Shuttle expression plasmids for genetic studies in Streptococcus mutans

Abstract

A set of shuttle plasmids containing four different constitutive promoters was generated to facilitate overexpression of foreign and native genes in streptococci, such as Streptococcus mutans. The four promoters that were chosen were: P_ami, P_spac, P₂₃ and P_veg. These promoters are active in many Gram-positive bacteria, and allow various levels of gene expression depending on the host bacterium. Shuttle plasmids were constructed based on two types of broad-host-range replication origins: a rolling-circle replicon (pSH71) and a theta replicon (pAMβ1). Shuttle plasmids derived from the pAMβ1 replicon were generated to avoid the structural and segregational stability problems associated with rolling-circle replication, since these problems may be encountered during large gene cloning. In a complementation assay, we used one such plasmid to express a gene in trans to show the utility of these plasmids. In addition, a series of plasmids was generated for the expression of recombinant proteins with an N-terminal 6×His tag or a C-terminal Strep-tag fusion, and, using a gene derived from S. mutans, we showed a high level of recombinant protein expression in S. mutans and Streptococcus pyogenes. Since these plasmids contain broad-host-range replication origins, and because the selected promoters are functional in many bacteria, they can be used for gene expression studies, such as complementation and recombinant protein expression.

INTRODUCTION

The genus Streptococcus constitutes a wide variety of bacteria ranging from beneficial organisms, such as Streptococcus thermophilus, to pathogenic organisms such as Streptococcus mutans, Streptococcus pneumoniae and Streptococcus pyogenes (Facklam, 2002). Like many other bacteria, streptococci have developed complex regulatory networks of gene expression that allow them to survive under diverse environments. To understand how the regulatory networks function, and to study interconnection among the various networks, one needs to express the regulatory genes under heterologous promoters that are separated from the regulatory controls of the genes. For expression of genes at different levels, foreign promoters with varying strength are normally employed. While the availability of foreign promoters for gene expression is not an issue for Gram-negative bacteria, such as Escherichia coli, or for the well-studied Gram-positive bacterium Bacillus subtilis, for many other Gram-positive bacteria, including streptococci, the availability of suitable promoters for gene expression remains a challenge.

In many instances, promoters isolated from one Gram-positive organism have been shown to work efficiently in other Gram-positive organisms. For example, the veg promoter (P_veg) of B. subtilis is also functional in S. pyogenes (Opdyke et al., 2003). Similarly, lactococcal promoters P₂₃ and P₅₉ are functional in many Gram-positive bacteria, including Staphylococcus aureus (Que et al., 2000). In addition, there are a few inducible heterologous promoters, such as those derived from nisin or tetracycline genes (Apfel et al., 2001; Eichenbaum et al., 1998), that have been shown to work in S. pneumoniae or S. pyogenes; however, these promoters have high basal levels of expression, and a very low titrable range, and this makes them incompatible for regulated gene expression studies.

In the case of S. mutans, which is the principal aetiological agent of human dental caries (Loesche, 1986), gene expression studies remain very difficult because of the limited availability of heterologous promoters. Moreover, there has not been any comparative study regarding gene expression under different promoters in this pathogen. Here, we constructed a series of shuttle plasmids that contain four different promoters derived from B. subtilis (P_veg and P_spac), Lactococcus lactis (P₂₃) and S. pneumoniae (P_ami). Using a gusA reporter gene, we showed that each of the four promoters is active in S. mutans, as well as in S. pyogenes. Furthermore, using these promoters, we also constructed vectors that allow the expression of a gene of interest as N- or C-terminal-tagged fusion proteins.

METHODS

Bacterial strains, growth conditions and genetic transformation

E. coli strains DH5α and NEB5α were grown in Luria–Bertani medium supplemented, when necessary, with ampicillin (100 μg ml⁻¹), kanamycin (50 μg ml⁻¹), erythromycin (300 μg ml⁻¹) or chloramphenicol (20 μg ml⁻¹). S. mutans UA159 and S. pyogenes JRS4 were routinely grown in Todd Hewitt medium (BBL; Becton Dickinson) supplemented with 0.23% yeast extract (THY). When necessary, kanamycin (300 μg ml⁻¹), chloramphenicol (20 μg ml⁻¹) or erythromycin (5–10 μg ml⁻¹) was included in the growth medium. S. mutans was transformed with the addition of competence-stimulating peptide (CSP), as described previously (Biswas et al., 2007b). S. pyogenes cultures for transformation were grown overnight in THY broth in static conditions at 37 °C. The overnight cultures were then diluted 100-fold, and grown at 37 °C until the OD₅₉₅ reached approximately 0.6. Cultures were then harvested, washed three times with ice-cold glycerol (103%, w/v), and resuspended in 103% glycerol to 1/100th of the original culture volume. A 100 μl aliquot was electroporated with approximately 5 μg plasmid DNA, using an Eppendorf electroporator at the 1.75 kV setting.

Construction of promoter plasmids for chromosomal integration

Plasmid pIB107 was selected to construct plasmids for chromosomal integration with a gusA reporter fusion (Biswas & Biswas, 2006). DNA fragments containing the promoter of interest were amplified from the appropriate plasmids containing the desired promoters, digested with BamHI and XhoI, and cloned into BamHI/XhoI-digested pIB107. For cloning of the P_ami promoter into pIB107, the primer pair Pami-Bam-Apa-F and Pami-Xho-R2 (for all the primers, see Table 1) was used to amplify a 55 bp fragment containing the promoter from the pAL2 plasmid (Beard et al., 2002). To clone the P_spac promoter into pIB107, plasmid pEU308 (Eichenbaum et al., 1998) was used as a template to amplify a 202 bp fragment with Pspac-Bam-Apa-F and Pspac-Xho-R2 primers. A 180 bp fragment containing the P₂₃ promoter was amplified from the pOri23 plasmid (Que et al., 2000) using P23-Bam-Apa-F and P23-Xho-R2 primers, for cloning into pIB107. Finally, the P_veg promoter region (180 bp) was amplified from pJRS1315 plasmid (Opdyke et al., 2003) using primers Pveg-F1 and Xho-Pveg-R2, and cloned into pIB107. The resultant plasmids were linearized with BglI, and transferred to UA159 by transformation, as described previously (Biswas & Biswas, 2006). The presence of the promoters in the chromosome of UA159 was verified by PCR amplification and DNA sequencing.

Table 1.

Oligonucleotides and templates for PCR amplification of promoter regions

Promoter	Template DNA	Primer	Sequence (5′ → 3′)*
P23	pOri23	P23-Bam-Apa-F	ACACAGGATCCGGGCCCCGAAAAGCCCTGACAACC
		P23-Bam-R1	ACACAGGATCCAACATCATTGTCATTCAT
		P23-Xho-R2	ACACACTCGAGAACATCATTGTCATTCAT
Pami	pAL2	Pami-Bam-Apa-F	ACACAGGATCCGGGCCCGAAAATTTGTTTGCTTTT
		Pami-Bam-R1	ACACAGGATCCTAGATTGACAACCATTATA
		Pami-Xho-R2	ACACACTCGAGTAGATTGACAACCATTATA
Pspac	pEU308	Pspac-Bam-Apa-F	ACACAGGATCCTGGGCCCCTGCAGATGAATTAATT
		Pspac-Bam-R1	ACACAGGATCCAATTGTTATCCGCTCACAATT
		Pspac-Xho-R2	ACACACTCGAGAATTGTTATCCGCTCACAATT
Pveg	pJRS1315	Pveg-F1	ACACAGGATCCGGGCCCCTGCAGAGGAGAGCTATGCGGG
		Xho-Pveg-R2	ACACACTCGAGATTTATTGTACAACACGAGCCCATT
MCS-A	–	MCS-F	CGATCCCCGCGGTACCCGGGAATTCTAGAGCTCGAGATCTATCGATA
		MCS-R	AGCTTATCGATAGATCTCGAGCTCTAGAATTCCCGGGTACCGCGGG
MCS-B	pASK-IBA-43+	Bgl-pASK43-F2	ACACAAGATCTCTTTAAGAAGGAGGATATACAAATGGCTAGCAGAGGATCGC
		pASK-IBA-R	CGCAGTAGCGGTAAACG
Others+	pIB175-pIB178	pJRS-F	TAAGGCTATTGGTGTTTATGGC
		pJRS-R	GTACAATTGTCATCACCATTCTT
	UA159	Eco-Smu487-F1	GCGGAATTCATGCTGATGAGTAAAACAAAAGTTATACTGG
		Bam-Smu487-R2	CGCGGATCCAACTCTGTTGGTTTTGAAACAACAGC

^*Bold letters indicate incorporated restriction sites.

⁺Primers used in the construction of various plasmids.

Construction of shuttle plasmids with heterologous promoters

To construct shuttle plasmids containing the various promoters under study, plasmid pJRS1315 was chosen. This plasmid is a derivative of pLZ12 (Perez-Casal et al., 1991), and it replicates via rolling-circle mechanism using the pSH71 replicon, which is active in both Gram-negative and Gram-positive bacteria (de Vos, 1987). pJRS1315 also contains a chloramphenicol-resistance gene, which is active in both Gram-positive and Gram-negative bacteria, and the P_veg promoter from B. subtilis (Moran et al., 1982). To facilitate promoter cloning, an intermediate plasmid, pIB144, was generated by cloning a BamHI/HindIII-digested fragment amplified from pIB81, which is a derivative of pASK43 (IBA) carrying the S. mutans covR gene, into BamHI/HindIII-digested pJRS1315 (Biswas et al., 2007a). Thus, plasmid pIB144 contains the P_veg promoter with extra restriction sites. Different promoter fragments, along with two small fragments carrying multiple-cloning sites (MCS), were cloned into pIB144 in two steps. First, the promoters of interest were amplified using the following primer pairs: P_ami, Pami-Bam-Apa-F and Pami-Bam-R1; P_spac, Pspac-Bam-Apa-F and Pspac-Bam-R1; and P₂₃, P23-Bam-Apa-F and P23-Bam-R1. Amplified PCR fragments were digested with ApaI and BamHI, and cloned into ApaI/BamHI-digested pIB144, so that the resident P_veg promoter was replaced with the respective promoter of interest. For the P_veg promoter, pIB144 was used directly for the second step, which involved cloning of DNA fragments containing two types of MCS into the intermediate plasmids. An MCS linker fragment was generated by annealing two primers, MCS-F and MCS-R, followed by cloning of the linker into BamHI/HindIII-digested intermediate plasmids (or pIB144) to create pIB164 (P_ami), pIB166 (P₂₃), pIB168 (P_spac) and pIB170 (P_veg), all of which contain the MCS to facilitate cloning of foreign genes for expression under the promoters of choice. Another MCS was also amplified from pASK43, this time using Bgl-pASK43-F2 and pASK-IBA-R primers and digestion with BglII and HindIII. It was then cloned into BamHI/HindIII-restricted intermediate plasmids (or pIB144) to generate plasmids pIB163 (P_ami), pIB167 (P_spac), pIB165 (P₂₃) and pIB169 (P_veg). This MCS allows cloning of genes for expression under different promoters as His-tagged proteins.

Various promoter-containing shuttle plasmids were also generated based on a theta-replicating plasmid pOri23 (Que et al., 2000). Plasmid pOri23 replicates in E. coli and Gram-positive bacteria, and contains an ermAM gene that confers erythromycin resistance in both E. coli and Gram-positive bacteria (see Fig. 2a). To clone heterologus promoters, primers pJRS-F and pJRS-R were used to amplify the P₂₃ promoter using pIB166 as a template, or to amplify the P_veg promoter using pIB170 as a template. PCR fragments were then cloned into EcoRI/SalI-digested and T4-blunted pOri23 to construct plasmids pIB184 (P₂₃) and pIB185 (P_veg). The correct orientation of the promoter fragments was verified by PCR analyses, followed by DNA sequencing. A fragment containing the ORF of the covR gene, which encodes a response regulator (Ajdic et al., 2002; Biswas & Biswas, 2006), was amplified from UA159, digested with BamHI and EcoRI, and cloned into EcoRI- and BamHI-digested pIB184 to create pIB609 (I. Biswas, unpublished), which was used for the complementation study.

Fig. 2. RCR-based shuttle vectors suitable for gene expression in streptococci.

RCR-based shuttle vectors suitable for gene expression in streptococci. (a) Generic physical maps of the RCR- and theta-based vectors illustrating the origins of replication and selectable markers for E. coli and Gram-positive bacteria. The locations of the promoters (P) upstream of MCS are shown. The RCR plasmids are based on the pSH71 replicon, and the theta plasmids are based on the pAMβ1 replicon. (b) RCR-based vectors, along with the corresponding promoters and MCS. B, BamHI; H, HindIII; B*, hybrid BamHI/ BglII. (c) The DNA sequences encoded in the MCS regions. The sequence between BamHI and HindIII of MCS-A, along with the unique restriction sites available for cloning, are shown in the upper panel. The sequence between hybrid BamHI/BglII and HindIII of the MCS-B region, along with available restriction sites, are shown in the lower panel. The ribosome-binding site (RBS), 6×His tag and Strep-tag sequence are underlined. The underlined GGATCT sequence indicates the BamHI/BglII (B*) hybrid site. (d) Western blot detection of His-Smu487 expression. Whole-cell lysates were prepared from S. pyogenes (GAS), S. mutans (Smu) and E. coli containing plasmids with Smu487 (pIB181) and without Smu487 (pIB180) under the control of the Pveg promoter. Approximately 50 μg whole-cell lysate was separated by using 123% SDS-PAGE, and transferred to a nitrocellulose membrane. The membrane was probed with anti-6×His antibody.

For promoter expression studies, a gusA reporter gene was isolated from pJRS462 by SmaI and SacI digestion, and it was cloned into SmaI/SacI-digested pIB164, pIB166, pIB168 and pIB170 to generate pIB175 (P_ami-gusA), pIB176 (P₂₃-gusA), pIB177(P_spac-gusA) and pIB178 (P_veg-gusA), respectively.

To generate promoter fusions in the theta replicon, plasmid pIB185 was used. Promoter sequences fused to the gusA gene were amplified directly using pJRS-F and pJRS-R primers from the plasmids pIB175, pIB176, pIB177 and pIB178, and they were cloned into ApaI/BglII-digested pIB185 to generate pIB188 (P_ami-gusA), pIB187 (P₂₃-gusA), pIB189 (P_spac-gusA) and pIB186 (P_veg-gusA), respectively.

A fragment containing the smu487 gene, which encodes a response regulator (Ajdic et al., 2002; Biswas et al., 2008), was amplified from UA159 using primers Eco-Smu487-F1 and Bam-Smu487-R2. This fragment was restricted with EcoRI and HindIII, and cloned into EcoRI/HindIII-digested pIB169 to generate pIB181.

Assay for β-glucuronidase

The β-glucuronidase (Gus) assay was modified from a published method (Cutting & Vander-Horn, 1990). Bacterial cultures were grown in THY overnight at 37 °C with the appropriate antibiotics. Cultures were diluted 1 in 20, and grown at 37 °C until the OD₅₉₅ reached approximately 0.6. A 1 ml volume of culture was harvested, washed in saline, and stored at −20 °C until all samples were ready to be processed. The pellet was then thawed, resuspended in Z-buffer (Cutting & Vander-Horn, 1990) containing lysozyme (0.5 mg ml⁻¹ final concn), and incubated at 37 °C for 10 min. Triton X-100 was then added to a final concentration of 0.013%, and the suspension was incubated on ice for 5 min. The enzymic reaction was initiated at 30 °C by addition of p-nitrophenyl-β-D-glucoside (Sigma) to a final concentration of 0.8 mg ml⁻¹; as soon as a yellow colour developed in the reaction mixture, the reaction was stopped by the addition of Na₂CO₃ to a final concentration of 0.25 M. The absorbance at 420 nm, and the time of Na₂CO₃ addition, were noted. Gus activity was expressed as [1000 × A₄₂₀]/[time (min) × cell OD₅₉₅] in Miller units (MU).

RESULTS

Measuring the relative strength of the heterologous promoters

We selected four different promoters, P_ami, P_spac, P_veg and P₂₃, for constructing shuttle plasmids suitable for gene expression. These promoters were chosen because they are highly expressed, and functional in various heterologous hosts (Barnett et al., 2007; Eichenbaum et al., 1998; Halfmann et al., 2007; Opdyke et al., 2003; Que et al., 2000; Takahashi et al., 2002). To evaluate the relative strength of these promoters, we made use of a previously described transcriptional reporter plasmid, pIB107, that allows for the fusion of promoters of interest to a promoterless gusA reporter gene (Biswas & Biswas, 2006). This plasmid also allows for insertion of the promoter–gusA reporter construct into the chromosome of S. mutans at the smu1405 locus for single-copy expression. Expression of gusA was quantified from these promoters by measuring Gus activity. For comparison purposes, we also included two well-studied native promoters from S. mutans: P_gbpC and P_gtfB (Biswas et al., 2007a; Biswas & Biswas, 2006). These promoter–gusA fusions were generated in the same way as the heterologous promoter fusions. As shown in Fig. 1, we found that each of the four promoters was functional in S. mutans, and expressed the gusA gene with a wide range of activity. The lowest activity, which was observed with the P_spac promoter (20 MU ml⁻¹), was comparable with the strength of the native promoter isolated from gbpC (Fig. 1). Since the expression of gbpC is moderately high, this suggests that the expression from the P_spac promoter is adequate for gene expression studies. On the other hand, the highest activity was observed with the P_veg promoter (104 MU ml⁻¹), and this was comparable with the expression from the P_gtfB promoter, which is highly expressed in S. mutans (Biswas & Biswas, 2006; Goodman & Gao, 2000; Li & Burne, 2001) (Fig. 1). Thus, our results showed that all four heterologous promoters were functional and highly active, making them suitable for gene-expression studies in S. mutans.

Fig. 1. Activity of various promoters in S. mutans.

Activity of various promoters in S. mutans. Promoter sequences were fused to a promoterless gusA reporter gene, and inserted into the S. mutans UA159 chromosome. Expression driven from the promoters was evaluated by measuring Gus activity at mid-exponential growth phase. The relevant regions of the promoters under investigation are shown. The −35 and −10 motifs, as well as the start site of transcription, are in bold. Underlined TG sequences indicate extended −10 regions. The Gus values are means (±sd) for experiments repeated at least twice. Background values for the control strain containing the gusA reporter gene without a promoter were less than 0.5 MU ml⁻¹.

Construction of rolling-circle-replicon-based shuttle plasmids containing different promoters

Some rolling-circle replicon (RCR) plasmids have a wide host range, and are known to replicate in both Gram-negative and Gram-positive bacteria (Gruss & Ehrlich, 1989). These are generally high-copy-number plasmids that are small in size, making them suitable for cloning purposes. To construct the shuttle plasmids, we chose an RCR-based vector, pJRS1315, which contains the pSH71 replicon, and which replicates well in E. coli and many streptococci (de Vos, 1987; Opdyke et al., 2003). In addition, this plasmid encodes a chloramphenicol-resistance gene that is functional in both E. coli and Gram-positive bacteria (Fig. 2a). We cloned the promoters of interest in pJRS1315 by replacing the resident promoter and an unrelated gene (sigX) to generate intermediate plasmids. To facilitate the cloning of target genes in these intermediate plasmids, we introduced DNA fragments containing two distinct MCS. The first MCS fragment (MCS-A) allows for cloning using 11 unique restriction sites (Fig. 2c), while the other MCS fragment (MCS-B) is derived from an E. coli expression plasmid pASK43+ that allows for the flexible cloning of fragments of interest, and also allows the expression of genes as fusion proteins (Fig. 2c).

Plasmids containing MCS-A, along with their respective promoters, are shown in Fig. 2(b). To evaluate the efficacy of these plasmids, we expressed the gusA reporter gene (~2 kb) under the control of the promoters. Resultant plasmids containing the respective promoter fused to gusA were introduced into S. mutans and S. pyogenes, followed by measurement of Gus activity. As shown in Table 2, with the exception of P₂₃ promoter, the levels of Gus activities obtained from these promoters in S. mutans were lower compared with the activities obtained with single-copy expression from the integrated chromosomal location. The Gus activities were also lower in S. pyogenes; however, they were higher than the respective activities measured in S. mutans. In both S. mutans and S. pyogenes, the level of gusA activity from P_spac was lower (but above the background level) compared with the levels from the other promoters. Taken together, our results indicate that these shuttle plasmids can be used for the expression of genes in both S. mutans and S. pyogenes. However, our results also suggest that cloning of gusA may have caused some plasmid instability, since the overall gusA activity was lower than expected when expressed from the plasmid, as opposed to chromosomal single-copy expression.

Table 2. Gus activity values obtained from various promoter containing RCR-based and theta-based vectors.

Mean values (±SD) are based on measurements of at least three independent cultures at mid-exponential growth phase.

Promoter	Gus activity (MU ml−1)
	S. mutans	S. pyogenes
RCR-based plasmid
Pami	29.9±1.8	150.2±1.6
Pspac	6.3±0.3	9.9±0.4
P23	191.0±8.0	21.5±1.4
Pveg	8.7±0.7	42.1±2.6
Theta-based plasmid
Pami	128.9±2.5	125.9±2.8
Pspac	41.1±6.0	2.1±0.3
P23	653.3±0.3	517.9±24.9
Pveg	179.5±15.5	109.7±7.7

The efficacy of the plasmids containing MCS-B for the expression of genes as fusion-tagged proteins was evaluated. For this, we selected Smu487, which is a response regulator from S. mutans (Biswas et al., 2008). A DNA fragment containing the smu487 ORF was cloned into the plasmid pIB169, which bears the P_veg promoter, to create plasmid pIB181, in which Smu487 is fused with the 6×His tag. Plasmid pIB181 was introduced into E. coli, S. mutans and S. pyogenes, whole-cell lysates were prepared, and expression of His-Smu487 was monitored by Western blot analysis, using anti-6×His antibody (Sigma). As shown in Fig. 2(d), His-Smu487 was expressed well in both S. mutans and S. pyogenes. The level of expression was about twofold higher in S. pyogenes compared with S. mutans. The expression of His-Smu487 was highest in E. coli, and this was about 5- to 10-fold higher than in the Gram-positive counterparts. Taken together, our results suggest that the plasmid sets generated here, with various promoters fused to MCS-B, can be used for expression of fusion proteins in streptococci, with high efficiency.

Construction of theta-replicon-based shuttle plasmids containing different promoters

Since we experienced some plasmid instability when we cloned the gusA gene into RCR-derived plasmids, we concluded that these plasmids may not be suitable for the cloning of certain genes, such as large genes, or genes with high G+C content (Gruss & Ehrlich, 1988). To circumvent this problem, we chose to use theta-replicating plasmids, which are structurally and segregationally more stable than RCR plasmids, and can accommodate large DNA inserts (Kiewiet et al., 1993). To this end, we used plasmid pOri23, which is a derivative of pIL253, a relatively high-copy-number variant of pAMβ1 (Simon & Chopin, 1988). Plasmid pOri23 also contains the colE1 replicon of pBluescript (Stratagene) for replication in E. coli, since pAMβ1 cannot replicate in E. coli, unlike pSH71. However, pAMβ1 can replicate in a wide range of Gram-positive bacteria, including streptococci (Luchansky et al., 1988; Perez-Arellano et al., 2001). Various pOri23 derivatives containing different promoters were generated, as described in Methods. These plasmids also contained MCS-A to aid in flexible gene cloning. To evaluate whether this series of plasmids can be efficiently used for gene expression in S. mutans and S. pyogenes, the gusA reporter gene was cloned into these plasmids, and introduced into these bacteria to quantify relative promoter strength. As shown in Table 2, high levels of gusA expression were observed in S. mutans from all of the promoters. The highest activity was obtained with P₂₃ promoter (653 MU ml⁻¹), while the lowest activity was obtained with the P_spac promoter (41 MU ml⁻¹). Depending on the promoter used, the levels of Gus expression were approximately 2- to 10-fold higher compared with their chromosomally integrated counterparts (Fig. 1 and Table 2). Except for the P_spac promoter, which also showed lower activity in RCR plasmids, higher Gus expression was also observed in S. pyogenes from the theta-replicating plasmids containing the promoters under study.

To demonstrate that these plasmids can be used for complementation studies, we complemented a covR-deficient strain of S. mutans. CovR is an important response regulator that regulates many virulence-associated genes in S. mutans (Biswas et al., 2007a; Biswas & Biswas, 2006). The covR-deficient strain selected for our study was IBS132, and this also contains a P_gbpC–gusA reporter fusion (Biswas et al., 2007a). The promoter P_gbpC is negatively regulated by CovR; therefore, CovR complementation can be studied by measuring transcription from P_gbpC. To achieve complementation, covR was expressed from plasmid pIB609, which contains covR under the control of the P₂₃ promoter in plasmid pIB184.

Transcription from PgbpC was measured from the wild-type (IBS131), the covR mutant (IBS132), and the complemented IBS132/pIB609 strain. For comparison purposes, we also included an IBS132 strain containing plasmid pIB30, which expresses covR under its own promoter (Biswas et al., 2007a). As expected, Gus activity was fourfold higher in IBS132 (mean±SD, 109.4±1.4 MU ml⁻¹), as compared with the wild-type IBS131 (27.1±1.2 MU ml⁻¹). When covR was expressed in trans in IBS132 from plasmid pIB609, the Gus activity value decreased approximately threefold to 37.1±0.7 MU ml⁻¹. This reduction was similar to the values that were obtained when covR was expressed from its own promoter in plasmid pIB30 (34.9±0.1 MU ml⁻¹). However, the Gus activity in the IBS132 strain containing plasmid vector pIB184 without covR was 108.3±1.9 MU ml⁻¹. Thus, our results indicate that expression of CovR from pIB609 is similar to the expression from pIB30. Therefore, the promoter-containing theta-replicating vectors described here can be used for genetic studies in streptococci.

DISCUSSION

We have explored the possibility of using four heterologous promoters for the expression of genes in streptococci. The promoters that were selected for this purpose originate from various Gram-positive bacteria, and are known to work in heterologous hosts (Barnett et al., 2007; Eichenbaum et al., 1998; Halfmann et al., 2007; Opdyke et al., 2003; Que et al., 2000; Takahashi et al., 2002). We found that each of the four promoters was functional in S. mutans, as well as in S. pyogenes, as expected. In S. mutans, the highest activity was obtained with the P_veg promoter in the single-copy expression system. The level of expression of the P_veg promoter was about fivefold higher than expression from the P_gbpC promoter, which is a moderately expressed native promoter (Biswas et al., 2007a; Li & Burne, 2001; Sato et al., 2000; Yoshida & Kuramitsu, 2002). However, the expression level of the P_veg promoter was almost equivalent to that of the P_gtfB promoter, which is a highly expressed native promoter (Biswas et al., 2007a; Biswas & Biswas, 2006; Li & Burne, 2001; Sato et al., 2000; Yoshida & Kuramitsu, 2002). The P_veg promoter, which originated from B. subtilis (Moran et al., 1982), contains a perfect −35 motif (TTGACA) and a near-perfect −10 motif (TACAAT) of the canonical σA consensus. The other promoters were also highly active in the single-copy expression system, with highest expression observed from P_ami, followed by P₂₃ and P_spac. Although the strength of the P_spac promoter was the lowest of the promoters tested, expression from P_spac was comparable with expression from the P_gbpC promoter. The P_spac promoter was originally constructed as an inducible promoter system for gene expression in B. subtilis, and it has been shown to work in S. pyogenes (Eichenbaum et al., 1998); however, the level of induction is much lower in S. pyogenes than in B. subtilis (Eichenbaum et al., 1998). We also found that, in S. pyogenes, P_spac promoter activity was very low compared with the other promoters.

We observed that when the promoters were present on RCR-based plasmids, with the exception of P₂₃, the activities of the promoters on the plasmids were much lower than when they were integrated in the S. mutans chromosome for single-copy expression. On the other hand, increased expression from the P₂₃ promoter was observed when P₂₃ was present on each of the RCR plasmids. RCR plasmids are known to exhibit structural or segregational instability after the cloning of relatively small DNA fragments (Gruss & Ehrlich, 1988). This is due to the nature of the replication mechanism involving ssDNA intermediates; this mechanism can lead to plasmid deletions (Michel & Ehrlich, 1986) or generation of high-molecular-mass plasmid multimers that are structurally unstable (Gruss & Ehrlich, 1988). Some DNA sequences as small as 7–10 bp in length can trigger high-molecular-mass plasmid multimer formation in bacteria (Biswas et al., 1995; Dabert et al., 1992). We speculate that the low level of promoter activity observed in the RCR plasmids is due to plasmid instability triggered by the cloning of DNA fragments containing the promoter sequences. Although the RCR plasmids suffer from this instability problem, in some instances, RCR plasmids are better suited for cloning and gene expression when compared with theta-replicating plasmids. For example, RCR plasmids are easier to introduce into various hosts compared with theta-replicating plasmids (Luchansky et al., 1988; O’Sullivan & Fitzgerald, 1999; Turgeon et al., 2006). Moreover, theta-replicating plasmids can lead to slower growth of the host, in addition to their limited host range, compared with RCR plasmids (O’Sullivan & Fitzgerald, 1999). Therefore, the RCR-based shuttle plasmids constructed here are useful, despite potential risks associated with plasmid instability.

The theta-replicating plasmid that was used for construction of our shuttle plasmids is derived from pIL253, which contains a pAMβ1 replicon (Simon & Chopin, 1988). Plasmid pIL253, and its derivatives, are known to be present as a moderate-copy-number plasmid in Gram-positive bacteria (Simon & Chopin, 1988); however, the copy number of this plasmid can vary depending on the host (Perez-Arellano et al., 2001). Since these plasmids can accommodate large DNA inserts, and are stable, they are the preferred choice for gene expression, specifically for those streptococci that are naturally transformable, such as S. mutans.

A series of shuttle plasmids was also designed so that proteins of interest could be expressed as tagged fusion proteins, as both N-terminal- and C-terminal-tagged proteins. The efficacy of the system was verified by the expression of Smu487 as a 6×His-tagged protein in the three hosts S. mutans, S. pyogenes and E. coli. Smu487 was expressed well in each of the three hosts, and was easily detected, although the expression was highest in E. coli. When it comes to protein expression and purification in Gram-positive bacteria, the choice of available systems is very restricted (Brockmeier et al., 2006; Fujimoto & Ike, 2001). Moreover, the available systems may not work in pathogenic streptococci, such as S. mutans. The gene-expression system developed here can be used for the successful expression and purification of proteins from Gram-positive bacteria, including pathogenic streptococci. This expression system should greatly facilitate gene regulation studies in this group of bacteria. Regulatory proteins may be associated with other indigenous accessory factors, and therefore purification from autologous hosts, rather than E. coli, offers immense advantages.

In conclusion, we report here a set of easy-to-clone vectors that permit high-level gene expression in S. mutans. Since these plasmids are based on constitutively expressed promoters that are carried by broad-host-range-based vectors, they can be used in many Gram-positive pathogens for gene-regulation studies. We are currently developing an inducible system that will allow us to further fine-tune the gene expression in streptococci.