|Home | About | Journals | Submit | Contact Us | Français|
In order to facilitate genetic study of the opportunistic bacterial pathogen Pseudomonas aeruginosa, we isolated a conditional, temperature-sensitive plasmid origin of replication. We mutagenized the popular Pseudomonas stabilizing fragment from pRO1610 in vitro using the Taq thermostable DNA polymerase in a polymerase chain reaction (PCR). Out of approximately 23,000 potential clones, 48 temperature-sensitive mutants were isolated. One mutant was further characterized and the origin of replication was designated as mSFts1. The mutations that resulted in a temperature-sensitive phenotype in mSFts1 were localized to the 1.2 kb of minimum sequence required for replication in P. aeruginosa. The DNA sequence analysis revealed two mutations within the coding sequence of the Replication control (Rep) protein. Growth of P. aeruginosa carrying the temperature-sensitive plasmid at the non-permissive temperature of 42°C resulted in loss of the plasmid by greater than 99.9999% of the cells after 16 hours of growth. In order to facilitate its utilization, the mSFts1 was converted into a genetic cassette flanked by mirrored restriction endonuclease digestion sites of a pUC1918 derivative. We demonstrate utilization of the mSFts1 for genetic studies involving complementation and regeneration of a mutant in P. aeruginosa research.
Pseudomonas aeruginosa is an opportunistic bacterial pathogen that primarily infects patients who are immunocompromised or suffer from the genetic disorder cystic fibrosis (CF). This gram-negative bacterium is found almost ubiquitously in diverse environments including soil, in the hospital environment, and in association with various eukaryotic organisms.
P. aeruginosa has been the subject of intense research for the past several years as the model system for study of biofilms and quorum sensing mechanisms (Miller and Bassler, 2001, O’Toole et al., 2000). However, even with the advances in molecular technology such as DNA array and proteomic approaches, many of the studies on P. aeruginosa that highlight biofilms and quorum sensing are still performed using classical genetic tools, including transposons (Brint and Ohman, 1995, Lindsey et al., 2008, O’Toole and Kolter, 1998). Thus, it is important to continue developing better genetic tools to facilitate the study of P. aeruginosa. This is especially true for plasmid replicons. The two most commonly used plasmid replicons for P. aeruginosa research are pRK2 (pRP4) in the pLAFR series of vectors (Friedman et al., 1982, McIver et al., 1993, Vanbleu et al., 2004) and the pRO1610 replicon that is designated as the Pseudomonas stabilizing fragment (SF) (Olsen et al., 1982). The SF has been incorporated into the pUC series of E. coli plasmid cloning vectors to generate P. aeruginosa-E. coli shuttle vectors designated as the pUCP series of vectors (Schweizer, 1991, West et al., 1994). In addition, the 1.9 kb of SF, which carries the 3′-end of the bla gene has been incorporated into the unique PstI site of the bla gene of pBR322 plasmid to allow replication in P. aeruginosa without affecting the bla mediated resistance to β-lactam antibiotics (Olsen, et al., 1982). In 1994, it was determined that approximately1.2 kb of the original 1.9 kb of SF was necessary to mediate replication in P. aeruginosa (West, et al., 1994). This 1.2 kb fragment carries the origin of replication and the replication protein. During our studies to elucidate environmental stress mechanisms of P. aeruginosa, we sought to develop genetic tools that facilitate manipulation of the bacterium, including a conditional replicon of P. aeruginosa that can be cured relatively easily from the cell. Such a replicon could be used in a variety of studies in this bacterium. In this communication, we present the isolation, characterization, and utilization of a temperature-sensitive replicon SF that facilitates genetic manipulation of P. aeruginosa.
Bacteria were grown in L broth (LB) or LB supplemented with appropriate antibiotics at 37°C with aeration, unless otherwise indicated. The following antibiotic concentrations were used in this study (per milliliter): ampicillin (Apr), 100 μg for Escherichia coli; carbenicillin (Cbr), 250 μg for P. aeruginosa. The bacterial strains and plasmids used in this study are listed in Table 1.
For cloning, E. coli strain DH10B was routinely used as the host. Plasmids were purified with QIAprep Miniprep columns by Qiagen (Santa Clarita, CA) or the FastPlasmid Mini columns by Eppendorf AG (Hamburg, Germany). DNA fragments were excised from agarose gels and purified using the Qiaex II DNA gel extraction system (Qiagen) according to the manufacturer’s instructions. Restriction enzymes and DNA modification enzymes were purchased from New England Biolabs (Beverly, MA). The high-fidelity thermostable DNA polymerase Pfu from Stratagene (La Jolla, CA) was used to amplify DNA fragments for cloning, and Amplitaq thermostable DNA polymerase from Perkin Elmer (Foster City, CA) or Taq from NEB was used to screen for appropriate clones. Oligonucleotide linkers for SacI were purchased from NEB and oligonucleotide primers for PCR amplification were purchased from Operon, Inc. (Alameda, CA). DNA was introduced into E. coli via by electroporation. A standard electroporation procedure was used for E. coli using the E. coli Gene Pulser by Bio-Rad (Hercules, CA). Electroporation of plasmids into P. aeruginosa were performed as previously described (Suh et al., 1999). For transfer of plasmids into P. aeruginosa via conjugation, tri-parental conjugation was performed as previously described (Suh, et al., 1999) with the following modifications. Briefly, the E. coli donor and the helper strains were grown to mid-log phase of growth at 37°C with aeration, while the recipient P. aeruginosa strain in stationary phase of growth was mixed 1:1 in LB containing 40 mM of NaNO3 and incubated for at least 2 hours at 42°C without aeration. Then the donor, helper, and recipient cells were mixed at 7:7:1 ratio in 1 ml of saline, centrifuged, and resuspended in 25-50 μl of saline. The cell suspension was spotted on a LB plate and incubated overnight at 30°C. Following incubation, cells were scraped with a sterile cotton swab, resuspended in 2 ml of saline, and plated on selective plates.
Two different PCR amplifications were performed to isolate temperature-sensitive alleles of the 1.9 kb Pseudomonas SF: a standard amplification approach that relied on the relatively low fidelity of the Taq polymerase, and a mutagenic amplification approach as described by Leung et al. (1989). In the mutagenic approach, the reaction mix contained Mn+2 in addition to Mg+2 and contained a 5-fold lower concentration of dCTP relative to the other three dNTPs in the reaction. Amplitaq (Perkin-Elmer, Foster City, CA) was used as the thermostable polymerase in both amplifications. The SF was amplified with the following oligonucleotide primers containing PstI sites as noted by italics: SSO-62 (5′-AACGCTGCAGCAATGGCAACAACGTTGCGC-3′) and SSO-63 (5′-AACGCTGCAGAAAGGCAGGCCGGGCCGTGG-3′). Briefly, the SF was amplified from pLJR0 (Runyen-Janecky et al., 1997) via either standard or mutagenic conditions, digested with PstI, and ligated into the unique PstI site within the bla gene of pSS191, a pBR322 derivative that carries the 230 bp moriT of RP4 at the HindIII site. The resulting ligation mixtures were electroporated into E. coli DH10B and approximately 30,000 Apr transformants were pooled. The pooled mixture was conjugated into P. aeruginosa, and then plated on selective medium (PIA Cb) to obtain 200-300 Cbr colonies per plate. Following growth at 30°C, the Cbr colonies were replica plated onto fresh PIA Cb plates and growth of the colonies at 30°C and 42°C were compared to identify putative SFts alleles.
To map the SFts1 mutation within the minimal 1.2 kb sequence necessary for replicon function (West, et al., 1994), the SFts1 was PCR amplified from pSS227 with SSO-47 (AAG GCA TGC TGT CCA ACC GCT CTG TAG GC) and SSO-48 (AAG GCA TGC CTG CAG AAA GGC AGG CCG GG) as a SphI fragment. Subsequent cloning of the 1.2-kb fragment into the unique SphI site of pSS191 generated pSS254, which exhibited the same temperature-sensitive phenotype as pSS191. This confirmed that the SFts mutation(s) was localized within the mSF sequence.
A genetic cassette of SFts1 was constructed by attaching a SacI linker (NEB) to the 1.2 kb mSFts1 and the resulting fragment was cloned into the unique SacI site of pUC1918 (Schweizer, 1993). This generated pSS255, which carries the mSFts1 as a cassette that can be liberated with various restriction enzymes common to multiple cloning sites of plasmids.
To measure the frequency by which plasmids carrying either the mSF or the mSFts1 are cured from P. aeruginosa, strain PAO1 carrying plasmids with each of the replicons was grown overnight at 30°C in LB Cb. The cultures were then diluted 1:2000 in LB and grown at either 30°C, 37°C, or 42°C for 16 hours and plated on LB and LB Cb plates to obtain colony forming units (cfu). Presence (Cbr) or absence (Cbs) of the plasmid was verified by PCR amplification of the mSF.
To complement the rhlR mutation in SS165 (PAO1 rhlR201::aacCI) (Chen et al., 2005), rhlR was PCR amplified from PAO1 with Pfu and cloned into pUC19 as a 1.3 kb EcoRI-XbaI fragment. The resulting plasmid, pSS10, was converted to a mobilizable plasmid, pSS440, by the addition of a moriT to the EcoRI site (Suh et al., 2004). Then either 1.2 kb mSF or mSFts1 was introduced as a HindIII fragment to pSS440 to generate pSS447 and pSS441, respectively. These plasmids were then introduced into SS165 to generate SS1673 and SS1921, respectively, for complementation tests.
The 1.9 kb PstI fragment carrying the SFts1 was sequenced with specific internal primers that corresponded to the SF. The plasmid clone that carried the SFts1 allele (pSS227) was used as the template. Each sequence of the SFts1 was covered four times to ensure accuracy. DNA sequencing was performed by the ACGT, Inc. (Northbrook, IL). DNA sequences were aligned and analyzed with the Lasergene software (DNA Star, Madison, WI).
P. aeruginosa rhlR201::aacCI mutant (SS165) complemented with either pSS447 (mSF) or pSS441 (mSFts1) was initially streaked and grown at 30°C with Cb selection to ensure the presence of plasmids. The next day, individual colonies were restreaked onto L-agar plates and grown at 42°C overnight without Cb selection to promote plasmid curing. Individual colonies were then subcultured into LB and grown at 37°C for 15-18 hrs. From these cultures, the number of viable cells (colony forming units) was measured and supernatants were collected to assay for elastase and pyocyanin productions. Elastase activity was determined via the elastin congo-red hydrolysis assay using ~20-30 μg of supernatant protein as previously described (Ohman et al., 1980). The elastase activity was calculated as increase in A495 min−1 g−1 of protein. Pyocyanin levels were determined as previously described (Suh, et al., 1999). Both elastase activity and pyocyanin productions were normalized to the number of viable cells.
We isolated temperature-sensitive alleles of a popular Pseudomonas replicon to provide a means for efficient elimination of plasmids. The temperature-sensitive Stabilizing Fragment (SFts) was isolated by random PCR mutagenesis of the 1.9 kb PstI fragment that contains the necessary functions for plasmid maintenance in P. aeruginosa (Olsen, et al., 1982). Two different approaches were taken to introduce random mutations in the SF during DNA amplification. In the first approach, we used the standard PCR amplification protocol and relied on the relatively low fidelity of the Taq polymerase. In the second approach, we followed the procedure of Leung et al. (Leung et al., 1989) to increase the potential error rate of Taq polymerase. The amplified fragments were then cloned into the unique PstI site located within the bla gene of a mobilizable pBR322 derivative, pSS191 and electroporated into E. coli. Approximately 30,000 Apr clones were initially isolated and pooled. To screen for temperature-sensitive mutants, the SF library was conjugated into PAO1 by triparental mating. Approximately 23,000 P. aeruginosa colonies from this conjugation were genetically screened for carbenicillin sensitivity following growth at 42°C indicating loss of the plasmid. From the genetic analysis, we identified 48 putative mutants that were unable to replicate at 42°C. On initial analysis, all 48 putative SF mutants behaved similarly in their Cb sensitivity at 42°C. Therefore, one of the constructs was characterized further and designated as SFts1. Subcloning of SFts1 and DNA sequencing revealed that the temperature-sensitive phenotype of the replicon was located within the 1.2 kb mSF sequence between StuI and PstI. This is the minimum replication sequence determined by West et al. (1994) to be required for function. DNA sequence analysis revealed two mutations that resulted in temperature-sensitive phenotype: a G:C→T:A transversion that changed Gly to Cys at residue 100 and a C:G→A:T transversion that changed Ser to Arg at residue 204 (Figure 1). Both of these mutations are located within the ORF that encodes for the replication protein. The mechanism by which the mutations cause the temperature-sensitive phenotype is unclear.
We measured efficiency of the loss of plasmids harboring the mSFts1 at 42°C in order to determine the usefulness of the mSFts1 carrying plasmids for genetic manipulation of P. aeruginosa. We found that when mSFts1 harboring cells were grown at 42°C for 16 hours without antibiotic selection, greater than 99.9999 % of the cells had lost the plasmid while only 2.7778% of the non-temperature-sensitive SF carrying plasmids were lost under the same condition (Table 2). At 37°C, the frequency of loss was similar at 99.9997% for mSFts1 and 3.0457% for mSF (data not shown). The loss of mSFts1 in these cells was verified via PCR (data not shown). Thus, the loss of plasmids that replicate via mSFts1 is very efficient in the absence of a selection. To cure the mSFts1 carrying plasmid from a cell, we routinely grow cells overnight at the non-permissive temperature of 42°C, dilute the culture (10−6 - 10−8) and plate the cells on L-agar to obtain individual colonies.
In order to facilitate use of the SFts1, we subcloned the 1.2 kb mSFts1 into the pUC1918 derivative as a SacI fragment to convert it into a genetic cassette that is flanked by mirrored restriction endonuclease sites of pUC19 with the exception of EcoRI (Figure 2). Therefore, the mSFts1 can be liberated as a cassette and cloned into any E. coli plasmid carrying P. aeruginosa genes. This will convert the plasmid into a conditional P. aeruginosa plasmid that can be utilized for complementation and then subsequent curing from the strain.
To demonstrate the utility of mSFts1 in genetic analysis of P. aeruginosa, we used the conditional replicon to first complement and then restore the original mutant phenotype of a rhlR mutant. RhlR is a global regulator of gene expression that is involved in cell-cell communication for optimal production of various virulence factors, including elastase and pyocyanin in P. aeruginosa (Brint and Ohman, 1995, Ochsner et al., 1994). To complement the PAO1 rhlR201::aacCI mutant (SS165) (Chen, et al., 2005), we cloned the mSF and mSFts1 into pSS440, a plasmid that carries a wild-type copy of rhlR for complementation and oriT of RP4 for conjugation, to generate pSS447 and pSS441, respectively. The plasmids were introduced into SS165 via triparental mating and selected for Cbr. The presence or either plasmids, pSS447 and pSS441, were verified by extraction and analysis of plasmids from the respective strains (data not shown). We then grew the bacterial cells at 42°C to promote plasmid curing as described in Materials and Methods, and determined the genetic complementation of rhlR mutant phenotype by measuring elastase activity and pyocyanin. As demonstrated in Figure 3, pSS447 which carries the mSF, was not cured during growth at 42°C as indicated by complementation of the rhlR mutation. In contrast, pSS441, which carries the mSFts1, was efficiently cured from the bacterium during growth at 42°C and no longer complemented the rhlR mutation. This resulted in restoration of the original SS165 (PAO1 rhlR201::aacCI) mutant phenotype.
Plasmids with temperature-sensitive origins of replication are useful tools for a variety of genetic manipulation of bacteria. One of the most popular utilizations of temperature-sensitive replicons is as a delivery vehicle for transposons (Buchan et al., 2008, Menendez et al., 2006). At non-permissive temperatures, the plasmids are unable to replicate. Thus, transposons that remain on the plasmid and do not integrate into the host genetic material are lost. This facilitates selection of transposon insertion mutants that have acquired the element from the delivery plasmid. In addition, temperature-sensitive replicons have been utilized for temporary production of proteins (Cherepanov and Wackernagel, 1995). Furthermore, temperature-sensitive replicons have been used for conditional expression of essential genes (Jasin and Schimmel, 1984, Pyne and Bognar, 1992). Finally, temperature-sensitive replicons can be used to generate mutations at specific regions of the chromosome by homologous recombination (Choi, 2008; Link, 1997; Wang, 2005). In this study we isolated a temperature-sensitive derivative of pRO1610 replicon, mSFts1, and expanded the potential utilization of conditional replicons by demonstrating temporary genetic complementation of a mutant.
Plasmids containing the pRO1610 replicon replicate in many Gram-negative bacteria, including Burkholderia spp. (Hassett et al., 1996). Unexpectedly, however, it was recently shown that a plasmid containing the mSFts1 replicon was non-replicative in Burkholderia thailandensis, even at the permissive temperature of 30°C (Choi et al., 2008). Furthermore, a plasmid containing a pRO1610 temperature-sensitive replicon selected in B. thailandensis replicated in P. aeruginosa at 30°C but did not exhibit a temperature-sensitive phenotype in this bacterium. These data demonstrate that molecular mechanism that governs the function of pRO1610 replicon in various Gram-negative bacteria is complex and a comprehensive structure and function study is needed to elucidate its function.
We thank Herbert Schweizer for critical reading of the manuscript. This study was supported by the NIH/NIAID Training Grant T32 AI-07617, the Auburn University Biogrant, and other funds from Auburn University (AU) awarded to S.-J. S.; AU CMB-USRS, AU COSAM-URF, Department of Biological Sciences FFE, and AU-URF awarded to B. E.; and Public Health Service grant AI-19146 from the National Institute of Allergy and Infectious Disease, and Veterans Administrations Medical Research Funds awarded to D. E. O.