Development of a Novel Targeting System for Lethal Photosensitization of Antibiotic-Resistant Strains of Staphylococcus aureus

ABSTRACT Light-activated antimicrobial agents (photosensitizers) are promising alternatives to antibiotics for the treatment of topical infections. To improve efficacy and avoid possible damage to host tissues, targeting of the photosensitizer to the infecting organism is desirable, and this has previously been achieved using antibodies and chemical modification of the agent. In this study we investigated the possibility of using a bacteriophage to deliver the photosensitizer tin(IV) chlorin e6 (SnCe6) to Staphylococcus aureus. SnCe6 was covalently linked to S. aureus bacteriophage 75, and the ability of the conjugate to kill various strains of S. aureus when exposed to red light was determined. Substantial kills of methicillin- and vancomycin-intermediate strains of S. aureus were achieved using low concentrations of the conjugate (containing 1.5 μg/ml SnCe6) and low light doses (21 J/cm2). Under these conditions, the viability of human epithelial cells (in the absence of bacteria) was largely unaffected. On a molar equivalent basis, the conjugate was a more effective bactericide than the unconjugated SnCe6, and killing was not growth phase dependent. The conjugate was effective against vancomycin-intermediate strains of S. aureus even after growth in vancomycin. The results of this study have demonstrated that a bacteriophage can be used to deliver a photosensitizer to a target organism, resulting in enhanced and selective killing of the organism. Such attributes are desirable in an agent to be used in the photodynamic therapy of infectious diseases.

Methicillin-resistant Staphylococcus aureus (MRSA) is a major cause of nosocomial infections, and in the United States the rates of MRSA infection in intensive care units have reached 50% (15,33). There is also an increasing incidence of community-acquired MRSA, and concern over community-acquired MRSA has led to much current research (28). Some strains of MRSA (fortunately still a minority) can also be resistant to teicoplanin and vancomycin, and the rate at which S. aureus can develop or acquire resistance to new antibiotics seems to be higher than the rate at which new antibiotics are discovered and developed (9,19). Thus, it is essential that new approaches to preventing, controlling, and treating infections due to MRSA be developed.
One possible antimicrobial approach is lethal photosensitization. This involves the use of a light-activatable compound, a photosensitizer, which generates singlet oxygen and free radicals when exposed to light of a suitable wavelength (11,16,37). Singlet oxygen is highly reactive and cytotoxic, damaging and destroying cell walls and membranes and thus resulting in cell death (2,3). The use of light-activated antimicrobials has been shown to be effective against a variety of gram-positive and gram-negative bacteria (16,36), including MRSA (14), oral bacteria (40), and cutaneous microbes (42). In earlier studies, we have demonstrated that lethal photosensitization can be effective against epidemic strains of MRSA and that the efficiency of the treatment can be increased by targeting the photosensitizer to the bacterial cell (7). This system used immunoglobulin G (IgG) as a targeting molecule to direct the photosensitizer tin(IV) chlorin e6 (SnCe6) to protein A in the cell wall of S. aureus, exploiting the IgG binding properties of protein A. Although this system could kill some of the MRSA strains tested, not all of the strains were efficiently killed, probably due to differences in protein A expression. We have also demonstrated that an antibody against the capsule of S. aureus can be used to deliver ScCe6 to the organism (8). Another possible means of delivering a photosensitizer to S. aureus is to use a bacteriophage. The use of a bacteriophage to deliver a photosensitizer to a target organism could have two important benefits: (i) more effective killing, since the singlet oxygen produced on irradiation would be concentrated at the bacterial cell wall, and (ii) selective killing, since binding of the phagephotosensitizer conjugate to the target organism would help to reduce collateral damage to host cells and members of the indigenous microbiota at the site.
The purposes of this study were (i) to determine whether phage could be used to deliver a photosensitizer, SnCe6, to S. aureus and thereby enhance killing compared to the photosensitizer itself; (ii) to determine if cell wall thickening in vancomycinintermediate S. aureus (VISA) strains resulted in insensitivity to the treatment; and (iii) to investigate the effects of a phage-SnCe6 conjugate and red light on human epithelial cells.
Phage 75 is a serogroup F staphylococcal phage used for typing staphylococci at the Laboratory of Hospital-Associated Infection, Central Public Health Laboratory, Health Protection Agency, United Kingdom. The phage was propagated on EMRSA-16 by using the phage overlay method (see below), purified by precipitation with 20% polyethylene glycol 8000 containing 1 M NaCl, resuspended in 0.85% NaCl, and filtered through a 0.45-m-pore-size filter (Nalgene). For propagation, 300 l of mid-exponential-phase EMRSA-16 was added to 15-ml Falcon tubes. Approximately 10 5 PFU of phage 75 was then added to the tubes and allowed to incubate at room temperature for 30 min to allow the phage to infect the bacteria. Cooled molten top NB2 agar (0.35% agar) (9 ml), containing 10 mM CaCl 2 , was added to the tubes, and the mixture was poured onto undried NB2 base agar (1% agar) plates. The plates were incubated at 37°C overnight. The top agar was removed by adding 1 ml of NB2 (containing 10 mM CaCl 2 ) to each plate to aid the scraping of the top agar into a small centrifuge tube. The collected agar was then centrifuged and the supernatant collected and passed through a 0.45-m filter to remove any bacterial cells. The resulting phage suspension was stored at 4°C.
The titer of the bacteriophage was determined by infecting EMRSA-16 by using the phage overlay method as follows. Tenfold serial dilutions of the phage suspension were made in sterile NB2 medium with 10 mM CaCl 2. A 30-l aliquot of each dilution was added to 300 l EMRSA-16 (OD 600 , approximately 1.0), and the mixtures were then added to 15-ml Falcon tubes containing 3 ml of molten NB2 agar containing 10 mM CaCl 2 . Each of these was poured onto dried NB2 agar plates, which were then incubated at 37°C overnight. The numbers of plaques on plates that contained between 30 to 300 plaques were then counted.
Ability of bacteriophage 75 to infect strains. Phage 75 is known to infect EMRSA-16, EMRSA-3, and EMRSA-15 (4a). The susceptibilities to infection by phage 75 of the other bacterial strains used in the study were determined by infecting mid-exponential-phase cultures of the above-described staphylococcal strains with 10 Ϫ1 to 10 Ϫ5 10-fold serial dilutions of a high-titer phage 75 stock, according to the method described above.
Laser and photosensitizer. The laser used in this study was a model 127 Stabilite helium-neon laser (SpectraPhysics, Mountain View, CA), with a measured output of 35 mW, which emits light with a wavelength of 632.8 nm. The photosensitizer was SnCe6, which was custom synthesized by, and purchased from, FrontierScientific (Lancashire, United Kingdom).
Preparation of phage-SnCe6 conjugate. Phage 75 was conjugated to the SnCe6 by using a zero-length cross-linker in an adaptation of the method described by Embleton et al. (7). Basically, 4 mg SnCe6 was dissolved in 2.4 ml conjugation buffer [0.1 M 2-(N-morpholino)ethanesulfonic acid pH 6.0, 0.5 M NaCl] along with 2 mM 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDAC) (Sigma, Poole, United Kingdom) and 5 mM sulfo-N-hydroxysuccinimide (Pierce, Cheshire, United Kingdom). After a 4-hour incubation with constant mixing at room temperature, ␤-mercaptoethanol was added to a concentration of 20 mM to quench the EDAC, and the reaction mix was adjusted to pH 7.6. Phage 75 in 0.85% NaCl (3 ml) was then added and the reaction mixture incubated overnight (16 h) with mixing at room temperature before quenching with ethanolamine at a concentration of 20 mM. The phage-SnCe6 conjugate was precipitated with 20% polyethylene glycol 8000 containing 1.0 M NaCl twice before being resuspended in sterile 0.85% NaCl and dialyzed against at least four changes of 0.85% NaCl so that free SnCe6 was eliminated from the preparation. The amount of SnCe6 present in the conjugate was measured by absorbance at 633 nm. To estimate the titer of the phage, an infection assay was performed. The concentration of the phage-SnCe6 conjugate was expressed in terms of the concentration of SnCe6 present in the conjugate.
Lethal photosensitization. The phage-SnCe6 conjugate (50 l) was added in duplicate to 50-l aliquots of bacterial suspension (containing 1 ϫ 10 7 CFU/ml) in a sterile 96-well plate. These were incubated with stirring for 30 min prior to exposure to a measured dose of red light from the HeNe laser (controls were not exposed to laser light). The typical dose was 21 J/cm 2 , which corresponds to a 5-min exposure using the aforementioned equipment (LϩSϩ). The following controls were also performed: addition of saline and no exposure to laser light (LϪSϪ), addition of unconjugated phage (at the same titer as the conjugate) in the absence of laser light, and addition of unconjugated SnCe6 and exposure to laser light. Following treatment, the survivors were enumerated by viable counting. Serial 1-in-10 dilutions in sterile phosphate-buffered saline were prepared from the contents of each well, and 25 l from each of these was plated in duplicate on blood agar. The number of colonies on each plate was counted following overnight incubation at 37°C. The limit of detection in all experiments was 400 CFU/ml. All experiments were carried out in duplicate on each occasion and on at least two separate occasions.
To compare killing of EMRSA-16 with epithelial cells (Hep2 cells) (see below), the experiments described above were repeated, using SnCe6 and phage-SnCe6 at final concentrations of 5 g/ml. However, the samples were mixed in the 96-well plates but were not stirred continuously. A lens was used to adjust the diameter of the beam to that of a well in a 96-well plate.
Killing of Mu3 and Mu50 grown in the presence of vancomycin. Mu3 and Mu50 were grown aerobically in NB2 broth (Oxoid) and were passaged through media containing increasing concentrations of vancomycin (0 to 16 mg/liter). The cells were harvested and exposed to phage-SnCe6 (5 g/ml SnCe6, 6.3 ϫ 10 8 PFU/ml) and laser light as described above, and the survivors were enumerated by viable counting.
Lethal photosensitization of epithelial cells. For these experiments, Hep-2 cells (a human epithelial cell line) were used. The Hep-2 cells were maintained in Dulbecco's minimal Eagle's medium (Gibco, Paisley, United Kingdom) supplemented with 10% fetal calf serum (Sigma Ltd.) and incubated at 37°C in 5% CO 2 -air. For experimental purposes, Hep-2 cells (in Dulbecco's minimal Eagle's medium without antibiotics) were plated into 96-well plates at a concentration of 2 ϫ 10 4 cells/well and left overnight to adhere to the surfaces of the wells. The following day, the medium was removed and the following additions made: 100 l of SnCe6 or phage-SnCe6 (in 0.85% NaCl, 10 mM CaCl 2 ) to a final concentration of 5 g/ml (LϩSϩ) and 100 l 0.85% NaCl, 10 mM CaCl 2 (LϪSϪ). LϩSϩ samples were exposed to 21 J/cm 2 laser light but were not stirred throughout the exposure. Instead, a lens was used to adjust the diameter of the beam to that of a well in a 96-well plate. LϪSϪ samples were not exposed to laser light.
Following the exposures and/or additions, the added solutions were removed and fresh medium was added. Cell viability was determined using the 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay as follows. A stock MTT solution (20 g/well) was added to the wells, and the plate contents were incubated for 4 h at 37°C. The medium was removed, acid-isopropanol (100 l of 0.04 M HCl in isopropanol) was added to elute the red formazan, and the suspension was mixed thoroughly. The concentration of formazan in the solution in each well was determined on a plate Titertek Multiskan PLUS plate reader, using a test wavelength of 570 nm and a reference wavelength of 620 nm.

RESULTS
Conjugation of bacteriophage 75 to SnCe6 resulted in suspensions containing between 10 and 15 g/ml of SnCe6 and 2.5 to 5 ϫ 10 9 PFU/ml phage 75.
Ability of phage 75 to infect strains. Phage 75 is a bacteriophage that is used for typing MRSA strains at the Laboratory of Hospital-Associated Infection, Central Public Health Laboratory, Health Protection Agency. The phage was found to be capable of infecting and lysing EMRSA-16 and, to a lesser extent, EMRSA-3, EMRSA-15, and NCTC 6571. It was not able to lyse any of the other strains used in the study.
Ability strain used (7,8). Therefore, the effectiveness of the phage-SnCe6 conjugate against different strains, including the VISA strain Mu50, the hetero-VISA strain Mu3, and the communityacquired MRSA strain MW2, was examined. As can be seen in Table 1, the phage-SnCe6 conjugate was able to achieve substantial reductions in the viable counts of a variety of strains of S. aureus, including MRSA, hetero-VISA, and VISA strains. Reductions of greater than 3 log 10 units in the viable count of EMRSA-1 and EMRSA-16 were achieved when the bacteria were irradiated with 21 J/cm 2 of red light in the presence of the phage-SnCe6 conjugate, whereas the phage itself, the phage-SnCe6 conjugate in the absence of laser light, and the SnCe6 (in the presence of light) did not have an appreciable effect on the viability of either of the two strains. From Table 1 it can be seen that the susceptibility of S. aureus to lethal photosensitization using the unconjugated SnCe6 was markedly strain dependent; EMRSA-15, Mu50, and NCTC 6571 were particularly susceptible (Ͼ1-log 10 -unit loss in the viable counts). All of the strains tested were susceptible to lethal photosensitization using the phage-SnCe6 conjugate, with kills of 4 log 10 units achieved in most cases. With phage 75 alone, the viable count of only EMRSA-16 was reduced. For all strains, the levels of killing achieved with the conjugate were much greater than those with unconjugated SnCe6.
Phage-SnCe6 conjugate is effective in killing EMRSA-16 in the presence and absence of CaCl 2 . The adsorption of phages onto bacteria is thought to require the presence of a divalent metal cation such as Ca 2ϩ , so it was reasoned that a divalent metal cation may be necessary for phage 75 to target SnCe6 to EMRSA-16 effectively. Therefore, experiments (using the conjugate at 1.5 g/ml) were performed with 0.85% NaCl containing either 10 mM CaCl 2 or 10 mM EDTA (Fig. 1). Following exposure to laser light, the phage-SnCe6 conjugate was able to reduce the viable counts of EMRSA-16 in either case, although the killing in the presence of CaCl 2 was much greater than that in the absence of divalent metal ions, with 99.9% and 75% reductions in viability, respectively. The presence or absence of divalent metal ions did not affect the efficiency of killing by unconjugated SnCe6, which killed approximately half of the cells in each case. A control of phage 75 alone, estimated to be at the same titer as phage in the phage-SnCe6 conjugate, was also used to determine the level of killing due to phage alone and the requirement of this for divalent metal ions. In the presence of CaCl 2 , the viable count was reduced by approximately 20%, while in the absence of divalent metal ion, no kills due to phage 75 alone were observed.
Comparison of phage-SnCe6 conjugate with SnCe6. To further investigate whether targeting the SnCe6 to MRSA by using phage 75 increased the effectiveness of lethal photosensitization, EMRSA-16 was irradiated in the presence of increasing concentrations of either unconjugated SnCe6 or phage-SnCe6 (Fig. 2). Samples irradiated in the presence of the phage-SnCe6 conjugate had much lower viable counts than those irradiated in the presence of SnCe6: at a low concentration of 1 g/ml, the proportions of EMRSA-16 surviving irradiation with laser light in the presence of phage-SnCe6 and SnCe6 were 0.15% and 21%, respectively. As the concentration of the photosensitizer was increased, the extent of killing also increased, and the phage-SnCe6 performed consistently better than the free SnCe6.
Effect of growth phase. The susceptibility to lethal photosensitization of EMRSA-16 in different growth phases is shown in Fig. 3. There was little difference in the kills obtained in the different growth phases under the conditions employed, i.e., with the phage-SnCe6 conjugate containing 1.5 g/ml of SnCe6, phage at 1.46 ϫ 10 7 PFU/ml, and a light dose of 21 J/cm 2 .
Comparison of the effects of phage-SnCe6 on human epithelial cells and EMRSA-16. The effects of the conjugate and of free SnCe6 on a human epithelial cell line, Hep2, were tested by seeding a 96-well plate with Hep2 cells and then subjecting them to treatment with SnCe6 or phage-SnCe6 at a final concentration of 5 g/ml and with laser light (Fig. 4). Control experiments with the addition of saline only were also  a To six MRSA strains, one hetero-VISA strain, one VISA strain, and one MSSA strain, suspended in 0.85% NaCl-10 mM CaCl 2 , either phage, SnCe6 (final concentration ϭ 1.5 g/ml), or phage-SnCe6 (final concentration ϭ 1.5 g/ml [3.8 ϫ 10 8 PFU/ml]) was added as appropriate, and aliquots were exposed to 21 J/cm 2 HeNe laser light.
b The percent kill of each suspension is the mean of at least four repeats relative to a control, unirradiated suspension to which neither phage nor photosensitizer was added.
performed. The degree of killing was measured using the cell viability assay described above. In addition, a tandem experiment demonstrating the killing of EMRSA-16 under comparable conditions was also performed, using a colony-counting method to ascertain the level of killing. Under the conditions employed, the viability of the epithelial cells was unaffected by irradiation in the presence of either the SnCe6 or the phage-SnCe6 conjugate. In contrast, reductions of 2 and 4 log 10 units in the viable count of EMRSA-16 were obtained using the SnCe6 and phage-SnCe6 conjugate, respectively. Effect of growth with vancomycin on the susceptibility to lethal photosensitization of VISA strains. To examine the ef-fects of growing vancomycin-intermediate strains in the presence of vancomycin on their susceptibility to lethal photosensitization by phage-SnCe6 and red light, Mu3 and Mu50 were grown in media with and without vancomycin. Both strains were passaged through media containing increasing concentrations of vancomycin (0 to 16 mg/liter) and then irradiated with laser light (21J/cm 2 ) in the presence of the phage-SnCe6 conjugate (5 g/ml SnCe6, 6.3 ϫ 10 8 PFU/ml) (Fig. 5). Both Mu3 To either human epithelial cells (Hep2 cells) or EMRSA-16, SnCe6 or phage-SnCe6 conjugate was added to give a final concentration of 5 g/ml, and the cultures were then exposed to 21 J/cm 2 of laser light. To LϪSϪ samples, only saline was added and the suspensions were not exposed to laser light. Error bars indicate standard deviations. To bacterial suspensions in 0.85% NaCl-10 mM CaCl 2 , 0.85% NaCl (LϪSϪ) or phage-SnCe6 (LϩSϩ) was added and exposed to laser light. Numbers above the bars show the percent kills achieved. The limit of detection was 400 CFU/ml. Error bars indicate standard deviations. and Mu50 were very susceptible to killing in all cases, with a reduction in the viable count of over 3 log 10 units regardless of the vancomycin concentration.

DISCUSSION
The lethal photosensitization of microbes is one of several alternatives to antibiotics that are being developed for use in the treatment of infections due to antibiotic-resistant organisms, and a number of studies have demonstrated its effectiveness in animal models (12,17,20). The light used (usually generated by a laser) in such an approach must, obviously, not have any adverse effects (e.g., due to phototoxicity or heating) on host tissue. The helium/neon laser is a device emitting light with a wavelength of 632 nm, and, at the light energy doses and densities needed to activate photosensitizers such as tin chlorin e6, it does not result in damage to human tissue. Indeed, irradiation of wounds with light from a helium/neon laser (without the application of a photosensitizer) has met with some success as a means of stimulating wound healing in humans (29,30). One possible problem with the use of lightactivated antimicrobial agents is that the singlet oxygen produced during the process has the potential to also damage neighboring host cells. There is, therefore, great interest in developing methods of targeting the photosensitizer to the infecting organism. As well as achieving selectivity, another advantage of using a targeted photosensitizer is that the antimicrobial effectiveness of lethal photosensitization is also increased. This is because, following binding of the targeted photosensitizer to the organism, subsequent irradiation results in the generation of singlet oxygen only in the vicinity of the pathogen and not at extraneous sites. Consequently, less photosensitizer needs to be applied, and because there is less attenuation of the incident light by unbound photosensitizer, a lower light dose can be used. Those targeting systems which have shown promise in laboratory and in vivo studies include the use of antibodies (4,7,8) and chemical modification of the photosensitizer itself (12,32). In the present study we investigated the possibility of another targeting strategy, one based on the use of bacteriophages.
Bacteriophages, in their own right, are exciting prospects as antibacterial agents, although problems could arise due to the induction, or the prior existence, of an immune response to the particular bacteriophage being used (22,35). However, studies with humans have demonstrated that bacteriophages may be applied topically and systemically without inducing adverse effects (22,34,35). One of the problems with using bacteriophages in the treatment of infections due to S. aureus is their restricted host range. There are a number of polyvalent staphylophages which lyse a broad range of S. aureus strains, but again, certain strains remain resistant (26,27). All staphylophages are thought to recognize the same receptor located in the peptidoglycan-teichoic acid complex on the surface of staphylococci, and thus differences in the susceptibility of strains are believed to be due to postadsorption phenomena (26). Indeed, Pantucek et al. (27) have shown that the polyvalent phage 812, which was able to lyse over 80% of 782 S. aureus strains tested, did in fact bind to strains that were resistant to this staphylophage. It is apparent from these studies that the use of a polyvalent phage such as 812 as a treatment for staphylococcal infections would not provide full protection. However, since bacteriophages also bind to staphylococci that are not susceptible to infection, they could serve as a delivery system for photosensitizers. Bacteriophages have already been used successfully to deliver other antimicrobial agents to bacteria. For example, Westwater et al. (39) have used a nonlytic phage of Escherichia coli as a vehicle to deliver DNA encoding bactericidal proteins to E. coli in vitro and in a mouse bacteremia model and have found that target bacterial cell numbers were reduced by several orders of magnitude.
The results presented here have demonstrated that it is possible to use bacteriophage 75 to deliver the photosensitizer SnCe6 to various strains of S. aureus and thereby enable killing of the organisms upon irradiation with red light. Furthermore, not only was the phage-SnCe6 conjugate more effective at killing S. aureus than the unconjugated photosensitizer, it was effective at killing EMRSA-16 and under similar conditions (but in separate experiments) did not affect the viability of human epithelial cells.
The phage 75-SnCe6 conjugate was able to sensitize a variety of strains of S. aureus to killing by red light; these included strains responsible for epidemics in the United Kingdom (1,21), community-acquired MRSA strains (6,28), and vancomycin-intermediate strains (13,38). Killing was not the result of phage-mediated lysis, as no significant reduction in the viable count was achieved in control cultures containing only the phage (with the exception of EMRSA-16). Furthermore, strains that were not lysed by phage 75 (e.g., EMRSA-3, Mu3, and Mu50) were also killed on irradiation in the presence of the phage-SnCe6 conjugate, implying that the bactericidal effect was due to lethal photosensitization.
Binding of bacteriophage to S. aureus is known to be affected by Ca 2ϩ ions and/or other divalent metal ions. The presence of the metal ion chelator EDTA resulted in a decrease in the kills obtained using either phage 75 alone or the phage-SnCe6 conjugate, although it had no effect on killing by the unconjugated photosensitizer. That the enhanced killing with phage-SnCe6 required the presence of divalent metal ions further supports the idea that the phage-SnCe6 is indeed binding to the strains and that this is sufficient for targeted lethal photosensitization.
The effect of growth phase on the susceptibility of S. aureus to lethal photosensitization appears to depend on the nature of the photosensitizer used. For example, killing induced by aluminum disulfonated phthalocyanine is not affected by the growth phase of the organism (41), whereas in the case of SnCe6, lag-phase cells are less susceptible to light-induced killing (7). The efficacy of some targeted photosensitizers can also be growth phase dependent. For example, a conjugate of IgG and SnCe6 is a less effective photosensitizer of lag-phase cells of MRSA than of cells from other growth phases. As the IgG targets the protein A on the organism, this reduced effectiveness may be attributable to reduced production of protein A during the lag phase of growth. In contrast, SnCe6 conjugated to an antibody against the capsule of S. aureus has a reduced efficacy against stationary-phase cells of MRSA (8). The phage-SnCe6 conjugate appears to be equally effective against MRSA in all growth phases.
One of the features common to vancomycin-intermediate strains of S. aureus is thickening of the cell wall in the presence 3694 EMBLETON ET AL. ANTIMICROB. AGENTS CHEMOTHER.
of the antibiotic (5,18). Such thickening is thought to be a major contributing factor to vancomycin resistance, as it results in entrapment of molecules of the antibiotic, thereby preventing them from reaching their intracellular target. Cell wall thickening may result in reduced susceptibility of the organism to lethal photosensitization for a number of reasons. First, the major targets of singlet oxygen are the bacterial cell wall and membrane, so that any increase in the thickness of the cell wall could protect both structures against the harmful effects of singlet oxygen. Second, it may alter the number and/or distribution of phage receptors and so could affect the number of molecules of the phage-SnCe6 conjugate adsorbed to the bacterial surface. The results of this study revealed that after growth of the strains in up to 16 mg/liter of vancomycin, the Mu3 and Mu50 strains remained highly susceptible to lethal photosensitization using the phage-SnCe6 conjugate.
The results of a number of in vitro studies have shown that the light dose and photosensitizer concentration required to achieve killing of bacteria have little effect on the viability of host cells in tissue culture (31,32). Furthermore, studies with animals have reported little evidence of damage to host tissues following the successful lethal photosensitization of infecting bacteria (20,23,24). Nevertheless, it is important to take all possible precautions against the possibility of collateral damage to host tissues during antimicrobial photodynamic therapy, and this is one of the potential benefits of using a photosensitizer-targeting system. The results of this study have shown that a light dose and phage-SnCe6 concentration able to achieve appreciable killing of MRSA had no effect on the viability of human epithelial cells in vitro. These results imply that damage to host tissue during antimicrobial photodynamic therapy may be avoidable. However, the results of such studies are very preliminary and certainly do not constitute a model of the situation in vivo. Issues such as the effect of light on conjugatecoated bacteria attached to, or within, phagocytes and other host cells need to be addressed. It is essential, therefore, that further investigations be carried out using appropriate models so that the likely effects on the host of the combined effect of light and the photosensitizer conjugate can be assessed.
Use of appropriate bacteriophages could enable targeting of other microbes responsible for diseases of humans and other animals. Bacteriophage-photosensitizer conjugates could, therefore, provide an effective means of selectively killing infecting organisms without damaging host tissues.