Formulation Stabilisation and Encapsulation of Bacteriophage For Phage Therapy

Advances in Colloid and Interface Science 249 (2017) 100–133
Contents lists available at ScienceDirect
Advances in Colloid and Interface Science

journal homepage: www.elsevier.com/locate/cis
Historical perspective
Formulation, stabilisation and encapsulation of bacteriophage for phage T

therapy
Danish J. Malika,⁎, Ilya J. Sokolova, Gurinder K. Vinnera, Francesco Mancusoa,
Salvatore Cinquerruia, Goran T. Vladisavljevica, Martha R.J. Clokieb, Natalie J. Gartonb,
Andrew G.F. Stapleya, Anna Kirpichnikovac
a
Chemical Engineering Department, Loughborough University, LE11 3TU, UK
b
Department of Infection, Immunity and Inflammation, University of Leicester, University Road, Leicester LE1 7RH, UK
c
Department of Mathematics and Computer Science, Liverpool Hope University, Hope Park, Liverpool L16 9JD, UK
A R T I C L E I N F O A B S T R A C T
Keywords: Against a backdrop of global antibiotic resistance and increasing awareness of the importance of the human
Antibiotic resistance microbiota, there has been resurgent interest in the potential use of bacteriophages for therapeutic purposes,
Bacteriophage known as phage therapy. A number of phage therapy phase I and II clinical trials have concluded, and shown
Encapsulation phages don't present significant adverse safety concerns. These clinical trials used simple phage suspensions
Phage therapy
without any formulation and phage stability was of secondary concern. Phages have a limited stability in so-
Pharmacodynamics
lution, and undergo a significant drop in phage titre during processing and storage which is unacceptable if
phages are to become regulated pharmaceuticals, where stable dosage and well defined pharmacokinetics and
pharmacodynamics are de rigueur. Animal studies have shown that the efficacy of phage therapy outcomes
depend on the phage concentration (i.e. the dose) delivered at the site of infection, and their ability to target and
kill bacteria, arresting bacterial growth and clearing the infection. In addition, in vitro and animal studies have
shown the importance of using phage cocktails rather than single phage preparations to achieve better therapy
outcomes. The in vivo reduction of phage concentration due to interactions with host antibodies or other
clearance mechanisms may necessitate repeated dosing of phages, or sustained release approaches. Modelling of
phage-bacterium population dynamics reinforces these points. Surprisingly little attention has been devoted to
the effect of formulation on phage therapy outcomes, given the need for phage cocktails, where each phage
within a cocktail may require significantly different formulation to retain a high enough infective dose.
This review firstly looks at the clinical needs and challenges (informed through a review of key animal studies
evaluating phage therapy) associated with treatment of acute and chronic infections and the drivers for phage
encapsulation. An important driver for formulation and encapsulation is shelf life and storage of phage to ensure
reproducible dosages. Other drivers include formulation of phage for encapsulation in micro- and nanoparticles
for effective delivery, encapsulation in stimuli responsive systems for triggered controlled or sustained release at
the targeted site of infection. Encapsulation of phage (e.g. in liposomes) may also be used to increase the cir-
culation time of phage for treating systemic infections, for prophylactic treatment or to treat intracellular in-
fections. We then proceed to document approaches used in the published literature on the formulation and
stabilisation of phage for storage and encapsulation of bacteriophage in micro- and nanostructured materials
using freeze drying (lyophilization), spray drying, in emulsions e.g. ointments, polymeric microparticles, na-
noparticles and liposomes. As phage therapy moves forward towards Phase III clinical trials, the review con-
cludes by looking at promising new approaches for micro- and nanoencapsulation of phages and how these may
address gaps in the field.
1. Introduction infections may be regarded as a significant achievement of modern

medicine. Surgery, transplantation and chemotherapy rely heavily on
The discovery of antibiotics and the subsequent control of bacterial the control of bacterial infections. Broad spectrum antibiotics are highly
⁎
Corresponding author.
E-mail address: d.j.malik@lboro.ac.uk (D.J. Malik).
http://dx.doi.org/10.1016/j.cis.2017.05.014
Received 17 March 2017; Received in revised form 11 May 2017; Accepted 11 May 2017
Available online 14 May 2017
0001-8686/ © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).
D.J. Malik et al. Advances in Colloid and Interface Science 249 (2017) 100–133
attractive since they may be used against a wide range of bacteria purify the bacteriophage (e.g. for the removal of host cell proteins, host
without the need to identify the infection causing bacterial agent. This cell DNA or endotoxin for Gram negative bacteria). Typically phage
advantage has also resulted in significant misuse and overuse of anti- may then be re-suspended in simple saline or buffer and stored under
biotics contributing to the emergence of antibiotic resistance [1]. Bac- refrigerated conditions or processed further e.g. spray dried to improve
terial resistance to antibiotics has become a significant problem in the storage shelf life or encapsulated in micro- or nanoparticle formula-
treatment of a wide range of infections where the bacteria commonly tions.
causing the infections have become highly resistant to many classes of Phage are unique antimicrobials in that in the presence of host
antibiotics including third generation cephalosporins, carbapenems and bacteria, they are able to increase their numbers by infecting the bac-
fluoroquinolones [2]. Recent evidence of plasmid mediated colistin teria and producing virion progeny whilst minimally affecting the
resistance in Enterobacteriaceae is particularly troubling [3]. Between overall microbiota and body tissues. Phage carrying polysaccharide
1940 and 1962 > 20 new classes of antibiotics came to market; since depolymerases (polysaccharide degrading enzymes) in their structure
then the antibiotic pipeline has produced only two new classes [4]. may be able to disrupt biofilms [15–17]. For example Enterococcus and
Coates et al. [4] estimate that to stem the rising tide of antibiotic re- Staphylococcus phage capable of destroying biofilms have been reported
sistance 20 new classes of antibiotics are urgently needed. It is highly [18,19]. In addition to their potential as human biotherapeutics, phage
unlikely that this goal will be achievable even if governments provided are being developed for agricultural use to rid the environment and
stronger economic incentives to industry to develop new broad spec- domestic animals of pathogens that could contaminate the food supply
trum antibiotics. Czaplewski et al. [5] call for a sustained, concerted chain [20], in aquaculture for the treatment of fish pathogens [21] and
and coordinated international effort and discuss the need to invest in for the control of infections in intensively farmed poultry [22,23]. Re-
the development of alternatives to antibiotics (non-compound ap- cent advances in molecular biology and sequencing technology have
proaches) including probiotics, vaccines as well bacteriophage and improved our basic understanding of how bacteriophage interact with
phage derived lysins. bacteria and have opened new possibilities for utilising phage, in-
cluding genetically engineered phage, for potential therapeutic and
2. Bacteriophage as novel antimicrobials diagnostic applications [7,24].
Bacteriophages (viruses that infect bacteria) are highly abundant in 3. Phage therapy for acute and chronic infections
the environment and may be the source of low cost antimicrobials. The
focus of recent phage therapy approaches is on the use of lytic tailed Most of the recent phage therapy studies (using small vertebrate
phages all of which belong to the Caudovirales and include the animals) have investigated treatments focusing on acute infections
Myoviridae, Siphoviridae and Podoviridae families [6]. Members of the (Table 1). In acute infections the specific infection causing bacterium
Caudovirales have an icosahedral capsid head that contains double- may be identified using suitable rapid diagnostic methods (e.g. lateral
stranded DNA (15–500 kbp), and a tail with surface receptor proteins flow assays, PCR, MALDI-TOF Mass Spectrometry). In such instances,
that interact with surface features on the host bacterium [7]. Successful narrow spectrum antimicrobials such as bacteriophages may provide a
phage therapy requires interaction between the phage and the bac- suitable therapeutic alternative where organisms are resistant to
terium resulting in adsorption of the phage to the host bacterium fol- frontline antibiotics or to reduce the use of broad spectrum antibiotics
lowed by injection of the phage DNA. The pharmacodynamics of this as part of a global effort towards antibiotic stewardship. An example of
process has been modelled using mathematical models based on col- this is in cases of urinary tract infections where a significant proportion
loidal interactions [8–10]. Upon infection, the phage replication cycle of the cases are caused by a particular pathotype Escherichia coli with
ensues culminating (after a short latency period) in cell lysis and the specific virulence factors [25]. Animal studies have shown that phage
liberation of multiple phage virions (with a burst size that is typically may be effective in certain instances e.g. in treating acute respiratory
between 10 and 100 [11]). Gill and Hyman [12] and Weber-Dabrowska infections caused by Pseudomonas aeruginosa [26–28] and in the treat-
et al. [13] provide an overview of key considerations related to phage ment of systemic infections caused by S. aureus [29]. A significant focus
choice, isolation and purification for phage therapy. Phage therapy of phage therapy studies in animals has been around respiratory in-
practice relies on the isolation of naturally occurring phage abundantly fections, gastrointestinal infections and infections of the skin and
found in the environment [14]. Typically, phage are isolated from the wounds (Table 1). Phage therapy studies with animals has shown that
environment and screened against commonly occurring pathogenic in certain instances, it may help in reducing the densities of the in-
bacterial strains (to identify host ranges) and then evaluated using in fecting bacterial populations to levels that may allow the host immune
vitro and in vivo animal models. The limitation of host ranges is response to mount a successful defence and clear the infection
overcome through the use of phage cocktails to ensure sufficient cov- [26,27,30].
erage of commonly occurring strains and the use of phage mixtures A number of in vivo phage studies (with animals and humans) have
targeting different receptors reduces the probability of encountering suggested that phage therapy may be beneficial in the treatment of
phage resistant bacterial mutants. Phages are generally manufactured difficult to treat antibiotic resistant pulmonary infections (e.g. cystic
using standard fermentation process technology. In brief, host bacteria fibrosis [26] and pneumonia [31,32]), topical and wound infections
are grown in liquid culture in a bioreactor. During the log growth [33,34] and gastrointestinal infections [35].
phase, phages are added to the bioreactor to infect the bacteria. In-
cubation of phage with bacteria results in phage adsorption to the 3.1. Challenges of antibiotic resistance for respiratory infections
bacteria, infection and following a short lag period, release of bacter-
iophage virions. The resulting lysate contains the product (the ampli- Cystic fibrosis (CF) is a genetic disease of the lung resulting in re-
fied phage) along with bacterial debris and residual fermentation duced hydration and thickening of secretions covering the respiratory
media. Removal of cellular debris is typically done using centrifugation epithelium. Highly viscous mucus is not cleared by the epithelial cells
and/or filtration. Ion exchange, gel filtration etc. can be used to further and eventually leads to chronic inflammation and bacterial infections.
101
Two common bacterial strains isolated from sputum are Staphylococcus infections will also have great impact on TB control. Infection is es-
aureus and Haemophilus influenzae, which are also one of the first to tablished in a new host when inhaled Mtb bacilli in the lower re-
infect the lungs during childhood [36]. S. aureus can generally be spiratory tract are engulfed by alveolar macrophages [46]. The bacilli
controlled with antibiotics but infection usually predisposes chronic resist usual macrophage killing mechanisms and replicate, until such
colonization. The main pathogen infecting the pulmonary epithelium is time that the immune response triggers recruitment of systemic
Pseudomonas aeruginosa, which grows in biofilms within the CF lungs mononuclear cells which surround the focus of infection and form a
[37]. Several pulmonary pathogens are capable of forming biofilms in granuloma. The centre of the granuloma becomes hypoxic, pH drops
which the bacteria are encased in exopolysaccharide (EPS) and are less and conditions no longer support bacterial growth. Instead the Mtb
metabolically active [38]. The extracellular matrix in the biofilm im- bacilli remain in a dormant state with low metabolic activity, until such
pairs the action of antibiotics acting as a diffusional barrier. The low time that the immune system weakens, the granuloma breaks down and
metabolic activity of the bacteria in the biofilm also limits the efficacy with this the bacilli reactivate and replicate once more. During this
of many classes of antibiotics that target various metabolic pathways. active TB disease the patient excretes large number of bacilli when they
Some bacteriophage may be able to degrade biofilms and infect bacteria cough and can transmit the infection to new susceptible hosts. A
residing within them [39,40]. Burkholderia cenocepacia are opportu- number of recent studies have started looking at phage therapy for the
nistic pathogens in CF causing acute pulmonary disease and sepsis or treatment of TB infections [47,48] and strategies to tackle the challenge
chronic infection characterised by accelerated decline in lung function. of phage delivery to these intracellular bacteria.
Multidrug resistant phenotypes are exhibited by most strains. There is a
need for alternative therapies to treat B. cenocepacia; several virulent 3.2. Challenges of antibiotic resistance for wounds
phages for B. cenocepacia have been characterised. A number of in vivo
phage therapy studies have looked at P. aeruginosa [26,27,28,41] and B. Chronic wounds are a severe worldwide problem. A recent pre-
cenocepacia [42] infections in animal models. valence study estimated 2.2 million wounds managed by the UK
Ventilator-associated-pneumonia (VAP) occurs often in patients in National Health Service (NHS) in 2012/2013 costing the UK NHS £5.3
an intensive care unit setting. A clear respiratory airway is a necessity billion [49]. Studies suggest that wound management accounts for over
for patient survival; after being intubated and receiving mechanical half of community health nurse resources in European settings [50].
ventilation, patients suffer from pneumonia caused by infection with P. The majority of chronic wounds are colonized with bacteria that form
aeruginosa, S. aureus, Streptococcus pneumoniae, H. influenzae and biofilms, thereby compromising wound healing by making them more
Acinetobacter baumannii which take advantage of the weakened immune refractory to treatment and slow tissue repair by stimulating chronic
system of the patient. One of the major problems is antibiotic re- inflammation at the wound site [51]. Antimicrobials including anti-
sistance. Klebsiella sp., Enterobacter sp., Serrati sp., and Burkholderia sp., biotics and antiseptics are used clinically to decrease the bacterial load
have also been reported as infecting bacteria. Klebsiella pneumoniae is an and promote wound healing [52]. However, systemic antibiotic therapy
opportunistic bacterial pathogen responsible for much community-ac- may be of little benefit due to poor diffusion of antimicrobials through
quired pneumonia and a significant proportion of hospital-acquired the biofilm and the poor efficacy of antibiotics against bacteria with
pneumonia [31]. It is a leading cause of pneumonia morbidity and reduced metabolic activity. This reduced activity increases antibiotic
mortality with resistance to carbapenems and other antibiotics [43]. A tolerance since many classes of antibiotics are only effective against
number of in vivo phage therapy studies have looked at B. cepacia [42] actively dividing bacteria by targeting peptidoglycan produced in the
and K. pneumonia [31,32] infections in animal models. cell wall (β-lactams), protein (aminoglycoside) synthesis, or DNA re-
Tuberculosis (TB) is a global health problem; one-third of the plication (quinolones) [53]. Polymicrobial biofilms in chronic wounds
world's population is infected with Mycobacterium tuberculosis (Mtb), are recognised as an important factor in the failure to achieve wound
the agent of TB, and at risk of developing disease and transmitting in- healing in a timely fashion. As an example, diabetic foot infections
fection. Annually there are over 10 million new cases of TB and an (DFI) typically tend to be polymicrobial with aerobic Gram-positive
estimated 1.4 million deaths [44]. Multidrug resistant tuberculosis cocci, especially staphylococci, the most common causative micro-
cases are a significant healthcare problem with extensively drug re- organism [54]. S. aureus and P. aeruginosa are considered important
sistant M. tuberculosis practically untreatable in low-income countries pathogens in antibiotic resistant wound infections. A number of animal
[45]. Multi-drug resistant (MDR) Mtb strains are resistant to isoniazid studies have looked at phage therapy for treating S. aureus and P. aer-
and rifampicin, two of the frontline antibiotics used to treat TB. The uginosa wound infections [33,34].
World Health Organisation (WHO) estimate there were nearly 500,000
new cases of MDR TB in 2015 [44]. Most alarming are the development 3.3. Challenges of antibiotic resistance for gastrointestinal infections
of extensively drug resistant (XDR) strains, which are additionally re-
sistant to second line antibiotics. Although antibiotic treatment rapidly Enteric pathogens form a large part of the healthcare burden around
reduces the infectiousness of patients, those with drug-resistant infec- the globe. The most common infection causing bacteria are Clostridium
tions remain infectious during therapy until the drug resistance is difficile causing C. difficile associated diarrhoea [55], Shigella dysenteriae
identified and the treatment regimen is changed. Treatment for TB is causing shigellosis [56], Escherichia coli causing gastroenteritis [57],
protracted, requiring antibiotics to be taken for at least 6 months, rising Vibrio cholerae causing cholera [58], Salmonella enterica causing sal-
to two years for MDR infections. The antibiotics can have undesirable monellosis [59], Listeria monocytogenes causing listeriosis [60] and
side effects, and interruptions to therapy (intermittent supply or patient Helicobacter pylori causing chronic gastritis [61]. The primary route of
non-compliance) provide the opportunity for the development of drug infection in low income countries is through the faecal-oral route or
resistance and treatment failure. There is therefore, an urgent need for from contaminated food and water [62]. Treatment usually involves
new therapies to treat TB and also to shorten the length of treatment. antibiotics however, with emerging antibiotic resistance, infections
The Mtb bacillus is transmitted in aerosol droplets and therapies which with antibiotic resistant organisms is on the rise, hence the need for
reduce patient infectiousness or prevent the establishment of new alternative therapy options. In addition to humans, these enteric
102
pathogens are also capable of infecting farm animals resulting in high High phage densities are needed in order to arrest the growth of phage
antibiotic use for animal husbandry, emerging resistance and risk of susceptible bacteria (Fig. 1 a, b). We have indicated an arbitrary upper
contamination of the food chain. Numerous studies have evaluated the bacterial concentration threshold of 109 CFU/ml as an aid to guide the
potential of phage therapy for the treatment of gastrointestinal infec- reader's eye. Adsorption of phage to bacteria results in an initial drop in
tions caused by E. coli, S. enterica, V. cholera and C. difficile (Table 1). phage concentration (see experimental results by Cairns et al. [9] with
Campylobacter jejuni showing this). The timing of rapid in situ phage
3.4. Phage-bacterium population dynamics - the need for bacteriophage amplification is highly dependent on the concentration of the bacteria
encapsulation [9]. Successful amplification of phage is dependent on high bacterial
concentration hence efficient killing is not achieved until both phage
There are a number of specific features regarding phage-bacterium and bacteria concentration increase significantly (Fig. 1 b). See also
dynamics that are worth highlighting in the context of phage therapy. experimental results in support of this in the work by Tanji et al. [64].
Bacterial killing by phage is dependent on phage particles adsorbing to Resistant bacteria quickly become the dominant population (Fig. 1 a, b)
the target bacteria [9]. Phage-bacterium binding kinetics may be unless a phage mixture (cocktail) is used containing high doses of the
modelled as a simple first order process with respect to the con- different phage strains capable of killing the resistant mutant popula-
centrations of bacteria and phage populations respectively [8,63]. tion (Fig. 1 c, d). It is known that phage clearance mechanisms rapidly
Payne and Jansen [63] developed a mathematical model focusing on reduce the phage concentration in the absence of host bacteria [58].
the dynamics of phage infection of bacteria during the bacterial ex- Using high initial bacterial concentration, a high phage dose is
ponential growth phase. Levin and Bull [8] extended the model to in- quickly able to arrest the rise of bacterial growth (Fig. 2 a). Tanji et al.
clude the effect of bacterial mutation leading to phage resistant popu- [64] show this for E. coli using an in vitro chemostat experimental
lations and the effect of the immune response in combating bacterial system. However, if low starting concentrations of bacteria are present,
infections. Cairns et al. [9] fitted in vitro phage-bacteria population phage concentrations decay significantly and phage are unable to am-
dynamics (time series data for Campylobacter jejuni) to extract interac- plify and achieve a sufficiently high concentration to eradicate the in-
tion parameters; the model included the rate at which resistant cells fecting bacteria until the bacteria have had time to grow bacterial
arise by mutation from a susceptible population. Their model and that numbers substantially (Fig. 2 b). In such circumstances it takes time for
of Levin and Bull [8] helps explain the in vitro observations (see Tanji the bacterial population to grow before phage are able to adequately
et al. [64]) showing an increase in concentration of resistant bacteria in multiply in situ and then arrest bacterial growth (experimental results
the presence of high concentrations of phage. In in vitro studies, the by Cairns et al. [9] show this). During this delay phage concentrations
rate of growth of resistant bacteria tends to be similar to that of the may decay significantly thereby resulting in poor phage therapy out-
original susceptible population. The fitted rate of mutation was found to comes. If phage are administered prophylactically too early prior to
be of the order of 10− 5 h− 1 (which is relatively fast). An alternative infection or at the early onset of infection (when the bacteria con-
explanation may be the presence of a small number of resistant cells centration is low), clearance of phage by the host immune system or by
present at the start of the culture [9]. Measures to suppress growth of other mechanisms (e.g. shortening of intestinal transit times during
resistant bacteria would include use of multiple phage strains in- diarrhoea) may result in lowering of the in situ phage concentration
corporating phage that bind to different receptors. The purpose of the resulting in poor phage therapy outcomes (Fig. 2 b). Under such con-
modelling is to illustrate a number of important features that provide ditions encapsulation may be a good strategy (Fig. 2 c). Encapsulation
insights that may help inform in vivo phage therapy. Modelling can be of phage and their slow controlled release may help in ensuring that the
used to determine the effects of parameters such as the dose of phage in situ phage concentration remains at a therapeutically effective level
delivered, the rate at which it is delivered (e.g. controlled release) or (over a realistic time period) allowing phage to amplify once the bac-
bacteria growth rate, on phage-bacteria population dynamics and on teria concentration increases to levels sufficient for in situ phage am-
phage therapy outcomes. It may be used to help interpret results of in plification (Fig. 2 c).
vivo studies and inform the design of future studies. Our simulations show that the in vivo phage concentration at the
We have undertaken numerical simulations of phage-bacterium site of infection is an important parameter governing the efficacy of
population dynamics for an acute infection model to highlight key phage therapy. Loss of phage activity during formulation and storage,
features including the dynamics of phage interaction with bacteria or phage inactivation due to the in vivo environment (e.g. acidic pH of
during the exponential growth phase (model equations and parameters the stomach or enzymatic activity) may result in poor phage therapy
used for simulations are laid out in Appendix A). Our model is similar to outcomes. These issues have received little attention in phage therapy
that of Levin and Bull [8] for phage-bacteria interactions incorporating literature. Research is therefore needed in the area of phage formula-
phage resistant bacteria arising from bacteria mutation (we have used tion to ensure phage titre remains stable during storage (e.g. over a
the mutation rate used by Cairns et al. [9]; we haven't included a model 24 month period). Phage cocktails necessitate careful consideration of
of the immune response). Phage growth is determined by a number of formulation conditions as each type of phage making-up the cocktail
parameters, the phage-bacteria adsorption rate constant, burst size (the may require individually tailored formulations to ensure phage viability
number of phage released for each infected bacteria), latent period and stability during storage.
(following infection how long it takes for the phage replication cycle to
be completed in a bacterial cell), the bacteria growth rate, phage and 3.5. Phage therapy animal studies – acute infections
bacteria elimination rates due to host response and the initial con-
centrations of bacteria and phage [65]. A unique feature of our model is 3.5.1. Key points arising from reviewing phage therapy literature focusing
that we have also incorporated the effect of controlled release of phage on acute infections in animals
which to our knowledge has not previously been done. Most in vivo phage therapy studies (in animals) have delivered the
Our simulations demonstrate the conditions under which phage phage and bacteria dosed simultaneously together (Table 1). Typically,
therapy may be able to limit the proliferation of bacterial populations. a high ratio of phage to bacteria has been used (phage concentrations
103
10–1000 times higher than the infective dose of bacteria). Phage necessitate the use of phage cocktails rather than single phage treat-
treatment of acute infections caused by both Gram positive and Gram ments. This is because phage mixtures targeting different receptors
negative bacteria have been studied (Table 1). reduce the probability of encountering phage resistant bacterial mu-
Phage have been administered using different delivery routes in- tants (see modelling section above). Denou et al. [86] suggest that
cluding intramuscular [34,66–69], subcutaneous [29,31–34,42], in- cocktails containing between 10 and 16 phage may be needed to cover
traperitoneal [31,34,67,70–72], intravenous [29,68], intranasal between half and two thirds of E. coli strains (representing the five main
[26–28,42], orogastrically [57,58,64,70,73–81], intracranial [69] or pathogenic types isolated from patients suffering from diarrhoea) in
added to animal feed [75,82] or water [83]. The route of phage admin- order to cure E. coli related diarrhoea. Cocktails of phage may also
istration has been shown to affect efficacy outcomes [34,42,66]. In a provide synergistic effects such as access to bacteria in mucus and
number of studies intraperitoneal delivery was shown to result in higher biofilms with a phage possessing biofilm/mucus degrading enzymes
concentrations of systemic phage, delivered earlier, remaining in circu- enhancing access within mucus/biofilms for other phage better suited
lation for longer and allowing phage access to numerous host organs in- to killing the bacteria [87]. This point has significant implications for
cluding lungs, spleen and kidneys [34,32]. Better therapy outcomes were phage formulation since different phage within a cocktail may need
seen for lung infections when phage were dosed using the intraperitoneal different formulation ingredients to aid shelf life during storage in order
or systemic route instead of using pulmonary delivery [31]. to retain infectivity. A number of animal studies (discussed below) have
A number of in vivo animal studies have shown dose dependent shown that regular repeated doses of high phage titres resulted in better
recoveries with high doses of phage resulting in better clinical outcomes outcomes rather than a single dose. Encapsulation of phage in con-
[33,66–68]. Delaying the start of phage treatment post-infection typi- trolled release systems may allow constant release (zero order release)
cally resulted in poor outcomes [33,67,68]. This was attributed to high of phage e.g. using readily available technologies such as osmotic
growth rates resulting in high concentrations of replicating bacteria pumps or constant reservoir systems used in the pharmaceutical in-
overwhelming the immune system with bacterial toxins [68]. dustry. Vandenheuvel et al. [88] discuss the importance of phage sta-
Inoculation of control animals (not infected with bacteria) with high bility and the unacceptability of phage titre drop in view of pharma-
phage doses was shown not to result in adverse health symptoms in ceutical regulations. The results of animal studies (above) provide
animals [67]. A number of in vivo studies reported rapid drop in phage strong evidence of the need to deliver high titres of phage at the site of
titre in the absence of host bacteria suggesting clearance by the host infection. Phage dose-response studies in animals suggest that a drop in
immune response or other removal mechanisms [32,58]. Encapsulation titre of 1-log during storage could result in phage therapy failure. Re-
of phage in liposomes (for treating gastrointestinal infections) was duction in phage titres due to phage inactivation attributed to stomach
shown to increase the retention time of phage in the host with better acidity was partly blamed for failure of a recent clinical trial aiming to
therapeutic efficacy [31]. show reduction in acute bacterial diarrhoea symptoms in children using
In situ amplification of phage (at high phage doses) has been ob- phage therapy [35].
served in some instances [71]. Frequent administration of high phage The pharmacokinetics of phage delivery to various animal organs
doses was found to result in better outcomes [64,76,78,80]. This is an depended on the route of administration. In a number of studies, in-
important point considering in vivo phage-bacteria dynamics. Better traperitoneal delivery of phage resulted in higher concentrations of
phage therapy outcomes were noted when both phage and bacteria phage in systemic circulation and those found in different organs in-
concentrations tend to be high. Phage titre reduction in vivo may result cluding lungs. Encapsulation of phage in liposomes was found to in-
in poor phage therapy results when bacterial concentrations may be too crease the length of systemic circulation of phage. There is some evi-
low for rapid phage amplification. Reduction in phage titre may also dence in literature that phage encapsulated in liposomes may enhance
explain results of a number of studies that evaluated prophylactic phage delivery to intracellular infections (discussed later).
therapy with mixed results [70,73,75,79,80,83]. Use of phage cocktails Bacteriophages have a narrow host range. They are specific not only
rather than single phage was found to yield better results [58,64] and in in terms of killing a particular species, but often they only target a sub
some instances reduced the chances of finding phage resistant mutants. set of strains within the species. In an outbreak situation where the
A number of studies did find phage resistant mutants [66,69]. The causative agent and strain information is available, selection of a re-
complex architecture of the intestinal tract was found to present a levant phage for therapy may not be a problem [26]. However, as phage
challenge for phage therapy. Temporary reduction in bacterial faecal host range is narrow, rapid diagnosis of the causative agent and the
shedding attributed to phage treatment was typically followed by an strain would benefit accurate phage selection. Most phage therapy
increase in faecal bacterial concentration without clearance of gastro- studies in model animal systems (discussed above) have used simple
intestinal infections [57,58,75,77,83]. phage solutions prepared immediately prior to use. Phage banks may
Results of a number of animal studies suggest that stopping rapid need to regularly screen phage against common infection causing
amplification in bacterial numbers is an important factor in therapy strains and manufacture these in batches to make them readily avail-
outcomes [26–28]. Arresting infection by early administration of high able. The phage would need to be processed in such a manner so as to
phage dose when the bacteria are also highly susceptible to phage in- ensure long shelf life (typically 12 months) e.g. formulated in excipients
fection was found to correlate with better phage therapy outcomes. and spray dried or freeze dried to ensure accurate dosage at the time of
Administration of high concentrations of phage may result in multiple therapy.
phage particles adsorbing per bacterial cell which may lead to bacterial Characterisation of the kinetics of phage binding to target bacterium
cell lysis due to phage binding without phage virion production. Del- and in vitro evaluation of phage-bacterium population dynamics as part
bruck [84] coined the term "lysis from without". Large numbers of of the selection tests (e.g. host range, presence of integrase genes) as-
phage adsorbing to the bacterial cell wall may lead to cell wall damage certaining the suitability of phage for therapeutic use is often over-
due to the enzymatic action of tailed phage enzymes acting on the cell looked and needs greater attention.
envelope/wall [85]. A detailed list of literature studies covering phage therapy in various
Infection is generally induced in animal models with a single bac- animal models is provided in Table 1 and more detailed commentaries
terial species, however, even single bacterial triggered infections on some selected papers are now given as follows.
104
Table 1
Summary of phage therapy animal studies to treat acute local, systemic, gastrointestinal and pulmonary bacterial infections.
Host organism Animal Infection Delivery Phage therapy Outcome

b - bacteria – good
p/Φ – phage / encapsulated phage - modest
* - with antacid added – no effect
Single phage (*) / Cocktail ( )

Injected into thoracic air sac
Phage added to feed water
Reduction in concentration
Delayed phage treatment

of bacterium or mortality
Phage resistant mutants
Pprophylactic treatment
Clearance of infection
R – repeated dosage
Dose related studies

Bacteria Log(CFU)
E – encapsulated
Phage Log(PFU)
Gastrointestinal
B. cenocepacia
Intraperitoneal
Subcutaneous
P. aeruginosa
K. pneumonia
Orogastrically
Intramuscular
post infection
Tracheotomy
Amplification
Septicaemia
V. vulnificus
Intravenous
Intracranial
S. enterica
Pulmonary
E. faecium
V. cholera
Meningitis
Instillation
Intranasal
C. difficile
S. aureus
Chickens
Systemic
Hamster
C. jejuni
Wound
Mouse
Rabbit
E. coli
Cattle
Local
Aero
Ref.
Rat
Pig
[66] Ec Sy bp 7 2-8 *
b p
[69] Ec Se bp 6 2-6
[69] Ec M bp 3 6-8 *
[67] Ef Sy bp 9 8 *
[68] Vv Sy L p b 2-6 4-8 *
[71] Sa Sy bp 9 9-11 *
[33] Sa W bp 8 6-9 *
p b
Pa W bp 2 8
[34] p b
[29] Sa Sy L bp bp 8 7-9(R) *
[31] Kp Sy pΦ b 4 9, 7(E) *
[69] Ec G p b 10 10 *
[83] Ec G b p 4-8 5-6
[64] Ec G bp* 9 8-10(R)
[73] Cj G bp 5 9-10 */
Pa G 6-10 10 *
[70] bp bp
[74] Se G bp 4 8
[75] Ec G bΦ 11 10-11
[57] Ec G bΦ* 5 ~5(R)
[76] Cd G bp* 3 ~8(R) */
[78] Se G bΦ 7 10(R)
[77] Se G bΦ 7 10(R)
[58] Vc G bp* 5, 8 5-7, 8 */
[58] Vc G bp* 8 9 */
[59] Se G bΦ 9 ~10(R)
[80] Se G bp 6 ~10(R)
[80] Se G bp 5 ~10(R)
[92] Ec P p p p 6-8
[90] Ec P p b 4 7-9
[42] Bc P p p b b 7-9 7-9 *
[26] Pa P bp 7 7-9 *
[27] Pa P bp 7 8
[28] Pa P bp 6 6-8 *
[31] Kp P pΦ b 4 2-4 *
[32] Kp P p b 7-8 9-11 *
3.5.2. Parenteral phage delivery (animal studies) mutants were found at the site of infection but were considered much
A number of animal studies have looked at parenteral delivery of less virulent.
lytic bacteriophage following delivery of an infectious dose of bacteria Barrow et al. [69] inoculated (intramuscularly (septicemia model)
(Table 1). Early work by Smith et al. [66] demonstrated the potential of and intracranially (meningitis model)) newly hatched and 3 week old
intramuscular and intravenous administration of phage (108 PFU/ chickens with E. coli and phage and showed dose dependent recovery.
mouse) against E. coli infection (intramuscular injection of LD50 dose of Administration of higher concentration of phage resulted in sig-
107 CFU/mouse) in a mouse model. The effect of delaying treatment nificantly better outcomes. The same study also evaluated oral E. coli
(8 h delay in treatment) showed favourable results. Prophylactic administration to Aberdeen Angus Calves with concomitant in-
treatment up to 2 days before induced infection resulted in significant tramuscular phage delivery. The phage were unable to arrest the in-
protection in animals. Intramuscular injection of bacteria resulted in testinal growth of E. coli and the presence of phage resistant mutant
high concentration of bacteria at the site of infection and in the blood, strains were found in the gut one day following E. coli inoculation and
spleen and liver ~24 h post-inoculation. Within 5 min of phage ad- two days later in the blood (indicating translocation of phage from the
ministration, high phage numbers were found in the muscle, blood, gut into blood e.g. through intra-abdominal abscesses) of the infected
spleen and liver. They persisted for shorter durations in the blood and calves. This study showed the importance of phage dose and route of
liver compared with the inoculated muscle and spleen. Phage resistant delivery on therapy outcomes (mortality).
105
10 10 10 10
10 10 10 10
8 8 8 8
10 10 10 10
Concentration
6 6 6 6
10 10 10 10
4 4 4 4
10 10 10 10
2 2 2 2
10 10 10 10
0 0 0 0
10 10 10 10
0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20
time, hours time, hours time, hours time, hours
(a) (b) (c) (d)

Fig. 1. Population dynamics of bacteria and phage.
Blue lines – phage P1 (solid line) and P2 (dashed line). Red lines – bacteria B1 (solid line) and B2 (dashed line); B2 is resistant to P1.
(a) – low dose of single phage (P10 ≈ 103 PFU/ml) is insufficient to prevent bacteria concentration exceeding upper threshold;
(b) - high dose of phage (P10 ≈ 106 PFU/ml) arrests B1 growth, but phage resistant mutant bacteria B2 grows unchecked;
(c) and (d) - mixture of two different phage; (c) - low dose of P2 is insufficient to prevent resistant bacteria B2 concentration exceeding upper threshold; (d) - high dose of both phage
prevents both bacteria concentrations from reaching upper threshold. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
Using a Enterococcus faecium bacteremia mouse model Biswas et al. treatment (≥ 6 h) post-infection showed lack of efficacy and resulted in
[67] investigated the effect of phage dose in E. faecium infected mice treatment failure. Dose response studies with animals infected with
(concomitant delivery of phage and bacteria delivered via in- 103 CFU of bacteria and immediately given either 104, 106 or 108 PFU
traperitoneal injection) as well as the effect of delaying the start of of phage showed only the higher concentration of phage dose at
phage treatment (5 h, 8 h, 12 h, 16 h and 24 h after inoculation with 108 PFU afforded significant protection to the treated mice. The need
bacteria). The severity of illness symptoms exhibited by mice was for high phage dosages was attributed to the short doubling time of the
correlated with the bacterial concentration in the blood and this was bacterium (45 min) which resulted in a high growth rate of bacteria
shown to be dependent on the concentration of phage used for treat- post-infection. The study also evaluated the half-life of phage (2.2 h at
ment. Mice inoculated with high titres of phage suspension (without 108 PFU) injected intravenously and found that phage concentrations in
infection with bacteria) showed no adverse symptoms. 24 h delay in blood rapidly decreased due to clearance by the reticuloendothelial
phage treatment resulted in 60% (3 out of 5) of the mice dying after system. Phages are known to induce antibodies aiding their removal
96 h. The dosage and the timing of the delivery of phage were found to from circulation as well as neutralising their interaction with bacteria
be important in arresting bacterial growth and in mice recovery. [30,42,89].
Cerveney et al. [68] evaluated the efficacy of phage (used locally Matsuzaki et al. [71] showed dose dependent pharmacodynamics
and systemically) against the Gram negative bacterium Vibrio vulnificus using concomitant intraperitoneal injection of S. aureus followed by
(causes septicemia after ingestion of contaminated oysters and necro- phage (delivered within 60 min of bacterial inoculation) with high
tising fasciitis in contaminated wounds) in an iron-dextran-treated concentrations of phage observed in blood (at high phage doses
mouse model. Bacteria were injected subcutaneously whereas phages 1011 PFU to treat 109 CFU of S. aureus) indicating in situ amplification
were injected intravenously. A high phage dose (108 PFU) was given of phage. In a rabbit model of wound infection induced by S. aureus,
(the animal was subcutaneously given a dose of 106 CFU of a highly Wills et al. [33] used a high dose of phage (109 PFU, delivered sub-
virulent clinical isolate of the bacterium to create a skin lesion) and cutaneously) whilst simultaneously infecting with S. aureus (~ 108 CFU)
shown to reduce the bacterial burden of the lesion and in the bacterial to show protection against abscess formation. The study also carried out
CFU in the liver (14 h post-infection). Experiments with delayed a dose response study and showed that only at the highest phage dose
10 10 10
10 10 10
Concentration
Concentration
Concentration
5 5 5
10 10 10
0 0 0
10 10 10
0 5 10 15 0 5 10 15 0 5 10 15
Time, hours Time, hours Time, hours
(a) (b) (c)

Fig. 2. Population dynamics of bacteria and phage.
Blue lines – phage concentration; red lines – bacteria concentration.
(a), (b) – effect of phage dose (104–108 PFU/ml) and initial bacteria concentration (104, 107 CFU/ml);
(c) - effect of continuous sustained delivery of phage for 1 h, 4 h and 10 h (solid, broken and dashed lines accordingly). (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
106
was there an absence of abscess formation. The median area of abscess in polymer or liposome encapsulation systems [57,75,77,78]. Colom
formation and the concentration of bacteria found in the abscess (CFU/ et al. [77] showed that encapsulation of phage in liposomes resulted in
ml) showed an inverse correlation with the dose of phage given. The significantly longer periods of phage retention (several days longer) in
effect of delaying treatment (by giving a high phage dose (109 PFU) 6 h, the caecum of chickens and lower number of bacteria colonizing the
12 h and 24 h after bacterial infection) resulted in abscess formation in gut.
all animals. Chibani-Chennoufi et al. [83] showed survival of T4-like E. coli
McVay et al. [34] in a mouse burn wound model compared in- phage through the gastrointestinal tract of mice when added to
tramuscular, subcutaneous and intraperitoneal routes of phage delivery drinking water. The minimal oral dose for consistent faecal recovery
using a cocktail of Pseudomonas aeruginosa phage. Phage administered was 104 PFU/ml of drinking water. E. coli infected mice showed 105
by the intraperitoneal route (~ 108 PFU) afforded better protection fold amplification of phage with concomitant drop in faecal E. coli
against P. aeruginosa infection (~ 102 CFU injected subcutaneously at counts from 108 to 104 CFU/ml. However, phage sensitive E. coli con-
the burn site) with higher concentrations of phage found in the blood, centration recovered post-treatment even when residual high faecal
spleen and liver of animals. Phage resistant mutants were not found phage titres were present, suggesting phage sensitive E. coli were able to
even in animals that had died. Intraperitoneal delivery of phage was replicate in the gut and that the concentration of phage present in the
found to result in delivery of higher phage dose, delivered earlier into gut was not high enough to eliminate the E. coli. The authors suggested
the blood and other tissues, and delivered for a more sustained period sites in the gut where E. coli may be protected from the phage (cells
of time. growing as microcolonies in the mucin layer). Other studies have also
Capparelli et al. [29] intravenously infected mice with S. aureus reported an initial reduction in faecal shedding followed by recovery of
(dose, 108 CFU/mouse) followed by immediate intravenous injection of bacterial numbers [64,73,74].
phage (107, 108 or 109 PFU/mouse). All the mice in the control group Tanji et al. [64] used a chemostat (in vitro studies) to investigate the
(infected but not treated with phage) and those treated with the lowest dose and frequency of phage administration on phage-E. coli population
phage dose (107 PFU/mouse) died within 4 days (10/10 mice). The dynamics. The chemostat experiments suggested repeated high doses of
mice treated with the intermediate dose (108 PFU/mouse) were only phage were needed to keep the E. coli concentration in check. The re-
partially protected (2/5 mice survived) whilst mice treated with the sults informed phage therapy in mice where E. coli was orally ad-
highest dose (109 PFU/mouse) all survived (5/5 survived). 4 days after ministered (dose 109 CFU/mouse) using a plastic sonde directly into the
phage injection, high concentrations of bacteria (in kidney, heart, stomach followed 2 days later with phage cocktail delivery to the sto-
spleen and lung) were recorded for the untreated controls whereas no mach in a similar manner (single dose of 108 PFU/mouse or 1010 PFU/
bacteria were found in the same organs for mice treated with 109 PFU/ mouse or repeated daily doses of 1010 PFU/mouse). Daily administra-
mouse. Lack of phage amplification in blood (attributed to a replication tion of phage resulted in a stable concentration of phage in the faeces
threshold) was in all likelihood down to the low concentrations of and significant reduction in the E. coli concentration in the faeces. A
bacteria in blood making it difficult to accurately monitor phage am- cocktail containing three phage was employed however, the con-
plification under such conditions. At low levels of systemic infection (S. centration of phage in the faeces suggested one phage dominated and
aureus dose 106 CFU/mouse) a 10 day delay in phage therapy was present at a significantly higher concentration in comparison with
(109 PFU/mouse) cleared S. aureus infection in the animals. Capparelli the other two (indicating differences in the pharmacodynamics of the
et al. [29] also evaluated the efficacy of phage therapy for local S. three phage). Nine days after the start of phage therapy with daily
aureus infections (subcutaneous injection of bacteria and phage). Con- administration of phage, no E. coli were detected in the lower small
comitant injection of phage prevented formation of abscess in the mice. intestine and the colon of treated mice. A number of studies have used
Delayed phage treatment (phage administration 4 days after onset of repeated dosing of phage and shown reduction in intestinal coloniza-
infection) did not prevent abscess formation. There was a reduction in tion with bacteria [57,76–80]. This study stands out as one of very few
the bacterial load in the abscess and the weight of the abscess when where elimination of E. coli was observed for treatment of gastro-
multiple injections (4 injections/day) were used. Repeated doses were intestinal infections.
found to be more effective than a single dose in reducing the bacterial Wagenaar et al. [73] evaluated the prophylactic or therapeutic ef-
load in the abscess and the weight of the abscess. S. aureus can adopt an ficacy of phage treatment for control of Campylobacter jejuni (C. jejuni,
intracellular lifestyle. Capparelli et al. [29] also evaluated the efficacy which colonizes young broiler chickens, is an important human pa-
of phage activity in a S. aureus intracellular infection mouse peritoneal thogen and is the most common cause of bacterial gastroenteritis
macrophage cell line model. Phages were not able to penetrate the worldwide). A modest 1-log reduction in bacterial CFU counts
macrophages and kill the intracellular S. aureus. Using a Trojan horse (~ 109 CFU/g faeces) was observed in both the prevention and the
approach, S. aureus infected with phage were able to access the in- therapeutic groups compared with the control group (not given phage).
tracellular compartment of the macrophages and kill S. aureus cells Phage selection was on the basis of activity across a wide range of C.
residing within. This study suggests the possibility of encapsulating jejuni strains.
phage in systems that may facilitate access to intracellular compart- Watanabe et al. [70] investigated the efficacy of phage therapy
ments where infecting bacteria may be targeted (more on this later in using a murine model of gut-derived sepsis caused by P. aeruginosa
the section on encapsulation of phage in liposomes to target in- (resembling the clinical pathophysiology of septicemia in humans).
tracellular pathogenic bacteria). Administration of cyclophosphamide (to suppress the immune system)
and ampicillin (to disrupt the gut microflora) was used before oral in-
3.5.3. Oral phage delivery (animal studies) oculation with P. aeruginosa to achieve gut-derived sepsis. Intestinal
Animal studies using oral delivery of bacteriophage have focused on overgrowth in the gut resulted in P. aeruginosa crossing the gastro-
acute gastrointestinal infections (Table 1). The main concern here is intestinal mucosal barrier by bacterial translocation and spreading
phage inactivation due to the acidic and proteolytic environment of the systemically in the mice. Timing of phage treatment was found to in-
stomach. The use of antacids e.g. calcium carbonate or bicarbonate fluence the efficacy of phage therapy. Mice treated with a high dose of
prior to infection challenge or with phage delivery in order to neutralise phage 1 day post-infection with P. aeruginosa showed improved survival
the stomach acidity has been common practice in a number of studies rates over controls (mice not treated with phage). Average numbers of
[23,58,64,76,78]. A few studies have employed encapsulation of phage viable bacteria in blood, liver, spleen and mesenteric lymph nodes were
107
significantly lower in phage treated mice compared with controls. but continuing regular 8 h phage treatments delayed onset of symptoms
Faecal counts were also lower in phage treated mice as were cytokine by 33 h (due to toxin production) in phage treated animals compared to
levels in serum and liver. Compared to oral administration, intravenous untreated animals was observed. Complete protection was not
and intraperitoneal administrations of phage were found to result in achieved. The amount of phage delivered to the cecum or colon was not
better outcomes. The infectious dose was 106 CFU/mouse with phage quantified however, high titres of phage 107 PFU/ml were recovered
dose of 1010 PFU/mouse. Only simultaneous intraperitoneal adminis- from the cecum and colon.
tration of phage and bacteria resulted in improved mice survival. Recently Colom et al. [77] used a cocktail of three Salmonella phage
Andreatti Filho et al. [74] showed short term (24 h after infection) individually encapsulated in cationic liposomes to treat Salmonella in-
reduction in Salmonella enterica detection (dose ~ 104 CFU/chick) from fection in commercial broilers under farm-like conditions. Bacter-
caecal tonsils following phage administration (phage cocktail con- iophage retention in the chicken caecum was investigated by giving an
taining 45 different phage administered using oral gavage, 108 PFU/ oral dose of phage (1010 PFU/chicken) of either free phage (1:1:1
chick). However, no significant differences were observed between the mixture of the three phages ϕ20, ϕ78 and ϕ87) or liposome en-
phage treated and the control groups after 48 h. The cocktail used capsulated phage cocktail to 63 1-day old chickens. Caecum samples
wasn't sufficiently well characterised in terms of individual phage collected from euthanized animals showed significantly higher numbers
concentrations and the kinetics of individual phage binding and their of chickens treated with encapsulated phage had phage in their caecum
lytic activity; a recurring theme in the reading of literature in this area. 48 h and 72 h later. Bacteriophage therapy against Salmonella was
Stanford et al. [75] encapsulated E. coli phage in a commercial pH evaluated over 17 days in chickens orally administered either the free
responsive methacrylate polymer using a spray drying process (a 1 log phage cocktail or encapsulated phage cocktail (bacterial dose at day 0:
reduction in activity due to spray drying was noted). Oral inoculation of 107 CFU/chicken, phage dose 1010 PFU/chicken). Oral doses of phage
1011 CFU/steer of a five strain mixture of E. coli O157:H7 was followed were administered daily for 9 days (from day − 1 to day 7 after Sal-
either by encapsulated phage given as a single oral bolus (1010 PFU/ monella infection). Over the course of daily phage therapy (days 1, 3, 6)
steer) or mixed-in with feed (1011 PFU/steer). Higher concentration of both groups receiving phage treatment showed lower (~ 3 log reduc-
faecal shedding of phage (2-log higher) was noted for encapsulated tion) Salmonella concentration in the cecum compared with the control
phage given as part of the feed rather than bolus suggesting differences group. Once daily phage treatment had stopped (day 8 onwards), Sal-
in in situ phage concentrations. Phage were shown to replicate in situ monella concentration in the free phage treated group returned to levels
over a period in excess of ten days following treatment and continued to similar to the control group. However, the liposome encapsulated
be shed 42 days post-treatment (albeit at lower concentrations in com- phage treated group continued to show lower Salmonella concentration
parison with the first ten days) however, no significant differences in in the cecum compared with the control group up to day 15. Salmonella
reduction in E. coli shedding was observed between the phage treated infection was not eradicated by either treatment which was attributed
steer and the controls. Insufficient understanding of the in vivo phar- to high initial Salmonella concentration in the cecum (106 CFU/g of
macodynamics of the phage against the E. coli strains may have resulted cecum), which was considered higher than naturally occurring levels.
in poor outcomes of phage therapy. In a separate study, Colom et al. [78] used a cocktail of three Sal-
To address the issue of human gastroenteritis caused by E. coli monella phage individually encapsulated in alginate microparticles
transmitted through farm animals, meat and poultry Abdulamir et al. containing CaCO3 to treat Salmonella infection in commercial broilers
[57] studied the effect of a cocktail of 140 lytic E. coli specific phage under farm-like conditions. Bacteriophage retention in the chicken
encapsulated using a hydroxypropyl methylcellulose (HPMC) based caecum was investigated by giving an oral dose of phage (1010 PFU/
enteric coating. The capsules (ZeinPharma Germany, GmbH, Germany) chicken) of either free phage (1:1:1 mixture of the three phages ϕ20,
encapsulating the 140 E. coli specific phage were administered orally in ϕ78 and ϕ87) or alginate/CaCO3 encapsulated phage cocktail to 63 1-
rats. A concentration of 107 PFU/ml phage suspended in an aqueous day old chickens. Caecum samples collected from euthanized animals
solution of sodium bicarbonate and an injectable master mix in lambda showed significantly higher numbers of chickens treated with en-
buffer were compared with HPMC capsules carrying the phage cocktail capsulated phage had phage in their caecum 48 h and 72 h later. Bac-
in lambda buffer. HPMC encapsulated phage were found to give better teriophage therapy against Salmonella was evaluated over 17 days in
results with a reduction in E. coli shedding of up to 4-log. chickens orally administered either the free phage cocktail or en-
Nale et al. [76] used a hamster model for the oral delivery of either capsulated phage cocktail (bacterial dose of day 0: 107 CFU/chicken,
single or a mixture of phage against Clostridium difficile. A hamster phage dose 1010 PFU/chicken). Oral doses of phage were administered
model of acute C. difficile infection (CDI) was used. The model reflected daily for 9 days (from day −1 to day 7 after Salmonella infection). Over
some of the clinical features of CDI including toxin-mediated diarrhoea the 15 days, a significant reduction (between 2 log and 3 log) in Sal-
and tissue pathology. Animals were given sodium bicarbonate orogas- monella concentration was noted for the group treated with en-
trically to reduce stomach acidity. Animals were infected with capsulated phage. At days 8, 10 and 15 the encapsulated phage treated
~ 103 CFU/hamster of spores and subsequently treated with group had lower Salmonella concentration compared with the free
~ 108 PFU/hamster of either a single phage or a cocktail of phage. The phage treated group showing better therapy outcomes using the en-
first dose was given at the time of infection with further doses every 8 h capsulated phage. The mucoadhesiveness of the microparticles and the
until the scheduled end-point of 36 h (toxin production in untreated slow sustained release kinetics were thought to be important factors in
animals is limited up to this time point in the cecum and colon thereby explaining these results. Similar concentration levels of phage were
allowing assessment of the effect of phage treatment on colonization in noted in the caecum for both encapsulated and free phage
the absence of toxin related symptoms). At this time point, animals (~ 103–104 PFU/g). Phage persisted in the caecum even after phage
were culled and enumeration of total bacterial load (spores and vege- treatment was stopped with higher phage concentrations noted for the
tative cells) in the cecum and colon was undertaken. Phage treatment free phage treated animals. Salmonella infection was not eradicated by
with certain combination of phage mixtures was shown to reduce either treatment. The authors reported that they did not find any phage
bacterial load in the cecum and colon however, results were variable. resistant bacteria in the phage treated groups. Hence the persistence of
This variability was attributed to the ratio of phage to bacteria present Salmonella infection was not attributed to phage resistant mutants.
at various locations in the cecum and colon during the initial phase of A different study using alginate microencapsulated Salmonella
bacterial outgrowth. In a separate experiment using the earlier protocol phage to treat pigs [59] showed that co-administration of Salmonella
108
with a phage cocktail resulted in reduction in pig intestinal Salmonella (using an atomising nozzle and exposing the birds to the aerosol in a
concentration (~ 2 log reduction). Yen et al. [58] recently reported closed chamber resulting in uptake by inhalation) of a cocktail con-
studies with mice and rabbits on the prophylactic effect of phage taining two bacteriophage (isolated against avian pathogenic E. coli) to
therapy to treat cholera. Oral administration of a cocktail of three Vibrio prevent an E. coli respiratory tract infection in broiler chickens. A severe
cholera lytic phage up to 24 h prior to infection with V. cholera reduced dose of 104 CFU/chicken (age 1 week) followed by a cocktail of phage
colonization and cholera-like diarrhoea. Complete elimination of bac- (a mixture of two phage at 107:108 PFU/chicken) resulted in reduction
teria was observed at short prophylaxis times (3 h), high phage dose in observed mortality (at age 3 weeks) compared with untreated con-
(106–107 PFU/mouse) and low bacterial infection dose (105 CFU/ trols. The dose of phage was found to be important with lower con-
mouse). Phage administration in negative controls (i.e. not inoculated centrations of aerosolized phage affording worse protection. The timing
with bacteria) showed a significant drop in phage titre (3 log reduction of phage administration was also evaluated with phage treatment im-
in mice) 12 h following phage treatment, indicating phage elimination mediately following E. coli infection showing positive efficacy results.
by the host. However, results were mixed when phage were administered prior to or
3 days following E. coli challenge. Bacteria were cultured from swabs of
3.5.4. Pulmonary phage delivery (animal studies) liver and air sacs but data on phage titres from the tissue samples were
Pulmonary delivery of phage has been achieved using a variety of not given.
methods including nasal instillation [26–28,42], aerosolization of li- Carmody et al. [42] investigated the therapeutic potential of bac-
quids using a nebulizer [90,91] and intraperitoneal injection [31,32]. teriophage in an in vivo mouse model of acute B. cenocepacia pul-
Carmody et al. [42] found intraperitoneal delivery of phage resulted in monary infection. Carmody et al. [42] compared intranasal inhalation
reduction in an acute lung Burkholderia cenocepacia infection in mice of bacteriophage with intraperitoneal injection of phage. 9 to 12 week
compared with nasal instillation whereas Oliveira et al. [92] found old mice were infected with 107 or 108 CFU B. cenocepacia (via tra-
intramuscular injection better than oral administration or using a spray. cheotomy). 24 h after infection, mice were treated with either in-
The dose of phage, concentration of bacteria and the timing of tranasal inhalation (spray details not given) or intraperitoneal injection
phage therapy were found to have a significant impact on phage of phage suspended in buffer (100 times higher dose of phage with
therapy outcomes. In most cases delivering high doses of phage im- respect to the bacterial dose was given). 48 h after phage treatment,
mediately or soon after inoculation of lungs with bacteria resulted in mice were euthanized and lung tissue samples were analysed for B.
better outcomes [26,28,32]. Singla et al. [39] compared treatment of cenocepacia and phage. Mice treated with phage delivered via in-
induced acute K. pneumonia infection in mice with free phage versus traperitoneal injection had significantly lower lung bacterial counts
liposome encapsulated phage and showed that high dose of en- compared with untreated control mice 48 h after treatment. The phage
capsulated phage gave better outcomes given up to 48 h prior to K. concentration in the lungs of mice was not statistically significant from
pneumonia infection. A number of studies reported in situ amplification mock infected mice with similar dose of intraperitoneal treated phage.
of P. aeruginosa and K. pneumonia phage [26,28,39]. Phages were ra- The intranasal phage treatment did not result in a statistically sig-
pidly cleared from the lungs of mice [28] hence high phage dose was nificant reduction in lung bacterial counts, however, phage had am-
needed for prophylactic treatment 24 h prior to bacteria inoculation plified in the lungs and were present at concentrations above those in
[26]. Effective phage therapy reduced the host inflammatory response mock infected mice with similar dose of inhaled phage. 24 h after in-
[26,28,39]. fection (but before phage treatment), bacteria were primarily found in
Phages for treating respiratory infections caused by P. aeruginosa, B. lung parenchyma in peribronchiolar and perivascular areas, occasion-
cenocepacia, E. coli and K. pneumonia are available. Efficacy of phage ally co-localised with alveolar macrophages. Relatively few bacteria
activity against clinical strains obtained from CF sputum and in rela- were found in the airway lumen. Bacteria were found in microcolonies
tively early stage biofilms has been shown [27,39]. Phages with lytic in lung parenchyma 72 h after infection (without phage treatment). The
activity have been shown to possess hydrolases that degrade bacterial concentration of phage in lungs of mice treated intranasally was 2 log
exopolysaccharides. Animal studies have shown that a number of phage higher in comparison with intraperitoneal treated mice. Intranasally
delivery routes are effective for phage delivery to treat acute lung in- administered phages were found co-localised with alveolar macro-
fections. However, further work needs to be done to show the efficacy phages 48 h after treatment. This suggested that phage delivered in-
of phage therapy in chronic infections against more mature biofilms tranasally were sequestered by lung macrophages and therefore may be
containing well established polymicrobial communities in both in vitro unavailable to kill bacteria found in the lung parenchyma. Inhaled
and in vivo systems. phage did replicate in bacteria as the titres were found to be higher
Oliveira et al. [92] evaluated the effect of the route of administra- compared with the control group. Phage may have been able to re-
tion (oral, direct spray on beaks and intramuscular injection) and the plicate in bacteria taken-up by the macrophages. The phage in the
concentration of phage in the dose (106, 107 and 108 PFU/chicken) in macrophages remained viable as determined by the high phage titres.
the dissemination of 3 lytic E. coli phage in the organs of chickens In intraperitoneally treated mice, phages were found in vascular and
(lungs, liver, duodenum and spleen). The focus of the study was on perivascular areas and alveolar septa where phages were found co-lo-
controlling avian respiratory infections. Administration of phage using calised with degraded bacteria. Results from the study indicated better
intramuscular injection was found to be the best means of ensuring access of phage delivered intraperitoneally via systemic circulation to
phage dissemination in all tissues (including lungs). In the work by bacteria than those delivered via pulmonary delivery. The site of
Oliveira et al. [92], aerosol spray administration (details regarding the greatest bacterial killing may have been within the lung parenchyma
spray generation were not given) allowed phage to be directly delivered rather than the airway lumen. Phage delivered via the airway may not
to the lungs. Sufficient details regarding the formulation of phage for have been able to effectively penetrate the respiratory epithelium
spray dose were not given, e.g. whether the spray droplet size dis- where bacteria may have translocated from the lumen to the lung in-
tribution was suitable to aid droplet distribution in the lower re- terstitium. The lower concentration of phage in the lung following in-
spiratory tract. The phage type, administration route and dose delivered traperitoneal delivery was attributed to a rapid decrease in bacterial
were all found to affect the distribution of phage in the animal tissues. density and subsequent clearance of phage thereafter in the absence of
Phages were rapidly cleared from the organs 10 h after administration. sufficient bacterial hosts.
Huff et al. [90] determined the efficacy of aerosol administration Semler et al. [91] found aerosol delivery of B. cenocepacia phage in a
109
mouse lung infection model to be an effective means of phage delivery aeruginosa mucoid strain isolated from a cystic fibrosis patient. Within
resulting in reduction in B. cenocepacia concentration over a three day 20 h post-infection, bacteria had multiplied in the lungs by 2-log
period post-infection. They also evaluated intraperitoneal delivery of B. compared with the initial dose (found in broncho-alveolar lavages). A
cenocepacia phage which resulted in higher concentrations of phage dose dependent curative effect was observed when phage (dose
delivered to the lung supporting the findings of Carmody et al. [42] that 106 PFU/mouse or 107 PFU/mouse was given intranasally 2 h post-in-
intraperitoneal delivery of phage allows phage to access the lung. There fection). 20 h post-infection bacteria were detected in macrophages,
was however a discrepancy between the results of Semler et al. [91] and alveolae and extracellular spaces of lungs from untreated animals.
Carmody et al. [42] with regard reduction in mouse lung bacteria Phage treatment resulted in lower concentrations of bacteria, high
concentrations for phage delivered using the intraperitoneal route. This concentrations of phage (suggesting in situ amplification) and low
may be explained by low concentrations of bacteria present in the lung concentration of markers of host immune response (cytokines and lac-
for experiments carried out by Semler et al. [91] which may have re- tate dehydrogenase) with histological examination of lung tissue sam-
sulted in poor in situ phage amplification. ples indicating less severe damage to lung tissues for the phage treated
Debarbieux et al. [26] evaluated pulmonary phage therapy using a mice. Uninfected mice given phage intranasally cleared these at the rate
fluorescently labelled P. aeruginosa strain (to monitor in real-time the of 0.5 log/day. Phage given 4 days prior to mice being given an in-
spatial and temporal development of infection) in an acute mouse lung fective dose of bacteria needed high preventive phage doses (108 PFU/
infection model. The infectious dose was 107 CFU/mouse. In preventive mouse) to achieve 100% survival.
experiments, phages were given intranasally 24 h before infection. Page The use of bacteriophage as a therapeutic strategy to fight K.
lysates were simply purified using caesium chloride ultracentrifugation pneumoniae infection has previously been tested; a murine model of
and then suspended in phosphate buffered saline. Untreated mice died infection could be cleared [32]. However, in this study (as in other
within 48 h of infection. Mice treated with a phage to bacterium ratio of animal studies reviewed above), clearance of infection only occurred
1:10 died within 5 days after infection. Mice treated with a higher when bacteriophage were administered immediately after K. pneumo-
phage to bacterium ratio 1:1 (80–100%) and 10:1 (100%) survived niae infection which limits the potential of phage therapy for clinical
until the end of the experiment (12 days post-infection). Subsequent use.
experiments were carried out using the higher phage dose i.e. phage to Singla et al. [31] used liposome encapsulation of the lytic K. pneu-
bacterium ratio of 10:1. Phages were administered 2 h post-infection. moniae bacteriophage (belonging to the Podoviridae family). They in-
Within 6 h post-infection differences were observed in the amount of vestigated treatment efficacy of liposome-encapsulated phage com-
luminescent bacteria in the lungs of the infected mice suggesting rapid pared with free phage, to treat established K. pneumoniae infection in a
killing by the phage. After 24 h untreated mice were either dead or murine pneumonia model. Delay in phage treatment administered via
showed high levels of luminescence whereas phage treated mice intraperitoneal injection (6 h, 24 h and 48 h) post-intranasal adminis-
showed no or weak spots of light with tissue cultures showing bacterial tration of K. pneumonia was investigated. Free phage administration 6 h
load of 108 CFU/ml in untreated mice versus 102 CFU/ml in phage post-administration of K. pneumonia was shown to result in complete
treated mice at 24 h post-infection. Phage concentrations in bronchiolar elimination of K. pneumonia from lung samples (this was not the case for
lavage fluid were lower in uninfected controls (106 PFU/ml) versus the 24 h (~ 2 log reduction in CFU noted) and 48 h (CFU levels similar
infected and phage treated mice (107 PFU/ml) indicating phage am- to untreated positive controls) treated animals) delay is phage treat-
plification within the infected lungs. Measurement of inflammatory ments. Phages were observed in the vascular and perivascular areas and
cytokines IL-6 and TNF-alpha indicated attenuation of the host in- along alveolar septa; similar results were reported by other researchers
flammatory response due to phage treatment. Delaying phage treat- [42]. Phages were observed in lungs for longer (up to ~3–4 days) for
ment by 2 h, 4 h and 6 h post-infection resulted in all mice surviving animals infected with K. pneumonia compared with negative controls.
after 24 h (treated with phage 2 h post-infection) but only 75% of mice Phage amplification was noted in infected animals treated with free
surviving following phage treatment 4 h and 6 h post-infection. At 72 h phage. Animals treated with liposome encapsulated phage 6 h (infec-
post-infection survival was as follows 100% (2 h delay), 75% (4 h tion cleared in 24 h post-treatment), 24 h (infection cleared in 24 h
delay) and 25% (6 h delay). Phages were seen to remain effective in post-treatment), 48 h (reduction in CFU) and 72 h (reduction in CFU)
mice 24 h following treatment. Infection of mice 24 h post-phage after K. pneumonia infection showed variable levels of efficacy of
treatment (intranasal administration of 108 phages/mouse: 107 CFU/ treatment. Earlier treatment was better than delaying treatment 48 h or
mouse) resulted in 100% of mice surviving 16 days post-infection 72 h. In the K. pneumonia mouse infection model, bacterial concentra-
whereas all untreated mice had died. Non-invasive bioluminescence tions in control animals (untreated) were seen to peak at day 3 and fell
data suggested that rapid bacterial replication kinetics was an im- significantly by day 5. Phage amplification in lung samples was noted
portant factor in arresting infection by early administration of phage for phage encapsulated liposome treated animals. Liposome entrapped
when the bacteria were also highly susceptible to phage infection. High phage persisted in lungs for 5 days post-treatment. Prophylactic phage
dose of phages administered early during the onset of infection rapidly treatment (3 h, 6 h and 24 h prior to intranasal K. pneumonia infection)
reduced bacterial numbers and the associated inflammatory response in was studied [31]. No significant therapeutic effect was observed for
the host. mice treated with free phage 24 h prior to infection. The 3 h and 6 h
Alemayehu et al. [27] compared phage therapy efficacy in a murine treatment resulted in absence of infection. For phage encapsulated in
model using both mucoid and non-mucoid P. aeruginosa strains isolated liposomes no infection developed when mice were treated 6 h, 24 h and
from CF patients. Lungs of 6- to 8-week old mice were infected in- 48 h prior to administration of infection suggesting encapsulation had a
tranasally with lux-tagged Pseudomonas (dose 106 CFU/mouse). 2 h beneficial prophylactic therapeutic effect due to retention of the phage
following infection a binary mixture of phage (podovirus and myovirus) at therapeutic concentrations in the lung.
was delivered intranasally at a dose of 107 PFU/mouse. Phages were Singla et al. [31] showed evidence of reduction in levels of pro-
effective in reducing considerably the amount of bacteria associated inflammatory cytokines IL-6 and TNF-α with free phage treatment (0 h
bioluminescence in the first 6 h period post-infection. and 6 h prior to intranasal bacterial inoculation) and with liposome
Morello et al. [28] evaluated the efficacy of phage treatment in a treated animals (24 h and 48 h prior to intranasal bacterial inocula-
mouse lung-infection model (a dose of 106 CFU/mouse resulted in an- tion). Debarbieux et al. [26] showed reduction in host inflammatory
imal death 2 days following infection) against a multidrug resistant P. response due to reduction in the number of bacteria upon phage
110
treatment. 3.7. Phage therapy case studies with humans and recent clinical trials
Although the majority of K. pneumoniae bacilli during infection are
thought to be extracellular, it is recognised that K. pneumoniae can There are relatively few recent published randomised control trial
survive within macrophages and epithelial cells [93,94]. Liposome studies on the therapeutic use of phage involving humans. Typically
encapsulation may enhance access of phage to intracellular K. pneu- these tend to be case studies [96] however, recently two small rando-
monia. Using an ex vivo model of cultured murine peritoneal macro- mised controlled trials have been undertaken [35,97]. The formulations
phages infected with K. pneumonia, Singla et al. [95] showed that caused for these trials tend to be simple phage lysates that have been
tionic liposome encapsulated phage were able to kill greater numbers purified using crude laboratory methods including centrifugation, mi-
(~ 95%) of intracellular K. pneumonia compared with free phage crofiltration, ultracentrifugation and in some instances endotoxin re-
(~ 20%). A potential limitation of phage therapy for treatment of in- moval through affinity chromatography [98]. Phage cocktails used for
fection is the development of phage-neutralising antibodies which may the trials were simple suspensions in buffer without microencapsula-
result in reduction of the concentration of phage at the site of infection tion. The titre of phage s used was typically low and reliant on in situ
or in systemic circulation [30]. Singla et al. [95] demonstrated using ex amplification which may have resulted in therapy failure in at least one
vivo experiments that liposome encapsulation was effective at com- of the studies [35].
pletely protecting K. pneumonia phage from the neutralising effect of Wright et al. [97] conducted a small (24 patients) randomised (two
murine antibodies, whereas free phages were inactivated by serum groups of 12), double blind Phase I/II clinical trial to evaluate the safety
antibodies raised against the phage in mice. Liposome encapsulated and efficacy of a therapeutic phage preparation against chronic otitis
phage were able to inactivate K. pneumonia following 3 h exposure to due to antibiotic resistant Pseudomonas aeruginosa infection. Each
mouse serum containing antibodies. 0.2 ml dose contained 105 PFU/ml each of six bacteriophages sus-
pended in 10% glycerol in phosphate-buffered saline and was applied to
the infected ear using a standard 1 ml syringe equipped with a 27 gauge
3.6. Phage therapy for chronic infections spinal needle. Patients re-visited the clinic on 7, 21 and 42 days fol-
lowing treatment. Patient inclusion criteria included checking that the
Results of animal studies (discussed above) relate to acute, rapidly P. aeruginosa strain causing infection was susceptible to at least one
progressive infections. In chronic infections, bacterial populations may phage in the therapeutic cocktail. Bacteriophage amplification (~ 2 log)
have a greater diversity of bacterial strains as polymicrobial colonies was noted for bacteriophages in human subjects under trial conditions
encased within biofilms with significant bacterial populations main- [97]. There were no reported side effects and no evidence of local or
tained over long periods of time, years even decades. Although a systemic toxicity due to phage therapy. Phage therapy was shown to
measureable specific immune response to the presence of the antigens result in a modest reduction (from the baseline) in median P. aeruginosa
expressed by the infecting bacteria may be present, however, on its counts. Results were not overwhelmingly conclusive. In cases where
own, the host may be unable to clear the infection. infection was treated phage counts decreased to below the limit of
Alemayehu et al. [27] investigated the ability of phage to penetrate detection. At day 42, 3 out of 12 patients in both the phage therapy
and kill Pseudomonas cells growing as biofilms on the surface of a layer group and the placebo group had levels of P. aeruginosa below the limit
of human lung cystic fibrosis bronchial epithelial cells. Phage applied to of detection. High levels of P. aeruginosa were present in patients
24 h old biofilms (bacteria concentration ~ 107 CFU/well; phage do- ~109 CFU/g.
se ~ 108 PFU/well) were able to increase phage numbers by replicating A recent randomised phage therapy trial to reduce E. coli associated
in situ whilst killing the resident Pseudomonas (2-log reduction in ob- diarrhoea in children (6–24 month old males) failed to improve diar-
served bioluminescence and 3- to 4-log reduction in cell concentration rhoea outcomes. This was attributed to low intestinal E. coli con-
after 24 h incubation of phage with cells). It took longer to clear centration resulting in lack of significant phage amplification in vivo
Pseudomonas grown in biofilms in comparison with planktonic cells in [35]. Half of the patients contained phage susceptible E. coli in stool
the lung model. Singla et al. [95] evaluated phage activity against samples however, the stool CFU titres were low (median titre 105 CFU/
young (4 day old) and more mature (7 day old) K. pneumonia biofilms. g stool). Phage-bacteria pharmacodynamics dictates that conditions
Both free phage and liposome encapsulated phage showed similar levels where there is a low in situ bacterial concentration are not conducive to
of antimicrobial activity (~ 2 log reduction in CFUs) in the 4 day old significant phage amplification. Phage cocktails used in the study were
biofilm model compared with the antibiotic amikacin. For 7 day old given orally without antacid and were not protected from the stomach
biofilms the effect was less marked (~ 1 log reduction). acidity by encapsulation hence phage titres reaching the site of infec-
Cocktails of phage covering multiple species and having the correct tion may have been low. Rapid elimination of the phage from the host
host range for different strains may be needed to facilitate clearance of may also have further lowered the efficacy of the treatment.
chronic infections. Examples of such chronic infections include infec- Fish et al. [96] recently published results of six case series where
tions in cystic fibrosis (P. aeruginosa) and tuberculosis (M. tuberculosis). bacteriophage treatment of intransigent diabetic toe ulcers was suc-
Having suitably diagnosed the polymicrobial constitution of the infec- cessful. The study used a commercially available (Eliava Bioprepara-
tion causing bacteria, tailored phage formulations would need to be tion, Tbilisi, Georgia) single lytic phage Sb-1 (broad host range against
prepared and administered. Phage banks with well characterised phage S. aureus) in sterile solution (10 ml per vial containing ~107–108 PFU/
lines with long shelf life would be needed in order to prepare such ml). Treatment involved standard wound care, including weekly soft
formulated phage cocktails and supplied for clinical use. The challenge tissue debridement if necessary [96]. Phage preparation was applied
here is delivery of precise concentrations of formulated phage cocktails topically by dripping the phage solution into the wound cavity; gauze
at the site of infection e.g. for cystic fibrosis, which may require the soaked with phage solution (0.1–0.5 ml depending on ulcer volume)
phage to penetrate biofilms to gain access to the bacteria. Furthermore was packed into the cavity and the wound was wrapped with dry gauze.
the phage would need to be phagocytosed into macrophages to target The actual dose given was variable depending on the size of the toe
M. tuberculosis cells present in granulomas which poses an additional ulcer. The patient was instructed to leave the gauze in place for 48 h.
challenge. The protocol was repeated weekly in a hospital-based wound clinic
setting until the ulcer became too small to pack. All six cases reported
healed with an average time to heal of ~6 weeks.
111
Merabishvili et al. [98] prepared a well-defined cocktail (designated 4. Bacteriophage encapsulation

BFC-1) of lytic phage (containing a mixture of three broad host range
lytic phages; two Myoviridae (against P. aeruginosa and S. aureus) and Developing formulations that incorporate bacteriophage for ther-
one Podoviridae (against P. aeruginosa)) designed for treatment of P. apeutic applications requires an appreciation of the chemical and
aeruginosa and S. aureus infections in burn wound patients as part of a physical stresses phage may encounter both during processing as well as
pilot human clinical study. The cocktail contained 109 PFU/ml of each during storage once formulated. Phage inactivation and long term re-
bacteriophage applied (application by spraying using a syringe with a duction in phage titre upon storage is highly undesirable. The physical
spray adapter fitted at the end) in doses of 1 ml/50 cm2 wound bed with and chemical properties of the formulation need careful consideration
the assumption that this would result in a ratio of phage to bacterium of when selecting a technique for encapsulation. Accurate loading of
100:1. The cocktail formulation consisted of endotoxin free phage phage per particle requires particle monodispersity which is rarely
(pyrogenicity evaluated using a rabbit animal model) suspended in achieved in practice however, some techniques are considerably better
physiological buffer and stored at 2–8 °C in 3 ml ‘single use only’ vials. than others (discussed below). The particle morphology should be
Stored at 4 °C, the phage suspension titre remained stable over without deformities and the particles shouldn't aggregate or un-
12 months. A safety trial on infected burn wounds of eight patients had controllably fuse together with material in the surrounding environ-
shown no adverse events following application. No efficacy data was ment. A polymer or lipid may be used to coat an existing structure
reported in the paper. However, a recent paper [99] presented results containing the phage, for example Murthy and Engelhardt [82] sprayed
from the application of BFC-1 cocktail on colonized burn wounds in a phage on dried skimmed milk and then encapsulated them in a lipid
small clinical trial (nine acute burn wound patients, 10 applications). coating. Alternatively, phage may be incorporated in the formulation
Punch biopsies were taken to determine bacterial species identification within the droplet which upon drying or crosslinking results in phage
and antibiotic susceptibility tests. 1 ml of the BFC-1 solution was ap- entrapment in the particle core. Depending on the technique used,
plied per 50 cm2 of burn wound area. The other half received standard polymer gelling may be part of the technique or it may be a process
treatment protocols. Patients also received systemic antibiotic treat- occurring downstream. There are many techniques and processes that
ment. Following BFC-1 treatment, the wound was covered with dres- may be used for stabilising, immobilising and encapsulating phage. The
sings, gauze and bandages according to standard treatment protocols. most common methods are spray-drying, spray freeze drying, freeze
2–5 h later, the burn wound was uncovered and two biopsies taken next drying, extrusion dripping methods, emulsion and polymerisation
to (within 2 cm) of the previous ones. Bacterial load determination of techniques. The literature focusing on phage encapsulation, formula-
the biopsies was undertaken. Despite initial indications of wound co- tion and storage stability is reviewed below and summary information
lonization or infection, bacterial cultures taken before and after treat- is presented in Tables 2–4.
ment showed low levels of bacteria (in 8/10 applications). In all cases
bacterial load remained unchanged after BFC-1 applications as well as 4.1. Formulation of bacteriophage for preservation and storage
after standard treatment. The low bacterial load in the wound was at-
tributed to topical and systemic antibiotic treatment of patients prior to Phages are protein structures and they are therefore susceptible to
enrolment of the patients in the study (which took 7 days from initial factors known to denature proteins; these include exposure to organic
assessment of colonization/infection with target pathogen). Rose et al. solvents [100,101], high temperatures [102], pH [102,103], ionic
[99] found that the sprayed BFC-1 solution had a tendency to run off strength [103], and interfacial effects. Additionally, mechanical stresses
the burn wound. Better formulation e.g. as a gel or part of a dressing during formulation or encapsulation including shear stresses during
(e.g. with controlled release properties) was suggested for future trials. mixing and agitation, atomisation during spraying [104] and, desicca-
tion stresses during drying [105], need careful consideration. Drying
3.8. Summary and lyophilization of bacteriophage with suitable carriers and bulking
agents to improve storage life have been the focus of numerous studies
For phage therapy to work, bacteriophage concentrations at the site (Table 2). A number of encapsulation techniques (reviewed further on)
of infection must be sufficiently high for significant in situ phage am- have been used, followed by drying [75] or incorporating solvent
plification to occur and as a result reduce the rate of replication of the evaporation (e.g. during electrospinning, [106]) to improve phage
infecting population of bacteria. The phages need to be delivered to the storage. One of the most common and successful modes of long-term
site of infection and they need to have access to replicating bacteria preservation of bacteriophage is storage at 4 °C in Trypticase Soy Agar
susceptible to phage infection. Even when susceptible bacterial con- and Brain Heart Infusion broth whereas storage at − 80 °C requires
centrations are high at the site of infection, delivery of low con- 50% glycerol as a cryoprotectant [107] or freeze drying with excipients
centrations of phage may not be sufficient to increase phage con- (e.g. sucrose or trehalose) as lyophilization and cryoprotectants
centration rapidly enough to achieve rapid killing of bacteria. Self- [108,109]. Ackermann [109] observed that phage titres for one of the
amplification will not occur when concentrations of bacteria are low largest collections of tailed phages typically tended to drop by 1 log
[8]. In vivo reduction in phage concentration due to the host immune over the course of a year but then remained fairly stable (although there
response or other mechanisms also needs consideration. Under these were many individual variations).
conditions, steady or frequent delivery of high concentrations of phage Clark [107] working for the American Type Culture Collection (A-
may be needed for successful phage therapy. Phage resistant popula- TCC) evaluated storage stability of a wide variety of bacteriophages (16
tions are highly likely and require use of a number of phage strains that in total) for long term preservation and distribution. Phage specimens
bind to different receptors thereby killing the mutant bacteria. Use of were treated and stored for two years at room temperature and at 4 °C
different phage strains for therapeutic use requires careful formulation as (i) broth lysates; (ii) lysates diluted with 50% glycerol; (iii) satur-
of phages to retain phage stability such that viable phages at a suffi- ating filter paper with lysate and then drying; (iv) freeze dried by
ciently high concentration are present at the time of use. Rational de- mixing lysates with an equal volume of double strength skim milk
sign of micro- and nanocapsules loaded with precise dose of en- (routine practice at that time at the ATCC). Phage titres measured im-
capsulated phage in stimuli responsive controlled release systems could mediately post-processing indicated that freeze drying resulted in a
be engineered to do this. significant loss of titre (between 1 log and 2 log reduction). However,
112
freeze drying did produce stable phage titres over the course of 2 years remaining adsorbed water is removed in a secondary drying phase. In
when stored refrigerated. After two years, the titre of phage in the broth the freeze-dried state, the rate constants of most chemical and physical
lysates were found to be generally higher than those of glycerol or degradation reactions are dramatically reduced thereby permitting long
freeze dried preparations. Thermally dried preparations generally did term storage under refrigerated or ambient temperatures (25 °C). Tra-
not prove satisfactory. Preparations stored at 4 °C showed higher titres ditional freeze drying is a slow process since the lyophilization is car-
than those kept at room temperature. All titres declined with time re- ried out at temperatures below the glass transition temperature of the
gardless of the conditions of preservation. maximally freeze-concentrated bacteriophage-excipient solution (Tg′)
and below the collapse temperature (Tc) of the sample thereby yielding
4.1.1. Effect of thermal and environmental stresses on phage stability a porous material with microstructure that is conducive for drying.
Phage may show different susceptibility to ingredients used in for- Freeze dried material typically results in a cake that needs further
mulations. Knezevic et al. [103] examined the effects of environmental processing e.g. milling to achieve fine particles suitable for loading in
factors on adsorption and inactivation of four P. aeruginosa phages (one dry powder inhalers (DPI) [112].
Podoviridae (designated δ) and three Siphoviridae phages (designated J- A brief summary of the main issues surrounding the freeze drying of
1, σ-1 and 001A). Factors investigated included temperature (range phage is now given. Papers on phage and virus freeze drying have
7–44 °C, exposure for 30 min in SM buffer), pH (pH 1.5–pH 9, at 37 °C mainly focused on evaluation of formulations in order to stabilise phage
for 30 min), phage neutralisation by exposure to carbohydrates and for storage [105,109,113–115]. A number of papers have shown that
amino acids (potential phage receptor constituents; incubation at 37 °C addition of amino acids (e.g. sodium glutamate [113]), peptides (e.g.
for 30 min in SM buffer containing either glucose, rhamnose, mannose, peptone [108]) and proteins (casein and lactoferrin [116]) to the for-
galactose, glucosamine, glutamine or alanine), exposure to exopoly- mulation improves phage viability upon freeze drying and following
saccharides and lipopolysaccharides, silver nitrate (e.g. for combination rehydration (Table 2). Literature suggests that disaccharides e.g. lactose
therapy with phage), povidone-iodine (for combination therapy) and [116], sucrose [105,114,115,117] and trehalose [77,104,105,114,118]
exposure to serum of rats. All phage were completely inactivated after improve phage survival during freezing as well as subsequent lyophi-
30 min exposure at pH 1.5. Maximum viability was observed at pH 7. lization. Osmotic damage is an important factor that reduces phage
Significant differences in susceptibility for the phage were observed at surviving the freeze drying process [108]. This includes hydration
pH 3, 5 and 9. Phages σ-1 and δ were found to be less susceptible to low stresses which may be ameliorated through addition of sugars to the
pH than J-1 and 001A. Survival at pH 9 was higher than at pH 5. Phage hydration medium [119], something often overlooked in recent studies
δ was considerably neutralised by glucose, rhamnose, glucosamine, that use simple buffer or saline for rehydrating freeze dried samples
mannose and alanine. None of the phage survived at working silver [105,114,120]. A number of papers have shown that the concentration
nitrate concentrations of 0.3%. Phages δ and 001A were highly sensi- of sucrose in the formulation affects phage viability during freeze
tive to povidone-iodine; exposure for 30 min to even the lowest con- drying and subsequent rehydration [105,114], although different
centration 0.5% completely inactivated them. For phages σ-1 and J-1 phages would appear to prefer either high or low concentrations. Rapid
the PhI50 (i.e. concentration to achieve a 50% reduction in titre) con- rates of freezing at relatively low temperatures (< −20 °C) have been
centration values were ~4%. found to result in better phage survival compared with slow freezing.
Briers et al. [102] evaluated the stability of the structural pepti- This has been attributed to having less time for osmotic damage to
doglycan hydrolase gp181 of the P. aeruginosa Myoviridae bacter- occur [108].
iophage ϕKZ to temperature (range 25–90 °C), ionic strength A detailed listing of literature studies is provided in Table 2 and
(20–320 mM) and pH (5–8). The protein is supposed to degrade the more detailed commentaries on some selected papers are now given as
peptidoglycan layer locally during the infection process. Gradual loss of follows.
enzymatic activity due to thermal inactivation was observed from 100% Engel et al. [113] investigated the storage stability of mycobacter-
activity at 25 °C down to 0% at 70 °C when exposed for 60 min and iophages using freeze drying. The effect of various cryoprotectants and
down to < 20% activity when exposed for 10 min to 90 °C. Optimal lyoprotectants at varying phage titre (104, 105 and 109 PFU/ml, for
enzymatic activity at 25 °C was observed at pH 6.2 and ionic strength phage BK1) was investigated including sodium glutamate (with and
140 mM with variation in pH and ionic strength in either direction without gelatine), gelatine, peptone (with and without sorbitol), dex-
resulting in loss of activity. tran (with glucose), calf serum and skim milk. 53 different myco-
bacteriophages were finally tested using sodium glutamate (5 wt% with
4.2. Freeze drying of bacteriophage for storage and encapsulation 0.5 wt% gelatine). Storage of freeze dried phage at room temperature in
the dark for as long as 2.5 years resulted in only a modest change in
Freeze drying (lyophilization) is routinely used in the pharmaceu- phage viability.
tical industry [110] to dry proteins, vaccines, peptides or colloidal Puapermpoonsiri et al. [117] carried out a qualitative study evalu-
carriers such as liposomes, nanoparticles and nanoemulsions [111]. ating the effect of sucrose, PEG and gelatine during freeze drying on
Freeze drying is typically done from solution and involves a freezing short term viability of S. aureus and P. aeruginosa bacteriophage over a
step followed by a drying step. During the freezing step, the phage 30 day period. Primary drying (freezing and sublimation of frozen
containing solution is cooled and ice crystals of pure water form re- water at − 30 °C for 1000 min) was followed by a secondary drying
sulting in concentration of the remaining liquid. This typically results in cycle (at an elevated temperature of 25 °C for 6 h) to desorb adsorbed
an increase in viscosity as well as osmolarity of the concentrated so- water. The residual moisture content (4–6 wt%) of the freeze dried
lution which inhibits further crystallization and eventually results in a powder was found to correlate with better phage viability; but due to
frozen amorphous/crystalline phase. Preventing aggregation of bac- the qualitative nature of this work, it is difficult to accurately assess the
teriophage and inactivation due to high osmotic pressure changes needs magnitude of this effect, which needs further exploration. Gelatine was
careful consideration here. At the end of the freezing stage, drying found to play no role in maintaining phage viability following lyophi-
begins as a result of switching on a very low temperature condenser lization. Control of the residual moisture content in the lyophilized
which draws water from the sample over time. A vacuum is also pulled material was found to be important in order to balance chemical (re-
to remove air which slows mass transfer to the condenser. A primary activity of amino acids) and physical stability (e.g. unfolding of tail
drying stage results in direct sublimation of the ice crystals, and proteins) of bacteriophage resulting in loss of lytic activity. High
113
concentrations of sucrose and PEG were needed to prevent collapse of remained stable (variability within ~1 log, measured at time period 3,
freeze dried cakes due to secondary drying at 25 °C. High concentra- 7, 12 and 27 months). Phage ISP stability was also recorded for phage
tions of additives (e.g. sucrose) lead to a decrease in phage titre with stored in LB broth (1 log reduction) and 0.9% NaCl (~ 2 log reduction)
better results at lower sucrose concentrations. After 30 days of storage over the same period (after 37 months) stored at 4 °C. Malenovská
at 4 °C viable phage were still present. Dini and Urraza [105] in- [115] also reported good results with sucrose in combination with ge-
vestigated the effect of buffer systems (PBS and SM), buffers with/ latine during lyophilization of structurally different animal viruses. It is
without skimmed milk and addition of disaccharides (sucrose or tre- noteworthy that a commercially available freeze dried phage cocktail
halose) on a Podovirus coliphage stability during freeze drying (at preparation contains sucrose and gelatine [121].
− 80 °C for 24 h) and storage. The coliphage freeze dried in SM buffer Golshahi et al. [116] determined the feasibility of preparing stable
gave the highest titre compared with skimmed milk and PBS respec- freeze dried formulations of B. cepacia and P. aeruginosa phage for re-
tively. Addition of trehalose or sucrose to PBS did not improve phage spiratory delivery. A 2 log reduction in phage titre was achieved fol-
titre outcomes. This may have been due to hydration stresses due to lowing freeze drying in a formulation of lactose/lactoferrin (60:40 w/
osmotic shock rather than the freeze drying process. Freeze dried w). Freeze dried phage titre remained stable over a 3 month storage
samples were rehydrated using a simple buffer rather than a buffer with period at 4 °C and 22 °C. Addition of lactoferrin resulted in powder that
added sucrose to prevent osmotic shock. Addition of sucrose to PBS did did not require milling to make it respirable (size < 5 μm) with
stabilise the phage titre upon storage for 120 days (at 4 °C) with phage ~30 wt% of powder available in the free particle fraction (FPF) upon
freeze dried in PBS alone showing a progressive decrease in phage titre aerosolization.
over 120 days of storage (~ 2.5 log loss compared with the initial lyo-
philized product). Addition of sucrose at low concentrations (0.1 M) in
SM buffer was found to result in a high phage titre post-freeze drying 4.2.1. Freeze drying of encapsulated phage
(only a ~0.5 log drop in titre attributed to the drying step rather than A number of papers have looked at freeze drying phage in-
freezing step) with the titre remaining stable for up to 120 days upon corporated in polymer matrices [77,100,118,121,122]. Pua-
storage. Increasing the sucrose concentration in SM buffer to 0.3 M or permpoonsiri et al. [100] lyophilized microencapsulated S. aureus and
0.5 M resulted in a ~0.5 log lower titre compared with the 0.1 M su- P. aeruginosa phage (dispersed in a salt buffer with gelatine) in PLGA
crose sample. Sucrose containing powders had a higher residual (poly(lactic-co-glycolic acid)). The microspheres were frozen in liquid
moisture content, compared with samples dried using PBS or skimmed nitrogen followed by lyophilization. All phage viability was lost after
milk. Only a small portion of the overall loss of phage titre (~ 0.3 log) 7 days of storage regardless of storage temperature. Loss of phage via-
was attributed to the freezing step (for the 0.5 M sucrose-SM buffer bility was attributed to the use of an organic solvent in the emulsifi-
sample) with lyophilization accounting for the majority of the ~1.5 log cation process (discussed later). Alfadhel et al. [123] used freeze drying
phage titre loss. Whether this could be attributed to the rehydration as part of the encapsulation process. S. aureus phage containing nasal
step needs further exploration. inserts were prepared by freeze drying (freezing at −30 °C and drying
One of the only papers to look at rehydration was Cox et al. [119] at 10 °C) phage formulations containing hydroxypropyl methylcellulose
who looked at loss of phage viability arising from detachment of tails (HPMC) as the encapsulating polymer. Freeze drying of phages for-
from capsid heads due to rapid rehydration following freeze drying. mulated in the polymer solution resulted in ~1 log drop in titre im-
Slow rehydration of freeze dried phage by exposure to a controlled mediately after lyophilization and between 3 and 4 log reduction over a
humidity environment (100% RH at ambient temperature) for 60 min year of storage at 4 °C. Addition of mannitol or increasing the HPMC
followed by rehydration using broth or sucrose containing broth re- concentration was not found to improve phage stability during storage.
sulted in significantly higher phage titre. The effect of hydration is one Dai et al. [118] undertook encapsulation of E. coli phage in water
area that definitely needs further exploration. soluble polyvinylpyrrolidone (PVP) nanofibers prepared using electro-
Merabishvili et al. [114] investigated the stability of a S. aureus spinning. Phage containing PVP solutions (with and without trehalose
myovirus phage ISP after freeze drying. Different formulations con- in SM buffer) were frozen in liquid nitrogen and then freeze dried
taining two different concentrations of phage titres (108 and 109 PFU/ (drying conditions not given). Immediately after drying the PVP and
ml) were prepared with different excipients (each excipient at either PVP-trehalose solutions showed no significant drop in phage titre.
0.1 M or 0.5 M concentration: sucrose, trehalose, mannitol, glycine, and However, after 8 weeks of storage at room temperature (humidity <
polyvinylpyrrolidone (PVP) and PEG 6000 at 1% and 5%). Primary 40%) the PVP sample showed a 3 log drop in titre whereas the PVP-
drying was carried out at − 30 °C whereas for secondary drying, the trehalose sample showed no significant titre drop. Korehei and Kadla
temperature was gradually increased to 25 °C over a period of ~ 10 h [106] encapsulated E. coli phage using emulsion electrospinning and
followed by isothermal drying. Some of the excipients (PVP, mannitol co-axial electrospinning. In emulsion electrospinning, the phages were
and PEG 6000) proved to be unsuccessful. PVP was found to inactivate pre-encapsulated in alginate nanoparticles which were suspended in
the phage even before freezing. Complete inactivation of phage ISP polyethylene oxide (PEO)/chloroform solution for electrospinning. In
after lyophilization was observed using glycine as excipient (glycine co-axial electrospinning the aqueous phage suspension remained in the
crystallizes during freezing). Complete loss of phage ISP activity using core of the fiber whereas the shell was made from polyethylene oxide
the sugar alcohol mannitol at 0.1 M was observed with a 4 log titre (PEO). The phage containing fibers were freeze dried (details not
reduction using 0.5 M concentration. The phage titre thereafter re- given). Fibers prepared using co-axial electrospinning were stored at
mained stable over a 27 month period with a 2 log reduction after 20 °C, 4 °C and − 20 °C for up to 4 weeks. High phage titre was ob-
37 months of storage. Phage ISP titre in PEG 6000 showed a 1.8 log (1% served immediately after freeze drying of fibers prepared using emul-
PEG) and 5 log (5% PEG) reduction following lyophilization. After sion (106 PFU/ml) and co-axial (108 PFU/ml) electrospinning. Over
37 months of storage, further loss of titre was 3 log (1% PEG) and 4 weeks of storage the co-axial fibers showed no loss of phage titre
1.7 log (5% PEG). High concentrations (0.5 M) of trehalose and sucrose when stored at − 20 °C and 4 °C whereas a significant drop in tire
were found to be the most effective stabilisers for phage ISP. A 1 log loss (> 6 log reduction) was observed for the samples stored at 20 °C.
in phage titre was observed immediately after lyophilization with a Recently Colom et al. [77] showed promising freeze drying results
further 1 log loss following 37 months of storage. Different concentra- for trehalose containing liposomes with encapsulated S. enterica phage.
tions of trehalose and sucrose (0.3 M, 0.5 M, 0.8 M and 1.0 M) were Lyophilization resulted in a modest drop (< 1 log) in phage titre for the
studied further. Phage ISP freeze dried from 0.8 M and 1.0 M sugar liposome encapsulated phages. Trehalose containing liposomes showed
solutions showed the smallest titre reduction immediately after drying no loss of phage titre when stored at 4 °C in 10 mM MgSO4 buffer
(~ 0.5 log). Over the next 27 months of storage phage ISP titre (pH 6.1) for a period of 3 months.
114
Table 2
Summary of studies looking at effects of formulation, drying and storage on phage viability.
Result / PFU Log reduction

Host Bacteria Formulation Excipients Method
good moderate poor / value
Storage conditions
( ) -encapsulated brown/blue – low/high relative
Proteins. (casein, whey etc.)

humidity
Amino acids or peptides

Mycobacteriophage s
Drying on filter paper
ambient temperature
Polyvinylpyrrolidone
Spray freeze drying
Storage duration,
refrigerated ~4 C
(LT – long term)
Stored as liquid
Before Storage
Freeze drying,
P. aeruginosa
Broth / lysate;
temperatures
Spray drying
S. eneterica
C. ulcerans
V. cholerae
B. cepacia
Trehalose
S. aureus
Air drying
sub-zero
Mannitol
Gelatine
Glycerol
Sucrose
Lactose
Leucine
months
Others
Others
~25 C
Saline
E. coli
Buffer
Water
Ref.
[107] Ec M Pa Sa 12 B L 24
B Gl L
B P FD
B P FP
[104] Pa W T M Le SF NA
W T M Le SD
[113] M B Ge A FD 30
B P FD
B O FD
[108] Cu B A FD 3
B A S FD
[119] Ec B S FD NA
[109] Ec Pa Sa Vc B L NA LT
>400
B Gl L
[105] Ec Bu FD 4
Bu S FD
[114] Sa B S FD 37
B T FD
B PV
B A FD
Sa L
B L
B M FD
B O FD
[123] Sa Bu Ge M (FD) 12 -
Bu Ge M O (FD) -
[118] Ec Bu T PV FD 2
W T PV FD
Bu PV FD
W PV FD
[106] Ec Bu (FD) 1
[120] Pa W T Le SD 12 /
M Le SD /
[100] Pa Sa Bu (FD) 0.25
[117] Pa Sa Bu S FD 1
Bu FD
Bu G FD
Bu O FD
[75] Ec Bu (SD) 12
[128] Pa Sa Bu La (SD) - N/A
Bu T (SD) -
Bu O (SD) -
[132] -
Pa Sa Bu T SD N/A 12
-
[77] Se W T (FD) 3
[131] Bc Pa Bu P T Le O SD 3
[116] Bc Pa W P La FD 3
[157] Sa Bu P O (AD) - 0.5
Sa Bu P O (L) 1.5
Intensity of shading differentiates vertical sections of the table.
115
4.2.2. Spray freeze drying of bacteriophage [120] with high loss of phage titre observed at higher temperatures
Spray freeze drying may offer the possibility of a single step process [104,128,130,131]. Literature on spray drying of phage suspensions
to produce bacteriophage loaded porous powders with controlled par- typically includes excipients in the formulation to protect the phage
ticle-size distribution without subjecting the phage to high thermal from desiccation and thermal stress (Table 2). Upon spray drying, sugar
stresses encountered during conventional spray drying (discussed excipients form amorphous structures with high glass transition tem-
below). peratures (Tg), e.g. in their anhydrous amorphous form Tg of trehalose
Leung et al. [104] recently spray freeze dried (using an ultrasonic is 115 °C and Tg for lactose is 108 °C. Below the glass transition tem-
nozzle to atomise the solution) a P. aeruginosa podovirus (PEV2) by perature, vitrification due to the excipient stabilises the encapsulated
spraying the phage solution (formulations containing different con- bioactive agent and permits storage over long periods at ambient or
centrations of trehalose, leucine and mannitol in combination) directly lower temperatures. Adsorption of moisture results in lowering of the
into a dewar of liquid nitrogen; the frozen droplets were subsequently Tg and compounds that are hygroscopic e.g. lactose require handling
lyophilized (primary drying at −30 °C for 24 h and secondary drying at under controlled atmospheric conditions (low relative humidity) [131].
25 °C for 48 h) in a conventional freeze dryer. The ultrasonic nozzle Refrigerated storage of spray dried powders at relative humidity ex-
employed high frequency vibration (48 kHz) to produce fine droplets; a ceeding 20% has been shown to result in loss of phage titre [120,132]
significant 2 log reduction in phage titre was noted just due to the ul- whereas samples stored at low humidity were stable. Trehalose is the
trasonic spraying process itself (without freeze drying). The sprayed most frequently reported excipient used in the spray drying of phage
droplets froze immediately upon immersion in liquid nitrogen. Sub- and it has been shown to result in spray dried powders with high phage
sequent lyophilization resulted in highly porous dry powder (mean titre and good storage stability [104,120,131,132]. Trehalose has low
geometric particle diameter ~ 30 μm corresponding to the droplet size toxicity and has been shown to protect biological material including
generated by the ultrasonic nozzle). Particles with high trehalose con- proteins, probiotics and vaccines, against desiccation and thermal
tent (60%) were found to be highly hygroscopic and collapsed during stress. Crowe et al. [133] suggested that the efficacy of trehalose as an
storage whilst particles with lower trehalose content (40%) retained excipient in drying is partly due to its high glass transition temperature
their porous structure (examined using SEM). The higher trehalose at all water contents but also because trehalose binds residual water left
containing particles adsorbed more water at lower relative humidity over from the drying process to form a dihydrate which might other-
values. Leung et al. [104] suggested handling and storage of powders wise participate in lowering the glass transition temperature to below
below 20% RH to avoid moisture related recrystallization of excipients ambient [132]. Other excipients that have been used to improve the
damaging encapsulated phage. The loss in phage titre post-atomisation dispersibility of spray dried phage containing powders include dextrans
due to spray freeze drying was negligible; this suggested that phage [128], lactose (common excipient used for DPI powders; [112,128]),
activity was preserved by the excipients during the freeze drying step as glucose, sucrose, mannitol [104] and, leucine [104,131] (Table 2).
well as the rehydration step. The high phage titre in the spray freeze Reducing sugars such as lactose may be unsuitable excipients for bac-
dried powders was attributed to rapid freezing of the droplets in liquid teriophage [128]. Alternatives such as mannitol have been suggested in
nitrogen resulting in good distribution of the phage within the frozen the literature [134]. Phage spray drying studies using mannitol have
particles. Spray freeze dried powder with the higher trehalose content reported improved particle dispersion characteristics however, phage
also had the higher phage titre. The aerosol performance (using a viability results have been mixed [104,120]. Addition of proteins e.g.
commercial DPI device, an Osmohaler™) of the freeze dried powders casein in combination with trehalose have shown good results [131].
showed a reasonable fine particle fraction (size < 5 μm having 44% Refrigerated storage of spray dried powders was reported to yield
phage dose ~ 104–105 PFU dose delivered) with high total recovery of higher titres of phage during long term storage compared with storage
aerosolized viable phage dose (88%–97%). under ambient conditions [120]. Different phage strains formulated and
spray dried under identical conditions showed significant differences in
4.3. Spray drying of bacteriophage for storage and encapsulation the resulting phage titre suggesting the need for individually for-
mulating each phage to be used in a phage cocktail [128,132]. A
Spray drying processes atomise a liquid containing dissolved solids, number of studies have reported spray dried phage powders to have a
converting it into a fine mist which is contacted with a hot dry gas suitable mass median aerosolized diameter for pulmonary delivery of
(typically hot dry air) inside a drying chamber. Due to the high surface phage to treat respiratory infections [104,128,131,132,135].
area to volume ratio of the very small atomised droplets, the solvent Stanford et al. [75] took a cocktail of 5 E. coli phage (as lysates)
rapidly evaporates with each droplet forming a particle comprising the which were added to an aqueous pH responsive methacrylate polymer
non-volatile components (bacteriophage and excipients) along with a (Eudragit S100) containing an unspecified stabiliser, then spray dried
small amount of residual moisture. The particles are separated from the (co-current spray dryer fitted with a rotary atomiser) into a powder
airstream exiting the dryer, and cyclones or bag filters are typically (with 4% moisture content). No drying conditions were reported in the
used for this. Spray drying is a scalable industrial process technology paper. The resulting encapsulated phages (described as Ephage) were
and is widely used to produce fine powders for pharmaceutical appli- stored at 20 °C. Spray drying of the aqueous phage-polymer solution
cations including pulmonary delivery via dry powder inhalers (DPI) resulted in a 1 log reduction in phage titre post-drying. Encapsulated
[124]. Dry powders are favoured for respiratory drug delivery because phages were shown to survive 20 min exposure to an aqueous solution
they show relatively long storage stability without requiring refrigera- at pH 3. Storage at 20 °C for one year resulted in a decline in phage titre
tion [125,126] and DPI are simple to use and do not require regular of between 1 log and 3 log depending on the phage. Walbeck [130] in a
cleaning and disinfection [127]. recent patent (assigned to Omnilytics, USA) reported a 7 log loss in
Phages have been shown not to be able to withstand high shear phage viability post-spray drying using E. coli O157:H7 specific phage
stress for too long. The nebulization process has been shown to result in (the phage suspensions were unprocessed crude lysates including
loss of phage titre [104,128]. Phages are also sensitive to thermal phage, lysed bacteria and fermentation media, no excipients were
stresses and have been shown to partially lose activity at temperatures added, a rotary atomiser was used to generate the spray). The loss in
higher than 60 °C [108,129]. Low spray drying temperatures (~ 40 °C phage viability may be attributed to the high air inlet temperature
outlet air temperature) have been shown to result in higher phage titre (range 200–220 °C) used which resulted in the dry phage powders being
116
exposed to high air outlet temperatures (90–130 °C). et al. [104] used a two fluid nozzle for liquid atomisation which itself
Vandenheuvel et al. [128] spray dried S. aureus and P. aeruginosa resulted in a ~1 log loss in phage viability. The air inlet temperature
phages using lactose, trehalose and dextran as excipients. Dextran was was 60 °C which resulted in low outlet temperatures of 40–45 °C. The
used not only for its effect as an excipient in the spray drying process; in low outlet temperature was chosen to minimize thermal stress for the
the treatment of respiratory infections it may prevent P. aeruginosa from phage. However, titres were lower (due to the drying process) than
attaching to lung epithelial cells and may increase CF sputum clearance. recorded for spray freeze drying process reported using the same phage
The effect of atomisation air flow rate and drying air inlet temperatures and excipients in the same paper. Spray dried powders were collected
were investigated (using inlet air temperatures of either 85 °C or inside a relative humidity controlled chamber (RH < 20%) and stored
100 °C). At a relatively low inlet air temperature of 85 °C, the P. aeru- at 4 °C. Spray drying resulted in a further ~1.5 log drop in phage titre.
ginosa podovirus was found to be more resistant to spray drying stresses Leung et al. [120] further investigated the effect of humidity (0%, 22%
in comparison with a S. aureus myovirus. Lactose and dextran were and 60% RH) on long term storage (over 12 months) of spray dried
found to be unsuitable excipients, whereas phage viability results with Pseudomonas aeruginosa phage (at 4 °C) dry powders containing dif-
trehalose (4% wt) were more promising with < 0.5 log reduction for ferent amounts of trehalose (40%, 60% and 80%). Storage in a low
the podovirus and a 2 log reduction for the myovirus. The effect of air humidity environment (0% and 22%) was found to result in no loss of
inlet temperature was investigated further using trehalose only. The phage titre. High humidity (60% RH) resulted in complete loss of viable
viability of the myovirus was found to be strongly affected by the higher phage after 3 months.
air temperature. At an air inlet temperature of 100 °C a ~5 log reduc- Matinkhoo et al. [131] spray dried Myoviridae bacteriophage for P.
tion in phage titre was observed in comparison with a ~3 log reduction aeruginosa (phiKZ and D3) and for B. cepacia (KS4-M and KS14) using
at 85 °C using the same air inlet flow rate. The mean particle size (D50) trehalose, leucine and casein (a mixture of four similar phosphoproteins
was between 2 and 6 μm. Increasing atomisation air flow (which re- with MW between 19 kDa and 24 kDa) as excipients. Similar to Leung
duced particle size and would have increased atomisation stresses had a et al. [120], leucine was added to improve the dispersibility of the
mildly adverse effect on phage tire). This paper showed that two dif- powder. Surfactants were added to some formulations to aid in the
ferent types of bacteriophage (a myovirus and podovirus) processed dispersion of the bacteriophage in the feed solution [136]. A vibrating
under the same conditions show significant differences in stability to mesh atomiser was used resulting in atomised droplet diameters around
spray drying stresses and result in significantly different loss of titres. 7 μm. Mild spray drying conditions were used to provide a low outlet
The same group then assessed the storage stability (over 12 months) temperature of between 40 °C–45 °C (air inlet temperature of 75 °C) and
of the spray dried S. aureus (myovirus) and P. aeruginosa (podovirus) an outlet relative humidity of < 7% RH. The spray dryer was housed in
phage using trehalose as excipients [132]. They principally investigated a controlled humidity chamber with glove ports to allow collection of
the effect of relative humidity (RH 0% and RH 54%) and temperature and storage of powder without exposure to moisture. In vitro evaluation
(0 °C and 25 °C). Within 3 months of storage, the S. aureus myovirus of the deposition of the phage loaded spray dried powders was carried
titre fell by 2 log (4 °C, 0% RH) and 4 log (4 °C, 54% RH) respectively out using an experimental idealized throat and lung system (Alberta
but then stabilised thereafter. At 25 °C almost complete loss of phage Idealized Throat attached to an Anderson cascade impactor) using a
viability was observed (7 log) for both 0% RH and 54% RH samples commercially available DPI. Four formulations were tested: (i) Leucine
after 4 months of storage with a 2 log loss observed after the first (77 wt%):trehalose (19 wt%) (LT); (ii) leucine (76 wt%):trehalose
month. The podovirus was more stable with < 1 log reduction over the (19 wt%):tyloxapol (2 wt%) (LTX); (iii) leucine (76 wt%):trehalose
12 months for the 4 °C sample irrespective of the RH, and between 1 log (19 wt%):pluronic (2 wt%) (LTP); (iv) leucine (76 wt%):trehalose
and 2 log reduction at 25 °C. Powder X-ray diffraction analysis (phage (19 wt%):casein (2 wt%) (LTC). All powders had a well dispersed ap-
free powders dried under identical conditions to the phage containing pearance. Small batches of material were produced for each formula-
powders) showed that spray dried trehalose powder was non-crystal- tion; the manufacturing yield of the surfactant containing formulations
line. Differential scanning calorimetry data showed that the spray dried was significantly higher (~ 80 wt%) compared with LT and LTC
trehalose had a Tg between 116 °C and 117 °C i.e. the trehalose was in (~ 60 wt%). Between 20 and 30 wt% of aerosolized dose (LTX, LTP and
an amorphous state suitable for phage vitrification. After 8 months of LTC) did not reach the lung and was entrained in the mouth-throat
storage at 54% RH, the presence of an endothermic peak between 86 fraction. In all cases the total lung fraction exceeded 50 wt%. All
and 91 °C on DSC scans indicated the presence of stable trehalose di- Myoviridae spray dried phages showed a titre loss of < 1 log. The for-
hydrate crystals. Spray dried trehalose powder samples stored at 0% RH mulation containing trehalose, leucine and the milk protein casein was
did not show changes in DSC endotherms or XRD spectra. Trehalose is found to be the best formulation. All the spray dried powder formula-
known to protect bioactive agents from thermal and dehydration tions prepared by Matinkhoo et al. (2011) delivered > 50 wt% of dose
stresses through structural and conformational stabilisation of proteins in the correct (< 5 μm) size range using the DPI. The LTC formulation
by direct hydrogen bonding and the process of vitrification. Over time, was found to most consistently deliver single phage doses and cocktails
at 25 °C, phage titre dropped even where no crystallization of the of phage.
amorphous trehalose was observed. A number of in vitro studies have evaluated the use of nebulizers for
Leung et al. [104] spray dried a P. aeruginosa phage (Podoviridae) aerosolization of liquid phage formulations for pulmonary delivery.
designated PEV2 focusing on making phage containing powder for Sahota et al. [137] carried out an in vitro study delivering an aerosol
pulmonary delivery. The lytic podovirus (supplied by AmpliPhi Bios- containing two different Pseudomonas aeruginosa bacteriophages
ciences, Australia) was used at a titre of 107 PFU/ml suspended in dilute PELP20 and PELI40 (isolated from the commercially available
(2 wt% solids) aqueous mixtures of trehalose, mannitol and leucine Pyophage and Intestiphage preparations from the Eliava Institute,
(respective ratios either 60:20:20 or 40:40:20). The formulations were Tbilisi, Georgia). A commercial nebulizer was evaluated to deliver the
designed for use in dry powder inhalers (DPI), as leucine helps improve aerosolized phage dose into a cascade impactor (phage in the droplet
the dispersibility of the powders by reducing cohesive interactions be- size range < 5 μm were deemed suitable for respiratory delivery); 12%
tween particles. Mixing phage with excipients in solution resulted in a of the nominal dose was aerosolized with < 0.8% of dose found to be in
~ 0.5 log titre reduction, which was attributed to changes in ionic the correct size range for delivery to the lung. Golshahi et al. [138]
strength when the phage buffer was diluted in excipient solution. Leung undertook a study where they employed two commercial nebulizers (a
117
jet nebulizer (Pari LC star) and a vibrating mesh nebulizer (Pari eFlow)) removal (Table 3). The dispersed phase usually comprises of the core
and estimated that around 50% of nebulized Burkholderia cepacia phage material and solvent carrying the active agent (bacteriophage) and the
dose was in the droplet range (measured using Phase Doppler Anemo- polymer and a second phase which allows for the breakup of the inner
metry) that would result in deposition (estimated using mathematical phase into droplets. The emulsion may be water-in-oil (W/O) more
modelling) in either the tracheobronchial or alveolar region of the lung. typical for phage encapsulation [100], but could also be oil-in-water
(O/W) [142] and in some cases a third phase may also be present. Once
4.4. Drying bacteriophage on filter paper the droplets have formed, the solvent carrying the core material may be
removed, which leaves behind solid particles containing the active
A couple of early studies looked at stabilising phage by drying phage agent in the core. Different mechanical emulsification techniques may
suspensions by filtering onto filter paper and then drying the filter be used to break-up the inner phase into droplets, such as high-pressure
paper [107,139]. Prouty [139] investigated the long term storage of homogenisation, rotor-stator homogenisation, and ultrasonication.
bacteriophage of lactic acid streptococci by saturating filter paper with When two immiscible liquids are put together, shearing force breaks up
whey filtrate (pH 4.5–4.7) containing high concentration of phage. the inner phase into small droplets. The droplet size uniformity how-
Phage titre was not reported however, phages were found to remain ever, is low as the shear applied does not remain uniform throughout
viable for up to 78 months stored at ambient temperature. Further work the container resulting in droplet heterogeneity with some droplets
is needed to look at stabilising phage in porous substrates. being larger and some smaller. More advanced methods to produce
uniform droplets based on microfluidic and membrane based techni-
4.5. Key points regarding drying of bacteriophage formulations ques have thus far not been used for phage encapsulation (these
methods are discussed at the end of the paper).
Phage drying and encapsulation can be performed using spray Alternatively, phage containing polymer droplets have been pro-
drying, freeze drying and spray freeze drying techniques. Freeze drying duced via the extrusion technique [143]. The droplet may form in the
appears to be more successful that spray drying at maintaining high air or into another liquid. The dispersed phase containing the polymer
titre values, although the ability of spray drying (or spray freeze drying) and phage is extruded through a needle with a specific nozzle diameter
provides very small particles thereby allowing them to be used in DPI determining the size of the droplet [144]. Atomisation nozzles have also
for pulmonary delivery, without the need for further milling. The best been used to generate smaller droplets [78]. If the polymer requires a
conditions for the spray drying of phage have tended to be with low gelling or crosslinking agent then the droplet may be collected in a bath
outlet air temperatures (40–60 °C) which are much lower than typically of the gelling agent [143,145,146]. Examples of this method are pre-
used in the spray drying of biological materials (70–100 °C). Some of sented by the work of Ma et al. [143,145] and Dini et al. [122] where
the viability loss in spray drying may be due to the atomisation process. the extrusion technique was employed to extrude alginate containing
Survival rates during subsequent storage additionally depend on tem- phage droplets into a bath of calcium chloride which caused ionotropic
perature and humidity during storage and are important considerations gelation. The process of gelation can occur in a number of ways de-
for the long term storage of phage. A review of published work on pending on the properties of the polymer/hydrogel. Other gelling
drying of phage clearly indicates that phage survival varies from phage triggers apart from ionotropic gelation include, heating, cooling and
to phage and is highly dependent on drying conditions and formulation covalent crosslinking. Other techniques that have been used for phage
parameters, which need to be optimised for each phage (although for- encapsulation in synthetic and natural polymers include polymer pre-
mulations involving disaccharides tend to be most successful). This cipitation [147], photopolymerisation [148] and thermal phase inver-
presents a considerable challenge for phage therapy since cocktails of sion [142,149].
phage are typically needed. These need to be formulated for storage and Most papers on phage encapsulation in liposomes have used the
the active titre of individual phage may be highly variable if all phage thin-film hydration method [31,39,77,95,150]. Phage have been en-
are formulated together. capsulated in nanofibers using electrospinning techniques
[101,106,118,151–153], in films e.g. for wound healing and food bio-
4.6. Methods used for bacteriophage encapsulation in micro- and control of pathogens [154,155] and in thin film structures [148].
nanoparticles
Bacteriophages may be encapsulated in protective micro- and na- 4.7. Polymers used for bacteriophage encapsulation
noparticles to overcome adverse storage and physiological conditions
en route to delivering the phage load at the site of infection (e.g. lungs, Studies on phage encapsulation have used a variety of hydrophilic
intestinal delivery). Controlled release and sustained release strategies and hydrophobic polymers including agarose [148], alginate
for phage delivery applications may be achieved using a diverse array [78,122,143–145,149,155,157,158], chitosan [143,156,158], pectin
of strategies. These include systems based on diffusion controlled re- [122], whey protein [149,154,157,159], gelled milk protein [149],
lease (e.g. solvent diffusion based osmotic pumps), matrix dissolution hyaluronic acid methacrylate [148], hydroxypropyl methyl cellulose
and erosion-controlled systems, ion exchange swelling based systems. (HPMC) [57,123], poly(N-isopropylacrylamide) [147], Poly(DL-lacti-
Phage compatible formulation and encapsulation processes need to be de:glycolide) [100], polyesteramide [121,160], polyvinyl pyrrolidone
carefully designed to prevent damage to viral capsid and DNA/RNA [101,118], polyethylene oxide/polyvinyl alcohol [106,151–153], cel-
components and stabilisation of the structure of viral capsid and tail lulose diacetate [153], polymethyl methacrylate [75,161]. Phages have
proteins to prevent loss of phage viability during manufacturing op- been encapsulated in different morphologies including nanospheres,
erations. It is important that the carrier encapsulating the phage is able microspheres, nanofibers, microfibers, membranes and thin film struc-
to withstand adverse environmental conditions for the duration of ex- tures (Table 3). Triggers for phage release include polymer solvation,
posure and is capable of delivering the bacteriophage to the site of polymer dissolution and erosion [153], polymer hydrolysis [100,162],
infection upon arrival [140,141]. phase inversion induced by temperature [147], and pH triggered dis-
A number of phage encapsulation studies using synthetic and nat- solution of polymer and enzyme driven polymer degradation [148].
ural polymers have utilised the emulsification route followed by solvent
118
Table 3
Summary of bacteriophage microencapsulation literature in polymer carriers.
Host Organism Carrier system Encapsulation method Additives Target Release Dimensions Structure Key Highlights
application studies
hydroxypropyl water-in-oil emulsion polysaccharides food simulated sphere acid exposure (AE); cascade
methylcellulose (W/O); water-in-oil-in- (PS); contamination gastric fluid fibre impaction study (CI); storage
(HPMC); hyaluronic water double emulsion sucrose (Suc); (FC); (SGF); film studies (SS); nasal inserts (NI);
acid methacrylate other sugars (S); nasal carriage clinical case studies (CCS);
(W/O/W); extrusion (E);
Salmonella Typhimurium
(HYMA); Poly(DL- chloroform as of infectious intestinal thermally responsive polymer

lactide:glycolide) gelation (G); solvent (Chl); organism fluid (SIF); (TR); edible coating (EC);
(PLGA); Polyvinyl electrospinning (ES); wet others (O); (NC); exposure to degrading enzymes
L. monocytogenes
alcohol (PVA); spinning (WS); solvent aqueous gastrointestinal phosphate (EEnz ); exposure to bile salts
Staphylococcus
Streptococcus
P. aeruginosa
Polyvinyl evaporation (SW); solvent (Aq) / respiratory / buffer saline (BS); animal studies (AS);
pyrrolidone (PVP); wound (PBS); exposure to bile salts (BS);
S. enterica
precipitation
S. aureus
cellulose diacetate infections

Proteus
L. lactis
polymerisation (PP); liquid

E. coli
(CDA); polyethylene (G / R / W); +/- indicates phage protection /

phage filled capsules
Ref
oxide (PEO); whey damage

protein (WP) (LPFC)
[100] Pa Sa PLGA; PVA W/O/W – SE - R Buffer ~ 10 µm CI; SS

[123] Sa HPMC Freeze drying S+O NC Sterile N/A NI; SS;
[155] Ec L Alginate E+G - FC N/A N/A SS;
[160] Ec P S Polyamide Casting Suc W PBS N/A SS;
[121] P Pa S Polyamide Casting Suc + O W N/A ~ 100 µm CCS
[147] Se Polyamide PP - W Water ~500 nm TR
[154] Ec WP Imbibition O FC SGF/SIF ~100 µm EC; SS;
[158] Ec Chitosan + alginate E+G - G SGF/SIF N/A AE(+); EEnz (+); BS (+)
[149] Ll Skim milk + rennet W/O + G - G SGF/SIF ~ 100 µm AE(+); EEnz (+);
[149] Ll Alginate E+G - G SGF/SIF ~ mm AE(-); EEnz(-);
[149] Ll Alginate + WP E+G - G SGF/SIF ~ mm AE(+); EEnz(+);
[159] Sa Alginate + WP E+G Suc; S; PS; O G - < 1 mm AE (+); SS;
[145] Sa Alginate E+G CaCO3 G SGF/SIF < 1mm AE(+); EEnz(+); BS (+);
[122] Ec Alginate; Pectin E+G - G SGF/SIF ~ mm AE (+); EEnz (+)
[143] Se Chitosan + alginate E+G - G SGF/SIF < 1mm AE(+); EEnz (+); BS (+); SS;
[78] Se Alginate E+G CaCO3 G SGF/SIF ~100 µm AE (+); EEnz (+); BS (+); SS; AS
[59] Se Alginate E+G Other G - - AS;
[75] Ec Eudragit S100 Spray drying - G SGF - AS; SS;
[57] Ec HPMC LPFC - G - - AS
[157] Se Alginate + WP E+G PS G SGF/SIF < 1 µm AE (+); EEnz (+); BS (+); SS;
[101] Ec PVP ES / WS Aq - Buffer Ø~100nm/ ~10µm
[152] Ec PVA ES Aq - Broth Ø ~500 nm SS;
[151] Ec PVA ES Aq - Broth Ø ~500 nm SS;
[118] Ec PVP ES Trehalose; O - Buffer Ø ~100 nm SS;
[106] Ec Alginate + PEO W/O + E + G + ES Chl FC Buffer Ø ~ µm SS
[106] Ec PEO ES (coaxial) Chl FC Buffer Ø ~ 1 µm Core-shell; SS
[153] Ec PEO; CDA ES (coaxial) Chl FC Buffer Ø ~ 1 µm Core-shell
[148] Sa HYMA photo-crosslinking - - Buffer - Enzyme triggered release
[142] Sa Soya bean oil O/W surfactant W - - Nano-emulsions
4.8. Encapsulation of bacteriophage in polymeric microparticles emptying rates are another important factor and need to be considered
in the design of phage encapsulation systems for oral use. Previous
Polymer phage encapsulation literature has focused largely on studies indicate that in humans gastric emptying of small microcapsules
gastrointestinal infections (Table 3). The drivers for encapsulation are (< 2 mm) is rapid (~ 0.5–1.5 h) and is not greatly affected by the di-
the need to protect phage from the harsh stomach environment ren- gestive state of the individual [163]. Gastric emptying times and en-
dering free phage inactive or at any rate resulting in reduction in phage vironmental conditions in other animals may be quite different to those
titre. High doses of bacteriophage need to be delivered in a controlled in humans e.g. gastric emptying times in pigs is longer (~ 1.4–2.2 h)
manner at the site of infection in order to effectively reduce the con- [164]. A suitable biocompatible carrier encapsulating the phage may be
centration of infectious bacteria present there. This poses a consider- required to protect and circumvent natural processes of the biological
able formulation and delivery challenge. Animal studies focusing on system e.g. the acidity of the stomach environment [75]. Triggered
gastrointestinal infections (discussed above) have shown poor efficacy release of the phage at the site of infection needs to be engineered into
outcomes due to the challenges associated with in vivo phage therapy. the delivery system. For example, the gastrointestinal tract may be
Oral administration of bacteriophage for human or animal use requires modelled as a distinct set of compartments (duodenum, jejunum, ileum
careful consideration of a number of factors including: the acidic pH of etc.) with each compartment having different environmental conditions
the stomach, digestive enzymes (pepsin, proteases, lipases, amylase and (with different pH and transit times) [165]. Release of phage in the
trypsinogen), bile salts, pancreatic juices, residence time in different infected colon may be the desired outcome. High capsule erosion rates
intestinal compartments (duodenum, jejunum, ileum) and phage per- triggered by a high pH early in the gastrointestinal tract may result in
meability into the mucosal lining where the infection may reside. phage release too early with poor phage therapy outcomes. The harsh
Without proper formulation, phage may easily be inactivated due to environment of the digestive system renders many sensitive therapeutic
exposure to adverse environmental conditions. Carriers designed to agents e.g. proteins and phage inactive when administered without due
protect the active agents e.g. from the harsh gastrointestinal environ- consideration to proper formulation [122,166–168]. Owing to the
ment may also be used to trigger their release. An important example of nature of the infection, fluctuations in physiological conditions may
this is pH dependent release of encapsulated phage [75], exploiting the also need to be considered and may indeed be put to use to trigger
variation in pH throughout the GI tract; carriers may be designed to release at the site of infection [141]. Consideration of particular in-
respond to specific pH which differs from the stomach (pH 1–3) to the fection specific symptoms that might influence the delivery of the
small intestine (pH 5.5–6.5) and the colon (pH 6.5–7.2) [141]. Gastric phage needs to be taken into account. For example, the onset of
119
diarrhoea in humans causes the osmotic gradient between the epithelia Bifidobacteria, Eubacteria, Clostridia, Enterococci and Enterobacteria. To
and colon to decrease resulting in increased fluid leakage and shorter survive in this environment bacteria utilise undigested substrates pre-
transit times [169]. Changes in pH have been observed in the intestine sent in the small intestine. Carbohydrates like di- and tri-saccharides
during infection which can directly affect the microbial population as and polysaccharides are fermented by these organisms [187]. En-
well as transit times in the GI-tract [170]. capsulation using natural polysaccharides is attractive due to their
Pathogens that cause systemic diseases may spread by the faecal/ biodegradability by human enzymes. Microbes produce enzymes such
oral route and may cross the intestinal epithelial barrier by transloca- as β-glucuronidase, α-arabinosidase, β-xylosidase, β-galactosidase, ni-
tion. The small intestine and the ileum are scattered with lymphoid troreductase or urea dehydroxylase [188]. These enzymes metabolise
tissue that forms the Peyer's patch (PP) nodules. The PP nodules are polysaccharides which the body is unable to process. Naturally occur-
overlaid by the follicle-associated epithelium (FEM) displaying ring polysaccharides are popular because they are low-cost and readily
Microfold cells (M cells). A large number of pathogens infecting the available offering a wide variety of structures and properties [189,190].
human body take advantage of this route to gain access to the host Other polysaccharides used for drug delivery applications include;
tissues. Salmonella typhimurium [171], E. coli [172] and many other cellulose derived biopolymers e.g. hydroxypropyl methylcellulose,
pathogens have been shown to adhere preferentially to M cells. Tar- pectin derived biopolymers e.g. polygalacturonic acid, guar gum, car-
geting the M cells in the PPs may offer the opportunity to deliver phage rageenans, carob bean gum (galactomannans) and dextrin. Amylose, a
accurately at the site of infection. Using anti-M-cell monoclonal anti- derivative of starch has been used to develop commercially available
bodies as specific ligands, adsorption of phage containing micro- or drug delivery systems, COLAL® and COLAL-PRED® for colonic diseases.
nanoparticles (e.g. liposomes) into PPs may be used to target the M-cell These polysaccharides utilise the role of enzymatic action by microflora
membrane [173]. to release the encapsulated agent [191,192].
The micro/nanoparticles must possess specific physicochemical Alginate and chitosan have been used in a number of studies to
properties such as size, zeta potential and hydrophilicity/hydro- encapsulate bacteriophage (Table 3). Sodium alginate is an anionic
phobicity balance contained within a narrow range of values [174]. linear polysaccharide composed of alternating blocks of β-(1 → 4)-
Enhancing phage residence time inside the intestine may linked D-mannuronic acid (M) and α-(1 → 4)-linked L-guluronic (G)
exploit the mucoadhesive property of certain molecules e.g. alginates residues. The encapsulation of phage in alginates is achieved using mild
[78,175,176]. This approach has not met with universal acceptance processing conditions; alginate gelation is achieved by crosslinking of
since the superficial mucus layer in the intestine is cleared at a rate carboxylate anions of guluronic acid and calcium ions. Chitosan is a
between 50 and 270 min, potentially removing anything bound to it. cationic hydrophilic polysaccharide obtained by partial deacetylation
Nanocarriers coated with polyPEG, hyaluronic acid or decorated with of chitin; it has been studied for its muco-penetrative and tight-junction
vitamin B12 have been shown to have a greater ability to diffuse penetration properties [193,194]. Penetration of the mucosal anionic
through the mucus layer [177] thereby avoiding mucoadhesion. surface by chitosan is determined by the amino groups [195]. Lameiro
There is scope for the development of innovative strategies to de- et al. [196] showed encapsulation of an adenoviral vector; for mucosal
liver phage encapsulated in micro- and nanoparticles for different delivery of vaccines; encapsulation efficiency over 84% was achieved
target clinical applications. Chemically modified biopolymers provide with a chitosan coating. Premature release in aqueous media as a
many opportunities for phage encapsulation including aliphatic polye- consequence of the swelling characteristic of chitosan was observed.
sters, polyamides, polycarbonate and poly(amino acids). Synthetic Takeuchi et al. [197] showed that chitosan coated sub-micron lipo-
polymers, like Eudragit® are a family of pH responsive poly- somes interacted with the gut mucosa and had longer residence times in
methacrylate polymers; Eudragit® L100 dissolves at pH 6 and Eudragit® the GI tract. Pectin has been used for the encapsulation of phage [122].
S100 at pH 7. These polymers are commercially available (Evonik, Pectin is a polysaccharide of partial methyl ester of α-D-galacturonic
Darmstadt, Rohm Pharma Germany) and are widely used in the pre- acid (poly-Gal) interrupted with (1 → 2)-α-L-rhamnose units. Pectin
paration of microspheres or for their coating. Eudragit® has been suc- forms rigid gels through crosslinking of galacturonic acid with multi-
cessfully used for the production of microspheres encapsulating drugs valent cations. Pectin incorporated in liposome nanocomplexes has
for colon delivery [178–182]. Combining Eudragit® with other biode- been shown to result in increased intestinal mucoadhesion and pro-
gradable polymers (mentioned above) may allow tailored systems for longed retention in the intestinal mucosa [198].
phage intestinal delivery. “Smart” stimuli responsive polymers with
modifiable properties have been studied for various applications with 4.8.1. Protection of bacteriophage from stomach acid, pepsin and bile salts
tunable pharmacokinetics [147,148,183]. Alginate and chitosan have been widely used to encapsulate a
Particle size, surface charge and presence of ligands on the carrier variety of bacteriophage for oral delivery applications (Table 3). The
surface play an important role in delivery and have so far not been porosity of alginate gel micro-beads affects phage susceptibility to acid
extensively exploited for targeted phage delivery. A vast choice of damage (when encapsulated phages have been exposed to SGF) due to
natural and synthetic polymers is available offering considerable op- diffusion of acid into the microparticles [122,143,145]; hydrogel pores
portunities to tailor the encapsulation and subsequent release char- tend be in the 5–200 nm range depending on the degree of crosslinking
acteristics of phage for different biomedical applications. Many carrier [175]. Dini et al. [122] showed that acid exposure of E. coli phage
systems have been explored for controlled release of therapeutic agents; (pH 1.6, 30 min exposure) encapsulated in pure alginate microparticles
these include polysaccharide polymers, synthetic polymers, liposomes, resulted in complete loss of viable phage. L. lactis phages encapsulated
and micelles, thereby permitting use of a multitude of different strate- in alginate were rapidly inactivated upon exposure to SGF at pH 2
gies for encapsulation and release based on different stimuli. [149]. A number of studies have shown that incorporation of calcium
Polymer based controlled release systems have been engineered for carbonate (antacid) within alginate micro-beads helps protect en-
the delivery of therapeutic agents for enteric applications including capsulated phage from acid damage [78,145]. The antacid containing
treatment of ulcerative colitis, Crohn's disease [184], colorectal cancer encapsulated phage were significantly more stable exposed to SGF
and food poisoning as well as many others [185]. The concentration of acidic conditions in comparison with free phage.
bacterial species in the upper GI tract is approximately 103–104 CFU/ Blending alginates with other polymers such as neutral gums (e.g.
ml, largely composed of Gram positive facultative anaerobic bacteria. Guar gum [122]), pectin [122], chitosan [122,143,145,158], whey
In the colon however, this increases significantly to 1011–1012 CFU/ml protein [149,157,159] and Eudragit [178] has been used to improve
[186] and is largely composed of anaerobes such as Bacteroides, the acid stability of alginate microparticles.
120
A number of studies have indicated that phage encapsulated in al- mutant bacteria. Colom et al. [78] showed the in vitro release kinetics
ginate-chitosan microparticles improves phage stability upon exposure in SIF (40 min exposure, pH 8, 1 mg/ml pancreatin, 10 mM bile salts,
to acid conditions however, the protection was not complete 42 °C) for three Salmonella phages (ϕ20 (podovirus), ϕ78 (podovirus)
[122,143,145]. Ma et al. [143,145] showed that Salmonella and S and ϕ87 (myovirus)) individually encapsulated in alginate micro-
aureus phages are sensitive to the acidic pH of stomach and therefore particles containing CaCO3. Almost complete phage release was re-
required protection for enteric treatment; this was afforded by en- ported within 40 min of exposure to SIF.
capsulating phage in composite microparticles containing chitosan and Microparticles with different compositions of alginate-whey protein
alginate. Only a ~2.5 log decrease in Salmonella phage Felix O1 was encapsulating Salmonella Felix O1 phage were exposed to SIF (pH 6.8)
shown after 1 h exposure to simulated gastric fluid (containing pepsin) for 6 h [157]. The microparticles were found to swell and eventually
at pH 2.4 compared with release of phage at pH 6.8 [143]. Kim et al. disintegrate. Encapsulated Salmonella Felix O1 phage showed sustained
[158] reported promising results for E. coli (Myoviridae) phage en- release profiles which varied depending on the amount of polymer
capsulated in chitosan-alginate beads exposed to SGF (containing content of the particles. For the same alginate content, increasing the
pepsin, 1 h exposure at pH 2 and pH 2.5). For free phage, complete loss whey protein concentration was found to slow the phage release rate.
of viable phage was noted within 5 min of acid exposure at pH 2 and Whey protein hydrogels tend to swell readily at pH values above the
after 60 min exposure at pH 2.5. Additives such as maltose (dis- protein isoelectric point (IEP ~pH 5.1) [199]. Pancreatin may also
accharide) have been added to formulations to improve storage stability degrade whey protein provided it is able to diffuse into the hydrogel
of alginate encapsulated phage to prevent phage loss during air drying structure. This may explain increasing erosion rates after an initial
of microparticles. Ma et al. [143] exposed Salmonella Felix O1 phage period of exposure to SIF. Both whey protein and alginate possess ne-
encapsulated in chitosan-alginate micro-beads to simulated bile (ex- gatively charged functional groups which would result in electrostatic
posure for 1 h and 3 h). Free phage showed a 1 log (1 h exposure) and repulsion and swelling of the gel upon exposure to alkaline SIF. Algi-
2 log (3 h exposure) reduction in phage titre. Microencapsulation in nate-whey protein encapsulated S. aureus phage K released phage over a
chitosan-alginate beads resulted in complete protection of phage to bile 2 h period upon exposure to SIF (pH 6.8) [159]. The microspheres were
salt inactivation. Dini et al. [122] showed that acid exposure of E. coli seen to swell and eventually disintegrate with sustained release of
phage (pH 1.6, 30 min exposure) encapsulated in pectin afforded im- phage. Changes in formulation resulted in modest differences in the
proved resistance to acid exposure and protection from pepsin in- release profiles. Dried microspheres resulted in slower release of phage.
activation. Increasing concentrations of either alginate or whey protein retarded
The concentration of encapsulating agent was found to be an im- the release of phage K. L. lactis P008 phage encapsulated in alginate-
portant factor in affording phage protection from SGF. Increasing whey whey protein showed a sustained release rate upon exposure to SIF over
protein concentration in alginate-whey protein microparticles resulted a 2 h period [149]. Pre-exposure to SGF followed by exposure to SIF
in better acid protection by reducing permeation rates of acid and di- resulted in a ~ 1 log reduction in total phage released. Gelled milk
gestive enzymes [159]. Whey protein amino groups may sequester protein microspheres were also found to be suitably protective of P008
protons diffusing through the polymer matrix. Tang et al. [157] en- phage and showed sustained release of phage upon exposure to SIF over
capsulated Salmonella Felix O1 phage in alginate-whey protein micro- a 2 h period.
particles (different alginate: whey protein compositions) with high
phage loading 1010 PFU/g wet particles. Solutions with high polymer 4.8.3. Effect of size of beads
concentrations (~ > 5% w/v) were found to have high viscosity and Particle size was found to be an important factor influencing phage
posed difficulties in extrusion and lead to a wider distribution of par- protection from SGF for acid permeable beads [159]. Alginate-whey
ticle sizes [157]. Alginate-whey protein microspheres were found to be protein microparticles having the same composition but different mean
better at protecting Salmonella Felix O1 phage in SGF compared with sizes showed larger particles protected phage better compared with
alginate chitosan microspheres [143] prepared by the same group. smaller microparticles [159]. Larger particles keeping phage further
Exposure of encapsulated Felix O1 to bile acid (1 h and 3 h duration) away from the acid environment were better at protecting phage from
did not result in significant change in phage titre [157]. acid damage [157]. However, larger particles are expected to experi-
ence more mastication [159] in the mouth and longer gastric retention
4.8.2. Release kinetics of encapsulated bacteriophage [163]. Samtlebe et al. [149] encapsulated L. lactis P008 phage in en-
Most studies with alginate containing phage show slow release of zymatically gelled milk protein microcapsules with a small mean par-
encapsulated phage over a 2 h–6 h period upon exposure to SIF ticle size (~ 130 μm) which was shown to protect phage exposed to SGF
[143,158]. Most systems swell upon exposure to SIF and phage release at pH 2 for up to 2 h. Colom et al. [78] showed that three Salmonella
is due to diffusion and encapsulation matrix degradation. Ma et al. phage individually encapsulated in alginate microparticles (si-
[143] showed that S. aureus phage K encapsulated in chitosan-alginate ze ~ 150 μm) containing CaCO3 were able to protect the phage from
micro-beads showed a sustained release rate over a 6 h period (ex- exposure to SGF although reduction in phage titre was noted for phage
posure to SIF pH 6.8). Microspheres were seen to swell and disintegrate. ~3.1 log reduction ϕ78 and ~ 2.4 log reduction ϕ87. No loss of titre
Addition of calcium carbonate in alginate-chitosan micro-beads slowed was noted for phage ϕ20.
down the release rate. This was attributed to excess Ca2 + ions slowing Future studies may want to focus on developing methods to control
the alginate hydrogel dissolution [145]. and tailor the porosity of the microcapsule shell to protect encapsulated
Kim et al. [158] exposed E. coli (Myoviridae) phage encapsulated in phage from SGF acidity whilst controlling the phage release rate.
chitosan-alginate beads to SIF (0.1% bile and 0.4% pancreatin, pH 7.5,
37 °C, 6 h exposure). Sustained release of E. coil phage encapsulated in 4.8.4. Storage stability
chitosan-alginate beads over a 6 h period was noted [158]. Sustained Wet alginate-whey protein microspheres containing encapsulated S.
release of phage (from beads) was shown to result in a ~ 3 log reduction aureus phage K showed no loss in phage viability over a six week sto-
in the in vitro concentration of E. coli 157:H7 after 5 h (compared with rage period (stored at 4 °C) [159]. Dried microspheres (with added
~ 4 log reduction by free phage). The starting concentration of bacteria maltose) were shown to retain phage viability over a two week storage
was 106 CFU/ml. After 10 h of incubation in SIF, both free phage and period (no reduction in phage titre when stored at 4 °C and 20% re-
encapsulated phage were unable to arrest bacterial growth (comparison duction in phage titre when stored at 23 °C). Colom et al. [78] stored
with untreated controls) possibly due to presence of phage resistant Salmonella phage encapsulated in alginate/CaCO3 microparticles at 4 °C
121
for six months and showed only a slight (~ 25%) reduction in phage in the presence of hyaluronidase enzyme secreted by S. aureus (a rare
titre. Tang et al. [157] air dried (at 22 °C) Salmonella Felix O1 phage example of using a virulence factor secreted by the pathogen as a
encapsulated in alginate-whey protein microparticles and found an 8- trigger for phage release).
log reduction in phage titre. Addition of maltodextrin to the alginate- Milo et al. [203] describe the use of a pH responsive coating for
whey protein microspheres stabilised the phage resulting in a < 0.5 log urinary catheters which may immobilise phage and release these upon
reduction in titre post-air drying (drying time 30 h, at 22 °C). No sig- onset of infection by Proteus mirabilis. Urinary tract infections may re-
nificant loss in Salmonella Felix O1 phage titre was observed for air sult in an increase in pH due to bacterial urease which could be used as
dried particles over a 2 week storage period (stored at 4 °C and 23 °C). a trigger to dissolve an upper layer of pH responsive polymer (e.g.
Phage encapsulated in wet alginate-whey protein microspheres stored methyl methacrylate-co-methacrylic acid; Eudragit S100) as part of the
at 4 °C for 6 weeks showed no significant loss in phage titre. construction of the catheter hydrogel coating.
4.8.5. Bacteriophage encapsulation in synthetic polymers 4.9. Bacteriophage encapsulation in liposomes

Surprisingly few studies are available on phage encapsulation in
synthetic polymers. Puapermpoonsiri et al. [100] freeze dried lytic Relatively few published studies have looked at encapsulation of
Staphylococcus aureus and Pseudomonas aeruginosa phage encapsulated bacteriophage in liposomes (see Table 4). In nearly all cases the thin-
in PLGA (poly(lactic-co-glycolic acid)) microspheres prepared using an film hydration method has been used for liposome preparation. Briefly,
W/O/W double emulsion technique (using dimethylformamide, DMF as this entails dissolving lipids in an organic solvent (typically chloroform)
organic solvent). The microsphere size was designed to deliver the followed by solvent evaporation under vacuum leaving a dry lipid film.
encapsulated phage via inhalation using a dry powder inhaler (1-5 μm). Addition of an aqueous phage suspension to the dry lipid film results in
The phage encapsulated microparticles had low bulk density and the stacks of liquid crystalline lipid bilayers becoming fluid and swel-
aerodynamic diameter ~ 4 μm suitable for respiratory delivery. Phage ling. Agitation of the solution results in detachment and closure of the
activation was retained post-particle formation and sustained phage lipid layers thereby forming relatively large and heterogeneous multi-
release kinetics was noted upon exposure of particles to PBS buffer over lamellar liposomes. Extrusion of the liposome mixture through porous
a 6 h period. Phage release showed an initial burst release (within membranes has been be used to form smaller liposomes [77,150]. Nieth
30 min for microparticles suspended in buffer) followed by a longer et al. [47] employed a novel method proposed by Weinberger et al.
release over ~ 6 h suggesting either inhomogeneous distribution of [204] to prepare phage encapsulated in giant unilamellar vesicles
phage within the microparticles or reflecting the dissolution kinetics of (GUV) using gel-assisted formation of GUV by rehydrating lipid layers
the polymer microparticles [100]. The formation of phage loaded PLGA on polyvinyl alcohol gels. This technique afforded a number of ad-
microspheres necessitated a solvent evaporation step (accomplished vantages over the traditional thin film hydration method. Firstly, faster
using freeze drying). No viable phage were recovered from the particles hydration kinetics due to aqueous solvent penetration through lipid
after 7 days of storage at either 4 °C or 20 °C. Phage inactivation due to film stack was achieved due to the difference in chemical potentials
exposure to organic solvent (dichloromethane) posed a problem. Mat- between the dry PVA supporting the lipid stacks and the liquid phase
subara et al. [200] evaluated the kinetics of the loss of viability of a above [47]. Secondly, better encapsulation phage efficiency was
filamentous phage exposed to four water miscible organic solvents achieved with the phage incorporated within the cast PVA gel films
ethanol, acetonitrile, dimethylsulfoxide and DMF; these solvents dis- instead of the rehydration buffer. Confocal microscopy showed visual
rupt protein conformation and enzymatic activity [201]. evidence that this method allowed good encapsulation of a fluores-
Stanford et al. [75] used Eudragit ES100 (methyl methacrylate-co- cently labelled E. coli λ phage. When the E. coli λ phage was provided
methacrylic acid polymer) for Escherichia coli specific phage encapsula- externally with the swelling buffer, no encapsulation was observed
tion for cattle treatment. They showed high sensitivity of the phage to the suggesting that the lipid stacks themselves were impermeable to the
acid in the gastric environment. However, encapsulated phage showed phage present in the external solution. The GUV prepared using this
improved viability and in vivo reduction in faecal shedding of E. coli technique resulted in large vesicles (5 μm–50 μm). Extrusion of GUV
following induced infection (discussed earlier in the review). The poten- through a 5 μm pore size membrane resulted in reduction in vesicle size
tial of ES100 for encapsulation of phage is also reported in a patent [202]. to < 5 μm however, E. coli λ phage concentration was below the sen-
A number of other published studies have looked at phage encapsulation sitivity of the fluorescence microscope. No quantitative phage titres
in polyester amide films for wound healing applications [121,160]. Thin were carried out by breaking the liposomes and releasing their phage
biodegradable polymer films impregnated with a mixture of lytic bac- cargo. Nieth et al. [47] also employed the inverse emulsion liposome
teriophage (containing a mixture of lytic phage (trade name, Pyophage) preparation method of Pautot et al. [205] to encapsulate the fluorescent
against P. aeruginosa, E. coli, S. aureus, Streptococcus, and Proteus mirabilis, E. coli λ phage. GUV prepared using the inverse emulsion method had
sold under the trade name PhagoBioDerm) was employed Markoishvili an average diameter between (10 μm–15 μm). Nieth et al. [47] also
et al. [121] to treat skin infections. showed significantly greater intracellular uptake of liposome en-
Hathaway et al. [147] encapsulated S. aureus phage K in poly(N- capsulated fluorescent E. coli λ phage (compared with free phage) in a
isopropylacrylamide), a thermally responsive polymer. Phage were macrophage-like human monocytic cell line.
encapsulated in PNIPAM nanospheres (~ 500 nm) which undergo a Colom et al. [77] encapsulated S. enterica phage in cationic lipo-
fully reversible temperature dependent phase transition at a lower somes (liposome size range ~ 200–800 nm). They suggested that ca-
critical solution temperature (LCST) which results in a change in the tionic liposome loaded bacteriophage would achieve two objectives: (i)
particle size due to polymer shrinkage and expulsion of water from the to protect the bacteriophage from gastric acids and (ii) act as a pro-
nanospheres along with the phage contained therein. The LCST for moter for mucoadhesiveness owing to their positively charged surfaces
PNIPAM can vary between 32 °C to 36 °C depending on polymer thereby prolonging intestinal residence time of phage [77]. The Sal-
properties [147]. The PNIPAM encapsulated phage nanoparticles were monella phages were encapsulated in unilamellar liposomes using the
grafted on to non-woven polypropylene films (mimicking a wound thin film hydration method. In vitro studies showed that encapsulated
dressing film) [147]. Evidence of temperature dependent phage release phage were released upon exposure to SIF within 60 min. Exposure to
data was somewhat limited in this paper. SGF resulted in loss of phage viability with modest protection of S.
Bean et al. [148] encapsulated S. aureus phage K in an agarose slab enterica phages afforded by encapsulation in cationic liposomes com-
covered in a thin film of a photo-cross-linkable hyaluronic acid me- pared with free phages. The intestinal residence time of liposome en-
thacrylate (HAMA) polymer. The HAMA polymer was shown to degrade capsulated phages in the cecum of newly hatched chickens was
122
considerably longer compared with free phage (refer to earlier section
intracellular macrophage model;

Freeze drying; storage studies;
Giant unilamellar vesicles; PE
on animal studies). This was attributed to the small size and mu-
PE = phage encapsulation
Storage studies; PE (good)

coadhesive properties of the cationic liposomes. Nieth et al. [47] also
Ex vivo biofilm model;

encountered problems in terms of bacteriophage encapsulation in li-
posomes. Encapsulation efficiency, control over the size, phage titre,
were found to be key issues that need addressing in future studies.
Animal studies
animal studies
PE (modest)
Balcao et al. [206] encapsulated lytic bacteriophage (against Salmonella
PE (good)
outcome
and E. coli) by emulsifying an aqueous phage solution (water phase) in a
Of note
(good)
lipid melt (oil phase) using homogenisation technique resulting in na-
noparticles in the size range (85–200 nm). The size of the nanoparticles
~500–1000 nm
showed stability over a 3 month storage period at room temperature.
~200–800 nm
Quantitative results on the antibacterial activity of encapsulated phage
50–250 nm
~100 nm
~100 nm
5–50 μm
were not provided although lytic activity of encapsulated phage was
Size
observed following phage release.
Singla et al. [39] evaluated the encapsulation of K. pneumonia phage
Simulated gastric
(belonging to the Podoviridae family) in liposomes prepared using dif-
Release studies
ferent phospholipid: cholesterol ratios with additional charge inducing
fluid/buffer
agents to prevent liposome aggregation. The liposomes were prepared
using the film-hydration method and ranged in size between 500 and
1000 nm. Phage encapsulation efficiency was shown to be higher for
–
–
–
the electrostatically stabilised positively charged liposomes (containing
Target infection
Gastrointestinal
stearylamine) with higher lipid content. Phage encapsulated liposomes
Intracellular
Respiratory
Respiratory
Respiratory
Respiratory
stored at 4 °C for 9 weeks were stable both in terms of size as well as
phage titre. Storage at ambient temperature or 37 °C resulted in
changes in size (increase) as well as loss of phage titre. In vivo toxicity
testing and biodistribution of phage was carried out in mice. Briefly,
~ 108 PFU dose was delivered to mice via the intraperitoneal route.
Rectal temperatures were taken hourly (for the first 5 h) and daily (for
polyoxamers (Lutrol F68)
Tween 80; stearylamine;

Tween 80; stearylamine
Tween 80; stearylamine

Chloroform as solvent;
4 days post-phage treatment). No adverse change in temperature or
Tween 80; glycerol;
trehalose added as
symptoms of lethargy or sickness was noted for the liposome/phage
dicetylphosphate
cryoprotectant
treated animals. Phage distribution in lungs, liver, spleen and kidney
were quantified. Blood titres of phage reached ~ 10 PFU/ml shortly
Additives
after injection (within 1st hour) for both free phage and liposome en-
capsulated phage. Liposome encapsulated phage remained in blood and
–
other organs for longer periods compared with free phage treated mice.
Thin film hydration + PVA
Free phage were cleared from the blood and organs 48 h post-treatment
Encapsulation method
gel assisted hydration
whereas liposome encapsulated phage were detectable in blood and all

Water-in-oil-in-water
Thin film hydration


organs up until day 3 (albeit at low concentrations ~ 1 PFU/ml) post-
double emulsion
treatment and remained detectable in the spleen up until day 14.

+ extrusion
Liposomes are spherical structures delimited by a double layer of

bulky amphiphilic lipids which encloses an aqueous phase. Charge, size,
fluidity of the bilayer, lamellarity and surface functionalization are the
major parameters that regulate liposomes fate in vivo and in vitro. Size
DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine); DOPS
and lamellarity vary as a result of the manufacturing process. Fluidity

Mixture of DLPC:Chol-PEG600:cholesterol:cholesteryl
and superficial charge depend on the chemical nature of the lipids and
Summary of bacteriophage microencapsulation literature in liposome carriers.
their relative ratio. Liposomes are also highly biocompatible and they
have been widely used in the last fifty years to encapsulate hydrophilic
(1,2-dioleoyl-sn-glycero-3-phospho-L-serine)
or hydrophobic therapeutic agents in their aqueous core or within the

bilayer thereby enhancing drug bioavailability and stability over time.
Encapsulated liposomes are a promising drug delivery system and
Phosphatidylcholine:cholesterol
pharmaceutical carrier of choice for numerous applications [207].

Softisan 100™; soybean PC
From a clinical viewpoint the potential ability of liposome en-

molar ratio (10:1:2:7)
capsulated phage to enter live cells containing intracellular pathogens

is of crucial importance e.g. M. tuberculosis or other intracellular bac-
Carrier system
teria e.g. S. aureus or E. coil. The surface of liposomes may be functio-

nalized to achieve various goals: (i) attachment of specific ligands for
targeted delivery of liposome cargo; (ii) attachment of hydrophilic
polymers to increase systemic circulation time and slow release of
bacteriophage cargo; (iii) attachment of various labels to monitor the
Host organism
fate of the bacteriophage loaded liposomes; (iv) incorporation of posi-

K. pneumonia
K. pneumonia
K. pneumonia
S. enteritidis
tively charged lipid derivatives or positively charged polymers e.g. to

S. enterica
Smegmatis
E. coli; M.
aid mucosal adhesion thereby improving residence time in the gastro-

intestinal tract; (v) protection of bacteriophage from physical and
chemical stresses; (vi) liposome encapsulated bacteriophage may result
Table 4
[206]
[150]
[77]
[31]
[39]
[95]
in modulated immune response to the presence of the bacteriophage in

Ref
the body.
123
4.9.1. Liposome size, charge and surface modification e.g. PEGylation is even less clear; bacteria can remain stored for years in an asympto-
Previous studies have found that by reducing the size of particles matic state that can suddenly be triggered into a potentially life-
below 1 μm (i.e. nm scale) enhancement in liposome nanoparticle ac- threatening active state, especially for people with a weak or compro-
cess to infection sites deep in intestinal mucosa may be achieved mised immune system. Liposomes innately accumulate in MPS due to
thereby improving phage therapy efficacy by targeting bacterial re- their interaction with serum proteins that accelerates phagocytosis. It
fugees residing in these difficult to access niches [197,198]. The re- makes bacteriophage-encapsulated liposomes an appealing approach to
sidence time of nanoparticles has been shown to be longer compared to treat antimicrobial resistant bacterial infections that spread from within
larger microparticles; larger particles are more prone to peristaltic ac- the MPS. The rate of clearance depends on the physical/chemical
tion and rapid transit through the colon. Nanoparticles have been characteristics of the vesicle` surface. It has been shown that uptake by
shown to interact with epithelial cells and show increased retention MPS for liposomes is enhanced when phosphatidylserine incorporating
enabling selective delivery of therapeutic agents into the colitis tissue PEG is used. On the other hand, long-circulating liposomes with high
[208]. Conventional formulations have so far not exploited this ability PEG value can accumulate at the site of infection due to increased
causing deposition in regional areas. Smaller particle sizes are able to permeability. Agrawal et al. [215] have demonstrated that tuftsin, a
take advantage of endocytosis for internalization into the epithelial tetra-peptide, used as a homing-device improves liposomes uptake by
cells of the colon. They attach to specialised epithelial M cells (dis- MPS cells but also stimulates these cells to take action against infec-
cussed above) which are responsible for antigen transport from the tions. Leishmaniasis, Aspergillus fumigatus and Mycobacterium tubercu-
lumen to the immune system. These M cells are also responsible for losis infections have been treated with drug-loaded tuftsin-bearing li-
uptake of nanoparticles by transcytosis [209]. Microparticles have been posomes and tuftsin-free liposomes and in all cases the treatment lead
shown to adhere to the inflamed mucosal wall whereas nanoparticles to enhanced effectiveness.
are absorbed across the epithelial barrier [210]. Phage therapy has recently been considered as an alternative to, or a
Interactions between charged particles and the mucus may be an supplement for, antibiotic treatment for mycobacterial respiratory in-
appropriate strategy for intestinal delivery. Cationic delivery systems fections. However, phage therapy for TB and other respiratory infec-
adhere to the negatively charged intestinal mucosa thereby promoting tions caused by non-tuberculosis mycobacteria (NTM) is complicated by
cellular uptake and subsequent drug release [211]. The mucins which the intracellular nature of the infection. It is generally recognised that
comprise the mucosa are also known to be negatively charged. Anionic phage do not diffuse across eukaryotic cell membranes and therefore
systems adhere to inflamed tissue via electrostatic interactions. The would be unable to infect the intracellular mycobacteria [48]. Using a
presence of positively charged proteins in the inflamed regions typically non-pathogenic mycobacterium, Mycobacterium smegmatis, infected
increases. Some authors suggest that unlike cationic particles which get with a mycobacteriophage as a ‘Trojan horse’ to deliver phage to the
immobilised on the mucus, anionic particles are able to diffuse through macrophage compartments in which M. avium resides, showed promise
the mucus network due to decreased electrostatic interactions [170]. when tested in vitro with infected macrophages [216]. However, when
Other studies suggest that particles are entrapped within the mucin used in a murine model of M. avium infection, this approach was less
mesh no matter what charge they carry since the mucus displays ne- effective. It was deemed that the numbers of infected M. smegmatis cells
gatively and positively charged proteins. Hence, if a non-adherent ap- required, would deliver too large a dose of antigenic material to be
proach is needed, the nanoparticle surface could be decorated with acceptable [217]. Alternative methods of delivery of phage to in-
specific moieties that confer so called ‘stealth’ properties [212]. In- tracellular mycobacteria have been sought. Looking to the relatively
corporating poly(ethylene glycol) (PEG) on the surface of nanoparticles routine use of liposome encapsulation for intracellular delivery of small
has been shown to result in a hydrophilic surface chemistry that allows molecule drugs [218], Nieth et al. [150] recently suggested en-
unobstructed diffusion of nanoparticles through the epithelium. An capsulation of phage within liposomes may result in uptake of phage
explanation for this is that surface modification prevents strong inter- and delivery to the required intramacrophage compartment to initiate
action of nanoparticles with the mucus which typically prevents diffu- lytic infection of intracellular pathogens. To this end, they demon-
sion to the colitis tissue [213]. strated successful encapsulation of mycobacteriophage TM4 in giant
Active targeting by surface functionalization of phage containing unilamellar vesicles. Furthermore, by encapsulation of a lambda phage
nanoparticles e.g. liposomes with ligands for selective phage accumu- expressing yellow fluorescent protein and fluorescent labelling of
lation at the site of infection may help increase the local concentration macrophage endosomal proteins, they were able to demonstrate in-
of phage delivered thereby increasing their therapeutic efficacy. tracellular delivery and co-localisation of the lambda phage and the
Monoclonal antibodies and peptides specific for target sites have been endosomes; in Mtb infected macrophages, the mycobacterial phago-
used to increase specificity and enhance muco-penetrative properties some fuses with early endosomes. This demonstration that liposome-
[214]. It is thought that by incorporating ligands targeting receptors encapsulated phage could gain access to the target mycobacteria sa-
present at the site of infection may improve binding and thereby in- tisfies a requirement for effective therapeutic use of bacteriophage to
tensify endocytosis. Possible receptors for targets in the inflamed colon treat intracellular mycobacterial infections. A previous study using li-
may include, ICAM-1 which is upregulated during inflammation, posome encapsulation of frontline antituberculosis antibiotics, ri-
mannose receptors and macrophage galactose-type lectin (MGL) (ex- fampicin and isoniazid, showed enhanced treatment in a murine in-
pressed by activated macrophages). The combined effect of the small fection model of Mtb compared with treatment with antibiotics alone
particle size of nanoparticles, PEGylation and integration of mannose [219]. The efficacy of a liposome-encapsulated mycobacteriophage to
ligands has recently been shown to increase penetration ability for treat Mtb infection in macrophages or animal models remains to be
macrophage targeting [208]. demonstrated.
Functionalization of liposomes for specific targeting may have an Encapsulation of bacteriophage in liposomes may be a potential
attractive future for phage delivery applications. Some pathogenic mi- approach to tackle antibiotic resistant bacterial infections in wounds.
croorganisms such as M. tuberculosis and S. enterica (serovar Previous studies have shown that topical applications of buflomedil
Typhimurium) are able to survive specific antimicrobial machinery hydrochloride loaded liposomes in mice significantly accelerate wound
causing persistent infections. Long-term residence is achieved by in- epithelialization [220] in normal and ischemic tissue. Furthermore, by
ternalization in mononuclear phagocyte system (MPS). It is not fully modifying the liposome surface properties (e.g. charge), liposomes can
understood how these bacteria inhibit their clearance by the immune be trapped in gauzes [221] reducing the number of applications and
system but common mechanisms include prevention of the normal consequent costs. The surface of liposome vesicles can be modified with
maturation of the phagosomal environment or blockage of the fusion of hyaluronic acid (HA) in order to confer bioadhesive properties [222].
the phagosome with endosome and lysosome. What happens afterwards Indeed, HA-anchored liposomes were found to bind to the extracellular
124
matrix (ECM) without compromising any of the liposome properties. A electrospinning. The phage titre of PVP/SM buffer samples (with and
marked enhancement of liposomal adhesion (in vitro) to a monolayer of without trehalose) fell by no more than ~1 log following electrospin-
human epidermoid carcinoma cells was observed. Topical delivery of ning. Electrospinning phage-PVP solutions in deionised water resulted
bacteriophage-loaded liposomes would constitute a phage reservoir in a 4 log reduction in phage titre. Addition of trehalose to deionised
localised in the area of the wound releasing phage at a high con- water resulted in some improvement with a 2 log reduction post-elec-
centration over an extended period of time. trospinning, showing that trehalose had a protective effect. Dai et al.
[118] also compared phage stabilisation using electrospinning with
4.10. Bacteriophage encapsulated in electrospun fibers freeze drying. Phage encapsulated in electrospun fibers (made from SM
buffer with trehalose) were found to have similar titre compared with
Fabrication of phage encapsulated polymer fibers (diameters ran- freeze dried powder samples. Storage of phage T7 in electrospun fibers
ging from tens of nanometers to microns) may readily be achieved using for eight weeks (stored at 20 °C) resulted in a 3 log reduction in phage
electrospinning, by drawing a charged polymer-in-solvent solution onto titres for samples prepared from SM buffer with/without trehalose.
a grounded electrode whilst evaporating the solvent. A number of Korehei and Kadla [106] evaluated emulsion electrospinning (pre-
studies have shown the feasibility of encapsulating phage in nanofibers encapsulating phage in ~800 nm alginate nanoparticles followed by
using the electrospinning technique (Table 2). A number of different lyophilization and then re-suspension in a polymer containing solvent
polymers have been evaluated for encapsulating phage in fibers in- followed by electrospinning) and coaxial electrospinning (incorporating
cluding cellulose diacetate [153], polyethylene oxide [106,153], poly- the phage in a protective core in a core-shell fiber structure) processes
vinyl alcohol [151,152] and polyvinyl pyrrolidone [101,118]. Good to encapsulate E. coli T4 phage in polyethylene oxide fibers. The process
phage encapsulation results have been reported using water soluble or of encapsulating phage in alginate nanoparticles resulted in a 1 log drop
chloroform soluble polymers [106,118]. Rapid dehydration during in phage titre. Freeze drying the phage encapsulated in nanospheres did
electrospinning was shown to damage phages. Use of buffer and in- not result in a drop in phage titre. Electrospraying the alginate sus-
corporation of sugars e.g. trehalose was shown to improve protection of pension on its own (without polyethylene oxide) resulted in a large
phage [118]. Protection of phages in core-shell fibers using co-axial drop in phage titre from 108 PFU/ml to 103 PFU/ml. Polyethylene oxide
electrospinning technique was shown to be particularly promising (PEO) is soluble in both aqueous and organic solvents making it a
[106,153]. Using immiscible solvents for the core (phage contained in suitable polymer for electrospinning. Polyethylene oxide fibers en-
an aqueous core) and shell (shell polymer dissolved in an organic water capsulating alginate nanoparticles were found to result in fibers with a
immiscible solvent) was found to stabilise the polymer jet resulting in bead-on-string morphology (fiber diameter ~ 1 μm). Dissolution of fi-
the formation of a stable core-shell fiber structure [106,153]. In this bers in SM buffer showed a phage titre of 106 PFU/ml (a 2 log reduc-
process, aqueous solvent evaporation from the core is very slow thereby tion). Fibers with diameter ~ 2 μm were prepared using a core/shell
overcoming drastic changes in osmotic pressure in the phage containing approach with the fiber shell made using PEO in chloroform and the
core. Encapsulation of phages in nanoparticles formulated in an emul- core containing T4 phage in SM buffer [106]. Good lytic activity was
sion followed by electrospinning of the emulsion was another approach retained post-electrospinning with a high phage titre obtained
that was shown to result in improved phage protection during elec- (108 PFU/ml) with fibers dissolved in SM buffer. Storage stability of T4
trospinning [106]. Increasing polymer molecular weight or blending phage in the core-shell fibers after freeze drying (freeze drying condi-
different polymers (e.g. cellulose acetate/PEO) was shown to result in tions were not reported in the paper and no details were given on
changes in the phage release kinetics from electrospun fibers [153]. The whether any excipients were added) was investigated (stored at
main reported mechanism of phage release from encapsulated fibers is − 20 °C, 4 °C and 20 °C) over a period of 30 days. No significant loss in
typically swelling of fibers upon exposure to solvent followed by phage titre was observed for samples stored at − 20 °C and 4 °C. Storage
polymer erosion accompanied by phage diffusion and release. En- at 20 °C resulted in a significant loss (> 6 log reduction) in phage titre.
capsulating phage in hydrophilic water soluble polymers results in Upon exposure of PEO fibers to SM buffer, fibers were seen to swell
rapid phage release due to polymer dissolution whereas use of hydro- followed by disintegration (within 10 min) resulting in release of phage
phobic polymers or blends may allow tailoring of phage release to en- for the alginate-emulsion based fibers and the core-shell fibers. Initial
sure slower release over a prolonged period. rate of phage release was faster from the core-shell fibers (the resulting
The polymer and solvent combination used for fiber production final titre was also higher) compared with the alginate nanoparticle
were found to be important factors in the formation of nanofibers and encapsulated fibers. Korehei and Kadla [153] varied the molecular
retention of phage viability therein. Lee and Belcher [101] spun fibers weight (MW) of PEO (100 kDa–600 kDa) and blended hydrophilic PEO
(diameters 100–200 nm) containing E. coli M13 phage which retained with the hydrophobic polymer cellulose diacetate (CDA). E. coli T4
infectivity against bacterial host when re-suspended in buffer. The phage encapsulated in fibers produced using different PEO/CDA
phage containing fibers were spun from an aqueous polyvinyl pyrroli- showed slow release of phage depending on the PEO MW and the ratio
done buffered solution containing phage; the fibers were captured as a of PEO/CDA. Higher PEO MW (results in increased polymer entangle-
non-woven fibrous mat. Concentration of viable phage in the fibers was ments) and greater proportion of hydrophobic CDA (slower rate of
not reported. Salalha et al. [152] and Kuhn [151] encapsulated T4, T7, polymer swelling and erosion) resulted in slower release of phage.
and λ E. coli phage in electrospun nanofibers (diameter 250–400 nm)
prepared from aqueous poly(vinyl alcohol) (PVA) solutions. Nanofiber 4.11. Immobilisation of bacteriophage on surfaces including thin films
encapsulated phage were shown to retain viability and remained stable
for three months stored at −20 °C and −55 °C and nearly so at 4 °C. Microbial biofilms on medical devices e.g. urinary and central ve-
Salalha et al. [152] suggested that this method was a facile way of nous catheters, stents, implants, invasive health monitoring devices, are
preserving phage. Immediately after electrospinning, a 2 log loss in a significant healthcare problem and are responsible for a vast majority
phage viability was reported (T4 ~ 1%, T7 ~ 2%, and λ ~ 1%). The of medical device associated infections. Bioactive packaging materials
losses in phage viability during electrospinning may be attributed to are also attractive in other areas including food packaging e.g. ready-to-
drying stresses due to rapid evaporation of solvent and drastic changes eat meat, ready cut fruits and poultry [155]. Antibiofilm formation
in osmotic pressure experienced by phage. Dai et al. [118] investigated strategies include oriented immobilisation by covalent attachment of
electrospinning (diameter 100-200 nm) of E. coli T7 phage in SM buffer bacteriophage to polymer surfaces [223] or immobilisation of bacter-
using PVP as polymer (15% w/v) and trehalose (5% w/v) as excipient iophage in stimuli responsive materials where release is triggered in
as a means to improve phage stability for long term storage. Inclusion of response to an external stimulus [148]. Although phage immobilisation
SM buffer had a significant positive effect on phage viability post- on surfaces may readily be achieved through passive adsorption [224],
125
Fig. 3. Membrane emulsification (ME) methods that can be

used for phage encapsulation: (a) Cross flow ME; (b) premix
ME; (c) Shirasu Porous Glass (SPG) microkit; (d) Micropore
Dispersion Cell; (e) axially oscillating membrane tube.
the process is inefficient with loss of phage activity due to poor or- up orientation resulting in increased sensitivity of phage for bacteria
ientation of phage tails needed to recognise and infect target bacteria. and a low detection limit for E. coli of 103 CFU/ml. Tawil et al. [235]
Chemical biotinylation of the phage capsid head has been used for or- reported on the synthesis of colloidal phage-gold nanoparticle com-
iented immobilisation (to retain phage lytic activity by leaving the tail plexes for the detection of a single S. aureus bacterium in a mixture of S.
fibers free in order to capture bacteria) of a Salmonella enteritidis phage aureus and E. coli using Dark Field Microscopy. The potential of phage
on streptavidin-labelled magnetic beads [225]. Tolba et al. [226] used for biosensing applications including low-cost medical diagnostics may
phage display technology to introduce affinity tags on the capsid head require their coupling with novel materials (thin film constructs,
to immobilise E. coli phage on streptavidin coated magnetic and cel- magnetic, metallic, polymer nanoparticles, quantum dots) [24].
lulose beads. Pearson et al. [223] covalently attached lytic E. coli and S.
aureus phage through reaction of primary amine groups on the capsid 5. Future trends in bacteriophage formulation and encapsulation
head with R-COOH surface groups (generated using microwave plasma
in the presence of maleic anhydride) on polyethylene and polytetra- 5.1. Phage encapsulation using novel stimuli responsive materials
fluoroethylene surfaces. Anany et al. [227] and Lone et al. [155] uti-
lised the electrostatic interaction between the anionic capsid head of E. There is significant scope for future research on encapsulation of
coli O157:H7 and Listeria monocytogenes phage cocktails and poly- phage in novel stimuli responsive polymers utilising different triggers
vinylamine treated cellulose membranes. Modification of cellulose for phage release including light, temperature, pH, enzymes. A wide
membranes resulted in greater numbers of infective phage oriented range of external stimuli may be used to trigger self-assembly of phage
correctly. A recent patent focuses on controlled covalent attachment of containing micro- and nano-particle structures (e.g. in micelles, nano-
bioactive bacteriophage to a hydrogel coating material for regulating gels, capsules, vesicles, core-shell particles, hybrid particle-in-particle
biofilm development with examples focusing on biofilm reduction on structures, emulsions and foams) and may induce their reversible or
urinary catheters [228]. An earlier patent gave examples of phage im- irreversible disintegration, aggregation, swelling. Micro- and nano-
mobilisation on nylon sutures for wound healing applications [229]. particles with surface grafted ligands may be used for targeted delivery
Cooper et al. [230] used carbodiimide oriented covalent attachment of of phage cargo released at the site of infection. Stimuli responsive
a tailed P. aeruginosa bacteriophage to magnetised multiwalled carbon homo-polymer and mixed-polymer brushes grafted on to microparticles
nanotubes. Preliminary results showed antimicrobial efficacy of bac- may be used to trigger phage release in response to changes in en-
teriophage-nanocomposite conjugates against P. aeruginosa. Handa vironmental or interfacial conditions. Phage may be encapsulated in
et al. [231] employed chemical vapour deposition to make a smooth different architectures e.g. thin film structures as multi-layered films,
aminosilane monolayer on a glass substrate. Subsequently, monolayers hybrid films combining polymers and nano- or micro-particles.
of covalently bound Salmonella specific phage were attached to the
aminosilane monolayer using well-established sulfo-NHS (N-hydro- 5.2. Phage encapsulation using membrane emulsification
xysulfosuccinimide) and EDC (1-ethyl-3-[3-dimethylaminopropyl] car-
bodiimide hydrochloride) chemistry [231]. Techniques for attachment Advanced techniques may be applied for phage encapsulation to
of phage to gold substrates has been investigated to increase their di- precisely control the size and architecture of phage containing micro-
agnostic utility as part of quartz crystal microbalance (QCM), surface and nano-particles. To-date methods used for phage encapsulation have
plasmon resonance (SPR) transduction based diagnostic platforms not utilised state-of-the art processing techniques that could be applied
[232] or direct detection as part of virus electrodes [233]. Recently, for the encapsulation of phage to achieve controlled dosage forms in
Richter et al. [234] applied an electrical potential to align E. coli phage uniform micro- and nanoparticles. Membrane emulsification (ME) is
T4 and drive them to the electrode surface resulting in head down, tail- one such process that allows formation of uniform drops by injecting a
126
Fig. 4. Morphologies of complex drops: (a) Conventional

W /O/W multiple emulsion drop; (b) multiple emulsion drops generated
1 2
in microfluidic channels with controlled number of inner drops
W
1 [246] and distinct inner drops [247]; (c) drops generated in
W microfluidic channels with multiple concentric shells [249];
2
(d) Janus and ternary drops generated in microfluidic channels
O [250,251].
(a) (b)
Janus Ternary drop

drop
Janus core Janus shell
(c) (d)
dispersed phase liquid through a microporous membrane into the into contact with one another. This is followed by conversion of the
continuous phase (Fig. 3 a). Alternatively, a pre-emulsified mixture of generated drops into solid particles (using techniques discussed earlier).
the dispersed and continuous phase may be repeatedly injected through The variation of drop sizes in microfluidic devices is negligible (< 3%)
the membrane (Fig. 3 b). To encapsulate bacteriophage, the formed and encapsulation efficiency of bacteriophage can reach 100%. The
drops, typically composed of a mixture of the wall forming materials, drop productivity can exceed 10,000 drops/s [245], however, the flow
solvent(s) and phage, could be solidified under mild agitation using rate of the dispersed phase is modest (0.01–10 ml h− 1/drop generation
various solidification reactions or processes, such as free-radical poly- unit - DGU). The most common planar DGUs are co-flow drop makers in
merisation, polycondensation, ionotropic/thermal gelation, cooling which immiscible fluids meet in parallel streams, cross-flow drop ma-
crystallization, and molecular or particle self-assembly triggered by kers or microfluidic junctions in which the immiscible fluid streams
solvent evaporation [236]. The advantages of ME over standard meet at an angle to one another, and flow-focusing drop makers in
emulsification procedures using high-pressure valve homogenisers or which there is a geometric element, such as a constriction, that causes
rotor-stator devices are in higher drop size uniformity and lower energy the streams to accelerate, narrowing the inner fluid thread and causing
inputs and applied shear, which can be useful to preserve phage in- it to break into drops [243].
tegrity. It is well known that bacteriophages are sensitive to mechanical Microfluidic emulsification allows for precise fabrication of struc-
shearing [104,128]. In direct ME, the shear rate on the membrane tured multiple emulsions with controlled drop morphology.
surface is ~103–104 s− 1 however, uniform drops may be obtained Conventional multiple emulsions are composed of numerous drops of
without any shear by spontaneous droplet formation driven by Laplace the inner phase dispersed within larger drops of the middle phase,
pressure gradients [237]. The shear rate in high-shear mixers and col- which itself is dispersed in an outer phase (Fig. 4 a). The middle phase
loid mills is typically ~105 s− 1 and can exceed 107 s− 1 in micro- must be immiscible with both the inner and outer phase. A W1/O/W2
fluidizers. The most common membranes used in ME are Shirasu Porous emulsion consists of drops of the inner water phase W1 dispersed within
Glass (SPG) (Fig. 3 c) and microsieve metallic membranes [238]. Shear the oil drops, which are then dispersed in the outer water phase W2.
on the membrane surface required for drop detachment can be gener- Bacteriophage typically would be suspended in W1 phase and the oil
ated in various ways including; (i) using a paddle stirrer placed above phase would contain a suspended or dissolved encapsulating shell
the membrane surface (Fig. 3 d) [239]; (ii) rotating membrane [240] material such as a polymer. Microfluidic methods can be used to pro-
and (iii) oscillating membrane [241]. In the oscillating ME system, the duce multiple emulsion drops consisting of a controlled number of
tubular membrane can oscillate tangentially clockwise or counter- inner drops encapsulated within each large drop (Fig. 4 b) [246]. The
clockwise [242] or radially upward and downward (Fig. 3 e) [241], number and content of individual inner drops can be controlled thereby
with frequencies ranging from 10 to 90 Hz. allowing the loading of bacteriophage to be precisely manipulated
[246,247]. The distinct inner drops can be used to achieve simulta-
neous encapsulation of two or more types of bacteriophage in separate
5.3. Phage encapsulation using microfluidic emulsification
microenvironments and to control the loading for each of them. Thus
phage cocktails may be individually formulated to overcome the lim-
Bacteriophage containing microcapsules could be fabricated using
itations of one formulation having to suit all.
microfluidic devices with unprecedented uniformity and control over
Microfluidic methods can be used to prepare multiple emulsion
the size, shape and morphology [243,244]. The process involves crea-
drops with concentric onion-like shells around the core drop (Fig. 4 c).
tion of drops in a variety of geometries consisting of micrometre length
Droplets with different numbers of shells may be synthesised depending
scale channels designed to bring two or more immiscible fluid streams
127
of the number of immiscible phases within each drop e.g. double (W1/ 6. Conclusions
O/W2 or O1/W/O2), triple (W1/O2/W3/O4 or O1/W2/O3/W4), quad-
ruple (W1/O2/W3/O4/W5 or O1/W2/O3/W4/O5), and quintuple (W1/ Implementation of phage therapy necessitates the use of phage
O2/W3/O4/W5/O6 or O1/W2/O3/W4/O5/W6) [248]. High-order mul- cocktails to overcome the limited host range of an individual phage and
tiple emulsions may be useful in the production of complex capsules for the risk of phage resistant bacterial mutants. High doses of phage need
co-encapsulation and simultaneous or sequential release of several to be reliably delivered at the site of infection. Repeated frequent
multi-component actives (Fig. 4 c) [249]. Multiple shells would allow dosing of phage may be impractical hence, encapsulation of phage in
tailored formulations within a single capsule e.g. an outer acid stable sustained release systems is needed. Bacteriophage susceptibility to
shell may protect phage from the stomach acidity, an inner burst re- chemical and physical stresses differs among phage families and also
lease shell for bolus delivery of phage with an innermost sustained within a family. Careful consideration needs to be given to ensure the
release shell resulting in slow sustained phage release over a significant formulation chemistry and processing conditions are suited to phage
time period negating the need for repeated phage dosing. making-up the cocktail in order to ensure reproducible retention of high
Other approaches might include phage encapsulation in Janus phage titres and stability during long term storage. Further research is
particles composed of two hemispheres with various physical and urgently needed in the field of phage formulation and encapsulation to
chemical properties (Fig. 4 d). Janus drops typically have two distinct adequately support any future phage therapy field.
regions of roughly equal surface area, however, the term has expanded
to include all multi-segment droplet structures with regions of different Acknowledgement
composition co-existing in asymmetric geometries [250]. Janus parti-
cles may be used to encapsulate bacteriophage in different polymers We would like to acknowledge EPSRC support (Grant no. EP/
having different in vivo degradation rates to achieve burst release fol- M027341/1) Tackling Antimicrobial Resistance: An Interdisciplinary
lowed by controlled sustained release. Approach.
Appendix A. Mathematical model of bacteria/phage interaction
dS
dt
= g⋅S − a⋅S⋅(P1 + P2)
dR
dt
= g⋅R + m⋅g⋅S − a⋅R⋅P2
dP1
dt
= b⋅a⋅ST ⋅P1T − n⋅a⋅S⋅P1 − d⋅P1
dP2
dt
= b⋅a⋅(S + R)T ⋅P2T − n⋅a⋅(S + R)⋅P2 − d⋅P2 (A1)
Here
S, R - concentrations of susceptible and resistant bacteria accordingly [CFU/ml],

P1 - concentration of the phage affecting susceptible bacteria only [PFU/ml],
P2 - concentration of the phage affecting both susceptible and resistant bacteria [PFU/ml],
T - latent period [min]; index ‘T’ indicates the value of the variable T minutes in the past,
g - bacteria growth rate coefficient [min− 1],
a - (bacteria/phage) adsorption coefficient [ml min− 1 PFU− 1],
n - average number of phages adsorbed by an individual bacterium,
b - burst size,
m - probability of bacteria mutation resulting in resistance to phage P1,
d - phage death/removal coefficient [min− 1].
For both figures in the paper the following values were used: g = 0.02 min− 1, n = 1, b = 30, m = 10− 5, d = 0.02 min− 1.
For the results shown in Fig. 1: a = 10− 10 ml/min, T = 40 min.
For the results shown in Fig. 2: a = 0.5 ∗ 10− 10 ml/min, T = 60 min.
Also, for any given latency period T a discrete distribution in the range 0.76T–1.24T was used instead of the fixed T. It was T for 48% of infected
bacteria, 0.88T for 20%, 1.12T for 20%, 0.76T for 6% and 1.24T for 6% - as the figure below illustrates for T = 30 min.
50
40
30
%
20
10
0
0 10 20 30 40
T, min
128
References bacteriophage therapy on pseudomonas aeruginosa cystic fibrosis strains: first

steps towards treatment and prevention. PLoS One 2011;6. http://dx.doi.org/10.
1371/journal.pone.0016963.
[1] O’Neill J. Tackling Drug-Resistant Infections Globally: Final Report and [29] Capparelli R, Parlato M, Borriello G, Salvatore P, Iannelli D. Experimental phage
Recommendations. Rev Antimicrob Resist 2016:1–84. http://dx.doi.org/10.1016/ therapy against Staphylococcus aureus in mice. Antimicrob Agents Chemother
j.jpha.2015.11.005. 2007;51:2765–73. http://dx.doi.org/10.1128/AAC.01513-06.
[2] Moellering RC. N Engl J Med 2010;363:2377–9. http://dx.doi.org/10.1056/ [30] Smith HW, Huggins MB. Successful treatment of experimental Escherichia coli in-
NEJMp1002530. fections in mice using phage: its general superiority over antibiotics. Microbiology
[3] Liu YY, Wang Y, Walsh TR, Yi LX, Zhang R, Spencer J, et al. Emergence of plasmid- 1982;128:307–18. http://dx.doi.org/10.1099/00221287-128-2-307.
mediated colistin resistance mechanism MCR-1 in animals and human beings in [31] Singla S, Harjai K, Katare OP, Chhibber S. Bacteriophage-loaded nanostructured
China: a microbiological and molecular biological study. Lancet Infect Dis lipid carrier: improved pharmacokinetics mediates effective resolution of Klebsiella
2016;16:161–8. http://dx.doi.org/10.1016/S1473-3099(15)00424-7. pneumoniae-induced lobar pneumonia. J Infect Dis 2015;212:325–34. http://dx.
[4] Coates AR, Halls G, Hu Y. Novel classes of antibiotics or more of the same? Br J doi.org/10.1093/infdis/jiv029.
Pharmacol 2011;163:184–94. http://dx.doi.org/10.1111/j.1476-5381.2011. [32] Chhibber S, Kaur S, Kumari S. Therapeutic potential of bacteriophage in treating
01250.x. Klebsiella pneumoniae B5055-mediated lobar pneumonia in mice. J Med Microbiol
[5] Czaplewski L, Bax R, Clokie M, Dawson M, Fairhead H, Fischetti VA, et al. 2008;57:1508–13. http://dx.doi.org/10.1099/jmm.0.2008/002873-0.
Alternatives to antibiotics — a pipeline portfolio review. Lancet Infect Dis [33] Wills QF, Kerrigan C, Soothill JS. Experimental bacteriophage protection against
2016;16:239–51. http://dx.doi.org/10.1016/S1473-3099(15)00466-1. Staphylococcus aureus abscesses in a rabbit model experimental bacteriophage
[6] Wittebole X, De Roock S, Opal SM. A historical overview of bacteriophage therapy protection against Staphylococcus aureus abscesses in a rabbit model. Antimicrob
as an alternative to antibiotics for the treatment of bacterial pathogens. Virulence Agents Chemother 2005;49:1220–1. http://dx.doi.org/10.1128/AAC.49.3.1220.
2014;5:226–35. http://dx.doi.org/10.4161/viru.25991. [34] McVay CS, Velásquez M, Fralick JA. Phage therapy of Pseudomonas aeruginosa
[7] Young R, Gill JJ. Phage therapy redux — what is to be done? Science infection in a mouse burn wound model. Antimicrob Agents Chemother
2015;350:1163–4. http://dx.doi.org/10.1126/science.aad8222. 2007;51:1934–8. http://dx.doi.org/10.1128/AAC.01028-06.
[8] Levin BR, Bull JJ. Population and evolutionary dynamics of phage therapy. Nat [35] Sarker SA, Sultana S, Reuteler G, Moine D, Descombes P, Charton F, et al. Oral
Rev Microbiol 2004;2:166–73. http://dx.doi.org/10.1038/nrmicro822. phage therapy of acute bacterial diarrhea with two coliphage preparations: a
[9] Cairns BJ, Timms AR, Jansen VAA, Connerton IF, Payne RJH. Quantitative models randomized trial in children from Bangladesh. EBioMedicine 2016;4:124–37.
of in vitro bacteriophage-host dynamics and their application to phage therapy. http://dx.doi.org/10.1016/j.ebiom.2015.12.023.
PLoS Pathog 2009;5:1–10. http://dx.doi.org/10.1371/journal.ppat.1000253. [36] Donlan RM, Costerton JW. Biofilms: survival mechanisms of clinically relevant
[10] Bull JJ, Gill JJ. The habits of highly effective phages: population dynamics as a microorganisms. Clin Microbiol Rev. 2002;15:167–93. http://dx.doi.org/10.1128/
framework for identifying therapeutic phages. Front Microbiol 2014;5:1–12. CMR.15.2.167.
http://dx.doi.org/10.3389/fmicb.2014.00618. [37] Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the natural
[11] Hadas H, Einav M, Fishov I, Zaritsky A. bateriophage T4 development depends on environment to infectious diseases. Nat Rev Microbiol 2004;2:95–108.
the physiology of its host Escherichia coli. Microbiology 1997;143:179–85. [38] Peters BM, Jabra-Rizk MA, O'May GA, William Costerton J, Shirtliff ME.
[12] Gill JJ, Hyman P. Phage choice, isolation and preparation for phage therapy. Curr Polymicrobial interactions: impact on pathogenesis and human disease. Clin
Pharm Biotechnol 2010;11:2014. Microbiol Rev 2012;25:193–213. http://dx.doi.org/10.1128/CMR.00013-11.
[13] Weber-Dabrowska B, Jonczyk-Matysiak E, Zaczek M, Lobocka M, Lusiak- [39] Singla S, Harjai K, Raza K, Wadhwa S, Katare OP, Chhibber S. Phospholipid ve-
Szelachowska M, Gorski A. Bacteriophage procurement for therapeutic purposes. sicles encapsulated bacteriophage: a novel approach to enhance phage biodis-
Front Microbiol 2016;7:1–14. http://dx.doi.org/10.3389/fmicb.2016.01177. tribution. J Virol Methods 2016;236:68–76. http://dx.doi.org/10.1016/j.jviromet.
[14] Mattila S, Ruotsalainen P, Jalasvuori M. On-demand isolation of bacteriophages 2016.07.002.
against drug-resistant bacteria for personalized phage therapy. Front Microbiol [40] Chan BK, Abedon ST. Bacteriophage and their enzymes in biofilm control. Curr
2015;6:1–7. http://dx.doi.org/10.3389/fmicb.2015.01271. Pharm Des 2015;21:85–99.
[15] Sutherland IW, Hughes KA, Skillman LC, Tait K. The interaction of phage and [41] Furusawa T, Iwano H, Hiyashimizu Y, Matsubara K, Higuchi H, Nagahata H, et al.
biofilms. FEMS Microbiol Lett 2004;232:1–6. http://dx.doi.org/10.1016/S0378- Phage Therapy Is Effective in a Mouse Model of Bacterial Equine Keratitis. Appl
1097(04)00041-2. Environ Microbiol 2016;82:5332–9. http://dx.doi.org/10.1128/aem.01166-16.
[16] Cornelissen A, Ceyssens PJ, T'Syen J, van Praet H, Noben JP, Shaburova OV, et al. [42] Carmody LA, Gill JJ, Summer EJ, Sajjan US, Gonzalez CF, Young RF, et al. Efficacy
The t7-related pseudomonas putida phage φ15 displays virion-associated biofilm of bacteriophage therapy in a model of Burkholderia cenocepacia pulmonary in-
degradation properties. PLoS One 2011;6. http://dx.doi.org/10.1371/journal. fection. J Infect Dis 2010;201:264–71. http://dx.doi.org/10.1086/649227.
pone.0018597. [43] Martin RM, Cao J, Brisse S, Passet V, Wu W, Zhao L, et al. Molecular epidemiology
[17] Pires DP, Oliveira H, Melo LDR, Sillankorva S, Azeredo J. Bacteriophage-encoded of colonizing and infecting isolates of Klebsiella pneumoniae. mSphere 2016;1:1–12.
depolymerases: their diversity and biotechnological applications. Appl Microbiol http://dx.doi.org/10.1128/mSphere.00261-16.
Biotechnol 2016;100:2141–51. http://dx.doi.org/10.1007/s00253-015-7247-0. [44] WHO. Global tuberculosis report. 2016; 2016.
[18] Gutiérrez D, Vandenheuvel D, Martínez B, Rodríguez A, Lavigne R, García P. Two [45] Jassal M, Bishai WR. Extensively drug-resistant tuberculosis. Lancet Infect Dis
phages, phiIPLA-RODI and phiIPLA-C1C, lyse mono- and dual-species staphylo- 2009;9:19–30. http://dx.doi.org/10.1016/S1473-3099(08)70260-3.
coccal biofilms. Appl Environ Microbiol 2015;81:3336–48. http://dx.doi.org/10. [46] Russell DG, Barry CE, Flynn JL. Tuberculosis: what we don't know can, and does,
1128/AEM.03560-14. hurt us. Science (80- ) 2010;328:852–6.
[19] Khalifa L, Brosh Y, Gelman D, Coppenhagen-Glazer S, Beyth S, Poradosu-Cohen R, [47] Nieth A, Verseux C, Barnert S, Süss R, Römer W. A first step toward liposome-
et al. Targeting Enterococcus faecalis biofilms with phage therapy. Appl Environ mediated intracellular bacteriophage therapy. Expert Opin Drug Deliv 2015:1–14.
Microbiol 2015;81:2696–705. http://dx.doi.org/10.1128/AEM.00096-15. http://dx.doi.org/10.1517/17425247.2015.1043125.
[20] Smith HW, Huggins MB, Shaw KM. The control of experimental Escherichia coli [48] Nieth A, Verseux C, Römer W. A question of attire: dressing up bacteriophage
diarrhoea in calves by means of bacteriophages. J Gen Microbiol therapy for the battle against antibiotic-resistant intracellular bacteria. Springer
1987;133:1111–26. http://dx.doi.org/10.1099/00221287-133-5-1111. Sci Rev 2015;3:1–11. http://dx.doi.org/10.1007/s40362-014-0027-x.
[21] Nakai T, Sugimoto R, Park K-H, Matsuoka S, Mori K, Nishioka T, et al. Protective [49] Guest JF, Ayoub N, Mcilwraith T, Uchegbu I, Gerrish A, Weidlich D, et al. Health
effects of bacteriophage on experimental Lactococcus garvieae infection in yel- economic burden that different wound types impose on the UK's National Health
lowtail. Dis Aquat Organ 1999;37:33–41. Service. BMJ Open 2015;5:e009283. http://dx.doi.org/10.1111/iwj.12603.
[22] Smith HW, Huggins MB. Effectiveness of phages in treating experimental [50] Lindholm C, Searle R. Wound management for the 21st century: combining ef-
Escherichia coli diarrhoea in calves, piglets and lambs. J Gen Microbiol fectiveness and efficiency. Int Wound J 2016;13:5–15. http://dx.doi.org/10.1111/
1983;129:2659–75. http://dx.doi.org/10.1099/00221287-129-8-2659. iwj.12623.
[23] Atterbury RJ, Van Bergen MAP, Ortiz F, Lovell MA, Harris JA, De Boer A, et al. [51] Clinton A, Carter T. Chronic wound biofilms: pathogenesis and potential therapies.
Bacteriophage therapy to reduce Salmonella colonization of broiler chickens. Appl Lab Med 2015;46:277–84.
Environ Microbiol 2007;73:4543–9. http://dx.doi.org/10.1128/AEM.00049-07. [52] Wiegand C, Abel M, Ruth P, Hipler UC. HaCaT keratinocytes in co-culture with
[24] Peltomaa R, López-Perolio I, Benito-Peña E, Barderas R, Moreno-Bondi MC. Staphylococcus aureus can be protected from bacterial damage by polihexanide.
Application of bacteriophages in sensor development. Anal Bioanal Chem Wound Repair Regen 2009;17:730–8. http://dx.doi.org/10.1111/j.1524-475X.
2016;408:1805–28. http://dx.doi.org/10.1007/s00216-015-9087-2. 2009.00536.x.
[25] Ulett GC, Totsika M, Schaale K, Carey AJ, Sweet MJ, Schembri MA. Uropathogenic [53] Peterson LR. Review. Squeezing the antibiotic balloon: the impact of antimicrobial
Escherichia coli virulence and innate immune responses during urinary tract in- classes on emerging resistance. Clin Microbiol Infect Suppl 2005;11:4–16. http://
fection. Curr Opin Microbiol 2013;16:100–7. http://dx.doi.org/10.1016/j.mib. dx.doi.org/10.1111/j.1469-0691.2005.01238.x.
2013.01.005. [54] Lipsky BA, Berendt AR, Cornia PB, Pile JC, Peters EJG, Armstrong DG, et al. 2012
[26] Debarbieux L, Leduc D, Maura D, Morello E, Criscuolo A, Grossi O, et al. Infectious Diseases Society of America clinical practice guideline for the diagnosis
Bacteriophages can treat and prevent Pseudomonas aeruginosa lung infections. J and treatment of diabetic foot infections. Clin Infect Dis 2012;54:132–73. http://
Infect Dis 2010;201:1096–104. http://dx.doi.org/10.1086/651135. dx.doi.org/10.1093/cid/cis346.
[27] Alemayehu D, Casey PG, Mcauliffe O, Guinane CM, Martin JG, Shanahan F, et al. [55] Nale JY, Shan J, Hickenbotham PT, Fawley WN, Wilcox MH, Clokie MRJ. Diverse
Bacteriophage phiMR299-2 and phiNH-4 can eliminate Psuedomonas aeruginosa in temperate bacteriophage carriage in Clostridium difficile 027 strains. PLoS One
the murine lung and on cystic fibrosis lung airway. MBio 2012;3:1–9. http://dx. 2012;7:1–9. http://dx.doi.org/10.1371/journal.pone.0037263.
doi.org/10.1128/mBio.00029-12.Editor. [56] Curtis V, Cairncross S. Effect of washing hands with soap on diarrhoea risk in the
[28] Morello E, Saussereau E, Maura D, Huerre M, Touqui L, Debarbieux L. Pulmonary community: a systematic review. Lancet Infect Dis 2003;3:275–81.
129
[57] Abdulamir AS, Sabah J, Bakar Fatimah A. Novel approach of using a cocktail of 1128/AAC.48.7.2558.
designed bacteriophages against gut pathogenic E. coli for bacterial load biocon- [84] Delbruck M. The growth of bacteriophage and lysis of the host. J Gen Physiol
trol. Ann Clin Microbiol Antimicrob 2014;13:1–11. 1940;23:643–60.
[58] Yen M, Cairns LS, Camilli A, Zuckerman JN, Rombo L, Fisch A, et al. A cocktail of [85] Abedon ST. Lysis from without. Bacteriophage 2011;1:46–9. http://dx.doi.org/10.
three virulent bacteriophages prevents Vibrio cholerae infection in animal models. 4161/bact.1.1.13980.
Nat Commun 2017;8:14187. http://dx.doi.org/10.1038/ncomms14187. [86] Denou E, Bruttin A, Barretto C, Ngom-Bru C, Brüssow H, Zuber S. T4 phages
[59] Wall SK, Zhang J, Rostagno MH, Ebner PD. Phage therapy to reduce preprocessing against Escherichia coli diarrhea: potential and problems. Virology
Salmonella infections in market-weight swine. Appl Environ Microbiol 2009;388:21–30. http://dx.doi.org/10.1016/j.virol.2009.03.009.
2010;76:48–53. http://dx.doi.org/10.1128/AEM.00785-09. [87] Schmerer M, Molineux IJ, Bull JJ. Synergy as a rationale for phage therapy using
[60] Oliveira M, Viñas I, Colàs P, Anguera M, Usall J, Abadias M. Effectiveness of a phage cocktails. PeerJ 2014;2:e590. http://dx.doi.org/10.7717/peerj.590.
bacteriophage in reducing Listeria monocytogenes on fresh-cut fruits and fruit [88] Vandenheuvel D, Lavigne R, Brussow H. Bacteriophage therapy: advances in for-
juices. Food Microbiol 2014;38:137–42. http://dx.doi.org/10.1016/j.fm.2013.08. umulation strategies and human clinical trials. Annu Rev Virol 2015;2:599–618.
018. [89] Kucharewicz-Krukowska A, Slopek S. Immunogenic effect of bactriophage in pa-
[61] Hua Zou WW. Phage therapy: promising for H. pylori infection. Clin Microbiol tients subjected to phage therapy. Arch Immunol Ther Exp (Warsz)
Open Access 2013;2:10–1. http://dx.doi.org/10.4172/2327-5073.1000112. 1987;35:553–61.
[62] Kotloff KL, Nataro JP, Blackwelder WC, Nasrin D, Farag TH, Panchalingam S, et al. [90] Huff WE, Huff GR, Rath NC, Balog JM, Donoghue AM. Prevention of Escherichia
Burden and aetiology of diarrhoeal disease in infants and young children in de- coli infection in broiler chickens with a bacteriophage aerosol spray. Poult Sci
veloping countries (the Global Enteric Multicenter Study, GEMS): a prospective, 2002;81:1486–91. http://dx.doi.org/10.1093/ps/81.10.1486.
case-control study. Lancet 2013;382:209–22. http://dx.doi.org/10.1016/S0140- [91] Semler DD, Goudie AD, Finlay WH, Dennis JJ. Aerosol phage therapy efficacy in
6736(13)60844-2. Burkholderia cepacia complex respiratory infections. Antimicrob Agents Chemother
[63] Payne RJH, Jansen V. Understanding bacteriophage therapy as a density-depen- 2014;58:4005–13. http://dx.doi.org/10.1128/AAC.02388-13.
dent kinetic process. J Theor Biol 2001;208:37–48. http://dx.doi.org/10.1006/ [92] Oliveira A, Sereno R, Nicolau A, Azeredo J. The influence of the mode of admin-
jtbi.2000.2198. istration in the dissemination of three coliphages in chickens. Poult Sci
[64] Tanji Y, Shimada T, Fukudomi H, Miyanaga K, Nakai Y, Unno H. Therapeutic use 2009;88:728–33. http://dx.doi.org/10.3382/ps.2008-00378.
of phage cocktail for controlling Escherichia coli O157:H7 in gastrointestinal tract [93] Cano V, March C, Insua JL, Aguiló N, Llobet E, Moranta D, et al. Klebsiella pneu-
of mice. J Biosci Bioeng 2005;100:280–7. http://dx.doi.org/10.1263/jbb.100.280. moniae survives within macrophages by avoiding delivery to lysosomes. Cell
[65] Leung CY, Weitz JS. Synergistic elimination of bacteria by phage and the immune Microbiol 2015;17:1537–60. http://dx.doi.org/10.1111/cmi.12466.
system. bioRxiv 2016:57927. http://dx.doi.org/10.1101/057927. [94] Fumagalli O, Tall BD, Schipper C, Oelschlaeger TA. N-glycosylated proteins are
[66] Smith HW, Huggins MB. Successful Treatment of Experimental Escherichia coli involved in efficient internalization of Klebsiella pneumoniae by cultured human
Infections in Mice Using Phage: its General Superiority over Antibiotics. J Gen epithelial cells. Infect Immun 1997;65:4445–51.
Microbiol 1982;128:307–18. http://dx.doi.org/10.1099/00221287-128-2-307. [95] Singla S, Harjai K, Katare OP, Chhibber S. Encapsulation of bacteriophage in li-
[67] Biswas B, Adhya S, Washart P, Paul B, Trostel AN, Powell B, et al. Bacteriophage posome accentuates its entry in to macrophage and shields it from neutralizing
therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant antibodies. PLoS One 2016;11:e0153777. http://dx.doi.org/10.1371/journal.
Enterococcus faecium. Infect Immun 2002;70:204–10. http://dx.doi.org/10.1128/ pone.0153777.
IAI.70.1.204. [96] Fish R, Kutter E, Wheat G, Blasdel B, Kutateladze M, Kuhl S. Bacteriophage
[68] Cerveny KE, DePaola A, Duckworth DH, Gulig PA. Phage therapy of local and treatment of intransigent diabetic toe ulcers: a case series. J Wound Care
systemic disease caused by Vibrio vulnificus in iron-dextran-treated mice. Infect 2016;25:S27–33.
Immun 2002;70:6251–62. http://dx.doi.org/10.1128/IAI.70.11.6251. [97] Wright A, Hawkins CH, Änggård EE, Harper DR. A controlled clinical trial of a
[69] Barrow P, Lovell M. Use of lytic bacteriophage for control of experimental therapeutic bacteriophage preparation in chronic otitis due to antibiotic-resistant
Escherichia coli septicemia and meningitis in chickens and calves. Clin Diagn Lab Pseudomonas aeruginosa; a preliminary report of efficacy. Clin Otolaryngol
Immunol 1998;5:294–8. 2009;34:349–57. http://dx.doi.org/10.1111/j.1749-4486.2009.01973.x.
[70] Watanabe R, Matsumoto T, Sano G, Ishii Y, Tateda K, Sumiyama Y, et al. Efficacy [98] Merabishvili M, Pirnay J, Verbeken G, Chanishvili N, Tediashvili M, Lashkhi N,
of bacteriophage therapy against gut-derived sepsis caused by Pseudomonas aeru- et al. Quality-controlled small-scale production of a well-defined bacteriophage
ginosa in mice. Antimicrob Agents Chemother 2007;51:446–52. http://dx.doi.org/ cocktail for use in human clinical trials. PLoS One 2009;4:1–10.
10.1128/AAC.00635-06. [99] Rose T, Verbeken G, De Vos D, Merabishvili M, Vaneechoutte M, Jennes S, et al.
[71] Matsuzaki S, Yasuda M, Nishikawa H, Kuroda M, Ujihara T, Shuin T, et al. Experimental phage therapy of burn wound infection: difficult first steps. Int J
Experimental protection of mice against lethal Staphylococcus aureus infection by Burn Trauma 2014;4:66–73.
novel bacteriophage phi MR11. J Infect Dis 2003;187:613–24. http://dx.doi.org/ [100] Puapermpoonsiri U, Spencer J, van der Walle CF. A freeze-dried formulation of
10.1086/374001. bacteriophage encapsulated in biodegradable microspheres. Eur J Pharm
[72] Tsonos J, Vandenheuvel D, Briers Y, De Greve H, Hernalsteens JP, Lavigne R. Biopharm 2009;72:26–33. http://dx.doi.org/10.1016/j.ejpb.2008.12.001.
Hurdles in bacteriophage therapy: Deconstructing the parameters. Vet Microbiol [101] Lee SW, Belcher AM. Virus-based fabrication of micro- and nanofibers using
2014;171:460–9. http://dx.doi.org/10.1016/j.vetmic.2013.11.001. electrospinning. Nano Lett 2004;4:387–90. http://dx.doi.org/10.1021/nl034911t.
[73] Wagenaar JA, Bergen MAPV, Mueller MA, Wassenaar TM, Carlton RM. Phage [102] Briers Y, Miroshnikov K, Chertkov O, Nekrasov A, Mesyanzhinov V, Volckaert G,
therapy reduces Campylobacter jejuni colonization in broilers. Vet Microbiol et al. The structural peptidoglycan hydrolase gp181 of bacteriophage phiKZ.
2005;109:275–83. http://dx.doi.org/10.1016/j.vetmic.2005.06.002. Biochem Biophys Res Commun 2008;374:747–51. http://dx.doi.org/10.1016/j.
[74] Andreatti Filho RL, Higgins JP, Higgins SE, Gaona G, Wolfenden AD, Tellez G, bbrc.2008.07.102.
et al. Ability of bacteriophages isolated from different sources to reduce Salmonella [103] Knezevic P, Obreht D, Curcin S, Petrusic M, Aleksic V, Kostanjsek R, et al. Phages
enterica serovar enteritidis in vitro and in vivo. Poult Sci 2007;86:1904–9. of Pseudomonas aeruginosa: response to environmental factors and in vitro ability
[75] Stanford K, Niu YD, Johnson R. Oral delivery systems for encapsulated bacter- to inhibit bacterial growth and biofilm formation. J Appl Microbiol
iophages targeted at Escherichia coli O157:H7 in feedlot cattle. J Food Prot 2011;111:245–54. http://dx.doi.org/10.1111/j.1365-2672.2011.05043.x.
2010;73:1304–12. [104] Leung SSY, Parumasivam T, Gao FG, Carrigy NB, Vehring R, Finlay WH, et al.
[76] Nale JY, Spencer J, Hargreaves KR, Buckley AM, Trzepin P, Douce GR, et al. Production of inhalation phage powders using spray freeze drying and spray
Bacteriophage combinations significantly reduce Clostridium difficile growth in drying techniques for treatment of respiratory infections. Pharm Res
vitro and proliferation in vivo. Antimicrob Agents Chemother 2016;60:968–81. 2016;33:1486–96. http://dx.doi.org/10.1007/s11095-016-1892-6.
http://dx.doi.org/10.1128/AAC.01774-15.Address. [105] Dini C, de Urraza PJ. Effect of buffer systems and disaccharides concentration on
[77] Colom J, Cano-Sarabia M, Otero J, Cortés P, Maspoch D, Llagostera M. Liposome- Podoviridae coliphage stability during freeze drying and storage. Cryobiology
encapsulated bacteriophages for enhanced oral phage therapy against Salmonella 2013;66:339–42. http://dx.doi.org/10.1016/j.cryobiol.2013.03.007.
spp. Appl Environ Microbiol 2015;81:4841–9. http://dx.doi.org/10.1128/AEM. [106] Korehei R, Kadla J. Incorporation of T4 bacteriophage in electrospun fibres. J Appl
00812-15. Microbiol 2013;114:1425–34. http://dx.doi.org/10.1111/jam.12158.
[78] Colom J, Cano-Sarabia M, Otero J, Aríñez-Soriano J, Cortés P, Maspoch D, et al. [107] Clark WA. Comparison of several methods for preserving bacteriophages. Appl
Microencapsulation with alginate/CaCO3: a strategy for improved phage therapy. Environ Microbiol 1962;10:466–71.
Sci Rep 2017;7:41441. http://dx.doi.org/10.1038/srep41441. [108] Davies JD, Kelly MJ. The preservation of bacteriophage H1 of Corynebacterium
[79] Zhang JY, Kraft BL, Pan YY, Wall SK, Saez AC, Ebner PD. Development of an Anti- ulcerans U103 by freeze-drying. J Hyg (Lond) 1969;67:573–83. http://dx.doi.org/
Salmonella Phage Cocktail with Increased Host Range. Foodborne Pathog Dis 10.1017/S0022172400042030.
2010;7:1415–9. http://dx.doi.org/10.1089/fpd.2010.0621. [109] Ackermann H-W, Tremblay D, Moineau S. Long-term bacteriophage preservation.
[80] Bardina C, Spricigo DA, Cortés P, Llagostera M. Significance of the bacteriophage WFCC Newsl 2004;38:35–40.
treatment schedule in reducing salmonella colonization of poultry. Appl Environ [110] Wang W. Lyophilization and development of solid protein pharmaceuticals. vol.
Microbiol 2012;78:6600–7. http://dx.doi.org/10.1128/AEM.01257-12. 203. 2000; 2000. http://dx.doi.org/10.1016/S0378-5173(00)00423-3.
[81] Teixeira AP, Oliveira R, Alves PM, Carrondo MJT. Advances in on-line monitoring [111] Abdelwahed W, Degobert G, Stainmesse S, Fessi H. Freeze-drying of nanoparticles:
and control of mammalian cell cultures: supporting the PAT initiative. Biotechnol formulation, process and storage considerations. Adv Drug Deliv Rev
Adv 2009;27:726–32. http://dx.doi.org/10.1016/j.biotechadv.2009.05.003. 2006;58:1688–713. http://dx.doi.org/10.1016/j.addr.2006.09.017.
[82] Murthy K, Engelhardt R. US 2009/0130196 A1: bacteriophage composition. 2009; [112] Telko MJ, Hickey AJ. Dry powder inhaler formulation. Respir Care
2009. 2005;50:1209–27. http://dx.doi.org/10.2165/00128413-200615470-00016.
[83] Chibani-Chennoufi S, Sidoti J, Bruttin A, Kutter E, Sarker S, Bru H. In vitro and in [113] Engel HWB, Smith D, Berwald LG. The preservation of mycobacteriophages by
vivo bacteriolytic activities of Escherichia coli phages: implications for phage means of freeze drying. Am Rev Respir Dis 1974;109:561–6.
therapy. Antimicrob Agents Chemother 2004;48:2558–69. http://dx.doi.org/10. [114] Merabishvili M, Vervaet C, Pirnay JP, de Vos D, Verbeken G, Mast J, et al. Stability
130
of Staphylococcus aureus phage ISP after freeze-drying (lyophilization). PLoS One [139] Prouty CC. Storage of the bacteriophage of the lactic acid streptococci in the de-
2013;8:1–7. http://dx.doi.org/10.1371/journal.pone.0068797. siccated state with observations on longevity. Appl Microbiol 1953;1:250–1.
[115] Malenovská H. The influence of stabilizers and rates of freezing on preserving of [140] Ibekwe VC, Fadda HM, McConnell EL, Khela MK, Evans DF, Basit AW. Interplay
structurally different animal viruses during lyophilization and subsequent storage. between intestinal pH, transit time and feed status on the in vivo performance of
J Appl Microbiol 2014;117:1810–9. http://dx.doi.org/10.1111/jam.12654. pH responsive ileo-colonic release systems. Pharm Res 2008;25:1828–35. http://
[116] Golshahi L, Lynch KH, Dennis JJ, Finlay WH. In vitro lung delivery of bacter- dx.doi.org/10.1007/s11095-008-9580-9.
iophages KS4-M and φKZ using dry powder inhalers for treatment of Burkholderia [141] McConnell EL, Fadda HM, Basit AW. Gut instincts: explorations in intestinal
cepacia complex and Pseudomonas aeruginosa infections in cystic fibrosis. J Appl physiology and drug delivery. Int J Pharm 2008;364:213–26. http://dx.doi.org/
Microbiol 2010;110:106–17. http://dx.doi.org/10.1111/j.1365-2672.2010. 10.1016/j.ijpharm.2008.05.012.
04863.x. [142] Esteban PP, Alves DR, Enright MC, Bean JE, Gaudion A, Jenkins ATA, et al.
[117] Puapermpoonsiri U, Ford SJ, van der Walle CF. Stabilization of bacteriophage Enhancement of the antimicrobial properties of bacteriophage-K via stabilization
during freeze drying. Int J Pharm 2010;389:168–75. http://dx.doi.org/10.1016/j. using oil-in-water nano-emulsions. Biotechnol Prog 2014;30:932–44. http://dx.
ijpharm.2010.01.034. doi.org/10.1002/btpr.1898.
[118] Dai M, Senecal A, Nugen SR. Electrospun water-soluble polymer nanofibers for the [143] Ma Y, Pacan JC, Wang Q, Xu Y, Huang X, Korenevsky A, et al. Microencapsulation
dehydration and storage of sensitive reagents. Nanotechnology 2014;25:225101. of bacteriophage Felix O1 into chitosan-alginate microspheres for oral delivery.
http://dx.doi.org/10.1088/0957-4484/25/22/225101. Appl Environ Microbiol 2008;74:4799–805. http://dx.doi.org/10.1128/AEM.
[119] Cox CS, Harris WJ, Lee J. Viability and electron microscope studies of phages T3 00246-08.
and T7 subjected to freeze-drying, freeze-thawing and aerosolization. J Gen [144] Gul O, Dervisoglu M. Application of multicriteria decision technique to determine
Microbiol 1974;81:207–15. optimum sodium alginate concentration for microencapsulation of Lactobacillus
[120] Leung SSY, Parumasivam T, Gao FG, Carrigy NB, Vehring R, Finlay WH, et al. casei Shirota by extrusion and emulsification. J Food Process Eng 2017;40:1–10.
Effects of storage conditions on the stability of spray dried, inhalable bacter- http://dx.doi.org/10.1111/jfpe.12481.
iophage powders. Int J Pharm 2017. http://dx.doi.org/10.1016/j.ijpharm.2017. [145] Ma Y, Pacan JC, Wang Q, Sabour PM, Huang X, Xu Y. Enhanced alginate micro-
01.060. spheres as means of oral delivery of bacteriophage for reducing Staphylococcus
[121] Markoishvili K, Tsitlanadze G, Katsarava R, Morris JG, Sulakvelidze A. A novel aureus intestinal carriage. Food Hydrocoll 2012;26:434–40. http://dx.doi.org/10.
sustained-release matrix based on biodegradable poly(ester amide)s and im- 1016/j.foodhyd.2010.11.017.
pregnated with bacteriophages and an antibiotic shows promise in management of [146] Lotfipour F, Mirzaeei S, Maghsoodi M. Preparation and characterization of algi-
infected venous stasis ulcers and other poorly healing wounds. Int J Dermatol nate and psyllium beads containing Lactobacillus acidophilus.
2002;41:453–8. ScientificWorldJournal 2012;2012:680108. http://dx.doi.org/10.1100/2012/
[122] Dini C, Islan GA, de Urraza PJ, Castro GR. Novel biopolymer matrices for micro- 680108.
encapsulation of phages: enhanced protection against acidity and protease ac- [147] Hathaway H, Alves DR, Bean J, Esteban PP, Ouadi K, Sutton JM, et al. Poly (N-
tivity. Macromol Biosci 2012;12:1200–8. http://dx.doi.org/10.1002/mabi. isopropylacrylamide-co-allylamine) (PNIPAM-co-ALA) nanospheres for the ther-
201200109. mally triggered release of Bacteriophage K. Eur J Pharm Biopharm
[123] Alfadhel M, Puapermpoonsiri U, Ford SJ, McInnes FJ, Van Der Walle CF. 2015;96:437–41. http://dx.doi.org/10.1016/j.ejpb.2015.09.013.
Lyophilized inserts for nasal administration harboring bacteriophage selective for [148] Bean JE, Alves DR, Laabei M, Esteban PP, Thet NT, Enright MC, et al. Triggered
Staphylococcus aureus: in vitro evaluation. Int J Pharm 2011;416:280–7. http://dx. release of bacteriophage K from agarose/hyaluronan hydrogel matrixes by
doi.org/10.1016/j.ijpharm.2011.07.006. Staphylococcus aureus virulence factors. Chem Mater 2014;26:7201–8.
[124] Hoe S, Ivey JW, Boraey MA, Shamsaddini-Shahrbabak A, Javaheri E, Matinkhoo S, [149] Samtlebe M, Ergin F, Wagner N, Neve H, Küçükçetin A, Franz CMAP, et al. Carrier
et al. Use of a fundamental approach to spray-drying formulation design to fa- systems for bacteriophages to supplement food systems: encapsulation and con-
cilitate the development of multi-component dry powder aerosols for respiratory trolled release to modulate the human gut microbiota. LWT- Food Sci Technol
drug delivery. Pharm Res 2014;31:449–65. http://dx.doi.org/10.1007/s11095- 2016;68:334–40. http://dx.doi.org/10.1016/j.lwt.2015.12.039.
013-1174-5. [150] Nieth A, Verseux C, Barnert S, Süss R, Römer W, Nieth A, et al. A first step toward
[125] Chew NYK, Chan HK. The role of particle properties in pharmaceutical powder liposome-mediated intracellular bacteriophage therapy. Expert Opin Drug Deliv
inhalation formulations. J Aerosol Med 2002;15:325–30. 2015;12:1411–24. http://dx.doi.org/10.1517/17425247.2015.1043125.
[126] Klingler C, Müller BW, Steckel H. Insulin-micro- and nanoparticles for pulmonary [151] Kuhn JC, Zussman E. US 2009/0061496A1. Encapsulation of bacteria and viruses
delivery. Int J Pharm 2009;377:173–9. http://dx.doi.org/10.1016/j.ijpharm. in electrospun fibers. 2009; 2009.
2009.05.008. [152] Salalha W, Kuhn J, Dror Y, Zussman E. Encapsulation of bacteria and viruses in
[127] Weers J. Inhaled antimicrobial therapy - barriers to effective treatment. Adv Drug electrospun nanofibres. Nanotechnology 2006;17:4657–81. http://dx.doi.org/10.
Deliv Rev 2015;85:24–43. http://dx.doi.org/10.1016/j.addr.2014.08.013. 1088/0957-4484/17/18/025.
[128] Vandenheuvel D, Singh A, Vandersteegen K, Klumpp J, Lavigne R, Van Den [153] Korehei R, Kadla JF. Encapsulation of T4 bacteriophage in electrospun poly
Mooter G. Feasibility of spray drying bacteriophages into respirable powders to (ethylene oxide)/cellulose diacetate fibers. Carbohydr Polym 2014;100:150–7.
combat pulmonary bacterial infections. Eur J Pharm Biopharm 2013;84:578–82. http://dx.doi.org/10.1016/j.carbpol.2013.03.079.
http://dx.doi.org/10.1016/j.ejpb.2012.12.022. [154] Vonasek E, Le P, Nitin N. Encapsulation of bacteriophages in whey protein films
[129] Jończyk E, Kłak M, Międzybrodzki R, Górski A. The influence of external factors for extended storage and release. Food Hydrocoll 2014;37:7–13. http://dx.doi.
on bacteriophages—review. Folia Microbiol (Praha) 2011;56:191–200. http://dx. org/10.1016/j.foodhyd.2013.09.017.
doi.org/10.1007/s12223-011-0039-8. [155] Lone A, Anany H, Hakeem M, Aguis L, Avdjian AC, Bouget M, et al. Development
[130] Walbeck AK. US20138501453B2: methods for drying bacteriophage and bacter- of prototypes of bioactive packaging materials based on immobilized bacter-
iophage-containing compositions, the resulting dry compositions, and methods of iophages for control of growth of bacterial pathogens in foods. Int J Food
use. 2013; 2013. Microbiol 2016;217:49–58. http://dx.doi.org/10.1016/j.ijfoodmicro.2015.10.
[131] Matinkhoo S, Lynch KH, Dennis JJ, Finlay WH, Vehring R. Spray-dried respirable 011.
powders containing bacteriophages for the treatment of pulmonary infections. J [156] Kaikabo AA, AbdulKarim SM, Abas F. Evaluation of the efficacy of chitosan na-
Pharm Sci 2011;100:5197–205. noparticles loaded FKAZ14 bacteriophage in the biological control of colibacillosis
[132] Vandenheuvel D, Meeus J, Lavigne R, Van Den Mooter G. Instability of bacter- in chickens. Poult Sci 2017;96:295–302. http://dx.doi.org/10.3382/ps/pew255.
iophages in spray-dried trehalose powders is caused by crystallization of the ma- [157] Tang Z, Huang X, Baxi S, Chambers JR, Sabour PM, Wang Q. Whey protein im-
trix. Int J Pharm 2014;472:202–5. http://dx.doi.org/10.1016/j.ijpharm.2014.06. proves survival and release characteristics of bacteriophage Felix O1 encapsulated
026. in alginate microspheres. Food Res Int 2013;52:460–6. http://dx.doi.org/10.
[133] Crowe LM, Reid DS, Crowe JH. Is trehalose special for preserving dry biomater- 1016/j.foodres.2012.12.037.
ials? Biophys J 1996;71:2087–93. http://dx.doi.org/10.1016/S0006-3495(96) [158] Kim S, Jo A, Ahn J. Application of chitosan–alginate microspheres for the sus-
79407-9. tained release of bacteriophage in simulated gastrointestinal conditions. Int J Food
[134] Steckel H, Bolzen N. Alternative sugars as potential carriers for dry powder in- Sci Technol 2015;50:913–8. http://dx.doi.org/10.1111/ijfs.12736.
halations. Int J Pharm 2004;270:297–306. http://dx.doi.org/10.1016/j.ijpharm. [159] Tang Z, Huang X, Sabour PM, Chambers JR, Wang Q. Preparation and char-
2003.10.039. acterization of dry powder bacteriophage K for intestinal delivery through oral
[135] Walbeck AK. US 2009/0093041 A1. Methods for drying bacteriophage and bac- administration. LWT- Food Sci Technol 2015;60:263–70. http://dx.doi.org/10.
teriophage containing compositions, the resulting dry compositions, and methods 1016/j.lwt.2014.08.012.
of use. 2009; 2009. [160] Katsarava R, Alavidze Z. US6703040B2: polymer blends as boidegradable matrices
[136] Wong Y-L, Sampson S, Germishuizen WA, Goonesekera S, Caponetti G, Sadoff J, for preparing biocomposites. 2004; 2004.
et al. Drying a tuberculosis vaccine without freezing. Proc Natl Acad Sci U S A [161] Govender M, Choonara YE, Van Vuuren S, Kumar P, Du Toit LC, Pillay VA. A
2007;104:2591–5. http://dx.doi.org/10.1073/pnas.0611430104. gastro-resistant ovalbumin bi-layered mini-tablet-in-tablet system for the delivery
[137] Sahota JS, Smith CM, Radhakrishnan P, Winstanley C, Goderdzishvili M, of Lactobacillus acidophilus probiotic to simulated human intestinal and colon
Chanishvili N, et al. Bacteriophage delivery by nebulization and efficacy against conditions. J Pharm Pharmacol 2015;67:939–50. http://dx.doi.org/10.1111/jphp.
phenotypically diverse Pseudomonas aeruginosa from cystic fibrosis patients. J 12389.
Aerosol Med Pulm Drug Deliv 2015;28:353–60. http://dx.doi.org/10.1089/jamp. [162] Katsarava R, Beridze V, Arabuli N, Kharadze D, Chu CC, Won CY. Amino Acid-
2014.1172. Based Bioanalogous Polymers. Synthesis, and Study of Regular Poly(ester amide)s
[138] Golshahi L, Seed KD, Dennis JJ, Finlay WH. Toward modern inhalational bacter- Based on Bis(alpha-amino acid) alpha,omega-Alkylene Diesters, and Aliphatic
iophage therapy: nebulization of bacteriophages of Burkholderia cepacia complex. J Dicarboxylic Acids. J Polym Sci: Part A: Polym Chem 1999;37:391–407.
Aerosol Med Pulm Drug Deliv 2008;21:351–60. http://dx.doi.org/10.1089/jamp. [163] Davis SS, Hardy JG, Fara JW. Transit of pharmaceutical dosage forms through the
2008.0701. small intestine. Gut 1986;27:886–92. http://dx.doi.org/10.1136/gut.27.8.886.
131
[164] Davis SS, Illum L, Hinchcliffe M. Gastrointestinal transit of dosage forms in the pig. Release 1998;52:109–18. http://dx.doi.org/10.1016/S0168-3659(97)00203-4.
J Pharm Pharmacol 2001;53:33–9. http://dx.doi.org/10.1211/ [192] Calinescu C, Mateescu MA. Carboxymethyl high amylose starch: chitosan self-
0022357011775163. stabilized matrix for probiotic colon delivery. Eur J Pharm Biopharm
[165] Uriot O, Galia W, Awussi AA, Perrin C, Denis S, Chalancon S, et al. Use of the 2008;70:582–9. http://dx.doi.org/10.1016/j.ejpb.2008.06.006.
dynamic gastro-intestinal model TIM to explore the survival of the yogurt bac- [193] Bernkop-Schnürch A, Humenberger C, Valenta C. Basic studies on bioadhesive
terium Streptococcus thermophilus and the metabolic activities induced in the si- delivery systems for peptide and protein drugs. Int J Pharm 1998;165:217–25.
mulated human gut. Food Microbiol 2016;53:18–29. http://dx.doi.org/10.1016/j. http://dx.doi.org/10.1016/S0378-5173(98)00017-9.
fm.2015.05.007. [194] Thanou M, Verhoef JC, Junginger HE. Chitosan and its derivatives as intestinal
[166] Philip AK, Philip B. Colon targeted drug delivery systems: a review on primary and absorption enhancers. Adv Drug Deliv Rev 2001;50:91–101. http://dx.doi.org/10.
novel approaches. Oman Med J 2010;25:79–87. http://dx.doi.org/10.5001/omj. 1016/S0169-409X(01)00180-6.
2010.24. [195] Van der Lubben IM, Verhoef JC, Borchard G, Junginger HE. Chitosan and its de-
[167] Choonara BF, Choonara YE, Kumar P, Bijukumar D, du Toit LC, Pillay V. A review rivatives in mucosal drug and vaccine delivery. Eur J Pharm Sci 2001;14:201–7.
of advanced oral drug delivery technologies facilitating the protection and ab- http://dx.doi.org/10.1016/S0928-0987(01)00172-5.
sorption of protein and peptide molecules. Biotechnol Adv 2014;32:1269–82. [196] Lameiro MH, Malpique R, Silva AC, Alves PM, Melo E. Encapsulation of adenoviral
http://dx.doi.org/10.1016/j.biotechadv.2014.07.006. vectors into chitosan-bile salt microparticles for mucosal vaccination. J Biotechnol
[168] Amidon S, Brown JE, Dave VS. Colon-targeted oral drug delivery systems: design 2006;126:152–62. http://dx.doi.org/10.1016/j.jbiotec.2006.04.030.
trends and approaches. AAPS PharmSciTech 2015;16:731–41. http://dx.doi.org/ [197] Takeuchi H, Matsui Y, Sugihara H, Yamamoto H, Kawashima Y. Effectiveness of
10.1208/s12249-015-0350-9. submicron-sized, chitosan-coated liposomes in oral administration of peptide
[169] Van Citters GW, Lin HC. Ileal brake: neuropeptidergic control of intestinal transit. drugs. Int J Pharm 2005;303:160–70. http://dx.doi.org/10.1016/j.ijpharm.2005.
Curr Gastroenterol Rep 2006;8:367–73. http://dx.doi.org/10.1007/s11894-006- 06.028.
0021-9. [198] Thirawong N, Thongborisute J, Takeuchi H, Sriamornsak P. Improved intestinal
[170] Hua S, Marks E, Schneider JJ, Keely S. Advances in oral nano-delivery systems for absorption of calcitonin by mucoadhesive delivery of novel pectin-liposome na-
colon targeted drug delivery in inflammatory bowel disease: selective targeting to nocomplexes. J Control Release 2008;125:236–45. http://dx.doi.org/10.1016/j.
diseased versus healthy tissue. Nanomed Nanotechnol Biol Med 2015;11:1117–32. jconrel.2007.10.023.
http://dx.doi.org/10.1016/j.nano.2015.02.018. [199] Shi L, Li Z, Li D, Xu M, Chen H, Zhang Z, et al. Encapsulation of probiotic
[171] Brayden DJ, Jepson MA, Baird AW. Intestinal Peyer's patch M cells and oral Lactobacillus bulgaricus in alginate–milk microspheres and evaluation of the sur-
vaccine targeting. Drug Discov Today 2005;10:1145–57. http://dx.doi.org/10. vival in simulated gastrointestinal conditions. J Food Eng 2013;117:99–104.
1016/S1359-6446(05)03536-1. http://dx.doi.org/10.1016/j.jfoodeng.2013.02.012.
[172] Inman LR, Cantey JR. Specific adherence of Escherichia coli (strain RDEC-1) to [200] Matsubara T, Emoto W, Kawashiro K. A simple two-transition model for loss of
membranous (M) cells of the Peyer's patch in Escherichia coli diarrhea in the rabbit. infectivity of phages on exposure to organic solvent. Biomol Eng 2007;24:269–71.
J Clin Invest 1983;71:1–8. http://dx.doi.org/10.1172/JCI110737. http://dx.doi.org/10.1016/j.bioeng.2007.02.002.
[173] Pappo J, Ermak TH, Steger HJ. Monoclonal antibody-directed targeting of fluor- [201] Olofsson L, Ankarloo J, Andersson PO, Nicholls IA. Filamentous bacteriophage
escent polystyrene microspheres to Peyer's patch M cells. Immunology stability in non-aqueous media. Chem Biol 2001;8:661–71. http://dx.doi.org/10.
1991;73:277–80. 1016/S1074-5521(01)00041-2.
[174] Shakweh M, Ponchel G, Fattal E. Particle uptake by Peyer's patches: a pathway for [202] Waddell TE, Johnson R, Mazzocco A. US7601347 B2: methods and compositions
drug and vaccine delivery. Expert Opin Drug Deliv 2005;1:141–63. for controlled release of bioactive compounds. 2009; 2009.
[175] George M, Abraham TE. Polyionic hydrocolloids for the intestinal delivery of [203] Milo S, Thet NT, Liu D, Nzakizwanayo J, Jones BV, Jenkins ATA. An in-situ in-
protein drugs: alginate and chitosan - a review. J Control Release 2006;114:1–14. fection detection sensor coating for urinary catheters. Biosens Bioelectron
http://dx.doi.org/10.1016/j.jconrel.2006.04.017. 2016;81:166–72. http://dx.doi.org/10.1016/j.bios.2016.02.059.
[176] Ensign LM, Cone R, Hanes J. Oral drug delivery with polymeric nanoparticles: the [204] Weinberger A, Tsai FC, Koenderink GH, Schmidt TF, Itri R, Meier W, et al. Gel-
gastrointestinal mucus barriers. Adv Drug Deliv Rev 2012;64:557–70. http://dx. assisted formation of giant unilamellar vesicles. Biophys J 2013;105:154–64.
doi.org/10.1016/j.addr.2011.12.009. http://dx.doi.org/10.1016/j.bpj.2013.05.024.
[177] Lai SK, Wang YY, Hanes J. Mucus-penetrating nanoparticles for drug and gene [205] Pautot S, Frisken BJ, Weitz DA. Engineering asymmetric vesicles. Proc Natl Acad
delivery to mucosal tissues. Adv Drug Deliv Rev 2009;61:158–71. http://dx.doi. Sci 2003;100:10718–21. http://dx.doi.org/10.1073/pnas.1931005100.
org/10.1016/j.addr.2008.11.002. [206] Balcao VM, Glasser CA, Chaud MV, del Fiol FS, Tubino M, Vila MMDC. Biomimetic
[178] Jain D, Panda AK, Majumdar DK. Eudragit S100 entrapped insulin microspheres aqueous-core lipid nanoballoons integrating a multiple emulsion formulation: a
for oral delivery. AAPS PharmSciTech 2005;6:E100–7. http://dx.doi.org/10. suitable housing system for viable lytic bacteriophages. Colloids Surf B
1208/pt060116. Biointerfaces 2014;123:478–85. http://dx.doi.org/10.1016/j.colsurfb.2014.09.
[179] Silva CM, Ribeiro AJ, Ferreira D, Veiga F. Insulin encapsulation in reinforced al- 045.
ginate microspheres prepared by internal gelation. Eur J Pharm Sci [207] Torchilin VP. Recent advances with liposomes as pharmaceutical carriers. Nat Rev
2006;29:148–59. http://dx.doi.org/10.1016/j.ejps.2006.06.008. Drug Discov 2005;4:145–60. http://dx.doi.org/10.1038/nrd1632.
[180] Kietzmann D, Moulari B, Béduneau A, Pellequer Y, Lamprecht A. Colonic delivery [208] Xiao B, Laroui H, Ayyadurai S, Viennois E, Charania MA, Zhang Y, et al.
of carboxyfluorescein by pH-sensitive microspheres in experimental colitis. Eur J Mannosylated bioreducible nanoparticle-mediated macrophage-specific TNF-
Pharm Biopharm 2010;76:290–5. http://dx.doi.org/10.1016/j.ejpb.2010.06.013. alpha RNA interference for IBD therapy. Biomaterials 2013;34:7471–82. http://
[181] Chawla A, Sharma P, Pawar P. Eudragit S-100 coated sodium alginate micro- dx.doi.org/10.1016/j.biomaterials.2013.06.008.
spheres of naproxen sodium: formulation, optimization and in vitro evaluation. [209] Pichai MVA, Ferguson LR. Potential prospects of nanomedicine for targeted
Acta Pharm 2012;62:529–45. http://dx.doi.org/10.2478/v10007-012-0034-x. therapeutics in inflammatory bowel diseases. World J Gastroenterol
[182] De Arce Velasquez A, Ferreira LM, Stangarlin MFL, Da Silva CDB, Rolim CMB, 2012;18:2895–901. http://dx.doi.org/10.3748/wjg.v18.i23.2895.
Cruz L. Novel Pullulan-Eudragit S100 blend microparticles for oral delivery of [210] Schmidt C, Lautenschlaeger C, Collnot EM, Schumann M, Bojarski C, Schulzke JD,
risedronate: formulation, in vitro evaluation and tableting of blend microparticles. et al. Nano- and microscaled particles for drug targeting to inflamed intestinal
Mater Sci Eng C 2014;38:212–7. http://dx.doi.org/10.1016/j.msec.2014.02.003. mucosa - a first in vivo study in human patients. J Control Release
[183] Stuart M a C, WTS Huck, Genzer J, Müller M, Ober C, Stamm M, et al. Emerging 2013;165:139–45. http://dx.doi.org/10.1016/j.jconrel.2012.10.019.
applications of stimuli-responsive polymer materials. Nat Mater 2010;9:101–13. [211] Coco R, Plapied L, Pourcelle V, Jérôme C, Brayden DJ, Schneider YJ, et al. Drug
http://dx.doi.org/10.1038/nmat2614. delivery to inflamed colon by nanoparticles: comparison of different strategies. Int
[184] Lautenschläger C, Schmidt C, Fischer D, Stallmach A. Drug delivery strategies in J Pharm 2013;440:3–12. http://dx.doi.org/10.1016/j.ijpharm.2012.07.017.
the therapy of inflammatory bowel disease. Adv Drug Deliv Rev 2014;71:58–76. [212] Cu Y, Saltzman WM. Stealth particles give mucus the slip. Nat Mater 2009;8:11–3.
http://dx.doi.org/10.1016/j.addr.2013.10.001. http://dx.doi.org/10.1016/j.immuni.2010.12.017.Two-stage.
[185] de Souza DF, Goebel K, Andreazza IF. Development of enteric coated sustained [213] Cu Y, Saltzman WM. Controlled surface modification with poly(ethylene)glycol
release minitablets containing mesalamine. Braz J Pharm Sci 2013;49:529–36. enhances diffusion of PLGA nanoparticles in human cervical mucus. Mol Pharm
http://dx.doi.org/10.1590/S1984-82502013000300014. 2008;6:173–81. http://dx.doi.org/10.1021/mp8001254.
[186] Sekirov I, Russell S, Antunes L. Gut microbiota in health and disease. Physiol Rev [214] Saltzman WM, Radomsky ML, Whaley KJ, Cone RA. Antibody diffusion in human
2010;90:859–904. http://dx.doi.org/10.1152/physrev.00045.2009. cervical mucus. Biophys J 1994;66:508–15. http://dx.doi.org/10.1016/S0006-
[187] Nicholson JK, Holmes E, Kinross J, Burcelin R, Gibson G, Jia W, et al. Host-Gut 3495(94)80802-1.
microbiota metabolic interactions. Science 2012;336:1262–8. [215] Agrawal AK, Gupta CM. Tuftsin-bearing liposomes in treatment of macrophage-
[188] Mountzouris KC, Tsirtsikos P, Kalamara E, Nitsch S, Schatzmayr G, Fegeros K. based infections. Adv Drug Deliv Rev 2000;41:135–46. http://dx.doi.org/10.
Evaluation of the efficacy of a probiotic containing Lactobacillus, Bifidobacterium, 1016/S0169-409X(99)00061-7.
Enterococcus, and Pediococcus strains in promoting broiler performance and mod- [216] Broxmeyer L, Sosnowska D, Miltner E, Chacón O, Wagner D, McGarvey J, et al.
ulating cecal microflora composition and metabolic activities. Poult Sci Killing of Mycobacterium avium and Mycobacterium tuberculosis by a mycobacter-
2007;86:309–17. iophage delivered by a nonvirulent Mycobacterium: a model for phage therapy of
[189] Hovgaard L, Brondsted H. Current applications of polysaccharides in colon tar- intracellular bacterial pathogens. J Infect Dis 2002;186:1155–60. http://dx.doi.
geting. Crit Rev Ther Drug Carrier Syst 1996;13:185–223. org/10.1086/343812.
[190] Vandamme TF, Lenourry A, Charrueau C, Chaumeil JC. The use of polysaccharides [217] Danelishvili L, Young LS, Bermudez LE. In vivo efficacy of phage therapy for
to target drugs to the colon. Carbohydr Polym 2002;48:219–31. http://dx.doi.org/ Mycobacterium avium infection as delivered by a nonvirulent Mycobacterium.
10.1016/S0144-8617(01)00263-6. Microb Drug Resist 2006;12:1–6.
[191] Lorenzo-Lamosa ML, Remuñán-López C, Vila-Jato JL, Alonso MJ. Design of mi- [218] Owais M, Gupta CM. Targeted drug delivery to macrophages in parasitic infec-
croencapsulated chitosan microspheres for colonic drug delivery. J Control tions. Curr Drug Deliv 2005;2:311–8.
132
[219] Orozco LC, Quintana FO, Beltran RM, Moreno ID, Wasserman M, Rodriguez G. The bacteriophage and laser ablated nanoparticle complexes for pathogen detection. J
use of rifampicin and isoniazid entrapped in liposomes for the treatment of murine Phys Chem C 2015;119:14375–82. http://dx.doi.org/10.1021/acs.jpcc.5b02169.
tuberculosis. Tubercle 1986;67:91–7. [236] Vladisavljevic GT, Williams RA. Recent developments in manufacturing emulsions
[220] Roesken F, Uhl E, Curri SB, Menger MD, Messmer K. Acceleration of wound and particulate products using membranes. Adv Colloid Interface Sci
healing by topical drug delivery via liposomes. Langenbecks Arch Surg 2005;113:1–20. http://dx.doi.org/10.1016/j.cis.2004.10.002.
2000;385:42–9. http://dx.doi.org/10.1007/s004230050010. [237] Kukizaki M. Shirasu porous glass (SPG) membrane emulsification in the absence of
[221] Ferreira H, Matamá T, Silva R, Silva C, Gomes AC, Cavaco-Paulo A. shear flow at the membrane surface: influence of surfactant type and concentra-
Functionalization of gauzes with liposomes entrapping an anti-inflammatory drug: tion, viscosities of dispersed and continuous phases, and transmembrane pressure.
a strategy to improve wound healing. React Funct Polym 2013;73:1328–34. J Membr Sci 2009;327:234–43. http://dx.doi.org/10.1016/j.memsci.2008.11.
http://dx.doi.org/10.1016/j.reactfunctpolym.2013.05.012. 026.
[222] Yerushalmi N, Arad A, Margalit R. Molecular and cellular studies of hyaluronic [238] Vladisavljević GT, Kobayashi I, Nakajima M. Production of uniform droplets using
acid-modified liposomes as bioadhesive carriers for topical drug delivery in wound membrane, microchannel and microfluidic emulsification devices. Microfluid
healing. Arch Biochem Biophys 1994;313:267–73. http://dx.doi.org/10.1006/ Nanofluid 2012;13:151–78. http://dx.doi.org/10.1007/s10404-012-0948-0.
abbi.1994.1387. [239] Kosvintsev SR, Gasparini G, Holdich RG, Cumming IW, Stillwell MT. Liquid-liquid
[223] Pearson HA, Sahukhal GS, Elasri MO, Urban MW. Phage-bacterium war on poly- membrane dispersion in a stirred cell with and without controlled shear. Ind Eng
meric surfaces: can surface-anchored bacteriophages eliminate microbial infec- Chem Res 2005;44:9323–30.
tions? Biomacromolecules 2013;14:1257–61. [240] Vladisavljević GT, Williams RA. Manufacture of large uniform droplets using ro-
[224] Bennett AR, Davids FG, Vlahodimou S, Banks JG, Betts RP. The use of bacter- tating membrane emulsification. J Colloid Interface Sci 2006;299:396–402.
iophage-based systems for the separation and concentration of Salmonella. J Appl http://dx.doi.org/10.1016/j.jcis.2006.01.061.
Microbiol 1997;83:259–65. [241] Holdich RG, Dragosavac MM, Vladisavljević GT, Kosvintsev SR. Membrane
[225] Sun W, Brovko L, Griffiths M. Food-borne pathogens. Use of bioluminescent emulsification with oscillating and stationary membranes. Ind. Eng. Chem. Res.
Salmonella for assessing the efficiency of constructed phage-based biosorbent y,z. 2010;49:3810–7. http://dx.doi.org/10.1021/ie900531n.
Int J Food Microbiol 2001:273–5. [242] Silva PS, Dragosavac MM, Vladisavljevic GT, Bandulasena HCH, Holdich RG,
[226] Tolba M, Minikh O, Brovko LY, Evoy S, Griffiths MW. Oriented immobilization of Stillwell M, et al. Azimuthally oscillating membrane emulsification for controlled
bacteriophages for biosensor applications. Appl Environ Microbiol droplet production. AIChE J 2015;61:3607–15. http://dx.doi.org/10.1002/aic.
2010;76:528–35. http://dx.doi.org/10.1128/AEM.02294-09. [243] Anna SL. Droplets and bubbles in microfluidic devices. Annu Rev Fluid Mech
[227] Anany H, Chen W, Pelton R, Griffiths MW. Biocontrol of Listeria monocytogenes and 2016;48:285–309. http://dx.doi.org/10.1146/annurev-fluid-122414-034425.
Escherichia coli O157:H7 in meat by using phages immobilized on modified cel- [244] Engl W, Backov R, Panizza P. Controlled production of emulsions and particles by
lulose membranes. Appl Environ Microbiol 2011;77:6379–87. http://dx.doi.org/ milli- and microfluidic techniques. Curr Opin Colloid Interface Sci
10.1128/AEM.05493-11. 2008;13:206–16. http://dx.doi.org/10.1016/j.cocis.2007.09.003.
[228] Donlan RM, Lehman SM, Garcia AJ. US9457132B2: controlled covalent attach- [245] Yobas L, Martens S, Ong W-L, Ranganathan N. High-performance flow-focusing
ment of bioactive bacteriophage for regulating biofilm development. 2016; 2016. geometry for spontaneous generation of monodispersed droplets. Lab Chip
[229] Scott H, Mattey M. WO2003093462A2: immobilisation and stabilisation of virus. 2006;6:1073–9. http://dx.doi.org/10.1039/b602240e.
2003; 2003. [246] Chu LY, Utada AS, Shah RK, Kim JW, Weitz DA. Controllable monodisperse
[230] Cooper IR, Illsley M, Korobeinyk AV, Whitby RLD. Bacteriophage-nanocomposites: multiple emulsions. Angew Chem Int Ed 2007;46:8970–4. http://dx.doi.org/10.
an easy and reproducible method for the construction, handling, storage and 1002/anie.200701358.
transport of conjugates for deployment of bacteriophages active against [247] Wang W, Xie R, Ju X-J, Luo T, Liu L, Weitz DA, et al. Controllable microfluidic
Pseudomonas aeruginosa. J Microbiol Methods 2015;111:111–8. http://dx.doi.org/ production of multicomponent multiple emulsions. Lab Chip 2011;11:1587–92.
10.1016/j.mimet.2015.02.005. http://dx.doi.org/10.1039/c1lc20065h.
[231] Handa H, Gurczynski S, Jackson MP, Auner G, Walker J, Mao G. Recognition of [248] Abate AR, Weitz DA. High-order multiple emulsions formed in poly(dimethylsi-
Salmonella typhimurium by immobilized phage P22 monolayers. Surf Sci loxane) microfluidics. Small 2009;5:2030–2. http://dx.doi.org/10.1002/smll.
2008;602:1392–400. http://dx.doi.org/10.1016/j.susc.2008.01.036. 200900569.
[232] Singh A, Glass N, Tolba M, Brovko L, Griffiths M, Evoy S. Immobilization of [249] Kim SH, Weitz DA. One-step emulsification of multiple concentric shells with
bacteriophages on gold surfaces for the specific capture of pathogens. Biosens capillary microfluidic devices. Angew Chem Int Ed 2011;50:8731–4. http://dx.
Bioelectron 2009;24:3645–51. http://dx.doi.org/10.1016/j.bios.2009.05.028. doi.org/10.1002/anie.201102946.
[233] Yang LC, Tam PY, Murray BJ, Mcintire TM, Overstreet CM, Weiss GA, et al. Virus [250] Yang S, Guo F, Kiraly B, Mao X, Lu M, Leong KW, et al. Microfluidic synthesis of
electrodes for universal biodetection. Anal Chem 2006;78:3265–70. multifunctional Janus particles for biomedical applications. Lab Chip
[234] Richter Å, Matuła K, Lesniewski A, Kwasnicka K, Łos J, Łos M, et al. Ordering of 2012;12:2097. http://dx.doi.org/10.1039/c2lc90046g.
bacteriophages in the electric field: application for bacteria detection. Sens [251] Nie Z, Li W, Seo M, Xu S, Kumacheva E. Janus and ternary particles generated by
Actuators B 2016;224:233–40. http://dx.doi.org/10.1002/9780470015902. microfluidic synthesis: design, synthesis, and self-assembly. J Am Chem Soc
a0024053. 2006;128:9408–12. http://dx.doi.org/10.1021/ja060882n.
[235] Tawil N, Sacher E, Rioux D, Mandeville R, Meunier M. Surface chemistry of
133

Formulation Stabilisation and Encapsulation of Bacteriophage For Phage Therapy

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Formulation Stabilisation and Encapsulation of Bacteriophage For Phage Therapy

Uploaded by

Copyright:

Available Formats

Advances in Colloid and Interface Science 249 (2017) 100–133

Contents lists available at ScienceDirect

Advances in Colloid and Interface Science

Formulation, stabilisation and encapsulation of bacteriophage for phage T

1. Introduction infections may be regarded as a signiﬁcant achievement of modern

Host organism Animal Infection Delivery Phage therapy Outcome

Single phage (*) / Cocktail ( )

Delayed phage treatment

Phage resistant mutants

Dose related studies

(a) (b) (c) (d)

(a) (b) (c)

Merabishvili et al. [98] prepared a well-deﬁned cocktail (designated 4. Bacteriophage encapsulation

Result / PFU Log reduction

Proteins. (casein, whey etc.)

Amino acids or peptides

Drying on filter paper

Spray freeze drying

Intensity of shading diﬀerentiates vertical sections of the table.

(HYMA); Poly(DL- chloroform as of infectious intestinal thermally responsive polymer

cellulose diacetate infections

polymerisation (PP); liquid

(CDA); polyethylene (G / R / W); +/- indicates phage protection /

oxide (PEO); whey damage

[100] Pa Sa PLGA; PVA W/O/W – SE - R Buffer ~ 10 µm CI; SS

4.8.5. Bacteriophage encapsulation in synthetic polymers 4.9. Bacteriophage encapsulation in liposomes

considerably longer compared with free phage (refer to earlier section

intracellular macrophage model;

Storage studies; PE (good)

Ex vivo bioﬁlm model;

polyoxamers (Lutrol F68)

Tween 80; stearylamine;

Tween 80; stearylamine

Tween 80; glycerol;

gel assisted hydration

whereas liposome encapsulated phage were detectable in blood and all

Thin ﬁlm hydration

Thin ﬁlm hydration

Thin ﬁlm hydration

treatment and remained detectable in the spleen up until day 14.

Liposomes are spherical structures delimited by a double layer of

and lamellarity vary as a result of the manufacturing process. Fluidity

or hydrophobic therapeutic agents in their aqueous core or within the

pharmaceutical carrier of choice for numerous applications [207].

From a clinical viewpoint the potential ability of liposome en-

capsulated phage to enter live cells containing intracellular pathogens

teria e.g. S. aureus or E. coil. The surface of liposomes may be functio-

fate of the bacteriophage loaded liposomes; (iv) incorporation of posi-

tively charged lipid derivatives or positively charged polymers e.g. to

aid mucosal adhesion thereby improving residence time in the gastro-

in modulated immune response to the presence of the bacteriophage in

Fig. 3. Membrane emulsiﬁcation (ME) methods that can be

Fig. 4. Morphologies of complex drops: (a) Conventional

Janus Ternary drop

Janus core Janus shell

Appendix A. Mathematical model of bacteria/phage interaction

S, R - concentrations of susceptible and resistant bacteria accordingly [CFU/ml],

References bacteriophage therapy on pseudomonas aeruginosa cystic ﬁbrosis strains: ﬁrst

You might also like