Oligomeric amyloid-beta peptide disrupts phosphatidylinositol-4,5-bisphosphate metabolism

Abstract

Synaptic dysfunction caused by oligomeric assemblies of amyloid-beta peptide (Abeta) has been linked to cognitive deficits in Alzheimer's disease. Here we found that incubation of primary cortical neurons with oligomeric Abeta decreases the level of phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P2), a phospholipid that regulates key aspects of neuronal function. The destabilizing effect of Abeta on PtdIns(4,5)P2 metabolism was Ca2+-dependent and was not observed in neurons that were derived from mice that are haploinsufficient for Synj1. This gene encodes synaptojanin 1, the main PtdIns(4,5)P2 phosphatase in the brain and at the synapses. We also found that the inhibitory effect of Abeta on hippocampal long-term potentiation was strongly suppressed in slices from Synj1+/- mice, suggesting that Abeta-induced synaptic dysfunction can be ameliorated by treatments that maintain the normal PtdIns(4,5)P2 balance in the brain.

Figure 1

Oligomeric Aβ42 peptide causes a decrease in the levels of PtdIns(4,5)P₂. (a) The acute effect of soluble Aβ42 oligomers (200 nM) on anionic phospholipids in 2-week-old primary cortical neuronal cultures is shown (n = 6 for each time point). (b) We observed a decrease in the level of PtdIns(4,5)P₂ after 60 min of treatment with oAβ42 (n = 12, P < 0.01), but not with vehicle (n = 10), inverted peptide Aβ42Rev (n = 6, P = 0.858), shorter nonamyloidogenic peptide Aβ38 (n = 6, P = 0.871) or oAβ42 pre-incubated with monoclonal antibody 6E10 (n = 6, P = 0.778). 6E10 antibody by itself did not affect PtdIns(4,5)P₂ levels (n = 6, P = 0.804). (c) Lipid levels after a 3-d exposure of cortical neurons to 200 nM oAβ42. PtdIns(4,5)P₂ levels remained depressed after subchronic exposure to oAβ42 (P < 0.01), whereas no further changes were observed in the levels of DPtdG, phosphatidylserine and PtdIns4P (n= 6). There was also a trend for an increase in PtdA levels (P= 0.091). (d) The oAβ42-induced PtdIns(4,5)P₂ reduction was reversed when the peptide-containing medium of cells treated with oAβ42 for 3 d was replaced with medium from age-matched cortical cultures (not containing peptide) and treated for an additional 3 d (n= 6, P< 0.01). Error bars here and in the rest of the figures denote mean ± s.e.m. * indicates P < 0.05 and ** indicates P < 0.01.

Figure 2

Oligomers of Aβ are potent destabilizers of PtdIns(4,5)P₂. (a) Oligomeric (O, n =6, P = 0.016), but not monomeric (M, n = 6, P = 0.748) or fibrillary (F, n = 6, P = 0.127), forms of Aβ42 reduced PtdIns(4,5)P₂ levels in cortical neurons after a 72-h exposure. (b) Effects of oAβ42 on PtdIns(4,5)P₂ were prevented after a 3-d incubation with the scyllo enantiomer of inositol (scyllo + oAβ42, n = 6, P = 0.787), but not with the inactive stereoisomer chiro-inositol (chiro + Aβ42, n = 6, P < 0.01). Scyllo- and chiro-inositol alone did not have any effect on basal PtdIns(4,5)P₂ levels (n = 6, P = 0.689 and P = 0.733, respectively). (c) A 3-d incubation of wild-type (WT) N2a cells with conditioned medium (1:4 dilution) from the neuroblastoma N2a cell line expressing the APP^sw mutation containing naturally secreted Aβ oligomers triggered PtdIns(4,5)P₂ breakdown in WT N2a cells (APP^sw on WT, n = 6, P < 0.05); addition of the 6E10 antibody to the conditioned media abolished the effect (APP^sw + 6E10, n = 6, P = 0.879).

Figure 3

oAβ42-triggered PtdIns(4,5)P₂ deficits are calcium and NMDA receptor dependent. (a) PtdIns(4,5)P₂ levels decreased after 60 min in the presence of the calcium ionophores A23187 (2 μM; n = 6, P < 0.01) and ionomycin (iono, 2 μM; n = 6, P < 0.01), but not when the calcium chelator EGTA (2 mM) was incubated with oAβ42 (n = 3, P = 0.5243). No effect on basal PtdIns(4,5)P₂ was observed when EGTA alone was added to the cultures (n = 6, P = 0.9023). (b) Aβ-induced PtdIns(4,5)P₂ deficiency was partially rescued by incubation of the cultures with the NMDA receptor antagonist AP5 (10 μM) and oAβ42 (AP5 + oAβ42, n = 6, P < 0.05 compared with control or oAβ42). Addition of the AMPA receptor antagonist CNQX (10 μM) had no effect on the oAβ42-induced PtdIns(4,5)P₂ reduction (CNQX + oAβ42, n =6, P = 0.741 compared to oAβ42 alone).

Figure 4

Analysis of fluorescent PtdIns(4,5)P₂ and DAG probes after treatment with oAβ42 in PC12 cells. (a) Representative intensity profiles of the GFP-PH_PLCδ1 probe before (control) and after the addition of oAβ42 or ionomycin are shown. The relative decrease in plasma membrane fluorescence was calculated as a ratio between the plasma membrane fluorescence intensity (average between the two external peaks) and the average cytosolic fluorescence intensity (dashed line). (b) Graph showing the decrease in plasma membrane fluorescence of GFP-PH_PLCδ1 after oAβ42 (n = 32) or ionomycin (n = 15) treatments. The control peptide Aβ42Rev had no effect (n = 17). (c) PLC inhibitors U73122 (0.5 μM) and edelfosine (edel, 0.5 μM) blocked the effect of oAβ42 after 120 min (n = 17, P < 0.01 compared with oAβ42 alone) and had no effects in the absence of the peptide (n = 17). Aβ38 had no effect on GFP-PH_PLCδ1 translocation (n = 17). (d) Scyllo-inositol blocked the effect of oAβ42 (oAβ42 + scyllo, n = 17), whereas chiro-inositol had no effect (oAβ42 + chiro, n = 17, P < 0.01). The inositol stereoisomers had no effect when added alone (n = 17). (e) Left, PC12 cells were transfected with GFP–C1-PKCγ and analyzed after 10 min of oAβ42 treatment for DAG translocation. Alexa 594–WGA was used as a plasma membrane marker. Right, quantification of the oAβ42 effect on GFP–C1-PKCγ localization, showing a significant translocation of the probe from the cytoplasm to the membrane (n =17, P < 0.01).

Figure 5

Hippocampi from mice lacking one copy of Synj1 show normal LTP in the presence of oAβ42. (a) Left, representative western blot analysis of synaptojanin 1 and tubulin in adult Synj1^+/+ and Synj1^+/− brain extracts (50 μg). Right, quantitative analysis of the blots (n = 4, P < 0.05). (b) PtdIns(4,5)P₂ levels in cortical cultures from Synj1^+/+ and Synj1^+/− mice after a 2-h treatment with 200 nM oAβ42. Although levels of PtdIns(4,5)P₂ were reduced by oβ42 in Synj1^+/+ cultures (P < 0.05), the levels of this lipid in Synj1^+/− neurons were unaffected by the peptide (P = 0.853) (n = 8 and n = 6 in control and oAβ42-treated cultures from Synj1^+/+ mice, respectively; n = 9 and n = 6 in control and oAβ42-treated cultures from Synj1^+/− mice, respectively). (c) Although Synj1^+/+ slices (n = 7) showed a reduction of LTP following bath application of 200 nM oAβ42 (F_1,15 = 8.556, P = 0.0104, relative to vehicle), Synj1^+/− slices (n = 7) showed normal LTP in the presence of the peptide (F_1,16 = 0.121, P = 0.73, relative to vehicle). LTP was normal in Synj1^+/− slices (n = 11) compared to Synj1^+/+ slices (n = 10) in the presence of vehicle (F_1,19 = 0.026, P = 0.87). Basal synaptic transmission was not affected in the Syn1^+/− mice (Supplementary Fig. 6). fEPSP, CA1 field-excitatory postsynaptic potential. The bar represents the time of bath application of oAβ42. The three arrows represent the θ-burst stimulation used to induce potentiation. Animals were 3–4 months old.