Influence of the Natural Microbial Flora on the Acid Tolerance Response of Listeria monocytogenes in a Model System of Fresh Meat Decontamination Fluids

Abstract

Depending on its composition and metabolic activity, the natural flora that may be established in a meat plant environment can affect the survival, growth, and acid tolerance response (ATR) of bacterial pathogens present in the same niche. To investigate this hypothesis, changes in populations and ATR of inoculated (10⁵ CFU/ml) Listeria monocytogenes were evaluated at 35°C in water (10 or 85°C) or acidic (2% lactic or acetic acid) washings of beef with or without prior filter sterilization. The model experiments were performed at 35°C rather than lower (≤15°C) temperatures to maximize the response of inoculated L. monocytogenes in the washings with or without competitive flora. Acid solution washings were free (<1.0 log CFU/ml) of natural flora before inoculation (day 0), and no microbial growth occurred during storage (35°C, 8 days). Inoculated L. monocytogenes died off (negative enrichment) in acid washings within 24 h. In nonacid (water) washings, the pathogen increased (approximately 1.0 to 2.0 log CFU/ml), irrespective of natural flora, which, when present, predominated (>8.0 log CFU/ml) by day 1. The pH of inoculated water washings decreased or increased depending on absence or presence of natural flora, respectively. These microbial and pH changes modulated the ATR of L. monocytogenes at 35°C. In filter-sterilized water washings, inoculated L. monocytogenes increased its ATR by at least 1.0 log CFU/ml from days 1 to 8, while in unfiltered water washings the pathogen was acid tolerant at day 1 (0.3 to 1.4 log CFU/ml reduction) and became acid sensitive (3.0 to >5.0 log CFU/ml reduction) at day 8. These results suggest that the predominant gram-negative flora of an aerobic fresh meat plant environment may sensitize bacterial pathogens to acid.

The initial microbial contamination and the processing and storage conditions of meat affect the microbial flora that dominates, which also affects survival and growth of bacterial pathogens present (13, 27, 40, 47). The growth and metabolic activity of these dominating organisms either leads to spoilage (13, 42, 47) or contributes to preservation (32, 40). Regardless of its spoilage or beneficial potential, the population of the indigenous, nonpathogenic microbial flora may be several logarithmic cycles higher than that of the naturally occurring pathogens (40, 47). Thus, the natural flora is usually expected to have a competitive advantage over pathogens for nutrient uptake and may alter the intrinsic factors (pH, water activity, nutrient catabolism, production of inhibitory metabolites, etc.) of meat or other foods in ways that the survival and growth of pathogens may be stimulated (19, 33, 34), not affected (5, 34), or suppressed (4, 10, 11, 35, 40, 50). Similar associations may also develop in meat plant environments where reservoirs of microbial contamination may be established (41).

Although many predictive models have been developed with data derived from experiments with pure pathogenic cultures in liquid media or simulated food systems (12, 20, 36), recent studies have started to evaluate growth kinetics of bacterial pathogens in mixed cultures with spoilage bacteria such as pseudomonads and lactic acid bacteria (LAB) (8, 9, 16, 18). In addition, specific models have been proposed, designed, evaluated or revalidated to predict growth of mixed microbial populations in meat and other foods (1, 7, 28, 49, 51). This may indicate an increasing scientific interest in the effects of microbial competition on food preservation and safety.

Meat decontamination technologies, including spraying or rinsing of animal carcasses before evisceration and/or before chilling with cold or hot water or chemical solutions (e.g., organic acids), or exposure to steam, are used extensively in the United States to reduce microbial contamination in meat (15, 44, 45, 48). These interventions are integrated into meat safety management systems, such as hazard analysis critical control point, which is required by regulation in the United States (21, 46), and are proposed to be used in meat cuts before packaging and shipment to the market (39, 45, 48). Despite their established effectiveness in reducing carcass surface contaminants (by approximately 1 to 3 logs), the residual efficacy of decontamination interventions and their residual effects on the natural flora of the meat and the plant environment have not been established adequately (44, 45, 48). The potential development of stressed bacterial pathogens that may cross-contaminate and multiply on meat or colonize meat processing plants are important safety issues that have yet to be addressed (45, 48). The risk of creating stressed pathogens may be greater following organic acid decontamination due to the residual effect of acids on the meat or in the waste and the longer adaptation times of pathogens to acid, especially if microbial niches are established in the plant environment.

The impact of pathogen reduction strategies on microbial ecology of meat plant environments and resulting products and the relative safety of products with low compared to high numbers of background flora has been questioned (26). It has been suggested that a possible explanation for the increasing number of foodborne outbreaks in the United States is that food may now have low numbers of antagonists to suppress pathogen survival and growth, and that could be due to decontamination (25, 26). Indeed, high levels of natural flora were recently shown to inhibit growth of Escherichia coli O157:H7 in ground beef (50). However, to our knowledge, limited data exist on the interactions between the natural flora and responses of pathogenic bacteria to stresses (2, 17), even though it seems logical that such interactions should occur in food environments (3, 22, 47).

The objective of this study was to investigate the effect of the natural flora on the acid tolerance response (ATR) of Listeria monocytogenes in a model system simulating waste fluids that may be present in a fresh beef processing plant. Following inoculation (10⁵ CFU/ml) of the pathogen in fresh beef decontamination fluids (washings) with or without prior filter sterilization, bacterial growth was monitored at 35°C while the ATR was evaluated by periodic exposure of L. monocytogenes cells grown in the absence or presence of background flora to acid. High (35°C) incubation rather than lower (≤15°C) commercial temperatures were selected in order to maximize the growth potential and ATR of L. monocytogenes in presence versus absence of a competitive natural flora. Although E. coli O157:H7 and Salmonella are the main target organisms of meat decontamination (44, 48), L. monocytogenes was selected because of its high incidence on fresh meat and poultry (41, 47), its prevalence in meat processing plants (41), its potential to survive meat decontamination when present as a natural contaminant (39), and its resistance to multiple stresses and superior ability to form biofilms (30, 31).

(Part of this work was presented as poster P-055 at the 87th Annual Meeting of International Association for Food Protection, August 6 to 9, 2000, Atlanta, Ga.)

MATERIALS AND METHODS

Bacterial strains.

L. monocytogenes N-7155 (serotype 1/2b) isolated from meat (6) was used throughout this study. In addition, a streptomycin (800 μg/ml)-resistant derivative of another meat isolate (6) of L. monocytogenes, N-7144 (serotype 1/2b, N-7144Sm+), developed in previous studies (54) was used in some experiments to selectively detect the pathogen on all-purpose media in order to support findings on selective media. Selection of strains was based on their meat origin and increased acid tolerance compared to other L. monocytogenes strains tested, including strain Scott A commonly used in food research, as will be described later. Strains were available as frozen (−70°C) stock cultures in Trypticase soy broth (BBL, Becton Dickinson Co., Cockeysville, Md.) with 0.6% yeast extract (Difco, Becton Dickinson Co., Sparks, Md.) (TSBYE), supplemented with 20% glycerol. Strains were activated by transferring 0.05 ml of stock inoculum in 10 ml of TSBYE followed by overnight incubation at 35°C. Working cultures were kept on Trypticase soy agar (BBL) with 0.6% yeast extract (TSAYE) slants at 4°C and transferred monthly. Strains were subcultured twice in TSBYE at 35°C for 24 h before use in experiments.

Preparation of fresh meat decontamination fluids (washings).

Fresh (≤72 h postmortem), nondecontaminated top rounds of beef were obtained from a commercial plant or from the Meat Science Laboratory of Colorado State University and used to prepare washings within 24 h after transportation. Each top round was cut into four portions weighing approximately 2 kg each. Each portion was individually spray washed with 2 liter of one of the following solutions: (i) cold (10°C) tap water, (ii) hot (85°C) tap water, (iii) a 2% warm (55°C) solution of lactic acid (dl-lactic acid; 85% [wt/wt]; Sigma, St. Louis, Mo.) in tap water, or (iv) a 2% warm (55°C) solution of acetic acid (glacial acetic acid; 100%; Mallinckrodt Baker Inc., Paris, Ky.) in tap water. These decontamination treatments were selected because they are commonly used in meat plants in the United States. An additional reason for spraying meat with water at two different temperatures (10 or 85°C) was to account for potentially lower populations of natural flora, and potentially higher organic matter, in hot compared to cold water washings, due to the greater bacterial killing and nutrient extracting effect of 85°C spray water. Such potential differences might affect subsequent growth of L. monocytogenes inoculated in the washings after meat decontamination. Spraying of meat portions was done with hand spray washers (Contico Int., St. Louis, Mo.). During spraying, the nozzle of the spray washer was kept at approximately 20 cm from the meat surface, while it was moved slowly around the suspended meat to ensure uniform spraying of all sides. A plastic bowl was placed under the meat to collect the washings. The meat was allowed to drain for 5 min, and then the washings were distributed in 500-ml amounts in presterilized screw-capped bottles (Nalgene). The washings were kept at 4°C if handled on the same day, or stored at −30°C and used within 30 days, following overnight thawing at 4°C. For use in these experiments, each type of washing was divided in two parts. One part was inoculated with L. monocytogenes without any treatment (unfiltered washings), as described later, while the other part was filtered through Whatman no. 1 filter paper using a Buchner funnel under vacuum to remove large meat particles, and then each was centrifuged at 12,000 rpm (Beckman model J2-21 centrifuge) for 30 min at 4°C to remove small particles. The supernatant was filter (0.2-μm pore size; Nalgene) sterilized under vacuum.

After preparation, the uninoculated meat washings were tested microbiologically, as described below, to ensure sterility of the filter-sterilized washings or to determine the numbers of natural meat flora and any naturally occurring Listeria spp. in unfiltered washings. Also, the pH of the washings was measured before storage and during the experiments. A digital pH meter (Accumet 50; Fisher Scientific, Houston, Tex.) with a glass pH electrode (Hanna Instruments, Ann Arbor, Mich.) was used for measurement of pH.

Culturing of L. monocytogenes in the washings.

Portions (100 ml) of each type of filter-sterilized or unfiltered washings were distributed in 250-ml presterilized bottles (Nalgene) and inoculated with stationary-phase cultures (TSBYE at 35°C for 24 h) of L. monocytogenes N-7155 or N-7144Sm+ (approximately 10⁵ cells/ml). Treatments, including uninoculated (control) washings of meat, were incubated statically at 35°C for 8 days. Samples were taken for microbiological analysis and pH measurement at 0, 1, 4, and 8 days of incubation. Serial decimal dilutions in 0.1% buffered peptone water (Difco) were prepared and then plated in duplicate on TSAYE and PALCAM (Difco) agar plates to determine total bacterial counts and populations of inoculated L. monocytogenes, respectively. When strain N-7144Sm+ was inoculated in unfiltered washings, TSAYE supplemented with 800 μg of streptomycin sulfate (Sigma) per ml (TSAYE-Sm) was used, in addition to PALCAM, to selectively enumerate L. monocytogenes. Colonies on agar plates were counted after incubation at 35°C for 48 h. For counts below the analysis detection limit (<1.0 log CFU/ml), 5 ml of culture was added to 45 ml of Listeria enrichment broth (Difco) and incubated at 35°C for 48 h for enrichment. Portions (0.1 ml) of the enriched culture were spread in duplicate on PALCAM plates and incubated at 35°C for 48 h, after which plates were checked for Listeria growth.

Ten percent of colonies on countable TSAYE plates were tested for Gram reaction by mixing colonies with a 3% KOH solution to observe presence (gram negative) or absence (gram positive) of a slimy suspension. Representative gram-negative colonies (three to five) were subjected to the oxidase test by using an Oxy-Swab (Remel Inc., Lenexa, Kans.) or dry slide (BBL) kit. All gram-positive colonies on plates were subjected to the catalase test by dropping a 3% H₂O₂ solution directly onto them to observe effervescence. The microscopic appearance of representative colonies was checked in wet mount or Gram-stained cultures.

Assessment of acid tolerance.

As mentioned, acid challenge experiments were performed to select strains N-7155 and N-7144Sm+ as the most acid tolerant among other L. monocytogenes strains available in our laboratory and to assess the acid tolerance of stationary-phase cells used to inoculate meat washings (day 0). Screening included strain Scott A, as well as the parental strain N-7144 to ensure that resistance to streptomycin did not alter the acid tolerance of the Sm+ derivative. These experiments were also useful in the selection of the most appropriate challenge media for later use, as discussed in Results. More specifically, the screening was conducted with TSBYE acidified to pH 3.5 or 2.5 with lactic or acetic acid. Lactic and acetic acids were selected as acidulants because of their commercial application for meat decontamination. Selection of challenge pH 3.5 or 2.5 was based on the pH range of organic acid solutions applied in meat decontamination. It was also based on the pH of most challenge media reported in the literature for L. monocytogenes and other food pathogens (14, 29, 30, 31, 52), and for L. monocytogenes with lactic acid (TSBYE, pH 3.5) in particular (37). The organic acid reagents used for preparation of the challenge media (the same as those used to prepare the 2% acid solutions for meat decontamination) were added to TSBYE (pH 7.2) undiluted before distribution in tubes and autoclaving. Challenge media were checked after sterilization to ensure that their pH was within 0.05 unit of the desired value.

Following growth of L. monocytogenes strains in 10 ml of TSBYE for 24 h at 35°C, 1 ml containing approximately 10⁶ cells suspended in 0.1% buffered peptone water of each strain was pipetted into 9 ml of TSBYE acidified to pH 3.5 or 2.5 with lactic or acetic acid. This gave a concentration of approximately 10⁵ cells/ml exposed to acid (i.e., the inoculation level in meat washings). Addition of cells to TSBYE with no pH adjustment (pH 7.2) served as a control. During acid challenging, tubes were kept at 25°C. Samples (1 ml) were periodically (0, 30, 60, 90, and 120 min) taken from each tube, serially diluted in 9 ml of 0.1% buffered peptone water, and plated on TSAYE or PALCAM plates to determine the number of survivors and potential differences in recovery on PALCAM due to acid injury. Strain N-7144Sm+ was also plated on TSAYE-Sm to evaluate the effect of the antibiotic on pathogen recovery. Colonies on plates were counted after incubation at 35°C for 48 h. To account for potentially higher populations of L. monocytogenes grown in the washings and then acid challenged, the selected strains were also exposed (10⁸ CFU/ml) to TSBYE acidified (pH 3.5) with lactic acid and to unfiltered 2% lactic acid (pH 2.5) and 2% acetic acid (pH 3.2) washings of meat. The acid solution washings were used as challenge media since bacterial pathogens may be transferred to such waste fluids following decontamination of meat. Survivors were determined as described above.

In experiments with meat washings, the ATR of inoculated L. monocytogenes was comparatively determined in the absence (i.e., filter-sterilized washings) or presence (i.e., unfiltered washings) of natural flora after 1 and 8 days of storage at 35°C. The acid challenge procedure was as described above. One milliliter of each washing culture was pipetted into each of four tubes containing 9 ml of a selected acid challenge medium. Selected challenge media were TSBYE adjusted to pH 3.5 with lactic acid, unfiltered 2% lactic acid (pH 2.5) or 2% acetic acid (pH 3.2) washings, and TSBYE (pH 7.2), which served as a control. TSBYE adjusted to pH 3.5 with acetic acid, or to pH 2.5 with either lactic or acetic acid, was not used, for reasons presented in Results. Portions (1 ml) were taken from each challenge medium at 0, 30, 60, 90, and 120 min after inoculation and analyzed on TSAYE and PALCAM, as well as on TSAYE-Sm when strain N-7144Sm+ was used, as for pure cultures. Colonies (10%) on TSAYE plates from unfiltered samples were tentatively characterized as described above to determine the overall composition of the natural flora which occurred in coculture with L. monocytogenes and survived acid challenging. Especially for presumptive colonies of LAB (i.e., white, gram positive, catalase negative), representative cultures were isolated in MRS broth and then tested for growth on Enterococcus agar (Difco). All experiments were done in triplicate, by inoculating either different batches of washings with strain N-7155 or the same batch of washings with strains N-7155 and N-7144Sm+. Microbiological counts were converted to log CFU per milliliter, and the means and standard deviations (SD) were calculated.

RESULTS AND DISCUSSION

Microbial contamination of meat washings.

Cold (10°C) and hot (85°C) water meat washing fluids had average total microbial (TSAYE) counts after preparation of 3.9 ± 0.6 and 4.9 ± 0.6 log CFU/ml, respectively. The higher natural flora counts of hot compared to cold water washings may have been due to hot water having removed more bacteria in the washings without necessarily killing them. Filter-sterilized water washings were found to be free of background flora, and when incubated (35°C, 8 days) without inoculation, no microbial growth (<1.0 log CFU/ml) occurred (data not shown). The 2% lactate and 2% acetate solution unfiltered and uninoculated washings contained natural flora of <1.0 log CFU/ml, and no microbial growth (<1.0 log CFU/ml) occurred during storage at 35°C. Thus, the microbial populations of inoculated filter-sterilized water washings or filter-sterilized or unfiltered acid solution washings developing on TSAYE were only L. monocytogenes, unlike those of unfiltered water washings, in which the changes of the natural flora were followed in coculture with the inoculated strains.

Behavior of L. monocytogenes in meat washings with or without natural flora.

Inoculated L. monocytogenes died off (<1.0 log CFU/ml upon direct plating on PALCAM or TSAYE) within 24 h in all acid solution washings at 35°C, irrespective of acidulant present or filter sterilization (Table 1). In all experiments, the complete inactivation of the pathogen in acid solution washings was confirmed by negative culture enrichment (data not shown), which was done in parallel with direct plating on PALCAM to verify absence of the pathogen when counts were lower than 1.0 log CFU/ml. Additional new data from our laboratory (43) have confirmed that the 35°C incubation temperature had an accelerating effect on the pathogen inactivation rate in acid solution washings compared to lower (4 or 10°C) incubation temperatures. At 4 and 10°C, L. monocytogenes may survive for 2 and 4 days in 2% lactate or 2% acetate washings, respectively, while E. coli O157:H7 may survive longer than L. monocytogenes, as the former pathogen was countable after 7 days in 2% acetate washings stored at 4°C (43). Thus, compared to incubation at the optimum (35°C) growth temperature, purposefully used in this study, the generally lower temperatures encountered in meat plants may enhance survival of acid-adapted pathogens, which may serve as potential cross-contamination sources, if pathogen niches are established in the plant (43). It should be noted, however, that spots near motor and conveyor belts or other equipment in meat plants may reach temperatures higher than those found in most other areas of a plant.

TABLE 1.

Changes in populations of natural flora (TSAYE) and inoculated L. monocytogenes^a (PALCAM) in fresh beef decontamination fluids (spray-washings) stored at 35°C with or without previous filter sterilization^a

Presence of natural flora	Decontamination spray washingsb	Change (log CFU/ml ± SD; n = 3)
		Populations on TSAYE				Populations on PALCAM
		0c	1	4	8	0	1	4	8
No	Water (10°C)b	5.5 (0.1)	7.3 (0.1)	5.7 (0.7)	5.9 (0.7)	5.5 (0.1)	7.0 (0.5)	5.7 (0.6)	5.8 (0.7)
	Water (85°C)	5.4 (0.0)	6.7 (0.0)	5.9 (0.7)	5.8 (0.6)	5.4 (0.0)	6.3 (0.5)	5.5 (0.2)	5.3 (0.2)
	Lactic acid (2%, 55°C)	5.2 (0.1)	<1.0	<1.0	<1.0	4.9 (0.1)	<1.0	<1.0	<1.0
	Acetic acid (2%, 55°C)	5.4 (0.1)	<1.0	<1.0	<1.0	5.3 (0.1)	<1.0	<1.0	<1.0
Yes	Water (10°C)	5.4 (0.1)	8.5 (0.3)	8.6 (0.1)	8.5 (0.3)	5.4 (0.1)	7.2 (0.4)	6.9 (0.7)	6.8 (0.8)
	Water (85°C)	5.4 (0.1)	8.0 (0.2)	8.5 (0.3)	8.5 (0.3)	5.4 (0.1)	6.7 (0.5)	7.1 (0.4)	6.4 (0.4)
	Lactic acid (2%, 55°C)	5.1 (0.2)	<1.0	<1.0	<1.0	4.9 (0.3)	<1.0	<1.0	<1.0
	Acetic acid (2%, 55°C)	5.2 (0.1)	<1.0	<1.0	<1.0	5.1 (0.1)	<1.0	<1.0	<1.0

^aIn all three replicate experiments, filter-sterilized washings were inoculated with strain N-7155. Unfiltered washings were inoculated with strain N-7155 (two replicates, n = 2) or N-7144Sm+ (one replicate, n = 1). 

^bConditions of preparation of washings before use in the experiments. 

^cDays of storage at 35°C. 

In nonacid (water) washings, L. monocytogenes populations increased by approximately 0.9 to 1.8 log CFU/ml by day 1, irrespective of natural flora (Table 1). Then, pathogen numbers remained constant in unfiltered water washings or declined by approximately 1.0 log in filter-sterilized water washings. This decline may indicate that filter-sterilized water washings contained fewer nutrients than unfiltered washings following centrifugation and filtration. Overall, neither type (10 or 85°C) of water washings supported extensive pathogen growth, indicating that the nutrient requirements of L. monocytogenes may have been higher than the nutrient supply in the water washings of meat. This, however, was not the case with the natural microbial flora (TSAYE), which increased rapidly (>10⁸ CFU/ml), outgrew L. monocytogenes by at least 1 to 2 logs by day 1, and remained at high numbers throughout storage of unfiltered water washings at 35°C (Table 1). The behavior of the natural flora in the unfiltered water washings without inoculation with L. monocytogenes was very similar to that in the inoculated samples (data not shown). Colonies of natural flora on TSAYE plates of water washings stored for 1 to 8 days at 35°C were creamy or yellowish gram negative, while plates were practically free of L. monocytogenes colonies (flat, whitish, with an erased center). Preliminary characterization of this flora indicated that it was a mixture of oxidase-positive (i.e., Pseudomonas or related genera) and oxidase-negative (i.e., family Enterobacteriaceae) bacteria, with the former type being more numerous (60 to 75%) overall, especially in water (10°C) washings (75 to 85%). Other data from our laboratory showed that unfiltered water (10 to 85°C) meat washings inoculated with L. monocytogenes, E. coli O157:H7, or Salmonella and stored at 4 or 10°C were similarly dominated by gram-negative flora, which, however, was shifted to >90% oxidase positive (i.e., Pseudomonas-like bacteria) (43). Thus, at 35°C, the composition of the natural flora may have been more diverse between water treatments or replicates compared to 4 or 10°C (43), depending on the type of bacteria present in the water washings after spraying and the ability of most Enterobacteriaceae to compete better at high temperatures. Further studies are needed to characterize this background flora at the genus and species level and to evaluate its potential influence on pathogen behavior.

Several previous studies have shown that, depending on the food and its storage conditions, the background flora, composed mainly of Pseudomonas or LAB may inhibit (4, 10, 11, 35), stimulate (19, 33, 34), or have no effect (5, 34) on L. monocytogenes. Inhibition of L. monocytogenes in mixed cultures in broth was due to a lengthening of the lag phase, slowing of growth, and suppression of the maximum population density (MPD) (8, 9). The magnitude of the competitive or inhibitory effect of the background flora may be influenced by temperature, pH, and nutrient depletion (8, 9), while sometimes siderophores and bacteriocins produced by Pseudomonas (11) and LAB (24), respectively, may also enhance suppression of L. monocytogenes. In this study, there was no direct evidence that the background flora inhibited or stimulated growth of L. monocytogenes in unfiltered washings of meat at 35°C. In fact, L. monocytogenes growth seemed to be unaffected by the predominant gram-negative flora, as the MPD of the pathogen by day 1, in the presence or absence of competitors in water washings, was virtually the same (Table 1). However, the rapid and high (>8 logs) establishment of this type of natural flora in the washings during storage at 35°C appeared to induce important changes at the cellular level in L. monocytogenes that modulated the magnitude and timing of expression of ATR, as discussed below.

Effect of natural flora and L. monocytogenes on the pH of washings.

The initial pHs of the water washings were similar (6.0 to 6.1), regardless of filter sterilization or whether cold (10°C) or hot (85°C) water was used to generate them (Table 2). During storage at 35°C, however, major differences in pH were observed, which were dependent on the presence or absence of natural flora and inoculated L. monocytogenes. Following the increases (>10⁸ CFU/ml) in natural flora of the unfiltered water washings, their pH increased on average from 6.0 to 6.1 at inoculation (day 0) to 6.9 to 7.7 by day 8, irrespective of pathogen presence (Table 2). Interestingly, though, the pH of washings inoculated with L. monocytogenes was reduced 0.3 to 0.5 unit by day 1, while overall, this pH decrease did not occur in the corresponding uninoculated samples (Table 2). Probably, following its 5-log inoculation and approximate 1 to 2-log growth at 35°C within 24 h, L. monocytogenes may have utilized via fermentative pathways most of the available glucose in the washings (16) to cause the observed pH reduction. Since similar pH reductions were rare in unfiltered but uninoculated washings, it seems that the dominating gram-negative flora used alternative metabolic pathways to support its growth (13). Indeed, in all cases there was a pH reduction in the washings by day 1, which was followed by a shift toward high pH values and development of putrid off odors at later days of incubation at 35°C (i.e., as soon as by-products of the gram-negative flora accumulated) (Table 2). In our other study (43), L. monocytogenes inoculated in unfiltered water (10 or 85°C) washings at 10⁵ cells/ml increased by 0.6 to 1.3 logs at 4 or 10°C but caused no reductions in pH. These findings suggest that L. monocytogenes maximized its growth and competition for glucose uptake at 35°C (Table 1), while at 4 or 10°C (43) it grew more slowly than the better-adapted meat spoilage flora, which through its higher metabolic activity counteracted pH reductions in the washings. In accordance, the pH of inoculated filter-sterilized water washings was also reduced 0.4 to 0.7 unit at day 1, as a result of L. monocytogenes growth, and finally reached values of approximately 5.0 (Table 2), evidently because there was no natural flora to counteract the pH reduction through its own metabolic activity. Also, the pH of filter-sterilized but uninoculated washings did not change during storage (Table 2), confirming the observation that pH reductions in the inoculated samples were due to L. monocytogenes growth. Meanwhile, the pH of acid solution washings was practically unchanged during storage, irrespective of filter sterilization or pathogen inoculation (Table 2), reflecting the absence of natural flora and the rapid inactivation of the inoculated pathogen at 35°C (Table 1).

TABLE 2.

Changes in pH of fresh beef decontamination fluids (spray washings) inoculated with L. monocytogenes^a with or without previous filter sterilization and stored at 35°C

Presence of natural flora	Decontamination spray washings	Change in pH (mean ± SD; n = 3)
		Inoculated samples				Uninoculated samples
		0b	1	4	8	0	1	4	8
No	Water (10°C)	6.1 (0.1)	5.4 (0.1)	5.3 (0.1)	5.3 (0.1)	6.1 (0.1)	6.1 (0.1)	6.0 (0.1)	6.1 (0.0)
	Water (85°C)	6.0 (0.1)	5.6 (0.1)	5.1 (0.1)	5.1 (0.1)	6.1 (0.1)	6.1 (0.1)	6.1 (0.1)	6.0 (0.1)
	Lactic acid (2%, 55°C)	2.5 (0.1)	2.5 (0.1)	2.5 (0.1)	2.5 (0.1)	2.5 (0.1)	2.4 (0.1)	2.4 (0.1)	2.5 (0.1)
	Acetic acid (2%, 55°C)	3.3 (0.1)	3.3 (0.1)	3.2 (0.1)	3.2 (0.1)	3.2 (0.1)	3.3 (0.0)	3.2 (0.1)	3.2 (0.0)
Yes	Water (10°C)	6.1 (0.0)	5.8 (0.1)	6.6 (0.1)	7.3 (0.7)	6.1 (0.1)	6.8 (0.2)	7.4 (0.0)	7.7 (0.2)
	Water (85°C)	6.1 (0.0)	5.6 (0.2)	6.6 (0.1)	6.9 (0.1)	6.1 (0.1)	6.4 (0.6)	7.0 (0.2)	7.0 (0.2)
	Lactic acid (2%, 55°C)	2.3 (0.0)	2.4 (0.0)	2.3 (0.0)	2.4 (0.0)	2.3 (0.0)	2.3 (0.1)	2.4 (0.0)	2.4 (0.0)
	Acetic acid (2%, 55°C)	3.2 (0.0)	3.2 (0.0)	3.1 (0.0)	3.2 (0.0)	3.2 (0.0)	3.2 (0.0)	3.2 (0.1)	3.2 (0.1)

^aIn all three replicate experiments, filter-sterilized washings were inoculated with strain N-7155. Unfiltered washings were inoculated with strain N-7155 (two replicates, n = 2) or N-7144Sm+ (one replicate, n = 1). 

^bDays of storage at 35°C. 

Acid tolerance of L. monocytogenes inoculum.

Stationary-phase (grown in TSBYE at 35°C for 24 h) cells of L. monocytogenes were resistant to acid, but not under all conditions tested. While TSBYE acidified to pH 2.5 with lactic or acetic acid, or to pH 3.5 with acetic acid, completely inactivated (<1.0 log CFU/ml) all strains tested within 0 to 120 min (data not shown), most strains displayed a profound survival in TSBYE acidified to pH 3.5 with lactic acid. Specifically, the population reductions on TSAYE of strains N-7155, N-7144Sm+, and N-7144 after 2-h exposure to lactic acid (pH 3.5) were on average 1.2, 1.2, and 1.1 logs, respectively (Fig. 1A). The respective reductions on PALCAM were 3.2, 2.9, and 2.9 logs, indicating extensive acid injury. The survival of strain N-7144Sm+ on TSAYE-Sm, however, was similar to that on TSAYE (Fig. 1A), indicating that streptomycin was not inhibitory to the acid-injured cells. The acid tolerance of strains N-7155 and N-7144 was remarkably higher than that of strain Scott A (Fig. 1A) and other L. monocytogenes strains (data not shown), while that of strain N-7144Sm+ was similar to that of its parental N-7144 strain (Fig. 1A); thus, these strains were selected for further study. Notably, these findings suggest that strain Scott A, which is commonly used in food research (14, 30, 31, 53), may not be the most suitable for studies dealing with L. monocytogenes survival in acid foods. When 10⁸ TSBYE cells (35°C, 24 h) of strain N-7155 were exposed to TSBYE acidified (pH 3.5) with lactic acid and to the unfiltered 2% acetic (pH 3.2) or 2% lactic acid washings, the pathogen survived much better in the former two challenge media (Fig. 1B). Strain N-7144Sm+ showed similar results (data not shown).

FIG. 1.

(A) Comparative survival of L. monocytogenes N-7155, N-7144, N-7144Sm+_n and Scott A on TSAYE, PALCAM, and TSAYESm (strain N-7144Sm+ only) after exposure to TSBYE acidified with lactic acid (pH 3.5). Cells were challenged to acid after growth as pure cultures in TSBYE at 35°C for 24 h. Cell numbers exposed to acid at time zero were determined in TSBYE (pH 7.2, control), where no population reductions occurred within 120 min (data not shown for graph clarity). Values are the mean of three replicates (SD range, 0.1 to 0.5). (B) Comparative survival of high levels (10⁸ CFU/ml) of L. monocytogenes N-7155 on TSAYE or PALCAM after exposure to TSBYE acidified with lactic acid (pH 3.5) and to unfiltered 2% acetate (pH 3.2) or 2% lactate (pH 2.5) spray washings (AA-SW or LA-SW) of meat. Cells were challenged to acid after growth as pure cultures in TSBYE at 35°C for 24 h. Cell numbers exposed to acid at time zero were determined as above. Values are from one representative experiment.

These findings confirm the major differences in the buffering capacity of the challenge media used and suggest differences in the acidifying capacity and bactericidal mode of action of each acid, which depend on both the acid ionization constant (solution pH) and concentration in solution (53). Indeed, TSBYE (100 ml) required an average addition of 1.3 or 8.6 and 4.7 or 42.8 ml of lactic or acetic acid, respectively, to adjust the pH to 3.5 and 2.5, respectively. Conversely, 100 ml of 2% (wt/vol) acidic meat washings containing approximately 2 ml of each of lactic (85% [wt/wt]; density, 1.2) or acetic (100% glacial) displayed a pH of 2.5 or 3.2 after spraying, respectively. These results confirm that acetate is a weaker acid than lactate and that TSBYE has a higher buffering capacity compared to meat washings. TSBYE acidified to pH 2.5 with acetic acid was of immediate lethality (<1.0 log CFU/ml at time zero on all recovery media) to all L. monocytogenes strains, while TSBYE acidified (pH 2.5) with lactic acid resulted in no survivors (<1.0 log CFU/ml) at 60 min (data not shown). In contrast, the 2% lactate washings of pH 2.5 permitted survival for 30 to 60 min (Fig. 1B). Accordingly, TSBYE acidified to pH 3.5 with acetate was more lethal to L. monocytogenes than 2% acetate washings, although of higher pH. Thus, the concentration of acetic acid rather than its pH in solution was more decisive for L. monocytogenes inactivation, a result consistent with evidence that acetate kills the pathogen primarily by lowering the intracellular pH following its penetration through the membrane and dissociation inside the cell (53). In contrast, the 2% lactate washings (pH 2.5) were consistently more lethal to L. monocytogenes than acidified TSBYE (pH 3.5) (Fig. 1B), having a higher acid concentration and lower pH in solution. In conclusion, the stationary cells of strains N-7155 and N-7144Sm+ were acid tolerant at inoculation of meat washings (day 0), while TSBYE acidified (pH 3.5) with lactic acid and the original 2% lactate (pH 2.5) or acetate (pH 3.2) solution washings were the most suitable challenge media for use in the experiments.

Effect of natural flora on ATR of L. monocytogenes in water washings.

The acid tolerance of L. monocytogenes was evaluated in unfiltered or filter-sterilized inoculated water (10 or 85°C) washings but not in acid solution washings because the latter did not allow survival of the inoculum at day 1 or 8 at 35°C. Changes in ATR of the pathogen in the water washings during storage were dependent on the presence or absence of natural flora and associated pH changes. In filter-sterilized water (10 or 85°C) washings, L. monocytogenes N-7155 (PALCAM) increased its acid tolerance by 1.0 to 2.9 logs from days 1 to 8 (Fig. 2). Consistent with previous findings (Fig. 1B), the pathogen survived better in acidified TSBYE (pH 3.5) with lactic acid and 2% acetate washings (pH 3.2) than in 2% lactate washings (pH 2.5) and better on TSAYE (Fig. 3) than on PALCAM (Fig. 2). Importantly though, following its preincubation and growth in the water washings, the recovery of strain N-7155 on PALCAM (Fig. 2) was slightly lower than that on TSAYE (Fig. 3), which was not the case in the cells from TSBYE cultures used as inocula (Fig. 1A). This finding may indicate that the oligotrophic environment of the washings enhanced the acid tolerance of L. monocytogenes due to starvation (30, 31), and thus the proportion of acid-injured cells that were unable to grow on PALCAM to the total surviving populations was reduced.

FIG. 2.

Acid survivors of L. monocytogenes N-7155 (PALCAM) after exposure to TSBYE acidified with lactic acid (pH 3.5), 2% acetate solution spray washings of meat (pH 3.2) (AA-SW), or 2% lactate solution spray washings of meat (pH 2.5) (LA-SW). Cells were challenged as for Fig. 1 after 1 or 8 days of growth in filter-sterilized spray washings of meat previously decontaminated with cold (10°C; A) or hot (85°C; B) water. Cell numbers exposed to acid at time zero were determined as for Fig. 1. Values are the mean of three replicates (SD range, 0.1 to 0.8).

FIG. 3.

Acid survivors of L. monocytogenes N-7155 (TSAYE) after exposure to TSBYE acidified with lactic acid (pH 3.5), 2% acetate solution spray washings of meat (pH 3.2) (AA-SW), or 2% lactate solution spray washings of meat (pH 2.5) (LA-SW). Cells were challenged as for Fig. 1 after 1 or 8 days of growth in filter-sterilized washings of meat previously decontaminated with cold (10°C; A) or hot (85°C; B) water. Cell numbers exposed to acid at time zero were determined as for Fig. 1. Values are the mean of three replicates (SD range, 0.1 to 0.9).

An opposite trend was observed for L. monocytogenes grown in unfiltered water washings, where the pathogen (PALCAM) was appreciably more acid tolerant by day 1 than by day 8 (Fig. 4). Specifically, when L. monocytogenes was exposed to acidified TSBYE (pH 3.5) with lactic acid and 2% acetate washings (pH 3.2) at day 1, its average population reductions ranged from 0.3 to 1.4 logs. The respective reductions at day 8 ranged from 3.0 to >5.0 log CFU/ml (Fig. 4). These results indicated that L. monocytogenes was sensitized to acid by approximately 1.5 to 4.5 logs from days 1 to 8 at 35°C in the presence of natural flora. Notably, sensitization was greater with acetic acid at pH 3.2 (Fig. 4), which was not the case either with the inoculum (Fig. 1B) or with stationary cells in filter-sterilized washings by day 8 (Fig. 2). Lactate (2%) washings (pH 2.5) were again the most lethal challenge medium (Fig. 4). Inoculation of the same batch of unfiltered washings with strain N-7144Sm+ indicated that acid sensitization of L. monocytogenes was neither strain dependent nor overestimated due to potentially low recovery on PALCAM of viable but acid-injured cells (Fig. 5). Indeed, although populations on PALCAM were constantly lower than those on TSAYE-Sm, the trend of L. monocytogenes N-7144Sm+ to become acid sensitive following an 8-day exposure to unfiltered water washings was unchanged (Fig. 5).

FIG. 4.

Acid survivors of L. monocytogenes (PALCAM) after exposure to TSBYE acidified with lactic acid (pH 3.5), 2% acetate solution spray washings of meat (pH 3.2) (AA-SW), or 2% lactate solution spray washings of meat (pH 2.5) (LA-SW). Cells were challenged as for Fig. 1 after 1 or 8 days of growth in unfiltered spray washings of meat previously decontaminated with cold (10°C; A) or hot (85°C; B) water. Cell numbers exposed to acid at time zero were determined as for Fig. 1. Values are the mean of three replicates (SD range, 0.1 to 1.5).

FIG. 5.

Comparative survival of L. monocytogenes N-7144Sm+ on TSAYE-Sm (solid lines) or PALCAM (dotted lines) after exposure to TSBYE acidified with lactic acid (pH 3.5), 2% acetate solution spray washings of meat (pH 3.2) (AA-SW) or 2% lactate solution spray washings of meat (pH 2.5) (LA-SW). Cells were challenged as for Fig. 1 after 1 or 8 days of growth in unfiltered spray washings of meat previously decontaminated with cold (10°C; A) or hot (85°C; B) water. Cell numbers exposed to acid at time zero were determined as for Fig. 1A. Values are from one representative experiment.

The predominant, gram-negative flora of the water washings was very sensitive to acid (Fig. 6), a result in agreement with previous data (23). At day 1, colonies that survived on TSAYE plates after 120 min of exposure to acid (Fig. 6) were mainly L. monocytogenes and were slightly more numerous than colonies on PALCAM (Fig. 4). At day 8, however, following acid sensitization of L. monocytogenes (Fig. 4 and 5), numerous (10⁵ CFU/ml) white colonies other than those of the pathogen developed on TSAYE plates from all samples (Fig. 6). Evidently, these indigenous microorganisms increased in relatively high numbers during storage and, despite neutral pH conditions in the unfiltered washings by day 8 (Table 2), maintained a high ATR and survived even in 2% lactate washings (pH 2.5) throughout the acid-challenging period (Fig. 6). Based on preliminary characterization tests, these organisms were LAB because they were gram-positive, catalase-negative nonsporeformers and were able to grow and acidify (pH <5.0; 30°C, 24 h) MRS broth. The profound acid resistance of LAB compared to food spoilage gram-negative bacteria is well established (38). More than 80% of those isolates were cocci but not enterococci, because they did not grow on Enterococcus agar. Additional tests are required to better characterize these isolates and evaluate their influence on the behavior of L. monocytogenes under these conditions.

FIG. 6.

Acid survivors of natural flora (TSAYE) after exposure to TSBYE acidified with lactic acid (pH 3.5), 2% acetate solution spray washings of meat (pH 3.2) (AA-SW), or 2% lactate solution spray-washings of meat (pH 2.5) (LA-SW). Cells were challenged as for Fig. 1 after 1 or 8 days of growth in unfiltered spray washings of meat previously decontaminated with cold (10°C; A) or hot (85°C; B) water. Cell numbers exposed to acid at time zero were determined as for Fig. 1A. Values are the mean of three replicates (SD range, 0.1 to 1.8).

This study evaluated the potential of the natural flora to modulate the ATR of L. monocytogenes in a food environment. The results have potential implications for meat plant environments where the pathogen may be found in association with natural microbial flora in residues of meat spray-washing solutions. Establishment of such niches in a meat plant may have potential implications for food safety through product cross-contamination. Few studies have attempted to correlate pathogen response to stresses with the natural flora in foods. Duffy et al. (17) reported that the addition of 10⁸ CFU of viable competitors per ml protected an underlying, exponentially growing population (10⁵ CFU/ml) of Salmonella serovar Typhimurium against thermal inactivation, as the pathogen D values at 55°C increased from 0.43 to 2.09. Furthermore, Aldsworth et al. (2) showed that the presence of 10⁸, but not 10⁵ to 10⁷, CFU of mixed viable competitors per ml also protected Salmonella serovar Typhimurium (10⁵ CFU/ml) from freezing, but addition of equal populations of heat-killed competitors did not provide any protective effect. Thus, the observed protective effect to food-related stresses, such as heat and freezing, was associated with the metabolic activity of the competitive flora; most importantly, it did not correlate with the stationary-phase stress adaptation, because it was essentially instantaneous after mixing of Salmonella serovar Typhimurium cells with the competitors (2). It was also found that by rapidly reducing the levels of dissolved oxygen through active respiration, high levels of competitive flora reduced oxidative damage to exponential-phase cells of Salmonella serovar Typhimurium and decreased the RpoS induction time, thus arresting pathogen growth and conferring protection from stress (2). On this basis, a novel hypothesis has been advanced to explain the increased sensitivity to stress of exponential-phase compared to stationary-phase bacterial cultures, the suicide response (2, 3). According to this hypothesis, pure cultures of rapidly growing and respiring bacterial cells exposed to mild stresses will suffer growth arrest but their metabolism will continue to result in a burst of free-radical production that is lethal to the cells, rather than the stress itself (3). This condition compares favorably with another recent study (18) showing that E. coli O157:H7 was inhibited to a lesser extent in coculture with a competitive flora than in pure culture in a simulated fermentation broth (pH 5.8 to 4.8) at 37°C. The authors also suggested that the increased numbers of the competitive flora protected the pathogen from acid (18); however, the ATR of E. coli O157:H7 in the absence versus the presence of competitors was not examined. Furthermore, none of the above studies investigated whether the protective effect of the competitive flora on bacterial pathogens will continue to exist upon extended periods of coexistence of growth-arrested pathogenic cells with high competitor populations. This is a condition that may apply in meat environments, where mixed microbial associations may be present (1, 5, 13, 27, 32, 35, 40, 43, 47).

In this study, L. monocytogenes developed its acid tolerance in filter-sterilized nonacid (water) meat washings in the same manner, but more slowly than in pure broth culture (14, 37). By day 1 at 35°C, the pathogen in the oligotrophic environment of these washings seemed to be in a slow exponential phase and was thus of increased acid sensitivity. By day 8, however, the pathogen had entered into the stationary phase in filter-sterilized washings. Thus, surviving cells, although lower in numbers than at day 1 (Fig. 2 and 3), were of higher acid resistance, probably due to adaptation to starving and low pH (5.0 to 5.3) conditions (30) as well as to their stationary-phase-induced acid resistance, which is regulated by RpoS and is pH independent (14).

The reversal of ATR of L. monocytogenes in unfiltered water washings at 35°C (Fig. 4 and 5) may be an important observation. The pathogen was very acid tolerant at day 1, indicating that cells were protected to acid earlier than those in filter-sterilized washings, even though there were no major differences in MPD of the two cultures by day 1 at 35°C (Table 1). In fact, the pathogen at day 1 in the unfiltered washings was more acid tolerant on PALCAM (Fig. 4) than it was the inoculum culture at day 0 on PALCAM, or even on TSAYE (Fig. 1A). Apart from the potential starvation, this accelerated ATR of L. monocytogenes in unfiltered washings appears to be in complete agreement with studies discussed (2, 3, 17), indicating that high numbers of competitive flora increase resistance of bacterial pathogens to stress by arresting growth and accelerating entry into stationary phase. In agreement with these studies (2, 17), the induction of the acid-protective responses in L. monocytogenes may have occurred soon after the competitive flora reached 10⁸ CFU/ml in the washings (i.e., within the first 24 h of incubation at 35°C). However, there are some remarkable differences between the experiments of this study and previous studies (2, 17). In our study, the competitive flora was not mixed instantaneously at high numbers (8 logs) with lower (5 logs) numbers of L. monocytogenes in broth, but it initiated growth from populations as low as 3 to 5 logs, under competition of a 5-log L. monocytogenes inoculum in the washings. Also, the mixed cultures were incubated statically at 35°C, thus under reduced but not limited oxygen availability. In essence, although during the first incubation hours at 35°C L. monocytogenes might have metabolized more actively, the rapidly growing gram-negative flora would have steadily increased competition for oxygen to support cell respiration. This may have shifted the metabolism of L. monocytogenes, which is a typical microaerophilic organism though capable of increasing faster and more extensively under aeration (31), to oxygen-independent pathways (16, 31). Thus, the possibility that the metabolic activity of L. monocytogenes was accelerated to compete against its natural competitors for glucose uptake via fermenting pathways cannot be ruled out, given that experiments were done near the optimum temperature (35°C) of the pathogen (31). In other words, the lag phase of L. monocytogenes might have been shortened and its exponential growth might have been accelerated, strictly depending on, and at the expense of, available glucose (16) in the washings, while the pathogen was attempting to compete against its competitors. As mentioned, this may have caused the observed decrease in pH of inoculated unfiltered washings at day 1 at 35°C, which was uncommon in the corresponding samples without inoculation (Table 2). Probably, L. monocytogenes converted some glucose to lactate (16), whereas, depending on the proportional distribution of pseudomonads or enterobacteria in each occasion, the gram-negative flora may have also converted glucose to gluconate and oxygluconate (13) or lactate and acetate (13), respectively, before they started to attack meat proteins (13, 47).

An important finding of this study appears to be the inability of L. monocytogenes to maintain its high ATR (day 1), or develop a high stationary-phase acid resistance, in unfiltered nonacid (water) washings at day 8 (Fig. 4 and 5). As mentioned, previous studies did not address the potential changes in stress resistance of stationary-phase populations of bacterial pathogens underlying much higher populations of spoilage bacteria in foods or their processing plants for extended times. The observed acid sensitization may have been due to a shift in the intracellular activities and catabolism of L. monocytogenes to adapt and survive under the conditions created in the water washings by the gram-negative spoilage flora. The presence of meat particles in the washings did not seem to offer any protection for L. monocytogenes to subsequent acid exposure, although such a protective effect has been seen in enteric pathogens (52). Thus, the findings of this study reveal that L. monocytogenes may shift from acid resistant to acid sensitive following exposure in food environments of neutral pH, such as in nonacid (water) meat decontamination wastes. Research is in progress to determine if L. monocytogenes and other pathogens, such as E. coli O157:H7 and Salmonella serovar Typhimurium DT 104, are also sensitized to acid following growth in pure compared to mixed cultures with various spoilage bacteria, and in poor substrates compared to complex media, at temperatures of 4 to 15°C to simulate more prevalent plant conditions. Results have shown that acid sensitization of bacterial pathogens occurs following exposure to water meat washings stored at 10°C, with E. coli O157:H7 and Salmonella serovar Typhimurium DT 104 being sensitized more than L. monocytogenes, especially if previously acid adapted (J. Samelis, J. N. Sofos, P. A. Kendall, and G. C. Smith, unpublished data).

The findings of this study may be of practical importance to the meat industry. Since bacterial pathogens seem to be acid sensitized under the stressful conditions prevailing in nonacid meat washings, the use of water, steam, or other nonacid treatments without or with reduced use of acidic meat decontamination technologies may be advantageous to meat safety. This approach could potentially minimize the risk of establishment of acid-adapted pathogens in the plant, while the acid-sensitive pathogens surviving on decontaminated meat or transferred to nonacid wastes may be easily inactivated in subsequent processing steps. However, whether these potential benefits of nonacid decontamination technologies offset any potential benefits due to residual organic acid antimicrobial effects on treated carcasses and the resulting meat cuts (44, 45) requires further research.

In summary, the present findings suggest that depending on its composition, growth rate, and metabolic activity, the natural microbial flora of a food environment may protect or sensitize pathogens to acid stress and thereby affect food safety. Further research is needed to study competitive interactions of bacterial pathogens with different types of spoilage bacteria under various sets of controlled environmental conditions in order to better understand and monitor microbial competition as a component of food safety systems.