Reduction of Salmonella in ground chicken using a bacteriophage

Reduction of Salmonella in ground chicken using a bacteriophage Ar’Quette Grant,∗ Salina Parveen,∗,1 Jurgen Schwarz,∗ Fawzy Hashem,∗ and Bob Vimini† ∗

Department of Agriculture, Food, and Resource Sciences, University of Maryland Eastern Shore, Princess Anne, MD 21853; and † Perdue Foods, LLC, Salisbury, MD 21801 was observed when the bacteriophage was diluted in sterile tap water than in sterile filtered water: 0.39 Log CFU/cm2 and 0.23 Log CFU/cm2 reduction after 30 min, respectively (P < 0.05). The non-GC isolates showed reductions of 0.71 Log CFU/cm2 and 0.90 Log CFU/cm2 after 30 min and 8 h, respectively (P < 0.05). The GC isolates were less sensitive to the bacteriophage: 0.39 Log CFU/cm2 and 0.67 Log CFU/cm2 reductions after 30 min and 8 h, respectively (P < 0.05). In conclusion, bacteriophage reduction was dependent on water used to dilute the bacteriophage, Salmonella’s susceptibility to the bacteriophage, and treatment time.

ABSTRACT This study’s goal was to ascertain the effectiveness of a commercially available Salmonella bacteriophage during ground chicken production focusing on: water source, different Salmonella serovars, and time. Salmonella-free boneless, skinless chicken meat was inoculated with 4.0 Log CFU/cm2 of either a cocktail of 3 Salmonella isolates derived from ground chicken (GC) or a cocktail of 3 Salmonella strains not isolated from ground chicken (non-GC). Bacteriophages were spread onto the chicken using sterile tap or filtered water for 30 min or 8 h. Salmonella was recovered using standard plating method. Greater Salmonella reduction

Key words: Salmonella, Salmonelex, Bacteriophage, Ground Chicken 2017 Poultry Science 0:1–8 http://dx.doi.org/10.3382/ps/pex062

INTRODUCTION

S. Typhimurium, S. Newport, S. Heidelberg, and S. Enteritidis are on the list of top 10 Salmonella serovars isolated from meat and poultry (USDA-FSIS, 2015c), and S. Typhimurium has been isolated from ground chicken (Fratamico, 2003; Whichard et al., 2007; White et al., 2001) along with S. Kentucky, S. Heidelberg, and S. Schwarzengrund (Schlosser et al., 2000). RCPBS has prompted new proposed performance standards for NRTE comminuted chicken, which allows a maximum of 25% Salmonella positives (USDAFSIS, 2015a). Current antimicrobial methods utilized in the United States include immersion chilling coupled with chemical compounds, such as chlorine and various organic acids, as processing aids (USDA-FSIS, 2016). Considering the proposed performance standards, alternative methods of pathogen control should be explored to reduce the level of Salmonella in NRTE ground chicken. Bacteriophages are viruses that are specific obligate bacterial parasites and usually possess high specificity for one bacterial species. Generally, once the phage inserts its genome into the bacteria, it will use the bacteria’s replicating machinery to produce more phages, which are released upon bacterial cell lysis. Lytic phages infect bacteria resulting in rapid host death with minimal chance of phage transduction (Monk et al., 2010). There has been an increased use of lytic phages in agricultural research and application. The phages used for therapeutic purposes are usually strictly lytic, and this therapeutic use of phages to combat bacterial

In the United States, non-typhoidal salmonellosis is the number one foodborne illness that results in hospitalization and/or death, causing approximately 1 million illnesses, 19,000 hospitalizations, and 380 mortalities (CDC, 2012). Although it has been recovered from a variety of foods, Salmonella is traditionally associated with raw and undercooked poultry, and thus is a major concern for poultry processors. Poultry, more specifically chicken, is a major vehicle for Salmonella (Foley et al., 2011; Painter et al., 2013). Sold as either whole carcass, parts, or ground meat, broiler chickens are a multi-billion-dollar per year industry in the United States (USDA-NASS, 2013). In 2012, the U.S. Department of Agriculture (USDA) Food Safety Inspection Service (FSIS) conducted an 8-month Raw Chicken Parts Baseline Survey (RCPBS) to determine the prevalence of Salmonella on various raw chicken parts including not ready to eat (NRTE) comminuted poultry (USDAFSIS, 2015a). NRTE comminuted poultry includes raw ground chicken derived from chicken parts that have not been breaded or battered (USDA-FSIS, 2015b). RCPBS found that 49% of NRTE comminuted chicken had tested positive for Salmonella. According to the USDA,  C 2017 Poultry Science Association Inc. Received December 12, 2016. Accepted March 2, 2017. 1 Corresponding author: [email protected]

1

2

GRANT ET AL.

infections dates back to the early 20th century. Phages were used to treat several enteric diseases, such as cholera and dysentery, as well as food and agricultural applications during the 1920s to 1940s (Harper et al., 2014). With the advent and subsequent increase of synthetic antibiotics, the use of phage therapy saw a decline during the mid-1900s in the western hemisphere (Monk et al., 2010). There has been a recent resurgence of interest with bacteriophage therapy, and many more recent studies exist that demonstrate the ability of bacteriophages to reduced pathogens on post-harvest agricultural commodities (Boyacioglu et al., 2013; Chibeu et al., 2013; Ferguson et al., 2013; Goode et al., 2003; Hooton et al., 2011; Hudson et al., 2013; Sharma et al., 2015; Spricigo et al., 2013; Sukumaran et al., 2016; Yeh et al., 2017). Several commercially available bacteriophages have been regarded as generally recognized as safe (GRAS) by the USDA-FSIS, including SalmonelexTM , a Salmonella-specific bacteriophage cocktail produced by Micreos, Inc. (Wageningen, Netherlands). Information is limited about the bacteriophage inactivation of Salmonella considering the variables that should be accounted for during large-scale production in ground chicken using a commercially available bacteriophage. The aim of this study was to ascertain the effectiveness of commercially available bacteriophages during ground chicken production in regards to water source, Salmonella serotype, and time using Salmonella isolated from ground chicken (GC) and lab-based Salmonella (non-ground chicken or non-GC) isolates.

MATERIALS AND METHODS Salmonella Serovars Used in This Study Several Salmonella serovars were used to understand the bacteriophage host range, in the form of 2 cocktails: one with isolates obtained from GC and the other composed of laboratory isolates that were non-GC in origin. Salmonella isolated from commercially available ground chicken (GC) (S. Newport, S. Typhimurium, S. Thompson), as well as several strains (S. Heidelberg ATCC 8326, S. Enteritidis ATCC 13076, S. Typhimurium (Parveen et al. 2007)) not isolated from ground chicken were used in this study. Prior to use, all isolates were characterized by agglutination serotyping at the USDA National Veterinary Services Laboratories (NVSL) in Ames, Iowa. All isolates were then made resistant to a minimum concentration of 25 μg/mL of nalidixic acid for selective purposes by spread plating isolates on gradually increasing concentrations of nalidixic acid. Both mutant and wild type growth curves were recorded with an optical density of 600 (OD 600) during incubation in Luria-Bertani (LB) broth (BD, Franklin Lakes, NJ) without antibiotics. Measurements were recorded until the cultures reached stationary phase.

Antibiotic Susceptibility Testing of Salmonella Isolates All of the isolates used in this study were sent to the USDA-NVSL for antibiotic resistance determination after they were made resistant to nalidixic acid. A SensititreTM gram-negative NARMS plate (Trek Diagnostic Systems, Independence, OH) was used, which is a type of minimal inhibitory concentration (MIC) test. The MIC ranges of the following antibiotics were tested: amikacin (0.5 to 64 μg/mL), amoxicillin/ clavulanic acid (0.5/1 to 16/32 μg/mL), ampicillin (1 to 32 μg/mL), cefoxitin (0.5 to 32 μg/mL), ceftiofur (0.12 to 8 μg/mL), ceftriaxone (0.25 to 64 μg/mL), chloramphenicol (2 to 32 μg/mL), ciprofloxacin (0.015 to 4 μg/mL), gentamicin (0.25 to 16 μg/mL), kanamycin (8 to 64 μg/mL), nalidixic acid (0.5 to 32 μg/mL), streptomycin (32 to 64 μg/mL), sulfisoxazole (16 to 256 μg/mL), tetracycline (4 to 32 μg/mL), trimethoprim/ sulfamethoxazole (0.12/2.238 to 4/76 μg/mL).

Water Analysis Ammonia, nitrates, and mineral content were determined in the 2 different water types used in this study to dilute the bacteriophage prior to application: sterile tap water (stH2 O) and sterile filtered water (sfH2 O) that had undergone reverse-osmosis filtration. Both types of water were sterilized by autoclaving. Ammonia, as NH3 , and nitrates were determined usR ing (Lachat Instruments, Loveland, CO) QuikChem 8500 Series 2 Flow Injection Analyzer System according R salicylate-hypochlorite (10-107-06to the QuikChem 2-A) and Cd reduction (10-107-01-1-A) methods. Water mineral contents were analyzed using inductively coupled plasma (ICP) optical emission spectroscopy with Varian 730-ES Axil ICP Spectrometer (Varian, Palo Alto, CA). The following elements and their limits (mg L−1 ) were detected: Al (0.01), As (0.01), Ca (0.1), Cd (0.01), Cu (0.01), Fe (0.01), Hg (0.01), K (0.1), Mg (0.01), Mn (0.01), Mo (0.01), Na (0.1), Ni (0.01), P (0.01), Pb (0.1), Na (0.1), Ni (0.01), S (0.1), and Zn (0.01).

Sample Collection and Detection of Salmonella Commercially available boneless, skinless chicken legs and thighs were obtained from a local, large-scale poultry processing facility. Samples were analyzed for Salmonella prior to use. Briefly, 325 g of the chicken pieces were removed from the batch and placed in individual sterile bags and submerged in 1,625 mL of sterile buffered peptone water (BPW). After being shaken vigorously for 1 min, meat and rinsate were incubated at 37◦ C for 24 h (USDA-FSIS, 2014). Rinsate was sampled for the presence of Salmonella using

BACTERIOPHAGE FOR REDUCING SALMONELLA IN GROUND CHICKEN

BAX-PCR (DuPont Qualicon, Wilmington, DE), per the manufacturer’s instructions. Screening was completed in quadruplicate per batch of chicken. Samples that tested positive for Salmonella were discarded and not used in the study. Samples that were Salmonella negative were stored at –20◦ C until sampled. Rather than using naturally contaminated chicken, it was in the interest of this study to establish a baseline of bacteriophage reduction using artificially inoculated samples.

Sample Preparation and Salmonella Inoculation The chicken was thawed by refrigeration until completely defrosted, and all thawed parts were trimmed to uniform size of approximately 50 cm2 surface area. The chicken was allowed to dry under a sterile chemical hood for 20 min per side at room temperature to reduce any excess moisture that may have accumulated during freezing. The Salmonella isolates used in this study were stored at –80◦ C in 20% glycerol (Difco, Sparks, MD) and 80% LB broth. Overnight cultures of individual isolates were serial diluted to equal concentration in sterile LB broth. For each cocktail, equal volumes of each strain was combined in a sterile Falcon tube (BD). One milliliter was applied to each 50 cm2 piece of meat (0.5 mL on the top and 0.5 mL on the bottom) to obtain a final concentration of ∼104 CFU/cm2 . Samples were allowed to air dry under a sterile chemical hood for 20 min per side at room temperature to give the Salmonella time to attach to the surface of the chicken. The samples were then placed in the refrigerator until the surface temperature reaches 1–4◦ C.

Bacteriophage Reduction of Salmonella SalmonelexTM , a commercially available Salmonella specific bacteriophage cocktail, was obtained from Micreos Inc. and kept at 4◦ C until used. Prior to each study, the bacteriophage titration was performed using a standard plaque assay to determine the start concentration. After titration, the bacteriophage was serial diluted in either sterile filtered water (sfH2 O), that underwent reverse-osmosis filtration and was obtained from a regional poultry processor, or sterile tap water (stH2 O), obtained from the same regional poultry processor. Both types of water were obtained prior to chlorination by the poultry processor. The diluted bacteriophage was applied to both sides of the Salmonella inoculated chicken to reach a final concentration that was suggested by the manufacturer (∼107 PFU/cm2 ). The pH of the different water samples was assessed without the presence of the bacteriophage (Orion Star, Radnor, PA). The control group was spread with equal volumes of either sfH2 O or stH2 O water without the bacteriophage. Samples were then in-

3

cubated at 4◦ C for 30 min (the minimum phage inactivation time recommended by the manufacturer) or 8 h. After incubating for the specified amount of time, the chicken was ground using a benchtop grinder (Waring Laboratory Science, Winsted, CT). After grinding, 25-g aliquots were placed in separate sterile bags and 225 mL of sterile BPW. The bags were stomached (Seward Laboratory Systems Inc., Davie, FL) at maximum speed for 1 min before sampling (USDA-FSIS, 2013). Immediately after stomaching, 12 mL of homogenate was centrifuged at 12,000 × g for 5 min and the supernatant containing any bacteriophage was removed prior to spread plating on XLT4 (Difco) in triplicate. The XLT4 was supplemented with 25 μg/mL of nalidixic acid. Samples were then incubated at 37◦ C for 18 to 24 h and enumerated as CFU/cm2 .

Statistical Analysis All treatments were independently performed in quadruplicate using a randomized block design. IBM SPSS statistical software (Armonk, NY) was used to determine statistical significant differences between the treatments (P < 0.05) as well as pairwise comparisons with a 95% confidence interval.

RESULTS Prior to beginning the study, all bacteria strains were first confirmed to be Salmonella. Two different Salmonella cocktails were used: a GC-cocktail and a non-GC cocktail. After characterization, all Salmonella serovars were made resistant to a minimum concentration of 25 μg/mL nalidixic acid for selective purposes. The growth rates of the nalidixic acid resistant strains were determined to be similar to the growth rates of their respective wild-type (P > 0.05) (results not shown). All of the tested isolates were resistant to nalidixic acid (MIC > 32.0 μg/mL). Of the GC isolates, S. Typhimurium and S. Newport also had intermediate resistance to ciprofloxacin (MIC = 0.25 and 0.21 μg/mL, respectively), while S. Thompson was resistant to sulfisoxaxole (MIC > 256 μg/mL), tetracycline (MIC > 32.0 μg/mL), and had intermediate resistance to chloramphenicol (MIC = 16.0 μg/mL). The non-GC serovars also had varying patterns of antibiotic resistance. S. Typhimurium was resistant to: gentamicin (MIC > 16.0 μg/mL), streptomycin (MIC > 64.0 μg/mL), and sulfisoxazole (MIC > 256.0 μg/mL). S. Heidelberg ATCC 8326 was intermediately resistant to ciprofloxacin (MIC = 0.5 μg/mL), and S. Enteritidis ATCC 13076 was only resistant to nalidixic acid at the aforementioned concentration.

4

GRANT ET AL. Table 1. Average minerals, ammonia, and nitrate content of stH2 O and sfH2 0. Ca

Cu

K

Mg

Na

Ni

P

S

Si

Ammonia

Nitrates

Min Max

0.10 600

0.01 6.00

0.10 120

0.01 120

0.10 120

0.01 6.00

0.01 120

0.10 600

0.010 30.0

mg/L

mg N/L

stH2 O sfH2 O

9.38 −

0.43 −

7.94 0.7

4.87 −

+ 18.7

0.11 −

0.63 −

1.25 −

7.62 1.75

0.124 0.118

0.080 0.034

Minimum and maximum limits of detection are given for the listed minerals. All mineral concentrations are given as μ g/mL, ammonia is mg/L and nitrates are mg N/L. (+) indicates results were above the limits of detection and (−) indicate results were below the limits of detection.

Table 2. Reduction of Salmonella cocktail consisting of serovars isolated from ground chicken by bacteriophage diluted in two sources of water, applied to the chicken, then ground. Treatment

pH

stH2 O Control stH2 O with Phage sfH2 O Control sfH2 O with Phage

8.14 7.40

Remaining Salmonella population after 30 min (Log CFU/cm2 ) 4.00a 3.61b 4.00a 3.77c

Bacteriophage solution was diluted in either sterile tap water (stH2 O) or sterile water filtered through a reverse-osmosis system (sfH2 O). Different letter designations indicate a statistically significant different means (P < 0.05) among all tested conditions.

Effects of Water Source on Salmonella Reduction The results of the mineral analysis of the stH2 O and the sfH2 O are illustrated in Table 1. Samples were analyzed in duplicate, and the averages were reported. The minerals mentioned in Section 2.3 of the methods that are not listed in Table 1 indicate that their levels were below the limit of detection for both samples and thus, were not included in the table. All of the minerals listed in Table 1 were present in the stH2 O, albeit Na levels were higher than the set limit of detection, while the only minerals identified in sfH2 O were K, Na, and Si. Before each study, the bacteriophage concentration was confirmed to be at the manufacturer’s guaranteed levels (results not shown). Table 2 shows the reduction of Salmonella based on the different water sources used to dilute the bacteriophage. It is not uncommon for poultry processors to use water filtration systems in lieu of tap water to minimize variations between processing locations. Considering this study was focused on the reduction of Salmonella in ground chicken, the GC-cocktail was used when testing the effects of water source. Salmonella cocktail was significantly reduced further when the bacteriophage was diluted in stH2 O water compared to the results reported when the phage was diluted in sfH2 O. The remaining Salmonella population after interacting with phage diluted in either stH2 O and sfH2 O was 3.61 Log CFU/cm2 and 3.77 Log CFU/cm2 , respectively. Control samples were not challenged with the bacteriophage, but with equal volumes of water. Based on these results, stH2 O water was used for the duration of the study.

Bacteriophage Reduction of Different Salmonella Cocktails It is not uncommon for studies to use target organisms that are not environmental isolates, but rather laboratory strains (Spricigo et al., 2013, Hudson et al., 2013, Hooton et al., 2011, Goode et al., 2003). One aspect of this study was to determine if there would be a difference in bacteriophage susceptibility between GC and non-GC Salmonella. Reduction of the 2 Salmonella cocktails by bacteriophage was determined after 30-min and 8-h treatments. Thirty minutes is the minimum interaction time required by the bacteriophage, according to the manufacturer, and 8 h is the maximum time to be practical in commercial poultry processing using 8-h shifts. The results show a greater reduction (P < 0.05) with the non-GC cocktail compared to the GC cocktail at both the 30-min and 8-h time points (Figure 1). Both cocktails experienced greater Log reductions (P < 0.05) after 8 h compared to their 30-min counterparts. After 30 min, there was a 0.39 Log CFU/cm2 reduction and a 0.71 Log CFU/cm2 reduction in the GC and nonGC cocktails, respectively, when compared to the control. Also, after 8 h, the non-GC isolates maintained the larger reduction (P < 0.05) with 0.90 Log CFU/cm2 compared to the 0.67 Log CFU/cm2 reduction observed with the GC cocktail.

Bacteriophage Reduction of Individual Serovars from the GC Cocktail The isolates that comprised the GC cocktail were further investigated for their susceptibility to the bacteriophage cocktail and were individually tested with a bacteriophage interaction of 30 min and 8 h (Figure 2). More variation was detected after 30 min of interaction, with S. Thompson having the greatest reduction at 0.53 Log CFU/cm2 , followed by S. Newport at 0.47 Log CFU/cm2 and S. Typhimurium, which had the lowest reduction of 0.32 Log CFU/cm2 . Interestingly, after 8 h there was little difference (P > 0.05) in the reduction levels between the 3 strains. Both S. Thompson and S. Newport were reduced by 0.77 Log CFU/cm2 and S. Typhimurium experienced a 0.80 Log CFU/cm2 reduction when compared to the control, even though it had the least amount of reduction after 30 min.

BACTERIOPHAGE FOR REDUCING SALMONELLA IN GROUND CHICKEN

5

Figure 1. Recovery of 2 Salmonella cocktails after being challenged with Salmonelex bacteriophage for either 30 min or 8 h. Non-ground chicken cocktail consisted of the non-ground chicken isolates: S. Heidlburg (ATCC8326), S. Enteritidis (ATCC13076), and S. Typhimurium. The ground chicken cocktail consisted of Salmonella serovar isolated from ground chicken: S. Newport, S. Thompson, and S. Typhimurium. Different letter designations at the top of each column indicate a statistically significant difference (P < 0.05).

Figure 2. Recovery of the individual serotypes that comprised the ground chicken cocktail: S. Newport (Newp.), S. Thompson (Thom), and S. Typhimurium (Typhm.). Each isolate was individually challenged with the Salmonelex bacteriophage for either 30 min or 8-h. Different letter designations at the top of each column indicate a statistically significant difference (P < 0.05)

DISCUSSIONS Performance standards for Salmonella in poultry are becoming more stringent in the United States. In 2012, the USDA-FSIS reported an 8-month survey and analysis of raw chicken parts titled Nationwide Microbiological Baseline Data Collection Programs: Raw Chicken Parts Baseline Survey. Data from that survey showed a prevalence of 49% Salmonella in comminuted chicken. Since then, the USDA has proposed that compliance would consist of no more than 13 of 52 Salmonella-positive samples for NRTE comminuted chicken (USDA-FSIS, 2015a). The U.S. Food

and Drug Administration (FDA) recently reported Salmonella recovery in retail poultry is the lowest it has been since 2002, which can be attributed to improved manufacturing practices by poultry processors (FDA, 2016). Currently, the poultry industry uses an assortment of organic acids (Ricke, 2003; Theron et al., 2010), essential oils (Gutierrez et al., 2008; Gutierrez et al., 2009) and chlorines (FDA, 2007; USDAFSIS, 2016) based interventions to mitigate the presence of pathogens. However, poultry processors must find alternative means to comply with these standards while meeting consumers’ demands for wholesome, clean-label, and safe products that are readily

6

GRANT ET AL.

available. Bacteriophages are proving to be a promising antibacterial intervention. Bacteriophages are being utilized in various aspects of food production, and tailed bacteriophages were shown to maintain stability in adverse environmental conditions (Jo´ nczyk et al., 2011), conditions that may be encountered during food processing. Types of water used to dilute bacteriophages is one of those conditions to be considered. Large poultry processors may have production plants located in different counties, regions, states, and/or countries. Water processing between facilities could be different based on plant location and sources of water. For instance, a facility located in one area may use filtered water, while a plant located in a different region may use the county’s well water supply. The results of this study show that sterile tap water had a greater (P < 0.05) antibacterial effect when compared to sterile water that has undergone reverse-osmosis filtration. Bacteriophage inactivation can be caused by osmotic shock (Whitman and Marshall, 1971), and diluting the bacteriophage in filtered water may have been rendered the bacteriophage less effective. This could have occurred when SalmonelexTM bacteriophages were transferred from their stock solution to the filtered water, which could cause some bacteriophage inactivation. Solution salt concentrations may affect the elastic and electrostatic pressure in the viral capsid that is required for injection of genetic information into a host (Evilevitch et al., 2008; Leibo and Mazur, 1966; Petrov et al., 2013). Leibo and Mazur (1966) studied the effects of different osmotic pressures on the ability of bacteriophages T4D, T4B, and T4B01 to inject their genetic material. All 3 bacteriophages are variations of E. coli specific T4 bacteriophage from the contractile tail having Myoviridae family. All 3 bacteriophages were inactivated when quickly transferred from a univalent salt concentration to a highly-diluted concentration, and the researchers of that study concluded that the concentration of the initial solution compared to the more dilute final solution effected bacteriophage inactivation. The levels of sodium and magnesium in this present study were below the limits of detection (Table 1), and the literature has shown that the lack of divalent cations, especially Mg2+ and Ca2+ , negatively impact phage attachment because these 2 ions help stabilize the interaction of the viron and host cell wall receptors (Li and Zhang, 2014; Moldovan et al., 2007). For example, when supplemented with 10 mM of calcium chloride, Li and Zhang (2014) observed a significant increase in Staphylococcus aureus-specific phage infection when compared to the control, and the lack of calcium ions (Ca2+ ) negatively affected phage propagation and bacterial lysis after adsorption, according to Cvirkait˙eKrupoviˇc et al. (2010). The minimized bacteriophage activity could have been caused by the absence of calcium and magnesium ions present in the filtered water (Table 1). However, future studies are needed to determine the virucidal and/or virostatic effect this specific sterile distilled water has on the SalmonelexTM bacte-

riophages. It is noteworthy that the bacteriophage was more effective in the sterile tap water even though the pH was slightly more alkaline than the sterile filtered water. The bacteriophages that comprise SalmonelexTM were able to survive a more alkaline environment. However, poultry processor should consider the pH of the water used in their facilities, because pH is another important factor that could reduce bacteriophage stability and thus its effectiveness (May et al., 2014). Time is a limiting factor during poultry production where the goal is to produce a quality product that maintains desirable organoleptic properties at the time of retail. Chicken must be ground within hours of deboning, but the increase in surface area of chicken particles after grinding greatly reduces the likelihood of bacteriophage interacting with its host, thus grinding of the meat greatly reduces the efficacy of the bacteriophage and minimizes the chances of interactions between microorganisms and bacteriophages (Sharma et al., 2015). Intralytix (Baltimore, MD), another company that sells bacteriophages for food protection, suggests applying the bacteriophage prior to grinding poultry and red meat for optimum bacterial control (Intralytix, 2014). Reduction of bacteria by bacteriophage action will typically have to occur within several hours of application during production to maintain a wholesome product, however, the more time equates to greater susceptibility to phages. This study used 8 h as the maximum time for bacteriophage reduction, because to produce a wholesome product, processors may only be able to apply the bacteriophage for the durations of time. This study assumed a maximum time frame of an 8-h working shift for the phage to interact with the Salmonella before the chicken is ground and the phage is rendered ineffective. In this study, 8 h resulted in a reduced (P < 0.05) Salmonella populations compared to 30 min. This trend was maintained with both cocktails and when the individual isolates were tested. None of the reductions were greater than 1 Log within an 8-h incubation. Similar results were echoed by Sharma et al. (2015). After 5-min incubation followed by grinding using bacteria and bacteriophage concentrations similar to this study, Sharma et al. (2015) reported insignificant reductions of S. Heidelberg in ground turkey, which was similar to Zinno et al. (2014) findings in minced chicken. However, with increased multiplicity of infection (MOI) and increasing reduction time to 48 h, Zinno et al. (2014) found a 2 Log CFU/g reduction. Similar to other studies, these findings suggest that poultry processors should find a compromise between time and reduction levels; too little time for the bacteriophage to work will result in too little reduction, and too much time could result in the loss of shelf-life and/or desirable organoleptic properties of the product. Temperature is another consideration during processing. Hudson et al. (2013) applied E. coli specific phages to E. coli O157:H7 inoculated sides of beef during cold storage, mimicking the temperatures experienced

BACTERIOPHAGE FOR REDUCING SALMONELLA IN GROUND CHICKEN

during large-scale production. The researchers determined that phage inactivation of E. coli O157:H7 was based more on phage concentration and not on temperature. Although bacteriophages used in our study reduced Salmonella during cold storage (≤4◦ C), that reduction did not reach 1 Log CFU/cm2 for either Salmonella cocktail at either 30 min or 8 h. The results indicate that the use of bacteriophages, like the ones used in our study and by Hudson et al. (2013), will likely not be effected by the cold temperatures at which poultry production facilities are typically maintained (0–4◦ C). This study recovered several serovars of Salmonella from GC, and produced a GC-cocktail and compare them to a non-GC cocktail. A greater reduction (P < 0.05) was observed in the non-GC strains than the GC strains. This suggests the incorporation of environmentally isolated pathogens should be considered when conducting bacteriophage studies. As previously mentioned, it is not uncommon for studies to use laboratory Salmonella isolates that may or may not have been isolated from poultry. These findings suggest that results could differ depending on bacterial isolates used and the host range of the bacteriophage. More importantly, the effectiveness of commercially available bacteriophages may not be consistent if the host range of the bacteriophage does not encompass the bacteria, in this case Salmonella, that could be encountered during production. Bacteriophage reduction does not depend on antibiotic resistance profile of its target organism (Burrowes et al., 2011; Chan et al., 2013; Kesselheim and Outterson, 2010). The Salmonella isolates in the GC cocktail had different antibiotic resistance profiles. Although the reduction between isolates was greater (P < 0.05) after 30 min, after 8 h the reduction between isolates were similar (P > 0.05). Reduction of the GC isolates was independent of antimicrobial resistance. Reduction of these isolates was independent of antibiotic resistance. Bacteriophage reduction is most likely caused by the amount and density of the bacteriophage receptors in the bacterial cell wall (Rakhuba et al., 2010) as well as the bacteriophage concentration. Goode et al. (2003) showed increased reduction of S. Enteritidis was optimal at MOI of 100–1,000, and these results were echoed by Sharma et al. (2015), who tested 3 different Salmonella serovars against a commercially available bacteriophage, and they showed optimal reduction at an increased MOI of 10,000. It is noteworthy that the abovementioned studies applied bacteriophage to a single serovar of Salmonella at a time. In conclusion, the Salmonella bacteriophage (SalmonelexTM ) used in our study was able to reduce Salmonella populations in ground chicken production, however, a 1.0 Log reduction was not accomplished within 8-h. Greater bacterial reductions could be achieved if incorporated in conjunction with the antimicrobial interventions currently used in the poultry industry. Prior to implementing bacterio-

7

phages into production, poultry processors should take several limiting factors into consideration, such as: the type of water used in their facilities, time, and the bacteriophage’s host range. Future directions of this project will be to run full scale trials that would incorporate the use of other antibiotic interventions with naturally contaminated Salmonella loads. This will help determine the full effects that commercially available bacteriophages have on reducing Salmonella loads.

ACKNOWLEDGEMENTS We thank Micreos for providing the bacteriophage and the technical support. We would also like to thank United States Department of Education Title III and Perdue Farms, Inc. for the funding of this project. Special thanks to Dr. Ray Bryant (USDA ARS) and Dr. Arthur Allen (UMES) for their assistance in the chemical analysis of water samples.

REFERENCES Boyacioglu, O., M. Sharma, A. Sulakvelidze, and G. Ipek. 2013. Biocontrol of Escherichia coli O157:H7 on fresh-cut leafy greens: Using a bacteriophage cocktail in combination with modified atmosphere packaging. Bacteriophage. 3:e246201–e246207. Burrowes, B., D. R. Harper, J. Anderson, M. McConville, and M. C. Enright. 2011. Bacteriophage therapy: Potential uses in the control of antibiotic-resistant pathogens. Expert Rev. Anti. Infect. Ther. 9:775–785. CDC. 2012. About foodborne illness. Accessed on October 13, 2013. Available at http://www.cdc.gov/foodsafety/facts.html#top. Chan, B. K., S. T. Abedon, and C. Loc-Carrillo. 2013. Phage cocktails and the future of phage therapy. Future Microbiol. 8:769– 783. Chibeu, A., L. Agius, A. Gao, P. M. Sabour, A. M. Kropinski, and S. Balamurugan. 2013. Efficacy of bacteriophage LISTEX P100 combined with chemical antimicrobials in reducing Listeria monocytogenes in cooked turkey and roast beef. Int. J. Food Microbiol. 167:208–214. Cvirkait˙e-Krupoviˇc, V., M. Krupoviˇc, R. Daugelaviˇcius, and D. H. Bamford. 2010. Calcium ion-dependent entry of the membranecontaining bacteriophage PM2 into its Pseudoalteromonas host. Virology. 405:120–128. Evilevitch, A., L. T. Fang, A. M. Yoffe, M. Castelnovo, D. C. Rau, V. A. Parsegian, W. M. Gelbart, and C. M. Knobler. 2008. Effects of salt concentrations and bending energy on the extent of ejection of phage genomes. Biophys. J. 94:1110–1120. Ferguson, S., C. Roberts, E. Handy, and M. Sharma. 2013. Lytic bacteriophages reduce Escherichia coli O157:H7 on fresh cut lettuce introduced through cross contamination. Bacteriophage 3:e24323. Foley, S. L., R. Nayak, I. B. Hanning, T. J. Johnson, J. Han, and S. C. Ricke. 2011. Population dynamics of Salmonella enterica serotypes in commercial egg and poultry production. Appl. Environ. Microbiol. 77:4273–4279. Fratamico, P. M. 2003. Comparison of culture, polymerase chain reaction (PCR), TaqMan Salmonella and Transia Card Salmonella assays for detection of Salmonella spp. in naturally-contaminated ground chicken, ground turkey, and ground beef. Mol. Cell. Probes. 17:215–221. Goode, D., V. M. Allen, and P. A. Barrow. 2003. Reduction of Experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic bacteriophages. Appl. Environ. Microbiol. 69:5032–5036. Gutierrez, J., C. Barry-Ryan, and P. Bourke. 2008. The antimicrobial efficacy of plant essential oil combinations and interactions with food ingredients. Int. J. Food Microbiol. 124:91–97.

8

GRANT ET AL.

Gutierrez, J., C. Barry-Ryan, and P. Bourke. 2009. Antimicrobial activity of plant essential oils using food model media: Efficacy, synergistic potential and interactions with food components. Food Microbiol. 26:142–150. Harper, D. R., B. H. Burrowes, and E. M. Kutter. 2014. Bacteriophage: Therapeutic uses in The Encyclopedia of Life Sciences eLS ed. John Wiley & Sons, doi: 10.1002/9780470015902.a0020000.pub2. Hooton, S. P. T., R. J. Atterbury, and I. F. Connerton. 2011. Application of a bacteriophage cocktail to reduce Salmonella Typhimurium U288 contamination on pig skin. Int. J. Food Microbiol. 151:157–163. Hudson, J. A., C. Billington, A. J. Cornelius, T. Wilson, S. L. W. On, A. Premaratne, and N. J. King. 2013. Bacteriophage to inactivate Escherichia coli O157:H7 on beef. Food Microbiol. 36:14–21. Intralytix. 2014. SalmoFresh. Accessed on October 2015. Available at http://www.intralytix.com/files/prod/02SP/02SP-Desc.pdf. and A. G´ orski. 2011. The Jo´ nczyk, E., M. Klak, R. Miedzybrodzki,  influence of external factors on bacteriophages—review. Folia Microbiol. 56:191–200. Kesselheim, A. S., and K. Outterson. 2010. Fighting antibiotic resistance: Marrying new financial incentives to meeting public health goals. Health Aff. 29:1689–1696. Leibo, S. P., and P. Mazur. 1966. Effect of osmotic shock and low salt concentration on survival and density of bacteriophages T4B and T4Bo1. Biophys. J. 6:747–772. Li, L., and Z. Zhang. 2014. Isolation and characterization of a virulent bacteriophage SPW specific for Staphylococcus aureus isolated from bovine mastitis of lactating dairy cattle. Mol. Biol. Rep. 41:5829–5838. May, E. R., K. Arora, and C. L. Brooks, 3rd. 2014. pH-induced stability switching of the bacteriophage HK97 maturation pathway. J. Am. Chem. Soc. 136:3097–3107. Moldovan, R., E. Chapman-McQuiston, and X. L. Wu. 2007. On kinetics of phage adsorption. Biophys. J. 93:303–315. Monk, A. B., C. D. Rees, P. Barrow, S. Hagens, and D. R. Harper. 2010. Bacteriophage applications: Where are we now? Lett. Appl. Microbiol. 51:363–369. Painter, J. A., R. M. Hoekstra, T. Ayers, R. V. Tauxe, C. R. Braden, F. J. Angulo, and P. M. Griffin. 2013. Attribution of foodborne illnesses, hospitalizations, and deaths to food commodities by using outbreak data, United States, 1998–2008. Emerg. Infect. Dis. J. 19:407. doi 10.3201/eid1903.111866. Parveen, S., M. Taabodi, J. Schwarz, T. P. Oscar, J. Harter-Dennis, and D. G White. 2007. Prevalence and antimicrobial resistance of Salmonella recovered from processed poultry J. Food Prot. 70:2466–2472. Petrov, A. S., S. S. Douglas, and S. C. Harvey. 2013. Effects of pulling forces, osmotic pressure, condensing agents and viscosity on the thermodynamics and kinetics of DNA ejection from bacteriophages to bacterial cells: a computational study. J. Phys. Condens. Matter. 25:115101. doi 10.1088/0953-8984/25/11/115101. Rakhuba, D. V., E. I. Kolomiets, E. Szwajcer Dey, and G. I. Novik. 2010. Bacteriophage receptors, mechanisms of phage adsorption and penetration into host cell. Polish J. Microbiol. 59:145–155. Ricke, S. C. 2003. Perspectives on the use of organic acids and short chain fatty acids as antimicrobials. Poult. Sci. 82:632–639. Schlosser, W., A. Hogue, E. Ebel, B. Rose, R. Umholz, K. Ferris, and W. James. 2000. Analysis of Salmonella serotypes from selected carcasses and raw ground products sampled prior to implementation of the pathogen reduction; Hazard analysis and critical control point final rule in the U.S. Int. J. Food Microbiol. 58:107–111. Sharma, C. S., J. Dhakal, and R. Nannapaneni. 2015. Efficacy of lytic bacteriophage preparation in reducing Salmonella in vitro, on turkey breast cutlets, and on ground turkey. J. Food Prot. 78:1357–1362. Spricigo, D. A., C. Bardina, P. Cortes, and M. Llagostera. 2013. Use of bacteriophage cocktail to control Salmonella in food and the food industry. Int. J. Food Microbiol. 165:169–174. Sukumaran, A. T., R. Nannapaneni, A. Kiess, and C. S. Sharma. 2016. Reduction of Salmonella on chicken breast fillets stored

under aerobic or modified atmosphere packaging by the application of lytic bacteriophage preparation SalmoFreshTM . Poult. Sci. 95:668–675. Theron, M. M., J. Lures, and F. Rykers. 2010. Application of organic acid in food preservation. Pages 51–95. Organic Acids and Food Preservation, M. M. Theron, and J. F. Rykers Lues, eds. CRC Press, Boca Raton, FL. U.S. FDA. 2007. Food and color additive final rule in Registrar citation number 72FR67572. Available at http://www.fda.gov/Food/ GuidanceRegulation/GuidanceDocumentsRegulatoryInformation /IngredientsAdditivesGRASPackaging/ucm074093.htm U.S. FDA. 2016. FDA NARMS retail meat interim report for Salmonella shows encouraging early trends continue; Includes whole genome sequencing data for the first time. Center for Veterinary Medicine. http://www.fda.gov/ AnimalVeterinary/NewsEvents/CVMUpdates/ucm498038.htm. USDA-FSIS. 2013. Isolation and identification of Salmonella from meat, poultry, pasturized egg, and catfish products and carcass and environmental sponges. Accessed on October 7, 2013. Available at: http://www.fsis.usda.gov/wps/wcm/connect/700c05fe06a2-492a-a6e1-3357f7701f52/MLG-4.pdf?MOD=AJPERES. USDA-FSIS. 2014. Laboratory guidebook: Notice of change. Accessed on October 7, 2015. Available at http:// www.fsis.usda.gov/wps/wcm/connect/700c05fe-06a2-492a-a6e13357f7701f52/MLG-4.pdf?MOD=AJPERES. USDA-FSIS. 2015a. Changes to the Salmonella and Campylobacter verification testing program: Proposed performance standards for Salmonella and Campylobacter in not-ready-to-eat comminuted chicken and turkey products and raw chicken parts and related agency verification procedures and other changes to agency sampling Pages 3940–3950 in Fed. Reg 80. Docket No. FSIS-2014-0023U.S. Department of Agriculture, Washington, DC. USDA-FSIS. 2015b. Pathogen reduction- Salmonella and Campylobacter performance standards verification testing. Accessed on October 7, 2015. Available at http://www.fsis.usda.gov/wps/ wcm/connect/b0790997-2e74-48bf-9799-85814bac9ceb/28 IM PR Sal Campy.pdf?MOD=AJPERES. USDA-FSIS. 2015c. Serotypes profile of Salmonella isolates from meat and poultry products January 1998 through December 2013. Available at: http://www.fsis.usda.gov/wps/portal/fsis/ topics/data-collection-and-reports/microbiology/annualserotyping-reports. USDA-FSIS. 2016. Safe and suitable ingredients used in the production of meat, poultry and eggs products. FSIS Directive 7120.1, revision 36. Available at: http://www.fsis.usda.gov/wps/wcm/ connect/bab10e09-aefa-483b-8be8-809a1f051d4c/71201.. pdf?MOD=AJPERES. USDA-NASS. 2013. Poultry production and value. Accessed on October 29, 2013. Available at http://usda01.library.cornell.edu/ usda/current/PoulProdVa/PoulProdVa-04-29-2013.pdf. Whichard, J. M., K. Gay, J. E. Stevenson, K. J. Joyce, K. L. Cooper, M. Omondi, F. Medalla, G. A. Jacoby, and T. J. Barrett. 2007. Human Salmonella and concurrent decreased susceptibility to quinolones and extended-spectrum cephalosporins. Emerg. Infect. Dis. 13:1681–1688. White, D. G., S. Zhao, R. Sudler, S. Ayers, S. Friedman, S. Chen, P. F. McDermott, S. McDermott, D. D. Wagner, and J. Meng. 2001. The isolation of antibiotic-resistant Salmonella from retail ground meats. N. Engl. J. Med. 345: 1147–1154. Whitman, P. A., and R. T. Marshall. 1971. Characterization of two psychrophilic Pseudomonas bacteriophages isolated from ground beef. Appl. Environ. Microbiol. 22:463–468. Yeh, Y., P. Purushothaman, N. Gupta, M. Ragnone, S. C. Verma, and A. S. de Mello. 2017. Bacteriophage application on red meats and poultry: Effects on Salmonella population in final ground products. Meat Sci. 127:30–34. Zinno, P., C. Devirgiliis, D. Ercolini, D. Ongeng, and G. Mauriello. 2014. Bacteriophage P22 to challenge Salmonella in foods. Int. J. Food Microbiol. 191:69–74.