Salt‐tolerant endopeptidases from xerophilic Aspergillus sydowii for potential applications in meat processing

Abstract

Background

Dashigara koji is a fermented product made from katsuobushi grounds using the fungus Aspergillus sydowii MA0196. It contains bioactive peptides generated through the proteolysis of katsuobushi proteins. However, the enzymatic basis of this proteolysis remains unclear. This study characterized two major endopeptidases secreted by A. sydowii MA0196: a subtilisin‐type protease (Oryzin MA0196) and a chymotrypsin‐like serine protease (Tryp MA0196).

Results

Recombinant Oryzin (rOryzin MA0196) expressed in Pichia pastoris, and wild‐type Tryp MA0196, purified from the native strain, were analyzed biochemically. Both enzymes exhibited broad pH tolerance and stability under high salt conditions, indicating xerotolerance. Tryp MA0196 retained high activity at up to 12.5% (w/v) NaCl (2.14 mol L⁻¹), whereas the activity of both enzymes gradually decreased with increasing glycerol concentration. Neither enzyme was affected by high histidine concentrations, which may reflect adaptation to katsuobushi fermentation. Activity assays showed that Tryp MA0196 acted on a broad range of protein substrates, whereas rOryzin MA0196 displayed high levels of activity toward casein, myofibrillar proteins, and type I collagen. Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis confirmed efficient hydrolysis of katsuobushi proteins, including myosin‐derived peptides, by both enzymes.

Conclusions

rOryzin MA0196 showed high levels of activity toward intact myofibrillar and collagen proteins, supporting its potential application as a meat‐tenderizing enzyme. Tryp MA0196 showed superior activity toward physicochemically modified myosin‐derived proteins in katsuobushi, indicating a specific role in katsuobushi protein degradation and potential applications in the valorization of katsuobushi byproducts. © 2026 Society of Chemical Industry.

ABBREVIATIONS

AEBSF

4‐(2‐aminoethyl)benzenesulfonyl fluoride hydrochloride

Oryzin MA0196

subtilisin‐type serine protease from Aspergillus sydowii MA0196

PMSF

phenylmethylsulfonyl fluoride

pNA

p‐nitroanilide

rOryzin MA0196

recombinant Oryzin MA0196

TLCK

tosyl‐l‐lysyl‐chloromethane hydrochloride

Tryp MA0196

chymotrypsin‐like serine protease from Aspergillus sydowii MA0196

INTRODUCTION

Xerophilic filamentous fungi, which can thrive in conditions of extremely low water activity, remain relatively underexplored compared with other industrially important microorganisms. ¹ , ² Nevertheless, these fungi are increasingly recognized as promising sources of robust enzymes that can function under harsh processing conditions such as high salinity and low water activity. ¹ , ² Among microbial proteases, alkaline serine proteases dominate the global enzyme market, accounting for nearly two‐thirds of industrial protease applications, due to their high catalytic activity, stability, broad substrate range, and abundant secretion by microbial hosts. ³ , ⁴ They are widely employed in detergents, leather processing, and the food industry, largely because of their versatility under diverse environmental conditions. ³ , ⁴

Although salt‐tolerant proteases from halophilic bacteria and archaea have been studied extensively, ⁵ , ⁶ , ⁷ reports on such enzymes from xerophilic fungi remain scarce. ⁸ This gap highlights the need to explore xerophilic Aspergillus species as potential sources of novel proteases with unusual catalytic properties. Aspergillus sydowii, a representative xerophilic fungus, has been studied as a valuable resource for industrial enzymes, including feruloyl esterase, ⁹ γ‐glutamyl transpeptidase, ¹⁰ and pectinase. ¹¹ Katsuobushi, a traditional Japanese smoked and fermented fish product, is typically made from Katsuwonus pelamis (skipjack tuna). ¹² In previous studies, the valorization of katsuobushi grounds (dashigara), a protein‐rich byproduct generated after extracting soup stock (dashi) from katsuobushi, was explored through fermentation with xerophilic Aspergillus species. ¹² The proteins in dashigara were hydrolyzed effectively by xerophilic Aspergillus, yielding bioactive peptides with potential nutritional and functional benefits. ¹² Fermented dashigara can be regarded as a natural cocktail of extracellular proteases, in which multiple endo‐ and exopeptidases act synergistically; endopeptidases cleave proteins at internal peptide bonds, whereas exopeptidases release amino acids or small peptides sequentially from the N‐ or C‐terminus. Such enzyme mixtures may provide versatile biocatalytic functions, not only for peptide generation, but also for broader applications in food processing.

Protein hydrolysis in such processes is initiated primarily by endopeptidases, which randomly cleave peptide bonds to generate oligopeptides for subsequent processing by exopeptidases. In previous work, ¹² two major endopeptidases were identified, secreted by A. sydowii MA0196: a subtilisin‐type protease (Oryzin MA0196) and a chymotrypsin‐like protease (Tryp MA0196). Despite their evident importance, the catalytic properties of these enzymes have not yet been elucidated.

The present study characterized rOryzin MA0196 and Tryp MA0196 from A. sydowii MA0196, focusing on their catalytic properties, stability in different environmental conditions, and substrate specificity. Their potential for application in meat processing was also assessed – particularly their ability to hydrolyze beef sarcoplasmic, myofibrillar, and collagen proteins – to expand the potential utility of xerophilic fungal proteases.

MATERIALS AND METHODS

Materials and reagents

Katsuobushi powder and residual katsuobushi grounds after broth extraction were obtained from Marutomo Co. (Ehime, Japan). Toyopearl DEAE‐650 M and Toyopearl Butyl‐650 M resins were purchased from Tosoh (Tokyo, Japan). Protein substrates, including bovine serum albumin (BSA), casein (Hammerstein grade), collagen, elastin, hemoglobin, myoglobin, and p‐nitroanilide (pNA) derivatives such as l‐leucine pNA, l‐arginine pNA, and Nα‐benzoyl‐dl‐arginine pNA, were obtained from Sigma‐Aldrich (St Louis, MO, USA). All other chemical reagents were of analytical grade.

Microorganisms and culture conditions

Aspergillus sydowii MA0196 was used as the source of Oryzin MA0196, Tryp MA0196, and their encoding genes. Strain MA0196 was cultured by solid‐state fermentation in previously optimized conditions. ¹² Cultivation in a solid medium consisting of residual katsuobushi grounds was carried out for 18–21 days, and the cultures were stirred weekly with a sterilized bar to promote fungal growth throughout the medium. ¹² Crude enzyme preparations were extracted from the katsuobushi solid‐state culture as described below for the purification of Tryp MA0196.

Transformants of Pichia pastoris GS115 heterologously expressing protease were cultivated in buffered methanol‐complex medium (BMMY), which consisted of 10 g L⁻¹ yeast extract, 20 g L⁻¹ polypeptone, and 13.4 g L⁻¹ BD Difco yeast nitrogen base (Becton, Dickinson & Co., Franklin Lakes, NJ, USA) and 100 mmol L⁻¹ potassium phosphate (pH 5.0). Induction of recombinant protease expression was performed by adding 0.5% (v/v) methanol, as previously described. ¹³ The culture supernatant was collected for purification of recombinant protease.

Expression of recombinant Oryzin MA0196

Total RNA from strain MA0196 was extracted using Sepasol‐RNA I Super G Plus (Nacalai Tesque Inc., Kyoto, Japan), as previously reported. ¹² First‐strand cDNA was synthesized with the PrimeScript 1st Strand cDNA Synthesis Kit (Takara Bio Inc., Kusatsu, Japan). Target peptidase genes were amplified using the cDNA mixture as the template and PrimeSTAR GXL DNA polymerase (Takara Bio Inc.). Attempts were made to obtain active recombinant Oryzin MA0196 (rOryzin MA0196) and Tryp MA0196 using the P. pastoris pPICZαA and Escherichia coli pCold I expression systems, respectively. However, only rOryzin MA0196 was successfully expressed in active form in P. pastoris.

The cDNA encoding mature Oryzin MA0196 (lacking the 20‐residue prepeptide) was amplified by polymerase chain reaction (PCR) using primers oryzin_F and oryzin_R (Supporting Information, Table S1). The linearized pPICZαA vector was generated by PCR using primers pPICZαA_anti and pPICZαA_sense (Supporting Information, Table S1). Purified PCR products were ligated into the linearized vector following the In‐Fusion Snap Assembly protocol (Takara Bio Inc.). Recombinant plasmids were introduced into E. coli HST08. The expression vector was linearized by PmeI digestion and transformed into P. pastoris GS115 by electroporation (MicroPulser, Bio‐Rad Laboratories, Hercules, CA, USA). Briefly, a mixture of 10 μg PmeI‐linearized plasmid DNA and 80 μL electrocompetent cells was subjected to electroporation using the Pichia yeast program. Electroporated cells were plated on yeast extract–peptone–dextrose–sorbitol (YPDS) agar plates (10 g L⁻¹ yeast extract, 20 g L⁻¹ peptone, 20 g L⁻¹ glucose, 1 mol L⁻¹ d‐sorbitol, and 15 g L⁻¹ agar) supplemented with 100 μg mL⁻¹ Zeocin (Thermo Scientific Inc., Waltham, MA, USA) and incubated at 30 °C for 2–4 days. Transformants were cultivated in yeast extract–peptone–dextrose (YPD) liquid medium, and protease‐producing strains were selected based on enzyme activity detected in the culture supernatant after methanol induction.

Purification of proteases from A. sydowii MA0196

rOryzin MA0196 was purified by ion‐exchange chromatography as described below. Tryp MA0196 could not be expressed heterologously in active form and was instead purified from crude enzyme preparations of strain MA0196. Protein concentrations were determined by using a protein assay Lowry kit (Nacalai Tesque Inc.).

Wild‐type Tryp MA0196 was purified from crude enzyme preparations (90 mL) obtained from strain MA0196 cultures. ¹² The purification procedure, including ion‐exchange and hydrophobic chromatography steps, is summarized in Supporting Information, Table S2. Enzyme purity was assessed by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE).

rOryzin MA0196 was purified using culture supernatant (30 mL), which was dialyzed against 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5; buffer A) with two buffer exchanges. The dialyzed enzyme solution was applied to a Toyopearl DEAE‐650 M column (1.5 × 20 cm) equilibrated with buffer A. After washing with 90 mL buffer A, proteins were eluted with a 0–0.3 mol L⁻¹ NaCl gradient in 450 mL buffer A. Fractions (6 mL) were collected at a flow rate of 1.0 mL min⁻¹. The most active fractions, together with adjacent fractions showing more than 70% relative activity, were pooled (see Supporting Information, Table S2).

Enzyme assays

Hydrolysis of protein substrates by rOryzin MA0196 and Tryp MA0196 was assayed in 50 mmol L⁻¹ Tris–HCl buffer, pH 7.5, using casein as the substrate. The reaction mixture (0.55 mL) contained 5 g L−1 casein in 50 mmol L⁻¹ Tris–HCl buffer (pH 7.5) and enzyme solution. After incubation at 30 °C for 60 min, the reaction was terminated by addition of 0.45 mL of 100 g L−1 trichloroacetic acid. The mixture was placed on ice for 10 min and centrifuged at 15 000 × g for 10 min. Supernatant (0.2 mL) was combined with 2.5 mL of 20 g L−1 Na₂CO₃ in 0.1 mol L⁻¹ NaOH and 0.2 mL of Folin–Ciocâlteu reagent, followed by incubation at room temperature for 30 min. Release of casein peptides was determined by measuring absorbance at 600 nm. A calibration curve was generated using l‐tyrosine as the standard. One unit (U) of protease activity was defined as the amount of enzyme required to release 1 μmol of tyrosine equivalents per min under the assay conditions (50 mmol L⁻¹ Tris–HCl buffer, pH 7.5, at 30 °C). Other proteins (Table 1) were also used as substrates in place of casein, and hydrolysis was evaluated under the same conditions.

Table 1.

Substrate specificity of rOryzin MA0196 and wild‐type Tryp MA0196 toward protein substrates

Substrate	Enzyme activity (U mg⁻¹)
	rOryzin	Tryp
Casein	16 ± 0.3a	5.37 ± 0.09a
Hemoglobin	1.70 ± 0.05b	1.32 ± 0.03c
Myoglobin	1.45 ± 0.03c	0.67 ± 0.02e
BSA	1.49 ± 0.08bc	1.79 ± 0.02b
Elastin	0.46 ± 0.14e	0.39 ± 0.03f
Collagen	0.12 ± 0.06f	0.25 ± 0.02f
Katsuobushi protein	1.17 ± 0.11d	0.96 ± 0.09d

Notes: Enzyme activity was measured using casein, hemoglobin, myoglobin, bovine serum albumin (BSA), elastin, collagen, or soluble katsuobushi protein. Reaction mixtures contained 0.15% (w/v) substrate protein. Results are expressed as specific activity (U mg⁻¹, mean ± SD, n = 5). Different letters indicate significant differences within each column. P < 0.05, one‐way analysis of variance (ANOVA) followed by Fisher's protected least significant difference (PLSD) test.

Hydrolysis of synthetic substrates was performed using a reaction mixture (1.1 mL) containing 0.5 mmol L⁻¹ pNA derivative (Table 2) in 50 mmol L⁻¹ sodium–potassium phosphate buffer (pH 7.5) and enzyme solution, as described previously. ¹⁴ After incubation at 40 °C for 30 min, the release of p‐nitroaniline was quantified by measuring absorbance at 405 nm. A calibration curve was prepared using p‐nitroaniline as the standard. One unit (U) of protease activity was defined as the amount of enzyme required to release 1 μmol of p‐nitroaniline per min.

Table 2.

Substrate specificity of rOryzin MA0196 and wild‐type Tryp MA0196 toward synthetic substrates

	Enzyme activity (U mg⁻¹)
Substrate	rOryzin	Tryp
Leu‐pNA	<0.001	<0.001
Gly‐Phe‐pNA	<0.001	<0.001
Phe‐Pro‐Ala‐pNA	<0.001	<0.001
Ala‐pNA	<0.001	<0.001
Arg‐pNA	<0.001	<0.001
Bz‐L‐Arg‐pNA	<0.001	0.032 ± 0.004
Sac‐Ala‐Ala‐Ala‐pNA	<0.001	<0.001

Notes: The aminopeptidase assay was performed using various p‐nitroanilide (pNA) derivatives in 50 mmol L⁻¹ sodium potassium phosphate buffer (pH 7.5), as described by Blinkovsky et al. ¹⁴ Values are expressed as means ± SDs (n = 5). No statistical comparisons were performed.

Preparation of soluble protein fractions for proteolysis assay

Katsuobushi soluble proteins were prepared as previously described. ¹² Briefly, 1 g of katsuobushi powder was mixed with 10 mL of 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5). After incubation at room temperature for 1 h, the soluble protein fraction was collected by centrifugation (20 000 × g for 10 min). During the katsuobushi manufacturing process, fish muscle proteins were physicochemically fragmented through repeated heating–drying and smoking–cooling cycles. ¹² The main peptides identified in this fraction were derived from the coiled‐coil region of the myosin heavy chain tail (accession no. XP_042261668.1). ¹² The amount of enzyme added to the reaction mixture was standardized based on its specific activity toward casein (Table 1). Tryp MA0196 was added at approximately three times the protein mass of rOryzin MA0196 to achieve comparable caseinolytic activity in the assays.

Soluble protein fraction from beef was prepared as described by Uytterhaegen et al. ¹⁵ Briefly, 6 g beef was mixed with 3 g quartz sand and ground in a chilled mortar for 5 min. The homogenate was extracted with 25 mL of buffer (0.8 mmol L⁻¹ ethylenediaminetetraacetic acid disodium salt dihydrate, 250 mmol L⁻¹ sucrose, 50 mmol L⁻¹ Tris–HCl, pH 7.6), stirred thoroughly, vortexed, and centrifuged (15 000 × g, 20 min, 4 °C). The supernatant (approximately 75 mL total, after two extractions) was designated as fraction I.

To prepare the myofibrillar protein fraction from beef, the pellet from fraction I was washed and extracted with 25 mL buffer (70 mmol L⁻¹ sodium lauryl sulfate, 10 mmol L⁻¹ imidazole, 100 mmol L⁻¹ 2‐mercaptoethanol, 50 mmol L⁻¹ Tris–HCl, pH 7.0). After stirring and centrifugation (15 000 × g, 20 min, 4 °C), the supernatant (approximately 75 mL total after two extractions) was designated as fraction II.

Collagen recovery from fraction II was low, so commercially obtained pepsin‐solubilized type I bovine skin collagen (Fujifilm Wako Pure Chemical Corp., Osaka, Japan) was used.

Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis and Tricine–SDS‐PAGE

The purified peptidases (approximately 1–5 μg of protein) from strain MA0196 were denatured and loaded onto an e‐PAGEL precast gel E‐T12.5 L. Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis was performed using an ATTO AE6530 electrophoresis system (ATTO Corp., Tokyo, Japan) following the manufacturer's instructions. The soluble protein fraction (approximately 100 μg of protein) from katsuobushi powder or beef, and hydrolyzed samples, were also analyzed by SDS‐PAGE and Tricine–SDS‐PAGE. Before electrophoresis, each sample was mixed with SDS‐PAGE sample buffer at a 2:1 ratio. Tricine–SDS‐PAGE was carried out using a p‐PAGEL precast gel P‐T16.5S and the AE6530 system, following the manufacturer's instructions.

Effect of various factors on peptidase activity and stability

1. The effect of temperature on enzyme activity was determined in 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5) across a temperature range of 15–70 °C. To assess thermostability, purified peptidase was preincubated in 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5) at 15–70 °C for 30 min. Residual activity was then measured in 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5) at 30 °C, as described above.

2. To investigate the effect of pH on enzyme activity assays were carried out using purified protease in 50 mmol L⁻¹ buffer solutions at different pH values: 50 mmol L⁻¹ sodium acetate buffer (pH 4.0–6.0), 50 mmol L⁻¹ sodium potassium phosphate buffer (pH 6.0–7.5), 50 mmol L⁻¹ Tris–HCl buffer (pH 7.5–9.0), and 50 mmol L⁻¹ glycine–NaOH buffer (pH 9.0–12.0). The specific activity of rOryzin MA0196 and Tryp MA0196 at pH 7.5 was 16 and 5.4 U mg⁻¹, respectively, and the relative activity was calculated based on the specific activity. To determine pH stability, purified protease was dialyzed overnight at 4 °C against disodium phosphate–citric acid buffers with various pH values. Following dialysis, residual activity was measured in sodium citrate buffer (pH 6.0).

3. The effects of NaCl, glycerol, and histidine concentrations on enzyme activity were investigated. Enzyme assays were performed using the purified peptidases in a reaction mixture containing NaCl (0–20% (w/v), 0–3.42 mol L⁻¹), glycerol (0–50% (v/v), 0–6.8 mol L⁻¹), or l‐histidine monohydrochloride monohydrate (0–10% (w/v), 0–0.48 mol L⁻¹).

Substrate specificity

For rOryzin MA0196 and wild‐type Tryp MA0196, protein substrates (hemoglobin, myoglobin, BSA, elastin, collagen, and katsuobushi proteins) were tested as alternative substrates to casein (Table 1). Substrate specificity toward artificial substrates was tested using pNA derivatives as described previously (Table 2). ¹⁴

Statistical analysis

All experiments were performed using three independent biological replicates. Each biological replicate was analyzed in technical triplicate. Data are presented as means ± standard deviations (SD). Statistical significance was evaluated by one‐way analysis of variance (ANOVA) followed by Fisher's protected least significant difference (PLSD) test for multiple comparisons, where appropriate. Differences were considered statistically significant at P < 0.05. All statistical analyses were performed using GraphPad Prism version 6.02 for Windows (GraphPad Software, La Jolla, CA, USA).

RESULTS AND DISCUSSION

Sequence analysis of Oryzin MA0196 and Tryp MA0196 from A. sydowii strain MA0196

All sequence data were deposited in the DNA Data Bank of Japan (DDBJ) database under accession numbers LC764757 (Tryp MA0196) and LC764758 (Oryzin MA0196). InterPro database analysis (EMBL‐EBI) indicated that Oryzin MA0196 belongs to the subtilisin‐like serine protease family (S8) (PR00723), which typically hydrolyzes peptide bonds in a broad range of substrates using a catalytic triad and is active under alkaline conditions. Oryzin MA0196 consists of a 20‐residue prepeptide and a 383‐residue mature protein. The catalytic triad residues Glu162, His193, and Ser348 (protein residue numbering including the preproprotein sequence) were identified by comparison with Aspergillus Oryzins and other subtilisin‐type proteases. ¹⁶ The amino acid sequence of Oryzin MA0196 shares 84% identity with Oryzin from A. nidulans FGSC A4 (Q00208), and less than 80% with other Aspergillus Oryzins (Supporting Information, Fig. S1). Phylogenetic analysis placed Oryzin MA0196 in the same clade as Oryzin FGSC A4 (Fig. 1). Several Oryzins (aspergillopeptidase B, EC 3.4.21.63) have been characterized in terms of their catalytic and structural properties, ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ providing a basis for comparison with rOryzin MA0196. Supporting Information, Fig. S1 summarizes details of sequence identities.

Figure 1.

Phylogenetic analysis of (A) Oryzin MA0196 from Aspergillus sydowii MA0196 and Aspergillus serine proteases, and (B) Tryp MA0196 from A. sydowii MA0196 and fungal and insect trypsin‐type serine proteases. Amino acid sequences of proteases were retrieved from the UniProtKB/Swiss‐Prot database and aligned using GENETYX version 13. A phylogenetic tree was constructed by the neighbor‐joining method with 1000 bootstrap replications. Accession numbers and sequence identities relative to Oryzin MA0196 and Tryp MA0196 are provided in the legends of Supporting Information Figs S1 and S2.

Tryp MA0196 was classified as a chymotrypsin‐like serine protease (PR00722), which preferentially cleaves peptide bonds adjacent to aromatic residues such as phenylalanine, tyrosine, or tryptophan. Tryp MA0196 consists of a 16‐residue prepeptide and a 232‐residue mature protein, and the catalytic triad residues His60, Asp105, and Ser202 (protein residue numbering including the preproprotein sequence) are consistent with trypsins from Anopheles gambiae (e.g., P35036). ²¹ The sequence of Tryp MA0196 showed 70% identity with a putative trypsin‐like serine protease (XP_659970) from A. nidulans FGSC A4 (Supporting Information, Fig. S1). Phylogenetic analysis placed Tryp MA0196 in the same clade as trypsins (40% to 44% identity) from Fusarium spp. Previously characterized Fusarium trypsins, including those from F. oxysporum ²² and Fusarium sp. BLB‐2006a, ²³ provide reference information for evaluating the catalytic properties of Tryp MA0196 (Fig. 1 and Supporting Information Fig. S1). Tryp MA0196 exhibits only moderate sequence identity (<45%) with characterized fungal trypsins (from Fusarium spp.), and no close orthologs have been functionally characterized in Aspergillus, suggesting that it represents a distinct lineage of chymotrypsin‐like proteases adapted to xerophilic environments.

Purification of rOryzin from P. pastoris and wild‐type Tryp MA0196 from A. sydowii MA0196

Attempts were made to express both Oryzin MA0196 and Tryp MA0196 heterologously using recombinant expression systems. However, only Oryzin was obtained successfully in active recombinant form (rOryzin MA0196). Optimal heterologous expression was achieved by cultivation in BMMY medium (pH 5.0) for 5 days with 0.5% (v/v) methanol and shaking at 140 rpm. In contrast, expression of Tryp MA0196 in E. coli using isopropyl β‐d‐1‐thiogalactopyranoside induction (0.1–1.0 mmol L⁻¹ at 15 °C) resulted in the accumulation of the protein in insoluble inclusion bodies. Based on structural comparison with F. oxysporum trypsin, ²² Tryp MA0196 contains three disulfide bonds (Cys45‐Cys61, Cys171‐Cys187, and Cys198‐Cys222) (Supporting Information, Fig. S1) (residues numbered according to the full‐length sequence including the preproprotein), and improper disulfide bond formation likely explains the lack of soluble active enzyme in E. coli. Consequently, Tryp MA0196 was purified directly from wild‐type A. sydowii strain MA0196 extracts obtained by solid‐state fermentation.

Under standard assay conditions (50 mmol L⁻¹ Tris–HCl buffer, pH 7.5, 30 °C), the specific activity of purified rOryzin MA0196 toward casein was 16 U mg⁻¹, whereas that of wild‐type Tryp MA0196 was 5.4 U mg⁻¹. Supporting Information, Table S2, summarizes the purification procedure and associated parameters, including total activity and recovery. The calculated molecular mass of rOryzin MA0196 was 40 336 Da, which was consistent with the experimental value of 39.9 kDa determined by SDS‐PAGE (Supporting Information, Fig. S2). The calculated molecular mass of Tryp MA0196 was 23 515 Da, which was lower than the apparent molecular mass of approximately 35.2 kDa determined by SDS‐PAGE (Supporting Information, Fig. S2). The sequence of Tryp MA0196 contains two N‐glycosylation motifs (¹³⁰Asn‐Ala‐Ser¹³² and ¹⁵³Asn‐Val‐Ser¹⁵⁵), and periodic acid–Schiff staining confirmed protein glycosylation (Supporting Information, Fig. S2). Together, these results demonstrated that both rOryzin and Tryp MA0196 were purified to homogeneity, providing reliable preparations for subsequent biochemical characterization and comparative analysis with Aspergillus Oryzins and Fusarium trypsin‐type serine proteases.

Characterization of rOryzin MA0196 and wild‐type Tryp MA0196

Effects of pH and temperature, and pH and thermal stability

The optimum temperatures for rOryzin MA0196 and Tryp MA0196 activity with casein as substrate were 40–45 °C (Supporting Information, Fig. S3). rOryzin MA0196 and Tryp MA0196 were stable at up to 40 and 30 °C, respectively, with an incubation time of 30 min (Supporting Information, Fig. S3). The optimum pH values for activity of rOryzin MA0196 and Tryp MA0196 were 7.0–7.5 (Supporting Information, Fig. S4). The pH stability ranges were 6.0–12.0 for rOryzin MA0196 and 5.0–11.0 for Tryp MA0196 (Supporting Information, Fig. S4). For comparison, Oryzin from A. nidulans HA‐10 retained stability at pH 6.0–10.0 and at temperatures up to 50 °C. ¹⁶ A hydrogen peroxide‐stable alkaline serine protease from A. flavus (similar to Oryzin from A. flavus NRRL 3357 based on sequence analysis) was active at pH 5.0–10.5, with the optimum at pH 10, and exhibited maximum activity at 45 °C. ¹⁸ Similarly, an alkaline serine protease from A. nidulans PW1, closely related to Oryzin from A. nidulans FGSC A4, exhibited maximum activity at pH 8.5 and 40 °C. ¹⁹

Effects of protease inhibitors on enzyme activity

rOryzin MA0196, like the Aspergillus alkaline serine proteases described in earlier literature, ¹⁷ , ¹⁸ , ¹⁹ was completely inhibited by phenylmethylsulfonyl fluoride (PMSF) (Table 3). Tryp MA0196 was not inhibited by PMSF, but its inhibition profile with other protease inhibitors, such as 4‐(2‐aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF), aprotinin, and tosyl‐l‐lysyl‐chloromethane hydrochloride (TLCK), was similar to that of the serine protease from F. oxysporum var. lini (residual activity with AEBSF, 14%; aprotinin, 20%; TLCK, 60%). ²⁴

Table 3.

Effect of protease inhibitors on rOryzin MA0196 and wild‐type Tryp MA0196 activity

	Remaining activity (%)
Inhibitor	rOryzin	Tryp
AEBSF	93.0 ± 1.0c	20.4 ± 1.6d
Aprotinin	93.8 ± 3.2c	25.2 ± 1.3d
Benzamidine HCl	99.2 ± 1.2ab	81.7 ± 1.4a
Leupeptin	96.3 ± 1.5bc	31.4 ± 1.8c
PMSF	< 1%d	60.5 ± 1.9b
TLCK HCl	100.8 ± 2.5a	34.4 ± 1.3c

Note: Purified recombinant Oryzin MA0196 or wild‐type Tryp MA0196 was incubated with 5 mmol L⁻¹ protease inhibitor in 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5) at 4 °C for 1 h. Residual enzyme activity (%) toward casein was then measured. Values are expressed as means ± SDs (n = 5). Different letters indicate significant differences within each column. P < 0.05, one‐way analysis of variance (ANOVA) followed by Fisher's protected least significant difference (PLSD) test. AEBSF, 4‐(2‐aminoethyl)benzenesulfonyl fluoride hydrochloride; PMSF, phenylmethylsulfonyl fluoride; TLCK, tosyl‐l‐lysyl‐chloromethane hydrochloride.

Substrate specificity toward protein and synthetic substrates

Substrate specificity toward protein substrates indicated that, among the tested proteins, rOryzin MA0196 preferentially hydrolyzed casein (Table 1). Consistently, an alkaline protease from A. sojae GIM3.33 (homologous to Oryzin from A. oryzae) exhibited 15.8% activity toward BSA relative to its activity toward casein. ²⁵ Tryp MA0196 exhibited lower activity toward casein than rOryzin MA0196, but showed relatively broad activity toward various proteins (Table 1). Although Tryp MA0196 was purified to homogeneity from crude enzyme preparations, the final preparation may have contained trace amounts of exopeptidase(s). Its activity was therefore tested further using synthetic substrates for aminopeptidase, carboxypeptidase, and dipeptidyl peptidase.

Tryp MA0196 exhibited detectable activity toward the trypsin‐like substrate Bz‐L‐Arg‐pNA (Table 2). However, its specific activity (0.032 ± 0.004 U mg⁻¹) was much lower than that of a trypsin‐like serine protease from F. oxysporum var. lini (6.3 U mg⁻¹) ²⁴ and was comparable to that of another F. oxysporum serine protease that was reported to lack activity toward Bz‐L‐Arg‐pNA. ²² These results suggest that Tryp MA0196 may possess limited trypsin‐like substrate specificity despite its classification as a chymotrypsin‐like protease. Further analysis using substrates specific for chymotrypsin‐like proteases would provide additional insight into its substrate specificity. Although determination of kinetic parameters such as K _m and k _cat would provide further insight into enzyme characteristics, the present study focused on comparative evaluation of substrate specificity and environmental tolerance. Detailed kinetic analysis will be addressed in future studies.

Effect of NaCl on enzyme activity

Aspergillus sydowii MA0196 is one of the predominant molds involved in katsuobushi fermentation, in which water activity is low. ²⁶ Accordingly, its extracellular hydrolytic enzymes, including rOryzin MA0196 and Tryp MA0196, are expected to function under moderately saline conditions. At 7.5% (w/v) NaCl (1.28 M), rOryzin MA0196 retained approximately 50% of its activity compared with its activity in the absence of NaCl (Fig. 2), similar to oryzins from A. oryzae and A. sojae used in soy sauce fermentation ²⁷ and the alkaline serine protease from A. flavus. ¹⁸ Tryp MA0196 retained relatively high activity – up to 12.5% (w/v) NaCl (2.14 mol L⁻¹) – although activity decreased at higher concentrations (Fig. 2).

Figure 2.

Effects of NaCl, glycerol, and histidine on the activity of recombinant Oryzin MA0196 (rOryzin MA0196) and wild‐type Tryp MA0196 from Aspergillus sydowii MA0196. (A) NaCl concentration 0–20% (w/v), 0–3.42 mol L⁻¹, (B) glycerol concentration 0–50% (v/v), 0–6.8 mol L⁻¹, and (C) histidine concentration 0–10% (w/v), 0–0.48 mol L⁻¹. Enzyme assays were performed using casein as the substrate in 50 mmol L⁻¹ Tris–HCl buffer (pH 7.5) containing the indicated additives. Relative activity toward casein was calculated based on specific activity in standard conditions (16 and 5.4 U mg⁻¹ for rOryzin MA0196 and Tryp MA0196, respectively). The activity of wild‐type Tryp MA0196 toward Nα‐benzoyl‐DL‐arginine p‐nitroanilide (Bz‐L‐Arg‐pNA) was also determined.

Salt‐tolerant serine proteases have been reported predominantly from bacteria, particularly Bacillus species, and from halophilic archaea such as Haloarcula and Halococcus. ⁶ Reports from filamentous fungi are relatively scarce, although Penicillium citrinum YL‐1 was reported to secrete a salt‐tolerant alkaline serine protease ²⁸ and A. oryzae 3.042 was shown to produce salt‐tolerant neutral and acid proteases. ⁸ The salt‐tolerant properties of rOryzin MA0196 and, in particular, the superior NaCl tolerance of Tryp MA0196, highlight the potential of xerophilic Aspergillus species as sources of robust proteases for high salt‐fermentation processes.

The high salt tolerance of Tryp MA0196 may be attributable to its amino acid composition. An increased proportion of acidic residues on the protein surface promotes ion–water binding interactions, stabilizing hydration shells and maintaining enzyme solubility and activity in saline environments. ⁶ The calculated ratio of acidic to basic residues, A/B = (Glu + Asp)/(Lys + His + Arg), ²⁹ was 1.25 for Oryzin MA0196 and 2.0 for Tryp MA0196, indicating a greater abundance of acidic residues in Tryp MA0196 (Supporting Information, Table S3). This structural feature is consistent with its superior salt‐tolerance and adaptation to the high‐salt, low‐water‐activity environment of katsuobushi fermentation.

Effects of glycerol and histidine on enzyme activity

Both rOryzin MA0196 and Tryp MA0196 maintained strong hydrolytic activity in low‐water‐activity conditions simulated by the addition of glycerol (Fig. 2). Although increasing glycerol concentration can create an environment favorable for peptide synthesis, it may also reversibly inhibit hydrolysis by limiting water availability. ³⁰ Although enzyme activity decreased with increasing glycerol concentration, both MA0196 proteases retained measurable activity even in 50% (v/v) glycerol (6.8 M), suggesting partial adaptation to xerophilic environments.

Histidine is one of the major free amino acids accumulated during katsuobushi fermentation, ¹² and high concentrations may interfere with proteolysis through competitive interactions or pH buffering effects. In this context, the observed tolerance of both MA0196 proteases to histidine may reflect adaptation to such environments, although further evidence is required to confirm this relationship. However, both rOryzin MA0196 and Tryp MA0196 were unaffected by high histidine concentrations (up to 100 mmol L⁻¹), indicating that these enzymes are tolerant to conditions mimicking the katsuobushi fermentation matrix. This characteristic further supports their functional relevance in the proteolytic degradation of katsuobushi proteins in complex physicochemical environments.

Evaluation of rOryzin MA0196 and wild‐type Tryp MA0196 from A. sydowii MA0196 for application in meat processing

Fungal alkaline proteases have diverse industrial applications, particularly in the detergent and food industries. ⁴ To further explore the functional properties of Oryzin MA0196 and Tryp MA0196 in conditions relevant to katsuobushi fermentation and food processing, this study examined their catalytic behavior.

Analysis using SDS‐PAGE (Fig. 3) indicated that Tryp MA0196 was more effective than rOryzin MA0196 in hydrolyzing katsuobushi proteins, particularly the coiled‐coil region of the myosin heavy chain tail. As Table 2 reports, the final preparation of Tryp MA0196 did not contain detectable exopeptidase activity. The observed hydrolysis can therefore be attributed to Tryp MA0196 itself, which efficiently degraded katsuobushi proteins into peptides with molecular masses of less than 17 kDa (Fig. 3). Table 1 shows that the specific activity of rOryzin MA0196 toward katsuobushi proteins was slightly higher than that of Tryp MA0196. However, in the experiments shown in Fig. 3, enzyme dosages were adjusted based on casein‐specific activity to allow a fair comparison. Consequently, approximately threefold more Tryp MA0196 was applied, which accounts for its apparently greater hydrolytic effect in Fig. 3. Although heterologous expression of Tryp MA0196 was not successful in the tested expression systems, developing an alternative expression system using A. oryzae or other koji molds could enable efficient enzyme production. Such a system would not only provide sufficient enzyme yields for practical applications but would also allow direct enzymatic hydrolysis of katsuobushi byproducts (dashigara). This approach could facilitate the valorization of underused fish resources by producing functional peptides with potential applications in the development of seasoning and in meat processing. ³¹ , ³²

Figure 3.

Electrophoretic analyses of hydrolysis of soluble katsuobushi proteins by endopeptidases from A. sydowii MA0196. (A) SDS–PAGE analysis using purified rOryzin MA0196. (B) SDS–PAGE analysis using wild‐type Tryp MA0196. (C) Tricine–SDS–PAGE analysis using purified rOryzin MA0196. (D) Tricine–SDS–PAGE analysis using wild‐type Tryp MA0196. Aliquots were collected after proteolysis for 0, 1, 3, and 24 h, mixed with sample buffer, and subjected to electrophoresis as described in the text. Lanes: BM, broad‐range protein marker; LM, low‐range protein marker; PM, peptide marker; C, control.

Fungal alkaline serine proteases, similar to bacterial proteases, have been applied widely in meat tenderization because of their ability to hydrolyze connective tissue and muscle fibers, improving the tenderness and palatability of beef while shortening the aging period. ³ , ³³ , ³⁴ To contextualize the proteolytic potential of rOryzin MA0196 and Tryp MA0196, their activity was compared with that of widely used plant‐derived meat tenderizers, such as papain, bromelain, actinidin, and zingibain. ³⁵

In the sarcoplasmic fraction, rOryzin MA0196 exhibited limited hydrolysis of intact actin (approximately 44 kDa) generating several smaller fragments, and Tryp MA0196 showed no detectable activity toward actin (44 kDa), myosin light chain (29 kDa), or troponin C (20 kDa) (Fig. 4). Commercial papain and actinidin readily hydrolyzed myosin heavy chain‐1 and actin, whereas bromelain displayed only partial activity toward these proteins. ³⁵ In the myofibrillar fraction, rOryzin MA0196 was more effective than Tryp MA0196 in degrading intact myosin heavy chains (200 kDa); residual fragments of approximately 100 kDa remained after Tryp MA0196 treatment for 24 h (Fig. 5). These findings, together with the results for katsuobushi proteins (Fig. 3), suggest that rOryzin MA0196 preferentially hydrolyzes intact fibrillar proteins, whereas Tryp MA0196 is more effective against physicochemically fragmented or partially denatured proteins, such as those generated during katsuobushi processing. In collagen hydrolysis, both enzymes exhibited activity toward type I collagen after 1 h of incubation, and Tryp MA0196 was slightly more effective than rOryzin MA0196 in degrading type I collagen (Fig. 6). Plant‐derived papain, bromelain, and actinidin hydrolyzed collagen type I chain A (approximately 260 kDa), but not chain B (approximately 130 kDa). ³⁵

Figure 4.

Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) analysis of beef sarcoplasmic proteins after hydrolysis with rOryzin MA0196 (A) or wild‐type Tryp MA0196 (B) from Aspergillus sydowii MA0196. Aliquots were collected after 0, 1, 3, and 24 h of proteolysis and subjected to SDS‐PAGE as described in the text.

Figure 5.

Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) analysis of beef myofibrillar proteins after hydrolysis with rOryzin MA0196 (A) or wild‐type Tryp MA0196 (B). Aliquots were collected after proteolysis for 0, 1, 3, and 24 h and subjected to SDS–PAGE.

Figure 6.

Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) analysis of type I bovine skin collagen after hydrolysis with rOryzin MA0196 (A) or wild‐type Tryp MA0196 (B). Aliquots were collected after proteolysis for 0, 1, 3, and 24 h and subjected to SDS–PAGE.

Together, these results indicate that rOryzin MA0196 is more effective than Tryp MA0196 in degrading beef muscle proteins, particularly myofibrillar proteins and collagen (Figs. 5 and 6), highlighting its potential utility as a meat‐tenderizing enzyme. Approximately 7 mg of purified rOryzin MA0196 can be obtained from 30 mL of culture supernatant using one‐step ion‐exchange chromatography (Supporting Information, Table S2). Analysis with SDS‐PAGE (Supporting Information, Fig. S2) indicated that rOryzin MA0196 was the predominant protein in the culture supernatant, and acetone precipitation provided an alternative, simplified, purification route. In contrast, purification of wild‐type proteases such as Tryp MA0196 requires multiple chromatographic steps. ⁴ Reported applications of alkaline serine proteases from Aspergillus spp. include fibrillar protein degradation, ³⁶ dehairing, ³⁷ detergent formulations, ³⁸ and keratin hydrolysis. ³⁹ This appears to be the first demonstration that an Oryzin‐type alkaline serine protease can contribute to beef tenderization, suggesting a potential extension of their functional repertoire. Plant proteases such as papain, bromelain, and ficin have long been studied as meat tenderizers, ⁴⁰ although their collagenolytic activity is mainly limited to type I collagen chain A. ³⁵ In this context, rOryzin MA0196 may serve as a complementary candidate for future industrial applications in meat processing.

CONCLUSIONS

This study characterized two major endopeptidases, rOryzin MA0196 and wild‐type Tryp MA0196, from the xerophilic fungus A. sydowii MA0196, highlighting xerophilic molds as underexplored yet valuable sources of robust proteases. rOryzin MA0196 exhibited strong hydrolytic activity toward intact myofibrillar and collagen proteins, underscoring its potential as a meat‐tenderizing enzyme for food processing applications. In contrast, Tryp MA0196 showed superior activity toward physicochemically modified myosin‐derived proteins in katsuobushi (Fig. 3), suggesting a specific role in katsuobushi protein degradation and the valorization of katsuobushi byproducts (dashigara). Overall, the results of this study demonstrate that rOryzin MA0196 is a promising candidate for meat tenderization, and Tryp MA0196 – distinct among filamentous fungal proteases – may serve as both an indicator enzyme for selecting functionally suitable katsuobushi molds and a tool for the efficient use of fishery byproducts, contributing to sustainable food processing and resource recycling.

AUTHOR CONTRIBUTIONS

Shinji Takenaka: conceptualization, methodology, investigation, formal analysis, bioinformatics, writing – original draft, reviewing and editing. Yasuhiro Oribe: conceptualization, methodology, data collection, writing – original draft. Jun‐ichi Matsumoto and Mikiharu Doi: conceptualization, reviewing – original draft. All authors reviewed and approved the final manuscript.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

Supporting information

Figure S1. Multiple sequence alignment of (A) Oryzin MA0196 from A. sydowii MA0196 with fungal and insect trypsin proteases, and (B) Tryp MA0196 from A. sydowii MA0196 with Aspergillus serine proteases. The catalytic triad Glu‐His‐Ser residues in Aspergillus Oryzins and His‐Asp‐Ser residues in fungal and insect Tryps were shown in red. Three disulfide bonds (Cys45‐Cys61, Cys171‐Cys187, and Cys198‐Cys222) and two N‐glycosylation motifs (130Asn‐Ala‐Ser132 and 153Asn‐Val‐Ser155) in Tryp MA0196 were shown in yellow and cyan, respectively.

Figure S2. SDS‐PAGE analysis confirming the purity of (A) rOryzin MA0196 expressed in P. pastoris transformant and (B) wild‐type Tryp MA0196 purified from A. sydowii MA0196. Protein samples from each purification step were separated on a 12.5% acrylamide gel using an ATTO AE6530 electrophoresis system (ATTO Corp.) according to the manufacturer's instructions. After electrophoresis, gels were stained with EzStain Aqua or by the periodic acid–Schiff (PAS) method. (A) Lanes: BM, protein molecular mass markers (broad range, 6.5–200 kDa; Takara Bio); 1, culture supernatant; 2, active fraction from DEAE‐Toyopearl chromatography; LM, protein molecular mass markers (low range, 14.3–97.2 kDa; Takara Bio). (B) Lanes: BM, protein molecular mass markers (broad range); 1, crude enzyme preparation; 2, dialyzed enzyme solution; 3, active fraction from DEAE‐Toyopearl chromatography; 4, active fraction from Butyl‐Toyopearl chromatography; LM, protein molecular mass markers (low range); 5, purified Tryp band stained by PAS reagent.

Figure S3. Effect of temperature on enzyme activity and thermostability. (A) Optimal temperatures of rOryzin MA0196 and wild‐type Tryp MA0196. Enzyme activity was assayed in 20 mmol L⁻¹ Tris–HCl buffer (pH 7.5) across a temperature range of 15–70 °C. The specific activities of rOryzin MA0196 and Tryp MA0196 at 40 °C were 23.6 and 6.9 U/mg, respectively, and relative activities were calculated based on these values. (B) Residual activity after 30‐min incubation. To evaluate thermostability, purified enzymes were pre‐incubated in 20 mmol L⁻¹ Tris‐HCl buffer (pH 7.5) at 15–70 °C for 30 min. The specific activities of rOryzin MA0196 and Tryp MA0196 at pH 7.5 were 16 and 5.4 U/mg, respectively, and relative activities were calculated accordingly.

Figure S4. Effect of pH on enzyme activity and pH stability. (A) Optimal pH values. Enzyme activity of purified rOryzin MA0196 and Tryp MA0196 was assayed in 50 mmol L⁻¹ buffer systems at different pH values: sodium acetate (pH 4.0–6.0), sodium potassium phosphate (pH 6.0–7.5), Tris–HCl (pH 7.5–9.0), and glycine‐NaOH (pH 9.0–12.0). The specific activities of rOryzin MA0196 and Tryp MA0196 at pH 7.5 were 16 and 5.4 U/mg, respectively, and relative activities were calculated based on these values. (B) Residual activity after 1‐h incubation at various pH values. To evaluate pH stability, purified enzymes were dialyzed overnight at 4 °C against disodium phosphate–citric acid buffers of varying pH. Following dialysis, residual activity was measured in 50 mmol L⁻¹ Tris–HCl buffer (pH 7.5). The specific activities of rOryzin MA0196 and Tryp MA0196 at pH 7.5 were 16 and 5.4 U/mg, respectively, and relative activities were calculated accordingly.

Table S1. Oligonucleotides used in this study Primer Sequence (5′ to 3′).

Table S2. Purification summary of recombinant Oryzin MA0196 expressed in Pichia pastoris transformant and wild‐type Tryp MA0196 from A. sydowii MA0196.

Table S3. Amino acid composition, A/B ratio, and theoretical isoelectric point (pI) of Oryzin MA0196 and Tryp MA0196 from A. sydowii MA0196.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.