The role of singlet oxygen and hydroxyl radical in the photobleaching of meso-substituted cationic pyridyl porphyrins in the presence of folic acid

Abbreviations

1. Introduction

Cancer is a major health threat, with conventional treatments like surgery, chemotherapy, and radiation therapy posing a negative impact on healthy cells, poor selectivity¹ and potential resistance from cancer cells.² Photodynamic therapy (PDT) offers a less invasive alternative.¹

PDT is a technique that relies on the simultaneous action of a photosensitizer (PS), light, and oxygen that cause the generation of reactive oxygen species (particularly singlet oxygen — ¹O₂) that lead to cancer cell death.³

During PDT, PS absorbs a photon, causing a PS to its energetically less stable excited singlet state (Sx), from which the molecule returns to its initial state. Deactivation occurs through vibrational relaxation, dissipation of energy as heat, fluorescence emission, and intersystem crossing to a triplet state (T₁) through phosphorescence. T₁ states of PS interact with substrates, transferring energy and electrons/hydrogen after the formation of ROS through Type-II and Type-I pathways, respectively.⁴ Type-I reactions produce superoxide anions, hydrogen peroxide (H₂O₂), and hydroxyl radicals (^•OH). The last one, could damage cells and diffuse through membranes. In the Type-II photochemical reaction, energy transfer (rather than electron transfer) to molecular oxygen (with an energy threshold of 22.5kcalmol${}^{-1}$) leads to the generation of singlet oxygen (¹O₂),^4,5 thought to be the primary mechanism.⁶

Self-photosensitization also occurs when reactive oxygen intermediates interact with the PS, causing alteration and/or destruction of PS, which is called photobleaching.⁷ The light-induced decrease in absorption or emission intensity occurs during photobleaching through two pathways: Type I involves reactive oxygen species and Type II involves ¹O₂.^7,8 Low photodegradation is optimal for extended radiation periods, although it could be associated with prolonged photosensitivity in the organism.⁶ Photobleaching affects dosimetry of the PDT.⁷ Therefore, it is worth studying the photobleaching of PSs.

It is important to determine if the hydroxyl radical or ¹O₂ is the main agent responsible for the commencement of the photobleaching. The differentiation between ¹O₂ and hydroxyl radicals is challenging due to complex processes. Strategies involve using quenchers like L-histidine for singlet oxygen and D-mannitol for hydroxyl radicals.⁹ Alternatively, probes like Singlet oxygen sensor green (SOSG) and 3${}^{\prime }$-(4-hydroxyphenyl) fluorescein (HPF) aid detection.^10,11 SOSG’s cell-impermeability limits intracellular studies.¹² Otherwise, semiconducting polymer dots doped with SOSG offer sensitive intracellular detection.¹¹ Various probes show promise, but detecting intracellular singlet oxygen remains challenging due to its low concentration and short lifetime.¹¹ Direct singlet oxygen luminescence detection at $\sim 1270$nm faces challenges despite being a “gold standard”.^11,13 For hydroxyl radical detection, HPF is more selective for hydroxyl radicals and does not react with singlet oxygen like the other hydroxyl radical probe 3${}^{\prime }$-(4-aminophenyl) fluorescein (APF).¹¹ Electron paramagnetic resonance method could be utilized as well for both, hydroxyl radicals and singlet oxygen that required an expensive instrument and complicated operating procedures.¹³ It should be noted that probe specificity and scavenger compatibility with biological systems, especially live cells or lipid membranes, pose issues.¹⁴

¹O₂ acts like an electrophile, and readily interacts with electron-rich double bonds.¹⁵ The quenching of ¹O₂ occurs by its inactivation through physical or chemical quenching. In chemical quenching, ¹O₂ reaction with the quencher leads to product formation. During the physical quenching, ¹O₂ only deactivates to its ground state, with no consumption of oxygen or formation of product.⁷ L-histidine (L-His) is the primary target of ¹O₂ oxidation on the proteins: it quenches ¹O₂ both physically and chemically (75%).¹⁶ It was assumed that ¹O₂ interaction by [$4+2$] cycloaddition with the L-His, produced a 2,5 endoperoxide. This molecule rearranges to a hydroperoxide that degrades to the alcohol.¹⁷

Following PDT, the formation of hydroxyl radicals (^•OH) takes place as well. The generation of hydroxyl radicals could take place with different mechanisms and is reviewed in the work of Dąbrowski, 2017.⁴

To assess the role of hydroxyl radicals, the effect of hydrogen donors as inhibitors should be studied.³ D-mannitol as hydrogen donors effectively scavenge ^•OH radicals by combining with one ^•OH radical and producing water molecule.¹⁵ We chose the D-mannitol as hydroxyl radical’s quencher.

D-mannitol and L-histidine can be used (and are the most suitable) as quenchers for the mechanistic study of photobleaching in the solution.

To improve PDT, the main strategy is to increase PS selectivity toward tumor cells to reduce side effects. The cancer cells exhibit fast growth and cell division and for this numerous types of cancer cell lines (on prostate, brain, lung, nose, ovary, and colon) over-express folic acid (FA) receptors.¹⁸ Notably, overexpression of FA receptor $\alpha$ isoform is characteristic of nonmucinous tumors of the ovary, uterus and cervix.¹⁹ This makes FA receptor a promising target for diagnostic imaging and cancer therapeutics.²⁰ In light of this, there is a significant interest in studying the interaction between drugs (such as PSs) and folic acid.

Porphyrins are a class of PSs crucial in biological processes. Transition metal ions (Mn, Fe, Co, Ni, Cu, Zn) can easily be introduced into the structure of the porphyrins. They are widely used in biomedicine and cancer imaging.²¹

The study intended to find out the impact of hydroxyl radicals and ¹O₂ on light-induced modifications of meso-substituted cationic pyridyl porphyrins in the presence of FA. Cationic porphyrins, known for their high DNA affinity are capable of initiating apoptotic cell death²² and a potential in antimicrobial PDT.²³ However, the photodegradation of cationic porphyrins in the presence of FA has not been studied previously.

The novelty of the work lies in the study of the effect of FA on photodegradation and the ability to generate ¹O₂ by cationic porphyrins, which were first synthesized in Armenia at Yerevan State Medical University named after M. Heratsi by Dr. R. Ghazaryan^24,25,26 and exhibited an exceptionally high level of ¹O₂ (${\gamma }_{\mathrm{\Delta }}$) formation.

2. Materials and Methods

2.1. Chemicals

The following cationic metalloporphyrins have been developed in the UK (ARLABION Co. Ltd., UK) and Armenia (Yerevan State Medical University named after M. Heratsi, Armenia): zinc meso-tetra [4-N-(2’-oxyethyl) pyridyl] porphyrin (Zn-TOEt4PyP) and zinc meso-tetra [3-N-butyl pyridyl] porphyrin (Zn-TBut3PyP).^24,25,26 The chemical composition is presented in Fig. 1.

FA has been purchased from ACROS Organics (Folic Acid, 97% pure; Product code 216630100, CAS Number: 59-30-3). L-histidine and D-mannitol were obtained from Sigma-Aldrich.

2.2. Noncovalent interaction of FA and PS’s

The interaction products were gained by the method developed in our laboratory. FA and PS were co-incubated for 24h, in the dark, at 20^∘C, with a 5-fold excess of FA. Glycerol with the final 20% concentration was added and the reaction mixture was incubated additionally for 24h. Glycerol was used as a stabilizer of FA. Since glycerol is an optical clearing agent utilized to increase the depth of light beam penetration into tumor tissue, a synergistic effect is expected when it is used with PDT at both stages of treatment — with preliminary fluorescent mapping of cancer cells and their photodestruction.^27,28,29,30 The purification from unbound components was performed by Al₂O₃ (aluminum oxide) column chromatography. 0.1M phosphate-buffered saline (PBS) with 20% glycerol was used as eluent. The Cary 60UV–Vis Spectrophotometer (Agilent Co., USA) was used to obtain the absorption spectra of the acquired fractions for the wavelength range of 200–700nm. PS and FA concentrations were calculated by the UV–V is spectrum from extinction coefficients.^31,32

2.3. Spectroscopic measurements of complexes and their components

Absorption spectra were taken with MC 122 spectrophotometer (Proscan, Belarus), and also on Cary 60 UV–Vis Spectrophotometer (Agilent Co., USA); fluorescence spectra were obtained on spectrofluorimeter Fluorolog-3 (HORIBA Scientific, Japan, USA, and France), as well as on the Cary Eclipse fluorescence spectrometer (Agilent Co., USA). Porphyrin excitation was near the Soret band maximum. For the fluorescence quantum yield (${\mathrm{\Phi }}_f$) determination, we used a relative method. The standard was meso-tetra-[N-methyl-4-pyridyl]porphyrin (H₂TMe4PYP) tosylate in H₂O, for which ${\mathrm{\Phi }}_f=0.044$.³³

¹O₂ time-resolved luminescence was recorded in the near-infrared region (NIR) using an NIR laser spectrometer described earlier.^34,35 Briefly, PS was excited at a wavelength of 532nm by laser pulses with a duration of 10ns, an energy of $\le 1$$\mu$J and a frequency of 2.5kHz (Nd:YAG-laser DTL-314QT, LAS a photomultiplier tube, PMT (model H10330A-45, Hamamatsu Photonics K.K.), operated in the photon-counting mode. A bandpass filter with a maximum of 1270nm and a halfwidth of 34nm was used for a ¹O₂ luminescence spectral selection. The absorbance values at the excitation wavelength (532nm) varied from 0.04 to 0.1. Quantum yields of ¹O₂ (${\gamma }_{\mathrm{\Delta }}$) formation were determined by a relative method using meso-tetra-(N-methyl-4-pyridyl)porphyrin tosylate as a standard, ${\gamma }_{\mathrm{\Delta }}=0.77$.³⁶

IRF-22Abbe Refractometer was used to measure the refractive index of buffer solutions with different concentrations of glycerol, n, at the wavelength of the yellow sodium line ${\lambda }_D=589.3$nm at a temperature of ($293\pm 1$) K.³⁷

We have studied solutions of porphyrins and their noncovalent interaction products with FA in 0.1M PBS that contains 20% glycerol, at room temperature (293^∘K) in quartz cuvettes ($10\times 10$mm). The effect of glycerol on the ¹O₂ luminescence was also investigated. The kinetics of ¹O₂ luminescence, photosensitized by the studied porphyrins in buffer solutions with different concentrations of glycerol (0%, 10%, 15%, and 20%), were studied.

2.4. Photobleaching study

Sample illumination was performed under similar conditions and the absorbance changes were recorded. The solvent was 0.1M PBS that contained 20% glycerol, and the spectra were recorded at room temperature. All experiments with spectroscopic characteristics were carried out in thermostated cuvettes at room temperature of 20^∘C (thermostat MLW UH4, Germany). The light focused directly on the samples.

Illumination was performed by a halogen lamp with a wavelength range of 380–850nm, by an irradiance of 30mW/cm², for a total of 30-min duration. Irradiation was carried out with a small-sized halogen quartz lamp with a flat filament body with the following characteristics: voltage 24 volts, power 150 watts, luminous flux 5000 lumens. The luminous flux of the lamp was focused onto the sample using a lens, achieving a uniform illumination of the surface. These characteristics make it possible to provide the necessary luminous flux (power) at the surface of the cuvette (in which the samples are located), enabling infusion of at least 94kJ of energy (the well-known Yablonsky scheme). This level of energy is sufficient to excite PSs and enable energy transfer to molecular oxygen that lead to the generation of ¹O₂, which typically requires 10 min or more of irradiation time. The power of the luminous flux was measured with a luxmeter Sunche Digital Light Meter HS1010 (China).

For L-histidine, four concentrations were as follows: 0.01, 0.02, 0.04, and 0.08mM, and for D-mannitol 10 times higher (0.1, 0.2, 0.4, and 0.8mM).

2.5. Statistical analysis

Every experiment was conducted with repetitions to have at least triplicate measurements. The result was reported as $\mathrm{\text{mean}}\pm \mathrm{\text{standard}}$ deviation and analyzed using ANOVA Two-Factor with Replication test. $p<0.001$ was considered statistically significant.³⁸

Statistics were done for different concentrations of L-His and D-mannitol, each concentration relative to its nonilluminated control. Statistics were not performed for the different concentrations regarding each other; it was done regarding their nonilluminated control. There are 4 concentrations; all of them are done relative to their respective control, group by group.

3. Results

3.1. The formation of noncovalent products

We have obtained noncovalent interaction products of FA and PSs. Upon purification from unbound components on the aluminum oxide column, the molar ratio of FA and PSs in the studied solutions were as follows:

(1)	(FA: Zn-TBut3PyP) → 6:1;
(2)	(FA: Zn-TOEt4PyP) → 1:1.

Absorption spectra of Zn-TBut3PyP and (Zn-TBut3PyP $+$ FA) noncovalent interaction products are presented in Fig. 2. Similar spectra were obtained for other porphyrins.

3.2. Study of the photosensitized singlet oxygen (¹O₂) formation by (PS + FA) and its components

The initial investigation focused on the effect of glycerol on the ¹O₂ luminescence, since the interaction products were obtained in 0.1M PBS with 20% glycerol. The kinetics of ¹O₂ luminescence, photosensitized by one of the studied porphyrins with different concentrations of glycerol, is presented in Fig. 3. Similar data were obtained for other porphyrins.

The kinetic parameters of the luminescence were analyzed by the following equation³⁹ :

where A is the coefficient that depends on the initial concentrations of O₂ and PS triplet state, ${\tau }_T$ and ${\tau }_{\mathrm{\Delta }}$ are the lifetimes of PS triplet state and ¹O₂, respectively.

We found that, in the aqueous solutions containing 20% glycerol, the lifetime of the PS triplet state, ${\tau }_T$, increased by about 2 times for all studied porphyrins, while the lifetime of ¹O₂ remained almost unchanged.

It is well known that the triplet state lifetime of porphyrin in a homogeneous medium is determined by the following expression :

where ${\tau }_{T0}$ is the triplet state lifetime in an oxygen-free medium, [O₂] is the concentration of dissolved molecular oxygen, and $k_q$ is the medium-dependent bimolecular rate constant of triplet state quenching by O₂. $k_q$ is directly proportional to the diffusion constant, D, for O₂ in the aqueous solutions.

To avoid uncertainty in determining the quantum yields of ¹O₂ formation by porphyrins in the complexes with FA, the porphyrins in 0.1M PBS with 20% glycerol were used as a standard. The results indicate that, within the experimental error, the quantum yields of ¹O₂ formation for the (FA + Zn-TOEt4PyP) noncovalent interaction products coincide with the value ${\gamma }_{\mathrm{\Delta }}$ for porphyrins (without FA) but the value ${\gamma }_{\mathrm{\Delta }}$ for (FA + Zn-TBut3PyP) is reduced by 16% (Table 1).

3.3. The effect of L-histidine on photobleaching

In 0.1M PBS containing 20% glycerol, we studied FA photobleaching in a concentration-dependent manner. The lighting of FA resulted in photoinduced spectral alterations at FA’s distinctive peaks (280nm and 350nm). The photo-induced absorption changes (reduction in absorption at 280nm and increase in absorption at 350nm) are more pronounced at low concentrations (up to 43.7% increase at 350nm), and diminish with concentration rise. It should be noted that the spectral changes at 280nm for all concentrations were less than 10%. The spectral alterations in the UV–Vis spectral region increased with increasing light time for all concentrations. The stated FA spectrum alterations should be interpreted as evidence of FA molecule photolysis, which appears stronger at lower concentrations of this substance.

The results of the concentration-dependent FA photolysis provided in SI (Figs. S1–S6).

We conducted a photobleaching study of (FA + PS) and free components after continuous illumination for 5, 15, and 30min. As glycerol has a photostabilizing effect on the interaction of FA and PSs, photobleaching was performed for the samples that contain 20% glycerol.⁴⁰ As the duration of exposure increased, the absorption of PSs gradually decreased (both in free components and in the case of interaction of PSs with FA).

To study the contribution of ¹O₂ to the photobleaching phenomenon of our PSs, L-histidine was used as a chemical quencher. It was added to the studied samples, obtaining the final concentrations.

In Fig. 4, the data for the samples photobleaching in the presence of L-histidine were presented, following 30min of illumination.

Figure 4 indicates that in the case of FA, the addition of L-histidine has little effect on the change of its absorbance at 280nm and that the ¹O₂ likely has little/or no effect on the photolysis of free FA.

Among all studied concentrations, 0.08mM was the most effective in almost all cases of photoprotection of free PSs and PSs in noncovalent interaction with FA. It should be noted that L-histidine exhibits more effective photoprotection for Zn-TBut3PyP and for (Zn-Tbut3PyP + FA). The percentage of photobleaching, following 30min of illumination, was 2.3 times lower for Zn-Tbut3PyP and 1.6 times for its noncovalent interaction product, when compared with corresponding samples without L-histidine. Given the observed reduction in photobleaching, we inferred that L-histidine was reacting with generated ¹O₂ and quenching it.

In Fig. 5, the UV–Vis spectrum of (Zn-Tbut3PyP + FA) noncovalent product was presented before and after 30min of illumination with and without L-histidine. It can be noted that with L-histidine, the absorption decrease was less than in the sample that has no L-histidine.

For metalloporphyrin Zn-TOEt4PyP, the 30min of illumination causes 1.2 times less photobleaching for both porphyrin and porphyrin that interacts with FA.

3.4. The effect of D-mannitol on photobleaching

To study the role of D-mannitol as a scavenger of ^•OH radical, four concentrations were tested. For all studied samples the highest (0.8mM) used concentration of the D-mannitol was the most effective in respect of quenching and causing less decrease of absorption following the illumination. D-mannitol’s effect on an absorption decrease and thus on photobleaching following 30min of illumination, are presented in Fig. 6.

For Zn-TOEt4PyP, D-mannitol had more quenching effect compared with L-histidine. Percentage of photobleaching was decreased by 1.8 and 1.37 times for porphyrin and for the noncovalent interaction products with FA, respectively, in case of 0.8mM D-mannitol. In contrast to the aforementioned, 0.8mM D-mannitol was a less effective quencher for Zn-TBut3PyP than 0.08mM L-histidine: the photobleaching percentage was less: by 1.36 and 1.2 times for porphyrin and for porphyrin interacting with FA, respectively (comparison with the samples where no D-mannitol was added).

In Fig. 7, we can see that D-mannitol decreased the photobleaching rate (e.g., the absorption decrease) of the sample by effectively reacting with generated hydroxyl radicals.

Therefore, by using quenchers the involvement of reactive oxygen species in the photobleaching was studied.

4. Discussion

4.1. The formation of noncovalent products

The variation in molar ratio of FA and PS in the complexes can be attributed to the difference in the side substituents of the porphyrins (the butyl or oxyethyl groups) that affected the noncovalent binding nature.

It was found that, after the separation of unbound components, small redshifts (1–3nm) in the absorption spectra of porphyrins were observed. The absorption bands in the region of 200–400nm belong to FA (Fig. 2). The fluorescence spectra and fluorescence quantum yield of the interaction products are not changed significantly. It can be speculated that such minor changes may be associated with the interaction of FA with the side substituents of the porphyrin without affecting the porphyrin macrocycle.

4.2. Study of the photosensitized singlet oxygen (¹O₂) formation by (PS + FA) and its components

It is clear that the addition of glycerol results in changes in the form and intensity of the kinetic of ¹O₂ luminescence.

According to the literature,⁴¹ the solubility of molecular oxygen drops by about 1.2 times in aqueous solutions containing 18% glycerol. It is well known that the diffusion constant, D, depends inversely on the viscosity of the solution. The addition of 20% glycerol in aqueous solutions leads to an increase in the viscosity of up to 1.8 times.⁴² Therefore, taking into account Eq. (2), the observed increase in ${\tau }_T$ correlates with the changes in the oxygen solubility and the viscosity of the solutions, containing glycerol.

The changes in the intensity of the ¹O₂ luminescence can be caused not only by the variation in the quantum yield of ¹O₂ formation but also by the changes in the rate constant of ¹O₂ luminescence. It is known that there is a correlation between the rate constant of ¹O₂ luminescence and the refractive index of the solution.${}^{43}$ Our measurements showed that the refractive index of the buffer solution increases from 1.335 to 1.367 after adding 20% glycerol. Such changes in the refractive index should lead to an increase in the rate constant of ¹O₂ luminescence by a factor of $\sim 1.6$.⁴⁴ Based on the obtained experimental data, it can be concluded that the observed 1.3x–1.4x increase in the intensity of the ¹O₂ luminescence in 20% glycerol aqueous solutions for all studied porphyrins is in good agreement with the increase in the refractive index of the solutions. However, since many factors affect the shape and the intensity of ¹O₂ luminescence, it cannot be ruled out that small changes in the quantum yield of ¹O₂ formation may occur upon the addition of glycerol.

One of the reasons for the decrease in ${\gamma }_{\mathrm{\Delta }}$ is a quenching of singlet oxygen by FA. However, analysis of the kinetic of singlet oxygen luminescence showed that the lifetime of ¹O₂ does not change in the presence of FA. According to Cabrerizo et al.,⁴⁵ the quenching constant of singlet oxygen by FA is ($30\pm 3$) 10⁶ M${}^{-1}$ cm${}^{-1}$. It can be shown that the concentration of FA in the (FA + Zn-TBut3PyP) sample is too small to decrease the luminescence intensity of ¹O₂ and its lifetime by more than 2%. Consequently, the decrease in ${\gamma }_{\mathrm{\Delta }}$ is caused by changes in the photophysical properties of the porphyrin in the complex with FA. To determine the influence of FA on the photophysical properties of Zn-TBut3PyP, additional experiments will be carried out with different concentrations of FA. The detailed analysis of the results of such experiments will be the subject of a separate publication.

It was also noted that FA has a low quantum yield of ¹O₂ ($\le 0.02$).⁴⁶ FA’s ${\gamma }_{\mathrm{\Delta }}$ has significantly low values compared with the other pterin derivatives. Fluorescence quantum yields of FA (${\lambda }_{\mathrm{\text{ex}}}=350$nm), in both acidic and alkaline media, are lower than for pterin, 6-formylpterin, 6-carboxypterin, and neopterin. This could be explained by FA’s long chain substituent that acts as an “internal fluorescence quencher”, increasing the radiationless deactivation of the singlet excited state. As a result, intersystem crossing becomes very inefficient and FA operates like a weak ¹O₂ sensitizer.⁴⁶ Therefore, in the complexes with porphyrins, FA will not act as PS.

4.3. The effect of L-histidine on photobleaching

Our previous study has shown that extended illumination periods (up to 1h) lead to a more significant reduction in absorption across all bands, resulting in more intense photobleaching of the compounds.⁴⁰ In this study, we limited the illumination exposure to 30min as the timeframe is commonly employed in vitro and in vivo experiments involving cell and mice irradiation. Furthermore, the chosen exposure duration was deemed optimal for modeling purposes as it pertains to the destruction of cancer cells.

According to Khaled et al.,⁸ the 1mM L-histidine results in a less photobleaching for Sn(IV) chlorin e₆’s. Besides, the porphyrin photobleaching could be more intense in the presence of the photooxidizable substrate (e.g., histidine) as a result of the attack on the porphyrin macrocycle by reactive ¹O₂ photooxidation products of the substrates itself.⁴⁷

L-histidine reduces Chlorin p6 photobleaching in serum medium, while mannitol has no effect, suggesting a Type-II mechanism for that compound.⁴⁸

It is worth mentioning that we observed the nonlinear photoprotection depending on the concentration of L-histidine. Our purpose was to find the optimal concentration for photoprotection of PSs. We expected a linear relationship between photoprotection and the presence of L-histidine. In samples lacking quenchers, we observed that as the duration of illumination increased, so did the photobleaching.

It has been noted that depending on the concentration, quencher could have the reverse effect. Particularly, in the study of Chekulayeva et al.,⁴⁷ it was noted that the porphyrin photobleaching could be more intense by the attack of reactive ¹O₂ photooxidation products of the substrates on the porphyrin.⁴⁰ From the data in Fig. 4, it can be seen that for Zn-TOEt4PyP and Zn-TOEt4PyP+FA in the presence of 0.02mM of L-histidine the same tendency of the photobleaching is observed. One can also see that Zn-TOEt4PyP+FA had more photobleaching than the free compounds, which was also consistent with our previous study.⁴⁰ Previous research has shown that the irradiation can lead to FA conversion into p-aminobenzoyl-L-glutamic acid (PGA) and 6-formylpterin (FPT) that further oxidation lead to pterin-6-carboxylic acid (PCA) formation.³² The complexity of PS–FA system increased with an increase of the FA concentration and irradiation exposure. This is because FPT and PCA from FA photolysis’ products act as a PS themselves, resulting in more pronounced photobleaching. The shifts from linear nature from the concentration of quencher can also be explained by the complicated nature of these FA photolysis products. We have more “PSs” (FA photolysis products and porphyrin) that in low concentrations of quenchers not only did not quench but also led to the reverse effect.

Although these findings provide valuable insights, further research will be required to fully understand the underlying mechanisms and to confirm the proposed hypotheses.

In summary, photodegradation of FA can operate as an additional source of ROSs, as evidenced by the observed increase in photobleaching for the Zn-TOEt4PyP+FA sample. The photodegradation of the Zn-TOEt4PyP is enhanced in the presence of FA. For Zn-TBut3PyP + FA, the photodegradation in the presence of FA is lower; which is consistent with the data regarding ¹O₂ quantum yield indicating that the latter is lower than for free porphyrin.

4.4. The effect of D-mannitol on photobleaching

It was noted by Lassalle⁷ that for effective trapping of ^•OH radical, the concentration of the D-mannitol should be very high. Therefore, 10-fold high D-mannitol concentration was added compared with L-histidine.

Our study showed effective quenching for Zn-TOEt4PyP with D-mannitol, and in contrast to this L-histidine was effective for Zn-TBut3PyP as a quencher of the ¹O₂. Furthermore, it is worth to mention that for Zn-TBut3PyP the ¹O₂ quantum yield (${\gamma }_{\mathrm{\Delta }}=0.98$) higher than for Zn-TOEt4PyP (${\gamma }_{\mathrm{\Delta }}=0.84$ for Zn-TOEt4PyP) (Table 1). In summary, both quenchers affected the photobleaching rate, so ¹O₂ and ^•OH radicals play an important role in the photobleaching process of studied porphyrins.

It should be noted that D-mannitol is also used as an optical clearing agent,⁴⁹ and its hyperosmotic properties can disrupt the blood–brain barrier, allowing for the delivery drugs or PS to brain tumors.⁴⁸ This may have a synergistic effect in the PDT treatment of brain tumors.

To clarify the role of D-mannitol and glycerol as potential optical clearing agents for PDT treatment of brain tumors we should briefly discuss available literature.^{49,50,51,52,53,54,55,56,57} In particular, it is known that hyperosmotic agents could be applied in acute cerebral ischemia for brain edema treatment. Mannitol and glycerol are widely used to control intracranial pressure (ICP).⁵¹ Reduction of brain water content, increased cerebrospinal fluid absorption and cerebral blood flow are putative beneficial mechanisms of action of hyperosmotic agents. Beneficial effect of mannitol against hemorrhagic stroke brain injury and neuroinflammation was revealed too. Intraperitoneal glycerol injection following the stroke has a protective effect on brain from hemorrhagic injury.⁵² To address its role in brain treatment, it is worth mentioning that glycerol has several beneficial properties, such as: (1) The readily movement of the glycerol through the blood brain barrier into the brain,⁵¹ which is an important characteristic for cancer treatment. (2) Glycerol’s potential ability to improve neurologic outcomes, as it promotes brain tissue metabolism, leads to less oxygen requirements and provides energy. In contrast to glycerol, mannitol did not possess such properties. However, there is insufficient evidence for the glycerol therapy.⁵³ Reducing cerebral edema and blood flow improvement caused by glycerol.⁵⁴ Glycerol used for the brain tumor treatment.⁵⁴

In the study of Moulton et al., it was noted that application of glycerol to porcine skin caused no changes in skin absorption coefficient while reflectance decrease and transmittance increase occur within the spectral range of 500–600nm, caused by decrease of the reduced scattering coefficient.⁵⁵ This improved the sensitivity of detection of photonic emission from transformed Salmonella through porcine skin compared to untreated porcine skin and can be proposed for use in PDT. Glycerol had the significant optical clearing effect in in vivo animal and human studies,^27,28,29,30 in particular fluorescence imaging of labeled cancer cells in mice.^29,56

Mannitol can be used to induce brain relaxation in patients undergoing supratentorial brain tumor resection. It is widely used to reduce ICP in patients with cerebral edema. It should be considered that mannitol has adverse effects as well, including hypochloremic metabolic alkalosis associated with volume contraction and diuresis, hypernatremia, hypokalemia, and renal failure.¹¹ Brain relaxation which is necessary for the patients that undergo craniotomy, could be achieved by mannitol, which removes water from brain tissue or decreases cerebral blood flow. Seo et al.⁵⁷ showed that although the high doses of mannitol lead to better relaxation of brain, it can also cause adverse effects.

Regarding the role of the mannitol and glycerol in light penetration properties and also the role during brain tumor conditions, it proposed that these compounds could have a significant role in the PDT of tumors,^27,30 especially for PDT of brain tumors.

5. Conclusion

The use of quenchers for ¹O₂ and hydroxyl radical can be used for mechanistic study of photobleaching. ¹O₂ and hydroxyl radicals are involved in the mechanism of photobleaching of studied porphyrins and their noncovalent interaction products with FA. It has been shown that the photobleaching of Zn-TOEt4PyP is increased in the presence of FA, whereas FA reduces the photobleaching of Zn-TBut3PyP. The reasons for such phenomenon are discussed. On the one hand, FA photolysis products can act as a PS themselves and lead to the increase in photobleaching of Zn-TOEt3PyP. On the other hand, the interaction of Zn-TBut3PyP with FA, leads to the decrease of ¹O₂ quantum yield and thus protects porphyrin from photodegradation.

Both pharmaceutical preparations (glycerol and mannitol) are widely used in medicine and have multifunctional properties. They can synergistically enhance the effectiveness of PDT due to their simultaneous positive effect on PS stability, increased permeability through biological barriers, suppression of photobleaching, and increased optical transparency of the tumor and surrounding tissues. This is achieved through the temporary and reversible suppression of light scattering. The latter makes it possible to observe the fluorescence of PS-labeled cancer cells from a greater depth and to more effectively irradiate the entire tumor volume with light.

Acknowledgments

This work was supported by the RA MESCS Science Committee and Belarusian Republican Foundation for Fundamental Research in the frames of the joint research project SC No. 21SC-BRFFR-1F007 and BRFFR Grant No. $\mathrm{\Phi }$21ARM-014 accordingly, as well as from the Ministry of Science and Higher Education of Russian Federation within the framework of a state assignment (project No. FSRR-2023-0007). Authors express their gratitude to Dr. Anna Zakoyan for performing the statistical analysis, as well as to Dr. Robert Ghazaryan for providing the cationic porphyrins.

Supplemental Materials

The Supplemental Materials are available at: https://www.worldscientific.com/doi/suppl/10.1142/S1793545824400029

Conflicts of Interest

The authors declare no conflict of interest.

ORCID

Lusine Mkrtchyan  https://orcid.org/0000-0003-0173-7090

Torgom Seferyan  https://orcid.org/0009-0002-9169-7888

Marina Parkhats  https://orcid.org/0009-0003-0378-1367

Sergei Lepeshkevich  https://orcid.org/0000-0001-6766-4731

Boris Dzhagarov   https://orcid.org/0009-0000-7326-9527

Gagik Shmavonyan  https://orcid.org/0000-0001-5207-6857

Elena Tuchina  https://orcid.org/0000-0003-4498-2846

Valery Tuchin  https://orcid.org/0000-0001-7479-2694

Grigor Gyulkhandanyan  https://orcid.org/0000-0003-2725-0555