Neuroprotective secreted amyloid precursor protein acts by disrupting amyloid precursor protein dimers

Abstract

The amyloid precursor protein (APP) is implied both in cell growth and differentiation and in neurodegenerative processes in Alzheimer disease. Regulated proteolysis of APP generates biologically active fragments such as the neuroprotective secreted ectodomain sAPPalpha and the neurotoxic beta-amyloid peptide. Furthermore, it has been suggested that the intact transmembrane APP plays a signaling role, which might be important for both normal synaptic plasticity and neuronal dysfunction in dementia. To understand APP signaling, we tracked single molecules of APP using quantum dots and quantitated APP homodimerization using fluorescence lifetime imaging microscopy for the detection of Förster resonance energy transfer in living neuroblastoma cells. Using selective labeling with synthetic fluorophores, we show that the dimerization of APP is considerably higher at the plasma membrane than in intracellular membranes. Heparan sulfate significantly contributes to the almost complete dimerization of APP at the plasma membrane. Importantly, this technique for the first time structurally defines the initiation of APP signaling by binding of a relevant physiological extracellular ligand; our results indicate APP as receptor for neuroprotective sAPPalpha, as sAPPalpha binding disrupts APP dimers, and this disruption of APP dimers by sAPPalpha is necessary for the protection of neuroblastoma cells against starvation-induced cell death. Only cells expressing reversibly dimerized wild-type, but not covalently dimerized mutant APP are protected by sAPPalpha. These findings suggest a potentially beneficial effect of increasing sAPPalpha production or disrupting APP dimers for neuronal survival.

FIGURE 1.

Investigation of APP dimerization using APP-mGFP. A, confocal image of a B103 cell expressing APP-mGFP. B–G, wide-field images of B103 cells expressing APP-mGFP alone (B–D) or in combination with APP-mCherry (E–G). B and E, mGFP channel. C and F, mCherry channel. D and G, mGFP lifetime. Scale bars: 10 μm. H, histograms of FRET efficiencies in different experimental conditions. Histograms (thin lines) were fitted to Gaussian curves (thick lines). APP donor only, 36 cells; APP FRET, 48 cells; K624C-APP donor only, 27 cells; K624C-APP FRET, 21 cells. PDF, probability density function.

FIGURE 2.

Investigation of cell surface APP dimers using APP-ACP. A–D, confocal image of a B103 cell expressing APP-ACP-mGFP and labeled with CoA-biotin and 1 nm quantum dots (QD655). A, mGFP fluorescence; B, QD655-labeled APP fluorescence. C, phase contrast image; untransfected cells are devoid of quantum dot labeling. D, overlay of images A and B. Note the selective plasma membrane staining by CoA-647-labeled APP-ACP (red) compared with APP-mGFP (green). E–J, wide-field images of APP-ACP-expressing B103 cells labeled with CoA-488 alone (E–G) or in combination with CoA-547 (H–J). The focal plane was chosen to give the strongest signal from the plasma membrane, and therefore, does not include neurites. E and H, CoA-488 channel. F and I, CoA-547 channel. G and J, lifetime of CoA-488-labeled APP-ACP. Scale bars:10 μm. K, regression analysis of the fluorescence lifetime of CoA-488-labeled APP-ACP-expressing B103 cells. The lifetime of each pixel was multiplied with its SNR and plotted against SNR. The slope of the linear fit gives the background-corrected lifetime (see “Experimental Procedures”). Different experimental conditions were vertically displaced for easier display, and parallels (broken lines) to the APP donor-only fit (red) were drawn to facilitate comparison between the fits. Black, APP donor-only; orange, APP FRET exposed to sAPPα; green, APP FRET; red, K624C-APP FRET. L, histograms of FRET efficiencies in different experimental conditions. Histograms (thin lines) were fitted to Gaussian curves (thick lines). APP donor-only, 55 cell; APP FRET, 98 cells; APP FRET + sAPPα, 35 cells; K624C-APP donor-only, 29 cells; K624C-APP FRET, 50 cells. PDF, probability density function.

FIGURE 3.

Proportion of dimerized APP in different experimental conditions. Left, total cellular APP (cf. Fig. 1). Right, cell surface APP (cf. Fig. 2). Black, quantitatively dimerized K624C mutant of APP, normalized to 100%. Dark gray, K624C-APP in the presence of sAPPα. Light gray, wild-type APP in the absence of biological ligands. White, wild-type APP in the presence of sAPPα. Error bars indicate the S.E. (left) or the error of fit (right). ^*, p < 0.05; ^**, p < 0.001.

FIGURE 4.

Tracking of APP using quantum dots. A, confocal image of a B103 cell expressing APP-ACP labeled with CoA-biotin and a 100 pm concentration of a mixture of quantum dots emitting at 585 nm (green) and 655 nm (red). Note the colocalization of both quantum dots at several spots. The z plane is through the middle of the cell. B, time tracks of the quantum dots shown in panel A, projected into the xy plane (duration ≤190 s). Inset, APP marked with quantum dots in a lower focal plane (z = 0 μm, t = 0 s) delineates the contour of the cell. Cell surface APP monomers and dimers display random walks. ^*, APP dimer undergoing endocytosis after 28 s; note the transition from random movement to fast directional transport. Ellipse, internalized quantum dots. Scale bar, 1 μm. C, APP dimer shown in the white square in panels A and B at higher magnification (1.2-μm side length) at different time points. Images are maximum projections into the xy plane. Blinking of 655-nm quantum dot (t = 2.2 s and 30.4 s) and 585-nm quantum dot (t = 17.4s) indicates that these are single quantum dots.

FIGURE 5.

Removal of heparin reduces APP dimerization. A–H, wide-field images of APP-ACP-expressing B103 cells labeled with CoA-488 and CoA-547. The focal plane was chosen to give the strongest signal from the plasma membrane and, therefore, does not include neurites. A, C, E, and G, CoA-488 intensity channel. B, D, F, and H, lifetime of CoA-488-labeled APP-ACP. A and B, wild-type APP-ACP in control condition. C and D, wild-type APP-ACP after heparinase treatment. E and F, K624C-APP-ACP in control condition. G and H, K624C-APP after heparinase treatment. Scale bars: 10 μm. I, proportion of dimerized APP. Left, K624C mutant of APP-ACP; right, wild-type APP-ACP. Black, control condition; white, after incubation with heparinase. Error bars indicate the S.E. (^*, p = 0.011, n = 10–20 cells per condition).

FIGURE 6.

Effect of sAPPα on survival of neuroblastoma cells. Cells transfected with control plasmid (left), wild-type APP (middle), or K624C-APP (right) and a fluorescent marker plasmid were starved of nutrients and serum for 48 h in the absence (open bars) or the presence of sAPPα (solid bars). Cell numbers are normalized to control cells in the absence of sAPPα. S.E. are indicated. ^**, p < 0.05; ^***, p < 0.001 (n = 20 fields of view, 9 wells).

FIGURE 7.

Hypothetical model of sAPPα action. Left, in the basal state APP mainly exists as dimers in the plasma membrane, which may activate cell death pathways. Part of transmembrane APP is cleaved by α-secretase, liberating sAPPα into the extracellular space. Right, when sAPPα binds to transmembrane APP, the dimers are disrupted, leading to increased cell survival.