Targeting the BAG2/CHIP axis promotes gastric cancer apoptosis by blocking apoptosome assembly

Abstract

Apoptosis has been shown to play an important role in the treatment of gastric cancer, and BCL2-associated athanogene 2(BAG2) has been found to be able to inhibit apoptosis by interacting with multiple apoptosis regulators. In this study, we demonstrate that BAG2 functions as an independent prognostic factor, correlating with unfavorable clinical outcomes in patients with gastric cancer (GC). We demonstrate that BAG2 upregulation inhibited apoptosis and increased proliferation, migration, and invasion of GC cells, whereas the opposite results were obtained in BAG2-deficient GC cells. Mechanistically, BAG2 interacts with the c-terminus of HSP70-interacting protein(CHIP) to inhibit the ubiquitination degradation of Heat shock protein70(HSP70) and increase the binding of HSP70 to apoptotic protease-activating factor 1(Apaf1). The reduced ubiquitination degradation of HSP70 reduces the release of mitochondrial cytochrome C (Cytc), which ultimately inhibits the formation of apoptotic bodies assembled by Cytc and Apaf1. The above effects of BAG2 inhibit the formation of Cytc and Apaf1-assembled apoptotic bodies. Furthermore, we screened FIIN-2, an inhibitor of the BAG2 complex, which effectively halts the malignant development of GC triggered by reduced apoptosis by blocking BAG-CHIP binding. In conclusion, this study highlights BAG2's key role in regulating apoptosis and confirms FIIN-2's effectiveness in GC-targeted therapy.

Keywords: BAG2; FIIN-2; apoptosis; apoptosome; gastric cancer.

Figure 1

Overexpressed BAG2 correlates with poor clinical outcomes in GC, promoting GC cell line proliferation, invasion, and migration. (A) BAG2 transcription level in TCGA database (https://portal.gdc.cancer.gov). (B) Prognostic analysis of BAG2 in GC revealed a correlation between high BAG2 expression and poor prognosis, as reported in TCGA. (C) Kaplan–Meier survival analysis was conducted to compare the survival outcomes of patients categorized into BAG2-low (n = 75) and BAG2-high (n = 77) groups. (D) Representative images of H&E staining and IHC analysis were used to visualize BAG2 expression, and the resulting IHC scores were statistically analyzed. The p values were determined using a two-sided nonparametric test, based on a sample size of 152 independent biological samples. (E) The growth curves of several GC cell lines were plotted. Data are presented as means ± SEM. The p values were determined using a two-sided nonparametric test and one-way ANOVA (n = 6 independent biological samples). (F) Colony formation assays were conducted in several GC cell lines, and the resulting colonies were counted and statistically analyzed. Data are presented as means ± SEM. The p values were determined via one-way ANOVA (n = 3 independent biological samples). (G, H) The effects of BAG2 on GC cell invasion and migration were assessed. Data are presented as means ± SEM. The p values were determined via one-way ANOVA (n = 3 independent biological samples). **p<0.01, ***p<0.001.

Figure 2

BAG2 KO promotes apoptosis in GC cells. (A) Annexin V-FITC/propidium iodide (PI) and Annexin-V APC/7AAD staining tested HGC-27 and AGS cell apoptosis using flow cytometry. Data are presented as means ± SEM. The p values were determined via one-way ANOVA (n = 3 independent biological samples). (B) TUNEL staining showed apoptotic changes in HGC-27 GC cells. Scale bar: 100 nm. (C) Transmission electron microscopy was used to observe the apoptosome morphology in HGC-27 cells. Scale bar: 50 nm. (D) Immunoblotting analysis was performed to examine the protein expression levels of apoptosis factors (including caspase-3 and caspase-9) in BAG2-KO and wild-type cell lines. (E) Representative images of xenograft mice carrying WT or BAG2-KO MKN-45 cell xenografts. Mice were sacrificed when tumors reached 100 mm3 in size; data are presented as means ± SEM. The p values were determined through a two-sided nonparametric test (n = 6 independent mice per group). (F, G) Tumor weight and growth curve of xenografts of MKN-45 cells with WT or BAG2-KO in mice are presented as means ± SEM. The p-value was determined via two-tailed nonparametric testing and one-way ANOVA (n = 6 independent biological samples). (H) Representative intratumor IHC images of caspase-3, caspase-9, and Ki-67, along with quantification of Ki-67-positive cells, are presented for mice xenograft tumors treated with BAG2 KO. (I, J) TUNEL staining was used to detect the effect of BAG2 KO on apoptosis in tumor tissues (n=3). Scale bar: 100 nm. (K) Representative western blot images of caspase-3, caspase-9, and GAPDH protein expressions in tumor tissues in BAG2-KO and control groups. ***p<0.001.

Figure 3

BAG2 regulates GC cell apoptosis through the CHIP-HSP70-Apaf1/Cyt-c axis. (A) Co-immunoprecipitation assays of Flag/HA-tagged BCL2, BAG2, CHIP, or HSP70 co-expressed in HEK293T cells. IP, immunoprecipitation; WCL, whole-cell lysates. (B) Immunofluorescence colocalization of BAG2 and CHIP/HSP70 in AGS cells. Cells were immunostained with anti-BAG2 antibody (red), anti-CHIP antibody (red or green), anti-HSP70 antibody (green) and DAPI (blue). (C) Co-immunoprecipitation of HA-tagged CHIP and FLAG-tagged HSP70 was performed in HEK-293 cells transfected with No-tagged BAG2. (D) HEK-293T cells were transiently transfected with plasmids encoding HA-tagged HSP70, along with the indicated amounts of a plasmid encoding Flag-BAG2 24 h after transfection. Cell lysates were analyzed via western blotting with the indicated antibodies. (E) HEK-293T cells were transiently transfected with plasmids encoding Flag-tagged HSP70 along with plasmids encoding HA-tagged wild ubiquitin or indicated mutant ubiquitin. Sixteen hours after transfection, cells were treated with MG132 for 8 h (10 μM). Cell lysates were analyzed via immunoprecipitation with anti-Flag and western immunoblotting with indicated antibodies. (F) Co-immunoprecipitation assays of Flag-tagged HSP70 and HA-tagged Apaf1 co-expressed in HEK-293T cells and co-immunoprecipitation assays of Flag-tagged Apaf1 and HA-tagged Cytc co-expressed in HEK-293T cells. IP, immunoprecipitation; WCL, whole-cell lysates. (G) Immunofluorescence colocalization of DAPI, Cytc, and mitochondria in AGS cells. Cells were immunostained with anti-Cytc antibody (green), MitoTracker^® Red CMXRos (red), and DAPI (blue).

Figure 4

BAG2 regulates the proliferation and apoptosis of GC cells dependent on HSP70. (A) Representative H&E and IHC staining of BAG2 and HSP70 in GC TMAs. (B) Scatter plots of BAG2 versus HSP70 H scores in human gastric TMAs. P values were determined using a two-sided Spearman’s rank correlation test (n = 185 independent biological samples). (C) Western blot was performed to validate the knockdown efficiency of three specific siRNAs against HSP70 in GC cells (n=3). (D) An MTT assay was conducted to validate the effect of HSP70 knockdown on the proliferation of GC HGC-27 cells. (E) Flow cytometry was performed to validate the effect of HSP70 knockdown on the apoptosis of HGC-27 cells(n=3). *p<0.05, **p<0.01, ***p<0.001.

Figure 5

FIIN-2 blocks the BAG2-CHIP interaction. (A) Interactions between CHIP and different BAG2 deletion mutants analyzed via co-immunoprecipitation assays. WT, wild type; IP, immunoprecipitation; WCL, whole-cell-lysates. (B) Interactions between BAG2 and different CHIP deletion mutants were examined using co-immunoprecipitation assays. (C) The bound conformation of BAG2 and CHIP as predicted by the Cluspro algorithm. BAG2 is displayed in yellow, and CHIP is displayed in green. (D) This schematic diagram shows the amino acids that interact between BAG2 and CHIP. On the binding surface, the CHIP residue bonds are highlighted in green, while those of BAG2 are in yellow. (E) Interactions between BAG2 and different CHIP deletion mutants containing residues on the binding surface of the mode were analyzed via co-immunoprecipitation assays. (F) Flow diagram of BAG2-CHIP complex inhibitor screening. (G) Computational modeling showcases the interactions between FIIN-2 and CHIP. CHIP is displayed in green, and FIIN-2 is displayed in pink. (H) Microscale thermophoresis (MST) was utilized to ascertain the kinetic constant (Kd) for the interaction between FIIN-2 and CHIP. (I) Co-immunoprecipitation assays of the BAG2-CHIP interaction in cells treated with FIIN-2 at the indicated concentrations in HGC-27. IP, immunoprecipitation; WCL, whole-cell lysates. (J) Western blotting was conducted to assess HSP70 expression levels in cells post-treatment with different FIIN-2 concentrations in HGC-27. (K) An in vitro ubiquitination assay was performed to determine the impact of FIIN-2 (C = 10 μM) on HSP70 ubiquitination, using specified recombinant proteins in HGC-27.

Figure 6

FIIN-2 is efficient in GC treatment in vitro. (A) Sensitivity of HGC-27 and AGS cells to FIIN-2 at various concentrations. Cell proliferation was evaluated after 48h of treatment. The results are expressed as means ± SEM. The p values were determined via one-way ANOVA (n = 4 independent biological samples). (B-D) Effect of FIIN-2 treatment on tumor (B) proliferation, (C) invasion, and (D) migration ability were detected via colony formation, Transwell, and wound-healing assays. Student’s t-test was used to examine statistical significance (mean ± SD, n = 3, ****p < 0.0001, ***p < 0.001,**p < 0.01, *p < 0.05). (E, F) Effect of FIIN-2 treatment on tumor apoptosis were detected via flow cytometry and western blot analysis (n=3).

Figure 7

FIIN-2 is efficient in treating GC in vivo and in organoids. (A-C) MKN45 cells were intratumorally injected in nude mice. The compound was dosed by intraperitoneal injections at a single dose of 20mg/kg/day×21, and DMSO was used as a control group. Shown are (A) representative images, (B) tumor weights, and (C) tumor volumes. Data are the mean ± s.e.m. The p values were determined by one-way ANOVA (n = 6 independent biological samples). (D) The effect of drug administration on caspase-3 and caspase-9 expression in tumor tissues was assessed using western blot analysis. (E) Representative intratumor IHC images of caspase-3, caspase-9, and Ki-67, along with quantification of Ki-67-positive cells, are presented for mice xenograft tumors treated with FIIN-2. Data represent the mean ± SEM. The p values were determined using one-way ANOVA with n = 6 independent biological samples. (F) The effect of FIIN-2 on GC organoids at various concentrations(n=3). (G) A schematic diagram of the mechanism by which FIIN-2 obstructs the progression of GC by blocking the BAG2-CHIP complex and regulating apoptosis mediated by HSP70 ubiquitination. *p<0.05 is used to indicate a difference, **p< 0.01 to indicate a significant difference.