Influence of Two-Component Signal Transduction Systems of Lactobacillus casei BL23 on Tolerance to Stress Conditions

Abstract

Lactobacillus casei BL23 carries 17 two-component signal transduction systems. Insertional mutations were introduced into each gene encoding the cognate response regulators, and their effects on growth under different conditions were assayed. Inactivation of systems TC01, TC06, and TC12 (LCABL_02080-LCABL_02090, LCABL_12050-LCABL_12060, and LCABL_19600-LCABL_19610, respectively) led to major growth defects under the conditions assayed.

Lactobacillus casei is a facultative heterofermentative lactic acid bacterium used in the food industry as a starter culture for milk fermentation and the maturation of cheeses and as a probiotic (4). Probiotic microorganisms must survive the industrial production processes (3, 23) and transit through the gastrointestinal tract (3). Bacteria have evolved sophisticated mechanisms to detect and adapt to environmental changes, and among them, two-component systems (TCS) play a central role (25). TCS typically consist of a sensor kinase (HK) and a response regulator (RR) (25). HKs monitor environmental signals and, in response to a stimulus, autophosphorylate and subsequently transfer the phosphoryl group to the RR, thus modulating its activity.

The role of TCS in Lactobacillus is not well understood, although they are likely involved in quorum sensing, production of bacteriocins (5, 16, 22, 26), and possibly in the stress response (1, 18, 20). The availability of complete genome sequences of two L. casei strains (15, 17) enables a more comprehensive study of the role of TCS in the stress response of this organism. In this study, we used a broad and operational definition of stress, as follows: any deviation from optimal growth conditions that results in a reduced growth rate or lower level of biomass (7).

The genome sequences of L. casei strains BL23 and ATCC 334 harbor 17 putative TCS. For simplicity, we have numerically renamed them TC01 to TC17 (see Table S1 in the supplemental material). TC17 corresponds to the previously characterized system MaeKR (11). The strains and plasmids used in this study are listed in Table S2 in the supplemental material. The primers used are listed in Table S3 in the supplemental material. Insertion mutants were obtained by cloning internal DNA fragments of each RR-encoding gene in plasmid pRV300 (14) and introduced in L. casei BL23 by electroporation (21). BL23-derivative strains harboring complete deletions of the LCABL_02080 (RR01), LCABL_12050 (RR06), and LCABL_19600 (RR12) genes were also obtained by insertion of plasmid pRV300 harboring the regions immediately upstream and downstream of each target gene and subsequent internal recombination. Complementation of the RR01 and RR06 deletions was achieved by cloning of the corresponding genes into the expression vector pT1NX (24).

The growth of L. casei BL23 and that of its derivative RR-defective mutants under reference conditions (MRS at 37°C without shaking) and using MRS supplemented with 0.5% bile, MRS supplemented with 0.6 M NaCl, MRS adjusted to pH 3.75, and MRS at 42°C were compared. Growth was monitored by changes in the optical density at 595 nm (OD₅₉₅) in a microtiter plate reader. At least three independent replicates of each growth curve were obtained. Maximal growth rates (μ_max) and the increment in OD values were considered to compare the performance of L. casei BL23 and its derivative mutants. Significant differences in growth parameters under the reference condition between the wild-type strain and each of the mutants were determined by one-way analysis of variance (ANOVA). Levene's test was used to assess the equality of error variances. To determine whether the responses of the mutant strains to each stress condition assayed were significantly different from that of the wild type, pairwise two-way ANOVA analyses were performed, testing the growth of L. casei BL23 and that of each mutant strain under the reference condition and each of the other stress conditions. We considered a significant difference to be detected if the analysis estimated that both the strain variable and interaction were below P values of 0.01.

Resistance to vancomycin, bacitracin, gramicidin, or nisin was determined in MRS using serial dilutions of the antimicrobial agents. The assays were performed in 96-well microtiter plates incubated for 24 h. The MIC (expressed as the number of μg ml⁻¹) was defined as the lowest concentration of the antimicrobial agent needed to totally inhibit the growth of the bacterial strain. The 50% inhibitory concentration (IC₅₀) was considered the concentration of the antimicrobial agent that diminished the maximal growth rate (μ_max) to 50% of its value under reference growth conditions.

Insertion mutants were obtained for each RR, thus indicating that none of the TCS are essential for growth. RR16 is homologous to RR YycF/VicR, which is essential for growth in other low-G+C-content Gram-positive bacteria (27). Inactivation of the YycF homolog-encoding gene (rrp-3) in Lactobacillus sakei did not result in any significant differences with the parental strain under a number of stress conditions (18). This suggests that the YycFG TCS is not essential in lactobacilli, although it is in the closely related enterococci and streptococci.

The growth rates of the different mutants were similar to that of the wild-type strain under reference conditions, except those of the TC04 and TC11 mutants, which were significantly reduced, and TC12, in which the maximum cell density was significantly lower than that of the wild-type strain (Fig. 1 A; see also Table S4 in the supplemental material). The inactivation of systems TC01, TC06, and TC12 led to major growth defects under stress conditions (Fig. 1 and Table 1; see also Tables S5A and B in the supplemental material).

FIG. 1.

Growth of L. casei BL23 and TCS-defective mutants under different conditions (only strains that displayed significant differences with strain BL23 are shown). Error bars indicate standard deviations (at least three replicates).

TABLE 1.

Results of the growth assays carried out with L. casei BL23 and selected TCS-defective mutants under different growth conditions^a

Condition	Growth parameter	Resultb
		BL23	TC01	TC06	TC12
Bile, 0.5%	μmax	0.17 ± <10−2	NG	NG	0.08 ± <10−2
	Change in OD	1.15 ± 0.14	NG	NG	0.52 ± 0.03
0.6 M NaCl	μmax	0.23 ± 0.01	NG	NG	0.21 ± 0.01
	Change in OD	1.83 ± 0.04	NG	NG	1.81 ± 0.03
pH 3.75	μmax	0.07 ± <10−2	0.03 ± <10−2	0.03 ± 0.01	NG
	Change in OD	0.56 ± 0.07	0.19 ± 0.06	0.14 ± 0.01	NG
42°C	μmax	0.26 ± 0.01	0.24 ± <10−2	0.17 ± 0.01	0.22 ± 0.02
	Change in OD	2.19 ± 0.10	2.07 ± 0.15	0.46 ± 0.03	1.02 ± 0.05

^aSee Tables S4A and B in the supplemental material for data of all mutants and the results of the ANOVA.

^bThe data shown are means and standard deviations. NG, no growth.

The effect of the inactivation of L. casei TCS on tolerance against antibiotics targeted to the cell envelope was also investigated. BL23 and its derivative mutants were resistant to bacitracin and vancomycin, but IC₅₀s for vancomycin in the TC01, TC06, and TC12 mutants were lower than that in the parental strain (see Table S6 in the supplemental material). In contrast, L. casei BL23 was sensitive to bacitracin and nisin (see Table S6), and the responses of the TCS mutants varied. The TC01, TC09, and TC10 mutants were more sensitive to both of the antimicrobials than BL23, whereas three other mutants (TC06, TC11, and TC12) were more sensitive to nisin only. Three mutants were more resistant to bacitracin (TC15, TC16, and TC17), and one mutant was more resistant to both of the antimicrobials (TC04).

To determine the possible polar effects of the insertional inactivation of systems TC01, TC06, and TC12, strains carrying deletions of RR01, RR06, and RR12 (ΔRR01, ΔRR06, and ΔRR12, respectively) and the corresponding complemented strains (except for the ΔRR12 mutant, which was impervious to transformation) were obtained. BL23 and ΔRR01 strains grew similarly under reference conditions. The growth of the ΔRR01 mutant under different stress conditions was similar to that observed for the insertional mutant (see Fig. S1A to D and Table S6 in the supplemental material), and the effects of the mutation were relieved in the complemented strain ΔRR01-c, except in the presence of 0.6 M NaCl, where the ΔRR01 mutant was able to grow (see Fig. S2B in the supplemental material). Therefore, the growth defect observed with salt was possibly due to a polar effect on the expression of some of the genes located downstream. System TC01 is homologous to the rrp-31 hpk-31 (LSA0277-78) system of L. sakei and the CroRS system of Enterococcus faecalis (2, 12, 13). Inactivation of the cognate RR in L. sakei led to premature arrest of growth under reference conditions (MRS at 30°C), poor growth at a high temperature (39°C), sensitivity to heat shock, aeration, H₂O₂, and higher resistance to vancomycin (18). These results contrast with our observations of the ΔRR01 mutant, thus suggesting that these homologous TCS have different physiological roles. The sensitivity of the ΔRR01 mutant to bile and the cell envelope-targeted antimicrobials bacitracin and vancomycin suggests that TC01 is involved in cell envelope stress tolerance.

The ΔRR06 and ΔRR12 mutants showed phenotypes very similar to those of their corresponding insertional mutant strains (see Fig. S2 and Table S6 in the supplemental material). Complementation of the ΔRR06 mutant with pT1-RR06 relieved the effects of the mutation, and we conclude that the effects observed were due to the inactivation of RR06 and not to polar effects on downstream genes. Though complementation of the ΔRR12 mutant was not achieved, the similarity of the phenotypes of TC12 and ΔRR12 strains suggests that the observed effects are due mainly to the inactivation of RR12.

System TC06 is homologous to the Bacillus subtilis YclJ-YclK TCS, which is activated under oxygen limitation conditions (10), and E. faecalis system Err06-Ehk06. Inactivation of this system in E. faecalis V583 resulted in a heat- and sodium dodecyl sulfate (SDS)-sensitive phenotype (6). Furthermore, it has been shown that Err06 is involved in resistance to H₂O₂ in E. faecalis JH2-2 (19). L. casei TC06 was very sensitive under most stress conditions assayed. In this sense, it is worth noting the presence of a putative transfer-messenger RNA (tmRNA)-carrying gene located upstream of the TC06-encoding genes. Interestingly, the involvement of a homologous tmRNA of Escherichia coli in the cell envelope stress response has been recently demonstrated (8), although both the involvement of TC06 in the regulation of tmRNA expression and the actual role of this tmRNA remain to be established.

System TC12 is paralogous to system TC09, but in contrast to TC12, inactivation of TC09 resulted only in higher sensitivity to bacitracin and nisin than that of BL23 (see Table S6 in the supplemental material). Systems TC09 and TC12 are homologous to the three paralogous TCS of B. subtilis, BceRS, YvcPQ, and YxdJK, involved in the cell envelope stress response (9). These three TCS are located next to genes encoding ABC transporters. Similarly, both TC09 and TC12 are located next to genes encoding putative ABC transporters. The lower resistances of TC09 against bacitracin and nisin and of TC12 against nisin suggest that these systems may also be involved in the cell envelope stress response. However, the growth defects of the TC12 mutant, particularly at a low pH, suggest that the functional role of this system in L. casei is quite different than that of its homolog, BceRS in B. subtilis.

In summary, this study shows that some TCS play a major role in the physiology of L. casei and its adaptation under changing environmental conditions. The detailed study of these systems should provide valuable insight into understanding the performance of this organism under conditions of industrial production and in the gastrointestinal habitat.