Download Plasmid pIP501 Encoded Transciptional Repressor CopR Binds to

Article No. mb982122
J. Mol. Biol. (1998) 283, 595±603
Plasmid pIP501 Encoded Transcriptional Repressor
CopR Binds to its Target DNA as a Dimer
Katrin Steinmetzer1*, Joachim Behlke2 and Sabine Brantl1
1
Institut fuÈr Molekularbiologie
Friedrich-Schiller-UniversitaÈt
Jena, Winzerlaer Str. 10
D-07745 Jena, Germany
2
Max-DelbruÈck-Zentrum fuÈr
Molekulare Medizin
Robert-RoÈssle-Straûe 10
D-13122 Berlin-Buch, Germany
The CopR protein is one of the two regulators of pIP501 copy number.
It acts as transcriptional repressor at the essential repR promoter pII.
Previously, we found that CopR contacts two consecutive major grooves
(site I and site II) on the same face of the DNA. In spite of identical
sequence motifs in these sites, neighboring bases were contacted differently. Furthermore, we showed that CopR can dimerize in solution. We
demonstrate by two independent methods that CopR binds the DNA as
a dimer. We present data that suggest that the sigmoidal CopR-DNA
binding curve published previously is the result of two coupled equilibria: dimerization of CopR monomers and CopR dimer-DNA binding.
A KD-value of 1.44(0.49) 10ÿ6 M for CopR dimers was determined by
analytical ultracentrifugation. Based on this value and the binding curve,
the equilibrium dissociation constant K2 for the CopR-DNA complex was
calculated to be 4(1.3) 10ÿ10 M. Quantitative Western blot analysis
was used to determine the intracellular concentration of CopR in Bacillus
subtilis. This value, 20 10ÿ6 to 30 10ÿ6 M, is 10 to 20-fold higher than
the equilibrium constant for dimer dissociation, suggesting that CopR
binds in vivo as a preformed dimer.
# 1998 Academic Press
*Corresponding author
Keywords: CopR; transcriptional repressor; binding stoichiometry; plasmid
replication control; DNA-protein-interaction
Introduction
Speci®c recognition of DNA sequences by DNAbinding proteins is a substantial part in the
processes of regulation of gene activity. Although
several prokaryotic transcriptional repressor proteins have been studied extensively with respect to
DNA-binding, e.g. lac-repressor, l repressor, 434
repressor and others (Harrison, 1991), little is
known about the class of repressor proteins that is
involved in copy number control of plasmids. Such
proteins have been identi®ed in many plasmids:
pLS1 (CopG, del Solar et al., 1990), R1 (CopB, Light
& Molin, 1982; Riise & Molin, 1986) and plasmids
of the inc18 family (Brantl et al., 1990): pIP501
(CopR, Brantl & Behnke, 1992), pAMb1 (CopF,
Swin®eld et al., 1990) and pSM19035 (CopS,
Ceglowski et al., 1993). Both CopG and CopB were
characterized as classical repressor proteins with
typical helix-turn-helix motifs and symmetric operator sites (13 bp in the case of CopG) overlapping
Abbreviations used: EMSA, electrophoretic mobility
shift assay.
E-mail address of the corresponding author:
[email protected]
0022±2836/98/430595±09 $30.00/0
the corresponding rep promoters. However, structural requirements of the protein-DNA binding
mechanisms have not yet been elucidated.
In the case of plasmid pIP501, originally isolated
from Streptococcus agalactiae (Horodniceanu et al.,
1976), the CopR protein is one of the two regulators of plasmid copy number (Figure 1). Whereas
the antisense RNA (RNAIII) induces premature
termination of repR mRNA transcription (Brantl
et al., 1993; Brantl & Wagner, 1994), the 10.4 kDa
CopR protein represses transcription from the
essential repR promoter pII about tenfold (Brantl,
1994). Furthermore, CopR binding prevents convergent transcription from pII from interfering
with antisense promoter pIII activity, thereby
indirectly increasing transcription of RNAIII
(Brantl & Wagner, 1996, 1997). Deletion of either
control element, RNAIII or CopR, leads to a 10 to
20-fold increase of pIP501 copy number, but simultaneous deletions do not show an additive effect.
The other two related Cop proteins, CopF and
CopS, exhibit more than 95% sequence identity
with CopR. Whereas the CopF operator was con®ned to a region of 31 bp (Le Chatelier et al., 1994),
the function and properties of CopS have not yet
been characterized in detail.
# 1998 Academic Press
596
Previously, we showed that CopR binds asymmetrically at two consecutive major grooves of the
DNA in its operator region (Steinmetzer & Brantl,
1997). The contacted bases and DNA-backbone
phosphate residues were determined by chemical
footprinting. Both binding sites share the sequence
motif 50 CGTG 30 , but neighboring bases were
found to be contacted differently, and half-site II
proved to be more extended than half-site I.
Here, we present data that CopR, which can
dimerize in solution, also binds the DNA as a
dimer. The sigmoidal binding curve (Steinmetzer
& Brantl, 1997) indicative of cooperative binding
was reconsidered. This curve is apparently in¯uenced by a CopR monomer-dimer equilibrium. The
equilibrium dissociation constant of the CopR
dimer was determined by analytical ultracentrifugation to be 1.44(0.49) 10ÿ6 M. This value was
used to calculate the equilibrium dissociation constant for the DNA-dimer complex from the binding
curve. Furthermore, the estimation of the intracellular concentration of CopR in Bacillus subtilis
allowed us to suggest that CopR binds to the DNA
in vivo also as a preformed dimer.
Results
Binding stoichiometry of the
CopR-DNA complex
To determine the binding stoichiometry of the
CopR-DNA complexes, we used two different
approaches: (i) we used a truncated CopR mutant
to create hetero-oligomers of the two proteins following an approach introduced by Hope & Struhl
(1987); and (ii) we performed EMSA (electrophor-
CopR Binding Stoichiometry
ectic mobility shift assays) at high CopR and DNA
concentrations.
First, we created hetero-oligomers of wild-type
His6-CopR and a truncated His6-CopR20 mutant
protein that lacks the 20 C-terminal amino acid
residues, but shows wild-type activity both in vivo
and in vitro. A mixture of varying amounts of
both puri®ed proteins was incubated for 30 minutes at room temperature and subsequently used
in EMSA. As shown in Figure 2, three complexes
of different electrophoretic mobility were formed.
The upper and the lower complex correspond to
the complexes obtained with His6-CopR or
His6-CopR20 alone, whereas a third one with
intermediate electrophoretic mobility represents a
complex with a His6-CopR-His6-CopR20 heterodimer. No higher-order complex was found.
These data indicate that CopR binds to the DNA
as a dimer.
Based on the assumption that CopR binds to
DNA as a dimer, titration experiments at a constant DNA concentration and at a constant CopR
concentration were performed. The results are
shown in Figure 3. The exact determination of
both protein and DNA concentration is crucial
for the correct interpretation of the results. The
protein concentration was determined on a Biotronik-analyser (acidic hydrolysis) and the DNA
concentration was calculated from absorbance
measurements of highly concentrated stocks. Both
data sets were averaged from two independent
experiments. The saturation of CopR with the
DNA target was reached at a molar ratio of
CopR monomer to DNA of 1.8. From the titration
experiment at constant DNA concentration, we
calculated a ratio of 2.25. The titration to satur-
Figure 1. Working model on copy number control of plasmid pIP501. Filled boxes, promoters; open rectangles,
open reading frames; thick arrows, RNAs; hatched/stippled, proteins; oriR, replication origin; ®lled arrows, activation/positive interactions; horizontal bar, repression. ATT indicates the position of induced termination of RNAII.
The negative regulation by CopR is transcriptional and is exerted at pII. Negative regulation by RNAIII is induced
by transcriptional attenuation. Repressed RNAII transcription in the presence of CopR permits increased RNAIII transcription as described by Brantl & Wagner (1997).
597
CopR Binding Stoichiometry
Figure 2. Heterodimer-DNA complex formation.
EMSA with varying concentrations of His6-CopR and
truncated His6-CopR20 protein. The positions of
unbound DNA, wild-type complex, His6-CopR20 complex and of the heterodimer complex are indicated. The
labeled DNA fragment KS1 was used.
KD represents the dissociation constant of the
CopR dimer, K2 is the respective constant for the
CopR-DNA complex. D, C and C2 are the free
DNA, free protein monomer and free protein
dimer, respectively. C2D symbolizes the complex
formed by one CopR dimer and one DNA
molecule.
ation experiments show that both, target DNA
and CopR protein, were about 90% active.
Determination of the molecular mass of CopR
by analytical ultracentrifugation
To study the molecular mass of CopR, different
CopR concentrations in the range of 0.036 to
0.214 g/l were analyzed by analytical ultracentrifugation using the sedimentation equilibrium technique. The radial concentration distributions
obtained at three different wavelengths were ®tted
simultaneously using the program POLYMOLE
(Behlke et al. 1997). The molecular mass values
differ with the initial CopR concentration as shown
in Figure 4. The data were analyzed using the
following equation according to a monomer-dimer
equilibrium:
KD ˆ 2a2 c0 =…1 ÿ a†
where a is the dissociation ratio and c0 is the total
concentration of loaded protein monomers. An
average KD of 1.44(0.49) 10ÿ6 M was calculated.
Equilibrium dissociation constant of the
CopR-DNA complex
The observed stoichiometry of the CopR-DNA
complex (Figures 2 and 3) suggests that our earlier
view of His6-CopR binding to target DNA
(Steinmetzer & Brantl, 1997) must be reconsidered.
There we interpreted the sigmoidal shape of the
binding curve as an indication of positive cooperative CopR binding to its target. If the protein exists
as a dimer and binds the DNA as a dimer, a
sigmoidal binding curve may be due to that CopRDNA binding is in¯uenced by dissociation of
active CopR dimers into inactive monomers.
CopR-dimer dissociation into monomers and
CopR-dimer binding to the DNA are described by
the following reaction scheme:
KD
K2
2C „ C2 ‡ D „ C2 D
Figure 3. Binding stoichiometry. A, Titration at a constant DNA concentration of 0.25 mM with increasing
His6-CopR concentrations. An ethidium bromide-stained
gel of the EMSA and the respective graph are shown.
B, Titration at a constant His6-CopR concentration of
0.5 mM with increasing DNA concentrations. An ethidium bromide-stained gel of the EMSA and the respective graph are shown. The data for both graphs were
averaged from two independent experiments. In both
cases the 62 bp non-labeled DNA fragment KS1 was
used as CopR target. The relative error for the determination of protein concentration and DNA concentration
was 10%.
598
CopR Binding Stoichiometry
Figure 4. Concentration-dependence of the average
molecular mass of CopR determined by analytical ultracentrifugation. Equilibrium sedimentation measurements
were performed at different His6-CopR concentrations in
the presence of 75 mM NaCl. The obtained data, averaged from scans at three different wavelengths, were
used to calculate the average molecular mass for each
concentration. The plot shows a curve ®t based on
the assumption of a monomer-dimer equilibrium.
M1 ˆ 11.900 kDa (His6-CopR).
Figure 5. Binding curve. The binding curve was
obtained by titration of a constant concentration of the
radioactively labeled DNA-fragment KS1 (0.5 nM) with
increasing concentrations of puri®ed His6-CopR. The
reaction mixture, the electrophoresis buffer and the
polyacrylamide gel contained 75 mM NaCl.
Determination of the intracellular
concentration of CopR
The existing equilibria can be described by the
following equations:
KD ˆ ‰CŠ2 =‰C2 Š
…1†
K2 ˆ ‰C2 Š‰DŠ=‰C2 DŠ
…2†
KD K2 ˆ ‰CŠ2 ‰DŠ=‰C2 DŠ
…3†
‰D0 Š ˆ ‰DŠ ‡ ‰C2 DŠ
…4†
‰C0 Š ˆ ‰CŠ ‡ 2‰C2 Š
…5†
[C0] and [D0] are the total protein concentration
and the total DNA concentration, respectively.
Since at every single protein concentration the
fraction of bound protein is < 2% of total protein
concentration, [C0] is approximately [C] ‡ 2[C2].
The following expression can be derived from the
above equations:
To determine the intracellular CopR concentration, B. subtilis strain DB104 (pCOP4) was
grown logarithmically and the cell titer was determined. Subsequently, crude cell extracts containing
CopR were prepared by sonication and analyzed
together with different concentrations of puri®ed
His6-CopR on an SDS/17.5% polyacrylamide gel
followed by Western blotting. Figure 6 shows a
typical Western blot. Crude extracts from two
independently grown B. subtilis cultures were analyzed in up to six parallels. As an internal control
(data not shown here), extracts were used from
B. subtilis strain DB104(pCOP7) containing a
pIP501 derivative, which replicates at 10 to 20-fold
higher copy number due to the lack of RNAIII
(Brantl & Behnke, 1992) and, therefore, produces
‰C2 DŠ=…‰DŠ ‡ ‰C2 DŠ† ˆ 1=…1 ‡ KD K2 =……KD =4†
p
…ÿ1 ‡ 1 ‡ 8‰C0 Š=KD ††2 †
…6†
This equation was applied to ®t the binding
curve of His6-CopR to the wild-type target KS1,
which was obtained in the presence of 75 mM
NaCl (Figure 5). Based on the value of
KD ˆ 1.44(0.49) 10ÿ6 M obtained by analytical
ultracentrifugation at the same salt concentrations,
K2 was calculated to be 4(1.3) 10ÿ10 M.
Figure 6. Quanti®cation of the intracellular CopR concentration. Western blot. Lanes 1 to 3: 12, 6 and 2 ng of
puri®ed His6-CopR, respectively. Lanes 4 and 5: 5 and
10 ml of crude extracts from DB104(pCOP4) containing
wild-type CopR.
599
CopR Binding Stoichiometry
10 to 20-fold more CopR. The loss of CopR during
the preparation of the crude lysates was approximately 50%, as determined by adding a known
amount of puri®ed His6-CopR to a plasmid-free
(copRÿ) B. subtilis culture, followed by immediate
sonication and subsequent quantitative analysis
analogously to the CopR containing extracts.
For the wild-type DB104(pCOP4), the number of
CopR monomers per cell was calculated to be
approximately 15000. Given an average B. subtilis
cell volume of 1.00(0.17) 10ÿ15 l (Abril et al.,
1997), an intracellular CopR concentration of
roughly 20 to 30 10ÿ6 M was estimated.
Non-specific DNA-binding mode of CopR
The relatively high intracellular CopR concentration prompted us to perform DNA-binding
experiments with CopR concentrations corresponding to the in vivo conditions. For direct
comparison we performed EMSA with the wildtype target and the mutated DNA targets at a
concentration of 0.5 and 6 10ÿ6 M His6-CopR.
The result is shown in Figure 7. At a protein concentration of 0.5 10ÿ6 M, His6-CopR is able to
bind selectively to its wild-type target and to
some of the mutated targets, but, with the exception of KS9, with a lower af®nity (A). However,
at higher protein concentrations of 6 10ÿ6 M, a
new complex with decreased electrophoretic
mobility compared to that at lower protein concentrations was observed (B). This complex was
formed with all mutated targets, even with KS5,
which is not recognized and bound by CopR at
lower protein concentrations. Since the mutated
targets still bear a high level of sequence homology to the wild-type target, we tested whether
CopR is able to bind to a completely unrelated
sequence at cellular protein concentrations.
Figure 7C shows a comparison of CopR binding
to the wild-type target and to a 41 bp sequence
from pIP501, located downstream (nucleotides
299 to 340) from the CopR target sequence. The
speci®c CopR-dimer DNA complex was formed
only with the wild-type target, but the nonspeci®c complex was observed with both DNA
fragments. Titration of 6 10ÿ6 M His6-CopR
with 0.4 10ÿ6 to 4 10ÿ6 M unlabeled wildtype target indicated that the observed higherorder complex consists of four CopR monomers
bound to the DNA (data not shown).
Discussion
CopR binds to the DNA as a dimer
Figure 7. Non-speci®c binding mode of CopR.
A, His6-CopR (0.5 mM) binding to wild-type target KS1
and to mutated targets. Bound and unbound DNA
species are indicated by arrows. B, Interaction of His6CopR (6 mM) with wild-type target and mutated targets.
Bound and unbound DNA species are indicated by
arrows. C, Titration of a DNA fragment (1 mM) of completely unrelated sequence and of the wild-type target
(1 mM) with increasing concentrations of His6-CopR. The
unbound DNA, the speci®c complex, formed only with
wild-type target and the non-speci®c complex, formed
at higher protein concentrations with both targets, are
indicated by arrows.
In the hetero-oligomer experiment, the formation
of three speci®c complexes was observed (Figure 2).
The migration of the intermediate complex
suggests that it contains a heterodimer composed
of one monomer wild-type His6-CopR and one
monomer of His6-CopR20. We can, therefore,
conclude that CopR binds the DNA as a dimer.
Based on this assumption, the results of titration to
saturation experiments at either constant DNA or
constant CopR concentration (Figure 3) show that
both protein and DNA were about 90% active. The
link between protein assembly and DNA-binding
was demonstrated also for many other gene regulatory systems such as Cro (Jana et al., 1997), lac
repressor (Chakerian & Matthews, 1992), l cI
repressor (Bain & Ackers, 1994) and the tryptophan
repressor (LeTilly & Royer, 1993).
CopR dimers are the predominant
species in vivo
Analytical ultracentrifugation showed that at the
protein concentrations used in the experiments
CopR exists as an equilibrium mixture of monomers and dimers in solution. The in vitro equili-
600
brium dissociation constant KD for CopR dimers
was determined to be 1.44(0.49) 10ÿ6 M. Thus,
the interactions between the two monomers are
relatively weak. The equilibrium dissociation constant of the CopR-DNA complex was calculated to
be 4(1.3) 10ÿ10 M. In this concentration range
CopR is mostly monomeric.
To clarify whether these assumptions also re¯ect
the in vivo conditions, we calculated the intracellular concentration of CopR by quantitative Western
blotting (Figure 6). The estimated number of CopR
molecules per B. subtilis cell is approximately
15,000, which corresponds to an intracellular concentration of 20 10ÿ6 to 30 10ÿ6 M. This is 10
to 20-fold higher than KD for CopR-dimer dissociation and suggests that CopR is present mainly
as a dimer in B. subtilis cells. Therefore, we suggest
that CopR in vivo is likely to bind to its operator as
a preformed dimer. As shown previously, CopR
reduces the plasmid copy number about 10 to
20-fold in B. subtilis (Brantl & Behnke, 1992) and,
as expected from this ®nding, decreases the
amount of repR-mRNA about 20-fold (Brantl,
1994). These effects can be suf®ciently explained by
an in vivo concentration of CopR in the range of 20
to 30 mM.
Transcriptional repressors have been found
mostly in much lower intracellular concentrations:
e.g. l cI repressor with 100 to 200 molecules per
cell (Backman et al., 1976), CytR with 100 molecules per cell (Valentin-Hansen et al., 1978) and
trp repressor with 50 to 300 molecules per cell
(Kelley & Yanofsky, 1982). However, there are also
examples for higher intracellular concentrations as
the leucine-responsive regulatory protein Lrp (6000
molecules per cell; Willins et al., 1991) or the Vibrio
cholerae protein Fur (2500 molecules per cell;
Watnick et al., 1997).
CopB encoded by the enterobacterial plasmid
R1 is the only other transcriptional repressor controlling plasmid copy number, for which DNA
binding constant (KD of 1 10ÿ10 M, Riise &
Molin, 1986) and intracellular concentration
(about 1 mM) were estimated (Light & Molin,
1982). Both values are between 5 and 20 times
lower than in the case of CopR. It is not unexpected that a protein with a higher binding af®nity is less abundant in the cell.
With the calculation of the intracellular concentration of CopR, the concentrations of both inhibitory components, RNAIII and CopR, as well as the
concentration of the target of control, repR-mRNA,
have now been determined for plasmid pIP501.
Whereas for plasmid ColE1 such calculations were
carried out several years ago (Brenner &
Tomizawa, 1991), pIP501 is the ®rst plasmid replicating in Gram-positive bacteria where such a
quantitative assessment for all components
involved in plasmid copy number control has been
performed (this report and Brantl & Wagner,
1996).
CopR Binding Stoichiometry
CopR-DNA complex formation at higher
protein concentrations
Under the conditions used in our experiments
we found evidence for two different CopR binding modes: (i) at submicromolar protein concentrations speci®c binding of CopR to its target
DNA was observed; and (ii) at micromolar concentrations CopR was shown to bind non-speci®cally to the DNA. Whereas the speci®c complex
at lower protein concentrations was formed
exclusively with the wild-type target and with
some closely related mutated targets, at micromolar protein concentrations, a protein-DNA complex with decreased electrophoretic mobility
compared to the speci®c complex was observed
with all (related and unrelated) DNA targets. The
narrow concentration range at which formation
of the higher-order complex occurs and nearly
complete binding of the DNA is achieved might
be the result of cooperative binding of CopR
dimers to the DNA. However, it is not yet clear
whether this non-speci®c binding occurs in vivo,
since cellular components would be able to in¯uence the DNA binding status of CopR.
Several CopR mutant proteins have been found
to bind exclusively non-speci®cally in vitro and
were completely inactive in vivo in copy number
control (S.B. & K.S., unpublished data), but the corresponding strains did not show growth defects.
Therefore, it is very likely that even the high
(determined) in vivo concentrations of CopR do not
negatively in¯uence other cellular processes and
that target speci®city can be maintained.
Materials and Methods
DNA preparation and manipulation
Plasmid DNA was isolated from B. subtilis as
described (Brantl et al., 1990). DNA manipulations
(restriction enzyme cleavage, ligation, etc.) were carried
out at the conditions speci®ed by the manufacturer or
according to standard protocols (Sambrook et al., 1989).
A GenAmp polymerase chain reaction (PCR) kit from
Perkin Elmer/Cetus was used. DNA sequencing was
performed according to the dideoxy-chain termination
method (Sanger et al., 1977) with a Sequenase kit from
U.S. Biochemical.
Construction and copy number determination of
plasmid pCOP
20
Construction of plasmid pCOP20 for the expression
of a 30 -truncated copR gene was performed in two subsequent steps:
Firstly, two PCR fragments were generated on plasmid pUC119-F as template with either oligonucleotide
C312-32 (50 ACAGAACCAGAACCATAAACAGAACAAGTAAC) and the reverse sequencing primer or oligonucleotide C313-32 (50 GTTACTTGTTCTGTTTATGGTTCTGGTTCTGT) and the universal sequencing primer
and used as templates for a second PCR reaction with
reverse and universal sequencing primer to obtain a
2.3 kb fragment. This fragment was cleaved with EcoRI
601
CopR Binding Stoichiometry
and BamHI and inserted into EcoRI/BamHI-digested
plasmid pBT48 yielding plasmid pPR20. The obtained
point mutation (181 G ! T) was con®rmed by sequencing. Secondly, a functional copR gene was reconstituted
by insertion of a KpnI/XbaI fragment from plasmid
pCOP4 into KpnI/XbaI digested pPR20 resulting in
plasmid pCOP20.
The copy number of plasmid pCOP20 in B. subtilis
DB104 was determined as described before (Brantl &
Behnke 1992) and shown to be wild-type, indicating that
both binding and oligomerization motifs are intact,
although a slight impairment in DNA-binding or oligomerization or protein stability cannot be totally
excluded.
Construction of pQE
20 to overexpress CopR
20
in Escherichia coli
A single PCR step using pCOP20 as template and
oligonucleotides B618-30 (50 GAATTCGGATCCGAACTAGCATTTAGAGAA) and B568-28 (50 TCTAGAGGATCCTTTATTCAGTTCGTTG) was performed to
obtain a 300 bp DNA fragment encoding a C-terminal
truncated copR20 gene, which was subsequently
cleaved with BamHI and inserted into the unique BamHI
site of vector pQE9 (Quiagen). The correct insert orientation was con®rmed by sequencing. The resulting plasmid was designated pQE20 and used for the
overexpression and puri®cation of His6-CopR20 which
carries an N-terminal histidine-tag.
Preparation of labeled wild-type and mutant
CopR targets
Synthetic oligonucleotides were 50 end-labeled with
[g-32P]ATP (Sambrook et al., 1989) and puri®ed from
denaturing 8% polyacrylamide gels. Pairwise combinations of labeled/unlabeled oligonucleotides were
annealed (68 C for one hour, slow cooling overnight),
resulting in the wild-type DNA fragment KS1 and the
corresponding mutant DNA fragments shown in Table 1.
All oligonucleotides, except the 41 nt and the 44 nt long
oligonucleotides, carry four G or C residues, respectively, at their ends to facilitate correct annealing and to
promote additional stability.
Unlabeled double-stranded DNA fragments were prepared by mixing equimolar concentrations of the respective oligonucleotides followed by reannealing as
described above. The concentration of the DNA fragments was determined based on the molar extinction
coef®cient. The concentration of the used DNA stock
solutions were 5 to 10 mM for the 62 bp fragments and
95 mM for the 44 bp and the 41 bp fragments.
CopR-DNA binding reaction and EMSA
Binding reactions were performed in a ®nal volume of
20 ml containing 1 nM labelled or 50 to 700 nM
unlabeled DNA and 20 to 7000 nM protein. After incubation at 30 C for 30 minutes in electrophoresis buffer
(0.5 TBE), aliquots of the reaction mixtures were separated on native 8% polyacrylamide gels run at room
temperature for 1.5 hours (16 V/cm). The binding curve
for the determination of the equilibrium dissociation constant for the protein-DNA complex was obtained in the
presence of 75 mM NaCl in the reaction mixture and in
the electrophoresis buffer. To ensure that the reaction
reached equilibrium, aliquots of the mixtures were
analyzed on polyacrylamide gels after several hours of
incubation, and no changes in the results were observed.
Gels containing non-labeled DNA fragments were
stained with ethidium bromide, digitalized on a Stratagene EAGLE EYETM scanner and quanti®ed with the
TINA-Pcbas 2.0 software. Gels containing labeled DNA
fragments were visualized and quanti®ed on a Fuji PhosphorImager. Autoradiograms were made from dried
gels.
Protein purification
Wild-type and mutant CopR protein (His6-CopR and
His6-CopR20) expressed from pQEC and pQE20,
respectively, carrying an N-terminal His6-tag were puri®ed by af®nity chromatography on Ni-NTA-Agarose
(Qiagen). Since the purity of the resulting proteins was
not suf®cient for further analysis, a subsequent puri®cation by HPLC reversed-phase chromatography on a
C18-column was performed. Elution was carried out with
an acetonitrile/water gradient, both solvents containing
0.1% (v/v) tri¯uroactetic acid. After elution, the protein
was lyophilized and stored at 4 C. For analysis, the protein was resuspended in phosphate buffer (50 mM,
Table 1. Sequences of the DNA fragments used in this
work
Only the sequence of the top strand is indicated. KS1 has the
wild-type target sequence. The bases of site I and II contacted by
CopR are boxed and the positions of base exchanges in the
mutated targets are highlighted.
602
pH 7.9, 150 mM NaCl), aliquoted and stored at ÿ70 C.
This CopR preparation, as analyzed by Coomassiestained SDS-PAGE, was judged to be >99% pure.
The protein content was determined by amino acid
analysis (acidic hydrolysis) on an amino acid analyzer
LC 3000 (Eppendorf/Biotronik, Germany). The estimated
relative error for this method is 10%. Furthermore, the
concentration determined by Bradford analysis was compared with the result of the amino acid analysis and
found to be reliable too. The difference was within the
range of 10% relative error. For fast determination of
protein concentration with small amounts of protein, a
calibration curve for the Bradford assay was measured
using His6-CopR. Based on this calibration curve, the
protein concentration of the HPLC-puri®ed mutant
CopR20 was determined.
Analytical ultracentrifugation
The molecular mass of CopR was analyzed by means
of an analytical ultracentrifuge XL-A (Beckman, Palo
Alto, CA) using the sedimentation equilibrium technique. CopR samples (100 ml) dissolved in 0.5 TBE
buffer (pH 8.0) containing 75 mM NaCl were ®lled in
six-channel cells and centrifuged for two hours at
30,000 rpm (over-speed) followed by about 20 hours at
26,000 rpm (equilibrium speed). The radial absorbance
distribution curves at sedimentation equilibrium were
recorded at 225, 230 and 235 nm. The three different
curves were ®tted simultaneously using the program
POLYMOLE (Behlke et al., 1997). The concentrationdependent apparent molecular mass values (MW) were
analyzed according to a monomer-dimer equilibrium.
An M1 value of 11.900 kDa for His6-CopR was used.
Determination of the intracellular concentration of
CopR in B. subtilis
B. subtilis strain DB104 (pCOP4) was grown in TY
medium with phleomycin. Cells from the logarithmic
growth phase were harvested and the cell titer was
determined by plating independent parallels of different
dilutions on selective TY medium to be 1.5 108/ml.
A sample (20 ml, corresponding to 3 109 cells) was pelleted, suspended in 400 ml of sonication buffer (Brantl,
1994) and sonicated three times for 30 seconds. After
centrifugation at 4 C, the supernatant was obtained,
80 ml of SDS-loading buffer was added and heated for
®ve minutes at 95 C. Different amounts of this crude
extract containing wild-type CopR were subjected to
SDS-PAGE on a 17.5% gel, in parallel with different
dilutions of HPLC-puri®ed His6-CopR of known concentration. Western blotting was performed using PVDF
membrane (NEN Dupont), a polyclonal peptide antiserum against wild-type CopR and a chemoluminescence kit from NEN Dupont. X-ray ®lm was exposed to
the blot for different times to ensure a linear response on
the ®lm. Films were digitalized with a MICROTEK ScanMaker E6 scanner and analyzed with TINA-Pcbas 2.0
software. To estimate CopR losses during sonication,
B. subtilis strain DB104 (copRÿ) was grown under the
conditions described above. Cells were harvested and a
known amount of puri®ed His6-CopR was added prior
to sonication. A loss of about 50% was calculated.
CopR Binding Stoichiometry
Acknowledgements
We thank E. Birch-Hirschfeld (Institut fuÈr Virologie,
Jena) for synthesizing the oligodeoxyribonucleotides,
Torsten Steinmetzer for help with the HPLC puri®cation of wild-type and mutant CopR proteins, and
Lydia Seyfarth (both Institut fuÈr Biochemie, Jena) for
the amino acid analysis of CopR. We are grateful to
Gerhart Wagner (SLU Uppsala, Sweden) for critically
reading the manuscript. This work was supported by
grant Br 1552/2-1 from the Deutsche Forschungsgemeinschaft (to S.B.).
References
Abril, A. M., Salas, M., Andreu, J. M., Hermoso, J. M. &
Rivas, G. (1997). Phage 29 protein p6 is in a
monomer-dimer equilibrium that shifts to higher
association states at the millimolar concentrations
found in vivo. Biochemistry, 36, 11901 ±11908.
Backman, K., Ptashne, M. & Gilbert, W. (1976).
Construction of plasmids carrying the cI gene of
bacteriophage l. Proc. Natl Acad. Sci. USA, 73,
4174± 4178.
Bain, D. L. & Ackers, G. (1994). Self-association and
DNA binding of lambda cI repressor N-terminal
domains reveal linkage between sequence-speci®c
binding and the C-terminal cooperativity domain.
Biochemistry, 33, 14679± 14689.
Behlke, J., Ristau, O. & SchoÈnfeld, H.-J. (1997). Nucleotide-dependent complex formation between the
Escherichia coli chaperonins GroEL and GroES studied under equilibrium conditions. Biochemistry, 36,
5149± 5156.
Behnke, D., Malke, H., Hartmann, M. & Walter, F.
(1979). Post-transformational rearrangement of an
in vitro reconstructed group-A streptococcal erythromycin resistance plasmid. Plasmid, 2, 605± 616.
Brantl, S. (1994). The CopR protein of plasmid pIP501
acts as transcriptional repressor at the essential repR
promoter. Mol. Microbiol. 14, 473± 483.
Brantl, S. & Behnke, D. (1992). Copy number control of
the streptococcal plasmid pIP501 occurs at three
levels. Nucl. Acids Res. 20, 395± 400.
Brantl, S. & Wagner, E. G. H. (1994). Antisense RNAmediated transcriptional attenuation occurs faster
than stable antisense/target RNA pairing: an in vitro
study of plasmid pIP501. EMBO J, 13, 3599± 3607.
Brantl, S. & Wagner, E. G. H. (1996). An unusually
long-lived antisense RNA in plasmid copy number
control: in vivo RNAs encoded by the streptococcal
plasmid pIP501. J. Mol. Biol. 255, 275± 288.
Brantl, S. & Wagner, E. G. H. (1997). Dual function of
the copR gene product of plasmid pIP501. J. Bacteriol.
179, 7016± 7024.
Brantl, S., Behnke, D. & Alonso, J. C. (1990). Molecular
analysis of the replication region of the conjugative
Streptococcus agalactiae plasmid pIP501 in Bacillus
subtilis. Comparison with plasmids pAMb1 and
pSM19035. Nucl. Acids Res. 18, 4783±4790.
Brantl, S., Birch-Hirschfeld, E. & Behnke, D. (1993).
RepR protein expression on plasmid pIP501 is controlled by an antisense RNA-mediated transcription
attenuation mechanism. J. Bacteriol, 175, 4052± 4061.
Brenner, M. & Tomizawa, J. (1991). Quantitation of
ColE1± encoded replication elements. Proc. Natl
Acad. Sci. USA, 88, 405± 409.
CopR Binding Stoichiometry
Ceglowski, P., Lurz, R. & Alonso, J. C. (1993). Functional analysis of pSM19035 derived replicons in
Bacillus subtilis. FEMS Microbiol. Letters, 109, 2 ± 3.
Chakerian, A. E. & Matthews, K. S. (1992). Effect of lac
repressor oligomerization on regulatory outcome.
Mol. Microbiol. 6, 963± 968.
del Solar, G. H., Perez-Martin, J. & Espinosa, M. (1990).
Plasmid pLS1-encoded RepA protein regulates transcription from repAB promoter by binding to a
sequence containing a 13-base pair symmetric
element. J. Biol. Chem. 265, 12569± 12575.
Harrison, S. C. (1991). A structural taxonomy of DNAbinding domains. Nature, 353, 715± 719.
Hope, I. A. & Struhl, K. (1987). GCN4, a eukaryotic
transcriptional activator protein, binds as a dimer to
target DNA. EMBO J. 6, 2781± 2784.
Horodniceanu, T., Bouanchaud, D. H., Bieth, G. &
Chabbert, Y. A. (1976). R-plasmids in Streptococcus
agalactiae (group B). Antimicrob. Agents Chemother,
10, 795± 801.
Jana, R., Hazbun, T. R., Mollah, A. K. & Mossing, M. C.
(1997). A folded monomeric intermediate in the formation of lambda Cro dimer-DNA complexes.
J. Mol. Biol. 273, 402±416.
Kelley, R. L. & Yanofsky, C. (1982). trp aporepressor
production is controlled by autogenous regulation
and inef®cient translation. Proc. Natl Acad. Sci. USA,
79, 3120± 3124.
Le, Chatelier E., Ehrlich, S. D. & JannieÂre, L. (1994). The
pAMb1 Cop repressor regulates plasmid copy number by controlling transcription of the repE gene.
Mol. Microbiol. 14, 463±471.
LeTilly, V. & Royer, C. A. (1993). Fluorescence anisotropy assays implicate protein-protein interactions
in regulating trp repressor DNA binding. Biochemistry, 32, 7753± 7758.
603
Light, J. & Molin, S. (1982). The sites of action of the
two copy number control functions of plasmid R1.
Mol. Gen. Genet. 187, 486± 493.
Riise, E. & Molin, S. (1986). Puri®cation and characterization of the CopB replication control protein, and
precise mapping of its target site in the R1 plasmid.
Plasmid, 15, 163±171.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Sanger, F., Nicklen, E. F. & Coulson, A. R. (1977). DNA
sequencing with chain terminating inhibitors. Proc.
Natl Acad. Sci. USA, 74, 5463± 5467.
Steinmetzer, K. & Brantl, S. (1997). Plasmid pIP501
encoded transcriptional repressor CopR binds
asymmetrically at two consecutive major grooves of
the DNA. J. Mol. Biol. 269, 684± 693.
Swin®eld, T. J., Oultram, J. D., Thompson, D. E., Brehm,
J. K. & Minton, N. P. (1990). Physical characterisation of the replication region of the Streptococcus
faecalis plasmid pAMb1. Gene, 87, 79 ± 90.
Valentin-Hansen, P., Svenningsen, B. A., MunchPetersen, A. & Hammer-Jespersen, K. (1978). Regulation of the deo operon in Escherichia coli. Mol. Gen.
Genet. 159, 191± 202.
Watnick, P. I., Eto, T., Takahashi, H. & Calderwood,
S. B. (1997). Puri®cation of Vibrio cholerae Fur and
estimation of its intracellular abundance by antibody sandwich enzyme-linked immunosorbent
assay. J. Bacteriol. 179, 243± 247.
Willins, D. A., Ryan, C. W., Platko, J. V. & Calvo, J. M.
(1991). Characterization of Lrp, an Escherichia coli
regulatory protein that mediates a global response
to leucine. J. Biol. Chem. 266, 10768± 10774.
Edited by T. Richmond
(Received 23 March 1998; received in revised form 27 July 1998; accepted 29 July 1998)