Emma Bell, Janice Spencer and Michael Mattey*
Fixed-Phage Ltd. Royal college Building R5.55, 204 George Street, Glasgow, G1 1XW, UK
Received Date: September 03, 2011; Accepted Date: October 11, 2011; Published Date: October 19, 2011
Citation: Bell E, Spencer J, Mattey M (2011) Better Luck Next Time:The Tesurrection of Bacteriophage Therapy. J Bioanal Biomed S6:001. doi: 10.4172/1948-593X.S6-001
Copyright: © 2011 Bell E, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Bioanalysis & Biomedicine
The potential of bacteriophages as therapeutic agents has been recognized since their discovery in 1915 and the idea of using bacteriophages to cure bacterial diseases has beenthe dominant concept until recently. The original research on bacteriophage therapy occurred at a time when bacterial diseases still killed many people, as such diseases had historically. Untreated, bacterial disease still has the potential to kill on the scale of the"Black Death" of 1347, which killed about 25 million people in Europe, at a time when thetotal population was about 60 million*. The pioneering uses of bacteriophages occurred against a background of bacterial disease unimaginable today, when a badly infected cutwas a death sentence, when most army deaths were not the result of the human enemy butthe bacterial one. Cholera, typhoid, plague and leprosy still cause deaths today, whiletuberculosis still kills about 1.6 million people worldwide, even with antibiotics available.A modern re-introduction of bacteriophage therapy would take place in competition withthe chemical armoury of antibiotics, which are largely fungal metabolites, evolved toadvance fungal competition against bacteria. The other major difference between the pre-antibiotic era and today is the regulatory oversight now in place, designed mainly to protectconsumers (patients) against unwanted side effects of chemical treatments. The two questions asked in this article are "Can bacteriophages provide a useful and effectivetreatment" and "Can bacteriophages succeed in a commercial sense"; do they work and will they be made?
One modern concern might make the introduction of bacteriophage therapy seem a "no-brainer" and that is the rise of antibiotic resistance amongst bacteria, particularly in the hospital setting. Coupled with the scarcity of new antibiotics with novel mechanisms of action, the need for an alternative treatment seems obvious. Whether the antibiotic scarcity is economic, in that the development and regulatory costs outweigh the return on investment, or more fundamental, in that few new targets for antibacterial action exist, or a combination of both, is a subject for debate outside this article, however the reality is that antibiotics are the undisputed first choice for the treatment of bacterial infections; the choice is "which antibiotics, in which order". As a proportion of the number of bacterial infections very few are truly untreatable with antibiotics, so the clinical and economic pressure for an alternative therapy is largely absent.
Bacteriophage therapy, in the human context, can be thought of as "internal" or"external" which are quite different conceptually. External application to surface wounds or infection sites or even taking bacteriophages by mouth or suppository (the alimentary canal is technically "outside") does not generally expose the bacteriophages to the immune defences of the body whereas injections for systemic infections does.
Although therapy is usually defined as "medical treatment", with the implication of human medical treatment, the term could be applied to veterinary medicine with some justification. Another factor is at play with veterinary applications and that is the regulatory pressure to limit the use of antibiotics to appropriate human medicine and to reduce residues in food animals. Both these pressures mean the "antibiotic gap" is greater in veterinary treatmentsand so the commercial hurdles to widespread bacteriophage use are lower.
The use of bacteriophages to treat bacterial diseases of plants could also be considered as"therapy", though the "medical" concept is more tenuous. However since there is no effective treatment for plant bacterial diseases the commercial scenario is akin to that in the early era of bacteriophage discovery.
Bacteriophages are bacterial viruses that invade bacterial cells and either directly or indirectly bring about cell lysis. The first hints of the existence of bacterial viruses was in 1896, when the bacteriologist Ernest Hankin  noted the antibacterial activity of water from the river Ganges against Vibrio cholerae, an activity which persisted despite filtration through fine porcelain filters, but which was removed by heating. A similar observation was made by other bacteriologist in this period.These observations were not pursued until 1915 when Fredrick Twort published a paper in the Lancet "An investigation on the nature of ultramicroscopic viruses" . Twort, a medically trained microbiologist, did not pursue the research, possibly because of financial problems and two years later Felix D'Herelle independently discovered bacteriophages in a paper presented to the French Academy of Science "Sur un microbe invisible antagoniste des bacilles dysentériques" . This work was based investigations carried out by D'Herelle in Mexico during 1910 while studying locust control. Several soldiers contracted dysentery and D'Herelle was asked to investigate the outbreak. He incubated extracts of the patients faeces, filtered to remove bacteria, with Shigella strains isolated from the faeces. Part of the mixture was used to try to develop a vaccine and part was spread on agar plates to observe the growth of the bacteria. On these plates D'Herelle observed small clear areas which he later called plaques. D'Herelle had no doubt that the plaques were caused by a virus infecting the bacteria in its locality, and he coined the name "bacteriophage". D'Herelle used bacteriophages to treat dysentery in a series of studies at the Hopital des Enfants-Malades in Paris in 1919, with Professor Victor Henri Hutinel as clinician. The bacteriophage preparation was tested for safety by D'Herelle, Hutinel and others by the simple expedient of drinking it themselves. A single dose was given to a 12 year old boy with severe dysentery, which led to a full recovery within days. These results were not reported immediately and the first published treatment with bacteriophages was made by Richard Bruynghe and Joseph Maisin  who treated Staphylococcal skin infections by surgically opening the boils and injecting bacteriophages.D'Herelle was appointed to the staff at the Pasteur Institute and became Professor of Protobiology at Yale. He established bacteriophage therapy centres in France, the USA and Georgia, and his work was fictionalised in the Sinclair Lewis novel "Arrowsmith", which won a Pulitzer Prize.
Pre antibiotic commercial products
D'Herelle and others continued the therapeutic use of bacteriophages though not always with the promising results of the early studies . In addition, a number of commercial companies began the production of bacteriophage based preparations. For example in the early 1940's Eli Lilly produced seven preparations based on bacteriophage lysates of the target bacteria and D'Herelle's commercial operation produced five preparations which were marketed by L'Oréal. Commercial bacteriophage preparations were not reliable and the efficacy of bacteriophage products was questioned. By the time antibiotics became widely available bacteriophage production which was already in decline, ceased in USA and Western Europe but continued in Eastern Europe and Russia. Two of the best known bacteriophage institutions in Eastern Europe are the Eliava Institute in Tiblisi, Georgia and the Hirszfeld Institute in Wroclaw, Poland. These and other institutions have built up a body of experience in treatments using bacteriophages . Bacteriophage therapy is used to a limited extent in Poland and published reports substantiate the efficacy of such treatments.
Although these reports are not double- blind trials and lack placebo controls in some cases,the overall evidence is positive. In Georgia the use of bacteriophages, particularly for control of dysentery, seems to have been more extensive but is largely anecdotal.
One advantage of bacteriophages as a treatment is their great variety and relative ease of isolation. Bacteriophages are the most abundant genetic entity on Earth with the total population estimated to be 1031 . This number accounts for a significant mass of organic matter. Additionally, it is not an inert mass but is responsible for the stability, death and evolution of, at very least, the microbial communities it infects and, on a greater scale, the entire ecosystem it inhabits . From the time of their discovery by D'Herelle and Twort in the 1910s , these ubiquitous, bacterial viruses have been detected in nearly all environments: soil, thermal vents and sulfur springs are just a few examples.
It is with relative ease that bacteriophages of commonly used and researched pathogenic bacteria such as E.coli and S.aureus can be isolated from seawater, soil, sewage and infection sites using simple enrichment procedures. The means of isolating phages from a sample by enrichment relies on inoculation of a host organism coupled with incubation for a period of time in order to amplify indigenous phages by their infection of that host. As a result, it is not quantitative. Alternatively direct isolation, which yields virions present at the time of isolation may be used, this method provides a quantitative measure of phage infecting a particular host and as a result the harvest may be very low .
Even more challenging isolation results when searching for bacteriophages in more extreme conditions. This is particularly true when their hosts are less abundant and more difficult to propagate in a laboratory setting; indeed there is little research on bacteriophages of anaerobic species despite their potential in bacteriophage therapy for infections of the gut .
Classification and taxonomy of bacteriophages
The first question in any regulatory process is "What is the active ingredient?" Theassumption behind the question is that, for biological entities, a Linnaean hierarchicalsystem exists that places the organism in the context of others and gives insight into its properties whilst also giving consistency in future production. Bacteriophages are loosely grouped by concepts lacking in distinct boundaries, thereby resulting in quasispecies  which is a problem for the regulatory requirements in commercial bacteriophage therapy. Consequently, this dilemma is subject to individual interpretation and creates debate amongst bacteriophage biologists. Phenotypic features have traditionally been, and perhaps still remain, the most widely used characteristics in clustering bacteriophages. Some of the first studies relied solely upon host range, however, with technological advances; more features began to be considered. In 1961, soon after the development of the electron microscope, Bradley's classification system was proposed in order to group bacteriophages into one of six morphologies .
Although this system remains to be used when describing almost all newly characterized bacteriophages it is not included within the universal taxonomic system which was created by the International Congress of Microbiology. This group established the International Committee for the Taxonomy of Viruses whose objective, since 1971, is to continually update taxonomic guidelines. The ICTV are true to the original Linnaean hierarchical system which placed the tailed bacteriophages under the order of the Caudovirales. Within this order there are three families: the Myoviridae, with long, contractile tails, the Siphoviridae with long, non-contractile tails, and finally, the Podoviridae with short, stubbed tails and a striking lack of features.
Each of these three families may be further broken down into genera by considering their host range, genome size and genome form. Using this grouping there is 1 order, 13 families and 31 genera. An example of ICTV classification is the Podoviridae which contains both the Salmonella P22 bacteriophage and the coliphage T7 [13,14]. They both fall under this order due to their short tails, however, it is known that P22 shares great genetic homology with the Siphoviriade _ bacteriophage . Indeed, their similarity is so great that recombination between them results in fully functional hybrids , yet according the ICTV they belong to different groups. There are several similar examples demonstrating the fact that taxonomic groupings according to ICTV may not reflect common ancestry and therefore this type of classification may be a flawed, although a widely recognized method of speciation. Additionally, morphology of bacteriophages is essential for classification according to this system. However, in the bacteriophage genome database, a large majority have never been examined under the microscope and consequently cannot be taxonomically grouped according to ICTV .
An obvious alternative in classing viruses would be to identify their overall similarities and group them according to these, rather than using a single core trait. The bacteriophage proteomic taxonomic tree is a system which exploits bacteriophage similarity and is genome-based. Its construction involved all 105 sequenced bacteriophages and the comparison of their predicted protein sequences. No single common protein was found in all the genomes; the closest, with 56 matches, was the YomI, a putative transglycosylase from Bacillus subtilis SPBc2 . Despite solving problems such as those previously discussed in the case of the P22, T7, classification according to ICTV, this system has been criticised. Lawrence et al.  drew up two hypothetical, nearly identical phylogenetic trees based on overall similarity. This comparison resulted in strikingly misleading information by the simple loss of a single bacteriophage from one of the trees, illustrating problems with the whole concept of overall similarity.
The inadequacy of almost all viral speciation is evident in the fact that small changes ingenetic code may result in dramatic morphological differences - for bacteriophage therapythe important possibility of lytic to lysogenic conversion is technically possible. The reverseis also true - environmental pressures may result in the maintenance of some bacteriophagemorphologies despite genetic differences. Yet another contribution to taxonomic confusion is due to genomic recombination - either homologous, or by illegitimate exchange. The results of this recombination are mixed viral populations with vast amounts of diversity.This alone reinforces a failure of the hierarchal system altogether. It appears that anysimple categorisation is flawed . The peculiar N15 bacteriophage, as already described,illustrates the difficultly in classifying viruses. N15 possesses a mixed genome: 50% is highly related to _ while the other half is related to a linear plasmid . Clearly any currently used categorisation of this bacteriophage would not adequately acknowledge this amalgamation within its genome. There are huge varieties of bacteriophages, each lacking or abundant in specific characteristics, making the whole taxonomic process severelyproblematic.
It has been proposed that one way of creating a bacteriophage taxonomical system is by exploiting the mosaicism, organization and recombination events of bacteriophage genomes and recognising their common ancestry . It does seem that the genomic taxonomy is a far more appropriate method as it encases not only common ancestry but also the phenotypic features for which these genomes encode. It has been suggested that the idea of a core set of genes for viral taxonomical purposes should be abandoned in favour of relatedness by means of a shared pool. Indeed, if bacteriophages are subject to genetic exchange amongst themselves, there are likely to be some shared aspects of their lifestyles. It appears that more recent proposals for bacteriophage taxonomy methodology recognise that a great amount of involvement is required before a virus can be clearly placed into any one group . Controversially, some even suggest that bacteriophages should belong to more than one taxonomic group at the same level . However in order to create the ideal taxonomical system for bacteriophages, it is first necessary to obtain all the information in order to complete the dataset required. One way of doing this is by studying and classifying their hosts before attempting bacteriophage taxonomy- extensive work required for "simple" classification.
However, when addressing the implications of bacteriophage taxonomy for the purpose of bacteriophage therapy, the most important grouping is that of lifestyle in terms of lysogeny. Once within the host the bacteriophage replicates to generate new bacteriophages or, in the case of temperate bacteriophages, integrates into the host genome, replicating its DNA as the host replicates. In this state the bacteriophage is termed a "prophage". A key issue for bacteriophage therapy is if the bacteriophage for use as therapy is capable of turning its host into a prophage. However this mechanism is only made possible by the possession of an integrase gene within the bacteriophage genome and as many bacteriophages (virulent bacteriophages) completely lack this, not all are capable of horizontal gene transfer between bacteria. As a result, with the exclusive use of these bacteriophages for bacteriophage therapy there is no real threat of passing virulence onto bacteria. However, the possibility that DNA from a "dead" lytic bacteriophage breakdown (free DNA produced) could be incorporated into a prophage sequence by transformation exists in theory, but the probability of an entire bacteriophage genome being available is very low, the probability of transformation about 10-8, and the probability of integration in the appropriate position to regain the lysogenic genes with the concomitant expulsion of prophage genes is again very low. Such an event is very unlikely. Therefore, the exclusively lytic nature of a particular bacteriophage would be demonstrated by the absence of integrase from its geneticsequence.
The now common practice of bacteriophage genome sequencing has allowed their uses to be applied not only as a therapeutic agent but also as a molecular cloning tool, display vector and use as a sensor of potentially pathogenic bacteria with the aid of a reporter gene.The structure of bacteriophage genomes makes their study and manipulation relatively simple. When compared to other organisms, the most striking feature of bacteriophage genomes is the mosaic structure they exhibit in relation to one another. The first indication of this was by Simon et al in 1971 with the use of electron microscopy, which revealed DNA-DNA heteroduplexes in E.coli bacteriophages . With the advent of molecular genetics and bacteriophage sequencing this mosaicism was observed with genes found in blocks. When two bacteriophage sequences are aligned, areas of sequence similarity are clustered, followed by distinct transitions to regions of little or no sequence similarities. An illustration of this is found within the major capsid genes of bacteriophages HK97 and HK022, which share 99% similarity nucleotide sequence. However, the nucleotide sequences for their DNA replication proteins share only 33% similarity . Due to their mosaic structure with neatly ordered clusters of genes according to functionality their lifestyle and relationships are readily identified.
Concerning bacteriophage therapy, one of the most useful aspects of bacteriophage molecular biology is with the rapid detection of lytic and lysogenic genes by applying PCR.This technique is rapid and gives a more definitive answer concerning the lifestyle of the bacteriophage. Indeed, it addresses one of the most obvious problems concerning the regulatory aspect of bacteriophage treatments.
Reporter bacteriophages have been developed to aid the rapid detection and enumeration of bacteria. Such systems have been developed in just a few organisms such as E.coli, M.tuberculosis and L.monocytogenes [24-26]. All of these systems have exploited the light producing properties of luciferase. The lux operon consists of seven genes, two of which are regulators, three are part involved with substrate synthesis and the remaining two; luxA and luxB, encode the subunits of luciferase . The system of detection of L.monocytogenes utilised only luxAB in a myovirus. In vivo homologous recombination between the genes and bacteriophage was used to insert the two genes into the wild type bacteriophage genome. These genes were carefully placed downstream of the major capsid protein without disrupting any of the other bacteriophage genes. It was found that as little as 500 bacterial cells could be detected from pure culture within two hours; however, a preenrichment stage was required for lower levels . Using bacteriophages in this way enables inexpensive, rapid detection systems for, potentially, all micro-organisms.
Mode of action
It is the specificity of bacteriophages that is part of the appeal of bacteriophage therapy; indeed it is this that often defines their application and value for therapeutic uses. Bacteriophages are structurally the simplest genetic entity; however, they act as efficient and impressive machines. Their specificity is attributed to their receptor binding protein(RBP). This protein sits on the bacteriophage base plate, which is at the tip of the tail. This is very specific to the individual bacteriophages' host range, and is the initial step in the infection process . Some base plates are known to rotate for efficient locking onto the host receptor; bacteriophage T4 base plate is capable of such maneuvers [29,30]. This triggers the injection of bacteriophage DNA into the host via the tail tube  and the lifecycle continues.
Large-scale production of bacteriophages
For efficient commercial purposes the development of large scale bacteriophage production is vital. Despite this there has been a distinct lack of recent studies examining this aspect of bacteriophage propagation. Even those which have been conducted in the past have studied a maximum of 200L , although the majority examines 5L batches . Overall these studies do illustrate the importance of media composition, multiplicity of infection (MOI), temperature and pH conditions for large-scale production. However, for commercially sound production cost effectiveness a crucial factor. As a result, media composites should be kept to a minimum whilst remaining capable of high bacteriophage titre production.Indeed the ultimate successful production would be in the form of continuous bacteriophage culture in large scale. This has been achieved by Bujanover  who reported of bacteriophage production for large-scale commercial production on semisolid medium. However, a liquid form would prove even more effective.
Can bacteriophages provide a useful and effective treatment?
The use of bacteriophage therapy falls into two categories, as do bacterial infections;external, usually localised applications and internal or systemic use where whole body infection occurs.
Wounds have been the subject of several studies. Wounds are categorised into open andclosed; open, usually subdivided according to the cause of the wound, for exampleincisions, lacerations, abrasions or puncture wounds and closed, typically contusions,hematomas, or crush injury. Chronic and Acute wounds are the result of injuries that disrupt the tissue. Chronic wounds include pressure, venous, and diabetic ulcers result when infection takes hold at the site of injury and becomes a chronic abscess. Acute usually result from infection occurring during the injury. Once the infection hits a critical point, it can spread locally or become systemic.
Markoishvili et al.  showed that bacteriophages soaked into a biodegradable film healed ulcers in 70% of 96 patients. These were recalcitrant ulcers that had not been healed by conventional treatment. Microbiological assessment in 22 of the patients found healing was associated with the elimination of the pathogen from the wound. The dressing contained ciprofloxacin and a commercial bacteriophage preparation ('PyoPhage') active against strains of Psuedomonas aeruginosa, Escherichia. coli, Staphylococcus aureus, Streptococcus and Proteus. The healing of these wounds could not be solely attributed to the bacteriophages.
Earlier Soothill  had reported that skin grafts infected by P. aeruginosa consistently failed in a guinea pig model. When these infected grafts were treated with 106 pfu of specific bacteriophage, 6 out of 7 grafts were successful. Soothill and colleagues also demonstrated efficacy with a bacteriophage against S. aureus in a rabbit abscess model . The sewerage derived bacteriophage reduced the abscess area and the count of S. aureus in the abscess in a dose dependant manner.
Marza et al.  reported a case of a 27 year old man with a 50% total body surface burns. The area had been excised and covered with skin grafts but these had become infected with P. aeruginosa after several months. These grafts had repeatedly broken down despite antibiotic treatment before an application of purified bacteriophage was tried. Three days after the application of the bacteriophage preparation P. aeruginosa could not be isolated from the wound and subsequent the grafting was successful.
Many bacteriophage studies have targeted gut infections; as with wounds it is relatively straightforward to introduce bacteriophages to the site of infection, either by irrigation or by ingestion. Smith et al.  reported that bacteriophages could cure severe diarrhoea caused by an introducing as enteropathogenic E. coli (EPEC) strain in calves. The single dose of bacteriophage was given at the same as the bacteria. Other cows that were feeding on litter contaminated by these bacteriophage infected cows were protected against the diarrhoea as well. This study showed the importance of matching the bacteriophages to the EPEC strain inoculated, as there was a bacteriophage -resistant EPEC strain also present in the cows which remained as virulent as the original strain inoculated.
A recent study by Denou et al.  has shown that oral treatment with a bacteriophage cocktail had no negative impact on the murine gut microbial flora after one month exposure. Bacteriophages were found in high titres in the caecum and colon and lower titresin the small intestine, but were not detected in the blood, liver or spleen. Serum antibodies towards the bacteriophage were not detected. This study shows that bacteriophage have no adverse effect on the microbial flora and would only target the pathogenic bacteria intended. The lack of serum antibodies detected also highlights that repeated dosing could be used without any negative effect.
Other studies on external sites have examined ear infections. Marza et al.  reported treatment of a dog with chronic bilateral otitis external caused by a Pseudomonas. Aeruginosa infection that had failed to respond to repeated courses of topical and systemic antibiotics. After inoculation of 400pfu of bacteriophage into the auditory canal there was marked improvement in the clinical signs, 27 hr after treatment.
Recently (2009) a controlled human trial was carried out by a private company based in the UK. Biocontrol Ltd performed a clinical trial including 24 patients treated for chronicotitis . This was a randomised, double blind, placebo controlled trial approved by UK Medicines and Healthcare products Regulatory Agency (MHRA) the first of its kindreported.
All patients had chronic otitis caused by antibiotic resistant P. aeurginosa strains which had proven untreatable by usual methods. Patients were randomised into two groups and treated with either a single dose of a bacteriophage cocktail or placebo and the symptoms and microbiological indicators were monitored for 42 days after administration.
Results showed that 3 patients treated with bacteriophages were symptom and bacteria free.All other patients receiving the bacteriophage therapy showed significantly lower bacterial counts and improvement of symptoms compared with the patients receiving the placebo control. This trial concluded that there were no adverse effects in patients receiving bacteriophage therapy. These results support the need and opportunity for bacteriophage therapy to become a routine treatment for difficult to treat infections.
The treatment of systemic infections is the most challenging environment for bacteriophage action, with issues of compartmentation, host defence and kinetics, which are not present to the same degree in topical applications.
There are three aspects to consider when evaluating bacteriophage therapy; the comparison with existing antibiotic treatments, the efficacy of bacteriophage treatment and the cost to benefit of the treatment.
Trials of bacteriophage therapy in humans have been limited to studies in Eastern Europeand Russia. Reviews of the Soviet literature [42-44] indicate a high level of success for various forms of bacteriophage therapy but these have mainly involved oral bacteriophage delivery for dysentery or local treatment of wounds. Slopek et al.  reported bacteriophage treatment of septicaemia with a 90% success rate but such trials have not been sufficiently rigorous to be of regulatory usefulness. What is clear from such work is that bacteriophage therapy has potential in the treatment of acute or chronic infections but well designed trials are needed to validate this claim.
Animal trials of injected bacteriophages have been carried out in a number of species with encouraging results. Intramuscular injections of lytic bacteriophages were used to treat experimental E. coli septicaemia in chickens and calves . Chickens given equal doses of bacteriophages and bacteria showed no morbidity and significant protection was shown by doses with a 1% ratio of bacteriophage to bacteria probably indicating bacteriophage multiplication in vivo. With calves, intramuscular injection of bacteriophages delayed the onset of an experimental infection of E coli infections and increased survival times.
Mouse models of bacterial infection have been successfully demonstrated; a significant study on VRE Enterococcus faecium infection  demonstrated that intraperitoneal administration of bacteriophages against VRE- E faecium prevented death.
When given 45 minutes after infection the bacteriophages prevented bacteraemia with a reduction in circulating bacteria. Fully effective treatment was observed up to five hours after infection, but was less effective beyond this time. The kinetics of bacteriophages in model situations was examined by Payne et al.  who showed that the outcome of treatment with a self replicating agent such as bacteriophages will depend critically on the initial numbers and replication rates. Descriptions of their interactions follow a predator- prey relationship rather than classical pharmacokinetics.
Another mouse model was used by Smith and Huggins (49) to show that a bacteriophage lytic for E. coli strain O18:K1:H7, which causes generalised infections in both man and animals, was more effective than conventional antibiotics.
The bacteria were injected into the brain of groups of 32 mice. At the same time the micewere either injected with a single dose of anti-K1 bacteriophage intramuscularly or bymultiple intramuscular doses of antibiotics (streptomycin, tetracycline, ampicillin,chloramphenicol or trimethoprimsulfafurazole). In controls (no bacteriophage or antibiotics) 31 out of 32 mice died; in the bacteriophage treated group 13 mice died while inthe antibiotic treated group 18 mice dying in the streptomycin treatment group and 26-28dying in the other antibiotic treated groups. Significantly, a bacteriophage resistant strain of E coli was isolated from the bacteriophage treated group. This strain did not produce any K1 capsule, as shown by the failure to precipitate with anti-K1 antiserum. Previous studies have shown than K1 mutants had greatly reduced virulence when injected into mice or chickens due to their inability to invade tissues .
Soothill  reported the protective ability of bacteriophages lytic against strains of Acinetobacter baumanii, P. aeruginosa and S. aureus in experimental infections of mice to investigate their potential for the treatment of infections of man. As few as 102 particles of an acinetobacter bacteriophage protected mice against 5 LD50 (1 x 108 cfu) of a virulent strain of A. baumanii, and bacteriophage was demonstrated to have multiplied in the mice.
A pseudomonas bacteriophage protected mice against 5 LD50 of a virulent strain of P. aeruginosa, with a PD50 of 1.2 x 107 particles. A staphylococcal bacteriophage failed toprotect mice infected with a strain of S. aureus. These studies support the view that bacteriophages could be useful in the treatment of human infections caused by antibioticresistant strains of bacteria.
Morello et al.  used a mouse lung infection model caused by a multidrug resistant P.Aeruginosa mucoid strain isolated from a cystic fibrosis patient to evaluate the efficacy of bacteriophage treatment. One single dose administered 2h after the onset of the infection allowed over 95% survival. A four day preventative treatment (one dose) resulted in 100% survival. This work provides an incentive to develop clinical studies on pulmonary bacteriophage therapy to combat multidrugresistant lung infections.
Colonisation of sheep by Escherichia coli O157:H7 could be reduced using a bacteriophage cocktail . The cocktail consisted of two bacteriophages isolated from ruminants showing no colonisation with E coli O157:H7. The animals which received the cocktail had a more effective reduction in E coli O157:H7 levels than animals receiving a single bacteriophage or the untreated bacteriophage-free controls.
Can bacteriophages succeed in a commercial sense?
As can be seen from the preceding section there is good evidence that bacteriophage therapy can work but several obstacles will have to be overcome before it becomes an accepted and available therapy. These can be considered under two headings, clinical obstacles and economic ones.
Clinically it is tempting to regard bacteriophages as alternatives to antibiotics especially with the apparent rise in antibiotic resistant bacterial infections. The host specificity of bacteriophages however places them in a very different category to antibiotics. While antibiotics have a relatively wide spectrum of antibacterial activity bacteriophages are relatively specific. The best "wide spectrum" bacteriophages will only kill about 70% of a given bacterial species, and to mimic the average antibiotic in specificity will require a cocktail of perhaps as many as 20 to 30 bacteriophages. While such a cocktail is perfectlypossible the regulation and manufacture of such a product would be challenging, especially if an injectable formulation was required. Each additional bacteriophage would add regulatory and manufacturing costs to the final product. The host specificity of bacteriophages is both a strength and weakness, the strength is that an infection could be targeted with some precision without damaging the rest of the bacterial flora in for example the gut or on skin, the weakness is that such an approach requires individual determinations of the bacterial infection before treatment. While personalised medicine is a much discussed and desirable development, the reality of how this could be afforded for everyone is not so clear. Frequently bacterial infection present as acute events, where the speed of response is important to the outcome. To use bacteriophages in such a situation will be difficult. This may restrict bacteriophages to treatment of chronic conditions where numbers of patients are lower and time is available for accurate diagnosis. This leads however into the economic problem.
The current attempts to introduce bacteriophages into medical practice take place against adifferent background to that in the historic situation when bacteriophages were first introduced. We now have a dominant product in the market, antibiotics. Despite the rise of resistance antibiotics are generally still effective and they are not expensive. To displace antibiotics several factors need to be in place; the new product needs to be better, cheaper or easier to use. The market in which bacteriophages are "better" than antibiotics is in antibiotic resistant infections, which is still a small proportion of the total doses of antibiotics sold which would make bacteriophage only products expensive, possibly prohibitively so. In other situations they may be "as good as" antibiotics in the primary aim of treating an infection. The specificity of bacteriophages is only important in the recovery from infection where the side effects of treatment are seen.
So are they cheaper? The cost of bacteriophages is an area of much debate. The main cost of both antibiotics and bacteriophages is in the recovery of costs associated with the regulation of the use and manufacture of the products; development costs are significant as is the actual manufacture, but these are not dominant factors. Unless a major and unlikely shift in regulatory trends occurs, the costs of bacteriophages will be similar to that of cheaper antibiotics; there is not a lot of difference between a fermentation antibiotic like penicillin and a bacteriophage fermentation in cost or complexity of process and given that a cocktail of bacteriophages may be involved, each with different media for their respectivehost bacteria, the costs for bacteriophage products may be higher. The regulatory consequences of pathogenic host bacterial production on a large scale have not really been considered but are unlikely to be cheap.
The other displacement advantage might be ease of use, but in practice this will be a level playing field, the administration methods, the distribution chains etc will be similar for bothproduct types. In short, bacteriophages are significantly better only for the minority of infections where antibiotic resistance is a factor, they are not likely to be cheaper or easierto use. It is unlikely therefore that they will displace antibiotics in the short or medium term.
To displace antibiotics in the long term, say 20 years, would require significant new developments in bacteriophage technology. This may result from trends in personalized medicine or genetic engineering of bacteriophages or as yet unsuspected rise in resistance to antibiotics. It may also reflect the ease of developing new bacteriophages compared with the difficulty of finding new antibiotics. The possibility of coformulating bacteriophages with antibiotics is a more likely scenario than simple bacteriophage antibacterials. Such products could combine the wide spectrum benefits of antibiotics with the specific ability to kill antibiotic resistant strains.
Other applications of bacteriophages are likely to come about before therapeutic ones. The use of bacteriophages in decontamination, food safety, agriculture and horticulture and other areas where a "green" biocide would be advantageous is easier and economically safer than the highly regulated and controlled therapeutic market. These products will however create a climate where bacteriophages become familiar and increase the possibilityof therapeutic applications.
* In the absence of census data the actual numbers are the subject of debate but most authorities agree that 30 to 60% of the population perished.
Bacteriophage therapy, which was a major treatment in the first half of the twentieth century, was largely displaced by antibiotics, which proved more effective. With increased perception of antibiotic resistance amongst bacteria as a significant problem and regulatory pressures for product with minimal environmental consequences, bacteriophages have again become fashionable. Evidence is reviewed that suggests that bacteriophage therapy,when properly understood and applied, can be effective, though formal evidence is limited. However, consideration of the economic case suggests that, at best, bacteriophages will not displace antibiotics are the therapy of choice in the short or medium terms.
All three authors are employed by Fixed-Phage Ltd. a company which specializes in bacteriophage immobilization.