Synergism of the Combination of Traditional Antibiotics and Novel Phenolic Compounds against Escherichia coli

Pathogenic Escherichia coli (E. coli)-associated infections are becoming difficult to treat because of the rapid emergence of antibiotic-resistant strains. Novel approaches are required to prevent the progression of resistance and to extend the lifespan of existing antibiotics. This study was designed to improve the effectiveness of traditional antibiotics against E. coli using a combination of the gallic acid (GA), hamamelitannin, epicatechin gallate, epigallocatechin, and epicatechin. The fractional inhibitory concentration index (FICI) of each of the phenolic compound-antibiotic combinations against E. coli was ascertained. Considering the clinical significance and FICI, two combinations (hamamelitannin-erythromycin and GA-ampicillin) were evaluated for their impact on certain virulence factors of E. coli. Finally, the effects of hamamelitannin and GA on Rattus norvegicus (IEC-6) cell viability were investigated. The FICIs of the antibacterial combinations against E. coli were 0.281–1.008. The GA-ampicillin and hamamelitannin-erythromycin combinations more effectively prohibited the growth, biofilm viability, and swim and swarm motilities of E. coli than individual antibiotics. The concentration of hamamelitannin and GA required to reduce viability by 50% (IC50) in IEC-6 cells was 988.54 μM and 564.55 μM, correspondingly. GA-ampicillin and hamamelitannin-erythromycin may be potent combinations and promising candidates for eradicating pathogenic E. coli in humans and animals.


Introduction
Escherichia coli (E. coli) is a common, rod-shaped, gram-negative bacteria belonging to the Enterobacteriaceae family. Many E. coli strains in the intestinal tract of humans and animals are considered as commensal members of the microbiota [1]. However, some strains, including enteropathogenic E. coli, enteroinvasive E. coli, enteroaggregative E. coli, enterotoxigenic E. coli, and enterohemorrhagic E. coli, are highly pathogenic, and may initiate a wide spectrum of diseases ranging from self-limiting to life-threatening intestinal and extra-intestinal illnesses, such as enteritis, pyelonephritis, enterotoxaemia, mastitis, cystitis, meningitis, septicemia, and arthritis [1][2][3]. Antibiotics have long been considered the first line of defense to prevent pathogenic E. coli-associated infections.

Antibacterial Effects of Combinations of Phenolic Compound and Antibiotic
The combination effects of commercially available antibacterial agents and phenolic compounds against E. coli ATCC25922 were determined from the fractional inhibitory concentration index (FICI) values by performing checkerboard microdilution assays. The results of the combined activities of these commercial antibacterial agents and phenolic compounds are shown in Table 2. The GA-thiamphenicol, HAMA-ERY, and HAMA-thiamphenicol combinations represented synergistic antibacterial effects against this strain with their FICI values of 0.281, 0.375, and 0.50, respectively. Nevertheless, 17 combinations were found to exhibit additive antibacterial effects (FICI: 0.502-0.625) against this strain. Other combinations against this bacterial strain demonstrated indifferent antibacterial effects, with no evidence of antagonistic effects from any combination. In view of their clinical and commercial significance, two combinations (AMP-GA, and ERY-HAMA), which were found to have synergistic or additive effects, were chosen for more studies.

Effects of Phenolic Compound-Antibiotic Combinations on Concentration-and Time-Dependent Bacterial Inhibition Rate
The effects of the HAMA-ERY and GA-AMP combinations on the inhibition rates of bacteria over designated time intervals are shown in Figure 1. Compared with the untreated control, an almost 6-fold reduction in the cell density of E. coli ATCC 25922 was found after treatment for 24 h with hamamelitannin (1× MIC). In comparison with the drug-free control, >6-fold reduced density of E. coli cells was obtained after treatment for 24 h with ERY (1× MIC). The treatment of this bacterium with the combination of HAMA ( 1 2 × MIC) + ERY ( 1 2 × MIC) for 24 h exhibited an approximate 5-fold reduced growth compared with the drug-free control. Furthermore, nearly 4-fold growth of the bacterium was inhibited by the collective effects of both antibacterial agents ( 1 4 × MIC of both), which confirmed the promising antibacterial effect of this drug combination. Similarly, E. coli ATCC 25922 cells treated with GA (1× MIC) for 24 h exhibited an approximate 5-fold reduction in cell density compared with the untreated control. The treatment of E. coli cells with AMP (1× MIC) for 24 h showed an approximate 5-fold reduction in growth in comparison with the untreated control. Compared with the untreated control, the growth of this bacterium was inhibited by approximately 5-fold after treatment for 24 h with the GA ( 1 2 × MIC) + AMP ( 1 2 × MIC) combination.

Effects of Phenolic Compound-Antibiotic Combinations on Bacterial Cell Morphology
To evaluate the effect of the HAMA-ERY and GA-AMP combinations on the cellular architecture, the morphology and ultrastructure of Escherichia coli ATCC 25922 cells treated with these antibacterial combinations were examined. Figure 2 presented the representative scanning electronic microscope (SEM) images of E. coli cells. As shown in the SEM images, the untreated (Figure 2A,F), HAMA (1× MIC)-treated ( Figure 2C) and GA (1× MIC)-treated ( Figure 2H) E. coli cells were found to be separated with perfect symmetry, and have a rod-like shape. Binary fissions of the bacterial cells were also found in these SEM images, which are indicated with yellow-colored arrows (

Effects of Phenolic Compound-Antibiotic Combinations on the Inhibition and Viability of Biofilm
The formation of E. coli ATCC 25922 biofilms under the treatment of HAMA-ERY combination and GA-AMP combination were investigated. Figure    The effects of GA-AMP combination and HAMA-ERY combinations on the viability of E. coli biofilm cells were investigated by scanning the biofilm cells with confocal laser scanning microscope (CLSM) after staining the cells with BacLight live/dead stain. The reductions in viability of E. coli biofilm cells caused by the application of the HAMA-ERY and GA-AMP combinations were shown by the result attained with the CLSM. The images of E. coli biofilms obtained by CLSM are shown in Figure 4; the percentage biomass of E. coli biofilms that developed on the glass surfaces and when incubated in the presence of HAMA-ERY and GA-AMP combinations are presented in Figure 5. The CLSM-generated images of E. coli biofilm grown for 48 h and treated with HAMA-ERY and GA-AMP combinations for 24 h showed fewer ratios of live cells (green fluorescent cells) compared with the proportion in untreated controls. These images were analyzed by IMARIS software for determining the dead biofilm (fluorescent red cells) biomass and live biofilm (fluorescent green cells) biomass. The sum of the live biofilm biomass and dead biofilm biomass was considered as total biofilm biomass in a sample. It is clearly revealed from this result ( Figure 5) that all biofilm cells in the untreated control sample were live, whereas a significant portion of biofilm cells were killed with the treatment of HAMA-ERY and GA-AMP combinations. The percentages of the biomasses of dead biofilm after treatment with the HAMA-ERY (58.27%) and GA-AMP (68.46%) combinations were noticeably higher than the percentages of the dead biofilm biomasses after treatment with the individual antibacterial agents (39.74-45.07%).  In each test group, biomasses of "Total biofilm" were thought to be 100%, and with this consideration the percent (%) biomasses of dead (Propidium Iodide-stained) biofilm and live (SYTO9-stained) biofilm were calculated. Data were analyzed by one-way analysis of variance with F-test. Data represented as mean ± standard deviation of three (n = 3) independent experiments. * Represents significance difference (P < 0.05) among the effects of individual drugs and combination drugs within each concentration group. ** Represents significance difference (P < 0.05) compared with the untreated control group. AMP, ampicillin; ERY, erythromycin; GA, gallic acid; HAMA, hamamelitannin; MIC, minimum inhibitory concentration.

Effects of Phenolic Compound-Antibiotic Combinations on Bacterial Cell Motility
The impacts of HAMA-ERY combination and GA-AMP combination on the swarming and swimming motilities of E. coli were assessed. Representative images of swarm cells and swim cells, which were either treated or untreated with combination antibacterial, are shown in Figure 6. The diameters of the swarm and swim zones are shown in Table 3. The swarm and swim motilities of E. coli were substantially restrained by the HAMA-ERY and GA-AMP combinations. Furthermore, the antibacterial combinations at sub-MIC concentration inhibited swarm and swim motilities of E. coli more intensely than those antibacterials individually could with their 1×MIC concentrations.  Data were analyzed by one-way analysis of variance with F-test. Data represented as mean ± standard deviation of three (n = 3) independent experiments. *, ** and *** represent significance differences among the effects of individual drugs and combination drug with the P-value of (P < 0.05), (P < 0.001), respectively. # Represents significance difference (P < 0.05) compared with the untreated control group. AMP, ampicillin; ERY, erythromycin; GA, gallic acid; HAMA, hamamelitannin; MIC: minimum inhibitory concentration; SD, standard deviation.

Effects of Phenolic Compound-Antibiotic Combinations on IEC-6 Cell Viability
The viabilities of small intestine cells (IEC-6) of Rattus norvegicus in presence of HAMA-ERY and GA-AMP combinations were studied. The rates of the viability of IEC-6 cells and the 50% inhibition concentration (IC 50 ) of the tested antibacterial agents are shown in Figure 7 and Table S2. Individual treatment with HAMA (≤62.5 µg/mL) and GA (≤62.5 µg/mL) could not considerably affect the viabilities of the cells. Similarly, individual application of ERY and AMP up to concentrations of 62.5 and 250 µg/mL could not distress the viabilities of the cells. The GA-AMP combination (250 and 500 µg/mL) resulted in increased cell viability compared with the GA alone ( Figure 7B). In contrast, the viabilities of the cells in presence of GA-AMP combination (250 and 500 µg/mL) were decreased in comparison with the viabilities found from the AMP alone. The viability rates of the cells after individual and combination treatment of HAMA and ERY were reduced compared to the untreated control. However, the viability of the cells after treatment with ERY or HAMA alone or with the combination was not significantly different. The IC 50

Discussion
Several recent reports indicate the formation of resistance by many bacterial strains to the diverse antibacterial drugs that have been widely utilized for treating the bacteria-associated infectious diseases in animal and humans [26]. Hence, there is an urgent need to develop alternative antibacterial agents able to combat resistant pathogen-associated infectious diseases [27,28]. The MIC values obtained in this study manifested the antibacterial effects of the tested phenolic-compounds against all these E. coli strains that were found to be resistant to between 2 and 10 of 10 traditional antibiotics ( Table 1).
The potential effects of the tested compounds were more obviously shown by their combination effects with traditional antibiotics, where they exhibited synergistic effects with ERY and thiamphenicol; either additive or indifferent interactions with the rest of the tested antibiotics against E. coli. It was also found that in the concentration-and time-dependent bacterial inhibition tests, the traditional antibiotics in combination with the phenolic compounds suppressed the growth of E. coli more potently than those antibiotics could inhibit individually. Moreover, the traditional antibiotics in combination with phenolic compounds demonstrated stronger inhibition of swarm motility, swim motility, and biofilm formation in E. coli.
The MIC of five phenolic compounds (GA, HAMA, epigallocatechin, epicatechin gallate, and epicatechin) was evaluated against five clinical strains and a QC strain of E. coli. GA has been reported to be effective in the suppression of the growth of many bacteria, including P. aeruginosa, Salmonella typhi, E. coli, and methicillin-sensitive and methicillin-resistant Staphylococcus aureus [29].
The MIC values obtained in this study (Table 1) showed that GA among the tested phenolic compounds enhanced the antibacterial activity the most, followed by epigallocatechin, HAMA, epicatechin gallate, and then epicatechin. Not only GA, but also epigallocatechin, exhibited very satisfying MIC values against E. coli. The MIC value of GA against E. coli in this study (1024 µg/mL) was lower than the MIC values (2500 µg/mL) reported previously [29]. The mean MIC value of plant-derived epigallocatechin against E. coli was reported as 733 ± 121 µg/mL, which was slightly lower and comparable with the results obtained in this study [30]. The MICs of epigallocatechin in resisting E. coli were lower in this study relative to the MICs of plant-extracted epigallocatechin in the previously published report. The variation in MICs of this compound in resisting E. coli may be due to the purity of the utilized epigallocatechin. Similarly, the MICs of epicatechin in resisting E. coli were 2500 µg/mL, which showed the similarity of the results obtained in our study with previously published results [29].
The treatment of bacterial infection with the combination of antimicrobial agents is suggested for preventing the development of antibiotic resistance and for improving the efficacies of antibacterial compounds as it is generally considered that the combination of antimicrobials is more efficacious when they possess synergistic effects; thus, it is a preferred technique [31]. As an example, the GA-thiamphenicol, HAMA-ERY, and HAMA-thiamphenicol combinations represented synergistic antibacterial effects in resisting E. coli with their FICI values of 0.281, 0.375, and 0.50, respectively. Moreover, additive antibacterial effects (FICI: 0.502-0.625) against E. coli were found from 17 combinations of antibacterial agents. Other antibacterial combinations in our study showed indifferent effects against E. coli. Outstanding in vitro antibacterial effects, together with the synergistic interactions with traditional antibiotics, underscore the potential appropriateness of these opportunistic molecules for treating the infections that are associated of E. coli. The combinations of HAMA-ERY and GA-AMP, which showed synergistic/additive effects in resisting E. coli, were carefully chosen for additional examinations based on their commercial and clinical significance.
The time-dependent killing assay is used to decide whether the effect of an antibacterial is bactericidal or bacteriostatic and to determine the pharmacodynamic features of new antimicrobial agents [32]. Bacteriostatic effects against the tested bacteria were observed in this study from GA and HAMA alone, and from the HAMA-ERY and GA-AMP combinations, as ≥99.9% reduction of the inoculum density was not found after 24 h of treatment in comparison with the control growth. Detectable growth of E. coli was not found at 12 h after treatment with the HAMA (1× MIC) and ERY (1× MIC) combination; beyond 12 h, the log phase had started and consequently reached its initial density within 24 h of treatment. Compared with the control, the density of bacterial cells was reduced by several-fold after treatment with GA or HAMA alone for 24 h, and the GA-AMP and HAMA-ERY combinations, which showed that the tested phenolic compounds, individually and together with ERY and AMP, exerted bacteriostatic effects.
Evaluation of the changes in the morphology, integrity, surface characteristics, and permeability of bacterial cell membrane yield important information regarding the mode of action of an antibacterial agent [33,34]. The morphological and physiological modifications in E. coli cells were determined by scanning the cells with SEM after treating the cells with HAMA or GA alone, and with the combinations of HAMA-ERY and GA-AMP. Both the HAMA-ERY and GA-AMP combinations resulted in direct effects on the morphology of the tested pathogen. In comparison with the untreated control cells, the treated E. coli cells had clear morphological variations. Bacterial cells treated with HAMA-ERY and GA-AMP combinations were revealed to be rope-like, long, and without binary fission, which was absolutely unlike the control cells. The results of this assay demonstrated that the antibacterial combinations had a vital impact on the cell division of E. coli. Similar effects were also found on E. coli cell morphology after the application of silver and gold nanoparticles [35]. The alteration in septum formation and the scarcity of the detachment of daughter cells at the time of cell division are two possible causes of this cell elongation effect [35].
It is reported that GA can inhibit the biofilm of E. coli and that HAMA has anti-biofilm effects against certain bacteria [36,37]. The effects of the HAMA-ERY and GA-AMP combinations on the formation of E. coli biofilms were determined in this study. In most cases, the antibacterial combinations more potently inhibited both the planktonic and biofilm cells of E. coli than the individual antibacterial agent could. The dead and live biofilm populations in presence of the combinations of HAMA-ERY and GA-AMP were ascertained by scanning the stained biofilm in CLSM. The enormous effects of the combination of GA-AMP on the E. coli biofilm cells was probably because of the smaller molecular weight of GA (170.12 g/mol), that can straightforwardly penetrate into the bio-films. Consequently, the antibacterial combinations appear to devastate the matrix of biofilm, which causes cell dissociation, and therefore the biofilm cells are further uncovered to the drugs and more vulnerable.
A pathogenic property of bacteria that is related to the dispersion and migration of bacterial cells is motility. Bacterial cells escape from the host immune response with the help of this pathogenic property [38]. The swimming motility is known to be driven by flagella. The flagella also functions in swarming motility, as well as in the formation of biofilm [30]. According to recently published reports, different types of antibiotics exhibit greater resistance to swarming cells than to biofilm cells [39,40]. The capability of the HAMA-ERY and GA-AMP combinations to inhibit the swarm and swim motilities of E. coli was evaluated in this study. The swarm motility and swim motility of E. coli were significantly inhibited by treatment with the HAMA-ERY and GA-AMP combinations (Figure 6), suggesting that these antibacterial combinations may have some impact on flagella-associated processes; specifically, flagella biosynthesis, chemotaxis, and rotation, which could lead to the decrease in swarming and swimming motilities.
In the study of the pharmacological effects of any drug, it is essential and desirable to determine the toxicity/safety profiles of that agent. In this study, the combinations of HAMA-ERY and GA-AMP were shown to have inhibitory effects against enteric pathogenic bacteria. If the ingested chemicals/toxins pass through the intestinal wall, then all tissues and organs are exposed to those chemicals/toxins in general. However, the highest concentration of ingested toxins/chemicals is first exposed and experienced by the gut among all organs. The functions of intestine in both humans and animals are affected by many of these toxins/chemicals [41]. Therefore, toxicological investigation focused on the intestine is of specific concern to this research. The impacts of HAMA and GA in IEC-6 cell viability were studied to determine any possible adverse effects of these compounds on intestinal cells. In IEC-6 cells, the IC 50 values of HAMA and GA were 988.54 and 564.55 µM, correspondingly. The combination of AMP and GA at their highest tested concentrations (125-500 µg/mL) resulted in improved viability of IEC-6 cells compared with GA treatment alone. The combination of AMP with GA reduced the IC 50 value of GA in IEC-6 cells. Likewise, the combination of HAMA (65-1032 µM) with ERY increased the IC 50 value of ERY in IEC-6 cell. The impacts of HAMA and GA on the viability of various cell lines have been determined previously, and these compounds exerted no adverse effects or cytoprotective effects on cell viability [42,43]. Therefore, the effects of ERY-HAMA combination and AMP-GA combination on IEC-6 cells viability found in this study were reliable and logical.
Considering all the results of this investigation, we can conclude that HAMA-ERY and GA-AMP combinations are potent, promising, and novel agents for the eradication of pathogenic E. coli. The HAMA-ERY and GA-AMP combinations could more efficiently inhibit E. coli biofilm formation than ERY or AMP alone, which is crucial for the improvement of the efficacies of traditional antibiotics and/or the development of new antibacterials to minimize the pathogenic effects of E. coli as well as its impact on livestock and public health. Further study is suggested for confirming their safety and efficacy in in vivo systems to enhance their effective and safe utilization as medicines.

Reagents, Chemicals, and Bacterial Strains
All reagents, chemicals, and microbiological culture media used in this study were from Sigma-Aldrich (St. Louis, MO, United States) unless otherwise mentioned. This study used a quality control strain ATCC 25922, and clinical strains V10-13-A02-002-008, V17-13-S02-002-009, V03-13-A01-002-003, V03-13-A03-002-009, and V16-13-S01-002-002 of E. coli, which were collected from different farms throughout the Republic of Korea. Clinical isolated strains V10-13-A02-002-008, V17-13-S02-002-009, V03-13-A01-002-003, V03-13-A03-002-009, and V16-13-S01-002-002 were from pig feces, chicken feces, cattle feces, chicken feces, and cattle carcass, respectively. Feces of animals were immediately collected after excretion in the farm. Similarly, carcass samples were instantly collected from freshly slaughtered cattle. Both kinds of specimens were immediately kept in iceboxes and then 10 g of each of these samples were transferred to sterile tubes. Nutrient broth (Kisan Bio Co., Ltd., Seoul, South Korea) with a volume of 45 mL were added to these tubes, and homogenized. One milliliter of these homogenates was transferred to a tube in 9 mL of buffered peptone water (Neogen, Lansing, MI, United States) and aerobically cultured for overnight at 37 • C. Then, 1 loop of cultures from each of these samples was streaked on eosin methylene blue agar (Kisan Bio Co., Ltd., Seoul, South Korea) and incubated for 24 h at 37 • C in aerobic condition. After incubating, the suspected colonies were biochemically identified as E. coli by using API 20E (bioMerieux Vitek Inc., Hazelwood, MO, United States). The isolated clinical strains were cultured for 20 h in Mueller Hinton broth (MHB; Becton Dickinson and Company, Becton Drive, NJ, United States) in a rotary incubator at 200 rpm and 37 • C before the experiments were performed.

Minimum Inhibition Concentrations of Phenolic Compounds and Conventional Antibiotics
The broth microdilution method of the CLSI was utilized to determine the MICs of the tested antibacterial agents [44]. Based on the CLSI guidelines, 5 × 10 5 colony forming unit (CFU)/mL inoculum density in MHB was maintained for this assay. The solutions of the different antibacterial agents were diluted serially in 100 µL volumes in 96-well plates. The starting concentrations of HAMA, GA, epigallocatechin, epicatechin gallate, epicatechin, thiamphenicol, penicillin G, norfloxacin, marbofloxacin, florfenicol, ERY, ceftiofur, cefotaxime, AMP, and amoxicillin after the addition of diluted bacterial suspension in 96-well plates were 2048, 1024, 1024, 1024, 1024, 2048, 1024, 32, 32, 256, 2048, 256, 64, 1024, and 512 µg/mL, correspondingly, against all the tested strains. The suspensions were adjusted to 0.5 McFarland units by dilution of the cultures of different bacterial strains, and these cell density-adjusted diluted cultures were again diluted 100 times. Each of the antibiotic solutions (100 µL) was individually transferred to the wells of 96-well plate. The diluted suspensions of each of these E. coli strains with a volume of 100 µL were dispensed to the designated wells of these 96-well plates. The bacterial cells in different drug solutions were incubated for 18 h at 37 • C and the turbidity in all wells was analyzed. The lowest concentration of each antibacterial agent that fully prohibited any escalation in turbidity was determined as the MIC.

Fractional Inhibition Concentration (FIC) Index of Phenolic Compounds and Conventional Antibiotics
The combination effects of phenolic compounds and traditional antibiotics on E. coli ATCC 25922 were determined by the previously reported checkerboard microdilution method, with slight modifications [45]. For the comparison between phenolic compounds and traditional antibiotics, one was applied horizontally and the other one applied vertically in serial dilutions to obtain mixtures of these two antibacterial agents at various ratios in each well of a 96-well plate. The starting concentrations of these phenolic compounds and traditional antibiotics were the same as those used in the MIC determination described in the previous subsection. Similar dilutions of these antibacterial agents individually and the drug-free control wells (medium only) were also added in each test plate. The cultures of E. coli strains in their initial log phase were diluted, and these diluted suspensions of each of the E. coli strains with a volume of 100 µL were dispensed to each wells of the 96-well plates, where the final density of bacterial cells after dispensing to each well was 5 × 10 5 CFU/mL. These bacterial cells in different drug solutions in 96-well plates were incubated at 37 • C for 18 h. The fractional inhibitory concentration (FIC) and the FICI were calculated from the following formulas [12].

Effects of Phenolic Compound-Antibiotic Combinations on the Inhibition Rate of E. coli
Following the method of a previously reported article, the time-and concentration-dependent inhibitory effects of the HAMA-ERY and GA-AMP combinations on E. coli ATCC 25922 were investigated [12]. Cultures of E. coli in their initial log phase were diluted and re-suspended in the drug-containing broth in a Falcon tube, to a final bacterial cell density of 5 × 10 5 CFU/mL. Drug-free MHB broth (10 mL) in 15 mL Falcon tubes with the same bacterial density as the drug-containing broth was employed as the control. Bacterial cells in drug-supplemented medium were incubated in a shaking incubator at 200 rpm and 37 • C. In accordance with a reported procedure, the cells were collected, washed, and dried [47]. Scanning electron microscopes (models EDX-350 and S-4300; Hitachi, Japan) were used to determine the morphology of the ultrastructure of control-and drug-treated E. coli cells.

Effects of Phenolic Compound-Antibiotic Combinations on the Growth and Viability of E. coli Biofilm
A previously described spectrophotometric method was slightly modified and used to determine the effects of HAMA-ERY and GA-AMP combinations on the formation of biofilm by E. coli [48,49]. In brief, antibacterial agents were added to wells of a 96-well microplate where 100 µL of trypticase soy broth (TSB; Becton Dickinson and Company, Becton Drive, NJ, United States) were added earlier to each well of a 96-well microplate. Three separate wells were used for each concentration of antibacterial agent. The same antibacterial concentrations as stated in Section 4.4 were applied in this assay. The culture of E. coli in the initial log phase was diluted in TSB and re-suspended in the drug-containing broth in a 96-well microplate to a final bacterial cell density of 5 × 10 5 CFU/mL after inoculation. Immediately after bacterial inoculation, the optical density (OD) value of the bacterial suspensions in the wells of a 96-well microplate were measured by using VersaMax microplate reader (Molecular Devices, San Jose, CA, United States) at a wavelength of 600 nm. The OD values of the bacterial suspensions in the wells of a 96-well microplate were measured again after incubation for 24 h at 37 • C to determine the planktonic cell growth. Subsequently, the supernatant of each well was carefully discarded without disturbing the biofilms that were attached to the inner surfaces of the well. The wells were washed three times in sterile phosphate buffered saline (PBS, pH 7.2) to remove the adherent drug and media components. Then, 200 µL of methanol (99%, v/v) was added to each well for 20 min to fix the biofilms. Crystal violet (100 µL; 0.2%, w/v) solution was added to the wells and kept for 15 min at room temperature to stain the biofilms. The wells were washed four times with PBS to remove the unbound or excess crystal violet. Then, 100 µL of ethanol (95%, v/v) was added to each of the wells to extract the crystal violet stain on the biofilm. The OD values of extracted crystal violet stain were measured and represented the comparative measurements of biofilm formation in comparison with the control.
The effects of the HAMA-ERY and GA-AMP combinations on the viability of the E. coli biofilms were evaluated in accordance with a previously reported biofilm viability assay [11,50]. Briefly, 2 mL of sterile TSB broth was dispensed to each of the chambers of a Nunc™ Lab-Tek™ II Chambered coverglass (ThermoFisher Scientific, Waltham, MA, USA). The culture of E. coli in its initial log phase was diluted in TSB and then inoculated into the broth of different chambers to a final bacterial cell density of 5 × 10 5 CFU/mL after inoculation. The E. coli cells in TSB in different chambers were incubated for 48 h at 37 • C to allow biofilm formation. The broth medium used in the formation of biofilm was replaced by sterile and fresh medium every 24 h. After incubation for 48 h, the planktonic cells and the media were discarded, and 1× PBS was used to wash the chamber cover glasses. Then The BacLight live/dead staining kit contains propidium iodide and SYTO9 stains. We have described the principle of the BacLight live/dead staining assay in our previously published article [18]. In brief, the SYTO9 (green fluorescence)-stained biofilm cells are considered live cells owing to their membrane integrity. Conversely, the propidium iodide (red fluorescence)-stained cells are recognized as dead cells owing to their damaged membranes. A ZEISS LSM 700 CLSM (Carl Zeiss, Jena, Germany) was used for imaging. To image SYTO9-stained cells, an emission filter of 495-550 nm and a 488 nm laser were used, and for the scanning of propidium iodide-stained cells an emission filter of 560-600 nm and a 561 nm laser were used. The image acquisition and subsequent image handling were performed using ZEN 5.5 software (Carl Zeiss, Jena, Germany). Images were randomly taken from different parts of cover glass. The IMARIS 9.1 software package (Bitplane, Zurich, Switzerland) was used to analyze the images for quantification of the biomasses of live and dead biofilms. The sum of the live biofilm biomass and dead biofilm biomass was considered to represent the biomass of total biofilm.

Effects of Phenolic Compound-Antibiotic Combinations on Swim and Swarm Motilities of E. coli
The effects of phenolic compound-antibiotic combinations on the swim and swarm motilities of E. coli ATCC 25922 were investigated by slightly modifying a previously reported method [38,39]. Luria-Bertani (Becton Dickinson and Company, Becton Drive, NJ, United States) 0.8%, with 0.6% agar (Becton Dickinson and Company, Becton Drive, NJ, USA) and 0.5% glucose (Scharlab, Barcelona, Spain) was the composition of the medium used for the swarming assay of E. coli. The medium used for the evaluation of E. coli swim activity was composed of 1% Tryptone broth, 0.3% agar, and 0.5% NaCl were added in molten swimming/swarming medium in petri dishes. Negative control plates used for the swimming and swarming assay contained no antibacterial agents. The molten medium in petri dishes was kept inside a clean bench for 1 h to dry. Then, E. coli culture (2 µL) in its initial log phase was inoculated onto the swimming and swarming media in petri dishes. The E. coli cell-inoculated swarm and swim plates were kept at 37 • C in an incubator for 18 h and 10 h, respectively, for the incubation of motile cells. After the incubation period, the diameters of the swim and swarm zones were measured using calibrated digital slide callipers (Mitotoyo, Japan), and photographs of the swim and swarm zones were taken.

Effects of Phenolic Compound-Antibiotic Combinations on the Viability of Small Intestine Cells (IEC-6)
In accordance with the standard EZ-Cytox (EZ-1000; Daeillab Service Co. Ltd., Jeonju, South Korea) assay procedure, the effects of phenolic compound-antibiotic combinations on the viability of Rattus norvegicus small intestine cell line (IEC-6 cells; American Type Culture Collection CRL-1592, VA, United States) were determined in vitro. Briefly, IEC-6 cells were inoculated in Dulbecco's Modified Eagle's medium (DMEM; ThermoFisher Scientific, Waltham, MA, United States), supplemented with 4.5 g/L glucose, 1.5 g/L sodium bicarbonate (Carolina Biological Supply Company, Burlington, NC, United States), 4 mM L-glutamine (ThermoFisher Scientific, Waltham, MA, United States), and 0.1 units/mL fetal bovine serum (10%) and bovine insulin (90%). The cells were cultured in a humidified atmosphere at 37 • C with 5% carbon dioxide (CO 2 ). Each week, the cells were passaged twice at a ratio of 1:5. The IEC-6 cell (2 × 10 4 cells/mL) with a volume of 100 µL was suspended in the abovementioned DMEM in 96-well plates for 24 h at 5% CO 2 and 37 • C. The medium from each well was aspirated and the cells were washed twice. Various concentrations of the antibacterial agents (100 µL) in the abovementioned DMEM were dispensed into each well. Then, IEC-6 cells in the presence of antibacterial agents in the abovementioned DMEM medium were incubated for 24 h at 5% CO 2 and 37 • C. The EZ-Cytox dye (10 µL) was dispensed to each well and the IEC-6 cells were incubated again for 2 h. The OD of the cell suspension in each well was measured using a VersaMax microplate reader (Molecular Devices, San Jose, CA, United States) at a wavelength of 450 nm. Cells not treated with any drugs were assigned as the control. The cell viability (%) was calculated from the following formula where OD is the optical density [51]: Cell viability (%) = OD of drug − treated sample OD of untreated sample × 100 (4)

Statistical Analysis
The results are presented as the mean ± standard deviation (SD) of the triplicate analysis. Statistical analysis was performed using SAS software (SAS Institute Inc., Cary, NC, USA). One-way analysis of variance (ANOVA) followed by F-test was used to compare the results. Statistical significance was considered for P-values of < 0.05.