Pathological mitophagy disrupts mitochondrial homeostasis in Leber's hereditary optic neuropathy

Abstract

Leber's hereditary optic neuropathy (LHON), a disease associated with a mitochondrial DNA mutation, is characterized by blindness due to degeneration of retinal ganglion cells (RGCs) and their axons, which form the optic nerve. We show that a sustained pathological autophagy and compartment-specific mitophagy activity affects LHON patient-derived cells and cybrids, as well as induced pluripotent-stem-cell-derived neurons. This is variably counterbalanced by compensatory mitobiogenesis. The aberrant quality control disrupts mitochondrial homeostasis as reflected by defective bioenergetics and excessive reactive oxygen species production, a stress phenotype that ultimately challenges cell viability by increasing the rate of apoptosis. We counteract this pathological mechanism by using autophagy regulators (clozapine and chloroquine) and redox modulators (idebenone), as well as genetically activating mitochondrial biogenesis (PGC1-α overexpression). This study substantially advances our understanding of LHON pathophysiology, providing an integrated paradigm for pathogenesis of mitochondrial diseases and druggable targets for therapy.

Keywords: CP: Neuroscience; LHON; autophagy; cybrids; iPSCs; mitochondria; mitophagy; mtDNA; optic nerve; retinal ganglion cells; therapy.

Graphical abstract

Figure 1

Autophagy is pathologically increased in cells from LHON-affected patients but not in unaffected mutation carriers (A–F) Detection of autophagy activity through immunoblot and GFP-LC3 puncta count in LHON fibroblasts carrying 3460 (A and B) and 11778 mutations (C–F). In the latter, the autophagy levels were also analyzed in fibroblasts obtained from the two unaffected mutation-carrier brothers (identified as #1 and #2). Where not indicated, the representative western blots come from couple #2, while the histograms represent the average of the data from the two pairs #1 and #2. Where indicated, cells were starved (STARV.) for 1 h. (G) Autophagy levels were also detected in idNeurons harboring LHON mutations. (H) Finally, autophagic levels were measured by immunoblot in ex vivo PBMCs obtained from healthy individuals (CTRLS) (n = 17), LHON-affected (n = 10), and LHON-carrier (n = 14) (the representative image shown has been cropped to invert the order of samples loading). (I and J) ELISA was performed on serum samples from CTRLS (n = 10), LHON-carrier (n = 9), and LHON-affected (n = 7) patients to detect ATG5 (I) and ATG7 (J). Data are presented as means ± SEM. n = at least 3 independent experiments for western blots or 5 visual fields per at least 3 independent samples per condition for GFP-LC3 experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 2

LHON disease is characterized by excessive mitophagy levels (A, B, D, and E) Confocal microscopy assessment of mitophagy respectively in fibroblasts and 35-day-old idNeurons carrying 3460 (A and D) and 11778 (B and E) mutations were performed by loading cells with LysoTracker red and MitoTracker green to visualize lysosomes and mitochondria, respectively. (C) Mitophagy levels were analyzed by detecting the amount of fluorescent YFP-Parkin localized on the mitochondrial surface in fibroblasts from one LHON-affected patient carrying the 11778 mutation and of the non-affected (carrier) brother carrying the same 11778 mutation. (F) A similar methodological approach with LysoTracker red and MitoTracker green was used to investigate whether mito-autophagosomes were present in specific regions of the idNeurons (soma and axon-dendrite (Ax-De) regions). ###p < 0.001 to CTRL soma, §§§p < 0.001 to CTRL Ax-De, ^∗∗∗p < 0.001 Ax-De to its own soma. (G and H) Increased levels of mitophagy marker Parkin (G) and Optineurin (H) were detected in serum samples of 11778 LHON-affected patients (n = 7) compared with CTRLS (n = 10) and 11778 LHON-carriers (n = 9). Data are presented as means ± SEM. n = at least 5 visual fields per at least 3 independent samples per condition for colocalization experiments. Data to evaluate region-specific mitophagy were obtained from 78 regions of interest (ROIs) for iPSC-derived control neurons, 297 ROIs for iPSC-derived 3460 neurons, and 225 ROIs for iPSC-derived 11778 neurons and analyzed using one-way ANOVA. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 3

Autophagy and mitophagy results increased in cybrids (A–D) Autophagy detection by immunoblot (A) and fluorescent microscopy (B) was performed in cybrids carrying 3460 and in cybrids derived from fibroblasts from one LHON-affected patient carrying the 11778 mutation and of the non-affected (carrier) brother carrying the same 11778 mutation (C and D). Where indicated, the cells were STARV. for 1 h. (E) Confocal microscopy assessment of mitophagy in control and mutant cybrids harboring 3460 LHON mutations. (F) Similar experiments were achieved in LHON-affected cybrids carrying the 11778 mutation and of the non-affected (carrier) brother carrying the same 11778 mutation. Data are presented as means ± SEM. n = at least 3 independent experiments for western blots or 5 visual fields per at least 3 independent samples per condition for fluorescent microscopy experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 4

Complex I deficiency leads to altered mitochondrial function in LHON-affected individuals, which is compensated in carriers (A–D) Measurements of mitochondrial ROS production in LHON-derived 3460 fibroblasts (A) and cybrids (B) and in 11778 fibroblasts (C) and cybrids (D) by using MitoSOX red probe. (E–H) The mitochondrial transmembrane potential (Ψ_m) of 3460 and 11778 fibroblasts and cybrids was detected by using the Ψ_m-sensitive probe TMRM. When indicated, the cells were STARV. for 30 min before TMRM loading. Data are presented as means ± SEM. n = at least 5 visual fields per at least 3 independent samples per condition. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 5

Compensatory therapeutic approaches targeting autophagy reverts LHON cells’ predisposition to apoptotic death (A and B) 11778 fibroblasts (A) and cybrids (B) harboring LHON mutations were treated with different autophagic inhibitors (3-MA [inhibitor of autophagy at early steps], chloroquine [CQ], and clozapine [CLOZ] [inhibitors of autophagy at late steps]). After 48 h, detection of autophagic activity through immunoblot technique was performed. (C and D) Detection of apoptotic process activity on 11778 LHON fibroblasts (C) and cybrids (D) treated with anti-autophagic agents by immunoblotting with antibodies against PARP and CAS3 apoptotic markers. (E and F) Cell viability in fibroblasts (E) and cybrids (F) pretreated with anti-autophagy compounds was performed at different time points. Data are presented as means ± SEM. n = at least 3 independent experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.

Figure 6

Oxidative stress modulation contributes to decrease in LHON cells’ autophagic activity and apoptotic death (A–D) After treatment with reduced idebenone (IDEB.) (10 μM for 3 h), fibroblasts (A and B) and iNeurons (C and D) were harvested and immunoblotted for the autophagic marker LC3 and against apoptotic markers PARP and CAS3. (E and F) Measurements of mitochondrial ROS production (E) and mitochondrial transmembrane potential (F) were studied in IDEB.-treated 11778 fibroblasts. (G–J) The same idebenone treatment was repeated in cybrids, where autophagy (G), apoptosis (H), ROS production (I), and mitochondrial transmembrane potential (J) were detected. Data are presented as means ± SEM. n = at least 3 independent experiments for western blots or 5 visual fields per at least 3 independent samples per condition for fluorescent microscopy experiments. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001.