Photodynamic Inactivation of Microorganisms Using Semisynthetic Chlorophyll a Derivatives as Photosensitizers

In this study, we describe the semisynthesis of cost-effective photosensitizers (PSs) derived from chlorophyll a containing different substituents and using previously described methods from the literature. We compared their structures when used in photodynamic inactivation (PDI) against Staphylococcus aureus, Escherichia coli, and Candida albicans under different conditions. The PSs containing carboxylic acids and butyl groups were highly effective against S. aureus and C. albicans following our PDI protocol. Overall, our results indicate that these nature-inspired PSs are a promising alternative to selectively inactivate microorganisms using PDI.

PDI has been used in the treatment of diseases caused by various microorganisms, including Gram-positive and Gram-negative bacteria and fungi [19], such as the inactivation of Candida albicans, which causes diseases in patients with low immunity [20,21]. It has also been used in the inactivation of Aedes aegypti mosquito larvae, a vector of the dengue, Zika, and chikungunya arboviruses [22][23][24][25][26]. Furthermore, PDI has gained attention in the inactivation of viruses, which have also shown resistance to drugs, such as an alternative to antiviral treatments against human papillomavirus and hepatitis B virus [27]. Recent studies have also shown the use of PDI for the treatment of severe acute respiratory syndrome caused by the new coronavirus (COVID-19) [28]. As PDI can inactivate DNA-or RNA-based viruses, these studies suggest considerable potential for use in virus photoinactivation in the future [29].
In general, photodynamic reactions require the presence of a photosensitizer, which is activated by light at a specific wavelength, allowing the production of reactive oxygen species (ROS). The main ROS are singlet oxygen, superoxide anions, hydroxyl radicals, and hydrogen peroxide. The quantum yields of each of these ROS depend on the PS and the conditions of the medium. After the formation of ROS, they interact with the target cells, causing death [30][31][32][33][34].
Chlorophyll a is formally a chlorin derivative, with four nitrogen atoms surrounding a central magnesium atom, along with numerous attached side chains and a hydrocarbon chain. Chlorins are excellent photosensitizers, and several synthetic chlorin analogues, such as m-tetrahydroxyphenylchlorin and mono-L aspartyl chlorin e6, have been used. Some substances, such as porphyrins, chlorins, and bacteriochlorins, stand out for their application in PDI: they present selected photophysical characteristics, allowing structural modifications to promote better solubility and amphiphilicity and improve their properties in several treatments [42]. Natural products are sources of inspiration for the development of several drugs [43]. The natural chlorophyll pigments from the cyanobacterium Spirulina maxima are abundant and easy to obtain. According to our protocol, [44] once dried, S. maxima is treated with methanol with 5% sulfuric acid, it gives rise to methyl pheophorbide a, and, after treatment with different amines or molecular oxygen, generates methyl pheophorbide a derivatives, including purpurin-18.
In this study, we propose the diverse semisynthesis of chlorophyll a derivatives, using simple reactions to evaluate these photosensitizers against microorganisms. All the chemical modifications performed with chlorophyll a were also aimed at conferring amphiphilicity to the PS, as this strategy has succeeded well in many approaches for PDT or PDI studies [45][46][47][48].
The obtained PS 2-4 and 6-8, were characterized using NMR, UV-vis, and high-resolution mass spectrometry (HRMS Q-TOFF) and presented absorption bands in the red region. See more details on the characterizations in Supplementary Materials, in which we present our compound band wavelengths and additional literature data (Table  S1). Overall, these photosensitizers were semisynthesized at a low cost because we performed small structural modifications in the natural chlorophyll a, having the methyl-pheophorbide a (1) as a direct and versatile molecular template. With these modifications, we obtained six chlorin derivatives, 2-4 and 6-8, all with absorption bands near 660 nm, different substituents, and amphiphilicity, which are desirable for use in PDT (Figures 1 and 2) [57]. The photostability of all compounds was also checked, showing that the photobleaching was not measurable even after 10 min of irradiation (see Supplementary Materials, Figure S18). The obtained PS 2-4 and 6-8, were characterized using NMR, UV-vis, and highresolution mass spectrometry (HRMS Q-TOFF) and presented absorption bands in the red region. See more details on the characterizations in Supplementary Materials, in which we present our compound band wavelengths and additional literature data (Table S1). Overall, these photosensitizers were semisynthesized at a low cost because we performed small structural modifications in the natural chlorophyll a, having the methyl-pheophorbide a (1) as a direct and versatile molecular template. With these modifications, we obtained six chlorin derivatives, 2-4 and 6-8, all with absorption bands near 660 nm, different substituents, and amphiphilicity, which are desirable for use in PDT (Figures 1 and 2) [57]. The photostability of all compounds was also checked, showing that the photobleaching was not measurable even after 10 min of irradiation (see Supplementary Materials, Figure S18).

Photodynamic Inactivation
The compounds 2-4 and 6-8 were evaluated against three microorganisms-S. aureus, C. albicans, and E. coli-for the inactivation of these microorganisms. First, we investigated the dark toxicity of the chlorophyll derivatives. The microorganisms were incubated for 20 min in the dark with the respective photosensitizer (10 μM) and their mortality was evaluated after 24 h. No mortality was observed after 24 h with only the irradiation of the microorganisms without photosensitizers (30 Jcm −2 ). The inactivation

Photodynamic Inactivation
The compounds 2-4 and 6-8 were evaluated against three microorganisms-S aureus, C. albicans, and E. coli-for the inactivation of these microorganisms. First, w investigated the dark toxicity of the chlorophyll derivatives. The microorganisms wer incubated for 20 min in the dark with the respective photosensitizer (10 μM) and thei mortality was evaluated after 24 h. No mortality was observed after 24 h with only th irradiation of the microorganisms without photosensitizers (30 Jcm −2 ). The inactivatio

Photodynamic Inactivation
The compounds 2-4 and 6-8 were evaluated against three microorganisms-S. aureus, C. albicans, and E. coli-for the inactivation of these microorganisms. First, we investigated the dark toxicity of the chlorophyll derivatives. The microorganisms were incubated for 20 min in the dark with the respective photosensitizer (10 µM) and their mortality was evaluated after 24 h. No mortality was observed after 24 h with only the irradiation of the microorganisms without photosensitizers (30 Jcm −2 ). The inactivation study was then performed with each microorganism. All photosensitizers 2-4 and 6-8 were used in the evaluation of S. aureus (Gram-positive bacteria), and different concentrations of photosensitizer (1 µM and 10 µM) and light fluences (15 Jcm −2 and 30 Jcm −2 ) were utilized. The results obtained from these initial photoinactivation studies ( Figure 3) show that with the increase in the carbon chain (from four to eight C atoms), the photosensitizers presented a decrease in the inactivation levels for the derivatives 2-4 and 6-8. study was then performed with each microorganism. All photosensitizers 2-4 and 6-8 were used in the evaluation of S. aureus (Gram-positive bacteria), and different concentrations of photosensitizer (1 μM and 10 μM) and light fluences (15 Jcm −2 and 30 Jcm −2 ) were utilized. The results obtained from these initial photoinactivation studies ( Figure 3) show that with the increase in the carbon chain (from four to eight C atoms), the photosensitizers presented a decrease in the inactivation levels for the derivatives 2-4 and 6-8. In addition, we observed that the photoinactivation of both S. aureus ( Figure 3) and C. albicans ( Figure 4) was influenced by light fluence and photosensitizer concentration, with increased photoinactivation at high concentrations and fluences. In addition, we observed that the photoinactivation of both S. aureus ( Figure 3) and C. albicans ( Figure 4) was influenced by light fluence and photosensitizer concentration, with increased photoinactivation at high concentrations and fluences.
Evaluating the photoinactivation of S. aureus ( Figure 3), we observed that at 1 µM and 15 Jcm −2 , the methyl pheophorbide derivatives 2-4 did not present relevant inactivation, whereas the purpurin-18 derivatives 6-8 allowed significant inactivation, with a reduction of 3 log using PS 8, 3.5 log with PS 7, and 4 log with PS 6. Maintaining the same concentration of photosensitizer and increasing the dose of light from 15 Jcm −2 to 30 Jcm −2 resulted in the inactivation of microorganisms by derivatives 2-4; however, the photoinactivation promoted by PS 8 was approximately 4 log, and that by derivatives 6 and 7 was approximately 5 log.
It is possible to observe in Figure 3 that at 10 µM, both light doses (15 Jcm −2 and 30 Jcm −2 ) were not very effective in the photoinactivation of S. aureus with photosensitizers 2-4. In contrast, purpurin-18 derivatives 6-8 completely inhibited the growth of these microorganisms, proving that these derivatives are much more effective than those derived from methyl pheophorbide 2-4. Evaluating the photoinactivation of S. aureus (Figure 3), we observed that at 1 µM and 15 Jcm −2 , the methyl pheophorbide derivatives 2-4 did not present relevant inactivation, whereas the purpurin-18 derivatives 6-8 allowed significant inactivation, with a reduction of 3 log using PS 8, 3.5 log with PS 7, and 4 log with PS 6. Maintaining the same concentration of photosensitizer and increasing the dose of light from 15 Jcm −2 to 30 Jcm −2 resulted in the inactivation of microorganisms by derivatives 2-4; however, the photoinactivation promoted by PS 8 was approximately 4 log, and that by derivatives 6 and 7 was approximately 5 log.
It is possible to observe in Figure 3 that at 10 µM, both light doses (15 Jcm −2 and 30 Jcm −2 ) were not very effective in the photoinactivation of S. aureus with photosensitizers 2-4. In contrast, purpurin-18 derivatives 6-8 completely inhibited the growth of these microorganisms, proving that these derivatives are much more effective than those derived from methyl pheophorbide 2-4.
These results suggest that the purpurin-18 derivatives 6-8, due to the presence of a carboxylic acid group in the molecules may facilitate their microorganism uptake, whereas the derivatives of methyl pheophorbide 2-4 have an ester group with lower uptake.
The results presented in Figure 4 for C. albicans were similar to those obtained for S. aureus, with purpurin-18 derivatives 6-8 presenting better results in terms of photoactivation than methyl pheophorbide a derivatives 2-4. However, PS 2 with the ester and butyl group also showed promising inactivation of C. albicans.
Overall, the results were similar to those for S. aureus and the higher the carbon chain present in the photosensitizers, the more hydrophobic the PS is used, and lower microorganism uptake was observed. As a consequence, we observed lower photoinactivation.
Photosensitizers 2-4 and 6-8 were also tested against E. coli, (Gram-negative bacteria), at 500 μM and light dose of 45 Jcm −2 . No photoinactivation by the photosensitizers 2-4 and 6-8 was observed, even using high doses of light, which was completely expected as Gram-negative microorganisms are preferentially inactivated by These results suggest that the purpurin-18 derivatives 6-8, due to the presence of a carboxylic acid group in the molecules may facilitate their microorganism uptake, whereas the derivatives of methyl pheophorbide 2-4 have an ester group with lower uptake.
The results presented in Figure 4 for C. albicans were similar to those obtained for S. aureus, with purpurin-18 derivatives 6-8 presenting better results in terms of photoactivation than methyl pheophorbide a derivatives 2-4. However, PS 2 with the ester and butyl group also showed promising inactivation of C. albicans.
Overall, the results were similar to those for S. aureus and the higher the carbon chain present in the photosensitizers, the more hydrophobic the PS is used, and lower microorganism uptake was observed. As a consequence, we observed lower photoinactivation.
Comparing among photosensitizers, when methylene blue was evaluated against the microorganism S. aureus at a concentration of 50 µM and a light dose of 9 J at 660 nm [60], a reduction of approximately 1.5 log CFU was obtained, whereas the photosensitizers 6, 7, and 8 used in this study completely inactivated the S. aureus microorganism at a concentration of 10 µM and a light dose of 15 Jcm −2 , demonstrating it was more effective using these semisynthetic PS. When using 50 µM and a light dose of 9 J to perform photodynamic inactivation with methylene blue for C. albicans, a reduction of approximately 1 log CFU was obtained, whereas in this study, PS 7 at a concentration of 10 µM and a light dose of 15 Jcm −2 allowed a reduction of approximately 2 log CFU. Using the PS 3, we obtained a reduction of approximately 1 log CFU, and PS 2 and 6 completely inactivated the microorganism.
In an in vitro study, the inactivation of C. albicans was evaluated with ICG at 1 µg mL −1 using a light dose of 228 J/cm −2 (810 nm); a satisfactory result showing a reduction of 1.2 log was obtained, similar to the results with nystatin. Compared to the control, the elimination of C. albicans increased by 92% when treated with ICG (1 mg/mL) with infrared (IR) laser irradiation (810 nm, 55 J/cm −2 ).

Materials and Methods
Nuclear magnetic resonance (NMR) analyses were performed on a Bruker Avance 400 spectrometer at 400.15 MHz ( 1 H) and 100.13 MHz ( 13 C). Tetramethylsilane was used as an internal reference.
Spirulina maxima powder was purchased from Pharma Nostra (Rio de Janeiro, Brazil), and other reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Semisynthesis of Photosensitizers
Isolation of methyl pheophorbide a (1) from Spirulina maxima (Mepheo a): 300 g of the Spirulina maxima was treated with a 1.5 L of the 5% methanolic solution of H 2 SO 4 for 24 h at room temperature. This mixture was filtered and washed with methanol (900 mL) and ethyl acetate (900 mL), and the organic phases were evaporated under reduced pressure. After that, 150 g of crushed ice was added to the crude residue. The residue was neutralized with solid NaHCO 3 and placed on a silica gel plug. The chlorophyll derivatives were retained on the plug, and the residual proteins and peptides (pale yellow in color) were eluted with water. The chlorophyll derivatives were then eluted with ethyl acetate (900 mL) and washed with water (3 × 400 mL). The organic phase was separated, dried over Na 2 SO 4 , and the solvent evaporated under reduced pressure. The methyl pheophorbide a was purified by silica gel flash chromatography using as eluent toluene:ethyl acetate (9:1), yielding the methyl pheophorbide a (1) (2.4 g, 3.9 mmol, 0.8% yield from natural dried Spirulina maxima) [44,63,64].    (2). To a solution of 30 mg (0.049 mmol) of 1 in 3 mL of dry tetrahydrofuran, 0.5 mL of butylamine was added and the reaction was stirred for 20 min at room temperature. After that, the solvent was evaporated, the mixture was diluted with dichloromethane and washed with HCl solution (1%). The resulting solution was dried over anhydrous sodium sulfate, and the solvent evaporated under reduced pressure. Compound 2 was isolated by chromatography over silica gel using toluene:ethyl acetate (9:1) as eluent (22 mg, 72% yield) [50,51].
UV-vis (CH 2 Cl 2 ) λmax (nm): 662, 607, 528, 498, 399 [66].  (3). To a solution of 40 mg (0.066 mmol) of 1 in 3 mL of dry tetrahydrofuran, 0.5 mL of hexylamine was added and the reaction was stirred for 20 min at room temperature. After that, the solvent was evaporated, the mixture was diluted with dichloromethane and washed with HCl solution (1%). The resulting solution was dried over anhydrous sodium sulfate, and the solvent evaporated under reduced pressure. Compound 3 was isolated by chromatography over silica gel using toluene:ethyl acetate (9:1) as eluent (45 mg, 95% yield) [50,51].  (4). To a solution of 30 mg (0.049 mmol) of 1 in 3 mL of dry tetrahydrofuran, 0.5 mL of octylamine was added and the reaction was stirred for 20 min at room temperature. After that, the solvent was evaporated, the mixture was diluted with dichloromethane and washed with HCl solution (1%). The resulting solution was dried over anhydrous sodium sulfate, and the solvent evaporated under reduced pressure. Compound 4 was isolated by chromatography over silica gel using toluene:ethyl acetate (9:1) as eluent (28 mg, 78% yield) [51].  a 1 (151 mg, 0.25 mmol) was dissolved in 500 mL of diethyl ether and 5 mL of pyridine. After, a potassium hydroxide solution in 1-propanol (2 g of KOH was dissolved in 10 mL of 1-propanol) was added into the first solution and oxygen was bubbled into the resulting reaction mixture for 1 h. The reaction mixture was extracted with water (500 mL). The aqueous layer was collected and the pH adjusted to 2-4 using cold H 2 SO 4 solution (25%). The aqueous layer was extracted with CH 2 Cl 2 and the solvent evaporated to give a purple residue. The product was purified by chromatography over silica gel using hexane:ethyl acetate 3:1 as eluent. After that, the carboxylic acid precursor was obtained in 55% yield (0.078 g, 0.119 mmol). This product was further reacted with a diazomethane solution in dichloromethane to produce purpurin-18 methyl ester (5) for 10 min at 0 ºC. The residue was crystallized with dichloromethane/hexane, thus obtaining 5 in 60% yield (63.0 mg, 0.109 mmol) as purple red crystals [51]. Method 2: Pigments were extracted twice from S. maxima dried powder (10 g) with acetone (4 × 100 mL) under magnetic stirring at 60 • C (4 × 30 min). The dark green extract was filtered off and the filtrate (ca. 400 mL) was reduced to 200 mL by partial evaporation under reduced pressure. NaOH (40 mL, 6 M) was added to the previous extract (200 mL); the mixture was vigorously stirred and oxygen was bubbled during 3 h. The solution was then acidified with concentrated HCl. The oxidized extract was evaporated to dryness. Carotenoids and part of xanthophylls were removed by extraction with petroleum ether (2 × 100 mL). The resulting residue was purified by flash chromatography (eluent CH 2

Photodynamic Inactivation
Staphylococcus aureus (American Type Culture Collection, ATCC 25923) and Escherichia coli (ATCC 25922) were grown in brain and heart infusion media. Candida albicans (ATCC 10231) was grown in Sabouraud dextrose broth. For experimental purposes, the microorganism concentration was adjusted to 10 7 -10 8 cells/mL in sterile distilled water, and 500 µL of each microorganism culture was added to 24-well plates with photosensitizers 2-4 and 6-8. Solutions were prepared by diluting the photosensitizer powder (1 mg for S. aureus and C. albicans and 2 mg for E. coli) in 100 µL of dimethyl sulfoxide (DMSO) (to dissolve the PS) and 900 µL of sterile water; the initial concentration of DMSO in the stock solutions was 10%. After dilution of these stock solutions to final concentrations of 1 µM and 10 µM for S. aureus and C. albicans, respectively, and of 500 µM for E. coli, the final concentration of DMSO was less than 2% in all the solutions. After preparing the solutions, they were protected from light. The 24-well plates were kept in the dark at 37 • C for 20 min.
A homemade LED-based device with emission centered at 660 nm was used to irradiate the culture plates. The 24-well plates were irradiated at 30 mWcm −2 using this device for 8, 16, and 25 min, resulting in fluences of 15, 30, and 45 Jcm −2 , respectively, which were used in the PDI against microorganisms. The fluence levels used were 15 Jcm −2 and 30 Jcm −2 for S. aureus and C. albicans, and 45 Jcm −2 for E. coli. After irradiation, 10-fold serial dilutions were performed and cells were cultured in agar plates. The colony-forming units (CFUs) were determined 24 h after initiation of the experimental procedure. All experiments were performed in triplicate.
In the control group (no treatment), 24-well plates were maintained at room temperature for 32 min. Using the same incubation time, the dark toxicity of the photosensitizers was evaluated in 24-well plates covered with aluminum foil to avoid light exposure. Phototoxicity was determined by irradiation at 30 Jcm −2 for S. aureus and C. albicans, and 45 Jcm −2 for E. coli.
Survival fractions (SFs) were expressed as ratios of CFUs of treated groups to the control group. The SF at 0 J/cm 2 provides a measure of the dark toxicity of chlorins.

Statistical Analysis
All the results reported in Figures 3 and 4 were statistically analyzed using the RStudio software (R version 4.1.1 (10 August 2021), R Core Team (2021), R: A language and environment for statistical computing; R Foundation for Statistical Computing, Vienna, Austria (https://www.R-project.org, accessed on 10 August 2021)) using a significance level of at least p < 0.05 and a confidence level of approximately 95%. The data were analyzed and approved by the normality test. Comparisons between the experimental groups were verified by Tukey's test.

Conclusions
The photosensitizers 2-4 and 6-7 derived from chlorophyll a were successfully semisynthesized. The characterization of the compounds is in accordance with data described in the literature and the main photophysical data are compiled and organized in Table S1 (SI). Overall, methyl pheophorbide a or purpurin-18 derivatives with different side chains (from the butyl to octyl groups) were prepared and studied. Subsequently, we investigated the photoinactivation of S. aureus and observed that the methyl pheophorbide derivatives 2-4 did not show great inactivation, whereas the purpurin-18 derivatives 6-8 allowed significant PDT inactivation. For C. albicans, the purpurin-18 derivative with the butyl group showed relevant inactivation, and the methyl pheophorbide with the butyl group also exhibited PDI. Photosensitizers 2-4 and 6-8 were also tested against the Gram-negative bacterium E. coli; however, no significant photoinactivation was observed. In general, the higher the carbon chain present in the photosensitizers, the more hydrophobic the compounds, with consequently lower photoinactivation efficacy. These results suggest that the use of chlorin derivatives with lower hydrophobic properties can be more effective for the photoinactivation of such microorganisms as S. aureus and C. albicans.