Abstract
Photodynamic therapy (PDT) is a light-activated modality that kills cells through photosensitizer-triggered reactive oxygen species, but its efficacy is strongly shaped by oxygen availability and by whether photochemistry proceeds through Type I or Type II pathways. In oxygen-rich settings, Type II PDT mainly generates singlet oxygen, whereas hypoxia suppresses this route and shifts interest toward Type I mechanisms that produce radical species such as superoxide, hydroxyl radicals, and hydrogen peroxide. This review argues that hypoxia-driven resistance in vitro is not explained by oxygen depletion alone, but by interactions among photochemical pathway choice, cellular adaptation, photosensitizer uptake and localization, and assay limitations. It highlights that hypoxia can reshape membrane composition, vesicular trafficking, and organelle targeting, which in turn alters reactive oxygen species patterns and photodamage outcomes. Methodologically, the work recommends Controlled hypoxia models, including 2D low oxygen cultures and 3D spheroids, together with orthogonal readouts of hypoxia and oxidative damage. For ROS profiling, it supports comparing normoxic and hypoxic conditions using probes for total ROS, superoxide, hydroxyl radicals, and singlet oxygen, because no single assay captures the full mechanistic picture. Finally, it emphasizes that lipid and protein oxidation should be interpreted with caution in hypoxic PDT, since FOX, TBARS, iodometric methods, and BODIPY-C11 mainly report secondary or localized products and may miss Type I driven damage unless combined with imaging, membrane integrity assays, and time-resolved measurements.
Keywords: PDT, Hypoxia experimental protocol, ROS, Biomolecule oxidation
Introduction
Photodynamic therapy (PDT) is a light-activated therapeutic modality that relies on the excitation of a photosensitizer (PS) to generate reactive oxygen species (ROS) capable of inducing oxidative damage and cell death. Upon photoexcitation, PSs populate excited singlet and triplet states, as described by the Jablonski diagram (Fig. 1) (Jabłoński 1935), enabling two major photochemical pathways: the Type II mechanism, based on energy transfer to molecular oxygen to produce singlet oxygen (1O2), and the Type I mechanism, which involves electron or hydrogen transfer reactions generating radical species such as superoxide (O2•⁻), hydroxyl radicals (•OH), and hydrogen peroxide (H2O2) (Baptista et al. 2017).
Fig. 1.

Representation of Jablonski diagram in PDT context
The production of 1O2 depends on triplet-triplet energy transfer (TETT) between triplet oxygen (3O2) and the PSs in triplet state excited (PS 3T1*) by intersystem crossing (ISC) as shown in Equations (1) and (2) (Wu et al. 2017). In the Type II mechanism, energy transfer from the PS in the triplet excited state to triplet oxygen results in the generation of singlet oxygen (1O2):
| 1 |
| 2 |
When the PS is in the excited triplet state (PS 3T1*), electron transfer can occur from PS to the suppressor Q (e.g., vitamin C, riboflavin, reduced coenzymes (Pereira et al. 2017), amino acids, and nitrogenous bases to be oxidized), producing a radical pair: PS 3T1* + Q → PS•− + Q•+. Starting from this radical pair, a cascade reaction of electron transfer can occur (Equations 3-5). The O2•⁻ radical anion can then be formed by electron transfer between the radical anion of PS and oxygen (PS•− + O2 → PS 1S0 + O2•⁻). O2•⁻ is considered the primary reactive oxygen species generated through the Type I pathway and can subsequently give rise to hydroperoxyl radicals (HOO•), hydrogen peroxide (H2O2), and hydroxyl (•OH) through successive reactions (Equations 3–5) (Ochsner 1997; Da and Machado 2000).
| 3 |
| 4 |
| 5 |
O2•− and H2O2 have limited biological toxicity that oppositely endows them broad diffusion range (O2•− and H2O2 diffusion rate in H2O at 298 K 1.5 10−8 m2 s−1 and 2 10−8 m2 s−1, respectively) to generate the most reactive HO• (diffusion rate = 3 10−9 m2 s−1 H2O at 298 K). •OH is extremely destructive to almost all biological molecules (Li et al. 2022a), which enables the limited oxygen in the hypoxic tumor to be fully utilized. But it is still oxygen-dependent. The term “oxygen-independent” should not be used. As for Type II pathway, the PS 3T1* transfers its energy to surrounding oxygen, forming the 1O2 with the decent cytotoxicity but short action range.
Traditionally, PDT has been conceptualized as a predominantly Type II process (Teng et al. 2025), implicitly assuming sufficient oxygen availability for efficient 1O2 generation. However, this paradigm is increasingly challenged by the recognition that oxygen availability is highly limited in solid tumor microenvironments, and both pre-existing hypoxia and additional oxygen consumption during illumination severely compromise 1O2 generation, making oxygen dependence a central cause of PDT failure in solid tumors (Lv et al. 2018; Li et al. 2022b).
In response to this limitation, recent studies have focused from purely 1O2‑centric designs toward PSs and photochemical schemes that either economize oxygen use or operate efficiently at low to negligible O2 levels (Li et al. 2020, 2022b). Comprehensive analyses of Type I PDT now show that appropriately engineered PSs can favor charge-separated states and electron transfer over energy transfer (Teng et al. 2025), thereby amplifying radical ROS such as O2•⁻ and •OH and maintaining photocytotoxicity under hypoxic conditions where classical Type II agents fail (Yu et al. 2022). At the same time, hybrid Type I/II systems capable of generating both 1O2 and O2•⁻, such as boron-dipyrromethanes (BODIPYs) (Wang et al. 2012), naphthalimide derivatives (Nguyen et al. 2019), biotinylated porphyrinoids (An et al. 2022), and the methylene blue derivatives ENBS-B (Li et al. 2018a), ENBOS (Li et al. 2019), and SORgenTAM (Li et al. 2020), have demonstrated that radical-mediated photochemistry can partially circumvent oxygen dependence, revitalizing PDT efficacy under hypoxic conditions, underscoring the potential of mixed photochemical regimes to overcome hypoxia‑induced resistance.
These advances have stimulated rational design principles for Type I and Type I/II photodynamic agents, including tuning redox potentials, managing intersystem crossing and charge-separation lifetimes, and exploiting aggregation-induced emission to stabilize long-lived radical intermediates (Li et al. 2018b; Chen et al. 2021b; Teng et al. 2025). Yet, PS photophysical characterization remains dominated by measurements performed in air-saturated solutions—molar absorptivity, triplet lifetimes, and 1O2 quantum yields (ΦΔ)—while oxygen-dependent radical yields and their evolution under hypoxia are rarely quantified in a systematic fashion. As a result, PSs optimized to maximize ΦΔ in normoxic media often display disappointing performance in hypoxic cellular assays, whereas Type I–biased candidates show robust activity that is not predictable from conventional ΦΔ alone (Li et al. 2018b; Teng et al. 2025). These findings have shifted attention toward rational PS design strategies that favor electron-transfer processes, redox cycling, and intracellular radical amplification rather than exclusive reliance on 1O2 generation (Li et al. 2018b; Teng et al. 2025).
Modeling hypoxia in vitro represents a critical step toward understanding PDT resistance at the cellular level. Hypoxia can be induced using physical approaches, such as hypoxic chambers or incubators with controlled oxygen atmospheres (typically 1–5% O2) (Matthiesen et al. 2021a; Yu et al. 2022). While conventional two-dimensional (2D) monolayer cultures remain the most widely used models in PDT research, they fail to reproduce oxygen gradients, diffusion limitations, and metabolic heterogeneity characteristic of tumors. Three-dimensional (3D) spheroids and multicellular aggregates provide more physiologically relevant hypoxic microenvironments, often exhibiting intrinsic resistance to PDT compared to 2D cultures (Kucinska et al. 2021). Furthermore, prolonged hypoxic conditioning can select resistant cell populations with altered redox balance, mitochondrial function, and stress-response pathways, making cell line choice and culture architecture decisive variables in vitro PDT studies (Ferguson et al. 2018; Shirmanova et al. 2024).
Hypoxic tumor environments not only limit oxygen availability for Type II photoprocesses but also remodel the cellular microenvironment in ways that affect PS uptake and subcellular localization (Xu et al. 2019; Sun et al. 2020). Under low oxygen tension, hypoxia-inducible factor 1 (HIF-1)–driven pathways reprogram cellular metabolism (Infantino et al. 2021), alter lipid composition of membrane (Botto et al. 2008; Seo et al. 2022), modulate vesicular trafficking (Muñiz-García et al. 2022), and change the expression of transporters and stress-response proteins (Parks et al. 2016), which in turn may modify PS internalization efficiency, intracellular trafficking, and retention compared with normoxic conditions. Recent studies further show that redirecting PS localization among mitochondria (Xu et al. 2019), lysosomes (Lv et al. 2016a), or endoplasmic reticulum (Li et al. 2022c) markedly reshapes the spatial and temporal patterns of ROS formation and photodynamic response under hypoxia conditions (Ihata et al. 2022; Alva et al. 2024).
At the molecular level, PDT resistance in hypoxic settings is linked both to shifts in the dominant ROS species (Chen et al. 2021b; Alva et al. 2024) and to an altered antioxidant defense capacity (Kilinç and Güney 2019; Shekhovtsov et al. 2024). Moreover, many classical assays of oxidative damage to proteins and lipids were originally validated under normoxic conditions and may underestimate or distort the extent and nature of injury under reduced O2 (Hong et al. 2022; Du et al. 2022; Li et al. 2023).
Taken together, these observations indicate that PDT resistance observed in vitro under hypoxia is not merely a direct consequence of reduced oxygen availability for 1O2 formation, but rather arises from a complex interplay between competing photochemical pathways, hypoxia-driven cellular adaptations, and methodological constraints inherent to the models and assays used to evaluate PDT efficacy.
This review examines hypoxia-induced resistance to PDT in vitro cell models, with a particular focus on how experimental design and methodological choices influence the interpretation of PDT efficacy. By integrating these perspectives, the aim is to provide an experimental roadmap that helps in selecting and benchmarking next‑generation PSs in hypoxic tumor settings highlighting common pitfalls in current evaluation pipelines. Such an approach is essential for understanding and guiding the rational development of new PS.
In vitro hypoxia methods
In vitro hypoxia models are essential to dissect how oxygen availability and microenvironmental architecture shape PDT response (Kucinska et al. 2021), especially when comparing Type II versus Type I/II PSs. Conceptually, these models can be grouped into the commonly 2D cell culture, where oxygen concentration is externally defined, and 3D platforms that generate intrinsic oxygen gradients and diffusion barriers closer to solid tumors (Wulftange et al. 2019; Salyha and Oliynyk 2023a).
2D monolayer
In 2D monolayers cell culture, hypoxia is most commonly achieved using gas-controlled incubators (Fig. 2A) or modular chambers that maintain cells at 1–5% O2 for 24–48 h before PDT, yielding reproducible stabilization of HIF-1α and metabolic adaptation (Wu and Yotnda 2011). Commercial modular incubator chambers (Fig. 2B), such as polycarbonate units connected to premixed gas cylinders (Kaur et al. 2023), allow precise adjustment of O2, CO2, and humidity and can be placed inside standard CO2 incubators, providing a cost-effective alternative to dedicated hypoxia workstations for PDT studies. In addition, “vacuum bag” techniques (Fig. 2C), where culture vessels are sealed in low-permeability storage bags, evacuated, and backfilled with defined gas mixtures, offer a simple way to generate low oxygen environments without specialized incubators and have been successfully used to expose cells and tissues to controlled hypoxic conditions (Bakmiwewa et al. 2015; Matthiesen et al. 2021b). Under these conditions, several studies have reported marked reductions in 1O2 production and PDT efficacy for classical Type II PSs, consistent with an oxygen-limited regime (Nguyen et al. 2021; Chen et al. 2021b; Li et al. 2022b; Teng et al. 2025).
Fig. 2.

Experimental approaches used to establish hypoxic conditions in 2D cell culture models for PDT studies: (A) gas-controlled hypoxia incubators or workstations, (B) modular hypoxia chamber connected to premixed gas cylinders, and (C) vacuum-sealed hypoxia system using low-permeability bags and oxygen indicators. These methods are commonly used to maintain cells under low O2 conditions prior to PDT irradiation
Chemical hypoxia mimetics provide a complementary approach in conventional incubators by stabilizing HIF-1α or decreasing mitochondrial oxygen consumption. Cobalt chloride (CoCl2) in the 100–600 µM range (Wu and Yotnda 2011; Kaur et al. (2023) is widely used to mimic chronic hypoxia in tumor cell lines by inhibiting prolyl hydroxylases and promoting HIF-1α accumulation, whereas desferrioxamine (DFO) at ~ 100 µM chelates iron and indirectly suppresses oxidative metabolism (Guo et al. 2006; Rinderknecht et al. 2021). These treatments reproduce key transcriptional hallmarks of hypoxia and are useful for probing hypoxia-driven resistance pathways; however, they do not fully recapitulate diffusion-limited oxygen gradients and may introduce off-target toxicity that must be carefully controlled in PDT experiments.
Beyond controlling external oxygen tension, it is essential to verify that cells are genuinely experiencing hypoxia at the molecular level using orthogonal readouts. Stabilization and nuclear accumulation of HIF-1α are commonly assessed by Western blot or immunofluorescence after 12–24 h at 1–5% O2 or in the presence of hypoxia mimetics (Wu and Yotnda 2011; Matthiesen et al. 2021b; Rinderknecht et al. 2021), providing a first-line confirmation that hypoxia-responsive signaling is activated.
In addition to HIF-1α immunodetection, several complementary assays are available to confirm that cells are truly hypoxic rather than merely exposed to reduced ambient O2. Oxygen-sensitive fluorescent probes such as Image-iT™ Green Hypoxia Reagent become fluorescent under low intracellular O2 concentration and have been used to visualize and quantify hypoxia in live cultures (Matthiesen et al. 2021b; Rinderknecht et al. 2021). Luciferase-based hypoxia reporters, in which firefly luciferase activity reflects HIF-dependent transcription, allow quantitative assessment of hypoxia-responsive signaling in living cells and have also been applied to validate hypoxic culture protocols (Wu and Yotnda 2011). In parallel, adduct formation by the 2-nitroimidazole pimonidazole under low pO2 can be detected by immunostaining or flow cytometry and is widely used as an independent marker of chronic or severe hypoxia (Aguilera and Brekken 2014).
3D spheroids cell model
Studies involving cell culture are widely used for clinical trials to aid in the development of new diagnostic tools and drug discovery. Two-dimensional (2D) cell culture approaches are gradually being replaced by three-dimensional (3D) cell culture methods because they provide more accurate data regarding the cell growth scenario (Chandra and Singh 2017; Mohammad-Hadi et al. 2018; Becceneri et al. 2020, 2024; Kucinska et al. 2021). As an in vitro model, three-dimensional cell culture allows researchers to conduct more biologically relevant studies than two-dimensional cell cultures (Engrácia et al. 2023).
In the 2D cell culture method (Fig. 3), the cell shape tends to be flat and elongated due to its ability to grow only on a two-dimensional surface, while in 3D cell culture (Fig. 3), the natural shape of the cell will be preserved during growth, allowing cell-to-cell interactions (Engrácia et al. 2023). In vivo, solid tumors are heterogeneously oxygenated and contain hypoxic regions that can lead to necrosis (Pereira et al. 2017). In order to mimic and better understand the action of photosensitizing drugs in tumor models using PDT, 3D spheroids have been used (Evans 2015; Pereira et al. 2017).
Fig. 3.

Schematic representation of conventional 2D monolayer and 3D spheroid culture models, illustrating oxygen and nutrient diffusion and cellular organization
Spheroids can closely resemble the properties of tumors, as they are three-dimensional cell aggregates with a necrotic core surrounded by a layer of viable cells (Evans 2015; Pereira et al. 2017) and serve as excellent models for measuring the effects of oxygen diffusion and consumption, both during normal growth and PDT (Evans 2015).
In addition to the oxygen consumption in the architecture of 3D cell models, which mimic in vivo dynamics, and the penetration of visible light into 3D models, since the constructs are relatively thin (typically < 1 mm), partly to allow for oxygen replenishment, the attenuation of light for PDT activation becomes negligible (Mohammad-Hadi et al. 2018).
Some cell lines have the ability to produce spheroids, and other cell types will only produce cell aggregates. The cell heterogeneity and metabolic gradients (including oxygen, glucose, and ATP) become apparent in spheroids only when they are larger than 150 µm. Spheroids with a necrotic and hypoxic central region become apparent only at a diameter of ~ 500 µm. The necrotic central core model in 3D models occurs due to (i) nutrient and metabolite deficiency, (ii) variation in oxygen concentration, (iii) accumulation of waste products, and (iv) low pH. In in vivo solid tumor models, the necrotic core appears at a distance > 100 µm from a blood supply (Raza et al. 2017).
The spheroids’ size and hypoxic regions determine how they grow, behave, interact, and respond to the external environment, making it a crucial parameter for evaluating various properties. One of the challenges when working with 3D culture relates to image quality, where the images obtained must show accuracy in thickness and size laterally and perpendicularly. In cases of large spheroids, light transmission decreases (Yoo et al. 2024).
However, one of the challenges of working with 3D cells is visualizing cell growth and interactions compared to 2D models (Yu et al. 2022; An et al. 2022). Imaging techniques can provide valuable information about their dynamics and interactions with other cell types, helping to improve our understanding of tissue regeneration.
Specifically, studies involving PSs in 3D models are experimentally helpful since many PSs exhibit fluorescence emission and can be employed to dynamically visualize the uptake and localization of these agents at the cellular and subcellular level (Evans 2015). 3D spheroid cultures, which simulate avascular tumors, have been helpful in modeling the uptake and penetration of PSs through multiple cellular and subcellular levels.
Xiao et al. (2005), using a spheroid bladder cancer model with a diameter of less than 400 µm, examined the uptake distributions of several porphyrin PSs using confocal microscopy. The choice of different types of PSs was based on their lipophilic nature. Aluminum phthalocyanine chloride (AlPC) falls into the hydrophobic group. Protoporphyrin-IX (PpIX) and benzoporphyrin derivative monoacid ring A (BPD-MA) are on the borderline between the hydrophobic and amphiphilic groups. Hypocrellins (HBEA-R1) is amphiphilic. The results of the distribution of PSs in the spheroid showed that the hypocrellins and benzoporphyrin groups presented a greater intratumor distribution compared to AlPC and PpIX (Xiao et al. 2005).
In a recent study by Li et al. (2024), spheroids of MCF-7 human breast tumor cells were constructed and different doses of violet light at 408 nm and red light at 625 nm were applied under normal and hypoxic oxygen conditions in the presence of a PS. The cell viability results indicated that the light irradiated at 408 nm was more efficient at killing cells in the spheroids under normal oxygen conditions, while cells exposed to the hypoxic microenvironment exhibited minimal PDT-induced death (Li et al. 2024).
Thus, the systematic incorporation of spheroids into preclinical PDT studies represents a fundamental step toward increasing the translational relevance of experimental results. The continuous improvement of culture, imaging, and tumor microenvironment control techniques should further expand the potential of these models, contributing to the rational development of more effective PSs and optimized therapeutic protocols.
Measuring ROS under normoxia and hypoxia conditions
ROS is a collective term for oxygen-derived molecules that are more reactive than molecular oxygen (O2), this group 1O2, O2⁻•, H2O2, •OH, and HOO• (Baptista et al. 2017). In PDT, the relative contribution of these species depends strongly on oxygen availability. Under hypoxia, 1O2 formation through classical Type II pathways decreases with falling pO2, whereas electron-transfer mediated Type I pathways may remain operative and become comparatively more relevant (Chen et al. 2021b; Teng et al. 2025). Therefore, comparing ROS generation under normoxic and hypoxic conditions may help distinguish whether a PS acts predominantly through Type II, Type I, or mixed photosensitizing mechanisms. The following subsections summarize experimental approaches used to detect ROS and evaluate oxidative processes in PDT systems.
Detection of total ROS
2′,7′-Dichlorofluorescein diacetate (DCFH-DA) is a widely used fluorogenic probe for detecting ROS and assessing overall oxidative stress in cells (Zhu et al. 2017; Murphy et al. 2022). This membrane-permeable compound diffuses into cells, where it is hydrolyzed by intracellular esterases to form the non-fluorescent 2′7′-dichlorofluorescin (DCFH), as depicted in Fig. 4A. Once formed, DCFH is retained within the cell and reacts with various ROS to generate the highly fluorescent 2′,7′-dichlorofluorescein (DCF), which can be quantified using flow cytometry, fluorescence plate readers, or fluorescence microscopy (Zhu et al. 2017; Ng and Ooi 2021; Yu et al. 2021). DCF fluorescence is typically measured at 538 nm following excitation at 488 nm.
Fig. 4.

Schematic representation of the chemical mechanisms involved in the detection and quantification of reactive oxygen species (ROS) in vitro: (A) ROS detection using DCFH-DA; (B) superoxide anion (O2•⁻) detection using DHE; (C) superoxide anion (O2•⁻) detection using NBT; (D) hydroxyl radical (•OH) detection using APF and HPF fluorescent probes; (E) singlet oxygen (1O2) detection using the SOSG probe and F RNO assay
DCFH-DA assays are commonly used as an initial, semi-quantitative screening probe to compare overall ROS generation under normoxic and hypoxic conditions (Lv et al. 2018; Teng et al. 2025). Under normoxia, increased fluorescence generally reflects the global oxidative burst induced by irradiation and is often used to validate light-triggered ROS production and evaluate the overall photosensitizing performance of a PS (Lv et al. 2018). In this context, DCFH-DA measurements are commonly complemented by more selective probes, such as dihydroethidium (DHE) for O2•⁻ (Yu et al. 2022; An et al. 2022) or Singlet Oxygen Sensor Green (SOSG) for 1O2 (Yu et al. 2022), in order to distinguish between Type I and Type II pathways. Normoxic experiments are also frequently used to establish baseline photochemical parameters, including total ROS generation, 1O2 quantum yield, and cellular phototoxicity. For example, Lv et al. (2018) showed that the cyclometalated Ru(II) complex Ru2 generated higher total ROS under normoxic conditions despite a relatively modest 1O2 yield, supporting the view that some PSs can act as efficient Type I biased ROS generators even when Type II activity is limited (Lv et al. 2018).
Under hypoxia, persistence of DCF fluorescence may indicate low-O2-dependent ROS generation pathways associated with Type I photochemistry (Li et al. 2022b; An et al. 2022). This comparison is particularly relevant because hypoxic tumor environments strongly limit classical Type II PDT, which depends on molecular oxygen for 1O2 production (Lv et al. 2018; Li et al. 2022b). Moreover, irradiation itself can aggravate local oxygen depletion through continuous O2 consumption (Lv et al. 2018). Consequently, comparing DCFH-DA signals under normoxia and hypoxia has become a useful strategy for evaluating whether a PS retains photodynamic activity under low oxygen conditions (Lv et al. 2018). At least one study reported measurable DCF fluorescence at 1% O2, suggesting that electron-transfer mediated ROS generation pathways remain operative under hypoxia (An et al. 2022).
It is important to note that DCFH-DA is a broad-spectrum indicator of oxidative stress and lacks specificity for individual reactive species. It does not selectively detect H2O2, lipid peroxides, 1O2, or O2•⁻. Therefore, the inclusion of appropriate controls is recommended. For instance, oxidative activity can be simulated using H2O2 or tert-butyl hydroperoxide (TBHP), depending on the cell type and analysis duration (Eruslanov and Kusmartsev 2010; Ng and Ooi 2021). Negative controls designed to suppress the ROS-generating process, including gene knockouts, knockdowns, inhibitors, or radical scavengers can help confirm which ROS are responsible for the oxidative damage (Murphy et al. 2022). Additionally, although DCFH-DA assays are widely employed in PDT studies, they are generally considered semi-quantitative rather than truly quantitative measurements of ROS production. Recent reviews on Type I PDT have further emphasized that robust quantitative methodologies for assessing Type I ROS generation and quantum yields are still lacking (Lv et al. 2018; Chen et al. 2021b). To complement its use, other probes have been applied in an attempt to detect specific reactive species (Eruslanov and Kusmartsev 2010).
Detection of superoxide anion (O2•⁻)
O2•⁻ exhibits selective reactivity, as it does not readily oxidize most biological molecules, but can specifically damage certain enzymes containing Fe–S clusters (Murphy et al. 2022). In simplified systems, O2•⁻ quantified through the SOD-inhibitable reduction of cytochrome c (Nauseef 2014). However, in cellular systems, detection is commonly achieved using the fluorescence generated by the oxidation of dihydroethidium (DHE) (Fig. 4B) or its mitochondria-targeted analogue, MitoSOX (Murphy et al. 2022).
Hydroethidine (HE)-based probes are widely employed for intracellular O2•⁻ detection (Bilski et al. 2009; Price and Kessel 2010; Zielonka and Kalyanaraman 2010; Kessel et al. 2011; Kalyanaraman et al. 2014; Ortega-Villasante et al. 2018). However, their specificity is often limited, as these dyes can also react with other ROS, including 1O2, •OH, and H2O2 (Bilski et al. 2009; Zielonka and Kalyanaraman 2010; Ortega-Villasante et al. 2018). Consequently, the resulting oxidation products may not exclusively reflect O2•⁻ activity. Currently, high-performance liquid chromatography (HPLC) is considered the most reliable method for identifying specific HE oxidation products (Kalyanaraman et al. 2014), since fluorescence microscopy alone does not allow definitive discrimination of intracellular O2•⁻ generation (Zielonka and Kalyanaraman 2010).
In the context of PDT, detection of O2•⁻ has become particularly relevant for distinguishing hypoxia-tolerant Type I pathways from conventional oxygen-dependent Type II mechanisms (Li et al. 2022b; Teng et al. 2025). Under normoxia, DHE fluorescence is commonly employed to validate the ability of a PS to generate O2•⁻ through electron-transfer reactions (Li et al. 2020; Yu et al. 2022). In these conditions, O2•⁻ measurements are frequently interpreted together with total ROS and 1O2 assays to establish mechanistic reference points for PS activity (Li et al. 2020; Yu et al. 2022). Under hypoxia, O2•⁻ detection becomes especially informative because several Type I PS retain O2•⁻ generation even when 1O2 production is strongly suppressed by oxygen limitation (Yu et al. 2022; An et al. 2022). In some cases, DHE signals persist or even increase under low oxygen conditions, supporting the interpretation that electron-transfer-mediated ROS pathways remain active in hypoxic environments (Li et al. 2020; An et al. 2022).
Mamalis et al. (2016) investigated intracellular O2•⁻ generation in human dermal fibroblasts treated with aminolevulinic acid (ALA)-based PDT, employing DHE as a probe due to its relative selectivity for O2•⁻. Cells were pre-incubated with ALA, followed by DHE loading and subsequent irradiation. O2•⁻ production was then quantified by flow cytometry, using fluorescence intensity as a proxy for intracellular O2•⁻ levels (Mamalis et al. 2016). Similarly, Dolgachev et al. (2005) evaluated O2•⁻ formation in cells subjected to PDT with a phthalocyanine-based PS (Dolgachev et al. 2005). More recently, Type I PSs such as SORgenTAM (Li et al. 2020; Chen et al. 2021b) and NI-TA (Yu et al. 2022) demonstrated sustained O2•⁻ generation under hypoxic conditions, reinforcing the importance of O2•⁻ detection as a mechanistic indicator of a low O2-dependent mechanism.
The Organization for Economic Cooperation and Development (OECD) adopted Test Guideline No. 495: Reactive Oxygen Species (ROS) Assay for Photoreactivity, last reviewed and confirmed in 2019, which recommends the nitroblue tetrazolium (NBT) assay for detecting O2⁻•. In this assay, O2⁻• reduces NBT via a one-electron transfer reaction, forming a partially reduced monoformazan derivative (NBT⁺), as illustrated in Fig. 4C (OECD 2024). This stable intermediate absorbs light at 560 nm, allowing spectrophotometric quantification of O2⁻• (OECD 2024).
In the context of PDT, some NBT assays were conducted in solution rather than in cell-based systems, often using NADH as a positive control to induce O2⁻• generation (Ikeda et al. 2017; Wen et al. 2017; Puttaswamy et al. 2023). In contrast, other studies applied the assay in cellular environments. For example, Malacarne (2023) and Panzarini et al. (2011) monitored ROS production in cells following PDT treatment using NBT (Panzarini et al. 2011; Malacarne et al. 2023). Kessel (2011) also employed NBT to detect intracellular O2⁻• formation, reporting higher selectivity compared to DHE (Kessel et al. 2011). Despite its sensitivity to O2⁻• (and to a lesser extent to •OH), the non-fluorescent nature of NBT limits its utility in fluorescence-based applications (Kessel et al. 2011). Nevertheless, because O2⁻• generation is frequently preserved under hypoxia in Type I PDT systems, NBT assays may provide valuable complementary information when comparing ROS production under normoxic and hypoxic conditions.
Detection of hydroxyl radical (•OH)
•OH radicals are highly reactive radical species that interact indiscriminately with nearby molecules at nearly diffusion-controlled rates (Murphy et al. 2022). To detect these radicals, Setsukinai et al. (2003) developed two fluorescent probes: 2-[6-(4’-hydroxy)phenoxy-3H-xanthen-3-on-9-yl]benzoic acid (HPF) and 2-[6-(4’-amino)phenoxy-3H-xanthen-3-on-9-yl]benzoic acid (APF) (Setsukinai et al. 2003). Initially non-fluorescent, these probes emit fluorescence upon reaction with •OH, forming stable O-dearylation products (Fig. 4D). The use of hydroxyl radical scavengers, such as dimethyl sulfoxide (DMSO), significantly reduces fluorescence, supporting their preferential reactivity with •OH. Due to their stability and resistance to autooxidation, HPF and APF are especially suited for use in cellular environments (Setsukinai et al. 2003; Gomes et al. 2005). Although APF is considered relatively selective for •OH, particularly in the presence of HRP, studies have also suggested that it may interact with 1O2 as well (Price and Kessel 2010).
The detection of •OH in PDT-treated systems has become increasingly relevant in studies investigating Type I photochemical pathways under hypoxia (Chen et al. 2021b; Teng et al. 2025). Unlike 1O2 generation, which strongly depends on molecular oxygen, •OH may arise from downstream radical cascades involving O2•⁻ disproportionation, Haber–Weiss reactions, and Fenton-type processes (Ochsner 1997; Yu et al. 2022; Li et al. 2022b). Consequently, the persistence of •OH generation under low oxygen conditions is often interpreted as evidence of “oxygen-independent” or partially oxygen-independent ROS production (Lv et al. 2018; Li et al. 2022b).
The detection of ROS in PDT-treated systems has been explored using APF, a cell-permeable probe sensitive to highly reactive species such as •OH and 1O2. Price et al. (2009) demonstrated its utility in mammalian cells by incubating them with both the PS and APF, with •OH scavengers like DMSO as a negative control. Following irradiation, fluorescence microscopy confirmed ROS generation, and the inclusion of DMSO enabled the assessment of •OH involvement specifically (Price et al. 2009). In another approach applied to microbial systems, Sułek et al. (2020) employed APF in bacterial PDT studies using dual-color flow cytometry. Bacteria were incubated with a porphyrin-based PS and APF, then irradiated, allowing ROS detection and characterization of oxidative stress responses via flow cytometry (Sułek et al. 2020).
Type I PDT systems active under hypoxia have demonstrated significant •OH formation through downstream radical cascades. Lv et al. (2018) reported that the cyclometalated Ru(II) PS Ru2 retained PDT efficacy under both normoxia and hypoxia, and DMPO-EPR detected DMPO-OH adducts, supporting •OH as a major contributor to phototoxicity (Lv et al. 2018). Similarly, NI-TA was proposed to generate O2⁻• that feeds SOD, Haber–Weiss, and Fenton chemistry to produce •OH under hypoxia (Yu et al. 2022). In that study, HPF was used as the •OH fluorescent probe, while DMPO was used for EPR detection of O2⁻• (Yu et al. 2022).
Detection of singlet oxygen (1O2) in cells
Two singlet states of O2 are known, but only the 1Δg state, although not a free radical, is of major biological significance. This singlet configuration makes O2 far more reactive than its ground-state triplet form. 1O2 can be generated via photosensitization, where a PS absorbs light and transfers the energy to ground-state O2, as well as through chemical reactions involving HOO• radicals, among others (Murphy et al. 2022).
Detecting 1O2 is challenging due to its high reactivity and the lack of selectivity of most probes. However, Singlet Oxygen Sensor Green (SOSG) has emerged as a relatively selective probe. In aqueous solution, SOSG emits weak blue fluorescence, initially quenched by photo-induced electron transfer from the anthracene donor to the fluorophore. Upon reaction with 1O2, a stable endoperoxide is formed (Fig. 4E), which suppresses this quenching and leads to a strong green fluorescence with emission peaks at 525–536 nm (excitation: 375 and 393 nm) (Ragàs et al. 2009; Kim et al. 2013b).
SOSG has been widely used to detect intracellular 1O2 in vitro using fluorescence microscopy (Flors et al. 2006; Hatz et al. 2007; Gollmer et al. 2011; Prasad et al. 2018; Chen et al. 2021a, 2025). Although traditionally considered cell-impermeable, evidence suggests that SOSG can be internalized under specific conditions, such as cell stress or in the absence of proteins (Peng et al. 2020). For example, SOSG has been incorporated into HGC-27 gastric cancer cells to detect 1O2 generated by polypyridyl Os(II) complexes, using a concentration of 2.5 µM in DMSO (Chen et al. 2025).
Similarly, SOSG uptake has been demonstrated in HeLa cells when cultured in standard maintenance medium (SMM), a protein-free culture medium (Hatz et al. 2007). In contrast, no significant uptake was observed when cells were maintained in conventional Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with fetal calf serum and antibiotics. This suggests that SOSG binds strongly to proteins, which can inhibit its internalization and alter its intracellular behavior (Prasad et al. 2018). Therefore, SOSG has been utilized both in cell-free systems, where potassium iodide served as a negative control (Wen et al. 2017), and in cellular environments, where its application was combined with confocal fluorescence microscopy to monitor intracellular 1O2 generation (Chen et al. 2015).
In PDT studies, 1O2 detection under normoxic conditions is commonly used to evaluate classical Type II photochemical activity and determine baseline photodynamic efficiency (Li et al. 2022b). Under air-saturated conditions, strong SOSG signals are generally interpreted as evidence of efficient energy transfer from the excited PS to molecular oxygen (Yu et al. 2022). Consequently, normoxic experiments are frequently used to determine 1O2 quantum yields and compare the Type II performance of different PSs (Lv et al. 2018).
In contrast, hypoxic experiments serve as an important stress test for oxygen dependence in PDT systems (Lv et al. 2018; An et al. 2022). Because Type II PDT relies directly on molecular oxygen, 1O2 generation is typically attenuated under low oxygen conditions, and irradiation itself may further aggravate oxygen depletion (Lv et al. 2018; Li et al. 2020). In some hypoxia-tolerant PDT systems, reduced SOSG fluorescence under hypoxia was accompanied by persistent DHE or DCFH-DA signals, supporting the interpretation that ROS generation shifted from Type II to Type I pathways (Yu et al. 2022; An et al. 2022). These observations reinforce the idea that 1O2 production alone is insufficient to evaluate PDT efficiency under hypoxic tumor conditions (Lv et al. 2018; Yu et al. 2022; Li et al. 2022b).
The Organization for Economic Cooperation and Development (OECD) adopted Test Guideline No. 495: ROS Assay for Photoreactivity last reviewed and confirmed in 2019 (OECD 2024), which recommends the p-nitrosodimethylamine (RNO) assay for detection of 1O2. This method is based on the procedure developed by Kraljic and El Mohsni, in which the generation of 1O2 is monitored by the spectrophotometric measurement of RNO bleaching, observed as a decrease in absorbance at 440 nm (Kraljić and Mohsni 1978; Onoue et al. 2008).
Although 1O2 does not directly react with RNO, the bleaching effect results from the interaction of 1O2 with an imidazole ring, leading to the formation of a trans-annular peroxide intermediate. This reactive intermediate subsequently oxidizes RNO, causing its decolorization (Fig. 4F). In the context of PDT, this assay is typically performed in aqueous solution in the presence of histidine, which serves as the imidazole source (Liu et al. 2008; Kim et al. 2013a; Simaremare et al. 2015; Urazaliyeva et al. 2024).
Cellular uptake and subcellular localization
Cellular uptake and subcellular localization of PSs are key determinants of PDT efficacy, as they govern both intracellular PS concentration and proximity to critical biomolecular targets (Castano et al. 2005; Benov 2015). Given the short diffusion radius and lifetime of ROS, oxidative damage is largely confined to the immediate microenvironment of the photoexcited PS, making this spatial PS location particularly important for treatment outcome (Tsubone et al. 2019a, b, 2021).
Many PSs inherently fluoresce upon exposure to specific excitation wavelengths, rendering flow cytometry (Salvati et al. 2018) or microplate readers (Shin et al. 2022), both effective methods for quantifying their uptake by mammalian cells or bacteria (Kiesslich et al. 2013). When a PS lacks intrinsic fluorescence, it can be conjugated to a fluorescent marker to facilitate detection and analysis. This process involves measuring the fluorescence intensity of individual cells after incubation with the PS for a defined period. Critical experimental variables, such as PS concentration, incubation duration, and environmental conditions (e.g., hypoxia, temperature, pH), must be optimized based on the specific goals and properties of the PS.
Fluorescence intensity is directly proportional to the number of PS accumulated within each cell. The flow cytometry data are typically represented in histograms or scatter plots, where the x-axis indicates fluorescence intensity (reflecting PS uptake), and the y-axis represents cell count or frequency. Cells with higher fluorescence intensities indicate greater uptake of the PS. The data can be used to quantify the percentage of cells that have internalized the PS and to compare uptake levels across different cell types or experimental conditions, such as normoxia or hypoxia (Kiesslich et al. 2013).
Confocal laser scanning fluorescence microscopy has improved the analysis of PS subcellular localization, offering higher sensitivity and spatial resolution compared to nonconfocal methods (Woodburn et al. 1991). Colocalization with organelle-specific fluorescent probes—whose emission spectra differ from that of the PS—allows precise identification of the accumulation site (Tsubone et al. 2019b). These probes can also highlight damaged regions postirradiation, supporting detailed assessment of PDT effects. Excitation and emission wavelengths must be carefully selected based on the optical properties of both the PS and the probes (Tsubone et al. 2020).
PSs, in general, can localize in mitochondria, lysosomes, endoplasmic reticulum, Golgi apparatus and plasma membranes (Kurutos et al. 2022). The probes more commonly used in studies are: mitochondria (Mito-tracker™ Red, MitoTracker™ Deep Red FM, MitoTracker™ Orange or rhodamine 123) (Teiten et al. 2003; Li et al. 2018c; Martins et al. 2019), lysosomes (LysoTracker™ Green DND-26 or LysoTracker™ Red DND-99) (Kiesslich et al. 2013), endoplasmic reticulum (ER Tracker Blue/White™ or DiOC6) (Carneiro et al. 2018), and golgi apparatus (BODIPY® FL C5-ceramide, BPC) (Teiten et al. 2003).
For the analysis, selected cell lines should be cultured in an appropriate growth medium and incubated under controlled conditions (normoxia or hypoxia) until they reach the desired confluence. After achieving the proper confluence, the cells must be treated with the PS at the established concentration and incubated for the necessary period to allow for adequate internalization. Following this, specific fluorescent probes targeting organelles of interest should be incubated with the cells. Once the incubation is complete, confocal microscopy should be employed to analyze the PS and the organelle markers. This enables overlay assessment, providing insights into the subcellular localization of the PS. For optimal results in confocal microscopy, it is essential to select the appropriate laser wavelength for excitation based on the fluorescent probes used and an emission filter set compatible with these probes. The resulting images will reveal the spatial distribution of the PSs within the cell, facilitating the analysis of their subcellular localization (Woodburn et al. 1991; Moserova and Kralova 2012; Chota et al. 2024).
To interpret these data, the PS images should be overlaid (or colocalized) with images of the organelle-specific probes. Visual colocalization is initially identified by the appearance of structures whose color reflects the combined contribution of both species when the images of each probe are overlaid (or “merged”). For example, if the PS exhibits red fluorescence and the organelle-specific probe shows green fluorescence, colocalization will be observed as an orange or yellow color resulting from the combination of both fluorescence signals (Teiten et al. 2003). The obtained images can then be analyzed using ImageJ or ZenSoftware to quantify the colocalization between the PS and organelle-specific probes (Schneider et al. 2012; Pucelik et al. 2020). Although visual inspection of merged fluorescence images provides an initial indication of PS localization, qualitative assessment alone may be subjective. Therefore, colocalization should be supported by quantitative metrics that objectively describe the spatial overlap between the PS and organelle-specific markers.
For quantitative analysis of colocalization, ImageJ plugins such as Colocalization Test (Fay et al. 1997) and Colocalization Threshold (Costes et al. 2004) can be employed. The Pearson correlation coefficient (PCC) measures the correlation between the fluorescence intensities of the PS and the organelle marker on a pixel-by-pixel basis, providing an estimate of the overall degree of colocalization. In contrast, Manders’ colocalization coefficients (M1 and M2) quantify the fraction of one fluorophore overlapping with the other, thereby indicating the proportion of PS associated with a given organelle and vice versa. Because PCC and Manders’ coefficients provide complementary information, they are commonly reported together to objectively support the qualitative interpretation of merged confocal images (Dunn et al. 2011).
Most protocols to measure cellular uptake and subcellular localization of PS are often characterized under normoxic culture conditions, implicitly assuming that both processes remain unchanged at reduced oxygen levels. Although hypoxia does not directly modify the intrinsic physicochemical properties of PSs—such as charge, lipophilicity, or aggregation propensity—which remain the primary determinants of cellular entry and organelle targeting (MacDonald et al. 1999; Ezzeddine et al. 2013; Mallidi et al. 2016; Martinez De Pinillos Bayona et al. 2017; Chen et al. 2023), it influences overall PS accumulation and retention indirectly through hypoxia-induced cellular responses. Stabilization of hypoxia-inducible factor 1α (HIF-1α) drives extensive metabolic and transcriptional reprogramming (Infantino et al. 2021), including shifts lipid composition of membranes (Botto et al. 2008; Seo et al. 2022), modulates vesicular trafficking (Muñiz-García et al. 2022), and changes the expression of transporters and stress-response proteins (Parks et al. 2016). These changes can modulate passive diffusion and endocytic uptake pathways, thereby affecting the intracellular concentration and aggregation state of PSs without altering their chemical identity.
Hypoxia also affects subcellular localization primarily as a consequence of global cellular stress responses rather than through direct relocalization of the PS itself. Mitochondria, which are frequent targets for cationic and lipophilic PSs, undergo functional remodeling under hypoxia, including depolarization of the mitochondrial membrane potential (ΔΨm), structural fragmentation, and reduced oxidative activity (Fuhrmann and Brüne 2017). These changes can diminish mitochondrial accumulation of PSs that rely on ΔΨm-driven uptake, leading to attenuated Type II photochemistry and reduced phototoxicity (Lv et al. 2016b; An et al. 2022). In parallel, hypoxia-induced activation of autophagy alters vesicular trafficking and organelle turnover, potentially sequestering PSs into lysosomal (Lv et al. 2016b; An et al. 2022) or endoplasmic reticulum (Zhuang et al. 2020; Li et al. 2022c) compartments and modifying their effective photochemical microenvironment.
Methodologically, the techniques used to quantify uptake and localization in normoxia—such as fluorescence spectroscopy of cell lysates, flow cytometry, confocal microscopy, and colocalization with organelle-selective dyes—remain applicable under hypoxia, but their readouts must be interpreted with additional care. Fluorescence quenching by hypoxia-induced changes in protein or lipid environment, altered autofluorescence, and differences in probe behavior at low O2 can all bias apparent PS concentrations or colocalization coefficients (Peskova et al. 2021), particularly in 3D spheroids where light scattering and absorption are significant (Kucinska et al. 2021; Abolhassani et al. 2023). Whenever possible, uptake studies in hypoxic models should therefore combine quantitative approaches (for example, flow cytometry or lysate fluorescence calibrated to standard curves) with high-resolution imaging in 2D and 3D and should validate organelle assignment using multiple markers or genetically encoded reporters rather than a single commercial dye.
Finally, hypoxia-adapted or PDT-experienced cell populations may acquire stable changes in transporter expression and membrane organization that further reshape PS uptake profiles over time. In such systems, comparing PS accumulation and localization between parental and hypoxia-adapted lines, or between 2D monolayers and size-matched spheroids, can help distinguish the direct oxygen effects from longer-term adaptive remodeling. Integrating these uptake and localization measurements with oxygen-controlled ROS profiling and organelle-specific damage readouts is therefore crucial for mechanistically sound evaluation of PSs in hypoxia-relevant in vitro models.
Oxidative damage to biomolecules under normoxia and hypoxia
Many methods are available to evaluate oxidative stress in biomolecules—proteins, lipids, and DNA—by detecting both primary (e.g., hydroperoxides, radicals) and secondary (e.g., aldehydes, carbonyls) oxidation products (Juan et al. 2021). During PDT, the ROS generated are primarily responsible for inducing oxidative damage to these biomolecules (Tsubone et al. 2019a), ultimately leading to cell death (Tsubone et al. 2021). However, hypoxia fundamentally alters which ROS predominate and, consequently, which biomolecular targets are most affected depending on subcellular localization.
Under normoxic conditions, Type II photochemistry dominates for most clinically used PSs, generating 1O2 that exhibits relatively selective reactivity toward electron-rich amino acid residues—particularly histidine, methionine, tryptophan, tyrosine, and cysteine—with rate constants exceeding 107 dm3 mol−1 s−1 (Davies 2004). In contrast, under hypoxia, 1O2 formation is markedly suppressed due to limited ground-state oxygen availability, and Type I pathways involving direct electron or hydrogen transfer from the excited PS to substrates become comparatively more important (Davies 2004; Teng et al. 2025). The resulting radical species—O2−•, •OH, and HOO•—are less selective and react with a broader range of biomolecular targets, initiating chain reactions that can propagate oxidative damage beyond the immediate vicinity of the photoexcited PS (Juan et al. 2021).
This mechanistic shift has direct implications for the pattern and extent of oxidative injury observed in hypoxic PDT models. Whereas 1O2-mediated damage tends to produce well-defined oxidation products on specific amino acid residues (Davies 2004; Ronsein et al. 2006; Riske et al. 2009), radical-driven oxidation under hypoxia can lead to widespread protein carbonylation, thiol oxidation across multiple sites, and—critically—enhanced lipid peroxidation via chain-propagating HOO• radicals (Bacellar and Baptista 2019; Rezende et al. 2022). Recent studies on Type I PSs and PDT-induced ferroptosis-like cell death demonstrate that lipid hydroperoxide accumulation, glutathione depletion, and GPX4 inactivation can become dominant features of hypoxic PDT, even when conventional markers of 1O2-mediated damage are attenuated (Shui et al. 2021a; Qi et al. 2023).
Photoinduced oxidative damage to proteins by PSs
Among the biomolecules mentioned, proteins are considered one of the main photodynamic targets, both due to their abundance—making up about 70% of the dry weight of cells—and due to their high reaction rates with 1O2 and other ROS (Szokalska et al. 2009). Under normoxia, the amino acid residues of cysteine, histidine, methionine, tyrosine, and tryptophan are the primary ones involved in protein photooxidation (Benov 2015). Since tryptophan and tyrosine are aromatic amino acid residues that emit intrinsic fluorescence around 350 nm in proteins (Lakowicz 1994), photosensitized protein oxidation can be evaluated by monitoring the decrease in fluorescence intensity (e.g., in HSA) (Hirakawa et al. 2020, 2022). As example, Hirakawa and collaborators (2022) evaluate the protein photodamage to HSA induced by two different water-soluble cationic phthalocyanine PSs—2,3,6,7,10,11,14,15-octaquiso-[N-methyl-(4-methylpyridinium-3-yl-oxy)phthalocyaninato] chloride gallium(III) iodide (GaPyPc) and 2,3,6,7,10,11,14,15-octaquiso-[N-methyl-(4-methylpyridinium-3-yl-oxy)phthalocyaninato] chloride zinc(II) iodide (ZnPyPc) in the presence of sodium azide (NaN3), which is used to quench the 1O2 generated during PDT application (Hirakawa et al. 2022).
Under hypoxia, however, the relative contribution of radical-mediated oxidation increases. Hydroxyl radicals (•OH) oxidize tryptophan to hydroxytryptophan, which is subsequently cleaved by residual oxygen into N-formylkynurenine—a pathway that can proceed even at low O2 levels and does not require 1O2. Similarly, cysteine thiol groups, which are highly susceptible to both 1O2 and radical attack, may undergo oxidation via different mechanisms depending on oxygen availability: 1O2 favors sulfenic acid formation, whereas •OH and O2−• can generate thiyl radicals that propagate disulfide cross-linking and mixed oxidation products (Davies 2004; Shui et al. 2021a).
Therefore, the photodamage to protein is induced by both 1O2 and electron transfer, and the oxidation of protein containing tryptophan via electron transfer can generate N-formylkynurenine (Fig. 5A). More specifically, hydroxyl radicals oxidize tryptophan to hydroxytryptophan, which is subsequently cleaved by oxygen into N-formylkynurenine, as shown in Fig. 5B (Lowry et al. 1951).
Fig. 5.

Schematic representation of the chemical reactions involved in the assessment of oxidative damage to biomolecules. Protein oxidation: (A) oxidation of tryptophan by hydroxyl radicals (•OH) and singlet oxygen (1O2), illustrating distinct oxidative pathways; (B) reactions underlying the DTNB assay for quantification of protein thiol oxidation through detection of free sulfhydryl groups; and (C) reactions involved in the DNPH assay for detection of protein carbonyl groups as markers of oxidative protein damage. Lipid peroxidation: (D) reactions underlying the FOX assay for detection of lipid hydroperoxides; (E) reactions involved in the TBARS assay for evaluation of lipid peroxidation through malondialdehyde (MDA) detection; and (F) oxidation reactions of BODIPY-C11 used to assess lipid peroxidation in vitro
Another approach to evaluate protein oxidative damage induced by photosensitization involves the detection or quantification of amino acid residues containing thiol groups, such as cysteine and methionine. A decrease in –SH group concentration following PDT treatment reflects the extent of protein oxidation (Saczko et al. 2004; Ion et al. 2012). The most widely used method for thiol quantification is the 5,5′-dithiobis(2-nitrobenzoic acid) assay, commonly known as Ellman’s reagent (referred to herein as DTNB assay) (Moan and Vistnes 1986; Rice-Evans et al. 1991; Daicoviciu et al. 2011), which will be discussed in detail in the next subsection (Sect. “Protein thiol groups or DTNB assay”). Protein photooxidation can also be assessed by measuring carbonyl contents. The 2,4-dinitrophenylhydrazine (DNPH) assay is a widely used method for measuring protein oxidation by detecting carbonyl groups in proteins (Tseng et al. 2009) and will also be described below (Sect. “DNPH assay”).
Protein thiol groups or DTNB assay
Thiol groups play a crucial role in the function of metabolic enzymes and cellular signaling processes that mediate responses to environmental factors. Among these, cysteine thiol groups are particularly important due to their reversible redox state, which directly influences structural, catalytic, and regulatory protein functions. However, these groups are highly susceptible to oxidation, which can lead to alterations in protein activity and cellular homeostasis (Moan and Vistnes 1986; Rice-Evans et al. 1991; Daicoviciu et al. 2011).
One of the most commonly used spectrophotometric methods for detecting thiol groups in both protein and non-protein molecules is the Ellman’s reagent assay (Rice-Evans et al. 1991). This assay utilizes 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB), which undergoes a disulfide exchange reaction with thiol groups, producing the yellow-colored 5-thio-2-nitrobenzoate anion (TNB) (Fig. 5C). TNB exhibits a high molar absorption coefficient (ε = 13,600 M−1 cm−1 at 412 nm), enabling its concentration to be determined from absorbance measurements according to the Beer-Lambert law. Because one mole of TNB is produced for each reacting thiol group, the absorbance directly reflects the thiol concentration in the sample (Rice-Evans et al. 1991). The reaction is highly specific for thiol groups and is most effective at pH 8.0–9.0, as lower pH values reduce the chromophore’s absorbance (Rice-Evans et al. 1991). Additionally, the rate of TNB production can be used to assess the accessibility of thiol groups within proteins, where lower reaction rates indicate sterically hindered or inaccessible thiol groups (Rice-Evans et al. 1991).
In PDT studies, protein thiol oxidation has been used as a marker of oxidative damage (Moan and Vistnes 1986; Du et al. 2003; Daicoviciu et al. 2011). Ellman’s method is applied to evaluate PDT-induced modifications in protein thiol content in cells (Moan and Vistnes 1986; Daicoviciu et al. 2011), often in combination with Western Blot analysis. In these assays, cells are treated with a PS and light, followed by incubation with DTNB to detect changes in free thiol availability. A modified protocol by Schothorst (1982) involves exposing cells to DTNB for 20 min at 20 °C, a reduction from the standard 60-min incubation time. This adjustment allows for the detection of early thiol oxidation events before secondary reactions and cellular repair mechanisms can significantly alter results. Control experiments with non-irradiated samples showed comparable results for 20-min and 60-min incubations, supporting the validity of the shorter incubation time (Moan and Vistnes 1986; Daicoviciu et al. 2011).
In addition to whole-cell assays, thiol oxidation can be assessed in lysed cells (Moan and Vistnes 1986). Sonicated cell samples undergo centrifugation to remove debris before the absorbance of the supernatant is measured at 412 nm. Scanning the full absorption spectrum enables correction for potential background interference or scattering from detached cells. Moreover, assays performed at 4 °C yielded similar relative thiol oxidation values but were less effective due to incomplete color development, reinforcing the preference for a 20 °C incubation temperature in most experiments (Moan and Vistnes 1986).
The Ellman’s reagent (DTNB) assay remains applicable for detecting protein thiol oxidation under hypoxia, but interpretation requires caution (Salyha and Oliynyk 2023b). Because radical-mediated thiol oxidation may produce a heterogeneous mixture of products—including sulfenic, sulfinic, and sulfonic acids, as well as inter- and intramolecular disulfides—the net decrease in free –SH groups detected by DTNB may not fully reflect the complexity of oxidative modifications. Additionally, hypoxia-adapted cells often exhibit elevated glutathione (GSH) pools and enhanced thioredoxin/glutaredoxin activity, which can partially buffer or reverse early thiol oxidation events before assay readout (Zhang et al. 2022; Wolny et al. 2024). Whenever possible, DTNB-based thiol quantification should be complemented by mass spectrometry or specific antibody-based detection of oxidized cysteine products to capture the full spectrum of modifications induced by hypoxic PDT.
DNPH assay
Protein carbonylation is an early marker of oxidative stress, once it occurs before major structural changes (Tseng et al. 2009). Unlike other oxidative modifications, such as thiol oxidation, protein carbonylation is irreversible, making it a reliable indicator of oxidative stress (Tseng et al. 2009). Therefore, protein carbonylation measured by the DNPH assay provides a robust and irreversible marker of oxidative protein damage that is equally relevant under normoxia and hypoxia. Importantly, carbonyl formation can arise from both 1O2-mediated and radical-mediated pathways: direct oxidation of lysine, arginine, proline, and threonine side chains (Juan et al. 2021), as well as secondary modifications by reactive aldehydes (e.g., malondialdehyde, 4-hydroxynonenal) generated during lipid peroxidation (Rezende et al. 2022).
As initially reported by Levine (Levine et al. 1994), the determination of carbonyl groups is performed using 2,4-dinitrophenylhydrazine (DNPH), which reacts with carbonyl groups, forming 2,4-dinitrophenylhydrazone (DNP)—Fig. 5D. Carbonyl content is measured around 360 nm, using an extinction coefficient of 22,000 M−1 cm−1 (Levine et al. 1994), allowing quantifying protein oxidation.
This method is applicable to both isolated proteins and cellular proteins. In cellular analysis, proteins are extracted by lysing cells through mild sonication, followed by centrifugation to remove insoluble debris. The resulting supernatant is divided into two aliquots: one reacts with DNPH to detect carbonyl groups, while the other serves as a control. Samples are then precipitated with trichloroacetic acid (TCA), washed with ethanol/ethyl acetate, and dissolved in guanidinium hydrochloride for spectrophotometric analysis.
In PDT studies, protein carbonyl analysis has been applied to bacteria (Tseng et al. 2009). Following photodynamic inactivation (PDI), bacterial cells are disrupted by sonication to obtain crude extracts for analysis. The assay follows the DNPH-labeling procedure, where crude extracts are incubated DNPH in HCl for 1 h at 37 °C. Protein blanks are prepared using HCl alone. After incubation, proteins are precipitated with 10% TCA, washed three times with ethanol/ethyl acetate (1:1) to remove excess DNPH, and dissolved in guanidine hydrochloride at 37 °C for 30 min. Carbonyl content is then quantified using absorbance technique (Tseng et al. 2009).
Similarly, in plasma protein carbonyl analysis, DNPH-based detection can also be used after PDT (Du et al. 2003). Plasma samples are incubated with DNPH in HCl for 1 h at room temperature in the dark. After protein precipitation with TCA and centrifugation, the washed protein pellet is dissolved in 6 M guanidine hydrochloride, and absorbance is measured at 360 nm. Protein concentration in each sample is estimated by absorbance readings at 280 nm, using a bovine serum albumin standard curve. Protein carbonyls values are then expressed as nmol carbonyl per mg of protein (Du et al. 2003).
In conclusion, protein carbonylation measurement is a tool for assessing oxidative damage across different biological contexts, including cellular proteins, plasma, and bacterial systems in PDT studies. Its irreversible nature makes it a robust biomarker for oxidative stress, facilitating research on oxidative damage and therapeutic interventions.
Photoinduced oxidative damage to lipids by PSs
Lipid peroxidation is one of the most frequently used indirect indicators of free radical activity in cells, involving the degradation of polyunsaturated fatty acids (PUFAs) in membrane phospholipids, lipoproteins, and other lipid structures (Daicoviciu et al. 2011). Among various cellular targets, lipid peroxidation is particularly harmful because its byproducts promote the rapid propagation of free radicals, amplifying oxidative damage (Saczko et al. 2004). This complex process generates a variety of stable yet toxic oxidized biomolecules, which can severely compromise membrane integrity (Daicoviciu et al. 2011). Under normoxia and Type II PS, 1O2 reacts directly with PUFA double bonds via a concerted “ene” reaction to form lipid hydroperoxides (LOOHs) without generating free radical intermediates—a non-chain-propagating mechanism that produces relatively defined hydroperoxide isomers (Fig. 6A) (Bacellar and Baptista 2019; Rezende et al. 2022).
Fig. 6.

Schematic representation of (A) Type I and Type II photoinduced lipid oxidation pathways occurring in biological membranes. Abbreviations: LH, unsaturated lipid; LOOH, lipid hydroperoxide; OLOOH, epoxyallylic hydroperoxide; OLOH, epoxyallylic alcohol; LO, lipid ketone or aldehyde; LOH, lipid alcohol; L•, lipid radical; LOO•, peroxyl radical; OLOO•, epoxyallylic peroxyl radical; OLO•, epoxyallylic oxyl radical; and LO•, alkoxyl radical. PSs in the ground, excited triplet, and excited singlet states are represented by PS 1S0, PS 3T1*, and PS 1S1*, respectively. B General stages involved in the lipid peroxidation radical chain reaction, including initiation, propagation, and termination processes
Under hypoxia, Type I photochemistry becomes comparatively more important, and radical-initiated lipid peroxidation can dominate. In this pathway, abstraction of a bis-allylic hydrogen by •OH or HOO• radicals generates a carbon-centered lipid radical (L•), which reacts with residual O2 to form a peroxyl radical (LOO•) that propagates the chain by abstracting hydrogen from adjacent PUFAs (Fig. 6B) (Bacellar and Baptista 2019; Rezende et al. 2022). This chain reaction can amplify oxidative damage beyond the immediate site of PS localization and generates a broader spectrum of primary (hydroperoxides) and secondary (aldehydes, epoxides) products compared with 1O2-mediated peroxidation (Fig. 6). Recent work linking PDT to ferroptosis—an iron-dependent form of cell death driven by uncontrolled lipid peroxidation—highlights the biological relevance of this pathway in hypoxic settings. Type I PSs that generate O2−• and •OH have been shown to deplete GSH, inactivate GPX4, and trigger ferroptosis-like death in hypoxic tumor cells, providing a mechanistic rationale for sustained PDT efficacy under low oxygen conditions (Shui et al. 2021a; Qi et al. 2023). Notably, even at low O2 concentrations (< 2%), sufficient molecular oxygen may remain to sustain LOO• radical formation and chain propagation, meaning that Type I–initiated lipid peroxidation can proceed under conditions where 1O2 generation is severely compromised (Shui et al. 2021a). Some reactive intermediates and end-products formed during photosensitized lipid oxidation exhibit inherent absorbance and/or fluorescence, allowing their detection and quantification through colorimetric or fluorometric methods, as outlined below.
Widely used assays such as FOX (Ferrous Oxidation−Xylenol orange), thiobarbituric acid reactive substances (TBARS) or iodometric methods are primarily sensitive to secondary or stable lipid peroxidation products and may underestimate or misrepresent lipid damage generated via Type I mechanisms (Rubio and Cerón 2021; Murphy et al. 2022). Fluorescent probes such as BODIPY-C11 provide improved spatial resolution but remain sensitive to probe localization (Greene et al. 2018). Accordingly, interpretation of lipid oxidation data in hypoxic PDT models requires careful consideration of both oxygen tension and photochemical pathway. Combining multiple readouts—such as live-cell imaging, membrane integrity assays, and time-resolved measurements—can help distinguish between reduced oxidation efficiency and altered oxidative mechanisms.
FOX assay
The FOX (Ferrous Oxidation−Xylenol orange) assay relies on the oxidation of ferrous ions (Fe2+) to ferric ions (Fe3⁺) by hydroperoxides under acidic conditions at room temperature (Fig. 5F) (Niki 2009, 2014). The dye Xylenol Orange binds to Fe3⁺, forming a chromophore complex with strong absorbance in the 540–600 nm range, which can be detected spectrophotometrically. Since hydroperoxide measurement is linked to free radicals and other reactive oxygen species, it serves as a key indicator of oxidative stress in biological samples (Halliwell and Chirico 1993; Bou et al. 2008). Therefore, its quantification provides an early and precise assessment of oxidative status (Bou et al. 2008). Additionally, this method is applicable to the determination of lipid hydroperoxides in liposomes, plasma, and lipoproteins (Jiang et al. 1991, 1992; Nourooz-Zadeh et al. 1994; Wolff 1994).
The FOX assay has two main versions. The first, FOX-1, uses an aqueous reagent containing sorbitol, which enhances the detection of water-soluble hydroperoxides such as hydrogen peroxide, butyl, and cumyl hydroperoxide (Jiang et al. 1991; Nourooz-Zadeh et al. 1994; Bleau et al. 1998; Banerjee et al. 2004; Cheeseman 2006). However, it is unsuitable for measuring low levels of lipid-soluble hydroperoxides in the presence of non-peroxidized lipids. To address this, FOX-2 was developed, utilizing a methanolic solution (90%, v/v) with butylated hydroxytoluene (BHT) as an antioxidant to prevent artificial oxidation. This version, widely applied to liposomes, cells and tissues, offers greater versatility (Hermes-Lima et al. 1995; Arab and Steghens 2004; Ayuyan and Cohen 2006; Fukuzawa et al. 2006; Bou et al. 2008; Angeli et al. 2011).
When quantifying hydroperoxides using the FOX method in PDT studies, special attention should be given to pigments that strongly absorb within the same spectral range as the Fe–XO complex. For example, anthocyanins at low pH (490–550 nm) and chlorophyll derivatives (with absorbance peaks at 400–455 and 645–665 nm) exhibit spectra spanning the orange to blue regions (Wrolstad et al. 2005; Bou et al. 2008). While these compounds do not strongly overlap with the Fe–XO absorbance range, high concentrations can still interfere with the assay, potentially affecting result accuracy.
The FOX-2 assay has been applied in PDT studies using PSs that induce photooxidation via a Type I mechanism (Vignoni et al. 2018; Sosa et al. 2023). This process involves the direct interaction of an excited-state PS, typically in its triplet state, with a target molecule or substrate. Through electron transfer or hydrogen abstraction, radicals are generated, which then participate in various competitive reactions, ultimately forming stable oxidized photoproducts (Sosa et al. 2023).
This mechanism follows a contact-dependent pathway, requiring direct interaction between the excited PS and the substrate, leading to significant membrane damage (Vignoni et al. 2018; Sosa et al. 2023). Increasing the lipophilicity of these PSs enhances their affinity for biomembranes, thereby promoting lipid hydroperoxide formation and further compromising membrane integrity (Sosa et al. 2023).
Other studies have applied the FOX-2 assay in PDT using porphyrins, a class of PSs that primarily undergo Type II mechanisms (Melo et al. 2013). Phospholipid hydroperoxides were identified as the predominant photoproducts generated during photosensitization with cationic porphyrins. Moreover, the formation of HOO• derivatives correlated with increased photosensitization efficiency in bacterial inactivation (Melo et al. 2013). Since phospholipids are key targets in photosensitization, their oxidation can serve as a valuable marker for assessing and monitoring photodamage in biological systems.
In PDT studies, the FOX-2 assay is typically performed in lipid vesicles (LUVs) containing specific concentrations of the PS, which are then exposed to radiation. The procedure generally follows a modified version of the method described by Wolff (Wolff 1994). The FOX-2 reagent is prepared by dissolving ferrous ammonium sulfate and sulfuric acid in water, followed by the addition of butylated hydroxytoluene (BHT), xylenol orange, and methanol to achieve the appropriate reaction conditions (Melo et al. 2013; Vignoni et al. 2018; Sosa et al. 2023).
The FOX-2 assay can be a valuable tool for assessing oxidative damage in PDT studies, particularly in evaluating lipid hydroperoxide formation. However, because 1O2- and radical-mediated peroxidation generate overlapping but non-identical hydroperoxide profiles, the assay cannot distinguish between these mechanisms without additional controls (Salyha and Oliynyk 2023b). Furthermore, under prolonged hypoxia, enhanced antioxidant defenses (elevated GSH, GPX4 activity) may reduce steady-state LOOH levels even when initial peroxidation rates are high, potentially leading to underestimation of oxidative damage if sampling is delayed. Time-course measurements and inclusion of GPX4 inhibitors (e.g., RSL3) as positive controls can help interpret FOX assay data in hypoxic models (Shui et al. 2021a; Hong et al. 2022).
TBARS assay
Lipid peroxidation is one of the most significant forms of oxidative damage, involving the degradation of phospholipids, cholesterol, and other unsaturated lipids. This process leads to the formation and breakdown of lipid hydroperoxides (LOOHs), generating stable and toxic oxidized compounds (Girotti 1990; Du et al. 2003) that compromise membrane integrity by reducing fluidity, increasing permeability, and decreasing membrane potential. Prolonged oxidative stress ultimately results in the destruction of cellular and organelle membranes, leading to cell death (Daicoviciu et al. 2011; Ion et al. 2012). Among the byproducts of lipid peroxidation, malondialdehyde (MDA) stands out as a toxic compound commonly used as a biomarker of oxidative stress (Saczko et al. 2004).
One widely used method for assessing lipid peroxidation is the thiobarbituric acid reactive substances (TBARS) assay, which detects MDA and other secondary peroxidation products. This test measures both the MDA naturally present in the sample and that generated from lipid hydroperoxides under assay conditions. Since MDA is a low-molecular-weight end product of lipid peroxidation, the TBARS assay provides an initial indication of oxidative damage and is particularly sensitive for detecting lipid peroxidation products such as 2,4-decadienals and, to a lesser extent, saturated aldehydes formed during the Type I mechanism (Ion et al. 2012). The reaction between thiobarbituric acid (TBA) and peroxidation products like MDA and 4-hydroxynonenal (HNE) results in TBARS, forming a pink-red trimethine adduct that can be quantified either by absorbance at 532 nm or fluorescence at 555 nm (Fig. 5G) (Girotti 1990; Babizhayev 1996). This reaction is conducted under high temperatures and acidic conditions to enhance sensitivity by promoting MDA adduct formation and accelerating fatty peroxide decomposition.
In a typical experiment, liposomes or cells are exposed to the PS at the desired concentration. After an incubation period of PS, the samples are submitted to irradiation at the appropriate wavelength for the PS used. At predetermined time points post-irradiation, the liposomes or cells are lysed using a lysis buffer containing an antioxidant such as BHT under controlled conditions. The samples are then centrifuged to remove insoluble material, and the supernatant is collected for analysis (Bacellar et al. 2014).
For MDA quantification, a calibration curve is prepared from a standard solution, diluted to obtain appropriate concentrations. To form the MDA-TBA adduct, a TBA solution is added to each standard or sample, followed by incubation at 95 °C for a fixed period. After cooling to room temperature, the absorbance or fluorescence of the MDA-TBA adduct is measured (Bacellar et al. 2014).
Bacellar et al. (2014) conducted a TBARS assay in a PDT study using liposomes prepared from soy lecithin films. The lipid films were hydrated to generate liposome suspensions, which were then treated with or without a PS and exposed to irradiation. Lipid peroxidation was evaluated by collecting samples at various time points postirradiation and analyzing the TBARS reaction products. Absorbance was measured at 532 nm, with background correction at 800 nm, and results were normalized to express relative levels of lipid oxidation across samples (Bacellar et al. 2014).
In PDT in vitro studies, MDA quantification can also be performed at different time points post-treatment to assess oxidative damage dynamics (Rapozzi et al. 2009; Karagöz et al. 2020; Shui et al. 2021b; de Melo Gomes et al. 2023; Kojima et al. 2024). Variations included the use of different TBA concentrations, incubation times, and wavelengths for detection, allowing adaptation to specific experimental conditions (Rapozzi et al. 2009; Karagöz et al. 2020; Shui et al. 2021b; de Melo Gomes et al. 2023; Kojima et al. 2024; Zhou et al. 2024). The data were typically expressed as MDA concentration relative to total protein or as absolute values in nmol/µL of TBARS. Despite methodological differences, all approaches aimed to assess MDA levels as an indicator of oxidative stress, with necessary controls and adjustments implemented to improve assay sensitivity and specificity. Under hypoxia, elevated TBARS levels relative to normoxic controls can thus serve as an indirect indicator of Type I pathway engagement, although the assay’s limited specificity means that other sources of reactive aldehydes must be controlled.
BODIPY-C11 (lipid peroxidation sensor)
Boron dipyrromethene (BODIPY) derivatives function as oxidation-sensitive probes for lipid peroxidation, undergoing a fluorescence shift from red to green upon oxidation (Cummins et al. 2013). These probes offer significant advantages, including intense and narrow fluorescence emission, as well as stability across a wide pH range and in various solvents (Cummins et al. 2013). Intracellular lipid peroxidation can be assessed using C11-BODIPY581/591 (Fig. 5H), with various methodological adaptations available to suit different cell types and implemented to ensure reliable detection across different experimental conditions.
In PDT studies, the oxidation-sensitive probe is typically added post-illumination to prevent direct photobleaching or photochemical reactions (Sakharov et al. 2005; Shui et al. 2021b; Ma et al. 2024; Xiong et al. 2024; Zhou et al. 2024). In fluorescence microplate assays (Rapozzi et al. 2009; dos Santos et al. 2020), cells are treated with C11-BODIPY and incubated under similar conditions, typically at a final concentration of 1 to 10 µM for 30 min at 37 °C after PDT treatment. Lipid peroxidation is quantified either by measuring the percentage decay of red fluorescence (non-oxidized form) in PDT samples compared to untreated controls (Rapozzi et al. 2009) or by assessing oxidized C11-BODIPY fluorescence at 488/510 nm (excitation/emission)—Fig. 5 (dos Santos et al. 2020). Fluorescence kinetics are monitored using microplate readers at 37 °C over a period of 1 h, with readings taken every 5 min to dynamically track oxidative changes. H2O2 treatment serves as a positive control for oxidative stress. Flow cytometry assays follow a similar protocol; C11-BODIPY is added post-PDT and probe incubation times can range from 30 min to 6 h at 37 °C. After C11-BODIPY incubation, cells are washed with PBS and re-suspended for flow cytometry analysis (Lange et al. 2019; Shui et al. 2021b; Zhang et al. 2023; Zhou et al. 2024).
Most C11-BODIPY-based lipid peroxidation assays employ fluorescence confocal microscopy. The probe is typically prepared in fetal calf serum at 20 µM and applied to cell cultures at a final concentration of 1 to 10 µM. Incubation lasts 10 to 30 min at 37 °C, either after PDT or, in some protocols, before irradiation (Zheng et al. 2022). Oxidized C11-BODIPY emits green fluorescence (excitation at 488 nm, emission at 500–530 nm), while the non-oxidized form emits red fluorescence (excitation at 543–561 nm, emission at 580–630 nm) (Sakharov et al. 2005; Rapozzi et al. 2009; Shui et al. 2021b; Ma et al. 2024; Xiong et al. 2024).
The BODIPY-C11 probe provides the advantages of real-time and spatially resolved detection of lipid peroxidation in live cells. This probe is particularly valuable for hypoxic PDT studies because it can capture the dynamics of peroxidation during and immediately after irradiation, before cellular repair mechanisms or secondary reactions obscure the initial oxidative burst (Shui et al. 2021a). However, because BODIPY-C11 is oxidized by both 1O2 and LOO• radicals, it does not discriminate between Type I and Type II mechanisms without additional controls such as 1O2 quenchers (e.g., sodium azide) or radical scavengers (e.g., vitamin E analogs).
Photoinduced oxidative damage to DNA by PSs
In PDT, PS-generated ROS can oxidize multiple classes of biomolecules, including lipids, proteins, and nucleic acids (Chen et al. 2021b; An et al. 2022; Moloudi et al. 2025). However, because ROS reactivity depends on the species generated, their intracellular localization, and the surrounding biochemical environment, the extent of DNA damage cannot be reliably inferred from total ROS production alone or from commonly reported endpoints such as mitochondrial depolarization, membrane disruption, or loss of cell viability (El-Hussein et al. 2012; Digby et al. 2019; Teng et al. 2025; El-Sadek et al. 2025). Lipids and proteins are generally the primary and most accessible targets of photooxidative reactions, whereas oxidative damage to nuclear DNA represents a more restricted subset of events (El-Hussein et al. 2012; Li et al. 2022b). When DNA is reached by ROS, PDT can induce oxidative base lesions, single- and double-strand breaks, DNA cross-linking, and other forms of genomic damage, which can be detected using comet assays and lesion-specific enzymes such as FPG and Endonuclease III El-Hussein et al. 2012). Indeed, DNA photodamage depends not only on ROS abundance and identity but also on ROS proximity to the nucleus, chromatin accessibility, nucleotide reactivity, and the efficiency of DNA repair mechanisms (El-Hussein et al. 2012; Digby et al. 2019; Moloudi et al. 2025).
Among the mechanisms involved in DNA oxidation during PDT, Type I photochemistry is considered a major contributor. Electron-transfer reactions generate radical species that can ultimately produce highly reactive oxidants, including •OH, which are recognized as major mediators of oxidative damage to DNA bases and the sugar-phosphate backbone under oxidative stress conditions (El-Hussein et al. 2012; Lv et al. 2018; Teng et al. 2025; Moloudi et al. 2025). Consequently, although several ROS may contribute to DNA oxidation, •OH is considered one of the most relevant candidates for inducing strand breaks and oxidative base lesions during Type I PDT (Lv et al. 2018; Li et al. 2020; An et al. 2022; Teng et al. 2025).
The extent of DNA photodamage is strongly influenced by the subcellular localization and DNA-binding properties of the PS. For example, the DNA-binding PS Br-DAPI induced extensive double-strand breaks only when physically associated with DNA (Digby et al. 2019). Mitochondrial DNA may be particularly susceptible because mitochondria are major sites of ROS generation and mtDNA lacks the protection provided by histones, while mitochondrial damage can additionally promote membrane permeabilization, cytochrome c release, caspase activation, and secondary nuclear DNA fragmentation (Moloudi et al. 2025).
This issue is especially relevant under hypoxia, where the suppression of 1O2 production can limit conventional Type II PDT, but Type I PSs may preserve phototoxicity through electron-transfer reactions and the generation of O2−• and •OH (Teng et al. 2025; El-Sadek et al. 2025). For instance, the cyclometalated Ru(II) PS Ru2 maintained PDT efficacy under both normoxic and hypoxic conditions, with •OH formation identified as an important contributor to cell death (Lv et al. 2018). Similarly, oxygen-economizing systems such as SORgenTAM can generate O2−•, promote its conversion into •OH through intracellular catalytic cascades, and reduce the oxygen dependence of PDT (Li et al. 2020). Nevertheless, persistent ROS production under hypoxia does not necessarily demonstrate that DNA is the primary intracellular target of photodamage (El-Hussein et al. 2012; Digby et al. 2019; An et al. 2022). Therefore, the evaluation of photoinduced DNA oxidation should rely on direct evidence, including strand-break assays (David et al. 2005; El-Hussein et al. 2012), DNA damage signaling markers (Moloudi et al. 2025), oxidative base modification assays (El-Hussein et al. 2012), and experimental confirmation that the PS localizes to, or exerts its photochemical activity near, nuclear DNA (Digby et al. 2019). Under hypoxic conditions, such analyses are particularly important because DNA damage may arise not only from residual 1O2 production but also from Type I radical chemistry, subcellular localization of the PS, mitochondrial dysfunction, and activation of the cellular DNA damage response.
Conclusions and future perspectives
PDT remains a powerful and versatile therapeutic modality; however, its biological efficacy is fundamentally constrained by oxygen availability and by the experimental paradigms used to evaluate PSs. As critically discussed in this review, the widespread reliance on normoxic in vitro models has led to a systematic overestimation of PDT performance and an implicit bias toward Type II, 1O2–dominated photochemistry. Under hypoxic conditions, which more closely reflect the tumor microenvironment even at the cellular scale, PDT efficacy is reshaped by coordinated changes in PS uptake, subcellular localization, ROS profiles, and biomolecular damage patterns.
A central conclusion emerging from this analysis is that hypoxia-driven resistance to PDT does not arise solely from oxygen depletion, but from broader cellular reprogramming involving metabolism, membrane organization, and stress-response pathways. These adaptations alter the intracellular location of PSs and may impact the dominant photochemical mechanisms from oxygen-dependent Type II pathways toward radical-based Type I processes. This mechanistic shift has important implications for how oxidative damage to biomolecules is generated and detected. Under hypoxia, lipid oxidation may proceed through radical-mediated, while protein oxidation emerges as a persistent and functionally relevant target of Type I photochemistry. Many commonly used analytical tools—including ROS probes and assays for lipid and protein oxidation—were developed and validated under normoxic conditions and therefore require careful adaptation and interpretation to capture hypoxia-dominated PDT mechanisms and to accurately assess PDT outcomes under low oxygen conditions.
Looking forward, advancing PDT toward hypoxia-relevant applications requires both conceptual and methodological realignment. From a PS design perspective, greater emphasis should be placed on molecules capable of sustaining photodynamic activity under low oxygen concentration, including those favoring Type I pathways, targeting organelles outside the most oxygen-depleted compartments, or combining hybrid Type I/Type II reactivity. Importantly, intrinsic photophysical properties must also be evaluated alongside PS behavior in hypoxia-adapted cellular environments, rather than inferred from solution-based measurements alone.
Equally critical is the refinement of in vitro experimental models. Controlled oxygen culture systems, hypoxia-adapted cell lines, and three-dimensional (3D) spheroids should be incorporated into PDT evaluation pipelines to capture oxygen gradients, metabolic heterogeneity, and diffusion constraints absent in conventional monolayers. Comparative analyses across normoxic and hypoxic conditions, as well as between 2 and 3D architectures, are essential to disentangle intrinsic PS properties from context-dependent cellular responses.
In summary, overcoming hypoxia-driven resistance in PDT is not simply a matter of improving oxygen delivery or maximizing Type I-PS, but of embracing the complexity of cellular adaptation and photochemical diversity under low oxygen conditions. By redefining how PDT is modeled, measured, and interpreted in vitro, the field can move toward more mechanistically sound and biologically relevant strategies for developing next-generation PSs capable of operating effectively in hypoxic environments.
Acknowledgements
The authors gratefully acknowledge the financial support provided by FAPEMIG (APQ-02393-24 and APQ-00864-25), CNPq (407282/2023-8). This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. Tayana M. Tsubone acknowledges CNPq for a Productivity Fellowship (308094/2025-5) and the grant awarded through the L’Oréal Brazil–UNESCO–ABC For Women in Science Program (2023 edition, Chemistry). IVC (CAPES Grant No. 88887.131921/2025-00) acknowledges CAPES for a Ph.D. scholarship. MMC acknowledges CAPES for a graduate scholarship. NCB (FAPESP Grant No. 2025/06781-1), MVPV (FAPESP Grant No. 2025/02222-8) acknowledge FAPESP for graduate scholarships. Renata G. de Lima acknowledges the grant from FAPESP (Grant No. 2025/09760-5).
Author contribution
I.V.C. led the writing process, integrated and harmonized the different sections of the review, prepared all figures, conceived the review and contributed substantially to the sections on reactive oxygen species detection and oxidative damage to biomolecules under normoxic and hypoxic conditions. T.M.T. defined its scope and structure, coordinated the development of the manuscript, supervised the writing process, conceived the review and critically revised the manuscript. R.G.L. contributed to the discussion of photochemical mechanisms in photodynamic therapy, particularly Type I and Type II pathways, and to the section on three-dimensional spheroid models. J.C.B. contributed to the section on singlet oxygen detection in cells. L.B.N. contributed to the section on cellular uptake and subcellular localization of photosensitizers. P.A.M. contributed to the sections discussing lipid peroxidation assays, including TBARS and BODIPY-C11 methodologies. M.D.S.F. contributed to the section on reactive oxygen species detection. N.C.B., M.M.C., and M.V.P.V. contributed to the section on photoinduced oxidative damage to proteins. All authors contributed to literature analysis and discussion within their respective areas of expertise, critically revised the manuscript, approved the final version, and agreed to be accountable for all aspects of the work.
Funding
The Article Processing Charge (APC) for the publication of this research was funded by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) (ROR identifier: 00x0ma614). This work was supported by the FAPEMIG (APQ-02393-24 and APQ-00864-25), FAPESP (2025/09760-5, 2025/06781-1 and 2025/02222-8), and CNPq (407282/2023-8 and 308094/2025-5). This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001.
Data availability
No datasets were generated or analyzed during the current study.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- Abolhassani H, Zaer M, Shojaosadati SA, Hashemi-Najafabadi S (2023) Rapid generation of homogenous tumor spheroid microtissues in a scaffold-free platform for high-throughput screening of a novel combination nanomedicine. PLoS ONE 18:e0282064. 10.1371/journal.pone.0282064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aguilera K, Brekken R (2014) Hypoxia Studies with Pimonidazole in vivo. Bio Protoc 4:. 10.21769/BioProtoc.1254 [DOI] [PMC free article] [PubMed]
- Alva R, Wiebe JE, Stuart JA (2024) Revisiting reactive oxygen species production in hypoxia. Pflugers Arch 476:1423–1444. 10.1007/s00424-024-02986-1 [DOI] [PubMed] [Google Scholar]
- Angeli JPF, Garcia CCM, Sena F, et al (2011) Lipid hydroperoxide-induced and hemoglobin-enhanced oxidative damage to colon cancer cells. Free Radic Biol Med 51:. 10.1016/j.freeradbiomed.2011.04.015 [DOI] [PubMed]
- An J, Tang S, Hong G, et al (2022) An unexpected strategy to alleviate hypoxia limitation of photodynamic therapy by biotinylation of photosensitizers. Nat Commun 13:. 10.1038/s41467-022-29862-9 [DOI] [PMC free article] [PubMed]
- Arab K, Steghens JP (2004) Plasma lipid hydroperoxides measurement by an automated xylenol orange method. Anal Biochem 325. 10.1016/j.ab.2003.10.022 [DOI] [PubMed]
- Ayuyan AG, Cohen FS (2006) Lipid peroxides promote large rafts: Effects of excitation of probes in fluorescence microscopy and electrochemical reactions during vesicle formation. Biophys J 91:. 10.1529/biophysj.106.087387 [DOI] [PMC free article] [PubMed]
- Babizhayev MA (1996) Failure to withstand oxidative stress induced by phospholipid hydroperoxides as a possible cause of the lens opacities in systemic diseases and ageing. Biochim Biophys Acta Mol Basis Dis 1315. 10.1016/0925-4439(95)00091-7 [DOI] [PubMed]
- Bacellar IOL, Baptista MS (2019) Mechanisms of Photosensitized Lipid Oxidation and Membrane Permeabilization. ACS Omega 4. 10.1021/acsomega.9b03244 [DOI] [PMC free article] [PubMed]
- Bacellar IOL, Pavani C, Sales EM, et al (2014) Membrane damage efficiency of phenothiazinium photosensitizers. Photochem Photobiol 90. 10.1111/php.12264 [DOI] [PubMed]
- Bakmiwewa SM, Heng B, Guillemin GJ et al (2015) An effective, low-cost method for achieving and maintaining hypoxia during cell culture studies. Biotechniques 59:223–229. 10.2144/000114341 [DOI] [PubMed] [Google Scholar]
- Banerjee D, Jacob J, Kunjamma G et al (2004) Measurement of urinary hydrogen peroxide by FOX-1 method in conjunction with catalase in diabetes mellitus-a sensitive and specific approach [2]. Clin Chim Acta. 10.1016/j.cccn.2004.07.026 [DOI] [PubMed] [Google Scholar]
- Baptista MS, Cadet J, Di Mascio P et al (2017) Type I and type II photosensitized oxidation reactions: guidelines and mechanistic pathways. Photochem Photobiol 93:912–919. 10.1111/php.12716 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Becceneri AB, Fuzer AM, Plutin AM et al (2020) Three-dimensional cell culture models for metallodrug testing: induction of apoptosis and phenotypic reversion of breast cancer cells by the: <trans>-[Ru(PPh3)2(N, N -dimethyl- N -thiophenylthioureato-k2O,S)(bipy)]PF6</trans>complex. Inorg Chem Front 7:2909–2919. 10.1039/d0qi00502a [DOI] [Google Scholar]
- Becceneri AB, Martin MT, Graminha AE et al (2024) The effect of light irradiation on a nitro-ruthenium porphyrin complex in the induced death of lung cancer cells in two- and three-dimensional cultures: insights into the effect of nitric oxide. Dalton Trans 53:11264–11275. 10.1039/D4DT00381K [DOI] [PubMed] [Google Scholar]
- Benov L (2015) Photodynamic therapy: Current status and future directions. In: Medical Principles and Practice 24(suppl 1):14–28. 10.1159/000362416 [DOI] [PMC free article] [PubMed]
- Bilski PJ, Karriker B, Chignell CF (2009) Quenching and generation of singlet oxygen by hydroethidine and related chromophores. Chem Phys Lett 475. 10.1016/j.cplett.2009.05.032 [DOI]
- Bleau G, Giasson C, Brunette I (1998) Measurement of hydrogen peroxide in biological samples containing high levels of ascorbic acid. Anal Biochem 263. 10.1006/abio.1998.2801 [DOI] [PubMed]
- Botto L, Beretta E, Bulbarelli A et al (2008) Hypoxia-induced modifications in plasma membranes and lipid microdomains in A549 cells and primary human alveolar cells. J Cell Biochem 105:503–513. 10.1002/jcb.21850 [DOI] [PubMed] [Google Scholar]
- Bou R, Codony R, Tres A et al (2008) Determination of hydroperoxides in foods and biological samples by the ferrous oxidation-xylenol orange method: a review of the factors that influence the method’s performance. Anal Biochem. 10.1016/j.ab.2008.02.029 [DOI] [PubMed] [Google Scholar]
- Carneiro J, Gonçalves A, Zhou Z, et al (2018) Synthesis and in vitro PDT evaluation of new porphyrins containing meso-epoxymethylaryl cationic groups. Lasers Surg Med 50. 10.1002/lsm.22824 [DOI] [PMC free article] [PubMed]
- Castano AP, Demidova TN, Hamblin MR (2005) Mechanisms in photodynamic therapy: Part two - Cellular signaling, cell metabolism and modes of cell death. Photodiagnosis Photodyn Ther 2(1):1–23. 10.1016/S1572-1000(05)00030-X [DOI] [PMC free article] [PubMed]
- Chandra A, Singh N (2017) Cell microenvironment pH sensing in 3D microgels using fluorescent carbon dots. ACS Biomater Sci Eng 3:3620–3627. 10.1021/acsbiomaterials.7b00740 [DOI] [PubMed] [Google Scholar]
- Cheeseman JM (2006) Hydrogen peroxide concentrations in leaves under natural conditions. J Exp Bot 57. 10.1093/jxb/erl004 [DOI] [PubMed]
- Chen B, Yang Y, Wang Y, et al (2021a) Precise Monitoring of Singlet Oxygen in Specific Endocytic Organelles by Super-pH-Resolved Nanosensors. ACS Appl Mater Interfaces 13:. 10.1021/acsami.1c01730 [DOI] [PubMed]
- Chen D, Xu Q, Wang W, et al (2021b) Type I Photosensitizers Revitalizing Photodynamic Oncotherapy. Small 17,2006742. 10.1002/smll.202006742 [DOI] [PubMed]
- Chen F, Cui K, Si S et al (2025) Polypyridyl Os(II) complexes as efficient human non-small cell lung cancer photosensitizers with enhanced singlet oxygen generation via the fused π-ring elongation. Inorg Chim Acta 578:122537. 10.1016/J.ICA.2025.122537 [DOI] [Google Scholar]
- Chen H, Tian J, He W, Guo Z (2015) H2O2-activatable and O2-evolving nanoparticles for highly efficient and selective photodynamic therapy against hypoxic tumor cells. J Am Chem Soc 137:. 10.1021/ja511420n [DOI] [PubMed]
- Chen Z, Han F, Du Y et al (2023) Hypoxic microenvironment in cancer: molecular mechanisms and therapeutic interventions. Signal Transduct Target Ther 8:70. 10.1038/s41392-023-01332-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chota A, Abrahamse H, George BP (2024) Green synthesis and characterization of AgNPs, liposomal loaded AgNPs and ZnPcS4 photosensitizer for enhanced photodynamic therapy effects in MCF-7 breast cancer cells. Photodiagnosis Photodyn Ther. 10.1016/j.pdpdt.2024.104252 [DOI] [PubMed] [Google Scholar]
- Costes S V., Daelemans D, Cho EH, et al (2004) Automatic and quantitative measurement of protein-protein colocalization in live cells. Biophys J 86:. 10.1529/biophysj.103.038422 [DOI] [PMC free article] [PubMed]
- Cummins TD, Higdon AN, Kramer PA, et al (2013) Utilization of fluorescent probes for the quantification and identification of subcellular proteomes and biological processes regulated by lipid peroxidation products. Free Radic Biol Med 59:56–68. 10.1016/j.freeradbiomed.2012.08.014 [DOI] [PMC free article] [PubMed]
- da Hora Machado (2000) Terapia fotodinâmica: princípios, potencial de aplicação e perspectivas. Quím. Nova 23(2). 10.1590/S0100-40422000000200015 [DOI]
- Daicoviciu D, Filip AG, Ion RM et al (2011) Oxidative photodamage induced by photodynamic therapy with methoxyphenyl porphyrin derivatives in tumour-bearing rats. Folia Biol (Praha). 10.14712/fb2011057010012 [DOI] [PubMed] [Google Scholar]
- David O, Bourré L, Krika Y et al (2005) DNA damages after SIM01 photodynamic treatment. Photodiagnosis Photodyn Ther 2:25–33. 10.1016/S1572-1000(05)00011-6 [DOI] [PubMed] [Google Scholar]
- Davies MJ (2004) Reactive species formed on proteins exposed to singlet oxygen. Photochem Photobiol Sci 3:17–25. 10.1039/b307576c [DOI] [PubMed] [Google Scholar]
- de Melo Gomes LC, de Oliveira Cunha AB, Peixoto LFF, et al (2023) Photodynamic therapy reduces cell viability, migration and triggers necroptosis in prostate tumor cells. Photochemical and Photobiological Sciences 22:. 10.1007/s43630-023-00382-9 [DOI] [PMC free article] [PubMed]
- Digby EM, Rana R, Nitz M, Beharry AA (2019) DNA directed damage using a brominated DAPI derivative. Chem Commun (Camb) 55:9971–9974. 10.1039/c9cc03942b [DOI] [PubMed] [Google Scholar]
- Dolgachev V, Oberley LW, Huang TT et al (2005) A role for manganese superoxide dismutase in apoptosis after photosensitization. Biochem Biophys Res Commun 332:411–417. 10.1016/J.BBRC.2005.04.141 [DOI] [PubMed] [Google Scholar]
- dos Santos AF, de Almeida DRQ, Terra LF et al (2020) Fluence rate determines PDT efficiency in breast cancer cells displaying different GSH levels. Photochem Photobiol. 10.1111/php.13182 [DOI] [PubMed] [Google Scholar]
- Du HY, Olivo M, Tan BKH, Bay BH (2003) Hypericin-mediated photodynamic therapy induces lipid peroxidation and necrosis in nasopharyngeal cancer. Int J Oncol 23:. 10.3892/ijo.23.5.1401 [DOI] [PubMed]
- Dunn KW, Kamocka MM, McDonald JH (2011) A practical guide to evaluating colocalization in biological microscopy. Am J Physiol Cell Physiol. 10.1152/ajpcell.00462.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Du Y, Han J, Jin F, Du Y (2022) Recent strategies to address hypoxic tumor environments in photodynamic therapy. Pharmaceutics. 10.3390/pharmaceutics14091763 [DOI] [PMC free article] [PubMed] [Google Scholar]
- El-Hussein A, Harith M, Abrahamse H (2012) Assessment of DNA damage after photodynamic therapy using a metallophthalocyanine photosensitizer. Int J Photoenergy. 10.1155/2012/281068 [DOI] [Google Scholar]
- El-Sadek MZ, El-Aziz MKA, Shaaban AH et al (2025) Advancements and emerging trends in photodynamic therapy: innovations in cancer treatment and beyond. Photochem Photobiol Sci 24:1489–1511. 10.1007/s43630-025-00765-0 [DOI] [PubMed] [Google Scholar]
- Engrácia DM, Pinto CIG, Mendes F (2023) Cancer 3D Models for Metallodrug Preclinical Testing. Int J Mol Sci 24(15):11915. 10.3390/ijms241511915 [DOI] [PMC free article] [PubMed]
- Eruslanov E, Kusmartsev S (2010) Identification of ROS using oxidized DCFDA and flow-cytometry. Methods in Molecular Biology 594:. 10.1007/978-1-60761-411-1_4 [DOI] [PubMed]
- Evans CL (2015) Three-dimensional in vitro cancer spheroid models for photodynamic therapy: Strengths and opportunities. Front Phys 3:15. 10.3389/fphy.2015.00015 [DOI]
- Ezzeddine R, Al-Banaw A, Tovmasyan A et al (2013) Effect of molecular characteristics on cellular uptake, subcellular localization, and phototoxicity of Zn(II) N-alkylpyridylporphyrins. J Biol Chem 288:36579–36588. 10.1074/jbc.M113.511642 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fay FS, Taneja KL, Shenoy S, et al (1997) Quantitative digital analysis of diffuse and concentrated nuclear distributions of nascent transcripts, SC35 and poly(A). Exp Cell Res 231:. 10.1006/excr.1996.3460 [DOI] [PubMed]
- Ferguson DCJ, Smerdon GR, Harries LW et al (2018) Altered cellular redox homeostasis and redox responses under standard oxygen cell culture conditions versus physioxia. Free Radic Biol Med 126:322–333. 10.1016/J.FREERADBIOMED.2018.08.025 [DOI] [PubMed] [Google Scholar]
- Flors C, Fryer MJ, Waring J, et al (2006) Imaging the production of singlet oxygen in vivo using a new fluorescent sensor, Singlet Oxygen Sensor Green®. In: J Exp Bot 57(8):1725–1734. 10.1093/jxb/erj181 [DOI] [PubMed]
- Fuhrmann DC, Brüne B (2017) Mitochondrial composition and function under the control of hypoxia. Redox Biol 12:208–215. 10.1016/j.redox.2017.02.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fukuzawa K, Fujisaki A, Akai K, et al (2006) Measurement of phosphatidylcholine hydroperoxides in solution and in intact membranes by the ferric-xylenol orange assay. Anal Biochem 359:. 10.1016/j.ab.2006.09.011 [DOI] [PubMed]
- Girotti AW (1990) Photodynamic lipid peroxidation in biological systems. Photochem Photobiol. 10.1111/j.1751-1097.1990.tb01744.x [DOI] [PubMed] [Google Scholar]
- Gollmer A, Arnbjerg J, Blaikie FH et al (2011) Singlet oxygen sensor green®: photochemical behavior in solution and in a mammalian cell. Photochem Photobiol. 10.1111/j.1751-1097.2011.00900.x [DOI] [PubMed] [Google Scholar]
- Gomes A, Fernandes E, Lima JLFC (2005) Fluorescence probes used for detection of reactive oxygen species. J Biochem Biophys Methods 65:45–80. 10.1016/j.jbbm.2005.10.003 [DOI] [PubMed]
- Greene LE, Lincoln R, Cosa G (2018) Spatio-temporal monitoring of lipid peroxyl radicals in live cell studies combining fluorogenic antioxidants and fluorescence microscopy methods. Free Radic Biol Med 128:124–136. 10.1016/J.FREERADBIOMED.2018.04.006 [DOI] [PubMed] [Google Scholar]
- Guo M, Song LP, Jiang Y et al (2006) Hypoxia-mimetic agents desferrioxamine and cobalt chloride induce leukemic cell apoptosis through different hypoxia-inducible factor-1α independent mechanisms. Apoptosis 11:67–77. 10.1007/s10495-005-3085-3 [DOI] [PubMed] [Google Scholar]
- Halliwell B, Chirico S (1993) Lipid peroxidation: Its mechanism, measurement, and significance. American Journal of Clinical Nutrition 57:. 10.1093/ajcn/57.5.715s [DOI] [PubMed]
- Hatz S, Lambert JDC, Ogilby PR (2007) Measuring the lifetime of singlet oxygen in a single cell: addressing the issue of cell viability. Photochem Photobiol Sci. 10.1039/b707313e [DOI] [PubMed] [Google Scholar]
- Hermes-Lima M, Willmore WG, Storey KB (1995) Quantification of lipid peroxidation in tissue extracts based on Fe(III)xylenol orange complex formation. Free Radic Biol Med 19:. 10.1016/0891-5849(95)00020-X [DOI] [PubMed]
- Hirakawa K, Katayama A, Yamaoka S et al (2022) Photosensitized protein damage by water-soluble phthalocyanine zinc(II) and gallium(III) complexes through electron transfer and singlet oxygen production. Chem Phys Lett 802:139764. 10.1016/J.CPLETT.2022.139764 [DOI] [Google Scholar]
- Hirakawa K, Takai S, Horiuchi H, Okazaki S (2020) Photooxidation activity control of dimethylaminophenyltris-(N-methyl-4-pridinio)porphyrin by pH. ACS Omega 5:27702–27708. 10.1021/acsomega.0c04303 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hong L, Li J, Luo Y et al (2022) Recent advances in strategies for addressing hypoxia in tumor photodynamic therapy. Biomolecules. 10.3390/biom12010081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ihata T, Nonoguchi N, Fujishiro T, et al (2022) The effect of hypoxia on photodynamic therapy with 5–4 aminolevulinic acid in malignant gliomas 40:103056. 10.1016/j.pdpdt.2022.103056 [DOI] [PubMed]
- Ikeda A, Satake S, Mae T, et al (2017) Photodynamic Activities of Porphyrin Derivative-Cyclodextrin Complexes by Photoirradiation. ACS Med Chem Lett 8:. 10.1021/acsmedchemlett.7b00098 [DOI] [PMC free article] [PubMed]
- Infantino V, Santarsiero A, Convertini P et al (2021) Cancer cell metabolism in hypoxia: role of HIF-1 as key regulator and therapeutic target. Int J Mol Sci. 10.3390/ijms22115703 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ion RM, Daicoviciu D, Filip AG, et al (2012) Oxidative stress effects of fullerene-porphyrin derivatives in photodynamic therapy. J Porphyr Phthalocyanines 16:. 10.1142/S1088424612500939 [DOI]
- Jabłoński A (1935) Über den Mechanismus der Photolumineszenz von Farbstoffphosphoren 94:38–46. 10.1007/BF01330795 [DOI]
- Jiang ZY, Hunt J V., Wolff SP (1992) Ferrous ion oxidation in the presence of xylenol orange for detection of lipid hydroperoxide in low density lipoprotein. Anal Biochem 202:. 10.1016/0003-2697(92)90122-N [DOI] [PubMed]
- Jiang ZY, Woollard ACS, Wolff SP (1991) Lipid hydroperoxide measurement by oxidation of Fe2+ in the presence of xylenol orange. Comparison with the TBA assay and an iodometric method. Lipids. 10.1007/BF02536169 [DOI] [PubMed] [Google Scholar]
- Juan CA, Pérez de la Lastra JM, Plou FJ, Pérez-Lebeña E (2021) The Chemistry of Reactive Oxygen Species (ROS) Revisited: Outlining Their Role in Biological Macromolecules (DNA, Lipids and Proteins) and Induced Pathologies. Int J Mol Sci 22:4642. 10.3390/ijms22094642 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kalyanaraman B, Dranka BP, Hardy M et al (2014) HPLC-based monitoring of products formed from hydroethidine-based fluorogenic probes - the ultimate approach for intra- and extracellular superoxide detection. Biochim Biophys Acta Gen Subj. 10.1016/j.bbagen.2013.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Karagöz ID, Yilmaz Y, Sanusi K (2020) Anticancer Activity Study and Density Functional/Time-Dependent Density Functional Theory (DFT/TD-DFT) Calculations of 2(3),9(10),16(17),23(24)-Tetrakis-(6-Methylpyridin-2-Yloxy)Phthalocyaninato Zn(II). J Fluoresc 30:. 10.1007/s10895-020-02584-1 [DOI] [PubMed]
- Kaur B, Miglioranza Scavuzzi B, Abcouwer S, Zacks D (2023) A simplified protocol to induce hypoxia in a standard incubator: a focus on retinal cells. Exp Eye Res 236:109653. 10.1016/J.EXER.2023.109653 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kessel D, Andrzejak M, Price M (2011) Biological consequences of PDT: tying up the loose ends. In: Optical Methods for Tumor Treatment and Detection: Mechanisms and Techniques in Photodynamic Therapy XX:788602. 10.1117/12.876209 [DOI]
- Kiesslich T, Gollmer A, Maisch T, et al (2013) A comprehensive tutorial on in vitro characterization of new photosensitizers for photodynamic antitumor therapy and photodynamic inactivation of microorganisms. Biomed Res Int 840417:1–17. 10.1155/2013/840417 [DOI] [PMC free article] [PubMed]
- Kilinç S, Güney S (2019) Effects of hypoxic preconditioning on oxidant-antioxidant systems in rat lung. Turk Klinikleri J Med Sci 39:194–201. 10.5336/medsci.2018-62968 [DOI] [Google Scholar]
- Kim JY, Il CW, Kim M, Tae G (2013a) Tumor-targeting nanogel that can function independently for both photodynamic and photothermal therapy and its synergy from the procedure of PDT followed by PTT. J Control Release 171:113–121. 10.1016/J.JCONREL.2013.07.006 [DOI] [PubMed] [Google Scholar]
- Kim S, Fujitsuka M, Majima T (2013b) Photochemistry of singlet oxygen sensor green. Journal of Physical Chemistry B 117:. 10.1021/jp406638g [DOI] [PubMed]
- Kojima Y, Tanaka M, Sasaki M, et al (2024) Induction of ferroptosis by photodynamic therapy and enhancement of antitumor effect with ferroptosis inducers. J Gastroenterol 59:. 10.1007/s00535-023-02054-y [DOI] [PubMed]
- Kraljić I, Mohsni S El (1978) A NEW METHOD FOR THE DETECTION OF SINGLET OXYGEN IN AQUEOUS SOLUTIONS. Photochem Photobiol 28:. 10.1111/j.1751-1097.1978.tb06972.x [DOI]
- Kucinska M, Plewinski A, Szczolko W et al (2021) Modeling the photodynamic effect in 2D versus 3D cell culture under normoxic and hypoxic conditions. Free Radic Biol Med 162:309–326. 10.1016/J.FREERADBIOMED.2020.10.304 [DOI] [PubMed] [Google Scholar]
- Kurutos A, Shindo Y, Hiruta Y, et al (2022) Organelle-selective near-infrared fluorescent probes for intracellular microenvironment labeling. Dyes and Pigments 204:. 10.1016/j.dyepig.2022.110424 [DOI]
- Lakowicz J (1994) Topics in fluorescence spectroscopy, protein fluorescence [Google Scholar]
- Lange C, Lehmann C, Mahler M, Bednarski PJ (2019) Comparison of cellular death pathways after mTHPC-mediated photodynamic therapy (PDT) in five human cancer cell lines. Cancers (Basel) 11:. 10.3390/cancers11050702 [DOI] [PMC free article] [PubMed]
- Levine RL, Williams JA, Stadtman EP, Shacter E (1994) [37] Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol 233:346–357. 10.1016/S0076-6879(94)33040-9 [DOI] [PubMed] [Google Scholar]
- Li J, Zhuang Z, Zhao Z, Tang BZ (2022a) Type I AIE photosensitizers: Mechanism and application. VIEW 3, 20200121. 10.1002/VIW.20200121 [DOI]
- Li M, Shao Y, Kim JH et al (2020) Unimolecular photodynamic O2-economizer to overcome hypoxia resistance in phototherapeutics. J Am Chem Soc 142:5380–5388. 10.1021/jacs.0c00734 [DOI] [PubMed] [Google Scholar]
- Li M, Xia J, Tian R et al (2018a) Near-infrared light-initiated molecular superoxide radical generator: rejuvenating photodynamic therapy against hypoxic tumors. J Am Chem Soc 140:14851–14859. 10.1021/jacs.8b08658 [DOI] [PubMed] [Google Scholar]
- Li M, Xiong T, Du J et al (2019) Superoxide radical photogenerator with amplification effect: surmounting the Achilles’ heels of photodynamic oncotherapy. J Am Chem Soc 141:2695–2702. 10.1021/jacs.8b13141 [DOI] [PubMed] [Google Scholar]
- Li M, Xu Y, Peng X, Kim JS (2022b) From low to no O2-dependent hypoxia photodynamic therapy (hPDT): a new perspective. Acc Chem Res 55:3253–3264. 10.1021/acs.accounts.2c00531 [DOI] [PubMed] [Google Scholar]
- Li S, Chen Y, Wu Y, et al (2022c) An Endoplasmic Reticulum Targeting Type I Photosensitizer for Effective Photodynamic Therapy against Hypoxic Tumor Cells. Chemistry - A European Journal 28:. 10.1002/chem.202202680 [DOI] [PubMed]
- Liu F, Zhou X, Chen Z et al (2008) Preparation of purpurin-18 loaded magnetic nanocarriers in cottonseed oil for photodynamic therapy. Mater Lett 62:2844–2847. 10.1016/J.MATLET.2008.01.123 [DOI] [Google Scholar]
- Li X, Chen L, Huang M et al (2023) Innovative strategies for photodynamic therapy against hypoxic tumor. Asian J Pharm Sci. 10.1016/j.ajps.2023.100775 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li X, Kwon N, Guo T et al (2018b) Innovative strategies for hypoxic-tumor photodynamic therapy. Angew Chem Int Ed Engl 57:11522–11531. 10.1002/anie.201805138 [DOI] [PubMed] [Google Scholar]
- Li Y, Wang KN, Liu B et al (2018c) Mitochondria-targeted two-photon fluorescent probe for the detection of biothiols in living cells. Sens Actuators B Chem 255:193–202. 10.1016/J.SNB.2017.08.041 [DOI] [Google Scholar]
- Li Y, Yang T, Liu S, et al (2024) Assays on 3D tumor spheroids for exploring the light dosimetry of photodynamic effects under different gaseous conditions. J Biophotonics 17:. 10.1002/jbio.202300552 [DOI] [PubMed]
- Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275. 10.1016/s0021-9258(19)52451-6 [DOI] [PubMed] [Google Scholar]
- Lv W, Zhang Z, Zhang KY et al (2016a) A mitochondria-targeted photosensitizer showing improved photodynamic therapy effects under hypoxia. Angew Chem 128:10101–10105. 10.1002/ange.201604130 [DOI] [PubMed] [Google Scholar]
- Lv W, Zhang Z, Zhang KY et al (2016b) A mitochondria-targeted photosensitizer showing improved photodynamic therapy effects under hypoxia. Angew Chem Int Ed 55:9947–9951. 10.1002/anie.201604130 [DOI] [PubMed] [Google Scholar]
- Lv Z, Wei H, Li Q et al (2018) Achieving efficient photodynamic therapy under both normoxia and hypoxia using cyclometalated Ru(II) photosensitizer through type I photochemical process. Chem Sci 9:502–512. 10.1039/c7sc03765a [DOI] [PMC free article] [PubMed] [Google Scholar]
- MacDonald IJ, Morgan J, Bellnier DA et al (1999) Subcellular Localization Patterns and Their Relationship to Photodynamic Activity of Pyropheophorbide-a Derivatives. Photochem Photobiol 70:789. 10.1562/0031-8655(1999)070<0789:SLPATR>2.3.CO;2 [DOI] [PubMed] [Google Scholar]
- Malacarne MC, Mastore M, Gariboldi MB, et al (2023) Preliminary Toxicity Evaluation of a Porphyrin Photosensitizer in an Alternative Preclinical Model. Int J Mol Sci 24:. 10.3390/ijms24043131 [DOI] [PMC free article] [PubMed]
- Mallidi S, Anbil S, Bulin A-L et al (2016) Beyond the barriers of light penetration: strategies, perspectives and possibilities for photodynamic therapy. Theranostics 6:2458–2487. 10.7150/thno.16183 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mamalis A, Koo E, Sckisel GD, et al (2016) Temperature-dependent impact of thermal aminolaevulinic acid photodynamic therapy on apoptosis and reactive oxygen species generation in human dermal fibroblasts. British Journal of Dermatology 175:. 10.1111/bjd.14509 [DOI] [PubMed]
- Martinez De Pinillos Bayona A, Mroz P, Thunshelle C, Hamblin MR (2017) Design features for optimization of tetrapyrrole macrocycles as antimicrobial and anticancer photosensitizers. Chem Biol Drug des 89:192–206. 10.1111/cbdd.12792 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martins WK, Santos NF, Rocha CdeS et al (2019) Parallel damage in mitochondria and lysosomes is an efficient way to photoinduce cell death. Autophagy. 10.1080/15548627.2018.1515609 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matthiesen S, Jahnke R, Knittler MR (2021a) A straightforward hypoxic cell culture method suitable for standard incubators. Methods Protoc 4:. 10.3390/mps4020025 [DOI] [PMC free article] [PubMed]
- Matthiesen S, Jahnke R, Knittler MR (2021b) A straightforward hypoxic cell culture method suitable for standard incubators. Methods Protoc 4:. 10.3390/mps4020025 [DOI] [PMC free article] [PubMed]
- Ma Y, Zhao X, Tian P, et al (2024) Laser-Ignited Lipid Peroxidation Nanoamplifiers for Strengthening Tumor Photodynamic Therapy Through Aggravating Ferroptotic Propagation and Sustainable High Immunogenicity. Small 20:. 10.1002/smll.202306402 [DOI] [PubMed]
- Melo T, Santos N, Lopes D, et al (2013) Photosensitized oxidation of phosphatidylethanolamines monitored by electrospray tandem mass spectrometry. Journal of Mass Spectrometry 48:. 10.1002/jms.3301 [DOI] [PubMed]
- Moan J, Vistnes AI (1986) Porphyrin photosensitization of proteins in cell membranes as studied by spin‐labelling and by quantification of dtnb‐reactive sh‐groups. Photochem Photobiol 44:. 10.1111/j.1751-1097.1986.tb03558.x [DOI] [PubMed]
- Mohammad-Hadi L, MacRobert AJ, Loizidou M, Yaghini E (2018) Photodynamic therapy in 3D cancer models and the utilisation of nanodelivery systems. Nanoscale 10:1570–1581. 10.1039/C7NR07739D [DOI] [PubMed] [Google Scholar]
- Moloudi K, Abrahamse H, George BP (2025) Overcoming photodynamic therapy resistance via DNA repair inhibitors: advanced review. Photodiagnosis Photodyn Ther. 10.1016/j.pdpdt.2025.104724 [DOI] [PubMed] [Google Scholar]
- Moserova I, Kralova J (2012) Role of er stress response in photodynamic therapy: Ros generated in different subcellular compartments trigger diverse cell death pathways. PLoS One 7:. 10.1371/journal.pone.0032972 [DOI] [PMC free article] [PubMed]
- Muñiz-García A, Romero M, Falcόn-Perez JM, et al (2022) Hypoxia-induced HIF1α activation regulates small extracellular vesicle release in human embryonic kidney cells. Sci Rep 12:. 10.1038/s41598-022-05161-7 [DOI] [PMC free article] [PubMed]
- Murphy MP, Bayir H, Belousov V, et al (2022) Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo. Nat Metab 4:. 10.1038/s42255-022-00591-z [DOI] [PMC free article] [PubMed]
- Nauseef WM (2014) Detection of superoxide anion and hydrogen peroxide production by cellular NADPH oxidases. Biochim Biophys Acta Gen Subj. 10.1016/j.bbagen.2013.04.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ng NS, Ooi L (2021) A simple microplate assay for reactive oxygen species generation and rapid cellular protein normalization. Bio Protoc 11:. 10.21769/BioProtoc.3877 [DOI] [PMC free article] [PubMed]
- Nguyen VN, Qi S, Kim S et al (2019) An emerging molecular design approach to heavy-atom-free photosensitizers for enhanced photodynamic therapy under hypoxia. J Am Chem Soc 141:16243–16248. 10.1021/jacs.9b09220 [DOI] [PubMed] [Google Scholar]
- Nguyen VN, Yan Y, Zhao J, Yoon J (2021) Heavy-atom-free photosensitizers: from molecular design to applications in the photodynamic therapy of cancer. Acc Chem Res 54:207–220. 10.1021/acs.accounts.0c00606 [DOI] [PubMed] [Google Scholar]
- Niki E (2014) Biomarkers of lipid peroxidation in clinical material. Biochim Biophys Acta Gen Subj. 10.1016/j.bbagen.2013.03.020 [DOI] [PubMed] [Google Scholar]
- Niki E (2009) Lipid peroxidation: physiological levels and dual biological effects. Free Radic Biol Med. 10.1016/j.freeradbiomed.2009.05.032 [DOI] [PubMed] [Google Scholar]
- Nourooz-Zadeh J, Tajaddini-Sarmadi J, Wolff SP (1994) Measurement of plasma hydroperoxide concentrations by the ferrous oxidation-xylenol orange assay in conjunction with triphenylphosphine. Anal Biochem 220:. 10.1006/abio.1994.1357 [DOI] [PubMed]
- Ochsner M (1997) Photophysical and photobiological processes in the photodynamic therapy of tumours. J Photochem Photobiol B 39:1–18. 10.1016/S1011-1344(96)07428-3 [DOI] [PubMed] [Google Scholar]
- OECD (2024) Guidance Document on an integrated approach on testing and assessment (IATA) for Phototoxicity Testing. OECD Series on Testing and Assessment, No. 397, OECD Publishing, Paris.
- Onoue S, Kawamura K, Igarashi N et al (2008) Reactive oxygen species assay-based risk assessment of drug-induced phototoxicity: Classification criteria and application to drug candidates. J Pharm Biomed Anal 47:967–972. 10.1016/J.JPBA.2008.03.026 [DOI] [PubMed] [Google Scholar]
- Ortega-Villasante C, Burén S, Blázquez-Castro A et al (2018) Fluorescent in vivo imaging of reactive oxygen species and redox potential in plants. Free Radic Biol Med. 10.1016/j.freeradbiomed.2018.04.005 [DOI] [PubMed] [Google Scholar]
- Panzarini E, Inguscio V, Dini L (2011) Timing the multiple cell death pathways initiated by Rose Bengal acetate photodynamic therapy. Cell Death Dis 2:. 10.1038/cddis.2011.51 [DOI] [PMC free article] [PubMed]
- Parks SK, Cormerais Y, Marchiq I, Pouyssegur J (2016) Hypoxia optimises tumour growth by controlling nutrient import and acidic metabolite export. Mol Aspects Med 47–48:3–14. 10.1016/J.MAM.2015.12.001 [DOI] [PubMed] [Google Scholar]
- Peng W, de Bruijn HS, Ten Hagen TLM, et al (2020) In-vivo optical monitoring of the efficacy of epidermal growth factor receptor targeted photodynamic therapy: The effect of fluence rate. Cancers (Basel) 12:. 10.3390/cancers12010190 [DOI] [PMC free article] [PubMed]
- Pereira PMR, Berisha N, Bhupathiraju NVSDK, et al (2017) Cancer cell spheroids are a better screen for the photodynamic efficiency of glycosylated photosensitizers. PLoS One 12:. 10.1371/journal.pone.0177737 [DOI] [PMC free article] [PubMed]
- Peskova NN, Brilkina AA, Gorokhova AA et al (2021) The localization of the photosensitizer determines the dynamics of the secondary production of hydrogen peroxide in cell cytoplasm and mitochondria. J Photochem Photobiol B 219:112208. 10.1016/J.JPHOTOBIOL.2021.112208 [DOI] [PubMed] [Google Scholar]
- Prasad A, Sedlářová M, Pospíšil P (2018) Singlet oxygen imaging using fluorescent probe Singlet Oxygen Sensor Green in photosynthetic organisms. Sci Rep 8:. 10.1038/s41598-018-31638-5 [DOI] [PMC free article] [PubMed]
- Price M, Kessel D (2010) On the use of fluorescence probes for detecting reactive oxygen and nitrogen species associated with photodynamic therapy. J Biomed Opt. 10.1117/1.3484258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Price M, Reiners JJ, Santiago AM, Kessel D (2009) Monitoring singlet oxygen and hydroxyl radical formation with fluorescent probes during photodynamic therapy. Photochem Photobiol 85:. 10.1111/j.1751-1097.2009.00555.x [DOI] [PMC free article] [PubMed]
- Pucelik B, Arnaut LG, Dąbrowski JM (2020) Lipophilicity of bacteriochlorin-based photosensitizers as a determinant for pdt optimization through the modulation of the inflammatory mediators. J Clin Med 9:. 10.3390/jcm9010008 [DOI] [PMC free article] [PubMed]
- Puttaswamy NY, Mahanta P, Sarma P, et al (2023) Structure-based biological investigations on ruthenium complexes containing 2,2′-bipyridine ligands and their applications in photodynamic therapy as a potential photosensitizer. Chem Biol Drug Des 102:. 10.1111/cbdd.14341 [DOI] [PubMed]
- Qi F, Yuan H, Chen Y et al (2023) Type I photoreaction and photoinduced ferroptosis by a Ru(II) complex to overcome tumor hypoxia in photodynamic therapy. CCS Chem 5:1583–1591. 10.31635/ccschem.022.202202074 [DOI] [Google Scholar]
- Ragàs X, Jiménez-Banzo A, Sánchez-García D et al (2009) Singlet oxygen photosensitisation by the fluorescent probe Singlet Oxygen Sensor Green®. Chem Commun. 10.1039/b822776d [DOI] [PubMed] [Google Scholar]
- Rapozzi V, Miculan M, Xodo LE (2009) Evidence that photoactivated pheophorbide a causes in human cancer cells a photodynamic effect involving lipid peroxidation. Cancer Biol Ther 8:. 10.4161/cbt.8.14.8628 [DOI] [PubMed]
- Raza A, Colley HE, Baggaley E et al (2017) Oxygen Mapping of Melanoma Spheroids using Small Molecule Platinum Probe and Phosphorescence Lifetime Imaging Microscopy. Sci Rep 7:10743. 10.1038/s41598-017-11153-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rezende LG, Tasso TT, Candido PHS, Baptista MS (2022) Assessing photosensitized membrane damage: available tools and comprehensive mechanisms†. Photochem Photobiol 98:572–590. 10.1111/php.13582 [DOI] [PubMed] [Google Scholar]
- Rice-Evans CA, Diplock AT, Symons MCR (1991) Laboratory techniques in biochemistry and molecular biology: techniques in free radical research. Elsevier [Google Scholar]
- Rinderknecht H, Ehnert S, Braun B et al (2021) The Art of Inducing Hypoxia Oxygen 1:46–61. 10.3390/oxygen1010006 [DOI] [Google Scholar]
- Riske KA, Sudbrack TP, Archilha NL et al (2009) Giant vesicles under oxidative stress induced by a membrane-anchored photosensitizer. Biophys J 97:1362–1370. 10.1016/j.bpj.2009.06.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ronsein GE, Miyamoto S, Bechara E et al (2006) Oxidação de proteínas por oxigênio singlete: mecanismos de dano, estratégias para detecção e implicações biológicas. Quim Nova 29:563–568. 10.1590/S0100-40422006000300027 [DOI] [Google Scholar]
- Rubio CP, Cerón JJ (2021) Spectrophotometric assays for evaluation of Reactive Oxygen Species (ROS) in serum: general concepts and applications in dogs and humans. BMC Vet Res 17:226. 10.1186/s12917-021-02924-8 [DOI] [PMC free article] [PubMed]
- Saczko J, Kulbacka J, Chwiłkowska A et al (2004) Levels of lipid peroxidation in A549 cells after PDT in vitro. Rocz Akad Med Bialymst 49(Suppl):1 [PubMed] [Google Scholar]
- Sakharov DV, Elstak EDR, Chernyak B, Wirtz KWA (2005) Prolonged lipid oxidation after photodynamic treatment. Study with oxidation-sensitive probe C11-BODIPY581/591. FEBS Lett. 10.1016/j.febslet.2005.01.024 [DOI] [PubMed] [Google Scholar]
- Salvati A, Nelissen I, Haase A et al (2018) Quantitative measurement of nanoparticle uptake by flow cytometry illustrated by an interlaboratory comparison of the uptake of labelled polystyrene nanoparticles. Nanoimpact 9:42–50. 10.1016/J.IMPACT.2017.10.004 [DOI] [Google Scholar]
- Salyha N, Oliynyk I (2023a) Hypoxia modeling techniques: a review. Heliyon. 10.1016/j.heliyon.2023.e13238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Salyha N, Oliynyk I (2023b) Hypoxia modeling techniques: a review. Heliyon 9:e13238. 10.1016/j.heliyon.2023.e13238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 10.1038/nmeth.2089 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seo J, Yun JE, Kim SJ, Chun YS (2022) Lipid metabolic reprogramming by hypoxia-inducible factor-1 in the hypoxic tumour microenvironment. Pflugers Arch 474:591–601. 10.1007/s00424-022-02683-x [DOI] [PubMed] [Google Scholar]
- Setsukinai K, Urano Y, Kakinuma K et al (2003) Development of novel fluorescence probes that can reliably detect reactive oxygen species and distinguish specific species. J Biol Chem. 10.1074/jbc.M209264200 [DOI] [PubMed] [Google Scholar]
- Shekhovtsov SV, Vorontsova YL, Slepneva IA et al (2024) The impact of long-term hypoxia on the antioxidant defense system in the Siberian frog Rana amurensis. Biochemistry (Mosc) 89:441–450. 10.1134/S0006297924030052 [DOI] [PubMed] [Google Scholar]
- Shin HJ, Kwak M, Joo S, Lee JY (2022) Quantifying fluorescent nanoparticle uptake in mammalian cells using a plate reader. Sci Rep 12:. 10.1038/s41598-022-24480-3 [DOI] [PMC free article] [PubMed]
- Shirmanova M V., Lukina MM, Sirotkina MA, et al (2024) Effects of Photodynamic Therapy on Tumor Metabolism and Oxygenation Revealed by Fluorescence and Phosphorescence Lifetime Imaging. Int J Mol Sci 25:. 10.3390/ijms25031703 [DOI] [PMC free article] [PubMed]
- Shui S, Zhao Z, Wang H et al (2021a) Non-enzymatic lipid peroxidation initiated by photodynamic therapy drives a distinct ferroptosis-like cell death pathway. Redox Biol 45:102056. 10.1016/J.REDOX.2021.102056 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shui S, Zhao Z, Wang H, et al (2021b) Non-enzymatic lipid peroxidation initiated by photodynamic therapy drives a distinct ferroptosis-like cell death pathway. Redox Biol 45:. 10.1016/j.redox.2021.102056 [DOI] [PMC free article] [PubMed]
- Simaremare ES, Djalil AD, Tjahjono DH (2015) Photophysical and physicochemical properties of Cu(II)chlorin e4 and Cu(II)chlorin e6 as a lead compound of photosensitizer for PDT. Indones J Pharm. 10.14499/indonesianjpharm26iss1pp29 [DOI] [Google Scholar]
- Sosa MJ, Fonseca JL, Sakaya A, et al (2023) Alkylation converts riboflavin into an efficient photosensitizer of phospholipid membranes. Biochim Biophys Acta Biomembr 1865:. 10.1016/j.bbamem.2023.184155 [DOI] [PubMed]
- Sułek A, Pucelik B, Kobielusz M, et al (2020) Photodynamic inactivation of bacteria with porphyrin derivatives: Effect of charge, lipophilicity, ros generation, and cellular uptake on their biological activity in vitro. Int J Mol Sci 21:. 10.3390/ijms21228716 [DOI] [PMC free article] [PubMed]
- Sun Y, Zhao D, Wang G et al (2020) Recent progress of hypoxia-modulated multifunctional nanomedicines to enhance photodynamic therapy: opportunities, challenges, and future development. Acta Pharm Sin B 10:1382–1396. 10.1016/J.APSB.2020.01.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Szokalska A, Makowski M, Nowis D et al (2009) Proteasome inhibition potentiates antitumor effects of photodynamic therapy in mice through induction of endoplasmic reticulum stress and unfolded protein response. Cancer Res. 10.1158/0008-5472.CAN-08-3439 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Teiten MH, Bezdetnaya L, Morlière P, et al (2003) Endoplasmic reticulum and Golgi apparatus are the preferential sites of Foscan® localisation in cultured tumour cells. Br J Cancer 88:. 10.1038/sj.bjc.6600664 [DOI] [PMC free article] [PubMed]
- Teng KX, Niu LY, Li J et al (2025) Rational design of type-I photodynamic agents. Angew Chem Int Ed. 64e202509416. 10.1002/anie.202509416 [DOI] [PubMed]
- Tseng SP, Teng LJ, Chen CT, et al (2009) Toluidine blue O photodynamic inactivation on multidrug-resistant Pseudomonas aeruginosa. Lasers Surg Med 41:. 10.1002/lsm.20765 [DOI] [PubMed]
- Tsubone TM, Baptista MS, Itri R (2019a) Understanding membrane remodelling initiated by photosensitized lipid oxidation. Biophys Chem 254:. 10.1016/j.bpc.2019.106263 [DOI] [PubMed]
- Tsubone TM, Martins WK, Baptista MS (2019b) Identifying specific subcellular organelle damage by photosensitized oxidations. Yale J Biol Med 92:(3):413–422 [PMC free article] [PubMed]
- Tsubone TM, Martins WK, Franco MSF et al (2021) Cellular compartments challenged by membrane photo-oxidation. Arch Biochem Biophys. 10.1016/j.abb.2020.108665 [DOI] [PubMed] [Google Scholar]
- Tsubone TM, Zhang Z, Goyal R, et al (2020) Porphyrin-Loaded TyroSpheres for the Intracellular Delivery of Drugs and Photoinduced Oxidant Species. Mol Pharm 17:. 10.1021/acs.molpharmaceut.0c00338 [DOI] [PubMed]
- Urazaliyeva A, Kanabekova P, Beisenbayev A, et al (2024) All organic nanomedicine for PDT–PTT combination therapy of cancer cells in hypoxia. Sci Rep 14:. 10.1038/s41598-024-68077-4 [DOI] [PMC free article] [PubMed]
- Vignoni M, Urrutia MN, Junqueira HC et al (2018) Photo-oxidation of unilamellar vesicles by a lipophilic pterin: deciphering biomembrane photodamage. Langmuir. 10.1021/acs.langmuir.8b03302 [DOI] [PubMed] [Google Scholar]
- Wang J, Hou Y, Lei W et al (2012) DNA photocleavage by a cationic BODIPY dye through both singlet oxygen and hydroxyl radical: new insight into the photodynamic mechanism of BODIPYs. ChemPhysChem 13:2739–2747. 10.1002/cphc.201200224 [DOI] [PubMed] [Google Scholar]
- Wen X, Zhang X, Szewczyk G, et al (2017) Potassium iodide potentiates antimicrobial photodynamic inactivation mediated by Rose bengal in in vitro and in vivo studies. Antimicrob Agents Chemother 61:. 10.1128/AAC.00467-17 [DOI] [PMC free article] [PubMed]
- Wolff SP (1994) Ferrous ion oxidation in presence of ferric ion indicator xylenol orange for measurement of hydroperoxides. Methods Enzymol 233:. 10.1016/S0076-6879(94)33021-2 [DOI]
- Wolny D, Stojko M, Zajdel A (2024) Novel strategies of glutathione depletion in photodynamic and chemodynamic therapy: a review. Adv Clin Exp Med 34:1213–1221. 10.17219/acem/191025 [DOI] [PubMed] [Google Scholar]
- Woodburn KW, Vardaxis NJ, Hill JS et al (1991) Subcellular localization of porphyrins using confocal laser scanning microscopy. Photochem Photobiol. 10.1111/j.1751-1097.1991.tb02081.x [DOI] [PubMed] [Google Scholar]
- Wrolstad RE, Acree TE, Decker EA et al (2005) Handbook of food analytical chemistry :pigments, colorants, flavors, texture, and bioactive food components 2:1–624
- Wu D, Yotnda P (2011) Induction and testing of hypoxia in cell culture. J vis Exp. 10.3791/2899 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wulftange WJ, Rose MA, Garmendia-Cedillos M et al (2019) Spatial control of oxygen delivery to three-dimensional cultures alters cancer cell growth and gene expression. J Cell Physiol 234:20608–20622. 10.1002/jcp.28665 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wu W, Shao X, Zhao J, Wu M (2017) Controllable photodynamic therapy implemented by regulating singlet oxygen efficiency. Adv Sci. 10.1002/advs.201700113 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xiao Z, Hansen CB, Allen TM, Miller GG, Moore RB (2005) Distribution of photosensitizers in bladder cancer spheroids: implications for intravesical instillation of photosensitizers for photodynamic therapy of bladder cancer. J Pharm Pharm Sci 8(3):5362–43 [PubMed]
- Xiong T, Chen Y, Peng Q, et al (2024) Lipid Droplet Targeting Type I Photosensitizer for Ferroptosis via Lipid Peroxidation Accumulation. Advanced Materials 36:. 10.1002/adma.202309711 [DOI] [PubMed]
- Xu F, Li H, Yao Q et al (2019) Hypoxia-activated NIR photosensitizer anchoring in the mitochondria for photodynamic therapy. Chem Sci 10:10586–10594. 10.1039/c9sc03355f [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yoo J, Choi JW, Kim E et al (2024) Evaluating cell growth and hypoxic regions of 3D spheroids via a machine learning approach. Mach Learn Sci Technol 5:035063. 10.1088/2632-2153/ad718e [DOI] [Google Scholar]
- Yu D, Zha Y, Zhong Z et al (2021) Improved detection of reactive oxygen species by DCFH-DA: new insight into self-amplification of fluorescence signal by light irradiation. Sens Actuators B Chem 339:129878. 10.1016/J.SNB.2021.129878 [DOI] [Google Scholar]
- Yu L, Xu Y, Pu Z et al (2022) Photocatalytic superoxide radical generator that induces pyroptosis in cancer cells. J Am Chem Soc 144:11326–11337. 10.1021/jacs.2c03256 [DOI] [PubMed] [Google Scholar]
- Zhang Y, Zhu J, Sun H, Li J (2022) Modulation of tumor hypoxia and redox microenvironment using nanomedicines for enhanced cancer photodynamic therapy. Appl Mater Today 29:101687. 10.1016/j.apmt.2022.101687 [DOI] [Google Scholar]
- Zhang Z, Liu Z, Huang Y et al (2023) Magnetic resonance and fluorescence imaging superparamagnetic nanoparticles induce apoptosis and ferroptosis through photodynamic therapy to treat colorectal cancer. Mater Today Phys. 10.1016/j.mtphys.2023.101150 [DOI] [Google Scholar]
- Zheng J, Du J, Ge H, et al (2022) Viscosity-dependent photocatalysis triggers ferroptosis and Type-I photodynamic therapy to kill drug-resistant tumors. Chemical Engineering Journal 449:. 10.1016/j.cej.2022.136565 [DOI]
- Zhou W, Lim A, Elmadbouh OHM, et al (2024) Verteporfin induces lipid peroxidation and ferroptosis in pancreatic cancer cells. Free Radic Biol Med 212:. 10.1016/j.freeradbiomed.2024.01.003 [DOI] [PMC free article] [PubMed]
- Zhuang Z, Dai J, Yu M et al (2020) Type I photosensitizers based on phosphindole oxide for photodynamic therapy: apoptosis and autophagy induced by endoplasmic reticulum stress. Chem Sci 11:3405–3417. 10.1039/D0SC00785D [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhu H, Fang Y, Miao Q, et al (2017) Regulating Near-Infrared Photodynamic Properties of Semiconducting Polymer Nanotheranostics for Optimized Cancer Therapy. ACS Nano 11:. 10.1021/acsnano.7b03507 [DOI] [PubMed]
- Zielonka J, Kalyanaraman B (2010) Hydroethidine- and MitoSOX-derived red fluorescence is not a reliable indicator of intracellular superoxide formation: another inconvenient truth. Free Radic Biol Med. 10.1016/j.freeradbiomed.2010.01.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No datasets were generated or analyzed during the current study.
