Mecheliolide elicits ROS-mediated ERS driven immunogenic cell death in hepatocellular carcinoma

Abstract

The nonnegligible reason for the poor prognosis of hepatocellular carcinoma (HCC) is resistance to conventional chemotherapy. Immunogenic cell death (ICD) is a rare immunostimulatory form of cell death that can reengage the tumor-specific immune system. ICD can improve the clinical outcomes of chemotherapeutics by promoting a long-term cancer immunity. The discovery of potential ICD inducers is emerging as a promising direction. In the present study, micheliolide (MCL), a natural guaianolide sesquiterpene lactone, was screened out by the virtual screening strategies, identified as an inhibitor of thioredoxin reductase (TrxR) and was evaluated to have high potential to induce ICD. Here, we showed that MCL induced ICD-associated DAMPs (damage-associated molecular patterns, such as CRT exposure, ATP secretion and HMGB1 release). MCL significantly triggered the regression of established tumors in an immunocompetent mouse vaccine model, and induced ICD (DCs maturation, the stimulation of CD4⁺, and CD8⁺ T-cells responses) in vivo. Mechanistically, we found that the magnitude of ICD-associated effects induced upon exposure of HCC cells to MCL was dependent on the generation of reactive oxygen species (ROS)-mediated endoplasmic reticulum stress (ERS). In addition, the suppression of ROS normalized MCL-induced ERS, in contrast, the downregulation of TrxR synergized with the ERS driven by MCL. We also systematically detected the H₂O₂ generation using Hyper7 sensors in HCC cells exposed to MCL. Notably, MCL inhibited the development of HCC organoids. Collectively, our results reveal a potential association between the TrxR inhibitors and ICD, presenting valuable insights into the MCL-activated ICD in HCC cells.

Graphical abstract

Fig. 1

The comprehensive screening process of TrxR inhibitors. (A) The screening process of discovering potential TrxR inhibitors from TCM database. The figure was created with BioRender.com. (B) Ten potential TrxR inhibitors were finally screened out and their structures were exhibited. (C) The key reaction between the Sec residue of TrxR and MCL were displayed by the molecular docking.

Fig. 2

MCL inhibits the TrxR activity and the growth of two HCC cells. (A) and (B) Analysis of cell viability using CCK-8 following in a time-dependent manner. Each panel shows the HCC cells (HepG2 and Hepa 1–6) were treated with MCL for 24 h, 48 h and 72 h. (C) The TrxR activity was tested through the DTNB monitoring through UV absorption. (D) The kinetic curve of TrxR inhibition by MCL. (E) The Sec depletion mutant of TrxR activity was evaluated in a concentration-dependent manner following the DTNB assay. (F) The pharmacophore of MCL (α-methylene-γ-lactone) was reduced, the new compound named Re-MCL was synthesized, and the Re-MCL against TrxR activity was tested. (G) and (H) The binding affinity of MCL with intracellular TrxR was assessed by the CETSA assays in HepG2 and Hepa 1–6 cells. The TrxR expression was tested by the Western blots. Statistics of grey intensity was normalized in different temperatures. n = 3; ns > 0.05; *, P < 0.05; **P < 0.01; ***, P < 0.001 when compared with control group.

Fig. 3

MCL covalently binds with Sec and Cys of TrxR. (A) MCL was incubated with wild type TrxR, and the adducts were analyzed by the MS. The adduct with m/z (372.7) was Cys-MCL; and the adduct with m/z (418.9) was Sec-MCL. (B) and (C) MCL was incubated with two TrxR mutants (USec498 TrxR and Truncated form of TrxR), two adducts with m/z (371.1) were Cys-MCL of USec498 TrxR and Truncated form of TrxR, respectively.

Fig. 4

MCL inhibits the TrxR in two HCC cells (HepG2 and Hepa 1–6) and triggers the ROS generation. (A) and (B) After the treatment of cells with MCL in a concentration manner for 24 h, the TrxR was detected by the TrxR green-probe. (C) and (D) MCL could inhibit the intracellular TrxR in HepG2 and Hepa 1–6 cells after the exposure of 24 h. (E) and (F) ROS generation (DCFH-DA staining) of MCL toward HCC cells, control knockdown cells and TrxR knockdown cells. These cells were treated with MCL (30 μM) (Scale bars: 50 μm). (G) and (H) The fluorescence intensity of ROS level was subsequently analyzed. (I) and (J) Growth inhibition in HCC cells, control knockdown cells and TrxR knockdown cells. These cells were treated with MCL for 24 h n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with control group. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

MCL selectively reduces the viability of HCC cells. (A) The TrxR expression was analyzed in three cell lines. (B) The viability of three cell lines was investigated after MCL treatment for 24 h n = 3; **, P < 0.01; ***, P < 0.001 when compared with the LO₂ group.

Fig. 6

MCL ROS-dependently induces apoptosis effects of two HCC cells. (A) and (E) HepG2 and Hepa 1–6 cells were treated with MCL (concentration-dependent) for 24 h followed by the assessment of Annexin V/PI staining. NAC pretreatment was performed at 4 mM for 1 h. (D) and (H) The quantification of cell death rate following Annexin V/PI staining. (B) and (F) The signature proteins of apoptosis in two HCC cells challenged with MCL (concentration-dependent) were analyzed by the Western blot assays. (C) and (G) The quantification of the immunoblots was analyzed. n = 3; ns > 0.05; * or ^#, P < 0.05; ** or ^##, P < 0.01; ***, P < 0.001 when compared with control group.

Fig. 7

MCL induces the ROS generation in a concentration-dependent manner. (A) and (B) After the exposure to MCL in two HCC cells (A: HepG2; B: Hepa 1-6), the ROS level was detected by the DCFH-DA probe. NAC pretreatment was performed at 4 mM for 1 h. (C) and (E) The ROS level of each group was analyzed by the flow cytometry, (D) and (F) the median was then calculated and analyzed. n = 3; ns > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with control group. Scale bar: 50 μm.

Fig. 8

H₂O₂ generation (HyPer response) in different cell compartments of HepG2 and Hepa 1–6 cells. (A) Imaging of two HCC cells H₂O₂ sensitive HyPer in different cell localizations after the exposure to the MCL (30 μM). NAC pretreatment was performed at 4 mM for 1 h. HyPer response was used to display the location in different cell compartments (mitochondria, nucleus, peroxisome and ER membrane (green); cytosol (red)). (B) The flow cytometry was carried to analyze the HyPer response in different cell localizations. (C) and (D) After the treatment of 30 μM MCL, the relative H₂O₂ response in different localizations of two HCC cells are shown. Data were the results of flow cytometry analysis. Each bar represents the mean of three replicates. Scale bar: 10 μM. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 9

MCL induces ROS-dependent ERS in HepG2 cells. (A) The images of HepG2 cells exposed to MCL through the electron microscopy, the arrow decorations represent the ER. (B) and (E) ER stress-related proteins of HepG2 cells were analyzed by the Western blot assays. The cells were treated with different concentrations of MCL. NAC pretreatment was performed at 4 mM for 1 h. Densitometric quantification of ERS-related proteins. (C) and (F) ER stress-related proteins of HepG2 cells were by the Western blot assays. The cells were treated with MCL (30 μM) for indicated time points. Densitometric quantification of ERS-related proteins. (D) and (G) The ERS-related proteins were assessed following the transfection of HepG2 cells with siRNA against TrxR before exposed to MCL (30 μM). Densitometric quantification of ERS-related proteins. n = 3; ns > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with control group (The Alpha-Tubulin and phospho-eIF2α were used as loading control).

Fig. 10

The prerequisite steps of ICD effects in HCC cells.The figure was created with BioRender.com.

Fig. 11

MCL induces DAMPs of HepG2 and Hepa 1–6 cells. (A) and (B) After the treatment of MCL (30 μM) toward two cells for 24 h, the images of CRT and HMGB1 were detected by the immunofluorescence. (C–F) The flow cytometry analysis of CRT and HMGB1 in two cells following exposed to MCL (30 μM) for 24 h is exhibited. (G–J) The mean of the flow cytometry results was calculated and analyzed. (K) and (L) DAMPs related proteins of HepG2 and Hepa 1–6 cells were analyzed by the Western blot assays. The cells were treated with MCL in different concentrations. NAC pretreatment was performed at 4 mM for 1 h. (M) and (N) Densitometric quantification of proteins (HSP 90, CRT and HMGB1). n = 3; ns > 0.05; * or ^#, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with control group (The Alpha-Tubulin was used as loading control). Scale bar: 10 μM.

Fig. 12

MCL promotes DCs maturation. (A) The scheme of murine DCs isolation, differentiation, and maturation analysis. 200× magnification. The figure was created with BioRender.com. (B) After the MCL or DOX treatment, the treated Hepa 1–6 cells were then co-cultured with immature DCs for another 24 h. The CD80 and CD86 positive cells were analyzed by the flow cytometry. (C) Quantification of mature DCs following the CD80 and CD86 staining. (D) The intracellular ATP level (HepG2 and Hepa 1–6 cells) was analyzed after MCL treatment in a concentration-dependent manner. (E) After the two HCC cells exposed to the MCL (30 μM) in different time points, the intracellular ATP was analyzed by the ATP ELISA assays. n = 3; ns > 0.05; **, P < 0.01; ***, P < 0.001 when compared with control group.

Fig. 13

MCL shows specific immune response (vaccine effects) in advanced HCC suppression. (A) The scheme of tumor incubation and treatment approach. (B) The tumor free curve of Hepa 1–6 tumor on the left side in mice after the different treatments (n = 7). (C) The renascent tumor volume in the left side of each group. (D) The changes of mice body weight.

Fig. 14

Therapeutic efficacy of MCL on suppressing cells in Hepa 1–6 tumor-bearing C57BL/6 mouse model. (A) Representative photographs of dissected tumors of each group. (B) H&E and TUNEL immunofluorescence staining of the tumor tissues in each group. Scale bar: 100 μm. (C–E) The tumor volume changes; the tumor weight and the body weight of mice were recorded. (F) and (H) The TrxR immunofluorescence stainings of tumor tissues after various treatments were analyzed. Scale bar: 100 μm. (G) The Western blot analysis of TrxR expression in tumor tissues of each group. (I) The TrxR activity of tumor tissues in three groups were analyzed. (J) The DHE and DCFH-DA immunofluorescence staining of tumor tissues and fluorescence intensity were detected and analyzed. Scale bar: 100 μm n = 4; ns > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with model group.

Fig. 15

MCL elicits ERS and induces DAMPs (ICD-related markers) in vivo. (A) The proteins (Grp78, Calnexin, ATF-4, p-eIF2α and Chop) immunofluorescence staining of the tumor tissues. Scale bar: 100 μm. (B) The ERS associated proteins of tumor tissues were analyzed by the Western blot assay. Densitometric quantification of ERS-related proteins. (C) The DAMPs (HSP 90, CRT and HMGB1) immunofluorescence staining of the tumor tissues. Scale bar: 100 μm. (D) The Western blot assays were carried out to assess the DAMPs expression of tumor tissues of each group. n = 4; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with model group.

Fig. 16

MCL remodels immune response of Hepa 1–6 tumor-bearing C57BL/6 mice. (A) The flow cytometric examination of mature DCs (gated on CD80⁺ and CD86⁺ cells) of spleen tissues. (B) and (C) The CD4 T cells (gated on CD4⁺ and CD3⁺ cells) and the CD8 T cell infiltration (gated on CD8⁺ and CD3⁺ cells) in tumor tissues were analyzed by the flow cytometry. (D) and (E) The proinflammatory cytokines including TNF-α and IFN-γ in the serum were analyzed by the ELISA assays. n = 4; *, P < 0.05; **, P < 0.01; ***, P < 0.001 when compared with model group.

Fig. 17

MCL significantly inhibits the growth of HCC organoids (50 ×). n = 3; ns > 0.05; **P < 0.01; ***, P < 0.001 when compared with control group.

Fig. 18

The schematic illustration of underlying mechanism of ICD effects inducing by MCL. The figure was created with BioRender.com.