Kalirin-7 prevents dendritic spine dysgenesis induced by amyloid beta-derived oligomers

Abstract

Synapse degeneration and dendritic spine dysgenesis are believed to be crucial early steps in Alzheimer's disease (AD), and correlate with cognitive deficits in AD patients. Soluble amyloid beta (Aβ)-derived oligomers, also termed Aβ-derived diffusible ligands (ADDLs), accumulate in the brain of AD patients and play a crucial role in AD pathogenesis. ADDLs bind to mature hippocampal neurons, induce structural changes in dendritic spines and contribute to neuronal death. However, mechanisms underlying structural and toxic effects are not fully understood. Here, we report that ADDLs bind to cultured mature cortical pyramidal neurons and induce spine dysgenesis. ADDL treatment induced the rapid depletion of kalirin-7, a brain-specific guanine-nucleotide exchange factor for the small GTPase Rac1, from spines. Kalirin-7 is a key regulator of dendritic spine morphogenesis and maintenance in forebrain pyramidal neurons and here we show that overexpression of kalirin-7 prevents ADDL-induced spine degeneration. Taken together, our results suggest that kalirin-7 may play a role in the early events leading to synapse degeneration, and its pharmacological activation may prevent or delay synapse pathology in AD.

Keywords: Alzheimer's disease; Rac1; guanine-nucleotide exchange factors; synapse.

Fig. 1.. ADDLs bind to cortical neurons.

Cultured rat cortical neurons (DIV 28) were treated with 500 nM ADDLs for 2 h, then immunostained with antibodies against ADDLs and GFP. (a-b) ADDLs bind to both the dendrite and the dendritic spine. Inset: heat map of ADDL mean intensity. Arrow heads indicate spines. Scale bar = 20 µm (a) and 10 µm (b). (c) Quantification of ADDL + spines indicates that approximately 75% of spines were bound by ADDL. (d-e) High magnification images further illustrate that ADDLs bind to the head of dendritic spines (arrowheads) and along the dendrite (arrows). Data were collected from 8 dendrite segments from 4 neurons. Scale bar = 5 µm.

Fig. 2.. Endogenous kalirin-7 in dendritic spine heads is rapidly reduced by ADDL treatment.

Rat cortical neurons (DIV 28) were treated with vehicle or 500 nM ADDLs for 2 h, then immunostained with antibodies against kalirin-7 and GFP. (a) Endogenous kalirin-7 is present throughout the dendrite and in the majority of dendritic spines. (b) Addition of ADDLs reduces endogenous kalirin-7 in spines. (c) The total levels of endogenous kalirin-7 following ADDL treatment is quantified by measuring mean intensity of kalirin-7 signal in dendrites and dendritic spines. (d) Ratio of endogenous kalirin-7 in spines:dendrites is reduced in ADDL-treated cultures. Data were collected from 13–18 dendrite segments from 5–7 neurons. Scale bar = 10 µm. ***p<0.001.

Fig. 3.. Kalirin-7 protects against ADDL-inducted dendritic spine morphology in cortical neurons.

(a) Rat cortical neurons (DIV 28) were transfected with eGFP-N2 plasmid ± kalirin-7-myc plasmid and treated with vehicle or 500 nM ADDLs for 16 h. Scale bar = 20 µm. (b) High-magnification images of vehicle- and ADDL-treated neurons. Scale bar = 5 µm. (c) Addition of ADDLs increases dendritic spine length, an effect that is blocked by overexpression of kalirin-7. (d) ADDLs also decreased dendritic spine density; loss of dendritic spines was partially prevented by kalirin-7 overexpression. (e) Further analysis revealed that ADDL treatment reduced the percent of mushroom spines, but this effect is occluded by the overexpression of kalirin-7. (f) Filopodia represented a larger percentage of total spines in ADDL-treated neurons and this increase was occluded by kalirin-7 overexpression. (g-h) ADDLs did not alter thin or stubby spines. Means + SEMs, *p=0.02, **p<0.01, ***p<0.001. Data were collected from 2–6 dendrite segments totaling approximately 100 µm from 11–14 neurons.

Fig. 4.. Proposed model of kalirin-7 protection.

Under basal conditions kalirin-7 is localized to both the spines and the dendrites. Previous studies show that kalirin-7 colocalizes with PSD-95 and is activated by multiple mechanisms, including by EphB2 and NMDA receptors. Kalirin-7 can stabilize dendritic spines through Rac1-PAK-mediated signaling. Further, ADDLs reduce levels of PSD-95 as well as EphB2 and NMDA receptors. In the current manuscript, we show that ADDL treatment alters the equilibrium of endogenous kalirin-7 in spines compared to dendrites. ADDLs also induce an immature dendritic spine phenotype characterized by a loss of mushroom shaped spines and an increase filopodial-like spines and spine length. Overexpression of kalirin-7 prevents this ADDL-induced spine atrophy, suggesting that ADDLs destabilize spines by interfering with kalirin-7 localization and signaling.