Involvement of the Novel Two-Component NsrRS and LcrRS Systems in Distinct Resistance Pathways against Nisin A and Nukacin ISK-1 in Streptococcus mutans

Abstract

The novel two-component systems NsrRS and LcrRS are individually associated with resistance against the distinct lantibiotics nisin A and nukacin ISK-1 in Streptococcus mutans. NsrRS regulates the expression of NsrX, which is associated with nisin A binding, and LcrRS regulates the expression of the ABC transporter LctFEG.

TEXT

Streptococcus mutans, a commensal bacterium in the oral cavity, is a cariogenic pathogen in humans (1, 2, 3). S. mutans forms dental plaques with other bacterial species. Therefore, S. mutans has evolved mechanisms that facilitate competition or cooperation with other oral bacteria in dental plaques. Many bacteria produce antibacterial agents, known as bacteriocins, to ensure survival within this community (4, 5). Bacteriocins are primarily classified into classes I and II (6). Class I bacteriocins (peptides of <5 kDa), called lantibiotics, contain a ring bridged by lanthionine and 3-methyllanthionine residues (7), whereas class II bacteriocins comprise unmodified amino acids (8). S. mutans produces several types of bacteriocins known as mutacins (9–12). However, the mechanism underlying the resistance of S. mutans to the bacteriocins of other bacteria has not been elucidated. Recently, two-component system (TCS)-based resistance mechanisms against antibacterial agents, including bacteriocins, have been identified in several bacterial species (13–17). In this study, we evaluated the roles of the S. mutans TCSs in resistance to several types of bacteriocins.

We first examined 14 S. mutans TCS mutants for susceptibility to class I and II bacteriocins by using previously described direct and MIC methods (18, 19) (Table 1 and Fig. 1). The strains, plasmids, and primers are listed in Tables S1 and S2 in the supplemental material. The methods for the construction of the mutants are described in the supplemental materials. Nisin A and nukacin ISK-1 were purified as previously described (20). One TCS encoded by the SMU.659-660 mutant (designated nsrRS [nisin A-resistant TCS] in this study) displayed increased susceptibility to nisin A compared with the wild type. Because we identified the function of this TCS, previously named SpaKR (21), we designated this TCS NsrRS. Another TCS encoded by the SMU.1146-1145 mutant (designated lcrRS [lacticin 481-resistant TCS]) displayed increased susceptibility to nukacin ISK-1 and lacticin 481. Each complemented strain was able to restore the respective mutation (Fig. 1). The susceptibilities of all TCS mutants against individual class II bacteriocins were not significantly different than those of wild-type cells.

Table 1.

Susceptibilities of TCS deletion mutants to various bacteriocins

Strain	Susceptibilitya to:
	Class I			Class IIa, munditicin	Class IIb, mutacin IV	Class IIc, lactocyclicin Q	Class IId, lacticin Q
	Nisin A	Nukacin ISK-I	Lacticin 481
UA159	1 (64)	1 (64)	1	5	0	3	8
TCS45	1 (64)	1 (64)	1	5	0	3	8
TCS1	1 (64)	1 (64)	1	5	0	3	8
TCS2	1 (64)	1 (64)	1	5	0	3	8
TCS3	8 (4)	1 (64)	1	5	0	3	8
TCS4	1 (64)	1 (64)	1	5	0	3	8
TCS5	1 (64)	1 (64)	1	5	0	3	8
TCS6	1 (64)	1 (64)	1	5	0	3	8
TCS7	1 (64)	1 (64)	1	5	0	3	8
TCS8	1 (64)	8 (8)	7	5	0	3	8
TCS9	1 (64)	1 (64)	1	5	0	3	8
TCS10	1 (64)	1 (64)	1	5	0	3	8
TCS11	1 (64)	1 (64)	1	5	0	3	8
TCS12	1 (64)	1 (64)	1	5	0	3	8
TCS13	1 (64)	1 (64)	1	5	0	3	8
TCS14	1 (64)	1 (64)	1	5	0	3	8

^aSusceptibility was based on direct analysis (the diameter of inhibition [in mm]), with the exception of the two class I drugs on which this study was focused, nisin and nukacin ISK-I, for which MICs were also determined (MIC values [in μg/ml] are shown in parentheses).

Fig 1.

Evaluation of nisin A and nukacin ISK-1 susceptibilities. The characteristics of the nsrRS (A) and lcrRS (B) loci in the mutant strains used in this study are indicated. The dotted arrows indicate the same operon. The MICs of nisin A and nukacin ISK-1 are also shown.

DNA microarray experiments were performed to characterize the transcriptional control mediated by NsrRS and LcrRS. The Agilent eArray platform was used to design a microarray; 14,028 probes (60-mers) were designed for the 2,012 protein-coding genes of S. mutans UA159 (up to seven probes per gene) (detailed methods are provided in the supplemental materials). A comparison of the transcription profiles of UA159 and nsrRS mutant cells revealed that the nisin A-mediated induction of seven of these genes (including SMU.654-657, which encodes an ABC transporter, and nsrXRS) was reduced in the nsrRS mutant (Table 2). A comparison of the transcription profiles of UA159 and lcrRS mutants demonstrated that the nukacin ISK-1-mediated induction of six of these genes (including lctEFG, which encodes an ABC transporter, and lcrXRS) was reduced in the lcrRS mutant (Table 2). We further confirmed the expression of these genes by using quantitative PCR (data not shown).

Table 2.

Genes up- or downregulated in S. mutans UA159 following nisin A or nukacin ISK-1 treatment

Treatment, regulation change, and gene IDa	Gene name	Regulation by TCS	Fold differenceb	Gene function
Nisin A
Upregulated genes
SMU.655	mutE	+c	17.3	ABC transporter, permease
SMU.1961	levD	−	17.1	Fructose-specific enzyme IIA component
SMU.654	mutF	+	16.8	ABC transporter, ATP binding
SMU.656	mutE2	+	14.8	ABC transporter, ATP binding
SMU.657	mutG	+	14.6	ABC transporter, permease
SMU.658	nsrX	+	12.9	Conserved hypothetical protein
SMU.1958	levF	−	10.8	Fructose-specific enzyme IIC component
SMU.659	nsrR	+	10.2	Response regulator
SMU.1957	levG	−	9.0	Fructose-specific enzyme IID component
SMU.1960	levE	−	7.6	Fructose-specific enzyme IIB component
SMU.1956		−	7.2	Conserved hypothetical protein
SMU.660	nsrS	+	6.3	Histidine kinase
SMU.1195		−	4.0	ABC transporter permease protein
SMU.1193	yhcF	−	3.9	Transcriptional regulator, GntR family
SMU.1194	yurY	−	3.9	ABC transporter, ATP binding
SMU.862		−	3.5	Conserved hypothetical protein
Nukacin ISK-1
Upregulated genes
SMU.1148	lctF	+d	74.4	ABC transporter, ATP binding
SMU.1149	lctE	+	67.3	ABC transporter, permease
SMU.1150	lctG	+	53.3	ABC transporter, permease
SMU.1705		−	6.5	Conserved hypothetical protein
SMU.1706		−	5.7	Conserved hypothetical protein
SMU.1704		−	5.2	Conserved hypothetical protein
SMU.1147	lcrX	+	4.7	Hypothetical protein
SMU.1146	lcrR	+	4.4	Response regulator
SMU.1145	lcrS	+	4.2	Histidine kinase
SMU.753		−	4.1	Conserved hypothetical protein
Downregulated genes
SMU.1960	levE	−	0.5	Fructose-specific enzyme IIB component
SMU.1961	levD	−	0.3	Fructose-specific enzyme IIA component
SMU.1877	manL	−	0.3	Mannose PTS component IIAB
SMU.1878	manM	−	0.3	Mannose PTS component IIC
SMU.1879	manN	−	0.3	Mannose PTS component IID

^aGene identification (ID) numbers are from the GEO database of NCBI (http://www.ncbi.nlm.nhi.gov/geo/).

^bThe fold change between bacteriocin-treated cells versus nontreated cells.

^cRegulated by NsrRS.

^dRegulated by LcrRS.

To identify the genes directly involved in nisin A resistance, we constructed SMU.654-657 and nsrXRS deletion mutants. The nsrXRS deletion mutant displayed increased susceptibility to nisin A, while the susceptibilities of the three SMU.654-657-related mutants were not altered (Fig. 1A). We also constructed an nsrX single mutant through substitution with the terminatorless Em^r gene. We confirmed that the mutation did not affect nsrRS expression downstream of nsrX (data not shown). Similar to the nsrXRS deletion mutant, the nsrX single mutant also displayed increased susceptibility to nisin A. The complementation of nsrX in the nsrX single mutant and nsrXRS mutant restored nsrX expression and susceptibility to nisin A (Fig. 1A). These results indicated that nsrX is associated with the NsrRS-regulated resistance to nisin A. NsrX comprises 280 amino acids and nine predicted transmembrane helices. NsrX does not show homology with any other known immunity proteins, such as NukH, NisI, or LtnI (22, 23, 24). Instead, NsrX shares homology with several acetyltransferases, including those of the TraX family, which is associated with F pilin acetylation in Escherichia coli (25). Thus, we performed a nisin A-binding assay to characterize the binding of nisin A to S. mutans cells, using a previously described method with some modifications (22, 26). The results of the nisin A-binding assay revealed that more nisin A bound to MM3055 cells (an nsrX deletion mutant chromosomally complemented with the cloned nsrX gene) than to MM3019 cells (an nsrX deletion mutant) (Table 3). This finding suggested that NsrX or an as-yet-unidentified factor modified through NsrX binds to nisin A and inhibits the binding of nisin A to lipid II. To identify the genes that are directly involved in nukacin ISK-1 resistance, we constructed an lctFEG deletion mutant, which exhibited a reduced MIC against nukacin ISK-1 (Fig. 1B). We determined that the LcrRS-LctFEG system is involved in lacticin 481 resistance in Lactococcus lactis CNRZ481 (Table 2). The structure of lacticin 481 is similar to nukacin ISK-1 (6), indicating that LcrRS responds to the lacticin 481 group of type AII lantibiotics.

Table 3.

Quantitative nisin A-binding assay results with the nsrX deletion mutant and nsrX-expressing cells

Strain	Genotype	Fraction	Nisin A bindinga (μg/ml)	P valueb
MM3019	nsrX mutant	Bound	10.7 ± 0.11
MM3055	nsrX+ in nsrX mutant	Bound	17.8 ± 0.23	0.008
MM3019	nsrX mutant	Unbound	21.1 ± 0.70
MM3055	nsrX+ in nsrX mutant	Unbound	16.5 ± 0.94	0.015

^aMean ± standard deviation binding activity, as measured by liquid chromatography/mass spectrometry.

^bStatistical significance for the difference between the nsrX deletion mutant versus the nsrX-expressing strain, based on Student's t test.

We performed a coculture assay using the method described in the supplemental material to determine whether TCS-mediated resistance was directly involved in colocalization with other bacteriocin-producing bacteria. S. mutans UA159 wild-type, nsrRS mutant, or lcrRS mutant cells were cocultured with nisin A-producing or nonproducing L. lactis strains. The nsrRS mutant exhibited dramatically decreased population ratios than wild-type or lcrRS mutant cells when cocultured with the nisin A-producing strain (L. lactis ATCC 11454) but not when cocultured with the non-nisin A-producing L. lactis strain (Fig. 2A). Similar results were obtained when S. mutans UA159 and mutant cells were cocultured with the nukacin ISK-1-producing or nonproducing strains (Fig. 2B).

Fig 2.

Coculture of S. mutans with L. lactis or Staphylococcus warneri. The methods for the coculture assay are described in the supplemental material. Populations of S. mutans were determined after L. lactis (A) or S. warneri (B) cells were cocultured with UA159, nsrRS mutant, or lcrRS mutant cells. *, P < 0.01, as determined using Dunnett's method, for the percentage of the S. mutans population.

In conclusion, we identified two novel TCSs involved in resistance to nisin A and nukacin ISK-1 in S. mutans and demonstrated that these two TCSs are important for coexistence with other class I bacteriocin-producing bacteria. These results highlight the roles of bacteriocins in the interactions between different species of oral bacteria and the importance of TCSs in these interactions.