Introduction

As one of the most widely distributed bacteria in the natural environment and human community, Streptococcus currently contains >100 recognized species [29]. Numerous Streptococcus species, such as S. pneumoniae, S. pyogenes, S. agalactiae, and S. suis, were considered as important human pathogens and could cause a broad range of diseases, such as pharyngitis, septicemia, pneumonia, meningitis, and toxic streptococcal syndrome [2, 33]. Notably, the re-emergence of S. pyogenes has been considered as a public health problem [33]. Other non-pathogenic Streptococcus, such as S. thermophilus, S. oralis, and S. mutans, have also been frequently isolated from human epidermis and cavities as commensal bacteria [20]. These non-pathogenic Streptococcus act as genetic donors for the phenotypic changes of pathogenic acceptors [12, 14]. However, limited by the technologies and efficient molecular tools available for the classification, conservation, and culture of Streptococcus, the functional analysis of its major genes associated with pathogenic mechanism, antimicrobial resistance, serotype, and other phenotypes is still insufficient [7, 10, 24]. Therefore, a powerful genetic tool with the ability to easily express certain genes could accelerate the study of functional genomics in these species.

The most characterized genetic tool in the functional analysis of bacteria is plasmid, which was used to investigate the expression of native and heterologous genes in different microorganisms during functional genomics studies. In Gram-negative bacteria, thousands of plasmids were developed in the model species Escherichia coli, such as pEX1/2/3, pET-5a, and pKK223-3 [9]. The common structure of these plasmids contained a typical origin of replication, which initiated the theta-type replicons not only in E. coli but also in most Gram-negative bacteria [5]. In contrast, rolling circle replicating (RCR) plasmids are abundant in Gram-positive bacteria, although they have also been reported in many Gram-negative microorganisms and in archaea. Streptococcus, as a Gram-positive genus, mainly employed the RCR mechanism for the replication of circular plasmids [13]. A few non-integrative RCR plasmid vectors have been designed for this genus. They were derived from pVA380-1 (S. orals) or pMV158 (S. agalactiae) [15, 25]. In addition, several theta-replicated plasmids have also been isolated from Gram-positive bacteria, while a series of plasmid vectors have been designed for streptococcal species based on theta replicons [4].

To obtain an efficient plasmid with the ability to replicate and express heterologous genes in Streptococcus, a dual-replicon shuttle vector system was designed and synthesized in this study by combining the theta and RCR replicons. This vector system could be well maintained in both Gram-positive and Gram-negative bacteria, and could successfully express heterologous genes. This system may be widely used in the functional genetic analysis of Streptococcus and other Gram-positive bacteria.

Materials and Methods

Bacterial Strains, Plasmids, and Culture Conditions

Fourteen bacterial strains and three plasmids were used in this study (Table 1). Gram-positive strains, including Streptococcus, Bacillus subtilis, Lactococcus lactis, and Staphylococcus aureus, were cultured on Columbia Blood Agar Base (Oxoid, Basingstoke, UK) supplemented with 8% sheep blood or a brain heart infusion (BHI) medium (Oxoid Ltd., Hampshire, UK). E. coli strain TOP10 (Invitrogen, Paisley, UK) was cultured on a Luria–Bertani (LB) medium. This study was approved by the local Ethics Committee of Beijing Ditan Hospital, Capital Medical University. No Human Participants or Animals were involved in this study.

Table 1 Bacterial strains and plasmids used in this study

DNA Extraction and PCR Amplification

The bacterial genomic DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. A PCR strategy was implemented to detect the target gene in the plasmid or the transformed bacteria, and to amplify the functional elements of the plasmid using Q5 High-Fidelity DNA Polymerase (New England Biolabs, Beverly, MA, USA) (Table S1).

Construction of Plasmids Using a Gibson Protocol

Two DNA segments were synthesized based on the sequence of pLS1ROM plasmid (accession number: JN381945.1) [22]. Gene synthesis method was applied by synthesizing reverse complementary short DNA oligos using an automatic DNA synthesizer firstly, then short double-strand DNA fragments were obtained by anneal and joined by an one-step thermocycled in vitro recombination system [1]. The large segment was assembled according to Gibson Assembly method [8]. Another DNA segment derived from pSMART-LCKan was amplified and linearized using PCR method.

A total of 100 ng of linear DNA fragments were mixed into a 20µL reaction system and assembled using NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs). The mixture was transformed into E. coli TOP10 competent cells via a chemical transformation method. Positive strains were screened using a LB medium containing 50 µg/ml ampicillin. The plasmid was purified with a GenElute Plasmid Miniprep Kit (Sigma, St Louis, USA) and validated using Sanger sequencing on an ABI 3730 machine (Applied Biosystems, Carlsbad, CA, USA). The sequence has been submitted to the GenBank database (accession number: MG799394). The primers designed for pDRE sequencing are listed in Table S2.

Cloning of Target Genes into the Constructed Vector

Two genes, the erythromycin resistance gene (erm) from plasmid pMG36e and the enhanced green fluorescent protein gene (egfp) from plasmid pUC57-egfp (Table 1), were amplified and cloned into the constructed vector to validate its function as an expression tool (Table S1). The two genes were inserted at BsaI restriction sites using Golden Gate Assembly Mix (New England Biolabs) and at HindIII/EcoRI restriction sites using NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs), respectively. The plasmid was verified by restriction enzyme digestion.

Transformation of Gram-Positive Bacteria with the Vector Containing erm/egfp

Competent Gram-positive bacterial cells were exposed to a single electrical pulse generated by a Gene-Pulser (Bio-Rad Laboratories, Richmond, CA, USA). The electric field strength of the pulse was set at 12 kV/mm. The transformed cells were plated on Columbia Blood Agar Base supplemented with 8% sheep blood and subsequently screened using erythromycin. The concentration of erythromycin was 2 µg/mL for L. lactis, Streptococcus and S. aureus, and 20 µg/mL for B. subtilis.

Validation of the Expression of the Vector Containing erm/egfp

After transforming the target bacteria with pDRE containing erm/egfp, the expression of erm and egfp in the bacteria was detected to confirm the presence of the plasmid. Real-time PCR of erm was performed to measure the relative amount of the plasmid in transformed colonies [19]. The 16S rDNA was used as the internal reference, and the relative DNA contents of the vector in the hosts were normalized as per chromosomic equivalent via multiplying the relative quantity by the copy number of 16S rDNA in different bacterial genomes [31]. A similar real-time PCR protocol was used to detect the expression of target genes. The cDNA was obtained using an One-Step reverse-transcription-PCR kit (Qiagen), and 2×EvaGreenFast qPCR Master Mix (Biorise, Beijing, China) was used to amplify erm and egfp. The 16S rRNA was used as the internal reference. All samples were run on a Roche LightCycler 480 Instrument (Roche Diagnostics, Indianapolis, IN, USA). The green fluorescence was detected using a fluorescence microscope (Zeiss, Oberkochen, Germany). Fluorescence emission was recorded at 480 nm. Primers designed for plasmid validation are listed in Table S3. The \({2^{ - \Delta \Delta {C_{\text{T}}}}}\) method was used to calculate the relative expression of the target genes (erm/egfp), the amplification efficiency of the erm and 16s rDNA were tested [19], and the amplification curve is shown in Fig. S1.

Results and Discussion

Reconstruction of E. coli Replicable Plasmid with Ampicillin Resistance

Due to the increased penicillin resistance of Gram-positive bacteria in clinic application and natural environments [16, 30], pSMART-LCKan was modified to pHCAmp in this study to adapt to the resistant profiles in Gram-positive bacteria. According to its Manual (http://www.lucigen.com/docs/manuals/MA004-CloneSmart-HCLC-Blunt-Cloning.pdf), pSMART-LCKan is a 1968 bps plasmid with the low copy number pBR322 origin of replication (ori). During the modification of this plasmid, the ori, cat promoter, bidirectional terminator (tonB), and another two terminators (T3Te and T7Te) were retained (Fig. 1a). In this step, the rop gene was removed to maintain the high copy number of the plasmid in E. coli. Using the primers ligated with two adaptors, a 1152 bp sequence was amplified for the subsequent ligation. Meanwhile, ampR gene was amplified from plasmid pJ444-01 with the same adaptors (Fig. 1b). After the two DNA fragments were assembled and transformed into E. coli TOP10, ampicillin-resistant colonies were screened out to extract the plasmid, which was named pHCAmp (Fig. 1c, d). This plasmid modification provides an efficient choice of kanamycin resistance to screen transformed strains in the future when limited antimicrobial susceptibility is exhibited by penicillin resistant Gram-positive bacteria [32].

Fig. 1
figure 1

Construction of pHCAmp. a The sequence from cat to T7 terminator in pSMART-LCKan was linearized by PCR amplification using primers P1 and P2 (Table S1). The terminators (T7Te, T3Te, and tonB), cat promoter, and ori in pSMART-LCKan were retained to allow the replication and regulation of pHCAmp in E. coli. b Terminators were represented by gray ellipses, while cat, ori, rop, and kanR were represented by gray arrows. The ampR obtained from plasmid pJ444-01 was represented by a pink arrow. Primers with overlapping sequences (the two blocks shown in red and blue) between pSMART-LCKan and ampR gene were designed and named as P3 and P4 (Table S1). c The above two fragments were assembled and transformed into E. coli TOP10, while the positive strains were screened out on LB plates containing ampicillin (AMP). d The extracted plasmid was sequenced and annotated. (Color figure online)

Construction of Plasmid pDRE

Theta and RCR are two major replication mechanisms for circular plasmids. pSMART-LCkan harbors a ColE1 replicon, which belongs to theta class B and requires DNA polymerase I to initiate replication. During theta replication, the DNA duplex is opened by the transcription of a long (~ 600 bp) pre-primer RNA II, which is then extended by DNA Pol I to initiate the leading-strand synthesis. Subsequently, DNA Pol III is loaded on the complex, and maintains the leading-strand synthesis and initiates the lagging-strand synthesis [17]. Nevertheless, in many Gram-positive bacteria, another replication mechanism named RCR is accessible for circular plasmids [6, 21]. The two-step replication proceeds as follows: (i) Leading-strand replication is initiated at the strand-specific double-strand origin (dso) until it generates a single-stranded plasmid DNA. (ii) Lagging-strand synthesis is started at the single-strand origin (sso) [26].

Therefore, to extend the host range of the constructed plasmid, a new plasmid, which was named pDRE and contained the above two replication origins, was constructed in this study. First, pHCAmp was linearized using the PCR method. Second, two DNA fragments, FA and FB, were synthesized based on the core elements, including ssoA, dso, copG, repB, tet promoter, tet terminator, maltose operator, and Pm promoter [26] (Figs. 2a, S2), of the pLS1ROM plasmid [22]. The sequences of these elements were the same as those in pLS1ROM. In this step, two short intergenic fragments were removed according to the comparison with pMV158 (accession number: NC_010096.1), so as to reduce the size of the plasmid (Fig. S2). FA contained dso, ssoA, copG, a Pm promoter (Pm), a maltose operator, and multiple cloning sites (MCS) different from those in pLS1ROM (Fig. 2a). Among these elements, dso and ssoA were of RCR origin. The gene copG encoded a transcriptional regulator, which could regulate the copy number of the plasmid by limiting the availability of the replication protein RepB [26]. The MCS was designed to include frequently used restricted sites for the insertion of exogenous genes (Fig. 2b). The malR gene encoded the maltose operon regulator, while the Pm promoter was assigned to the upstream of MCS to regulate the expression of inserted exogenous genes. FB contained repB, the malR gene with tet promoter and terminator, and two BsaI restriction sites. The repB encoded a replication initiator protein, and malR gene was designed to regulate the expression of cloned gene in MCS. Finally, these fragments (linearized pHCAmp, FA and FB) with overlapping sequences (symbolized by different colors in Fig. 2a) were assembled using the Gibson assembly method. The mixture was transformed into E. coli strain TOP10, and the positive strain was successfully screened out on LB plates containing penicillin. After extraction, the plasmid was sequenced and annotated, and named pDRE (Fig. 2b). Fig. S2 showed the similarities between the synthesized sequences of FA/FB and their original sequences in pLS1ROM and pMV158.

Fig. 2
figure 2

Construction of pDRE. a Three fragments (linearized pHCAmp, FA and FB) were used to assemble pDRE. Overlapping sequences flanking the fragments were presented by blocks in blue, red, and green. FA contained dso, ssoA, and copG for RCR replication, Pm promoter, operator, and MCS. FB included repB, tet promoter, malR, and tet terminator. b The final plasmid, pDRE, was assembled using the Gibson method. The circular sketch map was shown on the right, with the locus of MCS marked. (Color figure online)

The Genetic Characters of pDRE

This constructed shuttle vector, pDRE, was a 5329 bp circular plasmid containing two individual replication origins that could be used as a shuttle vector in Streptococcus and some of the Gram-negative bacteria. RCR plasmid pMV158 uses repB and dso to initiate its replication in Streptococcus [11], while ori plays the same role in the theta replication of plasmids. Two types of restriction sites, MCS and BSaI/BSaI, were also included in pDRE to make it easy to operate. The malR and operator, and the Pm promoter, were introduced into pDRE to control the expression of inserted genes. MalR, which belongs to the LacI–GalR family of transcriptional repressors, specifically binds to operator. In the absence of maltose, the complex of MalR–operator acts as a repressor of inserted genes [23]. Thus, the concentration of maltose in the culture medium determines the expression level of inserted genes [22].

By combining these elements, the plasmid pDRE has the potential to act as a shuttle vector in both Gram-positive and Gram-negative bacteria. The expression level of the genes carried by the plasmid could be altered using different culture conditions (Table 2).

Table 2 Characteristics of pDRE

E. coli Expression of Heterologous Genes Carried by pDRE

To validate that pDRE has the ability to express heterologous genes in E. coli, two genes were inserted into pDRE, egfp at HindIII/EcoRI and erm at BsaI/BsaI restriction sites, to generate a plasmid termed pDRE (erm/egfp), which was then transformed into E. coli TOP10 (Figs. 3a, S3). The erm gene was cloned in BsaI/BsaI restriction sites with its own promoter, while egfp was cloned in the MCS under the control of Pm promoter. The E. coli cells carrying pDRE (erm/egfp) were cultured on LB medium containing 0.2 mg/ml erythromycin. The cultured cells were then washed and diluted (1:10) in LB medium containing maltose (3 mM). The cultures were grown to OD600 = 0.5 and harvested by centrifugation. Subsequently, the expression of inserted genes was successfully detected at DNA and RNA levels (Fig. 3b, c). In the next step, the phenotype of the transformed strain was identified, and both erythromycin resistance and green fluorescence signal could be observed at 510 nm in transformed cells (Fig. 3d). These data showed that the inserted genes were expressed in E. coli.

Fig. 3
figure 3

Validation of the expression of inserted genes (egfp/erm) in E. coli. a Circular sketch map of constructed pDRE (erm/egfp) plasmid. The egfp gene (light green arrow) was inserted between HindIII and EcoRI sites, and erm (dark purple arrow) was inserted at the BSaI/BSaI sites of pDRE. Subsequently, the constructed plasmid was transformed into E. coli TOP10. b Detection of egfp and erm genes in transformed E. coli by PCR using the primers in Table S3. The erm gene was detectable in the transformed strains, thus validated its presence in the strains transformed by pDRE (erm/egfp). Both erm and egfp genes were absent in the original strains (data not shown). c The relative RNA expressions of egfp and erm were calculated by RT-qPCR using 16S rRNA as the internal reference (mean±SD, n = 3). The sizes of amplicons of egfp and erm were 447 bps and 508 bps, respectively. d Green fluorescence was successfully detected. (Color figure online)

Expression of Heterologous Genes in Streptococcus Using pDRE

To validate that pDRE has the ability to express heterologous genes in Streptococcus, ten Streptococcus strains were chosen: S. agalactiae (five strains), S. dysgalactiae (ATCC 35666), S. pyogenes (three strains), and S. thermophilus (ATCC 14485). Subsequently, the plasmid pDRE (erm/egfp) used in E. coli above was used to transform the Streptococcus strains, and transformed cells were cultured on BHI medium containing erythromycin (2 µg/ml) and maltose (3 mM). The positive strains were isolated by verifying the presence of erm gene in bacterial total DNA (Fig. S4). The presence of erm DNA and RNA was detected in all strains. The relative DNA quantities of erm were 2.23–9.76, indicating the different copy number of the plasmid in these strains. When compared with 16S rRNA, the relative level of erm expression was 0.002–0.28 (Fig. 4). The highest level of DNA abundance was observed in S. dysgalactiae, followed by S. agalactiae and S. pyogenes, whereas the highest level of expression was observed in S. pyogenes, followed by S. agalactiae. Green fluorescence could be observed in all strains. The above results confirmed that pDRE could be used as a tool to achieve gene expression in a broad range of streptococcal species.

Fig. 4
figure 4

The expression of inserted genes in pDRE in Streptococcus. Plasmid pDRE (erm/egfp) was transformed into ten strains: S. agalactiae (five strains), S. dysgalactiae (one strain), S. pyogenes (three strains) and S. thermophilus (one strain), while positive colonies were selected on the basis of erythromycin resistance; the plasmid could be extracted from the strains above. The relative DNA content of erm (top) and the relative mRNA expression level of erm (bottom) were calculated, using 16S rDNA and 16S rRNA as the internal reference, respectively (mean±SD, n = 3). The results of negative controls with untransformed cells were at the left of those of the transformed cells, respectively

In Streptococcus, shuttle vectors based on two major types of replicons have been designed for functional studies. One type of replicons involves RCR replication, such as those derived from pVA380-1 of S. oralis origin [15, 22]. These plasmids can replicate in a broad range of Gram-positive bacterial hosts but can rarely be expressed in Gram-negative bacteria [26]. Given their incompatibility with RCR plasmids already present in the hosts, plasmid vectors using theta replication have been developed, such as those derived from pAMβ1 of Enterococcus faecalis origin, pFW213 from S. parasanguinis, and its derivative, pCG1 [3, 4]. These plasmids have been widely used in Gram-negative bacteria [27, 28]. Considering that RCR replication is frequent in Gram-positive bacteria [13], we constructed the plasmid pDRE contains both types of replicons, theta replicon from E. coli and RCR replicon synthesized based on the sequence of pLS1ROM replicon. It should be mentioned that plasmid pMV158 is a RCR plasmid carrying two different ssos, ssoU and ssoA, for the initiation of lagging-strand synthesis [18]. Compared with ssoU, ssoA is more Streptococcus-specific [26]. To achieve better expression in Streptococcus, ssoA was chosen in this study to initiate the replication of the lagging strand. The expression of both inserted genes, erm and egfp, showed that pDRE could be expressed in E. coli and most Streptococcus.

Validation of pDRE Expression in Other Gram-Positive Genera

The replication of pDRE and the expression of cloned genes were also observed in multiple Gram-positive bacterial species, such as B. subtilis, L. lactis, and S. aureus. Compared with 16S rDNA, the relative quantities of target gene erm cloned in the BsaI/BsaI restirction sites were 6.37, 2.44, and 3.62 in B. subtilis, L. lactis, and S. aureus, respectively, whereas the relative expression levels of erm were 2.21, 0.07, and 0.29. These results showed that pDRE may achieve similar levels of expression in Lactococcus and Staphylococcus, while the expression of pDRE was higher in Bacillus (Fig. 5).

Fig. 5
figure 5

The inserted genes in pDRE could be expressed in other Gram-positive strains. pDRE (erm/egfp) was transformed into Bacillus subtilis (ATCC 6051), Lactococcus lactis (ACCC 11902), and Staphylococcus aureus (ATCC 6538P). The positive strains were obtained by erythromycin selection. The relative DNA content of erm (left) and relative mRNA expression of erm (right) were calculated using 16S rDNA and 16S rRNA as the internal reference, respectively (mean±SD, n = 3). The results of negative controls with untransformed cells were at the left of those of the transformed cells, respectively

Validation of the Function of the Maltose Inducible Promoter in pDRE

Four representative isolates harboring pDRE (erm/egfp) were selected to analysis the relative expression of cloned egfp gene under the control of the maltose inducible promoter. E. coli TOP10, S. agalactiae (ATCC 12513), L. lactis (ACCC 11902), and S. aureus (ATCC 6538P) were cultured in BHI medium containing erythromycin (0.2 mg/ml for E. coli TOP10 and 2 µg/ml for the other three strains). The cultures were grown to OD600 = 0.5 and harvested by centrifugation, then washed and diluted (1:10) in BHI medium containing maltose at final concentrations of 0.3, 3, and 30 mM. After the cultures were grown to the same OD, the mRNA levels of egfp were measured and the relative mRNA expression values were compared with those of erm gene (Fig. S5). Then the green fluorescence signal of the induced isolates were observed and the fluorescent intensity were detected. In contrast, the fluorescence were extremely low (0.18–1.8) in the four isolates when no maltose was added (Fig. 6).

Fig. 6
figure 6

egfp gene expression could be induced by adding maltose in transformed isolates. Detection of fluorescence in E. coli TOP10, S. agalactiae (ATCC 12513), Lactococcus lactis (ACCC 11902), and Staphylococcus aureus (ATCC 6538P) harboring pDRE (erm/egfp) after culturing in BHI medium containing different concentrations of maltose

Plasmid-based genetic tools with the ability to express both native and heterologous genes in different microorganisms can play key roles in both basic functional research and specific application such as vaccine development [34]. In most Gram-positive bacteria, one of the difficulties encountered in genetic functional analysis is the lack of an efficient overexpression system in the host. Once a shuttle vector with the ability to replicate in both Gram-negative and Gram-positive bacteria has been constructed, various exogenous genes can be cloned into the plasmid and transformed into E. coli with high quantitative efficiency. Subsequently, the plasmid can be introduced into an appropriate bacterial host to analyze the functions of target genes. The expression and replication of pDRE in B. subtilis and S. aureus indicated that this plasmid could be widely used in Gram-positive bacteria.

In conclusion, a shuttle vector system, pDRE, was constructed in this study for the heterologous gene expression in Streptococcus and other bacteria. The genes (erm and egfp) carried by pDRE could be expressed in seven Gram-positive species, including four Streptococcus spp. and three other species. Furthermore, the maltose inducible promoter could regulate the expression of cloned gene in transformed isolates, suggesting that this shuttle vector could become a useful genetic tool for genetic and functional studies in Streptococcus and other Gram-positive bacteria.