Pilar García, Instituto de Productos Lácteos de Asturias (IPLA-CSIC), Apdo 85, 33300-Villaviciosa, Asturias, Spain. E-mail: email@example.com
In recent years it has become widely recognized that bacteriophages have several potential applications in the food industry. They have been proposed as alternatives to antibiotics in animal health, as biopreservatives in food and as tools for detecting pathogenic bacteria throughout the food chain. Bacteriophages are viruses that only infect and lyse bacterial cells. Consequently, they display two unique features relevant in and suitable for food safety. Namely, their safe use as they are harmless to mammalian cells and their high host specificity that allows proper starter performance in fermented products and keeps the natural microbiota undisturbed. However, the recent approval of bacteriophages as food additives has opened the discussion about ‘edible viruses’. In this review, we examine the promising uses of phages for the control of foodborne pathogens and the drawbacks on which more research is needed to further exploit these biological entities.
The current technologies employed to inactivate bacterial pathogens in foods are not infallible, as proved by the continuous increase in several foodborne diseases caused by pathogens, such as Salmonella, Campylobacter, Escherichia coli, Listeria and others that have an enormous impact on public health (DuPont 2007). Contaminating bacteria can get access to food during slaughtering, milking, fermentation, processing, storage or packaging. Over the last few years, a number of strategies to minimize the microbial load of raw products have been explored as the use of antibiotics is restricted due to the negative impact on human antimicrobial therapies. Problems of acceptability and deterioration of the organoleptic properties have been described after physical treatments such as steam, dry heat and UV light. Moreover, the extensive use of sanitizers has led to the development of resistant bacteria rendering these procedures less effective. On the other hand, some approaches often used in the food industry to reduce contamination by foodborne pathogens cannot be directly applied to fresh fruits, vegetables and ready-to-eat products. Hence, despite recent advances to avoid transmission of bacterial pathogens throughout the food chain, novel strategies are still required to fulfil consumer demands for minimally processed foods with fewer chemical preservatives.
Recently, research on phage molecular biology has fuelled multiple biotechnological applications in very diverse fields including nanotechnology, vaccine development, therapeutic delivery, bacterial detection systems, novel antimicrobials against antibiotic-resistant bacteria, etc. Another promising field of application is the use of phages as natural antimicrobials in food to inhibit undesirable bacteria, which is likely to be acceptable to consumers (Hagens and Loessner 2007; Strauch et al. 2007). This review aims to provide an insight into how phages can be exploited to generate antibacterial agents to enhance microbial food safety.
Antimicrobial activity of phages: past and present
Bacteriophages are viruses that invade bacterial cells and, in the case of lytic phages, disrupt bacterial metabolism and cause the bacterium to lyse. They were discovered by Ernest Hankin (1896) and Frederick Twort (1915) who described their antibacterial activity. However, Felix d’Herelle (1919) was probably the first scientist who used bacteriophages as a therapy to treat severe dysentery. At that time, several companies then actively started up the commercial production of phages against various bacterial pathogens for human use. However, phage production was quickly displaced by the discovery of antibiotics in most of the Western world. Nevertheless, phage therapy is still an on-going practice in Eastern Europe and countries from the former Soviet Union. Several institutions in these countries have been involved in phage therapy research and production, with activities centralized at the Eliava Institute of Bacteriophage, Microbiology and Virology (Tbilisi, Georgia, http://www.evergreen.edu/phage/home.htm) and the Hirszfeld Institute of Immunology and Experimental Therapy (Wroclaw, Poland). Their work in this field has recently been extensively reviewed (Kutter and Sulakvelidze 2005).
The current threat of antibiotic-resistant bacteria has renewed the interest in exploring bacteriophages as biocontrol agents in Western countries (Matsuzaki et al. 2005; Sulakvelidze and Kutter 2005). In fact, some products based on bacteriophages are already commercially available (‘PhageBioderm’, ‘Bacteriophagum Intestinalis Liquidum’, ‘Pyobacteriophagum Liquidum’). Additionally, some care centres are particularly specialized in phage therapy (for example, Southwest Regional Wound Care Centre, Texas, http://www.woundcarecenter.net).
Besides phage therapy, the use of bacteriophages as antimicrobial agents and tools for detecting pathogens in feed and foodstuffs is also expanding with several companies having been created recently (Table 1). Fields of application comprise of water and food safety, agriculture and animal health. An example is OmniLytics, Inc. that gained US Environmental Protection Agency approval for the use of its product AgriPhage against plant pathogenic bacteria. In food manufacturing industry, EBI Food Safety recently marketed Listex™ P100 for controlling Listeria in meat and cheese products (Carlton et al. 2005). In August 2006, the US Food and Drug Administration (FDA) approved the use of a phage preparation targeting Listeria, LMP 102 (Intralytix, Inc.), in ready-to-eat meat and poultry products.
Table 1. Companies involved in bacteriophage application
Bacteriophages for biocontrol of pathogens in food
Bacteriophage-based biocontrol measurements have a great potential to enhance microbiological safety based, namely, on their long history of safe use, relatively easy handling and their high and specific antimicrobial activity (Table 2).
Table 2. Main advantages of bacteriophages as biocontrol tools in food safety
History of safe use
Ubiquitous in nature including food ecosystems
Natural commensals of humans and animals
Extensive clinical use in Eastern Europe
Highly active and specific
No adverse effects on the intestinal microbiota
Innocuous to mammalian cells
Can be active against biofilms
Versatile use along the food chain
Phage therapy, biosanitation, biopreservation
Tools for detecting pathogens
Source of potent antimicrobials
Endolysins and other peptidoglycan hydrolases
The concept of combating pathogens in food by means of phages can be addressed at all stages of production in the classic ‘farm to fork’ approach throughout the entire food chain (Fig. 1). Bacteriophages are suitable (i) to prevent or reduce colonization and diseases in livestock (phage therapy), (ii) to decontaminate carcasses and other raw products, such as fresh fruit and vegetables, and to disinfect equipment and contact surfaces (phage biosanitation and biocontrol) and (iii) to extend the shelf life of perishable manufactured foods as natural preservatives (biopreservation). Bacteriophages should also be considered in hurdle technology in combination with different preservation methods (Leverentz et al. 2003; Martínez et al. 2008). Several examples of phage application throughout the food chain are outlined below and summarized in table 3.
Table 3. Bacteriophages application in food safety and reported effects
Escherichia coli O157:H7
Intestinal reduction by oral administration (sheep)
Bacteriophage to control E. coli O157:H7 contamination
The emergence of pathogens, such as E. coli O157:H7, remains a continuous public health threat because its ingestion at concentrations as low as 10–100 cells may result in potent toxin exposure. Ruminants comprise the principal reservoir for this strain and contamination of animal products occurs during milking or slaughtering. Phage treatment aims to reduce pathogen contamination prior to slaughtering (Table 3). After the oral administration of phage CEV1, a 2-log-unit reduction in intestinal E. coli O157:H7 was achieved within 2 days in sheep (Raya et al. 2006). However, orally administered phage KH1 was not effective. When a combination of phages KHI and SH1 were administered rectally to cattle and phages were also maintained at 106 PFU ml−1 in the drinking water, significantly lower cell numbers were observed (Sheng et al. 2006).
Phages were also applied to the meat surface to avoid pathogen development. A mixture of three different phages was applied to beef contaminated with 103 CFU g−1E. coli O157:H7. In most of the samples, no viable cells could be recovered after storage at 37°C (O’Flynn et al. 2004).
Bacteriophage to control Salmonella contamination
Salmonella can be isolated from numerous animal species and is the major cause of food poisoning. Phage cocktails reduced the Salmonella enteritidis average faecal counts by 0·3–1·3 log units, but the pathogen was not completely eradicated even when more than 107 PFU of phage were present per gram of cecal content (Sklar and Joerger 2001). More recently, artificially infected broilers treated orally with high numbers of bacteriophages (1011 PFU) (Fiorentin et al. 2005). Although the bacteria were not eradicated from the birds, both studies showed that phage treatment would decrease levels of the pathogen bacteria entering the poultry production line.
The activity of the Salmonella phage SJ2 was tested in cheddar cheese manufacturing (Modi et al. 2001). In the presence of phages (MOI 104), Salmonella did not survive in the pasteurized cheeses after 89 days, whereas about 50 CFU ml−1 were still viable in raw milk cheeses.
Salmonella phage cocktails have been also assayed on fruits. Phage numbers remained relatively stable on melon and gave a significant reduction of target bacteria. On the contrary, a quick decline of infective phage particles was observed on apples due to the lower pH of this fruit (Leverentz et al. 2001). In another study, Salmonella phage Felix-O1 was tested in biocontrol experiments with Salmonella typhimurium on chicken frankfurters contaminated with 300 CFU. Bacterial count reductions of 1·8 and 2·1 log units were achieved (Whichard et al. 2003). Phages infecting Salm. typhimurium PT160 and Campylobacter jejuni were added at a low or high (10 or 104) MOI to either low or high (<100 or 104 cm−2) densities of host bacteria inoculated onto raw and cooked beef. Significant host inactivation of the order of 2–3 log units at 5°C and >5·9 log units at 24°C was achieved (Bigwood et al. 2008).
Bacteriophage to control Campylobacter contamination
The colonization of broiler chickens by the enteric pathogen Camp. jejuni is widespread and difficult to prevent. Oral infection with this pathogen has become the most common cause of foodborne disease in industrialized countries. The presence of phages negatively correlated with the levels of Campylobacter in cecal contents indicating that some degree of biocontrol may occur (Atterbury et al. 2005). Bacteriophages have been effective to decrease Campylobacter counts recovered from cecal contents, thereby lowering the risk of cross contamination during slaughtering (Loc Carrillo et al. 2005; Wagenaar et al. 2005). Significant decreases have been reported, although total clearance of the host could not be achieved by phage treatments.
Other authors have shown the efficacy of bacteriophage F2 or several lytic bacteriophages applied on artificially contaminated chicken skin (Atterbury et al. 2003). The antibacterial phage activity was detected at 4°C and −20°C using a very high MOI (MOI 105). At room temperature, lower MOIs of around 100 were already effective and more than 95% of the target cells were killed (Goode et al. 2003).
Bacteriophage to control Listeria monocytogenes contamination
Foodborne infection due to L. monocytogenes is often associated with either fresh or minimally processed foods, such as dairy products or salads, or with processed foods which are stored at low temperatures. Listeriosis usually has a low incidence. However, due to its high mortality rate, up to 30%, Listeria is considered as a relevant pathogen. Leverentz et al. (2003) evaluated a phage cocktail to control Listeria on fresh, cut apples and melon in combination with the bacteriocin nisin. Phage population rapidly declined on sliced apple likely due to the low pH. However, eradication of the pathogen was achieved in both fruits. Besides, the phages were shown to work synergistically with the bacteriocin. Similarly, a synergistic effect was seen in broth cultures but a phage-nisin mixture was not effective on ground beef stored at 4°C (Dykes and Moorhead 2002). Hence, phage activity should be optimized for each food type on a case-by-case basis.
Recently, a commercial product named Listex P100 was approved by the FDA as a food biopreservative and granted as GRAS (Generally Recognised As Safe) (Federal Register: August 18, 2006. Volume 71, Number 160, pp. 47729–47732). This product is based on the virulent phage P100 (Carlton et al. 2005) and, depending on dosage and treatment, a complete eradication of target cells was achieved. A bacteriophage-based preparation (LMP 102™) of six bacteriophages isolated from the environment has also been developed as an additive for ready-to-eat foods (Lang 2006).
Bacteriophage to control Enterobacter sakazakii contamination
Enterobacter sakazakii is an uncommon, but often fatal, invasive pathogen that causes bloodstream and central nervous system infections. A recent study has addressed the issue of Ent. sakazakii in reconstituted infant formula milk (Kim et al. 2007a). Newly isolated phages against this pathogen were able to effectively suppress growth in prepared infant formula, both at 24 and 37°C. The highest phage concentration tested (109 PFU ml−1) was the most effective and able to completely eradicate the target organism (Kim et al. 2007b).
Bacteriophage to control Staphylococcus aureus contamination
Staphylococcus aureus often causes food poisoning and is also a major etiologic agent in opportunistic and nosocomial infections. Staphylococcal food poisoning is due to the absorption of staphylococcal enterotoxins preformed in the food matrices (Le Loir et al. 2003). Mastitis caused by Staph. aureus is a major concern to the dairy industry and the most important source of milk contamination by this pathogen. The ability of the lytic Staph. aureus bacteriophage K to eliminate bovine Staph. aureus intramammary infection during lactation was evaluated, but there was significant degradation or inactivation of the infused phage within the gland (Gill et al. 2006a). Phage K inactivation was also reported in raw milk, likely due to the adsorption of whey proteins to the cell surface that interfere with phage attachment (O’Flaherty et al. 2005; Gill et al. 2006b). However, a cocktail of two lytic phages of dairy origin added at MOI 102 successfully inhibited Staph. aureus in acid and enzymatic curd manufacturing processes (García et al. 2007).
Conclusions and future prospects
The use of biological control measurements such as bacteriophage biocontrol seems to be a promising alternative for the management of food contamination as the use of chemical preservatives becomes restricted. Future developments involve safety and technological issues as well as expanding the antimicrobial skills of bacteriophages (Table 4).
Table 4. Future challenges of phage-based biopreservation
Lack of undesirable traits
Better knowledge of gene flow phage-host
Blocking gene dissemination systems
Large-scale safer production systems
Use of nonvirulent hosts
Enhance activity in food systems
Modelling phage behaviour
Same environment as phage source
Better knowledge of phage-host physiology
Expanding host range
Use of phage mixtures
Engineering tail fibre genes
New phage-derived antimicrobials
Inhibitors of host metabolism
Bacteriophages may act as vectors of undesirable traits (virulence and antibiotic-resistance genes) and temperate phages mediate lysogenic conversion that have raised safety concerns. Recent advance in genomics and phylogenic studies make it possible to understand gene flow among phages and hosts and potentially harmful bacteriophages could be avoided or re-designed without undesirable traits and lacking any gene dissemination systems. As an example, bacteriophages could be genetically engineered to block phage replication once the host has been killed (Hagens et al. 2004). This would prevent the release of large numbers of phages in a particular environment.
Technological challenges to strengthen the future use of bacteriophages as biopreservatives will require the establishment of safe large scale production processes. It is advisable to develop nonvirulent, genetically well-characterized bacteria as hosts in phage propagation. On the other hand, the antimicrobial activity of phages observed in laboratory conditions could be greatly reduced in food systems. Limiting factors are reduced diffusion rates that decrease the chance of host-phage collisions, the microbial load which might also act as a mechanical barrier by providing unspecific phage binding sites and other adverse factors such as temperature, pH and inhibitory compounds. As is the case of any food biopreservatives, bacteriophage efficacy in food should be evaluated on a case-by-case basis. To guarantee proper phage performance, it is wise to isolate phages from the same niche/habitat, as those phages will probably be better adapted to replicate and survive in those conditions. Bacterial fitness and physiological status has a clear influence on the phage infection rate. Phages that selectively bind to the host and replicate in different physiological states should be chosen.
There are several strategies to improve and expand the antimicrobial activity of phages. Comparative genomics of phage tail fibre genes involved in the recognition of specific host receptors will lead to approaches to expand the host range. A detailed analysis of the bacterial receptors will also help to understand and predict the development of bacteria insensitive mutants often due to the loss or mutation of these molecules. Both approaches will enhance the use of phages as antimicrobials as well as bacterial detection systems (Rees and Dodd 2006). New phage-derived antibacterial strategies based on phage-encoded proteins that commit cellular metabolism to phage proliferation, such as endolysins, are currently being evaluated (Liu et al. 2004; Loessner 2005).
Despite the fact that current research is basically at the experimental stage, the increasing number of publications and emerging companies in this area are paving the way for the future of phage-based biopreservation technologies. Landmarks for the next step towards the establishment of expanded commercial applications will be influenced by the accurate knowledge of bacteriophage biology to allow consumers to feel confident about the safety of ‘edible viruses’.