Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Isolation and Characterization of Two Lytic Bacteriophages, φSt2 and φGrn1; Phage Therapy Application for Biological Control of Vibrio alginolyticus in Aquaculture Live Feeds

  • Panos G. Kalatzis,

    Affiliations Institute of Marine Biology, Biotechnology and Aquaculture, Hellenic Centre for Marine Research, Former American Base of Gournes, Heraklion 71003, Crete, Greece, Marine Biological Section, University of Copenhagen, Helsingør, Denmark

  • Roberto Bastías,

    Affiliations Institute of Marine Biology, Biotechnology and Aquaculture, Hellenic Centre for Marine Research, Former American Base of Gournes, Heraklion 71003, Crete, Greece, Institute of Biology, Pontificia Universidad Católica de Valparaíso, Valparaíso, Chile

  • Constantina Kokkari,

    Affiliation Institute of Marine Biology, Biotechnology and Aquaculture, Hellenic Centre for Marine Research, Former American Base of Gournes, Heraklion 71003, Crete, Greece

  • Pantelis Katharios

    Affiliation Institute of Marine Biology, Biotechnology and Aquaculture, Hellenic Centre for Marine Research, Former American Base of Gournes, Heraklion 71003, Crete, Greece

Isolation and Characterization of Two Lytic Bacteriophages, φSt2 and φGrn1; Phage Therapy Application for Biological Control of Vibrio alginolyticus in Aquaculture Live Feeds

  • Panos G. Kalatzis, 
  • Roberto Bastías, 
  • Constantina Kokkari, 
  • Pantelis Katharios


Bacterial infections are a serious problem in aquaculture since they can result in massive mortalities in farmed fish and invertebrates. Vibriosis is one of the most common diseases in marine aquaculture hatcheries and its causative agents are bacteria of the genus Vibrio mostly entering larval rearing water through live feeds, such as Artemia and rotifers. The pathogenic Vibrio alginolyticus strain V1, isolated during a vibriosis outbreak in cultured seabream, Sparus aurata, was used as host to isolate and characterize the two novel bacteriophages φSt2 and φGrn1 for phage therapy application. In vitro cell lysis experiments were performed against the bacterial host V. alginolyticus strain V1 but also against 12 presumptive Vibrio strains originating from live prey Artemia salina cultures indicating the strong lytic efficacy of the 2 phages. In vivo administration of the phage cocktail, φSt2 and φGrn1, at MOI = 100 directly on live prey A. salina cultures, led to a 93% decrease of presumptive Vibrio population after 4 h of treatment. Current study suggests that administration of φSt2 and φGrn1 to live preys could selectively reduce Vibrio load in fish hatcheries. Innovative and environmental friendly solutions against bacterial diseases are more than necessary and phage therapy is one of them.


The intensification of aquaculture production has dramatically increased the incidences of microbial diseases causing substantial economic losses to the industry. One of the biggest problems in intensive fish culture is the mass mortalities in fish larvae caused by bacterial infections [13]. In marine aquaculture, vibrios are major pathogens causing vibriosis which is the most common disease in marine fish and invertebrate hatcheries [37].

Vibrio alginolyticus is a ubiquitous bacterium found in marine environment that has been associated with disease in aquatic animals but also in humans, causing tissue damages in skin, ear and internal organs [811].

V. alginolyticus is also one of the most common species found in marine hatchery water [12,13] and it is considered as an important pathogen for marine organisms [14], especially by being opportunistic invader of already damaged fish tissues [15]. There are several reports for V. alginolyticus causing significant mortalities in cultured gilthead seabream, S. aurata, especially during early life stages [1620]. Larval enteropathy (LE) is the most important pathology affecting this species at hatcheries, which is responsible for great reduction in survival rates. V. alginolyticus alone, or in synergy with other bacteria such as Aeromonas hydrophila, constitutes the major causative agent of LE [19]. Apart from cultured gilthead seabream, V. alginolyticus infection has been recorded during early rearing stages (≤3 g) of sharpsnout seabream, Diplodus puntazzo [21]. Mortality due to V. alginolyticus has also been recorded in ornamental fish [2224] and several invertebrates such as Penaeus monodon [25] and Macrobrachium rosenbergii [26]. In aquaculture there is a general consensus that V. alginolyticus enters the system through live prey (artemia and rotifers) which serve as vehicles for introducing the bacteria into the hatchery tanks [2729]. There are several studies demonstrating that Artemia nauplii are vectors for potentially harmful bacteria such as Vibrio spp. [30]. V. alginolyticus has been reported as the dominant member of the cultivable bacterial community of Artemia [13,29,31,32].

Disinfection techniques (filters, ozone, UV etc.) in marine hatcheries cannot offer a completely bacteria—free environment [33] and may lead to microbial imbalance leaving environmental niche wide open for the proliferation of opportunistic pathogens [34,35]. Administration of antibiotics has traditionally been the most commonly applied strategy against bacterial infections. Today, antibiotic usage is becoming increasingly obsolete in aquaculture as many economically important pathogens evolve resistance, including strains belonging to the genera Aeromonas and Vibrio [18,3638]. Development of multi—drug resistant strains, disturbance of natural microbiota, ecological and public health issues are some of the most important problems caused by the excessive use of chemotherapy [3941]. Thus, bacterial disease outbreaks could be ideally managed by limiting or even excluding pathogenic bacteria, as Vibrio, from the system without affecting the beneficial microbes. In fragile systems like marine hatcheries the use of bacteriophages, viruses that infect bacteria, is a promising alternative since they can selectively remove their bacterial hosts, while leaving natural microbiota unaffected [41,42]. Up to date, phage therapy has been applied in numerous cases of vibriosis against many pathogenic Vibrio spp. such as V. harveyi, V. parahaemolyticus and V. anguillarum, with positive results [4348].

The objective of this study is the isolation and characterization of lytic bacteriophages against the dominant vibrios of the live feeds of a marine fish hatchery and the assessment of their efficacy to reduce vibrio load in live feeds prior to their administration to fish larvae.

Materials and Methods

Bacterial strains and growth conditions

Vibrio alginolyticus, strain V1 was used as host for bacteriophage isolation. This strain was isolated previously from a vibriosis episode in juvenile gilthead seabream, S. aurata and it has already been fully sequenced [49]. Twenty-five different bacterial strains belonging to seven Vibrio species (V. anguillarum, V. harveyi, V. alginolyticus, V. ordalii, V. parahaemolyticus, V. splendidus and V. owensii) were used in the current study (Table 1). These bacteria were either purchased from international collections, or belong to HCMR’s collection and include both clinical and environmental isolates. V. splendidus strains were a kind offer from Dr. Frédérique Le Roux (Roscoff Marine Station). All bacterial species have been identified using biochemical (BIOLOG GEN III) and/or molecular tools [5052]. All bacterial strains were cultured in artificial sea water (23.4 gL-1NaCl, 24.7 gL-1 MgSO4 x 7H2O, 1.5 gL-1KCl and 1.43 gL-1CaCl2 x 2H2O), supplemented with 1% tryptone (Difco) and 0.5% yeast extract (Difco) at 25°C with reciprocal shaking [48].

Table 1. Bacterial strains of the genus Vibrio used in the current study.

T: type strain.

Isolation and propagation of bacteriophages

Two novel bacteriophages were isolated from water samples collected from two locations of the north coastline of Crete, Greece (35°20'16.2"N, 25°06'36.6"E and 35°20'05.3"N, 25°16'30.3"E) through standard enrichment methodology [53]. No special permission was required for sampling water. Briefly, samples were supplemented with 1% tryptone (Difco) and 0.5% yeast extract (Difco) and then inoculated with bacterial host, V. alginolyticus strain V1. Samples were incubated overnight at 25°C with reciprocal shaking, following centrifugation at 6,000 x g for 10 min. Supernatants were filtered (0.22 μm) and 100 μL were plated by standard double-layer agar method and incubated overnight at 25°C to detect and enumerate plaque forming units (pfu). Isolated plaques were picked and purified by re-plating five times to ensure clonal phage stocks. For phage propagation, 50 mL of a bacterial host liquid culture in early exponential phase (~108 cells mL-1) was infected at a multiplicity of infection (MOI) of 10 and incubated overnight at 25°C with reciprocal shaking. After centrifugation of the cultures, the supernatants were filtered (0.22μm), tittered and stored at 4°C.

Host range and efficiency of plating (EOP)

Bacterial lawns of each bacterial strain tested were prepared on Petri dishes of artificial sea water (23.4 gL-1NaCl, 24.7 gL-1 MgSO4 x 7H2O, 1.5 gL-1KCl and 1.43 gL-1CaCl2 x 2H2O) supplemented with 1% tryptone (Difco) and 0.5% yeast extract (Difco), and 20 μL drops of each phage were added on them, following overnight incubation at 25°C. EOP assay was also performed to obtain a quantitative measure of phage’s lytic activity and to assess possible “lysis from without” phenomenon [54,55]. EOP was determined for each phage—sensitive bacterial strain, by dividing the infectivity of phages vs tested strains to the infectivity of phages vs host strain V1 [56].

Morphological characterization of bacteriophages

Virion morphology of isolated phages was observed by Transmission Electron Microscopy (TEM). Samples were prepared on collodium copper grids, negatively contrasted with 2% uranyl acetate, and examined using an electron microscope (JEOL JEM2100) at 80 kV and an instrumental magnification of 120,000.

Viral whole genome extraction and RFLP analysis

Bacteriophages’ genome was extracted using phenol-chloroform protocol [48] and visualized in agarose gel 0.4% at 30 mV compared to a high range genome size ladder (Thermo Scientific GeneRuler High Range DNA Ladder). Restriction Fragment Length Polymorphism analysis (RFLP) was performed using six different restriction enzymes HpaII, Sau3AI, HincII, HaeIII, BgIII and BamHI, according the manufacturer’s instructions (Promega) for digesting phage genome. The genome of the phages was tested for resistance against RNAse A (Qiagen) and DNAse (Qiagen) treatment.

One step growth curve of the bacteriophages

Bacteriophages were added to 1mL host bacterial culture in early exponential phase, with MOI = 0.01, following incubation at 25°C for 15 min and centrifugation at 6,000 x g for 10 min. Supernatant containing free phages was discarded whereas phages that managed to attach to the bacteria during interaction time were pelleted on the bottom of the tube. Pellet was suspended in 0.5 mL and then transferred in 20 mL fresh liquid medium. This moment was considered as t = 0 and thereafter, 20 μL drops of serial dilutions were placed on Petri dishes containing bacterial lawn of the host, every 10 min for total duration of 80 min. Phage plaques were counted following overnight incubation at 25°C. The experiment was repeated three times for each phage.

In vitro efficacy vs host strain V1

In vitro lysis assay was performed in sterile 96-well plates using a TECAN microplate reader (Infinite PRO 200) equipped with temperature control. Briefly, 24 wells were used per each condition. Wells were loaded with 200 μL of freshly prepared culture of the host bacteria. The plate was placed in the reader and incubated at 25°C with orbital shaking. Cultures were infected at 3 different MOIs (1, 10 and 100 in sextuplicates) when the bacterial culture was at the exponential phase. Twelve wells were not infected and served as control. The growth curve of the cultures was monitored in real—time over 1000 min and OD560 measurements were recorded every 20 min.

In vitro efficacy vs presumptive vibrios from A. salina culture

The lytic effect of phages φGrn1 and φSt2 was also tested in vitro against bacterial isolates from live feed A. salina culture, grown on the selective medium for Vibrio TCBS (Thiosulphate Citrate Bile Salt). A sample from HCMR’s Artemia live feed culture was serially diluted on TCBS and incubated at 25°C for 24 h. Twelve different single bacterial colonies covering the whole range of various morphologies were picked and recultured twice on artificial sea water (23.4 gL-1NaCl, 24.7 gL-1 MgSO4 x 7H2O, 1.5 gL-1KCl and 1.43 gL-1CaCl2 x 2H2O), supplemented with 1% tryptone (Difco) and 0.5% yeast extract (Difco). Bacterial cultures of isolated presumptive vibrios were infected in their early exponential phase by the mixture of φSt2 and φGrn1 using a MOI = 100, with both phages’ titer being approximately the same. The experiment was done as described previously for the host strain in a 96 –well plate for 850 min organized in triplicates per bacterium.

In vivo administration of φGrn1 and φSt2 in A. salina culture

The brine shrimp, Artemia is a zooplanktonic organism widely used as live feed. It can be hatched within 24 hours from dormant cysts (batch culture) which can be easily distributed and stored for prolonged periods of time [30]. The cultures of Artemia salina (SEP Art, INVE), were obtained from the department of live feed cultures of the Hellenic Centre for Marine Research.

The presumptive Vibrio load of the Artemia cultures had been estimated earlier to be approximately 105 cfu mL-1. Six plastic containers were used, each containing 1L of A. salina cultures supplied with intense aeration. The total presumptive Vibrio count in each container was assessed at t0 = 0 h and at t1 = 4 h following serial dilutions of 1 mL samples and plating in TCBS. Phage mixture of φGrn1 and φSt2 was directly administered at a MOI = 100. The experiment was done in triplicates, phage treated vs untreated controls Artemia cultures.

Statistical analysis

Statistical analysis was performed to assess statistical significance of the difference between the two groups (control and phage treated) of the in vivo experiment. A two—way ANOVA (factor A: time, factor B: treatment, dependent variable: Vibrio population) was performed following assessment of normality of the data distribution. All statistical analyses were performed using SigmaPlot, version 13.0, Systat Software, Inc., San Jose, California, USA,


Isolation and characterization of bacteriophages φSt2 and φGrn1

Enrichment method was used to process 20 water samples. Two of the filtrates producing zones of clearing were selected for purification and further characterization. Bacteriophages φSt2 and φGrn1 were successfully isolated and purified after five—times plaque re-plating. The presence of phages was confirmed by single negative colonies formation on the bacterial lawn.

The morphology of the virions of φSt2 and φGrn1 under TEM (Fig 1) classified them to Myoviridae family. Both φGrn1 and φSt2 have elongated head and contractile tail. The head of bacteriophage φSt2 was approximately 81 nm wide by 151 nm long and the tail was about 132 nm long with a diameter of 20 nm. In the case of bacteriophage φGrn1, the head was approximately 74 nm wide by 138 nm long and the tail was about 134 nm long with a diameter of 20 nm. Furthermore, both phages possessed a filamentous appendage on the top of the head as observed under TEM examination (Fig 1). Both phages produced similar pinhead’s size clear plaques on the host bacterial lawn.

Fig 1. Transmission Electron Microscopy micrographs of φSt2 (left) and φGrn1 (right).

Both phages have similar morphology and are classified at the Myoviridae family. White arrows indicated the filamentous appendage which is present on both phages’ heads.

Twenty-five bacterial strains were used to define the lytic spectrum of the novel phages φSt2 and φGrn1. Both phages presented the same host range infecting different strains of V. alginolyticus, one strain of V. harveyi and one of V. parahaemolyticus. On the contrary, none of them was able to infect any V. anguillarum, V. ordalii, V. owensii and V. splendidus strains (Table 2). The lytic activity of phages φSt2 and φGrn1 is evident in 40% (10 out of 25) of the bacterial strains tested.

Table 2. Host range of phages φSt2 and φGrn1 against 25 bacterial strains from 7 different Vibrio species: V. anguillarum, V. ordalii, V. harveyi, V. alginolyticus, V. parahaemolyticus, V. splendidus and V. owensii.

Dark grey colour indicates clear lysis, light grey colour indicates turbid lysis and white colour indicates no inhibition. EOP is expressed as: the fraction of phages’ infectivity vs tested strains to phages’ infectivity vs host strain.

Both bacteriophages contain a double stranded DNA sensitive to DNAse and resistant to RNAse A treatment. HpaII, HaeIII and BamHI did not digest the genomes of the phages (Fig 2). However, both phages’ genomes were digested by Sau3AI, HincII and BgIII indicating the genetic differentiation of these phages (Fig 2). According to unpublished results about the sequences of φSt2 and φGrn1 their genome sizes are 250,485 bp and 248,605 bp, respectively.

Fig 2. Restriction endonuclease digestion profile of φSt2 and φGrn1 DNAs.

1: φSt2—HpaII, 2: φGrn1—HpaII, 3: φSt2—Sau3AI, 4: φGrn1—Sau3AI, 5: φSt2—HincII, 6: φGrn1—HincII, 7: φSt2—HaeIII, 8: φGrn1—HaeIII, 9: φSt2—BgIII, 10: φGrn1—BgIII, 11: φSt2—BamHI, 12: φGrn1—BamHI.

The replication parameters of these phages were determined by one step growth curves showed in Fig 3. Both φSt2 and φGrn1 had an approximate latency time of 30 min. However, their burst sizes were quite different since in the case of φSt2 it was 97 pfu / cell whereas in the case of φGrn1 the burst size was 44 pfu / cell.

Fig 3. One step growth curve for bacteriophages φSt2, latency time: 30 min, burst size: 97 phages per cell and φGrn1, latency time: 30 min and burst size: 44 phages per cell.

All values are means ± standard deviation of three independent experiments.

In vitro lytic effect on Vibrio strains

The lytic effect on the host bacteria was tested by infecting fresh cultures of V. alginolyticus strain V1 at the early exponential phase (OD560 ~ 0.15) with φSt2 and φGrn1 separately. Lysis was proportional to the MOI used with the lowest (MOI = 1) resulting to no effect and the highest (MOI = 100) to almost complete inhibition of bacteria growth for the whole duration of the experiment (1000 min). The response of the host was similar for both phages with high reproducibility of the results among the replicates (Fig 4a and 4b).

Fig 4. In vitro cell lysis experiment of phage (a) φSt2 and (b) φGrn1 vs host strain V. alginolyticus V1.

The conditions tested are: control (V1 strain without any phage addition) and 3 different MOI rates used for infecting V1 (MOI = 1, MOI = 10, MOI = 100). The values are means ± standard deviation of the six replicates.

The phages were tested in vitro against twelve bacterial isolates grown in TCBS originated from live feed A. salina culture. A phage mixture of φSt2 and φGrn1 at MOI = 100 significantly affected the growth of all twelve bacterial strains tested (Fig 5a–5l). In all cases, there was a delay in the exponential phase and even when the cultures reached a plateau of growth, the density of phage—treated bacteria was lower compared to their corresponding controls.

Fig 5. (a) A1, (b) A2, (c) A3, (d) A4, € A5, (f) A6, (g) A7, (h) A8, (i) A9, (j) A10, (k) A11 and (l) A12.

In vitro cell lysis experiment of phage mixture (φSt2 and φGrn1) against 12 presumptive Vibrio strains (A1–A12) isolated from the live feed culture, A. salina. The values are means ± standard deviation of the three replicates.

In vivo efficacy of phages to control Vibrio load in A. salina culture

The effect of the phage mixture (φSt2 and φGrn1 at MOI = 100) was in vivo examined by direct administration in live prey cultures, A. salina. At t0 = 0, the presumptive Vibrio count in TCBS plates was 8.1 x 104± 2.7 x 104 cfu mL-1 for the control and 7.6 x 104± 0.6 x 104 cfu mL-1 for the phage-treated cultures with no statistically significant difference (p: 0.782). After 4 h of incubation, presumptive Vibrio count did not change in the control cultures while it became significantly lower in the phage treated ones (p < 0.05). Total Vibrio load in phage treated live prey was 5.3 x 103±3.1 x 103 cfu mL-1, which is approximately 1.3 orders of magnitude or 93% decrease of the initial total presumptive cultivable Vibrio load (Fig 6).

Fig 6. In vivo efficacy levels of phages φSt2 and φGrn1 administered directly in live feed cultures of A. salina.

The values are means ± standard deviation of triplicates experiment (p < 0.05).


Live feed organisms like Artemia are able to bio-accumulate bacteria from the water column [57] acting as a vehicle for pathogen transfer into the hatchery facilities. Bacteriophage therapy in such fragile systems can be a reasonable alternative for the control of microbial diseases and for the prevention of multi—drug resistant bacteria spreading in aquaculture [58]. Since antibiotic resistance mechanisms are irrelevant to mechanisms of phage infection, phages could be successfully employed even against antibiotic—resistant pathogens [59].

In the current study, two novel bacteriophages φSt2 and φGrn1 were isolated and characterized. Both phages produced plaques with identical morphology in host bacterial lawn. Observation under TEM resulted in one common virion morphology, similar to the V. alginolyticus phage φpp2 [60]; all three phages belong to Myoviridae family, but the sizes of φSt2 and φGrn1 virions were much larger than φpp2. Bacteriophage φSt2 is larger than φGrn1 in both head length (151 nm vs 138 nm) and head width (81 nm vs 74 nm). However, both phages had the same tail length and diameter. They are different from other V. alginolyticus phages described in literature like φA318 and VAP11 which are classified as Podoviridae viruses with diameter of head 50–55 nm and 60 nm, respectively [61,62]. The head—associated filamentous appendage that was present on all virions’ heads under TEM, is not a common morphological feature of bacteriophages. However, similar devices have been recorded in Caulobacter crescentus phages acting as tool for the initial attachment of the phages to the bacterial flagella of the host [63]. Whether this is also its function in the case of φSt2 and φGrn1 needs further investigation.

Host range results were identical for φSt2 and φGrn1 suggesting that they recognize identical or very similar structures as receptors in their host. Lytic activity of the phages was not limited to the host strain V1, since they were able to infect all 8 V. alginolyticus strains tested, as well as bacterial strains that belong to V. harveyi and V. parahaemolyticus species. In terms of EOP, φSt2 and φGrn1 were almost equally efficient against all phage—sensitive bacterial strains, producing plaques within one order of magnitude. According to host range and EOP results, bacteriophages φSt2 and φGrn1 are broad host range phages, being able to infect in total 10 out of the 25 strains (40%) tested. However, it should be mentioned that only one strain for V. owensii and one strain of V. parahaemolyticus were examined against φSt2 and φGrn1, thus, their lytic potential against these two harveyi clade bacterial species needs to be further explored. The conservative nature of the structure of LPS phage receptors on the outer membrane of many of Gram negative bacterial species, could be a possible explanation for this broad lytic spectrum [64]. Further, genetic similarity of these bacterial species which belong to the harveyi clade may also have contributed to this observation [65,66]. The best candidates for phage therapy should be viruses able to lyse the majority of the bacterial target’s strains [6769]. Thus, φSt2 and φGrn1 are suggested as potential phage therapy candidates against infections caused by V. alginolyticus, since their lytic spectrum include all the V. alginolyticus strains tested and even strains from other species. Moreover, these phages were able to affect the growth of several strains isolated from an Artemia live feed system. It is known that several marine phages can infect different strains of the same species or different closely related species [53,70], however, specificity of phage-host interactions in marine waters is not fully elucidated and needs further investigation [71]. Another similar case of broad host range Vibrio—phage is φA318, isolated using V. alginolyticus host strain, able to infect several V. harveyi strains [61]. Interestingly, φSt2 and φGrn1 had strong lytic effect also against V. alginolyticus type strain DSM 2171, isolated in Japan from a spoiled horse mackerel responsible for food poisoning [72] expanding therefore their potential applications beyond aquaculture.

RFLP analysis showed that φSt2 and φGrn1 are two distinct phages. Three of the restriction enzymes used could not digest the genome of these phages. Bacteriophages refractory to restriction enzyme activity could be explained by several reasons. Among the most prevalent is the natural loss of restriction sites during evolution [73] but also the integration of unusual bases in the viral DNA [74,75]. More importantly, the expression of specific methyltransferase genes that may be contained in the viral genome will cause epigenetic modifications on the viral DNA affecting the restriction sites that endonucleases can recognize [76]. Comparative genomic analysis of these phages would provide further insights to their genetic identity.

One step growth curves defined the biological parameters of φSt2 and φGrn1 replicative cycle (latency time and burst size). Bacteriophages φSt2 and φGrn1 have similar and relatively short latency time (in both cases about 30 min) compared to other phages [61,77], but differ significantly in their burst sizes (φSt2, 97 phages per cell and φGrn1, 44 phages per cell). Phages with short latency period and high burst size are the most appropriate candidates for phage therapy, however, usually high burst sizes are followed by a more extensive latency period [78]. Thus, φSt2 and φGrn1 can be considered as good candidates for phage therapy applications since they have relatively short latent period and their burst size seems satisfactory apart from their wide host range.

In vitro efficacy of φSt2 and φGrn1 showed that both phages were notably effective against their bacterial host V. alginolyticus strain V1, at MOI = 10 and MOI = 100. On the contrary, at MOI = 1 the effect of phages in the bacterial growth was minimum. According to the literature high MOI rates, usually MOI = 100, are mostly used when phage therapy applied [44,79]. In our case, when MOI = 100 was applied in vitro against V1, the bacterium was almost completely eliminated after 100 min and did not recover until the end of the experiment (1000 min). Both φSt2 and φGrn1 exhibited a very similar but intense bactericidal activity. The use of phage combination is proposed in order to avoid bacterial resistance against phage infection [70], and there are several reports suggesting the use of phage cocktails to control vibriosis [43,44,69].

Prior to the in vivo phage administration, a selection of 12 presumptive Vibrio isolates from the A. salina culture were tested in vitro against the combined lytic activity of φSt2 and φGrn1. All isolates showed a variable degree of sensitivity to the cocktail of φSt2 and φGrn1. Although the strains were not identified to species level it was presumed that they were vibrios based on their efficacy to grow on the Vibrio selective medium but also on colony morphology. These strains were possibly closely related to V. alginolyticus since as stated earlier according to the literature, this species is considered to be the most prevalent bacterial component in A. salina cultures [13,31,32].

In vivo phage therapy application resulted in significant decrease of the presumptive Vibrio load in phage treated live preys. The initial presumptive Vibrio load in phage treated Artemia was decreased by 93% meaning that 93% less Vibrio will enter the hatchery system. Therefore, a treatment with phages in the live feeds prior to their introduction in the hatchery system could effectively reduce the Vibrio load in the larval rearing tanks. Controlling and reducing vibrios in fish and invertebrates hatcheries is critical for the final survival and quality of the produced larvae since the possibility for disease outbreak increases proportionally with the increase of the pathogen load. Our experiment did not show a total elimination of Vibrio in the live feed. Nevertheless, the reduction but not the complete eradication of the vibrios may also be beneficial since at a later stage, the maturing fish immune system will be exposed to low levels of bacteria helping the fish to develop immunity before their transfer to the open sea. From a practical point of view there are unique advantages in the use of phages as a preventive measure against vibriosis. Firstly, phage treatment procedure requires a relatively short period of time (4 h in our case), which offers the opportunity of disinfecting the live prey exactly before its administration to the larvae in the hatchery. The latter is that Artemia is a batch culture and as such, bacterial resistance development against phage infection is rather difficult. A phage treatment protocol could be established on a regular basis for each growth cycle of Artemia and phages should be added to the live prey 4 h before administration to the fish. Besides, the use of phages against vibrios would specifically reduce the load of potentially pathogenic bacteria, leaving other species unaltered. Numerous studies using bacteriophages as treatment agents, verify that applications in many aquaculture systems have been successful against many pathogenic bacteria such as V. harveyi, V. parahaemolyticus and V. anguillarum [43,46,48,58,68].

The apparent lytic nature of the phages (also confirmed by the initial genomic analysis not presented here) is an advantage but also a prerequisite for safe phage therapy development since proliferation of resistant bacteria is considered one of the main drawbacks in the use of bacteriophages to control bacteria [80]. It is commonly accepted that bacteriophages are a key factor to determine the bacterial populations in the environment [81,82]. Drawbacks and issues of concern are the eventual resistance development from bacteria against phage infection and the possibility of prophage induction, mutations and horizontal gene transfer [59]. These are events occurring naturally in the environment which however may be augmented and accelerated with careless use of phages, however variation in phage cocktails has the potential to delay or even inhibit the emergence of phage resistance [83,84].

Currently, it is commonly accepted that the need of an innovative and environmentally friendly alternative to antibiotics has become more than necessary. Presence of bacteriophages in environments where pathogenic bacteria occur should be seen as an opportunity that may lead to the development of a successful innovative and environmentally friendly solution. However, phage therapy should always be applied with great caution. We hope that further knowledge of phage—bacterium interactions will contribute to our ability to handle our own microbiota, when bacteriophages are used as biocontrol agents.


The newly isolated bacteriophages, φSt2 and φGrn1, are characterized by broad host range and compelling biological attributes making them potential candidates for phage therapy application. The positive results obtained by in vitro and in vivo trials, advocate to the fact that phage therapy in aquaculture can be an alternative to antibiotics. Thus, φSt2 and φGrn1 can effectively be used in the biological control of the Vibrio load in marine hatcheries, through of live prey disinfection.

Next step will be the analysis of the full genome sequencing of φSt2 and φGrn1 as well as the further evaluation of the novel disinfection technique on the survival of fish larvae in a controlled experiment.


We would like to thank Prof. Lone Gram from Danish Technical University for providing the VIB391 and DY05 strains and Dr. Frédérique Le Roux (Roscoff Marine Station) for providing the V. splendidus, Dr. Manolis Mandalakis for assisting in in vitro cell lysis assays, Dimitris Skliros for assisting and participating in sequencing analysis of the bacteriophages φSt2 and φGrn1 and Prof. George Chalepakis and the staff of the Laboratory of Electron Microscopy of the University of Crete.

Author Contributions

Conceived and designed the experiments: PK RB. Performed the experiments: PGK CK. Analyzed the data: PGK PK. Contributed reagents/materials/analysis tools: PK. Wrote the paper: PGK PK RB.


  1. 1. Planas M, Pérez-Lorenzo M, Vázquez JA, Pintado J. A model for experimental infections with Vibrio (Listonella) anguillarum in first feeding turbot (Scophthalmus maximus L.) larvae under hatchery conditions. Aquaculture. Elsevier; 2005;250: 232–243.
  2. 2. Talpur AD, Memon AJ, Khan MI, Ikhwanuddin M, Danish Daniel MM, Abol-Munafi AB. A novel of gut pathogenic bacteria of blue swimming crab Portunus pelagicus (Linneaus, 1758) and pathogenicity of Vibrio harveyi a transmission agent in larval culture under hatchery conditions. Res J Appl Sci. 2011;6: 116–127.
  3. 3. Goulden EF, Hall MR, Bourne DG, Pereg LL, Høj L. Pathogenicity and Infection Cycle of Vibrio owensii in Larviculture of the Ornate Spiny Lobster (Panulirus ornatus). Appl Environ Microbiol. 1752 N St., N.W., Washington, DC: American Society for Microbiology; 2012;78: 2841–2849.
  4. 4. Schiewe MH, Trust TJ, Crosa JH. Vibrio ordalii sp. nov.: A causative agent of vibriosis in fish. Curr Microbiol. 1981;6: 343–348.
  5. 5. Toranzo AE, Magariños B, Romalde JL. A review of the main bacterial fish diseases in mariculture systems. Aquaculture. 2005;246: 37–61.
  6. 6. Reid HI, Treasurer JW, Adam B, Birkbeck TH. Analysis of bacterial populations in the gut of developing cod larvae and identification of Vibrio logei, Vibrio anguillarum and Vibrio splendidus as pathogens of cod larvae. Aquaculture. Elsevier B.V.; 2009;288: 36–43.
  7. 7. Torrecillas S, Makol A, Caballero MJ, Montero D, Dhanasiri KS, Sweetman J, et al. Effects on mortality and stress response in European sea bass, Dicentrarchus labrax (L.), fed mannan oligosaccharides (MOS) after Vibrio anguillarum exposure. J Fish Dis. 2012;35: 591–602. pmid:22690841
  8. 8. Lopes CM, Rabadão EM, Ventura C, da Cunha S, Côrte-Real R, Meliço-Silvestre AA. A Case of Vibrio alginolyticus Bacteremia and Probable Sphenoiditis Following a Dive in the Sea. Clin Infect Dis. 1993;17: 299–300. pmid:8399897
  9. 9. Scheftel JM, Ashkar K, Boeri C, Monteil H. Phlegmon au doigt à Vibrio alginolyticus consécutif à une blessure chez un patient de retour du Maroc. Journées Francophones de Microbiologie des Milieux Hydriques, 23–24 novembre 2006, Agadir, Maroc. 2006.
  10. 10. Lee DY, Moon SY, Lee SO, Yang HY, Lee HJ, Lee MS. Septic shock due to Vibrio alginolyticus in a cirrhotic patient: The first case in Korea. Yonsei Med J. 2008;49: 329–332. pmid:18452273
  11. 11. Sabir M, Ennaji Moulay M, Cohen N. Vibrio Alginolyticus: An Emerging Pathogen of Foodborne Diseases. Int J Sci Technol. 2013;2: 302–309.
  12. 12. Skjermo J, Vadstein O. Characterization of the bacterial flora of mass cultivated Brachionus plicatilis. In: Gilbert JJ, Lubzens E, Miracle MR, editors. Rotifer Symposium VI SE—25. Springer Netherlands; 1993. pp. 185–191.
  13. 13. Villamil L, Figueras A, Planas M, Novoa B. Control of Vibrio alginolyticus in Artemia culture by treatment with bacterial probiotics. Aquaculture. 2003;219: 43–56.
  14. 14. Kong W, Huang L, Su Y, Qin Y, Ma Y, Xu X, et al. Investigation of possible molecular mechanisms underlying the regulation of adhesion in Vibrio alginolyticus with comparative transcriptome analysis. Antonie Van Leeuwenhoek. 2015; 1197–1206.
  15. 15. Austin B, Austin DA. Vibrionaceae representatives. Bacterial Fish Pathogens. 2012. pp. 357–411.
  16. 16. Colorni A, Paperna I, Gordin H. Bacterial infections in gilt-head sea bream Sparus aurata cultured at Elat. Aquaculture. 1981;23: 257–267.
  17. 17. Balebona MC, Andreu MJ, Bordas M a, Zorrilla I, Moriñigo M, Borrego JJ. Pathogenicity of Vibrio alginolyticus for cultured gilt-head sea bream (Sparus aurata L.). Appl Environ Microbiol. 1998;64: 4269–4275. pmid:9797276
  18. 18. Zorilla I, Chabrillon M, Arijo S, Dıaz-Rozales P, Martinez-Manzanares E, Balebona MC, et al. Bacteria recovered from diseased cultured gilthead sea bream (Sparus aurata L.) in southwestern Spain. Aquaculture. 2003;218: 11–20.
  19. 19. Varvarigos P. Vibrio alginolyticus combined with the motile Aeromonads Aeromonas sobria and Aeromonas hydrophila comprise the main bacterial taxa responsible for gilthead sea bream (Sparus aurata) larval enteropathy (LE). European Aquaculture Fish Pathology 13th International Conference of Fish and Shellfish diseases. 2007. p. O–126.
  20. 20. Akayli T, Timur G, Aydemir B, Kiziler AR, Coskun O, Albayrak G, et al. Characterization of Vibrio alginolyticus isolates from diseased cultured gilthead sea bream, Sparus aurata. Isr J Aquac. 2008;60: 89–94.
  21. 21. Athanassopoulou F, Prapas T, Rodger H. Diseases of Puntazzo puntazzo Cuvier in marine aquaculture systems in Greece. J Fish Dis. 1999;22: 215–218.
  22. 22. Sonia GAS, Lipton AP. Pathogenicity and antibiotic susceptibility of Vibrio species isolated from the captive-reared tropical marine ornamental blue damsel fish, Pomacentrus caeruleus (Quoy and Gaimard, 1825). Indian J Mar Sci. 2012;41: 348–354.
  23. 23. Martins ML, Mouriño JLP, Fezer GF, Buglione Neto CC, Garcia P, Silva BC, et al. Isolation and experimental infection with Vibrio alginolyticus in the sea horse, Hippocampus reidi Ginsburg, 1933 (Osteichthyes: Syngnathidae) in Brazil. Braz J Biol. 2010;70: 205–209. pmid:20231979
  24. 24. Balcázar JL, Gallo-Bueno A, Planas M, Pintado J. Isolation of Vibrio alginolyticus and Vibrio splendidus from captive-bred seahorses with disease symptoms. Antonie van Leeuwenhoek, Int J Gen Mol Microbiol. 2010;97: 207–210.
  25. 25. Lee K-K, Yu S-R, Chen F-R, Yang T-I, Liu P-C. Virulence of Vibrio alginolyticus Isolated from Diseased Tiger Prawn, Penaeus monodon. Curr Microbiol. Springer-Verlag; 1996;32: 229–231.
  26. 26. Jayaprakash NS, Rejish Kumar VJ, Philip R, Bright Singh IS. Vibrios associated with Macrobrachium rosenbergii (De Man, 1879) larvae from three hatcheries on the Indian southwest coast. Aquac Res. Blackwell Publishing Ltd; 2006;37: 351–358.
  27. 27. Munro PD, Birkbeck TH, Barbour A. Influence of rate of bacterial colonisation of the gut of turbot larvae on larval survival. Reinertsen H, Dahle LA, Jørgensen L, Tvinnereim K (Eds) Fish Farming Technology AA Balkema, Rotterdam. 1993. pp. 85–92.
  28. 28. Prol-García MJ, Planas M, Pintado J. Different colonization and residence time of Listonella anguillarum and Vibrio splendidus in the rotifer Brachionus plicatilis determined by real-time PCR and DGGE. Aquaculture. Elsevier B.V.; 2010;302: 26–35.
  29. 29. Snoussi M, Chaieb K, Mahmoud R, Bakhrouf A. Quantitative study, identification and antibiotics sensitivity of someVibrionaceae associated to a marine fish hatchery. Ann Microbiol. 2006;56: 289–293.
  30. 30. Dhont J, Dierckens K, Støttrup J, Van Stappen G, Wille M, Sorgeloos P. Advances in Aquaculture Hatchery Technology [Internet]. Advances in Aquaculture Hatchery Technology. Elsevier; 2013.
  31. 31. Thomson R, Macpherson HL, Riaza A, Birkbeck TH. Vibrio splendidus biotype 1 as a cause of mortalities in hatchery-reared larval turbot, Scophthalmus maximus (L.). J Appl Microbiol. 2005;99: 243–250. pmid:16033454
  32. 32. Høj L, Bourne DG, Hall MR. Localization, abundance and community structure of bacteria associated with Artemia: Effects of nauplii enrichment and antimicrobial treatment. Aquaculture. Elsevier B.V.; 2009;293: 278–285.
  33. 33. Schulze AD, Alabi AO, Tattersall-Sheldrake AR, Miller KM. Bacterial diversity in a marine hatchery: Balance between pathogenic and potentially probiotic bacterial strains. Aquaculture. 2006;256: 50–73.
  34. 34. Olafsen J a. Interactions between fish larvae and bacteria in marine aquaculture. Aquaculture. 2001;200: 223–247.
  35. 35. Kennedy B, Venugopal MN, Karunasagar I, Karunasagar I. Bacterial flora associated with the giant freshwater prawn Macrobrachium rosenbergii, in the hatchery system. Aquaculture. 2006;261: 1156–1167.
  36. 36. Akinbowale OL, Peng H, Barton MD. Antimicrobial resistance in bacteria isolated from aquaculture sources in Australia. J Appl Microbiol. 2006;100: 1103–1113. pmid:16630011
  37. 37. Heuer OE, Kruse H, Grave K, Collignon P, Karunasagar I, Angulo FJ. Human Health Consequences of Use of Antimicrobial Agents in Aquaculture. Clin Infect Dis. Oxford University Press; 2009;49: 1248–1253.
  38. 38. Sørum H, L’Abée-Lund TM. Antibiotic resistance in food-related bacteria—a result of interfering with the global web of bacterial genetics. Int J Food Microbiol. 2002;78: 43–56. pmid:12222637
  39. 39. Park SC, Shimamura I, Fukunaga M, Mori K-I, Nakai T. Isolation of bacteriophages specific to a fish pathogen, Pseudomonas plecoglossicida, as a candidate for disease control. Appl Environ Microbiol. 2000;66: 1416–1422. pmid:10742221
  40. 40. Perreten V. Resistance in the food chain and in bacteria from animals: relevance to human infections. White DG, Alekshun MN, McDermott PF (eds) Frontiers in antimicrobial resistance American Society for Microbiology, Washington, DC. 2005.
  41. 41. Oliveira J, Castilho F, Cunha A, Pereira MJ. Bacteriophage therapy as a bacterial control strategy in aquaculture. Aquac Int. 2012;20: 879–910.
  42. 42. Golkar Z, Bagasra O, Gene Pace D. Bacteriophage therapy: A potential solution for the antibiotic resistance crisis. J Infect Dev Ctries. 2014;8: 129–136. pmid:24518621
  43. 43. Karunasagar I, Shivu MM, Girisha SK, Krohne G, Karunasagar I. Biocontrol of pathogens in shrimp hatcheries using bacteriophages. Aquaculture. 2007;268: 288–292.
  44. 44. Silva YJ, Costa L, Pereira C, Mateus C, Cunha A, Calado R, et al. Phage therapy as an approach to prevent Vibrio anguillarum infections in fish larvae production. PLoS One. Public Library of Science; 2014;9: e114197.
  45. 45. Raghu Patil J, Desai SN, Roy P, Durgaiah M, Saravanan RS, Vipra A. Simulated hatchery system to assess bacteriophage efficacy against Vibrio harveyi. Dis Aquat Organ. Inter-Research; 2014;112: 113–9.
  46. 46. Lomelí-Ortega CO, Martínez-Díaz SF. Phage therapy against Vibrio parahaemolyticus infection in the whiteleg shrimp (Litopenaeus vannamei) larvae. Aquaculture. Elsevier; 2014;434: 208–211.
  47. 47. Martínez-Díaz SF, Hipólito-Morales A. Efficacy of phage therapy to prevent mortality during the vibriosis of brine shrimp. Aquaculture. 2013;400–401: 120–124.
  48. 48. Higuera G, Bastías R, Tsertsvadze G, Romero J, Espejo RT. Recently discovered Vibrio anguillarum phages can protect against experimentally induced vibriosis in Atlantic salmon, Salmo salar. Aquaculture. 2013;392–395: 128–133.
  49. 49. Castillo D, D’Alvise P, Kalatzis PG, Kokkari C, Middelboe M, Gram L, et al. Draft Genome Sequences of Vibrio alginolyticus Strains V1 and V2, Opportunistic Marine Pathogens. Genome Announc. 2015;3.
  50. 50. Di Pinto A, Ciccarese G, Tantillo G, Catalano D, Forte VT. A collagenase-targeted multiplex PCR assay for identification of Vibrio alginolyticus, Vibrio cholerae, and Vibrio parahaemolyticus. J Food Prot. 2005;68: 150–153. pmid:15690817
  51. 51. Pang L, Zhang X-H, Zhong Y, Chen J, Li Y, Austin B. Identification of Vibrio harveyi using PCR amplification of the toxR gene. Lett Appl Microbiol. Blackwell Publishing Ltd; 2006;43: 249–255.
  52. 52. Hong GE, Kim DG, Bae JY, Ahn SH, Bai SC, Kong IS. Species-specific PCR detection of the fish pathogen, Vibrio anguillarum, using the amiB gene, which encodes N-acetylmuramoyl-L-alanine amidase. FEMS Microbiol Lett. 2007;269: 201–206. pmid:17326755
  53. 53. Comeau AM, Chan AM, Suttle C. Genetic richness of vibriophages isolated in a coastal environment. Environ Microbiol. 2006;8: 1164–1176. pmid:16817925
  54. 54. Abedon ST. Lysis from without. Bacteriophage. 2011;1: 46–49. pmid:21687534
  55. 55. Kutter E. Phage Host Range and Efficiency of Plating. Bacteriophages: Methods 448 and Protocols, Volume 1: Isolation, Characterization, and Interactions ed Clokie M, New York: Springer Science. 2009. pp. 141–149.
  56. 56. Tan D, Gram L, Middelboe M. Vibriophages and their interactions with the fish pathogen Vibrio anguillarum. Appl Environ Microbiol. 2014;80: 3128–40. pmid:24610858
  57. 57. Goulden EF, Høj L, Hall MR. Advances in Aquaculture Hatchery Technology [Internet]. Advances in Aquaculture Hatchery Technology. Elsevier; 2013.
  58. 58. Nakai T, Park SC. Bacteriophage therapy of infectious diseases in aquaculture. Res Microbiol. 2002;153: 13–18. pmid:11881893
  59. 59. Loc-Carrillo C, Abedon ST. Pros and cons of phage therapy. Bacteriophage. Taylor & Francis; 2011;1: 111–114.
  60. 60. Lin Y, Lin C. Genome-wide characterization of Vibrio phage Phi-pp 2 with unique arrangements of the mob-like genes. Bio Med Cent Genomics. 2012;
  61. 61. Lin Y-R, Chiu C-W, Chang F-Y, Lin C-S. Characterization of a new phage, termed ϕA318, which is specific for Vibrio alginolyticus. Arch Virol. Springer Vienna; 2012;157: 917–926.
  62. 62. Heo YJ, Lee CH, Baek MS, Ahn HM, Hwang YS, Park KH, et al. Morphological characterization of Vibrio alginolyticus specific bacteriophage isolated from fish farms on west coast of Korea. J Fish Pathol Korean Soc Fish Pathol. 2012;25: 165–172.
  63. 63. Guerrero-Ferreira RC, Viollier PH, Ely B, Poindexter JS, Georgieva M, Jensen GJ, et al. Alternative mechanism for bacteriophage adsorption to the motile bacterium Caulobacter crescentus. Proc Natl Acad Sci. 2011;108: 9963–9968. pmid:21613567
  64. 64. Rakhuba DV, Kolomiets EI, Szwajcer Dey E, Novik GI. Bacteriophage receptors, mechanisms of phage adsorption and penetration into host cell. Polish J Microbiol. 2010;59: 145–155.
  65. 65. Hoffmann M, Monday SR, Fischer M, Brown EW. Genetic and phylogenetic evidence for misidentification of Vibrio species within the Harveyi clade. Lett Appl Microbiol. Blackwell Publishing Ltd; 2012;54: 160–165.
  66. 66. Urbanczyk H, Ogura Y, Hayashi T. Taxonomic revision of Harveyi clade bacteria (family Vibrionaceae) based on analysis of whole genome sequences. Int J Syst Evol Microbiol. 2013;63: 2742–2751. pmid:23710045
  67. 67. Sulakvelidze A. The challenges of bacteriophage therapy. Ind Pharm. 2011;45: 14–18.
  68. 68. Vinod MG, Shivu MM, Umesha KR, Rajeeva BC, Krohne G, Karunasagar I, et al. Isolation of Vibrio harveyi bacteriophage with a potential for biocontrol of luminous vibriosis in hatchery environments. Aquaculture. 2006;255: 117–124.
  69. 69. Chan BK, Abedon ST, Loc-carrillo C. Phage cocktails and the future of phage therapy. Future Microbiol. 2013; 769–783. pmid:23701332
  70. 70. Crothers-Stomps C, Høj L, Bourne DG, Hall MR, Owens L. Isolation of lytic bacteriophage against Vibrio harveyi. J Appl Microbiol. 2010;108: 1744–1750. pmid:19886890
  71. 71. Holmfeldt K, Middelboe M, Nybroe O, Riemann L. Large variabilities in host strain susceptibility and phage host range govern interactions between lytic marine phages and their Flavobacterium hosts. Appl Environ Microbiol. 2007;73: 6730–6739. pmid:17766444
  72. 72. Miyamoto Y, Nakamuma K, Takizawa K. Pathogenic Halophiles. Proposals Of A New Genus “Oceanomonas” And Of The Amended Species Names. Jpn J Microbiol. 1961;5: 477–486.
  73. 73. Moineau S, Pandian S, Klaenhammer TR. Restriction/modification systems and restriction endonucleases are more effective on lactococcal bacteriophages that have emerged recently in the dairy industry. Appl Environ Microbiol. 1993;59: 197–202. pmid:16348842
  74. 74. Jensen EC, Schrader HS, Rieland B, Thompson TL, Lee KW, Nickerson KW, et al. Prevalence of broad-host-range lytic bacteriophages of Sphaerotilus natans, Escherichia coli, and Pseudomonas aeruginosa. Appl Environ Microbiol. 1998;64: 575–580. pmid:9464396
  75. 75. Shivu MM, Rajeeva BC, Girisha SK, Karunasagar II, Krohne G, Karunasagar II. Molecular characterization of Vibrio harveyi bacteriophages isolated from aquaculture environments along the coast of India. Environ Microbiol. 2007;9: 322–331. pmid:17222131
  76. 76. Sails D, Wareing DR a, Bolton FJ, Fox J, Curry A. Characterisation of 16 Campylobacter jejuni and C. coli typing bacteriophages. J Med Microbiol. 1998;47: 123–128. pmid:9879954
  77. 77. Phumkhachorn P, Rattanachaikunsopon P. Isolation and partial characterization of a bacteriophage infecting the shrimp pathogen Vibrio harveyi. Afr J Microbiol 2010;4: 1794–1800.
  78. 78. Abedon ST, Herschler TD, Stopar D. Bacteriophage Latent-Period Evolution as a Response to Resource Availability. Appl Environ Microbiol. American Society for Microbiology; 2001;67: 4233–4241.
  79. 79. Mateus L, Costa L, Silva YJ, Pereira C, Cunha A, Almeida A. Efficiency of phage cocktails in the inactivation of Vibrio in aquaculture. Aquaculture. 2014;424–425: 167–173.
  80. 80. Levin BR, Bull JJ. Population and evolutionary dynamics of phage therapy. Nat Rev Micro. 2004;2: 166–173.
  81. 81. Jensen MA, Faruque SM, Mekalanos JJ, Levin BR. Modeling the role of bacteriophage in the control of cholera outbreaks. Proc Natl Acad Sci United States Am. 2006;103: 4652–4657.
  82. 82. Rodriguez-Valera F, Martin-Cuadrado A-B, Rodriguez-Brito B, Pasic L, Thingstad TF, Rohwer F, et al. Explaining microbial population genomics through phage predation. Nat Rev Micro. Nature Publishing Group; 2009;7: 828–836.
  83. 83. Tanji Y, Shimada T, Fukudomi H, Miyanaga K, Nakai Y, Unno H. Therapeutic use of phage cocktail for controlling Escherichia coli O157:H7 in gastrointestinal tract of mice. J Biosci Bioeng. 2005;100: 280–7. pmid:16243277
  84. 84. Meaden S, Koskella B. Exploring the risks of phage application in the environment. Front Microbiol. 2013;4: 358. pmid:24348468