Magnetic Properties of Reactive Oxygen Species and their Possible Role in Cancer Therapy

Cancerous tissue can be removed quite effectively by implementing a simple drug–the organic molecule of photosensitizer which needs to be elucidated by UV-visible light in the vicinity of the tissue [9]. This photodynamic therapy (PDT) utilizes the light source with an optical waveguide to produce photoexcitation of the drug. The excited photosensitizer being in contact with O₂ molecules (always present in tissue) can transfer its electronic energy to O₂ providing the singlet excited oxygen O₂(¹Δ_g), Figure 2. The second singlet excited ${}^1\Sigma {}_g^{+}$ state quickly relaxes transferring energy to the low-lying ¹Δ_g state [9]. Internal relaxation between singlet states is spin-allowed.

Electronic transition ${}^{\text{1}}{\Delta }_{\text{g}}\to {}^3\Sigma {}_g^{-}$ is strictly forbidden by spin and by orbital selection rules [2]. It is observed as a very weak near-infrared emission at 1270 nm with special sensitive photodetectors [9]. In a great difference from usual molecular spectra, this forbidden transition intensity is determined by weak magnetic interactions inside the O₂ molecule [2] which are the focus of our review. The low rate of spontaneous oxygen emission ¹Δ_g → ${}^3\Sigma {}_g^{-}$ determines the long lifetime (^τΔ) of the singlet excited oxygen O₂(¹Δ_g) in the rarefied gas phase (about an hour). In dense gases (few seconds) and in solvents it is shortened; in water ^τΔ =3 μs [9]. This lifetime is enough to react with cancer tissue cells and kill them since the singlet excited oxygen has no spin restrictions for chemical reactions with singlet organic molecules. The cancer cells include highly unsaturated polymers and π-conjugated molecules, which are easy to be oxidized by O₂(¹Δ_g) species that possess an energy excess of 94 kJ/mole. The spin and energy are important factors of singlet oxygen reactivity with cancer tissues.

Light absorption by photosensitizer (PS) being in collision with an O₂ molecule can lead to the charge-transfer (CT) excited state PS⁺ - O₂- of the triplet spin nature (since O₂ is triplet and PS is singlet in the ground states). This triplet radical pair (RP) ³(PS⁺• - O₂-•) possesses an intrinsic magnetic force (IMF) which can induce the spin-flip in the superoxide ion-radical (O₂-•), thereby providing the triplet-singlet (T-S) transition in this RP [6,7]. Thus the singlet RP has no spin restriction for chemical reactions with the tissue’s organic molecules. The singlet excited oxygen O₂(¹Δ_g) and superoxide ion-radical (O₂-•) produced via the photosensitizer as a result of its electronic excitation are the main reactive oxygen species (ROS) which have no spin restrictions for organic oxidation in great contrast to the ground triplet state O₂( ${}^3\Sigma {}_g^{-}$ ) oxygen which we breathe. The superoxide ion radical can be produced in dark biochemical reactions that accompany the electron transport chain (ETC) and other enzymatic redox processes [2,10]. Oxidative phosphorylation as the last step of cellular respiration, Eq.(1), provides the opportunity for superoxide leakage from the Krebs cycle. Thus, the random O₂-• leakages always accompany normal metabolism. That is why the superoxide dismutase (SD) regulating superoxide concentration is the oldest enzyme on the Earth and occurred at the very beginning of aerobic evolution [1]. The lifetime of the ion-radical O₂-• is comparable with ^τΔ in many biological liquids. These reactive oxygen species usually give rise to other ROS [1,7], like OH radicals (through the Fenton reaction [11]), OOH, or peroxynitrite [12]. All these processes depend on the spin and magnetic properties of the parent oxygen and its reduced derivatives which are highly reactive.

Singlet oxygen and some reactive oxygen species are often produced via the electronic excitation of photosensitizer dye (Figure 2). In the past two decades, tremendous efforts have been paid towards the design, synthesis, and development of new photosensitizer molecules for effective cancer treatment with PDT.

The PS typically is an organic molecule with the singlet ground state (S0); it possesses an even number of electrons paired with opposite spins as being located on the most energetically favorable molecular orbitals (MO). After absorbing light, one of the electrons is excited and transferred to a higher energy orbital with the spin orientation preserved (the first excited singlet state, S1) (Figure 2). PS in its singlet excited state is relatively unstable and can return to the ground state either by emission of photon (Fluorescence) or by release of the excitation energy in the form of chemical bond vibrations and heat dissipation (Internal Conversion, IC). Besides IC, the first excited singlet state can alternatively undergo intersystem crossing (ISC, electron spin reorientation) to the first excited triplet state T₁. The driving force for the spin-flip during the ISC process is the internal magnetic interaction induced by spin-orbit coupling (SOC) in the PS electronic shell. It depends on the orbital reorganization during the S₁ → T₁ quantum transition [2]. SOC is higher when atomic orbital rotation (of the px-py type) accompanies the S₁ → T₁ transition. The T₁ state lifetime is relatively long in the cell liquid (about a millisecond or longer) to enable either direct energy transfer (Type II) or electron transfer (Type I) reactions with oxygen resulting in ROS generation [13,14]. The quantum yield and relative contributions of these photochemical reactions depend on the PS structure, concentration of the PS, light intensity, and dose of molecular oxygen in the nearest micro-environment [14]. Most of the porphyrins generate mainly singlet O₂(¹Δ_g) oxygen through the Type II mechanism, while the palladium bacteriochlorophyll derivative is a pure generator of Type I superoxide ion products. Other bacteriochlorins generate both singlet oxygen and oxygen-centered radicals [13-17].

Optimization of the PS triplet T₁ state is the main strategy currently employed in the photosensitizer design to increase their efficiency in ROS generation [13,15]. Introduction of heavy transition-metal ions, such as Pt(II), Ir(III), Zn(II), and Pd(II) into the porphyrin ring increases the ISC rate and enhances the yield of the long-lived triplet T₁ state [13-16]. The metal ion choice can play a crucial role in the photochemical and redox properties of photosensitizers. It affects the spin density distribution in the triplet state of the PS molecule, which in turn influences its magnetic properties, ISC rate, and ability to perform an electron or energy transfer during ROS production according to Type I or Type II mechanisms. The oxidation potential of the metal ion is also an important factor determining the charge-transfer contribution to spin-orbit coupling (SOC) matrix element (ME) between S and T states [4,7]. The redox properties determine how easily the PS molecule can donate or accept electrons, which is important for the rate of energy transfer to oxygen molecules. Of course, SOC ME as an important magnetic parameter strongly depends on the atomic number (Z) of heavy metal ions (as Z⁴) [2]. However, the oxidation potential of the ion influences the atomic orbitals mixing and rotation through the CT contributions as was shown in the vase of ^τΔ calculation in singlet oxygen quenching by amines [18].

PDT mechanisms of tumor destruction

Various reactive oxygen species (ROS) being generated within the tumor tissue after its PS doping and irradiation are involved in the oxidation of numerous biological structures. The metastable O₂(¹Δ_g) molecules can be produced through PDT in such amounts which is sufficient to control various types of cancer cells without causing damage to healthy tissue. Some PS dyes are very effective in the high production of singlet O₂(¹Δ_g) oxygen through the Type II mechanism. Such natural dyes as Rose Bengal and various phthalocyanines served as useful PS drugs since the early days of PDT applications [17]. Illumination of PS induces the generation of the excited triplet T₁ state (³PS*) which can transfer electrons, protons, or electronic energy, leading to the production of ROS. Numerous ROS cause apoptosis or necrosis of tumor cells by photochemical oxidation.

Modern PSs operate in both mechanistic Types I and II; they can lead to oxidative stress within irradiation time and determine the PDT anticancer activity, which consists of three main mechanisms: (1) effect on cancer cells, (2) effect on tumor blood vessels, (3) effect on the host immune system [15,17]. Compared to antibiotics, PDT does not develop bacterial resistance. Operating independently, these mechanisms have the potential to interact with one another; their contributions to the overall tumor response are currently under serious study [15,19,20]. The most significant features governing a given mechanism include the general parameters of the PS (its electronic structure, lipophilicity, size, and charge), its redox and photochemical properties, the tendency to aggregation, and the applied time intervals from the PS administration to irradiation (drug-to-light interval; DLI) [21,22]. A comprehensive combination of these factors is necessary to achieve optimal long-term tumor remission, particularly in the case of metastatic cancers. Different molecules have been approved for their clinical use as photosensitizers; they include the well-known porphyrin oligomer mixtures, such as Photofrin® or the Verteporfin® (Visudyne) and Temoporfin® (Foscan) porphyrins [20]. Hydrophilic compounds (e.g. padeliporfin, F2BOH, TPPS, AlPcS3) primarily bound to albumin and often exhibiting low stability, are commonly employed in vascular-targeted therapy (V-PDT, DLI ≤ 15 min). Conversely, hydrophobic compounds (e.g. Temoporfin, ZnPC) which are localized within the inner lipid core of lipoproteins, particularly low-density lipoproteins (LDL), are regarded to require a solubilizing carrier such as liposomes or polymeric micelles [19]. These compounds are typically employed in cellular-targeted photodynamic therapy (long-acting C-PDT, DLI ≥ 12 h) [20]. Some other PSs have entered different stages of clinical trials. Among the different families of metallophotosensitizers, Ru(II) complexes with azaheterocyclic ligands have become recently available [19]. Ruthenium(II) ion as a heavy element with a large Z⁴ value possesses a strong SOC-induced magnetic field being an efficient T₁ state provider. Under illumination this PS produces various ROS being adsorbed near the intracellular membrane. Particular photosensitizers accumulate within specific organelles, while others may undergo redistribution over DLI time. Anyway, the primary intracellular targets for various PSs include the plasma membrane, endoplasmic reticulum, Golgi apparatus, lysosomes, and mitochondria. The singlet O₂(¹Δ_g) oxygen lifetime, its diffusion length and its DLI strongly depend on solvent microenvironment, pH, and organelle redox parameters. Spin-orbit coupling between ¹Δ_g and ${}^3\Sigma {}_g^{-}$ states is forbidden in isolated oxygen molecules; thus, there is no magnetic force to induce ${}^{\text{1}}{\Delta }_{\text{g}}\to {}^3\Sigma {}_g^{-}$ transition in oxygen gas at near-zero pressure [2]. Internal magnetic interactions in the O₂ molecule can be induced by charge-transfer admixtures to the wave functions of the ¹Δ_g and ${}^3\Sigma {}_g^{-}$ states [18], thus generating the spin-flip and quenching the singlet excited O₂(¹Δ_g) ${}^3\Sigma {}_g^{-}{}^1\Sigma {}_g^{+}$ oxygen. The SOC-induced mixing between and ... states is also very important for the O₂(¹Δ_g) state lifetime [2]. This SOC ME energy is relatively high (2.13 kJ/mole) and is dominative IMF for the whole ROS chemistry.

Superoxide ion-radical O₂-• produced by Type I photo reactions from excited ³PS* possesses doubly degenerate ground states with two electronic configurations shown in Figure 3.

From the first glance both states in Figure 3 possess the same energy when we account for only Coulomb electric interaction. However, the account of intrinsic magnetic forces in the form of spin-orbit coupling provides a small energy difference (1.91 kJ/mole). This fine splitting was measured in the O₂-• ionic beam by the electron time-of-flight spectroscopy [21]. Such a small energy is low compared to the dissociation of a chemical bond (300-500 kJ/mole) or even to the singlet O₂ excitation (Figure 1). We must emphasize that the SOC energy can be used not as a brute chemical force, but as a “tiny key” whose gentle rotation “opens the door” to the ordinary organic chemistry of singlet states. Simply speaking, superoxide is produced from the excited ³PS* through the 3(CT) state of the triplet radical pair ³(PS⁺• - O₂-•) [22,23]. Before dissociation this triplet RP has to undergo the T-S quantum transition to provide a fast chemical reaction allowed only in the singlet state. The intrinsic magnetic energy (1.91 kJ/mole) is high enough to induce a fast spin flip in the superoxide ion-radical O₂-•, thereby providing the triplet-singlet (T-S) transition in this radical pair [2,7]. The strength of magnetic interaction 1.91 kJ/mole is an order of magnitude higher than typical SOC matrix elements for T-S transitions observed in organic dyes in the form of intense phosphorescence. For example, the efficient ³(nπ*)→S0 transition in formaldehyde is governed by SOC ME equal to 0.23 kJ/mole [2]. The square of the SOC ME value determines the rate of quantum T-S transition. Thus, we see that the T-S conversion rate constant in the superoxide radical pair is two orders of magnitude higher than the typical rate of the measured organic phosphorescence. These estimates convince us that the T→S conversion inside the radical pair (RP) must be fast enough to compete with the rate of RP decomposition into two radicals. Since the Type I mechanism of PDT with electron transfer to oxygen occurs with the participation of the photosensitizer adsorbed on the tumor, the transition to the singlet state covers the entire ³PS*-O₂-cancer system. Now the tumor molecules in the singlet state are open to direct oxidation without spin inhibition. It is important to emphasize that the photosensitizer does not contribute to the SOC enhancement, it is restored and not consumed in such a PDT mechanism.

Possible pathways of PDT effect on cancer cells

The direct effect of ROS on cancer cells is to damage proteins, lipids, and DNA, resulting in an imbalance of homeostasis changes in cellular metabolism, and ion transportation [20]. The molecular response to oxidative stress consists of the activation of mitogen-activated protein kinase (MAPK) signal transduction pathways. They release several cytokines and mediators responsible for the processes of apoptosis, necrosis, and autophagy. These pathways are not mutually exclusive and can occur simultaneously in the same cell population. The interplay between apoptosis and necrosis after the PDT action is influenced by various accompanying factors. The most important parameters include the intracellular distribution of the PS, drug dose, light dose, DLI (influence the ISC and IC rate and total time of illumination), and the level of oxygen in the tissue. Moreover, there are some newly discovered types of regulated cell death important to PDT outcome, such as necroptosis [15], ferroptosis (cell death driven by iron-dependent lipid peroxidation) [19], pyroptosis (cell death triggered by inflammation) [21]. Activation of the programmed cell death can be induced either by the receptor or by mitochondrial pathways, where new superoxide ion-radicals can be additionally generated from enzymes of the Krebs cycle (besides the Type I mechanism of PDT). The death receptor pathway requires the presence of the so-called death receptors on the cell surface which, once combined with the proper ligand, initiate the caspase activation cascades leading to apoptosis [19]. The involvement of death receptor-dependent signaling is a consequence of cytokine expression by cells undergoing PDT-induced oxidative stress [17,24]. The mitochondrial apoptosis pathway is a very common mechanism in PDT since the primary target for many photosensitizers are these particular subcellular structures.

Technical comments. Two electronic configurations in Figure 3 differ by MS quantum number (in the left part M_S =1/2, in the wright part M_S =-1/2). Being in contact with the PS⁺• radical (with M_S =1/2) the left state of the superoxide ion-radical produces the triplet state of the radical pair ³(PS⁺• - O₂-•), while the Wright electronic configuration in Figure 3 produces the singlet state of the similar radical pair ¹(PS⁺• - O₂-•). The SOC ME between these T and S states includes the single-electron integral of the L_z operator between πgx-πgy orbitals, where L_z is an operator of the orbital angular momentum projection on molecular O-O axis. Such orbital integral provides the maximum possible value of the SOC matrix element for the oxygen atomic number Z=8. This SOC ME coincides with the observed SOC-induced splitting in the ground doublet state of the superoxide ion [21] according to theory [2]. A simple qualitative interpretation of these quantum-mechanical results looks like this. The T and S states of radical pair ³(PS⁺• - O₂-•) and ¹(PS⁺• - O₂-•) differ by electron “jump” from πgx to πgy orbital with simultaneous spin-flip (Figure 3). Such electron “jump” corresponds to rotation of “electron cloud” around molecular axis; this rotation creates a magnetic field (magnetic torque) which induces the spin-flip [7,22].

In this mini-review, we have emphasized the role of spin-depending internal magnetic interactions in such reactive oxygen species as singlet O₂(¹Δ_g) oxygen and superoxide ion-radical O₂-• generated in the course of photodynamic therapy. These two species, as many other ROS, are paramagnetic particles; the singlet state O₂(¹Δ_g) molecule – because of the orbital angular momentum, all others – because of non-paired spin. The necessity of ROS occurrences in aerobic life is determined by an evident paramagnetic nature of the O₂( ${}^3\Sigma {}_g^{-}$ ) molecule. Oxygen from the air can easily oxidase paramagnetic metals but is unable to react with diamagnetic organic staff in the absence of radicals because of the spin prohibition. The oxygen molecule itself demonstrates outstanding magnetic property – the SOC-induced mixing between ${}^3\Sigma {}_g^{-}$ and ${}^1\Sigma {}_g^{+}$ states (Figure 1) which affects the entire photobiology of oxygen in the cell, including quenching of singlet O₂(¹Δ_g) and its phosphorescence.

Numerous dye photosensitizers upon illumination generate both types of ROS - O₂(¹Δ_g) and O₂-• ion (as well as OH, OOH, NO, OONO-, H₂O₂). So far in PDT practice nobody paid attention to the outstanding magnetic property of the superoxide ion – the SOC-induced splitting of its ground doublet state. The observed energy of this splitting is relatively high and comparable with the SOC matrix element between the ${}^3\Sigma {}_g^{-}$ and ${}^1\Sigma {}_g^{+}$ states in the molecular ancestor O₂; both magnetic parameters have the same origin - the symmetry properties of the π_gx-π_gy degenerate orbitals in the O₂ system.

An important consequence of quantum perturbation theory, based on the above data, is that the energy of the SOC splitting in superoxide ion determines the rate of the triplet-singlet transition in the radical pair produced by an electron transferring from the triplet-excited photosensitizer to oxygen. This removes the spin prohibition for direct oxidation of membrane components or tumor organelles on which the ³PS* was adsorbed. This is possible since all active components, ROS, PS and tumor tissue arrange the common photo-reactive system.

1. Mittler R. ROS are good. Trends Plant Sci. 2017;22(1):11-19. Available from: https://doi.org/10.1016/j.tplants.2016.08.002
2. Minaev BF. Electronic mechanisms of molecular oxygen activation. Russ Chem Rev. 2007;76(11):988-1010. Available from: https://www.russchemrev.org/RCR3720pdf
3. Wigner E, Witmer EE. Zeits. F. Physik. 1928;51:859-864.
4. Minaev BF, Minaeva VA. Spin-dependent binding of dioxygen to heme and charge transfer mechanism of spin-orbit coupling enhancement. Ukr Biol Acta. 2008;2(1):56-64. Available from: https://www.bioorganica.org.ua/UBAdenovo/pubs_6_2_08/Minaev_2008_2.pdf
5. Zadeh-Haghighi H, Simon C. Magnetic field effects in biology from the perspective of the radical pair mechanism. J R Soc Interface. 2022;19(193):20220325. Available from: https://doi.org/10.1098/rsif.2022.0325
6. Prabhakar R, Siegbahn PE, Minaev BF. A theoretical study of the dioxygen activation by glucose oxidase and copper amine oxidase. Biochim Biophys Acta. 2003;1647(1-2):173-8. Available from: https://doi.org/10.1016/s1570-9639(03)00090-6
7. Minaev BF. Magnetic torque in superoxide ion is the main driving force of dioxygen activation in aerobic life. Biomed J Sci Tech Res. 2021;38(4):121-131. Available from: http://dx.doi.org/10.26717/BJSTR.2021.38.006171
8. Zheng J. Energy metabolism of cancer: Glycolysis versus oxidative phosphorylation (Review). Oncol Lett. 2012 Dec;4(6):1151-1157. Available from: https://doi.org/10.3892/ol.2012.928
9. Krasnovsky AJ, Jr. Primary mechanisms of photoactivation of molecular oxygen. History of development and the modern status of research. Biochemistry (Moscow). 2007;72:1065-1080. Available from: https://doi.org/10.1134/s0006297907100057
10. Lane N. Oxygen: The Molecule that Made the World. Oxford: Oxford University Press; 2002. p. 365. Available from: https://books.google.co.in/books/about/Oxygen.html?id=LXPrLcFdIPwC&redir_esc=y
11. Zakharov II, Kudjukov KYu, Bondar VV, Tyupalo NF, Minaev BF. DFT-based thermodynamics of Fenton reactions rejects the ‘pure’ aquacomplex models. Comput Theor Chem. 2011;964:94-99. Available from: https://doi.org/10.1016/j.comptc.2010.12.004
12. Sukhina MC. Spin-orbit coupling induced intersystem crossing in peroxynitrite. XXVI Ukrainian Conference of Young Scientists; Cherkasy, 2024;158-161.
13. Minaev BF. Photochemistry and spectroscopy of singlet oxygen in solvents. Recent advances which support the old theory. Chem Chem Technol. 2016;10(4):519-530. Available from: http://dx.doi.org/10.23939/chcht10.04si.519
14. Foot C. Mechanisms of photosensitized oxidation: There are several different types of photosensitized oxidation which may be important in biological systems. Science. 1968;162:963-970. Available from: https://doi.org/10.1126/science.162.3857.963
15. Riebe J, Bädorf B, Löffelsender S, et al. Molecular folding governs switchable singlet oxygen photoproduction in porphyrin-decorated bistable rotaxanes. Commun Chem. 2024;7:171. Available from: https://www.nature.com/articles/s42004-024-01247-7
16. Loboda O, Tunell I, Minaev BF, Ågren H. Theoretical study of triplet state properties of free-base porphin. Chem Phys. 2005;312(1-3):299-309. Available from: http://dx.doi.org/10.1016/j.chemphys.2004.11.041
17. Yashchuk LB. Possible electronic mechanisms of generation and quenching of luminescence of singlet oxygen in the course of photodynamic therapy: ab initio study. Biopolym Cell. 2006;22(3):231-235.
18. Minaev BF. Spin-orbit coupling of charge-transfer states and the mechanism for quenching singlet oxygen by amines. Theor Exp Chem. 1984;20(2):199-201. Available from: https://link.springer.com/article/10.1007/BF00592809
19. Pang Y, Li C, Deng H, Sun Y. Recent advances in luminescent metallacycles/metallacages for biomedical imaging and cancer therapy. Dalton Trans. 2022;51:16428-16438. Available from: https://doi.org/10.1039/D2DT02766F
20. Dos Santos AF, de Almeida DRQ, Terra LF, Baptista MS, Labriola L. Photodynamic therapy in cancer treatment - an update review. J Cancer Metastasis Treat. 2019;5:25. Available from: https://doi.org/10.20517/2394-4722.2018.83
21. Land JE, Raith W. Fine structure of O2− measured by electron time-of-flight spectroscopy. Phys Rev Lett. 1973;30:349-352. Available from: https://doi.org/10.1103/PhysRevLett.30.349.2
22. Panchenko OO, Minaev BF. Enzymatic spin-catalysis in flavin-containing oxidases and magnetic orientation of birds. Cherkasy Univ Bull Biol Sci Ser. 2018;1:114-120. Available from: https://bio-ejournal.cdu.edu.ua/article/view/2804
23. Minaev BF, Lunell S, Kobzev GI. Collision-induced intensity of the b1Σg+−a1Δg transition in molecular oxygen: Model calculations for the collision complex O2+H2. Int J Quantum Chem. 1994;50(4):279-292. Available from: https://doi.org/10.1002/qua.560500405
24. Minaev BF. Quantum-chemical investigation of the mechanisms of the photosensitization, luminescence, and quenching of singlet 1Δg oxygen in solutions. J Appl Spectrosc. 1985;42(5):518-523. Available from: https://link.springer.com/article/10.1007/BF00661398