Survival of Salmonella Enteritidis and Escherichia coli in cactus cladodes under domestic marketing conditions in Mexico

1 Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias , Campo Experimental Bajío. Km. 6.5 Carretera Celaya-San Miguel de Allende, C.P. 38110, Celaya, Guanajuato, México 2 Centro de Investigaciones Regionales "Dr. Hideyo Noguchi", Universidad Autónoma de Yucatán. Mérida, Yucatán Avenida Itzáes # 490 x Calle 59, Colonia Centro C.P. 97000 Mérida, Yucatán, Mexico. 3 Universidad Autónoma Metropolitana. Depto. de Biotecnología, Laboratorio de Enología y Alimentos Fermentados. México, D.F. México. Av. San Rafael Atlixco No. 186, Col. Vicentina, Del. Iztapalapa. C.P. 09340. México, D.F. México. 4 Colegio de Postgraduados, Campus Montecillo, Colegio de Postgraduados, Carretera México-Texcoco, Km. 36.5, Texcoco, México.


INTRODUCTION
Cactus cladodes (Opuntia ficus-indica) are a crop of great economic importance for Mexico; the species is cultivated in an area of 12,038 ha, with a production value of $1,617.645MXN (SIAP, 2015).Most of the fresh cactus cladodes are sold with spines in domestic markets; their storage time is longer than for cladodes without spines, where periods of commercialization are about three to five days (Valencia-Sandoval et al., 2010).Although, there is no information on outbreaks of foodborne illnesses associated with the consumption of cactus cladodes, some bad practices in the handling of this vegetable, especially in storage and elimination of spines, can involve the risk of product contamination by foodborne pathogens (Angeles-Núñez et al., 2014).The survival and growth of these pathogens in other vegetables have been associated with temperature, time and presentation of product (Corbo et al., 2005).
Foodborne pathogens have mechanisms to protect themselves within the plant and continue their proliferation.Studies have demonstrated the ability of Salmonella to survive in cactus leaves through the formation of biofilms after 24 h of incubation (De los Santos et al., 2012).This bacterium was found in the tissue of the cactus leaf cladodes, persisting for up to 14 days at room temperature (Landa-Salgado et al., 2013).Particularly, Salmonella was able to survive and proliferate under refrigerated storage conditions, increasing its population up to 3 log CFU at 4°C for periods longer than three weeks (Kroupitski et al., 2009;Liao et al., 2010); while Escherichia coli increased its population up to 5 log CFU at < 8°C for more than three weeks (Corbo et al., 2005, Liao et al., 2010).
Some authors have suggested that the survival of pathogens depends on the availability of nutrients in foods and the presence of secondary metabolites that may inhibit the survival of bacteria.Pad extracts of Nopalea cochenillifera have flavonoids and tannins that inhibit the growth of E. coli and Salmonella Typhimurium (Gómez-Flores et al., 2006).According to some studies with cladodes of O. ficus indica, the presence of protocatechuic, gallic, 4-hydroxybenzoic feluric, chlorogenic, syringic and sinapic acids and the epicatechin and quercetin flavonoids have been detected (Qiu et al., 2003;Guevara-Figueroa et al., 2010).These phenolic compounds have antimicrobial action through enzymatic inhibition processes and protein transport; some compounds and destabilization of cell membranes have the ability to inhibit biofilm formation (Othman et al., 2010).
In order to determine the survival ability of S. Enteritidis and E. coli in cactus cladodia under temperature conditions associated with marketing in Mexico, both bacteria were inoculated into cactus leaves with spines and without spines to assess survival for 16 days under two temperatures, refrigeration (4°C) and environmental (18°C).Also, the presence of phenolic compounds in cladodes with antagonistic potential for these bacteria was determined.The results of this study demonstrated higher survival of S. Enteritidis and E. coli on cactus cladodes at 18°C, suggesting the importance of refrigerated storage during commercialization to reduce the risk of the growth of foodborne pathogens.Additionally, the low survival of pathogens in cactus without spines suggests an antimicrobial effect provided by the leakage of phenolic compounds from plant tissue due to the peeling process.

Inoculum
The S. Enteritidis isolate (C-4153) w as obtained from the bacterial culture collection of the

Cactus cladode preparation and inoculation
Cactus cladores of var.Atlixco w ere purchased from a local market (State of Mexico), and carried to the laboratory in a cooler box.Cactus cladodes w ere immersed in 1% NaClO solution for tw o minutes to disinfect them, and then dried on a disinfected surface.The cactus cladodes w ere divided into tw o groups.In the first, spines w ere removed using sterile gloves and knives, w hile the second group w as preserved w ith spines.Later, cladodes w ith and w ithout spines w ere stored at 4 or 18°C; 4 lots w ith 36 cladodes each w ere used.Circles of 2 cm in diameter w ere inoculated w ith bacterial solutions.Cladodes w ere packed individually in plastic zipper bags.Survival evaluations of S. Enteritidis and E. coli w ere made by triplicate on days 0, 3, 6, 8, 10, 12, 14 and 16.

Bacterial enum eration
The inoculated area of the cactus cladode w as cut and diluted w ith 50 mL of sterile peptone w ater (0.8%) in sterile plastic bags and homogenized w ith a stomacher for 1 min.Serial dilutions w ere prepared, the bacterial dilution 10 -5 w as inoculated (1 mL) onto a Petri dish containing specific media.Hektoen Enteric Agar (Difco, BBL) w as used for identification of S. Enteritidis and Agar Eosinmethylene blue (Merck) for E. coli.Incubation w as carried out at 37°C for 24 h.Colony forming units (CFU) w ere enumerated.The results w ere expressed in log CFU mL -1 .

Extraction of phenolic com pounds for HPLC analysis
Physiologically mature cladode tissue (18-20 cm long) w as used to prepare extracts by the method of conventional extraction (reflux distillation).Different conditions w ere evaluated to determine the optimal conditions for extraction of total phenols: amount of tissue *Corresponding author.E-mail: martinez.talina@inifap.gob.mx.
Author(s) agree that this article remains permanently open access under the terms of the Creativ e Commons Attribution License 4.0 International License   (2, 6 and 10 g), 30 mL of solvent (w ater, methanol and ethanol), extraction temperature (40 and 60°C) and time (1, 2, 3, 4 and 5 h).Quantification of total phenols w as performed in triplicate by the Folin-Ciocalteu method (Kuskoski et al., 2005) and expressed in terms of equivalent amounts of gallic acid.

HPLC analysis of phenolic com pounds
The phenolic compounds in the extracts w ere determined by HPLC using a UV detector (Thermo Separations Products, USA) at 280 nm (Agilent 1100 series, Hew lett Packard Co., USA).The separation w as conducted in an Alltech Lichrosorb C18 column (250 x 4.6 mm), all extracts and solvents w ere filtered through a 0.47 µm filter (Varian) prior to analysis.In accordance w ith the methodology of Ndhlala et al. (2007), tw o mobile phases w e used: A: w ater : acetic acid (98:2 v/v) and B: w ater : acetonitrile : acetic acid (68:30:2).The flow rate w as 2 mL min -1 and 20 μL of each sample w ere injected.Standard solutions (0.02 mg mL -1 ) of gallic, protocatechuic, 4-hydroxybenzoic, caffeic, feluric, chlorogenic, syringic, ρ-coumaric and sinapic acids, and (-) epicatechin and quercetin w ere dissolved using methanol as the solv ent (HPLC degree).All phenolics w ere identified by comparing the UV spectral properties and retention times to those of authentic standards.

Experim ental design and statistical analysis
Tw o statistical test w ere performed, a 2x2 factorial design w as used to determine the influence of the cladode cactus presentation (w ithout spines and w ith spines), and temperature storage (4 and 18°C) on the survival of S. Enteritidis and E. coli every other day for 16 days (three repetitions in three individuals each time), an analysis of variance for repeated measures w as used (p≤0.05) and Tukey Mean Difference tests.A 3x3x2x6 factorial design w as used to determine the optimal conditions for the extraction of phenolic compounds, using ANOVA (p≤0.05) and Tukey mean difference tests.The SAS 9.1 program w as used to perform the analyses.

Evaluation of S. Enteritidis and E. coli survival
The populations (CFU) of S. Enteritidis and E. coli on cactus cladodes stored at 4°C were lower than at 18°C; the minimal survival was on cactus cladodes without spines stored at 4°C.The analysis of variance showed interaction between storage time, temperature and presentation of cactus cladodes in the survival of S.
Enteritidis and E. coli (Table 1).The maximum growth of S. Enteritidis was observed at 10 days after inoculation; the population of Salmonella in cactus cladodes without spines at 4°C declined with respect to the initial concentration 0.15 log CFU after 3 days of storage and 1.5 log CFU at final storage (16 days) (Figure 1).In cactus cladodes with spines, the population of S.
Enteritidis increased approximately 0.21 log CFU from 3 to 10 days, and declined 1.2 log CFU at the end of storage (Figure 2).S. Enteritidis was able to survive in the storage times and temperatures of domestic marketing in cactus cladodes with and without spines.The ability to survive at 4°C has been observed from nine days to eight weeks of storage; in this time, the populations declined approximately 0.5 to 2 log CFU (Liao et al., 2010).Salmonella has been able to survive at temperatures lower than 4°C.Strawn and Dayluk (2010) demonstrated viability of the bacteria in papaya and mango after 180 days at -20°C, and Kimber et al. (2012) found Salmonella on almonds and pistachios stored at -19 and 4°C at least one year after inoculation.In contrast, at 18°C, S. Enteritidis increased by 0.45 log CFU after 3 to 14 days in both presentations of cactus cladodes (Figures 1 to 2).Liao et al. (2010) observed a similar situation in jalapeño pepper, where populations of Salmonella Saintpaul increased around 3 log CFU at 20°C in just 48 h.E. coli presented a similar behavior to S. Enteritidis; the population of E. coli decreased significantly (p <0.001) at 4°C in both presentations of cactus cladodes.However, the population of this bacterium decreased to 0.94 log CFU at 3 days, and viable cells were not found at 16 days on cactus cladodes without spines (Figure 3).On cladodes with spines, the population increased by 0.3 log CFU on days 3 and 5, followed by a gradual decrease:  1.06 log CFU was detected on day 16, nearly the initial concentration (Figure 4).At 18°C, in cactus cladodes without spines and with spines, the bacterial population had increased at day 3; maximum growth was found on day 6, after this time the population decreased considerably, with remnants of 2.20 and 2.53 log CFU without and with spines, respectively, on day 16.The survival ability of E. coli on cactus cladodes has not been documented; however, on fruits of prickly pear without peel, E. coli O157: H7 was able to survive and increase its population to 4.5 and 5 log CFU g -1 in storage at 4 and 8°C, respectively (Corbo et al., 2005).In other vegetables at <8°C, this bacterium showed decreased growth but was able to survive (Khalil and Frank, 2010;Liao et al., 2010;Corbo et al., 2005).The ability of both pathogens to survive at the same time was studied by Hsu et al. (2006) at 4°C on aromatic herbs; the populations of both bacteria decreased to about <0.8 log at 5 days of  storage; however, E. coli O157: H7 decreased rapidly over time; bacteria were detected even after 24 days on rotting tissue.Other studies reported that the ability to survive was related to type of tissue.Khalil and Frank (2010) observed major growth on spinach leaves at 8°C (1.18 log CFU), while in lettuce, cilantro and parsley leaves, the bacterium did not grow at 8°C.With the information obtained in this study on  temperature and time of survival, there is a latent risk of the presence of foodborne pathogens in cactus cladodes, particularly when they are transported or kept in poor hygienic conditions or unrefrigerated, and consumed fresh in salads and juices.The presentation of cactus cladodes during storage was another important factor in the survival of both bacteria; survival was significantly greater (p≤0.001) in cactus cladodes with spines.The surface wax of the cladodes controls the transpiration and reflects solar radiation, and prevents the penetration of microorganisms into the surface tissue.The bacteria probably produce biofilms that allow them to survive on this wax.Some bacteria have the ability to form biofilms on the surface of the epidermis of fruits, leaves, stems and flower organs, as an adhesion and protection mechanism, also to trap nutrients for feeding (Ávila-Quezada et al., 2010).Hernandez et al. (2009) documented biofilm formation by Salmonella Typhimurium and S. Javanica 24 h after inoculation; their results showed faster adhesion.Few studies have been conducted to estimate the survival of both bacteria on the surface of this vegetable; however, in other products such as cucumber, mango, guava and tomato, S. Enteritidis was capable of developing biofilms on surfaces after inoculation (Tang et al., 2012).If these bacteria are established on the cuticle of cactus cladodes with biofilms, there is a risk of internalization of foodborne pathogens into tissue during the removal of the spines, which would remain there until consumption.The survival of S. Enteritidis and E. coli in spineless cactus cladode was significantly less at 4°C; this behavior is probably attributable to the decrease in bacterial growth influenced by the temperature; however, the hypothesis of liberation of metabolites as result of mechanical damage in the spine removal process should not be rejected.The production of these compounds is activated as a defense mechanism against a wide variety of microorganisms and increases their survival.The presence of metabolites with microbiological properties has been documented in several species of the genus Opuntia.Some extracts of O. cochenillifera (Syn.: Nopalea cochenillifera) showed in vitro inhibition of the growth of Candida albicans, Candida glabrata, E. coli, Salmonella Typhimurium, S. Typhi, Micrococcus sp., Klebsiella pneumoniae, Staphylococcus aureus, Saccharomyces cerevisiae and others (Gomez-Flores et al., 2006;Necchi et al., 2012).Also, extracts of O. stricta presented antimicrobial activity against S. aureus, E. coli, C. albicans, Bacillus sp., Pseudomonas aeruginosa and Enterococcus faecalis (Koubaa et al., 2015), while extracts of O. ficus-indica presented bactericidal activity against Campylobacter jejuni, Campylobacter coli, S. aureus, E. coli, P. aeruginosa, K. pneumoniae, Proteus mirabilis, Salmonella spp., E. faecalis, Citrobacter freundii, Acinetobacter baumannii, Streptococcus pneumoniae, Enterococcus faecium and Enterobacter cloacae (Wasnik and Tumane, 2016).Hayek and Ibrahim (2012) determined the antimicrobial potential of Opuntia matudae (xoconostle) against E. coli O157: H7.The authors attributed this inhibition to organic acids and polyphenols, especially flavonoids and tannins.Phenolic compounds caused cell wall degradation and disruption of the cytoplasmic membrane (Cetin-Karaca and Newman, 2015).Other mechanisms are enzyme inactivation, inhibition of DNA and RNA synthesis, electron transport chain and biofilm formation, and neutralization of toxins (Gutiérrez-Larraínz ar et al., 2012).

Determination of phenolic compounds in cladodes of cactus O. ficus-indica var. Atlixco
The comparison of means showed water as the best solvent to extract phenolic compounds.However, this solvent yielded an excessive amount of mucilage which obstructed the passage of the sample in the HPLC columns; therefore, the second option was methanol as solvent.The phenolic compound extraction used 10 g of tissue, solvent methanol, extraction time and temperature of 60°C for 6 h (Table 2).

Conclusions
The effect of temperature and presentation of cactus cladodes was significant on the survival of S Enteritidis and E. coli.S. Enteritidis and E. coli were able to grow and survive for 16 days at 4 and 18°C in cactus leaves with spines.In cactus leaves without spines, S. Enteritidis survived for 16 days at 4 and 18°C, while E. coli only survived during this period at 18°C.Opuntia ficus-indica (L.) Mill var.Atlixco presents phenolic compounds with antimicrobial potential that could reduce the pathogenicity of S. Enteritidis and E. coli associated with consumption of fresh cactus.
spines, same letters are not significantly different.

Figure 4 .
Figure 4. Survival of E. coli in cactus cladodes w ith spines stored at 4 and 18°C.Bars denote standard deviation, n = 9.

Table 1 .
Comparison of means of survival of Salmonella Enteritidis and E. coli in cactus leaf w ith and w ithout spines to 18 and 4°C.

Table 2 .
Comparison of means from the content of total phenols of aqueous and methanolic extracts.The means of total phenolics are show n according to statistical analysis.

Table 3 .
Concentration of phenolic compounds analyzed by HPLC in extracts of cactus cladodes var.'Atlixco'.