Introduction

Cardiovascular diseases such as acute myocardial infarction, ischemic heart disease, and high blood pressure are the major causes of human death throughout the world. Fibrin accumulation in the blood vessels is the main cause of the diseases. Thrombolytic agents such as tissue plasminogen activator (t-PA), urokinase (UK), and streptokinase (SK) have been widely used for the treatment of thrombosis. However, these agents can cause serious side effects, such as bleeding, short half-lives, high cost, allergic reactions, and large therapeutic doses [15]. Recently, fibrinolytic enzymes from Bacillus species have been actively studied because of their potential as fibrinolytic agents. Nattokinase secreted by Bacillus subtilis natto is the most well-known example [32]. Similar fibrinolytic enzymes are also secreted by Bacillus species and they have been purified. Their properties, including molecular weight, optimum pH and temperature, stability, and substrate specificity, have been reported [1,36]. Nattokinase is produced as a preproenzyme and processed into an active enzyme with 275 amino acids. Nattokinase has a catalytic triad consisting of Asp32, His64, and Ser221 residues and two calcium-binding sites that stabilize the three-dimensional structure [25].

The activity and stability of fibrinolytic enzymes are important properties to be considered if the enzymes are intended to be used for the prevention and treatment of thrombosis. Engineering of a fibrinolytic enzyme has been reported [7]. Error-prone PCR is a method widely used to introduce various mutations into a specific gene [4]. The method does not require detailed information on the structure of a protein and accurate predictions of the amino acid substitutions at the proper sites [20]. The half-life of a lipase from Bacillus sp. at 40℃ was increased by 3-fold and the catalytic activity of the mutant increased by 2~5-fold over the wild type after error-prone PCR [14]. The thermostability and pH stability of a xylanase from Thermomyces lanuginosus DSM 5826 were improved by error-prone PCR [31].

Bacillus sp. HK176 was isolated from cheonggukjang, a Korean fermented soyfood, as a strain secreting highly active fibrinolytic enzymes (unpublished result). In this work, the structural gene of the major fibrinolytic enzyme, aprE176, was cloned from HK176 and overexpressed in E. coli. Error-prone PCR was done with aprE176 to get mutants with increased fibrinolytic activities. One mutant, M179, showed increased fibrinolytic activity and thermostability. The cloning of aprE176 and production of M179 are described. The increased thermostability of M179 and a possible cause for the improved thermostability are also discussed.

Materials and Methods

Bacterial Strains and Plasmids

E. coli DH5α, E. coli BL21(DE3), and B. subtilis HK176 were grown in LB (Bacto tryptone 10 g, NaCl 10 g, yeast extract 5 g, per liter) broth at 37℃ with aeration. For cultivation of cells harboring pHY300PLK (Takara Bio Inc., Shiga, Japan), tetracycline was included in the culture medium at the concentration of 12 µg/ml. For cultivation of cells harboring pET-26b(+) (Novagen, Madison, WI, USA) or pET176, kanamycin was included at the concentration of 30 µg/ml.

Isolation of HK176 from Cheonggukjang

Commercial cheonggukjang products produced in several provinces of Korea were purchased and used as sources of bacilli with fibrinolytic activities. One gram of cheonggukjang was mixed with 9 ml of sterile 0.1% peptone water and homogenized using a stomacher (Seward, London, UK), and the homogenate was serially diluted using 0.1% peptone water. Aliquots (0.1 ml) of diluted samples were spreaded onto LB agar plates containing skim milk (2% (w/v)) and the plates were incubated for 12 h at 37℃. Colonies with large lysis zones were selected and further examined for the fibrinolytic activities by spotting on fibrin plates. The fibrinolytic activity was determined by the fibrin plate method [11]. Fibrin plates were incubated for 18 h at 37℃ and the size of lysis zone was compared with those caused by plasmin (Sigma, St. Louis, MO, USA) with known concentrations.

Construction of pET176

The aprE176 gene was amplified by PCR using total DNA from B. subtilis HK176 as the template and the primer pair 51F (5’-AGGATCCCAAGAGAGCGATTGCGGCTGTGTAC-3’, BamHI site underlined) and 51R (5’-AGAATTCTTCAGAGGGAGCCACCCGTCGATCA-3’, EcoRI site underlined) [16]. aprE176 without its signal sequence was amplified using the primer pair petF (5’-AGAGGATCCGATGGCAGGGAAATCA-3’, BamHI site underlined) and petR (5’-AGACTCGAGCTGAGCTGCCGCCTG -3’, XhoI site underlined). The PCR conditions were as follows: 94℃ for 5 min, followed by 30 cycles of 94℃ for 0.5 min, 60℃ for 0.5 min, and 72℃ for 1 min. Amplified aprE176 and the aprE176 without a signal sequence were cloned into pET26b(+) (5.36 kb, Kmr), respectively. E. coli DH5α and E. coli BL21(DE3) competent cells were prepared and transformed by electroporation as described previously [6]. Expression of aprE176 in E. coli was examined by measuring the fibrinolytic activity of the transformant, and the activity was expressed as plasmin unit/mg.

Purification of AprE176

E. coli BL21(DE3) cells harboring recombinant pET26b(+) with aprE176 or a mutant gene were cultivated in LB broth (500 ml) containing kanamycin (30 µg/ml) at 37℃. When the OD600 value of the culture reached 0.8, IPTG (isopropyl β-D-1-thiogalactopyranoside) was added (1 mM) to induce gene expression. After 15 h incubation at 20℃, cells were harvested by centrifugation and resuspended in 20 mM sodium phosphate buffer with 0.5 M NaCl and 10 mM imidazole (pH 7.4). Cells were disrupted by sonication (Sonoplus GM2070; Bandelin, Berlin, Germany) and the resulting cell extract was centrifuged at 12,000 ×g for 10 min at 4℃. The supernatant and cell pellet were obtained. AprE176 and M179 (a mutant) were purified from the supernatant using a Ni-NTA column (GE Healthcare, Uppsala, Sweden). The Bradford method [2] was used to determine the protein concentration of the enzyme preparation, and bovine serum albumin was used as the standard.

Properties of AprE176

Purified enzyme (1 µg) was incubated in buffers with different pH values (pH 3-12) for 2 h at 37℃. Buffer of 50 mM concentration was used: citrate–NaOH (pH 3–5), sodium phosphate (pH 6–7), Tris–HCl (pH 8–9), and glycine–NaOH (pH 10–12). After 2 h, the remaining activity was measured by the fibrin plate method. The optimum temperature for the fibrinolytic activity was determined by measuring the activity after 30 min incubation (pH 8.0) at temperatures ranging from 37℃ to 60℃. The thermostability of the enzyme was evaluated by measuring the remaining activity of enzyme, which was incubated at 45℃ (pH 8.0) for different time periods. The effects of metals and inhibitors on the activity were examined by incubating AprE176 in the presence of 5 mM metal ions or 1 mM inhibitors for 30 min at 40℃ (pH 8.0) and measuring the remaining activities.

Error-Prone PCR of aprE176

Error-prone PCR was performed using pET176, a pET26b(+) containing aprE176, as the template and according to the method of Cadwell and Joyce [5]. The reaction mixture (100 µl) consisted of 10 mM Tris–HCl (pH 8.3), 50 mM KCl, 7 mM MgCl2, 0.01% gelatin, 0.2 mM dATP and dGTP, 1 mM dCTP and dTTP, pET176 (10 ng), 0.5 mM MnCl2, and Taq polymerase (Takara, 5 units). petF and petR were used as the primers (0.5 µM, each). The PCR was carried out for 30 cycles of 94℃ for 1 min, 60℃ for 0.5 min, and 72℃ for 1 min. The PCR products were ligated with pET26b(+) after being digested with BamHI and XhoI. The ligation mixture was introduced into E. coli BL21(DE3) by electroporation. Transformants, growing on LB plates containing kanamycin (30 µg/ml) and skim milk (2% (w/v)), were screened for their proteolytic activities. After 48 h at 37℃, colonies with larger lysis zones than control were selected and inoculated into 5 ml of LB broth using sterile toothpicks. Plasmid DNA was prepared from each transformant for DNA sequencing.

Hydrolysis of Fibrinogen

Hydrolysis of fibrinogen was examined using purified AprE176 and M179 at 37℃. Fibrinogen (1 mg, bovine; MP Biochemicals, Illkirch, France) was dissolved in 1 ml of 20 mM Tris-HCl (pH 8.0) and digestion started by adding enzyme (50 ng/ml). Aliquots were taken at time points, mixed with 5× SDS sample buffer, and boiled for 5 min. SDS-PAGE was done using 12% acrylamide gel and the gel was stained with Coomassie Brilliant Blue R-250.

Kinetics and Amidolytic Activity

The amidolytic activity of AprE176 and M179 was assayed using the following substrates: N-Succinyl-Ala-Ala-Pro-Phe p-nitroanilide (Sigma, S7388), N-Benzoyl-Phe-Val-Arg p-nitroanilide hydrochloride (Sigma, B7632), N-Benzoyl-Pro-Phe-Arg p-nitroanilide hydrochloride (Sigma, B2133), and N-(p-tosyl)-Gly-Pro-Lys 4-nitroanilide acetate salt (Sigma, T6140). Fifty microliters of 10 mM substrate in 50 mM Tris-HCl (pH 8.0) was mixed with 10 µl of enzyme (1 µg) and 440 µl of Tris-HCl (pH 8.0). After 10 min at 37℃, 500 µl of citrateNaOH (pH 3.0) was added and the mixture was put on ice immediately and centrifuged at 12,000 ×g for 5 min. The optical density (OD410) of the supernatant was measured and the degree of hydrolysis was calculated from the absorbance value and molar extinction coefficient value of p-nitroanilide (8,800 M-1cm-1). Kinetic parameters (Vmax and Km ) of AprE176 and M179 were determined by measuring the release of p-nitroaniline from N-Succinyl-Ala-Ala-Pro-Phe p-nitroanilide (0.01 to 0.9 mM), which was dissolved in 100 mM phosphate buffer (pH 8.0) containing 4% (v/v) DMSO at 37℃ [32]. The Vmax value was converted to the kcat value from the relationship, kcat = Vmax/(enzyme).

Protein Structure Prediction

The structures of AprE176 and M179 were predicted by using SWISS-MODEL Workspace [30], and Subtilisin Nat (PDB ID: 4DWW) was used as the template structure. Predicted structures were compared using the PyMOL Molecular Graphics System (ver. 1.5.0.4, Schrödinger, LLC., NY, USA) and exported as images.

Statistical Analysis

All data are expressed as the mean ± SD values. The statistical analyses were performed using the SPSS program (SPSS, Inc., Chicago, IL, USA). Duncan’s multiple range test was used to examine the difference among treatments. P values of 0.05 were considered significant, if not otherwise stated.

Results and Discussion

Isolation and Identification of HK176

cheonggukjang products produced in different provinces of Korea in 2012 were purchased for the screening of bacilli with strong fibrinolytic activities. Colonies with strong protease activities were selected using skim-milk plates. Selected isolates were further examined for their fibrinolytic activities using fibrin plates. Among the isolates, HK176 was the best in terms of the fibrinolytic activity. For the identification of HK176, the 700 bp recA gene [21] and 1,300 bp 16S rRNA gene were amplified and the sequences were determined. BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) search for both genes indicated that HK176 was either a B. subtilis (99% identity) or a B. tequilensis (99% identity) (results not shown). The RA PD-PCR profile of HK176 obtained by S30 primer matched well with that of B. subtilis reference strains (results not shown) [21]. From these results, HK176 was identified as a B. subtilis strain.

Cloning of aprE176

aprE176 was amplified from genomic DNA of B. subtilis HK176 by PCR. The 1.5 kb fragment was cloned into an E. coli-Bacillus shuttle vector, pHY300PLK, after digestion with EcoRI and BamHI, resulting in pHY176 (6.43 kb). Sequence analysis showed that the fragment was 1,491 bp in length and contained the major fibrinolytic gene, aprE176. The sequence was deposited into GenBank under the accession number of KJ572414. aprE176 encodes a protein of 382 amino acids (aa) consisting of signal sequence (30 aa), propeptide (77 aa), and mature enzyme (275 aa). This configuration was expected from comparison of the amino acid sequence of AprE179 with those of similar fibrinolytic enzymes from bacilli [9,29]. The size of mature and active AprE176 was estimated to be 27 kDa after processing. The ribosome-binding site (RBS) and putative promoter sequences (-35 and -10) were located upstream of the ORF (Fig. 1) [11]. The nucleotide sequence of aprE176 showed 99% similarity to those of fibrinolytic genes of B. amyloliquefaciens CH51 (EU414203), B. amyloliquefaciens MJ5-41 (JF739176), B. subtilis JZ06 (EU386607), and B. subtilis D-2 (JQ730856). The translated amino acid sequence of AprE176 was aligned with those of other homologous proteins (Fig. 2). AprE176 showed higher similarity with other fibrinolytic enzymes such as 99.7% with AprE3-17 from B. licheniformis CH3-17 (ACU32756) [13], 99.4% with AprE51 from B. amyloliquefaciens CH51 (ACA34903) [16], 98.9% with peptidase S8 from B. amyloliquefaciens LFB112 (AHC41550), subtilisin DFE from B. amyloliquefaciens DC-4 (AAZ66858) [29], and subtilisin DJ-4 from Bacillus sp. DJ-4 (AAT45900) [17], 86.3% with nattokinase from B. subtilis subsp. natto (ACJ11220) [26], and 85.6% with subtilisin E from B. subtilis 168 (CAA74536). The nucleotide sequence of aprE176 was similar to those of B. amyloliquefaciens and B. subtilis strains and the deduced amino acid sequence showed the highest similarity to that of AprE3-17 from B. licheniformis. When the amino acid sequences of mature enzymes were compared, AprE176 differed from AprE3-17 at a single amino acid, and differed from AprE51 at two amino acids (Fig. 2).

Fig. 1.Nucleotide sequence of aprE176. The deduced amino acid sequence is shown above the nucleotide sequence. Putative -35 and -10 promoter sequences are underlined. The RBS and transcription terminator are also underlined. The ends of the pre (▼) and pro sequences (▽) are marked.

Fig. 2.Alignment of amino acid sequence of AprE176 with homologous enzymes. AprE51 (B. amyloliquefaciens CH51, ACA34903), AprE3-17 (B. licheniformis CH3-17, ACU32756), peptidase S8 (B. amyloliquefaciens LFB112, YP_008949402), subtilisin DFE (B. amyloliquefaciens DC-4, AAZ66858), subtilisin DJ-4 (Bacillus sp. DJ-4, AAT45900), nattokinase (B. subtilis subsp. natto, ACJ11220), and subtilisin E (B. subtilis 168, CAA74536). Amino acids different from those of other proteins are marked in a box.

proaprE176 without the signal sequence was amplified for overexpression in E. coli BL21(DE3). The 1.1 kb fragment was cloned into pET26b(+), resulting in pET176. In pET176, the pelB signal sequence in the vector directed the secretion of AprE176 into the periplasm of E. coli, and aprE176 was transcribed by the T7 promoter. The T7 promoter system was also used for the expression of other fibrinolytic genes in E. coli [8,22].

Purification of AprE176

After centrifugation of sonicated E. coli cells, the supernatant and cell pellet were obtained. Fibrinolytic activity was detected from both the supernatant and pellet, but higher activity was detected from the supernatant (data not shown). Thus, the supernatant was used for the purification of AprE176. A Ni-NTA column was used for the purification of AprE176, and bound AprE176 was eluted as described by the supplier. A 27 kDa single band was observed when the eluate was analyzed by SDS-PAGE (data not shown). The specific activity of the purified AprE176 was 216.8 ± 5.4 plasmin unit/mg protein. For the M179 purification, the exact same procedures were used.

Properties of AprE176

AprE176 maintained more than 80% of its activity after 2 h at pH between 7.0 and 10.0, but the activity decreased rapidly at pH below 6.0 (Fig. 3A). The optimum pH was 8.0. The optimum temperature was 40℃ at pH 8.0 (Fig. 3B). The activity decreased rapidly at 50℃ and was completely inactivated at 55℃. The optimum pH and temperature of AprE176 are similar to those of nattokinase from B. subtilis natto B-12 [35]. These activities are lower than those of an enzyme from B. amyloliquefaciens DC-4 (pH 9.0 and 48℃, respectively) [28] and an enzyme from Bacillus sp. CK11-4 (pH 10.5 and 70℃, respectively) [19]. The optimum pH of AprE176 (pH 8.0) was quite different from that of AprE3-17 (pH 6.0) [13] and AprE51 (pH 6.0) [16]. The mature AprE176 differs from AprE3-17 and AprE51 at a single amino acid and two amino acids, respectively. The result indicates that even a single amino acid difference can cause significant difference in the enzyme properties. AprE176 was completely inactivated by 1 mM PMSF (phenylmethylsulfonyl fluoride), a well-known inhibitor of serine proteases. EDTA (ethylenediaminetetraacetic acid) and EGTA (ethylene glycol tetraacetic acid) also inhibited the fibrinolytic activity of AprE176. The results showed that AprE176 is a serine protease and also a metalloprotease. Unlike AprE176, AprE3-17 retained more than half of its initial activity after 30 min exposure to PMSF and EDTA [13], and AprE51 retained 83% activity after 1 h exposure to EDTA at 45℃ [16]. The result again proves the uniqueness of AprE176 from other similar enzymes. Ca2+ increased the activity by 17%, but Cu2+, Mn2+, and Zn2+ reduced the activity by 7%, 7%, and 43%, respectively (Table 1). Ca2+ increased the activity of many fibrinolytic enzymes [10,18]. Zn2+ inhibitedseveral fibrinolytic enzymes, including AprE176. For example, 5 mM Zn2+ completely inactivated nattokinase from B. subtilis YJ1 [37] and decreased the activity of subtilisin DJ-4 from Bacillus sp. DJ-4 by 72.88% [17].

Fig. 3.Effects of pH and temperature on the activity of AprE176. (A) Residual fibrinolytic activities were measured after 2 h at different pH values. (B) Residual fibrinolytic activities were measured after 30 min in Tris-buffer (pH 8.0) at different temperatures.

Table 1.aThe counterion for the tested metals was chloride. All values are the mean ± SD (n = 3).

Error-Prone PCR of aprE176

Error-prone PCR was done with aprE176 for the purpose of obtaining mutant enzymes with improved fibrinolytic activities. E. coli BL21(DE3) transformants (550 colonies) were initially screened using skim-milk plates and further screened using fibrin plates. Subsequently, three clones (M57, M111, M179) that showed higher fibrinolytic activities than wild type were selected by measuring the clear zone on the fibrin plate. In addition, a mutant (M11) showing significantly decreased activity was also selected for comparison. Sequence analysis of four mutants showed that the amino acid substitutions were V84A/K213M, K213N, G157S, and A176T for M11, M57, M111, and M179, respectively. The specific activities of purified M11, M57, M111, and M179 were 10.9 ± 1.6, 277.1 ± 15.7, 300.4 ± 21.2, and 479.8 ± 12.2 plasmin unit/mg protein, respectively. Among them, M179 showed the highest fibrinolytic activity, and this value was 2.2-fold higher than that of wild-type AprE176. M111 showed 1.4-fold higher activity and M57 showed 1.3-fold higher activity. Changes in the 84th and 213th amino acids at the same time almost completely destroyed the activity (M11).

Hydrolysis of Fibrinogen by AprE176 and M179

Fibrinogen is the precursor of fibrin, composed of three pairs of disulfide-bonded polypeptide chains (Aα, Bβ, and γ). Both AprE176 and M179 quickly degraded the Aα chain of fibrinogen, which disappeared within 30 sec (Fig. 4). The Aα chain was hydrolyzed first, followed by the Bβ chain. This indicated that AprE176 possesses strong α-fibrinogenase and moderate β-fibrinogenase activities. The degradation pattern was similar to those observed in some other fibrinolytic enzymes such as AprE2 from B. subtilis CH3-5 [12], TPase from B. subtilis TP-6, and plasmin [18]. M179 degraded Aα and Bβ chains completely within 20 min, indicating that M179 had improved β-fibrinogenase activity. Both enzymes failed to degrade the γ-chain in 20 min.

Fig. 4.Hydrolysis of fibrinogen by AprE176 (A) and M179 (B). Lane M, size marker (DokDo-MARK, ELPIS-Biotech. Inc., Daejeon, Korea.); 1, no enzyme treatment; 2, 30 sec; 3, 1 min; 4, 2 min; 5, 3 min; 6, 4 min; 7, 5 min; 8, 10 min; 9, 20 min. A 12% acrylamide gel was used.

Amidolytic Activities and Kinetics of AprE176 and M179

The amidolytic activities of AprE176 and M179 were determined using four different synthetic substrates. The most sensitive substrate was N-succinyl-Ala-Ala-Pro-Phe-pNA, a preferred substrate for subtilisin and chymotrypsin. AprE176 and M179 showed no activity for the other substrates, indicating that AprE176 is a subtilisin-like serine protease. The amidolytic activities of AprE176 and M179 were 28.63 ± 0.20 and 39.84 ± 0.04 mM/min/mg protein, respectively, when N-succinyl-Ala-Ala-Pro-Phe-pNA was used as the substrate. M179 showed 1.39-fold higher amidolytic activity than AprE176. The kinetic constants, Km and kcat, were determined for AprE176 and M179 by measuring the initial rates of hydrolysis of N-succinyl-Ala-Ala-Pro-Phe-pNA (Table 2). Although M179 showed enhanced fibrinolytic activity and catalytic turnover (kcat), it had a lower substrate affinity (Km) than AprE176. As a result, the catalytic efficiency (kcat/Km) of M179 was not different from that of AprE176.

Table 2.All values are the mean ± SD (n = 3). Values in the same column with different superscripts are statistically different by Duncan’s multiple range test (p = 0.05).

In the case of a keratinase from B. licheniformis BBE11-1, a single amino acid change (N122Y) caused the increase of the catalytic efficiency by 5.6-fold compared with wild-type keratinase. The 122th residue of mutant (N122Y) was located 6.42 Å, 6.63 Å, and 2.45 Å from Asp32, His63, and Ser220, respectively. Some intramolecular interactions among the aromatic 122th residue, the active site (Asp32, His63, Ser220), and substrate influenced to increase the catalytic efficiency [23], whereas other substitutions (N217S, A193P, and N160C), which were located far from the active site, did not significantly change the Km or kcat [23]. For a serine alkaline protease from B. pumilus CBS, the catalytic efficiency was increased 42-fold by mutations (L31I/T33S/N99Y) [10]. The substitutions occurred at residues surrounding the catalytic residue, Asp32, in serine alkaline proteases. For the above examples, changes occurred near the active site, and this enhanced the catalytic efficiency. In the case of M179, the 176th residue was located far from the catalytic site, and this might be the reason why the catalytic efficiency did not increase.

Thermostability of M179

M179 possessed improved thermal resistance compared with AprE176. Purified M179 retained 67% residual activity after 150 min incubation at 45℃, whereas AprE176 had 50% activity (Fig. 5). When the incubation time was extended to 5 h at 45℃, M179 still retained 36% activity but AprE176 had only 11% activity. The thermostability of an alkaline protease (BgAP) from Bacillus gibsonii was increased more than 90-fold (half-life at 60℃) by directed evolution [24]. The loop area of the BgAP surface was replaced by substitutions to negatively charged amino acids, which contributed to the increase of the thermostability, probably through increasing ionic and hydrogen bond interactions in mutated BgAP. The half-life of a keratinase at 60℃ from B. licheniformis BBE11-1 was increased 8.6-fold by four amino acid substitutions (N122Y, N217S, A193P, N160C) [23]. Increase in the thermostability was explained to be the result of increased interactions, such as hydrophobic interaction, cation-pi interaction, and hydrogen bond, in the mutant. Interactions such as hydrogen bonds, hydrophobic interactions, and electrostatic interactions are well known to be the main factors for protein thermostability [33].

Fig. 5.Temperature stability of AprE176 and M179. Thermal stability was determined at 45℃ in Tris-HCl buffer (pH 8.0) for various times. AprE176, -●-; M179, -■-.

To explain the increased thermostability of M179 compared with wild-type ArpE176, molecular modeling was used. Three-dimensional models of AprE176 and M179 were constructed using the known structure of nattokinase (4DWW). Nattokinase and AprE176 share 76% identity in amino acid sequences, and homology-based model structures could be built using SWISS-MODEL Workspace (http://swissmodel.expasy.org/workspace/). Based on the model structures, A176T mutation might be responsible for the increased thermostability of M179. According to the model structure of AprE176, three amino acids (Gly169, Tyr171, and Val174) are located within 3.0 Å radius from a calcium ion (Fig. 6A). For M179, the hydroxyl group of Thr176 is located closely (2.9 Å) to the calcium ion, whereas Ala176 of AprE176 does not show any interaction with calcium ions. Bacillus fibrinolytic enzymes (subtilisin E [34], Carlsberg [25], and BPN [3]) have two calcium-binding sites, Ca A and Ca B sites. The Ca A site is located near the N-terminus and has higher calcium-binding affinity, whereas Ca B site has weaker binding affinity. The occupation of the Ca B site is an important factor for the increase of thermal resistance. Increased thermal stability by increased calcium concentration was shown by Pantoliano et al. [27] at calcium concentrations ranging from 0.1 to 100 mM. It was also reported, in the X-ray crystal structure study using subtilisin, that the occupancy of the Ca B site was increased as the calcium concentration was increased.

Fig. 6.Predicted structures of calcium-binding pocket. (A) AprE176, (B) M179. The purple ball indicates a calcium ion.

From the model structure, it is suspected that Thr176 of M179 is more likely located in the Ca B site. The existence of an extra amino acid with favorable interaction with calcium ions could possibly increase the affinity with calcium ions, resulting in increase of the thermostability of the enzyme. Increased thermostability is desirable if AprE176 and its derivatives are incorporated into foods that are expected to be heat treated.

AprE176 possesses strong fibrinolytic activity, which can be utilized for the production of fermented foods or drugs preventing cardiovascular diseases caused by fibrin clots. Mature AprE176 shows different physicochemical properties, such as pH optimum and inhibition by chemicals, from enzymes with quite similar amino acid sequences. The property of AprE176 can be improved by directed evolution methods, and M179, a derivative of AprE176, possesses improved thermostability. Further improvements are necessary, and studies on the roles of important amino acids for the activity of AprE176 should also be continued.