JOURNAL OF BACTERIOLOGY, Nov. 1997, p. 7016–7024
0021-9193/97/$04.0010
Copyright © 1997, American Society for Microbiology
Vol. 179, No. 22
Dual Function of the copR Gene Product of Plasmid pIP501
SABINE BRANTL1*
AND
E. GERHART H. WAGNER2
Institut für Molekularbiologie, Friedrich-Schiller-Universität Jena, Jena D-07745, Germany,1
and Department of Microbiology, Uppsala Genetic Center, Swedish University
of Agricultural Sciences, S-75007 Uppsala, Sweden2
Received 7 July 1997/Accepted 11 September 1997
i.e., an effect of opposing transcriptional activity. Sense RNA
transcription through the antisense promoter sequence interfered with normal copA expression. The original suggestion was
that this effect was on the level of CopA termination, giving
rise to longer (200-nucleotide [nt]) CopA transcripts that were
presumed to be inactive in inhibition, but later work (4) indicated that CopA activity was decreased due to its lower intracellular concentration.
Convergent transcription may also occur in copR mutant
derivatives of pIP501. If this is the case, repression of pII in the
presence of CopR should permit an increase in RNAIII transcription.
The aim of our present work was to test this assumption. The
following data reported here support the idea that convergent
transcription occurs: (i) the half-life of RNAIII is independent
of CopR; (ii) wild-type, but not a truncated, CopR protein
supplied in cis or in trans affects RNAIII accumulation when
the test plasmid encodes both rnaII and rnaIII; and (iii) a
fourfold reduction of RNAIII levels by RNAII transcription is
obtained only when rnaII and rnaIII are encoded in cis. Additionally, we show that a heterologous plasmid expressing only
rnaIII exerted stronger incompatibility with wild-type pIP501
than one expressing both rnaII and rnaIII. Experiments with
the gyrase inhibitor novobiocin indicated that transcription of
RNAII and RNAIII was increased by the antibiotic, consistent
with the possibility that the convergent transcription effect on
RNAIII accumulation is on the level of induced alteration of
DNA supercoiling in the pIII promoter region.
The activity of the copR gene product of pIP501 thus has a
dual effect: (i) partial repression of the synthesis of RNAII
results in decreased RepR synthesis, and (ii) prevention of
convergent transcription allows for increased initiation frequency at the antisense promoter, thus amplifying the inhibitory effect. The interplay of the RNAIII and CopR parts of the
regulatory circuit of copy number control of pIP501 is discussed.
Replication of the broad-host-range plasmid pIP501 (19)
is controlled at a step subsequent to transcription initiation
by CopR (7) and posttranscriptionally by an antisense RNA
(RNAIII) (11). RNAIII induces transcriptional attenuation of
RNAII within the leader region of the repR mRNA (11, 12).
CopR binds to a 44-bp sequence upstream of and partially
overlapping the repR promoter pII, thereby repressing RNAII
synthesis about 10-fold (7). Deletion of copR resulted in a 10to 20-fold increase of pIP501 copy number (9), consistent with
the elevated synthesis of RNAII, which encodes the rate-limiting RepR protein (10). Recently, we measured the steady-state
concentrations of RNAIII in Bacillus subtilis cells in high-copynumber and low-copy-number pIP501 derivatives. Since RNAIII
is synthesized constitutively, its concentration should be directly correlated with plasmid copy number. Surprisingly, we
found that this was not the case: both pPR1 (rnaIII1 copR; 50
to 100 copies/cell) and pCOP4 (rnaIII1 copR1; 5 to 10 copies/
cell) expressed about 1,000 to 2,000 molecules of RNAIII/cell
(1.4 to 2.8 mM [13]). This suggested that, in addition to its
function as a repressor of RNAII transcription, CopR is required for RNAIII accumulation. Previous analyses using
transcriptional promoter-lacZ fusions ruled out the idea that
CopR activates transcription from pIII (7). Hence, CopR must
act either by affecting the stability of RNAIII or by indirectly
increasing transcription (see below).
In the case of the enterobacterial plasmid R1, a repressor
protein (CopB) and an antisense RNA (CopA) are involved in
controlling copy number (26). Deletion of copB was shown to
result in a decrease in total CopA activity. This was inferred to
be due to a phenomenon called convergent transcription (31),
* Corresponding author. Mailing address: Institut für Molekularbiologie, Friedrich-Schiller-Universität Jena, Winzerlaer Strasse 10, Jena
D-07745, Germany. Phone: 49 3641 657576/78. Fax: 49 3641 657520.
E-mail: [email protected].
7016
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
Replication of plasmid pIP501 is regulated at a step subsequent to transcription initiation by an antisense
RNA (RNAIII) and transcriptionally by a repressor protein, CopR. Previously, it had been shown that CopR
binds to a 44-bp DNA fragment upstream of and overlapping the repR promoter pII. Subsequently, we found
that high-copy-number pIP501 derivatives lacking copR and low-copy-number derivatives containing copR produced the same intracellular amounts of RNAIII. This suggested a second, hitherto-unknown function of CopR.
In this report, we show that CopR does not affect the half-life of RNAIII. Instead, we demonstrate in vivo that,
in the presence of both pII and pIII, CopR provided in cis or in trans causes an increase in the intracellular
concentration of RNAIII and that this effect is due to the function of the protein rather than its mRNA. We
suggest that, in the absence of CopR, the increased (derepressed) RNAII transcription interferes, in cis, with
initiation of transcription of RNAIII (convergent transcription), resulting in a lower RNAIII/plasmid ratio.
When CopR is present, the pII promoter is repressed to >90%, so that convergent transcription is mostly abolished and RNAIII/plasmid ratios are high. The hypothesis that RNAII transcription influences promoter pIII
through induced changes in DNA supercoiling is supported by the finding that the gyrase inhibitor novobiocin
affects the accumulation of both sense and antisense RNA. The dual role of CopR in repression of RNAII transcription and in prevention of convergent transcription is discussed in the context of replication control of pIP501.
VOL. 179, 1997
DUAL FUNCTION OF CopR
7017
TABLE 1. Plasmids used in this study
Plasmid
a
Reference
r
pIP501 derivative, lacking the copR sequence upstream of the HpaI site; Cm
E. coli cloning vectors, Apr, multiple cloning site
pBT48 (8), carrying the complete replication region of pIP501; Pmr
pCOP4 derivative carrying a frameshift mutation within the copR gene
pPR1 derivative lacking the 210 region of promoter pIII (bp 470 to 475)
pGK13 (21) derivative, Emr, containing a transcriptional terminator as EcoRI fragment
pGK14 derivative, comprising pIII/rnaIII
Similar to pGKIII/1 but carrying a point mutationa (G408-A)
pGT1 derivative, comprising pII/IR1, pIII, IR2, and 100 bp downstream of IR2
pGTW derivative, carrying a point mutationa (G408-A)
Shuttle vector of pUB110 and pUC18; Kmr Pmr
pUCB1 derivative, carrying the wild-type rnaIII gene
pUC18 derivative carrying a segment of the Streptococcus equisimilis streptokinase region
pGTW derivative lacking the 210 region of promoter pIII
pGT1 derivative, comprising pI, pII/IR1, pIII, IR2, and 100 bp downstream of IR2
pGTC4 derivative carrying a frameshift mutation within the copR gene
pGKIII/1 derivative comprising the copR gene
pUC19 derivative carrying a transcriptional terminator
pTERM1 derivative with copR gene
pUCB1 derivative with copR gene
3
27
3
9
9
13
13
13
13
13
13
13
30
This
This
This
This
This
This
This
study
study
study
study
study
study
study
Mutations as manifested in the RNAIII sequence (see reference 8).
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions. Plasmids used in this study
to construct new derivatives of pIP501 are listed in Table 1. Plasmids were
propagated in B. subtilis DB104 (20), and Escherichia coli TG1 (27) was used for
subcloning and mutagenesis. All strains were routinely grown at 37°C on tryptone-yeast (TY) medium. For the isolation of total RNA, DB104 transformants
were cultivated on minimal medium containing Spizizen salts (1), 0.5% glucose,
0.2% Casamino Acids, and 5 mg of L-histidine per ml.
FIG. 1. Determination of the half-life of RNAIII in the presence and absence of CopR provided in cis. Half-lives were determined as described in Materials and
Methods. (Upper panel) The Northern blot analysis is shown. Samples were taken at indicated times after rifampin addition. Plasmids used are indicated. Hybridization
was with labelled RNAII. Reprobing to correct for loading errors was performed as described in Materials and Methods (data not shown). (Middle panel) Graphic
representation of the decay of RNAIII in the presence or absence of copR provided in cis. (Lower panel) Schematic linear maps of the two plasmids used. Arrows,
direction of transcription; small black boxes, promoters; jagged line, deletion of the last codon of copR and hence the deletion of the copR target upstream of pII.
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
pGB354
pUC18
pCOP4
pCOP8
pPR7
pGT1
pGKIII/1
pGK408
pGTW
pGT408
pUCB1
pCBW
pSU31
pGT7
pGTC4
pGTC8
pGKIIIC/2
pTERM1
pCOPT1
pCBCD
Description
7018
BRANTL AND WAGNER
J. BACTERIOL.
TABLE 2. Relative amounts of RNAIII in the presence
or absence of CopR supplied in cis
Plasmid
RNAs encoded
CopR
in cis
Relative signal
of RNAIIIa
pGTCW
pGTC8
pGTC4
RNAII and RNAIII
RNAII and RNAIII
RNAII and RNAIII
None
Truncated
Wild type
1.03
1.33
3.83
a
The amount of RNAIII was corrected for loading errors (Materials and
Methods). 3, corrected hybridization signal (pGTW); coefficients indicate the
corresponding increase. Calculated values represent averages from five independent preparations. Figure 2 shows an example of such an analysis.
DNA preparation and manipulation. DNA manipulations, restriction enzyme
cleavage, ligation, labelling with kinase, etc., were carried out under the conditions specified by the manufacturer or according to standard protocols (27). A
GenAmp PCR kit from Perkin-Elmer/Cetus was used for PCRs. DNA sequencing was performed according to the dideoxy chain termination method (28) with
a Sequenase kit from U.S. Biochemical.
Plasmid constructions. Plasmids for quantification of RNAIII, in the presence
or absence of CopR provided in cis, and for the calculation of RNAII in the
absence of RNAIII were constructed as follows. Plasmids pGTC4 and pGTC8
comprising pI, pII/IR1, and pIII/IR2 were constructed by cloning a 1.16-kb
BamHI/SalI PCR fragment into BamHI/SalI-digested vector pGT1. To this end,
plasmids pCOP4 and pCOP8 were used as PCR templates with oligonucleotides
B 844-29 (59 TCTAGAGGATCCAAATTCCCACTAAGCGC) and B 845-31 (59
GAATTCGTCGACCGTCATGAAGCACAGTTTC), respectively. For the construction of plasmid pGT7, a single PCR step was performed on pPR7 DNA with
oligonucleotides B 846-29 (59 GAATTCGGATCCAACAGAACCAGAACC
AG) and B 845-31. The resulting 500-bp PCR fragment encompassing pII/IR1
and pIII/IR2 was cleaved with BamHI and SalI and inserted into pGT1 as
described above.
A plasmid for tests of the effect of CopR on the half-life of RNAIII was
constructed as follows. The CopR target region overlaps the last codon and the
transcriptional terminator of the copR gene (7). To prevent CopR binding, we
removed its last codon and the transcriptional terminator. Previous experiments
showed that even longer truncations yield normal CopR activity in vivo (7a).
Plasmid pGKIII/1 (13) was used to insert the copR gene as a BamHI fragment
obtained by PCR on plasmid pCOP4 as template with oligonucleotides SB20 (59
GAATTCGGATCCTTAGAAATCATTGCTTTTTTCTTCTTCC) and SB21
(59 GAATTCGGATCCTGCAGCAAATTCCCACTAAGCGC 39) into the
unique BamHI site, resulting in plasmid pGKIIIC2 (copR gene in the same
orientation as the rnaIII gene). Gel shift assays with crude extracts from
DB104(pGKIIIC2) confirmed that plasmid pGKIIIC2 encodes a functional
CopR protein that can bind to its DNA target (data not shown). This means that
the deletion of the last copR codon did not affect the activity of CopR.
Plasmid pCBCD was constructed to provide CopR in trans. The last codon and
the transcriptional terminator of copR were replaced by a heterologous tran-
scriptional terminator, thus destroying the CopR target. The construction was
performed in three steps. First, a single PCR with oligonucleotides B 842-33 (59
AAGCTTGAATTCGTCGACTTCTAAAACGATGAG 39) and SB22 (59 AAG
CTTGAATTCTGCAGCTAGTCTCTACAACTAAG 39) on pSU31 DNA, carrying the bidirectional terminator of the streptokinase region (30), gave a 100-bp
fragment, which was cleaved with EcoRI and inserted into pUC19. Second,
pTERM1, carrying the terminator in the correct orientation, was used to insert
the BamHI fragment of plasmid pGKIIIC2 into the BamHI site of the pUC19
polylinker, resulting in plasmids pCOPT1 (copR gene in the proper orientation
with respect to the terminator) and pCOPT2 (opposite orientation from insert).
Since PCR introduced a PstI site at the 59 terminus of copR, the whole copR
terminator segment could be subcloned as a PstI fragment into pUCB1, yielding
pCBCD. Sequencing confirmed the orientation of the copR gene within pUCB1.
The activity of CopR was again tested in crude extracts prepared from B. subtilis
DB104(pCBCD).
Test for incompatibility. Strain DB104 carrying a wild-type pIP501 derivative
(pGB354, Cmr) was transformed with RNAIII or RNAII-RNAIII donor plasmids (pGKIII/1 and pGTW). Transformants were selected on erythromycin
(EM), encoded on the incoming plasmid. Subsequently, 200 transformants were
streaked in parallel on TY agar plates containing either EM or chloramphenicol.
Additionally, 12 transformants for each plasmid were grown overnight on TY
with EM, and plasmids were isolated.
Isolation of total RNA and Northern blot analyses. Isolation of total RNA,
Northern blot analysis, and reprobing of the filters to correct loading errors were
performed as described elsewhere (13). For the determination of RNAIII halflife, rifampin was added to a final concentration of 100 mg/ml to log-phase
cultures and time samples were taken, frozen in liquid nitrogen, and used for the
preparation of total RNA.
Novobiocin experiments. B. subtilis strains were grown on minimal medium to
early exponential phase (optical density at 560 nm, 0.1 to 0.2); freshly prepared
novobiocin (Sigma) was added to a final concentration of 10 mg/ml; and time
samples were taken after 10, 20, 30, 40, 50, and 60 min. A control culture was
grown without novobiocin and harvested after 60 min. Samples were frozen on
liquid nitrogen, and subsequent analyses were performed as described above.
RESULTS
(i) CopR protein does not affect the half-life of RNAIII.
Since previous experiments showed that CopR provided in
trans did not activate transcription initiation in pIII-lacZ fusions in the absence of the pII promoter (7), it was important
to analyze whether the higher RNAIII levels accumulating
from copR-proficient plasmids were due to a change in the
stability of RNAIII. B. subtilis DB104 transformants harboring
either plasmid pGKIII/1 (rnaIII1 copR) or pGKIIIC2 (rnaIII1
copR1) were used to determine the half-life of the antisense
RNA in the absence or presence of CopR. The use of these
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
FIG. 2. Steady-state concentration of RNAIII in the presence or absence of CopR supplied in cis. (Upper panel) The Northern blot is shown. Hybridization,
reprobing for 5S rRNA, and calculations were as described in Materials and Methods. The plasmids used are indicated. For pGTW and pGTC4, four samples and, for
pGTC8, three independent samples were taken and prepared separately. (Lower panel) Schematic linear maps (not to scale) of the plasmids used. Black boxes,
promoters; arrows, direction of transcription; stem loop, heterologous transcription terminator inserted 100 bp downstream of the pII attenuator. The terminator was
needed to obtain shortened full-length RNAII for Northern blot analyses (pGTW). The jagged line in the copR gene (pGTC8) indicates a frameshift mutation which
results in synthesis of a truncated (CopR*) protein.
VOL. 179, 1997
DUAL FUNCTION OF CopR
7019
TABLE 3. Relative amounts of RNAIII in the presence
or absence of CopR supplied in transa
Plasmid 1
RNA(s)
encoded
pCBCD
pGKIII
pGKIII
pGTW
pGTW
pGTW
None
RNAIII
RNAIII
RNAII and -III
RNAII and -III
RNAII and -III
Plasmid 2
pCBCD
pUCB1
pCBCD
CopR
in trans
Relative signal
of RNAIII
No
No
Yes
No
Vector
Yes
0
4.53
4.93
1.03
1.23
3.83
a
Relative RNAIII signals were calculated as for Table 2. The calculated values
represent averages from five independent preparations (three parallel samples
each). Figure 3 shows an autoradiogram of such an analysis.
two plasmids permits an assessment of possible interactions
between RNAIII and CopR without interference from other
components of the replication machinery of pIP501. From the
slope of semilogarithmic plots shown in Fig. 1 (lower panel),
the half-life of RNAIII was calculated to be 12 min (pGKIII/1)
and 13 min (pGKIIIC2). The insignificant difference between
these values suggests that CopR does not increase the metabolic stability of RNAIII.
(ii) Wild-type, but not truncated, CopR protein provided in
cis or in trans increases the steady-state concentration of RNAIII.
The pGK14 derivative pGT1 (compatible with pIP501) was
used as a carrier of different fragments of the pIP501 replication region placed upstream of a heterologous transcriptional
terminator (Materials and Methods). The resulting plasmids
were called pGTC4 (rnaIII1 copR1), pGTC8 (rnaIII1 copR
truncation) and pGTW (rnaIII1 copR). All inserted fragments
contained pII, pIII, the terminator IR2, and 100 bp downstream of IR2. In the case of the pGTC derivatives, pI and the
wild-type (pGTC4) or truncated (pGTC8) copR gene were
present. The pGT1 derivatives do not depend on pIP501 elements to replicate in B. subtilis; hence, copy number effects on
RNAIII concentration are eliminated. All plasmids were transferred into B. subtilis DB104, and total RNA was isolated and
subjected to Northern blotting.
As shown in Fig. 2 and Table 2, RNAIII was about fourfold
more abundant in cells containing pGTC4 encoding an active
CopR protein than in cells carrying pGTW (no copR) or
pGTC8 (frameshift mutation in the copR gene). Apparently,
the presence of a functional CopR protein permits the synthesis of higher intracellular amounts of RNAIII than those obtained in the absence of the protein.
A two-plasmid system was used to investigate whether the
effect of intact CopR on the accumulation of RNAIII was
indirect. The pIP501 copR gene lacking its own target region
was inserted into pUCB1, resulting in pCBCD, to provide CopR
in trans. A second plasmid was either pGTW containing a pII
and a pIII/rnaIII segment or pGKIII/1 carrying only the rnaIII
gene. Double plasmid strains were constructed, and Northern
blot analyses were performed. Figure 3 and Table 3 indicate
that RNAIII accumulation from pGKIII/1 is unaffected when
CopR is supplied in trans, again supporting the idea that CopR
does not activate transcription from pIII under conditions in
which this promoter is isolated from pII (7). In contrast,
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
FIG. 3. Steady-state concentration of RNAIII in the presence or absence of transcription from pII with CopR supplied in trans. (Upper panel) The Northern blot
is shown. Hybridization, reprobing for 5S rRNA, and calculations were as described in Materials and Methods. Three independently prepared sets of samples are shown.
(Lower panel) Schematic linear maps (not to scale) of the plasmids used (for details of construction, see Materials and Methods and Table 1). Arrows, direction of
transcription; black boxes, promoters; stem loop, heterologous transcription terminator inserted 100 bp downstream of the pII attenuator and at the end of the copRD
gene. In the case of pCBCD, the terminator was used for correct transcription termination of the copRD gene (in pCBCD), which lacks its own transcriptional terminator.
7020
BRANTL AND WAGNER
FIG. 4. Steady-state concentration of RNAIII in the presence or absence of
transcription from pII in trans. (Upper panel) Northern blot analysis was performed as described in Materials and Methods. Plasmids present are indicated.
The membrane was successively reprobed for RNAIII, RNAII, and 5S rRNA,
respectively. RNAIIF and RNAIIT refer to readthrough transcripts and terminated RNAII, respectively (13). Three arbitrarily chosen samples are presented.
Values for corrected signals (RNAIII/5S rRNA) are shown; the ratio of the
signals in the middle left panel was set to 1.0. (Lower panel) Schematic linear
plasmid maps (not to scale). Arrows, direction of transcription; black boxes,
active promoters; white box, inactive promoter pIII (deletion of 210 box).
in trans was present in the cell, the fraction of terminated over
readthrough RNAII remained unchanged.
When transcription occurred simultaneously from promoters pII and pIII (pGTW [Fig. 5]), RNAIII levels decreased
about 10 to 20 min after addition of novobiocin, and—indicating lower inhibitory activity of RNAIII—the ratio of terminated (T) to full-length (F) RNAII transcripts (13) changed
drastically, from 60 to 10% termination (Table 4). Thus, transcriptional activity from pII through the pIII region causes a
decrease in pIII activity (see above), and this effect dominates
over that caused by novobiocin alone.
To circumvent the problem that the long half-life of RNAIII
could mask an effect of supercoiling on de novo transcription,
the experiment was repeated with pGK408 and pGT408. As
the 408 mutant allele encodes an unstable variant of RNAIII
(half-life of about 2 min) (13), one expects that rapid changes
in RNAIII synthesis could be monitored more easily. As shown
in Fig. 6 and Table 5, the effect of novobiocin on the accumulation of RNAIII (pGK408; fivefold increase) was even more
pronounced than in the case of pGKIII/1, indicating that initiation at pIII contributes greatly to the observed accumulation
of RNAIII. In the case of pGT408 (transcription from both the
sense and the antisense promoter), RNAIII accumulation was
decreased in a manner similar to that in the case of pGTW.
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
RNAIII synthesis from pGTW was significantly (about fourfold) increased in the presence (pCBCD), but not in the absence, of a copR gene in trans (pUCB1).
Interestingly, RNAIII synthesized in cells carrying pGKIII/1
was about four- to fivefold more abundant than that in those
harboring pGTW, suggesting that RNAII synthesis or accumulation interferes with either RNAIII synthesis (transcription)
or accumulation. Figure 3 and Table 3 also demonstrate clearly
that active CopR permits RNAIII accumulation by repression
of pII, allowing for higher transcriptional activity from pIII.
(iii) RNAII transcription affects RNAIII levels only if both
are encoded on the same DNA segment. The results presented
so far suggest that, whenever RNAII is transcribed in cis, the
synthesis of RNAIII is decreased. If this is due to interference
of the RNA polymerase transcribing the rnaII region with the
activity of the pIII promoter, RNAII transcription from a second plasmid should not affect RNAIII accumulation. Figure 4
shows an experiment to test the effect of RNAII transcription
in trans. Filters with immobilized, extracted RNA from strains
carrying either a plasmid encoding only rnaIII (pCBW) or two
plasmids encoding rnaII and rnaIII separately (pGT7 and
pCBW) were probed for RNAIII, RNAII, and 5S rRNA (loading control). The quantification of the hybridization signals
clearly showed that RNAII transcription of a second template
fails to affect RNAIII accumulation significantly.
(iv) A plasmid expressing only rnaIII exerts stronger incompatibility with pIP501 than a plasmid expressing both rnaII
and rnaIII. Since RNAIII is the major incompatibility determinant of pIP501, its activity can be measured independently
from the Northern blot experiments above. Either one of the
two plasmids, expressing only rnaIII (pGKIII/1) or both rnaII
and rnaIII (pGTW), was introduced into B. subtilis to challenge
a resident wild-type pIP501 derivative (pGB354) (see Materials and Methods for details). When only rnaIII was expressed
from the incoming plasmid (pGKIII/1), only 12% of the tested
transformants carried both plasmids, whereas expression of
both rnaII and rnaIII (pGTW) failed to entail strong incompatibility (94% of the colonies still contained both plasmids).
Accordingly, isolation of plasmids from 12 transformants
grown overnight in liquid TY medium selecting for the incoming plasmid (pGKIII/1) showed no pGB354 (wild-type pIP501)
plasmid band, whereas in the case of pGTW, 11 of 12 transformants contained both plasmids. Hence, pGTW is inefficient
in displacing the wild-type pIP501 derivative after overnight
cultivation, as expected from four- to fivefold-lower RNAIII
accumulation when pII is present and active (see above and
Fig. 3 and Table 3).
(v) The gyrase inhibitor novobiocin affects accumulation of
RNAII and RNAIII. Bacterial promoters can vary with respect
to their sensitivity to DNA supercoiling (e.g., 18, 29). Novobiocin is an inhibitor of subunit B of DNA gyrase. To test whether
transcription from pII or pIII and, hence, accumulation of
RNAII and RNAIII are affected by changes in DNA supercoiling, B. subtilis DB104(pGKIII) (only pIII/rnaIII), DB104
(pGT7) (only pII; segment of rnaII), and DB104(pGTW) (both
pII and pIII/rnaIII and segment of rnaII) were grown on minimal medium to early logarithmic phase, novobiocin was added
to 10 mg/ml, time samples were withdrawn, and RNA was isolated and subjected to Northern blot analysis. As shown in Fig.
5A, RNAIII levels (pGKIII/1) increased after 20 min of novobiocin treatment and remained at a two- to threefold-higher
level until about 50 min. RNAII levels (pGT7) also increased
at 10 min after addition of novobiocin and decreased between
50 and 60 min to the initial level. Since pGT7 does not produce
RNAIII, and no second compatible plasmid providing RNAIII
J. BACTERIOL.
VOL. 179, 1997
DUAL FUNCTION OF CopR
7021
DISCUSSION
Previous studies showed that plasmid pIP501 is unusual in
that it encodes a regulatory antisense RNA that is very stable
(13). We argued previously that long-lived antisense RNAs
should be poor regulators since fortuitous decreases in plasmid
copy number in individual cells would be only slowly corrected,
resulting in unstable maintenance. The presence of the copR
gene in pIP501 and the results presented in this communication indicate a strategy used by pIP501 to cope with the risk of
unstable inheritance. We suggest that the effect of CopR is
exerted, by the same molecular event, on two levels: transcriptional repression of repR mRNA synthesis and accumulation of
RNAIII by prevention of convergent transcription.
A surprising result from previous work (13) was the similar
intracellular concentrations of RNAIII in cells carrying either
a high-copy-number copR mutant or a low-copy-number copR1
plasmid. This indicated that CopR either affected the stability
of RNAIII or exerted another, more indirect effect on its
synthesis. The results shown in Fig. 1 ruled out an effect on the
metabolic stability of RNAIII. A comparison of the intracellular concentrations of RNAIII in the presence or absence of
pII and the RNAIII target region and in the presence or
absence of CopR in cis or in trans showed that accumulation of
RNAIII was affected by RNAII transcription only in cis (Fig. 3
and 4 and Table 3). A key result is presented in Fig. 4: synthesis
of RNAII per se (for instance, if RNAII is provided in trans by
pGT7) does not affect the concentration of RNAIII. In contrast, when transcription of RNAII is permitted in cis—either
because the pII promoter is derepressed in the absence of
copR (pGTW [Fig. 2 and 3]) or because the CopR protein is
inactive due to truncation (pGTC8 [Fig. 2])—RNAIII levels
are four- to fivefold lower than those in the absence of RNAII
transcription (pGKIII [Fig. 3]). Hence, transcriptional activity
from the pII promoter rather than, e.g., RNAII-RNAIII interaction interferes with RNAIII synthesis. Functional tests were
in line with the RNAII transcription-induced decrease of
RNAIII concentration, since plasmid pGKIII/1 (only rnaIII)
showed strong incompatibility with a resident pIP501 wild-type
plasmid, whereas pGTW (rnaII/rnaIII) showed poor activity.
The phenomenon characterized by mutual or directional
effects of opposing transcriptional activity on RNA synthesis is
termed convergent transcription, and its effect on plasmid replication was first observed and described for the enterobacterial plasmid R1 (31). The results presented here are probably
pertinent to the same molecular events. The cis dependence of
the decrease of the amount of antisense RNA on transcription
of the sense RNA has, however, not been shown previously.
Mechanistically, the results shown here suggest that the crucial
factor in convergent transcription is the movement of the RNA
polymerase towards or through the pIII promoter region.
Three possibilities can be entertained: (i) promoter occlusion, (ii) colliding polymerases, and (iii) transcription-induced
changes in DNA supercoiling affecting initiation of transcription from pIII. Promoter occlusion can be ruled out since the
distance between the promoters pII and pIII suggests that
simultaneous binding should be possible. The second model is
not ruled out experimentally but is less likely since initiation
from pIII is approximately as frequent as initiation from pII
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
FIG. 5. Effect of novobiocin on accumulation of RNAII and RNAIII in the presence or absence of transcription from promoters pII and pIII. The autoradiograms
of the Northern blots probed for either RNAIII (A) or RNAII (B) are shown. Samples were taken at indicated times after novobiocin addition or without novobiocin
(D). RNAIIF, major band of full-length, readthrough RNAII (360 nt; the two or three minor bands above result from incomplete termination and are included in the
calculation); RNAIIT, terminated RNAII (260 nt). Plasmids used and promoters present are indicated. Reprobing to correct loading errors was performed as described
in Materials and Methods (data not shown).
7022
BRANTL AND WAGNER
J. BACTERIOL.
TABLE 4. Relative amounts of RNAIII and RNAII
after addition of novobiocina
Time after addition
of novobiocin
(min)
0
10
20
30
40
50
60
D
Relative amt of:
RNAIII
RNAII [% termination]
pGKIII
pGTW
pGTW
pGT7
1.03
1.23
2.03
2.03
2.03
2.53
1.53
1.03
0.33
0.23
0.13
0.073
0.0753
0.0753
0.0753
0.233
1.0y (60.3)
2.0y (13.6)
3.0y (7.8)
2.6y (9.3)
1.7y (13.4)
1.3y (11.0)
1.1y (13.6)
0.5y (42.0)
3.0y (17.6)
6.0y (25.0)
6.3y (23.7)
4.5y (22.4)
4.5y (18.0)
3.6y (20.4)
2.7y (17.1)
0.5y (24.0)
(supported by recent analysis of pII and pIII-lacZ transcriptional fusions [7]), and, therefore, a hypothetical collision between RNA polymerases on the same DNA template should
affect the levels of both transcripts. In addition, the concentrations of sense and antisense RNAs in plasmid-carrying B. subtilis cells and the half-lives of the RNAs suggest that RNA
polymerases initiate at either of the two promoters for only a
short fraction of the generation time. Thus, it appears unlikely
that simultaneous initiation events occur on the same DNA
template at a significant frequency. In contrast, it is probable
that RNAII transcription, by passing through the pIII region,
affects local DNA supercoiling. According to the twin supercoiled domain model for transcription (25), regions of positive
and negative supercoils are generated ahead of and behind, respectively, the transcribing polymerase. Such changes in DNA
supercoiling affect many promoters (e.g., 5, 18, 29). Antibiotics
that target DNA gyrase, like novobiocin, therefore often alter
promoter activity (18). The effect of novobiocin shown in Fig.
5 and Table 4 indicates that pIII is supercoiling sensitive. Thus,
we suggest that transcription-induced changes in DNA supercoiling may be responsible for the convergent transcription
effect in plasmid pIP501. A variant of this model is that supercoiling per se is not the ultimate cause of a decrease in pIII
activity, but that supercoiling-induced topological changes,
e.g., cruciform extrusion (e.g., 6), at one of several inverted
repeat sequences in or near the pIII promoter region (8) may
decrease initiation frequency. More work will be required to
elucidate the details of the mechanism.
Irrespective of the molecular mechanism by which convergent transcription occurs, the results presented here suggest a
way by which plasmid pIP501 can use the concerted action of
RNAIII and the CopR protein to efficiently regulate its copy
number. RNAIII alone is able to correct increases in copy
number: At higher plasmid concentrations, more RNAIII is
synthesized, which in turn increases transcriptional attenuation
of RNAII, thus decreasing the replication frequency. In contrast, fortuitous copy number decreases cannot rapidly be corrected by RNAIII, since its long half-life (13) will result in a
disproportionately high concentration of the inhibitor, which
threatens to yield a replication frequency inappropriately low
for the current copy number. CopR, due to its dual effect indicated above, acts in a complementary fashion to correct
downward fluctuations in plasmid concentration: decreased
synthesis of CopR will derepress the pII promoter. This single
event of derepression has two consequences: (i) increased
transcription of RNAII (12) and (ii) convergent transcription,
which decreases transcription of RNAIII. Both effects increase
RepR synthesis and, consequently, the replication frequency:
i.e., the molecular event of derepression of the pII promoter
works as an amplifier. The dual effect of CopR, together with
the previously recognized importance of the antisense RNA
RNAIII, suggests the model for pIP501 copy number control
shown in Fig. 7. Based on the concerted action of these two
regulatory elements, the stable inheritance of pIP501 can be
accounted for. Plasmids pSM19035 (2) and pAMb1 (14), which
share the overall organization of pIP501 (inc18 plasmid family
[8]) as well as the same mode of regulation (e.g., 22, 23), may
employ the same strategy to secure their stable maintenance.
Several other plasmids carry both antisense RNAs and repressor proteins as copy number control genes, e.g., plasmids
R1 (26) and pLS1 (15). Functional analysis of CopB of plasmid
R1 indicated that, although it prevents convergent transcription in a manner similar to that of CopR of pIP501, its primary
role is that of a rescue device at dangerously low copy numbers
and/or after conjugal transfer of R1 (24). During normal
steady-state conditions, the short half-life of CopA (28a) is
sufficient to cope with copy number deviations. In the case of
pLS1, CopG and the antisense RNA (RNAII) act independently, and the combined action of both elements is required
for copy number regulation (16). So far, no half-life has been
determined for RNAII (17), nor is it clear whether CopG is
required for stable maintenance. RNAII concentrations were
reported to correlate approximately with the copy numbers of
TABLE 5. Relative amounts of RNAIII408 (half-life mutant)
synthesized after addition of novobiocina
Relative amt of RNAIII408
Time after addition
of novobiocin
(min)
Plasmid pGK408/
rnaIII only
Plasmid pGT408/
rnaIII and rnaII
0
10
20
30
40
50
60
1.03
1.83
2.53
4.23
5.53
5.03
5.03
1.03
1.03
0.83
0.63
0.63
0.33
0.53
a
Calculation of relative amounts of RNAIII was done as for Table 2. Figure
6 shows the autoradiogram of the corresponding gel.
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
a
Calculation of relative amounts of RNAII and RNAIII and determination of
percent termination were done as for Table 2. Note that plasmid pGT7 expresses
only RNAII, whereas plasmid pGTW expresses both RNAII and RNAIII. x,
corrected hybridization signal at 0 min for RNAIII (pGTW); y, corrected hybridization signal at 0 min for RNAII (pGTW); coefficients indicate the corresponding increase or decrease. Figure 5 shows the autoradiogram of the corresponding gel.
FIG. 6. Effect of novobiocin on accumulation of RNAIII408 (half-life mutant) in the presence or absence of transcription from promoters pII and pIII.
The Northern blot is shown. Samples were taken at the indicated times after
novobiocin addition or without novobiocin (D). Hybridization, reprobing for 5S
rRNA, and calculations were done as described in Materials and Methods.
Plasmids and promoters are indicated.
VOL. 179, 1997
DUAL FUNCTION OF CopR
7023
the corresponding derivatives: copG-deficient plasmids replicating at a fivefold-higher copy number produced threefold
more RNAII than the wild-type plasmid (17). This may indicate that CopG does not significantly affect antisense RNA
levels by convergent transcription. A similar mode of regulation can be inferred for the other plasmids of the pLS1 family,
such as pE194, pADB201, etc. (16).
In summary, we can conclude that the inc18 plasmids represented by pIP501—in contrast to other antisense RNA-regulated plasmids—may have evolved a specific strategy to ensure stable plasmid maintenance that employs the concerted
action of an unusually stable antisense RNA and a transcriptional repressor protein with a dual function.
ACKNOWLEDGMENTS
We thank Eckhard Birch-Hirschfeld, Institut für Virologie, FriedrichSchiller-Universität Jena, for synthesizing the oligodeoxyribonucleotides. The excellent technical assistance of Bärbel Ukena is gratefully
acknowledged.
This work was supported by grant BR 1552/2-1 from Deutsche
Forschungsgemeinschaft (to S.B.).
REFERENCES
1. Anagnostopoulos, C., and J. Spizizen. 1961. Requirements for transformation in Bacillus subtilis. J. Bacteriol. 81:741–746.
2. Behnke, D., H. Malke, M. Hartmann, and F. Walter. 1979. Post-transformational rearrangement of an in vitro reconstructed group-A streptococcal
erythromycin resistance plasmid. Plasmid 2:605–616.
3. Behnke, D., M. S. Gilmore, and J. J. Ferretti. 1981. Plasmid pGB301, a new
multiple resistance streptococcal cloning vehicle and its use in cloning of a
gentamicin/kanamycin resistance determinant. Mol. Gen. Genet. 182:414–421.
4. Blomberg, P., E. G. H. Wagner, and K. Nordström. 1990. Control of replication of plasmid R1: the duplex between the antisense RNA, CopA, and its
target, CopT, is processed specifically in vivo and in vitro by RNaseIII.
EMBO J. 9:2331–2340.
5. Bowater, R. P., D. Chen, and D. M. J. Lilley. 1994. Modulation of tyrT
promoter activity by template supercoiling in vivo. EMBO J. 13:5647–5655.
6. Bowater, R. P., D. Chen, and D. M. Lilley. 1994. Elevated unconstrained
supercoiling of plasmid DNA generated by transcription and translation of
the tetracycline resistance gene in eubacteria. Biochemistry 33:9266–9275.
7. Brantl, S. 1994. The CopR protein of plasmid pIP501 acts as transcriptional
repressor at the essential repR promoter. Mol. Microbiol. 14:473–483.
7a.Brantl, S. Unpublished data.
8. Brantl, S., D. Behnke, and J. C. Alonso. 1990. Molecular analysis of the
replication region of the conjugative Streptococcus agalactiae plasmid pIP501
in Bacillus subtilis. Comparison with plasmids pAMb1 and pSM19035. Nucleic Acids Res. 18:4783–4790.
9. Brantl, S., and D. Behnke. 1992. Copy number control of the streptococcal
plasmid pIP501 occurs at three levels. Nucleic Acids Res. 20:395–400.
10. Brantl, S., and D. Behnke. 1992. The amount of RepR protein determines
the copy number of plasmid pIP501 in Bacillus subtilis. J. Bacteriol. 174:
5474–5478.
11. Brantl, S., E. Birch-Hirschfeld, and D. Behnke. 1993. RepR protein expression on plasmid pIP501 is controlled by an antisense RNA-mediated transcription attenuation mechanism. J. Bacteriol. 175:4052–4061.
12. Brantl, S., and E. G. H. Wagner. 1994. Antisense RNA-mediated transcriptional attenuation occurs faster than stable antisense/target RNA pairing: an
in vitro study of plasmid pIP501. EMBO J. 13:3599–3607.
13. Brantl, S., and E. G. H. Wagner. 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.
14. Clewell, D. B., Y. Yagi, G. M. Dunny, and S. K. Schultz. 1974. Characterization of three plasmid deoxyribonucleic acid molecules in a strain of Streptococcus faecalis: identification of a plasmid determining erythromycin resistance. J. Bacteriol. 117:283–289.
15. del Solar, G., and M. Espinosa. 1992. The copy number of plasmid pSL1 is
regulated by two trans-acting plasmid products: the antisense RNAII and the
repressor protein, RepA. Mol. Microbiol. 6:83–94.
16. del Solar, G., P. Acebo, and M. Espinosa. 1995. Replication control of
plasmid pSL1: efficient regulation of plasmid copy number is exerted by the
combined action of two plasmid components, CopG and RNAII. Mol. Microbiol. 18:913–924.
17. del Solar, G., P. Acebo, and M. Espinosa. 1997. Replication control of
plasmid pLS1: the antisense RNAII and the compact rnaII region are involved in translational regulation of the initiator RepB synthesis. Mol. Microbiol. 23:95–108.
18. Free, A., and C. J. Dorman. 1994. Escherichia coli tyrT gene transcription is
sensitive to DNA supercoiling in its native chromosomal context: effect of
DNA topoisomerase IV overexpression on tyrT promoter. Mol. Microbiol.
14:151–161.
19. Horodniceanu, T., D. H. Bouanchaud, G. Bieth, and Y. A. Chabbert. 1976.
R-plasmids in Streptococcus agalactiae (group B). Antimicrob. Agents Chemother. 10:795–801.
20. Kawamura, F., and R. H. Doi. 1984. Construction of a Bacillus subtilis double
mutant deficient in extracellular alkaline and neutral proteases. J. Bacteriol.
160:442–444.
21. Kok, J., J. M. B. van der Vossen, and G. Venema. 1984. Construction of
plasmid cloning vectors for lactic streptococci which also replicate in Bacillus
subtilis and Escherichia coli. Appl. Environ. Microbiol. 48:726–731.
22. Le Chatelier, E., S. D. Ehrlich, and L. Janniere. 1994. The pAMb1 Cop
repressor regulates plasmid copy number by controlling transcription of the
repE gene. Mol. Microbiol. 14:463–471.
23. Le Chatelier, E., S. D. Ehrlich, and L. Janniere. 1996. Countertranscriptdriven attenuation system of the pAMb1 repE gene. Mol. Microbiol. 20:
1099–1112.
24. Light, J., E. Riise, and S. Molin. 1985. Transcription and its regulation in the
basic replicon region of plasmid R1. Mol. Gen. Genet. 198:503–508.
25. Liu, L. F., and J. C. Wang. 1987. Supercoiling of the DNA template during
transcription. Proc. Natl. Acad. Sci. USA 84:7024–7027.
26. Nordström, K., S. Molin, and J. Light. 1984. Control of replication of
bacterial plasmids: genetics, molecular biology and physiology of the plasmid
R1 system. Plasmid 12:71–90.
27. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009
FIG. 7. Working model of copy number control of pIP501. The circuitry of replication control and the involvement of the genes and gene products are explained
in more detail in the text. Filled boxes, promoters; open rectangles, open reading frames; thick arrows, RNAs; hatched or checkered boxes, proteins; oriR, replication
origin; filled 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 (presence
of CopR) permits increased RNAIII transcription as described in this communication.
7024
BRANTL AND WAGNER
laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, N.Y.
28. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with
chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467.
28a.Söderbom, F. Unpublished data.
29. Spirito, F., and L. Bossi. 1996. Long-distance effect of downstream transcription on activity of the supercoiling-sensitive leu-500 promoter in a topA
J. BACTERIOL.
mutant of Salmonella typhimurium. J. Bacteriol. 178:7129–7137.
30. Steiner, K., and H. Malke. 1995. Transcription termination of the streptokinase gene of Streptococcus equisimilis H46A: bidirectionality and efficiency in
homologous and heterologous hosts. Mol. Gen. Genet. 246:374–380.
31. Stougaard, P., J. Light, and S. Molin. 1982. Convergent transcription interferes with expression of the copy number control gene, CopA, from plasmid
R1. EMBO J. 1:323–328.
Downloaded from jb.asm.org at AMERICAN SCIENTIFIC PUBLICATIO on July 3, 2009