The replicon of pSW800 from Pantoea stewartii

Cheng-Yeu Wu1,2, Jen-Fen Fu2 and Shih-Tung Liu2

Graduate Institute of Microbiology and Immunology, National Yang-Ming University, Shih-Pai, Taipei, 112, Taiwan1
Department of Microbiology and Immunology, Chang-Gung University, Kwei-Shan, 333, Taiwan2

Author for correspondence: Shih-Tung Liu. Tel: +886 3 328 0292. Fax: +886 3 328 0292. e-mail: cgliu{at}mail.cgu.edu.tw


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A 2019 bp DNA fragment containing the replicon of pSW800 from Pantoea stewartii SW2 was cloned and characterized. This replicon contains two genes – repA and repB, which encode a 36·5 kDa replication initiation protein (RepA) and a peptide of 18 aa, respectively. These two genes overlap by 8 bases with repB situated upstream. The replicon also transcribes an antisense RNA (RNAI) that inhibits the expression of repA and repB. The ribosome-binding sequence (RBS) of repA is likely to be hidden in a stem–loop structure, inhibiting the translation of repA. Furthermore, translation of repB is likely to disrupt the stem–loop structure, which is one of the criteria allowing the translation of repA to begin. A mutagenesis study revealed that a sequence (5'-GCACGGG-3') located 111 nt upstream from repA is crucial; mutation of this sequence prevented the translation of repA. Additionally, this region and the stem–loop structure containing the RBS of repA may form an RNA pseudoknot. Results in this study demonstrate that a mechanism similar to that regulating plasmid replication in the IncB, IncI{alpha} and IncL/M groups also regulates pSW800 replication.

Keywords: plasmid replication, replication control

The GenBank accession number for the sequence of the minimal replicon of pSW800 is AF310258.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Pantoea stewartii subsp. stewartii, a corn pathogen, causes necrosis and systemic wilting, known as Stewart’s wilt. This organism generally contains a large number of plasmids. For example, strain SW2 has 13 plasmids, ranging in size from 4 to 320 kb (Coplin et al., 1981 ). Because Pantoea is a member of the Enterobacteriaceae, many of these plasmids can replicate and be stably maintained in Escherichia coli (Frederick & Coplin, 1986 ). Although their total length is approximately 870 kb, these plasmids remain cryptic and no function has been identified for them, except for plasmid replication, conjugation and mobilization (Coplin et al., 1985 ; Fu et al., 1995 , 1996 , 1997 , 1998 ). Among the 13 plasmids in strain SW2, pSW100 (4272 bp) and pSW200 (4367 bp) contain a ColE1-like replicon (Fu et al., 1995 , 1998 ). Plasmid pSW500 (35 kb) has a copy number of two per cell. The replicon of this plasmid encodes a 36·5 kDa replication protein and contains seven iterons regulating plasmid replication (Fu et al., 1996 ). Additionally, plasmid pSW1200 (106 kb) is an IncY plasmid with a copy number of one per cell (Fu et al., 1997 ), while plasmid pSW800 (52 kb), also named pDC250 (Coplin et al., 1985 ), is self-transmissible and can mobilize pCR1, pBR322, pSW100 and pSW200 from P. stewartii to E. coli (Coplin et al., 1985 ; Fu et al., 1995 , 1998 ).

As is generally known, the expression of replication initiation proteins (Rep) is often closely regulated to maintain plasmid copy number and stability (Asano & Mizobuchi, 1998a ; Burian et al., 1999 ; del Solar et al., 1998 ). Plasmids in the IncB, IncI{alpha} and IncL/M groups, which are regulated post-transcriptionally by antisense RNA, are typical examples (Asano et al., 1999 ; Athanasopoulos et al., 1995 ; Siemering et al., 1993 ). In these plasmids the gene encoding the Rep protein overlaps with an upstream gene that is less than 90 bp in length (Asano & Mizobuchi, 1998b ) and contains a ribosome-binding sequence (RBS) hidden in a stem–loop structure preventing the translation of Rep (Asano et al., 1999 ; Athanasopoulos et al., 1995 ; Siemering et al., 1993 ). Although the peptide translated from the upstream gene is unimportant for plasmid replication, the translation process is believed to destablize the stem–loop structure (Praszkier et al., 1992 ). However, translation of this small peptide is still insufficient for the translation of Rep; formation of an RNA pseudoknot structure between the stem–loop region and a sequence upstream is also required to further destabilize the structure to allow the translation of Rep to occur (Athanasopoulos et al., 1999 ). Plasmids regulated by this mechanism also transcribe an antisense RNA molecule, forming a duplex RNA structure to prevent the formation of an RNA pseudoknot (Asano & Mizobuchi, 2000 ; Asano et al., 1998 ) and thus inhibiting the translation of Rep. This work demonstrates that a mechanism similar to that regulating plasmid replication in the IncB, IncI{alpha} and IncL/M groups also regulates pSW800 replication.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
P. stewartii subsp. stewartii SW2 is a wild-type strain (Coplin et al., 1981 ). The E. coli strains used herein were HB101 (F- hsdS20 supE44 recA13 ara14 proA2 rpsL20 xyl-5 mtl-1) (Boyer & Roulland-Dessoix, 1969 ) and CSH50 [rpsL {Delta}(lac-pro)] (Miller, 1972 ). Plasmids used in this study are listed in Table 1. Plasmid pSW800 was digested with PstI and the resulting restriction fragments were ligated with a PstI fragment containing a kanamycin resistance gene isolated from pUC4-KISS, thus generating pSW810 (23 kb). Plasmid pSW810 was deleted with HindIII to produce pSW820 (9 kb) and was digested with PvuII and religated to generate pSW830 (3 kb). The 2 kb PvuII fragment from pSW830, which contains the minimal replicon of pSW800, was cloned into the SmaI site in pGEM-7Zf(+) to generate pSW891, while pSW891 was subsequently digested with SmaI and HindIII to generate pSW893, in which the 5' 686 bp of the 2 kb PvuII fragment was deleted. A DNA fragment containing repA was amplified by PCR with primers rep-800U (5'-ACCCCTTTGCACCGGCATAA-3') and rep-L (5'-ATAGCGACAGGCTTTGCTGA-3'). The PCR fragment was inserted into the SmaI site in pGEM-7Zf(+) to generate pSW894. Additionally, a DNA fragment containing the repB-repA region was amplified by PCR with primers rep-U (5'-CCAAAAGCAGGCGGCGTGCC-3') and rep-L, and the fragment was inserted into the SmaI site of pGEM-7Zf(+) to generate pSW895. PCR was performed at 94 °C for 45 s, 50 °C for 45 s and 72 °C for 1 min for 30 cycles. The EcoRI–SmaI fragment in the replicon (pSW830) was subcloned into pGEM-7Zf(+) to generate pSW896 and into pSW102 to generate pSW897. Meanwhile, the plasmids were prepared according to the alkaline lysis method of Birnboim & Doly (1979) , followed by CsCl/ethidium bromide centrifugation (Sambrook & Russell, 2000 ). Finally, plasmid screening was performed according to the method of Kado & Liu (1981) .


View this table:
[in this window]
[in a new window]
 
Table 1. Plasmids used in this study

 
Electroporation.
P. stewartii was cultured to mid-exponential phase in LB broth at 30 °C. The cells were incubated on ice for 10 min, washed once with H2O and twice with 10% glycerol at 4 °C and finally resuspended in 10% glycerol. Approximately 2x109 c.f.u. and 0·1 µg pBR322 were mixed in a final volume of 100 µl. Electroporation was performed in a 0·1 cm cuvette with an electroporator (GenePulser; Bio-Rad) under conditions of 1500 V, 40 µF and 250 {Omega}. Cells were cultured for 1 h, then plated on LB agar containing 50 µg ampicillin ml-1 and 15 µg tetracycline ml-1.

Transfer of pSW800 and selection of E. coli HB101(pSW800).
P. stewartii SW800(pBR322) and E. coli HB100 were cultured in LB broth overnight. Approximately 1x108 c.f.u. were taken from each strain and mixed. Cells were incubated for 1 h at 30 °C and transconjugants were selected on LB agar containing 50 µg ampicillin ml-1, 15 µg tetracycline ml-1 and 25 µg streptomycin ml-1.

Southern blot analysis.
The 2 kb PvuII fragment from pSW830 was labelled with 50 µCi [{alpha}-32P]dCTP (3000 Ci mmol-1; Amersham), using a random-prime labelling kit (Amersham). DNA in an agarose gel was then transferred to Zeta-Probe membrane (Bio-Rad) according to an alkaline-transfer method described by the manufacturer. Finally, hybridization was performed according to a method described by Sambrook & Russell (2000) .

Determination of plasmid copy number.
Plasmids in P. stewartii SW2 were isolated by using the method of Kado & Liu (1981) and separated by agarose gel electrophoresis. Plasmid bands were then photographed over a UV transilluminator by a digital camera after staining the gel with ethidium bromide. The intensity of the plasmid bands was measured using Gel-Pro Software (Media Cybernetics) on an Apple Macintosh computer. Plasmids pSW500 and pSW1200, with copy numbers of two per cell and one per cell, respectively (Fu et al., 1996 , 1997 ), were used as references to determine the copy number of pSW800 in P. stewartii SW2. The same method was used to measure the copy number of pSW800 in E. coli HB101 with pMAK705, which has a copy number of five per cell, serving as a reference.

DNA sequencing.
DNA sequencing was performed according to the chain-termination method (Sanger et al., 1977 ), using double-stranded plasmid DNA as template.

Incompatibility analysis.
Plasmid pSW891 and its deletion derivatives (Apr) were transformed into E. coli HB101(pSW820) and transformants were selected on LB agar containing ampicillin and kanamycin. The cells were then subcultured in LB-ampicillin broth and the presence of pSW820 in the transformants was determined by replica-plating the colonies on LB agar containing ampicillin and kanamycin.

Northern blot analysis.
Cells were cultured in LB broth at 37 °C for 3 h to exponential phase. Total RNA was prepared from the cells according to the acid/phenol extraction method of Aiba et al. (1981) . RNA (50 µg) was then separated on a 1·5% formaldehyde/1% agarose gel at 27 mA for 2 h, with 23S rRNA (2904 nt), 16S rRNA (1473 nt) and 5S rRNA (128 nt) as size markers. The RNA was transferred and cross-linked to Zeta-Probe membrane and probes specific to RNAI and RNAII were transcribed in vitro with 50 µCi [{alpha}-32P]CTP (3000 Ci mmol-1; Amersham) and SP6 and T7 RNA polymerase, respectively, using SmaI-digested pSW891 as template. Hybridization was performed according to a method described by Chang et al. (1998) .

Primer extension.
Primers P-RNA-II (5'-GTTTTTGCATTCAGGGGTTAG-3'), complementary to the region between map locations 112 and 132, and P-RNAI (5'-CTCGAACTTGCCGGGACGAA-3'), which comprises a sequence between map locations 156 and 175 (see Fig. 2), were end-labelled with 3 µCi [{gamma}-32P]ATP (6000 Ci mmol-1; Amersham) and 5 U T4 polynucleotide kinase (Promega). Primer extension was then performed with 50 µg RNA and 1 U AMV reverse transcriptase (Promega) according to the method described by Lin et al. (1999) . Finally, labelled cDNA was analysed on an 8% urea/polyacrylamide gel (Sambrook & Russell, 2000 ).

Prediction of RNA structure.
The structures of the untranslated region of RNAII were predicted using SeqWeb version 1.1 MFold software [Wisconsin Package Version 10.0, Genetics Computer Group (GCG), Madison, WI, USA].

Construction of fusion plasmids.
A repB-lacZ translation fusion (pSW851) was constructed by inserting the PvuII–BsrFI fragment from pSW830 (see Fig. 2) into the HindIII site of pKM005. A repA-lacZ translational fusion (pSW852) was then generated by inserting the PvuII–XmnI fragment (see Fig. 2) into the HindIII site of pKM005. Meanwhile, transcriptional fusions in repB (pSW855) and repA (pSW856) were constructed by inserting the PvuII–BsrFI and PvuII–XmnI fragments into the XbaI site in pKM005. The -35 sequence of the RNAI promoter in pSW851 and pSW852 was changed to 5'-CTCAGA-3' by a PCR mutagenesis method of Ho et al. (1989) to generate pSW851M and pSW852M, respectively. The initiation codon of repB in pSW851M and pSW852M was changed to CCG by the same method to generate pSW853 and pSW854, respectively. The initiation codon of repA in pSW852M was also mutated to CCG to generate pSW857. Regions Z12–Z17 in pSW852M were substituted with a 5'-CTGCAG-3' sequence to generate pSW861–pSW866, respectively. Sequences in regions a, b, c and d in stem–loop VI in pSW852M were mutated to GCA (pSW871), CCC (pSW872), TGC (pSW873) and GGG (pSW874), respectively. The region e sequence in pSW873 was mutated to GCA to generate pSW875. The region f sequence in pSW874 was mutated to CCC to generate pSW876. Plasmid pSW877 was generated by mutating the region a sequence in pSW875 to GCA; the region b sequence in pSW876 was mutated to CCC to generate pSW878. These fusion plasmids were transformed into E. coli CSH50 and ß-galactosidase activity was assayed as described by Miller (1972) . The assay was performed in duplicate and each experiment was repeated at least three times.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Isolation and characterization of pSW800
Plasmid pSW800 is self-transmissible and can mobilize pBR322 from P. stewartii to E. coli by conjugation (Coplin et al., 1985 ). Consequently, pBR322 was transformed into P. stewartii SW2 (Fig. 1a, lane 1) and then a conjugation experiment was performed to transfer pSW800 and pBR322 into E. coli HB101 (Fig. 1a, lane 3). The transconjugants containing both pSW800 and pBR322 were selected on LB-ampicillin plates and then subcultured in LB broth for 80 generations to cure pBR322 from the cells. Gel electrophoresis revealed that the size of the remaining plasmid (52 kb) was identical to pSW800 (Fig. 1a, lane 2). Southern blot hybridization analysis further revealed that a 2 kb PvuII fragment subcloned from this plasmid (Fig. 1b) was homologous to pSW800 in P. stewartii SW2 (Fig. 1b, lane 1), confirming that the plasmid in E. coli HB101 was indeed pSW800. This plasmid has a copy number of one per cell in both P. stewartii SW2 and E. coli HB101. To verify the stability of pSW800, P. stewartii SW2 and E. coli HB101(pSW800) were cultured for 140 generations in LB broth and then 200 colonies were randomly selected from each culture. All of the selected colonies were found to contain pSW800, indicating that pSW800 is stable in both organisms.



View larger version (91K):
[in this window]
[in a new window]
 
Fig. 1. Isolation of pSW800. (a) P. stewartii was transformed with pBR322 (lane 1). After mating between P. stewartii SW2(pBR322) and E. coli HB101, pSW800 and pBR322 were transferred to E. coli HB101 (lane 3). Plasmid pBR322 was subsequently cured from E. coli HB101 by continuous culturing in LB broth (lane 2). Southern blot analysis revealed that the plasmid purified from E. coli HB101 was homologous to pSW800 in P. stewartii SW2 (b). Southern blot hybridization was performed using a 2 kb 32P-labelled PvuII fragment isolated from pSW830 as a probe. Lane 4, pBR322.

 
Cloning and sequence analysis of the replicon
The replicon of pSW800 was cloned by deleting the plasmid with PstI, HindIII and PvuII, and then ligating the restriction fragments with a kanamycin resistance gene isolated from pUC4-KISS (see Methods). This procedure eventually produced a 3 kb plasmid, pSW830. Because pSW830 was stable in E. coli HB101, pSW830 probably contains all the genetic information necessary for autonomous replication and stable maintenance. The sequence necessary for pSW800 replication resides in a 2019 bp PvuII fragment, which contains a 54 bp repB gene and a 939 bp repA gene (Fig. 2). The small peptide encoded by repB has no sequence homology with sequences in GenBank. repB starts at map location 217 and has an RBS (AGGA) located 10 nt upstream from the initiation codon (Fig. 2). repB overlaps, by 8 bp in the 3' region, with repA (Fig. 2), which starts from a GUG codon located at map location 266 (Fig. 2). This gene encodes a protein of 314 aa with a molecular mass of 36·5 kDa, which probably acts as a replication initiation protein since its sequence is approximately 30% identical to the sequences of the Rep proteins of pUM407.1, p307 and ColIb-P9 (Athanasopoulos et al., 1995 ; Hama et al., 1990 ; Saadi et al., 1987 ). A DnaA box is present downstream from repA at map location 1368 (Fig. 2). The fragment also contains three 9 bp repeats (5'-TACCCGCAA-3') 3' to the DnaA box (Fig. 2). The region between map locations 1497 and 1530 in the pSW800 replicon is A+T-rich with a G+C content of only 12%. The region between map locations 1669 and 2019 is probably unimportant, since deleting this region does not affect plasmid replication and stability.



View larger version (57K):
[in this window]
[in a new window]
 
Fig. 2. The sequence of the minimal replicon of pSW800. Boxed sequences -10 and -35 represent the promoters of RNAI and RNAII, while +1 represents the transcription start site. repB and repA are the two genes essential for pSW800 replication. RNAI and RNAII are the two RNA species transcribed by the replicon. Three 9 bp repeats downstream from repA are underlined. DnaA, DnaA box.

 
Analysis of the inc region
As expected, pSW820 and pSW891, both of which contain the entire pSW800 replicon, were incompatible; when E. coli HB101(pSW820) was transformed with pSW891, the transformants rapidly lost their kanamycin resistance following subculture in LB-ampicillin broth (Fig. 3). However, plasmid pSW893, which contains a deletion in the 5' region of the 2 kb PvuII fragment, was compatible with pSW820 (Fig. 3). The transformants did not lose the kanamycin-resistance phenotype following subculturing. A plasmid containing only repA (pSW894) or both repA and repB (pSW895) was also compatible with pSW820 (Fig. 3), indicating that neither repA nor repB functions as an inc determinant. However, plasmid pSW896 (Fig. 3), which contains the region upstream of repB, was incompatible with pSW820, indicating that the inc region is located upstream of repB. Furthermore, pSW800 in P. stewartii SW2 could be cured from the cells by pSW830. Moreover, this curing did not alter the morphology of the colonies and cells without pSW800 grew normally on M9-glucose agar. Since the replicon of pSW830 only contains the genes necessary for plasmid replication, this curing result suggests that pSW800 does not carry a gene that influences the nutritional requirements of P. stewartii SW2.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 3. Analysis of the inc region of pSW800. Plasmid pSW891 was constructed by subcloning the minimal replicon of pSW800 (a 2019 bp PvuII fragment) into pGEM-7Zf(+). In pSW893, the 5' 686 bp region (a PvuII–SmaI fragment) was deleted. Plasmid pSW894 contains repA, while pSW895 contains repB and repA, and plasmid pSW896 contains a 0·6 kb EcoRI–SmaI fragment of the region. Among these plasmids, pSW891 and pSW896 were incompatible with pSW820, a plasmid containing the replicon of pSW800. Boxes represent the location of repA and repB. Numbers indicate map locations.

 
Northern blot and primer extension analyses
RNA transcribed from the pSW800 replicon was analysed by Northern blot hybridization. A hybridization experiment with a 32P-labelled RNA probe transcribed with SP6 RNA polymerase from pSW891 revealed an RNA molecule approximately 80 nt long (Fig. 4, lane A). Meanwhile, primer extension analysis with primer P-RNAI revealed that this RNA (RNAI) molecule is transcribed leftwards from map location 199 (Figs 2 and 5a). A 32P-labelled RNA probe transcribed with T7 RNA polymerase from pSW891 detected an RNA molecule (RNAII) of approximately 1·7 kb (Fig. 4, lane B). Finally, primer extension analysis with an end-labelled primer (P-RNAII) revealed that the transcription start site of this RNA is at map location 69 (Figs 2 and 5b).



View larger version (61K):
[in this window]
[in a new window]
 
Fig. 4. Transcription of RNAI and RNAII. RNA was isolated from E. coli HB101(pSW830) (lanes A and B) and E. coli HB101(pSW852M) (lane C). Northern blot hybridization was performed with RNA probes complementary to RNAI (lanes A and C) and RNAII (lane B). Plasmid pSW52M has a mutation in the -35 region that abolishes transcription of RNAI.

 


View larger version (78K):
[in this window]
[in a new window]
 
Fig. 5. Determination of transcription start sites by primer extension analysis with P-RNAI (a) and P-RNAII (b). A, C, G and T denote the dideoxynucleotides used to terminate the reaction. Asterisks indicate the 5' terminus of the RNA and the sequences at the side indicate the cDNA products of primer extension.

 
repA-lacZ and repB-lacZ fusions
A repB-lacZ translational fusion (pSW851), in which the ninth codon of repB is fused in phase with the seventh codon of lacZ, expressed ß-galactosidase at a background level with a value of 17 Miller units, approximately equal to that exhibited by pKM005 (7 Miller units) (Table 2). A repA-lacZ translational fusion (pSW852) contains a fusion between the fifth codon of repA and the seventh codon of lacZ. This fusion also expressed a background level of ß-galactosidase (1 Miller unit) (Table 2), indicating that both repA and repB are either not translated or translated at a low level. On the other hand, transcriptional fusions generated in repB and repA (pSW855 and pSW856) exhibited markedly higher levels of ß-galactosidase activity, at 1085 and 1238 Miller units, respectively (Table 2). These values are approximately 155- to 177-times higher than that exhibited by pKM005, suggesting that the mRNA containing repA and repB is transcribed at a high level. These results reveal that expression of repA and repB is regulated post-transcriptionally.


View this table:
[in this window]
[in a new window]
 
Table 2. ß-Galactosidase expressed from repA-lacZ and repB-lacZ transcriptional and translational fusions

 
Inhibition of repB and repA translation by RNAI
Since primer extension analysis confirmed that RNAI is transcribed from the inc region, we examined whether or not transcription of RNAI inhibits the translation of repB and repA. The -35 sequence (5'-TTGTAT-3') of the RNAI promoter in pSW851 and pSW852 was mutated to 5'-CTCAGA-3' and produced two plasmids, pSW851M and pSW852M, respectively. Although reducing the transcription of RNAI by pSW852M to a level undetectable by Northern blot analysis (Fig. 4, lane C), this mutation increased the expression of ß-galactosidase from a background value (17 Miller units) to 1199 Miller units (Table 2). A similar mutation in pSW851 (pSW851M) also increased the expression of ß-galactosidase from 1 to 2135 Miller units (Table 2). Meanwhile, transformation of pSW897, which transcribes RNAI, reduced the expression of ß-galactosidase from pSW851M and pSW852M to 529 and 563 Miller units, respectively. On the other hand, transformation of pSW102, a cloning vector used to construct pSW897, did not influence the expression of ß-galactosidase from pSW851M and pSW852M, demonstrating that RNAI transcribed from pSW897 inhibits the expression of repA and repB. Furthermore, the initiation codon of repB in pSW851M and pSW852M was mutated to CCG (pSW853 and pSW854, respectively). This mutation reduced ß-galactosidase expression from these two plasmids to a background level (Table 2), confirming that repB must be translated for repA to be expressed. Changing the codon of repA in pSW852M at map location 266 from GTG to CCG (pSW857) also reduced ß-galactosidase activity to a background level, suggesting that translation of repA initiates from this codon.

Analysis of the sequence in stem–loop III
The target region in RNAII for RNAI is probably located between +58 and +130 of RNAII (see Discussion; Fig. 6a). This region potentially contains two stem–loop structures, III and IV (Fig. 6a). To elucidate the importance of stem–loop III, we changed the sequence of every 6 nt from +67 (Fig. 6b) (map location 136, Fig. 2) to 5'-CTGCAG-3' (a PstI site) in pSW852M by site-directed mutagenesis, subsequently generating six consecutive mutants, Z12–Z17 (Fig. 6b). The Z12 mutation, which contains a sequence change between +67 and +72 (pSW861), did not influence the expression of ß-galactosidase. The enzyme activity exhibited by the plasmid (2084 Miller units) was roughly equal to that displayed by the parent plasmid, pSW852M (Table 3). However, the Z13 mutation (pSW862), which contains a sequence change from +73 (Fig. 6b), reduced ß-galactosidase expression by 58% to 872 Miller units (Table 3). Furthermore, the Z14 mutation (pSW863), which contains a sequence change from +79 (Fig. 6b), reduced ß-galactosidase expression by 65% to 711 Miller units (Table 3). Moreover, the Z15 mutation (pSW864), which contains a sequence change from +85 (Fig. 6b), reduced the expression even further, decreasing activity by 83% to 364 Miller units (Table 3). Finally, the Z16 mutation reduced the ß-galactosidase expression by 58%, while the Z17 mutation decreased it by 76% to 496 Miller units (Table 3). These results suggest that the structure of stem–loop III and the nucleotide sequence in loop III (Fig. 6a) are essential for repA expression. Similar mutations were also generated in stem–loop IV (Fig. 6a), but reduced the expression of ß-galactosidase less than the mutations in stem–loop III (data not shown), indicating that stem–loop IV is less important than stem–loop III in regulating the translation of repA.



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 6. Predicted secondary structure in the 5' untranslated region of RNAII (a). Roman numerals mark the putative loops formed in the region and Arabic numerals indicate the map locations relative to the +1 site of RNAII. The target region for RNAI is located between +58 and +130 and is marked by two open triangles. The initiation codon of repA is in bold type. a, b, c, d depict the four regions potentially from an RNA pseudoknot structure; e and f, regions pairing with regions c and d. (b) Regions (Z12–Z17) in stem–loop III were changed to 5'-CTGCAG-3' to examine the function of this stem–loop structure in plasmid replication.

 

View this table:
[in this window]
[in a new window]
 
Table 3. ß-Galactosidase expressed from the mutants generated in stem–loops III and VI

 
Interaction between stem–loops III and VI
Since the sequence in loop III is crucial to the translation of repA, additional mutations were generated within this loop to investigate whether they influenced the expression of repA. Notably, the UGC sequence (region a) and the GGG sequence (region b) in loop III are complementary to the GCA sequence located at +173 (region c) and the CCC sequence located at +166 (region d), respectively (Fig. 6a). Changing the sequence in region a from UGC to GCA in pSW852M reduced the expression of ß-galactosidase by 94% to a background level. A change in region b from GGG to CCC also reduced expression by 85% to 311 Miller units (Table 3), demonstrating the importance of regions a and b in repA expression. However, mutating the sequence in region c from GCA to UGC in pSW852M did not influence the expression, and the activity expressed by the mutant plasmid (pSW873) was approximately equal to that exhibited by pSW852M (Table 3). Similarly, mutation in region d (pSW874) decreased the expression of ß-galactosidase by 11% (Table 3). These results strongly suggested that disrupting the structure of stem–loop VI has little effect on the translation of repA. Since a sequence change in regions c or d disrupts the structure of stem–loop III, a second mutation on the opposite side of the stem in regions e and f (Fig. 6a) was generated to restore the stem–loop structure. Mutations in both regions c and e (pSW875) reduced the expression of ß-galactosidase by 89% to 229 Miller units, while mutations in regions d and f (pSW876) decreased it by 63% to 781 Miller units (Table 3). A third mutation was subsequently generated in pSW875 and pSW876 to restore the complementarity of the sequence between regions a and c (pSW877), and between regions b and d (pSW878). Plasmid pSW877 exhibited a ß-galactosidase activity 6·6-fold higher (1603 Miller units) than that displayed by pSW875 (Table 3). Finally, the level of ß-galactosidase expression by pSW878 was also 2·1-fold higher (1603 Miller units) than that expressed by pSW876.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The genes in the replicon of pSW800 and IncB, IncI{alpha} and IncL/M plasmids are similarly organized
The replicon of the plasmids in the IncB, IncI{alpha} and IncL/M groups typically contains a small gene, usually less than 90 bp, which overlaps in the 3' region with a gene encoding the Rep protein (Asano & Mizobuchi, 1998b ; Athanasopoulos et al., 1995 ; Praszkier et al., 1992 ). The genes in the replicon of pSW800 are also organized similarly (Fig. 2). Furthermore, a mutation in the -35 region of the RNAI promoter was found to be unobtainable unless a lacZ gene was inserted into repA (pSW851M and pSW852M), suggesting that RepA is toxic to host cells if synthesis of RepA is no longer repressed by RNAI. This mutation is likely to result in an unregulated synthesis of RepA and uncontrolled, runaway plasmid replication. A similar phenomenon has also been observed in ColIb-P9 and NR1 (Hama et al., 1990 ; Wu et al., 1985 ).

Translation of repA and repB is coupled
Translation of repA and repB is likely to be coupled in pSW800 since mutation of the initiation codon of repB (pSW853 and pSW854) prevents the translation of repA (Table 2). Translation coupling is also important for expressing a replication initiation protein from the plasmids in the IncFII (Blomberg et al., 1992 ), IncI{alpha}, IncL/M and IncB groups (Asano & Mizobuchi, 1998 ; Athanasopoulos et al., 1995 ; Praszkier et al., 1992 ). In these plasmids, and in pSW800, repB (repY) typically overlaps with repA (repZ). This unusual organization seems crucial for the translation of Rep, since mutation of the termination codon or generation of a nonsense mutation in repB in pUM604, pUM720 and repY in ColIb-P9 significantly reduces the expression of the Rep protein (Asano et al., 1999 ; Athanasopoulos et al., 1995 ; Siemering et al., 1993 ). These studies suggest that, in translation coupling, ribosome translation of the upstream gene has to terminate at an exact location so that the ribosome can translate the downstream gene directly without leaving the mRNA (Athanasopoulos et al., 1995 ). Furthermore, the RBS of repA or repZ of pMU604, pMU720, ColIb-P9 and pSW800 is typically hidden in a stem–loop structure (Fig. 6a; Asano & Mizobuchi, 2000 ; Athanasopoulos et al., 1999 ; Praszkier et al., 1992 ). The RBS becomes available for ribosome binding, allowing the translation of Rep to occur, if the stem–loop structure is destroyed by mutagenesis (Table 2; pSW873 and pSW874). However, translation coupling between repB and repA of pMU604 and pMU720 also depends on the formation of an RNA pseudoknot structure between the sequence in the stem–loop structure and a sequence located upstream (Athanasopoulos et al., 1995 ; Praszkier et al., 1992 ). In the case of pUM604, a lack of repA translation when an RNA pseudoknot structure is absent is partly attributed to the poor RBS sequence of the gene (Athanasopoulos et al., 1995 ). Because repA of pSW800 also contains a short Shine–Dalgarno sequence (Fig. 2; 5'-CGGA-3'), this may explain why pSW871 and pSW872 express ß-galactosidase at a background level (Table 3) and formation of an RNA pseudoknot structure is required for the translation of repA (Table 3; pSW877 and pSW878).

Expression of repA and repB is regulated post-transcriptionally
Translational fusions (pSW851 and pSW852) revealed that repA and repB are translated at a low level (Table 2), which can probably be attributed to the transcription of RNAI, since a -35 mutation in the RNAI promoter significantly raises the expression, and presence of pSW897, which transcribes RNAI, can again repress the expression from pSW8521M and pSW852M (Table 2). Furthermore, transcriptional fusions generated in repA and repB (pSW855 and pSW856) also revealed that repA and repB are regulated post-transcriptionally at the translational level. According to our analysis, transcription of RNAI begins from map location 199 (Fig. 5a), which is only 4 nt upstream from the RBS of repB (Fig. 2). Although RNAI does not directly form a duplex with the region containing the RBS of repB, the duplex region in RNAII is probably close enough to impede the binding of ribosome to the repB RBS, thus inhibiting the translation of repB, and ultimately repA. This may indeed be the case, since the RBS of repB and the transcription start site of RNAI in pMU604 are also separated by 4 nt (Athanasopoulos et al., 1995 ).

RNase III is not required for control of repA expression
Since interaction between antisense and target RNAs will create double-stranded RNA structures, it was conceivable that cleavage by the double-stranded RNA-specific RNase III could affect control of repA expression. To test whether RNase III influences the replication, we transformed pSW830 into E. coli HT115 (Franch et al., 1999 ), an RNase III-deficient strain, revealing that the plasmid replicates normally and is stably maintained in the strain, and thus suggesting that RNase III does not influence pSW800 replication.

The sequence in loop III is essential for repA translation
Transcription of RNAI from pSW800 is likely to terminate at the 3' region of stem–loop III, since the stem–loop structure resembles a {rho}-independent transcription stop signal (Fig. 6a). If transcription of RNAI terminates here, RNAI should be 73 nt long, consistent with the estimated size from Northern blot analysis (Fig. 4a). The target area for RNAI in RNAII of pSW800 thus contains stem–loops III and IV (Fig. 6b). Linker scanning analysis of stem–loop III reveals that the Z12 mutation in pSW852M, which does not significantly influence the stem–loop structure, does not influence expression (Fig. 6b and Table 3). However, the mutations that disrupt the stem–loop structure in pSW852M, including mutations Z13, Z16 and Z17, significantly reduced ß-galactosidase expression (Fig. 6b and Table 3). Varying the sequence in loop III reduced the expression even further, to just 17% of the level expressed by pSW852M (Table 3). These experimental results clearly demonstrate not only the critical role the stem–loop structure and the loop sequence play, but also that the interaction between loop III and a downstream region affecting the translation of repA, probably stem–loop VI, are crucial.

Interaction between stem–loops III and VI is important for the expression of repA
In the plasmids of IncI{alpha}, IncL/M and IncB groups, formation of an RNA pseudoknot structure is required before Rep proteins can be expressed (Asano & Mizobuchi, 1998b ; Athanasopoulos et al., 1995 ; Siemering et al., 1993 ). These RNA pseudoknot structures are frequently formed between an upstream loop and the stem of the stem–loop structure containing the RBS of repA (Asano & Mizobuchi, 1998b ), and these two regions are normally separated by approximately 100 nt (Asano & Mizobuchi, 1998 ; Athanasopoulos et al., 1995 ; Praszkier et al., 1992 ). However, the length of the sequence involving pseudoknot formation in these plasmids generally varies. For example, in ColIb-P9, only five of the seven complementary bases are important for pseudoknot formation (Asano & Mizobuchi, 1998b ). Meanwhile, pseudoknot formation in pUM604 requires six of the eight complementary bases (Athanasopoulos et al., 1999 ). If pseudoknot formation is also required for repA translation in pSW800, a sequence that complements loop III and stem VI may appear. Our results reveal that mutation in region c (Fig. 6a, Table 3), which partially disrupts the structure of stem–loop VI and destroys the pairing between regions a and c, does not influence the expression of ß-galactosidase from pSW852M (Table 3). Meanwhile, mutation of the sequence of region e (pSW875), to restore the structure of stem–loop VI in pSW873, reduced the expression of ß-galactosidase by 89%, indicating that the structure of stem VI prevents the translation of repA. Changing the region a sequence (pSW875) to restore the pairing between regions a and c in pSW875 again increases the expression of ß-galactosidase (Table 3). This behaviour demonstrates that the two regions must interact, probably forming a pseudoknot structure, for repA to be expressed. Since a mutation in region c (pSW873) causes repA translation, this mutation may destabilize the structure of stem–loop VI to a degree that a pseudoknot structure is no longer necessary for repA translation. Similar to the mutation in region c, a mutation in region d does not significantly influence expression. However, in contrast to the sequence changes in regions c and e (pSW875), mutations in regions of d and f (pSW876) reduced the expression of ß-galactosidase only by 58% (Table 3). It is unclear why a mutation in regions d and f does not decrease the expression as much as a mutation in regions c and e. It is likely that the structure in regions d and f may also be influenced by the ribosome that occupies the region where repB terminates, explaining why changing the sequence alone does not completely abolish the translation. We also found that restoring the base pairing between regions b and d (pSW878) in pSW876 still increases ß-galactosidase expression to 86% of that exhibited by pSW874 (Table 3), showing that interaction between regions b and d can enhance the translation of repA.

Results in this study have demonstrated that a mechanism similar to that regulating plasmid replication in the IncB, IncI{alpha} and IncL/M groups also regulates pSW800 replication. Translational coupling between repA and repB and interactions between loop III and stem VI are required for repA of pSW800 to be expressed.


   ACKNOWLEDGEMENTS
 
The authors want to thank Dr T. Franch for providing E. coli HT115. This research was supported by the Chang-Gung Memorial Hospital (CMRP-720IV) and the National Science Council (NCS-2311-B-182-002) of the Republic of China.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aiba, H., Adhya, S. & de Crombrugghe, B. (1981). Evidence for two functional gal promoters in intact Escherichia coli cells. J Biol Chem 256, 11905-11910.[Abstract/Free Full Text]

Asano, K. & Mizobuchi, K. (1998a). Copy number control of Inc{alpha} plasmid ColIb-P9 by competition between pseudoknot formation and antisense RNA binding at a specific RNA site. EMBO J 17, 5201-5213.[Abstract/Free Full Text]

Asano, K. & Mizobuchi, K. (1998b). An RNA pseudoknot as the molecular switch for translation of the repZ gene encoding the replication initiator of IncI{alpha} plasmid ColIb-P9. J Biol Chem 273, 11815-11825.[Abstract/Free Full Text]

Asano, K. & Mizobuchi, K. (2000). Structural analysis of late intermediate complex formed between plasmid ColIb-P9 Inc RNA and its target RNA. How does a single antisense RNA repress translation of two genes at different rates? J Biol Chem 275, 1269-1274.[Abstract/Free Full Text]

Asano, K., Niimi, T., Yokoyama, S. & Mizobuchi, K. (1998). Structural basis for binding of the plasmid ColIb-P9 antisense Inc RNA to its target RNA with the 5'-rUUGGCG-3' motif in the loop sequence. J Biol Chem 273, 11826-11838.[Abstract/Free Full Text]

Asano, K., Hama, C., Inoue, S., Moriwaki, H. & Mizobuchi, K. (1999). The plasmid ColIb-P9 antisense Inc RNA controls expression of the RepZ replication protein and its positive regulator repY with different mechanisms. J Biol Chem 274, 17924-17933.[Abstract/Free Full Text]

Athanasopoulos, V., Praszkier, J. & Pittard, A. J. (1995). The replication of an IncL/M plasmid is subject to antisense control. J Bacteriol 177, 4730-4741.[Abstract]

Athanasopoulos, V., Praszkier, J. & Pittard, A. J. (1999). Analysis of elements involved in pseudoknot-dependent expression and regulation of the repA gene of an IncL/M plasmid. J Bacteriol 181, 1811-1819.[Abstract/Free Full Text]

Birnboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7, 1513-1523.[Abstract]

Blomberg, P., Nördstrom, K. & Wagner, E. G. H. (1992). Replication control of plasmid R1: RepA synthesis is regulated by CopA RNA through inhibition of leader peptide translation. EMBO J 11, 2675-2683.[Abstract]

Bolivar, F., Rodriguez, R. L., Greene, P. J., Betlach, M. C., Heyneker, H. L. & Boyer, H. W. (1977). Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2, 95-113.[Medline]

Boyer, H. W. & Roulland-Dessoix, D. (1969). A complementation analysis of the restriction and modification of DNA in Escherichia coli. J Mol Biol 41, 459-472.[Medline]

Burian, J., Stuchlik, S. & Kay, W. W. (1999). Replication control of a small cryptic plasmid of Escherichia coli. J Mol Biol 294, 49-65.[Medline]

Chang, P. J., Chang, Y. S. & Liu, S. T. (1998). Role of Rta in the translation of bicistronic BZLF1 of Epstein–Barr virus. J Virol 72, 5128-5136.[Abstract/Free Full Text]

Coplin, D. L., Rowan, R. G., Chisholm, D. A. & Whitmoyer, R. E. (1981). Characterization of plasmids in Erwinia stewartii. Appl Environ Microbiol 42, 599-604.[Medline]

Coplin, D. L., Frederick, R. D. & McCammon, S. L. (1985). Characterization of a conjugative plasmid from Erwinia stewartii. J Gen Microbiol 131, 2985-2991.

Franch, T., Thisted, T. & Gerdes, K. (1999). Ribonuclease III processing of coaxially stacked RNA helices. J Biol Chem 274, 26572-26578.[Abstract/Free Full Text]

Frederick, R. D. & Coplin, D. L. (1986). Transformation of Escherichia coli by plasmid DNA from Erwinia stewartii. Mol Plant Pathol 76, 1353-1356.

Fu, J. F., Chang, H. C., Chen, Y. M., Chang, Y. S. & Liu, S. T. (1995). Sequence analysis of an Erwinia stewartii plasmid, pSW100. Plasmid 34, 75-84.[Medline]

Fu, J. F., Chang, H. C., Chen, Y. M., Chang, Y. S. & Liu, S. T. (1996). Characterization of the replicon of plasmid pSW500 of Erwinia stewartii. Mol Gen Genet 250, 699-704.[Medline]

Fu, J. F., Ying, S. W. & Liu, S. T. (1997). Cloning and characterization of the ori region of pSW1200 of Erwinia stewartii: similarity with plasmid P1. Plasmid 38, 141-147.[Medline]

Fu, J. F., Hu, J. M., Chang, Y. S. & Liu, S. T. (1998). Isolation and characterization of plasmid pSW200 from Erwinia stewartii. Plasmid 40, 100-112.[Medline]

Hama, C., Takizawa, T., Moriwaki, H., Urasaki, Y. & Mizobuchi, K. (1990). Organization of the replication control region of plasmid ColIb-P9. J Bacteriol 172, 1983-1991.[Medline]

Hamilton, C. M., Aldea, M., Washburn, B. K., Babitzke, P. & Kushner, S. R. (1989). New method for generating deletions and gene replacements in Escherichia coli. J Bacteriol 171, 4617-4622.[Medline]

Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K. & Pease, L. R. (1989). Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51-59.[Medline]

Kado, C. I. & Liu, S. T. (1981). Rapid procedure for detection and isolation of large and small plasmids. J Bacteriol 145, 1365-1373.[Medline]

Lin, T. P., Chen, C. L., Chang, L. K., Tschen, J. S. & Liu, S. T. (1999). Functional and transcriptional analyses of a fengycin synthetase gene, fenC, from Bacillus subtilis. J Bacteriol 181, 5060-5067.[Abstract/Free Full Text]

Masui, Y., Coleman, J. & Inouye, M. (1983). Multipurpose Expression Cloning Vehicles in Escherichia coli. New York: Academic Press.

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Oka, A., Sugisaki, H. & Takanami, M. (1981). Nucleotide sequence of the kanamycin resistance transposon Tn903. J Mol Biol 147, 217-226.[Medline]

Praszkier, J., Wilson, I. W. & Pittard, A. J. (1992). Mutations affecting translational coupling between the rep genes of an IncB miniplasmid. J Bacteriol 174, 2376-2383.[Abstract]

Saadi, S., Maas, W. K., Hill, D. F. & Bergquist, P. L. (1987). Nucleotide sequence analysis of RepFIC, a basic replicon present in IncFI plasmids P307 and F, and its relation to the RepA replicon of IncFII plasmids. J Bacteriol 169, 1836-1846.[Medline]

Sambrook, J. & Russell, D. (2000). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sanger, F., Nicklen, S. & Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74, 5463-5467.[Abstract]

Siemering, K. R., Praszkier, J. & Pittard, A. J. (1993). Interaction between the antisense and target RNAs involved in the regulation of IncB plasmid replication. J Bacteriol 175, 2895-2906.[Abstract]

del Solar, G., Giraldo, R., Ruiz-Echevarria, M. J., Espinosa, M. & Diaz-Orejas, R. (1998). Replication and control of circular bacterial plasmids. Microbiol Mol Biol Rev 62, 434-464.[Abstract/Free Full Text]

Wu, R. P., Womble, D. D. & Rownd, R. H. (1985). Incompatibility mutants of IncFII plasmid NR1 and their effect on replication control. J Bacteriol 163, 973-982.[Medline]

Received 6 March 2001; revised 1 June 2001; accepted 12 June 2001.