1701

Journal of Food Protection, Vol. 75, No. 9, 2012, Pages 1701–1708

doi:10.4315/0362-028X.JFP-11-520
Copyright G, International Association for Food Protection

Research Note

Evaluation of Lactic Acid as an Initial and Secondary Subprimal

Intervention for Escherichia coli O157:H7, Non-O157 Shiga
Toxin–Producing E. coli, and a Nonpathogenic E. coli Surrogate
for E. coli O157:H7

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021

C. I. PITTMAN, I. GEORNARAS, D. R. WOERNER, K. K. NIGHTINGALE, J. N. SOFOS, L. GOODRIDGE, AND K. E. BELK*

Center for Meat Safety & Quality, Department of Animal Sciences, Colorado State University, Fort Collins, Colorado 80523-1171, USA

MS 11-520: Received 28 November 2011/Accepted 12 April 2012

ABSTRACT
Lactic acid can reduce microbial contamination on beef carcass surfaces when used as a food safety intervention, but
effectiveness when applied to the surface of chilled beef subprimal sections is not well documented. Studies characterizing
bacterial reduction on subprimals after lactic acid treatment would be useful for validations of hazard analysis critical control
point (HACCP) systems. The objective of this study was to validate initial use of lactic acid as a subprimal intervention during
beef fabrication followed by a secondary application to vacuum-packaged product that was applied at industry operating
parameters. Chilled beef subprimal sections (100 cm2) were either left uninoculated or were inoculated with 6 log CFU/cm2 of a
5-strain mixture of Escherichia coli O157:H7, a 12-strain mixture of non-O157 Shiga toxin–producing E. coli (STEC), or a 5-
strain mixture of nonpathogenic (biotype I) E. coli that are considered surrogates for E. coli O157:H7. Uninoculated and
inoculated subprimal sections received only an initial or an initial and a second ‘‘rework’’ application of lactic acid in a custom-
built spray cabinet at 1 of 16 application parameters. After the initial spray, total inoculum counts were reduced from 6.0 log
CFU/cm2 to 3.6, 4.4, and 4.4 log CFU/cm2 for the E. coli surrogates, E. coli O157:H7, and non-O157 STEC inoculation groups,
respectively. After the second (rework) application, total inoculum counts were 2.6, 3.2, and 3.6 log CFU/cm2 for the E. coli
surrogates, E. coli O157:H7, and non-O157 STEC inoculation groups, respectively. Both the initial and secondary lactic acid
treatments effectively reduced counts of pathogenic and nonpathogenic strains of E. coli and natural microflora on beef
subprimals. These data will be useful to the meat industry as part of the HACCP validation process.

Carcasses and meat from healthy animals is initially
sterile but becomes contaminated with bacteria when
exposed to the environment during fabrication and process-
ing (13). Extensive research has been conducted on beef
carcass decontamination, investigating various methodolo-
gies such as spot, thermal, and chemical decontamination
and other novel techniques (13). Although single interven-
tions reduce bacterial populations, the presence of residual
bacteria remains a concern. The use of multiple sequential
interventions has been more effective than the use of
individual interventions (1). Therefore, many beef decon-
tamination systems in the United States employ multiple
combinations of knife trimming, steam vacuuming, hot
water washes, and chemical sprays to take advantage of the
additive effects of these interventions.
Antimicrobial interventions have traditionally focused
on animal hides or carcasses. Sustained refrigeration
temperatures throughout the dressing process and fabrica-
tion have been considered sufficient for controlling bacterial
growth on meat (11). However, surviving bacterial
* Author for correspondence. Tel: 970-491-6244; Fax: 970-491-5326;
E-mail: keith.belk@colostate.edu.
populations may continue to grow during temperature abuse
periods. Contact with employees, improperly sanitized
contact surfaces such as belts, tables, saws, cutting boards,
knives, or hooks, and other carcasses could reintroduce
pathogenic bacteria to meat surfaces (14). Gill et al. (4)
reported that Escherichia coli counts on beef carcasses
increased from immediately before entry into the fabrication
process to the subsequent exit from the process line. In a
similar study, Gill et al. (5) evaluated bacterial populations
on the surface of beef carcasses and primal cuts before and
after fabrication. Total coliform and E. coli counts increased
from 4.0 and 3.5 log CFU/500 carcasses to .6.0 and
5.5 log CFU/500 cuts, respectively. The increased popula-
tions on primal cuts were attributed to contact with cutting
surfaces such as tables (5). Thus, interventions for
controlling bacterial growth before, during, and after
fabrication are as important as whole carcass interventions.
On 25 July 1996, the U.S. Department of Agriculture
Food Safety Inspection Service (USDA-FSIS) (15) pub-
lished the pathogen reduction, hazard analysis and critical
control point (HACCP) systems final rule. This regulation
states that each establishment must evaluate and validate the
effectiveness of its HACCP system for controlling food
1702 PITTMAN ET AL.
safety hazards. In May 2010, the USDA-FSIS (16) issued
the draft guidance, HACCP systems validation document,
which clarified the expectation of validation. This document
addressed the importance of validating the entire HACCP
system, including prerequisite programs. It also stipulated
what was necessary for validation and defined validation
as ‘‘the process of demonstrating that the HACCP system
as designed can adequately control identified hazards to
produce a safe, unadulterated product.’’ Validation has two
elements: (i) scientific or technical support for the HACCP
system and (ii) practical in-plant demonstration that the
HACCP system is effective. When the scientific documen-
tation defines a particular parameter, that parameter must be
used for the process. In addition to proving that the HACCP
system is theoretically sound and based on scientific
support, the establishment must show through in-plant
demonstration that its HACCP system is able to achieve the
desired effect. The first step of in-plant validation is defining
critical operational parameters, including treatment time,
temperature, pressure, and concentration or microbial log
reduction. In-plant validation must provide data sufficient to
prove that the process can operate effectively on a daily
basis (16).
To effectively demonstrate that a HACCP system is
working properly to control pathogens, surrogate indicator
organisms are essential. Surrogate organisms can include
Enterobacteriaceae, coliforms, or nonpathogenic E. coli.
An effective indicator organism proves the ability of the
HACCP system to reduce pathogen populations without
artificial introduction of pathogens into the plant. Recently,
nonpathogenic E. coli biotype I isolates were described as
surrogates for E. coli O157:H7. Marshall et al. (9) evaluated
bacteria isolated from beef hides and identified five
nonpathogenic E. coli isolates as surrogates for estimating
E. coli O157:H7 growth and reduction after antimicrobial
interventions. These isolates were further evaluated during
cooking, fermentation, freezing, and refrigerated storage of
meat by Keeling et al. (8), who found no difference between
two of the isolates (BAA-1428 and BAA-1430) and E. coli
O157:H7 during frozen storage. These surrogates even
survived at a slightly higher rate than did E. coli O157:H7.
Under refrigeration, all five isolates were similar to E. coli
O157:H7. No difference was found between the isolates and
E. coli O157:H7 during fermentation, and the isolates
survived at higher levels than E. coli O157:H7, adding an
additional layer of security from overprediction (8).
On 20 September 2011, the USDA-FSIS (17) published
a proposed rule for Shiga toxin–producing E. coli (STEC) in
certain raw beef products, declaring six E. coli serotypes
(O26, O45, O103, O111, O121, and O145) of non-O157
STEC as adulterants of nonintact raw beef products. Because
of this new rule, validation of the ability of an antimicrobial
intervention to control these bacteria is very important.
A decontamination step is often implemented before
packaging of subprimal meat cuts to control recontamination
that might have occurred during fabrication. Limited
published information is available on the validation of
chemical antimicrobial treatments applied in a spray cabinet
to chilled subprimals. Lactic acid is a common antimicrobial
J. Food Prot., Vol. 75, No. 9
agent used to reduce bacterial populations on meat. However,
its ability to reduce bacterial populations on reworked
subprimals needs to be validated. The objective of this study
was to validate the use of lactic acid as an initial subprimal
intervention and as a secondary intervention on reworked
product under industry operating parameters. Each operating
parameter was evaluated separately to determine its effec-
tiveness and to provide validating information to the meat
industry. Because of the proposed requirements for HACCP
validation by the USDA-FSIS (16), a study listing the effects
of each parameter on various microorganisms is needed for
the meat industry. The draft guidance states that ‘‘Care
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021
should be taken to ensure that the supporting validation
documentation is sufficiently related to the process, product
and hazard identified in the hazard analysis. . . . To be
effective, the process procedures should relate and adhere to
the specifications in the supporting documentation. If the
documentation listed a particular critical parameter such as
concentration of an antimicrobial, that concentration should
be used in the process. . . . If, for example, the process
specifications described in the supporting documentation are
not implemented in the same or similar enough way in the
establishment’s process, additional research studies need to
be conducted and documented to ensure the modified
implementation achieves the desired result.’’ The present
study was conducted to evaluate the effect of each parameter
combination and the process as a whole to fulfill this
requirement.
Limited published information exists on non-O157
STEC. Data are needed on the effectiveness of antimicrobial
interventions currently used by the meat industry against
non-O157 STEC and on whether the antimicrobial effects
obtained are similar to those obtained for E. coli O157:H7
or the biotype I E. coli surrogates. Therefore, this study also
included a comparison of the ability of chemical interven-
tions to reduce E. coli O157:H7, non-O157 STEC, and the
biotype I E. coli surrogates when applied to chilled
subprimals using industry parameters. The information
obtained should be extremely useful as a validation tool
for industry systems.
MATERIALS AND METHODS
Bacterial strains and preparation of inocula. Three inocula
were used in this study. One comprised five E. coli O157:H7
strains (ATCC 43888, ATCC 43895, C1-057, C1-072, and C1-
109) isolated from bovine feces (3). The second comprised five
strains of nonpathogenic E. coli (ATCC BAA-1427, ATCC BAA-
1428, ATCC BAA-1429, ATCC BAA-1430, and ATCC BAA-
1431) that have been used as surrogates for E. coli O157:H7 (9).
The third comprised 12 strains of non-O157 STEC (i.e., 2 strains
each of serotypes O26, O45, O103, O111, O121, and O145)
(Table 1). The non-O157 STEC strains were originally provided
by Michigan State University. Rifampin-resistant cultures of each
strain were developed, based on the method described by Kaspar
and Tamplin (7), to allow selection and differentiation of inoculum
populations from natural flora associated with meat. The rifampin-
resistant strains were individually cultured and subcultured (35uC
for 24 ¡ 2 h) in 10 ml of tryptic soy broth (Difco, BD, Sparks,
MD) supplemented with 100 mg/ml rifampin (Sigma-Aldrich Inc.,
St. Louis, MO). Cell cultures (10 ml) of each strain were harvested
J. Food Prot., Vol. 75, No. 9 LACTIC ACID INTERVENTION FOR E. COLI O157:H7, STEC, AND SURROGATES
TABLE 1. Strains included in the non-O157 STEC inoculum
Non-O157 STEC serotype Source
O26:H11 Human
O45:NM Human
O103:HN Human
O111:NM Human
O121 Human
O121:H19 Human
O145:H28 Human
O26:H11 Cattle
O145:NM Cattle
O45:NM Cattle
O103:N Cattle
O111 Cattle
individually by centrifugation (4,629 | g for 15 min at 4uC)
(Eppendorf model 5810 R, Brinkmann Instruments Inc., Westbury,
NY), washed with 10 ml of phosphate-buffered saline (PBS,
pH 7.4: 0.2 g/liter KH2PO4, 1.5 g/liter Na2HPO4?H2O, 8.0 g/liter
NaCl, and 0.2 g/liter KCl), recentrifuged, and suspended in PBS to
obtain 8 log CFU/ml. Inocula were then prepared by adding 10 ml
of the respective strains within each inoculum type together and
vortexing vigorously for 2 min.
Sample inoculation and treatment. Two types of subpri-
mals were used in the study: beef round peeled knuckle (IMPS
167A) and beef brisket flats (IMPS 120A). Chilled subprimals
were obtained before postchill antimicrobial intervention at three
different times, but postharvest aging was held constant (48 to 72 h
postmortem). The subprimals were portioned into five sections
with 100 cm2 of exposed lean surface and were spot inoculated
(50 ml of inoculant cocktail on both sides of sections) to a target
level of approximately 6 log CFU/cm2, with the pathogenic E. coli
O157:H7 mixture, the pathogenic non-O157 STEC mixture, or the
surrogate nonpathogenic E. coli mixture. Sterile PBS was placed
on a fourth set of sections (50 ml on both sides of sections) assigned
as the noninoculated controls to mimic the inoculated sections.
Samples were held at 4uC for 30 min to allow bacterial cell
attachment.
Uninoculated and inoculated beef knuckle or brisket sections
were treated with lactic acid in a custom-built spray cabinet with a
stainless steel conveyor belt (Chad Co., Olathe, KS) at 1 of 16
combinations of two lactic acid concentrations (2.0 or 5.0%), two
lactic acid temperatures (22 or 48uC), two pressures (1.03 or 4.83
bar), and two flow rates (0.22 or 6.22 liters/min). Concentrated
lactic acid (88% L-lactic acid; Purac, Linconshire, IL) was mixed
with tap water to the desired concentrations. The pH values of the
2.5% lactic acid spray were 2.25 and 3.23 at 25 and 55uC,
respectively. The pH values of the 5.0% lactic acid sprays were
2.04 and 3.06 at 25 and 55uC, respectively. Lactic acid was applied
to all sections via four 1/8 MEG 2510 WashJet HSS nozzles (Spray
Systems, Wheaton, IL) above the sample and three of the same
nozzles below. Sections were allowed to drip for 10 s and were
then vacuum packaged in vacuum bags (20.3 by 35.6 cm; Cryovac,
Duncan, SC), sealed in a single chamber vacuum packager
(Hollymatic, Countryside, IL), and stored at 4uC until bacteria
were enumerated or a second lactic acid treatment was applied 24 h
after the first application.
Microbiological analyses. Sections (100 cm2) from each
subprimal (beef knuckle or brisket) were sampled before
inoculation to determine initial total bacterial populations. Sections
1703
were placed in 625-ml filter bags (19 by 30 cm; Nasco Whirl-Pak,
Modesto, CA), 100 ml of D/E neutralizing broth (Difco, BD) was
added, and the mixture was pummeled (Masticator, IUL Industries,
Barcelona, Spain) for 2 min. Sample homogenates were serially
diluted 10-fold in 0.1% buffered peptone water (Difco, BD), and
appropriate dilutions were surface plated on tryptic soy agar (TSA;
Difco, BD). After inoculation and before the initial lactic acid
treatment, samples collected from the subprimal sections were
surface plated on TSA plus rifampin (100 mg/ml) for enumeration
of rifampin-resistant E. coli inoculum populations, thus determin-
ing the inoculation level. Samples collected immediately after the
initial lactic acid treatment but before the second treatment (i.e.,
after vacuum-packaged storage at 4uC) and those collected after the
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021
second treatment were plated on TSA and TSA plus rifampin
(100 mg/ml) for enumeration of total bacterial populations and
rifampin-resistant E. coli inoculum populations, respectively.
Samples also were collected from the uninoculated treated sections
in the same way and surface plated on TSA to enumerate total
bacteria. Samples were incubated at 37uC for 18 to 24 h before
enumeration. The detection limit for this method was 1.0 log CFU/
cm2.
Statistical analysis. Samples were analyzed in triplicate, and
the entire experiment was conducted twice, for a total of six
samples per treatment. Main effects of inoculum, subprimal,
sampling time, and temperature, pressure, rate, and concentration
of lactic acid application were analyzed. After initial analysis,
temperature, pressure, rate, and concentration of lactic acid
application were analyzed together as the application parameter.
All interactions between inoculum, subprimal, time, and applica-
tion parameter were analyzed. Statistical analysis was conducted
using the general linear model of SAS version 9.2 (12). Differences
were considered statistically significant when P values were less
than 0.05.
RESULTS AND DISCUSSION
For the natural microflora on the uninoculated samples,
the mean (¡standard deviation) total plate count (TPC) was
3.5 ¡ 0.2 log CFU/cm2 across all samples. After the first
lactic acid treatment, the TPC from the uninoculated
samples was reduced to 2.4 ¡ 0.3 log CFU/cm2 (Fig. 1).
Similarly, the TPC before the second treatment was 3.4 ¡
0.2 log CFU/cm2 and was reduced to 2.3 ¡ 0.3 log CFU/
cm2 across all samples (Fig. 1). Some differences (P ,
0.05) were found between samples treated under different
application parameters. After the first application of lactic
acid with various parameters, TPCs were 0.9 ¡ 0.3 to 1.5
¡ 0.2 log CFU/cm2 lower than the starting counts for
samples (Table 2). Similarly, after the second application of
lactic acid, TPCs were 1.0 ¡ 0.1 to 1.2 ¡ 0.5 log CFU/cm2
lower than the starting counts, depending on the application
parameter (Table 3). After treatment, the TPCs mirrored the
total inoculum counts (TICs), indicating that the majority of
the bacteria enumerated in the TPCs were the artificially
inoculated bacteria. After the initial spray application of
lactic acid, mean TICs were reduced from 6.0 log CFU/cm2
before treatment to 3.6 ¡ 0.1, 4.4 ¡ 0.1, and 4.4 ¡ 0.1 log
CFU/cm2 for the E. coli surrogates, E. coli O157:H7, and
non-O157 STEC inoculation groups, respectively, after
treatment (Table 4). After the first lactic acid treatment,
there was a significant difference (P , 0.05) of 0.1 ¡ 0.0
1704 PITTMAN ET AL. J. Food Prot., Vol. 75, No. 9

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021

FIGURE 1. Differences in total plate counts after the first and second lactic acid application on sections of chilled beef briskets and
peeled beef knuckles inoculated with cocktails of Escherichia coli surrogates, E. coli O157:H7, or Shiga toxin–producing E. coli (STEC) at
initial levels of 6.0 log CFU/cm2.

and 0.1 ¡ 0.0 log CFU/cm2 on different subprimals for the
TPC and TIC, respectively. Although the difference was
significant, it was not meaningful on a microbiological scale (7).
Similarly, differences (P , 0.05) were observed in
TICs when lactic acid was applied at different parameters.
There was no difference in TPCs for the uninoculated
samples after the first application when lactic acid was
applied at different concentrations or pressures. However, a
difference of 0.1 ¡ 0.0 log CFU/cm2 was observed when
lactic acid was applied at different rates, and the same
difference in TPCs was observed when lactic acid was
applied at different temperatures. The small differences (0.0
to 0.1 ¡ 0.0 log CFU/cm2) between counts at different
application parameters were significant but not large enough
to affect differences in microbial growth (10).
For the E. coli surrogates and E. coli O157:H7, TICs
were 0.1 ¡ 0.0 and 0.1 ¡ 0.0 log CFU/cm2 lower,
respectively, when lactic acid was applied at 5.0 versus
2.0% (Table 4). Conversely, non-O157 STEC TICs were
0.1 ¡ 0.1 log CFU/cm2 lower when lactic acid was applied
at 2.0 versus 5.0%. After the first application at different
pressures, the E. coli surrogates and non-O157 STEC TICs
were 0.2 ¡ 0.1 log CFU/cm2 higher when lactic acid was
applied at 4.83 versus 1.03 bar. When the first lactic acid
treatment was applied at 1.03 bar, E. coli O157:H7 TIC was
0.1 ¡ 0.0 log CFU/cm2 higher than that after application at
4.83 bar. Similar differences were obtained with different
application rates. TICs were 0.2 ¡ 0.1, 0.1¡ 0.0, and 0.0
¡ 0.0 log CFU/cm2 higher when lactic acid was applied at
6.22 versus 0.22 liters/min (Table 4). Likewise, application

TABLE 2. Total plate counts recovered on tryptic soy agar from uninoculated subprimal samples treated with lactic acid a
Mean ¡ SD total plate counts (log CFU/cm2)

First application Second application

Lactic acid Application rate
concn (%) Temp (uC) Pressure (bar) (liters/min) Before treatment After treatment Before treatment After treatment

2.0 22 1.03 0.22 3.6 ¡ 0.2 C 2.4 ¡ 0.3 AB 3.6 ¡ 0.1 B 2.6 ¡ 0.1 B
4.83 0.22 3.4 ¡ 0.2 E 2.3 ¡ 0.3 B 3.3 ¡ 0.1 D 2.2 ¡ 0.2 F
1.03 6.22 3.3 ¡ 0.1 F 2.4 ¡ 0.2 A 3.3 ¡ 0.1 D 2.2 ¡ 0.2 F
4.83 6.22 3.6 ¡ 0.1 C 2.4 ¡ 0.3 AB 3.5 ¡ 0.2 BC 2.5 ¡ 0.2 C

48 1.03 0.22 3.8 ¡ 0.2 A 2.3 ¡ 0.2 B 3.8 ¡ 0.2 A 2.8 ¡ 0.2 A
4.83 0.22 3.2 ¡ 0.2 G 2.3 ¡ 0.3 B 3.0 ¡ 0.3 F 1.9 ¡ 0.5 G
1.03 6.22 3.6 ¡ 0.1 C 2.4 ¡ 0.3 AB 3.5 ¡ 0.2 BC 2.5 ¡ 0.2 C
4.83 6.22 3.5 ¡ 0.1 D 2.3 ¡ 0.3 B 2.9 ¡ 0.4 F 1.7 ¡ 0.5 H

5.0 22 1.03 0.22 3.6 ¡ 0.1 C 2.4 ¡ 0.3 AB 3.4 ¡ 0.2 C 2.4 ¡ 0.3 D
4.83 0.22 3.5 ¡ 0.1 D 2.3 ¡ 0.3 B 3.2 ¡ 0.1 E 2.2 ¡ 0.2 F
1.03 6.22 3.6 ¡ 0.3 BC 2.4 ¡ 0.2 A 3.6 ¡ 0.3 B 2.6 ¡ 0.3 B
4.83 6.22 3.6 ¡ 0.1 C 2.4 ¡ 0.3 AB 3.2 ¡ 0.3 E 2.1 ¡ 0.5 F

48 1.03 0.22 3.7 ¡ 0.2 B 2.3 ¡ 0.2 B 3.4 ¡ 0.3 C 2.3 ¡ 0.3 E
4.83 0.22 3.7 ¡ 0.2 B 2.3 ¡ 0.3 B 3.4 ¡ 0.4 C 2.3 ¡ 0.5 DE
1.03 6.22 3.3 ¡ 0.1 F 2.4 ¡ 0.3 AB 3.2 ¡ 0.1 E 2.2 ¡ 0.2 F
4.83 6.22 3.4 ¡ 0.1 E 2.3 ¡ 0.3 B 3.3 ¡ 0.2 D 2.3 ¡ 0.2 E

a
Within a column, means with different letters are significantly different (P , 0.05).
 J. Food Prot., Vol. 75, No. 9

TABLE 3. Total plate counts recovered on tryptic soy agar from inoculated subprimal samples treated with lactic acid a
Mean ¡ SD E. coli surrogates (log CFU/cm2) Mean ¡ SD E. coli O157:H7 (log CFU/cm2) Mean ¡ SD non-O157 STEC (log CFU/cm2)

Lactic acid Pressure Application After first Before second After second After first Before second After second After first Before second After second
concn (%) Temp (uC) (bar) rate (liters/min) application application application application application application application application application

2.0 22 1.03 0.22 4.1 ¡ 0.1 A 3.4 ¡ 0.1 F 2.6 ¡ 0.0 E 4.4 ¡ 0.1 C 3.9 ¡ 0.0 G 2.9 ¡ 0.1 G 4.3 ¡ 0.0 D 4.4 ¡ 0.0 C 3.6 ¡ 0.0 C
4.83 0.22 3.6 ¡ 0.2 E 3.8 ¡ 0.2 B 2.9 ¡ 0.1 B 4.4 ¡ 0.1 C 4.4 ¡ 0.3 E 3.2 ¡ 0.2 F 4.2 ¡ 0.1 E 4.3 ¡ 0.1 D 3.6 ¡ 0.0 C
1.03 6.22 3.5 ¡ 0.0 F 3.6 ¡ 0.0 D 2.8 ¡ 0.0 C 4.4 ¡ 0.1 C 4.6 ¡ 0.1 C 3.3 ¡ 0.1 E 4.4 ¡ 0.2 C 4.4 ¡ 0.2 C 3.7 ¡ 0.1 B
4.83 6.22 3.6 ¡ 0.1 E 3.7 ¡ 0.1 C 2.8 ¡ 0.1 C 4.6 ¡ 0.1 A 4.7 ¡ 0.1 B 3.5 ¡ 0.2 C 4.2 ¡ 0.1 E 4.3 ¡ 0.1 D 3.6 ¡ 0.0 C

48 1.03 0.22 3.2 ¡ 0.2 I 3.5 ¡ 0.1 E 2.7 ¡ 0.1 D 4.3 ¡ 0.1 D 4.5 ¡ 0.0 D 3.3 ¡ 0.1 E 4.3 ¡ 0.1 D 4.3 ¡ 0.1 D 3.6 ¡ 0.0 C
4.83 0.22 3.7 ¡ 0.1 D 3.8 ¡ 0.0 B 2.9 ¡ 0.0 B 4.3 ¡ 0.0 D 4.5 ¡ 0.0 D 3.3 ¡ 0.0 E 4.3 ¡ 0.2 D 4.4 ¡ 0.2 C 3.6 ¡ 0.0 C
1.03 6.22 3.3 ¡ 0.1 H 3.5 ¡ 0.1 E 2.7 ¡ 0.0 D 3.9 ¡ 0.2 F 4.3 ¡ 0.1 F 2.7 ¡ 0.3 H 3.9 ¡ 0.2 H 3.9 ¡ 0.1 H 3.1 ¡ 0.1 E
4.83 6.22 3.8 ¡ 0.1 C 3.9 ¡ 0.1 A 3.0 ¡ 0.1 A 4.4 ¡ 0.1 C 4.6 ¡ 0.1 C 3.4 ¡ 0.2 D 4.0 ¡ 0.1 G 4.0 ¡ 0.1 G 3.1 ¡ 0.0 E

5.0 22 1.03 0.22 3.8 ¡ 0.2 C 3.9 ¡ 0.1 A 3.0 ¡ 0.1 A 4.4 ¡ 0.1 C 4.6 ¡ 0.1 C 3.4 ¡ 0.1 D 4.6 ¡ 0.0 B 4.6 ¡ 0.0 B 3.8 ¡ 0.0 A
4.83 0.22 3.7 ¡ 0.2 D 3.8 ¡ 0.1 B 2.9 ¡ 0.1 B 4.4 ¡ 0.1 C 4.6 ¡ 0.1 C 3.4 ¡ 0.1 D 4.6 ¡ 0.1 B 4.6 ¡ 0.1 B 3.8 ¡ 0.0 A
1.03 6.22 3.6 ¡ 0.0 E 3.7 ¡ 0.0 C 2.9 ¡ 0.0 B 4.2 ¡ 0.2 E 4.6 ¡ 0.3 BC 3.9 ¡ 0.1 A 4.7 ¡ 0.1 A 4.7 ¡ 0.1 A 3.8 ¡ 0.0 A
4.83 6.22 3.7 ¡ 0.1 D 3.8 ¡ 0.1 B 2.9 ¡ 0.0 B 4.3 ¡ 0.1 D 5.0 ¡ 0.1 A 2.9 ¡ 0.1 G 4.6 ¡ 0.1 B 4.6 ¡ 0.1 B 3.8 ¡ 0.0 A

48 1.03 0.22 3.4 ¡ 0.1 G 3.6 ¡ 0.1 D 2.8 ¡ 0.0 C 4.5 ¡ 0.1 B 4.7 ¡ 0.1 B 3.5 ¡ 0.1 C 4.6 ¡ 0.1 B 4.6 ¡ 0.1 B 3.8 ¡ 0.0 A
4.83 0.22 3.7 ¡ 0.1 D 3.8 ¡ 0.0 B 2.9 ¡ 0.0 B 4.6 ¡ 0.1 A 4.7 ¡ 0.1 B 3.6 ¡ 0.1 B 4.6 ¡ 0.1 B 4.7 ¡ 0.1 A 3.8 ¡ 0.0 A
1.03 6.22 3.5 ¡ 0.1 F 3.7 ¡ 0.1 C 2.8 ¡ 0.0 C 4.6 ¡ 0.1 A 4.7 ¡ 0.1 B 3.6 ¡ 0.1 B 4.1 ¡ 0.2 F 4.2 ¡ 0.2 E 3.6 ¡ 0.0 C
4.83 6.22 3.9 ¡ 0.1B 3.9 ¡ 0.1 A 3.0 ¡ 0.1 A 4.5 ¡ 0.0 B 4.7 ¡ 0.0 B 3.5 ¡ 0.0 C 4.1 ¡ 0.1 F 4.1 ¡ 0.1 F 3.2 ¡ 0.0 D

a
Within a column, means with different letters are significantly different (P , 0.05).
LACTIC ACID INTERVENTION FOR E. COLI O157:H7, STEC, AND SURROGATES
1705

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021

 1706
PITTMAN ET AL.

TABLE 4. Total inoculum counts recovered on tryptic soy agar plus rifampin from inoculated subprimal samples treated with lactic acid a
Mean ¡ SD E. coli surrogates (log CFU/cm2) Mean ¡ SD E. coli O157:H7 (log CFU/cm2) Mean ¡ SD non-O157 STEC (log CFU/cm2)

Lactic acid Pressure Application After first Before second After second After first Before second After second After first Before second After second
concn (%) Temp (uC) (bar) rate (liters/min) application application application application application application application application application

2.0 22 1.03 0.22 3.3 ¡ 0.2 G 4.0 ¡ 0.1 A 2.9 ¡ 0.1 A 4.3 ¡ 0.1 C 4.4 ¡ 0.1 D 3.2 ¡ 0.1 D 4.4 ¡ 0.0 C 4.4 ¡ 0.0 C 3.5 ¡ 0.0 C
4.83 0.22 3.6 ¡ 0.2 DE 3.5 ¡ 0.1 F 2.5 ¡ 0.1 C 4.3 ¡ 0.1 C 4.4 ¡ 0.1 D 3.2 ¡ 0.2 D 4.4 ¡ 0.1 C 4.4 ¡ 0.1 C 3.5 ¡ 0.1 C
1.03 6.22 3.6 ¡ 0.2 DE 3.6 ¡ 0.2 DE 2.5 ¡ 0.2 CD 4.6 ¡ 0.0 A 4.3 ¡ 0.2 DE 3.6 ¡ 0.0 A 4.4 ¡ 0.3 BC 4.4 ¡ 0.2 BC 3.5 ¡ 0.2 BC
4.83 6.22 3.8 ¡ 0.1 B 3.4 ¡ 0.1 G 2.4 ¡ 0.1 C 4.3 ¡ 0.1 C 4.5 ¡ 0.1 C 3.2 ¡ 0.1 D 4.4 ¡ 0.2 C 4.4 ¡ 0.2 BC 3.5 ¡ 0.1 C

48 1.03 0.22 3.6 ¡ 0.1 E 3.6 ¡ 0.1 E 2.6 ¡ 0.1 B 4.4 ¡ 0.0 B 4.3 ¡ 0.1 E 3.2 ¡ 0.0 D 4.3 ¡ 0.1 D 4.4 ¡ 0.1 C 3.5 ¡ 0.1 C
4.83 0.22 3.7 ¡ 0.1 C 3.6 ¡ 0.1 E 2.6 ¡ 0.1 B 4.4 ¡ 0.1 B 4.3 ¡ 0.1 E 3.2 ¡ 0.1 D 4.4 ¡ 0.1 C 4.5 ¡ 0.1 B 3.6 ¡ 0.1 B
1.03 6.22 3.7 ¡ 0.0 C 3.5 ¡ 0.1 F 2.5 ¡ 0.1 C 4.4 ¡ 0.1 B 4.3 ¡ 0.1 E 3.3 ¡ 0.1 C 4.2 ¡ 0.2 E 4.2 ¡ 0.2 D 3.4 ¡ 0.1 D
4.83 6.22 4.0 ¡ 0.1 A 3.8 ¡ 0.2 C 2.8 ¡ 0.2 A 4.4 ¡ 0.2 BC 4.5 ¡ 0.1 C 3.3 ¡ 0.2 CD 4.5 ¡ 0.1 B 4.5 ¡ 0.1 B 3.6 ¡ 0.1 B

5.0 22 1.03 0.22 3.2 ¡ 0.1 G 3.7 ¡ 0.1 CD 2.6 ¡ 0.1 B 4.2 ¡ 0.1 E 4.6 ¡ 0.1 B 3.1 ¡ 0.2 E 4.5 ¡ 0.0 B 4.5 ¡ 0.0 B 3.6 ¡ 0.0 B
4.83 0.22 3.5 ¡ 0.2 DEF 3.6 ¡ 0.1 E 2.6 ¡ 0.1 B 4.3 ¡ 0.1 C 4.6 ¡ 0.1 B 3.1 ¡ 0.2 E 4.5 ¡ 0.1 B 4.5 ¡ 0.1 B 3.6 ¡ 0.1 B
1.03 6.22 3.6 ¡ 0.2 DE 3.7 ¡ 0.1 CD 2.7 ¡ 0.1 B 4.6 ¡ 0.0 A 4.5 ¡ 0.1 C 3.5 ¡ 0.0 B 4.5 ¡ 0.2 B 4.5 ¡ 0.2 AB 3.6 ¡ 0.1 B
4.83 6.22 3.7 ¡ 0.1 CD 3.5 ¡ 0.1 F 2.5 ¡ 0.1 C 4.3 ¡ 0.1 C 4.7 ¡ 0.1 A 3.1 ¡ 0.1 E 4.5 ¡ 0.1 B 4.5 ¡ 0.1 B 3.6 ¡ 0.1 B

48 1.03 0.22 3.5 ¡ 0.1 F 3.7 ¡ 0.1 CD 2.7 ¡ 0.1 B 4.3 ¡ 0.0 D 4.5 ¡ 0.1 C 3.2 ¡ 0.0 D 4.5 ¡ 0.1 B 4.5 ¡ 0.1 B 3.6 ¡ 0.1 B
4.83 0.22 3.6 ¡ 0.1 E 3.7 ¡ 0.1 CD 2.7 ¡ 0.1 B 4.3 ¡ 0.1 C 4.5 ¡ 0.1 C 3.2 ¡ 0.2 D 4.5 ¡ 0.1 B 4.6 ¡ 0.1 A 3.6 ¡ 0.0 B
1.03 6.22 3.7 ¡ 0.0 C 3.6 ¡ 0.1 E 2.6 ¡ 0.1 B 4.4 ¡ 0.1 B 4.5 ¡ 0.1 C 3.2 ¡ 0.1 D 4.4 ¡ 0.1 C 4.4 ¡ 0.1 C 3.5 ¡ 0.1 C
4.83 6.22 3.5 ¡ 0.2 DEF 3.9 ¡ 0.1 B 2.8 ¡ 0.1 A 4.4 ¡ 0.2 BC 4.6 ¡ 0.0 B 3.2 ¡ 0.3 CDE 4.6 ¡ 0.1 A 4.6 ¡ 0.1 A 3.7 ¡ 0.1 A

a
Rifampin was added to the agar at 100 mg/ml. Within a column, means with different letters are significantly different (P , 0.05).
J. Food Prot., Vol. 75, No. 9

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021

J. Food Prot., Vol. 75, No. 9 LACTIC ACID INTERVENTION FOR E. COLI O157:H7, STEC, AND SURROGATES
of lactic acid at different temperatures resulted in TICs that
were 0.1¡ 0.0, 0.0 ¡ 0.0, and 0.0 ¡ 0.0 log CFU/cm2
higher after the first application at 48 versus 22uC for the E.
coli surrogates, E. coli O157:H7, and non-O157 STEC
inoculation groups, respectively (Table 4). During storage at
4uC, TICs decreased by 0.1 ¡ 0.1, 0.2 ¡ 0. 1, and 0.0 ¡
0.1 log CFU/cm2 for the E. coli surrogates, E. coli O157:H7,
and non-O157 STEC inoculation groups, respectively,
compared with the TICs before storage.
After the second lactic acid application, TICs were 2.6
¡ 0.1, 3.2 ¡ 0.1, and 3.6 ¡ 0.1 log CFU/cm2 for the E. coli
surrogates, E. coli O157:H7, and non-O157 STEC inocula-
tion groups, respectively. However, lower counts were
obtained when lactic acid was applied at 5.0 versus 2.0%.
After the first and second lactic acid applications, a
significant difference was observed when lactic acid was
applied at different parameters. However, because the
differences in the TIC after treatment were 0.3 and 0.2 log
CFU/cm2 for the first and second application, respectively,
the effects of changes in application parameters was not
considered microbiologically meaningful (10). Similarly,
the differences between inoculants were significant but not
meaningful because they were less than 1.0 log CFU/cm2
(10). The difference between subprimal types was signifi-
cant (P , 0.05) but also not meaningful on a microbiolog-
ical scale. After the first lactic acid application, TICs for
beef brisket samples were slightly higher than TICs for beef
knuckles (3.7 and 3.6 log CFU/cm2, respectively). After the
second application, TICs for the beef brisket samples were
0.04 CFU/cm2 higher than those for the beef knuckle
sections, a statistically significant (P , 0.05) but microbi-
ologically meaningless difference.
Bacon et al. (2) found that lactic acid treatment, which
is used as a whole carcass intervention, resulted in minimal
reductions (,0.5 log CFU/100 cm2) of TPCs, total coliform
counts, and E. coli counts from initial mean values of 5.7,
3.8, and 3.3 log CFU/100 cm2, respectively, on top sirloin
butts. In contrast, at the application parameters used in the
present study, lactic acid reduced counts of all three inocula
and the uninoculated background flora. Bacon et al. (2)
found that counts of E. coli O157:H7 were reduced from 5.8
to 4.7 log CFU/g and from 4.2 to 2.7 log CFU/g for a high-
and low-level inoculation group, respectively, when lactic
acid was applied at 2% and 55uC. Similar results were
observed in the present study when lactic acid was applied
at 1 of 16 combinations of two concentrations (2.0 and
5.0%), two temperatures (22 and 48uC), two pressures (1.03
and 4.83 bar), and two flow rates (0.22 and 6.22 liters/min)
to beef brisket and knuckle sections. Heller et al. (6)
inoculated beef outside rounds with a three-strain cocktail
of E. coli O157:H7 and applied lactic acid at 55uC at two
concentrations, 2.5 and 5.0%. E. coli O157:H7 mean
populations were reduced by 0.9 to 1.1 log CFU/100 cm2
from 3.6 log CFU/100 cm2. The authors did not find any
differences between results of applied treatments.
After the initial lactic acid treatment in our study, the
surrogate inoculum of nonpathogenic E. coli had a lower
TIC than both of the pathogenic inoculants. This result was
most likely due to the fact that the surrogates were
1707
developed for use on hot beef carcass tissue but were used
in this study on chilled beef subprimals and to the use of
rifampin-resistant derivatives of the strains. However, the
behavior of the E. coli surrogates still closely resembled that
of the other inocula as a result of the lactic acid action on the
chilled beef subprimals. Because the counts on the sections
inoculated with the surrogates were lower after the initial
lactic acid application, the counts after the second
application were also lower.
Regardless of the combination of application parame-
ters, lactic acid was an effective antimicrobial. Both the
initial and secondary lactic acid treatments resulted in lower
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021
after-treatment TPCs and TICs. A second lactic acid spray,
applied as a rework procedure to product treated previously
with lactic acid, can be applied with the same operating
parameters regardless of the type of subprimal being treated.
Although the lactic acid treatments did not reduce microbial
populations to below detectable limits, the double lactic acid
spray treatment was effective as a primary and rework
intervention against E. coli O157:H7 and non-O157 STEC.
These data would be useful to industry as part of the
HACCP validation process.
REFERENCES
1. Bacon, R. T., K. E. Belk, J. N. Sofos, R. P. Clayton, J. O. Reagan,
and G. C. Smith. 2000. Microbial populations on animal hides and
beef carcasses at different stages of slaughter in plants employing
multiple-sequential interventions for decontamination. J. Food Prot.
63:1080–1086.
2. Bacon, R. T., J. N. Sofos, K. E. Belk, and G. C. Smith. 2002.
Commercial application of lactic acid to reduce bacterial populations
on chilled beef carcasses, subprimal cuts and table surfaces during
fabrication. Dairy Food Environ. Sanit. 22:674–682.
3. Carlson, B. A., K. K. Nightingale, G. L. Mason, J. R. Ruby, W. T.
Choat, G. H. Loneragan, G. C. Smith, J. N. Sofos, and K. E. Belk.
2009. Escherichia coli O157:H7 strains that persist in feedlot cattle
are genetically related and demonstrate an enhanced ability to adhere
to intestinal epithelial cells. Appl. Environ. Microbiol. 75:5927–5937.
4. Gill, C. O., M. Badoni, and J. C. McGinnis. 1999. Assessment of the
adequacy of cleaning of equipment used for breaking beef carcasses.
Int. J. Food Microbiol. 46:1–8.
5. Gill, C. O., J. C. McGinnis, and J. Bryant. 2000. Contamination of
beef chucks with Escherichia coli during carcass breaking. J. Food
Prot. 64:1824–1827.
6. Heller, C. E., J. A. Scanga, J. N. Sofos, K. E. Belk, W. Warren-Serna,
G. R. Bellinger, R. T. Bacon, M. L. Rossman, and G. C. Smith.
2007. Decontamination of beef subprimal cuts intended for blade
tenderization or moisture enhancement. J. Food Prot. 70:1176–1180.
7. Kaspar, C. W., and M. L. Tamplin. 1993. Effects of temperature and
salinity on the survival of Vibrio vulnificus in seawater and shellfish.
Appl. Environ. Microbiol. 59:2425–2429.
8. Keeling, C., S. E. Niebuhr, G. R. Acuff, and J. S. Dickson. 2009.
Evaluation of Escherichia coli biotype 1 as a surrogate for
Escherichia coli O157:H7 for cooking, fermentation, freezing, and
refrigerated storage in meat processes. J. Food Prot. 72:728–732.
9. Marshall, K. M., S. E. Niebuhr, G. R. Acuff, L. M. Lucia, and J. S.
Dickson. 2005. The identification of Escherichia coli O157:H7 meat
processing indicators for fresh meat through the comparison of the
effects of selected antimicrobial interventions. J. Food Prot. 68:
2580–2586.
10. National Advisory Committee on Microbiological Criteria for Foods.
2009. Parameters for determining inoculated pack/challenge study
protocols. J. Food Prot. 73:140–202.
11. Palumbo, S. A. 1986. Is refrigeration enough to restrain foodborne
pathogens? J. Food Prot. 49:1003–1009.
1708 PITTMAN ET AL.
12. SAS Institute. 2011. SAS version 9.2 (TS2M3). SAS Institute Inc.,
Cary, NC.
13. Sofos, J. N., K. E. Belk, and G. C. Smith. 1999. Processes to
reduce contamination with pathogenic microorganisms in meat, p.
596–605. In Proceedings of the 45th International Congress of
Meat Science and Technology, Yokohama, Japan, 1 to 6 August
1999.
14. Upmann, M., P. Jakob, and G. Reuter. 2000. Microbial transfer
during cutting and deboning of pork in a small-scale meat processing
plant. Dairy Food Environ. Sanit. 20:14–23.
J. Food Prot., Vol. 75, No. 9
15. U.S. Department of Agriculture, Food Safety and Inspection Service.
1996. Pathogen reduction: hazard analysis and critical control point
(HACCP) system. Fed. Regist. 61:38805–38989.
16. U.S. Department of Agriculture, Food Safety and Inspection Service.
2010. Draft guidance: HACCP systems validation, p. 6–28. Available at:
http://www.fsis.usda.gov/PDF/HACCP_Validation_ltrs.pdf. Accessed 14
November 2011.
17. U.S. Department of Agriculture, Food Safety and Inspection Service.
2011. Shiga toxin–producing Escherichia coli in certain raw beef
products. Fed. Regist. 76:58157–58165.
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/75/9/1701/1686771/0362-028x_jfp-11-520.pdf by Colombia user on 15 February 2021