Phenolics with Bactericidal Activity Alter Motility and Biofilm Formation in Enterotoxigenic,Enteropathogenic,and Enterohemorrhagic Escherichia coli

Abstract

Most Escherichia coli strains are innocuous to human beings; however, some strains can cause diarrhea and are grouped into pathotypes. Since current trends promote the use of natural-origin compounds to control bacteria, in this study, the effects of the phenolic compounds (PCs) tannic acid (TA), gallic acid (GA), methyl gallate (MG), and epigallocatechin gallate (EG) on the growth, swarming motility, biofilm formation, and expression of selected virulence genes of three E. coli pathotypes (enteropathogenic Escherichia coli [EPEC], enterohemorrhagic Escherichia coli [EHEC], and enterotoxigenic Escherichia coli [ETEC]) were evaluated. Minimum bactericidal concentrations (MBCs) were determined by using microtiter plates, and the effects of sublethal PC concentrations on swarming motility were evaluated on Luria-Bertani agar. Biofilm formation was assessed in microtiter plates via crystal violet staining, and the expression levels of genes involved in biofilm formation (flhC, fliA, fliC, and csgA) and swarming motility (csgD and cyaA) were evaluated via quantitative PCR. All PC were bactericidal with minimal bactericidal concentrations ranging from 0.07 to 2.1 mg/mL. At concentrations lower than the MBC, PCs decreased swarming motility (14.8–100%). GA reduced biofilm formation in all of the tested strains; however, TA, MG, and EG induced biofilm formation in some strains at specific concentrations. TA induced the overexpression of csgA, csgD, and cyaA, whereas the other PCs did not have any effects or reduced their expression levels. The PCs tested in this study showed potential to control E. coli strains belonging to the EHEC, ETEC, and EPEC pathotypes by affecting their growth, swarming motility, and virulence gene expression; however, proper concentrations must be used to avoid the induction of undesirable virulence factor genes.

Introduction

Foodborne diseases are a public health concern worldwide due to their morbidity and/or mortality rates and economic losses. It has been estimated that during 2018, 25,606 foodborne infections occurred in the United States alone (Tack et al., 2019).

Although Escherichia coli is usually an innocuous resident of the gastrointestinal tract, some strains can cause gastrointestinal and extra-intestinal diseases. Diarrheagenic E. coli strains have been classified into pathotypes based on the diseases caused in humans and their pathogenic characteristics: enteropathogenic Escherichia coli (EPEC), enteroinvasive Escherichia coli, enterotoxigenic Escherichia coli (ETEC), enterohemorrhagic Escherichia coli (EHEC), enteroaggregative Escherichia coli, diffusely adherent Escherichia coli, and adherent-invasive Escherichia coli (Croxen et al., 2013). EHEC, ETEC, and EPEC are the most common pathotypes worldwide and are primarily associated with consumption of contaminated water or food (Tack et al., 2019).

The hallmark of EHEC is its ability to produce Shiga-toxins (1 and/or 2, Stx), and these strains are responsible for clinical manifestations that can progress to bloody diarrhea with important sequelae, such as hemolytic uremic syndrome (Carter et al., 2011).

ETEC strains express virulence factors such as surface adhesins, which act as colonization factors, and heat-stable and heat-labile enterotoxins. These strains cause moderate to severe diarrhea mainly in infants and travelers worldwide (Robins-Browne et al., 2016). ETEC infections generally occur after consumption of contaminated water, ice, and/or food, and, although uncommon, person-to-person contact (Vidal et al., 2016).

The members of the pathotype EPEC are also important causes of watery, nonbloody infant diarrhea, especially in developing countries. In these strains, the protein intimin plays an important role in virulence, together with characteristic attaching-effacing lesions (Robins-Browne et al., 2016). Although this bacterium is transmitted via a fecal–oral route, several outbreaks have occurred via ingestion of contaminated food or water (Croxen et al., 2013).

To control these pathogens, current trends suggest the replacement of chemical preservatives in foods with those of natural origin (Gyawali and Ibrahim, 2014). Many of these active compounds have been isolated and are currently synthetized artificially. Most studies of the antimicrobial activity of natural products focus on inhibiting growth or killing microorganisms (Chouhan et al., 2017); however, to treat or prevent diseases, it is important not only to kill the pathogen but also to inhibit the production of its virulence factors. This concept, known as “anti-virulence” therapy, is based on the use of compounds at sublethal concentrations that can still inhibit physiological processes such as biofilm formation, swarming motility, quorum sensing, and the production of toxins, pigments, surfactants, and enzymes (Roy et al., 2011; García-Heredia et al., 2013; Silva et al., 2016).

E. coli forms biofilms on foods and contact surfaces and attaches to and colonizes the human intestine. Biofilm-embedded bacteria are generally more tolerant to environmental stress, sanitizing agents, and antimicrobials (Singh et al., 2017). Bacterial swarming motility is a collective mode of movement through which bacteria rapidly migrate in coherent patterns; as the fastest mode of surface translocation, swarming motility allows bacterial colonization of nutrient-rich environments (Verstraeten et al., 2008). Because swarmer cells are highly resistant to antimicrobial agents, swarming bacterial communities are problematic in environmental and clinical contexts (Kim and Surette, 2003; Kim et al., 2003; Lai et al., 2009).

Many studies indicate that phenolic compounds (PCs) are responsible for the antimicrobial activity of many plants (Díaz-Gómez et al., 2014; Ortega-Ramirez et al., 2014). These compounds are antimicrobial due to the activity of their functional groups and its synergies. The most commonly identified mechanism of action is to cause alterations in the bacterial membrane, whose activity is essential for vital cellular processes; thus, it is reasonable to analyze the feasibility of their use to target E. coli pathotypes.

The identification of compounds that can impede targets such as bacterial motility and biofilm formation at sublethal concentrations while killing the bacteria at higher concentrations will be very useful for controlling environmental pathogens (Corzo-Ariyama et al., 2019). We wondered whether phenolics usually found in plants had an effect on growth and virulence factors of pathogenic E. coli. In this study, the lethal concentrations, and the effects of the sublethal concentrations of the PCs tannic acid (TA), gallic acid (GA), methyl gallate (MG), and epigallocatechin gallate (EG) on the growth, swarming mobility, biofilm formation, and expression of virulence genes of three E. coli pathotypes (EPEC, EHEC, and ETEC) were assessed.

Materials and Methods

Polyphenolic compounds

TA (403040), GA (G7384), MG (274194), and EG (E4268) were purchased from Sigma-Aldrich (México). Stock solutions (50 mg/mL) were diluted in 35% ethanol and stored at 4°C in amber flasks. Working solutions were prepared by taking appropriate volumes from the stock solutions and diluted in water immediately before assays.

Microbial strains and culture conditions

The EHEC serotype O157:H7 strain (ATCC 43895) was kindly provided by Dr. Lynn McLasborougth, Department of Food Science, University of Massachusetts, (Amherst, MA). The EPEC strain serotype O111:NM (ATCC 43887) and the ETEC strain serotype O78:H11 (ATCC 35401) were obtained commercially. All of the strains were stored at −80°C in brain-heart infusion broth (BHI; Bioxon, Becton-Dickinson, Mexico) with 20% glycerol (Sigma-Aldrich). Before use, an aliquot of culture was used to inoculate BHI agar tubes, which were then incubated for 48 h at 37°C and then stored at 4°C. For the assays, a loop of culture was transferred to tubes with 5 mL of slant Muller-Hinton (MH; Difco, Becton-Dickinson, USA) agar or Luria-Bertani (LB) broth (EMD Millipore Corporation, Germany) and incubated for 24 h at 37°C. For working solutions, colonies from MH agar were suspended in saline solution (SS; 0.85% p/v), and the suspension was adjusted to 1 × 10⁸ colony-forming unit (CFU)/mL (T₆₀₀ = 74).

Minimum bactericidal concentration

Aliquots (125 μL) of various concentrations of the PCs (50–1250 μg/mL) were added to sterile 96-well polystyrene U-microtiter plates (Nunclon™ Delta Surface; Thermo Fisher Scientific, Marietta, OH) filled with 122.5 μL of 2 × MH broth. Next, fresh cultures of bacteria (2.5 μL, 1.0 × 10⁸ CFU/mL) were used to inoculate each well, and the plates were subsequently incubated at 37°C for 24 h. After incubation, 10-μL aliquots from each well were dropped onto MH agar plates and then incubated at 37°C for 24 h (Sánchez et al., 2013). The minimum bactericidal concentration (MBC) was defined as the lowest PC concentration that prevented any visible bacterial growth on the MH agar plate after 24 h of incubation (Kang et al., 2018).

Once the MBCs were determined, the effects of sub-MBC concentrations (0.75 and 1.25 mg/mL for TA and GA, and 0.05 and 0.2 mg/mL for MG and EG) on bacterial growth in MH broth were determined. Aliquots were taken from these cultures every 2 h (37°C), diluted in SS, and plated on MH agar. After incubation at 37°C/24–48 h, the CFU numbers were enumerated. The pH of the medium with compounds added at their MBC were: TA 6.7; GA 5.3; MG 7.0; and EG 7.0.

Effects of PCs on swarming motility

Swarming motility was analyzed via the method of García-Heredia et al. (2016) with minor modifications. Briefly, Petri plates were prepared with soft LB agar (0.35%) containing various PC concentrations. Five microliters of an active culture (1.0 × 10⁸ CFU/mL) was placed onto the center of a plate followed by incubation for 18–24 h at 37°C. Next, the colony diameter was measured and compared with the diameter of the control colony (without PC) to determine the percent of reduction of the area of swarming motility.

Biofilm assay

Sterile 96-well polystyrene U-microtiter plates were filled with 125 μL 2 × LB broth supplemented with 0.2% glucose (Sigma-Aldrich), 1% sodium citrate (Sigma-Aldrich), and 150 μL of PC at final concentrations lower than the MBC (previously determined). The culture medium was inoculated with 3 μL of active E. coli cultures (1.0 × 10⁸ CFU/mL), and it was incubated for 24 h at 37°C. After measuring the A_630nm values (Bio-Tek Epoch, BioTek^® Winooski, VT), the medium was removed from the wells, and the plate was washed twice with distilled water and left to dry for 24 h. Two hundred fifty microliters of 0.1% crystal violet (CV; Sigma-Aldrich) was added to the wells followed by incubation for 15 min at 37°C. Each plate was washed twice, and 250 μL of 95% ethanol was added to dissolve the residual CV inside the biofilm matrix. The A_570nm values were recorded (García-Heredia et al., 2016). Medium with and without PCs, and bacteria without compounds were used as controls.

Biofilm formation was quantified by determining the biofilm formation index (BFI) obtained from the formula: BFI = (AB – CW)/G (García-Heredia et al., 2013), where AB is the A_570nm of the stained attached microorganism, CW is the A_570nm of the stained control (medium without bacteria), and G is the A_630nm of the cell growth in liquid culture medium without PC. According to Teh et al. (2010), BFI values >1.10 indicate strong biofilms, BFI values from 0.70 to 1.09 correspond to moderate biofilms, BFI values from 0.35 to 0.69 indicate weak biofilms, and BFI values <0.35 indicate no biofilm formation.

Gene expression analysis

The effects of the PCs on the expression levels of genes associated with swarming motility (flhC, fliA, fliC _EH, and fliC _ET) and biofilm formation (csgA, csgD, and cyaA) were analyzed. The oligonucleotides used in this study (Table 1) were designed with the support of the National Center for Biotechnology Information database (Altschul et al., 1990) by using the Primer3Plus software (Untergasser et al., 2007) and Primer-BLAST (Ye et al., 2012). To assess primer efficiency, the ΔG was determined by using the OligoAnalyzer v3.0 software (PrimerQuest^® Program; IDT, Coralville, IA), and ΔGs lower than −6 for heterodimerizations and ΔGs lower than −3 for homodimerizations were required.

Table 1.

Primers Used to Determine the Effects of Phenolics on the Expression Levels of Swarming Motility- and Biofilm Formation-Associated Genes

Target gene	Relevant physiological process	Primers 5′-3′	Assay
flhC	Transcription activator	CGGCAGGATTCTGGGAAAGT	Swarming motility
flhC	Transcription activator	CTGGACATTGGTGCGGTTTG
fliA	Flagellar biosynthesis sigma factor	TTAGGGATCGATATTGCCGATT
fliA	Flagellar biosynthesis sigma factor	CGTAGGAGAAGAGCTGGCTGTT
fliC_EH	Flagellin (EHEC O157:H7)	F-TCAAGTTGCCTGCATCGTCT
fliC_EH	Flagellin (EHEC O157:H7)	R-ACCCTTCATGCTGATGTGGG
fliC_ET	Flagellin (ETEC O78:H11)	F-TGTCGCTGTTGACCCAGAAT
fliC_ET	Flagellin (ETEC O78:H11)	R-AGAGAGACGTTCAATGGCGG
csgA	Major curli subunit precursor	F-TGGCAGGTGTTGTTCCTCAGT	Biofilm formation
csgA	Major curli subunit precursor	R-GTCAGAGTTACGGGCATCAGTTT
csgD	Transcriptional regulator	F-CCGCTTGTGTCCGGTTTT
csgD	Transcriptional regulator	R-GAGATCGCTCGTTCGTTGTTC
cyaA	Adenylate cyclase	F-AGCTTTGGCGAGAATCAAAA
cyaA	Adenylate cyclase	R-CGCAGATGCTGGCTATAACA
rrsH	rRNA 16S	F-CGATGCAACGCGAAG	Constitutive gene
rrsH	rRNA 16S	R-CCGGACCGCTGGCAA	Constitutive gene

EHEC, enterohemorrhagic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; F, forward; R, reverse.

To evaluate the expression of genes associated with swarming motility, bacterial samples were taken from the edge of the swarming colonies exposed to PC. For biofilm formation, aliquots (400 μL) of activated bacterial cultures were inoculated into tubes containing 4 mL of fresh LB broth supplemented with 0.2% glucose added with PCs at various concentrations. The cultures were incubated for 24 h at 37°C. RNA was extracted via the Trizol chloroform technique (Czapski and Trun, 2014). The integrity of the obtained RNA was determined by measuring the A_260/280nm ratio (NanoDrop 2000), and the RNA was then used for DNA synthesis (iScript™ cDNA Synthesis Kit; Bio-Rad Laboratories, Mexico). Quantitative PCR was performed by using a real time-PCR system (PikoReal 96; Thermo Scientific, Mexico) with Q SYBR Green Supermix (Bio-Rad Laboratories). The PCR cycling conditions were 95°C for 3 min, followed by 40 cycles at 95°C for 15 s and one cycle of 60°C for 45 s.

Changes in gene expression between the treated and untreated bacteria were determined via the delta Ct method (Livak and Schmittgen, 2001). The Escherichia coli 16S rRNA housekeeping gene was used to normalize the levels of the target genes. Untreated bacteria were used as controls and as a baseline to compare the gene expression levels.

Statistical analysis

All of the experiments were performed at least three times independently, and each time the experiment was tested in triplicate. Comparisons between treatments were analyzed via analysis of variance, and the Tukey–Kramer test was used to compare the means via Number Cruncher Statistical System version 11 (NCSS, LLC). The significance used was 95%.

Results

Minimum bactericidal concentrations

All the PCs analyzed were bactericidal against the E. coli pathotypes studied (Table 2). MG showed an inhibitory effect at lower concentrations (0.4–0.07 mg/mL) compared with the concentrations required for the other compounds (e.g., TA 1.8–2.5 mg/mL). In general, all of the strains exhibited similar sensitivities to the compounds with only minor variations. Sub-inhibitory concentrations (0.75 and 1.25 mg/mL for TA and GA and 0.05 and 1.25 for MG and EG) did not cause growth reductions for the analyzed strains (p ≥ 0.05); thus, these concentrations were used for further assays.

Table 2.

Minimal Bactericidal Concentrations of Phenolic Compounds Against Three Escherichia coli Pathotypes (EPEC, EHEC, and ETEC)

Polyphenolic compound	MBC (mg/mL)
Polyphenolic compound	EHEC (O157:H7)	ETEC (O78:H11)	EPEC (O111:NM)
Tannic acid	2.5 ± 0.01	2.4 ± 0.03	1.8 ± 0.02
Gallic acid	2.1 ± 0.02	2.2 ± 0.04	2.1 ± 0.01
Methyl gallate	0.4 ± 0.02	0.6 ± 0.01	0.07 ± 0.004
Epigallocatechin gallate	0.3 ± 0.008	0.6 ± 0.01	0.11 ± 0.003

± standard deviation.

EPEC, enteropathogenic Escherichia coli; MBC, minimum bactericidal concentration.

Effects of PCs on swarming motility

These assays were performed only with the EHEC and ETEC strains, because the EPEC strain is non-motile. All the PCs affected swarming motility in the tested strains (Table 3); the ETEC strain was affected more strongly (with reductions from 89.6% to 100%) than was the EHEC strain (with reductions from 14.8% to 100%).

Table 3.

Reduction (%) and Effects of Subinhibitory Polyphenol Concentrations on Swarming Motility and Biofilm Formation in Three Escherichia coli Pathotypes

Polyphenolic compound	Concentration (mg/mL)	EHEC (O157:H7)		ETEC (O78:H11)		EPEC (O111:NM)^a
Polyphenolic compound	Concentration (mg/mL)	Swarming motility in mm (% reduction)	BFI (type^b)	Swarming mobility in mm (% reduction)	BFI (type^b)	Biofilm formation index (type^b)
Control	—	69.3 ± 3.8^c	0.5 ± 0.06^d	74.0 ± 0.0^c	1.23 ± 0.08^d,e	1.1 ± 0.05^e
Control	—	(0)	(WB)	(0)	(SB)	(SB)
Tannic acid	0.75	0.3 ± 0.6^f	1.1 ± 0.06^c	0.0 ± 0.0^g	0.90 ± 0.00^g	1.3 ± 0.08^c
	0.75	(99.5)	(SB)	(100)	(MB)	(SB)
	1.25	0.0 ± 0.0^f	0.5 ± 0.16^d	0.0 ± 0.0^g	1.1 ± 0.17^e,g	1.2 ± 0.03^d
	1.25	(100)	(WB)	(100)	(SB)	(SB)
Gallic acid	0.75	33.0 ± 10^g,h	0.2 ± 0.03^e,g	2.7 ± 0.57^e,g	0.7 ± 0.08^h	1.1 ± 0.09^d,e
	0.75	(52.4)	(NB)	(96.3)	(MB)	(SB)
	1.25	4.3 ± 0.6^f	0.3 ± 0.05^e	1.0 ± 0.00^g	0.4 ± 0.04^f	0.8 ± 0.08^g
	1.25	(93.7)	(WB)	(98.3)	(WB)	(MB)
Methyl gallate	0.05	40.0 ± 5.0^e,g	0.2 ± 0.04^g	5.0 ± 0.00^d,e	1.56 ± 0.03^c	0.3 ± 0.04^h
	0.05	(41.8)	(NB)	(93.2)	(SB)	(NB)
	0.2	24.0 ± 7.5^h	0.2 ± 0.03^e,g	7.3 ± 0.57^d	1.4 ± 0.30^c,d	NDⁱ
	0.2	(65.8)	(NB)	(90.0)	(SB)
Epigallocatechin gallate	0.05	59.0 ± 2.7^c,d	1.0 ± 0.13^c	6.3 ± 3.06^d	1.4 ± 0.07^c,d	0.9 ± 0.09^g
	0.05	(14.8)	(MB)	(91.4)	(SB)	(MB)
	0.2	55.3 ± 4.0^d,e	0.2 ± 0.05^e,g	7.7 ± 0.57^d	1.4 ± 0.04^c,d	ND
	0.2	(23.0)	(NB)	(89.6)	(SB)

± standard deviation.

Strain is not motile.

Type of biofilm according to BFI values: No biofilm formed, (NB), (BFI <0.35); weak biofilm, (WB), (BFI >0.35 to <0.69); moderate biofilm, (MB), (BFI >0.7 to <1.09); and strong biofilm, (SB) (BFI >1.1).

c–h

Indicate significant differences (p ≤ 0.05). Comparisons were performed between the control and subinhibitory polyphenolic concentrations.

Assay was not performed, because the concentration to be used was higher than the MBC.

BFI, biofilm formation index; MBC, minimum bactericidal concentration; ND, not detected.

AT was the most effective compound for inhibiting swarming motility (99.5–100%) at both concentrations in the strains analyzed, followed by GA (52.4–98.3%), MG (41.8–93.2%), and EG (14.8–91.4%). The compounds had different effects on the tested pathotypes. The effects of the two PC doses were similar (p ≥ 0.05) for the ETEC strain, whereas dose dependence was observed in most of the treatments for the EHEC strain.

Biofilm formation

In general, most compounds induced weak or non-formed biofilm in the EHEC strain, whereas the ETEC and EPEC strains formed strong biofilms (Table 3). In general, the PCs either decreased (41% of treatments, 9/22) or did not affect (in 50% of treatments, 11/22) biofilm formation in the tested strains. In general, the ETEC biofilm formation was the most weakly affected, followed by the EPEC strain, and the EHEC was the most sensitive. The lower TA concentration (0.75 mg/mL), but not the higher concentration, increased EHEC biofilm formation from weak to strong (BFI = 0.5–1.1). In general, no dose dependence was observed at any of the analyzed PC concentrations.

Gene expression analysis

In general, the expression levels of the tested genes did not follow a regular pattern, as only the expression levels of most of the motility genes were decreased after GE treatment in all the tested strains. No dose dependence was observed in most of the treatments (Table 4). An interesting finding was that although swarming motility was inhibited or decreased via treatment with the PCs, swarming-related genes were overexpressed (p ≤ 0.05) on most of the treatments, especially in the EHEC and ETEC strains. TA greatly increased the expression levels of flhC and fliA in all of the tested strains, particularly in the EPEC strain. The fliA expression was also increased via GA and MG treatment in the EHEC and ETEC strains.

Table 4.

Fold Changes in the Expression Levels of Virulence-Related Genes for Escherichia coli Pathotypes After Addition of Four Polyphenols

Polyphenolic compound	Concentration (mg/mL)	Expression level fold changes of selected virulence genes
		Motility			Biofilm
		flhC	fliA	fliC	CsgA	CsgD	CyaA
EHEC (O157:H7)
Tannic acid	0.75	6.12 ± 2.00^a,b	0.55 ± 0.59^b	−0.08 ± 0.41^c	173.0 ± 42.2^a	0.22 ± 0.02^b	−.31 ± 0.09^a
Tannic acid	1.25	6.11 ± 2.06^a,b	9.86 ± 9.83^b	−0.68 ± 0.08^c	99.77 ± 3.17^b	4.09 ± 1.17^a	−0.59 ± 0.06^a
Gallic acid	0.75	9.62 ± 1.29^a	54.94 ± 15.23^a	48.33 ± 16.96^a	35.88 ± 12.92^c	−0.99 ± 0.00^c	4.93 ± 2.14^a
Gallic acid	1.25	1.07 ± 0.12^c,d	12.41 ± 1.61^b	8.36 ± 0.19^b,c	5.18 ± 3.44^b	−0.99 ± 0.00^c	9.84 ± 11.02^a
Methyl gallate	0.05	2.45 ± 2.90^b–d	22.37 ± 11.22^b	3.60 ± 1.58^c	1.97 ± 0.89^c	−0.97 ± 0.01^c	−0.13 ± 0.64^a
Methyl gallate	0.2	4.29 ± 1.35^b,c	18.97 ± 9.21^b	39.6 ± 27.4^a,b	22.39 ± 11.65^c	−0.94 ± 0.00^b,c	−0.90 ± 0.01^a
Epigallocatechin gallate	0.05	−0.79 ± 0.23^d	−0.83 ± 0.09^b	−0.81 ± 0.05^c	43.61 ± 15.75^c	−0.88 ± 0.00^b,c	−0.94 ± 0.01^a
Epigallocatechin gallate	0.2	−0.51 ± 0.26^d	−0.83 ± 0.00^b	−0.74 ± 0.08^c	−0.50 ± 0.10^c	−0.97 ± 0.00^c	−0.96 ± 0.00^a
ETEC (O78:H11)
Tannic acid	0.75	7.19 ± 3.25^a	12.42 ± 2.99^a,b	0.21 ± 0.55^c,d	42.70 ± 12.46^a	2.77 ± 3.18^a	0.46 ± 0.35^a
Tannic acid	1.25	4.42 ± 1.37^a,b	15.20 ± 3.93^a	2.14 ± 2.79^b–d	17.57 ± 4.74^b	1.52 ± 1.39^a,b	−0.04 ± 0.46^a,b
Gallic acid	0.75	1.62 ± 0.55^b–d	11.12 ± 0.96^a,b	9.99 ± 3.82^b,c	−0.97 ± 0.00^c	−0.91 ± 0.01^b	−0.95 ± 0.03^b
	1.25	−0.25 ± 0.37^c,d	1.16 ± 1.89^a,b	3.96 ± 2.24^b–d	0.98 ± 0.44^c	−0.97 ± 0.00^b	−0.97 ± 0.01^b
Methyl gallate	0.05	1.06 ± 0.29^b–d	12.55 ± 7.33^a,b	11.74 ± 7.25^b	−0.00 ± 0.17^c	1.50 ± 0.77^a,b	−0.94 ± 0.04^b
Methyl gallate	0.2	3.33 ± 1.24^b,c	11.53 ± 10.84^a,b	24.86 ± 4.42^a	−0.95 ± 0.01^c	−0.56 ± 0.09^a,b	−0.44 ± 0.49^a,b
Epigallocatechin gallate	0.05	−0.82 ± 0.03^d	0.00 ± 0.49^b	−0.00 ± 0.44^c,d	−0.18 ± 0.29^c	−0.58 ± 0.33^a,b	−0.64 ± 0.19^b
Epigallocatechin gallate	0.2	−0.76 ± 0.03^d	−0.63 ± 0.04^b	−0.75 ± 0.01^d	9.19 ± 1.88^b,c	0.27 ± 0.58^a,b	−0.29 ± 0.51^a,b
EPEC (O111:NM)
Tannic acid	0.75	43.65 ± 4.08^a	157.09 ± 14.54^a		30.58 ± 6.18^a	5.62 ± 3.07^b,c	6.28 ± 5.54^a
Tannic acid	1.25	11.67 ± 0.35^b	49.94 ± 3.03^b		23.65 ± 8.24^a	7.13 ± 1.2^b,c	5.93 ± 2.4^a
Gallic acid	0.75	0.29 ± 0.038^c	−0.06 ± 0.07^c		−0.98 ± 0.00^b	−0.67 ± 0.15^b,c	−0.96 ± 0.03^b
Gallic acid	1.25	−0.40 ± 0.063^c	0.74 ± 0.11^c		−0.94 ± 0.02^b	−0.94 ± 0.00^c	−0.95 ± 0.03^b
Methyl gallate	0.05	−0.18 ± 0.20^c	0.00 ± 0.21^c		3.58 ± 1.50^b	32.44 ± 6.52^a	3.65 ± 0.50^a,b
Methyl gallate	0.2	ND^e	ND		ND	ND	ND
Epigallocatechin gallate	0.05	−0.79 ± 0.00^c	−0.70 ± 0.02^c		20.82 ± 5.99^a	9.60 ± 2.15^b	5.36 ± 0.91^a
Epigallocatechin gallate	0.2	ND	ND		ND	ND	ND

± standard deviation.

a–d

Indicate significant differences (p ≤ 0.05). Comparisons were performed between the same genes in E. coli strains analyzed for each compound tested.

Assay was not performed, because the concentration used was higher than the MBC.

The biofilm-related genes also exhibited highly variable expression patterns, and no dose dependence was observed; only GA treatment (at both concentrations) resulted in lower csgD expression in all the tested strains, whereas csgA expression increased in response to TA exposures in all of the strains (Table 4).

Discussion

The MBCs of the PCs varied from strain to strain and ranged from 0.07 to 2.5 mg/mL. This variation could be a consequence of the physical and chemical characteristics of the PCs, such as molecule size, active groups, and hydrophobicity (Daglia, 2012). Differences in the bacterial isolates could also influence the degree of strain susceptibility to the PCs. Susceptibility differences between strains to natural compounds have been commonly reported (Choi et al., 2014).

The MBCs determined in this study are consistent with those previously reported for TA against E. coli (Taguri et al., 2004) and for EG (Taguri et al., 2004; Lee et al., 2009); however, the MBC of GA was slightly lower (2.1–2.2 mg/mL compared with 5 mg/mL; Borges et al., 2013). These minor discrepancies could be due to variations between the strains and the reagents used for the MBC determination (Borges et al., 2013).

Among the tested PCs, MG showed the highest antimicrobial activity for all the strains (MBC = 0.07 to 0.6 mg/mL). An extract from Toona sureni leaves exhibited an MBC of 7.5 mg/mL against E. coli (Ekaprasada et al., 2015). The main compound with antimicrobial activity in these leaves was MG. In a study performed by Sánchez et al. (2013), the MBC values of MG against Vibrio cholerae strains 569-B and 1837 were 30 and 50 μg/mL, respectively. MG induces cell membrane alterations that increase their permeability, thus decreasing the cytoplasmic pH and promoting cell membrane hyperpolarization, which, in turn, lead to lower cellular adenosine triphosphate concentrations (Castillo et al., 2017).

Sublethal concentrations may induce atypical biological responses in bacteria (Bernier and Surette, 2013). Although the sub-inhibitory concentrations used in this study did not affect bacterial growth, these concentrations were sufficient to alter physiological processes. Bacterial motility is a complex process that contributes to biofilm formation, which is critical for host–microbe interactions and pathogenesis (Shi et al., 2017; Kang et al., 2018). All of the subinhibitory PC concentrations analyzed in this study decreased the EHEC and ETEC swarming motility. Although the PCs did reduce motility, they did not always limit biofilm formation. The relationship between motility and biofilm formation tends to be complex, and these processes might share components at specific stages and under specific conditions, thus allowing bacteria to choose between swarming and biofilm formation (Verstraeten et al., 2008). Discrepancies between biofilm formation and swarming motility have been previously observed (Dusane et al., 2015). Physical interactions between TA and the Pseudomonas aeruginosa flagellum have been reported (O'May et al., 2012), and these interactions can modify processes such as secondary metabolite synthesis, block swarming, and increase the bacteria's ability to form biofilms; as a consequence, the bacteria are protected against antimicrobial effects.

The intracellular signaling molecule c-di-GMP, which is synthesized by diguanylate cyclases and degraded by specific phosphodiesterases (EAL- or HD-GYP domain proteins), is a secondary messenger that can respond to extracellular signals to regulate multicellular behavior in bacteria. It has been linked to enhanced biofilm formation and with reductions in swarming motility and virulence in bacteria (Verstraeten et al., 2008; O'May et al., 2012; Defoirdt, 2018). In general, high c-di-GMP concentrations correlate with enhanced biofilm formation and reduced swarming motility and virulence. Some bacteria have multiple proteins with GGDEF or EAL domains that act on c-di-GMP to modulate various physiological processes, thus resulting in an inverse relationship between swarming and biofilm formation (Verstraeten et al., 2008), such as that observed in our assays. The opposite effect, that is, increased swarming with reduced biofilm formation, has also been thoroughly documented (Caiazza et al., 2007; Kuchma et al., 2007; Verstraeten et al., 2008; Shrout et al., 2011); however, we did not observe this effect in our study.

In Salmonella, MG can inhibit DNA gyrase or interfere with the ATPase system (Choi et al., 2014). These effects create stress conditions under which biofilms provide protection to bacterial populations to increase their chances of surviving. On the other hand, other authors have demonstrated that GA can chelate metal ions that serve as cofactors of enzymes, and such metal ion chelation can inhibit bacterial growth and provoke reduced swarming motility and biofilm formation (Liu et al., 2003; Zaidi–Yahiaoui et al., 2008).

Sub-inhibitory EG concentrations induce morphological changes in Gram-positive and Gram-negative bacteria, including Escherichia coli O157:H7, in the form of perforations and grooves in the cell envelope (Cui et al., 2012). In our study, this compound affected swarming motility in ETEC and EHEC strains and reduced biofilm formation in the EHEC and EPEC strains.

The primary function of the flhDC master operon is to control flagellum biogenesis (Fraser et al., 2000), whereas the flhDC gene products form a key regulator of swarmer cell differentiation in several Enterobacteriaceae species. fliA is a flagellar biosynthesis sigma factor; fliC gene encodes flagellin, which is necessary to form flagella. No expression of fliC was observed in the EPEC isolate, which was in accordance with a lack of motility in this isolate. In this study, EG provoked reduced expression of motility-related genes in EHEC and ETEC strains, which was consistent with an observed reduction in swarming. However, this behavior was not observed for the other PCs, and, in some cases, overexpression of motility genes occurred concomitantly with decreased swarming motility.

The csgA, csgD, and cyaA genes are important for biofilm formation. In our assays, GA induced a reduced expression of these genes in the EPEC and ETEC strains, which was consistent with the observed reduction in biofilm formation; however, in the EHEC strain, csgA and cyaA were overexpressed without an increase in biofilm formation. A similar effect was observed for MG and csgA expression. A possible cause of this observation is that metal ion chelation by PC might hinder the formation of stronger biofilms, as these ions are also necessary for growth and enzymatic activity (Liu et al., 2003; Zaidi–Yahiaoui et al., 2008).

In conclusion, TA, GA, MG, and EG exhibited bactericidal activity against EHEC, EPEC, and ETEC E. coli pathotypes, and at low concentrations, these compounds affected important virulence factors such as biofilm formation and swarming motility without significant reductions in the cell populations. However, it is important to consider that at some concentrations, biofilm formation could increase. At 1.25 mg/mL, GA simultaneously inhibited swarming and biofilm formation. These compounds are promising candidates for use in controlling these pathogenic bacteria in food, provided the appropriate concentrations are used. This is to greatly reduce the probability that sublethal concentrations are used commercially. Studies involving a greater number of strains of these and other pathotypes are being carried out to analyze their variability in biofilm formation and swarming motility, and the effect of these compounds.

Footnotes

Acknowledgment

C.E.G.S. acknowledges a postgraduate fellowship from CONACYT.

Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by a grant from the Consejo Nacional de Ciencia y Tecnología de México (CONACYT) grant no. CB-2014-236642 and the Universidad Autónoma de Nuevo León.

References

Altschul

, Gish

, Miller

, Myers

, Lipman

. Basic local alignment search tool. J Mol Biol, 1990; 215:403–410.

Bernier

, Surette

. Concentration-dependent activity of antibiotics in natural environments. Front Microbiol, 2013; 4:20.

Borges

, Ferreira

, Saavedra

, Simoes

. Antibacterial activity and mode of action of ferulic and gallic acids against pathogenic bacteria. Microb Drug Resist, 2013; 19:256–265.

Caiazza

, Merritt

, Brothers

, O'Toole

. Inverse regulation of biofilm formation and swarming motility by Pseudomonas aeruginosa PA14. J Bacteriol, 2007; 189:3603–3612.

Carter

, Brandl

, Louie

, Kyle

, Carychao

, Cooley

, Parker

, Bates

, Mandrell

. Distinct acid resistance and survival fitness displayed by curli variants of enterohemorrhagic Escherichia coli O157: H7. Appl Environ Microbiol, 2011; 77:3685–3695.

Castillo

, Dávila-Aviña

, Heredia

, Garcia

. Antioxidant activity and influence of Citrus byproduct extracts on adherence and invasion of Campylobacter jejuni and on the relative expression of cadF and ciaB. Food Sci Biotechnol, 2017; 26:453–459.

Choi

, Mun

, Chahar

, Bharaj

, Kang

, Kim

, Shin

, Kwon

. Methyl gallate from Galla rhois successfully controls clinical isolates of Salmonella infection in both in vitro and in vivo systems. PLoS One, 2014; 9:e102697.

Chouhan

, Sharma

, Guleria

. Antimicrobial activity of some essential oils—present status and future perspectives. Medicines, 2017; 4:58.

Corzo-Ariyama

, García-Heredia

, Heredia

, García

, León

, Jaykus

, Solís-Soto

. Phylogroups, pathotypes, biofilm formation and antimicrobial resistance of Escherichia coli isolates in farms and packing facilities of tomato, jalapeño pepper and cantaloupe from Northern Mexico. Int J Food Microbiol, 2019; 290:96–104.

10.

Croxen

, Law

, Scholz

, Keeney

, Wlodarska

, Finlay

. Recent advances in understanding enteric pathogenic Escherichia coli . Clin Microbiol Rev, 2013; 26:822–880.

11.

Cui

, Oh

, Lim

, Youn

, Lee

, Pak

, Park

, Jo

, Park

. AFM study of the differential inhibitory effects of the green tea polyphenol (−)-epigallocatechin-3-gallate (EGCG) against Gram-positive and Gram-negative bacteria. Food Microbiol, 2012; 29:80–87.

12.

Czapski

, Trun

. Expression of csp genes in E. coli K-12 in defined rich and defined minimal media during normal growth, and after cold-shock. Gene, 2014; 547:91–97.

13.

Daglia

Polyphenols as antimicrobial agents. Curr Opin Biotech, 2012; 23:174–181.

14.

Defoirdt

Quorum-sensing systems as targets for antivirulence therapy. Trends Microbiol, 2018; 26:313–328.

15.

Díaz-Gómez

, Toledo-Araya

, López-Solís

, Obreque-Slier

. Combined effect of gallic acid and catechin against Escherichia coli . LWT-Food Sci Technol, 2014; 59:896–900.

16.

Dusane

, O'May

, Tufenkji

. Effect of tannic and gallic acids alone or in combination with carbenicillin or tetracycline on Chromobacterium violaceum CV026 growth, motility, and biofilm formation. Can. J Microbiol, 2015; 61:487–494.

17.

Ekaprasada

, Nurdin

, Ibrahim

, Dachriyanus

. Antibacterial activity of methyl gallate isolated from the leaves of Toona sureni . Int J Adv Sci Eng Inf Technol, 2015; 5:280–282.

18.

Fraser

, Furness

, Hughes

Swarming migration by Proteus and related bacteria. In: Brun

, Shimkets

, eds. Prokaryotic Development. Washington, DC: ASM Press, 2000, pp. 381–401.

19.

García-Heredia

, García

, Merino-Mascorro

JÁ

, Feng

, Heredia

. Natural plant products inhibits growth and alters the swarming motility, biofilm formation, and expression of virulence genes in enteroaggregative and enterohemorrhagic Escherichia coli . Food Microbiol, 2016; 59:124–132.

20.

García-Heredia

, Orue

, Heredia

, García

. Efficacy of citrus-based disinfectants to inhibit growth, swarming, and biofilm formation of Salmonella and decontaminate parsley. J Food Agric Environ, 2013; 11:1295–1299.

21.

Gyawali

, Ibrahim

. Natural products as antimicrobial agents. Food Control, 2014; 46:412–429.

22.

Kang

, Li

, Liu

, Jin

, Wang

, Sun

. The specific effect of gallic acid on Escherichia coli biofilm formation by regulating pgaABCD genes expression. Appl Microbiol Biot, 2018; 102:1837–1846.

23.

Kim

, Killam

, Sood

, Surette

. Swarm-cell differentiation in Salmonella enterica serovar Typhimurium results in elevated resistance to multiple antibiotics. J Bacteriol, 2003; 185:3111–3117.

24.

Kim

, Surette

. Swarming populations of Salmonella represent a unique physiological state coupled to multiple mechanisms of antibiotic resistance. Biol Proced Online, 2003; 5:189.

25.

Kuchma

, Brothers

, Merritt

, Liberati

, Ausubel

, O'Toole

. BifA, a cyclic-Di-GMP phosphodiesterase, inversely regulates biofilm formation and swarming motility by Pseudomonas aeruginosa PA14. J Bacteriol, 2007; 189:8165–8178.

26.

Lai

, Tremblay

, Déziel

. Swarming motility: A multicellular behaviour conferring antimicrobial resistance. Environ Microbiol, 2009; 11:126–136.

27.

Lee

, Kim

, Lim

, Nam

, Youn

, Nam

, Kim

, Park

. Antipathogenic properties of green tea polyphenol epigallocatechin gallate at concentrations below the MIC against enterohemorrhagic Escherichia coli O157: H7. J Food Protect, 2009; 72:325–331.

28.

Liu

, Hsu

, Tsai

, Chan

, Liu

, Thomas

, Tomlinson

, Lo

, Lin

. Antihypertensive effects of tannins isolated from traditional Chinese herbs as non-specific inhibitors of angiontensin converting enzyme. Life Sci, 2003; 73:1543–1555.

29.

Livak

, Schmittgen

. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods, 2001; 25:402–408.

30.

O'May

, Ciobanu

, Lam

, Tufenkji

. Tannin derived materials can block swarming motility and enhance biofilm formation in Pseudomonas aeruginosa . Biofouling, 2012; 28:1063–1076.

31.

Ortega-Ramirez

, Rodriguez-Garcia

, Leyva

, Cruz-Valenzuela

, Silva-Espinoza

, Gonzalez-Aguilar

, Siddiqui

, Ayala-Zavala

. Potential of medicinal plants as antimicrobial and antioxidant agents in food industry: A hypothesis. J Food Sci, 2014; 79:R129–R137.

32.

Robins-Browne

, Holt

, Ingle

, Hocking

, Yang

, Tauschek

. Are Escherichia coli pathotypes still relevant in the era of whole-genome sequencing?. Front Cell Infect Microbiol, 2016; 6:141.

33.

Roy

, Adams

, Bentley

. Developing next generation antimicrobials by intercepting AI-2 mediated quorum sensing. Enzyme Microb Technol, 2011; 49:113–123.

34.

Sánchez

, Heredia

, Camacho-Corona

, García

. Isolation, characterization and mode of antimicrobial action against Vibrio cholerae of methyl gallate isolated from Acacia farnesiana . J Appl Microbiol, 2013; 115:1307–1316.

35.

Shi

, Sun

, Liu

, Guo

, Sun

, Chen

, ZhangW, Wen

, Peng

. Xia X. Inhibition of Cronobacter sakazakii virulence factors by citral. Sci Rep, 2017; 7:43243.

36.

Shrout

, Tolker-Nielsen

, Givskov

, Parsek

. The contribution of cell-cell signaling and motility to bacterial biofilm formation. MRS Bullet, 2011; 36:367–373.

37.

Silva

, Zimmer

, Macedo

, Trentin

. Plant natural products targeting bacterial virulence factors. Chem Rev, 2016; 116:9162–9236.

38.

Singh

, Singh

, Chowdhury

, Singh

. Understanding the mechanism of bacterial biofilms resistance to antimicrobial agents. Open Microbiol J, 2017; 11:53.

39.

Tack

, Marder

, Griffin

, Cieslak

, Dunn

, Hurd

, Scallan

, Lathrop

, Muse

, Ryan

, Smith

. Tobin-D´Angelo M, Vugia DJ, Holt KG, Wolpert BJ, Tauxe R, Geissler AL. Preliminary incidence and trends of infections with pathogens transmitted commonly through food—Foodborne Diseases Active Surveillance Network, 10 U.S. sites, 2015–2018. Am J Transplant, 2019; 19:1859–1863.

40.

Taguri

, Tanaka

, Kouno

. Antimicrobial activity of 10 different plant polyphenols against bacteria causing food-borne disease. Biol Pharm Bull, 2004; 27:1965–1969.

41.

Teh

, Flint

, French

. Biofilm formation by Campylobacter jejuni in controlled mixed-microbial populations. Int J Food Microbiol, 2010; 143:118–124.

42.

Untergasser

, Nijveen

, Rao

, Bisseling

, Geurts

, Leunissen

. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Res, 2007; 35:W71–W74.

43.

Verstraeten

, Braeken

, Debkumari

, Fauvart

, Fransaer

, Vermant

, Michiels

. Living on a surface: Swarming and biofilm formation. Trends Microbiol, 2008; 16:496–506.

44.

Vidal

, Chamorro

, Girón

. Enterotoxigenic Escherichia coli . In: Torres

, ed. Escherichia coli in the Americas. Cham: Springer International Publishing, 2016, pp. 1–26.

45.

, Coulouris

, Zaretskaya

, Cutcutache

, Rozen

, Madden

. Primer-BLAST: A tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics, 2012; 13:134.

46.

Zaidi–Yahiaoui

, Zaidi

, Bessai

. A Influence of gallic and tannic acids on enzymatic activity and growth of Pectobacterium chrysanthemi (Dickeya chrysanthemi bv. chrysanthemi). Afr J Biotechnol, 2008; 7:482–486.