A Comparative Evaluation of the Photosensitizing Efficiency of Porphyrins, Chlorins and Isobacteriochlorins toward Melanoma Cancer Cells

Skin cancer is one of the cancers that registers the highest number of new cases annually. Among all forms of skin cancer, melanoma is the most invasive and deadliest. The resistance of this form of cancer to conventional treatments has led to the employment of alternative/complementary therapeutic approaches. Photodynamic therapy (PDT) appears to be a promising alternative to overcome the resistance of melanoma to conventional therapies. PDT is a non-invasive therapeutic procedure in which highly reactive oxygen species (ROS) are generated upon excitation of a photosensitizer (PS) when subjected to visible light of an adequate wavelength, resulting in the death of cancer cells. In this work, inspired by the efficacy of tetrapyrrolic macrocycles to act as PS against tumor cells, we report the photophysical characterization and biological assays of isobacteriochlorins and their corresponding chlorins and porphyrins against melanoma cancer cells through a photodynamic process. The non-tumoral L929 fibroblast murine cell line was used as the control. The results show that the choice of adequate tetrapyrrolic macrocycle-based PS can be modulated to improve the performance of PDT.


Introduction
Worldwide, the cancer incidence rate has increased, being one of the main causes of death. Currently, some means of treating cancer include surgery in the case of localized solid tumors, and radiotherapy and chemotherapy (for non-localized tumors) [1]. The development of new therapeutic agents that are selective, efficient, and cause minimal damage to the patient has become one of the central challenges of medicinal chemistry. Melanoma is a type of skin cancer that originates in melanocytes [2], which are found in the basal layer of the epidermis [2,3]. The resistance of this form of cancer to conventional treatments has led to the employment of some alternatives [3].
Photodynamic therapy (PDT) is a therapeutic methodology that requires the combination of a photosensitizer (PS), dioxygen ( 3 O 2 ), and light which, under specific conditions, generate reactive oxygen species (ROS) that are highly cytotoxic, inducing a reduction in the viability of cancer cells [1,[4][5][6]. This is a therapeutic approach that has been pointed out by the scientific community as a promising complement or alternative to the conventional anticancer approaches, such as surgery or radiotherapy. PDT is a non-invasive therapy with high selectivity for cancer cells and reduced side effects due to its controllability and high spatiotemporal approach.
Porphyrins and related compounds are known for their excellent performance as photosensitizers (PS) in PDT [7][8][9][10][11][12][13]. Currently, applied PS are often based on porphyrins, chlorins, In the present work, we decided to evaluate and compare how the efficiency of isobacteriochlorin, chlorin, and porphyrin analogues against melanoma cancer cells would be affected by the type of meso-substituted scaffold selected-A4-type versus A3Btype. For this, Por1, Chl1, and Iso1 were chosen as examples of meso-substituted A4-type macrocycles while for the A3B-type series, Por2, Chl2, and Iso2 were selected. To investigate that influence, the photophysical/photochemical and biological features of both meso-substituted series A4-type and A3B-type derivatives were performed. The carried-out assays allowed us to compare how the photophysical and biological properties of the studied derivatives are modulated by inducing changes in one of the substituents at the meso-position. This comparative study can be a driving force for further studies using both the PS design and PDT assays as targets.

Photosensitizers: Synthesis and Characterization
5,10,15,20-Tetrakis(pentafluorophenyl)porphyrin (Por1) and 5,10,15tris(pentafluorophenyl)-20-(4-pyridyl)porphyrin (Por2) were obtained directly by condensation of pyrrole with the adequate aldehydes in acidic conditions, according to previously described procedures [32,36,37]. The pyrrolidine-fused chlorins (Chl1 and Chl2) and isobacteriochlorins (Iso1 and Iso2) were prepared throughout the 1,3-dipolar cycloaddition of the appropriate porphyrin (Por1 or Por2) and the azomethine ylide generated from N-methylglycine and paraformaldehyde ( Figure 1). The reduced derivatives were attained in yields similar to those reported in the literature (70% for Chl1 and 49% for Chl2; 18% for Iso1 and 15% for Iso2). The structures of all the compounds were confirmed by 1 H and 19 F NMR, UV-Vis spectroscopy, and mass spectrometry and agreed with the data reported in the literature [32,33,[37][38][39][40]. The absorption, emission, and excitation spectra of porphyrins (Por1,2) and their reduced analogs, chlorins (Chl1,2), and isobacteriochlorins (Iso1,2), were acquired in N,Ndimethylformamide (DMF) and are summarized in Table 1. The photophysical characterization of Por1 and Por2 was assessed since they are the chemical precursors of the reduced derivatives Chl1, Chl2, Iso1, and Iso2. The absorption spectra of Por1 and Por2 exhibit the typical features of free-base meso-substituted porphyrin derivatives [41], displaying an intense Soret band at 410 and 412 nm, respectively, due to S 0 →S 2 transitions, The absorption, emission, and excitation spectra of porphyrins (Por1,2) and their reduced analogs, chlorins (Chl1,2), and isobacteriochlorins (Iso1,2), were acquired in N,N-dimethylformamide (DMF) and are summarized in Table 1. The photophysical characterization of Por1 and Por2 was assessed since they are the chemical precursors of the reduced derivatives Chl1, Chl2, Iso1, and Iso2. The absorption spectra of Por1 and Por2 exhibit the typical features of free-base meso-substituted porphyrin derivatives [41], displaying an intense Soret band at 410 and 412 nm, respectively, due to S 0 →S 2 transitions, followed by four weak Q bands in the visible region ranging from 504 to 638 nm, attributed to the S 0 →S 1 transitions. The Soret band on the absorption spectra of chlorins (Chl1,2) was slightly blue-shifted (ca. 5 nm) when compared with the corresponding porphyrin precursor ( Figure S1). Concerning the Q bands region, significant red shifts (~14 nm) were observed for the two bands at higher wavelengths. The UV-Vis spectrum of both isobacteriochlorins Iso1 and Iso2 shows Soret bands even more blue-shifted (20-25 nm) than those of the corresponding chlorin derivatives Chl1 and Chl2; the last Q bands also show significant red shifts (10-15 nm) when compared with porphyrins. This red shift in the last Q band for both reduced series is of high interest for PDT, since this transition absorbs in the so-called therapeutic window (600-800 nm). As expected, Chl1 and Chl2 display the last Q band at ca. 650 nm with a relatively high intensity when compared with the other Q bands. The pyridyl group at the meso-position of the porphyrin macrocycle did not induce noticeable changes in the absorption features of the free-base porphyrin and reduced derivatives.
Concerning the steady-state emission spectra, Por1, Por 2, and Iso2 showed similar emission spectra with two well-defined bands at 638 and~700 nm. The Iso1 emission spectrum displayed a similar profile; however, the first vibrational mode of the fluorescence was blue-shifted 38 nm to 600 nm. Both chlorin derivatives Chl1 and Chl2 only had one emission band at 654 and 649 nm, respectively ( Figure S2). The emission spectra of the studied compounds are typical of free-base meso-tetraarylporphyrins and of their reduced derivatives [42][43][44].
The fluorescence quantum yields (Φ F ) ( Table 1) were determined in DMF using mesotetraphenylporphyrin (TPP) as the standard (Φ F = 0.11 in DMF) [45,46]. The fluorescence properties of each compound type appeared quite similar, which is consistent with the similarity of their chemical structure. The higher fluorescence quantum yields observed for Chl1, Chl2, Iso1, and Iso2, if compared to Por1 and Por2, can be ascribed to the differences in their electronic structures and molecular geometries. Chlorins and isobacteriochlorin generally have a more distorted and non-planar structure compared to porphyrins [47]. This distortion can lead to a decrease in the non-radiative relaxation pathways, such as vibrational relaxation and internal conversion, thereby enhancing the fluorescence efficiency. Compound Iso2 shows the highest fluorescence quantum yield (Φ F = 0.21), followed by Chl2 (Φ F = 0.16), Chl1 (Φ F = 0.15), and finally, Iso1 (Φ F = 0.13). In fact, all the isobacteriochlorin and chlorin-type derivatives presented a greater fluorescence quantum yield (Φ F ) than TPP (0.11). However, opposite behavior was observed for the starting porphyrins Por1 and Por2, which showed a lower fluorescence quantum yield (Φ F ) than TPP (respectively, 0.01 and 0.06 versus 0.11). The fluorescence lifetime of a singlet state (τ) found for porphyrins varied between 10 ns and 11.1 ns. The lifetime reduction observed for Chl1 (6.02 ns) and Chl2 (6.90 ns) when compared to Por1 and Por2 can be explained by the presence of a reduced pyrrole-type ring, leading to a reduction in the π-conjugation effect and improving the macrocycle distortion in the excited state, thus resulting in a decrease in radiative decay rates [48]. The emission decay profile for Iso1 and Iso2 derivatives can be fitted ( Figure S3) by two decay components. For Iso1, we have 5.4 ns (~97%) and 1.2 ns (~3%), whereas the short-lived component for Iso2 had a contribution of 77%, and the contribution of the long-lived component was lower when compared to Iso1. The reduction observed for the fluorescence lifetimes of isobacteriochlorins Iso1 and Iso2 can be ascribed to the presence of an additional reduced pyrrolic-type ring at the macrocycle core. The increase in the number of reduced pyrrolic units conducts, in general, to an increase in the conformational flexibility (lower rigidity) of the macrocycle, affecting the electronic structure. Pyrrolidine-type units should decrease the π-conjugation effect, and they may improve the macrocycle distortion in the excited state, contributing to the reduction observed for the fluorescence lifetimes of isobacteriochlorins Iso1 and Iso2 and of chlorins Chl1 and Ch2, leading to a fast internal conversion.
The long-lived triplet state's (τ T ) profile decay of the studied compounds range between 0.50 and 0.98 µs. For Por2, bearing a pyridyl group at one of the meso-positions, the value of the triplet lifetime is lower than the one displayed by Por1, with no pyridyl substituent (0.76 versus 0.98 µs). However, for the reduced derivatives, a different behavior is observed. Chl2 (0.74 µs) and Iso2 (0.63 µs) display higher triplet excited state lifetime values when compared with the corresponding analogs Chl1 (0.51 µs) and Iso1 (0.50 µs). The remarkable dependence of the triplet lifetime as a function of the substituent and its position is well established in the literature [44,49,50].
The efficiency of a PS in PDT is also related to its ability to generate reactive oxygen species (ROS), mainly singlet oxygen ( 1 O 2 ). ROS acts as signaling molecules, but they can also promote cellular damage by rapidly oxidizing cellular components. The measurement of the 1 O 2 quantum yield (Φ ∆ ) was assessed by the luminescence method, measuring the 1 O 2 phosphorescence at 1270 nm ( Figure 2) and using ZnPc (zinc(II) phthalocyanine) as the reference. It is worth noting that this approach is not dependent on the dye concentration but only on the number of photons absorbed [51]. In general, the Φ ∆ increased when the pyridyl group was attached to the macrocycle backbone (except for Iso2) in the following order: Table 1). Additionally, the triplet lifetime (τ T ) increased for chlorin-type derivatives as a function of the substituent, which consequently increased the Φ ∆ [52] . The presence of halogens influences the ability to generate 1 O 2 [53], photostability, and photophysical features [16,19]. The differences observed for the Φ ∆ are probably related to the macrocycle planar distortion in Por1, Chl1, and Iso1 derivatives. The presence of halogen atoms leads to an improved intersystem crossing from the photosensitizer s excited singlet and triplet states [53]; however, for the evaluated compounds, the presence of more halogen atoms did not favor 1 O 2 generation of Por1 and its reduced analogs (Chl1 and Iso1) when compared with those bearing a pyridyl unit (Por2, Chl2, and Iso2). This is undoubtedly related to the lower amount of T 1 states quenched by 3 O 2 due to the occurrence of competing processes, reducing the formation of 1 O 2 . Probably, different restrictions in the aryl ring rotations ascribed to the presence of three C 6 F 5 and one pyridyl unit, higher asymmetry, and different electronic effects contributed to improve the ability of Por2, Chl2, and Iso2 to generate 1 O 2 when compared with the corresponding Por1 and reduced analogs Chl1 and Iso1 with four C 6 F 5 substituents [54]. Singlet oxygen generation is strongly related to the probability of molecular dioxygen ( 3 O 2 ) colliding with the PS in the triplet state, which can be affected by steric effects, often associated with a decrease in the amount of the excited triplet state. Data in Table 1 agree with this and can be related to the presence of the pyrrolidinefused rings on both reduced derivatives, resulting in a decrease in collision frequencies between the macrocycles and 3 O 2 . The high Φ ∆ of Chl2 seems to indicate higher success in the collisions involving 3 O 2 and this reduced derivative. The best 1 O 2 generator was Chl2 (Φ ∆ = 0.81); however, all the compounds studied are able to generate this highly cytotoxic species (ROS), which makes them suitable to be used in PDT against cancer cells.

PDT Assays
With the exception of Por1, due to its low solubility in the RPMI/DMSO (1%) medium (vide infra), the ability of all the other derivatives Chl1, Iso1, Por2, Chl2, and Iso2 to act as PS in the PDT assays was investigated, considering their efficiency to generate 1 O 2 . The cytotoxicity of the compounds was assessed by an in vitro MTT colorimetric assay [55]. The viability of B16F10 murine melanoma cells was evaluated after treatment with Chl1, Iso1, Por2, Chl2, and Iso2 at different concentrations (1.5, 6.2, 25, 50, and 100 µM), without and under red light irradiation (660 nm; 5.4 J/cm 2 ) (Figures 3 and 4). A non-tumoral fibroblast L929 murine cell line was used as the control.
Molecules 2023, 28, x FOR PEER REVIEW 6 of 16 decrease in collision frequencies between the macrocycles and 3 O2. The high ΦΔ of Chl2 seems to indicate higher success in the collisions involving 3 O2 and this reduced derivative. The best 1 O2 generator was Chl2 (Φ = 0.81); however, all the compounds studied are able to generate this highly cytotoxic species (ROS), which makes them suitable to be used in PDT against cancer cells.
The results show that in general, the compounds displayed cytotoxicity under nonirradiated conditions (dark) against both melanoma cells and non-tumoral cells at 25 μM or higher concentrations. However, in general, photo-stimulation increased their cytotoxicity. Of all the PS studied, globally speaking, Chl1 was the least cytotoxic ( Figure  3A,B) in the dark at 100 μM (38% and 31% cell death for L929 and B16F10, respectively). Both chlorins were effective toward B16F10 at 25 μM under red light irradiation (660 nm) with a reduction in the cell viability of 43% (Chl1) and 42% (Chl2), taking into account their toxicity under dark conditions ( Figures 3B and 4D). This phototoxicity improves at higher concentrations, attaining cell deaths of ca 64% at 100 μM, considering also the cytotoxicity under dark conditions. Despite the efficiency of both chlorins, these PS displayed small selectivity for B16F10 cells, causing the death of the non-tumoral cells ranging from 30% to 35% (Figures 3A and 4C).
Although both isobacteriochlorins induced a reduction in tumoral cells, Iso2 was more efficient and selective for B16F10 cells (p < 0.001) ( Figure 4E,F) than Iso1 ( Figure  3C,D). For instance, Iso1 at 25 μM displayed cytotoxicity under both dark (~36%) and light conditions (~49%) toward B16F10 cells and similar phototoxicity toward the non-tumoral cells (cell death of ~50%). Although there was less cytotoxicity for non-tumoral cell lines, the photocytotoxicity was more relevant than the cytotoxicity at higher concentrations. No improvement in the selectivity toward B16F10 cells was observed for isobacteriochlorin Iso1. Moreover, Iso2 at 25 μM presented lower cytotoxicity in the dark and was able to cause 72% of cell death of the treated melanoma cells, and no significative cell death (20%) was achieved in non-tumoral cell lines under similar conditions. Therefore, when comparing porphyrin and chlorin-type derivatives using the same concentration, chlorins seem to be more efficient PS after being irradiated with red light (660 nm) due to their strongest absorption in the red region. In general, the more photocytotoxic compounds contain the pyridyl group in their structure (Chl2 and Iso2). In fact, there is a strong relationship between structure and activity. The efficiency of Chl2 can be attributed to its photophysical and photochemical properties, mainly the production of 1 O2, but no selectivity for B16F10 cancer cells was observed; this PS was also able to induce a significant reduction in the non-tumoral fibroblast L929 murine cell line. Additionally, its dark activity can be due to the formation of adducts with biological structures that affect the cell survival [56], since the production of ROS is limited. Therefore, cellular uptake and cellular sublocalization should also be considered. . The results are presented as mean ± standard deviation. Significant differences relative to control cell cultures are presented with an # and relative to irradiated and non-irradiated by *. Statistical significance: # p < 0.05, * p < 0.05, *** p < 0.001, and **** p < 0.0001.
Several factors can influence the PS activity; the formation of aggregates is the feature that contributes the most to the decrease in its efficiency, due to the limitation of ROS production and cellular uptake [57,58]. Aggregation studies were performed in order to understand the potential effect of the porphyrin-type scaffold selected (meso-substituted-A4-type versus A3B type) on the cytotoxicity observed. UV-Vis studies assessing different biological PS concentrations in RPMI medium and DMSO (1%) were performed (see ESI, . The results are presented as mean ± standard deviation. Significant differences relative to control cell cultures are presented with an # and relative to irradiated and non-irradiated by *. Statistical significance: # p < 0.05, * p < 0.05, *** p < 0.001, and **** p < 0.0001.
Iso2. Por1 showed low solubility in biological conditions; so, it was decided not to evaluate this compound in the further PDT assays. The aggregation studies showed that the pyridyl group increased the PS solubility of Por2, but no noticeable changes were observed for Chl2, although leading to a decrease in the solubility of derivative Iso2 when compared with the corresponding A4-type derivative.
The results show that in general, the compounds displayed cytotoxicity under nonirradiated conditions (dark) against both melanoma cells and non-tumoral cells at 25 µM or higher concentrations. However, in general, photo-stimulation increased their cytotoxicity. Of all the PS studied, globally speaking, Chl1 was the least cytotoxic ( Figure 3A,B) in the dark at 100 µM (38% and 31% cell death for L929 and B16F10, respectively). Both chlorins were effective toward B16F10 at 25 µM under red light irradiation (660 nm) with a reduction in the cell viability of 43% (Chl1) and 42% (Chl2), taking into account their toxicity under dark conditions (Figures 3B and 4D). This phototoxicity improves at higher concentrations, attaining cell deaths of ca 64% at 100 µM, considering also the cytotoxicity under dark conditions. Despite the efficiency of both chlorins, these PS displayed small selectivity for B16F10 cells, causing the death of the non-tumoral cells ranging from 30% to 35% (Figures 3A and 4C).
Although both isobacteriochlorins induced a reduction in tumoral cells, Iso2 was more efficient and selective for B16F10 cells (p < 0.001) ( Figure 4E,F) than Iso1 (Figure 3C,D). For instance, Iso1 at 25 µM displayed cytotoxicity under both dark (~36%) and light conditions (~49%) toward B16F10 cells and similar phototoxicity toward the non-tumoral cells (cell death of~50%). Although there was less cytotoxicity for non-tumoral cell lines, the photocytotoxicity was more relevant than the cytotoxicity at higher concentrations. No improvement in the selectivity toward B16F10 cells was observed for isobacteriochlorin Iso1. Moreover, Iso2 at 25 µM presented lower cytotoxicity in the dark and was able to cause 72% of cell death of the treated melanoma cells, and no significative cell death (20%) was achieved in non-tumoral cell lines under similar conditions. Therefore, when comparing porphyrin and chlorin-type derivatives using the same concentration, chlorins seem to be more efficient PS after being irradiated with red light (660 nm) due to their strongest absorption in the red region. In general, the more photocytotoxic compounds contain the pyridyl group in their structure (Chl2 and Iso2). In fact, there is a strong relationship between structure and activity. The efficiency of Chl2 can be attributed to its photophysical and photochemical properties, mainly the production of 1 O 2, but no selectivity for B16F10 cancer cells was observed; this PS was also able to induce a significant reduction in the non-tumoral fibroblast L929 murine cell line. Additionally, its dark activity can be due to the formation of adducts with biological structures that affect the cell survival [56], since the production of ROS is limited. Therefore, cellular uptake and cellular sublocalization should also be considered.
Several factors can influence the PS activity; the formation of aggregates is the feature that contributes the most to the decrease in its efficiency, due to the limitation of ROS production and cellular uptake [57,58]. Aggregation studies were performed in order to understand the potential effect of the porphyrin-type scaffold selected (meso-substituted-A 4type versus A 3 B type) on the cytotoxicity observed. UV-Vis studies assessing different biological PS concentrations in RPMI medium and DMSO (1%) were performed (see ESI, Figures S4-S8). It was noticed that the PS strictly followed the Beer-Lambert law to established concentrations, thus suggesting no aggregation in RPMI at concentrations below 3 µM for Chl1, 12.5 µM for Iso1, 6.0 µM for Por2, 3.0 µM for Chl2, and 3.0 µM for Iso2. Por1 showed low solubility in biological conditions; so, it was decided not to evaluate this compound in the further PDT assays. The aggregation studies showed that the pyridyl group increased the PS solubility of Por2, but no noticeable changes were observed for Chl2, although leading to a decrease in the solubility of derivative Iso2 when compared with the corresponding A 4 -type derivative.
As expected, more fluorine atoms increase the lipophilicity of the PS, which is corroborated by cellular uptake studies ( Figure S9).
The octanol:water partition coefficients (Log p) were calculated for all the macrocycles using Molinspiration WebME Editor 3.81 [47]. The A 4 -type macrocycles exhibited Log p values between 9.98 and 9.75, while for the A 3 B-type macrocycles, the Log p values ranged from 9.69 to 9.39. Considering the possibility that at physiological pH the pyrrolidine units must be almost entirely in the cationic form (pyrrolidine pKa = 11.3), partition coefficients were also calculated for chlorin and isobacteriochlorin derivatives while taking that into account. A slight decrease was observed for Log p values of both series of protonated compounds, being 9.21 (Chl1) and 9.01 (Iso1) for A 4 -type reduced macrocycles and 8.60 and 7.93 for the corresponding A 3 B-type macrocycles Chl2 and Iso2, respectively. The uptake of Chl1 and Iso1 increased slightly as a function of concentration. On the other hand, both Chl2 and Iso2 showed a reduction in the uptake at 25 µM after 4 h of incubation in B16F10 cells evaluated by fluorescence; if compared to the concentrations of 6.1 µM and 12.5 µM, this behavior can be partially justified by the aggregation phenomena, corroborating the main findings in aggregation studies.
When the stability of the PS in RPMI/DMSO (1%) mixture was evaluated in the dark over 24 h ( Figure S10), a noticeable decrease (~40%) in the stability was observed for Chl1 and Chl2, and even after this long period in the solution, Iso1 and Por2 showed an acceptable stability, with a decrease of around 20%. Iso2 showed the lowest stability in solution, with a decrease of around 45% in the absorption percentage of this PS after the same period. These data show that, under dark conditions, Iso1 and Por2 were the most stable in the RPMI/DMSO (1%) mixture at 3.1 µM after 24 h. When irradiated for 30 min under the same irradiation conditions of biological assays, the studied compounds showed good stability. Apart from Chl2 ( Figure S11D), which displays a noticeable decrease in the absorption intensity at Soret band, the UV-Vis spectra of the other compounds remained almost unchanged ( Figure S11).
To demonstrate the intracellular distribution in B16F10 cells, images of fluorescence microscopy were acquired. Red fluorescence was observed in all cases, which led us to suggest that the PS was efficiently internalized ( Figure 5). The two panels on the left side displayed the PS internalized in the B16F10 cells, with Hoechst staining the nucleus (blue) and Rhodamine staining the mitochondria (green). Chlorins (Chl1 and Chl2) and isobacteriochlorins (Iso1 and Iso2) showed similar fluorescence patterns with the mitochondria probe. The overlapping fluorescence of Chl2 and Iso2 suggests that the compounds are preferentially localized in regions near mitochondria, while Chl1 and Iso1 are apparently spread throughout the cell, including the nucleus.

Generalities
Absorption spectra were obtained on a Shimadzu UV-2501PC spectrophotometer in the 350-800 nm range. The fluorescence spectra were recorded in DMF in 1 cm × 1 cm quartz optical cells under normal atmospheric conditions on a computer-controlled Despite the lack of selectivity for cancer cells (B16F10) for most of the tested PS at high concentrations, we propose to employ metronomic PDT to improve the tumor-specific response. Wilson et al. created the term metronomic PDT, which consists of administering both the PS and light during an extended period at very low doses and over many hours in order to increase the selective reduction on the viability of cancer cells by apoptosis. In their study, the authors aimed to compare standard or acute PDT with metronomic PDT. At the end, they found that metronomic PDT enhanced tumor-specific cell death, while decreasing the harm to adjacent normal tissues [59]. Therefore, herein we suggest that metronomic delivery or several PDT treatments are required to increase the selectivity of the PS we have tested, namely, of the chlorin-type derivatives.

Generalities
Absorption spectra were obtained on a Shimadzu UV-2501PC spectrophotometer in the 350-800 nm range. The fluorescence spectra were recorded in DMF in 1 cm × 1 cm quartz optical cells under normal atmospheric conditions on a computer-controlled F4500-Hitachi spectrofluorometer. The widths of both excitation and emission slits were set at 3.0 nm. To calculate the fluorescence quantum yield (Φ F ), TPP in DMF was used as reference using Equation (1). In this equation, Φ F is the fluorescence quantum yield of the sample; Φ st is the fluorescence quantum yield of TPP (λ exc = 420 nm, Φ F = 0.11 in DMF); A st is the absorbance of TPP and A is the absorbance of the sample at the excitation wavelength; S st and S represent the integrated emission band of the TPP and sample, respectively.
Laser flash photolysis experiments for detection of the triplet state in solution were performed on a system using a Quanta Ray Lab-130 4 Hz Nd:YAG laser at 420 nm from Spectra Physics as the excitation source. The fluorescence lifetime was determined on a Fluorescence Correlation Spectroscopy/Fluorescence Lifetime Imaging coupled to a laser with 420 nm excitation wavelength. Direct measurement of singlet oxygen was performed by luminescence method from Equation (2), as described in the literature [60], using ZnPc as standard (λ exc = 660 nm, Φ ∆ = 0.56 in DMF) [61,62].
where Φ ∆ is the singlet oxygen quantum yield of the sample; Φ st the singlet oxygen quantum yield of ZnPc; I st is the phosphorescence intensity of 1 O 2 at 1270 nm for ZnPc; and I s the phosphorescence intensity for the sample.

Synthesis of the Photosensitizers (PS)
The photosensitizers were synthesized as described in the literature [32,36,37].

Cell Culture
The B16F10 (melanoma murine) and L929 (fibroblast murine) cells were obtained from the American Type Culture Collection. The cell line was cultured in RPMI medium with 10% supplemental fetal bovine fetal serum, 100 IU mL −1 of penicillin G, 100 mg mL −1 of streptomycin, and 1 µg mL −1 of amphotericin. Cells were seeded until 75-90% confluence in 96-well plates and cultured in a humidified incubator at 37 • C with 5.0% CO 2 for 24 h.

Cell Viability Assay
Evaluation of cell cytotoxicity by porphyrin derivatives was performed against two different cell lines: the murine melanoma, B16F10, and murine fibroblast, L929. To this end, 2 × 10 4 cells were incubated for 24 h in 96-well cell culture plates. After this period, the treatments with Chl1, Iso1, Por2, Chl2, and Iso2 that were previously dissolved in DMSO (1.5, 6.2, 25, 50, and 100 µM), then dissolved in the culture medium, and serially diluted to the appropriate concentration to give a final DMSO concentration of 1%, were conducted with and without red light irradiation (emitted by an array of 96 light-emitting diodes (LEDs), λ = 660 nm). After the indicated treatment, the cells were incubated at 37 • C for 3 h in a culture medium containing 10 mol L −1 MTT in RPMI without supplemental fetal bovine fetal serum. The blue MTT formazan precipitate was then dissolved in 50 µL of DMSO, and the absorbance was measured at 480 nm with a multi-well plate reader. The cell viability was expressed as the percentage of the absorption values in the treated cells relative to the non-treated (control) cells. The data are presented as an average of three independent experiments with replicates.

Statistical Analysis
Statistical analysis was performed by two-way ANOVA. Equality of variance was assumed with Bonferroni's post hoc test for pair-wise comparisons. Results with p < 0.05 were considered statistically significant.

Fluorescence Studies
Cellular uptake of porphyrin derivatives by B16F10 cells was performed by fluorescence spectroscopy and microscopy, as described in the literature [32]. The fluorescence was measured with a microplate reader (Spectra Max Paradigm) set at λ exc = 420 nm, matching the Soret band and corresponding emission wavelength to each PS at the initial time = after 4 h of incubation. For fluorescence microscopy, specific organelles' staining probes were used. Mitochondria were stained with Rhodamine 123, and Hoechst 33342 was used as the nuclei probe. After washing two times with PBS, the cells were examined by fluorescence microscopy (Nikon Eclipse Ti Microscope model TI-FL) using the following filters: DAPI (λ exc at 340/380 nm and λ em at 435 to 485 nm), FITC (λ exc at 488 nm and λ em at 505 to 527 nm), and Cy5 (λ exc at 620/660 nm and λ em at 662.5/737.5 nm) for porphyrin derivatives detection.

Stability Studies
The stability of each porphyrin derivative was verified by UV-Vis at regular intervals for up to 24 h under dark conditions. Photostability studies of the porphyrin derivatives were performed by irradiation of a solution of each derivative in RPMI and DMSO (1%) under the same conditions of PDT assays. The solutions were stirred and kept at ambient temperature.

Conclusions
In this work, two different series of neutral tetrapyrrolic macrocycle-based PS were successfully prepared and characterized. All the compounds are able to generate ROS, namely 1 O 2 , which shows their potential to be used as PS in PDT against tumor cancer cells. However, due to the high aggregation observed for the A 4 -type Por1 under the conditions where biological assays were carried out, the PS activity of this compound was not evaluated. So, the PDT efficiency of the A 4 -type reduced derivatives (Chl1 and Iso1) as well as all the A 3 B-type porphyrinic PS prepared (Por2, Chl2, and Iso2) was evaluated toward murine melanoma cells (B16F10 cells) and non-tumor fibroblast murine cell lines (L929).
The cytotoxicity of the compounds evaluated is strongly dependent on their structure, level of reduction in the tetrapyrrolic macrocycle, and concentration. Iso2 and Chl2 were the most efficient PS against B16F10 cancer cells; however, Chl2 also displayed higher (photo)cytotoxicity for non-tumor L929 cells.
Although both chlorins can be considered efficient PS against B16F10 cancer cells, they also showed high (photo)cytotoxicity for non-tumoral L929 cells. A different situation was observed with Iso2, whereby at 25 µM, it displayed good selectivity for B16F10 tumoral cells, being able to induce a cell viability reduction of 72% and only 20% of cell death of non-melanoma L929 cells; however, Iso1, with the same degree of reduction and efficiency to generate 1 O 2 , was cytotoxic under both dark and light conditions. These results confirm how a simple manipulation of the porphyrin core can affect the photodynamic action of this type of reduced PS.
In sum, the results showed that the choice of an adequate tetrapyrrolic macrocycle can be modulated to improve the performance of PDT. The presence of a pyridyl unit affords tetrapyrrolic macrocycle-based PS with improved features to be used in PDT. Additionally, the PDT efficiency was highly dependent on the subcellular localization mechanism and cellular uptake.
This study is a pertinent contribution to both tetrapyrrolic macrocycle-based PS development and PDT fields, providing significant data to modulate the tetrapyrrolic macrocycle core and aiming to fine-tune its photophysical/photochemical properties. In the future, it will help the scientific community to design and better prepare PS with improved features and PDT activity to overcome the challenges of effectively reducing the viability of cancer cell lines.

Conflicts of Interest:
The authors declare no conflict of interest.