Inactivation of Escherichia coli O157:H7 and Salmonella on Fresh Herbs by Plant Essential Oils

Abstract

Consumer awareness of fresh herbs and its demand has increased in recent years due to health benefits and distinct aroma in prepared food. There are specific markets for local growers, especially for organically grown herbs. Shiga-toxigenic Escherichia coli and Salmonella spp. have been detected and associated with foodborne outbreaks from fresh herbs. Limited treatment options are available in the processing of fresh herbs to prevent the spread of foodborne pathogens. In this study, plant-based essential oils were evaluated on fresh herbs for their antimicrobial activities against Salmonella and E. coli O157:H7. Fresh herbs (basil, cilantro, dill, parsley, and tarragon) were inoculated with cocktails of either Salmonella or E. coli O157:H7 and then dip treated with chlorine (50 ppm), cinnamaldehyde (0.3 and 0.5%), and carvacrol (0.1 and 0.3%). Control herb samples were dipped in sterile water. Samples were collected on days 0, 2, 7, and 14 for enumeration of pathogens during 4°C storage. The bactericidal efficacy differed with herbs and antimicrobial concentrations. Treatments with 0.3% carvacrol or 0.5% cinnamaldehyde reduced E. coli O157:H7 and Salmonella by 5 log CFU/g (p > 0.05%) on cilantro and dill leaves from their initial inoculum level. Bactericidal efficacy of 0.1% carvacrol and 0.3% cinnamaldehyde was significant against Salmonella compared with chlorine on all herb leaves. E. coli O157:H7 and Salmonella populations were reduced further during storage of treated herbs. There was no visual difference in herbs treated with 0.3% cinnamaldehyde or 0.1% carvacrol from control samples. Results indicate that 0.3% cinnamaldehyde and 0.1% carvacrol are effective against E. coli O157:H7 and Salmonella, retain color attributes of fresh herbs, and, therefore, may be an alternative wash treatment for fresh herbs.

Introduction

Foodborne illnesses associated with consumption of fresh fruits and vegetables have increased in previous decades. Increase in consumption of fresh produce to meet healthy dietary habits and the year-around availability of fresh produce to consumers due to an increasingly globalized produce industry has resulted in an increased number of foodborne illnesses and outbreaks. The Centers for Disease Control and Prevention estimates that approximately 48 million people in the United States become ill each year from foodborne pathogens, causing 128,000 hospitalizations and 3000 deaths (Scallan et al., 2011).

Fresh herbs are used by consumers and in commercial establishments for their flavoring properties and health benefits. Certain herbs such as cilantro, basil, and parsley have been implicated in foodborne outbreaks (Pezzoli et al., 2008; Elviss et al., 2009). Consumption of basil contaminated with Salmonella Anatum and enterotoxigenic Escherichia coli resulted in >200 human illnesses in Denmark (Zweifel and Stephan, 2012). Contamination of cilantro with Salmonella Thompson was responsible for 76 illnesses in the United States (Campbell et al., 2001). Two foodborne outbreaks in 2011 caused by Shigella sonnei in Norway were associated with consumption of pesto made from fresh basil (Guzman-Herrador et al., 2013).

Previous investigations have revealed the prevalence of enteric pathogens in fresh herbs. A longitudinal survey of 1700 samples of different vegetables revealed Salmonella contamination in 12% of parsley and 11% of cilantro samples collected in Mexico (Quiroz-Santiago et al., 2009). Salmonella was also detected in fresh herbs sold at retail stores in Canada (Denis et al., 2016), the United Kingdom (Willis et al., 2015), and farmers' markets in the United States (Levy et al., 2015). Approximately 1.6% of 592 pre-packed basil and coriander leaves imported from Israel and Cyprus were contaminated with Salmonella and shiga-toxigenic E. coli (Delbeke et al., 2015). Microbial assessment of 774 fresh herb samples collected at retail markets in the United Kingdom revealed Salmonella contamination in curry leaves, basil, walleria, and cilantro (Willis et al., 2015). Enterotoxigenic E. coli were recovered in 4 of the 1200 parsley samples collected in the United States (Feng and Reddy, 2014).

Chlorinated water is commonly used to wash and decontaminate fresh produce; however, chlorine is not effective in completely removing bacteria. Further, byproducts of chlorine during produce wash may have human and environmental safety concerns (Yossa et al., 2012). Consumers' preferences for minimum chemicals in food systems have led research on environmentally friendly natural antimicrobials as produce wash for controlling pathogens in fresh produce (Yossa et al., 2013). Carvacrol and cinnamaldehyde, the major components of oregano and cinnamon oils, respectively, have shown antimicrobial activity against Salmonella enterica on organic leafy greens (Moore-Neibel et al., 2013; Rada et al., 2016). The purpose of this study was to investigate the antimicrobial effects of cinnamaldehyde and carvacrol against E. coli O157:H7 and Salmonella on fresh herbs.

Materials and Methods

Bacterial cultures

The cocktails of five nalidixic acid-resistant strains of E. coli O157:H7 (RM 4406, RM 4688, RM 1918, RM 4407, RM 5279) and five Salmonella spp. (Salmonella Braenderup, Salmonella Newport, Salmonella Negev, Salmonella Thompson, Salmonella Tennessee) from our culture collections were used for inoculation study. Strains were grown individually in 10 mL tryptic soy broth (Neogen, MI) at 37°C overnight. After incubation, cultures were centrifuged (7500g for 10 min, 10°C), and cell pellets were suspended in 0.1 M sterile phosphate buffer (pH 7.0, Neogen) to obtain OD₆₀₀ of 1 (∼8 log CFU/mL). Equal volumes (20 mL) of individual strains were mixed to prepare E. coli O157:H7 or Salmonella cocktails for inoculation study. Three independent trials were conducted by using E. coli O157:H7 or Salmonella cocktails for inoculation on fresh herbs.

Antimicrobials

Cinnamaldehyde and carvacrol (99% w/v; Sigma-Aldrich, St. Louis, MO) were used as natural antimicrobials for comparison with chlorine (Clorox, Oakland, CA) in this study. Cinnamaldehyde (0.3%, 0.5%) and carvacrol (0.1%, 0.3%) were prepared in sterile 0.1% Tween-20 (Fisher Scientific, NJ); a 50-ppm chlorine was prepared fresh in sterile distilled water for use on fresh herbs.

Fresh herbs inoculation and treatment

Fresh organic herbs (basil, parsley, cilantro, tarragon, and dill) were purchased at a retail grocery store and kept at 4°C overnight before the onset of the experiment. The herbs were washed thoroughly for 2 min with running tap water, and they were then air-dried under UV light for 30 min. Fresh herbs (30 g) were immersed for 2 min in 300-mL sterile water containing a cocktail of Salmonella or E. coli O157:H7 (6 log CFU/mL), and they were then air-dried for 30 min. Inoculated herb leaves were dipped in treatment solutions (200 mL) for 2 min with manual agitation followed by brief washing (30 s) in D/E neutralizing broth (Neogen) to remove residual antimicrobials. The excess water in herb leaves was removed by spinning in a salad-spinner for 1 min. Treated leaves were placed in sterile Whirl-Pak filter bags (Nasco Whirl-Pak, WI). The bags were sealed and stored at 4°C for 14 days. Samples washed with sterile water served as control.

Microbial analysis of treated herbs

Fresh herbs were analyzed during the incubation period at 0 (after wash treatment), 2, 7, and 14 days for populations of E. coli O157:H7 and Salmonella. Herb leaves (4 g) were pummeled in 36-mL sterile peptone water for 2 min in a stomacher (Interscience, France), and serially diluted samples were spiral plated (Whitley Scientific, England) on Xylose-lysine-tergitol-4 agar (Neogen) for Salmonella; and on Sorbitol MacConkey agar (Neogen) supplemented with 0.05 mg/L of cefixime (Sigma-Aldrich, MO), 2.5 mg/L of potassium tellurite (Sigma-Aldrich), and 50 ppm nalidixic acid (Sigma-Aldrich) for E. coli O157:H7. Colonies were counted after 24 h of incubation at 37°C by using Protocol colony counter (Microbiology International, Inc., MD).

Color measurement of treated herbs

Color values (L, a, b) of herbs treated with antimicrobials essential oils were measured on days 0, 2, 7, and 10 by using a CR-20 chroma meter (Minolta, Inc. Japan). Illuminant D65 and 10° observer angle were used. At least twenty measurements were made on each of control, chlorine-treated, and essential oil-treated herb leaves.

Statistical analysis

The experiment was repeated three times for each treatment and storage period. Reductions in E. coli O157:H7 and Salmonella (log CFU/g) from initial populations were compared among treatment-time combinations by a three-way ANOVA using “proc-mixed” procedure (SAS 9.4, Cary, NC). Color data were analyzed similarly by the proc mixed procedure. The level of statistical significance was set at p < 0.05 in all cases.

Results

Effect of antimicrobials on E. coli O157:H7 populations on fresh herbs

Initial E. coli O157:H7 populations on fresh herbs were 5.03–5.44 log CFU/g. Washing these fresh herbs with sterile distilled water reduced E. coli O157:H7 by 0.6–1.5 log CFU/g (Fig. 1A–E). Chlorine treatment resulted in 0.7–2.0 log CFU/g reduction in E. coli on fresh herbs. E. coli O157:H7 recovered from basil treated with 0.3% cinnamaldehyde (4.17 log CFU/g), 0.5% cinnamaldehyde (3.46 log CFU/g), and 0.1% carvacrol (3.92 log CFU/g) were not significantly different from control basil (4.75 log CFU/g) (Fig. 1A). Basil washed with 0.3% carvacrol significantly reduced E. coli O157:H7 to below the level of detection (<0.57 log CFU/g). Washing cilantro with 0.5% cinnamaldehyde and 0.3% carvacrol significantly reduced E. coli O157:H7 to 0.85 log CFU/g and <0.57 log CFU/g, respectively (Fig. 1D). Treatment with 0.5% cinnamaldehyde or 0.3% carvacrol significantly reduced E. coli O157:H7 in other fresh herbs also when compared with chlorine treatment. E. coli O157:H7 populations in 0.5% cinnamaldehyde- and 0.3% carvacrol-treated dill (<0.57 log CFU/g), parsley (1.65 and 0.67 log CFU/g), and tarragon (2.36 log and <0.57 log CFU/g) were significantly lower than E. coli O157:H7 recovered from chlorine-treated dill, parsley, and tarragon (3.06, 3.73, and 4.44 log CFU/g, respectively).

FIG. 1.

Survival of Escherichia coli O157:H7 on fresh herbs (A) basil, (B) parsley, (C) tarragon, (D) cilantro, and (E) dill after a wash with antimicrobials and during storage at 4°C. Values plotted on each sampling point are an average of three replicates. Error bar represents standard deviations from the means. (ART; Cinn, cinnamaldehyde; Carv, carvacrol). Treatment columns marked with * are significantly different from control within the storage period.

Lower recovery of E. coli O157:H7 was reported from all treated herbs except parsley after 2 days of refrigerated storage (Fig. 1A–E). E. coli O157:H7 recovered on these herbs treated with 0.3% carvacrol or 0.5% cinnamaldehyde were significantly lower than E. coli recovered from control- or chlorine-treated herbs. At 7 days, E. coli populations were below detection level (<0.57 CFU/g) in all fresh herbs when washed with 0.5% cinnamaldehyde or 0.3% carvacrol. A wash treatment with cinnamaldehyde or 0.3% carvacrol reduced E. coli O157:H7 to a non-detectable level in cilantro, dill, parsley, and tarragon at 14 days; however, it was recovered in chlorine-treated dill (1.43 log CFU/g), parsley (0.57 log CFU/g), and tarragon (1.49 log CFU/g).

Effect of antimicrobials on Salmonella populations on fresh herbs

Initial Salmonella populations on fresh herbs were 5.03–5.46 log CFU/g. Washing with sterile distilled water removed 0.4–0.9 log CFU/g Salmonella on fresh herbs (Fig. 2A-E). Washing fresh herbs with chlorine removed 0.9–2.5 log CFU/g Salmonella on fresh herbs. Fresh herb treatment with carvacrol (0.1% and 0.3%) and cinnamaldehyde (0.3% and 0.5%) significantly reduced Salmonella populations on all herbs as compared with control. Further, washing herbs with 0.3% carvacrol reduced Salmonella by 5 log CFU/g to an undetectable level (<0.57 log CFU/g) in all herbs except cilantro. Treatment with 0.5% cinnamaldehyde reduced Salmonella by ca. 4 log CFU/g in all herbs, resulting in a significant reduction compared with chlorine treatment. The antimicrobial effect of these antimicrobials at lower concentrations (0.1% carvacrol and 0.3% cinnamaldehyde) was superior to chlorine with all herbs and was significantly different on cilantro, parsley, and tarragon leaves.

FIG. 2.

Survival of Salmonella on fresh herbs (A) basil, (B) parsley, (C) tarragon, (D) cilantro, and (E) dill after a wash with antimicrobials and during storage at 4°C. Values plotted on each sampling point are an average of three replicates. Error bar represents standard deviations from the means. (ART; Cinn, cinnamaldehyde; Carv, carvacrol). Treatment columns marked with * are significantly different from control within the storage period.

After 2 days at 4°C, Salmonella were further reduced by 0.2–0.5 log CFU/g in herbs washed with water (Fig. 2A–E). Treatment with carvacrol (0.1%, 0.3%) and cinnamaldehyde (0.3%, 0.5%) resulted in significantly lower recovery of Salmonella in basil, cilantro, parsley, and tarragon compared with the control. Salmonella were undetectable in all fresh herbs when washed with 0.3% carvacrol. Lower recovery of Salmonella was observed with further storage at 7 and 14 days. At 7 days, Salmonella populations in cinnamaldehyde- or carvacrol-treated cilantro, dill, parsley, and tarragon were significantly lower compared with Salmonella in corresponding control- or chlorine-treated herbs. After 14 days of storage, Salmonella were recovered at ≤0.57 log CFU/g in these herbs when treated with carvacrol or cinnamaldehyde; whereas Salmonella populations were significantly higher in chlorine-treated basil (2.78 log CFU/g), cilantro (2.70 log CFU/g), dill (1.13 log CFU/g), and parsley (1.85 log CFU/g).

Effect of antimicrobials on color of fresh herbs

The color coordinate value l (lightness) of fresh herbs treated with 0.1% carvacrol, 0.3% cinnamaldehyde, or 0.5% cinnamaldehyde was not significantly different from corresponding herbs treated with water (Table 1). However, treatment with 0.3% carvacrol significantly affected lightness values of basil, cilantro, dill, and tarragon leaves. The difference in lightness was more evident during storage. The lightness of cilantro, parsley, and tarragon treated with 0.1% carvacrol was not significantly different from that of corresponding control herbs after 7 and 10 days of storage. The effects of 0.3% carvacrol and 0.5% cinnamaldehyde on lightness of fresh herbs were more pronounced than at lower concentrations of these antimicrobials. The initial greenness (a value) of cilantro, dill, and parsley leaves treated with these antimicrobials was not different from their control leaves; however, a 0.3% carvacrol wash of basil and tarragon significantly lowered their greenness values. The greenness of control herb leaves remained consistent during storage, whereas it was reduced significantly on basil, cilantro, and tarragon when treated with antimicrobials after 7 days. The blueness (b values) was significantly affected when basil and tarragon were treated with 0.3% carvacrol or dill treated with 0.5% cinnamaldehyde. Lower b values (p < 0.05) of basil and dill treated with antimicrobials were observed at 7 and 10 days of storage irrespective of treatment and concentrations. Cilantro-, parsley-, and tarragon-treated with 0.3% carvacrol had significantly lower b values than control leaves after 10 days.

Table 1.

Color Measurements (L, a, b values) of Fresh Herbs Treated with Natural Antimicrobials During Storage at 4°C

		Color parameters
		Day 0			Day 2			Day 7			Day 10
Herb	Treatment	Day 0	a	b	L	a	b	L	a	b	L	a	b
Basil	Water	41 ± 1 ax	−8 ± 1 ax	23 ± 3 ax	40 ± 2 ax	−7 ± 1 ax	21 ± 2 ax	38 ± 2 ax	−7 ± 1 ax	21 ± 2 ax	37 ± 2 ax	−5 ± 2 ay	21 ± 2 ax
	Carv-0.1	38 ± 2 ax	−8 ± 1 ax	19 ± 1 abx	29 ± 2 cy	−4 ± 3 by	13 ± 3 by	31 ± 5 by	−1 ± 1 cz	9 ± 3 cy	28 ± 2 by	0 ± 1 cz	10 ± 1 cy
	Carv-0.3	31 ± 3 bx	0 ± 1 bx	9 ± 5 bx	28 ± 1 cx	0 ± 1 cx	9 ± 3 bx	27 ± 2 bx	0 ± 0 cx	9 ± 2 cx	29 ± 1 bx	1 ± 0 cx	8 ± 1 cx
	Cin-0.3	41 ± 2 ax	−10 ± 1 ax	25 ± 2 ax	35 ± 2 by	−8 ± 1 ax	19 ± 2 ax	31 ± 3 by	−3 ± 3 by	13 ± 4 bcy	31 ± 2 by	−2 ± 3 by	12 ± 4 bcy
	Cin-0.5	42 ± 1 ax	−9 ± 0 ax	23 ± 1 ax	35 ± 2 by	−7 ± 1 ax	19 ± 1 axy	30 ± 3 bz	−4 ± 3 by	16 ± 4 by	31 ± 4 bz	−3 ± 4 by	15 ± 5 by
Cilantro	Water	41 ± 1 ax	−9 ± 0 ax	23 ± 2 ax	39 ± 1 ax	−8 ± 0 ax	19 ± 2 ax	39 ± 2 ax	−9 ± 0 ax	25 ± 2 ax	36 ± 2 ax	−8 ± 0 ax	24 ± 3 ax
	Carv-0.1	40 ± 2 ax	−9 ± 0 ax	21 ± 2 ax	38 ± 1 ax	−8 ± 0 ax	19 ± 2 ax	39 ± 2 ax	−7 ± 0 ax	20 ± 3 ax	36 ± 2 ax	−7 ± 0 ax	21 ± 4 ax
	Carv-0.3	33 ± 3 bx	−8 ± 1 ax	20 ± 1 ax	31 ± 2 bxy	−6 ± 0 bx	17 ± 0 axy	29 ± 1 bxy	−1 ± 0 cy	14 ± 1 by	25 ± 2 by	0 ± 0 cy	12 ± 1 by
	Cin-0.3	37 ± 1 abx	−7 ± 0 abx	19 ± 2 ax	36 ± 2 ax	−7 ± 1 ax	17 ± 2 axy	31 ± 2 by	−4 ± 1 by	14 ± 2 by	29 ± 2 by	−3 ± 1 by	13 ± 1 by
	Cin-0.5	35 ± 1 bx	−6 ± 1 bx	19 ± 2 ax	34 ± 3 abx	−5 ± 1 bx	18 ± 2 ax	29 ± 1 by	−2 ± 0 cy	14 ± 1 by	29 ± 2 by	−1 ± 0 cy	14 ± 1 by
Dill	Water	35 ± 1 axy	−8 ± 0 ax	19 ± 1 ax	36 ± 2 ax	−7 ± 0 ax	19 ± 2 ax	33 ± 1 axy	−7 ± 1 ax	20 ± 1 ax	32 ± 1 ay	−7 ± 0 ax	20 ± 1 ax
	Carv-0.1	34 ± 1 ax	−8 ± 0 ax	20 ± 1 ax	31 ± 2 bx	−7 ± 1 ax	19 ± 2 ax	25 ± 1 by	−4 ± 0 by	13 ± 1 by	25 ± 2 bcy	−3 ± 1 by	12 ± 2 by
	Carv-0.3	24 ± 1 bx	−7 ± 0 ax	17 ± 1 abx	25 ± 1 cx	−6 ± 1 ax	17 ± 1 ax	25 ± 1 bx	−2 ± 0 cy	15 ± 1 bxy	22 ± 1 cx	−2 ± 0 by	12 ± 0 by
	Cin-0.3	32 ± 1 ax	−8 ± 1 ax	22 ± 1 ax	31 ± 1 bx	−7 ± 1 ax	19 ± 1 axy	26 ± 2 by	−5 ± 1 abx	16 ± 1 by	27 ± 2 by	−6 ± 1 ax	16 ± 2 aby
	Cin-0.5	32 ± 1 ax	−5 ± 1 bx	15 ± 1 bx	29 ± 1 bxy	−4 ± 0 bx	17 ± 1 ax	26 ± 2 by	−3 ± 0 by	15 ± 1 bx	26 ± 1 by	−1 ± 0 ay	14 ± 0 bx
Parsley	Water	36 ± 1 ax	−7 ± 1 ax	17 ± 3 ay	37 ± 1 ax	−8 ± 0 ax	19 ± 2 axy	36 ± 2 ax	−8 ± 1 ax	22 ± 2 axy	37 ± 2 ax	−9 ± 1 ax	25 ± 6 ax
	Carv-0.1	34 ± 1 ax	−8 ± 0 ax	17 ± 3 ay	33 ± 2 abx	−8 ± 0 ax	18 ± 2 ay	35 ± 2 ax	−8 ± 0 ax	21 ± 4 ax	38 ± 1 ax	−9 ± 0 ax	23 ± 3 ax
	Carv-0.3	34 ± 2 ax	−8 ± 1 ax	18 ± 1 ax	34 ± 2 abx	−6 ± 1 ax	18 ± 1 ax	32 ± 4 ax	−3 ± 1 by	17 ± 1 bx	33 ± 2 ax	−1 ± 0 by	18 ± 1 bx
	Cin-0.3	34 ± 1 ax	−8 ± 0 ax	17 ± 1 abx	34 ± 1 abx	−8 ± 0 ax	18 ± 1 ax	34 ± 1 ax	−9 ± 0 ax	20 ± 4 ax	36 ± 1 ax	−9 ± 0 ax	21 ± 2 ax
	Cin-0.5	34 ± 1 ax	−7 ± 0 ax	16 ± 2 ax	32 ± 2 bx	−7 ± 1 ax	16 ± 1 ax	33 ± 1 ax	−9 ± 0 ax	18 ± 2 abx	37 ± 1 ax	−9 ± 0 ax	20 ± 2 abx
Tarragon	Water	40 ± 1 ax	−8 ± 0 ax	23 ± 4 ax	37 ± 1 ax	−7 ± 1 ax	21 ± 2 ax	37 ± 1 ax	−8 ± 0 ax	22 ± 3 ax	38 ± 1 ax	−8 ± 0 ax	23 ± 4 ax
	Carv-0.1	38 ± 2 abx	−8 ± 0 ax	20 ± 2 ax	38 ± 1 ax	−8 ± 1 ax	21 ± 1 ax	37 ± 1 ax	−7 ± 1 ax	20 ± 0 ax	38 ± 2 ax	−7 ± 2 ax	21 ± 1 ax
	Carv-0.3	35 ± 1 bx	−4 ± 1 bx	14 ± 2 bx	35 ± 1 ax	−2 ± 1 cy	14 ± 1 bx	29 ± 2 by	−1 ± 1 by	14 ± 1 bx	34 ± 2 abx	1 ± 1 by	13 ± 2 bx
	Cin-0.3	39 ± 1 abx	−8 ± 0 ax	21 ± 1 ax	34 ± 1 axy	−5 ± 1 by	14 ± 1 by	31 ± 1 by	−2 ± 0 bz	12 ± 1 by	31 ± 1 by	−1 ± 1 bz	12 ± 2 by
	Cin-0.5	36 ± 1 abx	−6 ± 1 abx	15 ± 2 bx	34 ± 2 axy	−3 ± 1 cy	14 ± 1 bx	31 ± 1 by	−1 ± 1 byz	12 ± 2 bx	32 ± 1 bxy	0 ± 1 bz	11 ± 2 bx

Values are mean–SD. Each experiment was replicated 10 times.

Values in the same column within herb not followed by the same letter (abc) are significantly different, values in the same row for the same parameter not followed by the same letters (xy) are significantly different (p < 0.05).

Carv, carvacrol; Cin, cinnamaldehyde.

Discussion

A reduction in E. coli O157:H7 and Salmonella up to 1.5 and 0.9 log CFU/g, respectively, was observed when herbs were washed with water. Our results agree with other studies that reported 0.3–0.7 log CFU/g reduction in E. coli O157:H7 (Bhargava et al., 2015) and Salmonella (Rada et al., 2016) when fresh produce was washed with water. The chlorine wash (200 ppm) reduced only ∼1.5 log CFU/g of E. coli O157:H7 on cilantro, and the reduction was not significantly different from control water wash (Foley et al., 2004). Chlorine (200 ppm) wash of parsley resulted in >1.5 log CFU reduction of E. coli O157:H7, Salmonella, and Listeria monocytogenes in studies conducted by Orue et al. (2013). We used 50 ppm chlorine based on our previous studies (Yossa et al., 2012) and observed 0.7–2.0 log CFU/g reduction in E. coli O157:H7 when fresh herbs were washed with 50 ppm chlorine.

Other studies examined the efficacy of several essential oils against Salmonella on organic lettuce and spinach. The results showed that essential oils were significantly effective in killing Salmonella on iceberg and romaine lettuce, and fresh spinach (Moore-Neibel et al., 2011–2013). Oregano extract reduced E. coli and Salmonella by 1.5–2.0 log CFU/g on cilantro and parsley (Orue et al., 2013). A combination treatment of cinnamaldehyde with olive extract resulted in a significant reduction in Salmonella populations on organic fresh produce (Rada et al., 2016). Similarly, a combination of cinnamaldehyde with acetic acid was more effective in killing Salmonella and E. coli O157:H7 on fresh spinach leaves than using cinnamaldehyde alone (Yossa, et al., 2012). The concentrations of cinnamaldehyde (0.3% and 0.5%) and carvacrol (0.1% and 0.3%) used in this study were based on minimum inhibitory concentration results (Andrews, 2001; data not shown). In this study, the bactericidal efficacy of cinnamaldehyde increased with an increase in cinnamaldehyde concentration as observed by other researchers (Yossa et al., 2012) and it varied with the type of fresh herbs. Differences in leaf structure and surface hydrophobicity of these herbs could have resulted in differences in the recovery of E. coli O157:H7 and Salmonella after a wash with chlorine or natural antimicrobials.

Carvacrol has been previously evaluated to kill enteric pathogens on fresh produce. It was superior to cinnamaldehyde and reduced E. coli O157:H7 on organic leafy greens, by 3 log CFU/g at 0.1% concentration and to an undetectable level at 0.3% level (Denton et al., 2015). Siroli et al. (2015) observed a significant reduction in spoilage bacteria of lamb's lettuce with natural antimicrobials; however, the carvacrol treatment negatively affected the quality and sensory properties of lettuce. In our study, treatment with 0.3% carvacrol reduced E. coli O157:H7 and Salmonella by 5 log CFU/g; however, it affected the visual quality of herbs. Herb wash with 0.1% carvacrol was superior to chlorine in reducing pathogens and it did not affect the quality of certain herbs.

We used dip-inoculation of pathogens on fresh herbs as dipping results in higher population of attached pathogens on produce surface compared with spray- or spot-inoculation (Lang et al., 2004). In general, fresh herbs are likely to be dipped in a dump tank for less than 3 min (Duffy et al., 2005). Cinnamon oil treatment of organic fresh produce resulted in a greater reduction in Salmonella when treatment time was increased from 1 to 2 min (Todd et al., 2013). In our study, we dipped herb leaves for 2 min to minimize potential internalization of these pathogens as Salmonella internalization in parsley was increased with dip time from 3 to 15 min (Duffy et al., 2005). We also used Tween-20 to increase the solubility of essential oil, and helping them to penetrate bacterial cell walls and membrane. Previous research in our laboratory has demonstrated additional 1 log CFU killing of E. coli O157:H7 on spinach leaves when cinnamaldehyde was incorporated with Tween-20 (Yossa et al., 2013).

A reduction in pathogen populations during refrigerated storage of fresh produce has been reported. Shigella sonnei were reduced by 2.5 log CFU/g on inoculated parsley during 14 days of storage at 4°C (Wu et al., 2000), whereas only 1 log CFU/g reduction in Salmonella and E. coli O157:H7 was observed on parsley leaves stored at 4°C for 16 days (Hsu et al., 2006). A moderate reduction (1 log CFU/g) of Salmonella on fresh basil was also observed during 9-days refrigerated storage (Eckner et al., 2015). We observed more than 1.5 log CFU/g reduction in E. coli O157:H7 and Salmonella populations during storage of these herbs. Differences in pathogen reduction during storage could be attributed to inoculum level, strain variation, herb cultivar, and storage conditions employed in these studies.

The antimicrobial activity of essential oils could be attributed to more than one specific mechanism due to their complex chemical structures and volatile phenolic compounds, monoterpenes, and alcohols. Essential oils may cause membrane damage, resulting in cellular leakage; changes in intracellular pH, membrane potential, and adenosine tryphosphate (ATP) synthesis (Sanchez et al., 2010; Hyun et al., 2015). We observed greater antimicrobial activity of carvacrol against pathogens on fresh produce compared with cinnamaldehyde, which could be due to their different modes of bactericidal action. Carvacrol dissolved the phospholipid layer of cells, causing membrane destabilization and increased membrane permeability, resulting in ATP release (Ultee et al., 2002; Friedman, 2006). Consequently, cinnamaldehyde does not damage the cell membrane but interferes with amino acid decarboxylases activity in bacterial cells (Gill and Holley, 2004). The sensory impact of natural antimicrobials in produce is important for their application as a produce wash. The sensory quality of bell pepper and lettuce washed with thyme was unacceptable (Uyttendaele et al., 2004; Gutierrez et al., 2008). Lettuce treated with oregano and thyme was rejected after 7 days due to poor sensory qualities (Gutierrez et al., 2009). In general, fresh herbs or foods associated with spices or seasoning might be least affected by the incorporation of natural antimicrobials; however, we observed poor sensory quality in fresh herbs that were treated at higher concentrations of cinnamaldehyde or carvacrol and stored for more than 7 days.

Conclusion

Fresh herbs contaminated with enteric pathogens pose serious health concern as chlorine is marginally effective against pathogens on fresh herbs. Cinnamaldehyde and carvacrol are superior to chlorine in reducing enteric pathogens on herb leaves. These antimicrobials at lower concentrations do not affect the color of fresh herbs. The effect of cinnamaldehyde and carvacrol on other sensory qualities requires further investigation to determine the overall acceptability of these antimicrobial-treated herbs.

Footnotes

Acknowledgments

This work was supported by the USDA-ARS project plan “Characterization and Mitigation of Bacterial Pathogens in the Fresh Produce Production and Processing Continuum”.

Disclosure Statement

No competing financial interest exist.

References

Andrews

. Determination of minimum inhibitory concentrations. J Antimicro Chemother, 2001; 48:5–16.

Bhargava

, Conti

, da Rocha

SRP

, Zhang

. Application of an oregano oil nanoemulsion to the control of foodborne bacteria on fresh produce. Food Microbiol, 2015; 47:69–73.

Campbell

, Mohle-Boetani

, Reporter

, Abbott

, Farrar

, Brandl

, Mandrell

, Werner

. An outbreak of Salmonella serotype Thompson associated with fresh cilantro. J Infect Dis, 2001; 183:984–987.

Delbeke

, Ceuppens

, Jacxsens

, Uyttendaele

. Microbiological analysis of pre-packed sweet basil and coriander leaves for the presence of Salmonella spp. and Shiga toxin-producing E. coli . Int J Food Microbiol, 2015; 208:11–18.

Denis

, Zhang

, Leroux

, Trudel

, Bietlot

. Prevalence and trends of bacterial contamination in fresh fruits and vegetables sold at retail in Canada. Food Control, 2016; 67:225–234.

Denton

, Ravishankar

, Friedman

, Jaroni

. Efficacy of plant-derived compounds against Escherichia coli O157:H7 during flume-washing and storage of organic leafy greens. J Food Process Preserv, 2015; 39:2728–2737.

Duffy

, Cisneros-Zevallos

, Castillo

, Pillai

, Ricke

, Acuff

. Survival of Salmonella transformed to express green fluorescent protein on Italian parsley as affected by processing and storage. J Food Prot, 2005; 68:687–695.

Eckner

, Hogasen

, Begum

, Olkand

, Kofitsyo

, Johannessen

. Survival of Salmonella on basil plants and in pesto. J Food Prot, 2015; 78:402–406.

Elviss

, Little

, Hucklesby

, Sagoo

, Surman-Lee

, Pinna

, Threlfall

. Microbiological study of fresh herbs from retail premises uncovers an international outbreak of salmonellosis. Int J Food Microbiol, 2009; 134:83–88.

10.

Feng

PCH

, Reddy

. Prevalence and diversity of enterotoxigenic Escherichia coli strains in fresh produce. J Food Prot, 2014; 77:820–823.

11.

Foley

, Euper

, Caporaso

, Prakash

. Irradiation and chlorination effectively reduce Escherichia coli O157:H7 inoculated on cilantro without negatively affecting quality. J Food Prot, 2004; 67:2092–2098.

12.

Friedman

. Antibiotic activities of plant compounds against non-resistant and antibiotic-resistant foodborne pathogens. In: Juneja

, Cherry

, and Tumik

, eds. Advances in Microbial Food Safety. Washington, DC: American Chemical Society, 2006, pp. 167–183.

13.

Gill

, Holley

. Mechanisms of bactericidal action of cinnamaldehyde against Listeria monocytogenes and of eugenol against L. monocytogenes and Lactobacillus sakei . Appl Environ Microbiol, 2004; 70:5750–5755.

14.

Gutierrez

, Bourke

, Lonchamp

, Barry-Ryan

. Impact of plant essential oils on microbiological, organoleptic and quality markers of minimally processed vegetables. Innov Food Sci Emerg, 2009; 10:195–202.

15.

Gutierrez

, Rodriguez

, Barry-Ryan

, Bourke

. Efficacy of plant essential oils against foodborne pathogens and spoilage bacteria associated with ready-to-eat vegetables: Antimicribal and sensory screening. J Food Prot, 2008; 71:1846–1854.

16.

Guzman-Herrador

, Nilsen

, Cudjoe

, Jensvoll

, Kvamme

, Lindegard

, Lindstedt

, Nygard

, Severinsen

, Werner-Johanse

, Wester

, Wiklund

, Vold

. A Shigella sonnei outbreak traced to imported basil- the importance of good typing tools and produce traceability systems, Norway 2011. Euro Surveill, 2013; 18:1–7.

17.

Hsu

, Simonne

, Jitareerat

. Fates of seeded Escherichia coli O157:H7 and Salmonella on selected fresh culinary herbs during refrigerated storage. J Food Prot, 2006; 69:1997–2001.

18.

Hyun

, Bae

, Song

, Yoon

, Lee

. Antibacterial effect of various essential oils against pathogens and spoilage microorganisms in fresh produce. J Food Saf, 2015; 35:206–219.

19.

Lang

, Harris

, Beuchat

. Survival and recovery of Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis, and treatment with chlorinated water. J Food Prot, 2004; 67:1092–1103.

20.

Levy

, Beck

, Kossik

, Patti

, Meschke

, Calicchia

, Hellberg

. Microbial safety and quality of fresh herbs from Los Angeles, Orange county and Seattle farmers' markets. J Sci Food Agric, 2015; 95:2641–2645.

21.

Moore-Neibel

, Gerber

, Patel

, Friedman

, Jaroni

, Ravishankar

. Antimicrobial activity of oregano oil against antibiotic-resistant Salmonella enterica on organic leafy greens at varying exposure times and storage temperatures. Food Microbiol, 2013; 34:123–129.

22.

Moore-Neibel

, Gerber

, Patel

, Friedman

, Ravishankar

. Antimicrobial activity of Lemongrass oil against Salmonella enterica on organic leafy greens. J Appl Microbiol, 2012; 112:485–492.

23.

Moore-Neibel

, Patel

, Jaroni

, Friedman

, Ravishankar

. Antimicrobial activity of Apple, Hibiscus, Olive, and hydrogen peroxide formulations against Salmonella enterica on organic leafy greens. J Food Prot, 2011; 74:1676–1683.

24.

Orue

, Garcia

, Feng

, Heredia

. Decontamination of Salmonella, Shigella, and Escherichia coli O157:H7 from leafy green vegetables using edible plant extracts. J Food Sci, 2013; 78:M290–M296.

25.

Pezzoli

, Elson

, Little

, Yip

, Fisher

, Yishai

. Packed with Salmonella-investigation of an international outbreak of Salmonella Senftenberg infection linked to contamination of prepacked basil in 2007. Foodborne Pathog Dis, 2008; 5:661–668.

26.

Quiroz-Santiago

, Rodas-Surez

, Vazques

, Fernandez

, Quinones-Ramirez

, Vazquez-Salinas

. Prevalence of Salmonella in vegetables from Mexico. J Food Prot, 2009; 72:1279–1282.

27.

Rada

, Todd-Searle

, Friedman

, Patel

, Jaroni

, Ravishankar

. Combining essential oils and olive extract for control of multi-drug resistant Salmonella enterica on organic leafy greens. SDRP J Food Sci Technol, 2016; 1:1–9.

28.

Sanchez

, Garcia

, Geredia

. Extracts of edible and medicinal plants damage membranes of Vibrio cholerae . Appl Environ Microbiol, 2010; 76:6888–6894.

29.

Scallan

, Hoekstra

, Angulo

, Tauxe

, Widdowson

, Roy

, Jones

, Griffin

. Foodborne illness acquired in the United States-major pathogens. Emerg Infect Dis, 2011; 17:7–15.

30.

Siroli

, Patrignani

, Serrazanetti

, Tappi

, Rocculi

, Gardini

, Lanciotti

. Natural antimicrobials to prolong the shelf-life of minimally processed lamb's lettuce. Postharvest Biol Technol, 2015; 103:35–44.

31.

Todd

, Friedman

, Patel

, Jaroni

, Ravishankar

. The antimicrobial effects of cinnamon leaf oil against multi-drug resistant Salmonella Newport on organic leafy greens. Int J Food Microbiol, 2013; 166:193–199.

32.

Ultee

, Bennik

, Moezelaar

. The phenolic hydroxyl grup of carvacrol is essential for action against the food-borne pathogen Bacillus cereus . Appl Environ Microbiol, 2002; 68:1561–1568.

33.

Uyttendaele

, Neyts

, Vanderswalmen

, Notebaert

, Debevere

. Control of Aromonas on minimally processed vegetables by decontamination with lactic acid, chlorinated water or thyme essential oil solution. Int J Food Microbiol, 2004; 90:263–271.

34.

Willis

, Sadler-Reeves

, Elviss

, Aird

, Fox

, Kaye

, de Pinna

, Lane

, McLauchlin

. An assessment of the microbiological safety of fresh whole-leaf herbs from retail premises in the United Kingdom with a focus on Salmonella spp. J Appl Microbiol, 2015; 119:827–833.

35.

, Doyle

, Beuchat

, Wells

, Mintz

, Swaminathan

. Fate of Shigella sonnei on parsley and methods of disinfection. J Food Prot, 2000; 63:568–572.

36.

Yossa

, Patel

, Millner

, Lo

. Essential oils reduce Escherichia coli O157:H7 and Salmonella on spinach leaves. J Food Prot, 2012; 75:488–496.

37.

Yossa

, Patel

, Millner

, Ravishankar

, Lo

. Antimicrobial activity of plant essential oils against Escherichia coli O157:H7 and Salmonella on lettuce. Foodborne Pathog Dis, 2013; 10:88–96.

38.

Zweifel

, Stephan

. Spices and herbs as source of Salmonella-related foodborne diseases. Food Res Int, 2012; 45:765–769.