Microbiological quality of ready-to-eat salads: An underestimated vehicle of bacteria and clinically relevant antibiotic resistance genes

International Journal of Food Microbiology 166 (2013) 464–470

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Short communication

Microbiological quality of ready-to-eat salads: An underestimated vehicle of bacteria and clinically relevant antibiotic resistance genes Joana Campos a, Joana Mourão a, Nazaré Pestana a, Luísa Peixe a, Carla Novais a, Patrícia Antunes a,b,⁎ a b

REQUIMTE, Laboratório de Microbiologia, Faculdade de Farmácia, Universidade do Porto, Rua Jorge Viterbo Ferreira n° 228, 4050-313 Porto, Portugal Faculdade de Ciências da Nutrição e Alimentação, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal

a r t i c l e

i n f o

Article history: Received 27 March 2013 Received in revised form 9 July 2013 Accepted 7 August 2013 Available online 17 August 2013 Keywords: Ready-to-eat salads Antibiotic resistance blaSHV-2 qnrB9 E. coli D/ST69

a b s t r a c t The increase demand for fresh vegetables is causing an expansion of the market for minimally processed vegetables along with new recognized food safety problems. To gain further insight on this topic we analyzed the microbiological quality of Portuguese ready-to-eat salads (RTS) and their role in the spread of bacteria carrying acquired antibiotic resistance genes, food products scarcely considered in surveillance studies. A total of 50 RTS (7 brands; split or mixed leaves, carrot, corn) were collected in 5 national supermarket chains in Porto region (2010). They were tested for aerobic mesophilic counts, coliforms and Escherichia coli counts as well as for the presence of Salmonella and Listeria monocytogenes. Samples were also plated in different selective media with/ without antibiotics before and after enrichment. The E. coli, other coliforms and Enterococcus recovered were characterized for antibiotic resistance profiles and clonality with phenotypic and genetic approaches. A high number of RTS presented poor microbiological quality (86%—aerobic mesophilic counts, 74%—coliforms, 4%—E. coli), despite the absence of screened pathogens. In addition, a high diversity of bacteria (species and clones) and antibiotic resistance backgrounds (phenotypes and genotypes) were observed, mostly with enrichment and antibiotic selective media. E. coli was detected in 13 samples (n = 78; all types and 4 brands; phylogenetic groups A, B1 and D; none STEC) with resistance to tetracycline [72%; tet(A) and/or tet(B)], streptomycin (58%; aadA and/or strA-strB), sulfamethoxazole (50%; sul1 and/or sul2), trimethoprim (50%; dfrA1 or dfrA12), ampicillin (49%; blaTEM), nalidixic acid (36%), ciprofloxacin (5%) or chloramphenicol (3%; catA). E. coli clones, including the widespread group D/ST69, were detected in different samples from the same brand or different brands pointing out to a potential cross-contamination. Other clinically relevant resistance genes were detected in 2 Raoultella terrigena carrying a blaSHV-2 and 1 Citrobacter freundii isolate with a qnrB9 gene. Among Enterococcus (n = 108; 35 samples; Enterococcus casseliflavus—40, Enterococcus faecalis—20, Enterococcus faecium—18, Enterococcus hirae—9, Enterococcus gallinarum—5, and Enterococcus spp.—16) resistance was detected for tetracyclines [6%; tet(M) and/or tet(L)], erythromycin [3%; erm(B)], nitrofurantoin (1%) or ciprofloxacin (1%). The present study places ready-to-eat salads within the spectrum of ecological niches that may be vehicles for antibiotic resistance bacteria/genes with clinical interest (e.g. E. coli-D-ST69; blaSHV-2) and these findings are worthy of attention as their spread to humans by ingestion cannot be dismissed. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Fruits and vegetables are essential components of a human healthy diet and nutrition. Currently, there is a growing demand for these products, causing an expansion of the market for minimally processed vegetables ready-to-use or ready-to-eat (Little and Gillespie, 2008; Olaimat and Holley, 2012). Ready-to-eat fresh vegetables are globally produced and commercialized all year round and have been accompanied by new food safety threats since they are eaten raw and usually without other washing/decontamination procedures (Little and Gillespie, 2008). In ⁎ Corresponding author at: Universidade do Porto, Faculdade de Ciências da Nutrição e Alimentação, Rua Dr. Roberto Frias, 4200 Porto, Portugal. Tel.: +351 22 5074320; fax: +351 22 5074329. E-mail address: [email protected] (P. Antunes). 0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.08.005

fact, fresh vegetables (e.g., salads and sprouts) have recently become increasingly recognized as potential vehicles of foodborne diseases, which are highlighted by large and serious international outbreaks such as the associated with the fenugreek seed sprouts contaminated with Escherichia coli O104:H4 in 2011 in Europe and tomato and spinach with Salmonella and E. coli O157 in America (Olaimat and Holley, 2012; EFSA/AIT, 2013). On the other hand, a growing food safety concern is related with the role of food in human exposure to antimicrobial resistant bacteria, including zoonotic pathogens and/or commensal and environmental bacteria as a reservoir of resistance genes (EFSA, 2008, 2012; DANMAP, 2012). Although food products of animal origin are the principal foods carrying such antimicrobial resistant bacteria/genes, contamination during preparation, handling and processing of fresh food of plant origin, such as salads, is becoming a current concern (EFSA, 2008). The contribution of minimally processed vegetables for

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

the spread of antibiotic resistance remains unclear and has been scarcely considered in surveillance studies. The aim of this study was to evaluate the microbiological quality of bagged ready-to-eat salads available for sale in Portugal and to investigate their contribution for the dissemination of bacteria resistant to clinically relevant antibiotics. 2. Materials and methods 2.1. Sampling plan Fifty samples of bagged ready-to-eat salads belonging to seven brands (arbitrarily designated I to VII) and available in five national supermarket chains in Porto region, were collected from February through May of 2010. They included leaves (n = 26) carrot (n = 10), mixture of leaves + carrot (n = 5) and of leaves + carrot + corn (n = 9). All samples were randomly purchased before their bestbefore date, transported to the laboratory in their original package in a maximum period of 1 h and analyzed immediately after their arrival.

465

30 μg, teicoplanin—30 μg, nitrofurantoin—300 μg, erythromycin—15 μg, minocycline—30 μg and quinuspristin-dalfopristin—15 μg was only tested for Enterococcus spp. and susceptibility to gentamicin—10 μg, kanamycin—30 μg, streptomycin—10 μg, nalidixic acid—30 μg, sulfamethoxazole—300 μg and trimethoprim—5 μg only for E. coli and other coliforms. Minimal inhibitory concentration (MIC) to ciprofloxacin was performed by the E-test method and interpreted following EUCAST guidelines (EUCAST, 2012). Multidrug resistance (MDR) was considered when the isolates were resistant to three or more antibiotics of different families. The study of resistance to several extended spectrum β-lactams (ceftazidime, ceftriaxone, cefotaxime, cefepime, cefoxitin, aztreonam and imipenem) and the double disk synergy test (DDST) for detection of extended-spectrum β-lactamases (ESBL) was performed in MuellerHinton agar (bioMérieux) with and without cloxacillin (250 μg/ml) in E. coli resistant to amoxicillin and other coliforms recovered from the chromID Coli medium with ceftazidime and/or cefotaxime. 2.5. Characterization of antibiotic resistance and virulence genes

2.2. Microbiological quality analysis Twenty-five grams of each sample was added to a culture medium/diluent (1:10; homogenized for 2 min in a Stomacher), in agreement with specific standard methods for aerobic mesophilic count (ISO 4833:2003 protocol), coliforms (AFNOR/NF BIO 12/20-12/06), E. coli (ISO, 16649-2:2001) and the pathogenic bacteria Salmonella (ISO 6579:2002; AFNOR BIO 12/01-04/94 protocol) and Listeria monocytogenes (ISO, 11290-1:2004; AFNOR BIO-12/11-03/04 protocol). The samples were classified as satisfactory, borderline and unsatisfactory considering the national microbiological guidelines for ready-to-eat foods (Santos et al., 2005). 2.3. Screening of antibiotic resistant bacteria and species identification The study of antibiotic resistance was conducted in Gram positive (Enterococcus) and Gram negative bacteria (E. coli and other coliforms) broadly used in surveillance programs (DANMAP, 2012; EFSA, 2012). Samples were plated on different culture media directly from the sample suspension and also after an enrichment step (Buffered Peptone Water 37 °C/18 h). For the search of Gram negative bacteria resistant to clinically relevant antibiotics, 0.1 ml of each sample/enrichment was plated on chromID Coli medium (bioMérieux) without antibiotics and also supplemented with tetracycline (8 μg/ml), sulfamethoxazole (256 μg/ml), ceftazidime (2 μg/ml) or cefotaxime (2 μg/ml). For the search of antibiotic resistant Enterococcus 0.1 ml of each sample/ enrichment was plated on Slanetz–Bartley agar medium without antibiotics and also supplemented with vancomycin (4 μg/ml), ampicillin (8 μg/ml) or gentamicin (128 μg/ml). From every plate with colonies, different morphotypes were selected to recover isolates from different species and/or resistance phenotypes, present within the same salad sample. Identification of Gram negative bacteria to the species level was conducted by API ID32GN/API 20E and/or 16SrDNA/ gyrB sequencing. Enterococcus was identified by Gram stain, catalase reaction, hydrolysis of esculin and growth on NaCl 6.5%. The species were identified by PCR as previously described (Novais et al., 2005). Positive and negative controls were included in all PCR reactions. 2.4. Study of antibiotic susceptibility Antimicrobial susceptibility tests were performed by the disk diffusion method following Clinical and Laboratory Standards Institute guidelines (CLSI, 2011). E. coli ATCC 25922 and Enterococcus faecalis ATCC 29212 were used as control strains. The susceptibility to amoxicillin—10 μg, ciprofloxacin—5 μg, chloramphenicol—30 μg and tetracycline—30 μg was studied both to Enterococcus spp., E. coli and other coliforms. Susceptibility to gentamicin—120 μg, streptomycin—300 μg, vancomycin—

Antibiotic resistance genes were searched by PCR following the conditions previously described (Aarestrup et al., 2000; Antunes et al., 2006, 2007, 2011; Novais et al., 2005). The search of genes encoding resistance to glycopeptides (vanA, vanB, vanC1, vanC2), tetracyclines [tet(M), tet(L), tet(O), tet(S), tet(K)] and macrolides [erm(B)] in Enterococcus spp. and to sulfamethoxazole (sul1, sul2 and sul3), tetracycline [tet(A), tet(B) and tet(G)], chloramphenicol (floR, cmlA and catA), amoxicillin (blaTEM, blaSHV, and blaCTX-M), streptomycin (aadA and strA-strB), trimethoprim (dfrA1 and dfrA12) and ciprofloxacin [qnrA, qnrB, qnrS, qepA and aac(6′)-Ib-cr] in E. coli and coliforms was performed. The detection and characterization of class 1 integrons was performed by PCR, PCR/restriction fragment length polymorphism and sequencing in E. coli and coliform resistant to sulfamethoxazole (Antunes et al., 2006, 2007). Positive and negative controls were included in all PCR reactions. Detection of virulence gene markers of STEC strains (stx1, stx2 and eaeA) was also searched by PCR as previously described (Nguyen et al., 2009). 2.6. Clonal relatedness The assignation of E. coli phylogenetic groups was carried out by the multiplex PCR assay previously described by Clermont et al. (2000). Clonal relatedness among E. coli and specific coliform isolates were assessed by pulsed field gel electrophoresis (PFGE) following XbaI digestion of genomic DNA according to the standard 1 day protocol of the CDC (Antunes et al., 2006). Salmonella Braenderup H9812 (CDC) was used as molecular size marker. E. coli isolates belonging to phylogroup D (often associated with extra-intestinal infections) were further characterized by multilocus sequence typing (MLST) by using the standard fumC gene (analyzed for the presence of specific nucleotide polymorphisms associated with E. coli epidemic clonal lineages) or the complete scheme, according to the available database (http://mlst.ucc.ie/mlst/dbs/Ecoli) in specific cases. 2.7. Statistical analysis Differences among the rates of antibiotic resistance among E. coli isolates recovered from different plates were analyzed by the Fisher exact test (α = 0.05) using the GraphPad Prism software, version 6.0a. 3. Results 3.1. Microbiological evaluation of ready-to-eat-salads A high number of packed ready-to-eat salads of all types of vegetables were classified as unsatisfactory due to the presence of N106 CFU/g

466

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

aerobic mesophilic microorganisms (n = 43; 86%) and N 104 CFU/g coliforms (n = 37; 74%) (Santos et al., 2005) (Fig. 1). Comparison between the types of salads revealed that about 95% (23 out of 24) of those carrying carrots exceed those guideline values (Santos et al., 2005). Two samples (4%) carried E. coli at unsatisfactory levels of ≥102 CFU/g (1.2 × 102 CFU/g and 1.3 × 104 CFU/g), corresponding to mixed leaves (n = 1) and mixed leaves + carrot + corn (n = 1) from the same brand and one sample (leaves) at borderline level. Nevertheless, after enrichment step 26% (n = 13; all types and 4 brands) carried E. coli isolates. Salmonella and L. monocytogenes were not detected in any ready-to-eat salad samples studied (Fig. 1). 3.2. Detection and characterization of antibiotic resistant bacteria 3.2.1. E. coli E. coli (n = 78 isolates) from different phylogenetic groups (A, n = 20; B1, n = 26; D, n = 32) were recovered from 26% of the samples studied (n = 13; all types and 4 brands). Among those isolates none was simultaneously positive for genes associated with STEC isolates stx1, stx2 and eae. Table 1 shows the occurrence of antibiotic resistant E. coli isolates recovered from different selective media with or without enrichment. The enrichment step allowed the detection of most E. coli isolates (n = 63) from diverse phylogenetic groups and isolates resistant to tetracycline or sulfamethoxazole were better detected using agar plates containing each of these two compounds and after enrichment (p b 0.05) than in media without any supplement. Most of those E. coli isolates presented resistance to tetracycline, streptomycin, sulfamethoxazole and/or trimethoprim contrasting with all but one obtained with the standard counting methods (without enrichment and from selective media without antibiotics) which were susceptible to all the antibiotics tested. The presence of resistant isolates of E. coli (n = 58/78; 74%) was observed in 11 samples, corresponding to 22% of the ready-to-eat salads of different types and brands, and multidrug resistance in 47 (81%) of those isolates (9 samples). Because enrichment steps and different medium allowed potential selection of more than one isolate of the same strain from the same sample, from each sample we select E. coli isolates of different phylogenetic groups and/or antibiotic resistance phenotypes for further

86%

74%

4%

studies (Table 2). The 29 isolates belonged to three phylogenetic groups (A, n = 11; B1, n = 10; D, n = 8) and were associated with 21 PFGE types (A = 8; B1 = 9; D = 4). Isolates from D phylogroup showed a diversity of fumC sequences. Two of them showed a fumC allele compatible with ST69 (CC69), a main clonal complex of E. coli, which was confirmed by MLST complete scheme. Three clones (PFGE types A-7, B1-6 and D-3), including clone D/ST69, were detected in different ready-to-eat salads samples of the same brand and one clone (PFGE type D-2) was isolated in two brands throughout the study (Table 2). Fifteen E. coli isolates (9 samples of all types of salads) of different phylogenetic groups and clones, including 2 isolates from the widespread group D/ST69, were MDR. They carried genes encoding for resistance to amoxicillin (blaTEM), tetracyclines [tet(A), tet(B)], aminoglycosides [strA-strB, aadA1, aadA2], sulfamethoxazole (sul1, sul2) and trimethoprim (dfrA1, dfrA12), some inserted in two class 1 integron types (dfrA1-aadA1; dfrA12-orfF-aadA2) of 13 isolates (Table 2). Extended spectrum β-lactamase (ESBL) and plasmid-mediated quinolone resistance (PMQR) mechanisms were not observed. 3.2.2. Other coliforms Other Gram negative β-galactosidase-producing isolates, considered as coliforms (n = 30), were recovered from the selective plates containing ceftazidime (n = 9) or cefotaxime (n = 21) of 26 samples. The different species included Rahnella aquatilis (n = 13), Citrobacter freundii complex (n = 8), Raoultella terrigena (n = 2), Hafnia alvei (n = 2), Enterobacter cloacae (n = 3), Enterobacter aerogenes (n = 1) and Cronobacter sakazakii (n = 1). The DDST test to ESBL detection was positive for 15 isolates identified as R. aquatilis (n = 13) and R. terrigena (n = 2) obtained from different salads and brands. All R. aquatilis presented only the intrinsic chromosomal gene blaRAHN-1 and identical β-lactams resistance profile. The ESBL phenotype of the two R. terrigena, clonally related, was associated with an acquired blaESBL gene identified as blaSHV-2 (GenBank Accession NO. KC879153) conferring resistance to the β-lactams ampicillin and cefotaxime. Those isolates were MDR and among them were detected genes codifying for resistance to tetracycline [tet(B)] and chloramphenicol (cmlA) in addition to a class 1 integron carrying genes for trimethoprim (dfrA1), streptomycin (aadA1) and sulphonamide (sul1) resistance.

0%

0%

Fig. 1. Percentage of RTE-salads types classified as unsatisfactory (“US”), borderline and satisfactory (“S”) for microbiological parameters. The percentage of samples classified as unsatisfactory for each parameter is presented below. Microbiological guidelines (Santos et al., 2005) for RTE-salads (CFU/g): Aerobic mesophilic count, “US” N 106 and “S” ≤ 104; Coliforms, “US” N 104 and “S” ≤ 102; Escherichia coli, “US” ≥ 102 and “S” ≤ 10; Salmonella spp., “S” absence in 25 g; Listeria monocytogenes, “S” absence in 25 g.

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

467

Table 1 Antibiotic resistance of the Escherichia coli isolates recovered by different cultural approaches. Basis of isolate selection

No antibiotic Without enrichment

With enrichment

Tetracycline (8 mg/L) Without enrichment

With enrichment

Sulfametoxazole (256 mg/L) Without enrichment

With enrichment

PHGa (n° isolates; n° samples)

Antibioticb resistance [n° isolates (%)] CIP

NAL

AML

TET

CHL

SUL

TMP

STR

GEN

KAN

B1 (n = 9; 3) Total (n = 9; 3)

0 (0) 0 (0)

0 (0) 0 (0)

1 (11) 1 (11)

1 (11) 1 (11)

0 (0) 0 (0)

0 (0) 0 (0)

0 (0) 0 (0)

1 (11) 1 (11)

0 (0) 0 (0)

0 (0) 0 (0)

A (n = 2; 2) B1 (n = 6; 6) D (n = 3; 3) Total (n = 11; 11)

0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 2 (67) 2 (18)

0 (0) 0 (0) 2 (67) 2 (18)

0 (0) 2 (33) 2 (67) 4 (36)

0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 2 (67) 2 (18)

0 (0) 0 (0) 2 (67) 2 (18)

0 (0) 0 (0) 2 (67) 2 (18)

0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 0 (0) 0 (0)

A (n = 1; 1) B1 (n = 2; 1) Total (n = 3; 1)

1 (100) 0 (0) 1 (33)

1 (100) 0 (0) 1 (33)

0 (0) 0 (0) 0 (0)

1 (100) 1 (50) 2 (67)

0 (0) 0 (0) 0 (0)

0 (0) 1 (50) 1 (33)

0 (0) 1 (50) 1 (33)

0 (0) 1 (50) 1 (33)

0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 0 (0)

A (n = 8; 5) B1 (n = 4; 3) D (n = 12; 4) Total (n = 24; 10)

3 (38) 0 (0) 0 (0) 3 (13)

4 (50) 0 (0) 9 (75) 13 (54)

2 (25) 3 (75) 9 (75) 14 (58)

8 (100) 4 (100) 12 (100) 24 (100)

1 (13) 0 (0) 0 (0) 1 (4)

1 (13) 0 (0) 9 (75) 10 (42)

1 (13) 0 (0) 9 (75) 10 (42)

3 (38) 3 (75) 9 (75) 15 (63)

0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 0 (0) 0 (0)

B1 (n = 3; 1) Total (n = 3; 1)

0 (0) 0 (0)

0 (0) 0 (0)

0 (0) 0 (0)

2 (67) 2 (67)

0 (0) 0 (0)

2 (67) 2 (67)

2 (67) 2 (67)

2 (67) 2 (67)

0 (0) 0 (0)

0 (0) 0 (0)

A (n = 9; 4) B1 (n = 2; 2) D (n = 17; 6) Total (n = 28; 9)

0 (0) 0 (0) 0 (0) 0 (0)

1 (11) 0 (0) 11 (65) 12 (43)

6 (67) 0 (0) 15 (88) 21 (75)

8 (89) 0 (0) 15 (88) 23 (82)

1 (11) 0 (0) 0 (0) 1 (4)

8 (89) 1 (50) 15 (88) 24 (86)

8 (89) 1 (50) 15 (88) 24 (86)

8 (89) 1 (50) 15 (88) 24 (86)

0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 0 (0) 0 (0)

Total (n = 78; 13)

4 (5)

28 (36)

38 (49)

56 (72)

2 (3)

39 (50)

39 (50)

45 (58)

0 (0)

0 (0)

a

PHG, phylogenetic groups. b Abbreviations: CIP, ciprofloxacin; NAL, nalidixic acid; AML, amoxicillin; TET, tetracycline; CHL, chloramphenicol; SUL, sulfamethoxazole; TMP, trimethoprim; STR, streptomycin; GEN, gentamicin; KAN, kanamycin.

The presence of the PMQR genes was positive in one C. freundii (obtained from a mixed salad) non-ESBL producer, with decreased susceptibility to ciprofloxacin (MIC 0.25 μg/mL) and carrying qnrB9 gene (GenBank Accession NO. KC688257). 3.2.3. Enterococcus Enterococcus were found in 70% of the samples (n = 35), mainly when an enrichment step was included (31 out of 35 samples). Different species were identified (≥2 species in 34% of samples) and comprised the clinically relevant species E. faecalis and Enterococcus faecium (Table 3). Resistant isolates (n = 8/108 isolates) were detected in 4 samples (all leaves), corresponding to 7% of the ready-to-eat salad samples. Antibiotic resistance was detected for tetracyclines, erythromycin, nitrofurantoin or ciprofloxacin and co-resistance to tetracyclines and erythromycin in 3 isolates (2 E. faecalis and 1 E. faecium) from the same sample (Table 3). 4. Discussion This study shows that most of ready-to-eat salads (all types and brands) analyzed (86%) presented unsatisfactory microbiological quality according to the Portuguese guidelines (Santos et al., 2005). Poor microbiological quality was more often detected among salads containing carrot, as described (Abadias et al., 2008), which may be related to its direct contact with soil/water or multiple processing stages (e.g. peeling and slicing) favoring the contact with microorganisms. In fact, although aerobic mesophilic count is not related to the “safety” of the product it is indicative of good practices throughout the processing, from production to marketing, including the storage temperature (Abadias et al., 2008; De Giusti et al., 2010). Also the occurrence of a large number of coliforms in most samples as well as E. coli exceeding the 102 UFC/g limit in 4% of the salads is of concern, since E. coli is defined by EU microbiological criteria (Commission Regulation (EC) No, 2073/2005) as

the indicator microorganism for fecal contamination during the manufacturing process (“process hygiene criteria”) for this product category (precut, ready-to-eat fruits and vegetables). Those results observed in this and other European studies (Abadias et al., 2008; Caponigro et al., 2010; De Giusti et al., 2010; Sagoo et al., 2003) emphasize the need for improvements in production hygiene and selection of raw materials in manufacturing processes. In accordance with the EU “food safety criteria” at the market place (Comission Regulation (EC) No, 2073/2005) pathogenic foodborne microorganisms Salmonella and/or L. monocytogenes were not detected in Portuguese ready-to-eat salads, contrasting with other European studies where those bacteria were found in lower levels (Abadias et al., 2008; Caponigro et al., 2010; De Giusti et al., 2010; Sagoo et al., 2003). Nevertheless, outbreaks associated with ready-to-eat vegetables or sprouts were reported in Europe (Little and Gillespie, 2008; Olaimat and Holley, 2012) and spread of pathogenic microorganisms due to international foodproducts trades could not be discarded. Recent studies from lessdeveloped countries, which are important vegetable suppliers, showed that most of the ready-to-eat salads had a poor microbiological quality, including human pathogenic bacteria (Castro-Rosas et al., 2012; Oliveira et al., 2011), highlighting the need to implement more restricted control hygienic standards and measures for imported vegetables. Furthermore, salads as potential vehicle of antibiotic resistant bacteria were scarcely considered in surveillance studies (DANMAP, 2012; EFSA, 2012) and this is one of the few reports (Bezanson et al., 2008; Boehme et al., 2004; Schwaiger et al., 2011) alerting for the presence of those bacteria in this type of food. This work shows E. coli and other opportunistic pathogens from coliform group carrying acquired genes conferring resistance to clinically relevant antibiotics, some of them usually associated with mobile elements and with potential for transferring to pathogens, as well as Enterococcus spp. with acquired resistance. However, standard counting procedures strongly underestimate the multiplicity of antibiotic resistant phenotypes/genotypes (E. coli,

468

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

Table 2 Characterization of the selected Escherichia coli isolates by sample type of RTE-salads. Type/Branda

Leaves I

Sample

S35 S40 S41b S52b

PHGc

PFGE typed

D B1 A B1 A D B1 B1 B1 D D A A A B1

D-2 B1-3 A-6 B1-4 A-7 D-3 B1-5 B1-6 B1-9 D-2 D-1 A-2 A-3 A-1 B1-1

MLSTe

fumC100

ST69 (CC69)

Genotype

AML, NAL, STR, SUL, TET, TMP TET CLO, NAL, STR, SUL, TET, TMP (–) CIP, NAL, TET AML, STR, SUL, TET, TMP (–) AML, STR, TET STR, SUL, TMP AML, NAL, STR, SUL, TET, TMP (–) AML, STR, SUL, TET, TMP AML, SUL, TET, TMP STR, TET (–)

blaTEM, aadA, strA,-strB, sul1,sul2, tet(A), dfrA1 tet(B) catA, aad A, strA-strB, sul1-sul2, tet(A)-tet(B), dfrA1 tet(B) blaTEM, aadA, strA-strB, sul1-sul2, tet(A), dfrA1

II III VII

S33 S14 S26

Carrot I

S34

D

D-2

II

S39

A B1

A-5 B1-2

AML, TET TET

Leaves/Carrot I

S54

II

S31

A A A D A

A-8 A-8 n-ty D-4 A-4

Leaves/Carrot/Corn I S32

D

D-2

A B1 B1 B1

A-7 B1-7 B1-6 B1-8

D

D-3

b

S53

fumC100

Antibiotic resistance Phenotypef

fumC100

ST297

fumC100

AML, NAL, STR, SUL, TET, TMP

1500 bp (dfrA1-aadA1) 1500 bp (dfrA1-aadA1) 1500 bp (dfrA1-aadA1)

2000 bp (dfrA12-orfF-aadA2) 1500 bp (dfrA1-aadA1) 1500 bp (dfrA1-aadA1) 1500 bp (dfrA1-aadA1)

blaTEM, aadA, strA,-strB, sul1-sul2, tet(A), dfrA1 blaTEM, tet(A) tet(A)

1500 bp (dfrA1-aadA1)

STR, SUL, TET, TMP AML, STR, SUL, TMP (–) TET (–)

aadA, sul1, tet(A), dfrA12 blaTEM, strA-strB, sul2, dfrA1

2000 bp (dfrA12-orfF-aadA2) intI1-dfrA1-aadA1

AML, NAL, STR, SUL, TET, TMP

blaTEM, aadA, strA-strB, sul1-sul2, tet(A), dfrA1 tet(B)

tet(B)

CIP, NAL, TET (–) AML, STR, TET STR, SUL, TET, TMP ST69 (CC69)

blaTEM, strA-strB, tet(B) aadA, sul1, dfrA12 blaTEM, aadA, strA-strB, sul1-sul2, tet(A), dfrA1 blaTEM, aadA, sul1, tet(A), dfrA1 blaTEM, aadA, sul1, tet(A), dfrA1 tet(B)

Class 1 integron — bp (gene cassettes)

AML, STR, SUL, TET, TMP

blaTEMstrA-strB, tet(B) aadA, strA-strB, sul1-sul2, tet(A), dfrA1 blaTEM, aadA, strA-strB, sul1-sul2, tet(A), dfrA1

1500 bp (dfrA1-aadA1)

1500 bp (dfrA1-aadA1) 1500 bp (dfrA1-aadA1)

a

Brands are identified from I to VII. Samples classified as unsatisfactory (S52 and S53) or borderline (S41) for the quality control parameter E. coli. c PHG, phylogenetic groups. d PFGE types in bold are those shared by different samples; n-ty, non-typable by PFGE. e ST, sequence type and CC, clonal complex, as identified by Multilocus Sequence Typing. f AML, amoxicillin; CIP, ciprofloxacin; CLO, chloramphenicol; STR, streptomycin; SUL, sulfamethoxazole; NAL, nalidixic acid; TET, tetracycline; TMP, trimethoprim; (–), isolates susceptible to all antibiotics tested. b

Enterococcus spp.), species (Enterococcus spp.) and clonal lineages (E. coli) occurring in ready-to-eat salads, which were mainly recovered after an enrichment step and from plates supplemented with different antibiotics. These data should be considered for future antibiotic resistance surveillance studies in different foodstuffs, including readyto-eat salads. The prevalence rates of antibiotic resistance in fresh vegetables from this and other studies (Schwaiger et al., 2011) are lower than those described for other sources (e.g., humans and animals) (DANMAP, 2012; EFSA, 2012; Novais et al., 2005, 2006). The exception was E. coli, which antibiotic resistance prevalence rates were equivalent to those obtained from isolates implicated in human infections (DANMAP, 2012; Rodríguez-Baño et al., 2012; Valverde et al., 2009) or recovered from other settings (e.g. animal production environment, meat thereof) (DANMAP, 2012; EFSA, 2012). These data pointed out human and/or animal as potential sources of resistant isolates in the farms (e.g. untreated irrigation water, inappropriate organic fertilizers/manure, wildlife and domestic animals) and during the processing and packaging (e.g. human and equipment contamination) of the products. Also, the high percentage of resistance to antibiotics often used as empiric agents in human therapy (e.g. amoxicillin, sulfamethoxazole/trimethoprim) is of concern because this species is frequently implicated in urinary

tract infections admittedly caused by host's own intestinal tract isolates and potentially acquired by the food chain (Manges and Johnson, 2012; Nordstrom et al., 2013). The occurrence of particular MDR clones in different samples and brands, indicates cross-contamination between production lines (e.g., during handling, slicing, washing or disinfection) or a common source of the vegetables (e.g. acquisition from the same farm supplying for different brands). These strains harbored common genetic elements (e.g., integrons) including genes encoding for MDR and two of them belong to a worldwide disseminated lineage commonly observed in extraintestinal human infections, D-ST69 E. coli clone (Manges and Johnson, 2012; Valverde et al., 2009), here first described in ready-toeat salads. Recent studies have also been increasingly describing MDR E. coli infections associated with strains of phylogenetic groups A and B1 (Rodríguez-Baño et al., 2012), those mostly found in this study. A foodborne origin for E. coli D-ST69 clone and other clonal lineages belonging to phylogenetic groups A and B1 and/or their potential transmission to humans via the food supply is scarcely explored and cannot be discarded. Besides MDR E. coli, ready-to-eat salads carried other coliforms with genes coding for resistance to antibiotics considered critically important for human medicine by WHO (WHO, 2012), as fluoroquinolones and 3rd generation cephalosporins, and stresses the role of non-human

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

469

Table 3 Characterization of the selected Escherichia coli isolates by sample type of RTE-salads. Species (No. Isolates) E. casseliflavus (n = 40)

Type of samples (No.)

Leaves (23); carrot (3); leaves/carrot (11); leaves/carrot/corn (3)

a

Antibiotic resistance [nº isolates (%)] TET

MIN

ERY

VAN

0(0)

0(0)

0(0)

0(0)

STR

GEN

AML

NIT

Q/D

CHL

CIP

0(0)

TEC

0(0)

0(0)

0(0)

0(0)

1(3)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

16(80)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

1(6)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

vanC2-40 (100) E. faecalis (n = 20)

Leaves (8); carrot (3); leaves/carrot (4); leaves/carrot/corn (5)

2(10)

1(5)

tet(M)-2 (10) tet(L)-2 (10)

E. faecium (n = 18)

Leaves (12); leaves/carrot (4); leaves/carrot/corn (2)

1(6)

1(6)

tet(M)-1 (6) tet(L)-1 (6)

E. hirae (n = 9)

Leaves (7); leaves/carrot/corn (2)

3(33)

3(33)

2(10)

b

erm(B)-2 (10)

1(6) erm(B)-1 (6)

tet(M)-3 (33) tet(L)-3 (33)

E. gallinarum (n = 5)

Leaves (5)

0(0)

0(0)

vanC1-5 (100) Enterococcus spp. (n = 16) Total (n = 108)

Leaves (8); carrot (4); leaves/carrot (2); leaves/carrot/corn (2)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

0(0)

6 (6)

5(5)

3(3)

0(0)

0(0)

0(0)

0(0)

0(0)

1(1)

17(16)

0(0)

0(0)

a

Abbreviations: TET, tetracycline; MIN, minocycline; ERY, erythromycin; VAN, vancomycin; TEC, teicoplanin; STR, high level of resistance (HLR) to streptomycin; GEN, HLR to gentamicin; AML, amoxicillin; NIT, nitrofurantoin; Q/D, quinuspristin-dalfopristin; CHL, chloramphenicol; CIP, ciprofloxacin. b E. faecalis has natural resistance to Q/D.

sources and environment coliforms as potential reservoirs for ESBL and PMQR genes. As far as we know, this is the first report of the presence of blaSHV-2 in Raoultella, an emerging opportunist pathogen (Castanheira et al., 2009), both in ready-to-eat salads and other niches. The presence of this acquired gene conferring resistance to extended-spectrum βlactams, located in an IncHI2 plasmid of 310 Kb (data not shown), frequently associated with members of Enterobacteriaceae and broad ecological niches (EFSA, 2011; Poirel et al., 2012), enhances the possibility of horizontal transfer to human pathogens. Also important is the occurrence of qnrB9 in a foodborne Citrobacter isolate presenting MIC to ciprofloxacin above the ECOFF, as it has been proposed as the likely source of plasmid-mediated qnrB genes (Jacoby et al., 2011) occurring in a diversity of pathogenic Gram negative bacteria, including those associated with human infections. The multiplicity of bacteria genus found in ready-to-eat salads suggests an ineffective washing/disinfection of the samples and/or survival of bacteria under different physical/chemical conditions. Although the goal of this study did not include tolerance to disinfectants used in food industry (e.g. chlorine), the two MDR D-ST69 E. coli found were strong biofilm producers (Novais et al., 2013), indicating that particular clones can accumulate different adaptive features to efficiently resist industrial processes and/or survive on fresh products (Olaimat and Holley, 2012). The selection of particular bacterial populations potentially associated with the use of disinfectants (e.g. chlorine) could also explain the very low rates of resistance detected for Enterococcus spp., comparing to other non-hospitalar niches (e.g. animal production, healthy human feces) where resistance reported in our country was significatively higher (Novais et al., 2005, 2006). In conclusion, this is one of the few studies (Bezanson et al., 2008; Boehme et al., 2004; Schwaiger et al., 2011) demonstrating the possible role of ready-to-eat salads in the dissemination of bacteria within kitchen environment and placing salads within the spectrum of food products that may be vehicles for antibiotic resistant bacteria/genes with

clinical interest. The extension of the possible reservoirs of antibiotic resistant D-ST69 E. coli clone is worthy of attention since the spread to humans by ingestion of ready-to-eat salads cannot be dismissed. More studies are needed to better understand the colonization ability and the pathogenic potential of particular Enterobacteriaceae clonal lineages transmitted by the food chain, including fresh vegetables. The results of this study also stress the need of multilevel strategies that includes i) implementation of good manufacturing practices to prevent contamination at the pre- and post-harvested level of production and processing; and ii) effective decontamination methods (e.g. washing and disinfection) before packaging.

Acknowledgements This study was funded by Fundação para a Ciência e a Tecnologia (FCT), which belongs to the Ministry of Education and Science from Portugal, through grant no. PEst-C/EQB/LA0006/2011 and supported by Universidade do Porto/Santander Totta “Projectos Pluridisciplinares 2009”. Joana Mourão was supported by a fellowship from Fundação para a Ciência e a Tecnologia (SFRH/BD/77518/2011). The authors are grateful to Ana Cristina Silva, Carmen Costa, Heloísa Nunes and João Pires for technical assistance.

References Aarestrup, F.M., Agerso, Y., Gerner-Smidt, P., Madsen, M., Jensen, L.B., 2000. Comparison of antimicrobial resistance phenotypes and resistance genes in Enterococcus faecalis and Enterococcus faecium from humans in the community, broilers, and pigs in Denmark. Diagn. Microbiol. Infect. Dis. 37, 127–137. Abadias, M., Usall, J., Anguera, M., Solsona, C., Viñas, I., 2008. Microbiological quality of fresh, minimally-processed fruit and vegetables, and sprouts from retail establishments. Int. J. Food Microbiol. 123, 121–129. AFNOR BIO 12/01-04/94, 2010. VIDAS Salmonella (VIDAS SLM) — Ref. 30702 Dual Selective Enrichment for the Rapid Detection of Salmonella.

470

J. Campos et al. / International Journal of Food Microbiology 166 (2013) 464–470

AFNOR BIO 12/11-03/04, 2010. VIDAS Listeria monocytogenes II (VIDAS LMO2) — Ref. 30704 Enrichment Stage at 37 °C. AFNOR/NF BIO 12/20-12/06, 2010. chromID™ Coli medium agar — chromogenic and selective medium designed for the enumeration of coliforms and β-glucuronidase positive E. coli from food samples. Validated for the Enumeration of Coliforms at 37 °C. Antunes, P., Machado, J., Peixe, L., 2006. Characterization of antimicrobial resistance and class 1 and 2 integrons in Salmonella enterica isolates from different sources in Portugal. J. Antimicrob. Chemother. 58, 297–304. Antunes, P., Machado, J., Peixe, L., 2007. Dissemination of sul3-containing elements linked to class 1 integrons with an unusual 3′CS region among Salmonella isolates. Antimicrob. Agents Chemother. 51, 1545–1548. Antunes, P., Mourão, J., Machado, J., Peixe, L., 2011. First description of qnrS1-IncN plasmid in a ST11 Salmonella enteritidis clinical isolate from Portugal. Diagn. Microbiol. Infect. Dis. 69, 463–465. Bezanson, G.S., MacInnis, R., Potter, G., Hughes, T., 2008. Presence and potential for horizontal transfer of antibiotic resistance in oxidase-positive bacteria populating raw salad vegetables. Int. J. Food Microbiol. 30, 37–42. Boehme, S., Werner, G., Klare, I., Reissbrodt, R., Witte, W., 2004. Occurrence of antibiotic-resistant enterobacteria in agricultural foodstuffs. Mol. Nutr. Food Res. 48, 522–531. Caponigro, V., Ventura, M., Chiancone, I., Amato, L., Parente, E., Piro, F., 2010. Variation of microbial load and visual quality of ready-to-eat salads by vegetable type, season, processor and retailer. Food Microbiol. 27, 1071–1077. Castanheira, M., Deshpande, L.M., DiPersio, J.R., Kang, J., Weinstein, M.P., Jones, R.N., 2009. First descriptions of blaKPC in Raoultella spp. (R. planticola and R. ornithinolytica): report from the SENTRY Antimicrobial Surveillance Program. J. Clin. Microbiol. 47 (12), 4129–4130 (Dec). Castro-rosas, J., Cerna-Cortés, J.F., Méndez-Reyes, E., Lopez-Hernandez, D., Gómez-Aldapa, C.A., Estrada-Garcia, T., 2012. Presence of faecal coliforms, Escherichia coli and diarrheagenic E. coli pathotypes in ready-to-eat salads, from an area where crops are irrigated with untreated sewage water. Int. J. Food Microbiol. 156, 176–180. Clermont, O., Bonacorsi, S., Bingen, E., 2000. Rapid and simple determination of the Escherichia coli phylogenetic group. Appl. Environ. Microbiol. 66, 4555–4558. Clinical and Laboratory Standards Institute (CLSI), 2011. Performance standards for antimicrobial susceptibility testing. M100-S21 21st Informational Supplement, 31. Clinical and Laboratory Standards Institute, Wayne, PA. Commission regulation (EC) N° 2073/2005 of 15 November 2005 on microbiological criteria for foodstuffs. Off. J. Eur. Union 338, 1–26 (http://eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri=OJ:L:2005:338:0001:0026:EN:PDF). Danish Integrated Antimicrobial Resistance Monitoring, Research Programme (DANMAP), 2012. DANMAP 2011 — Use of Antimicrobial Agents and Occurrence of antimicrobial Resistance in Bacteria from Food Animals, Food and Humans in Denmark. (http://www.danmap.org). De Giusti, M., Aurigemma, C., Marinelli, L., Tufi, D., De Medici, D., Di Pasquale, S., De Vito, C., Boccia, A., 2010. The evaluation of the microbial safety of fresh ready-to-eat vegetables produced by different technologies in Italy. J. Appl. Microbiol. 109, 996–1006. European Committee on Antimicrobial Susceptibility Testing (EUCAST), 2012. Breakpoints Tables for Interpretation of MICs and Zone Diameters (Version 2.0). European Food Safety Authority (EFSA), 2008. Scientific opinion of the panel on biological hazards on a request from the European Food Safety Authority on foodborne antimicrobial resistance as a biological hazard. EFSA J. 765, 1–87 (http://www.efsa.europa. eu/en/efsajournal/doc/765.pdf). European Food Safety Authority (EFSA), 2011. Panel on Biological Hazards (BIOHAZ); scientific opinion on the public health risks of bacterial strains producing extended-spectrum β-lactamases and/or AmpC β-lactamases in food and food-producing animals. EFSA J. 9 (8), 2322 (http://www.efsa.europa.eu/ en/efsajournal/doc/2322.pd). European Food Safety Authority (EFSA), 2012. European Food Safety Authority and European Centre for Disease Prevention and Control; The European Union Summary Report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2010. EFSA J 10 (3), 2598 (http://www.efsa.europa.eu/en/ efsajournal/doc/2598.pdf).

European Food Safety Authority (EFSA) and Austrian Institute of Technology (AIT) GmbH, 2013. Food of plant origin: production methods and microbiological hazards linked to food-borne disease. Reference: CFT/EFSA/BIOHAZ/2012/01 Lot 1 (food of plant origin with high water content such as fruits, vegetables, juices and herbs). Supporting Publications 2013:EN-402. ISO 11290-1, 2004. Horizontal method for the detection and enumeration of Listeria monocytogenes. Part 1: Detection Method. ISO 16649-2, 2001. Microbiology of food and animal feeding stuffs — horizontal method for the enumeration of beta-glucuronidase-positive Escherichia coli. Part 2: Colony-count Technique at 44 °C Using 5-bromo-4-chloro-3-indolyl beta-D-glucuronide. ISO 4833, 2003. Microbiology of Food and Animal Feeding Stuffs — Horizontal Method for Enumeration of Microorganisms — Colony Count Technique at 30 °C. ISO 6579, 2002. Microbiology of food and animal feeding stuffs. Horizontal Method for the Detection of Salmonella spp. Jacoby, G.A., Griffin, C.M., Hooper, D.C., 2011. Citrobacter spp. as a source of qnrB alleles. Antimicrob. Agents Chemother. 55, 4979–4984. Little, C.L., Gillespie, I.A., 2008. Prepared salads and public health. J. Appl. Microbiol. 105, 1729–1743. Manges, A.R., Johnson, J.R., 2012. Food-borne origins of Escherichia coli causing extraintestinal infections. Clin. Infect. Dis. 55, 712–719. Nguyen, T.V., Le Van, P., Le Huy, C., Gia, K.N., Weintraub, A., 2009. Detection and characterization of diarrheagenic Escherichia coli from young children in Hanoi. Vietnam. J. Clin. Microbiol. 43, 755–760. Nordstrom, L., Liu, C.M., Price, L.B., 2013. Foodborne urinary tract infections: a new paradigm for antimicrobial-resistant foodborne illness. Front. Microbiol. 4, 29. Novais, C., Coque, T.M., Costa, M.J., Sousa, J.C., Baquero, F., Peixe, L.V., 2005. High occurrence and persistence of antibiotic-resistant enterococci in poultry food samples in Portugal. J. Antimicrob. Chemother. 56, 1139–1143. Novais, C., Coque, T.M., Sousa, J.C., Peixe, L.V., 2006. Antimicrobial resistance among faecal enterococci from healthy individuals in Portugal. Clin. Microbiol. Infect. 12, 1131–1134. Novais, A., Vuotto, C., Pires, J., Montenegro, C., Donelli, G., Coque, T.M., Peixe, L., 2013. Diversity and biofilm-production ability among isolates of Escherichia coli phylogroup D belonging to ST69, ST393 and ST405 clonal groups. BMC Microbiol. 13, 144. Olaimat, A.N., Holley, R.A., 2012. Factors influencing the microbial safety of fresh produce: a review. Food Microbiol. 32, 1–19 (Review). Oliveira, M.A.D., Maciel de Souza, V., Morato Bergamini, A.M., De Martinis, E.C.P., 2011. Microbiological quality of ready-to-eat minimally processed vegetables consumed in Brazil. Food Control 22, 1400–1403. Poirel, L., Bonnin, R.A., Nordmann, P., 2012. Genetic support and diversity of acquired extended-spectrum β-lactamases in Gram-negative rods. Infect. Genet. Evol. 12, 883–893. Rodríguez-Baño, J., Mingorance, J., Fernández-Romero, N., Serrano, L., López-Cerero, L., Pascual, A., E.S.B.L.-R.E.I.P.I. Group, 2012. Virulence profiles of bacteremic extendedspectrum β-lactamase-producing Escherichia coli: association with epidemiological and clinical features. PLoS One 7, e44238. Sagoo, S.K., Little, C.L., Ward, L., Gillespie, L.A., 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. Santos, M.I., Correia, C., Cunha, M.I.C., Saraiva, M.M., Novais, M.R., 2005. Valores Guia para avaliação da qualidade microbiológica de alimentos prontos a comer preparados em estabelecimentos de restauração. Rev. Ordem dos Farm 64, 66–68. Schwaiger, K., Helmke, K., Hölzel, C.S., Bauer, J., 2011. Antibiotic resistance in bacteria isolated from vegetables with regards to the marketing stage (farm vs. supermarket). Int. J. Food Microbiol. 15, 191–196. Valverde, A., Cantón, R., Garcillán-Barcia, M.P., Novais, A., Galán, J.C., Alvarado, A., de la Cruz, F., Baquero, F., Coque, T.M., 2009. Spread of bla(CTX-M-14) is driven mainly by IncK plasmids disseminated among Escherichia coli phylogroups A, B1, and D in Spain. Antimicrob. Agents Chemother. 53, 5204–5212. WHO, 2012. Critically Important Antimicrobials for Human Medicine, 3rd edition. WHO Advisory Group on Integrated Surveillance of Antimicrobial Resistance (AGISAR).