Cross-linking of cell surface amyloid precursor protein leads to increased β-amyloid peptide production in hippocampal neurons: implications for Alzheimer's disease

Abstract

The accumulation of the β-amyloid peptide (Aβ) in Alzheimer's disease (AD) is thought to play a causative role in triggering synaptic dysfunction in neurons, leading to their eventual demise through apoptosis. Aβ is produced and secreted upon sequential cleavage of the amyloid precursor protein (APP) by β-secretases and γ-secretases. However, while Aβ levels have been shown to be increased in the brains of AD patients, little is known about how the cleavage of APP and the subsequent generation of Aβ is influenced, or whether the cleavage process changes over time. It has been proposed that Aβ can bind APP and promote amyloidogenic processing of APP, further enhancing Aβ production. Proof of this idea has remained elusive because a clear mechanism has not been identified, and the promiscuous nature of Aβ binding complicates the task of demonstrating the idea. To work around these problems, we used an antibody-mediated approach to bind and cross-link cell-surface APP in cultured rat primary hippocampal neurons. Here we show that cross-linking of APP is sufficient to raise the levels of Aβ in viable neurons with a concomitant increase in the levels of the β-secretase BACE1. This appears to occur as a result of a sorting defect that stems from the caspase-3-mediated inactivation of a key sorting adaptor protein, namely GGA3, which prevents the lysosomal degradation of BACE1. Together, our data suggest the occurrence of a positive pathogenic feedback loop involving Aβ and APP in affected neurons possibly allowing Aβ to spread to nearby healthy neurons.

Figure 1.

Neurotoxic and synaptotoxic effects of APP cross-linking in cultured rat PHNs. a, Dose–response curve for 22C11 treatment of PHNs >24 h, showing a significant amount of cell death at 1.0 μg/ml compared with control PHNs treated with PBS. Lower concentrations (100 pg/ml and 100 ng/ml) had no significant neurotoxic effects after 24 h. Cell death was assessed with the commercially available LIVE/DEAD cell viability assay kit and quantified from at least three independent experiments and expressed as the mean ± SEM number of viable cells per field (*p < 0.05). b–f, Synaptic changes in PHNs treated for 24 h with a 100 ng/ml solution of intact 22C11 or isolated Fab fragments compared with control PHNs treated with PBS. Synaptic effects were assessed by a direct count of dendritic spine protrusions in diolistically labeled PHNs and by measuring the protein levels of two dendritic spine markers, PSD-95 and drebrin A. b, Representative images of diolistically labeled PHNs showing a decrease in dendritic spine protrusions in 22C11-treated compared with PBS-treated or Fab-treated PHNs. c, Dendritic spines were quantified from an average of 12 neurons taken from at least three independent experiments and expressed as the mean ± SEM number of spines/μm (**p = 0.009). d, Representative Western blots showing a decrease in the synaptic markers PSD-95 and drebrin A in 22C11-treated compared with PBS-treated PHNs. e, f, Densitometric analysis of PSD-95 and drebrin A levels was quantified from three independent experiments and expressed as the mean ± SEM band intensity (**p = 0.005 for PSD-95; ***p < 0.001 for drebrin A).

Figure 2.

Treatment with 22C11 increases Aβ levels in cultured rat PHNs. Treatment of PHNs for 24 h with a 100 ng/ml solution of 22C11 resulted in a significant increase in Aβ levels compared with control PHNs treated with PBS. Intracellular Aβ levels from PHN lysates were assessed by IP with the commercially available anti-Aβ antibody 4G8 and by ELISA. a, Representative Western blots from IP of Aβ showing an increase in Aβ levels in 22C11-treated compared with PBS-treated PHNs. b, Densitometric analysis of the levels of Aβ was quantified from at least three independent experiments and expressed as the mean ± SEM band intensity (**p = 0.039). c, Aβ_(x-42) levels from 22C11-treated and PBS-treated PHN lysates measured by ELISA were quantified from four independent experiments, expressed as the mean ± SEM signal intensity in arbitrary units (AU) (*p = 0.033).

Figure 3.

Increased Aβ levels in 22C11-treated cultured rat PHNs correlate with increased BACE1 protein levels and not with increased γ-secretase activity. Treatment of PHNs for 8 h with a 100 ng/ml solution of 22C11 did not significantly affect PS1 cleavage compared with PBS-treated PHNs. a, Representative Western blots showing no significant changes in the levels of the N-terminal fragment of PS1 in 22C11-treated PHNs compared with PBS-treated PHNs. b–d, Treatment of PHNs for 8 h with a 100 ng/ml solution of 22C11 did not significantly affect γ-secretase activity in 22C11-treated PHNs compared with PBS-treated PHNs. b, Scheme depicting the in vitro γ-secretase activity assay: solubilized membranes from C100-myc overexpressing HEK293 cells were mixed with solubilized 22C11-treated (100 ng/ml; 8 h) or PBS-treated PHNs and incubated for 2 h, resulting in the formation of AICD-myc fragments detectable by Western blot. c, Representative Western blots showing no observable difference in AICD-myc production in 22C11-treated compared with PBS-treated PHNs. d, Densitometric analysis of the AICD-myc levels was quantified from three independent experiments, expressed as the relative mean ± SEM signal intensity (ns). e–g, Treatment of PHNs for 8 h with a 100 ng/ml solution of 22C11 resulted a significant increase in BACE1 protein levels with a concomitant increase in secreted APPβ fragments. e, Representative Western blots showing increased levels of BACE1 and sAPPβ in 22C11-treated compared with PBS-treated PHNs. f, g, Densitometric analysis of BACE1 levels was quantified from at least three independent experiments and expressed as the mean ± SEM band intensity (***p < 0.001 for BACE1; *p = 0.05 for sAPPβ).

Figure 4.

Increased BACE1 levels in 22C11-treated cultured rat PHNs occur through increased stability of the protein, not through increased synthesis. Cycloheximide (CHX) degradation time course: treatment of PHNs with a 100 ng/ml solution of 22C11 significantly delayed the degradation of BACE1 following CHX treatment, while increasing APP processing. a, Representative Western blots showing the degradation of BACE1 in PHNs at various time points (3, 6, and 18 h) after the addition of CHX (150 μm) alone or CHX (150 μm) plus 22C11 (100 ng/μl). b, Densitometric analysis of BACE1 levels was quantified from at least three independent experiments and expressed as the mean ± SEM signal intensity (*p = 0.05). c, Representative Western blots showing the processing of APP in PHNs at various time points (3, 6, and 18 h) after the addition of CHX alone (150 μm) or CHX (150 μm) plus 22C11 (100 ng/μl). d, Densitometric analysis of the APP levels was quantified from at least three independent experiments and expressed as the mean ± SEM signal intensity. e, Quantification of APP levels expressed as the percentage change after treatment with CHX alone (150 μm) or CHX (150 μm) plus 22C11 (100 ng/μl), indicating increased APP turnover by 22C11 treatment (**p = 0.003 and *p = 0.01 for 6 and 18 h time points, respectively).

Figure 5.

Inhibition of lysosomal hydrolase leads to increased levels of endogenous BACE1 in cultured rat PHNs. Treatment of PHNs for 8 h with various protease inhibitors (chloroquine and NH₄Cl for lysosome; MG132 for proteasome) resulted in a significant increase in BACE1 protein levels in PHNs in which lysosomal degradation was blocked, suggesting that lysosomal degradation is the preferred pathway for BACE1 in PHNs. a, Representative Western blots showing increased levels of BACE1 and APP fragment C99 in PHNs treated for 8 h with the lysosomal inhibitors chloroquine (100 μm) and NH₄Cl (500 μm). b, Densitometric analysis of BACE1 levels was quantified from three independent experiments and expressed as the mean ± SEM band intensity (**p = 0.002 for chloroquine; **p = 0.004 for NH₄Cl). c, Representative Western blots showing BACE1 levels in PHNs treated for 8 h with the proteasomal inhibitor MG132 (1.0 μm). d, Densitometric analysis of BACE1 levels was quantified from three independent experiments and expressed as the mean ± SEM band intensity (NS).

Figure 6.

22C11 treatment of cultured rat PHNs triggers the caspase-3-dependent cleavage of GGA3. a–c, Treatment of PHNs with a 100 ng/ml solution of 22C11 resulted in increased cleavage and activation of caspase-3 after 8 h. a, Representative Western blots showing an increase in cleaved caspase-3 levels after 8 h in 22C11-treated PHNs compared with PBS-treated PHNs. b, Densitometric analysis of cleaved caspase-3 levels was quantified from three independent experiments and expressed as the mean ± SEM (**p = 0.002) band intensity. c, NucView assay for in vivo caspase-3 activity: PHNs in chamber slides were treated with 22C11 (100 ng/ml) for 8 h before incubating with the DEVD-NucView caspase substrate. Upon cleavage of this substrate by caspase-3, the NucView moiety enters the nucleus and fluoresces upon binding with DNA. Treated PHNs showed a marked increase in fluorescence compared with PBS-treated neurons, indicative of increased caspase-3 activity. Pretreatment for 1 h with the caspase-3 inhibitor Z-DEVD-FMK blocked increased NucView fluorescence, confirming that fluorescence was a result of caspase activity. d, g, Treatment of PHNs with a 100 ng/ml solution of 22C11 resulted in increased cleavage of GGA3 compared with PBS-treated neurons in a caspase-3-dependent manner. d, Representative Western blots showing the downregulation of caspase-3 in PHNs after an 8 h treatment with penetratin-1-linked caspase-3 siRNA (siCasp3; 80 nm). e, Densitometric analysis of caspase-3 levels was quantified from three independent experiments and expressed as the mean ± SEM (*p = 0.01) band intensity. f, Representative Western blots with a C-terminal-specific GGA3 antibody showing increased production of the ∼37 kDa fragment in 22C11-treated compared with PBS-treated PHNs. This effect was blocked by downregulation of caspase-3. g, Densitometric analysis of cleaved GGA3 (∼37 kDa fragment) levels was quantified from three independent experiments and expressed as the mean ± SEM (**p = 0.001 for Control vs 22C11; *p = 0.014 for 22C11 vs 22C11 plus siCasp3) band intensity.

Figure 7.

Increased BACE1 and Aβ levels in 22C11-treated neurons is caspase-3 and GGA3 dependent. a–c, Cultured rat PHNs were treated with 22C11 (100 ng/ml) only or with 22C11 plus siCasp3 for 8 h. a, Representative Western blots and IP analysis showing increased levels of BACE1 and Aβ in 22C11-treated compared with PBS-treated PHNs. Increases in both BACE1 and Aβ levels were blocked by downregulation of caspase-3. b, c, Densitometric analysis of BACE1 and Aβ levels was quantified from three independent experiments and expressed as the mean ± SEM band intensity (BACE1: *p = 0.013 for Control vs 22C11; *p = 0.052 for 22C11 vs 22C11 plus siCasp3; Aβ: ***p < 0.0001 for Control vs 22C11; *p = 0.01 for 22C11 vs 22C11 plus siCasp3). d--f, Mouse neuroblastoma cells (B103) stably expressing APP (B103^APP) were transfected with 4.0 μg of GGA3 plasmid or with 4.0 μg of empty vector plasmid and treated with 500 ng/ml of a 22C11 solution 48 h post-transfection. d, Representative Western blots showing increased BACE1 levels in mock-transfected cells, as well as increased Aβ production (determined by IP). Overexpression of human GGA3 in the cells blocked both BACE1 increase and Aβ production. e, f, Densitometric analysis of BACE1 and Aβ (IP) levels was quantified from three independent experiments and expressed as the mean ± SEM band intensity (Aβ: *p = 0.036 for 22C11 vs 22C11 plus GGA3; BACE1: **p = 0.002 for 22C11 vs 22C11 plus GGA3).

Figure 8.

Treatment of cultured rat PHNs with 22C11 induces the accumulation of BACE1 in early endosomal compartments. a–l, PHNs in chamber slides (Control or pretreated with siCasp3) were treated with a 100 ng/ml solution of 22C11 for 8 h. Following fixation and permeabilization, PHNs were stained with antibodies for BACE1 (a, e, i, green) and the subcellular early endosome marker EEA1 (b, f, j, red) and imaged and analyzed by confocal microscopy (c, g, k). Colocalization of BACE1 with EEA1 (d, h, l) was markedly increased in PHNs treated with 22C11-treated (d) compared with PBS-treated PHNs (h), consistent with an endosome-to-lysosome trafficking defect. Downregulation of caspase-3 by siRNA completely blocked colocalization of BACE1 with EEA1 (l). m, Colocalization of BACE1 and EEA1 was quantified from an average of 10 neurons from three independent experiments using the Volocity software and expressed as the mean ± SEM percentage of BACE1 colocalized with EEA1 (*p = 0.033 for Control vs 22C11; *p = 0.013 for Control vs 22C11 plus siCasp3).

Figure 9.

Schematic illustration of 22C11-induced BACE1 accumulation and Aβ production in neurons. Following synthesis in the endoplasmic reticulum, pro-BACE1 traffics through the trans-Golgi network (TGN), where it matures before localizing to the plasma membrane. a, Under normal conditions, surface BACE1 is reinternalized through the endosomal pathway and may be recycled from endosomes to the TGN back to the plasma membrane. Alternatively, BACE1 can interact with GGA3, which targets BACE1 to lysosomes for degradation. b, Dimerization of APP and subsequent activation of caspase-3 results in the cleavage of GGA3, preventing GGA3 from shuttling BACE1 to lysosomes. This allows BACE1 to accumulate in endosomes, resulting in increased cleavage of APP and increased Aβ production.