Investigation into the toxicity and membrane interactive properties of a range of cationic photosensitisers with potential application to Photoantimicrobial Chemotherapy

Abstract

The emergence of bacterial strains with resistance to traditional antibiotics has necessitated fresh approaches to the treatment of infections. In the present study, the therapeutic potential of a series of phenothiazinium based photosensitisers (PhBPs) to act as novel photodynamic antibacterial agents is investigated. It is shown that Azure A, Azure B, Azure C, Brilliant Cresy! Blue, Neutral Red, Pyronin Y, To!uidine Blue 0, New Methylene B!ue and Dimethylmethylene Blue possess light adsorption maxima (Xm), which range from 536 nm - 648 nm, acceptab!e for therapeutic use.
Under dark conditions, all the PhBPs examined were able to kill E. coli with MLCs in the range 0.98 ltM and 1000 MM., indicating significant levels of inherent toxicity. When illuminated (fluence rate of 1.7 mW emj, except for Azure A and Brilliant
Cresyl Blue, these PhBPs showed reductions in their MLCs that varied from two-fold to sixtyfour-fold indicating high levels of photo-toxicity to E. co/i. The relative singlet oxygen yield of these PhBPs was generally 0.77 or above, consistent with
high levels of singlet oxygen production and suggesting that some may utilise type II mechanisms of photosensitisation. The lipophilieity (log P) of these PhBPs was generally 0.7 or above, showing these photosensitisers to be lipojihilic and suggesting an ability to interact with lipid or membranes. Studies on the potential of these PhBPs for lipid / membrane action showed that all of these dyes possessed inherent abilities for haemolytic activity (ranging between 6% and 19%) and bacteriolytic activity when directed against E. co/i (ranging between 6% and 13%). However, upon illumination (fluence rate of 1.7 mW em"). only Dimethyl Methylene Blue showed increased levels of haemolytic ability (11% to 61%) whilst only New Methylene Blue and Dimethylmethylene Blue showed increased levels of bacteriolysis (from 6% to 9%, and 9% to 11%, respectively). Consistent with their inherent lytic abilities, each of the PhBPs studied showed the ability to interact with lipid monolayers mimetic of bacterial membranes (DOPG / DOPE, 30:70 molar ratio, surface pressure 30 mN M 5, inducing initial changes in surface pressure that ranged between 3 mN M' and 13 mN M', consistent with monolayer association / penetration. However, in most cases, subsequent monolayer decay was observed with decay rates that ranged between 0.5
10-3 and 2.9 x lO- mN M' se6* Generally consistent with their lack of light driven haemolytic and bacteriolytic abilities, the exposure of majority of the PhBPs studied induced to light induced no changes in the saturated to unsaturated ratio of either C18 fatty acids, or C8 fatty acids, that had been isolated from membranes of E. coil HBIOI. However, illuminated Toluidine Blue 0, New Methylene Blue and Dimethylmethylene Blue (fluence rate of 1.7 mW cm') were found to induce changes to these ratios, suggesting photodynamic attack on E. coil membrane lipid. For C16 fatty acids, the dark ratios of saturated to unsaturated fatty acids in the presence of PhBPs were 0.14, 0.15 and 0.27 respectively, which under light conditions increased to 0.20, 0.25 and 0.35, respectively. In addition, illuminated Dimethylmethylene Blue (fluence rate of 1.7 mW cm 1) was found to induce changes in the saturated to unsaturated ratio of C18 fatty acids, which was 0.45 under dark conditions and increased to 0.65 under light conditions.
It is suggested that whilst the majority of PhBPs studied are lipid / membrane interactive, light driven lipid attack may make only a minor contribution to the phototoxicity of Toluidine Blue 0, New Methylene Blue and Dimethylmethylene Blue when directed against E. coil. However, it is further suggested that, Azure A and Brilliant Cresyl Blue are not photo-toxic to E. coil and in general, bacterial membrane lipid is not a key site of photodynamic action for the PhBPs studied. Whilst there is the possibility that other membrane components may be the key targets of these PhBPs, which are photo-toxic to E. coil, based on the results obtained here, it is suggested that the PhBPs studied may be taken up by this organism to attack intracellular targets. Incidental damage arising from such uptake and I or membrane association may then be responsible for the inherent cellular lysis and toxicity to E. coil observed for these dyes.