Presence and Correlation of Some Enteric Indicator Bacteria

609

Journal of Food Protection, Vol. 78, No. 3, 2015, Pages 609–614 doi:10.4315/0362-028X.JFP-14-229 Copyright G, International Association for Food Protection

Research Note Presence and Correlation of Some Enteric Indicator Bacteria, Diarrheagenic Escherichia coli Pathotypes, and Salmonella Serotypes in Alfalfa Sprouts from Local Retail Markets in Pachuca, Mexico

ESMERALDA RANGEL-VARGAS,1 CARLOS A. GO´ MEZ-ALDAPA,1 M. DEL REFUGIO TORRES-VITELA,2 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 ANGE´ LICA VILLARRUEL-LO´ PEZ,2 ALBERTO J. GORDILLO-MARTI´NEZ,1 AND JAVIER CASTRO-ROSAS1*

1Centro de Investigaciones Quı´micas, Instituto de Ciencias Ba´sicas e Ingenierı´a, Universidad Auto´noma del Estado de Hidalgo, Ciudad del Conocimiento, Carretera Pachuca-Tulancingo Kilometer 4.5, 42183 Mineral de la Reforma, Hidalgo, Mexico; and 2Laboratorio de Microbiologı´a Sanitaria, Centro Universitario de Ciencias Exactas e Ingenierı´a, Universidad de Guadalajara, Marcelino Garcı´a Barraga´n 1451, 44430 Guadalajara, Jalisco, Mexico

MS 14-229: Received 18 May 2014/Accepted 3 December 2014

ABSTRACT Data on the presence of diarrheagenic Escherichia coli pathotypes (DEPs) in alfalfa sprouts and correlations between the presence of coliform bacteria (CB), fecal coliforms (FC), E. coli, DEPs, and Salmonella in alfalfa sprouts are not available. The presence of and correlations between CB, FC, E. coli, DEPs, and Salmonella in alfalfa sprouts were determined. One hundred sprout samples were collected from retail markets in Pachuca, Hidalgo State, Mexico. The presence of indicator bacteria and Salmonella was determined using conventional culture procedures. DEPs were identified using two multiplex PCR procedures. One hundred percent of samples were positive for CB, 90% for FC, 84% for E. coli, 10% for DEPs, and 4% for Salmonella. The populations of CB ranged from 6.2 up to 8.6 log CFU/g. The FC and E. coli concentrations were between ,3 and 1,100 most probable number (MPN)/g. The DEPs identified included enterotoxigenic E. coli (ETEC; 2%), enteropathogenic E. coli (EPEC; 3%), and Shiga toxin–producing E. coli (STEC; 5%). No E. coli O157:H7 strains were detected in any STEC-positive samples. In samples positive for DEPs, the concentrations ranged from 210 to 240 MPN/g for ETEC, 28 to 1,100 MPN/g for EPEC, and 3.6 to 460 MPN/g for STEC. The Salmonella isolates identified included Salmonella enterica serotype Typhimurium in three samples and Salmonella enterica serotype Enteritidis in one. STEC and Salmonella Typhimurium were identified together in one sample. Positive correlations were observed between FC and E. coli, between FC and DEPs, and between E. coli and DEPs. Negative correlations occurred between CB and DEPs and between CB and Salmonella. Neither FC nor E. coli correlated with Salmonella in the sprout samples. To our knowledge, this is the first report of ETEC, EPEC, and STEC isolated from alfalfa sprouts and the first report of correlations between different indicator groups versus DEPs and Salmonella.

The consumption of vegetable seed sprouts has been Diarrheagenic E. coli pathotypes (DEPs) (enterotoxi- growing in global popularity over the past 10 years due to genic E. coli [ETEC], enteropathogenic E. coli [EPEC], consumer demands for novel flavors and because consumers Shiga toxin–producing E. coli [STEC], enteroinvasive E. perceive health benefits of sprouts. However, a number of coli [EIEC], enteroaggregative E. coli [EAEC], diffuse sprout-associated foodborne illness outbreaks have been adherent E. coli, and enteroaggregative-hemorrhagic E. coli reported since the 1990s, and Salmonella and Escherichia [EAHEC]), are important foodborne pathogens (3, 28). A coli O157:H7 strains have been the most commonly recent foodborne outbreak of hemolytic-uremic syndrome associated pathogens (8, 19). Alfalfa and mung bean sprouts caused by EAHEC O104:H4 in several European countries are most often implicated in outbreaks in North America (8, that involved sprouts highlights the importance of screening 19, 25). Sprout-related outbreaks of foodborne illnesses for DEPs in sprouts (3). This EAHEC outbreak involved have also been reported in other parts of the world. An more than 4,000 persons in 16 countries, and raw fenugreek outbreak of E. coli O157:H7 infection from radish sprouts in sprouts were implicated as the vehicle of EAHEC. However Japan in 1996 led to 9,451 illnesses and 12 deaths (33). In the outbreak strain could not be identified on fenugreek Mexico, Salmonella has been isolated from raw alfalfa seeds from the implicated lot (3). Greater frequencies of sprouts (6) and from raw mung bean sprouts (9). Sprouts are incidence data for DEPs are still needed for a wide variety often eaten raw; this increases the potential infection risk of vegetables that are consumed raw, such as sprouts. In associated with sprout consumption. Mexico, DEPs have been associated with diarrheal illness both in native children (15, 16, 41) and in visitors to the * Author for correspondence. Tel: z52 771 717 2000, Ext 6501; Fax: z52 country (32, 42). ETEC, EPEC, STEC, and enteroinvasive 771 717 2000, Ext 6502; E-mail: [email protected]. E. coli (EIEC) are important causes of diarrhea in developed 610 RANGEL-VARGAS ET AL. J. Food Prot., Vol. 78, No. 3 countries and in visitors from regions where DEPs are not these 4 large markets (total of 20 retailers) during a 10-week study endemic (15, 16, 35, 41). ETEC is the leading cause of period in the summer. Samples were purchased every 2 weeks at weanling diarrhea in developing countries and of traveler’s each retailer, resulting in 20 samples per day and a total of 100 diarrhea around the world (35). The predominant reservoirs alfalfa sprout samples for the entire study period. At the time of of ETEC are pigs and cattle (34). EPEC strains have been purchase, the sprouts were packaged in plastic containers and sold at room temperature. The sprouts had been packaged in plastic isolated from a variety of animal species, such as cattle, containers by the retailers or sprout producers. The plastic goats, sheep, chickens, pigeons, and gulls and EPEC (12), containers containing sprouts were placed in sterilized plastic bags strains have been responsible for outbreaks of diarrhea in in a cooler with frozen gel packs for transport to the laboratory and both developing and developed countries (35). STEC strains analyzed no more than 1 h after purchase. are human pathogens associated with foodborne illness; some strains can cause hemorrhagic colitis, which in some Sample preparation. Subsamples (100 g) from each sample cases may progress to hemolytic-uremic syndrome (35). were placed in a sterile plastic bag, and 900 ml of sterilized lactose Ruminants, primarily cattle, are the predominant reservoir broth (Bioxon, Becton Dickinson, Mexico) was added to each Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 of STEC (4). EIEC cause an invasive dysenteric form of sprout subsample. The subsamples were pummeled in a stomacher diarrhea in humans (35). Humans are a major reservoir for (Seward, Worthing, UK) at 260 rpm for 1 min, and sample EIEC (28). Foodborne bacteria, including DEPs, are usually dilutions for microorganism counts were prepared using sterile transmitted by contaminated food. Fecal contamination of peptone water (0.1%). Subsamples were analyzed to detect the presence of CB, FC, E. coli, Salmonella, ETEC, EPEC, EIEC, food and drinking water is the major route of infection by STEC, and EAEC as previously described (9, 23). pathogenic bacteria for humans (30). The DEPs are classified based on their unique virulence Microbiological analyses. CB, FC, and E. coli were factors and can be identified by these traits alone. However, determined exactly as described in the U.S. Food and Drug each group has different pathogenic properties, different Administration Bacteriological Analytical Manual (BAM) (48). virulence factors, different epidemiologies, and different Briefly, for CB, 1 ml of each dilution was transferred to petri reservoirs (28, 35). dishes, and violet red bile agar was then poured into each plate, Recently, we isolated ETEC, EIEC, and STEC and followed by incubation at 35uC for 48 h. FC and E. coli were Salmonella from raw mung bean sprouts in Mexico (9); analyzed by the most-probable-number (MPN) procedure: 1 ml of however, with the exception of this study and with the serial dilutions in 0.1% sterile peptone water of each subsample homogenate was inoculated into nine tubes containing lactose exception of E. coli O157:H7 investigations in sprouts (8, broth (Bioxon) (three tubes from dilution 1021, three tubes from there are no reports in the relevant literature on the 22 23 25, 33), dilution 10 , and three tubes from dilution 10 ) and into presence of other DEPs in sprouted alfalfa seeds intended Durham tubes. After incubation at 37uC for 48 h, a loopful of for human consumption. positive culture suspension (determined by turbidity and gas On the other hand, E. coli has been used as an indicator production) was transferred to tubes containing Fluorocult brilliant of fecal contamination of food for many years (13). green 2%–bile lactose broth (Merck, Darmstadt, Germany). After Numerous studies correlate fecal coliform levels with the incubation at 44.5 ¡ 0.2uC for 48 h, tubes that were positive for presence of E. coli. Nevertheless, the value of the fecal growth and gas production were considered positive for FC. For coliform assay as a fecal contamination indicator is nullified determining the presence of E. coli specifically, FC-positive tubes when nonfecal-origin bacteria are the principal microbes were used to identify indole formation (48). All tubes that were detected by the assay (36, 38). It has been reported that positive for indole and gas production were streaked onto eosin sprouts are generally contaminated with coliforms (6, 7, 9, methylene blue agar (Bioxon). Two to three presumptive E. coli colonies were selected from methylene blue agar plates and 36, 44). These microorganisms are mainly nonfecal-origin biochemically characterized by the indole, methyl red, Voges- bacteria from seeds (44), and they increase in concentration Proskauer, citrate tests (IMViC; Bioxon) (48). Biochemical during the sprouting of seeds (7). In addition, Salmonella confirmation of presumptive E. coli isolates was done with the strains have been isolated from sprout samples that tested API 20E test (bioMe´rieux, Hazelwood, MO). All E. coli isolates negative for E. coli (9). This means that FC and E. coli may formed typical colonies. FC and E. coli levels were calculated not be indicators of fecal pathogenic bacteria (such as DEPs using the MPN tables described in BAM (48). We isolated one to or Salmonella) on sprouts. The objective of this study was to three E. coli strains from each E. coli–positive tube of Fluorocult measure the presence of coliform bacteria (CB), fecal brilliant green bile lactose broth. It is known that the confirmation coliforms (FC), E. coli, DEPs (ETEC, EPEC, EIEC, STEC, of only one colony as E. coli is sufficient to regard that broth tube EAEC, and diffuse adherent E. coli), and Salmonella on as positive for E. coli. These positive tubes were used to calculate alfalfa sprouts from public markets and to determine the the E. coli concentrations as described in BAM (48). All confirmed correlations between CB, FC, E. coli, DEPs, and Salmo- E. coli strains were streaked on tryptic soy agar slants, incubated at 37 C for 24 h, and maintained at 3 to 5 C until they were used for nella. u u PCR. MATERIALS AND METHODS Multiplex PCRs for DEP locus identification. All E. coli Sprout samples. A total of 100 alfalfa (Medicago sativa) strains isolated and confirmed by biochemical tests were analyzed sprout samples (250 g each) were purchased in the four largest using two multiplex PCRs to identify several genetic loci public markets in Pachuca, Hidalgo State, Mexico. The city of associated with diarrheagenic E. coli exactly as previously Pachuca has 13 public markets, but only 4 of these are sizable. described (23). Briefly, the first multiplex PCR identifies the Sprout samples were collected from 5 vegetable retailers in each of following loci: heat-stable and heat-labile enterotoxins (st and lt) J. Food Prot., Vol. 78, No. 3 DIARRHEAGENIC E. COLI PATHOTYPES AND SALMONELLA IN SPROUTS 611

TABLE 1. Populations and frequencies of coliform bacteria, fecal coliforms, E. coli, diarrheagenic E. coli pathotypes, and Salmonella on alfalfa sprout samplesa

Microorganism or indicator Minimum Median Maximum Frequency (%)

CB 6.2 6.9 8.6 100 FC ,3 41 1,100 90 E. coli ,3 11 1,100 84 DEPs ,3 ,3 1,100 10 Salmonella ND ND ND 4 a Minimum, median, and maximum values are in log CFU per gram for coliform bacteria and in most probable number (MPN) per gram for fecal coliforms and E. coli (n ~ 100 samples). CB, coliform bacteria; FC, fecal coliforms; DEPs, diarrheagenic E. coli pathotypes; ND, not determined. for ETEC, intimin (eaeA) and bundle-forming pilus (bfp) for coli, DEPs, and Salmonella. A P value of ,0.05 was considered Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 EPEC, Shiga toxin 1 and 2 (stx1 and stx2) and intimin (eaeA) for significant. Statistical analysis was performed using Statistica STEC, and invasion-associated loci (ial) for EIEC. The second (version 8, StatSoft, Inc., Tulsa, OK). multiplex PCR identifies three EAEC plasmid-borne virulence genes, including the master regulon (aggR), dispersin (aap), and RESULTS AND DISCUSSION the autotransporter Tol C (aatA). STEC isolates were further In the 100 samples analyzed, CB, FC, E. coli, DEPs, characterized by the expression of the O157 lipopolysaccharide and H7 flagellar antigens using only the E. coli O157:H7 latex and Salmonella were isolated from 100, 90, 84, 10, and 4 agglutination test kit (RIM E. coli O157:H7 Latex Test Kit, Remel, samples, respectively (Table 1). The populations of CB Lenexa, KS). All reagents, nucleoside triphosphates, primers, and ranged from 6.2 to 8.6 log CFU/g. The FC and E. coli Taq polymerase used in PCRs were from Invitrogen (AccesoLab, concentrations were between ,3 and 1,100 MPN/g. The CB Me´xico D.F., Mexico), and a thermal cycler Px2 (Thermo Electron levels observed are in agreement with the 7.3 to 8.5 log Corporation, Beverly, MA) was used. CFU/g of alfalfa sprouts reported previously (6). In a study The MPN for ETEC, EPEC, and STEC was calculated done in the United States, FC were isolated from alfalfa following the MPN method in BAM (48). As we described above, sprout samples at concentrations ranging from 1.1 | 103 PCR confirmation of only one E. coli strain as a DEP was to 2.4 | 105 MPN/100 g of sprouts (44); Klebsiella sufficient to regard the original broth tube from which was isolated pneumoniae isolates were found in most of the FC-positive the E. coli strain analyzed as positive for ETEC, EPEC, or STEC. samples in that study. Thus, we used this information to calculate the DEP concentration The DEPs identified included ETEC, EPEC, and with the MPN tables (48). STEC; these were isolated from 2, 3, and 5% of samples, Salmonella isolation and identification. The presence of respectively. In these positive samples, the concentrations Salmonella was analyzed by following the method described in ranged from 210 to 240 MPN/g for ETEC, 28 to BAM (48). Briefly, bags containing sprout samples in lactose broth 1,100 MPN/g for EPEC, and 3.6 to 460 MPN/g for STEC. were incubated for 18 h at 37uC (preenrichment). Aliquots (1 ml) No E. coli O157:H7 isolates were detected in any STEC- from preenriched subsamples were then enriched in tetrathionate positive samples (Table 2). For the STEC, only the stx2 (10 ml; Bioxon) and selenite cystine (10 ml; Bioxon) broths and locus was detected in one strain, the stx1 locus was detected incubated for a further 24 h at 43uC (tetrathionate) or 37uC in three strains, and both stx1 and stx2 were detected in one (selenite cystine). Both enrichment broths were then streaked onto strain (Table 2). STEC strains have been identified from Hektoen enteric agar (Bioxon), xylose lysine desoxycholate agar raw foods in Mexico, including vegetable salads (5) and (Bioxon), and bismuth sulfite agar (Bioxon) plates and incubated at carrot juice (46). The pathogenesis of STEC is linked to 37uC for 24 to 48 h. Three typical Salmonella colonies and one atypical colony per plate were inoculated in triple sugar iron agar different virulence factors, such as Shiga toxins (the and lysine iron agar and incubated for 24 h at 35uC (48). The products of stx1 and stx2) (35). It has been reported that Salmonella isolates with typical biochemical reactions in triple the products of stx1 and stx2 cause different degrees and sugar iron agar and lysine iron agar were then biochemically types of tissue damage (31); the product of stx2 is more toxic identified with a commercial biochemical kit (API 20E, bio- than that of stx1 to human renal endothelial cells (31). Other Me´rieux, Me´xico D.F., Mexico). Cultures presumptively identified important virulence determinants in STEC include the locus as Salmonella with the API 20E test were then serologically of enterocyte effacement (LEE), which is shared by EPEC. identified using both the somatic polyvalent (O) test and the This 35- to 45-kb pathogenicity island is responsible for the Salmonella serological flagellar (H) test (O and H antisera were formation of attaching and effacing lesions on intestinal purchased from the Institute of Epidemiological Diagnosis and epithelial cells (35). It contains the eae gene encoding the Reference of the Secretariat of Health, Me´xico D.F., Mexico). outer membrane adhesin intimin, which mediates tight Salmonella isolates were streaked onto tryptic soy agar, incubated contact between STEC or EPEC and intestinal epithelial at 35uC for 24 h, and maintained at 3 to 5uC. Salmonella isolates were sent to the Institute of Epidemiological Diagnosis and cells (27). Although the LEE island is carried by STEC Reference for serotype identification. strains, its presence is not essential for pathogenesis, and some eae (LEE)–negative strains can cause hemolytic- Statistical analysis. The Pearson correlation coefficient was uremic syndrome and occasional outbreaks (2, 18, 29, 35, used to compare relationships between the presence of CB, FC, E. 43). These eae-negative STEC strains are postulated to have 612 RANGEL-VARGAS ET AL. J. Food Prot., Vol. 78, No. 3

TABLE 2. Diarrheagenic E. coli pathotypes, loci associated with harvest) or secondary contamination (during washing, diarrheagenic E. coli, DEP concentrations, and Salmonella slicing, soaking, packaging, and preparation). However, a serotypes identified in alfalfa sprout samples sprout-associated outbreaks have largely been linked to Sample DEP/locus (DEP concn [MPN/g]) Salmonella serotype seeds contaminated with pathogenic microorganisms (36) rather than postproduction contamination. Field sources of 4 EPEC/eaeA (120) — seed contamination with pathogenic microorganisms in- 7 Non-O157 STEC/stx1 (460) — clude soil, water, wild and domestic animals, drift and 12 — Typhimurium runoff from adjacent farms, and manure (1, 38). In addition, 19 ETEC/st (210) — microorganisms may contaminate sprouts during packaging, 27 Non-O157 STEC/stx2 (3.6) — 39 — Typhimurium distribution, and sale. An additional concern is the 42 ETEC/st (75) — possibility that small producers in Mexico use nonpotable water for sprout production. Cifuentes et al. (10) have 50 Non-O157 STEC/stx1 (460) — 63 EPEC/eaeA (28) — reported that households use untreated wastewater, treated Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 77 Non-O157 STEC/stx1 (240) Typhimurium wastewater effluent from interconnected reservoirs, or 82 Non-O157 STEC/stx1, stx2 (11) — natural rainfall as irrigation water in central Mexico. Also, 89 EPEC/eaeA (1,100) — it is necessary to consider the potential of the DEPs to 94 — Enteritidis survive on seeds and grow during the sprouting of seeds. In a —, not detected; st, thermostable enterotoxin locus; eaeA, intimin effect, we have observed that EAEC, non-O157 STEC, EIEC, EPEC, and ETEC strains survived at least 90 days on locus; stx1 and stx2, Shiga toxin 1 and 2 loci. mung bean seeds at 25 ¡ 2uC (21). In addition, we have observed that STEC, EIEC, ETEC, and EPEC strains all other putative adherence- and virulence-associated factors, grew during the germination and sprouting of alfalfa seeds, which include STEC agglutinating adhesin (Saa) and reaching counts of approximately 5 and 6 log CFU/g after subtilase cytotoxin (SubAB) (43). Many STEC strains also 1 day at 20 and 30uC, respectively (22). However, the fact produce enterohemolysin, encoded by the plasmid-borne that DEP and E. coli were recovered at low levels would ehxA gene, but its role in pathogenesis remains uncertain suggest postharvest contamination of sprouts. (11, 17). Considering the uncertainties regarding the role of No correlation existed between CB and DEPs or stx subtypes and putative virulence factors in STEC between CB and Salmonella. Neither FC nor E. coli pathogenesis, it is not clear whether many of the eae- correlated with Salmonella in the sprout samples. FC and negative STEC strains can cause severe human illness. E. coli have been used as indicators of fecal contamination EPEC is a leading cause of diarrhea in developed of food for many years. Nevertheless, the value of the fecal countries (16, 47). Typical EPEC strains contain both eaeA coliform assay as a fecal contamination indicator is nullified and bfp, while atypical EPEC strains contain only eaeA (26). when nonfecal-origin bacteria are the principal microbes Unlike typical EPEC strains, which are found only in detected by the assay (37, 39). On the other hand, in this humans, atypical EPEC strains have been isolated from a study, Salmonella was isolated in the absence of E. coli in variety of animal species (12). ETEC represent another major one sprout sample. This coincides with our previous reports cause of diarrhea in developing countries and in visitors from about isolation of Salmonella from mung bean sprout regions where ETEC is not endemic (traveler’s diarrhea) (35). samples that tested negative for E. coli (9) and with reports ETEC are of human origin and, hence, are indicators of of Salmonella being isolated from different foods that tested contamination from fecal sources. They are usually transmit- negative for E. coli (14, 20, 40, 45). Many researchers agree ted by contaminated food. The ETEC pathotype has been on the need to expand the search for pathogenic microor- isolated from mung bean sprouts (9) and carrot juice (46). ganisms beyond just FC and E. coli to include species such In this study, Salmonella was detected in 4% of the as Salmonella. Our results suggest that CB, FC, and E. coli alfalfa sprout samples and DEPs (ETEC, EPEC, or STEC) are not adequate indicators of the possible presence of in 10% (Table 1). This is in agreement with our previous Salmonella in alfalfa sprouts. reports about the isolation of Salmonella (1.1%) in alfalfa In addition, the results indicate that alfalfa sprouts in sprouts from supermarkets of a different Mexican city Pachuca, Mexico, pose a potential risk of foodborne illness (Queretaro City) (6) and with the isolation of Salmonella in the local population and visitors. This because some (5%) and DEPs (10%) from mung bean sprouts in markets alfalfa samples were contaminated with Salmonella and of Pachuca City (9). The Salmonella serotypes identified DEPs. The E. coli pathotypes found in the alfalfa sprout included Salmonella enterica serotype Typhimurium in samples have pathogenic potential. The minimal infective three samples and Salmonella enterica serotype Enteritidis dose is between 105 and 109 CFU for ETEC and EPEC and in one (Table 2). Both STEC and Salmonella Typhimurium between 1 and 100 CFU for STEC (35). Although the were identified in one sample (Table 2). The Salmonella infective dose for ETEC and EPEC is high, this level could serotypes isolated from sprout samples have been associated be reached in 100 g of some positive alfalfa sprout samples with many cases of human illness and have been isolated (1,100 MPN per g | 100 g ~ 105) (Table 2). On the other from different foods in Mexico (24). hand, the concentrations of STEC per gram in positive Bacterial pathogens can contaminate fresh vegetables alfalfa sprout samples were within the minimal infective by primary contamination (while growing and during dose (Table 2). J. Food Prot., Vol. 78, No. 3 DIARRHEAGENIC E. COLI PATHOTYPES AND SALMONELLA IN SPROUTS 613

Prevention and mitigation of the risk associated with 12. Cortes, C., R. De la Fuente, J. Blanco, M. Blanco, J. E. Blanco, G. DEP in sprouts requires the incorporation and consistent Dhabi, A. Mora, P. Justel, A. Contreras, A. Sanchez, J. C. Corrales, and J. A. Orden. 2005. Serotypes, virulence genes and intimin types application of good agricultural practices and good of verotoxin-producing Escherichia coli and enteropathogenic E. coli manufacturing practices throughout the sprout production isolated from healthy dairy goats in Spain. Vet. Microbiol. 110:67– process, from crop to harvest to retailer (36). Proper sprout 76. handling and processing practices must be promoted 13. Dog˘an-Halkman, H. B., I. C¸ akir, F. Keven, R. W. Worobo, and A. K. and implemented by both Mexican sprout growers and Halkman. 2003. Relationship among fecal coliforms and Escherichia consumers (36). coli in various foods. Eur. Food Res. Technol. 216:331–334. 14. Endley, S., L. Lu, E. Vega, M. E. Hume, and S. D. Pillai. 2003. Male- Finally, although the number of alfalfa sprout samples specific coliphages as an additional fecal contamination indicator for analyzed in the current study may be considered small, screening fresh carrots. J. Food Prot. 66:88–93. sufficient information was obtained to conclude that the 15. Estrada-Garcı´a, T., J. F. Cerna, L. Paheco-Gil, R. F. Vela´zquez, T. J. incidence of DEPs and Salmonella in alfalfa sprouts Ochoa, J. Torres, and H. L. DuPont. 2005. Drug-resistant represents a potential public health hazard in Pachuca, diarrheogenic Escherichia coli, Mexico. Emerg. Infect. Dis. 11: 1306–1308. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 Mexico. 16. Estrada-Garcı´a, T., C. Lopez-Saucedo, R. Thompson-Bonilla, M. Abonce, D. Lopez-Hernandez, J. I. Santos, J. L. Rosado, H. L. ACKNOWLEDGMENT DuPont, and K. Z. Long. 2009. Association of diarrheagenic Escherichia coli pathotypes with infection and diarrhea among This work was supported by Programa de Mejoramiento del Mexican children and association of atypical enteropathogenic E. coli Profesorado de la Secretaria de Educacio´n Pu´blica (PROMEP-SEP), with acute diarrhea. J. Clin. Microbiol. 47:93–98. Mexico, grant 2013 (Redes Tema´ticas de Colaboracio´n, red de inocuidad alimentaria). 17. Ethelberg, S., K. E. Olsen, F. Scheutz, C. Jensen, P. Schiellerup, J. Enberg, A. M. Petersen, B. Olesen, P. Gerner-Smidt, and K. Mølbak. 2004. Virulence factors for hemolytic uremic syndrome, Denmark. REFERENCES Emerg. Infect. Dis. 10:842–847. 1. Beuchat, L. R. 1996. Pathogenic microorganisms associated with 18. Feng, P., S. D. Weagant, and S. R. Monday. 2001. Genetic analysis fresh produce. J. Food Prot. 59:204–216. for virulence factors in Escherichia coli O104:H21 that was 2. Bonnet, R., B. Souweine, G. Gauthier, C. Rich, V. Livrelli, J. Sirot, implicated in an outbreak of hemorrhagic colitis. J. Clin. Microbiol. B. Joly, and C. Forestier. 1998. Non-O157:H7 Stx2-producing 39:24–28. Escherichia coli strains associated with sporadic cases of hemolytic- 19. Fett, W. F. 2005. Interventions to ensure the microbiological safety of uremic syndrome in adults. J. Clin. Microbiol. 36:1777–1780. sprouts, p. 187–210. In G. M. Sapers, J. R. Gorny, and A. E. Yousef 3. Buchholz, U., H. Bernard, D. Werber, M. M. Bohmer, C. (ed.), Microbiology of fruit and vegetables. CRC Press, Boca Raton, FL. Remschmidt, H. Wilking, Y. Delere, M. An der Heiden, C. Adlhoch, 20. Garcia-Villanova Ruiz, B., R. Galvez Vargas, and R. Garcia- J. Dreesman, J. Ehlers, S. Ethelberg, M. Faber, C. Frank, G. Fricke, Villanova. 1987. Contamination of fresh vegetables during cultiva- M. Greiner, M. Hohle, S. Ivarsson, U. Jark, M. Kirchner, J. Koch, G. tion and marketing. Int. J. Food Microbiol. 4:285–291. Krause, P. Luber, B. Rosner, K. Stark, and M. Kuhne. 2011. German 21. Go´mez-Aldapa, C. A., E. Rangel-Vargas, H. Bautista-De Leo´n, Ma. outbreak of Escherichia coli O104:H4 associated with sprouts. N. E. Va´zquez-Barrios, A. J. Gordillo-Martı´nez, and J. Castro-Rosas. Engl. J. Med. 365:1763–1770. 2013. Behavior of enteroaggregative Escherichia coli, non-O157- 4. Caprioli, A., S. Morabito, H. Brugere, and E. Oswald. 2005. shiga toxin-producing E. coli, enteroinvasive E. coli, enteropatho- Enterohaemorrhagic Escherichia coli: emerging issues on virulence genic E. coli and enterotoxigenic E. coli strains on mung bean seeds and modes of transmission. Vet. Res. 36:289–311. and sprout. Int. J. Food Microbiol. 166:364–368. 5. Castro-Rosas, J., J. F. Cerna-Corte´s, E. Me´ndez-Reyes, D. Lo´pez- 22. Go´mez-Aldapa, C. A., E. Rangel-Vargas, M. R. Torres-Vitela, A. Herna´ndez, C. A. Go´mez-Aldapa, and T. Estrada-Garcı´a. 2012. Presence Villarruel-Lo´pez, and J. Castro-Rosas. 2013. Behavior of non-O157 of faecal coliforms, Escherichia coli and diarrheagenic E. coli pathotypes Shiga toxin-producing Escherichia coli, enteroinvasive E. coli, in ready-to-eat salads, from an area where crops are irrigated with enteropathogenic E. coli, and enterotoxigenic E. coli strains on untreated sewage water. Int. J. Food Microbiol. 156:176–180. alfalfa sprouts. J. Food Prot. 76:1429–1433. 6. Castro-Rosas, J., and E. F. Escartı´n. 1999. Incidence and germicide 23. Go´mez-Aldapa, C. A., M. R. Torres-Vitela, O. A. Acevedo-Sandoval, E. sensitivity of Salmonella typhi and Vibrio cholerae O1 in alfalfa Rangel-Vargas, A. Villarruel-Lo´pez, and J. Castro-Rosas. 2013. sprouts. J. Food Saf. 19:137–146. Presence of Shiga toxin-producing Escherichia coli, enteroinvasive E. 7. Castro-Rosas, J., and E. F. Escartı´n. 2000. Survival and growth of coli, enteropathogenic E. coli, and enterotoxigenic E. coli on tomatoes Vibrio cholera O1, Salmonella typhi and Escherichia coli O157:H7 from public markets in Mexico. J. Food Prot. 76:1621–1625. in alfalfa sprouts. J. Food Sci. 65:162–165. 24. Gutie´rrez-Cogco, L., E. Montiel-Va´zquez, P. Aguilera-Pe´rez, and 8. Centers for Disease Control and Prevention. 2012. Multistate foodborne M. C. Gonza´lez-Andrade. 2000. Serotipos de Salmonella identifica- outbreaks. Available at: http://wwwn.cdc.gov/foodborneoutbreaks/ dos en los servicios de salud de Me´xico. Salud Pu´blica Me´x. 42:490– Default.aspx. Accessed 3 January 2015. 495. 9. Cerna-Cortes, J. F., C. A. Go´mez-Aldapa, E. Rangel-Vargas, E. 25. Health Canada. 2012. Policy on managing health risk associated with the Ramı´rez-Cruz, and J. Castro-Rosas. 2013. Presence of indicator consumption of sprouted seeds and beans. Available at: http://www.hc-sc. bacteria, Salmonella and diarrheagenic Escherichia coli pathotypes gc.ca/fn-an/alt_formats/hpfb-dgpsa/pdf/legislation/sprouts_pol_pousses- on mung bean sprouts from public markets in Pachuca, Mexico. Food eng.pdf. Accessed 3 January 2015. Control 31:280–283. 26. Hernandes, R. T., W. P. Elias, M. A. Vieira, and T. A. Gomes. 2009. 10. Cifuentes, E., U. Blementhal, G. Ruiz-Palacios, S. Bennett, and M. An overview of atypical enteropathogenic Escherichia coli. FEMS Quigley. 2000. Health risks in agricultural villages practicing Microbiol. Lett. 297:137–149. wastewater irrigation in central Mexico: perspectives for protection. 27. Ito, K., M. Iida, M. Yamazaki, K. Moriya, S. Moroishi, J. Schriftenr. Ver. Wasser. Boden. Lufthyg. 105:249–256. Yatsuyanagi, T. Kurazono, N. Hiruta, and O. A. Ratchtrachenchai. 11. Cookson, A. L., J. Bennett, F. Thomson-Carter, and G. T. Attwood. 2007. Intimin types determined by heteroduplex mobility assay of 2007. Molecular subtyping and genetic analysis of the enterohemo- intimin gene (eae)-positive Escherichia coli strains. J. Clin. lysin gene (ehxA) from Shiga toxin-producing Escherichia coli and Microbiol. 45:1038–1041. atypical enteropathogenic E. coli. Appl. Environ. Microbiol. 73: 28. Kaper, J. B., J. P. Nataro, and H. L. Mobley. 2004. Pathogenic 6360–6369. Escherichia coli. Nat. Rev. Microbiol. 2:123–140. 614 RANGEL-VARGAS ET AL. J. Food Prot., Vol. 78, No. 3

29. Karch, H., P. I. Tarr, and M. Bielaszewska. 2005. Enterohaemor- 40. Orozco, L., L. Rico-Romero, and E. F. Escartı´n. 2008. Microbiolog- rhagic Escherichia coli in human medicine. Int. J. Med. Microbiol. ical profile of greenhouses in a farm producing hydroponic tomatoes. 295:405–418. J. Food Prot. 71:60–65. 30. Kuhnert, P., P. Boerlin, and J. Frey. 2000. Target genes for virulence 41. Paniagua, G. L., E. Monroy, O. Garcı´a-Gonza´lez, J. Alonso, E. assessment of Escherichia coli isolates from water, food and the Negrete, and S. Vaca. 2007. Two or more enteropathogens are environment. FEMS Microbiol. Rev. 24:107–117. associated with diarrhoea in Mexican children. Ann. Clin. Microbiol. 31. Louise, C. B., S. A. Kaye, B. Boyd, C. A. Lingwood, and T. G. Antimicrob. 6:17. Obrig. 1995. Shiga toxin associated hemolytic uremic syndrome: 42. Paredes-Paredes, M., P. C. Okhuysen, J. Flores, J. A. Mohamed, R. S. effect of sodium butyrate on sensitivity of human umbilical vein Padda, A. Gonzalez-Estrada, C. A. Haley, L. G. Carlin, P. Nair, and endothelial cells to Shiga toxin. Infect. Immun. 63:2766–2769. H. L. DuPont. 2011. Seasonality of diarrheagenic Escherichia coli 32. Merson,M.H.,G.K.Morris,D.A.Sack,J.G.Wells,J.C.Feeley, pathotypes in the US students acquiring diarrhea in Mexico. J. Travel R. B. Sack, W. B. Creech, A. Z. Kapikian, and E. J. Gangarosa. 1976. Med. 18:121–125. Travelers’ diarrhea in Mexico. A prospective study of physicians and 43. Paton, A. W., M. C. Woodrow, R. M. Doyle, J. A. Lanser, and J. C. family members attending a congress. N. Engl. J. Med. 294:1299–1305. Paton. 1999. Molecular characterization of a Shiga-toxigenic 33. Michino, H., K. Araki, S. Minami, S. Takaya, N. Sakai, M. Miyazaki, Escherichia coli O113:H21 strain lacking eae responsible for a

A. Ono, and H. Yanagawa. 1999. Massive outbreak of Escherichia cluster of cases of hemolytic-uremic syndrome. J. Clin. Microbiol. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/3/609/1683680/0362-028x_jfp-14-229.pdf by guest on 29 September 2021 coli O157:H7 infection in school children in Sakai City, Japan, 37:3357–3361. associated with consumption of white radish sprouts. Am. J. 44. Patterson, J. E., and M. J. Woodburn. 1980. Klebsiella and other Epidemiol. 150:787–796. bacteria on alfalfa and bean sprouts at the retail level. J. Food Sci. 45: 34. Nagy, B., and P. Z. Fekete. 1999. Enterotoxigenic Escherichia coli 492–495. (ETEC) in farm animals. Vet. Res. 30:259–284. 45. Sagoo, S. K., C. L. Little, L. Ward, I. A. Gillespie, and R. T. Mitchell. 35. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. 2003. Microbiological study of ready-to-eat salad vegetables from Clin. Microbiol. Rev. 11:142–201. retail establishments uncovers a national outbreak of salmonellosis. J. 36. National Advisory Committee on Microbiological Criteria for Foods. Food Prot. 66:403–409. 1999. Microbiological safety evaluations and recommendations on sprouted seeds. Int. J. Food Microbiol. 52:123–153. 46. Torres-Vitela, M. D. R., C. A. Go´mez Aldapa, J. F. Cerna-Cortes, A. 37. National Advisory Committee on Microbiological Criteria for Foods. Villarruel-Lo´pez, E. Rangel-Vargas, and J. Castro-Rosas. 2013. 1999. Microbiological safety evaluations and recommendations on Presence of indicator bacteria, diarrheagenic E. coli pathotypes and fresh produce. Food Control 10:117–143. Salmonella in fresh carrot juice from Mexican restaurants. Lett. Appl. 38. Natvig, E. E., S. C. Ingham, B. H. Ingham, L. R. Cooperband, and Microbiol. 56:180–185. T. R. Roper. 2002. Salmonella enterica serovar Typhimurium and 47. Trabulsi, L. R., R. Keller, and T. A. T. Gomes. 2002. Typical and Escherichia coli contamination of root and leaf vegetables grown in atypical enteropathogenic Escherichia coli. Emerg. Infect. Dis. 8: soils with incorporated bovine manure. Appl. Environ. Microbiol. 68: 508–513. 2737–2744. 48. U.S. Food and Drug Administration. 2013. Enumeration of 39. Nguyen-the, C., and F. Carlin. 2000. Fresh and processed vegetables, Escherichia coli and the coliform bacteria, chap 4. In Bacteriological p. 620–684. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould analytical manual. Available at: http://www.fda.gov/Food/Food (ed.), The microbial safety and quality of food, Aspen Publishers, ScienceResearch/LaboratoryMethods/ucm064948.htm. Accessed 17 Gaithersburg, MD. November 2014.