Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lorenzo-Díaz, F. Articles by Espinosa, M. Search for Related Content PubMed PubMed Citation Articles by Lorenzo-Díaz, F. Articles by Espinosa, M.

 Previous Article  |  Next Article 

Journal of Bacteriology, February 2009, p. 720-727, Vol. 191, No. 3
0021-9193/09/$08.00+0     doi:10.1128/JB.01257-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Lagging-Strand DNA Replication Origins Are Required for Conjugal Transfer of the Promiscuous Plasmid pMV158

Fabián Lorenzo-Díaz and Manuel Espinosa^*

Centro de Investigaciones Biológicas, Consejo Superior de Investigaciones Científicas, Ramiro de Maeztu 9, E-28040 Madrid, Spain

Received 8 September 2008/ Accepted 9 November 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The promiscuous streptococcal plasmid pMV158 is mobilizable by auxiliary plasmids and replicates by the rolling-circle mechanism in a variety of bacterial hosts. The plasmid has two lagging-strand origins, ssoA and ssoU, involved in the conversion of single-stranded DNA intermediates into double-stranded plasmid DNA during vegetative replication. Transfer of the plasmid also would involve conversion of single-stranded DNA molecules into double-stranded plasmid forms in the recipient cells by conjugative replication. To test whether lagging-strand origins played a role in horizontal transfer, pMV158 derivatives defective in one or in both sso's were constructed and tested for their ability to colonize new hosts by means of intra- and interspecies mobilization. Whereas either sso supported transfer between strains of Streptococcus pneumoniae, only plasmids that had an intact ssoU could be efficiently mobilized from S. pneumoniae to Enterococcus faecalis. Thus, it appears that ssoU is a critical factor for pMV158 promiscuity and that the presence of a functional sso plays an essential role in plasmid transfer.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Conjugation of bacterial plasmids is, together with transposition, the most important source of horizontal gene transfer among bacteria of the same or of different species (42). Conjugation implies the unidirectional transfer of one plasmid DNA strand from a donor to a recipient cell. This is initiated by the activity of a plasmid-encoded protein generically termed relaxase, in a process that resembles replication by the rolling-circle mechanism (13, 29). In the case of numerous, small plasmids (<10 kb) isolated primarily from gram-positive bacteria, two pioneer findings led to the discovery of the rolling-circle mechanism of replication (RCR) plasmids (reviewed in references 20 and 21). First, the strand-specific single-stranded DNA (ssDNA) molecules which act as replication intermediates were identified (41) and, second, the relaxing activity on the supercoiled DNA via the recognition of a specific sequence (the double-strand origin) of the Rep initiator proteins were described (22). Most RCR plasmids are not self-transmissible; instead, they encode not only the Rep topoisomerase-like initiator but also a Mob protein with relaxase activity involved in mobilization mediated by auxiliary plasmids. Such is the case of the promiscuous plasmid pMV158, which can be mobilized between various bacterial species by the pMV158-encoded MobM protein and by helper conjugative plasmids belonging to the Inc18 plasmid family, such as pAMβ1 (15), or even by IncP plasmids, such as RP4 (11). The relaxing activity of MobM on supercoiled DNA of pMV158 and the site of cleavage were first demonstrated in vitro (5, 17), and later the same activity of the MobA protein of the Staphylococcus aureus RCR plasmid pC221 was demonstrated (3, 39).

Initiation of transfer, like initiation of RCR, involves cleavage of the phosphodiester bond of a specific dinucleotide on one of the plasmid strands. Cleavage is mediated either by the plasmid-encoded Mob protein at the origin of transfer (oriT) during conjugation or by the plasmid-encoded Rep protein at the dso during replication. In both processes, this initial stage is followed by displacement of the cleaved strand in a unidirectional manner (8, 21, 29, 36). Thus, RCR and conjugal transfer are equivalent processes in the sense that they generate strand-specific ssDNA plasmid intermediates that correspond only to the cleaved strand (9, 16, 41). The ssDNA intermediates are generated in the plasmid host by the activity of the Rep initiator protein (replication) or generated and transferred to the recipient cell (T-DNA) and closed by the Mob relaxase (conjugation), where they are converted into double-stranded plasmid DNA (dsDNA) molecules by lagging-strand synthesis. Replication of the lagging strand is initiated at the single-strand origins (sso) by the host RNA polymerase (RNAP), upon recognition of a specific site on ssDNA and synthesis of a short RNA primer (pRNA). The pRNA is used by DNA polymerase I for limited extension synthesis, followed by replication of the lagging strand by DNA polymerase III (27). Features of the sso include the potential to generate stem-loop structures on ssDNA (9, 16, 41) that can conform a ssDNA promoter, which is inactive in the dsDNA configuration. This kind of promoter was described in the coliphage N4 (18), as recognized by the virion RNAP (4, 14). A different kind of ssDNA promoter, F_rpo, was reported for the Escherichia coli plasmid F and was demonstrated to be used for gene expression and appeared to play a role during plasmid conjugation (34). The presence of ssDNA promoters has also been demonstrated in plasmids pMV158 (27) and ColI-P9 (1, 35). The organization of this kind of promoters showed that they are placed on the DNA strand that is partially complementary to the template strand.

The first sso was described in the staphylococcal RCR plasmid pT181, in which a deletion located out of the replicon led to instability, reduction in copy number, and accumulation of ssDNA intermediates (16). Plasmid pMV158 exhibits two sso's, ssoA and ssoU (23). Two conserved regions were found in the ssoA of pLS1 plasmid (a nonmobilizable pMV158 derivative lacking ssoU): a short region termed recombination site B, RS_B, supposedly involved in plasmid cointegration (16, 38), and a 6-nucleotide (nt) consensus sequence (5'-TAGCGT-3', termed CS-6). Determination of the roles of these two conserved sites showed that, whereas RS_B was the primary site of RNAP binding (located at the stem of the hairpin), CS-6 was the termination site for the synthesis of a 20-nt pRNA in the loop of the hairpin (27). The predicted intrastrand pairings in the pMV158-ssoA showed the presence of an ssDNA promoter in the vicinity of the RS_B, which would have a consensus –35 region (5'-TTGACA-3') but a weak –10 region (5'-TAcgcT-3'). With this situation, RNA synthesis should start and proceed in the direction toward the binding site of RNAP, being thus opposite to RNA synthesis from classic promoters (27) (see Fig. 1A). Sites homologous to RS_B and CS-6 were later observed in pMV158-ssoU (24).

View larger version (20K): [in this window] [in a new window]

FIG. 1. Features of pMV158 and its two lagging-strand origins. (A) Schematic map of the plasmid transfer module indicating relevant restriction sites and the relative positions (shadowed) of the two lagging-strand origins of replication (ssoA and ssoU). Plasmid-encoded MobM protein (arrow below the map) and the position of the oriT are depicted. Direction of DNA transfer is indicated. The EcoRI fragment deleted to construct the pLS1 derivative (28) and the positions of primers used are also shown. A representation of the secondary structures of ssoA (left) and of ssoU (right) indicates the positions of the CS-6 and the RSB regions (boxed). The locations of the G3 and G7 mutations in the ssoA and of the restriction sites used to generate deletions in the ssoU are shown. The start point and direction for the RNA primer (pRNA) synthesis, downstream to CS-6 sequence, is indicated by a wavy arrow. (B) Relevant sequence features of the two pMV158 sso. The RSB and CS-6 sequences are shown in boxes. The restriction sites of BsaI and DraI used to generate ssoU-BD mutant are also indicated, as well as the nucleotides changed (boldface) to construct the ssoA-G3G7 mutant (sequence indicated beneath).

In the present study we have addressed the question of whether and, eventually, which of the two pMV158-ssos plays a role in conjugal transfer. With this objective, we constructed pMV158-derivatives defective in one or both sso's and tested their role on intra- and interspecies mobilization. Whereas either sso supported transfer between strains of Streptococcus pneumoniae with the same efficiency as the parental pMV158, only the ssoU could do so when conjugal transfer was assayed between S. pneumoniae and Enterococcus faecalis. Our findings show that the functionality of ssoU is a critical factor in the colonization of a broad range of gram-positive bacteria for the pMV158 promiscuous plasmid and demonstrate that efficient transfer and replication in enterococci depend upon a functional ssoU. We suggest that sso's lacking functionality for vegetative replication in a specific host should not be efficient in conjugative transfer and vice versa, since both events are mechanistically identical. As far as we know, this is the first report that shows the effect of sso functionality on horizontal gene transfer by plasmid conjugation, as well as the efficiency of the ssoA and ssoU in E. faecalis.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains, plasmids, and DNA manipulations. Bacteria and plasmids used are listed in Table 1. AGCH and ESTY (Pronadisa, Spain) media and the growth conditions for S. pneumoniae and E. faecalis have been described (28, 33). Competent S. pneumoniae 708 cells bearing plasmid pAMβ1 were transformed with DNA from pMV158wt and its derivatives as reported elsewhere (10). These strains were used as donors in filter-mating experiments. To construct S. pneumoniae MP3008, the novobiocin-resistant (Nov^r) strain MP517, which is unable to grow in maltose as the only carbon source, was transformed with the PstI DNA fragment from plasmid pLS70 that contains part of the pneumococcal wild-type mal operon (40). S. pneumoniae MP3008 (Nov^r) and E. faecalis OG1RF (resistant to rifampin) were used as recipients in intra- and interspecies transfer assays, respectively. Cultures of transconjugants were used to determine plasmid copy numbers and to detect intracellular ssDNA intermediates as described previously (9). Selection for plasmids pAMβ1 and pMV158 was performed using 1 µg of erythromycin/ml and 1 µg of tetracycline/ml in S. pneumoniae and 1 µg of erythromycin/ml and 4 µg of tetracycline/ml in E. faecalis. Purified pMV158 plasmid DNA was prepared by two consecutive CsCl-ethidium bromide gradients as described previously (7). Plasmid pMV158BD, a derivative of pMV158 lacking the ssoU was constructed by deletion of a 205-bp BsaI-DraI DNA fragment (coordinates 3223 to 3428 of pMV158; see Fig. 1). To generate pMV158-derivatives in the ssoA, the small 1,132-bp EcoRI fragment from pMV158 was cloned into plasmid pLS1G3G7 carrying nucleotide changes in the RS_B and CS-6 conserved sequences of the ssoA (27) (see Fig. 1). The resulting plasmid, pMV158G3G7, was functionally defective in the ssoA. Similarly, the small EcoRI restriction fragment from pMV158BD (927 bp) was cloned into pLS1G3G7, thus generating pMV158G3G7BD, with defective functionality of both sso genes. All constructions were rescued by transformation of competent pneumococcal cells, and the mutations were confirmed by sequencing with specific primers. U1 (GGGATCAACTTTGGGAGAGA) and U2 (GCGTCTCAAAAACACGTTCA) were used to confirm the ssoU deletion, A1 (TCACAACGCTCACCTCCA) was used to confirm the G3G7 mutations of ssoA, and U1 and M1 (AAAGCACCCCTCACATGC) were used to confirm the orientation of the small EcoRI fragment (see Fig. 1A).

View this table: [in this window] [in a new window]

TABLE 1. Bacterial strains and plasmids used in this study

Filter-mating experiments. Mobilization assays of pMV158 and its derivates from S. pneumoniae donor cells harboring pAMβ1 as auxiliary plasmid were performed as described previously (37) with minor modifications. Donor and recipient cultures were grown without aeration at 37°C to 5 x 10⁸ cells/ml. Cells were centrifuged and resuspended in prewarmed AGCH medium supplemented with 10 mM MgCl₂, 2 mg of bovine serum albumin/ml, and 100 U of DNase I. Donor-recipient mixtures (1/5 ratios) were filtered onto sterile 25-mm nitrocellulose filters (0.22-µm pore size). The filters were then placed cell-side-down over another filter previously placed on a plate with conjugation medium (AGCH with 10 mM MgCl₂, 2 mg of bovine serum albumin/ml, and 2% agar). After 4 h of incubation at 37°C, the cells were recovered by washing the filters in 1 ml of AGCH medium. A recent method to perform multiple simultaneous conjugations was also applied (33), where several donor cell densities were mixed with a fixed recipient cell density and placed onto a multiwell plate equipped with a 0.22-µm-pore-size nitrocellulose filter (Millipore), filtered, and incubated for 4 h at 37°C. With this device, eight transfer experiments were performed for each condition, resulting in a high degree of reproducibility and enough repetitions of the same experiment. Transconjugants were selected on AGCH medium with tetracycline (1 µg/ml), novobiocin (10 µg/ml), 0.3% maltose, and 1.5% agar for S. pneumoniae MP3008 or on ESTY plates with tetracycline (4 µg/ml), 0.3% glucose, and 1.5% agar for E. faecalis OG1RF. After serial dilutions, the number of CFU of recipient strains was calculated in the same media, but without selection for tetracycline resistance. Conjugative mobilization efficiencies were calculated as the number of transconjugant cells per recipient.

Determination of plasmid copy number and intracellular ssDNA accumulation. Total DNA preparations from cultures harboring plasmids (28) were loaded on 0.7% agarose gels in Tris-borate buffer containing 0.5 µg of ethidium bromide/ml. After electrophoresis, the plasmid copy number was determined by fluorescence densitometry with the Gel-Doc system and QuantityOne software (Bio-Rad). The DNAs in the same gels were denatured and transferred to positively charged nylon membranes (Roche), instead of the nitrocellulose ones used earlier (9, 41) since the nylon membranes offered superior performance. Membranes were hybridized with a specific ³²P-labeled PCR-DNA probe (coordinates 609 to 924 of pMV158), amplified with P1 (GCACGGTTATGCTACT) and P2 (CAGCTCCCAGTCGCTT) primers. The total amounts of ssDNA and dsDNA were quantified with PhosphorImager equipment and using ImageQuant software (Molecular Dynamics).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Construction of pMV158 derivatives deficient in sso functionality. Plasmid pMV158 contains a gene cassette devoted to its mobilization, which is composed of the origin of transfer, oriT, and the gene encoding the MobM relaxase (Fig. 1A). This cassette is encompassed by the two sso genes in such a way that, during conjugation, ssoA, generated in the transferred ssDNA, enters the recipient cell first, whereas ssoU is the last plasmid region transferred (Fig. 1A; see also Fig. 5). Computer-assisted and structural analyses showed that both sso's could generate complex secondary structures on the ssDNA intermediates by intrastrand pairing. In these stem-loop structures the consensus sequences RS_B (the RNAP binding sites) is partially paired and the CS-6 would be unpaired and placed on the loop of the hairpin (24, 27); schematized in Fig. 1A. In the case of ssoA, a single hairpin would be generated, whereas up to five hairpins may be formed in the region encompassing ssoU (Fig. 1A). Homologies between the two lagging-strand origins were observed especially around the RS_B conserved sequences, homologies that were partially maintained when the four types of sso reported thus far were aligned (24). A strong interaction between the host RNAP and the lagging-strand origin(s) could be an important factor in determining the host range of RCR plasmids (24). This, in conjunction with the genetic structure surrounding the mobM cassette and the promiscuity of pMV158, led us to hypothesize that the ssoU could be involved in the determination of the broad host range of ssoU-containing RCR plasmids, thus contributing to the horizontal spread between the different hosts they colonize (6). To address this hypothesis, we constructed pMV158 derivatives carrying modifications in their sso, either individually or together, that affected their functionality. To modify ssoA, a 6-nt change was introduced in the unpaired RS_B sequence (mutation G7), as well as another 9-nt change introduced into the CS-6 sequence (mutation G3), as shown in Fig. 1B. These changes did not alter the high potential for secondary structure formation within ssoA, and only local changes in the general organization of this region were predicted by computer analysis (not shown). However, the mutations hindered binding of RNAP (changes at the RS_B) and affected termination of the pRNA (changes at CS-6) (27). The resulting mutant plasmid, pMV158G3G7 (Fig. 1B and Table 1), has a nonfunctional ssoA. In the case of the pMV158-ssoU, a fine characterization of the nucleotides important for its functionality has not yet been made. Thus, we generated a 205-bp deletion that affected hairpins I, II, and III (Fig. 1A). The resulting plasmid, termed pMV158BD, lacks the two most important sequences (RS_B and CS-6) within the ssoU. Finally, a derivative affected in both origins was also generated (plasmid pMV158G3G7BD). The wild-type pMV158 and its three derivatives were rescued in competent S. pneumoniae cells, and the mutations were confirmed by sequencing of the affected regions.

View larger version (20K): [in this window] [in a new window]

FIG. 5. Proposed model for reconstitution of pMV158 dsDNA after transfer. After nicking by the MobM relaxase, the cleaved strand is piloted by MobM into the recipient cell. The ssoA lagging strand enters first and as ssDNA undergoes intramolecular pairing and then acts as a conversion signal for the host-recipient RNAP to synthesize the pRNA which initiates lagging-strand replication. If the recipient RNAP does not recognize efficiently ssoA (crossed), then only when the entering ssoU is properly folded may the synthesis of the pRNA take place.

Defects in sso functionality lead to reduced mobilization frequency. To analyze the usage of the sso's during conjugative mobilization, each of the plasmids (pMV158wt and its derivatives) were next transferred to competent S. pneumoniae 708 cells carrying pAMβ1 (the plasmid providing the auxiliary functions for conjugation), and intra- and interspecific conjugal transfer assays were performed using S. pneumoniae MP3008 (Nov^r) or E. faecalis OG1RF as recipients. To have several transfer experiments with high reproducibility, we made use of a recently developed multiwell plate setup coupled with a filter device so that up to eight transfers per plasmid and per donor-recipient cells were done simultaneously (33). The results obtained have a high confidence index and a low experimental standard deviation (usually below 10%). When intraspecies transfers were assayed between S. pneumoniae strains, it was evident that plasmids lacking both sso's exhibited a near-150-fold reduction in the conjugation frequencies (Fig. 2). No significant reduction in transfer frequencies were found for plasmids impaired in the functionality of either ssoA or ssoU (Fig. 2). These findings indicate that in the pneumococcal host, both pMV158-ssos were functionally replaceable, and that a 205-bp deletion encompassing most of the ssoU did not affect the functionality of the pMV158 transfer module.

View larger version (26K): [in this window] [in a new window]

FIG. 2. Transfer frequencies of pMV158 and derivatives from S. pneumoniae harboring pAMβ1 as donor cells to S. pneumoniae and to E. faecalis as recipient cells. Donor and recipient cells were incubated together on nitrocellulose filters (0.22-µm pore size) at a ratio 1:5, for 4 h at 37°C. The transfer frequencies (plotted in logarithmic scale) were calculated as the number of transconjugants CFU per ml of recipient cells. The bar above each column indicates the standard error of eight independent experiments in a 96-well filter plate. The reduction in transfer frequencies of plasmids with mutations in ssoU relative to frequencies of pMV158wt from S. pneumoniae to E. faecalis were statistically significant (P < 0.001). The numbers of colonies of S. pneumoniae MP3008 and E. faecalis OG1RF recipients were 5.5 x 108 and 3.2 x 108 CFU per filter, respectively.

In interspecies transfer experiments from pneumococci to E. faecalis, we took advantage of this bacteria being aerophilic. Since pneumococcus is unable to grow on the surface of agar plates (being a microaerophilic bacteria), there was no need to apply selection for the recipients other than growing them on the surface of the plates, a strategy that has proven to be useful in plasmid transfer from S. pneumoniae to aerobic bacteria (12, 33). In the case of pMV158wt, the frequencies of transfer were similar to or even greater than, i.e., ca. 10^–4 transconjugants of recipient cells per ml (Fig. 2), those observed for transfer between pneumococci. This is around the maximum value ever attained for pMV158 transfers (11, 37, 43). The values obtained for pMV158wt were nearly identical to those observed when the plasmid carried a nonfunctional ssoA (pMV158G3G7; Fig. 2). Other mutations in the ssoA were also tested (i.e., mutations in the CS-6 or in the RS_B), and the results obtained did not differ from the ones obtained for pMV158G3G7 (not shown). However, there was a strong 30-fold reduction in the transfer frequency when the plasmid had a defective ssoU (pMV158BD) with a further decrease (nearly 60-fold) when the plasmid tested lacked both origins (pMV158G3G7BD; Fig. 2). We conclude that, in E. faecalis, there is a strong preference for the use of ssoU as the plasmid lagging-strand replication signal. Although the copy number of pMV158wt was lower in E. faecalis than in S. pneumoniae, the plasmid was segregationally stable in the former host, since 100% of the cells retained the plasmid after 100 generations in the absence of selective pressure (data not shown). Further, the ssDNA/dsDNA ratio, which indicates efficient replication, was the same for the two bacterial hosts tested here (Table 2).

View this table: [in this window] [in a new window]

TABLE 2. Copy number, molecular ssDNA/dsDNA ratios, and ACa

Intracellular accumulation of ssDNA depends on the plasmid sso-host interactions. The reduction in the conjugation efficiency when ssoU-deficient plasmids were mobilized to E. faecalis indicates that, in contrast to mobilization to S. pneumoniae, the ssoU origin may have a critical role. If this were the case, we would expect ssoU-deficient plasmids in E. faecalis to generate large amounts of ssDNA intermediates. To test this hypothesis, colonies of transconjugants harboring pMV158wt or its derivatives were selected and grown for 30 generations (the minimum period for a colony to become a full-grown liquid culture [9]). Total DNA was prepared, and the different DNA forms were separated by agarose gel electrophoresis in the presence of ethidium bromide. Gels were recorded (gels marked as "L" in Fig. 3 and 4), and the DNA was transferred to filters and analyzed by Southern hybridization. The plasmid DNA forms bound to the membranes showed that ssDNA intermediates had a higher electrophoretic mobility than supercoiled circular covalently closed (ccc) monomeric forms ("R" gels in Fig. 3 and 4). In the case of S. pneumoniae, transconjugants harboring plasmids with either sso intact accumulated ssDNA intermediates in amounts similar to pMV158wt, which were very low and only detectable after long exposures (Fig. 3A, R gels). These findings demonstrate that both origins were equally functional in S. pneumoniae, in agreement with the conjugation frequencies observed (Fig. 2), allowing us to conclude that both lagging-strand origins supported postconjugative conversion of ssDNA to dsDNA in S. pneumoniae with similar levels of efficiency. However, cells harboring pMV158G3G7BD exhibited a 3 to 4-fold reduction in copy number (from 30 to 8 to 10 copies per genomic equivalent, see below) and the ss/ds DNA ratios increased from nearly undetectable for pMV158wt to large amounts (Fig. 3B, R gels). A very different picture was observed for E. faecalis transconjugants (Fig. 4). In this case, it was apparent that cells harboring pMV158BD (defective in ssoU) accumulated substantially increased levels of ssDNA (Fig. 4A, R gels). This amount was similar to that found for plasmids with both origins inactivated (pMV158G3G7BD; Fig. 4B, R gels), indicating that the ssoU was the only highly efficient functional origin in the enterococcal host.

View larger version (63K): [in this window] [in a new window]

FIG. 3. Accumulation of intracellular ssDNA intermediates in S. pneumoniae MP3008 carrying pMV158 wild-type (wt) or derivatives with mutations in ssoA (G3G7) or in ssoU (BD) (A) and mutations in both origins (B). Total DNA was prepared from plasmid-containing pneumococcal cultures and the different forms were separated by electrophoresis on 0.7% agarose gels in 1x Tris-borate-EDTA buffer with 0.5 µg of ethidium bromide/ml (L, left). After denaturation, the DNA was transferred to nylon membranes and hybridized with a 32P-labeled probe (R, right). The various DNA forms are indicated: chromosomal (chr), open circle (oc), circular covalently closed supercoiled (ccc), and ssDNA intermediates.

View larger version (58K): [in this window] [in a new window]

FIG. 4. Intracellular ssDNA accumulated in E. faecalis OG1RF transconjugants harboring wild-type or mutations in ssoA (G3G7) or in ssoU (A) and mutations in both origins (B). Total DNA prepared from enterococcal cells was treated as in Fig. 3.

Comparative quantifications of plasmid copy numbers and ssDNA/dsDNA ratios were calculated from determination of the plasmid copy numbers and of the radioactivity counted in the different plasmid bands. A parameter, termed the accumulation coefficient (AC), was introduced to calculate the relationship between the ssDNA/dsDNA ratios of the different sso mutants with respect to the ssDNA/dsDNA ratio of pMV158wt (Table 2). In the pneumococcal transconjugants, a fourfold reduction in plasmid copy number was found only for pMV158G3G7BD concomitant with a strong (40-fold) increase in the AC ratio. No significant variations in copy numbers were observed for plasmids defective in either sso, although a slight increase in the AC ratios was detected (Table 2). In the case of the E. faecalis transconjugants, the eightfold reduction in plasmid copy numbers was greater than the values determined for pneumococci, although the number of copies of pMV158wt was lower in enterococci (around 17 copies per genome equivalent) than in pneumococci (around 30 copies). Furthermore, a severe drop in copy numbers (from 17 to 4) was measured for plasmid pMV158BD, with a further twofold reduction when neither plasmid origin was functional (pMV158G3G7BD; Table 2). Decreases in the AC ratios show that ssoA might still be partially functional (AC ratio of 2 for pMV158G3G7) but not so for ssoU (AC ratio of 14 for plasmid pMV158BD). The null mutant for both sso's (pMV158G3G7BD) exhibited a further twofold reduction in the AC ratio. These results are consistent with the observed drop in the conjugation frequencies from S. pneumoniae to enterococcal cells (Fig. 2). From these results we may draw the following conclusions: (i) in S. pneumoniae, both ssoA and ssoU are equally functional and can replace each other, and (ii) in E. faecalis, ssoU is the main origin used by the plasmid, but there is a partial functionality of ssoA, as shown by further reductions in copy numbers and AC coefficients when the plasmid bears mutations in both origins.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Our understanding of a successful conjugative transfer of pMV158 not only contemplates that the transferring ssDNA molecule (the T-DNA) physically penetrates the recipient cell but also implies an efficient establishment in the new host. In this sense, after recircularization of the T-DNA by the strand transfer activity of the MobM relaxase, efficient conversion of ssDNA to dsDNA would be critical for the process to reconstitute the pMV158 plasmid that would undergo vegetative replication, repopulation of plasmid molecules, and antibiotic resistance expression in the recipient cells. Thus, the sso's would be the first elements participating into the transfer process within the recipients. To analyze the sso functionality on replicative transfer and/or on vegetative replication, we have taken into account three parameters: (i) plasmid transfer frequencies, (ii) copy numbers, and (iii) ssDNA accumulation of the wild-type and derivatives of pMV158. From the results obtained we can conclude that both pMV158-ssos could support "postconjugative" conversion of ssDNA to dsDNA in S. pneumoniae but not in E. faecalis. Here, we have conscientiously used the term "postconjugative" to include both events, conjugative and vegetative lagging-strand replication, because they are mechanistically identical. In the case of transfer between pneumococci, either origin was functional, whereas deletion of ssoU in plasmids with an intact ssoA (pMV158BD) resulted in a 60-fold reduction in the interspecies transfer frequency. The ssoA origins seem to function efficiently only in the plasmid natural host, and it was hypothesized that specific host factors may be required for the ssoA functionality present in their native hosts and the absence of such factors would be responsible for their poor functionality in heterologous hosts (9). However, while pMV158-ssoA was not functional in S. aureus cells, it supported lagging-strand synthesis on staphylococcal cell extracts, pointing to a problem of efficiency rather than to a lack of specificity of the initiation process (25, 26). Thus, an efficient RNAP-ssoA interaction could be a determinant of the host specific functionality of the ssoA-type origins. In the case of ssoU, it would appear that its ability to interact efficiently with RNAPs from various hosts could provide these plasmids with an expanded host range. The mobilization cassette (oriT and gene mobM) of the pMV158 derivatives used here was intact. If transfer is independent of plasmid copies in the donor, then the number of mobilization events from S. pneumoniae donors to the recipient cells should be similar, independent of the activity of the sso. The transferred molecules (as ssDNA) of a plasmid lacking a functional sso would not support conversion to dsDNA or would do so very poorly, thus compromising its establishment in the new host because of its low copy number prior to cell division, since the distribution of RCR plasmid copies is based on random events (8). Alternatively, it could be that plasmid copy number in the donors may affect the number of transfer events. If this were the case, plasmids replicating with a similar efficiency in S. pneumoniae, such as pMV158BD and pMV158wt (Table 2), should maintain a number of transfer events comparable to those of the recipients. However, the frequency of transfer of pMV158BD to E. faecalis was significantly lower than those of pMV158wt and pMV158G3G7 (Fig. 2), demonstrating that the absence of the ssoU signal is essential for a successful conjugation. Plasmid pMV158G3G7BD (with nonfunctional sso) transferred very poorly and accumulated ssDNA (30- to 40-fold) in both E. faecalis and S. pneumoniae. Thus, it appears that, whereas pneumococcal intraspecies transfer was equally efficient, provided that the plasmid bears a functional sso, it was not the case for interspecies transfer from pneumococci to enterococci, where a strong dependence of an intact ssoU is essential for pMV158 propagation. Since the ssoU functions efficiently in both hosts, this origin seems to be an important determinant for the promiscuity of pMV158. However, other factors may contribute to the extraordinary host range of this plasmid (it has been established in more than 20 different hosts thus far). Stable inheritance of a plasmid after colonizing a new host does not necessarily need a functional lagging-strand origin (19), so that establishment of a plasmid bearing a functional ssoU does not seem to be enough for productive replication in S. pneumoniae. This view is supported by the following: (i) a derivative of plasmid pVA380-1 (isolated from Streptococcus ferus and carrying a kanamycin gene [30, 31]) could be easily established in pneumococci, where it stably replicates with a high copy number (15); and (ii) attempts to transfer the staphylococcal RCR plasmid pUB110 (which bears an ssoU identical to that of pMV158, a highly homologous mob cassette and a kanamycin resistance gene) to S. pneumoniae, have failed (our unpublished observations). Thus, we have invoked the fitness and/or adaptation of the bacterium-plasmid pair as one of the main reasons for the plasmid broad host range (6).

The current model for conjugation predicts that, at least in small RCR plasmids such as pMV158, reconstitution of a dsDNA plasmid after transfer takes place by synthesis of a pRNA starting at one of the plasmid sso's by a mechanism equivalent to the vegetative plasmid lagging-strand replication (Fig. 5). Therefore, we could predict that plasmids with nonfunctional sso's for vegetative replication should not be efficient in conjugative transfer and vice versa. The results obtained for the plasmid lacking both sso's (pMV158G3G7BD) corroborated this assumption, since vegetative replication and conjugative transfer were negatively affected, independently of the host tested. If pMV158 colonizes a new pneumococcal host by the transformation of competent cells, its DNA should be taken up as ssDNA segments by the degradative process of DNA transport (32). Thus, intracellular reconstitution of an intact plasmid molecule by DNA synthesis and/or recombination is followed by vegetative leading-strand replication from the dso by the RCR mechanism. Entry of plasmid DNA by electrotransformation (in a dsDNA conformation), as in E. faecalis, would also result in vegetative leading-strand replication. However, colonization of a new host by conjugation would imply Mob-mediated closing of a full-length ssDNA intermediate followed by (or simultaneous with) lagging-strand (postconjugative) replication from the sso. If this kind of replication were independent of the sso efficiency, we should expect the same conjugation frequencies for plasmids with or without defects in the sso. However, no increased numbers of transconjugants were obtained in any case by prolonged incubation times, which mirrors a direct effect of ssoU functionality on plasmid transfer. In addition, we have analyzed the transformation efficiency by electroporation (where plasmid DNA enters as a double strand) of E. faecalis OG1RF with pMV158 and sso mutants. The results showed no differences in the number transformed colonies recovered with pMV158BD and pMV158, although the colony size of the former was smaller and the plasmid copy number was lower, as expected. Thus, it could be envisaged that the role of the sso on plasmid establishment may be more relevant in the first stages of plasmid colonization of a new host via conjugative transfer.

 
We thank A. Bravo for critical reading of the manuscript and members of Espinosa's lab for helpful discussions.

This research was financed by grants BFU2007-63575 and CSD2008-00013-INTERMODS from the Spanish Ministry of Science and Innovation and EU-CP223111 (CAREPNEUMO, European Union). F.L.-D. was the recipient of a fellowship (BF03/00529) from the Carlos III Health Institute (Spain).

 
* Corresponding author. Mailing address: Department of Protein Science, Centro de Investigaciones Biológicas, Consejo Superior de Investigaciones Científicas, Ramiro de Maeztu 9, E-28040 Madrid, Spain. Phone: (34) 918373112. Fax: (34) 915360432. E-mail: mespinosa{at}cib.csic.es

Published ahead of print on 21 November 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Bates, S., A. Roscoe, N. Althorpe, W. Brammar, and M. Wilkins. 1999. Expression of leading region genes on Incl1 plasmid Col1b-P9: genetic evidence for single-stranded DNA transcription. Microbiology 145:2655-2662.[Abstract/Free Full Text]
2
2. Burdett, V. 1980. Identification of tetracycline-resistant R-plasmids in Streptococcus agalactiae (group B). Antimicrob. Agents Chemother. 18:753-760.[Abstract/Free Full Text]
3
3. Caryl, J. A., M. C. A. Smith, and C. D. Thomas. 2004. Reconstitution of a staphylococcal plasmid-protein relaxation complex in vitro. J. Bacteriol. 186:3374-3383.[Abstract/Free Full Text]
4
4. Davydova, E. K., T. J. Santangelo, and L. B. Rothman-Denes. 2007. Bacteriophage N4 virion RNA polymerase interaction with its promoter DNA hairpin. Proc. Natl. Acad. Sci. USA 104:7033-7038.[Abstract/Free Full Text]
5
5. de Antonio, C., M. E. Farias, M. G. de Lacoba, and M. Espinosa. 2004. Features of the plasmid pMV158-encoded MobM, a protein involved in its mobilization. J. Mol. Biol. 335:733-743.[CrossRef][Medline]
6
6. del Solar, G., J. C. Alonso, M. Espinosa, and R. Díaz-Orejas. 1996. Broad host range plasmid replication: an open question. Mol. Microbiol. 21:661-666.[CrossRef][Medline]
7
7. del Solar, G., R. Díaz, and M. Espinosa. 1987. Replication of the streptococcal plasmid pMV158 and derivatives in cell-free extracts of Escherichia coli. Mol. Gen. Genet. 206:428-435.[CrossRef][Medline]
8
8. del Solar, G., R. Giraldo, M. J. Ruiz-Echevarría, M. Espinosa, and R. Díaz-Orejas. 1998. Replication and control of circular bacterial plasmids. Microbiol. Mol. Biol. Rev. 62:434-464.[Abstract/Free Full Text]
9
9. del Solar, G., A. Puyet, and M. Espinosa. 1987. Initiation signals for the conversion of single stranded to double-stranded DNA forms in the streptococcal plasmid pLS1. Nucleic Acids Res. 15:5561-5580.[Abstract/Free Full Text]
10
10. Espinosa, M., P. López, M. T. Pérez-Ureña, and S. A. Lacks. 1982. Interspecific plasmid transfer between Streptococcus pneumoniae and Bacillus subtilis. Mol. Gen. Genet. 188:195-201.[CrossRef][Medline]
11
11. Farías, M. E., and M. Espinosa. 2000. Conjugal transfer of plasmid pMV158: uncoupling of the pMV158 origin of transfer from the mobilization gene mobM, and modulation of pMV158 transfer in Escherichia coli mediated by IncP plasmids. Microbiology 146:2259-2265.[Abstract/Free Full Text]
12
12. Farías, M. E., E. Grohmann, and M. Espinosa. 1999. Expression of the mobM gene of the streptococcal plasmid pMV158 in Lactococcus lactis subsp. lactis. FEMS Microbiol. Lett. 176:403-410.[Medline]
13
13. Francia, M. V., A. Varsaki, M. P. Garcillan-Barcia, A. Latorre, C. Drainas, and F. de la Cruz. 2004. A classification scheme for mobilization regions of bacterial plasmids. FEMS Microbiol. Rev. 28:79-100.[CrossRef][Medline]
14
14. Glucksmann-Kuis, M. A., P. Makiewicz, C. Malone, and L. B. Rothman-Denes. 1992. Specific sequences and a hairpin structure in the template strand are required for N4 virion RNA polymerase promoter recognition. Cell 70:491-500.[CrossRef][Medline]
15
15. Grohmann, E., L. M. Guzmán, and M. Espinosa. 1999. Mobilisation of the streptococcal plasmid pMV158: interactions of MobM protein with its cognate oriT DNA region. Mol. Gen. Genet. 261:707-715.[CrossRef][Medline]
16
16. Gruss, A. D., H. F. Ross, and R. P. Novick. 1987. Functional analysis of a palindromic sequence required for normal replication of several staphylococcal plasmids. Proc. Natl. Acad. Sci. USA 84:2165-2169.[Abstract/Free Full Text]
17
17. Guzmán, L., and M. Espinosa. 1997. The mobilization protein, MobM, of the streptococcal plasmid pMV158 specifically cleaves supercoiled DNA at the plasmid oriT. J. Mol. Biol. 266:688-702.[CrossRef][Medline]
18
18. Haynes, L. L., and L. B. Rothman-Denes. 1985. N4 virion RNA polymerase sites of transcription initiation. Cell 41:597-605.[CrossRef][Medline]
19
19. Hernández-Arriaga, A. M., M. Espinosa, and G. del Solar. 2000. A functional lagging strand origin does not stabilize plasmid pMV158 inheritance in Escherichia coli. Plasmid 43:49-58.[CrossRef][Medline]
20
20. Khan, S. A. 2005. Plasmid rolling-circle replication: highlights of two decades of research. Plasmid 53:126-136.[CrossRef][Medline]
21
21. Khan, S. A. 2000. Plasmid rolling-circle replication: recent developments. Mol. Microbiol. 37:477-484.[CrossRef][Medline]
22
22. Koepsel, R. R., R. W. Murray, W. D. Rosenblum, and S. A. Khan. 1985. The replication initiator protein of plasmid pT181 has sequence-specific endonuclease and topoisomerase-like activities. Proc. Natl. Acad. Sci. USA 82:6845-6849.[Abstract/Free Full Text]
23
23. Kramer, M. G., G. del Solar, and M. Espinosa. 1995. Lagging-strand origins of the promiscuous plasmid pMV158: physical and functional characterization. Microbiology 141:655-662.[Abstract/Free Full Text]
24
24. Kramer, M. G., M. Espinosa, T. K. Misra, and S. A. Khan. 1999. Characterization of a single-strand origin, ssoU, required for broad host range replication of rolling-circle plasmids. Mol. Microbiol. 33:466-475.[CrossRef][Medline]
25
25. Kramer, M. G., M. Espinosa, T. K. Misra, and S. A. Khan. 1998. Lagging strand replication of rolling-circle plasmids: specific recognition of the ssoA-type origins in different gram-positive bacteria. Proc. Natl. Acad. Sci. USA 95:10505-10510.[Abstract/Free Full Text]
26
26. Kramer, M. G., S. A. Khan, and M. Espinosa. 1998. Lagging-strand replication from the ssoA origin of plasmid pMV158 in Streptococcus pneumoniae: in vivo and in vitro influences of mutations in two conserved ssoA regions. J. Bacteriol. 180:83-89.[Abstract/Free Full Text]
27
27. Kramer, M. G., S. A. Khan, and M. Espinosa. 1997. Plasmid rolling circle replication: identification of the RNA polymerase-directed primer RNA and requirement of DNA polymerase I for lagging strand initiation. EMBO J. 16:5784-5795.[CrossRef][Medline]
28
28. Lacks, S. A., P. López, B. Greenberg, and M. Espinosa. 1986. Identification and analysis of genes for tetracycline resistance and replication functions in the broad-host-range plasmid pLS1. J. Mol. Biol. 192:753-765.[CrossRef][Medline]
29
29. Lanka, E., and B. M. Wilkins. 1995. DNA processing reactions in bacterial conjugation. Annu. Rev. Biochem. 64:141-169.[CrossRef][Medline]
30
30. Le Blanc, D. J., L. N. Lee, and A. Abu-Al-Jaibat. 1992. Molecular, genetic, and functional analysis of the basic replicon of pVA380-1, a plasmid of oral streptococcal origin. Plasmid 28:130-145.[CrossRef][Medline]
31
31. Le Blanc, D. J., Y. Y. M. Chen, and L. N. Lee. 1993. Identification and characterization of a mobilization gene in the streptococcal plasmid pVA380-1. Plasmid 30:296-302.[CrossRef][Medline]
32
32. López, P., M. Espinosa, D. L. Stassi, and S. A. Lacks. 1982. Facilitation of plasmid transfer in Streptococcus pneumoniae by chromosomal homology. J. Bacteriol. 150:692-701.[Abstract/Free Full Text]
33
33. Lorenzo-Díaz, F., and M. Espinosa. 2009. Large-scale filter mating assay for intra- and inter-specific conjugal plasmid transfer in Gram-positive bacteria. Plasmid 61:65-70.[CrossRef][Medline]
34
34. Masai, H., and K. Arai. 1997. Frpo: a novel single-stranded DNA promoter for transcription and for primer RNA synthesis of DNA replication. Cell 89:897-907.[CrossRef][Medline]
35
35. Nasim, M. T., I. C. Eperon, B. M. Wilkins, and W. J. Brammar. 2004. The activity of a single-stranded promoter of plasmid ColIb-P9 depends on its secondary structure. Mol. Microbiol. 53:405-417.[CrossRef][Medline]
36
36. Novick, R. P. 1998. Contrasting lifestyles of rolling-circle phages and plasmids. Trends Biochem. Sci. 23:434-438.[CrossRef][Medline]
37
37. Priebe, S. D., and S. A. Lacks. 1989. Region of the streptococcal plasmid pMV158 required for conjugative mobilization. J. Bacteriol. 171:4778-4784.[Abstract/Free Full Text]
38
38. Projan, S. J., and R. P. Novick. 1988. Comparative analysis of five related staphylococcal plasmids. Plasmid 19:203-221.[CrossRef][Medline]
39
39. Smith, M. C. A., and C. D. Thomas. 2004. An accessory protein is required for relaxosome formation by small staphylococcal plasmids. J. Bacteriol. 186:3363-3373.[Abstract/Free Full Text]
40
40. Stassi, D. L., P. Lopez, M. Espinosa, and S. A. Lacks. 1982. Cloning of chromosomal genes in Streptococcus pneumoniae. Proc. Natl. Acad. Sci. USA 78:7028-7032.[CrossRef]
41
41. te Riele, H., B. Michel, and S. D. Ehrlich. 1986. Are single-stranded circles intermediates in plasmid DNA replication? EMBO J. 5:631-637.[Medline]
42
42. Thomas, C. M. 2000. The horizontal gene pool. Harwood Academic Publishers, Amsterdam, The Netherlands.
43
43. van der Lelie, D., H. A. B. Wosten, S. Bron, L. Oskam, and G. Venema. 1990. Conjugal mobilization of streptococcal plasmid pMV158 between strains of Lactococcus lactis subsp. lactis. J. Bacteriol. 172:47-52.[Abstract/Free Full Text]

---

Journal of Bacteriology, February 2009, p. 720-727, Vol. 191, No. 3
0021-9193/09/$08.00+0     doi:10.1128/JB.01257-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lorenzo-Díaz, F. Articles by Espinosa, M. Search for Related Content PubMed PubMed Citation Articles by Lorenzo-Díaz, F. Articles by Espinosa, M.