Tongguanteng injection exerts anti-osteosarcoma effects through the ER stress-associated IRE1/CHOP pathway

Abstract

Background

In China, Tongguanteng injection (TGT) is widely used in the treatment or adjuvant treatment of various types of cancer. However, the effect and mechanism of TGT in osteosarcoma is not clear.

Methods

The 143B and MG-63 cells were treated with different concentrations of TGT. Cell proliferation, migration, invasion and apoptosis were detected using CCK8 assay, transwell assay and flow cytometry. Differentially expressed genes (DEGs) were screened using RNA sequencing (RNA-seq). The identified mRNA and protein expression associated with the IRE1/CHOP pathway was validated by RT-PCR and western blot assay. To explore the underlying mechanisms, 4-phenylbutyric acid (4-PBA) was selected as a specific endoplasmic reticulum (ER) stress inhibitor. Small interfering RNA (siRNA) or pEX‐3-ERN1 plasmid was transfected into 143B cells to silence or overexpress IRE1, respectively. The potential downstream proteins, including CHOP, and apoptosis associated proteins, caspase-3 and PARP1 were determined. Furthermore, the effect of TGT was demonstrated in 143B cell tumor-bearing mice in vivo. H&E staining, TUNEL staining and immunohistochemistry were conducted in tumor tissues obtained from the xenograft mouse model.

Results

TGT was shown to dramatically suppress the proliferation, migration and invasion, and induce apoptosis of osteosarcoma 143B and MG-63 cells in vitro. The identified DEGs included HSPA5 (encoding BiP) and ERN1 (encoding the IRE1 protein), as well as apoptosis-associated gene DDIT3 (encoding the CHOP protein). The term “IRE1-mediated unfolded protein response” was screened to be the most enriched biological process GO term. The expression of ER stress-associated proteins including ATF6, BiP, p-IRE1, XBP1s and CHOP, as well as apoptosis-associated cleaved caspase-3 and cleaved PARP1 proteins, was significantly upregulated by TGT treatment in osteosarcoma 143B cells, suggesting that TGT might promote the apoptosis of osteosarcoma 143B cells through the IRE1/CHOP pathway. Furthermore, knocking down IRE1 with si-IRE1 or inhibiting of ER stress with 4-PBA suppressed the expression of ATF6, BiP, XBP1s and CHOP induced by TGT, as well as the expression of cleaved caspase-3 and cleaved PARP1. On the contrary, overexpressing IRE1 promoted CHOP expression and induced osteosarcoma cell apoptosis. Consistent with in vitro results, TGT dramatically inhibited the tumor growth and promoted the expression of p-IRE1 and CHOP in tumor-bearing mice.

Conclusion

The findings suggest that TGT exerts an anti-osteosarcoma effect in vitro and in vivo. The underlying mechanism might be associated with the activation of IRE1/CHOP pathway in ER stress. Our findings suggest that targeting IRE1/CHOP pathway might be a potential novel approach for osteosarcoma treatment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12906-024-04689-7.

Introduction

Osteosarcoma, the most common malignant bone tumor, generally occurs in children and young adults aged from 10 to 30 years [1]. Surgery and chemotherapy are the main treatments for osteosarcoma, which have improved the overall 5-year survival rate up to 60% among patients with localized osteosarcoma [2]. However, the 5-year survival rate dramatically diminishes to less than 20% in patients with recurrent or metastatic osteosarcoma [1, 3]. In addition, the curative treatments inevitably cause long-term complications, including neurocognitive dysfunction, cardiomyopathy, and liver and kidney toxicity, all of which may influence the prognosis of cancer patients [4]. Although measures have been taken to improve the survival of osteosarcoma patients in the past four decades, treatment of osteosarcoma still faces numerous challenges, including limited effective drugs, lack of specific therapeutic targets, and uncertainty regarding the efficacy of immunotherapy. Accordingly, exploring effective anti-osteosarcoma drugs that possess ideal therapeutic efficacy and low toxicity is urgently needed.

Endoplasmic reticulum (ER) stress is commonly observed in tumor cells as a result of imbalance in homeostasis induced by various endogenous and exogenous factors [5]. To maintain ER homeostasis, three principal signaling pathways have been identified to trigger the unfolded protein response (UPR), including the inositol-requiring enzyme 1α (IRE1α, referred to as IRE1 hereafter), the protein kinase RNA-like ER kinase (PERK) and the activating transcription factor 6 (ATF6)-mediated signaling pathways, as well as signaling components such as the downstream transcription factors X-box binding protein 1 (XBP1), ATF4, and C/EBP homologous protein (CHOP) [6]. However, when UPR is incapable of restoring ER function, cell apoptosis is triggered. ER stress appears to be a double-edged sword in tumor development and progression. Persistent ER stress facilitates cancer cell survival, while promoting ER stress to initiate apoptotic pathways may be an effective antitumor strategy [7, 8]. In recent years, increasing evidence has suggested that drugs or chemicals, such as CYT997 (Lexibulin), targeting ER stress are potential treatments for osteosarcoma [9–11]. The major ER chaperone binding immunoglobulin protein (BiP), also described as 78-kDa glucose-regulated protein (GRP78), is encoded by the heat shock protein family A (Hsp70) member 5 (HSPA5) gene. BiP, a protein frequently overexpressed in several types of cancer, has been demonstrated to be an ER stress detector that directly binds to the UPR activator proteins, IRE1 and PERK, leading to UPR activation [12–14].

IRE1, which is encoded by the ER to nucleus signaling 1 (ERN1) gene, is a critical component of the UPR pathway that has dual protein kinase and RNase activities [15, 16]. Upon the accumulation of misfolded proteins, activated IRE1 (p-IRE1) catalyzes the XBP1 mRNA through unconventional splicing to generate a mature form of XBP1, XBP1s [17–19]. XBP1s binds to the CHOP promoter and upregulates CHOP expression, eventually leading to cell apoptosis [20, 21]. Additionally, IRE1 can activate its downstream apoptotic-signaling kinase 1 (ASK1), which subsequently induces Jun-N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (p38 MAPK), leading to the activation of CHOP [20]. CHOP, which is encoded by DDIT3, is the major regulator of ER stress-induced apoptosis. The latest research showed that curcumin regulates the expression of CHOP through ATF6, another key regulator of the ER stress response [22]. Studies have also suggested that ATF6 fine-tunes the UPR, which is primarily modulated by the predominant IRE1α- and PERK-dependent pathways [23]. There are numerous studies on triggering ER stress through IRE1 or CHOP to promote tumor cell apoptosis [24–27]. Surfactin, a natural component from Bacillus subtilis, stimulates cell apoptosis through the ER stress-associated IRE1-ASK1-JNK pathway in human osteosarcoma cells [9]. Coiled-coil domain containing 170 may activate the IRE1α-XBP1s signaling pathway to promote apoptosis in MCF7 breast cancer cells [28]. 2-bromopalmitate is proved to modulate CHOP and promote adriamycin-induced osteosarcoma cell apoptosis [29].

Tongguanteng injection (referred to as TGT), derived from the traditional Chinese medicine Marsdenia tenacissima (Roxb.) Wight et Arn., is authorized by the National Medical Products Administration (NMPA) for cancer therapy for over 20 years in China [30, 31]. Until now, nearly 196 chemical components have been discovered from the plant, with C₂₁ steroidal glycosides identified as characteristic compounds that exert antitumor effects through multiple mechanisms [30, 32]. TGT is often used in conjunction with chemotherapy to treat various cancers including liver cancer, gastric cancer, and colon cancer. Our previous studies revealed that TGT exerts a synergistic effect with chemotherapy and reverses drug resistance in ovarian cancer [31, 33]. T. Huang’s research revealed for the first time that TGT could trigger apoptosis in osteosarcoma cells, albeit being less effective than doxorubicin [34]. In our recently published study, network pharmacology and in vitro experiments confirmed that TGT induces dose-dependent apoptosis in osteosarcoma cells, and 6 kinds of C₂₁ steroidal glycosides were identified as predominant bioactive constituents [35]. However, the anti-osteosarcoma effect of TGT in vivo remains unclear, and the underlying mechanism needs to be further studied.

In the current research, osteosarcoma cells and xenograft nude mouse model were constructed to explore the antitumor activity of TGT. Additionally, to elucidate the underlying mechanism, the ER stress-associated IRE1/CHOP pathway was investigated by RNA-seq assay. Further results demonstrated that TGT regulated the IRE1/CHOP pathway to trigger osteosarcoma cell apoptosis. Our findings indicate that TGT is a potential treatment for osteosarcoma, however, further clinical validation through randomized clinical trials is necessary.

Materials and methods

Drugs and reagents

TGT (No. 202108031) was acquired from Nanjing Sanhome Pharmaceutical Co., Ltd. (1 mL of the mixture contains 1 g of dried raw herb powder). The stem of M. tenacissima was obtained from Yunnan, China. A voucher exemplar (200907-T009-05) was stored in the herbarium of Sanhome Pharmaceutical Co. Ltd and was authenticated by Professor De-Kang Wu, from Nanjing University of Chinese Medicine. 4-PBA (No. HY-A0281) was purchased from MedChemExpress (MCE, New Jersey, USA). Cisplatin (No.1M0696B03) was obtained from Qilu Pharmaceutical Co., Ltd (Jinan, China). ATF6 (No. A0202), p-IRE1 (No. AP0878), IRE1 (No. A17940), PARP1 (No. A0942), BCL-2 (No. A19693), BAX (No. A19684), β-Actin (No. AF0003) and β-Tubulin (No. A12289) antibodies were obtained from ABclonal Technology Co., Ltd. (Wuhan, China). BiP (No. 3177S) and cleaved caspase-3 (No. 9662 S) antibodies were purchased from Cell Signaling Technology, Inc. (Boston, USA). CHOP antibody (No. 15204-1-AP) was acquired from Proteintech Group, Inc. (Wuhan, China). The XBP1 antibody (No. ab220783) was obtained from Abcam Plc. (Cambridge, UK).

Cells and culture

Human osteosarcoma 143B and MG-63 cells were acquired from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Human bone marrow mesenchymal stem cells (BMSCs) were acquired from Procell Life Science & Technology Co., Ltd. (Wuhan, China). All the cells were incubated in DMEM (HyClone, USA) containing 1% penicillin/streptomycin (NCM Biotech., China) and 10% FBS (Thermo Fisher Scientific, USA) at 37 ℃ with 5% CO₂. The cells were subcultured every two days at a ratio of 1:3.

Cell counting kit-8 (CCK-8) assay

Cells were seeded separately into 96-well plates (8 × 10³ cells/well) and incubated overnight until completely adherent to the walls. Then the cells were treated with increasing concentrations of TGT (20, 40, 60, 80, 100, 120, or 140 mg · mL^− 1). The cells that did not receive TGT treatment were set as the control group. 10 µL of CCK-8 reagent (Beyotime Biotech., China) was added to each well after incubating with TGT for 24 h. The absorbance value was determined via a microplate reader (BioTek, Winooski, USA) at 450 nm after incubation at 37 ℃ for 1.5 h.

Plate colony formation

Osteosarcoma cells were seeded into 6-well plates (5 × 10³ cells/well) and incubated overnight until completely adherent to the wall. Afterwards, the osteosarcoma cells were exposed to diverse concentrations of TGT (40, 60, or 80 mg · mL^-1). The cells that did not receive TGT treatment were set as the control group. After incubating for 24 h, the medium was superseded by fresh complete medium until each colony contained more than fifty cells. Then cells were fixed for 15 min with 4% paraformaldehyde, and dyed for 30 min with 0.1% crystal violet. The number of colonies were calculated via an ImageJ software. The colony formation ratio was displayed as (number of clones/number of cells inoculated) × 100%.

Cell migration and invasion assay

The lateral migration capacity of the cells was assessed by the wound healing assay. Osteosarcoma cells (3 × 10⁵ cells/well) were seeded in 6-well plates and cultured until reaching 90% confluence. Cell scratches were produced perpendicularly with a sterilized 10 µL pipette tip. The wound areas were recorded using an inverted microscope (Leica DMi8, Wetzlar, Germany) at 0 and 24 h after 40, 60 and 80 mg · mL^-1 of TGT treatment. The cells that did not receive TGT treatment were set as the control group. The images were recorded via ImageJ software, and the scratch healing rate was determined.

Transwell assay was carried out to evaluate the longitudinal migration and invasion ability of osteosarcoma cells [36]. The transwell chambers with 8.0 μm pore (Corning Costar, USA) were placed into 24-well plates with or without 60 µL of Matrigel solution (3 mg · mL^-1) added to the upper chamber for invasion and migration assay, respectively. 200 µL of serum-free cell suspensions (1 × 10⁵ cells/mL) containing different concentrations of TGT were added to the upper chamber, and 600 µL of complete medium containing 10% FBS was added to the lower chamber. After 24 h of incubation, the chambers were fixed with a 4% paraformaldehyde solution for 15 min, and dyed with 0.1% crystal violet for 30 min. Then the number of cells invading the chamber membrane was calculated via ImageJ software. This experiment was conducted three times.

Hoechst 33342 staining

Osteosarcoma cells (3 × 10⁵ cells/well) were cultured in 6-well plates to approximately 80% confluence, and treated with TGT (40, or 80 mg·mL^-1) for 24 h. The cells that did not receive TGT treatment were set as the control group. Then, 20 µL of Hoechst 33342 staining reagent was added and incubated for 15 min. Afterwards, the staining reagent was removed. The cell image was recorded via a fluorescence microscope.

Flow cytometry assay

After being seeded into 6-well plates (3 × 10⁵ cells/well) and growing to approximately 80% confluence, osteosarcoma cells were treated with diverse concentrations of TGT (40 or 80 mg·mL^-1) for 24 h. The cells that did not receive TGT treatment were set as the control group. Subsequently, both the floating cells in the supernatant, and the attached cells digested by trypsin digestion were harvested. Apoptotic cells were distinguished and quantified applying the FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, USA) following the producer’s protocol. Afterwards, the cell apoptotic rate was calculated via a flow cytometry (BD Biosciences, CA, USA).

RNA sequencing (RNA-seq)

Osteosarcoma 143B cells were exposed to TGT (40 or 80 mg·mL^-1) for 24 h. For both the treatment and control groups (without TGT treatment), three biological replicates were carried out. Utilizing an EZ-press RNA Purification Kit (EZBioscience, Roseville, USA), total RNA was extracted following the producer’s protocol. A microplate reader was also utilized to determine the purity and quantity of the extracted RNA. A VAHTS Universal V5 RNA-seq Library Prep Kit (Vazyme Biotech., China) was utilized to establish transcriptome libraries: the extracted mRNA was fragmented, reverse transcribed into cDNA, adenylated at the 3’ Ends, ligated to adaptors and selectively enriched. An Agilent 2100 Bioanalyzer (CA, USA) was used to determine library quality and fragment size. OE Biotech Co., Ltd. (Shanghai, China) carried out the RNA-seq and analysis. Using an Illumina NovaSeq 6000 for library sequencing, 150 bp double-ended reads were produced, resulting in nearly 50 raw reads per sample. The cleaned reads were mapped to the reference genome using the HISAT2 software and quantification of gene expression levels was performed based on FPKM values. HTSeq-count was used to obtain the read counts for each gene. R (version 3.2.0) was used for PCA analysis to evaluate biological replicates of the samples. DESeq2 software was utilized to normalize the number of genes in each sample, calculate fold changes, and test for significant differences based on a negative binomial distribution. For DEGs, a criterion of q-value < 0.05 and |log₂ (fold change)| > 1 (|log₂ FC| > 1) was established. DEGs were subjected to hierarchical cluster analysis using R (version 3.2.0).

Short time-series expression miner (STEM) software was applied to screen for genes whose expression was affected by TGT in a dose-dependent manner. Based on the hypergeometric distribution, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of the TGT-affected genes were carried out.

Real-time PCR (quantitative PCR, q-PCR)

Utilizing the EZBioscience Color Reverse Transcription Kit (EZBioscience, Roseville, USA), total RNA was reverse transcribed on a PCR instrument (for 15 min at 42 ℃ and heat inactivation for 30 s at 95 ℃). The q-PCR reaction mixture was prepared in a 384-well plate according to the instructions for 2× Color SYBR Green qPCR Master Mix (ROX2 plus) (EZBioscience, Roseville, USA). q-PCR was conducted with the following amplification program: 95 °C for 5 min, 10 s at 95 °C and 30 s at 60 °C for 40 cycles. GAPDH was used as a normalization reference. Sangon Biotech (Shanghai, China) synthesized all of the primers, which are listed in Table 1. The average Ct value was obtained, and the relative expression was determined with the 2^-ΔΔCt method. The q-PCR assays were conducted for three biological and three technical replications.

Table 1.

Primers for randomly selected DEGs used for q-PCR analysis

Gene	Sequence (5’ to 3’)
MMP3	Forward	AGTCTTCCAATCCTACTGTTGCT
	Reverse	TCCCCGTCACCTCCAATCC
RHOJ	Forward	CTGATGAGCTACGCCAACGA
	Reverse	GTCACAGTAACTGCATAGTGGTC
KLHL4	Forward	GTGAGAAACGCGCACAAGATT
	Reverse	AACCAACCTATGGGCTGGGAT
KCNK10	Forward	GTGGTGGTCTACCTTGTCACT
	Reverse	GGGCTCACACAGACATGATCC
HSPA5	Forward	CATCACGCCGTCCTATGTCG
	Reverse	CGTCAAAGACCGTGTTCTCG
TIMP3	Forward	CATGTGCAGTACATCCATACGG
	Reverse	CATCATAGACGCGACCTGTCA
ERN1	Forward	CACAGTGACGCTTCCTGAAAC
	Reverse	GCCATCATTAGGATCTGGGAGA
DDIT3	Forward	GGAAACAGAGTGGTCATTCCC
	Reverse	CTGCTTGAGCCGTTCATTCTC

Transfection of IRE1 siRNAs and IRE1 overexpression (IRE1-OE) plasmid

Table 2 displays the sequences of human IRE1 siRNAs that were produced by GenePharma Co., Ltd. (Shanghai, China). Human ERN1 (GenBank Accession: NM_001433.5) was cloned and inserted into the pEX-3 (pGCMV/MCS/EGFP/NEO) vector to construct the IRE1-OE plasmid pEX‐3-ERN1. The 143B cells were seeded into 6-well plates (3 × 10⁵ cells/well) containing reduced serum medium (Gibco, USA) without antibiotics, and transfected with reaction mixture following the protocol of Lipofectamine™ 3000 reagent (Thermo Fisher Scientific, USA). The siRNA transfection complex contained 5 µL of Lipofectamine™ 3000 Reagent and 100 nM human IRE1 siRNA. The cells without transfection were set as the blank control. The cells transfected with siNC were set as the negative control group. The IRE1-OE plasmid transfection complex contained 5 µL of Lipofectamine™ 3000, 5 µL of P3000™ Reagent, and 2.5 µg of IRE1-OE plasmid. The cells transfected with empty plasmid were set as the control group. After 6 h of transfection, the media was replaced with fresh DMEM complete media (HyClone, USA) and cultured for 48 h. Then the proteins were extracted and the transfection efficiency was calculated via a western blot experiment. However, cells transfected with IRE1-OE plasmid were screened with 200 µg/mL G-418 disulfate (MCE, USA), and stably transfected cells were utilized for western blot experiment.

Table 2.

Sequences of human IRE1 siRNAs

siRNA	Sequence (5’ to 3’)
siIRE1-1	sense	CAGACAGACCUGCGUAAAUUC
	antisense	GAAUUUACGCAGGUCUGUCUG
siIRE1-2	sense	AUGGAGCUGAGGGCACAAUUG
	antisense	CAAUUGUGCCCUCAGCUCCAU
siIRE1-3	sense	GCCUGACGAAACUUCCUUUUA
	antisense	UAAAAGGAAGUUUCGUCAGGC

Western blot assays

Supplemented with protease and phosphatase inhibitor, the RIPA lysis buffer cocktail (Beyotime Biotech., China) was used to lyse 143B cells on ice for 30 min (100 µL per well, 6-well plates). Cell lysates were scraped into EP tubes and centrifuged at 12,000 rpm for 20 min at 4 ℃. Then the total protein was extracted, and protein concentration was detected by a BCA protein quantification kit (Beyotime Biotech., China). Protein concentrations in all samples were adjusted to 1 µg/mL with a loading dose of 15 µg. Loading buffer (ABclonal Biotech., China) was added to the protein samples, which were subsequently denatured by boiling at 95 ℃ for 10 min in a metal bath. Proteins were transferred to PVDF membranes after being separated by SDS-PAGE. The membranes were blocked for 15 min with rapid blocking reagent (NCM Biotech., China), and then they were incubated overnight at 4 °C with diluted primary antibodies at a dilution of 1: 1000. The primary antibodies included IRE1, p-IRE1, ATF6, PARP1 and β-Tubulin purchased from ABclonal Technology Co., Ltd. (Wuhan, China), BiP and cleaved caspase-3 obtained from Cell Signaling Technology, Inc. (Boston, USA), CHOP antibody acquired from Proteintech Group, Inc. (Wuhan, China), and XBP1 antibody obtained from Abcam Plc. (Cambridge, UK). β-Tubulin was chosen as an internal control. Then, the membranes were incubated with diluted HRP-labeled secondary antibodies (1: 5000) for 1 h at room temperature. The luminescence process was assessed by ECL luminescence (NCM Biotech., China). Grayscale analysis was conducted using ImageJ software.

Construction of a xenograft tumor model

Shanghai Laboratory Animal Center (SLAC, Shanghai, China) provided male BALB/c nude mice that aged 5 to 6 weeks. The mice were kept under monitored conditions with a 12/12 h light/dark cycle, a temperature range of 20 ~ 26 °C and a humidity range of 40 ~ 70% in a pathogen-free environment. The mice were given water and food ad libitum. To maintain a hygienic environment, bedding and cages were replaced every three days. The animal researches were accredited by the Animal Welfare Ethics Committee of Shanghai Sixth People’s Hospital (No. SYXK 2018-0028). All animal experiments were conducted rigorously in compliance with Provision and General Recommendation of Chinese Experimental Animals Administration Legislation. The 143B cells were resuspended and washed thrice with PBS. Then, the cells (1 × 10⁶ cells/100 µL) were subcutaneously inoculated into the armpits of the nude mice. The mice were segmented into 5 groups at random when the tumor volume reached 100 ~ 300 mm³: the control group (normal saline), cisplatin group (3 mg·kg^-1), low-dose TGT group (10 g·kg^-1), medium-dose TGT group (20 g·kg^-1), high-dose TGT group (40 g·kg^-1), and 8 mice in each group. Cisplatin was administered once every 3 days via intraperitoneal injection, while the other groups were injected once daily. The tumor size was measured every day with a vernier caliper (tumor volume = length × width²/2). After 14 d of continuous administration, the mice received 3% halothane to induce anesthesia. Then the euthanasia was conducted by cervical dislocation. The tumors were dissected, photographed and weighed. Finally, all tumor tissues were stored in liquid nitrogen or fixed in 4% paraformaldehyde.

H&E staining, TUNEL staining and immunohistochemistry (IHC)

H&E staining was performed to obtain histological information. The tumor tissue was immersed in paraffin, sectioned into serial 4-µm-thick slices, and placed on coverslips on microscope slides. After dewaxing in xylene and rehydrating in alcohol baths, paraffin sections of the liver, heart, kidneys and lungs were stained successively with hematoxylin and eosin. The sections were dehydrated in gradient alcohol, cleaned in xylene and covered with coverslips. Images were recorded via a microscope (Olympus Corporation, Tokyo, Japan).

TUNEL staining was conducted to evaluate nuclear fragmentation. After dewaxing in xylene, the paraffin sections were rehydrated with gradient alcohol. Endogenous catalase was inactivated in 3% H₂O₂. Sections from different groups of mice were stained via a In Situ Cell Death Detection Kit, POD (Roche, Penzberg, Germany) following the producer’s instructions. DAB staining solution (Beyotime Biotech., China) was used to enhance the staining intensity.

IHC analysis was implemented to assess the expressions of p-IRE1, CHOP and cleaved caspase-3. Immersed in sodium citrate buffer, the tissue sections were boiled for 20 min. Then the sections were blocked in 5% BSA for 30 min. Cleaved caspase-3, p-IRE1, and CHOP antibodies were incubated with the sections overnight at 4 ℃. After rinsing three times with PBS, secondary antibodies were added, and the mixture was incubated for 1 h. Finally, DAB staining solution (Beyotime Biotech., China) was used for color development.

Statistical analysis

Graphpad Prism 8.0 software was used to compute the original data. The raw data were represented by the mean ± standard deviation (SD). For multiple group comparisons, one-way analysis of variance (ANOVA) was performed to test for statistically significant differences, followed by the Bonferroni correction. Statistical significance was considered at a value of p < 0.05.

Results

TGT inhibits the proliferation of osteosarcoma cells

To explore the effect of TGT on the growth of osteosarcoma cells, osteosarcoma cells (143B and MG-63) were exposed to diverse concentrations of TGT for 24 h. CCK-8 experiments were carried out. The results showed that the osteosarcoma cell viability declined sharply in a dose-dependent manner. For 143B and MG-63 cells, the IC₅₀ values were 92.53 and 87.44 mg·mL^− 1, respectively. While, in human BMSCs, which served as normal control cells, TGT at concentrations ranging from 20 to 140 mg·mL^− 1 did not significantly suppress the proliferation of BMSCs, indicating no toxicity in normal cells (Fig. 1A). Moreover, in the plate colony formation assay, the number of cell colonies formed in the TGT groups were remarkably reduced (Fig. 1B), indicating that TGT could significantly inhibit the growth of osteosarcoma cells.

Fig. 1.

TGT attenuates the growth of osteosarcoma cells. A 143B, MG-63 and BMSC cells were treated with TGT at different concentrations for 24 h. The cell viability was evaluated and the IC₅₀ value was calculated. B After treatment with TGT (40, 60, or 80 mg·mL^− 1) for 24 h, the colony formation ability of the 143B and MG-63 cells was detected. Data are expressed as the mean ± SD (n = 3). **p < 0.01, vs. the control group

TGT inhibits the migration and invasion of osteosarcoma cells

The wound healing assay revealed that the lateral migration capacity of the TGT treatment groups was markedly reduced compared to the control group (Fig. 2A). Furthermore, in the transwell assay, the number of migrating and invading cells in the TGT treatment groups was also significantly decreased (Fig. 2B and C). These results indicated that TGT could effectively inhibit the migration and invasion of osteosarcoma cells.

Fig. 2.

TGT inhibits the migration and invasion of osteosarcoma cells. A TGT suppressed the lateral migration capacity of 143B and MG-63 cells in the scratch test. Scale bar = 100 μm. B TGT inhibited the longitudinal migration of osteosarcoma cells according to the transwell assay. Scale bar = 50 μm. C TGT suppressed the invasion ability of osteosarcoma cells according to the transwell assay. Scale bar = 50 μm. Data are presented as the mean ± SD (n = 3). **p < 0.01, vs. the control group

TGT induces osteosarcoma cell apoptosis

To explore the impact of TGT on the apoptosis of osteosarcoma cells (143B and MG-63), Hoechst staining was carried out to detect the apoptosis in osteosarcoma cells treated with TGT for 24 h. As was shown, compared with the control group, the TGT intervention group exhibited apoptotic morphology characterized by a reduced cell number and densely stained nuclei (Fig. 3A). A higher TGT concentration was associated with more obvious morphological changes in apoptotic cells. Consistent with results of Hoechst staining, flow cytometry analysis indicated that TGT treatment promoted osteosarcoma cell apoptosis in a dose-dependent manner (Fig. 3B). The increased expression of BAX protein and decreased expression of BCL-2 protein in TGT treatment group further validated the results of the phenotypic data (Fig. 3C).

Fig. 3.

TGT promotes apoptosis in osteosarcoma cells. A TGT induced the apoptosis of 143B and MG-63 cells, as determined by a Hoechst staining assay. Scale bar = 25 μm. B TGT increased the apoptosis rate of osteosarcoma cells according to the flow cytometry assay. C The expression of BAX protein was enhanced and BCL-2 protein was reduced by TGT treatment, as detected by western blot experiment. Data are expressed as the mean ± SD (n = 3). *p < 0.05, **p < 0.01, vs. the control group

RNA-seq of 143B cells treated with TGT

In present research, RNA-seq was conducted on osteosarcoma 143B cells to screen the genes affected by TGT that were involved in the process of apoptosis. Two concentrations of TGT (40 and 80 mg·mL^− 1) were set up to identify DEGs by pairwise comparisons with the untreated group. A threshold of q-value < 0.05 and |log₂ FC| > 1 was set for screening DEGs. Compared with the control group, a total of 63 and 328 DEGs were identified (Supplementary Table S1-2), respectively. As illustrated in the volcano plot (Fig. 4A), the fold change in expression of the apoptosis-associated gene DDIT3 was highly ranked in both groups. Highly upregulated expression of ERN1 and HSPA5 was detected in the 80 mg·mL^− 1 TGT group (Fig. 4A). To validate the DEGs screened from the RNA-seq data, 8 randomly selected DEGs (4 upregulated and 4 downregulated) were selected for q-PCR assay. The relative expression was calculated with the 2^−ΔΔCt method. One-way ANOVA and Bonferroni correction for comparisons test were utilized, and the result were found to be consistent with the RNA-seq data except for KLHL4 (Fig. 4B). In line with the findings exhibited in Fig. 4A, ER stress-associated genes ERN1 and HSPA5 were detected to be highly expressed (Fig. 4B). A total of 16 expression trends were identified using STEM software (Supplementary Fig. S1). The gene expression trend in profile 13, 12 and 15 was positively correlated with ER stress response pathway. GO enrichment analysis of genes classified as profile 15 revealed that “IRE1-mediated UPR” was the most enriched biological process, and “Protein processing in ER” was identified as the most enriched KEGG pathway (Fig. 4C).

Fig. 4.

RNA-seq analysis of 143B cells treated with TGT for 24 h. A Volcano plot of DEGs with q-value < 0.05 and |log₂ FC|>1 as threshold in the TGT 40 and 80 mg·mL^− 1 group. B Heatmap and q-PCR validation of the ER stress-associated genes and randomly selected DEGs. C GO and KEGG analysis of genes categorized into profile 15. Data are denoted as the mean ± SD (n = 3). *p < 0.05, **p < 0.01, vs. the control group

TGT activates the IRE1/CHOP pathway and mediates cell apoptosis

To investigate the relationship between ER stress and TGT-induced apoptosis in osteosarcoma cells, the protein expression of ATF6, BiP, IRE1, XBP1 and CHOP was measured by western blot experiment. After treatment with TGT, the expression of the ATF6, BiP, p-IRE1, XBP1s, and CHOP was significantly upregulated in osteosarcoma 143B cells (Fig. 5A), suggesting that the ER stress was triggered. Meanwhile, TGT dramatically upregulated the expression levels of cleaved caspase-3 and cleaved PARP1 in osteosarcoma 143B cells (Fig. 5B). These results indicated that TGT might promote the phosphorylation of IRE1, thereby activating the IRE1/CHOP pathway and inducing ER stress, leading to apoptosis of osteosarcoma 143B cells.

Fig. 5.

TGT triggers the ER stress response pathway and activates apoptotic proteins in 143B cells. A After TGT treatment, the expression levels of ATF6, BiP, p-IRE1, XBP1s and CHOP were enhanced, as shown by western blot experiment. B The expression of cleaved caspase-3 and cleaved PARP1 was enhanced by TGT treatment, as detected by western blot experiment. Data are displayed as the mean ± SD (n = 3). *p < 0.05, **p < 0.01, vs. the control group

IRE1 knockdown attenuates the ER stress response and cell apoptosis

To explore the role of IRE1 in TGT-induced ER stress, siIRE1s were designed and successfully transfected. siIRE1-3, which had the highest siRNA knockdown efficiency, was selected for transfection into osteosarcoma cells to interfere the ER stress response pathway (Fig. 6A). As illustrated in Fig. 6B, compared to the TGT treatment group, the protein expression levels of ATF6, BiP, p-IRE1, XBP1s and CHOP decreased after IRE1 knockdown. Furthermore, the expression levels of cleaved caspase-3 and cleaved PARP1 were significantly reduced, as well as the cell apoptosis rate (Fig. 6C and D). The most well-known ER stress inhibitors are 4-PBA and tauroursodeoxycholic acid (TUDCA), which are widely used in the study of a variety of diseases, including cancer, and are proved to display outstanding safety profiles in vivo [37–39]. Previous studies have shown that 4-PBA inhibits ER stress in osteosarcoma, while TUDCA has not been reported to date [40, 41]. Treatment with 4-PBA had slightly stronger effect on the expression of apoptosis-related proteins and cell apoptosis rate than that of IRE1 knockdown (Fig. 6C and D). Therefore, it was reasonable to speculate that TGT could trigger ER stress-induced apoptosis in osteosarcoma 143B cells through the IRE1/CHOP pathway.

Fig. 6.

Knocking down of IRE1 or inhibiting of ER stress suppresses the 143B cell apoptosis. A The knockdown efficiency of three siIRE1s was measured via a western blot experiment. B The expression levels of proteins involved the regulation of ER stress, including ATF6, BiP, p-IRE1, XBP1s and CHOP, were significantly decreased according to western blot analysis. C The expression levels of cleaved caspase-3 and cleaved PARP1 were significantly reduced by knockdown of IRE1, as detected by western blot experiment. D The apoptosis rate was significantly decreased by IRE1 knockdown and 4-PBA treatment, as determined via flow cytometry analysis. Data are displayed as the mean ± SD (n = 3). ^#p < 0.05, ^##p < 0.01, vs. the control group; *p < 0.05, **p < 0.01, vs. the TGT alone group

IRE1 overexpression enhances CHOP expression and apoptosis in 143b cells

To further illustrate the relationship between the IRE1/CHOP pathway and the ER stress or 143B cell apoptosis, the IRE1-OE plasmid was transfected into 143B cells. The results demonstrated that the expression level of CHOP markedly enhanced upon IRE1 overexpression. In addition, cleaved caspase-3 and cleaved PARP1, apoptosis-associated proteins, were also detected to be highly expressed (Fig. 7A). A plate cloning formation assay and flow cytometry assay indicated that overexpression of IRE1 reduced colony formation and induced apoptosis in 143B cells (Fig. 7B and C). The results indicated that overexpression of IRE1 improved the CHOP expression and induced apoptosis in 143B cells.

Fig. 7.

IRE1 overexpression enhances CHOP expression and apoptosis in 143B cells. A Western blot analysis was used to evaluate the expression levels of IRE1, CHOP, cleaved caspase-3 and cleaved PARP1 in 143B cells transfected with the IRE1-OE plasmid. B The colony formation capacity of 143B cells was decreased upon IRE1 overexpression. C The apoptosis level of the 143B cells was reduced upon IRE1 overexpression, as determined by flow cytometry. Data are displayed as the mean ± SD (n = 3). *p < 0.05, **p < 0.01, vs. the control group

TGT plays an anti-osteosarcoma role by regulating the IRE1/CHOP pathway in vivo

To validate the therapeutic effect of TGT on osteosarcoma, a nude mouse xenograft model was constructed. Cisplatin was used as a positive control drug. TGT treatment was observed to significantly decrease the tumor volume and weight, but not as effective as those in the cisplatin group. However, the mice treated with TGT did not exhibit significant weight loss, which were similar to the control group, while the cisplatin group showed significant weight loss (Fig. 8A). H&E staining experiment displayed that the stained images of the heart, liver and lungs in the low, medium and high-dose treatment groups of TGT were not significantly different compared to the control group. There was also no distinct tissue damage to kidneys in the low-dose treatment group of TGT, despite occasional observation of eosinophilic substances within the renal tubules in the medium and high-dose treatment groups. Meanwhile, no tissue necrosis or inflammatory cell infiltration were observed in heart, liver, lungs and kidneys. However, cisplatin treatment group exhibited certain toxicity to the major organs, including the heart (with a small amount of myocardial cell necrosis), liver (hepatocyte swelling), lungs (scattered infiltration of granulocytes), and kidneys (with eosinophilic substances present within renal capsules and tubules) (Fig. 8B). The results indicated that TGT inhibited the growth of osteosarcoma in mice, exhibiting no obvious toxicity and an excellent safety in vivo. In line with the results of the in vitro experiments, TUNEL staining experiments also demonstrated that TGT treatment enhanced the apoptosis rate of tumor cells. The immunohistochemical results also illustrated that TGT treatment promoted the expression levels of p-IRE1, CHOP, and cleaved caspase-3 (Fig. 8C). Collectively, these results illustrated that TGT also implemented an anti-osteosarcoma function in vivo and the latent mechanism might be associated with ER stress activation.

Fig. 8.

TGT inhibits osteosarcoma growth in vivo through the IRE1/CHOP pathway. A Tumor morphology, body weight and tumor volume curve were displayed (mean ± SD, n = 8). B Representative H&E staining images of the heart, liver, lung and kidney tissues from mice in each group (400×, n = 3, Scale bar = 20 μm). C TGT-induced cell apoptosis (TUNEL staining, 400×) and enhanced cleaved caspase-3, p-IRE1 and CHOP levels (IHC staining, 400×) in the xenograft model (mean ± SD, n = 3, Scale bar = 20 μm). *p < 0.05, **p < 0.01, vs. the control group

Discussion

Osteosarcoma is the most common primary malignant cancer of the bone. Surgery and chemotherapy are standard therapies for osteosarcoma and result in an overall survival of 60 ~ 75% [42, 43]. However, limited therapeutic improvements have been achieved in the previous 40 years. Therefore, investigating novel strategy and new medicine for the treatment of osteosarcoma is necessary. Owing to its significant antitumor effects, TGT has shown considerable application prospects in clinical cancer treatment. Previous studies indicated that TGT promotes apoptosis in osteosarcoma 143B and SAOS2 cells in a dose-dependent manner, which was in accordance with the results of our study [34, 35]. However, the mechanism of TGT on osteosarcoma has not been investigated to date.

In the current research, TGT suppressed the proliferation, migration and invasion of 143B and MG-63 osteosarcoma cells in vitro. In order to explore the potential target of TGT in osteosarcoma, RNA-seq was applied to identify the DEGs in 143B cells that treated with TGT, and the potential target genes including DDIT3 and ERN1 were ultimately screened. The CHOP protein, which is encoded by DDIT3, is an important transcription factor involved in the ER stress response. Under natural conditions, the expression level of endogenous CHOP is generally very low. However, ER stress can be triggered under pathological conditions or upon microbial infection and is accompanied by increased CHOP expression and activated cell apoptosis [20]. Excessive expression of CHOP can also induce osteosarcoma cell apoptosis via multiple signaling pathways [11, 29]. In current research, HSPA5 and ERN1 were detected to be upregulated by TGT, which was verified by RT-PCR experiment. Similarly, high expression of HSPA5 and ERN1 was observed in the TGT 40 mg·mL^− 1 group by q-PCR verification experiment. Considering the enrichment of the biological process “IRE1-mediated unfolded protein response”, it is reasonable to speculate that TGT may regulate the IRE1/CHOP pathway to trigger osteosarcoma cell apoptosis [44].

Mounting evidence strongly suggests that ER stress can play both pro-survival and pro-apoptotic roles in cancer cells [7, 11, 45, 46]. Three kinds of signaling pathways, mediated by IRE1, PERK and ATF6, involved in the induction of the UPR are discovered, namely, the IRE1-, PERK- and ATF6-mediated signaling pathways. The UPR attempts to restore protein homeostasis but if unsuccessful may trigger cell apoptosis via a cascade that includes upregulation of CHOP, cleavage of caspase-3 and the inactivation of PARP1 [47, 48]. In IRE1-mediated ER stress, XBP1s is identified as a pivotal role. Notably, XBP1 is initially induced by ATF6 and spliced by IRE1 to generate its mature form, XBP1s, which is capable of binding to the promoter region of CHOP. The biding, in turn, upregulates CHOP expression, ultimately leading to cell apoptosis [49]. Consequently, the expression levels of p-IRE1, ATF6, BiP, XBP1s and CHOP proteins, which are involved in ER stress, were investigated in the present study. The results showed that the expression of the proteins was significantly increased after TGT treatment in osteosarcoma 143B cells, and IRE1 protein underwent phosphorylation modification. Studies have demonstrated that phosphorylation is required for the activation of IRE1 protein [50], indicating that TGT promoted the activity of IRE1 and subsequently activated downstream ER stress-associated signaling pathways. Furthermore, TGT also upregulated the expression of cleaved caspase-3 and cleaved PARP1, which indicated the induction of cell apoptosis. Previous studies have suggested that ATF6 fine-tunes the UPR, which is modulated mainly by the predominant IRE1α- and PERK-dependent responses [21, 23, 51]. However, the PERK expression was not affected by TGT in our study. Reports have shown that triggering ER stress through the IRE1 pathway promotes tumor cell apoptosis [9]. Therefore, this research indicated that TGT may enhance the dissociation of the IRE1-BiP complex to activate IRE1 and its downstream transcription factors XBP1 and CHOP.

IRE1 can inhibit or promote tumorigenesis in different contexts [52–54]. To clarify whether TGT induced osteosarcoma cell apoptosis through ER stress mediated by IRE1/CHOP pathway, IRE1 expression was knocked down by siRNA. The transfection of si-IRE1 led to decreased expression of the downstream transcription factors XBP1s and CHOP. Knockdown of IRE1 also resulted in decreased expression of ATF6 and BiP. BiP is a major ER chaperone and act as a primary sensor in the activation of the UPR. In resting cells, three key UPR activators, ATF6, PERK and IRE1, bind to BiP to maintain an inactive state. Upon ER stress, BiP is dissociated from the complex, leading to an enhancement of its expression level [55, 56]. Accordingly, the downregulation of IRE1 inhibited the occurrence of ER stress, leading to decreased BiP expression. A decrease in the expression of ATF6 and XBP1s contributed to the attenuation of ER stress size. Moreover, the expression of cleaved caspase-3 and cleaved PARP1 was markedly reduced in TGT treatment coupled with si-IRE1 group, as well as the apoptosis rate. These results were also further validated in the ER stress inhibitor, 4-PBA, treatment group [44]. Therefore, IRE1 might play a pivotal role in TGT-induced IRE1/CHOP pathway mediated ER stress, thereby regulating the apoptosis of osteosarcoma cells.

To further confirm our findings, IRE1 was overexpressed in 143B cells, which enhanced the expression of CHOP and apoptosis-associated proteins, cleaved caspase-3 and cleaved PARP1. Caspase-3, as a key executioner caspase in the apoptosis process, plays a crucial role in ER stress-induced apoptosis. ER stress activates caspase-3 indirectly or directly, which then cleaves substrate proteins (such as PARP1), subsequently leading to the execution of apoptosis [57, 58]. Consistent with previous studies in which CHOP promoted caspase-3 and PARP1 activation, these results indicated that overexpression of IRE1 promoted CHOP expression and triggered cell apoptosis. To conclude, the abovementioned results demonstrated that TGT stimulated osteosarcoma cell apoptosis through ER stress-associated IRE1/CHOP pathway in vitro.

Furthermore, the effect of TGT on transplantation tumor model was constructed. The in vivo validation confirmed that TGT stimulated osteosarcoma cell apoptosis through the ER stress-associated IRE1/CHOP pathway in vivo, accompanied by an increase in the expression of p-IRE1 and CHOP in tumor tissues. Interestingly, for the TUNEL staining, 40 g·kg^− 1 of TGT treatment resulted in a lower apoptosis rate compared to 20 g·kg^− 1 of TGT group. The reason might associate with the TGT-induced non-apoptotic dependent cell death, such as autophagy, necrosis or ferroptosis, which couldn’t be detected by the TUNEL assay staining. Our previous study showed that TGT could induce autophagy in 143B cells, although the autophagy was not the leading cause of TGT-induced cell death, which validated the speculation [36].

Increasing evidence strongly suggests that targeting ER stress is a promising strategy for cancer treatment [59]. There have been many attempts to target ER stress-associated proteins, including IRE1. Induced hyperactivation of IRE1 to switch a pro-survival condition to a pro-apoptosis condition may be a potential antitumor strategy [28, 60]. Consistent with these results, our study revealed that IRE1 protein underwent phosphorylation modifications in the treatment with TGT, which promoted the activity of IRE1 and subsequently activated downstream ER stress-associated signaling pathways, ultimately promoted osteosarcoma cell apoptosis to exert anti-tumor effect. However, due to different tumor type and context, efforts to improve cancer therapy by targeting IRE1 have primarily been limited to inhibiting its activity thus far [25]. CHOP, a pro-apoptotic regulator, is critical in inducing cancer cell apoptosis [26, 27]. In our study, CHOP was proved to be a downstream protein regulated by IRE1 in osteosarcoma. Our research also provided evidence that TGT targeted IRE1/CHOP pathway and exerted significant anti-osteosarcoma effect. Nevertheless, further clinical validation, including clinical efficacy, specificity and toxicity, is definitely required. There is still a long way to go from experimental research to clinical application.

Conclusion

Our study demonstrated that TGT exerts an effective anti-osteosarcoma effect by activating the IRE1/CHOP pathway, ultimately triggering the apoptosis of osteosarcoma cells (Fig. 9). Our results provide novel perspectives on the antitumor mechanisms of TGT and suggest its potential as a promising strategy for the treatment of osteosarcoma. However, the specific components in TGT responsible for the anti-osteosarcoma effects remains unknown. Further studies should prioritize identifying these components to better understand the mechanism and potential therapeutic applications. Additionally, the specific roles of CHOP and other proteins regulated by IRE1 in osteosarcoma, and their impact on cell fate, require further investigation.

Fig. 9.

TGT inhibits the growth of osteosarcoma through the ER stress-associated IRE1/CHOP pathway

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

TGT

Tongguanteng injection

DEG

Differentially expressed gene

4-PBA

4-phenylbutyric acid

RNA-seq

RNA sequencing

ER

Endoplasmic reticulum

siRNA

Small interfering RNA

IHC

Immunohistochemistry

UPR

Unfolded protein response

IRE1

Inositol-requiring enzyme 1

PERK

Protein kinase RNA-like ER kinase

ATF6

Activating transcription factor 6

XBP1

X-box binding protein 1

CHOP

C/EBP homologous protein

BiP

Binding immunoglobulin protein

GRP78

78-kDa glucose-regulated protein

HSPA5

Heat shock protein family A (Hsp70) member 5

ERN1

Endoplasmic reticulum to nucleus signaling 1

CCK-8

Cell counting kit-8

q-PCR

Quantitative PCR

STEM

Short time-series expression miner

GO

Gene ontology

KEGG

Kyoto encyclopedia of genes and genomes

IRE1-OE

IRE1 overexpression

SD

Standard deviation

BMSCs

Bone mesenchymal stem cells

Author contributions

All the listed authors contributed to the conception of the study and its design. X.-C. Xue, Y.-Y. Zhou, L.-Y. Wei and J. Yang performed the experiments. Y.-L. Han, J.-J. Chen, L.-Y. Xu, M.-Y. Wang and Y.-J. Hu were responsible for data analysis. X.-C. Xue, Y.-Y. Zhou and X.-Q. Zhang wrote the manuscript. Y.-L. Han, J.-J. Chen and L.-Y. Xu revised the manuscript. All authors read and approved the final version of the manuscript.

Funding

This study was supported by grants from the National Natural Science Foundation of China (No. 82003987), The Science and Technology Development Fund of Shanghai Pudong New Area (No. PKJ2020-Y08, No. PKJ2023-Y51), Health Industry Special Project of Shanghai Pudong New Area Health Commission (No. PW2021E-03), Shanghai Children’s Foundation (No. 202011).

Data availability

The RNA-seq data is publicly available at the NCBI Sequence Read Archive (http://www.ncbi.nlm.nih.gov/sra/) with the accession number PRJNA1105705.

Declarations

Ethics approval and consent to participate

The animal researches were accredited by the Animal Welfare Ethics Committee of Shanghai Sixth People’s Hospital (No. SYXK 2018-0028). All animal experiments were conducted rigorously in compliance with Provision and General Recommendation of Chinese Experimental Animals Administration Legislation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.