Integrative proteomics highlight presynaptic alterations and c-Jun misactivation as convergent pathomechanisms in ALS

Abstract

Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disease mainly affecting upper and lower motoneurons. Several functionally heterogeneous genes have been associated with the familial form of this disorder (fALS), depicting an extremely complex pathogenic landscape. This heterogeneity has limited the identification of an effective therapy, and this bleak prognosis will only improve with a greater understanding of convergent disease mechanisms. Recent evidence from human post-mortem material and diverse model systems has highlighted the synapse as a crucial structure actively involved in disease progression, suggesting that synaptic aberrations might represent a shared pathological feature across the ALS spectrum. To test this hypothesis, we performed the first comprehensive analysis of the synaptic proteome from post-mortem spinal cord and human iPSC-derived motoneurons carrying mutations in the major ALS genes. This integrated approach highlighted perturbations in the molecular machinery controlling vesicle release as a shared pathomechanism in ALS. Mechanistically, phosphoproteomic analysis linked the presynaptic vesicular phenotype to an accumulation of cytotoxic protein aggregates and to the pro-apoptotic activation of the transcription factor c-Jun, providing detailed insights into the shared pathobiochemistry in ALS. Notably, sub-chronic treatment of our iPSC-derived motoneurons with the fatty acid docosahexaenoic acid exerted a neuroprotective effect by efficiently rescuing the alterations revealed by our multidisciplinary approach. Together, this study provides strong evidence for the central and convergent role played by the synaptic microenvironment within the ALS spinal cord and highlights a potential therapeutic target that counteracts degeneration in a heterogeneous cohort of human motoneuron cultures.

Keywords: ALS; Motoneuron; Proteomics; Spinal cord; Synapse; hiPSC.

Fig. 1

Enrichment analysis of the human ALS spinal cord synaptic proteome. a Schematic diagram shows the experimental workflow. Synaptically-enriched fractions were collected using fresh frozen human spinal cord tissue from age and gender-matched controls, sporadic ALS (sALS) and C9ORF72-RE + ve donors. After Tandem Mass Tag (TMT) labelling, proteome was examined using liquid chromatography, mass spectrometry-based quantification and identification (LC–MS/MS). Collected MS data were analysed using MaxQuant proteomics software. b Heatmap shows genotypic-specific protein changes using hierarchical clustering with Euclidean distance. c The number of up- and down-regulated proteins were counted using the control normalized ratiometric values on a threshold of 20% ≤ change in each direction. d, e Bioinformatics analysis revealed the top enriched KEGG pathways and Gene Ontology (GO) terms using the up- and down-regulated protein IDs. Colorized dot plots and bar graphs were created based on fold enrichment and gene counts/pathway. f Synaptic enrichment analysis using down-regulated protein IDs by the SynGO database shows enhancement in presynaptic terms including synaptic vesicle cycle

Fig. 2

Synaptic proteome from hiPSC-derived MNs highlights presynaptic alterations in ALS. a Heatmap showing the significantly altered proteins identified in the ALS synaptic fraction compared to the control group. b Volcano plot represents the differential levels of the significant synaptic proteins in ALS MNs, with most of them being down-regulated. c ALS-specific enrichment based on down-regulated synaptic proteins reveals GO (biological function) terms linked to mitochondria and presynaptic function. d SynGO analysis reveals the sub-synaptic fractions (cellular component) significantly altered in ALS. e Protein–protein analysis based on the down-regulated molecules n ALS identifies a network of proteins significantly related to synaptic vesicles and neurotransmitter release. f Crucial members of the SNARE machinery are down-regulated in ALS

Fig. 3

Alignment of the human post-mortem and hiPSC c9orf72 cell line proteomics dataset reveals synaptic vesicular dysfunction. a C9orf72 hiSC cell line ALS^C9orf72 shows 89% overlap with the human post-mortem dataset. b Heatmap demonstrates ALS group-specific protein changes using the 2501 congruent proteins. c-e Aligned dataset were used for further enrichment investigation. Based on fold enrichment and gene counts (dot size) top 10 KEGG pathways (c) were visualized by colorized dot plots. Different Gene Ontology (GO) terms (d-e) plotted by fold enrichment and the bar graph is colorized by gene number/pathway. f Synaptic enrichment analysis was created using SynGO database, revealing enriched synaptic functional terms such as synaptic vesicle cycle, presynaptic endocytosis and postsynaptic density. g Heatmap shows specific protein changes in terms of synaptic vesicle release. h Protein–protein interaction network built on the violet cluster identified in g

Fig. 4

Phosphoproteome analysis highlights convergent JUN activation in ALS. a Principal Component Analysis (PCA) of the normalized expression values showing separation between the healthy and ALS phosphoproteomes. b Venn diagram shows the unique and shared phosphosites in both genotypes. c Enrichment analysis displaying the of the top down- and up-regulated GO (biological processes) terms in ALS. d, e Kinome tree and top enriched KEGG pathways using predicted kinases linked to the ALS phosphoproteome. f, g Top 10 significantly enriched kinases (and their targets) in ALS MNs. h Immunocytochemistry confirms that ALS MNs have significantly higher nuclear levels of phospho-c-Jun^Ser63 than healthy controls at DIV 42. Welch's t-test **P < 0.01. n = 3 independent cultures. Scale bar: 25 µm

Fig. 5

Docosahexaenoic acid's neuroprotective effects in fALS. a Representative images and quantitative data for the impact of DHA on the SYP levels in 12 fALS cell lines. DHA treatment significantly increases SYP intensity in all treated cell lines. Paired t-test ***P < 0.001. n = 3 independent treatments with each hiPSC line. Scale bar: 5 µm. b, c Treatment with DHA exerts a neuroprotective effect in fALS cultures by decreasing the levels of p-c-Jun^Ser63 and improving MN survival in mutant cultures. Paired t-test ***P < 0.001. n = 3 independent treatments with each hiPSC line. Scale bar: 25 µm. d Evaluation of DHA efficacy in 12 fALS cell lines with multiple ALS-related phenotypes. n = 3 independent treatments with each hiPSC line. To adjust the maximum value of phenotype rescue to 100% when accumulating three endpoints, the upper limit of phenotype rescue for each item was compressed to 33.3%