Amitriptyline-mediated cognitive enhancement in aged 3×Tg Alzheimer's disease mice is associated with neurogenesis and neurotrophic activity

Abstract

Approximately 35 million people worldwide suffer from Alzheimer's disease (AD). Existing therapeutics, while moderately effective, are currently unable to stem the widespread rise in AD prevalence. AD is associated with an increase in amyloid beta (Aβ) oligomers and hyperphosphorylated tau, along with cognitive impairment and neurodegeneration. Several antidepressants have shown promise in improving cognition and alleviating oxidative stress in AD but have failed as long-term therapeutics. In this study, amitriptyline, an FDA-approved tricyclic antidepressant, was administered orally to aged and cognitively impaired transgenic AD mice (3×TgAD). After amitriptyline treatment, cognitive behavior testing demonstrated that there was a significant improvement in both long- and short-term memory retention. Amitriptyline treatment also caused a significant potentiation of non-toxic Aβ monomer with a concomitant decrease in cytotoxic dimer Aβ load, compared to vehicle-treated 3×TgAD controls. In addition, amitriptyline administration caused a significant increase in dentate gyrus neurogenesis as well as increases in expression of neurosynaptic marker proteins. Amitriptyline treatment resulted in increases in hippocampal brain-derived neurotrophic factor protein as well as increased tyrosine phosphorylation of its cognate receptor (TrkB). These results indicate that amitriptyline has significant beneficial actions in aged and damaged AD brains and that it shows promise as a tolerable novel therapeutic for the treatment of AD.

Figure 1. AMI affects Aβ and tau in aged 3×TgAD mice.

Vehicle (veh)- and AMI-treated Aβ expression in 3×TgAD hippocampus (A), frontal cortex subiculum (B), CA1 region (C), and amygdala (D). (E) AMI/veh effects on Aβ load in the frontal cortex. (F) AMI/veh effects on hippocampal PHF load. Histograms depict relative quantification of associated panels (A–F). AMI effects on hippocampal (G) or cortical (H) Aβ disposition. (I) AMI/veh effects upon total soluble hippocampal tau levels. (J) AMI/veh effects upon soluble phosphorylated PHF tau. For this and subsequent analyses, values in histograms represent mean ± SEM (n = 6), *p<0.05, ** p<0.01, ***p<0.001.

Figure 2. AMI alteration of various synaptic factors.

AMI and vehicle (veh)-mediated effects upon hippocampal expression of (A) synaptophysin, (B) synapsin I, (C) PSD95 and (D) spinophilin. AMI effects on PSD95 (E) and synapsin I (F) expression in primary hippocampal cells. AMI/veh-mediated effects upon cortical expression of (G) synaptophysin, (H) synapsin I, (I) PSD95 and (J) spinophilin. AMI effects on PSD95 (K) and synapsin I (L) expression in primary cortical cells. AMI modulation of hippocampal BDNF (M), NGF (M) levels, TrkB tyrosine phosphorylation (O) and Akt-serine (Ser)-473 phosphorylation (P). AMI effects upon TrkB tyrosine phosphorylation in primary hippocampal (U) or cortical neurons (V). AMI regulation of cortical BDNF (Q) or NGF (R) levels, TrkB tyrosine phosphorylation (S) and Akt-1 Ser-473 phosphorylation (T). AMI effects upon Akt-1 Ser-473 phosphorylation in primary hippocampal (W) or cortical neurons (X). Western band intensities were quantified as actin-normalized arbitrary absorbance units (AU).

Figure 3. AMI improves learning and memory in 3×TgAD mice.

(A) Morris water maze (MWM) used. (B) AMI/veh effects upon day 1 platform escape latency. (C) AMI/veh effects upon MWM platform acquisition measured by escape latency in seconds. Grey panel indicates 3 time points after the seventh day of MWM testing for the probe trial, as described in Methods . (D) AMI effects upon time spent in MWM platform quadrant during 3 probe trial time periods after MWM acquisition. (E) AMI effects upon number of platform traverses during the 3 probe trial time periods after MWM acquisition. (F) Novel object preference (NOP) protocol. (G) AMI/veh effects upon 3×TgAD NOP index. (H) Open field test employed. AMI/veh or effects on open field performance: distance traveled (I); time spent in specific zone (J); total ambulatory activity counts (K); vertical activity counts (L) and total vertical activity time (M). (N) Elevated plus maze employed (arms 1 and 3 = dark). (O) AMI effects upon total time spent in open arms 2 and 4.

Figure 4. Hippocampal and cortical AMI genotropic actions.

(A) Venn analysis of AMI-significantly-regulated transcripts in 3×TgAD cortex (c) and hippocampus (h) (red-upregulation: green-downregulation; blue-diverse regulation). (B) Commonly AMI-regulated genes and z-ratios (cortex-black: hippocampus-white). KEGG signaling pathway population, quantified as a ‘hybrid’ score (see Methods ) by hippocampus (C) and cortex (D) AMI-regulated genes. (E) MSigDB PAGE collection population, quantified as aggregate z-score, by significant AMI-regulated hippocampal (E) or cortical (F) genesets (PAGE collections directly associated with AD-highlighted in red). Percentage gene population of PAGE collection by the input AMI-regulated geneset is indicated for each red histogram bar. (G) Latent semantic indexing (LSI) GeneIndexer term interrogation of significantly AMI-regulated hippocampal or cortical genes. Numbers of AMI-regulated genes implicitly correlating with the specific interrogation term are depicted by the size of the colored bars (red-upregulated: green-downregulated). Total LSI correlation score of all identified genes is indicated above each histogram bar. (H) Dendrogram association of AMI-regulated hippocampal/cortical neurogenesis-associated genes.

Figure 5. AMI activates neuronal developmental processes.

(A) AMI (10 nM, 72 hrs) effect upon primary hippocampal (veh-treated, 1: AMI-treated, 2) and cortical (veh-treated, 3: AMI-treated, 4) cell morphology, indicated with anti-MAP2 immunoreactivity. (B) AMI/veh effects upon adult neurogenesis in hippocampal DG (veh: NeuN-1, BrdU-2, overlay-3; AMI: NeuN-4, BrdU-5, overlay-6). AMI-mediated alterations of hippocampal Ndrg4 (C), Flot2 (D), Grlf1 (E) and Neo1 (F) expression. AMI-mediated alterations of cortical Ndrg4 (G), Rtn4 (H) and Elavl2 (I) expression. Western band intensities were quantified as actin-normalized arbitrary absorbance units (AU:n = 3 individual experiments per protein assessed). Two week treatment of murine neural progenitor cells with AMI (10–20 nM: grey bars) or BDNF (10 ng/mL: white bars) affects Sox2 (J), synapsin I (K) and synaptophysin (L) expression, compared to vehicle treated (black bars) cells. The associated histograms depict the mean ± SEM western band intensity (normalized AU) data from three independent experiments.