Photodynamic Inactivation of Bacillus Spores, Mediated by Phenothiazinium Dyes

Spore formation is a sophisticated mechanism by which some bacteriasurvive conditions of stress and starvation by producing a multilayeredprotective capsule enclosing their condensed DNA. Spores arehighly resistant to damage by heat, radiation, and commonlyemployed antibacterial agents. Previously, spores have alsobeen shown to be resistant to photodynamic inactivation usingdyes and light that easily destroy the corresponding vegetativebacteria. We have discovered that Bacillus spores are susceptibleto photoinactivation by phenothiazinium dyes and low doses ofred light. Dimethylmethylene blue, methylene blue, new methyleneblue, and toluidine blue O are all effective, while alternativephotosensitizers such as Rose Bengal, polylysine chlorin(e6)conjugate, a tricationic porphyrin, and a benzoporphyrin derivative,which easily kill vegetative cells, are ineffective. Sporesof Bacillus cereus and B. thuringiensis are most susceptible,B. subtilis and B. atrophaeus are also killed, and B. megateriumis resistant. Photoinactivation is most effective when excessdye is washed from the spores, showing that the dye binds tothe spores and that excess dye in solution can quench lightdelivery. The relatively mild conditions needed for spore killingcould have applications for treating wounds contaminated byanthrax spores, for which conventional sporicides would haveunacceptable tissue toxicity.

INTRODUCTION

Top

Introduction

The genus Bacillus comprises a group of gram-positive endospore-formingrod-shaped aerobic bacteria and is widely distributed in soilenvironments. Sporulation is a protective mechanism wherebybacteria can survive in the environment for long time periods,even at physical extremes for life forms on earth. Bacillusanthracis is a pathogen, and in 1887, was the first bacteriumshown to be the cause of a disease by Robert Koch, who demonstratedits ability to form endospores and produced experimental anthraxby injecting it into animals (31). B. anthracis has been proposedto be an important organism in biological warfare and bioterrorismdue to its ability to be weaponized by producing a finely groundhighly infective spore powder that could be widely dispersed(12). B. anthracis is a member of a closely related family ofsix species (22), including B. cereus, which is less pathogenicto humans (1, 5), and B. thuringiensis, which is pathogenicto insects (45).

Spores are relatively resistant to many of the disinfectantand antiseptic agents that routinely destroy vegetative bacteria,such as alcohols, phenols, chlorhexidine, and benzalkonium compounds(37, 39). Agents commonly cited as being sporicidal includeheat greater than 121°C, formaldehyde or glutaraldehyde,strong hypochlorite solutions, chlorine dioxide, and ionizingor UV radiation (38, 55).

Photodynamic therapy (PDT) is a therapy for cancer and otherdiseases that has received regulatory approvals for severalindications in many countries (10). It employs nontoxic dyesknown as photosensitizers (PS) and visible light of the appropriatewavelength to excite the PS molecule to the excited singletstate. This excited state undergoes intersystem crossing tothe long-lived triplet state, which can react with molecularoxygen to produce cytotoxic species such as singlet oxygen,superoxide, and radicals (9, 11). These reactive oxygen speciescan then oxidize many biological molecules, such as proteins,nucleic acids, and lipids, leading to cell death. PDT has theadvantage over other therapies of dual selectivity: not onlyis the PS targeted to the affected tissue or cell type, butthe light can also be accurately delivered to the appropriatearea. Although originally developed as a cancer treatment, PDThas been successful as a treatment for age-related macular degenerationin ophthalmology (2) and is under investigation as a treatmentfor infectious disease (16).

The rapidly increasing emergence of antibiotic resistance amongpathogenic bacteria may be bringing to an end a period extendingover the past 50 years termed "the antibiotic era" (59) andhas led to a major research effort to find alternative antibacterialtherapeutics (6), such as PDT, to which we have hypothesizedbacteria will not be easily able to develop resistance (8).Indeed, Lauro et al. failed to demonstrate the development ofresistance in Actinobacillus actinomycetemcomitans and Peptostreptococcusmicros after 10 cycles of subtotal photodynamic inactivation(PDI) (28). Because the delivery of visible light is almostby definition a localized process, PDT for infections is likelyto be applied exclusively to localized disease by local deliveryof the PS into the infected area by methods such as topicalapplication, instillation, interstitial injection, or aerosoldelivery (16).

It has been known since the first days of PDT early in the 1900sthat certain microorganisms can be killed by the appropriatecombination of dyes and light in vitro, but in the last 15 yearsprogress has been made in defining the precise molecular featuresof the PS necessary for binding to and subsequently killingby photodynamic action microbes such as bacteria, fungi, andviruses (51). Bacillus subtilis spores, however, have been reportedto be resistant to PDT using the dye Rose Bengal under conditionswhich easily led to the killing of vegetative cells (40).

We decided to carry out a study examining candidate antimicrobialPS for their ability to mediate the light-induced destructionof bacterial spores. We describe here for the first time theefficient photoinactivation (PDI) of Bacillus spores, includingthose closely related to anthrax, mediated by members of thefamily of phenothiazinium dyes.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains and culture conditions.
The following Bacillus strains were used: B. atrophaeus (ATCC9372), B. cereus (ATCC14579), B. megaterium (ATCC14581), B.thuringiensis (ATCC 33740), and B. subtilis (ATCC 6051). Vegetativecells were routinely cultured in brain heart infusion (BHI)broth with shaking and on BHI agar, both at 37°C. For initialexperiments, spores of B. atrophaeus and B. cereus were purchasedfrom SGM Biotech, Inc. (Bozeman, MT). Subsequently, spores ofall species were prepared in our laboratory, using sporulationbroth for B. atrophaeus, B. megaterium, and B. subtilis andsporulation agar (4, 32) for B. cereus and B. thuringiensis.The sporulation medium consisted of 16.0 g nutrient broth (Difco),2.0 g KCl, and 0.5 g MgSO₄ per liter; 17 g of agar was addedif needed. The pH of the medium was adjusted to 7, and thenthe medium was autoclaved and cooled. After the medium was cooled,1 ml of 1 M Ca₂(NO₃)₂, 1 ml of 0.1 M MnCl₂ · 4H₂O, 1ml of 1 mM FeSO₄, and 2 ml of 50% glucose were added. Bacteriawere grown in sporulation medium for 3 days.

For spore purification, the mixture of spores and cells wascentrifuged at 1,300 x g for 20 min, washed with 5 volumes 1M KCl-0.5 M NaCl, rinsed with sterile deionized water, washedwith 1 M NaCl, and rinsed with sterile deionized water again.Lysozyme (50 µg/ml) was added in the presence of buffer(5 volumes 0.05 M Tris-Cl, pH 7.2) and incubated with constantstirring at 4°C overnight. Lysozyme was removed by centrifugationeight times (at 1,300 x g) and washing with sterile deionizedwater. In a control experiment, B. cereus spores were preparedwithout the use of lysozyme. The purity of spore suspensionswas checked using Wirtz-Conklin stain (5% aqueous malachitegreen and safranine O) (20) and transmission electron microscopy.Spores were >95% pure, with the remaining material beingfragments of vegetative sporangium (data not shown). Sporeswere then frozen at –80°C in phosphate-buffered salinewithout Ca²⁺ or Mg²⁺ (PBS) with 10% glycerol and stored untiluse. To avoid germination, spores were used immediately afterdefrosting.

Photosensitizers and light sources.
Eight PS were employed (Fig. 1 shows their structures). RoseBengal (RB), toluidine blue O (TBO), methylene blue (MB), newmethylene blue N (zinc chloride double salt; NMBN), and 1,9-dimethylmethyleneblue chloride (DMMB) were purchased from Sigma (St. Louis, MO).These PS were dissolved in distilled water to give stock solutionswith a dye concentration of 2 mM. 5-Phenyl-10,15,20-tris(N-methyl-4-pyridyl)porphyrinchloride [TriP(4)] was generous gift of G. M. T. Smijs and R.van der Steen (Leiden University Medical Center, The Netherlands),and a stock solution of 1 mM was prepared in PBS. The poly-L-lysinechlorin(e6) conjugate had an average of two ce6 molecules attachedper pl chain, with an average degree of polymerization of 167lysine residues, and was prepared as described previously (17)and stored as a 1.6 mM stock solution in water. A benzoporphyrinderivative (BPD) was a gift of QLT Inc. (Vancouver, Canada)and was dissolved in dimethyl sulfoxide at a concentration of300 µM.

View larger version (30K):
[in this window]
[in a new window]
FIG. 1. Chemical structures of the PS used.

All PS stock solutions were stored at 4°C in the dark forno more than 2 weeks, and immediately before experiments, werediluted in PBS without Ca²⁺ or Mg²⁺. Spectra of stock solutionsdiluted 140- to 280-fold in methanol were recorded on a UV-visiblespectroscopy system (Waldbronn, Germany). A noncoherent lightsource with interchangeable fiber bundles (LumaCare, London,United Kingdom) was employed. Thirty-nanometer-band-pass filtersat ranges of 540 ± 15 nm for RB, 635 ± 15 nm forTBO, DMMB, and TriP(4), 660 ± 15 nm for MB, NMBN, andpLce6, and 690 ± 15 nm for BPD were used. The total poweroutput provided out of the fiber bundle ranged from 300 to 700mW.

Uptake studies.
Uptake studies were performed with all spore species and TBO.Preliminary studies had shown that inactive PS such as RB gaveno detectable uptake of dye by the spores (data not shown).Spores at a density of 10⁷/ml were incubated for 3 h with PSin the dark at 37°C. Unbound PS was washed out by centrifugationof the mixture of dye and microorganisms for 6 min at 1,550x g, followed by the resuspension of washed pellets in 10 mlPBS without Ca²⁺ or Mg²⁺. Aliquots (200 µl) of these suspensionswere used for PDI experiments, the remaining suspensions werecentrifuged again, and the pellets were dissolved in 6 ml 0.1M NaOH-1% sodium dodecyl sulfate (SDS) for at least 24 h togive a clear solution. These solutions were used for PS uptakemeasurements. The fluorescence of dissolved pellets was measuredon a spectrofluorimeter (FluoroMax3; SPEX Industries, Edison,NJ). The excitation wavelength was 620 nm, and the range foremission was 627 to 720 nm. A calibration curve was constructedfrom pure TBO at different concentrations dissolved in NaOH-SDSand used for determination of the PS concentration in dissolvedpellets. Uptake values were obtained by dividing the numberof nmol of PS in the dissolved pellet by the number of CFU obtainedby serial dilutions and the number of PS molecules/cell, calculatedby using Avogadro's number.

PDI studies.
Suspensions of spores or bacteria at 10⁷/ml were incubated withPS in the dark at 37°C. The incubation time was 3 h, andthe PS concentration was varied from 5 to 1,600 µM. Illuminationwas performed either after or before excess dye was washed out.Fluence levels ranged from 0 to 200 J/cm² at an irradiance of200 to 400 mW/cm². During illumination after defined fluenceshad been delivered, aliquots of 20 µl were taken to determinethe CFU. The contents of the wells were mixed before sampling.The aliquots were serially diluted 10-fold in PBS without Ca²⁺or Mg²⁺ to give dilutions of 10^–1 to 10^–6 timesthe original concentrations and were streaked horizontally onsquare BHI agar plates as described by Jett et al. (24). Plateswere incubated at 37°C overnight. Colonies were counted,and the survival fraction was determined relative to an untreatedcontrol. No additional treatment to enhance spore germinationwas used.

Two types of control conditions were employed, namely, illuminationin the absence of PS and incubation with PS in the dark. PSwere generally not toxic for spores or vegetative cells in thedark, and light alone did not cause cell destruction. All experimentswere performed in triplicate.

RESULTS

Top

Results

Spores are resistant to photoinactivation, except that by TBO.
We initially screened a panel of PS to test PDI of B. cereusspores (see chemical structures in Fig. 1). BPD is a clinicallyused PS that has found wide applicability in ophthalmology andhas undergone testing against cancer. We previously reportedthat it was active in mediating PDI against gram-positive bacteria(44). TriP(4) is a cationic porphyrin that has been reportedto be active in mediating PDI against both gram-positive andgram-negative bacteria (27) as well as fungi (43) and viruses(47). The pL-ce6 conjugate is a member of a class of macromolecularconjugates between polycationic polymers and PS that show ahigh degree of efficiency in mediating PDI of both classes ofbacteria in vitro and in vivo (16, 18, 19). RB was previouslyreported to be ineffective at killing B. subtilis spores butis known to kill vegetative bacterial cells efficiently (40).TBO is the most frequently used member of the class of phenothiaziniumdyes for mediating PDI of bacteria and fungi (52, 57, 58). Becausewe initially did not know whether the dyes would bind to thespores, we did not wash the spores after incubation with thedye.

As shown in Fig. 2, only TBO (50 µM) demonstrated a light-dose-dependentloss of viability of B. cereus spores, with 40 J/cm² of 630-nmlight leading to 99.999% killing. All of the other PS, evenwhen used at a 100 µM concentration and with 100 J/cm²of the appropriate light, led to no loss of viability whatsoever.We repeated the TBO PDI experiment with B. cereus spores thathad been prepared without the use of lysozyme. There was nodifference in killing between spores that had been treated withlysozyme and those that had not (data not shown). This showsthat lysozyme does not remove an outer coat layer or part ofthe cortex from the spores, thus increasing the penetrationof the dye into the spore interior. We checked to be sure thatthe incubation of B. cereus spores for 3 h with 100 µMTBO solution at 37°C did not cause germination, using phase-contrastmicroscopy, and that incubated spores were as resistant to 70%ethanol as control spores (data not shown).

View larger version (27K):
[in this window]
[in a new window]
FIG. 2. PDI of B. cereus spores incubated with different PS at a concentration of 100 µM (50 µM for TBO) for 3 h at 37°C. The spores were illuminated without a wash. The appropriate wavelength of light was used for each dye (see Materials and Methods), and the fluence rate was 200 mW/cm².

Phenothiazinium dyes as a class photoinactivate spores.
Since TBO had shown significant and surprising activity as anantimicrobial PS against B. cereus spores, we examined otherrelated members of the class of phenothiazinium dyes (see structuresin Fig. 1). MB has been used almost as often as TBO for antimicrobialPDT (30, 46, 48, 61). DMMB was found by O'Neill et al. to bethe most effective PS compared to other phenothiazinium dyestested against the gram-positive bacterium Streptococcus sanguis(34). NMBN was found to be the most active PS compared to otherphenothiazinium dyes tested against the gram-negative bacteriumYersinia enterocolitica (54). Because our first experimentshad shown that TBO appeared to bind to B. cereus spores, wewashed the spores after phenothiazinium dye incubation and resuspendedthem in PBS before PDI.

Figure 3 shows that there are three broad levels of effectivenessof the different phenothiazinium dyes tested (50 µM) inkilling B. cereus spores. MB is the least powerful, killingonly about 99% of spores after treatment at 40 J/cm²; DMMB hadan almost identical efficiency to that found with TBO (99.999%)after treatment with 40 J/cm². NMBN was the most powerful ofthe compounds we tested, leading to >99.999% killing withonly half the amount of light (20 J/cm²).

View larger version (27K):
[in this window]
[in a new window]
FIG. 3. PDI of B. cereus spores incubated with different phenothiazinium dyes at 50 µM for 3 h at 37°C. The spores were illuminated after a wash. The appropriate wavelength of light was used for each dye (see Materials and Methods), and the fluence rate was 200 mW/cm².

Differential susceptibility of Bacillus spores.
We next asked whether spores from other Bacillus species werealso susceptible to PDI with TBO. B. thuringiensis is a speciesthat is closely related to B. cereus (and B. anthracis) (22).B. subtilis and B. atrophaeus (formerly B. subtilis var. niger)are a pair of closely related species (3). B. megaterium haslarge cells both in the vegetative state and as spores (33,35). Again, these experiments were carried out with a wash afterincubation. There were large differences in susceptibility toTBO-mediated PDI between spores of different Bacillus species.In order to display these in Fig. 4, we used a series of TBOconcentrations in the incubation mixture and a constant fluenceof 40 J/cm² of 630-nm light. B. cereus and B. thuringiensiswere both highly susceptible to PDI, with a TBO concentrationof 50 µM leading to an almost complete loss of viability.In contrast, B. subtilis and B. atrophaeus were much less sensitiveand needed concentrations as high as 1.6 mM to achieve killingof >99.9% of cells. B. atrophaeus was significantly moresensitive than B. subtilis (P < 0.05 for comparison of slopesby nonlinear regression analysis). B. megaterium spores werecompletely resistant to PDI under these conditions.

View larger version (27K):
[in this window]
[in a new window]
FIG. 4. PDI of different Bacillus spores incubated with different concentrations of TBO for 3 h at 37°C. The spores were illuminated after a wash. The light used was 635 ± 15 nm, and 40 J/cm² was delivered at a fluence rate of 200 mW/cm². *, B. subtilis and B. atrophaeus had significantly different killing curves (P < 0.05 for comparison of slopes by nonlinear regression analysis).

Photoinactivation is better after a wash.
Most of the experiments described above were carried out afterthe spores were incubated with the dye and excess dye in solutionwas removed by pelleting the spores by centrifugation and resuspendingthem in PBS before illumination. This demonstrated that thedye actually bound to the spores, rather than generating reactiveoxygen species in solution that subsequently attacked the spores.We compared the effectiveness of this technique with the alternativemethod of leaving the dye in the incubation mixture. Figure5 (top panel) shows that in the case of B. cereus spores, wherewe only needed to use TBO concentrations of up to 50 µM,the washing out of excess dye gave a somewhat better rate ofkilling than leaving the dye in solution. In contrast, for B.atrophaeus (Fig. 5, bottom panel), where much higher TBO concentrationswere required, we found that when the dye was left in solution,killing reached a plateau with a value of only 80% at 200 µMTBO, while it continued to increase with increasing TBO concentrationswhen the spores were washed free from excess dye. We interpretedthese results by postulating a major absorbance of the lightby the dye in solution, thus significantly reducing the amountof light that can penetrate to the dye-containing spores insuspension. The absorbance of the B. atrophaeus dye mixturewas much higher than that of the B. cereus dye mixture due tothe higher concentrations of TBO required to kill B. atrophaeus,and therefore the optical screening effect of the excess dyewas more pronounced, leading to almost no additional killingof B. atrophaeus spores at concentrations of TBO above 100 µM.

View larger version (15K):
[in this window]
[in a new window]
FIG. 5. Comparison of PDI of spores with and without a wash. (Top) B. cereus spores were incubated with TBO at 50 µM for 3 h at 37°C. Suspensions were either washed or not and were illuminated with 635- ± 15-nm light at a fluence rate of 200 mW/cm². (Bottom) B. subtilis spores were incubated with TBO at different concentrations for 3 h at 37°C. Suspensions were either washed or not and were illuminated with 40 J/cm² of 635- ± 15-nm light at a fluence rate of 200 mW/cm².

Comparison of photosensitivity between spores and vegetative cells.
We compared the relative sensitivities to PDI of spores andvegetative cells from two Bacillus species (B. cereus and B.subtilis). Because there is such a large difference in the concentrationsof TBO needed to kill these four entities, we plotted the TBOconcentrations on a log scale (Fig. 6). The sensitivities ofvegetative cells from the two species were fairly similar, exceptthat B. cereus was killed more than B. subtilis at low TBO concentrations(up to 8 µM). There was a very large difference in susceptibilitybetween spores of B. cereus and spores of B. subtilis, whichwas similar to that found previously (Fig. 4) and meant thatthere was a much smaller difference in sensitivity between sporesand vegetative cells of B. cereus (the TBO concentration neededto kill spores was 3 to 4 times higher than that needed to killvegetative cells) than between spores and vegetative cells ofB. subtilis (>100 times the TBO concentration was neededto kill spores compared to vegetative cells).

View larger version (27K):
[in this window]
[in a new window]
FIG. 6. Comparison of PDI of spores and vegetative cells. B. cereus and B. subtilis spores and vegetative cells were incubated with TBO at different concentrations for 3 h at 37°C. Cells were washed and illuminated with 40 J/cm² of 635- ± 15-nm light at a fluence rate of 200 mW/cm².

Dye uptake by Bacillus spores.
The fact that the spores could be washed free of dye in theincubation mixture by centrifugation allowed us to quantifythe amount of dye taken up by each spore by previously establishedmethods of chemical extraction and fluorescence spectrophotometry.Figure 7 shows the relationship between the uptake of TBO byspores after incubation and increasing concentrations of dye.No detectable uptake could be measured with spores of B. megaterium.Although there was a significant difference between the uptakevalues of B. cereus and B. thuringiensis at high TBO concentrations(>100 µM), at lower concentrations where PDI is performed(50 µM) the uptakes were similar, which explains why B.cereus and B. thuringiensis showed similar susceptibilitiesto PDI (Fig. 4). Both B. subtilis and B. atrophaeus had muchlower uptakes, and in fact it is possible to estimate an uptakevalue in the region of 3 x 10⁷ molecules per spore being necessaryfor PDI to occur (see the dotted line in Fig. 7). However, althoughthere is clearly a correlation between the uptake of dye byspores and susceptibility to killing, we cannot exclude thatthere are also intrinsic differences in susceptibility betweendifferent Bacillus spores independent of dye uptake.

View larger version (33K):
[in this window]
[in a new window]
FIG. 7. Uptake of TBO by spores. Spores were incubated with TBO at different concentrations for 3 h at 37°C, washed, and dissolved in NaOH-SDS, and dye uptake was measured by spectrofluorimetry. Spore numbers were measured by plating, and a calibration curve gave moles of dye. Points are means of three determinations, and bars are standard errors of the means. The dotted line indicates the approximate level of dye uptake necessary for efficient PDI. *, significant differences (P < 0.01 by unpaired two-tailed Student's t test) between B. cereus and B. thuringiensis and between both of these and B. subtilis/B. atrophaeus.

DISCUSSION

Top

Discussion

This study has demonstrated PDI of Bacillus spores for the firsttime. We discovered that dyes from the phenothiazinium structuralgroup were the only compounds that were active among severalPS we tested that had previously been shown to be efficientagainst vegetative bacterial cells. Previously, Schafer et al.had shown that spores of B. subtilis were resistant to PDI byusing 2 µM of RB and 90 J/cm² visible light (40). Thephenothiazinium dyes bound to the spores and in fact were moreactive after excess dye was washed from the spore suspensions.This was likely due to the dye remaining in solution actingas an optical screen, thus preventing the light from reachingthe spores containing active bound dye. It was necessary toincubate the spores for 3 h with dyes at a temperature of 37°C.Shorter incubation times and lower incubation temperatures wereless effective (data not shown), suggesting that the dyes penetratedinto the spores by a process of passive diffusion. Diffusionis significantly time dependent and is enhanced at higher temperatures.A future study will systematically examine the effects of variousincubation times and temperatures on the process of spore PDI.The difficulty that certain exogenous molecules face in diffusinginto spores is due to the impermeable nature of the spore coatand is thought to be responsible for much of the resistanceof spores to biocidal chemicals (41).

Phenothiazinium dyes are small molecules that have an intrinsicmolecular cationic charge that allows them to bind to the generallyanionic outer membranes of bacteria and fungi (36, 53, 54).We investigated the possibility that small cationic PS as abroad class may therefore be effective against spores. However,we failed to demonstrate any PDI of spores using a tricationicporphyrin (another small molecule) that has been effectivelyused to mediate PDI of various bacterial (27) and fungal (43)cells. There is therefore clearly another molecular featureinvolved besides the cationic charge and small molecular size.This may be related to the ability of phenothiazinium dyes toform dimers and oligomers. Usacheva et al. (48-50) showed thatTBO was a better antimicrobial PS than MB partly because itwas able to self-associate into dimers that were supposed topreferentially interact with lipopolysaccharide and other componentsof bacterial membranes.

The differential sensitivity of spores from different Bacillusspecies is probably related to differences in spore structure.B. cereus and B. thuringiensis are closely related species froma family that also contains B. anthracis (23, 26) and threeother species (21). They are characterized by the presence ofan outer structure beyond the spore coat termed an exosporium.The exosporium of B. cereus contains two main layers: the outerlayer is made up of a nap of hairlike projections, while theinner layer has a hexagonally perforated surface pattern ofholes, and the intact basal membrane is about 19 nm thick (15).It is possible that the presence of the exosporium is involvedin the high dye accumulation found with B. cereus and B. thuringiensisspores and not found with nonexosporium-forming spores. If theexosporium traps more dye than the "naked" spore, then it mayallow more dye diffusion into the interior of the spore, wherefatal photodamage could occur. The nonsusceptibility of B. megateriumspores may simply be related to their much larger size. We havefound that Candida albicans cells are less sensitive to PDIthan both gram-positive and gram-negative bacterial cells, partlybecause the fungal cells are 20 to 30 times bigger than thebacterial cells (7). However, the complete lack of detectableTBO uptake by B. megaterium spores suggests that the dye eitherdoes not bind to the spore or cannot penetrate to a sufficientextent to carry out photodamage of essential components. Itis possible that differences in spore coat permeability areprimarily responsible for the marked differences in PDI betweenspecies. In particular, B. megaterium spores are thought tohave a lower permeability to small molecules, as described byGerhardt and coworkers (14, 25). Another distinguishing factorthat might explain the nonsusceptibility of B. megaterium sporesis the much more hydrophilic nature of their surfaces than thoseof spores of B. subtilis and B. cereus (56). Since phenothiaziniumdyes (in common with most PS) have a planar aromatic molecularstructure, they will have amphiphilic regions that can bindto hydrophobic surfaces despite the polar cationic charges thatare also on the molecule. If the dyes failed to bind to thespores, then the process of diffusion into the spore would bemuch reduced. Setlow and coworkers have demonstrated that B.subtilis spores are susceptible to agents that produce reactiveoxygen species. These include aqueous ozone (60), hydrogen peroxide(29), peroxynitrite (13), and the Fenton reagent, consistingof cuprous chloride and ascorbic acid (42). The fact that PDIproduces reactive oxygen species such as singlet oxygen, superoxide,and hydroxyl radicals is in agreement with the observed sporicidaleffects of oxidizing agents.

Future work will investigate the mechanisms of PDI of Bacillusspores and will also study spores of Clostridium species. Itis fairly well established that PDI of vegetative bacterialcells involves damage to the cytoplasmic membrane and leakageof essential cellular constituents into the medium. It may bepossible to identify which proteins in the spore coat or thespore core are damaged by the reactive oxygen species producedby PDI. Spores produced by mutant bacteria are known that havedeficiencies in the permeability barrier, and these may givesome useful insights into the mechanisms of phenothiaziniumdye uptake by spores.

ACKNOWLEDGMENTS

This work was funded by the Department of Defense’s Programto Develop Biomedical Applications of the Free Electron Laser(contract number N00014-94-1-0927) and by the National Instituteof Allergy and Infectious Diseases (grant number R01 AI050875).

We are grateful to Maleha Khan for technical assistance andto Tayyaba Hasan for support. We thank QLT Inc. for the giftof BPD and Saskia Lambrechts, Threes Smijs, and Ruud van derSteen for the gift of TriP(4). We thank George P. Tegos fora critical reading of the manuscript.

FOOTNOTES

* Corresponding author. Mailing address: Wellman Center for Photomedicine, Massachusetts General Hospital, BAR414, 40 Blossom Street, Boston, MA 02114. Phone: (617) 726-6182. Fax: (617) 726-8566. E-mail: hamblin{at}helix.mgh.harvard.edu.

REFERENCES

Top