Pre-Exposure of Foodborne Staphylococcus aureus Isolates to Organic Acids Induces Cross-Adaptation to Mild Heat

ABSTRACT Staphylococcus aureus is a typical enterotoxin-producing bacterium that causes food poisoning. In the food industry, pasteurization is the most widely used technique for food decontamination. However, pre-exposure to an acidic environment might make bacteria more resistant to heat treatment, which could compromise the bactericidal effect of heat treatment and endanger food safety. In this work, the organic acid-induced cross-adaptation of S. aureus isolates to heat and the associated mechanisms were investigated. Cross-adaptation area analysis indicated that pre-exposure to organic acids induced cross-adaptation of S. aureus to heat in a strain-dependent manner. Compared with other strains, S. aureus strain J15 showed extremely high heat resistance after being stressed by acetic acid, citric acid, and lactic acid. S. aureus strains J19, J9, and J17 were found to be unable to develop cross-adaptation to heat with pre-exposure to acetic acid, citric acid, and lactic acid, respectively. Analysis of the phenotypic characteristics of the cell membrane demonstrated that the acid-heat-cross-adapted strain J15 retained cell membrane integrity and functions through enhanced Na+K+-ATPase and FoF1-ATPase activities. Cell membrane fatty acid analysis revealed that the ratio of anteiso to iso branched-chain fatty acids in the acid-heat-cross-adapted strain J15 decreased and the content of straight-chain fatty acids exhibited a 2.9 to 4.4% increase, contributing to the reduction in membrane fluidity. At the molecular level, fabH was overexpressed with preconditioning by organic acid, and its expression was further enhanced with subsequent heat exposure. Organic acids activated the GroESL system, which participated in the heat shock response of S. aureus to the subsequent heat stress. IMPORTANCE Cross-adaptation is one of the most important phenotypes in foodborne pathogens and poses a potential risk to food safety and human health. In this work, we found that pretreatment with acetic acid, citric acid, and lactic acid could induce subsequent heat tolerance development in S. aureus. Various S. aureus strains exhibited different acid-heat cross-adaptation areas. The acid-induced cross-adaptation to heat might be attributable to membrane integrity maintenance, stabilization of the charge equilibrium to achieve a normal internal pH, and membrane fluidity reduction achieved by decreasing the ratios of anteiso to iso fatty acids. The fabH gene, which is involved in fatty acid biosynthesis, and groES/groEL, which are related to heat shock response, contributed to the development of the acid-heat cross-adaptation phenomenon in S. aureus. The investigations of the stress cross-adaptation phenomenon in foodborne pathogens could help optimize food processing to better control S. aureus.

With further thermal treatment, it was found that after cross-adaptation to acid and heat, the leakage amount of nucleic acids and proteins in the S. aureus strain J15 was still lower than that of the control-group strains (J19, J9, and J17). The nucleic acid leakage amounts in S. aureus strains J19, J9, and J17 were 1.50, 1.28, and 1.65 times that of S. aureus strain J15, respectively, and the protein leakage amounts in the control group were 2.47, 2.19, and 2.11 times that of the acid-heat-cross-adapted strain J15, respectively. Mild acidic stress has been shown to increase the expression of heat stress-related genes and to activate heat shock proteins such as DnaK, GroELS, and Clp protease, leading to resistance to heat stress (25).
The cell membrane potential of cross-adapted S. aureus. Rhodamine 123, a cationic fluorescent dye, was used for monitoring the membrane potential of S. aureus cells. Rhodamine 123 was released when the membrane potential was compromised by external stresses, resulting in a decrease in the fluorescence intensity (26). As shown in Fig. 1A to C, after exposure to acid stress, the membrane potential of all S. aureus strains was decreased compared with that of the unstressed strains. The average fluorescence intensities of rhodamine 123 in S. aureus strain J15 under acetic acid, citric acid, and lactic acid stress were 286. 49, 293.41, and 270.62 arbitrary units (AU), respectively, which were significantly higher (P , 0.05) than those of the control strains J19, J9, and J17. This result indicated that the variations in the ability to cope with acid stress might result in differences in the membrane potentials of various bacterial strains during a state of cross-adaptation between acid and heat. With subsequent exposure to heat stress, the rhodamine 123 mean fluorescence intensity of S. aureus strain J15 was further decreased, but it was still lower than that of the control strains (J19, J9, and J17) (P , 0.05) ( Fig. 1A to C). Acid-adapted S. aureus strains activated the channels that regulate ion transport on the cell membrane to maintain the charge balance through the transmembrane transport of differently charged ions (27). Under subsequent heat stress, acid-adapted S. aureus strains are able to maintain cell homeostasis, minimize changes in membrane potential, and improve bacterial survival (27).
The cell membrane fluidity serves as a marker for molecular mobility inside the lipid bilayer, and it is strongly correlated with the physiological processes of bacterial cells, including ion transport, growth, and reproduction (28). In this work, the cell membrane fluidity was analyzed by a 1,6-diphenyl-1,3,5-hexatriene (DPH) probe, which could be inserted into the acyl chain of membrane fatty acids. The reduction in cell membrane fluidity could compromise the interaction between cell membrane lipids and DPH molecules. The rotation of DPH was restricted, which led to an increase in the polarization signal of DPH. As shown in Fig. 1D to F, the polarization values of S. aureus strain J15 under acetic acid-, citric acid-, and lactic acid-heat stress were 0.35, 0.34, and 0.34, respectively, while those of the control strains J19, J9, and J17 were 0.32, 0.31, and 0.31, respectively. The polarization value was inversely proportional to the cell membrane fluidity. The cell membrane fluidity of the acid-heat-adapted S. aureus strain was lower than that of the susceptible counterparts. The reduction in membrane fluidity has been attributed to the adaptation to acid stress (29). With subsequent heat treatment, the polarization values of acetic acid-, citric acid-, and lactic acid-stressed S. aureus strain J15 were further enhanced to 0.37, 0.38, and 0.38, respectively, which were higher than those of S. aureus strains J19, J9, and J17. This result indicated that the strains with cross-adaptation phenomenon exhibited decreased membrane fluidity. Alvarez-Ordonez et al. (30) also discovered that acid pretreatment decreased the membrane fluidity (with an increase in the polarization values) of Salmonella strain CECT 4384 by lowering the ratios of unsaturated to saturated fatty acids in cell membrane, which promoted the development of heat tolerance.
Activities of Na + K + -ATPase and F o F 1 -ATPase of cross-adapted S. aureus. Na 1 K 1 -ATPase and F o F 1 -ATPase are critical transmembrane enzymes that perform critical functions in maintaining the membrane potential, and their activities were evaluated in this work. Na 1 K 1 -ATPase is responsible for exporting sodium ions and importing potassium ions to keep a stable membrane potential in bacterial cells (31). F o F 1 -ATPase is a multisubunit enzyme and functions as a membranous channel for proton translocation (32). It plays an important role in the acid adaptation of bacterial cells through the efflux of protons to maintain the internal pH to a normal state (33). As shown in Fig. 2A to C, all strains of S. aureus exhibited an increase in the activity of Na 1 K 1 -ATPase when exposed to acid stress. The enzyme activities of S. aureus strain J15 under acetic acid-, citric acid-, and lactic acid-heat stress were 0.316, 0.133, and 0.182 mmol/h, respectively, while those of the control group (S. aureus strains J19, J9, and J17) were 0.165, 0.058, and 0.062 mmol/h, respectively. The increase in the activity of Na 1 K 1 -ATPase contributed to maintaining the normal function of ion exchange inside and outside the cell, which put S. aureus in a stable physiological state in an acid environment (31). With further thermal treatment, strain J15 under acetic acid-, lactic acid-and citric acid- Acid-Heat Cross Adaptation of S. aureus Microbiology Spectrum heat stress had Na 1 K 1 -ATPase activities of 0.320, 0.172, and 0.098 mmol/h, respectively, which were higher than those of strains J19, J9, and J17 (P , 0.05). As shown in Fig. 2D to F, after exposure to acetic acid, citric acid, and lactic acid, the F o F 1 -ATPase activity of S. aureus strain J15 was increased to 1.327, 1.031, and 0.892 mmol P i /min/mg protein, respectively, which were significantly higher than those of the control strains (P , 0.05). Intracellular protons were pumped with the aid of F o F 1 -ATPase to maintain the harmony of the interior pH and increase the bacterial survival rate (33). Under acid-heat cross-adaptation, the activity of F o F 1 -ATPase in the S. aureus strain J15 was significantly higher than that in the control groups (J19, J9, and J17) (P , 0.05). With further heat treatment, as shown in Fig. 2D to F, the F o F 1 -ATPase activities of strain J15 with acid-heat cross-adaptation phenomenon were increased to 1.464, 1.221, and 1.021 mmol P i /min/mg protein, respectively, which demonstrated that F o F 1 -ATPase might play an important role in the development of acid-heat crossadaptation. F o F 1 -ATPase has been reported as the auxiliary chaperone of heat shock protein (HSP) complexes (34). HSP recognizes proteins with abnormal conformations and prevents nonspecific aggregation induced by heat stress (34,35). Upon acid-heat cross-adaptation, F o F 1 -ATPase may be related to the increase in the expression of HSPs, thereby enhancing the resistance of S. aureus to subsequent heat exposure.
The relative content and change of fatty acid profiles in the S. aureus cell membrane. Modification of cell membrane fatty acid is one of the important strategies employed by bacterial cells against external stress (36). In this work, the fatty acid profiles of S. aureus before and after treatments were analyzed. The cell membrane of S. aureus consists of straight-chain fatty acids (SCFAs), branched-chain fatty acids (BCFAs), and a small amount of unsaturated fatty acids (USFAs) (Tables 3 to 5). BCFAs were further divided into iso-BCFAs and anteiso-BCFAs (36). The ratio of anteiso-BCFAs to iso-BCFAs was related to the cell membrane fluidity of S. aureus. The methyl groups of anteiso-BCFAs were farther from the end of the fatty acid chain than those of a -, not detected. Different letters indicate significant differences (P , 0.05) in the ratios of SCFAs, anteiso-BCFAs, or iso-BCFAs or the ratios of anteiso-BCFAs to iso-BCFAs (anteiso/iso).
iso-BCFAs. Thus, the decrease in anteiso-BCFAs contributed to the compromise in the cell membrane fluidity (37). When exposed to the organic acids, the ratio of anteiso-BCFAs to iso-BCFAs was significantly decreased in S. aureus strain J15 (P , 0.05). The control strains J19, J9, and J17 did not show as great a decrease. The decrease in anteiso-BCFAs was mainly attributed to the reduction of anteiso-C 15:0 and anteiso-C 19:0 . The increase of iso-BCFAs mainly resulted from the increase of iso-C 15:0 and iso-C 17:0 . Therefore, the reduction in the ratio of anteiso-BCFAs to iso-BCFAs could explain the decrease in the cell membrane fluidity of strain J15, as indicated above.
In addition, a decrease in SCFAs has also been associated with an increase in membrane fluidity (36). After exposure to acetic acid-, citric acid-, and lactic acid-heat stress, the content of SCFAs in the cell membrane of S. aureus strain J15 increased from 17.38%, 16.88%, and 15.58% to 21.47%, 19.79%, and 20.02%, respectively. The increase in SCFAs was mainly attributed to the increases in C 16:0 and C 18:0 in the cell membrane of strain J15. The contents of SCFAs in the control S. aureus strains (J19, J9, and J17) decreased slightly from 17.79%, 8.82%, and 17.89% to 17.33%, 8.76%, and 16.69%, respectively. It has been reported that Gram-positive bacteria such as S. aureus promote the synthesis of SCFAs to reduce the fluidity of cell membranes upon adverse stress (e.g., high temperature) (38). Therefore, the enhancement of SCFAs in cell membranes could contribute to the emergence of the acid-heat cross-adaptation phenomenon.
The regulation of cell membrane fatty acid synthesis. In S. aureus, BCFAs accounted for the largest amount of the total fatty acids, and the synthesis of BCFAs performed an important role in the modification of cell membrane fluidity when exposed to external stimuli (39). The synthesis of BCFAs began with the amination of isoleucine, valine, and leucine by the branched-chain-amino-acid aminotransferase BAT (encoded by ilvE). The short branched-chain acyl coenzyme A (acyl-CoA) derivatives 2-methylbutyryl-CoA, isobutyryl-CoA, and isovaleryl-CoA were then produced through the oxidative decarboxylation  by the branched-chain a-keto acid dehydrogenase BKD (encoded by lpd, bkdA1, bkdA2, and bkdB). Subsequently, these acyl-CoA substances were converted into malonyl-CoA, which was further catalyzed to malonyl-ACP by the malonyl-CoA:ACP transacylase FabD (encoded by fabD) (40,41). b-Ketoacyl-ACP was then produced from the b-ketoacyl-acylcarrier-protein ACP synthase III (encoded by fabH)-induced condensation of malonyl-ACP and acetyl-CoA/2-methylbutyryl-CoA to initiate the elongation cycles (involving fabI, fabG, and fabF) to produce BCFAs or SCFAs (42). When subjected to external stress, the bacterial cells developed resistance by regulating the length and saturation of fatty acid chains and changing the composition of fatty acids to alter the membrane fluidity. To further understand the causes of membrane phenotypic changes, the expression of membrane fatty acid synthesis-associated genes in S. aureus under acid-heat stress was investigated in this work. As shown in Fig. 3A, after exposure to acetic acid, the fabD and fabH genes were upregulated by 2.56-and 3.49-fold, respectively, in S. aureus strain J15. The fabD and fabH genes are responsible for encoding the malonyl-CoA:ACP transacylase FabD and b-ketoacyl-ACP synthase III, which are involved in type 2 fatty acid synthesis pathway for the biosynthesis of BCFAs and SCFAs (42). Genes associated with the biosynthesis of BCFAs were upregulated in strain J15 after exposure to acetic acid. For instance, the gene of bkdB, which was shown to be upregulated by 3.26-fold, and is involved in the production of BKD, a multisubunit enzyme complex that consists of a dehydrogenase (E1a), a decarboxylase (E1b), a dihydrolipoamide acyltransferase (E2), and a dihydrolipoamide dehydrogenase (E3) and participates in the early stages of BCFA biosynthesis (43). Further heat treatment dramatically increased the expression of genes (e.g., fabH) in strain J15, which might be associated with the enhancement in SCFA content and the changes in the ratio of anteiso-BCFAs to iso-BCFAs measured previously.
After citric acid exposure, the genes involved in the biosynthesis of BCFAs were  highly expressed in S. aureus strain J15, which exhibited the citric acid-heat cross-adaptation (Fig. 3B). The bkdA gene was significantly upregulated by a factor of 4.08, while the genes related to BCFA synthesis in the control strain did not appear to be changed. The bkdA and bkdB genes encode the dehydrogenase E1a and the decarboxylase E1b, respectively, which are the polypeptide components of BKD complex and are critical in the synthesis of BCFAs (44). The fabH and fabI genes in the elongation cycle of the fatty acid biosynthesis were upregulated by 2.91-and 2.34-fold, respectively, in S. aureus strain J15 under citric acid stress. The fabI gene encodes an NADPH-dependent trans-2-enoyl-ACP reductase, which reduces 2-enoyl-ACP to fatty acyl-ACP at the expense of NADPH (41). Following heat treatment, as exhibited in Fig. 3B, the genes related to fatty acid biosynthesis were still highly expressed in strain J15. After lactic acid exposure, the fabH and fabI genes were upregulated by 2.90-and 2.32-fold, respectively, in S. aureus strain J15. The ilvE, bkdA, bkdB, and ipd genes, which are related to the biosynthesis of BCFAs, were also upregulated in strain J15, among which bkdA was the most significantly upregulated, 5.89-fold. For the control strain J17, no upregulation was observed in the BCFA biosynthesis-associated genes (Fig. 3C). Acid stress probably activates the fatty acid biosynthesis pathway to adjust the SCFA content or the ratio of anteiso-BCFAs to iso-BCFAs in the cell membrane in order to defend against the acid stress (45). With further thermal exposure, the fabH and fabI genes were still overexpressed in strain J15, and there was a significant difference (P , 0.05) from those of the control strain J17. The expression of the fabH gene was upregulated under three organic acid-heat cross-adaptation conditions. It caused changes in the

Acid-Heat Cross Adaptation of S. aureus
Microbiology Spectrum fatty acid profiles of S. aureus, reduced cell membrane fluidity, enhanced heat tolerance, and produced an acid-heat cross-adaptation phenomenon. The ipd gene associated with the biosynthesis of BCFAs remained upregulated, while the others were downregulated, as exhibited in Fig. 3C. Analysis of heat stress response gene expression. Heat stress response-associated regulators in bacterial cells include chaperones (DnaK, GroES, and GroEL) under the control of HrcA repressors, the general stress proteins (requiring sigma factors), and the thermoprotease (ClpAP) regulated by the class III stress gene repressor CtsR (35). Pre-exposure to acid stress might induce bacteria to synthesize specific chaperones so as to protect or repair intracellular macromolecules. Heat stress proteins, such as DnaK, DnaJ, GrpE, HrcA, GroEL, GroES, and Clp, have been reported to be crucial acid resistance factors that act as molecular chaperones to promote the repair of nucleic acids and proteins during acid stress, thus maintaining survival through the heat stress (46). The regulation of heat stress response-associated genes in S. aureus under acid-heat stress was explored in this work.
As shown in Fig. 4A, the heat stress response-associated genes were upregulated in acetic acid-stressed S. aureus strain J15, except the HrcA and CstR repressor encoding genes (hrcA and cstR), which were downregulated. The clpC gene was the most significantly upregulated, 6.93-fold, in strain J15, after acetic acid stress. DnaK-GroESL operonassociated genes (dnaK, groES, and groEL) were upregulated in strain J15, which was attributed to the downregulation of HrcA and CstR repressors. In S. aureus, both the dnaK and groESL operons were dually regulated by CtsR and HrcA (47), which could explain the upregulation of dnaK, groES, and groEL and the downregulation of hrcA and cstR in this work. With further thermal exposure, the heat stress responserelated genes remained upregulated in strain J15 (Fig. 4A). The groES gene was upregulated by 7.48-fold upon exposure to acetic acid-heat treatment. GroES, an oligomeric protein, contains seven identical subunits, and it is the cochaperonin of GroEL. The GroEL-GroES complex participates in the folding and conformation maintenance of cellular proteins, especially in response to external stress (e.g., high temperature, pH change) (47,48). The overlapping induction of stress response genes caused by organic acid and heat contributes to the acid-heat cross-adaptation phenomenon in S. aureus. After exposure to citric acid, most of the heat stress response-related genes were upregulated in S. aureus strain J15 (Fig. 4B). The groES, groEL, and dnaK genes were upregulated by 2.69-, 2.65-, and 3.65-fold, respectively, values which were much higher than those for the control strain J9. With the subsequent heat treatment, the sigB and clpP genes were upregulated by 7.66-and 2.90-fold, respectively. The s B factor, encoded by sigB, is an important contributor involved in the regulation of a series of genes in response to stimuli (e.g., heat and osmosis) (49,50). The clpP gene encodes the ATPdependent Clp protease proteolytic subunit, which is involved in the degradation of misfolded proteins induced by an environmental stressor (e.g., high temperature) (51). The expression of clpP has been determined to be negatively controlled by CtsR (52). As seen in Fig. 4B, the CtsR-encoding gene ctsR was highly repressed in acid-heatcross-adapted strain J15 cells, which probably contributed to the upregulation of clpP gene and prompted acid-heat cross-adaptation in strain J15.
When strain J15 was exposed to lactic acid, the groEL, groES, and clpC genes were upregulated by 2.73-, 3.74-, and 6.56-fold, respectively (Fig. 4C). Similarly, Rode et al. (53) also observed the overexpression of groEL, groES, and clpC in S. aureus upon treatment with lactic acid (pH 4.5) for 180 min. With further thermal exposure, the groES, sigB, and dnaK genes in strain J15 were upregulated by 3.62-, 2.98-, and 2.77-fold, respectively, which might give rise to the heat stress response. The groES gene was significantly upregulated in S. aureus strain J15 undergoing the three organic acid pretreatments and remained upregulated upon subsequent heat exposure, suggesting that groES may be one of the key contributors to the acidheat cross-adaptation of S. aureus.
Conclusions. In this work, the pretreatment of S. aureus with acetic acid, citric acid, and lactic acid stress induced the development of tolerance to subsequent heat exposure at 60°C. Various S. aureus strains exhibited different acid-heat cross-adaptation areas. Phenotypic traits, such as cell membrane structure and function, enzyme activity, and the fatty acid profile of the cell membrane, were further compared between the acid-heat-cross-adapted strain and acid-heat-susceptible strains (Fig. 5). The acid-induced cross-adaptation to heat might be attributed to the repair of the cell membrane to maintain the membrane integrity and to avoid the leakage of intracellular macromolecules, the equilibrium of the internal and external charge differences to achieve a normal internal pH, and the reduction of cell membrane fluidity through reducing the ratio of anteiso-BCFAs/iso-BCFAs. Gene expression analysis demonstrated that the gene fabH, involved in the biosynthesis of fatty acids, as well as groES and groEL, related to the heat shock response, contributed to the development of the acid-heat cross-adaptation phenomenon in S. aureus strains. The results of this work provide a theoretical basis for optimizing food processing and preventing incomplete sterilization due to the induction of bacterial stress resistance.

MATERIALS AND METHODS
Bacterial strains and culture conditions. S. aureus strain ATCC 25923, designated SS, and S. aureus isolates from food, designated J1 to J19, were used in this study. The sources of the S. aureus isolates are shown in Table 6. All the S. aureus strains were mixed with 50% glycerol solution at a ratio of 1:1 (vol/vol) Acid-Heat Cross Adaptation of S. aureus Microbiology Spectrum and stored at 280°C. S. aureus was then recovered on Baird-Parker medium supplemented with 5% egg yolk tellurite emulsion and incubated at 37°C for 24 h. A single colony was transferred to nutrient broth. After incubation at 37°C and 180 rpm for 18 h, S. aureus cells were harvested through centrifugation at 2,320 Â g for 10 min at 4°C and washed three times by resuspension in sterile phosphate-buffered saline (PBS). The initial concentrations of each S. aureus strain were approximately 10 9 CFU/mL. Acid and thermal treatments. Glacial acetic acid (99.5%), anhydrous citric acid powder, and L-lactic acid (86%) were added to sterile water, and the mixture was adjusted to a pH of 4 with a pH meter (Mettler, Toledo, OH, USA). All the prepared acid solutions were filtered with 0.22-mm syringe filters for microbial removal and stored at 4°C before use. S. aureus was treated with each acid solution, and PBS solution (pH 7.0)-treated S. aureus were used as blank controls. The cells were collected by centrifugation (2,320 Â g, 25°C, 10 min). S. aureus cells were washed twice by resuspending in sterile PBS and were resuspended in 10 mL of each acid treatment solution. The acidadapted S. aureus cells were incubated at 60°C for 6 min and then placed in ice to end the thermal stress. Microbiological analysis. After acid treatment, the sample was washed twice by centrifugation at 2,320 Â g for 10 min at 4°C and resuspended in sterile PBS. The bacterial solution was diluted to an appropriate concentration and plated on tryptone soy agar (TSA) (Hope Bio-Technology Co., Ltd., Qingdao, China), followed by incubation at 37°C for 24 h.
Ce calculation of the cross-adaptation area. The inactivation data of S. aureus after exposure to various treatments were fitted to a log-logistic model (54) with the use of Origin 9.1 software (OriginLab, Massachusetts, USA): where A represents the maximum inactivation level that can be achieved by acids under certain conditions; M is the maximum inactivation rate, log(CFU/mL)/log h; and N is the logarithm of the time at which the maximum inactivation rate reachs. The cross-adaptation area is defined as the area between the stress treatment and control curves, which equals the region under the treatment curve minus the area under the control curve (Fig. 6).
Analysis of cell membrane characteristics. (i) Cell membrane integrity estimation. After centrifugation at 5,568 Â g for 2 min, the supernatant was obtained for absorbance measurement at 260 nm and 280 nm with a Nanodrop spectrophotometer (Thermo Fisher, USA), which indicated the nucleic acid and protein leakage amounts for S. aureus, respectively.
(ii) Cell membrane potential analysis. Based on the method by Zhang et al. (55) with a minor modification, the untreated and acid-treated groups (0.5 mL) were centrifuged at 2,320 Â g for 10 min, washed twice with PBS (pH 7.0), and then suspended in 1 mL rhodamine 123 solution (2 mg/mL in PBS) for a 30-min incubation at room temperature. The stained samples were then centrifuged and washed with PBS to remove the excess rhodamine 123. Fluorescence spectra of each sample in the range of 500 to 600 nm were measured with a fluorospectrophotometer (Cary Eclipse, Varian, USA) at an excitation wavelength of 480 nm, and the peak value of fluorescence intensity at an emission wavelength of 530 nm was used to analyze the membrane potential changes (MPC) of bacteria, calculated as F/F 0 Â 100, where F is the fluorescence intensity of the treated sample and F 0 is the fluorescence intensity of the untreated strain.
(iii) Cell membrane fluidity analysis. Cell membrane fluidity was measured using the 1,6-diphenyl-1,3,5-hexatriene (DPH) fluorescence probe-based method described by Wang et al. (56) with minor mod- ifications. Bacterial suspensions (approximately 10 8 CFU/mL) from the treated and untreated groups were centrifuged at 2,320 Â g for 10 min and resuspended in PBS to adjust the optical density at 600 nm (OD 600 ) to 0.5. Then, the DPH probe (2.0 mM) was added, and the mixture was incubated at 37°C in the dark for 1 h. After centrifugation, the supernatant was removed, and the pellets were resuspended in 4 mL of PBS buffer solution. The samples were analyzed by a SpectraMax M5 multifunctional fluorescent plate reader at an excitation wavelength of 360 nm and an emission wavelength of 430 nm (slit width, 5.0/5.0 nm). Fluorescence polarization (P) is calculated as (I VV 2 GI VH )/ (I VV 1 GI VH ), where G is the grating factor, I VV represents the fluorescence intensity obtained when the polarizer and polarizer optical axis in the same vertical direction, and I VH is the fluorescence intensity obtained when the optical axis of the polarizer and polarizer in vertical and horizontal directions, respectively.
(iv) Cell membrane fatty acid analysis. The bacterial suspensions of the treated and untreated groups were centrifuged at 2,320 Â g for 10 min at 4°C, and the supernatant was discarded. Each sample precipitate was washed twice with PBS buffer. Based on the method of Sasser (57), the saponifying agent (5.13 M sodium chloride in methanol) was added to the bacterial pellets, followed by brief shaking. The mixture was placed into a boiling water bath for 5 min and shaken violently for 5 to 10 s, followed by a 30-min boiling treatment. After cooling to room temperature, 2 mL methylation reagent (3.25 M hydrochloric acid in methanol) was added to the mixture, followed by brief shaking and then incubation at 80°C for 10 min. After cooling, the extract (n-hexane-methyl tert-butyl ether = 1:1 [vol/ vol]) with a volume of 1.25 mL was added and tumbled for 10 min. Subsequently, sodium hydroxide (0.3 M) was added to the remaining organic phase and the upper n-hexane/methyl-tert-butyl ether phase containing the fatty acid methyl esters (FAMEs) was transferred for analysis by Sherlock Microbial Identification System.
Determination of Na + K + -ATPase and F o F 1 -ATPase activities. (i) Na + K + -ATPase activities. The bacterial suspensions of the treated and untreated groups were centrifuged at 2,320 Â g for 10 min, and the precipitate was resuspended in PBS buffer to achieve a concentration of around 10 8 CFU/mL. Then, the bacterial cells were lysed by ultrasound, and the Na 1 K 1 -ATPase activity was determined with an ultrafine Na 1 K 1 -ATPase kit (Nanjing Jiancheng Institute of Biological Engineering, China). One unit of enzyme activity is the amount of Na 1 K 1 -ATPase decomposing ATP to produce 1 mmol inorganic phosphorus per h per 10,000 bacterial cells.
(ii) F o F 1 -ATPase activities. F o F 1 -ATPase activities were determined using the colorimetric method described by Price and Driessen (28) with minor modifications. The bacterial suspensions of the treated and untreated groups were centrifuged at 2,320 Â g for 10 min at 4°C, washed with 50 mM piperazine-N,N9-bis(2-ethanesulfonic acid) (PIPES) buffer, and suspended in 10 mL of PIPES buffer. The bacterial cells were then lysed by ultrasound (300 W) for 25 min. Centrifugation was performed at 4°C, and 2,320 Â g for 10 min, and the supernatant was obtained for further analysis. The composition of reaction system A included 50 mM KCl, 5 mM MgSO 4 , 10 mM carbonyl cyanide m-chlorophenylhydrazone (CCCP), 0.9 mM G-strophanthin, 25 mM HEPES, 1 mM oligomycin, 50 mM KNO 3 , and 4 mM ATP. In the B reaction system, no oligomycin was added, and the other components were the same as those in the A reaction system. The A and B reaction systems were incubated at 37°C for 15 min, and 100 mL of sample was added, followed by a 1-h reaction time. A mixture of 40 mM ammonium molybdate, 1.5 mM malachite green, and 5% (vol/vol) Triton X-100 was added to the A and B reaction systems. After a 1-min reaction, 1.5 mM citric acid was added and incubated for 20 min. A spectrophotometer was used for colorimetry analysis at 660 nm, and the phosphorus content of the solution was calculated according to the phosphorus standard curve. The protein concentrations of the samples were determined with a bicinchoninic acid (BCA) assay kit. F o F 1 -ATPase activity (nmol P i / min/mg protein) was measured by the amount of inorganic phosphorus released through the hydrolysis by ATPase.
RT-qPCR analysis of gene expression. After centrifugation, 1 mL TRIzol reagent was added to the treated and untreated bacterial pellets and oscillated vigorously. Chloroform (0.2 mL) was added and mixed thoroughly, followed by a 5-min incubation at room temperature. After centrifugation at 5,568 Â g and 4°C for 15 min, the supernatant was transferred to a new tube, and 0.5 mL of isopropanol was added and mixed thoroughly. The pellets were collected after a centrifugation at 5,568 Â g and 4°C for 10 min, and 75% alcohol was used to wash the pellets three times. After drying, RNase/DNase-free water was added to dissolve the extracted RNA. The integrity and purity of RNA in the sample were determined by gel electrophoresis and absorbance measurements at 260 nm and 280 nm. The extracted RNA was reverse transcribed to cDNA using a SuperScript III first-strand synthesis SuperMix kit (Thermo Fisher, USA). The appropriate diluted cDNA was used for further qPCR analysis. The qPCR system consisted of 25 mL qPCR mix solution (Platinum SYBR green qPCR SuperMix; Thermo Fisher, USA), 1 mL forward/reverse primers, 1 mL cDNA, and 22 mL RNase/DNase-free water. The PCR procedure was as follows: 50°C, 2 min; 95°C, 2 min; 40 cycles of 95°C, 15 s, 60°C, 30 s; and a melting curve analysis. The primers for each gene used in this study were designed with the NCBI Primer BLAST Online system (https://www.ncbi.nlm.nih.gov/tools/ primer-blast) and are listed in Table 7. All primers were obtained from Sangon Biotech (Shanghai) Co., Ltd.
Statistical analysis. All experiments were conducted with three replicates. The statistical significance of the obtained data was analyzed through a one-way analysis of variance (ANOVA) and Duncan's multiple-range tests with SPSS 21.0 software (SPSS Inc., IBM Corporation, Armonk, NY, USA). A P value of ,0.05 was considered statistically significant.