Abstract

The protective effects of bacteriophages were assessed against experimental Staphylococcus aureus infection in mice. Of the S. aureus phages isolated in the study, ϕMR11 was representatively used for all testing, because its host range was the most broad and it carries no genes for known toxins or antibiotic resistance. Intraperitoneal injections (8×108 cells) of S. aureus including methicillin-resistant bacteria, caused bacteremia and eventual death in mice. In contrast, subsequent intraperitoneal administration of purified ϕMR11 (MOI ⩾0.1) suppressed S. aureus–induced lethality. This lifesaving effect coincided with the rapid appearance of ϕMR11 in the circulation, which remained at substantial levels until the bacteria were eradicated. Inoculation with high-dose ϕMR11 alone produced no adverse effects attributable to the phage. These results uphold the efficacy of phage therapy against pernicious S. aureus infections in humans and suggest that ϕMR11 may be a potential prototype for gene-modified, advanced therapeutic S. aureus phages

The worldwide spread of pathogenic bacteria that are resistant to a variety of antibiotics and the significant problems involved with their control threaten to reduce modern medicine to a state reminiscent of the preantibiotic era. Although novel antibiotics directed against such drug-resistant bacteria can be developed when extensive funds are committed for research, the pathogens will ultimately become resistant to the new drugs. To break this vicious cycle, it will be necessary to adopt chemotherapy-independent remedial strategies to combat bacterial infections

Bacteriophages (phages) are viruses that specifically infect and lyse bacteria. Phage therapy is a method of harnessing phages as bioagents for the treatment of bacterial infectious diseases. Phage therapy was originally introduced ∼80 years ago by Felix d’Herelle, a discoverer of phages [1, 2]. Despite all his efforts, this therapeutic initiative was later abandoned in Western countries, in part because of the subsequent highly successful discovery and mass production of many kinds of effective antibiotics in the 1940s. Furthermore, most early research into the therapeutic use of phages was poorly organized or uncontrolled, and our understanding of the basis for phage biology was immature. These factors combined to produce a negative outcome for phage therapy [2–4 ]. Nevertheless, phages have been used for practical purposes in the former Soviet Union and eastern Europe to the present day [2–6 ]. The ongoing active bacterial evolution of multidrug resistance also has recently motivated the Western scientific community to reevaluate the therapeutic potential of phages for diverse bacterial infections that are virtually incurable by conventional chemotherapy [2–4 ]. Phage therapy may be an alternative to antibiotics, because it has proved to be medically superior to antibiotic therapy in many ways [7–10 ]. In financial terms, the developmental costs of phage therapy are expected to be much less than those involved in the development of novel antibiotics

Growing evidence has suggested 4 different clinical applications for phages, as follows: (1) the administration of living bactericidal (virulent) phages or their natural mutants [11–18 ]; (2) the topical use of purified phage-encoded bacteriolytic peptidoglycan (cell wall) hydrolases (collectively referred to as endolysin or lysin), such as amidase and transglycosidase [19–22 ]; (3) the clinical use of phage structural proteins as a metabolic inhibitor to the key enzyme(s) of bacterial peptidoglycan synthesis [23]; and (4) the use of the phage-display system of the filamentous M13 phage, which expresses coat proteins fused to specific antibodies directed against bacterial antigens [24]. Besides these experimental achievements, clinical trials of phage therapy also may improve its prospects in the West [25]

Staphylococcus aureus is a pathogen of pyogenic inflammatory diseases, food poisoning, and toxic-shock syndrome; it is also a major causative agent for opportunistic and/or nosocomial infections, often with a high mortality rate [26]. According to one report, >50% of clinical S. aureus isolates in Japan today carry multidrug resistance, typically known as methicillin-resistant S. aureus (MRSA) [27]. Moreover, certain MRSA strains also have already acquired resistance to vancomycin (vancomycin-resistant S. aureus [VRSA]), a unique antibiotic previously considered to be effective against MRSA [28]

A study conducted in Poland has recently reported a summary of their 20-year experience of the clinical application of phages in humans, demonstrating their therapeutic efficacy, including S. aureus phages [5, 6]. Although these case-oriented reports are very attractive and valuable, additional basic information, such as details of the phages used and clinical data, should be presented, preferably in a publicized form, for the wider acceptance of phage therapy in the medical community. In the 1980s, Smith et al. [11–14 ] undertook rigorous investigations into phage therapy for pathogenic Escherichia coli infections in a veterinary context, thereby reopening this field of research in Western countries. Since Smith’s reevaluation, there has been, to the best of our knowledge, only 1 published report examining phage efficacy against experimental infections of S. aureus [15]. Using a mouse model, Soothill [15] examined the protective effects conferred by the S. aureus phage, ϕ-131, but, unfortunately, he failed to demonstrate the therapeutic efficacy of the phage (addressed again in Discussion). Therefore, the authenticity of phage therapy for S. aureus–induced diseases has not yet been confirmed, at least not on the animal experiment level

Our goal was to establish an effective and advanced phage-therapy system against life-threatening infections caused by antibiotic-resistant bacteria. We first focused on and scrutinized the therapeutic potency of phages directed against S. aureus infections, including MRSA and VRSA, in a mouse model. We isolated a number of wild S. aureus phages, extensively examined their bacteriolytic activity, and identified some with a broad host range. Using one of these, designated ϕMR11, our study provides the first experimental evidence that the administration of phages can successfully protect mice against fatal S. aureus infection without any adverse effects, thereby implying the clinical applicability of phage to human S. aureus–induced diseases

Materials and Methods

Culture mediaAll reagents and constituents of culture media used in our study were purchased from Nacalai Tesque, unless otherwise stated. Tryptic soy broth (TSB) and heart infusion broth (HIB) were obtained from Difco Laboratories. TSBM is TSB medium supplemented with 20 mM MgCl2, and HIBMC is HIB medium supplemented with 20 mM each MgCl2 and CaCl2. For phage-plaque formation and spot test, TSBM-based solid medium containing 1.5% or 0.5% agarose was used for the lower or upper layers, respectively

Measurements of bacterial growth and cell numberBacterial growth was monitored by measuring turbidity with a Klett-Summerson colorimeter (filter 54). One Klett unit was assumed to be equivalent to 6.4×106S. aureus cells/mL. This conversion formula was based on a previously standardized correlation between turbidity and bacterial cell numbers counted directly with a Petroff-Hausser counting chamber (Hausser Scientific)

Bacterial strainsThe bacterial strains used in our study included 43 methicillin-sensitive S. aureus (MSSA) strains and 29 MRSA strains, denoted by SA or MR preceding each strain number, respectively. Most MSSA strains were derived from nasal swab samples obtained from 162 healthy individuals, and all MRSA samples were derived from clinical specimens obtained from patients in Kochi Medical School Hospital. Nasal swab samples and clinical specimens were spread on mannitol salt agar supplemented with egg yolk (Nissui Pharmaceutical), which is a selective and semispecific growth medium for Staphylococcus. All the isolated strains were confirmed as S. aureus by use of the API-STAPH identification kit (API Systems), together with the coagulase test and light-microscopic observations. Identification of MRSA strains was achieved by colony formation on a selective salt agar plate containing 6 μg/mL oxacillin (MSO medium; Nissui). Strains that yielded colonies on the MSO plates were tested further for the mecA gene by use of a polymerase chain reaction–based detection kit (MecA test Wakunaga; Wakunaga Pharmaceutical). Strains incapable of growing on MSO were judged to be MSSA. A standard MSSA strain, 209P (obtained from the Research Institute for Microbial Diseases of Osaka University), and 2 VRSA strains, Mu3 and Mu50 [28] (kindly provided by Professor K. Hiramatsu, Juntendo University), also were used

In the present study, MSSA strain SA37 (mecA negative) was isolated from the nares of a healthy volunteer and served as the propagation host and/or experimental target of our particular S. aureus phage ϕMR11 (see below), unless otherwise stated, because SA37 is sensitive to most of the phages isolated, including ϕMR11. Two variants of SA37 were established for some animal experiments. For the examination of bacterial dynamics in vivo, a rifampicin-resistant mutant of SA37 (SA37-RIF2) was used to distinguish the injected strain from normal flora-derived strains. SA37-RIF2 was isolated by seeding the parental SA37 on selective TSB agar containing 2 μg/mL rifampicin. This mutant was confirmed as having the same growth properties and sensitivity to ϕMR11 phage as SA37. The other variant, the ϕMR11-lysogen of SA37 (SA37/ϕMR11), was used in a comparative study of the in vivo efficacy of ϕMR11. This variant was derived from a rare bacterial colony that had appeared among ϕMR11-exposed SA37 cells spread on a TSBM plate. Preliminary confirmation of the lysogeny of ϕMR11 in SA37/ϕMR11 was made by HindIII- or XbaI-digestion analysis, which showed that the cleaved DNA fragment pattern of the phage released from SA37/ϕMR11 by mitomycin C (MMC) induction (see below) was identical to that of the original ϕMR11. The lysogen had the same biological properties as parental SA37 but was insensitive to lysis by ϕMR11

Isolation of S. aureusbacteriophagesWe attempted to induce the replication of possible lysogenic phages by treating 72 S. aureus strains (43 MSSAs and 29 MRSAs) with MMC. In brief, bacterial cells were grown at 37°C until the late logarithmic phase (corresponding to 100 Klett units) in TSB medium. MMC then was added to the cultures to a final concentration of 1 μg/mL, followed by further incubation for 30 min. After the cells were washed once with TSB, they were resuspended in the same culture medium and were incubated again at 37°C for 4 h. The cultures then were centrifuged to remove cell debris filtered through a 0.45-μm pore-sized membrane, and the resultant supernatants were subjected to spot-test screening for bacteriolytic activity by use of the 72 S. aureus wild strains and the 209P, Mu3, and Mu50 strains as indicators. If the supernatants contained phages, this screening also facilitated an estimation of their bacteriolytic host range. About 5 μL of each preparation was placed directly onto lawns of the bacterial indicators with a disposable loop, and the production of lytic spots was assessed after incubation for 24 h at 37°C [29]

Lysis-positive supernatant samples sorted by the spot test then were subjected to single-plaque isolation using an appropriate host bacterium. The picked plaque was propagated in a larger volume of culture medium. The plaque-purified lytic agents then were examined by electron microscopy, regardless of whether they had the shape that is characteristic of phages. Of the identified phages, a phage designated ϕMR11 was representatively used for all subsequent experiments, because it was shown by the spot test to have the broadest host range (see Results)

Biological characterization of ϕMR11The adsorption rate, latent period, and burst size of ϕMR11 were determined essentially according to the method of Adams [30]. All incubations for the analyses were carried out in TSBM medium at 37°C. In brief, for examination of the adsorption rate, ϕMR11 (1.5×103 pfu) were mixed with SA37 cells (109/mL), and the number of free infectious phage virions was measured in the phage-cell mixture after treatment with chloroform [31]. To determine the latent period and burst size, SA37 cells (5.8×108 /mL) were exposed to ϕMR11 (2.4×103 pfu) for 5 min, washed thoroughly with cold TSBM medium to remove unbound phages, and then resuspended in fresh medium. An aliquot of the cell suspension was harvested regularly during incubation at 37°C to be titrated for newly produced phages, including both released and cell-associated phages, on a lawn of SA37

Large-scale purification of phage particlesPhage ϕMR11 was purified essentially according to the procedure used for several T4-type phages [29, 32–36 ]. SA37 host cells suspended at 1.2×108 cells/mL in 500 mL TSBM medium were exposed to a crude preparation of ϕMR11 at an MOI of 0.01 and were vigorously shaken for ∼4 h at 37°C, resulting in complete lysis of the bacteria. After 5 min of treatment with 1% chloroform at 37°C, the culture fluid was centrifuged at 8000 g for 10 min at 4°C to remove cell debris. Polyethylene glycol (average molecular weight, 6000) and NaCl were added to the supernatant to final concentrations of 10% and 0.5 M respectively, and kept at 4°C for 3 days. The resultant precipitate containing the phage particles was collected by centrifugation at 8000 g for 20 min at 4°C, resuspended in 10 mM Tris-HCl and 5 mM MgCl2 (pH 7.2) buffer, and treated with 20 μg/mL DNase I (Type II; Sigma Chemical) and 10 μg/mL RNase A (Type IA; Sigma) for 30 min at 37°C. The phage suspension then was placed on top of a discontinuous CsCl gradient (ρ=1.3, 1.5, and 1.7) and centrifuged at 100,000 g for 1 h at 4°C. The phage band was collected and dialyzed against saline that contained 20 mM MgCl2 and 20 mM CaCl2 (SMC) for 2 h at 4°C. CsCl-gradient separation and dialysis (1 h) were repeated, and further dialysis against HIBMC medium was carried out for 1 h at 4°C. The purified phage suspension was filtered (0.45-μm pore size), divided into aliquots, and stored at 4°C until used. The samples were appropriately diluted with HIBMC just before use for infections. The titers (pfu/mL) of the purified samples were determined by inoculating them into bacterial strain SA37. In experiments that involved the inoculation of mice with MRSA strains, ϕMR11 was propagated once with each MRSA before its administration, to minimize any possible bacterial restriction-modification, which is known to inhibit phage replication [37], and then purified as described. Plaque-forming units of host-adapted ϕMR11s were each titrated with the corresponding MRSA

Bioassay for toxins of S. aureusSome bacterial strains used in the study were checked for the production of toxins known to be associated with S. aureus, using reversed passive latex agglutination (RPLA) [38]. Enterotoxin and toxic-shock syndrome toxin–1 (TSST-1) of S. aureus were measured by use of the bacterial culture supernatants harvested after incubation at 37°C for 18–20 h, strictly according to the protocols for the assay kits (SET–RPLA Seiken and TST–PRLA Seiken; Denka Seiken)

Animal experimentsFor infection experiments, 6–8-week-old BALB/c female mice (body weight, ∼20 g) were used. S. aureus cells were grown in 100 mL TSB medium at 37°C and were centrifuged at 8000 g for 5 min at the early stationary phase (∼250 Klett units). The cell pellet was washed with 100 mL saline, centrifuged again under the same conditions, and finally resuspended in 5 mL saline. After appropriate dilution, turbidity (in Klett units) was measured to determine bacterial cell numbers, as described above. Varying numbers of bacterial cells suspended in 0.5 mL saline were injected into the peritoneal cavities of mice through one side of the abdomen, and purified phage suspensions in 1 mL HIBMC medium were injected through the other side. As controls, equal volumes of saline or HIBMC alone were injected intraperitoneally on all test occasions. The test animals were observed for between 1 week and 1 month

Approximately 0.5 mL of blood was taken by puncturing the orbital plexus of test mice with a capillary tube and mixed immediately with 50 μL heparin (1000 U/mL; Aventis Pharma). After the heparinized blood was diluted with SMC, colony-forming units of SA37-RIF2 were measured on rifampicin-TSB plates, and plaque-forming units of ϕMR11 were measured by use of SA37 as the host on TSBM plates. Blood samples were also collected from untreated mice, to ensure that mice used in the experiments were free of naturally or accidentally contaminating phage or bacteria

The data of survival rates of mice and dynamics of phage/host bacteria were analyzed statistically by use of Fisher’s exact test and a 2-way analysis of variance, respectively. The analyses were performed with SPSS 11.0J for Windows

Preparation of a mechanical lysate of S. aureusS. aureus SA37 was grown in 100 mL of TSB at 37°C to ∼250 Klett units, centrifuged at 12,000 g for 5 min at 4°C, washed with saline, and resuspended in saline at the concentration of 2×109 cells/mL. The cell pellet from 5 mL of the suspension was destroyed by grinding with a mortar and pestle at 4°C, and the volume was adjusted again to 5 mL with saline. Then the lysate was filtered through a membrane with a pore-size of 0.45 μm to remove residual living cells. The filtrate was used as a mechanical lysate of S. aureus

Electron microscopyPurified phage samples in SMC were fixed in 5% formalin and negatively stained with 2% uranyl acetate (pH 4.0). Electron micrographs were taken with a transmission electron microscope (Hitachi H-7100; Hitachi) at 75 kV

DNA sequencingGenome DNA of ϕMR11 was extracted, as described elsewhere [29]. HindIII-digested fragments of the genome were cloned into the pUC18 vector. Each cloned fragment was sequenced by PCR amplification (primer walking) by use of fluorescently labeled dideoxy chain-terminating nucleotides (ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit; Applied Biosystems), according to the specifications of the supplier. PCR products were purified by AutoSeq G-50 column (Amersham-Pharmacia Biotech) and were analyzed on an automated DNA sequencer (model 310; Applied Biosystems)

Results

Isolation of S. aureus–specific phages with therapeutic potentialCultures of 72 S. aureus strains (MSSA and MRSA) were treated with MMC, and the supernatants were harvested and then subjected to the spot test. The indicator panel for screening was composed of the same 72 S. aureus strains and strains 209P (MSSA), Mu3 (hetero-VRSA), and Mu50 (VRSA) [27]. Thirty-six of 72 supernatant samples clearly produced lytic spots on the lawn of at least 1 bacterial indicator, implying the possible induction of lysogenic phage replication by MMC. Of the 36 spot-positive supernatants, 8 formed lytic spots on lawns of most of the bacterial indicators and proved to contain phages at the time of testing, as ascertained by a subsequent plaque-formation assay and electron microscopy (see below). One of 8 phages identified as bacteriolytic for a broad host range, designated ϕMR11, formed spots on almost all S. aureus strains examined, indicating that this phage can adsorb efficiently to a wide range of S. aureus strains via the receptor molecule(s) ubiquitously expressed on the bacteria (see below). Moreover, ϕMR11 also produced large clear plaques on the greatest number of bacterial indicators (30/75 S. aureus strains tested). The discrepancy between the results for the spot test and the plaque-formation assay reflects the 2 different mechanisms underlying bacteriolysis by phage [20, 39]. Lytic spot formation represents a combination of both lysis from within associated with the phage replication (lytic) cycle and lysis from without caused by phage binding per se to bacteria [39]. Plaque formation, on the other hand, indicates the lysis from within mechanism only [39]. Therefore, a spot-positive but plaque-negative result implies the occurrence of lysis simply as the result of phage attachment to the bacterial cell surface, with subsequent inhibition of any multiplicative process(es) of the progeny virus. Abortion of the fully lytic cycle is an unfavorable event for the practical application of therapeutic phage. However, we expect that it may be largely overcome by modifying the phage genome—for example, by introducing a mutation at the repressor-binding sites critical to lytic cycle initiation. To this end, phages with relatively small genomes, such as ϕMR11 (see below), will be an advantage because their genomes are easy to manipulate by standard genetic engineering techniques

Morphological, biological, and genetic characterizations of ϕMR11As shown in figure 1, electron microscopy demonstrated that ϕMR11 has an isometrically hexagonal head, 56±2 nm in diameter, and a noncontractile tail of 175±5 nm in length, with a knoblike structure at its distal end. Therefore, it should be assigned to the family Siphoviridae morphotype B1 [40]. Restriction-enzyme analysis revealed that the ϕMR11 genome is circularly permuted double-stranded DNA, with an expected size ∼43 kbp. Sequencing the entire genome has been recently completed, and the exact genome size is 43,011 bp. ϕMR11 gene organization is unique (authors’ unpublished data) compared with 8 other S. aureus phages, the sequences of which have been registered in the nucleotide sequence databases. More important, the genomic sequence has verified that ϕMR11 carries neither antibiotic-resistance genes nor known toxin genes often associated with certain S. aureus phages, such as enterotoxin [41], leukocidin [42, 43], or exofoliative toxin [44, 45]. In this regard, the sequence data are consistent with the fact that the parental S. aureus strain, MR11, did not detectably produce enterotoxin (or TSST-1) on the RPLA assay (<1 ng/mL of bacterial culture fluid)

Figure 1

Electron micrograph of the Staphylococcus aureus–derived phage ϕMR11. Purified phage particles were negatively stained with 2% uranyl acetate. Bar 100 nm

Figure 1

Electron micrograph of the Staphylococcus aureus–derived phage ϕMR11. Purified phage particles were negatively stained with 2% uranyl acetate. Bar 100 nm

Biological studies clarified other features of ϕMR11 (figure 2), such as the following: (1) a rapid adsorption rate (when sensitive bacterial hosts were present in sufficient numbers, >90% of ϕMR11 particles bound to them within 1 min); (2) a short latent period (∼25 min); and (3) a relatively large burst size (>100). The efficient host-cell lysis characteristics of ϕMR11 are reminiscent of typical virulent phages, such as the T-even coliphages [40]. Of interest, ϕMR11 even lysed sensitive bacteria efficiently during the stationary phase, at which stage phage replication generally declines in concert with the cessation of host cell growth (data not shown)

Figure 2

Virological properties of the ϕMR11 phage. A Adsorption rate. ϕMR11 phages were mixed with excess Staphylococcus aureus SA37 cells, and then nonadsorbed infectious phages were serially counted. Data are percentage of nonadsorbed ϕMR11 relative to the initial input dose of phages. B Latent period and burst size. After a 5-min adsorption of ϕMR11 to a sufficient number of SA37 cells, the ϕMR11-exposed bacterial cells were thoroughly washed with tryptic soy broth medium supplemented with 20 mM MgCl2 to remove free phages and then were resuspended in fresh medium. Newly synthesized phage virions, including both released and cell-associated virions, then were measured at regular intervals during the incubation, using harvested culture samples. In both panels, each black circle represents an average of 2 samples. SE bars are too small to be visible behind the circles

Figure 2

Virological properties of the ϕMR11 phage. A Adsorption rate. ϕMR11 phages were mixed with excess Staphylococcus aureus SA37 cells, and then nonadsorbed infectious phages were serially counted. Data are percentage of nonadsorbed ϕMR11 relative to the initial input dose of phages. B Latent period and burst size. After a 5-min adsorption of ϕMR11 to a sufficient number of SA37 cells, the ϕMR11-exposed bacterial cells were thoroughly washed with tryptic soy broth medium supplemented with 20 mM MgCl2 to remove free phages and then were resuspended in fresh medium. Newly synthesized phage virions, including both released and cell-associated virions, then were measured at regular intervals during the incubation, using harvested culture samples. In both panels, each black circle represents an average of 2 samples. SE bars are too small to be visible behind the circles

On the evidence cited above, ϕMR11 was considered to be a suitable candidate for the therapeutic phage, or its prototype, for the treatment of human S. aureus infections. Therefore, the following experiments were undertaken using ϕMR11 in an animal model

Mouse model of S. aureus–induced diseaseThe dose of S. aureus lethal to mice was determined by injecting mice with varying numbers of strain SA37, ranging from 2×102 to 2×1010 cells per dose (figure 3). Intraperitoneal injections of 2×102 –108 SA37 did not reduce the survival rate of mice during the subsequent 7-day observation period. By contrast, injections of 3×108 –1×109 cells lowered the survival rate in a dose-dependent manner. Because the injection of 8×108 SA37 cells was fatal in >80% of mice within 24 h and in 100% within 7 days of injection, this level of challenge was considered to be optimal for observing the phage effect on bacterial lethality (figure 3). Therefore, the dose of S. aureus was fixed at 8×108 cells throughout the following experiments. A more precise time-chase analysis showed that intraperitoneal injection of 8×108 SA37 cells killed most mice between 6 and 7 h after injection, with associated preceding bacteremia (figure 4 and below). Dissection of mice that died from bacterial infection 6 h after injection revealed severe systemic congestion, with splenomegaly and acute ascites (data not shown)

Figure 3

Determination of the challenge dose of Staphylococcus aureus lethal to mice. Serially diluted suspensions of S. aureus SA37 cells were injected intraperitoneally into mice, and mouse fatalities were observed. White and black circles represent the survival rates 1 and 7 days after injection, respectively. Numerals associated with symbols indicate the nos. of mice tested. In the control experiment, denoted “0” on the horizontal axis, only 0.5 mL of saline was injected

Figure 3

Determination of the challenge dose of Staphylococcus aureus lethal to mice. Serially diluted suspensions of S. aureus SA37 cells were injected intraperitoneally into mice, and mouse fatalities were observed. White and black circles represent the survival rates 1 and 7 days after injection, respectively. Numerals associated with symbols indicate the nos. of mice tested. In the control experiment, denoted “0” on the horizontal axis, only 0.5 mL of saline was injected

Figure 4

A precise time-chase analysis of Staphylococcus aureus–induced killing of mice. Ten mice were each inoculated with a fixed no. of S. aureus SA37 cells (8×108 ), and their mortality was traced for up to 24 h. No. of surviving mice dropped suddenly 6 h after bacterial challenge, and death was preceded by obvious bacteremia (figure 7). These data are also relevant to the next experiment (see figure 6)

Figure 4

A precise time-chase analysis of Staphylococcus aureus–induced killing of mice. Ten mice were each inoculated with a fixed no. of S. aureus SA37 cells (8×108 ), and their mortality was traced for up to 24 h. No. of surviving mice dropped suddenly 6 h after bacterial challenge, and death was preceded by obvious bacteremia (figure 7). These data are also relevant to the next experiment (see figure 6)

Protective effects of phage administration against S. aureus infection in mice and associated safety issuesPurified ϕMR11 was injected into the mouse peritoneal cavity immediately after challenge with SA37, at various MOIs ranging from 0.01 to 200. Administration of ϕMR11 at an MOI of 0.1–200 significantly protected mice from SA37-induced death (figure 5). On the other hand, administration of a large amount (up to 2.0×1011 pfu) of ϕMR11 alone to 30 mice did not affect their physical condition or survival during the 1-month observation period. Consistent with the physical state of the mice, various tissues and organs of mice killed at 6 h and 1, 7, and 14 days after phage-only treatment did not differ in pathology from those of control mice untreated or injected with HIBMC medium only (data not shown). These findings indicate that ϕMR11 itself, at least when inoculated into the peritoneal cavity, does not give rise to any detectable adverse effects, despite its rapid entry into the circulation and its presumably systemic distribution in vivo (see below and figure 7)

Figure 5

Dose-dependent lifesaving effects of ϕMR11 against experimental Staphylococcus aureus infection in mice. After injection of S. aureus SA37 (8×108 ) cells, ϕMR11 were injected into the mouse peritoneal cavity at various MOIs, and the fate of the mouse was observed. Numerals adjacent to circles indicate the nos. of mice examined. Mice injected only with heart infusion broth supplemented with 20 mM MgCl2 and 20 mM CaCl2, which was used to prepare the phage suspensions, served as control mice and are represented by circles at an MOI of 0. There is a statistically significant difference in survival rates between mice treated with ϕMR11 at an MOI ⩾0.1 and untreated control mice (P<.012)

Figure 5

Dose-dependent lifesaving effects of ϕMR11 against experimental Staphylococcus aureus infection in mice. After injection of S. aureus SA37 (8×108 ) cells, ϕMR11 were injected into the mouse peritoneal cavity at various MOIs, and the fate of the mouse was observed. Numerals adjacent to circles indicate the nos. of mice examined. Mice injected only with heart infusion broth supplemented with 20 mM MgCl2 and 20 mM CaCl2, which was used to prepare the phage suspensions, served as control mice and are represented by circles at an MOI of 0. There is a statistically significant difference in survival rates between mice treated with ϕMR11 at an MOI ⩾0.1 and untreated control mice (P<.012)

Figure 7

Rapid appearance of phage ϕMR11 in circulation. Target bacterium (8×108 cells) and/or ϕMR11 (MOI, 200) were injected intraperitoneally into 4 groups of mice, each of which consisted of 5 mice. A rifampicin-resistant mutant of SA37 (SA37-RIF2; see text) was used as the target in this experiment. A peripheral blood sample was taken from 1 mouse from each group 2, 4, 6, and 24 h after injection and was titrated to estimate the nos. of phages and bacteria in the circulation. Black and white circles represent plaque-forming units of ϕMR11 in the presence and absence of SA37-RIF2, respectively. Black and white squares represent colony-forming units of the host bacteria with and without the administration of phage, respectively. All mice injected with only SA37-RIF2 died during the observation period within 6 h of injection (†). Each circle and square represent an average of 2 samples. SE bars are too small to be visible behind the symbols. The 2-way analysis of variance revealed statistically significant differences between the in vivo dynamics of ϕMR11 in the presence and in the absence of SA37-RIF2 (P<.001) and between the dynamics of SA37-RIF2 in the presence and absence of ϕMR11 (P<.001)

Figure 7

Rapid appearance of phage ϕMR11 in circulation. Target bacterium (8×108 cells) and/or ϕMR11 (MOI, 200) were injected intraperitoneally into 4 groups of mice, each of which consisted of 5 mice. A rifampicin-resistant mutant of SA37 (SA37-RIF2; see text) was used as the target in this experiment. A peripheral blood sample was taken from 1 mouse from each group 2, 4, 6, and 24 h after injection and was titrated to estimate the nos. of phages and bacteria in the circulation. Black and white circles represent plaque-forming units of ϕMR11 in the presence and absence of SA37-RIF2, respectively. Black and white squares represent colony-forming units of the host bacteria with and without the administration of phage, respectively. All mice injected with only SA37-RIF2 died during the observation period within 6 h of injection (†). Each circle and square represent an average of 2 samples. SE bars are too small to be visible behind the symbols. The 2-way analysis of variance revealed statistically significant differences between the in vivo dynamics of ϕMR11 in the presence and in the absence of SA37-RIF2 (P<.001) and between the dynamics of SA37-RIF2 in the presence and absence of ϕMR11 (P<.001)

Possible mechanisms behind the lifesaving effects of ϕMR11A previously published in vitro analysis using human materials showed that a crude lysate of S. aureus infected by phage provoked an immediate cell-mediated immune response [46]. Therefore, we investigated whether the direct bactericidal activity of ϕMR11 was actually responsible for its protective effects. Intraperitoneal injections of mechanical lysate, an analogue of phage lysate, prepared from 1×109 SA37 cells, was followed by the injection of an equivalent number of living SA37 cells. The mechanical lysate did not prevent death in the mice. Mice injected with mechanical lysate prepared only from twice the number of SA37 (2.0×109 ) were not affected, indicating that the mouse-killing effect was not caused simply by the excessive bacterial burden imposed in this experiment. Therefore, it is unlikely that bacterial components, even those released by phage lysis, were involved in the protective effect. To study another critical possibility, that ϕMR11 virions themselves stimulate an antibacterial immune response, a further experiment was conducted using the ϕMR11-lysogen of SA37 (SA37/ϕMR11) as the target bacterium. This SA37 derivative has biological properties and a genetic background identical to the parental SA37 strain, except for its nonsusceptibility to ϕMR11-induced lysis. If the net lifesaving effect is mediated, not through ϕMR11-induced bacteriolysis but via a virion-stimulated immune response (e.g., production of cytokines), the administration of phage should be effective against SA37/ϕMR11, as well as against SA37 infections. Injection of purified ϕMR11 (MOI, 10) rescued mice challenged with SA37 from death but not mice challenged with SA37/ϕMR11. These results support the conclusion that direct bactericidal activity exerted by ϕMR11 is the prime and perhaps the sole determinant of the protective effect observed in the present study

Temporal sequence and in vivo distribution of ϕMR11 and S. aureusPurified ϕMR11 was administered intraperitoneally to mice at different times (up to 60 min) after SA37 injection (figure 6). Although ϕMR11 administration proved to be effective at MOIs ranging widely from 1 to 200 (figure 5), the highest MOI of 200 was applied in this and subsequent investigations (see below), under the assumption that patients were in an extreme situation of systemic S. aureus infection such as sepsis. After 24 h, most mice treated with phages at any time point were still alive, whereas 80%–100% of mice not treated with phages were dead. The therapeutic efficacy of ϕMR11 was even discernible in mice treated 60 min after injection with bacteria, when all the control mice injected with SA37-only already exhibited signs of physical deterioration, such as reduced activity and ruffled hair. Although a few ϕMR11-treated mice died during the following 6 days, the survival rates among mice treated with phages at any time point were always significantly higher than those of the untreated controls (figure 6)

Figure 6

Protective effects with delayed administration of ϕMR11. Purified ϕMR11 (MOI, 200) were administered to 5 mice at the various time intervals indicated, after challenge with Staphylococcus aureus SA37 (8×108 cells). As a control, 1 mL of phage-free heart infusion broth supplemented with 20 mM each MgCl2 and CaCl2 was injected into mice. Survival rates were determined after 1 (A) and 7 (B) days. Shaded and hatched columns represent the phage-treated and -untreated mouse groups, respectively. Asterisks signify statistically significant differences compared with that of the controls: *P<.05, **P<.01, and ***P<.002

Figure 6

Protective effects with delayed administration of ϕMR11. Purified ϕMR11 (MOI, 200) were administered to 5 mice at the various time intervals indicated, after challenge with Staphylococcus aureus SA37 (8×108 cells). As a control, 1 mL of phage-free heart infusion broth supplemented with 20 mM each MgCl2 and CaCl2 was injected into mice. Survival rates were determined after 1 (A) and 7 (B) days. Shaded and hatched columns represent the phage-treated and -untreated mouse groups, respectively. Asterisks signify statistically significant differences compared with that of the controls: *P<.05, **P<.01, and ***P<.002

On the basis of the above results, the in vivo dynamics of the bacteria and phages were investigated in detail using SA37-RIF2 as the target cell. Mice were injected intraperitoneally with 8×108 SA37-RIF2 alone, with SA37-RIF2 plus ϕMR11 (1.6×1011 pfu), or with ϕMR11 alone. Bacteremia occurred within 2 h and then persisted in the SA37-RIF2–injected mice, regardless of phage treatment (figure 7). However, the bacterial loads in the blood were significantly lower in ϕMR11-treated mice than in untreated mice throughout our observations (figure 7). Compatible with the result shown in figure 4, all mice injected only with SA37-RIF2 died within 6 h, whereas ϕMR11-treated mice were invariably saved, concomitantly with a subsiding of septicemia within 24 h. However, after ϕMR11 was injected into the peritoneal cavity, a significant number of infectious ϕMR11 was readily detected 2 h later in blood specimens from both the SA37-RIF2–infected and uninfected mice at titers of 1.3 and 7.7×108 pfu/mL, respectively. This initial discrepancy between the titers of circulating phages probably represents an acute “consumption” of administered ϕMR11 by the inoculated target bacteria. It is noteworthy, however, that subsequent relative levels of circulating ϕMR11 were reversed in the 2 mouse groups: the titer rapidly decreased with time in mice not inoculated with SA37-RIF2, whereas it remained higher in SA37-RIF2–injected mice. Typical titers measured after 6 h were 4.7×105 pfu/mL in the absence of SA37-RIF2 versus 5.5×108 pfu/mL in the presence of SA37-RIF2 (figure 7). These results illustrate the in vivo kinetics, whereby intraperitoneal ϕMR11 was amplified by the coexisting target bacteria and migrated into the bloodstream, perhaps resulting in the systemic dissemination of the phages. Under these circumstances, the circulating ϕMR11 was sustained at a significant level until the target cells were eradicated, which must have counteracted the progression of bacteremia. The beneficial in vivo kinetics of ϕMR11 in curing the S. aureus infection also may be applicable in general to phages given therapeutically to counteract other bacterial infections [47]

Efficacy of ϕMR11 for MRSA infectionFinally, we investigated whether ϕMR11 is also efficacious against MRSA infections. One of 4 clinical MRSA strains—MR1, MR13, MR18, or MR28—was injected intraperitoneally (8×108 cells) into mice. For this experiment, ϕMR11 was propagated once beforehand with each corresponding MRSA strain to eradicate as much as possible any bacterial restriction-modification [36]. Phages then were purified through a CsCl gradient. As shown in figure 8, all mice treated with these “adapted” ϕMR11s (MOI, 50) survived for 7 days after MRSA injection, in contrast to the >90% mortality in the phage-untreated mice, which suggests that ϕMR11 also is potentially useful for human MRSA infections

Figure 8

Therapeutic efficacy of ϕMR11 for methicillin-resistant Staphylococcus aureus (MRSA) infections in mice. Four clinically isolated MRSA strains, MR1, MR13, MR18, and MR28, were used as host bacteria. ϕMR11 was propagated once with each MRSA, to minimize possible phage inactivation by the bacterial restriction-modification system, followed by CsCl purification. Each “adapted” ϕMR11 was administered (MOI, 50) intraperitoneally to mice immediately after challenge with the corresponding MRSA strain (8×108 cells). Survival rates were determined 7 days after injection of the bacterium and/or phage. Shaded and hatched columns represent the phage-treated and -untreated mouse groups, respectively. Each group was composed of 10 mice. A statistically significant difference (P<.0002) in survival rates was observed between paired groups of phage-treated and -untreated mice for all MRSAs, as indicated by horizontal brackets with asterisks

Figure 8

Therapeutic efficacy of ϕMR11 for methicillin-resistant Staphylococcus aureus (MRSA) infections in mice. Four clinically isolated MRSA strains, MR1, MR13, MR18, and MR28, were used as host bacteria. ϕMR11 was propagated once with each MRSA, to minimize possible phage inactivation by the bacterial restriction-modification system, followed by CsCl purification. Each “adapted” ϕMR11 was administered (MOI, 50) intraperitoneally to mice immediately after challenge with the corresponding MRSA strain (8×108 cells). Survival rates were determined 7 days after injection of the bacterium and/or phage. Shaded and hatched columns represent the phage-treated and -untreated mouse groups, respectively. Each group was composed of 10 mice. A statistically significant difference (P<.0002) in survival rates was observed between paired groups of phage-treated and -untreated mice for all MRSAs, as indicated by horizontal brackets with asterisks

Discussion

The emergence of multidrug-resistant bacteria will have a serious impact on various aspects of medical practice. There are several potential chemotherapy-independent schemes with which to address this clinical problem, one being the prevention of colonization by drug-resistant bacteria by using microbial interference [48, 49]. In the present study, using a newly isolated phage, we experimentally reevaluated the therapeutic potential of phages in mice infected with S. aureus including MRSA

A series of rigorous studies into phage therapy by Smith et al. [11–14 ] in the 1980s made a significant contribution that led to reevaluation of phage efficacy against infections of E. coli [16, 17], Acinetobacter baumanii [15], Pseudomonas aeruginosa [15], Klebsiella species [50, 51], Lactococcus garvieae [47], and Enterococcus faecium [18] in animal models or in natural animal targets of these virulent microbes. These, together with the present study, support the potential of phage therapy against various bacterial infectious diseases; in fact, successful treatment for humans has been reportedly achieved in eastern Europe and the former Soviet Union [2–6 ]. These reports are certainly noteworthy in the clinical context; however, basic features of the phages used for individual bacterial pathogens, including the pharmacokinetics involved, are still largely unclear. In addition, as described in the literature [5, 6], the phages were administered in a “made-to-order” manner, which may provide some clinical merit, but as crude preparations. For verification of the scientific as well as clinical validity of the therapy, well-characterized and purified phages will be desirable for therapeutic use

The present study has shown that the selected phage ϕMR11 is effective in treating S. aureus infections in an animal model. There has been only one report published in 1992 on experimental phage therapy for S. aureus infection, by Soothill [15]. In that study, phage ϕ-131 (unpurified), which is commonly used in Poland, was administered to mice artificially infected with S. aureus strains 6409 or M60. However, this experiment failed to reproduce the efficacious effects reported in humans [5, 6], although therapeutic efficacy was successfully shown for phages specific to P. aeruginosa and A. baumanii. This negative result using S. aureus may be explained, in part, by the low dose of ϕ-131 administered: an MOI of 0.018 for strain 6409 and 0.086 for strain M60 [15]. Furthermore, the MOIs calculated in that study may have been overestimates because S. aureus tends to form cell clusters. Our data (figure 5), which are consistent in this context with Soothill’s findings [15], indicate that an MOI of 1 is the minimum required for ϕMR11 to yield a fully protective effect, at least in the mouse model. Taken together, these data imply that determination of the appropriate dose of phage is a prerequisite for successful phage therapy

Some phage lysates of S. aureus have been reported to be capable of stimulating a cell-mediated immune response [46]. This raises the question of whether the protective effect induced by ϕMR11 is attributable to direct bacteriolysis by the phage or associated with vaccinelike immune activation by dispersed bacterial components. However, such a bacterial antigen-induced immune reaction can be discounted, at least as a principal mechanism, on the basis of our own results, which show that a “mechanical” lysate of S. aureus induced no antibacterial effect. The irrelevancy of immune mediation in ϕMR11 treatment also is supported by the fact that bacterial antigens were completely removed from the ϕMR11 preparations during repeated CsCl-gradient centrifugation purification. Furthermore, any activation by phage particles themselves of an antibacterial immune response also was negligible, because ϕMR11 had no therapeutic effect against infections with the phage-lysogenic S. aureus substrain SA37-ϕMR11

Our study also demonstrated that ϕMR11 can rapidly enter into the circulation (within 2 h), even when administered intraperitoneally, which implies rapid systemic distribution of the phage in vivo. In fact, infectious ϕMR11 were subsequently also detected in various tissues and organs such as the spleen, liver, kidney, brain, and skeletal muscle (105–106 pfu/g; authors’ unpublished data), which suggests an extension of the medical application of phage to systemic infections. Merril et al. [16] observed similar in vivo translocation of E. coli λ phage and Salmonella phage P22 administered in a mouse system. As an unfavorable consequence, however, these phages were rapidly eliminated from the blood, making the capture of circulating phages by the splenic reticuloendothelial system an anticipated problem of phage therapy [7, 16, 52]. Of interest, in the same article, Merril et al. also reported that a small population of mutant λ phages survived in the circulation, with a concomitant alteration to major head protein E. Therefore, the authors postulated that such “serially passaged” immune-escape mutants may facilitate improvements to the therapeutic efficacy of phages [16]. Our data, on the other hand, indicate that infectious ϕMR11 was maintained at high levels in the bloodstream in the presence of target bacteria, at up to 3 orders of magnitude above levels measured in the absence of target cells. ϕMR11 titers in the circulation were substantial, ∼104 pfu/mL, even 24 h after injection (figure 7). As pointed out by Barrow and Soothill [9], our data and those of others [46] support the view that the immunologic elimination of certain phages in vivo may not be a serious issue for practical applications

To establish an efficient phage-therapy system generally applicable against human S. aureus–induced diseases, several other obstacles must be overcome in the future. The occurrence of phage resistance is unquestionably a problem in phage therapy [53]. However, according to data described elsewhere [54] and our own data (unpublished), the incidence of bacteria that resist phage attack is ∼10-fold lower than the incidence of bacteria insensitive to an antibiotic. Even if bacteria acquire phage resistance, new mutant phages that act lytically against these bacteria can be isolated quickly (usually in a few days). It will be necessary to create advanced therapeutic phages to circumvent other inevitable problems, at least at present, such as the lysogenicity of therapeutic phages and the restriction-modification systems of bacteria. We believe that it is possible to render prototype phages (i.e., with broad host ranges) nonlysogenic by artificial gene alteration, such as via the introduction of mutations into and/or deletion of genomic elements essential for phage integration into the host cell DNA. ϕMR11 will be an eligible candidate for a prototype with which to pursue these objectives, because its genome size is relatively small and allows for easy genetic manipulation. Of greater importance, ϕMR11 uses a binding receptor(s) commonly expressed on S. aureus as indicated by its highly positive reactivity in the spot test. It also lacks known toxin and drug-resistance genes. Finally, to minimize phage inactivation by the bacterial restriction-modification system [34], we are now constructing an alternative host bacterium for the production of ϕMR11 with methylated CG bases throughout the phage genome

The data presented here, derived from using an animal model for S. aureus infections, support the therapeutic use of phages in humans. Phages are conceivably present in all kinds of bacteria. To search for and characterize further phages with therapeutic potential may cast new light on treating diverse bacterial infections that are uncontrollable by currently available antibiotics. Moreover, further scientific proof of the in vivo efficacy and safety of phage therapy will validate its clinical use in humans, thereby possibly greatly reducing antibiotic use, which is the major culprit in the recent increase in multidrug-resistant bacteria. Advanced therapeutic phages will provide, as d’Herelle hoped, a promising remedial resource to overcome the “modern plagues” on the condition that an appropriate phage strain and a relevant method for its administration are selected [15, 16, 55–57 ]. This revitalized therapy also may become a powerful weapon against bioterrorism based on pathogenic bacteria with intentionally introduced antibiotic resistance [22]

Acknowledgments

We thank K. Ito for help with manuscript preparation. We are deeply indebted to K. Hiramatsu of Juntendo University for the kind gift of bacterial strains Mu50 and Mu3 and to the students of the Kochi Medical School for their generous assistance in collecting nasal swab samples

References

1
Summers
WC
Félix d’Herelle and the origins of molecular biology
1999
New Haven, CT
Yale Univ. Press
2
Ho
K
Bacteriophage therapy for bacterial infections
Perspect Biol Med
 , 
2001
, vol. 
44
 (pg. 
1
-
16
)
3
Alisky
J
Iczkowski
K
Rapoport
A
Troitsky
N
Bacteriophage show promise as antimicrobial agents
J Infect
 , 
1998
, vol. 
36
 (pg. 
5
-
15
)
4
Sulakvelidze
A
Alavidze
Z
Morris
JG
Jr
Bacteriophage therapy
Antimicrob Agents Chemother
 , 
2001
, vol. 
45
 (pg. 
649
-
59
)
5
Slopek
S
Weber-Dabrowska
B
Dabrowski
M
Kucharewicz-Krukowska
A
Results of bacteriophage treatment of suppurative bacterial infections in the years 1981–1986
Arch Immunol Ther Exp
 , 
1987
, vol. 
35
 (pg. 
569
-
83
)
6
Weber-Dabrowska
B
Mulczyk
M
Górski
A
Bacteriophage therapy of bacterial infections: an update of our institute’s experience
Arch Immunol Ther Exp
 , 
2000
, vol. 
48
 (pg. 
547
-
51
)
7
Phage therapy [editorial]
Lancet
 , 
1983
, vol. 
2
 (pg. 
1287
-
8
)
8
Lederberg
J
Smaller fleas…ad infinitum: therapeutic bacteriophage redux
Proc Natl Acad Sci USA
 , 
1996
, vol. 
93
 (pg. 
3167
-
8
)
9
Barrow
PA
Soothill
JM
Bacteriophage therapy and prophylaxis: rediscover and renewed assessment of potential
Trends Microbiol
 , 
1997
, vol. 
5
 (pg. 
268
-
71
)
10
Pirisi
A
Phage therapy: advantages over antibiotics
Lancet
 , 
2000
, vol. 
356
 pg. 
1418
 
11
Smith
HW
Huggins
MB
Successful treatment of experimental Escherichia coli infections in mice using phage: its general superiority over antibiotics
J Gen Microbiol
 , 
1982
, vol. 
128
 (pg. 
307
-
18
)
12
Smith
HW
Huggins
MB
Effectiveness of phages in treating experimental Escherichia coli diarrhoea in calves, piglets, and lambs
J Gen Microbiol
 , 
1983
, vol. 
129
 (pg. 
2659
-
75
)
13
Smith
HW
Huggins
MB
Shaw
KM
Factors influencing the survival and multiplication of bacteriophages in calves and in their environment
J Gen Microbiol
 , 
1987
, vol. 
133
 (pg. 
1127
-
35
)
14
Smith
HW
Huggins
MB
Shaw
KM
The control of experimental Escherichia coli diarrhoea in calves by means of bacteriophages
J Gen Microbiol
 , 
1987
, vol. 
133
 (pg. 
1111
-
26
)
15
Soothill
JS
Treatment of experimental infections of mice with bacteriophages
J Med Microbiol
 , 
1992
, vol. 
37
 (pg. 
258
-
61
)
16
Merril
CR
Biswas
B
Carlton
R
, et al.  . 
Long-circulating bacteriophage as antibacterial agents
Proc Natl Acad Sci USA
 , 
1996
, vol. 
93
 (pg. 
3188
-
92
)
17
Barrow
P
Lovell
M
Berchieri
A
Jr
Use of lytic bacteriophage for control of experimental Escherichia coli septicemia and meningitis in chickens and calves
Clin Diagn Lab Immunol
 , 
1998
, vol. 
5
 (pg. 
294
-
8
)
18
Biswas
B
Adhya
S
Washart
B
, et al.  . 
Bacteriophage therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant Enterococcus faecium
Infect Immun
 , 
2002
, vol. 
70
 (pg. 
204
-
10
)
19
Das
P
Bacteriophage therapy offers new hope for streptococcal infections
Lancet
 , 
2001
, vol. 
357
 pg. 
938
 
20
Nelson
D
Loomis
L
Fischetti
VA
Prevention and elimination of upper respiratory colonization of mice by group A streptococci by using a bacteriophage lytic enzyme
Proc Natl Acad Sci USA
 , 
2001
, vol. 
98
 (pg. 
4107
-
12
)
21
Loeffler
JM
Nelson
D
Fischetti
VA
Rapid killing of Streptococcus pneumoniae with a bacteriophage cell wall hydrolase
Science
 , 
2001
, vol. 
294
 (pg. 
2170
-
2
)
22
Schuch
R
Nelson
D
Fischetti
VA
A bacteriolytic agent that detects and kills Bacillus anthracis
Nature
 , 
2002
, vol. 
418
 (pg. 
884
-
9
)
23
Bernhardt
TG
Wang I-N
Struck
DK
Young
R
A protein antibiotic in the phage Qβ virion: diversity in lysis target
Science
 , 
2001
, vol. 
292
 (pg. 
2326
-
9
)
24
Cao
J
Sun
Y
Berglindh
T
, et al.  . 
Helicobacter pylori–antigen-binding fragments expressed on the filamentous M13 phage prevent bacterial growth
Biochim Biophys Acta
 , 
2000
, vol. 
1474
 (pg. 
107
-
13
)
25
Sulakvelidze
A
Morris
JG
Brown
T
Development and commercialization of phage therapy in the United States [abstract]
Program and abstracts of the 10th Evergreen International Phage Biology Meeting
 , 
2000
Olympia, WA
Evergreen State College
26
Noble
WC
Collier
L
Staphylococcal diseases
Microbiology and microbial infections
 , 
1998
, vol. 
Vol 3
 
New York
Oxford University Press
(pg. 
231
-
56
)
27
Hashimoto
H
Drug resistance of methicillin-resistant Staphylococcus aureus (MRSA) in Japan until 1993
Jpn J Antibiot
 , 
1994
, vol. 
47
 (pg. 
575
-
84
)
28
Hiramatsu
K
Aritaka
N
Hanaki
H
, et al.  . 
Dissemination in Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin
Lancet
 , 
1997
, vol. 
350
 (pg. 
1670
-
3
)
29
Matsuzaki
S
Tanaka
S
Koga
T
Kawata
T
A broad–host-range vibriophage, KVP40, isolated from sea water
Microbiol Immunol
 , 
1992
, vol. 
36
 (pg. 
93
-
7
)
30
Adams
KH
Bacteriophages
1959
New York
Wiley-Interscience
31
Inoue
T
Matsuzaki
S
Tanaka
S
A 26-kDa outer membrane protein, OmpK, common to Vibrio species is the receptor for a broad-host-range vibriophage, KVP40
FEMS Microbiol Lett
 , 
1995
, vol. 
125
 (pg. 
101
-
6
)
32
Matsuzaki
S
Inoue
T
Tanaka
S
A vibriophage, KVP40, with major capsid protein homologous to gp23* of coliphage T4
Virology
 , 
1998
, vol. 
242
 (pg. 
314
-
8
)
33
Matsuzaki
S
Inoue
T
Kuroda
M
Kimura
S
Tanaka
S
Cloning and sequencing of major capsid protein (mcp) gene of a vibriophage, KVP20, possibly related to T-even coliphages
Gene
 , 
1998
, vol. 
222
 (pg. 
25
-
30
)
34
Matsuzaki
S
Kuroda
M
Kimura
S
Tanaka
S
Major capsid proteins of certain Vibrio and Aeromonas phages are homologous to the equivalent protein, gp23*, of coliphage T4
Arch Virol
 , 
1999
, vol. 
144
 (pg. 
1647
-
51
)
35
Matsuzaki
S
Kuroda
M
Kimura
S
Tanaka
S
Vibriophage KVP40 and coliphage T4 genomes share a homologous 7-kbp region immediately upstream of the gene encoding the major capsid protein
Arch Virol
 , 
1999
, vol. 
144
 (pg. 
2007
-
12
)
36
Matsuzaki
S
Inoue
T
Tanaka
S
, et al.  . 
Characterization of a novel Vibrio parahaemolyticus phage, KVP241, and its relatives frequently isolated from seawater
Microbiol Immunol
 , 
2000
, vol. 
44
 (pg. 
953
-
6
)
37
Matsuzaki
S
Inoue
T
Tanaka
S
Evidence for the existence of a restriction-modification system common to several species of the family Vibrionaceae
FEMS Microbiol Lett
 , 
1992
, vol. 
73
 (pg. 
191
-
4
)
38
Fujikawa
H
Igarashi
H
Rapid latex agglutination test for detection of staphylococcal enterotoxins A to E that uses high-density latex particles
Appl Environ Microbiol
 , 
1988
, vol. 
54
 (pg. 
2345
-
8
)
39
Young
R
Bacteriophage lysis: mechanism and regulation
Microbiol Rev
 , 
1992
, vol. 
56
 (pg. 
430
-
81
)
40
Ackermann H-W
DuBow
MS
Viruses of prokaryotes. I. General properties of bacteriophages
1987
Boca Raton
FL CRC Press
41
Betley
MJ
Mekalanos
JJ
Staphylococcal enterotoxin A is encoded by phage
Science
 , 
1985
, vol. 
229
 (pg. 
185
-
7
)
42
Kaneko
J
Kimura
T
Kawakami
Y
Tomita
T
Kamio
Y
Panton-Valentine leukocidin genes in a phage-like particle isolated from mitomycin C–treated Staphylococcus aureus V8 (ATCC 49775)
Biosci Biotechnol Biochem
 , 
1997
, vol. 
61
 (pg. 
1960
-
2
)
43
Kaneko
J
Kimura
T
Narita
S
Tomita
T
Kamio
Y
Complete nucleotide sequence and molecular characterization of the temperate staphylococcal bacteriophage ΦPVL carrying Panton-Valentine leukocidin genes
Gene
 , 
1998
, vol. 
215
 (pg. 
57
-
67
)
44
Yamaguchi
T
Hayashi
T
Takami
H
, et al.  . 
Phage conversion of exfoliative toxin A production in Staphylococcus aureus
Mol Microbiol
 , 
2000
, vol. 
38
 (pg. 
694
-
705
)
45
Yoshizawa
Y
Sakurada
J
Sakurai
S
Machida
K
An exfoliative toxin A–converting phage isolated from Staphylococcus aureus strain ZM
Microbiol Immunol
 , 
2000
, vol. 
44
 (pg. 
189
-
91
)
46
Dean J
H
Silva
JS
McCoy
JL
, et al.  . 
In vitro human reactivity to staphylococcal phage lysate
J Immunol
 , 
1975
, vol. 
115
 (pg. 
1060
-
4
)
47
Nakai
T
Sugimoto
R
Park K-H
, et al.  . 
Protective effects of bacteriophage on experimental Lactococcus garvieae infection in yellowtail
Dis Aquat Org
 , 
1999
, vol. 
37
 (pg. 
33
-
41
)
48
Uehara
Y
Nakama
H
Agematsu
K
, et al.  . 
bacterial interference among nasal inhabitants: eradication of Staphylococcus aureus from nasal cavities by artificial implantation of Corynebacterium sp
J Hosp Infect
 , 
2000
, vol. 
44
 (pg. 
127
-
33
)
49
Uehara
Y
Kikuchi
K
Nakamura
T
, et al.  . 
Inhibition of methicillin-resistant Staphylococcus aureus colonization of oral cavities in newborns by viridans group streptococci
Clin Infect Dis
 , 
2001
, vol. 
32
 (pg. 
1399
-
407
)
50
Bogovazova
GG
Voroshilova
NN
Bondarenko
VM
The efficacy of Klebsiella pneumoniae bacteriophage in the therapy of experimental Klebsiella infection
Zh Mikrobiol Epidemiol Immunobiol
 , 
1991
, vol. 
4
 (pg. 
5
-
8
)
51
Bogovazova
GG
Voroshilova
NN
Bondarenko
VM
, et al.  . 
Immunobiological properties and therapeutic effectiveness of preparations from Klebsiella bacteriophages
Zh Mikrobiol Epidemiol Immunobiol
 , 
1992
, vol. 
3
 (pg. 
30
-
3
)
52
Geier
MR
Trigg
ME
Merril
CR
Fate of bacteriophage lambda in non-immune germ-free mice
Nature
 , 
1973
, vol. 
246
 (pg. 
221
-
3
)
53
Lowbury
EJL
Hood
AM
The acquired resistance of Staphylococcus aureus to bacteriophage
J Gen Microbiol
 , 
1953
, vol. 
9
 (pg. 
524
-
35
)
54
Carlton
RM
Phage therapy: past history and future prospects
Arch Immunol Ther Exp
 , 
1999
, vol. 
47
 (pg. 
267
-
74
)
55
Levin
BR
Bull
JJ
Phage therapy revisited: the population biology of a bacterial infection and its treatment with bacteriophage and antibiotics
Am Nat
 , 
1996
, vol. 
147
 (pg. 
881
-
98
)
56
Payne
RJH
Phil
D
Jansen
VAA
Phage therapy: the peculiar kinetics of self-replicating pharmaceuticals
Clin Pharmacol Ther
 , 
2000
, vol. 
68
 (pg. 
225
-
30
)
57
Payne
RJH
Jansen
VAA
Understanding bacteriophage therapy as a density-dependent kinetic process
J Theor Biol
 , 
2001
, vol. 
208
 (pg. 
37
-
48
)
Informed consent was obtained from all patients and volunteers participating in the study. The animal experimentation guidelines of the Ministry of Education, Science, Sports and Culture, Japan, were followed during the research. All animal experiments were approved by the Kochi Medical School Animal Use Committee
Financial support: Mochida Memorial Foundation for Medical and Pharmaceutical Research; Magobei Kobayashi Memorial Medical Foundation; Life Fund of Kochi Newspaper and Broadcasting; President Research Fund of Kochi Medical School Hospital