Extended-Spectrum β-Lactamase,AmpC,and MBL-Producing Gram-Negative Bacteria on Fresh Vegetables and Ready-to-Eat Salads Sold in Local Markets

Abstract

We investigated the occurrence of extended-spectrum β-lactamase (ESBL), AmpC, and carbapenemase-producing Gram-negative bacteria isolated from 160 samples of fresh vegetables (n = 80) and ready-to-eat (RTE) prepacked salads (n = 80). Phenotypic and genotypic analyses were carried out on the isolates in terms of the species present and relative resistance. Resistance to β-lactam antibiotics was found in only 44 (24 from fresh vegetables and 20 from RTE salads) of a total of 312 Gram-negative strains (14.1%). The prevalence of ESBL-producing strains from fresh vegetables was 83.3% (20/24) and 16.7% (4/24) for AmpC. Among the 20 bacterial isolates from RTE salads, 80% (16/20) were identified as ESBL-producing strains and the remaining 20% (4/20) as MBL-producing strains. PCR and sequencing confirmed the presence of bla_SHV-12, bla_CTX-M-1, bla_CTX-M-15, bla_RHAN_-1, bla_ACC-1, bla_DHA-1, bla_VIM-1, and bla_IMP-1. Seven different replicons were identified, where IncHI1, FIA, and I1 were the most representative types; when compared with the Inc types, isolates from fresh vegetables and RTE salads were similar. The location of genes on a conjugative plasmid was confirmed by positive results obtained with conjugation assays.

Our study has demonstrated the occurrence and distribution of ESBL/AmpC and MBL strains in fresh vegetables and RTE salads in Italy and possible public health risks associated with consumption of these fresh products.

Introduction

Antibiotic use in human medicine, agriculture, aquaculture, and livestock industries has led to the diffusion in the environment of antibiotic resistance genes (ARGs).¹ Extensive use of antibiotics in plant agriculture contributes to the selection of resistant bacteria.² Applying manure from animal farming to agricultural fields could also influence spreading of antibiotic-resistant bacteria to vegetables. Edible plants have often been debated as vectors for transfer of pathogenic- and/or antibiotic-resistant bacteria. The transfer of pathogens to humans can be demonstrated by associating the source and the microorganism responsible for the outbreaks (e.g., Escherichia coli O157, Salmonella spp.).^3–6

The microbiological hazard behind the consumption of ready-to-eat (RTE) food has emerged in recent years and several conspicuous international outbreaks of pathological microorganisms have highlighted the fact that fresh vegetables are potential vehicles of microorganisms responsible for foodborne diseases.

A handful of studies have underlined the presence of extended-spectrum β-lactamase (ESBL)-producing E. coli in fruit and vegetables at retail sale.⁷ More specifically, E. coli with CTX-M-14 or CTX-M-1 ESBLs were isolated from parsley and tomatoes, respectively, collected from markets in Tunisia.⁸ However, in a larger study of 1,216 vegetables on retail sale in the Netherlands, ESBL-producing E. coli was isolated only from a single batch of blanched celery.⁹

Fresh produce frequently harbors nonpathogenic environmental microorganisms.¹⁰ During growth and harvesting, vegetables and fruits can also become contaminated with pathogenic and commensal bacteria from animals and humans.

This contamination can occur in the field by direct contact with the soil and by using manure and wastewater as biofertilizers.^9,11 In addition, microorganisms of crops are subjected to a high selection pressure caused by exposure to antibiotic residues indirectly coming from manure and wastewater or directly from phytopharmaceutical agents.^12–15 It has been proved that antimicrobial agents different from antibiotics have the ability to promote a co-selection process, indirectly selecting for antibiotic resistance. Heavy metal contaminations are widely spread, both in agriculture¹⁶ and in aquaculture, contributing to that environmental burden.¹⁷ Moreover, fresh produce, such as fruits and salads that are grown close to the soil, is often consumed raw, exposing consumers to the risk of infection by contaminating organisms.^18,19 However, it should be noted that fresh produce might also be contaminated in stores, following the distribution process or by incorrect human manipulation.^9,20

The use of antibiotics as growth promoters is forbidden in European countries since 2006 thanks to Regulation (EC) No. 1831/2003²⁰ of the European Parliament on additives for use in animal nutrition. Antibiotics are therefore used only for therapeutic or occasionally metaphylactic purposes. As a consequence, also raw food types, as fresh produce, can harbor significant populations of antibiotic-resistant bacteria and are now considered important vehicles for rapid transfer of antibiotic resistance determinants.^21,22 Considering that fresh produce is usually consumed without applying a sanitization process (e.g., cooking), this can lead to the transfer of bacteria harboring antibiotic resistance genes (ARGs) in the human gastrointestinal tract.²³

The aim of this study was to assess the occurrence of ESBL, AmpC-type β-lactamase, and carbapenamase-producing Gram-negative bacteria isolated from fresh vegetables and RTE prepacked salads sold in the local markets of Modena (Italy).

Methods

Sampling of fresh vegetables and RTE salads

In this study, 80 different fresh vegetable samples, comprising carrot, spring onion, arugula, chicory, Batavia green lettuce, Batavia red lettuce, frisee salad, red chicory, and iceberg lettuce, as well as 80 RTE prepacked salads (mixed salads comprising 3 to 6 different types of the following vegetables: Batavia green lettuce, arugula, carrot, chicory, frisee salad, spring onion, Batavia red lettuce, and red chicory), were purchased from 20 local food markets in the city of Modena (Italy) from October 2015 to August 2016. All samples were produced in Italy.

Each sample (25 g) was placed in a sterile plastic bag together with 225 ml buffered peptone water (Scharlab, Milan, Italy) and then homogenized for 2 min in a laboratory blender Stomacher^® 400 Circulator (Seward Ltd., Worthing, UK).

One hundred microliters of the appropriate dilutions was spread on plates of MacConkey agar (Scharlab) supplemented with cefotaxime (1 μg/ml) and, following incubation at 37°C for 24 hr, up to 3 colonies were selected and subcultured onto the same medium again at 37°C for 24 hr. The identification of species and testing of antimicrobial susceptibility were carried out using the Vitek 2 system and AST-GN041 card (bioMerieux, Florence, Italy).

Phenotypic identification of ESBLs and AmpC

ESBL production was tested by the double-disc synergy test (DDST), as recommended by the Clinical and Laboratory Standards Institute (CLSI).²⁴ DDST was performed by placing discs of cefotaxime (CTX, 30 μg; Beckton, Dickinson and Company, Breda, The Netherlands), cefotaxime with clavulanic acid (CTX-CLA, 30/10 μg), ceftazidime (CAZ, 30 μg), and ceftazidime with clavulanic acid (CAZ-CLA, 30/10 μg) on plates of Mueller-Hinton agar (MH, Scharlab, Milan, Italy). ESBL production was considered positive when, following incubation at 37°C for 24 hr, the growth inhibitory zone, either around the CTX-CLA or the CAZ-CLA disc, increased by 5 mm or more compared with the diameter around the disc containing CTX or CAZ alone. The original DDST was further modified to detect AmpC producers using the double synergy differential test.²⁵ This test was performed using both CTX (30 μg) and CAZ (30 μg) discs alone and in combination with boronic acid (CTX-BA and CAZ-BA; 30 μg plus 10 μl of a 60-mg/ml solution of benzo(b)thiophene-2-boronic acid in dimethyl sulfoxide; Sigma, Milan, Italy). For detection of AmpC β-lactamase, results were interpreted based on an increase of the growth inhibitory zone by 5 mm or more around the CTX-BA and/or CAZ-BA discs compared with the diameter around the disc of the corresponding cephalosporin alone.

In Gram-negative organisms, DDST may not show positivity in the presence of AmpC together with ESBL since the AmpC type of β-lactamase inhibits the action of clavulanate. Hence, the synergistic effect of clavulanic acid and third-generation cephalosporins, which are used, is obscured. For this reason, detection of ESBLs was performed by applying the combination disc test, as recommended by EUCAST,²⁶ which utilized cefotaxime, ceftazidime, and cefepime with an amoxicillin–clavulanate disc. ESBL detection was considered positive when inhibition zones around any of the cephalosporin discs were augmented in the direction of the disc containing clavulanic acid.

Phenotypic identification of MBL and KPC

We applied a simple phenotypic method to detect and differentiate the production of metallo-β-lactamases (MBL), Klebsiella pneumoniae carbapenemase (KPC), or MBL and KPC together.²⁷ More specifically, the test was performed by inoculating the Mueller-Hinton agar with the standard diffusion method and then preparing one disc of meropenem (MER, 10 μg) without any inhibitor and three discs containing 400 μg of phenylboronic acid (PBA; Sigma) or 292 μg of EDTA (Sigma) or both 400 μg of PBA and 292 μg of EDTA. The agar plates were incubated at 37°C overnight. The diameters of the growth inhibitory zone around the MER discs with inhibitors were compared with the plain MER disc.

Detection of ARGs

DNA was extracted from the samples using a standard heat lysis protocol.²⁸

ESBL genes (bla_TEM, bla_SHV, and bla_CTX-M),²⁹ plasmid-mediated AmpC-type genes (bla_MOX bla_CIT, bla_DHA, bla_ACC, bla_EBCM, and bla_FOX),^28,30 and carbapenamase genes (bla_KPC; bla_IMP, bla_VIM, bla_OXA-48, and bla_NDM)^30,31 were detected, as previously reported.

PCR-positive amplicons were purified by the QIAquick PCR Purification Kit according to the manufacturer's instructions (Qiagen, Milan, Italy) and directly sequenced using amplification primers on the 3130 Genetic Analyzer (Applied Biosystem, Milan, Italy). Purification and sequencing were carried out by Genex s.r.o. (CZ, Czech Republic). Sequence alignment and analysis were performed online using the BLAST program of the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov).

Plasmid replicon typing

Plasmids were identified by PCR-based replicon typing, as previously described.³² Plasmid DNA was amplified by five multiplex and three simplex PCRs using 18 pairs of primers that have been recognized as Inc replicon types: FIA, FIB, FIC, HI1, HI2, I1-Ic, L/M, N, P, W, T, A/C, K, B/O, X, Y, F, and FIIA. The resulting plasmid replicons were visualized by gel electrophoresis on 1% agarose gels and stained with GelRed™ (Biotum, Fremont, CA).

Conjugation assay

Transferability of ESBL/AmpC and MBL genes by conjugation was performed by matting out assays using E. coli J53-2 (met⁻, pro⁻, rif ^R) as the recipient strain. Transconjugants were selected on MacConkey agar containing cefotaxime (2 mg/L; Sigma), cefoxitin (32 mg/L; Sigma), or imipenem (2 mg/L) plus rifampicin (100 mg/L; Sigma). The means of the conjugation frequency (transconjugants/recipients) were evaluated. Gene transfer was confirmed by PCR and sequencing, as described above.

Statistical analysis

The prevalence of ESBL and AmpC from fresh vegetables and RTE salads was analyzed by performing the chi-square test using SPSS software (version 21.0; SPSS, Inc., Chicago, IL). The level of significance was set at a p-value under 0.05.

Results

A total of 312 Gram-negative strains were isolated and identified by means of their biochemical properties and subsequently confirmed with Vitek 2.

Of these, 160 isolates were recovered from fresh vegetables, with a high proportion of Citrobacter freundii (26%). The other strains identified were Enterobacter cloacae (22%), Pantoea agglomerans (16%), Rahnella aquatilis (12%), E. coli (10%), Hafnia alvei (5%), Enterobacter aerogenes (5%), and Pseudomonas fluorescens (4%).

The remaining 152 isolates were recovered from RTE salads, with a high prevalence of E. cloacae (26.3%) and E. coli (25%). The other strains identified were E. aerogenes (22.3%), P. fluorescens (7.9%), Klebsiella ozaenae (7.9%), P. agglomerans (5.3%), and C. freundii (5.3%).

In this study, only 44 (24 of fresh vegetables and 20 of RTE salads) of 312 (14.1%) isolates were considered for final analysis, following the results of the Vitek 2 Advanced Expert System (AES), which detected a phenotype of ESBL, AmpC, and carbapenamase-producing Gram-negative bacteria; no bacterial strains with a phenotype for intrinsic β-lactam antibiotic resistance were included. Moreover, these strains were successively screened by phenotypic tests (Table 1).

Table 1.

Antimicrobial Resistance, Double Synergy Differential Test, and Combined Disc Test of Strains Isolated from Fresh Vegetables (Samples 1–24) and Ready-to-Eat Salads (Samples 1a–20a)

			β-lactam/MIC value (mg/l)										DSDT		Combined disc
Sample number	Species	Origin	AM	AMC	TZP	FOX	CAZ	CTX	FEP	IMP	MER	ERT	SYN-CLA	SYN-BA	SYN-BOR	SYN-EDTA
1	Rahnella aquatilis	Carrot	>32	16	8	8	2	4	<1	<0.25	<0.25	<0.25	+	−	−	−
2	Escherichia coli	Arugula	>32	16	8	4	2	4	<1	<0.25	<0.25	<0.25	+	−	−	−
3	R. aquatilis	Carrot	>32	>32	16	4	>64	4	<1	<0.25	<0.25	<0.25	+		−	−
4	E. coli	Arugula	>32	8	8	<4	>64	4	<1	<0.25	<0.25	<0.25	+	−	−	−
5	E. coli	Iceberg lettuce+arugula	>32	8	8	<4	>64	4	<1	<0.25	<0.25	<0.25	+	−	−	−
6	E. coli	Iceberg lettuce+arugula	>32	16	8	<4	2	>64	4	<0.25	<0.25	<0.25	+	−	−	−
7	C. freundii	Arugula	>32	16	8	<4	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
8	C. freundii	Arugula	>32	>32	16	<4	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
9	C. freundii	Arugula	>32	16	16	<4	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
10	C. freundii	Arugula	>32	2	8	16	>32	>64	<1	<0.25	<0.25	<0.25	−	+	−	−
11	C. freundii	Arugula	>32	4	16	8	>32	>64	<1	0.5	<0.25	<0.25	−	+	−	−
12	C. freundii	Arugula	>32	>32	64	8	4	>16	>16	0.5	<0.25	<0.25	+	−	−	−
13	Enterobacter cloacae	Frisee salad	>32	>32	16	8	2	>64	8	<0.25	<0.25	<0.25	+	−	−	−
14	E. cloacae	Frisee salad	>32	>32	64	8	2	16	2	<0.25	<0.25	<0.25	+	−	−	−
15	Pantoea agglomerans	Frisee salad	>32	>32	32	4	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
16	P. agglomerans	Frisee salad	>32	>32	16	4	4	>64	16	<0.25	<0.25	<0.25	+	−	−	−
17	P. agglomerans	Frisee salad	>32	>32	>32	8	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
18	P. agglomerans	Frisee salad	>32	>32	32	32	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
19	P. agglomerans	Carrot	>32	16	8	8	<1	8	2	<0.25	<0.25	<0.25	+	−	−	−
20	P .agglomerans	Carrot	>32	4	8	>64	16	>64	2	<0.25	<0.25	<0.25	−	+	−	−
21	Hafnia alvei	Arugula	>32	4	8	>64	16	>64	2	<0.25	<0.25	<0.25	−	+	−	−
22	H. alvei	Arugula	>32	16	16	8	2	4	<1	<0.25	<0.25	<0.25	+	−	−	−
23	R. aquatilis	Arugula	>32	16	16	4	2	2	<1	<0.25	<0.25	<0.25	+	−	−	−
24	R. aquatilis	Arugula	>32	16	8	8	2	4	<1	<0.25	<0.25	<0.25	+	−	−	−
1A	P. fluorescens	Mixed salad (6 different types of vegetables)	>32	8	8	<4	1	1	<1	>8	>8	>4		−	−	+
2B	P. fluorescens	Mixed salad (6 different types of vegetables)	>32	8	8	<4	1	1	<1	>8	>8	>4		−	−	+
3A	C. freundii	Mixed salad (6 different types of vegetables)	>32	16	>64	<4	8	>64	4	<0.25	<0.25	<0.25	+	−	−	−
4A	C. freundii	Mixed salad (6 different types of vegetables)	>32	>16	32	<4	4	32	2	<0.25	<0.25	<0.25	+	−	−	−
5A	C. freundii	Mixed salad (6 different types of vegetables)	>32	>32	16	<4	8	32	2	<0.25	<0.25	<0.25	+	−	−	−
6A	C. freundii	Mixed salad (6 different types of vegetables)	>32	16	16	<4	8	32	2	<0.25	<0.25	<0.25	+	−	−	−
7A	C. freundii	Mixed salad (3 different types of vegetables)	>32	>16	8	<4	>32	>64	<1	<0.25	<0.25	<0.25	+	−	−	−
8A	C. freundii	Mixed salad (3 different types of vegetables)	>32	4	16	8	>32	>64	<1	0.5	<0.25	<0.25	+	+	−	−
9A	E. cloacae	Mixed salad (2 different types of vegetables)	>32	>32	>64	<4	8	32	>16	0.5	<0.25	<0.25	+	−	−	−
10A	E .cloacae	Mixed salad (2 different types of vegetables)	>32	>32	16	8	4	>64	<1	<0.25	<0.25	<0.25	+	−	−	−
11A	Pseudomonas fluorescens	Mixed salad (3different types of vegetables)	>32	>32	>64	2	2	2	<1	>8	>8	>4	+	−	−	+
12A	P. fluorescens	Mixed salad (3different types of vegetables)	>32	>32	32	>4	<1	2	<1	>8	>8	>4	−	−	−	+
13A	E. coli	Mixed salad (4 different types of vegetables)	>32	>32	16	4	4	>64	16	<0.25	<0.25	<0.25	−	−	−	−
14A	E. coli	Mixed salad (4 different types of vegetables)	>32	>32	32	8	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
15A	P. agglomerans	Mixed salad (2 different types of vegetables)	>32	>32	32	32	4	>64	2	<0.25	<0.25	<0.25	+	−	−	−
16A	P. agglomerans	Mixed salad (2 different types of vegetables)	>32	16	8	8	<1	8	2	<0.25	<0.25	<0.25	+	−	−	−
17A	K.. ozaenae	Mixed salad (6 different types of vegetables)	>32	4	8	>64	16	>64	2	<0.25	<0.25	<0.25	+		−	−
18A	K. ozaenae	Mixed salad (6 different types of vegetables)	>32	4	8	>64	16	>64	2	<0.25	<0.25	<0.25	+		−	−
19A	K. ozaenae	Mixed salad (3 different types of vegetables)	>32	16	16	8	2	4	<1	<0.25	<0.25	<0.25	+	−	−	−
20A	K. ozaenae	Mixed salad (3 different types of vegetables)	>32	16	16	4	2	2	<1	<0.25	<0.25	<0.25	+	−	−	−

AM, ampicillin; AMC, amoxicillin/clavulanic acid; CAZ, ceftazidime; CTX, cefotaxime; DSDT, double synergy differential test; ERT, ertapenem; FEP, cefepime; FOX, cefoxitin; IMP, imipenem; MER, meropenem; SYN-BA, synergy with boronic acid; SYN-CLA, synergy with clavulanic acid; SYN-EDTA, synergy with EDTA; TZP, piperacillin/tazobactam.

Among the 24 isolates from fresh vegetables, 20 showed an increase (>5 mm) in the inhibition zone diameter for cefotaxime, ceftazidime, and/or cefepime in the presence of amoxicillin/clavulanic acid (AMC), which was indicative of ESBL production, while four isolates showed enlargement of the inhibition zone in the presence of boronate and were classified as AmpC producers; (Table 1). The ESBL/AmpC-producing isolates belonged to the following genes (species/number of isolates): bla_CTX-M-15 (C. freundii/4; E. coli/1; P. agglomerans/6), bla_SHV-12 (E. coli/3), bla_RAHN-1 (R. aquatilis/4), bla_CTX-M-1 (E. cloacae/2), and bla_DHA-1 (E. cloacae/2) (Table 2). Among all 24 microbial strains, we identified two H. alvei harboring a bla_ACC gene. In this microbial species, the bla_ACC gene is located on the chromosome,³³ thus it was not further characterized.

Table 2.

Overview of Vegetable Types in Relation to ESBL, AmpC, MBL-Encoding Genes, and Plasmids

		Genotypes
Species	Origin	ESBLs	AmpC	MBL	Plasmid replicons
E. coli	Iceberg lettuce+arugula	SHV-12 (3)CTX-M-15 (1)			I1 (3)FIA (1)
C. freundii	Arugula	CTX-M-15 (4)			HI (4)
E. cloacae	Frisee salad	CTX-M-1 (2)	DHA-1 (2)		HI1 (2); L/M (2)
P. agglomerans	Frisee salad+carrot	CTX-M15 (6)			A/C (6)
R. aquatilis	Arugula	RAHN-1 (4)			ND
P. fluorescens	Mixed salad (6 different types of vegetables)+Mixed salad (3 different types of vegetables)			VIM-1(2) IMP-1(2)	ND
E. coli	Mixed salad (4 different types of vegetables)	SHV-12 (2)CTX-M-15 (2)			I1 (2)FIA (2)
C. freundii	Mixed salad (6 different types of vegetables)Mixed salad (3 different types of vegetables)	CTX-M-1 (2)CTX-M-15 (2)			HI1 (2)FIA (2)
K. ozaenae	Mixed salad (6 different types of vegetables)+Mixed salad (3 different types of vegetables)	CTX-M-1 (4)			N (4)
E. cloacae	Mixed salad (2 different types of vegetables)	CTX-M-1 (2)			HI1 (2)
P. agglomerans	Mixed salad (2 different types of vegetables)	CTX-M-15 (2)			A/C (2)

Numbers in parentheses ( ) refer to the number of isolates.

ESBL, extended-spectrum β-lactamase; MBL, metallo β-lactamase; ND, not determined.

Sixteen of 20 RTE salad isolates displayed an ESBL phenotype and four strains were positive for the combined disc with EDTA, suggesting production of the MBL-type enzyme (Table 1). Among the four E. coli isolated, two carried a bla_SHV-12 and two a bla_CTX-M-15. The gene bla_CTX-M-1 was detected in C. freundii (n = 2) as well as in K. ozaenae (n = 4) and E. cloacae (n = 2). Moreover, two C. freundii and two P. agglomerans harbored a bla_CTX-M-15 gene. The MBL genes, identified only in four strains of P. fluorescens, were as follows: two bla_VIM-1 and two bla_IMP-1 (Table 2).

A summary of replicons detected among the 42 isolates is given in Table 2. Overall, seven different replicons were identified, where IncHI1, FIA, and I1 were the most representative types; other Inc groups included HI, L/M, A/C, and N.

When compared with the Inc types, isolates from fresh vegetables and RTE salads were similar among each other. The FIA and I1 plasmids were detected in all E. coli strains; HI, FIA, and HI1 for the ACC-1 gene were found in C. freundii; A/C replicons were detected in all P. agglomerans strains; HI1 and L/M plasmids were found in E. cloacae; and N plasmids were identified in K. ozaenae (Table 2).

The four R. aquatilis and two H. alvei harboring chromosomal level bla_RAHN-1 and bla_ACC genes, respectively, were not characterized by the conjugation assay.^33,34 Thirty-eight bacterial strains were able to transfer genes encoding ESBL/AmpC and MBL by conjugation. The transfer frequency for these strains was of ∼10⁻⁵ for recipient cell. PCR and sequencing confirmed that transconjugants carried the SHV-12, CTX-M-15, CTX-1, DHA-1, VIM-1, and IMP-1 genes. The high transfer frequency highlights the plasmid potential of diffusion and dissemination among susceptible isolates, also of different species.

Finally, a significant difference (χ² = 8.97, p < 0.05) emerged when analyzing sample type (fresh vegetables vs. RTE prepacked salads) with respect to different types of ESBL, AmpC, and MBL determinants.

Discussion

The current study has demonstrated the higher occurrence of ESBL and AmpC-producing strains in bacteria from fresh vegetables than RTE salads. Our results correlate with other studies indicating vegetables as a possible route for the dissemination of resistant genes in the community.^12,13,23,35

Pathogen contamination of fresh and RTE vegetables has been widely described, and in recent years, outbreaks linked to these foods have been reported. The production and processing practices applied to fresh and RTE vegetables are not sufficient to ensure complete microbial safety, and human pathogens can be associated with this type of produce. Many causes may be involved in this issue, such as preharvest contamination (soil, wild or domestic animals, field workers, harvesting equipment, manure and soil amendments, and irrigation water etc.). Improper handling during processing (cutting, washing/sanitizing, and packaging/storing) can also compromise the safety of a product and can provide many opportunities for microbial cross-contamination.³⁶

In recent years, there has been a strong expansion of scientific research applied to food safety. Epidemiological investigations have demonstrated the increasing crucial role of food types in the etiology of specific infectious diseases to the extent of placing them within a large group of syndromes called foodborne diseases.³⁷ These diseases have become an important problem for public health not only in underdeveloped countries but also in high-performing socioeconomic developed countries, where microbial contamination is frequent and difficult to control.³⁸ Contamination of fresh produce by pathogenic or opportunistic microorganisms is now emerging as a major food safety problem and the presence of antibiotic-resistant bacteria can represent an additional risk factor, in particular for old and immunocompromised people. The microbiological quality of food remains one of the most important factors to keep under control as foodborne illnesses are both one of the major causes of morbidity and mortality and an economic damage for food companies.

In our bacterial isolates, the resistant genes present in vegetables mainly belonged to the CTX-M family, and more specifically, CTX-M-15 was found both in fresh vegetables and in RTE salads. CTX-M-15 has been identified as the most common type of ESBL Gram-negative bacteria in Europe and has been increasingly described in community isolates.^39–44 We could hypothesize that contamination by ESBL/AmpC or carbapenemase of vegetable samples and farm environments could be mainly of human source. For example, fecal carriage of CTX-M-producing E. coli in nursing homes in Bavaria was reported by Valenza et al., and the same authors also demonstrated by PFGE typing that a person-to-person transmission or a common source of infection for ESBL-producing microorganisms may occur.⁴⁵ Similarly, cross-contamination during handling by colonized workers is an event that has to be taken into consideration.

Previous studies have reported the isolation of Enterobacteriaceae harboring CTX-M-14 and CTX-M-15 ESBL in fresh culinary herbs from Southeast Asia, highlighting the potential human health risk associated with their consumption.⁴⁶ In addition, another study carried out in fresh vegetables refers the detection of the same ESBL variants (CTX-M-15, CTX-M-14, CTX-M-1, SHV-12, and RAHN-1) as the one detected in our study.^{8,18,38,47,48} DHA-1 AmpC β-lactamases have also been reported in E. cloacae by previous findings.⁴⁹ The link between antibiotic use in animal production and human health has been widely described,⁵⁰ and livestock, in particular, is considered to be an important reservoir for ESBL/AmpC or, to a lesser extent, carbapenemase-producing strains transferred to humans through the food chain. The potential implication of wastewater used for irrigation cannot be discarded as a route for dissemination of these antibiotic-resistant bacteria, and recent studies also suggest that irrigation groundwater constitutes a source of antibiotic-resistant human pathogens, that may enter the food chain through vegetable ingestion.⁵¹ Additionally, an increase in the carriage of ARGs in soil along the years⁵² has been reported, highlighting the role of this element as a reservoir of antibiotic-resistant bacteria and genes. Last, given that the community could also be a reservoir of these microorganisms,⁵³ postharvest contamination may also occur in fresh vegetables and RTE salads, a particularly handled and processed food.

We also found two strains of R. aquatilis with chromosomal ESBL genes; human infections caused by these bacteria have already been described, mainly in immunocompromised patients.^54,55 The main risk associated with these species of Enterobacteriaceae, once they have been ingested with contaminated food, is that although they are not pathogenic strains, they may facilitate the dissemination of ESBL genes to Gram-negative opportunistic pathogens (e.g., by conjugation) during their (temporary) transit in the human intestine.¹² This claim is corroborated by our results of conjugation assays that reported a high transfer frequency of plasmids harboring ESBL and other resistant genes. Accordingly, these genes can spread within a bacterial population and from one environment to another, such as food and human gut microbiota, by exchanging genetic information.

The isolation of carbapenemase-producing microorganisms in RTE salads is alarming and constitutes a food safety issue. In the present study, we detected P. fluorescens MBL-producing Gram-negative bacteria in RTE salads for the first time. Although this organism may not be considered a pathogen, its contribution to the resistome and potential for lateral gene transfer also to clinically relevant bacteria is certainly a cause for concern.

Previous findings^8,34 have demonstrated that ESBL-producing Enterobacteriaceae (ESBL-E) carried several different plasmids, some of which were also found in Enterobacteriaceae strains isolated here from fresh vegetables and RTE salads.

Transfer of the bla_CTX-M-1 gene in E. coli strains was associated with acquisition of the IncI1 replicon plasmid, as has already been reported by other authors in E. coli isolates of human and animal origins.^47,56,57 The variation and distribution of plasmids and genes found in the present investigation are consistent with the findings of Carattoli et al.³²

The impact of food processing on the transfer of antimicrobial resistance to humans is widely reported⁵⁸ and, in this context, the possible link between microbiological hygiene parameters and occurrence of ESBL bacteria could be also considered. As previously reported, fresh produce can be contaminated in several ways before, during, and after food processing. Environmental postcontamination and cross-contamination with other foods and during human handling represent important sources of pathogenic bacteria, also endowed with antimicrobial resistance features.²³ In particular, fresh fruit and vegetables are more perishable than other types of foods and, usually, customers touch them to determine maturity. These raw products may be ingested without any prior processing or preservation treatment, and the increasing demand for raw and minimally processed food contributes to the possibility of introducing multiresistant pathogenic bacteria. Therefore, minimal processing technologies can lead to stressed and/or sublethally damaged cells, and several studies have demonstrated that stress may also impact the phenotypic antimicrobial resistance of microorganisms, increasing horizontal transmission of plasmids containing antimicrobial resistance genes by conjugation.^59,60 These bacteria can also survive the ingestion process and may contribute to the spread of ARGs to the resident bacterial colonizing the intestinal tract. Last, the occurrence of biofilms in the food industry is largely involved in the antimicrobial resistance problem. This ecosystem represents an evolutionary advantage for microorganisms because it grants protection from different adverse conditions and the sessile life in community also confers an ideal state for resistance transfer. In conclusion, surveillance remains the key factor to address the issue of antibiotic resistance in fresh vegetables, fruits, and salads and should be employed at all levels of food processing in compliance, applying the community safety rules. Even for this criticism, the application of good hygiene and manufacturing practices, based on Hazard Analysis and Critical Control Point (HACCP) principles, should be implemented to contribute to achieving food safety.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Kummerer

2004. Resistance in the environment. J. Antimicrob. Chemoth., 54:311–320.

Mc Manus

P.S.

, Stockwell

V.O.

, Sundin

G.W.

, and Jones

A.L.

. 2002. Antibiotic use in plant agriculture. Annu. Rev. Phytopathol., 40:443–465.

Ackers

M.L.

, Mahon

B.E.

, Leahy

, Goode

, Damrow

, Hayes

P.S.

, Bibb

W.F.

, Rice

D.H.

, Barrett

T.J.

, Hutwagner

, Griffin

P.M.

, and Slutsker

. 1998. An outbreak of Escherichia coli 0157:H7 infections associated with leaf lettuce consumption. J. Infect. Dis., 6:1588–1593.

Delaquis

, Bach

, and Dinu

L.D.

. 2007. Behavior of Escherichia coli O157:H7 in leafy vegetables. J. Food Prot., 70:1966–1974.

Sagoo

S.K.

, Little

C.L.

, Ward

, Gillespie

I.A.

, and Mitchell

R.T.

. 2003. Microbiological study of ready-to-eat salad vegetables from retail establishments uncovers a national outbreak of salmonellosis. J. Food. Prot., 66:403–409.

Brendan

R.J.

, Griffin

P.M.

, Cole

, Walsh

K.A.

, and Chai

J.S.

. 2013. Outbreak-associated Salmonella enterica Serotypes and Food Commodities, United States, 1998–2008. Emerg. Infect. Dis., 8:1239–1244.

Walia

, Rana

S.W.

, Maue

, Rana

, Kumar

, and Walia

S.K.

. 2013. Prevalence of multiple antibiotic-resistant Gram-negative bacteria on bagged, ready-to-eat baby spinach. Int. J. Environ. Health. Res., 23:108–111.

Ben Said

, Jouini

, Klibi

, Dziri

, Alonso

C.A.

, Boudabous

, Ben Slama

, and Torres

. 2015. Detection of extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae in vegetables, soil and water of the farm environment in Tunisia. Int. J. Food. Microbiol., 203:86–92.

van Hoek

A.H.

, Veenman

, van Overbeek

W.M.

, Lynch

, de Roda Husman

A.M.

, and Blaak

. 2015. Prevalence and characterization of ESBL- and AmpC-producing Enterobacteriaceae on retail vegetables. Int. J. Food Microbiol., 204:1–8.

10.

Aserse

A.A.

, Räsänen

L.A.

, Aseffa

, Hailemariam

, and Lindström

. 2013. Diversity of sporadic symbionts and nonsymbiotic endophytic bacteria isolated from nodules of woody, shrub, and food legumes in Ethiopia. Appl. Microbiol. Biotechnol., 97:10117–10134.

11.

Berendonk

T.U.

, Manaia

C.M.

, Merlin

, Fatta-Kassinos

, Cytryn

, Walsh

F.H.

, Bürgmann

, Sørum

, Norström

, Pons

, Kreuzinger

, Huovinen

, Stefani

, Schwartz

, Kisand Baquero

, and Martinez

J.L.

. 2015. Tackling antibiotic resistance: the environmental framework. Nat. Rev. Microbiol., 13:310–317.

12.

Blaak

, van Hoek

A.H.

, Veenman

, Docters van Leeuwen

A.E.

, Lynch

, van Overbeek

W.M.

, and de Roda Husman

A.M.

. 2014. Extended spectrum ß-lactamase- and constitutively AmpC-producing Enterobacteriaceae on fresh produce and in the agricultural environment. Int. J. Food Microbiol., 3:168–169.

13.

Finley

R.L.

, Collignon

, Larsson

D.G.

, McEwen

S.A.

, Li

Z.X.

, Gaze

W.H.

, Reid-Smith

, Timinouni

, Graham

D.W.

, and Topp

. 2013. The scourge of antibiotic resistance: the important role of the environment. Clin. Infect. Dis., 57:704–710.

14.

Gaze

W.H.

, Zhang

, Abdouslam

N.A.

, Hawkey

P.M.

, Calvo-Bado

, Royle

, Brown

, Davis

, Kay

, Boxall

A.B.

, and Wellington

E.M.

. 2011. Impacts of anthropogenic activity on the ecology of class 1 integrons and integron-associated genes in the environment. ISME J., 5:1253–1261.

15.

Berger

C.N.

, Sodha

S.V.

, Shaw

R.K.

, Griffin

P.M.

, Pink

, Hand

, and Frankel

. 2010. Fresh fruit and vegetables as vehicles for the transmission of human pathogens. Environ. Microbiol., 12:2385–2397.

16.

Baker-Austin

, Wright

M.S.

, Stepanauskas

, and McArthur

J.V.

. 2006. Co-selection of antibiotic and metal resistance. Trends Microbiol., 14:176–182.

17.

Burridge

, Weis

J.S.

, Cabello

, Pizarro

, and Bostick

. 2010. Chemical use in salmon aquaculture: a review of current practices and possible environmental effects. Aquaculture, 306:7–23.

18.

Randall

L.P.

, Lodgeb

M.P.

, Elviss

N.C.

, Lemma

F.L.

, Hopkins

K.L.

, Teale

C.J.

, and Woodford

. 2017. Evaluation of meat, fruit and vegetables from retail stores in five United Kingdom regions as sources of extended-spectrum beta-lactamase (ESBL)-producing and carbapenem-resistant Escherichia coli. Int. J. Food Microbiol., 214:283–290.

19.

Zekar

F.M.

, Granier

S.A.

, Marault

, Yaici

, Gassilloud

, Manceau

, Touati

, and Millemann

. 2017. From farms to markets: gram-negative bacteria resistant to third-generation cephalosporins in fruits and vegetables in a region of north africa. Front. Microbiol., 8:1569.

20.

Regulation no 1831/2003 of the European Parliament and of the council of 22 September 2003 on additives for use in animal nutrition. Official Journal of the European Union, 2003, L 268/29, 18.10.2003.

21.

Boehme

, Werner

, Klare

, Reissbrodt

, and Witte

. 2004. Occurrence of antibiotic-resistant enterobacteria in agricultural foodstuffs. Mol. Nutr. Food Res., 48:522–531.

22.

Falomir

M.P.

, Gozalbo

, and Rico

. 2010. Coliform bacteria in fresh vegetables: from cultivated lands to consumers. A. (ed.). Curr. Res. Technol. Educ. Top., 2:1175–1181.

23.

Schwaiger

, Helmke

, Hölzel

C.S.

, and Bauer

. 2011. Antibiotic resistance in bacteria isolated from vegetables with regards to the marketing stage (farm vs. supermarket). Int. J. Food Microbiol., 148:191–196.

24.

Clinical and Laboratory Standards Institute (CLSI). 2010. Performance standards for antimicrobial susceptibility testing. M100-S20., Wayne, PA.

25.

Sabia

, Gargiulo

, and Sarti

. 2012. Evaluation of a double synergy differential test (DSDT) for differential detec-tion of ESBL and AmpC-type b-lactamases in Escherichia coli, Klebsiella pneumoniae and Proteus mirabilis. New Microbiol., 35:221–225.

26.

European Committee on Antimicrobial Susceptibility Testing (EUCAST). 2013. Guideline for detection of resistance mechanisms and specific resistances of clinical and/or epidemiological importance of resistance mechanisms v1. 0:pp. 1–40.

27.

Tsakris

, Poulou

, Pournaras

, Voulgari

, Vrioni

, Themeli-Digalaki

, Petropoulou

, and Sofianou

. 2010. A simple phenotypic method for the differentiation of metallo-beta-lactamases and class A KPC carbapenemases in Enterobacteriaceae clinical isolates. J. Antimicrob. Chemother., 65:1664–1671.

28.

Perez-Perez

J.F.

, and Hanson

N.D.

. 2002. Detection of plasmid-mediated AmpC β-lactamase genes in clinical isolates by using mulpiplex PCR. J. Clin. Microbiol., 40:2153–2162.

29.

Kim

, Jeon

, Lee

, Park

, Lee

, and Kim

. 2009. Rapid detection of extended spectrum b-lactamase (ESBL) for Enterobacteriaceae by use of a multiplex PCR based method. Infect. Chemother., 41:181–184.

30.

Dallenne

, Da Costa

, Decre

, Favier

, and Arlet

. 2010. Development of a set of multiplex PCR assays for the detection of genes encoding important b-lactamases in Enterobacteriaceae. J. Antimicrob. Chemother., 65:490–495.

31.

Nordmann

, Poirel

, Carrër

, Toleman

M.A.

, and Walsh

T.R.

. 2011. How to Detect NDM-1 Producers J. Clin. Microbiol., 49:718–772.

32.

Carattoli

, Bertini

, Villa

, Falbo

, Hopkins

K.L.

, and Threlfall

E.J.

. 2005. Identification of plasmids by PCR-based replicon typing. J. Microbiol. Methods, 63:219–228.

33.

Jacoby

G.A.

2009. AmpC beta-lactamases. Clin. Microbiol. Rev., 22:161–168.

34.

Bellais

, Poirel

, Fortineau

, Decousser

J.W.

, Nordmann

. 2001. Biochemical–genetic characterization of the chromosomally encoded extended-spectrum class A β-lactamase from Rahnella aquatilis. Antimicrob. Agents Chemother., 45:2965–29658.

35.

Kim

H.S.

, Chon

J.W.

, Kim

Y.J.

, Kim

D.H.

, Kim

M.S.

, and Seo

K.H.

. 2015. Prevalence and characterization of extended-spectrum-β-lactamase-producing Escherichia coli and Klebsiella pneumoniae in ready-to-eat vegetables. Int. J. Food Microbiol., 207:83–86.

36.

Jung

, Jang

, and Matthews

K.R.

. 2014. Effect of the food production chain from farm practices to vegetable processing on outbreak incidence. Microb. Biotechnol., 7:517–527.

37.

Newell

D.G.

, Koopmans

, Verhoef

, Duizer

, Aidara-Kane

, Sprong

, Opsteegh

, Langelaar

, Threfall

, Scheutz

, van der Giessen

, and Kruse

. 2010. Food-borne diseases—the challenges of 20 years ago still persist while new ones continue to emerge. Int. J. Food Microbiol., 139:3–15.

38.

EFSA/ECDC. 2017. The European Union summary report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2015. EFSA J., 15:4694.

39.

Ruimy

, Brisabois

, Bernede

, Skurnik

, Barnat

, Arlet

, Momcilovic

, Elbaz

, Moury

, Vibet

M.A.

, Courvalin

, Guillemot

, and Andremont

. 2010. Organic and conventional fruits and vegetables contain equivalent counts of Gram-negative bacteria expressing resistance to antibacterial agents. Environ. Microbiol., 12:608–615.

40.

Rossi

, Rizzotti

, Felis

G.E.

, and Torriani

. 2014. Horizontal gene transfer among microorganisms in food: current knowledge and future perspectives. Food Microbiol., 42:232–243.

41.

Lascols

, Hackel

, Hujer

A.M.

, Marshall

S.H.

, Bouchillon

S.K.

, Hoban

D.J.

, Hawser

S.P.

, Badaland

R.E.

, and Bonomo

. 2012. Using nucleic acid microarrays to perform molecular epidemiology and detect novel beta-lactamases: a snapshot of extended-spectrum beta-lactamases throughout the world. J. Clin. Microbiol., 50:1632–1639.

42.

Laxminarayan

, Duse

, Wattal

, Zaidi

A.K.

, Wertheim

H.F.

, Sumpradit

, Vlieghe

, Hara

G.L.

, Gould

I.M.

, Goossens

, Greko

, So

A.D.

, Bigdeli

, Tomson

, Woodhouse

, Ombaka

, Peralta

A.Q.

, Qamar

F.N.

, Mir

, Kariuki

, Bhutta

Z.A.

, Coates

, Bergstrom

, Wright

G.D.

, Brown

E.D.

, and Cars

. 2013. Antibiotic resistance-the need for global solutions. Lancet Infect. Dis., 13:1057–1098.

43.

Livermore

D.M.

, Canton

, Gniadkowski

, Nordmann

, Rossolini

G.M.

, Arlet

, Ayala

, Coque

T.M.

, Kern-Zdanowicz

, Luzzaro

, Poirel

, and Woodford

. 2007. CTX-M: changing the face of ESBLs in Europe. J. Clin. Microbiol., 59:165–174.

44.

Peirano

, Costello

, and Pitout

J.D.

. 2010. Molecular characteristics of extended-spectrum beta-lactamase-producing Escherichia coli from the Chicago area: high prevalence of ST1 producing CTX-M-15 in community hospitals. Int. J. Antimicrob. Agents., 36:19–23.

45.

Valenza

, Nickel

, Pfeiferb

, Pietschb

, Voigtländerc

, Lehner-Reindla

, and Höllerd

. 2017. Prevalence and genetic diversity of extended-spectrum b-lactamase (ESBL)-producing Escherichia coli in nursing homes in Bavaria, Germany. Vet. Microbiol., 200:138–141.

46.

Veldman

, Kant

, Dierikx

, van Essen-Zandbergen

, Wit

, and Mevius

. 2014. Enterobacteriaceae resistant to third-generation cephalosporins and quinolones in fresh culinary herbs imported from Southeast Asia. Int. J. Food Microbiol., 177:72–77.

47.

Reuland

E.A.

, al Naiemi

, Raadsen

S.A.

, Savelkoul

P.H.M.

, Kluytmans

J.A.J.W.

, and Vandenbroucke-Grauls

C.M.J.E.

. 2014. Prevalence of ESBL-producing Enterobacteriaceae in raw vegetables. Eur. J. Clin. Microbiol. Infect. Dis., 33:1843–1846.

48.

Raphael

, Wong

L.K.

, and Riley

L.W.

. 2011. Extended-spectrum Beta-lactamase gene sequences in Gram-negative saprophytes on retail organic and nonorganic spinach. Appl. Environ. Microbiol., 77:1601–1607.

49.

Mohd Khari

F.I.

, Karunakaran

, Rosli

, and Tee Tay

. 2016. Genotypic and Phenotypic Detection of AmpC β-lactamases in Enterobacter spp. Isolated from a Teaching Hospital in Malaysia. PLoS One, 3:1–13.

50.

Wallmann

2014. Antimicrobial resistance: challenges ahead. Vet. Rec., 175:323–324.

51.

Araújo

, Henriques

I.S.

, Leandro

S.M.

, Alves

, Pereira

, and Correia

. 2014. Gulls identified as major source of fecal pollution in coastal waters: a microbial sourcetracking study. Sci. Total Environ., 470–471:84–91.

52.

Knapp

C.W.

, McCluskey

S.M.

, Singh

B.K.

, Campbell

C.D.

, Hudson

, and Graham

D.W.

. 2011. Antibiotic resistance gene abundances correlate with metal and geochemical conditions in archived Scottish soils. PLoS One, 6:e27300.

53.

Rahman

M.A., and El-Sherif

R.H.

. 2011. High rates of intestinal colonization with Extended-Spectrum Lactamase-Producing Enterobacteriaceae among healthy individuals. J. Investig. Med., 59:1284–1286.

54.

Gaitan

J.I.

and Bronze

M.S.

. 2010. Infection caused by Rahnella aquatilis. Am. J. Med. Sci., 339:577–579.

55.

Chang

C.L.

, Shin

J.H.

, Lee

E.Y.

, and Son

H.C.

. 1999. Rahnella aquatilis Sepsis in an Immunocompetent Adult. J. Clin. Microbiol., 12:4161–4416.

56.

Ben Sallem

, Ben Slama

, Rojo-Bezares

, Porres-Osante

, Jouini

, Klibi

, Boudabous

, Sáenz

, and Torres

. 2014. IncI1 plasmids carrying bla_CTX-M-1 or bla_CMY-2 genes in Escherichia coli from healthy humans and animals in Tunisia. Microb. Drug Resist., 20:495–500.

57.

Mnif

, Harhour

, Jdidi

, Mahjoubi

, Genel

, Arlet

, and Hammami

. 2013. Molecular epidemiology of extended-spectrum beta-lactamase-producing Escherichia coli in Tunisia and characterization of their virulence factors and plasmid addiction systems. BMC Microbiol., 13:147.

58.

Verraes

, Van Boxstael

, Van Meervenne

, Van Coillie

, Butaye

, Catry

, de Schaetzen

M.A

, Van Huffel

, Imberechts

, Dierick

, Daube

, Saegerman

, De Block

, Dewulf

, and Herman

. 2013. Antimicrobial resistance in the food chain: a review. Int. J. Environ. Res. Public Health, 10:2643–2669.

59.

McMahon

M.A.S.

, Blair

I.S.

, Moore

J.E.

, and McDowell

D.A.

. 2007. The rate of horizontal transmission of antibiotic resistance plasmids is increased in food preservation-stressed bacteria. J. Appl. Microbiol., 103:1883–1888.

60.

Ganjian

, Nikokar

, Tieshayar

, Mostafaei

, Amirmozafari

, and Kiani

. 2012. Effects of salt stress on the antimicrobial drug resistance and protein profile of Staphylococcus aureus. Jundishapur. J. Microbiol., 5:328–331.