Doxorubicin-induced carbonylation and degradation of cardiac myosin binding protein C promote cardiotoxicity

Abstract

Dose-dependent oxidative stress by the anthracycline doxorubicin (Dox) and other chemotherapeutic agents causes irreversible cardiac damage, restricting their clinical effectiveness. We hypothesized that the resultant protein oxidation could be monitored and correlated with physiological functional impairment. We focused on protein carbonylation as an indicator of severe oxidative damage because it is irreversible and results in proteasomal degradation. We identified and investigated a specific high-molecular weight cardiac protein that showed a significant increase in carbonylation under Dox-induced cardiotoxic conditions in a spontaneously hypertensive rat model. We confirmed carbonylation and degradation of this protein under oxidative stress and prevention of such effect in the presence of the iron chelator dexrazoxane. Using MS, the Dox-induced carbonylated protein was identified as the 140-kDa cardiac myosin binding protein C (MyBPC). We confirmed the carbonylation and degradation of MyBPC using HL-1 cardiomyocytes and a purified recombinant untagged cardiac MyBPC under metal-catalyzed oxidative stress conditions. The carbonylation and degradation of MyBPC were time- and drug concentration-dependent. We demonstrated that carbonylated MyBPC undergoes proteasome-mediated degradation under Dox-induced oxidative stress. Cosedimentation, immunoprecipitation, and actin binding assays were used to study the functional consequences of carbonylated MyBPC. Carbonylation of MyBPC showed significant functional impairment associated with its actin binding properties. The dissociation constant of carbonylated recombinant MyBPC for actin was 7.35 ± 1.9 μM compared with 2.7 ± 0.6 μM for native MyBPC. Overall, our findings indicate that MyBPC carbonylation serves as a critical determinant of cardiotoxicity and could serve as a mechanistic indicator for Dox-induced cardiotoxicity.

Keywords: ROS; cancer; cardioprotection.

Fig. 1.

Dox-induced protein carbonylation in cardiac tissue. (A) Coomassie blue staining of 10% (wt/vol) polyacrylamide Tris-glycine gel for cardiac tissue samples under reducing conditions. Lane 1, protein marker; lane 2, saline-treated sample without 2,4-dinitrophenyl hydrazine (DNPH); lane 3, saline-treated sample with DNPH; lane 4, Dox-treated sample with DNPH. The arrow indicates the specific protein that shows significant Dox-induced carbonylation. (B) Western blot analysis using anti-DNP antibody. Lanes 1 through 4 correspond to the same order as in A. The arrow indicates the specific protein that shows Dox-induced carbonylation. (C) Total protein carbonylation normalized to total protein. (D) Quantification of specific 140-kDa protein carbonylation from B normalized to total protein from A. (E–G) Dual-color Western blot using goat anti-DNP antibody and rabbit anti-MyBPC antibody. Donkey anti-goat 800CW (green) and donkey anti-rabbit 680LT (red) were used as secondary antibodies. Lanes 1–4 correspond to the same order as in A and B. An Odyssey scan at channel 800 (E), an Odyssey scan at channel 700 (F), and overlap of channels 700 and 800 (G) are shown (n = 5; *P < 0.05).

Fig. 2.

Dox-induced carbonylation and degradation of MyBPC in cardiac tissue. (A) Representative Western blot of duplicate samples after immunoprecipitation (IP) of MyBPC from saline- and Dox-treated cardiac tissue. After immunoprecipitation, carbonyls in MyBPC were derivatized with DNPH and DNP-derivatized MyBPC was detected using goat anti-DNP antibody. WB, Western blot. (B) Quantification of MyBPC-specific carbonyl from A normalized to total immunoprecipitated MyBPC for each treatment. (C) Representative Western blot of duplicate samples from saline and Dox-treated cardiac tissue using anti-MyBPC antibody. (D) Quantification of relative MyBPC protein level normalized to tubulin from C (n = 4; *P < 0.05).

Fig. 3.

Dox-induced carbonylation of MyBPC in HL-1 cells. Immunoprecipitated MyBPC was derivatized with DNPH and detected using goat anti-DNP antibody in immunoblots. Representative Western blots of immunoprecipitated duplicate samples of HL-1 cells are shown after treatment with 200 nM Dox for 24 h (A) and 500 nM Dox for 6 h (C). Cont, control. (E) Representative Western blot of immunoprecipitated MyBPC after treatment of HL-1 cells with vehicle (DMSO), 500 nM Dox, or 500 nM Dox and 50 μM dexrazoxane (Dex) for 24 h. B, D, and F are quantitative representations of the relative carbonylation of MyBPC from A, C, and E, respectively, normalized to total immunoprecipitated MyBPC for the corresponding treatment. The error bars represent ±SD from three separate experiments (*P < 0.05 relative to control; **P < 0.05 relative to Dox alone).

Fig. 4.

Dox-induced cytotoxicity and degradation of MyBPC in HL-1 cells. Representative Western blots are shown after treatment of HL-1 cells with 200 nM to 1 μM Dox for 2–24 h to determine caspase-3 cleavage (A) and the level of MyBPC (B). Tubulin was used as a loading control. (C) Quantification of MyBPC level from B normalized to tubulin. (D) Representative Western blot of HL-1 cells treated with 500 nM Dox alone and in combination with 500 nM MG-132 or 50 μM Dex for 24 h to determine the MyBPC protein level using anti-MyBPC antibody. (E) Quantification of relative MyBPC level from D normalized to tubulin. All values are relative to the control normalized to 100. The error bars represent ±SD from four separate experiments (*P < 0.05 relative to control; **P < 0.05 relative to Dox alone).

Fig. 5.

Metal-catalyzed carbonylation and degradation of recombinant rat MyBPC. (A) Purified recombinant cardiac MyBPC after chromatographic purification. Lane 1, protein marker; lanes 2–7, fractions from gel filtration chromatography. (B) Representative Western blots of recombinant MyBPC and carbonyl levels with or without induction of oxidative stress using the indicated concentration of hydrogen peroxide, ferrous sulfate, or ascorbate for 1 h on ice. Relative carbonyl is the representation of total MyBPC-specific carbonyl divided by total MyBPC for corresponding treatments. (C) Quantification of relative carbonyl level for each treatment divided by corresponding total MyBPC. Asc, ascorbate. (D) Relative average changes in carbonylation and degradation of recombinant MyBPC with increasing oxidative stress conditions. All values are relative to the control (unoxidized) MyBPC normalized to 1. The error bars represent ±SD from three separate experiments.

Fig. 6.

Actin binding assay using MyBPC and oxidized MyBPC. (A) Representative Western blot to determine actin binding to immunoprecipitated MyBPC from saline- and Dox-treated cardiac tissue. Lane 1, cardiac tissue lysate without immunoprecipitation; lane 2, protein A/G PLUS agarose beads (Santa Cruz Biotechnology, Inc.) with antibody; lane 3, saline-treated cardiac sample after immunoprecipitation; lane 4, Dox-treated cardiac sample after immunoprecipitation. The level of actin was detected using anti-actin antibody in immunoprecipitated samples. MyBPC-bound actin in the immunoprecipitated MyBPC was normalized to total MyBPC in the corresponding saline- and Dox-treated samples. (B) Representative Coomassie blue staining gel for supernatant and pellets after an ultracentrifugation experiment for an actin binding assay using recombinant native and oxidized (carbonylated) MyBPC. Lanes P1 to P5 represent the pellets after ultracentrifugation of 5 μM actin (P1), 4 μM recombinant MyBPC (P2), 4 μM recombinant oxidized MyBPC (P3), 4 μM recombinant MyBPC + 5 μM actin (P4), and 4 μM recombinant oxidized MyBPC + 5 μM actin (P5), respectively. Lanes S1 to S5 represent the supernatant of the corresponding samples. The binding curves for the dissociation constants of MyBPC (C) and oxidized MyBPC (D; MyBPC-ox) with actin are shown. mol, mole.

Fig. 7.

Schematic representation of cardiotoxicity through Dox-mediated metal-catalyzed oxidation (MCO) and MyBPC carbonylation.