ABSTRACT
The
rpoA
gene, encoding the
[alpha]
subunit of RNA polymerase, was cloned from
Streptomyces coelicolor
A3(2). It is preceded by
rpsK
and followed by
rplQ
, encoding ribosomal proteins S11 and L17, respectively, similar to the gene order in
Bacillus subtilis
. The
rpoA
gene specifies a protein of 339 amino acids with deduced molecular mass of 36
510 Da, exhibiting 64.3 and 70.7% similarity over its entire length to
Escherichia coli
and
B.subtilis
[alpha]
subunits, respectively. Using T7 expression system, we overexpressed the
S.coelicolor
[alpha]
protein in
E.coli.
A small fraction of this protein was found to be assembled into
E.coli
RNA polymerase. Antibody against
S.coelicolor
[alpha]
protein crossreacted with that of
B.subtilis
more than with
the
E.coli
[alpha]
subunit. The ability of recombinant
[alpha]
protein to assemble
[beta]
and
[beta]'
subunits into core enzyme
in vitro
was examined by measuring the core enzyme activity. Maximal reconstitution was obtained at
[alpha]
2:
[beta]
+
[beta]'
ratio of 1:2.3, indicating that the recombinant
[alpha]
protein is fully functional for subunit assembly. Similar results were also
obtained for natural
[alpha]
protein. Limited proteolysis with endoproteinase Glu-C revealed that
S.coelicolor
[alpha]
contains a tightly folded N-terminal domain and the C-terminal region is more protease-sensitive than that of
E.coli
[alpha]
.
DNA dependent RNA polymerases of prokaryotes are multisubunit enzymes composed of core enzyme with a subunit composition of [alpha]
2
[beta][beta]' and [sigma] subunit specifying promoter recognition (
1
,
2
). The [alpha] subunit is the most characterized of the
Escherichia coli
RNA polymerase subunits, serving several functions. Initially, [alpha] subunit has been known to play a role in the assembly of multisubunit RNA
polymerase complex, providing a scaffold for the assembly of [beta] and [beta]' in the following order; 2[alpha] -> [alpha]
2
-> [alpha]
2
[beta] -> [alpha]
2
[beta][beta]' (
3
). It contains determinants for protein-protein interaction for its dimerization, for [alpha]-[beta] interaction, and possibly, for [alpha]-[beta]' interaction. These
determinants reside in the N-terminal two-thirds of [alpha] protein (
4
-
7
). The C-terminal third is involved in the interaction with various transcriptional
activators (
8
-
10
), binding to specific DNA sequences (
11
,
12
), as well as its own dimerization (
12
). It has been demonstrated that the two regions constitute two separate domains
connected by a flexible linker (
12
,
13
). The structure of the C-terminal domain (CTD) has been determined by NMR and was shown to contain
a recognition loop and helices for both transcriptional activators and the DNA UP element, distinctly different from other DNA binding proteins (
14
). It has also been shown that the contact sites for both DNA and transcriptional
activators occupy the same surface on [alpha]CTD (
14
-
16
).
Gram-positive bacteria of the genus
Streptomyces
undergo a complex morphological and physiological differentiation (
17
). When the growth of the substrate mycelial colony on a solid surface nearly
ceases, usually triggered by nutrient limitation, the colonies develop aerial
mycelia, utilizing nutrients primarily provided from the hydrolysis of
substrate mycelia. The aerial mycelia further develop into chains of spores.
The biosynthesis of various antibiotics by Streptomycetes usually occurs concomitantly with the development of aerial mycelia and spores. These differentiation
processes involve a wide range of regulatory mechanisms including alternative [sigma] factors (
18
). Even though several lines of growing evidence indicate that the interaction
with various regulatory factors is central to this regulation, only a few
transcription factors were identified until now (
19
-
22
). A better understanding of
Streptomyces
gene regulation will require not only the analysis of
cis
- and
trans
-acting regulatory factors, but also the detailed characterization of
Streptomyces
RNA polymerase as a whole as well as its subunits. In this study we present our
work on cloning and characterization of the gene for the [alpha] subunit. The overproduced recombinant [alpha] protein was demonstrated to contain full activity for core enzyme
assembly, and a somewhat different structure from
E.coli
.
Escherichia coli
DH5[alpha] was used for all initial transformation and propagation of plasmids.
Strain BL21 (DE3) pLysS, a derivative of lambda lysogen carrying an IPTG
inducible gene for T7 RNA polymerase, was used to overexpress the [alpha] protein (
23
). The
Streptomyces coelicolor
strain used was M145, a prototrophic, SCP1
-
SCP2
-
derivative of the wild-type A3(2) strain. For RNA polymerase preparation,
S.coelicolor
A3(2) M145 was grown in 3.5 l YEME medium (
24
) containing 34% sucrose and 5 mM MgCl
2
under vigorous aeration in a fermenter for 23 h.
For cloning the
rpoA
gene, oligonucleotide primers [alpha]1 (5'-GT(G/C) CT(G/C) CAC GA(G/A) TAC AGC AC-3') and [alpha]2 (5'-CTT (G/C)AG GCA GTT GTA GCT
(G/C)CG-3') were designed as to contain mixed compositions of 50% each of G
and C or G and A at each degenerate codon position, and synthesized by Oligos,
Etc. (USA). The PCR was performed in the reaction mixture (100 [mu]l) containing 20 pmol each of [alpha]1 and [alpha]2 primers, four dNTPs at 0.2 mM each, 10 mM Tris-HCl (pH 9.0), 1.5 mM MgCl
2
, 50 mM KCl, 0.1% Triton X-100, 100 ng
S.coelicolor
genomic DNA and 2.5 U
Taq
DNA polymerase (Poscochem, Korea). The reaction mixture was subjected to 30
cycles of polymerization consisting of 1 min denaturation at 94oC, 1 min annealing at 37oC, and 1.5 min extension at 72oC. The 4 kb
Pst
I/
Bam
HI genomic DNA fragment hybridizing with the PCR product was cloned into pUC18
and further subcloned into pGEM-7zf(+). To construct [alpha]-expressing pET plasmid (pET-[alpha]), an oligonucleotide primer [alpha]
Nde
I (5'-ACTGAAGGATCCC-
DNA sequencing was carried out on double-stranded DNA fragments cloned into pGEM-7zf(+) by both manual and automated sequencing (ALFexpress, Pharmacia). The sequence was deposited to EMBL under accession number X92107. Sequence comparison with the database was done using BLAST program (
25
), and CLUSTAL V for multiple alignments (
26
).
Escherichia coli
BL21 (DE3) pLysS cells containing pET-[alpha] plasmid were grown to OD
600
of 0.6. Following induction by 0.4 mM isopropyl-[beta]-d-thiogalactopyranoside (IPTG), cells were harvested and stored at -70oC until use. The frozen cells were suspended
in the lysis buffer [10 mM Tris-HCl (pH 7.9 at 4oC), 1 mM EDTA, 0.1 M NaCl, 0.3 mg/ml of lysozyme and 1 mM
phenylmethylsulfonyl fluoride (PMSF)] for 20 min on ice and sodium deoxycholate
was added to a final concentration of 0.2% and further incubated for 20 min
before sonication. After removing cell debris by centrifugation, proteins were
precipitated with 60% saturated ammonium sulfate and resolubilized in and
dialyzed against buffer TGED [10 mM Tris-HCl (pH 7.9 at 4oC), 0.1 mM EDTA, 0.1 mM DTT, 5% glycerol]. The proteins were then
subjected to DEAE-Sepharose column chromatography, eluted with a linear gradient of 0-0.5 M NaCl. Fractions containing [alpha] protein were eluted at 0.3 M NaCl and were further
purified on the preparative polyacrylamide gel electrophoresis system (BioRad).
Purified [alpha] protein was electrophoresed on 10% SDS-PAGE. Gel slice containing [alpha] was crushed and resuspended in phosphate buffered saline (0.8% NaCl, 0.02% KCl, pH 7.2). A total of
250 [mu]g protein was used to immunize two mice for antibody production. The Western
blot analysis was performed as described by Blake
et al
. (
27
). Blotted filters were incubated overnight with polyclonal mouse antibody against [alpha] (1:1000 dilution) in TBS buffer containing 20 mM Tris-HCl (pH 7.5), 500 mM NaCl, 0.5% BSA and 0.05% Tween 20. Bound
antibody was detected by anti-mouse IgG goat antibodies conjugated with alkaline phosphatase.
RNA polymerase was prepared by a combined modification of two procedures
developed for the purification of holoenzymes from
E.coli
and
S.coelicolor
(
28
,
29
). In brief,
S.coelicolor
mycelial cells (40 g wet weight) were suspended in 400 ml lysis buffer [10 mM
Tris-HCl (pH 7.9 at 4oC), 2 mM EDTA, 1 mM PMSF, 1 mM 2-mercaptoethanol, 0.2 M NaCl, 1 mM DTT, 5% glycerol] containing
1 g lysozyme and were incubated at 4oC for 30 min. The cells were disrupted by grinding with aluminum oxide in a
commercial mixer. Crude extracts were rapidly mixed with Polymin P at a final
concentration of 0.35% (v/v) and subjected to salt extraction, ammonium
sulphate precipitation, DEAE-Sephacel, and heparin-Sepharose chromatographies. For further purification, the fractions containing intact [beta] and [beta]' subunits were pooled and applied via a 1 ml superloop
to a MonoQ HR 1/1 anion exchange column on FPLC (Pharmacia). The column was
washed with 4 ml TGED plus 0.2 M NaCl and then a 20 ml linear gradient of 0.2-0.6 M NaCl was applied at a flow rate of 0.5 ml/min. Fractions containing pure RNA polymerase were diluted with two times
concentrated storage buffer (TGED with 50% glycerol and 0.15 M NaCl) and stored
at -70oC.
Escherichia coli
BL21 cells containing pET-[alpha] plasmid were grown to OD
600
of 0.6 and treated with 0.01 mM IPTG for 6 h. Cells from 200 ml culture were
harvested and lysed in the same buffer used for large scale preparation of RNA
polymerase (
28
). Following Polymin P fractionation, the polymerase fraction was loaded on 1 ml
heparin-Sepharose column and proteins were eluted with step gradient of KCl from
0.3 to 0.8 M. The pooled fractions containing RNA polymerase were further fractionated on glycerol gradient centrifugation as described by Glass
et al
. (
30
).
The [alpha], [beta] and [beta]' subunits were separated from the core RNA polymerase as described by Lill
et al
. (
31
) with slight modifications. Core enzyme (3 mg) in 3 ml of storage buffer was diluted with 6 ml of deionized 9 M urea (final concentration 6 M) and adjusted to 10 mM DTT.
Following 15 min incubation at 30oC, the mixture was dialyzed for 3 h against buffer TG
20
ED [TGED with 20% (v/v) glycerol] containing 6 M urea. The dissociated enzyme
was applied to a phosphocellulose column (0.9 * 4 cm) equilibrated with the same buffer. The column was then washed with
TG
20
ED plus 6 M urea. Subunit [alpha] came out in the flow-through fraction whereas subunits [beta] and [beta]' were eluted at 0.2 and 0.7 M KCl, respectively.
The fractions containing each subunit were dialyzed overnight against the renaturation buffer [50 mM Tris-HCl (pH 7.9 at 4oC), 0.1 mM EDTA, 1 mM DTT, 10 mM MgCl
2
, 0.3 M KCl, 20% glycerol] to remove urea. They were concentrated by ammonium sulphate precipitation. The protein pellet was redissolved in TGED buffer and dialyzed overnight against the storage buffer and kept at -20oC.
To reconstitute core enzyme
in vitro
, either natural or recombinant [alpha] protein was mixed with [beta] and [beta]' subunits at various stoichiometric molar ratios in
reconstitution buffer containing 7-10 mM Tris-HCl (pH 7.9 at 25oC), 7 mM MgCl
2
, 25 mM MnCl
2
, 0.07-0.1 mM EDTA, 0.3 M KCl, 10 mM DTT and 40% glycerol. The final
concentration of total protein was adjusted between 0.2 and 0.5 mg/ml. The
mixture was incubated at 30oC for 30 min to 1 h. The RNA synthesizing activity of the resulting enzyme was determined as described by Burgess and Jendrisak (
32
).
Comparison of amino acid sequences of [alpha] proteins from
E.coli
,
Bacillus subtilis
and plant chloroplasts revealed several conserved regions (
33
-
37
). Based on this observation, oligonucleotide primers were designed from two
such conserved sequences to amplify the intervening region of
S.coelicolor
[alpha] gene. Degenerate oligonucleotide primers corresponding to
E.coli
residues 64-70 ([alpha]1) and 265-271 ([alpha]2) were synthesized (Fig.
2
). PCR reaction on chromosomal DNA from
S.coelicolor
A3(2) M145 produced a single species of product of 620 bp, the size of which is
as predicted from the sequence
of
rpoA
gene from
E.coli
. The PCR product was cloned and partially sequenced. The translated amino acid
sequence was 35 and 41% identical to the sequences of the corresponding regions
of
E.coli
(34 of 98 amino acid residues) and
B.subtilis
(40 of 98 amino acid residues) [alpha] subunit, respectively. This PCR product was used as a probe for the
Southern hybridization of
S.coelicolor
chromosomal DNA digested with
Kpn
I
, Pst
I,
Bam
HI or
Pst
I/
Bam
HI. A single distinct band was detected for each digest (data not shown). The hybridizing
Pst
I-
Bam
HI fragment of 4 kb was cloned into pUC18 and further subcloned into pGEM-7zf(+).
Nucleotide sequencing identified an open reading frame for the [alpha] subunit (
rpoA
),
preceded by a truncated sequence from ribosomal protein S11 (
rpsK
) with 132 bp of intercistronic region, and followed by another truncated
sequence from ribosomal protein L17 (
rplQ
)
with 191 bp of intercistronic region. The restriction map and the gene
organization of a 1.6 kb
Pst
I-
Bcl
I fragment were shown in Figure
1
. The C-terminal 30 residues encoded by
S.coelicolor
rpsK
exhibited high similarity to the C-terminal residues of
E.coli
ribosomal protein S11 (20 identical residues) (
33
). The N-terminal 71 residues encoded by
S.coelicolor
rplQ
also exhibited 54% similarity to the N-terminal residues of
E.coli
ribosomal protein L17 (
33
).
Using the T7 expression system, as described in Materials and Methods, we were able to obtain a large amount of
S.coelicolor
[alpha] protein in
E.coli
. Following induction with
0.4 mM IPTG,
S.coelicolor
[alpha] protein made up to 35% of the total protein in
E.coli
cell extracts. The recombinant [alpha] protein was purified to >95% homogeneity as shown in Figure
3
. The apparent molecular mass was determined to be ~46 kDa, substantially larger than the predicted one. We compared the
antigenicity among [alpha] proteins from
S.coelicolor
,
B.subtilis
, and
E.coli
by immunoblotting (Fig.
4
). The polyclonal antibodies raised against the recombinant
S.coelicolor
[alpha] protein cross-reacted with the natural [alpha] protein in purified RNA polymerase, demonstrating that the
recombinant [alpha] protein was antigenically indistinguishable from the natural one (Fig.
4
B, lanes 1 and 2). We found that the antibody against
S.coelicolor
[alpha] protein cross-reacted better with
B.subtilis
[alpha] than that of
E.coli
(Fig.
4
B, lanes 3 and 4), suggesting that the structural difference in [alpha] subunit between
S.coelicolor
and
E.coli
is rather substantial. The deduced pI value of
S.coelicolor
[alpha] from the amino acid composition is 4.36, which is most acidic when
compared with those of
E.coli
subunits; [beta]' (6.85), [beta] (5.30), [alpha] (4.65) and [sigma]
70
(4.40) (
39
). This acidic property of the
S.coelicolor
[alpha] protein was confirmed by two-dimensional gel electrophoresis performed in parallel with
E.coli
RNA polymerase under the denaturing condition (8 M urea) (data not shown).
The incorporation of the
S.coelicolor
[alpha] subunit into
E.coli
RNA polymerase holoenzyme
in vivo
was examined to verify its ability to assemble into RNA polymerase.
E.coli
BL21 cells containing pET-[alpha] were slightly induced for
S.coelicolor
[alpha] protein with 0.01 mM IPTG instead of 0.4 mM. Under this condition, most of the recombinant [alpha] protein exists in soluble fractions. RNA polymerase holoenzyme was purified from these cells through Polymin-P fractionation, ammonium sulfate precipitation, heparin- Sepharose chromatography and glycerol gradient
ultracentrifugation. Figure
5
A demonstrated a SDS-PAGE profile of purified RNA polymerase from
E.coli
BL21 (lane 2)
and
S.coelicolor
as a control (lane 1). In lane 2, a very faint band of 46 kDa was visible along
with thicker band of
E.coli
[alpha] which migrates faster. The presence of
S.coelicolor
[alpha] protein in
E.coli
RNA polymerase preparation was more clearly demonstrated by immunoblotting with
antibody against
S.coelicolor
[alpha] protein (Fig.
5
B, lane 2). The result indicates that the
S.coelicolor
[alpha] subunit is able to assemble with
E.coli
[beta] and [beta]' subunits to form hybrid core enzymes. We checked each
fraction of glycerol gradient for the presence of
S.coelicolor
[alpha] by immunoblotting and found out that the
S.coelicolor
[alpha] protein was present only in the bottom fractions where
E.coli
[beta] and [beta]' subunits cosedimented as well as in the top fractions
existing as free subunits (data not shown). This rules out the possibility that
S.coelicolor
[alpha] might have been contaminated in the
E.coli
RNA polymerase preparation.
In order to test further the assembly function of the recombinant [alpha] protein, we performed
in vitro
reconstitution of
S.coelicolor
RNA polymerase core enzyme from its subunits. [beta] and [beta]' subunits were dissociated from the purified RNA polymerase
by 6 M urea and isolated through phosphocellulose column chromatography as described in Materials and Methods. The separation of isolated subunits was checked by SDS-PAGE as shown in Figure
6
A. We assigned the higher salt-eluted subunit as [beta]' as in
E.coli
(
40
) and noted that [beta]' migrated slightly faster than [beta], unlike in
E.coli
. The isolated [beta] and [beta]' and the recombinant [alpha] proteins were then mixed in the absence of urea at
different molar ratios of [beta][beta]'/[alpha]
2
and measured for RNA polymerizing activity (Fig.
6
B). The active enzyme was recovered only when all the subunits were combined.
Urea was not required in the reconstitution process as observed by Lill
et al
. (
31
) for
E.coli
RNA polymerase. The optimal recovery was achieved when the subunits were mixed
at the [alpha]
2
:[beta][beta]' ratio of 1:2.3, instead of the stoichiometric 1:1 ratio. This indicates that the amount of `competent' [beta] and/or [beta]' molecules in these preparations was ~43% of the total [beta] and [beta]' proteins. More
excess amounts of [alpha] over [beta] and [beta]' subunits were found to be inhibitory to enzyme
assembly or activity. The maximum activity of reconstituted enzyme corresponded
to 40-60% of the activity of core enzyme treated in parallel with 6 M urea
without column separation of subunits, comparable to the efficiency with
E.coli
RNA polymerase (
41
). When the natural [alpha] subunit dissociated from the native core enzyme replaced the recombinant [alpha] protein, there was no difference in the efficiency of
reconstitution within experimental error. This indicates that the recombinant [alpha] protein is indeed capable of assembling functional RNA polymerase with [beta] and [beta]' subunits, just like natural [alpha] protein.
To obtain preliminary information on the subdomain structure of [alpha] subunit, the purified [alpha] protein was partially digested with endoproteinase Glu-C which cuts at Glu residues. The reaction products were run
on SDS-polyacrylamide gel and compared with
E.coli
[alpha] protein subjected to the same treatment in parallel (Fig.
7
).
Streptomyces coelicolor
[alpha] was cleaved into one major (33 kDa
app
) and two minor (~28 kDa
app
) fragments, while
E.coli
[alpha] was cleaved into two major fragments, 28 and 8 kDa
app
as observed by Blatter
et al
. (
12
). Previous studies on domain mapping of
E.coli
[alpha] have demonstrated that the protease resistant 28 kDa fragment is from
the highly structured N-terminal two-thirds of [alpha], and the smaller 8 kDa fragment is from the C-terminal region suggesting that [alpha] consists of two major domains, N-terminal domain (NTD) and C-terminal domain (CTD), linked by protease-sensitive flexible linker from
residue 235 to 248 (
12
,
13
). Both the 33 and 28 kDa fragments were derived from the N-terminal portion of
S.coelicolor
[alpha], since various lengths of deletions from the C-terminal end up to residue 250 barely changed the pattern of proteolysis (data not shown). We were not able to
observe any smaller fragments around 10 kDa for
S.coelicolor
[alpha] (Fig.
7
, lane 2).
We isolated the gene for [alpha] subunit protein of
S.coelicolor
RNA polymerase and demonstrated that its gene product expressed in
E.coli
can indeed be assembled into a functional RNA polymerase core enzyme. The
rpoA
gene was found to be preceded by
rpsK
(S11) and followed by
rplQ
(L17). This gene order is the same as in
B.subtilis
and
Chlamydia trachomatis.
In
E.coli
,
rpoA
lies in an operon with four ribosomal protein genes in the order,
rpsM
(S13),
rpsK
(S11),
rpsD
(S4),
rpoA
and
rplQ
(L17) (
33
). The organization of
B.subtilis rpoA
operon differs from
E.coli
in that it lacks
rpsD
gene preceding
rpoA
(
42
). In the genomes of both
B.subtilis
and
C.trachomatis
,
the
rpoA
gene is preceded by the
rpsK
gene (
43
). In this respect,
S.coelicolor
is more closely related with these groups than
E.coli.
The absence of
rpsD
is interesting because S4 has been shown to act as a translational repressor of
the [alpha] operon in
E.coli
, inhibiting the translation of the three ribosomal protein cistrons preceding
the [alpha] and L17 cistron distal to [alpha] (
44
,
45
). This autogenous regulation of the [alpha] operon by S4 results in differential expression of ribosomal proteins
and [alpha] in
E.coli
. Further investigation is needed as to whether the ribosomal proteins and [alpha] are differentially regulated in
S.coelicolor
,
and if so, what is the mechanism of action.
The sequence comparison of
S.coelicolor
[alpha] protein with three other known bacterial [alpha] proteins revealed that there are several conserved regions. Mutation studies on
E.coli
[alpha] have demonstrated that the residues in the N-terminal portion are involved in the assembly of the core RNA
polymerase (
5
-
7
). Point mutations within the C-terminal domain that affected transcriptional activation by regulatory
factors or binding to UP element identified individual residues of [alpha] as potential contact sites (
8
,
10
,
12
). Most of these residues are conserved in
S.coelicolor
[alpha] as well. One example is the binding site for cyclic AMP (cAMP) receptor
protein (CRP). The residues between 258 and 270, known to be involved in the
interaction with CRP at class I CRP-dependent promoters (
16
,
46
,
47
), are very well conserved. These are residues at 258(D), 259(D), 260(L),
261(E), 262(L), 265(R), 268(N) and 269(C) in
E.coli
, corresponding to
S.coelicolor
residues at 251(E), 252(E), 253(L), 254(E), 255(L), 258(R), 261(N) and 262(C),
respectively. The residues involved in the recognition of the UP element of
rrnB
P1 promoter in
E.coli
, 260(L), 262(L), 264(V), 265(R), 266(S), 268(N), 269(C), are all
correspondingly conserved in
S.coelicolor
(
14
-
16
). One of the two residues at 297(K) and 298(K) in
E.coli
, known to be involved in both UP element and CRP recognition, are conserved in
S.coelicolor
at 291. Whether these conservations have any meaning in
S.coelicolor
is not known, since so far neither a homolog of CRP nor CRP-dependent promoters have been reported for
Streptomyces
spp. Furthermore, intracellular concentration of cAMP is relatively constant and does not appear to be involved in catabolite
repression in
Streptomyces
spp. (
48
). The presence of UP element in
S.coelicolor
promoters has not been reported either. One of the two residues, Pro-322 and Pro-323 of
E.coli
[alpha], which is involved in OmpR activation (
49
) is also conserved at 316. Lys-271 involved in interactions with the activators CysB, AraC and MelR in
E.coli
is conserved in
S.coelicolor
at residue 264 (
50
). Although no homologs of these activators have been described to date in
Streptomyces
spp., the work in
E.coli
to define the activator contact sites on [alpha] can help assign a role on
S.coeliclor
[alpha] that might interact similarly with different array of transcriptional
activators.
S.coelicolor
[alpha] subunit protein contains an extra C-terminal tail of 17 amino acids. This extended C-terminal tail is also observed in the Chlamydial [alpha] protein, with complete lack of homology between the
two (
43
). The prominent feature in the C-terminal tail of
S.coelicolor
[alpha] is the abundance of acidic residues (six out of 17).
Even though the peptide sequences are rather highly conserved between
S.coelicolor
and
E.coli
[alpha] proteins, the structural difference is manifested by low immuno-crossreactivity and the difference in susceptibility to
endoproteinase. The less compact structure in C-terminal domain of
S.coelicolor
[alpha] can be postulated, and the different interaction between
S.coelicolor
[alpha] and DNA or other transcriptional regulators can be inferred. Further genetic and biochemical studies are necessary in this respect.
We thank Dr A. Ishihama for helpful discussions and providing pGEMAX185 for the
overproduction of the
E.coli
[alpha] subunit; Dr D. Hopwood for providing
S.coelicolor
M145 A3(2) strain; Drs N. Fujita and J.-W. Suh for valuable discussions; Dr H.-D. Youn for technical assistance in protein purification; J.H. Lee
for automatic sequencing; other members of our lab for the sequence data
analysis and helpful discussions. This work was supported by a grant from the
Korea Science and Engineering Foundation for SRC (Research Center for Molecular
Microbiology) to J.-H. Roe.
*To whom correspondence should be addressed. Tel: +82 2 880 6706; Fax: +82 2 888
4911; Email: jhroe@alliant.snu.ac.kr
+
Present address: Department of Biological Chemistry and Molecular Pharmacology,
Harvard Medical School, Boston, MA 02115, USA
REFERENCES
Return