503

Journal of Food Protection, Vol. 72, No. 3, 2009, Pages 503509

Escherichia coli O157:H7 Strains Isolated from Environmental

Sources Differ Significantly in Acetic Acid Resistance Compared
with Human Outbreak Strains
DEOG-HWAN OH,1 YOUWEN PAN,2 ELAINE BERRY,3 MICHAEL COOLEY,4 ROBERT MANDRELL,4
AND FREDERICK BREIDT, JR.5*

1Department of Food Science and Biotechnology, College of Bioscience and Biotechnology, Kangwon National University, Hyoja 2-dong,

Chuncheon 200-701, Republic of Korea; 2Department of Microbiology, North Carolina State University, Raleigh, North Carolina 27695-7615; 3U.S.
Department of Agriculture, Agricultural Research Service, Roman L. Hruska U.S. Meat Animal Research Center, P.O. Box 166, Spur 18D,
Clay Center, Nebraska 68933-0166; 4U.S. Department of Agriculture, Agricultural Research Service, Western Regional Research Center,
800 Buchanan Street, Albany, California 94710; and 5U.S. Department of Agriculture, Agricultural Research Service, Department of Food,
Bioprocessing, and Nutrition Sciences, North Carolina State University, Raleigh, North Carolina 27695-7624, USA

MS 08-255: Received 3 June 2008/Accepted 7 November 2008

ABSTRACT
A number of studies on the influence of acid on Escherichia coli O157:H7 have shown considerable strain differences,
but limited information has been reported to compare the acid resistance based on the different sources of E. coli O157:H7
isolates. The purpose of this study was to determine the survival of E. coli O157:H7 strains isolated from five sources (foods,
bovine carcasses, bovine feces, water, and human) in 400 mM acetic acid solutions under conditions that are typical of acidified
foods. The isolates from bovine carcasses, feces, and water survived acetic acid treatment at pH 3.3 and 30C significantly (P
0.05) better than did any food or human isolates. However, resistance to acetic acid significantly increased as temperature
decreased to 15C for a given pH, with little (P 0.05) difference among the different isolation sources. All groups of E.
coli O157:H7 strains showed more than 1.8- to 4.5-log reduction at pH 3.3 and 30C after 25 min. Significantly reduced (less
than 1-log reduction) lethality for all E. coli O157:H7 strain mixtures was observed when pH increased to 3.7 or 4.3, with
little difference in acetic acid resistance among the groups. The addition of glutamate to the acetic acid solution or anaerobic
incubation provided the best protection compared with the above conditions for all groups of isolates. These results suggest
that temperature, pH, and atmospheric conditions are key factors in establishing strategies for improving the safety of acidified
foods.

Enterohemorrhagic Escherichia coli O157:H7 has de- acid preservation of foods (19, 20, 27, 29, 34). Concern
veloped multiple mechanisms to survive under low-pH con- about the survival of E. coli O157:H7 in acidified foods
ditions. Because of its low infectious dose, an important has led to recent research to define the heat treatments (13)
component of E. coli O157:H7 pathogenesis is thought to required to assure a 5-log reduction in the numbers of this
be the ability to survive in extremely acidic environments, and other acid-resistant pathogens. Similarly, acidified
such as in stomach acid or in areas of the intestine that foods with a pH of 3.3 or below need not be heat processed
contain organic acids (14, 21, 23, 35). Disease outbreaks to assure safety, but do require a holding time after the
caused by E. coli O157:H7 in acid foods (3, 8, 28, 38) have addition of acetic acid (12).
led to increased research on acid resistance mechanisms of Waterborne outbreaks of E. coli O157:H7 have been
this organism (2, 17, 36, 40, 41). Acidified foods such as reported from irrigation or drinking water contaminated
pickled cucumbers and peppers are typically prepared in with infected feces, although the specific source of the fecal
hermetically sealed containers that hold acetic acid as the contamination is sometimes unclear (18). The major res-
primary acidulant, lack dissolved oxygen, contain NaCl at ervoir of E. coli O157:H7 is the bovine gastrointestinal
2 to 3%, and have pH values between 3.0 and 4.6. Several tract, and human illness has been associated with the con-
studies have shown that adaptation to acidic conditions can sumption of undercooked ground beef and unpasteurized
further enhance the survival of E. coli O157:H7 and other milk. Human infections have also been associated with a
pathogens in foods that are preserved by low pH and acids, variety of ready-to-eat foods, including fresh vegetable
as well as the cross-protection against heat, salt, and organic
products, as well as acid foods, such as apple cider and
* Author for correspondence. Tel: 919-513-0186; Fax: 919-513-0810;
fermented meats (3, 8, 28, 38).
E-mail: [email protected]. Organic acids have been used in foods as acidulants,
Mention of a trademark or proprietary product does not constitute a flavor enhancers, and to enhance microbial safety. Organic
guarantee or warranty of the product by the U.S. Department of Agri- acids are often used in foods with other preservatives or
culture or North Carolina Agricultural Research Service, nor does it
imply approval to the exclusion of other products that may be suitable.
preservation systems, such as acidic pH, drying, heat, an-
Paper no. FSR08-09 of the Journal Series of the Department of Food aerobiosis, chemical preservatives, or refrigeration. Weak
Science, North Carolina State University, Raleigh, NC 27695-7624. acids often have antimicrobial activity associated with their
504 OH ET AL. J. Food Prot., Vol. 72, No. 3

undissociated form, which can cross cell membranes freely cell suspension into 180 l of 0.1 M 3-(N-morpholino)propane-
(10, 15). Factors affecting the antimicrobial activity of or- sulfonic acid buffer (pH 7.2; Sigma, St. Louis, MO) with 0.85%
ganic acids include pH, acid concentration, and ionic saline in a 96-well microplate. Numbers of viable cells were de-
strength, as well as the physiology of the target microor- termined after additional dilution and plating on TSA supple-
mented with 1% glucose, with a spiral plater (model 4000, Spiral
ganisms and their environment (growth phase, aerobic or
Biotech, Inc., Norwood, MA). Plates were incubated for 24 to 48
anaerobic atmosphere, acid adaptation, and so forth) (11, h at 37C, and an automatic plate reader (QCount, Spiral Biotech)
14, 30, 31). Temperature is a primary factor influencing was used to count the colonies.
acid activity, with rising temperature typically resulting in For the strain mixture study, equal volumes of bacterial cell
increasing effectiveness of organic acids (1, 12, 26). suspensions of four to six strains from each isolation source (iden-
Previous studies on the influence of organic acids on tification numbers shown in boldface in Table 1) were pooled and
E. coli O157:H7 have shown strain differences (9, 26, 33, mixed prior to the acid challenge studies. The procedure of the
37, 42), but limited information has been reported to com- acetic acid challenge for the mixed strains was the same as that
pare the acid resistance based on the different sources of for individual strains, but additional conditions were tested, in-
E. coli O157:H7 isolates under the exposure to various en- cluding different pH levels (pH 3.3, 3.7, and 4.3), different tem-
vironmental conditions that simulate those found in foods. peratures (15, 22, and 30C), and different atmospheric conditions
(aerobic and anaerobic). Anaerobic solutions were incubated at
The purpose of this study was to determine the survival of
room temperature in an anaerobic chamber (COY, Grass Lake,
E. coli O157:H7 strains isolated from five different sources MI) with a mixed anaerobic gas (5% carbon dioxide, 10% hydro-
(food, bovine carcasses, bovine feces, water, and humans) gen, and the balance nitrogen) for 24 h prior to use, reducing
in acetic acid solution exposed under different environmen- dissolved oxygen from 5 mg/liter (typical of tap water) to less
tal conditions. than 0.05 mg/liter as determined with a dissolved oxygen probe
(31).
MATERIALS AND METHODS
E. coli isolates and culture conditions. A total of 52 E. coli Genomic fingerprint analysis. E. coli O157 isolate finger-
O157:H7 strains, two E. coli O157 strains, and five E. coli non- prints were generated and analyzed by pulsed-field gel electro-
O157:H7 strains from various sources including foods, bovine car- phoresis (PFGE) of SpeI-digested genomic DNA by using the
casses, bovine feces, water, and humans were used in this study PulseNet procedure (http://www.cdc.gov/pulsenet/). Pulsed-field
(Table 1). The non-O157:H7 strains were all commensal non- gel certified agarose was obtained from Bio-Rad (Hercules, CA);
pathogenic E. coli strains of bovine feces origin and included in Tris-borate-EDTA running buffer and lysozyme were purchased
the study for comparative purposes. Additional details regarding from Sigma. SpeI was purchased from New England Biolabs
some of the outbreak strains and strains isolated from watersheds (Beverly, MA). Lambda concatemers (Bio-Rad) were used as size
have been reported previously (22). All strains were stored at markers. E. coli banding patterns were analyzed and comparisons
80C in Bacto tryptic soy broth (TSB, pH 7.2; BD Biosciences, were made by using Molecular Analyst software (Bio-Rad), em-
San Jose, CA) supplemented with 16% glycerol. Each culture was ploying the Dice similarity coefficient in conjunction with the un-
streaked from frozen stocks onto plates of Bacto tryptic soy agar weighted pair group method by using arithmetic averages for clus-
(TSA; BD Biosciences) and incubated at 37C for 24 h. To prepare tering.
cells for the acid challenge, an overnight culture of each E. coli
Statistical analysis. Three or more replications of the tests
strain was inoculated into 10.0 ml of TSB supplemented with 1%
for each strain were carried out. Strains were grouped by source.
glucose and incubated statically for 18 h at 37C to induce acid
Data analysis was done with a one-way analysis of variance by
resistance (16). Each culture was washed twice by centrifugation
using SAS (version 6.12 software, SAS Institute, Inc., Cary, NC).
at 3,500 g for 10 min at 10C, and finally suspended in phys-
Differences (P 0.05) between means of log reduction for the
iological saline (8.5 g/liter of NaCl). The optical density at 600
various groupings of strains were analyzed by using the least-
nm of each cell suspension was adjusted to 0.8 (ca. 109 cells per
significant difference test.
ml). Actual starting concentrations were confirmed by plating se-
rial dilutions on TSA. RESULTS
Preparation of acetic acid solution. The acetic acid solu- The survival of E. coli strains, measured as log-CFU-
tions were prepared essentially as described, but with a few mod- per-milliliter reduction of each individual strain exposed to
ifications (11). The ionic strength in the acetic solutions (400 mM)
an acetic acid solution typical of nonpasteurized acidified
was held constant by adjusting the NaCl concentration based on
acid anion concentrations as determined by using pHtools, a
vegetable products (400 mM acetic acid, pH 3.3, ionic
MATLAB routine (24). The pH of the acetic acid solutions was strength 0.34) for 25 min, with or without glutamic acid (2
adjusted with HCl and NaOH. For the treatment of the acid so- mM), at 30C is shown in Table 1. The mean value of the
lution in the presence of glutamic acid, the concentration of glu- viable cell reductions of the eight bovine carcass isolates
tamic acid was 2 mM. after acetic acid treatment in the absence of glutamic acid
was 1.18 log CFU/ml, showing the highest resistance com-
Acid challenge studies. For the pure (individual) culture ex-
pared with isolates from other sources. The survival of bo-
periments, aliquots of 200 l of each cell suspension were added
into 1.8 ml of acetic acid solutions (400 mM, pH 3.3) with or
vine carcass isolates was significantly (P 0.05) different
without glutamic acid in 12-well tissue culture plates (flat-bottom from that of food isolates (the mean reduction was 2.17 log
plates; no. 351172 Falcon, Franklin Lakes, NJ) and incubated aer- CFU/ml) and from human isolates, which had a mean re-
obically at 30C for 25 min. After incubation, the cells were im- duction of 3.18 log CFU/ml (P 0.0001), but not signif-
mediately diluted 10-fold to rapidly neutralize the pH prior to icantly different from the bovine feces isolates (which in-
plating. This was done by transferring 20 l of the acid treated cluded the non-O157 isolates; the mean reduction was 1.4
J. Food Prot., Vol. 72, No. 3 ACID RESISTANCE OF E. COLI O157:H7 505

TABLE 1. Escherichia coli strains used in the study and survival in 400 mM acetic acid solutions (pH 3.3 at 30C for 25 min)
Isolatesa Reduction in log CFU/mlb

ID no.c Original ID no. Serotype Source (reference) Without glutamic acid With glutamic acid

B0201 SRCC 1675 O157:H7 Apple cider, October 2002 3.56 0.16 2.66 0.20
B0349 NMSLD-1 O157:H7 Spinach 2.91 0.22 0.02 0.01
B0264 RM1484 O157:H7 Apple juice, associated with 1996 outbreak 2.80 0.20 0.02 0.01
B0204 SRCC 1941 O157:H7 Pork, September 2002 2.52 0.16 0.1 0.03
B0202 SRCC 1486 O157:H7 Salami, October 2002 2.14 0.06 2.89 0.06
B0203 SRCC 2061 O157:H7 Ground beef, October 2002 1.22 0.32 0.09 0.07
B0348 38094 O157:H7 Salami 1.12 0.13 0.04 0.03
B0350 960212 O157:H7 Sakai 1.11 0.27 0.03 0.02
B0243 3004-98 O157:H7 Bovine carcass 1.63 0.24 0.29 0.06
B0257 234AB1 O157 Bovine carcass 1.62 0.11 0.06 0.09
B0235 5A-1 non-O157 Bovine feces (6) 1.50 0.13 0.07 0.07
B0256 183 (H18-1) O157 Bovine carcass 1.41 0.20 0.02 0.01
B0242 233AC1 O157:H7 Bovine carcass (4, 25) 1.31 0.12 0.11 0.06
B0240 294RC1 O157:H7 Bovine carcass (4, 25) 1.17 0.15 0.17 0.08
B0239 258AC1 O157:H7 Bovine carcass (4, 25) 1.13 0.35 0.39 0.10
B0237 1C-3 non-O157 Bovine feces (6) 0.95 0.15 0.08 0.01
B0238 97AC1 O157:H7 Bovine carcass (4, 25) 0.64 0.07 0.03 0.02
B0241 28RC1 O157:H7 Bovine carcass (4, 25) 0.46 0.11 0.08 0.10
B0258 F6B-2 O157:H7 Bovine feces (5) 1.64 0.09 0.16 0.10
B0234 3A-1 non-O157 Bovine feces (6) 1.60 0.19 0.02 0.02
B0259 F7B-1 O157:H7 Bovine feces (5) 1.56 0.33 0.32 0.19
B0254 106-4-1 O157 Bovine feces 1.54 0.24 0.29 0.13
B0262 NS-P410 O157 Bovine feces 1.53 0.01 0.08 0.07
B0260 F15B-1 O157 Bovine feces (5) 1.42 0.12 0.28 0.13
B0236 3C-3 non-O157 Bovine feces (6) 1.40 0.20 0.02 0.01
B0261 902-2 O157 Bovine feces (7) 1.33 0.17 0.16 0.19
B0253 1307-8(8-1) O157 Bovine feces 1.19 0.20 0.23 0.06
B0255 207-10-2 O157 Bovine feces 1.12 0.45 0.28 0.15
B0233 1A-5 non-O157 Bovine feces (6) 1.08 0.23 0.07 0.02
B0301 RM5630 O157:H7 Water (18) 3.57 0.32 0.12 0.00
B0307 RM5875 O157:H7 Water (18) 2.41 0.40 0.03 0.02
B0306 RM5850 O157:H7 Water (18) 2.38 0.42 0.04 0.03
B0309 RM5714 O157:H7 Water (18) 2.13 0.20 0.20 0.15
B0302 RM5667 O157:H7 Water (18) 1.40 0.12 0.10 0.04
B0297 RM5450 O157:H7 Water (18) 1.66 0.15 0.14 0.03
B0299 RM5607 O157:H7 Water (18) 1.66 0.38 0.35 0.39
B0285 RM4886 O157:H7 Water (18) 1.38 0.22 0.07 0.02
B0275 RM4859 O157:H7 Water (18) 1.36 0.36 0.34 0.08
B0305 RM5754 O157:H7 Water (18) 1.35 0.45 0.05 0.05
B0281 RM4876 O157:H7 Water (18) 1.28 0.13 0.11 0.11
B0289 RM5036 O157:H7 Water (18) 1.21 0.16 0.05 0.04
B0280 RM4864 O157:H7 Water (18) 1.10 0.07 0.10 0.00
B0287 RM4888 O157:H7 Water (18) 0.80 0.15 0.12 0.06
B0283 RM4884 O157:H7 Water (18) 0.50 0.33 0.05 0.00
B0269 RM4263 O157:H7 Human, outbreak, 2000, waterborne 4.71 0.08 0.34 0.34
B0273 RM4688 O157:H7 Human, outbreak, 2002, leafy vegetable 4.13 0.46 0.07 0.07
B0247 3159-98 O157:H7 Human outbreak (22) 3.63 0.34 0.08 0.08
B0296 RM5279 O157:H7 Human, outbreak, 2005, leafy vegetable 3.58 0.32 0.17 0.10
B0311 RM6011 O157:H7 Human, outbreak, 2006, leafy vegetable 3.43 0.13 0.35 0.03
B0246 3139-98 O157:H7 Human outbreak (22) 3.35 0.21 0.05 0.01
B0271 RM4406 O157:H7 Human, outbreak, 2003, leafy vegetable 3.15 0.34 3.75 0.06
B0250 3261-98 O157:H7 Human outbreak 2.90 0.31 0.11 0.13
B0263 RM1242 O157:H7 Human, sporadic, 1997 2.85 0.37 0.09 0.11
B0251 3361-91 O157:H7 Human outbreak (22) 2.82 0.52 0.09 0.11
B0249 3187-95 O157:H7 Human outbreak 2.75 0.25 0.07 0.09
B0266 RM2189 O157:H7 Human, outbreak, 1999, taco meat 2.73 0.51 0.19 0.20
B0245 3055-93 O157:H7 Human outbreak (22) 2.72 0.29 0.28 0.04
B0265 RM1918 O157:H7 Human, outbreak, 1999, lettuce 2.57 0.34 0.11 0.15
B0244 3014-93 O157:H7 Human outbreak (22) 2.43 0.04 0.11 0.15
a ID, identification.
b Values are means standard deviations of the results from three independent replicate tests for each strain.
c The strains in boldface were selected to make mixtures representing the corresponding sources.
506 OH ET AL. J. Food Prot., Vol. 72, No. 3

FIGURE 1. The effect of 400 mM acetic

acid on the survival of the mixtures of E.
coli O157:H7 strains from different sources
at different pH values (white bars, pH 3.3;
gray bars, pH 3.7; black bars, pH 4.3) and
different temperatures. (A) 30C. (B) 22C.
The bars indicate the mean values, and the
error bars indicate the standard deviations
(n 3). Different letters indicate a signif-
icant (P 0.05) difference.

log CFU/ml, P 0.998) or the water isolates (mean re- senting the five different source groups to the acetic acid
duction was 1.65 log CFU/ml, P 0.75). Surprisingly, the solution (400 mM, pH 3.3, and 30C for 25 min) had the
human isolates were the most sensitive to the acetic acid same trend as that of the individual strains from the cor-
solution, and their survival was significantly (P 0.0001) responding groups (Table 1 and Figure 1). When the mixed
lower than that of any of the other four groups of isolates cells of human outbreak isolates were exposed to the acetic
(to any group). The sensitivity of each of the 15 human acid solution, the viable cell density was decreased by ap-
isolates was significantly greater than that of the eight bo- proximately 4.5 orders of magnitude, making this the group
vine carcass isolates and 13 bovine feces isolates. most sensitive to the acid solution under these conditions.
To further investigate the survival of the E. coli strains The viable cell densities for the mixtures of food, bovine
from different sources, four to five strains from each source carcass, bovine feces, and water isolates were reduced by
were selected to make a mixture representing their corre- 3.3, 1.8, 1.8, and 1.9 log CFU/ml, respectively. However,
sponding source (the selected strain identification numbers the sensitivity of E. coli O157:H7 to the acetic acid solution
are those shown in boldface in Table 1) and were treated was significantly decreased when pH increased to 3.7 and
under additional conditions chosen for their relevance to 4.3 at 30C. There was no significant (P 0.05) difference
acidified foods. The effect of 400 mM acetic acid on the in the survival of the isolate mixtures from different sources
survival of the cocktail of E. coli O157:H7 strains from (Fig. 1A) at a relatively higher pH (3.7 or 4.3). When the
different sources at different pH values, temperatures, and challenge test was performed at lower temperatures, the
atmospheric conditions are illustrated in Figures 1 and 2. sensitivity of E. coli O157:H7 to the 400 mM acetic acid
The resistance of the mixed-strain suspensions repre- solution was also significantly reduced. At 22C, the acetic
acid solution reduced the population of the human isolate
mixture by approximately 1.2 log CFU/ml. Fewer than 90%
of the cells in the strain mixtures from the other four groups
were inactivated (Fig. 1B). At 15C, there was little or no
detectable inhibitory effect(s) on any mixtures of E. coli
O157:H7 strains, regardless of the pH values (data not
shown).
We compared the sensitivity of E. coli O157:H7 strains
from different sources in 400 mM acetic acid solution (pH
3.3, 30C for 25 min) in the presence and absence of dis-
solved oxygen. Under anaerobic conditions (Fig. 2), the E.
coli O157:H7 cells from all five sources showed little or
no reduction in cell numbers, as we have previously ob-
served (31). There was a significant (P 0.05) difference
in acetic acid resistance among the mixtures of E. coli
O157:H7 strains from different sources between anaerobic
and aerobic conditions, however. We also determined the
survival of the mixtures of E. coli O157:H7 strains from
different sources in 400 mM acetic acid containing 2 mM
glutamic acid (Fig. 2). Acetic acid containing glutamic acid
FIGURE 2. The effect of anaerobic (striped bars) incubation or
showed minimal inhibitory effects on all of the E. coli
addition of glutamic acid (gray bars) on the acetic acid resistance
of the mixture of E. coli O157:H7 strains from different sources
O157:H7 strains, irrespective of pH values or temperatures,
at pH 3.3 and 30C. The white bars represent the control treat- and there was no significant difference in acetic acid resis-
ment (aerobic, no glutamic acid) as shown in Figure 1A (pH 3.3). tance among the groups tested with glutamic acid.
The bars indicate the mean values and the error bars indicate the PFGE was applied for molecular typing of 24 E. coli
standard deviations (n 3). Different letters indicate a significant O157:H7 isolates from different sources (Fig. 3). Using di-
(P 0.05) difference. gestion with SpeI, a total of 20 restriction enzyme digestion
J. Food Prot., Vol. 72, No. 3 ACID RESISTANCE OF E. COLI O157:H7 507

FIGURE 3. Dendogram of PFGE data for

the isolates used for the group analysis.
The alignment of the bands and the den-
dogram were generated with Molecular
Analyst software, as described in the text.
The numbers next to each lane are strain
designations (B0-numbers from Table 1),
which are followed by serotype and source
isolation.

patterns were identified. The diversity index of PFGE typ- cedure of Buchanan and Edelson (16) to induce acid resis-
ing was 0.83 (20 types identified among 24 isolates: 20 of tance during preparation of the bacterial cells. Our results
24 0.83). Isolates 275, 280, and 297, with the same showed that temperature, pH, and atmosphere are the sig-
PFGE-SpeI pattern, were isolated from water; similarly, iso- nificant factors that can alter acid resistance of E. coli
lates 253 and 258 from bovine feces, and isolates 201 and O157:H7 in acetic acid solution. Cheville et al. (19) showed
264 from food (apple cider) were matched based on source. that resistance to heat and acid challenges can be influenced
However, all of the five human clinical isolates showed dif- by the incubation temperature of the microorganism prior
ferent PFGESpeI types. The 10 isolates from bovine to an acid challenge. Previously, we determined that dis-
sources were found to have 8 different types. These data solved oxygen influences significantly the acid sensitivity
indicated that isolates from the same source having similar of E. coli O157:H7 (31). Wild-type E. coli O157:H7 cul-
acid resistance do not show an obvious trend to be grouped tures that were grown aerobically required less acetate to
together by PFGE pattern. induce extreme acid resistance than did those grown an-
aerobically, and the addition of the reducing agent cysteine
DISCUSSION to the media greatly increased the requirement for acetate
Differences in acid tolerance among isolates of E. coli (23). Small et al. (41) reported that anaerobic growth under
have been reported (9, 21, 26, 33, 37, 42), but it is difficult acidic conditions restored acid resistance of a nonpathogen-
to compare these data, because the studies were performed ic E. coli rpoS mutant.
with different microorganisms, growth media, physiological For comparisons between bacterial strains isolated
status of bacteria, organic acids, pH, and incubation tem- from different sources, we held temperature, pH, and ionic
peratures. Our studies focused on the acetic acid resistance strength constant to control protonated acid concentrations,
of a mixture of E. coli O157:H7 isolates from different as described previously (11). The adaptation of E. coli
sources. Interestingly, the non-O157:H7 strains from bovine O157:H7 strains for growth or survival in different envi-
feces had acid resistance characteristics similar to the O157: ronments, including rumen, meat, water, or human gastro-
H7 strains (Table 1). Acetic acid is the most common or- intestinal tract, may contribute to the differences in sensi-
ganic acid present in acidified foods. Contrary to our re- tivity to acetic acid that we observed. Exposure to environ-
sults, McKellar and Knight (37) reported that human out- mental stresses other than acid may lead to altered acid
break E. coli O157:H7 strains were significantly (P 0.05) resistance due to cross-protection (19, 20). Because the
better able to survive acid treatment than were strains from genes regulating responses to acid stress are controlled by
animal or human sources. However, our study used the pro- global regulatory systems (29, 32, 39), it is possible that
508 OH ET AL. J. Food Prot., Vol. 72, No. 3

changes in expression of key regulatory genes, such as rpoS better control of the occurrence, growth, or survival of this
may be responsible for the stain differences observed. Lin organism in foods.
et al. (35, 36) developed assays to separate three different
acid resistance mechanisms by which E. coli can survive ACKNOWLEDGMENTS
in low-pH environments for extended periods: arginine and The authors acknowledge Dr. Weiming Gu for his technical support
glutamate decarboxylasedependent systems and a glucose of PFGE procedure, Ms. Sandra Parker for excellent secretarial assistance,
catabolite-repression system. They suggested that these and Diana Carychao for characterization of some of the strains. This in-
vestigation was supported in part by a research grant from Pickle Packers
mechanisms promote survival in low-pH environments, International, Inc., Washington, DC, Kangwon National University, and
such as stomach and acid foods. There was no significant U.S. Department of Agriculture, Cooperative State Research, Education,
difference between the groups in overall decarboxylate-de- and Extension Service project number 2006-01240.
pendent system of acetic acid resistance (Table 1, column
6); however, several strains (B0201, B0202, and B0271 list- REFERENCES
ed in Table 1) apparently lacked the glutamate decarbox- 1. Ahamad, N., and E. H. Marth. 1989. Behavior of Listeria monocy-
ylation system (30), or if it were present, may have not togenes at 7, 13, 21 and 35C in tryptose broth acidified with acetic,
been functional. E. coli O157:H7 strains lacking the glu- citric and lactic acids. J. Food Prot. 52:688695.
2. Amold, K. W., and C. W. Kaspar. 1995. Starvation- and stationary-
tamate decarboxylase system were also found by Bhagwat
phaseinduced acid tolerance in Escherichia coli O157:H7. Appl.
et al. (9). These strains are currently the subject of further Environ. Microbiol. 61:20372039.
research, and the role of the glutamate decarboxylase sys- 3. Anonymous. 1995. Escherichia coli O157:H7 outbreak linked to
tem in survival of E. coli strains in food systems remains commercially distributed dry-cured salami. Morb. Mortal. Wkly.
unclear (40). Our results confirmed that significant differ- Rep. 44:157160.
ences in acid resistance among E. coli O157:H7 isolates 4. Barkocy-Gallagher, G. A., T. M. Arthur, G. R. Siragusa, J. E. Keen,
R. O. Elder, W. W. Laegreid, and M. Koohmaraie. 2001. Genotypic
from different sources exist. For example, the human out- analyses of Escherichia coli O157:H7 and O157 nonmotile isolates
break strains, in general, were more sensitive in the absence recovered from beef cattle and carcasses at processing plants in the
of glutamic acid than were any other group of strains (Table Midwestern states of the United States. Appl. Environ. Microbiol.
1). One outbreak strain, B0271, apparently lacked detect- 67:38103818.
able glutamic acid decarboxylase activity (Table 1). It is 5. Berry, E. D., and G. A. Barkocy-Gallagher. 2002. Acid resistance
status of Escherichia coli O157 in bovine feces as shed from natu-
interesting that the human isolates were less acid resistant rally contaminated cattle. American Dairy Science Association
than were the environmental isolates tested in this study, American Society of Animal ScienceCanadian Society of Animal
because acid resistance is a virulence factor for E. coli Science Joint Meeting, Quebec City, Quebec, Canada, 21 to 25 July
O157:H7, aiding in transit of the organism through the 2002.
6. Berry, E. D., G. A. Barkocy-Gallagher, and G. R. Siragusa. 2004.
stomach (20).
Stationary-phase acid resistance and injury of recent bovine Esche-
It is worth noting also that the outbreak strains were richia coli O157 and non-O157 biotype 1 Escherichia coli isolates.
diverse based on spatial, temporal, and genotypic differ- J. Food Prot. 67:583590.
ences. Eight of the human outbreaks strains (Table 1, the 7. Berry, E. D., and D. N. Miller. 2005. Cattle feedlot soil moisture
strains designated RM) represented eight different out- and manure content. II. Impact on Escherichia coli O157. J. Environ.
Qual. 34:656663.
breaks occurring between 1997 and 2006, and associated
8. Besser, R. E., S. M. Lett, J. T. Weber, M. P. Doyle, T. J. Barrett, J.
with different food or environmental sources (taco meat, G. Wells, and P. M. Griffin. 1993. An outbreak of diarrhea and he-
various leafy vegetables, and water), and multiple geno- molytic uremic syndrome from Escherichia coli O157:H7 in fresh
types (22). Whiting and Golden (42) reported previously on pressed apple cider. JAMA 269:22172220.
the amount of variation existing among strains of E. coli 9. Bhagwat, A. A., L. Chan, R. Han, J. Tan, M. Kothary, J. Jean-Gilles,
and B. D. Tall. 2005. Characterization of enterohemorrhagic Esch-
O157:H7, showing D-values in acid and preservative treat-
erichia coli strains based on acid resistance phenotypes. Infect. Im-
ments at pH 4.8 ranging from 24 to 189 h. Large et al. (33) mun. 73:49935003.
observed that the strain-to-strain variation within a clonal 10. Booth, I. R., and R. G. Kroll. 1989. The preservation of foods by
group was larger than variation between replicates or due low pH, p. 119160. In G. W. Gould (ed.). Mechanisms of action
to experimental error. These data are consistent with our of food preservation procedures. Elsevier Science Publishers, New
York.
results indicating no significant correlation of resistance
11. Breidt, F., J. S. Hayes, and R. F. McFeeters. 2004. Independent ef-
with genotype. Genetic analysis of several of these isolates fects of acetic acid and pH on survival of Escherichia coli in sim-
(RM strains) has shown differences between PFGE and ulated acidified pickle products. J. Food Prot. 67:1218.
multilocus variable-number tandem-repeat analysis studies 12. Breidt, F., J. S. Hayes, and R. F. McFeeters. 2007. Determination of
(22). Thus, conclusions regarding the genetic differences in 5-log reduction times for food pathogens in acidified cucumbers dur-
ing storage at 10 and 25C. J. Food Prot. 70:26382641.
the acid resistance of E. coli O157:H7 groups of bacteria
13. Breidt, F., Jr., J. S. Hayes, J. A. Osborne, and R. F. McFeeters. 2005.
will require examination of additional isolates from a va- Determination of 5-log pathogen reduction times for heat-processed,
riety of sources. acidified vegetable brines. J. Food Prot. 68:305310.
The development of acid tolerance by foodborne path- 14. Brudzinski, L., and M. A. Harrison. 1998. Influence of incubation
ogenic bacteria may be significant at several points along conditions on survival and acid tolerance response of Escherichia
coli O157:H7 and non-O157:H7 isolates exposed to acetic acid. J.
the farm-to-table continuum of food production. It is im-
Food Prot. 61:542546.
portant to understand how previous environment and pro- 15. Brul, S., and P. Coote. 1999. Preservative agents in foods: mode of
cessing conditions can affect the acid tolerance status of action and microbial resistance mechanism. Int. J. Food Microbiol.
foodborne E. coli O157:H7 in order to devise strategies for 50:117.
J. Food Prot., Vol. 72, No. 3 ACID RESISTANCE OF E. COLI O157:H7 509

16. Buchanan, R. L., and S. G. Edelson. 1996. Culturing enterohemor- 29. Hengge-Aronis, R. 1993. Survival of hunger and stress: the role of
rhagic Escherichia coli in the presence and absence of glucose as a rpoS in early stationary phase gene regulation in E. coli. Cell 72:
simple means of evaluating the acid tolerance of stationary-phase 165168.
cells. Appl. Environ. Microbiol. 62:40094013. 30. Hersh, B. M., F. T. Farooq, D. N. Barstad, D. L. Blankenshorn, and
17. Buchanan, R. L., and S. G. Edelson. 1999. pH-dependent stationary- J. L. Slonczewski. 1996. A glutamate-dependent acid resistance gene
phase acid resistance response of enterohemorrhagic Escherichia coli in Escherichia coli. J. Bacteriol. 178:39783981.
in the presence of various acidulants. J. Food Prot. 62:211218. 31. Kreske, A. C., K. Bjornsdottir, F. Breidt, and H. Hassan. 2008. Ef-
18. Centers for Disease Control and Prevention. 1999. Public health dis- fects of pH, dissolved oxygen, and ionic strength on the survival of
patch: outbreak of Escherichia coli O157:H7 and Campylobacter Escherichia coli O157:H7 in organic acid solutions. J. Food Prot.
among attendees of the Washington County FairNew York, 1999. 71:24042409.
Morb. Mortal. Wkly. Rep. 48:803804. 32. Lange, R., and R. Hengge-Aronis. 1991. Identification of a central
19. Cheville, A., K. W. Armold, C. M. Cheng, and C. W. Kasper. 1996. regulator of stationary-phase gene expression in Escherichia coli.
rpoS regulation of acid, heat, and salt tolerance in Escherichia coli Mol. Microbiol. 5:4959.
O157:H7. Appl. Environ. Microbiol. 62:18221824. 33. Large, T. M., S. T. Walk, and T. S. Whittam. 2005. Variation in acid
20. Chung, H. J., W. Bang, and M. A. Drake. 2006. Stress response of resistance among Shiga toxinproducing clones of pathogenic Esch-
Escherichia coli. Comp. Rev. Food Sci. Food Saf. 5:5264. erichia coli. Appl. Environ. Microbiol. 71:24932500.
21. Conner, D. E., and J. S. Kotrola. 1995. Growth and survival of Esch- 34. Leyer, G. J., and E. A. Johnson. 1993. Acid adaptation induces cross-
erichia coli O157:H7 under acidic conditions. Appl. Environ. Mi- protection against environmental stress in Salmonella typhimurium.
crobiol. 61:382385. Appl. Environ. Microbiol. 59:18421847.
22. Cooley, M., D. Carychao, L. Crawford-Miksza, M. T. Jay, C. Myers, 35. Lin, J., I. Lee, J. Frey, J. L. Slonczewski, and J. W. Foster. 1995.
C. Rose, C. Keys, J. Farrar, and R. E. Mandrell. 2007. Incidence and Comparative analysis of extreme acid survival in Salmonella typhi-
murium, Shigella flexneri, and Escherichia coli. J. Bacteriol. 177:
tracking of Escherichia coli O157:H7 in a major produce production
40974104.
region in California. PloS ONE 2:e1159.
36. Lin, J., M. P. Smith, K. C. Chapin, H. S. Baik, G. N. Bennett, and
23. Diez-Gonzalez, F., and J. B. Russell. 1999. Factors affecting the ex-
J. W. Foster. 1996. Mechanisms of acid resistance in enterohemor-
treme acid resistance of Escherichia coli O157:H7. Food Microbiol.
rhagic Escherichia coli. Appl. Environ. Microbiol. 62:30943100.
16:367374.
37. McKellar, R., and K. Knight. 1999. Growth and survival of various
24. Dougherty, D. P., E. Ramos Da Conceicao Neta, R. F. McFeeters, S.
strains of enterohemorrhagic Escherichia coli in hydrochloric and
R. Lubkin, and F. Breidt, Jr. 2006. Semi-mechanistic partial buffer
acetic acid. J. Food Prot. 62:14661469.
approach to modeling pH, the buffer properties, and the distribution
38. Miller, L. G., and C. W. Kasper. 1994. Escherichia coli O157:H7
of ionic species in complex solutions. J. Agric. Food Chem. 54:
acid tolerance and survival in apple cider. J. Food Prot. 57:460
60216029. 464.
25. Elder, R. O., J. E. Keen, G. R. Siragusa, G. A. Barkocy-Gallagher, 39. Price, S. B., C. M. Cheng, C. W. Kasper, J. C. Wright, F. J. DeGraves,
M. Koohmaraie, and W. W. Laegreid. 2000. Correlation of entero- T. A. Penfound, M. P. Castanie-Cornet, and J. W. Foster. 2000. Role
hemorrhagic Escherichia coli prevalence in feces, hides, and car- of rpoS in acid resistance and fecal shedding of Escherichia coli
casses of beef cattle during processing. Proc. Natl. Acad. Sci. USA O157:H7. Appl. Environ. Microbiol. 66:632637.
97:29993003. 40. Price, S. B., J. C. Wright, F. J. DeGraves, M. P. Castanie-Cornet, and
26. Entani, E., M. Asai, S. Tsujihata, Y. Tsukamoto, and M. Ohta. 1998. J. W. Foster. 2004. Acid resistance systems required for survival of
Antimicrobial action of vinegar against foodborne pathogenic bac- Escherichia coli O157:H7 in the bovine gastrointestinal tract and in
teria including Escherichia coli O157:H7. J. Food Prot. 61:953959. apple cider are different. Appl. Environ. Microbiol. 70:47924799.
27. Farber, J. M., and F. Pagotto. 1992. The effect of acid shock on the 41. Small, P., D. Blankenhorn, D. Welty, E. Zinser, and J. L. Slonczew-
heat resistance of Listeria monocytogenes. Lett. Appl. Microbiol. 15: ski. 1994. Acid and base resistance in Escherichia coli and Shigella
197201. flexneri: role of rpoS and growth pH. J. Bacteriol. 176:17291737.
28. Glass, K. A., J. M. Loeffelholz, J. P. Ford, and M. P. Doyle. 1992. 42. Whiting, R. C., and M. H. Golden. 2002. Variation among Esche-
Fate of Escherichia coli O157:H7 as affected by pH or sodium chlo- richia coli O157:H7 strains relative to their growth, survival, thermal
ride and in fermented, dry sausage. Appl. Environ. Microbiol. 58: inactivation, and toxin production in broth. Int. J. Food Microbiol.
25132516. 75:127133.