Response to high-pressure, low-temperature treatment in vegetables: determination of survival rates of microbial populations using flow cytometry and detection of peroxidase activity using confocal microscopy


Dr Guadalupe Préstamo, Instituto del Frío, Ciudad Universitaria, 28040 Madrid, Spain.


Application of high hydrostatic pressure (200, 300, 350 and 400 MPa) at 5 °C for 30 min to different micro-organisms, including Gram-positive and Gram-negative bacteria, moulds and yeasts, proved to be more effective in inactivating these organisms than treatments at 20 °C for 10 min and at 10 °C for 20 min. Moulds, yeasts, Gram-negative bacteria and Listeria monocytogenes were most sensitive, and their populations were completely inactivated at pressures between 300 and 350 MPa. The same conditions of pressure, temperature, and time were applied to different vegetables (lettuce, tomato, asparagus, spinach, cauliflower and onion), achieving reductions of from 2–4 log units in both viable mesophiles and moulds and yeasts at pressures of between 300 and 400 MPa. Sensory characteristics were unaltered, especially in asparagus, onion, tomato and cauliflower, though slight browning was observed in cauliflower at 350 MPa. Flow cytometry was applied to certain of the microbial populations used in the above experiment before and after the pressurization treatment. The results were indicative of differing percentage survival rates depending on micro-organism type, with higher survival rates for Gram-positive bacteria, except L. monocytogenes, than in the other test micro-organisms. Growth of survivors was undetectable using the plate count method, suggesting that micro-organisms suffering from pressure stress were metabolically inactive though alive. The pressurization treatments did not inactivate the peroxidase responsible for browning in vegetables. Confocal microscopic examination of epidermal tissue from onion showed that the enzyme had been displaced to the cell interior. Use of low temperatures and moderately long pressurization times yielded improved inactivation of micro-organisms and better sensorial characteristics of the vegetables, and should lower industrial costs.


Early studies on the resistance of micro-organisms to high hydrostatic pressure with a view to preserving foodstuffs were carried out by Certes in 1884 (Cheftel 1995). Hite (1899) and Hite et al. (1914) also investigated the effect of high pressures on different micro-organisms in foods, particularly in milk and vegetables.

High hydrostatic pressure has been proposed as an effective alternative to conventional heat treatments as a means of preserving foodstuffs (Cheftel 1992, 1995; Knorr 1994). Today, the food industry is expressing growing interest in high-pressure treatments for food products, and there are many high-pressure applications, despite certain drawbacks, i.e. low efficacy in destroying bacterial spores and inactivating certain enzymes. On the other hand, pressure treatments are better than conventional thermal processing in reducing undesirable effects on the nutritional and sensory attributes of foods, and could yield commercially stable products of high quality. In addition, high-pressure treatments are less energy-intensive than conventional heat treatments and may in future be more commercially competitive (Mozhaev et al. 1994).

The resistance of micro-organisms to high pressure varies considerably. The degree of inactivation depends on a number of different factors, such as micro-organism type, level of pressure, temperature, time, and water activity and composition of the foodstuff (Carlez 1994). Bacteria in the early log phase of growth are normally more sensitive to pressure than they are during the stationary phase (Zobell 1970). With certain exceptions, e.g. Listeria monocytogenes, Gram-positive bacteria are more high-pressure resistant than Gram-negative bacteria, moulds and yeasts. Bacterial spores are capable of withstanding pressures of 1000 MPa (Cheftel 1992, 1995).

In an earlier study, Arroyo et al. (1997) reported that application of pressures of 400 MPa at 20 °C for 10 min lowered populations of Gram-negative bacteria from 108 to 109 cfu ml−1 to 102 cfu ml−1. On the other hand, treatment with 350 MPa at 10 °C for 20 min caused the number of detectable survivors of most Gram-negative bacterial populations to fall to zero. In that same study, application of 400 MPa at 20 °C for 10 min and 400 MPa at 10 °C for 20 min yielded similar results for Gram-positive bacteria, with survival rates of 102 and 103 cfu ml−1. The number of spores of Bacillus cereus and Bacillus subtilis fell to less than one log unit after treatment at 400 MPa at 10 °C for 20 min. Moulds and yeasts were completely inactivated by treatment with 300 and 350 MPa at 10 °C for 20 min.

Populations of contaminant viable aerobic mesophiles, moulds and yeasts, contaminating such vegetables as lettuce and tomatoes, fell to one log unit after pressurization at 350 MPa at 20 °C for 10 min. Similar results were achieved using 300 MPa at 10 °C for 20 min. However, at this latter pressure, the sensory characteristics of the products began to be affected, particularly in the case of the lettuce (Arroyo & Préstamo 1996; Arroyo et al. 1997).

It therefore appears that in vegetables, large reductions in the number of micro-organisms can be achieved by prolonging the pressurization time while lowering the temperature, which should, in turn, help avoid some of the undesirable effects on the sensory properties of those foods.

The plate count method used to verify the reduction in microbial populations following pressurization suggests that many reported results may be false negatives, because some micro-organisms may not be irreversibly affected by pressurization and may be able to reactivate under certain favourable conditions. Using populations in the stationary growth phase, many of the micro-organisms will be dead, others still alive but metabolically inactivated, and colleagues both alive and metabolically active. For that reason, it would be useful to determine the proportions of the different components of the populations after pressurization (Cheftel 1995).

Flow cytometry has been widely used in microbiology to analyse the cell cycle of Escherichia coli (Steen et al. 1990). It has proved to be a fast and effective method of verifying the viability of micro-organisms present in water and soil (Porter et al. 1995; Deleo & Baveye 1996), and of determining the level of contamination of works of art prior to restoration (Arroyo et al. 1995).

The enzymes present in fruits and vegetables are partially responsible for the changes in texture, colour and flavour which take place during ripening, senescence and cold storage. Some enzymes are sensitive to pressure, e.g. chymotrypsin is inactivated at pressures of around 200 MPa (Mozhaev et al. 1994). Other enzymes, in contrast, are particularly resistant to pressure, e.g. peroxidase, pectin methyl sterase, lipoxygenase and polyphenoloxidase. To inactivate this last-mentioned enzyme, Gomes & Ledward (1996) had to apply 800 MPa for 5 min.

The object of the present study was to ascertain the efficacy of pressure treatments of 200, 300, 350 and 400 MPa at 5 °C for 30 min on different populations of saprophytic, pathogenic and phytopathogenic micro-organisms. With that objective in mind, reductions in the numbers of contaminating micro-organisms in different vegetables following pressurization treatments using those same conditions of pressure, temperature and time were examined, and the effect of the treatments on the sensory characteristics of the vegetables was determined. Flow cytometry was employed before and after pressurization to ascertain the metabolic state of the microbial populations and the extent of damage caused by the treatments. In addition, confocal microscopy was used to establish the effect of pressure on certain enzymes.

Materials and methods

Test organisms

High pressure was applied to 18 selected microbial strains that included five Gram-positive bacteria and nine Gram-negative bacteria; the remaining four were moulds and yeasts. The strains were from the Spanish Type Culture Collection and from the working collection of the Department of Microbiology of the Madrid Complutense University and were as follows: Bacillus cereus LAB 56; Bacillus subtilis CECT 39; Staphylococcus aureus LAB 327; Micrococcus luteus CECT 245; Listeria monocytogenes NCTC 11994; Escherichia coli CECT 99; Salmonella typhimurium LAB 36; Salmonella enteritidis LAB 151; Yersinia enterocolitica CECT 500; Aeromonas hydrophila CECT 839; Pseudomonas aeruginosa LAB 64; Pseudomonas fluorescens ATCC 13525; Erwinia herbicola LAB 31; Xanthomonas campestris ATCC 33913; Saccharomyces cerevisiae CECT 1383; Aspergillus niger LAB 104; Aspergillus flavus LAB 107; Penicillium spp. LAB 102.

Sample preparation

Pure cultures.

The bacterial strains were cultured on tryptone soy agar (TSA) slants at 30 °C for 24–48 h; moulds and yeasts were cultured on Sabouraud agar (SA) slants at 28 °C for 24–72 h. These cultures were then used to prepare stock suspensions in tryptone soy broth (TSB). Serial decimal dilutions were prepared from these suspensions. A loopful of decimal dilution was streaked on Petri dishes containing TSA for bacteria and SA for moulds and yeasts; the former were incubated at 30 °C for 24–48 h, the latter at the same temperature for 24–72 h, to obtain isolates. A single colony was taken from each dish and incubated in tryptone soy broth (TSB) at 30 °C for 18 h. These cultures were then used to prepare serial decimal dilutions, yielding bacterial populations of 105–109 cfu ml−1 and mould and yeast populations of 103–104 cfu ml−1. Eppendorf tubes containing 1·5 ml of each micro-organism suspension were introduced in flexible tubes filled with sterile water and pressurized. The number of micro-organisms was determined by plating 0·1 ml of dilutions on TSA or SA before and after treatment. Three replicates of all samples were performed. The pressure levels employed were 200, 300, 350 and 400 MPa at 5 °C for 30 min.

Vegetable samples.

Baby lettuce, tomatoes, spinach, asparagus, onions and cauliflowers were analysed before and after high-pressure treatment. An amount of 10 g of sample was weighed out using sterile instruments under aseptic conditions, added to 90 ml TSB and homogenized in sterile bags using a Stomacher model 400 Lab blender for 3 min. Counts of viable aerobic mesophiles were obtained by means of serial dilutions plated on TSA and incubated at 30 °C for 24–48 h; moulds and yeasts on SA were incubated at 28 °C for 24–78 h. Similarly, 10 g aliquots of sample were sealed in latex tubes for pressurization. The same microbiological analysis procedure described above was employed after pressurization. Samples were subjected to 300, 350 or 400 MPa at 5 °C for 30 min. Samples of asparagus and cauliflower were cooled and stored for 1 week at 20 °C after the pressurization treatment.

Sensory evaluation

Changes in colour, flavour and texture were evaluated subjectively (Dethmers 1981). The two-step rating scale employed consisted of two categories, i.e. the same as, or different from, untreated product. Evaluations were carried out before and after pressurization treatment.

High hydrostatic pressure treatment

A high-pressure apparatus (ACB GEC Alsthon, Nantes, France) and thermostatic circuit were used to generate high pressure levels with two hydro-pneumatic pumps. Samples were immersed in a fluid of low compressibility (water) in a steel container (100 mm × 300 mm, and 2·35 l volume). Temperature was held constant using a water-bath. Temperature and pressure were recorded in a Lab Tech notebook program (Laboratory Technologies Corporation, Wilmington, MA, USA). The micro-organism cultures were introduced into sealed Eppendorf tubes, and these tubes were introduced into other flexible tubes (latex tubes filled with distilled water) to prevent the Eppendorf tubes breaking, isolate them from the fluid in the container and ensure transmission of the pressure throughout the entire sample. The pressure machine reached 400 MPa in 2–3 min and the depressurization process took 5–8 min. Depressurization is necessary to avoid the samples becoming frozen.

Flow cytometry

Flow cytometry (FCM) was performed using a FACScan (Becton-Dickinson) flow cytometer equipped with a 25 mW argon laser operated at a wavelength of 488 nm, using CELLQUEST software (Becton-Dickinson) for data acquisition and analysis. A total of 10 000 events were registered per sample.

Cell lysis and viability were measured by PI (propidium iodide) as a measure of membrane integrity. Only cells that have lost selective permeability due to lysis take up the dye, which stains nucleic acids. PI-positive cells fluoresce and can be quantified using flow cytometry.

Bacterial cultures were prepared for flow cytometry by removing 1 ml of each culture and centrifuging at 3000 rev min−1 for 5 min, before and after the high-pressure treatment. The bacterial sediment was then washed in phosphate-buffered saline solution (PBS) at pH 7·4 and resuspended in 200 Fl of PBS. Micro-organisms were stained with propidium iodide (final concentration: 5 Fg l−1) (Sigma). Cells were incubated at room temperature for 30 min, washed, and resuspended in PBS before examination using FCM. Flow cytometry counts were compared with the results of the conventional plate count method using nutrient agar.

Peroxidase activity

Peroxidase (POD) activity was determined following the method described by Préstamo & Manzano (1993) using o-dianisidine as chromogenic indicator. Total reaction volume was 3 ml, containing: 2·7 ml 50 mmol l−1 sodium acetate, pH 6 buffer; 0.·1 ml 0·5% hydrogen peroxide; 0·1 ml enzyme extract; and 0·1 ml of 0·25% o-dianisidine (w/v). The reaction was measured by spectrophotometry at 460 nm. Peroxidase activity was expressed in units of POD ml−1. A unit of POD was defined as an increase of 0·001 unit of absorbance per minute.

Confocal microscopy

The fluorescent dye 2-7, dichlorofluoresceine diacetate was used to stain onion tissue to detect POD before and after the high-pressure treatment. Samples were immersed in the dye for 30 min and then washed in PBS. Fluorescence emitted by the POD stained by the dye was observed in an MRC-1000 confocal microscope (Bio-Rad), equipped with an argon/ krypton ion laser as light source. Digital analysis was performed using Comos software (Bio-Rad). Digital images were photographed using a Lasergraphics (Irvine, CA, USA) picture printer.

Statistical analysis

The results were analysed using the Statgraphics program (version 5.0, Duncan test, one way, P= 0.05).


Effect of pressure on the pure suspensions of micro-organisms

Gram-positive bacteria.

Figure 1 plots the effects of the different pressure treatments at 5 °C for 30 min and the statistical study (Duncan test) is depicted in Table 1. Application of a pressure of 200 MPa lowered the initial populations (> 108 and > 107 cfu ml−1) of B. cereus, B. subtilis, Staph. aureus and M. luteus by 2 or 3 log units and significant differences were observed in all samples except for L. monocytogenes and Staph. aureus where no significant differences were detected. Pressures of 300 and 350 MPa yielded survival rates for these four micro-organism strains of > 103 cfu ml−1. At 300 Mpa, no significant differences were found in B. cereus and M. luteus and at 350 Mpa, no significant differences were found in B. cereus, Staph. aureus and M. luteus. At 400 Mpa, the survival rate was > 102 cfu ml−1. Under the conditions employed, Staph. aureus and B. cereus were slightly more sensitive, presenting higher significant differences to the action of the high-pressure treatment than the other two strains (M. luteus and B. subtilis).

Figure 1.

Reduction of Gram-positive populations of (◆), Bacillus cereus; (▪), B. subtilis; (▴), Staphylococcus aureus; (○), Micrococcus luteus and (□), Listeria monocytogenes after application of high pressure (200, 300, 350 and 400 MPa) at 5 °C for 30 min

Table 1.  Statistical results of Gram-positive populations (Bacillus cereus, B. subtilis, Staphylococcus aureus, Micrococcus luteus and Listeria monocytogenes) after high pressure treatment (200, 300, 350 and 400 MPa) at 5°C for 30 min
 0 MPa200 MPa300 MPa350 MPa400 MPa
  1. Mean±s.d.; n = 3; P = 95%.

Bacillus cereus8·72±0·11a4·83±0·26c3·60±0·07c2·95±0·16b2·04±0·23c
B. subtilis7·61±0·20b4·10±0·08d3·83±0·10b3·74±0·08a2·74±0·09a
Staphylococcus aureus7·73±0·09b5·60±0·15a4·14±0·05a3·00±0·01b1·94±0·11c
Micrococcus luteus7·85±0.45b5·19±0·15b3·70±0·06c3·00±0·05b2·47±0·07b
Listeria monocytogenes7·02±0·16c5·53±0·19a3·00±0·02d0·00±0·00c0·00±0·00d

In the case of L. monocytogenes, the initial population was > 107 cfu ml−1 and the rate of decrease was 2 log units at 200 MPa and 4 log units at 300 MPa. Higher pressures (350 and 400 MPa) decreased the population below detection limit, and significant differences were observed in comparison with the other strains.

Gram-negative bacteria.

Figure 2(a,b) shows the results for the Gram-negative bacteria at the different pressure levels at 5 °C for 30 min, and the statistical results (Duncan test) are plotted in Tables 2a and 2b. Figure 2(a) plots the effects of pressure on the populations of E. coli, Salm. typhimurium, Salm. enteritidis, Y. enterocolitica and A. hydrophila (> 108 and > 107 cfu ml−1). A pressure of 200 MPa reduced the populations of E. coli and Salm. typhimurium by 3 log units, and the populations of Salm. enteritidis and Y. enterocolitica by 2 log units. For A. hydrophila, the reduction was 4 log units and significant differences were observed in comparison with the other strains. At 300 Mpa, survival rates were > 102 cfu ml−1, except in E. coli in which the survival rate was only > 101 cfu ml−1, presenting significant differences in E. coli and Salm. enteritidis in comparison with the other strains, Salm. typhimurium, Y. enterocolitica and A. hydrophila, where no significant differences were detected. Application of higher pressures (350 and 400 MPa) eliminated the populations of all these strains below the detection limit.

Figure 2.

Figure 2.

Reduction of Gram-negative populations of (a) (◆), Escherichia coli, (▪), Salmonella typhimurium; (▴)Salm. enteritidis; (○), Yersinia enterocolitica; (□), Aeromonas hydrophila and (b) (◆), Pseudomonas aeruginosa, (▪), Ps. fluorescens; (▴), Erwinia herbicola; (○), Xanthomonas campestris after application of high pressure (200, 300, 350 and 400 MPa) at 5 °C for 30 min

Figure 2.

Figure 2.

Reduction of Gram-negative populations of (a) (◆), Escherichia coli, (▪), Salmonella typhimurium; (▴)Salm. enteritidis; (○), Yersinia enterocolitica; (□), Aeromonas hydrophila and (b) (◆), Pseudomonas aeruginosa, (▪), Ps. fluorescens; (▴), Erwinia herbicola; (○), Xanthomonas campestris after application of high pressure (200, 300, 350 and 400 MPa) at 5 °C for 30 min

Table 2a.  Statistical results of Gram-negative populations (a) Escherichia coli, Salmonella typhimurium, Salm. enteritidis, Yersinia enterocolitica and Aeromonas hydrophila) and (b) Pseudomonas aeruginosa, Ps. fluorescens, Erwinia herbicola and Xanthomonas campestris) after high pressure treatment (200, 300, 350 and 400 MPA) at 5 °C for 30 min
 0 MPa200 MPa300 MPa350 MPa400 MPa
  1. Mean±s.d.; n = 3; P = 95%.

  2. Table 2b

Escherichia coli8·57±0·04a5·21±0·13a1·04±0·08c00
Salmonella typhimurium8·07±0·29b5·27±0·04a2·00±0·08b00
Salm. enteritidis7·45±0·29ab5·21±0·23a2·47±0·10a00
Yersinia enterocolitica7·10±0.11c5·41±0·10a2·00±0·24b00
Aeromonas hydrophila8·49±0.37a4·10±0·11b2·20±0·12b00

Figure 2(b) depicts the behaviour of Ps. aeruginosa, Ps. fluorescens, Erw. herbicola and X. campestris and in Table 2b, the statistical results (Duncan test) are given. The initial population of Ps. aeruginosa (> 107 cfu ml−1) dropped 3 log units at 200 MPa and 5 log units at 300 MPa. At 200 MPa, no significant differences were observed in Ps. aeruginosa and X. campestris. After pressurization at 350 and 400 MPa, the population of Ps. aeruginosa was no longer detectable. Pseudomonas fluorescens and Erw. herbicola followed similar trends to Ps. aeruginosa. The initial populations (> 106 cfu ml−1) were reduced to 103 and 102 cfu ml−1, respectively, at 300 MPa, and dropped to undetectable levels at 350 and 400 MPa. Xanthomonas campestris was the strain most susceptible to pressure and the initial population (> 108 cfu ml−1) was eliminated to below the detection limit at 300 MPa.

Moulds and yeasts.

Figure 3 illustrates the results obtained for the moulds and yeasts. The initial population of S. cerevisiae was lowered 3 log units at a pressure of 300 MPa and decreased below detection limit at 400 MPa. The statistical results (Duncan test) are shown in Table 3. At 200 MPa, no significant differences were detected in A. niger and A. flavus. At 300 MPa, A. niger and Penicillium showed no significant differences. Survival rates of the initial populations of A. niger, A. flavus and Penicillium spp. (> 104 cfu ml−1) were > 101 and > 102 cfu ml−1, respectively, at a pressure of 300 MPa. The populations were eliminated below the detection limit at 350 MPa, except for S. cerevisiae which presented significant differences in comparison with the other strains.

Figure 3.

Reduction of mould and yeast populations (◆), Saccharomyces cerevisiae (▪), Aspergillus niger (▴), A. flavus and (○) Penicillium spp. after application of high pressure (200, 300, 350 and 400 MPa) at 5 °C for 30 min

Table 3.  Statistical results of mould and yeast populations (Saccharomyces cerevisiae, Aspergillus niger, A. flavus and Penicillium spp. after high pressure treatment (200, 300, 350 and 400 MPa) at 5 °C for 30 min
 0 MPa200 MPa300 MPa350 MPa400 MPa
  1. Mean±s.d.; n = 3; P = 95%.

Saccharomyces cerevisiae5·43±0·07a4·06±0·06a2·69±0·12a0·95±0·08a0
Aspergillus niger4·04±0·07c2·81±0·04c1·34±0·15c0·00±0·00b0
A. flavus4·66±0·04b3·00±0·16c2·30±0·08b0·00±0·00b 

Effects of pressure on the micro-organisms in vegetables

Viable aerobic mesophiles.

The initial populations of viable aerobic mesophiles (Fig. 4) in lettuce, spinach, cauliflower and asparagus were > 105 cfu g−1, lower in tomato and onion (> 103). Pressurization of the former group of vegetables at 300 MPa yielded a survival rate of > 103 cfu g−1, except in the case of cauliflower, in which the initial population decreased below detection limit; the statistical results are shown in Table 4. In tomato and onion, the population decreased by one log unit and no significant differences were observed between them. At 350 MPa, survival rates were > 103 cfu g−1 in all the vegetables in the first group, except cauliflower, and lower in the others, > 102 cfu g−1 in tomato and > 101 cfu g−1 in onion; significant differences were observed in cauliflower, tomato and onion in comparison with asparagus, lettuce and spinach. In asparagus, lettuce and spinach no significant differences were observed. Pressurization at 400 MPa brought about complete elimination (below the detection limit) in onion and nearly complete elimination (below the detection limit) in tomato (> 10 cfu g−1), and no significant differences were detected in spinach and lettuce, cauliflower and onion.

Figure 4.

Decrease of mesophilic aerobic micro-organisms contaminating different vegetable products (◆) spinach, (▪) lettuce, (▴) cauliflower, (○) asparagus, (□) tomato and (•) onion after application of high pressure (300, 350, 400 MPa) at 5 °C for 30 min

Table 4.  Statistical results of mesophilic aerobic micro-organisms contaminating different vegetables (spinach, lettuce, cauliflower, asparagus, tomato and onion) after high pressure treatment (200, 300, 350 and 400 MPa) at 5 °C for 30 min
 0 MPa300 MPa350 MPa400 MPa
  1. Mean±s.d.; n = 3; P = 95%.


Moulds and yeasts.

Moulds and yeasts (Fig. 5) presented initial populations of > 105 cfu g−1 in spinach, > 104 cfu g−1 in lettuce, cauliflower and asparagus, and > 103 cfu g−1 in tomato and onion. Their behaviour in response to pressurization was similar to that of the viable mesophiles; the statistical results are shown in Table 5. At 300 MPa, significant differences were detected in all samples. At 350 MPa, no significant differences were observed in asparagus and lettuce. Pressurization at 400 MPa eliminated, below the detection limit, the populations in tomato, cauliflower and onion, and significant differences were observed in comparison with lettuce, spinach and asparagus.

Figure 5.

Decrease of moulds and yeasts contaminating different vegetable products (◆) spinach, (▪) lettuce, (▴) cauliflower, (○) asparagus, (□) tomato and (•) onion after application of high pressure (300, 350, 400 MPa) at 5 °C for 30 min

Table 5.  Statistical results of moulds and yeasts contaminating different vegetables (spinach, lettuce, cauliflower, asparagus, tomato and onion) after high pressure treatment (200, 300, 350 and 400 MPa) at 5 °C for 30 min
 0 MPa300 MPa350 MPa400 MPa
  1. Mean±s.d.; n = 3; P = 95%.


Effects of pressure on sensory characteristics

Pressurization did bring about some changes in the sensory characteristics of the vegetables. At pressures of 350 MPa and higher, the appearance of most of the vegetables was similar to the appearance after cooking. Texture was firm in lettuce and spinach at 300 MPa, and in tomato, cauliflower, asparagus and onion at 350 MPa. At 400 MPa, texture was firm in cauliflower, asparagus and onion. In tomato, the skin loosened, but texture remained rather firm.

The colour of lettuce and spinach was altered at 300 MPa and browning began. At 350 MPa, cauliflower underwent slight browning of the outer portions. Colour was unchanged in asparagus, tomato and onion.

Flavour was good in all the vegetables, even after pressurization at 400 MPa.

Application of flow cytometry to the pure microbial populations

Some bacterial cultures were selected and tested for survival rates after pressurization at 200, 300 and 400 MPa using flow cytometry. The micro-organisms selected were: E. coli, Salm. typhimurium, X. campestris, L. monocytogenes, B. cereus and Staph. aureus. The results are shown in Tables 6 and 7. Table 6 shows that live cells made up 72·04% of the initial population of E. coli and of those, 69·35% were single cells while 2·68% were cell aggregates. Dead cells accounted for 27·97% and of those, 26% were single cells while 1·97% were cell aggregates. Table 6 shows that after pressurization at 400 MPa, the percentage of dead cells was 72·44% (of that percentage, 70·63% were single cells and 1·81% were cell aggregates). The survival rate was 27·56% (27·37% single cells and 0·19% cell aggregates). The initial E. coli population dropped by 44·48% after pressurization at 400 MPa. Pressure did appear to affect cell aggregation.

Table 6.  Determination of survival (%) of Gram-negative micro-organisms using flow cytometry
 %Control200 MPa300 MPa400 MPa
  1. LC, live cells; DC dead cells; *cell aggregates

Salmonella typhimuriumLC75·7671·7235·8330·26
   0·92* 0·58* 0·13* 0·28*
   0·64* 0·82* 0·93* 1·17*
Escherichia coliLC69·3551·1628·6427·37
   2·68* 0·06* 1·69* 0·19*
   1·97* 0·32*13·92* 1·81*
Xanthomonas campestrisLC57·7846·2631·1429·14
   1·18* 0·02* 1·71* 0·01*
   1·69* 0·26* 8·16* 8·16*
Table 7.  Determination of survival (%) of Gram-positive micro-organisms using flow cytometry
 %Control200 MPa300 MPa400 MPa
  1. LC, live cells; DC dead cells; *cell aggregates

Listeria monocytogenesLC71·8842·0626·7226·72
   0·00* 0·01* 0·01* 0·00*
   0·08* 0·18* 0·21* 0·11*
Staphylococcus aureusLC53·5149·0426·0623·72
   6·69* 1·09* 0·03* 0·01*
  19·10* 6·96* 2·60* 2·04*
Bacillus cereusLC 9·6011·3026·0623·72
   0·49* 0·13* 0·03* 0·01*
  17·60*28·22* 2·60* 2·04*

Salmonella typhimurium (Table 6) exhibited similar behaviour. The initial population consisted of 76·68% live cells of which 0·92% were cell aggregates. After pressurization at 400 MPa, the proportion of live cells dropped to 30·54% (0·28% cell aggregates), representing a decrease of 46·14% in the initial population. As in the preceding case, cell aggregation was affected by pressurization.

Table 6 presents the results obtained for X. campestris. Live cells made up 59·76% of the initial population (57·58% being single cells, 1·18% being cell aggregates). After pressurization at 400 MPa, the percentage of live cells was 29·15%, only 0·01% of which were cell aggregates. The reduction was 30·61%.

Table 7 depicts the results for the Gram-positive bacteria selected for the experiment. After pressurization at 400 MPa, the initial population of L. monocytogenes, comprising 71·88% live cells, dropped to a final percentage of 26·72% of live cells, representing a decrease of 45·16%.

The results for B. cereus were unexpected. The initial population of live cells (10·09%) was considerably lower at the start of the experiment than the population of dead cells (89·90%); 72·30% of the dead cells were single cells, 17·60% cell aggregates. The percentage reduction after pressurization was similar at all levels of pressure and similar to the results for the control group.

Table 7 plots the results for Staph. aureus. The initial percentage of 60·20% of live cells (6·69% cell aggregates) fell to 23·73% (0·01% cell aggregates), representing a decrease of 37·47% in the initial population.

The findings showed that on the whole, the largest reduction was achieved at 300 MPa, and those results were verified using the conventional plate count method.

Peroxidase activity as measured by confocal microscopy

Activity levels for peroxidase, an enzyme highly resistant to both heat and pressure, were measured in tomato, lettuce and onion before and after the pressurization treatment. At 300–350 MPa at 5 °C for 30 min, activity increased; 400 MPa was insufficient to inactivate this enzyme.

Confocal microscopy demonstrated high fluorescence emissions by the stained peroxidase in the control sample of onion, with most activity located in the cell wall (Fig. 6). After pressurization at 400 MPa, fluorescence emissions by the stained peroxidase were spread throughout the cytoplasm (Fig. 7), with no emissions from the cell wall except for some granular accumulations.

Figure 6.

Location of peroxidase in the cell wall of onion before treatment at high pressure

Figure 7.

Location of peroxidase after pressure treatment at 400 MPa. The peroxidase was spread throughout the cytoplasm


Interest in the application of high-pressure treatments to biological material and food constituents and food systems is growing, one important reason being the possibility of inactivating micro-organisms at low and moderate temperatures. Though little work has been carried out on the causes of microbial inactivation (Hoover et al. 1989), certain morphological changes have been observed at higher pressures, including compression of gas vacuoles, cell elongation, separation of the cell membrane from the cell wall, contraction of the cell wall and resultant pore formation, alterations of the cytoskeleton, nucleus, and cell organelles, coagulation of plasma proteins, and release of cell constituents to the exterior (Shimada & Shimahara 1991).

The degree of inactivation depends on different factors: micro-organism type, amount of pressure, treatment temperature and time, and composition of the dispersion medium. As reported by Arroyo & Préstamo (1996) and Arroyo et al. (1997), treatment of different strains of micro-organisms and different vegetables (lettuce and tomato) with pressures of 100, 200, 300, 350 and 400 MPa at a temperature of 20 °C for 10 min yielded varying results.

Spores presented the highest resistance, followed by Gram-positive bacteria; moulds and yeasts and Gram-negative bacteria were the most sensitive. However, those conditions of pressure, temperature and time were insufficient to reduce the microbial populations in vegetables, and also, at 350 MPa or higher, they gave rise to undesirable sensory effects. Treatment using the same pressures at 10 °C for 20 min decreased the resistance of Gram-positive bacteria and greatly affected Gram-negative bacteria, moulds and yeasts, for which no survivors were detectable at pressures of between 300 and 400 MPa, and at the same time, improved the organoleptic characteristics of the vegetables tested. In the present study, application of 200,300, 350 and 400 MPa at 5 °C for 30 min achieved higher reduction rates than in the above studies (Figs 1, 2, 3, 4 and 5) and in addition, treatment at 5 °C proved to be more effective for inactivating vegetative cells. These findings corroborated the results of others who have reported the greatest alterations at temperatures from 30 °C down to around 0 °C, with peak microbial resistance at temperatures of 15–30 °C and significantly lower levels of resistance at higher or lower temperatures. According to Ludwig et al. (1992), the inactivation rate did not depend on the solution employed, and similar results were achieved using both nutrient broth and physiological saline solution. Citrobacter freundii, Ps. fluorescens, Listeria inocua and E. coli inoculated into minced meat or dispersed in buffer solution presented lower resistance to pressurization at 3 and 5 °C than at 20 °C. On the other hand, the endogenous flora in food was more resistant to pressure than the organisms inoculated into the food or TSB (Carlez et al. 1992, 1993; Carlez 1994; Arroyo & Préstamo 1996; Arroyo et al. 1997). The lower resistance of micro-organisms at lower temperatures may be attributable to alterations in the composition of the cell membrane, with resultant phase transitions from liquid into gel as reported by Eze (1990) for the membrane of E. coli. Changes in membrane fluidity may be caused by weakening of hydrophobic forces and crystallization of phospholipids (Ludwig et al. 1992; Carlez et al. 1992, 1993; Smelt & Rijke 1992; Tonello et al. 1992). The achievement of similar levels of inactivation using low-temperature treatments is of considerable interest to industry as it would lower costs. In addition, lower temperatures contribute to fewer alterations in texture in vegetables.

Another factor to be considered is the protection against micro-organisms conferred by certain ingredients of food. At pressures above 200 MPa, concentrations of 40% sucrose and 10–13% NaCl exert a pronounced inhibitory effect (Hoover et al. 1989; Oxen & Knorr 1993; Patterson et al. 1996). On the other hand, the presence of sugars and salts in vegetables exerts a baroprotective effect on contaminating micro-organisms in foods (Horie et al. 1991; Takahashi et al. 1993). Ogawa et al. (1990) and Oxen & Knorr (1993) also reported increased resistance to pressure by fungi in the presence of sugars (sucrose, fructose and glucose), which could result in baroresistant or barosensitive foodstuffs. Overall, findings have been extremely variable. According to Styles et al. (1991), milk also exerts a baroprotective effect, with a treatment of 340 MPa at 23 °C for 80 min being necessary to inactivate L. monocytogenes (6 log cycles) in milk. For Y. enterocolitica in buffer solution, an inactivation rate of 5 log cycles was achieved after 15 min at 275 MPa. For that same treatment time, 350 MPa was needed for Salm. typhimurium, 375 MPa for L. monocytogenes, 450 MPa for Salm. enteritidis and 700 MPa for E. coli O157:H7 and Staph. aureus. To obtain similar results in TSB in our study at 5 °C, 30 min at pressures between 300 and 350 MPa were required. The baroprotective effect of water activity also needs to be considered in the case of certain micro-organisms capable of surviving at low aw values. The baroprotective effect depends not only on the level of pressure applied and treatment time, but also on interactions with other intrinsic and extrinsic variables that condition responses by micro-organisms (Knorr 1993, Knorr (1994; Rovere & Maggi 1995; Palou et al. 1997). Listeria monocytogenes was resistant to pressure only at high aw values. Psychrotrophic strains may have adaptations in the cell membrane and fluidity levels that may increase their tolerance to pressure (Lanciotti et al. 1996). The response of micro-organisms to pressurization has been observed to consist of two phases, both by other workers and in the present study. In a first phase, a decline of 4–6 log units (Figs 1, 2, 3, 4 and 5) usually occurs at 300 MPa, but in a second phase, cells prove to be much more resistant, particularly Gram-positive bacteria (Arroyo & Préstamo 1996; Arroyo et al. 1997). Other researchers have reported that same effect in E. coli, Salmonella spp., V. parahaemolyticus, L. monocytogenes, S. cerevisiae, Rhodotorula rubra, and other moulds and yeasts (Butz & Ludwig 1986; Hoover et al. 1989; Metrick et al. 1989; Ogawa et al. 1990; Honma & Haga 1991; Shimada & Shimahara 1991; Styles et al. 1991; Ludwig et al. 1992; Oxen & Knorr 1993; Yukisaki et al. 1994). Metrick et al. (1989) isolated the second, more pressure-resistant phase, cultured it, subjected it to a second pressurization treatment and again observed the same two-phase response. This response has been attributed to genotypic variations in the resistance of individual cells in a population. Furthermore, high pressures result in synthesis of a protein, porin, suggesting a genetic basis for the response to pressure (Mozhaev et al. 1994).

At 400 MPa, cells conserve their membrane potential and integrity but lose their ability to grow on nutrient agar. Using a differential numeration method in TSA with and without NaCl, Patterson et al. (1996) found a population of live cells that showed the effects of stress following exposure to sublethal pressure; the proportion of cells exhibiting signs of stress fell when they were exposed to pressurization in milk. Survival rates differed when measured immediately after application of the pressurization treatment, or after a recovery period in foodstuffs or on enriched nutrient media.

The phenomenon of microbial stress also contains an adverse component, in that the action of pressurization may be over-estimated by an inability to evaluate the numbers of micro-organisms suffering from stress yet still capable of growth during food storage. Carlez (1994) studied the development of the natural flora in minced meat exposed to 400–450 MPa at 20 °C for 20 min and found that after pressurization, Pseudomonas grew at a temperature of 3 °C during a storage period of 10–15 d. It was also found that residual contamination did not decrease in certain vegetables stored at 4 °C for 10 d after pressurization at 400 MPa (Arroyo & Préstamo 1996; Arroyo et al. 1997). Indeed, the authors decided to use flow cytometry to evaluate the percentage of cells suffering from stress yet still alive after observing growth by certain micro-organisms theoretically killed by pressurization following revitalization after a few hours in TSB. The results obtained (Tables 6 and 7) indicate that cytometry was in fact capable of detecting a substantial proportion of cells that were unable to grow in culture media subsequent to pressure stress. The percentage of dead cells was higher in Gram-positive bacteria than in Gram-negative bacteria. The percentage values for B. cereus differed from those for the other strains because many vegetative cells had sporulated in the stationary phase, and spores are much more resistant to pressurization. Bacterial spores are not inactivated at pressures of 1000 MPa (Cheftel 1995). Flow cytometry has been shown to be a fast, excellent means of obtaining counts (Arroyo et al. 1995) and in the present study, it also proved to be effective in determining the percentages of dead, living and metabolically-inactive cells, of supreme importance in establishing the efficacy of pressurization.

Finally, pressurization may also alter sensory characteristics, for instance, colour. Pigments such as carotenoids, chlorophylls and anthocyanins appear to resist pressurization. Colour, texture and firmness are more sensitive in leafy vegetables (lettuce and spinach) and may undergo adverse or beneficial changes. Certain vegetables that contain gas vacuoles are severely and irreversibly compressed as a result of the high compressibility of the gases. In consequence, they undergo morphological alterations, tissue disruptions and cell exudation, in many cases related to enzymatic activity. Enzyme inactivation depends on the pressure employed. Some enzymes are relatively easy to destroy, namely, certain hydrogenases in E. coli (100 MPa), yeast carboxypeptidases (400 MPa), and Na/K dependent ATP-ase located in the lipid layer, involved in active transport across the cell membrane (Chong et al. 1985). Further investigation into enzymatic inactivation and protection in the presence of various substances, in pressurized tissues or solutions, is still needed. In certain denaturation-inactivation phenomena, the two effects may be antagonistic; for instance, α-amylase destroyed by heat can be reactivated by pressure (Mori et al. 1991; Hayakawa et al. 1992). Enzymatic browning reactions in some vegetables or vegetable extracts appear to be activated by pressurization treatments (Asaka & Hayashi 1991). Peroxidase is relatively heat-resistant and has conventionally been used in the blanching test for vegetables, and also as a measure of browning (Burnette 1977; Lee & Hammes 1979; Préstamo & Manzano 1993). Certain of the vegetables employed in the experiment (cauliflower, spinach and lettuce) displayed slight browning at pressures of 350 and 400 MPa. To obtain information on peroxidase behaviour, the activity of this enzyme was measured before and after treatment. After treatment, there was an increase in enzymatic activity as a result of enhanced sample extractability and accelerated reactions following pressurization (Mozhaev et al. 1994). Pressurization at 400 MPa was not sufficient to inactivate peroxidase. Confocal microscopy showed that fluorescence emissions by stained peroxidase in onion before treatment were located mainly in the cell wall. After treatment at 400 MPa, fluorescence emissions by stained peroxidase were located in the cytoplasm with no emissions from the cell wall except for a few granulations. A possible explanation for this effect is decreased permeability of the cell wall after treatment, giving the tissue a soaked appearance. Decompartmentalization of the peroxidase may have contributed to the browning recorded for the samples.

The best results for destruction of the pure microbial population samples and the contaminating populations in vegetables were obtained using a treatment temperature of 5 °C and a treatment time of 30 min. According to the results of flow cytometry, the destruction rate was lower than recorded previously because many cells that were alive after the pressurization treatment did not survive in TSA. Good results can be achieved in vegetables if pressurization is followed by cold storage or is combined with other technologies such as modified atmospheres. In certain cases, inactivation of peroxidase could avoid unwanted browning.