Andrey A Filippov*, Kirill V Sergueev and Mikeljon P Nikolich
Department of Emerging Bacterial Infections, Bacterial Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, Maryland, USA
Received Date: December 27, 2012; Accepted Date: January 18, 2013; Published Date: January 21, 2013
Citation: Filippov AA, Sergueev KV, Nikolich MP (2013) Bacteriophages against Biothreat Bacteria: Diagnostic, Environmental and Therapeutic Applications. J Bioterr Biodef S3:010. doi:10.4172/2157-2526.S3-010
Copyright: © 2013 Filippov AA, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Bioterrorism & Biodefense
Plague, anthrax and brucellosis are severe bacterial infections presenting a serious threat to public health. Their causative agents can be weaponized and a number of drug-resistant strains have been described. This requires improvement of existing and development of new methods of diagnostics, strain characterization, prophylaxis and therapy of these infections. This review article focuses on lytic bacteriophages (phages) active against Yersinia pestis, Bacillus anthracis and Brucella including the uses of phages for diagnostics, strain typing, specific decontamination, and antibacterial therapy.
Yersinia pestis; Bacillus anthracis; Brucella; Bacteriophages; Phage diagnostics; Phage typing; Phage decontamination; Phage therapy
ATP: Adenosine Triphosphate; CDC: Centers for Disease Control and Prevention; CFU: Colony-forming Unit; EOP: Efficiency of Plating; FDA: Food and Drug Administration; FSU: Former Soviet Union; h: hour; LD50: 50% Lethal Dose; LPS: Lipopolysaccharide; PCR: Polymerase Chain Reaction; qPCR: Quantitative Real-time Polymerase Chain Reaction; PFU: Plaque-forming Unit; RTD: Routine Test Dilution; sbsp.: subspecies; UA: Unavailable
Bioterrorism or biothreat agents are highly infectious and pathogenic microorganisms (bacteria, viruses, and fungi) and their toxins that can be used by individuals or groups of terrorists or as biowarfare agents in military operations. The Centers for Disease Control and Prevention (CDC) has established two groups of the most dangerous biothreat agents, categories A and B [1,2]. The highest-priority category A, in addition to botulinum toxin and several especially dangerous viruses, includes three biothreat bacteria, the causative agents of bubonic and pneumonic plague (Yersinia pestis), anthrax (Bacillus anthracis) and tularemia (Francisella tularensis). Category B comprises several toxins, viruses and bacterial pathogens including the causative agents of brucellosis (Brucella species), glanders (Burkholderia mallei), and melioidosis (Burkholderia pseudomallei). In this review, we focus on plague, anthrax and brucellosis. Plague [3] and anthrax [4] are severe fulminant primarily zoonotic infections that can be transmitted to humans and easily disseminated in human populations, mainly via natural or artificial aerosolization of the bacteria, resulting in epidemics with high mortality rates. Brucellosis is a globally spread zoonosis which is also prevalent in humans in the form of severe systemic disease. There are four species of Brucella pathogenic for humans, Brucella melitensis, Brucella suis, Brucella abortus, and Brucella canis [5]. The serious public health concern about these biothreat infections is aggravated by the emergence of multidrug-resistant strains of Y. pestis [6,7] and isolates of Brucella resistant to some clinically relevant antibiotics [8,9], as well as by easy induction of antibiotic resistance in B. anthracis [10], and a potential for the application of genetically engineered multidrugresistant strains of biothreat bacteria in biological attacks [11-13]. The severity of these biothreat bacterial infections requires efficient biosurveillance and biodefense, including availability of a rich arsenal of up-to-date methods of rapid detection and identification of the bacteria, strain characterization, diagnostics, efficient prophylaxis, and therapy of these infections. Many bacterial viruses (bacteriophages or phages) active against Y. pestis, B. anthracis and Brucella species have been described. Characteristics and practical applications of such phages are the subject of this review. Due to the lack of data on lytic phages of F. tularensis and very limited information on practical importance of phages capable of lysing B. pseudomallei and B. mallei, this review does not cover the literature on these bacteria.
The Most important bacteriophages lytic for Y. pestis, B. anthracis and Brucella species are listed in table 1. The listed bacteriophages are members of the same order, Caudovirales (tailed viruses) but belong to five different families (Myoviridae, Podoviridae, Siphoviridae, Tectiviridae, and Inoviridae) and different groups. The first three families have different tail structures: long contractile, short noncontractile, or long noncontractile tails, respectively. The features of the Tectiviridae phages are a double coat and a tail equivalent, and the Inoviridae family representatives have a filamentous structure. Phages of these families contain linear double-stranded DNA, except the Inoviridae members which carry circular single-stranded DNA [83]. The genomes of many phages active against biothreat bacteria were sequenced (Table 1). The bacterial cell surface receptors were identified for Y. pestis-splecific phages φA1122 [16,84], Pokrovskaya, L-413C, Y, φJA1, R, and PST [16], mainly in different sugar residues of the LPS outer and inner core. It was found that φA1122 is specific for Y. pestis when grown at 26-28°C but can efficiently lyse both Y. pestis and Yersinia pseudotuberculosis (the closest phylogenetic relative of Y. pestis) at 37°C [17,20-22,84], due to repressed production of O-antigen which blocks the Y. pseudotuberculosis phage receptor in the LPS inner core when grown at lower temperatures [84]. The receptor for B. anthracis-specific phage gamma was localized in a cell surface-anchored protein, GamR [85]. The AP50 phage of B. anthracis was recently shown to require for its adsorption the intact structure of another cell surface-anchored protein, CsaB, important for S-layer assembly [53]. Since the main difference between smooth and rough strains of Brucella is the presence or absence of the lipopolysaccharide (LPS) O antigen, receptors of brucellaphages should at least partially involve LPS; the receptors for S-specific phages of Brucella were roughly associated with a protein-phospholipid-LPS complex [79,86].
Host bacterium | Bacteriophage | Family | Group | Sequence No. | Reference(s) |
---|---|---|---|---|---|
Y. pestis | Pokrovskaya (YepE2, YpP-G) | Podoviridae | T7 | NC_011038; JQ9657021 | [14-18] |
Y. pestis | φA1122 | Podoviridae | T7 | NC_004777 | [16,17,19-23] |
Y. pestis | Y | Podoviridae | T7 | JQ9657001 | [16-18,24,25] |
Y. pestis2 | R | Podoviridae | T7 | JQ9657011 | [16-18,24,26-28] |
Y. pestis2 | d’Herelle-m (YpsP-G) | Podoviridae | T7 | JQ9657031 | [17,18,27,29,30] |
Y. pestis | Yep-phi | Podoviridae | T7 | HQ333270 | [31] |
Y. pestis2 | PST | Myoviridae | T4 | UA1 | [16,17,24,32,33] |
Y. pestis | φJA1 | Myoviridae | T4 | UA1 | [16,17] |
Y. pestis3 | PY100 | Siphoviridae | T1 | AM076770 | [34] |
Y. pestis | L-413C | Myoviridae | P2 | NC_004745 | [15-17,22,23,35,36] |
B. anthracis4 | β (Wβ; beta) | Siphoviridae | γ | DQ289555 | [37-40] |
B. anthracis4 | γ (Wγ; gamma) | Siphoviridae | γ | DQ221100 DQ222851 DQ222855 DQ289556 NC_007458 | [38,39,41-47] |
B. anthracis | Fah | Siphoviridae | γ | NC_007814 | [46-48] |
B. anthracis | AP50 | Tectiviridae | ? | NC_011523 | [49-53] |
B. anthracis5 | 8a | Myoviridae | ? | UA | [54] |
B. anthracis | Nk | Myoviridae | ? | UA | [55] |
B. anthracis | DB | Myoviridae | ? | UA | [55] |
B. anthracis | MH | Podoviridae | ? | UA | [55] |
B. anthracis | JRB7 | Inoviridae | ? | UA | [56,57] |
B. anthracis | K | ? | ? | UA | [43,47,58,59] |
B. anthracis | VA-9 | ? | ? | UA | [38,47] |
B. abortus6 | Tb (Tbilisi) | Podoviridae | Tb | NC_019446 | [60-67] |
B. abortus7 | Np (Nepean) | Podoviridae | Tb | UA | [65] |
B. abortus8 | Fz (Firenze) | Podoviridae | Tb | UA9 | [63,65,68-70] |
B. abortus10 | Pr (Perote) | Podoviridae | Tb | NC_019447 | [67] |
B. suis11 | Wb (Weybridge) | Podoviridae | Tb | UA9 | [63-65,69-71] |
B. suis11 | S708 | Podoviridae | Tb | UA9 | [69,70,72,73] |
B. suis12 | 1330 | Podoviridae? | Tb? | UA | [74-76] |
B. melitensis13 | Bk (Berkeley) | Podoviridae | Tb | UA9 | [63,64,70,77] |
B. melitensis14 | 311 | Podoviridae? | Tb? | UA | [75,76,78] |
B. abortus15 | R/O | Podoviridae | Tb | UA | [64,70,79-81] |
B. abortus16 | R/C | Podoviridae | Tb | UA9 | [65,70,79-81] |
B. melitensis17 | Iz (Izatnagar) | Podoviridae | Tb | UA | [65,70,82] |
Notes: UA, unavailable. 1We (A.A. Filippov, K.V. Sergueev, and M.P. Nikolich, unpublished data) recently sequenced the genomes of PST and φJA1 (for the first time),
as well as of the Pokrovskaya, Y, R, and d’Herelle-m phages. The last four genomes were virtually identical to the sequences JQ965702, JQ965700, JQ965701, and
JQ965703, respectively, published by another group [18].
2Phages R, d’Herelle-m and PST have slightly higher efficiencies of plating (EOP) on Yersinia pseudotuberculosis than on Y. pestis [17].
3Phage PY100 lyses cultures of Y. pestis and several other species of Yersinia [34].
4Both β and γ phages are lytic derivatives of a temperate phage
W isolated from Bacillus cereus [37,41].
58a is a spore-binding phage [54].
6Tb lyses B. abortus at both RTD (routine test dilution, the highest phage dilution producing
semiconfluent lysis) and 104 RTD, as well as lyses B. suis and B. neotomae only at 104 RTD [70].
7Np is highly specific for B. abortus [65].
8Fz lyses B. abortus and B.
neotomae (at RTD and 104 RTD) and B. suis (only at 104 RTD) [70].
9We (J. Farlow, A.A. Filippov, K.V. Sergueev, and M.P. Nikolich, manuscript in preparation) recently
sequenced the genomes of brucellaphages Fz, Wb, S708, Bk, and R/C and observed a high level of identity with previously sequenced phages Tb and Pr [67].
10Pr produces clear plaques on B. abortus and B. suis and turbid plaques on B. melitensis [67].
11Wb and S708 lyse B. suis, B. abortus and B. neotomae (at RTD and 104 RTD);
Wb can lyse some B. melitensis strains (also both at RTD and 104 RTD) [70].
12Phage 1330 has a host range similar to that of Wb and S708 [74,76].
13Bk lyses B. melitensis, B. abortus, B. suis, and B. neotomae at both titers [70,77].
14?Phage 311 has a host range similar to that of Bk [75,76].
15?R/O lyses B. ovis and rough strains of B. abortus at
RTD and 104 RTD [70,80,81].
16R/C lyses B. ovis, B. canis and rough strains of B. abortus at both concentrations [70,80,81].
17Iz lyses smooth B. abortus, as well as both
smooth (better) and rough (less efficiently) strains of B. neotomae, B. suis and B. melitensis [70,82].
Table 1: Bacteriophages active against biothreat bacteria.
Lytic Y. pestis phages were isolated from animals infected with plague [87] or their carcasses [14], blood of patients [19], rodent burrow soil [88], and from sewage [16,17]. Several groups of temperate phages were isolated from Y. pestis cultures with subsequent selection of virulent (lytic) mutants from some of them [35,89-91]. Phages capable of lysing Y. pestis are similar to T7, T4, T1 or P2 phages that typically grow on enteric bacteria such as Escherichia coli and Shigella, with a prevalence of highly lytic T7-like phages (Table 1). Some of “plague phages” can be considered host range mutants of coliphages. For example, we isolated a Y. pestis-specific T4-like phage φJA1 from Maryland sewage [16,17].
The sources of isolation of anthrax phages were lysogenic cultures of B. anthracis [92-94] and Bacillus cereus [37,41], soil [49,54,55,95], sewage [96,97], and the intestinal tract of an earthworm [98]. These phages are very variable and belong to all the five above-mentioned families, however, only genomes of three B. anthracis-specific phages have been sequenced to date (not including prophages; see table 1). Interestingly, phage gamma contains a functional fosfomycin resistance gene, which is evidence of lysogenic conversion and could be the cause of the spread of fosfomycin resistance among B. anthracis strains that is, in turn, important for survival in soil in the presence of Streptomyces producing this antibiotic [39]. Moreover, the infection with temperate phages and lysogenization of B. anthracis can cause multiple phenotypic alterations: blockade or promotion of sporulation, induction of exopolysaccharide production and biofilm formation associated with the ability for long-term colonization of soil and the earthworm gut in a non-sporulating form [99].
The majority of Brucella typing phages were isolated from the bacterial cultures including B. abortus [65,68,72,100], B. suis [71-74], and B. melitensis [72,77,78,101] or host range mutants were derived from known phages following propagation on the bacterial cultures of interest, sometimes with the use of chemical mutagenesis [69,79,80,102]. Phages were isolated from the blood of 16 brucellosis patients out of 23 tested and in one case of severe neurobrucellosis, the phage was isolated both from blood and cerebrospinal fluid [103]. Several brucellaphages were isolated from aborted fetuses of farm animals [104], milk, urine, and feces of bovines [105] and farm sewage [60]. In contrast to the phages of Y. pestis and B. anthracis, brucellaphages have very similar or nearly identical morphology and a high degree of DNA homology despite their differences in the sources of isolation and host ranges; they all seem to belong to the same group of Tb-like phages (Table 1). A very interesting feature was discovered in phages Tb and Pr: the presence of putative DnaA-like protein and oriC-like origin of replication in their genomes suggesting that these phages can replicate by the same mechanism as bacterial chromosomes [67]. We sequenced five more brucellaphages (Fz, Wb, S708, Bk, and R/C) and found the same conserved sequences (J. Farlow, A.A. Filippov, K.V. Sergueev, and M.P. Nikolich, manuscript in preparation) that makes this group unique in the phage world.
Routine bacteriophage lysis tests have been used for identification of biothreat bacteria, discrimination of relative species and differentiation of typical and atypical strains for more than 80 years [96,100,106]. In particular, the phage lysis assay is an essential part of Y. pestis identification and bacteriological diagnosis of plague [107,108]. The specificity of four routinely used plague diagnostic phages was demonstrated on thousands of strains: Pokrovskaya and L-413C [15,107], φA1122 [20,108], and Yep-phi [31]. These phages are capable of lysing practically all Y. pestis strains and allow differentiation from Y. pseudotuberculosis. Phage φA1122 is recommended for plague diagnosis in the USA by the CDC, whereas Pokrovskaya and L-413C are widely used in the FSU countries, and Yep-phi is applied in China. We performed efficiency of plating (EOP) tests for 11 phages on Y. pestis, Y. pseudotuberculosis and E. coli and showed that the Y and φJA1 phages also have high diagnostic value [17]. Phages can be used not only for identification but also for typing of Y. pestis. For example, LPS-specific phages FP1 and FP3 were successfully used for differentiation of “main subspecies”, Y. pestis sbsp. pestis, from sbsp. microti and discrimination of several biovars within the latter subspecies [109].
A standard test of phage lysis requires a pure Y. pestis culture that extends this assay time up to two or three days [107,108]. We recently developed a rapid (4 h) and simple indirect phage-based method of detection and identification of Y. pestis [22]. The approach is based on the detection of φA1122 or L-413C amplification in the presence of Y. pestis by quantitative real-time PCR (qPCR) targeting the phage DNA but omitting the step of DNA isolation. The method is more specific than PCR tests targeting bacterial DNA, because in this case qPCR registers the titer increase of specific phage amplifying if live Y. pestis cells are available. The detection limit was as low as one bacterial cell per 1 μl sample (103 CFU/ml) and can be even lowered by sample concentration. The clinical relevance of this method was demonstrated on artificially spiked samples of human blood [22], sheep serum and mouse spleen suspensions with as high serum assay sensitivity as 104 CFU/ml with sample preparation and DNA isolation steps omitted [110]. We also proved that this method works well with the use of JBAIDS platform (Idaho Technology Inc.), a portable field-deployable thermocycler, with the same sensitivity as when using a full-sized qPCR instrument (Light Cycler 2.0, Roche) [110]. A genetically engineered phage φA1122 expressing luciferase reporter genes was recently exploited for fluorescent labeling and indirect detection of Y. pestis [21]. The method was found to be very rapid (1 h) and allowed detection of 820 or more Y. pestis cells, while the fluorescent signals were also observed with two Y. pseudotuberculosis strains and one Yersinia enterocolitica strain. In the latest publication of the same group, a second generation reporter phage was described with a higher level of the bioluminescent signal response providing a 10-fold increase in the assay sensitivity and a potential to discriminate drug-susceptible and drug-resistant strains of Y. pestis due to the correlation of bioluminescent signals with antibiotic concentration [111].
A standardized gamma phage lysis test [45] was approved by the FDA for anthrax diagnostics in 2005 [112]. This phage lysed 96-100% of B. anthracis strains tested including capsulated cultures and 2-8% of strains that belong to relative species: B. cereus, Bacillus megaterium, Bacillus mycoides, and Bacillus subtilis [45,47,113]. Despite its limited specificity, gamma is globally used for the identification of B. anthracis [43,45,47,113-116]. Phages Fah [47,113] and AP50 [117] are strictly B. anthracis-specific but have more narrow lytic spectra: they lyse only 73-89.5% and 33% of strains, respectively. Phages K and VA-9 lyse all B. anthracis strains but have slightly lower specificity than gamma; the K and VA-9 phages are widely used for anthrax diagnosis in the FSU countries in addition to gamma and Fah [43,47].
The routine phage lysis test is a long procedure that requires isolation of a pure B. anthracis culture (sometimes including spore germination) and its overnight exposure to the phage that overall takes two to four days [114]. Several novel approaches were used to improve phagebased diagnostics. Indirect detection of B. anthracis Sterne, using qPCR targeting the gamma phage DNA took only 5 h. Its sensitivity for pure culture was 2×104 bacterial cells per milliliter, but the specificity of phage propagation and subsequent bacterial detection was not tested on other B. anthracis strains and representatives of other Bacillus species [118]. Using a luxAB-tagged reporter phage Wβ, the detection of B. anthracis was observed within 60 minutes and at low concentration, 103 CFU/ml [40,119]. The lysin of the gamma phage, PlyG, was exploited for rapid (1 h) and sensitive (100 bacteria) detection of B. anthracis based on the release of ATP from the lysed bacterial cells, which resulted in light emission in the presence of luciferase and luciferin [44]. C-terminal binding domain of PlyG was fused to glutathione S-transferase and used with horseradish peroxidase to bind to and detect ca. 103 CFU of both non-encapsulated and encapsulated B. anthracis within 3 h [120]. A short synthetic polypeptide representing the putative PlyG binding motif was coupled with Qdot-nanocrystals and used for the detection of single B. anthracis cells under fluorescent microscope [121].
There is a set of typing phages routinely and globally used for the identification and speciation of Brucella including Tb, Fz, Wb, S708, Bk, R/C, and Iz (Table 1). The character of lysis of Brucella cultures with the use of these phages at RTD and 104 RTD (see the notes to table 1) allows definitive identification of smooth and rough strains of B. abortus, B. suis, B. melitensis, B. neotomae, B. canis, and B. ovis [63,70,81,122] and even provides the differentiation of some biovars within B. suis [70,123]. Phage typing of Brucella is a multi-day procedure including isolation of a pure bacterial culture of slow-growing Brucella and the phage lysis tests. We recently developed a qPCR assay targeting S708 or Bk phage DNA for indirect identification of B. abortus (K.V. Sergueev, A.A. Filippov and M.P. Nikolich, manuscript in preparation). The assay allowed definitive identification of single cells of B. abortus during 72 hours in pure culture, blood and tissue samples.
There are two important fields in environmental applications of bacteriophages: detection of biothreat bacteria in environmental samples and phage-based decontamination. In contrast to PCR assays targeting bacterial DNA, phage-based techniques that rely on phage propagation detect only live bacterial cells. This is a definite advantage when testing the activity of natural foci of disease and for forensic purposes, such as during the investigation of bioterrorist attacks. For example, natural plague foci cover 6 to 7% of the dry surface of the Earth and require permanent epizootiological monitoring [124,125]. Employing the methods described in the previous section with the use of wild-type [22,110] and engineered [21,111] Y. pestis-specific phages is promising for determining the status of natural plague foci (e.g., when testing soil specimens from rodent burrows) and for the bacterial detection in environmental samples in case of intentional release of Y. pestis. The development of a field-deployable technology of indirect detection of Y. pestis [110] should facilitate the work in plague foci. Similarly, indirect detection methods with the use of intact [118] and genetically modified [40,119] phages or phage lysins [44,120,121] should help to identify B. anthracis in soil and other environmental samples in cases of natural anthrax or bioterrorist attacks. Since B. anthracis persists within soil both as vegetative cells and as spores [99,126,127], and spores can remain infective in soil as long as 70 years [128], the methods of spore detection are particularly important for environmental studies. A filamentous phage JRB7 was shown to specifically bind to B. anthracis spores [129] and was used in phagecoated magnetoelastic micro-biosensors for real-time detection of down to 103 spores per milliliter of water suspension [56,57]. Another spore-binding phage, 8a, was recently isolated and studied in terms of mechanisms of DNA injection [54]. Work is in progress on genome sequencing of the 8a phage and its genetic modification for developing an efficient method of B. anthracis spore detection [130]. Recent identification in the gamma phage lysin PlyG a domain for recognizing B. anthracis spores [131] can also help to improve phage-based spore detection. A successful method to identify live Brucella cells using phage-based qPCR (K.V. Sergueev, A.A. Filippov and M.P. Nikolich, manuscript in preparation) could be used for very sensitive and specific testing of agricultural and environmental samples in brucellosis foci.
Bacteriophages were used for specific disinfection of medical catheters [132,133], food products [134,135], and even corals in aquarium experiments [136]. In the cases of bioweapon or bioterrorist attacks, there is a problem of the “soft” disinfection of military, industrial and medical equipment, especially expensive sensitive equipment (like optical devices) that cannot withstand chemically aggressive disinfectants. The use of antibiotic solutions is not going to be efficient because there is a high probability of the application of natural or engineered multidrug-resistant strains. Recently, a phage cocktail including Pokrovskaya, Y, R, d’Herelle-m, and PST (see table 1) was employed for decontamination of hard surfaces (glass, gypsum board and stainless steel) spiked with Y. pestis, and the complete elimination of the bacteria was observed after a 5-min exposure to the cocktail at concentrations of 109, 108 and 107 PFU/ml [18]. A cocktail of unspecified phages isolated from topsoil retained viability at temperatures from –20°C to 37°C and when applied to B. anthracis spores, prevented their germination [137]. This approach could be expanded for the formulation of a phage cocktail active against multiple biothreat agents.
Given the emergence of drug-resistant strains of biothreat agents, the development of alternative antibacterials has become an important goal. This problem is particularly acute for plague, because at least three multidrug-resistant strains of Y. pestis have been isolated from patients and a wild rodent [6,7]. Easy transmissibility of the resistance genes from these isolates was demonstrated, and one unique isolate was shown to be resistant to all the chemotherapeutic agents recommended for plague treatment and prophylaxis [6]. The widespread dissemination of extensively drug-resistant bacteria has driven recent progress in bacteriophage therapy against experimental, veterinary and human bacterial infections [138,139]. The first therapeutic trial of bubonic plague with bacteriophage was conducted in 1925: four patients received injections of phage suspension into buboes and recovered in a few days [140]. Several more attempts at phage therapy against plague in the 1920’s-1930’s failed or provided limited therapy efficiency which could be explained by a lack of proper laboratory practices with the phages [141]. After the successful treatment of plague with a sulfonamide was first implemented in 1938 [142], studies on bacteriophages as potential therapies against plague were discontinued until recently.
We have addressed the challenges of plague phage therapy from several angles [16,17,22,23,143]. We tested 11 phages lytic for Y. pestis and demonstrated the high lytic activity of T7-like phages and L-413C (see table 1) towards the plague bacterium [17,22], the high specificity of Pokrovskaya, L-413C, φJA1, φA1122, and Y [17,22], and the high stability of purified phage suspensions [23] prepared not only in a phage storage buffer (SM) but also in phosphate-buffered saline supplemented with gelatin suitable for parenteral injection [144]. Phages Pokrovskaya, L-413C, φJA1, φA1122, Y, and d’Herelle-m were shown to have low EOP on E. coli at 37°C [17] and thus, in contrast to antibiotics, they should not affect the normal microflora in animals and humans. We also addressed a potential problem of phage-resistant mutants by: i) demonstrating an extremely low frequency of mutation to φA1122 resistance, ii) identifying seven bacterial cell surface receptors for nine different phages, and iii) finding that most of phage-resistant mutants are attenuated for virulence in mice [16]. These data suggested that: i) potential phage resistance in the case of φA1122 is not a significant problem; ii) Y. pestis mutants resistant to one specific phage should remain susceptible to the other phages, thus the use of these phages in a therapeutic cocktail is likely to prevent cross-resistance [145]; and iii) the majority of phage-resistant mutants are attenuated and will be removed by the immune system and thus should not pose a problem for the efficacy of therapy. Plague diagnostic phages were shown to be safe in cytotoxicity and acute toxicity tests [23]. Intraperitoneal administration of φA1122 provided longer persistence and one log higher concentrations of the phage in mouse spleen, liver and blood vs. intramuscular injections [23]. Pharmacodynamic and therapeutic trials revealed the propagation of φA1122 in mice infected with Y. pestis, a marked reduction of the bacteria in blood and organs, or total clearance of them, and a therapeutic effect of the phage. The most effective result was observed after a challenge with 103 LD50 of Y. pestis and subsequent (1 h later) single intraperitoneal injection of the phage at each of two doses (5×108 and 5×109 PFU): the survival of 20-40% mice and an 84% extension of mean time to death in nonsurvivors [23]. We consider this result very promising, because this is the first attempt of phage therapy against a fulminant plague infection in mice (LD50=2 bacteria). A single injection of the φA1122 phage provided an efficacy of therapy that can be compared with the results of multiple daily injections of a high dose of antibiotic over five days [146,147]. As for possible ways to improve the efficacy of plague therapy, we [143] suggest testing intravenous and intranasal routes of phage administration, as well as injections into or around the buboes [140], the selection of longer (10-14 days) circulating phage mutants [148], the application of different phages and phage cocktails, and the use of alternative animal models (rats and guinea pigs) providing higher lethal doses of Y. pestis and longer infectious processes.
Currently, the concern about drug resistance for B. anthracis and Brucella is not as eminent as in the case of plague. However, the potential introduction of multidrug resistance to current therapeutic options is a serious threat with either pathogen species which could provide dramatic problems in outbreak response. If this problem becomes more serious, there are lytic phages active against both B. anthracis and Brucella as potential alternative therapies. The use of phage lysins is a very promising area because this approach helps to circumvent the problem of phage resistance. The gamma phage PlyG lysin was successfully used in preliminary trials for intravenous therapy of experimental B. cereus and attenuated B. anthracis infection: 70-90% of treated vs. 0-10% of untreated mice survived [44,149]. Interestingly, brucellaphage were isolated from blood of the vast majority of brucellosis patients and a phage titer rise correlated with the improvement of patients’ condition [103]. Administration of phage to guinea pigs and rabbits with experimental brucellosis stimulated the immunity (resulted in higher agglutination titers) and increased the number of animals cleared from Brucella [150].
Based on the current body of literature and state of the field, one can conclude that bacteriophages lytic for Y. pestis, B. anthracis and Brucella species are routinely used for the detection, identification and typing of the host bacteria. New highly efficient indirect approaches with the use of genetically engineered or wild type phages have advanced diagnosis of the biothreat infections but require acceptance to be moved into practical use. Phage cocktails can be successfully used for the specific decontamination of chemically sensitive surfaces and expensive equipment. Highly lytic phages and their lysins are promising antibacterials for the treatment of drug-resistant plague, anthrax and brucellosis.
This work was supported in part by the Defense Threat Reduction Agency, Joint Science and Technology Office, Medical S&T Division. The findings and opinions expressed herein belong to the authors and do not necessarily reflect the official views of the WRAIR, the U.S. Army or the Department of Defense.
Make the best use of Scientific Research and information from our 700 + peer reviewed, 黑料网 Journals