The phage target specificity means that they cause minimal disruption to the normal microbiomes of animals, thus preserving the beneficial microbial niche
[72][5]. The precise bacterial selection by the phage is achieved by recognising specific receptor proteins on the host bacterium which the phage adsorb to using specialised tail fibers; after which they penetrate and release their genetic material into the host
[73][6]. Generally, phages of most
S. aureus strains interact with a unique cell wall teichoic acid, which is different from other coagulase-negative staphylococci (CoNS) and blocks recognition by phages specific to CoNS using the tail tip complex
[74][7]. For studies specifically conducted on bovine mastitis,
S. aureus phages utilise three domains located on the endolysin sequences: cysteine; histidine-dependent amidohydrolase/peptidase (CHAP); amidase 2 (N-acetylmuramyl-L-alanine amidase); and SH3b for host cell wall recognition
[42][8].
Sequel to successful adsorption and penetration, lytic phages immediately hijack their host DNA replication machinery to synthetise genetic materials and structural proteins during the latent period. The time taken to achieve this has been reported to vary in bovine mastitis phages and could range from 10 (
S. aureus), 15 (
E. faecum), 20 (
P. aeruginosa phage), to 30 min (
S. agalactiae)
[7,50,67,69][9][10][11][12]. Subsequently, after viral synthesis, numerous phage particles are assembled and eventually released by the lysis of the host through a combined activity of the endolysin and holin enzymes that degrade the bacteria cell wall
[75][13]. For bovine mastitis, phage progeny or burst (number of phage particles synthesised and released per single bacterial cell) varies from 20 to 100 PFU/cell within ~175 min
[7,27,50,67,69][9][10][11][12][14]. The ability of lytic phages to ultimately lyse infected bacteria and amplify after infection ensures the clearance of the bacterial pathogens as well as continual increased supply of infective phages (auto-dosing) at infection sites
[26,71][15][16]. Furthermore, the shorter replication time demonstrated by the phages can reduce product development timeframes to provide an opportunity for rapid customised or tailored treatments to target specific strains of bacteria
[26][15].
For bovine mastitis, there is also a range in the types of phages which have been identified as being candidates for treatment of one or more of the organisms known to be an infectious agent. Characteristics of bacterial species which cause mastitis, together with the types of research studies conducted on therapeutic phages are summarised in
Figure 1. Several phages have been identified for these purposes, but a greater proportion of work has been conducted on
S. aureus being the common aetiological agent causing this infection. This approach has also been described for bacterial species such as
A. viridans [68][17],
E. coli [60][18] and
K. oxytoca [9][19].
Figure 1. Characteristics of bacterial communities that cause bovine mastitis and aspects of bacteriophage therapeutic studies conducted in the area to control the infection. (A) The microbial niche of bovine mastitis consists of polybacterial strains which can aggregate and produce biofilms and are mostly resistant to several antimicrobials which affect antibiotic efficiency. (B) Phages that target and kill the bacterial species have been isolated either directly or via enrichment of samples from sewage, wastewater and sewage effluent, manure and faeces of cattle and pigs, and dairy products such as cheeses and raw or unpasteurised milk, via prophage induction of bacterial cultures with mitomycin C or commercially sourced. Phage characterisation focuses on determining lysis specificity and efficacy, phage infection kinetics to ascertain the adsorption, replication/amplification and growth, host range coverage (black cells showing bacterial lysis by phages and white cells showing no lysis), and genome analyses to ascertain gene functions, expressions and relationships. Therapeutic activities of single and combinations of phages were tested in various infection models in vitro (bacterial clearance in pure cultures, milk and biofilms), ex vivo (in cell cultures of MAC-T and bMEC) and in refined (Galleria mellonella larvae) and established (mice and cattle) in vivo models. The mouse and cattle images were downloaded from Microsoft PowerPoint resources under the license CC BY-NC and no modifications were made.
2.2. Isolation of Phages from a Wide Range of Sources
Phages are the most abundant entities on earth with a ~10
31 PFU/mL reported concentration
[76,77,78][20][21][22]. For bovine mastitis, various sample sources have been explored for the purpose of isolating phages for the various pathogens responsible for the infection. Most of the work has focused on screening raw milk samples obtained from a confirmed mastitis cattle, either directly after centrifuging and filtering of samples or via enrichment procedures to amplify and isolate phages
[28,34][23][24]. Phages have been reported to actively bind to, lyse and amplify in milk constituents, and huge successes of phage isolation for
S. aureus,
S. agalactiae and
S. arlettae have been recorded from this source through this method
[7,28,29,34,40,41,46,49,63][9][23][24][25][26][27][28][29][30]. However, in one instance, no phage was isolated from the milk samples examined
[48][31]. The reason for this may be attributed to the bovine whey protein which may prevent attachment of some phages
[58][32]. This may also simply be the lack of phages specific for the bacterial host used as a target for the isolation or the occurrence of the phages in very low titers requiring an enrichment procedure to enable viral amplification and enhance detection
[48][31].
Milk products have also been examined and yielded phages for bovine mastitis pathogens
Staphylococcus and
Streptococcus via enrichment of Cabrales and Peñamellera cheeses, although all phages isolated from this method yielded temperate phages
[34][24]. Moreover,
S. aureus and
S. arlettae phages of lysogenic origin have been isolated from milk as well
[34[24][26][27][30][33],
40,41,56,63], although for regulatory purposes, lytic phages are preferred to temperate phages due to the possibility of lysogeny occurring and transfer of virulence genes via horizontal gene transfer. However, where strictly lytic phages are not isolated, temperate phages showed potential therapeutic efficacy and are particularly useful for treatment
[79,80][34][35].
Sewage, sewage effluent, sewage water, barn flushes, wastewater, cowshed water and manure from dairy farms have yielded a large quantity of phages targeting mastitis-causing pathogens which may be attributed to the microbial richness in these sources
[32,37,42,43,48,51,52,53][8][31][36][37][38][39][40][41]. Other very odd sources such as pig manure have been a good source to isolate phages for the infection; this may reveal the interconnection of niches for these organisms
[50,54,55][10][42][43].
Phages from commercial sources such as 23361 (ATCC), BP39 (PhageLux) and SAML-4, SAML-12, SAML-150, SAML-4229 and SATA-8505 (StaphLyse) have been investigated for potential usage for bovine mastitis
S. aureus [30,31][44][45].
2.3. Cocktail Optimisation to Improve Therapeutic Activity
Therapeutic activity of single-phage treatments can significantly reduce bacterial load in many infection models using optimal multiplicity of infection (MOI; the ratio of infecting phages to bacteria in a given infection challenge) as shown in many studies
[32,43,48,49,51,52,53,60][18][29][31][36][38][39][40][41]. However, phage resistance was detected within as early as 2 hours after phage treatment as indicated by the regrowth of cultures after lysis which can negatively impact therapeutic efficacy
[49][29]. To curtail resistance and lysogeny development, broaden host target coverage and specificity, and to improve lysis efficiency, a cocktail of diverse phages can be optimised
[62,81,82,83][46][47][48][49]. This strategy has proven successful, and various combinations of diverse phage morphologies have shown beneficial combinatorial effects in clearing several bacteria causing bovine mastitis. For example, a four-phage cocktail was developed for
E. coli and cocktails of two or three phages were shown to be more effective than single-phage treatments for
S. aureus [30,34,35,54,55,61,62][24][42][43][44][46][50][51]. Similarly, the therapeutic efficacy of a phage cocktail was shown to be comparable to that of the antibiotic ceftiofur sodium for
E. coli in cattle and
S. aureus in mice
[35,61][50][51]. This has been extrapolated further, with a cocktail of four phages together with the lactic acid bacterium
Lactiplantibacillus plantarum proving effective
[54][42]. Phage activity on
S. aureus was shown to be delayed by
IgG-dependent aggregation using single-phage treatment. While in contrast, the use of a cocktail showed no significant effect with or without
IgG in milk
[30][44].
2.4. Characterisation of Phage Lysis and Stability in Pure Cultures
Several therapeutic assessments have investigated the efficacy and safety of phages for the targeted eradication of bovine mastitis. Fundamental research has been conducted regarding phage activity in pure cultures to determine lysis capabilities by individual phages and in combination with other phages. Host range analysis mainly focuses on phage lysis activity using spot test with the double-layer agar method (application of phage samples to confluent cultures of bacteria in a semi-solid agar medium overlayed on solid agar medium). This is to ascertain the range of relevant bacterial strains the phages can lyse with some demonstrating broad or narrower host coverage
[28,36,37][23][37][52]. Besides phage coverage on a wide range of strains, other phages of
S. aureus showed inter-species lysis, targeting
S. sciuri and
Rothia terrae [28][23] as well as
E. coli [43][38], and
K. oxytoca phage P2 lysing
E. aerogenes as well
[9][19].
Further work was also directed to stability (in various temperature and pH ranges) and killing assays in pure cultures in broth or liquid media and milk (pasteurised and unpasteurised) using MOI assays in a given infection model to provide an insight into the dosage
[54,55][42][43]. Data showed a wide range of effectiveness of MOI range of 0.001 to 100 in vitro
[43,48,49,51,52,53,63,68][17][29][30][31][38][39][40][41]. However, optimal effectiveness was at MOI of 10 in vivo for some of the data
[60][18]. Other reports showed that efficacy was achieved in a phage-dose-dependent manner in milk using an
S. aureus phage
[31][45].
2.5. Phage Therapeutic Activity in Biofilms
The pathogens causing mastitis can aggregate in vitro and in vivo in extracellular polysaccharide-containing biofilm matrixes which restricts antibiotic access to bacteria
[17,18,47,84][53][54][55][56]. Phages have been shown to prevent or penetrate established biofilms produced by mastitis bacteria in vitro and in vivo, hence showing the potential to be used as a standalone treatment or to supplement antibiotic use and enhance therapeutic efficacy
[47,49][29][55]. The phages can lyse bacteria early in the culture to prevent biofilm formation or may disrupt established biofilms which can enhance bacterial killing or provide pathogens access within the biofilm matrix
[72,82][5][48]. In
S. aureus biofilms, treatment using a single phage or a cocktail of phages significantly reduced bacterial load in planktonic cultures as well as established biofilms on polystyrene surfaces, in milk and on mammary glands
[47,49][29][55].
As well as the issue of potential protection from biofilms, mastitis-causing bacteria have been shown to be afforded some level of protection from bacterial aggregation
[56][33], including during the times when
S. aureus was exposed to phage infection. However, previous work showed a total kill of
S. aureus, which has a few cells and have survived phage treatment, probably by some level of aggregation. This means that the numbers remaining are sufficiently low for them in turn to be removed by the animal’s own immune system
[30][44].
2.6. Phage Therapeutic Assessments in Mastitis Ex Vivo and In Vivo Models
Phages have low inherent toxicity to the immune system, and they are potentially cheaper to isolate and develop, which provides an economic advantage over antibiotics
[72][5]. To contextualise and provide insight into the therapeutic safety and efficacy of phages, relevant ex vivo models involving bovine cells lines were investigated. The studied cell lines for bacterial and phage interactions for this are the mammary alveolar cells-large T antigen (MAC-T) and bovine mammary epithelial (bMEC) cell lines
[51,52,65][39][40][57]. A cocktail of two phages, CM8-1 and SJT-2 was shown to reduce
K. pneumonia numbers and consequently reduce adhesion, invasion, and cytotoxicity in bMEC cells
[65][57].
S. aureus phages were shown to migrate intercellularly and could reach the nucleus within 3 h after exposure to MAC-T cell lines and have an endocytotic activity of 12% in a bovine ex vivo model
[51,52][39][40].
Studies on
S. aureus-colonised
G. mellonella larvae showed a 50% survival rate four days after treatment with a single phage
[44][58]. The in vivo model that has been extensively studied for bovine mastitis phage therapy is the mouse model mainly because this model has itself been well established for infection since the 1970s
[35,44,50,60,64,68][10][17][18][50][58][59]. Results in mice showed favourable outcomes for phage therapy with reduced colonisation and reduced inflammatory cytokines as soon as 24 hours after treatment. The mouse model has also been reported to be a more time- and cost-effective model than those of larger mammals with comparable symptoms, inflammatory indicators, colonisation, and histopathological characteristics. Therapeutic efficacies have been achieved in cattle as well
[57,61][51][60].
3. Barriers/Challenges to Therapeutic Phage Application to Control Bovine Mastitis
Researchers outlined the advantages of phage therapy and research work conducted in the area to control bovine mastitis. However, a degree of caution needs to be applied by anyone considering using it as a potential prophylactic treatment. It has been reported that the infusion of a phage sample into unaffected quarters in the udder of lactating dairy cattle resulted in an increase in the somatic cell count in the milk from that quarter
[57][60]. This suggests that there has been some form of immune response taking place in that particular quarter of the cow’s udder. A comparable increase in somatic cell count was not seen in animals infused with a phage sample where the animal had some level of mastitis infection, even at a sub-clinical level
[57][60].
The situation in terms of using phage as a treatment for mastitis is complicated, yet evidence exists to show that in
S. aureus the whey proteins in milk can adhere to the surface of cells, thereby blocking potential attachment sites for the phage
[58][32]. Moreover, it was shown that in raw milk, as opposed to milk which has been heat-treated, phage K which has the potential to infect and kill
S. aureus was less successful
[59][61]. It is thought that this is due to the clumping of the bacteria on fat globules within the milk and some sort of presumed protection from this activity.
On the other hand, the lysogenisation of the bacterial host by temperate phages could potentially cause the exchange of virulence factors via horizontal gene transfer as stated above. However, the use of phages can come with additional complications. One such example of this was seen where a phage which entered the lysogenic phase was also found to contain a gene which conferred resistance to multiple types of antibiotics
[12][62]. Therefore, although there is a clear potential for usage of phages as a means of killing bacteria causing mastitis infections, there needs to be considerable research undertaken before using these phages as treatments. Temperate phages can access the lytic life cycle via induction through treatment with mitomycin C as shown in
S. galactiae [56][33] or activation of the repressor or deletion of the integrase genes. Unfortunately, they are unsuitable for therapeutic purposes in their wild form. However, genetic engineering has provided avenues for genetic manipulation to help develop therapeutically acceptable phages where strictly lytic ones are not available
[85,86,87][63][64][65].
The polymicrobial niche of bovine mastitis is also a challenge to overcome
[5,6,16][66][67][68]. Most work conducted to date has focused on a single bacterial species in relevant infection model systems, except for example where
S. aureus phages showed interspecies lysis on
S. sciuri and
Rothia terrae [28][23], and
E. coli [43][38]. Whilst this is informative and provides useful insights into the therapeutic potential of the phages, it is still unclear how these single bacterial species targets would alleviate bovine mastitis. More work is therefore needed on multispecies targets through phage cocktail optimisations to clear the bacterial communities as standalone treatments or as adjunct to antibiotics for the effective clearance of bovine mastitis infection.