Background

Dental caries is an infective transmittable bacterial disease characterized by a multi-factorial pathology, and Streptococcus mutans (S. mutans) is considered as the primary aetiological agent of dental caries [1, 2]. Adhesion to a tooth surface and biofilm formation by S. mutans are the initial steps in caries development [2]. Pac (also called P1 and SpaP) is a multi-functional adhesive and is considered the primary factor that mediates the early attachment to tooth enamel [3]. Glucan binding protein C (GbpC), wall-associated protein A (wapA) and dextranase have been demonstrated to be closely related to adherence and biofilm properties [4–6]. The aforementioned proteins all contain a conserved LPXTG motif [7, 8]. The sortase A (SrtA) enzyme has been demonstrated as an essential transpeptidase that recognizes the LPXTG motif and responsible for sorting and anchoring those proteins to the cell wall of S. mutans [9]. Inactivation of the srtA gene could result in defective pathogenesis [10]. For example, Pac from S. mutans srtA inactivated strain could not attach to cell wall, which inhibits the ability of the mutant strain to colonize teeth and form a biofilm, and consequently reduces the occurrence of caries [11, 12]. Therefore, SrtA is thought to take a critical role in pathogenesis of S. mutans.

The various genotypes of S. mutans are involved in the susceptibility to dental decay [13, 14], and the distribution of genotypes of S. mutans differs by population. In our previous studies, we compared the srtA gene of S. mutans strains isolated from caries-free children and children with high-severity caries. Chromosomal DNA of S. mutans strains were extracted and amplified by PCR (polymerase chain reaction) to obtain the srtA gene. Then the purified PCR products were sequenced. The srtA gene sequence of S. mutans UA159 was selected as a reference sequence. The srtA gene sequences of S. mutans clinical isolates were compared with that of S. mutans UA159 using Variant Reporter™ Software (Applied Biosystems, CA, USA) (accession numbers: KP301259 - KP301500). The distributions of missense mutations were compared between the groups [15, 16]. A total of 17 missense mutation sites were found and remarkably, the prevalence of the point mutations T168G and G470A significantly differed between the two groups [16]. The total length of the srtA gene in S. mutans UA159 is 741 bp. T168G is a point mutation at the 168th base in the srtA gene; this base was T in S. mutans UA159, while some clinical isolates had a G base substitution at that site. Additionally, G470A denotes a G base at the 470th base in the srtA gene of S. mutans UA159, while an A base is substituted in the srtA gene of some clinical isolates. The frequency of mutations at the 168 locus was significantly higher in the caries-free group than in the high-severity caries group. Moreover, strains with the locus 470 polymorphism exhibited a significantly higher mutation frequency in the high-severity caries group.

Since SrtA is closely associated with adherence and biofilm formation, we hypothesized that the missense mutations T168G and G470A in the srtA gene might affect the function of the SrtA enzyme and consequently lead to the changes in the cariogenicity of S.mutans. Based on our previous study, we constructed T168G and G470A missense mutations using the srtA gene of S. mutans UA159 as a template, and investigated the effects of the two missense mutations on SrtA activity in S. mutans.

Methods

Bacterial strains, plasmids, and culture conditions

S. mutans UA159 (ATCC700610) (Guangdong Culture Collection Centre of Microbiology, Guangzhou, China) was used as the source of chromosomal DNA for the PCR. The Escherichia coli (E. coli) BL21 (TaKaRa, Kyoto, Japan) as a host of gene operation and expression vector pET32a (Novagen, Madison, WI, USA) were used for gene expression. E. coli BL21 strains were grown in Luria-Bertani (LB) broth and plated onto LB medium containing 1.5 % (w/v) agar at 37 °C. Ampicillin was added when needed at 100 μg/mL (final concentration).

Construction of srtA △N120(UA159) and mutant expression vectors

SrtA is a membrane-anchoring protein containing an N-terminal signal peptide that can decrease its hydrophilicity. Therefore, full-length SrtA is difficult to purify and is unstable [17]. However, the transpeptidase activity of the truncated SrtA enzyme is not influenced by the absence of the N-terminal signal peptide because the deleted hydrophobic N-terminal region of SrtA functions as a signal peptide for secretion and a stop-transfer signal for membrane anchoring [18, 19]. Thus, to decrease the hydrophobicity of SrtA, the truncated SrtA lacking the N-terminal 40 amino acids was expressed in this study according to previous studies [20–22].

This study protocol was approved by the Ethics Committee of Guanghua School of Stomatology, Sun Yat-sen University (ERC-[2012]-13). Based on our previous epidemiological investigation [16], srtA △N120(T168G) and srtA △N120 (G470A) were constructed using the srtA gene of S. mutans UA159 as a template. The chromosomal DNA of S. mutans UA159 was extracted and amplified DNA fragment which contains truncated SrtA coding gene according to previously described methods [20, 21] with modifications. In brief, the primers 5′-CGGGATCC GCTTGGAATACCAATAGATATCAG-3′ (BamHI site is italic) and 5′-CCGCTCGAG TTAAAATGATATTTGATTATAGGACTGC-3′ (XhoI site is italic) were used to amplify the truncated srtA fragment (621 bp) from S. mutans UA159 chromosomal DNA by PCR. The srtA fragment was cloned into linearized pET32a vector by digested with BamHI and XhoI to generate the X6 HIS tagged recombinant plasmid pET32a-srtA △N120(UA159) . The site-directed mutagenesis of T168G and G470A was performed using the MutanBEST Kit (TaKaRa, Kyoto, Japan) with pET32a-srtA △N120(UA159) as a template plasmid to construct pET32a-srtA △N120(T168G) and pET32a-srtA △N120(G470A) by following the manufacturer’s instructions. The primers 5′-GCAAGAAAGAGG ATTGAACACAACAAGGC-3′ (mutated base is italic) and 5′-TAACATTAGAAACCTG ATATCTATTGGTATTCCAAG-3′ were used to generate the T168G mutation, and the primers 5′-CCTTTAGAACATGCAAAAGAAGGCATGG-3′ (mutated base is italic) and 5′-TGAAAAGAGCATCTGTGAAGATCCGGTC-3′ were used to generate the G470A mutation. As expressed products of these gene mutations, the D56E and R157H mutants of SrtA were generated. All primers used in this study were synthesised by Shanghai Sangon Company (Shanghai, China). The plasmids were sequenced by Shanghai Sangon Company to verify that the expected sites were mutated.

Expression and purification of SrtA△N40(UA159) and mutant enzymes

The pET32a-srtA △N120(UA159) and mutant constructs were introduced into E. coli BL21 by chemical transformation by the manufacturer’s protocol. The transformed cells were grown in Luria broth medium containing 50 μg/mL ampicillin at 37 °C until the OD600 reached 0.6. The expression of truncated SrtA in E. coli BL21 was induced with 0.2 mM isopropyl β-D-1-thiogalactopyranoside (IPTG), and the cells were grown for another 6 h at 15 °C. The supernatant was then collected and centrifuged at 16,000 × g for 15 min. All soluble recombinant SrtA enzymes were purified on a Ni Sepharose 6 Fast Flow column (GE healthcare, Piscataway, NJ, USA) according to the manufacturer’s instructions. The purity and specificity of the SrtA△N40(UA159) and mutant enzyme preparations were verified by sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analysis using a rabbit anti-His tag monoclonal antibody (Abgent, San Diego, CA, USA).

Non-denaturing polyacrylamide Gel analysis

The soluble recombinant SrtA enzymes were subjected to non-denaturing polyacrylamide gel electrophoresis (native PAGE) described previously [23, 24] with modifications. The proteins were loaded onto 4–16 % Bis-Tris gels and resolved by electrophoresis at 4 °C. The gels were stained with Coomassie Brilliant Blue R250 (Bio-Rad, Hercules, CA, USA) and the protein bands were visualized. The ratios of dimers/monomers of SrtA proteins were evaluated by comparing the densities of dimer bands and monomer bands using ImageJ software (National Institutes of Health, Bethesda, MD, USA). All the reported ratios are the means of triplicate assays.

SrtA activity assay

The activities of the purified SrtA△N40(UA159) and the two point mutated enzymes were monitored as described previously [21, 22] with modifications. The synthetic peptide 4-(4-dimethylamino phenylazo) benzoic acid (Dabcyl)-QALPETGEE-5-[(2-aminoethyl)amino]naphthalene-1-sulphonic acid (Edans) (Dabcyl-QALPNTGEE-Edans) (Jiershenghua, Shanghai, China) was used as the substrate to determine SrtA activity. The substrate Dabcyl-QALPNTGEE-Edans contains a fluorescent luminophore and a fluorescence quencher. When Dabcyl-QALPNTGEE-Edans is cleaved by SrtA, the fluorophore Edans group is separated from the quencher Dabcyl group, which enhances the fluorescence signal. Dabcyl-QALPNTGEE-Edans was added to the kinetic reaction at a final concentration from 0.2 μM to 12.8 μM. Reactions were performed in 400 μL of reaction buffer (50 mM Tris–HCl, 5 mM CaCl2, 150 mM NaCl, pH 7.5) containing varying concentrations of fluorescent peptide substrate Dabcyl-QALPNTGEE-Edans (0.2-12.8 μM), 1.2 μM purified SrtA, and 0.2 M NH2OH. The experiments were performed for 30 min at 37 °C at an excitation wavelength at 350 nm and an emission wavelength at 495 nm. SrtA activity was assessed by quantifying increases in fluorescence intensity using a Victor3 1420 multilabel counter (PerkinElmer, Waltham, MA, USA). The maximum velocity (V max ) and Michaelis constant (K m ) were calculated based on the curve fit for the Michaelis-Menten equation using Origin 8 software (OriginLab, Northampton, MA, USA):

$$ v= Vmax\left[S\right]/\left({K}_m+\left[S\right]\right) $$

where v is the slope during the linear phase of cleavage and [S] is the substrate concentration. The catalytic rate constant (k cat ) was calculated based on the ratio of V max to the enzyme concentration, and the catalytic efficiency was determined based on the k cat /K m ratio. All reported values are the means of triplicate assays.

Results

Site-specific mutation of srtA △N120(UA159)

The sortase coding gene srtA △N120(T168G) and srtA △N120(G470A) were generated by point mutation procedure from srtA gene of S. mutans UA159. The mutated nucleotide positions in each srtA genes were indicated in Fig. 1.

Fig. 1
figure 1

Point mutations in mutants compared with the srtA gene of S. mutans UA159. Detailed legend: srtA △N120(T168G) contains a point mutation at base 168, and srtA △N120 (G470A) contains a point mutation at base 470 (labelled in green)

Full size image

Expression and purification of SrtA△N40 (UA159) and mutants

The SrtA△N40(UA159) and point mutated sortase SrtA△N40(D56E) and SrtA△N40(R157H) were expressed as the recombinant protein that coded by srtA △N120(UA159) , srtA △N120(T168G) and srtA △N120 (G470A) , respectively. Compared with the amino acid sequence of SrtA△N40(UA159), the mutant enzyme SrtA△N40(D56E) contains a single amino acid substitution from aspartate (D) to glutamate (E) at the 56th amino acid residue, while SrtA△N40(R157H) contains a single amino acid substitution from arginine (R) to histidine (H) at the 157th amino acid residue. The amino acid sequences are shown in Fig. 2. SDS-PAGE analysis of expressed and purified SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) is shown in Fig. 3.

Fig. 2
figure 2

Point mutations in mutant enzymes compared with the SrtA enzyme of S. mutans UA159. Detailed legend: SrtA△N40(D56E) contains a single mutation at the 56th amino acid residue, and SrtA△N40(R157H) contains a single mutation at the 157th amino acid residue (labelled in green). The putative catalytic domain of SrtA is shown in red

Full size image
Fig. 3
figure 3

SDS-PAGE analysis of SrtA enzyme expression. Detailed legend: Expression of non-purified and purified SrtA analysed by SDS-PAGE. Lane M: Protein marker. Lane 1: Supernatant from E. coli BL21 transfected with pET32a. Lanes 2, 4, 6: Supernatants from E. coli BL21 transfected with pET32a-srtA △N120(UA159) , pET32a-srtA △N120(T168G) and pET32a-srtA △N120(G470A) , respectively, before purification. Lanes 3, 5, 7: Supernatants from E. coli BL21 transfected with pET32a-srtA △N120(UA159) , pET32a-srtA △N120(T168G) and pET32a-srtA △N120(G470A) , respectively, after purification using Ni sepharose 6 Fast Flow columns. New bands appeared in the supernatants from E. coli BL21 containing pET32a-srtA △N120(UA159) , pET32a-srtA △N120(T168G) and pET32a-srtA △N120(G470A) compared with the supernatant from E. coli BL21 containing pET32a. The estimated molecular weight of SrtA△N40(UA159) and the mutant enzymes was approximately 42 kDa as indicated by SDS-PAGE

Full size image

As shown in Fig. 3, the estimated molecular weight of the purified enzymes was approximately 42 kDa, which was consistent with the theoretical molecular weight. Western blot analysis for SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) probed with a rabbit anti-His tag monoclonal antibody revealed anti-His antibody-reactive bands (Fig. 4).

Fig. 4
figure 4

Western blot of SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) probed with rabbit anti-His tag monoclonal antibody. Detailed legend: Lane M: Marker. Lane 1: Supernatant from E. coli BL21 transfected with pET32a. Lanes 2, 3, 4: Supernatants from E. coli BL21 transfected with pET32a-srtA △N120(UA159) , pET32a-srtA △N120(T168G) and pET32a-srtA △N120(G470A) , respectively

Full size image

Native PAGE analysis

Native PAGE was used to evaluate the native status of purified SrtA△N40(UA159) and mutants. The dimer/monomer ratios of recombinant SrtA enzymes were estimated by comparing the densities of dimer bands and monomer bands. As shown in Fig. 5, the recombinant SrtA enzymes primarily existed as monomers and dimers. The dimer/monomer ratios of SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) were 3.25 ± 0.16, 3.28 ± 0.25 and 3.21 ± 0.32, respectively. The dimer/monomer ratios of SrtA△N40(D56E) and SrtA△N40(R157H) were close to that of SrtA△N40(UA159) on native gel.

Fig. 5
figure 5

Native PAGE of SrtA△N40(UA159) and mutant proteins. Detailed legend: Lane 1: Dimer and monomer of SrtA△N40(UA159). Lane 2: Dimer and monomer of SrtA△N40(D56E). Lane 3: Dimer and monomer of SrtA△N40(R157H). The band labelled with D indicates the band of dimeric protein; the band labelled with M indicates the band of monomeric protein

Full size image

SrtA activity assay

To measure the rate of cleavage by SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H), the substrate Dabcyl-QALPNTGEE-Edans was incubated with the purified enzymes, and the kinetic constants were calculated for the hydrolysis catalysed by SrtA△N40(UA159) and the mutant enzymes.

The kinetic parameters for SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) are compared in Table 1. The cleavage activity of the mutant enzymes was reduced compared with SrtA△N40(UA159), and the cleavage activity of SrtA△N40(R157H) was more similar to SrtA△N40(UA159) than SrtA△N40(D56E).

Table 1 Comparison of the enzymatic activities of SrtA△N40(UA159) and the mutant enzymes
Full size table

The k cat values of SrtA△N40(UA159) and SrtA△N40(R157H) were approximately 3.1-fold and 2.0-fold higher than the k cat of SrtA△N40(D56E), respectively, whereas the k cat of SrtA△N40(UA159) was only 0.3-fold higher than the k cat of SrtA△N40(R157H).

The K m values of SrtA△N40(D56E) and SrtA△N40(R157H) showed negligible decreases compared with the K m of SrtA△N40(UA159), suggesting that the affinities of SrtA△N40(D56E) and SrtA△N40(R157H) for substrate Dabcyl-QALPNTGEE-Edans were approximately equal to those of SrtA△N40(UA159).

SrtA△N40(UA159) and SrtA△N40(R157H) catalysed the sorting reaction more efficiently than SrtA△N40(D56E), as indicated by the k cat /K m ratios. The most apparent effect on SrtA catalysis was produced by the mutation D56E. The k cat /K m values of SrtA△N40(UA159) and SrtA△N40(R157H) were approximately 4.1 and 3.0 times the k cat /K m of SrtA△N40(D56E), respectively. The k cat /K m of SrtA△N40(UA159) was slightly higher than the k cat /K m of SrtA△N40(R157H).

Discussion

S. mutans is the primary pathogen of dental caries, and because of a variety of different genetic events, S. mutans strains exhibit considerable phenotypic variation and differ in cariogenicity [13]. The protease SrtA is an important virulence factor that catalyses the cell wall anchoring of surface proteins containing an LPXTG motif [7] and the polymorphisms of the srtA gene could lead to variations in cariogenic capacity [1].

Previously, we performed two epidemiological investigations to explore and compare the genetic polymorphisms of the srtA gene among clinical strains of S. mutans that were isolated from children with distinct caries status [15, 16]. The results indicated that almost all clinical isolates harboured point mutations when S. mutans UA159 served as the template strain. Although the locations and periods of the two epidemiological investigations differed, the two epidemiological investigations yielded similar results. The T168G mutation was primarily observed in the caries-free group, whereas the G470A mutation was mainly detected in the caries-active group [16]. Based on these results, this study was conducted to assess the effects of missense mutations T168G and G470A in the srtA gene on the activity of the SrtA enzyme, which has not been previously reported.

A single point mutation in srtA gene was demonstrated to be able to completely change the enzyme activity. The srtA genes in S. mutans Ingbritt and S. mutans NG5 both contain nonsense mutations that cause premature termination and result in the production of incomplete SrtA enzymes and defective cell wall sorting activity [14, 25]. The srtA gene of S. mutans NG5 contains a stop codon arising from a single base substitution from G to T at a GAA codon that is 70 amino acids upstream of the putative active site of the enzyme [14]. The generation of a new termination codon in the srtA gene of S. mutans Ingbritt arises from a deletion of 11 bp [25]. In Staphylococcus aureus, mutations at H120, C184 and R197, the catalytic triad of SrtA, could affect the enzymatic activity. For example, the point mutated sortases SrtA△N24(H120Q), SrtA△N24(C184S) and SrtA△N24(R197A) were expressed as the recombinant protein that coded by srtA △N72(C360A) , srtA △N72(T550A) , and srtA △N72(C589G/G590C) , respectively. Compared with the wild-type SrtA, the enzymatic activities of these point mutated sortases decreased dramatically [26]. However, the mutation that is not at the active site of SrtA could also influence the enzymatic activity. For instance, the average activity of point mutated sortase SrtA△N59(I123G), which was expressed as the recombinant protein that coded by srtA △N177(A367G/T368G) , was also lower than the average activity of wild-type SrtA [23].

Our study showed that missense mutations arising from single base substitutions of T168G and G470A in the srtA gene in S. mutans could also result in changes in enzyme activity. Compared with the amino acid sequence of SrtA△N40(UA159), the mutant enzymes SrtA△N40(D56E) and SrtA△N40(R157H) contained single amino acid substitutions from D to G at the 56th amino acid residue and from R to H at the 157th amino acid residue, respectively. Although the D56E and R157H mutations in the amino acid sequence did not lead to marked changes in the affinity of SrtA for the Dabcyl-QALPNTGEE-Edans substrate, the catalytic efficiency of SrtA△N40(D56E) was decreased compared with those of SrtA△N40(UA159) and SrtA△N40(R157H). The D56E mutation significantly affected SrtA catalysis. And the enzymatic activity of SrtA(D56E), which is primarily expressed in the caries-free group, was notably decreased compared with the enzymatic activity of SrtA△N40(UA159) and SrtA△N40(R157H), which is primarily expressed in the caries-active group (Table 1). This difference may be responsible for the significantly lower incidence of caries in the caries-free group than in the caries-active group. Nevertheless, dental caries is an infective bacterial disease characterized by a multi-factorial pathology, and many factors other than the srtA gene of S. mutans contribute to dental caries.

The mutation data may be explained through correlation with the physical-chemical characteristics of amino acids [27]; the physical-chemical characteristics such as charge and size more or less differ among D, E, R and H. However, the mechanisms by which amino acid mutations affect protein function are complex and related to many factors [23, 27–30], including protein dimerization and structure, among others. Additionally, some cases could not be easily explained or identified by structure alone. Multiple amino acid sequence alignments of sortase enzymes with determined structures in closely related species indicated that the catalytic triad of SrtA in S. mutans was composed of H139, C205 and R213 [22]. Thus, amino acid residues D56 and R157 do not belong to the catalytic triad, which is important for enzyme activity. To preliminarily explore the possible mechanism of the effects of the D56E and R157H amino acid mutations on SrtA enzyme activities, we analysed the generated SrtA△N40(UA159) and mutants for dimerization using native PAGE. The results showed that the recombinant SrtA enzymes existed primarily in both dimeric and monomeric forms, which was consistent with previous study [23]. No apparent differences in the ratios of dimeric/monomeric SrtA△N40(UA159), SrtA△N40(D56E) and SrtA△N40(R157H) were found, indicating that SrtA△N40(D56E) and SrtA△N40(R157H) were similar to SrtA△N40(UA159) in terms of monomer-dimer equilibrium and that the effects of the D56E and R157H mutations on the enzyme activities may not be related to SrtA dimerization. Further in-depth studies are needed to investigate the exact mechanism by which the D56E and R157H amino acid mutations affect the enzyme activities.

This study was subject to certain limitations. Research limitations precluded us from investigating the effects of D56E and R157H mutations on the structure or conformation of SrtA. Mutant strains of S. mutans should be constructed in future studies to observe the effects of the D56E and R157H mutations in SrtA on strain phenotypes, such as adhesion and biofilm formation. However, the selection of mutation sites and mutation types in our study were based on two previous clinical epidemiological investigations whereas traditional studies of random mutations and specific studies of enzyme active sites or domains did not rely on such epidemiological investigations [15, 16]. Therefore, the results of this study were an approximation of the clinical condition and manifestation of caries, which is significant for the guidance of clinical preventive services. To the best of our knowledge, the effects of the D56E and R157H mutations on the catalytic activity of SrtA have not yet been studied in the context of missense mutations of srtA in S. mutans.

Conclusions

The reasons for the differences in cariogenicity among clinical isolates of S. mutans are complex. The results of the present study suggest that the diversity of the srtA gene can lead to the differences in enzyme activity in clinical isolates of S. mutans. Specifically, the T168G mutation in the srtA gene of S. mutans can decrease the enzyme activity.

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article and its Additional file 1.