Isolation, identification and characterization of lytic, wide host range bacteriophages from waste effluents against Salmonella enterica serovars

Food Control 38 (2014) 67e74

Contents lists available at ScienceDirect

Food Control journal homepage: www.elsevier.com/locate/foodcont

Isolation, identification and characterization of lytic, wide host range bacteriophages from waste effluents against Salmonella enterica serovars Mastura Akhtar, Stelios Viazis, Francisco Diez-Gonzalez* Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN 55108, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 July 2013 Received in revised form 26 September 2013 Accepted 28 September 2013

The use of bacteriophages is considered as a viable alternative to chemical antimicrobials against foodborne pathogens. The objective of this study was to develop a collection of lytic bacteriophages which will be able to infect different pathogenic Salmonella enterica serovars. Phages were isolated from animal feces and sewage samples, purified, characterized morphologically and by DNA fingerprinting, and host ranges were determined. Spot test and efficiency of plaquing (EOP) data indicated that two phages, SEA1 and SEA2 had the broadest host range against Salmonella among all isolated phages. SEA2 was highly efficient to infect S. Typhimurium DT104 (0.5e1 EOP value). Only phage SSA1 was able to infect S. Montevideo. Transmission electron microscopy (TEM) revealed the phages in the collection were mostly (4 out of 6) Siphoviridae, while SEA1 and SEA2 were Myoviridae T4-like phages. SEA1 and SEA2 had the largest genome sizes in the collection, 190 and 170 kb, respectively. Pulsed field gel electrophoresis (PFGE) analysis demonstrated distinct digestion profiles with EcoRI for phages SSA1, STD3, STE3 and STF1. However, SEA1 and SEA2 shared a similar restriction enzyme (RE) digestion pattern with same morphotype, but distinct profiles in lysing Salmonella strains. These anti-Salmonella phages were highly host specific with few exceptions of lytic phages that were able to infect a wide variety of Salmonella. These phages have potential for use in applications controlling Salmonella on different matrices. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Phages Salmonella Antimicrobial Anti-Salmonella phages Biocontrol Food safety

1. Introduction Salmonella infections are related to a wide diversity of food products and their source of contamination ranges from human through food processing facilities (Miao & Miller, 1999; Tauxe, Doyle, Kuchenmüller, Schlundt, & Stein, 2010). The estimated annual cost due to foodborne Salmonella infections is $2.4 billion in the United States ((USDA-ERS), 2001). According to the most recent report, non-typhoidal Salmonella has been identified as the leading cause of foodborne illnesses in USA with estimates of 1 million cases and 378 deaths per year (Scallan et al., 2011). Outbreaks have been associated with multidrug resistant serovars such as S. Typhimurium DT104 that often cause serious problems because of the limited treatment options (CDC, 2001). Though preventative measures are available, the recurring outbreaks of Salmonella pose

* Corresponding author. Department of Food Science and Nutrition, University of Minnesota, 225A, 1334 Eckles Ave., St. Paul, MN 55108, USA. Tel.: þ1 612 624 9756; fax: þ1 612 625 5272. E-mail address: [email protected] (F. Diez-Gonzalez). 0956-7135/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodcont.2013.09.064

an important public health risk that needs great attention and a pursuit for effective antimicrobial alternatives. In this context, bacteriophages have potential as an alternative to antibiotics or other conventional chemical control methods against bacterial pathogens. Bacteriophages are viruses capable of lysing bacteria, and specific lytic phages can kill pathogenic bacteria in their own habitat. Phages are ubiquitous in nature and can often be found in a variety of environments related to their host such as soil, sewage, water, manure, animal and produce farms, as well as different food processing plant effluents. The application of bacteriophages as a food safety intervention has been recently investigated and a few commercial preparations have been approved and marketed. Bacteriophages are often used in high concentrations to inactivate foodborne pathogens, such as Escherichia coli O157:H7, Salmonella, Listeria, and Campylobacter in different foods (Carlton, Noordman, Biswas, De Meester, & Loessner, 2005; Greer, 2005; O’Flynn, Coffey, Fitzgerald, & Ross, 2006; O’Flynn, Ross, Fitzgerald, & Coffey, 2004). Also, in production facilities, phages also have been used to control specific bacteria at pre-harvest and post-harvest stages of food production and storage (Greer, 2005). The Food and Drug Administration (FDA) has

68

M. Akhtar et al. / Food Control 38 (2014) 67e74

approved Listeria specific phages for use in foods (FDA, 2006) indicating the promising use of bacteriophages in food applications. Salmonella specific phages have been isolated and a few lytic wideactivity range phages have been identified. Bielke et al. found only two potent bacteriophages from the isolated pool were able to lyse 7 out of 10 Salmonella isolates (Bielke, Higgins, Donoghue, Donoghue, & Hargis, 2007). Flynn et al. described that two lytic phages, st104a and st104b were screened from 100 fecal samples and evaluated as effective therapy to reduce Salmonella enterica serovar Typhimurium DT104 in vitro (O’Flynn et al., 2006). However, despite the existing inadequate library of lytic Salmonella phages, literature search indicated that there is no currently available phage treatment to control the diverse group of Salmonella serovars. In this study, waste effluents from cattle and swine manure pit and raw sewage from the wastewater treatment plant were screened to isolate a collection of broad spectrum lytic Salmonella phages. The selected potent lytic phages were identified morphologically and further characterized to evaluate their lysis efficacy for infecting a wide range of S. enterica serovars. This library of lytic phages may be prospective to use with other isolated lytic phages for further food-related applications against Salmonella. 2. Materials and methods 2.1. Bacterial strains and culture conditions A total of 34 different of S. enterica strains belonging to eight serovars were used to serve as host strains and to screen phages from isolation sources (Table 1). All bacterial strains were streaked from the frozen culture stock onto tryptic soy agar (TSA; Neogen Corp., Lansing, MI). Prior to experiments, each strain was grown by picking an isolated colony from TSA plates and inoculating into tryptic soy broth (TSB; Neogen Corp.) and incubated at 37  C to obtain fresh overnight cultures. 2.2. Sample collection Several waste effluent samples were collected from 1) cattle and swine manure pit from the dairy cattle and swine barns of the University of Minnesota; 2) city sewage treatment plant in Rice

Table 1 List of bacterial strains that were used for this study. Salmonella serovars

Strain ID number

Number of strains

Source of strains

S. Agona S. Typhimurium

Agona FDA UK-1, 3019907 ATCC 14028, ATCC 700408 I598, I503, I527, I534, I535, I649, I526, I758, I536, I740, I600 E2009005811 2009595, 95657613, I823 FSIS10 95573473 AMO7076, AMO7073, AMO5104, AMO5313, B4442 Newport FDA 2006036 E2008001236, E2008001177, E2003002913, E2009010674, E2008001358 E200700502

1 2 2 11

FDA FSML ATCC VBS

1 3 1 1 5

MDH FSML FSML MDH CDC

1 1 5

FDA FSML MDH

1

MDH

S. S. S. S.

Enteritidis Heidelberg Montevideo Newport

S. Saintpaul

S. Tennessee

Abbreviation: FDA, Food and Drug Administration; FSML, Food Safety Microbiology Laboratory, University of Minnesota; ATCC, American Type Culture Collection; VBS, Department of Veterinary and Biomedical Sciences, University of Minnesota; MDH, Minnesota Department of Health; CDC, Centers for Disease Control and Prevention.

Lake City, Wisconsin; and 3) raw sewage sludge, wastewater treatment plant, City of St. Paul. Samples were collected by several visits throughout the spring and summer seasons. 2.3. Enrichment, isolation, purification and preparation of bacteriophages All waste samples were collected and centrifuged at 10,000  g for 10 min to remove solid particles and then filtered with 0.45 mm pore syringe filters (Nalgene, Thermo Fisher Scientific Inc. USA). For enrichment, isolation and purification of bacteriophages, modified methods were used from previously published reports (Klieve, 2005; Mullan, 2002; Twest & Kropinski, 2009). Salmonella strains were grown overnight at 37  C in TSB to obtain pure bacterial cultures. 0.1 mL of overnight pure cultures of Salmonella strains were inoculated into 10 mL TSB and incubated at 37  C shaker for 3e4 h to grow exponential phase cultures. Filtered sample supernatant (4.5 mL) was then mixed with 0.5 mL exponential phase bacterial cultures and 0.5 mL 10 concentrated TSB, and incubated at 37  C for 24e48 h. After incubation, samples were centrifuged at 10,000  g for 10 min, supernatants were filtered with a 0.45 mm filter syringe, and used as enriched phage (EP) samples. Initially, spot testing was used to isolate the phages. The host bacterial lawn was made by using a tryptone top agar (TTA) (containing per liter: bacteriological agar, 4; tryptone, 10 g; yeast extract, 5 g; NaCl, 7.5 g; Glucose, 10 g; 12 mmol MgSO4; 12 mmol CaCl2) containing host bacterial suspensions that were overlaid on top of TSA agar plates. When the agar overlays were solidified, several EP samples were spotted (1:10, phage:host) onto the lawns in a row and plates were incubated at 37  C for 18e24 h. After incubation, all plates were examined for positive and clear spot formation. The lysed agar spot was cut and dissolved into phosphate buffer saline (PBS; Neogen, Inc.), centrifuged and filtered again to collect the supernatants. The supernatants were used as spot filtrate. Plaque assays were conducted to isolate and purify individual phages from spot filtrates. For plaque assay, a series of 10-fold dilutions were made of spot filtrates. Exponential phase host cultures were mixed with dilutions of spot filtrates as host: spot filtrate (2:1) and incubated at 37  C for 15 min for propagation. These phages and bacterial suspensions were mixed with the TTA agar and poured onto TSA. Plates were incubated at 37  C for 24 h for plaque formation. To purify the phages, isolated plaques were picked by using a pipette or a wire loop and suspended with PBS. The suspensions were centrifuged and the supernatants were filter sterilized and used as a single phage culture. This purification process of an individual plaque through plaque assay, centrifugation and resuspension was repeated at least 3 times to ensure a pure phage stock. The high titer phage stocks were prepared by inoculating 1 mL of overnight host bacterial cultures with 100 ml of purified phage stock (105e106 PFU mL1) into TSB (with 5 mol l1 CaCl2) and incubated at 37  C for 18e20 h. When lysis of liquid TSB cultures was visually observed, a few drops of CHCl3 were added for complete lysis of the bacterial cells. The amplified, purified phages were centrifuged at 10,000  g for 10 min to pellet debris and supernatants were filtered to remove bacterial contaminants. The filtered supernatants were stored as high titer (108e1010 PFU mL1) phage stocks at 4  C and used for the different analysis throughout the study. 2.4. Determination of host range Spot testing was conducted to measure the ability of individual phages to infect different serovars of Salmonella. First, bacterial

M. Akhtar et al. / Food Control 38 (2014) 67e74

lawns were made with exponential phase cultures on TSA plates as described above and purified phages were spotted as 1:10 (phage:host) on each plate. All plates were incubated at 37  C for 24 h. After incubation, all clear spots were identified as positive and lytic spots on each strain of bacterial lawns and then analyzed to report the host range of isolated phages. Phages were selected based on the spot test for further analysis by efficiency of plaquing (EOP) using a modified version of previously published protocols (O’Flynn et al., 2004; Viazis, Akhtar, Feirtag, Brabban, & Diez-Gonzalez, 2011). EOP was conducted on selected phages to determine their ability of plaque formation on a diverse set of Salmonella serovars. When a phage formed a plaque on the lawn of specific Salmonella strain of interest, the efficiency of plaquing (EOP) was measured by using the Salmonella host strain as the reference strain. The EOP was determined by expressing the phage titer of the susceptible strain relative to the phage titer of the reference strain. EOP of bacteriophages was identified on the tested strains from eight different serovars, those are Typhimurium (N ¼ 15), Saintpaul (N ¼ 1), Tennessee (N ¼ 1), Enteritidis (N ¼ 3), Montevideo (N ¼ 1), Newport (N ¼ 4), Agona (N ¼ 1), Heidelberg (N ¼ 1) (Table 4).

69

England Biolabs, Ipswich, MA, USA) following the manufacturer’s protocol. All digested plugs were run on a PFGE gel in a CHEF-DR II PFGE (BioRad, Inc) for 18 h and imaged on an ethidium bromidestained agarose gel (1%, wt/vol) by a BioRad gel imaging system. The RE digested DNA images were analyzed to identify the distinct DNA features among purified phages. 2.7. Data analysis All reported results were calculated by averaging two independent experimental data in which duplicate samples with duplicate counts were tested at each time interval. Mean and standard deviations were determined using Microsoft Excel (Version 14.0; Microsoft, Corp; Redmond, WA). Two factor Analysis of variances (ANOVA) of EOP values among the phages were calculated in respect of infecting different serovars of Salmonella. First, all tested Salmonella strains were grouped by serovars. Mean EOP were measured for different Salmonella serovars for each phage, and ANOVA was performed by using Microsoft Excel. 3. Results

2.5. Morphology 3.1. Collection and isolation of phages Stock cultures of purified phages were used to image the morphology by transmission electron microscopy (TEM). Phages were diluted into Phosphate buffer saline (PBS) to avoid the gelatin microstructures and negatively stained with either ammonium molybdate or 0.5% phosphotungstic acid or 3% uranyl acetate (Ackermann, 2009b). The fixed phages on copper grid were pictured and analyzed by a Philips CM12 transmission electron microscope (TEM) with 60 kV, at the Imaging Center, College of Biological Sciences, and by a JEOL 1200 EXII TEM, with accelerating voltage from 40 kV to 120 kV, at the Characterization Facility, College of Science and Engineering, University of Minnesota. 2.6. Phage genome size determination The genome sizes of the phages were measured by pulsed field gel electrophoresis (PFGE). Phage DNA plugs were made by mixing purified phage (1:1) with 1.4% (w/v) molten PFGE grade agarose (Ultra Pure DNA Grade Agarose: BioRad, Hercules, CA). DNA plugs were treated with proteinase K (0.1 mg mL1) and lysis buffer to extract the phage DNA and stored in 0.5  Tris EDTA (TE) at 4  C for further analysis. Phage DNA was tested by PFGE to ensure the purity of isolated phage. Restriction enzyme digestion was used for further characterization of the isolated phages. These sets of DNA plugs were made with higher DNA concentration than the genome size to obtain the clear multiple bands of digested DNA. Phage DNA plugs were then digested with restriction enzyme (RE) EcoRI (New

As many as 52 waste samples were tested for the presence of bacteriophages against Salmonella. A total of 59 phages were isolated from natural waste. Twelve phages were isolated from dairy cattle manure and nine phages from swine manure using S. Typhimurium UK-1 and S. Typhimurium ATCC 14028 as host strains. A third group of phages (N ¼ 38) were obtained from raw sewage samples of a wastewater treatment facility. Eleven strains of four different Salmonella serovars were used as host for phage isolation from sewage samples. Thirty phages were isolated using S. Typhimurium, five phages with S. Enteritidis, one with S. Newport, and two with S. Saintpaul (Table 2). We could not isolate any phage with the Tennessee serovar. 3.2. Host range of phages 3.2.1. Spot testing All 59 isolated phages were capable of lysing their host strains throughout the purification process. When these phages were screened by spot testing against a total of thirty strains of six Salmonella serovars including Typhimurium, Enteritidis, Newport, Heidelberg, Saintpaul and Tennessee (Table 3), less than 25% of isolated phages (13 out of 59) formed clear plaques and were capable to lyse at least two serovars (Table 3) while the rest of them were highly specific in infecting only their host serovars and plaques were not well defined. Among phages against Typhimurium,

Table 2 Collection and isolation of bacteriophages against different Salmonella serovars from natural sources. Host bacterial strain

List of isolated phages

Number of phages

S. S. S. S. S. S. S. S. S. S. S.

STA1, STA2, STA3, STA9/STA4, STA5, STA6, STA7, STA8, STA10/STA11, STA12, STA13, STA14 STB1, STB2, STB3, STB4, STB5, STB9, STB10, STB11/STB6, STB7, STB8/STB13, STB14, STB15, STB16, STB17, STB18, STB19 STC1, STC2, STC3, STC4, STC5, STC6, STC7, STC8, STC9, STC10, STC11, STC12, STC13 STD1, STD2, STD3 STE1 STF1 STG1 SEA1, SEA2, SEA3 SEB1, SEB2 SNA1 SSA1, SSA2

14 18 13 3 1 1 1 3 2 1 2

Typhimurium UK-1 Typhimurium ATCC 14028 Typhimurium DT104 Typhimurium I600 Typhimurium I536 Typhimurium I527 Typhimurium 758 Enteritidis 2009595 Enteritidis 95657613 Newport AMO7076 Saintpaul E2008001236

70

M. Akhtar et al. / Food Control 38 (2014) 67e74

Table 3 Sensitivity of Salmonella serovars to lysis by selected isolated bacteriophages determined by spot testing. Phage

STA1 STA3 STA9 STA10 STA11 SEA1 SEA2 SSA1 SSA2 STD3 STE1 STF1

% Of positive spots against Salmonella serovars Typhimurium (N ¼ 14)

Enteritidis (N ¼ 3)

Newport (N ¼ 6)

Heidelberg (N ¼ 1)

Saintpaul (N ¼ 5)

Tennessee (N ¼ 1)

92.9 92.9 92.9 64.3 85.7 64.3 64.3 78.6 78.6 78.6 78.6 35.7

0 33.3 0 66.7 100 100 100 100 100 100 100 33.3

0 16.7 0 33.3 0 83.3 66.7 50 50 50 50 16.7

100 100 100 100 100 0 100 100 100 100 100 0

20 0 0 80 80 100 100 80 100 100 80 100

0 0 0 0 0 100 100 100 100 100 100 0

STD3 and STE1 were capable of lysing at least three other serovars in addition to the host. Two S. Enteritidis phages, SEA1 and SEA2 were isolated from two different samples using one strain, S. Enteritidis 2009595. These two phages displayed similar host range activity, but SEA1 did not lyse the Heidelberg strain. The S. Saintpaul phages, SSA1 and SSA2 had broad host range and lysed 50e 100% strains of all six serovars. SEA1 and SEA2 infected lower number of Typhimurium strains but produced larger clearing spots than SSA1 and SSA2. SEA2, SSA1, SSA2, STD3 and STE1 were the phages that formed visibly clear spots on the majority of bacterial lawns (from 50 to 100%) of all six tested serovars. 3.2.2. EOP In the process of identifying the most effective phages, the selected phages were further analyzed to determine the host range and lytic capacity by EOP. The EOP assay was conducted on several representative phages which had lysed 50% or more tested strains. Results are presented for nine phages (SEA1, SEA2, SSA1, SSA2, STE1, STA3, STA9, STA10 and STD3) which were the most efficient among all the isolated 59 phages in generating plaques against tested strains (Tables 4 and 5). SEA1 and SEA2 were found to have the broadest spectrum of lytic ability. SEA1 was found to be the most efficient with a wide range of lysis ability among all the isolated phages (according to ANOVA, p < 0.05). SEA2 showed medium to high efficiency (EOP ¼ 0.2e1) to lyse all Newport, Enteritidis, Tennessee, Saintpaul strains and the majority of tested Typhimurium strains. SEA1 was medium to highly efficient in producing plaques from most Newport, Enteritidis, Saintpaul, Tennessee and few Typhimurium strains. Similar to the spot test, SSA1 and SSA2 also lyzed a wide variety of hosts in EOP analysis. SSA1 had medium to highly efficiency (0.2e1) to infect majority of Typhimurium strains, but EOP values were low (0.001e0.2) or negative for other serovars. SSA2 had the greatest efficiency (EOP ¼ 0.5e1) infecting Typhimurium UK-1, among the phages that had been isolated with that strain. Only SSA2 phage had the capacity to infect the serovar Montevideo strain with very high efficiency (EOP ¼ 120.58). STA3, STA9 and STA10 were highly infection ability with EOP values from 0.5 to 1 against most S. Typhimurium strains. These Typhimurium phages had very low EOP values when tested against other serovars. Interestingly, STA3 and STA9 had high efficiency values (EOP ¼ 0.5e 1) while infecting the Heidelberg strain. STE1 was the only phage isolated with S. Typhimurium strain I536, and also the only phage with higher EOP values than 0.2 against three additional strains of the same serovar. However, it was the only Typhimurium phage that had medium to low EOP values against Tennessee, Newport and Enteritidis strains. The variances of EOP values among the phages in respect of infecting different Salmonella serovars were statistically different (P  0.05) with ANOVA.

3.3. Characterization of phages 3.3.1. Morphology TEM was conducted after negative staining to classify the selected phages by morphological characteristics. All of the phages that were identified by TEM (Fig. 1) were tailed phages and appeared to fall into the order Caudovirales. The isolated phages belonged to two different families, Myoviridae and Siphoviridae (Ackermann, 2009a). SEA1 and SEA2 resembled the typical morphological features of T4 Myoviridae phages with icosahedral head and contractile tail with fixation structures. The head and tail length were approx. 111 and 100 nm, respectively, for these Myoviridae phages. The Saintpaul-specific phage, SSA1 belonged to Siphoviridae with a non-contractile tail. TEM pictures of Typhimurium phages, STA9, STA1 and STA2 indicated that they were Siphoviridae, non-contractile, long tailed phages. 3.3.2. Genome size We determined the genome size for a variety for isolated phages adopting the narrow and broad host range. According to the TEM pictures (Fig. 1), phages were tailed Myoviridae or Siphoviridae phages. These typical phages have been described previously containing double stranded DNA (Ackermann, 2009a). The most variable genome sizes were observed among the screened Typhimurium phages with a range from 35 to 190 kb. The Enteritidis phages had comparatively larger genomes and had minimal size differences among each other, SEA1 and SEA2 sizes were approx. 190 and 170 kb, respectively. The Saintpaul-specific phage, SSA1 genome size was estimated to be 125 kb. 3.3.3. Restriction enzyme digestion Isolated DNAs of selected phages were digested with restriction endonuclease EcoRI. The compared digestion patterns were quite different for the digested phage DNA. EcoRI digestion of SSA1 produced17 visible bands, 22 fragments are obtained from STE2, and phage STD3 DNA were cut into 3 clear fragments. SEA1, SEA2 and STE1 were not digested at all. 4. Discussion This study isolated and partially characterized virulent and broad-spectrum Salmonella-specific bacteriophages from natural sources with the ultimate goal of developing a collection of phages to biocontrol Salmonella. It has long been recognized that the presence of bacteriophages is tightly associated with their natural hosts. Because Salmonella is a natural inhabitant of the gastrointestinal tract of animals (Guttman, Raya, & Kutter, 2005) and abundant in animal feces, various waste effluents typically provide

M. Akhtar et al. / Food Control 38 (2014) 67e74

71

Table 4 Efficiency of plaque formation (EOP) by selected bacteriophages against Salmonella serovars. Bacterial strains

SEA1

SEA2

SSA1

SSA2

STE1

STA3

STA9

STA10

STD3

S. Typhimurium I598 S. Typhimurium I527 S. Typhimurium I534 S. Typhimurium I503 S. Typhimurium I535 S. Typhimurium I649 S. Typhimurium I526 S. Typhimurium E2009005811 S. Typhimurium 14028 S. Typhimurium I758 S. Typhimurium I536 S. Typhimurium I740 S. Typhimurium DT104 S. Typhimurium UK-1 S. Typhimurium I600 S. Saintpaul E2008001236 S. Tennessee E200700502 S. Enteritidis 95657613 S. Enteritidis 2009595 S. Enteritidis I823 S. Montevideo 95573473 S. Newport AMO7076 S. Newport AMO7073 S. Newport B4442CDC S. Newport AMO5313 S. Agona FDA S. Heidelberg FSIS10

þþ þþþ þþþ  þþþ  þ þ þþþ   þþþ    þþ þþþ þ Host þþ  þþþ þþþ  þþþ  þ

þþþ þþþ þþþ  þþþ  þþþ þþ þþþ  þþ þþþ þþþ þþ þþ þþ þþþ þþþ Host þþ  þþþ þþþ þþ þþþ þþ 

þþþ þþþ þþ þ þ þþ þþþ þ þ þþ þþþ þþ  þ þþþ Host þ þ þ   þ   þ  

þ þ þþ þ þ þþ þ   þ þ þþ  þþþ þ Host  þþþ þ  þþþ  þ þ þ þ þþ

        þþ þþþ Host þþþ     þþ þ  þ    þþ þþ  

þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþ þþþ þþ þþþ Host þþþ            þþþ

þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþ þþþ þþ þþþ þþ þþþ Host þþþ            þþþ

þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþþ þþ þþþ þþ þ Host þþþ            þþ

   þ  þ    þ þþþ þ þ  Host þ           

þþþ, EOP 0.5 to 1.0; þþ, EOP 0.2 to 0.5; þ, EOP 0.001 to 0.2; , (<0.001) bacterial strain was not susceptible to phage attack.

the best source for phage isolation. In this study, swine manure, cattle manure and sewage samples were collected at various time points to isolate the lytic bacteriophages against a selected group of Salmonella serovars. Though the same set of Salmonella host strains were used for screening all effluent samples (Table 2) for isolation purposes, sewage samples were found to have the most diverse phages. Wide host range phages against Typhimurium, Enteritidis, Saintpaul and Newport serovars with clear plaques were predominantly isolated from sewage effluents. In earlier reports, pooled fecal samples have been continuously investigated as a rich source for collecting lytic Salmonella phages (Dhillon, Dhillon, Chau, LI, & Tsang, 1976; O’Flynn et al., 2006). However, Ibrahim et al. evaluated three different sources, fecal samples, intestinal contents of turkey poults, and carrier cultures of Salmonella and E. coli to determine the suitability of different habitats for lytic Salmonella phage isolation. In their study, fecal samples were the least diverse source and carrying only S. Typhimurium phages. However, carrier cultures were identified to be the most diverse source with the isolation of S. Typhimurium, S. Copenhagen and S. Heidelberg phages. The authors also reported that phages from fecal samples did not show wide host range and were able to infect 20% of the tested strains compared to 57% lysis activity of carrier culture phages (Ibrahim, 1969). Feces from swine commercial farms were investigated by Callaway et al. and indicated that phages are ubiquitous in pig manure but the population of lytic phages against S. Typhimurium might be very low (Callaway et al., 2010). On the other hand, studies with wastewater samples were found to be more successful attempts for phage isolation. Bacteriophages from sewage were able to infect broad host range and not highly serovar specific (Carey-Smith, Billington, Cornelius, Hudson, & Heinemann, 2006) or genus-restricted (Bielke et al., 2007). In our study, manure samples carried mostly temperate phages with strict host specificity which corresponds with previous findings. In contrast, lytic and wide host range phages were predominantly isolated from sewage samples of community wastewater plant. We speculate that this difference mirror the adaptation

to different environment and exploitation of different vectors. In fact, phages present in animal feces would need their bacterial host as a vector to better survive and disseminate from one GI tract to another. On the other hand, phages in water sewage can reach a wide variety of bacterial hosts (e.g., different Salmonella serovars). Lysogenic bacterium plays a role in dissemination of phage genes which is also in the evolutionary interest of the bacteriophage (Boyd & Brüssow, 2002; Desiere, Mcshan, Van Sinderen, Ferretti, & Brüssow, 2001). All of our fecal samples were collected from a single swine and dairy barn, which may also limit the presence of wide variety of Salmonella serovars in comparison to the samples from a community wastewater treatment plant. Studies with E. coli O157:H7 phages also provided similar conclusions; the virulent phage, SP15 was isolated from activated sludge at wastewater treatment plant and showed to infect a diverse range of E. coli strains whereas PP17 was found in pig feces lysing only O157:H7 strains (Yoichi et al., 2004). Several methods have been adopted previously to characterize isolated phages (Mclaughlin, Balaa, Sims, & King, 2006; O’Flynn et al., 2004). Our study was run through a specific screening scheme effluent samples were first filtered, centrifuged and enriched for phage isolation. To ensure the phage isolation from low concentration population from effluent samples, initial filtering was done to have a rich phage flora as well as removing debris and other background bacterial flora. This method also stated earlier to isolate lytic and diverse Salmonella phages from wastewater samples (Bielke et al., 2007). In comparison, chloroform was also used to clean the background flora by other researchers (Callaway et al., 2010) prior to enrichment. Studies were also described the isolation method of two lytic O157:H7 phages by processing the different fecal samples for direct enrichment without any pre filtering step (O’Flynn et al., 2004) might be indicative of having less success of isolating phages from a lower concentrated population. Interestingly, in a previous research of Salmonella phage isolation, E. coli B host strain was used in parallel to enrich phages followed by isolation with direct spot testing against S. Typhimurium. The

72

M. Akhtar et al. / Food Control 38 (2014) 67e74

Table 5 Determination of phage genome size by pulsed field gel electrophoresis (PFGE). Bacteriophage

Whole genome size (wkb)

SEA1, SEA2, STA1, SSA1 STA4, STD3

190 170 145 125 45 35

STA11 STF1 STA2, STB3, STB4, STE1 STA3, STA3, STA4, STB5, STB6, STB7

authors suggested that E. coli B was used as a very sensitive strain to aid the phage isolation from a lower phage concentrated environment (Callaway et al., 2010). In our approach, host strains of eight different Salmonella serovars were used as a mechanism to increase the chances of finding lytic phages that can infect a variety of serovars. In this study, lytic phages were initially isolated based on the capacity of forming clear spots and plaques on host lawns. While purifying, plaque uniformity was carefully observed because uniform plaques are considered (Ibrahim, 1969) to be critical to for obtaining a single phage strain isolation. A large portion of isolated phages was not selected for further characterization due to their inability to make clear plaques with host serovars. The host range of the phages was determined by spot testing and EOP for a wide variety of Salmonella serovars. Spot testing was used as the first criteria to select wide host range phages which were eventually selected for EOP analysis to determine the efficiency of infecting capacity relative to host strains (O’Flynn et al., 2004; Viazis et al., 2011; Viscardi et al., 2008). Based on the spot test (Table 3) and EOP results (Table 4), SEA1 and SEA2

exhibited the most extensive lysis over a wide range of Salmonella compared to other screened phages. SSA1 was efficient to infect several Typhimurium strains other than host Saintpaul. In contrast, SSA2 lysed serovars Enteritidis and Montevideo strains as well as Typhimurium. Spot test results indicated that Typhimurium phages were moderately specific in infecting different Salmonella serovars, however, most of them appeared to be strictly host specific in EOP tests. Interestingly, STA3 and STA9 were able to lyse Heidelberg which was a rare phenomenon for other isolated phages. Similar to this study, previously isolated virulent phages, st104a, st104b and Felix 01 were also reported to be infective over a wider range of serovar hosts (O’Flynn et al., 2006). However, supporting to our findings, several research groups also observed the high degree of specificity among Salmonella phages. In one report, 34 Salmonella phages were isolated from sewage samples and host range was determined by spot testing against 53 different serovars. Host range comparison data revealed a high specificity of most of those phages were only effective to lyse the host strains, but only 5 phages had a wider range and none of them were able to infect all tested strains (Heringa, Kim, Jiang, Doyle, & Erickson, 2010). In accordance with our findings, Callaway et al. reported the lytic phages against S. Typhimurium showed a high degree of specificity (Callaway et al., 2010). In contrast, Ibrahim (1969) observed a variation in host range of isolated E. coli phages that were capable of lysing Salmonella as well. One study (Mclaughlin & King, 2008) showed an interesting phenomenon that phages isolated from swine lagoons were able to infect the ATCC reference strains but not effective against the lagoon strains of Salmonella. The authors of the latter study hypothesized that long term co-culture of host and phages

Fig. 1. Transmission electron micrographs of negatively stained bacteriophages that caused lysis to Salmonella serovars.

M. Akhtar et al. / Food Control 38 (2014) 67e74

can be a selection factor for developing a phage-resistant bacterial population. Our results indicated that the majority of the phages were specific for the host serovars, somewhat specific to the strain and few were able to cross-infect other serovars. The distinct profiles of Salmonella susceptibility may be explained by the non-specific binding receptors on the bacterial host or different resistant mechanisms during phage infection (Duckworth, Glenn, & Mccorquodale, 1981). Most dsDNA phage DNAs are known to be cased into an icosahedral shell of protein attached to a tail (Calender & Inman, 2005). The tail-end contains fiber proteins which help them to recognize the receptor molecules on the bacterial cell wall and also restrain them to bind onto non-specific bacterial cells (Kutter & Sulakvelidze, 2005; Singh et al., 2010). Also, bacteria could resist the phage infections through superinfection exclusion (Sie) systems. Sie systems are predicted to be the membrane associated proteins which prevent the phage DNA to enter into bacterial hosts. Though the molecular mechanisms are not clearly described yet, but the prophages of some Enterobacteriaceae species were identified to have the genes that encode the Sim and SieA proteins. The virulent coliphage, T4 possess two different Sie systems which offers resistance against other T-even like phages by blocking the phage DNA injection in Gram negative bacteria. Interestingly, S. enterica subsp. Enterica serovar Typhimurium carrying lysogenic phage P22 was also reported to have SieA systems and insensitive to other phage (e.g. L, mG178 and mG40) infections (Hofer et al., 1995; Labrie, Samson, & Moineau, 2010). All the TEM pictures of isolated phages resembled either Siphoviridae or Myoviridae phages with tail and icosahedral heads. All samples contained tailed phages, most of those were Siphoviridae. STA9 appeared similar to previously identified virulent phage MB78 (Joshi, Siddiqi, Rao, & Chakravorty, 1982), and all of Siphoviridae phages looked dissimilar to the virulent phage of S. Typhimurium, Felix-O1 (Lindberg, 1967; O’Flynn et al., 2006) as well as to the known temperate phage, P22. Only two Enteritidis phages, SEA1 and SEA2 appeared as Myoviridae T4 like phage (O’Flynn et al., 2004) and isolated from sewage. These results suggest the possible presence of diverse phage morphotype in sewage and manure samples as identified by others (Mclaughlin et al., 2006; O’Flynn et al., 2006). The EcoRI digestion patterns of the Saintpaul and Typhimuriumspecific phages were distinct and did not match with other digested phages. The two Myoviridae phage restriction profiles were similar; EcoRI could not restrict the DNA of SEA1 and SEA2 (Fig. 1). Interestingly, the digestion profiles of SEA1 and SEA2 matched with the phage appearance, they shared the same morphotype. However, spot test and EOP data (Tables 3 and 4) indicated these two phages exhibited differences in infection capacity. This phenomenon was also identified in previous study where two isolated Salmonella phages were indistinguishable with NdeI and DraI digestion but showed differences in host range analysis, the authors suggested that those two phages were isolated from different sources, but they could be closely related with enough divergent to exhibit the differences in infecting hosts. 5. Conclusions Our study described that bacteriophages were ubiquitous among different waste samples but finding a potent lytic phage infecting a wide range of bacteria might be critical. The high degree of specificity among majority of phages may pose a barrier to reduce the contamination of the large group of Salmonella serovars. Adding these lytic phages to the existing small bacteriophage library would help to treat different Salmonella infections.

73

This study presents an alternate option to biocontrol of the undesirable foodborne pathogen, Salmonella. Any effective implementation of these phages must have a clear understanding with high quality supporting data and effective trials to meet the regulatory standards. Though the results have demonstrated the lysis efficiency of these phages, most importantly, the ability to infect a wide range of Salmonella hosts; these isolated phages need to be further explored as an effective antimicrobial to control the target pathogens on different matrices. It is fundamental to characterize and understand the microbial ecosystem in molecular level prior to utilize these bacteriophages as biological control approach to improve food safety. Acknowledgment We thank Mike Magee and Mike Larose, Rice Lake Wastewater Treatment facilities, Rice Lake, WI for their cooperation and help to provide sewage samples for this study. We would like to thank Dr. Andrew Brabban and Dr. Elizabeth Kutter of Evergreen State College, WA for all of their suggestions help with developing the methods. This project was funded by the USDA’s Integrated Organic Program under award No. 2007-51300-03796. References Ackermann, H.-W. (2009a). Phage classification and characterization. In M. R. J. A. Clokie, & A. M. Kropinski (Eds.), Bacteriophages, Methods and protocols (vol. 1; pp. 127e140). New york, NY 10013, USA: Humana Press, c/o Springer ScienceþBusiness Media, LLC. Ackermann, H.-W. (2009b). Basic phage electron microscopy. In M. R. J. A. Clokie, & A. M. Kropinski (Eds.), Bacteriophages, Methods and protocols (vol. 1; pp. 127e140). New york, NY 10013, USA: Humana Press, c/o Springer ScienceþBusiness Media, LLC. Bielke, L., Higgins, S., Donoghue, A., Donoghue, D., & Hargis, B. M. (2007). Salmonella host range of bacteriophages that infect multiple genera. Poultry Science, 86, 2536e2540. Boyd, E. F., & Brüssow, H. (2002). Common themes among bacteriophage-encoded virulence factors and diversity among the bacteriophages involved. Trends in Microbiology, 10, 521e529. Calender, R., & Inman, R. (2005). Phage biology. In M. K. Waldor, D. I. Friedman, & S. L. Adhya (Eds.), Phages. their role in bacterial pathogenesis and biotechnology (pp. 18e36). Washington, DC, USA.: ASM Press. Callaway, T. R., Edrington, T. S., Brabban, A., Kutter, E., Karriker, L., Stahl, C., et al. (2010). Occurrence of Salmonella-specific bacteriophages in swine feces collected from commercial farms. Foodborne Pathogens and Disease, 7, 851e856. Carey-Smith, G. V., Billington, C., Cornelius, A. J., Hudson, J. A., & Heinemann, J. A. (2006). Isolation and characterization of bacteriophages infecting Salmonella spp. FEMS Microbiology Letters, 258, 182e186. Carlton, R. M., Noordman, W. H., Biswas, B., De Meester, E. D., & Loessner, M. J. (2005). Bacteriophage P100 for control of Listeria monocytogenes in foods: genome sequence, bioinformatic analyses, oral toxicity study, and application. Regulatory Toxicology and Pharmacology, 43, 301e312. Centers for Disease Control and Prevention (CDC). (2001). Outbreaks of multidrugresistant Salmonella Typhimurium associated with veterinary facilities-Idaho, Minnesota, and Washionton, 1999. Morbidity and Mortality Weekly Report, 50, 701e704. Desiere, F., Mcshan, W. M., Van Sinderen, D., Ferretti, J. J., & Brüssow, H. (2001). Comparative genomics reveals close genetic relationships between phages from dairy bacteria and pathogenic Streptococci: evolutionary implications for prophage-host interactions. Virology, 288, 325e341. Dhillon, T. S., Dhillon, E. K., Chau, H. C., LI, W. K., & Tsang, A. H. (1976). Studies on bacteriophage distribution: virulent and temperate bacteriophage content of mammalian feces. Applied and Environmental Microbiology, 32, 68e74. Duckworth, D. H., Glenn, J., & Mccorquodale, D. J. (1981). Inhibition of bacteriophage replication by extrachromosomal genetic elements. Microbiological Reviews, 45, 52e71. Greer, G. G. (2005). Bacteriophage control of foodborne bacteria. Journal of Food Protection, 68, 1102e1111. Guttman, B., Raya, R., & Kutter, E. (2005). Basic phage biology. In E. Kutter, & A. Sulakvelidze (Eds.), Bacteriophages: Biology and applications (pp. 29e66). USA: CRC Press. Heringa, S. D., Kim, J., Jiang, X., Doyle, M. P., & Erickson, M. C. (2010). Use of a mixture of bacteriophages for biological control of Salmonella enterica strains in compost. Applied and Environmental Microbiology, 76, 5327e5332. Hoffer, B., Ruge, M., & Dreiseikelmann, B. (1995). The superinfection exclusion gene (sieA) of bacteriophage P22: identification and overexpression of the gene and localization of the gene product. Journal of Bacteriology, 177, 3080e3086. Ibrahim, A. A. (1969). Bacteriophage typing of Salmonella. I. Isolation and host range study of bacteriophages. Applied Microbiology, 18, 444e447.

74

M. Akhtar et al. / Food Control 38 (2014) 67e74

Joshi, A., Siddiqi, J. Z., Rao, G. R., & Chakravorty, M. (1982). MB78, a virulent bacteriophage of Salmonella typhimurium. Journal of Virology, 41, 1038e1043. Klieve, A. (2005). Bacteriophages. In H. Makkar, & C. Mcsweeney (Eds.), Methods in Gut microbial Ecology for Ruminants (pp. 39e46). Netherlands: IAEA. Labrie, S. J., Samson, J. E., & Moineau, S. (2010). Bacteriophage resistance mechanisms. Nature Reviews Microbiology, 8, 317e327. Lindberg, A. A. (1967). Studies of a receptor for felix O-1 phage in Salmonella minnesota. Journal of General Microbiology, 48, 225e333. Mclaughlin, M. R., Balaa, M. F., Sims, J., & King, R. (2006). Isolation of Salmonella bacteriophages from swine effluent lagoons. Journal of Environmental Quality, 35, 522e528. Mclaughlin, M. R., & King, R. A. (2008). Characterization of Salmonella bacteriophages isolated from swine lagoon effluent. Current Microbiology, 56, 208e213. Miao, E., & Miller, S. (1999). Bacteriophages in the evolution of pathogen-host interactions. Proceedings of the National Academy of Sciences of the United States of America, 96, 3. Mullan, W. M. A. (2002). Dairy science and food technology. Bacteriophage enumeration [On-line] UK: Available: http://www.dairyscience.info/phage/ enumeration.asp?Tab¼2. O’Flynn, G., Coffey, A., Fitzgerald, G. F., & Ross, R. P. (2006). The newly isolated lytic bacteriophages st104a and st104b are highly virulent against Salmonella enterica. Journal of Applied Microbiology, 101, 251e259. O’Flynn, G., Ross, R. P., Fitzgerald, G. F., & Coffey, A. (2004). Evaluation of a cocktail of three bacteriophages for biocontrol of Escherichia coli O157:H7. Applied and Environmental Microbiology, 70, 3417e3424. Scallan, E., Hoekstra, R. M., Angulo, F. J., Tauxe, R. V., Widdowson, M. A., Roy, S. L., et al. (2011). Foodborne illness acquired in the United Statesemajor pathogens. Emerging Infectious Diseases, 17, 7e15.

Singh, A., Arya, S. K., Glass, N., Hanifi-Moghaddam, P., Naidoo, R., Szymanski, C. M., et al. (2010). Bacteriophage tailspike proteins as molecular probes for sensitive and selective bacterial detection. Biosensors & Bioelectronics, 26, 131e138. Tauxe, R. V., Doyle, M., Kuchenmüller, T., Schlundt, J., & Stein, C. E. (2010). Evolving public health approaches to the global challenge of foodborne infections. International Journal of Food Microbiology, 139(Suppl. 1), S16eS28. The Food and Drug Administration (FDA). (2006). FDA approval of Listeria-specific bacteriophage preparation on ready-to-eat (RTE) meat and poultry products. http://www.cfsan.fda.gov/..dms/opabacqa.html. Twest, R., & Kropinski, A. (2009). Bacteriophage enrichment from water and soil. In M. R. J. A. Clokie, & A. M. Kropinski (Eds.), Bacteriophages, Methods and protocols (vol 1; pp. 15e22). New york, NY 10013, USA: Humana Press, c/o Springer ScienceþBusiness Media, LLC. United States Department of Agriculture-Economic Research Service (USDA-ERS). (2001). ERS estimates food borne disease costs at $6.9 billion per year. http:// www.ers.usda.gov/publications/aer741/aer741.pdf Accessed 16.10.09. Viazis, S., Akhtar, M., Feirtag, J., Brabban, A. D., & Diez-Gonzalez, F. (2011). Isolation and characterization of lytic bacteriophages against enterohaemorrhagic Escherichia coli. Journal of Applied Microbiology, 110, 1323e1331. Viscardi, M., Perugini, A. G., Auriemma, C., Capuano, F., Morabito, S., Kim, K. P., et al. (2008). Isolation and characterisation of two novel coliphages with high potential to control antibiotic-resistant pathogenic Escherichia coli (EHEC and EPEC). International Journal Antimicrobial Agents, 31, 152e157. Yoichi, M., Morita, M., Mizoguchi, K., Fischer, C., Unno, H., & Tanji, Y. (2004). The criterion for selecting effective phage for Escherichia coli O157:H7 control. Biochemical Engineering Journal, 19, 221e227.