Solasonine inhibits NSCLC malignant progression by regulating SLC7A11 ubiquitination degradation and inducing ferroptosis

Background and purpose

This study responds to the pressing need for novel therapeutic strategies to address key challenges in non-small cell lung cancer (NSCLC) treatment. It explores effective traditional Chinese medicine, investigates its molecular mechanism, and pursues the modernization of traditional Chinese medicine. The current research investigated how Solasonine (SS) extracted from Solanum nigrum L. may counter NSCLC by activating the ferroptosis pathway, while also elucidating the fundamental molecular mechanisms involved.

Methods

Proliferation (CCK8) and metastasis (wound healing) of A549/HCC1833 cells were assessed. Reactive oxygen species (ROS) production, proteomic sequencing, co-immunoprecipitation (Co-IP), and molecular docking mechanisms of the SS treatment group were analyzed. SLC7A11 regulation was evaluated using ferroptosis inhibitors and by ubiquitination assays. SS efficacy was tested in a subcutaneous/lung metastasis nude mouse model.

Results

SS suppressed NSCLC proliferation/metastasis by inducing ferroptosis. Proteomics revealed SS downregulated ferroptosis-related proteins and SLC7A11. SS promoted SLC7A11 ubiquitination/degradation via USP10/TRIM25 interactions, confirmed by Co-IP/docking. In vitro/vivo, SS increased ROS, inhibited tumor growth/metastasis, and activated ferroptosis pathways (reduced SLC7A11/GPX4/GSR/GSS). Immunohistochemistry confirmed the presence of ferroptosis markers in tumors.

Conclusion

SS triggers ferroptosis in NSCLC by disrupting USP10/TRIM25-mediated SLC7A11 stability. This study proposes a novel treatment strategy for NSCLC with a traditional Chinese medicine monomer.

Graphical Abstract

The graphical abstract of the present study. Solasonine, as one of the most important components of Solanum nigrum L., competitively binds to SLC7A11 with the deubiquitinating enzyme USP10, thereby promoting TRIM25 to bind to SLC7A11 and causes SLC7A11 to be ubiquitinated and degraded by TRIM25, stimulating ROS production and causing ferroptosis to inhibit the malignant progression of NSCLC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-026-03516-6.

Highlights

1. Solasonine inhibits non-small cell lung cancer (NSCLC) growth and spread by triggering a specific cell death form termed ferroptosis.

2. This anti-cancer effect involves Solasonine reducing levels of the key protein SLC7A11 and enhancing its degradation through ubiquitination.

3. The novel molecular mechanism reveals that Solasonine promotes SLC7A11 ubiquitination via the USP10/TRIM25 regulatory axis, thereby targeting ferroptosis pathways in NSCLC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-026-03516-6.

Introduction

Non-small cell lung cancer (NSCLC) raises a significant therapeutic challenge because of its refractoriness to conventional treatments [13]. The conventional drugs primarily aim to induce apoptosis in cancer cells; however, these therapies are frequently hindered by multiple limitations. Notably, epigenetic multidrug resistance, low drug targeting efficiency, and high metastasis rates constitute major barriers to effective NSCLC treatment [3, 14, 21]. These issues underscore the urgent need for novel therapeutic strategies that can overcome these limitations [17].

To address these challenges, various modes of cell death for cancer treatment have been investigated, including apoptosis, necrosis, and autophagy. Among these, ferroptosis has emerged as a promising target. Ferroptosis is regulated by various molecular pathways involving intracellular iron accumulation and glutathione (GSH) synthesis suppression-induced lipid peroxidation [15]. This unique mode of cell death offers the potential to treat cancers that develop conventional chemotherapy resistance through potentially delaying or even reversing drug resistance.

Recent advancements in drug-induced ferroptosis have gained considerable attention, particularly in the context of identifying compounds that can effectively trigger this cell death pathway. Among the various agents studied, traditional Chinese medicines have shown promising results in inducing ferroptosis [16]. For example, oxymatrine suppresses liver cancer progression through SIRT1/YY1/GPX4 axis-mediated ferroptosis regulation [6]; shikonin inhibits anaplastic thyroid carcinoma cell growth via enhancing ferroptosis and suppressing glycolysis [27]; and Piceatannol inhibits neuronal ferroptosis resulting from cerebral ischemia–reperfusion in mice, and other studies have been reported in this regard [33]. These natural compounds, with their complex and often multifaceted mechanisms of action, may provide a novel approach to targeting NSCLC through ferroptosis induction.

Solasonine, as a major biological component of Solanum nigrum L., has demonstrated anticancer effects against several malignancies [1, 2, 4, 23, 24, 28]. It is currently approved by the FDA for preclinical experimental use in oncology. In this paper, we focus on the progress made in ferroptosis induction by Solasonine, and by examining the potential of Solasonine in promoting ferroptosis in NSCLC cells, we aim to provide a new therapeutic strategy that could overcome some of the major issues associated with conventional treatments. This study not only contributes to the understanding of ferroptosis-inducing mechanisms but also provides the potential to develop more effective and innovative NSCLC treatment options.

Materials and methods

Ethics statement

The BALB/c nude mice used in the research were obtained from Shanghai Jihui Experimental Animal Feeding Co., Ltd. (Shanghai, China). The mice were aged 4–6 weeks, weighing 15–20 g. Experiments were approved by the Ethics Committee of Shanghai University of Medicine and Health Sciences of Shanghai, China (Approval no.: 2023-GZR-18–340406198707142817-SUMHS-2023–05–08). All intraoperative actions were guided by humanitarian principles to maximize the protection and welfare of the animals. Nude mice were anesthetized with 30 mg/kg pentobarbital sodium, and animals' suffering was minimized throughout the experiment.

Cell culture assays

Normal human lung epithelial Beas-2B cells and NSCLC A549 and HCC1833 cells used in this study were provided by the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in DMEM (Shanghai Basal Media Technologies Co., Ltd., Shanghai, China) and supplemented with penicillin–streptomycin (Gibco, USA) and 10% FBS (bio-explorer, USA). A549 and HCC1833 cells were treated for 24 h with specific Solasonine (CAS: 19121–58-5, S9144, purity: 99.62%, Selleck, USA) concentrations in DMSO (0–50 μM), Cisplatin (CAS: 15663–27-1, S1166, purity: 99.84%, Selleck, USA) in DMSO (0–50 μM).

Quantitative reverse transcription PCR (RT-qPCR) analysis

Total RNA was extracted from tumor cells or tissues. RNA sample purity and concentrations were tuned using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). Retrotranscriptional reaction conditions were: 37℃, 15 min × 3; 85 °C, 5 s at 4 °C.

Thermal cycle conditions of RT-qPCR were 95.0 °C for 30 s (hold stage); 95.0 °C for 10 s and 60.0 °C for 30 s (PCR stage for 40 cycles); 95.0 °C for 15 s and 60.0 °C for 60 s (melt curve stage).

RNA was transcribed reversely into complementary DNA using SuperScript II Reverse Transcriptase (Thermo Fisher Scientific). RT-qPCR was performed through the AB 7300 Real-Time System (Applied Biosystems, Foster City, CA, USA) using primer pairs for SLC7A11 and β-actin (Shanghai Shenggong Biological Engineering Co., Ltd., Shanghai, China) and TaqMan Universal PCR Master Mix (Thermo Fisher Scientific). Gene expression was quantified via the 2 − ΔΔCt approach. Primers used to assay SLC7A11 expression were (forward/reverse): 5′-TCCTGCTTTGGCTCCATGAACG-3′/5′-AGAGGAGTGTGCTTGCGGACAT-3′. Additionally, primers utilized to assay β-actin expression were 5′-CACCATTGGCAATGAGCGGTTC-3′/5′-AGGTCTTTGCGGATGTCCACGT-3′.

Western blot (WB) assays

Total protein from the NSCLC cells was extracted with RIPA lysis buffer. Protein samples were separated by 10% SDS-PAGE, which were transferred to PVDF membranes (Bio-Rad Laboratories, Hercules, CA, USA). The membranes were blocked applying skimmed dry milk (Yili Milk Company, Inner Mongolia, China) for 1 h and hybridized using SLC7A11 monoclonal antibody (1:1000 dilution; AF301472; AiFang Biological, China), GPX4 monoclonal antibody (1:1000 dilution; 67763–1-Ig; Proteintech, China), GSS monoclonal antibody (1:5000 dilution; 67598–1-Ig; Proteintech), GSR polyclonal antibody (1:1000 dilution; 18257–1-AP), USP10 polyclonal antibody (1:5000 dilution; 67917–1-Ig), TRIM25 monoclonal antibody (1:5,000 dilution; 67314–1-Ig), and β-actin antibody (1:20,000 dilution; 66009–1-1 g). Immunoreactive bands were visualized through chemiluminescence (Thermo Fisher Scientific). For pairwise comparisons, least significant difference (LSD) t-test was applied. For comparisons across multiple groups, one-way ANOVA was employed for datasets following normal distributions. ImageJ was utilized for image analyses.

Co-immunoprecipitation (Co-IP) assays

In the Co-IP experiment, we obtained protein samples from cells utilizing a special cracking buffer in the Co-IP kit (Biolinkedin, China). Protein A/G magnetic beads were washed in immunoprecipitation (IP) buffer before being used. Protein concentration of lysates following lysis was determined using a BCA kit (Yeason, China). Lysate was incubated with the IP antibody or IgG homologous control at 4 °C and subsequently incubated. Immunoprecipitates were analyzed by WB, or the samples were tested for ubiquitination omics by mass spectrometry.

Transwell assays

Transwell culture chambers (Guangzhou Jet Bio-Filtration Co., Ltd., China) were used for migration experiments. In each chamber, 30,000 cells were allowed to proliferate. After 1 d, they were washed twice using pre-cooled PBS and dyed with crystal violet. Pre-cooled PBS was used for washing and the cells were air dried. The analysis was performed using software ImageJ.

Scratch wound assays

For scratch wound assay, A549 and HCC1833 cells (3 × 10⁵) were cultured in DMEM in six-well plates. When a confluency of 90 ~ 95% was achieved, a “scratch” was made utilizing a pipette tip. Subsequently, wells were washed twice with PBS to clean detached cells, which were incubated with Solasonine in DMSO (0, 7.5, and 15 μM for A549; 0, 15, and 30 μM for HCC1833). Wound images were obtained utilizing an inverted microscope.

Lentiviral transfection

Each A549 and HCC1833 cell group was inoculated with 2 × 10⁵cells/well. Following a 6-h starvation, 20 µl PGMLV-CMV-MCS-ZsGreen1-T2A-Puro as a carrier of SLC7A11 lentiviral fluid or empty liquid was added. This was mixed well and incubated overnight. Cells were screened with 2 µg/mL puromycin for 3 d, transfection was confirmed, and transfection efficiency was evaluated by WB and RT-qPCR.

Proteomic data collection and analyses

HCC1833 cells were seeded and treated with or without Solasonine (30 μM). Cells were washed with PBS and lysed with RIPA. Protein concentration was determined using a BCA protein assay kit (Pierce, IL, USA) following the manufacturer’s protocol. Cleared lysates were stored at − 80 °C and subsequently 250 μg total protein was reduced with 10 mM DTT and alkylated. The sediment was treated with 1.5 mL prechilled 100% acetone and centrifuged three times. Each sample was air-dried for 2 ~ 3 min and diluted. Peptides were desalted using SPE C18 cartridges (Thermo Fisher Scientific). This was performed with three replicates to verify protein identification accuracy, reliability, and reproducibility.

Ubiquitinomics

We established two sample groups, Control and SS treatment, and performed Co-IP with anti-SLC7A11 antibody. The key steps were: 1) Cell lysis and protein extraction; 2) Immunoprecipitation of SLC7A11-protein complexes; 3) Washing and elution; 4) SDS-PAGE separation and tryptic digestion; 5) LC–MS/MS-based ubiquitinome profiling; 6) Bioinformatics analysis of differential ubiquitination sites. Samples were subjected to ubiquitinomics detection to identify SS-induced ubiquitination alterations.

Molecular docking analysis

The USP10 structure is based on an alpha-fold. Molecular docking was performed with Autodock vina, while Solasonine’s three-dimensional (3D) structure was built using Chem 3D, and the MM2 module in Chem 3D was utilized for energy minimization. During docking, the parameters were set to a 60 Å × 60 Å × 60 Å grid box, the box size including the protein pocket. Grid spacing was set to 0.375 Å, and subsequent data were analyzed by Pymol.

Molecular dynamics simulation

Molecular dynamics simulation was performed on a super-computing platform. The complexes were solvated using an SPC water model. The protein topology was generated through GROMOS96 53a6 force field and the ligand topology using a PRODRG server. To eliminate steric clashes, systems were subjected to the steepest energy minimization and provided a maximum force < 1000 kJ/mol/nm. Particle Mesh Ewald was applied to compute long-range electrostatic forces. After energy minimization, position restraint simulation of 5000 steps was implemented with constant particle, volume, and temperature conditions. Lastly, a 300 ns MD simulation was performed on each ensemble. The root meaned square deviation (RMSD) and relative root mean square fluctuation (ΔRMSF) were calculated by GROMACS rmsd and rmsf tools.

Binding free-energy calculation:

The binding free energies for WT and mutated α-L-rhamnosidase complexes were computed based on the MD trajectory approach for each complex, of which MM/PBSA was obtained using the following equations:

The free energy of the molecules is represented by

Molecular mechanics potential energy, E_MM, including both the bonded and non-bonded energy, was calculated with the following equation:

Here, , , and are bonded interactions consisting of bond, angle, and dihedral, respectively. Non-boned interactions include van der Waals () and Coulomb charge effects ()

MM/PBSA solvability consists of polar () and apolar (). G_polar is determined from the Poisson-Boltzmann (PB) equation. G_apolar is estimated following the surface area. is a coefficient related to the solvent surface tension. A is the solvent accessible surface area (SASA), and b is the fitting parameter.

Briefly, the MM/PBSA calculation formula is as follows,

Pulmonary metastasis assays

Stably transfected light-emitting labeled HCC1833 (Luc-HCC1833) cells were employed to evaluate metastasis. Nude mice were injected with light-emitting labeled HCC1833 (Luc-HCC1833) or Luc-HCC1833 cells, which overexpressed SLC7A11, through the tail vein. Lung metastases were confirmed 4 weeks after injection using a bioluminescence imaging system. Solasonine dissolved in DMSO and cosolvent (40 mg/kg) was injected every 3 days, and the Control, LV-Ctrl, and LV-SLC7A11 groups were injected with DMSO and cosolvent for Control. The intact lung tissues were stained using hematoxylin and eosin (HE).

CCK8 assays

Beas-2B, A549, and HCC1833 cells were seeded into 96-well plates. These cells were maintained for 24 h with various Solasonine concentrations in DMSO (0–50 μM), or Cisplatin in DMSO (0–50 μM) for 0, 1, 2, and 3 d before follow-up experiments. The CCK8 assay (Yeasen Biotechnology Co., Ltd., Shanghai, China) was applied to determine cell viability. The absorbance was determined at 450 nm.

HCC1833 xenograft model

BALB/c nude mice were housed, and each mouse was injected with HCC1833 cells, which have tumor formation capacity, suspended in saline and the animals were assigned randomly to: (1) control group was injected with the same solvent but without the target drug as a vehicle control; (2) intervention groups receiving different drugs in an equivalent volume of mixed liquor. All the treatment routes were via caudal intravenous injection. Animals were treated five times, once every second day. Tumor sizes were calculated every 2 days. At 7 days post cell microinjections, we performed the experiments. After 12 days, the mice were euthanized, and tumors were weighed. Tumor size was determined every 2 days with a vernier caliper from the equation, V = length × width²/2. The concentrations of the injected drugs were as follows: In the Solasonine group, the Low-dose group was 20 mg/kg, the Middle-dose group was 40 mg/kg, and the High-dose group was 60 mg/kg. Except for the animal experiments in Fig. 1, in the remaining animal experiments, the injected Solasonine doses were all of that used in the Middle-dose group.

Fig. 1.

Solasonine treatment inhibited NSCLC malignant progression. A CCK8 data show the proliferation ability regarding both A549 and HCC1833 cells after treatment with various Solasonine concentrations for 24 h (n = 3), each experiment was independently repeated three times. B CCK8 data show the proliferation ability regarding both A549 and HCC1833 cells post treatment with the IC₅₀ concentration for 0, 24, 48, and 72 h (n = 3), each experiment was independently repeated three times. C-F Colony formation data revealing the cell proliferation ability of both A549 (C-D) and HCC1833 (E–F) cells after treatment with different Solasonine concentrations (0, 7.5, and 15 μM for A549 and 0, 15, and 30 μM for HCC1833) for 24 h. G-H Representative HCC1833 tumor formation figures in nude mouse xenografts (n = 5). I Tumor volume was tracked every 2 days. J Tumor weight was determined 12 days after injection. K-L The immunohistochemical data illustrating Ki67-positive cell percentage. M–N The immunofluorescence data illustrating the TUNEL-positive cell percentage. O-P Live image detection showing luc-H1833 cell pulmonary metastasis (n = 5). Q-R The numbers of metastatic foci in lung tissues were calculated according to the H&E staining. Data are denoted by means ± SD. ns, no significant difference. **P < 0.01, ***P < 0.001

Immunohistochemical analysis

Tumor specimens were fixed in paraffin and sectioned into slices of 5 μm thickness. The relevant index of each nude mouse post various treatments was validated using an AxioPhot light microscope (Carl Zeiss AG, Oberkochen, Germany). Positive cell proportions within the total population were utilized as the standard in statistical analyses. For comparisons, LSD t-test and one-way ANOVA were applied. ImageJ was utilized for image processing.

Immunofluorescence cytochemistry

For protein immunofluorescence, tissues were fixed in 4% polyformaldehyde. Post permeabilization with 0.1% Triton X-100 in PBS, blocking was performed with 3% BSA. The tissues were incubated with antibodies against SLC7A11 (1:100 dilution; AF301472; AiFang Biological) or GPX4 (1:400 dilution; 67,763–1-Ig; Proteintech). For the secondary antibody, Alexa Fluor 594 anti-rabbit IgG was utilized. The detection of TUNEL levels in tumor tissues was performed through a TUNEL Assay Kit (Fluorescence, 488 nm, #25,879, CST, USA). Fluorescence was determined via Image Xpress High Content Imaging (Molecular Devices, CA, USA) or images were obtained with a Leica confocal microscope TCS SP5 via objectives. Laser intensity, magnification, and microscope settings for each channel were maintained.

EdU analysis

The EdU Cell Proliferation Assay Kit (UE, China) was employed to determine cell proliferation. A549 and HCC1833 cells were seeded into 24-well plates. Once 60–70% confluence was attained, the cells were incubated with Solasonine in DMSO. Cells were fixed in 4% paraformaldehyde and images were obtained with an inverted fluorescence microscope (Leica). The proliferation rate was determined as the ratio of EdU-positive cell numbers to the total Hoechst 33,342-positive cell counts.

Production of reactive oxygen species (ROS)

ROS levels were determined using a ROS detection kit (Bioscience Biotechnology, Co., Ltd., Shanghai, China). Images were obtained using an inverted fluorescence microscope (Leica Microsystems GmbH).

Colony formation assay

HCC1833 and A549 cells were seeded into six-well plates and post-treated with Solasonine in DMSO. Following colony formation, the cells were fixed in 4% paraformaldehyde, dyed, washed three times with PBS, air dried, and images obtained for statistical analysis.

Statistical analyses

Statistics analyses were performed using Prism (GraphPad, California, USA). For pairwise comparisons, the LSD t-test was applied. For comparisons across several groups, one-way ANOVA was employed for datasets adhering to normal distribution. Friedman test was used for non-normally distributed data. A p-value < 0.05 was regarded as being statistically significant.

Results

Solasonine treatment significantly suppressed NSCLC malignant progression

CCK8 assays were performed to evaluate SS (the crystal structure is shown in Fig S2A) cytotoxicity (0, 10, 20, 30, 40, and 50 μM) against Beas-2B, HCC1833, and A549 cells. SS cytotoxicity increased gradually in a concentration-dependent manner (Fig. 1A), with IC₅₀ values for the A549 and HCC1833 cells of 14.82 and 30.62 μM, respectively. When the same concentration gradient of SS was used to determine its cytotoxicity in normal human lung epithelial Beas-2B cells, the IC₅₀ concentration was 32.08 μM (Fig S2B), which was different from the sensitivity of the two lung cancer cells. Sensitivity of A549 and HCC1833 cells to Cisplatin, a commonly used clinical chemotherapy drug, was respectively detected with the same concentration gradient. As shown in Fig S2C, the IC₅₀ concentration of A549 was 17.42 μM and that of HCC1833 was 26.77 μM. The sensitivity of these two types of cells to DDP was somewhat different from that to SS. Furthermore, SS inhibitory effects on A549 (15 μM) and HCC1833 (30 μM) cell proliferations were tested over a prolonged time period (0, 24, 48, and 72 h) (Fig. 1B). SS increasingly suppressed A549 and HCC1833 cell proliferation with time. Furthermore, clone formation experiments (Fig. 1C-F), EdU incorporation assays (Fig S1A-B), cell scratch tests (Fig S1E–G), and transwell migration experiments (Fig S1C-D) collectively demonstrated that SS exhibits inhibitory effects upon both cell migration and proliferation.

Subsequently, the SS therapeutic efficacy was assessed in nude mice bearing subcutaneous xenografts of HCC1833 tumors (Fig. 1G). Mice were stratified into four groups: a negative control (NC) group and three Solasonine groups of different doses, namely low dose, middle dose, and high dose, with five mice in each group. Once the tumor volume in the mice attained 40 mm³, the groups were administered either no-load solvent (NC group) or SS by intravenous injection every 2 days for 10 days. As depicted in Fig. 1H-J, a notable tumor volume and weight reduction was detected in the SS groups compared to the NC group, suggesting a superior therapeutic effect of SS. Post-treatment, the tumors were harvested for Ki67 histological analysis. Immunohistochemical analysis revealed a substantial decrement in Ki67 expression following SS treatment (Fig. 1K-L). Specifically, the TUNEL positivity rate increased significantly in the SS groups compared to the NC group (Fig. 1M–N), indicating that SS effectively inhibits tumor proliferation in mice harboring HCC1833 tumors. The results showed that, as expected, the therapeutic effect of medium and high dose drug injections was more significant, and there was no significant difference in therapeutic effects between the middle and high doses. Adhering to the therapeutic principle of choosing the lowest dose that achieves a therapeutic effect, we consequently adopted the middle dose as that used for treating animals.

Furthermore, a metastatic lesion model was employed to investigate SS's influence on HCC1833 cell invasion from the tail vein to the lungs following tail vein injection. An anti-lung metastasis experiment with SS was conducted, and a lung invasion model of HCC1833 tumors was established by injecting Luci-labeled HCC1833 cells via the tail vein. Concurrently, SS was administered via tail vein injection for intervention, once every 3 days for a total of five administrations. The impact of SS on lung metastatic tumors was evaluated by in vivo fluorescence imaging. A marked decrease in vivo fluorescence intensity was observed after SS treatment (Fig. 1O-P), indicating that SS significantly inhibited the lung metastasis of HCC1833 cells. Additionally, H&E staining of lung tissues confirmed a reduction in the number of lesions (Fig. 1Q-R), further validating that SS effectively suppresses lung metastasis of HCC1833 cells.

Proteomics showed that Solasonine promotes ferroptosis by down-regulating SLC7A11 protein levels

On the basis of previous experimental findings, it was evident that the compound SS exhibits potent inhibitory effects upon NSCLC cell proliferation and migrations, specifically A549 and HCC1833 cells. To elucidate the underlying mechanism of SS's anti-NSCLC activity, proteomic sequencing was conducted. Analysis of the volcano plot (Fig. 2A) derived from the sequencing data revealed a statistically significant difference in protein levels between HCC1833 cells treated with SS (30 μM) and untreated controls. Furthermore, the bubble plot (Fig. 2B) indicated that the impact of SS on protein levels is predominantly focused on pathways associated with ferroptosis. The proteomic heatmap (Fig. 2C) highlighted the downregulation of a ferroptosis-related protein, SLC7A11. Previous studies have reported that SS induces ferroptosis in tumor cells through modulating SLC7A11, albeit with an incomplete understanding of the specific mechanism [30]. Therefore, the focus of the present research was to study the precise mechanisms by which SS regulates SLC7A11.

Fig. 2.

Solasonine promotes ferroptosis by down-regulating SLC7A11 protein levels in vitro. A Volcano plots illustrated the proteins that were differentially expressed after proteomics analysis. B Bubble plots illustrated KEGG pathway enrichment for the proteomics result of HCC1833 cells with or without Solasonine treatment. C Clustered heat map of HCC1833 cells post Solasonine treatments compared with control cells. D-G Western blot experiments showing the SLC7A11, GPX4, GSS, and GSR protein levels in both A549 and HCC1833 cells with or without drug (Solasonine at IC₅₀, Fer-1, and DMSO) treatment, n = 3. H-I ROS production was detected in different drug treatment (Solasonine at IC₅₀, Fer-1 at 16 μM, and DMSO) groups. J GSH levels were detected in different drug treatment (Solasonine at IC₅₀, Fer-1 at 16 μM) groups in A549 and HCC1833 cells. Data are denoted by means ± SD. ns, no significant difference. *P < 0.05, **P < 0.01, ***P < 0.001

Solasonine promotes ferroptosis by down-regulating SLC7A11 protein levels in vitro

To ascertain whether SS inhibits the SLC7A11 protein level, thereby facilitating the ferroptosis pathway in A549 and HCC1833 cells, the ferroptosis inhibitor Fer-1 was utilized, and WB experiments were conducted. The results demonstrated that SS reduces the SLC7A11 protein level (Fig. 2D-E), and the addition of Fer-1 reverses this effect, confirming that SS influences ferroptosis by downregulating SLC7A11. Additionally, the impact of SS on key proteins associated with ferroptosis, such as GPX4, GSR, GSS, and GSH, was investigated. The WB results indicated that SS decreases the GPX4, GSR, and GSS protein levels (Fig. 2F-G). Notably, there was a linear correlation between the increasing Fer-1 concentration and the reversal of the downregulation of the GPX4, GSR, and GSS protein levels. Concurrently, a reagent kit was employed to determine GSH levels (Fig. 2J), a crucial indicator of ferroptosis, and a similar trend was observed. Furthermore, the effect of SS on cellular ROS was examined. SS significantly elevated the intracellular ROS levels, which was reversed by Fer-1 (Fig. 2H-I), indicating that SS promotes ROS production. These data provide new insight into mechanisms underlying SS's anti-NSCLC activity and its role in regulating ferroptosis-associated proteins and pathways.

Having established that SS inhibits the functionality of A549 and HCC1833 cells via the ferroptosis pathway, we confirmed the ferroptosis impacts on proliferation and migration of these cell lines through the ferroptosis inhibitor Fer-1. Through the utilization of EdU (Fig. 3A-C), clone formation (Fig. 3D-E), transwell (Fig S1J-K), and scratch assays (Fig S1H-I), it was unequivocally demonstrated that Fer-1 significantly reverses the inhibitory effects of SS on A549 and HCC1833 cell proliferation and migration.

Fig. 3.

Solasonine promotes ferroptosis by down-regulating SLC7A11 protein levels in vitro. A-C EdU detection shows the proliferation ability of both A549 and HCC1833 cells after different drug treatment (Solasonine at IC₅₀, Fer-1 at 16 μM) for 24 h. D-E Colony formation data revealing the cell proliferation ability of both A549 and HCC1833 cells after different drug treatment (Solasonine at IC₅₀, Fer-1 at 16 μM) for 24 h. F RT-qPCR assay shows the mRNA level of SLC7A11 in the A549 cell line before and after its overexpression. G-H Western blot experiment data showed the SLC7A11 protein levels in the A549 cell line before and after overexpression, n = 3. I-J Western blot experiment data showed the SLC7A11, GPX4, GSS, and GSR protein levels with or without Solasonine (IC₅₀) treatment in the A549 cell line before and after overexpression, n = 3. K-L ROS production was detected in different groups which were with or without Solasonine (IC₅₀) treatment in the A549 cell line before and after overexpression. (M) GSH levels were detected in different groups which were with or without Solasonine (IC₅₀) treatment in the A549 cell line before and after overexpression. Data are denoted by means ± SD. ns, no significant difference. *P < 0.05, **P < 0.01, ***P < 0.001

To validate SLC7A11’s roles in the context of SS-induced ferroptosis, a rescue experiment was designed and executed. Initially, an SLC7A11-overexpressing A549 cell line (designated as LV-SLC7A11) was successfully generated via lentiviral transfection. As evident from Fig. 3F–H, SLC7A11 mRNA and protein levels in modified A549 cells were markedly upregulated, confirming successful establishment of the LV-SLC7A11 cell line. Subsequently, the effects of SS intervention on key ferroptosis-related proteins, including GPX4, GSR, GSS, and GSH, were investigated in the LV-SLC7A11 cells through WB analysis. SS decreased SLC7A11, GPX4, GSR, and GSS protein levels (Fig. 3I-J). However, the high expression of SLC7A11 attenuated the impact of SS on the protein levels of these ferroptosis-associated markers. Similarly, the GSH level (Fig. 3M), a crucial indicator of ferroptosis, exhibited a comparable trend. Consistent with these findings, EdU (Fig S1P-Q), clone formation (Fig S1N–O), transwell (Fig S1L-M), scratch (Fig S1R-S), and ROS (Fig. 3K-L) experiments all demonstrated that the elevated SLC7A11 expression diminishes the inhibitory effects of SS on A549 cells.

Solasonine promotes TRIM25-mediated ubiquitination degradation of SLC7A11 by targeting USP10

Previous investigations have established that SS promotes ferroptosis in A549 and HCC1833 cells by downregulating the SLC7A11 protein level. Ubiquitination is an indispensable mode of protein post-translational modification and is important in ferroptosis regulations [5, 8–11, 32, 34, 35]. 12. E3 ubiquitin ligases and deubiquitinating enzymes are the most significant enzymes in the ubiquitin system, and their dysregulation is closely associated with the progression of many cancers [7, 8, 19, 29, 31]. During these ubiquitin-system enzymes affect cell sensitivities to iron death through regulating the ubiquitination status of specific proteins. The ubiquitination degradation pathway has been implicated in this process, however, its precise mechanism remains unclear. To elucidate the degradation dynamics of the SLC7A11 protein, we employed CHX and MG132 in conjunction with WB experiments [8, 12, 21, 22]. Prolonged CHX treatments failed to prevent SS-induced downregulation regarding SLC7A11 protein (Fig. 4A-B), showing a consistent decline. By contrast, with the extension of the MG132 treatment time (Fig. 4C-D), the SS-mediated down-regulation of the SLC7A11 protein level was reversed and accompanied by a significant upward trend. These findings suggested that SS intervention may facilitate the ubiquitination degradation of SLC7A11 protein.

Fig. 4.

Solasonine promotes TRIM25-mediated ubiquitination degradation of SLC7A11 by targeting USP10. A-B Western blot experiment data showed the SLC7A11 protein levels after different drug treatment (Solasonine at IC₅₀, CHX at 40 μM) with different treatment times in A549 cells, n = 3. C-D Western blot experiment data showing the SLC7A11 protein levels after different drug treatment (Solasonine at IC₅₀, MG132 at 10 μM) with different treatment times in A549 cells, n = 3. E Ubiquitination-related protein detection. SLC7A11 with magnetic beads enriched the total protein from A549 cells that was collected and detected by protein mass spectrometry. F Schematic diagram of molecular docking data. Solasonine was bound to the SLC7A11 active site by five hydrogen bonds to form a complex. G-H Co-IP experiment showed that SLC7A11 was directly bound to USP10 and TRIM25 before and after Solasonine (IC₅₀) treatment in A549 cells, n = 3. I-J Western blot experiment showed the SLC7A11, USP10, and TRIM25 protein levels before and after Solasonine (IC₅₀) treatment in A549 cells, n = 3. Data are denoted by means ± SD. ns, no significant difference. *P < 0.05, **P < 0.01, ***P < 0.001

Ubiquitination-related protein detection of SLC7A11 with magnetic beads enriched the total protein obtained from A549 cells and it was detected by protein mass spectrometry. The proteins that were bound relatively strongly to the SLC7A11 protein were analyzed. To further identify the underlying ubiquitination machinery, after analyzing the mass spectrum results (Fig. 4E), TRIM25 was selected as one of its E3 ligases with the greatest binding force, and USP10, the only deubiquitinating enzyme, for follow-up study. To substantiate the binding of SS to USP10, protein molecular docking experiments were performed (Fig. 4F). The results revealed a binding energy of − 8.360 kcal/mol between SS and USP10, facilitated by hydrogen bonding interactions involving Asp523, Glu525, and Lys687, and the results of multiple molecular docking experiments of different forms and molecular dynamics simulation experiments all confirmed our conjecture (Fig S2D-H).

Additionally, CO-IP experiments were conducted to demonstrate the binding interactions among SLC7A11, USP10, and TRIM25 of A549 cells. SS intervention significantly reduced the binding of SLC7A11 to USP10 while concurrently enhancing its direct binding to TRIM25 (Fig. 4G-H). The results obtained from the INPUT group experiments (Fig. 4I-J) further supported this hypothesis. It is proposed that SS may downregulate SLC7A11 expression by competitively binding to USP10, a deubiquitinase, thereby disrupting the interaction between SLC7A11 and USP10. This separation triggers SLC7A11 degradation via the TRIM25 proteasome pathway.

Solasonine promotes ferroptosis by down-regulating SLC7A11 levels in vivo

Furthermore, to validate the ferroptosis roles in an HCC1833 subcutaneous tumor xenograft mouse model, we employed the ferroptosis inhibitor Fer-1 (Fig. 5A). Mice were stratified into four groups: NC, Fer-1, SS, and Fer-1 + SS groups, with five animals per group. Once the tumor volume attained 40 mm³, mice in the respective groups were administered intravenous injections of no-load solvent (NC group), SS, or Fer-1 every 2 days for a duration of 10 days. As illustrated in Fig. 1H, post-treatment digital images of the tumors demonstrated a significant inhibition of tumor growth in the SS group. Conversely, in the Fer-1 + SS group, tumor volume inhibition was not significant compared to the SS group, suggesting that tumor volume inhibition reappeared (Fig. 5B-C). This finding was corroborated by the changes in tumor weight (Fig. 5D).

Fig. 5.

Solasonine promotes ferroptosis by down-regulating SLC7A11 levels in vivo. A-B Representative HCC1833 tumor formation figures in nude mouse xenografts (n = 5). C Tumor volume summaries were tracked every 2 days. D Tumor weight was determined 12 days after injection. E–F The immunohistochemical data illustrating Ki67-positive and TUNEL-positive cell percentages. G-H The immunofluorescence detection showed SLC7A11 expression in tumor tissues. I-J The immunofluorescence detection showed GPX4 expression in tumor tissues. K-L Live image detection showing luc-HCC1833 cell pulmonary metastasis (n = 5). M–O The numbers of metastatic foci in lung tissues were calculated according to the H&E staining. P-Q The immunofluorescence detection showed SLC7A11 expression in lung tissues. R-S The immunofluorescence detection showed GPX4 expression in tumor tissues. Data are denoted by means ± SD. ns, no significant difference. *P < 0.05, **P < 0.01, ***P < 0.001

Subsequent immunohistochemical analysis by Ki67 and TUNEL staining evaluated the impact of SS-induced ferroptosis on HCC1833 tumors. The SS treatment group exhibited a notable decrease in Ki67 expression and increase in TUNEL expression compared to the Fer-1 + SS group (Fig. 5E–F). Specifically, the Ki67 positivity rate decreased and the TUNEL positivity rate increased substantially in the SS group relative to the NC group. Additionally, immunohistochemical analysis of the SLC7A11 and GPX4 proteins was conducted to assess SS intervention effects upon ferroptosis-associated markers in HCC1833 tumors. The SLC7A11 and GPX4 expression levels were downregulated significantly in the SS treatment group (Fig. 5G-J). Importantly, the co-administration of Fer-1 reversed the SLC7A11 and GPX4 downregulations, further supporting the notion that SS inhibits HCC1833 tumors by inducing ferroptosis in HCC1833 cells at the tissue level.

To investigate the influence of SS-induced ferroptosis on HCC1833 cell invasion into the lungs, a metastatic lesion model was utilized with the ferroptosis inhibitor Fer-1. An anti-lung metastasis experiment was conducted, involving the establishment of a lung invasion model of HCC1833 tumors through tail vein injection of Luci-labeled HCC1833 cells. Concurrent interventions included tail vein injections of SS or Fer-1, administered once every 3 days for a total of five administrations. The effect of SS on lung metastatic tumors was evaluated through in vivo fluorescence imaging. A marked decrease in vivo fluorescence intensity was observed in the SS treatment group when comparing to the Fer-1 + SS group Fig. 5K-L, indicating that SS significantly inhibits lung metastasis of HCC1833 cells by inducing the ferroptosis pathway. These findings were further confirmed by H&E staining of lung tissues and whole lung tissue images (Fig. 5N-S).

Discussion

The poor prognosis associated with NSCLC underscores the urgent demand with respect to novel therapeutic strategies to overcome the major obstacles posed by multidrug resistance, low drug targeting efficiency, and high metastatic rates in chemotherapy [18, 20, 25, 26]. In the current research, we identified SS as a promising agent with exceptional antitumor properties. Our findings demonstrated that SS effectively inhibits tumor proliferation, growth, and metastasis particularly in animal models where it suppressed not only the growth of HCC1833 xenografts but also their pulmonary metastasis.

The proteomic sequencing conducted to elucidate the molecular mechanisms underlying SS's antitumor efficacy revealed a predominant impact on pathways associated with ferroptosis. Notably, the downregulation of a key ferroptosis-associated protein, SLC7A11, was highlighted in the proteomic heatmap. After conducting an extensive literature research, we found that previous studies have suggested that SS's role in inducing ferroptosis is by modulating SLC7A11, albeit with a limited understanding of the specific mechanisms involved [30]. The current investigation determined if Solasonine actually reduces the SLC7A11 level as predicted and by what means it achieves this goal. Our current research deepens this understanding by demonstrating that SS indeed inhibits SLC7A11 expression and that this inhibition is attenuated with the ferroptosis inhibitor Fer-1, thereby confirming SS-mediated tumor suppression through the induction of ferroptosis in A549 and HCC1833 cells.

Furthermore, our results elucidated that SS promotes ferroptosis by inhibiting additional ferroptosis-associated protein expressions, including GPX4, GSR, GSS, and GSH. The use of Fer-1 to intervene in SS's therapeutic effect demonstrated a reduction not only in SS's ability to inhibit tumor proliferation, growth, and metastasis but also in its inhibitory effect on SLC7A11 and the aforementioned ferroptosis-associated proteins. These findings collectively suggested that SS activates the ferroptosis pathway by inhibiting SLC7A11, thereby facilitating tumor cell death. Animal studies further corroborated this process.

To gain insights into the molecular mechanisms by which SS downregulates SLC7A11, we conducted experiments with CHX and MG132, which revealed that SS inhibits SLC7A11 expression through ubiquitination degradation. Ubiquitinomics mass spectrometry identified USP10 and TRIM25 as key players in the ubiquitination degradation of SLC7A11. Molecular docking simulations indicated a strong affinity between SS and USP10, while Co-IP experiments demonstrated that SS intervention significantly reduces the binding between SLC7A11 and USP10 while enhancing direct binding between SLC7A11 and TRIM25. On the basis of these observations, we speculate that SS might downregulate SLC7A11 expression via competitively binding to the deubiquitinase USP10, disrupting the interaction between SLC7A11 and USP10, and thereby inducing SLC7A11 degradation via the TRIM25 proteasome pathway.

Conclusion

Our research demonstrated that SS, a steroidal alkaloid derived from S. nigrum L., exhibits significant antitumor activity against NSCLC by inducing ferroptosis through the downregulation of SLC7A11 and other ferroptosis-associated proteins. Proteomic analysis and further experiments elucidated the mechanism by which SS promotes SLC7A11 ubiquitination and degradation via interactions with USP10 and TRIM25. Our research results indicated that Solasonine is a candidate drug for clinical application in the treatment of NSCLC, which provides a new therapeutic strategy for the clinical treatment of NSCLC. (Graphic Abstract).

Abbreviations

CCK8

Cell Counting Kit 8

CHX

Cycloheximide

DTT

DL-Dithiothreitol

Edu

5-Ethynyl-2'-deoxyuridine

Fer-1

Ferrostatin-1

GPX4

Phospholipid hydroperoxide glutathione peroxidase 4, PHGPx 4

GSH

Glutathione

GSR

Glutathione reductase

GSS

Glutathione synthetase

HMOX1

Heme oxygenase 1

IC₅₀

50% Inhibitory concentration

NSCLC

Non-small cell lung cancer

ROS

Reactive Oxygen Species

RT-qPCR

Quantitative reverse transcription PCR

SIRT1

Silent information regulator 1

SLC3A2

Solute carrier family 3 member 2

SLC7A11

Solute Carrier Family 7 Member 11

SS

Solasonine

TCM

Traditional Chinese medicine

TF

Transferrin

TRIM25

Tripartite motif-containing 25

USP10

Ubiquitin-specific peptidase 10

WB

Western blot.

YY1

Yin Yang 1

Authors’ contributions

MJ, FX, GH, and XX conceived and performed the experiments and data analyses and drafted the manuscript; YW designed and performed the experiments and data analysis; XL, CZ, MQ, ZZ, NW, and WY performed the experiments and data analysis. All authors agree to be accountable for all aspects of this work, ensuring its integrity and accuracy.

Funding

This word was funded by National Natural Science Foundation of China (No. 52103171, 82127807), Shanghai "Science and Technology Innovation Action Plan" Natural Science und Project (24ZR1429400), Shanghai Municipal Health Commission Medical New Technology research and Transformation Seed Program Project Plan Assignment (2024ZZ2056), Shanghai Science and Technology program (23010502600), National Key Research and Development Program of China (2020YFA0909000) and Shanghai Key Laboratory of Molecular Imaging (18DZ2260400).

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

All experiments were approved by Ethics Committee in Shanghai University of Medicine and Health Sciences of Shanghai, China (Certificate no.: 2023-GZR-18–340406198707142817).

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.