4-Methoxydalbergione Induces Dual Activation of Apoptosis and Autophagy-Dependent Cell Death via ROS–MAPK Signaling in Human Neuroblastoma Cells

Abstract

Neuroblastoma, the predominant extracranial solid malignancy in the pediatric population, remains a major clinical challenge due to pronounced intratumoral heterogeneity and intrinsic therapeutic resistance. 4-Methoxydalbergione (4-MD), a benzoquinone derivative isolated from Dalbergia odorifera, has demonstrated anticancer activity in several tumor models; however, its effects and underlying cell death mechanisms in neuroblastoma remain unclear. Here, we investigated the cytotoxic effects of 4-MD in human neuroblastoma cells using cell viability assays, flow cytometry, immunoblotting, and fluorescence microscopy. 4-MD reduced cell viability in a dose- and time-dependent manner and induced caspase-3 cleavage accompanied by MAPK activation, indicating apoptotic cell death. Concurrently, 4-MD promoted autophagosome accumulation, as evidenced by LC3-II accumulation, acidic vesicular organelle formation, ATG5 upregulation, and p62 degradation, in association with activation of the AMPK/mTOR/ULK1 signaling axis. Pharmacological inhibition of autophagy significantly attenuated 4-MD-induced cytotoxicity without affecting caspase-3 activation, demonstrating a caspase-independent, pro-death role of autophagy. Reactive oxygen species (ROS) acted as a critical upstream mediator, as antioxidant treatment suppressed both apoptotic and autophagic signaling. Moreover, inhibition of Na⁺,K⁺-ATPase with ouabain selectively reduced autophagy-dependent cell death, implicating autosis as an additional mechanism. Notably, 4-MD exhibited minimal toxicity toward primary cortical neurons. Collectively, these findings demonstrate that 4-MD engages multiple, non-redundant cell death pathways through coordinated ROS–MAPK–AMPK/mTOR/ULK1 signaling, highlighting its potential to overcome therapeutic resistance in heterogeneous neuroblastoma cells.

1. Introduction

Neuroblastoma is the most common extracranial solid tumor in children, typically diagnosed during infancy or early childhood [1,2]. As an embryonal neuroendocrine neoplasm arising from neural crest progenitors, it arises anywhere along the sympathetic nervous system, with the adrenal medulla being the most frequent site. At the time of diagnosis, many tumors already exhibit metastatic spread, reflecting both their aggressive biology and clinical heterogeneity.

Although neuroblastoma represents approximately 6–10% of pediatric cancers, it is responsible for 12–15% of childhood cancer-related mortality, highlighting its high lethality. Advances in molecular characterization—particularly the identification of prognostic markers such as MYCN amplification, segmental chromosomal aberrations, and ALK mutations—have facilitated more precise risk stratification and the development of targeted therapeutic approaches [3,4]. Current multimodal treatments for high-risk cases include surgery, chemotherapy, radiotherapy, autologous stem cell transplantation, differentiation therapy (e.g., retinoids), and immunotherapy (e.g., anti-GD2 monoclonal antibodies). Nevertheless, outcomes for patients with high-risk or relapsed disease remain poor, primarily due to intrinsic tumor heterogeneity and the acquisition of drug resistance. These limitations highlight the urgent need for novel therapeutic agents with broad and durable efficacy against neuroblastoma.

Therapeutic failure in high-risk neuroblastoma is frequently linked to impaired apoptotic responses and adaptive survival mechanisms [3,4]. Apoptosis is primarily mediated through mitochondrial dysfunction, leading to cytochrome c release and subsequent caspase activation, often downstream of stress-responsive MAPK signaling cascades, including ERK, p38, and JNK [5,6]. In parallel, autophagy is primarily regulated by the AMPK–mTOR–ULK1 signaling axis, which integrates cellular energy status and stress responses. AMPK is activated by phosphorylation under metabolic or oxidative stress and functions as a negative regulator of mTOR complex 1 (mTORC1) [7]. Activated AMPK suppresses mTORC1 activity both indirectly via TSC2 [8] and directly through phosphorylation of Raptor, thereby reducing its kinase activity [9]. Functionally, inhibition of mTORC1 is reflected by decreased phosphorylation of its downstream substrate S6K [10]. Suppression of mTORC1 relieves its inhibitory phosphorylation of ULK1, enabling activation of the ULK1 complex and initiation of autophagosome formation [11]. This process engages core autophagy machinery, including ATG5, which is essential for autophagosome elongation [12]. Dysregulation of these interconnected apoptotic and autophagic pathways may confer survival advantages to neuroblastoma cells, thereby contributing to therapeutic resistance.

4-Methoxydalbergione (4-MD; 2-[1-(4-hydroxyphenyl)prop-2-enyl]-5-methoxycyclohexa-2,5-diene-1,4-dione) is a benzoquinone-based natural product isolated from the heartwood of Dalbergia odorifera, a medicinal plant traditionally used in Korea and China for cardiovascular and anti-inflammatory disorders [13]. Previous studies have reported diverse biological activities of 4-MD, including antioxidant, antimicrobial, anti-inflammatory, and antitumor effects [13,14,15,16,17,18,19,20]. For example, 4-MD suppresses lipopolysaccharide (LPS)-induced inflammatory responses in macrophages through inhibition of the NF-κB signaling pathway [14] and reduces IL-33-induced cytokine production in mast cells [15].

Recent research has also revealed potent anticancer effects of 4-MD in multiple tumor types, including human bladder cancer [16], astroglioma [17], esophageal cancer [18], hepatocellular carcinoma [19], and non-small-cell lung cancer [20]. In hepatocellular carcinoma, 4-MD induces apoptosis by upregulating GADD45G expression [19], while in A549 lung cancer cells and xenograft models, it promotes ferroptosis via inhibition of DNMT1 [20]. Despite these promising findings, the potential anticancer effects of 4-MD in neuroblastoma have not yet been explored. Therefore, in this study, we investigated the cytotoxic activity of 4-MD against human neuroblastoma SH-SY5Y cells and elucidated the underlying cellular and molecular mechanisms.

2. Materials and Methods

2.1. Materials and Chemicals

As previously described [21], 4-MD was isolated from the dried heartwood of D. odorifera. The chemical identity and purity of the isolated 4-MD were confirmed by ¹H NMR spectroscopy and HPLC, and the purity was determined to be 96% (Supplementary Figures S1 and S2 and Table S1). A stock solution of 4-MD was prepared in dimethyl sulfoxide (DMSO, Sigma, St. Louis, MO, USA) and stored at −20 °C. The working solution was prepared by diluting the stock solution in serum-free medium. Dulbecco’s Modified Eagle’s Medium (DMEM) and all medium supplements were purchased from Gibco BRL (Grand Island, NY, USA). 3-(4, 5–dimethylthiazol–2–yl)-2,5-diphenyl tetrazolium bromide (MTT, #298931) was purchased from Duchefa Biochemie (Haarlem, The Netherlands). Acridine Orange (AO, #A6014), N-acetyl L-cysteine (NAC, #A7250), L-N^G-monomethyl arginine citrate (L-NMMA, #M7033), 3-methyladenine (3-MA, #M9281), bafilomycin A1 (#B1793), PD98059 (#P215), SB203580 (#S8307), SP600125 (#S5567), sulfanilamide (#S925), N-(1-Naphthyl)ethylenediamine dihydrochloride (NEDD, #222488), and 2′-7′-dichlorodihydrofluorescein diacetate (DCFDA, #D6883) were bought from Sigma-Aldrich (St.Louis, MO, USA). Caspase-inhibitor Z-VAD-FMK (#627610) was purchased from Calbiochem (San Diego, CA, USA). (4,6-diamidino-2-phenylindole (DAPI, #112002) was purchased from Cell Biolabs Inc. (San Diego, CA, USA). The LDH assay kit (#K311), caspase-3 activity assay kit (#K106), TMRE mitochondrial membrane potential assay kit, and mitochondria/cytosol fractionation kit (#K256) were purchased from BioVision (Milpitas, CA, USA). Primary antibodies against PARP (#9542S), LC3B (#2775S), GAPDH (#2118S), ERK (#9102S), p-ERK (#9101S), p38 (#9212S), p-p38 (#9211S), JNK (#9252S), p-JNK (#9251S), pAMPK (#2535), AMPK (#5832), mTOR (#2972), p-Raptor (#2083), Raptor (#2280), pULK1-Ser-317 (#12753), p-ULK1-Ser-757 (#6888), ULK1 (#8054), ATG5 (#2630), P62 (#5114), Beclin1 (#3738), pS6 (#2215), and S6 (#2317) were purchased from Cell Signaling Technology (Danvers, MA, USA). Alpha-spectrin (#MAB1622) antibody was purchased from EMD Millipore (Temecula, CA, USA). Caspase-3 (#sc-7148), actin (#sc-1616), cytochrome c (#sc-7159), and COX-IV (#sc-376731) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

2.2. Cell Culture

The human neuroblastoma cell lines SH-SY5Y (Korean Cell Line Bank (KCLB) No. 22266) and SK-N-SH (KCLB No. 30011) were obtained from the Korean Cell Line Bank (Seoul, Republic of Korea). Cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37 °C in a humidified incubator with 5% CO₂.

Primary cortical neurons were prepared from embryos of an ICR mouse at embryonic day 17 (E17). The animals were housed under standard laboratory conditions with a 12 h light/dark cycle and had ad libitum access to food and water. Prior to tissue collection, the mouse was deeply anesthetized with isoflurane and euthanized by cervical dislocation in accordance with approved ethical guidelines, in order to minimize pain and distress. All animal experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Wonkwang University, Republic of Korea (Approval No. WKU13-49), approved on 9 May 2023, and were conducted in accordance with the guidelines of the National Institute of Toxicological Research, Korea Food and Drug Administration. Following dissection, the cerebral cortex was isolated, and the meninges were carefully removed. The tissue was transferred to Hank’s balanced salt solution and enzymatically digested with trypsin for 30 min. Primary cortical neurons were prepared using standard trypsinization without the use of antimitotic or glia-selective treatments. The dissociated cells were suspended in DMEM containing 5% FBS and seeded onto poly-D-lysine-coated culture dishes, followed by incubation at 37 °C in a humidified atmosphere with 5% CO₂. After 5 h, the medium was replaced with Neurobasal medium supplemented with 0.4 mM glutamine and 0.5× B-27 supplement. To inhibit astrocyte proliferation, 0.5 μM cytarabine was added 24 h after plating. Treatments with 4-MD and other reagents were performed 24 h thereafter. Primary cortical neuron culture experiments were independently repeated three times.

2.3. MTT Assay

Cell viability was assessed using the MTT assay. Cells were seeded at 1 × 10⁵ cells per well in 96-well plates and treated with varying concentrations of 4-MD for 12 h, or with 20 μM 4-MD for up to 24 h. For inhibitor studies (NAC, PD98059, SP600125, SB203580, bafilomycin, and 3-MA), cells were pretreated with inhibitors for 1 h prior to 12 h 4-MD exposure. After treatment, the medium was replaced by 100 μL of MTT solution (1 mg/mL in DMEM medium) per well and incubated at room temperature for 2 h. The solution was carefully removed, and the formazan crystals were solubilized with 100 μL of DMSO. After 20 min, absorbance was measured at 540 nm using a microplate ELISA reader (SPECTRAmax190, Sunnyvale, CA, USA).

2.4. LDH Release Assay

LDH release was measured using a cytotoxicity assay kit. SH-SY5Y cells at 5 × 10⁵ cells per well in 24-well plates were seeded. After 24 h, cells were treated with 4-MD using the same protocol as for the MTT assay. Triton X-100 (1%) was used as a positive control. Following treatment, 100 μL of culture medium was collected into 96-well plates, and the reagent mixture was added according to the manufacturer’s instructions. After 30 min incubation in the dark, absorbance was evaluated at 490 nm in microplate ELISA reader.

2.5. Cell Morphology

SH-SY5Y cells were seeded at 2 × 10⁶ cells per well in 6-well plates and incubated for 24 h. Morphological changes were examined under a bright-field microscope (Nikon Eclipse TS100 FL, Tokyo, Japan) and photographed at 20× magnification after 12 h of 4-MD treatment.

2.6. Caspase-3 Activity Assay

Apoptosis was evaluated by measuring caspase-3 activity using a Caspase-3/CPP32 colorimetric assay kit (Biovision, Milpitas, CA, USA). At 2 × 10⁶ cells per well, SH-SY5Y cells were seeded in 6-well plates and treated with 4-MD at various concentrations for 12 h. For inhibitor studies, cells were pretreated for 1 h before 4-MD exposure. Caspase-3 activity was quantified according to the manufacturer’s protocol.

2.7. Measurement of Intracellular ROS

Intracellular ROS generation was measured using DCFDA fluorescence dye. Cells were seeded at 1 × 10⁵ cells per well in 96-well plates, incubated for 24 h, and treated with 4-MD for 12 h at different concentrations. As a positive control, H₂O₂ was added during the last 45 min. Cells were incubated with 20 μM DCFDA in serum-free medium for 20 min in the dark. Fluorescence was measured at 485/535 nm (excitation/emission) using a fluorescence microplate reader. Images were captured with an inverted fluorescence microscope.

2.8. Analysis of Mitochondrial Membrane Potential

Mitochondrial membrane potential was assessed using TMRE mitochondrial membrane potential assay kit according to the manufacturer’s instructions. SH SY5Y cells were seeded at 2 × 10⁵ cells per well in 96-well plates and incubated for 24 h. Cells were treated with 4-MD at varying concentrations for 12 h, then incubated with 200 nM TMRE dye for 20 min. FCCP (20 μM) was used as a negative control. Fluorescence was measured at 549/575 nm (excitation/emission).

2.9. Cytosolic and Mitochondrial Fractionation

Cell fractionation was performed using a mitochondria/cytosol fractionation kit according to the manufacturer’s instructions. SH-SY5Y cells were seeded at 2 × 10⁶ cells per well in 6-well plates and treated with 4-MD at various concentrations for 12 h before fractionation.

2.10. Immunocytochemistry

SH-SY5Y cells were seeded at 1 × 10⁶ cells per well on glass coverslips in 6-well plates. After 24 h, cells were treated with 20 μM 4-MD for 12 h, fixed with 200 μL methanol (−20 °C) for 15 min, blocked with 1% BSA in PBST for 1 h, and incubated overnight at 4 °C with anti-LC3 antibody (1:500). After washing, cells were incubated with secondary antibody (1:200) for 1 h at room temperature, counterstained with DAPI (1:1000, 3 min), and mounted with ProLong Gold antifade reagent. Images were captured using a confocal microscope (Olympus, Tokyo, Japan) at 40× magnification. DAPI was taken on the blue/DAPI channel, and LC3B was taken on the Green channel (488 nm).

2.11. Acridine Orange Staining

SH SY5Y cells were seeded at 2 × 10⁶ cells per well in a 6-well plate and treated with 4-MD for 12 h. Cells were stained with AO (1 μg/mL in PBS) for 20 min at 37 °C, washed, and observed by fluorescence microscopy (488 nm filter). Stained cells were then collected and analyzed by flow cytometry (FACS Calibur, BD Bioscience, San Jose, CA, USA). Green fluorescence was detected in the FL1 channel, while red fluorescence from acidic vesicular organelles was detected in the FL3 channel using a 610 nm emission filter.

2.12. Western Blotting

Cells were lysed in RIPA buffer (150 mM NaCl, 1% triton X-100, 1% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl, and 2 mM EDTA) supplemented with protease and phosphatase inhibitors (Roche). Lysates were centrifuged at 15,000 rpm for 10 min at 4 °C, and protein concentrations were determined using BCA assay (Thermo Scientific, Waltham, MA, USA). Equal protein amounts (15–30 μg) were separated on SDS-PAGE gels (8–13%), transferred to PVDF membranes, blocked with 5% skim milk, and incubated overnight with specific primary antibodies (1:1000 in 5% skim milk) at 4 °C. After washing, membranes were incubated with HRP-conjugated secondary antibodies (1:1000–2000) for 1 h. Blots were developed using chemiluminescent substrate (Thermo Scientific) and visualized on X-ray film (Fujifilm, 9-7-3 Akasaka, Minato-ku, Tokyo, Japan). Band intensities were quantified using ImageJ software version 1.54m 5 (NIH, Bethesda, MD, USA). All target proteins were detected using freshly prepared PVDF membranes. GAPDH was used as a loading control and, when necessary, was detected by stripping and reprobing the same membrane after analysis of the target proteins. Stripping and reprobing conditions were carefully optimized to minimize signal carryover and ensure reliable detection.

2.13. Statistical Analysis

All experiments were repeated at least three times independently. Data are expressed as means ± standard deviation (S.D.), and representative results are shown. Statistical significance was determined by one-way ANOVA, followed by Tukey’s test for comparing the paired sets of data and Dunnett test for multiple sets of data using GraphPad Prism 8 software (La Jolla, CA, USA). A p-value less than 0.05 was considered statistically significant.

3. Results

3.1. 4-MD Induces Cytotoxicity in Neuroblastoma SH-SY5Y Cells

4-MD was isolated from the heartwoods of Dalbergia odorfera (Figure 1A) [21]. To examine its anti-cancer effects, SH-SY5Y neuroblastoma cells were exposed to increasing concentrations of 4-MD (5–40 μM) for 12 h, or to 20 μM 4-MD for varying durations (1–24 h). As shown in Figure 1B,D, 4-MD significantly reduced SH-SY5Y cell viability in a dose- and time-dependent manner. Consistently, LDH release was markedly increased under these conditions, further confirming cytotoxicity (Figure 1C,E). Morphological analyses revealed progressive changes with higher concentrations of 4-MD, including cell shrinkage, loss of dendritic processes, and detachment from the culture substrate (Figure 1F). Furthermore, we evaluated the effect of cyclophosphamide as a positive control in our experiment, where SH-SY5Y cells treated with cyclophosphamide reduced the cell viability in a dose-dependent manner (Figure S3A).

Consistent with these findings, similar cytotoxic effects of 4-MD were observed in another neuroblastoma cell line, SK-N-SH (Figure 1G). Both SH-SY5Y cell and SK-N-SH were MYCN non-amplified neuroblastoma cell lines, so to cover the wide heterogeneity of neuroblastoma, we evaluated the effect of 4-MD in the BE(2)-C cell, a MYCN-amplified cell line. Similar cytotoxic effects of 4-MD were observed in the BE(2)-C cell line (Figure 1H). Collectively, these results indicate that 4-MD exerts potent cytotoxic activity against multiple neuroblastoma cells.

By contrast, primary cortical neurons were largely resistant to 4-MD, showing no significant viability loss up to 30 μM and only a slight reduction at 40 μM, which remained markedly lower than the cytotoxicity observed in neuroblastoma cells (Figure 1I), whereas cyclophosphamide treatment showed similar reduced cell viability in primary cortical neurons and neuroblastoma cells (Figure S3A,B). Also, to demonstrate the vulnerability of primary cortical neurons to some neurotoxic reagents, we treated doxorubicin and showed decreased cell viability (Figure S3C). These findings suggest that 4-MD exerts cancer-selective cytotoxicity in neuroblastoma cells while sparing normal neuronal cells.

3.2. 4-MD Induces MAPK Activation and Apoptosis in SH-SY5Y Cells

To investigate the mechanisms of 4-MD-induced cytotoxicity, we examined MAPK signaling, which is closely associated with apoptotic cell death. Western blot analysis revealed that 4-MD dose-dependently increased phosphorylation of ERK, p38, and JNK (Figure 2A), suggesting MAPK activation contributes to 4-MD-induced cell death.

Consistently, 4-MD treatment enhanced PARP cleavage and increased the generation of α-spectrin breakdown products (150 kDa, calpain-dependent; 120 kDa, caspase-3-dependent) (Figure 2B). Moreover, 4-MD reduced pro-caspase-3 levels, elevated cleaved caspase-3 expression (Figure 2B), and markedly enhanced caspase-3 enzymatic activity (Figure 2C). These results demonstrate that 4-MD triggers caspase-dependent apoptosis in SH-SY5Y cells.

Importantly, co-treatment with Z-VAD-FMK, a pan-caspase inhibitor, partially rescued cell viability (Figure 2D) and completely abolished PARP cleavage (Figure 2E). These findings indicate that caspase activation plays a central role in 4-MD-induced apoptosis, though additional caspase-independent pathways may also contribute to cell death.

3.3. 4-MD-Induced Cell Death Is Mediated by ROS Production

Because intracellular ROS are major mediators of cellular damage and apoptosis [22,23], we examined ROS production in 4-MD-treated cells. DCFDA fluorescence analysis revealed a dose-dependent increase in intracellular ROS levels (Figure 3A), which was further confirmed by immunocytochemical staining (Figure 3B).

To assess mitochondrial involvement, we measured mitochondrial membrane potential using TMRE fluorescence. 4-MD significantly reduced mitochondrial membrane potential (Figure 3C) and promoted cytochrome c release from mitochondria into the cytosol (Figure 3D). These findings suggest that 4-MD disrupts mitochondrial integrity, leading to ROS accumulation and contributing to its cytotoxic effects in neuroblastoma cells.

To validate the role of ROS in 4-MD-mediated cytotoxicity, SH-SY5Y cells were co-treated with the ROS scavenger N-acetylcysteine (NAC). NAC markedly restored cell viability and reduced LDH release compared with 4-MD treatment alone (Figure 3E,F). In addition, NAC co-treatment abolished 4-MD-induced PARP and α-spectrin cleavage and significantly suppressed caspase-3 activation (Figure 3G,H). Importantly, NAC also attenuated the phosphorylation of ERK, p38, and JNK (Figure 3I). Together, these findings demonstrate that ROS generation is a critical mediator of 4-MD-induced MAPK activation and apoptotic cell death in SH-SY5Y neuroblastoma cells.

3.4. 4-MD Induces Apoptotic Cell Death Through ERK and p38 Activation

To define the contribution of individual MAPKs, SH-SY5Y cells were co-treated with the ERK inhibitor PD98059, the p38 inhibitor SB203580, or the JNK inhibitor SP600125. All three inhibitors partially rescued 4-MD-induced cytotoxicity, as shown by MTT and LDH assays (Figure 4A,B). Notably, PD98059 and SB203580 significantly reduced PARP cleavage and caspase-3 activity (Figure 4C,D), indicating that ERK and p38 activation play critical roles in 4-MD-induced apoptosis. By contrast, SP600125 failed to suppress PARP cleavage or caspase-3 activity, despite partially improving cell viability. These results suggest that 4-MD triggers apoptotic cell death predominantly via ERK and p38 signaling pathways.

3.5. 4-MD Induces Autophagy via Activation of the AMPK/mTOR/ULK1 Pathway

Autophagy is increasingly recognized as a programmed cell death mechanism distinct from apoptosis [24]. To determine whether 4-MD induces autophagy, we first analyzed LC3B conversion. Western blotting revealed a dose- and time-dependent increase in the LC3B-II/LC3B-I ratio following 4-MD treatment (Figure 5A,B). Consistently, immunocytochemistry showed enhanced LC3 puncta formation, indicating autophagosome maturation (Figure 5C). Acridine orange staining further confirmed an increase in acidic vesicular organelles (AVOs), as shown by both flow cytometry and fluorescence microscopy (Figure 5D,E).

We next examined autophagy-related signaling pathways. 4-MD enhanced phosphorylation of AMPK, Raptor, and ULK1 at Ser317, while reducing phosphorylation of S6K only at higher concentrations (≥30 μM) and ULK1 at Ser757 (Figure 5F). In addition, 4-MD upregulated Beclin-1 and ATG5, and reduced p62 expression, indicating activation of autophagy-related signaling pathways. These results suggest that 4-MD modulates autophagy in SH-SY5Y cells through the AMPK/mTOR/ULK1 signaling pathway.

3.6. 4-MD-Induced Autophagic Cell Death Is Mediated by p38 and JNK

To determine whether 4-MD-induced autophagy contributes to its cytotoxic effects, SH-SY5Y cells were co-treated with the autophagy inhibitors bafilomycin A1 or 3-methyladenine (3-MA). Bafilomycin A1, a lysosomal inhibitor, prevents the fusion of autophagosomes with lysosomes, whereas 3-MA, a class III PI3K inhibitor, blocks the conversion of LC3B-I to LC3B-II [25]. Both inhibitors partially attenuated 4-MD-induced loss of cell viability and LDH release (Figure 6A,B), suggesting that autophagy partially contributes to 4-MD-mediated cytotoxicity. However, neither inhibitor prevented PARP cleavage or caspase-3 activation (Figure 6C,D), indicating that apoptotic signaling proceeds independently of autophagy blockade.

In contrast, the antioxidant NAC effectively suppressed 4-MD-induced LC3B-II accumulation and inhibited activation of the AMPK/mTOR/ULK1 pathway (Figure 6E,F), demonstrating that ROS generation is essential for autophagy induction. Furthermore, pharmacological inhibition of p38 (SB203580) or JNK (SP600125) significantly reversed LC3B-II accumulation (Figure 6G), implicating both p38 and JNK as critical mediators of 4-MD-induced autophagy. Together, these findings indicate that 4-MD triggers ROS-dependent autophagic cell death through activation of the AMPK/mTOR/ULK1 axis, with additional regulation by the p38 and JNK signaling pathways.

3.7. 4-MD Induces Autosis in SH-SY5Y Cells

We next examined whether Na⁺/K⁺-ATPase contributes to the anti-cancer effects of 4-MD, since autosis, a distinct form of autophagy-dependent cell death, is characterized by its dependence on Na⁺/K⁺-ATPase activity [26]. Co-treatment with ouabain, a specific Na⁺/K⁺-ATPase antagonist, significantly attenuated 4-MD-induced cell death (Figure 7A) and reduced LC3B-II accumulation, without affecting PARP cleavage (Figure 7B). These findings indicate that 4-MD triggers Na⁺/K⁺-ATPase-dependent autosis in SH-SY5Y neuroblastoma cells.

We then investigated the relationship between Na⁺/K⁺-ATPase-dependent autosis and MAPK signaling in response to 4-MD. Cells were pretreated with ouabain for 1 h, followed by 4-MD treatment for 12 h, and phosphorylation levels of ERK, JNK, and p38 were assessed. Ouabain pretreatment dose-dependently reduced 4-MD-induced ERK and JNK phosphorylation, while paradoxically enhancing p38 phosphorylation (Figure 7C). Interestingly, ouabain alone increased p38 phosphorylation at 100 nM and suppressed ERK phosphorylation in a dose-dependent manner, suggesting that Na⁺/K⁺-ATPase inhibition itself modulates MAPK activity independently of 4-MD. In contrast, the attenuation of JNK phosphorylation by ouabain in the presence of 4-MD supports a critical role for JNK in mediating Na⁺/K⁺-ATPase-dependent autosis (Figure 7C).

4. Discussion

Neuroblastoma is the most common extracranial solid tumor in childhood, and approximately one-third of affected patients are classified as high-risk due to therapeutic resistance that largely arises from profound intratumoral heterogeneity [1,2]. Such heterogeneity enables distinct tumor subpopulations to evade therapies targeting a single death pathway, thereby limiting long-term efficacy. In this context, therapeutic strategies capable of engaging multiple, non-redundant cell death mechanisms may offer a more effective approach. In the present study, we show that 4-MD exhibits potent cytotoxic activity in human neuroblastoma SH-SY5Y cells through the concomitant induction of apoptosis and autophagy, both driven by ROS generation and downstream MAPK signaling activation (Figure 8). 4-MD-induced ROS provoked mitochondrial dysfunction, resulting in cytochrome c release and caspase-3 activation via ERK and p38 signaling, thereby driving apoptotic cell death. Concurrently, ROS-mediated AMPK activation suppressed mTORC1 activity through Raptor phosphorylation and modulated ULK1 phosphorylation, promoting ATG5 upregulation and LC3-II accumulation, to facilitate autophagic cell death, including Na⁺/K⁺-ATPase-associated autosis. Collectively, these findings extend previous observations and provide several conceptual insights into the regulation of neuroblastoma cell death.

First, our findings indicate that 4-MD does not simply trigger parallel cellular stress responses, but rather concurrently activates apoptosis and autophagy as functionally independent yet cytotoxic death pathways. Importantly, pharmacological blockade of either pathway did not lead to compensatory activation of the other, suggesting that 4-MD may circumvent a commonly observed resistance mechanism in which tumor cells evade apoptosis by shifting toward cytoprotective autophagy, or conversely suppress autophagy to favor alternative survival programs [27,28,29]. In neuroblastoma, accumulating evidence indicates that therapy-induced autophagy frequently serves a protective role. For instance, in high-risk neuroblastoma cells, conventional chemotherapeutic agents such as cisplatin, cyclophosphamide, and etoposide have been shown to enhance apoptotic cell death when autophagy is pharmacologically inhibited using chloroquine, underscoring the contribution of cytoprotective autophagy to chemoresistance [30]. Similarly, genipin has been reported to induce both apoptosis and autophagy in SK-N-SH neuroblastoma cells; however, inhibition of autophagy further augments genipin-induced cytotoxicity, indicating that autophagy acts predominantly as a survival mechanism in this context [31]. In contrast, ABTL0812, a novel anticancer agent, has been shown to induce cytotoxic autophagy in SK-N-BE(2) neuroblastoma cells, where autophagy activation directly contributes to cancer cell death rather than promoting cell survival [32]. Collectively, these studies illustrate the dual, context-dependent role of autophagy; notably, our findings suggest that autophagy induced by 4-MD contributes directly to cell death rather than serving as a compensatory protective response. Although the molecular determinants underlying this functional switch remain to be fully elucidated, the concurrent activation of non-compensatory apoptotic and autophagic pathways distinguishes 4-MD from many previously reported neuroblastoma-targeting agents, which typically induce autophagy as an adaptive survival response that constrains therapeutic efficacy.

Second, this study reveals a functional segregation within ROS–MAPK signaling, whereby ERK and p38 predominantly contribute to apoptotic execution, while p38 and JNK preferentially regulate autophagy-related signaling cascades associated with autophagosome accumulation. This division of labor provides mechanistic insight into how a common upstream signal—ROS—can be differentially decoded into distinct lethal outcomes. While excessive ROS is known to induce oxidative damage and cell death through MAPK cascades, the relative contribution of ERK, p38, and JNK varies substantially depending on cellular context and pharmacological stimulus [33]. For example, apatinib induces both apoptosis and autophagy in neuroblastoma cells primarily through suppression of ERK phosphorylation [34], whereas meriolin1 enhances p38 and JNK activation while inhibiting ERK signaling [35]. In contrast, cearoin selectively increases ERK activation without significantly altering JNK phosphorylation in SH-SY5Y cells [36]. Notably, although 4-MD suppresses ERK phosphorylation in human bladder cancer cells [16], it robustly activates ERK, p38, and JNK in SH-SY5Y cells, underscoring the context-dependent nature of MAPK signaling. Importantly, JNK activation in 4-MD-treated neuroblastoma cells contributed to cytotoxicity without participating in canonical apoptotic signaling, suggesting that non-apoptotic MAPK functions play a critical role in mediating 4-MD-induced cell death. Such layered signaling architecture may be particularly advantageous in heterogeneous tumors, where reliance on a single downstream effector frequently facilitates adaptive resistance.

Consistent with this complexity, NAC fully restored cell viability, whereas LDH release and MAPK phosphorylation were only partially suppressed. This discrepancy may reflect assay-dependent sensitivity, inter-experimental variability, or temporal divergence between upstream signaling events and downstream survival outcomes. In addition, while our data support ROS as the primary upstream regulator of MAPK signaling in response to 4-MD, the incomplete suppression of MAPK activation by NAC suggests that minor ROS-independent mechanisms may also contribute.

Third, our data identify Na⁺,K⁺-ATPase-dependent autosis as a previously unrecognized component of 4-MD-mediated cytotoxicity in neuroblastoma. Autosis has been predominantly described in ischemic injury [26,37], and its relevance to neuroblastoma has remained largely unexplored. Although autosis-inducing agents have not been extensively investigated as direct anticancer therapeutics, autosis triggered by oncolytic virus-infected T cells has been shown to mediate potent bystander killing of tumor cells without affecting non-tumor cells [38]. The observation that ouabain selectively attenuated autophagic cell death without affecting apoptotic markers supports the presence of a distinct autotic program activated by 4-MD. This finding expands the spectrum of cell death modalities that may be therapeutically exploitable in neuroblastoma and suggests autosis as a potential vulnerability in this disease. Notably, ouabain suppressed 4-MD-induced ERK and JNK phosphorylation while enhancing p38 activation, consistent with prior reports demonstrating that ouabain independently activates p38 signaling [39,40,41]. While these observations do not establish a direct causal relationship, they suggest that MAPK signaling may be functionally associated with 4-MD-induced autosis and warrant further mechanistic investigation.

Finally, the preferential cytotoxicity of 4-MD toward neuroblastoma cells, accompanied by minimal toxicity in primary cortical neurons, suggests a potentially favorable therapeutic index. Neuroblastoma cells are characterized by elevated basal oxidative stress and dysregulated autophagy, features that may render them particularly vulnerable to additional perturbations in redox homeostasis and autophagy-related signaling pathways [34,35]. In contrast, neurons depend on tightly regulated autophagy to preserve cellular integrity and maintain long-term survival [36,37]. It is therefore plausible that 4-MD amplifies ROS-dependent autophagic signaling beyond a tolerable threshold in malignant cells, while normal neurons remain relatively resilient under the experimental conditions tested. Such differential sensitivity may reflect intrinsic differences in metabolic state, redox buffering capacity, and autophagy regulation between rapidly proliferating tumor cells and post-mitotic neurons. It should also be noted that the primary cortical neurons used in this study were derived from mice, whereas SH-SY5Y, SK-N-SH, and BE(2)-C cells are of human origin. Therefore, species-dependent differences may partially contribute to the differential sensitivity to 4-MD. However, control experiments using cyclophosphamide demonstrated comparable reductions in cell viability in both mouse primary cortical neurons and human SH-SY5Y cells under identical experimental conditions (Supplementary Figure S3A,B), indicating that cross-species responsiveness to cytotoxic chemotherapy is broadly similar within our experimental system. Accordingly, while species-related factors cannot be entirely excluded, the relative resistance of primary cortical neurons to 4-MD is unlikely to be solely explained by species origin, but rather reflects cell-type- and context-dependent effects of 4-MD. Although the precise molecular determinants underlying this selectivity remain to be fully elucidated, our findings suggest that 4-MD preferentially exploits tumor-associated vulnerabilities without broadly compromising neuronal viability. This selective cytotoxicity is particularly relevant for pediatric malignancies such as neuroblastoma, in which preservation of long-term neurological function represents a critical clinical priority.

Beyond cell-type selectivity, the pronounced molecular heterogeneity of neuroblastoma represents a central obstacle to durable therapeutic responses. MYCN amplification, which defines an aggressive high-risk subtype, is one of the most clinically significant determinants of this heterogeneity. While SH-SY5Y and SK-N-SH cells are MYCN-non-amplified models, we additionally evaluated the effect of 4-MD in BE(2)-C cells, a MYCN-amplified neuroblastoma cell line. Notably, 4-MD induced comparable cytotoxic effects in BE(2)-C cells (Figure 1H), indicating that its anti-tumor activity is not restricted to MYCN-non-amplified backgrounds. These findings suggest that 4-MD-mediated cytotoxicity may extend across molecularly distinct neuroblastoma subtypes, potentially overcoming one dimension of tumor heterogeneity. Rather than targeting a single oncogenic driver, 4-MD appears to exploit shared vulnerabilities associated with oxidative stress and dysregulated autophagy, mechanisms that are broadly relevant across genetically diverse neuroblastoma populations. Nevertheless, validation across a broader panel of neuroblastoma cell lines would further strengthen the generalizability of our conclusions regarding the cytotoxic efficacy of 4-MD across heterogeneous neuroblastoma subtypes.

Importantly, the present study was designed as a mechanistic in vitro investigation aimed at defining the signaling pathways and cell death modalities engaged by 4-MD at the cellular level. While these results provide a foundational framework for understanding 4-MD-induced cytotoxicity, comprehensive in vivo validation will be essential to rigorously evaluate therapeutic efficacy, pharmacokinetics, systemic toxicity, and potential neurotoxicity, particularly within the context of the developing nervous system. Accordingly, future studies employing appropriate neuroblastoma animal models, including xenograft and survival analyses, will be necessary to determine whether the selective vulnerability observed in vitro can be translated into a safe and effective therapeutic strategy in vivo.

Collectively, our findings support a model in which 4-MD acts as a multi-modal amplifier of cell death, converting intrinsic oxidative stress into concurrent apoptotic, autophagic, and autotic programs through coordinated ROS–MAPK–AMPK/mTOR/ULK1 signaling. Rather than targeting a single oncogenic pathway, 4-MD engages multiple non-redundant death mechanisms, potentially reducing the likelihood of adaptive escape in heterogeneous neuroblastoma populations.

5. Conclusions

In conclusion, this study advances current understanding of neuroblastoma cell death regulation by demonstrating that excessive activation of autophagy—traditionally viewed as cytoprotective in this context—can function as a dominant cytotoxic mechanism when integrated with apoptotic and autotic pathways. These findings position 4-MD as a promising prototype for multi-pathway-engaging therapeutic strategies and support further in vivo evaluation in models of therapy-resistant neuroblastoma.