97

Journal of Food Protection, Vol. 78, No. 1, 2015, Pages 97–103

doi:10.4315/0362-028X.JFP-14-041
Copyright G, International Association for Food Protection

Bacteriophage Cocktail for Biocontrol of Salmonella in

Dried Pet Food
SERENA HEYSE,1 LEIGH FARRIS HANNA,1 JOELLE WOOLSTON,2 ALEXANDER SULAKVELIDZE,2
AND DUANE CHARBONNEAU1*

1The Procter & Gamble Company, 8700 Mason-Montgomery Road, Mason, Ohio 45040; and 2Intralytix, Inc., 701 East Pratt Street, Baltimore,

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021

Maryland 21202, USA

MS 14-041: Received 22 January 2014/Accepted 1 September 2014

ABSTRACT
Human salmonellosis has been associated with contaminated pet foods and treats. Therefore, there is interest in identifying
novel approaches for reducing the risk of Salmonella contamination within pet food manufacturing environments. The use of lytic
bacteriophages shows promise as a safe and effective way to mitigate Salmonella contamination in various food products.
Bacteriophages are safe, natural, highly targeted antibacterial agents that specifically kill bacteria and can be targeted to kill food
pathogens without affecting other microbiota. In this study, we show that a cocktail containing six bacteriophages had a broad-
spectrum activity in vitro against a library of 930 Salmonella enterica strains representing 44 known serovars. The cocktail was
effective against 95% of the strains in this tested library. In liquid culture dose-ranging experiments, bacteriophage cocktail
concentrations of $108 PFU/ml inactivated more than 90% of the Salmonella population (101 to 103 CFU/ml). Dried pet food
inoculated with a mixture containing equal proportions of Salmonella serovars Enteritidis (ATCC 4931), Montevideo (ATCC
8387), Senftenberg (ATCC 8400), and Typhimurium (ATCC 13311) and then surface treated with the six-bacteriophage cocktail
($2.5 ¡ 1.5 | 106 PFU/g) achieved a greater than 1-log (P , 0.001) reduction compared with the phosphate-buffered saline–
treated control in measured viable Salmonella within 60 min. Moreover, this bacteriophage cocktail reduced natural
contamination in samples taken from an undistributed lot of commercial dried dog food that tested positive for Salmonella. Our
results indicate that bacteriophage biocontrol of S. enterica in dried pet food is technically feasible.

In the United States, about one in six people contract a
foodborne illness each year (12). Among the five most
significant foodborne pathogens, nontyphoidal Salmonella
is the second most frequent domestically acquired food-
borne illness and is responsible for some of the most serious
health outcomes (12). Although the vast majority of cases of
salmonellosis in humans are associated with food consump-
tion, human contact with pet foods has been identified as a
potential source of infection (15). In 2004 to 2005, an
outbreak of human Salmonella Thompson infections in
Washington State and Western Canada was linked to
animal-derived raw pet treats (1, 2). A subsequent multistate
outbreak of human Salmonella Schwarzengrund infections
in 2006 to 2007 was connected to dried dog food from a
single manufacturer (3, 8) and led to a product recall (4, 5).
In 2012, the U.S. Food and Drug Administration conducted
a nationwide survey of Salmonella contamination of dried
pet foods, pet treats, and supplements to determine the
prevalence, serotypes, genetic fingerprints, and antibiotic
resistance profiles of the organisms and, if necessary, to
remove contaminated products from commerce (42).
Because the risk of Salmonella contamination in pet food
* Author for correspondence. Tel: 513-622-2896; Fax: 513-622-5578;
E-mail: charbonneau.dl@pg.com.
is of concern, developing improved methods to detect and
reduce the presence of Salmonella in commercial pet foods
is important.
Research and commercial interest in bacteriophage-
mediated control of foodborne pathogens has accelerated in
recent years, and significant progress has been made in their
development as an effective, natural, environmentally friendly
technology to improve food safety (17, 30, 37). Bacterio-
phages are the most abundant organisms on earth (about 1031
particles compared with 107 humans or 1030 bacteria) and are
present in food (26), water (9), and the environment, where
they play a vital role in the earth’s microbial ecology (13, 16).
It is estimated that bacteriophages are responsible for killing
20 to 40% of the total population of marine bacteria on a daily
basis (39). Additionally, bacteriophages are normal and
ubiquitous constituents of the human microbiota. Bacterio-
phages can be found in the oral cavity and the human
gastrointestinal tract, where they are an integral part of the
ecology of the gut microbiota (32).
Lytic bacteriophages, which infect, lyse, and kill their
specific bacterial hosts, are well suited to biocontrol
purposes. They are highly targeted, replicating agents that
control the pathogens of interest without affecting other,
often beneficial, constituents of the bacterial flora (26).
Bacteriophage preparations have also recently gained
98 HEYSE ET AL.
regulatory acceptance for commercial application to pre-
vent foodborne pathogens in the U.S. marketplace. These
include bacteriophages that control Listeria monocytogenes
(ListShield, Listex) (38, 40), Escherichia coli O157:H7
(EcoShield and Finalyse) (23, 43), and Salmonella (Salmo-
Fresh) (24, 44, 46).
Our laboratories are investigating the safety and effective-
ness of bacteriophage biocontrol of Salmonella contamination
of dried pet food products. In this article, we report on (i) the
activity of a bacteriophage cocktail against a spectrum of
Salmonella enterica strains; (ii) the efficacy of the bacterio-
phage cocktail against low Salmonella populations; and (iii) its
effectiveness in reducing Salmonella contamination of labora-
tory-inoculated dried pet food as well as naturally contaminated
commercial dried pet food.
MATERIALS AND METHODS
Bacteria and bacteriophages. A library of 930 strains of S.
enterica representing more than 44 known serovars was obtained
from a variety of research and public health laboratories. All strains
were stored at 280uC in 30% glycerol and in 70% Luria-Bertani
broth (EMD Chemicals, Gibbstown, NJ) or 10 g/liter tryptone (BD,
Franklin Lakes, NJ), 5 g/liter yeast extract (BD), and 10 g/liter NaCl
(Morton Salt Inc., Chicago, IL). The bacteriophage cocktail
(SalmoLyse) and its component monophages (SBA-1781, SPT-1,
SSE-121, STML-198, STML-13-1, and SKML-39) were obtained
from Intralytix, Inc. (Baltimore, MD) and were refrigerated at 4uC
and protected from light, prior to use.
Range of activity of the bacteriophage cocktail and
individual monophages against S. enterica strains. Individual
S. enterica strains from the above-referenced library were grown in
Luria-Bertani broth at 35 ¡ 2uC with 150 ¡ 25 rpm agitation to
early log phase (optical density at 600 nm of 0.3), and then 0.1-ml
aliquots were added to 5 ml of top agar (Luria-Bertani broth plus
0.7% agar) and were plated onto individual Luria agar plates. Once
the agar solidified, 10 ml of the bacteriophage cocktail or the
component monophages was spotted on the bacterial lawn at a
concentration of 1.25 ¡ 0.75 | 109 PFU/ml and allowed to dry.
Plates were incubated at 35 ¡ 2uC for 24 ¡ 4 h. The appearance
of smaller plaques to confluent lysis within the spot of phage
application indicated Salmonella strain susceptibility to either the
cocktail or the monophage.
Genomic screening of the component monophages. DNA
samples from the constituent monophages in the bacteriophage
cocktail were screened for potentially undesirable genetic sequences,
including genes identified in the Code of Federal Regulations, 40
CFR part 725.421 (14), and antibiotic resistance genes and 16S
bacterial ribosomal RNA genes. Genome sequences were obtained
by 454 sequencing at SeqWright, Inc. (Houston, TX). Analysis of
the assembled genomes for undesirable sequences was also
completed at SeqWright, Inc., by identifying candidate open reading
frame (ORF) sequences with National Center for Biotechnology
Information (NCBI) ORF Finder (http://www.ncbi.nlm.nih.gov/
projects/gorf/). Annotation of the candidate ORFs was completed
with NCBI BLASTX (http://blast.ncbi.nlm.nih.gov/Blast.cgi) using
the nonredundant protein database. Results with an E-value of less
than 1024 were screened for alignment with an undesirable sequence
by a key word search for results matching the defined sequence
names provided in the Code of Federal Regulations, words matching
16S bacterial ribosomal RNA encoding sequences (search words
J. Food Prot., Vol. 78, No. 1
‘‘16S,’’ ‘‘RNA’’), or antibiotic resistance encoding sequences
(search words ‘‘antibiotic,’’ ‘‘drug,’’ ‘‘mycin,’’ ‘‘resist,’’ ‘‘trans-
port,’’ or ‘‘pump’’). All annotated sequences identified by the key
word search were manually reviewed for predicted protein function
to identify the presence or absence of any undesirable genetic
sequences.
Influence of host and bacteriophage cocktail concentra-
tions on the inactivation of Salmonella enterica Typhimurium
in liquid culture. Stationary-phase cultures of Salmonella
Typhimurium (ATCC 13311) grown overnight (16 ¡ 4 h) at 37
¡ 2uC with 150 rpm agitation were 10-fold serially diluted into
tryptic soy broth (BD, Franklin Lakes, NJ), resulting in bacterial
populations ranging from approximately 101 to 103 CFU/ml. The
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021
bacteriophage cocktail was 10-fold serially diluted in 0.9% saline
to obtain concentrations of 1.25 ¡ 0.75 | 105 to 1010 PFU/ml.
Five milliliters of each Salmonella population of 101 to 103 CFU/
ml was mixed with 5 ml of the bacteriophage cocktail dilutions
ranging from 1.25 ¡ 0.75 | 105 to 1010 PFU/ml. In total, 18
individual combinations of bacteriophage and Salmonella Typhi-
murium dilutions were examined. Mixtures were incubated at room
temperature (approximately 22uC), and samples were taken 60 min
postmixing. Surviving Salmonella Typhimurium pathogens were
enumerated by quantitative plating on tryptic soy agar (TSA). With
bacterial populations of 101 and 102, cells were concentrated on a
0.2-mm-pore-size filter (MicroFunnel, Pall Co., Port Washington,
NY) that was then transferred to a TSA plate for enumeration.
Experiments were carried out in triplicate, on two independent
occasions, and the results were averaged.
Efficacy of the bacteriophage cocktail on dried pet food
inoculated with various strains of S. enterica. Samples of dried
pet food (kibble) were inoculated at approximately 1 | 104 CFU/g
with a mixture of equal proportions of Salmonella serovars
Enteritidis (ATCC 4931), Montevideo (ATCC 8387), Senftenberg
(ATCC 8400), and Typhimurium (ATCC 13311). From overnight
(16 ¡ 4 h) cultures grown at 37uC with 150 rpm agitation, sterile
10-fold dilutions were performed to achieve a final stock population
of approximately 1 | 106 CFU/ml. Populations of Salmonella were
verified by enumeration using standard microbiological plate count
on TSA. Two-milliliter portions of each of these stock populations
were mixed together, and 7 ml of 0.9% saline was added to the
mixed culture to prepare the kibble inoculum. This mixed
Salmonella culture was added to 150 g of dog kibble (5 to 6.5 mm
thick and 10 to 12.5 mm in diameter) over five separate additions of
0.9 ml (4.5 ml total added). Before each addition, the kibble sample
was mixed to ensure even Salmonella distribution over the surface of
the kibble. Inoculated kibble was allowed to equilibrate for 60 min at
room temperature prior to bacteriophage treatment. The bacterio-
phage cocktail (initial concentration 1.25 ¡ 0.75 | 1010 PFU/ml)
was diluted 10-fold in 0.9% saline to prepare dilutions of 1.25 ¡
0.75 | 107, 108, and 109 PFU/ml. The inoculated kibble was then
treated with a surface spray application of either the bacteriophage
cocktail (to attain a final concentration of 2.5 ¡ 1.5 | 105, 106, or
107 PFU/g) or with phosphate-buffered saline (PBS; negative
control) followed by thorough mixing. Treated samples were
incubated at room temperature for 1 h prior to enumeration. Samples
(25 g) were added to 225 ml of buffered peptone water and then
stomached for 2 min at 260 rpm, and the Salmonella population was
quantified with the rapid mini modified semisolid Rappaport-
Vassiliadis most-probable-number (MPN) assay as described by
Shashidhar et al. (34), modified so that positive wells were
confirmed with the MicroSEQ 16S rDNA identification system for
Salmonella detection (Life Technologies, Carlsbad, CA) (6, 7).
J. Food Prot., Vol. 78, No. 1 BACTERIOPHAGE COCKTAIL FOR BIOCONTROL OF SALMONELLA
Experiments were completed in duplicate, on three independent
occasions, and the results were averaged.
Efficacy of bacteriophage cocktail on noncommercially
distributed kibble containing Salmonella. Noncommercially
distributed kibble that tested positive for Salmonella was treated
with the bacteriophage cocktail, using an air-atomized spray
application while the kibble was mixing in a 40-liter batch mixer.
Kibble batches of 12.5 kg were treated with 250 ml of the
bacteriophage cocktail dilutions 1.25 ¡ 0.75 | 108, 109, or 1010
PFU/ml over a period of 2 min to obtain a final dosage of either
2.5 ¡ 1.5 | 106, 107, or 108 PFU/g. Gamma-irradiated kibble
served as a sterilized control. For this purpose, samples were
treated with a minimum of 10 kGy at a contract facility (Sterigenics
International, Inc., Westerville, OH). Fifty 25-g random samples of
kibble were collected from each treatment and separately enriched
overnight (20 ¡ 4 h) at 37uC in 225 ml of buffered peptone water.
Following enrichment, samples were assayed for detectable
Salmonella using the MicroSEQ Salmonella spp. assay (7).
Statistical analysis. A linear regression model was used to
conduct an analysis of variance for the CFU data obtained from the
influence of host and bacteriophage cocktail concentrations on the
inactivation of S. enterica in liquid culture experiments. The
dependent variable was log of Salmonella (CFU per milliliter) for
each bacteriophage-treated sample minus log of Salmonella for the
corresponding saline-treated sample; zero counts were replaced with
0.5. Main effects used were day, starting Salmonella concentration,
bacteriophage concentration, and replicate (all categorical). Interac-
tion terms in the model were starting Salmonella concentration by
day and by bacteriophage concentration.
Estimation of the MPNs and log reductions for the inoculated
dry pet food and the contaminated undistributed commercial lots
data was completed using the Poisson binomial model formulation
as described in Jarvis et al. (25). Statistical significance and
confidence limits were determined using the alpha ~ 0.05 level.
The statistical impact of bacteriophage treatment on a
contaminated commercial lot was analyzed using standard methods
for proportions and the Jonckheere-Terpstra test for trend (22) with
significance determined using the alpha ~ 0.05 level.
RESULTS
Activity of a Salmonella bacteriophage cocktail
against a library of 930 S. enterica strains and genomic
screening of the component monophages. Table 1 depicts
the susceptibility of 930 strains of S. enterica, consisting of
44 known serovars, to the lytic activity of the bacteriophage
cocktail or the component monophages found in the cocktail
at a concentration of 1.25 ¡ 0.75 | 109 PFU/ml. At this
concentration of the bacteriophage cocktail, lysis was
observed on 882 (95%) of the Salmonella strains tested.
The genomes of the cocktail monophages were screened for,
and found to be free of, potentially undesirable sequences,
including toxins, antibiotic resistance genes, and bacterial
sequences (see ‘‘Materials and Methods’’).
Influence of host and bacteriophage cocktail con-
centrations on the inactivation of Salmonella Typhimur-
ium in liquid culture. Bacterial contamination in foods
processed under modern sanitary protocols, when identified,
is likely at a very low level. In order to be effective at
controlling the low levels of contamination, sufficient levels
99
of bacteriophage must be applied to ensure contact with the
targeted bacteria (20). To evaluate the concentration of
bacteriophage cocktail required to inactivate low levels of
Salmonella Typhimurium (ATCC 13311) under laboratory
conditions, dose-ranging experiments were completed in
liquid culture at bacterial populations ranging from 101 to 103
CFU/ml by applying serial dilutions of the bacteriophage
cocktail. At all concentrations of Salmonella Typhimurium
examined, greater than 90% inactivation was achieved at
bacteriophage levels of 108 PFU/ml or greater after a 1-h
incubation (Fig. 1).
Efficacy of the bacteriophage cocktail on dried pet
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021
food inoculated with four strains of Salmonella. When
individual samples of dried kibble were inoculated with a
mixture of four strains of Salmonella (Typhimurium ATCC
13311, Senftenberg ATCC 8400, Montevideo ATCC 8387,
and Enteritidis ATCC 4931) and then treated by a surface
spray application of the bacteriophage cocktail, a dose
response in the efficacy of the cocktail on the reduction of
the Salmonella population was observed (Fig. 2). At a
bacteriophage treatment concentration of 2.5 ¡ 1.5 | 105,
106, and 107 PFU/g, a decrease in the Salmonella population
of 0.75 ¡ 0.13, 1.37 ¡ 0.14, and 2.05 ¡ 0.17 log MPN/g
was obtained in comparison to the PBS-treated controls. The
reductions (versus PBS) observed in the Salmonella
populations are statistically significant (P , 0.001) for each
bacteriophage treatment concentration examined. In addition,
the reductions across the bacteriophage treatments were
significantly different from each other (P , 0.03).
Efficacy of the bacteriophage cocktail on Salmonel-
la in an undistributed lot of contaminated commercial
dried pet food. Batches from an undistributed lot of kibble
contaminated with Salmonella were spray treated with three
different concentrations of the bacteriophage cocktail and then
analyzed for residual contamination compared with an
untreated control group and a gamma-sterilized control group.
Twenty-eight percent of the 50 samples of untreated kibble had
detectable Salmonella, whereas all irradiated samples were
negative (Fig. 3). Relative to the untreated controls, each
concentration of bacteriophage cocktail tested (2.5 ¡ 1.5 |
106, 107, and 108 PFU/g) reduced the percentage of
Salmonella-positive samples in each treatment group to 20,
16, and 14%, respectively. The decreasing trend in the number
of positive samples with an increasing concentration of
bacteriophage treatment is statistically significant as determined
by a Jonckheere-Terpstra test for trend (P ~ 0.03). Also, using
a Poisson binomial model formulation to calculate the MPN
value per 25-g sample indicated that each bacteriophage
cocktail concentration examined decreased the Salmonella
population present, with an MPN/25 g value of 0.33 for the
untreated control sample versus 0.22, 0.17, and 0.15 for
bacteriophage cocktail concentrations of 2.5 ¡ 1.5 | 106, 107,
and 108 PFU/g, respectively (Table 2).
DISCUSSION
We report here for the first time that a cocktail of six
naturally derived, lytic bacteriophages with broad-scale
100 HEYSE ET AL. J. Food Prot., Vol. 78, No. 1

TABLE 1. Range of activity of the bacteriophage cocktail and its individual monophage components against a collection of Salmonella
enterica strains a
No. of strains susceptible to each respective bacteriophage
Total
Serotype no. of strains SPT-1 SSE-121 SBA-1781 STML-198 SKML-39 STML-13-1 Cocktail

Agona 11 6 7 5 9 0 1 11
Alachua 5 1 3 1 4 1 0 4
Braenderup 1 1 0 1 1 0 0 1
Brandenburg 2 1 0 0 1 0 2 2
Bredeney 2 2 0 2 0 0 0 2
Choleraesuis 1 1 0 1 0 0 0 1
Dublin 1 0 0 0 0 0 0 0
Enteritidis 137 127 119 25 66 3 126 137

Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021

Gallinarum 3 2 0 2 1 0 2 3
Georgia 19 16 2 15 17 0 17 17
Grampian 8 0 1 0 2 5 0 8
Hadar 116 52 47 56 105 28 17 115
Havana 2 0 0 0 1 0 0 1
Heidelberg 41 30 26 9 3 0 29 41
Indiana 1 0 0 0 1 0 1 1
Infantis 5 4 2 5 5 0 0 5
Javiana 1 1 0 1 0 0 0 1
Kedougou 1 0 1 1 1 0 0 1
Kentucky 25 3 3 11 1 16 0 25
Larochelle 2 0 0 0 2 0 0 2
Liverpool 2 0 1 0 2 0 0 2
Livingstone 5 2 2 2 5 0 0 5
Mbandaka 3 0 0 0 3 0 0 3
Meleagridis 2 0 0 0 1 0 0 1
Montevideo 4 1 0 2 1 0 0 2
Muenster 1 0 0 0 0 0 0 0
Newport 27 22 21 17 23 18 0 27
Othmarschen 1 1 1 1 1 0 0 1
Paratyphi B 1 0 0 0 0 0 0 0
Poona 1 0 1 0 0 0 0 1
Putten 1 0 0 0 0 0 0 0
Reading 5 3 0 0 5 0 4 5
Rissen 1 0 0 0 0 0 0 0
Schwarzengrund 5 4 4 4 4 0 2 4
Senftenberg 8 5 5 4 6 0 0 8
Stanley 1 1 0 1 1 0 0 1
Tennessee 2 1 1 1 2 0 0 2
Thompson 6 1 5 4 3 0 1 6
Tilburg 1 0 0 0 1 0 0 1
Typhi 2 0 0 0 2 0 0 2
Typhimurium 183 125 114 120 77 0 161 183
Virchow 2 0 0 0 2 0 0 2
West Hampton 1 1 1 1 1 0 0 1
Worthington 2 1 1 1 2 0 0 2
Unknown 279 145 156 108 122 22 76 245
All 930 560 524 401 484 93 439 882
a
A spot test was performed by applying 10 ml of 109 PFU/ml on bacterial lawns of individual Salmonella enterica strains on agar and
recording zones of lysis.

effectiveness against Salmonella significantly reduced Sal-
monella populations on laboratory-inoculated dried pet food
and also was effective in reducing Salmonella in an
undistributed lot of contaminated commercial dried pet food.
Bacteriophage treatment shows significant promise for
reducing foodborne pathogens in various foods. Bacterio-
phages target and kill only their specific bacterial hosts;
moreover, they do not alter the taste, appearance, or smell of
foods, nor do they have corrosive effects on environmental
surfaces or manufacturing equipment. The efficacy of
bacteriophage treatment to reduce contamination by food-
borne bacteria has been demonstrated experimentally for a
variety of raw (21, 27, 35, 36) and ready-to-eat (18, 19)
foods, and several bacteriophage preparations have recently
gained regulatory approval in the United States for food
contact applications (41).
J. Food Prot., Vol. 78, No. 1 BACTERIOPHAGE COCKTAIL FOR BIOCONTROL OF SALMONELLA
FIGURE 1. Influence of the concentrations of the host and
bacteriophage on the inactivation of Salmonella Typhimurium in
liquid culture. The average percent survival with standard error
bars is represented.
Bacteriophages are not motile and must reach their
targets by diffusion or direct application; thus, the efficiency
of bacteriophage infection and bacterial lysis in food will
depend on the host concentration, the characteristics of the
food matrix, the concentrations of bacteriophage applied,
and the bacteriophage application technique employed. Of
these variables, bacteriophage concentration and the mode
of application can be manipulated in manufacturing plants.
For inactivation of small Salmonella populations ranging
from 101 to 103 CFU/ml in liquid culture, we found a
threshold bacteriophage concentration of approximately 108
PFU/ml (Fig. 1). Likewise, other investigators have report-
ed that inactivation of Salmonella by bacteriophage P7 was
independent of bacterial populations within this range at
bacteriophage levels of approximately 5 | 108 PFU/ml
(10). These observations suggest that biocontrol of low
numbers of pathogens may be possible by using sufficiently
high bacteriophage levels.
FIGURE 2. Efficacy of the bacteriophage cocktail surface applied
at a total bacteriophage concentration of 2.5 ¡ 1.5 | 105, 106, or
107 PFU/g onto dried pet food inoculated with a mixture of equal
proportions of Salmonella serovars Enteritidis (ATCC 4931),
Montevideo (ATCC 8387), Senftenberg (ATCC 8400), and Typhimur-
ium (ATCC 13311). The average log MPN per gram with standard
error bars is shown. Dark gray bars indicate the average log MPN
per gram with PBS treatment, and light gray bars indicate the
average log MPN per gram with bacteriophage cocktail treatment.
101
FIGURE 3. Effect of the bacteriophage cocktail treatment on the
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021
number of Salmonella-positive samples among groups of 50 drawn
randomly from naturally contaminated commercial dried pet food.
As proof of concept in pet food, a greater than 1-log
reduction in S. enterica contamination was achieved by
surface treating laboratory-inoculated dried pet kibble with
the bacteriophage cocktail at a concentration $2.5 ¡ 1.5 |
106 PFU/g. The fact that this concentration of bacteriophage
is lower than that required to observe a similar reduction in
the S. enterica population in the liquid culture dose-ranging
experiment is likely due to the Salmonella population being
concentrated on the outer surface of the kibble, which
provides a smaller relative surface area where the bacterio-
phage must localize in order to attach to the bacterial target.
Several recent studies have demonstrated the efficacy of
various bacteriophage preparations against a diverse array of
Salmonella-contaminated food matrices, including fruits
(27), sausages (18, 45), cheese (33), turkey deli meat (18),
seafood (18), eggs (18), and milk (18, 29). Our study expands
on those previous studies and demonstrates that bacterio-
phages also can be effective in significantly reducing
Salmonella levels in dried pet food. We further demonstrated
that surface application of the bacteriophage cocktail to dried
kibble (2.5 ¡ 1.5 | 106, 107, and 108 PFU/g), from an
undistributed, contaminated commercial lot that tested
positive for Salmonella, reduced the number of positive
kibble samples (Fig. 3), thus demonstrating that the lytic
bacteriophage cocktail lowered the prevalence and concen-
tration of Salmonella observed in commercial production.
A version of the bacteriophage cocktail examined here,
SalmoFresh (Intralytix, Inc., Baltimore, MD), has been
designated generally recognized as safe (GRAS) for direct
application onto poultry, fish and shellfish, and processed
fruits and vegetables (GRAS notice no. GRN 000435). No
deleterious effects on indigenous flora or health are
expected. Animal studies of bacteriophage administration
support this conclusion (28, 32). Moreover, the gut is
regularly exposed to lytic bacteriophages in food and water.
Notably, prior to the advent of modern antibiotics,
bacteriophage therapy was administered to treat bacterial
infections (reviewed in (11)) and has maintained a long
tradition in Eastern Europe. A recent metagenomic analysis
of a commercial bacteriophage therapy from a Russian
manufacturer revealed 18 distinct bacteriophage types with
no undesirable genes, and a small human volunteer trial
102 HEYSE ET AL.
TABLE 2. Impact of the bacteriophage cocktail treatment on the MPN value of Salmonella-positive samples from naturally contaminated
commercial dried pet food
Treatment (control or
bacteriophage concn [PFU/g]) MPN/25 g sample
Untreated 0.33
106 0.22
107 0.17
108 0.15
found no adverse effects associated with oral bacteriophage
exposure (31).
In conclusion, we have demonstrated that bacteriophage
biocontrol of Salmonella on dried kibble is technically
feasible. The treatment process is simple in that it involves
surface spraying and mixing of the treated foodstuff to
achieve bacterial inactivation. Therefore, bacteriophages
could serve as an effective additional tool for reducing
Salmonella contamination risk in commercial pet food.
ACKNOWLEDGMENTS
The authors acknowledge Deborah Hutchins of Hutchins &
Associates LLC (Cincinnati, OH) for technical writing assistance and
Dan Schnell of Procter & Gamble (Cincinnati, OH) for statistical analysis.
J. Woolston and A. Sulakvelidze hold an equity stake in Intralytix, Inc., a
Maryland corporation developing bacteriophage preparations (including
SalmoLyse) for various applications.
REFERENCES
1. Anonymous. 2006. Human salmonellosis associated with animal-
derived pet treats—United States and Canada, 2005. Morb. Mortal.
Wkly. Rep. 55:702–705.
2. Anonymous. 2006. An international outbreak of human salmonellosis
associated with animal-derived pet treats—Canada and Washington
State, 2005. Can. Commun. Dis. Rep. 32:150–155.
3. Anonymous. 2008. Multistate outbreak of human Salmonella
infections caused by contaminated dry dog food—United States,
2006–2007. Morb. Mortal. Wkly. Rep. 57:521–524.
4. Anonymous. 2008. Mars recalls pet food because of potential link
with human illness. J. Am. Vet. Med. Assoc. 233:1198.
5. Anonymous. 2008. Update: recall of dry dog and cat food products
associated with human Salmonella Schwarzengrund infections—
United States, 2008. Morb. Mortal. Wkly. Rep. 57:1200–1202.
6. Balachandran, P., Y. Cao, L. Wong, M. R. Furtado, O. V.
Petrauskene, and R. S. Tebbs. 2011. Evaluation of applied biosystems
MicroSEQ real-time PCR system for detection of Salmonella spp. in
food. J. AOAC Int. 94:1106–1116.
7. Balachandran, P., M. Friberg, V. Vanlandingham, K. Kozak, A.
Manolis, M. Brevnov, E. Crowley, P. Bird, D. Goins, M. R. Furtado,
O. V. Petrauskene, R. S. Tebbs, and D. Charbonneau. 2012. Rapid
detection of Salmonella in pet food: design and evaluation of
integrated methods based on real-time PCR detection. J. Food Prot.
75:347–352.
8. Behravesh, C. B., A. Ferraro, M. Deasy III, V. Dato, M. Moll, C. Sandt,
N. K. Rea, R. Rickert, C. Marriott, K. Warren, V. Urdaneta, E. Salehi,
E. Villamil, T. Ayers, R. M. Hoekstra, J. L. Austin, S. Ostroff, and I. T.
Williams. 2010. Human Salmonella infections linked to contaminated
dry dog and cat food, 2006–2008. Pediatrics 126:477–483.
9. Bergh, O., K. Y. Borsheim, G. Bratbak, and M. Heldal. 1989. High
abundance of viruses found in aquatic environments. Nature 340:
467–468.
10. Bigwood, T., J. A. Hudson, and C. Billington. 2009. Influence of host
and bacteriophage concentrations on the inactivation of food-borne
pathogenic bacteria by two phages. FEMS Microbiol. Lett. 291:59–
64.
J. Food Prot., Vol. 78, No. 1
95% lower confidence limit 95% upper confidence limit
0.19 0.56
0.12 0.42
0.09 0.35
0.07 0.32
11. Campbell, A. 2003. The future of bacteriophage biology. Nat. Rev.
Genet. 4:471–477.
12. Centers for Disease Control and Prevention. 2011. 2011 Estimates of
foodborne illness in the United States. Available from: http://www.
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021
cdc.gov/Features/dsFoodborneEstimates/. Accessed 3 January 2014.
13. Chibani-Chennoufi, S., A. Bruttin, M. L. Dillmann, and H. Brussow.
2004. Phage-host interaction: an ecological perspective. J. Bacteriol.
186:3677–3686.
14. Code of Federal Regulations. Title 40. Reporting requirements and
review processes for microorganisms. General exemptions for new
microorganisms. Introduced genetic material, part 725.421. U.S.
Government Printing Office, Washington, DC. Available from: http://
www.ecfr.gov/cgi-bin/text-idx?c~ecfr&SID~41f4e8877ad4dbc9472
bebb6da815be2&rgn~div8&view~text&node~40:32.0.1.1.13.7.1.3&
idno~40. Accessed 30 August 2013.
15. Finley, R., R. Reid-Smith, and J. S. Weese. 2006. Human health
implications of Salmonella-contaminated natural pet treats and raw
pet food. Clin. Infect. Dis. 42:686–691.
16. Fuhrman, J. A. 1999. Marine viruses and their biogeochemical and
ecological effects. Nature 399:541–548.
17. Greer, G. G. 2005. Bacteriophage control of foodborne bacteria. J.
Food Prot. 68:1102–1111.
18. Guenther, S., O. Herzig, L. Fieseler, J. Klumpp, and M. J. Loessner.
2012. Biocontrol of Salmonella Typhimurium in RTE foods with the
virulent bacteriophage FO1-E2. Int. J. Food Microbiol. 154:66–72.
19. Guenther, S., D. Huwyler, S. Richard, and M. J. Loessner. 2009.
Virulent bacteriophage for efficient biocontrol of Listeria monocyto-
genes in ready-to-eat foods. Appl. Environ. Microbiol. 75:93–100.
20. Hagens, S., and M. J. Loessner. 2010. Bacteriophage for biocontrol of
foodborne pathogens: calculations and considerations. Curr. Pharm.
Biotechnol. 11:58–68.
21. Higgins, J. P., S. E. Higgins, K. L. Guenther, W. Huff, A. M.
Donoghue, D. J. Donoghue, and B. M. Hargis. 2005. Use of a
specific bacteriophage treatment to reduce Salmonella in poultry
products. Poult. Sci. 84:1141–1145.
22. Hollander, M., and D. A. Wolfe. Nonparametric statistical methods,
2nd ed. John Wiley and Sons, New York.
23. Intralytix, Inc. 8 February 2011. Intralytix receives FDA regulatory
clearance for phage-based E. coli technology [press release].
Available at: http://www.intralytix.com/Intral_News_PR020811.
htm. Accessed 29 September 2013.
24. Intralytix, Inc. 27 February 2013. Intralytix wins regulatory approval
for phage-based food safety product effective against Salmonella
[press release]. Available at: http://www.intralytix.com/Intral_News_
PR022713.htm. Accessed 23 January 2014.
25. Jarvis, B., C. Wilrich, and P. T. Wilrich. 2010. Reconsideration of the
derivation of most probable numbers, their standard deviations,
confidence bounds and rarity values. J. Appl. Microbiol. 109:1660–
1667.
26. Kennedy, J. E., Jr., and G. Bitton. 1987. Bacteriophages in food,
p. 289–316. In S. M. Goyal, C. P. Gerba, and G. Bitton (ed.), Phage
ecology. John Wiley and Sons, New York.
27. Leverentz, B., W. S. Conway, Z. Alavidze, W. J. Janisiewicz, Y.
Fuchs, M. J. Camp, E. Chighladze, and A. Sulakvelidze. 2001.
Examination of bacteriophage as a biocontrol method for Salmonella
on fresh-cut fruit: a model study. J. Food Prot. 64:1116–1121.
28. Mai, V., M. Ukhanova, L. Visone, T. Abuladze, and A. Sulakvelidze.
2010. Bacteriophage administration reduces the concentration of
J. Food Prot., Vol. 78, No. 1 BACTERIOPHAGE COCKTAIL FOR BIOCONTROL OF SALMONELLA
Listeria monocytogenes in the gastrointestinal tract and its translo-
cation to spleen and liver in experimentally infected mice. Int. J.
Microbiol. 2010:624234.
29. Marti, R., K. Zurfluh, S. Hagens, J. Pianezzi, J. Klumpp, and M. J.
Loessner. 2013. Long tail fibres of the novel broad-host-range T-even
bacteriophage S16 specifically recognize Salmonella OmpC. Mol.
Microbiol. 87:818–834.
30. Maura, D., and L. Debarbieux. 2011. Bacteriophages as twenty-first
century antibacterial tools for food and medicine. Appl. Microbiol.
Biotechnol. 90:851–859.
31. McCallin, S., S. Alam Sarker, C. Barretto, S. Sultana, B. Berger,
S. Huq, L. Krause, R. Bibiloni, B. Schmitt, G. Reuteler, and H.
Brussow. 2013. Safety analysis of a Russian phage cocktail: from
MetaGenomic analysis to oral application in healthy human subjects.
Virology 443:187–196.
32. Mills, S., F. Shanahan, C. Stanton, C. Hill, A. Coffey, and R. P. Ross.
2013. Movers and shakers: influence of bacteriophages in shaping the
mammalian gut microbiota. Gut Microbes 4:4–16.
33. Modi, R., Y. Hirvi, A. Hill, and M. W. Griffiths. 2001. Effect of
phage on survival of Salmonella enteritidis during manufacture and
storage of cheddar cheese made from raw and pasteurized milk.
J. Food Prot. 64:927–933.
34. Shashidhar, R., I. Srivastava, and J. R. Bandekar. 2011. Quantifica-
tion of Salmonella in food samples from India using the MINI-MSRV
MPN and modified MINI-MSRV MPN methods. J. Food Sci. 76:
M564–M567.
35. Soni, K. A., and R. Nannapaneni. 2010. Bacteriophage significantly
reduces Listeria monocytogenes on raw salmon fillet tissue. J. Food
Prot. 73:32–38.
36. Soni, K. A., R. Nannapaneni, and S. Hagens. 2010. Reduction of
Listeria monocytogenes on the surface of fresh channel catfish fillets
by bacteriophage Listex P100. Foodborne Pathog. Dis. 7:427–434.
37. Sulakvelidze, A. 2013. Using lytic bacteriophages to eliminate or
significantly reduce contamination of food by foodborne bacterial
pathogens. J. Sci. Food Agric. 93:3137–3146.
38. Sulakvelidze, A., and G. Pasternack. 2010. Industrial and regulatory
issues in bacteriophage applications in food production and
103
processing, p. 297–326. In P. M. Sabour and M. W. Griffiths (ed.),
Bacteriophages in the control of food- and waterborne pathogens.
ASM Press, Washington, DC.
39. Suttle, C. A. 2005. Viruses in the sea. Nature 437:356–361.
40. U.S. Department of Agriculture. 2009. Safe and suitable ingredients
used in production of meat and poultry products. USDA Food Safety
and Inspection Service (FSIS) directive 7120.1, Amend. 19. Available
at: http://www.fsis.usda.gov/OPPDE/rdad/FSISDirectives/7120.1Amend19.
pdf. Accessed 28 August 2013.
41. U.S. Food and Drug Administration. 2006. Food additives permitted
for direct addition to food for human consumption; bacteriophage
preparation, p. 47729–47732. 21 CFR Part 172. Docket no. 2002F-
0316 (formerly 02F-0316). Available at: http://edocket.access.gpo.
gov/2006/E6-13621.htm. Accessed 28 August 2013.
42. U.S. Food and Drug Administration. 24 October 2011. FY 2012
nationwide assignment to collect and analyze samples of pet foods, pet
Downloaded from http://meridian.allenpress.com/jfp/article-pdf/78/1/97/1688497/0362-028x_jfp-14-041.pdf by Indonesia user on 26 March 2021
treats, and supplements for pets from interstate commerce in the United
States for Salmonella [memorandum]. Available at: http://www.fda.
gov/AnimalVeterinary/Products/AnimalFoodFeeds/Contaminants/ucm
277264.htm. Accessed 30 August 2013.
43. U.S. Food and Drug Administration. 2011. Inventory of effective
food contact substance (FCS) notifications. FCN no. 1018. Available
at: http://www.accessdata.fda.gov/scripts/fcn/fcnDetailNavigation.
cfm?rpt~fcslisting&id~1018. Accessed 29 September 2013.
44. U.S. Food and Drug Administration. 2013. GRAS notice inventory.
GRN no. 435. Available at: http://www.accessdata.fda.gov/scripts/
fcn/fcnNavigation.cfm?filter~435&sortColumn~&rpt~graslisting.
Accessed 29 September 2013.
45. Whichard, J. M., N. Sriranganathan, and F. W. Pierson. 2003.
Suppression of Salmonella growth by wild-type and large-plaque
variants of bacteriophage Felix O1 in liquid culture and on chicken
frankfurters. J. Food Prot. 66:220–225.
46. Woolston, J., A. R. Parks, T. Abuladze, B. Anderson, M. Li, C.
Carter, L. F. Hanna, S. Heyse, D. Charbonneau, and A. Sulakvelidze.
2013. Bacteriophages lytic for Salmonella rapidly reduce Salmonella
contamination on glass and stainless steel surfaces. Bacteriophage 3:
e25697.