The Photodynamic Anticancer and Antibacterial Activity Properties of a Series of meso-Tetraarylchlorin Dyes and Their Sn(IV) Complexes

A series of tetraarylchlorins with 3-methoxy-, 4-hydroxy- and 3-methoxy-4-hydroxyphenyl meso-aryl rings (1-3-Chl) and their Sn(IV) complexes (1-3-SnChl) were synthesized and characterized so that their potential utility as photosensitizer dyes for use in photodynamic therapy (PDT) and photodynamic antimicrobial chemotherapy (PACT) can be assessed. The photophysicochemical properties of the dyes were assessed prior to in vitro PDT activity studies against MCF-7 breast cancer cells through irradiation with Thorlabs 625 or 660 nm LED for 20 min (240 or 280 mW·cm−2). PACT activity studies were performed against both planktonic bacteria and biofilms of Gram-(+) S. aureus and Gram-(−) E. coli upon irradiation with Thorlabs 625 and 660 nm LEDs for 75 min. The heavy atom effect of the Sn(IV) ion results in relatively high singlet oxygen quantum yield values of 0.69−0.71 for 1-3-SnChl. Relatively low IC50 values between 1.1−4.1 and 3.8−9.4 µM were obtained for the 1-3-SnChl series with the Thorlabs 660 and 625 nm LEDs, respectively, during the PDT activity studies. 1-3-SnChl were also found to exhibit significant PACT activity against planktonic S. aureus and E. coli with Log10 reduction values of 7.65 and >3.0, respectively. The results demonstrate that the Sn(IV) complexes of tetraarylchlorins merit further in depth study as photosensitizers in biomedical applications.


Introduction
Cancer causes more deaths each year than malaria, tuberculosis, and HIV combined [1]. Traditional treatments can have severe side effects. In the mid-1970s, photodynamic therapy (PDT) emerged as an alternative non-invasive treatment option for treating cancers that can be readily reached by laser light via a fiber optic. Upon photoexcitation, the T 1 triplet state of a photosensitizer (PS) dye is populated through intersystem crossing and singlet oxygen is generated as the main cytotoxic agent from molecular dioxygen by energy transfer [2]. Photofrin ® was the first PS dye to be clinically approved. It is comprised of a series of porphyrin oligomers that exhibit significant dark toxicity and slow clearance from the body. The lowest energy Q band is relatively weak and only absorbs at the edge of the therapeutic window (620-850 nm), where human tissue has high transparency to incident laser light. Endogenous chromophores, such as heme, absorb significantly at <600 nm [3], while H 2 O absorbs significantly at longer wavelengths in the near-infrared (NIR) region. Other optical properties, such as the scattering coefficients of particles in biological tissues, exhibit a strong wavelength dependence at higher energy. As a result, new PS dyes for PDT should absorb significantly in the phototherapeutic window.
The second generation of PS dyes was mainly comprised of phthalocyanines, since their intense Q bands lie in the 650−700 nm range [4]. These dyes have solubility issues related to π-π stacking related aggregation, which can make their photophysicochemical properties less suitable for PS dye applications. Attempts have been made to address this by modifying the structures of PS dyes to enhance the singlet oxygen quantum yield (Φ ∆ ) and decrease aggregation effects. Chlorins (Scheme 1), such as chlorin e6 and Purlytin ® [5][6][7], have emerged as PS dyes for PDT due to their relatively intense Q bands at ca. 650 nm [3,8,9]. In recent years, we have reported a series of studies on the PDT activity properties of the Sn(IV) complexes of readily synthesized porphyrin analogues [10], including N-confused porphyrins [11,12], corroles [13,14], and chlorins [15][16][17]. This research has focused on Sn(IV) complexes, since the presence of two axial ligands through trans-axial on the Sn(IV) ion hinders π-π stacking and hence aggregation, and the heavy atom effect of this ion enhances the generation of singlet oxygen [10].
Molecules 2023, 28, x FOR PEER REVIEW 2 of 20 The second generation of PS dyes was mainly comprised of phthalocyanines, since their intense Q bands lie in the 650−700 nm range [4]. These dyes have solubility issues related to π-π stacking related aggregation, which can make their photophysicochemical properties less suitable for PS dye applications. Attempts have been made to address this by modifying the structures of PS dyes to enhance the singlet oxygen quantum yield (ΦΔ) and decrease aggregation effects. Chlorins (Scheme 1), such as chlorin e6 and Purlytin ® [5][6][7], have emerged as PS dyes for PDT due to their relatively intense Q bands at ca. 650 nm [3,8,9]. In recent years, we have reported a series of studies on the PDT activity properties of the Sn(IV) complexes of readily synthesized porphyrin analogues [10], including N-confused porphyrins [11,12], corroles [13,14], and chlorins [15][16][17]. This research has focused on Sn(IV) complexes, since the presence of two axial ligands through trans-axial on the Sn(IV) ion hinders π-π stacking and hence aggregation, and the heavy atom effect of this ion enhances the generation of singlet oxygen [10]. Our previous studies on chlorins have focused on ligands with meso-thienyl andmethylthiophenyl substituents [15][16][17]. In this study, the PDT activities of the Sn(IV) complexes of a series of structurally analogous chlorins with 3-methoxy-, 4-hydroxy-and 3methoxy-4-hydroxyphenyl meso-aryl rings (1-3-SnChl) are compared to the corresponding chlorins (1-3-Chl), which were prepared by reducing one of the peripheral pyrrole bonds of the corresponding porphyrins (1-3-Por) (Scheme 1). The suitability of 1-3-SnChl for use in photodynamic antimicrobial chemotherapy (PACT) [18][19][20] is also examined against both planktonic S. aureus and E. coli and their biofilms as typical Gram-(+) and Gram-(−) strains, respectively. Unusually for non-cationic species [21,22], high Log10 reduction values were reported against E. coli in our previous studies involving Sn(IV) chlorins with meso-aryl substituents that contain sulfur atoms [16,17].
The 1 H NMR and MALDI-TOF MS data confirmed the structures of target chlorin dyes 1-Chl and 2-Chl and Sn(IV) complexes 1-3-SnChl ( Figures S1 and S2). The anticipated proton signals for the synthesized chlorin dyes can be readily assigned. The chemical shifts at 8.00−8.71 ppm in each 1 H NMR spectrum are assigned to the six protons from the Our previous studies on chlorins have focused on ligands with meso-thienyl and -methylthiophenyl substituents [15][16][17]. In this study, the PDT activities of the Sn(IV) complexes of a series of structurally analogous chlorins with 3-methoxy-, 4-hydroxy-and 3methoxy-4-hydroxyphenyl meso-aryl rings (1-3-SnChl) are compared to the corresponding chlorins (1-3-Chl), which were prepared by reducing one of the peripheral pyrrole bonds of the corresponding porphyrins (1-3-Por) (Scheme 1). The suitability of 1-3-SnChl for use in photodynamic antimicrobial chemotherapy (PACT) [18][19][20] is also examined against both planktonic S. aureus and E. coli and their biofilms as typical Gram-(+) and Gram-(−) strains, respectively. Unusually for non-cationic species [21,22], high Log 10 reduction values were reported against E. coli in our previous studies involving Sn(IV) chlorins with meso-aryl substituents that contain sulfur atoms [16,17].
The 1 H NMR and MALDI-TOF MS data confirmed the structures of target chlorin dyes 1-Chl and 2-Chl and Sn(IV) complexes 1-3-SnChl ( Figures S1 and S2). The anticipated proton signals for the synthesized chlorin dyes can be readily assigned. The chemical shifts at 8.00−8.71 ppm in each 1 H NMR spectrum are assigned to the six protons from the β-pyrrole positions, while the aromatic protons from the meso-aryl groups lie between 7.00−8.00 ppm at slightly lower frequencies. The hydroxy (OH) protons from the mesoaryl substituent groups lie at higher frequencies between 8.84−10.00 ppm for 2-3-Chl and the 2-3-SnChl complexes ( Figure S1). The methyl (CH 3 ) protons from the meso-aryl substituent groups lie between 3.83−4.07 ppm in the aliphatic region for 1-Chl and 3-Chl and their corresponding Sn(IV) chlorin complexes ( Figure S1). The characteristic four pyrroline protons for chlorin dyes lie between 4.12−4.37 ppm as singlet signals ( Figure S1). These proton peaks confirm the synthesis of chlorin dyes from their parent porphyrin structures [26,27]. Singlet peaks for the two NH core protons for 1-3-Chl lie between −1.42 to −1.57 ppm in the negative low-frequency region. Upon insertion of Sn(IV) ions, the two NH protons disappeared in the spectra of 1-3-SnChl ( Figure S1). This confirms their successful synthesis. MALDI-TOF mass data confirmed the structures for the target chlorin dyes. The anticipated parent ion peaks are observed, which agree closely with the theoretical masses of chlorin dyes. The MALDI-TOF data for 1-3-Chl and their 1-3-SnChl complexes are provided in Figure S2. The MS data for 1-3-Chl free-base chlorin dyes exhibit mainly [M+H] + or [M] + molecular ion species, while the 1-3-SnChl complexes exhibited [M−Cl+H] + or [M−Cl] + parent peak species due to the loss of a chloride ion.

Photophyiscochemical Properties
The ground state absorption spectra of the studied A 4 free base meso-tetraaryl chlorins are typical of chlorin dyes, with intense B bands at ca. 425 nm and a set of four Q bands at low energy region and a characteristic slightly more intense Q band at ca. 650 nm, in contrast to the parent porphyrin dyes [28] (Figure 1). For Sn(IV) chlorins (1-3-SnChl), there are red shifts in the B bands, and the four Q bands of free-base ligands collapse into three with the characteristic slightly more intense Q band blue shifting to ca. 630 nm ( Figure 1). The greater intensity of the Q band arises from hydrogen reduction on one of the peripheral β-β pyrrole double bonds of the parent porphyrin ring to form a low-symmetry chlorin dye [16,28]. Table 1 summarizes the optical properties of the chlorin dyes studied in DMSO. There are minor spectral changes due to inductive and mesomeric effects from the meso-aryl substituents of the chlorins studied. The absorption spectra of meso-4-hydroxyphenyl (2) and meso-4-hydroxy-3-methoxyphenyl (3) chlorin derivatives are red shifted in contrast to those of the meta-substituted meso-3-methoxyphenylchlorin (1). This is due to more favorable mesomeric interactions between the chlorin core and meso-phenyl groups with para-substituents [29].
7.00−8.00 ppm at slightly lower frequencies. The hydroxy (OH) protons from the meso-aryl substituent groups lie at higher frequencies between 8.84−10.00 ppm for 2-3-Chl and the 2-3-SnChl complexes ( Figure S1). The methyl (CH3) protons from the meso-aryl substituent groups lie between 3.83−4.07 ppm in the aliphatic region for 1-Chl and 3-Chl and their corresponding Sn(IV) chlorin complexes ( Figure S1). The characteristic four pyrroline protons for chlorin dyes lie between 4.12−4.37 ppm as singlet signals ( Figure S1). These proton peaks confirm the synthesis of chlorin dyes from their parent porphyrin structures [26,27]. Singlet peaks for the two NH core protons for 1-3-Chl lie between −1.42 to −1.57 ppm in the negative low-frequency region. Upon insertion of Sn(IV) ions, the two NH protons disappeared in the spectra of 1-3-SnChl ( Figure S1). This confirms their successful synthesis. MALDI-TOF mass data confirmed the structures for the target chlorin dyes. The anticipated parent ion peaks are observed, which agree closely with the theoretical masses of chlorin dyes. The MALDI-TOF data for 1-3-Chl and their 1-3-SnChl complexes are provided in Figure S2. The MS data for 1-3-Chl free-base chlorin dyes exhibit mainly [M+H] + or [M] + molecular ion species, while the 1-3-SnChl complexes exhibited [M−Cl+H] + or [M−Cl] + parent peak species due to the loss of a chloride ion.

Photophyiscochemical Properties
The ground state absorption spectra of the studied A4 free base meso-tetraaryl chlorins are typical of chlorin dyes, with intense B bands at ca. 425 nm and a set of four Q bands at low energy region and a characteristic slightly more intense Q band at ca. 650 nm, in contrast to the parent porphyrin dyes [28] (Figure 1). For Sn(IV) chlorins (1-3-SnChl), there are red shifts in the B bands, and the four Q bands of free-base ligands collapse into three with the characteristic slightly more intense Q band blue shifting to ca. 630 nm ( Figure 1). The greater intensity of the Q band arises from hydrogen reduction on one of the peripheral β-β pyrrole double bonds of the parent porphyrin ring to form a low-symmetry chlorin dye [16,28]. Table 1 summarizes the optical properties of the chlorin dyes studied in DMSO. There are minor spectral changes due to inductive and mesomeric effects from the meso-aryl substituents of the chlorins studied. The absorption spectra of meso-4-hydroxyphenyl (2) and meso-4-hydroxy-3-methoxyphenyl (3) chlorin derivatives are red shifted in contrast to those of the meta-substituted meso-3-methoxyphenylchlorin (1). This is due to more favorable mesomeric interactions between the chlorin core and meso-phenyl groups with para-substituents [29].  evidenced by slight red shifts from CHCl 3 < DMSO. Band broadening is observed in the absorption spectra of free-base chlorins 1-2-Chl in 1% DMSO/H 2 O relative to those of the 1-2-SnChl dichlorotin(IV) chlorin complexes ( Figure 1 and Table S1), but is relatively limited for both sets of dyes. This is likely related to aggregation effects, which are decreased by the dichloro axial ligands of Sn(IV) chlorin complexes since face-to-face π-π stacking is hindered [10]. The additional broadening that is observed in the spectra of 3-Chl and 3-SnChl may be related to the hydrogen bonding of the 4-hydroxy-3-methoxyphenyl rings. The chlorins studied have two-band emission profiles typical of chlorin dyes ( Figure 2) [28]. The emission spectra of Sn(IV) chlorins are blue shifted in contrast to the free-base chlorins due to the introduction of the Sn(IV) ion resulting in rigid planar structures. The trends in the emission profiles of the chlorin dyes studied correspond closely to those in the UV-visible absorption spectra, with red shifts observed for 2-Chl, and 3-Chl relative to 1-Chl ( Figure 1, Table 1).
The solvatochromic absorption data in CHCl3 and 1% DMSO/H2O of the chlorins studied are provided in Figure 1. The main absorption bands of the chlorin dyes studied exhibit minor solvatochromic effects with a slight dependence on solvent polarity, as evi denced by slight red shifts from CHCl3 < DMSO. Band broadening is observed in the ab sorption spectra of free-base chlorins 1-2-Chl in 1% DMSO/H2O relative to those of the 1 2-SnChl dichlorotin(IV) chlorin complexes ( Figure 1 and Table S1), but is relatively limited for both sets of dyes. This is likely related to aggregation effects, which are decreased by the dichloro axial ligands of Sn(IV) chlorin complexes since face-to-face π-π stacking is hindered [10]. The additional broadening that is observed in the spectra of 3-Chl and 3 SnChl may be related to the hydrogen bonding of the 4-hydroxy-3-methoxyphenyl rings The chlorins studied have two-band emission profiles typical of chlorin dyes ( Figure  2) [28]. The emission spectra of Sn(IV) chlorins are blue shifted in contrast to the free-base chlorins due to the introduction of the Sn(IV) ion resulting in rigid planar structures. The trends in the emission profiles of the chlorin dyes studied correspond closely to those in the UV-visible absorption spectra, with red shifts observed for 2-Chl, and 3-Chl relative to 1-Chl ( Figure 1, Table 1). The photophysicochemical properties of 1-3-Chl and 1-3-SnChl in DMSO are sum marized in Table 2. The free-base chlorins (1-3-Chl) have relatively low fluorescence quan tum yield (ΦF) values and relatively long fluorescence lifetimes (τF) ( Table 2). This is con sistent with the reported trends for chlorin dyes and is related to their flexible structures [30][31][32]. The ΦF and τF values decreased significantly on metalation of the dyes to form 1 3-SnChl due to the Sn(IV) ion heavy atom effect [16]. The triplet lifetime (τT) values of 1 3-Chl are also longer than those of 1-3-SnChl due to the heavy atom effect [16]. The free base chlorins (1-3-Chl) have relatively high ΦΔ values that lie in the 0.59−0.62 range ( Table  2). This can be attributed to effective quenching processes in the transfer of energy by the triplet state of chlorin dyes to molecular oxygen [31]. The ΦΔ values of 1-3-SnChl, on the other hand, are higher than those of the corresponding free-base chlorins due to the heavy atom effect [16]. The photostabilities in 1% DMSO/H2O solutions of the chlorins studied were determined by determining the extent of photobleaching of the B band after 30 min photoirradiation with a 660 nm (280 mW·cm −2 ) Thorlabs LED lamp mounted onto a Mod ulight 7710-680 medical laser system housing in a similar manner to the PDT and PACT The photophysicochemical properties of 1-3-Chl and 1-3-SnChl in DMSO are summarized in Table 2. The free-base chlorins (1-3-Chl) have relatively low fluorescence quantum yield (Φ F ) values and relatively long fluorescence lifetimes (τ F ) ( Table 2). This is consistent with the reported trends for chlorin dyes and is related to their flexible structures [30][31][32]. The Φ F and τ F values decreased significantly on metalation of the dyes to form 1-3-SnChl due to the Sn(IV) ion heavy atom effect [16]. The triplet lifetime (τ T ) values of 1-3-Chl are also longer than those of 1-3-SnChl due to the heavy atom effect [16]. The free-base chlorins (1-3-Chl) have relatively high Φ ∆ values that lie in the 0.59−0.62 range ( Table 2). This can be attributed to effective quenching processes in the transfer of energy by the triplet state of chlorin dyes to molecular oxygen [31]. The Φ ∆ values of 1-3-SnChl, on the other hand, are higher than those of the corresponding free-base chlorins due to the heavy atom effect [16]. The photostabilities in 1% DMSO/H 2 O solutions of the chlorins studied were determined by determining the extent of photobleaching of the B band after 30 min photoirradiation with a 660 nm (280 mW·cm −2 ) Thorlabs LED lamp mounted onto a Modulight 7710-680 medical laser system housing in a similar manner to the PDT and PACT activity measurements ( Table 2). The free-base chlorins have low photostability values (63-67%), in contrast to those of their Sn(IV) chlorin complexes. This is due to the flexible structural properties of free-base chlorins since they lack the stabilization effect at the frontier π-MOs from a high valent metal ion [28]. The photostabilities of Sn(IV) chlorins are higher due to their greater structural rigidity and the heavy atom effect [16]. Table 2. Photophysicochemical and photostability properties of 1-3-Chl and 1-3-SnChl in DMSO.   (Table 3). This could be due to the reported favorable effects of stronger absorption of light by a PS dye at longer wavelength deep into the therapeutic window, since it can readily penetrate into the tissues resulting in enhanced phototherapy effects [16,33,34]. activity measurements ( Table 2). The free-base chlorins have low photostability values (63-67%), in contrast to those of their Sn(IV) chlorin complexes. This is due to the flexible structural properties of free-base chlorins since they lack the stabilization effect at the frontier π-MOs from a high valent metal ion [28]. The photostabilities of Sn(IV) chlorins are higher due to their greater structural rigidity and the heavy atom effect [16].   (Table 3). This could be due to the reported favorable effects of stronger absorption of light by a PS dye at longer wavelength deep into the therapeutic window, since it can readily penetrate into the tissues resulting in enhanced phototherapy effects [16,33,34].   1-3-Chl free-base chlorins also exhibited relatively favorable PDT activities (Table 3).  (Table 3). As would be anticipated based on their high Φ ∆ values and stronger absorption in the NIR region, 1-3-SnChl significantly outperformed their 1-3-Chl free-base chlorin ligands (Table 3). For instance, upon illumination at 660 nm ( Figure 3, Table 3), 1-3-SnChl had IC 50 values of 1.1−4.1 µM, and PI values of 16.7−24.8, and low cell viability values of 3.9−11.1% at 25 µM. The PDT activity trends due to the different meso-aryl substituents of 1-3-Chl and 1-3-SnChl are consistent with the anticancer properties of vanillic and phenolic derivatives [35][36][37]. For example, meso-vanillic chlorin derivatives (3-Chl, 3-SnChl) followed by phenolic chlorins (2-Chl, 2-SnChl) have significantly higher PDT activities than the meso-methoxyphenylchlorins (1-Chl and 1-SnChl).

Lipophilicity
The lipophilicity (Log P o/w ) values of 1-3-SnChl were determined by the shake flask method (see Section 4.6), since relatively low IC 50 values were obtained ( Table 3). The Log P o/w values lie between 0.96 and 1.39 (Table 4) in the optimal lipophilic balance range for drugs [38], since the dichloro axial ligands can minimize aggregation effects at higher concentrations of the dyes and enhance cellular uptake [10,39,40]. The lower Log P o/w values for 2-SnChl and 3-SnChl can be explained by the scope for hydrogen bonding with water through the -OH groups on the meso-aryl rings.         a Illumination time in minutes required to achieve a Log 10 reduction value corresponding to 0% cell viability is provided after the @ character. Otherwise, the Log10 reduction value at 75 min is provided.
1-3-Chl free-base chlorin ligands also exhibit moderate Log 10 reduction values under illumination at 660 nm for planktonic E. coli in the 2.20−2.65 range after illumination for 75 min, which denotes ≥99% cell reduction and ≤0.9 cell survival. This is in contrast to the favorable performance 1-3-Chl exhibited toward planktonic S. aureus. This trend is anticipated in the context of neutral PS dyes toward Gram-(−) E. coli, since E. coli bacteria have a permeability barrier, which restricts PS dyes from penetrating the cells [41,42]. On the other hand, 1-3-SnChl exhibited favorable PACT activities toward planktonic E. coli with higher Log 10 reduction values of ≥3.50 after 75 min illumination with Thorlabs M625L3 and M660L4 LEDs (Table 5). These values are greater than the optimal ≥3 Log 10 (99.9% cell reduction) United States Food and Drug Administration's guideline for antimicrobial agents [43,44]. This suggests that 1-3-SnChl can be efficient PS dyes for the eradication of planktonic E. coli. The effects of different meso-aryl substituents on the PACT activities of the chlorin dyes against planktonic S. aureus and E. coli are consistent with those observed for their PDT activities. The meso-4-hydroxyphenyl (2-Chl, 2-SnChl) and -4-hydroxy-3methoxyphenylchlorins (3-Chl, 3-SnChl) have slightly higher PACT activities against E. coli than the meso-methoxyphenylchlorin (1-Chl, 1-SnChl) ( Figure 5, Table 5). This can be attributed to the hydrophilic vanillic and phenolic meso-aryl groups of the chlorin PS dyes, which can interact favorably with the planktonic S. aureus and E. coli cell wall resulting in slightly enhanced PS penetration and cellular uptake [39,45].

In Vitro PACT Cytotoxicity Studies against Biofilms of S. aureus and E. coli Bacteria
The optimal concentrations of 1-3-Chl and 1-3-SnChl used during the time-dependence studies were 25 µM for S. aureus biofilms and 50 µM for E. coli biofilms (Figures 6-8). The Log 10 reduction values for S. aureus and E. coli biofilms are ≥1 Log 10 with ≥90% cell reduction and ≤10% cell survival (Table 6). These Log 10 reduction values are much lower than those observed against the planktonic S. aureus and E. coli bacteria. This is normally anticipated in the context of S. aureus and E. coli biofilms, since biofilms are less susceptible to antimicrobial agents due to their self-securing polymeric matrix, which restricts drug penetration [46][47][48]. The photoexcitation of chlorin dyes at 660 nm resulted in slightly higher Log 10 reduction values for S. aureus and E. coli biofilms than those obtained with 625 nm irradiation (Table 6). This observation agrees with previous reports, which explained that NIR illumination facilitates deeper penetration of light into the tissue resulting in enhanced PACT activities [16,36,37]. The meso-4-hydroxyphenyl and -4-hydroxy-3-methoxyphenylchlorins (2-3-Chl, 2-3-SnChl) exhibited slightly higher PACT activities toward biofilm cells of S. aureus and E. coli than the meso-methoxyphenylchlorins (1-Chl, 1-SnChl) ( Table 6). This could be associated with favorable interactions between the hydrophilic vanillic and phenolic groups of the chlorin dyes and the biofilm cells, resulting in slightly enhanced drug penetration and cellular uptake [45]. As anticipated based on their high ΦΔ values, the Log10 reduction values for 1-3-SnChl are slightly higher toward biofilm cells of S. aureus and E. coli in contrast to the 1-3-Chl free-base ligands (Figures 7 and 8, Table 6). LEDs with viable colonies quantified by the viable colony count method. The error bars represent the mean ± standard deviation of three independent assays each with three replicates. The meso-4-hydroxyphenyl and -4-hydroxy-3-methoxyphenylchlorins (2-3-Chl, 2-3-SnChl) exhibited slightly higher PACT activities toward biofilm cells of S. aureus and E. coli than the meso-methoxyphenylchlorins (1-Chl, 1-SnChl) ( Table 6). This could be associated with favorable interactions between the hydrophilic vanillic and phenolic groups of the chlorin dyes and the biofilm cells, resulting in slightly enhanced drug penetration and cellular uptake [45]. As anticipated based on their high Φ ∆ values, the Log 10 reduction values for 1-3-SnChl are slightly higher toward biofilm cells of S. aureus and E. coli in contrast to the 1-3-Chl free-base ligands (Figures 7 and 8, Table 6). The meso-4-hydroxyphenyl and -4-hydroxy-3-methoxyphenylchlorins (2-3-Chl, 2-3-SnChl) exhibited slightly higher PACT activities toward biofilm cells of S. aureus and E. coli than the meso-methoxyphenylchlorins (1-Chl, 1-SnChl) ( Table 6). This could be associated with favorable interactions between the hydrophilic vanillic and phenolic groups of the chlorin dyes and the biofilm cells, resulting in slightly enhanced drug penetration and cellular uptake [45]. As anticipated based on their high ΦΔ values, the Log10 reduction values for 1-3-SnChl are slightly higher toward biofilm cells of S. aureus and E. coli in contrast to the 1-3-Chl free-base ligands (Figures 7 and 8, Table 6).

Discussion
The PDT activities of 1-3-Chl and 1-3-SnChl are broadly similar to those of other tetraarylchlorins with thien-2-yl and methylthiophenyl rings that we have reported previously using a similar approach with a Thorlabs M660L4 LED (Table 7) [15][16][17]. In a similar manner, the PI values for 1-3-SnChl are significantly larger than those for 1-3-Chl, demonstrating again that coordination of a Sn(IV) ion enhances the PDT activities. This may be related to their higher ΦΔ values (Table 2) and scope for a decrease in aggregation due to trans-axial ligation [10]. In a similar manner, the PACT activity properties of 1-3-Chl and 1-3-SnChl against planktonic S. aureus and E. coli (Table 8) are broadly similar to those reported previously for dyes with thien-2-yl and methylthiophenyl meso-aryl rings [15][16][17]. The Log10 reduction values corresponding to ≈ 0% cell survival were achieved more quickly in the context of Gram-(+) S. aureus and at lower concentration than was the case previously with Sn(IV) tetramethylthiophenylchlorin [17], while moderately high Log10 reduction values > 3.0 were obtained for 1-3-SnChl against E. coli (Table 8). Values > 3.0 are consistent with the FDA's guidelines for an effective antibacterial agent [43,44]. In contrast with Sn(IV) tetramethylthiophenylchlorin [17]; however, 0% cell survival was not

Discussion
The PDT activities of 1-3-Chl and 1-3-SnChl are broadly similar to those of other tetraarylchlorins with thien-2-yl and methylthiophenyl rings that we have reported previously using a similar approach with a Thorlabs M660L4 LED (Table 7) [15][16][17]. In a similar manner, the PI values for 1-3-SnChl are significantly larger than those for 1-3-Chl, demonstrating again that coordination of a Sn(IV) ion enhances the PDT activities. This may be related to their higher Φ ∆ values ( Table 2) and scope for a decrease in aggregation due to trans-axial ligation [10]. In a similar manner, the PACT activity properties of 1-3-Chl and 1-3-SnChl against planktonic S. aureus and E. coli (Table 8) are broadly similar to those reported previously for dyes with thien-2-yl and methylthiophenyl meso-aryl rings [15][16][17]. The Log 10 reduction values corresponding to ≈0% cell survival were achieved more quickly in the context of Gram-(+) S. aureus and at lower concentration than was the case previously with Sn(IV) tetramethylthiophenylchlorin [17], while moderately high Log 10 reduction values > 3.0 were obtained for 1-3-SnChl against E. coli (Table 8). Values > 3.0 are consistent with the FDA's guidelines for an effective antibacterial agent [43,44]. In contrast with Sn(IV) tetramethylthiophenylchlorin [17]; however, 0% cell survival was not achieved against Gram-(−) E. coli for Sn(IV) complexes 1-3-SnChl with 3-methoxy-, 4-hydroxyand 3-methoxy-4-hydroxyphenyl meso-aryl rings. These results demonstrate that Sn(IV) tetraarylchlorin complexes merit further in depth study for use as PS dyes for PDT and against both Gram-(+) and Gram-(−) bacteria in the context of PACT. Table 7. IC 50 and PI values against MCF-7 cancer cells for 1-3-Chl and 1-3-SnChl and a series of tetraarylchlorins and Sn(IV) tetraarylchlorin complexes reported previously [15][16][17].

Instrumentation
1 H NMR spectra were measured with Bruker Avance II TM 600 MHz and AMX 400 MHz instruments with the solvent residual as the internal reference (δ = 7.26 ppm for CDCl 3 ). ACS spectral grade solvents were used for spectroscopic measurements. UV-visible spectra were recorded with a Shimadzu UV-2550 spectrophotometer and an Evolution 350-UV-Vis spectrophotometer from Thermo Fischer Scientific. Mass spectrometry data were measured on a Bruker AutoFLEX III Smartbeam TOF/TOF mass instrument in the positive ion mode with α-cyano-4-hydroxycinnamic acid as the MALDI matrix. Fluorescence spectra were measured with a Varian Eclipse ® spectrofluorometer with the optical density at the B band maxima adjusted to~0.05. Φ F values were determined in DMSO with a comparative method by using ZnTPP as the standard (Φ F = 0.039 in DMSO [49]). τ F values were determined with a Picoquant FluoTime 300 TCSPC setup. Decay curves were measured at the emission band maxima and deconvoluted using the Picoquant FluoFit software package version 4.6.6.0. Laser flash photolysis in DMSO was used to determine the τ T values at 500 nm with an Edinburgh Instruments LP980 instrument by fitting an exponential to the decay curve. An Ekspla NT-342B laser (2.0 mJ, 7 ns and 20 Hz) provided a probe beam of ca. 430 nm at the B band maxima. The Φ ∆ values were determined using a comparative method in DMSO with DPBF as the scavenger and Rose Bengal as a standard (Φ ∆ = 0.76 in DMSO [50]). An Ekspla NT-342B laser with an OPO provided monochromatic light at a spectral crossover between the standard and the sample. Photostability tests were carried out in a 1 × 1 cm quartz cuvette, and the solutions were irradiated with a 660 nm (280 mW·cm −2 ) Thorlabs LED lamp mounted onto a Modulight 7710-680 medical laser system housing in a similar manner to the PDT and PACT activity measurements. The 1% DMSO/H 2 O solutions were prepared in the dark by dissolving 1 mg of sample in 6 mL of solvent. Photobleaching of the dyes at the B band maxima was determined by UV-visible absorption spectroscopy. The percentages of the initial B band intensities observed after 30 min of photoirradiation are provided in Table 2.

Synthesis of Free-Base Chlorins
1-3-Por and 2-Chl have been reported previously and were prepared according to literature procedures [23,24]. 1-Chl was synthesized following a procedure described in the literature [25] (Scheme 1). 5,10,15,20-tetrakis(3-methoxyphenyl)chlorin (1-Chl) was synthesized in this study using 1-Por (74 mg, 1 mmol) dissolved in dry pyridine (40 mL), and K 2 CO 3 (8 mmol) was added and stirred. The mixture was then refluxed in an inert nitrogen environment, and the same amount of p-toluenesulfonyl hydrazide (4 mmol) was added at 3 h intervals for 12 h. The reaction progress was monitored regularly using TLC and UV-visible absorption spectroscopy and stopped when porphyrin peaks disappeared, and the reduced bacteriochlorin peak emerged at ca. 730 nm. The reaction mixture was cooled to room temperature, extracted with CHCl 3 , washed three times with 0.1 M HCl solution, and lastly with Millipore water to remove pyridine and salts. The organic layer was dried over anhydrous sodium sulfate, and p-chloranil was added until the band at 730 nm disappeared. The mixture was filtered, and the organic solvent was evaporated. The crude product was purified by using silica gel chromatography with 3:1 CH 2 Cl 2 /petroleum ether eluent to afford the 1-Chl target compound as a purple solid. Yield: 31 mg (43%). 3-Chl was synthesized in a similar manner to that described for 1-Chl (Scheme 1). 5,10,15,20-Tetra(4-hydroxy-3-methoxyphenyl)chlorin 3-Chl was obtained in 35% yield.

PDT Studies
MCF-7 cells were cultured as previously described [51] in DMEM containing 4.5 g·L −1 glucose with L-glutamine and phenol red. The media was supplemented with 10% (v/v) heat-inactivated FCS and 5% 100 unit mL −1 penicillin 100 µg·mL −1 streptomycin amphotericin B. The cells were grown in a T75 cm 2 vented flask (Porvair) and incubated in a humidified 5% CO 2 atmosphere at 37 • C until 80% confluence was achieved. The cells were rinsed with DPBS before routine standard trypsinization during subculturing and cell harvesting. The trypsinized cells were treated with trypan blue dye exclusion assay (0.40% trypan blue solution), and a hemocytometer was used to count viable cells. Cells were seeded at a density of 10,000 cells/well in 96-well plates in supplemented red-phenol DMEM medium and incubated under a humidified 5% CO 2 for 24 h at 37 • C to enable cell attachment to the wells. The attached cells were then rinsed twice using 100 µL DPBS, and subsequently, 100 µL of the drugs (the studied compounds) in the DMEM medium were administered over gradient concentration ranges. The drug stock solutions were prepared in DMSO since they are insoluble in aqueous media. After dilution, appropriate drug aliquot volumes were prepared in DMEM media such that the highest gradient concentrations consisted of <1% (v/v) DMSO. The control cells were supplemented with DMEM medium alone, with the effect of DMSO on the cells investigated by incubating the cells in 1% (v/v) DMSO-DMEM media. The plates were then re-incubated for 24 h in the dark for further studies that include in vitro dark cytotoxicity and PDT activity studies.
The MCF-7 cancer cells treated with the different drug concentrations were rinsed thrice with 100 µL DPBS to remove any residual drug. DMEM media was added with no red phenol. For in vitro dark cytotoxicity studies, the drug-treated cells were not photoirradiated, while for light studies, the treated cells were photoirradiated with a M625L3 or M660L4 Thorlabs LED mounted onto a Modulight 7710-680 medical laser system housing. The excitations of the PS dyes were performed at the red end of the visible since the dyes used in this work are intended for use in the treatment of deepseated infected soft tissues and tumors [52,53]. Both Thorlabs M625L3 and M660L4 LEDs were used for 20 min photoirradiations of cells treated with 1-3-Chl and 1-3-SnChl. The DMEM was replaced with red phenol, and the cells were incubated further for 24 h. Cell viabilities were determined using the MTT assay protocols following the manufacturer's specifications by measuring the absorbances of the assays at 540 nm on a Synergy 2 multimode microplate reader (BioTek ® ). All experiments were carried out in triplicate, with the data analyzed statistically using Student's t-test and ANOVA. The percentage cell viabilities were calculated as the percentage ratio of the absorbance of the drug-treated cells against the untreated controls [51], as described in Equation (1).

% Cell Viability =
Absorbance of samples at 540 nm Absorbance of control at 540 nm × 100 (1) Since the IC 50 value is another important parameter for determining drug efficacy, it was calculated for the dyes by nonlinear regression analysis using GraphPad Prism 5. During in vitro PDT studies, the IC 50 value provides a half-maximal inhibitory concentration of a PS dye for inhibiting the growth or killing 50% of cancer cells after drug incubation and light treatment.

Lipophilicity Studies
The lipophilicity tests were carried out in triplicate for each dye complex studied by using the "shake-flask" method [54]. The sample solutions were prepared using 0.5 mg of each dye complex dissolved in 10 mL of dry CHCI 3 . For each 3 mL CHCl 3 solution of the complex, the absorbance at the Soret band maxima (A o ) was measured. Then, 3 mL of Millipore water was added to each 3 mL CHCl 3 solution, followed by stirring of the mixture at room temperature for 4 h. Thereafter, the mixture was centrifuged at 5000 rpm for 10 min to allow the separation of the water and CHCl 3 phases. Then, the absorbance of the CHCl 3 layer (A f ) was measured for each complex mixture, with the value for the water phase determined from the difference between the A o and A f values of the dye complex in CHCl 3 . The partition coefficient values (Log P o/w ) for the complexes in CHCl 3 :H 2 O were derived using Equation (2) [54]. P octanol = [1.343 + Log P chloroform ]/1.126 (2)

PACT Studies
The PACT activities of the PS dyes were carried out against planktonic and biofilms cells of Gram-(+) S. aureus (ATCC ® 25923 TM ) and Gram-(−) E. coli bacteria (ATCC ® 25922 TM ) strains. Photoinactivation experiments were performed as described in the literature with slight modifications [55][56][57]. The viable bacteria colonies were estimated using the direct viable colony count and by an indirect method with the crystal violet staining assay. The planktonic bacteria cells were estimated using the direct method only by counting the viable colonies formed on the agar plates using a bacteria counter, as described previously in the literature [55][56][57], while for biofilms cells, both the viable colony count and indirect methods were used [35,[58][59][60]. The biofilm biomass was quantified indirectly via the crystal violet assay by measuring the absorbance intensity at 590 nm on a Synergy 2 multi-diode microplate reader (BioTek). The PS dye drug stock solutions were first prepared in DMSO since they were insoluble in PBS alone with the appropriate drug aliquots volumes prepared to 1% (v/v) DMSO-PBS solution after serial dilution. Photoinactivation experiments were carried out in the dark, and after photoirradiation with an M625L3 or M660L4 Thorlabs LED mounted onto a Modulight 7710-680 medical laser system housing. For all in vitro dark cytotoxicity and photoinactivation activity studies of the dyes against both the planktonic and biofilm cells of S. aureus and E. coli bacteria strains, the effect of PS dye drug concentrations at a fixed fluence (light) dosage and the effect of different irradiation times at a fixed drug dye concentration were investigated.

Planktonic Bacteria
S. aureus and E. coli bacteria strains were grown anaerobically on agar plates to obtain colonies of each by following the manufacturer's specifications. A single colony of each strain to be studied was inoculated into a 5 mL freshly prepared Lura nutrient broth. The culture mixture was vortexed and incubated at 37 • C with agitation (ca. 200 rpm) in a rotary shaker incubator for 18 and 48−72 h for S. aureus and E. coli, respectively. Aliquots from the bacteria culture mixtures were taken regularly to measure their optical densities until the bacterial growth reached a mid-logarithmic phase (OD 620 nm, 0.6−0.8). The bacteria pellets were then harvested through centrifugation for 15 min at 3000 rpm and washed three times with PBS to remove residual nutrient broth. The bacteria pellets were re-suspended in 4 mL PBS and further diluted to 1:1000 (v/v) in PBS to obtain the working bacteria culture stock solutions. The viable colonies count of the freshly prepared S. aureus and E. coli bacteria cultures were determined by serial dilution of the bacteria culture stock solutions by factors of 10 −4 , 10 −5 , 10 −6 , 10 −7 , 10 −8 , and 10 −9 . A 100 µL aliquot of each sample solution was aseptically inoculated on the agar plates in triplicate and incubated at 37 • C for 24 h to determine the optimum bacteria count. The viable bacteria colonies were counted on a Scan 500 ® series Automatic Colony Counter. Optimized bacteria solutions of colonyforming units (CFU/mL) ranging from 2.78 to 3.01 × 10 8 and 1.62 to 2.01 × 10 8 CFU/mL for S. aureus and E. coli, respectively, were used in the subsequent studies.
The experimental procedure followed for sample preparation for in vitro dark and light studies of the dyes for photoinactivation of planktonic S. aureus and E. coli bacteria strains was the same. For time-dependence studies, a 5 mL bacteria suspension (10 8 CFU/mL) of S. aureus or E. coli bacteria strains in 1% DMSO/PBS solution was incubated with appropriate concentrations of the porphyrinoid dye drugs at 37 • C in a rotary shaking incubator at 200 rpm for 30 min in the dark. Half (2.5 mL) of the drugincubated bacteria suspension was transferred into a 24-well plate for light studies, and the remaining half (2.5 mL) was kept in another 24-well plate in the dark for in vitro dark cytotoxicity studies.
During light studies, the drug-treated bacteria suspensions were irradiated with M625L3 or M660L4 Thorlabs LEDs mounted onto a Modulight 7710-680 medical laser system housing over different time intervals. For 1-3-Chl and 1-3-SnChl, 1 µM of the dye was administered against S. aureus and 5 µM against E. coli bacteria suspensions. For the bacteria suspensions treated with 1-3-Chl and 1-3-SnChl, both Thorlabs M625L3 and M660L4 LEDs were used for photoirradiation at 5, 10, 15, 30, 45, 60, and 75 min intervals. After light and dark treatments at the considered time intervals, a 100 µL aliquot of the samples was aseptically inoculated on the agar plate, which was incubated for 18 h at 37 • C. Viable bacteria colonies were counted with a Scan 500 ® series Automatic Colony Counter to determine colony-forming unit (CFU/mL) values. The controls were S. aureus and E. coli bacteria suspensions (colony-forming units ≈ 10 8 CFU/mL) with no PS dye drug. The cell survival fractions were calculated by comparing the drug-treated bacteria with the control. The Log 10 reductions were calculated using Equation (3): A and B are the number of viable colonies (CFU/mL) of bacteria for the untreated and treated samples, respectively. All the experiments were carried out in three independent triplicates, and data were analyzed statistically using student t-test and ANOVA. Prior to time-dependence studies, the effect of gradient concentrations of the dyes was analyzed at fixed light intensity (fluence) to establish the optimal concentrations for study. For bacteria suspensions treated with 1-3-Chl and 1-3-SnChl, gradient concentrations of 0.5, 1, 1.5, 2, 2.5, and 5 µM were administered against S. aureus, and 1, 2.5, 5, 10, and 15 µM against E. coli with photoirradiation carried out for 45 min using a Thorlabs M625L3 LED mounted onto a Modulight 7710-680 medical laser system housing.

Biofilm Bacteria
The freshly prepared planktonic S. aureus and E. coli bacteria cultures were diluted to 1 × 10 6 CFU/mL in Tryptic Soy Broth (TBS). The TBS-supplemented bacteria cultures (200 µL/well) were seeded into the 96-well plates and incubated anaerobically under static conditions at 37 • C with 5% CO 2 for 24 and 72 h for S. aureus and E. coli, respectively, to form S. aureus and E. coli single-species bacteria biofilms. The wells with media alone provided the negative controls. After incubation, the culture media was discarded, and the wells were rinsed gently twice with PBS to remove the residual TBS and the non-adherent planktonic bacteria cells. The plates were air-dried at room temperature for 30 min to fix the adherent cells.
The biofilm bacteria biomass was quantified using the crystal violet assay. First, 150 µL/well of an aliquot of 0.1% (w/v) of crystal violet was used to stain the biofilmcoated wells for 15 min at room temperature. The crystal violet was then discarded, and the wells were washed three times using 200 µL PBS/well to remove the residual dye. 150 µL/well of PBS was again added to the stained wells, and the absorbance of well solutions was measured at 590 nm on a Synergy 2 multi-mode microplate reader (BioTek ® ). All experiments were carried out in triplicate, and the Z-score method was used to identify outlier values. Non-outlier values were used to calculate the average biofilm formed. The average negative control value was subtracted from the obtained biofilms absorbance data, and the data were provided as mean absorbance ± standard deviation. Biofilms were considered to have formed when the absorbance value at 590 nm was three times the standard deviation of the negative control mean absorbance [61]. All experimental data were analyzed statistically using a student t-test and ANOVA. Viable cell counts for the single species S. aureus and E. coli biofilms were also estimated by the viable colony count method. The biofilms formed in the 96-well plates were solubilized by adding 200 µL/well of PBS and sonicating the plates vigorously for 10 min, followed by scraping of the wells to dislodge the adherent cells. The resultant well suspension was vortexed to homogenize the solution and serial diluted 10-fold times, and a 100 µL aliquot of the diluted sample was then aseptically inoculated on the agar plate in triplicates followed by incubation at 37 • C for 18 h. The viable colonies count (CFU/mL) was carried out on a Scan 500 ® series Automatic Colony Counter. All the experimental data were analyzed statistically with a student t-test and ANOVA.
Initially, the effect of the various dye concentrations on inhibiting the growth of the single-species biofilms of S. aureus and E. coli strains experiments were carried out. For these concentration studies, the single species biofilms of S. aureus and E. coli strains were incubated with 100 µL/well aliquots of the dye series at various concentrations for 30 min. During the light studies, the drug-incubated 96-well plate biofilms were photoirradiated with a 625 or 660 nm Thorlabs LED for 30 min, while for the dark studies, the drug-treated plates were kept in the dark (in the absence of irradiation) for 30 min to determine the dark cytotoxicities of the drugs. For 1-3-Chl and 1-3-SnChl, the single-species S. aureus and E. coli biofilms were treated with drug concentrations of 25, 50, 100 and 200 µM with both M625L3 and M660L4 Thorlabs LEDs. The controls were the biofilm-coated wells with PBS media alone (untreated biofilms). After the dark and light treatments, the supernatant liquid was discarded, and the wells were gently rinsed with PBS. The biofilm cell densities were then quantified using crystal violet assay by measuring the absorbance of the assay at 590 nm on a Synergy 2 multi-mode microplate reader (BioTek ® ), and also by using the viable colony count method as previously described in Section 4.7.1. For time-dependence studies, 1-3-Chl and 1-3-SnChl, 25 and 50 µM of the dyes were administered against S. aureus and E. coli biofilms, respectively. Both Thorlabs M625L3 and M660L4 LEDs were used for the photoirradiations. The ability of the dyes to inhibit the single species S. aureus and E. coli biofilms was quantified using the crystal violet assay and viable colonies count methods as previously described, and the data were compared with the controls. All experiments were performed with three independent triplicates and compared with the control, and data were analyzed statistically with Student's t-test and ANOVA.

Conclusions
The photoexcitation of chlorin dyes with 3-methoxy-, 4-hydroxy-and 3-methoxy-4hydroxyphenyl meso-aryl rings (1-3-Chl) at the red end of the electromagnetic spectrum results in favorable PDT activities against MCF-7 breast cancer cells and PACT activities against Gram-(+) S. aureus and Gram-(−) E. coli bacterial strains. Sn(IV) chlorin complexes 1-3-SnChl have more favorable PDT and PACT activities than the corresponding free base 1-3-Chl dyes. This is probably due to their high singlet oxygen quantum yields and decreased aggregation due to trans-axial ligation, which favorably enhances the PDT and PACT activities of the chlorin dyes. The data demonstrate that readily synthesized tetraarylchlorins and their metal complexes merit further in-depth study for use as PS dyes in biomedical applications. It is particularly noteworthy that high Log 10 reduction values >3.0 were obtained for 1-3-SnChl against Gram-(−) E. coli despite the absence of cationic charges in the structures. Data of this type have only previously been reported for tetrarylchlorins and their Sn(IV) complexes [10,15,17] with sulfur-atom-containing thien-2-yl, 5-bromothien-2-yl, tetramethylthiophenyl meso-aryl rings.