Synaptic changes contribute to persistent extra-motor behaviour deficits in amyotrophic lateral sclerosis

Abstract

Extra-motor symptoms are increasingly recognised in amyotrophic lateral sclerosis (ALS), encompassing cognitive, social, and behavioural deficits. TAR DNA binding protein 43 (TDP-43) pathology is the central disease marker of almost all cases of ALS and approximately half of frontotemporal dementia (FTD). However, the mechanisms linking TDP-43 pathology with extra-motor symptoms in TDP-43-associated neurodegenerative diseases remain unresolved. In this study, we used the rNLS8 mouse model, which expresses human TDP-43 with an ablated nuclear localisation sequence (hTDP-43^∆NLS) in a doxycycline-regulatable manner causing progressive motor decline reminiscent of ALS, to delineate molecular changes associated with disease-relevant phenotypes. We found that in addition to previously reported dramatic motor decline, rNLS8 mice also develop extra-motor phenotypes consistent with FTD, including disinhibition-like and anxiety-like behaviours, and social interaction impairments. These changes began in the earliest disease stages and remained readily detectable even when rNLS8 mice became severely motor impaired. Notably, extra-motor deficits persisted in rNLS8 mice that had recovered motor function upon hTDP-43^∆NLS transgene suppression. This correlates with widespread mis-splicing of RNA in rNLS8 cortex at disease onset with n = 814 genes showing differential exon usage, a molecular phenotype of TDP-43 loss of function. Mis-splicing persists in the rNLS8 cortex in recovery and may represent lasting impacts of cytoplasmic TDP-43 expression. Further, proteomics analysis of the cortex of rNLS8 mice revealed depletion of synaptic proteins, particularly those involved in glutamatergic signalling pathways, which also persisted following hTDP-43^∆NLS transgene suppression. Similar changes to the glutamatergic pathway were detected in transcriptomic and proteomic datasets from human ALS and FTD post-mortem cortex. Our findings suggest that targeting glutamatergic synaptic components may be an avenue to correct extra-motor deficits associated with TDP-43 pathology.

Supplementary Information: The online version contains supplementary material available at 10.1186/s40478-025-02150-5.

Keywords: ALS; Extra-motor phenotypes; Frontotemporal dementia; Motor neuron disease; Proteomics; Synapse; TDP-43; Transcriptomics.

Fig. 1

rNLS8 mice at disease onset show recoverable motor deficits but persistent hyperlocomotion. A Experimental schematic. Mice were behaviourally evaluated at 0 and 2 WOD, and +2, +4, +6 weeks back on dox. Results for the hindlimb neurological score B and latency to fall in seconds in the rotarod test C are shown. Mice that explored the open field arena for over 30 min were analysed for the total distance travelled D, rearing number E and relative central travel distance F. Data are shown as mean ± SD. Control (5 M, 6F), rNLS8 (3 M, 8F). *p < 0.05, *p < 0.01, ***p < 0.001 between control and rNLS8 mice at the same timepoints by repeated one-way ANOVA. ###p < 0.001 of rNLS8 mice relative to other timepoints by repeated one-way ANOVA

Fig. 2

The cortical transcriptome at disease onset is strongly associated with neuroinflammation and this is completely normalised in recovery in rNLS8 mice. A, B Volcano plots of transcript mean log fold change (LFC; rNLS8/Con) and significance level [-log10(Pvalue)] of cortex from A disease onset (2 WOD) and B recovery (2 WOD + 6 weeks back on dox). Significantly upregulated genes are in red and downregulated in blue. C Gene ontology analysis of biological process enrichment at disease onset (2 WOD) of significantly upregulated (top) and downregulated (bottom) genes. Size of the data points are relative to the number of genes in the term. D Protein–protein interaction networks of significantly upregulated genes at disease onset (2 WOD). Genes are coloured by membership to given biological processes. Data are from n = 4 mice per group

Fig. 3

Differential exon usage in genes associated with axonogenesis and neuronal projection organisation at disease onset and in recovery cortex of rNLS8 mice. A Metascape gene ontology of onset (n = 814) and recovery (n = 175) genes that show significant alternative splicing. Bubble sizes increase with enrichment ratio and colour gradients are scaled to the range of P-values. B Protein–protein interaction networks of genes with significant DEU events. All significant networks relate to DEU genes in onset (2 WOD) cortex. C Number of genes with DEU events at onset (red), recovery (blue) or both onset and recovery (purple). List of genes with DEU events in both onset and recovery cortex are listed alphabetically and neuronal genes are in purple. D Differential exon usage of Cyfip2 019 correlates with a significant decrease in Cyfip2 expression at 2WOD (fragments per kilobase of transcript per million; FKPM). Analysis of a published longitudinal proteomic dataset from control and rNLS8 cortex [29] shows a gradual decrease in cytoplasmic FMR1-interacting protein 2 (CYFIP2) abundance over time and partial recovery in rNLS8 mice allowed to recover for 2 weeks back on dox. E. Differential exon usage of Unc13a at E050 correlates with a significant decrease in Unc13a expression (FKPM) at 2WOD and a gradual decrease in UNC13A protein abundance over time [29]. Data represent n = 4–5 mice per group. *p < 0.05 by t-test

Fig. 4

rNLS8 mice exhibit disinhibition-like phenotypes, and social and executive deficits despite impaired motor function. A The schematic of the rNLS8 mouse model. B The data of rotarod, grip strength and inverted grid tests at 6 WOD and 6 + 6 weeks back on dox (n ≥  6F per group). The open field analyses revealed the C total distance, D rearing number, and E relative central travel of the experimental mice that travelled in the open field arena over 30 min at baseline before the removal of dox (0 WOD), at 1-to-6 WOD and at additional 1-to-6 weeks back on dox. Control (8F), rNLS8 (6F). Data as mean ± SD. *p < 0.05, **p < 0.01 by two-way ANOVA

Fig. 5

rNLS8 mice exhibit social and executive deficits despite impaired motor function. A Schematic of three-chamber social interaction tests. B Sociability session to compare the interaction time (in seconds) that the experimental mice spent with the non-social object (Object) versus the social object (Mouse) at 1, 2, 4 WOD and 6 WOD + 6 weeks back on dox. C Social recognition session to assess the interaction time (in seconds) that the experimental mice spent with a novel (Nov) mouse versus a familiar (Fam) mouse at 1, 2, 4 WOD and + 6 weeks back on dox. D Schematic of Y maze tests. E Spontaneous alteration (%), F total entry number, and G total travel distance (m) at 1, 2, 4 WOD and + 6 weeks back on dox. Control (5 M, 6F), rNLS8 (5 M, 6F). Data as mean ± SD. *p < 0.05, **p < 0.01 by two-way ANOVA

Fig. 6

Depletion of glutamate signalling proteins persists in rNLS8 mice despite motor recovery. A Heatmap of the relative protein abundance of quantified proteins in “persistent” and “recovered” protein subsets in rNLS8 mice at 1, 2, 4, and 6 WOD (WOD) and in recovery (6 WOD + 2 weeks on dox) compared to littermate controls (Ctrl). The persistent subset of proteins represents those that are significantly decreased (fold change > 1.2, p < 0.05) in rNLS8 mice in late disease (6 WOD) and recovery (6 WOD + 2 weeks on dox). The recovered subset of proteins represents those that are significantly decreased in rNLS8 mice in late disease (6 WOD) but are not significantly different to control levels in recovery (6 WOD + 2 weeks on dox). Each column represents data from an individual mouse (n = 5/group), and red = high and blue = low relative protein abundance. B Volcano plots of mean log fold change (rNLS8/Con) and p (− log10) from late disease (6 WOD) and recovery (6 WOD + 2 weeks on dox) timepoints. All proteins (grey), proteins belonging to the glutamatergic synapse (black) gene ontology and the synapse organisation (green) gene ontology terms are shown. (C) Metascape gene ontology analysis of persistent and recovered subsets of proteins. The size of the bubble indicates the number of proteins in each term. (D) Protein–protein interaction network of components of the glutamatergic synapse (red), which are persistently significantly decreased in disease (6 WOD) and recovery (6 WOD + 2 weeks on dox). Proteomics data [29] were re-analysed to identify persistently decreased and recovered proteins in the rNLS8 mouse cortex

Fig. 7

Glutamatergic synapse proteins that are persistently decreased in rNLS8 mice are also significantly decreased in human ALS and FTD post-mortem tissues. A subset of n = 40 persistent significantly decreased proteins from the rNLS8 cortex “glutamatergic synapse” gene ontology term were input into the TDP-map webtool [29]. A total of n = 33 proteins/genes were detected in the transcriptomic [40] and proteomic [39] datasets of human post-mortem brain tissue and abundance is plotted. Datapoints are colour-scaled to illustrate the log fold change where red is increased and blue is decreased compared to controls. Circle size is dependent on the magnitude of log fold change, the larger the change the larger the circle size. A (*) indicates significance (P < 0.05) and (-) indicates that the protein/gene was not detected in the datasets