Morphological and Physiological Changes Induced by High Hydrostatic Pressure in Exponential- and Stationary-Phase Cells of Escherichia coli: Relationship with Cell Death

The relationship between a loss of viability and several morphologicaland physiological changes was examined with Escherichia colistrain J1 subjected to high-pressure treatment. The pressureresistance of stationary-phase cells was much higher than thatof exponential-phase cells, but in both types of cell, aggregationof cytoplasmic proteins and condensation of the nucleoid occurredafter treatment at 200 MPa for 8 min. Although gross changeswere detected in these cellular structures, they were not relatedto cell death, at least for stationary-phase cells. In additionto these events, exponential-phase cells showed changes in theircell envelopes that were not seen for stationary-phase cells,namely physical perturbations of the cell envelope structure,a loss of osmotic responsiveness, and a loss of protein andRNA to the extracellular medium. Based on these observations,we propose that exponential-phase cells are inactivated underhigh pressure by irreversible damage to the cell membrane. Incontrast, stationary-phase cells have a cytoplasmic membranethat is robust enough to withstand pressurization up to veryintense treatments. The retention of an intact membrane appearsto allow the stationary-phase cell to repair gross changes inother cellular structures and to remain viable at pressuresthat are lethal to exponential-phase cells.

INTRODUCTION

Top

Introduction

With increasing consumer demand for high-quality products withfresh characteristics, there has been an increasing interestin new mild methods of food preservation. One of the most promisingis high hydrostatic pressure, and this is now being increasinglyapplied commercially for the processing of foodstuffs such assauces, fruit juices, oysters, and meat products (27). The mainadvantage of this technology is that it can better preservethe taste, color, and original texture of the product (6, 14,27). Also under study are combinations of high pressure withother methods, such as heat, low pH, or antimicrobial peptides(10, 21, 25). However, to exploit these new approaches fully,we need to better understand the effects of high pressure onmicrobes.

The inactivation of bacteria by high pressure has been investigatedby numerous workers (14, 18, 32). It is generally acknowledgedthat bacterial spores are more resistant to high pressure thanare vegetative cells and that gram-positive species are moreresistant than gram-negative ones. However, there are many exceptionsto this general rule, and great variation exists among strains(1, 4, 27, 29). For any given strain, resistance is affectedby its physiological state and the composition of the suspendingmedium during pressurization. Cells in the stationary phaseof growth are more resistant than those in the exponential phase(4, 24). Also, some studies on the influence of the pressurizingmedium have been done, and they showed that complex media richin ions, proteins, lipids, and carbohydrates are more protectivethan buffers (10, 12, 27, 35). However, despite much effort,the main cellular targets and events responsible for cell deathhave not yet been identified, and therefore it is difficultto understand, for instance, the wide variation in resistancewithin the same species or the influence of the compositionof the pressurizing medium. Data published in the literatureindicate that envelopes could be an important target for high-pressureinactivation (2, 16, 34). Several authors have reported damageto the cytoplasmic membrane, such as a loss of osmotic responsiveness,uptake of vital dyes (propidium iodide, ethidium bromide, oroxonol), or loss of intracellular material, after pressurization(4, 24, 33). The loss of function of some proteins, includingthe F₁-F₀ ATPase and multidrug efflux pumps, has also been described(34, 38, 39). Moreover, Ritz et al. (28) reported the loss ofproteins from both the outer and inner membranes of Salmonellaenterica serovar Typhimurium, as well as the formation of budson the cell surface, after pressure treatments. However, ithas also been reported that in some cases, pressure-killed stationary-phasecells retain a functional membrane (24), and this has led tothe consideration of other structures inside the cell as potentialkey targets for inactivation. Some authors have reported similaritiesbetween cell inactivation and protein denaturation kinetics(19, 36), and changes in the conformation of various structuresin the bacterial cell, such as the nucleoid and ribosomes, havebeen reported for electron microscopy and differential scanningcalorimetry studies (20, 23, 26). However, it is still not clearwhich, if any, of these changes is responsible for the inactivationof the cell.

The aim of this work was to examine various morphological andphysiological changes that occur in pressurized Escherichiacoli cells that might be correlated with cell death. Changesin gross morphology, the loss of membrane integrity, the lossof osmotic responsiveness, changes in the conformation of nucleicacids, and protein coagulation were studied in exponential-and stationary-phase cells after exposure to different pressuretreatments.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains and growth conditions.
E. coli J1, an acid-resistant commensal strain, was kindly providedby Ian Booth, University of Aberdeen, Aberdeen, United Kingdom.Stationary-phase cultures were prepared by inoculating 10 mlof tryptone soy broth (Oxoid, Basingstoke, United Kingdom) supplementedwith 0.3% yeast extract (TSB-YE) with a loopful of growth fromtryptone soy agar supplemented with 0.3% yeast extract (Oxoid)and incubating the resulting culture at 37°C for 6 h ina shaking incubator. Fifty microliters of this culture was inoculatedinto 50 ml of fresh TSB-YE and incubated for 18 h under thesame conditions, which resulted in a stationary-phase culturecontaining approximately 5 x 10⁹ cells/ml. Exponential-phasecells were prepared by inoculating 50 µl of the stationary-phaseculture into 50 ml of fresh TSB-YE and incubating the culturefor 2.5 to 3.0 h, until the optical density at 680 nm reached0.3, corresponding to 10⁸ cells/ml.

Pressure treatment.
Cells were centrifuged at 3,000 x g for 20 min at 4°C, andthe pellets were resuspended in phosphate-buffered saline (PBS;pH 7.2). Cell suspensions (0.5 to 2.0 ml) were placed in sterileplastic pouches that were heat sealed and placed on ice beforepressurization. Samples were pressure treated in a 300-ml pressurevessel (model S-FL-850-9-W; Stansted Fluid Power, Stansted,United Kingdom). The pressure-transmitting fluid was ethanol-castoroil (80:20). Cells were exposed to pressures from 100 to 600MPa for 8 min. Pressure treatments were done at an ambient temperature(approximately 20°C), and the transient increase in temperatureof the pressurization fluid caused by adiabatic heating wasmeasured with a thermocouple. The maximum temperature reachedduring pressurization at 600 MPa was 43°C. After decompression,the pouches were removed from the unit and placed on ice beforeviable counts and other tests were performed.

Viable counts.
Samples were serially diluted in maximum recovery diluent (Oxoid)and plated onto tryptone soy agar-yeast extract supplementedwith 0.1% (wt/vol) sodium pyruvate. Colonies were counted after48 h of incubation at 37°C. Data presented are the meanvalues from at least three independent experiments, and errorbars correspond to standard deviations of the means.

Cell staining.
Several fluorescent dyes were used to visualize different structuresin the cells. Samples in PBS were mixed 1:1 (vol/vol) separatelywith the dye solutions, at the concentrations described below,and were immediately prepared for microscopic examination. Nucleoidswere visualized by staining with 100 µM 4',6-diamidino-2-phenylindole(DAPI) (Sigma-Aldrich, Poole, Dorset, United Kingdom). Cellenvelopes were stained with a 165 µM solution of the lipophilicdye FM 4-64 (Molecular Probes, Eugene, Oreg.). Protein was stainedwith a 130 µM solution of fluorescein isothiocyanate (FITC)(Sigma-Aldrich). A solution of 83 µM acridine orange (Polysciences,Eppelheim, Germany) was used to simultaneously visualize double-stranded(green fluorescence) and single-stranded (red fluorescence)nucleic acids. Samples were prepared for microscopy as follows:1 µl of either unstained or stained cell suspension wasspread on a glass coverslip and immobilized by placement ofthe coverslip onto a glass slide coated with a very thin flatlayer of 1% technical agar.

Examination of the ability of cells to plasmolyze.
To study the ability of the cells to respond to an osmotic shock,we spread 1 µl of sample onto a glass coverslip, gentlyplaced the coverslip on a slide coated with 1% agar containing0.75 M NaCl, and observed it under a microscope.

Microscopic examination.
Samples were examined under a Microphot-SA microscope (NikonCorporation, Tokyo, Japan) equipped with phase-contrast opticsand an epifluorescence unit. For unstained and plasmolyzed cells,phase-contrast optics alone were used. For cells stained withFM 4-64, FITC, and acridine orange, epifluorescence light withthe appropriate filters was used, and for DAPI staining, cellswere viewed simultaneously by phase-contrast and epifluorescencemicroscopy, as described by Hiraga et al. (13). In all cases,a x100 objective was used with immersion oil, giving a totalmagnification of x1,000. Images were captured with a 12-bit1,392- by 1,040-pixel monochrome charge-coupled device camera(CoolSnap Pro_cf) and were processed with Image-Pro Plus, v.4.5 (Media Cybernetics, Silver Spring, Md.).

Quantification of protein released after pressurization.
Suspensions of exponential- and stationary-phase cells grownas described above were adjusted to a final concentration ofapproximately 2 x 10⁹ cells/ml in PBS and were pressure treatedat either 200 MPa for 8 min (exponential- and stationary-phasecells) or 600 MPa for 8 min (stationary-phase cells only). Fivehundred microliters of the pressurized samples was centrifugedat 3,000 x g for 20 min at 4°C to pellet whole cells. Pelletswere resuspended in 500 µl of sterile PBS, and the proteinconcentrations in both fractions (supernatant and cells) werequantified by the Bradford assay (Sigma), with bovine serumalbumin as the standard.

RESULTS

Top

Results

Resistance to high hydrostatic pressure.
The inactivation of exponential- and stationary-phase cellsof E. coli J1 after treatment for 8 min at different pressuresup to 600 MPa is shown in Fig. 1. Stationary-phase cells weremuch more resistant to high pressure than were exponential-phasecells. For example, after 8 min at 300 MPa, >70% of stationary-phasecells remained viable compared to <0.01% of exponential-phasecells. No decrease in viability was observed after pressurizationat 100 MPa for either exponential- or stationary-phase cells.Pressures higher than 100 MPa caused a sharp decrease in theviability of exponential-phase cells, while stationary-phasecells survived up to 500 MPa with little loss of viability (viabilityreduction of 57%). Treatment at 600 MPa inactivated 99.99% ofstationary-phase cells. We cannot discount a combined effectof pressure plus heat at the higher pressures tested. Althoughnonlethal temperatures are reached during adiabatic heatingdue to the pressure increase, i.e., 43°C at 600 MPa and40°C at 500 MPa, other authors have demonstrated that pressuretreatments at moderately high temperatures are more bactericidalthan those at low temperatures (7).

View larger version (16K):
[in this window]
[in a new window]
FIG. 1. Pressure resistance of E. coli J1 during the exponential ( ${blacktriangleup}$ ) and stationary (•) phase. Data shown are percentages of survivors after 8 min of treatment at different pressures. Each point corresponds to the mean of at least three independent experiments. The initial inoculum levels were 5 x 10⁹ cells/ml for stationary-phase cells and 10⁸ cells/ml for exponential-phase cells.

In the next series of experiments, exponential-phase and stationary-phasecells were compared after treatments that resulted in approximatelythe same degree of inactivation (4- to 5-log decrease). Exponential-phasecells were treated at 200 and 300 MPa, and stationary-phasecells were treated at 600 MPa. We found that the various changesin exponential-phase cells were evident after treatment at 200MPa, but since the appearance of the cells treated at 200 MPawas essentially similar to that of those treated at 300 MPa,either pressure was chosen for illustrative purposes for thefigures.

Morphological changes visible under phase-contrast optics.
Figure 2 shows the effect of pressure on the appearance of cellsviewed by phase-contrast microscopy. Untreated stationary-phasecells (Fig. 2B) were smaller and more rounded than those inthe exponential phase (Fig. 2A), as reported by others, butwere not spherical in form as occurs after starvation in a minimalmedium (15). The appearances of exponential- and stationary-phasecells after pressure treatments at 300 and 600 MPa, respectively,are shown in Fig. 2C and D. Untreated cells of both types presenteda dark homogeneous cytoplasm. Pressurization caused a generallightening of the cytoplasm, which was more evident in exponential-phasecells (Fig. 2C) than in stationary-phase ones (Fig. 2D). Exponential-phasecells showed a slight granularity of the cytoplasm, but in stationary-phasecells, distinct small rounded areas of high refractive abilitywere observed (Fig. 2D). An image analysis of the cells beforeand after pressurization showed no significant changes in size(cross-sectional area) for either exponential- or stationary-phasecells (data not shown).

View larger version (106K):
[in this window]
[in a new window]
FIG. 2. Effect of pressure on general appearance of E. coli J1 by phase-contrast microscopy of exponential- and stationary-phase cells. (A) Untreated exponential-phase cells (100% viable cells); (B) untreated stationary-phase cells (100% viable cells); (C) exponential-phase cells treated with 300 MPa for 8 min (0.002% viable cells); (D) stationary-phase cells treated with 600 MPa for 8 min (0.01% viable cells). Bars, 2 µm.

Membrane integrity of pressure-treated cells.
Figure 3 shows the osmotic responses of untreated and pressurizedcells. Untreated cells that possessed an intact cytoplasmicmembrane showed a very refractile cytoplasm when placed on agarcontaining 0.75 M NaCl which was caused by the loss of waterand consequent concentration of the cytoplasm under hypertonicconditions (Fig. 3A and C). Exponential-phase cells appeareduniformly refractile and slightly shrunken in appearance (Fig.3A), whereas in stationary-phase cells the refractile cytoplasmhad apparently become separated from the poles of the cell (Fig.3C). In cells with a nonfunctional membrane, water and saltcan pass freely through the membrane, and there is no osmoticresponse and no condensation of the cytoplasm, which thereforeappears dark gray. Exponential-phase cells pressurized at 100MPa remained viable and retained the ability to respond to anosmotic challenge (data not shown), but those pressurized at200 MPa, by which >99.99% of the population was killed, hadcompletely lost the osmotic response (Fig. 3B). Stationary-phasecells maintained the ability to plasmolyze with all treatmentsup to 500 MPa (Fig. 3D and E, in which the percentages of viablecells were 79 and 43%, respectively). Only the cells pressurizedat 600 MPa showed an effect of pressure on osmotic responsiveness,as can be seen in Fig. 3F, in which some of the cells do notshow a refractile cytoplasm. However, although >99.99% ofthe cells pressurized at 600 MPa were dead, approximately 60%were still able to plasmolyze.

View larger version (120K):
[in this window]
[in a new window]
FIG. 3. Effect of pressure on the osmotic response of E. coli J1. Phase-contrast photomicrographs of exponential- and stationary-phase cells placed on agar containing 0.75 M NaCl are shown. (A) Untreated exponential-phase cells (100% viable cells); (B) exponential-phase cells treated with 200 MPa for 8 min (0.02% viable cells); (C) untreated stationary-phase cells (100% viable cells); (D) stationary-phase cells treated with 200 MPa for 8 min (79% viable cells); (E) stationary-phase cells treated with 500 MPa for 8 min (43% viable cells); (F) stationary-phase cells treated with 600 MPa for 8 min (0.01% viable cells). Bars, 2 µm.

Figure 4 shows the appearances of untreated and pressurized(300 MPa) exponential-phase cells stained with FM 4-64, a lipophilicdye that binds to both outer and cytoplasmic membranes, butpreferentially to the cytoplasmic membrane (9). Pressurizationcaused a visible disruption of the envelopes of exponential-phasecells, including the formation of vesicles of a lipidic materialcoming out from the cells and also an appearance of membranethickening and the formation of local structures resemblinginvaginations protruding into the cytoplasm. However, for stationary-phasecells, no structural changes in the membrane were detected withinthe range of pressures studied (data not shown).

View larger version (69K):
[in this window]
[in a new window]
FIG. 4. Effect of pressure on membrane integrity of exponential-phase cells of E. coli J1. Fluorescence photomicrographs of cells stained with the lipophilic membrane probe FM 4-64 are shown. (A) Untreated cells (100% viable cells); (B) cells pressurized at 300 MPa for 8 min (0.002% viable cells). Bars, 2 µm. Arrows show internal and external vesicles.

Effect of pressure on the nucleoid.
Figure 5 shows phase-contrast-fluorescence micrographs of untreatedand pressurized cells stained with DAPI, a permeative intercalatingprobe that binds to DNA. Staining with DAPI revealed gross changesin the nucleoid conformation for both types of cells. In untreatedexponential- or stationary-phase cells (Fig. 5A and B), thefluorescence was evenly distributed inside the cytoplasm. However,pressures higher than 100 MPa caused condensation of the nucleoid,which was very dramatic in exponential-phase cells pressurizedat 200 MPa or more (Fig. 5C and E). Condensation of the nucleoidalso occurred in stationary-phase cells, starting at 200 MPa,and became progressively more intense at increasing pressures(Fig. 5D and F). In pressure-treated cells, the nucleoids wereoften skewed across the cell or pulled to one side next to themembrane.

View larger version (86K):
[in this window]
[in a new window]
FIG. 5. Effect of pressure on nucleoid configuration of E. coli J1. Phase-contrast-fluorescence photomicrographs of exponential- and stationary-phase cells stained with DAPI are shown. (A) Untreated exponential-phase cells (100% viable cells); (B) untreated stationary-phase cells (100% viable cells); (C) exponential-phase cells treated with 300 MPa for 8 min (0.002% viable cells); (D) stationary-phase cells treated with 300 MPa for 8 min (73% viable cells); (E) exponential-phase cells treated with 400 MPa for 8 min (<0.0001% viable cells); (F) stationary-phase cells treated with 400 MPa for 8 min (61% viable cells). Bars, 2 µm.

Effect of pressure on RNA distribution.
For the examination of RNA conformations, cells were stainedwith acridine orange, which binds to single- or double-strandednucleic acids, giving red or green fluorescence, respectively.DNA thus appears green and RNA appears red. When we used thisdye, a long-pass filter (590 nm) was placed in the light pathto allow for selective visualization of the red-staining RNA.Untreated cells stained rather poorly with acridine orange,but the red fluorescence (RNA) was evenly distributed throughoutthe cytoplasm, with no evident areas of higher intensity (datanot shown). In pressure-treated exponential-phase cells, thered-stained material had a condensed appearance (Fig. 6A). Itwas also noticeable that a considerable amount of red-stainedmaterial was released into the extracellular medium. This wasobserved only for exponential-phase cells (Fig. 6A), and norelease of material was observed for pressurized stationary-phasecells (Fig. 6B).

View larger version (63K):
[in this window]
[in a new window]
FIG. 6. Effect of pressure on RNA configuration of E. coli J1. Fluorescence photomicrographs of exponential- and stationary-phase cells stained with acridine orange are shown. (A) Exponential-phase cells treated with 200 MPa for 8 min (0.02% viable cells); (B) stationary-phase cells treated with 600 MPa for 8 min (0.01% viable cells). Bars, 2 µm. The arrow indicates stained material released from the cell.

Effect of pressure on intracellular protein.
Figure 7 shows cells stained with FITC, a fluorescent dye thatbinds specifically to proteins. Untreated cells had an evenlydistributed low-intensity fluorescence within the cytoplasm(data not shown). Pressurization caused the condensation ofproteins in both exponential- and stationary-phase cells. Inexponential-phase cells, clumps of aggregated protein were distributedinside the whole cell but were preferentially close to the membrane(Fig. 7A). In stationary-phase cells, protein aggregates werebigger and more intensely stained (Fig. 7B and C). The proteinaggregates appeared in the whole population of cells at pressuresas low as 200 MPa, and their appearance did not change appreciablyat higher pressures. The amount of protein released from cellsafter pressurization was also measured, and the results areshown in Table 1. Exponential-phase cells lost a much largerproportion of cellular protein to the external medium than didstationary-phase cells. No attempt was made to solubilize proteinaggregates present in the cell, which may have led to an underestimateof the protein content of pressure-treated cells and probablyexplains why the amount of cytoplasmic protein apparently decreasedmore than the amount lost from the cell.

View larger version (48K):
[in this window]
[in a new window]
FIG. 7. Effect of pressure on cellular protein of E. coli J1. Phase-contrast-fluorescence photomicrographs of exponential- and stationary-phase cells stained with FITC are shown. (A) Exponential-phase cells treated with 200 MPa for 8 min (0.02% viable cells); (B) stationary-phase cells treated with 300 MPa for 8 min (73% viable cells); (C) stationary-phase cells treated with 600 MPa for 8 min (0.01% viable cells). Bars, 2 µm.

View this table:
[in this window]
[in a new window]
TABLE 1. Protein content of whole cells and pressurizing medium (supernatant) of untreated and pressurized exponential- and stationary-phase cells of E. coli J1

DISCUSSION

Top

Discussion

In an attempt to distinguish between events that are associatedwith a loss of viability and those that are not, we examinedseveral morphological and physiological changes that occur whencells of E. coli are exposed to different pressures and comparedthese changes to the loss of viability of exponential- and stationary-phasecells. The results are summarized in Table 2.

View this table:
[in this window]
[in a new window]
TABLE 2. Morphological and physiological changes of E. coli strain J1 cells exposed to different pressures for 8 min

The pressure for the onset of rapid inactivation was between100 and 200 MPa for exponential-phase cells and between 500and 600 MPa for stationary-phase cells, a difference of 400MPa. This difference in resistance is very similar to that seenfor exponential- and stationary-phase cells of E. coli O157strain C9490 (24). In general, the resistance of exponential-phasecells varies very little between strains (4, 24), but largedifferences occur in stationary-phase cells, due mainly to differencesin rpoS status (29), and this has a large effect on the relativeresistances of exponential- and stationary-phase cells.

The main differences in the accompanying morphological and physiologicalchanges were as follows: a complete loss of the plasmolysisresponse occurred in exponential-phase cells between 100 and200 MPa, whereas only a partial loss was seen in stationary-phasecells at the much higher pressure of 600 MPa; and a visibleperturbation of membrane structure and a loss of RNA occurredat 100 to 200 MPa in exponential-phase cells, but correspondingchanges were absent from stationary-phase cells. Condensationof the nucleoid and the formation of protein aggregates occurredin both cell types with about 200 MPa of pressure.

In this work, we observed markedly different behaviors of themembranes of exponential- and stationary-phase cells towardpressurization. Membranes of exponential-phase cells showedphysical disruption after pressurization, with the formationof vesicles, areas of engrossment, and also invaginations towardthe cytoplasm (Fig. 4). The observed vesicles resembled thoseobserved by Katsui et al. (17) with heat-treated cells, whichwere formed from lipids from the outer membrane. It is possiblethat vesicles observed in pressurized exponential-phase cellsare from outer membrane materials, whereas engrossment areas,which occupy part of the cytoplasm, are from the cytoplasmicmembrane (Fig. 4). The membranes of stationary-phase cells remainedvirtually intact. Pagán and Mackey (24) previously suggestedthat the role of the cytoplasmic membrane as a critical targetfor the inactivation of cells by high pressure was differentin exponential- and stationary-phase cells, but the reasonsfor the apparent differences in membrane behavior in responseto pressure are unknown.

Recent work on the role of membrane fluidity on cellular pressureresistance (5) provided evidence that pressure resistance increaseswith membrane fluidity. However, membrane fluidity seems tobe less important for stationary-phase cells than for exponential-phaseones, because unspecified changes, possibly linked to RpoS activity,occur during stationary phase that have a greater influenceon pressure resistance than membrane fluidity and hence maskits effect (5, 29). In fact, according to the microscopic studiesof Beney et al. (3) on phospholipid vesicles, the factor thatdetermines whether gross structural perturbations of the membranevesicles occur under pressure is the difference in compressibilitybetween the membrane and the water inside. Those authors demonstratedthat vesicles with a more compressible membrane lose part oftheir aqueous content during pressurization and change in shape(produce buds) upon decompression due to the excess of membranein relation to the remaining aqueous content. However, vesiclescontaining cholesterol, which have less compressible membranes,undergo a smaller decrease in volume and do not change in shapeupon decompression. Therefore, if bacterial cells were to behavesomewhat like these vesicles, one could infer that exponential-phasecells would have more compressible membranes than those in stationaryphase, and in consequence, would undergo more drastic changesin shape and form vesicles from the excess membrane materialupon decompression. Conversely, stationary-phase cells wouldbe expected to have less compressible membranes or ones thatcould resist the loss of material after the release of pressure.

When cells enter the stationary phase of growth, they developa generalized resistance to a variety of stresses, partly dueto the action of the alternative sigma factor ${sigma}$ ^s, which controlsthe transcription of >50 genes (15). A range of morphologicaland physiological changes have been described that might possiblyaccount for the increase in pressure resistance. The cell envelope(outer membrane, cell wall, and cytoplasmic or inner membrane)drastically changes upon entry into the stationary phase. Thereis an increase in cardiolipin, synthesized from phosphatidylglycerol(8), and the unsaturated fatty acids are converted into theircyclopropyl fatty acid derivatives (11). Stationary-phase cellsalso have a higher protein/lipid ratio in their membranes, whichmakes them less prone to lateral phase separation (37), anda higher degree of cross-linking among membrane proteins (22).Also, the accumulation of trehalose, the increased level ofstabilizing polyamines, and the thickening of the wall duringstationary phase (four to five layers of peptidoglycan comparedto only two or three during growth) (15, 22) could contributeto the high level of stability of the cell envelopes of stationary-phasecells.

From the results obtained in this work, we can conclude thatin exponential-phase cells the loss of viability is always accompaniedby a loss of the physical integrity of the membrane, whereasin stationary-phase cells membranes can remain physically intact,even in dead cells. This leads to the suggestion that the inactivationof stationary-phase cells by high pressure may occur by a mechanismother than the permanent loss of membrane integrity.

One of the most obvious changes in both exponential- and stationary-phasecells was the condensation of the nucleoid (Fig. 5). The condensednucleoids of pressure-treated cells were often irregularly positionedwithin the cell, unlike those seen in chloramphenicol-treatedcells, which are generally rather symmetrical and central tothe cell axis. This asymmetric appearance has also been observedwith pressure-treated cells of S. enterica serovar Thompsonexamined by electron microscopy (20). Condensation of the nucleoidwas also observed for the gram-positive organisms Listeria monocytogenesand Lactobacillus plantarum (20, 39). Exponential-phase cellsshowed a more drastic change in nucleoid shape than those inthe stationary phase, possibly due to the absence of compactingDNA-binding proteins, such as Dps, which is synthesized in responseto oxidative stress but is also synthesized during stationaryphase under the complex control of the stationary-phase regulator ${sigma}$ ^s (15). The difference in appearance of the nucleoid could alsobe explained by a different initial conformation of the DNAin exponential-phase cells, in which active transcription andtranslation are continuously taking place in the cytoplasm.There may also be attachments between some nascent proteinsand the membrane. In any case, after 8 min at 400 MPa, about60% of stationary-phase cells were still alive, but >90%of the cells contained a highly condensed nucleoid. We can concludethat in exponential-phase cells the loss of viability is correlatedwith changes in nucleoid conformation, but for stationary-phasecells this morphological change occurs at pressures below thosewhich cause death. The most straightforward conclusion fromthese results is that changes in the nucleoid are not relatedto cell death in stationary-phase cells.

It is noteworthy that neither the loss of membrane integritynor nucleoid condensation happened at 100 MPa. A pressure valuebetween 100 and 200 MPa seemed to be the threshold for the initiationof nucleoid changes in both types of cells, despite the enormousdifference in pressure sensitivity between them. Previous work(24) demonstrated that stationary-phase cell membranes becomepermeable during pressurization but are able to reseal afterwards.This transient permeabilization phenomenon happened only atpressures above 100 MPa. We can speculate, then, that some changeshappening in the cells at pressures above 100 MPa could be indirectlyrelated to the permeabilization of the membrane during pressurizationand to consequent changes in the intracellular environment.For example, membrane damage could lead to the loss of magnesiumions, which are essential for the maintenance of nucleoid andribosome structure and are known to protect cells against high-pressureinactivation (12).

Staining with acridine orange revealed further changes in thecytoplasm of the cells. Exponential-phase cells lost RNA tothe extracytoplasmic medium (Fig. 6A). This leakage was neverobserved with stationary-phase cells. These results confirmthat membranes of exponential-phase cells are much more pressuresensitive than are those of stationary-phase cells and alsoindicate that both cytoplasmic and outer membranes are extensivelydamaged, allowing the leakage of large molecules such as RNAfrom the cytoplasm. In stationary-phase cells, pressurizationcaused the appearance of strongly fluorescent areas of condensedRNA inside the cytoplasm, which could reflect changes in ribosomeconformation. This would agree with calorimetric studies showingthat pressure causes denaturation of the ribosome (23).

The cytoplasm of pressure-treated stationary-phase cells containednumerous protein-containing aggregates throughout the cytoplasm,often close to the membrane (Fig. 7B and C). Further work willbe needed to determine the types of protein contained in theaggregates and to establish whether other components, such asRNA, are also present. This coagulated protein was also observedin exponential-phase cells, but to a lesser extent. This maybe explained partly by the fact that exponential-phase cellslose a much larger amount of protein to the extracellular medium(Table 1), but it could also possibly be caused by a differentstate of the cytoplasmic protein in the two types of cell. Theamount and distribution of denatured protein in stationary-phasecells appeared to be similar in cells pressurized at nonlethal(300 MPa) and lethal (600 MPa) pressures, indicating again thatthis kind of damage is not lethal for the cell.

From all these data, we can conclude that exponential- and stationary-phasecells undergo several common changes, such as condensation ofthe nucleoid and the aggregation of protein, but there are someother changes that are specifically associated with exponential-phasecells and are correlated with the loss of viability of thesecells. These include all of the changes related to membraneintegrity, such as the formation of vesicles and irregularitiesin the membrane ultrastructure, the loss of osmotic responsiveness,and the loss of protein and RNA to the external medium. Otherworkers have also suggested that the cell membrane is a criticaltarget for the inactivation of cells by high pressure and havesuggested, for example, that death may be associated with theloss of specific membrane functions, such as ATPase activityor multidrug efflux systems (34, 38). From the results presentedhere, we propose a simpler explanation, at least for exponential-phasecells: we suggest that death results from a catastrophic andirreversible loss of physical membrane integrity.

In contrast, the ability of stationary-phase membranes to resealafter pressure treatment may allow a proportion of the cellpopulation to maintain homeostasis and to reverse the deleteriouseffects, such as protein aggregation or nucleoid condensation,of pressure. The ability of stationary-phase cells to recoverafter pressure treatment may also be assisted by the presenceof molecular chaperones, such as the DnaK protein, that havean essential role in stress resistance and are specificallyinduced during stationary phase (30, 31). However, after a certainpressure level, at about 600 MPa, the damage is irreparablein a majority of the population and there is a drastic decreasein viability. Further work is required to clarify the causesof this drastic decrease in the viability of stationary-phasecells.

ACKNOWLEDGMENTS

We thank the European Commission for providing Pilar Mañaswith a Marie Curie postdoctoral grant in the Quality of Lifeprogram.

FOOTNOTES

* Corresponding author. Mailing address: School of Food Biosciences, University of Reading, Whiteknights, Reading RG6 6AP, United Kingdom. Phone: 441189357229. Fax: 441189357222. E-mail: b.m.mackey{at}reading.ac.uk.

Present address: Departamento de Producción Animal yCiencia de los Alimentos, Facultad de Veterinaria, 50013 Zaragoza,Spain.

REFERENCES

Top