Anthracycline-containing chemotherapy causes long-term impairment of mitochondrial respiration and increased reactive oxygen species release in skeletal muscle

Abstract

Anticancer treatments for childhood acute lymphoblastic leukaemia (ALL) are highly effective but are now implicated in causing impaired muscle function in long-term survivors. However, no comprehensive assessment of skeletal muscle mitochondrial functions in long-term survivors has been performed and the presence of persistent chemotherapy-induced skeletal muscle mitochondrial dysfunction remains a strong possibility. Non-tumour-bearing mice were treated with two drugs that have been used frequently in ALL treatment (doxorubicin and dexamethasone) for up to 4 cycles at 3-week intervals and euthanized 3 months after the 4th cycle. Treated animals had impaired growth and lower muscle mass as well as reduced mitochondrial respiration and increased reactive oxygen species production per unit oxygen consumption. Mitochondrial DNA content and protein levels of key mitochondrial membrane proteins and markers of mitochondrial biogenesis were unchanged, but protein levels of Parkin were reduced. This suggests a novel pattern of chemotherapy-induced mitochondrial dysfunction in skeletal muscle that persists because of an acquired defect in mitophagy signaling. The results could explain the observed functional impairments in adult survivors of childhood ALL and may also be relevant to long-term survivors of other cancers treated with similar regimes.

Figure 1. Chemotherapy treatment causes sustained impairment of whole animal and muscle growth.

(A) Changes in body weight over time in control vs. chemotherapy treated animals. Arrows indicate timings of each cycle of chemotherapy. (B) Representative photographs of control (left) and treated (right) mice taken at the end of our long-term chemotherapy protocol. Gastrocnemius muscle weights obtained in control and treated animals in the early (C) and late (D) chemotherapy treatment groups. (E) Representative triple MHC labeling (MHCI: blue; MHCIIa: red; MHCIIx: purple; MHCIIb: green) performed on gastrocnemius muscle serial cross sections obtained in control (images on the left) and treated (images on the right) mice in the late group. Scale bars: 50 μm. (F) Quantification of gastrocnemius muscle fiber size in control and treated mice. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01.

Figure 2. Chemotherapy treatment causes generalized chronic impairment in skeletal muscle mitochondrial function.

(A, B) Mitochondrial respiration was assessed in permeabilized myofibers prepared from the red gastrocnemius muscle regions in control and treated mice in early (A) and late (B) groups. Mitochondrial substrates and inhibitors were subsequently added as follows: G + M, ADP, succinate (Succ), AA, and ascorbate + TMPD (TMPD). (C, D) Acceptor control ratio (ACR), a parameter indicative of mitochondrial coupling efficiency, was determined in control and treated mice in early (C) and late (D) groups. ACRs values were obtained by dividing the respiration rate with G + M + ADP (ADP, state 3) by the respiration rate with G + M (state 2). (E, F) Mitochondrial H₂O₂ production was assessed in permeabilized myofibers prepared from the red gastrocnemius muscle regions of control and treated mice in early (E) and late (F) groups. Mitochondrial substrates and inhibitors were subsequently added as follows: G + M, succinate (Succ), 0.1 mM ADP, 1 mM ADP, and AA. (G, H) Mitochondrial free radical leak was determined by dividing H₂O₂ emission rates by their corresponding respiration rates (data from panel A and B). Data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Figure 3. Chemotherapy treatment-induced chronic intrinsic mitochondrial functional changes in skeletal muscle are not mediated by increased mtDNA mutations or reduced mitochondrial content.

(A, B) Quantification of citrate synthase activity performed in gastrocnemius muscle of control and treated mice in early (A) and late (B) groups. (C) Quantification of the mitochondrial DNA copy number obtained in the gastrocnemius muscle of control and treated mice in the late group. (D) Left panel: representative immunoblot for complex I (CI), complex II (CII), complex III (CIII), complex IV (CIV) and the ATP synthase (ATPs) performed on gastrocnemius samples obtained from control and treated mice in the late group. Right panel: the corresponding Ponceau stain was used to ensure equal protein loading between lanes. E) Quantification of complex I, II, III, IV and ATPs contents in the gastrocnemius muscles obtained from control and treated mice in the late group. (F, G). Quantification of the relative PGC-1α mRNA expression in the gastrocnemius muscles obtained from control and treated mice in early (F) and late (G) groups. (H) Quantification of the relative PGC-1α mRNA expression in the gastrocnemius muscles obtained from control and treated mice in the late group. (I) Representative sequential complex IV and complex II stain performed in gastrocnemius cross-sections of control (left) and treated (right) mice in the late group. Note that any fibers with high accumulation of mitochondrial DNA mutations should appear blue. Note that out of the 10 treated mice analyzed in the late group, only one showed 2 blue fibers. (J) Long-range PCR amplification of a fragment of the major arc of the mitochondrial DNA, very often affected by mtDNA deletions in pathological conditions, performed on samples from the plantaris of control and treated mice in the late group. Data in graphs are presented as means ± SEM. *P < 0.05.

Figure 4. Altered mitochondrial quality control mechanisms rather than increased cellular oxidative damage are implicated in chemotherapy-induced chronic intrinsic mitochondrial dysfunction.

(A) Immunoblot for 4-Hydroxynonenal (HNE), a marker of lipid peroxidation used to assess oxidative stress, performed on gastrocnemius muscle samples obtained from control (1) and treated (2) mice in the late group. (B) Corresponding Ponceau stain that was used to ensure equal protein loading between lanes. (C) Quantification of the HNE content in gastrocnemius muscle samples obtained from control and treated mice in the late group. (D) Representative Parkin, VDAC and β-tubulin immunoblots performed on gastrocnemius muscle samples obtained from control and treated mice in the late group. (E, F) Quantification of the protein content of Parkin (E) and VDAC (F), normalized to β-tubulin protein content, in gastrocnemius muscle samples obtained from control and treated mice in the late group. (G) Quantification of the Parkin over VDAC content in gastrocnemius muscle samples obtained from control and treated mice in the late group. Data are presented as means ± SEM. *P < 0.05; **P < 0.01.