Fig 1.
The Becn1 F121A mutation inhibits the BECN1-BCL2 interaction in vivo.
(A) Schematic representation of the strategy for hyperactive autophagy via the Becn1F121A knockin allele. F121A blocks binding of BECN1 with its inhibitor BCL2, which leads to upregulated BECN1 function and constitutively high autophagy. (B) Co-immunoprecipitation of BECN1 by BCL2 in skeletal muscle and brain tissues from wild-type (WT) and Becn1F121A mice. Less BECN1F121A is immunoprecipitated by BCL2 than WT BECN1, quantified by the BECN1/BCL2 ratio in the IP samples from 3 independent experiments. FA/FA, Becn1F121A homozygous knock-in mice. ***, P<0.001, t test.
Fig 2.
Becn1F121A leads to autophagy hyperactivation in vivo.
(A, B) Representative images (left panel) and quantification (right panel) of GFP-LC3 puncta (autophagosomes) in skeletal muscle (A) and brain (B) of GFP-LC3 Becn1+/+ and GFP-LC3 Becn1FA/FA mice at non-autophagy-inducing conditions (fed and rested), after 90-min exercise, or after 48 hours of starvation. Scale bar: 25 μm. Results represent mean ± s.e.m. N = 5. *, P<0.05, **, P<0.01, t test. (C) Western blot analysis (left panel) and quantification (right panel) of LC3 and p62 levels in skeletal muscle from Becn1+/+ and Becn1FA/FA mice injected with one dose of PBS or 50 mg/kg lysosomal inhibitor chloroquine. The autophagy flux is measured by the difference in the p62 and LC3 levels between mice injected with PBS and with chloroquine. Results represent mean ± s.e.m. N = 3. **, P<0.01, two-way ANOVA for comparison of magnitude of changes between different groups in mice of different genotypes.
Fig 3.
Becn1FA/FA mutation ameliorates cerebral Aβ accumulation, memory deficits and mortality of Alzheimer’s mouse models.
(A-B) Dot-blot assays and quantification of soluble (A) and insoluble (B) Aβ42 levels in homogenated brain samples of 6-month old 5XFAD Becn1+/+ and 5XFAD Becn1FA/FA mice, immunostained with anti-Aβ42 antibody. Total protein loading was labeled by Ponceau S. Triplicate experiments from 6 mice in each group were shown. ***, P<0.001, t test. (C) ELISA analyses of total (soluble and insoluble) Aβ42 levels in the cortex of 6-month old 5XFAD Becn1+/+ and 5XFAD Becn1FA/FA mice. N = 7. (D) Representative images (upper) and quantification (lower) of amyloid plaques stained by Thioflavin S in brain (cortex and hippocampus) of 5XFAD Becn1+/+ and 5XFAD Becn1FA/FA mice. Magnification of the enclosed regions is shown on the right. N = 6–8. Scale bar: 500 μm. ***, P<0.001, t test. (E) Dot-blot assays and quantification of total Aβ42 levels in homogenated brain samples of 6-month old 5XFAD Becn1FA/FA mice treated with the autophagy inhibitor SBI-0206965 or vehicle once per day for 7 days. Total protein loading was labeled by Ponceau S. N = 5. **, P<0.01, t test. (F) Morris water maze test of 6-month old WT, 5XFAD Becn1+/+ and 5XFAD Becn1FA/FA mice. Escape latency and total distance traveled in visible platform test and hidden platform test are shown. N = 11–20. Results represent mean ± s.e.m. Two-way repeated measures ANOVA.
Fig 4.
Becn1F121A improves the survival rate of PDAPP AD mice.
Kaplan-Meier survival curve of PDAPP mice with normal (PDAPP, N = 51), hyperactive (PDAPP Becn1FA/FA, N = 34), or deficient (PDAPP Bcl2AAA, N = 33) autophagy monitored over time for 9 months. Statistical significance was analyzed by the log-rank test.
Fig 5.
Autophagosomal sequestration of Aβ42 in brain of autophagy-hyperactive mice.
(Left) Scheme of immunoisolation of autophagosomes from brain of 12-week old 5XFAD Becn1FA/FA mice expressing GFP-LC3. Briefly, post-nucleus extracts of the brain lysates was obtained by centrifugation at a low speed of 1,000 xg. Autophagosomes were enriched by centrifugation at a high speed of 20,000 xg, and pulled down by an anti-GFP antibody using magnetic beads. (Right) Western blot detection of Aβ42 fibrillar and oligomeric species inside autophagosomes immunoprecipitated by GFP antibody as in the scheme. A known autophagy cargo p62 serves as a positive control, and a cytosolic enzyme GAPDH is a negative control.
Fig 6.
An autophagy-inducing compound ML246 reduces amyloid load in an autophagy-dependent manner in vitro and in vivo.
(A) Chemical structure of ML246. (B) Dot-blot assays (left) and quantification (right) of secreted Aβ42 levels in conditioned media of HEK293 cells stably expressing APP treated with vehicle (DMSO) or ML246 for 24 h, immunostained with anti-Aβ42 antibody. Cells were transfected with non-targeting control (NC) or ATG7 siRNA 24 h prior to ML246 treatment. Results are quantified from 4 independent experiments. (C) Representative images (left) and quantification (right) of TUNEL signals (red) in WT primary cortical neurons treated with conditioned media from (B) for 24 h. Nuclei were stained with DAPI. Scale bar, 100 μm. N = 10 fields (each field containing 20–30 neurons). (D, E) Representative images (upper) and quantification (lower) of dot-blot assays on soluble (D) and insoluble (E) Aβ42 levels in brain samples of 6-month old 5XFAD and 5XFAD Becn1+/- KO mice after 5 weeks of ML246 treatment, immunostained with anti-Aβ42 antibody. Total protein loading was labeled by Ponceau S. Triplicate experiments from 4–5 mice in each group were shown. Results represent mean ± s.e.m. NS, not significant; *, P<0.05; **, P<0.01; ***, P<0.001, t test.
Fig 7.
ML246 and voluntary exercise decrease cerebral amyloid plaques and ameliorate memory deficits in 5XFAD AD mice.
(A, B) Representative images (upper) and quantification (lower) of dot-blot assays on soluble (A) and insoluble (B) Aβ42 levels in brain samples of 6-month old 5XFAD and 5XFAD Becn1+/- KO mice after 4 months of voluntary running, immunostained with anti-Aβ42 antibody. Total protein loading was labeled by Ponceau S. Triplicate experiments from 4–5 mice in each group were shown. (C) Representative images (left) and quantification (right) of amyloid deposits stained by Thioflavin S in brain of 6-month old 5XFAD mice, and 5XFAD mice subject to 5 weeks of ML246 treatment or 4 months of voluntary exercise. Scale bar: 500 μm. Results represent mean ± s.e.m. N = 6–8. *, P<0.05; **, P<0.01, t test. (D) Morris water maze test of 6-month old WT, 5XFAD, and 5XFAD mice after 5 weeks of ML246 treatment or 4 months of voluntary running. Escape latency and total distance traveled in visible platform test and hidden platform test are shown. N = 10–11. Results represent mean ± s.e.m. Two-way repeated measures ANOVA.