Enhancing autophagy maturation with CCZ1-MON1A complex alleviates neuropathology and memory defects in Alzheimer disease models.

Rationale: Impairment of autophagy maturation has been implicated in Alzheimer's disease (AD) pathogenesis. However, the mechanism for this impairment has not been elucidated, and whether enhancing autophagy maturation is a viable therapeutic strategy for AD has not been verified.
Methods: We examined the autophagosome maturation process in AD cell and mouse models by immunoblotting. To further understand the changes in autophagy in AD brains, we analyzed the transcriptome by RNA-sequencing and measured the expression of RAB7, CCZ1 and MON1A. We performed brain stereotaxic injections of AAV into 3xTg AD mouse brain and WT mouse brain to over-express MON1A/CCZ1 or knockdown MON1A. For in vitro studies, we purified autophagosomes, and determined GTP-RAB7 level in autophagosome fractions by GST-R7BD affinity-isolation assay.
Results: We report that the active form of RAB7 was selectively decreased in autophagosome fractions isolated from cells and tissues of AD models, and that this decrease was accompanied by impaired activity of its guanine nucleotide exchange factor (GFE) CCZ1-MON1A. Overexpressing CCZ1-MON1A increased the active form of RAB7, enhanced autophagosome maturation, and promoted degradation of APP-CTFs, Aβ and P-tau in an autophagy-dependent manner in cells and a mouse AD model. Conclusions: Our data reveals that CCZ1-MON1A-RAB7 complex dysfunction is a potential mechanism for autophagosome maturation defects in AD, and advances the possibility that enhancing autophagosome maturation is a novel therapeutic strategy against AD. © The author(s).

Introduction

Alzheimer's disease (AD) is the most common neurodegenerative disorder worldwide. It is characterized by two main histological hallmarks: amyloid plaques composed of Aβ peptides, and neurofibrillary tangles (NFTs) formed by hyperphosphorylated tau protein 1, 2. Aβ is generated through cleavage of APP by β- and γ-secretases. APP is first cleaved by BACE1/β-secretase, which generates soluble sAPPβ and membrane-bound C99 APP-CTFβ. Subsequently APP-CTFβ is cleaved into Aβ1-40 or Aβ1-42 by γ-secretase 3. Tremendous efforts have been spent on the development of therapies targeting Aβ and P-tau for AD treatment; however, these attempts have largely failed. There is a huge need for the discovery of novel therapeutic strategy(s) to slow disease progression. Macroautophagy (henceforth referred to as autophagy) is an evolutionarily conserved pathway that regulates the degradation of cytoplasm materials including long-lived proteins, damaged cellular organelles and aggregated or misfolded proteins 4. Autophagy features the formation of double-membrane autophagosomes that fuse with lysosomes for degradation. It can be divided into three stages: initiation, extension, and maturation. Massive accumulation of abnormal autophagosomes in neurons has been observed in the brains of AD patients 5, implicating impairment of autophagy, especially at the autophagosome maturation stage, in the pathogenesis of AD 6. Furthermore, increasing evidence has revealed that disruption of the autophagy-lysosome pathway occurs very early in disease progression 7. Inger Lauritzen demonstrated a link between accumulating C99 and the endosomal-autophagic-lysosomal pathology. They found that C99-mediated autophagy impairment is independent of Aβ 8. Similarly, Loan Vaillant-Beuchot found that APP-CTF accumulation also impaired autophagy/mitophagy as shown by the accumulation of LC3-II and SQSTM1/p62, impaired Parkin and PINK1 recruitment to mitochondria, up-regulation of membrane and matrix mitochondrial proteins, and defective fusion of mitochondria with lysosomes 9. However, the mechanisms underlying autophagy impairment in AD are not fully understood.

Modulation of the autophagy pathway has been recognized as a potential therapeutic strategy for treating neurodegenerative diseases including AD. Many studies have been performed to test whether enhancement of autophagy by genetic or pharmacological approaches can lead to improvement of AD phenotypes in AD models 10-12. However, autophagy is a double-edged sword in AD development, in that simply enhancing autophagosome induction without restoring the autophagosome maturation defect may have detrimental consequences for AD 13. A wiser strategy for modulating autophagy in AD may be specifically targeting the autophagosome maturation process. However, the feasibility and efficacy of this strategy has not yet been verified. RAB7 is a small GTPase required for both endosome and autophagosome maturation 14. CCZ1-MON1A complex is the RAB7 guanine nucleotide exchange factor (GEF) which promotes the exchange of GDP-RAB7 (inactive state) to GTP-RAB7 (active state) 15, 16. Krisztina Hegedus found that, in starved drosophila fat cells, autophagosomes accumulate due to impaired fusion with lysosomes upon loss of the CCZ1-MON1A-RAB7 module. They also found that CCZ1-MON1A recruits RAB7 to autophagosomes and regulates autophagosome-lysosome fusion in these cells 17. The yeast homolog of the members of the mammalian LC3 protein (Atg8) directly binds to CCZ1 and is indeed a primary determinant to recruitment of CCZ1-MON1A to autophagosomes 18. Our previous research showed that over-expression of the constantly active form of RAB7 rescued autophagosome maturation defects in a cellular AD model lacking autophagy gene NRBF2. Meanwhile, our data revealed that APP interacts and co-localizes with CCZ1 and MON1A in N2S cells 19. These data highlight the role of CCZ1-MON1A-RAB7 in regulating autophagosome maturation for efficient degradation of AD-associated proteins. We hence propose that the CCZ1-MON1A complex is a key determinant of autophagosome maturation in AD and that enhancing this complex's activity could be an effective approach to facilitating autophagosome maturation in AD models for the alleviation of AD pathologies.

In this paper, we report the reduction of the active form RAB7 in the autophagosome fraction isolated from cellular and mice AD models, as well as the inhibition of CCZ1-MON1A GEF activity in cellular and mice AD models. Our results verify the effects of CCZ1-MON1A in regulating autophagosome maturation, in degrading AD-related protein, and in improving neuropathological and behavioral performance in AD models. Our results suggest that CCZ1-MON1A, via activating autophagosome maturation, facilitates AD-related protein degradation and alleviates memory defects and neuropathological abnormalities in AD. Our findings provide evidence that enhancement of autophagosome maturation is of potential therapeutic value in AD treatment.

Results

AD cell and mice models exhibit inhibition of autophagosome maturation

Autophagy dysfunction has been observed in the brain tissue of AD patients as well as in several AD models 20, 21. To provide a comprehensive analysis of the autophagic activity in both Aβ- and P-tau-based AD models, we examined the autophagosome maturation process in N2a mouse neuroblastoma cells stably expressing the human Swedish mutant APP695 (N2S cells) (), in a 3xTg AD mice (APP Swedish, MAPT P301L, and PSEN1 M146V) model () and in a 5xFAD model (M146L and L286V mutations in PSEN1, and Swedish (K670N/M671L), Florida (I716V), and London (V717I) mutations in APP) (). We found that both LC3-II and SQSTM1 accumulation in 3xTg AD mice (Figure ), 5xFAD mice (Figure ) and N2S cells (Figure ). Immunofluorescence staining showed massive accumulation of SQSTM1 in N2S cells. Induction of autophagy by HBSS reduced SQSTM1 puncta effectively in N2a cells but less efficiently in N2S cells (Figure ). Interestingly, almost all SQSTM1 puncta co-localized with LC3 puncta in N2S cells (), indicating the inhibition of autophagic degradation. The RFP-GFP-LC3 probe has been widely used to monitor autophagy flux 22, based on the fact that GFP is more rapidly quenched than RFP in acidic environments in lysosomes. We used the RFP-GFP-LC3 probe to monitor autophagy flux in N2S cells and found that the number of yellow puncta (both GFP- and RFP-positive) increased (Figure ) while the number of red-only puncta decreased in N2S cells (Figure ). Meanwhile, we found that colocalization of autophagosome marker LC3 and lysosome marker LAMP1was decreased in N2S cells (Figure ). Taken together, these findings indicate that autophagosome maturation is impaired in AD cell and mouse models.

Active form of RAB7 in autophagosome fractions and MON1A-CCZ1 GEF activity are decreased in AD cell and mice models

Our and others previous studies have revealed the critical role of RAB7 in regulating autophagosome maturation 19. To investigate how genes involved in the CCZ1-MON1A-RAB7 complex are affected in AD, we examined their transcriptional changes in four brain regions of AD patients and controls from the Mount Sinai Brain Bank (MSBB) cohort (PMID) 23. Both MON1A (Rho = -0.19, adj.p = 0.01) and RAB7 (Rho = -0.22, adj.p = 0.003) mRNA expression demonstrated a negative correlation with CDR score only in the parahippocampal gyrus (BM36) which is particularly vulnerable to AD (Figure -B), but not in other brain regions, indicating a region-specific impact of AD on these genes. CCZ1 expression was not significantly affected by AD in any of the four brain regions. The data suggest progressive declines of CCZ1-MON1A GEF activity and of the active form of RAB7 might occur during the course of AD in the parahippocampal gyrus region.

To understand whether RAB7 activity is impaired in AD cell and mouse models, we applied the GST-tagged RAB7-binding domain (GST-R7BD) of RILP 19, 24 and GTP beads 19, 25, which selectively bind GTP-binding proteins, to monitor the level of the GTP form of RAB7 (GTP-RAB7) in autophagosome fractions (). We found that GST-R7BD pulled down much less GTP-RAB7 in autophagosome fractions isolated from N2S cells and from the hippocampus tissue of 3xTg AD mice, compared with that isolated from N2a cells and WT mice (Figure ), without affecting the total RAB7 protein level. Consistently, GTP beads also pulled down less GTP-RAB7 in autophagosome fractions isolated from N2S cells and hippocampus tissue of 3xTg AD mice (). Interestingly, the total amount of GTP-RAB7 increased rather than decreased in N2S cells and hippocampus tissue of 3xTg mice (), indicating impaired autophagosome maturation may trigger a compensation mechanism in AD models.

CCZ1-MON1A complex is the GEF for RAB7 and has been reported to interact with Atg8 (mammalian homolog LC3) and to localize on autophagosomes in yeast 18. We confirmed that the CCZ1-MON1 complex co-localizes with autophagosome marker LC3 in N2a cells (Figure ) and interacts with LC3 in N2a cells and mice brain tissue (Figure ). To understand whether reduced GTP-RAB7 on autophagosomes is due to impaired CCZ1-MON1 complex function, firstly, we determined the localization of CCZ1-MON1 complex on autophagosomes in N2S cells by confocal microscopy. Secondly we immune-purified the CCZ1-MON1 complex from N2S cells, 3xTg mice and 5XFAD mice models to perform in vitro GEF assays 17, 19. As expected, colocalization between GFP-CCZ1/GFP-MON1A and RFP-LC3 was reduced in N2S cells (Figure ), and the CCZ1-MON1 complex GEF activity was decreased in N2S cells (Figure ), 3xTg mice (Figure ) and 5XFAD mice (Figure ) models.

Our previous work revealed that CCZ1 interacts with PIK3C3, which produces PtdIns3P to regulate autophagosome formation and maturation 19. PtdIns3P has been reported to recruit the CCZ1-MON1A complex to regulate RAB7 activation 26. We wondered whether VPS34-associated CCZ1-MON1A GEF activity was reduced in AD mouse models. CCZ1-MON1A protein immunopurified by VPS34 antibody was used to perform the GEF assay. From the results we can see PIK3C3-associated CCZ1-MON1A GEF activity was markedly decreased in 3xTg AD mouse brain tissue (Figure ). Consistently, the CCZ1-associated PIK3C3 kinase activity is reduced in 3xTg AD mice (Figure ). Collectively, the above results indicate that the active form of RAB7 (GTP-RAB7) is reduced specifically on the autophagosomes in AD cell and mice models due to impaired CCZ1-MON1A GEF activity.

CCZ1-MON1A is activated during autophagy and positively regulates autophagosome maturation

Our data show that CCZ1-MON1A puncta partially co-localized with autophagic structures. However, the effect of the CCZ1-MON1A complex on autophagy has not been fully explored. By using GST-R7BD affinity-isolation assay as described previously, we found that GTP-RAB7 was significantly increased on autophagosomes in starvation-induced autophagy conditions, without affecting total RAB7 protein level (Figure ). PIK3C3-associated GEF activity and CCZ1-associated VPS34 kinase activity both increased in starvation-induced autophagy conditions (Figure ). More interestingly, we found that co-localization between CCZ1-MON1A and LC3 was increased in starvation-induced autophagy conditions (Figure ). To understand the effect of CCZ1-MON1A on autophagy, we over-expressed CCZ1-MON1A in N2a cells and observed increased GTP-RAB7 as expected (Figure ). We applied an RFP-GFP-LC3 probe to examine autophagic flux in CCZ1-MON1A over-expression cells (Figure ). Compared with the empty vector group, CCZ1-MON1A over-expression led to increase of yellow puncta (Figure ), of red-only puncta (Figure ), and of the ratio of red-only puncta to yellow puncta (Figure ), indicating the enhancement of autophagosome maturation.

To confirm the function of CCZ1-MON1A in regulating autophagic degradation in cells, we examined the levels of the autophagy substrate SQSTM1 in N2a (Figure ) and N2S cells (). We found that CCZ1-MON1A over-expression significantly reduced SQSTM1 protein levels and SQSTM1 puncta in N2a and N2S cell (Figure ). At the same time, we showed that over-expression of CCZ1 or MON1A alone did not affect SQSTM1 level (). Finally, we confirmed that the CCZ1-MON1A over-expression-mediated degradation of SQSTM1 is dependent on autophagy as the effect can be blocked by Atg5 SiRNA transfection or CQ treatment (Figure ).

In addition, we knocked down (KD) MON1A expression in N2a cells by using Mon1a shRNA and found that Mon1a KD significantly decreased GTP-RAB7 level and increased LC3-II and SQSTM1 levels (Figure ). Autophagosome maturation in N2a and Mon1a KD N2a cells was determined by the RFP-GFP-LC3 (Figure ). The number of total puncta increased (Figure ) while the number of red-only puncta (Figure ) decreased in Mon1a KD cells. Immunofluorescence staining showed the number of SQSTM1 puncta significantly increased in Mon1a KD N2a cells (Figure ). We enhanced autophagy flux by starving cells with HBSS which enhanced the degradation of SQSTM1 in N2a cells but not in N2a KD Mon1a cells (Figure ). Taken together, these results demonstrate that CCZ1-MON1A positively regulates autophagosome maturation in neuronal cells.

CCZ1-MON1A positively regulates autophagic degradation of APP-CTFs, Aβ and P-tau

To understand whether impaired CCZ1-MON1A-RAB7 module function is responsible for the accumulation of AD-related protein APP-CTFs, Aβ and P-tau in AD models, we overexpressed CCZ1-MON1A complex in N2S cells. We found that over-expression of CCZ1-MON1A significantly reduced APP-CTF (Figure ) and Aβ levels (Figure ). Meanwhile, we examined the effect of CCZ1-MON1A over-expression on inducible HEK 293 cells expressing pTRE3G-mcherry-BI promoter-EGFP Tau P301L (HEK 293 3G-EGFP-Tau P301L/mcherry) and found that over-expression of CCZ1-MON1A significantly reduced the GFP-P-tau level (). To further confirm the function of CCZ1-MON1A in APP-CTFs degradation and Aβ production, we generated Mon1a knockdown (KD) N2S cell using shRNA-Mon1a (Figure ). KD of Mon1a led to dramatic accumulation of APP-CTFs and Aβ in N2S cells (Figure ). These results indicate that CCZ1-MON1A is an important determinant of APP-CTFs, Aβ and P-tau contents. Interestingly, the increase of CTFs levels are more dramatic than Aβ levels in Mon1a knockdown (KD) N2S cells, suggesting that there is a difference in the fate of the two catabolites and that additional proteases may target accumulated Aβ for degradation.

To better understand how CCZ1-MON1A mediates autophagy and promotes APP-CTFs degradation, we examined whether CCZ1-MON1A would recruit APP/APP-CTFs into autophagosomes and lysosomes, as previous data showed that CCZ1-MON1A interacts with both LC3 (Figure ) and APP 19. As the results show, the co-localization of APP/APP-CTFs with autophagosome marker LC3 and lysosome marker LAMP1 was significantly increased in CCZ1-MON1A-overexpressed cells (Figure ). Interestingly, over-expression of CCZ1-MON1A significantly decreased the percentage of APP/APP-CTFs present in the RAB5-positive early endosomes while it increased distribution of APP/APP-CTFs in the RAB7-positive late endosomes (), indicating that CCZ1-MON1A also promoted endocytic maturation of APP/APP-CTFs. These results suggest that over-expression CCZ1-MON1A promotes recruitment of APP and APP-CTFs into autophagic-lysosome structures.

CCZ1-MON1A regulates APP-CTFs and Aβ levels through autophagy

We and others have found that enhanced autophagy promotes Aβ degradation 27. Our previous data showed that CCZ1-MON1A is a positive regulator of autophagy and promotes degradation of APP-CTF and Aβ. To determine whether CCZ1-MON1A-mediated APP-CTF degradation is dependent on autophagy, we treated CCZ1-MON1A over-expression cells with chloroquine (CQ) and SAR405 to inhibit autophagy (Figure ). We found that the CCZ1-MON1A over-expression-mediated reductions of APP-CTFs and Aβ were abolished by CQ and SAR405 treatment in N2S cells (Figure ). Furthermore, KD of Atg5 and Atg7, two important proteins for autophagosome initiation, blocked CCZ1-MON1A over-expression-induced APP-CTFs degradation (Figure ). These results indicate that autophagy is required for CCZ1-MON1A-mediated degradation of APP-CTFs and Aβ.

AAV-mediated CCZ1-MON1A over-expression in the hippocampus alleviated autophagy impairment and AD-related behavioral and neuropathological changes in 3xTg AD mice

To confirm the effects of CCZ1-MON1A in regulating AD-related behavioral and neuropathological changes, AAV9-Flag-ccz1 /AAV9-Flag-mon1a (AAV-CM) or empty vector AAV-FLAG (AAV-VE) were stereotaxically injected into both sides of the hippocampus of 3xTg AD mice and WT C57 mice 28. There were nine mice in each group. Two months after injection, the mice were subjected to spatial memory testing, and their brains were used for biochemical and neuropathological analyses. Immunostaining of the brain slices revealed that injection of the virus into the DG regions led to highly restricted hippocampal expression ().

The Morris water maze test (MWM) 29 was used to evaluate the spatial memory of 3xTg mice receiving AAV-CM or AAV-FLAG injections. The 3xTg AD mice receiving AAV-CM injections travelled a significantly shorter time to find the hidden platform compared with those receiving AAV-Flag treatment, while WT mice receiving AAV-CM injections did not display any alternation in the time to find the hidden platform (Figure ). In the probe trial, the CM-AAV-injected 3xTg AD mice spent a significantly greater percentage of time in the target quadrant when compared to the vehicle-treated 3xTg mice (Figure ). While the CM-AAV-injected WT mice showed the similar time spent in the target quadrant compared to the vehicle-treated WT mice (Figure ). We next investigated short-term memory using the Y-maze text 30. 3xTg AD mice receiving CM-AAV injection more frequently entered the novel arm and spent more time there (Figure ). These results suggest that over-expression of CCZ1-MON1A in the hippocampus selectively improved cognitive function of 3xTg AD mice.

Anxiety is one of the clinical features of AD patients, and 3xTg AD mice also display anxious behavior 31. We used an open field test to test whether CCZ1-MON1A over-expression might exert anxiolytic effects. We found that over-expression of CCZ1-MON1A significantly increased the time in the center area (Figure ). This data suggested that over-expression of CCZ1-MON1A decreased anxiety in 3xTg AD mice.

Next, we examined autophagy status. We found that over-expression of CCZ1-MON1A significantly increased the protein level of LC3-II and reduced SQSTM1 level in 3xTg AD mice and WT mice (Figure ), suggesting that autophagy was enhanced in CCZ1-MON1A-injected 3xTg AD mice and WT mice. Furthermore, over-expression of CCZ1-MON1A decreased the levels of FL-APP, APP-CTFs (Figure ), and intracellular Aβ1-42/Aβ1-40, and of the Aβ plaque load significantly (Figure ). The levels of phosphorylated tau (S205/T202, S214, T181) also dramatically decreased in CCZ1-MON1A over-expression mice. (Figure ). Conversely, AAV-mediated Mon1a KD exacerbated autophagy impairment in 3xTg AD mice, and AD-related behavioral and neuropathological changes in 3xTg AD mice. AAV-GFP-sh-Mon1a injection aggravated the spatial memory impairment and anxiety as shown by the MWM, Y-maze and open field test results (). The injection also inhibited autophagy as evidenced by accumulation of LC3 and SQSTM1 (), resulting in significant accumulation of FL-APP, APP-CTFs, Aβ plaques, intracellular Aβ1-42/Aβ1-40 and P-tau (). Collectively, CCZ1-MON1A over-expression in the hippocampus improved cognitive and emotional performance, restored autophagy function, and alleviated neuropathology in a 3xTg AD mouse model, indicating that CCZ1-MON1A over-expression is neuroprotective in this model.

Material and Methods

Reagents

Antibodies for Immunoblotting and IP, anti-ACTB/β-actin (sc-47778), and anti-CCZ1 antibody (sc-514290), were purchased from Santa Cruz Biotechnology. Anti-flag antibody (14794), anti-GFP (2956), anti-GST antibody (2622), anti-RAB7 antibody (9467), ATG5 antibody (9980), ATG7 antibody (8558), anti-rabbit IgG (7074), anti-mouse IgG (7076) and normal rabbit IgG (2729) were purchased from Cell Signaling Technology. Anti-LC3B (NB100-2220) antibody and anti-MON1A (NBP1-52007) were purchased from Novus Biologicals. Anti-hVPS34 antibody (Z-R016 and Z-R015) was purchased from Echelon Bioscience. Anti-SQSTM1/p62 (P0067) was purchased from Sigma-Aldrich. For immunofluorescence studies, Alexa Fluor® 488 goat anti-rabbit/mouse IgG (A-11008, A-11001), Alexa Fluor® 555 goat anti-rabbit/mouse IgG (A-21429, A-21424) and Alexa Fluor® 647 goat anti-rabbit/mouse IgG (A-21244, A-21235) were purchased from Thermo Fisher Scientific. For ELISA, anti-Aβ1-16 monoclonal antibody (6E10, 803001) and anti-beta amyloid polyclonal antibody (CT695, 51-2700) were purchased from Thermo Fisher. Anti-beta-amyloid 1-42 antibody (AB5078P) was purchased from Millipor. Anti-beta-amyloid 1-40 antibody (0060-100/bA4(40)-5C3) was purchased from Nanotools. DMEM (11965-126). Dynabeads® protein G for immunoprecipitation (10003D), fetal bovine serum (10270-106), opti-MEM (31985-070), G418 (10131-035), lipofectamine™ 3000 Transfection Reagent (L3000008), mant-GDP (M12414) and GTP (18332015) were purchased from Thermo Fisher Scientific.

Cell culture and transfection

N2S (N2a stably transfected with human Swedish mutant cell) cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal bovine serum and 200 ug/ml G418, in a humidified atmosphere of 5% CO2 at 37 °C. Hela-GFP-LC3 cells and N2a cells were cultured in DMEM containing 10% fetal bovine serum in a humidified atmosphere of 5% CO2 at 37 °C. Cells were transfected with plasmids using Lipofectamine® 3000 Transfection Reagent (L300008) from Thermo Fisher according to the manufacturer's instructions. 24-72 h after transfection, cells were analyzed.

Immunostaining

Cell lines were trypsinized and transferred to 24-well dishes containing coverslips followed by transient transfection with indicated plasmids for 24-72 h. Then cells were fixed using ice-cold formaldehyde for 10 min. Cell were then permeabilized in 0.2% Triton X-100 (Sigma, T8787). Cells were incubated with primary antibodies at 4 °C overnight, washed with PBS buffer and then incubated with secondary antibodies for 2 h at room temperature. Cells were then washed 3 times with PBS and stained with DAPI at room temperature for 5 min. Slides were examined with a confocal microscope.

Immunoblotting

Cells or tissues were homogenized in ice-cold RIPA buffer (Cell Signaling Technology, 9806) with complete protease inhibitor mixture (Roche Applied Science, 04693124001). The proteins were separated by 10-15% SDS-polyacrylamide gel electrophoresis and transferred to PVDF membranes. The membranes were blocked with 5% nonfat milk dissolved in TBST for 2 h and incubated with primary antibodies overnight at 4 °C. Finally, the blots were incubated with secondary antibody 2 h at room temperature. Blots were visualized using ECL kit (GE Healthcare Life Sciences, RPN2235).

MON1, RAB7 mRNA analysis

Gene expression was examined at the mRNA level in multiple brain regions of healthy control and AD patient samples from the Mount Sinai Brain Bank (https://www.synapse.org/#!Synapse:syn3157743). Spearman correlation analysis was performed to examine the relationship between gene expression and CDR score with Benjamini-Hochberg correction for multiple testing.

Purification of autophagosomes from cells and mouse brains

Cells or brain tissue was homogenized in homogenization buffer (0.25 M sucrose, 10 mM HEPES, 1 mM EDTA, protease inhibitors pH 7.5), then centrifuged at 1000 g for 5 min. Lysosomes and nuclei were removed by glycyl-phenylalanine-naphthylamide (GPN). We removed mitochondria using different concentrations of Nycodenz. The final autophagosome proteins were extracted using ice-cold RIPA buffer (Cell Signaling Technology, 9806) 32.

GTP-agarose pull-down for RAB7 activity assay

GTP-RAB7 was measured by GTP-agarose beads (Sigma, G9768) following a published protocol 24. Briefly, cells or animal brains were collected and lysed in a buffer containing 50 mM Tris-HCl pH 7.5, 5 mM MgCl2, 0.5% Triton X-100, 250 mM NaCl, and protease inhibitors. The solution was then incubated with GTP-beads overnight at 4 °C. The beads were washed with lysis buffer and boiled in SDS-loading buffer before SDS-PAGE for immunoblotting.

GST-R7BD pull-down assay

GTP-Rab7 was purified using GST-R7BD bait. Mammalian cells were lysed by the lysis (pull down) buffer (100 mM NaCl, 20 mM HEPES, 1% Triton X-100, 5 mM MgCl2, protease inhibitors). Each pull-down was performed in 1 mL with 500 µg of cell lysate and 100 µL of beads pre-equilibrated in pull-down buffer. Beads were rotated overnight at 4 °C, and washed twice with cold pull-down buffer. Bound proteins were eluted by adding 1× sample buffer, and then incubated at 95 °C for 8 min. The amount of GTP- RAB7 was measured by blotting with a RAB7-specific antibody and appropriate secondary antibody.

GEF assay

GEF assays were performed as follows. Mant-GDP-bound RAB7 was diluted to a concentration of 50 nM in exchange buffer [20 mM HEPES (pH 7.5), 0.5 mM MgCl2 and 150 mM NaCl 15. The dissociation of mant-GDP from RAB7 was determined by measuring the decrease in fluorescence signal that accompanies release of mant-GDP in the presence of a large excess of GTP. Nucleotide release reactions were initiated by addition of GTP (1 mM final concentration) to the reaction buffer. The reaction buffer was then incubated with purified CCZ1-MON1A from cell or mouse brain. Samples were excited at 360 nm, and the emission was monitored at 440 nm.

PIK3C3/VPS34 kinase assay

PIK3C3/VPS34 kinase activity was determined by Class Ⅲ PI3K ELISA kit (Echelon, cat: K-3000). Briefly, VPS34 protein was immunoprecipitated by VPS34 antibody or CCZ1 antibody. Prepare PI for kinase reaction. The kinase reaction was allowed to proceed for 2 h at 37 °C. The kinase reaction was quenched by adding EDTA. This solution was then incubated with PI3P detector at room temperature for 1 h. The solution was discarded, each plate was washed 3 times with TBST. Secondary detector was added, and solutions were incubated for 30 min at room temperature. The solution was again discarded, and each plate was washed 3 times with TBST. TMB solution was added, and incubated for 30 min at room temperature. H2SO4 was added to stop the reaction. Absorbance values were read at 450 nm.

ELISA assay of Aβ1-40 and Aβ1-42

The brain tissues and cells were lysed in lysis buffer (Cell Signaling Technology, 9806). ELISA plates (Nunc) were coated with 6E10 antibody, and incubated at 4 °C overnight. Plates were washed 3 times with PBST and blocked with Block Ace for 2 h at room temperature. Plates were again washed with PBST 3 times. Samples or standards were incubated at 37 °C for 2 h. Plates were washed with PBST 3 times. Secondary antibody was added, and incubated for 2 h at 37 °C. Plates were washed 4 times with PBST. Streptavidin-conjugated horseradish peroxidase was prepared in a ratio of 1:4000 in assay diluent, and dispensed 100 μl/well. Wells were incubated for 1 h at 37 °C. The wells were washed 4 times with PBST. A solution of Buffers A and B in 1:1 ratio was prepared and added, 100 μl/well; wells were incubated for 30 min in dark. H2SO4 was added to stop the reaction. Absorbance at 450 nm was read and recorded.

Mice feeding

3xTg AD mice and C57 mice were housed in individually ventilated cages with standard rodent bedding. The room temperature was kept at approximately 22 °C and the relative humidity between 40 and 70%. Mice were housed under constant light-cycle (12 h light/dark). All experiments were carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the University of Macau research ethics committee (Approval No. UMARE-013-2019).

Stereotactic injection of adeno-associated virus

Adeno-associated virus (AAV) was stereotactically, injected bilaterally into 3xTg AD mouse hippocampi as described. AAV-Flag-ccz1, AAV-Flag-mon1a, AAV-Flag, AAV-GFP-sh-mon1a and AAV-GFP-sh-Ctrl were stereotactically injected into the hippocampi of 6-month-old 3xTg AD mice at the following coordinates: anterior posterior, 2.0 mm; medial lateral, ± 1.7 mm; dorsal ventral, 2.0 mm.

Morris water maze

The Morris water maze was used to evaluate spatial learning memory. The experiments were performed in a 100 cm diameter, 50 cm deep tank filled with opaque water kept at 21 °C. The 10 cm diameter platform was submerged 1 cm under the water surface. Training trials lasted 60 sec, and were performed 4 times a day on 5 consecutive days. The platform location was held constant within each pair of daily tests, but the location of the animal's starting point was changed for each test. To assess long-term spatial memory retrieval, a probe trial (60 s) was conducted 24 h a after the last training session. The time spent in the target quadrant was recorded.

Y-maze test

Mice short-term memory was assessed using the Y maze. There are 3 arms in a Y maze: novel arm, starting arm, and other arm. During the first stage, the novel arm is blocked. Mice entering the starting arm are allowed to move into the open arm for 5 min. After 2 h, the novel arm is opened, starting the second phase. The time spent in each arm, and the number of entries into each arm is recorded.

Open field test

Anxiety-related behavior of the mice was measured by the open field test. A 60 cm× 60 cm× 25 cm box, with a 30 cm× 30 cm outlined center area, was constructed. Each mouse was placed in the center area and left for 5 min. The time spent in the center area and at the margins was recorded.

Statistical analysis

Each experiment was performed at least 3 times, and the results are presented as mean ± SEM. One-way analysis of variance (ANOVA) was followed by the Student-Newman-Keuls test using the Sigma Plot 11.0 software package. A probability value of P< 0.05 was considered to be statistically significant.

Discussion

Accumulating evidence implicates autophagic-lysosomal pathway dysregulation in AD 33. Consistent with this evidence, our findings reported here showed that LC3-II and p62 accumulated in multiple cellular and animal models of AD (Figure 1). Furthermore, our data reveals, for the first time, that the CCZ1-MON1A-RAB7 module is impaired in AD models, suggesting a novel pathogenic mechanism for autophagosome maturation inhibition. We showed that over-expression of CCZ1-MON1A enhanced autophagosome maturation, promoted APP-CTFs, Aβ and P-tau degradation, and alleviated memory defects and neuropathology in AD models. This study highlights the critical role of the CCZ1-MON1A-RAB7 module in the regulation of neuronal autophagy to maintain neuron homeostasis and counteract neurodegeneration.

Figure 1

AD cell and mice models exhibit inhibition of autophagosome maturation. (A) Immunoblotting was used to detect the expression of LC3 and SQSTM1 in the hippocampus region of 12-month-old C57BL/6, 3xTg AD mice. (C-B) Quantitative data in panel Figure 1A. Data are quantified as mean ± SEM (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 vs. the relative control. (D) Immunoblotting was used to detect the expression of LC3 and SQSTM1 in the hippocampus region of 12-mo-old 5xFAD AD mice. (E-F) Quantitative data in panel Figure 1D. Data are quantified as mean ± SEM (n =3). *P < 0.05, **P < 0.01, ***P < 0.001 vs. the relative control. (G) Immunoblotting was used to detect the expression of LC3 and SQSTM1/p62 in the APP over-expressed N2S cells. (H-I) Quantitative data in panel Figure 1G. Data are quantified as mean ± SEM (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 vs. the relative control. (J-K) SQSTM1 levels were determined by immunofluorescence. Quantification data are presented as the mean ± SEM, n = 20-25 cells from 3 independent experiments. ***P < 0.001 vs. the relative control, ns, not significant. Scale bar, 7.5 μm. (L-N) Autophagosome maturation in N2a and N2S cells was determined by the GFP-RFP-LC3 probe. Quantification data are presented as the mean ± SEM, n = 20-25 cells from 3 independent experiments. Scale bar, 7.5 μm. (O-P) N2a and N2S cells were transiently transfected with GFP-Lc3 and mCherry-LAMP1, the colocalization of autophagosomes and lysosomes was visualized under confocal microscope. Quantification data are presented as the mean ± SEM, n = 20-25. *P < 0.05, **P < 0.01, ***P < 0.001 vs. the relative control. Scale bar, 7.5 μm.

The CCZ1-MON1A complex is a RAB7 GEF which promotes the transformation of GDP-RAB7 (inactive state) into GTP-RAB7 (active state). RAB7 plays a key role in endosome and autophagosome maturation by promoting vesicle transportation 15. RAB7 is required for the transfer of the cargo from late endosomes to lysosomes, and RAB7 is activated by the CCZ1-MON1A complex on late endosomes and it dissociates from lysosomes in mammalian cells 34. CCZ1 directly interacts with Atg8 to become localized on autophagosomes 18. However, the role of CCZ1-MON1A in the regulation of autophagy in mammals is still unclear. Here, our results show that, in neuronal cells and mouse brains, CCZ1-MON1A over-expression increased the active form of RAB7 and promoted autophagosome maturation, while Mon1a depletion blocked autophagosome maturation. Our previous study found that APP interacts with CCZ1, and the interaction may contribute to the redistribution of APP and APP-CTF in neuronal cells. Indeed, over-expression of CCZ1-MON1A increased the recruitment of APP/APP-CTFs to autophagosomes and lysosomes, and this phenomenon was more obvious after CQ treatment. These data indicate that CCZ1-MON1A, by generating an active form of RAB7, plays a key role in regulating autophagosome maturation. Interestingly, over-expression of CCZ1-MON1A attenuated Aβ and tau pathology in cellular and mice models of AD, and alleviated memory impairment in 3xTg AD mice. The results suggest that CCZ1-MON1A is a critical factor in the regulation of Aβ and tau pathology, as well as in memory in AD. However, further study is needed to clarify whether CCZ1-MON1A exerts anti-AD effects mainly through autophagy or whether other mechanisms are involved.

Abnormal accumulation of autophagosomes can be observed in AD animal models and AD patients, implying the interruption of autophagosome maturation 5. Several mechanisms have been found to be responsible for the impaired autophagosome maturation, including lysosomal inhibition. Though enhancement of autophagy has been regarded as a therapeutic strategy against AD, simply activating autophagosome biogenesis may be problematic as it will lead to autophagosome accumulation when maturation is inhibited. How to selectively restore impaired autophagosome maturation is still an unsolved problem. By revealing the impaired CCZ1-MON1A GEF activity in AD models and showing that over-expression of CCZ1-MON1A promoted autophagosome maturation and alleviated autophagic substrates accumulation in cellular and mice models of AD, we show that modulation of CCZ1-MON1A expression or CCZ1-MON1A GEF activity can be an approach to restoring autophagosome maturation in AD. A further task would be to identify small molecule CCZ1-MON1A activator(s) for therapeutic applications.

In summary, our study revealed the dysregulation of CCZ1-MON1A-RAB7 in AD models, and showed that CCZ1-MON1A is the positive regulator of autophagosome maturation for efficient clearance of APP-CTF, Aβ and P-tau in AD models. Our work thus establishes an important link between CCZ1-MON1A dysfunction, autophagosome maturation impairment and AD development. More broadly, our study results support the concept that enhancing autophagosome maturation might be a viable therapeutic strategy against AD.

Supplementary figures.

Click here for additional data file.