Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 33 (43), 17042-51

Mitochondrial Alterations Near Amyloid Plaques in an Alzheimer's Disease Mouse Model

Affiliations

Mitochondrial Alterations Near Amyloid Plaques in an Alzheimer's Disease Mouse Model

Hong Xie et al. J Neurosci.

Abstract

While accumulation of amyloid-β (Aβ) deposited as senile plaques is a hallmark feature of Alzheimer's disease (AD), the neurotoxicity of these deposits remains controversial. Recent in vitro studies suggested a link between elevated Aβ and mitochondrial dysfunction that might contribute to the pathogenesis of AD. However, the in vivo evidence for mitochondria dysfunction caused by Aβ is still missing. Using intravital multiphoton imaging with a range of fluorescent markers, we systematically surveyed mitochondrial structural and functional changes in AD mouse models. We observed severe impairments to be limited to the vicinity of Aβ plaques, which included reduction of both numbers and membrane potential of mitochondria and the emergence of dystrophic and fragmented mitochondria. Both neuronal soma and neurites with oxidative stress show severe alterations in mitochondrial membrane potential in amyloid precursor protein mice. These results provide in vivo evidence revealing Aβ plaques as focal sources of toxicity that lead to severe structural and functional abnormalities in mitochondria. These alterations may contribute to neuronal network dysfunction and warrant further investigation as possible targets for therapeutic intervention in AD.

Figures

Figure 1.
Figure 1.
Mitochondrial loss and structural abnormalities in living APP/PS1 transgenic mouse brain. A, B, Representative images of lenti-mtGFP-labeled mitochondria are shown in the living brain. A, 3D reconstruction from an image stack in the somatosensory cortex. Scale bar, 50 μm; block size, 200 × 200 × 500 μm. B, There is a dramatic reduction in mitochondrial signal in the vicinity of dense-core plaques in APP/PS1 transgenic mice. Top, mtGFP (green)-labeled mitochondria and methoxy-XO4-stained plaques (blue) at low and high magnification. Scale bar, 20 μm. Bottom, Demonstrates how quantitative measures were performed. To the left, plaques were outlined and concentric bands at 4 μm intervals were generated. To the right are summary data of the quantitative analysis of mitochondrial density or “burden” in each band (n = 13 plaques from 4 animals). Scale bar, 20 μm. C, Mitochondrial swelling in living APP/PS1 transgenic mouse. Top, Live imaging of AAV-mtGFP (green)-labeled neuronal mitochondria and methoxy-XO4-stained plaques (blue). Arrows show swollen mitochondria. Bottom, Live imaging of AAV-GFP (green)-labeled neurites and methoxy-XO4-stained plaques (blue). Arrows show dystrophic neurites. Scale bar, 20 μm. D, More dystrophic neurites than swollen mitochondria were observed near plaques in Tg mice. E, Immunofluorescence of COX IV (red) showing loss of COX IV surrounding plaques. Scale bar, 40 μm.
Figure 2.
Figure 2.
Mitochondrial fragmentation in living APP/PS1 transgenic mouse brain. A, Examples of lenti-mtGFP-labeled mitochondrial structure in neurons in a Wt and APP/PS1transgenic mouse. In the top images, neurons in the brain of a Wt mouse exhibit normal mitochondrial morphology. The bottom images show a neuron near plaques in a Tg mouse brain that shows severe fragmentation of mitochondria as a result of excessive mitochondrial fission. Scale bar, 40 μm. B, Frequency distribution of mitochondrial length from neurons in A. Red, neurons with mitochondrial fragmentation; black, neurons with normal mitochondria. C, Results of counting the number of neurons with mitochondrial fragmentation per volume in Tg versus Wt mice (n = 6 mice per genotype).
Figure 3.
Figure 3.
MMP was not altered in areas far from amyloid plaques in the APP/PS1 transgenic mice. A, FCCP reduced the MMP in vivo as indicated with the MTR/MTG signal ratio in Wt mouse cortex. MTR (MMP sensitive), in red; MTG, in green; Hoechst, in blue. FCCP (10 μm) was applied on cortex for 15 min. The inset frames at higher magnification (20 × 20 μm) demonstrate that both MTR and MTG exhibit mitochondrial localization. To the right, the graph shows quantification of the signals in FCCP-treated versus untreated brains (n ≥ 8 imaging sites from 2 animals, p < 0.001). Scale bar, 10 μm. B, FCCP reduced the MMP in vivo as indicated with the JC-1 ratio (red/green) in Wt mouse cortex. FCCP (10 μm) was applied topically to the cortex for 15 min. The ratio of J-aggregate (red) to J-monomer (green) changed in all mitochondria in the affected cortex. Right, The quantification of the JC-1 ratio (n ≥ 11 imaging sites from 2 animals, p < 0.001). Scale bar, 15 μm. C, No differences in the JC-1 ratio were observed in areas distant to Aβ plaques (>50 μm) in Tg mouse, compared with Wt mouse, demonstrating that most mitochondria have a normal resting membrane potential. Left, JC-1 staining in 8- to 10-month-old Wt and Tg mouse. Right, Quantification of JC-1 ratio (n = 12 imaging sites from 3 to 5 animals per genotype). D, No difference in the MTR/MTG ratio confirms the lack of a detectable alteration in MMP in Tg versus Wt brains in areas distant from Aβ plaques (>50 μm; n ≥ 60 imaging sites from 3 to 4 animals per genotype). Scale bar, 15 μm.
Figure 4.
Figure 4.
Impairment of MMP near amyloid plaques in living APP/PS1 transgenic mouse. Representative examples of functional alterations in mitochondria near plaques are shown. A, The MMP-sensitive JC-1 ratio was decreased near plaques. Top, J-aggregates (J–a) in red; J-monomers (J-m) in green; methoxy-XO4 in blue. Quantification of JC-1 ratio around the plaques (n = 17 plaques from 5 mice), divided into concentric bands, with 6 μm intervals. Bottom, Magnified JC-1 image and JC-1 ratio map. White line showed the edge of the plaque and white arrowheads showed punctuate mitochondria with low JC-1 ratio. Scale bars: top, 30 μm; bottom, 12 μm. B, A loss in membrane potential is confirmed using the fluorescent dye TMRE that shows a reduced signal near plaques in Tg mice. Top, TMRE, in red; methoxy-XO4 in blue. Top right, quantification of TMRE signals, n = 46 plaques from three mice. Bottom, heatmap of TMRE fluorescence. White line shows the edge of the plaque and white arrowheads show punctuate mitochondria with lower signal. Scale bar, 20 μm. C, Using a third ratiometric approach for monitoring MMP, we demonstrate the loss of MMP using MTR and MTG staining. Left, MTR, in red; MTG, in green; methoxy-XO4, in blue. Right, Quantification of MTR/MTG ratio around the plaques (n = 18 plaques from 4 mice), divided into concentric bands, with 6 μm intervals. Scale bar, 20 μm. D, CAA does not affect mitochondrial function in smooth muscle cells. TMRE (red) and methoxy-XO4 (blue) staining in an 8-month-old APP/PS1 mouse. Vessels are masked with white dash lines. Left, Vessel without CAA; middle, vessel with CAA; right, comparison of TMRE fluorescence in vessels with and without CAA (n ≥ 21 vessels from 5 mice). Scale bar, 30 μm.
Figure 5.
Figure 5.
Reduced mitochondrial membrane potential in dystrophic neurites in living APP/PS1 transgenic mouse. Dystrophic neurites show severe alterations in MMP as shown in these representative images. A, A typical image of TMRE staining in AAV-GFP-labeled neurites in a living Tg mouse. Right insert indicates that no significant change of MMP induced by AAV-mediated GFP expression (n = 5 animal, p < 0.001). Scale bar, 13 μm. B,TMRE staining decreased in dystrophic neurites surrounding plaque. The white solid mask indicates dystrophic neurites and white dashed mask indicates normal neurites. Heatmap of TMRE fluorescence with masks, where blues are low MMP and greens to yellow and red are high MMP. C, Summary data that indicate that compared with normal neurites, the TMRE fluorescence in the dystrophic neurites was dramatically decreased and widely distributed (n = 5 animal, p < 0.001). Scale bar, 20 μm.
Figure 6.
Figure 6.
Oxidative stress was accompanied by mitochondrial dysfunction in living APP/PS1 transgenic mouse. Both neuronal soma and neurites with oxidative stress show severe alterations in MMP as shown in these representative images. A, A typical image of TMRE staining in AAV-roGFP-labeled neurons in a living Tg mouse. The emission fluorescence of roGFP excited at 800 (green) and 900 nm (red) was used as an index of redox potential (Xie et al., 2013). The white masks indicates normal neurons without oxidative stress and white mask with * indicates an oxidized neuron. Scale bar, 14 μm. B, A typical image of TMRE staining in AAV-roGFP-labeled neurites in a living Tg mouse. The white dashed mask indicates normal neurites without oxidative stress and white solid masks indicate oxidized neurites. Scale bar, 14 μm. C, MMP significantly decreased in oxidized neurites and neurons (n = 8 oxidized neurites from 3 animals, p < 0.001; n = 12 oxidized cells from 3 animals, p < 0.001).

Similar articles

See all similar articles

Cited by 51 PubMed Central articles

See all "Cited by" articles

Publication types

Substances

Feedback