Reducing Salmonella Horizontal Transmission During Egg Incubation by Phage Therapy

Abstract

Salmonella is a serious problem for both animal production and public health worldwide. Contaminated poultry is the main vehicle of Salmonella and the most important serotype is Salmonella enterica serovar Enteritidis. In order to test the efficiency of bacteriophages to treat Salmonella enterica serovar Enteritidis infections in poultry, a cocktail of two phages, F1055S and F12013S, isolated from chicken litter was applied by aerosol spray on fertile eggs challenged with Salmonella Enteritidis. To the best of our knowledge, this is the first experiment in which phages were applied by spray on fertile eggs. Two groups of eggs were challenged with Salmonella (3×10⁸ colony-forming units [CFU]/mL) and one of them was treated with the phage suspension (2×10⁶ plaque-forming units [PFU]/mL). A third group was used as nonchallenged and nontreated control. The phage treatment of challenged Salmonella eggs reduced the disease symptoms in the chicks. The arthritis and pasting after 8 days in the challenged and treated group were similar to those normally occurring in chicks (nonchallenged and nontreated chick control group) (p=1.000 and p=0.828, respectively, for arthritis and pasting) and were significantly lower than the challenged but nontreated ones (p=0.017 and p=0.002 for arthritis and pasting, respectively). The phage-treated group did not lose weight, showing an average weight similar to that of the nonchallenged control group and higher than that of the challenged nontreated group. The results of this study suggest that the application of phages by aerosol spray during the transfer of the eggs from incubators to hatchers may be an effective and inexpensive approach for reducing the horizontal transfer of Salmonella in poultry.

Introduction

S almonella is a major cause of foodborne disease in the developed world, causing gastrointestinal infections, commonly acquired through the consumption of undercooked food products derived from poultry (Favrin et al., 2001).

Salmonella may be introduced into poultry flocks horizontally from different sources, including direct transmission between flocks, contaminated feeds, biologic vectors (insects, rodents, wild birds, humans), as well as via vertical transmission (Toro et al., 2005). Salmonella Enteritidis is one of the most important Salmonella serotypes associated with chicken meat and eggs (Fiorentin et al., 2005a). The factors that are responsible for its epidemic spread still remain unclear. Epidemiological investigations indicate that laying flocks become infected directly from the farm environment (Van de Giessen et al., 1994; Omwandho and Kubota, 2010). However, it is also known that egg contamination by Salmonella Enteritidis may occur by vertical transmission in the reproductive tract before deposition of the shell (Omwandho and Kubota, 2010).

Traditional measures of salmonellosis prevention and control include the use of prebiotics and probiotics, safety measures, vaccination and, mainly, antibiotics (Borie et al., 2008). However, concerns related to antibiotic-resistant bacteria have stimulated interest in alternative treatments for bacterial infections. Among these therapies, a special interest has been given to phage therapy (Fiorentin et al., 2005b).

Recent research using bacteriophages has been focused on treating enteric infections in poultry (Bielke et al., 2007). The bacteriophages could be useful as therapeutic or prophylactic agents providing a natural, nontoxic, feasible, and inexpensive approach for Salmonella control in poultry (Borie et al., 2009). Nonetheless, additional work is needed in order to develop effective and practical phage therapy protocols as well as to select the appropriate method of phage delivery.

In most studies undertaken to control Salmonella Enteritidis, phages were administered orally. Several studies have shown that phages thus administered significantly reduced (by 3.5–4 log) Salmonella Enteritidis from ceca of chickens (Fiorentin et al., 2005a; Atterbury et al., 2007; Filho et al., 2007). Higgins et al. (2007) also observed reductions of Salmonella Enteritidis from the ceca of chicks treated orally, although the differences were not significant. Sklar and Joerger (2001) showed that the oral administration of phages to control infections of Salmonella Enteritidis in chickens reduced the cecal counts (reductions of 0.3–1.3 log), but only when phages were administered through feed particles (as delivery vehicles). One concern arising when phages are applied is the fact that the viability of orally administered phage may be rapidly reduced under the acidic conditions of the gastrointestinal tract of the chicken, in the presence of enzymes as well as of other digestive compounds such as bile. To address this issue, some authors ascertained the phage-based aerosol-spray treatment of chickens. Goode et al. (2003) determined that phages could control Salmonella Enteritidis on chicken skin by spray application (reductions up to 2 log). Borie et al. (2008, 2009) used a cocktail of three phages to reduce Salmonella Enteritidis in chickens, thus showing that the treatment by coarse spray significantly reduced the incidence of Salmonella in the ceca.

To our knowledge, no attempt was made to assess the efficiency of phages to control Salmonella infections in poultry by spraying fertile eggs during incubation. This is a period in which infection spreads easily, particularly during the hatching process (Russell, 2010).

The objective of this study was to assess the ability of bacteriophages administered by aerosol spray on fertile eggs during incubation in reducing Salmonella Enteritidis infection in poultry.

Materials and Methods

Experimental chicken eggs

Fertile eggs from Cobb heavy breeders free of Salmonella were used. Chickens were cared for by using procedures designed and conducted in accordance with the principles and specific guidelines of animal welfare of the Federation of Laboratory Animal Science Association (Van Zutphen et al., 2001). Such guidelines are based on the European Council Directive of November 24, 1986 (86/609/EEC) and regard the protection of animals used for scientific experimental purposes. The work was carried out in the Laboratory Controlvet, which is licensed by the Portuguese General Directorate of Veterinary for the care and use of animals (birds) under number 36 of the decree no.1005/92 of October 23.

Salmonella isolation

Salmonella Enteritidis was isolated from chicken meat, feces, litter, fertile eggs, and organs of infected birds collected from Portuguese industrial poultry facilities and domestic fowl across the country (North, Center, and South of Portugal). The detection of Salmonella was performed according to the International Organization for Standardization (ISO, 2002). Twenty-five grams of each sample were inoculated in 225 mL of buffered peptone water (Biokar Diagnostics, BK131GC) and incubated at 37°C for 21±1 h. One milliliter of enriched sample was inoculated in 9 mL of Müller-Kauffmann Tetrathionate-Novobiocin Broth (Biokar Diagnostics, BK169) and incubated at 37°C for 21±1 h. A 10-μL loop of tetrathionate medium was seeded in SM ID Medium (bioMérieux, 43291) and incubated at 37°C for 21±1 h. Typical Salmonella colonies were confirmed by inoculation on triple sugar iron agar (Biokar Diagnostics, BK059HA) at 37°C for 24 h and by slide agglutination using serotype-specific antisera (BD, 228181). Salmonella isolates were stored in Nutrient Broth (Oxoid, CM0067) with 20% glycerol (AppliChem, A1123) at −80°C.

Bacteriophage isolation and purification

Phages were isolated from samples of chicken litter collected from local poultry houses. Samples were added to 250 mL of a Salmonella Enteritidis culture (3–4 h culture) in Buffered Peptone Water (Biokar Diagnostics, BK131GC), incubated overnight at 37°C, stirred (120 rpm), and centrifuged at 9000×g for 10 min (Selecta, Mixtasel). The supernatant was filtered through a 0.22-μm syringe filter (Orange Scientific, 1520012). The spot test was used as an initial method to check for the presence of phages. Layers of 3 mL of Mueller Hinton Agar (Biokar Diagnostics, BK048HA), previously inoculated with 100 μL of each Salmonella strain (6–8 h culture), were spotted with 10 μL of the phage filtrate. Plates were incubated overnight at 37°C. A clear zone in the plate indicated the presence of phages (Vieira et al., 2012).

To isolate the phages, serial 10-fold dilutions of phage filtrate were prepared in Tryptone-salt Broth (Biokar Diagnostics, BK014HA). One hundred microliters of the respective host strains were grown in buffered peptone water for 3–4 h at 37°C and inoculated on Mueller Hinton Agar plates. Each plate was spotted with 10 μL of bacteriophage dilution and incubated overnight at 37°C. Isolated lysis plaques with different morphologies were selected. Each distinct lysis plaque was sequentially passed on plates three times so as to purify the phage. Five pure bacteriophage stocks were produced and two (F1055S and F12013S) were selected. Both encompassed the broadest host range against Salmonella Enteritidis as well as uniform lysis plaques in order to be used in the phage therapy experiments. The two phages were pooled for use as a cocktail (cocktail, F1055S and F12013S titer of ∼2×10⁶ PFU mL⁻¹) and stored at 4°C. Bacteriophage stocks were kept at −20°C with 20% of glycerol (AppliChem, A1123) for long-term use.

Bacteriophage host range determination

Bacterial susceptibility to bacteriophage was assayed for the 640 Salmonella isolates obtained from the meat, feces, litter, fertile eggs, and organs of infected birds. The spot test was used to detect bacterial infection. Three independent experiments were carried out for each phage.

Bacteriophage morphological characterization

Bacteriophage particles were sedimented at 25,000×g for 60 min using a Beckman J2-21 (Palo Alto, CA) centrifuge. Bacteriophages were washed twice in 0.1 M ammonium acetate (pH 7.0), deposited on copper grids (Ernest F, Fullam, Clifton Park, NY) with carbon-coated Formvar films (Canemco & Marivac, Quebec, Canada), stained with 2% potassium phosphotungstate (pH 7.2) and examined in a Philips EM 300 electron microscope. The phages were classified in accordance with the International Committee of Taxonomy of Viruses guidelines.

Characterization of bacteriophage lysis plaques

The size and turbidity of the lysis plaques were determined by using Mueller Hinton Agar following overnight incubation at 37°C in the presence of the respective host.

Phage aerosol spray treatment of fertile eggs

After 18 days of incubation, 90 fertile eggs weighing the same, taken from Salmonella-free chickens were randomly divided into three groups and incubated at 37.5°C±0.2°C. The control group (Group 1) was maintained in a separate room within a small incubator measuring 32×33×10 cm. The other two groups (Groups 2 and 3) were placed in a different room, in distinct incubators with the same dimensions of that of Group 1, and challenged with 100 mL of 3×10⁸ CFU/mL of Salmonella Enteritidis by aerosol spray. The control group (Group 1) was sprayed with the same volume of Salmonella free-buffered peptone water. Immediately afterwards, the challenged group (Group 3) was treated with 100 mL of bacteriophage suspension of 2×10⁶ PFU/mL by aerosol spray. The other two groups (Groups 1 and 2) were treated with the same volume of phage free-buffered peptone water. The Salmonella culture, phage suspension, and buffered peptone water were applied with a manual spray, delivering approximately 1 mL per second.

After hatching, all chicks were placed in brooder rings with both feed and water ad libitum and maintained at an appropriate temperature and humidity for 8 days. During this time, the chicks were examined for limping (arthritis) and accumulation of feces around the cloacae (pasting). Afterwards, all chicks were humanely euthanized with isoflurane (Abbott, 05260-05) to obtain samples of heart, liver, spleen, and ceca to detect Salmonella. The heart, liver, and spleen of each chick were pooled and analyzed as one sample. These samples were analyzed for the presence of Salmonella as described previously.

Statistical analyses

All data were analyzed with SPSS v15.00 software. The Kruskal–Wallis test was used to evaluate the differences among treatments. The Mann–Whitney test was used to analyze whether the differences between groups were significant. Significance was determined at p<0.05.

Results

Isolation and characterization of bacteriophages

A total of five Salmonella phages were isolated from the poultry litter. The host range of each phage was determined against the 640 Salmonella Enteritidis isolates. On the basis of the lytic spectrum data, two phages, F1055S and F12013S, with the widest host range against Salmonella, were selected for phage therapy experiments. Examination of the phages by transmission electron microscopy indicated that both bacteriophages (head ∼55 nm and tail length ∼95 nm) are members of the Siphoviridae family of double-stranded DNA phages (Fig. 1). The two phages produced uniform lysis plaques and different infection spectra against the Salmonella Enteritidis isolates. The phages F1055S and F12013S infected 67% and 78% of the Salmonella isolates, respectively. The two phages infected 93% of the Salmonella isolates.

FIG. 1.

Electron microphotographs of bacteriophages isolated from poultry litter. Left side shows phage F1055S and right side shows phage F12013S (magnification 200,000×).

Phage aerosol spray treatment of fertile eggs

Some of the incubated fertile eggs did not hatch (nine, seven, and six in Groups 1, 2, and 3, respectively) and so were not considered (Table 1).

Table 1.

Number of Surviving Birds, Average Weight, Number of Birds with Arthritis and Pasting (Number of Positive/Percentage of Positive), and Presence of Salmonella in the Ceca and Other Organs (Pooled Heart, Liver, and Spleen) in the Three Groups at the End of the Experiment

Bird group	Treatment	No. of birds	Average weight (g)	Arthritis	Pasting	Birds positive for Salmonella in ceca	Birds positive for Salmonella in other organs
1	Nonchallenged and nontreated control	21	103.81	0/0	3/14	0/0	0/0
2	Challenged and nontreated	23	95.65	5/22	14/61	19/83	23/100
3	Challenged and treated	24	104.20	0/0	4/17	10/42	21/88

The average weight of the chicks in the nonchallenged control group (Group 1) at the end of the experiment was similar (p=0.820) to that of the challenged and phage-treated group (Group 3). The average weight of the challenged nontreated birds (Group 2) was significantly lower than that of Groups 1 (p=0.018) and 3 (p=0.002).

The number of chicks with arthritis and pasting at the end of the experiment was significantly higher in Group 2 than in Groups 1 (p=0.025 and p=0.002 for arthritis and pasting, respectively) and 3 (p=0.017 and p=0.002, respectively). The number of birds with arthritis and pasting was similar in Groups 1 and 3 (p=1.000 and p=0.828).

The birds in Group 1 remained free of Salmonella throughout the experiment (Table 1). In Group 2, Salmonella was isolated from 83% of cecal samples and from 100% of the other organ (pooled heart, liver, and spleen) samples (Table 1). In Group 3, Salmonella was taken from 42% of the cecal samples and from 88% of the pooled samples (Table 1). The differences in the frequency of recovery of Salmonella between Groups 1 and 2 were significant for ceca and for the other organ samples (p=0.000 for both ceca and pooled samples). The differences between Group 2 and Group 3 were significant for cecal samples (p=0.009) but not for other organ pooled samples (p=0.083).

Discussion

The selection of appropriate bacteriophages, the phage delivery method, and the life stage (eggs, chicks, or chickens) at which phage therapy is applied are key factors in the success of phage-mediated control of Salmonella in poultry. In this study, phages were administered through aerosol spray on fertile eggs toward the end of their incubation.

The two phages applied produced different infection spectra against several Salmonella Enteritidis isolates, suggesting that they were not the same. Both phages were mixed and used as a phage cocktail to reduce the possibility of bacterial resistance developing (Filho et al., 2007).

The results indicated that the two bacteriophages can reduce Salmonella infections in poultry. Both phages F1055S and F12013S infected several Salmonella Enteritidis isolates obtained from chicken meat, feces, litter, fertile eggs, and organs, reducing Salmonella Enteritidis infection in chicks. The phages were applied by spray. This raises the possibility that Salmonella reduction is due to physical removal and/or inhibitory components present in the buffered peptone water. However, this is unlikely because the challenged nontreated group was sprayed in the same way with phage-free buffered peptone water. Although research on bacteriophage treatment of poultry diseases has focused mainly on oral administration (Sklar and Joerger, 2001; Fiorentin et al., 2005a; Atterbury et al., 2007; Filho et al., 2007; Higgins et al., 2007), this delivery method, even when effective bacteriophages are used, can affect the viability of the phages due to the harsh conditions of the gastrointestinal tract. This can explain the inconsistent results of phage therapy experiments, when phages are administered orally. Actually, Sklar and Joerger (2001) showed a higher efficacy of phage therapy against Salmonella Enteritidis through oral administration, when phages were administered by using feed particles (as delivery vehicles). The same phages were not as effective when administered in the starter feed or in drinking water.

In this study, phages were administered through aerosol spray, which allows bacteria to be inactivated in the environment of the egg/chick, avoiding phage exposure to the gastrointestinal tract. The results suggest that Salmonella-specific bacteriophages applied by spray can be an efficient, safe, and inexpensive method to reduce Salmonella infections in poultry. The phage treatment by spraying challenged eggs with Salmonella Enteritidis after 18 days of incubation reduced the disease symptoms in the chicks in comparison to the nontreated challenged control group. Eight days after hatching, the arthritis and pasting were similar to that normally occurring in the poultry and were significantly lower in phage-treated chicks than in nontreated ones. Moreover, the phage-treated group did not lose weight. The average weight of phage-treated chicks was similar to that of the control group and higher than that of the challenged nontreated group.

The treatment via spray with the phage cocktail reduced Salmonella recovered from chick ceca (reduction of 50% relatively to the challenged nontreated chicks). However, the reduction in the pool of the other organs (heart, liver, and spleen) was not as effective. Because a high dose of bacteria was used, it is understandable that bacteriophages did not clear all the Salmonella from the infected chicks. However, only the frequency of Salmonella recovered was determined and not the reduction in the concentration of Salmonella in the different organs. Consequently, it is not possible to determine the magnitude of Salmonella reduction in the treated chick organs. Nevertheless, similar studies of phage therapy to control Salmonella Enteritidis in poultry have shown that, for higher frequencies of Salmonella recovery in ceca, liver, and spleen, the reduction of Salmonella concentration of treated birds, with a single dose of phages (as in this study), was of 3.5 log (Fiorentin et al., 2005a). Moreover, the low detection rate of Salmonella in ceca in comparison to the other organs could also be due to the presence of viable phages in the ceca and not in the other internal organs. In order to achieve a better understanding of the differences between ceca and other organs, more studies including phage counts in the different organs are needed.

In most studies of phage therapy to control infections by Salmonella Enteritidis, the challenge and the phage treatment have been done on chicks. In this study, they were done to the eggs shortly before the end of their incubation. As Salmonella Enteritidis infection in poultry is transmitted primarily within the farm environment (Van de Giessen et al., 1994) and spreads during the process of egg incubation, particularly during the hatching process (Russell, 2010), the results suggest that the application of phages by spray during the transfer of the eggs from incubators to hatchers can be an effective and inexpensive approach in reducing the horizontal transfer of Salmonella Enteritidis.

Although the purpose of the phage treatment of the eggs was to reduce the horizontal transfer of Salmonella Enteritidis in poultry, it is believed that the phages applied to the eggs can seep through the porous eggshell into the egg itself. Such has been observed in bacteria that are much larger than phages (Messens et al., 2005), thus mitigating the effect of vertical transmission (i.e., if the phages maintain their viability inside the eggs until hatching). In this way, phages could inactivate Salmonella present in infected eggs, and even reduce the intestinal colonization and subsequent fecal excretion of Salmonella in chicks after their hatching. The reduction in fecal excretion will lead to a decrease in the environment contamination and consequently the risk of horizontal contamination might also decline.

Further studies should be performed to evaluate the lysogenic potential of the two phages in order to develop a safe treatment, thus avoiding the expression of possible virulence genes (i.e., evaluate the lysogenic conversion potential). In addition, it would also be important to determine the survival of the phages inside noninfected eggs.

Footnotes

Acknowledgments

We would like to thank the Controlvet Laboratory that financed the study and CESAM for funding the research group of University of Aveiro (project Pest-C/MAR/LA0017/2011). We are grateful to Professor Hans Ackerman (Laval University, Quebec, Canada) for the TEM observation and morphological characterization of the phages and to Dr. Ana Paiva for the correction of the English text.

Disclosure Statement

No competing financial interests exist.

References

Atterbury

, Van Bergen

MAP

, Ortiz

, Lovell

, Harris

, De Boer

, Wagenaar

, Allen

, Barrow

. Bacteriophage therapy to reduce Salmonella colonization of broiler chickens. Appl Environ Microbiol, 2007; 73:4543–4549.

Bielke

, Higgins

, Donogue

, Hargis

, Tellez

. Use of wide-host-range bacteriophages to reduce Salmonella on poultry products. Int J Poult Sci, 2007; 6:754–757.

Borie

, Zurita

, Sánchez

, Rojas

, Santander

, Robeson

. Prevention of Salmonella enterica subspecie enterica serotype Enteritidis (Salmonella Enteritidis) infection in chickens using a bacteriophage. Arch Med Vet, 2008; 40:197–201.

Borie

, Sánchez

, Navarro

, Ramírez

, Morales

, Retamales

, Robeson

. Aerosol spray treatment with bacteriophages and competitive exclusion reduces Salmonella Enteritidis infection in chickens. Avian Dis, 2009; 53:250–254.

Favrin

, Jassim

, Griffiths

. Development and optimization of a novel immunomagnetic separation bacteriophage assay for detection of Salmonella enterica serovar Enteritidis in broth. Appl Environ Microbiol, 2001; 67:217–224.

Filho

, Higgins

, Gaona

, Wolfenden

, Tellez

, Hargis

. Ability of bacteriophages isolated from different sources to reduce Salmonella enterica serovar Enteritidis in vitro and in vivo. Poult Sci, 2007; 86:1904–1909.

Fiorentin

, Vieira

, Barioni

Jr . Use of lytic bacteriophages to reduce Salmonella Enteritidis in experimentally contaminated chicken cuts. Braz J Poult Sci, 2005a. 7:255–260.

Fiorentin

, Vieira

, Barioni

Jr . Oral treatment with bacteriophages reduces the concentration of Salmonella Enteritidis PT4 in caecal contents of broilers. Avian Pathol, 2005b. 34:258–263.

Goode

, Allen

, Barrow

. Reduction of experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic bacteriophages. Appl Environ Microbiol, 2003; 69:5032–5036.

10.

Higgins

, Higgins

, Bielke

, Hargis

. Selection and application of bacteriophages for treating Salmonella enteritidis infections in poultry. Int J Poult Sci, 2007; 6:163–168.

11.

[ISO] International Organization for Standardization. ISO 6579. Microbiology of food and animal feeding stuffs—Horizontal method for detection of Salmonella spp. 6579. 2002.

12.

Messens

, Grijspeerdt

, Herman

. (2005b) Eggshell penetration by Salmonella: A review. World's Poultry Sci J, 2005; 61:71–85.

13.

Omwandho

, Kubota

. Salmonella enterica serovar Enteritidis: A mini-review of contamination routes and limitations to effective control. JARQ, 2010; 44:7–16.

14.

Russell

. Preventing Salmonella Colonization of Chickens Electrostatic Application of Electrolyzed Oxidative Acidic Water. University of Georgia: Bulletin, 1313. 2010.

15.

Sklar

, Joerger

. Attempts to utilize bacteriophage to combat Salmonella enterica serovar Enteritidis infection in chickens. J Food Saf, 2001; 21:15–29.

16.

Toro

, Price

, McKee

, Hoerr

, Krehling

, Perdue

. Use of bacteriophages in combination with competitive exclusion to reduce Salmonella from infected chickens. Avian Dis, 2005; 49:118–124.

17.

Van de Giessen

, Ament

AJHA

, Notermans

SHW

. Intervention strategies for Salmonella Enteritidis in poultry flocks: A basic approach. Int J Food Microbiol, 1994; 21:145–154.

18.

Van Zutphen

LFM

, Baumans

, Beynen

. Principles of Laboratory Animal Science. 2nd. Amsterdam: Elsevier, 2001.

19.

Vieira

, Silva

, Cunha

, Gomes

NCM

, Ackermann

, Almeida

. Page therapy to control multidrug-resistant P aeruginosa skin infections: In vitro and ex vivo experiments. Eur J Clin Microbiol Infect Dis, 2012; 31:3241–3249.