Synergistic and Antagonistic Effects of Combined Subzero ...

0 downloads 0 Views 156KB Size Report
Aug 2, 2005 - The combined effects of subzero temperature and high pressure on the ... The amount of mechanical energy transferred to the cell system is ...
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 2006, p. 150–156 0099-2240/06/$08.00⫹0 doi:10.1128/AEM.72.1.150–156.2006 Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Vol. 72, No. 1

Synergistic and Antagonistic Effects of Combined Subzero Temperature and High Pressure on Inactivation of Escherichia coli Marwen Moussa, Jean-Marie Perrier-Cornet,* and Patrick Gervais Laboratoire de Ge´nie des Proce´de´s Alimentaires et Biotechnologiques, ENSBANA, Universite´ de Bourgogne, 1, Esplanade Erasme, 21000 Dijon, France Received 2 August 2005/Accepted 26 September 2005

The combined effects of subzero temperature and high pressure on the inactivation of Escherichia coli K12TG1 were investigated. Cells of this bacterial strain were exposed to high pressure (50 to 450 MPa, 10-min holding time) at two temperatures (ⴚ20°C without freezing and 25°C) and three water activity levels (aw) (0.850, 0.992, and ca. 1.000) achieved with the addition of glycerol. There was a synergistic interaction between subzero temperature and high pressure in their effects on microbial inactivation. Indeed, to achieve the same inactivation rate, the pressures required at ⴚ20°C (in the liquid state) were more than 100 MPa less than those required at 25°C, at pressures in the range of 100 to 300 MPa with an aw of 0.992. However, at pressures greater than 300 MPa, this trend was reversed, and subzero temperature counteracted the inactivation effect of pressure. When the amount of water in the bacterial suspension was increased, the synergistic effect was enhanced. Conversely, when the aw was decreased by the addition of solute to the bacterial suspension, the baroprotective effect of subzero temperature increased sharply. These results support the argument that water compression is involved in the antimicrobial effect of high pressure. From a thermodynamic point of view, the mechanical energy transferred to the cell during the pressure treatment can be characterized by the change in volume of the system. The amount of mechanical energy transferred to the cell system is strongly related to cell compressibility, which depends on the water quantity in the cytoplasm. ters, a synergistic effect could be achieved, reducing the pressures and treatment times required. The combined effects of high pressure and low or subzero temperatures on microbial inactivation have been studied by some authors. A synergistic effect between these parameters has generally been reported in the inactivation of microorganisms in the vegetative state (9, 11, 34, 39, 41). In some cases, the initial microbial populations were completely inactivated with a combined treatment of high pressure and low or subzero temperature, whereas only a slight microbial inactivation was achieved under the same pressure conditions at room temperature (34). The magnitude of this synergistic effect is strongly dependent on the type of microorganism (41). The interaction of high pressures and subzero temperatures in microbial inactivation is complex, and possible phase-transition phenomena must be taken into account. Some authors have recently demonstrated that freezing under hyperbaric conditions is an effective way to reduce microbial contamination (20, 35). In addition to the antimicrobial effects of combining high pressure and subzero temperature treatments, these treatments when combined offer various processing advantages such as rapid freezing and thawing and cold storage of foods under liquid conditions. These applications have been reviewed by a number of authors (8, 16, 17). Although high pressure combined with subzero temperatures appears to be a promising field of investigation, the efficiency of this combination on microbial inactivation has not yet been thoroughly studied. Moreover, there remains a major unanswered question about the mechanisms involved in the combined pressure-temperature inactivation of microorganisms.

Food processing under high hydrostatic pressure is an emerging technology that has stimulated considerable interest within the food industry over the past 15 years. There are currently some interesting commercial opportunities and research challenges in the high-pressure processing of foods (42). This processing technique allows the manufacture of innovative food products while preserving the texture, color, raw flavoring agents, and nutritional value of the food, which are all aspects valued by consumers (22, 36). Furthermore, high-pressure treatments cause the denaturation of several enzymes that are responsible for quality deterioration, as well as the inactivation of pathogenic and spoilage microorganisms (15). However, to achieve high or complete microbial inactivation, high pressures and/or long treatment times are required, so that the cost of the process seriously limits its industrial applications. For this reason, it would be beneficial to optimize high-hydrostatic-pressure processes. Numerous studies have demonstrated the temperature dependence of the antimicrobial effects of high pressure (19, 25, 39). Moreover, the efficiency of high-pressure treatments is controlled by other process parameters such as the applied pressure and the kinetics of pressurization (31), as well as by the physicochemical properties of the medium being treated, such as pH (1, 18) and water activity (10, 26, 43). Precise control of these parameters is necessary to ensure efficient treatment. With appropriate combinations of these parame-

* Corresponding author. Mailing address: Laboratoire de Ge´nie des Proce´de´s Alimentaires et Biotechnologiques, ENSBANA, Universite´ de Bourgogne, 1, Esplanade Erasme, 21000 Dijon, France. Phone: 33 3 80 39 68 45. Fax: 33 3 80 39 68 98. E-mail: [email protected]. 150

HIGH-PRESSURE E. COLI INACTIVATION AT SUBZERO TEMPERATURE

VOL. 72, 2006

151

FIG. 1. Schematic diagram of the experimental setup for combined high pressure and subzero temperature treatments.

The aim of the present study was to investigate the interaction between the effects of subzero temperatures and high pressures on the inactivation of Escherichia coli K12TG1. The bacterium was exposed to high-pressure treatments (50 to 450 MPa, 10-min holding time) at ambient and subzero temperatures (⫺20°C under liquid conditions). The combined high pressure-low temperature treatments were performed at different hydration levels (aw 0.850, 0.992, and ca. 1.000) to clarify the interaction between high pressure and low temperature and to examine the involvement of the thermodynamic properties of water in the high-pressure inactivation of microorganisms. MATERIALS AND METHODS Bacterium and culture conditions. Escherichia coli K-12TG1 [supE hsd ⌬5 thi ⌬(lac-proAB) F⬘(traD36 proA⫹B⫹ lacIq lacZ⌬M15)] (Laboratoire de Microbiologie, Ensbana, Dijon, France) was used as the bacterial model. This strain of E. coli was chosen based on its high survival variability in the studied highpressure range. A stock culture maintained at ⫺80°C in a 20% glycerol solution was subcultured onto Luria-Bertani (LB) agar plates (Sigma Aldrich, France) every 3 weeks. These plates were incubated at 37°C for 24 h and then stored at 4°C until subcultures were required for experiments. Liquid subcultures were prepared by transferring a single colony of E. coli from the subculture plate to autoclaved test tubes containing 9 ml of LB broth at approximately pH 6.7 (not adjusted) and an aw of 0.992 ⫾ 0.003. These tubes were then incubated statically for 16 h at 37°C. Liquid cultures were then prepared by injecting 0.2 ml of subculture into 20 ml of LB broth, and this was grown statically at 37°C for 24 h, until it reached stationary phase (optical density at 600 nm ⫽ 0.7). High-pressure treatments. (i) Sample preparation. Samples of about 800 ␮l of liquid bacterial culture were transferred aseptically to polyethylene bags (Samco, United Kingdom). These were heat sealed after exclusion of air bubbles using a TEW TISH-200 impulse sealer (Electric Heating Equipment Co., Ltd., Taiwan). Samples were then placed at room temperature and pressure until pressurization. Controls were prepared in the same way to evaluate any bacterial inactivation that occurred during sample preparation. The final cell concentration ranged from 2 ⫻ 108 to 5 ⫻ 108 CFU ml⫺1. For some experiments, pressurization was carried out in medium with reduced aw (0.850 ⫾ 0.003). This was performed in order to decrease the intracellular water quantity by replacing it partially with glycerol (Sigma Aldrich, 99% [wt/

vol]), an aw depressor known to be freely permeant. The concentration of glycerol (781.49 g liter⫺1) necessary to lower water activity was calculated with the Norrish equation (27). Samples (20 ml) of bacterial culture were centrifuged twice at 2,880 ⫻ g for 5 min. After each centrifugation step, the pellet was resuspended in 19 ml of sterile water-glycerol solution. Pressurization was also performed in medium with high aw (ca. 1.000 ⫾ 0.003). Similarly, the cell pellet was harvested with the same centrifugation conditions and resuspended in sterile distilled water. The handling of bacteria and all centrifugation steps were carried out under ambient conditions. The water activity of the suspensions was measured with an AquaLab-CX2 osmometer (⫾0.003 accuracy; Decagon Devices) before sample preparation. (ii) Pressure and temperature monitoring. Figure 1 shows a schematic diagram of the experimental setup. High-pressure treatments were performed by using a high-pressure vessel (Mazilly Constructions Me´caniques, Venarey-LesLaumes, France; 80-mm outside diameter, 10-mm inside diameter, 100-mm height, and 55-mm depth). The inner volume of the chamber was 4 ml with a pressure tolerance of up to 600 MPa. A hand pump (Novaswiss, Switzerland) was used for the high-pressure treatment, with 97% ethanol as the pressure-transmitting fluid. The bags used to hold the microbial samples were checked regularly to ensure that there was no leakage of ethanol. The pressure inside the sample chamber was monitored by using a pressure gauge (Sedeme, France). The temperature was measured by using a type K thermocouple (NiCr/NiAl; response time, 70 ms; Thermocoax, France) passed through the upper plug of the vessel and positioned close to the sample. Pressure and temperature data were recorded over the entire pressurization period with a data acquisition device (instruNet; GW Instruments, Massachusetts). The global accuracy of each acquisition channel was about 0.5°C for temperature and 0.5% for pressure. (iii) High-pressure treatment at room temperature. Pressures of up to 450 MPa were applied to the bacterial samples. The pressure was increased by slowly operating the hand pump. Temperature was maintained at 25°C by immersing the high-pressure vessel in a temperature-controlled water bath. Compression heat was minimal (⬍2°C) at all pressure targets, and this was rapidly dissipated due to the slow compression rate and the temperature-controlled water bath. Treatment time was measured from immediately after the target pressure was reached. Pressure was then held constant for 10 min before the decompression phase was initiated. The come-up and come-down times were 3 min whatever the pressure-target. (iv) High-pressure treatment at subzero temperatures in liquid conditions. Combined high pressure and subzero temperature treatments were performed as described by Perrier-Cornet et al. (34). In brief, temperature was lowered by first immersing the high-pressure vessel in a cryostat (F81-HP; Julabo, Germany) that

152

APPL. ENVIRON. MICROBIOL.

MOUSSA ET AL.

RESULTS

FIG. 2. Pressure (light line) and temperature (heavy line) profiles for combined high pressure and subzero temperature treatment (200 MPa, ⫺20°C, 10 min).

was loaded with ethanol and kept at ⫺60°C in order to accelerate heat transfer. The cooling rate was about 20°C min⫺1. Pressure was increased while the temperature was lowered. The high-pressure vessel was then transferred to a second cryostat (RC6CP; Lauda, Germany) that was loaded with ethanol and maintained at ⫺20°C to maintain a constant temperature during the treatment process. The target pressure (50 to 450 MPa) and temperature (⫺20°C) were reached at approximately the same time. For treatments at atmospheric pressure, the temperature was lowered to ⫺20°C in the supercooled region as facilitated by the experimental design. Precise pressure and temperature monitoring of the samples was carried out to ensure that the samples did not freeze. After the required holding time (10 min) at the appropriate temperature and pressure, the vessel was reheated by immersion in a water bath maintained at 27°C. The pressure was simultaneously released by slowly operating the hand pump. Ambient pressure and temperature were reached at approximately the same time (Fig. 2). Viable cell counts. The viability of E. coli K12TG1 was determined by counting CFU on LB agar plates. Pressurized samples were serially diluted and appropriate dilutions were then plated onto two LB agar plates each. The plates were incubated for 22 h at 37°C, after which the CFU were counted. The numbers of colonies on the two replicate plates were averaged. The microbial inactivation achieved with each treatment was expressed as the logarithmic viability decrement: log10(N/N0), where N is the number of CFU ml⫺1 after treatment and N0 is the number of CFU ml⫺1 of untreated bacteria. Results are reported as the means of three to six separate experiments. The 90% confidence intervals of the means were determined based on N/N0 quotient calculations. A Student test was performed to determine statistical significance (P ⬍ 0.05) between the inactivation data under different high-pressure treatments. Compressibility measurements. Compressibility measurements were made by using a high-pressure setup for differential thermal analysis designed by the Institute of High-Pressure Physics (Warsaw, Poland). This apparatus was designed to measure volume, temperature, and pressure simultaneously. It consists of two high-pressure cells (piston cylinder devices of about 5 ml of inner volume and 700 MPa pressure tolerance), two actuators, a support with two movement transducers, and a hand pump equipped with a pressure transducer (Sitec, Switzerland). The lid of each cell is equipped with a venting valve and the base is equipped with three copper-constantan thermocouples to monitor the temperature directly inside the cells. The actuators, which are mounted on the support structure and are provided with two movement transducers, are piped from the pump unit through capillary tubes. They generate high pressure in the sample cells and allow precise on-line monitoring of sample volume. The sample cell was filled with a known volume of degassed liquid (water or water-glycerol mixture). After the sample cell was mounted into the actuator, this latter was placed on the support. The whole system was immersed in a temperature-controlled bath. Isothermal measurements of variations in volume (␦V) with changes in pressure (␦P) were made and used to calculate isothermal compressibility according to the following equation: KT ⫽ ⫺1/V (␦V/␦P)T.

High-pressure inactivation of E. coli K12TG1 in standard culture medium (aw ⴝ 0.992) at 25°C and ⴚ20°C. Bacterial samples were subjected to different hyperbaric treatments. Pressure levels ranging from 50 to 450 MPa were applied to each sample for 10 min at either 25°C or ⫺20°C. Cells were pressurized in the standard culture medium (aw ⫽ 0.992). For experiments at ⫺20°C, the pressure treatments were performed under supercooled conditions. As previously described by Perrier-Cornet et al. (34), the very small volume of the vessel and the static conditions ensured that no crystallization occurred in the liquid even though the temperature was less than the liquid’s theoretical freezing point. Indeed, changes in pressure and temperature in the treatment vessel were used to indicate the onset of freezing in the samples. If freezing (hexagonal ice I) occurred, for example, increases in pressure (⬎5 MPa) and in temperature (2°C) were immediately recorded. The accuracy of this method was verified by measuring the optical properties of the sample with an optical high-pressure device designed by the Institute of High-Pressure Physics (Warsaw, Poland). Figure 3A shows the inactivation rate of E. coli K12TG1 as a function of pressure and temperature. At room temperature, pressures of up to 150 MPa did not induce significant loss of viability (P ⬍ 0.05). At higher pressures, microbial inactivation increased approximately linearly with pressure and reached a maximum of 5.3 log cycles at 450 MPa. At ⫺20°C and atmospheric pressure, when the sample remained in a liquid state, a slight inactivation of E. coli K12TG1 was observed (0.2 log cycle, i.e., 37%). This microbial inactivation was induced solely by the effect of subzero temperature because no crystallization occurred. In some samples in which freezing occurred (data not shown), the mechanical stress associated with freezing resulted in a higher inactivation of E. coli K12TG1 (0.5 log cycle, i.e., 70%). This inactivation rate corresponds approximately to that observed by Suppes et al. (40) in E. coli DH5␣ after a 5-min treatment at ⫺20°C and atmospheric pressure under freezing conditions. The onset of pressure-induced inactivation at ⫺20°C began at 100 MPa. The logarithmic decrease in viability did not show a linear response to increasing pressure. Indeed, a pressure increase of 150 MPa, from 100 to 250 MPa and then from 250 to 400 MPa, resulted in an increase in the inactivation rate of about 2.8 and 0.5 log cycles, respectively. At pressures of up to 350 MPa, inactivation of E. coli was more effective at ⫺20°C than at 25°C. For example, the pressure required to reach an inactivation rate of 3 log cycles at ⫺20°C was about 100 MPa lower than that required at 25°C. On the other hand, a pressure of 250 MPa, for example, induced a significantly higher inactivation rate (P ⬍ 0.05) at ⫺20°C than at 25°C. The area delimited by the curves in Fig. 3A illustrates the synergistic interaction between high pressure and subzero temperature for pressures up to 350 MPa. This interaction reached a maximum at 250 MPa and then decreased. The two curves intersect slightly above 350 MPa. This pressure characterizes the equilibrium state between the synergistic and antagonistic effects observed at lower and higher pressures, respectively. At 400 and 450 MPa, the inactivation rate of E. coli was lower at

VOL. 72, 2006

HIGH-PRESSURE E. COLI INACTIVATION AT SUBZERO TEMPERATURE

FIG. 3. Inactivation of Escherichia coli K-12TG1 by high-pressure treatment for 10 min at 25°C (Œ) and ⫺20°C (F) at aw ⫽ 0.992 (A; standard culture medium), aw ⫽ 0.850 (B; bacteria suspended in glycerol-water solution), and aw ⫽ approximately 1.000 (C; bacteria suspended in distilled water). The dashed lines show the limit of detection by the CFU count.

⫺20°C than at 25°C (P ⬍ 0.05). Thus, subzero temperatures counteracted the effects of high pressures on microbial inactivation. The baroprotective effects of subzero temperatures can be compared to that conferred by a change in water activity. Therefore, we explored whether the protective effects of subzero temperatures persist in medium with reduced water activity. High-pressure inactivation of E. coli K12TG1 in medium with reduced water activity (aw ⴝ 0.850) at 25°C and ⴚ20°C. E. coli K12TG1 was pressurized in a sterile water-glycerol solution adjusted to a water activity of 0.850. Suspension of the bacterial pellet in this hyperosmotic medium did not induce any loss of viability, as confirmed by plate counts of control

153

samples. This corroborated the finding of Mille et al. (23) of 100% viability of E. coli K12TG1 after osmotic shock in liquid medium (final osmotic pressure of 26 MPa at 30°C, i.e., aw near 0.830). Pressure levels ranging from 50 to 450 MPa were applied for 10 min at 25°C and ⫺20°C (under liquid conditions). Figure 3B shows the inactivation rate of E. coli K12TG1 as a function of pressure and temperature at aw ⫽ 0.850. The pressure sensitivity of E. coli K12TG1 decreased significantly at reduced water activity (P ⬍ 0.05). For example, at 450 MPa and 25°C, the inactivation rate was about 3.5 log cycles at an aw of 0.850, whereas it exceeded 5 log cycles when the aw was 0.992 (Fig. 3A). At pressures of up to 250 MPa, there was no significant difference (P ⬍ 0.05) between microbial inactivation at 25°C and at ⫺20°C (⬍0.5 log cycles at both temperatures). At higher pressures, only an antagonistic interaction between pressure and subzero temperature was observed. The magnitude of this interaction, for example, at 450 MPa, was higher (P ⬍ 0.05) at an aw of 0.850 (3 log cycles) than under standard conditions (1.7 log cycle at an aw of 0.992, Fig. 3A). This set of experiments revealed a cumulative protective effect of reduced water activity and subzero temperature against pressure-induced inactivation. It is also of interest to investigate the coupled effects of high pressures and subzero temperatures on the inactivation of E. coli in a medium with high water activity. High-pressure inactivation of E. coli K12TG1 in medium with high water activity (aw of approximately 1.000) at 25°C and ⴚ20°C. Bacterial cells were harvested by centrifugation and suspended in sterile distilled water. Figure 3C shows the inactivation rate of E. coli K12TG1 as a function of pressure (from 50 to 450 MPa, 10-min holding time) and temperature (25°C and ⫺20°C, under liquid conditions) at an aw of approximately 1.000. It is noteworthy that suspending the bacteria in distilled water prior to pressure treatment caused no loss of viability. These observations corroborate the findings of Mitchell and Moyle (24), who reported that E. coli remained viable in distilled water if harvested in the stationary phase of growth, as was the case in the sample preparation for the present study. At 25°C, a pressure of 200 MPa induced microbial inactivation of less than 1 log cycle. A further increase in pressure of up to 350 MPa resulted in an increase in the inactivation rate of more than 8 log cycles. Interestingly, the combination of subzero temperature with high pressure enhanced microbial inactivation regardless of the pressure. Indeed, at ⫺20°C and 200 MPa, the viability of E. coli was reduced by 3.6 log cycles, more than fourfold the inactivation rate at 25°C. A pressure of 250 MPa was sufficient to decrease viability to below the detection limit. Hence, a synergistic interaction between high pressure and subzero temperature was observed, similar to that observed at an aw of 0.992 (Fig. 3A). This interaction made it possible to reduce the pressure significantly (P ⬍ 0.05) to achieve the same inactivation rate obtained at 25°C. However, no antagonistic effect between pressure and subzero temperature was observed. We can reasonably infer that this phenomenon will occur at pressures higher than those tested in the current experiment. It was not possible to measure the effects of higher pressures on microbial viability because the entire

154

MOUSSA ET AL.

initial population of E. coli K12TG1 was inactivated at either 250 MPa (⫺20°C) or 350 MPa (25°C). DISCUSSION This study evaluated the combined effects of high hydrostatic pressures and subzero temperatures on the viability of E. coli K12TG1. When bacterial cells were treated in culture medium (aw ⫽ 0.992) at 25°C, the inactivation rate reached 5.3 log cycles at 450 MPa, the highest pressure tested. The strain of E. coli studied here seemed to be less pressure sensitive than some other strains of the same bacterium harvested in stationary phase of growth (2, 30). At ⫺20°C, in the supercooled region, the pressure sensitivity of E. coli K12TG1 was greater than at 25°C. This synergism between high pressure and subzero temperature made it possible to reduce the pressure and/or improve the pressure-mediated inactivation. Irrespective of the inactivation rate, our findings corroborate the observations of Takahashi (41), who examined the inactivation of E. coli after pressure treatment (200 MPa, 20 min) at ⫺20°C and at room temperature. Takahashi observed a much higher inactivation rate at the subzero temperature (8 log cycles in samples that had previously been frozen) than at room temperature (4 log cycles). Hashizume et al. (9) reported that the same degree of inactivation of yeast could be achieved at a lower pressure at ⫺20°C (in samples that had previously been frozen) than at 25°C. More recently, Perrier Cornet et al. (34) reported that at a fixed pressure of 150 MPa, an initial population of Saccharomyces cerevisiae was completely inactivated at ⫺20°C (more than 8 log cycles under liquid conditions), whereas it was only slightly inactivated at 25°C (less than 0.5 log cycles). These authors also examined the inactivation of a gram-positive bacterium, Lactobacillus plantarum, and reported that at 150 MPa, less than 1 log cycle inactivation occurred at 25°C, whereas inactivation of more than 7 log cycles occurred at ⫺15°C. In contrast, in the present study, E. coli K12TG1 showed no loss of viability at up to 150 MPa and 25°C, whereas inactivation of approximately 1.4 log cycles occurred at ⫺20°C. This result obviously contradicts the generally accepted fact that gram-negative bacteria exhibit higher pressure sensitivity than gram-positive bacteria (2). Furthermore, the viability of E. coli K12TG1 was less affected by the synergism between high pressure and subzero temperature than were the viabilities of L. plantarum and S. cerevisiae (34). Unlike the effects in L. plantarum and S. cerevisiae, the magnitude of the synergistic effect in E. coli K12TG1 appeared to be pressure dependent. It increased gradually with increasing pressure up to 250 MPa and then decreased between 250 and 350 MPa. At ⬎350 MPa, the synergistic effect was completely neutralized by an antagonistic effect. Accordingly, E. coli K12TG1 cells were more resistant to higher than to lower pressure levels. A similar observation was described by Paga´n and Mackey (30) for E. coli H1071 cells in stationary phase of growth after pressure treatments at room temperature. The unusual pattern of survival of E. coli K12TG1 cells after combined high pressure and subzero temperature treatments was observed consistently in many experiments. These observations reflected a baroprotective effect at subzero temperatures. Pressure sensitivity of E. coli is related to intracellular water content. The pressure sensitivity of E. coli K12TG1 was highly

APPL. ENVIRON. MICROBIOL.

dependent on the water activity of the system. When the bacterium was suspended in a water-glycerol solution with an aw of 0.850, it appeared to be more pressure resistant than at an aw of 0.992. This finding underscores the baroprotective effect of solutes, previously described for E. coli (37, 43), Rhodotorula rubra (29), and Zygosaccharomyces bailii (32). The combination of subzero temperature and high pressure at an aw of 0.850 caused a cumulative protective effect of solute and subzero temperature against pressure-induced inactivation. However, when pressurized in a distilled water (aw of approximately 1.000), E. coli K12TG1 showed a much higher pressure sensitivity than at lower water activities, especially at ⫺20°C. Glycerol is a permeant solute that penetrates the cell by simple diffusion (24). Therefore, it can be inferred that water in the cell cytoplasm was partly replaced by glycerol when the bacterium was harvested in a water-glycerol solution. On the other hand, when the bacterium was harvested in distilled water, the amount of intracellular water increased. Accordingly, the pressure sensitivity of E. coli K12TG1 was dependent on the extent of water in the cell cytoplasm. Parallel change with pressure and temperature of protein behavior, microbial inactivation, and water structure. Several studies have highlighted the crucial role of water in the pressure-induced denaturation of biological systems. Oliveira et al. (28) reported that protein denaturation decreased linearly with a decrease in water concentration. Similarly, Kinsho et al. (13) observed that the removal of water by the addition of polyols or small cationic ions had an efficient protective effect against enzyme inactivation at high pressures and subzero temperatures. These latter authors also reported that cold-inactivation mechanisms were pressure dependent and differed at pressures less than 200 MPa from those at pressures greater than 200 MPa. Moreover, a maximum stability temperature was evidenced for different proteins and a bell-shaped dependence of protein stability on temperature was observed (38). A parallel has been proposed between the structure of water and the thermal denaturation of proteins (14). In fact, among other similarities, the graph of liquid water density follows a bellshaped curve at atmospheric pressure with a maximum at 4°C (3). Some authors emphasized the effect of pressure on water density as a key for understanding cold denaturation of proteins at high pressure (21). The properties of water under pressure vary and are largely a function of the pressure range (7). Indeed, the effect of increasing pressure on the behavior of cold water is to systematically push the temperature of maximum density to lower and lower temperatures (5). So-called particular properties are observed for pressures below 200 MPa. However, above a 200MPa pressure, water loses its particular characteristics and behaves like a classic hydrogen-bonded liquid. The addition of solutes causes the formation of hydration shells, leading to a new organization of water molecules. This phenomenon is strongly enhanced when the pressure is increased and, accordingly, it cancels out the particular properties of pure water in the pressure range 0.1 to 200 MPa (12). The variation in water properties with pressure, temperature, and the presence of solutes reflects changes in the arrangement of water molecules. From a biological point of view, this could explain the baroprotective effects of solutes on proteins and microorganisms under denaturing conditions. The

VOL. 72, 2006

HIGH-PRESSURE E. COLI INACTIVATION AT SUBZERO TEMPERATURE

155

FIG. 4. Isothermal compressibility of water-glycerol mixtures at 25°C and 100 MPa (■), 200 MPa (Œ), 300 MPa (F), and 400 MPa (}).

mechanisms of pressure-induced microbial inactivation may involve denaturation of some critical life processes such as enzyme reactions as suggested by some authors (9, 34). Also, a parallel between water properties and microbial inactivation can be identified. For a known set of hydration conditions, a synergistic effect was observed at pressures up to a critical level (250 MPa for an aw of 0.992), whereas antagonism occurred at pressures higher than this critical level. The consequence of increasing the hydration rate at a fixed pressure was to enhance the synergism and increase the pressure threshold that marked the crossover between synergism and antagonism. Below this threshold, pressure and temperature affect microbial viability in a similar manner and, in the same way, water behaves as a singular liquid. Above this threshold, pressure and temperature have roughly opposite effects on microbial viability and, at the same time, water behaves as a classic hydrogen-bonded liquid. Involvement of water compression. From a thermodynamic point of view, the mechanical energy transferred to the cell system during pressure treatment may be characterized by the change in volume of the system. This means that the compressibility of the cell cytoplasm is the predominant factor determining the amount of energy transferred. The compressibility of water, considered the major constituent of the cell cytoplasm, is obviously important. Numerous studies have examined variations in water compressibility with pressure and temperature (5, 6, 12). Bridgman (6) observed that the compressibility of water decreased with increasing pressure and increased with decreasing temperature. Figure 4 shows experimental data on the variation in compressibility of water-glycerol mixtures over a range of pressures at 25°C. It is noteworthy that the isothermal compressibility of such mixtures decreases with increasing glycerol fraction, regardless of the pressure between 100 and 400 MPa. This result corroborates the results of Whalley and Heath (44). By replacing part of the intracellular water with glycerol, which is less compressible than water, a decrease in cell compressibility can reasonably be expected. On the other hand, an increase in water content in the cell cytoplasm (aw of approximately 1.000) should result in an increase in its compressibility and thus in the mechanical energy transferred to

the cell system. The compressibility of the cell itself is a promising field of investigation, as suggested by Hayakawa et al. (10). The difference in compressibility between the cell envelope and the cytoplasm could perturb the area/volume ratio, which seems to be involved in cell inactivation (4, 33). Conclusions. This study shows that combined high-pressure and subzero temperature treatment is a promising way to optimize high-hydrostatic-pressure processes, since such a combination made it possible to reduce the pressure magnitude and/or improve the pressure-mediated inactivation. Nevertheless, the interaction between high pressure and subzero temperature appears to be complex. Indeed, it was pointed out that, depending on pressure level and aw of the medium being treated, subzero temperature counteracted the inactivation caused by high pressure. This unexpected phenomenon leads to the necessity to take into account the process parameters to ensure efficient treatment. The structure of water versus the stability of proteins and the microbial inactivation allowed to propose the involvement of water compression in the mechanism of pressure-mediated inactivation. However, this attractive approach could not explain the baroprotective effect at subzero temperatures. Further work should be undertaken with a view to better elucidate this phenomenon. REFERENCES 1. Alpas, H., N. Kalchayanand, F. Bozoglu, and B. Ray. 2000. Interactions of high hydrostatic pressure, pressurization temperature, and pH on death and injury of pressure-resistant and pressure-sensitive strains of food-borne pathogens. Int. J. Food Microbiol. 15:33–42. 2. Alpas, H., N. Kalchayanand, F. Bozoglu, A. Sikes, C. P. Dunne, and B. Ray. 1999. Variation in resistance to hydrostatic pressure among strains of foodborne pathogens. Appl. Environ. Microbiol. 65:4248–4251. 3. Angell, C. A. 1982. Supercooled water, p. 1–81. In F. Franks (ed.), Water: a comprehensive treatise, vol. 7. Plenum Press, Inc., New York, N.Y. 4. Beney, L., J.-M. Perrier-Cornet, M. Hayert, and and P. Gervais. 1997. Shape modification of phospholipid vesicles induced by high pressure: influence of bilayer compressibility. Biophys. J. 72:1258–1263. 5. Bridgman, P. W. 1912. Water in the liquid and five solid forms under pressure. Proc. Am. Acad. Arts. Sci. XLVII:439–558. 6. Bridgman, P. W. 1914. Thermodynamic properties of liquid water to 80° and 12000 KGM. In Collected experimental papers, vol. 1. Harvard University Press, Cambridge, Mass. 7. Cavaille, D., and D. Combes. 1996. Effect of high hydrostatic pressure and additives on the dynamics of water: a spectroscopy study. J. Raman. Spectrosc. 27:853–857. 8. Cheftel, J.-C., M. Thiebaud, and E. Dumay. 2002. Pressure-assisted freezing

156

9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19.

20.

21. 22. 23. 24. 25.

26.

27.

MOUSSA ET AL. and thawing of foods: a review of recent studies. High Pressure Res. 22:601– 611. Hashizume, C., K. Kimura, and R. Hayashi. 1995. Kinetic analysis of yeast inactivation by high pressure treatment at low temperatures. Biosci. Biotech. Biochem. 59:1455–1458. Hayakawa, I., O. Mitsuhiko, and J. Kajihara. 1992. Study of pressure bearability of bakers yeast under buffer conditions, p. 47–56. In I. Hayakawa (ed.), Food processing by ultrahigh pressure twin screw extrusion. Technomic Publishing Company, Lancaster, Pa. Hayakawa, K., Y. Ueno, S. Kawamura, T. Kato, and R. Hayashi. 1998. Microorganism inactivation using high-pressure generation in sealed vessels under sub-zero temperature. Appl. Microbiol. Biotechnol. 50:415–418. Kanno, K., and A. Angell. 1979. Water: anomalous compressibilities to 1.9 kbar and correlation with supercooling limits. J. Chem. Phys. 70:4008–4016. Kinsho, T., H. Ueno, R. Hayashi, C. Hashizume, and K. Kimura. 2002. Sub-zero temperature inactivation of carboxypeptidase Y under high hydrostatic pressure. Eur. J. Biochem. 269:4666–4674. Klotz, I. M. 1999. Parallel change with temperature of water structure and protein behavior. J. Phys. Chem. 103:5910–5916. Knorr, D. 1993. Effects of high-hydrostatic-pressure processes on food safety and quality. Food Technol. 47:156–161. LeBail, A., D. Chevalier, D. M. Mussa, and M. Ghoul. 2002. High pressure freezing and thawing of foods: a review. Int. J. Refrig. 25:504–513. Li, B., and D.-W. Sun. 2002. Novel methods for rapid freezing and thawing of foods: a review. J. Food Eng. 54:175–182. Linton, M., J. M. J. McClements, and M. F. Patterson. 1999. Inactivation of Escherichia coli O157:H7 in orange juice using a combination of high pressure and mild heat. J. Food Prot. 62:277–279. Ludwig, H., C. Bieler, K. Hallbauer, and W. Scigalla. 1992. Inactivation of microorganisms by hydrostatic pressure, p. 25–32. In C. Balny, R. Hayashi, K. Heremans, and P. Masson (ed.), High pressure and biotechnology. Colloque INSERM. John Libbey and Co., Ltd., London, England. Luscher, C., A. Balasa, A. Fro ¨hlig, E. Ananta, and D. Knorr. 2004. Effect of high-pressure-induced ice I-to-ice III phase transitions on inactivation of Listeria innocua in frozen suspension. Appl. Environ. Microbiol. 70:4021– 4029. Marque´s, M. I., J. M. Borreguero, H. E. Stanley, and N. V. Dokholyan. 2003. Possible mechanism for cold denaturation of proteins at high pressure. Phys. Rev. Lett. 91:138103. Matser, A. M., B. Krebbers, R. W. van den Berg, and P. Bartels. 2004. Advantages of high pressure sterilisation on quality of food products. Trends Food Sci. Tech. 15:79–85. Mille, Y., L. Beney, and P. Gervais. 2002. Viability of Escherichia coli after combined osmotic and thermal treatment: a plasma membrane implication. Biochim. Biophys. Acta 1567:41–48. Mitchell. P., and J. Moyle. 1955. Osmotic function and structure in bacteria. Symp. Soc. Gen. Microbiol. 6:150–180. Moerman, F., B. Mertens, L. Demey, and A. Huyghebaert, A. 2001. Reduction of Bacillus subtilis, Bacillus stearothermophilus and Streptococcus faecalis in meat batters by temperature–high hydrostatic pressure pasteurization. Meat Sci. 59:115–125. Molina-Ho ¨ppner, A., W. Doster, R. F. Vogel, and M. G. Ganzle. 2003. Protective effect of sucrose and sodium chloride for Lactococcus lactis during sublethal and lethal high-pressure treatments. Appl. Environ. Microbiol. 70:2013–2020. Norrish, R. S. 1966. An equation for the activity coefficients and equilibrium relative humidities of water and confectionary syrups. J. Food Technol. 1:25–39.

APPL. ENVIRON. MICROBIOL. 28. Oliveira, A. C., L. P. Gaspar, A. T. Da Poian, and J. L. Silva. 1994. Arc repressor will not denature under pressure in the absence of water. J. Mol. Biol. 240:184–187. 29. Oxen, P., and D. Knorr. 1993. Baroprotective effects of high solute concentrations against inactivation of Rhodotorula rubra. Lebensm.-Wiss. u-Technol. 26:220–223. 30. Paga ´n, R., and B. Mackey. 2000. Relationship between membrane damage and cell death in pressure-treated Escherichia coli cells: differences between exponential- and stationary-phase cells and variation among strains. Appl. Environ. Microbiol. 66:2829–2834. 31. Palou, E., A. Lo ´pez-Malo, G. V. Barbosa-Ca ´novas, J. Welti-Chanes, P. M. Davidson, and B. G. Swanson. 1998. High hydrostatic pressure come-up time and yeast viability. J. Food Prot. 61:1657–1660. 32. Palou, E., A. Lo ´pez-Malo, G. V. Barbosa-Ca ´novas, J. Welti-Chanes, and B. G. Swanson. 1997. Effect of water activity on high hydrostatic pressure inhibition of Zygosaccharomyces bailii. Lett. Appl. Microbiol. 24:417–420. 33. Perrier-Cornet, J. M., M. Hayert, and P. Gervais. 1999. Yeast cell mortality related to a high-pressure shift: occurrence of cell membrane permeabilization. J. Appl. Microbiol. 87:1–7. 34. Perrier-Cornet, J. M., S. Tapin, G. Serenella, and P. Gervais. 2004. Highpressure inactivation of Saccharomyces cerevisiae and Lactobacillus plantarum at subzero temperatures. J. Biotechnol. 115:405–412. 35. Picart, L., E. Dumay, J.-P. Guiraud, and J.-C. Cheftel. 2004. Combined high pressure–sub-zero temperature processing of smoked salmon mince: phase transition phenomena and inactivation of Listeria innocua. J. Food Eng. 68:43–56. 36. Polydera, A. C., E. Galanou, N. G. Stoforos, and P. S. Taoukis. 2004. Inactivation kinetics of pectin methylesterase of Greek navel orange juice as a function of high hydrostatic pressure and temperature process conditions. J. Food Eng. 62:291–298. 37. Satomi, M., T. Yamaguchi, M. Okuzumi, and T. Fujii. 1995. Effect of conditions on the barotolerance of Escherichia coli. J. Food Hyg. Soc. Jpn. 36:29–34. 38. Smeller, L. 2002. Pressure-temperature phase diagrams of biomolecules. Biochim. Biophys. Acta 1595:11–29. 39. Sonoike, K., T. Setoyama, Y. Kuma, and S. Kobayashi. 1992. Effect of pressure and temperature on the death rates of Lactobacillus casei and Escherichia coli, p. 297–301. In C. Balny, R. Hayashi, K. Heremans, and P. Masson (ed.), High pressure and biotechnology. Colloque INSERM. John Libbey and Co., Ltd., London, England. 40. Suppes, G. J., S. Egan, A. J. Casillan, K. W. Chan, and B. Seckar. 2003. Impact of high pressure freezing on DH␣ Escherichia coli and red blood cells. Cryobiology 47:93–101. 41. Takahashi, K. 1992. Sterilization of microorganisms by hydrostatic pressure at low temperature, p. 303–307. In C. Balny, R. Hayashi, K. Heremans, and P. Masson (ed.), High pressure and biotechnology. Colloque INSERM. John Libbey and Co., Ltd., London, England. 42. Torres, J. A., and G. Velazquez. 2005. Commercial opportunities and research challenges in the high pressure processing of foods. J. Food Eng. 67:95–112. 43. Van Opstal, I., S. C. M. Vanmuysen, and W. Michiels. 2003. High sucrose concentration protects Escherichia coli against high pressure inactivation but not against high pressure sensitization to the lactoperoxidase system. Int. J. Food Microbiol. 88:1–9. 44. Whalley, E., and J. B. R. Heath. 1976. Isothermal compressibility of waterglycerol mixture at 60.0°C measured relative to water. Can. J. Chem. 54: 2249–2251.