1. Aβ Secretion and Plaque Formation Depend on Autophagy
Per Nilsson, Krishnapriya Loganathan, Misaki Sekiguchi, Yukio Matsuba, Kelvin Hui, Satoshi Tsubuki, Motomasa Tanaka, Nobuhisa Iwata, Takashi Saito, Takaomi C. Saido 
Cell Reports 
Volume 5, Issue 1, Pages (October 2013)
DOI: /j.celrep
Copyright © 2013 The Authors Terms and Conditions

2. Cell Reports 2013 5, 61-69DOI: (10.1016/j.celrep.2013.08.042)
Copyright © 2013 The Authors Terms and Conditions

3. Figure 1 Aβ Plaque Formation Depends on Autophagy
(A) Immunohistological analysis of Aβ plaque (4G8 antibody) in 20-month-old Atg7flox/flox × APP and Atg7flox/flox; CamKII-Cre × APP mouse brains. Aβ plaque staining was quantified (n = 4, ∗∗p < 0.005).
(B) Aβ ELISA measurements of hippocampal and cortical brain homogenates of 20-month-old Atg7flox/flox × APP and Atg7flox/flox; CamKII-Cre × APP mice (n = 6, ∗p < 0.05, ∗∗p < 0.005).
(C) Brain sections of 6-month-old Atg7flox/flox × APP and Atg7flox/flox; CamKI-Cre × APP mice were immunostained with N1D-Aβ antibody (upper panels) and Aβ40 antibody (lower panels). Inset shows cortical neurons at 40× magnification. Aβ staining in CA1 was quantified (n = 5, ∗p < 0.05).
(D) Aβ ELISA measurements of hippocampal brain homogenates of 6-month-old mice with genotypes as indicated (n = 6, ∗p < 0.05).
(E) APP, APP-CTF, PS1, and BACE1 levels in hippocampal brain homogenates from 6-month-old mice with genotypes as indicated were determined by quantitative western blot (n = 5, no significant difference).
Scale bars represent 500 μm (A) and 100 μm (C). Data are represented as mean ± SEM. See also Figures S1, S2, and S3.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

4. Figure 2 Autophagy Influences Aβ Secretion
(A) Release of endogenous Aβ from Atg7flox/flox and Atg7flox/flox; Nes-Cre cortical/hippocampal primary neurons was determined by ELISA measurements of conditioned media (n = 3, ∗∗∗p < ).
(B and C) Lenti-Atg7 expression in primary neurons activates autophagy and increases release of Aβ. Autophagy activation was measured by monitoring LC3 metabolism by quantitative western blot (n = 6, ∗p < 0.05, ∗∗p < 0.01).
(D) Aβ ELISA measurements of conditioned media from wild-type primary neurons infected with SFV-APP and treated with pharmacological compounds as indicated (n = 4, ∗p < 0.05, ∗∗p < 0.01). Activation and modulation of autophagy were determined by measuring LC3II/I, LC3II/β-actin and p-p70/p70 levels, respectively, by quantitative western blot (n = 6; ∗p < 0.05, ∗∗p < 0.01).
(E) Atg7flox/flox and Atg7flox/flox; Nes-Cre primary neurons were infected with SFV-APP and stained for Aβ (Aβ40 antibody). Three representative neurons per genotype are shown.
Data are represented as mean ± SEM.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

5. Figure 3
Amyloidosis Inhibits Autophagy and Activates Neurodegenerative Processes
(A) Western blot analysis of LC3 in cortical brain homogenates from Atg7flox/flox and Atg7flox/flox × APP mice (representative samples from two individuals per genotype are shown). LC3 immunoreactivity was quantified (n = 5, ∗p < ).
(B and C) Coimmunostaining of Aβ (4G8 antibody) and LC3 (B) and Aβ and p62 (C) using brain sections of 15-month-old Atg7flox/flox × APP mice.
(D and E) Immunostaining for cleaved caspase 3 (D) and RIPK1 (E) of 15-month-old brain sections with genotypes as indicated. The intensities were quantified (n = 5, ∗p < 0.01).
(F) Quantitative western blot analysis of cleaved caspase 3 (n = 3, ∗p < 0.05).
(G) Fluor-jade C staining of 15-month-old brain sections with indicated genotypes. Insets show 40× magnification.
Scale bar represents 50 μm (B), 25 μm (C, upper panel), 4 μm (C, lower panel), and 100 μm (D, E, and G). Data are represented as mean ± SEM.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

6. Figure 4
Amyloidosis Exacerbates Autophagy-Deficiency-Induced Neurodegeneration
(A) Wet weights of dissected brain tissue from 3- and 15-month-old mice with genotypes as indicated (n = 5/genotype).
(B) Representative sections stained by H&E from 15-month-old mice with genotypes as indicated. The cortical and hippocampal thickness were quantified.
(C) Hippocampal volume was calculated from T2 MRI data.
(D) Cell counting of p62-positive neurons in CA1 of 20-month-old Atg7flox/flox; CamKII-Cre and Atg7flox/flox; CamKII-Cre × APP mice.
(E) Morris water maze was performed with 15-month-old littermates with genotypes as indicated (n = 10). The performance on day 5 was analyzed by ANOVA followed by Tukey’s post hoc test.
Scale bar represents 500 μm. Data are represented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < See also Figure S4.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

7. Figure S1
Effects of Neuron-Specific Conditional Knock Down of Atg7, Related to Figure 1
(A–D) Western blot analysis of cortical homogenates (A and B) and immunohistochemical experiments using paraffin embedded brain sections (C and D) revealed lowered LC3 metabolism (n = 5, p∗ < 0.05)) and accumulation of p62- and ubiquitin-positive aggregates in cortex and hippocampus in 3-month-old mice. The proteasome levels were not affected by autophagy deficiency. Interestingly, Atg7 expression is high in hippocampus indicating high autophagic activity in hippocampus. Scale bar represents 500 μm (C, upper panel), 100 μm (C, lower panel and D). Data are represented as mean ± SEM.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

8. Figure S2
Formation of Inclusion Bodies in CA1 of Atg7flox/flox; CamKII-Cre Mice, Related to Figure 1
(A) Hematoxylin and eosin (H&E) and Cresyl violet staining of paraffin embedded brain sections and fresh frozen toluidine blue stained sections revealed IB in the pyramidal cells in CA1 and dendate gyrus (DG) at three and nine months of age respectively.
(B and C) The IBs are positive for p62 and ubiquitin (B) and p-S403-p62 (C).
(D and E) Specificities of antibodies used in the study. Generation of phospho-S403-p62 specific antibody (D). One, ten and 100 ng of phosphorylated and non-phosphorylated peptide, containing the peptide sequence surrounding S403 in p62 (CKKKESLSQMLSMGFSDEGKKK) were spotted on a PVDF membrane and incubated with purified antibody obtained from sera after immunization with the phosphorylated peptide. Antibody purification and dot blot was performed as previously described (Pagans et al., Methods 53, 91). Specificities of Aβ antibodies used in the study (E). Ten μg of delipidated brain homogenate from APP23 mouse was loaded in lane 1 and 3 and one ng of synthetic Aβ40 was loaded in lane 2 and 4. The membrane containing lane 1 and 2 was probed with an antibody raised against N1D-Aβ diluted 1: (right panel, Saido et al., 1994) and reprobed with an antibody recognizing the C-terminal of APP, diluted 1: (left panel, Sigma cat # A8717) and the membrane containing lane 3 and 4 was probed with antibody raised against Aβ40 1: 5000 (right panel, Invitrogen cat # 44348A) and reprobed with anti-C-terminal APP antibody 1: (left panel, Sigma cat # A8717).
Scale bar represents 20 μm.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

9. Figure S3
Increased Astrocytosis in Cortex of 3-Month-Old Atg7flox/flox; CamKII-Cre Mice, Related to Figure 1
(A and B) Immunostaining for GFAP using paraffin embedded brain sections shows increased number of astrocytes (n = 5, ∗p < 0.01) (A) and increased levels of GFAP by Western blot analysis with β-actin used as loading control (B).
(C) Immunohistochemical analysis of 3-month-old paraffin embedded brain sections reveals p62-positive aggregates colocalizing with histone deacetylase 6 (HDAC6) in dendate gyrus but not with IB in CA1. HDAC6 is upregulated in the mossy fibers (MF) of CA3 (n = 5, p∗ < 0.05).
Scale bar represents 100 μm (A) and 250 μm (C). Data are represented as mean ± SEM.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions

10. Figure S4
Amyloidosis Exacerbates Autophagy-Deficiency-Induced Neurodegeneration, Related to Figure 4
(A) Body weight measurements of 3- and 15-month-old Atg7flox/flox, Atg7flox/flox; CamKII-Cre, Atg7flox/flox x APP and Atg7flox/flox; CamKII-Cre x APP mice were performed (n = 5/genotype, ∗p < 0.001, ∗∗p < ). The body weights of Atg7flox/flox; CamKII-Cre and Atg7flox/flox; CamKII-Cre x APP mice are significantly reduced at 15 months of age. Data are represented as mean ± SEM.
(B) Representative T2 MRI sections of 15-month-old Atg7flox/flox, Atg7flox/flox; CamKII-Cre, Atg7flox/flox x APP and Atg7flox/flox; CamKII-Cre x APP mice as indicated. Scale bar 1 mm.
Cell Reports 2013 5, 61-69DOI: ( /j.celrep )
Copyright © 2013 The Authors Terms and Conditions