Differential disruptions in population coding along the dorsal-ventral axis of CA1 in the APP/PS1 mouse model of Aβ pathology

Abstract

Alzheimers Disease (AD) is a progressive neurodegenerative disorder characterized by a range of behavioral alterations including memory loss as well as cognitive and psychiatric symptoms. While there is growing evidence that cellular and molecular pathologies, such as the accumulation of amyloid beta (A{beta}) plaques may contribute to AD, it remains unclear how this histopathology can give rise to such disparate behavioral deficits. One hypothesis for these diverse behavioral presentations is that A{beta} accumulation has differential effects on neuronal circuits across brain regions, depending on the diverse neurophysiological properties and connections within and between the neurons in these different areas. To test this, we recorded from large neuronal populations in the dorsal and ventral CA1 regions of the hippocampus, areas that are known to be structurally and functionally diverse, in both APP/PS1 animals, a mouse model of A{beta} pathology, and age-matched C57BL/6 controls. Although we found similar levels of A{beta} pathology in the two subregions, populations of neurons in dorsal and ventral CA1 in APP/PS1 mice showed distinct signatures of disrupted neuronal activity as animals navigated a virtual reality environment. In dorsal CA1, pairwise correlations and entropy, a measure of the diversity of activity patterns, were decreased in the APP/PS1 mice. However, in ventral CA1, the opposite findings were observed; pair-wise correlations and entropy was increased in APP/PS1 mice relative to C57BL/6 controls. When we attempted to connect the microscopic features of population activity (the correlations) with the macroscopic features of the population code (the entropy) using a pairwise Ising model, we found that the models performance decreased in predicting dorsal CA1 activity but increased in predicting vCA1 activity in APP/PS1 mice, as compared to the C57BL/6 animals. Taken together, these findings suggest that A{beta} pathology exerts distinct effects across different hippocampal regions. Our results suggest that the diverse behavioral deficits associated with AD and the cellular pathology that arises from A{beta} accumulation may be mechanistically linked by studying the dynamics of neural activity within these diverse hippocampal circuits.