Purification and characteristics of a novel milk-clotting metalloprotease from Bacillus velezensis DB219

Milk-clotting enzyme (MCE) is the essential active agents in dairy processing. The traditional MCE is mainly obtained from animal sources, in which calf ren-net is the most widely used in cheese industry. Traditional MCE substitute is becoming necessary due to its limited production and increased cheese consumption. A novel traditional MCE substitute was produced from Bacillus velezensis DB219 in this study. The DB219 MCE exhibited a notable specific activity of 6,110 Soxhlet units/mg and 3.16-fold purification yield with 28.87% recovery through ammonium sulfate fractionation and DEAE-Sepharose Fast Flow. The purified DB219 MCE was a metalloprotease with a molecular weight of 36 kDa. The DB219 MCE was weak acid resistance and stable at pH 6.0 to 10.0 and temperature <45°C. The highest milk-clotting activity was observed in substrate at pH 5.5 added with 20 to 30 m M CaCl 2 . The Michaelis constant and maximal velocity for casein were 0.31 g/L and 14.22 μmol/min. The DB219 MCE preferred to hydrolyze β-casein instead of α-casein. The DB219 MCE hydrolyzed α-casein, β-casein, and κ-casein to generate significantly different peptides in comparison with calf rennet and ES6023 MCE (fungal MCE) through SDS-PAGE and reversed-phase HPLC analysis. The DB219 MCE mainly cleaved Thr124-Ile125 and Ser104-Phe105 bonds in κ-casein and had unique casein cleavage sites and peptide composition through LC-MS/MS analysis. The DB219 MCE was potential to be a new milk coagulant and enriched kinds of traditional MCE substitute.


INTRODUCTION
Milk-clotting enzyme (MCE) plays an important role in food and dairy processing, especially cheese production.The MCE leads to milk clotting and influences cheese quality (e.g., yield, texture, and flavor; Liburdi et al., 2018).The traditional MCE (e.g., calf rennet, lamb MCE, and camel MCE) is mainly from animal and calf rennet and is the most commonly used in the cheese industry (Kumari et al., 2016).However, traditional MCE substitutes are becoming necessary and attracting great interest due to limitation in the humanistic morality, growth cycle, and production cost of traditional MCE.
The traditional MCE substitutes are mainly obtained from plants and microorganisms.Most plant-derived MCE show high proteolytic activity (PA) and hydrolyze casein to generate many bitter peptides, which is a limitation in cheese making (Salehi et al., 2017).Microbial MCE has attracted much attention as potentially traditional MCE substitutes in the recent years.Most fungal MCE has low heat-sensitivity, which leads to high residual enzyme activity and bitterness.A few MCE (e.g., Rhizomucor miehei and Rhizomucor pusillus) are reported to be used in cheese making (Liu et al., 2021).Bacterial MCE is superior to fungal MCE in production cost, biodiversity, gene modification, material usage, and fermentation control (Wehaidy et al., 2018).Moreover, some bacterial MCE has high milkclotting activity (MCA) and suitable characteristics for cheese production (da Silva et al., 2019).However, the reports of bacterial MCE are less than those of fungal MCE.There is little research on cleavage sites of casein and hydrolytic peptide composition on bacterial MCE.Therefore, it is necessary to find various bacterial MCE for development of traditional MCE substitutes and making different cheeses.
In this study, a novel DB219 MCE produced by Bacillus velezensis DB219 isolated from soil was found and identified a kind of metalloprotease.The DB219 MCE was purified and had notable specific activity, high MCA, weak acid resistance, and strong affinity to the casein substrate.The unique casein hydrolysis, cleavage sites, function peptides and peptide composition indicated DB219 MCE was potential to be a new milk coagulant as traditional MCE substitute.

MATERIALS AND METHODS
No human or animal subjects were used, so this analysis did not require approval by an Institutional Animal Care and Use Committee or Institutional Review Board.

Determination of MCA
The MCA was measured by the method of Arima et al. (1970), which was according to the appearance of first discontinuous particles and expressed in Soxhlet units (SU).The enzyme solution was incubated in a 35°C for 10 min and added to skim milk (10% (wt/vol), 10 mM CaCl 2 ), which was pre-incubated at 35°C for 5 min.The milk clotting was stopped when discontinuous particles were formed.The calculation formula is as follows: SU = (2,400 × V S × N)/(T × V E ), where V S = skim milk volume (mL); N = MCE dilution multiple; T = milk-clotting time (s); and V E = MCE volume (mL).

Strain and Crude MCE Preparation
Bacillus velezensis DB219 (CGMCC No. 24624) was a MCE-producing strain and isolated from soil samples in dairy cattle pastures in our previous study (Zhang et al., 2022).The seed liquid of the strain was activated in the improved tryptone yeast cystine medium and inoculated with 5% (vol/vol) into the fermentation medium containing wheat bran (60 g/L).It was cultured at 37°C and 180 rpm.The broth was centrifuged at 6,000 × g and 4°C for 20 min to obtain the MCE in the supernatant.Ammonium sulfate (60% saturation) was added to the supernatant and precipitated at 4°C for 3 h.The precipitate was centrifuged at 6,000 × g and 4°C for 20 min.Then, it was dissolved in Tris-HCl buffer (20 mM and pH 7.0).The crude MCE solution was packed into dialysis bags (8-14 kDa), dialyzed at 4°C with stirring and lyophilized to powder for further purification.

MCE Purification
The crude MCE was dissolved in 20 mM Tris-HCl buffer of pH 7.0.The fully dissolved crude MCE was filtrated by a 0.22 μm membrane filter and injected into DEAE-Sepharose Fast Flow column (5 × 5 mL) through AKTA pure (Cytiva).The column was preequilibrated with Tris-HCl (20 mM, pH 8.5) buffer before injection.The samples were eluted with Tris-HCl (20 mM, pH 8.5) buffer and eluted linearly with Tris-HCl buffer containing 0-1.2 M NaCl at a flow rate of 0.5 mL/min.The absorbance of each tube at 280 nm was recorded and the MCA was determined.The higher MCA fractions were pooled, dialyzed, lyophilized, and stored at −20°C for subsequent experiments.

MCE Purity and Concentration Determination
The molecular mass and purity of MCE were measured through SDS-PAGE (12%) according to the method of Hang et al. (2016) with some modification.The concentration was measured at 280 nm and determined by Bradford method (Bradford, 1976).Briefly, different concentrations of BSA solution were used to make standard curve.Bradford reagent (200 uL) was reacted with 20 uL of sample for 10 min and determined at 595 nm.The protein concentration was calculated according to the BSA standard curve.

Effect of pH on MCE Stability
The DB219 MCE, calf rennet, and ES6023 MCE solutions were adjusted to pH 3.0 to 13.0 with 0.1 M NaOH and HCl, respectively.Each solution was incubated at 25°C for 40 min.The change of MCA was determined and compared after incubation and adjusting MCE solutions to pH 7.0.The MCA (pH 7.0) was defined as the control.

Effect of CaCl 2 on MCE
Skim milk solutions at intervals of 10 mM CaCl 2 (0-100 mM) were taken as substrates to detect the changes in MCA of DB219 MCE, calf rennet and ES6023 MCE, respectively.The MCA (10 mM CaCl 2 ) was defined as control.

Effect of Milk Substrate pH on MCE
The skim milk solutions of pH 5.5, 6.0, 6.5, 7.0, and 7.5 were taken as the substrates respectively to determine the changes in MCA of DB219 MCE, calf rennet, and ES6023 MCE, respectively.The MCA (skim milk pH 6.5) was defined as control.

Effect of Temperature on DB219 MCE
The DB219 MCE was incubated at 35-60°C (intervals of 5°C) for 10-50 min (intervals of 10 min) with 10 mM CaCl 2 skim milk solution as the substrate to detect thermal stability.The MCA without incubation of DB219 MCE was defined as 100%.The optimal temperature for DB219 MCE was determined by incubating the 10 mM CaCl 2 skim milk solution at 30-85°C (intervals of 5°C) and measuring changes in MCA.

DB219 MCE Toward Protease Inhibitors
The PMSF of 1 and 2 mM, EDTA of 1, 5, 10, and 25 mM, pepstatin A of 0.01, and 0.02 and 1 mM DTT were added to DB219 MCE with the same MCA, respectively, according to the report of Salehi et al. (2017).Each solution was incubated at 25°C for 30 min and measured MCA.The MCA without protease inhibitor treatment was defined as control.

Kinetic Measurement for DB219 MCE
The Michaelis-Menten constant was measured by the method of He et al. (2011) with a little modification.The casein solution at 0.3, 0.4, 0.7, 1.0, and 2.0 g/L was used as the substrate for the measurement of PA according to the method of Arima et al. (1970).Michaelis constant (K m ) and maximal velocity (V m ) were calculated according to double-reciprocal plot of Lineweaver-Burk equation (Lineweaver and Burke, 1934).

Identification of Cleavage Sites and Peptide Composition
The cleavage sites and peptides from α-CN, β-CN, and κ-CN hydrolysis caused by DB219 MCE were identified through the method of Meng et al. (2021) and Hang et al. (2016) with some modification.The mass spectrum data were retrieved by ProteinPilot (V4.5) and algorithm was Paragon.The MS/MS data were searched through bovine protein database of MASCOT (http: / / www .matrixscience.com).The peptide distribution of α-CN, β-CN, and κ-CN hydrolyzed by DB219 MCE was aligned through COBLAST (https: / / www .ncbi.nlm.nih.gov/tools/ cobalt/ ) and charted by Adobe Illustrator CS6 and PEAKS Studio 8.5.The functional peptides were obtained through peptide composition analysis and aligning functional peptides in casein reported in the previous reports.

Statistical Analysis
Every experiment was carried out in triplicate.The results were expressed as means ± standard deviations.The obtained data were analyzed by the software SPSS 17.0 (version 17.0).

MCE Preparation and Purification
As shown in Figure 1, the crude MCE showed 4 protein peaks at 280 nm through the separation of DEAE-Sepharose Fast Flow.Only the second peak had MCA.The mature DB219 MCE was not combined with DEAE-Sepharose Fast Flow and eluted with Tris-HCl (20 mM and pH 8.5).The MCE in the second peak exhibited a single molecular weight of 36 kDa band through SDS-PAGE gel analysis (Supplemental Figure S1, http: / / dx .doi.org/ 10 .17632/8fkj2zmkfy .1;Zhang, 2023).Table 1 shows the crude DB219 MCE was purified 3.16-fold and the specific activity reached 6,110 SU/mg after (NH 4 ) 2 SO 4 precipitation and DEAE-Sepharose FF separation.
The DB219 MCE behaved higher MCA and better purification results compared with earlier reports.The pure DB219 MCE was 6,110 SU/mg of specific activity, which was higher than Bacillus licheniformis BL312 (5,291 SU/mg; Zhang et al., 2019a) and Bacillus subtilis natto MCE (12 SU/mg; Wu et al., 2013).The DB219 MCE showed higher purification fold and recovery rate in comparison with Bacillus subtilis MTCC 10422 (Kumari et al., 2016) and Withania coagulans fruit (Salehi et al., 2017).The fast purification were necessary for MCE production.High specific MCA accelerated milk clotting and cheese manufacture.Moreover, it is interested that mature DB219 MCE was likely not combined with ion-exchange column and other useless protein bound to the column.This phenomenon was different from most reported MCE under similar purification parameters (Cavalcanti et al., 2004;Yu et al., 2010).This characteristic of DB219 MCE was beneficial to enhance purification efficiency (Ding et al., 2012).Meanwhile, it is indicated that DB219 MCE had high isoelectric point and was potential to be traditional MCE substitute because of high MCA and great purification.

MCE Stability and MCA Toward pH
Figure 2A showed that the DB219 MCE was stable at a similar pH range (6.0-10.0) to ES6023 MCE and at a wider pH range than calf rennet (4.0-7.0).The DB219 MCE exhibited the greatest stability at a higher pH (8.0) than calf rennet (6.0) and ES6023 MCE (6.0).At pH 5.0-6.0, the MCA of DB219 MCE was kept about 67-87%.The MCA significantly decreased at pH <5.0 or pH >10.0 and almost disappeared at pH <4.0 or pH >12.0.
The ionic stability of enzyme in acidified colloidal solution affected enzyme usage.The milk substrate pH was adjusted to 6.0-6.5 before cheese manufacture, which meant the MCE kept high MCA at pH 6.0-6.5 (Ding et al., 2012).Poor cheese yield and cheese quality caused by high pH sensitization of MCE (Hang et al., 2016).The DB219 MCE showed excellent pH stability at pH 6.0-10.0,indicating its applicability for cheese manufacture.The cheese is usually in the acid condition during ripening.The DB219 MCE activity decreased enormously at pH <5.0, which led to low residual PA and bitter peptides from α-CN and β-CN hydrolysis (Merheb-Dini et al., 2010).This characteristic (pH <5.0) was similar to calf rennet and ES6023 MCE.

MCA Toward CaCl 2 and Milk Substrate pH
The DB219 MCE exhibited the maximal MCA at 20-30 mM CaCl 2 , which was similar to ES6023 MCE (30 mM) and lower than calf rennet (40 mM; Figure 2B).The maximal MCA of DB219 MCE was 1.2-fold higher than that of the control.As shown in Figure 2C, the effect of milk substrate pH on DB219 MCE was  similar to that on calf rennet and ES6023 MCE.They showed the maximal MCA at milk substrate pH 5.5 and decreased below this pH value.
It is necessary to use Ca 2+ for the cheese manufacture.The MCE could be stabilized by Ca 2+ through strengthening interaction among binding sites of MCE molecule (Guo et al., 2019).Moreover, Ca 2+ played a bridge role in casein micelles and promoted forming milk curd (Alexandraki and Moatsou, 2018).Therefore, 20-30 mM CaCl 2 was selected to add to milk substrate during milk clotting.In addition, the pH of fresh milk substrate was generally 5.5-6.5 (Rangnoi et al., 2014).It is indicated that DB219 MCE showed high MCA immediately and shortened milk-clotting time during cheese making.It was important characteristic for DB219 MCE as traditional MCE substitute to apply in cheese production.

DB219 MCE Toward Temperature
Figure 3A indicated that the DB219 MCE was stable at 35-45°C, but decreased at above 50°C.The MCA decreased by 44% at 55°C for 20 min and 77% at 60°C for 10 min.The DB219 MCE was almost inactive at 60°C for 30 min.As shown in Figure 3B, the optimal temperature for DB219 MCE exhibiting maximal activ-ity was 75°C.The DB219 MCE had similar optimal temperature with Bacillus subtilis MCE (Kumari et al., 2016).The fresh milk substrate after pasteurization was clotted through MCE at a temperature range of 35°C to 75°C (Sun et al., 2014), which laid in the optimal temperature of DB219 MCE.Moreover, the great MCE was sensitive to temperature, which could inhibit PA by regulating temperature (Sun et al., 2014).It could decrease the negative effects of residual MCE on cheese flavor and texture during ripening (Kumari et al., 2016).The DB219 MCE possessed weak thermal stability, indicating a good characteristic as commercial MCE.

DB219 MCE Toward Metal Ions
Figure 3C shows that Ca 2+ , Co 2+ , Mn 2+ , and Li + significantly increased the DB219 MCE MCA to about 109, 124, 128, and 193% at 0.25 M, respectively.The 0.25 M K + , Zn 2+ , and Mg 2+ had no influence on DB219 MCE.The Mn 2+ and Co 2+ increased the MCA by 7% and 70% at 0.5 M, respectively.The other kinds of metal ions decreased the MCA at both 0.25 and 0.5 M. According to the report of Hashem (2000), Mn 2+ , Li + , and Co 2+ could enhance MCA of Penicillium oxalicum MCE, which was similar to our results.It is noted that low Ca 2+ concentration increased the MCA of DB219 MCE.However, the MCA decreased at 0.5 M Ca 2+ .It might be due to the entire neutralization of negative charged residues on the external casein micelles.The Ca 2+ created both iso-electric conditions and ion bridges (Sun et al., 2014).Sodium chloride played an important role in cheese manufacture, which affected microbial growth, moisture content, enzyme activity, and protein variety in cheese (Salehi et al., 2017).The 0.25 M NaCl was used in making Camembert cheese (Mane and Mcsweeney, 2020).The NaCl was beneficial to inhibited MCE activity through changing conformation of casein.The DB219 MCE inhibited by Na + (Figure 3C), indicating its feasibility for cheese production.

Kinetic Parameters of DB219 MCE
As indicated in Figure 4, a linear regression curve is generated from a Lineweaver-Burk double-reciprocal plot of velocity versus substrate amount with intercepts on the x axis (−K m ) and y axis (1/V m ), from which K m and V m can be calculated.The K m and V m values of the DB219 MCE for casein were 0.31 g/L and 14.22 μmol/ min at 35°C and pH 6.0, respectively (Figure 4B).The K m represents the affinity of the enzyme and the substrate.A high K m indicates a low affinity and vice versa (Ritchie and Prvan, 1996).The K m of DB219 MCE was lower than that of B. amyloliquefaciens D4 MCE (0.471 g/L; He et al., 2011), Paenibacillus sp.BD3526 MCE (1.36 g/L; Hang et al., 2016) and Bacillus subtilis MTCC 10422 MCE (5 g/L; Kumari et al., 2016).It is indicated that DB219 MCE had a higher affinity to the casein substrate, which led to shorter milk-clotting time for cheese manufacture.

Hydrolysis Characteristics of DB219 MCE
Figure 5 showed the RP-HPLC chromatograms of peptides from casein hydrolysis.The α-CN, β-CN, and κ-CN hydrolysis levels for various MCE were different.The α-CN degradation level hydrolyzed by DB219 MCE was weaker than that by ES6023 MCE, but higher than that for calf rennet.The α-CN hydrolysates at about 30 min were significantly different for ES6023 MCE (Figure 5B).The DB219 MCE hydrolyzed β-CN to a greater degree than ES6023 MCE and calf rennet (Figure 5D).For κ-CN hydrolysis, ES6023 MCE was slightly higher than DB219 MCE and calf rennet.The DB219 MCE exhibited similar hydrolysis level of κ-CN to calf rennet (Figure 5F).Sodium dodecyl sulfate-PAGE analysis was shown in Figure 6.The α-CN, β-CN, and κ-CN degradation levels hydrolyzed by DB219 MCE, calf rennet, and ES6023 MCE were in accordance with the RP-HPLC results.The hydrolysis level was getting weaker with the increase of α-CN: MCE, β-CN: MCE, and κ-CN: MCE ratios.The hydrolysis of DB219 MCE and ES6023 MCE was κ-CN > β-CN > α-CN.The calf rennet hydrolyzed κ-CN much more than α-CN and β-CN.The casein hydrolysates were significantly different in the same casein: MCE ratio.The DB219 MCE hydrolyzed α-CN to produce mainly about 17 and 25 kDa, which were similar to calf rennet except hydrolysates of 25 kDa and different from those generated by ES6023 MCE (about 14 and 20 kDa; Figure 6A).The DB219 MCE and ES6023 MCE hydrolyzed β-CN to produce hydrolysates about 12 and 20 kDa.The β-CN was hydrolyzed into 12-24 kDa fragments by calf rennet (Figure 6B).Moreover, the DB219 MCE and calf rennet hydrolyzed κ-CN to mainly generate para-κ-CN (14 kDa) and glycomacropeptides (GMP; 17 kDa).The κ-CN was also hydrolyzed into clear fragments (18-24 kDa) by ES6023 MCE except para-κ-CN and GMP (Figure 6C).

CONCLUSIONS
The novel MCE was extracted and purified from Bacillus velezensis DB219 with high yield and recovery.The DB219 MCE was metalloprotease (about 36 kDa) and remarkable specific activity of 6,110 SU/mg.The DB219 MCE was weak acid resistance and stable at pH 6.0-10.0 and temperature <45°C.The highest MCA appeared at milk substrate pH 5.5 and 20-30 mM CaCl 2 .The optimal temperature was 75°C.The K m and V m values for casein were 0.31 g/L and 14.22 μmol/min, respectively.The DB219 MCE hydrolyzed β-CN to a greater level than α-CN, indicating that it was suitable for making hard cheese.The casein cleavage sites and peptide composition generated by DB219 MCE were significantly different.DB219 MCE is a novel MCE that enriches kinds of calf rennet substitute.

Figure 7
Figure 7 indicated the differences of casein cleavage sites between DB219 MCE and R. miehei MCE.The