100-500 mM imidazole. Fractions (2 mL) containing recombinant protein were dialyzed at 4°C for 1 day against 1 L of 50 mM Tris-HCl buffer (pH 8.0) containing 5 mM 2-mercaptoethanol and 10% (v/v) glycerol. The buffer was changed once during dialysis, and dialyzed fractions were centrifuged at 10,000 × g for 10 min at 4°C to pellet proteins that were denatured during dialysis. Supernatants were recovered as purified enzyme, were mixed with an equal volume of 100% (v/v) glycerol, and were used immediately for enzyme assays or were stored at -20°C until use.
Purification of the recombinant protein to near -homogeneity was confirmed using SDSpolyacrylamide gel electrophoresis, and protein concentrations of purified enzyme preparations were determined using Bio-Rad Protein Assay Kits (Bio-Rad Laboratories, Hercules, CA, USA) with BSA as a standard.
Enzyme activity assays 5 Amino acid racemase activity was assayed by measuring L-aspartate, L-glutamate, Lserine, and L-alanine after formation from respective enantiomers in reaction mixtures. Oxidase and dehydrogenase activities against D-glutamate were determined using a colorimetric method for 2-oxoglutaric acid production as described previously 3 . Briefly, the reaction mixture was prepared as described above and 10 mM D-glutamate was used as the substrate. Mixtures were incubated at 37ºC for 2 h, and 10 μL of 100% (w/v) trichloroacetic acid was added to stop the reaction. The produced 2-oxoglutaric acid was reacted with 2,4-dinitrophenylhydrazine and was quantitated by measuring A445 against a blank mixture that lacked D-glutamate. The limit of quantitation of this method was 0.019.
Glutamine synthetase and decarboxylase activities were determined by measuring Lglutamine, D-glutamine, and/or γ-aminobutyric acid contents of reaction mixtures.
Briefly, reaction mixtures were prepared as described above and 10 mM D-glutamate was 7 used as a substrate. Mixtures were then incubated at 37ºC for 2 h, and 600 μL of 100% (v/v) methanol was added to stop the reaction. Subsequently, mixtures were incubated at -80ºC for 1 h and centrifuged at 20,000 × g for 10 min at 4°C to remove precipitated proteins. Supernatants (600 μL) were then filtered through 0.45 μm Millex-LH filters, and filtrates were appropriately diluted with H2O. Diluents were then applied to highperformance liquid chromatography (HPLC) analyses following OPA/Boc-L-cysteine derivatization as described above.
Glutamate cyclase activity was determined by measuring the reduction of D-or Lglutamate contents in reaction mixtures. Briefly, reaction mixtures were prepared as described above using 10 mM D-or L-glutamate as substrate. Unless otherwise noted, mixtures were incubated at 37ºC for 2 h, and then 600 μL of 100% (v/v) methanol was added to stop the reaction. Subsequently, mixtures were incubated at -80ºC for 1 h and centrifuged at 20,000 × g for 10 min at 4°C to remove precipitated proteins. Supernatants (600 μL) were then filtered through 0.45-μm Millex-LH filters, and filtrates were appropriately diluted with H2O. Diluents were then analyzed using HPLC after OPA/Boc-L-cysteine derivatization as described above.
Glutamate cyclase activity was also determined by measuring the formation of D-or Lglutamate from 5-oxo-D-or 5-oxo-L-proline, respectively. Briefly, reaction mixtures were 8 prepared as described above, and 10 mM 5-oxo-D-or 5-oxo-L-proline were used as substrates. Mixtures were incubated at 37ºC for 2 h, and 600-μL aliquots of 100% (v/v) methanol were added to stop the reactions. Subsequently, mixtures were incubated at -80ºC for 1 h and centrifuged at 20,000 × g for 10 min at 4°C to remove precipitated proteins. Supernatants (600 μL) were then filtered through 0.45 μm Millex-LH filters, and filtrates were appropriately diluted with H2O. Diluents were then analyzed using HPLC after OPA/Boc-L-cysteine derivatization as described above.
D-Glutamate cyclase activities were assayed by measuring the formation of 5-oxo-Dproline from D-glutamate in reaction mixtures. Specifically, reaction mixtures were prepared as described above, and 10 mM D-glutamate was used as the substrate.
Mixtures were incubated at 37ºC for 2 h, and 600 μL of 100% (v/v) methanol was then added to stop the reaction. Subsequently, mixtures were incubated at -80ºC for 1 h and centrifuged at 20,000 × g for 10 min at 4°C to remove precipitated proteins. Supernatants (600 μL) were then filtered through 0.45 μm Millex-LH filters, and 100-μL aliquots of filtrates were evaporated to dryness. Subsequently, 65 μL aliquots of 2-propanol were added to residues to dissolve 5-oxo-D-proline but not D-glutamate. Suspensions were then vigorously vortexed for 20 s, followed by sonication in a water bath for 5 min and vigorous vortexing for another 20 s to extract 5-oxo-D-proline products. Suspensions were then centrifuged at 20,000 × g for 5 min at 4°C, and supernatants (60 μL) were transferred to fresh 1.5 mL microtubes. Original pellets were then vigorously vortexed in 60 μL of 2propanol again for 20 s to further extract 5-oxo-D-proline products. Suspensions were then centrifuged at 20,000 × g for 5 min at 4°C and supernatants (60 μL) were transferred to fresh 1.5 mL microtubes. The extraction procedure was repeated again, and supernatants were combined. A total of 180 μL of recovered supernatant was then filtered through a 0.45 μm Millex-LH filter, and amino acids were derivatized as