The replication origin of a repABC plasmid
© Cervantes-Rivera et al; licensee BioMed Central Ltd. 2011
Received: 15 April 2011
Accepted: 30 June 2011
Published: 30 June 2011
Skip to main content
© Cervantes-Rivera et al; licensee BioMed Central Ltd. 2011
Received: 15 April 2011
Accepted: 30 June 2011
Published: 30 June 2011
repABC operons are present on large, low copy-number plasmids and on some secondary chromosomes in at least 19 α-proteobacterial genera, and are responsible for the replication and segregation properties of these replicons. These operons consist, with some variations, of three genes: repA, repB, and repC. RepA and RepB are involved in plasmid partitioning and in the negative regulation of their own transcription, and RepC is the limiting factor for replication. An antisense RNA encoded between the repB-repC genes modulates repC expression.
To identify the minimal region of the Rhizobium etli p42d plasmid that is capable of autonomous replication, we amplified different regions of the repABC operon using PCR and cloned the regions into a suicide vector. The resulting vectors were then introduced into R. etli strains that did or did not contain p42d. The minimal replicon consisted of a repC open reading frame under the control of a constitutive promoter with a Shine-Dalgarno sequence that we designed. A sequence analysis of repC revealed the presence of a large A+T-rich region but no iterons or DnaA boxes. Silent mutations that modified the A+T content of this region eliminated the replication capability of the plasmid. The minimal replicon could not be introduced into R. etli strain containing p42d, but similar constructs that carried repC from Sinorhizobium meliloti pSymA or the linear chromosome of Agrobacterium tumefaciens replicated in the presence or absence of p42d, indicating that RepC is an incompatibility factor. A hybrid gene construct expressing a RepC protein with the first 362 amino acid residues from p42d RepC and the last 39 amino acid residues of RepC from SymA was able to replicate in the presence of p42d.
RepC is the only element encoded in the repABC operon of the R. etli p42d plasmid that is necessary and sufficient for plasmid replication and is probably the initiator protein. The oriV of this plasmid resides within the repC gene and is located close to or inside of a large A+T region. RepC can act as an incompatibility factor, and the last 39 amino acid residues of the carboxy-terminal region of this protein are involved in promoting this phenotype.
Proteins that are involved in the initiation of DNA replication are essential to cells. These proteins recognize the origin of replication, destabilize double-stranded DNA, and recruit the replisome, which is the machinery directly involved in DNA replication .
Both the activity and concentration of the initiator proteins are highly regulated because the genetic material needs to be replicated only once per generation. A failure in this process could accelerate the production of new DNA molecules with a concomitant increase in the number of new origins of replication, which could be used in new rounds of replication and leading to cell death (i.e., "runaway replication") .
Initiator proteins control the replication rate using several mechanisms that limit either their own synthesis or their availability. The initiator proteins can directly auto-regulate the transcription of their own genes or trigger the production of negative regulators, antisense-RNAs or proteins, which are co-transcribed with the initiator genes. The activity of the initiator proteins can be controlled by covalent modifications or by titrating out their availability using DNA sites that resemble origins of replication. In addition, the DNA initiation rate can be controlled by blocking or hiding the origins of replication [3, 4].
The initiation of replication of the Escherichia coli chromosome and of some of its plasmids has been studied extensively. However, our knowledge of other bacterial replication systems is limited. Research on new replicons that are not found in E. coli or its close relatives would yield new insights into the regulation of initiation of replication in bacteria. The present work concerns repABC replicons, which are present on large, low copy-number plasmids and on some secondary chromosomes in at least 19 α-proteobacterial genera. Some bacterial strains contain more than one repABC replicon, indicating that this plasmid family encompasses several incompatibility groups [5–7].
The basic replicon of repABC plasmids is compact because all of the elements required for replication and segregation are encoded in a single operon, the repABC operon [8, 9]. However, this operon is controlled by a complex regulatory mechanism. The first two genes of the repABC operon encode for proteins belonging to a type Ia segregation system . RepA and RepB have been implicated in the negative transcriptional regulation of the repABC operon [9, 11].
RepC is a limiting replication factor and thus has been suggested to be the initiator protein [8, 12, 13]. The members of the repABC family contain a centromeric-like sequence (parS) in three possible locations: downstream of and close to the stop codon of repC [14, 15], between repA and repB, or upstream of repA [16, 17]. A conserved sequence between the repB and repC genes is present in all known repABC replicons and contains an antisense RNA (ctRNA) gene, the product of which negatively modulates the expression of RepC [18–20]. Regulatory role of the ctRNA depends on its pairing with the repABC mRNA. In the absence of the ctRNA, the mRNA section corresponding to the repB-repC intergenic region folds into a large stem-loop structure so that the predicted repC Shine-Dalgarno (SD) sequence and the repC initiation codon remain single-stranded, allowing repC translation. In contrast, when the ctRNA hybridizes with the repABC mRNA, the repC leader sequence forms an intrinsic terminator, blocking repC transcription .
Many aspects of the biology of these plasmids remain unknown, especially the details of the replication or segregation of these genetic elements. In this paper, we demonstrate the following: A) RepC is the only element encoded in the repABC operon of the Rhizobium etli p42d plasmid (formally pRetCFN42d) that is necessary and sufficient for plasmid replication. B) RepC is an incompatibility factor. C) The RepC carboxy-terminal region is involved in the incompatibility phenotype. D) The origin of replication of the repABC plasmid resides in a large A+T-rich region located at the central section of the repC gene.
Bacterial strains and plasmid used in this work
Rhizobium etli CE3
Streptomycin resistant derivative of CFN42 strain
R. etli CFNX101
recA::Ω-Spectinomycin derivative of CE3
R. etli CFNX107
recA:: Ω-Spectinomycin derivative of CE3, laking plasmid p42a and p42d.
E. coli S17-1
Plasmid donor in conjugations
A chloramphenicol resistant suicide vector derived from pBC SK(+), and containing oriT
pDOP derivative with the intergenic region repB-repC, the complete repC gene under Placpromoter, and 500 pb downstream repC stop codon.
pDOP derivative carrying a 5.6 Kb HindIII with repABC operon of R. etli plasmid p42d.
pDOP derivative with the intergenic region repB-repC and the complete repC gene under Plac promoter.
pDOP carrying repC gen of plasmid p42d, with a SD sequence (AGGA) and under Plac promoter.
Similar to pDOP-C but with a repC gene carrying a deletion from codon 2 to codon 29
Similar to pDOP-C but with a repC gene carrying a deletion from codon 372 to codon 401
pDOP containing a repC fragment from codon 2 to codon 110, with a SD consensus sequence under Plac promoter.
pDOP containing a repC fragment from codon 2 to codon 209, with a SD consensus sequence under Plac promoter.
pDOP containing a repC fragment from codon 2 to codon 309, with a SD consensus sequence under Plac promoter.
pDOP containing a repC fragment from codon 310 to codon 403, with a SD consensus sequence under Plac promoter.
pDOP containing a repC fragment from codon 210 to codon 403, with a SD consensus sequence under Plac promoter.
pDOP containing a repC fragment from codon 111 to codon 403, with a SD consensus sequence under Plac promoter.
Similar to pDOP-C but without the SD sequence
Similar to pDOP-C but with a mutant repC gene carrying
silent mutations to increase its CG content
Similar to pDOP-C but with repC gene, carrying a frameshift mutation at the BglII restriction site
Similar to pDOP-C but with repC gene, carrying a frameshift mutation at the SphI restriction site
pDOP derivative carrying repC gen of the Agrobacterium
tumefaciens C58 linear chromosome, with a SD sequence (AGGA) and under Plac promoter.
pDOP derivative carrying repC gen of the Sinorhizobium meliloti 1021 pSymA, with a SD sequence (AGGA) and under Plac promoter.
pDOP with a hybrid repC gene, encoding the first 140 amino acid residues of the pSymA RepC protein and the rest of p42d.
pDOP with a hybrid repC gene, encoding the first 140 amino acid residues of the p42d RepC protein and the rest of pSymA.
pDOP with a hybrid repC gene encoding, the first 140 amino acid residues of the pSymA RepC protein, the next 140 amino acid residues from the p42d RepC protein and the rest from the pSymA RepC protein.
pDOP witha hybrid repC gene, encoding the first 140 amino acid residues of the p42d RepC protein, the next 140 amino acid residues from the pSymA RepC protein and the rest from the p42d RepC protein.
pDOP derivative with a hybrid repC gene, encoding the first 280 aminoacid residued of pSymA RepC and the rest of p42d RepC protein.
pDOP derivative with a hybrid repC gene, encoding the first 330 amino acid residues of p42d RepC protein and the rest of pSymA RepC protein.
pDOP derivative with a hybrid repC gene, encoding the first 362 amino acid residues of p42d RepC protein and the rest of pSymA RepC protein.
pDOP derivatives were introduced into Rhizobium by conjugation using E. coli S17-1 as a donor strain . The strains were grown in the proper antibiotic-free liquid medium to stationary phase, mixed in a donor-recipient ratio of 1:2 on antibiotic-free PY plates, and incubated at 30°C overnight. The cells were resuspended in PY medium, and serial dilutions were plated on the appropriate selective PY medium.
Oligonucleotides used in these work
Insert of plasmid pDOP-αC was generated by amplifying the incα-repC region with the primers ALFAU2 and Mal-C2Kpn. The repC (p42d) gene present in pDOP-C was amplified by PCR with the primers RBS-C and Mal-C2. The repC gene of pSymA, present in construct pDOP-CsA, was obtained by PCR with the primers C-SymA and K-Syma-L and the genomic DNA of S. meliloti 2011 as the template. The repC gene of the linear chromosome of Agrobacterium tumefaciens C58 was obtained by PCR with the primers repCATBamU and repCATHinL and genomic DNA as the template.
The insert of the plasmid pDOP-C/D1UM with a deletion in its 5'-end was obtained with the oligonucleotides repC-D1U and Mal-C2. The repC gene present in the plasmid pDOP-C/RD1L was amplified with the primers RBS-C and repC-D1L. Six plasmids carrying fragments of the repC gene were constructed: pDOP-C/F1 insert was obtained with primers repC-F1U and repC-F1L. The insert of plasmid pDOP-C/F1-F2 was obtained with primers repC-F1U and repC-F2L. Inserts of plasmids pDOP-C/F1-F3, pDOP-C/F4, pDOP-C/F4-F3, and pDOP-C/F4-F2 were obtained with the following primer pairs: repC-F1U and repC-F3L, repC-F4U and repC-F4L, repC-F3U and repC-F4L, repC-F2U and repC-F4L, respectively.
The insert of the plasmid pDOP-Cs/SD was acquired by PCR with the primers repCd-sSDU and Mal-C2.
Plasmid pDOP-CBglII, was constructed digesting pDOP-C with BglII and filling in 5'-overhangs with T4 DNA polymerase (Fermentas). The blunted plasmid was ligated again with T4 ligase (Fermentas). Plasmid pDOP-CSphI was constructed in a similar way but digesting pDOP-C with SphI.
The pDOP-TtMC insert was obtained by an overlap extension PCR as described by Horton et al. (1989) . The first PCR was performed using the primers Ttrack1-U and Mal-C2Kpn, and pH3 DNA as initial template. This product was purified and used as template for a second PCR with the oligonucleotides Mal-C2Kpn and Ttrack2-U; the amplification product was named T2-U. A third PCR amplification product obtained with the primers RBS-C and Ttrack1-L, and pH3 DNA as the template, was purified and used as a template in a new PCR reaction with the primers RBS-C and Ttrack2-L. The amplification product was named T2-L.
Finally, PCR products T2-U and T2-L were then mixed and used as the template for the last PCR. In this reaction, the primers Mal-C2Kpn and RBS-C were used, and the final PCR product was cloned into pDOP.
Overlap extension PCR was also employed to obtain repC hybrid genes. RepC gene amplification products from pSymA were obtained using pDOP-CsA as the template, and the repC p42d products were obtained using pH3 as the template. Most of the hybrid genes described here required the overlap of two PCR products. The insert of plasmid pDOP/C420-1209 was obtained using the primers C-SymA and AL-2Uc for the first PCR product and AL-2U and Mal-C2 for the second product. The final PCR product was obtained with the external primers C-SymA and Mal-C2. The insert of plasmid pDOP/C1-420 was constructed with primers RBS-C and 1L-B2c and the primers 1L-B2 and K-SymAL for the first and second PCR products, respectively. These products were combined using the primers RBS-C and K-SymAL. The pDOP/C841-1209 insert was constructed with the primers C-SymA and BL-3Uc for the first PCR product and BL-3U and Mal-C2 for the second. These products were joined in a third PCR with the primers C-SymA and Mal-C2. The hybrid gene in pDOP/C1-990 was acquired with the primers RBS-C and Sal-CdL for the first PCR product and Sal-CdU and Mal-C2 for the second. These PCR products were integrated in a third PCR with the primers RBS-C and Mal-C2. Similarly, the hybrid gene of pDOP/C1-990 was obtained with the primers RBS-C and Cd-1086 for the first amplification product. To obtain the second PCR product, the primers Cs-1087U and Mal-C2 were used, and both PCR products were fused with the primers RBS-C and Mal-C2. The inserts of two of the constructs, pDOP/C421-840 and pDOP/Cs421-840, required the fusion of three PCR products. The hybrid gene located in pDOP/C421-840 required the primers C-SymA and AL-2Uc for the first PCR product, the primers AL-2U and AL-2Uc for the second PCR product, and the primers 2L-CU and K-SymA for the third PCR product. The three PCR products were fused in the final PCR with the primers C-SymA and K-SymA. The hybrid gene present in pDOP/Cs421-840 was obtained using the primers RBS-C and 1L-B2c for the first PCR product, the primers 1L-B2 and B2-3Uc for the second PCR product, and the primers BL-2U and Mal-C2 for the third PCR product. These PCR products were linked using the primers RBS-C and Mal-C2 in the final PCR. DNA sequences of the inserts of all constructs were obtained to corroborate their correctness.
The plasmid profiles of four transconjugants from each cross were visualized on agarose gels according to the protocol described by Hynes and McGregor .
Plasmid DNA was isolated using the High Pure Plasmid Isolation Kit (Roche) according to the manufacturer's instructions. Restriction and ligation reactions were performed under the conditions specified by the enzyme manufacturer (Fermentas). PCR was performed using Platinum High Fidelity Taq Platinum Polymerase or ThermalAce™ DNA Polymerase (Invitrogen). PCR products were cloned using the TOPO TA Cloning Kit (Invitrogen).
The incompatibility properties of the constructs were determined as described in Ramírez-Romero et al. .
To determine the replication capabilities of the pDOP derivatives in R. etli, the plasmids were introduced into CFNX107 by conjugation. The plasmid profiles of at least four transconjugants from each cross were analyzed. A recombinant plasmid was considered capable of replicating in R. etli if the plasmid profiles of the transconjugants showed a new band of the expected size.
The plasmid copy numbers of the CFNX107 transconjugants containing pDOP derivatives were evaluated as follows: the total DNA of each transconjugant was isolated, digested with HindIII endonuclease, resolved in a 1% agarose gel and transferred to Hybond-N+ membranes (Amersham). The blot was then simultaneously hybridized with an Ω- spectinomycin cassette located within the recA gene (chromosome-encoded) and with a fragment of pDOP; both probes were of the same size and GC content. The hybridization signals were quantified with a PhosphorImager SI (Molecular Dynamics). The plasmid copy-number was calculated from the ratio of the integrated hybridization signal of the recombinant plasmid and the integrated hybridization signal of the chromosome.
Alignments were performed with Clustal-W  at the WWW service of the European Bioinformatics Institute http://www2.ebi.ac.uk/clustalw. Protein secondary structure predictions were made with PSIPRED  at the WWW service of the Bioinformatics Group, UCL Department Of Computer Science http://bioinf.cs.ucl.ac.uk/psipred/. The DNA duplex helical stability profile was calculated using WEB-THERMODYN: sequence analysis software for profiling DNA helical stability http://www.gsa.buffalo.edu/dna/dk/WEBTHERMODYN/.
To identify the minimal region of p42d that is capable of independent replication (putting aside the properties of the parental plasmid), we further explored the region between the repB stop codon and the 500 bp downstream of the repC stop codon. Three PCR products that possessed parts of this region were amplified and cloned into pDOP, a mobilizable suicide vector, under the control of the Plac promoter, which behaves as a constitutive promoter in Rhizobium. The first construct (pDOP-αC) contained the repB-repC intergenic region (inc-alpha) and the complete repC gene. The second construct, pDOP-SDnC, contained the repC open reading frame (ORF), including its putative repC Shine-Dalgarno (SD) sequence (AGGUG). The third construct contained the repC ORF but with a SD sequence that was more similar to the Rhizobium etli SD consensus (AGGAA) positioned 6 bp prior to the repC initiation codon (pDOP-C). As a control, we introduced a HindIII fragment of 5.6 Kb that carried the entire repABC of p42d into pDOP conferring it the ability to replicate in Rhizobium (Figure 1) .
To prove that RepC is essential for replication, two repC deletions and two frame-shift mutants were constructed and cloned into pDOP under the control of the Plac promoter. Plasmid pDOP-C/D1UM contained a repC gene with a deletion of 14 codons (from codon 2 to 14), and plasmid pDOP-C/RD1L contained a repC gene with a deletion at its 3'end of 14 codons (from codon 388 to 401). The construct pDOP-CBglII possessed a repC gene with a frame-shift mutation at nucleotide 948, while plasmid pDOP-CSphI carried a frame-shift mutation at nucleotide 277. All of these constructs contained the same SD sequence as construct pDOP-C and were in the same relative orientation with respect to PLac in the vector. All plasmids were mated into the R. etli CFNX107 strain, but no transconjugants were obtained, indicating that the complete RepC product is crucial for replication.
To demonstrate that these observations were not specific to the p42d repC sequence, the repC genes of S. meliloti 1021 pSymA and the A. tumefaciens C58 linear chromosome were amplified by PCR and introduced into pDOP under Plac control and downstream of a SD sequence. The recombinant plasmids were conjugated into R. etli strain CFNX107, and the plasmid profiles of the transconjugants were analyzed. Both recombinant plasmids were capable of replication in Rhizobium, as was pDOP-C (Figure 2). These results clearly suggest that the presence of an origin of replication (oriV) within repC is a general property of repABC operons.
To circumscribe the origin of replication (oriV) of the repABC plasmids, we performed an in silico analysis to search for three sequence features that are characteristic of the oriV in low copy-number plasmids: a set of tandem direct repeat sequences (iterons), a region of high A+T content, and DnaA boxes. We only detected a region of high A+T content between positions 450 and 850 of the repC coding region. However, we did not find any trace of even highly degenerated direct repeat sequences or of DnaA boxes.
The identification of an oriV sequence is generally based on its ability to facilitate replication when present on a plasmid that otherwise could replicate only if the appropriate replication factors (e.g., an initiator protein) were provided in trans. To more precisely locate the oriV within repC, we cloned a collection of internal segments of repC into the suicide vector pDOP (Figure 1). This collection was conjugated into an R. etli strain containing the parental plasmid (CFNX101) as the source of all the trans elements required for replication, but we were unable to obtain transconjugants.
To determine if the activation of oriV requires transcription (i.e., the repC mRNA also acts as a replication primer), we constructed a pDOP derivative that contained a repC gene but lacked a SD sequence (pDOP-Cs/SD) (Figure 1). This plasmid was also incapable of replicating in R. etli CFNX101. Similarly, the two plasmids with repC frame-shift mutations, pDOP-CBglII and pDOP-CSphI, were also conjugated into R. etli CFNX101 without success. Overall, these results indicate that RepC exerts its action in cis.
Plasmid incompatibility, or the inability of two replicons to coexist in the same cell line, results from the sharing of elements involved in plasmid replication, partitioning or control . The repC open reading frame of p42d, when cloned in a vector capable of replicating in R. etli, CFNX101, can coexist with p42d . However, all of our attempts to introduce the construct pDOP-C into R. etli CFNX101 failed. In contrast, CFNX101 transconjugants carrying a similar construct (pDOP-CsA) that contained the repC gene pSymA of S. meliloti 2011 were easily obtained. The frequencies with which CFNX101/pDOP-CsymA and CFNX107/pDOP-CsymA transconjugants were obtained were similar (average 5 × 10-3). Moreover, the plasmid profiles of the transconjugants showed that pDOP-CsA replicated in these strains as an independent entity. These observations indicate that pDOP-C and its parental plasmid p42d are incompatible, while that of pDOP-CSymA and p42d are compatible.
Plasmids in which the oriV is located in the gene encoding an initiation protein are uncommon but not exceptional. The Enterococcus faecalis pheromone-responding plasmid pAD1  (Francia, et al., 2004), the Staphylococcus xylosus plasmid pSX267 , the plasmids pAMβ1 and pLS32 from Bacillus subtilis [33–35], and the Staphylococcus aureus multiresistance plasmids pSK1 and pSK41 [36, 37] fall into this category. However, the origins of replication in all of these plasmids have recognizable iterons, and an insert that contains some or all of the iterons from these plasmids is usually capable of driving plasmid replication if the initiator protein is provided in trans. The minimal replicon of the p42d plasmid is the repC ORF sequence driven by a constitutive promoter (Plac) with an SD sequence that we designed. Frame shift and deletion mutants of the repC gene disrupted the capacity for replication of the minimal replicon, indicating that RepC is essential for replication and is likely the initiator protein. To confirm this function, it will be necessary to demonstrate that this protein binds the oriV, melts the double-stranded DNA, and recruits the initiation host factor.
A DNA sequence analysis of the repC gene clearly showed the absence of iterons or other large, perfect or imperfect, repetitive sequences (>8 bp), which are the typical DNA-binding sites of plasmid initiator proteins .
The replication of several bacterial plasmids, such as P1, F, R6K, RK2, Rts1, pMU720, and pSC101, requires a crucial and concerted participation of DnaA and the plasmid-encoded initiator protein. These plasmids contain at least one DnaA box in their oriV sequences [38–43]. For other plasmids, DnaA participates only as an accessory, but these plasmids also contain DnaA boxes in their origins of replication (e.g., pR1) . However, we failed to identify such DnaA boxes within the repC-coding region, suggesting that DnaA does not have a role in p42d replication.
A common property of theta-replicating plasmids is an A+T rich region close to the origin of replication, which is necessary for strand melting and the assembly of host initiation factors . The repC ORF sequence of p42d contains a large A+T rich region that is crucial for plasmid replication. A construct carrying silent mutations that partially eliminated the A+T rich region was unable to promote replication in R. etli strains with or without the symbiotic plasmid, indicating that this region is an essential part of the oriV. However, a sequence analysis of other repC genes located in repABC operons revealed that an A+T rich region was present in all of the analyzed plasmids but its relative location was not conserved (data not shown).
The p42d minimal replicon (pDOP-C) has two intriguing properties. First, the construct resulted in enhancing the plasmid copy-number to around six, in contrast parental plasmid, which was maintained at 1-2 copies per chromosome. Second, the strain carrying this construct has a longer duplication time and a lower yield when the cells reach stationary phase than the strain without this construct.
While describing the observed increase in the plasmid copy-number, we must bear in mind that the repC gene in pDOP-C was expressed by a constitutive promoter.
In addition, the negative transcriptional regulation of the repC gene expression mediated by RepA and RepB was eliminated, and the antisense RNA (ctRNA), which also plays a negative role in the expression of repC, was removed. In the absence of these layers of negative regulation, it is expected that the plasmid replication would accelerate resulting in the production of new DNA molecules with a concomitant increase in the number of new origins of replication, which in turn, could be used to promote new rounds of replication, leading to cell death. However, in the present study, with the use of the minimal replicon (pDOP-C) we did not observe cell death, and the plasmid copy-number increased only moderately. This observation suggests the existence of a posttranslational mechanism that limits RepC activity, thus preventing over-initiation.
Growth kinetics of CFNX101 and CFNX107 were identical (data not shown), however, when pDOP-C was introduced into CFNX1017 growth of the bacterium was inhibited. The growth rate and yield diminution observed in strain CFNX107/pDOP-C relative to CFNX107 is not likely caused by the metabolic burden imposed by pDOP-C replication. The size of the parental plasmid (p42d) is approximately 374 Kb, while the size of pDOP-C is approximately 5.57 Kb; even if we take into consideration the 6-fold increase in plasmid copy-number, the amount of DNA required for replication in CFNX107/pDOP-C is several fold lower than the amount of DNA required for replication in CFNX101. Based on these observations it can be hypothesized that RepC, being an initiator protein, must perform three tasks: recognize the origin of replication, unwind the DNA at the origin, and recruit the replisome. An excess of RepC could lead to the formation of more of replication "bubbles". However, if one or more elements of the replisome are suboptimal in the growing cell, then, some replication forks will be stalled resulting in inhibition of cell division and growth.
We demonstrated that pDOP-C was capable of autonomous replication in an R. etli strain lacking the parental plasmid (p42d). However, we could not introduce this construct into an R. etli strain harboring the parental plasmid. In contrast, a similar construct that contained the repC gene of S. meliloti pSymA replicated autonomously with the same behavior in both strains. This result indicates that RepC is an incompatibility factor that prevents the coexistence of p42d and pDOP-C and that the incompatibility phenomenon is replicon-specific. Additionally, a construct (pDOP-C1-1086) expressing a chimeric protein consisting of the amino-terminal region of p42 RepC and 39 aa residues of the carboxy-terminal region of the pSymA RepC protein was capable of replicating as an independent entity with the same efficiency in R. etli strains, with or without p42d. This result indicates that the last 39 aa residues of the RepC carboxy-terminal region are directly involved in the incompatibility phenotype. A close inspection of this region in the RepC proteins of pSymA and p42d shows that they share 62.5% of identity, indicating that 15 amino acid residues or less are critical in promoting the incompatibility phenotype. Interestingly, however, in spite of the variations in 15 aa residues, RepC proteins of p42d and pSymA have a similar secondary structure: both possess two alpha helices of ten amino acid residues each, separated by a coiled region of six amino acid residues, in the same relative positions.
Our current hypothesis linking incompatibility and the RepC posttranslational regulation is as follows: RepC, like many other plasmid-encoded initiator proteins, exists in two forms, an active monomer and an inactive dimer, and protein thermodynamics favors dimer formation . The RepC carboxy-terminal region is involved in dimer formation, and the dimerization process is replicon-specific. The introduction of pDOP-C into a strain containing p42d displaces the RepC monomer-dimer equilibrium that favors the inactive form, preventing the establishment of the incoming plasmid. A similar introduction of a construct with the RepC of a compatible plasmid will not affect the monomer-dimer equilibrium and will allow the establishment of the new plasmid.
Another unusual observation was the inability to complement the repC ORF in trans for replication. One possibility is that the repC transcript acts as an RNA primer for replication or assists in DNA melting at the oriV. However, the construct pDOP-Cs/SD, which lacks a SD sequence, could not replicate in CFNX101, suggesting that translation is required for the newly synthesized RepC protein to be located at the oriV. To the best of our knowledge, the only initiator protein that functions only in cis is RepA from prophage N15 . At this stage we cannot determine which of these possibilities is more likely, and further experiments are needed to resolve these questions.
RepC is the only element encoded in the repABC operon of the Rhizobium etli p42d plasmid that is necessary and sufficient for plasmid replication and is likely the initiator protein. The oriV of this plasmid resides within the repC gene and is located close to or inside of a large A+T region. This architecture is shared by other repABC plasmids. Our results also indicate that RepC can act as an incompatibility factor and that the last 39 aa of the carboxy-terminal region of this protein are involved in this phenotype.
This work was supported by the Consejo Nacional de Ciencia y Tecnología (CONACyT, México) (Grant number: 000000000100099); and by the Programa de Apoyo a Proyectos de Investigación e Inovación Tecnológica (PAPIIT-UNAM, México) (Grant number IN205611-3) to M.A. C. R. C-R, F. P-L and G P-S were supported during the Ph.D. program (Programa de Doctorado en Ciencias Biomédicas-Universidad Nacional Autónoma de México) with scholarships from Consejo Nacional de Ciencia y Tecnología and Dirección General de Estudios de Posgrado (México). We are greatly indebted to Ángeles Pérez-Oseguera for her technical support, and to Dr. Pallavolu Maheswara Reddy for his critical review of the manuscript.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.