RNA dysregulation, largely driven by the dysfunction of RNA-binding proteins, is emerging as a key pathological feature of many neurodegenerative diseases. In amyotrophic lateral sclerosis (ALS), the hallmark pathology is intraneuronal aggregates of the RNA/DNA-binding protein TDP-43. While TDP-43 has several biological functions, its repression of non-conserved “cryptic” exons during RNA splicing has garnered considerable interest [1]. The nuclear clearance and cytoplasmic aggregation of TDP-43 observed in ALS and other TDP-43 proteinopathies disrupts this function, resulting in the incorrect splicing of hundreds of genes. TDP-43 inclusion pathology is observed in the primary sites of neurodegeneration (motor cortex and spinal cord) in 97% of ALS patients. Among non-motor brain regions, some (e.g. prefrontal cortex, hippocampus) demonstrate variable incidence of TDP-43 inclusions across ALS patients while others (e.g. occipital cortex, cerebellum) are consistently spared of this pathology.
We recently conducted a multi-region brain transcriptomic analysis of ALS to better understand the alterations occurring across brain regions variably affected by TDP-43 inclusion pathology [2]. We performed mRNA-seq on five brain regions (motor cortex, prefrontal cortex, hippocampus, occipital cortex, cerebellum) each derived from 22 sporadic ALS cases and 11 non-neurological controls (n=165 samples) [3]. Gene expression changes and corresponding enriched pathways, were largely concordant across brain regions, suggesting that ALS-linked mechanisms including inflammation, mitochondrial dysfunction and oxidative stress, are also dysregulated in non-motor brain regions. Intriguingly, despite the absence of TDP-43 inclusion pathology in the cerebellum, this region demonstrated the greatest number of ALS-specific alternative splicing events (5356 events, 23.6% of which were classified as de novo). This included a switch in POLDIP3 transcript usage, an established marker of TDP-43 loss of function.
To further explore our observations, we have extended our cohort to include 24 sporadic ALS cases, 15 non-neurological controls and importantly, nine ALS cases carrying a pathogenic repeat expansion in C9orf72 (the major genetic cause of ALS). A pathological feature specific to C9orf72 ALS cases is abundant RNA foci and TDP-43-negative, p62-positive dipeptide-repeat protein inclusions in the cerebellum. We have isolated RNA from frozen post-mortem cerebellum and generated parallel PacBio long-read (Kinnex full-length RNA, Revio system) and Illumina short-read RNA-seq for all 48 samples. Isoform-specific expression and alternative splicing analyses incorporating short- and long-read data will extensively characterise the ALS cerebellum transcriptome, including potential similarities between sporadic and C9orf72 ALS cases. Our findings challenge the notion that the cerebellum is an experimental “control” region for ALS and suggests broader TDP-43 dysfunction.