Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 7, 11483

Deregulation of Mitochondrial F1FO-ATP Synthase via OSCP in Alzheimer's Disease

Affiliations

Deregulation of Mitochondrial F1FO-ATP Synthase via OSCP in Alzheimer's Disease

Simon J Beck et al. Nat Commun.

Abstract

F1FO-ATP synthase is critical for mitochondrial functions. The deregulation of this enzyme results in dampened mitochondrial oxidative phosphorylation (OXPHOS) and activated mitochondrial permeability transition (mPT), defects which accompany Alzheimer's disease (AD). However, the molecular mechanisms that connect F1FO-ATP synthase dysfunction and AD remain unclear. Here, we observe selective loss of the oligomycin sensitivity conferring protein (OSCP) subunit of the F1FO-ATP synthase and the physical interaction of OSCP with amyloid beta (Aβ) in the brains of AD individuals and in an AD mouse model. Changes in OSCP levels are more pronounced in neuronal mitochondria. OSCP loss and its interplay with Aβ disrupt F1FO-ATP synthase, leading to reduced ATP production, elevated oxidative stress and activated mPT. The restoration of OSCP ameliorates Aβ-mediated mouse and human neuronal mitochondrial impairments and the resultant synaptic injury. Therefore, mitochondrial F1FO-ATP synthase dysfunction associated with AD progression could potentially be prevented by OSCP stabilization.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Figure 1
Figure 1. Loss of OSCP in AD subjects and 5xFAD mouse brain mitochondria.
(a) Densitometric quantification of immunoreactive bands of OSCP in protein extracts from the temporal lobes of non-AD, MCI and AD patients. β-Actin was used to indicate the loading amount of proteins. The lower panel is representative of six non-AD, six MCI and four AD patients. (b) Immunohistochemical staining using anti-OSCP on brain sections from human temporal lobes showed substantial reduction of OSCP in neurons. The right panel shows representative images. Scale bar, 40 μm. (c,d) Densitometric quantification of OSCP immunoreactive bands shows the age-dependent OSCP reduction in synaptic (c) and nonsynaptic (d) mitochondria from 5xFAD mice. TOM40 was used to show the enrichment of mitochondrial fractions. Coomassie blue staining was employed to indicate the loading amount of mitochondrial proteins. Five nonTg and five 5xFAD mice at 4 months old and 5 nonTg and 6 5xFAD mice at 9 months old were used in the experiments. Error bars represent s.e.m.
Figure 2
Figure 2. OSCP deficiency impacts neuronal mitochondrial function.
(a) OSCP expression was downregulated in primary cultured mouse neurons by using lentivirus carrying OSCP shRNA (OSCP KD (OSCP knockdown neurons)) and the control neurons were treated by lentivirus carrying nonTarget shRNA. n=6–9 samples. The lower panel shows the representative image of immunoreactive bands of OSCP. TOM40 was used as the loading control. (b) OSCP downregulation induced reduction in mitochondrial membrane potential. n=36–51 neurons from at least three independent experiments. The right panel shows representatives of TMRM staining and phase contrast images. Scale bar, 20 μm. (c) OSCP deficiency also induced decreased neuron ATP production (n=13–22 samples from at least three independent experiments). (d) OSCP deficiency induced increased mitochondrial superoxide levels (n=10–15 neurons from at least three independent experiments). The right panel shows representatives of Mitosox Red staining (red). NISSL (Blue) was used to identify neurons. Scale bar, 20 μm. (e) Loss of OSCP induced reduction in neuritic mitochondrial population. The right panel shows representative images of MAP2 (green, dendrites) and Mito-Dsred (red, mitochondria) staining. n=27–43 neurons from at least three independent experiments. Scale bar, 20 μm. (f) mPTP formation demonstrated by the drop in mitochondrial calcein intensity in the exposure of 2 μM ionophore. CsA was used at 1 μM. *P<0.01 versus OSCP KD neurons in the presence or absence of CsA. #P<0.05 versus control neurons treated by CsA. n=5–10 independent experiments. Error bars represent s.e.m.
Figure 3
Figure 3. OSCP deficiency impacts synaptic function.
(a) Loss of OSCP induced a decline in synaptic density. The staining of vGlut1 (blue) and PSD95 (red) were used to identify the pre- and post- synaptic components of synapses, respectively. Neuronal dendrites were identified by staining for MAP2 (green). Scale bar, 5 μm. n=27–43 neurons from at least three independent experiments. (be) Loss of OSCP affected mEPSCs. (b) Representative traces of mEPSCs in control (upper panel) and OSCP KD neurons (lower panel). Scale bars represent 200 ms and 20 pA. Data were collected from 10 control neurons and 7 OSCP knockdown (OSCP KD) neurons from at least three different cultures. (c) Quantitative analysis of mEPSC frequency. (d) Quantitative analysis of mEPSC amplitude. (e) Analysis of the cumulative fraction of mEPSC amplitude distribution. (f) Loss of OSCP reduced the amplitude of EPSCs evoked by puff-application of glutamate. Scale bars represent 500 ms and 100 pA. Data were collected from seven control neurons and eight OSCP KD neurons. The right panel shows quantitative analysis of EPSC amplitude. Error bars represent s.e.m.
Figure 4
Figure 4. OSCP/Aβ interactions impact OSCP function in vivo and in vitro.
(a) Co-immunoprecipitation of OSCP and Aβ in AD patient temporal lobes. Results shown are representatives from three non-AD and three AD patients. Aβ1–42 peptide and AD brain extracts were used as positive controls for Aβ immunoreactive bands. (b) Co-immunoprecipitation of OSCP and Aβ in 5xFAD mouse neocortex. Results shown are representatives from three mice in each group. Aβ1–42 peptide was used as a positive control for Aβ immunoreactive bands. (c) Colocalization of OSCP (green) and Aβ (red) in neocortex and hippocampus from 5xFAD mice. Neurons were identified by staining for NISSL (blue). Scale bar, 10 μm. (d) OSCP and Aβ interaction determined by an in vitro pull-down assay. (e) Amino-acid residue numbers are given for mature OSCP protein (with known mitochondrial signalling peptide removed). SP is the mitochondrial signalling peptide. Wild-type OSCP and OSCPΔ107–121 were used for pull-down assay. OSCPΔ107–121 displayed lowered capacity to interact with Aβ compared to wild-type OSCP. (f) In vitro pull-down assay showed that Aβ1–42 suppresses the ability of OSCP, but not OSCPΔ107–121, to bind α- and β-subunits as demonstrated by decreased α- and β-immunoreactive bands. n=6–10 samples per group. Error bars represent s.e.m.
Figure 5
Figure 5. F1FO-ATP synthase deregulation in 5xFAD mouse synaptic mitochondrial.
(a) Synaptic mitochondria from 5xFAD mice demonstrated an age-dependent decrease in their respiratory control ratio (RCR). Six nonTg and 5 5xFAD mice at 4 months old and 6 nonTg and 5 5xFAD mice at 9 months old were used. (b) Synaptic mitochondria from 5xFAD mice had an age-dependent decrease in ATP synthesis. Six nonTg and 6 5xFAD mice at 4 months old and those from six nonTg and seven 5xFAD mice at 9 months old were used in the experiments. (c) Synaptic mitochondria from 5xFAD mice demonstrated an early decrease in the F1FO-ATP synthase catalytic activity at 4 months old which was exacerbated at 9 months old. Five mice of each group at 4 months old and seven nonTg and nine 5xFAD mice at 9 months old were used in the experiment. (d) Age-dependent increase in oligomycin-insensitive respiration of synaptic mitochondria from 5xFAD mice. Six nonTg and five 5xFAD mice at 4 months old, and six nonTg and five 5xFAD mice at 9 months old were used in the experiments. (e,f) Decreased oligomycin sensitivity of synaptic mitochondria from 5xFAD mice at 4 (e) and 9 months old (f). All data are presented as percentage of the activity of the corresponding vehicle-treated mitochondrial fractions. Six nonTg and five 5xFAD mice at 4 months old, and seven nonTg and seven 5xFAD mice at 9 months old were used in the experiments. (g,h) Increased F1 dissociation in synaptic mitochondria from 5xFAD mice. (g) The analysis of free F1. (h) The left panel is the representative of images collected from seven nonTg and six 5xFAD mice at 4 months old, and six nonTg and six 5xFAD mice at 9 months old. F1 was determined by probing with anti-β subunit antibody and the molecular weight of the bands. The same amount of samples was used for SDS–PAGE and Tom40 and β subunit were detected to show the loading amount of mitochondrial proteins. The right panel is the coomassie blue staining before immunoblotting to indicate the loading amount of samples. Error bars represent s.e.m.
Figure 6
Figure 6. OSCP overexpression ameliorates Aβ-induced mitochondrial dysfunction in mouse neurons.
Mouse cortical neurons were exposed to 500 nM oligomeric Aβ1–42 for 24 h. (a) Attenuated oligomeric Aβ1–42-induced OSCP reduction by OSCP overexpression. Western blot images show OSCP expression level in mouse primary neurons which is representative of seven samples in each group. (b) Shows immunofluorescent staining of OSCP (green) in primary cultured neurons. COXIV (red) was used to determine the localization of OSCP in mitochondria. *P<0.01 versus other groups. Scale bar, 20 μm. (c) OSCP overexpression protected mitochondrial membrane potential against oligomeric Aβ1–42 toxicity. The upper panel of representative images shows the TMRM staining and the lower panel shows phase contrast images. Scale bar, 20 μm. *P<0.01 versus other groups. n=40–64 neurons from at least three independent experiments. (d) OSCP overexpression protected neuron ATP reduction against oligomeric Aβ1–42 toxicity. *P<0.01 versus other groups. n=12 samples of each group from at least three independent experiments. (e) Attenuated Aβ-induced reduction in neuritic mitochondrial population by OSCP overexpression. *P<0.01 versus other groups. n=22–27 neurons from at least three independent experiments. The upper panel of representative images shows the merged staining of MAP2 (green, dendrite) and Mito-Dsred (red, mitochondria). Scale bar, 20 μm. The middle panel shows Mito-Dsred staining and the lowest panel shows enlarged images from the indicated areas. (f) Ameliorated Aβ-sensitized mPTP formation by OSCP overexpression. Ionophore (2 μM) was used to trigger mPTP formation. *P<0.05 versus vehicle-treated control and OSCP OE neurons. #P<0.05 versus Aβ-treated OSCP OE neurons. n=4–6 independent experiment. Control refers to neurons infected with control lentivirus carrying backbone vector. OSCP OE refers to neurons infected with lentivirus carrying OSCP cDNA. Error bars represent s.e.m.
Figure 7
Figure 7. OSCP overexpression protects mouse neurons against Aβ induced synaptic dysfunction.
(a) Attenuated Aβ-induced synaptic density reduction in OSCP OE neurons. *P<0.01 versus other groups. n=26–43 neurons collected from at least three independent experiments. Synapses were visualized by the staining for vGlut1 (Blue) and PSD95 (red) to identify the pre- and postsynaptic components of synapses, respectively. Neuronal dendrites were determined by MAP2 (green). Scale bar, 5 μm. (be) OSCP overexpression protected mEPSCs from Aβ toxicity. (b) Representative traces of mEPSC recordings from control and OSCP OE neurons in the presence or absence of Aβ. Scale bars represent 200 ms and 20 pA. The data were collected from 10 vehicle-treated control neurons, 8 Aβ-treated control neurons, 7 vehicle-treated OSCP OE neurons and 8 Aβ-treated OSCP OE neurons. (c) Quantitative analysis of mEPSC frequency. (d) Quantitative analysis of mEPSC amplitude. *P<0.05 versus other groups. (e) The cumulative fraction of mEPSC amplitude distribution. (f) Overexpression of OSCP prevented the reduction in amplitude of glutamate-evoked EPSCs that results from Aβ toxicity in control neurons. Scale bars represent 500 ms and 100 pA. Data were collected from seven vehicle-treated control neurons, nine Aβ-treated control neurons, five vehicle-treated OSCP OE neurons and seven Aβ-treated OSCP OE neurons. Error bars represent s.e.m.
Figure 8
Figure 8. OSCP overexpression ameliorates Aβ-induced mitochondrial dysfunction in human neurons.
(a) Human neurons were exposed to 500 nM oligomeric Aβ1–42 for 4 days, then subjected to immunoblotting detection of OSCP levels. Aβ induced a significant reduction in OSCP expression level. The lower panel shows representative immunoblotting images. Tom40 was used as the loading control. n=4–5 samples of each group. (b) Co-immunoprecipitation of OSCP and Aβ in Aβ-treated human neurons. (c) Significantly increased OSCP expression in the OSCP overexpressing neurons. The lower panel shows representative immunoblotting images. Tom40 was used as the loading control. n=4 samples of each group. (d) Preserved ATP production by OSCP overexpression. *P<0.01 versus other groups. n=6–10 samples of each group. (e) Preserved mitochondrial membrane potential by OSCP overexpression. *P<0.01 versus other groups. n=17–23 neurons from at least 3 time independent experiments. Right panels are representative images of TMRM staining (upper row) and phase contrast (lower row). Scale bar, 20 μm. (f) Preserved neuritic mitochondrial population by OSCP OE. *P<0.05 versus other groups. Right panels are representative images. Dendrites were determined by the staining of MAP2 (green); and mitochondria were identified by mito-Dsred (red). Scale bar, 20 μm. (g) Attenuated Aβ-sensitized mPTP formation by OSCP overexpression. Ionophore (2 μM) was used to trigger mPTP formation. *P<0.05 versus vehicle-treated control and OSCP OE neurons. #P<0.05 versus Aβ-treated OSCP OE neurons. n=4–7 independent experiments. Error bars represent s.e.m.
Figure 9
Figure 9. Schematic summary.
With the progress of AD, brain mitochondria gradually undergo OSCP loss and Aβ accumulation in AD-relevant conditions. OSCP loss and OSCP/Aβ interaction impair OSCP function to keep F1FO complex integrity. This leads to severe mitochondrial dysfunction, including decreased ATP production, collapsed mitochondrial membrane potential, and increased ROS production and release, as well as the activation of mitochondrial permeability transition pore formation. Such mitochondrial deregulation results in dampened neuronal function and eventually neuronal death.

Similar articles

See all similar articles

Cited by 26 PubMed Central articles

See all "Cited by" articles

References

    1. Reddy P. H. Role of mitochondria in neurodegenerative diseases: mitochondria as a therapeutic target in Alzheimer’s disease. CNS. Spectr. 14, 8–13 (2009). - PMC - PubMed
    1. Lin M. T. & Beal M. F. Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature 443, 787–795 (2006). - PubMed
    1. Swerdlow R. H. et al. . Mitochondrial dysfunction in cybrid lines expressing mitochondrial genes from patients with progressive supranuclear palsy. J. Neurochem. 75, 1681–1684 (2000). - PubMed
    1. Yao J. et al. . Mitochondrial bioenergetic deficit precedes Alzheimer’s pathology in female mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. USA 106, 14670–14675 (2009). - PMC - PubMed
    1. Du H. et al. . Cyclophilin D deficiency attenuates mitochondrial and neuronal perturbation and ameliorates learning and memory in Alzheimer’s disease. Nat. Med. 14, 1097–1105 (2008). - PMC - PubMed

Publication types

Feedback