Mitochondrial deficits and abnormal mitochondrial retrograde axonal transport play a role in the pathogenesis of mutant Hsp27-induced Charcot Marie Tooth Disease

Abstract

Mutations in the small heat shock protein Hsp27, encoded by the HSPB1 gene, have been shown to cause Charcot Marie Tooth Disease type 2 (CMT-2) or distal hereditary motor neuropathy (dHMN). Protein aggregation and axonal transport deficits have been implicated in the disease. In this study, we conducted analysis of bidirectional movements of mitochondria in primary motor neuron axons expressing wild type and mutant Hsp27. We found significantly slower retrograde transport of mitochondria in Ser135Phe, Pro39Leu and Arg140Gly mutant Hsp27 expressing motor neurons than in wild type Hsp27 neurons, although anterograde movement velocities remained normal. Retrograde transport of other important cargoes, such as the p75 neurotrophic factor receptor was minimally altered in mutant Hsp27 neurons, implicating that axonal transport deficits primarily affect mitochondria and the axonal transport machinery itself is less affected. Investigation of mitochondrial function revealed a decrease in mitochondrial membrane potential in mutant Hsp27 expressing motor axons, as well as a reduction in mitochondrial complex 1 activity, increased vulnerability of mitochondria to mitochondrial stressors, leading to elevated superoxide release and reduced mitochondrial glutathione (GSH) levels, although cytosolic GSH remained normal. This mitochondrial redox imbalance in mutant Hsp27 motor neurons is likely to cause low level of oxidative stress, which in turn will contribute to, and indeed may be the underlying cause of the deficits in mitochondrial axonal transport. Together, these findings suggest that the mitochondrial abnormalities in mutant Hsp27-induced neuropathies may be a primary cause of pathology, leading to further deficits in the mitochondrial axonal transport and onset of disease.

Figure 1

Cell survival in SH-SY5Y cells transfected with wild type (WT) and mutant Hsp27 constructs. (A) The images show the morphology of SH-SY5Y cells under i) control conditons and following exposure to ii) H₂O ₂, iii) cytochalasin D, and iv) colchicine-induced stress. The cells are immunostained for beta III tubulin staining (red) and co-stained for the nuclear marker, DAPI (blue). Scale bar=50µm. (B) The bar charts show the extent of cell death assessed using an LDH assay in which LDH release from control and stressed cells was determined 4 days following transfection with WT and mutant Hsp27, and 24-h treatment with the cell stressors. LDH release in experimental conditions is normalized to that of untransfected cells, thus values >0 signify elevated cell death compared to untransfected controls. Error bars= SEM. *P < 0.05; **P < 0.001.

Figure 2

Morphological characterization of primary motor neuron cultures expressing WT and mutant Hsp27. Expression of (A) beta III tubulin (red) and (B) neurofilament light chain (red) in virally infected motor neurons expressing i) WT Hsp27, or ii) Ser135Phe Hsp27; iii) Pro37Leu Hsp27, or iv) Arg140Gly mutant Hsp27. GFP: green. Scale bars=10 µm.

Figure 3

Axonal transport of mitochondria in mutant Hsp27 expressing primary motor neurons. (A) i) A typical primary mouse motor neuron expressing GFP (green), indicating transgene expression of WT Hsp27, and co-labelled using Mitotracker (red). ii) A higher power image of an axon of the neuron shown in i) (dotted rectangle). Scale bar = 10µm. (B) Live cell tracking of an individual mitochondria within an axonal segment using time lapse microscopy. i) Individual frames of an axon transport movie showing tracking of one depicted mitochondrion (arrows). ii) A kymograph generated using the whole mitochondrial transport movie, with vertical lines indicating stationary mitochondria and some shifted lines indicating movement of mitochondria between frames. (C) The proportion of moving mitochondria in motor neuron axons as a percentage of the total number of mitochondria in each experimental condition. (D) Relative frequency of the speed of mitochondrial movements. Speed data have been binned and their frequency is shown in each condition in the anterograde (left; green shaded area) and retrograde (right; purple shaded area) direction. (E) For each experimental condition, the mean number of movements in the anterograde and retrograde directions is summarised. (F) Pause analysis of mitochondrial movements in axons. The percentage of time each mitochondrion spent pausing is calculated for each experimental condition. Error bars= SEM; p > 0.05.

Figure 4

Axonal transport of p75NTR in mutant Hsp27 expressing motor neurons. (A) i) A primary mouse motor neuron expressing GFP (green), co- expressed with a WT Hsp27 construct and the intracellular fragment of P75NTR (red). ii) An axonal section selected from the cell in i) (dotted rectangle) is shown at higher magnification. Scale bar=10 µm. (B) Relative frequency of the speed of p75NTR movements. Speed data have been binned and their frequency is shown. Error bars =SEM. (C) Pause analysis of p75NTR movements in motor axons. The percentage of time each labelled receptor spent pausing is calculated for each experimental condition. Error bars= SEM *P < 0.05; **P < 0.001.

Figure 5

Mitochondrial membrane potential in mutant Hsp27 expressing motor neurons. (A) A primary mouse motor neuron expressing WT Hsp27 and GFP, labelled with the potentiometric dye TMRM (red). Scale bar= 20µm. (B) TMRM intensity of individual mitochondria measured in WT and Ser135Phe Hsp27 expressing motor neurons. TMRM intensity was measured in both mitochondria of the cell body (black bars) and in neurites (grey bars). (C) Changes in TMRM intensity (mitochondrial membrane potential) in response to oligomycin (Complex V blocker); rotenone (Complex II blocker) and FCCP (protonophore). (D) Mitochondrial Complex I activity in motor neurons expressing WT and mutant Hsp27. Error bars=SEM *P < 0.05.

Figure 6

Oxidative stress in mitochondria of mutant Hsp27 motor neurons. To detect mitochondrial superoxide a mitochondria selective superoxide probe (Mitosox) was used, in a system employing FACS analysis (A,B) and confocal microscopy (C). Ai) A scattergraph showing primary mouse motor neurons expressing WT Hsp27 and GFP under control conditions (FL1H= green channel to detect GFP expressing cells; FL3-H= red channel to detect red mitosox signal). A very small number of cells are above threshold, positive for both GFP and Mitosox. Aii) The bar chart shows the percentage of GFP and Mitosox positive cells expressed within the population of GFP-positive, infected cells, under control, unstimulated conditions. Bi) A scattergraph of the same set of cells as in Ai), following a 20-min incubation with Antimycin A. There is a shift in the number of cells above threshold for both green and red channels, indicating that more GFP positive cells are also positive for Mitosox. The number of GFP and Mitosox positive cells expressed following 20-min incubation with antimycin A is shown in Bii). (C) Confocal images of primary motor neurons expressing WT Hsp27 (GFP) stained for mitosox (red) under (i) control conditions and (ii) following 20-min incubation with Antimycin A. (Scale bars=SEM). (iii) The fluorescent intensity of Mitosox measured before and during the 20-min antimycin A treatment, shown as a fold increase to baseline for each experimental condition. (D) Primary motor neurons expressing WT Hsp27 (GFP) and GSH (white) in (i) whole cells and (ii) in mitochondria only. (Scale bars=10µm). (iii) For each set of experiments GSH intensity in whole cells (dark bars) and mitochondria (light bars) was expressed as a percentage of the fluorescence in WT Hsp27 expressing cells. (E) Primary motor neurons, infected with an empty vector (i) and (ii) a vector expressing Ser135Phe Hsp27 mutant stained for nitrotyrosine. (Scale bar=20) (iii) Western blot showing nitrotyrosine levels in primary motor neurons infected with lentiviral constructs.