Rescuing impaired cortical‐hippocampal interactions during sleep with 40Hz stimulation directed to hippocampus in 3xTg‐AD/PVcre mice

Abstract

AbstractBackgroundIn preclinical Alzheimer’s disease (AD), spatial learning and memory is impaired (Allison et al., 2016). We reported similar impairments in 6‐month 3xTg‐AD female mice on a virtual spatial‐reorientation‐task (VM) that requires memory for using landmarks to navigate (Cushing et al., 2020). Memory replay during sleep is critical for learning related plasticity (Ego‐Stengel & Wilson, 2009; Jadhav et al, 2012; Maingret et al, 2016), and hippocampal‐cortical dysfunction is a potential mechanism for memory impairments in AD (Gennaro et al, 2017; Khan et al, 2014). Consistent with this hypothesis, we previously found deficits in hippocampal‐parietal cortex (HPC‐PC) coordination during sleep coinciding with impairments on the VM (Cushing et al, 2020). 40Hz stimulation has been shown to clear AD pathology in mice (Iaccarino et al, 2016; Martorell et al, 2019), and improve functional connectivity in preclinical AD patients (He et al, 2021), but see (Soula et al, 2023).MethodThus, we assessed HPC‐PC coordination in 3xTg‐AD/PVcre mice learning the VM. We implanted a 16‐tetrode recording array targeting PC and HPC and an optical fiber targeting HPC. Daily recording sessions of rest‐task‐rest commenced as mice learned the VM, followed by HPC optogenetic stimulation, with either 40Hz or SHAM stimulation. We assessed sleep quality metrics, delta waves (DW) in PC, and HPC markers of memory replay (SWRs) during slow wave sleep (SWS).ResultIn SHAM stimulated mice SWR‐DW cross‐correlations were reduced, similar to 3xTg‐AD mice (Cushing et al, 2020); while in the 40Hz stimulated mice, this phase‐locking was rescued. Furthermore, the 40Hz stimulated mice have restored performance on the VM compared to SHAM mice. However, this rescued HPC‐PC coupling no longer predicted performance as in NonTg animals (Cushing et al, 2020). Instead, DWTs independently predicted performance in 40Hz animals, but not SHAM animals. We also found paradoxical reduction of amyloid beta in contralateral PC and ventral CA1, but no reduction in contralateral dorsal CA1, synaptically closer to where stimulation occurred.ConclusionThus, 40Hz stimulation of HPC rescued learning and memory related functional interactions in the HPC‐PC network during sleep and as a consequence rescued impairments in spatial navigation, despite a decoupling between HPC‐PC coupling and learning and memory.