Distinct mechanism of cervical cancer cell death caused by the investigational new drug SHetA2

Abstract

Drug-targetable vulnerabilities of cancer cells include their dependence on heat shock proteins (HSPs) to support elevated mitochondrial metabolism and counteract cell death factors. The investigational new drug SHetA2 targets these vulnerabilities in ovarian and endometrial cancer cells by disrupting complexes of the mortalin HSP with its client proteins (mitochondrial support proteins, metabolic enzymes, p53) leading to mitochondrial leakage of cytochrome c and apoptosis-inducing factor (AIF), and caspase-dependent apoptosis. Our objective was to evaluate the roles of mitochondrial damage and another SHetA2-target HSP protein, cytoplasmic heat shock cognate 70 (hsc70), in the mechanism of SHetA2 killing of cervical cancer cells. Cervical cancer cells responded to SHetA2 with excessive mitophagy that did not deter AIF leakage into the cytoplasm. Then, hsc70 was unable to prevent cytoplasmic AIF nuclear translocation and promotion of DNA damage and cell death, because SHetA2 disrupted hsc70/AIF complexes. The Cancer Genome Atlas analysis found that overexpression of hsc70, but not mortalin, was associated with worse cervical cancer patient survival. Use of specific inhibitors documented that AIF and mitophagy, but not caspases, contributed to the mechanism of SHetA2-induced cell death in cervical cancer cells. As validation, excessive mitophagy and lack of caspase activation were observed in SHetA2-inhibited xenograft tumors.

Keywords: SHetA2; apoptosis inducing factor; cell death; cervical cancer; heat shock cognate 70; heat shock protein; mitochondria; mitophagy.

Figure 1

SHetA2 induces mitochondrial depolarization and loss of ATP, (A): Relative fluorescence (RFU) of JC-1 dye, and ratios of JC-1 aggregate to monomer expressed as mean ± SD and analyzed using two-way ANOVA. (B), ATP levels in cells treated with SHetA2 (10 µM) or vehicle analyzed using t-tests. (C), Representative histograms from Flow cytometric analysis of MitoSOX staining of cells treated with SHetA2 (10 µM) or vehicle for 24 hours (left) and average (mean ± SD) data of three independent experiments were analyzed by t-tests (right). **p ≤ 0.01, ****p ≤ 0.0001 when compared with respective control.

Figure 2

SHetA2 reduces mitochondrial networking and mass: (A, B), Confocal imaging of MitoTracker™ Green staining of cells treated with 10 µM SHetA2 (A, upper panel). Representative imaging of mitochondrial networks analyzed using Image J software (A, lower panel). Maximum mitochondrial branch lengths were compared by t-tests (B). (C), C-33 A, Ca Ski and SiHa cervical cancer cells treated with 10 µM SHetA2 or vehicle for 24 hours were stained with MitoTracker™ Green and Hoechst. The fluorescence was imaged and analyzed by using the Operetta^®High Content Imaging System. Mitochondrial mass in cells treated with SHetA2 or vehicle for 24 hours were measured by ratios of MitoTracker™ Green to Hoechst staining, presented as mean ± SD and analyzed using t-tests. ****p ≤ 0.0001 when compared with respective control.

Figure 3

SHetA2 alters mitochondrial morphology and downregulate mitochondrial dynamics proteins: (A), TEM of SiHa cells treated with SHetA2 (10 µM) or vehicle for 4 and 8 hours. Representative images showing mitochondria (M), nucleus (N), nuclear condensation (NC) and autophagic vesicles (V) are shown (A, upper panel). For the quantification of mitochondrial size, mitochondria with cristae structure were measured for length, width and total area and compared by t-tests (A, lower panel). (B), Western blots of cells treated with DMSO or 10 µM SHetA2 for 48 hours (upper panel). GAPDH (for C-33 A, Ca Ski, SiHa, and C-4-II), or cyclophilin B (for SiHa) or α-tubulin (for ME-180) were used as loading controls (L. Control). Densitometric analysis of the bands are shown as mean ± SD and were compared using a t-test (lower panel). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001 NS-not significant when compared with respective control.

Figure 4

SHetA2 induces mitophagy in cervical cancer cells: (A, B), Mitophagy induction was demonstrated by confocal imaging of cervical cancer cells treated with SHetA2 (10 µM) for 24 hours and stained with Pink1 (green) mitotracker red (red) and DAPI (blue) (A), or with MitoView green (green) and Lyso Deep Red (red) dye (B).

Figure 5

SHetA2 induced mitophagy contribute to cervical cancer cell death: (A), Western blot analysis of Ca Ski and SiHa cell treated with SHetA2 (10 µM) for 48 hours (left panel). Densitometric analysis of the bands (Pink-1, p62 and LC3-II/LC3-I ratio) are shown as mean ± SD and were compared using a t-test (right panel). (B), Western blot analysis of cytoplasmic and mitochondrial fractions of Ca Ski and SiHa cell treated with SHetA2 (10 µM) for 48 hours (left panel). Densitometric analysis of the bands are shown as mean ± SD and were compared using a t-test (right panel). (C), Cervical cancer cells were pre-treated with 10 µM Mdivi-1 for 4 hours followed by SHetA2 (10 µM) for 72 hours and MTT assay was performed. ****p ≤ 0.0001 when compared with respective control.

Figure 6

SHetA2 inhibits metabolic viability of cervical cancer cell in caspase-independent manner: (A), Caspase 3 activity in cervical cancer cells treated for 48 hours with vehicle, 10 µM SHetA2, or a combination of pan-caspase inhibitor (Z-VAD-FMK, 30 µM, pre-treatment for 3 hours) and SHetA2. A one-way ANOVA was used for statistical analysis. (B), Representative dose response curves of C-33 A, Ca Ski and SiHa cells pretreated with 30 µM of pan-caspase inhibitor (Z-VAD-FMK) for 3 hours followed by SHetA2 treatment at different doses for 72 hours. ***p ≤ 0.001, ****p ≤ 0.0001 when compared with respective control.

Figure 7

SHetA2 induces AIF mediated DNA damage contributing cell death: (A), Confocal microscopy of cells stained with anti -AIF (green) or -γH2AX (green), mitotracker (red) and DAPI (blue) after treatment with 10 µM SHetA2 for 24 hours. (B), Western blots of cells transfected with either non-target si-RNA (si-CTL) or si-AIF and treated with SHetA2 (10 µM) (left panel) and analyzed by densitometry analysis (right panel). GAPDH (Ca Ski, SiHa), or cyclophilin B (SiHa) were used as loading controls (L. Control). (C), MTT assays of cells transfected with either si-CTL or si-AIF. (D), cervical cancer cells were transfected with either si-CTL or si-AIF. Following pretreatment with Mdivi-1, transfected cells were treated with SHetA2 at indicated concentration for 72 hours and MTT assay was performed. *p ≤ 0.05, ***p ≤ 0.001, ****p ≤ 0.0001 when compared with respective control.

Figure 8

Role of hsc70 in cervical cancer and SHetA2 mediated cell death mechanism: (A), Co-immunoprecipitation assays of cervical cancer cells treated with 10 µM SHetA2 for 24 hours. (B, C), UALCAN analysis of TCGA data for expression of mountain (HSPA9, B) and hsc70 (HSPA8, C upper panel) mRNA in cervical squamous cell carcinoma (CESC) compared to healthy tissue and comparison of their high versus low expression with cervical squamous cell cancer patient survival probability (lower panel B, C).

Figure 9

SHetA2 induces mitophagy and inhibits growth of cervical cancer tumor in-vivo: (A), Average tumor volumes of Ca Ski xenograft in mice treated with orally SHetA2 (30 and 60 mg/kg) or vehicle for 21 days. The average tumor volume during the treatment period (left panel) and tumor weight (right panel) were compared by two-way and one-way ANOVA, respectively. (B), expression of cleaved caspase-3 was measured on the xenograft tumor tissue treated with SHetA2 (60mg/kg group) or control by Immunohistochemistry and the representative images are shown (left panel). The H score was compared between both groups by using Student’s t-tests (right panel). (C), TEM images of Ca Ski xenograft tumors control or 30 mg/kg/day SHetA2 treatment groups. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ns; not significant when compared with respective control.

Figure 10

Model of SHetA2 mechanism of causing cervical cancer cell death.