Acetate Activates Lactobacillus Bacteriocin Synthesis by Controlling Quorum Sensing

ABSTRACT

Bacteriocins are useful for controlling the composition of microorganisms in fermented food. Bacteriocin synthesis is regulated by quorum sensing mediated by autoinducing peptides. In addition, short-chain fatty acids, especially acetic acid, reportedly regulate bacteriocin synthesis. Five histidine kinases that regulated the synthesis of bacteriocins were selected to verify their interactions with acetate. Acetate activated the kinase activity of PlnB, SppK, and HpK3 in vitro and increased the yield of their cognate bacteriocins plantaricin EF, sakacin A, and rhamnosin B in vivo. The antimicrobial activity against Staphylococcus aureus of the fermentation supernatants of Lactobacillus plantarum, Lactobacillus sakei, and Lactobacillus rhamnosus with addition of acetate increased to 298%, 198%, and 289%, respectively, compared with that in the absence of acetate. Our study elucidated the activation activity of acetate in bacteriocin synthesis, and it might provide a potential strategy to increase the production of bacteriocin produced by Lactobacillus.

IMPORTANCE Bacteriocins produced by lactic acid bacteria (LAB) are particularly useful in food preservation and food safety. Bacteriocins might increase bacterial competitive advantage against the indigenous microbiota of the intestines; at the same time, bacteriocins could limit the growth of undesired microorganisms in yogurt and other dairy products. This study confirmed that three kinds of histidine kinases were activated by acetate and upregulated bacteriocin synthesis both in vitro and in vivo. The increasing yield of bacteriocins reduced the number of pathogens and increased the number of probiotics in milk. Bacteriocin synthesis activation by acetate may have a broad application in the preservation of dairy products and forage silage.

INTRODUCTION

For thousands of years, humankind has prolonged foodstuff storage using lactic acid bacteria (LAB) to control microbial succession during fermentation or inhibit spoilage organisms’ growth (1, 2). The preservative effect of LAB depends on the antimicrobial substances that they produce, such as short-chain fatty acids, hydrogen peroxide, and bacteriocins, to expel competitors (3, 4). Bacteriocins are antimicrobial peptides or proteins that increase membrane permeability via pore formation, DNA transcription inhibition, or induction of reactive oxygen species production within cells (5). Bacteriocins provide a competitive advantage to LAB, enabling the colonization of new ecosystems or preventing the invasion of their own by exogenous microorganisms (6, 7).

The proposed synthesis of LAB bacteriocins is regulated by quorum-sensing systems consisting of three components: the autoinducing peptide (AIP) and a two-component regulatory system that has a membrane-bound histidine kinase (HK) sensor and an intracellular response regulator (RR) (8, 9). The AIP is secreted at a low but constant rate by cells in the bacterial population, reflecting the cell density during growth. At a critical threshold concentration, the AIP activates the HK and transfers phosphoryl groups to the response regulator. Phosphorylation of the RR typically results in activation of effector domain function, which is commonly DNA-binding activity, resulting in transcriptional regulation of bacteriocin genes involved in output responses (10–12). Additionally, recent reports on the biosynthesis of bacteriocin revealed that environmental conditions such as growth temperatures, ionic strength, and pH played a critical role in the synthesis of bacteriocin (11, 13). Several studies showed that, for some strains, the antimicrobial activity was measured only in solid plates that contained indicator strains, and no bacteriocins were detected in liquid culture. In addition, some researchers have confirmed that the production of bacteriocins in LAB cultured alone is deficient, while many Gram-positive bacteria triggered the synthesis of bacteriocins (14). Nilsson et al. found that acetate activates the synthesis of bacteriocin A9b in Carnobacterium piscicola, which shows a dose-dependent regulation at acetate concentrations of up to 10 to 20 μM (15). Therefore, more comprehensive views of the bacteriocin production mechanism are needed (16).

In a previous study (17), we found that acetate promoted bacteriocin synthesis in Lactobacillus plantarum. Based on a phylogenetic tree constructed with several HKs, which included PlnB from L. plantarum, four HKs, those that regulate the synthesis of sakacin P, enterocin A, nisin A, and rhamnosin B, were selected. The activation effect of acetate on these HKs was determined both in vitro and in vivo, and it was found to improve the production of bacteriocins.

RESULTS

Homologous analysis of HKs in Lactobacillus spp.

In a previous study, we analyzed the binding position between acetate and the membrane protein HK PlnB (from L. plantarum), and clarified the acetate regulation mechanism in plantaricin synthesis (our unpublished data). To gain more insight into the phenomenon by which acetate activates bacteriocin synthesis in LAB, the protein sequences of HKs were BLAST searched using the NCBI BLAST interface, and a phylogenetic tree was constructed (Fig. 1a). Based on the genetic relationships found, SppK, EntK, NisK, and HpK3 from Lactobacillus sakei, Enterococcus faecium, Lactococcus lactis, and Lactobacillus rhamnosus, which regulate the synthesis of sakacin A, enterocin A, nisin A, and rhamnosin B, respectively, were selected for further analysis. The transmembrane domains of the selected HKs were predicted, and five transmembrane domains were identified (see Table S1 in the supplemental material), except for NisK, which has two transmembrane domains. The extracellular loops of SppK, EntK, and HpK3 showed reorganization sites homologous to those of PlnB, containing basic amino acids, i.e., arginine, lysine, and histidine (Fig. 1b). However, the extracellular loop of NisK was different from those of the other HKs, which might lead to different enzyme properties.

FIG 1.

Neighbor-joining tree of PlnB of L. plantarum 163 and homologous histidine kinases. (a) PlnB of L. plantarum 163 was used to run a BLAST search at the NCBI BLAST interface, and 11 homologous kinases were selected and aligned by ClustalW to construct a neighbor-joining tree using the bootstrap method (1,000 bootstrap replications) and Poisson model of MEGA-X (version 10.1.1). (b) Extracellular loops of a subset of HKs were predicted, and hypothetical acetate recognition sites of HKs are shown in the red box.

Expression of HKs and determination of their enzyme properties.

plnB, sppK, entK, hpk3, and nisK were amplified, ligated to the vector pET30a(+), and expressed in Escherichia coli C43(DE3). Each recombinant protein was purified using nickel-nitrilotriacetic acid (Ni-NTA) columns (Fig. 2a to e), and then their His tags were removed using an enterokinase (Fig. 2f). Subsequently, the optimal acetate concentration, pH, and temperature of HKs were analyzed to determine each enzyme’s properties.

FIG 2.

Expression and purification of HKs. (a to e) Expression and purification of SppK, EntK, NisK, HpK3, and PlnB, respectively. Lane 1, membrane proteins of E. coli C43(DE3) without IPTG. Lane 2, HKs induced by IPTG. Lane 3, purified HKs (using an Ni-NTA column). (f) His tags were removed using enterokinase, and SDS-PAGE was used to determine the purity of the HKs. Lanes 1 to 5, SppK, EntK, NisK, HpK3, and PlnB, respectively. M, protein marker 26616.

The optimal pH was 7.0, and the temperature ranged from 35 to 40°C (Fig. 3a and b). In addition, the optimal acetate concentrations were 12 μM for PlnB and EntK and 6 μM for SppK and HpK3 (Fig. 3c). However, NisK cannot be activated by acetate (Fig. S1). The K_m of the HKs ranged from 242 to 274 μM, whereas the k_cat ranged from 76.58 to 119.53 min⁻¹ (Table 1; also, see Fig. S2).

FIG 3.

Enzyme characteristics of HKs. (a) Relative activities of SppK, EntK, HpK3, and PlnB with 12 μM acetate at different temperatures. (b) Relative activities of SppK, EntK, HpK3, and PlnB with 12 μM acetate at different pHs. (c) Relative activities of SppK, EntK, HpK3, and PlnB at optimal temperature and pH in the presence of 0 to 600 μM acetate.

TABLE 1.

Enzyme properties of 4 histidine kinases^a

Kinase	Optimal acetate concn (μM)	Optimal pH	Optimal temp (°C)	Km of ATP (μM)	kcat (min−1)
PlnB	12	7.0	40	272 ± 39	76.58 ± 5.16
SppK	6	7.0	40	254 ± 29	90.30 ± 4.76
EntK	12	7.0	35	274 ± 42	84.52 ± 5.95
HpK3	6	7.0	35	242 ± 63	119.53 ± 13.92
NisK	ND	ND	ND	ND	ND

^aND, not detected.

Effect of short-chain fatty acids on the activity of HKs in vitro.

LAB produce various organic acids during their growth process, such as formic acid, acetic acid, and lactic acid (18). To assess the activation of short-chain fatty acids on HKs, formate, acetate, propionate, butyrate, or lactate was used to activate SppK, PlnB, EntK, HpK3, and NisK. The concentration of ADP in the sample was measured by the luciferase assay, given that the kinase activity of HKs can be determined by calculating ATP consumption (Fig. S1). The results showed that formate, propionate, propionate, butyrate, and lactate had weaker activation effects on HKs than acetate, except for NisK, which could not be activated by any of the short-chain fatty acids tested (Fig. 4; also, see Fig. S3). For example, the activities of SppK induced by propionate and EntK induced by lactate were about 50% of those obtained with acetate. This suggested that short-chain fatty acids, mainly acetate, effectively activated the HKs in the quorum-sensing system in vitro, which phosphorylated the response regulator by hydrolyzing ATP (Fig. S4).

FIG 4.

Activation of HKs by short-chain fatty acids. (a to d) Relative activities of SppK, PlnB, EntK, and HpK3 induced by optimal concentrations of formate, propionate, butyrate, and lactate, as measured by luciferase assay. The HK activity induced by acetate was considered the positive control. Lowercase letters in each graph indicate statistically significant differences in HK activity (P < 0.05).

Acetate activates the synthesis of bacteriocins in vivo.

The transcription of the bacteriocin genes under study was detected to confirm the acetate activation of HKs in vivo. The results showed that the mRNA levels of sppA, entA, rhmB, plnE, and plnF increased significantly in the presence of acetate. For example, the mRNA level of plnF increased by 14.8-fold in the logarithmic growth phase (12 h) and 9.5-fold in the stationary phase (24 h) after addition of acetate for 2 h. In contrast, acetate could not activate the transcription of nisA (Fig. 5a), in accordance with the in vitro results for NisK. Subsequently, the production of bacteriocins was measured by high-performance liquid chromatography (HPLC), indicating that acetate increased the yield of sakacin A, rhamnosin B, plantaricin E, and plantaricin F to 174.7%, 127.7%, 142.6%, and 161.8%, respectively. (Fig. 5b; also, see Fig. S5 and Tables S4 to S7). The antimicrobial activities of L. sakei, L. plantarum, and L. rhamnosus against Staphylococcus aureus increased to 198%, 298%, and 289% and those against E. coli increased to 156%, 205% and 213%, respectively (Fig. 5c and d). Surprisingly, the antimicrobial activity of the bacteriocin produced by E. faecium against S. aureus did not significantly increase after addition of extra acetate, even when the mRNA levels of entA had increased (Fig. 5c).

FIG 5.

Acetate activates bacteriocin synthesis in Lactobacillus spp. (a) Effect of acetate on the transcription of bacteriocin genes of Lactobacillus strains grown in MRS2 medium. (b) Bacteriocins were purified and concentrated by SPE and detected by HPLC-mass spectrometry. (c and d) Relative antimicrobial activity of clear supernatant against S. aureus ATCC 14458 and enterotoxigenic E. coli strain CICC 10667 grown in MRS2 broth supplemented with different acetate concentrations.

DISCUSSION

Bacteria within a population act socially by coordinating their activities through quorum sensing, which improves access to nutrients, promotes defense against competitors, and enhances survival under adverse environmental conditions. Generally, some bacteriocins produced by lactic acid bacteria regulate quorum sensing (two-component system) depending on the production and extracellular release of AIP, such as PlnA and pINC8IF of L. plantarum, inducer pheromone (IP-673 and IP-706) of L. sakei, EntF of Enterococcus faecium and nisin A of L. lactis. Also, research has found that bacteriocin of L. plantarum NC8 cannot be synthesized in pure culture (11), yet the antibacterial activity increased to 2,560 bacteriocin units (BU)/ml when it is cocultured with L. lactis MG1363 (14). This indicated that bacteriocin synthesis was regulated by signals other than AIP.

Acetic acid has multiple functions in the intestine; for example, it coordinates neutrophil and ILC3 responses against Clostridium difficile through its cognate receptor-free fatty acid receptor 2 and protects against respiratory syncytial virus infection through a GPR43-type 1 interferon response (19, 20). Additionally, acetate is a signaling molecule that increases the yield of 2,3-butanediol by upregulating the transcription of alsR and alsS (21). Acetate is also an activator of the synthesis of bacteriocin A9b in C. piscicola (22). In this study, acetate activated bacteriocin synthesis in L. plantarum, L. sakei, and L. rhamnosus both in vivo and in vitro.

Bacteriocin production is a powerful weapon for LAB to compete with pathogens. Controlling bacteriocin synthesis may be a simple way to decrease the number of pathogens in dairy products (23). Additionally, LAB bacteriocins are diverse, show a broad spectrum of antibacterial activity, and are generally recognized as safe preservatives (24). Purified bacteriocins and bacteriocin-producing LAB have been widely used in food preservation for inhibiting the growth of pathogens such as Listeria and S. aureus (18, 25). Improving the concentration of bacteriocins is a feasible method for food safety. Here, acetate activated the kinase activity of SppK, PlnB, and HpK3 in vitro, which regulated the synthesis of sakacin P, plantaricin F, and rhamnosin B in vivo. These bacteriocins enhanced the competitiveness of Lactobacillus spp. and inhibited the growth of pathogens in yogurt. LAB in dairy products is an effective method to inhibit pathogens' growth and improve food safety. In many cases, LAB has been shown to inhibit the growth of S. aureus and L. monocytogenes by reducing the pH and producing hydrogen peroxide and bacteriocins (26). According to our findings, in fermented dairy products, fermented vegetables, and silage, the number of Lactobacillus organisms, especially that of L. plantarum, can be increased by adding sodium acetate, thereby improving the probiotic function of the product, or inhibiting the growth of pathogens.

Conclusion.

LAB quorum-sensing HKs, such as PlnB, SppK, EntK, and HpK3, were expressed in E. coli and purified to detected their activities induced by acetate. The result showed that PlnB, SppK, EntK, and HpK3 were activated by 6 to 12 μM acetate in vitro. However, only PlnB, SppK, and HpK3 were activated to improve the mRNA level and yield of bacteriocin in vivo. The yield of sakacin A, rhamnosin B, plantaricin E, and plantaricin F increased to 174.7%, 128.7%, 142.6%, and 161.8%, respectively.

MATERIALS AND METHODS

Microbial strains and medium.

Microbial strains used to amplify HKs and for yogurt fermentation are shown in Table 2. LAB were cultured in an deMan-Rogosa-Sharpe (MRS) medium at 30°C. E. coli and S. aureus were cultured in Luria-Bertani (LB) medium at 37°C. The restriction enzymes BamHI and XhoI and T4 DNA ligase were obtained from TaKaRa (Dalian, China). Primers were synthesized by GenScript (Nanjing, China). MRS2 medium corresponded to MRS medium without sodium acetate.

TABLE 2.

Strains and vectors^a

Strain or vector	Description	Reference
Strains
Lactobacillus plantarum 163	Identified from koumiss, producing plantaricin EF, amplifies PlnB	30
Lactobacillus sakei ATCC 15521	Produces sakacin P, amplifies PlnB	Lab stock
Lactobacillus rhamnosus LGG	Produces bacteriocin rhamnosin B and C, amplifies histidine kinase HpK3	Lab stock
Enterococcus faecium ATCC 19434	Produces enterocin A, amplifying EntK	Lab stock
Lactococcus lactis ATCC 11454	Produces nisin A, amplifies NisK	Lab stock
Escherichia coli C43(DE3)	F− ompT hsdSB (rB− mB−) gal dcm(DE3); expresses histidine kinases	Zoman Biotechnology, Beijing, China
Staphylococcus aureus Rosenbach ATCC 144580	Indicator strain; produces enterotoxin B	Lab stock
E. coli CICC 10667 (ETEC)	Indicator strain; contains heat-labile enterotoxin gene (LT) and human heat-stable enterotoxin gene (STIb)	Lab stock
Vectors
pET30a(+)	Expression vector, Kanr, T7 promoter, lac operator	Novagen
pET30a-plnB	pET30a with PlnB	This study
pET30a-SppK	pET30a with SppK	This study
pET30a-EntK	pET30a with EntK	This study
pET30a-NisK	pET30a with NisK	This study
pET30a-HpK3	pET30a with HpK3	This study

^aATCC, American Type Culture Collection; CICC, China Center of Industrial Culture Collection; ETEC, enterotoxigenic Escherichia coli.

Selecting homologous kinases of PlnB from LAB.

The protein sequence of PlnB of L. plantarum 163 was BLAST searched using the NCBI BLAST tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Homologous proteins of PlnB (GenBank numbers CCC77923.1, AAW79055.1, AXT15952.1, ALZ52862.1, CAR88282.1, KRL52345.1, YP_193512.1, BAF74793.1, ACT99734.1, ATH76030.1, and CAA80467.1) were selected and aligned by ClustalW. A neighbor-joining tree was constructed by the bootstrap method (1,000 bootstrap replications) and Poisson model of MEGA-X (version 10.1.1). Then, SppK (AAW79055.1), EntK (ACT99734.1), NisK (CAA80467.1), and HpK3 (CAR88282.1) were selected for further research. Among these, SppK, EntK, and NisK are related to the synthesis of sakacin P, enterocin A, and nisin A, respectively. whereas HpK3 regulates rhamnosin B synthesis (the gene cluster was analyzed [Fig. S6]).

Expression and purification of PlnB-homologous kinases.

As described in Table 2, HKs were amplified using different primers sets (Table S2) and digested with BamHI and XhoI. They were then linked to pET30a and transformed into E. coli C43(DE3) (27), and positive clones were selected on LB medium plates supplemented with 50 μg ml⁻¹ kanamycin. Then, 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) was added to the medium, and cultures were incubated for 16 h at 16°C to express HKs. Subsequently, the cells were harvested by centrifugation and washed twice with Tris-EDTA (TE) buffer. Cells were disrupted by ultrasonication, and the membrane was harvested by ultracentrifugation at 100,000 × g for 30 min. The precipitate was dissolved in TE buffer containing 2% dodecyl maltoside, and HKs were purified using an Ni-NTA column (Thermo Fisher, MA, USA) according to a previously described method (28). The His tags attached to purified proteins were removed by enterokinase (Merck, NJ, USA) and purified using an Ni-NTA column. The flowthrough was concentrated using an ultrafiltration tube (10 kDa; Millipore, MA, USA), and proteins were detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

Determination of the kinase activity of PlnB-homologous kinases.

Kinase activity (ATPase) of HKs was determined by analyzing the ADP concentration as follows: 12 μM sodium acetate, 10 mg liter⁻¹ HK, and 450 μM ATP in TE buffer (50 mM, pH 7.5) were mixed and incubated at 30°C for 30 min. Then, ADP-Glo reagent was added to terminate the kinase reaction and deplete the remaining ATP. The kinase detection reagent was added to convert ADP into ATP, and the newly synthesized ATP was measured using the luciferase/luciferin reaction based on the ADP-Glo kinase assay (Promega, Madison, USA) protocol (https://www.promega.com.cn/resources/protocols/technical-manuals/0/adp-glo-kinase-assay-protocol/).

Determination of the enzyme properties of purified HKs.

Kinase activities of HKs were detected at different concentrations of sodium acetate (0, 0.5, 1, 3, 6, 12, 60, 120, and 600 μM) to determine the optimal concentration of acetate. Then, the kinase activities of HKs were detected in a series of pH values (4, 5, 6, 7, 8, and 9) and temperatures (20, 25, 30, 35, 40, and 45°C). The K_m and V_max of HKs were detected at different ATP concentrations (0, 50, 100, 200, 300, 400, and 500 μM). The K_m and V_max of HKs were calculated using the Michaelis-Menten equation (Origin 8.0; OriginLab, MA, USA).

Identification of the activator of PlnB-homologous kinases.

Formate, acetate, propionate, butyrate, and lactate were added to the reaction system to final concentrations of 3, 6, 12, 60, and 120 μM. Then, the kinase activity of the HKs was tested using the ADP-Glo kinase assay as described above. The kinase activity of HKs induced by acetate was used as a positive control.

Detection of the mRNA levels of the bacteriocin genes by real-time PCR.

L. plantarum 163, L. sakei ATCC 15521, L. rhamnosus LGG, E. faecium ATCC 19434, and L. lactis ATCC 11454 were cultured in MRS2 medium for 48 h at 30°C. They were then inoculated into fresh MRS2 medium and cultured to mid-logarithmic phase, after which they were induced with 12 μM sodium acetate for 2 h. Subsequently, cells were harvested by centrifugation at 5,000 × g for 10 min and washed twice with TE buffer (10 mM Tris-HCl, 1 mM EDTA [pH 8.0]). The RNA was extracted using TRIzol according to the TRIzol Plus RNA purification kit manual (Invitrogen, CA, USA). Next, the RNA was reverse transcribed to DNA, and the transcription of the bacteriocin genes plnE, plnF, sppA, entA, nisA, and rhmB was detected by real-time PCR and compared with that of the reference 16S gene (BACT1369F and PROK1541R) (Table S3) using a PowerUp SYBR green master mix (Thermo Fisher, MA, USA) and a StepOnePlus real-time PCR system (Applied Biosystems, CA, USA).

Detection of the bacteriocin levels by SPE-HPLC.

L. plantarum 163, L. sakei ATCC 15521, L. rhamnosus LGG, and E. faecium ATCC 19434 were cultured in MRS2 medium for 48 h at 30°C in the presence of 12 μM sodium acetate, whereas the control group was grown in the absence of sodium acetate. The supernatant of each broth was then loaded onto a preprocessed solid-phase extraction (SPE) column (Oasis MCX, 60 mg/3 ml; Waters, MA, USA). The impurities were removed using 3 ml of methanol, followed by 3 ml of deionized water. Bacteriocins were eluted with 3 ml of methanol (containing 5% [vol/vol] ammonia). Subsequently, bacteriocins were measured by HPLC (Dionex, CA, USA) in 90% water (0.1% trifluoroacetic acid) and 10% methanol at 259 nm using an Agilent Eclipse XDB C₁₈ (5-μm, 4.6- by-250 mm) column. The antimicrobial activity of each bacteriocin for S. aureus Rosenbach ATCC 14458 was determined through microdilution in 96-well microtiter plates according to the Clinical and Laboratory Standards Institute procedures (29).

Statistical test.

The experiment employed a randomized design with four treatments and four repeats per treatment, and the average was calculated at each point. The kinase activity and relative activity of HKs were compared using the one-way analysis of variance (ANOVA), and post hoc comparisons were made using Tukey’s multiple-comparison tests in SPSS (IBM, version 17.0). Differences were considered statistically significant at a P value of <0.05.