SGM Journals
CELL AND MOLECULAR BIOLOGY OF MICROBES

A simple plasmid-based system that allows rapid generation of tightly controlled gene expression in Staphylococcus aureus

Andrew T. F. Liew, Torsten Theis, Slade O. Jensen, Jorge Garcia-Lara, Simon J. Foster, Neville Firth, Peter J. Lewis, Elizabeth J. Harry

  • 1The ithree Institute, University of Technology Sydney, Ultimo, New South Wales 2007, Australia
  • 2School of Biological Sciences, University of Sydney, Sydney, New South Wales 2006, Australia
  • 3Microbiology and Infectious Diseases, School of Medicine, University of Western Sydney, New South Wales 2751, Australia
  • 4Department of Molecular Biology and Biotechnology, University of Sheffield, Sheffield S10 2TN, UK
  • 5School of Environmental and Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia
  • Correspondence
    Elizabeth J. Harry
    liz.harry{at}uts.edu.au

    • Microbiology 2011; 157(3):666–676 · https://doi.org/10.1099/mic.0.045146-0

    View at publisher PubMed

    Abstract

    We have established a plasmid-based system that enables tightly controlled gene expression and the generation of GFP fusion proteins in Staphylococcus aureus simply and rapidly. This system takes advantage of an Escherichia coli–S. aureus shuttle vector that contains the replication region of the S. aureus theta-mode multiresistance plasmid pSK41, and is therefore a stable low-copy-number plasmid in the latter organism. This vector also contains a multiple cloning site downstream of the IPTG-inducible Pspac promoter for insertion of the gene of interest. Production of encoded proteins can be stringently regulated in an IPTG-dependent manner by introducing a pE194-based plasmid, pGL485, carrying a constitutively expressed lacI gene. Using GFP fusions to two essential proteins of S. aureus, FtsZ and NusA, we showed that our plasmid allowed tightly controlled gene expression and accurate localization of fusion proteins with no detrimental effect on cells at low inducer concentrations. At higher IPTG concentrations, we obtained sixfold overproduction of protein compared with wild-type levels, with FtsZ–GFP-expressing cells showing lysis and delocalized fluorescence, while NusA–GFP showed only delocalized fluorescence. These results show that our system is capable of titratable induction of gene expression for localization or overexpression studies.

    • ↵A supplementary figure, showing the construction of plasmid pLOW-GFP, is available with the online version of this paper.

    • †These authors contributed equally to this work.

    Edited by: J. Lindsay

    INTRODUCTION

    Many cellular processes in bacteria, such as cell division and DNA replication, require tightly regulated protein levels in the cell for their proper function (Harry et al., 2006). Many proteins have also been found to interact with each other in a particular stoichiometry, and severe phenotypes are often observed if this stoichiometry is perturbed. Phenotypes that are associated with protein overproduction can often lead to useful insights about protein interactions and/or protein function. For example, in Escherichia coli, overproduction of the cell division protein FtsZ is toxic to cells. However, overproduction of the cell division protein FtsA counteracts this toxic effect (Dai & Lutkenhaus, 1992; Dewar et al., 1992). This led to the suggestion that these proteins interact with each other in a particular stoichiometry in the cell. The severe toxicity associated with FtsZ overproduction also highlights the importance of gene expression systems that allow tightly regulated overproduction of proteins when induced, while causing minimal effects on cell viability in the absence of inducer.

    Staphylococcus aureus is a Gram-positive pathogen and is a significant cause of a number of diseases ranging from minor skin infections to life-threatening bacteraemia (García-Lara et al., 2005). The ability of S. aureus to develop resistance to current clinically approved antibacterials makes the treatment of these infections increasingly difficult. Despite the clinical significance of S. aureus, several factors have hampered molecular analysis of this Gram-positive pathogen. With the exception of strain RN4220, S. aureus contains several restriction–modification systems that prevent genetic manipulation using recombinant DNA isolated from E. coli (Kreiswirth et al., 1983; Veiga & Pinho, 2009). However, strain RN4220 is heavily mutagenized, and is not representative of pathogenic S. aureus strains (Lindsay, 2008). Construction of genomically integrated constructs of ‘non-RN4220’ S. aureus strains often requires a two-step process involving phage transduction and screening for the loss of a resistance marker (Lindsay, 2008), which can be time-consuming.

    A number of regulated systems which allow the tight control of gene expression and thus protein production have been described for Bacillus subtilis and E. coli (Cha & Stewart, 1997; Muchová et al., 2002). However, relatively few systems are available for controlled overexpression of genes in S. aureus. In general, these systems either utilize an ectopically integrated genomic copy of the gene, expressed under the control of a regulatable promoter (Puyang et al., 2003), or express the gene of interest on a high-copy-number plasmid (Zhang et al., 2000). The integration of DNA constructs into the genome, however, can be time-consuming, and when a high-copy plasmid system is used there is the risk of overproducing proteins, which could then lead to cell death.

    In this report, we present a quick and easy plasmid-based system that allows the overexpression of genes placed under an IPTG-inducible promoter in natural S. aureus isolates. This system uses a low-copy-number plasmid, called pLOW, which contains an IPTG-inducible promoter, Pspac, to allow regulated gene expression. The system also includes a second high-copy-number plasmid (pGL485) expressing lacI, the gene encoding the lac repressor protein, to fully repress the Pspac promoter. We show using a β-galactosidase assay and quantitative immunoblotting that this plasmid-based system enables tight repression as well as titratable induction of genes placed under the control of the Pspac promoter of plasmid pLOW. We also show that this system, at low inducer concentrations, can be used to accurately localize two proteins (FtsZ and NusA) involved in different cellular processes, when fused with a GFP tag, as well as to examine the phenotypic effects on S. aureus cells of overproduction of both proteins when high inducer concentrations are used.

    METHODS

    Bacterial strains and culture conditions.

    E. coli strain DH5α (Sambrook & Russell, 2001) and S. aureus strains SH1000 (Horsburgh et al., 2002) and RN4220 (Kreiswirth et al., 1983) were employed in this study. E. coli strains were cultured at 37 °C in Luria–Bertani (LB) broth or on LB agar containing, where appropriate, 100 μg ampicillin (Ap) ml−1. S. aureus strains were cultured at 37 °C in brain heart infusion (BHI) broth or on BHI agar containing, where appropriate, 5 μg erythromycin (Em) ml−1, 25 μg lincomycin (Lm) ml−1 and 10 μg chloramphenicol (Cm) ml−1.

    E. coli was transformed by standard procedures (Sambrook & Russell, 2001) and S. aureus RN4220 by electroporation, as previously described (Grkovic et al., 2003). Plasmids used to transform the S. aureus LH607 (protein A-deficient NCTC 8325 S. aureus strain) and SH1000 backgrounds were purified from RN4220 strains carrying the corresponding plasmids. Briefly, S. aureus RN4220 was first lysed with lysostaphin [0.2 mg ml−1 (Sigma)] at 37 °C for 30 min, and then plasmid DNA was purified using the QIAprep Spin Miniprep kit (Qiagen). Purified DNA was then electroporated into the S. aureus LH607 and SH1000 backgrounds.

    DNA isolation and manipulation.

    Plasmid DNA was isolated from E. coli using the QIAprep Spin Miniprep kit (Qiagen). Restriction enzymes, shrimp alkaline phosphatase (all New England Biolabs), T4 DNA ligase (Promega) and Pfx50 DNA polymerase for PCR amplification (Invitrogen) were each used according to the manufacturers' instructions. Oligonucleotides (Table 1⇓) were purchased from Sigma, PCR products were purified with QIAquick PCR Purification columns (Qiagen), and DNA fragments were isolated from agarose gels using the QIAquick Gel Extraction kit (Qiagen). Relevant sequences were verified by automated DNA sequencing, performed by the Australian Genome Research Facility (University of Queensland, Australia).

    Table 1.

    Primer sequences

    Underlined type, restriction endonuclease recognition sites; bold type, RBS sequences; italic type, stop codons changed to a tryptophan.

    Construction of pLOW, pLOW-BgaB, pLOW-GFP-Control and pLOW-GFP.

    Construction of pLOW and pLOW-GFP is depicted in Supplementary Fig. S1. The S. aureus Em-resistance gene ermC was amplified from plasmid pNL9164 (Yao et al., 2006) using primers ermCf-ClaI and ermCr-KpnI (Table 1⇑), producing a DNA fragment equivalent to basepairs 2067–3300 of the published sequence of pNL9164. This fragment was cloned into the ClaI and KpnI sites of pAPNC213 (Morimoto et al., 2002), resulting in plasmid pAPNC213-erm (Table 2⇓). Plasmid pAPNC213-erm was digested with EcoRI and BglII, and the fragment containing the promoter Pspac, the Em-resistance determinant ermC and the gene encoding the repressor LacI was cloned into the EcoRI and BamHI sites of the pSK41 minireplicon pSK7700 (Ni et al., 2009), resulting in plasmid pLOW (Table 2⇓). The gene bgaB, encoding a thermostable β-galactosidase, was introduced in order to evaluate the capability of LacI to repress the Pspac promoter. The gene was amplified from plasmid pMAD (Arnaud et al., 2004) using primers bgaBf and bgaBr-EcoRI. The DNA fragment was cloned into the SalI and EcoRI sites of pLOW, resulting in plasmid pLOW-BgaB. Plasmid pLOW expressing untagged GFP (pLOW-GFP-Control) was constructed by introducing the gfpmut-1 gene PCR-amplified from plasmid pSG1164 (Lewis & Marston, 1999) using the GFP controlf-SalI/GFP controlr-BamHI primer pair. The PCR product was then cloned into plasmid pLOW using the SalI and BamHI restriction sites.

    Table 2.

    Plasmids

    Plasmid pLOW-GFP was constructed by inserting the gfpmut-1 gene downstream of the Pspac promoter. The gene gfpmut-1, encoding GFP, was amplified from plasmid pSG1164 (Lewis & Marston, 1999) using the primers gfpr-EcoRI and gfpf-XmaI. The amplicon was cloned into the EcoRI and XmaI sites of plasmid pLOW, resulting in plasmid pLOW-GFP (Fig. 1⇓). The copy number of pLOW in S. aureus SH1000 was determined as described previously (Grkovic et al., 2003) and found to be maintained at approximately 10 copies per cell.

    Figure image not available in archive
    Fig. 1.

    (a) Plasmid pLOW, designed for regulated expression of S. aureus proteins. The MCS is shown in full, with the reading frame indicated by the triplet arrangement of the nucleotide sequence. Only unique restriction sites are shown. Genes: ApR, E. coli blaM gene for β-lactamase; ColEI-HC, high-copy-number E. coli plasmid origin of replication; EmR, S. aureus gene originating from plasmid pE194 and encoding Em-resistance gene ermC; pSK41 ori, low-copy-number S. aureus origin of replication; LacI, lac repressor protein constitutively expressed from the penicillinase promoter Ppcn. (b) Plasmid pLOW-GFP, designed for producing protein–GFP fusions in S. aureus. GFP, gfpmut-1 gene encoding GFP.

    Construction of ftsZ–gfp and nusA–gfp fusions on plasmid pLOW.

    To construct the C-terminal FtsZ–GFP fusion, the ftsZ gene was amplified from S. aureus SH1000 genomic DNA using the primer pair ftsZf-SalI/ftsZr-BamHI. The forward primer was designed to incorporate the native ftsZ RBS (Table 2⇑) in the final ftsZ–gfp construct. The stop codon of the ftsZ gene was modified to encode a tryptophan residue, to create a translational fusion of ftsZ to GFP. The plasmid containing the in-frame ftsZ–gfp fusion was called pLOW-FtsZ-GFP. GFP was separated from the genes of interest by a short linker peptide (Trp-Ile-Pro-Gly).

    The C-terminal NusA–GFP was constructed in a similar way to the FtsZ–GFP fusion, except that the primer pair nusAf-SalI/nusAr-BamHI was used to PCR-amplify the nusA gene from S. aureus SH100 genomic DNA.

    β-Galactosidase reporter assay.

    S. aureus cells were grown to OD600 0.5. The expression of β-galactosidase was induced by the addition of IPTG at concentrations of 50, 100, 200, 400, 800 and 1000 μM. Cultures were incubated in the presence of inducer for an additional 3 h at 37 °C. Samples of ∼500 μl were collected from each culture. The number of cells in each sample was equalized based on the OD600 readings. A quantitative ONPG β-galactosidase assay was performed as described by Bhavsar et al. (2001). The A420 of each sample was determined with a PowerWave HT microplate spectrophotometer (BioTek).

    Quantitative immunoblotting for FtsZ–GFP expression.

    An overnight culture of S. aureus LH607 cells carrying plasmid pLOW-FtsZ-GFP was diluted to a starting OD600 of 0.05, and grown to mid-exponential phase (OD600 ∼0.40) with different inducer concentrations (0, 5, 10, 20, 50, 100, 200, 400 and 1000 μM). The OD600 of the different cultures was normalized to ensure that similar amounts of proteins were present in each sample. About 5 ml of each culture was harvested by centrifugation at 4000 r.p.m. for 5 min at room temperature. After removal of the supernatant, the bacterial cell pellet was resuspended in 150 μl WL buffer [0.3 mg lysostaphin ml−1 (Sigma), 1× protease inhibitor (Roche), 25 mM Tris, 0.3 mg PMSF ml−1]. The resuspension was then incubated at 37 °C for 30 min. A 50 μl volume of 4× NuPAGE buffer (Invitrogen) was then added to the cell lysate and the samples were boiled at 95 °C for 5 min.

    Using the appropriate dilution of the sample, proteins were then separated on a 4–12 % NuPAGE Novex Tris-Acetate precast gel (Invitrogen) under reducing conditions and blotted to nitrocellulose membranes using the iBlot Blotting System (Invitrogen). Blocking and incubation with primary (1 : 10 000) rabbit polyclonal anti-FtsZ antibodies and secondary (1 : 2500) horseradish peroxidase-conjugated anti-rabbit IgG (Promega) were done in 5 % milk in PBS containing 0.05 % Tween 20 for 2 h and 1 h, respectively, at room temperature. Protein visualization was performed with the ECL kit (GE Healthcare) as per the manufacturer's instructions, and densitometric analysis of the bands was performed with the Kodak 1D Molecular Imaging Software, version 4.5 (Kodak).

    Establishing the linear range of detection for quantitative immunoblotting.

    To establish the linear range of detection, i.e. the range of band intensities over which a change in protein concentration generates a proportional change in intensity, a series of dilutions were prepared for one of the cell lysates (grown with 1000 μM IPTG) and probed with anti-FtsZ antibodies (as described above). The observed band intensities, measured by densitometry, were plotted against the lysate concentration to determine the linear range of detection for FtsZ–GFP and native FtsZ, which was subsequently used for quantification of FtsZ–GFP and native FtsZ from immunoblots.

    Growth rate analysis of S. aureus.

    A 50 ml volume of BHI medium containing the appropriate antibiotics was inoculated with a stationary phase culture of S. aureus to a final OD600 of 0.005. Subsequently, cell growth was monitored by OD600 measurements every 30 min. Generation times were calculated, and data were analysed for statistical significance using a Student's paired t test with a two-tailed distribution.

    Fluorescence microscopy.

    Live-cell microscopy was performed as described previously for B. subtilis (Peters et al., 2007). Exponentially growing S. aureus cells from 1 ml culture were collected by centrifugation and resuspended in 20 μl LB broth. A 3 μl volume of the cell suspension was transferred onto a 2 % (w/v) agarose pad [prepared by dissolving type 1 agarose (low electroendosmosis, Sigma) in LB broth] within a 65 μl Gene Frame (Integrated Sciences). Where necessary, S. aureus cells were stained with 1 μg ml−1 4′,6-diamidino-2-phenylindole (DAPI) to visualize nucleoids.

    Immunofluorescence microscopy (IFM) was performed as described previously (Peters et al., 2007; Pinho & Errington, 2003), with modifications. Briefly, 0.5 ml of mid-exponential culture was first harvested and fixed with ice-cold methanol for 1 h at −20 °C. After washing with PBS, cells were resuspended in 90 μl GTE buffer (50 mM glucose, 20 mM Tris/HCl, pH 7.5, 10 mM EDTA). A gentle lysis was performed using lysostaphin (Sigma) at a final concentration of 30 ng ml−1 for 1 min on a polylysine-treated slide. Cells were then treated with 2 % BSA/PBS Tween 20 for 15 min and washed with PBS. Primary anti-FtsZ antibodies (raised against B. subtilis FtsZ in rabbits) were then added and incubated overnight at 4 °C. Following this, the cells were washed 10 times with PBS and incubated with 5 % goat serum in PBS for 30 min. Secondary antibody (donkey anti-rabbit FITC-conjugated antibody) was then added at a dilution of 1 : 100 and incubated for 1 h in the dark at room temperature. Finally, cells were washed again with PBS and mounted in 50 % (v/v) glycerol.

    Phase contrast and fluorescence microscopy were performed using a Zeiss Axioplan 2 fluorescence microscope equipped with a 100× phase contrast objective and an AxioCam MRm cooled charge-coupled device (CCD) camera controlled through AxioVision software (version 4.5, Carl Zeiss). GFP and DAPI fluorescence was visualized with filter sets 488009 and 488002 (Carl Zeiss), respectively. Image analysis, processing and preparation for publication were performed using AxioVision version 4.5. Data from the cell-size measurements were analysed for statistical significance using a Student's paired t test with a two-tailed distribution.

    RESULTS AND DISCUSSION

    Features of plasmid pLOW

    As depicted in Fig. 1⇑, plasmid pLOW features pUC19 components that allow replication and selection in E. coli, i.e., a high-copy-number origin of replication, ColEIHC, and an Ap-resistance determinant (Ni et al., 2009; Yanisch-Perron et al., 1985). Replication of pLOW in S. aureus occurs by the theta mode of replication from sequences derived from the low-copy-number multiresistance plasmid pSK41 (Firth et al., 2000). Furthermore, the plasmid contains a number of restriction sites located downstream from the Pspac promoter to facilitate the introduction of the gene under investigation (Fig. 1a⇑). The gene encoding LacI, constitutively expressed from the penicillinase promoter Ppcn, was introduced into pLOW to limit the detrimental effects of low-level expression of S. aureus genes in the cloning host E. coli DH5α as well as to effectively repress the expression of genes placed under the control of Pspac in S. aureus in the absence of IPTG. A selectable resistance marker for use in S. aureus is provided by the ermC gene. This marker was chosen to allow for the use of the pLOW vector together with the tetracycline-selectable S. aureus integrative vectors, which can be used to ectopically introduce other gene constructs onto the S. aureus chromosome (Lee et al., 1991).

    Construction of plasmid pLOW

    Plasmid pLOW was constructed using two different plasmids and is depicted in Supplementary Fig. S1. The first plasmid, pAPNC213, is a B. subtilis integrative plasmid containing the Pspac promoter, a multiple cloning site (MCS) and the gene encoding the lacI repressor. The second plasmid (pSK7700) contains the staphylococcal pSK41 minireplicon. Construction of plasmid pLOW first involved the introduction of the resistance determinant ermC for selection of the plasmid in S. aureus into plasmid pAPNC213. The Pspac-ermC-lacI cassette was then excised from plasmid pAPNC213 and cloned into plasmid pSK7700 containing the staphylococcal pSK41 minireplicon, creating plasmid pLOW. Plasmid pLOW-GFP was constructed by inserting the PCR-amplified gfpmut-1 gene downstream of the Pspac promoter. For complete details of pLOW plasmid construction see Methods and Supplementary Fig. S1.

    Regulated and titratable induction of protein expression with plasmid pLOW

    It was important to ensure that genes placed under the control of the IPTG-regulated promoter Pspac of pLOW showed titratable gene expression in response to the level of IPTG. To test this, we placed the bgaB gene, encoding the β-galactosidase enzyme from Bacillus stearothermophilus, under the control of the Pspac promoter of plasmid pLOW and assayed β-galactosidase activity using different inducer concentrations. This assay has been shown to be a reliable measure of promoter induction in S. aureus (Zhang et al., 2000). However, no transcriptional control in response to different IPTG concentrations was detected even though lacI was expressed from the pLOW construct, suggesting that the repressor concentration was too low to effectively repress Pspac (Fig. 2⇓). This is not surprising as earlier studies have shown the need for a multi-copy plasmid containing constitutively expressed lacI to repress the Pspac promoter (Jana et al., 2000). Therefore, a second copy of lacI was introduced on a co-resident rolling circle (RC) plasmid, pGL485 (Table 2⇑). This plasmid has been shown to exist at a copy number of about 20 copies per cell in S. aureus (Weisblum et al., 1979). The additional copy of lacI provided by pGL485 resulted in titratable and IPTG-dependent bgaB expression from Pspac, with a basal level of ∼4 % of the maximum β-galactosidase activity. Maximal enzyme activity was achieved between 200 and 400 μM IPTG (Fig. 2⇓), which resulted in a 20-fold increase in activity compared with the uninduced sample. Interestingly, maximal enzyme activity, in the presence of lacI overexpression (from plasmid pGL485), was slightly lower than the activity measured in the absence of plasmid pGL485. This is likely due to lacI expression from pGL485 preventing full release of LacI from the Pspac promoter in pLOW, even at high IPTG concentrations (400–1000 μM). Nevertheless, this assay demonstrates the utility of this two-plasmid system for controlled high through to very low level expression of genes.

    Figure image not available in archive
    Fig. 2.

    Titratable induction of the Pspac promoter system in S. aureus, using the thermostable β-galactosidase bgaB from B. stearothermophilus as a reporter gene. IPTG concentrations tested were 0, 50, 100, 200, 400, 800 and 1000 μM. The repressor protein LacI was constitutively expressed either from plasmid pLOW-BgaB alone (○) or from pLOW-BgaB and pGL485 (•). Points and error bars represent the mean of three measurements and the sem, respectively.

    Quantification and phenotypic effects of protein overexpression: FtsZ

    Having established that the Pspac system of plasmid pLOW/pGL485 was capable of titratable induction of gene expression utilizing the β-galactosidase assay, it was important to demonstrate the utility of the plasmid for the tightly controlled expression of natural S. aureus genes. We chose the cell division protein FtsZ (Haydon et al., 2008) as an example to demonstrate the applicability of our plasmid system for controlled gene expression. FtsZ is a highly conserved, tubulin-like protein which polymerizes into a ring called the Z ring at the midcell division site, and has been shown to be essential for cell viability in S. aureus (Pinho & Errington, 2003). We chose this protein for this study because FtsZ is currently a promising candidate for anti-staphylococcal therapy (Singh & Panda, 2010), and therefore a tool to evaluate the effects of regulated ftsZ expression in live cells in response to potential FtsZ inhibitors may be useful in drug–protein interaction studies. Secondly, the concentration of FtsZ has been shown to be important for proper cell division to occur in E. coli and B. subtilis. High-level overexpression (two- to fivefold overexpression) of ftsZ in these organisms results in a perturbation of cell growth (Ma et al., 1996; Ward & Lutkenhaus, 1985; Weart & Levin, 2003), indicating that cellular levels of FtsZ have to be tightly regulated.

    To demonstrate tightly controlled ftsZ expression using the pLOW Pspac system, we quantified the cellular FtsZ levels present when different inducer concentrations were used. We used an FtsZ–GFP fusion so that we could visualize FtsZ production in vivo. We introduced the gfpmut-1 gene into plasmid pLOW downstream of the Pspac promoter, creating plasmid pLOW-GFP (Fig. 1b⇑). The S. aureus ftsZ gene was then PCR-amplified and inserted upstream of the GFP tag, creating an ftsZ–gfp fusion in the plasmid.

    To determine the level of FtsZ–GFP produced from plasmid pLOW, we used quantitative immunoblotting and anti-FtsZ antibodies. Cells were incubated with different IPTG concentrations (5, 10, 20, 50, 100, 200, 400 and 1000 μM), harvested and probed with anti-FtsZ antibodies (Fig. 3a⇓). Two bands corresponding to the molecular mass of FtsZ–GFP (∼68 kDa) and native FtsZ (∼41 kDa) were observed in the immunoblot. A third band with a higher molecular mass than that of native FtsZ (∼49 kDa) was detectable in cells incubated with higher IPTG levels (100–1000 μM). This protein likely corresponds to a proteolytically cleaved FtsZ–GFP (Cristea et al., 2005), and was no more than 11±1 % of total FtsZ–GFP. Whilst the amount of cleaved FtsZ–GFP product was small, we combined the levels of cleaved FtsZ–GFP and FtsZ–GFP and utilized this value to calculate the ratio of FtsZ–GFP levels to those of native FtsZ to quantify the protein expression level from the Pspac promoter.

    Figure image not available in archive
    Fig. 3.

    (a) Quantitative immunoblot showing different levels of FtsZ–GFP overexpression in S. aureus using anti-FtsZ antibodies. Lanes contained the following IPTG concentrations: 1, 0 μM; 2, 5 μM; 3, 10 μM; 4, 20 μM; 5, 50 μM; 6, 100 μM; 7, 200 μM; 8, 400 μM; 9, 1000 μM. WT, wild-type S. aureus LH607; FtsZ–GFP′ refers to a possible proteolysed product of FtsZ–GFP. (b) Fold overexpression of FtsZ in S. aureus obtained by calculating the ratio of FtsZ levels (FtsZ–GFP+native FtsZ) in the FtsZ–GFP-expressing strain to native FtsZ levels of wild-type LH607. IPTG concentrations tested were as for (a). Points and error bars represent the mean value obtained from three separate experiments and the sem, respectively. (c) Longer exposure of a partial immunoblot from (a) showing the low but detectable levels of FtsZ–GFP in lanes 2–4. Lanes: 1, 0 μM; 2, 5 μM; 3, 10 μM; 4, 20 μM; 5, 50 μM IPTG. The contrast of the image was enhanced to visualize faint FtsZ–GFP bands in lanes 2–4.

    As depicted in Fig. 3(b, c)⇑, at concentrations between 0 and 20 μM IPTG, the cellular level of FtsZ–GFP was only about 4 % of the level of native FtsZ (that is, the level of total FtsZ in the FtsZ–GFP-expressing strain was ∼0.04-fold higher than wild-type FtsZ levels). In contrast, the level of FtsZ–GFP increased to 80 % of the native FtsZ level (that is the level of total FtsZ in the FtsZ–GFP-expressing strain was ∼1.8-fold higher than wild-type FtsZ levels) at 50 μM IPTG. At 1000 μM IPTG, a maximal sixfold overproduction was achieved. These results demonstrate both the tight regulation of gene expression at low levels of IPTG and the substantial overexpression, in this case sixfold overproduction of FtsZ, at high levels of inducer.

    We then determined the phenotypic effect of the various levels of FtsZ–GFP expression in S. aureus SH1000 cells by examining growth rate (Fig. 4a⇓), cell size (Fig. 4b⇓) and FtsZ–GFP localization (Fig. 5⇓). Carriage of pLOW-GFP and pGL485 had no significant effect on the size of S. aureus cells (P>0.05 compared with S. aureus SH1000). However, the cells grew slightly slower (P<0.05 compared with S. aureus SH1000), probably due to plasmid replication and gene expression. In the absence of IPTG (Fig. 5a⇓), FtsZ–GFP could not be visualized, which is consistent with the tight repression of gene expression of the Pspac promoter of plasmid pLOW in the presence of pGL485 in the absence of inducer. Expression of ftsZ–gfp at IPTG levels up to 100 μM (3.5-fold higher than cellular levels of FtsZ in wild-type cells) had no significant effect on either cell size or generation time (P>0.05 compared with S. aureus carrying pLOW-GFP and pGL485), as shown in Fig. 4(a, b)⇓. Consistent with this, FtsZ–GFP could be seen localizing as a ring at the future division site (Fig. 5b–d⇓). The localization pattern looked identical to that of S. aureus FtsZ observed with IFM (Fig. 5g⇓) and previously published FtsZ immunolocalization data (Pinho & Errington, 2003).

    Figure image not available in archive
    Fig. 4.

    Generation times (a) and cell diameters (b) of S. aureus strains expressing different levels of FtsZ–GFP and NusA–GFP from the plasmids pLOW-FtsZ-GFP and pLOW-NusA-GFP, respectively. Except for SH1000, all S. aureus strains analysed for generation time and cell diameter carried the RC plasmid pGL485 from which an additional copy of lacI was constitutively expressed. (a) Bars represent the mean of three measurements; (b) bars represent the mean of at least 100 measurements. Error bars, sem.

    Figure image not available in archive
    Fig. 5.

    (a–f) Live-cell microscopy of a strain expressing FtsZ–GFP grown with different IPTG concentrations (shown in μM). Yellow arrows indicate cell lysis, red arrows show non-ring-like structures and blue arrows show delocalized FtsZ–GFP. A phase/GFP fluorescence overlay is shown in the left-hand panel and a corresponding phase image in the right-hand panel. (g) IFM of S. aureus FtsZ. Bars, 1 μm.

    Pspac derepression at higher IPTG concentrations (>200 μM) resulted in severe cell division defects. We observed a 40 % increase in cell size (Fig. 4b⇑) and generation times of up to 126 min compared with 35 min for S. aureus strains carrying pLOW-GFP/pGL485 (Fig. 4a⇑). In addition, cell lysis was frequently observed (yellow arrows in Fig. 5e, f⇑). These observations are consistent with the cell division defects observed with high-level FtsZ overproduction in B. subtilis and E. coli (Ma et al., 1996; Ward & Lutkenhaus, 1985; Weart & Levin, 2003). This increase in cell size has also been observed with FtsZ-depleted S. aureus cells (Pinho & Errington, 2003). FtsZ also appeared delocalized and formed non-ring-like structures (blue and red arrows in Fig. 5e, f⇑). The large increase in cell size and delocalized FtsZ signal are likely to be associated with aberrant (non-septal) peptidoglycan synthesis caused by the delocalization of the peptidoglycan synthetic machinery when FtsZ is delocalized (Pinho & Errington, 2003). Our results suggest that this high level of FtsZ production results in a phenotype similar to FtsZ depletion in S. aureus.

    Phenotypic effects of protein overexpression: NusA

    We also examined the localization pattern of a transcription protein of S. aureus, NusA. NusA is an RNA polymerase (RNAP)-associated protein that is highly conserved across many bacterial species (Davies et al., 2005). It has been shown to play a role in both transcription antitermination and pausing (Quan et al., 2005). In E. coli and B. subtilis, NusA acts to stimulate intrinsic termination of RNA transcription (Yakhnin et al., 2008), and has been shown to localize to the nucleoid due to its interaction with RNAP (Davies et al., 2005; Watt et al., 2007).

    Interestingly, unlike FtsZ–GFP, maximal expression of NusA–GFP only resulted in a very mild phenotype. As depicted in Fig. 4(a, b)⇑, there was no significant change in the generation times (P>0.05 compared with S. aureus maintaining pLOW-GFP and pGL485), and only a slight, but statistically significant, increase in cell size (P<0.05 compared with S. aureus maintaining pLOW-GFP and pGL485).

    The localization pattern of NusA–GFP was consistent with the staining pattern observed with the DNA stain DAPI (Fig. 6a, b⇓), suggesting that NusA–GFP in S. aureus also localizes to the nucleoid. To ensure that the NusA–GFP localization in S. aureus was specific for the nucleoid and not just localizing within the cytoplasm, we compared the localization pattern of NusA–GFP with that of untagged GFP expressed at the same inducer concentration (500 μM IPTG) from plasmid pLOW-GFP-Control (Fig. 6d, e⇓). Although untagged GFP had a uniform staining pattern like that of NusA–GFP, it consistently occupied more of the cell and extended outside the nucleoid-staining region. Taken together, this shows that like B. subtilis NusA–GFP, S. aureus NusA–GFP expressed from plasmid pLOW localizes to the nucleoid.

    Figure image not available in archive
    Fig. 6.

    Localization of NusA (a–c) and a GFP control (d–f) in S. aureus grown in the presence of 50 μM IPTG. Exponentially growing cells of S. aureus harbouring the plasmids pLOW-NusA-GFP (a–c) and pLOW-GFP (d–f) were examined by fluorescence microscopy. DNA staining using DAPI (a, d), GFP fluorescence (b, e) and an overlay of the two images (c, f) corresponding to the same cells are shown. Bars, 1 μm. Arrows in (d) and (e) show that, unlike NusA–GFP, the untagged GFP shows no clear gap in GFP fluorescence when there is a clear separation of nucleoids. This indicates that NusA localizes specifically to the nucleoid in these cells.

    Similarities in NusA localization between B. subtilis and S. aureus indicate that NusA was being correctly localized in S. aureus. However, in B. subtilis, NusA appears to concentrate into discrete foci, which represent sites of rRNA transcription (Davies et al., 2005). Utilizing various inducer concentrations (5–500 μM), no such transcription foci were detected in S. aureus expressing NusA–GFP. We attribute the lack of transcriptional foci to differences in rRNA operon number and distribution over the genome. B. subtilis has 10 rRNA operons, and seven of these are clustered near oriC, leading to the observed transcription foci in the presence of NusA–GFP. S. aureus has only five rRNA operons, located between approximately 6 ° and 27 ° on the 36 ° map of the chromosome. Fewer rRNA operons and the absence of rRNA operon clusters are most probably the reasons for no clearly defined NusA–GFP foci in S. aureus. We cannot rule out the possibility that the increase in cellular NusA, even at the lowest levels of IPTG (5 μM), prevents the formation of these foci. Regardless of this, the presence of NusA over the entire nucleoid in S. aureus is consistent with its localization in other organisms.

    Conclusion

    We have shown that the plasmid pLOW system allows quick and easy titratable expression of genes in S. aureus, demonstrating the utility of the plasmid for high- and low-level gene expression. Although several overexpression systems are available for S. aureus (Lindsay, 2008; Zhang et al., 2000), we have shown that our plasmid system is capable of tight repression of gene expression in the absence of inducer, through the use of a second high-copy-number plasmid constitutively expressing lacI (pGL485), while allowing up to a sixfold increase in protein overproduction. We have also demonstrated the usefulness of the pLOW-GFP plasmid for protein localization in live S. aureus cells without perturbing cellular homeostasis when low inducer concentrations are used.

    We envisage that this system would be useful for studying drug–protein interaction dynamics, namely to determine the cellular targets of inhibitory compounds either by determining the localization of the protein in the presence of the compound or by determining whether overexpression of the protein target confers resistance to the chemical (Haydon et al., 2008; Stokes et al., 2005). This system would be especially useful for low-level expression of GFP-fused proteins that may not be fully functional, but are used to ‘decorate’ the chromosomally encoded native protein and allow its accurate localization. Since expression of GFP on a plasmid in addition to the wild-type protein might result in artefacts in terms of protein localization, the localization should be confirmed with further experiments. In fact, this plasmid-based system would be an ideal first-pass approach for high-throughput, genome-wide analysis of protein localization and protein interaction networks in S. aureus, similar to the impressive study conducted in Saccharomyces cerevisiae (Huh et al., 2003). The localization patterns of candidate proteins identified in such a study should then be further evaluated/confirmed by chromosomally tagging these proteins with fluorescent tags (Pereira et al., 2010). Overall, we present here a useful tool for studying the effects of protein overexpression as well as localizing proteins in S. aureus.

    Acknowledgments

    This work was supported by the National Health and Medical Research Council (Australia) via Project Grant 455597 to E. J. H. and P. J. L. The anti-FtsZ antibody was a gift from Shigeki Moriya.

    References

    1. Arnaud, M., Chastanet, A. & Débarbouillé, M. (2004). New vector for efficient allelic replacement in naturally nontransformable, low-GC-content, Gram-positive bacteria. Appl Environ Microbiol 70, 6887–6891.
    2. Bhavsar, A. P., Zhao, X. & Brown, E. D. (2001). Development and characterization of a xylose-dependent system for expression of cloned genes in Bacillus subtilis: conditional complementation of a teichoic acid mutant. Appl Environ Microbiol 67, 403–410.
    3. Cha, J. H. & Stewart, G. C. (1997). The divIVA minicell locus of Bacillus subtilis. J Bacteriol 179, 1671–1683.
    4. Cooper, E. L., García-Lara, J. & Foster, S. J. (2009). YsxC, an essential protein in Staphylococcus aureus crucial for ribosome assembly/stability. BMC Microbiol 9, 266.
    5. Cristea, I. M., Williams, R., Chait, B. T. & Rout, M. P. (2005). Fluorescent proteins as proteomic probes. Mol Cell Proteomics 4, 1933–1941.
    6. Dai, K. & Lutkenhaus, J. (1992). The proper ratio of FtsZ to FtsA is required for cell division to occur in Escherichia coli. J Bacteriol 174, 6145–6151.
    7. Davies, K. M., Dedman, A. J., van Horck, S. & Lewis, P. J. (2005). The NusA : RNA polymerase ratio is increased at sites of rRNA synthesis in Bacillus subtilis. Mol Microbiol 57, 366–379.
    8. Dewar, S. J., Begg, K. J. & Donachie, W. D. (1992). Inhibition of cell division initiation by an imbalance in the ratio of FtsA to FtsZ. J Bacteriol 174, 6314–6316.
    9. Firth, N., Apisiridej, S., Berg, T., O'Rourke, B. A., Curnock, S., Dyke, K. G. H. & Skurray, R. A. (2000). Replication of staphylococcal multiresistance plasmids. J Bacteriol 182, 2170–2178.
    10. García-Lara, J., Masalha, M. & Foster, S. J. (2005). Staphylococcus aureus: the search for novel targets. Drug Discov Today 10, 643–651.
    11. Grkovic, S., Brown, M. H., Hardie, K. M., Firth, N. & Skurray, R. A. (2003). Stable low-copy-number Staphylococcus aureus shuttle vectors. Microbiology 149, 785–794.
    12. Harry, E., Monahan, L. & Thompson, L. (2006). Bacterial cell division: the mechanism and its precision. Int Rev Cytol 253, 27–94.
    13. Haydon, D. J., Stokes, N. R., Ure, R., Galbraith, G., Bennett, J. M., Brown, D. R., Baker, P. J., Barynin, V. V., Rice, D. W. & other authors (2008). An inhibitor of FtsZ with potent and selective anti-staphylococcal activity. Science 321, 1673–1675.
    14. Horsburgh, M. J., Aish, J. L., White, I. J., Shaw, L., Lithgow, J. K. & Foster, S. J. (2002). σB modulates virulence determinant expression and stress resistance: characterization of a functional rsbU strain derived from Staphylococcus aureus 8325-4. J Bacteriol 184, 5457–5467.
    15. Huh, W. K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S. & O'Shea, E. K. (2003). Global analysis of protein localization in budding yeast. Nature 425, 686–691.
    16. Jana, M., Luong, T. T., Komatsuzawa, H., Shigeta, M. & Lee, C. Y. (2000). A method for demonstrating gene essentiality in Staphylococcus aureus. Plasmid 44, 100–104.
    17. Kreiswirth, B. N., Löfdahl, S., Betley, M. J., O'Reilly, M., Schlievert, P. M., Bergdoll, M. S. & Novick, R. P. (1983). The toxic shock syndrome exotoxin structural gene is not detectably transmitted by a prophage. Nature 305, 709–712.
    18. Lee, C. Y., Buranen, S. L. & Ye, Z. H. (1991). Construction of single-copy integration vectors for Staphylococcus aureus. Gene 103, 101–105.
    19. Lewis, P. J. & Marston, A. L. (1999). GFP vectors for controlled expression and dual labelling of protein fusions in Bacillus subtilis. Gene 227, 101–109.
    20. Lindsay, J. A. (2008). Staphylococcus: Molecular Genetics. Norfolk, UK. : Caister Academic Press.
    21. Ma, X., Ehrhardt, D. W. & Margolin, W. (1996). Colocalization of cell division proteins FtsZ and FtsA to cytoskeletal structures in living Escherichia coli cells by using green fluorescent protein. Proc Natl Acad Sci U S A 93, 12998–13003.
    22. Morimoto, T., Loh, P. C., Hirai, T., Asai, K., Kobayashi, K., Moriya, S. & Ogasawara, N. (2002). Six GTP-binding proteins of the Era/Obg family are essential for cell growth in Bacillus subtilis. Microbiology 148, 3539–3552.
    23. Muchová, K., Kutejová, E., Scott, D. J., Brannigan, J. A., Lewis, R. J., Wilkinson, A. J. & Barák, I. (2002). Oligomerization of the Bacillus subtilis division protein DivIVA. Microbiology 148, 807–813.
    24. Ni, L., Jensen, S. O., Ky Tonthat, N., Berg, T., Kwong, S. M., Guan, F. H., Brown, M. H., Skurray, R. A., Firth, N. & Schumacher, M. A. (2009). The Staphylococcus aureus pSK41 plasmid-encoded ArtA protein is a master regulator of plasmid transmission genes and contains a RHH motif used in alternate DNA-binding modes. Nucleic Acids Res 37, 6970–6983.
    25. Pereira, P. M., Veiga, H., Jorge, A. M. & Pinho, M. G. (2010). Fluorescent reporters for studies of cellular localization of proteins in Staphylococcus aureus. Appl Environ Microbiol 76, 4346–4353.
    26. Peters, P. C., Migocki, M. D., Thoni, C. & Harry, E. J. (2007). A new assembly pathway for the cytokinetic Z ring from a dynamic helical structure in vegetatively growing cells of Bacillus subtilis. Mol Microbiol 64, 487–499.
    27. Pinho, M. G. & Errington, J. (2003). Dispersed mode of Staphylococcus aureus cell wall synthesis in the absence of the division machinery. Mol Microbiol 50, 871–881.
    28. Puyang, X., Fan, J., Schumacher, T., Borsari, B., Davey, M. & Ling, L. L. (2003). Overexpression of murG in S. aureus does not alter the ramoplanin MIC. In 43rd Annual Interscience Conference on Antimicrobial Agents, 14–17 September 2003, Chicago, IL.
    29. Quan, S., Zhang, N., French, S. & Squires, C. L. (2005). Transcriptional polarity in rRNA operons of Escherichia coli nusA and nusB mutant strains. J Bacteriol 187, 1632–1638.
    30. Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY. : Cold Spring Harbor Laboratory.
    31. Singh, P. & Panda, D. (2010). FtsZ inhibition: a promising approach for antistaphylococcal therapy. Drug News Perspect 23, 295–304.
    32. Stokes, N. R., Sievers, J., Barker, S., Bennett, J. M., Brown, D. R., Collins, I., Errington, V. M., Foulger, D., Hall, M. & other authors (2005). Novel inhibitors of bacterial cytokinesis identified by a cell-based antibiotic screening assay. J Biol Chem 280, 39709–39715.
    33. Veiga, H. & Pinho, M. G. (2009). Inactivation of the SauI type I restriction–modification system is not sufficient to generate Staphylococcus aureus strains capable of efficiently accepting foreign DNA. Appl Environ Microbiol 75, 3034–3038.
    34. Ward, J. E., Jr & Lutkenhaus, J. (1985). Overproduction of FtsZ induces minicell formation in E. coli. Cell 42, 941–949.
    35. Watt, R. M., Wang, J., Leong, M., Kung, H. F., Cheah, K. S. E., Liu, D., Danchin, A. & Huang, J. D. (2007). Visualizing the proteome of Escherichia coli: an efficient and versatile method for labeling chromosomal coding DNA sequences (CDSs) with fluorescent protein genes. Nucleic Acids Res 35, e37.
    36. Weart, R. B. & Levin, P. A. (2003). Growth rate-dependent regulation of medial FtsZ ring formation. J Bacteriol 185, 2826–2834.
    37. Weisblum, B., Graham, M. Y., Gryczan, T. & Dubnau, D. (1979). Plasmid copy number control: isolation and characterization of high-copy-number mutants of plasmid pE194. J Bacteriol 137, 635–643.
    38. Yakhnin, A. V., Yakhnin, H. & Babitzke, P. (2008). Function of the Bacillus subtilis transcription elongation factor NusG in hairpin-dependent RNA polymerase pausing in the trp leader. Proc Natl Acad Sci U S A 105, 16131–16136.
    39. Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.
    40. Yao, J., Zhong, J., Fang, Y., Geisinger, E., Novick, R. P. & Lambowitz, A. M. (2006). Use of targetrons to disrupt essential and nonessential genes in Staphylococcus aureus reveals temperature sensitivity of Ll.LtrB group II intron splicing. RNA 12, 1271–1281.
    41. Zhang, L., Fan, F., Palmer, L. M., Lonetto, M. A., Petit, C., Voelker, L. L., St John, A., Bankosky, B., Rosenberg, M. & McDevitt, D. (2000). Regulated gene expression in Staphylococcus aureus for identifying conditional lethal phenotypes and antibiotic mode of action. Gene 255, 297–305.