The first phenothiazinium dyes were synthetized by the end of the 19th century in Germany in the context of a booming German textile industry. Hereby, Alizarin, which was developed in 1868 by Carl Graebe and Carl Liebermann, worked as some kind of precursor, when further research in the purpose of its manufacture and commercialization led to the development of new dyes (López-Muñoz et al., 2005). One of these new dyes was so-called “Methylene Blue (MB)”—the first phenothiazinium dye—whose synthesis protocol was patented by Heinrich Caro in the 1870s (Caro, 1878).

Besides other applications, e.g., anti-malarial drugs or antipsychotic agents, phenothiazinium derivatives have been employed as PS due to their strong absorption in the red spectral region (ca. 600–680 nm) (Felgenträger et al., 2013). Phenothiazinium dyes are single positively charged and composed of a three-ring π-system with attached auxochromic side groups, whereby their singlet oxygen quantum yield Φ_Δ is around 0.5, thus acting according to type I and type II mechanisms (Wilkinson et al., 1993). Examples for phenothiazinium dyes, which have been tested for inactivation of biofilms in vitro, are foresaid MB [3,7-bis(dimethylamino)-phenothiazin-5-ium chloride] or Toluidine Blue [TBO; 3-amino-7-(dimethylamino)-2-methyl-phenothiazin-5-ium chloride] (see Figure 2).

Using MB, Fontana et al. investigated its effect on ex vivo polymicrobial biofilms cultivated from dental plaque samples taken from patients with chronic periodontitis. After a cultivation period of 7 days, the biofilms were incubated with MB in concentrations of 25 or 50 μg/ml for 5 min and illuminated with a diode laser (1 W; 665 nm) for another 5 min, whereby a light dose of 30 J/cm² was applied, which resulted in a maximal inactivation rate of 32 % CFU only. The authors concluded that this low susceptibility of complex dental biofilms toward aPDT may be overcome by novel delivery and targeting approaches (Fontana et al., 2009).

For testing the efficacy of MB on yeast biofilms, Rossoni et al. (2014) cultured biofilms of Candida albicans serotype A and B strains for 48 h. After that period, biofilms were incubated with MB at a concentration of 300 μ M for 5 min and subsequently irradiated with a gallium-aluminum-arsenide (GaAlAs) laser (35 mW; 660 nm) for 285 s, applying a light dose of 26.3 J/cm². This resulted in reductions of less than 1 log₁₀ (64%) for serotype A biofilms, whereas for serotype B biofilms there was a reduction of more than 2 log₁₀. The authors explained these distinct inactivation rates of Candida albicans serotype A and B biofilms by the differences in cell wall mannan structure between both serotypes. Furthermore, this difference in sensitivity to aPDT may be due to ultrastructural differences in EPS composition (Rossoni et al., 2014). However, in a comment to this study Mariusz Grinholc opposed that (i) only single reference strains of both serotypes and (ii) only one PS were tested, wherefore it may not be possible to conclude a general serotype-dependent sensitivity toward aPDT in cells of Candida albicans (Grinholc, 2014).

Employing TBO as a PS, the Brazilian group of Iriana C. J. Zanin conducted a number of studies (Zanin et al., 2005, 2006; Teixeira et al., 2012): In 2005 they examined its antimicrobial effect in combination with either a helium/neon (HeNe) gas laser (32 mW; 632.8 nm) or a light-emitting diode (LED; 32 mW; 620–660 nm with a maximum at 638.8 nm) on Streptococcus mutans biofilms grown on hydroxyapatite discs in a constant-depth film fermentor for 3, 7, or 10 days. After the respective culture period biofilms were incubated with TBO (100 mg/l) for 5 min and afterwards irradiated for 5, 15, or 30 min with HeNe laser or LED light (light doses of 49, 147, or 294 J/cm²). This resulted in reductions in viability of CFU of 2 to 5 log₁₀ steps depending on biofilm age, light source and irradiation period (Zanin et al., 2005). One year later, they tested the effect of TBO and LED light on monospecies biofilms of Streptococcus mutans, Streptococcus sobrinus, and Streptococcus sanguinis. Here, biofilms were cultivated for 5 days on enamel slabs. After incubation with TBO (100 mg/l) for 5 min, biofilms were exposed to LED light (32 mW; 620–660 nm, maximum: 638.8 nm) for 7 min (light dose: 85.7 J/cm²). Results showed inactivation rates of approximately 1 log₁₀ step for Streptococcus mutans and Streptococcus sobrinus biofilms and more than 3 log₁₀ for Streptococcus sanguinis biofilms. The authors tried to explain these varying results by the ability of mutans streptococci like Streptococcus mutans and Streptococcus sobrinus to produce EPS to a greater extent than Streptococcus sanguinis, which is not among mutans streptococci (Zanin et al., 2006). In 2012, the same group published a combined in vitro and in situ report on the photodynamic effect of TBO: (i) biofilms of Streptococcus mutans were grown on hydroxylapatite discs for 5 days, (ii) volunteers wore intraoral devices with enamel slabs for 7 days under cariogenic challenge by dropping a sucrose solution 8 times per day. After the respective culture period biofilms were incubated with TBO (100 mg/l) for 5 min and irradiated with a LED light source (40 mW, 620–660 nm, maximum: 638.8 nm) for 15 min (light dose 55 J/cm²). For in vitro biofilms CFU were reduced by more than 5 log₁₀ steps after aPDT, whereas for in situ biofilms reduction was less than 1 log₁₀ step, when the numbers of CFU of Streptococcus mutans and total streptococci were examined. These distinct results were explained by the authors due to the thickness of the referring biofilms with in situ biofilms (1000 μm) being approximately tenfold as thick as in vitro biofilms (100 μm); therefore, they concluded that this problem could be solved e.g., by developing a PS being able to penetrate through the EPS (Teixeira et al., 2012).

As MB and TBO are designated to be potential efflux pump substrates in a variety of microbial species (Tegos and Hamblin, 2006; Prates et al., 2011) and due to the enhancing effect of efflux pump inhibitors on photodynamic inactivation of planktonic cells of Gram positives (Tegos et al., 2008), Kishen et al. (2010) tested the potentiating effect of an efflux pump inhibitor (verapamil hydrochloride) on aPDT with MB against Enterococcus faecalis monospecies biofilms. Hereby, 4 days old biofilms were incubated with MB or anionic RB (see below, Chapter Eosin Y, Erythrosine, Rose Bengal) as a control PS at concentrations of 100 μ M for 15 min in the dark and subsequently irradiated with a noncoherent light source with 30 nm bandpass filters (300–600 mW; MB: 660 ± 15 nm; RB: 540 ± 15 nm), employing light doses from 10 to 40 J/cm². For MB, there was a killing efficacy of more than 5 log₁₀ steps, which could even be enhanced slightly by an additional incubation with the efflux pump inhibitor. Employing RB exhibited only marginally worse results (≈5 log₁₀). However, CLSM analysis showed that aPDT with these PS had different effects on the Enterococcus faecalis biofilms: While aPDT with RB produced no substantial destruction of biofilm structure, aPDT with MB resulted in damage of biofilm structure to a greater extent (Kishen et al., 2010).

In a study investigating the efficacy of two commercially available systems for aPDT, Meire et al. (2012) treated 24 h old biofilms of Enterococcus faecalis either with PAD™ (Dentofex, Inverkeithing, UK) or HELBO® (HELBO® Photodynamic Systems, Bredent medical, Senden, Germany) according to the recommendations of the manufacturers. For PAD™, biofilms were incubated for 2 min with TBO at a concentration of 12.7 mg/l and afterwards illuminated for 150 s with a soft diode laser (100 mW) emitting at 635 nm. For HELBO®, incubation of the Enterococcus faecalis biofilms was with MB (10 mg/ml) for 3 min with a subsequent irradiation for 2 min with a soft laser (HELBO® Theralite Laser) obtaining a wavelength of 660 nm and an output-power of 75 mW. Results showed a 2 log₁₀ step reduction for HELBO®, whereas PAD™ exhibited only an inactivation rate of less than 1 log₁₀ step. When in addition the antibacterial efficacy of irradiation with Er:YAG (2940 nm; 50 or 100 mJ; 15 Hz; 40 s) and Nd:YAG (1064 nm; 2 W; 15 Hz; 40 s) lasers was tested, there was a 4 log₁₀ inactivation rate for Er:YAG treatment using 100 mJ pulses, whereas Er:YAG using 50 mJ pulses and Nd:YAG had no effect at all (<1 log₁₀). In contrast, treatment with NaOCl (2.5%; 1, 5, 10, or 30 min) showed reductions of more than 6 log₁₀ regardless which immersion periods were used. However, a prolonged action of NaOCl beyond the respective treatment periods cannot be ruled out, since no reagent was used for stopping its antimicrobial effect after the respective immersion period (Meire et al., 2012).

However, in addition to the above-mentioned batch-culture studies, phenothiazinium derivatives have also been examined in more applied models: Zand et al. (2014) used extracted human maxillary and mandibulary incisors, decoronated them and instrumented their root canals up to #60 K-file. After an autoclavation process Enterococcus faecalis biofilms were formed in these root canals for 4, 6 or 8 weeks. Subsequently, canals were incubated with TBO (25 μg/ml) for 5 min and irradiated with a 625 nm diode laser (100 mW; light dose 214.28 J/cm²) twice for 2.5 min intermitted by a 2.5 min break, applying a total of 2 × 15 J, which led to a complete elimination of Enterococcus faecalis below detection limit of the CFU assay (Zand et al., 2014).

In contrast, Fimple et al. (2008) investigated the photodynamic effects of MB on multispecies biofilms comprising Actinomyces israelii, Fusobacterium nucleatum, Prevotella intermedia, and Porphyromonas gingivalis in infected root canals of extracted human teeth. Teeth were decoronated, instrumented up to an apical file size of 0.465, autoclaved and infected with the foresaid pathogens. After a culture period of 3 days canals were incubated with MB (25 μg/ml) for 10 min and irradiated by means of a diode laser (1 W; 665 nm) coupled to an optical fiber for 2.5 min followed by a 2.5 min break and a second light exposure of 2.5 min, applying a total light dose of 30 J/cm². Results showed a CFU reduction by 80%. The authors concluded that aPDT with MB could be an effective adjunct to conventional endodontic treatment, when the parameters get optimized (Fimple et al., 2008).

Up to date, phenothiazinium derivatives have to be regarded as the most studied PS in respect of inactivation of biofilms. Moreover, recently Voos et al. (2014) examined Safranine O (3,7-diamino-2,8-dimethyl-5-phenyl-phenazinium chloride), a structure analog to classical phenothiazinium dyes, where a nitrogen atom replaces the sulfur atom, which centers the delocalized electron system, for aPDT against biofilms. In this study, ex vivo biofilms were cultured from subgingival plaque samples for 24 or 72 h. After the respective incubation period the biofilms were incubated with Safranine O (10 μ M) for 15 min followed by irradiation with a diode laser (0.5 W; 532 nm) for 40 or 100 s, applying light doses of 20 or 50 J/cm², respectively. In 24 h old biofilms there was a reduction of 2 log₁₀ steps with a light dose of 20 J/cm² and 3 log₁₀ steps with 50 J/cm², whereas treatment with CHX (0.2%; 3 min) generated a reduction by 1 log₁₀ step only. In contrast, in 72 h old biofilms neither aPDT nor treatment with CHX led to any antibacterial effect (Voos et al., 2014). (Please find a summary of all studies described in this section in Table 1).

Porphyrins, chlorins, and phthalocyanines are structurally comparable heterocyclic macrocycles, whereby porphyrins and phthalocyanines are composed of four pyrrole subunits, while chlorins comprise three pyrrole cycles and one pyrroline (see Figure 3). Many of these compounds are naturally occurring. For example, one of the widest known porphyrins is heme, the pigment in erythrocytes, on which the red color of blood is based. Likewise, some bacterial species form porphyrins in their metabolism.

Oral bacteria that are able to synthetize for example black-pigmented species like Porphyromonas gingivalis and Prevotella spp. (Soukos et al., 2005; Lennon et al., 2006) or Aggregatibacter actinomycetemcomitans (Cieplik et al., 2013a). Consequently, it has been shown in vitro that these bacterial species can be inactivated by irradiation with light only, whereby it is a commonly accepted hypothesis that endogenous porphyrins among other substances may act as PS and lead to a lethal auto-photosensitization process (König et al., 2000; Soukos et al., 2005; Lennon et al., 2006; Cieplik et al., 2013a). Porphyrins and chlorins have an intense absorption maximum at approximately 405 nm, known as Soret band, and smaller peaks at wavelengths longer than 500 nm (Q bands) (Gouterman, 1961); their singlet oxygen quantum yields are in a range between 0.5 and 0.8 (Fernandez et al., 1997), thus acting predominantly according to type II mechanism.

There are some studies evaluating the efficacy of cationic PS being based on porphyrin or chlorin structures concerning the inactivation of biofilms. With respect to porphyrins, TMPyP [5,10,15,20-tetrakis(1-methyl-4-pyridinium)-porphyrin tetra-(p-toluenesulfonate]—a fourfold positively charged derivative—has to be regarded as the most commonly used PS (Di Poto et al., 2009; Collins et al., 2010; Maisch et al., 2012; Cieplik et al., 2013b; Eichner et al., 2013; Gonzales et al., 2013). Di Poto et al. (2009) cultured biofilms from three distinct strains of Staphylococcus aureus for 24 h, incubated them with TMPyP at a concentration of 10 μ M for 15 min and irradiated them with increasing doses of white light (150–200 J/cm²) isolated from the emission of a tungsten lamp (166 mW/cm²; 400–800 nm), whereby this treatment exhibited a rate of bacteria killing of 1 to 2 log₁₀ steps at the highest light dose depending on the strain, which was tested. Hereby, CLSM analysis showed the presence of dead cells throughout the biofilm implying that there was no hindrance for the PS to diffuse into the biofilms; consequently, the authors suggested that the reduced susceptibility of biofilm bacteria compared to that, which was observed when planktonic bacteria were treated (≥6 log₁₀), may be due to differences in cell wall composition, growth rate or EPS-components hindering uptake of the PS through the cell walls. Furthermore, aPDT displayed a second anti-biofilm mechanism in this study leading to detachment of parts of the biofilm and therefore to a disruption of biofilm architecture (Di Poto et al., 2009).

Collins et al. (2010) tested the effect of aPDT with TMPyP on Pseudomonas aeruginosa biofilms, which were cultured for 24 h from a wild type or a mutant strain, respectively. After that period, 225 μ M TMPyP was applied, immediately followed by illumination with a mercury vapor lamp fitted with a colored glass filter blocking wavelengths shorter than 400 nm (100 W; ≥400 nm) for 10 min (220–240 J/cm²), which resulted in a pronounced inactivation rate of about 4 log₁₀ steps for both strains. In addition, CLSM analysis revealed that aPDT resulted in substantial disruption of wild-type biofilms, whereby TMPyP without irradiation did not have that effect (Collins et al., 2010).

In contrast, when our group used TMPyP for aPDT against monospecies biofilms of Enterococcus faecalis (as a control PS for evaluating the antibacterial efficacy of SAPYR, which will be discussed below) we found no reduction of CFU at all. In this study, after a cultivation period of 72 h, biofilms were incubated with TMPyP (100 μ M) for 60 min and irradiated for 2 min with a LED light-curing unit for dental resins (1360 mW/cm²; 460 ± 20 nm), whereby the intensity of light reaching the biofilms was 600 mW/cm². This inefficacy may be due to the large molecular structure of TMPyP, which may hinder the penetration of this PS throughout the EPS due to steric reasons. Furthermore, drug diffusion may also be delayed due to strong electrostatic interactions between fourfold positively charged TMPyP and negatively charged EPS-molecules. The emission of the LED light-curing unit was also not ideal for excitation of either SAPYR or TMPyP; nevertheless, this type of light source was chosen, as it is widely applied in dental practice (Cieplik et al., 2013b).

Pereira Gonzales et al. (2013) compared the antimicrobial efficacy of XF-73 [5,15-bis-[4-(3-trimethylammoniopropyloxy)-phenyl]-porphyrin chloride]—a porphyrin derivative with two positive charges only—to that of TMPyP for inactivation of Candida albicans biofilms. These biofilms were cultured for 24 h, incubated with increasing concentrations of PS for 4 h and illuminated with an incoherent light source (13.4 mW/cm²; 418 ± 20 nm) for 60 min, whereby a light dose of 48.2 J/cm² was applied. Results showed CFU reductions of more than 5 log₁₀ for concentrations of 1.0 μ M of XF-73 at least, whereas for TMPyP concentrations of 50 μ M were needed at the minimum for obtaining equal antimicrobial results. This was explained due to stronger electrostatic interactions of TMPyP with the EPS compared to XF-73 leading to a lower degree of diffusion of the PS into the biofilm (Gonzales et al., 2013).

Referring to chlorins, Photodithazine® is a commercially available cationic chlorin-e6 derivative with two absorption maxima at around 400 nm (Soret band) and 660 nm (slightly smaller peak, Q-band), which has been evaluated in two studies conducted by the group of Ana C. Pavarina (Dovigo et al., 2013; Quishida et al., 2013): Quishida et al. formed multispecies biofilms from Streptococcus mutans, Candida albicans and Candida galbrata for 48 h; afterwards they were incubated with Photodithazine® at concentrations from 100 to 250 mg/l for 20 min and exposed to red light from a LED light source (71 mW/cm²; 660 nm) for 9 min (light dose 37.5 J/cm²). At a PS-concentration of 200 mg/ml aPDT-treatment showed reductions of 1 or 2 log₁₀ steps for Candida spp. or Streptococcus mutans, respectively (Quishida et al., 2013). In another study from the same group, Dovigo et al. cultured monospecies biofilms of clinical isolates of Candida spp. according to the protocol outlined above. These were incubated with Photodithazine® (125 mg/l) for 20 min and afterwards illuminated with a LED device (25 mW/cm2; 660 ± 20 nm) applying a light dose of 37.5 J/cm². Results showed reductions of approximately 1 log₁₀ step (Dovigo et al., 2013). However, in both of these studies Photodithazine® was excited by red light at its minor absorption peak at 660 nm without this point being discussed by the authors. Consequently, irradiation with blue light at 400 nm may lead to better results (Dovigo et al., 2013; Quishida et al., 2013).

In contrast to porphyrins and chlorins, phthalocyanines are highly hydrophobic, which compromises their water solubility (Ribeiro et al., 2013b). Therefore, methods for solubilization of phthalocyanines are mandatory for applying these compounds as PS for aPDT, whereby one option is their entrapment in nanoemulsions. Ribeiro et al. evaluated chloro-aluminum phthalocyanine (ClAlPc) encapsulated in cationic and anionic nanoemulsions (final concentration of ClAlPc in nanoemulsions: 31.8 μ M) in two studies regarding its aPDT-efficacy on biofilms of Candida albicans (Ribeiro et al., 2013a,b). However, in these studies the antimicrobial efficacy was only examined by detecting metabolic cell activity with XTT reduction assay. Consequently, this data is not comparable to CFU-assay data, which is mandatory, since—according to the ASM—any novel approach has to accomplish a CFU reduction rate of more than 3 log₁₀ in order to use the terms “antibacterial” or “antimicrobial” (Table 2).

Eosin Y, Erythrosine and Rose Bengal (RB) are anionic xanthene dyes derived from fluorescein. Hereby, Eosin Y and Erythrosine are red dyes resulting from the action of either bromine or iodine on fluorescein, whereas RB is a pink dye eventuating from a tetrachlorination of Erythrosine (see Figure 4). All these dyes show intense absorption bands in the green wavelength range (480–550 nm) (DeRosa and Crutchley, 2002); their singlet oxygen quantum yields are between 0.6 and 0.8, therefore mainly acting according to type II mechanism (Wilkinson et al., 1993).

Employing Erythrosine (ERY; 2,4,5,7-tetraiodofluorescein) as a PS for aPDT, Wood et al. (2006) studied its effect on 200 μm thick biofilms of Streptococcus mutans, which were grown for up to 288 h in a constant-depth film fermentor, and compared it to MB and Photofrin, a Porphyrin derivative. The biofilms were incubated with 22 μ M ERY, MB, or Photofrin for 15 min and exposed to white light from a tungsten filament lamp (ERY: 22.7 mW/cm² in the wavelength range 500–550 nm; MB, Photofrin: 22.5 mW/cm² at 600–650 nm) for 15 min. Hereby, ERY-mediated aPDT resulted in a reduction of about 3 log₁₀ steps for 288 h old biofilms, whereas aPDT with MB and Photofrin showed only 2.6 or 1.1 log₁₀ steps inactivation, respectively. Contrary to expectations, 48 h old biofilms were less susceptible to aPDT, regardless which PS was used (ERY: 2.2 log₁₀; MB: 1.5 log₁₀; Photofrin: 0.5 log₁₀). The authors explained this as young biofilms may be more metabolically active, thus having more effective repair systems. Moreover, older biofilms may contain voids and channels through the EPS, allowing greater amount of penetration of the respective PS (Wood et al., 2006). However, negatively charged ERY being the most effective PS in this study is quite contradictory to the above-mentioned general opinion, how a “perfect” PS should be composed, as the positive charge of a PS appears to promote an electrostatic interaction with negatively charged sites at the outer membrane of bacterial cells (Maisch et al., 2004). In contrast, anionic fluorescein derivatives like ERY or RB may be efficient though due to their high lipophilicity. In another study of the same group, Metcalf et al. (2006) investigated the effect of light-fractionation on aPDT with ERY on Streptococcus mutans biofilms formed under equal experimental parameters. ERY was used at 22 μ M with a 15 min incubation period, too. Biofilms were illuminated with the same tungsten filament lamp (22.7 mW/cm²; 500–550 nm), whereby a light dose of 6.75 J/cm² was applied for every 5 min of irradiation. Here, it was shown that aPDT with ERY can be potentiated by light fractionation: Streptococcus mutans could be inactivated by about 2 log₁₀ steps, when ERY was irradiated continuously for 5 min, whereas rates of 3 log₁₀ and 3.7 log₁₀ could be achieved, when there were either 5 × 1 min light pulses with 5 min recovery periods or 10 × 30 s light pulses with 2 min recovery breaks between pulses. This was explained by the authors due to replenishment of target molecules such as oxygen for the PS during the dark periods (Metcalf et al., 2006).

In order to compare the photodynamic efficacy of ERY and RB (4,5,6,7-tetrachloro-2′,4′,5′,7′-tetraiodofluorescein), Pereira et al. (2013) cultured biofilms of Streptococcus mutans and Streptococcus sanguinis for 48 h and incubated them with either with ERY or RB at concentrations of 5 μ M for 5 min followed by irradiation with a blue LED (200 mW; 455 ± 20 nm) for 180 s (light dose: 95 J/cm²; applied energy: 36 J). This treatment exhibited small effects of less than 1 log₁₀ for both species regardless of whether ERY or RB was used. However, it has to be considered that blue light from a 455 nm LED may not be optimal for excitation of ERY and RB with green light being more appropriate (Pereira et al., 2013). In a study from the same group, Freire et al. (2013) compared RB with Eosin Y (2,4,5,7-tetrabromofluorescein) concerning their potency for inactivation of Candida albicans biofilms. After culturing these biofilms for 48 h, treatment was done with 200 μ M RB or Eosin Y for 5 min as a pre-irradiation period and subsequent illumination with a green LED (90 mW; 532±10 nm) for 180 s (light dose: 42.63 J/cm²; applied energy: 16.2 J), which resulted in inactivation rates of 0.22 and 0.45 log₁₀ for RB and Eosin Y, respectively (Freire et al., 2013). In contrast to those results, Kishen et al. achieved a high killing efficacy of approximately 5 log₁₀ steps, when using RB against Enterococcus faecalis monospecies biofilms, as mentioned above (see Chapter Phenothiazinium Derivatives) (Kishen et al., 2010).

However, as cationic PS are needed for good antimicrobial photodynamic activity, the group of Annie Shrestha and Anil Kishen synthesized a polycationic chitosan-conjugated Rose Bengal (CSRB) for mounting a positive charge on anionic RB (Shrestha and Kishen, 2012; Shrestha et al., 2012). This conjugate was evaluated against Enterococcus faecalis and Pseudomonas aeruginosa biofilms, whereby its efficacy was compared to RB and MB. Biofilms were cultured for 7 days, incubated for 15 min with the respective PS at concentrations of 0.3 mg/ml (CSRB) or 10 μ M (RB, MB) and irradiated by 540 nm (CSRB, RB) or 660 nm (MB) fibers. Applying a light dose of 40 J/cm² resulted in inactivation rates of about 3 log₁₀ steps for MB in both biofilms, while for RB there were reductions of 3 log₁₀ steps in Pseudomonas aeruginosa biofilms and 4 log₁₀ steps in Enterococcus faecalis biofilms. In contrast, when using CSRB there was an inactivation of 9 log₁₀ steps of Pseudomonas aeruginosa and 5 log₁₀ steps of Enterococcus faecalis. The authors explained these strikingly better results of CSRB compared to anionic RB and cationic MB due to the greater ability of CSRB to adhere to bacterial cells and the synergistic antibacterial effects of chitosan and RB (Shrestha and Kishen, 2012) (Table 3).

Recently, new classes of PS have been introduced as PS for aPDT: Curcumins, perinaphthenone derivatives and fullerenes (see Figure 5).

Curcumin [1E,6E-1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione] is a naturally occurring intense-yellow dye, isolated from the rootstocks of the plant Curcuma longa, which has been used as medicine, spice and food-colorant for hundreds of years (Crivello and Bulut, 2005; Aggarwal et al., 2007). As its absorption spectrum is in the UV/blue wavelength range (300–500 nm with a maximum at 430 nm), it has been thought to be a feasible PS for aPDT (Araújo et al., 2012). However, curcumin is a nearly quantitative type I mechanism PS, whose photodynamic effect is due to hydrogen peroxide without generation of singlet oxygen (Dahl et al., 1989).

Araújo et al. (2014) evaluated the susceptibility of Streptococcus mutans and Lactobacillus acidophilus, either grown in multispecies biofilms or in artificial dentine carious lesions, toward aPDT with a curcumin-solution (consisting of 66.7% glucamin, 17.8% curcumin, 15.5% demethoxy and bisdemethoxy curcumin; concentration: 0.75 to 5 g/l). Biofilms were cultured for 7 days, exposed to the curcumin-solution for 5 min and irradiated for 5 min with blue light derived from a LED (central wavelength: 450 nm; light dose: 5.7 J/cm²). With curcumin-concentrations above 3 g/l, CFU were reduced by more than 3 log₁₀ steps in the multispecies biofilms. However, when the dentine carious lesion was exposed to curcumin (5 g/l) with subsequent irradiation, there was only a small reduction of 69%; this reduced effect on bacteria located in demineralized dentine was explained by the authors due to decreased penetration depth of the PS, diminished binding to bacterial cells or attenuated penetration of light (Araújo et al., 2014).

In contrast to curcumins, perinaphthenones are only slight yellowish and exhibit singlet oxygen quantum yields close to unity, therefore acting nearly exclusively according to type II mechanism. Consequently, PNS (1H-phenalen-1-one-2-sulfonic acid), which was the first derivative of perinaphthenone being employed as a PS (originally described by Nonell et al., 1993), has been proposed as a universal standard for singlet oxygen studies (Oliveros et al., 1999). However, PNS is negatively charged, whereas positively charged PS are essential for good antimicrobial activity (Alves et al., 2009). Therefore, recently SAPYR [2-((4-pyridinyl)methyl)-1H-phenalen-1-one chloride] was introduced as the lead compound of a batch of positively charged derivatives based on a 7-perinaphthenone-structure, eliminating the drawback of PNS, but conserving the high singlet oxygen quantum yield of Φ_Δ 0.99 (Cieplik et al., 2013b; Späth et al., 2014). Our group evaluated SAPYR for its anti-biofilm properties against Enterococcus faecalis and Actinomyces naeslundii grown in monospecies or in polyspecies biofilms consisting of Enterococcus faecalis, Actinomyces naeslundii, and Fusobacterium nucleatum (Cieplik et al., 2013b). These biofilms were cultured for 72 h, afterwards incubated for 60 min with the respective PS at 100 μ M (TMPyP served as a reference PS) and irradiated for 120 s with a blue LED, usually applied for polymerization of dental resins (light intensity at sample-level: 600 mW/cm²). This light source was used, as it is widely applied in dental practice, although its emission spectrum shows only partial overlap with the absorption spectra of SAPYR and TMPyP. Nevertheless, aPDT with SAPYR resulted in CFU-reductions of more than 5 log₁₀ for Enterococcus faecalis in both types of biofilm; Actinomyces naeslundii was inactivated by more than 2 log₁₀ in monospecies and by more than 4 log₁₀ in polyspecies biofilms. In contrast, as mentioned above (Section Porphyrins, Chlorins, and Phthalocyanines), TMPyP had no effect at all (Cieplik et al., 2013b).

As a completely distinct class of PS, fullerenes are ball-shaped cage molecules composed of carbon atoms, whereby fullerene C60 has to be regarded as the most studied compound (Sharma et al., 2011). Recently, the group of Michael R. Hamblin introduced several functionalized cationic fullerenes as PS for aPDT, which showed notable results in vitro against planktonic bacteria and yeasts (Mizuno et al., 2011). These compounds absorb extensively in the visible and UV wavelength range (Sharma et al., 2011) and react—in contrast to perinaphthenone derivatives—predominantly according to type I mechanism, thus producing mainly superoxide (Mizuno et al., 2011). However, until now fullerenes have not been tested for their antimicrobial photodynamic efficacy against biofilms (Table 4).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.