ABSTRACT D. GARCÍA, N. GÓMEZ, P. MAÑAS, S. CONDÓN, J. RASO AND R. PAGÁN ª 2005 The Society for Applied Microbiology

Transcription

1 Journal of Applied Microbiology, 99, 9 doi:./j.-7...x Occurrence of sublethal injury after pulsed electric fields depending on the micro-organism, the treatment medium ph and the intensity of the treatment investigated D. García, N. Gómez, P. Mañas, S. Condón, J. Raso and R. Pagán Departamento de Producción Animal y Ciencia de los Alimentos, Facultad de Veterinaria, Universidad de Zaragoza, Zaragoza, Spain /: received November, revised 7 December and accepted December ABSTRACT D. GARCÍA, N. GÓMEZ, P. MAÑAS, S. CONDÓN, J. RASO AND R. PAGÁN.. Aims: The objective was to investigate the occurrence of sublethal injury after pulsed electric field (PEF) depending on the treatment time, the electric field strength and the ph of the treatment media in two Gram-positive (Bacillus subtilis ssp. niger, Listeria monocytogenes) and six Gram-negative (Escherichia coli, Escherichia coli O7:H7, Pseudomonas aeruginosa, Salmonella serotype Senftenberg 77W, Salmonella serotype Typhimurium, Yersinia enterocolitica) bacterial strains. Methods and Results: A characteristic behaviour was observed for the Gram-positive and Gram-negative bacteria studied. Whereas Gram-positive bacteria showed a higher PEF resistance at ph 7Æ, the Gram-negative were more resistant at ph Æ. In these conditions, in which bacteria showed their maximum resistance, a large proportion of sublethally injured cells were detected. In most cases, the longer the treatment time and the higher the electric field applied, the greater the proportion of sublethally injured cells that were detected. No sublethal injury was detected when Gram-positive bacteria were treated at ph Æ and Gram-negative at ph 7Æ. Conclusions: Sublethal injury was detected after PEF so, bacterial inactivation by PEF is not an Ôall or nothingõ event. Significance and Impact of the Study: This work could be useful for improving food preservation by PEF. Keywords: Bacillus subtilis, Escherichia coli O7:H7, Listeria monocytogenes, Pseudomonas aeruginosa, pulsed electric fields, Salmonella, sublethal injury, Yersinia enterocolitica. INTRODUCTION Pulsed electric fields (PEF) are considered one of the most promising nonthermal processes for food preservation (Barbosa-Cánovas et al. 999). The process consists of the application of short duration pulses ( ls) of a high electric field ( 7 kv cm ) ) to food placed between two electrodes. PEF are able to inactivate micro-organisms (Wouters and Smelt 997) and enzymes (van Loey et al. ) without altering sensorial and nutritional properties of foods, so attention is beginning to focus on the possibility of using PEF treatments to pasteurize or sterilize food. Correspondence to: Dr Rafael Pagán Tomás, Dpto. PACA. Facultad de Veterinaria, Universidad de Zaragoza, C/Miguel Servet, 77,, Zaragoza, Spain ( pagan@posta.unizar.es). Conventional heat pasteurization treatments have been used in a wide range of products to ensure safe inactivation of pathogenic and spoilage micro-organisms. Traditionally, these heat treatments had been designed to inactivate at least 7 log cycles of the most heat-resistant pathogen contaminating foods. More recently, for fruit and vegetable juices, an inactivation of log cycles in the bacteria associated with food-borne illness has been proposed (NACMCF 997). As a result, the interest in inactivating at least -log cycles of pathogens contaminating foods by traditional or emerging technologies has increased (Evrendilek et al. 999; Stringer et al. ; Iu et al. ). In spite of the capacity of PEF treatments to inactivate micro-organisms, some authors have demonstrated the high PEF resistance of some pathogenic bacteria such as Listeria ª The Society for Applied Microbiology

2 SUBLETHAL INJURY IN PEF-TREATED CELLS 9 monocytogenes in milk (Bendicho et al. ), Escherichia coli O7:H7 in apple juice or cider (Iu et al. ; Ravishankar et al. ) or Salmonella serotype Senftenberg 77W in laboratory media (Álvarez et al. ). Therefore, the use of this technology to pasteurize foods would be compromised. As a result, in recent years food technologists have been trying to develop more complex processes, combining thermal and nonthermal treatments with other preservative factors at mild doses to ensure the microbial safety and stability of the products with a minimal loss of quality attributes (Raso and Barbosa-Cánovas ). This approach, known as Ôhurdle technologyõ, has already been applied successfully using traditional techniques of food preservation (Leistner and Gorris 99). The total preservative effect of combining several preservation techniques could be merely additive, but a synergistic effect is preferable. Leistner and Gorris (99) suggested that a synergistic effect could be expected when the hurdles applied affect different targets within the cell, impairing the reparation of the damages inflicted and the restoration of the cellular homeostasis. The design of the optimum hurdle combination requires finding out the mechanism of inactivation of each hurdle in order to determine the order of application and the intensity of the hurdles. Microbial inactivation by PEF is believed to be due to the effects of PEF on the cell envelopes. PEF can cause formation of pores, affecting the integrity and functionality of the membrane (Hamilton and Sale 97; Sale and Hamilton 97). These pores can be reversible or irreversible, depending on the degree of membrane damage (Weaver and Chizmadzhev 99). The occurrence of reversible pores implies that a proportion of cells reseal to a greater or lesser extent after PEF treatment. Therefore micro-organisms that survive PEF treatments might be sublethally injured, which would allow the design of combined processes with a synergistic lethal effect. As sublethal injury is supposed to be related to the higher sensitivity of survivors to stress conditions after treatment, the success of a combined treatment should be correlated with the degree of sublethal injury caused by the hurdles in the bacterial population (Wuytack et al. ). However, under suitable conditions, sublethally injured cells might be repaired, which is a very important aspect that needs to be taken into account regarding food safety. A few cells being capable of repairing damage after PEF treatment could result in infective concentrations. The most common method used to determine the occurrence of sublethal injury consists of plating survivors after treatments into two culture media: a nonselective one, which allows cells to repair sublethal damages and recover, and a selective one, in which survivors are not capable of repairing their damages and finally die (Mackey ). Sublethally injured cells are estimated by the difference in the number of survivors obtained after plating treated cells in both media. Although the formation of reversible pores under PEF, as described above, would suggest so, most of published data (Simpson et al. 999; Dutreux et al. a,b; Russell et al. ; Ravishankar et al. ; Ulmer et al. ; Wuytack et al. ; Aronsson et al. ) have not in fact demonstrated the occurrence of sublethal injury after PEF treatment using the selective medium plating technique. Only Damar et al. () detected scarce sublethal injury, lower than % of surviving cells, when Escherichia coli and Staphylococcus aureus were PEF treated at kv cm ) in peptone solution. Also, Unal et al. (), Liang et al. () and Ravishankar et al. () have observed sublethally injured cells when PEF were combined with ozone or heat. In this case, it is not clear whether sublethal injury is caused by PEF, or a consequence of heat and ozone, which might be responsible for the occurrence of sublethal injuries per se. On the contrary, some results obtained by our research group (García et al. a,b; Zerrouk et al. ) have demonstrated that PEF-treated E. coli, Salmonella serotype Senftenberg 77W and E. coli O7:H7 cell populations were sublethally injured in a large proportion. For example, more than 99Æ9% of survivors were sublethally injured when E. coli cells were treated in McIlvaine buffer at ph for ls at9kvcm ). The disagreements about the occurrence of sublethal injury after PEF treatments among published data could be related to differences in the methodology employed by the authors but is most likely because of the different treatment conditions and micro-organisms investigated. In this way, García et al. (a) demonstrated that the PEF resistance of E. coli NCTC 9 and the occurrence of sublethal injury were related and depended on the treatment medium ph. Given that the occurrence of sublethal injury is a relevant aspect to be taken into account in the development of appropriate combined processes, a study was carried out in order to describe when sublethal injury occurs and how it might be influenced by treatment conditions. The aim of this work was to investigate the occurrence of sublethal injury after PEF treatments depending on the treatment time, the electric field strength and the ph of the treatment media in two Gram-positive and six Gramnegative bacterial strains. MATERIALS AND METHODS Micro-organisms and growth conditions The strains of Bacillus subtilis ssp. niger (NCTC 7), Escherichia coli (NCTC 9), Listeria monocytogenes (ATCC ), Pseudomonas aeruginosa (ATCC ), Salmonella ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

3 9 D. GARCÍA ET AL. serotype Senftenberg 77W (ATCC 8), Salmonella serotype Typhimurium (ATCC ) and Yersinia enterocolitica (ATCC 9) used in this investigation were supplied by the Spanish Type Culture Collection. The strain of E. coli O7:H7 used in this investigation is a VTEC (phage type ) isolated by Dr Chapman (Chapman et al. 99) and was kindly provided by Dr B. Mackey (University of Reading, UK). During this investigation the strains were maintained on slants of tryptic soya agar (Biolife, Milan, Italy) with Æ% of yeast extract added (Biolife) (TSAYE) (E. coli, E. coli O7:H7, L. monocytogenes, Ps. aeruginosa, Y. enterocolitica) or nutrient agar (NA; Biolife) (B. subtilis ssp. niger, Salm. Senftenberg 77W, Salm. Typhimurium). Broth subcultures were prepared by inoculating, with one single colony from a plate, a test tube containing ml of sterile tryptic soya broth (Biolife) with Æ% of yeast extract added (Biolife) (TSBYE) (E. coli, E. coli O7:H7, L. monocytogenes, Ps. aeruginosa, Y. enterocolitica) or nutrient broth (NB; Biolife) (B. subtilis ssp. niger, Salm. Senftenberg 77W, Salm. Typhimurium). After inoculation the tubes were incubated at 7 C for h. With these subcultures, ml Erlenmeyer flasks containing ml of TSBYE (E. coli, E. coli O7:H7, L. monocytogenes, Ps. aeruginosa, Y. enterocolitica) or NB (B. subtilis ssp. niger, Salm. Senftenberg 77W, Salm. Typhimurium) were inoculated to a final concentration of cells ml ). These flasks were incubated under agitation ( rev min ) ; Selecta, mod. Rotabit, Barcelona, Spain) at 7 C until the stationary growth phase was reached. PEF equipment The PEF equipment used in this investigation has been previously described by Raso et al. (). High electric field pulses were produced by discharging a -lf capacitor (Maxwell, San Diego, CA, USA) via an IGBT switch (HTS 8-ÇFI; Behlke, Frankfurt, Germany) into a treatment chamber. A function generator (AGF ; Tektronix, Wilsonville, OR, USA) delivered the on-time signal to the switch. The capacitor was charged using a high voltage dc power supply (HCK M ; FUG, Rosenhein, Germany). A cylindrical plastic tube closed with two polished stainless steel electrodes was used as the treatment chamber. The distance between electrodes was Æ cm and the electrode area was Æ cm. The circuit configuration generated square waveform pulses at different frequencies and pulse widths. Actual electric field strength applied was measured in the treatment chamber with a high voltage probe connected to an oscilloscope (TDS ; Tektronix). The PEF equipment includes provisions for measuring sample temperature. Immediately after the treatment a thermocouple type K of Æ9 mm diameter, pneumatically activated, enters into the treatment chamber and the temperature is measured in the centre of the chamber. PEF treatments Before treatment, bacterial cultures were centrifuged at g for min suspended in citrate phosphate McIlvaine buffer of ph 7Æ, Æ, Æ, Æ and Æ at room temperature (Dawson et al. 97). whose concentration was adjusted to an electric conductivity of ms cm ). Next, Æ ml of the microbial suspensions, at a concentration of c. 9 CFU ml ), were placed in the treatment chamber with a sterile syringe as has been previously described (Raso et al. ). A pulse width of ls and a pulse repetition rate of Hz (when applying kv cm ) ) or Hz (when applying or 9 kv cm ) ) was used in this study. Experiments started at room temperature. In all cases the temperature of the samples after treatment was lower than C. Viable counts After treatments, samples were adequately diluted and Æ ml samples were pour plated onto TSAYE (E. coli, E. coli O7:H7, L. monocytogenes, Ps. aeruginosa, Y. enterocolitica) or NA (B. subtilis ssp. niger and Salm. Senftenberg 77W, Salm. Typhimurium). Plates were incubated for h (E. coli, E. coli O7:H7, Ps. aeruginosa, B. subtilis ssp. niger, Salm. Senftenberg 77W, Salm. Typhimurium) or 8 h (L. monocytogenes, Y. enterocolitica) at7 C. Previous experiments showed that longer incubation times did not influence survival counts. After incubation, colony-forming units (CFU) were counted with an improved image analyser automatic counter (Protos, Analytical Measuring Systems, Cambridge, UK) as previously described (Condón et al. 99). The error bars on the figures indicate the mean standard deviations for the data points obtained from at least two independent experiments. Survival counts shown in Fig. were based on mean values obtained from three to six independent experiments. Detection of sublethal injury In order to determine bacterial cell injury, treated samples were also plated on TSAYE (E. coli, E. coli O7:H7, L. monocytogenes, Ps. aeruginosa, Y. enterocolitica) orna (B. subtilis ssp. niger, Salm. Senftenberg 77W, Salm. Typhimurium) with,,,,,, or % of sodium chloride (Probus, Barcelona, Spain) added respectively (TSAYE-SC, NA-SC). These were the maximum noninhibitory sodium chloride concentrations for native cells previously determined for each bacterial species investigated (data not shown). Plates containing selective media were incubated for h more than those containing nonselective media. Previous experiments showed that longer incubation times did not influence survival counts. The number of sublethally injured cells was estimated by the difference in the number of CFU obtained after plating ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

4 SUBLETHAL INJURY IN PEF-TREATED CELLS 97 PEF-treated cells in the nonselective (TSAYE, NA) and the selective (TSAYE-SC, NA-SC) media. The proportion of sublethally injured cells was estimated by the difference in the number of log cycles of CFU obtained after plating PEF-treated cells in the same media. RESULTS Occurrence of sublethal injury after PEF treatments in different bacterial species Figure shows the number of log cycles of inactivation of B. subtilis ssp. niger, L. monocytogenes, E. coli and Salm. Senftenberg 77W after PEF treatments at 9,, 9 and kv cm ), respectively, for ls in citrate phosphate buffer of ph Æ and 7Æ, and recovered in the nonselective and the selective medium. These treatment conditions had been selected from a previous work (data not shown) in order to attain log cycle of inactivation at the treatment medium ph in which micro-organisms showed the maximum PEF resistance: ph 7Æ for the Gram-positive bacteria (B. subtilis ssp. niger (a) log cycles of inactivation (c) log cycles of inactivation 7 7 (b) log cycles of inactivation (d) log cycles of inactivation 7 7 Fig. Log cycles of inactivation of Bacillus subtilis ssp. niger (a), Listeria monocytogenes (b), Escherichia coli (c) and Salmonella serotype Senftenberg 77W (d) and cells after PEF treatments at 9,, 9 and kv cm ), respectively, for ls in citrate phosphate buffer of ph Æ and 7Æ and recovered in the nonselective (white bars) and the selective (black bars) medium and L. monocytogenes) and ph Æ for the Gram-negative (E. coli and Salm. Senftenberg 77W). As seen in Fig., the Gram-positive bacteria recovered in the nonselective medium showed a higher PEF resistance when treated at ph 7Æ, whereas the Gram-negative bacteria were more PEF resistant when treated at ph Æ. At the ph of maximum PEF resistance, more than 9% of survivors of each bacterial strain studied were sublethally injured as survival counts obtained in the selective media were at least log cycle lower than in the nonselective media. On the contrary, no sublethal injury was detected when the Gram-positive bacteria were PEF treated at ph Æ and the Gram-negative treated at ph 7Æ. Under these treatment conditions, no statistically significant differences (P > Æ) were observed between the number of survivors recovered in the nonselective or the selective medium. Effect of the electric field strength on the occurrence of sublethal injury after PEF treatments Figure shows the number of log cycles of inactivation of B. subtilis ssp. niger, L. monocytogenes, E. coli and Salm. Senftenberg 77W suspended in citrate phosphate buffer of ph 7Æ and Æ, after PEF treatments at different electric field strengths for ls, and recovered in the nonselective and the selective medium. As shown in Fig., when cells were suspended in the ph of maximum PEF resistance, B. subtilis ssp. niger and L. monocytogenes in ph 7 and Salm. Senftenberg 77W in ph (Fig. a,c,f,h respectively), and exposed to a range of electric field strengths the number of survivors recovered in the nonselective media kept constant or decreased slightly with the increasing electric field strength. However, the number of survivors recovered in the selective media decreased in a higher proportion up to the maximum electric field strength applied. As a result, the higher the electric field strength, the greater the proportion of sublethally injured cells. In contrast, because of the lower PEF resistance of E. coli cells, the proportion of dead cells increased to a higher degree with the increase in the electric field strength, so the proportion of sublethally injured cells was at its maximum at 9 kv cm ) and decreased at kv cm ). After a PEF treatment at 9 kv cm ) for ls the proportion of sublethally injured E. coli cells was higher than 99Æ9%. However, sublethal injury was barely observed at any electric field strength when PEF resistance was evaluated at the ph in which micro-organisms showed the lowest PEF resistance: B. subtilis ssp. niger and L. monocytogenes PEF treated at ph and E. coli and Salm. Senftenberg 77W at ph 7 (Fig. b,d,e,g respectively). Under these experimental conditions, PEF resistance decreased strongly with the increase in the electric field strength. The number of log ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

5 98 D. GARCÍA ET AL. (a) (b) (c) Inactivation (log ) Electrical field strength (kv cm ) Electrical field strength (kv cm ) Electrical field strength (kv cm ) Inactivation (log ) (d) (e) (f) Inactivation (log ) Inactivation (log ) Electrical field strength (kv cm ) Electrical field strength (kv cm ) Electrical field strength (kv cm ) Inactivation (log ) Inactivation (log ) (g) Inactivation (log ) (h) Inactivation (log ) 9 9 Electrical field strength (kv cm ) Electrical field strength (kv cm ) Fig. Log cycles of inactivation of Bacillus subtilis ssp. niger (a and b), Listeria monocytogenes (c and d), Escherichia coli (e and f) and Salmonella serotype Senftenberg 77W (g and h) cells after PEF treatments at different electric field strengths for ls in citrate phosphate buffer of ph 7Æ (a, c, e and g) and ph Æ (b, d, f and h) recovered in the nonselective (white bars) and the selective (black bars) medium cycles of dead cells reached was very similar to that observed when micro-organisms were PEF treated at the ph of the highest PEF resistance and recovered in the selective media. Therefore, when selective recovery conditions are used, the influence of the treatment medium ph on the PEF resistance of the Gram-positive or Gram-negative bacteria tested might disappear. At the maximum electric field tested for ls the number of killed plus sublethally injured cells was > log cycles, independently of the micro-organism investigated. Effect of the treatment time on the occurrence of sublethal injury after PEF treatments Figure shows the number of log cycles of inactivation of B. subtilis ssp. niger and L. monocytogenes suspended in citrate phosphate buffer of ph 7Æ ande. coli and Salm. Senftenberg 77W in ph Æ, after different PEF treatment times, and recovered in the nonselective and the selective medium. Figure shows the influence of the treatment time on the occurrence of sublethal injury when micro-organisms were suspended in the ph of maximum PEF resistance. As shown in Fig., the number of dead cells kept constant or increased slightly with the duration of the PEF treatment, whereas the proportion of sublethally injured cells increased drastically up to the maximum treatment time applied. The higher proportion of sublethally injured cells, more than 99% of survivors, was observed when E. coli cells were PEF treated at 9 kv cm ) or Salm. Senftenberg 77W at kv cm ) for ls. Again, when B. subtilis ssp. niger and L. monocytogenes were PEF treated for different times (,, and ls) in citrate phosphate buffer of ph ÆorE. coli and ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

6 SUBLETHAL INJURY IN PEF-TREATED CELLS 99 Fig. Log cycles of inactivation of Bacillus subtilis ssp. niger (a), Listeria monocytogenes (b), Escherichia coli (c) and Salmonella serotype Senftenberg 77W (d) cells after PEF treatments at 9,, 9 and kv cm ), respectively, for different times in citrate phosphate buffer of ph 7Æ (a and b) and ph Æ (c and d) and recovered in the nonselective (white bars) and the selective (black bars) medium (a) log cycles of inactivation (c) log cycles of inactivation Treatment time (ms) Treatment time (ms) (b) log cycles of inactivation (d) log cycles of inactivation Treatment time (ms) Treatment time (ms) Salm. Senftenberg 77W at ph 7Æ, none or very scarce sublethal injury was detected at any treatment time investigated (data not shown). Effect of the ph of the treatment medium on the occurrence of sublethal injury after PEF treatments Figure shows the survival fraction of B. subtilis ssp. niger, L. monocytogenes, E. coli and Salm. Senftenberg 77W after PEF treatments at 9,, 9 and kv cm ), respectively, for ls, in citrate phosphate buffer of different ph, and recovered in the nonselective and the selective medium. Data for PEF inactivation of B. subtilis ssp. niger and L. monocytogenes at ph Æ are not included as these microorganisms died at this low ph. Figure illustrates that whereas B. subtilis ssp. niger and L. monocytogenes showed their maximum resistance at ph 7Æ, E. coli and Salm. Senftenberg 77W showed their maximum resistance at ph Æ. The proportion of suble- Fig. Survival fraction of Bacillus subtilis ssp. niger (a), Listeria monocytogenes (b), Escherichia coli (c) and Salmonella serotype Senftenberg 77W (d) cells after PEF treatments at 9,, 9 and kv cm ), respectively, for ls, in citrate phosphate buffer of different ph and recovered in the nonselective (s) and the selective (d) medium log of survival fraction log of survival fraction (a) 7 (c) 7 log of survival fraction log of survival fraction (b) 7 (d) 7 ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

7 D. GARCÍA ET AL. thally injured cells was also at its maximum at the ph in which micro-organisms showed the highest PEF resistance. There was also more than 9% of sublethally injured cells when Salm. Senftenberg 77W was PEF treated at ph Æ. However, it should be noted that, with the exception of L. monocytogenes, the survival fraction obtained after recovery in the selective media was practically independent of the ph of the treatment medium. Occurrence of sublethal injury after PEF treatments in other Gram-negative bacteria As the behaviour of the Gram-negative bacteria tested was very unusual because of their higher resistance under acidic conditions, the study was extended to other four Gram-negative bacterial strains. Figure shows the number of log cycles of inactivation of Y. enterocolitica, Salm. Typhimurium, Ps. aeruginosa and E. coli O7:H7 after PEF treatments at different intensities in citrate phosphate buffer of ph Æ and7æ, and recovered in the nonselective and the selective medium. (a) log cycles of inactivation (c) log cycles of inactivation 7 7 (b) log cycles of inactivation (d) 7 7 Fig. Log cycles of inactivation of Yersinia enterocolitica (a), Salmonella serotype Typhimurium (b), Pseudomonas aeruginosa (c) and Escherichia coli O7:H7 (d) cells after PEF treatments at kv cm ) for ls, 9 kv cm ) for ls, kv cm ) for ls and kv cm ) for ls, respectively, in citrate phosphate buffer of ph Æ and 7Æ and recovered in the nonselective (white bars) and the selective (black bars) medium log cycles of inactivation As shown in Fig., the four Gram-negative bacterial strains investigated were also more PEF resistant when treated at ph Æ showing from 9 to 99Æ9% of sublethally injured cells. DISCUSSION Most published data (Simpson et al. 999; Dutreux et al. a,b; Russell et al. ; Ravishankar et al. ; Ulmer et al. ; Wuytack et al. ; Aronsson et al. ) have not demonstrated the occurrence of sublethal injury after PEF treatments using the selective medium plating technique, concluding that bacterial inactivation by PEF may be an Ôall or nothingõ event. In the present investigation, sublethal injury has been detected after PEF depending on the treatment conditions and the micro-organism investigated. In fact, a characteristic behaviour has been described for the two Gram-positive bacteria studied, which is the opposite of the response for the six Gram-negative ones. As mentioned above, the study was carried out on six Gram-negative strains in order to confirm their unusual microbial resistance under acidic conditions rather than under neutral. Despite the fact some micro-organisms were more PEF resistant than others, the two Gram-positive bacteria were more PEF resistant at ph 7Æ than at ph Æ, showing a high proportion of sublethally injured cells under these treatment conditions. In contrast, the six Gram-negative bacteria were more PEF resistant at ph Æ than at ph 7Æ, and only showed sublethally injured cells when treated at ph Æ. These results suggest that bacterial inactivation by PEF is not an Ôall or nothingõ event and in any case, it would depend on the micro-organism and the treatment conditions investigated. Sublethal injury cells were not detected when B. subtilis ssp. niger and L. monocytogenes were PEF treated at ph Æ or Æ, or E. coli and Salm. Senftenberg 77W were treated at ph Æ or7æ. Also, independently of the micro-organism and the treatment medium ph investigated, the application of short treatment times or low electric field strengths did not cause extensive sublethal injury. The nonexistence of sublethal injury under these conditions might explain most published data as in most cases, Gram-negative bacteria have been PEF treated at neutral ph: Salm. Typhimurium in distilled water or mmol l ) Tris-maleate buffer at ph 7Æ (Simpson et al. 999; Russell et al. ) or in HEPES buffer at ph 7Æ (Wuytack et al. ), E. coli in phosphate buffer at ph Æ8 or fat-free milk at ph Æ8 (Dutreux et al. a) or in nutritive treatment medium at ph Æ, Æ or7æ (Aronsson et al. ). In some other cases, Gram-positive bacteria have been PEF treated at acid ph: Lact. plantarum in model beer at ph Æ oræ (Ulmer et al. ). Therefore, the results of most published data would agree with conclusions of this study. However, ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

8 SUBLETHAL INJURY IN PEF-TREATED CELLS Simpson et al. (999) and Russell et al. () did not detect sublethal injury after PEF when L. monocytogenes was PEF treated in distilled water or mmol l ) Tris-maleate buffer at ph 7Æ. In our opinion, the electric field strength applied, kv cm ), might be not have been high enough. Our results have revealed that more than 9% of sublethally injured L. monocytogenes cells were detected when at least kv cm ) for ls were applied. Preliminary results have shown that other factors such us growth temperature or the treatment medium composition may also influence PEF resistance and the magnitude of the occurrence of sublethal injury in Gram-positive or Gramnegative bacteria (data not shown). For example, L. monocytogenes cells grown at C were as PEF resistant as those grown at 7 C, whereas the percentage of sublethally injured cells after ls atkvcm ) increased by two more log cycles, up to 99Æ9% of survivors. Regarding the influence of the treatment medium composition, L. monocytogenes or E. coli PEF resistance dramatically decreased and sublethal injury completely disappeared when Æ% of NaCl was present in the treatment media at the same conductivity. Any contribution to the occurrence of sublethal injury because of overheating or medium electrolysis during PEF treatment should be discarded. The PEF equipment employed includes provisions for measuring sample temperature immediately after PEF treatment. The temperature was always lower than C. Moreover, as shown in Fig. or Fig., for each micro-organism two treatment media were used under the same experimental conditions: one at ph and the other at ph 7. Both media had exactly the same conductivity so, under the same PEF treatment, the same electric energy, and hence the same overheating or medium electrolysis were delivered to the treatment medium. However, sublethal injury was detected in both media depending on the micro-organism investigated. Sublethal injury measured using a selective medium plating technique is supposed to be a consequence of loss of membrane integrity and functionality (Mackey ). Hence, this study would also confirm that membrane damage is an important event in bacterial inactivation by PEF. As the mechanism of microbial inactivation by PEF seems to be related to the effects on the cell envelopes, it seems reasonable to think that the differences in the structure and the composition between the cell wall of Gram-positive and Gram-negative bacteria are responsible for the different behaviour observed under PEF. More research is needed in order to determine the mechanisms of inactivation by PEF of Gram-positive and Gram-negative bacteria. Regardless of the influence of the ph of the treatment medium on microbial PEF resistance, data reported in the literature are inconclusive. Whereas according to several authors (Sale and Hamilton 97; Hülsheger et al. 98; Heinz and Knorr ; Ravishankar et al. ; Álvarez et al. ) the ph of treatment media did not affect microbial inactivation by PEF, some micro-organisms such as E. coli, Listeria innocua, L. monocytogenes or Lact. plantarum were more PEF sensitive at acidic than at neutral ph (Vega-Mercado et al. 99; Wouters et al. 999, ; Aronsson and Rönner ; Álvarez et al. ; Aronsson et al. ), and others, such as Salm. Enteritidis, Salm. Senftenberg 77W or E. coli, more PEF resistant at acidic than at neutral ph (Jeantet et al. 999; Álvarez et al. ; García et al. a,b). As shown in Figs,, and, our results prove that the effect of the ph of the treatment media depends on the electric field strength and the micro-organism investigated, particularly whether it is Gram-positive or Gram-negative bacteria. Moreover, during this investigation it has been observed that when PEF-treated Gram-negative cells were not immediately plated onto nonselective recovery agar, sublethally injured cells became sensitive to the subsequent maintenance in the treatment medium being inactivated exponentially with time. Under these conditions sublethally injured cells die, and are therefore not detected, so the differences in PEF resistance disappear and the effect of treatment medium ph is not observed. This fact was previously described in PEF-treated E. coli cells by García et al. (a). Therefore, when a study is carried out with different bacterial species in order to determine their relative PEF resistance in the same treatment conditions, this aspect should be considered as a possible cause of the differences observed. However, results of this study indicate that the higher PEF resistance observed in the Gram-positive bacteria at ph 7Æ and the Gram-negative bacteria at ph Æ seemed to be related to the capacity of these cells to repair sublethal injuries. As observed in Fig., or, in most cases the number of survivors at the ph of the lowest PEF resistance and the number of survivors recovered in the selective medium at the ph of highest PEF resistance coincided. It seems that micro-organisms were more PEF resistant at a certain ph because of the possibility of repairing sublethal injuries. In order to explain these results two hypotheses are suggested: Either (i) the intrinsic resistance of Grampositive and Gram-negative bacteria to PEF is different depending on the ph of the treatment medium, which would means that lesions inflicted by PEF at ph Æ in Gram-positive and at ph 7Æ in Gram-negative bacteria are not repairable; or (ii) the mechanisms of sublethal injury repair are directly affected by PEF at the ph of the lowest resistance and cells are not able to repair them. More research is needed in order to determine the mechanisms by which sublethal injury might be repairable. ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

9 D. GARCÍA ET AL. Nevertheless, when the selective medium was used as a recovery medium, among bacterial species investigated there still exist differences in PEF resistance under the same experimental conditions. So, as suggested by other authors, the intrinsic microbial resistance might also be related to other factors, such as the type of micro-organism (Hülsheger et al. 98; Wouters et al. 999; MacGregor et al. ), their cell size and shape (Kehez et al. 99; Qin et al. 998; Heinz et al. ), etc. Despite the apparent inefficacy on inactivating Grampositive bacteria at ph 7Æ or Gram-negative at ph Æ at the higher PEF intensity tested, results of this study have demonstrated that a large proportion of survivors to the PEF treatment were sublethally injured and thus susceptible to being inactivated by other hurdles used in combination with PEF. In this way, if the possibility of repairing sublethal injury were avoided, the maximum microbial inactivation could be reached independently of the influence of the ph of the treatment medium, which would simplify the use of this technology as a food preservation method. García et al. (a) have demonstrated that the combination of PEF treatment applied in McIlvaine buffer at ph Æ for ls at9kvcm ) and a subsequent storage at room temperature for h caused the inactivation of more than 99Æ99% of E. coli cells. Similar results were obtained after PEF treating Salm. Senftenberg 77W in the same treatment media for ls at kv cm ) (García et al. b). In both cases, the number of sublethally injured cells after PEF treatment and the number of extra dead cells after the subsequent storage coincided. As described by García et al. (a,b), these results would indicate that the sensitivity of PEF-treated Gram-negative cells to a subsequent acid incubation is also related to the occurrence of sublethal injury. In addition, Zerrouk et al. () have demonstrated that the combination of a PEF treatment at kv cm ) for ls in apple juice (ph Æ8) and a subsequent storage of the apple juice under refrigeration for 8 h allowed log cycles of inactivation to be achieved in the population of E. coli O7:H7. Therefore, the combination of PEF and maintenance under refrigeration of acid foods might be an effective pasteurization method, by sufficiently reducing the presence of spoilage and pathogenic micro-organisms. Gram-positive bacteria PEF treated at ph 7 were not sensitive to a subsequent incubation in the same treatment medium. However, a subsequent incubation of survivors at ph Æ demonstrated their sensitivity to acid incubation as injured cells were also inactivated (data not shown). These results would indicate that the lesions inflicted by PEF are unspecific, making either Gram-positive or Gram-negative bacterial cells sensitive to a subsequent incubation under stress conditions, such as the presence of low concentrations of salt or the reduction of the ph of the medium. We have described for the first time the important differences in the way that PEF might affect Gram-positive and Gram-negative bacteria. Our results have demonstrated the presence of a large proportion of sublethally injured cells after PEF treatments depending on the type of bacteria, the ph of the treatment medium, the treatment time and the electric field strength investigated. Moreover, these studies confirm that membrane damage is an important event in the inactivation of bacteria by PEF. The fact that sublethal injury has been detected after PEF treatments is a very relevant aspect to be taken into account, as from a practical point of view the occurrence of sublethally injured cells would mean that PEF might act synergistically with many other hurdles, such as antimicrobial substances, modified atmospheres, etc., proving to be valuable in developing appropriate combination processes. ACKNOWLEDGEMENTS This work was supported by CICYT (Project AGL ) and Ministerio Español de Educación, Cultura y Deporte, and Ciencia y Tecnología, which provided D. García and N. Gómez with a grant to carry out this investigation. Our thanks to Mrs P. Noone for her collaboration in the English correction of this work. REFERENCES Álvarez, I., Raso, J., Palop, A. and Sala, F.J. () Influence of different factors on the inactivation of Salmonella senftenberg by pulsed electric fields. Int J Food Microbiol,. Álvarez, I., Pagán, R., Condón, S. and Raso, J. () Environmental factors influencing the inactivation of Listeria monocytogenes by pulsed electric fieldsõ. Lett Appl Microbiol, Álvarez, I., Raso, J., Sala, F.J. and Condón, S. () Inactivation of Yersinia enterocolitica by pulsed electric fields. Food Microbiol, 9 7. Aronsson, K. and Rönner, U. () Influence of ph, water activity and temperature on the inactivation of Escherichia coli and Saccharomyces cerevisiae by pulsed electric fields. Innov Food Sci Emerg Technol,. Aronsson, K., Borch, E., Stenlöf, B. and Rönner, U. () Growth of pulsed electric field exposed Escherichia coli in relation to inactivation and environmental factors. Int J Food Microbiol 9,. Barbosa-Cánovas, G.V., Góngora, M.M., Pothakamury, U.R. and Swanson, B.G. (999) Preservation of Foods with Pulsed Electric Fields. San Diego, CA: Academic Press. Bendicho, S., Giner, J., Barbosa-Cánovas, G.V and Martín, O. () Procesado de leche mediante aplicación de pulsos eléctricos de alta intensidad de campo (Revisión). Alimentaria, 9. Chapman, P.A., Siddons, C.A.., Wright, D.J., Norman, P., Fox, J. and Crick, E. (99) Cattle as a possible source of verocytotoxin-producing Escherichia coli O7 infections in man. Epidemiol Infect, 9 7. Condón, S., Palop, A., Raso, J. and Sala, F.J. (99) Influence of the incubation temperature after heat treatment upon the estimated heat ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

10 SUBLETHAL INJURY IN PEF-TREATED CELLS resistance values of spores of Bacillus subtilis. Lett Appl Microbiol, 9. Damar, S., Bozoglu, F., Hizal, M. and Bayindirli, A. () Inactivation and injury of Escherichia coli O7:H7 and Staphylococcus aureus by pulsed electric fields. World J Microbiol Biotechnol 8,. Dawson, R.M.C., Elliot, D.C., Elliot, W.H. and Jones, K.M. (97) Data for Biochemical Research. Oxford, UK: Oxford at Clarendon Press. Dutreux, N., Notermans, S., Góngora-Nieto, M.M., Barbosa-Cánovas, G.V. and Swanson, B.G. (a) Effects of combined exposure of Micrococcus luteus to nisin and pulsed electric fields. Int J Food Microbiol, 7. Dutreux, N., Notermans, S., Wijtzes, T., Góngora-Nieto, M.M, Barbosa-Cánovas, G.V. and Swanson, B.G. (b) Pulsed electric fields inactivation of attached and free-living Escherichia coli and Listeria innocua under several conditions. Int J Food Microbiol, Evrendilek, G.A., Zhang, Q.H. and Richter, E.R. (999) Inactivation of Escherichia coli O7:H7 and Escherichia coli 879 in apple juice by pulsed electric fields. J Food Prot, García, D., Gómez, N., Condón, S., Raso, J. and Pagán, R. (a) Pulsed electric fields cause sublethal injury in Escherichia coli. Lett Appl Microbiol,. García, D., Gómez, N., Mañas, P., Condón, S., Raso, J. and Pagán, R. (b) Occurrence of sublethal injury in Salmonella senftenberg under pulsed electric fields. In IFT Annual Meeting, July, Illinois. pp.. Chicago, IL. Hamilton, W.A. and Sale, A.J.H. (97) Effects of high electric fields on microorganisms: II. Killing of bacteria and yeasts. Biochim Biophys Acta 8, Heinz, V. and Knorr, D. () Effect of ph, ethanol addition and high hydrostatic pressure on the inactivation of Bacillus subtilis by pulsed electric fields. Innov Food Sci Emerg Technol, 9. Heinz, V., Alvarez, I., Angersbach, A. and Knorr, D. () Preservation of liquid foods by high intensity pulsed electric fields basic concepts for process design. Trends Food Sci Technol,. Hülsheger, H., Potel, J. and Niemann, E.G. (98) Killing of bacteria with electric pulses of high field strength. Radiat Environ Biophys,. Iu, J., Mittal, G.S. and Griffiths, M.W. () Reduction in levels of Escherichia coli O7:H7 in apple cider by pulsed electric fields. J Food Prot, Jeantet, R., Baron, F., Nau, F., Roignant, M. and Brulé, G. (999) High intensity pulsed electric fields applied to egg white: effect on Salmonella enteritidis inactivation and protein denaturation. J Food Prot, 8 8. Kehez, M.M., Savic, P. and Johnson, B.F. (99) Contribution to the biophysics of the lethal effects of electric field on microorganisms. Biochim Biophys Acta 78, Leistner, L. and Gorris, L.G.M. (99) Food preservation by hurdle technology. Trends Food Sci Technol,. Liang, Z., Mittal, G.S. and Griffiths, M.W. () Inactivation of Salmonella Typhimurium in orange juice containing antimicrobial agents by pulsed electric fields. J Food Prot, van Loey, A., Verachtert, B. and Hendrickx, M. () Effects of high electric pulses on enzymes. Trends Food Sci Technol, 9. MacGregor, S.J., Farish, O., Fouracre, R., Rowan, N.J. and Anderson, J.G. () Inactivation of pathogenic and spoilage microorganisms in a test liquid using pulsed electric fields. IEEE Trans Plasma Sci 8, 9. Mackey, B.M. () Injured bacteria. In The Microbiological Safety and Quality of Food. Vol. I. ed. Lund, M., Baird-Parker, T.C. and Gould, G.W. pp.. Gaithersburg, MD: Aspen Publisher, Inc. National Advisory Committee on Microbiological Criteria for Food (NACMCF) (997) Recommendations for Controlling the Transmission of Pathogenic Microorganisms in Juices. Washington, DC: Food Safety and Inspection Service, US Department of Agriculture. Qin, B.-L., Barbosa-Cánovas, G.V., Swanson, B.G., Pedrow, P.D. and Olsen, R.G. (998) Inactivating microorganisms using a pulsed electric field continuous treatment system. IEEE Trans Ind Appl,. Raso, J. and Barbosa-Cánovas, G.V. () Nonthermal preservation of foods using combined processing techniques. Crit Rev Food Sci Nutr, 8. Raso, J., Alvarez, I., Condón, S. and Sala, F.J. () Predicting inactivation of Salmonella senftenberg by pulsed electric fields. Innov Food Sci Emerg Technol, 9. Ravishankar, S., Fleischman, G.J. and Balasubramaniam, V.M. () The inactivation of Escherichia coli O7:H7 during pulsed electric field (PEF) treatment in a static chamber. Food Microbiol 9,. Russell, N.J., Colley, M., Simpson, R.K., Trivett, A.J. and Evans, R.I. () Mechanism of action of pulsed high electric field (PHEF) on the membranes of food-poisoning bacteria is an Ôall-or-nothingÕ effect. Int J Food Microbiol,. Sale, A.J.H. and Hamilton, W.A. (97) Effects of high electric fields on microorganisms I. Killing of bacteria and yeast. Biochim Biophys Acta 8, Simpson, R.K., Whittington, R., Earnshaw, R.G. and Russell, N.J. (999) Pulsed high electric field causes all or nothing membrane damage in Listeria monocytogenes and Salmonella typhimurium, but membrane H + -ATPase is not a primary target. Int J Food Microbiol 8,. Stringer, S.C., George, S.M. and Peck, M.W. () Thermal inactivation of Escherichia coli O7:H7. J Appl Microbiol Symp Suppl 88, 79S 89S. Ulmer, H.M., Heinz, V., Gänzle, M.G., Knorr, D. and Vogel, R.F. () Effects of pulsed electric fields on inactivation and metabolic activity of Lactobacillus plantarum in model beer. J Appl Microbiol 9,. Unal, R., Kim, J.-G and Yousef, A.E. () Inactivation of Escherichia coli O7:H7, Listeria monocytogenes, and Lactobacillus leichmannii by combinations of ozone and pulsed electric field. J Food Prot, Vega-Mercado, H., Pothakamury, U.R., Chang, F-J., Barbosa-Cánovas, G.V. and Swanson, B.G. (99) Inactivation of Escherichia coli by combining ph, ionic strength and pulsed electric fields hurdles. Food Res Int 9, 7. Weaver, J.C. and Chizmadzhev, Y.A. (99) Theory of electroporation: a review. Bioelectrochem Bioenerg,. Wouters, P.C. and Smelt, J.P.P.M. (997) Inactivation of microorganisms with pulsed electric fields: potential for food preservation. Food Biotechnol, 9 9. ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x

11 D. GARCÍA ET AL. Wouters, P.C., Dutreux, N., Smelt, J.P.P.M. and Lelieveld, H.L.M. (999) Effects of pulsed electric fields on inactivation kinetics of Listeria innocua. Appl Environ Microbiol, 7. Wouters, P.C, Alvarez, I. and Raso, J. () Critical factors determining inactivation kinetics by pulsed electric field food processing. Trends Food Sci Technol,. Wuytack, E.Y., Duong Thi Phuong, L., Aertsen, A., Reyns, K.M.F., Marquenie, D., De Ketelaere, B., Masschalck, B., Van Opstal, L. et al. () Comparison of sublethal injury induced in Salmonella enterica serovar Typhimurium by heat and by different nonthermal treatments. J Food Prot, 7. Zerrouk, M., García, D., Alvarez, I., Condón, S. and Pagán, R. () Inactivation of Escherichia coli O7:H7 during the storage under refrigeration of apple juice treated by pulsed electric fields. In I Congreso Internacional de Ciencia, Tecnología y Seguridad Alimentaria, 7 November. pp. 9. Navarra, Pamplona, Spain. ª The Society for Applied Microbiology, Journal of Applied Microbiology, 99, 9, doi:./j.-7...x