Inhibitory interneuron deficit links altered network activity and cognitive dysfunction in Alzheimer model

Abstract

Alzheimer's disease (AD) results in cognitive decline and altered network activity, but the mechanisms are unknown. We studied human amyloid precursor protein (hAPP) transgenic mice, which simulate key aspects of AD. Electroencephalographic recordings in hAPP mice revealed spontaneous epileptiform discharges, indicating network hypersynchrony, primarily during reduced gamma oscillatory activity. Because this oscillatory rhythm is generated by inhibitory parvalbumin (PV) cells, network dysfunction in hAPP mice might arise from impaired PV cells. Supporting this hypothesis, hAPP mice and AD patients had decreased levels of the interneuron-specific and PV cell-predominant voltage-gated sodium channel subunit Nav1.1. Restoring Nav1.1 levels in hAPP mice by Nav1.1-BAC expression increased inhibitory synaptic activity and gamma oscillations and reduced hypersynchrony, memory deficits, and premature mortality. We conclude that reduced Nav1.1 levels and PV cell dysfunction critically contribute to abnormalities in oscillatory rhythms, network synchrony, and memory in hAPP mice and possibly in AD.

Figure 1. Network Hypersynchrony Emerges during Reduced Intensity of Gamma Activity in hAPPJ20 Mice

EEG recordings from the left (L) and right (R) parietal cortex (PC) in NTG (n=8) and hAPPJ20 (n=11) mice. The intensity of gamma activity (20–80 Hz), frequency of epileptiform discharges (spikes), and exploratory activity were measured in 1-min intervals for 60 min. Numbers in bars are minutes of EEG recording (C, F) or mice (G). (A) EEG recordings (top), spike distribution (middle), and full frequency range (0–100 Hz) spectrogram (bottom) from an hAPPJ20 mouse. Spike rate increases during periods of reduced gamma intensity. (B) Longitudinal quantifications of gamma activity in hAPPJ20 (black) and NTG (green) mice illustrating abnormal patterns of gamma activity in hAPPJ20 mice. Each trace represents a mouse. (C) hAPPJ20 mice, but not NTG controls, displayed frequent spikes (arrowheads). (D) Longitudinal quantification of gamma activity (top trace) and spike rate (bottom histogram). (E) Linear regression analyses revealed an inverse relationship between intensity of gamma activity and spike rate in individual (left) and among (right) hAPPJ20 mice. (F) Minutes with low, intermediate (inter), or high intensity of gamma activity (left) had high, intermediate, or low spike rates (right), respectively. (G) Only hAPPJ20 mice with increased gamma activity during exploration (responders, RP) (left) had reductions in spikes (right). *p<0.05, **p<0.01, ***p<0.001 by unpaired (C) or paired (G) two-sample t test, or ANOVA and Tukey test (F). Bars represent mean ± SEM. See Figure S1 for supporting data.

Figure 2. Inhibitory Synaptic Impairments and PV Cell Dysfunction in hAPPJ20 Mice

(A–C) Synaptic alterations in layer II/III pyramidal neurons (Pyr) of the parietal cortex in hAPPJ20 (hAPPJ20/Pyr; red) and NTG (NTG/Pyr; black) mice. (A, B) Frequency of miniature (mini, m) and spontaneous (spont, s) excitatory (A) and inhibitory (B) postsynaptic currents (mIPSC, sIPSC, mEPSC, and sEPSC). #p<0.05, ##p<0.01 (Wilcoxon-Mann-Whitney two-sample rank test). (C) Recordings of mIPSCs (upper panels) and sIPSCs (lower panels). (D–G) Synaptic alterations in fast-spiking GABAergic cells of the parietal cortex in GAD67^eGFP transgenic mice without (NTG/GAD67^eGFP; black) or with hAPP (hAPPJ20/GAD67^eGFP; red) expression. (D) Resting membrane potential (RMP) of pyramidal cells and fast-spiking GABAergic cells. ***p<0.001 (t test). (E) Spike traces from fast-spiking GABAergic cells (red) evoked by 800-ms current of 380 pA. (F) Mean action potential amplitudes for the first spike at threshold and for the last spike of the train at each 20-pA-current step above threshold. Neurons were held at −70 mV between current steps (n=10–14 cells per genotype). **p<0.01 (t test), ***p<0.001 (two-way ANOVA). (G) Reduced action potential amplitudes for the first spike at threshold in fast-spiking GABAergic cells of hAPPJ20/GAD67^eGFP mice. Neurons were at RMP (left) or held at −70 mV (right) between current steps. *p<0.05, **p<0.01 (t test). Numbers in bars are cells (A, B, D, G). Values are mean ± SEM. See Figure S2 and Table S1 for supporting data.

Figure 3. Reduced Nav1.1 Levels in PV Cells of hAPPJ20 Mice and in AD Brains

(A–B) Parietal cortex from mice and inferior parietal cortex from humans was dissected and VGSC levels were determined by western blot analyses. (A) Representative western blots and quantification of Nav1.1, Nav1.2, Nav1.3, and Nav1.6 in hAPPJ20 and NTG mice. (B) Representative western blots and quantification of Nav1.1, Nav1.2, and Nav1.6 levels in nondemented controls and AD cases. (C–G) Double fluorescence in situ hybridization for Nav1.1 (red) and immunohistochemistry for PV (green) on cryosections from NTG, hAPPJ20, and GAD67^eGFP mice. (C) In NTG mice, Nav1.1 mRNA was expressed at high levels by PV cells. Arrowheads in bottom panels (merged) indicate single- (PV; green, Nav1.1; red) or double-labeled (yellow) cells. (D, E) In GAD67^eGFP mice, Nav1.1 (D) and Nav1.6 (E) mRNAs were highly expressed by GAD67^eGFP-positive (green) cells. (F) In hAPPJ20 mice, Nav1.1 mRNA expression was reduced in the parietal cortex (compare with C). (G) hAPPJ20 mice had no reductions of PV cells in the parietal cortex (left) but had fewer Nav1.1-positive PV cells (middle) and lower Nav1.1 mRNA expression in PV cells (right) than controls. *p<0.05, **p<0.01, ***p<0.001 (t test). Scale bars: 50 µm (C, F); 5 µm (D, E). Values are mean ± SEM. Numbers in bars are mice (A), human cases (B), or cells (G). See Figure S3 for supporting data.

Figure 4. Inhibition of VGSCs Reduces Gamma Activity and Enhances Hypersynchrony and Context-Dependent Memory Deficits in hAPPJ20 Mice

(A–C) EEG recordings from the left (L) and right (R) parietal cortex (PC) were performed in hAPPJ20 (n=11) and NTG (n=5) mice before and after injection of the VGSC blocker riluzole (RZL, 20 mg/kg, i.p.). (A) In hAPPJ20 mice, but not in NTG controls, riluzole induced intense epileptiform activity (spikes) within 10–15 min (arrowheads). Lower panels show the full frequency range spectrograms of the corresponding EEG recordings. (B) 1-min (left) and 60-min (right) interval quantifications of spikes 60 min before and after riluzole injection. Two-way repeated-measures ANOVA revealed significant effects of treatment (p<0.001), genotype (p<0.01), and treatment × genotype interaction (p<0.001) (left). (C) Intensity of gamma activity after riluzole treatment normalized to gamma intensity during the first minute of recording (100%). Riluzole reduced gamma activity more prominently in hAPPJ20 mice than in NTG controls. Treatment (p<0.001) and treatment × genotype interaction (p<0.001) effects by two-way repeated-measures ANOVA. (D–F) hAPPJ20 (n=15–16 per group) and NTG (n=12–15 per group) mice were treated with the VGSC blocker phenytoin (PHT) or vehicle (Veh) and tested to assess context-dependent habituation and dishabituation to a novel environment. (D) Phenytoin (85 mg/kg/day for 10 days) prevented habituation only in hAPPJ20 mice. *p<0.05 vs vehicle-treated hAPPJ20 mice (twoway repeated-measures ANOVA and Bonferroni test). (E) Phenytoin also increased dishabituation in hAPPJ20 mice. ***p<0.001 vs vehicle-treated hAPPJ20 mice (two-way repeated-measures ANOVA and Bonferroni test). (F) Phenytoin-treated hAPPJ20 mice fully habituated to the environment when they were exposed to it more frequently. However, after 10 days, phenytoin-treated hAPPJ20 mice showed increased dishabituation. *p<0.05, **p<0.01 (unpaired t test), ##p<0.01 (paired t test) (B). *p<0.05, ***p<0.001 vs vehicle-treated hAPPJ20 mice (ANOVA and Bonferroni test) (D–F). Bar or line graphs represent mean ± SEM. See Figure S4 for supporting data.

Figure 5. Increasing Levels of Nav1.1 in hAPPJ20 Mice by Nav1.1-BAC Transgene Overexpression

(A) Nav1.1(SCN1A)-BAC transgene construct and PCR detection of genotypes. (B–E) Single bright-field in situ hybridization for Nav1.1 (B, C) and double fluorescence in situ hybridization for Nav1.1 (red) and immunohistochemistry (green) (E) for parvalbumin (PV), calretinin (CR), calbindin (CB), or somatostatin (SOM) on cryosections from NTG, hAPPJ20, Nav1.1, and hAPPJ20/Nav1.1 (bigenic) mice. (B) As in NTG mice (Figure S3A), Nav1.1 mRNA expression in Nav1.1 mice was restricted to a few cells. (C) Nav1.1 mRNA expression patterns were similar in all genotypes. (D) Densitometric analysis of Nav1.1 mRNA signals in the parietal cortex (C) revealed that Nav1.1-BAC transgene expression increased Nav1.1 mRNA levels in mice with (p=0.018) or without (p=0.028) hAPP expression (two-way ANOVA). (E) In Nav1.1 and bigenic mice, Nav1.1 mRNA was highly expressed by PV cells but not by other interneuronal populations. Arrowheads indicate single- (white) or double-labeled (yellow) cells. Asterisks indicate very low or putative expression. (F) Representative western blots (left) and quantification (right) of Nav1.1 levels in the parietal cortex. Nav1.1 levels were increased by Nav1.1-BAC transgene expression in mice with (p<0.001) or without (p<0.001) hAPP (two-way ANOVA). **p<0.01, ***p<0.001 (one-way ANOVA and Tukey test). (G) Cortical Aβ_1-x levels in 3-month-old hAPPJ20 and bigenic mice determined by ELISA. Scale bars: 10 µm (B, E); 50 µm (C). Numbers in bars are mice (D, F, G). Bars represent mean ± SEM.

Figure 6. Enhancing Nav1.1 Levels Increases Inhibitory Synaptic Currents and Gamma Activity and Reduces Epileptiform Discharges in hAPPJ20 Mice

(A–C) Average frequency (A) and amplitude (C) of sIPSCs and cumulative probability of sIPSC interevent intervals (IEI) (B) recorded from layer II/III and V pyramidal cells. For statistical analyses, data from cells in layers II/III and V were combined (n=21–27 cells per genotype). Nav1.1-BAC expression prevented sIPCS frequency deficits in hAPPJ20 mice. *p<0.05, ***p<0.001 (ANOVA and Tukey test). (D–J) EEG recordings from parietal cortex in NTG (n=6), hAPPJ20 (n=7), Nav1.1 (n=9), and bigenic (n=6) mice. The intensity of gamma activity (20–80 Hz) and epileptiform discharges were measured in 1-min intervals for 60 min. (D) Intensity of gamma activity. (E, F) Cumulative probability of gamma intensity (E) and proportion of time spent at each gamma intensity (F) in hAPPJ20 and bigenic mice. Nav1.1-BAC transgene expression increased the intensity of gamma activity in hAPPJ20 mice. p<0.001 (two-sample Kolmogorov-Smirnov test). (G) Spike frequency. Nav1.1-BAC transgene expression reduced spike frequency in hAPPJ20 mice. (H) Spike frequency in hAPPJ20 and bigenic mice during periods of low, intermediate (inter) or high intensity of gamma activity. (I) Exploratory activity during EEG recording. (J) Significant increases in gamma activity during exploratory behavior (dark blue) were found in bigenic mice but not in hAPPJ20 mice. *p<0.05, **p<0.01, ***p<0.001 (ANOVA and Tukey test) (D, G, H). *p<0.05, **p<0.01 (paired t test) (J). Numbers in bars are cells (A, C), minutes of recording (D, G–I) or mice (J). Bars represent mean ± SEM.

Figure 7. Increasing Nav1.1 Levels Ameliorates Spatial and Context-Dependent Learning and Memory Deficits in hAPPJ20 Mice

NTG, hAPPJ20, Nav1.1, and hAPPJ20/Nav1.1 (bigenic) mice were tested to assess spatial learning and memory in the Morris water maze and context-dependent habituation and dishabituation to a novel environment in the open field. (A) Distance swum in the hidden (spatial) and visible (cued) platform components of the Morris water maze test. In the spatial component, bigenic mice performed better than hAPPJ20 mice. *p<0.05, **p<0.01, ***p<0.001 (two-way repeated-measures ANOVA and Bonferroni test). (B) Platform crossings during the probe trial (platform removed). Only hAPPJ20 mice had memory retention deficits. *p<0.05 vs. nontarget locations (paired two-tailed t test). (C) Activity of mice in an open field was assessed on days 1–2, 7, and 23–24. In these three sessions, bigenic mice habituated faster and more completely than hAPPJ20 mice. *p<0.05, **p<0.01, and ***p<0.001 (two-way repeated-measures ANOVA and Bonferroni test). Red asterisks indicate differences between NTG and hAPPJ20 mice, and yellow asterisks differences between hAPPJ20 and bigenic mice. (D) Only hAPPJ20 mice had memory retention deficits. ***p<0.001 by ANOVA and Tukey test. (E) Survival curves. Nav1.1-BAC expression reduced premature mortality of hAPPJ20 mice. ***p<0.001 (Kaplan-Meier test). Values are mean ± SEM. See Figure S5 for supporting data.