Mitochondrial vulnerability and increased susceptibility to nutrient-induced cytotoxicity in fibroblasts from leigh syndrome French canadian patients

Abstract

Mutations in LRPPRC are responsible for the French Canadian variant of Leigh Syndrome (LSFC), a severe disorder characterized biochemically by a tissue-specific deficiency of cytochrome c oxidase (COX) and clinically by the occurrence of severe and deadly acidotic crises. Factors that precipitate these crises remain unclear. To better understand the physiopathology and identify potential treatments, we performed a comprehensive analysis of mitochondrial function in LSFC and control fibroblasts. Furthermore, we have used this cell-based model to screen for conditions that promote premature cell death in LSFC cells and test the protective effect of ten interventions targeting well-defined aspects of mitochondrial function. We show that, despite maintaining normal ATP levels, LSFC fibroblasts present several mitochondrial functional abnormalities under normal baseline conditions, which likely impair their capacity to respond to stress. This includes mitochondrial network fragmentation, impaired oxidative phosphorylation capacity, lower membrane potential, increased sensitivity to Ca2+-induced permeability transition, but no changes in reactive oxygen species production. We also show that LSFC fibroblasts display enhanced susceptibility to cell death when exposed to palmitate, an effect that is potentiated by high lactate, while high glucose or acidosis alone or in combination were neutral. Furthermore, we demonstrate that compounds that are known to promote flux through the electron transport chain independent of phosphorylation (methylene blue, dinitrophenol), or modulate fatty acid (L-carnitine) or Krebs cycle metabolism (propionate) are protective, while antioxidants (idebenone, N-acetyl cysteine, resveratrol) exacerbate palmitate plus lactate-induced cell death. Collectively, beyond highlighting multiple alterations in mitochondrial function and increased susceptibility to nutrient-induced cytotoxicity in LSFC fibroblasts, these results raise questions about the nature of the diets, particularly excess fat intake, as well as on the use of antioxidants in patients with LSFC and, possibly, other COX defects.

Fig 1. Effect of the LRPPRC A354V mutation on LRPPRC content, mitochondrial content and cellular ATP levels.

(A) Immunoblot and densitometric analysis of LRPPRC content in whole cell lysates from control and LSFC fibroblasts (n = 4). Representative immunofluorescence images of control (B;C) and LSFC fibroblasts (D;E) labeled with anti-LRPPRC (green) and anti-pyruvate dehydrogenase (red) antibodies. Overlay images (yellow), and line scan analysis show the cellular distribution of LRPPRC in the mitochondrial compartment. (F) Cytochrome c oxidase (COX) enzyme activity in whole cell lysates from control and LSFC fibroblasts was normalized to that of the mitochondrial marker citrate synthase (CS) to take into account possible differences in mitochondrial content (n = 4). (G) Immunoblot and densitometric analysis of VDAC, a mitochondrial marker protein, in whole cell lysates from control and LSFC fibroblasts (n = 3). (H) Citrate synthase activity in whole cell lysates from control and LSFC fibroblasts (n = 4). (I) Cellular ATP content in control and LSFC fibroblasts (n = 4). Data are expressed as means ± S.E. Experiments were performed in one control (EBS-4) and one LSFC cell line (AL-006). Difference between control and LSFC cells was assessed with a paired t-test. ** Significantly different from the control group p ≤ 0.01. Statistical power for LRPPRC and COX was 97%.

Fig 2. Effect of the LRPPRC A354V mutation on basal mitochondrial network morphology and functions.

Representative live cell images of MTG-loaded control (A) and LSFC (B) fibroblasts used for quantitative analysis of mitochondrial network morphology. (C) Form Factor (FF) values calculated using the equation FF = 4π*Area/perimeter2 (n = 6). Representative live cell images of control (D) and LSFC (E) fibroblasts labeled with TMRE (red) and MTG (green). (F) Mitochondrial membrane (ΔΨ) potential expressed as the ratio of TMRE to MTG (n = 5). Lower values are indicative of reduced ΔΨ. (G) Mean fluorescence intensity of the mitochondria-specific superoxide probe MitoSOX in control and LSFC fibroblasts (n = 5). (H) Maximal ADP-driven respiration in digitonin-permeabilized fibroblasts energized with complex I (5 mM glutamate—2.5 mM malate; Glut-Mal; n = 15) or complex II substrates in presence of the complex I inhibitor rotenone (5 mM succinate + 1 μM rotenone; Succ+Rot; n = 14). Inset shows representative respirometry traces confirming that respiratory rates increased promptly in response to the addition of respiratory substrates, and were potently inhibited by complex I (rotenone), and complex II (malonate) blockers. (I) Mitochondrial calcium retention capacity (CRC) in control and LSFC fibroblasts exposed to progressive Ca2+ loading (n = 8). Inset shows representative Ca2+ kinetic tracings observed in control and LSFC fibroblasts. Tracings show progressive Ca2+ accumulation followed by PTP-induced release of accumulated Ca2+. Each spike indicates the addition of a calcium pulse of 83 nmoles. All experiments were performed in one control (EBS-4) and one LSFC (AL-006) cell line, except for the determination of ΔΨ, which was performed in EBS-3 and AL-002. Data are expressed as means ± S.E. Difference between control and LSFC cells was assessed with a paired t-test. Significantly different from the control group: * p < 0.05, ** p ≤ 0.01. Statistical power: C: 92%; F: 85%; G: 80%; H: Glut-Mal 80%; Succ+Rot: 96%; I: 73%.

Fig 3. Stress-induced cell death in control and LSFC fibroblasts.

(A) Lactate dehydrogenase (LDH) release in control and LSFC fibroblasts exposed for 48 h to factors relevant to acidotic crises. (B) Mean LDH release (n = 5), (C) Caspase 3/7 activity (n = 7), (D) Cellular ATP content (n = 7) and (E) COX/CS activity ratio (n = 3) assessed at baseline and after exposure to PL (palmitate 1 mM; lactate 10 mM) for 34 h (LDH release) or 24 h (other parameters). All experiments were performed in one control (EBS-4) and one LSFC (AL-006) cell line. Data are expressed as means ± S.E. Statistical significance of differences between groups or conditions was assessed with a two-way ANOVA for repeated measures followed by a Bonferroni post hoc test. Significantly different from the control cells in the same experimental condition: * p < 0.05, ** p < 0.01. Significantly different from baseline within the same experimental group: ^ρ p < 0.05, ^ρρ p < 0.01, ^ρρρ p < 0.001.

Fig 4. Effect of selected therapeutic agents on palmitate plus lactate (PL)-induced cytotoxicity.

(A) Lactate dehydrogenase (LDH) release in control and LSFC fibroblasts exposed to PL in presence or absence of therapeutic compounds. The numbers in brackets (ex: control: n = 2:6) refer to the number of cell lines tested and of experiments, respectively. (B) Mean LDH release (n = 5), (C) Caspase 3/7 activity (n = 5), (D) Cellular ATP content (n = 7) and (E) COX/CS activity ratio (n = 3) measured after exposure to PL (palmitate 1 mM; lactate 10 mM) in absence or presence of CPMB (1 mM L-carnitine; 0.2 mM propionate; 125 nM Methylene blue) for 34 h (LDH release) or 24 h (other parameters). All measurements were performed in 4 control and 3 LSFC cell lines for the aforementioned number of experiments. Data are means ± S.E. and expressed as fold-change relative to the mean value of the PL condition in control (ctrl) cells. Statistical significance of differences between groups and conditions were analyzed with a linear hierarchical model (cf. Methods section). Significantly different from the control cells in the same experimental condition: * p < 0.05, ** p < 0.01. Significantly different from baseline within the same experimental group: ^ρ p < 0.05, ^ρρ p < 0.01, ^ρρρ p < 0.001.