Metformin Protects Cardiomyocyte from Doxorubicin Induced Cytotoxicity through an AMP-Activated Protein Kinase Dependent Signaling Pathway: An In Vitro Study

Doxorubicin (Dox) is one of the most widely used antitumor drugs, but its cumulative cardiotoxicity have been major concerns in cancer therapeutic practice for decades. Recent studies established that metformin (Met), an oral anti-diabetic drug, provides protective effects in Dox-induced cardiotoxicity. Met has been shown to increase fatty acid oxidation, an effect mediated by AMP activated protein kinase (AMPK). Here we delineate the intracellular signaling factors involved in Met mediated protection against Dox-induced cardiotoxicity in the H9c2 cardiomyoblast cell line. Treatment with low dose Met (0.1 mM) increased cell viabilities and Ki-67 expressions while decreasing LDH leakages, ROS generations and [Ca2+]i. The protective effect was reversed by a co-treatment with compound-C, an AMPK specific inhibitor, or by an over expression of a dominant-negative AMPKα cDNA. Inhibition of PKA with H89 or a suppression of Src kinase by a small hairpin siRNA also abrogated the protective effect of the low dose Met. Whereas, with a higher dose of Met (1.0 mM), the protective effects were abolished regardless of the enhanced AMPK, PKA/CREB1 and Src kinase activity. In high dose Met treated cells, expression of platelet-derived growth factor receptor (PDGFR) was significantly suppressed. Furthermore, the protective effect of low dose Met was totally reversed by co-treatment with AG1296, a PDGFR specific antagonist. These data provide in vitro evidence supporting a signaling cascade by which low dose Met exerts protective effects against Dox via sequential involvement of AMPK, PKA/CREB1, Src and PDGFR. Whereas high dose Met reverses the effect by suppressing PDGFR expression.


Introduction
Doxorubicin (Dox), an anthracycline antibiotic, has been established as an agent against a wide range of cancers [1]. However, the severe cardiotoxicity of Dox is a major factor limiting its use in the treatment of many malignancies [2].
Intensive investigations of Dox-induced cardiotoxicity have been carried out. The different lines of evidence have provided putative mechanisms, but the precise mechanism underlying Doxinduced cardiotoxicity is not fully elucidated. Most studies favor the hypothesis that free radical-induced oxidative stress plays a pivotal role. This is supported by the chemical structure of Dox and its tendency to generate reactive oxygen species (ROS) during drug metabolism [3][4][5]. Recent findings indicate that endothelial nitric oxide synthase (eNOS) reductase domain converts Dox to an unstable semiquinone intermediate that favors ROS generation [5]. Although gaining less attention than ROS has received, a number of studies suggested that Dox-mediated alteration of Ca 2+ homeostasis is another possible mechanism of cardiotoxicity. Recent studies have demonstrated that Dox-mediated ROS generation induces increase of intracellular Ca 2+ ([Ca 2+ ] i ), which plays a critical role in damage of cardiomyocytes [6]. Metformin (Met) is an oral biguanide anti-hyperglycemic drug that is widely used for the management of type 2 diabetes mellitus. The therapeutic effects of Met have been attributed to a combination of improved peripheral uptake and utilization of glucose, decreased hepatic glucose output, decreased rate of intestinal absorption of carbohydrate, and enhanced insulin sensitivity [7,8]. Beyond its glucose lowering effects, Met has been shown to exhibit antioxidant properties in various tissues and acts to decrease lipid peroxidation, an effect that is independent of its effect on insulin sensitivity [9. 10]. Further, Met has been demonstrated to exert cardioprotective effects that could be due to its direct beneficial effects on cellular and mitochondrial function and therefore be independent of its insulin-sensitizing effect [11].
Through its activation of 59-adenosine monophosphate-activated protein kinase (AMPK), Met reduces the generation of ROS in cultured endothelial cells [12] and in animal models of heart failure [13,14] and protects cardiomyocytes from oxidative stress induced by H 2 O 2 or TNFa [14,15]. However, the specific mechanism by which Met activates AMPK and the corresponding antioxidant effect has not been established. These antioxidant effects suggest that Met could offer a protection against the cardiotoxicity of Dox, although no data are available to support additional benefits of Met in patients being treated with the anthracycline.
The present study was undertaken to delineate signaling pathways by which Met treatment evokes protective effects against the Dox induced cardiotoxicity. For this purpose, we studied Doxinduced in vitro toxicity in a fetal rat cardiomyoblast cell line, H9c2, human fetal cardiomyocyte cell line, RL-14 and rat neonatal primary cardiomyocyte. The results of this study provide evidence that the cardioprotective effects of Met are mediated by activation of the AMPK, PKA Src and platelet-derived growth factor receptor (PDGFR). Furthermore, the protective effects are suppressed with high dose Met (1 mM) treatment secondary to reduced cellular PDGF-receptor (PDGFR) expression.

Reagents and antibodies
Unless otherwise specified, all materials were reagent grade and obtained from Sigma-Aldrich (St. Louis, MO, USA). Anti-Ki67 antibody was obtained from BD Biosciences (San Jose, CA, USA). Alkaline phosphatase (ALP) conjugated horse anti-mouse IgG antibody was obtained from Vector Laboratory (Burlingame, CA, USA). Anti-phosphorylated/total AMPKa, anti-phosphorylated/ total acetyl-CoA carboxylase (ACC) and anti-phosphorylated PDGF receptor b (PDGFRb) antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA). Anti-phosphorylated/total CREB1, c-Src and total PDGFR-b antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-phosphorylated tyrosine antibody was obtained from Millipore (Billerica, MA, USA).

Cell viability assay
Cell viabilities were estimated using CellTiter-Blue Cell Viability Assay (Promega, Fitchburg, WI, USA). Briefly, viable cells retain the ability to reduce resazurin into resorufin, which is highly fluorescent. Nonviable cells rapidly lose metabolic capacity, do not reduce the indicator dye, and thus do not generate a fluorescent signal. Buffered solution containing highly purified resazurin was added to cells growing on 96-well microplates. The spectral properties of the buffer changed upon reduction of resazurin to resorufin. Fluorescence which was emitted from resorufin was measured with maximum excitation and emission spectra of 560 nm and 590 nm, respectively.

Lactate dehydrogenase release
Lactate dehydrogenase (LDH) is a cellular enzyme released upon membrane damage and a recognized marker of cell damage or death [16]. LDH released into the incubation medium was estimated using an assay kit from Sigma-Aldrich. In brief, LDH reduces nicotinamide adenine dinucleotide, which is then converted a tetrazolium dye to a soluble, colored formazan derivative; this was measured using a micro plate reader (490 nm).

Reactive Oxygen Species Assay
Cellular ROS was measured using a detection assay kit (Abcam, Cambridge, MA, USA). In brief, 29,79-dichlorofluorescein diacetate (DCFDA), a fluorogenic dye that measures hydroxyl, peroxyl and other ROS activity within the cell, was added to the cells growing in the 96-well plates. After diffusion into the cells, DCFDA was deacetylated by cellular esterases to a non-fluorescent compound, which was later oxidized by ROS into 29,79dichlorofluorescin (DCF), a highly fluorescent compound. Fluorescence from the DCF was detected by fluorescence micro plate reader with maximum excitation and emission spectra of 495 nm and 529 nm, respectively.

Determination of [Ca 2+ ] i
Levels of [Ca 2+ ] i were measured using fluo-4 (Molecular Probe, Eugene, OR, USA), a fluorescent Ca 2+ -indicator dye. Briefly, after removing the growth medium from the cells growing in 96-well microplates, 100 mL of dye loading solution (16 fluo-4 dye with 2.5 mM of probenecid) was added to each well. After incubation for one hour at 37uC, fluorescence from the fluo-4 was detected with a fluorescence micro plate reader with maximum excitation and emission spectra of 494 nm and 516 nm, respectively.

Immunohistochemistry
H9c2 cells were seeded on poly-L-lysine coated chamber slides. The cells were fixed with 2% formaldehyde and permeabilized by 0.2% TritonX-100, and incubated with a mouse monoclonal anti-Ki67 antibody overnight at 4uC in a humidified chamber. The cells were then incubated with an ALP conjugated anti-mouse IgG (H+L) secondary antibody for 30 min at room temperature. Bound antibody was detected using the ALP substrate kit (Vector Laboratories) and lightly counterstained with veronal acetate buffered 1% methyl green solution, pH 4.0. Permount (Fisher Scientific, Ottawa, Ontario, Canada) was used as the mounting media and sections were cover slipped. The immunohistochemical studies were repeated four times on samples prepared from independent cultures. The labeling index or the proportion of Ki67 positive cells was calculated according to the following formula: 1006 (the number of Ki67-positive nuclei/total number of nuclei). Each image was analyzed four times to obtain an average labeling index.

AMPK activity assay
Cellular AMPK activities were measured using an AMPK kinase assay kit (Cyclex, Ina, Nagano, Japan). Briefly, cell lysate samples were added to plates coated with a substrate-peptide corresponding to surrounding mouse IRS-1 serine 789 (S789), which contains serine residues that can be phosphorylated by AMPK. After washing, anti-phosphorylated mouse IRS-1 S789 monoclonal antibody was added, then the colorimetric reaction was developed by peroxidase conjugated anti-mouse IgG and tetramethylbenzidine substrate (TMB). The absorbance was measured at 450 nm using a micro plate reader.

Western blotting
Protein levels of the cell lysates for Western blotting (50 mg/ lane) were measured as described [17]. GelCode Blue (Thermo Scientific, Waltham, MA, USA) stain of the post transfer gel was used as the loading control for total and phosphorylated PDGFRb blotting. The results were visualized with Super Signal West Pico chemiluminescent substrate (Thermo Scientific) and analyzed with the UN-SCAN-IT gel software for Windows (Silk Scientific Inc., Orem, UT, USA).

PKA activity assay
Cellular PKA activity was measured using a PKA kinase activity kit (Enzo Life Sciences, Farmingdale, NY, USA). Briefly, cell lysates to be assayed were added to PKA substrate coated micro plate wells, followed by the addition of ATP to initiate the phosphorylation reaction. After terminating the kinase reaction, a phosphorylated substrate specific antibody was added to the wells. The phosphor-specific antibody was subsequently bound by peroxidase conjugated secondary antibody. The assay was developed with TMB. The absorbance was measured at 450 nm using a micro plate reader.

Src activity assay
Protein lysates (1 mg) from H9c2 cells were immunoprecipitated with anti-cSrc antibody. Kinase activity was determined by measuring phosphorylation of a specific Src substrate (KVEKI-GEGTYGVVYK) using a Src assay kit (Millipore). Briefly, the cSrc immunoprecipitated beads were incubated with a [c-32 P]ATP-ATP-Mg 2+ mix at 30uC for 10 min. A sample without the substrate peptide was included as a background control. Reactions were terminated by adding 40% trichloroacetic acid. After centrifugation the supernatants that include phosphorylated substrate were transferred onto Whatman P81 ionexchange cellulose chromatography paper circles (GE Healthcare, Little Chalfont, UK). The paper circles were washed five times in 0.75% phosphoric acid and once in acetone, and then counted in a liquid scintillation counter.

PI3K assay
PI3K activity was determined with in vitro immunoprecipitation lipid kinase assay. Briefly, cell lysates (0.5 mg) were immunoprecipitated with anti-phosphorylated tyrosine antibodycoated protein G-sepharose (GE Healthcare), and the beads were resuspend in assay buffer containing 300 mM adenosine to inhibit phosphoinositide 4-kinase (PI4K) activity [18]. L-a-phosphoinositide (Avanti Polar Lipid, Alabaster, AL, USA) was used as the lipid substrate (2 mg/reaction). After incubation, the final extracted reaction mixtures were spotted on to silica gel coated TLC plates (GE Healthcare), and were run in TLC buffer (65% n-propanol and 0.54 M acetic acid). The results were analyzed by phosphorimaging. Densitometric analysis was performed by using the UN-SCAN-IT gel software.

Stable transfection
Constructs of wild type (WT), dominant-negative (DN) and constitutively active (CA) AMPKa1 in pcDNA3.1 expression vector were generously provided by Prof. David Carling (MRC Clinical Sciences Centre, Imperial College, London, UK). A construct of constitutive active Src (Y529F) in pUSEamp-was purchased from Millipore.
Stable transfections of these vectors were performed using Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions. Individual single cells were isolated and selected with G418 (AMPKa1s and active Src transfected cells, 500 mg/ml) or puromycin (shRNA transfected cells, 5 mg/ml). Phenotypes of the transfected cells were evaluated by AMPK and Src activities (Fig.  S1, S2).

Statistical analysis
Statistics of the densitometric analysis were generated from four independent experiments. Statistical evaluations of the other assays (cell viability, LDH leakage, ROS generation, [Ca 2+ ] i , AMPK activity, PKA activity, and Src activity) were performed from four independent experiments which tested at least 10 samples each time.
Statistical significance of the difference among groups was analyzed by the paired Student's t test or parametric ANOVA and Ryan's multiple comparison test using Microsoft Exel (Microsoft, Redmond, WA, USA) and ANOVA4 on the Web (http://www. hju.ac.jp/,kiriki/anova4/). All data were represented as the mean 6 SD of at least four different experiments. A probability of p,0.05 was considered to represent a significant difference.

Effects of Met on Dox-induced cardiomyocyte toxicity
H9c2 cells were seeded in 96-well microplates (3610 3 cells/well) and quietened overnight in serum-free medium. In order to minimize the influence of serum on the metabolism of cells while keeping the cells in proliferative status, medium supplemented with reduced (1%, v/v) FBS was used in the experiment [19]. A concentration of Dox (10 nM) was determined to induce up to 40% of growth suppression after 72-96 hours (Fig.S3). The Met concentrations (0.1 and 1.0 mM) used in the experiments were adopted from published in vitro studies [20,21]. Cells were cultured in the reduced serum medium under various combinations of Dox and Met concentrations for up to 72 hours. After treatment with 10 nM of Dox for 72 hours, the cell viability was suppressed to 43.065.0% of the vehicle control level. Coincubation with 0.1 mM of Met reduced the suppression level to 31.166.2%. Co-incubation with higher concentration (1.0 mM) of Met, however, did not affect the effect of Dox on cell viability (46.063.7% of the control level; Fig. 1A). Dox induced a significant increase in LDH leakage to culture supernatants, another index of cellular damage, after 24 hours of incubation (373.16115.3% of the control level; Fig. 1B). The increase in LDH leakage was lessened by co-incubation of lower dose of Met (213.6644.9%) but not by the higher dose of Met (347.86104.6%). Furthermore, Dox induced a significant reduction of Ki-67 positive cells, (28.665.7% vs 5.161.4%, CTR vs Dox). The decrease in Ki-67 staining, again, was lessened by coincubation of lower dose of Met (14.161.1%) but not by the higher dose of Met (5.862.8%; Fig. 1C). Since Dox-induced cardiotoxicity may be related to cellular ROS generation [3][4][5] (Fig.S4A-D).
In order to confirm the effects of Met on other cardiomyocytes in altered stages of differentiation, cell viability, LDH leakage, ROS generation and [Ca 2+ ] i were evaluated using RL14 human fatal cardiomyocytes and rat neonatal primary cardiomyocytes. As shown in Figure 2A-D, Met showed similar effects with these cells.

Effects of AMPK activity on protection against Doxinduced cardiomyocyte toxicity
Recent advances in the understanding of Met action have centered on the discovery that Met leads to increased phosphorylation and activation of AMPKa [22,23]. Therefore, the relationship between the protection against Dox-induced cardiomyocyte toxicity and AMPK activity were examined in this study.
Co-incubation of Met (0.1 and 1.0 mM) with 10 nM Dox significantly enhanced cellular AMPK activity in H9c2 cells after 72 hour incubation (0.1 mM Met; 584.0688.3%, 1.0 mM Met; 1009.36127.2% of the control level), while incubation with Dox alone showed no effect (91.261.6%; Fig.3A). AMPK phosphorylates and inhibits acetyl-CoA carboxylase (ACC), the key enzyme that controls generation of malonyl-CoA from acetyl-CoA. As malonyl-CoA decreases fatty acid oxidation through inhibition of carnitine palmitoyl transferase-1 (CPT-1), phosphorylation of ACC relieves the inhibition of CPT-1, favoring fatty acid oxidation. In the current study, Met increased AMPK (0.1 mM;  (Fig. 3B). In order to verify the effects of AMPK activity on the Met induced effects, cells were co-incubated with of compound-C, an AMPK inhibitor. A concentration of the compound-C (10 mM) was determined that did not affect the cell viability while higher concentrations (20 mM, 40 mM) significantly reduce cell viability at 72-hour (Fig. S5). As shown in Fig. 3C, coincubation of compound-C completely reversed the effects of Met in attenuating Dox-induced reduction in cell viability (compound C-; 68.466.9%, compound C+; 51.965.6% of the control level). Furthermore, H9c2 cells were stably transfected with plasmids with wild type (WT), dominant-negative (DN) or constitutively active (CA) AMPKa1 cDNAs (Fig. S1). In DN-AMPKa cells the protective effect of 0.1 mM Met was completely abrogated (Fig. 3D). These findings suggested that the 0.1 mM Metmediated protective effects were dependent on the AMPK activity. Some clones of the CA-AMPKa1 transfected H9c2 cells showed extremely high AMPK activities which were comparable to those of cells treated with 1.0 mM of Met (Fig. S1). Interestingly, the CA-AMPKa1 transfected cells which obtained extremely high AMPK activities did not show protective effects against Doxinduced toxicity (Fig. 3D, Fig. S6). These results suggest that Met protected cardiomyocytes through moderately enhanced AMPK activities and the protective effect was reversed if the AMPK activity exceeded a certain threshold.

Effects of PKA activity on protection against Dox-induced cardiomyocyte toxicity
In previous studies, we described PKA as a crucial factor in cell survival of isoproterenol stimulated H9c2 cells [24,25]. To determine the involvement of PKA in the Met mediated protective effects, PKA activities were measured by in vitro kinase assay. In the cells co-incubated with Met, PKA activities were significantly elevated (Met 0.1 mM; 187.6637.6%, Met 1.0 mM; 211.6681.5% of the control level; Fig. 4A). Dox treatment alone had no effect on either PKA activity (108.2%623.  Fig.4D). Furthermore, the Met induced PKA activities were reversed by coincubation with compound-C (compound C-; 187.6637.6%, compound C+; 112.7619.1% of the control level; Fig. 4E). These data suggested that the protective effect of Met was mediated by the PKA activation, and the PKA activation was dependent on the AMPK activity.

Met-mediated protective effect is dependent on Src family tyrosine kinase
We have shown that the Src-family tyrosine kinase is involved in bAR-mediated anti-apoptosis in H9c2 cells [25]. In the present study, to explore the role of Src in Met-mediated cell protection, a series of experiments were initiated. First, cells were cultured with vehicle, Dox alone, or Dox in combination with Met for 72 hours, and then the cellular Src activity was measured. As shown in Second, cSrc was knocked down by shRNA transfection. We have reported effective knockdown of Src at the RNA and protein levels in H9c2 cardiomyocyte with this approach [24,25]. In this study, H9c2 cells were stably transfected with a scrambled oligo control vector (pSilencer) or shSrc. Knock down of Src effectively obliterated the protective effect of 0.1 mM Met on the Doxinduced toxicity (pSilencer; 69.066.0%, shSrc; 35.766.0% cell viability of the control level; Fig. 5B).  Fig.5D), which was observed in non-transfected H9c2 cells (Fig. 5C), suggesting PKA acts upstream of Src in this signaling pathway. These observations suggest that Src is a critical factor in the Metinduced anti-cytotoxic effect by functioning downstream of AMPK/PKA signaling pathway.

Expression of PDGFR is down regulated by co-incubation with 1.0 mM Met
We have shown that PDGFR plays a pivotal role in survival of H9c2 cells [24,25]. In order to explore the roles of PDGFR signaling in Met-induced survival in Dox-treated H9c2 cells, experiments were performed as follows. Cellular expression and phosphorylation levels of PDGFR were evaluated by western blotting. As shown in Fig. 6A, the phosphorylation levels of PDGFRb was increased in 0.1 mM Met treated rat neonatal primary cardiomyocytes and H9c2 cells. In contrast, the expression levels of the receptor were significantly suppressed in 1.0 mM Met treated cells. The PDGFR expression was also downregulated in the CA-AMPK transfected H9c2 cells which showed extremely high AMPK activity (Fig. S7). With the hypothesis that an AMPK/PKA/Src/PDGF sequence was a critical pathway for the Met induced cardiomyocyte protection and to verify the functional insufficiency of PDGFR response against PKA stimulation in 1.0 mM Met treated cells. We performed an experiment using activities of phosphoinositide 3kinase (PI3K), a downstream molecule of PDGFR signaling, as an index for PDGFR sensitivities against forskolin. In 0.1 mM Met treated cells, forskolin stimulation induced a significant increase in PI3K activity while 1.0 mM Met treated cells showed no effect (Fig. 6B). Furthermore, co-incubation with AG1296, a PDGFR specific antagonist, abrogated 0.1 mM Met induced protective effect against Dox-mediated cytotoxicity in H9c2 cardiomyocytes (Fig. 6C). These findings suggested that PDGFR signaling is a crucial factor in Met-induced protective effect against Doxmediated cytotoxicity of H9c2 cells. The reversal of the effect with the higher concentration (10 mM) of Met is a result of the down regulation and the functional loss of PDGFR.

Discussion
In this study, we demonstrated that metformin (Met), an antidiabetic agent, has a clearly protective effect against doxorubicininduced toxicity on cardiomyocytes through activation of AMPK. Whether Met, as a consequence of its modulated metabolism, influences cardiomyocyte survival remains unknown. We have previously reported a pro-survival/proliferation pathway in cardiomyocyte [24,25] and renal mesangial cells [26,27] involving G-protein coupled receptor (GPCR)-PKA-Src-receptor tyrosine kinase. The present study provides evidences that AMPK and PKA are activated sequentially following Met treatment. However, the mechanisms of AMPK dependent PKA activation are not fully clarified. A signaling interaction between AMPK and PKA was described in a hypothalamic cell line [28] and AMPK mediated CREB, a downstream transcription factor of a cAMP/ PKA signaling pathway, activations were reported in hepatocyte [29] and neuronal cells [30]. Met/AMPK-induced transactivations of several GPCRs were also described in a pancreatic b-cell line [31]. All of the findings in the previous publications suggest high probability for the existence of AMPK/PKA/CREB or AMPK/GPCR/PKA/CREB signaling cascades in some cells.
Src has been identified as a key effector of PKA signaling [32,33]. In this study, we showed a pivotal role of the Src kinase as we have previously reported [24][25][26]. We have demonstrated that GPCR/PKA/Src to receptor tyrosine kinase (RTK) link is a notable pro-survival signaling pathway in cardiomyocytes and renal mesangial cells. In the present study, we provide convincing evidence that AMPK activation is critical in Met-mediated resistance against the Dox-induced cytotoxicity and that this protective effect was accomplished via sequential activation of PKA/Src/PDGFR. A very novel and interesting finding in this study is the dual effects of Met on cardiomyocyte survival. We showed that, at lower concentrations (0.1 mM), Met protected cardiomyocytes from the Dox-induced toxicity, whereas a higher concentration (1.0 mM) of Met failed to do so despite the fact that higher concentration of Met induced increases in many parameters we measured including AMPK. PKA, CREB1 and Src with even more potent manners than those with a lower concentration of Met. Moreover and most notably, a higher concentration of Met showed similar effects on ROS generations and [Ca 2+ ] i in Dox intoxicated cardiomyocytes. Our data suggested that the biphasic effect was caused by dose dependent alteration in PDGFR expression. Excessive AMPK activity in 1.0 mM Met treated cells may induce the suppression of the PDGFR expression. And the attenuated PDGFR signaling may be a factor to wipe out the Met induced effects of decreased ROS generation and [Ca 2+ ] i against Dox treatment. In the meanwhile, the elevated PDGFR activity in cells treated with the lower concentration of Met may overcome the Dox induced cardiotoxicity to maintain cell viabilities. However, this hypothesis is still premature because of lack of the bibliographical evidences to support it. Further investigations should be addressed to provide a logical explanation for these unexpected findings.
PDGF was originally identified in serum and platelets as a strong mitogen for fibroblasts, smooth muscle cells, and glial cells [34]. PDGF signaling plays important roles in the pathogenesis of several proliferative and degenerative diseases such as tumorigenesis, arteriosclerosis, and fibrosis [35]. In the present study, we demonstrated that the higher dose of Met resulted in a significant reduction of the levels of PDGFR. In contrast, lower dose of Met did not reduce the levels of PDGFR but enhanced the cellular activity of the receptor tyrosine kinase. More important, the protective effect of lower dose of Met is abrogated by PDGFR antagonist; clearly PDGFR signaling is important for the dual effect of Met-mediated protection against Dox toxicity.
In the last few years, several studies concerning about the protective effects of Met against the Dox toxicity has been published elsewhere [36][37][38]. Interestingly, in these papers, they demonstrated that even a dose of 4 mM of Met were able to protect cardiomyocytes in culture from the cytotoxic effect of doxorubicin. The cause of the discrepancy between our and their findings is not elucidated at present. Supposedly, differences in the lineage of the cells used (H9c2 vs HL-1) and dosage of Dox applied (10 nM vs 5 mM) may deduce to the inconsistency. Elucidating details in this discrepant action of Met on cardiomyocyte will benefit further understanding a mechanism of the protective effect of Met against cardiotoxicity of Dox.
The major proteolytic pathway involving the ubiquitin-proteasome system (UPS) is dependent on ATP [39]. Activation of AMPK results in the stimulation of a variety of cellular processes involved in the production of ATP, e. g., glucose uptake [40], protein synthesis [41] and UPS-mediated protein degradation [42]. In the present study, AMPK activity in 1.0 mM Met treated cells was significantly higher than those in 0.1 mM Met treated cells (Fig. 3A). Furthermore, constitutively active AMPKa cDNA transfected cells, which had even higher AMPK activities than those of 1.0 mM Met treated cells, showed suppressed PDGFR expression (Fig. S1, S6). Considering these findings, AMPK activities beyond a certain threshold may promote PDGFR degradation in H9c2 cardiomyocytes. Elucidating further details in this effect on PDGFR expression should be addressed in the future.
We have investigated the roles of AMPK and PKA as crucial factors in Met induced resistance against Dox toxicity on H9c2 cardiomyocytes. We demonstrated that PDGFR transactivation is involved in this pathway. We further established that Src played a pivotal role in the signaling pathway by functioning between PKA and PDGFR. We also described cellular PDGFR expression levels as regulatory factor for the protective effect of Met on cardiomyocytes. Based on these findings, despite the fact that bibliographic references to support this hypothesis are limited at present, we propose a hypothetical pathway for the Met-mediated protective effect against the Dox-induced toxicity on cardiomyocyte (Fig. 7). Although there are other components needed to be identified in this signaling pathway, our findings nonetheless provide important information for the protective effects of Met which has attracted attention recently. Elucidating further details in this signaling pathway should lead to better understanding over the conventional chemotherapy for malignant neoplasms.