Identification and enzymatic properties of arginine decarboxylase from Aspergillus oryzae

The growth dependency of the synthesis of arginine and downstream metabolites in rice koji was examined by mass spectrometry imaging (MSI) (24). We focused on arginine, agmatine, putrescine, spermidine, 4-guanidinobutyric acid, and γ-aminobutyric acid (GABA), for which ionization conditions have been established, allowing identification based on m/z values. As shown in Fig. 2, these compounds were not detected in steamed rice; however, as A. oryzae grew, their accumulation was observed with increasing incubation time. Arginine and spermidine showed accumulation around the edges in the rice koji cultivated for 19 and 27 hours but were detected inside the rice koji cultivated for 45 hours. At 45 hours, where arginine was strongly detected in the rice koji, accumulation of polyamines (putrescine and spermidine) was observed. Agmatine was detected slightly in the peripheral region in the rice koji section. These results suggested that arginine and downstream metabolites are synthesized in response to mycelial growth.

In the previous study, the amount of agmatine in rice koji increased when cultivation pH was lowered to 3 (18). To examine the localization of arginine and agmatine, imaging patterns were compared. As shown in Fig. 3; Fig. S1, arginine was detected more strongly in rice koji at pH 3.0 and 6.0 than in steamed rice. On the other hand, agmatine localization was not observed in steamed rice and rice koji at pH 6.0, consistent with previous data (18). Furthermore, the accumulation of agmatine overlapped with parts of the region where arginine accumulation was observed. This result suggests that agmatine decarboxylase, the enzyme responsible for producing agmatine from arginine, is also produced in the interior of the steamed rice, likely by the nutrient mycelia.

As agmatine was detected in the interior region of rice koji by MSI analysis, ADC was expected to exist in the mycelium. To compare ADC activity in aerial and substrate mycelia, crude cell extract was obtained from each fraction. A. oryzae RIB40 conidia were inoculated into steamed rice and mixed well. Solid-phase cultivation was conducted to obtain rice koji. The obtained rice koji was suspended and vortexed vigorously in phosphate buffer and centrifuged (7,000 × g). By repeating those steps, aerial and substrate mycelial fractions were separated as supernatants and precipitates (Ps), respectively, as described in the Materials and Methods section. The collected precipitates were used for the substrate mycelial fraction. Aerial mycelium in the supernatant was collected by centrifuging at a higher speed (10,000 × g). Substrate and aerial fractions were frozen with liquid nitrogen and disrupted by a mixer and sonication on ice. The obtained mycelial and aerial mycelial fractions of rice koji were used for the ADC assay. As shown in Table 1, 83% of the total mycelium was obtained as substrate mycelia. Most of the ADC activity (98.4%) was detected in substrate mycelia. When each activity was compared per hyphal amount, 92.6% was also found in substrate mycelium. Based on the ratio, ADC was mainly expressed in substrate mycelia.

Since ADC activity was mainly detected in the substrate mycelia, purification of the enzyme was attempted using these mycelia. The substrate mycelia fraction obtained from 300 g of rice koji was frozen with liquid nitrogen, disrupted using a mixer, and sonicated on ice. The disrupted mycelial sample was centrifuged, and the supernatant was subjected to ammonium sulfate fractionation. The resulting active fraction was dialyzed and applied to an anion-exchange column, eluted with a linear gradient of NaCl. The fractions containing ADC activity were collected, concentrated, and further applied to a gel filtration column. Active fractions were concentrated and loaded onto an SDS-PAGE as shown in Fig. 4A. In gel-filtration chromatography, ADC activity was detected in several fractions in Fig. 4B. Fraction 14 exhibited the highest purification fold based on the ADC-specific activity (0.258 pmol^agmatine/min/mg^protein) in Fig. 4B.

The active fractions obtained after gel filtration were concentrated and applied to an SDS-PAGE preparatory gel. Two thick bands of fraction 14 were excised, dehydrated in acetonitrile, and alkylated. The resulting product was then digested with trypsin and desalted. The isolated peptides were analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The effluent from the high performance liquid chromatography (HPLC) was directly electrosprayed into the mass spectrometer, where peptides separated by LC were ionized. Experimental conditions are described in Table S1. MS1 survey scans were acquired, and peptides were identified. MS/MS fragmentation spectra were searched with Proteome Discoverer 2.4 (Thermo Fisher Scientific K.K.) against the A. oryzae RIB40 genome database. The results were further analyzed using Mascot™ v. 2.6 (Matrix Science, SC), and the identified proteins are listed in Table S2. Among them, gene AO090102000327 (A. oryzae RIB40 chromosome 4) encoding a 54 kDa protein showed the highest mascot score (895), and it was annotated as phosphatidylserine decarboxylase (PSD). In the A. oryzae genome database, no ADC was annotated, and six PSDs (AO090012000733, AO090005001124, AO090103000445, AO090102000327, AO090701000800, and AO090701000802) were annotated. To determine the phylogenetic relationships of PSDs in A. oryzae RIB40, PSDs from various organisms were selected, and their amino acid sequences were compared. Based on the aligned data, a phylogenetic tree was constructed as shown in Fig. 5. PSDs were classified into three major clades: the first one is the non-mitochondria-derived group, containing one gene (AO090012000733) in A. oryzae RIB40; the second one is the mitochondria-derived group, including bacteria, (25–27) with one gene (AO090005001124) in A. oryzae RIB40; and the third one is a unique clade, with no enzymatic activity reported for PSDs and containing four genes (AO090103000445, AO090102000327, AO090701000800, and AO090701000802) in A. oryzae RIB40 (28, 29). The protein identified in the present study (Gene AO090102000327) belongs to the third clade. PSD has been reported to function as a pyruvoyl-dependent decarboxylase, and unlike many decarboxylases, it does not require pyridoxal phosphate (PLP) (30, 31) as a cofactor. The inactive proenzyme of the pyruvoyl-dependent enzyme is cleaved into two active subunits, generally designated as α and β subunits. The cleavage occurs at the serine residue, termed nonhydrolytic serinolysis, and the NH₂ terminal of the serine residue is converted to ammonia and the pyruvoyl group (32). Gene AO090102000327, which was predicted to encode ADC in this study, had no introns in its open reading frame. Speculated amino acid sequence encoded by AO090102000327 possesses the Gly-Gly-Ser-Thr region that contains Ser functional for serinolysis of pyruvoyl-dependent enzyme (33–36), resulting in 441 amino acid peptide (49 kDa) and 42 amino acid one (5 kDa). To further elucidate the function of the enzyme encoded by AO090102000327, we attempted to express it as a recombinant form in Escherichia coli.

The full-length open reading frame of AO090102000327 was amplified via polymerase chain reaction (PCR) and introduced into the expression plasmid pET28a to enable the expression of the encoded protein in recombinant form. The resulting plasmid, designated pET28a-Ao-ADC1, was then introduced into E. coli BL21 CodonPlus (DE3)-RIL cells. Expression of the encoded protein with N-terminal His tags was induced by the addition of isopropyl β-D-1-thiogalactopyranoside (IPTG). Following induction, cells were harvested through centrifugation, resuspended in phosphate buffer, and disrupted using sonication. After the removal of cell debris via centrifugation, the supernatant and precipitate were combined with an SDS-containing sample buffer and analyzed using SDS-PAGE. Upon incubation of cells at 37℃, the induced protein was detected in insoluble precipitates as a 54 kDa protein (Fig. 6A). To mitigate protein aggregation, E. coli cells were cultivated at 20℃ following IPTG induction. The induced proteins were detected in the soluble fraction with sizes of 54 kDa and 49 kDa. The 49 kDa size aligned well with the anticipated size, resulting from the serinolysis of the pyruvoyl-dependent enzyme. However, the smaller fragment (5 kDa) was not detected in the SDS-PAGE, possibly due to being beyond the separation range. Protein purification was carried out using crude extracts obtained from cells cultivated at 20°C. After removing cell debris by centrifugation, the supernatant was applied to a column packed with Ni resin. Elution was performed stepwise with phosphate buffer supplemented with imidazole, followed by dialysis against phosphate buffer as described in the Materials and Methods section. The protein was purified to homogeneity, as shown in Fig. 6B. However, the 5 kDa subunit could not be confirmed in the normal SDS-PAGE. Therefore, Tris-Tricine SDS-PAGE was performed, and proteins were visualized by silver staining as shown in Fig. 6C. The 5 kD band corresponding to α subunit (shown as MA) was detected with 45 kD β subunit (shown as MB). To examine the enzymatic characteristics of the purified protein, the decarboxylation activity of arginine, ornithine, and phosphatidylserine was assessed, along with kinetic parameters (Km and Vmax). Reaction mixtures containing the substrate (arginine, ornithine, or phosphatidylserine) were incubated at various temperatures (20, 30, 40, 50, or 60℃) for different reaction times (5–60 min) with the purified protein in citrate buffer at various pH values (pH 3.0, 4.0, 5.0, or 6.0). The reactions were terminated by the addition of tricarboxylic acid (TCA), and the reaction mixtures were filtered. Contents of agmatine, putrescine, and phosphatidylethanolamine were determined to assess ADC, ODC, and PSD activities, respectively. Kinetic parameters (Km and Vmax) were then determined for arginine, ornithine, and phosphatidylserine. As shown in Table 2, the protein exhibited the lowest Michaelis constant (25 µM) and the highest maximum velocity (5.0 × 10⁻¹ mmol/min) when arginine was used as a substrate. On the other hand, no PSD activity was detected. Based on the obtained values, we concluded that AO090102000327 of A. oryzae RIB40 encodes arginine decarboxylase, not phosphatidylserine decarboxylase, and designated AO090102000327 as Ao-adc1. As shown in Table 3, no significant difference in decarboxylation activity was observed between the presence and absence of PLP, indicating that the protein is a pyruvoyl-dependent decarboxylase and does not require PLP for its activity. The optimal pH and temperature for ADC activity were also evaluated as shown in Fig. 7, showing that Ao-ADC1 optimally functions at pH 4.0 and 40℃.

The fungus produces agmatine during solid-state cultivation but not under submerged cultivation, and its productivity is notably enhanced in response to acidic conditions (18). To investigate the effect of pH on ADC1 expression in A. oryzae RIB40, we employed immunodetection using specific polyclonal antisera. A. oryzae RIB40 was cultured for 45 hours at 30°C under solid-state conditions with a pH range of 3.0–6.0. Disrupted crude cell extracts containing mycelia were obtained and subjected to SDS-PAGE, followed by electroblotting onto polyvinylidene difluoride (PVDF) membranes. Western blotting analysis was performed using these antibodies, and immunocomplexes were detected as shown in Fig. 8. ADC1 expression was detected under cultivation conditions with pH ranging from 4 to 6. However, at pH 3, ADC1 expression was notably lower.

A. oryzae RIB40 (19) was obtained from the National Research Institute of Brewing (Higashihiroshima, Hiroshima, Japan). E. coli strain TG1 (Funakoshi Co., Ltd., Tokyo, Japan) (Table S3), which was used for DNA manipulation to construct the plasmid pET28a derivatives to express Ao-adc1 gene from A. oryzae RIB40 (19), was routinely cultivated at 37°C in the lysogeny broth (LB) medium containing 20 µg/mL kanamycin. For the Ao-ADC1 production, E. coli BL21 Codon Plus (DE3)-RIL (Agilent, Santa Clara, CA, USA) (Table S3) cells harboring the expression plasmid were cultivated at 37°C or 20℃ on LB medium containing 20 µg/mL kanamycin and 30 µg/mL chloramphenicol. Plasmid pET28a was purchased from Merck (Darmstadt, Germany) and used for recombinant ADC1 expression in E. coli BL21 Codon Plus (DE3)-RIL cells.

Rice koji was prepared as previously described (37). Rice grains of 300 g polished to 90% of the total weight were soaked in tap water for 3 hours and steamed for 50 min. After cooling to 30°C, 3 g of the A. oryzae RIB40 conidia were inoculated into steamed rice and mixed well. The solid phase medium was cultivated with water-moistened gauze (15 × 23 cm), wrapped in a cotton cloth (35 × 35 cm) in a plastic container (16 × 23 × 5 cm), and incubated at 30°C. After 19 and 27 hours, the rice grains containing A. oryzae were mixed upside down to improve the aeration and lower the temperature. Finally, the solid culture was collected as rice koji after 45 hours cultivation and cooled at room temperature for 2 hours. The obtained rice koji was used in subsequent experiments.

Sample preparation was carried out according to the reported procedure (42) with slight modifications. One rice koji or steamed rice was first embedded in a mold (Base mold A, Falma, Tokyo, Japan) with 10% gelatin solution and then frozen with liquid nitrogen. The frozen sample block was attached to the sample holder using the optimal cutting temperature compounds (Surgipath FSC 22, Leica Biosystems, Nussloch, Germany) before sectioning using a cryostat (CM1850, Leica Biosystems). Longitudinal sections were obtained with 14 µm thickness from the approximate center of the solid phase medium grain. Adhesive cryofilm (Section-Lab, Yokohama, Kanagawa, Japan) was used to acquire the sample sections. Finally, the sections were attached to indium tin oxide (ITO) glass (Matsunami Glass, Osaka, Japan) using conductive double-sided tape (Shielding Non-woven Fabric Tape; 3M Company St. Paul, MN, USA). The matrix, α-cyano-4-hydroxycinnamic acid (CHCA) (Merck, Darmstadt, Germany), was supplied to the surface of the sections at 220°C with a thickness of 0.7 µm using a vacuum sublimation machine, iMLayer (Shimadzu, Kyoto, Japan). After that, 500 µL of 90% CHCA was sprayed evenly onto the glass slide using an airbrush. Then, the analysis was performed using iMScope QT (Shimadzu). All mass spectra were obtained in the m/z range (88.00–210.00) in positive ion mode. After analysis, analysis software IMAGEREVEAL MS (Shimadzu) was used to create the MS images by selecting m/z derived from target metabolites.

When aerial and substrate mycelial fractions were needed, obtained rice koji was suspended in 10 mL of 50 mM phosphate buffer (pH 7.0) in a 50 mL centrifuge tube and vortexed vigorously. The suspended solution was centrifugation at 7,000 × g, 4°C, for 10 min, and the supernatant and precipitate were separately collected. To the precipitate, 10 mL of 50 mM phosphate buffer (pH 7.0) was added again, vortexed vigorously, and centrifuged at 7,000 × g, 4°C, for 10 min. This step was repeated six times, and the final precipitate was frozen with liquid nitrogen, and the supernatants (S) (60 mL in total) were collected and centrifuged at 10,000 × g, 4°C for 10 min to collect the aerial mycelium as a precipitate. The precipitate (aerial mycelial fraction) was then frozen with liquid nitrogen and resuspended in 70 mL of 50 mM phosphate buffer (pH 7.0), followed by crushing with mixer and sonication on ice. The mixture was used as an aerial mycelial fraction. The precipitate containing substrate mycelia obtained by the centrifugation was frozen with liquid nitrogen. To 10 g of the precipitate, 100 mL of 50 mM phosphate buffer (pH 7.0) was added, followed by crushing with mixer and sonication on ice. The mixture was used as a substrate mycelial fraction.

The growth of A. oryzae was evaluated by determining the amount of N-acetylglucosamine (GlcNAc), which is the building block of the major fungal cell wall constituent, chitin (18). To measure the total amount of mycelia, 2 g of the rice koji was dried at 100°C for 1 hour and completely ground using a mortar. The resultant powder was suspended in 10 mL of 50 mM phosphate buffer (pH 7.0), vigorously vortexed for 10 seconds, and recovered by centrifugation (10,000 × g, 10 min). These washing steps were repeated three times, and the resultant pellet was resuspended in 10 mL 50 mM phosphate buffer (pH 7.0). Next, 10 mg of Yatalase (TaKaRa Bio, Tokyo, Japan) was added to the suspension and incubated at 37°C for 1 hour, with reciprocal shaking at 200 rpm. The supernatant was collected as the GlcNAc fraction by centrifugation (10,000 × g, 10 min). The amount of GlcNAc was determined using a colorimetric method. Then, 500 µL of the GlcNAc fraction was mixed with 100 µL of 0.8 M borate buffer (pH 9.1, adjusted with KOH) and heated in boiling water for 3 min. After cooling in tap water, 3 mL of a coloring solution composed of 10 mg/mL p-dimethylaminobenzaldehyde and 125 mM hydrochloric acid in glacial acetic acid was added to the mixture and further incubated at 37°C for 20 min, giving rise to purple coloration. The absorbance of the reaction mixtures at 585 nm was measured, and the amount of GlcNAc in cultures (in micrograms of GlcNAc per gram culture) was estimated based on a standard curve generated using 0.05, 0.1, 0.15, and 0.2 µmol GlcNAc (Wako Pure Chemical, Osaka, Japan).

Agmatine and putrescine amounts were quantified using HPLC, as previously described (18). To monitor ADC and ODC activities, the products of enzymatic reactions were analyzed by HPLC. The reaction mixtures (500 µL) contained 0.1 mM substrate (arginine or ornithine) in 25 mM citrate buffer (pH 3.0, 4.0, 5.0, or 6.0). Each purified protein (5 µg) or crude cell extract was added to the mixture and then incubated at 30°C for 60 min. When PLP dependency was examined, 0.1 mM PLP was added to the reaction mixture. The reaction was terminated by the addition of 50 µL of 10% (wt/vol) TCA, and the reaction mixture was filtered through a 0.45 µm pore-size filter (Millex LH filter; Millipore, Bedford, MA, USA). The agmatine and putrescine contents in the filtrate were determined by HPLC to assess the ADC and ODC activity, respectively.

PSD activity was measured by detecting phosphatidylethanolamine. The reaction mixtures (250 µL) containing 0.2 mM phosphatidylserine, 0.2% Triton X, 25 mM phosphate buffer (pH 7.0), and 50 µL of enzyme sample were incubated at 30°C for 10 min. When PLP dependency was examined, 0.1 mM PLP was added to the reaction mixture. The reaction was terminated by adding 25 µL of 2 N KOH, following centrifugation (20,000 × g, 15 min, 4°C). The supernatant was obtained, and produced phosphatidylethanolamine was detected according to the phosphatidylethanolamine assay kit (Abcam, Cambridge, UK).

The rice koji obtained from 300 g of steamed rice was frozen with liquid nitrogen and resuspended in 200 mL of 50 mM phosphate buffer (pH 7.0), followed by crushing with mixer and sonication on ice. The disrupted hyphal sample was centrifuged (7,000 × g, 15 min), and the supernatant was obtained as crude cell extract. After the supernatant was obtained by centrifugation (7,000 × g, 15 min), ammonium sulfate was added to 60% saturation. The supernatant was obtained by centrifugation (15,000 × g, 20 min), and ammonium sulfate was added to 80% saturation. The precipitate was collected by centrifugation (15,000 × g, 20 min), dissolved in 50 mM phosphate buffer (pH 7.0), and dialyzed against the same buffer. The solution was applied to anion-exchange column, 5 mL Hitrap Q (Cytiva, Tokyo, Japan), followed by elution with a linear gradient of NaCl (0 to 1.0 M). The fractions with containing ADC activity were collected, dialyzed against 50 mM phosphate buffer (pH 7.0). The fractions were collected, concentrated with an Amicon Ultra-3K device (Millipore), applied to a Superdex 200 HR 10/30 gel filtration column (Cytiva), and eluted with the same buffer. Fractions with enzymatic activity were collected and concentrated. The concentrations of purified proteins were determined using the Bradford dye-binding assay (43) with bovine serum albumin (BSA) as a standard. The active fractions were concentrated and applied to an SDS-PAGE preparatory gel. The thick bands were cut out, dehydrated in acetonitrile, and alkylated by incubation with 55 mM iodoacetamide for 90 min. The product was digested with trypsin (Trypsin Gold, Promega, WI) overnight at 37°C, and desalted by Zip-Tip (Millipore). The isolated peptides were analyzed by liquid chromatography mass spectrometry (LC–MS) (Table S1) using an Orbitrap MS (Thermo Fisher Scientific K.K.) equipped with Advance UHPLC HTSxt-PAL (AMR), and Mascot was used as the sequence database search engine. The flow rate of LC was set to 300 nL/min with a gradient of 5% solution A (99.9% water and 0.1% formic acid) and 95% solution B (100% acetonitrile). Solution B was increased up to 35% at 40 min and further to 90% at 41 min. After holding for 5 min, the mixture was washed with 100% solution B was conducted for 60 min. The effluent from the HPLC was directly electrosprayed into the mass spectrometer. Peptides separated by LC were ionized at 1.7 kV in the positive ion mode. MS1 survey scans were acquired from 350 to 1800 m/z at a resolution of 60,000. Peptides were identified with a precursor mass tolerance of 10 ppm, and a fragment m/z tolerance of 0.8. MS/MS fragmentation spectra was searched with Proteome Discoverer 2.4 (Thermo Fisher Scientific K.K.). The results were analyzed by the Mascot™ v. 2.6 (Matrix Science, SC) using the following databases: MASCOT MS/MS ion search/via PD2.4, Uniprot AspergillusOryzae_RIB40_UP000006564, and Genome_NCBI.

The full-length Ao-adc1 gene was amplified by PCR from A. oryzae RIB genomic DNA using the specific primer pairs AO-CAA-1-Rv and AO-GAC-1-Fw (Table S4). The amplified DNA was digested with restriction enzymes, EcoRI and NdeI, and introduced into EcoRI and NdeI sites of plasmid pET28a (Takara Bio) (Table S3). The plasmid construction was confirmed by nucleotide sequence determination by using synthetic primers (1Ao-GCC-Fw2, 2Ao-GAC-Fw2, 3Ao-GAA-Fw2, 4Ao-CTC-Fw2, and 5Ao-GTG-Fw2 listed in Table S4). The resultant plasmid was designated as pET28a-Ao-ADC1. The plasmid was introduced into E. coli BL21 CodonPlus (DE3)-RIL cells, and the transformant cells were grown in the LB medium containing 20 µg/mL kanamycin and 20 µg/mL chloramphenicol at 37℃. The expression of the ADC with N-terminal His tags was induced by the addition of 1 mM IPTG. After further incubation at 37℃ or 20℃ for 4 hours, the cells were harvested by centrifugation, resuspended in 50 mM phosphate buffer [pH 7.0], and disrupted by sonication. After removing the cell debris by centrifugation, each supernatant was applied to a column packed with 5 mL of His60 Ni Superflow Resin (Takara Bio) and eluted stepwise with the 50 mM phosphate buffer [pH 7.0] supplemented with 100, 200, or 500 mM imidazole. Eluted protein was dialyzed against 50 mM phosphate buffer [pH 7.0]. The concentrations of purified proteins were determined using the Bradford dye-binding assay with BSA as a standard. To monitor purification, normal Tris-glycine SDS-PAGE was conducted and visualized by Coomassie brilliant blue staining. For small peptide monitoring Tris-tricine SDS-PAGE was performed according to the reported procedure (44) and visualized by silver staining.

The effect of pH on ADC1 expression in A. oryzae RIB40 was examined by immunodetection with specific polyclonal antisera. The polyclonal antisera against Ao-ADC1 were obtained from Female New Zealand White rabbit (Japan SLC) and designed as anti-Ao-ADC1. A. oryzae RIB40 was cultured for 60 hours in solid state condition at the range of pH 3.0–6.0 by using 50 mM citric acid buffer. One gram of rice koji was frozen with liquid nitrogen and resuspended in 10 mL of 50 mM phosphate buffer (pH 7.0), followed by crushing with mixer and sonication on ice. Disrupted cells were centrifuged, and supernatants were obtained. These samples were subjected to electrophoresis in gels containing 0.1% SDS sulfate and 10% PAGE and then electroblotted onto polyvinylidene difluoride membranes (Bio-Rad, Tokyo, Japan). The membranes were blocked using 5% skimmed milk and incubated with rabbit anti-Ao-ADC1 serum at a 1:2,000 dilution subjected to SDS-PAGE, and the gel was stained with coomassie brilliant blue (CBB) to display the total protein. Western blotting was performed using antibodies that bind specifically. Detection of immunocomplexes was performed using Alexa Fluor 700 goat anti-rabbit IgG (Invitrogen, Carlsberg, CA), and the signal was visualized with an Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE).

The phylogenetic analysis was conducted using the RAxML method provided by GenomeNet operated by the Kyoto University Bioinformatics Center (https://www.genome.jp/). Bootstrap resampling was performed 1,000 times, and only the values observed in more than 50% of the replicas are shown at the branching points. Scale bar indicates 1 substitution per amino acid.