A Novel Form of Compensation in the Tg2576 Amyloid Mouse Model of Alzheimer’s Disease

Abstract

One century after its first description, pathology of Alzheimer’s disease (AD) is still poorly understood. Amyloid-related dendritic atrophy and membrane alterations of susceptible brain neurons in AD, and in animal models of AD are widely recognized. However, little effort has been made to study the potential effects of combined morphological and membrane alterations on signal transfer and synaptic integration in neurons that build up affected neural networks in AD. In this study spatial reconstructions and electrophysiological measurements of layer II/III pyramidal neurons of the somatosensory cortex from wild-type (WT) and transgenic (TG) human amyloid precursor protein (hAPP) overexpressing Tg2576 mice were used to build faithful segmental cable models of these neurons. Local synaptic activities were simulated in various points of the dendritic arbors and properties of subthreshold dendritic impulse propagation and predictors of synaptic input pattern recognition ability were quantified and compared in modeled WT and TG neurons. Despite the widespread dendritic degeneration and membrane alterations in mutant mouse neurons, surprisingly little, or no change was detected in steady-state and 50 Hz sinusoidal voltage transfers, current transfers, and local and propagation delays of PSPs traveling along dendrites of TG neurons. Synaptic input pattern recognition ability was also predicted to be unaltered in TG neurons in two different soma-dendritic membrane models investigated. Our simulations predict the way how subthreshold dendritic signaling and pattern recognition are preserved in TG neurons: amyloid-related membrane alterations compensate for the pathological effects that dendritic atrophy has on subthreshold dendritic signal transfer and integration in layer II/III somatosensory neurons of this hAPP mouse model for AD. Since neither propagation of single PSPs nor integration of multiple PSPs (pattern recognition) changes in TG neurons, we conclude that AD-related neuronal hyperexcitability cannot be accounted for by altered subthreshold dendritic signaling in these neurons but hyperexcitability is related to changes in active membrane properties and network connectivity.

Increased Signal Delays and Unaltered Synaptic Input Pattern Recognition in Layer III Neocortical Pyramidal Neurons of the rTg4510 Mouse Model of Tauopathy: A Computer Simulation Study With Passive Membrane

“…This earlier and our recent analyses suggest the two key proteins to affect layer III neocortical pyramidal neurons remarkably differently, leaving synaptic input pattern recognition unaltered in both cases. It was found that Aβ decreases membrane resistance and increases membrane capacitance (Somogyi et al, 2016), whereas in our current article, we report mutant tau protein not to change membrane resistance and capacitance in TG neurons. Aβ was shown to affect the morphology and membrane properties of pyramidal neurons in a way that morphological and membrane alterations compensate each other's pathological effects and subthreshold dendritic signaling, and the input pattern recognition/discrimination remains virtually unaltered (Somogyi et al, 2016).…”

“…It was found that Aβ decreases membrane resistance and increases membrane capacitance (Somogyi et al, 2016), whereas in our current article, we report mutant tau protein not to change membrane resistance and capacitance in TG neurons. Aβ was shown to affect the morphology and membrane properties of pyramidal neurons in a way that morphological and membrane alterations compensate each other's pathological effects and subthreshold dendritic signaling, and the input pattern recognition/discrimination remains virtually unaltered (Somogyi et al, 2016). Such a compensation of dendritic atrophy by parallel alterations in membrane properties is not possible in rTg4510 mice because of the unaltered nature of passive membrane properties.…”

“…Spines were not modeled as individual structures but by proportionally increasing specific membrane capacitance and conductance of dendritic compartments. This way, we accounted for electrical effects caused by increase in dendritic surface area due to dendritic spines without actually modifying length and diameter of cylindrical compartments in our computational models ( Shelton, 1985 ; Holmes, 1989 ; Larkman, 1991 ; Jaslove, 1992 ; Stuart and Spruston, 1998 ; Schmidt-Hieber et al, 2007 ; Somogyi et al, 2016 ). In the first step of this procedure, relative increase in total surface area ( q ) caused by spines was calculated for each dendritic compartment according to the following formula: q = ( A s + A d )/ A d , where A s is total surface area of spines received by the dendritic compartment, and A d is surface area of the “smooth” cylindrical compartment without spines.…”

“…In an earlier article, we had shown that synaptic input pattern recognition/discrimination did not get altered in layer III pyramidal neurons of the Tg2576 amyloid mouse model of AD either (Somogyi et al, 2016), where the human amyloid precursor protein with the Swedish mutation was overexpressed, leading to Aβ accumulation. These findings on the absence of alterations in synaptic input pattern recognition under the effect of pathological Aβ and tau, which are two important proteins of many neurodegenerative diseases and both of which are present in AD, may have important message as to the future directions of research toward therapeutics: Once integration of synaptic signals do not get altered by the pathological Aβ FIGURE 12 | Synaptic input pattern recognition/discrimination is unchanged in advanced tauopathy.…”

Multivariate discriminant analysis for branching classification of colonic tubular adenoma glands

“…There are several methods to group or classify similar data by using the characteristics of the data, such as discriminant analysis (DA) (Kennedy, Kaiser, Fisher, et al, 1980;Klecka, 1980;Lachenbruch & GOLDSTEIN, 1979;Teknomo, 2019), analysis of variation (ANOVA) (Ståhle & Wold, 1989), cluster analysis (Romesburg, 2004;Somogyi, Katonai, Alpár, & Wolf, 2016), and fractal analysis (Smith Jr., Marks, Lange, Sheriff Jr., & Neale, 1989). Multivariate DA is a statistical tool that is widely used in statistics to characterize two or more groups of objects by evaluating linear combinations of group features which affect classification.…”