Nanomagnetic Hyperthermia Sensitizes Gemcitabine Chemosensitivity in Pancreatic Cancer by Inhibiting HSPB1 to Amplify ACSL4‐Mediated Ferroptosis

ABSTRACT

Pancreatic cancer, one of the most malignant solid tumors with the poorest prognosis, is commonly treated with gemcitabine (GEM)‐based systemic chemotherapy. However, chemoresistance remains a significant therapeutic challenge. Ferroptosis, a novel form of programmed cell death, has demonstrated exceptional susceptibility in chemoresistant cancer cells. Due to their unique magnetic responsiveness, iron oxide nanoparticles can generate heat under alternating magnetic fields, which enables magnetic hyperthermia (MH) therapy. This approach not only directly induces tumor cell apoptosis but also enhances chemosensitization. In this study, we demonstrate that ferrimagnetic vortex‐domain iron oxide nanorings (FVIOs) as an ideal nanoplatform for MH, significantly improving GEM chemosensitivity in pancreatic cancer. Both in vitro and in vivo experiments demonstrated that MH not only synergizes with GEM to exhibit an antitumor effect in wild‐type (WT) pancreatic tumors, but also directly suppresses the proliferation of gemcitabine‐resistant (GR) counterparts while alleviating chemoresistance. High‐throughput transcriptomic profiling revealed an increased susceptibility of GR cells to ferroptosis induction. Mechanistically, MH suppressed HSPB1 expression in GR cells, thereby attenuating its regulatory role in the ubiquitination‐mediated degradation of ACSL4 protein and ultimately promoting ferroptosis. Comprehensive in vivo evaluations confirmed the multidimensional chemosensitizing efficacy of MH alongside favorable biosafety. Our findings highlight MH as a promising adjuvant strategy for pancreatic cancer, particularly in GEM‐resistant cases, by overcoming chemoresistance through the potentiation of ferroptosis.

1. Introduction

Pancreatic cancer remains one of the most lethal malignancies in the digestive system, with the reported 5‐year survival rate approximating only 13% [1] and accounting for nearly 5% of global cancer‐related mortality [2]. The majority of pancreatic cancer cases are ineligible for curative resection due to the presence of metastases at diagnosis and chemotherapy there of remains a cornerstone of current therapeutic strategies for this malignancy [3, 4]. However, the therapeutic process is frequently complicated by the rapid development of chemoresistance, resulting in low response rates to gemcitabine (GEM) chemotherapy [5]. The mechanisms underlying GEM chemoresistance often involve the crosstalk between malignant cells and components of the tumor microenvironment [6], yet the majority of current therapeutic strategies targeting this challenge have yielded limited clinical outcomes [7, 8].

In recent years, nanotechnology has witnessed rapid development in the biomedical field, where diverse nanomaterials have been effectively applied to cancer treatment based on their unique properties [9]. Among these, iron oxide nanoparticles have emerged as promising candidates due to their favorable biocompatibility and well‐defined metabolic mechanisms [10, 11], which enable the efficient loading of various therapeutic molecules as delivery carriers. Moreover, their exceptional magnetic responsiveness allows them to convert magnetic energy into localized heat through magnetic hysteresis loss and relaxation effects when exposed to an alternating magnetic field (AMF), a process known as magnetic hyperthermia (MH) therapy [12]. Compared to conventional hyperthermia approaches, MH therapy demonstrates superior spatial precision in confining thermal effects to tumor regions while minimizing damage to adjacent healthy tissues, an advantage that has motivated extensive research in oncological applications since its early conceptualization [13]. Another notable advantage of MH therapy lies in its potential for synergistic combination with traditional chemotherapy, and may help overcome limitations such as chemoresistance encountered in monotherapy regimens [14, 15]. This combinatorial strategy shows particular promise for challenging malignancies like pancreatic cancer.

Ferroptosis, a newly identified form of programmed cell death, is characterized by iron overload‐driven lipid peroxidation accumulation [16]. This process is driven by the iron‐catalyzed peroxidation of polyunsaturated fatty acids within cell membranes, which leads to the accumulation of membrane lipid peroxides and cell death [17]. Beyond known inducers such as Erastin and RSL3, emerging evidence suggests that variations in temperature and oxygen levels may also activate ferroptosis [18]. Indeed, the induction of ferroptosis via nanoplatforms has been increasingly explored in cancer therapy, including for prostate [19], breast [20], and thyroid cancers [21]. Notably, oncogenic and proliferation‐related signaling pathways in cancer cells have been closely linked to ferroptosis regulation, the targeted modulation of ferroptosis signaling a novel anticancer strategy [22, 23]. This approach exhibits particular efficacy against therapy‐resistant or highly metastatic cancer cells, which demonstrate aberrant susceptibility to ferroptosis [24]. Recent investigations reveal that GEM‐induced reactive oxygen species (ROS) accumulation synergizes with ferroptosis to enhance cancer cell cytotoxicity [25, 26]. This synergistic lethality involves molecular interactions with intrinsic drug resistance pathways in tumor cells [27]. Consequently, elucidating the intrinsic relationship between ferroptosis and GEM resistance warrants further exploration.

In this study, we demonstrate that ferrimagnetic vortex‐domain iron oxide nanorings (FVIOs) are an effective magnetic carrier for nanomagnetic hyperthermia and significantly sensitize pancreatic cancer to GEM chemotherapy. Notably, MH not only synergizes with GEM to exert an antitumor effect in wild‐type (WT) pancreatic cancer, but also directly inhibits the proliferation of GEM‐resistant (GR) tumors while substantially alleviates their chemoresistance. Mechanistically, GR pancreatic cancer cells exhibit heightened susceptibility to ferroptosis. MH suppresses the expression of heat shock protein HSPB1 in GR cells, thereby disrupting its inhibition of the ubiquitination degradation of the ACSL4 protein, ultimately leading to enhanced ferroptosis induction in GR populations. Comprehensive in vivo evaluations further confirm the biosafety profile of this therapeutic approach. Our findings support MH as a novel treatment strategy for pancreatic cancer, particularly effective against chemoresistant variants.

2. Results

2.1. Characterization of FVIOs and its Induction of MH in AMF

FVIOs, a magnetic iron oxide nanomaterial based on Fe₃O₄, exhibited a particle size of approximately 60–80 nm under TEM imaging and were internalized by cells through endocytosis (Figure 1A; Figure S1). When dispersed in deionized water at room temperature, the hydrodynamic diameter of FVIOs was measured to be about 91.4 nm, with minimal fluctuations in particle size across varying temperatures. Similar stability was observed in the zeta potential measurements (Figure 1B,C). Owing to the intrinsic magnetic properties of its iron oxide composition, FVIOs generated MH under an AMF. Notably, studies [28] have demonstrated that FVIOs exhibit superior heat generation performance compared to conventional magnetic iron oxide nanomaterials, such as superparamagnetic iron oxide nanoparticles. To evaluate their heating efficiency, FVIOs were prepared at varying concentrations (based on Fe content) (Figure 1D) and exposed to an AMF. Remote temperature monitoring revealed a concentration‐dependent heating profile, with a 50% particle concentration being notably more efficient, and the temperature increase was further enhanced with higher magnetic field intensity (Figure 1E,F). Additionally, MH measurements performed on cells that had internalized FVIOs showed no significant difference in heat production compared to FVIOs alone (Figure 1G,H).

FIGURE 1.

Characterization of FVIOs and its induction of MH in AMF. (A) TEM image of extracellular and intracellular FVIOs (scale bar: 50 nm). (B) Hydrodynamic diameter and (C) Zeta potential diagram of FVIOs measured at different temperatures. (D) Visual representations of FVIOs at different concentrations (Fe, 0–150 µg/mL). Temperature profiles of (E) different concentrations of FVIOs exposed to 250 Oe AMF, and (F) 50 µg/mL FVIOs exposed to different intensities of AMF (250–400 Oe) for 10 min. Temperature profiles of (G) different concentrations of FVIOs exposed to AMF of 250 Oe, and (H) 50 µg/mL FVIOs exposed to different intensities of AMF for 10 min after 12 h incubation with pancreatic cancer cells MIA PaCa‐2. FVIOs ferrimagnetic vortex‐domain iron oxide nanorings; MH magnetic hyperthermia; AMF alternating magnetic field; TEM transmission electron microscopy.

2.2. Magnetic Hyperthermia Synergizes with GEM to Exert Antitumor Effects in Pancreatic Cancer

Previous studies have demonstrated FVIOs‐mediated MH therapy in breast cancer [28] and hepatocellular carcinoma [29]. To investigate its therapeutic potential in pancreatic cancer, we treated pancreatic cancer cell lines (Panc‐1 and MIA PaCa‐2) with MH. Compared to FVIOs monotherapy, MH not only significantly suppressed cell viability but also reduced the IC50 values of GEM in both cell lines (Figure 2A). Based on the preliminary experimental results, GEM (0.02 µM) and MH (50 µg/mL Fe) were selected as the intervention concentrations for subsequent studies. Notably, MH exhibited tumor cell selectivity with minimal impact on the viability of hTERT‐HPNE pancreatic ductal epithelial cells (Figure S2). To further evaluate the synergistic antitumor effects of MH and GEM, combination therapy was administered to pancreatic cancer cells. High‐throughput synergy analysis revealed HSA synergy scores >10 across cell lines, indicating strong pharmacological synergy (Figure 2B). Colony formation assays confirmed that the combination therapy more potently inhibited tumorigenic proliferation compared to monotherapies (Figure 2C). Similarly, combined treatment significantly enhanced cytotoxicity and induced apoptosis (Figure 2D; Figure S3A,B). Western blotting analysis demonstrated downregulation of proliferation‐associated proteins (Cyclin‐D3/PCNA) and an increased apoptosis‐related protein ratios (Bcl‐2/Bax) in the combination group (Figure 2E).

FIGURE 2.

MH synergizes with GEM to exert antitumor effects in pancreatic cancer. (A) Effects of different concentrations of FVIOs with or without AMF on the cell viability of two pancreatic cancer cell lines Panc‐1 and MIA PaCa‐2 were detected by the CCK‐8 method, and the effect of MH on IC50 of GEM was also assessed (n = 3, two‐way ANOVA). (B) HSA synergy score of MH and GEM in Panc‐1 and MIA PaCa‐2 cells was fitted based on the inhibition of cell viability by the two treatments. GEM (0.02 µM), MH (50 µg/mL Fe) or their combination were used to treat MIA PaCa‐2 and Panc‐1 cells, and (C) the cell proliferation was measured by colony formation assay (scale bar: 5 mm; n = 3, two‐way ANOVA); (D) the cell apoptosis was analyzed by flow cytometry detection; (E) the expression level of proteins related to cell proliferation or apoptosis was detected using western blotting assay. (F) The subcutaneous xenograft model mice with Panc‐1 cells were treated with GEM (25 mg/kg), MH (3 mg/cm³), alone or in combination. The tumor volumes and body weight were measured every 2 days until the experiment was terminated. (G) Representative images of the subcutaneous xenograft tumors. (H) The average weights of the tumors (n = 5, one‐way ANOVA). (I) The changes in tumor volume (n = 5, two‐way ANOVA). (J) H&E and Ki‐67 staining of tumor tissues (scale bar: 500 µm). ns: Not significant. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. GEM gemcitabine; CCK‐8 cell counting kit‐8; IC50 half‐maximal inhibitory concentration; HSA highest single agent; PI propidium iodide; FITC fluorescein isothiocyanate; H and E hematoxylin and eosin.

To further confirm the synergistic antitumor effect in vivo, nude mice bearing subcutaneous xenografts were treated with GEM and MH monotherapies or combination therapy (Figure 2F). The combination group exhibited the most substantial suppression of tumor growth, with the smallest tumor volumes and weights (Figure 2G–I). Histopathological analysis revealed disorganized tumor architecture with extensive necrosis in H and E‐stained sections and significantly reduced Ki‐67 proliferation indices in the combination group (Figure 2J). In addition to the therapeutic effects, the side effects of MH treatment are of equal concern. Real‐time infrared thermography demonstrated that AMF exposure did not alter systemic body temperature, while localized tumor hyperthermia reached 42–45°C within 10 min of MH treatment (Figure S4A), indicating tumor‐specific thermal effects. Serum iron assay and Prussian blue staining of organs showed no significant fluctuations in Fe levels post‐MH treatment (Figure S4B,C). Moreover, MH treatment caused neither body weight changes (Figure S4D) nor histopathological abnormalities in major organs (Figure S4E), confirming its favorable safety profile. In conclusion, our findings establish that MH synergistically enhances the antitumor efficacy of GEM in pancreatic cancer, while maintaining excellent biosafety.

2.3. MH Directly Inhibits Proliferation and Induces Apoptosis in GR Pancreatic Cancer

Chemoresistance remains a major obstacle in the clinical GEM chemotherapy for pancreatic cancer. To investigate the therapeutic potential of MH against GR pancreatic cancer, we established stable GR cell lines (MIA GR and PANC GR) from their WT counterparts via low‐concentration, stepwise induction (Figure 3A). Both MIA GR and PANC GR exhibited markedly enhanced resistance to GEM, with IC50 values hundreds of times higher than those of their WT counterparts (Figure 3B,C). When MH was applied to both WT and GR cells, cell viability assays revealed that MH at 40 µg/mL significantly inhibited GR cell proliferation. The differential therapeutic effects between WT and GR cells became more pronounced with increasing MH concentrations (Figure 3D). Microscopic observation demonstrated reduced cell density and increased apoptotic morphology in MH‐treated GR cells, corroborated by Calcein/PI staining, which indicated stronger cytotoxicity toward GR cells (Figure 3E,F).

FIGURE 3.

MH inhibits proliferation and induces apoptosis in GR pancreatic cancer. (A) Schematic diagram of constructing GR pancreatic cancer cells. (B) Cell viability of WT and GR cell lines of Panc‐1 and MIA PaCa‐2 in response to GEM and (C) their respective IC50 were assessed. (D) Cell viability of respective WT and GR cell lines of Panc‐1 and MIA PaCa‐2 in response to MH mediated by different concentrations of FVIOs were detected by the CCK‐8 method (n = 3, two‐way ANOVA). WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with MH, and (E) the apoptotic morphology of the cells was observed under light microscope (scale bar: 100 µm); (F) the activity and toxicity to the cells were labelled using Calcein/PI dye respectively (scale bar: 250 µm); (G) the proliferative capacity of the cells was detected using colony formation assay (scale bar: 5 mm; n = 3, two‐way ANOVA); (H) the proportion of apoptotic cells was detected using flow cytometry; (I) the expression level of proteins related to cell proliferation or apoptosis was detected using western blotting assay. C57 mouse pancreatic orthotopic xenograft tumor models were constructed using the WT and GR cells of MIA PaCa‐2 and subsequently given MH treatment. (J) The representative images of the pancreatic orthotopic xenograft tumors and (K) statistics of their volume (n = 5, one‐way ANOVA). (L) H and E, Prussian Blue iron staining of tumor tissues (scale bar: 100 µm; enlarged image: 40 µm). ns: Not significant. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. GR gemcitabine‐resistant; WT wild‐type.

The colony‐formation assay further confirmed that MH more effectively suppressed tumorigenic proliferation in GR cells compared to WT ones (Figure 3G). Flow cytometry analysis revealed significantly higher apoptosis rates in MH‐treated GR cells, a finding validated by Western blotting (Figure 3H,I). To evaluate the efficacy of MH in vivo, orthotopic pancreatic tumor models were established using MIA WT and MIA GR cells in mice. Gross tumor morphology and volumetric analysis demonstrated that GR tumors responded more robustly to MH treatment than WT tumors (Figure 3J,K). H and E and Prussian blue staining of tumor tissues revealed more extensive tissue destruction, increased necrotic cells, and substantial iron accumulation in MH‐treated GR tumors compared to WT counterparts (Figure 3L). Taken together, these findings demonstrate that MH exerts significantly enhanced therapeutic effects against GR pancreatic cancer compared to its effects on WT tumors, providing a promising strategy for overcoming chemoresistance in this lethal malignancy.

2.4. MH Alleviates Drug Resistance in GR Pancreatic Cancer

To investigate whether MH exerts a direct therapeutic effect on GR pancreatic cancer while influencing its intrinsic drug resistance, we measured the IC50 of GEM in MIA GR and PANC GR cell lines following MH intervention. Compared to the untreated group, the IC50 of both GR cell lines were significantly reduced following MH treatment (Figure 4A). Western blotting further revealed that MH downregulated the expression of drug resistance‐associated proteins MDR1 (multidrug resistance protein 1) and RRM2 (ribonucleotide reductase subunit M2) in GR cells, while upregulating the expression of drug sensitivity‐related proteins ENT1 (equilibrative nucleoside transporter 1) and DCK (deoxycytidine kinase) (Figure 4B). Comet assay, commonly used to assess DNA double‐strand breaks caused by chemotherapy or genotoxic agents [30]. To evaluate the impact of MH on GEM‐induced DNA damage, we performed comet assays in both WT and GR pancreatic cancer cells (Figure 4C) and measured the expression level of γ‐H2AX by western blotting (a DNA double‐strand break marker) (Figure S5). The results demonstrated that MH not only enhanced GEM's cytotoxic effects in WT cells but also restored GEM responsiveness in previously resistant GR cells. Statistical analysis of tail DNA percentage and tail moment confirmed these significant changes.

FIGURE 4.

MH partially restores the chemotherapeutic response to GEM in GR pancreatic cancer. (A) GR pancreatic cancer cells MIA GR and PANC GR were treated with MH, and subsequently assessed for changes in IC50 against GEM. (B) MH was used to treat PANC GR and MIA GR cells, and the GEM resistance/sensitivity related protein expression levels were detected by western blotting assay. (C) WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with GEM or MH combined with GEM, the comet assay was used to detect different cellular states in response to chemotherapy, and some of the key indicators were statistically analyzed (scale bar: 200 µm; n = 10, two‐way ANOVA). (D) The subcutaneous xenograft model mice with MIA GR cells were treated with GEM, MH, alone or in combination, or with the control (PBS). The tumor volumes and body weight were measured every 2 days until the experiment was terminated. (E) Representative images of the subcutaneous xenograft tumors. (F) The changes in tumor volume (n = 5, two‐way ANOVA). (G) The average weights of the tumors (n = 5, one‐way ANOVA). (H) The changes in body weight of mice (n = 5, two‐way ANOVA). (I) H&E, Prussian Blue, RRM2, and MDR1 staining of tumor tissues (scale bar: 200 µm). (J) A schematic diagram of MH exerts multi‐dimensional chemosensitizing effects on GEM chemotherapy in pancreatic cancer. ns: Not significant. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. PBS phosphate buffer saline; RRM2 ribonucleotide reductase M2; MDR1 multi‐drug resistance‐1.

For in vivo validation, a subcutaneous xenograft model of GR pancreatic cancer was established in nude mice (Figure 4D). GEM monotherapy failed to inhibit tumor growth due to intrinsic drug resistance, whereas MH alone exhibited notable antitumor efficacy. Strikingly, sequential treatment with MH followed by GEM achieved the most significant tumor suppression, suggesting a partial reversal of GEM resistance. No significant weight loss was observed in any treatment group throughout the experiment (Figure 4H). H and E and Prussian blue staining of tumor tissues further highlighted the superior therapeutic effect of the MH‐GEM sequential regimen. Immunohistochemistry confirmed a reduced expression of RRM2 and MDR1 in MH‐treated GR tumors (Figure 4I). Collectively, these findings demonstrate that MH exerts multi‐dimensional chemosensitizing effects (Figure 4J), synergizing with GEM in WT pancreatic cancer while alleviating resistance in GR tumors, thereby restoring their sensitivity to GEM therapy.

2.5. GR Pancreatic Cancer Cells Exhibit Heightened Sensitivity to Ferroptosis‐Inducing Therapy

To compare the transcriptomic profiles between GR and WT pancreatic cancer cells, high‐throughput transcriptome sequencing was performed on paired cell lines. Gene Set Enrichment Analysis (GSEA) revealed significant enrichment of GR cells in peroxisome‐related pathways, reactive oxygen species (ROS) signaling, oxidative phosphorylation, and fatty acid metabolism compared to WT cells (Figure 5A). Differential expression analysis identified 3982 differentially expressed genes (DEGs), with 1486 upregulated and 2496 downregulated in GR vs. WT cells (Figure 5B). Given the emerging role of ferroptosis in chemotherapy resistance across cancers [31, 32], we investigated its potential link to gemcitabine resistance in pancreatic cancer. The intersection of DEGs with ferroptosis‐related genes from the FerrDb V2 database (http://www.zhounan.org/ferrdb/current/) yielded 88 overlapping genes (Figure 5C). PPI analysis via STRING demonstrated robust functional connectivity among these genes (Figure 5D). Further screening identified key ferroptosis markers, including ACSL4 (acyl‐CoA synthetase long‐chain family member 4) and FTH1 (ferritin heavy chain 1), whose expression levels were notably suppressed in GR cells (Figure 5E), suggesting impaired ferroptosis signaling in resistant cells.

FIGURE 5.

GR pancreatic cancer cells exhibit heightened sensitivity to ferroptosis‐inducing therapy. Transcriptome sequencing was performed on the pancreatic cancer cell lines MIA WT and MIA GR (3 replicates each) and DEGs between the two cells were analyzed. (A) GSEA analysis of overall transcriptome genes (MSigDB Database). (B) Volcano map of DEGs. (C) Intersection of DEGs with ferroptosis‐related genes. (D) Protein interaction analysis of common DEGs (STRING database). (E) Heatmap of ferroptosis signaling gene expression in common DEGs. WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with Erastin, and (F) the apoptotic morphology of the cells was observed under light microscope (scale bar: 100 µm); (G) the activity and toxicity to the cells were labelled using Calcein/PI dye respectively (scale bar: 250 µm) and (H) the proportions of cells in different states were counted; (I) the proportion of apoptotic cells was detected using flow cytometry. DEGs differentially expressed genes; GSEA gene set enrichment analysis; Erastin: a ferroptosis inducer.

To validate these findings, GR and WT cells were treated with the ferroptosis inducer Erastin. Microscopic observations revealed accelerated apoptotic morphology in GR cells as early as 24 h post‐treatment, with pronounced differences observed by 48 h (Figure 5F). Calcein/PI staining and flow cytometry confirmed significantly enhanced cytotoxicity in GR cells compared to their WT counterparts (Figure 5G–I). Collectively, these results indicate that GR pancreatic cancer cells exhibit attenuated ferroptosis signaling, rendering them hypersensitive to ferroptosis‐inducing therapies.

2.6. MH Induces Increased Protein Levels of ACSL4 and Enhanced Ferroptosis in GR Pancreatic Cancer Cells

Previous studies have demonstrated that the accumulation of lipid peroxides is a hallmark of ferroptosis [33], with ACSL4 playing a pivotal role in catalyzing polyunsaturated fatty acids into lipid peroxides to drive this process. Building on this, we investigated ACSL4 in GR pancreatic cancer. Analysis of TCGA data revealed significantly elevated ACSL4 expression in pancreatic tumor tissues compared to normal tissues. GEO derived survival data further indicated that high ACSL4 expression correlated with poor patient prognosis (Figure 6A,B). To validate these findings, we constructed tissue microarrays using tumor specimens from GEM‐treated pancreatic cancer patients. IHC staining demonstrated markedly lower ACSL4 expression in GR patients compared to WT counterparts (Figure 6C), a finding consistent with results of the RT‐PCR assay (Figure S6). Further survival analysis revealed that patients with low ACSL4 expression exhibited significantly shorter overall survival (OS) and progression‐free survival (PFS) than those with high ACSL4 levels (Figure 6D). To explore the regulatory interplay between MH and ferroptosis, we treated GR and WT pancreatic cancer cells with MH. RT‐qPCR analysis showed no significant changes in ACSL4 mRNA levels post‐MH treatment. However, Western blotting revealed a striking upregulation of ACSL4 protein in GR cells (Figure 6E,F), suggesting post‐transcriptional regulation. Intracellular immunofluorescence confirmed enhanced ACSL4 protein fluorescence intensity in MH‐treated GR cells (Figure S7).

FIGURE 6.

MH induces increased protein levels of ACSL4 and enhanced ferroptosis in GR pancreatic cancer cells. (A) Expression levels of ACSL4 in pancreatic cancer and normal tissues (TCGA and GTEx database, n = 179 vs. 171, one‐way ANOVA). (B) Correlation between ACSL4 expression and prognosis of pancreatic cancer patients (GEO database, n = 145 vs. 134, Log‐rank test). The expression of ACSL4 were obtained by immunohistochemical staining based on pancreatic cancer tissues from our center, and its association with (C) GEM chemoresistance (scale bar: 400 µm; enlarged image: 100 µm; n = 16 vs. 30, unpaired t test) and (D) patient prognosis was examined (n = 16 vs. 30, Log‐rank test). WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with MH, and (E) the mRNA expression level of ACSL4 was detected by RT‐PCR assay (n = 3, two‐way ANOVA), (F) the protein expression level was detected by western blotting assay. (G) WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with MH alone, or in combination with Fer‐1, Z‐VAD, and NSA, and their cell viability were detected by CCK‐8 method (n = 3, two‐way ANOVA). WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with MH or MH combined with Fer‐1, and (H) the levels of intracellular ROS were detected by DCFH probe (scale bar: 250 µm); (I) the levels of intracellular MDA were detected by TBA assay (n = 3, two‐way ANOVA); (J) the levels of cellular oxidative C11 were detected by BODIPY probe (n = 4, two‐way ANOVA); (K) the expression levels of ferroptosis pathway‐related proteins were detected by western blotting assay. (L) WT and GR cell lines of Panc‐1 and MIA PaCa‐2 were treated with MH and intracellular mitochondrial morphology was observed by TEM (scale bar: 2 µm; enlarged image: 200 nm). ns: Not significant. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. TCGA The Cancer Genome Atlas; GEO Gene Expression Omnibus; ACSL4 acyl‐CoA synthetase long chain family member 4; Fer‐1: ferrostatin‐1, a ferroptosis inhibitor; Z‐VAD: an apoptosis inhibitor; NSA: necrosulfonamide, a necrotic inhibitor; ROS reactive oxygen species; MDA malondialdehyde.

We further assessed MH‐induced ferroptosis in GR cells using functional assays. Co‐treatment with the ferroptosis inhibitor Fer‐1, the necroptosis inhibitor NSA, or the apoptosis inhibitor Z‐VAD demonstrated that MH's anti‐proliferative effects on GR cells were specifically and partially reversed by Fer‐1 (Figure 6G). Calcein/PI staining corroborated MH's selective cytotoxicity toward resistant cells, similarly attenuated by Fer‐1 (Figure s8A,B). MH treatment significantly elevated ROS levels while reducing GSH in GR cells, effects partially rescued by Fer‐1 (Figure 6H; Figure S8C). Lipid peroxidation markers, including MDA and oxidized BODIPY C11, were markedly increased in MH‐treated GR cells, again reversible with Fer‐1 (Figure 6I,J). Given the link between mitochondrial dysfunction and ferroptosis [34], JC‐1 staining revealed MH‐induced mitochondrial membrane depolarization in GR cells, partially mitigated by Fer‐1. TEM confirmed mitochondrial ultrastructural damage, including cristae loss and membrane disruption, in MH‐treated GR cells (Figure 6L; Figure s8 D,E). Western blotting demonstrated MH‐mediated downregulation of ferroptosis suppressors (FTH1, GPX4, SLC7A11) and upregulation of ACSL4 in GR cells, with Fer‐1 partially reversing these effects (Figure 6K; Figure S9A–D). Collectively, these findings establish that MH preferentially induces ACSL4‐mediated ferroptosis in GR pancreatic cancer cells, providing a mechanistic basis for its differential therapeutic efficacy against WT and GR malignancies.

2.7. MH Inhibits Ubiquitination Degradation of ACSL4 Protein by Down‐Regulating HSPB1

To validate the transcriptional changes induced by MH treatment in pancreatic cancer GR cells, we performed transcriptome sequencing on MH‐treated vs. untreated GR cells. Differential expression analysis identified 810 DEGs, including 163 upregulated and 647 downregulated genes (Figure 7A). KEGG pathway analysis confirmed that MH treatment significantly upregulated the ferroptosis pathway in GR cells (Figure 7B). GO and GSEA analysis revealed marked alterations in mitochondrial metabolic and respiratory processes, as well as pathways associated with cellular responses to biotic stimuli and inflammatory reactions post‐MH intervention (Figure S10A,B). To elucidate the mechanism underlying MH‐induced ferroptosis enhancement, we intersected the ferroptosis gene database with the identified DEGs using a Venn diagram approach, identifying 32 overlapping genes (Figure 7C,D). Among these, HSPB1—a recently reported negative regulator of ferroptosis [35, 36], exhibited significant downregulation following MH treatment. Subsequent RT‐qPCR and Western blotting validated that MH treatment markedly suppressed both mRNA and protein expression of HSPB1 in GR cells (Figure 7E,F), while it did not significantly affect ACSL4 mRNA levels (Figure S11). Based on pathway changes, we hypothesized that HSPB1 inhibition occurs in response to nanomagnetic hyperthermia‐induced heat shock, subsequently modulating ACSL4 protein levels.

FIGURE 7.

MH inhibits ubiquitination degradation of ACSL4 protein by down‐regulating HSPB1. Transcriptome sequencing was performed on the MH‐treated or untreated PANC GR cells (3 replicates each) and DEGs between the two cells were analyzed. (A) Volcano map of DEGs. (B) Dot plot of KEGG analysis of DEGs. (C) Intersection of DEGs with ferroptosis‐related genes. (D) Heatmap of gene expression in common DEGs. MIA GR and PANC GR cells were treated with MH or MH combined with Fer‐1, and (E) the mRNA expression level of HSPB1 was detected by RT‐PCR assay (n = 3, two‐way ANOVA), (F) the protein expression level of HSPB1 and ACSL4 were detected by western blotting assay. (G) Protein interaction analysis of HSPB1 and ACSL4 (STRING database). (H) Correlation of HSPB1 and ACSL4 expression in pancreatic cancer tissues (TCGA database, n = 184, Chi‐squared test). Typical representative images of high and low expression of (I) HSPB1 and (J) ACSL4 were obtained by immunohistochemical staining based on pancreatic cancer tissues from our center (scale bar: 400 µm; enlarged image: 100 µm); (K) Statistical analyses of the expression relevance between the two molecules were assessed (n = 46, Chi‐squared test); (L) Correlation between HSPB1 expression and prognosis of patients with pancreatic cancer (n = 25 vs. 21, Log‐rank test). (M) The mRNA expression levels of HSPB1 were detected in two HSPB1 overexpressing cells PANC GR and MIA GR. Based on the two HSPB1 overexpressing cells PANC GR and MIA GR (n = 3, two‐way ANOVA), (N) the mRNA expression level of ACSL4 was detected by RT‐PCR assay (n = 3, two‐way ANOVA), (O) the protein expression level of HSPB1 and ACSL4 were detected by western blotting assay. (P) MIA GR and PANC GR were given corresponding treatments according to different groups of VECTOR (empty plasmid control), HSPB1 (overexpression), HSPB1 combined with MG132 or CQ, and intracellular protein synthesis was blocked by CHX at specific time points, then the expression levels of ACSL4 protein were detected by western blotting assay, and (Q) the ACSL4 protein half‐life of each group were analyzed. (R) VECTOR control and HSPB1 overexpressing cell lines of MIA GR and PANC GR were treated with MH, and ACSL4‐specific protein ubiquitination levels were detected using the IP‐ubiquitination assay. ns: Not significant. ^* p < 0.05, ^*** p < 0.001, ^**** p < 0.0001. HSPB1 heat shock protein family B member 1; MG132: a proteasome inhibitor; CQ: chloroquine, an autophagy inhibitor; CHX cycloheximide; IP immunoprecipitation.

To explore the HSPB1‐ACSL4 relationship, STRING database analysis predicted potential indirect interactions between the two proteins through intermediate molecules. TCGA data further demonstrated a significant negative correlation between HSPB1 and ACSL4 expression in pancreatic cancer tissues (Figure 7G,H). Immunohistochemical staining of tissue microarrays confirmed this inverse correlation (Figure 7I–K). Database analyses and our patient cohort data indicated that high HSPB1 expression correlated with a poorer prognosis in pancreatic cancer patients (Figure 7L; Figure S12). To delineate the HSPB1‐ACSL4 mechanism, we generated HSPB1‐overexpressing GR cell lines via plasmid transfection. RT‐qPCR confirmed successful HSPB1 overexpression but did not alter ACSL4 mRNA levels (Figure 7M,N). However, Western blotting demonstrated that HSPB1 overexpression significantly reduced ACSL4 protein levels (Figure 7O). Given HSPB1's role as a chaperone facilitating clearance of misfolded proteins, we postulated that HSPB1 regulates ACSL4 degradation. Using cycloheximide (CHX) to block protein synthesis, time‐course Western blotting showed accelerated ACSL4 degradation in HSPB1‐overexpressing cells, which was reversed by the proteasomal inhibitor MG‐132 but not by the lysosomal inhibitor chloroquine (CQ) (Figure 7P,Q). Cellular fluorescence assays revealed co‐localization of HSPB1 and ACSL4 (Figure S13), while IP‐ubiquitination assays demonstrated increased ubiquitinated ACSL4 and reduced ACSL4 protein levels under HSPB1 overexpression (Figure 7R). Collectively, these findings indicate that HSPB1, as a downstream effector of MH, modulates ACSL4 stability via ubiquitination‐mediated degradation, linking it to pancreatic cancer prognosis.

2.8. HSPB1 Overexpression Weakens Ferroptosis Induced by MH Treatment

To determine the role of HSPB1 overexpression on MH‐induced ferroptosis in GR pancreatic cancer cells, we treated HSPB1‐overexpressing and control GR cells with MH. Calcein/PI staining demonstrated that HSPB1 overexpression significantly inhibited MH‐mediated cytotoxicity in GR cells (Figure S14A). Using DCFH and JC‐1 probes to assess intracellular ROS levels and mitochondrial membrane potential, respectively, we found that HSPB1 overexpression partially suppressed MH‐induced ROS accumulation and mitochondrial membrane depolarization (Figure 8A; Figure S14B). Furthermore, HSPB1 overexpression markedly attenuated the MH‐triggered elevation of MDA and oxidized BODIPY C11 levels in GR cells (Figure 8B,C). Western blotting analysis of ferroptosis‐related proteins revealed that MH‐induced alterations in the expression of SLC7A11, GPX4, FTH1, and ACSL4 were partially reversed by HSPB1 overexpression (Figure 8D).

FIGURE 8.

HSPB1 overexpression weakens ferroptosis induced by MH treatment. VECTOR control and HSPB1 overexpressing cell lines of MIA GR and PANC GR were treated with MH, and (A) the levels of intracellular ROS were detected by DCFH probe (scale bar: 250 µm); (B) the levels of intracellular MDA were detected by TBA assay (n = 3, one‐way ANOVA); (C) the levels of cellular oxidative C11 were detected by BODIPY probe (n = 3, one‐way ANOVA); (D) the expression levels of ferroptosis pathway‐related proteins were detected by western blotting assay. (E) Mouse pancreatic orthotopic xenograft tumor models were constructed using PANC GR cells, and subsequently treated with PBS (negative control), GEM, MH alone, or in combination with Fer‐1/ML346. (F) Representative images of the pancreatic orthotopic xenograft tumors and (G) statistics of their volume (n = 5, one‐way ANOVA). Detection of (H) MDA (n = 6, one‐way ANOVA) and (I) total Fe levels in serum (n = 5, one‐way ANOVA). (J) H&E, Prussian Blue, Ki‐67, and HSPB1 staining of tumor tissues (scale bar: 200 µm). (K) Immunohistochemical staining of tumor tissues for ACSL4, FTH1, GPX4, 4‐HNE, MDR1, and RRM2 (scale bar: 200 µm). ns: Not significant. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. ML346: a HSPB1 pro‐expression agent. FTH1 ferritin heavy chain 1; GPX4 glutathione peroxidase 4; 4‐HNE 4‐hydroxynonenal.

To validate these findings in vivo, we established orthotopic xenograft models in C57 mice using PANC GR cells and administered designated treatments (Figure 8E). MH treatment significantly suppressed GR pancreatic tumors, which were partially reversed by the ferroptosis inhibitor Fer‐1 and the HSPB1 inducer ML346 (Figure 8F,G). Serum analysis showed that MH significantly increased MDA levels but had no significant effect on total iron content, whereas both Fer‐1 and ML346 reduced MDA and iron levels to varying degrees (Figure 8H,I). Histopathological examination revealed disorganized tumor architecture with apoptotic/necrotic cells, iron deposition, and downregulated Ki‐67/HSPB1 expression in MH‐treated tumors. ML346 upregulated HSPB1 expression and counteracted MH's therapeutic effects similarly to Fer‐1. Immunohistochemistry further confirmed that the enhanced ferroptosis signaling and amelioration of chemoresistance induced by MH treatment were concomitantly attenuated by both Fer‐1 and ML346 (Figure 8J,K).

As a nanomagnetic hyperthermia modality not yet widely applied clinically, the in vivo safety profile of MH requires rigorous evaluation. To systematically assess the biosafety of MH, we performed H&E staining and Prussian blue iron staining on major organs from all treatment groups (Figure S15A,B). The results demonstrated intact architecture and clear morphology across all organs, with moderately elevated iron deposition observed in the spleen and liver. Furthermore, serum analysis of hepatic function markers (ALT, AST) and renal parameters (CRE, BUN) revealed no significant fluctuations in these biochemical indices despite therapeutic interventions (Figure S15C–F). Notably, longitudinal monitoring of body weight changes showed no statistically significant intergroup differences, confirming maintained physiological stability during treatment (Figure S15G,H). Collectively, our findings demonstrate that HSPB1 suppression is essential for MH‐enhanced ferroptosis in GR pancreatic cancer. MH exerts its therapeutic effects via the HSPB1‐ACSL4‐ferroptosis axis while maintaining favorable biosafety, highlighting its potential as a novel therapeutic strategy to overcome chemoresistance in pancreatic malignancies.

3. Discussion

Pancreatic cancer, a malignant solid tumor with a notoriously poor prognosis, frequently develops chemoresistance to conventional agents such as GEM, leading to unfavorable clinical outcomes. In this study, we employed a novel iron oxide nanoparticle‐FVIOs as a therapeutic carrier, leveraging its exceptional heat‐generating properties under AMF exposure to achieve precise nanomagnetic hyperthermia (MH) in pancreatic cancer. Over the past few decades, nanotechnology has been extensively explored to develop novel and advanced diagnostic and therapeutic tools in cancer research [37, 38]. As a nanoscale intra‐tumoral hyperthermia approach, MH has already transitioned into clinical practice, being approved for recurrent glioblastoma and currently under investigation in prostate cancer trials [39, 40]. Our study demonstrates that MH significantly enhances GEM chemosensitization in pancreatic cancer and partially overcomes GEM chemoresistance, supporting its clinical translation.

As an established MH mediator, FVIOs have shown therapeutic potential in other malignancies. Liu et al. [28]. reported that FVIOs‐mediated MH enhanced phagocytic clearance of breast cancer cells and sensitized tumors to PD‐L1 immunotherapy. Similarly, Tang et al. [29]. revealed that FVIOs‐mediated MH elevated intracellular ROS levels in hepatocellular carcinoma, promoting lipid peroxide accumulation and GPX4 inactivation to exacerbate ferroptosis. These findings underscore the broad applicability of FVIOs‐mediated MH. In pancreatic cancer, however, gemcitabine resistance remains a major clinical challenge [41]. Consequently, identifying effective treatments for GR pancreatic cancer remains an urgent clinical dilemma. Our study systematically evaluated the potential of MH against GR pancreatic cancer.

Notably, we found that GR pancreatic cancer exhibited a superior response to MH therapy compared to WT counterparts, with significant reversal of GEM chemoresistance. Beyond this finding, other mechanisms—such as cellular lipid transport/metabolic networks [42], and the disruption of S and G2/M cell cycle checkpoints [43]—can also be targeted to overcome gemcitabine chemoresistance. Mechanistically, previous studies have revealed that heat generated by iron oxide magnetic nanoparticles disrupts the stromal barrier—a key factor in pancreatic cancer chemoresistance [44]. More importantly, MH therapy induces distinct cell death pathways, particularly ferroptosis—an iron‐dependent programmed cell death mechanism, thereby overcoming GEM resistance in pancreatic cancer [45]. Transcriptomic sequencing revealed substantially lower basal ferroptosis levels in GR pancreatic cancer cells relative to WT cells. Intriguingly, MH induced significantly more pronounced ferroptosis in GR cells, which may represent the core mechanism underlying the differential therapeutic efficacy observed.

We further identified ACSL4 as a pivotal regulator of MH‐induced ferroptosis. Commonly, ACSL4 activates fatty acids and promotes lipid peroxidation, playing an essential role in ferroptosis [46]. Notably, certain metastatic and chemoresistant cancers demonstrate heightened susceptibility to ferroptosis inducers [47, 48]. Mechanistically, MH rapidly elevates intracellular thermal energy and ROS levels, activating ferroptosis signals [49]. The dual action of hyperthermia and ROS‐mediated oxidative stress also promotes immunogenic cell death [50], highlighting MH's multimodal potential.

Central to this process, MH downregulates the heat stress effector HSPB1 (HSP27). HSPB1 functions as an ATP‐independent molecular chaperone, binding misfolded proteins and directing them toward degradation [51]. Its knockout amplifies oxidative stress sensitivity [52] while its overexpression maintains redox homeostasis [53]. Importantly, elevated HSPB1 predicts GEM chemoresistance, and its knockdown enhances GEM sensitivity [54]. In breast cancer, HSPB1 shields cells from drug‐induced ferroptosis [36]. Through bioinformatic analysis and experimental validation, we identified an inverse correlation between HSPB1 and ACSL4 expression. IP‐ubiquitination assays confirmed that HSPB1 regulates ACSL4 ubiquitination levels. Given that ubiquitination‐dependent protein degradation influences tumor progression [55], the “HSPB1‐ACSL4” axis constitutes a core mechanism by which MH potentiates ferroptosis and reverses chemoresistance in GR pancreatic cancer.

While this study provides critical insights into MH‐mediated ferroptosis and chemosensitization, several mechanistic details warrant further exploration. Specifically, the upstream mechanisms by which MH suppresses HSPB1 expression remain unclear. Additionally, MH's direct molecular targets in reversing chemoresistance in GR pancreatic cancer need to be identified. Beyond ferroptosis, MH may exert immunomodulatory effects in pancreatic cancer [56, 57]. Despite the current limitations of immunotherapy in pancreatic cancer, magnetic nanoparticles like FVIOs, with intrinsic magnetic responsiveness and biocompatibility, can enhance tumor immune responses [58]. Also, it may hold particular advantages for magnetic resonance imaging (MRI), especially in the non‐invasive diagnosis of small liver tumors and other pathologies [59]. Unlike conventional hyperthermia, MH‐induced endogenous heating may more effectively trigger immunogenic cell death. Our future studies will focus on evaluating FVIOs’ ability to confer antigen specificity to pancreatic cancer cells, amplify T‐cell effector functions, and neutralize immunosuppressive factors within the tumor microenvironment, potentially advancing pancreatic cancer immunotherapy.

In summary, our in vitro and in vivo experiments demonstrate that FVIO‐mediated nanomagnetic hyperthermia multi‐dimensionally enhances GEM chemosensitivity in pancreatic cancer (Figure 9). Briefly, in WT pancreatic cancer, MH synergizes with GEM to suppress tumor progression. In GR pancreatic cancer, MH exerts direct cytotoxic effects: 1) it restores GEM chemo‐responsiveness in GR cells, and 2) it downregulates HSPB1, thereby inhibiting ACSL4 ubiquitination‐dependent degradation, which in turn amplifying ferroptosis. Through these mechanisms, MH significantly potentiates GEM efficacy and offers a promising strategy to improve clinical outcomes in pancreatic cancer therapy.

FIGURE 9.

Schematic diagram of the mechanism by which MH therapy mediated by FVIOs sensitizes GEM chemotherapy in pancreatic cancer. In this study, FVIOs demonstrated therapeutic potential by enabling intracellular hyperthermia (MH) under AMF exposure following cellular uptake, and the therapeutic mechanisms varied depending on the state of pancreatic cancer cells. In WT pancreatic cancer cells, MH exhibited synergistic antitumor effects when combined with GEM chemotherapy. For GR pancreatic cancer cells, MH operated through dual mechanisms: first, by restoring chemosensitivity to GEM; and second, by downregulating HSPB1 expression, which thus reduced the ubiquitin‐mediated degradation of ACSL4 protein. The elevated ACSL4 levels consequently enhanced ferroptosis through increased lipid peroxidation. The chemosensitizing effects of MH in GEM‐based pancreatic cancer treatment were notably validated in both in vitro and in vivo assays. PCC pancreatic cancer cells; Ub ubiquitin.

4. Materials and Methods

4.1. Cell Culture

The cell lines used in this study included the human pancreatic cancer cell lines Panc‐1 (PANC), MIA PaCa‐2 (MIA), and the human normal pancreatic ductal epithelial cell line hTERT‐HPNE (HPNE). Additionally, gemcitabine‐resistant pancreatic cancer cell lines, PANC GR and MIA GR, were established from their respective parental cell lines, PANC wild‐type (WT) and MIA WT, through continuous low‐dose GEM induction using a stepwise concentration escalation protocol [60, 61]. All cells were maintained in DMEM medium (Gibco, USA) supplemented with 10% FBS and cultured at 37°C in a humidified atmosphere containing 5% CO₂. Routine subculturing was performed according to standard protocols.

4.2. Characterization of FVIOs and MH

The FVIOs utilized in this study were synthesized by the Laboratory of Magnetic Nanobiomaterials and Nanomedicine at Northwestern University, a collaborating research group. The microstructure and subcellular distribution of the FVIOs were characterized using transmission electron microscopy (TEM) (JEOL 100CX). The hydrodynamic size and zeta potential of the FVIOs were measured with a Malvern Zeta‐sizer Nano ZS. Real‐time heat generation by the FVIOs under an AMF was monitored using a remote infrared thermal imaging system, with temperature change profiles plotted accordingly.

4.3. Cell Viability Assay

Cell viability was assessed using the Cell Counting Kit‐8 (CCK‐8) (GlpBio, USA) assay. After respective treatments, 10 µL of CCK‐8 solution was added to each well, and the cells were incubated at 37°C for approximately 30 min. The optical density (OD) values of each well were then measured at 450 nm using a spectrophotometer. The half‐maximal inhibitory concentration (IC50) of the drug was calculated using GraphPad Prism 9.1 software (USA). Cell viability under MH and GEM interventions was measured separately, and the synergy scores of the two treatments were evaluated using response surface methodology and the highest single agent (HSA) method [62]. A synergy score >10 was defined as indicating strong synergistic effects.

4.4. Cell Proliferation and Apoptosis Detection

The tumorigenic proliferative capacity of the cells was evaluated using a colony formation assay. Following corresponding interventions, single cells were cultured for approximately 2 weeks, then fixed and stained with crystal violet. Colonies were observed and counted under a microscope. Apoptosis was analyzed by flow cytometry. Treated cells were labeled with fluorescent dyes from an apoptosis detection kit (BD Biosciences, USA) and assessed using a flow cytometer (Agilent, USA) to determine the apoptotic cell ratio. Cellular viability/toxicity was examined using a live/dead assay: treated cells were loaded with Calcein AM/PI dual fluorescence probes, incubated in the dark for 30 min, and then imaged under a fluorescence microscope.

4.5. Western Blotting

Total proteins were extracted from pancreatic cancer cells using a Total Protein Extraction Kit (Beyotime, China). Protein concentrations were determined with a BCA Protein Assay Kit (CUSBio, China). The processed protein samples were separated by SDS‐PAGE and transferred onto PVDF membranes (Millipore, USA). The membranes were blocked with a rapid blocking buffer (YamayBio, China) to prevent nonspecific binding and then incubated overnight at 4°C with primary antibodies diluted according to the manufacturers’ instructions, including: Bcl‐2 (12789‐1‐AP), Bax (50599‐2‐Ig), Cyclin‐D3 (26755‐1‐AP), PCNA (10205‐2‐AP), MDR1/P‐gp (22336‐1‐AP), ENT1 (29862‐1‐AP), RRM2 (11661‐1‐AP), DCK (17758‐1‐AP), ACTIN (66009‐1‐Ig), GAPDH (10494‐1‐AP), ACSL4 (22401‐1‐AP), SLC7A11 (26864‐1‐AP), GPX4 (67763‐1‐Ig), FTH1 (11682‐1‐AP), HSPB1 (18284‐1‐AP), and ubiquitin (10201‐2‐AP) (all from Proteintech, China). After washing with TBST, the membranes were incubated with HRP‐conjugated secondary antibodies (SA00001‐1 or SA00001‐2, Proteintech, China) for 1 h at room temperature on a shaker. Following additional washes, protein signals were visualized using an ultrasensitive ECL chemiluminescent reagent (NCMBio, China) on a ChemiDoc XRS+ system (Bio‐Rad, USA). β‐actin (ACTIN) or GAPDH served as the loading control.

4.6. Comet Assay

GEM treated pancreatic cancer cells were collected and mixed with low‐melting‐point agarose (0.7%), then pipetted onto pre‐coated slides layered with normal‐melting‐point agarose (1%). A coverslip was applied to form a uniform thin layer, and the slides were solidified at 4°C. After removing the coverslip, the slides were immersed in pre‐cooled lysis buffer for 2 h at 4°C under light‐protected conditions. Post‐lysis, slides were incubated in alkaline electrophoresis buffer for 30 min to allow DNA unwinding. Electrophoresis was performed at 1 V/cm and 300 mA for 30 min. Subsequently, the slides were neutralized with a Tris‐HCl buffer (pH 7.5) and stained with propidium iodide (PI) for 10 min in the dark. Fluorescent microscopy was employed to visualize and capture comet images for the analysis of DNA damage.

4.7. High‐Throughput Sequencing

High‐throughput transcriptome sequencing was performed by Novogene (Beijing, China). Briefly, RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Poly(A)‐enriched mRNA was isolated using oligo‐dT magnetic beads, followed by fragmentation, double‐stranded cDNA synthesis, end repair, and adapter ligation to construct sequencing libraries according to standard protocols. Library quantification was carried out with a Qubit 3.0 Fluorometer (Thermo Fisher Scientific, USA). Finally, paired‐end sequencing (2 × 150 bp) was conducted on the Illumina NovaSeq 6000 platform (Illumina, USA).

4.8. Bioinformatics Analysis

Gene expression clustering analysis was performed using Gene Set Enrichment Analysis (GSEA) to identify expression patterns of functionally related gene clusters. Protein‐protein interaction (PPI) networks were constructed via the STRING database (https://string‐db.org), and functional associations among target proteins were elucidated. Transcriptomic data and clinical information from patients with pancreatic ductal adenocarcinoma (PDAC) were obtained from The Cancer Genome Atlas (TCGA) PDAC project to evaluate the expression levels of specific target molecules and their prognostic relevance. Visualization of the analytical results, including volcano plots, heatmaps, Venn diagrams, and bubble plots, was implemented using R software packages (version 4.3.0; R Foundation) to enhance interpretability.

4.9. Human Tissue Samples and Ethics Statement

Pancreatic cancer tissues and adjacent normal pancreatic tissues were obtained from 46 patients diagnosed with primary pancreatic ductal adenocarcinoma (PDAC) who underwent surgical resection at the Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi'an Jiaotong University (China). All patients received postoperative adjuvant chemotherapy with a gemcitabine‐based regimen, and their chemotherapy response profiles, along with prognostic clinical data, were systematically collected. This study was approved by the Ethics Committee of The First Affiliated Hospital of Xi'an Jiaotong University (No. XJTU1AF2021LSK‐457). Tissue specimens were orientationally labeled by senior pathologists, and selected regions were processed into tissue microarrays (TMAs) by a certified biotechnology company (Servicebio, China). Hematoxylin and eosin (H and E) staining and immunohistochemical (IHC) staining were subsequently performed. Kaplan–Meier (K–M) survival curves were generated by integrating patient prognostic data to evaluate clinical outcomes.

4.10. RT‐qPCR

Total RNA was extracted from pancreatic cancer cells using a dedicated RNA isolation kit (Fastagen, China) according to the manufacturer's protocol. cDNA was synthesized from the extracted RNA using a reverse transcription kit (Takara, Japan). RT‐qPCR was performed using SYBR Green‐based reaction mixtures, following the instructions provided by the manufacturer (TSINGKE, China). GAPDH was utilized as an internal control for normalization, and relative gene expression levels were calculated using the 2^−ΔΔCt method. The primer sequences used in this study were as follows: ACSL4 (Forward:5′‐CATCCCTGGAGCAGATACTCT‐3′; Reverse:5′‐TCA CTTAGGATTTCCCTGGTCC‐3′), HSPB1(Forward:5′‐ACGGTCAAGACCAAGGA TGG‐3′; Reverse: 5′‐AGCGTGTATTTCCGCGTGA‐3′), GAPDH (Forward:5′‐GAAA GCCTGCCGGTGACTAA‐3′; Reverse: 5′‐GCCCAATACGACCAAATCAGAG‐3′).

4.11. Analysis of Ferroptosis

Ferroptosis in pancreatic cancer cells was evaluated by measuring reactive oxygen species (ROS), reduced glutathione (GSH), malondialdehyde (MDA), oxidized lipids, mitochondrial membrane potential, ultrastructural changes of mitochondrial, and the expression levels of ferroptosis markers. Intracellular ROS levels were measured using the DCFH‐DA probe (Beyotime, China) following the manufacturer's instructions. Briefly, after designated treatments, cells were incubated with DCFH‐DA in the dark at 37°C for 30 min. The fluorescence intensity of DCF was then quantified via flow cytometry. The levels of reduced glutathione (GSH) and malondialdehyde (MDA) were determined using commercial assay kits (GSH: Beyotime, China; MDA: NjjcBio, China) according to the provided protocols. For oxidized lipid detection, cells were stained with BODIPY 581/591 C11 dye (Thermo Fisher Scientific, USA) at 37°C for 30 min under light‐protected conditions, followed by flow cytometric analysis of fluorescence intensity. Mitochondrial membrane potential was assessed using the JC‐1 fluorescent probe (Absin, China). Cells were incubated with JC‐1 (diluted in assay buffer as per instructions) at 37°C for 30 min, and fluorescence images were captured using a fluorescence microscope. Mitochondrial ultrastructure was observed by TEM. Protein expression levels of ferroptosis markers (e.g., GPX4, FTH1, SLC7A11) were analyzed by Western blotting. All experiments were performed in triplicate.

4.12. Immunofluorescence

Treated cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.01% Triton X‐100 for 10 min, and blocked with 5% bovine serum albumin (BSA) for 1 h at room temperature. Primary antibodies against ACSL4 (Abcam, ab155282) and HSPB1 (Proteintech, 18284‐1‐AP) were diluted according to the manufacturers’ instructions and incubated with the cells overnight at 4°C. The next day, the cells were washed and incubated with CoraLite488‐ or CoraLite594‐conjugated anti‐rabbit IgG secondary antibodies (Proteintech, SA00013‐2, SA00013‐4) at room temperature for 1 h under light‐protected conditions. Nuclei were counterstained with DAPI (Biosharp, China) for 5 min. Fluorescence images were captured using a fluorescence microscope.

4.13. Plasmids Transfection

The HSPB1 overexpression plasmid and the corresponding empty VECTOR plasmids were synthesized by MiaoLing Biotech (Wuhan, China). Transfection was performed using the Lipo8000 transfection reagent (Beyotime, China) following the manufacturer's protocol. GFP‐tagged plasmids enabled real‐time monitoring of transfection efficiency under a fluorescence microscope. At 48 h post‐transfection, cells were selected with 2 µg/mL puromycin for 7 days to establish stable overexpression lines.

4.14. Protein Half‐Life Detection

Cells from different groups, VECTOR, HSPB1 (overexpression), HSPB1 plus MG‐132 (20 µm), and HSPB1 plus CQ (50 µm) were treated with 100 µm cycloheximide (CHX) to inhibit de novo protein synthesis. Total cellular proteins were extracted at 0, 1.5, 3, 6, 12, and 24 h post‐CHX treatment. The expression levels of ACSL4 protein were analyzed using Western blotting, as previously described. Band intensities were quantified using ImageJ software (NIH, USA), and the relative protein expression at each time point was normalized to the baseline (0 h, untreated control). Protein half‐life was calculated using nonlinear regression analysis in GraphPad Prism 9.0 (GraphPad Software, USA).

4.15. Immunoprecipitation‐Ubiquitination Assay

Cells were divided into predefined groups and subjected to corresponding treatments. After 24 h, the cells were treated with 10 µm MG132 for 6 h. Total proteins (antigens) were extracted using a proprietary IP lysis buffer. The antigens were then incubated overnight at 4°C on a rotating shaker with magnetic beads conjugated to an anti‐ACSL4 antibody (22401‐1‐AP, Proteintech) to form antigen‐antibody complexes. The beads were collected for immunoprecipitation (IP), washed three times with the lysis buffer, and the bound proteins were heat‐denatured in loading buffer. Ubiquitination levels of ACSL4 were analyzed by Western blotting using an anti‐ubiquitin antibody (10201‐2‐AP, Proteintech).

4.16. Animal Experiments

Female BALB/c nude mice (4 weeks old, 18–22 g) and C57BL/6 mice (4 weeks old, 20–24 g) were obtained from GemPharmatech (Jiangsu, China) to establish subcutaneous and orthotopic pancreatic cancer xenograft models, respectively. For the subcutaneous model, Panc‐1 and MIA GR cells (2 × 10⁷ cells) were mixed 1:1 (v/v) with pre‐chilled Matrigel (Corning, USA) and injected into the right inguinal region of the nude mice. For the orthotopic model, MIA GR and PANC GR cells (2 × 10⁶ cells) suspended in Matrigel were injected into the pancreatic body of the C57BL/6 mice. Each group included five mice. Tumor volume and body weight were monitored weekly post‐implantation, followed by designated treatments. After the treatments, the mice were anesthetized, and whole blood was collected via the orbital artery to isolate serum. The mice were then euthanized, and various tissues (tumor, pancreas, heart, spleen, liver, lung, and kidney) were excised. Tumor tissues were measured, weighed, and fixed in 4% paraformaldehyde for further analysis. Serum levels of total iron (Fe), malondialdehyde (MDA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (CRE), and blood urea nitrogen (BUN) were quantified using proprietary assay kits according to the manufacturer's protocols. All animal experiments were approved by the Biomedical Ethics Committee of Xi'an Jiaotong University Health Science Center (No. XJTUAE2023‐2224).

4.17. Tissue Staining

Tissues harvested from mice or patients were fixed in 4% paraformaldehyde, dehydrated through a graded ethanol series, cleared in xylene, and embedded in paraffin. Sections (4–5 µm) were cut and mounted onto glass slides. Before staining, the slides were deparaffinized and rehydrated. For H and E staining, the sections were sequentially stained with hematoxylin and eosin; For Prussian blue staining, the slides were treated with a specific staining solution (Yeasen, China), followed by a DAB‐enhanced chromogenic reaction and hematoxylin counterstaining; For IHC staining, antigen retrieval and blocking were performed before overnight incubation at 4°C with primary antibodies diluted in a working solution (prepared as per manufacturer's instructions). Sections were then incubated with secondary antibodies, developed with DAB, and counterstained with hematoxylin. All stained slides were dehydrated, cover‐slipped, and subjected to microscopic examination and evaluation.

4.18. Statistical Analysis

Statistical analysis was performed using GraphPad Prism 9.0 (GraphPad Software, USA) and SPSS 22.0 (IBM, USA). Data were expressed as mean ± standard deviation. A Student's t‐test was used to assess differences between two groups, while one/two‐way analysis of variance (ANOVA) was applied for comparisons among three or more groups. The Log‐rank test is employed for survival analysis. A p‐value < 0.05 was considered statistically significant.

Author Contributions

Jiaqiang Ren conducted the majority of the experiments, analyzed and visualized the data, and authored the manuscript; Shuai Wu designed the experiment and the manuscript, contributing numerous ideas and providing technical guidance; Tong Su, Hugang Li, Jiachun Ding, Fan Chen, Jiantao Mo, and Jie Li participated in data analysis and organized the figures; Zheng Wang and Liang Han supervised the experiments and revised the manuscript; Zheng Wu reviewed the final manuscript and oversaw the submission process. All authors read and approved the final version of the manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (82400759); Generic Technology Research and Development Project of Shaanxi Province (2023GXJS‐01‐2); Institutional Foundation of The First Affiliated Hospital of Xi'an Jiaotong University (2024‐MS‐14); Clinical research project of the First Affiliated Hospital of Xi'an Jiaotong University (XJTU1AF‐CRF‐2022‐034). The authors are deeply grateful to Professor Xiaoli Liu and her research team for their generous provision of nanoparticles, experimental instruments, and related guidance.

Ethics Approval and Consent to Participate

All procedures involving animals in this research were conducted in accordance with the guidelines and protocols approved by the Animal Ethics Committee of the Health Science Center of Xi'an Jiaotong University (No. XJTUAE2023‐2224).

Consent for Publication

All authors are fully aware of the authorship of the paper and consent to its publication.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll72890‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.