Photodynamic Inactivation of Conidia of the Fungi Metarhizium anisopliae and Aspergillus nidulans with Methylene Blue and Toluidine Blue

Photodynamic Inactivation of Conidia of the Fungi Metarhizium anisopliaeand Aspergillus nidulans with Methylene Blue and Toluidine Blue

Fernanda P. Gonzales1, Sergio H. da Silva1, Donald. W. Roberts2 and Gilberto U. L. Braga*1

1Departamento de Analises Clınicas, Toxicologicas e Bromatologicas, Faculdade de CienciasFarmaceuticas de Ribeirao Preto, Universidade de Sao Paulo, Ribeirao Preto, SP, Brazil

2Department of Biology, Utah State University, Logan, UT

Received 22 May 2009, accepted 7 November 2009, DOI: 10.1111 ⁄ j.1751-1097.2009.00689.x

ABSTRACT

Antimicrobial photodynamic treatment (PDT) is a promisingmethod that can be used to control localized mycoses or kill fungiin the environment. A major objective of the current study was tocompare the conidial photosensitization of two fungal species(Metarhizium anisopliae and Aspergillus nidulans) with methy-lene blue (MB) and toluidine blue (TBO) under differentincubation and light conditions. Parameters examined weremedia, photosensitizer (PS) concentration and light source. PDTwith MB and TBO resulted in an incomplete inactivation of theconidia of both fungal species. Conidial inactivation reached upto 99.7%, but none of the treatments was sufficient to achieve a100% fungicidal effect using either MB or TBO. PDT delayedthe germination of the surviving conidia. Washing the conidia toremove unbound PS before light exposure drastically reduced thephotosensitization of A. nidulans. The reduction was muchsmaller in M. anisopliae conidia, indicating that the conidia ofthe two species interact differently with MB and TBO. Conidiaof green and yellow M. anisopliae mutants were less affected byPDT than mutants with white and violet conidia. In contrast towhat occurred in PBS, photosensitization of M. anisopliae andA. nidulans conidia was not observed when PDT was performedin potato dextrose media.

INTRODUCTION

Although yeasts of the genera Candida and Cryptococcuscontinue to be the main opportunistic fungi responsible forinvasive mycoses in humans, serious infections caused byAspergillus spp. and by other genera of filamentous fungi haveemerged all over the world. The primary reason is theincreased numbers of immunocompromised individuals (1).Additionally, the continuous and indiscriminate use of fungi-cides has favored the selection of fungal strains resistant tocurrently used fungicides (2). The selection of tolerant fungalstrains has occurred both in species pathogenic to humans andin phytopathogenic fungi.

Metarhizium anisopliae is a worldwide entomopathogenicdeuteromycete that produces green rod-shaped conidia. Thefungus has been used for decades in programs of agriculturalpest and disease-vector control in various countries (3).

Although M. anisopliae is not considered to be pathogenicfor humans or domestic animals, starting in the late 1990ssome rare cases of mycoses caused by the fungus in bothimmunocompromised and immunocompetent individuals havebeen reported (4). In none of these cases was the mycosisassociated with use of the fungus as a bioinsecticide.

The photodynamic inactivation of fungi is based on the useof a photosensitizer (PS) that binds to the wall or accumulatesinto the cell of the target microorganism. The PS is thenexposed to visible light of an appropriate wavelength in thepresence of oxygen, and this starts photochemical processesthat produce a series of reactive oxygen species (ROS) capableof damaging virtually all types of biomolecules and of killingthe microbial cells. The use of PS to kill fungal structures suchas conidia and hyphae and for the treatment of mycoses is arecent and promising application of photodynamic treatment(PDT) (5–9). PS such as methylene blue (MB) and toluidineblue (TBO) have been used to kill yeast (10–12) and filamen-tous fungus cells (13,14).

The physicochemical characteristics (i.e. chemical compo-nents, electrostatic charge, hydrophobicity) of the conidial walland surface differ greatly from those of vegetative yeast cellsand of filamentous-fungus hyphae. The walls of the conidia ofAspergillus nidulans and M. anisopliae are extremely hydro-phobic and have a negatively charged surface (15–17). Theouter surface of the conidium consists of a resilient layer ofwell-organized fascicles of rodlets composed of hydrophobin, afamily of small proteins that are amphiphiles with hydrophilicand hydrophobic parts and are among the most surface-activeproteins (16,18–20). Other molecules, such as sugars andlipids, are an integral part of the surface of M. anisopliae andA. nidulans conidia and can play a role in their physico-chemical properties (15,16). Both the chemical compositionand the structural organization of the wall vary widelyamong species, strains and fungal developmental stages(15,17,21,22).

Although the molecular mechanisms of the fungal PDT thatgive rise to the biological effects are largely unknown, severalattempts have been made to understand how different PSinteract with the fungal cells both before and during PDT andalso to determine the site of damage. The physicochemicalcharacteristics of the microbial cell surface influence theselectivity and efficacy of the various PS (7,8,23–28). Cell walland plasma membrane are negatively charged, which favorsthe binding of cationic PS like MB and TBO. TBO binds

*Corresponding author email: [email protected] (Gilberto U. L. Braga)! 2010TheAuthors. JournalCompilation.TheAmericanSociety ofPhotobiology 0031-8655/10

Photochemistry and Photobiology, 2010, 86: 653–661

653

instantaneously to polyphosphates localized outside theSaccharomyces fragilis plasma membrane after adding cellsto the dye solution (29). Ito (10) observed that TBO does notpenetrate the cells of Saccharomyces cerevisiae and attributedthe entire photodynamic activity of the dye to its actionstarting from the extracellular medium and ⁄ or from the outersurface of the cell. Smijs et al. (8) reported severe cell walldisruptions and deformation of both Trichophyton rubrummicroconidia and mycelium after lethal PDT with the cationicPS Sylsens B. This fungicidal effect was achieved after thebinding of this positively charged porphyrin to the negativelycharged outer wall of fungal microconidia and hyphae.

Information currently extant on photodynamic inactivationof filamentous fungi is very limited (5–9,13,14). The extensionof photosensitization studies to new fungus species, and toidentifying PS that could be used as fungicides, are importantfor the development of this technique. The objectives of thepresent study were: (1) to evaluate for the first time the PDT ofthe conidia of the fungusM. anisoplae with two relatively well-understood PS, MB and TBO, under different conditions(media, PS concentration and light source) and to comparethese results with those obtained with a model ascomycete,A. nidulans; and (2) to determine whether altered pigmentationof M. anisopliae conidia can influence susceptibility to PDT.

MATERIALS AND METHODSM. anisopliae and A. nidulans strains. M. anisopliae var. anisopliaestrain ARSEF 23 (wild-type strain with dark green conidia) and UV-Binduced color mutants of this strain, i.e. ARSEF 6849 (purple conidia),ARSEF 6994 (yellow conidia) and ARSEF 6998 (white conidia) wereobtained from USDA-ARSEF (U.S. Plant, Soil and NutritionLaboratory, Ithaca, NY). A. nidulans strain ATCC 10074 (wild strainwith dark green conidia) was obtained from ATCC (American TypeCulture Collection, Manassas, VA).

Production of conidia. The M. anisopliae and A. nidulans strainswere grown on potato dextrose agar medium (Acumedia Manufac-turers, Inc., Lansing, MI) supplemented with 1 g L)1 yeast extract(Technical; Difco Laboratories, Detroit, MI) (PDAY) at 28"C, for 12and 4 days, respectively. Conidia were suspended in Tween 80 solution(Sigma-Aldrich Chemie Gmbh, Germany) (0.01% vol ⁄ vol), conidialaggregates removed by filtration through a polycarbonate membrane(8 lm pore size; Nucleopore#, NJ), conidial concentration estimatedwith a hemocytometer and appropriate dilutions made with Tween 80solution.

Photosensitizers. MB (C16H18ClN3S) and TBO (C15 H16N3SCl)(Fig. 1) were obtained from Sigma-Aldrich, Inc. (St. Louis, MO).Stock solutions of MB and TBO were prepared with PBS, pH 7.4, atconcentrations 10-fold higher than the highest concentration used. Thesolutions were stored in the dark at )20"C for up to 2 weeks. Dilutionswere prepared with PBS, pH 7.4.

Visible light exposures. Light exposures were conducted in atemperature-controlled chamber (Nova Etica, Sao Paulo, Brazil).Light was provided by a 300 W halogen lamp (Osram, Brazil). Visiblelight irradiance (400–700 nm) was 50 W m)2, as measured with aUSB4000 spectroradiometer (Ocean Optics, Dunedin, FL).

Laser exposures. Light was provided by an Eagle diode laser(Quantum Tech, Brazil). Conidial suspensions were illuminated with20 J cm)2 of 675 nm light at an irradiance of 78 mW cm)2.

Conidial PDT. One hundred microliters of conidial suspension (106

conidia mL)1) and 900 lL of the PS solution in PBS (pH 7.4), or inpotato dextrose broth medium (Difco Laboratories) were placed in12 · 100 mm glass tubes (borosilicate; Normax, Portugal). The finalconcentrations of MB and TBO in the mixture were 0, 1, 13, 52, 100and 400 lg mL)1. The mixture was held in the dark for 30 min, at28"C, with shaking (50 rpm) (preincubation) before exposure to visiblelight (50 W m)2) for 30 or 60 min in the presence of dyes. ‘‘Darkcontrol’’ tubes were protected from light inside the chamber during the

exposures by wrapping them with aluminum foil. The temperatureinside the chamber was maintained at 20 ± 4"C. The temperatureinside the tubes was determined at 15 min intervals using a digitalthermometer with external readout (TFDA, Germany). In a typicalexperiment, the initial temperature of the conidial suspensions was24 ± 2"C and the temperatures increased up to 1 and 2"C (after30 min of exposure) and 2 and 4"C (after 60 min of exposure) in thedark control and in the light exposure tubes, respectively. Afterexposure, the tubes were centrifuged at 2000 g for 5 min, the conidiawashed once and resuspended in 1 mL PBS, pH 7.4. Four drops of theconidial suspension (20 lL each) were placed on the surface of 5 mL ofPDAY medium containing 0.008% (wt ⁄ vol) benomyl (Sigma) in petridishes (60 · 15 mm). This concentration of benomyl allowed germi-nation to be monitored for extended periods of time (30). The disheswere incubated in the dark at 28"C. Conidial germination was assessedat 400· magnification after incubation for 12 and 18 h for M. anisop-liae strains and after 8 and 14 h for A. nidulans. A total of 300 conidiaper treatment were evaluated. Additional experiments were performedusing PS concentrations of 13 and 53 lg mL)1 to evaluate conidialgermination at 24 h intervals up to 72 h. Conidia that did notgerminate within 72 h after PDT were considered inactivated by thetreatment.

Experiments in which the conidia were washed with PBS beforeexposure to light to wash out the unbound PS were also carried outwith a 100 lg mL)1 concentration of MB and TBO. In theseexperiments, the laser was used as a light source and exposure wascarried out on 24-well flat-bottomed plates (TPP, Switzerland).Conidial germination was evaluated up to 72 h after PDT. To confirmthat germination counts were comparable to colony-forming-unit(CFU) counts, CFU of the wild-type M. anisopliae (ARSEF 23) werecounted daily at 20· magnification for 12 days.

Statistical analysis. There were three independent experiments. Alinear mixed effects model was used to analyze the influence of thevariables (PS concentration, light dose and conidial color) on conidialphotosensitization. No data transformations were required for statis-tical analysis. All computations were done with PROC MIXED inSAS ⁄STAT version 9.0 (31). The comparison of treatment means wasperformed using Student’s t-tests. P < 0.05 was considered statisti-cally significant.

RESULTS

PDT with broad-spectrum visible light of M. anisopliaeand A. nidulans conidia in PBS

PDT of M. anisopliae conidia with MB and TBO. Figure 2a,bshows the effect of PDT with MB on M. anisopliae (ARSEF23) conidial germination. Conidia were preincubated in the

Figure 1. Chemical structures of methylene blue (a) and toluidine blue(b).

654 Fernanda P. Gonzales et al.

PS (1–400 lg mL)1) for 30 min and then exposed to visiblelight in the presence of the PS for 30 min (Fig. 2a) or60 min (Fig. 2b). Conidial germination was assessed at 12 h(data not shown) and 18 h (Fig. 2a,b) after PDT. Becausewe initially did not know whether the dyes would bind tothe M. anisopliae and A. nidulans conidia, we did not washthe conidia after incubation with the dyes. In the absence ofMB, exposure to light for 30 min (P = 0.675) (Fig. 2a) or60 min (P = 0.554) (Fig. 2b) did not delay conidial germi-nation. We observed a small delay in conidial germinationin the dark controls, but only at MB concentration of400 lg mL)1 (Fig. 2c). PDT for both 30 and 60 min delayedconidial germination at all MB concentrations (1–400lg mL)1) (P < 0.0001 for all concentrations) (Fig. 2a,b).The greatest effect was obtained at MB concentrations of13, 52 and 100 lg mL)1 after PDT for both 30 min (Fig. 2a)and 60 min (Fig. 2b). The lowest effect was obtained at theconcentrations of 1 and 400 lg mL)1 (Fig. 2a,b). In mosttreatments, germination counts were significantly higher at18 h than at 12 h after PDT (data not shown). The highestrecovery was observed at concentrations of 1 and400 lg mL)1. At some concentrations of the PS, the delayin germination caused by exposure for 60 min was higherthan that caused by exposure for 30 min. The greatestdifferences were observed at the concentrations of 13 (P =0.0097) and 400 lg mL)1 MB (P = 0.0003) (cf. Fig. 2awith 2b).

Additional experiments performed to evaluate conidialgermination at 24 h intervals up to 72 h showed that14.9 ± 7.1% and 15.6 ± 9.8% of the conidia germinated72 h after PDT with MB (60 min of exposure) at concentra-tions of 13 and 52 lg mL)1, respectively.

Figure 2c,d shows the effect of PDT with TBO on M. ani-sopliae (ARSEF 23) conidial germination. A delay in germi-nation was observed in the dark controls at TBOconcentrations >52 lg mL)1 (Fig. 2d). PDT with TBO forboth 30 and 60 min delayed conidial germination at all TBOconcentrations (1–400 lg mL)1) (P < 0.0001 for all concen-

trations). The greatest delays were observed at concentrationsof 13, 52 and 100 lg mL)1 light exposure for 30 (Fig. 2c) and60 min (Fig. 2d), and the delays at these three TBO concen-trations were significantly different from those with the otherTBO concentrations. Germination delays were lowest withTBO concentrations of 1 and 400 lg mL)1. Germination delaycaused by 60 min PDT was higher than that caused by 30 minPDT for all TBO concentrations (P < 0.001 for all concen-trations) (cf. Fig. 2c with 2d). For most treatments, germina-tion evaluated 18 h after photosensitization was higher thanafter 12 h (data not shown).

Additional experiments performed to evaluate conidialgermination up to 72 h at 24 h intervals showed that18.3 ± 0.9% and 27.8 ± 15.3% of the conidia germinated72 h after PDT (60 min of exposure) at concentrations of13 and 52 lg mL)1, respectively. Micrographs documentingthe effect of PDT with MB on conidial germination ofM. anisopliae (ARSEF 23) are shown in Fig. 3.

PDT of A. nidulans conidia with MB and TBO. Figure 4shows the effect of PDT with MB and TBO on A. nidulans(ATCC 10074) conidial germination. The effects of PDT withMB and TBO (13–400 lg mL)1) on the germination ofA. nidulans in general were very similar to those obtained forM. anisopliae. In the absence of PS, the conidia were notinactivated by exposure to light for 30 min or 60 min. In theabsence of light, MB did not delay germination at any of theconcentrations used (Fig. 4a,b). TBO was more toxic than MBfor conidia of A. nidulans. In the absence of light, TBOconcentrations higher than 52 lg mL)1 delayed germination ofthe conidia (Fig. 4c,d). PDT for both 30 and 60 min delayedconidial germination at all MB and TBO concentrations (13–400 lg mL)1) (P < 0.0001 for all MB concentrations andP < 0.001 for all TBO concentrations) (Fig. 4).

Conidial recovering after PDT was higher in A. nidulansthan in M. anisopliae. Additional experiments performed toevaluate conidial germination up to 72 h showed that62.4 ± 7.6% and 44.0 ± 13.5% of the conidia germinated

Figure 2. Effects of photodynamic treatment with methylene blue (MB) or toluidine blue (TBO) on conidial germination ofMetarhizium anisopliae(ARSEF 23). Conidia were preincubated in MB (a, b) or TBO (c, d) for 30 min and then exposed to broad-spectrum visible light (50 W m)2) for30 min (a, c) or 60 min (b, d) in the presence of the photosensitizer. Germination was evaluated after 18 h incubation.

Photochemistry and Photobiology, 2010, 86 655

72 h after PDT with MB (60 min of exposure) at concentra-tions of 13 and 52 lg mL)1, respectively, and that94.8 ± 3.1% and 70.2 ± 0.2% of the conidia germinated72 h after PDT with TBO at concentrations of 13 and52 lg mL)1, respectively.

PDT of wild and color-mutant conidia of M. anisopliae withMB and TBO. The effects of sublethal PDT doses withMB andTBO (1 lg mL)1) on conidial germination of wild-type (greenconidia) and color mutants (yellow, purple and white conidia)are presented in Fig. 5. Exposure to light alone did not delay thegermination of the wild-type or color-mutant conidia (data notshown). Exposure to MB alone delayed the germination of asmall fraction of the conidia of the purplemutant (P < 0.0001),but not the germination of the wild-type strain or the white andyellow mutants (cf. Fig. 5a with 5b). PDT with MB delayed thegermination of both the wild-type conidia and the conidia of allcolored mutants (P < 0.0001 for all treatments). The highestdelay was observed in white and purple conidia, with no

significant difference between them (P = 0.835) and the lowestdelay was observed in green and yellow conidia, with nosignificant difference between them (P = 0.896).

In the absence of light, TBO (1 lg mL)1) did not delay thegermination of wild-type or any color-mutant conidia. PDTwith TBO delayed both wild-type and all color-mutant conidialgermination (P < 0.0001 for all treatments) (cf. Fig. 5c with5d). The delay was highest for purple conidia and lowest foryellow conidia. The effect of PDT with TBO was lower thanwith MB for both green and color-mutant conidia. Thegreatest difference occurred for white conidia, which presentedless delay in germination and more recovery when photosen-sitized with TBO than with MB (cf. Fig. 5b with 5d).

PDT of M. anisopliae conidia in PDB medium. In contrast toPDT in buffer (PBS), none of the concentrations (13–400 lg mL)1) of MB or TBO dissolved in culture medium(PDB) delayed conidial germination of M. anisopliae orA. nidulans (data not shown).

Figure 3. Effect of photodynamic treatment (PDT) with methylene blue (MB) on conidial germination of Metarhizium anisopliae (ARSEF 23)(400· magnification). (a) Conidia not treated with MB and not exposed to light. (b) Conidia not treated with MB and exposed to broad-spectrumvisible light (50 W m)2, 60 min). (c) Conidia treated with MB (13 lg mL)1) and not exposed to light. (d–f) Conidia treated with MB and exposedto light. Bar = 80 lm. Pictures were taken 24 h after PDT (a, b, c, and d), 48 h after PDT (e) and 72 h after PDT (f).

656 Fernanda P. Gonzales et al.

PDT with laser of M. anisopliae and A. nidulans conidia in PBS

Comparison of PDT between conidia washed and unwashedbefore exposure to light. Conidia were treated with MB orTBO and then washed by centrifugation to remove theunbound PS before exposing the conidia to light, and theefficacy of photosensitization of these conidia was compared tothat obtained in experiments with no washing. Washing the

conidia drastically reduced the effect of PDT on A. nidulans byboth MB and TBO (Fig. 6). The marked reduction observedafter washing indicates that large part of the photodynamicactivity may be due to reactive species from the unbound dyein the extracellular medium. Washing had a much smallereffect on M. anisopliae conidia, indicating that the interactionsof MB and TBO with the conidia occur in different manners inthe two species (Fig. 6). Conidial germination evaluated up to

Figure 4. Effects of photodynamic treatment with methylene blue (MB) or toluidine blue (TBO) on conidial germination of Aspergillus nidulans(ATCC 10074). Conidia were preincubated in MB (a, b) or TBO (c, d) for 30 min and then exposed to broad-spectrum visible light (50 W m)2) for30 min (a, c) or 60 min (b, d) in the presence of the photosensitizer. Germination was evaluated after 18 h incubation.

Figure 5. Effects of photodynamic treatment with methylene blue (MB; 1 lg mL)1) (a, b) and toluidine blue (TBO) (c, d) on the germination ofwild-type (green) and color-mutant conidia of Metarhizium anisopliae strain ARSEF 23. The conidia were preincubated in MB for 30 min and thenexposed to broad-spectrum visible light (50 W m)2, 60 min) in the presence of the PS. Germination was evaluated after incubation for 12 h (a, c)and 18 h (b, d).

Photochemistry and Photobiology, 2010, 86 657

72 h after PDT with MB indicated that at the higherphotosensitization treatments (nonwashed conidia of bothspecies) conidial germination was 0.3 ± 0.6% for M. anisop-liae and 6 ± 2% for A. nidulans 72 h after PDT.

Determination of survival by CFU counts (rather than bygermination counts) revealed very low conidial survival. Thesurvival ofM. anisopliae conidia (ARSEF 23) treated with MB(100 lg mL)1) and light (laser) and nonwashed (the germina-tion data of this experiment are shown in Fig. 7) was0.8 ± 0.2%. The plates were examined each day for 12 days.No new colonies were observed after 4 days. It should bepointed out that despite the high inactivation levels achieved insome treatments none of the treatments with MB or TBOkilled 100% of conidia of either species. Surviving conidiagerminated and formed new colonies (Fig. 8).

DISCUSSION

Although extremely promising, antimicrobial photodynamictherapy is still in development (9,32,33). Most PDT studieshave been conducted in vitro. Data are limited regarding thein vivo use of PDT for the treatment of mycoses in animalmodels and in patients (11,32–35). To date, no PS has been

Figure 6. Comparison of photodynamic treatment of conidia with andwithout a buffer wash before light exposure. The conidia of Meta-rhizium anisopliae (a) and Aspergillus nidulans (b) were preincubatedwith methylene blue (MB) or toluidine blue (TBO) for 30 min.Conidial suspensions were either washed or not washed with PBS andwere illuminated with 20 J cm)2 of 675 nm light at an irradiance of78 mW cm)2. Light was provided by an Eagle diode laser (QuantumTech, Brazil). Germination was evaluated after incubation for 12 h (a)and 18 h (b).

Figure 7. Comparison of photodynamic treatment of conidia with andwithout a buffer wash before light exposure. The conidia of Meta-rhizium anisopliae (a) and Aspergillus nidulans (b) were preincubatedwith methylene blue for 30 min. Conidial suspensions were eitherwashed or not washed with PBS and were illuminated with 20 J cm)2

of 675 nm light at 78 mW cm)2. Light was provided by an Eagle diodelaser. Germination was evaluated up to 48 h after PDT.

Figure 8. Effect of conidial photodynamic treatment with methyleneblue (MB) on colony development of (a) Metarhizium anisopliae and(b) Aspergillus nidulans. The conidia were preincubated with MB(100 lg mL)1) for 30 min and were illuminated with 20 J cm)2 of675 nm light at 78 mW cm)2. Light was provided by an Eagle diodelaser. After PDT, conidia (10 lL, 105 mL)1) were spotted on potatodextrose agar and plates were incubated at 28"C.

658 Fernanda P. Gonzales et al.

licensed for use as an antimicrobial agent, and the methods forthe distribution of these drugs are still being established (36).

Some studies have shown that PDT can be used to killfilamentous fungi that can cause superficial and ⁄ or invasivemycoses in humans, e.g. T. rubrum (5–9,37), Aspergillusfumigatus (14).

We evaluated the effect of conidial photosensitization bycytological monitoring of germination. This method has theadvantage of permitting observation of the cytological effectsof photosensitization and of the recovery of conidia exposedto sublethal PDT. On the other hand, it has the disadvantageof not permitting the detection of several logs of killing,which is possible in experiments based on dilution, platingand CFU counts. The present results showed that it ispossible to inactivate conidia of M. anisopliae and A. nidulanswith MB and TBO. The effect of PDT with both MB andTBO was dependent on the concentration of the PS, dose,light source and the fungal strain ⁄ species. PTD with visiblelight was less effective than with laser. The use of a laser with675 nm wavelength, which is close to the absorption peaks ofMB (kmax 665) and TBO (kmax 636), improved the effect ofPDT. Additionally, this wavelength is weakly absorbed bybiomolecules and has a higher penetration in cell suspensionsand in tissues than white light. In the most effectivetreatments, conidial inactivation was >99%. These valueswere confirmed by CFU counting experiments with evalua-tions up to 12 days after PDT. Although none of thetreatments used killed 100% of the conidia, the possibility ofobtaining a successful PDT with MB and TBO should not beruled out. In the present study, we did not investigate otherkey factors involved in PDT efficacy such as pH and ionstrength of incubation media (7). PDT efficacy with TBO onS. cerevisiae is higher at pH 7.8 than at pH 5.8 (10).Literature regarding the PDT of conidia with phenothiazini-um dyes is extremely limited. The choice of the MB and TBOconcentrations used in the present experiment was based onprevious results of photosensitization of A. nidulans (13).Wild-type and mutant conidia of this fungus were photoin-activated by 12.5 lg mL)1 of MB and TBO. Conidia ofNeurospora crassa were completely photoinactivated by30 lMM TBO (60 min, 55 000 lux), while 10 lMM caused inacti-vation of about 80% and concentrations of less than 3 lMM

did not inactivate conidia (38). We observed with M. anisop-liae and A. nidulans that increased concentrations of the PS(MB and TBO) increased the efficiency of PDT up to anoptimum concentration, with an abrupt reduction in theefficiency of the process occurring thereafter. This type ofreduction, which has been observed in other biologicalsystems and with other PS, can be attributed mainly to theblockage of light caused by an excessive amount of the PS(23,25).

Washing the conidia before exposure to light drasticallyreduced the effect of PDT on A. nidulans by both MB andTBO. Washing had a much smaller effect on M. anisopliaeconidia. This indicates that despite having conidia withhydrophobic surface and negatively charged (which favorsbinding with positively charged PS), M. anisopliae andA. nidulans conidia interact differently with both PS. Addi-tional experiments are necessary to clarify the molecularaspects of M. anisopliae and A. nidulans conidial interactionswith MB and TBO.

When the conidia were washed before exposure, the effectof PDT with MB was higher than by TBO inM. anisopliae andlower in A. nidulans. The greater efficacy of TBO compared toMB has been previously observed in wild conidia of A. nidu-lans (13) and in gram-negative bacteria (24,27). In the lattercase, the difference in photobactericidal efficacy was attrib-uted, at least in part, to the differential binding of the PS tospecific components of the outer membrane, e.g. TB interactswith lipopolysaccharides more significantly than with MB(24,27).

Wild-type conidia with green pigmentation were moretolerant than white and purple mutants to PDT with MBand TBO. The determination of the absorption spectra of thepigments present in the wild-type and mutants would be ofhelp to understand how the different pigments interact withlight and influence conidial photosensitization. Unfortunately,despite several attempts, we were unable to extract thepigments of M. anisopliae conidia. Optical density at 665 nm(maximum absorbance of MB) of the suspension of whiteconidia was lower than that of the yellow and violet mutantsand of the wild-type strain (data not shown). The greaterpenetration of light through the suspension may have been oneof the factors responsible, at least in part, for the higher effectof PDT on the albino mutant.

The chemical nature of the pigments responsible for thegreen color of M. anisopliae conidia is unknown. The greencolor (similar to that of M. anisopliae) of conidia of otherascomycete species such as A. fumigatus and A. nidulans is dueto the deposition of pigments in the cell wall (39–41). Whitemutants of A. fumigatus and A. nidulans present changes insuperficial morphology, physicochemical characteristics andwall composition, and they are more sensitive to the stressesinduced by chemical and physical agents (21,40). The conidialwall of the white mutant strain of A. nidulans lacked melaninand a-1,3-glucan linkages, and contained twice as muchgalactose as the wild-type strain (21). M. anisopliae mutantsfor conidial color had marked morphological changes and hadincreased sensitivity to temperature and to UV radiation(30,42). White A. fumigatus conidia were killed more easily byoxidants and were more easily damaged by human monocytesin vitro than green wild-type conidia (39).

Direct interaction with incident light is not the only waypigments can influence photosensitization. Wild-type conidiaof A. nidulans were more tolerant to PDT both with MB andTBO than color mutants (13). As observed with M. anisopliaein the present study, the tolerance of the yellow mutant tophotoinactivation by MB and TB was higher than that of thewhite mutant, but similar to that of wild-type conidia. Thisseems to indicate that the yellow pigment is also able to confersome protection against photo-oxidation. Fuchs et al. (26)showed that a mutant of Cryptococcus neoformans with acompromised cell wall and less 1,3-b-DD glucan linkages thanthe wild-type was more susceptible to photodynamic inactiva-tion than the wild type. Taken together, the data indicate thatthe impairment of the wall integrity due to the absence ofpigments or of structural polysaccharides influences thesusceptibility of vegetative yeast cells and of conidia to PDT.

The presence of antioxidant pigments such as carotenoidsincreases the conidial tolerance of other species to PDT. Forexample, albino mutants of N. crassa were less tolerant toPDT with MB than the wild-type strain (43). In previous

Photochemistry and Photobiology, 2010, 86 659

experiments we did not detect any carotenoids (i.e. b-carotene,astaxanthin, lutein, lycopene) in conidia of the ARSEF 23strain (C. A. Carollo, A. L. A. Calil, L. A. Schiave,T. Guaratini, D. W. Roberts, N. P. Lopes and G. U. L.Braga, unpublished).

In contrast to what occurred in PBS, photosensitizationof M. anisopliae and A. nidulans conidia was not observedwhen PDT was performed in PDB medium. As the PDBmedium presented minimal absorption in the red region(data not shown), which includes wavelengths capable ofactivating MB and TBO, it is unlikely that the reduction inconidial photosensitization was due to a decrease in incidentlight. The most likely possibility is that the high concentra-tion of organic matter present in the PDB mediuminactivated the cytotoxic species produced in the extracellu-lar medium during PDT or compete with the microbial cellsfor dye binding (44). Shimizu et al. (32) observed thatbovine albumin protected N. crassa conidia against ROSformed during photosensitization with TBO. Albuminalso inhibited the photoinactivation of Staphylococcusaureus, Pseudomonas aeruginosa and Candida albicans withporphyrin (45).

The findings presented here, in our opinion, add additionalsupport to the concept that the potential of PDT for theselective inactivation of microorganisms in both therapeuticand nontherapeutic applications is a promising antimicrobialapproach that warrants further, intensified research.

Acknowledgements—We are grateful to Dr. Antonio Claudio Tedescofor supplying the Eagle diode laser. This work was supported by grants03 ⁄ 07702-9 from the State of Sao Paulo Research Foundation(FAPESP) and 47.6990 ⁄ 2004-1 from the Brazilian National Councilfor Scientific and Technological Development (CNPq). We sincerelythank CNPq for an MS fellowship to F.P.G.

REFERENCES1. Espinel-Ingroff, A., A. Fothergill, M. Ghannoum, E. Manavathu,

L. Ostrosky-Zeichner, M. Pfaller, M. Rinaldi, W. Schell and T.Walsh (2005) Quality control and reference guidelines for CSLIbroth microdilution susceptibility method (M-38-A Document)for amphotericin B, itraconazole, posaconazole and voriconazole.J. Clin. Microbiol. 43, 5243–5246.

2. Anderson, J. B. (2005) Evolution of antifungal-drug resistance:Mechanisms and pathogen fitness. Nat. Rev. Microbiol. 3, 547–556.

3. Roberts, D. W. and R. J. St Leger (2004) Metarhizium spp., cos-mopolitan insect-pathogenic fungi: Mycological aspects. Adv.Appl. Microbiol. 54, 1–70.

4. Osorio, S., R. de la Camara, M. C. Monteserin, R. Granados, F.Ona, J. L. Rodriguez-Tudela and M. Cuenca-Estrella (2007)Recurrent disseminated skin lesions due to Metarhizium anisopliaein an adult patient with acute myelogenous leukemia. J. Clin.Microbiol. 45, 651–655.

5. Smijs, T. G. and H. J. Schuitmaker (2003) Photodynamic inacti-vation of the dermatophyte Trichophyton rubrum. Photochem.Photobiol. 77, 556–560.

6. Smijs, T. G., R. N. van der Hass, J. Lugtenburg, H. Liu, R. L.de Jong and H. J. Schuitmaker (2004) Photodynamic treatmentof the dermatophyte Trichophyton rubrum and its microconidiawith porphyrin photosensitizers. Photochem. Photobiol. 80, 197–202.

7. Smijs, T. G. M., J. A. Bouwstra, M. Talebi and S. Pavel (2007)Investigation of conditions involved in the susceptibility of thedermatophyte Trichophyton rubrum to photodynamic treatment.J. Antimicrob. Chemother. 60, 750–759.

8. Smijs, T. G. M., A. A. Mulder, S. Pavel, J. J. M. Onderwater, H.K. Koerten and J. A. Bouwstra (2008) Morphological changes ofthe dermatophyte Trichophyton rubrum after photodynamictreatment: A scanning electron microscopy study.Med. Mycol. 46,315–325.

9. Smijs, T. G. M., S. Pavel, M. Talebi and J. A. Bouwstra (2009)Preclinical studies with 5,10,15-tris(4-methylpyridinium)-20-phe-nyl-[21H,23H]-porphine trichloride for the photodynamic treat-ment of superficial mycoses caused by Trichophyton rubrum.Photochem. Photobiol. 85, 733–739.

10. Ito, T. (1977) Toluidine blue: The mode of photodynamic actionin yeast cells. Photochem. Photobiol. 25, 47–53.

11. Teichert, M. C., J. W. Jones, M. N. Usacheva and M. A. Biel(2002) Treatment of oral candidiasis with methylene blue-medi-ated photodynamic therapy in immunodeficient murine model.Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 93, 155–160.

12. Souza, S. C., J. C. Junqueira, I. Balducci, C. Y. Koga-Ito, E.Munin and A. O. C. Jorge (2006) Photosensitization of differentCandida species by low power laser light. J. Photochem. Photobiol.B, Biol. 83, 34–38.

13. Al-Rubeai, M. and M. El-Hassi (1986) Inactivation of wild typeand mutant Aspergillus nidulans conidia by far-UV, near-UV,visible and sun lights. Environ. Exp. Bot. 32, 243–252.

14. Friedberg, J. S., C. Skema, E. D. Baum, J. Burdick, S. A.Vinogradov, D. F. Wilson, A. D. Horan and I. Nachamkin (2001)In vitro effects of photodynamic therapy on Aspergillus fumigatus.J. Antimicrob. Chemother. 48, 105–107.

15. Boucias, D. G., J. C. Pendland and J. P. Latge (1988) Nonspecificfactors involved in attachment of entomopathogenic deuteromy-cetes to host insect cuticle.Appl. Environ.Microbiol. 54, 1795–1805.

16. Girardin, H., S. Paris, J. Rault, M.-N. Bellon-Fontaine and J.-P.Latge (1999) The role of the rodlet structure on the physico-chemical properties of Aspergillus conidia. Lett. Appl. Microbiol.29, 364–369.

17. Shah, F. A., N. Allen, C. J. Wright and T. M. Butt (2007)Repeated in vitro subculturing alters spore surface properties andvirulence of Metarhizium anisopliae. FEMS Microbiol. Lett. 276,60–66.

18. Paris, S., J.-P. Debeaupuis, R. Crameri, M. Carey, F. Charles, M.C. Prevost, C. Schmitt, B. Philippe and J. P. Latge (2003) Conidialhydrophobins of Aspergillus fumigatus. Appl. Environ. Microbiol.69, 1581–1588.

19. Wosten, H. A. B. (2001) Hydrophobins: Multipurpose proteins.Annu. Rev. Microbiol. 55, 625–646.

20. Linder, M. B., G. R. Szilvay, T. Nakari-Setala and M. E. Penttila(2005) Hydrophobins: The protein-amphiphiles of filamentousfungi. FEMS Microbiol. Rev. 29, 877–896.

21. Claverie-Martin, F., M. R. Diaz-Torres and M. J. Geoghegan(1988) Chemical composition and ultrastructure of wild-type andwhite mutant Aspergillus nidulans conidial walls. Curr. Microbiol.16, 281–287.

22. Dague, E., A. Delcorte, J.-P. Latge and Y. F. Dufrene (2008)Combined use of atomic force microscopy, X-ray photoelectronspectroscopy, and secondary ion mass spectrometry for cell sur-face analysis. Langmuir 24, 2955–2959.

23. Usacheva, M. N., M. C. Teichert and M. A. Biel (2001) Com-parison of the methylene blue and toluidine blue photobactericidalefficacy against gram-positive and gram-negative microorganisms.Lasers Surg. Med. 29, 165–173.

24. Usacheva, M. N., M. C. Teichert and M. A. Biel (2003) Theinteraction of lipopolysaccharides with phenotiazine dyes. LasersSurg. Med. 33, 311–319.

25. Demidova, T. N. and M. R. Hamblin (2005) Photodynamicinactivation of Bacillus spores, mediated by phenothiazinium dyes.Appl. Environ. Microbiol. 71, 6918–6925.

26. Fuchs, B. B., G. P. Tegos, M. R. Hamblin and E. Mylonakis(2007) Susceptibility of Cryptococcus neoformans to photodynamicinactivation is associated with cell wall integrity. Antimicrob.Agents Chemother. 51, 2929–2936.

27. Usacheva, M. N., M. C. Teichert, Y. M. Usachev, C. E. Sievertand M. A. Biel (2008) Interaction of the photobactericides meth-ylene blue and toluidine blue with a fluorophore in Pseudomonasaeruginosa cells. Lasers Surg. Med. 40, 55–61.

660 Fernanda P. Gonzales et al.

28. Paardekooper, M., P. J. A. Van den Broek, A. W. De Bruijne, J.G. R. Elferink, T. M. A. R. Dubbelman and J. Van Steveninck(1992) Photodynamic treatment of yeast cells with the dye tolui-dine blue: All-or-none loss of plasma membrane barrier proper-ties. Biochim. Biophys. Acta 1108, 86–90.

29. Tijssen, J. P. F., H. W. Beekes and J. Van Steveninck (1981)Localization of polyphosphates at the outside of the yeast cellplasma membrane. Biochim. Biophys. Acta 649, 529–532.

30. Braga, G. U. L., D. E. N. Rangel, S. D. Flint, A. J. Anderson andD. W. Roberts (2006) Conidial pigmentation is important to tol-erance against solar-simulated radiation in the entomopathogenicfungus Metarhizium anisopliae. Photochem. Photobiol. 82, 418–422.

31. Statistical Analysis Systems Institute (2002) SAS ⁄STAT Version9.0. SAS Institute Inc., Cary, NC.

32. Zeina, B., J. Greenman, W. M. Purcell and B. Das (2001) Killingof cutaneous microbial species by photodynamic therapy. Br. J.Dermatol. 144, 274–278.

33. Hamblin, M. R. and T. Hassan (2004) Photodynamic therapy: Anew antimicrobial approach to infectious disease? Photochem.Photobiol. Sci. 3, 436–450.

34. Propst, C. and L. Lubin (1978) In vitro and in vivo photosensi-tized inactivation of dermatophyte fungi by heterocyclic dyes.Infect. Immun. 20, 136–141.

35. Komerik, N. and A. J. Macrobert (2006) Photodynamic therapyas an alternative antimicrobial modality for oral infections.J. Environ. Pathol. Toxicol. Oncol. 25, 487–504.

36. Harris, F., L. K. Chatfdield and D. A. Phoenix (2006) Phenothi-azinium based photosensitizers-photodynamic agents with amultiplicity of cellular targets and clinical applications. Curr. DrugTargets. 6, 615–627.

37. Calzavara-Pinton, P. G., M. Venturini and R. Sala (2005)A comprehensive overview of photodynamic therapy in the

treatment of superficial fungal infections of the skin. Photochem.Photobiol. 78, 1–6.

38. Shimizu, M., T. Egashira and U. Takahama (1979) Inactivation ofNeurospora crassa conidia by singlet molecular oxygen generatedby a photosensitized reaction. J. Bacteriol. 138, 293–296.

39. Jahn, B., A. Koch, A. Schmidt, G. Wanner, H. Gehringer, S.Bhakdi and A. Brakhage (1997) Isolation and characterization ofa pigmentless-conidium mutant of Aspergillus fumigatus withaltered conidial surface and reduced virulence. Infect. Immun. 65,5110–5117.

40. Goncalves, C. R. and S. R. Pombeiro-Sponchiado (2005) Anti-oxidant activity of the melanin pigment extracted from Aspergillusnidulans. Biol. Pharm. Bull. 28, 1129–1131.

41. Tsai, H.-F., Y. C. Chang, R. G. Washburn, M. H. Wheeler and K.J. Kwon-Chung (1998) The developmentally regulated alb1 geneof Aspergillus fumigatus: Its role in modulation of conidial mor-phology and virulence. J. Bacteriol. 180, 3031–3038.

42. Rangel, D. E. N., G. U. L. Braga, A. J. Anderson and D. W.Roberts (2005) Variability in conidial thermotolerance of Meta-rhizium anisopliae isolates from different geographic origins.J. Invertebr. Pathol. 88, 116–125.

43. Blanc, P. L., R. W. Tuveson andM. L. Sargent (1976) Inactivationof carotenoid-producing and albino strains of Neurospora crassaby visible light, blacklight, and ultraviolet radiation. J. Bacteriol.125, 616–625.

44. Moan, J., C. Rimington and A. Western (1985) The binding ofdihematoporphyrin ether (photofrin II) to human serum albumin.Clin. Chim. Acta 145, 227–236.

45. Lambrechts, S. A., M. C. Aalders, F. D. Verbraak, J. W.Lagerberg, J. B. Dankert and J. J. Schuitmaker (2005) Effectof albumin on the photodynamic inactivation of microorganismsby a cationic porphyrin. J. Photochem. Photobiol. B, Biol. 79, 51–57.

Photochemistry and Photobiology, 2010, 86 661