5-Aminolevulinic Acid-Mediated Photodynamic Therapy Induces Ferroptosis in Oral Leukoplakia and Oral Squamous Cell Carcinoma

TL;DR

This study shows that 5-aminolevulinic acid photodynamic therapy can trigger a type of cell death called ferroptosis in oral leukoplakia and oral squamous cell carcinoma.

Contribution

The study reveals that ALA-PDT induces ferroptosis in OLK and OSCC, suggesting ferroptosis agonists could improve treatment outcomes.

Findings

ALA-PDT increases ROS and decreases GSH/GSSG ratio in OLK and OSCC cells.

Mitochondrial morphology after ALA-PDT shows features of ferroptosis.

Ferroptosis agonists like erastin enhance the effects of ALA-PDT.

Abstract

5-Aminolevulinic acid (ALA)-mediated photodynamic therapy (PDT) is one of the treatment modalities for oral leukoplakia (OLK) and oral squamous cell carcinoma (OSCC). However, the role of ferroptosis in ALA-PDT for OLK and OSCC remains unclear. Therefore, this study aimed to investigate whether ALA-PDT can induce ferroptosis in OLK and OSCC. We detected relative cellular dehydrogenase activity (CCK-8 assay), long-term proliferative viability, reactive oxygen species (ROS) generation, glutathione levels, and mitochondrial morphology after ALA-PDT. The expression of ferroptosis-related proteins was detected using Western blot. A tongue OSCC model was established in male BalB/c nude mice, and then ALA-PDT was performed. Immunohistochemical staining of Ki67, GPX4 and FTH1 was conducted to evaluate the effect of ALA-PDT. Subsequently, OLK and OSCC cells were pre-treated with ferrostatin-1 (Fer-1) before ALA-PDT. Relative cellular dehydrogenase activity, ROS generation, lipid peroxidation, Fe2+ levels, and ferroptosis-related protein expression were measured. Finally, OLK and OSCC cells were treated with a combination of ALA-PDT and erastin, and mitochondrial function was evaluated. In vitro study showed that ALA-PDT increased ROS generation and decreased GSH/GSSG ratio in OLK and OSCC cells. After ALA-PDT, mitochondrial morphology exhibited typical characteristics of ferroptosis. In vivo experiments showed that immunohistochemistry (IHC) scores of Ki67, GPX4 and FTH1 in the tissues decreased after ALA-PDT. Moreover, pre-treatment with Fer-1 could reverse ROS levels, lipid peroxidation and intracellular Fe2+ accumulation in OLK and OSCC cells after ALA-PDT. Additionally, Fer-1 pre-treatment reversed the changes in protein expression induced by ALA-PDT. The combination of ALA-PDT and erastin significantly reduced mitochondrial O2•− production and decreased mitochondrial membrane potential. Above all, ALA-PDT can induce ferroptosis in OLK and OSCC. The use of ferroptosis agonists may enhance the therapeutic efficacy of ALA-PDT for OLK and OSCC.

Figures

ALA-PDT-induced ferroptosis in OLK and OSCC cells. (A) After co-incubating cells with different concentrations of ALA for 4 h, ALA-PDT was performed. To evaluate the long-term proliferative viability, clonogenic assay was performed (n = 3). (B) Cells were incubated with 1 mM ALA for 4 h in serum-free medium. Cells were then loaded with the DCFH-DA probe for 20 min and subjected to PDT irradiation. After irradiation, cells were cultured in complete medium. Intracellular ROS levels were observed under a fluorescence microscope at 4 h post-irradiation. Green: DCFH-DA, ROS. (C) Evaluated the cellular GSH/GSSG content 24 h after ALA-PDT (n = 3). (D) 24 h after ALA-PDT, the mitochondrial morphology of SCC9 and Leuk1 cells was observed by TEM. Red arrow: mitochondria with typical ferroptosis morphology. a: there is a statistical difference compared to the control group; b: there is a statistical difference compared to 0 mM ALA mediated PDT. ns: no significance.

ALA-PDT induces changes in ferroptosis-related proteins in tongue tumors of nude mice. (A) Schematic diagram of in vivo experiment; (B) Body weight of nude mice was monitored to assess the condition of the nude mice; (C,D) Observed and measured tumor volume (n = 5); (E–J) Immunohistochemical staining and IHC scoring were performed for Ki67, GPX4 and FTH1 to evaluate protein expression (n = 5). (K) Perls’ Prussian blue staining was performed to evaluate iron deposition. Arrows indicate the deposition of trivalent iron. ns: no significance.

Full text

1. Introduction

The World Health Organization (WHO) defines oral leukoplakia (OLK) as follows: it occurs in the oral mucosa, which is characterized by the appearance of persistent white plaques or patches on the oral mucosa that cannot be scraped off and cannot be attributed to any other clinically or histopathologically definable condition [1], with a global incidence rate of 4.11% [2]. OLK should not be underestimated as a potential malignant oral disease. Age, gender, smoking and drinking habits, mechanical stimulation, systemic diseases and other factors are risk factors for the malignant transformation of OLK, and may even lead to further development of OLK into oral cancer. Oral squamous cell carcinoma (OSCC) accounts for 90% of oral cancers, which is the most common pathological type [3]. Notably, with the progression of oral cancer, it is often accompanied by pain and impaired speech and mastication, which severely compromises patients’ quality of life.

Traditional treatment methods of OLK include topical medications, surgery, laser therapy and cryotherapy. However, these treatments have some problems, for example, the curative effect is not significant or cannot effectively inhibit the malignant transformation of OLK. At the same time, for large-area OLK, skin grafting is often required when surgical resection is used, which will also change the patient’s oral anatomical structure and affect the oral function.

Currently, the treatment of oral cancer is mainly based on surgery, radiotherapy and chemotherapy. But surgery usually requires removal of normal tissue near the tumor, and sometimes even tooth extraction or jaw removal, which has a serious impact on appearance and function of patients. In addition, there are problems such as nerve injury and sensory abnormalities in patients. Radiotherapy and chemotherapy can cause related oral mucositis, osteomyelitis or osteonecrosis of the jaw, and may damage the salivary glands, resulting in dry mouth, rampant caries and other complications [4]. Therefore, more and more attention has been paid to the effective treatment methods that not only eliminate the lesion but also are minimally invasive and capable of improving patients’ quality of life.

Photodynamic therapy (PDT), a promising treatment method with minimal trauma, has been used for oral potentially malignant diseases and oral cancer [5,6]. PDT causes minimal trauma by selectively targeting malignant and precancerous cells while sparing surrounding normal tissues. Side effects are limited to localized erosion and swelling at the treatment site, with generally mild post-procedural pain. These adverse reactions typically disappear within two weeks [7], thereby preventing scarring, eliminating the need for skin or bone grafts, and preserving both anatomical integrity and physiological function without significant structural alteration. At present, PDT is a promising treatment used in the precancerous lesions and cancerous stages of oral cavity, skin, lung, gastrointestinal tract and urinary tract because of its minimally invasive nature [4,8]. At the same time, PDT can also be combined with other treatment methods to strengthen the treatment efficacy of cancer [9].

The mechanism of PDT mainly involves apoptosis, necrosis, autophagy, pyroptosis and immunogenic cell death [10,11]. Ferroptosis is a type of iron-dependent programmed death that has received attention in recent years, which is characterized by excessive accumulation of ferrous iron and lipid peroxidation in cells [12]. With the progress of research, it was gradually realized that the mechanism of cell death caused by PDT may also include ferroptosis. Previous studies have shown that photosensitizers can induce apoptosis and ferroptosis of glioma cells and sarcoma cells in mice after illumination [13]. And 5-aminolevulinic acid-mediated PDT (ALA-PDT) increased reactive oxygen species (ROS), promoted lipid peroxidation and induced ferroptosis in keloid fibroblasts, while inhibiting the expression of solute carrier family 7 member 11 (SLC7A11/xCT, a component of system xc^−^) and glutathione peroxidase 4 (GPX4) in cells [14]. Another study found that PDT with hypericin as photosensitizer can trigger ferroptosis-like cell death characterized by lipid peroxidation and mitochondrial Fe^2+^ accumulation in lung adenocarcinoma cells, both in vitro and in vivo [15].

However, there is relatively little research on whether PDT causes ferroptosis in OLK and OSCC treatments. Therefore, this study aims to explore the role of ferroptosis in ALA-PDT, in order to provide a new perspective for PDT treatment of OLK and OSCC.

2. Materials and Methods

2.1. Cell Culture

Human oral squamous cancer cell lines SCC9, HN6 and CAL27, as well as human oral leukoplakia cell line Leuk1, were obtained from the Central Laboratory of Peking University School of Stomatology. SCC9 cells were cultured in DMEM/F12 medium (Gibco, Grand Island, NY, USA), while CAL27 and HN6 cells were cultured in DMEM medium (Gibco, Grand Island, NY, USA). Leuk1 cells were cultured in RPMI medium 1640 (Gibco, Grand Island, NY, USA). Complete medium used to culture cells was supplemented with 10% fetal bovine serum (FBS; ABW, Xiamen, China) and 1% penicillin and streptomycin (Solarbio, Beijing, China). Cells were cultured under conditions of 5% CO_2_ at 37 °C.

2.2. PDT Irradiation

In this study, ALA was used as the precursor of the photosensitizer. Cells were first incubated with a specific concentration of ALA in the dark for 4 h, and then irradiation was performed with a semiconductor laser (LD600-C, Wuhan Yage Optoelectronics Technology Co., Ltd., Wuhan, China) with an output power of 150 mW and a wavelength of 635 nm. Energy flux (J/cm^2^) = optical power density (W/cm^2^) × illumination time (s). The light dose in this study was 10 J/cm^2^.

2.3. CCK8 Assay

To detect the effect of different concentrations of ALA (Fudan-Zhangjiang, Shanghai, China), ferrostatin-1 (Fer-1, Sigma, St. Louis, MO, USA) or erastin (MCE, Monmouth Junction, NJ, USA) on relative cellular dehydrogenase activity during PDT process, CCK8 assay was performed. Cells were seeded into 96-well plates and treated with the corresponding treatment group after cells were confluent to 70–80%. After 24 h of culture, CCK8 reagent (Beyotime, Shanghai, China) and culture medium were mixed at a ratio of 1:10 and added to each well. Subsequently, cells were incubated in 37 °C incubator for 2 h. A microplate reader (BioTek, Winooski, VT, USA) was used to determine the OD value at 450 nm. Relative cellular dehydrogenase activity = (OD_t__reatment group_/OD_control group_) × 100%.

2.4. Clonogenic Assay

Cells were treated differently according to their respective experimental groups. Then, clonogenic assay was conducted to evaluate long-term proliferation capacity. Cells were seeded at 600 cells per well in 6-well plates, and the culture medium was replaced every 2–3 days. SCC9 cells were stained with crystal violet after 10 days, while Leuk1 cells were stained after 14 days. Finally, the number of colonies was counted.

2.5. ROS Observation

To evaluate the primary ROS burst and sustained production of ROS in cells, cells were seeded into 6-well plates. After co-incubating cells with ALA under serum-free culture conditions for 4 h, cells were washed once with PBS, then DCFH-DA probe (Beyotime, Shanghai, China) was diluted with serum-free medium at 1:1000 to make the final concentration of 10 μM. A total of 1 mL diluent was added to each well. Cells were incubated at 37 °C in the dark for 20 min. Subsequently, irradiation was performed after washing cells twice with PBS. After irradiation, the cells were immediately (within 0.5 h) observed by a fluorescence microscope (Olympus, Tokyo, Japan) or cultured in complete culture medium containing 10% serum for 4 h before observing ROS levels. The control group and the treatment groups experienced the same serum conditions.

2.6. Evaluation of GSH Depletion

Cells were seeded into 60 mm cell culture dishes. After cells were confluent to 70–80%, PDT irradiation was performed. Treated cells were cultured for 24 h. The glutathione content was detected with GSH/GSSG kit (Beyotime, Shanghai, China) according to the instruction.

2.7. Transmission Electron Microscopy (TEM) for Mitochondrial Morphology Observation

SCC9 and Leuk1 cells were seeded in 60 mm culture dishes and allowed to reach 70–80% confluence before PDT irradiation. At 24 h post-PDT irradiation, the culture medium was removed and cells were fixed with 2.5% glutaraldehyde at room temperature for 5 min. Cells were then gently scraped using a cell scraper, collected by centrifugation to form cell pellets, and further fixed in the dark at room temperature for 30 min. The samples were subsequently stored at 4 °C until processing. Finally, mitochondrial morphology was observed by TEM.

2.8. Lipid Peroxidation Assessment

BODIPY 581/591 C11 probe (Beyotime, Shanghai, China) was used to evaluate lipid peroxides in cells. Cells were pre-treated with 1 μM Fer-1 24 h in advance. Then, ALA-PDT was performed. Six h later, 1 μL BODIPY 581/591 C11 was added to 1 mL PBS to prepare the working solution. Cells were washed once with PBS. Then, 1 mL working solution was added to each well. Cells were incubated at 37 °C in the dark for 30 min. Cells were washed twice with PBS. A fluorescence microscope was used to evaluate the lipid peroxidation.

2.9. Fe2+ Level Measurement

Cells were pre-treated with 1 μM Fer-1 for 24 h. At 6 h post-PDT, the intracellular Fe^2+^ content was detected using the Ferro Orange probe (Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions. Briefly, the collected cell suspension was incubated with 1 μM Ferro Orange probe at 37 °C under light-protected conditions for 30 min. Cells were washed once with PBS, then fluorescence intensity was measured by flow cytometry.

2.10. Western Blot

For the preparation of total cell lysates, both adherent and detached cells were collected. Total proteins were extracted from cells using RIPA lysis buffer (Solarbio, Beijing, China). Protein samples were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). After incubating with blocking solution (Shanghai Epizyme Biomedical Technology Co., Ltd., Shanghai, China) at room temperature for 1 h, the PVDF membranes were incubated with primary antibodies (anti-GPX4, anti-FTH1, anti-TFRC, anti-cleaved caspase-3, anti-phosphorylated MLKL, anti-MLKL and anti-β-actin) at 4 °C overnight with gentle shaking, followed by incubation with horseradish peroxidase–conjugated secondary antibodies at room temperature for 1 h. Antibodies used in this study were all purchased from ABclonal (Wuhan, China). Finally, protein bands were visualized by ECL substrate (NCM Biotech, Suzhou, China) using the eBlot chemiluminescence imaging system. Band intensity was quantified by densitometric analysis using ImageJ software (version 1.54), and the relative expression levels of target proteins were normalized to β-actin as the internal control.

2.11. Flow Cytometric Analysis of Apoptosis

To evaluate cell apoptosis, ANNEXIN V-FITC/PI staining (Elabscience, Wuhan, China) was used. SCC9 and Leuk1 cells were cultured in 6-well plates. At 24 h post-irradiation, both adherent and detached cells were collected. Pellets were washed twice with PBS. After resuspending the cells in 500 μL Binding Buffer, 5 μL ANNNEXIN V-FITC and 5 μL PI were added. After incubating on ice in the dark for 20 min, flow cytometry was used for detection.

2.12. In Vivo Study

The in vivo experiments have been approved by the Laboratory Animal Ethics Committee of Peking University School of Stomatology (Ethical Approval Number: BDKQ-202507070748). Ten 4–5-week-old male BalB/c nude mice purchased from Peking University Health Science Center were used for establishing the tongue OSCC model. The mice were housed in a specific pathogen-free barrier facility using individually ventilated caging systems. The ambient temperature was 24 ± 2 °C with a relative humidity of 50 ± 10%. The nude mice were provided with autoclaved feed, sterilized water and autoclaved bedding. Following a one-week acclimation period, the experiments were initiated. Both the cell suspension injection and the irradiation procedures were conducted under continuous isoflurane anesthesia separately. The duration of anesthesia for each procedure was approximately 5 min. SCC9 cells were digested with trypsin and resuspended in PBS. A 25 μL aliquot of cell suspension (containing 10^6^ cells) was administered to mice via intralingual injection to establish an orthotopic xenograft model. Approximately two weeks post-inoculation, tumors became clearly visible and PDT irradiation was performed. As previously reported [5], ten nude mice were randomly allocated into two groups: the control group and the PDT group. Each group contained 5 mice. Every mouse of the PDT group received an intralingual injection of 25 μL of 20% ALA (prepared in saline) followed by irradiation with a 635 nm semiconductor laser at an output power of 150 mW for 2 min. The control group received an equal volume of saline alone. Body weight and tumor development in the tongue were regularly monitored. On the 7th day, body weight loss approached the 20% threshold (a predetermined humane endpoint). Mice were euthanized promptly to minimize distress, and tissues were collected for subsequent analysis.

Tumor tissues of tongue were fixed in 4% paraformaldehyde for 24 h, followed by dehydration and embedding. Immunohistochemical staining of Ki67, GPX4 and FTH1 was conducted to evaluate the expression of proteins. Protein expression within tumor tissues was assessed using immunohistochemistry (IHC) scoring. The scoring system was defined as follows: the proportion of positive cells was scored as 1 point (0–25%), 2 points (26–50%), 3 points (51–75%) or 4 points (76–100%). Staining intensity was graded as 1 point (weakly positive), 2 points (moderately positive) or 3 points (strongly positive). The final IHC score for each sample was calculated as the product of the proportion score and the intensity score. Perls’ Prussian blue staining was performed to evaluate iron deposition. Five samples of each experimental group were used for analysis.

2.13. Mitochondrial ROS Detection

Cells were seeded in 6-well plates and cultured until reaching 70–80% confluence, followed by different treatment. For mitochondrial superoxide detection, at 6 h post-PDT irradiation, cells were incubated with 1 μM MitoSOX Red (MCE, Monmouth Junction, NJ, USA) under light-protected conditions for 30 min, then cells were washed twice with serum-free medium. For mitochondrial H_2_O_2_ detection, cells were incubated with 0.1 μM MitoPeDPP (Dojindo, Kumamoto, Japan) at 37 °C under light-protected conditions for 15 min. Cells were washed twice with PBS. Nuclei were stained with DAPI (Solarbio, Beijing, China) and then examined by fluorescence microscopy.

2.14. Mitochondrial Membrane Potential Assay

Within 0.5 h after irradiation and at 6 h after irradiation, cells were incubated with TMRE working solution (Beyotime, Shanghai, China) at 37 °C for 30 min. After removing the working solution and replacing it with fresh culture medium, cells were observed using a fluorescence microscope.

2.15. Statistical Analysis

All data were statistically analyzed and visualized using GraphPad Prism version 10.0. Results are expressed as mean ± standard deviation from three independent experiments. Differences between the two groups were assessed by Student’s t-test, while comparisons among multiple groups were performed using one-way analysis of variance (ANOVA). A p-value of less than 0.05 was considered statistically significant.

3. Results

3.1. ALA-PDT-Induced Ferroptosis in OLK and OSCC Cells

Initially, three OSCC cell lines (SCC9, HN6 and CAL27) and an OLK cell line (Leuk1) were treated with PDT after incubation with varying concentrations of ALA. The results (Figure S1) demonstrated that, for OSCC cell lines, 0.5 mM ALA could reduce relative cellular dehydrogenase activity to approximately 50% following irradiation. However, relative cellular dehydrogenase activity of OSCC cell lines showed no significant further reduction with increasing ALA concentrations until reaching 10 mM. In contrast, in the OLK cell line (Leuk1), a concentration of 1 mM ALA was required to reduce the relative cellular dehydrogenase activity by approximately 50%. Notably, at 10 mM ALA, PDT irradiation paradoxically increased relative cellular dehydrogenase activity of Leuk1. Microscopic images at 24 h post-PDT mediated by different concentrations of ALA showed that, under the 10 mM ALA condition, Leuk1 cells exhibited fewer rounded cells and better overall cellular morphology (Figure S2). Furthermore, clonogenic assay was performed to evaluate long-term proliferative ability (Figure 1A). Results showed that, after PDT irradiation mediated by 1 mM, 2 mM and 4 mM ALA, the colony numbers of SCC9 cells and Leuk1 cells significantly decreased. A total of 10 mM ALA further reduced the colony numbers of SCC9, while the colony numbers of Leuk1 in the 10 mM group increased compared to the 1 mM, 2 mM and 4 mM groups. Therefore, 1 mM ALA was selected for subsequent experiments as it represents the lowest concentration within the maximal efficacy plateau, ensuring a robust PDT effect while minimizing nonspecific complications associated with supra-optimal doses.

To evaluate the primary ROS burst and sustained oxidative stress, a DCFH-DA probe was used to observe ROS levels both within 0.5 h and at 4 h after PDT irradiation. A strong and significant increase in ROS was detected immediately in SCC9 and Leuk1 cells after irradiation, representing the initial ROS burst. At 4 h post-PDT irradiation, the ROS signal decreased compared to 0.5 h, but still increased compared to the control group, indicating the persistence of secondary ROS and oxidative stress (Figure S3). Meanwhile, ROS signals can be detected in four types of cells at 4 h post-PDT irradiation (Figure 1B). TMRE staining was used to assess mitochondrial membrane potential. The results showed a decreasing trend in mitochondrial membrane potential in both SCC9 and Leuk1 cells after irradiation (Figure S4).

Furthermore, glutathione content assay revealed a significant decrease in the GSH/GSSG ratio in cells after ALA-PDT (Figure 1C). TEM was conducted to examine the mitochondrial morphology of SCC9 and Leuk1 cells before and after ALA-PDT. The images demonstrated that, after ALA-PDT, mitochondria exhibited typical characteristics of ferroptosis, namely, reduced mitochondrial volume, condensed matrix, increased membrane density and decreased mitochondrial cristae. However, these phenomena were not observed in the control group (Figure 1D). This suggests that ALA-PDT induces ferroptosis in OSCC and OLK cells.

ALA-PDT-induced ferroptosis in OLK and OSCC cells. (A) After co-incubating cells with different concentrations of ALA for 4 h, ALA-PDT was performed. To evaluate the long-term proliferative viability, clonogenic assay was performed (n = 3). (B) Cells were incubated with 1 mM ALA for 4 h in serum-free medium. Cells were then loaded with the DCFH-DA probe for 20 min and subjected to PDT irradiation. After irradiation, cells were cultured in complete medium. Intracellular ROS levels were observed under a fluorescence microscope at 4 h post-irradiation. Green: DCFH-DA, ROS. (C) Evaluated the cellular GSH/GSSG content 24 h after ALA-PDT (n = 3). (D) 24 h after ALA-PDT, the mitochondrial morphology of SCC9 and Leuk1 cells was observed by TEM. Red arrow: mitochondria with typical ferroptosis morphology. a: there is a statistical difference compared to the control group; b: there is a statistical difference compared to 0 mM ALA mediated PDT. ns: no significance.

Subsequently, ferroptosis-related proteins GPX4, FTH1 and TFRC were evaluated by Western blot at 6 h and 24 h after ALA-PDT. The results showed that, 6 h after PDT irradiation, the expression of GPX4 and FTH1 in four types of cells showed a decreasing trend, while the expression of TFRC increased. This protein expression pattern persisted at 24 h after PDT (Figure 2).

The above results indicate that ALA-PDT causes ferroptosis by affecting intracellular ROS and disrupting iron homeostasis in OLK and OSCC cells.

3.2. ALA-PDT Caused Changes in Ferroptosis-Related Proteins in Nude Mouse OSCC Model

To investigate the effect of ALA-PDT on ferroptosis-related proteins in vivo, we established an OSCC model by injecting SCC9 cell suspension into the tongues of nude mice and performed ALA-PDT. Tumor tissues were collected and analyzed for various indicators on day 7 after PDT (Figure 3A). During the modeling and treatment process, the body weight of the nude mice was monitored. The results indicated that the body weight of nude mice in both groups continued to decrease after PDT; however, no statistically significant difference was observed between the two groups. (Figure 3B). Measurement of tumor volume revealed that the tumor volume in the PDT group was smaller than that in control group, but the difference was not statistically significant (Figure 3C,D).

Subsequently, Ki67 immunohistochemical staining was performed to evaluate the cell proliferation ability in tumor tissues. The results showed that the IHC score of Ki67 in the PDT group was lower than that in control group (Figure 3E,F), indicating a slower tumor cell proliferation rate in the PDT group. Immunohistochemical staining of ferroptosis-related proteins GPX4 and FTH1 revealed decreased expression of both proteins in the PDT group, with lower IHC scores compared to the control group (Figure 3G–J). In addition, Perls’ Prussian blue staining showed that the ALA-PDT group has more iron deposition (Figure 3K). These results suggest the occurrence of ferroptosis in tumor tissues.

ALA-PDT induces changes in ferroptosis-related proteins in tongue tumors of nude mice. (A) Schematic diagram of in vivo experiment; (B) Body weight of nude mice was monitored to assess the condition of the nude mice; (C,D) Observed and measured tumor volume (n = 5); (E–J) Immunohistochemical staining and IHC scoring were performed for Ki67, GPX4 and FTH1 to evaluate protein expression (n = 5). (K) Perls’ Prussian blue staining was performed to evaluate iron deposition. Arrows indicate the deposition of trivalent iron. ns: no significance.

3.3. Fer-1 Partially Reversed the Effects of ALA-PDT on OLK and OSCC

To further validate the role of ferroptosis in ALA-PDT, four types of cells were treated with the ferroptosis inhibitor Fer-1. Firstly, to determine the concentration of Fer-1 used in this study, cells were pre-treated with 1 μM, 2 μM and 4 μM Fer-1 for 2 h, followed by ALA-PDT. Subsequently, relative cellular dehydrogenase activity was analyzed. Results indicated that 1 μM Fer-1 could partially reverse the relative cellular dehydrogenase activity of four types of cells after ALA-PDT, while increasing the concentration to 2 μM and 4 μM did not significantly improve relative cellular dehydrogenase activity (Figure 4A). Consequently, 1 μM Fer-1 was chosen for following experiments.

To observe ROS levels, a DCFH-DA probe was used. Within 0.5 h after irradiation, the ROS levels in both the PDT-alone group and the Fer-1 pre-treatment group increased significantly, with no significant difference between the two groups (Figure S5). To evaluate the sustained production of ROS in cells, observations were conducted again 4 h after PDT. The images showed that pre-treatment with Fer-1 followed by ALA-PDT resulted in reduced ROS levels compared to the PDT-alone group (Figure 4B). The intracellular Fe^2+^ content was evaluated by Ferro Orange assay, which showed that ALA-PDT could significantly increase the intracellular Fe^2+^ content, while Fer-1 could decrease it to some extent. But a statistically significant difference was only observed in Leuk1 cells (Figure 4C,D). C11 BODIPY 581/591 probe was used to detect cellular lipid peroxidation after ALA-PDT. The results showed that, after ALA-PDT, the red fluorescence representing a reduced state in cells decreased, while the green fluorescence representing the oxidized state increased, indicating an increase in cellular lipid peroxidation. This phenomenon could be partially reversed through pre-treating cells with Fer-1 (Figure 4E–H).

Analysis of ferroptosis-related proteins showed that, compared to the PDT group, pre-treatment with Fer-1 partially reversed the effects of ALA-PDT by increasing the expression of GPX4 and FTH1 and decreasing that of TFRC (Figure 5).

Based on these results, Fer-1 can reverse ferroptosis caused by ALA-PDT.

Furthermore, to investigate the role of other cell death modalities such as apoptosis and necrosis in this PDT condition, apoptosis detection was first performed. Results showed that ALA-PDT could cause apoptosis in SCC9 and Leuk1 cells. Interestingly, in SCC9 cells, the use of Fer-1 could reduce apoptotic cells, whereas, in Leuk1 cells, Fer-1 had no effect on apoptosis (Figure 6A–D). Western blot results showed that the expression of the apoptotic protein-cleaved caspase-3 increased in SCC9 and Leuk1 cells after PDT treatment, while Fer-1 could only reverse the expression of cleaved caspase-3 in SCC9 cells and had no effect on Leuk1 cells. The necrosis marker pMLKL showed no significant change after PDT (Figure 6E–H).

3.4. ALA-PDT Caused Sustained Mitochondrial Dysfunction in OLK and OSCC Cells

A ferroptosis agonist, erastin, was added to pre-treat SCC9 and Leuk1 cells for 48 h. Then, the IC50 of erastin was detected on both cells. Results showed that erastin had an IC50 of 4.814 μM for SCC9 cells and 23.55 μM for Leuk1 cells (Figure 7A,B). Subsequently, the effect of ALA-PDT combined with erastin on SCC9 cells and Leuk1 cells was detected by clonogenic assay. The results demonstrated that both PDT irradiation and erastin treatment independently decreased the clonogenic survival of SCC9 and Leuk1 cells. Furthermore, the combination of PDT and erastin resulted in a synergistic reduction in colony formation (Figure S6). Additionally, CCK8 assay was conducted 24 h after PDT irradiation. The results suggested that the combination of PDT and erastin led to a modest but statistically significant decrease in relative cellular dehydrogenase activity (Figure 7C,D).

Mitochondrial superoxide production was detected using MitoSOX Red at 6 h post-irradiation. The results showed that ALA-PDT enhanced red fluorescence in SCC9 cells, indicating increased mitochondrial superoxide, which could be reversed by Fer-1. Treatment with erastin alone also intensified red fluorescence in SCC9 cells, while the combination of PDT and erastin further enhanced red fluorescence in SCC9 cells (Figure 7E,H). Interestingly, in Leuk1 cells, neither PDT, Fer-1 nor erastin alone induced significant changes in red fluorescence, suggesting no substantial alteration in mitochondrial superoxide production. However, the combination of PDT and erastin significantly increased red fluorescence in Leuk1 cells (Figure 7F,H).

MitoPeDPP was used to detect the production of mitochondrial H_2_O_2_ at 6 h post-irradiation. The images presented that the green fluorescence of SCC9 cells was significantly enhanced after ALA-PDT, which elucidated an increase in mitochondrial H_2_O_2_. The green fluorescence was attenuated after Fer-1 pre-treatment. Erastin alone also markedly increased green fluorescence in cells; however, its combination with PDT did not further enhance the green fluorescence (Figure 7E,H). In Leuk1 cells, ALA-PDT slightly increased green fluorescence, but Fer-1 treatment did not reduce it. Erastin alone elevated green fluorescence in Leuk1 cells and, similarly to SCC9 cells, the combination of PDT and erastin did not further enhance green fluorescence in Leuk1 cells (Figure 7F,H).

To detect changes in mitochondrial membrane potential, TMRE staining was performed 6 h after irradiation. The results indicated that the red fluorescence in SCC9 and Leuk1 cells weakened after PDT, reflecting a decrease in mitochondrial membrane potential. Erastin alone also slightly reduced the mitochondrial membrane potential. Moreover, after the combination of PDT and erastin, the red fluorescence in SCC9 and Leuk1 cells almost disappeared, indicating a significant decrease in mitochondrial membrane potential (Figure 7I–K).

4. Discussion

OLK is among the most prevalent oral potentially malignant disorders and possesses a significant risk of malignant progression to OSCC if left untreated. Beyond conventional therapeutic strategies, PDT has recently emerged as a promising minimally invasive modality for the management of OLK and OSCC. Ferroptosis, an iron-dependent regulated cell death pathway, is primarily characterized by the accumulation of lipid peroxides and intracellular iron overload [12,16]. Although the mechanism of PDT has been increasingly investigated, the potential involvement of ferroptosis in PDT-mediated treatment of OLK and OSCC remains poorly understood. Our study provides evidence that ALA-PDT induces lipid peroxide accumulation, elevates intracellular Fe^2+^ levels, promotes mitochondrial ROS generation and ultimately triggers ferroptosis in OLK and OSCC cells. These findings not only elucidate a novel mechanistic basis for ALA-PDT in treating OLK and OSCC but also suggest potential strategies for combination therapies involving ALA-PDT.

4.1. Manifestations of ALA-PDT-Induced Ferroptosis in OLK and OSCC Cells

Ferroptosis is an emerging therapeutic target and prognostic marker in OLK and OSCC. By regulating iron metabolism, lipid peroxidation and the antioxidant system, ferroptosis can be induced to promote cell death and enhance the efficacy of existing treatments [17,18].

Numerous studies have confirmed that ALA-PDT can increase intracellular ROS generation. In addition to the immediate ROS burst, PDT triggers a cascade of sustained oxidative damage, with significant contributions from mitochondrial dysfunction and endoplasmic reticulum stress [19,20,21]. The sustained accumulation of ROS detected at 24 h post-PDT in previous studies provide evidence for persistent oxidative damage [21,22,23]. The findings of our study are consistent with those reports. We also observed ROS signals 4 h after ALA-PDT irradiation, which was consistent with those reports.

Glutathione, a key ROS scavenger, is ubiquitously present in cells and participates in various antioxidant responses, thereby regulating redox homeostasis [24,25]. Glutathione exists in two forms: reduced glutathione (GSH) and oxidized glutathione (GSSG), with GSH being the predominant form under normal physiological conditions [26]. During ferroptosis, cells exhibit significant depletion of GSH [27]. Our study found that, after ALA-PDT irradiation, the GSH/GSSG ratio significantly decreased in both OLK and OSCC cells (Figure 1C), indicating a reduction in GSH content and a disruption of cellular redox homeostasis.

Alterations in mitochondrial morphology can be seen as a hallmark of ferroptosis, characterized by reduced mitochondrial volume, loss or diminishment of cristae and increased mitochondrial membrane density [12,28]. Our research confirmed that Leuk1 and SCC9 cells exhibit these morphological changes following ALA-PDT (Figure 1D). Taken together, these results suggest that ALA-PDT may induce ferroptosis in OLK and OSCC cells.

Fer-1, a specific ferroptosis inhibitor, effectively suppresses iron-dependent lipid peroxidation through multiple mechanisms, including scavenging alkoxyl radicals, chelating Fe^2+^ and reducing the intracellular labile iron pool, thereby blocking the ferroptosis process [29]. In the present study, pre-treatment with Fer-1 partially rescued cell damage caused by ALA-PDT (Figure 4). Firstly, our study found that Fer-1 had no obvious effect on the instantaneous ROS burst but could reduce the generation of secondary ROS (Figures S5 and Figure 4B), which may be the reason why Fer-1 partially reverses the subsequent effects of ALA-PDT on cells. In addition, the assessment of Fe^2+^ content revealed that ALA-PDT elevated intracellular Fe^2+^ in both OLK and OSCC cells, which could be partially reversed by Fer-1 (Figure 4C,D). Furthermore, the evaluation of lipid peroxidation using C11 BODIPY 581/591 demonstrated that ALA-PDT promoted lipid peroxide accumulation, which was also suppressed upon Fer-1 treatment (Figure 4E–H). Collectively, these results indicate that Fer-1 counteracts the cellular effects of ALA-PDT in OLK and OSCC cells, further supporting the conclusion that ALA-PDT induces ferroptosis in these cell types.

Notably, an interesting phenomenon was observed in Leuk1 cells. Compared to 1 mM, 2 mM and 4 mM ALA, the killing effect of PDT mediated by 10 mM ALA on Leuk1 is reduced (Figure 1A and Figure S1). This phenomenon may be caused by the formation of high concentrations of PpIX aggregates or unmetabolized ALA outside the cell or on the cell membrane surface, leading to the inner filter effect. This effect prevents light from truly reaching the interior of the cell, thereby reducing cellular damage [30,31]. Additionally, the distribution of photosensitizers within cells can also affect the efficiency of PDT. For example, studies have found that, when the concentration of the photosensitizer AlPcS_2_ exceeds 1 μM, it aggregates within the cell and becomes more diffusely localized, resulting in reduced damage to the glioma cell line BMG-1 by PDT [32]. Furthermore, this phenomenon may also be attributed to the generation of a large amount of ROS by 10 mM ALA within a very short period of time after irradiation, which rapidly consumes oxygen, leading to cellular hypoxia and weakening the cytotoxic effect on cells. Dysart et al. confirmed that, under normal oxygenation conditions, ALA-PDT can effectively kill cells; however, under hypoxic conditions, the cytotoxic effect of ALA-PDT is significantly reduced [33]. The differential responses of Leuk1 cells and OSCC cell lines to varying ALA concentrations may be attributed to inherent metabolic differences between OLK cells and OSCC cells. Leuk1 cells may exhibit heightened sensitivity in photosensitizer uptake and metabolic pathways to high photosensitizer concentrations, readily triggering defense mechanisms and activating powerful endogenous protective mechanisms such as hormesis or cytoprotective autophagy [34,35]. In contrast, OSCC cells possess a more robust metabolism, enabling rapid conversion of ALA to PpIX with less pronounced photobleaching effects.

4.2. Changes in Ferroptosis-Associated Protein Expression Induced by ALA-PDT

GPX4, a selenoprotein, functions to scavenge intracellular peroxides and maintain cell survival. Loss of GPX4 activity leads to uncontrolled lipid peroxidation, which makes it an important regulator of ferroptosis [36]. FTH1 and TFRC are key molecules in iron metabolism regulation and directly participate in ferroptosis by modulating intracellular iron homeostasis. FTH1 is responsible for storing intracellular labile iron, thereby suppressing iron-dependent oxidative damage, such as the Fenton reaction. Degradation of FTH1 enhances cellular susceptibility to ferroptosis [37]. In contrast, TFRC, as a transport protein, facilitates iron uptake and promotes ferroptosis by expanding the intracellular active iron pool [38]. Our in vitro studies demonstrated that ALA-PDT downregulated GPX4 and FTH1 expression while upregulating TFRC in both OLK and OSCC cells (Figure 2). These alterations were reversed by Fer-1 treatment (Figure 5).

Furthermore, to simulate the clinical process of PDT for OSCC, we utilized an orthotopic tongue xenograft model, as the tongue is a common site of OSCC [39]. The established tumors were then subjected to PDT irradiation, after which the changes in the key indicators were monitored. Ki67 expression is correlated with the proliferative activity of tumor cells [40]. We observed decreased Ki67 expression following ALA-PDT in a nude mouse model of OSCC (Figure 3E,F), indicating reduced tumor cell proliferative capacity. The lack of statistically significant tumor volume reduction after ALA-PDT in our study (Figure 3D) can be explained by the constrained experimental timeline. The relatively short treatment and observation period were likely insufficient for the therapeutic effects to fully manifest as significant macroscopic regression. The immunohistochemical staining of ferroptosis-associated proteins confirmed that ALA-PDT reduced GPX4 and FTH1 expression in tumor tissues (Figure 3G–J), suggesting the occurrence of ferroptosis in vivo. The decrease in FTH1 expression implies increased iron release, which may amplify the Fenton reaction and promote lipid peroxide generation. However, weight loss could still be observed in both control and treatment groups as a consequence of tumor growth within the oral cavity, which inevitably affected feeding and swallowing in mice, finally leading to weight reduction.

These findings provide a mechanistic explanation for the marked increase in lipid peroxidation and elevated Fe^2+^ levels observed in OLK and OSCC cells after ALA-PDT.

Notably, the regulatory effects of ALA-PDT on GPX4 and FTH1 were less pronounced in Leuk1 cells compared to OSCC cells. This differential response may be attributed to lower ferroptosis sensitivity in OLK cells or could indicate that ALA-PDT induces ferroptosis in Leuk1 cells primarily through alterations in other genes. The precise underlying mechanisms, however, require further investigation.

4.3. ALA-PDT Induces Sustained Mitochondrial Dysfunction in OLK and OSCC Cells

Mitochondria play a crucial role in ferroptosis. As central hubs for cellular iron metabolism, mitochondria facilitate iron overload-induced ROS generation via the Fenton reaction, thereby initiating lipid peroxidation [41]. Previous studies have shown that the photosensitizer ALA exhibits mitochondrial targeting properties, promoting the synthesis of mitochondrial PpIX and subsequent substantial ROS production [42]. Furthermore, impairment of the mitochondrial electron transport chain can exacerbate ROS generation and accelerate lipid peroxidation [43]. In this study, we observed that ALA-PDT increased both mitochondrial superoxide and H_2_O_2_ levels in SCC9 cells (Figure 7E,G). In Leuk1 cells, however, ALA-PDT elevated mitochondrial H_2_O_2_ without significantly altering superoxide levels. A previous study showed that a decrease in mitochondrial membrane potential was observed 6 h after ALA-PDT [44]. Another study showed that, at 6 h post-MB-PDT treatment, mitochondrial superoxide significantly increased and mitochondrial membrane potential significantly decreased [45]. In fact, at several hours post-PDT, the detected mitochondrial ROS are no longer the initial photochemical products, such as singlet oxygen. Instead, they constitute a self-amplifying oxidative stress signaling network, sustained by the damaged mitochondria themselves. This network includes O_2_•^−^ originating from electron leakage due to persistently dysfunctional electron transport chains, H_2_O_2_ derived from superoxide and •OH generated via the Fenton reaction. These species collectively lead to exacerbate mitochondrial damage and further delayed cell death, forming a vicious cycle of escalating deterioration [46,47].

Moreover, combined treatment with ALA-PDT and erastin further enhanced mitochondrial superoxide production and decreased mitochondrial membrane potential. These findings suggest that both ALA-PDT and erastin disrupt mitochondrial physiological function, promote lipid peroxide accumulation and exacerbate ferroptosis. However, the synergistic effect of erastin on ALA-PDT seems limited, which may be due to the fact that both require the consumption of GSH. Our in vitro data with erastin suggest the potential for ferroptosis agonists to synergize with PDT, but this requires further validation, particularly in in vivo models. Although we lack data from in vivo experiments, a previous study by Yang et al. confirmed that erastin enhances the therapeutic efficacy of ALA-PDT in a 4-nitroquinoline-1-oxide-induced oral leukoplakia mouse model [18]. In addition, a study by Zhu et al. demonstrated that the combination of erastin with a Ce6-PDT regimen significantly enhanced antitumor efficacy in an OSCC xenograft model [48]. Therefore, the combination of ALA-PDT with erastin may represent a promising strategy to improve therapeutic outcomes for refractory OLK and OSCC. Additionally, previous research has demonstrated that the downregulation of GPX4, particularly its mitochondrial isoform, leads to uncontrolled mitochondrial lipid peroxidation [49], which corroborates our observations of decreased GPX4 expression and increased mitochondrial ROS following ALA-PDT irradiation.

4.4. Mixed Cell Death Mechanisms Can Be Observed in ALA-PDT

ALA-PDT exerts its therapeutic effects through multiple cell death modalities. The accumulation of ROS and subsequent oxidative stress within cells can impact various signaling pathways that regulate survival, ultimately leading to apoptosis, necrosis, autophagy and ferroptosis [46]. Therefore, these cell death pathways interact with each other.

Based on our findings, we propose that the sequence of events leading to ferroptosis upon ALA-PDT may be as follows. Initially, ALA-PDT irradiation generates a burst of ROS that directly damages multiple cellular components. Subsequently, the primary ROS cause immediate and sustained damage to mitochondria, contributing to loss of mitochondrial membrane potential and mitochondrial dysfunction. This damage also causes persistent secondary ROS production and lipid peroxidation. In parallel, cellular iron homeostasis is disrupted, resulting in the accumulation of Fe^2+^. The convergence of persistent lipid peroxidation and iron overload ultimately triggers ferroptosis.

In this study, in addition to ferroptosis, ALA-PDT-induced apoptosis was also observed. A noteworthy finding is that the effect of ferroptosis inhibitor Fer-1 on apoptosis exhibits cell-type specificity (Figure 6). In SCC9 cells, Fer-1 treatment significantly reduced the number of apoptotic cells after ALA-PDT, indicating that ferroptosis may act as an upstream trigger signal for apoptosis. A previous study found that erastin-induced ferroptosis in neuronal cell HT-22 is not independent of mitochondrial damage, but activates the pro-apoptotic protein BID, causing the mitochondrial outer membrane permeabilization and the release of pro-apoptotic factors such as cytochrome c, thereby amplifying apoptotic signals [50]. In contrast, in Leuk1 cells, Fer-1 failed to reduce the apoptosis level, indicating that ALA-PDT-induced apoptosis and ferroptosis are independent and parallel processes. This difference may be due to inherent differences in antioxidant capacity, iron metabolism homeostasis, mitochondrial function or death receptor signaling between the two types of cells. This result emphasizes the importance of considering cell background specificity when evaluating PDT mechanisms, and provides a theoretical basis for personalized PDT strategies targeting different oral precancerous lesions and malignant tumors in the future.

In order to achieve good therapeutic effects in clinical practice, the ALA-PDT conditions used are harsher. The PDT parameters used in this study are milder compared to clinical practice. Its core value lies in revealing a biological scenario that may be masked by high-intensity treatment but has important clinical significance: simulating the ‘sublethal’ or ‘suboptimal treatment area’ caused by tissue depth, heterogeneous light penetration or thickened keratinous barriers in clinical ALA-PDT. In these regions, the level of oxidative stress imposed on cells may be insufficient to cause immediate collapse but could precisely reach the threshold required to initiate regulated cell death programs, such as ferroptosis. This unveils a critical backup pathway that can be activated under non-lethal stress. Delayed death processes like ferroptosis may still contribute to partial lesion resolution. However, under clinically relevant high-intensity PDT conditions, the role of ferroptosis needs further validation.

4.5. Limitations of the Study

This study provides preliminary evidence that ALA-PDT can induce ferroptosis under relatively mild conditions, these parameters differ from the more intensive settings typically used in clinical practice. The role of ferroptosis under conditions closer to clinical reality, as well as whether ferroptosis agonists can definitively enhance the efficacy of ALA-PDT, requires further investigation. Additionally, while our research to some extent confirms that ALA-PDT triggers a hybrid form of cell death, the execution patterns may differ between OSCC and OLK cells, which also requires further research in the future.

5. Conclusions

Our research has found that the induction of ferroptosis is one of the mechanisms underlying the cytotoxicity of ALA-PDT, which works in concert with other established pathways such as apoptosis. In vitro studies suggest that the application of ferroptosis agonists may enhance the therapeutic efficacy of ALA-PDT against OLK and OSCC.