Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A number of lines of evidence suggest a role of oxidative stress in cardiac diseases. The level of oxidants increases in the myocardium as a result of ischemia and ischemic reperfusion (Flitter, 1993; Goldhaber, 1997). As estimated with experimental animals, the concentration of H₂O₂ reaches 6 µM with ischemia and rises to 11 µM during reperfusion in the myocardium (Goldhaber, 1997). The myocardium accumulates oxidative damage over the course of heart failure (Hill and Singal, 1996; Keith et al., 1998; Singh et al., 1995). Elevated levels of oxidative damage markers are observed in patients with congestive heart failure and in animal models of chronic heart failure (Dhalla et al., 1996; Hill and Singal, 1996; Keith et al., 1998). Oxidative stress is thought to be a major contributor to the cardiac toxicity of the antineoplastic drug doxorubicin, which can cause acute cardiac toxicity when administrated at high doses or cause chronic heart failure at mild doses or accumulated low doses above a certain threshold (Singal and Iliskovic, 1998; Singal et al., 2000). Although epidemiological studies show that antioxidant vitamins reduce the risk of coronary heart disease and the mortality rate associated with this disease (Rimm et al., 1993; Kushi et al., 1996), clinical trials provide promising evidence that vitamin E reduces the rate of secondary myocardial infarction (Stephens et al., 1996). Animal experiments reveal a beneficial effect of vitamin E therapy in reducing the rate of heart failure induced by pressure overload or doxorubicin (Dhalla et al., 1996; Singal et al., 2000). However, regardless of these positive outcomes of antioxidants in preventing heart failure, it is not clear how oxidants cause or accelerate heart disease at the mechanistic level.

Cardiac hypertrophy is a common endpoint of many cardiovascular diseases (Colucci and Braunwald, 1997). An increase in the size of cardiomyocytes is the key feature of most enlarged hearts. In end-stage heart failure, enlarged cardiomyocytes are often observed regardless of whether the heart volume or mass is increased (Colucci and Braunwald, 1997). Enlargement of cardiomyocytes has been reported to result from an increase in protein content and activation of p70S6K1 (Sadoshima and Izumo, 1995; Boluyt et al., 1997). p70S6K1 plays a major role in regulating the phosphorylation of 40S ribosomal S6 protein and selective translation of a family of mRNAs that contain an oligopyrimidine tract at the 5'-transcriptional start site (Dufner and Thomas, 1999). These mRNA species make up 20 to 30% of total translated mRNAs and encode components of the translational apparatus important for cell growth (Dufner and Thomas, 1999). Activation of p70S6K1 results from phosphorylation of specific threonine (Thr) and Serine (Ser) residues such as Thr389 and Thr421/Ser424 (Dufner and Thomas, 1999). Phosphorylation of Thr389 and activation of p70S6K1 can be inhibited by rapamycin, which inhibits an upstream regulator the Ser/Thr kinase mTOR (Chou and Blenis, 1995; Proud, 1996; Pullen and Thomas, 1997; Dufner and Thomas, 1999).

One of a few important upstream regulators of p70S6K1 is PI3K. PI3K can be activated by a number of receptor tyrosine kinases as well as G protein-coupled receptors. PI3K is a heterodimer composed of a 110-kDa catalytic subunit (p110) and an 85-kDa regulatory subunit (p85). PI3K catalyzes addition of a phosphate group to the 3' position of the sugar ring in a phosphoinositide. Its products act on multiple downstream effectors that interact with Src homology-2 and pleckstrin homology domains of serine/threonine and tyrosine kinases. Some of these kinases contribute to phosphorylation and activation of p70S6K1.

We found that the majority of cardiomyocytes in culture can survive a pulse treatment with low or mild doses of H₂O₂ but become enlarged over a course of 4 to 7 days (Chen et al., 2000b). The cells respond to H₂O₂ treatment by increasing cell volume and protein content. To understand the mechanism behind this phenomenon, we determined early changes in signal transduction pathways of oxidative stress by using primary cultured neonatal rat cardiomyocytes. The dose of 200 µM was chosen for most experiments because it appeared to induce maximal increases in cell size without killing the majority of the cells (Chen et al., 2000b). We demonstrate herein that H₂O₂ activates PI3K and p70S6K1 in cardiomyocytes. Inhibitor approaches indicate a critical role of these two kinases in oxidant-induced enlargement of cell size. To evaluate the immediate pharmacological significance of these findings, we also tested whether the anthracycline quinones, represented by doxorubicin and daunorubicin, induce cardiomyocyte hypertrophy via an oxidative stress mechanism.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents Chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise indicated. Stabilized H₂O₂ (H-1009; Sigma-Aldrich) was used and the concentration of the stock was verified by absorbency at 240 nm. Wortmannin, rapamycin, doxorubicin, and daunorubicin were obtained from Calbiochem (La Jolla, CA).

Cell Culture and Treatment. Cardiomyocytes were prepared from 1- to 2-day old neonatal Sprague-Dawley rats (Harlan Bioproducts for Science, Indianapolis, IN). Briefly, the heart tissue was cut into 1- to 2-mm pieces and washed with a cocktail containing 0.02% pancreatin (Invitrogen, Carlsbad, CA) and 0.045% collagenase to remove red blood cells. The heart tissue was digested in fresh cocktail at 37°C for 15 min. The digestion was repeated 6 to 10 times. Dissociated cells were collected by centrifugation and were resuspended in Ham's F-12 medium (Invitrogen) with 1.0% (w/v) bovine serum albumin (Invitrogen), 0.025% (w/v) Fetuin, 0.1 mM ascorbate, 100 units/ml penicillin G, and 100 units/ml streptomycin. The dissociated cells were placed in uncoated 100-mm dishes and incubated at 37°C in a 5% CO₂ incubator for 45 to 60 min. This procedure allows fibroblasts to attach to the dishes, whereas most myocytes remain unattached. The population of cells enriched in myocytes by this differential plating procedure was collected and counted. The cells were seeded at a density of 2 × 10⁶ cells/100-mm dish in DMEM with 1 mM pyruvate, 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 units/ml streptomycin. One to 2 days after plating, the media were replaced with fresh DMEM containing 0.5% FBS for 48 h and the cells were treated with H₂O₂, doxorubicin, or daunorubicin in the medium. For groups treated with N-acetylcysteine (NAC) or buthionine sulfoximine (BSO), 5 mM NAC or 100 µM BSO was added to cells during the 48-h low-serum (0.5%) incubation time. NAC or BSO was removed by changing medium before H₂O₂ or doxorubicin treatment. At the time of H₂O₂, doxorubicin or daunorubicin treatment, more than 90% of the cells were capable of expressing sarcomeric myosin as determined by immunocytochemistry with an antibody against sarcomeric myosin heavy chain.

p70S6K1 Activity Assay. The protocol of Oh et al. (1998) was adopted for measuring p70S6K1 activity. After H₂O₂ exposure, myocytes were lysed on ice in 0.5 ml of p70S6K1 lysis buffer (10 mM potassium phosphates, pH 7.4, 1 mM EDTA, 5 mM EGTA, 10 mM MgCl₂, 50 mM -glycerophosphate, 0.5% Triton X-100, 1 mM Na₃VO₃, 2 mM DTT, 1 mM phenylmethylsulfonyl fluoride, and 10 µg/ml aprotinin). After a 20-min incubation on ice, the lysates were centrifuged for 5 min at 13,000 rpm. Supernatants were collected and measured for protein concentration by the Bradford method (Bio-Rad, Hercules, CA) before being used for immunoprecipitation. An equal amount (500 µg) of proteins from each sample was used for a 2-h incubation on ice with 5 µl of p70S6K1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) and for an additional 1-h incubation after the protein A-Sepharose (30 µl/reaction tube) was added. The protein A-Sepharose immunocomplexes were washed twice with the kinase lysis buffer and twice with a kinase reaction buffer (20 mM 4-morpholinepropanesulfonic acid pH 7.2, 25 mM -glycerophosphate, 5 mM EDTA, 1 mM Na₃VO₃, and 1 mM DTT) before being resuspended in 30 µl of kinase reaction buffer containing 20 µM S6 kinase substrate peptide (RRRLSSLRA; Santa Cruz Biotechnology). The kinase reaction was initiated by addition of 10 µCi of [-³²P]ATP plus unlabeled ATP to a final concentration of 40 µM and MgCl₂ to a final concentration of 10 mM. After a 10-min incubation at 30°C, 20 µl of reaction mixtures was spotted onto p-81 phosphocellulose discs (Whatman, Clifton, NJ), which were then washed twice in 0.75% phosphoric acid and twice in acetone. After being dried in air, the discs were placed in 10 ml of scintillation fluid and phosphorylated products were quantified by liquid radioactive chromatography.

Assay for p70S6K1 Phosphorylation. After H₂O₂ exposure, myocytes were harvested in a lysis buffer (1% Triton X-100, 10 mM Tris pH 7.4, 5 mM EDTA pH 8.0, 50 mM NaCl, 50 mM NaF, 10 µg/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, and 2 mM Na₃VO₄). Protein concentration was measured by the Bradford method according to the manufacturer's instruction (Bio-Rad). An equal amount of proteins was loaded in each lane and separated by 8% SDS-polyacrylamide gel electrophoresis for Western blot as previously described (Chen et al., 2000a). After transferring the proteins to a polyvinylidene difluoride membrane, the membrane was incubated for 2 h with an antibody that recognizes phosphorylated Thr 389 or phosphorylated Thr421/Ser424 of p70S6K1 (1:1000 dilution; New England Biolabs, Beverly, MA). For determining the basal level of the protein, a duplicated membrane from the same experiment was incubated with an antibody that recognizes both phosphorylated and unphosphorylated forms of p70S6K1 (1:1000 dilution; New England Biolabs). Bound antibodies were detected by enhanced chemiluminescence reaction following the incubation with a secondary antibody conjugated with the horseradish peroxidase (Chen et al., 2000a).

PI3K Activity Assay. Myocytes were harvested by scraping on ice by using PI3K lysis buffer (20 mM HEPES pH 7.4, 2 mM EGTA, 50 mM -glycerophosphate, 1% Triton X-100, 10% glycerol, 1 mM DTT, 1 mM Na₃VO₄, 2 µM leupeptin, 10 unit/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride). Supernatants were collected for protein concentration determination as described above. Samples containing 500 µg of protein were incubated with 5 µl of anti-p85 antibody (Upstate Biotechnology, Lake Placid, NY) for 2 h on ice and for additional 1 h after addition of 40 µl of protein A-Sepharose. The kinase reaction was carried out at 30°C for 10 min in a 50-µl reaction mixture containing the immunoprecipitates, 0.2 mg/ml phosphatidylinositol, 10 µCi of [-³²P]ATP, and the kinase reaction buffer (20 mM Tris-HCl pH 7.4, 50 mM NaCl, 10 mM MgCl₂, 0.5 mM EGTA, 120 µM adenosine, and 50 µM ATP). The reaction was stopped by the addition of 100 µl of 1 M HCl. Phospholipids were extracted immediately using 200 µl of a chloroform and methanol mixture (1:1 volume ratio). An equal volume of organic phase from each sample was spotted onto a silica gel 60 plate (EM Separations Technology, Darmstadt, Germany) for thin layer chromatography by using a solvent containing chloroform, methanol, 25% ammonia hydroxide, and water (43:38:5:7 volume ratio). The results were obtained by autoradiography. An Instant Imager (Packard Instrument Co., Meriden, CT) was used to quantify the radioactivities in the product phosphatidylinositol 3-phosphate. The inhibitory effect of wortmannin was verified in vitro by incubating the immunoprecipitates with wortmannin for 30 min at 30°C before the kinase reaction.

Measurement of Phosphorylation of p85 Subunit of PI3K. Cells were harvested as described above using PI3K lysis buffer. For each sample, 500 µg of proteins was incubated with either anti-phosphotyrosine antibody or anti-p85 antibody (Upstate Biotechnology) for immunoprecipitation. Proteins were dissociated from the Sepharose by boiling and were separated by 8% SDS-polyacrylamide gel electrophoresis for Western blot as described (Chen et al., 2000a). The polyvinylidene difluoride membrane was incubated with an anti-p85 antibody and the bound antibody was detected by enhanced chemiluminescence reaction after incubation with a secondary antibody conjugated with horseradish peroxidase as described (Chen et al., 2000a).

Measurements of Cell Enlargement. Cell volume and protein content per cell were determined as previously described (Chen et al., 2000b). Briefly, the cells were allowed to recover for 5 days after H₂O₂, doxorubicin, or daunorubicin treatment. For the NAC-treated group, cells were incubated with 2.5 mM NAC during the 5-day recovery period. The adherent cells in six-well culture plates were detached by trypsin treatment and rounded cells were loaded onto a microslide field finder (Fisher Scientific, Pittsburgh, PA) for the measurement of cell diameters, which were used to calculate cell volume by using the equation of 4/3xxRadius³. Protein concentrations were measured by bicinchoninic acid method according to the manufacturer's instruction (Pierce Chemical, Rockford, IL). Protein content per cell was determined by dividing the total amount of protein in each well with the cell number, which was determined by a Coulter Counter or hemocytometer after trypsin treatment.

Statistics. One-way analysis of variance (ANOVA) was used to compare groups of means followed by the Student-Newman-Keuls method for multiple comparisons. Groups of means that are not significantly different from each other are indicated in the figure by a common letter symbol. Any mean with a letter designation different from others is significantly different from the others. When labeled with letters "ab", the mean is not significantly different from those labeled with a or b, although the mean with the letter designation a is significantly different from that with the letter b.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of p70S6K1 in Cardiomyocytes by H₂O₂ Treatment. Activation of p70S6K1 is an important event leading to cardiomyocyte hypertrophy induced by a variety types of inducers (Sadoshima and Izumo, 1995; Boluyt et al., 1997; Laser et al., 1998; Simm et al., 1998). We determined whether H₂O₂ activated p70S6K1 by measuring its activity with cell lysates harvested after H₂O₂ treatment. Time course studies indicated a significant increase in p70S6K1 activity after 30 min of 200 µM H₂O₂ treatment (Fig. 1A). The activity reached a plateau of 2-fold after 60 min and the elevation level remained at 120 min (Fig. 1A). The activity of p70S6K1 returned to the basal level 24 h after H₂O₂ treatment (data not shown). When cells were treated with various concentrations of H₂O₂ for 60 min, activation of p70S6K1 was detectable with 50 µM H₂O₂ and reached the highest level of 2.2-fold with 150 µM H₂O₂ (Fig. 1B). H₂O₂ at 100 or 200 µM concentration induced similar levels of activation as H₂O₂ at 150 µM (Fig. 1B).

View larger version (42K): [in this window] [in a new window]   Fig. 1.   H2O2-induced p70S6K1 (p70 S6 kinase) activity. Cardiomyocytes were treated with 200 µM H2O2 for the indicated time (A) or were treated for 60 min with various doses of H2O2 (B). The data are expressed as means ± standard deviations of percentage of increase in the activity from three independent experiments. Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test.

View larger version (48K): [in this window] [in a new window]   Fig. 2.   H2O2-induced phosphorylation of Thr389 and Thr421/Ser424 of p70S6K1 (p70 S6 kinase) . Cardiomyocytes were treated with 200 µM H2O2 for the indicated time (A) or were treated for 60 min with various doses of H2O2 (B). The cells were harvested for Western blots to detect phosphorylation or protein levels. The data are presented as the image from one experiment representative of three.

H₂O₂ Induces PI3K Activation in Cardiomyocytes. An important upstream regulator of p70S6K1 is PI3K (Chou and Blenis, 1995; Proud, 1996; Dufner and Thomas, 1999). H₂O₂ increases p70S6K1 activity and phosphorylation, suggesting the possibility that H₂O₂ may activate PI3K. To test whether H₂O₂ activates PI3K in cardiomyocytes, we measured PI3K activity by an in vitro kinase assay after isolating the PI3K complex by using an antibody against the p85 subunit (Oh et al., 1998). Cardiomyocytes were treated with 200 µM H₂O₂ for 5 to 60 min. An increase in the kinase activity was detected at 5 min and reached a plateau at 10 min but remained elevated for 60 min (Fig. 3). One experiment indicated that the kinase activity at 120 min of H₂O₂ exposure was comparable to that at 60 min but diminished 24 h post-H₂O₂ exposure (data not shown). The dose-response studies of 10-min H₂O₂ exposure indicated that the dose of 150 µM caused the highest activation (Fig. 4A). To ensure that an equal amount of PI3K protein in each sample was used for the in vitro kinase activity assay, the level of p85 protein was measured by Western blot (Figs. 3 and 4A, middle panels). The results showed that H₂O₂ induced PI3K activation without changing p85 protein level.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Time course of H2O2-induced PI3K (PI 3-Kinase) activation. Cardiomyocytes treated with 200 µM H2O2 for the indicated time were harvested for in vitro kinase activity assay. The 32P-labeled lipid product (PI3-P) was separated by thin layer chromatography from residual [32P]ATP (loading) that remained in the aqueous phase and served as a loading control (top). Aliquots of samples containing 20 µg of protein were used for Western blots to determine the protein level of the p85 subunit (middle). The data are from one experiment representative of three (top and middle) or are means ± standard deviation from three independent experiments (bottom). Means with a given letter designation are significantly different (p < 0.05) from other means with different letters determined by analysis of variance followed by multiple comparisons with the Student-Newman-Keuls test.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Dose-response of H2O2-induced PI3K activation (A), inhibitory effect of wortmannin (WTN, A), and tyrosine phosphorylation of p85 subunit (B). Cardiomyocytes treated with H2O2 at various concentrations for 10 min were harvested for in vitro PI3K (PI 3-kinase) activity assay (A). The 32P-labeled lipid product (PI3-P) was separated by thin layer chromatography from residual [32P]ATP (loading) that remained in the aqueous phase and served as a loading control (A, top). Aliquots of samples containing 20 µg of protein were used for Western blots to determine the protein level of the p85 subunit (A, middle). Cell lysates from 10-min, 200 µM H2O2-treated cardiomyocytes were used for p85 tyrosine phosphorylation assay via Western blots with an anti-p85 antibody after immunoprecipitation with an anti-phosphotyrosine antibody (B, top) or an anti-p85 antibody (B, bottom). The data presented are either from one experiment representative of three (top and middle, A and B) or are means ± standard deviation from three independent experiments (A, bottom). Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test (A).

Relationship between p70S6K1 and PI3K Activation. Our study reveals that H₂O₂ induces activation of p70S6K1 and PI3K in cardiomyocytes. The immediate question is whether PI3K is the upstream regulator of p70S6K1 under oxidative stress. A pharmacological inhibitor specific for PI3K (i.e., wortmannin) allows us to address this question. Rapamycin, an inhibitor of the key Ser/Thr kinase mTOR that regulates the activity of p70S6K1 and other cellular proteins, serves as a positive control for p70S6K1 inhibition. Induction of oxidative stress by H₂O₂ results in depletion of glutathione, which is the main source of cellular nonprotein sulfhydryls. To eliminate the possibility that wortmannin and rapamycin function as antioxidants to prevent oxidative stress, we measured cellular nonprotein sulfhydryl content by using Ellman's reagent (Chen and Stevens, 1991). Cells were pretreated 30 min with the inhibitors and were treated with H₂O₂ in the presence of the inhibitors. We found that a 2-h treatment of H₂O₂ at 200 µM or lower failed to cause a significant loss of nonprotein sulfhydryls in primary cultured rat neonatal cardiomyocytes or in the rat cardiomyocyte H9C2 cell line. Failure to detect loss of sulfhydryls is probably a result of limited sensitivity of the assay. In one representative experiment, H₂O₂ at 500 µM reduced nonprotein sulfhydryls by 24.5% at the end of 2-h treatment time in H9C2 cells. In the presence of 100 nM wortmannin or 10 ng/ml rapamycin, H₂O₂ reduced nonprotein sulfhydryls by 35 or 31.4%, respectively. These data indicate that the inhibitors at the concentrations used were unlikely to prevent H₂O₂ from inducing loss of nonprotein sulfhydryls.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effect of wortmannin (WTN) and rapamycin (RPM) on H2O2-induced p70S6K1 (p70 S6 Kinase activity) phosphorylation (A and B) and activation (C). Cardiomyocytes were pretreated 30 min with WTN (A) or RPM (B) at indicated concentrations. The cells were then treated with 200 µM H2O2 for 60 min in the presence of the inhibitors. Phosphorylation on Thr389 or Thr421/Ser424 and the protein level of p70S6K1 were determined by Western blot (A and B). Lysates harvested from cells treated with H2O2 in the absence or presence of 100 nM WTN or 10 ng/ml RPM were used for immunoprecipitation with anti-p70S6K1 antibody and in vitro p70S6K1 assay (C). The data are means ± standard deviation from three independent experiments (C). Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test (A).

Effect of p70S6K1 and PI3K on H₂O₂-Induced Cell Enlargement. Cells surviving H₂O₂ treatment become enlarged over the course of 5 days (Chen et al., 2000b). As in our previous studies, cardiomyocytes were cultured in a medium containing 10% FBS and the cells were maintained in the medium containing 10% FBS during and after H₂O₂ treatment. Differing from the previous studies, herein we cultured cardiomyocytes in a medium containing 0.5% FBS for 48 h before H₂O₂ treatment. This low-serum medium reduces the background for various kinase assays and prevents the complication of serum effects on cardiomyocyte hypertrophy. Cells were treated with 200 µM H₂O₂ for 90 min. H₂O₂ and oxidized medium were removed by placing cells in fresh DMEM containing 0.5% FBS. The cells were allowed to recover for 5 days in the low-serum medium before examination of cell surface areas, cell volumes, and the protein content per cell. To test the effect of wortmannin and rapamycin on H₂O₂-induced hypertrophy, cells were pretreated 30 min with these inhibitors and then treated with H₂O₂ in the presence of these inhibitors. During the 5-day recovery period, inhibitors were added back into the medium. Examinations of cell morphology, cell volumes, and the protein content per cell showed that wortmannin and rapamycin abolished cell enlargement (Fig. 6, A-C). These results suggest a role of PI3K and p70S6K1 in H₂O₂-induced cell enlargement.

View larger version (45K): [in this window] [in a new window]   Fig. 6.   Effect of wortmannin (WTN) or rapamycin (RPM) on H2O2-induced cell enlargement. Cardiomyocytes in DMEM containing 0.5% FBS were pretreated with 100 nM WTN or 10 ng/ml RPM for 30 min. The cells were treated with 200 µM H2O2 for 90 min in the presence of the inhibitors and were then placed in fresh medium containing 0.5% FBS and the inhibitors. The medium was changed every 1 to 2 days with freshly added inhibitors. Cells at 5 days after H2O2 treatment were used for morphology analysis (A), measurements of cell volumes (B), and protein content per cell (C). A digital camera attached to a phase-contrast microscope with 20× lens was used to acquire the images (A). At least 99 cells were measured for diameters randomly (B) and three groups of samples were measured for protein content per cell (C). The data are means ± standard deviations of 99 cells (B) or triplicates (C) from one experiment representative of three. Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test (A).

Induction of Cardiomyocyte Enlargement by Doxorubicin and Daunorubicin. The anthracycline quinones represented by doxorubicin and daunorubicin are frequently used drugs for several types of cancer. The use of these drugs is limited because of their side effect of inducing cardiac toxicity (Singal and Iliskovic, 1998; Singal et al., 2000). Although the acute toxicity, including various arrhythmias is clinically manageable, chronic heart failure develops weeks to years after the drug administration and is a major concern. Considerable evidence supports that oxidative stress mediates cardiac toxicity induced by anthracyclines (Singal and Iliskovic, 1998; Singal et al., 2000). The chemical structures of doxorubicin and daunorubicin indicate their ability to undergo redox cycling. Doxorubicin accepts electrons from oxoreductive enzymes to form semiquinone free radicals, which may initiate a chain of redox reactions. Doxorubicin has been shown to produce superoxide and H₂O₂ with the mitochondrial fraction of heart extracts (Doroshow and Davies, 1986). Daunorubicin, a structural analog of doxorubicin, produces less reactive oxygen species (Doroshow and Davies, 1986). We tested whether doxorubicin or daunorubicin induced cardiomyocyte hypertrophy by treating primary cultured neonatal rat cardiomyocytes for 6 h with the drugs at 1, 10 or 100 nM. After the treatment, the cells were allowed to recover for 5 days in the medium containing 0.5% FBS before measurements of cell volume and protein content per cell. The results showed that doxorubicin and daunorubicin at 10 or 100 nM can both induce cell enlargement as measured by cell volume and the protein content per cell (Fig. 7, A and B). A greater degree of cell volume increase or protein content increase was observed with doxorubicin in comparison with daunorubicin (Fig. 7, A and B).

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Induction of cell enlargement by doxorubicin (Dox) or daunorubicin (Dau). Cardiomyocytes in DMEM containing 0.5% FBS were treated with 1, 10, or 100 nM Dox or Dau for 6 h. The cells were allowed to recover for 5 days before measurement of cell volume (top) or protein content per cell (bottom). At least 100 cells were measured for diameter randomly and three groups of samples were measured for protein content per cell. The data are means ± standard deviations of 100 cells (top) or triplicates (bottom) from one experiment representative of two. Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Effect of BSO and NAC on cell enlargement induced by doxorubicin (Dox) or H2O2. Cardiomyocytes in DMEM containing 0.5% FBS were pretreated with 0.1 mM BSO or 5 mM NAC for 48 h before treatment with 1, 10, or 100 nM Dox for 6 h (A) or 50, 100, 150, or 200 µM H2O2 for 90 min (B). The cells were allowed to recover for 5 days and 2.5 mM NAC was added back to the cells. The data are means ± standard deviations of cell volumes measured randomly from 100 cells (top panels) or of triplicate measurements of protein content per cell from one experiment representative of two (bottom panels). Means with a given letter designation are significantly different (p < 0.05) from other means with different letters as determined by ANOVA followed by multiple comparisons with the Student-Newman-Keuls test.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our study shows that H₂O₂ can activate p70S6K1 and PI3K in cardiomyocytes. PI3K was activated within 5 min of 200 µM H₂O₂ treatment, whereas p70S6K1 was not activated until 30 min of H₂O₂ exposure. These changes preceded cell enlargement. Inhibiting PI3K with wortmannin blocked p70S6K1 activation. Wortmannin and rapamycin prevented H₂O₂ from inducing cell enlargement. These data suggest that PI3K-dependent activation of p70S6K1 plays an important role in cell size increases induced by oxidants.

Many of our experiments use H₂O₂ at the dose of 200 µM. This dose may appear to be higher than the steady-state concentration of H₂O₂ observed under pathophysiological conditions. With ischemia, cumulative doxorubicin treatment or chronic heart failure, it seems that the myocardium is exposed to oxidants at low doses chronically. Evidence supports the fact that the myocardium can accumulate oxidative damage with repeated exposure to low-dose oxidants. For example, cardiac toxicity of doxorubicin is usually determined by cumulative doses above 550 mg/m² of body surface area regardless of whether the drug is administrated at a high dose once (slowly) or at low doses repetitively. Based on this phenomenon, one may postulate that chronic exposure to oxidants above certain cumulative doses may ultimately result in damage similar to that induced by 200 µM H₂O₂. However, this postulation demands for experimental evidence. In human fibroblasts, two treatments with 75 µM H₂O₂ result in prolonged cellular and molecular changes similar to that induced by one treatment with 150 µM (Chen et al., 2001). Although one treatment with 200 µM H₂O₂ is convenient from the experimental prospective, we need to test whether repetitive exposures to H₂O₂ at the dose similar to that observed under pathophysiological conditions, for example 6 or 11 µM, cause enlargement of cardiomyocytes. Most importantly, whether PI3K and p70S6K1 play a role in low-dose oxidant effects in cardiomyocytes remains to be determined.

An important caveat of the present study is the experimental system using neonatal rat cardiomyocytes. Cardiomyocytes ex vivo enable investigation of the signal transduction pathways and molecular changes induced by oxidants. However, chronic heart failure or cardiac hypertrophy is a disease that most likely develops in the late stage of adulthood. Ideally, cardiomyocytes from adult animals should be used to study the mechanism of oxidative stress or cardiomyocyte hypertrophy. The differences between neonatal rat cardiomyocytes and adult rat cardiomyocytes include the rate of DNA synthesis and differentiation status. Early reports reveal that about 13% of neonatal cardiomyocytes undergo DNA synthesis, whereas the DNA synthesis figure is much lower (0.2-2%) in adult hearts (Anversa and Kajstura, 1998). Adult cardiomyocytes are fully differentiated and form myofibers (Mitcheson et al., 1998). Technically, it has been difficult to isolate a large quantity of cardiomyocytes from adult animals and to maintain their viability ex vivo over a long time course when cellular and molecular changes can be defined. In contrast, isolation of neonatal rat cardiomyocytes is relatively easy and the cells can be maintained in culture for a long time. The large quantity of homogeneous neonatal rat cardiomyocytes allows us to define cellular and molecular changes induced by oxidants. Because of the difference between neonatal cardiomyocytes and adult cardiomyocytes, caution should be exercised when extrapolating the results from neonatal cardiomyocytes to cases of adult cardiomyocyte hypertrophy.

Activation of p70S6K1 is a well known mitogenic event in tumor cells (Chou and Blenis, 1995). Oxidants have been shown to induce mitogenic responses and activate p70S6K1 in certain tumor cell lines (Bae et al., 1999). Differing from tumor cells, cardiomyocytes appear to grow in size instead of cell number in response to oxidant stimulation (Chen et al., 2000b). Our data are consistent with that of others, showing that activation of p70S6K1 plays an important role in cardiomyocyte hypertrophy. Furthermore, we have found that activation of p70S6K1 by H₂O₂ involves PI3K. PI3K-dependent activation of p70S6K1 has been reported with growth factors and endocrine factors (Chou and Blenis, 1995; Proud, 1996; Pullen and Thomas, 1997; Dufner and Thomas, 1999). With these factors, phosphorylation of Thr421/Ser424 appears to be necessary but not sufficient for p70S6K1 activation. In contrast, phosphorylation of Thr389 appears to correlate with increases in p70S6K1 activity. In our experimental system, wortmannin and rapamycin can completely block Thr389 phosphorylation and p70S6K1 activity. These data suggest a similarity between growth factors and oxidants in regulating p70S6K1 activity by Thr389 phosphorylation.

A few recent studies report an important role of p70S6K1 in cell size regulation. In the fruit fly and mouse, knocking out the p70S6K1 gene results in a delay in the development and a smaller body size due to a reduction in cell sizes (Shima et al., 1998; Montagne et al., 1999). In the heart, p70S6K1 can be activated by a variety of hypertrophy inducers such as angiotensin II, - or -adrenergic receptor agonists, mechanical stretch in vitro, and pressure overload in vivo (Sadoshima and Izumo, 1995; Boluyt et al., 1997; Laser et al., 1998; Simm et al., 1998). Rapamycin is capable of inhibiting the development of cardiomyocyte hypertrophy in these experimental models. Because p70S6K1 is thought to regulate protein translation, aberrant activation of protein translation may be a key factor for cell size increases.

One of the important findings from this study is the activation of PI3K by H₂O₂ in cardiomyocytes. Although this phenomenon is novel in cardiomyocytes, oxidants have been shown to activate PI3K in other cell types. PI3K activation is usually a mitogenic response triggered by growth factor receptors or G proteins (Fruman et al., 1998). The p85 subunit of PI3K is an adapter protein that can be phosphorylated by receptor tyrosine kinases and Src family tyrosine kinases. It has been shown that oxidants activate Src family kinases in cardiomyocytes (Aikawa et al., 1997). This may explain the observed tyrosine phosphorylation of p85 in H₂O₂-treated cardiomyocytes. It is thought that phosphorylation of the p85 subunit contributes to the relocation of p85/p110 complex from the cytosol to the plasma membrane where PI3K can contact its activator ras and its substrate inositol-containing lipids (Fruman et al., 1998). Deora et al. (1998) demonstrated that changes in redox status led to recruitment of the p85 subunit to the plasma membrane where it associated with ras protein and the p110 catalytic subunit of PI3K. It has been reported that oxidants activate the small G protein ras (Hardwick and Sefton, 1997). Taken together, it seems that src and ras are two potential upstream factors regulating PI3K activation by oxidants. Although activated src can phosphorylate p85 and cause p85 translocation to the plasma membrane, activated G proteins may contribute to changes in PI3K activities.

The signal transduction pathways of cardiomyocyte hypertrophy have attracted a great deal of attention because of the potential to develop novel pharmacological interventions against heart failure by targeting early signaling events of cardiomyocyte hypertrophy. The discovery of PI3K activation by H₂O₂ in cardiomyocytes is interesting and deserves further investigation. Whether PI3K plays a role in cardiomyocyte hypertrophy under pathophysiological conditions in vivo is not clear. A recent study using transgenic approaches shows that PI3K activity determines the size of cardiomyocytes and the size of the heart (Shioi et al., 2000). Transgenic mice with constitutively activated PI3K develop bigger hearts and bigger cardiomyocytes compared with normal littermates. In contrast, inactivating PI3K with dominant negative mutant genes results in smaller hearts and smaller cardiomyocytes (Shioi et al., 2000). These data suggest the possibility that PI3K plays an important role in cardiomyocyte hypertrophy in vivo.

Chronic heart failure is often a clinically intractable disorder and lacks effective therapeutic treatment. An important morphological feature of chronic heart failure is cardiomyocyte hypertrophy. Identifying the inducers of cardiomyocyte hypertrophy and the intracellular signaling cascade of the inducers is necessary for mechanism-based drug design. This study and others suggest that oxidative stress plays a role in cardiomyocyte hypertrophy in vitro. Oxidative stress has been viewed as an important factor contributing to chronic heart failure, which often involves multiple pathophysiological changes. Hypertension and increased activity of the renin-angiotensin system can trigger cardiac hypertrophy and ultimately heart failure. Increased levels of angiotensin II are known to cause cardiac hypertrophy in vitro and in vivo. In vascular smooth muscle cells, oxidants appear to mediate angiotensin II-induced signaling changes of hypertrophy (Griendling et al., 2000). Because angiotensin II activates a plasma membrane-associated NAD(P)H oxidase (Griendling and Ushio-Fukai, 2000), it is anticipated that angiotensin II may also produce oxidants in cardiomyocytes. Although further experiments will determine whether oxidative stress plays a role in cardiac hypertrophy and chronic heart failure induced by ischemia, ischemic reperfusion, or pressure overload in vivo, several pieces of experimental evidence have already demonstrated that doxorubicin induces cardiomyocyte hypertrophy in vivo and oxidative stress mediates cardiac hypertrophy induced by doxorubicin (Kang et al., 1996; Sun et al., 2001). These findings suggest that antioxidants may serve as useful therapeutic agents against heart failure induced by doxorubicin and perhaps other types of cardiovascular disease. The fact that oxidants activate PI3K and p70S6K1 may provide alternatives in developing pharmacological agents against cardiomyocyte hypertrophy and heart failure associated with oxidative stress.