Biosci. Biotechnol. Biochem.

, 70 (2), 433–440, 2006

Purification and Characterization of Two Novel Halotolerant

Extracellular Proteases from Bacillus subtilis Strain FP-133
Endang S ETYORINI,1 Shinji T AKENAKA,2 Shuichiro M URAKAMI,2 and Kenji A OKI2; y
1
Division of Life Science, Graduate School of Science and Technology, Kobe University,
Rokko, Kobe 657-8501, Japan
2
Laboratory of Applied Microbiology, Department of Biofunctional Chemistry,
Faculty of Agriculture, Kobe University, Rokko, Kobe 657-8501, Japan

Received August 5, 2005; Accepted October 25, 2005

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

Bacillus subtilis strain FP-133, isolated from a reports on halotolerant proteases from the main indus-
fermented fish paste, synthesized two novel halotolerant trial protease producer Bacillus subtilis.6)
extracellular proteases (expro-I and expro-II), showing B. subtilis is a highly favorable bacterium for protease
activity and stability at concentrations of 0–20% (w/v) production, because it is non-pathogenic, well explored
NaCl. Each protease was purified to homogeneity and as a model of gram-positive bacteria, and capable of
characterized. The purified expro-I was a non-alkaline synthesizing various types of protease.7–9) Since B. sub-
serine protease with an optimum pH of 7.5, although tilis is classified as a halotolerant bacterium growing
most serine proteases from Bacillus strains act at the well in a range of 0–10% of NaCl, it can be expected to
alkaline side. The molecular mass of expro-I was produce several types of halotolerant protease. To obtain
29 kDa. The purified expro-II was a metalloprotease an optimum performance of protease under saline
with a molecular mass of 34 kDa. It was activated by conditions, it is necessary that the character of the
Fe2þ , which has never been reported as a bacterial halotolerant protease is well understood.
protease activator. At a concentration of 7.5% (w/v) In this study, we purified and characterized two
NaCl, both proteases preferred animal proteins to halotolerant extracellular proteases (expro-I and expro-
vegetable proteins as natural substrates. In addition, II) from B. subtilis strain FP-133, isolated previously.
under saline conditions, expro-I and II showed high The hydrolytic activity of the enzymes for some natural
catalytic activity toward gelatin and casein respectively. proteins was also measured to determine potential
prospects for a wide range of industrial uses.
Key words: halotolerant protease; Bacillus subtilis; neu-
tral serine protease; Fe-containing metal- Materials and Methods
loprotease; endoprotease
Materials. Casamino acids were purchased from
The saline fermentation process, involved in the Difco Laboratories (Detroit, MI); yeast extract and
production of various protein-rich foods, is important to Polypepton were from Nihon Seiyaku (Tokyo); casein
prevent putrefaction and development of food poison- Hammerstein grade was from ICN Biomedicals (Aurora,
ing, as well as to produce desired flavor.1) That process OH); N-succinyl-Ala-Ala-Ala-p-nitroanilide was from
concerns hydrolysis of protein as the main process. Sigma-Aldrich (St. Louis, MO); Gly-Leu, was from
Proteases able to maintain high activities under moder- Peptide Institute (Osaka, Japan); DE52 cellulose and
ate saline conditions are thus essential. Halophilic CM52 cellulose were from Whatman (Madison, WI);
proteases are less suitable for this process, because they CM-Toyopearl 650S, Phenyl-Toyopearl 650M, and
need at least 12.5% (w/v) NaCl for expression of high Toyopearl HW-55 SF were from Tosoh (Tokyo); and
activity.2) Therefore, halotolerant proteases, which are other chemicals used were of guaranteed grade.
active at both low and high concentrations of NaCl, are
needed. Strain and culture conditions. Bacillus subtilis strain
Several halotolerant proteases have been purified FP-133 isolated from Terasi, a Southeast Asian tradi-
from microorganisms involved in saline food fermenta- tional fish paste fermented under saline conditions, was
tion. Some of them are produced by Filobacillus sp. used throughout this study. Various carbon and energy
RF2–5 isolated from fish sauce,3) Penicillium chrysoge- sources and nitrogen sources were evaluated to establish
num Pg222 from cured ham,4) and Pseudoalteromonas the optimum conditions for the production of halotoler-
sp. strain CP7 from soil,5) but there have been few ant proteases. The basal medium consisted of 5 g NaCl,
y
To whom correspondence should be addressed. Fax: +81-78-882-0481; E-mail: [email protected]
434 E. SETYORINI et al.

1 g KH2 PO4 , 0.3 g Na2 HPO4 .12H2 O, and 0.02 g for 10 min to obtain the supernatant as a protease source
.
MgSO4 7H2 O in 100 ml H2 O. The pH of the medium (fraction 1, 500 ml). Fraction 1 was brought to 85%
was adjusted to 5.5. Carbon and energy sources (1% w/v saturation of (NH4 )2 SO4 . The precipitate was collected
each at final concentration), including D-arabinose, D- by centrifugation at 15;000  g for 10 min and filtration
fructose, D-galactose, D-mannose, D-raffinose, D-glucose, through Whatman filter no. 2 paper (Whatman, Madi-
lactose, maltose, and glycerol and nitrogen sources (1% son, WI) and dissolved in buffer A. The solution was
w/v) including yeast extract, Polypepton, malt extract, dialyzed overnight against 20 mM Tris–HCl buffer
Casamino acids, and casein were added to the basal (pH 7.3) containing 10% sucrose (buffer B). The dia-
medium. lyzed solution (fraction 2, 80 ml) was loaded onto a
Four milliliters of pre-culture was transferred into DE52 cellulose column (1:6  12:5 cm) previously
400 ml of medium. It was then incubated at 30
C at equilibrated with buffer B. When the column was
140 rpm on a reciprocal shaker for 36 h. Cell growth was washed with buffer B, proteases were eluted and active
measured by monitoring absorbance at 660 nm (OD660 ). fractions were pooled (fraction 3, 50 ml). The other
Protease activity was measured as caseinolytic activity, proteases were not eluted with a linear gradient of 0–
as described below. 0.4 M NaCl in buffer B. Fraction 3 was dialyzed against

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

20 mM potassium phosphate buffer (pH 6.4) containing
Enzyme assays. Protease (caseinolytic) activity was 10% (w/v) sucrose (buffer C). The dialyzed solution
assayed by a modification of the method of Kunitz.10) (fraction 4, 50 ml) was loaded onto a CM52 cellulose
The reaction mixture (800 ml) consisted of 0.5% (w/v) column (2:2  16:5 cm) equilibrated with buffer C.
casein Hammerstein grade in 20 mM Tris–HCl buffer When the column was washed with buffer C, one peak
(pH 7.3) (buffer A) containing 7.5% (w/v) NaCl. The with protease activity (expro-II) appeared and the active
reaction was started by adding 100 ml of enzyme fractions were pooled (fraction 5, 35 ml). Other proteins
solution. After incubation for 30 min at 37
C, the were eluted with a linear gradient of 0–0.3 M NaCl in
reaction was stopped by adding 450 ml of 10% (w/v) buffer C. Another protease (expro-I) was obtained from
trichloroacetic acid, and then kept on ice for another this chromatography and the active fractions were
10 min, followed by centrifugation at 15;000  g for pooled (fraction 6, 45 ml). Fraction 6 were loaded onto
10 min. The absorbance of the supernatant was meas- a CM-Toyopearl column (2:2  13 cm) equilibrated
ured against a blank (non-incubated sample) at 280 nm. with buffer C. Proteins were eluted with a linear
One unit of activity was defined as the amount of gradient of 0–0.3 M NaCl in buffer C and the active
enzyme that made an increase in absorbance of 1 in 1 h fractions were pooled (fraction 7, 25 ml). Fraction 7 was
under the conditions described. Protein concentrations loaded onto a Phenyl-Toyopearl column (1:6  11:5 cm)
were measured by the method of Lowry et al.11) equilibrated with buffer C containing 1 M (NH4 )2 SO4 .
For the kinetic study of protease I (expro-I) produced Proteins were eluted with a linear gradient of 1.0–0 M
by strain FP-133, N-succinyl-Ala-Ala-Ala-p-nitroanilide (NH4 )2 SO4 . The purity of the enzyme in each fraction
was used as a substrate. The reaction mixture, composed was verified by polyacrylamide gel electrophoresis
of 1 ml of buffer A containing the substrate at various (PAGE). Fractions showing a single band on a poly-
concentrations and 7.5% (w/v) NaCl and 0.15 ml of acrylamide gel were pooled.
enzyme (3 mg) solution, was incubated at 45
C. Ab- Fraction 5, containing expro-II, was loaded onto a
sorbance at 410 nm was measured every 5 min to Phenyl-Toyopearl column (1:6  11:5 cm) equilibrated
estimate released p-nitroaniline.12) A molar extinction with buffer C, containing 0.8 M (NH4 )2 SO4 . Proteins
coefficient of 4:0  103 for p-nitroaniline was used. were eluted with a linear gradient of 0.8–0 M
The kinetic study of protease-II (expro-II) was carried (NH4 )2 SO4 , the active fractions were pooled, and the
out using Gly-Leu as a substrate, by a modification of solution was dialyzed overnight against buffer C. The
the method of Kessler and Yaron.13) A reaction mixture dialyzed solution (fraction 8, 20 ml) was loaded onto a
composed of 0.8 ml of buffer A containing the substrate DE52 cellulose column equilibrated with buffer C.
at various concentrations, 7.5% (w/v) NaCl, and 0.1 ml When the column was washed with buffer C, expro-II
of enzyme (2 mg) solution was incubated at 45
C. The was eluted. Fractions showing a single band on a SDS–
reaction was initiated by adding enzyme solution to the polyacrylamide gel were pooled. The other proteases
reaction mixture. After 30 min, the reaction was stopped were not eluted with a linear gradient of 0–0.3 M NaCl
by adding 1 ml each of ninhydrin reagent and 4 M in buffer C.
sodium acetate buffer (pH 5.5). The released amino
acids were then measured by the ninhydrin method.12) A Determination of molecular masses. The purified
molar extinction coefficient of 1:5  103 at 570 nm for enzyme solution was concentrated to 1.0 ml with a
an equimolar mixture of glycine and L-leucine was used. collodion bag (Sartorius, Goettingen, Germany). The
concentrated sample and size markers (catalase 210 kDa,
Purification of enzymes. All purification steps were
-globulin 160 kDa, bovine serum albumin 67 kDa,
performed at 4
C. The culture, harvested at an early ovalbumin 45 kDa, and myoglobin 17 kDa) were loaded
stationary growth phase, was centrifuged at 15;000  g onto a column (2  92 cm) of Toyopearl HW-55 SF
 Novel Halotolerant Proteases from B. subtilis 435

equilibrated with buffer C, containing 0.2 M NaCl and Results

eluted with the same solution. The molecular mass of the
subunit of the enzymes was determined by SDS– Effects of carbon, energy, and nitrogen sources on
PAGE.14) An LMW peptide calibration kit for SDS– protease production
PAGE (Amersham Biosciences, Little Chalfont, Buck- Among the nine carbon and energy sources tested,
inghamshire, U.K.) was used for size markers. lactose and D-galactose promoted the production of
extracellular proteases in the presence of 1% (w/v)
Determination of NH2 -terminal amino acid sequence. casein as nitrogen source, although the two substrates
The purified enzyme was electroblotted by the method did not increase the cell growth of B. subtilis strain FP-
of Matsudaira.15) The NH2 -terminal amino acid se- 133. On the other hand, the production of the proteases
quence was identified with a Shimadzu PPSQ-10 protein was not increased by D-glucose or D-fructose, despite
sequencer (Shimadzu, Kyoto, Japan). promotion of cell growth. We observed that 1% (w/v)
lactose was optimal for the production of halotolerant
Effects of temperature, pH, and NaCl concentration extracellular proteases.
on the activity and stability of the purified enzymes. The Then we tested the effect of nitrogen source on

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

effect of temperature on enzyme activity was determined protease production in the medium containing 1% (w/v)
by the method described above at temperatures from 30 lactose as carbon and energy source. Of the five nitrogen
to 80
C in 5
C increments. The optimum pH of the sources tested, yeast extract led to the highest activity of
enzyme was determined at 37
C using the following protease in the culture. Polypepton promoted protease
20 mM buffers: sodium acetate (pH 3.0–5.5), sodium- production moderately. Then we examined the optimum
potassium phosphate (pH 5.0–8.0), Tris–HCl (pH 7.0– combination of yeast extract and Polypepton to give the
9.5), and sodium carbonate-bicarbonate (pH 9.0–11.0). highest protease production. The culture medium con-
The effect of NaCl concentration on caseinolytic activity taining 1% (w/v) each of yeast extract and Polypepton
was tested using 20 mM sodium-potassium phosphate was most favorable for the production of proteases and
buffer (pH 6.5) containing 0–20% (w/v) NaCl. their recoverability; when B. subtilis strain FP-133 was
The effect of temperature on the stability of the incubated in this medium, proteins including proteases
enzyme was determined by incubating the enzyme in the culture were precipitated with (NH4 )2 SO4 without
(7.5 mg) at 0–75
C for 10 min. The remaining activity difficulty. On the basis of these results, the medium
was then measured under optimum conditions. The containing 1% (w/v) each of lactose, yeast extract, and
effect of pH on the stability of the enzyme was Polypepton was established for the production of extra-
determined by dialyzing the enzyme (7.5 mg) against cellular proteases. Under these conditions, OD660 of 9 in
buffers with various pHs at 4
C overnight, and the the cell growth and units/OD660 of 540 in the production
remaining activity was measured. For the study of of proteases were obtained. On the other hand, in the
halostability, the enzyme (7.5 mg) was incubated in original medium containing 1% (w/v) D-glucose as car-
buffer B, containing 0–20% (w/v) NaCl at 4
C for 24 h. bon and energy source and 1% (w/v) casein as nitrogen
The remaining activity was assayed. source, OD660 of 11 in the cell growth and units/OD660
of 21 in the production of proteases were obtained,
Substrate specificity. The reaction mixture consisted showing that under the optimum conditions, the protease
of 0.4% (w/v) protein in 400 ml of 50 mM sodium productivity of strain FP-133 increased 26-fold.
phosphate buffer (pH 6.5) containing 7.5% (w/v) NaCl
and 150 ml of enzyme (4 mg) solution. After incubation at Purification of proteases and their molecular proper-
37
C for 1 h, 0.1 ml of the mixture was withdrawn and ties
the increase in the amount of free amino groups was Table 1 is a summary of a typical enzyme purification
determined by the ninhydrin method.12) The activity for for expro-I and II from B. subtilis strain FP-133. The
each substrate was expressed as the amount of amino specific activities of the final preparations of expro-I and
group (mmol as glycine) per hmgenzyme. II were 270 and 120 units per mg with overall recoveries
The preparation of natural substrates for expro-I and of 6 and 0.7% respectively. The final enzyme prepara-
II was initiated with inactivation of possible enzymes tions of expro-I and II showed 100- and 44-fold
contained in the materials by heating at 120
C for increases respectively in their specific activities. The
10 min. The materials were ground in order to extract final preparation of expro-I showed single protein
proteins with 20 mM Tris–HCl buffer (pH 7.3). After bands on polyacrylamide and SDS–polyacrylamide gels
filtration of the extracts with Whatman filter no. 2 paper, (Fig. 1). That of expro-II also showed a single band on
the protein concentrations of the filtrate were estimated SDS–polyacrylamide gel. Expro-II was not observed by
by the method of Lowry et al.11) native PAGE using pH 9.5,16) 8.0,17) or 4.318) gels.
The molecular masses of expro-I and II were 29 and
Assay of iron. The Fe content in expro-II was 33 kDa by gel filtration on Toyopearl HW-55S and 29
measured with a Hitachi polarized Zeeman 180–80 and 34 kDa by SDS–PAGE respectively. These findings
atomic absorption spectrophotometer (Hitachi, Tokyo). indicate that the two proteases are monomers.
436 E. SETYORINI et al.
Table 1. Summary of Purification of Expro-I and II General properties of the purified proteases
Total Total Specific
The optimum pH, effect of temperature on enzyme
Recovery activity, pH stability, thermostability, kinetic values, and
activity protein activity
(%)
(U) (mg) (U/mg) N-terminal amino acid sequences of the purified expro-I
Culture supernatant 7,200 2,700 2.7 100 and II are listed in Table 2. Based on the DDBJ
(NH4 )2 SO4 1,400 150 5.3 20 database, the N-terminal amino acid sequence of expro-I
DE52 1,100 26 42 15 showed 76% identity to pro-subtilisin from Bacillus sp.
Expro-I DJ-4,19) under accession no. AY627764, whereas the N-
CM52 680 18 38 9 terminal amino acid sequence of expro-II did not show
CM-Toyopearl 440 7.6 58 6 any identity to those from other proteins. Because of the
Phenyl-Toyopearl 460 1.7 270 6 optimum pH for activity, the two proteases exhibited the
Expro-II characteristics of neutral protease.
CM52 290 11 27 4
Phenyl-Toyopearl 96 0.7 140 1.3
Effect of salt concentration on the activity and
2nd DE52 49 0.4 120 0.7
stability of the proteases

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

As Fig. 2 shows, expro-I and II maintained significant
activities at a concentration of 20% (w/v) NaCl,
although they exhibited the highest activity in the
(I) (II) absence of NaCl. In addition, high salt concentrations
were still favorable for the storage of the two enzymes.
(+) (-)
94 Expro-I maintained 100% activity even when it was
stored in 20% (w/v) NaCl for 24 h at 4
C, whereas
67
expro II kept 76% activity under the same conditions.
45
Effects of inhibitors, denaturing agents, and metal
30 ions on enzyme activity
Expro-I was strongly inhibited by serine protease
20 inhibitors, such as phenylmethylsulfonyl fluoride
kDa
(PMSF), diisopropyl fluorophosphate (DFP), and chy-
14
(-) mostatin (Table 3). However, other serine protease
(+)
inhibitors, chymotrypsin-like serine protease inhibitor,
A A B M N -p-tosyl-L-lysine chloromethyl ketone hydrochloride
(TLCK), and leupeptin, did not inhibit the protease. The
Fig. 1. Native PAGE (I) and SDS–PAGE (II) of Purified Expro-I (A)
and Expro-II (B).
metalloprotease inhibitor EDTA, and the cysteine pro-
I, Native PAGE. The purified enzyme (10 mg) was run on 7.5% tease inhibitors p-chloromercuribenzoic acid (PCMB),
(w/v) gels of pH 4.3 in a running buffer (pH 4.5) of -alanine and ICH2 COOH, and HgCl2 did not inhibit expro-I. In
acetic acid.18) II, SDS–PAGE. The purified enzymes (10 mg each) addition, expro-I was strongly inhibited by the denatur-
and markers (M) were run on 12.5% (w/v) gels containing 0.1% (w/
ing and reducing agents SDS, dithiothreitol (DTT), and
v) SDS in sodium phosphate buffer (pH 7.2).14) The gels used for
native PAGE and SDS–PAGE were stained with 0.25% (w/v) -mercaptoethanol, but not by urea (Table 3). On the
Coomassie Brilliant Blue R-250 in a solvent of ethanol–acetic acid– other hand, expro-II was markedly inhibited by the
H2 O (9:2:9, v/v). metalloprotease inhibitor EDTA. Expro-II was inhibited

Table 2. Characteristics of Expro-I and II

Parameter Expro-I Expro-II

Protease class Serine protease Metalloprotease
Molecular mass 29 kDa (native) 33 kDa (native)
29 kDa (denatured) 34 kDa (denatured)
Optimum pH 7.5 8
Temperature with highest 60
C 45
C
activity
pH stability ( 70% act.) 5.5–10 5.5–9
Thermostability 70
C 50
C
Kinetic value Km : 0.72 mM Km : 50 mM
Vmax : 1.2 mM Vmax : 270 mM glycine or L-
p-nitroaniline/minmg leucine/minmg
NH2 -terminal amino acid
sequence AESVPYGVSEIKAPAL ADATGXGGNQXTGQNY
 Novel Halotolerant Proteases from B. subtilis 437
Table 3. Effects of Inhibitors and Denaturing and Reducing Agents
on the Activities of Expro-I and II

Relative activity (%)

Compound
Expro-I Expro-II
a
Inhibitor
None 100 100
PMSF 0 125
Chymostatin 19 79
TLCK 95 104
Leupeptin 95 82
DFP 0 118
PCMB 104 107
ICH2 COOH 96 86
HgCl2 98 110
EDTA 2Na 71 25
Pepstatin A 94 82

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

n-Octyl alcohol 94 84
Phenacyl bromide 91 97
Denaturing and reducing
agentb
None 100 100
SDS 15 78
Urea 136 100
DTT 9 43
-mercaptoethanol 1 67
a
A reaction mixture (500 ml) containing 0.5% (w/v) casein, enzyme (4 mg),
Fig. 2. Effects of Salt Concentration on the Activity and Stability of 7.5% (w/v) NaCl, and 1 mM of each inhibitor in 50 mM sodium phosphate
Expro-I (A) and Expro-II (B). buffer (pH 6.5) was incubated at 37
C for 30 min. The remaining activity
The activities of expro-I and II were measured at various was then measured, as described in the text.
b
concentrations of KCl ( ) and NaCl ( ) at 37
C at pH 7.5. The A reaction mixture (400 ml) containing 0.5% (w/v) casein, enzyme (4 mg),
enzymes were incubated at various concentrations of NaCl ( ) at 7.5% (w/v) NaCl, and 2% (w/v) SDS, 4 M urea, 0.1% (w/v) DTT, or 0.1%
(v/v) -mercaptoethanol in 50 mM sodium phosphate buffer (pH 6.5) was
4
C overnight, and then the remaining activity was measured in the
incubated at room temperature for 1 h. The remaining activity was then
presence of 7.5% (w/v) NaCl.
measured.

by DTT and slightly by SDS. Expro-I and II were not

inhibited by the aspartic protease inhibitors pepstatin A,
phenacyl bromide,20) or n-octyl alcohol.20,21)
Table 4. Effects of Metal Ions on the Activities of Expro-I and IIa
Among the metal ions tested, expro-I was strongly
inhibited by Ni2þ , Ca2þ , Cu2þ , and Zn2þ (Table 4). On Relative activity (%)
Metal ion
the other hand, expro-II was not much inhibited by these Expro-I Expro-II
metal ions. The activity of expro-II increased in the
Holoenzyme
presence of Fe2þ . When expro-II was treated with None 100 100
10 mM EDTA, its activity was completely lost, but was Fe2þ 110 121
recovered by the addition of Fe2þ , Fe3þ , and Zn2þ Fe3þ 103 83
(Table 4). Other metal ions examined did not recover Mn2þ 85 106
the activity of the apoenzyme. Then we analyzed the Mg2þ 69 74
Ni2þ 7 60
iron in expro-II and found that the protease contained Ca2þ 11 97
1.0 mol of Fe per mol of protein. Cu2þ 0 38
Zn2þ 0 81
Substrate specificity Apoenzymeb
Expro-I and II hydrolyzed gelatin, casein, ovalbumin, None —c 0
and bovine serum albumin, with the highest activity for Fe2þ — 65
gelatin and casein respectively (Table 5). Among the Fe3þ — 74
Mn2þ — 0
natural substrates tested, mackerel meat, shrimp meat,
Zn2þ — 14
and tofu proteins were more susceptible to hydrolysis by
a
expro-I and II than the other proteins (Table 5). The experiments were carried out as described in Table 3, using 1 mM of
metal ions.
b
Apoenzyme was prepared by incubating expro-II (5 mg) with 10 mM
Discussion .
EDTA 2Na at 4
C for 1 h, when enzyme activity was completely lost.
.
After removal of EDTA 2Na by dialysis, the apoenzyme was dialyzed
against a buffer containing 1 mM of each metal ion for 4 h at 7
C, and then
In this study, we found two extracellular halotolerant the recovered enzyme activity was measured.
proteases, expro-I and II, in the culture of B. subtilis c
The apoenzyme of expro-I was not prepared.
438 E. SETYORINI et al.
Table 5. Substrate Specificity of Expro-I and II which show an optimum pH of 10.7,23,28)
mmol Gly/hmgenzyme
Expro-I and II showed high activity for gelatin,
Substrate casein, albumin, ovalbumin, and bovine serum albumin,
Expro-I Expro-II
indicating that the two enzymes are endoproteases.
Casein 44 42
Gelatin 52 36 Expro-II was strongly inhibited by EDTA, and its
Bovine serum albumin 25 17 identity as a metalloprotease was confirmed by treating
Ovalbumin 18 9.1 holo- and apo-types of this enzyme with metal ions
Mackerel meat protein 21 19 (Table 4). The prominent roles of Fe2þ in the enhance-
Shrimp meat protein 11 17 ment of activity of the holoenzyme and in the recovery
Shrimp shell protein 1.0 7.0 of activity of the apoenzyme were observed. In addition,
Soybean protein 0.19 6.1 this enzyme contained 1 mol of Fe per mol of protein.
Tofua protein 6.1 11
Okarab protein 1.8 4.5
This is the first report on a bacterial metalloprotease
Beef protein 11 5.1 containing Fe and activated by Fe2þ . Several proteases
a
from animals29) and Saccharomyces cerevisiae30) have
A traditional Japanese food made from soybeans; soybean curd.
been reported to retain Fe2þ in their protein molecules.

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

b
Waste produced in the process of tofu manufacture.
There were no NH2 -terminal amino acid sequences up to
16 residues of other proteins showing identity to that of
strain FP-133, and they were purified to homogeneity. expro-II.
The purified proteases showed activity at concentrations At any given pH, expro-II was adsorbed on neither
of 0–20% (w/v) NaCl. Some reports state that B. sub- anion nor cation exchangers. This feature was in accord
tilis secretes several types of extracellular protease, with the failure of expro-II protein bands to appear on
including the alkaline serine protease subtilisin, neutral pH 9.5, 8.0, and 4.3 gels in native PAGE. However,
protease A, bacillopeptidase F, metalloprotease, neutral when expro-II was run on the hydrophobic resin Phenyl-
protease B, and cell wall-associated extracellular pro- Toyopearl, it was adsorbed on the resin and eluted with
tease.9,22,23) However, only one halotolerant protease (NH4 )2 SO4 . These findings suggest that expro-II is
from Bacillus subtilis has been reported.6) It acts dominated by small uncharged and hydrophobic amino
optimally on the alkaline side, although this bacterium acid residues, such as L-alanine and glycine, which
was apparently equipped with halostress response probably bring the low net electric charges and hydro-
mechanisms like Kþ uptake and synthesis of compatible phobicity to the protein molecule of expro-II. In the near
solutes.24) This is the first report describing neutral future, we will clone the gene encoding expro-II and
halotolerant proteases produced by a B. subtilis strain, determine the complete amino acid sequences of the
because expro-I and II had an optimal pH between 7.5 protein. Another explanation for these characteristic has
and 8, which is favorable for industrial use. been suggested by Scopes,31) who describes slippery
Lactose and D-galactose promoted more protease proteins that are difficult to bind to any adsorbents due to
production by strain FP-133 than D-glucose or glycerol, non-electrostatic interactions and uneven distribution of
although lactose and D-galactose did not increase its cell charges over protein surfaces.
growth. In Penicillium expansum25) and Yersinia ruck- When we used Gly-Leu, a Km value of 50 mM was
eri,26) the synthesis of extracellular proteases is pro- obtained, which is higher than that of expro-I for N-
moted under malnutrition conditions rather than the best succinyl-Ala-Ala-Ala-p-nitroanilide, although different
condition for cell growth. In these microorganisms, substrates of peptide were used (Table 2). These results
including strain FP-133, protease production appears to suggest that expro-II has low affinity for peptides as
be sensitive to catabolite repression by rapidly assim- compared with expro-I.
ilable sugars such as D-glucose, D-fructose, and glycerol. The halotolerant proteases expro-I and II showed a
Expro-I was strongly inhibited by a broad spectrum of broad range of NaCl concentrations for the expression of
serine protease inhibitors. In addition, the NH2 -terminal activity and stability (Fig. 2) as compared with halo-
amino acid sequence of expro-I showed 76% identity philic proteases. For example, an extracellular serine
to pro-subtilisin from Bacillus sp. DJ-4.19) Therefore, protease from the halophilic archaeon Natrialba magadii
expro-I was classified as a serine protease. The reducing does not show any activity at a concentration less than
agents -mercaptoethanol and DTT inhibited the en- 5.8% (w/v) NaCl.32) Some reports suggest that the
zyme. -Mercaptoethanol has been reported to stabilize amino acid composition of a protein determines its
cysteine proteases by protecting the oxidation of halophilism or halotolerance.33) In general, the halo-
sulfhydryl groups in proteins, but -mercaptoethanol philism of an enzyme strengthens in the predominant
would cause the alteration of structure of enzymes by existence of acidic amino acid residues, an increased
reducing disulphide bonds in the proteins.27) Expro-I amount of small hydrophobic amino acid residues such
probably lost its activity due to alteration of the protein as glycine, L-alanine, and L-valine, and a decreased
structure in the presence of -mercaptoethanol and DTT. amount of aliphatic amino acid residues, but no reports
The optimum pH for expro-I was 7.5. This is different describing the characteristics of amino acid composition
from that of most serine proteases from B. subtilis, in halotolerant enzymes have been published. They will
 Novel Halotolerant Proteases from B. subtilis 439

be elucidated by determining the complete amino acid from Clostridium histolyticum. In ‘‘Methods in Enzy-
sequences of expro-I and II produced by B. subtilis mology’’ Vol. XLV part B, ed. Lorand, L., Academic
strain FP-133. Press, New York, pp. 544–545 (1976).
At a concentration of 7.5% (w/v) NaCl, expro-I and II 14) Weber, K., and Osborn, M., The reliability of molecular
weight determination by dodecyl sulfate–polyacrylamide
preferred to hydrolyze animal proteins rather than
gel electrophoresis. J. Biol. Chem., 244, 4406–4412
vegetable proteins. These results suggest the application (1969).
of the proteases in fish- or meat-based food fermenta- 15) Matsudaira, P., Sequence from picomole quantities of
tion. In addition, gelatin was markedly susceptible to protein electroblotted onto polyvinylidene difluoride
hydrolysis by the proteases. Since gelatin is a denatured membranes. J. Biol. Chem., 262, 10035–10038 (1987).
form of collagen, the two proteases should be useful for 16) Davis, B. J., Disc electrophoresis. II. Method and
the preparation of hydrolyzates from collagen, which application to human serum proteins. Ann. NY Acad.
is needed in the pharmaceutical, cosmetic, and food Sci., 121, 404–427 (1964).
industries. 17) Williams, D. E., and Reisfeld, R. A., Disc electro-
phoresis in polyacrylamide gels: extension to new
conditions of pH and buffer. Ann. NY Acad. Sci., 121,

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024

References 373–381 (1964).
18) Reisfeld, R. A., Lewis, U. J., and Williams, D. E., Disc
1) Margesin, R., and Schinner, F., Potential of halotolerant electrophoresis of basic proteins and peptides on poly-
and halophilic microorganisms for biotechnology. Ex- acrylamide gels. Nature, 195, 281–283 (1962).
tremophiles, 5, 73–83 (2001). 19) Choi, N. S., Chang, K. T., Maeng, P. J., and Kim, S. H.,
2) Ventosa, A. J., Nieto, J., and Oren, A., Biology of Cloning, expression and fibrinogenolytic properties of a
moderately halophilic aerobic bacteria. Micribiol. Mol. subtilisin DJ-4 gene from Bacillus sp. DJ-4. FEMS
Biol. Rev., 62, 504–554 (1998). Microbiol. Lett., 236, 325–331 (2004).
3) Hiraga, K., Nishikata, Y., Namwong, S., Tanasupawat, 20) Bankus, J. M., and Bond, J. S., Some commercially
S., Takada, K., and Oda, K., Purification and character- available proteases. In ‘‘Proteolytic Enzymes, a Practical
ization of serine protease from a halophilic bacterium Approach,’’ eds. Beynon, R., and Bond, J. S., Oxford
Filobacillus sp. RF2–5. Biosci. Biotechnol. Biochem., University Press, New York, p. 308 (2001).
69, 38–44 (2005). 21) Tang, J., Competitive inhibition of pepsin by aliphatic
4) Benito, M. J., Rodriguez, M., Nunez, F., Asensio, M. A., alcohols. J. Biol. Chem., 240, 3810–3815 (1965).
Bermudez, M. E., and Cordoba, J. J., Purification and 22) Park, C. H., Lee, S. J., Lee, S. G., Lee, W. S., and Byun,
characterization of an extracellular protease from Pen- S. M., Hetero- and autoprocessing of the extracellular
icillium chrysogenum Pg222 active against meat pro- metalloprotease (Mpr) in Bacillus subtilis. J. Bacteriol.,
teins. Appl. Environ. Microbiol., 68, 3532–3536 (2002). 186, 6457–6464 (2004).
5) Sanchez-Porro, C., Mellado, E., Bertoldo, C., 23) Wu, X. C., Lee, W., Tran, L., and Wong, S. L.,
Antranikian, G., and Ventoza, A., Screening and Engineering a Bacillus subtilis expression-secretion
characterization of the protease CP1 produced by the system with a strain deficient in six extracellular
moderately halophilic bacterium Pseudoalteromonas sp. proteases. J. Bacteriol., 173, 4952–4958 (1991).
strain CP76. Extremophiles, 7, 221–228 (2003). 24) Boch, J., Kempf, B., Schmid, R., and Bremer, E.,
6) Kembhavi, A. A., Kulkarni, P., and Pant, A., Salt- Synthesis of the osmoprotectant glycine betaine in
tolerant and thermostable alkaline protease from Bacillus Bacillus subtilis: characterization of the gbsAB genes.
subtilis NCIM no. 64. Appl. Biochem. Biotechnol., 38, J. Bacteriol., 178, 5121–5129 (1996).
83–92 (1993). 25) Dahot, M. U., Cultivation of Penicillium expansum on
7) Adinarayana, K., Ellajah, P., and Prasad, D. S., Purifi- rice husk powder for protease production. J. Islamic
cation and partial characterization of thermostable serine Academic Sci., 6, 29–33 (1993).
alkaline protease from a newly isolated Bacillus subtilis 26) Secades, P., and Guijarro, J. A., Purification and
PE-11. AAPS Pharm Sci. Tech., 4, 1–9 (2003). characterization of an extracellular protease from the
8) Kawamura, F., and Doy, R. H., Construction of Bacillus fish pathogen Yersinia ruckeri and effect of culture
subtilis double mutant deficient extracellular alkaline conditions in production. Appl. Environ. Microbiol., 65,
and neutral protease. J. Bacteriol., 160, 442–444 (1984). 3969–3975 (1999).
9) Valbuzzi, A., Ferrari, E., and Albertini, A. M., A novel 27) Scopes, R. K., ‘‘Protein Purification: Principle and
member of the subtilisin-like protease family from Practice,’’ Springer-Verlag, New York, pp. 195–196
Bacillus subtilis. Microbiol., 145, 3121–3127 (1999). (1982).
10) Kunitz, M., Crystalline soybean trypsin inhibitor. II. 28) Rao, M. B., Tanksale, A. M., Ghatge, M. S., and
General properties. J. Gen. Physiol., 30, 291–310 (1947). Deshpande, W., Molecular and biotechnological aspects
11) Lowry, O. H., Rosebrough, N. J., Farr, A. L., and of microbial proteases. Microbiol. Mol. Biol. Rev., 62,
Randall, R. J., Protein measurement with the Folin 597–635 (1998).
phenol reagent. J. Biol. Chem., 193, 265–275 (1951). 29) Gifford, A. H. T., Klippenstein, J. R., and Moore, M. M.,
12) Prescott, J. M., and Wilkes, H., Aeromonas amino- Serum stimulates growth and proteinase secretion by
peptidase. In ‘‘Methods in Enzymology’’ Vol. XLV part Aspergillus fumigatus. Infect. Immun., 70, 19–26 (2002).
B, ed. Lorand, L., Academic Press, New York, pp. 530– 30) Hiromura, M., and Sakurai, H., Intracellular metal
543 (1976). transport proteins. RIKEN Rev., 35, 23–25 (2001).
13) Kessler, E., and Yaron, A., Extracellular aminopeptidase 31) Scopes, R. K., ‘‘Protein Purification: Principle and
440 E. SETYORINI et al.
Practice,’’ Springer-Verlag, New York, pp. 67–90 magadii: purification and biochemical characterization.
(1982). Extremophiles, 4, 181–188 (2000).
32) Gimenez, M. I., Studdert, C. A., Sanchez, J. J., and 33) Madern, D., Ebel, C., and Zaccai, G., Halophilic
De Castro, R. E., Extracellular protease of Natrialba adaptation of enzymes. Extremophiles, 4, 91–98 (2000).

Downloaded from https://academic.oup.com/bbb/article/70/2/433/5953806 by guest on 16 May 2024