Probiotics and Antimicrobial Proteins (2019) 11:283–294

https://doi.org/10.1007/s12602-017-9382-7

Strain Screening from Traditional Fermented Soybean Foods

and Induction of Nattokinase Production in Bacillus subtilis MX-6
Li-Li Man 1 & Dian-Jun Xiang 2 & Chun-Lan Zhang 1

Published online: 6 February 2018

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
The plasminogen-free fibrin plate assay method was used to isolate Bacillus subtilis MX-6, a strain with high production of
nattokinase from Chinese douchi. The presence of aprN, a gene-encoding nattokinase, was verified with PCR method. The
predicted amino acid sequence was aligned with homologous sequences, and a phylogenetic tree was constructed. Nattokinase
was sublimated with ammonium sulfate, using a DEAE-Sepharose Fast Flow column, a CM-Sepharose Fast Flow column and a
Sephadex G-75 gel filtration column. SDS-PAGE analysis indicated that the molecular weight of the purified nattokinase from
Bacillus subtilis MX-6 was about 28 kDa. Fermentation of Bacillus subtilis MX-6 nattokinase showed that nattokinase produc-
tion was maximized after 72 h; the diameter of clear zone reached 21.60 mm on the plasminogen-free fibrin plate. Nattokinase
production by Bacillus subtilis MX-6 increased significantly after supplementation with supernatant I, supernatant II and soy
peptone but decreased substantially after the addition of amino acids. This result indicated that the nattokinase production by
B. subtilis MX-6 might be induced by soybean polypeptides. The addition of MgSO4 and CaCl2 increased B. subtilis MX-6
nattokinase production.

Keywords Nattokinase . Screening . Bacillus subtilis . Induction

Introduction transforming the passivating proenzyme plasminogen into

plasmin [7, 8]. The other thrombolytic type is a plasmin-like
Cardiovascular diseases (CVDs) that result from the coagulation protein that degrades the fibrin directly and dissolves thrombin
of fibrin have a severe impact on human health [1–3]. entirely. Both types of fibrinolytic agents have been used
Thrombolytic enzymes are therefore often used to treat CVDs extensively in the treatment of CVDs but have nonetheless
[4–6]. Thrombolytic agents may be divided into two categories several drawbacks including high cost, possible bleeding
based on their effects: one type contains urokinase and a tissue- complications and risk of acute coronary reocclusion [9, 10].
type plasminogen activator (t-PA), which decomposes fibrin by In light of these limitations, the fibrinolytic enzymes pro-
duced by various microbes deserve further attention. It is pos-
Li-Li Man, Dian-Jun Xiang and Chun-Lan Zhang contributed equally to sible to acquire these microbial fibrinolytic enzymes rapidly,
this work. reducing the energy input, processing time and overall cost of
CVD treatment [11]. The fibrinolytic enzymes produced by
* Dian-Jun Xiang Bacillus sp. are highly active and stable, making them attrac-
[email protected]
tive as thrombolytic agents. Bacillus sp. secretes a variety of
Li-Li Man
intracellular and extracellular proteases, including
[email protected] nattokinase. Nattokinase-producing strains are usually isolat-
ed from traditional fermented soybean products, for example,
Chun-Lan Zhang
[email protected] Chinese douchi, Japanese natto, Korean doen-jang [12] and
Korean Chungkook-jang [13].
1
College of Life Science, Inner Mongolia University for Nationalities, Nattokinases, subtilisin-type serine proteases, are currently
Tongliao 028042, People’s Republic of China primarily consumed in food or used as probiotics, but their
2
College of Agriculture, Inner Mongolia University for Nationalities, potential fibrinolytic properties have been the subject of sev-
Tongliao 028042, People’s Republic of China eral recent studies. Compared to other thrombolytic agents,
284 Probiotics & Antimicro. Prot. (2019) 11:283–294

nattokinase has many advantages including high activity, in- saline, plated onto the basal seed medium with 1.8% agar and
creased security, low cost and easy oral administration [14]. incubated at 37 °C for 24 h. Strains were separated and puri-
Nattokinase has shown promise as a thrombolytic enzyme fied, then cultivated in 250 mL Erlenmeyer flasks with 50 mL
both in vivo and in vitro [15]: it not only digests fibrin imme- of basal seed medium at 37 °C for 12 h in an orbital incubator
diately, but also possesses a plasminogen activator [2]. at 180 rpm (OD600 = 1.5–1.6). We added 5% (v/v) inoculum
Nattokinase has been screened for fibrinolytic activity using culture (OD600 = 1.5–1.6) to 250-mL Erlenmeyer flasks con-
the fibrin plate method and the fibrin degradation method. The taining 50 mL of basal fermentation medium and incubated
aprN gene that encodes nattokinase is homologous with genes these at 37 °C for 48, 72 and 96 h in an orbital incubator at
encoding fibrinolytic enzymes. Hence, a combination of 180 rpm. Cultures were then centrifuged at 5000×g for
plasminogen-free fibrin plate assays and PCR-based methods 10 min, and the supernatant was used as a crude enzyme to
should effectively screen for and identify nattokinase-producing detect fibrinolytic enzyme activity. The strain with the largest
Bacillus sp. strains. Few studies have investigated the effects of diameter on the plasminogen-free fibrin plate was selected for
soybean components on nattokinase production [5]. further study.
Our study had several aims. First, screen Bacillus sp.
strains from fermented soybean products (natto and douchi) Identification of Strains Producing Fibrinolytic
and identify the strains producing fibrinolytic enzymes. Enzymes
Second, sequence aprN, the gene that encodes nattokinase.
Third, purify the crude enzyme and determine the molecular We used Bergey’s Manual of Systematic Bacteriology to iden-
weight of nattokinase. Fourth, ascertain the fermentation char- tify bacterial strains based on morphology and physiology.
acteristics of nattokinase in Bacillus subtilis MX-6. Fifth, in- The micromorphology of individual cells of strain MX-6
vestigate the effects of soybean components and inorganic was examined using gram staining. The colony morphology
salts on nattokinase production in B. subtilis MX-6. of strain MX-6 was observed after cultivation on basal seed
medium with 1.8% agar at 37 °C for 24 h. The features of
strain MX-6 were identified using biochemical tubes.
Materials and Methods The total genomic DNA of strain MX-6 was extracted
with a TIANamp Bacteria DNA Kit (Tiangen, China).
Samples and Medium Based on known sequences of B. subtilis (GenBank acces-
sion numbers AY219900.1, EF423598.1, JN399219.1,
We purchased douchi and natto from supermarkets local to EF528288.1, DQ416796.1, JX402129.1, JQ435698.1,
Mudanjiang, China. We obtained fibrinogen and thrombin AB188212.1 and HM030757.1), we constructed novel de-
from Sigma-Aldrich (USA). Other chemicals used herein generate B. subtilis primers (MX-6-16S rDNA-F: 5′-
were analytical reagents (AR). AGAGKTTGAWYMTGGCTCAGGACG-3′ and MX-6-
The basal seed medium was 1% peptone, 0.3% beef 16S rDNA-R: 5′-ARGGAGGTGATCCAGCCRCACCTT-
extract and 0.5% NaCl (pH 7.2–7.4). The basal fermenta- 3′) to amplify 16S rDNA. PCR products were tested with
tion medium contained 2% peptone, 2% glucose and 0.5% agarose gel electrophoresis at 100 V and sequenced.
NaCl (pH 7.2–7.4). Sequence homology was evaluated with nucleotide BLAST
(BLASTN, NCBI).
Measurement of Fibrinolytic Enzyme Activity
Cloning of the Gene-Encoding Nattokinase
We used the plasminogen-free fibrin plate assay method of
Jeong and Kim [9] to test fibrinolytic enzyme activity. In brief, We next amplified the nattokinase-encoding gene of
10 μL of crude enzyme was poured into a plasminogen-free B. subtilis MX-6, aprN. The total genomic DNA of
fibrin plate and cultured at 37 °C for 18 h. The amount of B. subtilis MX-6 was extracted with a TIANamp Bacteria
fibrinolytic enzyme activity was indicated by the diameter of DNA Kit (Tiangen, China). We used primers aprN-F (5′-
the clear zone on the plasminogen-free fibrin plate. We used GTATGAAAATAGTTATT TCGAGTCTC-3′) and aprN-R
vernier callipers to measure the diameter of this clear zone, (5′-TCCGGTGCTTGTGAAGATTTTCAGA-3′) to amplify
expressed as a mean ± standard deviation (n = 3). aprN. The PCR program was as follows: 94 °C for 5 min;
30 cycles of 94 °C for 30 s, 52 °C for 30 s, 72 °C for 1 min;
Isolation of Strains Producing Fibrinolytic Enzymes 72 °C for 10 min; and 4 °C indefinitely. PCR products were
tested with agarose gel electrophoresis at 100 V. PCR products
Two grams of natto or douchi was mixed with 5 mL sterilized were sequenced, and sequence homology was evaluated with
normal saline and heated to 95 °C for 5 min. The supernatant nucleotide BLAST (BLASTN, NCBI). We used primer 5.0 to
was then collected and diluted in 0.9% (w/v) sterile normal infer the sequence of the protein encoded by aprN and Expasy
Probiotics & Antimicro. Prot. (2019) 11:283–294 285

to predict its molecular weight. We then constructed a multiple 70% ethanol and stirred at 55 °C for 30 min. Supernatant II
sequence alignment of our inferred protein and homologous was obtained via centrifugation.
proteins with ClustalX. We used the neighbour-joining meth-
od in MEGA 6.0 to construct a phylogenetic tree based on this Effect of Soybean Components on Nattokinase Production
alignment.
B. subtilis MX-6 was cultured in a 250 mL Erlenmeyer flask
Purification of Nattokinase and Determination with 50 mL of basal seed medium at 37 °C for 12 h in an
of Molecular Weight orbital incubator at 180 rpm (OD600 = 1.5–1.6). We added 5%
(v/v) inoculum culture (OD600 = 1.5–1.6) to 50 mL of each of
B. subtilis MX-6 was cultured in a 250-mL Erlenmeyer flask the following media: (1) basal fermentation medium with 1, 2,
with 50 mL of basal seed medium at 37 °C for 12 h in an 3, 4 or 5% SPI added; (2) basal fermentation medium with 1,
orbital incubator at 180 rpm (OD600 = 1.5–1.6). We added 5% 2, 3, 4 or 5% supernatant I added; (3) basal fermentation
(v/v) inoculum culture (OD600 = 1.5–1.6) to 50 mL of basal medium with 1, 2, 3, 4 or 5% supernatant II added; (4) basal
fermentation medium, and cultured it at 37 °C for 72 h in an fermentation medium in which 2% peptone was replaced by
orbital incubator at 180 rpm. We obtained the crude enzyme soy peptone (0.5%) peptone (1.5%), soy peptone (1.0%) plus
supernatant via centrifugal separation at 10,000×g for 10 min peptone (1.0%), soy peptone (1.5%) plus peptone (0.5%) or
at 4 °C. We used ammonium sulphate to precipitate proteins soy peptone (2.0%); and (5) basal fermentation medium with
including nattokinase. Precipitates were collected via centri- added glutamic acid (Glu), aspartic acid (Asp), arginine (Arg),
fugation at 12,000×g for 20 min after overnight settlement at glycine (Gly), alanine (Ala), serine (Ser) or valine (Val) at
4 °C and dissolved in 10 mL of 0.04 M Barbital sodium-HCl concentrations of 0.05, 0.1, 0.15 or 0.2%. We used nattokinase
buffer. This solution was added to a DEAE-Sepharose Fast production when B. subtilis MX-6 was cultured in basal fer-
Flow column and balanced with 20 mM Tris-HCl buffer. mentation medium as the control.
The parts with nattokinase activity were eluted and delivered All media combinations were incubated at 37 °C for 48, 72,
into a CM-Sepharose Fast Flow column. Next, active parts and 96 h in an orbital incubator at 180 rpm. We used centrif-
were collected and added to a Sephadex G-75 gel filtration ugal separation to obtain the crude enzyme supernatant and
column previously balanced with 20 mM Tris-HCl buffer in- used this to detect nattokinase activity. All experiments were
cluding 0.15 M NaCl. The molecular weight of crude enzyme performed in triplicate.
and purified nattokinase was measured with SDS-PAGE.
Regulation of Nattokinase Production with Inorganic Salts
Determination of the Fermentation Characteristics
of Nattokinase Production in B. subtilis MX-6 B. subtilis MX-6 was cultured in a 250-mL Erlenmeyer flask
with 50 mL of basal seed medium at 37 °C for 12 h in an orbital
B. subtilis MX-6 was cultured in a 250-mL Erlenmeyer flask incubator at 180 rpm (OD600 = 1.5–1.6). We added 5% (v/v)
with 50 mL of basal seed medium at 37 °C for 12 h in an inoculum culture (OD600 = 1.5–1.6) to 50 mL of the following
orbital incubator at 180 rpm (OD600 = 1.5–1.6). We added 5% media: (1) basal fermentation medium with added K2HPO4
(v/v) inoculum culture (OD600 = 1.5–1.6) to 50 mL basal fer- (0.1%) plus KH2PO4 (0.1%), K2HPO4 (0.2%) plus KH2PO4
mentation medium in a 250-mL Erlenmeyer flask, and this (0.1%) or K2HPO4 (0.6%) plus KH2PO4 (0.1%); (2) basal fer-
was continuously cultivated at 37 °C for 102 h in an orbital mentation medium with added 10−5, 10−4 or 10−3 mol/L
incubator with shaking at 180 rpm. Samples were taken every MnSO4·H2O; (3) basal fermentation medium with added
6 h. The crude enzyme supernatant was obtained via centrif- 0.02, 0.04 or 0.06% MgSO4; and (4) basal fermentation medi-
ugal separation and was used to detect nattokinase activity um with added 0.02, 0.04 or 0.06% CaCl2. We used nattokinase
with a plasminogen-free fibrin plate assay. production when B. subtilis MX-6 was cultured in basal fer-
mentation medium as the control. All media combinations were
Regulation of Nattokinase Production by Soybean incubated at 37 °C for 48, 72 and 96 h in an orbital incubator at
Components in B. subtilis MX-6 180 rpm. We used centrifugal separation to obtain the crude
enzyme supernatant and used this to detect nattokinase activity.
Preparation of Soybean Components All experiments were performed in triplicate.

To obtain soybean components, we soaked and ground 1000 g Statistical Analysis

soybeans in 4000 mL water. pH was controlled at 5.5 using
HCl, and the seriflux was agitated at 50 °C for 30 min. Soy We used SPSS 17.0 for statistical analyses of the experimental
protein isolate (SPI) and supernatant I were harvested sepa- results. One-way analysis of variance (ANOVA) was used to
rately via centrifugation. Some of supernatant I was added to evaluate the statistical significance for data comparison.
286 Probiotics & Antimicro. Prot. (2019) 11:283–294

Results and Discussion Amplification of aprN, the Gene-Encoding

Nattokinase in B. subtilis MX-6
Strain Isolation and Screening for Fibrinolytic Enzyme
Activity To amplify aprN, the gene-encoding nattokinase, we used
previously published B. subtilis aprN gene sequences to de-
Sixty strains of B. subtilis were isolated from 15 commercial sign novel aprN primers (aprN-F and aprN-R). These se-
natto and douchi products. Four of strains (herein named MX-1, quences were B. subtilis strain 29R (CP017763.1), B. subtilis
MX-3, MX-5 and MX-6) had significantly higher nattokinase strain KH2 (CP018184.1), B. subtilis subsp. natto strain
activity as compared to all other strains (P < 0.01) (data not CGMCC 2108 (CP014471.1) and B. subtilis subsp. natto
shown). Strain MX-6, isolated from douchi, had the most BEST195 DNA (AP011541.2). Using our novel primers, we
nattokinase activity and was thus selected for further study. successfully amplified aprN in B. subtilis MX-6, suggesting
Most strains of B. subtilis that produce fibrinolytic en- that aprN is conserved across strains/subspecies of B. subtilis.
zymes have been isolated from natto, a common Japanese The aprN gene sequence was 1473 bp (Fig. 1) and had 99%
fermented soybean food [16, 17]. Other strains that produce homology with B. subtilis 29R (CP017763.1), B. subtilis KH2
homologous fibrinolytic enzymes have been found in other (CP018184.1) and Bacillus subtilis subsp. Natto strain
fermented foods, including B. subtilis DC33 from traditional CGMCC 2108 (CP014471.1). The sequence of aprN has been
flavour-rich Chinese foods [18]; B. subtilis LSSE-22 from submitted to GenBank, accession number KY576911.
samples of soybean paste [2]; B. subtilis TP-6 from The start codon of aprN was GTG, followed by an open
Indonesian soybean tempeh [14]; B. subtilis from fermented reading frame of 1152 bp, and had three termination co-
red beans; Bacillus sp. DJ-4 from Korean doen-jang [14]; dons (TAA TAG TAA; Fig. 1). Similar to nattokinase in
Bacillus amyloliquefaciens LSSE-62 from Chinese soybean B. subtilis LSSE-22 [2], the aprN open reading frame
paste [2]; Bacillus megaterium KSK-07 from Egyptian kishk (1143 bp) in B. subtilis MX-6 had an inferred amino acid
[19]; Bacillus weihenstephanensis from Vietnamese shrimp sequence with 381 amino acids (aa), including a 29 aa pre-
paste; and Bacillus sp. CK11-4 and Bacillus vallismortis sequence, a 77 aa pro-peptide and a 275 aa mature peptide
Aceo2 from traditional Korean condiment chungkook-jang (Fig. 1). The predicted molecular weight of mature
[20]. In addition to Bacillus spp., other sources of this enzyme nattokinase was 27,712.72 Da. The inferred amino acid
include Pseudomonas sp. and some marine organisms [5]. We sequence was 99% homologous to B. subtilis nattokinase
selected natto and douchi products for the isolation of fibrino- (AA065246.1). This result suggested that the fibrinolytic
lytic enzyme-producing strains because these are both rich in enzyme produced by B. subtilis MX-6 was a nattokinase.
beneficial food microorganisms. Here, we isolated B. subtilis Our multiple sequence alignment showed that our inferred
MX-6, a strain with high fibrinolytic enzyme production, from amino acid sequence was homologous to that of B. subtilis
Chinese douchi. (WP 014479360.1), with only one different residue, sug-
gesting that the domains of nattokinase in B. subtilis are
relatively conserved (Fig. 2a).
Identification of Strain MX-6 Phylogenetic analysis of fibrinolytic enzymes across
strains of Bacillus recovered two subfamilies of enzymes
Strain MX-6 was identified using morphology, biochemistry, (Fig. 2b). One subfamily included B. subtilis subtilisin
and genetic homology. Individuals of strain MX-6 had spore, NAT (WP 014479360.1), B. subtilis nattokinase precursor
rod-type, gram-positive morphology. These cells were posi- (AA065246.1), B. subtilis subtilisin E (WP 009966941.1),
tive for catalase, β-galactosidase, lecithinase, urease, VP re- B. subtilis fibrinolytic enzyme MX-6 (this study), Bacillus
action and motility. Colonies of strain MX-6 were white, mojavensis peptidase S8 (WP 010333625.1),
round, opaque, and wrinkled, with wavy edges. Strain MX-6 Brevibacterium halotolerans (WP 044156525.1), Bacillus
did produce gas and ammonia from glucose and arginine. axarquiensis (WP 059352635.1), Bacillus tequilensis (WP
Casein, gelatine and starch were hydrolysed by strain MX-6. 024712916.1) and B. vallismortis (WP 010329279.1). The
We amplified and sequenced strain MX-6 16S rDNA using other subfamily included Bacillus atrophaeus peptidase S8
degenerate primers. The length 16S rDNA amplified by PCR (WP 063637320.1), Bacillus nakamurai (WP
was about 1.5 kb; sequencing confirmed that the length of 061521470.1), Bacillus velezensis serine alkaline protease
strain MX-6 16S rDNA was 1542 bp. This sequence was up subtilisin (WP 022553591.1), Bacillus licheni AprE 3-17
to 99% homologous with the 16S rDNA of B. subtilis IAM (ACU32756.1), and Geobacillus stearothermophilus alka-
12118 (NR112116.1) and B. subtilis 168 (NR102783.1). line serine protease (ABY25856.1). Thus, our phylogenet-
Based on our results, we identify strain MX-6 as B. subtilis. ic analysis indicated that the fibrinolytic enzyme produced
Our sequence of B. subtilis MX-6 16S rDNA has been sub- by B. subtilis MX-6 belonged to the subtilisin fibrinolytic
mitted to GenBank, accession number KY575151. enzyme family.
Probiotics & Antimicro. Prot. (2019) 11:283–294 287

Fig. 1 The nucleotide (upper line) and inferred amino acid (lower line) and mature peptide, respectively. The putative SD sequence is overlined,
sequences of the aprN gene, encoding nattokinase in B. subtilis MX-6. and the transcription terminator is indicated with arrows
‘Pre’, ‘pro’ and ‘mature’ indicate the predicted signal peptide, pro-peptide
288 Probiotics & Antimicro. Prot. (2019) 11:283–294
 Probiotics & Antimicro. Prot. (2019) 11:283–294 289

ƒFig. 2 Alignment (a) and phylogenetic tree analysis (b) of amino acid (31 kDa). The molecular weights of fibrinolytic enzyme pro-
sequences of the fibrinolytic enzyme produced by B. subtilis MX-6 and duced by B. amyloliquefaciens FCF-11 (18.2 kDa) [22],
other homologous enzymes. a The fibrinolytic enzyme produced by
B. subtilis MX-6 was aligned with homologous enzymes: B. subtilis
Rhizopus chinensis 12 (18.0 kDa) and Paenibacillus
subtilisin NAT (WP 014479360.1), B. mojavensis peptidase S8 (WP polymyxa (118 and 49 kDa) [23] were also inconsistent with
010333625.1), Brevibacterium halotolerans peptidase S8 (WP our results, as were the molecular weights of the fibrinolytic
044156525.1), B. vallismortis peptidase S8 (WP 010329279.1), enzymes produced by Lactococcus lactis, Vagococcus lutrae,
B. licheni AprE 3-17 (ACU32756.1) and B. nakamurai peptidase S8
(WP 061521470.1). b Phylogenetic analysis showed that the
Vagococcus fluvialis, Weissella thailandensis and Bacillus
fibrinolytic enzyme produced by B. subtilis MX-6 clustered with methylotrophicus (35 to 116 kDa) [21]. These conflicting re-
homologous enzymes: B. subtilis subtilisin NAT (WP 014479360.1), sults indicate that the molecular weights of fibrinolytic en-
B. subtilis nattokinase precursor (AA065246.1) and B. subtilis subtilisin zymes from different microbes vary.
E (WP 009966941.1)

Fermentation Characteristics of Nattokinase

Molecular Weight of Nattokinase Produced by B. subtilis MX-6

We detected several bands when the crude enzyme produced B. subtilis MX-6 produced nattokinase only at very low levels
by B. subtilis MX-6 was analysed with SDS-PAGE (Fig. 3). between 0 and 12 h. Production of nattokinase increased be-
Overall purification of nattokinase from the culture superna- tween 12 and 72 h, peaking at 72 h. At 72 h, the diameter of
tant was 53.4-fold, with a recovery rate of 11.7%. These re- clear zone on the plasminogen-free fibrin plate was 21.60 mm.
sults indicated that the purified protein was approximately Production of nattokinase decreased between 72 and 102 h.
28 kDa and had fibrinolytic activity.
Our analyses predicted that the molecular weight of the Regulation of Nattokinase Production by Soybean
mature nattokinase produced by B. subtilis MX-6 was Components
27,712.72 Da, similar to that measured with SDS-PAGE
(about 28 kDa). Previous studies have suggested that the mo- Different concentrations of SPI did not significantly increase
lecular weight of microbial fibrinolytic enzymes ranges from the production of nattokinase by B. subtilis MX-6 (P > 0.05)
14 to 52 kDa [21]; the molecular weight of B. subtilis MX-6 (Fig. 4a). Bacteriocin production was highest in the basal
nattokinase found in this study was within this range. Our fermentation medium supplemented with 2% SPI.
results were consistent with those of Dabbagh [16], who found B. subtilis MX-6 nattokinase production increased signifi-
that the molecular weight of nattokinase was 27.7 kDa. cantly with increased concentrations of supernatant I and
However, our results were incongruent with the previously supernatant II (P < 0.01) (Fig. 4b, c). Nattokinase production
reported weights of fibrinolytic enzymes in other B. subtilis by B. subtilis MX-6 also increased significantly with in-
stains including B. subtilis DC33 (about 30 kDa) [18], creased concentrations of soy peptone in the basal fermenta-
B. subtilis A1 (about 31 kDa) [9] and B. subtilis BK-17 tion medium (P < 0.01) (Fig. 4d). Supplementing the basal
fermentation medium with different concentrations of Glu,
Marker 1 2 Asp, Arg and Ser significantly decreased bacteriocin produc-
tion (Fig. 5a–c, f) (P < 0.01), but production increased when
the medium was supplemented with Gly (Fig. 5d).
Nattokinase production varied with increased concentrations
of Ala (Fig. 5e) and Val (Fig. 5g) in the basal fermentation
medium.
To our knowledge, few previous studies have considered
the effect of soybean components on nattokinase production.
31.0kDa Our results showed that both supernatant I and supernatant II
increased nattokinase production in B. subtilis MX-6
(P < 0.01), while SPI had no significant effect (P > 0.05).
This difference may be explained by a difference in molecule
size: SPI is macromolecular and so may not directly enter the
21.5kDa cell, while supernatants I and II are micromolecular sub-
stances (polypeptides and amino acids precipitated from soy-
bean protein), as such, they can immediately be absorbed by
the cell. Thus, we hypothesized that micromolecular sub-
Fig. 3 SDS-PAGE image of crude enzyme and purified nattokinase. Lane stances played an important role in the induction of
1, crude enzyme; lane 2, purified nattokinase nattokinase production.
 290 Probiotics & Antimicro. Prot. (2019) 11:283–294

a 48h 72h 96h b 48h 72h 96h

Diameter of clear zone (mm)

Diameter of clear zone (mm)

**
**
**
**
**
**
** ** ** **
**
** ** **
**

Control 1%SPI 2%SPI 3%SPI 4%SPI 5%SPI Control 1% 2% 3% 4% 5%

Supernatant Supernatant Supernatant Supernatant Supernatant

48h 72h 96h

c d
**
48h 72h 96h **

Diameter of clear zone (mm)

Diameter of clear zone (mm)

**
** ** ** **
**
** **
** ** ** **
** **
** ** **
** ** ** **
** ** **
**

Control 1% 2% 3% 4% 5% Control 0.5%soy peptone 1%soy peptone 1.5%soy peptone 2%soy peptone
+1.5%peptone +1%peptone +0.5%peptone +0%peptone
Supernatant Supernatant Supernatant Supernatant Supernatant

Fig. 4 Effect of various supplements on nattokinase production by nattokinase is expressed as mean mm ± SD (n = 3). Compared to controls,
B. subtilis MX-6 cultured in basal fermentation medium. a SPI. b double asterisks and single asterisk indicate the significant difference at
Supernatant I. c Supernatant II. d Soy peptone. The fibrinolytic the level of 0.01 (P < 0.01) and 0.05 (P < 0.05), respectively. No asterisk
diameter of nattokinase produced by B. subtilis MX-6 in basal indicates that the difference is not significant (P > 0.05) with respect to
fermentation medium was used as control. The fibrinolytic diameter of controls

To investigate this, we tested the effect of soy peptone, a Regulation of Nattokinase Production with Inorganic
mixture of polypeptides and amino acids, on nattokinase pro- Salts
duction. We obtained soy peptone by hydrolysing soybean
protein with either an acid, an alkali or an enzyme. Our results Supplementation of the basal fermentation medium with var-
indicated that soy peptone induced nattokinase production in ious concentrations of K2HPO4 and KH2PO4 did not signifi-
B. subtilis MX-6. Liu reported that soy peptone was a better cantly affect B. subtilis MX-6 nattokinase production
substrate for nattokinase production in B. subtilis [24]. Our (P > 0.05) (Fig. 6a). However, the addition of MnSO4·H2O
results strongly suggested that the micromolecular substances, to the basal fermentation medium significantly decreased
including polypeptides and amino acids, induced bacteriocin nattokinase production (P < 0.01) (Fig. 6b). In contrast,
production. nattokinase production of B. subtilis MX-6 increased with
To determine whether B. subtilis MX-6 nattokinase pro- the addition MgSO4 and CaCl2 to the basal fermentation me-
duction was induced by polypeptides or amino acids, we in- dium; maximum bacteriocin production was recorded in the
vestigated the effects of various amino acids on nattokinase media supplemented with 0.04% MgSO4 (Fig. 6c) and 0.04%
production. Soybeans contain a variety of amino acids includ- CaCl2 (Fig. 6d).
ing Glu, Asp, Arg, Gly, Ala, Ser and Val. We thus tested the Various microbes use inorganic salts as part of their consti-
effects of these amino acids on soybean production. Our re- tution or as modulators of physiological active substances.
sults demonstrated that such nattokinase production was Mg2+ affects not only substance oxidation but also protein
inhibited by amino acids. Nattokinase is an extracellular en- synthesis and is an efficient activator of many important en-
zyme. Extracellular enzymes, especially proteases and amy- zymes. Ca2+ is required by some microbes such as Bacillus
lases, are usually inhibited by various end products including proteus to synthesize proteases. Mn2+ is used both for absorp-
amino acids, consistent with our results. Therefore, tion of phosphorus and for spore formation in B. subtilis. The
nattokinase production by B. subtilis MX-6 was induced by permeability of the cell is regulated by K+. Here, we found that
soybean polypeptides in an induction and end-product feed- B. subtilis MX-6 nattokinase production was increased by the
back inhibition loop. addition of MgSO4 and CaCl2 (Fig. 6). In contrast, MnSO4·
 Probiotics & Antimicro. Prot. (2019) 11:283–294 291

a 48h 72h 96h b 48h 72h 96h

Diameter of clear zone (mm)

Diameter of clear zone (mm)

** **
**
** **
** ** **
** **
** ** ** **
** **
** ** **
**
** **
** **

Control 0.05%Glu 0.1%Glu 0.15%Glu 0.2%Glu Control 0.05%Asp 0.1%Asp 0.15%Asp 0.2%Asp

c 48h 72h 96h

d 48h 72h 96h

Diameter of clear zone (mm)

Diameter of clear zone (mm)

** ** ** **
**
** **
** ** **
** ** ** **

** ** **
** ** ** **
**
**
**
Control 0.05%Arg 0.1%Arg 0.15%Arg 0.2%Arg Control 0.05%Gly 0.1%Gly 0.15%Gly 0.2%Gly

e 48h 72h 96h f 48h 72h 96h

Diameter of clear zone (mm)

Diameter of clear zone (mm)

** ** **
** **
** **
**
** ** **
** **
** **
**
** ** ** **
** ** **
**
Control 0.05%Ala 0.1%Ala 0.15%Ala 0.2%Ala Control 0.05%Ser 0.1%Ser 0.15%Ser 0.2%Ser

g 48h 72h 96h

Diameter of clear zone (mm)

** ** ** **

** ** ** **

** ** ** **

Control 0.05%Val 0.1%Val 0.15%Val 0.2%Val

Fig. 5 B. subtilis MX-6 nattokinase production with the addition of SD of mm (n = 3). Compared to controls, double asterisks and single
various amino acids to the basal fermentation medium. a Glu. b Asp. c asterisk indicate the significant difference at the level of 0.01 (P < 0.01)
Arg. d Gly. e Ala. f Ser. g Val. The fibrinolytic diameter of nattokinase and 0.05 (P < 0.05), respectively. No asterisk indicates that the difference
produced by B. subtilis MX-6 in basal fermentation medium was used as is not significant (P > 0.05) with respect to controls
control. The fibrinolytic diameter of nattokinase is expressed as means ±
 292 Probiotics & Antimicro. Prot. (2019) 11:283–294

a b
48h 72h 96h 48h 72h 96h
Diameter of clear zone (mm)

Diameter of clear zone (mm)

**

** **
** **
** **
**
**
Control K2HPO4:KH2PO4 K2HPO4:KH2PO4 K2HPO4:KH2PO4 Control MnSO4· H2O MnSO4· H2O MnSO4· H2O
0.1%:0.1% 0.2%:0.1% 0.6%:0.1% 10-5mol/L 10-4mol/L 10-3mol/L

c 48h 72h 96h d 48h 72h 96h

**
** **
**
Diameter of clear zone (mm)

**

Diameter of clear zone (mm)

**
** ** ** **
** **
**
** ** **
** **

Control 0.02%MgSO4 0.04%MgSO4 0.06%MgSO4 Control 0.02%CaCl2 0.04%CaCl2 0.06%CaCl2

Fig. 6 B. subtilis MX-6 nattokinase production with the addition of used as a control. The fibrinolytic diameter of nattokinase is expressed
various inorganic salts to the basal fermentation medium. a K2HPO4 as mean mm ± SD (n = 3). Compared to controls, double asterisks and
plus KH2PO4 (0.1:0.1%, 0.2:0.1% or 0.6:0.1%). b MnSO4·H2O (10−5, single asterisk indicate the significant difference at the level of 0.01
10−4 or 10−3 mol/L). c MgSO4 (0.02, 0.04 or 0.06%). d CaCl2 (0.02, 0.04 (P < 0.01) and 0.05 (P < 0.05), respectively. No asterisk indicates that
or 0.06%). The fibrinolytic diameter of the nattokinase produced by the difference is not significant (P > 0.05) with respect to controls
B. subtilis MX-6 in unsupplemented basal fermentation medium was

H2O had a negative impact on nattokinase production, and translocation across the plasma membrane, nattokinase may
neither K2HPO4 nor KH2PO4 had an obvious influence. The bind to the external surface of the cell membrane, producing a
increase in nattokinase production in the presence of MgSO4 feedback control response that restrains nattokinase produc-
and CaCl 2 indicated that these salts are essential for tion. After the replacement of the membrane-bound
nattokinase production. Our results are consistent with a pre- nattokinase with Mg2+ and Ca2+, nattokinase may be contin-
vious report, which suggested that the production of the fibri- uously synthesized. Under these circumstances, it is clear that
nolytic enzyme AprE 3-17 synthesized by B. licheniformis the addition of Mg2+ and Ca2+ would increase nattokinase
CH 3-17 was slightly inhibited by K+, strongly inhibited by production.
Mn2+ and enhanced by Mg2+ and Ca2 [25]. In contrast, however,
nattokinase production by B. subtilis A1 was enhanced by Mg2+
but inhibited by Ca2+ and Mn2+ [19]. In Bacillus natto NRRL Conclusions
3666, nattokinase production increased with MgSO4 concentra-
tion. In addition, serine proteinase production by B. subtilis was Nattokinase, an enzyme typically produced by food-grade mi-
enhanced in the presence of Ca2+, K+, and Mg2+ [26]. It is crobes, has attracted increased attention both as a food addi-
probable that discrepancies in the effects of various inorganic tive and as an oral thrombolytic. Here, we isolated B. subtilis
salts are due to differences among bacterial strains. MX-6, a strain with high production of nattokinase, as identi-
In B. subtilis MX-6, Mg2+ and Ca2+ may play an important fied by a plasminogen-free fibrin plate assay and sequence
role in the regulation of nattokinase production. That is, upon analysis. We then amplified aprN, the gene-encoding
Probiotics & Antimicro. Prot. (2019) 11:283–294 293

nattokinase in B. subtilis MX-6, with novel primers. We con- in silico studies on fibrinolytic activity of nattokinase: a clot buster
from Bacillus sp. Front Biol 12(3):219–225. https://doi.org/10.
structed a multiple sequence alignment and a phylogeny of the
1007/s11515-017-1453-3
inferred protein encoded by aprN and other homologous pro- 6. Toymentseva AA, Danilova IV, Tihonova AO, Sharipova MR,
teins. Next, we purified the crude enzyme. The molecular Balaban NP (2016) Purification of recombinant extracellular prote-
weight calculated by SDS-PAGE was about 28 kDa. ases from Bacillus pumilus for ß-amyloid peptide cleavage. Russ J
B i o o rg C h e m 4 2 ( 1 ) : 5 3 – 5 8 . h t t p s : / / d o i . o rg / 1 0 . 11 3 4 /
Maximum nattokinase production by B. subtilis MX-6 was
S1068162015060175
observed at 72 h, when the diameter of clear zone reached 7. Chhikara A, Sharma S, Chandra J, Nangia A (2017) Thrombin
21.60 mm on the plasminogen-free fibrin plate. The addition activable fibrinolysis inhibitor in beta thalassemia. Indian J
of soybean polypeptides, Mg2+ and Ca2+ increased B. subtilis Pediatr 84(1):25–30. https://doi.org/10.1007/s12098-016-2208-x
8. Godier A, Parmar K, Manandhar K, Hunt BJ (2016) An in vitro
MX-6 nattokinase production.
study of the effects of t-PA and tranexamic acid on whole blood
We believe that other workers can use our results to in- coagulation and fibrinolysis. J Clin Pathol 70:154–161
crease nattokinase yields, decreasing overall cost of 9. Zhang F, Zhang J, Linhardt RJ (2015) Interactions between
nattokinase synthesis. In addition, although the positive effect nattokinase and heparin/GAGs. Glycoconj J 32(9):695–702.
https://doi.org/10.1007/s10719-015-9620-8
of soy peptone on nattokinase production has been demon-
10. Chen HS, Qi SH, Shen JG (2017) One-compound-multi-target:
strated, it was previously unclear which components of soy combination prospect of natural compounds with thrombolytic
peptone were causing this increase. Here, we showed that therapy in acute ischemic stroke. Curr Neuropharmacol 15(1):
nattokinase production might be increased by soybean poly- 134–156
peptides. Finally, it is now possible to continue this work by 11. Singh R, Kumar M, Mittal A, Mehta PK (2016) Microbial enzymes:
industrial progress in 21st century. 3 Biotech 6(2):174
investigating the transciptome(s) of the gene(s) controlling 12. Nguyen TT, Quyen TD, Le HT (2013) Cloning and enhancing
nattokinase yield after soybean component addition. production of a detergent- and organic-solvent-resistant nattokinase
from Bacillus subtilis VTCC-DVN-12-01 by using an eight-
Acknowledgements We express our gratitude to Yan-Chen Man, Chun- protease-gene-deficient Bacillus subtilis WB800. Microb Cell
Hua Guo and Qian-Rong Xiang for their technical assistance. Factories 12(1):79. https://doi.org/10.1186/1475-2859-12-79
13. Duan C, Feng Y, Zhou H, Xia X, Shang Y, Cui Y (2015)
Optimization of fermentation condition of man-made bee-bread
Author Contributions Li-Li Man performed the experiments and contrib-
by response surface methodology. Adv Appl Microbiol 333:
uted significantly to the analysis and manuscript preparation. Dian-Jun
353–363
Xiang conceived and designed the experiments. Chun-Lan Zhang per-
14. Vaithilingam M, Chandrasekaran SD, Gupta S, Paul D, Sahu P,
formed the data analyses.
Selvaraj JN, Babu V (2016) Extraction of Nattokinase enzyme from
Bacillus cereus isolated from rust. Natl Acad Sci Lett 39(4):263–
Funding Information This study was supported by the Mudanjiang 267. https://doi.org/10.1007/s40009-016-0476-7
Science and Technology Plan Project of China (Nos. G2013n0012 and 15. Devi CS, Mohanasrinivasan V, Sharma P, Das D, Vaishnavi B,
Z2013n018). Naine SJ (2016) Production, purification and stability studies on
nattokinase: a therapeutic protein extracted from mutant
Compliance with Ethical Standards Pseudomonas aeruginosa CMSS isolated from bovine milk. Int J
Pept Res Ther 22(2):263–269. https://doi.org/10.1007/s10989-015-
Conflict of Interest The authors declare that they have no conflict of 9505-5
interest. 16. Dabbagh F, Negahdaripour M, Berenjian A, Behfar A,
Mohammadi F, Zamani M, Ghasemi Y (2014) Nattokinase: pro-
duction and application. Appl Microbiol Biotechnol 98(22):9199–
9206. https://doi.org/10.1007/s00253-014-6135-3
References 17. Nagata C, Wada K, Tamura T, Konishi K, Goto Y, Koda S, Kawachi
T, Tsuji M, Nakamura K (2016) Dietary soy and natto intake and
cardiovascular disease mortality in Japanese adults: the Takayama
1. Cai D, Zhu C, Chen S (2017) Microbial production of nattokinase: study. Am J Clin Nutr 105:426–431
current progress, challenge and prospect. World J Microb 18. Wang C, Ji B, Li B, Ji H (2006) Enzymatic properties and identifi-
Biotechnol 33(5):84. https://doi.org/10.1007/s11274-017-2253-2 cation of a fibrinolytic serine protease purified from Bacillus
2. Wei X, Luo M, Xie Y, Yang L, Li H, Xu L, Liu H (2012) Strain subtilis DC33. World J Microbiol Biotechnol 22(12):1365–1371.
screening, fermentation, separation, and encapsulation for produc- https://doi.org/10.1007/s11274-006-9184-7
tion of nattokinase functional food. Appl Biochem Biotechnol 19. Kotb E (2015) Purification and partial characterization of serine fibri-
168(7):1753–1764. https://doi.org/10.1007/s12010-012-9894-2 nolytic enzyme from Bacillus megaterium KSK-07 isolated from
3. Albuquerque W, Nascimento T, Brandão-Costa R, Fernandes T, kishk, a traditional Egyptian fermented food. Appl Biochem
Porto A (2017) Static magnetic field effects on proteases with fibri- Microbiol 51(1):34–43. https://doi.org/10.1134/S000368381501007X
nolytic activity produced by Mucor subtilissimus. 20. Anh DBQ, Mi NTT, Van Hung P (2015) Isolation and optimization
Bioelectromagnetics 38(2):109–120. https://doi.org/10.1002/bem. of growth condition of Bacillus sp. from fermented shrimp paste for
22016 high fibrinolytic enzyme production. Arab J Sci Eng 40(1):23–28.
4. Kapoor R, Panda BP (2013) Bioprospecting of soybean for produc- https://doi.org/10.1007/s13369-014-1506-8
tion of nattokinase. Nutrafoods 12(3):89–95. https://doi.org/10. 21. Thokchom S, Joshi SR (2014) Screening of fibrinolytic enzymes
1007/s13749-013-0032-8 from lactic acid bacterial isolates associated with traditional
5. Mohanasrinivasan V, Mohanapriya A, Potdar S, Chatterji S, Konne fermented soybean foods. Food Sci Biotechnol 23(5):1601–1604.
S, Kumari S, Merlyn Keziah S, Subathra Devi C (2017) In vitro and https://doi.org/10.1007/s10068-014-0217-y
294 Probiotics & Antimicro. Prot. (2019) 11:283–294

22. Kotb E (2014) Purification and partial characterization of a NLSSE using statistical experimental methods. Process Biochem
chymotrypsin-like serine fibrinolytic enzyme from Bacillus 40(8):2757–2762. https://doi.org/10.1016/j.procbio.2004.12.025
amyloliquefaciens FCF-11 using corn husk as a novel substrate. 25. Jo HD, Kwon GH, Park JY, Cha J, Song YS, Kim JH (2011)
World J Microbiol Biotechnol 30(7):2071–2080. https://doi.org/ Cloning and overexpression of aprE3-17 encoding the major fibri-
10.1007/s11274-014-1632-1 nolytic protease of Bacillus licheniformis CH 3-17. Biotechnol
23. Lu F, Sun L, Lu Z, Bie X, Fang Y, Liu S (2007) Isolation and Bioproc 16(2):352–359. https://doi.org/10.1007/s12257-010-
identification of an endophytic strain EJS-3 producing novel fibri- 0328-0
nolytic enzymes. Curr Microbiol 54(6):435–439. https://doi.org/10. 26. Artemov AV, Samuilov VD (1990) Effect of polyelectrolytes on
1007/s00284-006-0591-7 serine proteinase secretion by Bacillus subtilis. FEBS Lett 262(1):
24. Liu J, Xing J, Chang T, Ma Z, Liu H (2005) Optimization of nutri- 33–35. https://doi.org/10.1016/0014-5793(90)80146-A
tional conditions for nattokinase production by Bacillus natto