ph-dependent Stationary-Phase Acid Resistance Response of Enterohemorrhagic Escherichia coli in the Presence of Various Acidulants

Transcription

1 211 Journal of Food Protection, Vol. 62, No. 3, 1999, Pages ph-dependent Stationary-Phase Acid Resistance Response of Enterohemorrhagic Escherichia coli in the Presence of Various Acidulants ROBERT L. BUCHANAN* AND SHARON G. EDELSON Food Safety Research Unit, U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, 600 East Mermaid Lane, Wyndmoor, Pennsylvania 19038, USA MS 98-48: Received 23 February 1998/Accepted 31 July 1998 ABSTRACT The effect of acidulant identity on the ph-dependent stationary-phase acid resistance response of enterohemorrhagic Escherichia coli was studied. Nine strains of E. coli (seven O157:H7, one O111:H-, and one biotype 1 reference strain) were cultured individually for 18 h at 37 C in tryptic soy broth (TSB) plus 1% dextrose and in TSB without dextrose to yield acid resistance induced and noninduced stationary-phase cells, respectively. These cultures were then inoculated into brain heart infusion broth (BHI) supplemented with 0.5% citric, malic, lactic, or acetic acid and adjusted to ph 3.0 with HCl. The BHI tubes were incubated at 37 C for up to 7 h and samples were removed after 0, 2, 5, and 7 h and plated for counting CFU on BHI agar and MacConkey agar (MA). The results were compared to data previously obtained with HCl only. Acid resistance varied substantially among the isolates, being dependent on the strain, the acidulant, and the induction of ph-dependent acid resistance. Hydrochloric acid was consistently the least damaging to cells; lactic acid was the most detrimental. The relative activity of the other acids was strain dependent. Inducing ph-dependent acid resistance increased the already substantial acid tolerance of stationary-phase E. coli. The extent of injury also varied with acid and strain, with as much as a 5-log-cycle differential between BHI agar and MA CFU counts. The accurate determination of the survival of enterohemorrhagic E. coli in acidic foods must take into account the biological variability of the microorganism with respect to its acid resistance and its ability to enhance survival through the induction of physiological stress responses. Acid resistance and acid tolerance are important virulence determinants that contribute to the survival and pathogenicity of infectious foodborne pathogens such as enterohemorrhagic Escherichia coli (EHEC), Salmonella, Shigella spp., and Listeria monocytogenes to cause disease. While there is no universally accepted definition, the term acid resistance (acid habituation) is generally used to refer to the extended exposure of a microorganism to moderately acidic conditions (e.g., ph 5.0) leading to it being able to withstand ph values of 2.5 (16). Conversely, acid tolerance refers to the enhanced survival of a microorganism exposed to ph values between 2.5 and 4.0 after a brief exposure to moderately acidic conditions. In addition to increasing the likelihood that these organisms will survive in acidic foods for extended periods (12, 17, 25, 29), acid resistance increases the portion of the population that survive the gastric environment (16, 21, 33) and further appears to increase infectivity once the pathogen attaches to the intestinal tract (14, 32, 39). In addition, induction of acid resistance or acid tolerance can increase bacterial re- * Author for correspondence. Present address: U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, Washington, D.C. Tel: ; Fax: ; rbuchana@bangate.fda.gov. Mention of brand or firm name does not constitute an endorsement by the U.S. Department of Agriculture above others of a similar nature not mentioned. sistance to other stresses such as heat, ionizing and nonionizing irradiation, and antimicrobial agents (7, 15, 20, 24, 27). Acid resistance has also been observed in a variety of other species of Enterobacteriaceae (31). Several physiological systems contribute to the acid tolerance and acid resistance of species of Enterobacteriaceae (4, 16). Maximal survival of E. coli, including EHEC, in strongly acidic environments appears to be associated with stationaryphase acid habituation, where both the constitutive rpos gene-regulated and inducible ph-dependent acid resistance are active (3, 4, 6, 13, 19, 25, 34). Survival of foodborne pathogens in acidic environments is influenced by a number of environmental and cultural factors (10, 12, 37, 38). One such factor is the identity of the acidulant used to modify the ph. However, most studies on acid resistance and/or tolerance have employed HCl, which is considered one of the gentler acidulants. Weak organic acids often have additional antimicrobial activity associated with their anions (10). Accordingly, the objectives of the present study was to assess the ability of EHEC cells to survive exposure to a ph 3.0 medium that contained one of four organic acids (citric, malic, lactic, or acetic) used commonly in foods and to compare the results with those acquired earlier by using HCl only. Prior growth in acidogenic and nonacidogenic media was used to assess the relative contributions of the ph-dependent and ph-independent stationary phase-acid resistance, respectively.

2 212 E. COLI ACID RESISTANCE J. Food Prot., Vol. 62, No. 3 TABLE 1. Escherichia coli strains evaluated Acid tolerance a when grown in: Strain Serotype Source S1, B1409 O157:H7 Clinical isolate, CDC Extreme Strong S2, O157:H7 Isolated from beef kidney, FSIS Extreme Strong S3, 30-2C4 O157:H7 Clinical isolate from dry-cured salami outbreak, Extreme Extreme CDC via FSIS S4, 932 O157:H7 Clinical isolate, CDC via M. P. Doyle Extreme Strong S5, Ent-C9490 O157:H7 Clinical isolate from Jack-in-the-Box outbreak, Extreme Extreme CDC S6, A9124-C1 O157:H7 Clinical isolate, CDC Extreme Moderate S7, 95JB1 O111:H- Clinical isolate from mettwurst outbreak, Adelaide Extreme Extreme Children s Hospital (Australia) via J. Paton S8, ATCC Not available Commercially available reference strain, ATCC via Strong Weak Difco S9, SEA13B88 O157:H7 Fresh apple juice isolate from Odwalla outbreak, FDA Extreme b Extreme b a Based on survival at 37 C in BHI adjusted to ph 3.0 using HCI (6). Extreme, 1-log reduction after 7 h; strong, 1- to 2-log reduction after 7 h; moderate, 2- to 4-log reduction after 7 h; weak, 3-log reduction after 2 h. b Same criteria as in footnote a but from current study. MATERIALS AND METHODS With the exception of the acid challenge medium, the study was performed by using techniques described previously (6) so that the results could be compared. These methods are outlined briefly below. Microorganisms. The seven EHEC and one non-ehec strains characterized for acid resistance earlier (11) using HCladjusted brain heart infusion broth (BHI; Difco Laboratories, Detroit, Mich.) were used in the present study (Table 1). E. coli O157:H7 SEA13B88 was also evaluated. Stock cultures were maintained in tryptic soy broth (TSB, Difco), stored at 2 C, and transferred bimonthly. For convenience, the strains will be referred to as S1 through S9 (Table 1). Preparation of inocula. E. coli strains were cultured individually in 125-ml Erlenmeyer flasks containing 50 ml of either TSB with 1% (wt/vol) dextrose () or TSB without dextrose (, Difco). The cultures were incubated without agitation at 37 C for 18 h to yield stationary-phase cells (10 8 to 10 9 CFU/ml). The final ph values of the acidogenic and nonacidogenic cultures were approximately 4.6 and 7.1, respectively, thus providing cells that were induced and not induced for ph-dependent, stationary phase acid resistance. Preparation of acid challenge media. Double-strength BHI was prepared in 250-ml portions and sterilized by autoclaving. The portions were then supplemented with 2.5 g of citric, malic, lactic, or acetic acid, 220 ml of sterile deionized distilled water was added, and the ph was adjusted to 3.0 with HCl. Sterile deionized water was added to bring the volume to 500 ml and achieve a final acidulant concentration of 5 g/liter (0.5% wt/vol). This concentration was selected on the basis of its common use in foods. Five grams per liter is equivalent to molar concentrations of 23.8, 37.3, 55.5, and 83.3 mm of citric, malic, lactic, and acetic acid, respectively. At ph 3.0, the calculated molar concentrations of fully undissociated acid are 13.6, 26.7, 48.4, and 81.8 mm, respectively. The media were then dispensed in 10-ml portions into sterile test tubes (16 by 125 mm) and preequilibrated to 37 C. The medium for the initial characterization of strain S9 was prepared in the same manner except that no organic acid was added. Assessment of acid resistance. Twelve tubes for each acidulant-strain-growth medium combination were inoculated with 0.1-ml portions of either the - or -grown cultures to achieve an approximate initial population density of 10 7 CFU/ ml. All tubes were incubated at 37 C. After 0, 2, 5, and 7 h, three tubes were removed and assessed for viable CFU by surface plating appropriate dilutions on duplicate plates of both BHI agar TABLE 2. Inactivation of exponential and stationary-phase cells of Escherichia coli O157:H7 Ent-C9490 in brain heart infusion adjusted to ph 3.0 with hydrochloric acid and supplemented with 0.5% lactic, acetic, or citric acid a Growth phase, growth medium Mean population density decrease: log CFU/ml b HCl only Lactic acid Acetic acid Citric acid Exponential, 4.16 (0.35) 4.15 c ( 0.49) 4.52 ( 0.43) 4.47 (0.39) Stationary, 0.03 (0.09) 4.42 ( 1.20) 0.73 (0.38) 0.20 (0.28) Stationary, 0.00 (0.05) 1.99 (0.74) 0.05 (0.02) 0.18 (0.09) a E. coli O157:H7 was cultured in and for 18 h; then fresh was inoculated with the 18-h culture. After an additional 3 h of incubation the cultures were used for the 2-h acid challenge. b Log(viable count differential): mean log(n 0 ) log(n 2 ). Number in parentheses is standard deviation; n 4. c One or more of the cultures had fallen below the lower limit of detection after 2-h challenge.

3 J. Food Prot., Vol. 62, No. 3 BUCHANAN AND EDELSON 213 FIGURE 1. The inactivation of Escherichia coli O157:H7 S1 (B1409) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). (BHIA) and MacConkey agar (MA) with a Spiral Plater (model 3000, Spiral Biotech, Bethesda, Md.). All plates were incubated for24hat37 C and then enumerated with an automated colony counter (Spiral Laser Counter, Spiral Biotech). The BHIA counts were used to determine total counts and MA counts were used to estimate the number of noninjured cells. The extent of injury was then estimated from the differential in the number of CFU between the two media. The survival data were compared with those from the earlier study. A preliminary study on the effect of growth phase on acid resistance was conducted with E. coli S5 by a modification of the techniques described above. S5 cells were cultured at 37 C in test tubes (16 by 125 mm) containing 10 ml of or. After 18 h, 50 l of the culture was used to inoculate a fresh tube of, and all cultures were incubated for an additional 3hat37 C. The cultures were then inoculated into tubes containing BHI adjusted to ph 3.0 with HCl and containing 0.5% lactic, citric, or acetic acid or HCl only. The acid challenge was limited to a 2-h exposure at 37 C. FIGURE 2. The inactivation of Escherichia coli O157:H7 S2 ( ) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). RESULTS Acid resistance. Strain S5 was used to provide a preliminary assessment of the effect of growth phase and phdependent stationary-phase acid resistance on the survival of E. coli in BHI adjusted to ph 3.0 and containing 0.5% lactic, acetic, or citric acid (Table 2). Exponential-phase cells grown in were relatively sensitive to the 2-h 37 C acid challenge; the population density decreased by more than 4 log cycles with each of the acids. Stationaryphase cells were substantially more resistant; population density decreases were less than 1 log cycle with all acids except lactic. Inducing ph-dependent stationary-phase acid resistance by growing the organism in further increased survival; the decrease in population density after exposure to 0.5% lactic acid was less than 2 log cycles. The effect of prior growth of E. coli in acidogenic or nonacidogenic on the survival of cells subsequently exposed to a ph % acid (lactic, acetic, citric, or malic) challenge at 37 C was studied by individually challenging nine strains for 7 h (Figs. 1 to 9). Included are the data for ph 3.0 BHI adjusted with HCl only from our earlier characterization of strains S1 through S8 (6). The extent of survival varied substantially with the acidulant employed, the strain evaluated, and the prior induction of ph-dependent stationary-phase acid resistance. The least damaging acid of the five was consistently HCl, whereas lactic acid had the greatest antimicrobial activity at this ph. The relative effectiveness of the other acids varied among the strains and with prior growth conditions (Table 3). All nine strains of E. coli displayed some degree of increased acid resistance when cultured in acidogenic. With some strain-acidulant combinations, the differences in the survival of cells grown in and in cells during the 7-h exposure were minimal (e.g., S9 with HCl, citric, and malic acids, Fig. 9). With other combinations, the levels of -grown cells were essentially unchanged while the -grown cells had declined by 4 to 5 log cycles (e.g., S1 with malic acid, Fig. 1).

4 214 E. COLI ACID RESISTANCE J. Food Prot., Vol. 62, No. 3 FIGURE 3. The inactivation of Escherichia coli O157:H7 S3 (30-2C4) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). FIGURE 5. The inactivation of Escherichia coli O157:H7 S5 (Ent- C9490) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). Injury. Colonies were counted on two media (BHIA and MA) to determine the extent of injury. When physiologically stressed, E. coli loses its resistance to bile salts. Thus, strains were plated on BHIA to estimate total counts of CFU and on MA to provide the numbers of uninjured cells. The extent of injury varied with strains, prior growth conditions, acid identity, and duration of exposure. Examples of the different types of injury responses (i.e., survival curves based on BHIA and MA CFU counts) encountered are depicted in Figure 10. The extent and time of maximal injury is summarized in Table 4. Inactivation was generally preceded by injury. However, for acidulant-strain combinations that were highly resistant, injury was observed with little inactivation (e.g., S7 -grown cells exposed to citric acid, Fig. 7, Table 3). Alternatively, lactic acid affected some strains so strongly that detection of injury was limited to the 0-h samples (e.g., S4 -grown cells, Fig. 4, Table 3). The extent of injury was generally in the range of 65 to 99% of the viable cells (an 0.5- to 2.0- log-cycle differential between BHIA and MA counts). However, with certain acidulant-strain combinations the extent of injury transitorily increased to as much as to % of the viable cells (4- to 5-log-cycle differential). The extreme degree of injury was limited largely to -grown cells that were exposed to lactic or acetic acids. DISCUSSION Recent outbreaks associated with the survival of E. coli O157:H7 in acidic foods such as unpasteurized apple cider and fermented meats have underscored the need for accurate data on the acid resistance of pathogenic bacteria, which requires the identification of conditions that lead to maximum acid resistance. For E. coli O157:H7, these conditions occur with stationary-phase cells where rpos geneassociated and ph-dependent resistance systems are both FIGURE 4. The inactivation of Escherichia coli O157:H7 S4 (932) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). FIGURE 6. The inactivation of Escherichia coli O157:H7 S6 (A9124-C1) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ).

5 J. Food Prot., Vol. 62, No. 3 BUCHANAN AND EDELSON 215 FIGURE 7. The inactivation of Escherichia coli O111:H- S7 (95JB1) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). FIGURE 8. The inactivation of the reference strain Escherichia coli S8 (ATCC 25922) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown-cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ). active (Table 2). In the present study, the survival of E. coli at 37 C in BHI adjusted to ph 3.0 was used as a model of the organism s likely behavior when exposed to acidic conditions. It approximates the conditions that might be expected in the stomach after consumption of a meal. Further, it provides a means of acquiring an accelerated assessment of the resistance characteristics of E. coli in acidic foods held at lower temperatures (6). The data indicate that the acid resistance of stationary-phase EHEC cells varies substantially among strains and is dependent on the acidulant present and whether the ph-dependent acid resistance system has been induced. Previously, we observed that EHEC strains with extreme acid resistance (S3, S5, and S7 (6) and S9 in the present study) (Figs. 3, 5, 7, and 9) exhibited little, if any, inactivation at 37 C after 7 h in HCl-adjusted BHI (ph 2.5 and 3.0), regardless of whether they were grown in or. This finding suggested that these strains might represent isolates where the mechanism responsible for inducible ph-dependent acid resistance had become constitutive. Alternatively, it was possible that the strains were so acid resistant that the 7-h exposure was insufficient to distinguish differences between - and grown cells. The current results indicate that the latter explanation is the correct one. The antimicrobial activity of the weak organic acids increased the rate of inactivation, making it possible to demonstrate increased stationaryphase acid resistance when cells were initially grown in acidogenic. For example, while there was no difference in S7 counts for - and -grown cells for BHI with HCl only, there were distinct differentials for BHI with 0.5% citric, lactic, acetic, or malic acid (Fig. 7). Hydrochloric acid was consistently the gentlest acid tested and thus provides a worst-case for acid inactivation kinetics (i.e., longest exposure times needed for inactivation). Conversely, lactic acid was consistently the most deleterious organic acid. Between these two extremes, the sensitivity of the strains to citric, acetic, and malic acid varied substantially. In general, monocarboxylic acetic acid had greater antimicrobial activity than tricarboxylic citric acid or dicarboxylic malic acid; however, this was not universal. For example, grown cells of S9 were equally resistant to HCl, citric acid, malic acid, and acetic acid (Table 3). One possibility for rationalizing the differences in acid resistance among the strains is that EHEC become animal or commodity adapted. Prior work with Listeria monocytogenes has suggested that polycarboxylic acids may have two effects: antimicrobial activity associated with the undissociated form of the molecule and a protective effect, possibly due to the chelation of metal ions by the dissociated form (8, 9, 11). Alternatively, malic acid has been reported to enhance the survival of Lactobacillus plantarum in an acidic environment by serving as an energy source that is not coupled to H -ATPase activity (18). The decreased antimicrobial activity of malic and citric acids may also reflect the lower concentrations of undissociated FIGURE 9. The inactivation of Escherichia coli O157:H7 S9 (SEA13B88) in brain heart infusion (ph 3.0) containing 0.5% lactic, acetic, citric, malic, or hydrochloric acid only when incubated for 7hat37 C. Closed symbols: -grown cells; open symbols: -grown cells; acids: lactic (, ), acetic (, #), citric (, ), malic (, ), and HCl only (, ).

6 216 E. COLI ACID RESISTANCE J. Food Prot., Vol. 62, No. 3 TABLE 3. Summary of the. effect of acid identity on the relative rate of inactivation at ph 3.0 when enterohemorrhagic Escherichia coli were incubated at 37 C for h Relative rate of inactivation by acid of cells grown in: E. coli strain a S1 HCl malic citric acetic lactic HCl citric malic acetic lactic S2 HCl malic citric acetic lactic HCl malic citric acetic lactic S3 HCl malic citric acetic lactic HCl malic citric acetic lactic S4 HCl citric malic acetic lactic HCl citric malic acetic lactic S5 HCl citric malic acetic lactic HCl citric malic acetic lactic S6 HCl citric malic acetic lactic HCl citric malic acetic lactic S7 HCl malic citric acetic lactic HCl malic citric acetic lactic S8 HCl citric acetic malic lactic HCl citric malic acetic lactic S9 HCl citric acetic malic lactic HCl citric malic acetic lactic a For strain identities, see Table 1. acid as compared to lactic acid. However, this would not account for lactic acid being more effective than acetic acid. The calculated concentration of undissociated acetic acid is almost twice that of lactic acid. At higher ph values (4.7 to 6.0), E. coli O157:H7 has been reported to be more sensitive to acetic acid than lactic acid (1). Similar observations have been made for Listeria monocytogenes (2, 24, 37) and Staphylococcus aureus (30). However, the present study indicated that lactic acid was more effective at ph 3.0. Goodson and Rowbury (19) also observed more rapid inactivation of E. coli with lactic acid than with acetic acid when cells were exposed to a ph of 3.5. The relative activity of different organic acids would also be dependent on the presence of acid-specific membrane transport systems (26). The ability of ph-dependent stationary-phase acid resistance to enhance the survival of E. coli when challenged with various organic acids is in agreement with earlier investigations. Goodson and Rowbury (17) observed that prior growth at ph 5.0 increased the survival of exponential cells of E. coli 1829 (a K-12 derivative) at ph 3.5 in the FIGURE 10. Examples of injury responses observed when enterohemorrhagic Escherichia coli were exposed to ph 3.0 at 37 C. Brain heart infusion agar counts: injured noninjured cells ( ); MacConkey agar counts: noninjured ( ). Strain S7, grown; acidulant, HCl ( ). Strain S5, -grown; acidulant, citric acid ( ). Strain S9, -grown; acidulant, acetic acid ( ). presence of 31 mm lactic acid and 12.5 mm benzoic, propionic, sorbic, acetic, or trans-cinnamic acids. Lin et al. (26) found that all tested strains of E. coli O157:H7 possessed oxidative glutamate decarboxylase and arginine decarboxylase acid resistance systems; however, the effectiveness of each of the systems varied among isolates. Further, the relative importance of the different acid resistance systems appeared to vary with acidulant. For example, they observed that the glutamate decarboxylase system was the primary mechanism for survival at ph 4.0 in the presence of 20 mm benzoic acid, whereas arginine decarboxylase and glutamate decarboxylic systems were both important for a mixture of fatty acids designed to mimic the environment of the small intestine (26). Guilfoyle and Hirshfield (22) reported that arginine decarboxylase- and lysine decarboxylase-deficient mutants of E. coli were more acid sensitive. The specific cellular sites and physiological functions protected by the induction of ph-dependent, stationaryphase acid resistance have still not been elucidated. The mechanism is likely to be complex since any explanation must account for the protective effect against both the low ph and anion effects of the various organic acids. DNA repair has been proposed as one of the mechanisms associated with increased survival in acidic environments (4). However, Sinha (34) examined the response of four classes of DNA-repair-deficient strains of E. coli exposed to mm acetic acid, 31.4 mm lactic acid, or 36.5 mm p-amino benzoic acid at ph 3.5. Their results suggested that only the polymerase repair system, but not excision repair, may be associated with acid resistance. Goodson and Rowbury (20) concluded that the increased resistance to UV irradiation that results when E. coli is made acid resistant is independent of the SOS response. The cell membrane of E. coli is another site that is believed to be intimately associated with its acid resistance (4). The loss of resistance to bile salts observed in the present study is indicative of membrane damage. The increased maximal injury observed with -grown cells exposed to lactic and acetic acids suggest that induction of ph-dependent acid resistance increases the cells ability to

7 J. Food Prot., Vol. 62, No. 3 BUCHANAN AND EDELSON 217 TABLE 4. The extent and time of maximal injury observed when enterohemorrhagic Escherichia coli grown in acidogenic (tryptic soy broth plus dextrose) and nonacidogenic (TSB without dextrose) media were incubated in brain heart infusion broth containing 0.5% lactic, acetic, citric, or malic acid and adjusted to ph 3.0 with hydrochloric acid a Maximum differential b : log no. CFU in BHI log no. CFU in MA, by acid (incubation time (h) of maximum rate of injury) E. coli strain c HCl only d Lactic acid Acetic acid Citric acid Malic acid S (5) 1.18 (5) 3.78 (2) 0.39 (0) 1.48 (2) 0.80 (2) 1.34 (7) 1.20 (2) 0.93 (7) 0.85 (5) S (5) 0.96 (3) 0.26 (0) 0.57 (0) 1.76 (2) 0.03 (0) 1.34 (2) 1.60 (2) 1.59 (7) 0.70 (7) S (7) 1.10 (7) 4.47 (2) 2.06 (0) 5.34 (5) 1.77 (2) 2.56 (5) 2.26 (7) 1.17 (5) 0.69 (5) S (2) 0.93 (3) 3.84 (2) 1.22 (0) 4.36 (5) 1.52 (2) 2.62 (7) 2.27 (5) 1.81 (7) 1.92 (5) S (7) 1.05 (5) 1.14 (2) 0.25 (2) 2.85 (7) 1.38 (5) 1.10 (7) 1.57 (7) 0.32 (5) 1.00 (7) S (7) 1.73 (7) 5.32 (2) 0.76 (0) 4.79 (5) 0.26 (0) 1.64 (5) 1.34 (2) 0.41 (7) 0.46 (2) S (5) 0.31 (7) 4.61 (2) 0.07 (0) 3.57 (5) 1.40 (5) 2.52 (2) 0.87 (7) 1.27 (7) 0.38 (5) S (2) 0.88 (5) 2.57 (2) 1.18 (0) 4.19 (7) 0.90 (0) 1.43 (7) 0.71 (2) 1.08 (7) 1.17 (0) S (7) 0.65 (7) 2.53 (2) 1.28 e (2) 4.17 (7) 3.38 (5) 0.26 (7) 0.68 (5) 0.35 (7) 0.16 (5) a Cultures were grown in acidogenic tryptic soy broth with 1% dextrose () and nonacidogenic tryptic soy broth without dextrose () for 18 h at 37 C. The final ph was approximately 4.6 to 4.8 and 7.0 to 7.2, respectively. b Each value represents the average of at least three replicates. c For strain identities, see Table 1. d Values for the HCl-only cultures for strains S1 through S8 are from reference (6). e No colonies were detected on the MacConkey agar plates for one or more of the replicates. Assigned a value equal to the lower limit of detection (log CFU/ml 1.00). remain viable despite significant membrane damage (Table 4). Recently, Brown et al. (5) reported that induction of ph-dependent stationary-phase acid resistance may involve modification of the cell s phospholipids such that there are increased levels of cyclopropane fatty acids that stabilize the cell membrane. The extent of injury transitorily observed (i.e., as much as a 100,000-fold) dramatically demonstrates the need for EHEC detection methods to include suitable recovery protocols if the microorganism was exposed to acidic conditions. Foods that rely on acid inactivation of pathogens have been consumed for millennia, with millions of servings eaten each day. However, the results of this study add to the increasing body of knowledge that is establishing the complex nature and fragility of this preservation technique. With the emergence of highly infectious, acid-resistant pathogens, there is an acute need to understand and predict how ph, acid identity, acid concentration, temperature, water activity, and antimicrobial compounds interact to control pathogens in the wide range of ready-to-eat acidic foods that are not thermally processed. Furthermore, the results of the present study demonstrate that even within a single serotype, the ability to survive in acidic environments can vary substantially among isolates and can be enhanced by physiological stress responses and prior growth conditions. Both the biological variability and stress response mechanisms of foodborne pathogens must be considered when designing formulations and systems for the production of foods that rely on acidity as the primary means of controlling microbiological hazards. REFERENCES 1. Abdul-Raouf, U. M., L. R. Beuchat, and M. S. Ammar Survival and growth of Escherichia coli O157:H7 in ground roasted beef as affected by ph, acidulants, and temperature. Appl. Environ. Microbiol. 59: Ahamad, N., and E. H. Marth Behavior of Listeria monocytogenes at 7, 13, 21 and 35 C in tryptose broth acidified with acetic, citric and lactic acids. J. Food Prot. 52: Arnold, K. W., and C. W. Kaspar Starvation- and stationaryphase-induced acid tolerance in Escherichia coli O157:H7. Appl. Environ. Microbiol. 61: Bearson, S., B. Bearson, and J. W. Foster Acid stress responses in enterobacteria. FEMS Microbiol. Lett. 147: Brown, J. L., T. Ross, T. A. McMeekin, and P. D. Nichols Acid habituation of Escherichia coli and the potential role of cyclopropane fatty acids in low ph tolerance. Int. J. Food Microbiol. 31: Buchanan, R. L., and S. G. Edelson Culturing enterohemorrhagic Escherichia coli in the presence and absence of glucose as a simple means of evaluating the acid tolerance of stationary-phase cells. Appl. Environ. Microbiol. 62: Buchanan, R. L., S. G. Edelson, and G. Boyd. Effects of ph and acid resistance on the radiation resistance of enterohemorrhagic Escherichia coli. J. Food Prot. 62: Buchanan, R. L., and M. H. Golden Interaction of citric acid concentration and ph on the kinetics of Listeria monocytogenes inactivation. J. Food Prot. 57: Buchanan, R. L., and M. H. Golden Interactions between ph and malic acid concentration on the inactivation of Listeria monocytogenes. J. Food Safety 18: Buchanan, R. L., M. H. Golden, and J. G. Phillips Expanded models for the nonthermal inactivation of Listeria monocytogenes. J. Appl. Microbiol. 82: Buchanan, R. L., M. H. Golden, and R. C. Whiting Differentiation of the effects of ph and lactic and acetic acid concentration on the kinetics of Listeria monocytogenes inactivation. J. Food Prot. 56: Buchanan, R. L., M. H. Golden, R. C. Whiting, J. G. Phillips, and J. L. Smith Non-thermal inactivation models for Listeria monocytogenes. J. Food Sci. 59: Cheville, A. M., K. W. Arnold, C. Buchrieser, C.-M. Cheng, and C. W. Kaspar rpos regulation of acid, heat and salt tolerance in Escherichia coli O157:H7. Appl. Environ. Microbiol. 62:

8 218 E. COLI ACID RESISTANCE J. Food Prot., Vol. 62, No del Portillo, F. G., J. W. Foster, and B. B. Finlay Role of acid tolerance response genes in Salmonella typhimurium virulence. Infect. Immun. 61: Farber, J. M., and F. Pagotto The effect of acid shock on the heat resistance of Listeria monocytogenes. Lett. Appl. Microbiol. 15: Foster, J. W Low ph adaptation and the acid tolerance response of Salmonella typhimurium. Crit. Rev. Microbiol. 21: Gahan, C. G. M., B. O Driscoll, and C. Hill Acid adaptation of Listeria monocytogenes can enhance survival in acidic foods and during milk fermentation. Appl. Environ. Microbiol. 62: Garcia, M. J., M. Zúñiga, and H. Kobayashi Energy production from L-malic acid degradation and protection against acidic internal ph in Lactobacillus plantarum CECT 220. J. Gen. Microbiol. 138: Goodson, M., and R. J. Rowbury Resistance of acid-habituated Escherichia coli to organic acids and its medical and applied significance. Lett. Appl. Microbiol. 8: Goodson, M., and R. J. Rowbury RecA-independent resistance to irradiation with u.v. light in acid-habituated Escherichia coli. J. Appl. Bacteriol. 70: Gordon, J., and P. L. C. Small Acid resistance in enteric bacteria. Infect. Immun. 61: Guilfoyle, D. E., and I. N. Hirshfield The survival benefit of short-chain organic acids and the inducible arginine and lysine decarboxylase genes for Escherichia coli. Lett. Appl. Microbiol. 22: Ita, P. S., and R. W. Hutkins Intracellular ph and survival of Listeria monocytogenes Scott A in tryptic soy broth containing acetic, lactic, citric and hydrochloric acids. J. Food Prot. 54: Leyer, G. J., and E. A. Johnson Acid adaptation induces crossprotection against environmental stresses in Salmonella typhimurium. Appl. Environ. Microbiol. 59: Leyer, G. J., L.-L. Wang, and E. A. Johnson Acid adaption of Escherichia coli O157:H7 increases survival in acidic foods. Appl. Environ. Microbiol. 61: Lin, J., M. P. Smith, K. C. Chapin, H. S. Baik, G. N. Bennett, and J. W. Foster Mechanisms of acid resistance in enterohemorrhagic Escherichia coli. Appl. Environ. Microbiol. 177: Lou, Y., and A. E. Yousef Adaptation to sublethal environmental stresses protects Listeria monocytogenes against lethal preservation factors. Appl. Environ. Microbiol. 63: Miller, A. J., J. E. Call, and R. C. Whiting Comparison of organic acid salts for Clostridium botulinum control in an uncured turkey product. J. Food Prot. 56: Miller, L. G., and C. W. Kaspar Escherichia coli O157:H7 acid tolerance and survival in apple cider. J. Food Prot. 57: Minor, T. E., and E. H. Marth Loss of viability by Staphylococcus aureus in acidified media. I. Inactivation by several acids, mixtures of acids, and salts of acids. J. Milk Food Technol. 35: Nojoumi, S. A., D. G. Smith, and R. J. Rowbury Tolerance to acid in ph 5.0-grown organisms of potentially pathogenic gramnegative bacteria. Lett. Appl. Microbiol. 21: O Driscoll, B., C. G. M. Gahan, and C. Hill Adaptive acid tolerance response in Listeria monocytogenes: isolation of an acidtolerant mutant which demonstrates increased virulence. Appl. Environ. Microbiol. 62: Peterson, W. L., P. A. Mackowiak, C. C. Barnett, M. Marling-Carson, and M. L. Haley The human gastric bactericidal barrier: Mechanisms of action, relative antibacterial activity, and dietary influences. J. Infect. Dis. 159: Sinha, R. P Toxicity of organic acids for repair-deficient strains of Escherichia coli. Appl. Environ. Microbiol. 51: Small, P., D. Blankenhorn, D. Welty, E. Zinser, and J. L. Slonczewski Acid and base resistance in Escherichia coli and Shigella flexneri: role of rpos and growth ph. J. Bacteriol. 176: Sorrels, K. M., D. C. Enigl, and J. R. Hatfield Effect of ph, acidulant, time and temperature on the growth and survival of Listeria monocytogenes. J. Food Prot. 52: Whiting, R. C Modeling bacterial survival in unfavorable environments. J. Ind. Microbiol. 12: Whiting, R. C., S. Sackitey, S. Calderone, K. Morely, and J. G. Phillips Model for the survival of Staphylococcus aureus in nongrowth environments. Int. J. Food Microbiol. 31: Wilmes-Riesenberg, M. R., B. Bearson, J. W. Foster, and R. Curtiss III Role of the acid tolerance response in virulence of Salmonella typhimurium. Infect. Immun. 64: