Doxorubicin induced ROS-dependent HIF1α activation mediates blockage of IGF1R survival signaling by IGFBP3 promotes cardiac apoptosis

Dox-induced ROS production and apoptosis in a dose-dependent manner in embryo-derived cardiac cells

Earlier studies indicated that Dox can induces oxidative stress and apoptotic cell death. To evaluate the impact of Dox on embryo-derived cardiac cells (the H9c2 cell line), we used the TUNEL assay. Treatment with Dox at 0.5, 1.0, or 1.5 μM for 24 h induced cell apoptosis in a dose-dependent manner in H9c2 cells (Figure 1A). Further, MitoSOX staining was performed to evaluate the effect of Dox challenge on mitochondrial superoxide generation. Mitochondrial superoxide generation was significantly increased by Dox administration in a dose-dependent manner (Figure 1B). Moreover, we analyzed the effect of Dox on various oxidative stress markers. Western blot analysis indicated that the protein expression levels of pro-oxidative stress markers, such as the NADPH oxidase subunits NOX2, p47, and p22^phox were elevated following Dox challenge (Figure 1C). Although there was a dose-dependently increased trend, however, only 1.5 μM reached to significant level. Additionally, we evaluated the impact of Dox on different apoptosis markers. Immunoblotting data showed that the protein expression level of the anti-apoptotic marker (Bcl-xL) was downregulated significantly, whereas that of the pro-apoptotic markers (Bax and caspase 3) were elevated by Dox challenge (Figure 1D), although there is no significant difference in the levels of Bax compared with control, while in the case of caspase 3, only 1 μM Dox dose had a significant difference. These results confirm that Dox can induce ROS generation and apoptosis mediated cell death in embryo-derived cardiac cells. Furthermore, following a 6-week Dox treatment, rat cardiac functions were evaluated using an echocardiogram (Table 1). We observed that several parameters, especially ejection fraction (EF) and fractional shortening (FS), that assess the function of the left ventricle, were significantly lower in Dox-treated rats. This is in line with the notion that Dox induces cardiac damage.

Figure 1. Doxorubicin increases generation of ROS and augments apoptosis in cardiac cells. H9c2 cells challenged with increasing doses of doxorubicin (Dox) for 24 h were harvested. (A) Apoptosis as detected using the TUNEL assay. (B) Mitochondrial superoxide production measured using MitoSOX staining. (C) The protein expression levels of the NADPH oxidase subunits NOX-2, p47, and p22 were detected using immunoblotting. (D) Protein levels of Bcl-xL, Bax, and cleaved caspase 3 were analyzed using western blot. Data are presented as mean ± standard deviation (n = 3). Scale bar represents 100 μm. Statistical significance is indicated as follows: ^*P < 0.05, ^**P < 0.01, ^***P < 0.001.

Dox increases cardiac IGFBP3 expression and secretion as well as decreases cardiomyocyte survival in vitro and in vivo

Expression of IGFBP3 and that of the components of IGF1 survival signaling pathway were assessed in cardiac cells exposed to 0.5, 1.0, and 1.5 μM Dox for 24 h. We found that IGFBP3 and cytochrome c protein levels were increased by Dox, whereas the protein expression levels of the survival pathway components, such as phosphorylated PI3K and AKT, were lower in the cells challenged with Dox (Figure 2A). Next, we measured the levels of these proteins in the ventricle heart tissue of Dox-treated and control rats using western blot and found similar changes to those seen in embryo-derived cardiac cells. Further, pro-apoptosis-related protein cleaved caspase 3 and Bax levels were highly increased in the ventricle tissue of Dox-administered rats (Figure 2B). Importantly, immunohistochemistry results showed that cardiac IGFBP3 accumulation was higher in the Dox-treated rats compared with that in the control animals (Figure 2C). Additionally, Dox treatment increased IGFBP3 secretion was revealed by higher IGFBP3 levels in both the cell culture medium of Dox-treated cardiac cells and serum of Dox-treated rats (Figure 2D). These results indicate that Dox could trigger cell apoptosis by increasing cardiac IGFBP3 expression and secretion coupled with the downregulation of IGF1 signaling.

Figure 2. Doxorubicin increases expression and secretion of IGFBP3, and decreases cardiac cell survival in vitro and in vivo. (A) H9c2 cells were incubated with Dox for 24 h at the indicated doses and protein expression of IGFBP3 and that of the components of survival signaling pathway were measured using immunoblotting. (B) Left ventricles of control and Dox-administered rats were isolated and the levels of IGFBP3 and survival-related proteins were analyzed using western blotting. (C) IGFBP3 expression from the rat left ventricle tissue was detected using immunohistochemistry. (D) H9c2 cells were challenged with Dox for 24 h at indicated doses and the amount of secreted IGFBP3 in both cell medium and in the animal serum of Dox-treated rats was examined using western blotting. The quantitative plot of IGFBP3 from the sera of doxorubicin challenged rats is included. Scale bar indicates 200 μm.

Downregulation of IGFBP3 prevents Dox-induced loss in IGF1 survival signaling, and increase in apoptosis and ROS generation

The above obtained data showed enhanced IGFBP3 expression and secretion following Dox treatment, which suppressed IGF1 signaling. Therefore, we hypothesized that IGFBP3, secreted in higher amounts into the extracellular space following exposure to Dox, binds IGF1 and blocks the survival pathway, thereby promoting cell apoptosis. Indeed, the co-IP data showed higher association of IGFBP3 with IGF1 in Dox-exposed cardiac cells (Figure 3A). Further, the levels of the phosphorylated IGF1 signaling pathway component phosphorylated AKT, were upregulated by the application of an anti-IGFBP3 antibody. Likewise, levels of the pro-apoptotic protein Bax and caspase 3, which were increased by Dox, were dose-dependently reduced by the treatment with an anti-IGFBP3 antibody (Figure 3B). Moreover, we applied Igfbp3 siRNA approach to verify that Dox downregulated IGF1 survival signaling and increased pro-apoptotic protein levels by potentiating IGFBP3 expression. Treatment with Si-Igfbp3 dose-dependently rescued the reduction of IGF1 survival signaling and increase in the levels of pro-apoptotic proteins (Figure 3C). Notably, TUNEL assay further revealed that Dox induced apoptosis (Figure 3D). Moreover, MitoSOX staining indicated that ROS generation was reverted to normal upon IGFBP3 inhibition using either anti-IGFBP3 or knockdown approach (Figure 3D, 3E). These results indicate that the Dox-induced inhibition of IGF1 survival signaling, cardiac cell apoptosis, and oxidative stress are possibly mediated by the enhancement of cardiac IGFBP3 expression.

Figure 3. Doxorubicin enhances extracellular binding of IGFBP3 to IGF1, blocking IGFI survival signaling and promoting cell apoptosis. H9c2 cells were treated with 1.0 μM Doxorubicin (Dox) for 24 h. (A) Extracellular association of IGF1 with IGFBP3 was detected using co-IP. (B, C) Dox-challenged cells were incubated with either (B) anti-IGFBP3 antibody at indicated amounts or transfected with (C) Igfbp3 siRNA. The levels of pro-survival and apoptosis-related proteins were analyzed using western blotting. (D, E) H9c2 cardiomyoblasts challenged with Dox (1.0 μM) for 24 h were either incubated with anti-IGFBP3 antibody and/or transfected with Si-Igfbp3 and thereafter probed with TUNEL and MitoSOX reagents to evaluate the apoptotic cell death (D) and mitochondrial superoxide generation (E). Data are expressed as mean ± standard deviation (n = 3). Scale bar represents 100 μm. Statistical significance is showed as follows: ^*P < 0.05, ^**P < 0.01.

Effects of Dox on ROS generation, IGFBP3 and mediated cardiac apoptosis are mediated by HIF1A

Previous studies have indicated that upon hypoxia exposure, cardiac HIF1A translocates from cytosol to the nucleus enhancing IGFBP3 expression and triggering cell apoptosis [14]. Therefore, we sought to verify whether HIF1A expression is induced by Dox. From the western blot analysis, we found that HIF1A protein expression was increased dose-dependently in H9c2 cardiac cells upon Dox challenge. Additionally, higher HIF1A levels were noted in the ventricles of Dox-treated rats compared with that in the normal group (Figure 4A). The observation of immunofluorescence microscopy demonstrated increment in Dox for 24 h induced HIF1A translocation from the cytosol to the nucleus (Figure 4B). Further, we examined the effects of Hif1a siRNA and HIF1A inhibitor on Dox-exposed cardiac cells. We found that Dox mediated decrease in the level of the pro-survival protein, phosphorylated AKT, was reverted using siHif1a and HIF1A inhibitor in a dose-dependent manner. In addition, exposure to the HIF1A inhibitor also reversed Dox-induced upregulation of IGFBP3 expression and secretion (Figure 4C, 4D). Moreover, TUNEL and MitoSOX staining revealed that Dox-induced cardiac apoptosis and the enhanced mitochondrial ROS generation were reversed upon Hif1a knockdown and inhibition (Figure 4E, 4F). These results indicate that Dox-induced ROS production enhanced IGFBP3 expression and secretion, downregulated IGF1 survival signaling, and induced apoptosis of embryo derived cardiac cells in a HIF1A dependent manner.

Figure 4. Dox-induced ROS production, cell apoptosis, downregulation of pro-survival signaling, and upregulation of IGFBP3 expression are dependent upon HIF1A activity. (A) H9c2 cells were incubated with increasing dose of Dox for 24 h. HIF1A protein levels were analyzed in H9c2 cells and animal cardiac tissue using western blotting. (B) Nuclear translocation of HIF1A in cells challenged with or without Dox were detected using fluorescence microscopy. Blue represents nucleus; green represents HIF1A. (C, D) Dox-exposed cells were treated with HIF1A inhibitor (C) or transfected with Hif1a siRNA (D). The levels of IGFBP3, pro-survival, and pro-apoptotic proteins in cells and cell culture medium (for secreted IGFBP3) were detected using western blotting. (E, F) Dox challenged H9c2 cells either treated with HIF1A inhibitor or transfected with siHif1a, were incubated with TUNEL and MitoSOX reagents to assess apoptosis mediated cell death (E) and mitochondrial ROS production (F). Statistical significance difference was shown as ^*P < 0.05, ^**P < 0.01 from at least three independent experiments.

Dox-induced ROS inhibits PHD, thereby enhancing HIF1A-IGFBP3 signaling, leading to increased cardiac apoptosis

Previous studies show that mitochondrial ROS suppress the expression of PHD proteins. Under normoxic conditions, HIF is hydroxylated by PHD-containing enzymes and undergoes polyubiquitination and proteasomal degradation by VHL E3 ubiquitin ligase [23]. Using western blotting, we measured the levels of the PHD protein in Dox challenged cardiac cells and tissue of rats administered with Dox and found that PHD expression was decreased in both models (Figure 5A). Further, we showed that the ROS scavenger NAC dose-dependently reversed the decrease in the levels of the phosphorylated pro-survival protein AKT, as well as attenuated the increase in the levels of HIF1A, IGFBP3 (cellular expression and secreted amount), and pro-apoptotic proteins Bak and caspase 3 induced by Dox exposure (Figure 5B). In order to determine the source of ROS following the exposure to Dox, we examined the effects of apocynin, an NADPH oxidase inhibitor that inhibits the cytosolic ROS, and rotenone, a complex 1 inhibitor that inhibits the generation of the mitochondrial ROS. We found that both inhibitors reversed the Dox-induced downregulation of PHD and p-PI3K levels and the upregulation of HIF1A, IGFBP3 (cellular expression and secreted amount), Bax, and pro-apoptotic caspase 3 (Figure 5C). Furthermore, we used PHD overexpression plasmid to investigate whether PHD downregulation is involved in the activation of HIF1A by Dox. Immunoblotting data showed that overexpression of PHD reversed the Dox-induced enhancement of the levels of HIF1A, IGFBP3 (cellular expression and secreted amount), and pro-apoptotic proteins as well as the attenuation of p-AKT levels (Figure 5D). These results indicate that ROS levels elevated by Dox stress, decreased the PHD expression, which in turn enhances HIF1A-IGFBP3 signaling leading to downregulation of cellular survival and stimulation of apoptosis in cardiac cells.

Figure 5. Dox-induced mitochondrial ROS stabilizes HIF1A through the downregulation of PHD and promotes IGFBP3-induced cardiac apoptosis. (A) H9c2 cells and tissues from rats administered with indicated concentrations of Dox were subjected to western blotting. (B, C) Dox-exposed cells were treated with the ROS scavenger NAC (B), mitochondria complex I inhibitor rotenone (Rot) or the NADPH oxidase inhibitor Apo (C). The harvested cellular extract was analyzed using western blotting. (D) Cells were transfected with a PHD overexpression plasmid of the indicated amount and levels of HIF1α, IGFBP3, pro-survival, and pro-apoptotic proteins in cells and cell culture medium (for secreted IGFBP3) were measured by western blot analysis. Data represents as the mean ± standard deviation of the mean (n = 3). Statistical significance is represented as follows: ^*P < 0.05, ^**P < 0.01.

Altogether, the current study suggests that Dox-induced ROS generation inhibits PHD expression, leading to translocation of HIF1A to the nucleus and enhancing the expression as well as secretion of IGFBP3 and thereby blocks IGF1 pro-survival signaling, resulting in increased cardiac cell apoptosis (Figure 6).

Figure 6. Graphical representation illustrating the mechanism underlying Dox-induced cell apoptosis. Dox induced ROS generation, which suppressed PHD. This stabilized nuclear HIF1A, promoted IGFBP3, and enhanced its extracellular association with IGF1. This interaction blocked survival signaling and resulted in cell apoptosis.

Cell culture experiments

H9c2, an rat embryo derived cardiomyoblast cell line acquired from the American Type Culture Collection (Manassas, VA, USA) were cultured in Dulbecco’s modified essential medium supplemented with 10% cosmic calf serum, 2 mM glutamine, 100 units/mL penicillin, and 100 μg/mL pyruvate in the humidified atmosphere of 95% air and 5% CO₂ at 37°C. During the treatment, H9c2 cells were exposed to 1 μM Dox for 24 h, harvested, and extracted for further analysis.

In vivo experiments

We used 8-week-old male Wistar rats that were purchased from the National Animal Breeding and Research Center (Taipei, Taiwan). All the animals were divided into two groups: normal group (N = 5) and Dox-treated group (N = 5). The animals were maintained under a 12-h light/dark cycle at an ambient temperature of 25°C. The treated group received intraperitoneal injections of Dox at a dose of 5 mg per kg body weight once a week for 6 weeks. The normal group received injections of the vehicle. All the experiments in this study were approved by the Ethical Committee Faculty of China Medical University [39, 40].

Tissue extraction

Cardiac tissue samples from the control and Dox-challenged rats were homogenized for protein extraction in a lysis buffer at a concentration of 100 mg tissue/mL buffer. The homogenates were placed on ice for 15 min and then centrifuged at 12,000 × g for 30 min at 4°C. The supernatant was collected and stored at −80°C for further analysis.

Lowry protein assay

Total protein concentration was determined with the Folin-Ciocalteu reagent for enhanced color development using the Lowry method. Diluted bovine serum albumin solutions (0, 0.1, 0.2, 0.3, 0.4, and 0.5 mg/mL) were used as standards. Proteins were first reacted with alkaline cupric sulfate in the presence of tartrate (2% Na-K tartrate: 1% CuSO₄ × 5H₂O: 2% Na₂CO₃ in 0.1 N NaOH = 1:1.98) for 10 min at room temperature. After the incubation, the Folin phenol reagent was added. The color enhancement occurs when the tetradentate copper complex transfers electrons to the phosphomolybdic/phosphotungstic acid complex. Following incubation for 40 min at room temperature in the dark, the absorbance was recorded at 750 nm.

Immunohistochemistry (IHC)

IHC staining was performed as described previously [41, 42]. Cardiac tissues from the rats were deparaffinized with xylene, followed by rehydration using gradient series of alcohol. Thereafter, they were permeabilized, blocked, and washed twice with phosphate buffered saline (PBS) followed by primary antibody incubation for 1 h. Next, tissues were washed with PBS and incubated with horseradish peroxidase-conjugated avidin biotin complex using Vectastain Elite ABC Kit and NovaRED chromogen (Vector Laboratories, Burlingame, CA, USA) followed by hematoxylin staining to stain the nucleus. Finally, the photomicrographs were acquired and processed using microscopy (OLYMPUS^® BX53, Tokyo, Japan).

Western blot analysis

Western blotting was performed as delineated in our previous studies [43, 44]. Briefly, cultured H9c2 cells with 70–90% confluency were harvested and lysed using a lysis buffer containing 50 mM Tris (pH 7.5), 0.5 M NaCl, 1.0 mM EDTA (pH 7.5), and 10% glycerol. The suspension was centrifuged at 12,000 rpm for 35 min and supernatant was collected. The collected proteins were separated by 10–12 % sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane. Nonspecific protein binding was inhibited using 5% skimmed milk. The membrane was probed with specific primary antibodies in the blocking buffer at 4°C overnight. Thereafter, PVDF membranes were washed and incubated with the secondary antibody for 1 h followed by densitometric analysis using Fuji LAS 3000 imaging system. For repeated blotting, PVDF membranes were stripped with the Restore Western Blot Stripping Buffer at room temperature for 30 min.

Transfection of plasmid and siRNAs

Cardiac cells with up to 50% confluency were replaced with fresh culture medium containing serum, treated with 1 μM Dox for 24 h, and transfected with either pCMV-PHD3 plasmid or siRNA against Hif1a and Igfbp3, using PureFection™ Nanotechnology-based Transfection Reagent (System Biosciences, CA, USA), according to the manufacturer’s protocol. In each experiment, the efficiency of gene overexpression was measured using western blot analysis. For the knockdown experiments, cells with 70–90% confluency were used Double-stranded siRNA sequences targeting Hif1a and Igfbp3 mRNAs were obtained from EMD Millipore Corporation. The knockdown efficiency was confirmed using western blot analysis.

Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay

To assess apoptosis, H9c2 cells grown on 6 mm plates were fixed with 4% formaldehyde solution for 30 min at room temperature. After washing thrice with PBS, cells were permeabilized in 0.1% Triton X-100 solution in 0.1% sodium citrate for 10 min at 4°C. Following two washes with PBS, the samples were first incubated with the TUNEL for 1 h at 37°C in the dark. Then, cells were stained with DAPI for 2 min for nuclear detection. Fluorescence microscopy images of TUNEL-positive cardiac myocytes and apoptotic bodies were obtained using excitation wavelength in the range of 450–500 nm and recording the emission spectrum in the range of 515–565 nm (green).

Measurement of mitochondrial ROS generation

Generation of mitochondrial ROS was evaluated in embryo-derived cardiac cells using MitoSOX (Invitrogen Molecular Probes, MA, USA). After H9c2 cells were challenged with Dox for 24 h, they were probed with MitoSOX reagent for 30 min at 37°C followed by DAPI staining for 5 min. ROS generation was evaluated using fluorescence microscopy (Olympus, Tokyo, Japan), by excitating the probes at 510 nm and recording the emission at 580 nm.

Immunofluorescence

Immunofluorescence was executed to analyze the expression level of HIF1A. H9c2 cardiac cells cultured in chamber slides (SPL Life Sciences, Korea) were challenged with Dox for 24 h. They were then fixed in 4% paraformaldehyde for 1 h, permeabilized with 0.1% Triton X-100 for 5 min followed by blocking for 1 h to avoid nonspecific binding. Next, H9c2 cells were incubated with the HIF1α primary antibody for 1 h and Alexa Fluor 488 Red anti-rabbit IgG secondary antibody for 1 h. Lastly, cells were washed, and stained with DAPI, and then imaged to analyze the expression of HIF1A.

Co-immunoprecipitation (Co-IP)

To evaluate protein-protein interactions, the total cellular extract from H9c2 cells was immunoprecipitation using Protein G magnetic beads (Millipore), according to the guidelines provided by the manufacturer. From the whole cell lysate 500 μg of protein was probed with 3 μg of the primary antibody overnight at 4°C. Finally, the protein complexes were washed with PBS and eluted at 95°C. The resultant elute was analyzed by the western blotting method as described above.

Statistical analysis

Data values are expressed as the means ± standard deviation. The significance of the treatment effect was assessed by one-way analysis of variance using the GraphPad Prism5. All western blots and immunofluorescence images were analyzed using ImageJ software. Differences were considered to be statistically significant for P < 0.05.