Syringaresinol Attenuates Aging‐Associated Ferroptosis‐Relevant Stress Through an HIF‐1α–GPX4 Defense Axis

ABSTRACT

Ferroptosis contributes to aging‐associated functional decline, yet compounds with robust organismal efficacy and defined upstream regulatory mechanisms remain limited. Here, we established a diethyl maleate (DEM)‐induced glutathione depletion model in wild‐type (N2) Caenorhabditis elegans as a survival‐based screening platform and identified syringaresinol (Syr) as a leading hit from an in‐house small‐molecule library. In nematodes, Syr improved survival under DEM challenge, reduced lipid peroxidation, reactive oxygen species (ROS), and malondialdehyde levels, and alleviated age‐associated oxidative lipid stress and iron imbalance during natural aging, accompanied by extended lifespan and improved healthspan‐related phenotypes. In primary human foreskin fibroblasts, Syr conferred dose‐dependent protection against RSL3‐ or erastin‐induced ferroptosis, preserved cellular integrity, suppressed lipid peroxidation and ROS, and restored expression of GPX4, SLC7A11, and ferritin. In two senescence models, Syr also attenuated senescence‐associated phenotypes and ferroptosis‐related oxidative lipid stress, concomitant with recovery of GPX4 expression. Network‐based prediction and functional perturbation identified HIF‐1α as a candidate mediator of Syr‐associated cytoprotection. HIF‐1α knockdown weakened Syr‐mediated protection and largely prevented GPX4 restoration, whereas GPX4 knockdown did not alter HIF‐1α abundance. These findings support a functional HIF‐1α–GPX4 defense axis in fibroblasts, while direct transcriptional regulation remains to be clarified. Overall, Syr attenuates ferroptosis‐relevant oxidative lipid stress and aging‐associated phenotypes in C. elegans and human fibroblast models, supporting further mechanistic and mammalian in vivo validation.

A 220‐compound screen in DEM‐challenged C. elegans identified syringaresinol (Syr) as a lead protective compound. Syr alleviated aging‐ and ferroptosis‐associated phenotypes in C. elegans and human foreskin fibroblasts by reducing lipid peroxidation, ROS accumulation, MDA levels, labile Fe²⁺, cell death, and SA‐β‐gal activity, while restoring GPX4/SLC7A11/ferritin‐associated defense responses. The proposed mechanism suggests that Syr attenuates senescence‐associated functional decline through a HIF‐1α‐associated GPX4 protective axis. The dashed arrow denotes a hypothesized mediator or interaction that remains to be experimentally validated.

Abbreviations

AD

Alzheimer's disease

C11‐BODIPY 581/591

boron‐dipyrromethene 581/591

CCK‐8

Cell Counting Kit‐8

C. elegans

Caenorhabditis elegans

DEM

diethyl maleate

D‐gal

D‐galactose

DHE

dihydroethidium

FBS

fetal bovine serum

Fer‐1

ferrostatin‐1

FTL

ferritin light chain

FUdR

5‐fluoro‐2′‐deoxyuridine

GO

gene ontology

GPX4

glutathione peroxidase 4

GSH

glutathione

HFFs

human foreskin fibroblasts

HIF‐1α

hypoxia‐inducible factor‐1α

KEGG

Kyoto Encyclopedia of Genes and Genomes

Lip‐1

liproxstatin‐1

MDA

malondialdehyde

MMP

matrix metalloproteinase

PGSK

PhenGreen SK

PPI

protein–protein interaction

qRT‐PCR

quantitative real‐time polymerase chain reaction

Rap

rapamycin

ROS

reactive oxygen species

SA‐β‐gal

senescence‐associated β‐galactosidase

SLC7A11

solute carrier family 7 member 11

Syr

syringaresinol

TNF‐α

tumor necrosis factor‐α

WT

wild type

1. Introduction

Aging is the predominant risk factor for most chronic diseases and is characterized by a progressive decline in cellular homeostasis. In addition to canonical forms of regulated cell death such as apoptosis and necroptosis, ferroptosis has emerged as a distinct mode of cell death driven by iron‐dependent phospholipid peroxidation and failure of GPX4‐centered antioxidant defense [1]. In the classical model, depletion of glutathione (GSH), the essential cofactor required for GPX4 activity, or direct inhibition of GPX4 impairs the detoxification of lipid peroxides, thereby leading to uncontrolled lipid oxidation, amplification of reactive oxygen species (ROS), and ferroptotic damage [2]. In parallel with these mechanisms, aged tissues commonly exhibit reduced redox buffering capacity, disturbed iron homeostasis, and lipid remodeling, all of which increase vulnerability to iron‐dependent lipid peroxidation and ferroptosis‐like injury [3]. In Caenorhabditis elegans ( C. elegans ), age‐associated increases in ferrous iron and alterations in glutathione homeostasis have been shown to promote ferroptosis‐like damage and frailty, supporting the concept that ferroptosis‐linked oxidative lipid stress is mechanistically integrated into aging trajectories rather than simply representing a passive consequence of physiological decline [4]. This view has been further strengthened by recent evidence showing that pharmacological inhibition of ferroptosis delays aging and extends healthspan across multiple species [5].

Despite rapid advances in ferroptosis research, the identification of ferroptosis inhibitors with robust in vivo efficacy remains a major challenge [6]. Many candidate compounds prioritized in reductionist cell‐based assays fail to translate across different stress conditions, tissues, or species, in part because conventional culture systems do not adequately capture compound uptake, metabolism, and organism‐level stress physiology [7]. Whole‐organism discovery platforms can complement these approaches by integrating such variables into phenotype‐based outputs, thereby enabling the prioritization of compounds with demonstrable protective activity under physiological constraints. C. elegans is particularly well suited for this purpose because it enables scalable and quantitative assessment of survival and functional phenotypes while retaining key evolutionarily conserved stress‐response pathways [8]. Notably, diethyl maleate (DEM) is a classical thiol‐reactive electrophile that depletes glutathione through conjugation chemistry, often mediated by glutathione S‐transferases, and is widely used to impose a controlled glutathione stress state [9]. This feature provides a defined oxidative window that can be exploited for organism‐level discovery of compounds with ferroptosis‐relevant protective activity [4].

Syringaresinol (Syr) is a naturally occurring lignan widely distributed in medicinal and edible plants, including Panax ginseng berry, sesame, Magnolia officinalis, Eucommia ulmoides , and other lignan‐rich botanical sources. Syr has been reported to exert antioxidant and anti‐inflammatory effects and to modulate aging‐associated phenotypes in mammalian systems, including oxidative stress‐induced skin aging and age‐related functional decline [10]. It has also been implicated in ferroptosis‐associated injury, with reported effects on endogenous lipid peroxide detoxification and GPX4‐centered defense pathways [11]. These findings are consistent with broader evidence that natural products and traditional medicine‐derived interventions can regulate ferroptosis‐related pathology in neurodegenerative disorders, including Alzheimer's disease‐related models [12, 13]. However, the scope and mechanistic organization of Syr‐mediated ferroptosis defense in aging‐relevant contexts remain insufficiently defined. In particular, it is unclear whether Syr can be identified and prioritized through an unbiased whole‐organism screening strategy and subsequently validated as a conserved anti‐ferroptotic activity across evolutionarily distant systems. It also remains unresolved how Syr exposure is hierarchically linked to GPX4‐centered defense reinforcement in senescence‐prone mammalian cells.

Hypoxia‐inducible factor‐1α (HIF‐1α) is a central regulator of metabolic adaptation and stress responses, with context‐dependent roles in both homeostasis and disease [14]. Emerging evidence suggests that HIF‐1α can intersect with ferroptosis regulation in specific pathological settings, including through reported transcriptional control of GPX4 in inflammatory epithelial contexts [15]. These observations raise the possibility that an HIF‐1α‐associated defense module may provide an upstream regulatory logic linking small‐molecule exposure to GPX4‐dependent ferroptosis resistance [16]. Whether such a hierarchy operates in aging‐relevant mammalian senescence states, and whether it is functionally required for natural product‐associated cytoprotection remains largely unknown.

Here, we established a DEM‐based glutathione depletion paradigm in wild‐type (N2) C. elegans as a whole‐organism screening platform for ferroptosis‐relevant protective compounds. Using this system, we identified Syr as a protective candidate from an in‐house small‐molecule library and subsequently validated its activity in nematode oxidative lipid stress and natural aging paradigms, as well as in ferroptosis and senescence models in primary human fibroblasts. We further examined whether Syr‐associated cytoprotection involves an HIF‐1α‐dependent regulatory relationship linked to restoration of GPX4 abundance and suppression of oxidative lipid stress. Together, these findings position Syr within an aging‐relevant ferroptosis defense framework and support a functional HIF‐1α–GPX4 defense axis associated with protection against aging‐related cellular decline.

2. Materials and Methods

2.1. Biological Materials

Human foreskin fibroblasts (HFFs) were kindly provided by Dr. Pei Xu, Affiliated Hospital of Southwest Medical University. The wild‐type C. elegans N2 was obtained from the Caenorhabditis Genetics Center (CGC, University of Minnesota, Minneapolis, MN, USA).

2.2. Reagents and Antibodies

Syringaresinol (Syr, #BBP00152) was purchased from Yunnan BioBiopha Biotechnology Co. Ltd. (Kunming, China). Liproxstatin‐1 (Lip‐1, #L407981), RSL3 (#R407911), and erastin (#E424821) were obtained from Shanghai Aladdin Biochemical Technology Co. Ltd. (Shanghai, China). Ferrostatin‐1 (Fer‐1, #GC10380) was purchased from GLPBIO Technology LLC (Montclair, CA, USA). Diethyl maleate (DEM, #D97703) was obtained from Sigma‐Aldrich (St. Louis, MO, USA). D‐galactose (D‐gal, #S11050) was purchased from Shanghai Yuanye Bio‐Technology Co. Ltd. (Shanghai, China). Rabbit antibodies against glutathione peroxidase 4 (GPX4, monoclonal, #T56959F, 1:1000), solute carrier family 7 member 11 (SLC7A11, polyclonal, #T57046F, 1:1000), and ferritin light chain (FTL, polyclonal, #T55648F, 1:1000) were purchased from Abmart Biopharmaceutical Co. Ltd. (Shanghai, China). Rabbit anti‐hypoxia‐inducible factor‐1α (HIF‐1α, polyclonal, #YP‐Ab‐01776, 1:500) was obtained from Youpin Biotechnology Co. Ltd. (Shenzhen, China). Rabbit anti‐p21 (monoclonal, #2947T, 1:1000), mouse anti‐β‐actin (monoclonal, #4970S, 1:5000), horseradish peroxidase (HRP)‐conjugated goat anti‐rabbit IgG (#7074P2, 1:5000), and HRP‐conjugated goat anti‐mouse IgG (#7076S, 1:5000) were purchased from Cell Signaling Technology (Danvers, MA, USA).

2.3. C. elegans Experiments

2.3.1. Maintenance and Synchronization

Wild‐type C. elegans (N2) were maintained on nematode growth medium (NGM) plates seeded with Escherichia coli OP50 at 20°C under standard conditions unless otherwise stated [17]. Age‐synchronized populations were generated by alkaline hypochlorite lysis. Briefly, mixed‐stage worms were collected and lysed in freshly prepared lysis buffer containing 1 mL of 1% sodium hypochlorite, 1 mL of 5 M NaOH, and 8 mL of M9 buffer. After vigorous vortexing for 2–3 min until adult carcasses were disrupted and eggs were released, eggs were collected by centrifugation, washed 2–3 times with M9 buffer, resuspended in M9 buffer, and incubated overnight at 20°C with gentle shaking (60 rpm) to obtain synchronized L1 larvae.

2.3.2. Establishment and Optimization of the DEM Challenge Model

To establish and optimize a DEM‐based ferroptosis‐relevant oxidative stress model, synchronized L1 larvae were transferred to OP50‐seeded NGM plates and cultured at 20°C for 48–52 h until the L4 stage. Age‐synchronized L4 worms were then transferred to 35‐mm NGM plates containing 5‐fluoro‐2′‐deoxyuridine (FUdR; stock concentration, 5 mg/mL; added at 1:1000 before agar solidification) to prevent progeny development. Late‐L4 or young adult worms were exposed to different DEM concentrations and exposure durations to identify a reproducible injury window with stable and quantifiable lethality while retaining sufficient dynamic range for pharmacological rescue. DEM was selected because it is a classical glutathione‐depleting electrophile and has been used to trigger acute GSH depletion‐associated ferroptosis‐like injury in C. elegans and related oxidative stress systems [4]. Based on these optimization experiments, the final screening condition was set at 100 mM DEM for 72 h. For model induction, 400 μL of solution containing DEM, with or without candidate compounds (100 μM), was evenly spread onto the plate surface and allowed to air‐dry at room temperature, followed by addition of 300 μL OP50 suspension (OD600 = 0.5). Worms were then incubated at 20°C for 72 h. Vehicle controls received the corresponding volume of DMSO (final concentration ≤ 0.1%), and Fer‐1 was used as a positive control for protection. Worms were scored as dead when they exhibited no spontaneous movement and failed to respond to gentle mechanical stimulation.

2.3.3. Whole‐Organism Screening and Lead‐Compound Validation

Using the optimized condition of 100 mM DEM for 72 h, an in‐house small‐molecule library containing 220 natural‐product‐derived or natural‐product‐inspired compounds was screened in a survival‐based assay. The library was assembled to cover diverse chemical classes, including lignans, flavonoids, phenolic acids, coumarins, terpenoids, saponins, alkaloids, quinones, and other structurally defined small molecules. Compounds were selected based on chemical diversity, availability with confirmed purity, reported or potential relevance to redox regulation, inflammation, aging, mitochondrial function, or regulated cell death, and adequate solubility in DMSO under the screening condition.

All compounds were initially tested at 100 μM, with matched vehicle controls and a consistent final DMSO concentration across groups. Before formal ranking, compounds showing visible precipitation on NGM plates or obvious basal toxicity in the absence of DEM were excluded from further prioritization. Each screening batch included vehicle, DEM‐only, and DEM + Fer‐1 controls to monitor assay performance. Primary hits were prioritized based on their ability to increase worm survival under DEM challenge relative to the DEM‐only group. Compounds ranked among the top candidates in the primary screen were independently retested under the same conditions. Syr emerged as the leading protective compound and consistently showed the strongest protective effect among the top 12 retested candidates. Syr was then subjected to dose‐ and time‐response validation for subsequent experiments.

2.3.4. Lifespan Assay

Approximately 150 synchronized L4 worms were transferred onto NGM plates containing FUdR (5 mg/mL stock, added at 1:1000 before agar solidification) and OP50, with or without the indicated concentrations of Syr. Worms were maintained at 20°C. Beginning on Day 1 of adulthood, worms were transferred to fresh plates every 2 days. Worms were scored as dead when pharyngeal pumping ceased and no response to gentle touch was observed. Missing worms were censored. The assay continued until all worms had died. Kaplan–Meier survival curves were generated using GraphPad Prism 9 and compared using the log‐rank test, following standard C. elegans lifespan approaches.

2.3.5. Healthspan Assessment

For pharyngeal pumping, 20 worms were randomly selected and the number of pharyngeal contractions during a 20‐s interval was counted under a stereomicroscope (Leica M205FA). For body‐bend analysis, individual worms were placed on bacteria‐free NGM plates containing a drop of M9 buffer. After a 30‐s acclimation period, the number of body bends crossing the longitudinal axis within 20 s was recorded. For locomotor trajectory analysis, 20‐s videos of freely moving worms were captured using a Leica M205FA stereomicroscope equipped with a camera. Videos were analyzed in ImageJ using Manual Tracking or the WormTracker plugin to calculate mean locomotion speed (μm/s).

2.3.6. Malondialdehyde Assay

Malondialdehyde (MDA) levels were measured using a thiobarbituric acid‐based assay kit (Beyotime, #S0131S). Four groups were included: control, DEM, DEM + Fer‐1, and DEM + Syr. Approximately 3000 adult worms per group were collected, washed three times with ice‐cold M9 buffer, and snap‐frozen in liquid nitrogen. Samples were homogenized in lysis buffer using a tissue disruptor (60 Hz, 10 s per cycle, 5 cycles total, with 30‐s intervals on ice), then centrifuged at 12000 × g for 15 min at 4°C. Supernatants were used for MDA measurement according to the manufacturer's instructions. Absorbance was read at 532 nm. MDA content was normalized to protein concentration determined by the BCA assay and expressed as μmol/mg protein. All procedures were performed under light‐protected conditions. Three biological replicates were analyzed per group.

2.3.7. ROS Detection

Intracellular reactive oxygen species were assessed in nematodes and cells using dihydroethidium (DHE; BIORIGIN Biotech, #BN1108), an established fluorescent probe for ROS imaging [18]. After treatment, samples were washed twice with pre‐warmed M9 buffer (nematodes) or PBS (cells). Samples were then incubated with 10 μM DHE for 30–40 min in the dark at 20°C for worms or 37°C for cells. After staining, excess probe was removed by three washes with M9 buffer or complete medium, respectively. Worms were anesthetized with 2% levamisole before imaging. Fluorescence images were acquired immediately under the red fluorescence channel (Ex 518 nm/Em 605 nm), and mean fluorescence intensity was quantified using ImageJ.

2.3.8. Lipid Peroxidation Assay

Lipid peroxidation was assessed using C11‐BODIPY 581/591 (Cayman Chemical, #27086), a widely used probe for detecting ferroptosis‐associated lipid peroxidation [19]. After washing with M9 buffer (worms) or PBS (cells), samples were incubated with 5 μM probe for 30 min in the dark at 20°C for worms or 37°C for cells. After three washes to remove excess dye, worms were anesthetized with 2% levamisole and mounted for imaging. Fluorescence images were acquired immediately using FITC and TRITC channels. The reduced probe emits red fluorescence (~590 nm), whereas the oxidized probe emits green fluorescence (~510 nm). Fluorescence intensities were quantified using ImageJ, and the green/red ratio was calculated as an index of lipid peroxidation.

2.3.9. Labile Iron Detection

The labile iron pool in worms was measured using PhenGreen SK (PGSK; Thermo Fisher Scientific, #P16035), a classical fluorescent probe used to assess chelatable iron pools [20]. Worms were washed with M9 buffer and incubated with 5 μM PGSK for 60 min at 20°C in the dark with gentle shaking. After three washes with M9 buffer, worms were anesthetized with 2% levamisole, mounted, and imaged immediately under the FITC channel (Ex 488 nm/Em 500–550 nm). All imaging was completed within 2 h after staining.

2.4. Cell Culture

HFFs were cultured in high‐glucose DMEM (Gibco, #C11995500BT) supplemented with 10% fetal bovine serum (FBS; Procell, #SA210623) at 37°C in a humidified incubator with 5% CO₂. For cryopreservation, logarithmically growing cells were collected and resuspended in cryopreservation medium (Biosharp, #BL203B), transferred to cryovials, stored overnight at −80°C, and then placed in liquid nitrogen for long‐term storage. For recovery, vials were rapidly thawed in a 37°C water bath, diluted with pre‐warmed complete medium, and centrifuged at 700–800 rpm for 3 min to remove cryoprotectant. Cells were then resuspended and seeded into culture vessels. For routine passage, cells at 80%–90% confluence were detached with 0.25% trypsin (Biosharp, #BL501A), neutralized with complete medium, collected by centrifugation, and passaged at a ratio of 1:3–1:5. Medium was replaced every 2–3 days. For drug treatment experiments, cells were seeded at appropriate densities according to the downstream assay. Cells grown in 96‐well plates were used for CCK‐8 assays, whereas cells grown in 6‐well plates were used for Western blotting and related analyses.

2.5. Establishment and Validation of Senescence Models

Senescence models were established in HFFs using D‐gal‐induced premature senescence and replicative senescence. For D‐gal‐induced senescence, HFFs at passages 5–8 in logarithmic growth were treated with 150 mM D‐gal for 5 days, with fresh D‐gal‐containing medium replaced every 48 h [21]. After induction, cells were washed three times with pre‐warmed PBS, cultured in regular complete medium for an additional 24 h, and then subjected to subsequent analyses; untreated cells served as controls. For replicative senescence, low‐passage HFFs (passages 8–10) were serially passaged at a split ratio of 1:2–1:3, and population doubling time was recorded at each passage. Cells were considered senescent when population doubling time had increased by at least twofold relative to early‐passage cells and when more than 70% of cells displayed typical senescent morphology, including enlarged and flattened cell bodies and increased cytoplasmic granularity. Under these conditions, stable replicative senescence was typically observed after passages 30–40 [22]. Senescence was validated by SA‐β‐gal staining using a commercial kit (Solarbio, #G1581) according to the manufacturer's instructions at pH 6.0. Multiple random fields were imaged under an inverted microscope, and the percentage of SA‐β‐gal‐positive cells was quantified.

2.6. Establishment of Ferroptosis Models in HFFs

To induce ferroptosis, HFFs in logarithmic growth were seeded at an appropriate density and allowed to adhere overnight. For the RSL3 model, cells were treated with 1.56 μM RSL3 for 24 h. For the erastin model, cells were treated with 25 μM erastin for 24 h. Both working concentrations were determined in preliminary dose–response experiments. RSL3 and erastin were selected as canonical ferroptosis inducers acting through GPX4 inhibition and system Xc⁻ inhibition, respectively. Untreated controls and solvent controls (final DMSO concentration < 0.1%) were included in each experiment. After treatment, cells were washed once with pre‐warmed PBS and subjected to downstream assays, including CCK‐8 analysis.

2.7. Cell Viability Assay

Cell viability was measured using the Cell Counting Kit‐8 (CCK‐8; APEXBIO, #K1018). HFFs in logarithmic growth were seeded into 96‐well plates at 3 × 10^3 cells/well and allowed to attach overnight. Cells were then treated with the indicated compounds or vehicle control for 24 or 48 h. At the endpoint, 10 μL of CCK‐8 reagent was added to each well, and plates were incubated at 37°C for 1–2 h in the dark. Absorbance was measured at 450 nm using a microplate reader (BioTek Synergy 2). Relative cell viability was calculated as follows: (OD_experimental − OD_blank)/(OD_control − OD_blank) × 100%. Each experiment was independently repeated at least three times, with three technical replicates per condition.

2.8. SA‐β‐Gal Staining

SA‐β‐gal staining was performed using a commercial kit (Solarbio, #G1581) according to the manufacturer's instructions. Briefly, culture medium was removed and cells were washed twice with cold PBS, followed by fixation with 1 mL of fixative solution for 15 min at room temperature. Cells were then washed three times with PBS and incubated with freshly prepared staining solution at 37°C in a non‐CO₂ incubator protected from light for 12–16 h. After staining, at least three non‐overlapping random fields were captured per group under an inverted microscope. Cells with distinct blue cytoplasmic precipitates were considered SA‐β‐gal‐positive. The number of positive cells and total cells were quantified using ImageJ, and the percentage of SA‐β‐gal‐positive cells was calculated.

2.9. Western Blotting

Cells were washed three times with ice‐cold PBS and lysed on ice for 30 min in RIPA buffer supplemented with protease and phosphatase inhibitors. Lysates were centrifuged at 12000 × g for 15 min at 4°C, and the supernatants were collected for protein quantification using the Bradford assay. Protein samples were adjusted to equal concentrations (2–4 μg/μL) and mixed with 5× SDS loading buffer. All samples except those used for SLC7A11 detection were denatured at 100°C for 10 min, whereas samples for SLC7A11 were heated at 60°C for 4 min to preserve antigenicity. Equal amounts of protein (20–40 μg) were separated on 7.5% or 15% SDS‐PAGE gels, with electrophoresis performed at 80 V through the stacking gel and 120 V through the resolving gel. Proteins were then transferred onto PVDF membranes by wet transfer at 250 mA for 1–2 h in an ice bath. PVDF membranes with pore sizes of 0.45 μm (> 50 kDa) or 0.22 μm (< 50 kDa) were selected according to target protein size. Membranes were blocked with 5% non‐fat milk in TBST for 1 h at room temperature and incubated overnight at 4°C with the primary antibodies listed in Section 2.2. After washing with TBST, membranes were incubated with HRP‐conjugated secondary antibodies (1:5000) for 1 h at room temperature. Protein bands were visualized using ECL substrate and captured using a chemiluminescence imaging system. Band intensities were quantified with Image Lab or ImageJ software, and target protein expression was normalized to β‐actin.

2.10. Quantitative Real‐Time PCR

Total RNA was extracted using TRIzol reagent (Yeasen, #18605ES20). One microgram of total RNA was reverse‐transcribed after genomic DNA removal using a two‐step cDNA synthesis kit (Vazyme, #R223‐01). Quantitative real‐time PCR was performed using SYBR Green PCR Master Mix (Vazyme, #Q‐712‐02), with three technical replicates per sample. GAPDH was used as the internal reference gene. Relative mRNA expression was calculated using the 2^‐ΔΔCt method. Primer sequences are listed in Table S1.

2.11. Transient Transfection and Gene Knockdown

To investigate the functional relationship between HIF‐1α and GPX4 in Syr‐associated cytoprotection, HFFs were transiently transfected with shRNA plasmids targeting HIF1A (Miaoling Bio, #ML‐shHIF1A‐001) or GPX4 (Miaoling Bio, #ML‐shGPX4‐003). A mock‐transfected group receiving transfection reagent alone without plasmid DNA was included as a control. One day before transfection, HFFs were seeded into 6‐well plates at 1 × 10^5 cells/well to achieve 70%–80% confluence at the time of transfection. Transfection was performed using ExFect Transfection Reagent (Vazyme, #T101‐02) according to the manufacturer's instructions. Briefly, 2 μg of plasmid DNA was mixed with ExFect Enhancer in 125 μL Opti‐DMEM, and 5 μL of ExFect Reagent was diluted separately in 125 μL Opti‐DMEM. After 5 min, the two solutions were combined and incubated for 15 min to allow complex formation, then added dropwise to the cells. Medium was replaced with fresh complete medium after 6 h. Cells were harvested 48 h after transfection. Knockdown efficiency was confirmed by Western blotting before subsequent functional assays.

2.12. Molecular Docking

The predicted full‐length structure of human HIF‐1α was obtained from the AlphaFold Protein Structure Database using UniProt accession Q16665. Because no experimentally resolved full‐length HIF‐1α structure is currently available, the AlphaFold‐predicted model was used for exploratory molecular docking, and the docking results were interpreted as structure‐based predictions rather than direct biochemical evidence of binding. The receptor structure was prepared by removing non‐protein molecules, adding polar hydrogens, and assigning Gasteiger charges using AutoDockTools 1.5.7. The three‐dimensional structure of syringaresinol was obtained from PubChem and energy‐minimized using Chem3D/Open Babel before docking. Molecular docking was performed using AutoDock Vina version 1.2.3.

A broad docking grid was defined to cover the predicted binding region containing His193, Asp238 and Gln320. The grid center was set to x = −2.001, y = −3.098 and z = 0.820, with grid dimensions of 100.0 × 120.0 × 126.0 Å. The exhaustiveness parameter was set to 8. The top‐ranked binding pose was selected based on the lowest predicted binding energy and visual inspection of ligand–residue interactions. Protein–ligand interactions were visualized using PyMOL version 2.5.4.

2.13. Statistical Analysis

All experiments were independently repeated at least three times with appropriate technical replicates. Except for lifespan assays, data are presented as mean ± SD. Statistical analyses and graphing were performed using GraphPad Prism 9.0. Lifespan data were analyzed using Kaplan–Meier survival curves and compared by the log‐rank (Mantel‐Cox) test. For other datasets, statistical analyses were performed as indicated in the figure legends. All tests were two‐tailed. A value of p < 0.05 was considered statistically significant, with p < 0.01 and p < 0.001 indicating higher levels of significance.

3. Results

3.1. Syr Suppresses Glutathione Depletion‐Induced Ferroptosis‐Relevant Injury in C. elegans

To establish an in vivo screening platform for ferroptosis‐relevant protective compounds, we first optimized a DEM‐induced glutathione depletion model in wild‐type (N2) C. elegans by testing different exposure times and concentrations (Figure S1A,B). Using the optimized DEM challenge as the screening window, we screened 220 compounds from our in‐house library and identified Syr as the leading protective hit. In the survival heatmap, Syr was among the compounds conferring the highest survival under DEM challenge and was selected for further validation (Figure 1A,B; Table S2). To confirm this result, we re‐tested the top 12 compounds ranked by survival in the primary screen; among these candidates, Syr consistently showed the strongest protective effect (Figure S1C,D; Table S3).

FIGURE 1.

Syr protects against DEM‐induced ferroptosis‐relevant injury in wild‐type C. elegans (N2). (A) Survival heatmap from high‐throughput screening of an in‐house compound library in wild‐type N2 worms under DEM challenge. The color scale indicates survival from low (green) to high (red). Syr was identified as a protective hit and is indicated by a yellow arrow. (B) Chemical structure of Syr. (C) Survival rates of N2 worms exposed to DEM (100 mM) in the presence or absence of ferrostatin‐1 (Fer‐1, 100 μM) or Syr (50, 100, and 200 μM). (D) Representative bright‐field images of worms at 24, 48, and 72 h after treatment with vehicle control (Ctrl), DEM (100 mM), Fer‐1 (100 μM), or Syr (100 μM). Red arrows indicate representative abnormal or dead worms under DEM challenge. (E) Quantification of survival rates for the groups shown in (D). (F) Representative C11‐BODIPY 581/591 fluorescence images of worms showing reduced probe fluorescence (red), oxidized probe fluorescence (green), and merged channels. (G) Quantification of lipid peroxidation expressed as the C11‐BODIPY 581/591 green/red fluorescence ratio. (H) Representative bright‐field and DHE fluorescence images of worms showing intracellular ROS. (I) Quantification of DHE fluorescence intensity shown in (H). (J) MDA levels in worms from the indicated groups, normalized to total protein content. Data are presented as mean ± SD. Multiple‐group comparisons were performed using one‐way ANOVA followed by Tukey's multiple‐comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant. Unless otherwise indicated, statistical significance was determined relative to the DEM‐treated group. Scale bars are shown in the images.

We next performed dose‐ and time‐response validation. Syr showed maximal protection at 200 μM; however, because the effects at 100 and 200 μM were comparable, 100 μM was selected for subsequent experiments (Figure 1C–E; Figure S1E). We then assessed ferroptosis‐relevant readouts, including lipid peroxidation, ROS, and MDA. Syr markedly reduced DEM‐induced lipid peroxidation, as indicated by a lower C11‐BODIPY 581/591 green/red fluorescence ratio, and significantly decreased MDA levels. In parallel, Syr suppressed ROS accumulation detected by DHE staining (Figure 1F–J). Together, these results indicate that Syr protects C. elegans against DEM‐induced ferroptosis‐relevant injury and is associated with reduced oxidative lipid damage under glutathione depletion conditions.

3.2. Syr Attenuates Age‐Associated Functional Decline and Ferroptosis‐Relevant Stress During Natural Aging in C. elegans

We next examined whether Syr could modulate ferroptosis‐relevant stress during natural aging in C. elegans . In untreated worms, lipid peroxidation progressively increased with age, as shown by C11‐BODIPY 581/591 staining, with older animals exhibiting stronger green fluorescence and weaker red fluorescence (Figure 2A,B). ROS levels also increased significantly during aging (Figure 2C,D). Syr treatment attenuated these age‐associated increases in both lipid peroxidation and ROS accumulation (Figure 2A–D).

FIGURE 2.

Syr attenuates age‐associated ferroptosis‐relevant oxidative stress, delays functional decline, and extends lifespan in naturally aging C. elegans . (A) Representative C11‐BODIPY 581/591 fluorescence images of worms from the Ctrl and Syr (100 μM) groups at adult Days 1, 7, and 14, showing reduced probe fluorescence (red), oxidized probe fluorescence (green), and merged channels. (B) Quantification of lipid peroxidation expressed as the C11‐BODIPY 581/591 green/red fluorescence ratio corresponding to (A). (C) Representative DHE fluorescence images showing ROS and PGSK fluorescence images indicating the labile Fe²⁺ pool in worms from the indicated groups and time points. (D, E) Quantification of DHE fluorescence intensity and PGSK fluorescence intensity, respectively, corresponding to (C). (F) Kaplan–Meier survival curves of worms treated with or without Syr (100 μM). (G) Violin plot comparing median survival between groups. (H) Representative images of lipofuscin autofluorescence in worms at adult Days 1, 7, and 14. (I) Quantification of lipofuscin fluorescence intensity corresponding to (H). (J) Representative 20‐s locomotor trajectories of individual worms at the indicated ages. (K) Quantification of mean locomotion speed. (L) Quantification of locomotion distance. (M) Pharyngeal pumping rate expressed as contractions per 20 s. (N) Body‐bend frequency expressed as bends per 20 s. Unless otherwise indicated, data are presented as mean ± SD. Comparisons among multiple groups were performed using one‐way ANOVA followed by Tukey's multiple‐comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001 versus Ctrl. Survival curves were analyzed using the log‐rank test. Scale bars are shown in the images.

We further assessed age‐related changes in the labile iron pool, a key determinant of ferroptosis susceptibility. Because PGSK fluorescence is quenched by labile Fe²⁺, the stronger green fluorescence observed in Syr‐treated worms was consistent with reduced age‐associated labile iron accumulation (Figure 2C,E). Lifespan analysis showed that Syr significantly extended survival in C. elegans , increasing median lifespan from 18 to 22 days and maximum lifespan from 26 to 29 days (Figure 2F,G). Detailed statistics, including the total number of worms, censored animals, median and maximum lifespan, and log‐rank p values, are provided in Table S4. Syr also improved multiple healthspan‐related phenotypes. Compared with age‐matched controls, Syr‐treated worms displayed reduced lipofuscin accumulation, more preserved locomotor activity, higher movement speed, longer locomotor trajectories, and improved pharyngeal pumping and body‐bend frequency at later ages (Figure 2H–N). Thus, in addition to extending lifespan, Syr delayed the progression of age‐associated functional decline across multiple behavioral and physiological readouts. Collectively, these data indicate that Syr improves both lifespan and healthspan in C. elegans and is associated with reduced oxidative lipid stress and ameliorated age‐associated labile iron accumulation during aging.

3.3. Syr Suppresses Ferroptosis in Human Foreskin Fibroblasts

To determine whether the anti‐ferroptotic activity of Syr is conserved in mammalian cells, we evaluated its effects in HFFs using two canonical ferroptosis inducers, RSL3 and erastin. CCK‐8 assays showed that both RSL3 and erastin reduced HFF viability in a concentration‐dependent manner, supporting their use for ferroptosis model establishment (Figure S2A,B). Based on these results, 1.56 μM RSL3 and 25 μM erastin were selected for subsequent experiments. Syr showed minimal basal cytotoxicity within the low‐to‐moderate concentration range, whereas higher concentrations reduced HFF viability; Lip‐1 showed minimal cytotoxicity up to 1 μM (Figure S2C,D; Table S5).

Under these conditions, Syr rescued the loss of viability induced by both ferroptosis inducers (Figure 3A,B). In the RSL3 model, cell viability was reduced to 45.58% ± 0.20%, whereas Syr restored viability to a maximum of 84.51% ± 0.71% at 50 μM. In the erastin model, viability decreased to 56.35% ± 7.74%, whereas Syr restored viability to a maximum of 83.92% ± 2.00% at 25 μM. Detailed numerical values and p values are provided in Table S5. Consistently, morphological examination showed that RSL3 and erastin caused typical ferroptosis‐associated damage, including cell shrinkage and loss of adhesion, whereas Syr preserved cell morphology and structural integrity within the effective concentration range, similar to Lip‐1 (Figure 3C,D).

FIGURE 3.

Syr attenuates RSL3‐ or erastin‐induced ferroptosis in HFF cells. (A) CCK‐8 assay showing viability of HFF cells treated with RSL3 (1.56 μM) in the presence of Lip‐1 (1 μM) or increasing concentrations of Syr (0.20–100 μM), with Ctrl as baseline. (B) CCK‐8 assay showing viability of HFF cells treated with erastin (25 μM) in the presence of Lip‐1 (1 μM) or increasing concentrations of Syr (0.20–100 μM). (C) Representative bright‐field images of HFF cells treated with RSL3 in the presence or absence of Lip‐1 or Syr (6.25, 12.5, and 25 μM). (D) Representative bright‐field images of HFF cells treated with erastin in the presence or absence of Lip‐1 or Syr (6.25, 12.5, and 25 μM). (E) Representative fluorescence images of C11‐BODIPY 581/591 staining (lipid peroxidation; red, reduced probe; green, oxidized probe; merge) and DHE staining (ROS) in HFF cells under RSL3 treatment with or without Lip‐1 or Syr (12.5, 25, and 50 μM). (F) Quantification of lipid peroxidation expressed as the C11‐BODIPY 581/591 green/red fluorescence ratio and ROS levels expressed as DHE fluorescence intensity corresponding to (E). (G) Representative immunoblots of GPX4, SLC7A11, and ferritin light chain in HFF cells treated with RSL3 in the presence or absence of Lip‐1 or Syr; β‐Actin was used as the loading control. (H) Densitometric quantification of the immunoblots shown in (G), normalized to β‐Actin. All experiments were independently repeated at least three times. Data are presented as mean ± SD. Statistical analyses were performed using one‐way ANOVA followed by Tukey's multiple‐comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001 versus the corresponding model group. Scale bars are shown in the images.

Because GPX4 is central to ferroptosis execution, we next focused on the RSL3 model to assess downstream readouts. Syr co‐treatment significantly reduced RSL3‐induced lipid peroxidation and ROS accumulation (Figure 3E,F). At the protein level, RSL3 decreased the expression of several key anti‐ferroptotic defense proteins, including GPX4, SLC7A11, and ferritin light chain (FTL), whereas Syr largely restored their expression (Figure 3G,H). These results indicate that Syr suppresses ferroptosis in HFFs and is associated with preservation of endogenous defense pathways involved in antioxidant buffering, lipid peroxide detoxification, and iron handling.

3.4. Syr Attenuates Senescence Phenotypes and Reduces Ferroptosis‐Associated Vulnerability in HFFs

After establishing the anti‐ferroptotic activity of Syr in HFFs, we next examined whether it could also alleviate cellular senescence. Before evaluating the anti‐senescent effects of Syr, we first examined the effects of D‐gal and rapamycin (Rap) on HFF viability to define suitable experimental conditions. CCK‐8 assays showed that both D‐gal and Rap influenced basal cell viability in a dose‐dependent manner (Figure S3), and the concentrations used in subsequent experiments were selected based on these preliminary dose–response assays.

Syr was then evaluated in two senescence models, D‐gal‐induced premature senescence and replicative senescence. In both models, SA‐β‐gal staining showed a marked increase in senescent cell positivity, whereas Syr treatment significantly reduced the proportion of SA‐β‐gal‐positive cells (Figure 4A–D). Syr also ameliorated the characteristic senescent morphological alterations, as reflected by reduced cell length and width, indicating partial reversal of the enlarged and flattened senescent phenotype (Figure 4B,D). These effects were generally comparable to those observed with Rap.

FIGURE 4.

Syr alleviates cellular senescence phenotypes and associated ferroptosis‐relevant oxidative stress in two HFF senescence models. (A) Representative SA‐β‐gal staining images of HFF cells in the Ctrl group, D‐gal‐induced senescence group, Rap‐treated group, and Syr‐treated groups (6.25, 12.5, and 25 μM) within the D‐gal model. (B) Quantification of SA‐β‐gal positivity and cell morphology parameters (cell length and width) corresponding to (A). (C) Representative SA‐β‐gal staining images of replicative senescent HFF cells (Rep) treated with or without Rap or Syr (6.25, 12.5, and 25 μM). (D) Quantification of SA‐β‐gal positivity and cell morphology parameters corresponding to (C). (E) Representative immunoblots of p21 in the D‐gal model and quantification of the p21/β‐Actin ratio. (F) Representative immunoblots of p21 in the Rep model and quantification of the p21/β‐Actin ratio. (G) Representative immunoblots of GPX4 in the D‐gal model and quantification of the GPX4/β‐Actin ratio. (H) Representative immunoblots of GPX4 in the Rep model and quantification of the GPX4/β‐Actin ratio. (I) Representative C11‐BODIPY 581/591 and DHE fluorescence images in the D‐gal model, showing lipid peroxidation and ROS, respectively. (J) Representative C11‐BODIPY 581/591 and DHE fluorescence images in the Rep model. (K) Quantification of the C11‐BODIPY 581/591 green/red fluorescence ratio and DHE fluorescence intensity corresponding to (I). (L) Quantification of the C11‐BODIPY 581/591 green/red fluorescence ratio and DHE fluorescence intensity corresponding to (J). Unless otherwise indicated, data are from at least three independent experiments and are presented as mean ± SD. Statistical analyses were performed using one‐way ANOVA followed by Tukey's multiple‐comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001 versus the corresponding model group. Scale bars are shown in the images.

To further evaluate the anti‐senescent effects of Syr at the transcriptional level, qRT‐PCR analysis showed that Syr significantly downregulated multiple senescence‐associated genes, including the cell‐cycle regulators p14 and p21, the inflammatory factors IL‐1α, CXCL3, and TNF‐α, and the matrix‐remodeling genes MMP‐3, MMP‐9, and MMP‐12 (Figure S4). Consistent with these findings, Western blot analysis showed that p21 protein abundance was increased in both senescence models and was reduced by Syr treatment (Figure 4E,F).

We then examined whether senescence in these models was accompanied by impaired ferroptosis defense. GPX4 protein expression was markedly decreased in both models, whereas Syr treatment restored GPX4 abundance (Figure 4G,H). Consistent with this change, C11‐BODIPY 581/591 staining showed increased lipid peroxidation in senescent cells, while DHE staining showed increased ROS accumulation; both changes were attenuated by Syr treatment in the two models (Figure 4I–L). Collectively, these results indicate that Syr alleviates senescence phenotypes in HFFs and is accompanied by restoration of GPX4 and reduction of lipid peroxidation and ROS, consistent with reduced ferroptosis‐associated vulnerability in senescent cells.

3.5. Network Pharmacology and Molecular Docking Implicate HIF‐1α as a Candidate Upstream Mediator of Syr‐Associated Protection

To explore potential upstream regulators underlying the anti‐ferroptotic and anti‐senescence effects of Syr, we applied a network pharmacology strategy to identify candidate targets shared by Syr, ferroptosis, and aging‐related pathways. Integration of Syr‐predicted targets with curated ferroptosis‐ and aging‐related gene sets yielded 17 overlapping candidates (Figure 5A), suggesting that Syr may modulate a shared molecular network relevant to both processes. Gene Ontology (GO) enrichment analysis showed that these candidates were mainly associated with lipid oxidation, inflammatory responses, and regulation of apoptosis (Figure 5B–D). KEGG pathway analysis further revealed significant enrichment in the HIF‐1 signaling pathway and arachidonic acid metabolism (Figure 5E).

FIGURE 5.

Network pharmacology and molecular docking identify HIF‐1α as a candidate hub linking Syr to aging‐ and ferroptosis‐related pathways. (A) Venn diagram showing the overlap among aging‐related genes, ferroptosis‐related genes, and predicted Syr targets, identifying 17 shared candidate genes. (B–D) Gene Ontology enrichment analyses of the 17 shared genes, including Biological Process (BP) (B), Cellular Component (CC) (C), and Molecular Function (MF) (D), presented as bubble plots. Bubble size indicates gene count and color indicates adjusted significance. (E) KEGG pathway enrichment analysis of the shared genes shown as a bubble plot. (F) Protein–protein interaction network of the shared targets, highlighting HIF‐1α as a central hub node. (G) Exploratory molecular docking model of Syr with the AlphaFold‐predicted full‐length HIF‐1α structure, showing the predicted binding conformation and key interaction residues, including His193, Asp238, and Gln320. The top‐ranked Vina docking score was −6.8 kcal/mol.

To identify key nodes within this predicted network, we constructed a protein–protein interaction network and performed topological analysis, which highlighted HIF‐1α as a prominent hub protein (Figure 5F). Exploratory molecular docking further suggested that Syr could be accommodated within a predicted binding region of HIF‐1α involving key interacting residues His193, Asp238, and Gln320, with a top‐ranked Vina docking score of −6.8 kcal/mol (Figure 5G). Because this analysis was computational, the docking result was interpreted as hypothesis‐generating and was used to prioritize HIF‐1α for functional validation, rather than as direct evidence of Syr–HIF‐1α target engagement.

3.6. HIF‐1α Is Required for Syr‐Associated Protection and Lies Upstream of GPX4

Based on the network prediction and docking results, we next tested whether HIF‐1α is functionally required for Syr‐associated protection in HFFs. In the RSL3‐induced ferroptosis model, RSL3 markedly reduced HIF‐1α protein abundance, whereas Syr co‐treatment antagonized this reduction and preserved HIF‐1α levels; the magnitude of this effect was comparable to that observed with Lip‐1 (Figure 6A,B). These findings suggest that Syr is associated with the maintenance of HIF‐1α abundance under ferroptotic stress.

FIGURE 6.

Syr‐associated protection in ferroptosis and senescence models depends on HIF‐1α and is accompanied by restoration of GPX4. (A, B) Representative immunoblots (A) and quantification (B) of HIF‐1α expression in HFF cells under RSL3‐induced ferroptosis conditions in the Ctrl, RSL3 (1.56 μM), Lip‐1 (1 μM), and Syr (25 μM) groups; β‐actin served as the loading control. (C, D) Representative SA‐β‐gal staining images (C) and percentages of SA‐β‐gal‐positive cells (D) in the D‐gal‐induced senescence model treated with Syr (25 μM), with or without HIF‐1α knockdown. (E, F) Representative SA‐β‐gal staining images (E) and percentages of SA‐β‐gal‐positive cells (F) in replicatively senescent HFF cells treated with Syr (25 μM), with or without HIF‐1α knockdown. (G, H) Representative C11‐BODIPY 581/591 and DHE fluorescence images in the D‐gal‐induced senescence model (G) and replicative senescence model (H), respectively, under the indicated treatment conditions. Oxidized C11‐BODIPY fluorescence is shown in green, reduced C11‐BODIPY fluorescence in red, and intracellular ROS detected by DHE staining in red. (I, J) Quantification of the C11‐BODIPY green/red fluorescence ratio and DHE fluorescence intensity corresponding to the D‐gal‐induced senescence model shown in (G) and the replicative senescence model shown in (H), respectively. (K, L) Representative immunoblots and densitometric quantification of GPX4 expression in the D‐gal‐induced senescence model (K) and replicative senescence model (L) treated with Syr (25 μM), with or without HIF‐1α knockdown. (M, N) Representative immunoblots and densitometric quantification of HIF‐1α expression in the replicative senescence model (M) and D‐gal‐induced senescence model (N) treated with Syr (25 μM), with or without GPX4 knockdown. β‐Actin was used as the loading control. Data are presented as mean ± SD. Statistical analyses were performed using one‐way ANOVA followed by Tukey's multiple‐comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant. Scale bars are shown in the images.

To determine whether HIF‐1α is required for Syr‐mediated protection, we established an HIF‐1α knockdown model in HFFs and verified knockdown efficiency by Western blotting. HIF‐1α protein abundance was markedly reduced in the knockdown group relative to the corresponding control group, confirming effective silencing (Figure S5A,B). In control cells, Syr substantially alleviated RSL3‐induced morphological injury, including cell shrinkage and detachment; however, this protective effect was markedly weakened after HIF‐1α knockdown (Figure S5C). Consistently, CCK‐8 analysis showed that Syr rescued the RSL3‐induced decline in cell viability in control cells, whereas this rescue was largely attenuated in HIF‐1α‐knockdown cells (Figure S5D). We then assessed whether HIF‐1α was also required for Syr‐associated protection in the two senescence models. In control cells, Syr reduced SA‐β‐gal positivity, improved senescent morphology, and suppressed lipid peroxidation and ROS accumulation. These effects were largely abolished after HIF‐1α knockdown (Figure 6C–J; Figure S5E–H), indicating that HIF‐1α is necessary for Syr‐associated protection in both ferroptotic stress and senescence‐related contexts.

Finally, we examined the relationship between HIF‐1α and GPX4. In senescent cells, GPX4 protein abundance was reduced and was restored by Syr treatment. Importantly, this restoration was largely lost after HIF‐1α knockdown, indicating that Syr‐associated GPX4 recovery requires HIF‐1α under these experimental conditions (Figure 6K,L). To further examine the directionality of this relationship, we established a GPX4 knockdown model and verified knockdown efficiency by Western blotting, which showed a marked reduction in GPX4 protein abundance compared with control cells (Figure S5I,J). By contrast, GPX4 knockdown did not alter HIF‐1α protein abundance (Figure 6M,N), consistent with a unidirectional functional relationship in which HIF‐1α acts upstream of GPX4 restoration. Together, these results support a functional requirement for HIF‐1α in Syr‐associated GPX4 recovery and ferroptosis protection, while not establishing direct transcriptional regulation of GPX4 by HIF‐1α.

4. Discussion

In the present study, we combined whole‐organism discovery in C. elegans with mechanistic interrogation in human foreskin fibroblasts to evaluate Syr in ferroptosis‐ and aging‐relevant settings. Previous studies have suggested that Syr exerts anti‐aging activity in mammalian tissues and modulates oxidative and ferroptosis‐related pathways in multiple disease contexts [11]. Our findings extend this body of work in several respects. First, Syr was identified through a phenotype‐first in vivo screening strategy rather than being examined only in preselected cell‐based systems. Second, its effects were not confined to an acute ferroptosis‐relevant injury model but were also associated with improved lifespan and healthspan‐related phenotypes during natural aging in C. elegans . Third, in human fibroblasts, Syr attenuated ferroptotic injury and senescence‐associated phenotypes, and these effects were functionally linked to HIF‐1α‐associated restoration of GPX4. Together, these findings position Syr not simply as a general antioxidant‐like compound but as a natural‐product candidate that attenuates aging‐relevant ferroptosis vulnerability across evolutionarily distinct experimental models.

Our results also address a question in the field: whether ferroptosis is merely a downstream manifestation of cellular decline or a mechanistically informative axis that can be pharmacologically engaged to influence aging‐related phenotypes. Aging is commonly accompanied by glutathione insufficiency, impaired redox buffering, iron dyshomeostasis, and lipid remodeling, all of which increase susceptibility to iron‐dependent lipid peroxidation [23]. In this context, the DEM‐based C. elegans platform provided a practical in vivo screening window that integrates compound uptake, metabolism, and survival under physiological constraints. This point is important because many candidate ferroptosis inhibitors identified in reductionist systems do not necessarily retain activity at the organismal level. An important strength of the present study, therefore, is that Syr was prioritized in vivo and then taken forward for cross‐species validation.

Another notable aspect of this work is that it links acute ferroptosis‐relevant injury with physiological aging within the same experimental framework. In the DEM challenge model, Syr reduced lipid peroxidation, ROS accumulation, and MDA generation, indicating protection against glutathione depletion‐associated oxidative lipid damage. When the analysis was extended to natural aging, the same general stress dimensions remained evident: aging worms showed increased lipid peroxidation, ROS accumulation, and altered labile iron status, whereas Syr attenuated these changes while improving lifespan and multiple healthspan‐related readouts. Although these findings do not indicate that physiological aging is equivalent to ferroptosis, they support the view that aging in C. elegans is accompanied by a progressive ferroptosis‐relevant stress state that is experimentally and pharmacologically tractable. In this sense, the significance of the study lies not only in the protective effect of Syr itself, but also in the internally coherent framework linking glutathione depletion, oxidative lipid stress, iron‐related changes, and functional aging outcomes within one organismal system.

The DEM model should nevertheless be interpreted with appropriate caution. DEM depletes glutathione and creates a cellular environment that favors lipid peroxide accumulation and ferroptosis‐relevant injury, but glutathione depletion can also activate broader oxidative stress pathways. Therefore, in C. elegans , we use the term “ferroptosis‐relevant oxidative lipid stress” to describe the DEM‐induced and aging‐associated phenotypes. The ferroptosis‐related interpretation is strengthened by the reduction of C11‐BODIPY oxidation, MDA, ROS, and labile iron‐related changes, as well as by the use of canonical ferroptosis inducers, RSL3 and erastin, in HFFs. Nevertheless, the organismal DEM model should not be viewed as exclusively ferroptotic.

The HFF data further broaden the significance of the study by showing that Syr‐associated protection is not restricted to nematode physiology. In human fibroblasts, Syr reduced cell death induced by both RSL3 and erastin, preserved cell morphology, suppressed lipid peroxidation and ROS accumulation, and restored GPX4, SLC7A11, and FTL expression. These findings indicate that the protective effects of Syr extend across distinct ferroptosis‐triggering mechanisms and are accompanied by reinforcement of endogenous defense systems involved in antioxidant buffering, lipid peroxide detoxification, and iron handling. Importantly, the study then moved beyond ferroptosis stress alone to senescence‐prone fibroblasts. In both D‐gal‐induced and replicative senescence models, Syr reduced SA‐β‐gal positivity, improved senescent morphology, and attenuated senescence‐associated molecular changes, while simultaneously restoring GPX4 and lowering oxidative lipid stress. This coupling of anti‐ferroptotic and anti‐senescence phenotypes is a notable feature of the present work, because it places Syr at the intersection of two processes that are often discussed together conceptually but are less often examined side by side in the same study. Nevertheless, although Syr simultaneously attenuated senescence‐associated phenotypes and ferroptosis‐relevant oxidative lipid stress, the causal relationship between ferroptosis suppression and senescence delay remains to be fully defined. The current data support a close association and functional overlap between these processes in the tested models, but do not establish that ferroptosis inhibition is the sole or primary mechanism underlying the anti‐senescent effect of Syr. This more cautious interpretation is also consistent with recent pharmacology studies showing that natural products can attenuate ferroptosis‐associated pathology in neurodegenerative settings [12].

The proposed HIF‐1α–GPX4 relationship is biologically plausible, although it should be interpreted with appropriate caution. HIF‐1α is a central regulator of metabolic adaptation and stress responses, but its role in aging is highly context‐dependent and varies across tissues, cell states, and exposure duration [24]. At the same time, independent work has shown that HIF‐1α can regulate GPX4 transcription and suppress ferroptosis in intestinal epithelial cells, supporting the feasibility of an HIF‐1α–GPX4 functional connection [25]. In addition, GPX4‐centered ferroptosis defense can be regulated at multiple upstream levels, including post‐transcriptional control of GPX4 pre‐mRNA maturation, as shown for DDX39B in hepatocellular carcinoma [26]. Other recent work has also highlighted that upstream RNA‐centered regulatory programs can modulate ferroptosis susceptibility in neurological injury models [27]. Natural products can likewise regulate ferroptosis through diverse upstream networks rather than a single conserved route, as illustrated by manoalide‐induced ferroptosis in lung cancer cells [28]. Our data are consistent with a unidirectional hierarchy in fibroblasts, because Syr‐associated restoration of GPX4 was largely weakened by HIF‐1α knockdown, whereas GPX4 knockdown did not reciprocally alter HIF‐1α abundance. Within the limits of the present models, these findings support the interpretation that Syr‐associated protection involves an HIF‐1α–GPX4 defense axis that reinforces ferroptosis resistance.

However, the present evidence should not be interpreted as demonstrating direct transcriptional regulation of GPX4 by HIF‐1α. The loss‐of‐function experiments indicate that HIF‐1α is required for GPX4 recovery and cytoprotection under the tested conditions, whereas additional assays, such as ChIP‐qPCR, promoter‐reporter analysis, and evaluation of HIF‐1α transcriptional cofactors, will be required to determine whether GPX4 is a direct transcriptional target of HIF‐1α in senescent fibroblasts. Similarly, although molecular docking suggested that Syr may be accommodated within a predicted HIF‐1α binding region involving key interacting residues His193, Asp238, and Gln320, this result remains computational and hypothesis‐generating. Direct target engagement has not yet been demonstrated, and additional biophysical and cellular approaches, such as SPR, MST, ITC, CETSA, or DARTS, will be required before direct binding can be concluded. It also remains possible that Syr affects HIF‐1α indirectly, for example through modulation of prolyl hydroxylases, VHL‐dependent degradation, mitochondrial redox balance, or other determinants of HIF‐1α stability.

Several additional limitations should be acknowledged. First, our mammalian evidence is currently limited to fibroblast models. Whether the same protective axis operates in vivo in mammalian aging remains an open question. Second, the concentrations of Syr used in this study are within the range commonly applied in natural‐product phenotypic screening and cellular validation studies, but they are relatively high from a translational perspective. Thus, the present work should be viewed as proof‐of‐concept evidence for biological activity rather than direct evidence of therapeutic feasibility. Future studies should determine the pharmacokinetic profile, tissue exposure, metabolic stability, safety margin, and long‐term efficacy of Syr in mammalian aging models. Third, ferroptosis defense is multi‐layered. Beyond GPX4, other systems such as FSP1‐CoQ10, GCH1‐BH4, and DHODH may also contribute to Syr‐associated phenotypes and should be examined in future work. Finally, because aging is heterogeneous across tissues and cell states, it will be important to define the boundary conditions of Syr action, including dose, exposure duration, stress context, and possible sex‐ or tissue‐specific effects.

Overall, our findings support a model in which Syr‐associated protection is linked to attenuation of ferroptosis‐relevant oxidative lipid stress across species and, in human fibroblasts, involves an HIF‐1α‐associated program coupled to GPX4 restoration. By integrating organismal screening, natural aging phenotypes, mammalian ferroptosis models, and functional dependency testing, this study expands the mechanistic framework through which Syr can be understood in aging‐related settings. More broadly, it supports the view that ferroptosis vulnerability is not merely a byproduct of aging‐associated decline, but an experimentally addressable component of that decline and a potentially useful entry point for natural‐product‐based intervention.

Author Contributions

Xiao‐Gang Zhou and Jian‐Ming Wu supervised and designed the experiments; Ya‐Ping Li and Fei‐Hong Huang collected and processed the data; Meng‐Ting Wu, Meng‐Yi Chen, Yong‐Ping Wen, Xiang Li, Lu Yu, and An‐Guo Wu participated in discussions for the paper; Ya‐Ping Li and Xiao‐Gang Zhou wrote the paper. All authors agree to be accountable for all aspects of work, ensuring integrity and accuracy.

Funding

This work was supported by the Central Nervous System Product Research and Development Key Laboratory of Sichuan Province, Sichuan Credit Pharmaceutical CO. Ltd. (Grant No. 260001‐01SZ), the Key Laboratory of Narcotics Control Technology of Liaoning Province, People's Republic of China (No. LNJD2025‐03), and the Science and Technology Strategic Cooperation Programs of Luzhou Municipal People's Government and Southwest Medical University (2025LZXNYDJC17).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Sub‐screening in the DEM‐induced ferroptosis model in C. elegans and validation of Syr. (A) Schematic timeline of the DEM‐induced ferroptosis paradigm in wild‐type C. elegans (N2), from synchronization to adulthood and DEM exposure, with survival scored at indicated time points. Worms were transferred onto fresh FUdR‐containing NGM plates prior to DEM treatment, as illustrated. (B) Time‐ and concentration‐dependent survival curves of N2 worms exposed to increasing concentrations of DEM (6.25, 12.5, 25, 50, 100, and 200 mM). Survival was recorded at 12, 24, 36, 48, 60, and 72 h after DEM administration. The condition yielding ~50% survival (DEM 100 mM) was selected for subsequent screening/validation experiments. Data are presented as mean ± SD (n = 30 worms per group). (C) Representative brightfield images of worms after DEM (100 mM) challenge with co‐treatment of Fer‐1 (100 μM, positive control) or candidate hits from the primary screen (plate positions labeled). Red arrows indicate typical abnormal/death phenotypes induced by DEM. Scale bars, as indicated. (D) Quantification of survival rates at 72 h corresponding to panel C. DEM markedly reduced survival, whereas Fer‐1 and several candidates significantly restored survival. (E) Representative brightfield images showing dose‐dependent protection by Syr (50, 100, and 200 μM) under DEM (100 mM) challenge, with Fer‐1 (100 μM) as a reference inhibitor. Red arrows indicate abnormal/death phenotypes. Scale bars, as indicated. Unless otherwise specified, experiments were independently repeated three times. Data are shown as mean ± SD. For panel B, two‐way ANOVA (time × concentration) with multiple‐comparisons correction was applied. For panel D, one‐way ANOVA with Tukey's multiple comparisons test was used. *p < 0.05, **p < 0.01, ***p < 0.001 versus DEM; ns, not significant.

Figure S2: Dose–response assessment of ferroptosis inducers/inhibitor and Syr cytotoxicity in HFF cells. (A) CCK‐8 analysis of HFF cell viability after treatment with increasing concentrations of RSL3, demonstrating a dose‐dependent reduction in viability. (B) CCK‐8 analysis of HFF cell viability after treatment with increasing concentrations of erastin, showing dose‐dependent cytotoxicity consistent with ferroptosis induction. (C) CCK‐8 analysis of HFF cell viability following treatment with increasing concentrations of Syr. Syr displayed minimal cytotoxicity at lower‐to‐moderate concentrations, whereas higher concentrations reduced viability. (D) CCK‐8 analysis of HFF cell viability following treatment with increasing concentrations of Lip‐1. Low‐to‐moderate concentrations showed minimal impact on basal viability, whereas high concentrations decreased viability. Data are normalized to Ctrl and presented as mean ± SD (n ≥ 3 independent experiments). One‐way ANOVA with multiple comparisons was applied. *p < 0.05, **p < 0.01, ***p < 0.001 versus Ctrl; ns, not significant.

Figure S3: Effects of Rap and D‐gal on basal HFF cell viability. (A) CCK‐8 assay of HFF cell viability after 24‐h exposure to increasing concentrations of Rap (μM). (B) CCK‐8 assay of HFF cell viability after 24‐h exposure to increasing concentrations of D‐gal (mM). Data are presented as mean ± SD from three independent experiments. One‐way ANOVA followed by Dunnett's multiple comparisons test was performed. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant (vs. Ctrl).

Figure S4: Syr suppresses SASP‐related transcription in D‐gal‐induced or replicative senescent HFF cells. (A) qRT‐PCR analysis of senescence markers and SASP/inflammatory genes (P14, P21, IL‐1α, CXCL3, TNF‐α, MMP‐3, MMP‐9, and MMP‐12) in D‐gal (150 mM)‐induced senescent HFF cells. D‐gal markedly increased the expression of these genes, whereas rapamycin (Rap, 10 μM) and Syr (25 μM) significantly reduced their induction. (B) qRT‐PCR analysis of the same gene panel in Rep HFF cells. Rep cells exhibited elevated transcription of multiple senescence‐ and SASP‐associated genes, and both Rap (10 μM) and Syr (25 μM) reduced their expression to varying degrees. Data are presented as mean ± SEM (n ≥ 3) and normalized to Ctrl. One‐way ANOVA with multiple comparisons was applied. *p < 0.05, **p < 0.01, ***p < 0.001 versus the corresponding model group (D‐gal or Rep), unless otherwise indicated.

Figure S5: HIF‐1α knockdown attenuates the protective effects of Syr against ferroptosis and senescence‐associated phenotypes. (A, B) Validation of HIF‐1α knockdown efficiency by Western blotting (A) with densitometric quantification of HIF‐1α/β‐actin (B). (C) Representative brightfield images of HFF cells under RSL3‐induced ferroptosis (1.56 μM) with or without Syr (25 μM) and HIF‐1α knockdown (HIF‐1α^KD). RSL3 induced ferroptosis‐associated morphological injury, which was alleviated by Syr, whereas HIF‐1α^KD weakened this protection. (D) CCK‐8 quantification of cell viability corresponding to (C), showing that Syr restores viability under RSL3 challenge, whereas HIF‐1α^KD attenuates the Syr‐mediated rescue. Conditions labeled “Exfect” indicate the transfection reagent/vehicle control as shown in the panel. (E, F) Quantification of cell length (E) and width (F) in the D‐gal (150 mM) senescence model. D‐gal increased cell size parameters, Syr partially normalized morphology, and HIF‐1α^KD reduced the magnitude of Syr's improvement. (G, H) Quantification of cell length (G) and width (H) in Rep HFF cells, showing similar trends: Syr improved senescence‐associated morphology, and HIF‐1α^KD diminished this effect. (I, J) Validation of GPX4 knockdown by Western blotting (I) with densitometric quantification of GPX4/β‐actin (J). Data are presented as mean ± SD (n ≥ 3). One‐way ANOVA with Dunnett's multiple comparisons test was used. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant. KD, transient knockdown; Ctrl/Con, empty‐vector control.

Table S1: Primer sequences for qRT‐PCR.

Table S2: Laboratory drug inventory and primary screening survival rate.

Table S3: Revalidation of primary screen hits (survival ⟩ 50%) in a DEM‐induced ferroptosis model.

Table S4: Summary statistics for the C. elegans lifespan assays.

Table S5: Numerical data and statistical results for HFF cell viability assays.

Data Availability Statement

All data supporting the findings of this study are included in the article and its Supporting Information S1. Additional raw data is available from the corresponding author upon reasonable request.