Food Microbiology 46 (2015) 383e394

Contents lists available at ScienceDirect

Food Microbiology journal homepage: www.elsevier.com/locate/fm

Mixed culture biofilms of Salmonella Typhimurium and cultivable indigenous microorganisms on lettuce show enhanced resistance of their sessile cells to cold oxygen plasma Iqbal Kabir Jahid a, b, Noori Han a, Cheng-Yi Zhang a, Sang-Do Ha a, * a b

School of Food Science and Technology, Chung-Ang University, 72e1 Nae-Ri, Daedeok-Myun, Anseong, Gyunggido 456e756, South Korea Department of Microbiology, Jessore University of Science and Technology, Jessore 7408, Bangladesh

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 March 2014 Received in revised form 4 August 2014 Accepted 11 August 2014 Available online 27 August 2014

Control of foodborne pathogens in fresh produce is crucial for food safety, and numerous Salmonella Typhimurium (ST) outbreaks have been reported already. The present study was done to assess effectiveness of cold oxygen plasma (COP) against biofilms of ST mixed with cultivable indigenous microorganisms (CIM). ST and CIM were grown at 15
C as monocultures and mixed cultures for planktonic state, biofilm on stainless steel, and lettuce leaves. Thereafter, the samples were treated with COP and surviving populations were counted using plate counting methods. Biofilms and stomatal colonization were examined using field emission scanning electron microscopy (FESEM) and food quality was assessed after treatment. Mixed cultures of ST and CIM showed an antagonistic interaction on lettuce but not on SS or in planktonic state. Mixed cultures showed significantly (p < 0.05) greater resistance to COP compared to monoculture biofilms on lettuce but not on SS or planktonic state. Shift from smooth to rugose colony type was found for planktonic and for biofilms on SS but not on lettuce for ST. Mixed culture biofilms colonized stomata on the inside as demonstrated by FESEM. Although, lettuce quality was not affected by COP, this technology has to be optimized for further development of the successful inactivation of complex multispecies biofilm structures presented by real food environment. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Salmonella Typhimurium Indigenous microorganisms Cold oxygen plasma Mixed culture biofilms Weibull model

1. Introduction Fresh produce is an important source of nutrients, vitamins, and fiber for human sustenance (Anonymous, 2013). The incidence of foodborne disease, including salmonellosis, linked to fresh produce has risen worldwide during the last few years because of high consumption, changing consumer habits, and broader worldwide distribution (FDA, 2013). In the USA alone, 1527 outbreaks were identified between 2009 and 2010, resulting in 29,444 cases of illness with 1118 hospitalizations and 23 deaths (MMWR, 2013). Salmonella spp. was the most common bacterial pathogen reported during this period, accounting for 30% of foodborne illnesses (MMWR, 2013). Salmonella enterica serovar Typhimurium (ST) is one of the most clinically important foodborne pathogens, with 6 outbreaks caused by ST in the USA in 2013 (CDC, 2013). The US

* Corresponding author. School of Food Science and Technology, Chung-Ang University, 72e1 Nae-Ri, Daedeok-Myun, Anseong, Gyunggido 456-756, South Korea. Tel.: þ82 031 670 4831; fax: þ82 031 675 4853. E-mail address: [email protected] (S.-D. Ha). http://dx.doi.org/10.1016/j.fm.2014.08.003 0740-0020/© 2014 Elsevier Ltd. All rights reserved.

Department of Agriculture Economic Research Service reported 1.4 million cases of salmonellosis in 2010, costing 2.65 billion dollars and resulting in 415 deaths (USDA-ERS, 2010). The European Food Safety Authority (EFSA, 2013) reported over 100,000 human cases of illness due to Salmonella sp., costing approximately V3 billion annually. Biofilms are a 3-dimensional sessile community of microorganisms attached to a solid surface shielded inside an endogenously produced extracellular polymeric substance (EPS) secreted by the colony, representing a common mode of microorganism growth in natural and industrial settings (Agle, 2007). Biofilm production on the surface of produce is a common phenomenon under natural conditions (Rayner et al., 2004), and 10e80% of the total natural population of leaves contains biofilms (Morris et al., 1998; Lindow and Brandl, 2003). Rayner et al. (2004) also identified natural biofilms on fresh tomatoes, carrots, and mushrooms. Artificially seeded biofilms of foodborne pathogens and methods for their reduction have been extensively studied and € reviewed (Niemira and Cooke, 2010; Olmez and Temur, 2010; Patel and Sharma, 2010; Jahid and Ha, 2012; Olaimat and Holley, 2012). The results of these studies indicate that foodborne pathogens in

384

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

the form of biofilms are resistant to common disinfectants, and therefore represent a substantial food safety problem (Jahid and Ha, 2012). Most of the clinical, environmental, and food sample studies on biofilms conducted thus far have examined pure culture biofilms; however, such biofilms are uncommon in regular food. Data obtained thus far on mixed culture biofilms in food have been reviewed previously (Manuzon and Wang, 2007), and the complex interactions between cell populations forming biofilms have been discussed elsewhere (Moons et al., 2009). In nature, certain microorganisms first colonize/attach to the surface, and other microbes then follow to form biofilms (Kolenblander, 2000). Compared to monocultures on food contact surfaces, mixed culture biofilms are more resistant to common disinfectants (Behnke et al., 2011; Behnke and Camper, 2012; Kostaki et al., 2012; van der Veen and Abee, 2011; Ibusquiza et al., 2012; Jahid et al., 2014b). Currently, there are no disinfectants that can achieve a 5-log reduction of foodborne pathogen biofilms on fresh produce; this state of affairs poses a major food safety problem (Jahid and Ha, 2012). Biofilms formed by foodborne pathogens are resistant to many disinfectants, particularly biofilms consisting of mixed culture. Novel alternative methods are therefore needed to address this growing health concern. Cold oxygen plasma (COP) has been used successfully to disinfect various foods such as lettuce ndez and Thompson, 2012; Jahid et al., 2014a), mangoes (Ferna (Perni et al., 2008), melons (Perni et al., 2008), apples (Critzer et al., 2007), and cantaloupe (Critzer et al., 2007). Few studies, however, have tested the ability of COP to combat biofilms (Vleugels et al., 2005). Many bacteria, such as Vibrio cholerae (Yildiz and Schoolnik, 1999), Vibrio parahaemolyticus (Chen et al., 2010), Escherichia coli, and ST (de Rezende et al., 2005) alter their phenotypes and shift from the smooth to rugose colony type and vice versa. Shifts from smooth to rugose variants can be induced by exposure to adverse conditions such as nutrient deprivation (Ali et al., 2002), temperature (Anriany et al., 2001), and predation (Matz et al., 2005). In general, compared to the smooth type, rugose colonies form thicker biofilms and exhibit lower motility, higher resistance to chlorine, and greater exopolysaccharide formation (Yildiz and Schoolnik, 1999). Although microbial inactivation on food and food contact surfaces can easily be described using linear regression analysis, most disinfectants have inactivation kinetics that are nonlinear; i.e., microbial survival curves exhibit upward or downward concavity and a sigmoid shape (Mafart et al., 2002; Buzrul and Alpas, 2007; van Boekel, 2002). Accordingly, many authors have used nonlinear modeling methods, such as logistic models (Cole et al., 1993), the Gompertz equation (Veen and Abee, 2011), the Weibull model (Buzrul and Alpas, 2007), and the Fermi equation (Peleg, 1996). Many researchers have employed the Weibull model for analyzing the inactivation kinetics of planktonic (Buzrul and Alpas, 2007; Chun et al., 2010; van Boekel, 2002) and biofilm populations (Vaid et al., 2010), whereas others have modified the Weibull model to fit the data (equation) on foods and on the food industry (Mafart et al., 2002; Albert and Mafart, 2005; Coroller et al., 2006). GInaFiT 1.6 software is available free of charge (http://cit.kuleuven.be/ biotec/downloads.php) and is easy to use for analyses of various linear and nonlinear equations in Microsoft Excel, including the Weibull model (Geeraerd et al., 2005). This software has previously been applied to analysis of biofilms on stainless-steel (SS) surfaces (Posada-Izquierdo et al., 2013). The present study focuses on the inactivation kinetics of mixed culture ST and cultivable indigenous microorganisms on lettuce in the state of planktonic growth and in biofilms. The experiments were conducted either on SS coupons or on lettuce treated with

COP; the data were analyzed using the Weibull model in addition to assessment of a shift to the rugose phenotype. 2. Materials and methods 2.1. Experimental design Indigenous microorganisms present on the surface of lettuce were isolated from cleaned lettuce leaves, and the isolated strains were representative of cultivable indigenous microorganisms (CIM). Lettuce juice broth (LJB) was then prepared by blending and filtering using a 0.22-mm filter, and used to grow either planktonic cells or biofilms on SS coupons. Lettuce leaves were cut to the same size and background flora was removed. Bacterial inocula of CIM and ST were then prepared and seeded as monocultures and mixed cultures for either planktonic growth or biofilm formation on SS coupons and lettuce leaves. After incubation and washing, the samples were treated with COP at different exposure times, and the inactivation kinetics were determined. Smooth and rugose colonies were counted and photographed. Field emission scanning electron microscopy (FESEM) was used to examine biofilm formation on SS coupons and on lettuce leaves and stomatal colonization on lettuce leaves. The color and texture were recorded to assess food quality of the lettuce after COP treatment. The entire experimental design is outlined in Fig. 1. 2.2. Isolation of cultivable indigenous microorganisms; media and growth conditions Fresh iceberg lettuce (Lactuca sativa) was purchased from a local grocery store in Anseong, Republic of Korea on the day of harvest, and transported to the laboratory within 30 min under refrigerated conditions. The core and outer 2 layers were discarded by hand, and the internal parts were cut into 5  3 cm2 sections using a sterile scalpel. The resulting lettuce coupons were washed with sterile distilled water (SDW) to remove unattached cells, and then immersed in Dulbecco's phosphate-buffered saline (DPBS; SigmaeAldrich, Inc.; St. Louis, MO) and incubated at 15
C or 20
C for 24 h. Microorganisms were then plated on brilliant green agar (BGA; Difco; Becton Dickinson, Franklin Lakes, NJ) with 25 mg/mL of nalixic acid (NA) and 25 mg/mL of novobiocin (NO) to identify STlike colonies present in the samples. ST-negative samples were then mixed together and the solution was termed CIM. ST used in this study was obtained from Chemical Regulation and Food Safety Center, Exponent, Inc. (Bowie, MD, USA) .Ha et al., 1995) which is resistance to NO and NA. The strain was selected for the study because of its properties of resistance to both antibiotics which was used to differentiate from CIM. All the strains were preserved at 70
C in T1N1 broth (1 g tryptose/100 mL and 1 g NaCl/100 mL) containing 15% glycerol, and one cryovial was thawed and resuspended in tryptic soy broth (TSB); incubation for 24 h for 20
C at 220 rpm was used in each experiment. 2.3. Preparation of lettuce and reduction of background flora Coupons were used immediately after preparation or stored at 4
C until use. To reduce background flora, coupons were placed on sterile petri dishes and treated with COP for 5 min on both sides. Sterility of lettuce coupons was then confirmed according to procedures described in Section 2.10. The results of this procedure showed that background microflora was removed successfully by COP, with counts below the detection level of 1 log CFU/mL or cm2.

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

385

Fig. 1. Experimental design.

2.4. Preparation of LJB LJB was prepared according to procedures described previously (Shen et al., 2012), with minor modifications. Briefly, small sections of lettuce were processed and blended in a commercial blender (Model: SMX-760J; clean sense Shinil electric mixture; Korea). The mixtures were then passed through a filter bag (Nasco Whirl-pak; Fort Atkinson, WI), and centrifuged twice (10,000  g for 30 min at 5
C) to remove bigger particles. Supernatants were filtersterilized by passing them through 0.22-mm filters (Millipore Corporation; Billerica, MA, USA), and the resulting LJB was diluted to 3% with sterile water before experiments. 2.5. Preparation of SS coupons SS coupons (2  2  0.1 cm, type: 302) were initially rubbed with ethanol (70%) for 1 h to remove waste, grease, and oil and were then rinsed with distilled water. The coupons were then washed with a commercial detergent for 1 h at room temperature. After that, the coupons were rinsed in distilled water, dried for 1 h in a dry oven at 55
C, and autoclaved at 121
C for 15 min. 2.6. Preparation of inocula and the inoculate for planktonic cells Cultures from TSB were centrifuged at 10,000  g for 10 min. The pellets were washed with 10 mL DPBS and resuspended in 5 mL DPBS. The ST monoculture, CIM monoculture, and mixed cultures ((1:1 ratio) were then diluted 1:50 in 5 mL of diluted LJB (3%) and incubated at 15
C for 24 h under static conditions for planktonic growth in 50-mL Falcon tubes.

on the SS coupons under static conditions. For biofilm formation on the surface of lettuce, inocula were prepared in the same way as described and diluted with 100 mL sterile-distilled water in beakers. Three lettuce coupons were submerged in each beaker for 5 min to get the bacteria attached. Beakers were then decanted and the coupons were transferred to petri dishes and dried in a laminar flow chamber (Vision bio-safety cabinet; Korea) for 5 min. After that, the coupons were transferred to a new petri dish and incubated at 85% relative humidity (RH) and 15
C for 24 h. 2.8. The COP device Biozone photoplasma technology produces long-wavelength ultraviolet (UV) light ranging from 100 to 280 nm, producing ozone and negative ions for destruction of bacteria (Biozone Scientific; Orlando, FL The device used in our study was purchased from Biozone Scientific International Inc., USA (Photoplasma, Model: Induct, ID 60; Fig. 2). The plasma was set up with a bench-scale collimated beam in a closed chamber. The distance between samples and the producing electrode is adjustable, thus the electrode was placed above the samples. The distance between the samples and light in this study was 9.5 cm. Emitted UV light was measured using a power meter (photoradiometer, HD2102.1; Delta Ohm; Padova, Italy), and doses between 1200 and 1250 mW s/cm2 were delivered. The light bulb was warmed up for at least 30 min prior to experiments to establish stable energy output. The ozone produced was monitored using an ozone analyzer (Excellence in Instrumentation; Norwood, MA) at a distance of 9.5 cm, and measurements showed approximately 3.0 ppm.

2.7. Biofilm formation on SS and lettuce coupons For biofilm formation on SS coupons, the prepared inocula (single and mixed) were diluted 1:50 in 10 mL LJB in 50-mL Falcon tubes containing an SS coupon completely submerged in LJB. The tubes were incubated at 15
C for 72 h to form biofilms

2.9. Treatment of planktonic cultures and biofilms on SS and lettuce coupons After incubation, planktonic cells were vortexed and 500 mL of planktonic populations were pipetted into petri dishes (90 

386

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

Fig. 2. Schematic view of the cold oxygen plasma (COP) device.

12 mm) for COP treatment. SS coupons were rinsed with 100 mL sterile-distilled water twice to remove unattached cells, and then treated with COP for different periods of time. The coupons were then turned over using sterile tweezers and treated again for the same length of time. After incubation (Section 2.7 above), each lettuce coupon was rinsed in 100 mL of sterile-distilled water twice in a beaker to remove unattached cells. Then, the lettuce coupons were placed in petri dishes and treated with COP for different periods of time. The coupons were then turned over using sterile tweezers and treated for the same length of time.

2.10. Microbiological analysis COP-treated SS coupons were transferred to a small petri dish (55  12 mm) containing 2 mL of 0.1% peptone water (PW), and then scrubbed using tweezers for 10 times right clockwise and 10 times anticlockwise to touching the petri dish and transferred to test tubes for ultrasonication to dispersal the aggegated bacteria. To remove biofilms from lettuce, coupons were placed in 25 mL PW (Oxoid; Basingstoke, Hants, UK) in a sterile stomacher bag (Nasco Whirl-pak; Atkinson, WI) and homogenized using a stomacher (Bagmixer, Interscience; Saint Nom, France) at the highest speed for 1 min. Serial dilutions of each planktonic sample, homogenized biofilm from SS, and lettuce samples were made in 0.1% peptone water and thereafter plated onto nonselective and selective media. The ST colonies were counted by means of spread plating on BGA containing 25 mg/mL of NA and 25 mg/mL of NO, followed by incubation at 37
C for 48 h. The size of total bacterial populations was determined using spread plating onto the R2A medium followed by incubation at 37
C for 48 h. CIM counts were calculated by subtracting the BGA counts from R2A counts. The data were transformed to log 10 values, and survivor populations were plotted against COP treatment time.

2.11. Determination of colony morphology To determine colony morphology (rugose and smooth), colonies were grown on BGA with appropriate antibiotics and incubated at 37
C for 48 h. Colonies on plates were photographed using an Olympus Microscope (Model CX 31) equipped with a digital camera (Tucsen; China) with appropriate software for microscopy analysis. 2.12. FESEM The biofilm formation on SS and lettuce coupons were prepared as described above. The samples with both non-inoculated (control) and inoculated with monoculture and mixed cultures without treatment were analyzed by electron microscopy. The samples were then processed and visualized using FESEM (Sigma; Carl Zeiss; Germany) according to the procedures described previously (Jahid et al., 2014a). 2.13. Food quality testing: color, texture, and sensory properties The color of the adaxial side of the lettuce was measured using a colorimeter (UltraScan Pro/Hunterlab; USA) in 3 randomly selected areas in the L*a*b* mode. The values of L* (lightness, whiteeblack), a* (greenered), and b* (blueeyellow) before and after COP treatment were recorded for further analysis. Texture before and after treatment was analyzed using a texture analyzer (Stable Micro Systems TA.XT express enhanced; Texture Technologies Corporation; Scarsdale, NY) with an SMSP/2 probe. The same size and types of lettuce coupons were used for texture analysis without inoculation with ST and CIM. The detection conditions were as follows: pretest speed at 1.0 mm/s, test speed at 1.0 mm/s, posttest speed 5.0 mm/s, and return distance at 15.0 mm/s. Texture of 5 replicates was measured for each sample. Sensory characteristics of lettuce before and after 5 min of COP exposure were evaluated by 10 trained and 10 untrained panelists using a hedonic scale according to methods described previously with a few modifications (Lee and Ha, 2008). The quality was

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

387

assessed using a subjective 7-point hedonic scale as follows: 1 ¼ “dislike extremely, not acceptable,” 4 ¼ “neither like nor dislike, the lower limit of the acceptable range,” 7 ¼ “like extremely, essentially free from any noxious effects, original quality is preserved.” The parameters evaluated were color, smell, texture, and overall acceptability of the lettuce. The sensory evaluation was conducted on the same days as microbial analyses were performed and immediately after COP treatment. Prior to sample evaluation, the trained panelists participated in orientation sessions to familiarize themselves with the scale parameters (color, texture, odor, and overall perception) using the hedonic scale and quality assessment. 2.14. Non-linear regression The Weibull model (a 2-parameter nonlinear model historically used for failure engineering) in a cumulative form is described as follows:



Log

Nt N0



 ¼

1 2:303

 n t a

(1)

where a is a scale parameter (unit is min or s), n is a unitless shape parameter (van Boekel, 2002), Nt is the number of microorganisms (colony-forming units (CFU)/mL or cm2) after treatment time t, N0 is the initial number of microorganisms (CFU/mL or cm2), and t is the treatment time (duration in min). Values of n equal to 1 correspond to linear survival curves, n values >1 correspond to downward concave (shoulder) survival curves, and n values 0.05). Indigenous microorganisms were previously found to have an inhibitory effect on the planktonic state of S. enterica, Listeria monocytogenes, and E. coli O157:H7 (Liao, 2007). We did not observe this inhibitory effect and subsequent antimicrobial actions in the planktonic state probably because of interstrain variation in CIM. Recently, it was reported that 30% of indigenous microorganisms on produce plants form moderate and strong biofilms (Liu et al., 2013). Nevertheless, populations of monospecies biofilms were similar to mixed species biofilms for ST and CIM on SS. The mean counts were 6.1 log CFU/ mL and 6.3 log CFU/mL for ST monoculture and mixed culture, respectively. There were significantly lower cell numbers in mixed culture biofilms compared to monocultures on lettuce in our study (p < 0.001). The mean cell numbers on lettuce were 5.9, 4.8, 7.1, and 6.0 for ST monoculture, ST mixed culture, CIM monoculture, and CIM mixed culture biofilms, respectively (Fig. 3). Epiphytic bacteria on spinach leaves interact antagonistically with E. coli O157:H7 because of competition for nutrients, resulting in the reduced growth of pathogens because of acid production (Lopez-Velasco et al., 2012). Our data showed that mixed culture biofilms on lettuce showed an antagonistic relationship, whereas in planktonic and SS biofilms, the interaction was neutral. Planktonic populations do not compete for nutrients, space, and oxygen availability, which could explain these neutral relationships as they could move the free space in broth. A previous study showed that natural microflora on pork interacts antagonistically with Salmonella spp. at 15
C or lower temperatures (Møller et al., 2013). The competition between mixed biofilms of E. coli O157:H7 and indigenous

388

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

microorganisms of spinach leaf analyzed using functional metagenomics, showed that limited macronutrients, such as nitrogen, carbon, and phosphorus, are the primary resources under competition (Carter et al., 2012). Because lettuce leaves contain all macronutrients, indigenous microorganisms compete with pathogens in this context, but not on SS, which has no nutrients (Fig. 3). It has also been documented that CIM form significantly stronger biofilms (p < 0.05) both in monoculture and mixed culture compared to ST (Fig. 3) which might be due to CIM to degrade the lettuce leave during colonization. It is plausible to suggest that CIM have genes to degrade the lettuce leaf components such as cellulose, pectin, chitin, and lignin (Carter et al., 2012). Thus, it can be assumed that CIMs have survival advantages over ST. 3.2. Death curves of planktonic growth and biofilms on SS and lettuce coupons To analyze the interactions between indigenous microorganisms and ST, indigenous populations were mixed with ST and incubated in both the planktonic and biofilm state, and the samples were then treated with COP for different periods of time. Biofilms are known to be resistant to many common disinfectants (Jahid and Ha, 2012); thus, COP could represent a novel approach for killing biofilm-forming microorganisms on food and food contact surfaces since this method produce a mixture of reactive agents constituted of photon, electron, ions, free radicals generated at room temperature and atmospheric pressure which might be effective for ~ o, 2012). Death destruction of biofilms populations (Brelles-Marin curves were obtained for monocultures and mixed cultures of CIM and ST, fitted to the Weibull model using the GInaFiT 1.6 software, and d and p values were calculated (Fig. 4 and Table 1). Because most inactivation curves are nonlinear, several freeware tools, such as GInaFiT 1.6 software (Geeraerd et al., 2005; http://cit.kuleuven. be/biotec/downloads.php), exist to easily test the Weibull and linear models. Most recently, the software was applied to determine the survival kinetics of biofilms of E. coli and Salmonella spp. on SS coated with vegetable juices (Posada-Izquierdo et al., 2013). The inactivation kinetics of planktonic culture and biofilms on SS and lettuce coupons was also determined using this software in our study. The graphs in Fig. 4 show the size of surviving populations as a function of duration of COP treatment for monocultures and mixed cultures of planktonic cells (Fig. 4A), biofilms on SS (Fig. 4B), and biofilms on lettuce coupons (Fig. 4C). Overall, the results show more effective microbial inactivation with longer COP treatment, with little resistance in planktonic cells (Fig. 4A) and biofilms on SS (Fig. 4B), and more notable resistance to COP of biofilms on lettuce (Fig. 4C). For planktonic populations, an approximate 5-log reduction was achieved within less than 1 min of COP treatment (Fig. 4A). Initial levels of both monocultures and mixed cultures in the planktonic state were 7e8 log CFU/mL, which was reduced to 1.7, 1.62, and 1.78 log CFU/mL after 50 s of COP treatment in ST monoculture, ST mixed culture, and CIM mixed culture, respectively, and to 0.40 log CFU/mL in CIM monoculture after 20 s of COP exposure (Fig. 4A). The planktonic cells of Staphylococcus aureus, Pseudomonas aeruginosa, and Candida albicans are reduced by > 5.0 log CFU/mL after COP treatment for 20 s (Kvam et al., 2012). No significant differences were observed between the mixed culture populations and monoculture populations of planktonic cells treated with COP (p > 0.05). Previous studies also observed insignificant effects on growth of mixing of cultures in the planktonic state (Behnke et al., 2011; Behnke and Camper, 2012). Biofilms on SS coupons showed the same microbial inactivation curve trends as those observed in planktonic cells, with very little resistance to COP (Fig. 4B). In other words, initial biofilm populations of approximately 6e7 log CFU/cm2 were reduced to approximately

Fig. 4. Microbial death rate of monocultures and mixed cultures of Salmonella Typhimurium (ST) and cultivable indigenous microorganisms (CIM) incubated at 15
C and then treated with different doses of cold oxygen plasma (COP). The points represent mean ± SEM of 3 independent experiments, and vertical bars indicate the duration of COP treatment. Curves were fit to a modified Weibull model using GInaFiT 1.6 software. (A) Planktonic populations, (B) biofilms on SS coupons, and (C) biofilms on lettuce leaves.

1.0 log CFU/cm2 after treatment with COP for 40 s (Fig. 4B). Previous studies showed that COP could penetrate 15-mm-thick biofilms, thereby effectively killing all biofilm cells (Xiong et al., 2011). Our data on SS biofilms confirmed this previously reported efficacy. It was also previously reported that COP could effectively penetrate 25-mm biofilms (Pei et al., 2012). We noted that both monocultures and mixed cultures biofilms populations on SS were equally

0.98 0.97 0.97 0.98 0.02 0.08 0.04 0.03 0.02qr 0.06q 0.03rs 0.02s ± ± ± ± 0.26 0.32 0.20 0.15

Here, r is the shape parameter and d is the scale parameter of modified weibull model by Mafart et al., 2002 and analyzed by Add-in GInaFiT 1.6 (Geeraerd et al., 2005; http://cit.kuleuven.be/biotec/downloads.php) software. RMSE is root mean square error to determine the fit of the model with death rate. Tr is time required to kill 90% population calculated from Weibull model which is analogous to traditional D-value. Here the Tr for 4 log reduction. c

a

0.97 0.98 0.95 0.98 0.03 0.03 0.05 0.02 13.0c 2.8d 10.5c 3.5d

0.32 ± 0.19m 0.0017n 0.14 ± 0.16m 0.18 ± 0.10m

0.37 0.17 0.27 0.33

± ± ± ±

0.04i 0.02k 0.05j 0.03ij

RMSE Adjusted-R

r ± SE Tr-reduction d ± SE (sec)c Adjusted-R RMSE

r ± SE d ± SE

0.07 ± 0.04ab 0.03 ± 0.04ab 0.01 ± 0.01b 0.00045b ST mono-culture CIM mono-culture ST mixed-culutre CIM mixed-culture

b

345.93v 309.87v 4901.36w 1056.21w 0.98 0.96 0.93 0.97 0.03 0.04 0.03 0.02 0.07ef 0.09fg 0.07g 0.10e ± ± ± ± 0.54 0.44 0.31 0.67 14.4tu 6.6u 22.8t 11.6tu

26.55 13.27 56.87 133.4

± ± ± ±

9.31z 8.61z 24.9y 22.4x

r ± SE Tr-reduction d ± SE (sec)c

Lettuce biofilms

2

SS biofilms

2 b a a

Planktonic Bacteria

Table 1 Weibull model parameters for mixed species of cultivable indigenous microorganisms and Salmonella Typhimurium (planktonic and biofilms) on stainless steel and lettuce coupons.

RMSE Adjusted-R2 Tr-reduction (sec)

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

389

sensitive to COP. These findings suggest that COP reactive agents would penetrate inside the thin biofilms populations on SS compared to lettuce biofilms (Fig. 4B). It is reasonably assumed that, even though, the thickness of biofilms formation on SS and lettuce leaves were not measured, bacteria sheltered and formed microenvironment inside the stomata of lettuce leaves (Jahid et al., 2014b). The results were similar to findings of Kostaki et al. (2012). Several other studies, however, have found that mixed culture biofilms on SS are more resistant to chlorine, benzalkonium chloride, and peracetic acid compared to monocultures (Ibusquiza et al., 2012; Behnke et al., 2011; Behnke and Camper, 2012; van der Veen and Abee, 2011). As shown in Fig. 4C, mean reductions of 3.74, 4.11, 1.74, and 1.63 log CFU/cm2 were achieved in ST monoculture, CIM monoculture, ST mixed culture, and CIM mixed culture biofilms, respectively, on lettuce after 5 min of COP treatment. Effects of COP on both planktonic growth and various foods have been summandez and Thompson, 2012). It should rized in a recent review (Ferna be noted that COP has not been able to reduce biofilm growth by more than 5.0 log in food, although COP treatment has previously been shown to inhibit growth by up to 7.0 log in the planktonic state (Rowan et al., 2007). The efficacy of COP treatment depends on food type, bacterial age, and COP characteristics; however, it was previously shown that higher stomatal colonization and internalization due to mixed culture of ST with natural bacteria might contribute to COP resistance (Jahid et al., 2014b). 3.3. Weibull kinetics of ST and CIM as planktonic growth and SS and lettuce biofilm formation with COP treatment Table 1 shows the Weibull model parameters and goodness-offit indexes of planktonic growth and biofilms of SS and lettuce coupons. In general, the Weibull model accurately predicted the planktonic and biofilm inactivation kinetics, with adjusted correlation coefficients (R2) of planktonic, SS, and lettuce biofilms 0.93. The estimated RMSE was 0.05). Planktonic cells of both mixed cultures and monocultures showed a 4-log reduction in viability within 15 s of COP treatment, whereas another study showed a 1-log reduction within 27 s for E. coli and Bacillus subtilis (Tseng et al., 2012). Cold atmospheric plasma treatment of 30 s was also shown to cause a 4- to 6-log reduction of C. albicans on agar €mpfl et al., 2012). In the present work, in SS biofilms, plates (Kla none of the parameters were statistically significant (p > 0.05) according to ANOVA, although higher numerical values were observed for ST than for CIM in both monocultures and mixed cultures (Table 1). The differences in the death rate among monocultures and mixed cultures of ST and CIM treated with COP were not significant (p > 0.05). These results are in agreement with previous studies (Shen et al., 2012; Kostaki et al., 2012), but other

390

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

Fig. 5. Smooth and rugose colony formation as a result of COP treatment. (A) Smooth colonies on a plate; (B) rugose colonies on a plate; (C) smooth colonies (40  magnification); (D) rugose colonies (40  magnification).

authors have reported that monocultures and mixed cultures show significant differences in biofilms on SS after chlorine, benzalkonium chloride, or peracetic acid treatment (Behnke and Camper, 2012; van der Veen and Abee, 2011; Giaouris et al., 2013). These apparently contradictory results might be the consequence of differences in experimental settings, intrastrain variation, or variation in disinfectants used across studies. Mixed culture CIM and ST biofilms formed on lettuce are highly resistant to COP, and approximately 3e20-fold higher resistance was observed in our study compared to monocultures (Table 1). As expected, ST showed higher resistance than did CIM according to ANOVA, although the differences were not significant (p > 0.05). A mechanism that might explain this phenomenon could be higher EPS production or more internalization, especially to stomata, compared to monocultures. This resistance of mixed culture biofilms on SS was not reported in previous studies (Behnke and Camper, 2012; van der Veen and Abee, 2011; Giaouris et al., 2013). Recently, it was noted that internalization to stomata at higher temperatures in strong biofilms was not reduced by COP treatment (Jahid et al., 2014a). 3.4. A shift from smooth to rugose colony morphology after treatment with COP Fig. 5 shows the smooth and rugose colony phenotypes of ST. In this study, ST colony morphology was smooth without treatment (Fig. 5A and C), whereas the rugose morphology manifested itself after COP treatment of planktonic cells and biofilms on SS in both monocultures and mixed cultures, even after only 10 s of treatment (Fig. 5B and D). On the other hand, COP treatment of lettuce

coupons did not result in any shift from smooth to rugose appearance of colonies after treatment for 5 min (data not shown). Planktonic growth and biofilms of CIM did not change from smooth to rugose morphology. The rugosity might depend on the strength of disinfectant treatment with COP. Because COP does not penetrate inside lettuce, this could explain the lack of rugose colony formation in biofilms on lettuce. It is noteworthy that ST forms rugose colonies for the overproduction of EPS, which eventually leads to more biofilm formation compared to smooth colonies (de Rezende et al., 2005). The shift from smooth to rugose morphology was documented to be the result of microbial stress conditions such as nutrient deprivation (Wai et al., 2005) or a higher environmental temperature (Ali et al., 2002) and was shown to be regulated by cyclic di-GMP (Nakhamchik et al., 2008). A Salmonella Senftenberg rugosecolony variety was isolated from high-salt environments at mussel-processing facilities (Martinez-Urtaza et al., 2004). The present report shows a shift from smooth to rugose colony morphology after treatment with disinfectants, and the reasons for the shift from the smooth to rugose form in natural settings is still somewhat of a mystery. It nevertheless appears that the food industry is attempting to kill the bacteria and that this action results in the formation of rugose colonies, which could enhance the occurrence of biofilms on food and food contact surfaces. 3.5. Lettuce surface image analysis by means of FESEM FESEM images were acquired to illustrate the form and arrangement of cells in monoculture and mixed culture biofilms on SS and lettuce (Fig. 6). In general, microbial monoculture biofilms

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394 Fig. 6. Field emission scanning electron microscopy (FESEM) images of biofilm formation on stainless steel (SS) and lettuce leaves colonized with Salmonella Typhimurium (ST) and cultivable indigenous microorganism (CIM) monocultures and mixed cultures. The black arrows indicate biofilm formation. (A) A negative control on SS, (B) ST monoculture on SS coupons, (C) CIM monoculture on SS coupons, (D) ST þ CIM mixed culture on SS coupons, (E) a negative control of the lettuce surface, (F) ST monoculture on lettuce, (G) CIM monoculture on lettuce, and (H) ST þ CIM mixed culture on lettuce.

391

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

on the surface of lettuce coupons were different from biofilms of mixed cultures. It is evident from SEM images that SS coupons were cleaned properly (Fig. 6A). It is also evident in the micrograph that ST monoculture formed 3-dimensional biofilms on SS coupon surfaces with the secretion of EPS (Fig. 6B). On the other hand, CIM formed mat-like biofilm structures with more EPS compared to ST monocultures (Fig. 6C). When both cultures were mixed together, mixed culture biofilms looked like the combination of both micrographs (Fig. 6D). L. monocytogenes monoculture is flat and homogenous, but S. aureus monocultures show colonized aggregation (Rieu et al., 2008). Nonetheless, the study by Rieu and colleagues showed that mixed culture SS biofilms containing a combination of strains show behavior similar to our findings (Rieu et al., 2008). In mixed culture biofilms, colony morphology suggested that CIM colonized the surface of SS coupons, and ST formed biofilms on the surface of CIM biofilms (Fig. 6D). The mixed cultures also formed more EPS compared to monocultures; this observation is in line with previous findings on P. aeruginosa and E. coli mixed cultures (Machado et al., 2012). Both types were apparently not embedded in each other judging by the morphology of the microorganisms. The micrograph of the lettuce surface without bacterial inoculation (Fig. 6E) shows that a reduction of natural microflora following UVC treatment at the initial stage was successful because no bacterium-like structures were observed. Fig. 6EeH shows biofilm structures of ST monoculture, CIM monoculture, and mixed cultures. Although all types of cultures formed biofilms, the appearance and structure were different among the types of culture. By contrast, artificial inoculation and incubation at 15
C with the ST monoculture formed a 3-dimensional, highly aggregated, and compacted biofilm structure embedded under the EPS matrix (Fig. 6F). Bacterial cells were shielded under the EPS material, strengthening the biofilm structures. As expected, CIM monoculture also formed 3-dimensional biofilms on lettuce coupon surfaces (Fig. 6G). On the other hand, mixed culture biofilms of CIM and ST did not form compacted biofilms, but rather were scattered, less densely aggregated, and covered a greater surface area (Fig. 6H). Mixed culture biofilms also formed EPS-like materials, showing a lack of intercalation with EPS materials. In contrast to monocultures, SEM images of mixed culture sessile cells showed numerous punctate cell aggregates across the lettuce leaf surface. 3.6. Lettuce stomatal colonization Representative micrographs of stomatal colonization of ST monoculture, CIM monoculture, and ST and CIM mixed cultures are shown in Fig. 7. The control (without inoculation) showed no bacterial affinity or colonization by natural bacteria (Fig. 7A). The cells seeded via artificial inoculation using dipping methods and incubation at 15
C were more extensively aggregated on the guard cells of stomata by ST with EPS-like net structures (Fig. 7B). Affinity to stomata of ST is also supported by other authors (Kroupitski et al., 2011; Jahid et al., 2014b). In contrast to ST monocultures, CIM had no affinity to stomata (Fig. 7C). As expected, mixed cultures showed strong affinity, with bacterial aggregation and stomatal wells filled with bacteria. The guard cells were also contaminated with bacteria and EPS-like materials (Fig. 7D). Previously, our laboratory reported that quorum sensing enhanced stomatal colonization and resistance to COP (Jahid et al., 2014a). Therefore, enhanced resistance to COP in mixed culture biofilms on lettuce might be due to internalization and extensive colonization in stomatal wells just like the resistance to UV-C (Jahid et al., 2014b). Comparatively greater internalization was also previously reported for iceberg lettuce and arugula, but not romaine lettuce, basil, parsley, or tomatoes (Golberg et al., 2011).

Fig. 7. Field emission scanning electron microscopy (FESEM) images of stomatal colonization on lettuce coupons by Salmonella Typhimurium (ST) and cultivable indigenous microorganism (CIM) monocultures and mixed cultures. (A) A negative control, (B) ST monoculture on lettuce, (C) CIM monoculture on lettuce, and (D) ST þ CIM mixed culture on lettuce.

392

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394 Table 2 Food quality measurement of lettuce before and after treatment with cold oxygen plasma for 5 min. Parameters

Control ± SEM**

Color lightness (L*) Color redness (a*) Color yellowness (b*) Texture (mN) Color Appearance Smell Overall acceptance

84.93 6.67 23.60 192.77 5.66 5.2 5.2 5.66

± ± ± ± ± ± ± ±

0.85c 0.35d 1.3e 62.5f 0.25g 0.41h 0.35i 0.28j

Treated ± SEM** 82 6.17 25.41 197.77 5.16 5.08 5.0 5.16

± ± ± ± ± ± ± ±

0.94c 0.35d 1.2e 63.34f 0.40g 0.35h 0.46i 0.34j

**The values are mean ± SEM of 3 independent experiments. The values with same letters within a row were not significant (p < 0.05) according to Duncan's multiplerange test.

3.7. Food quality assessment Lettuce quality before and after 5-min COP treatment is shown in Table 2. Color and texture were tested, and no significant differences were observed between controls and treated lettuce samples (p > 0.05; Table 2). Rød et al. (2012) reported that COP changes the redness (a*) of meat. In our study, no significant differences in visual quality or any other tested sensory quality attributes were noted by the panelists between samples treated with COP and controls (p > 0.05). Sensory evaluation was conducted after storage for 2 h following COP treatment. All scores after treatment were higher than 5, which is greater than the lowest acceptable limit. It should be noted that our preliminary experiments showed that sensory evaluation immediately after treatment resulted in some panelists' complaining about a bad smell due to O3 from COP treatment (data not shown). 4. Conclusion The results of this study show that ST and natural bacteria antagonistically interact, and eventually internalize to a significantly greater degree than do monocultures, leading to COP resistance. COP treatment also shifts the colony morphology when cells are exposed directly to COP in planktonic culture and biofilms on SS but not on the lettuce surface where the bacteria hide in stomata and internal surfaces. COP could not be used as an alternative technique to eliminate foodborne pathogens from lettuce, although further research is needed to assess the efficacy of this method against other foodborne pathogens in the presence of natural bacteria. Moreover, a better understanding of the mechanisms underlying mixed culture of ST and CIM is necessary to minimize resistance to COP treatment. The limitation of the study was uncharacterized CIM population but still it shows significant impact on present sanitizing methods in food industry. Acknowledgments This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2013005051). References Agle, M.E., 2007. Biofilms in the food industry. In: Blaschek, H.P., Wang, H.H., Agle, M.E. (Eds.), Biofilms in Food Environment. Blackwell Publishing Ltd., Ames, IA, USA, pp. 3e17. Albert, I., Mafart, P., 2005. A modified Weibull model for bacterial inactivation. Int. J. Food Microbiol. 100, 197e211. Ali, A., Rashid, M.H., Karaolis, D.K., 2002. High-frequency rugose exopolysaccharide production by Vibrio cholerae. Appl. Environ. Microbiol. 68 (11), 5773e5778.

393

Anonymous, 2013. http://www.nutrition-and-you.com/vegetable-nutrition.html (accessed 23.05.13). Anriany, Y.A., Weiner, R.M., Johnson, J.A., De Rezende, C.E., Joseph, S.W., 2001. Salmonella enterica serovar Typhimurium DT104 displays a rugose phenotype. Appl. Environ. Microbiol. 67 (9), 4048e4056. Behnke, S., Camper, A.K., 2012. Chlorine dioxide disinfection of single and dual species biofilms, detached biofilm and planktonic populations. Biofouling 28 (6), 635e647. Behnke, S., Parker, A.E., Woodall, D., Camper, A.K., 2011. Comparing the chlorine disinfection of detached biofilm clusters with those of sessile biofilms and planktonic populations in single- and dual-species cultures. Appl. Environ. Microbiol. 77 (20), 7176e7184. ~ o, G., 2012. Challenges in biofilm inactivation: the use of cold plasma Brelles-Marin as a new approach. J. Bioprocess Biotech. 2, e108. http://dx.doi.org/10.4172/ 2155e9821.1000-107. Buzrul, S., Alpas, H., 2007. Modeling inactivation kinetics of food borne pathogens at a constant temperature. LWT e Food Sci. Technol. 40 (4), 632e637. Carter, M.Q., Xue, K., Brandl, M.T., Liu, F., Wu, L., Louie, J.W., Mandrell, R.E., Zhou, J., 2012. Functional metagenomics of Escherichia coli O157:H7 interactions with spinach indigenous microorganisms during biofilm formation. PLoS One 7 (9), e44186. Centers for Disease Control and Prevention (CDC), 2013. Reports of Salmonella Outbreak Investigations from 2013. http://www.cdc.gov/salmonella/outbreaks2013.html. Chen, Y., Dai, J., Morris Jr., J.G., Johnson, J.A., 2010. Genetic analysis of the capsule polysaccharide (K antigen) and exopolysaccharide genes in pandemic Vibrio parahaemolyticus O3:K6. BMC Microbiol. 10, 274. Chun, H.H., Kim, H.J., Won, M., Chung, K.S., Song, K.B., 2010. A comparison of kinetic models of foodborne pathogen inactivation by aqueous chlorine dioxide, fumaric acid, and ultraviolet-C. J. Korean Soc. Appl. Biol. Chem. 53 (2), 243e248. Cole, M.B., Davies, K.W., Munro, G., Holyoak, C.D., Kilsby, D.C., 1993. A vitalistic model to describe the thermal inactivation of Listeria monocytogenes. J. Ind. Microbiol. 12, 232e239. Coroller, L., Leguerinel, I., Mettler, E., Savy, N., Mafart, P., 2006. General model, based on two mixed Weibull distributions of bacterial resistance, for describing various shapes of inactivation curves. Appl. Environ. Microbiol. 72 (10), 6493e6502. Critzer, F., Kelly-Wintenberg, K., South, S., Golden, D., 2007. Atmospheric plasma inactivation of foodborne pathogens on fresh produce surfaces. J. Food Prot. 70 (10), 2290e2296. de Rezende, C.E., Anriany, Y., Carr, L.E., Joseph, S.W., Weiner, R.M., 2005. Capsular polysaccharide surrounds smooth and rugose types of Salmonella enterica serovar Typhimurium DT104. Appl. Environ. Microbiol. 71 (11), 7345e7351. European Food Safety Authority (EFSA), 2013. Foodborne Zoonotic Diseases. Salmonella. http://www.efsa.europa.eu/en/topics/topic/salmonella.htm (accessed 16.05.13). ndez, A., Thompson, A., 2012. The inactivation of Salmonella by cold atmoFerna spheric plasma treatment. Food Res. Int. 45 (2), 678e684. Food and Drug administration, 2013. Outbreaks Associated with Fresh and Fresh-cut Produce. Incidence, Growth, and Survival of Pathogens in Fresh and Fresh-cut Produce available from: http://www.fda.gov/Food/FoodScienceResearch/ SafePracticesforFoodProcesses/ucm091265.htm (accessed 24.04.13). Geeraerd, A.H., Valdramidis, V.P., Van Impe, J.F., 2005. GInaFiT, a freeware tool to assess nonlog-linear microbial survivor curves. Int. J. Food Microbiol. 102, 95e105. Giaouris, E., Chorianopoulos, N., Doulgeraki, A., Nychas, G.-J., 2013. Co-culture with Listeria monocytogenes within a dual-Species biofilm community strongly increases resistance of Pseudomonas putida to benzalkonium chloride. PLoS ONE 8 (10), e77276. Golberg, D., Kroupitski, Y., Belausov, E., Pinto, R., Sela, S., 2011. Salmonella Typhimurium internalization is variable in leafy vegetables and fresh herbs. Int. J. Food Microbiol. 145 (1), 250e257. Ha, S.D., Pillai, S.D., Ricke, 1995. Growth response of Salmonella spp. To cycloheximide amendment in media. J. Rapid Methods Autom. Microbiol. 4, 77e85. nchez, D., Cabo, M.L., 2012. Adherence Ibusquiza, P.S., Herrera, J.J.R., V azquez-Sa kinetics, resistance to benzalkonium chloride and microscopic analysis of mixed biofilms formed by Listeria monocytogenes and Pseudomonas putida. Food Control 25, 202e210. Jahid, I.K., Ha, S.D., 2012. A review of microbial biofilms of produce: future challenge to food safety. Food Sci. Biotechnol. 21 (2), 299e316. Jahid, I.K., Han, N., Ha, S.D., 2014a. Inactivation kinetics of cold oxygen plasma depend on incubation conditions of Aeromonas hydrophila biofilm on lettuce. Food Res. Int. 55, 181e189. Jahid, I.K., Han, N., Srey, S., Ha, S.D., 2014b. Competitive interactions inside mixedspecies biofilms of Salmonella Typhimurium and cultivable indigenous microorganisms on lettuce enhance microbial resistance of their sessile cells to ultraviolet C (UV-C) irradiation. Food Res. Int. 55, 445e454. €mpfl, T.G., Isbary, G., Shimizu, T., Li, Y.F., Zimmermann, J.L., Stolz, W., Schlegel, J., Kla Morfill, G.E., Schmidt, H.U., 2012. Cold atmospheric air plasma sterilization against spores and other microorganisms of clinical interest. Appl. Environ. Microbiol. 78 (15), 5077e5082. Kolenblander, P.E., 2000. Oral microbial communities: biofilms, interactions, and genetic systems. Annu. Rev. Microbiol. 54, 413e437. Kostaki, M., Chorianopoulos, N., Braxou, E., Nychas, G.J., Giaouris, E., 2012. Differential biofilm formation and chemical disinfection resistance of sessile cells of

394

I.K. Jahid et al. / Food Microbiology 46 (2015) 383e394

Listeria monocytogenes strains under monospecies and dual-species (with Salmonella enterica) conditions. Appl. Environ. Microbiol. 78 (8), 2586e2595. Kroupitski, Y., Pinto, R., Belausov, E., Sela, S., 2011. Distribution of Salmonella typhimurium in romaine lettuce leaves. Food Microbiol. 28 (5), 990e997. Kvam, E., Davis, B., Mondello, F., Garner, A.L., 2012. Nonthermal atmospheric plasma rapidly disinfects multidrug-resistant microbes by inducing cell surface damage. Antimicrob. Agents Chemother. 56 (4), 2028e2036. Lee, M.J., Ha, S.D., 2008. Synergistic effect of vitamin B1 on sanitizer and disinfectant treatments for reduction of coliforms in rice. Food Control 19, 113e118. Liao, C.H., 2007. Inhibition of foodborne pathogens by native microflora recovered from fresh peeled baby carrot and propagated in cultures. J. Food Sci. 72 (4), M134eM139. Lindow, S.E., Brandl, M.T., 2003. Microbiology of the phyllosphere. Appl. Environ. Microbiol. 69, 1875e1883. Liu, N.T., Lefcourt, A.M., Nou, X., Shelton, D.R., Zhang, G., Lo, Y.M., 2013. Native microflora in fresh-cut produce processing plants and their potentials for biofilm formation. J. Food Prot. 76 (5), 827e832. Lopez-Velasco, G., Tydings, H.A., Boyer, R.R., Falkinham, J.O., Ponder, M.A., 2012. Characterization of interactions between Escherichia coli O157:H7 with epiphytic bacteria in vitro and on spinach leaf surfaces. Int. J. Food Microbiol. 153 (3), 351e357. Machado, I., Lopes, S.P., Sousa, A.M., Pereira, M.O., 2012. Adaptive response of single and binary Pseudomonas aeruginosa and Escherichia coli biofilms to benzalkonium chloride. J. Basic Microbiol. 52 (1), 43e52. Mafart, P., Couvert, O., Gaillard, S., Leguerinel, I., 2002. On calculating sterility in thermal preservation methods: application of the Weibull frequency distribution model. Int. J. Food Microbiol. 72, 107e113. Manuzon, M.Y., Wang, H.H., 2007. Mixed culture biofilms. In: Blaschek, H.P., Wang, H.H., Agle, M.E. (Eds.), Biofilms in Food Environment. Blackwell Publishing Ltd., Ames, IA, pp. 105e125.  n, A., Garcia-Martin, O., 2004. Detection Martinez-Urtaza, J., Peiteado, J., Lozano-Leo of Salmonella Senftenberg associated with high saline environments in mussel processing facilities. J. Food Prot. 67 (2), 256e263. Matz, C., McDougald, D., Moreno, A.M., Yung, P.Y., Yildiz, F.H., Kjelleberg, S., 2005. Biofilm formation and phenotypic variation enhance predation-driven persistence of Vibrio cholerae. Proc. Natl. Acad. Sci. U. S. A 102 (46), 16819e16824. Møller, C.O., Ilg, Y., Aabo, S., Christensen, B.B., Dalgaard, P., Hansen, T.B., 2013. Effect of natural microbiota on growth of Salmonella spp. in fresh pork - a predictive microbiology approach. Food Microbiol. 34 (2), 284e295. Moons, P., Michiels, C.W., Aertsen, A., 2009. Bacterial interactions in biofilms. Crit. Rev. Microbiol. 35 (3), 157e168. Morbidity and Mortality Weekly Report (MMWR), 2013. Surveillance for Foodborne Disease Outbreaks d United States, 2009e2010 available from: http://www. cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm?s_cid¼mm6203a1_w (accessed 25.01.13). Morris, C.E., Monier, J.M., Jacques, M.A., 1998. A technique to quantify the population size and composition of the biofilm component in communities of bacteria in the phyllosphere. Appl. Environ. Microbiol. 64 (12), 4789e4795. Nakhamchik, A., Wilde, C., Rowe-Magnus, D.A., 2008. Cyclic-di-GMP regulates extracellular polysaccharide production, biofilm formation, and rugose colony development by Vibrio vulnificus. Appl. Environ. Microbiol. 74 (13), 4199e4209. Niemira, B.A., Cooke, P.H., 2010. Escherichia coli O157:H7 biofilm formation on Romaine lettuce and spinach leaf surfaces reduces efficacy of irradiation and sodium hypochlorite washes. J. Food Sci. 75, M270eM277. Olaimat, A.N., Holley, R.A., 2012. Factors influencing the microbial safety of fresh produce: a review. Food Microbiol. 32, 1e19. € Olmez, H., Temur, S.D., 2010. Effects of different sanitizing treatments on biofilm and attachment of Escherichia coli and Listeria monocytogenes on green leaf lettuce. LWT-Food Sci. Technol. 43, 964e970. Patel, J., Sharma, M., 2010. Differences in attachment of Salmonella enterica serovars to cabbage and lettuce leaves. Int. J. Food Microbiol. 139 (1e2), 41e47.

Pei, X., Lu, X., Liu, J., Liu, D., Yang, Y., Ostrikov, K., Chu, P.K., Pan, Y., 2012. Inactivation of a 25.5 mm Enterococcus faecalis biofilm by a room-temperature, batteryoperated, handheld air plasma jet. J. Phys. D: Appl. Phys. 45 (16), 165205. Peleg, M., 1996. Evaluation of the Fermi equation as a model of doseeresponse curves. Appl. Microbiol. Biotechnol. 46, 303e306. Perni, S., Liu, D.W., Shama, G., Kong, M.G., 2008. Cold atmospheric plasma decontamination of the pericarps of fruit. J. Food Prot. 71 (2), 302e308. rez-Rodríguez, F., Posada-Izquierdo, G.D., Valero, A., García-Gimeno, R.M., Pe Zurera, G., 2013. Modelling survival kinetics of Staphylococcus aureus and Escherichia coli O157:H7 on stainless steel surfaces soiled with different substrates under static conditions of temperature and relative humidity. Food Microbiol. 33, 197e204. Posada-Izquierdo, G.D., P0 erez-Rodrıguez, F., Zurera, G., 2013. Mathematical quantification of microbial inactivation of Escherichia coli O157:H7 and Salmonella spp. on stainless steel surfaces soiled with different vegetables juice substrates. Food Res. Int. 54 (2), 1688e1698. Rowan, N.J., Espie, S., Harrower, J., Anderson, J.G., Marsili, L., MacGregor, S.J., 2007. Pulsed-plasma gas-discharge inactivation of microbial pathogens in chilled poultry wash water. J. Food Prot. 70 (12), 2805e2810. Rayner, J., Veeh, R., Flood, J., 2004. Prevalence of microbial biofilms on selected fresh produce and household surfaces. Int. J. Food Microbiol. 95, 29e39. Rieu, A., Lemaître, J.P., Guzzo, J., Piveteau, P., 2008. Interactions in dual species biofilms between Listeria monocytogenes EGD-e and several strains of Staphylococcus aureus. Int. J. Food Microbiol. 126 (1e2), 76e82. Rød, S.K., Hansen, F., Leipold, F., Knøchel, S., 2012. Cold atmospheric pressure plasma treatment of ready-to-eat meat: inactivation of Listeria innocua and changes in product quality. Food Microbiol. 30 (1), 233e238. Shen, C., Luo, Y., Nou, X., Bauchan, G., Zhou, B., Wang, Q., Millner, P., 2012. Enhanced inactivation of Salmonella and Pseudomonas biofilms on stainless steel by use of T-128, a fresh-produce washing aid, in chlorinated wash solutions. Appl. Environ. Microbiol. 78 (19), 6789e6798. Tseng, S., Abramzon, N., Jackson, J.O., Lin, W.J., 2012. Gas discharge plasmas are effective in inactivating Bacillus and Clostridium spores. Appl. Microbiol. Biotechnol. 93 (6), 2563e2570. United States Department of Agriculture-Economic Research Service (USDA-ERS), 2010. Foodborne Illness Cost Calculator for Salmonella and E. coli. CIDRAP Report. USDA-ERS, Washington, DC available at: http://www.cidrap.umn.edu/ cidrap/content/fs/fooddisease/news/may2410costest.html. Vaid, R., Linton, R.H., Morgan, M.T., 2010. Comparison of inactivation of Listeria monocytogenes within a biofilm matrix using chlorine dioxide gas, aqueous chlorine dioxide and sodium hypochlorite treatments. Food Microbiol. 27, 979e984. van Boekel, M.A.J.S., 2002. On the use of the Weibull model to describe thermal inactivation of microbial vegetative cells. Int. J. Food Microbiol. 74, 139e159. van der Veen, S., Abee, T., 2011. Mixed species biofilms of Listeria monocytogenes and Lactobacillus plantarum show enhanced resistance to benzalkonium chloride and peracetic acid. Int. J. Food Microbiol. 144 (3), 421e431. Vleugels, M., Shama, G., Deng, X.T., Greenacre, E., Brocklehurst, T., Kong, M.G., 2005. Atmospheric plasma inactivation of biofilm-forming bacteria for food safety control. IEEE Trans. Plasma Sci. 33 (2), 824e828. Wai, S.N., Mizunoe, Y., Takade, A., Kawabata, S.I., Yoshida, S.I., 2005. Vibrio cholerae O1 strain TSI-4 produces the exopolysaccharide materials that determine colony morphology, stress resistance, and biofilm formation. Appl. Environ. Microbiol. 64 (10), 3648e3655. Xiong, Z., Du, T., Lu, X., Cao, Y., Pan, Y., 2011. How deep can plasma penetrate into a biofilm? Appl. Phys. Lett. 98, 221503. Yildiz, F.H., Schoolnik, G.K., 1999. Vibrio cholerae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. Proc. Natl. Acad. Sci. U. S. A 96 (7), 4028e4033.