Previous Article | Next Article 
Applied and Environmental Microbiology, January 2000, p. 304-309, Vol. 66, No. 1
0099-2240/0/$04.00+0
Copyright © 2000, American Society for Microbiology. All rights reserved.
Efficient Production of Artificially Designed
Gelatins with a Bacillus brevis System
Tsutomu
Kajino,*
Haruo
Takahashi,
Masana
Hirai, and
Yukio
Yamada
Toyota Central Research and Development
Laboratories, Inc., Nagakute, Aichi 480-1192, Japan
Received 12 April 1999/Accepted 20 September 1999
 |
ABSTRACT |
Artificially designed gelatins comprising tandemly repeated
30-amino-acid peptide units derived from human 
I collagen were successfully produced with a Bacillus brevis system. The
DNA encoding the peptide unit was synthesized by taking into
consideration the codon usage of the host cells, but no clones having a
tandemly repeated gene were obtained through the above-mentioned
strategy. Minirepeat genes could be selected in vivo from a mixture of
every possible sequence encoding an artificial gelatin by randomly
ligating the mixed sequence unit and transforming it into
Escherichia coli. Larger repeat genes constructed by
connecting minirepeat genes obtained by in vivo selection were also
stable in the expression host cells. Gelatins derived from the
eight-unit and six-unit repeat genes were extracellularly produced at
the level of 0.5 g/liter and easily purified by ammonium sulfate
fractionation and anion-exchange chromatography. The purified
artificial gelatins had the predicted N-terminal sequences and amino
acid compositions and a solgel property similar to that of the native
gelatin. These results suggest that the selection of a repeat unit
sequence stable in an expression host is a shortcut for the efficient
production of repetitive proteins and that it can conveniently be
achieved by the in vivo selection method. This study revealed the
possible industrial application of artificially designed repetitive proteins.
| |
INTRODUCTION |
The relationship between function
and structure for proteins has been extensively studied, and the
consensus sequences for certain functions have become clear for some
proteins (6). Structural proteins, e.g., collagen and silk
proteins, generally have a repeat consensus sequence which is related
to the protein's unique properties. Gelatin can be obtained by the
partial hydrolysis of collagen as a mixture of tripeptide repeats with
various molecular masses and compositions. The physicochemical
properties of collagen and synthetic collagen analogs are well
understood and have been extensively reviewed (1). To create
new materials having novel properties, artificial repetitive proteins
have been designed on the basis of the consensus sequences of
structural proteins and expressed in Escherichia coli, which
acts as a host (4, 10, 11). However, little effort for the
efficient production of repetitive proteins has been made, so the
production levels of repetitive proteins remain low and thus are not
high enough for industrial applications.
A host-vector system involving Bacillus brevis as a host has
been used for the efficient extracellular production of many heterologous proteins, including prokaryotic and eukaryotic proteins (17). This system has two prominent features: heterologous
proteins are secreted directly into the culture medium in soluble and
biologically active forms, and the secreted proteins are often stable
because of the very low extracellular protease activity. However, no
application of this system to repetitive proteins has been reported,
except for the production of the casoxin D repeat by fusion to the
epidermal growth factor (9). Therefore, we explored the
possibility of efficient secretion of tandem repetitive proteins. Here,
we report the direct secretion and some properties of artificial
uniform gelatin analogs designed as tandem repeats of tripeptide units.
| |
MATERIALS AND METHODS |
Strains, plasmids, media, and reagents.
Plasmid pAN21 was
derived from pAN3 (17) by elimination of the restriction
sites PpuMI, AvaI, and BanII. Gene
cloning experiments were performed on E. coli XL1-Blue.
B. brevis 31-OK was used throughout this work for expression
studies (7). Plasmids pNU212 (17), pNU211L4
(14), and pNU211R2L4 (14) containing the signal
peptide of middle wall protein (MWP), the signal peptide L4, and the
signal peptide R2L4, respectively, were described previously. pNH326 was constructed from pNH300 (12) by changing the MWP signal peptide to an R2L6 (14) signal peptide. YC and YC-P2 media
(8) were used for the production of artificial gelatins.
Stability of insert DNA in the host cells.
The DNAs encoding
the repeat units of NEU were synthesized exactly according to the
sequences selected by the in vivo selection method. To compare the
stability in the host cells, a repeat unit DNA designed on the basis of
the codon usage of E. coli was also synthesized. These DNAs
were ligated into the BanII site in pAN21Neu under the same
conditions. E. coli HB101 was transformed with a ligation
reaction mixture containing 10 µg of insert DNA. The stability of an
insert DNA in the host cells was evaluated as the transformation
efficiency of each insert DNA.
Immunoblot analysis.
Immunoblot analysis was performed with
the use of anti-NEU antiserum and alkaline phosphatase-coupled goat
anti-rabbit immunoglobulin G (Bio-Rad Laboratories, Hercules, Calif.).
5-Bromo-4-chloro-3-indolylphosphate (BCIP) and nitroblue tetrazolium
were used for visualization. Anti-NEU antiserum was prepared as
follows. A 3-month-old New Zealand White rabbit was immunized with the
synthesized NEU peptide in Freund's complete adjuvant. After three
subsequent immunizations with the NEU peptide in Freund's incomplete
adjuvant, the rabbit was bled from the auricular artery, the blood was
allowed to clot, and then the antiserum was collected by centrifugation.
Purification of artificial gelatins.
Culture supernatants,
after 6 days of culture, were fractionated by the ammonium sulfate
precipitation method. Large portions of NEU8 and PHI6 were included in
the fractions obtained by 30 to 45% and 70 to 90% saturation with
ammonium sulfate, respectively. Each precipitate was dissolved in a
small volume of 20 mM Tris-HCl, pH 7.5 (buffer A) and then dialyzed
against the same buffer. After removal of the insoluble materials by
centrifugation, the dialyzed sample was applied to a Mono Q HR 5/5
column (Pharmacia) equilibrated with buffer A. Elution was performed
with an NaCl gradient (0 to 1 M) in buffer A. The eluate was dialyzed
against distilled water and then freeze-dried.
Amino acid sequencing and composition analysis and molecular mass
determination.
Amino acid sequencing was performed with an
automated amino acid sequencer (model 473A; Applied Biosystems). The
amino acid composition was determined by the method of Dreyer and Bynum
(3) with a Piko-Tag system (Millipore; Waters). Molecular
masses were determined by gel permeation chromatography. The purified
NEU8 and PHI6 were each put on a column of TSK gel G3000 SWXL (0.78 by
30 cm; Tosoh, Tokyo, Japan). Equilibration and elution were performed
with 100 mM sodium phosphate buffer (pH 7.0) containing 0.15 M sodium
chloride with or without 5 M guanidine hydrochloride (GnHCl) at the
flow rate of 1 ml/min. Ribonuclease A, ovalbumin, bovine serum albumin,
chymotrypsinogen, and catalase were used as molecular markers.
Viscosity of gelatin solutions.
The purified Neu8 and Phi6
were dissolved in phosphate buffer (100 mM
NaH2PO4, pH 7.0), and their viscosities were
measured with a corn-plate viscometer (EL; Tokisangyo, Tokyo, Japan)
after equilibrium was obtained at various temperatures. The viscosities of NEU8 and PHI6 solutions were compared with that of gelatin from
bovine skin (approximately 75 bloom, G6650; Sigma, St. Louis, Mo.).
Protein concentrations were determined by the dry weight method.
| |
RESULTS |
Design and gene synthesis of artificial gelatins.
On the basis
of the hydropathy of human I collagen, two units were selected from
a hydrophilic region, namely, PHI, and a moderately hydrophilic region,
NEU, for the construction of artificial gelatins. PHI and NEU each
comprise 30 amino acids of human I collagen, as follows: PHI,
1010-GESGREGAPGAEGSPGRDGSPGAKGDRGET-1039, and NEU,
1040-GPAGPPGAPGAPGAPGPVGPAGKSGDRGET-1069. The DNAs encoding NEU and PHI were designed by considering the codon usage of B. brevis in accordance with these two 30-amino-acid unit sequences attached to the C-terminal end of the MWP signal peptide for secretion. NcoI and HindIII restriction sites were
positioned at the 5' and 3' ends of the DNAs, respectively, for
insertion into the plasmid. BanII sites were used for random
oligomerization. The BspEI and XmaI site pair in
the Phi gene and the BamHI and BglII site pair in
the Neu gene were arranged for step-by-step tandem connection (Fig.
1A). The DNAs described above, NeuVec and
PhiVec, were each synthesized from four fully overlapping
oligonucleotides, N1 to N4 and P1 to P4, respectively, which were
initially in two pieces with overhangs, though the two gene pieces were
then combined. The N-terminal ends of the artificial gelatins were
fused in frame to the C-terminal end of the MWP signal peptide in the
constructs (Fig. 1A). Each resulting construct was inserted between the
NcoI and HindIII sites in pAN21, yielding the
plasmids pAN21Neu and pAN21Phi (Fig. 1B). These plasmids were used for
the construction of repetitive genes. To construct the repetitive genes
of Neu and Phi, the monomer units encoding NEU and PHI were initially synthesized by taking into consideration the codon usage of B. brevis, and then ligated at random into the BanII sites
of pAN21Neu and pAN21Phi, respectively. The resulting ligates were
transformed into E. coli. However, no transformant was
obtained even after 10 trials. Therefore, the repeat unit was
synthesized as a mixture of every possible sequence encoding an
artificial gelatin. Two pairs of oligonucleotides, N5 and N6 and P5 and
P6, were synthesized to construct the repeat units for the Neu and Phi
genes, respectively (Fig. 1C). Each pair of oligonucleotides was
designed so as to hybridize to each other in the complementary region
(Fig. 1C) at the 3' end of each fragment.
View larger version (35K):
[in this window]
[in a new window]
|
FIG. 1.
Construction of plasmids to oligomerize an artificial
gelatin unit. (A) Nucleotide and amino acid sequences of the artificial
gelatin unit. The gene was designed with reference to the codon usage
for cell wall proteins of B. brevis (16). The
gelatin unit was in frame following the MWP signal peptide. The
oligonucleotides synthesized are boxed and major restriction sites are
indicated. (B) The plasmids for oligomerizing the artificial gelatin
unit. The open and shadowed bars denote the signal sequence region of
the MWP gene of B. brevis 47 and the structural gene for an
artificial gelatin, respectively. Apr and ori represent the
ampicillin resistance gene and replication origin, respectively. (C)
Schematic diagram of the construction of a tandemly repeated gene.
First, a pair of oligonucleotides, N5 and N6, was annealed, and then
the rest of the bases were filled in with polymerase to synthesize a
monomer unit. Second, minirepeat genes were obtained by ligating the
monomer unit at random and in vivo selection. Third, larger repeat
genes were constructed by connecting the minirepeat genes step by step.
Dashed bars indicate the repeats of the artificial gelatin unit. X, the
number of repeat units. The exact sequences of N5 and N6 and P5 and P6
are indicated below the figure.
|
|
The rest of the bases were filled in with Vent polymerase (New England
Biolabs, Beverly, Mass.) after annealing, yielding
a double-stranded
DNA which encoded the 30-amino-acid repeat unit
of NEU (Fig.
1C). The
resulting DNA was digested with
BanII and
then randomly
ligated into the
BanII site in pAN21Neu.
E. coli was transformed with the resulting ligate (Fig.
1C). A total of
48 recombinant clones containing the tandem repeat of the Neu
gene (the
minirepeat gene) (2 clones with five repeats, 6 clones
with four
repeats, 16 clones with three repeats, and 24 clones
with two repeats)
were successfully obtained. The DNA sequences
of the cloned genes
revealed a bias of codon selection in the
wobble base of nearly all
codons (Fig.
2). Around the fixed
sequence
region, the codon bias was remarkable. This codon selection
bias
was obviously different from that of either
E. coli or
B. brevis.
The stability of sequences selected by the in
vivo selection method
was evaluated as the efficiency of transformation
into
E. coli (Table
1).
Although all three sequences (A, B, and C in Table
1) selected by the
in vivo selection method were efficiently
transformed and formed many
colonies, the sequence (D in Table
1) artificially designed on the
basis of the codon usage of
E. coli could not form any
colony. These findings suggest that the
sequences selected once by the
host cells are stable in the host.
View larger version (13K):
[in this window]
[in a new window]
|
FIG. 2.
Codon bias of the Neu gene repeat cloned in E. coli. A total of 82 Neu repeat units in 48 clones carrying the
correct structured plasmid were sequenced. The codon usage for each
amino acid of the repeat unit is shown. The codon bias was examined
statistically by means of the chi-square test with the null hypothesis
that the probabilities for codon selection are equal for all codons.
Asterisks show the statistically significant codon bias.
|
|
To construct larger repeat genes, the minirepeat genes obtained as
described above were step by step connected homogeneously
or
heterogeneously (
9,
10) (Fig.
1C). Thus,
E. coli
clones
containing larger repeat genes (8, 10, and 12 repeats, named
pAN21NEU8,
pAN21Neu10, and pAN21Neu12, respectively) were obtained, and
these
larger repeat genes were also stable in the cells. A clone
containing
the Phi repetitive gene (6 repeats, named pAN21Phi6) was
obtained
in the manner described above for the Neu repeat gene except
that
a different pair of oligonucleotides, P5 and P6, and plasmid
pAN21Phi
were
used.
Extracellular production of artificial gelatins by B. brevis.
These tandemly repeated genes were excised from the
plasmids by digestion with NcoI and HindIII
and then inserted between the NcoI and
HindIII sites in a B. brevis expression
vector, pNU212, and transformed into B. brevis 31-OK. No
B. brevis transformants containing larger repeat genes
composed of single minirepeats were obtained. A B. brevis
transformant carrying pNU212NEU8 or pNU212Neu10, which contained 8 or
10 repeats constructed by connecting a few minirepeat genes,
respectively, was obtained, but one containing 12 repeats was not. In
6-day culture supernatants of B. brevis carrying pNU212NEU8
and pNU212Neu10, a protein which cross-reacted with antiserum to the
synthesized NEU peptide was detected by sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Fig.
3A), indicating that these products were
derived from the Neu gene. We tried to increase the yield of NEU8 by
changing the expression vector to pNH326, which has a modified signal
peptide (R2L6) and only two of the five promoters of the MWP promoter region. The amount of NEU8 produced by pNH326NEU8 was about 10 times
greater than that produced by pNU212NEU8, reaching as much as 0.5 g of extracellular NEU8 protein per liter of culture on optimization of
the culture conditions, as estimated from the intensity of the band
compared with that of the purified NEU8 (Fig. 3B).
View larger version (39K):
[in this window]
[in a new window]
|
FIG. 3.
SDS-PAGE and immunoblot analysis of artificial gelatins
secreted by B. brevis. B. brevis 31-OK carrying an
expression vector was grown at 30°C. YC and YC-P2 media were used for
pNU212 and pNH326, respectively. Culture supernatants were subjected to
SDS-PAGE, followed by immunoblot analysis with anti-NEU antiserum (A)
or staining with Coomassie brilliant blue (B). Lanes 1 and 5, molecular
weight markers; lanes 2, 3, and 4, 5 µl of 6-day culture supernatants
of B. brevis carrying pNU212NEU8, pNU212Neu10, and pNU212,
respectively; lanes 6 and 12, 1 µg of purified NEU and PHI,
respectively; lanes 7 and 8, 5 µl of 3-day and 6-day culture
supernatants of B. brevis carrying pNH326NEU8; lane 9, 5 µl of a 6-day culture supernatant of B. brevis carrying
pNH326; lanes 10 and 11, 5 µl of 3-day and 6-day culture supernatants
of B. brevis carrying pNH326PHI6, respectively.
|
|
The synthetic gene encoding PHI6 was also excised with
NcoI
and
HindIII, and inserted between the
NcoI
and
HindIII sites of
pNH326, resulting in pNH326PHI6 for
the expression of PHI6. A
specific band of approx. 31.7 kDa was
detected by SDS-PAGE. The
sequence of the five N-terminal amino acids
of this protein matched
that deduced from the DNA sequence of the Phi
gene. Thus, we confirmed
that the 31.7-kDa product was derived from the
Phi gene. The amount
of PHI6 in the culture supernatant of pNH326PHI6
was estimated
to be 0.5 g/liter from the intensity of the band on
SDS-PAGE gels
(Fig.
3B). The amounts of NEU8 and PHI6 in the culture
supernatant
increased from the early stationary phase of growth (data
not
shown), and after 3 days the amounts of these gelatins had
increased
only slightly (Fig.
3B). These findings suggested that NEU8
and
PHI6 secreted into the medium are stable throughout the
cultivation.
Characterization of NEU8 and PHI6 produced by B. brevis.
The artificial gelatins were purified, yielding almost single-band
SDS-PAGE gels with one (NEU) or two (PHI) conventional steps, as
described in Materials and Methods, the yields being 81 and 53% of
total secreted NEU and PHI, respectively. The six N-terminal amino acid
residues of the purified NEU8 and PHI6 were identical to those deduced
from the DNA sequences, and this amino acid composition was also
similar to that deduced for the DNA. Although the molecular masses of
NEU8 and PHI6 estimated by SDS-PAGE were about 32.4 and 31.7 kDa,
respectively, which were about 1.5 and 2 times larger than the values
deduced from the DNA sequences, those in the presence of 5 M GnHCl were
estimated to be 20.3 kDa and 16.3 kDa, respectively, by gel filtration
chromatography (Fig. 4A). These values
were in good agreement with those deduced from the DNA sequences (19.3 and 15.6 kDa). In the absence of 5 M GnHCl, the molecular masses of
NEU8 and PHI6 were 59.6 and 50.3 kDa, respectively (Fig. 4B). Figure
5 shows the viscosity change with temperature. The viscosity of 1% solutions of NEU8 and PHI6 increased with decreasing temperature and decreased with increasing temperature, as in the case of the native gelatin solution. This reversible viscosity change was possible at least three times. These results suggested that the artificial gelatins have a solgel property like that
of the native gelatin. But the threshold temperature of the solgel
transition of an artificial gelatin was lower than that of the native
gelatin by about 15°C, and the viscosity changes of 1% solutions of
NEU8 and PHI6 from 60 to 5°C were similar to those of 0.3 and 0.1%
solutions of the native gelatin, respectively.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 4.
Gel filtration of the purified artificial gelatins on a
TSK gel G3000 SWXL column. Gel filtration chromatograms of the
artificial gelatins in the presence (A) and absence (B) of 5 M GnHCl.
Proteins were monitored as to the absorbance at 215 nm. BSA, bovine
serum albumin; OVA, ovalbumin; CT, chymotrypsinogen; RNase,
ribonuclease A.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
FIG. 5.
Viscosity change of gelatin solutions with temperature.
 , gelatin from bovine skin (1%);  , gelatin from bovine skin
(0.3%); ×, gelatin from bovine skin (0.1%);  , NEU8 (1%);  ,
PHI6 (1%). cp, centipoise.
|
|
| |
DISCUSSION |
We have successfully produced large amounts of repetitive
proteins, artificially designed uniform gelatins, with B. brevis. Codon selection according to the codon usage of the host
strain is generally good for efficient production. The stability in the host cells of a designed DNA is another important factor for the efficient production of heterologous proteins. However, we do not know
how to design a stable DNA in host cells because the sequences that are
stable in the host strain, especially repetitive ones, are not known.
In fact, tandem repeats composed of single units designed by taking
into consideration the codon usage of E. coli could not be
cloned into E. coli. However, we were able to obtain stable
tandem repeat DNA sequences encoding repetitive proteins by allowing
the host cells to select. The selected sequences exhibited a bias
skewed strongly from the codon usage of the host strain in the wobble
base, showing that stable sequences were selected and cloned by the
host cells. Based on these results, we emphasize that it is critically
important to select a stable sequence encoding a protein for expressing
a repetitive protein, and it is only possible to do so by the random
cloning method. Although six-unit repeats constructed from both single
and dual three-unit repeats could be cloned into E. coli,
only hetelorogous six-unit repeats derived from dual three-unit repeats
could be cloned into B. brevis. These results suggest that
B. brevis is more sensitive to repetitive sequences. We do
not know any obvious reason for these results, but speculate that
rec+ causes a phenotype of higher sensitivity to
a repetitive sequence in B. brevis than in E. coli XL1-Blue, which is a rec mutant strain.
The artificial gelatin productivity became obviously higher with the
use of pNH326, which contained a modified signal peptide and a weakened
promoter. Although the modified signal peptide increased the
productivity of many kinds of protein (7, 14, 15), in the
case of gelatin no transformant could be obtained with the use of
pNU211L4 or pNU211R2L4, which contained a modified signal peptide and
strong multiple promoters. These results suggest that a weakened
promoter is more favorable for the efficient production of repeative
proteins like artificial gelatins.
By SDS-PAGE, the molecular masses of the purified artificial gelatins
were estimated to be larger than those deduced from the DNA sequences.
This may be due to the mobilities of collagen-related peptides on
SDS-PAGE gels, which mobilities are lower than those of common proteins
used as molecular weight markers (5). The molecular masses
estimated in the presence of 5 M GnHCl, and the N-terminal amino acid
sequences and amino acid composition analysis results also coincided
with those of the gelatins produced by B. brevis. Gel
filtration chromatography with or without 5 M GnHCl revealed that both
NEU8 and PHI6 existed as aggregates and that these artificial gelatins
might form a collagen-like triple helical structure in solution at room temperature.
The artificial gelatins showed a thermoreversible viscosity change
similar to that of the native gelatin, suggesting that the artificial
gelatins have a solgel property like that of the native gelatin. But
the viscoelastic parameters of artificial gelatin and native gelatin
solutions were distinct from each other. The amino acid sequence and/or
composition of gelatin are well-known to affect the gelation property
of a solution of it (2), and the hydroxylation of proline
residues is also known to stabilize the gel structure (13).
The differences in the numbers of proline residues and the deficiency
of hydroxylation of proline residues in artificial gelatins seem to
cause the variation in the viscoelastic property.
This is the first report of the efficient extracellular production of
artificially designed gelatins, repetitive proteins. As the properties
of artificial gelatins can be manipulated by means of the amino acid
sequence, unique gelatins or collagen-related peptides having novel
properties, e.g., photochemically active materials for holography or
nonimmunogenic materials for medical use, can be produced by means of
the B. brevis system. Furthermore, this study showed that
the B. brevis system is useful for the production of other
repetitive proteins, e.g., elastin and silk protein, and revealed the
possible industrial application of artificially designed repetitive proteins.
| |
ACKNOWLEDGMENTS |
We are grateful to Shigezo Udaka of Tokyo University of
Agriculture and Kunihiko Gekko of Hiroshima University for valuable advice and fruitful discussions.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Toyota Central
Research and Development Laboratories, Inc., Nagakute, Aichi 480-1192, Japan. Phone: 81-561-63-8491. Fax: 81-561-63-6498. E-mail:
e0846{at}mosk.tytlabs.co.jp.
| |
REFERENCES |
| 1.
|
Bhatnagar, R. S., and R. S. Rapaka.
1976.
Synthetic polypeptide models of collagen: synthesis and applications, p. 479-523.
In
G. N. Ramachandran, and A. H. Reddi (ed.), Biochemistry of collagen. Plenum Press, New York, N.Y.
|
| 2.
|
Bhatnagar, R. S.,
N. Pattabiraman,
K. R. Sorensen,
R. Langridge,
R. D. MacElroy, and V. Renugopara-Krishnan.
1988.
Inter-chain proline: proline contacts contribute to the stability of the triple helical conformation.
J. Biomol. Struct. Dyn.
6:223-233[Medline].
|
| 3.
|
Dreyer, W., and E. Bynum.
1967.
High-voltage paper electrophoresis.
Methods Enzymol.
11:32-39.
|
| 4.
|
Goldberg, I.,
A. J. Salerno,
T. Patterson, and J. I. Williams.
1989.
Cloning and expression of a collagen-analog-encoding synthetic gene in Escherichia coli.
Gene
80:305-314[CrossRef][Medline].
|
| 5.
|
Hayashi, T., and Y. Nagai.
1980.
The anomalous behavior of collagen peptides on sodium dodecyl sulfate-polyacrylamide gel electrophoresis is due to the low content of hydrophobic amino acid residues.
J. Biochem.
87:803-808[Abstract/Free Full Text].
|
| 6.
|
Heslot, H.
1998.
Artificial fibrous proteins.
Biochimie
80:19-31[Medline].
|
| 7.
|
Kajino, T.,
K. Kato,
C. Miyazaki,
O. Asami,
Y. Yamada,
M. Hirai, and S. Udaka.
1999.
Isolation of a protease-deficient mutant of Bacillus brevis and efficient secretion of a fungal protein disulfide isomerase by the mutant.
J. Biosci. Bioeng.
87:37-42.
|
| 8.
|
Kajino, T.,
Y. Saito,
M. Hirai,
O. Asami,
Y. Yamada, and S. Udaka.
1998.
Extracellular production of an intact and biologically active human growth hormone by the Bacillus brevis system.
J. Ind. Microbiol. Biotechnol.
19:227-231.
|
| 9.
|
Kato, M.,
Y. Fujiwara,
A. Okamoto,
M. Yoshikawa,
H. Chiba, and S. Udaka.
1995.
Efficient production of Casoxin D, a bradykinin agonist peptide derived from human casein, by Bacillus brevis.
Biosci. Biotechnol. Biochem.
59:2056-2059[Medline].
|
| 10.
|
Lewis, R. V.,
M. Hinman,
S. Kothakota, and M. J. Fournier.
1996.
Expression and purification of a spider silk protein: a new strategy for producing repetitive proteins.
Protein Expr. Purif.
7:400-406[CrossRef][Medline].
|
| 11.
|
Martin, S. L.,
B. Vrhovski, and A. S. Weiss.
1995.
Total synthesis and expression in Escherichia coli of a gene encoding human tropoelastin.
Gene
154:159-166[CrossRef][Medline].
|
| 12.
|
Okamoto, A.,
A. Kosugi,
Y. Koizumi,
F. Yanagida, and S. Udaka.
1997.
High efficiency transformation of Bacillus brevis by electroporation.
Biosci. Biotechnol. Biochem.
61:202-203[Medline].
|
| 13.
|
Rosenbloom, J.,
M. Harsch, and S. A. Jimenez.
1973.
Hydroxyproline content determines the denaturation temperature of chick tendon collagen.
Arch. Biochem. Biophys.
158:478-481[CrossRef][Medline].
|
| 14.
|
Sagiya, Y.,
H. Yamagata, and S. Udaka.
1994.
Direct high-level secretion into the culture medium of tuna growth hormone in biologically active form by Bacillus brevis.
Appl. Microbiol. Biotechnol.
42:358-363[Medline].
|
| 15.
|
Takimura, Y.,
M. Kato,
T. Ohta,
H. Yamagata, and S. Udaka.
1997.
Secretion of human interleukin-2 in biologically active form by Bacillus brevis directly into culture medium.
Biosci. Biotechnol. Biochem.
61:1858-1861[Medline].
|
| 16.
|
Tsuboi, A.,
R. Uchihi,
T. Adachi,
T. Sasaki,
S. Hayakawa,
H. Yamagata,
N. Tsukagoshi, and S. Udaka.
1988.
Characterization of the genes for the hexagonally arranged surface layer proteins in protein-producing Bacillus brevis 47: complete nucleotide sequence of the middle wall protein gene.
J. Bacteriol.
170:935-945[Abstract/Free Full Text].
|
| 17.
|
Udaka, S., and H. Yamagata.
1993.
High-level secretion of heterologous proteins by Bacillus brevis.
Methods Enzymol.
217:23-33[Medline].
|
Applied and Environmental Microbiology, January 2000, p. 304-309, Vol. 66, No. 1
0099-2240/0/$04.00+0
Copyright © 2000, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Koide, T.
(2007). Designed triple-helical peptides as tools for collagen biochemistry and matrix engineering. Phil Trans R Soc B
362: 1281-1291
[Abstract]
[Full Text]
-
Li, C., Ng, M. L. P., Zhu, Y., Ho, B., Ding, J. L.
(2003). Tandem repeats of Sushi3 peptide with enhanced LPS-binding and -neutralizing activities. Protein Eng Des Sel
16: 629-635
[Abstract]
[Full Text]
-
Werten, M. W. T., Wisselink, W. H., Jansen-van den Bosch, T. J., de Bruin, E. C., de Wolf, F. A.
(2001). Secreted production of a custom-designed, highly hydrophilic gelatin in Pichia pastoris. Protein Eng Des Sel
14: 447-454
[Abstract]
[Full Text]
Top
Abstract
Introduction
