Methyl gallate and tylosin synergistically reduce the membrane integrity and intracellular survival of Salmonella Typhimurium

Nymphaea tetragona Georgi (Nymphaceae) is traditionally used in Asia for the treatment of diarrhea, dysentery and fever. The plant contains various active compounds, including methyl gallate (MG) which are reported to inhibit bacterial virulence mechanisms. This study aimed to evaluate the alterations on viability, membrane potential and integrity of Salmonella enterica Serovar Typhimurium exposed to MG in combination with Tylosin (Ty), which is relatively inactive against Gram-negative bacteria, but it is commonly used as a feed additive in livestock. Besides, the effects of sub-inhibitory concentrations of the combination (MT) on the interaction between S. Typhimurium and the host cell, as well as on the indirect host responses, were characterized. Flow cytometry, confocal and electron microscopic examinations were undertaken to determine the effects of MT on S. Typhimurium. The impacts of sub-inhibitory concentrations of MT on biofilm formation, as well as on the adhesion, invasion and intracellular survival of S. Typhimurium were assessed. The result demonstrated significant damage to the bacterial membrane, leakage of cell contents and a reduction in the membrane potential when treated with MT. Sub-inhibitory concentrations of MT significantly reduced (P < 0.05) the biofilm-forming, adhesive and invasive abilities of S. Typhimurium. Exposure to MT drastically reduced the bacterial count in macrophages. Up-regulation of interleukin (IL)-6, IL-8 and IL-10 cytokine genes were detected in intestinal epithelial cells pre-treated with MT. This report is the first to describe the effects of MT against S. Typhimurium. The result indicates a synergistic interaction between MG and Ty against S. Typhimurium. Therefore, the combination may be a promising option to combat S. Typhimurium in swine and, indirectly, safeguard the health of the public.


Introduction
Salmonella enterica Serovar Typhimurium is one of the causative agents of food poisoning that is transmitted to humans from food animals, such as pigs and chickens. Pigs  S. Typhimurium and the host cell, as well as on the indirect host responses, were characterized.

Chemicals and reagents
Unless specified, the chemicals and reagents used in the current study were purchased from Sigma (St. Louis, MO, USA). Stock preparations of 30 mg/mL of MG, PG and Ty were prepared in 50% ethanol and diluted in the appropriate medium (sterile distilled water and bacterial or cell culture medium). The proportion of ethanol in the final diluent never exceeded 0.5% (v/v).

Salmonella strains and culture conditions
The study was conducted using S. Typhimurium (LVPP-STI15) from pigs, a strain that has been described and used in our previous study [21], and S. Typhimurium (ATCC 14028). S. Typhimurium was routinely cultured in Luria-Bertani (LB) agar (Difco, Sparks, MD, USA). Prior to assays, it was grown overnight in LB broth at 37˚C.

Cell culture
Human colorectal cancer (Caco-2) and RAW 264.7 macrophage cells were obtained from the Korean Cell Line Bank (Seoul, Korea). Caco-2 cells were maintained in minimum essential medium (MEM; Gibco, Grand Island, NY, USA) supplemented with 1% penicillin-streptomycin, 1% non-essential amino acids and 20% fetal bovine serum (FBS). RAW 264.7 macrophages were grown in Roswell Park Memorial Institute (RPMI) medium, containing 1% penicillin-streptomycin and 10% FBS. Cells were grown at 37˚C in 5% CO 2 , and the medium was changed every other day.

MIC and minimum bactericidal concentration (MBC)
The MICs of Ty, MG and PG against S. Typhimurium strains were determined using the broth microdilution method [22], with an inoculum of approximately 1×10 5 colony-forming units per milliliter (CFU/mL) in Mueller-Hinton broth (MHB) and antibiotic-free cell culture medium. MIC was determined as the lowest concentration of the agent that inhibits visible growth, which appeared as non-turbid as judged by the naked eye after incubation at 37˚C for 24 h. Aliquots from wells not showing visible growth were plated on LB agar medium. The lowest dilution concentrations that killed the bacterial inoculum after a further incubation at 37˚C for 24 h were considered as the MBCs. MICs and MBCs were determined in triplicate in five independent experiments.

Checkerboard assay
The anti-bacterial effects and interactions between Ty and both compounds (MG and PG) were assessed by the checkerboard method. Serial dilutions of MG and PG (range, 1

Scanning electron microscope (SEM) analysis
The effect of MT on bacterial membrane integrity was further examined using an SEM, with a slight modification of the method described by [24]. S. Typhimurium (10 7 CFU/mL) was treated as described in the time-kill assay at 37˚C for 6 h. Bacteria were centrifuged (10,000 g for 15 min) and washed twice with 0.9% saline solution, re-suspended in 2.5% glutaraldehyde and maintained at -4˚C for 12 h. The resultant cells were centrifuged (10,000 g for 8 min) and dehydrated in ethanol (30,50,70,80, 90 and 100%) for 10 min. The samples were further dried and examined under an SEM (SU8200 Hitachi, Tokyo, Japan).

Determination of membrane potential (MP)
The change in the differences in electric potential that exists across the intact bacterial cell membrane (i.e., the MP) following MT treatment was determined using the BacLight bacterial MP kit (Molecular Probes). S. Typhimurium (10 7 CFU/mL) was diluted in phosphate-buffered saline (PBS) and treated similarly to the time-kill assay, for 2 h. Afterwards, 10 μL (3 mM) of 3,3-diethyloxacarbocyanine iodide was added to 1 mL of the bacterial suspension and mixed well. Ten microliters of 500 μM carbonyl cyanide 3-chlorophenylhydrazone was added into the control tube while an equal volume of PBS was added to the non-treated control. Following incubation at room temperature for 15 min, the samples were analyzed using flow cytometry (BD Biosciences, San Jose, CA, USA) at excitation and emission filters of 488 and 515 nm, respectively. The MP was determined based on the ratio of the red-to-green fluorescence intensity.

Biofilm inhibition and dispersion assay
The effect of sub-inhibitory concentrations of MG (2-128 μg/mL), Ty (2-256 μg/mL) and MT ( 1 / 128 × MT-1 / 4 × MT) on S. Typhimurium biofilm was evaluated using a minor modification of a previous method [25]. Overnight cultures of S. Typhimurium grown in tryptic soy broth were normalized to an OD 600 of 0.8. Two hundred microliters of a 1:100 dilution (in tryptic soy broth) of the normalized cultures was dispensed in 96-well plates, treated with different concentrations of MG, Ty and MT, and incubated at 30˚C under static condition. Control plates were not treated with any of the test substances. Following a 24-h incubation, the planktonic bacteria growth was confirmed using a microplate reader (optical density at 600 nm). The plates were then washed with sterile double-distilled water, fixed at 60˚C for 1 h and stained with 0.1% crystal violet (CV) in water. After incubation at room temperature for 15 min, the CV was solubilized using 30% acetic acid in water, and the absorbance was determined at 550 nm against the blank (30% acetic acid in water). The concentrations of MG, Ty and MT that resulted in 50% inhibition of the maximum biofilm formation (EC 50 ) were determined. Furthermore, the impact of MT on biofilm dispersion was measured in cells treated after 24 h of biofilm formation and incubated further for 24 h.

Percentage viability = (OD value of treated cells/OD value of control cells) ×100
where OD is the optical density.

Adhesion and invasion assay
The gentamicin protection assay [27] was conducted to evaluate the impacts of different concentrations of MG, Ty and MT on the adhesion and invasion capacity of S. Typhimurium to Caco-2 cells. Cells (10 5 /mL) were grown in 24-well plates in antibiotic-free medium. Cells were treated with 900 μL of MG (8, 16 and 32 μg/mL), Ty (16, 32 and 64 μg/mL), 1 / 16 × MT (8 μg/mL MG and 16 μg/mL Ty), 1 / 8 × MT (16 μg/mL MG and 32 μg/mL Ty) and 1 / 4 × MT (32 μg/mL MG and 64 μg/mL Ty) for 45 min. Then, S. Typhimurium was recovered by centrifugation at 10,000 g for 15 min, washed twice with 1× PBS and re-suspended in antibiotic-free MEM (10 7 CFU/mL). One hundred microliters of bacterial suspensions (MOI = 10:1) were added to each wells, centrifuged at 1000 g for 2 min and incubated at 37˚C for 45 min. The supernatant was discarded, and the cells were washed with 1× PBS to remove non-adhered bacteria. Cells were disrupted with 1% Triton X-100. The lysates were serially diluted using agar saline and plated on LB agar. The total number of adhering bacteria (CFU/mL) was determined after incubation at 37˚C for 16-18 h. For the invasion assay, treated and infected cells were incubated at 37˚C for 60 min. The supernatant was discarded, and non-adhered bacteria were removed by washing three times with 1× PBS, followed by addition of 300 μL gentamicin (100 μg/mL) in MEM and further incubation for 60 min to kill extracellular bacteria. Cell lysis and determination of the number of invading bacteria were performed similarly to the procedure mentioned above in the adhesion assay.

Intracellular survival assay
Survival of S. Typhimurium in macrophages pre-treated with MG, Ty and MT was determined by slight modifications of the methods described by Lu et al. [28]. Cells (10 5 /mL) grown without antibiotics in 24-well plates were treated and infected, as described above in the invasion assay. After incubation with gentamicin (100 μg/mL) at 37˚C for 60 min, cells were washed with 1× PBS, and 25 μg/mL gentamicin in RPMI was added for the remaining incubation time. At 2, 4 and 8 h post-incubation, cells were lysed, and the total number of surviving bacteria was determined as described above.

Detection of cytokine expression
Infection of intestinal epithelial cells with S. Typhimurium up-regulates the expression of many cytokine genes [29]. The effects of MG, Ty and MT on S. Typhimurium (ATCC 14028)induced cytokine (tumor necrosis factor alpha [TNFα], and interleukin [IL]-1β, IL-6, IL-8 and IL-10) gene expression in Caco-2 cells was evaluated using quantitative real-time reverse transcription PCR (qRT-PCR). For this purpose, cells (10 5 /mL) were treated and infected, as described above in the adhesion assay. After 8-h incubation, total RNA was extracted using TRIzol (Ambion Life Technologies, Carlsbad, CA, USA). The concentration and purity of RNA were determined using Nanophotometer (Implen GmbH, Munich, Germany). An A 260 / A 280 ratio of 1.8 to 2.0 was considered as good quality RNA and a total of 1μg was taken for cDNA synthesis. The cDNA was amplified, and qRT-PCR was performed in a CFX96 Touch real-time PCR detection system (Bio-Rad, Hercules, CA, USA) using IQ SYBR Green Supermix (Bio-Rad Laboratories [Singapore] Pte. Ltd.). Three hundred Nano molars of both the forward and reverse primers were used in 20 μL reaction mixture. The reaction conditions included denaturation at 95˚C for 3 min, 40 cycles of amplification at 94˚C for 10 s, 60˚C for 30 s and melting from 65 to 90˚C. The housekeeping genes β-actin and GADPH were used to normalize gene expression. The fold changes of cytokines gene expression were analysed using the 2 -ΔΔCT method and the statistical difference were determined using the Student's t-test in GraphPad Prism 6 (GraphPad Software, Inc., San Diego, CA, USA). The sequences of all primers used in this experiment are listed in S1 Table.

Data analysis
Data were analyzed using GraphPad Prism 6 (GraphPad Software, Inc., San Diego, CA, USA). One-way and two-way analyses of variance (ANOVA) were conducted to compare the mean values among treatment groups. P < 0.05 was considered statistically significant.

MIC and MBC
The MICs of MG, Ty and PG against the two S. Typhimurium strains were 512, 1024 and 128 μg/mL, respectively. The MBCs of MG and PG were twice the MIC values in both strains; however, the MBC of Ty exceeded 4 mg/mL. Similar MIC values were obtained for both of the tested agents in RPMI and MEM.

Checkerboard assay
MT was effective, as confirmed by the reduced MICs of 128 and 256 μg/mL for MG and Ty, respectively, for both strains. The FICI was calculated to be 0.5, indicating synergism between Ty and MG. Furthermore, an additive effect (FICI = 0.56) was found between Ty and PG, in which the MIC values were reduced to 64 μg/mL. Therefore, MT was chosen for subsequent experiments.

Time-kill assay
To confirm the results of the checkerboard assay, the in vitro time-kill activities of MT were performed for both strains (Fig 1). At 1× MT, a synergistic effect occurred that caused more than 2 log 10 CFU/mL reduction compared with Ty and MG alone. Conversely, the growth curves of S. Typhimurium treated with 64 and 128 μg/mL Ty were almost the same as those treated with 256 μg/mL Ty. Likewise, the differences in the growth curves of S. Typhimurium treated with 32, 64 and 128 μg/mL MG were not significant.

Bacterial membrane integrity
The confocal microscopic images of SYTO9/PI-stained S. Typhimurium exposed to MG, Ty and MT for 4 h are illustrated in Fig 2. S. Typhimurium with a damaged membrane (red fluorescent stain) were observed after 4-h exposure to 1 / 2 × MT (5.8 ± 1.2%) and 1× MT (21.2 ± 4.8%). In contrast, those grown in the presence of MG, Ty and 1/4× MT (green fluorescent stain) indicated that the cell membrane was intact.

Field emission scanning electron microscope (SEM) analysis
For both strains of S. Typhimurium, treatment with 1 / 2 × MT and 1× MT caused a remarkable modification in morphology, such as rough surface (red arrow), membrane blebbing (yellow arrow) and leakage of cellular contents (blue arrow) (Fig 3). Marked membrane disintegration with leakage of cellular contents was evident in the 1× MT-treated S. Typhimurium compared with the non-treated control, as well as those treated with MG and Ty alone.

Resting MP
The MP of S. Typhimurium (ATCC 14028) that had been incubated with different concentrations of MT for 2 h was determined by measuring the ratio of red-to-green fluorescence intensity (Fig 4). The ratio was decreased to 0.20 in S. Typhimurium treated with 1 / 2 × MT and 1× MT. The value was significantly (P < 0.001) lower when compared with fluorescence ratios of 0.80, 0.83 and 1.10 in MG-treated, Ty-treated and non-treated S. Typhimurium, respectively. However, treatment with 1 / 4 × MT did not produce a significant change in the fluorescence ratio (P > 0.05).

Biofilm formation
The study demonstrated an insignificant difference (P > 0.05) in planktonic bacterial growth (mean OD 600 ) in the presence and absence of sub-inhibitory concentrations of MG, Ty and MT. Particularly, MT reduced S. Typhimurium biofilm formation in a concentration-dependent manner. Biofilm formation by S. Typhimurium (ATCC 14028) treated with 1 / 8 × MT and 1 / 4 × MT was decreased by 57.3 and 55.9%, respectively (Fig 5). The EC 50 of MT was calculated for both strains of S. Typhimurium, considering 0 and 100% as the minimal and maximal biofilm inhibition, respectively. Accordingly, 1 / 10 × MT (a combination of 12.8 μg/mL MG and 25.6 μg/mL Ty) could cause half of the maximum inhibition (EC 50 ) of S. Typhimurium (LVPP-STI15) biofilms while the EC 50 value for S. Typhimurium ATCC 14028 biofilms was 1 / 12.5 × MT (a combination of 10.24 μg/mL MG and 20.5 μg/mL Ty). Conversely, the maximum biofilm inhibitions caused by MG and Ty were 6.86 and 6.48%, respectively (S1 Fig). There was no significant difference (P > 0.05) in the biofilm inhibition ability of MT between the two strains. Besides, neither the combination nor each agent dispersed the pre-formed biofilms of both strains.

Cytotoxicity assay
The effects of MG, Ty and MT on the viability of Caco-2 epithelial cells and RAW 264.7 macrophages are shown in Fig 6. Treatment of cells with MG and Ty individually, at initial concentrations of 2 mg/mL, did not produce a significant toxic effect. Likewise, greater than 90% of the cells were viable following 24 h of exposure to MT containing initial concentrations of 2 mg/mL MG and 2 mg/mL Ty.

Effects of MT on intracellular bacterial survival
After incubation for 2 h, treatment did not produce a significant difference in the percentage of surviving bacteria (Fig 8). However, a marked decrease in the percentage of intracellular S. Typhimurium was observed after incubation with various concentrations of MT for 4 and 8 h in comparison to MG and Ty alone. The bacterial load (ATCC 14028 and LVPP-STI15) in cells incubated with 1 / 16 × MT, 1 / 8 × MT and 1 / 4 × MT for 8 h was decreased significantly (P < 0.001) by 79.14 and 88.6%, 79.1 and 81.99%, and 82.7 and 87.75%, respectively, compared with the load in the infected and non-treated cells. Interestingly, MG significantly (P < 0.5) decreased the intracellular S. Typhimurium count from 4 h of incubation, in contrast with Ty, as well as the infected and non-treated cells.

Cytokine expression
S. Typhimurium induced the expression of IL-6, IL-8, IL-1β, TNFα and IL-10 genes in CaCo-2 cells after 8-h infection (Fig 9). A concentration-dependent up-regulation (P < 0.001) in the IL-6 gene expression was detected when infected cells were incubated with 1 / 8 × MT and 1 / 4 × MT compared to those of MG and Ty. The expression of IL-8 gene was significantly enhanced (P < 0.01) in 1

Discussion
The high MIC value of Ty (1024 μg/mL) found in this study is not surprising because the outer membrane of Enterobacteriaceae, such as Salmonella, has low permeability to macrolides, including Ty, which limits the activities of these drugs [30]. The MIC of MG (512 μg/mL) was comparable to the findings of Choi et al. [16] but much higher than the values (3.9-125 μg/ mL) reported by Choi et al. [17] and Choi et al. [18] on Salmonella isolates from chicken and pig. The difference of bacterial strains used between these studies could contribute to their difference in susceptibility to MG.
The FICI and the time-kill assay confirmed in vitro synergy since more than a 2-log-fold reduction in CFU/mL was found when MT was used against S. Typhimurium at concentrations lower than the MICs of the individual components. The SYTO9/PI staining and SEM analysis further supported these findings, in which bacterial death, membrane damage or disintegration and leakage of cellular contents were exhibited following MT treatment. Proteins and small molecules are believed to cause anti-bacterial action by perforating the bacterial membrane and causing leakage of cellular contents [31]. The ability of 1× MT to induce leakage of cellular contents indicates that it acts to cause bacterial membrane destabilization and perforations. This could lead to a further influx of MG and Ty across the damaged membrane to the intracellular target sites. Previous studies have confirmed the membrane-damaging activity of MG when used alone against V. cholera [20] and multidrug-resistant Shigella spp. [19] but at relatively higher concentrations (1× MBC-5× MBC).
Reduction in the ratio of red-to-green fluorescent of S. Typhimurium following SYTO9/PI staining is an indicator of an altered MP [32]. Studies on carvacrol and thymol confirmed that agents that contain free hydroxyl groups could act as a protonophore [32][33][34], enhancing insertion into the bacterial membrane and disrupting the physical, as well as chemical properties, of the membrane. The lipid layer becomes destabilized, resulting in an increase of passive proton flux across the membrane and contributes to alteration in the MP [34,35]. Therefore, the free hydroxyl groups in MG could contribute to the membrane destabilization effects of MT. In addition, MT-induced loss of S. Typhimurium membrane integrity could also lead to leakage of protons and potassium, which results in the change in MP. Studies on various bacterial species exhibited that natural products could alter the MP and ultimately lead to cell death [33][34][35]. Therefore, the anti-bacterial activity of MT could involve disruption of the MP, as well as membrane disintegration.
A number of studies have demonstrated the role played by biofilms in anti-microbial resistance and persistence. S. Typhimurium is known to be able to produce biofilms on diverse surfaces, including the gallbladder, epithelial cells and various host tissue compartments [36,37]. The CV staining demonstrated a concentration-dependent reduction of S. Typhimurium biofilm formation. The activity of MT against S. Typhimurium biofilms was not a function of the direct anti-bacterial activity, as it was confirmed in the time-kill and membrane integrity assays. QS-ability is suggested to play a critical role in biofilm formation. Furthermore, exopolysaccharides, such as cellulose, are one of the most important extracellular polymeric structures for biofilm formation in S. Typhimurium [37,38]. Our previous studies demonstrated Salmonella: infected, but non-treated cells. Results mean ± SEM of three independent experiments. MG = 32 μg/mL; Ty = 64 μg/mL; 1 / 16 × MT = 8 μg/mL MG and 16 μg/mL Ty; 1 / 8 × MT = 16 μg/mL MG and 32 μg/mL Ty; and ¼× MT = 32 μg/mL MG and 64 μg/mL Ty. � P < 0.05 compared to infected, but non-treated cells; τ P < 0.05 compared to Ty; π P < 0.05 compared to MG; and # P < 0.05 compared to both MG and Ty.
https://doi.org/10.1371/journal.pone.0221386.g009 that sub-MIC of MG reduced exopolysaccharide production and QS in P. aeruginosa biofilms [39] and S. Typhimurium [40], respectively. In vitro studies have also indicated that macrolides alter biofilm architecture through inhibition of polysaccharide synthesis in P. aeruginosa [41]. Moreover, the anti-QS [42] and biofilm inhibitory activities of sub-MIC concentrations of macrolides [43] and MG [39] are reported in various bacterial species. Therefore, MG and Ty could act synergistically to inhibit biofilm formation in S. Typhimurium.
The ability to adhere, invade and survive in intestinal epithelial cells and macrophages are critical for the pathogenesis of Salmonella. Thus, agents that could interfere with the adherence and invasion of S. Typhimurium to intestinal epithelial cells play a critical role to avoid the establishment of infection [44]. In addition, the importance of motility for bacterial invasion has already been confirmed [45]. In the current study, MT treatment reduced the adhesion, invasion and intracellular survival of S. Typhimurium. Similar effects were also found following treatment of cells with MG, but to a lesser extent compared with MT. Low levels of macrolides, such as erythromycin, are reported to reduce the adherence of P. aeruginosa to airway epithelial cells [46]. Moreover, our recent study confirmed the inhibitory effects of MG on S. Typhimurium motility and invasion into host cells [40]. Therefore, the synergistic interaction between the two agents could contribute to the significant reduction in the adhesion, invasion and survival of S. Typhimurium to the host cells.
The balance between the host response and the bacterial virulence mechanisms determines the outcome of infection. The S. Typhimurium-induced release of cytokines by intestinal epithelial cells are critical to initiate and orchestrate the inflammatory events that occur after acute bacterial infection [24]. The cellular immune response following infection and treatment was determined by measuring the gene expression of cytokines. Earlier studies confirmed the expression of various cytokines in Salmonella-infected intestinal epithelial cells [24,47]. Treatment of infected cells with MT up-regulated the gene expressions of IL-6, IL-8 and IL-10. Activation of these cytokines is crucial for initiating the host-defense against S. Typhimurium. Inflammatory bowel disease was observed in mice deficient with IL-10 [48]. The up-regulation of IL-6 and IL-8 is critical for the host because of high biological significance. Indeed, IL-6 plays a critical role in regulation of hematopoiesis, inflammation and immunity [49]. It is also correlated with the differentiation of B cells and macrophages. Likewise, IL-8 is a pro-inflammatory chemokine that attracts and activates neutrophils to the intestinal lumen [50].

Conclusions
Our findings suggest that MG enhances the membrane-destabilizing and biofilm inhibitory effects of Ty in S. Typhimurium. Sub-inhibitory concentrations of MT were also effective in reducing the adhesion and invasion of S. Typhimurium to intestinal epithelial cells. The direct anti-bacterial activity and indirect stimulation of the host response could be the mechanism of anti-Salmonella activity of MT. Therefore, co-administration of MG with Ty could be considered to reduce the fecal burden of S. Typhimurium in swine and thereby reduce transmission of infection to humans. Future work will aim at evaluating the in vivo efficacy of MT in target animal. Cell: noninfected and non-treated control. Salmonella: infected, but non-treated cells. Results mean ± SEM of three independent experiments. MG = 32 μg/mL; Ty = 64 μg/mL; 1 / 16 × MT = 8 μg/mL MG and 16 μg/mL Ty; 1 / 8 × MT = 16 μg/mL MG and 32 μg/mL Ty; and ¼× MT = 32 μg/mL MG and 64 μg/mL Ty. � P < 0.05 compared to infected, but non-treated cells; τ P < 0.05 compared to Ty; π P < 0.05 compared to MG; and # P < 0.05 compared to both MG and Ty. (JPG)