Significance of the 20-kDa subunit of heterodimeric 2-deoxy-scyllo-inosose synthase for the biosynthesis of butirosin antibiotics in Bacillus circulans

Article abstract of DOI:10.1271/bbb.66.1538

A gene (btrC2) encoding the 20-kDa subunit of 2-deoxy-scyllo-inosose (DOI) synthase, a key enzyme in the biosynthesis of 2-deoxystreptamine, was identified from the butirosin-producer Bacillus circulans by reverse genetics. The deduced amino acid sequence of BtrC2 closely resembled that of YaaE of B. subtilis, but the function of the latter has not been known to date. Instead, BtrC2 appeared to show sequence similarity to a certain extent with HisH of B. subtilis, an amidotransferase subunit of imidazole glycerol phosphate synthase. Disruption of btrC2 reduced the growth rate compared with the wild type, and simultaneously antibiotic producing activity was lost. Addition of NH₄Cl to the medium complemented only the growth rate of the disruptant, and both the growth rate and antibiotic production were restored by addition of yeast extract. In addition, a heterologous co-expression system of btrC2 with btrC was constructed in Escherichia coli. The simultaneously over-expressed BtrC2 and BtrC constituted a heterodimer, the biochemical features of which resembled those of DOI synthase from B. circulans more than those of the recombinant homodimeric BtrC. Despite the similarity of BtrC2 to HisH the heterodimer showed neither aminotransfer nor amidotransfer activity for 2-deoxy-scyllo-inosose as a substrate. All the observations suggest that BtrC2 is involved not only in the secondary metabolism, but also in the primary metabolism in B. circulans. The function of BtrC2 in the butirosin biosynthesis appears to be indirect, and may be involved in stabilization of DOI synthase and in regulation of its enzyme activity.

Full text of DOI:10.1271/bbb.66.1538

This article was downloaded by: [University of Auckland Library]
On: 09 October 2014, At: 13:42
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Significance of the 20-kDa Subunit of Heterodimeric
2-Deoxy-scyllo-inosose Synthase for the Biosynthesis
of Butirosin Antibiotics in Bacillus circulans
Hideyuki TAMEGAIa, Eriko NANGOa, Ayumi KOIKE-TAKESHITAa, Fumitaka KUDOa & Katsumi
KAKINUMAa
a Department of Chemistry, Tokyo Institute of Technology 2-12-1 O-okayama, Meguro-ku,
Tokyo 152-8551, Japan
Published online: 22 May 2014.
To cite this article: Hideyuki TAMEGAI, Eriko NANGO, Ayumi KOIKE-TAKESHITA, Fumitaka KUDO & Katsumi KAKINUMA
(2002) Significance of the 20-kDa Subunit of Heterodimeric 2-Deoxy-scyllo-inosose Synthase for the Biosynthesis of
Butirosin Antibiotics in Bacillus circulans , Bioscience, Biotechnology, and Biochemistry, 66:7, 1538-1545, DOI: 10.1271/
bbb.66.1538
To link to this article: http://dx.doi.org/10.1271/bbb.66.1538
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Biosci. Biotechnol. Biochem., 66 (7), 1538–1545, 2002
Signiˆcance of the 20-kDa Subunit of Heterodimeric 2-Deoxy-scyllo-inosose
Synthase for the Biosynthesis of Butirosin Antibiotics in Bacillus circulans
Hideyuki TAMEGAI, Eriko NANGO, Ayumi KOIKE-TAKESHITA
Fumitaka KUDO, and Katsumi KAKINUMA
,
†
Department of Chemistry, Tokyo Institute of Technology, 2-12-1 O-okayama, Meguro-ku,
Tokyo 152-8551, Japan
Received January 24, 2002; Accepted March 20, 2002
A gene (btrC2) encoding the 20-kDa subunit of 2-
deoxy-scyllo-inosose (DOI) synthase, a key enzyme in
the biosynthesis of 2-deoxystreptamine, was identiˆed
from the butirosin-producer Bacillus circulans by
reverse genetics. The deduced amino acid sequence of
BtrC2 closely resembled that of YaaE of B. subtilis, but
the function of the latter has not been known to date.
Instead, BtrC2 appeared to show sequence similarity to
a certain extent with HisH of B. subtilis, an amidotran-
sferase subunit of imidazole glycerol phosphate syn-
thase. Disruption of btrC2 reduced the growth rate
compared with the wild type, and simultaneously an-
tibiotic producing activity was lost. Addition of NH4Cl
to the medium complemented only the growth rate of
the disruptant, and both the growth rate and antibiotic
production were restored by addition of yeast extract.
In addition, a heterologous co-expression system of
sides can be classiˆed into two major groups. One is
those having an aglycone of fully-substituted
aminocyclitol which is shown or proposed to be
biosynthesized from myo-inositol. Streptomycin,
fortimicin, and spectinomycin are members of this
group, and genetic as well as enzymological studies
have been reported for their biosynthesis.3–5) The
other group of aminoglycosides has a common agly-
cone of 2-deoxystreptamine (DOS), and clinically
important major compounds are included in this
category, like neomycin, kanamycin, ribostamycin,
butirosin, gentamicin, tobramycin, sisomicin,
sagamicin etc. The biosynthesis of DOS-containing
aminoglycosides have been studied mostly by the use
of isotope-tracer technology with whole-cell systems
of producing organisms and blocked mutants.6–10)
However, until recently, no biosynthetic gene of the
DOS-containing aminoglycosides has been identiˆed.
Genetic analysis was once done on the butirosin-
producer Bacillus circulans and a region around the
self-resistance gene (butA) was analyzed.11) While a
butirosin-transporter gene (butB) was found in this
particular region, no butirosin biosynthetic gene was
identiˆed. The enzymes involved in the biosynthesis
of these antibiotics also remained unclear except for
some transamination-related enzymes.12,13) In our
laboratory, much study was devoted to 2-deoxy-
scyllo-inosose (DOI) synthase (Fig. 1), the key en-
zyme for the initial step of DOS biosynthesis, by
using isotope-labeled substrates or substrate ana-
logues.14–19) The DOI synthase catalyzes the car-
bocycle-forming reaction from glucose-6-phosphate
to DOI using NAD+ and Co2+ as cofactors. The mul-
ti-step enzyme process includes the ˆrst oxidation at
C-4 of glucose-6-phosphate, and subsequent elimina-
tion of a phosphate group from the activated ulose
intermediate follows to form an enol or enolate inter-
mediate. Subsequent reduction at C-4, the opening of
the hemiacetal ring, and the last aldol-type in-
tramolecular condensation between C-1 and C-6 take
place consecutively to form DOI.
btrC2 with btrC was constructed in Escherichia coli
.
The simultaneously over-expressed BtrC2 and BtrC con-
stituted a heterodimer, the biochemical features of
which resembled those of DOI synthase from B. circu-
lans more than those of the recombinant homodimeric
BtrC. Despite the similarity of BtrC2 to HisH the heter-
odimer showed neither aminotransfer nor amidotran-
sfer activity for 2-deoxy-scyllo-inosose as a substrate.
All the observations suggest that BtrC2 is involved not
only in the secondary metabolism, but also in the prima-
ry metabolism in B. circulans. The function of BtrC2 in
the butirosin biosynthesis appears to be indirect, and
may be involved in stabilization of DOI synthase and in
regulation of its enzyme activity.
Key words: butirosin-binsynthesis;
2-deoxy-scyllo-
inosose synthase; 20-kDa subunit
Aminoglycosides are among the most important
antibiotics in clinical demands for a long time. Even
today, the importance of aminoglycosides is not
diminished, for example, in the treatment of HIV
infection.1,2)
In view of their chemical structure, aminoglyco-
†
To whom correspondence should be addressed. Fax: +81-3-5734-3713; E-mail: kakinuma
＠
chem.titech.ac.jp

20-kDa Subunit of 2-Deoxy-scyllo-inosose Synthase
1539
Recently, we puriˆed DOI synthase from the
and intriguing to ˆgure out the features of the
20-kDa subunit.
20)
butirosin-producer B. circulans
.
The enzyme was
composed of two subunits with molecular mass of
40-kDa and 20-kDa, respectively, and NAD+ and
Co2+ were proved to be necessary for its enzymatic
activity. Further, we isolated the gene of the large
In this study, we identiˆed the gene of DOI syn-
thase 20-kDa subunit, constructed a gene disruptant,
and heterologously expressed it with btrC in E. coli
.
The properties of the 20-kDa subunit are also dis-
cussed on the basis of physiological studies.
subunit (btrC ), and over-expressed it in Escherichia
21)
coli
.
Subsequently, the butirosin-biosynthetic gene
22)
cluster was found in the locus of btrC
.
All these
Materials and Methods
works appeared to be the ˆrst for both the biosyn-
thetic enzymes and genes of 2-deoxystreptamine-
containing aminoglycoside antibiotics. Surprisingly,
heterologously-expressed BtrC showed the over-all
enzymatic activity by itself without the presence of
the 20-kDa subunit. However, there were some sig-
niˆcant diŠerences between the natural enzyme from
B. circulans and the recombinant BtrC expressed in
E. coli in views of protein stability and kinetic
parameters. Further, we were unable to identify the
gene responsible for the 20-kDa subunit around
btrC, at least in the regions of up- and downstreams
of about. 18 kb.22) Thus, it appeared to be important
Organisms. B. circulans SANK72073 was cultured
in a butirosin-producing medium as described previ-
ously,20) or in a glucose minimal medium.23) E. coli
JM109 was cultured in LB medium or on LB-agar
containing 50
mg ml of ampicilin or 30 mg ml of
W
W
kanamycin, when necessary.
Identiˆcation of a partial amino acid sequence of
DOI synthase 20-kDa subunit. The N-terminal amino
acid sequence of the 20-kDa subunit of DOI synthase
was analyzed as follows. Puriˆed DOI synthase from
B. circulans was separated into each subunit on
12.5
z
Tricine-SDS-PAGE,24) and subsequently
transferred to a PVDF membrane (Millipore) by
electroblotting. The membrane was stained with
Coomassie brilliant blue. A band of interest was ex-
cised, and its N-terminal amino acid sequence was
analyzed with
a
protein sequencer, PPSQ-21
(Shimadzu, Japan). Analysis of internal amino acid
sequences was done by the APRO Life Science In-
stitute (Japan).
Cloning of the gene. DNA manipulations were
done as described in the literature.25) On the basis of
partial amino acid sequences and alignment with
known ORFs, oligonucleotides for PCRs primers
were designed (Fig. 2, Table 1). Nested PCR (94
for 5 min, and 94 C for 30 sec, t C for 30 sec, 72
for 1 min, in a total of 35 cycles, and then 72 C for
9C
9
C
9
9
9
7 min) was done using these primers and B. circulans
choromosome as a template for the ˆrst PCR, and
the second PCR was done with the ˆrst PCR product
as a template. The second PCR products were sub-
cloned into pT7blue-T vector (Novagen) to obtain
pPCR1 to 8. The conditions for nested PCR were
Fig. 1. The Reaction of DOI Synthase.
Fig. 2. N-Terminal Amino Acid Sequence of DOI Synthase 20-kDa Subunit and Comparison with Related Proteins.
Arrows: designed PCR primers

1540
H. TAMEGAI et al.
summarized in Table 2. The sequences of pPCR1 to 8
were analyzed to design appropriate oligonucleotides
for further PCR. The third PCRs were done by using
genome libraries of B. circulans as templates, which
had previously been constructed by digestion of the
chromosome with EcoR I, Hind III, Kpn I, Pst I, Sac
I, and Xho I, respectively, and were then subcloned
into pBluescriptII SK+ (Stratagene). With one of the
designed primers (23k10, 23k11 and 23k20) and T3
primer (or T7 primer) (Tabel 1), PCR was further
Gene disruption. A DNA fragment (1.8 kb) con-
taining the target gene was ampliˆed by PCR (95
for 5 min, and 95 C for 1 min, 57 C for 1 min, 72
for 2 min, in total 30 cycles, and then 72 C for
7 min), using the oligonucleotides of 20kDIS-f and
20kDIS-r (Table 1) as primers. The PCR product was
5?-phosphorylated by T4 polynucleotide kinase, and
9
9
C
C
9
9
9
then subcloned into Sma I-digested pUC119. After
the sequence was conˆrmed, a BamH I-EcoR I
fragment was recovered and subcloned into a corre-
sponding site of pHB201 (Tanaka et al., unpub-
lished). The resulting plasmid was digested with Nae
I (328 bp from the starting codon of btrC2), and a
tetracycline-resistance gene cassette (1.9 kb, reco-
vered from pBEST30926) with Sma I digestion) was
inserted in the reverse direction to yield pHB20Tckd.
The direction of the tetracycline-resistance gene
cassette was conˆrmed by an EcoR I digestion using
EcoR I site located at the end of the gene cassette.
The resulting plasmid was introduced into wild type
of B. circulans, and a gene disruptant was construct-
ed by homologous recombination, according to the
method described previously.22) Disruption of the
done (94
30 sec, 72
then 72 C for 7 min). The PCR product was
9C for 5 min, and 94
9C for 30 sec, 569C for
9C for 2 min, in a total of 35 cycles, and
9
subcloned into pT7blue-T vector to obtain pPstC1,
pSacC1, pEcoC1, and pXhoC1, and sequenced.
From the resulting sequence, appropriate primers for
the fourth PCR were designed to amplify the 5?-ter-
minal region of the gene (Tabel 1, 23k21 and 23k22).
PCR was done similarly to the third PCR, and the
product was further subcloned into pT7blue-T vector
to obtain pSacN1 and pSacN2. Based on the se-
quence of the resulting plasmids, the full region of
the target gene was analyzed. In all cases, at least 3
individual clones were analyzed to avoid incorpora-
tion of a mismatch in the PCR.
target gene was conˆrmed by PCR (95
and 95 C for 1 min, 60 C for 1 min, 72
in a total of 30 cycles, and then 72 C for 7 min) using
9
C for 5 min,
9
9
9
C for 3 min,
9
the chromosome of the disruptant as a template, and
oligonucleotides of 20kDISc-f and 20kDISc-r
(Table 1) as primers.
Table 1. Designed PCR Primers
20kf1
ATGAARGTNGGNGTNCTNGC
20kf2
20f3
20f4
20f6
20r2
20r3
ATGAARGTNGGNGTNTTRGC
GTNYTNGCNYTNCARGGNGCNGTNGC
CARGGNGCNGTNGCNGARCAYAT
CAYATHMGNYTNATHGARGCNGTNGGNGG
TTYTCNACNACNCCNCCNCCNACNGC
CCNGGDATDATNARNGCRTA
Analysis of antibiotic activity. Wild type and dis-
ruptant of B. circulans were cultured in a butirosin-
producing medium described previously16) for 2 days.
×
g,
Each culture was then centrifuged (10,000
15 min), and the supernatant was tested for antibiotic
20r4
20r5
20Kr1
23k10
23k11
23k20
23k21
23k22
T3 primer
T7 primer
20kDIS-f
20kDIS-r
20kDISc-f
20kDISc-r
20k-fSD
20k-r
TCNGTNARYTCNGGRTGRAA
GCRTANARYTGNGCNARYTGYTCNGC
GGDATDATNARNGCRTA
CGAGGCGGTTGGCGGAGAAGGCG
GGAGAAGGCGTCGTTGTAAAGCG
CTCCGCAACCGCCTCGATCAGG
AATGCCCTTAACAGGCAGATCGG
CACGATTTGCCCGTTGTACTCGC
AATTAACCCTCACTAAAGGG
GTAATACGACTCACTATAGGG
GGATCCGACTGTTCTGGAAG
GAATTCCGGAACGTTCCCAT
CCGCAGGCGGTGTAGCGACT
CAGTGAGCTTTGGCCTCGAA
activity by a paper disk diŠusion method.27)
Simultaneous expression of btrC2 with btrC in
E. coli. To obtain btrC2 (including the SD sequence)
for heterologous expression, primers (20k-fSD and
20k-r, Table 1) were designed, and PCR (95
5 min, and 95 C for 1 min, 60 C for 1 min, 72
1 min in total 30 cycles, and then 72 C for 7 min) was
9
C for
9
9
9C for
9
done using the B. circulans chromosome as template.
PCR products were once subcloned into the pT7
blue-T vector, and digested with Xho I. These frag-
ments were further subcloned into the Xho I site of
pDS4,21) to yield pDS2002. E. coli BL21(DE3) car-
CCGCGGCTCGAGAGATCGGCCGAGGCAG
TGGAGCACTCGAGGTGCTTCTGCTT
Table 2. Primers and Annealing Tempareture (t
9
C) for Nested PCR
PCR lot
ˆrst PCR second PCR
obtained plasmid
×
×
PCR47
PCR49
PCR115
PCR122
PCR123
20kf2 20r3 (50
9
9
9
9
C)
C)
C)
C)
C)
20f4 20r2 (58
9
9
9
9
C)
C)
C)
C)
C)
pPCR7
pPCR8
pPCR1,2,3
pPCR4
pPCR5,6
×
×
20kf2 20r4 (50
20f4 20r2 (58
×
×
20kf1 20r3 (50
20f6 20r5 (53
×
×
20k‰ 20krl (50
20f4 20r5 (53
×
×
20kf1 20r3 (50
9
20f4 20r5 (53
9

20-kDa Subunit of 2-Deoxy-scyllo-inosose Synthase
1541
rying pDS2002 was cultivated in 4.8 l of LB medium
containing 30 g ml of kanamycin until OD600
reached to 0.6, and then isopropyl-
topyranoside was added to a ˆnal concentration of
1 m . After 2 hours of additional cultivation, cells
were harvested by centrifugation (10,000
and kept at „30 C until use.
and appropriate bands were extracted with 100
ml of
m
CH3CN. Two l of each solution was analyzed by
m
W
b
-
D
-thiogalac-
HPLC with a Senshu Pak ODS 1251 N column
×
(4.6 mm i.d. 25 cm, Sensyu Scientiˆc, Japan) and a
M
‰uorescent detector (excitation; 331 nm, emission;
523 nm). The elution was done with a linear gradient
×
g
, 10 min)
9
of 20–50
z
CH3CN in 0.2
M
sodium acetate with a
‰ow rate of 1 ml min.
W
Puriˆcation of the expressed protein. The frozen
cells (10 g) were thawed, homogenized in 50 m Tris-
M
Miscellaneous. DNA sequencing was done with a
LONG READIR 4200 (Li-Cor) according to the
manufacturer's protocol. PCR was performed by a
GeneAmp PCR System 9700 (Amersham Pharmacia)
using ExTaq DNA polymerase (Takara, Japan).
Oligo DNAs for PCR primer were synthesized in
Amersham Pharmacia. Puriˆcation of plasmids was
done with a GFX Micro Plasmid Prep Kit (Amer-
sham Pharmacia). Restriction enzymes and modiˆca-
tion enzymes were purchased from Takara (Japan).
Genome preparations were done with Dr. GENTLE
(Takara, Japan). Genetic analysis was done with
GENETYX-MAC ver. 10 (Software Development,
Japan), and database research was done by FASTA
and BLAST2 on the internet. All other reagents were
of the highest grade commercially available. Poly-
acrylamide gel electrophoresis in the absence of SDS
was done by the method of Davis,28) and poly-
acrylamide gel electrophoresis in the presence of SDS
was done by the method of Laemmli.29)
HCl buŠer (pH 7.7), and disrupted by a sonic oscilla-
tor (Soniˆer Type-250, Branson) for a total period of
20 min on an ice bath. Unbroken cells were removed
×
g, 10 min) to obtain a
by centrifugation (10,000
cell-free extract. The extract thus obtained was then
fractionated by (NH4)2SO4 precipitation. The
precipitate obtained by 40
tion was dissolved in 40 ml of 50 m
(pH 7.7), and the resulting solution was dialyzed
against the same buŠer at 4 C for 15 h. The dialysate
z–70
z
(NH4)2SO4 satura-
M
Tris-HCl buŠer
9
×
was then put on a column (2.5 12 cm) of DEAE-
Sepharose FF equilibrated with the same buŠer, and
absorbed proteins were eluted with a linear gradient
of NaCl (0–0.4
were combined, concentrated by ultraˆltration, and
then put on a column of Hi Load 26 60 Superdex
M
). Enzymatically active fractions
W
200 pg equilibrated with 50 m
M
Tris-HCl buŠer (pH
7.7) containing 0.1
M
of NaCl. Enzymatically active
fractions were used as enzyme preparations.
Enzyme assay. DOI synthase activity was meas-
ured as described previously.20) Transamination ac-
tivity was examined as follows. Reaction mixture A
Accession Number. The sequences analyzed in this
study have been deposited in the DDBJ EMBL Gen-
W
W
Bank nucleotide sequence databases with the acces-
sion number of AB066209.
was composed of 0.5 m
25 m of L-glutamine, 5 m
glucose-6-phosphate, 5 m of CoCl2, and 2.3 mg of
puriˆed BtrC2+BtrC in 50 m BES buŠer, pH 7.7.
Reaction mixture B contained 25 m of L-glutamine,
5 m of NADH, 10 m of glucose-
of NAD+, 5 m
6-phosphate, 5 m of CoCl2, and 2.3 mg of puriˆed
BtrC2+BtrC in 50 m BES buŠer, pH 7.7. Reaction
mixture C contained 25 m of L-glutamine, 5 m
of NAD+, 5 m
of NADPH, 10 m of glucose-6-
phosphate, 5 m of CoCl2, and 2.3 mg of puriˆed
BtrC2+BtrC in 50 m BES buŠer, pH 7.7. The total
volume of each mixture was 4 ml. Each reaction mix-
ture was incubated separately at 42 C for 3 h, and
M
of pyridoxal-5-phosphate,
M
M
of NAD+, 10 m
M
of
M
Results
M
M
Cloning of DOI synthase 20-kDa subunit gene
The N-terminal amino acid sequence of the DOI
synthase 20-kDa subunit was analyzed as Fig. 2. It
showed high similarity to the B. subtilis yaaE gene,
and a homologue of this latter gene has been found in
a number of organisms. However, the functions of
all of these genes are still unknown. On the basis of a
partial amino acid sequence and a conserved region
with other ORFs, oligonucleotides were designed as
primers for PCR (Fig. 2, Table 1). By using these
primers, nested PCR was done with B. circulans
chromosome as template for the ˆrst PCR (Table 2).
To obtain the full region of the gene, PCR was
done further, as described in Materials and Methods.
As a result, the locus of the DOI synthase 20-kDa
subunit gene (named as btrC2) was identiˆed (Fig. 3).
M
M
M
M
M
M
M
M
M
M
M
9
the enzyme was removed from the mixtures by
ultraˆltration. Six ml of dansyl chloride solution
(10 mg ml in CH3CN), 0.5 ml of 2 N NaOH, and
W
2 ml of H2O were added to the ˆltered reaction mix-
tures, and the mixtures were incubated separately at
9
35 C for 3 h. The solvent was removed by ‰ushing
with air. Each residue was dissolved in 50
m
l of
The location of the ORF was as same as yaaE in B.
31)
methanol and centrifuged (20,000
g
, 10 min) to
subtilis30) or B. halodurans
.
BtrC2 showed com-
×
remove debris. The supernatant thus obtained was
separated by preparative TLC with a Merck Kieselgel
60 F254 plates Art 1.05744 (CHCl3:methanol, 5:1),
paratively high similarity to YaaE of B. subtilis and
its homologue. To be noted was that BtrC2 was simi-
lar to HisH of B. subtilis, amidotransferase subunit

1542
H. TAMEGAI et al.
Fig. 3. Genetic Map of DOI Synthase 20-kDa Subunit Gene
btrC2) Locus and Comparison with B. subtilis
(
.
Fig. 4. Deduced Amino Acid Sequence of BtrC2 and Compari-
son with Related Proteins.
Underlines: glutamine amidotransfer domain37)
of imidazole glycerol phosphate synthase32) to some
extent (Fig. 4).
Fig. 5. Growth Proˆles of B. circulans SANK72073 and
DbtrC2
in a Glucose Minimal Medium (A), and in a Butirosin-producing
Medium (B).
Disruption of btrC2 in B. circulans
In order to investigate the physiological function
of btrC2, a disruptant of btrC2 was constructed by
insertion of a tetracycline-resistance gene cassette.
Disruption of btrC2 was conˆrmed by PCR (wild
type; 1.4 kb, disruptant; 3.3 kb) and subsequent
sequence analysis of the PCR product (data not
shown). The disruptant showed more retarded
growth than the wild type in a minimal medium
(Fig. 5A), and in a butirosin-producing medium


#
;
B. circulans SANK72703,
disruptant supplemented with 0.1
ruptant supplemented with 0.5 yeast extract,
ruptant supplemented with 10
;
btrC2 disruptant,
; btrC2
$
M
NH4Cl,
;

btrC2 dis-
; btrC2 dis-
z
m
g ml pyridoxine-HCl
W
gle band was excised and directly analyzed by poly-
acrylamide gel electrophoresis in the presence of SDS
(Fig. 6B). Apparently, two bands were observed,
which corresponded to BtrC2 and BtrC. It should be
noted that recombinant BtrC2+BtrC turned out to
be stable in the (NH4)2SO4 fractionation or dialyza-
tion, which was in contrast to the observation that
recombinant BtrC was easily aggregated by these
(Fig. 5B). Growth was restored by adding 0.5
extract (in a butirosin-producing medium) or 0.1
z
yeast
M
NH4Cl (in each medium). Addition of pyridoxine had
no eŠect on growth. The btrC2 disruptant showed no
antibiotic production even in the butirosin-producing
medium. Interestingly however, antibiotic activity
was recovered when the culture was done in a
butirosin-producing medium containing additional
treatments.21) In addition, although recombinant
21)
btrC lost its activity in a buŠer without Co2+
,
recombinant BtrC2+BtrC was stable under such
conditions. Kinetic parameters of recombinant
BtrC2+BtrC for the DOI synthase activity were
measured, and compared with those of recombinant
BtrC homodimer and native DOI synthase from B.
circulans (Table 3). Although various combinations
of reaction conditions has been attempted, recom-
binant BtrC2+BtrC was unable to catalyze any reac-
tions beyond the formation of DOI. Neither an ami-
nation product (2-deoxy-scyllo-inosamine) nor a new
product was obtained.
0.5
z
yeast extract.
Heterologous expression of btrC2 with btrC
To investigate the role of BtrC2 in butirosin
biosynthesis, btrC2 was heterologously expressed
simultaneously with btrC in E. coli, and the resulting
BtrC2 was puriˆed to homogeneity. Interestingly,
BtrC2 and BtrC co-migrated into same fractions dur-
ing all the steps of puriˆcation. The puriˆed enzyme
appeared as a single band in polyacrylamide gel elec-
trophoresis in the absence of SDS (Fig. 6A). This sin-

20-kDa Subunit of 2-Deoxy-scyllo-inosose Synthase
1543
Table 3. Kinetic Parameters of Each Enzyme
DOI
recombinant
BtrC2+BtrC
recombinant
BtrC
synthase
from
B. circulans
for glucose-6-phosphate
Km (M)
1.2
4.4
×
×
10„3
10„1
2.1
2.3
×
10„4
1.0
9.0
7.3
×
×
10„4
10„2
kcat (s„1
)
for NAD+
Km (M)
4.2
×
10„4
×
10„4
1.7
×
10„5
role of amidotransferase in the primary metabolism.
An ORF2 homologue was reported to participate
in pyridoxine biosynthesis.34) More recently, it has
been shown that the nitrogen atom of pyridoxine
originates from the amide nitrogen of glutamine in B.
35)
subtilis
.
Consequently, it seems that BtrC2 is in-
volved in the pyridoxine biosynthesis in B. circulans
,
Fig. 6. Polyacrylamide Gel Electrophoreses in the Absence of
SDS (A) and in the Presence of SDS (B).
probably as an amidotransferase. However, as
described above, supplementation of pyridoxine to
the culture of btrC2 disruptant could not cure its
growth defect. This may suggest that BtrC2 plays a
pleiotropic role in the pyridoxine biosynthetic system
and other primary metabolisms. In some bacteria, it
For (B), M; size marker, 1; recombinant BtrC2+BtrC
(excised from (A)), 2; DOI synthase from B. circulans
.
Discussion
In this study, we succeeded in identifying the gene
was already known that
p
-aminobenzoate synthase
common
＝
of DOI synthase 20-kDa subunit (ORF3 btrC2 in
and anthranilate synthase share
a
Fig. 3). The location of btrC2 appeared to be unrelat-
ed to the butrosin biosynthetic gene cluster having
btrC in the middle, which is the key catalytic subunit
of DOI synthase in the DOS biosynthesis. It should
be pointed out that sequences similar to btrC2 were
found in a number of organisms, and in the most
cases, gene similar to ORF2 was accompanied adja-
cently. Further, the other neighboring ORF1 and
amidotransferase subunit, which provides the capaci-
ty for amido-transfer to each enzyme.36)
Based on these results and previous works by
others, we ˆrst assumed a functional sharing between
BtrC and BtrC2, that BtrC catalyzes the conversion
of glucose-6-phosphate to DOI as already elucidated,
and BtrC2 could be an amidotransferase subunit, to
yield 2-deoxy-scyllo-inosamine from DOI. So far
as we attempted, however, the recombinant
BtrC2+BtrC was unable to show such an activity un-
der various conditions. In addition, we identiˆed the
ORF4 found in the locus of btrC2 in B. circulans is
31)
also shown in B. subtilis30) and B. halodurans
.
These facts suggested that this gene array seems to be
distributed widely, at least in Bacilli, and hence btrC2
may participate not only in butirosin biosynthesis,
but also in the primary metabolism.
L
-glutamine:2-deoxy-scyllo-inosose aminotransferase
gene btrS from B. circulans quite recently.
(
)
(Tamegai et al., to be published) These facts appear
to suggest that the original assumption for the func-
tion of BtrC2+BtrC heterodimer described above is
not the case. As to the function of BtrC2, it should be
emphasized that, btrC2 disruptant lost its capacity
for antibiotic production, and feeding of yeast ex-
tract to the culture was able to restore antibiotic
production of the disruptant. It seems likely, there-
fore, that BtrC2 may participate in the butirosin
biosynthesis, although in a rather indirect way. The
results of simultaneous expression of btrC2 with btrC
showed that BtrC2 and BtrC spontaneously form a
heterodimeric structure, and the heterodimer was
much more stable than the BtrC homodimer.
Further, kinetic parameters of recombinant
BtrC2+BtrC were more similar to those of DOI syn-
thase from B. circulans than to those of recombinant
BtrC. In conclusion, the major role of BtrC2 in
butirosin biosynthesis appears to be stabilization and
An important observation was that BtrC2 showed
signiˆcant similarity to HisH, the glutamine amido-
transferase subunit of imidazole glycerol phosphate
synthase. In the histidine biosynthetic pathway, the
HisFH complex catalyzes the reaction of N1-[(5
?-
phosphoribulosyl)-formimino]-5-aminoimidazole-4-
carboxamide ribonucleotide to imidazole glycerol
phosphate and 5-aminoimidazole-4-carboxamido-1-
b- -ribofuranosyl 5?-monophosphate in the presence
D
of glutamine, and further, HisF alone can catalyze
the same reaction in the presence of NH+4 .33) Appar-
ently, this fact suggests that the function of the
amidotransferase subunit HisH can be complement-
ed by addition of NH+4 . The results in this study that
btrC2 disruptant was defective in growth both in the
minimal medium and in the butirosin-producing
medium, and the growth was complemented by the
presence of NH+4 clearly showed that BtrC2 plays a

1544
H. TAMEGAI et al.
Uehara, Y., Ikeda, D., and Umezawa, H., Biosynthe-
regulation of DOI synthase.
sis of 2-deoxystreptamine. J. Antibiotics
,
38,
1211–1218 (1985).
Acknowledgments
13) Lucher, L. A., Chen, Y.-M., and Walker, J. B.,
Reactions catalyzed by puriˆed
L
-glutamine:keto-
This work was supported by a Grant-in-Aid for
Scientiˆc Research (No. 11356004 and 13760084)
from the Ministry of Education, Science, Sports and
Culture of Japan and a grant from the Agricultural
Chemical Research Foundation.
scyllo-inositol aminotransferase, an enzyme required
for biosynthesis of aminocyclitol antibiotics. An-
timicrob. Agents Chemother., 33, 452–459 (1989).
14) Yamauchi, N. and Kakinuma, K., Conˆrmation of in
vitro synthesis of 2-deoxy-scyllo-inosose, the earliest
intermediate in the biosynthesis of 2-deoxystrepta-
mine, using cell free preparations of Streptomyces
fradiae. J. Antibiotics, 45, 774–780 (1992).
References
1) Cho, J. and Rando, R. R., Speciˆcity in the binding
15) Yamauchi, N. and Kakinuma, K., Enzymatic
carbocycle formation in microbial secondary
of aminoglycosides to HIV-RRE RNA. Biochemistry
38, 8548–8554 (1999).
,
metabolism. The mechanism of the 2-deoxy-scyllo
-
2) Litovchick, A., Evdokimov, A. G., and Lapidot, A.,
Aminoglycoside-arginine conjugates that bind TAR
RNA: synthesis, characterization, and antiviral activ-
ity. Biochemistry, 39, 2838–2852 (2000).
3) Distler, J., Mansouri, K., Mayer, G., Stockmann,
M., and Piepersberg, W., Streptomycin biosynthesis
and its regulation in Streptomycetes. Gene, 115,
105–111 (1992).
inosone synthase reaction as a crucial step in the 2-
deoxystreptamine biosynthesis in Streptomyces fra-
diae. J. Org. Chem., 60, 5614–5619 (1995).
16) Kudo, F., Yamauchi, N., Suzuki, R., and Kakinuma,
K., Kinetic isotope eŠect and reaction mechanism of
2-deoxy-scyllo-inosose synthase derived from
butirosin-producing Bacillus circulans. J. Antibiotics
50, 424–428 (1997).
,
4) Dairi, T., Ohta, T., Hashimoto, E., and Hasegawa,
M., Self cloning in Micromonospora olivasterospora
of fms genes for fortimicin A (astromicin) biosynthe-
sis. Mol. Gen. Genet., 232, 262–270 (1992).
17) Iwase, N., Kudo, F., Yamauchi, N., and Kakinuma,
K., Substrate speciˆcity of 2-deoxy-scyllo-inosose
synthase, the starter enzyme for 2-deoxystreptamine
biosynthesis, toward deoxyglucose-6-phosphates and
proposed mechanism. Biosci. Biotechnol. Biochem.,
62, 2396–2407 (1998).
5) Ahlert, J., Distler, J., Mansouri, K., and Pieper-
sberg, W., Identiˆcation of stsC, the gene encoding
the
L
-glutamine:scyllo-inosose aminotransferase
18) Kakinuma, K., Ogawa, Y., Sasaki, T., Seto, H., and
Otake, N., Stereochemistry of ribostamycin biosyn-
from streptomycin-producing streptomycetes. Arch.
Microbiol., 168, 102–113 (1997).
2
thesis. An application of H NMR spectroscopy. J.
6) Rosi, D., Goss, W. A., and Daum, S. J., Mutational
biosynthesis by idiotrophs of Micromonospora pur-
purea I. Conversion of aminocyclitols to new
aminoglycoside antibiotics. J. Antibiotics, 30, 88–97
(1977).
7) Daum, S. J., Rosi, D., and Goss, W. A., Mutational
biosynthesis by idiotrophs of Micromonospora
purpurea II. Conversion of non-amino containing
Am. Chem. Soc., 103, 5614–5616 (1981).
19) Kakinuma, K., Ogawa, Y., Sasaki, T., Seto, H., and
Otake, N., Mechanism and stereochemistry of the
biosynthesis of 2-deoxystreptamine and neosamine C.
J. Antibiotics, 42, 926–933 (1989).
20) Kudo, F., Hosomi, Y., Tamegai, H., and Kakinuma,
K., Puriˆcation and characterization of 2-deoxy-scyl-
lo-inosose synthase derived from Bacillus circulans. A
crucial carbocyclization enzyme in the biosynthesis of
2-deoxystreptamine-containing aminoglycoside an-
tibiotics. J. Antibiotics, 52, 81–88 (1999).
21) Kudo, F., Tamegai, H., Fujiwara, T., Tagami, U.,
Hirayama, K., and Kakinuma, K., Molecular cloning
of the gene for the key carbocycle-forming enzyme in
the biosynthesis of 2-deoxystreptamine-containing
aminocyclitol antibiotics and its comparison with de-
hydroquinate synthase. J. Antibiotics, 52, 559–571
(1999).
22) Ota, Y., Tamegai, H., Kudo, F., Kuriki, H., Koike-
Takeshita, A., Eguchi, T., and Kakinuma, K.,
Butirosin-biosynthetic gene cluster from Bacillus cir-
culans. J. Antibiotics, 53, 1158–1167 (2000).
cyclitols to aminoglycoside antibiotics. J. Antibiotics
30, 98–105 (1977).
,
8) Furumai, T., Takeda, K., Kinumaki, A., Ito, Y., and
Okuda, T., Biosynthesis of Butirosins. II Biosynthet-
ic pathway of butirosins elucidated from cosynthesis
and feeding experiments. J. Antibiotics, 32, 891–899
(1979).
9) Fujiwara, T., Takahashi, Y., Matsumoto, K., and
Kondo, E., Isolation of an intermediate of 2-deoxys-
treptamine biosynthesis from a mutant of Bacillus cir-
culans. J. Antibiotics, 33, 824–829 (1980).
10) Kase, H., Iida, T., Odakura, Y., Shirahata, K., and
Nakayama, K., Accumulation of 2-deoxy-inosamine
by
a 2-deoxystreptamine-requiring idiotroph of
Micromonospora sagamiensis. J. Antibiotics, 33,
1210–1212 (1980).
23) Yoshida, K., Ishio, I., Nagakawa, E., Yamamoto,
Y., Yamamoto, M., and Fujita, Y., Systematic study
of gene expression and transcription organization in
the gntZ-ywaA region of the Bacillus subtilis genome.
Microbiology, 146, 573–579 (2000).
11) Aubert-Pivert, E. and Davies, J., Biosynthesis of
butirosin in Bacillus circulans NRRL B3312: identiˆ-
cation by sequence analysis and insertional mutagene-
sis of the butB gene involved in antibiotic production.
Gene, 147, 1–11 (1994).
24) Schaä gger, H. and von Jagow, G., Tricine-sodium
dodecyl sulfate-polyacrylamide gel electrophoresis
for the separation of proteins in the range from 1 to
12) Suzukake, K., Tokunaga, K., Hayashi, H., Hori, M.,

20-kDa Subunit of 2-Deoxy-scyllo-inosose Synthase
1545
100 kDa. Anal. Biochem., 166, 368–379 (1987).
25) Sambrook, J., Fritsch, E. F., and Maniatis, T.,
Molecular cloning: A laboratory manual. 2nd ed.
Cold Spring Harbor Laboratory, New York (1989).
26) Itaya, M., Construction of a novel tetracycline
resistance gene cassette useful as a marker on the
Bacillus subtilis chromosome. Biosci. Biotechnol.
Biochem., 56, 685–686 (1992).
27) Fujiwara, T. and Kondo, E., Biosynthetic pathway of
2-deoxystreptamine. J. Antibiotics, 34, 13–15 (1981).
28) Davis, B. J., Disk electrophoresis-II method and
application to human serum proteins. Ann. N. Y.
Acad. Sci., 121, 404–427 (1964).
32) Henner, D., J., Band, L., Flaggs, G., and Chen, E.,
The organization and nucleotide sequence of the
Bacillus subtilis hisH, tyrA and aroE genes. Gene, 49,
147–152 (1986).
33) Klem, T. J. and Davisson, V. J., Imidazole glycerol
phosphate synthase: the glutamine amidotransferase
in histidine biosynthesis. Biochemistry, 32, 5177–5186
(1993).
34) Ehrenshaft, M., Bilski, P., Li, M. Y., Chignell, C.
F., and Daub, M. E., A highly conserved sequence is
a novel gene involved in de novo vitamin B6 biosyn-
thesis. Proc. Natl. Acad. Sci. USA, 96, 9374–9378
(1999).
29) Laemmli, U. K., Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature, 227, 680–685 (1970).
30) Ogasawara, N., Nakai, S., and Yoshikawa, H., Sys-
tematic sequencing of the 180 kilobase region of the
Bacillus subtilis chromosome containing the replica-
tion origin. DNA Res., 1, 1–14 (1994).
31) Takami, H., Takaki, Y., Nakasone, Y., Sakiyama,
T., Maeno, G., Sasaki, R., Hirama, C., Fuji, F., and
Masui, N., Genetic analysis of the chromosome of
alkaliphilic Bacillus halodurans C-125. Extremo-
philes, 3, 227–233 (1999).
35) Tanaka, K., Tazuya, K., Yamada, K., and Kumaoka,
H., Biosynthesis of pyridoxine: origin of the nitrogen
atom of pyridoxine in microorganisms. J. Nutr. Sci.
Vitaminol., 46, 55–57 (2000).
36) Kaplan, J. B., Merkel, W. K., and Nichols, B. P.,
Evolution of glutamine amidotransferase genes:
nucleotide sequences of the pabA genes from
Salmonella typhimurium, Klebsiella aerogenes and
Serratia marcescens. J. Mol. Biol., 183, 327–340
(1985).
37) Kane, J. F., Regulation of a common amidotran-
sferase subunit. J. Bacteriol., 132, 419–425 (1977).