Abstract

   Targeting senescence has emerged as a promising strategy for liver
   cancer treatment. However, the lack of a safe agent capable of inducing
   complete senescence and being combined with senolytics poses a
   limitation. Here, we screened a natural product library and identified
   tryptanthrin (TRYP) as a potent inducer of cellular senescence in liver
   cancer cells both in vitro and in vivo. Mechanistically, Glutathione
   S-transferase P1 (GSTP1), a key regulator for redox homeostasis, was
   identified as a target protein for TRYP-induced senescence. TRYP
   directly bound to GSTP1 and inhibited its enzymatic activity, mediating
   reactive oxygen species (ROS) accumulation, followed by DNA damage
   response (DDR), consequently contributing to initiating primary
   senescence. Furthermore, TRYP triggered DNA damage-dependent activation
   of NF-κB pathway, which evoked senescence-associated secretory
   phenotype (SASP), thereby leading to senescence reinforcement.
   Importantly, TRYP exposed the vulnerability of tumor cells and
   sensitized senescent cells to apoptosis induced by senolytic agent
   ABT263, a Bcl2 inhibitor. Taken together, our findings reveal that TRYP
   induces cellular senescence via GSTP1/ROS/DDR/NF-κB/SASP axis,
   providing a novel potential application in synergizing with senolytic
   therapy in liver cancer.

   Keywords: Tryptanthrin, Cellular senescence, Liver cancer, GSTP1, SASP,
   ABT263

Graphical abstract

   Image 1
   [41]Open in a new tab

Highlights

     * •
       TRYP, as a novel pro-senescence agent, induces obvious cellular
       senescence in vitro and in vivo with a high safety profile.
     * •
       GSTP1 is a direct pharmacological target of TRYP triggering
       oxidative stress and cellular senescence.
     * •
       TRYP-induced senescent cells establish a pro-inflammatory
       environment to spread and reinforce the senescence.
     * •
       TRYP-induced senescence exposes the vulnerability of tumor cells
       and enhances their sensitivity to senolytic therapy.

1. Introduction

   Liver cancer ranked as the sixth most common cancer worldwide [[42]1].
   Due to its insidious onset, less than 30 % of liver cancer patients are
   suitable for radical treatment at first diagnosis [[43]2]. Targeted
   therapies are currently achieving significant success in a broad
   spectrum of tumors. Unfortunately, liver cancer is among the solid
   tumors with the fewest somatic mutations and lacks effective targeted
   therapeutic agents [[44]3]. Strikingly, the strategy of inducing and
   exploiting the vulnerability of tumor cells by targeting senescence
   holds great promise in liver cancer treatment [[45]4,[46]5]. Through
   the induction of cellular senescence, the senescent tumor cells expose
   the vulnerability to be subsequently targeted for elimination. This
   approach no longer relies on a specific gene mutation, thereby
   substantially expanding drug applicability.

   Cellular senescence serves as a state of irreversible growth arrest
   that plays a pivotal role in tumor suppression [[47]6,[48]7]. Various
   cellular stressors, such as genomic damage, epigenomic perturbations,
   or oxidative stress, can initiate the process of senescence. Senescent
   cells exhibit distinct phenotypic changes, including flattened and
   irregular cell morphology in vitro, loss of Lamin B1 protein, increased
   β-galactosidase activity, heterochromatic alterations, heightened
   metabolic activity, and the activation of the senescence-associated
   secretory phenotype (SASP) [[49][8], [50][9], [51][10]]. The SASP
   encompasses a group of secreted proteins, including pro-inflammatory
   cytokines, growth factors, and matrix metalloproteases [[52]11].
   Although many chemotherapeutic agents have the potential to induce
   senescence, apoptosis remains the predominant phenotype [[53]12].
   Inducing complete senescence in tumor cells is a crucial prerequisite
   for utilizing cellular senescence as a therapeutic target.
   Consequently, an increasing number of studies are focusing on screening
   senescence-inducing drugs [[54]5,[55]13,[56]14]. Natural products,
   characterized by their structural diversity and potential in drug
   development, represent a valuable resource for identifying effective
   senescence inducers. Therefore, our study aims to screen effective
   senescence inducers from natural products.

   It is widely recognized that various natural phytochemicals possess
   anti-cancer properties. Indigo naturalis, a traditional Chinese
   medicine with a history of over 1400 years in treating acute infectious
   diseases [[57]15], has been shown effective in treating hematologic
   tumors based on numerous clinical trials [[58]16]. One of the active
   ingredients identified in Indigo naturalis is tryptanthrin (TRYP)
   [[59]17]. Studies have emphasized the therapeutic potential and
   superior safety profile of TRYP, as it belongs to the class of
   alkaloids [[60]18]. TRYP and its derivatives have demonstrated many
   biological effects, particularly anti-tumor activity in vitro
   [[61][19], [62][20], [63][21], [64][22]]. TRYP is considered a highly
   promising small molecule compound derived from plants with anti-tumor
   properties [[65]23,[66]24]. However, it remains unknown whether TRYP
   can inhibit liver cancer in vivo and induce cellular senescence,
   despite its primary distribution in the liver [[67]25].

   In the present study, TRYP was identified as a potent senescence
   inducer through screening a library of natural products and
   significantly impeded tumor progression in liver cancer.
   Mechanistically, TRYP targeted and inhibited the enzymatic activity of
   GSTP1, leading to excessive accumulation of reactive oxygen species
   (ROS), which in turn triggered DNA damage and ultimately induced
   cellular senescence. Additionally, the DNA damage response (DDR) upon
   TRYP treatment activated NF-κB pathway, leading to SASP secretion and
   senescence reinforcement. Importantly, we revealed that TRYP was an
   excellent pro-senescence agent and proposed the synergistic combination
   of TRYP with senolytic therapy as an innovative and promising approach
   for liver cancer therapy.

2. Materials and methods

2.1. Cell culture

   Huh7 and HepG2 cell lines were obtained from National Collection of
   Authenticated Cell Cultures (Shanghai, China), Huh6 was purchased from
   Procell Life Science&Technology Co.,Ltd (Hubei, China), Hep3B and
   HCCLM3 were obtained from FuHeng Cell Center (Shanghai, China). All
   cell lines were cultured in DMEM or MEM (BasalMedia, Shanghai, China)
   supplemented with 10 % FBS (Excell Bio, China) and 100 μg/mL
   penicillin-streptomycin (BasalMedia, Shanghai, China). The cells were
   maintained at 37 °C in a humidified atmosphere with 5 % CO[2].

2.2. Antibodies and reagents

   The antibodies utilized in this study were as follows: p53 (Santa Cruz,
   Cat# sc-126), H2AX (Santa Cruz, Cat# sc-517336), γ-H2AX (CST, Cat#
   9718), PARP (CST, Cat# 9542), c-PARP (CST, Cat# 5625), p21 (CST, Cat#
   2947), p27 (CST, Cat# 2552), p-ATM (CST, Cat# 1981), Lamin B1
   (Proteintech, Cat# 112987-1-AP), p16 (Proteintech, Cat# 10883-1-AP),
   GSTP1 (Abclonal, Cat# A5691), β-actin (HuaBio, Cat# 68190504).

   The small molecule compounds used in this study: TRYP (PubChem CID:
   73549, 99.94 % purity, Selleck, Cat# S5686), danirixin (DA) (Selleck,
   Cat# S6620), reparixin (RE) (Selleck, Cat# S8640), KU-55933 (Selleck,
   Cat# S1092), BAY 11–7082 (Selleck, Cat# S2913), Etoposide (Selleck,
   Cat# S1225), N-Acetylcysteine (NAC) (MedChem Express, Cat# HY-B0215),
   TLK199 (MedChem Express, Cat# HY-13634A), ABT263 (Selleck, Cat# S1001).
   For in vitro studies, a stock solution of TRYP (10 mM) was fully
   dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, Germany) and
   stored at −80 °C as small aliquots before use. For in vivo studies,
   TRYP was freshly prepared and dissolved in saline containing 10 % DMSO
   with ultrasonic.

2.3. Tumor xenograft model

   Male BALB/c nude mice (5 weeks of age) were purchased from SLAC ANIMAL
   (Shanghai, China) and maintained and treated by established guidelines.
   The experimental protocol was reviewed and approved by the
   Institutional Animal Care and Use Committee of Shanghai University of
   Traditional Chinese Medicine (PZSHUTCM2311130005). The maximum
   allowable tumor volume was set at 2000 mm^3. Subcutaneous injection of
   2 × 10^6 Huh7 cells was performed on the flank of each mouse. Mice were
   randomized into two groups: a control group receiving daily
   intraperitoneal injections of vehicle, and a treatment group
   administered 100 mg/kg of TRYP (The dosage was chosen based on the
   pharmacokinetic profile of TRYP [[68]26] and the effective dosages
   utilized in a previous study [[69]27]). Every other day, the mice were
   weighed, and the length and width of the tumors were measured. Tumor
   volumes were calculated with the formula V = L × S^2/2 (where L is the
   longest diameter and S is the shortest diameter) by a digital caliper.
   At the end of the experiment, mice were euthanized, and tumors were
   excised, photographed, and weighed. To evaluate liver and kidney
   function, the livers and kidneys of the mice were fixed in 4 %
   paraformaldehyde and subjected to H&E staining. Additionally, serum
   samples were collected to assess liver and kidney function.

   For the in vivo combination experiment of TRYP and ABT263, mice were
   randomized into 4 groups: Control (Ctrl), TRYP (100 mg/kg, i.p., q.d.),
   ABT263 (25 mg/kg, i.g., q.3d.), and TRYP + ABT263 (TRYP combined with
   ABT263). Other operations are consistent with the above.

2.4. SA-β-gal staining

   Senescence-associated β-galactosidase (SA-β-gal) staining was conducted
   according to the manufacturer's instructions (C0602, Beyotime,
   Shanghai, China). Briefly, cells were seeded in 12-well plates and
   treated with TRYP for 72 h. Afterward, the cultured cells were fixed by
   β-galactosidase staining fixative at room temperature for 20 min,
   followed by incubation in the staining working solution for
   approximately 12–24 h at 37 °C. The stained cells were then observed
   and photographed under a light microscope. The quantification of
   SA-β-gal staining-positive cells was based on the analysis of five
   random images. For mouse subcutaneous tumor tissues, frozen sections
   were prepared, fixed, and stained following the same procedure as
   mentioned above.

2.5. Immunofluorescence staining

   Cells were seeded in 35 mm confocal dishes at a density of
   2 × 10^4 cells/dish. After 72 h of TRYP treatment, the cells were fixed
   and permeabilized with 0.1 % Triton X-100. Next, cells were blocked
   with 2 % BSA for 1 h and incubated with the γ-H2AX antibody overnight
   at 4 °C. The following day, the secondary antibody was incubated in a
   dark room for 1 h. Nuclei were stained blue with DAPI. Images were
   acquired using a fluorescent microscope (Keyence, Japan). The
   percentage of γ-H2AX foci cells was calculated using the ImageJ
   software.

2.6. Measurement of ROS generation

   Cells were collected and stained with CellROX Orange Reagent (Yeasen
   Biotechnology, Shanghai, China). After incubating at 37 °C for 30 min,
   the level of ROS was detected by flowmetry. Data were analyzed by
   FlowJo 10 software.

2.7. RNA sequencing

   Huh7 cells were treated with 1 ‰ DMSO or TRYP (5 μM) for 72 h, and
   total RNA was isolated using Trizol. After transcriptome sequencing
   library construction, the library preparations were sequenced on an
   Illumina sequencing platform and 150bp paired-end reads were generated.
   Reads were aligned to the GRCh38 human reference genome. RNA-seq raw
   count data was filtered with the criteria of the gene counts >10 at
   least in one experiment. The differentially expressed genes were
   analyzed by DESeq2 filtering with the criteria of FC ≥ 2 or ≤0.5, and
   p < 0.05. KEGG pathway enrichment analysis was performed using the R
   package clusterProfiler (version 4.2.2), and the enrichment analysis
   results were visualized in a dot plot by R package enrichplot (version
   1.14.2).

   Gene set enrichment analysis (GSEA) was performed by the R package
   clusterProfiler (version 4.2.2). The “SenMayo” gene set [[70]28] was
   used to assess the enrichment of senescence-associated genes in
   TRYP-treated versus control cells. Gene sets from Reactome pathways
   related to “cell cycle checkpoints” (R-HSA-69620) and “DNA
   double-strand break repair” (R-HSA-5693532) were used to evaluate the
   enrichment of genes associated with cell cycle arrest and DNA damage
   repair in TRYP-treated versus control cells. The
   “HALLMARK_REACTIVE_OXYGEN_SPECIES_PATHWAY” (M5938) gene set was used to
   assess the enrichment of genes up-regulated by ROS in TRYP-treated
   versus control cells. The “HALLMARK_INFLAMMATORY_RESPONSE” (M5932) and
   “HALLMARK_TNFA_SIGNALING_VIA_NFKB” (M5890) gene sets were used to
   evaluate the enrichment of genes associated with inflammatory response
   and NF-κB signaling pathway in TRYP-treated versus control cells.
   Normalized enrichment scores and adjusted P values were shown in the
   figures.

2.8. DARTS assay

   The DARTS assay was performed according to established procedures
   [[71]29]. In summary, Huh7 cells were lysed using M-PER buffer (Pierce,
   78501) containing protease inhibitors, and then incubated with DMSO or
   250 μM TRYP for 1 h on a rotator at 4 °C. Subsequently, the cell lysis
   was added with pronase and incubated at room temperature for 30 min. To
   prepare the DARTS samples for mass spectrometry analysis, protease
   inhibitors (11836153001, Roche) were added to the mixture on ice to
   halt the digestion process.

2.9. Molecular docking simulation

   The crystal structure of GSTP1 protein (PDB ID: [72]7BIA) was obtained
   from Protein Data Bank ([73]https://www.rcsb.org/), while the 3D
   structure of TRYP (CAS: 13220-57-0) was retrieved from the PubChem
   database ([74]https://pubchem.ncbi.nlm.nih.gov/). Both the GSTP1 and
   TRYP structures were preprocessed using AutoDock Tools 1.5.6.
   Subsequently, molecular docking was conducted using AutoDock Vina
   1.2.3. Reasonable structures with lower energies were selected, and the
   resulting visualization views were generated by Pymol 2.5.0. The
   protein hydrophilicity surface of GSTP1 was mapped by ChimeraX
   software, with different colors representing the hydrophilicity of
   amino acids.

2.10. Molecular dynamic simulation

   The complexes obtained from docking were utilized as initial structures
   for molecular dynamics simulations using Gromacs 2023.2. The GAFF small
   molecule force field and Amber14sb_parmbsc1 protein force field were
   employed to characterize small molecules and proteins, respectively.
   The TIP3P water model was used to fill the system with water molecules,
   and Na^+/Cl^− was added to ensure electrical neutrality. The simulation
   time was set to 100 ns at a constant temperature of 298.15 K.
   Trajectories were obtained by Pymol 2.5.0, while the Root Mean Square
   Deviation (RMSD) was calculated using Gromacs.

2.11. Microscale thermophoresis

   The coding sequence of GSTP1 was inserted into the pET-28a vector
   (Novagen). His-GSTP1 was expressed in Escherichia coli BL21 (DE3). To
   detect direct binding, MST was conducted according to the
   manufacturer's instructions (MO-L018, NanoTemper, German). Briefly,
   recombinant His-GSTP1 proteins were labeled with MST fluorescence dye.
   Subsequently, 10 μL TRYP samples at different concentrations were
   obtained by multiplicative dilution, followed by the addition of 10 μL
   labeled proteins, and mixed thoroughly. The fluorescence signal was
   detected by the Monolith NT.115 instrument (NanoTemper Technologies) at
   25 °C. The K[d] values were calculated by fitting a standard binding
   curve to the series of diluted ligands.

2.12. Biolayer interferometry

   The binding affinity between TRYP and GSTP1 protein was assessed using
   the BLI technique. Biotinylated GSTP1 was prepared following the
   manufacturer's instructions (1828 M, Genemore, Jiangsu, China). After
   pre-wetting the SSA biosensors with PBST (containing 0.02 % Tween 20),
   the biotinylated protein was directly immobilized on the biosensors in
   96-well black plates (655209, Greiner, Germany). TRYP was then diluted
   to the appropriate concentration with PBST in a final volume of 200 μL
   per well, while an equal volume of PBST was added to the control wells.
   The process involved three main steps: a 30-s baseline, followed by a
   60-s association, and finally a 60-s dissociation, which was repeated
   cyclically. Data acquisition and analysis were conducted using the
   ForteBio Octet Data Acquisition and Data Analysis software (Port
   Washington, NY, USA).

2.13. In vitro kinetic assay of GSTP1

   A GSTP1 kinetic assay was carried out according to the manufacturer's
   guidelines (Solarbio, Cat# BC0355). The principle underlying the GSTP1
   enzyme activity assay involves the GST-catalyzed conjugation of
   glutathione (GSH) to 1-chloro-2,4-dinitrobenzene (CDNB), which
   generates the product GS-DNB. This product is characterized by a peak
   absorbance at 340 nm. In brief, 2 μg of GSTP1 protein was incubated
   with DMSO or TRYP (3, 10, 30 μM) on ice for 1 h. Following this, the
   reaction buffer was added to a 96-well UV plate (Corning, Cat# 3635),
   and mixed with GSH and 1-chloro-2,4-xylene (CDNB). The enzyme activity
   kinetics assay was conducted by recording the absorbance at 340 nm
   every 10 s for 60 min. Finally, the enzyme activity of GSTP1 was
   calculated according to the manufacturer's instructions.

2.14. Senescence conditional medium-related experiments

   Huh7 cells were seeded in 6 cm dishes at a density of
   1.5 × 10^5 cells/dish and treated with DMSO or 10 μM TRYP for 96 h. The
   medium was then removed, cells were washed three times with PBS, and a
   fresh medium was added. After 24 h, the medium was collected and mixed
   1:1 with fresh medium to obtain a conditioned medium. The conditioned
   medium was added to 12-well plates with 1.5 × 10^4 Huh7 cells per well,
   and incubated for an additional 96 h. Finally, SA-β-gal staining assay
   was performed on the cells. The flowchart was drawn by Figdraw
   ([75]www.figdraw.com).

2.15. Quantitative and statistical analysis

   Data was analyzed using GraphPad Prism software. The data are presented
   as mean ± standard error of the mean. For normally distributed data,
   the differences between two groups were tested for statistical
   significance using independent-sample two-tailed t-tests (unpaired).
   Four levels of significance were used for all tests (*P < 0.05,
   **P < 0.01, ***P < 0.001, ns = no significance).

3. Results

3.1. TRYP induces cellular senescence and inhibits proliferation of liver
cancer cells

   Acute generation of senescent cells facilitates inhibition of
   tumorigenesis, recruitment of immune cells, and activation of
   anti-tumor immunity [[76]30]. In this study, we aimed to identify small
   molecules from natural products that induce senescence in liver cancer
   cells. Through screening a library of 1445 small compounds, TRYP
   ([77]Fig. 1A) was identified to strongly induce the senescence
   phenotype, piquing our interest. TRYP, a compound found in Indigo, is
   characterized by a simple structure and ease of modification. Previous
   studies reported a variety of biological activities associated with
   TRYP and its derivatives, such as anti-bacterial, anti-parasitic and
   anti-tumor [[78]31], indicating its potential as a highly promising
   antitumor agent. The IC[50] values of TRYP in three liver cancer cell
   lines (Huh7, HepG2, and Hep3B) were determined to be 11.060, 8.703, and
   7.273 μM, respectively ([79]Fig. 1B). Notably, TRYP was found to induce
   a senescence phenotype across a broad spectrum of liver cancer cell
   lines, as evidenced by positive staining for senescence-associated
   β-galactosidase (SA-β-gal) ([80]Fig. 1C–D, [81]Figs. S1A–B). Given that
   different senescence-inducing stimuli often lead to the loss of Lamin
   B1, along with frequent upregulation of p16 and p21 [[82][32],
   [83][33], [84][34], [85][35]], we further examined senescence-related
   markers and observed an increase in p21, p27, and p16, as well as a
   decrease in Lamin B1 upon TRYP treatment ([86]Fig. 1E–F, [87]Figs.
   S1C–D). Additionally, upregulation of p21 and p16 at the mRNA level was
   observed ([88]Fig. 1G). DNA damage is a known marker of cellular
   senescence, we indeed found that TRYP treatment induced DNA damage, as
   evidenced by γ-H2AX foci formation and increased γ-H2AX protein
   expression ([89]Fig. 1H–I, Fig. S1E-F). Furthermore, cell cycle
   analysis showed that TRYP induced G2/M-phase cell-cycle arrest
   ([90]Fig. S1G), which may lead to the initiation of cellular
   senescence.

Fig. 1.

   [91]Fig. 1
   [92]Open in a new tab

   TRYP is a novel pro-senescence agent in liver cancer cells

   (A) Identification of TRYP as a potential hit through screening a
   natural product library.

   (B) The half-maximal inhibitory concentrations (IC[50]) of TRYP in
   Huh7, HepG2 and Hep3B were shown.

   (C-D) Huh7 and HepG2 cells were treated with 0, 1.25, 2.5, 5, and 10 μM
   TRYP for 72 h respectively, followed by SA-β-gal staining (scale
   bar = 50 μm) (C). The proportion of staining-positive cells to the
   overall cells was shown in the bar plot (D).

   (E–F) The protein expressions of senescence markers (p21, p16, p27 and
   Lamin B1) in Huh7 cells treated with TRYP were detected by western
   blot, with β-actin as a loading control (E). The corresponding
   statistical analyses of protein expressions were shown (F).

   (G) Cells were treated with different doses of TRYP for 72h, and the
   mRNA expression levels of p21 and p16 were detected by qPCR.

   (H) Huh7 cells were treated with 1 ‰ DMSO or TRYP (5, 10 μM) for 72 h,
   then γ-H2AX foci was determined by immunofluorescence staining (green).
   DAPI was used to stain the nuclei (blue) (scale bar = 20 μm). The ratio
   of γ-H2AX foci to total cells was presented in a bar plot (right
   panel).

   (I) Huh7 cells were treated with 1 ‰ DMSO or TRYP (5 μM, 10 μM) for
   72 h. Protein expression of γ-H2AX and H2AX were detected by western
   blot, with β-actin as a loading control (left panel). The corresponding
   statistical analyses of the western blot bands were shown (right
   panel).

   (J) Huh7 and HepG2 cells were cultured with different concentrations of
   TRYP for 20 days and subjected to colony formation assay (left panel).
   A statistical graph of colonies was shown (right panel). (For
   interpretation of the references to color in this figure legend, the