Mol Gen Genet (1992) 234:105-112
OIGR3
© Springer-Verlag 1992
In vitro and in vivo analysis of transcription
within the replication region of plasmid piP501
Sabine Brantl 1, Beatriz Nuez 2, and Detlev Behnke 1
1 Institute for Molecular Biology, Beutenbergstr. 1I, 0-6900 Jena, FRG
2 Centro de Biologia Molecular, CSIC UAM, Universidad Aut6noma de Madrid, Canto Blanco, E-28049 Madrid, Spain
Received December 9, 1991
Summary. Derivatives of the conjugative streptococcal
plasmid piP501 replicate stably in Bacillus subtilis. The
region essential for replication of piP501 has been
narrowed down to a 2.2 kb D N A segment, the sequence
of which has been determined. This region comprises two
genes, copR and repR, proposed to be involved in copy
control and replication. By in vitro and in vivo transcriptional analysis we characterized three active promoters,
P~, PH and Pm within this region. A putative fourth promoter (P~v) was neither active in vitro nor in vivo. We
showed that copR is transcribed from promoter p~ while
the repR gene is transcribed from promoter PlI located
just downstream of copR. The Pn transcript encompasses
a 329 nucleotide (nt) long leader sequence. A counter
transcript that was complementary to a major part of this
leader was found to originate from a third promoter PHI.
The secondary structure of the counter transcript revealed several stem-loop regions. A regulatory function
for this antisense R N A in the control of repR expression
is proposed. Comparative analysis of the replication regions of pAMI3 1 and pSM19035 suggested a similar
organization of transcriptional units, suggesting that an
antisense R N A is produced by these plasmids also.
Key words: Bacillus subtilis - Plasmid replication region
- Transcriptional analysis - Promoter - Terminator
Introduction
Most of the plasmids used today as cloning vectors in
Bacillus subtilis have been derived from several antibiotic
resistance plasmids originally isolated from Staphylococcus aureus. Plasmids like pC194, pE194, and pUB110 or
pT181 are prototype plasmids, which provided the basis
for the construction of a large variety of cloning vectors
(Gruss and Ehrlich 1989; Alonso 1989; Bron 1990). All
Correspondence to: D. Behnke
of these plasmids have been found to replicate via a
rolling circle mechanism known from the single-stranded
D N A (ssDNA) bacteriophages of Escherichia coli. This
mode of replication involves the production of singlestranded replication intermediates, which appear to be
responsible for the segregational instability of many of
these vectors as well as for the structural instability noted c'
for recombinant plasmids carrying larger inserts (Bron
and Luxen 1985; Ehrlich et al. 1986).
A different family of cloning vectors for gram-positive
bacteria has been developed in our laboratory. These
vectors, which were originally constructed for use in
streptococci, were derived either from the large antibiotic
resistance plasmid pSM19035 isolated from Streptococcus pyogenes (Behnke et al. 1979; Behnke and Ferretti
1980) or the conjugative plasmid piP501 isolated from
Streptococcus agalactiae (Horodniceanu et al. 1976;
Behnke et al. 1981). Plasmid pSM19035 has a unique
structural feature. It carries large inverted repeat sequences that comprise approximately 90 % of the plasmid
genome (Behnke et al. 1980). Surprisingly, this structural
feature was stably maintained, as were even drastically
shortened derivatives of pSM19035 (Behnke and Klaus
1983). This suggested a well regulated mode of replication. Deletional analysis of both pSM19035 and piP501
allowed the sequences essential for replication to be
narrowed down to approximately 2 kb (Behnke and
Gilmore 1981; Behnke and Klaus 1983). Subsequent
sequence analysis revealed a high degree of identity between the replication regions of these two plasmids
(Brantl et al. 1989, 1990; Sorokin and Khazak 1989),
which is also shared by a third streptococcal plasmid,
pAMI31 (Clewell et al. 1974), the replication region of
which has recently been sequenced (Swinfield et al. 1990).
Plasmid pAM[31 has recently been suggested to replicate
via a theta-type mechanism as it is known for most
plasmids of gram-negative bacteria (Janni6re et al. 1990;
Bruand et al. 1991). This agrees with our own observations for pSM19035 or piP501 (S. Brantl and J. Alonso,
unpublished data). Vectors derived from these streptococcal plasmids replicate well in B. subtilis and exhibit
106
high segregational and structural stability. They may,
therefore, provide the basis for a new generation of stable
cloning vectors for B. subtilis and other gram-positive
bacteria.
In this communication we extend our analysis of the
replication region of plasmid piP501 and report on the
topological organization of transcriptional units within
this region. Among the three promoters identified in
vitro and in vivo, one directed synthesis of a non-coding
counter transcript, which is likely to regulate the expression of repR by an antisense mechanism.
Materials and methods
Bacterial strains andplasmids. The plasmids used in this
study have previously been described (Brantl et al. 1990).
B. subtilis strain DB104 his, nprR2, nprE18, aprA3 (Kawamura and Doi 1984) served as the standard host for all
plasmids. Subcloning and mutagenesis experiments were
carried out in E. coli TG2 (Sambrook et al. 1989).
DNA preparation, manipulation and sequencin 9. The isolation of plasmid D N A from B. subtilis has previously
been reported (Brantl et al. 1990). DNA manipulations
such as restriction enzyme cleavage, ligation, filling-in
reactions with Klenow fragment of DNA polymerase I
or end-labelling with polynucleotide kinase were either
carried out under the conditions specified by the manufacturer or according to standard protocols (Sambrook
et al. 1989). D N A sequencing followed the method of
Sanger et al. (1977). For size references included in nuclease S1 mapping experiments fragments of known
sequence were also chemically treated according to Maxam and Gilbert (1977).
In vitro transcription. The D N A fragments used as templates for in vitro transcription experiments are indicated
in Fig. 1A. B. subtilis cyA RNA polymerase was isolated
as described (Sogo et al. 1979). The protocol for in vitro
transcription has previously been published (Mellado et
al. 1986). Transcripts labelled with [a32p]UTP were
separated on 4% or 6% denaturing polyacrylamide gels
under standard conditions. Either end-labelled HindIII
fragments of bacteriophage 029 DNA or purine sequence
reactions of known D N A fragments were included as size
markers on each gel.
RNA isolation. Total R N A was prepared from B. subtilis
DB104(pGB354) for nuclease S1 mapping of promoters
Pn, Pin, Piv and for mapping of terminators of Pli and pro.
To map promoter PI, total R N A was prepared from
B. subtilis DB104 carrying the high copy number plP501
derivative pCOP8 (S. Brantl and D. Behnke 1992). Cultures (15 ml) were grown to a n O D 6 o o of 1.0. After the
addition of 220 gl of 2.5 M sodium azide, cells were
rapidly centrifuged on frozen 0.7% NaC1. The pellet was
dissolved in 1 ml of lysis buffer I (10% sucrose, 0.1 M
NaC1, 0.1 M EDTA, pH 8.0, containing 1 mg/ml oflysozyme) and incubated for 5 min at 37° C. Subsequently,
3 ml of lysis buffer II (10 mM TRIS-HC1, pH 7.4, 1 mM
EDTA, 100 mM sodium acetate, pH 7.0, 2% SDS) were
added and three extractions were carried out with hot
phenol (65 ° C) followed by two extractions with cold
phenol. After removal of phenol by two chloroform
extractions the R N A was precipitated with ethanol and
redissolved at a concentration of 30 mg/ml in distilled
water treated with diethylpyrocarbonate.
Nuclease S1 mapping. For nuclease S1 mapping the following D N A fragments were isolated: promoter Pl,
HincII-HindIII fragment (753 bp); promoter Pll,
KpnI-TaqI fragment (400 bp); promoter Pro, TaqI-PvuII
fragment (215 bp); promoter P~v, Hinfl fragment (243
bp). Fragments were labelled at their 5' ends with
[y32p]ATP and polynucleotide kinase. For mapping of
transcriptional terminators the following fragments were
labelled at their 3' ends by filling-in 5' overhanging sticky
ends: terminator of PIl, TaqI-PvuII fragment (215 bp);
terminator of PHI, KpnI TaqI fragment (400 bp). The
positions of these fragments within the replication region
of piP501 are shown in Fig. 1A. Strand separation,
hybridization and digestion with nuclease S1 followed
exactly the protocol outlined by Barthelemy et al. (1987).
Protected fragments were electrophoresed on 6% denaturing polyacrylamide gels with Maxam/Gilbert or Sanger sequencing reactions as size standards.
Construction of the Pzv deletion mutant. A fragment of
plasmid pUC118-F (Brantl et al. 1990) that comprised
the repR Shine-Dalgarno (SD) region and downstream
sequences was amplified by the polymerase chain
reaction (PCR) using the reverse sequencing primer
and
the oligonucleotide:
5' GGATAACAATTAATCTAGAAAGGAGAACAG3'. As a result, a 5'
flanking restriction site for Xbal was introduced. The
amplified fragment was joined to a D N A fragment carrying the DNA sequences of the piP501 replication region
upstream of the putative promoter P~v. The deletion of
promoter P~v was confirmed by DNA sequencing of the
final construct on plasmid pPR14. For detailed sequence
information see Fig. 4.
RNA secondary structure calculation. RNA secondary
structures were calculated with the PC-Gene program on
an IBM-PC. This work was done at the ZMBH in
Heidelberg.
Results
In vitro run-off transcription
A schematic representation of the replication region of
plasmid piP501 is shown in Fig. 1A. Computer-based
analysis of the DNA sequence (Brantl et al. 1990;
S. Brantl and D. Behnke 1992) encompassing this
region revealed the presence of four putative promoter
sequences. Three of them (Pl~, Pill, Plv) are located within
the intergenic region between copR and repR, while promoter Pl precedes the copR open reading frame. The
DNA sequences of all four promoters as well as the
107
A
Hit
H K Hf
II
L
L. W
-765
M Dr Dd
I
1
[
I
I
I
Hf
Pv R D r
I
l
I
I
HfM
II
I
613
RS
E
II
J
860 2269
ID
[
repR
copR
plV
pll
P. nl
pl
~'.nl
m
-- Lengthsof transcripts
ILJIm
4
H II
Dd M
400
]
I
T
pill
(in nucleotides)
K
I
2
I
3
K
i
,S
J
80
580,460, 270,240,
140
• Pv
i
f
460,330, 270, 240,
140
•
4
Hf
Dr
Hf
I
Ji
r
190,
140
.Hf
Hf
I
[
.~
400,270,240,140
4
Dr
M
1
190,140
M
pl
IP
I
Dd
Dd
I
I
240, 330,140
117
T
Pv
i
i
.J
10
B
65
Hf
Hf
I
[
--
promoter I
-10
-35
CAAGTGATTAATCACTTGTTTATTAAGATATTAAAAGCTATAA~FrAAATAAAGCGTG
premoter II
TAATGCACGAAATCATTGCTTATTTTTTTAAAAAGCGATATACTAGATATAACGAAAC
promoter III
TATCTTCTTAAAGACTTGAAATCCCCTCAAAAACCCGATATAATGGGTTTACAGATAT
promoter IV
TGTTGTGATTCAACTTTGATCGTAGCTTCTAACTAATTAATTTTCGTAAGAAAGGAGA
Fig. 1A and B. In vitro run-off transcription experiments with BaciL
lus subtilis (yg RNA polymerase. A The upper part presents a
schematic outline of the HincII-EeoRI fragment encompassing the
replication region of plasmid piP501. The - 3 5 and - 10 boxes of
the putative promoters are shown as filled and open rectangles,
respectively. Proposed directions of transcription are indicated by
arrows. The lower part shows the individual fragments used as
templates together with the lengths and positions of the transcripts
(arrows) obtained. Numbers on the left refer to the autoradiographs presented in Fig. 2. 1, HincII-KpnI fragment (nucleotides
[nt] - 7 6 5 to - 1); 2, KpnI-SnaBI fragment (nt 1 to 860); 3, KpnIPvuII fragment (nt 1 to 613); 4, DraI-RsaI fragment (nt 265 to 668);
5, MboII fragments (nt 223 to 517 and 517 to 805); 6, Hinfl
frangment (nt 67 to 558); 7, HinfI fragment cleaved by DraI (nt 67
to 265 and 265 to 558); 8, DdeI fragment (nt 345 to 533); 9,
TaqI-PvuII fragment (nt 399 to 613); 10, HinfI fragment (nt 558 to
801). B DNA sequence of the putative promoters identified by
computer analysis within the replication region of piP501. Abbreviations : repR and copR, genes involved in replication and copy
control of piP501; Dd, DdeI; Dr, DraI; E, EcoRI; HII, HincII;
HIII, HindIII; Hf, HiM'I; K, KpnI; M, MboII; Pv, PvuII; R, RsaI;
Sn, SnaBI; T, TaqI. Nucleotides are numbered according to the
originally published sequence of ,the piP501 replication region
(Brantl et al. 1990)
assumed directions of transcription are also summarized
in Fig. 1A, B.
To determine which of these sequences in fact constitute active promoters, in vitro transcription experiments were carried out with ~A R N A polymerase isolated
from B. subtilis. To this end various fragments of the
replication region were prepared that included one, two
or three of the putative promoters (Fig. 1A). A single
transcript of approximately 80 nucleotides (nt) in length
was obtained when the HincII-KpnI fragment encom-
passing the region upstream of copR was used as a template (Fig. 2, lane 1). The length of this transcript was
consistent with the location of promoter p~. In contrast,
most of the fragments that extended through part or all
of the intergenic region between copR and repR gave rise
to multiple transcripts. Stepwise reduction of the length
of the fragments used as templates, together with size
estimates of the RNAs obtained, allowed the different
transcripts to be assigned to promoters Pn or Pro. In
addition, a DraI site located within the spacer region of
108
460~
580-460--
230~
190--
2 70
240-140~I
~
11 7
140--
,:
80~
1
2
I
3
4
5
A
6
7
8
A
8
B
9
Fig. 2. Autoradiographs of in vitro run-off transcription experiments with Bacillus subtilis eA RNA polymerase. The DNA fragments of the piP501 replication region that were used as templates
are shown in Fig. 1A. Lane numbers of the in vitro transcription
reactions correlate with the respective fragment numbering in Fig.
1A (left-hand panel). Transcripts were separated on either 4 % or 6 %
denaturing polyacrylamide gels. The size standards used were either
radioactively labelled HindIII fragments of bacteriophage q~29
(lanes A) or purine/pyrimidine reactions of a DNA fragment of
known sequence (lane B). Reactions with inactive templates (number 10 in Fig. 1A) are not shown. Numbers indicate size estimates
of the run-off transcripts (nts)
PII (Fig. 1A) was used to inactivate this promoter. This
cleavage led to the expected loss of the transcript assigned to Pn that was observed with the same template
(HinfI fragment) not treated with DraI (Fig. 2, lanes 6
and 7). Finally, the use of two very short fragments
covering only Pm or Pm and Piv allowed the unequivocal
demonstration of the activity of promoter Pm (Fig. 2,
lanes 8 and 9). In all of the in vitro transcription experiments we failed, however, to observe any transcript that
might have originated from promoter P~v- A HinfI fragment carrying only P~v was completely inactive as a
template (Fig. 1A). These results suggested that Piv may
not be the promoter that directs transcription of the repR
gene as previously assumed (Brand et al. 1990). In fact,
as will be shown below, this putative promoter sequence
can be deleted without any effect on piP501 replication.
Full-length transcripts were always found to arise
from both promoters Pl] and Phi. Each of the two promoters, however, also gave rise to shorter transcripts that
terminated at distinct sites within the templates. Examination of the D N A sequences around these sites revealed
in all cases the presence of stem-loop structures resembling rho-independent terminators.
In vivo analysis of transcripts
Nuclease S1 mapping experiments were performed to
determine whether the promoters detected in vitro were
also active in vivo. A 214 bp TaqI-PvuII fragment was
isolated to probe for the transcriptional start site of
promoter Pro. The separated single strands of this fragment were hybridized to total R N A from B. subtilis
DB104 carrying plasmid pGB354. As shown in Fig. 3C
one strong and three weaker signals were obtained that
corresponded in size (59-63 nt) to the length of the
protected fragment expected to arise from hybridization
with an overlapping Pm transcript. Promoter Pin was,
therefore, clearly active in vivo. The in vitro transcriptional analysis did not, however, allow a reliable conclusion with regard to the length of the PIH transcript.
Thus, a KpnI-TaqI fragment was isolated that spanned
109
G
A
C
T
G/A
...B
T
C
A
I_/- i
G
G/A
m
O
O
qm~
-
....
G)
-4
msm
m
~mm
-10
AAAAGCTATAATTTAAATAAAGCGTGAA
-95
-10
TTTAAAAAGCGATATACTAGATATAACG
270
OOmto
IIIIIII
R N A II
RNA I
B
A
G/A C/T
G
A
C
T
O
"z
C9
E¢ j Oi
~
o
o
<
(.9
I-L9
I.-
J
o
L9
-10
AAAACCCGATATAATGGGTTTACAGATAT
480
.: . } e l O
RNA III
AACTGGTCGTGGCTCTTTTTTTGTATTC
oeeooooo
306
lb
I:,ee¢¢¢~
RNA III
C
the two termination points found in vitro for pro-directed
transcripts. Hybridization of the 3"-labelled fragment
(filled-in TaqI site) to total R N A from DB104(pGB354)
and subsequent nuclease S1 treatment gave rise to a
single group of protected fragments ranging in size between 74 and 81 nt (Fig. 3D). This size corresponded well
with the shortest (140 nt) Pin transcript observed in the
in vitro transcription experiments. Since no other signals
were detectable, termination at this site appeared to be
efficient in vivo. In the following the Iength of the Pn~
transcripts is defined as 136 nt, i.e. extending from the
strongest start signal (A at position 460) to the last strong
termination signal (T at position 325).
D
Fig. 3A-D. Nuclease S1 mapping of
transcriptional start sites of promoters Pl (A), Pn (B) and promoter
Pm (C) and of the termination site of
RNAIII (D). The lower panels show
the DNA sequences of the respective
parts of the piP501 replication region.
Arrows denote the direction of transcription. Open andfilled circles refer
to weak and strong signals, respectively, seen in the nuclease S1 experiments. Dideoxynucleotide sequencing
reactions or purine and/or pyrimidine
reactions according to Maxam and
Gilbert (1977) were used for size determinations of the protected fragments. The Maxam/Gilbert reactions
were done on the same fragments
that were used for nuclease S1 mapping
To analyse the transcripts directed in vivo by promoter Pn, the same 400 bp KpnI-TaqI fragment as used
above was 5'-labelled at the TaqI site and hybridized to
total R N A from DB104 (pGB354). After nuclease S1
treatment, two signals were detectable (Fig. 3B) corresponding to protected fragments of 113 and 114 nt in
length, thus confirming the position and activity of promoter PI[ in vivo. As noted for Pro, no reliable conclusion
on the length of the transcript originating from Pu could
be derived from the in vitro experiments. Therefore, a
3"-labelled 214 bp TaqI-PvuII fragment was isolated that
covered the termination point of the short p . transcript
seen in vitro. The nuclease S1 mapping experiment re-
110
A
Hf
-35
- 10
Pv
5" "i ~ ~ ~ i ' r G T T G T G A T T C A A C ~ T A G C T T C T A A C T A A ~ C G T A A G A A A G G A G A A C A G C T G A A T G A A T
547
559
~. t
601
SD
620
RepR
I
t
/
B
B
5" t I I
5~,7
,t
Xb
Pv
GTTGTGGGGATCCTCTAGAAAGGAGAACAGCTGAATGAAT
559
601
620
RepR
Fig. 4A and B. Comparison of the wild-type sequence (A) of the
region upstream of repR with that of the deletion mutant present
on plasmidpPR14 (B). The predicted -35 and - 10 regions of the
putative promoter Piv are boxed. The Shine-Dalgarno (SD)-
vealed a strong signal corresponding to full protection of
the TaqI-PvuII fragment, thus suggesting that pH-directed transcription extended beyond the PvuII site. Only a
very faint signal was noted at a position that correlated
with the upstream termination site observed in vitro
(data not shown). Promoter p., therefore, appeared to be
the promoter controlling transcription of the repR gene.
This conclusion was further supported by the nuclease S 1
mapping experiments designed to map promoter P~v. On
using a 243 bp long Hinfl fragment, we observed only
complete protection (data not shown), confirming a start
point for the repR mRNA upstream of P~vFinally, we were able to demonstrate in vivo the activity of promoter p~, assumed to direct transcription of the
copR gene. Nuclease S1 mapping experiments were carried out with a 753 bp HincII-HindIII fragment. For the
S 1 mapping, the fragment was 5'-labelled at the HindIII
site and hybridized to total RNA from B. subtilis DB104
carrying plasmid pCOP8. As shown in Fig. 3A protected
fragments ranging in size from 62-68 nt were obtained,
thus confirming the location and in vivo activity of promoter Pv
Deletion of promoter Piv
The failure to detect in vitro or in vivo transcripts originating from promoter P~v did not exclude the possibility
that this promoter is active at a low but essential level.
Therefore, a deletion was constructed that completely
removed promoter Plv. The DNA sequences of the wildtype Plv region and that of the deletion mutant are shown
in Fig. 4. The mutant plasmid pPR14, harbouring the
above deletion, was still able to replicate in B. subtilis and
no effect on the copy number was noted. We therefore
conclude that the proposed promoter P~v is not active
and is dispensable at least for piP501 replication in
B. subtilis.
Secondary structure analysis of RNAIII
Analysis of the nucleotide sequence of RNAIII, transcribed from Pn~, revealed neither an open reading frame
nor a structure resembling a translational initiation region. It therefore appeared likely that this non-coding
sequence and the translational start codon of repR are also indicated. Restriction sites: B, BarnHI; Hf, HinfI; Pv, PvuII; Xb,
XbaI. Numbering of the nucleotide positions refers to the original
sequence (Brantl et al. 1990)
O" O U
t~UO'd- C,
~UUA_
~oG
"C~.
_QO',3-CG5 ~
GO£,# .e'
aUUUc '
UA
AU
UA
GC
AU
U~j.C G /
A U
G C
/C G,,
U
U
"C O/
UG
GC
AU
AU
AU
GC
GU
U"U A"A
O A U" G
U"U A"A
A,A U"A
CG
5' A/
U A
XAUUAA"
/G Cx
"CCAAAUUUU
U 3'
5. Secondary structure of the antisense RNAIII identified
within the replication region of plasmid piP501
Fig.
RNA molecule served a regulatory function similar to
the antisense RNAs known to regulate the replication
initiation frequency in other plasmids (Novick et al.
1984; Novick 1987). Computer-based calculation of the
potential secondary structure of RNAIII, according to
Zuker and Stiegler (1981), revealed three stem-loop
structures, one of which was branched (Fig. 5), The free
energy of the folded RNAIII molecule was estimated to
be - 31.1 kcal/mol. It is interesting to point out that the
two terminal loops of the branched structure (Fig. 5)
were particularly GC-rich in an otherwise AT-rich environment. This was also true for the basic stem of the
branched structure, which is therefore expected to be
rather stable.
A comparison of the DNA sequence of the replication
region of piP501 with those of the two incompatible
plasmids pSM19035 and pAM[31 showed that the latter
plasmids contained putative promoters and terminators
equivalent to Pin of piP501 and its terminator. These are
likely to give rise to putative RNAIII transcripts of
similar length. The nucleotide sequence of the putative
RNAIII transcript from pSM19035 was highly homologous to that of piP501 (90.5% identity). However, only
67.5% similarity was found between the respective regions of piP501 and pAM~I. As a consequence, the
calculated secondary structure of the putative RNAIII of
pAM[~I differed significantly from that of the two other
plasmids.
111
Discussion
The topological organization of the replication region of
plasmid piP501 resembles to a large extent that of the
E. coli plasmid R1 (Masai and Arai 1988). The replication regions of both plasmids encompass coding regions
for two putative proteins, a smaller one, designated Cop,
which is involved in copy control and a larger one,
designated Rep, which is essential for initiation of replication (Masai and Arai 1988; Brantl et al. 1990). The
data presented in this communication extend the similarities between the two plasmids to the organization of
transcriptional units within the replication region. The
copR gene o f piP501 is transcribed from promoter p, the
activity of which could be clearly demonstrated both in
vitro and in vivo. In a previous publication, we assumed
that a putative promoter, Piv, located immediately upstream of the repR SD sequence controlled repR expression. However, we were not able to detect a transcript originating from Ply either in vitro or in vivo and
the putative promoter could be deleted without any apparent effect on piP501 replication. Transcription of
repR is instead directed by promoter Pn, located just
downstream of the copR coding sequence. Transcription
from promoter p . gives rise to a m R N A with a long
( ~ 329 nt) leader sequence preceding the region coding
for repR. Furthermore, we were able to detect a counter
transcript (RNAIII, 136nt) that originated from promoter P,I and that was complementary to a major part
of the repR m R N A leader region. It is, therefore, likely
that this short noncoding R N A serves a similar regulatory function to the copA antisense R N A of plasmid R1
(Stougaard et al. 1981; Persson et al. 1990a, b). In fact
deletional analysis of promoter PI,, and of the R N A I I I
region has recently provided evidence for a regulatory
role for this antisense R N A in copy control of piP501
(S. Brantl and D. Behnke 1992).
A comparison of the replication region of piP501 with
that of two related plasmids (pSM19035, Behnke et al.
1979; and pAM131, Clewell et al. 1974) that belong to the
same incompatibility group (incl8; Brantl et al. 1990)
revealed promoter sequences at positions very similar or
identical to promoters p . and Pm of piP501. It is, thus,
reasonable to assume that on these two plasmids the
transcriptional units are organized in the same way as on
piP501. As a consequence, replication ofpSM19035 and
pAM131 would also involve regulation by antisense
RNAs. The putative R N A I I I of plasmid pSM19035
showed extensive similarity ( > 9 0 % ) with R N A I I I of
piP501, while considerable less similarity (65%) was
noted in the case of pAM~ 1. Furthermore, the secondary
structure of the putative antisense R N A of pAM131 was
clearly different from that of the two other plasmids.
Since all of these plasmids are, however, incompatible, it
seems unlikely that the antisense RNAs are the components mediating the incompatibility. It seems more
probable that the trans-acting Rep protein may be the
limiting element for which the different plasmid species
compete. In this respect it is interesting to note that all
three plasmids share greater than 90 % similarity in their
respective rep genes both at the nucleotide and at the
amino acid level.
In a previous communication we reported loss of
autonomous replication as a consequence of the deletion
of 360 nt downstream of the KpnI site of piP501 (Brantl
et al. 1990, mutant plasmid p4). Resolution of the transcriptional units on piP501 now allows a more precise
interpretation of these data. The deletion described completely removed promoter PII, leading to loss of repR
expression rather than to impairment or inactivation of
the origin of replication. Interestingly, Swinfield et al.
(1990) reported that a similar deletion on plasmid
pAM131 (pMTL20CI3g) also impaired its ability to replicate. Examination of the nucleotide sequence of pAMI31
showed that the respective deletions inactivated the putative promoter PI, of pAM131 postulated from our comparative analysis.
Acknowledgements. We would like to thank M. Salas and F. Rojo
for experimental advice and J. Alonso for stimulating discussions.
The excellent technical assistance of Ina Poitz is gratefully acknowledged. This work was supported in part by the research exchange
programme between Germany and Spain and by grants from the
Max Planck Society (to D.B.), from the Direccion General de
Investigacion Cientifica y Tecnica (no. PB 87-0323 to M. Salas) and
an Institutional Grant from Fundacion Ramon Areces (to Centro
de Biologia Molecular). B.N. was a recipient of a predoctoral
fellowship from M.E.C.
References
Alonso JC (1989) DNA replication of plasmids from Gram-positive
bacteria in Bacillussubtilis. Plasmid pUB110 as a model system.
Microbiologia (SEM) 5: 5-12
Barthelemy I, Salas M, Mellado RP (1987) In vivo transcription of
bacteriophage q~29 DNA-transcription initiation sites. J Virol
60:874-879
Behnke D, Ferretti JJ (1980) Physical mapping of plasmid pDB 101 :
a potential vector plasmid for molecular cloning in streptococci.
Plasmid 4:130-138
Behnke D, Gilmore MS (1981) Location of antibiotic resistance
determinants, copy control and replication functions on the
double-selective streptococcal cloning vector pGB301. Mol Gen
Genet 184:115-120
Behnke D, Klaus S (1983) Double or triple sets of replication
functions as inverted and direct repeats on in vitro reconstructed
streptococcal MLS resistance plasmids. Z AUg Mikrobiol
23 : 539-547
Behnke D, Malke H, Hartmann M, Walter F (1979) Post-transcriptional rearrangement of an in vitro reconstructed group-A
streptococcal erythromycin resistance plasmid. Plasmid
2:605-616
Behnke D, Tomich PK, Clewell DB (1980) Electron microscopic
mapping of deletions on a streptococcal plasmid carrying extraordinarily long inverted repeats. Plasmid 4:139-142
Behnke D, Gilmore M, Ferretti JJ (1981) Plasmid pGB301, a new
multiple resistance streptococcal vehicle and its use in cloning
of a gentamycin/kanamycin resistance determinant. Mol Gen
Genet 182:41~421
Brantl S, Nowak A, Behnke D, Alonso JC (1989) Revision of
the nucleotide sequence of the Streptococcus pyogenes plasmid
pSM19035 repS gene. Nucleic Acids Res 17:10110
Brantl S, Behnke D, Alonso JC (1990) Molecular analysis of the
replication region of the conjugative plasmid piP501 of Streptococcus agalactiae in Bacillus subtilis. Comparative analysis
to plasmids pAM[31 and pSM19035. Nucleic Acids Res
18 : 47834790
Brantl S, Behnke D (1992) Copy number control of the streptococcal plasmid piP501 occurs at three levels. Nucleic Acids Res
20:395~400
112
Bron S (1990) Plasmids. In: Harwood CR, Cutting SM (eds) Molecular biology methods for Bacillus. John Wiley & Sons,
pp 75-138
Bron S, Luxen E (1985) Segregational instability of pUBll0derived recombinant plasmids in Bacillus subtilis. Plasmid
14:235-244
Bruand C, Ehrlich SD, Janni6re L (1991) Unidirectional theta
replication of the structurally stable Enterococcusfaecalis plasmid pAMI31. EMBO J 10:2171-2177
Clewell DB, Yagi Y, Dunny GM, Schultz SK (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
Ehrlich SD, Noirot P, Petit MA, Janni~re L, Michel B, te Riele H
(1986) Structural instability of Bacillus subtilis plasmids. In:
Setlow JK, Hollaender A (eds) Genetic engineering, vol. 8.
Plenum Publishing, New York, pp 71-83
Gruss A, Ehrlich SD (1989) The family of highly interrelated single-stranded deoxyribonucleic acid plasmids. Microbiol Rev
53 : 231-241
Horodniceanu T, Bouanchaud DH, Bieth G, Chabbert YA (1976)
R-plasmids in Streptococcus agalactiae (group B). Antimicrob
Agents Chemother 10:795-801
Janni6re L, Bruand C, Ehrlich SD (1990) Structurally stable Bacillus subtilis cloning vectors. Gene 87:53-61
Kawamura F, Doi RH (1984) Construction of Bacillus subtilis
double mutant deficient in extracellular alkaline and neutral
proteases. J Bacteriol 160: 442M44
Masai H, Arai K-I (1988) R1 plasmid replication in vitro, repA- and
dnaA-dependent initiation at oriR. In: Robb EM, Summers WC
(eds) DNA replication and mutagenesis. American Society for
Microbiology, Washington DC, pp 113-121
Maxam AM, Gilbert V (1977) A new method for sequencing DNA.
Proc Natl Acad Sci USA 74:560-564
Mellado RP, Barthelemy I, Salas M (1986) In vitro transcription of
bacteriophage q~29 DNA. Correlation between in vitro and in
vivo promoters. Nucleic Acids Res 14:4731~4741
Novick RP 0987) Plasmid incompatibility. Microbiol Rev
51 : 381-395
Novick RP, Adler GK, Projan SJ, Carleton S, Highlander S, Gruss
A, Khan SA, Iordanescu S (1984) Control of pT 181 replication.
I. The pT181 copy control function acts by inhibiting the synthesis of replication protein. EMBO J 3: 2399-2405
Persson C, Wagner EGH, Nordstr6m K (1990a) Control of replication of plasmid R1 : structures and sequences of the antisense
RNA, CopA, required for its binding to the target RNA, CopT.
EMBO J 9: 3767-3775
Persson C, Wagner EGH, Nordstr6m K (1990b) Control of replication of plasmid R1 : formation of an initial transient complex is
rate-limiting for antisense RNA-target RNA pairing. EMBO J
9:3777-3785
Sambrook J, Fritsch EF Maniatis T (1989) Molecular cloning.
A laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, New York
Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with
chain terminating inhibitors. Proc Natl Acad Sci USA
74: 5463-5467
Sogo JM, Inciarte MR, Corral J, Vinuela E, Salas M (1979) RNA
polymerase binding sites and transcription map of the DNA of
Bacillus subtilis phage q~29. J Mol Biol 127:411~436
Sorokin A, Khazak V (1989) Structure of pSM19035 replication
region and MLS resistance gene. In: Butler LO, Harwood C,
Moseley BEB (eds) Genetic transformation and expression.
Intercept, Andover, UK, pp 269-281
Stougaard P, Molin S, Nordstr6m (1981) RNAs involved in copy
number control and incompatibility of plasmid R1. Proc Natl
Acd Sci USA 78:6008-6012
Swinfield TJ, Oultram JD, Thompson DE, Brehm JK, Minton NP
(1990) Physical characterisation of the replication region of the
Streptococcus faecalis plasmid pAMI31. Gene 87 : 79-90
Zuker M, Stiegler P (1981) Description of the method used in
RNAFOLD. Nucleic Acids Res 9:133-148
C o m m u n i c a t e d b y H. B 6 h m e