FUS ALS-causative mutations impair FUS autoregulation and splicing factor networks through intron retention

Abstract

<jats:title>Abstract</jats:title> <jats:p>Mutations in the RNA-binding protein FUS cause amyotrophic lateral sclerosis (ALS), a devastating neurodegenerative disease. FUS plays a role in numerous aspects of RNA metabolism, including mRNA splicing. However, the impact of ALS-causative mutations on splicing has not been fully characterized, as most disease models have been based on overexpressing mutant FUS, which will alter RNA processing due to FUS autoregulation. We and others have recently created knockin models that overcome the overexpression problem, and have generated high depth RNA-sequencing on FUS mutants in parallel to FUS knockout, allowing us to compare mutation-induced changes to genuine loss of function. We find that FUS-ALS mutations induce a widespread loss of function on expression and splicing. Specifically, we find that mutant FUS directly alters intron retention levels in RNA-binding proteins. Moreover, we identify an intron retention event in FUS itself that is associated with its autoregulation. Altered FUS levels have been linked to disease, and we show here that this novel autoregulation mechanism is altered by FUS mutations. Crucially, we also observe this phenomenon in other genetic forms of ALS, including those caused by TDP-43, VCP and SOD1 mutations, supporting the concept that multiple ALS genes interact in a regulatory network.</jats:p>

FUS-ALS mutants alter FMRP phase separation equilibrium and impair protein translation

“…Our results highlight that Taf15, one of the three members of the FET family (along with Fus and Ewsr1), is increased in both Ribo-Fus ∆14/∆14 and Ribo-Fus -/datasets. Moreover, previous findings have shown that mutant FUS impairs its autoregulation leading to an increase in RNA levels (Humphrey et al, 2020); we here find that Fus is increased in the mutant translatome dataset ( Fig. 5D), indicating that impaired autoregulation could indeed alter FUS protein levels.…”

“…Importantly, FUS mutations, as well as loss of FUS, induce alterations in transcript expression, with the latter having a stronger effect (Humphrey et al, 2020). These changes can influence the MN translatome composition, nonetheless we found alterations in our translatome datasets that did not correlate with a change in the total spinal cord RNA (input) expression, both in this dataset, and in our previous high-depth RNA-seq data (Humphrey et al, 2020) ( Supplementary Table 1). Further, as MN-derived transcripts are a minority of the total spinal cord RNA, we set out to understand whether mutant FUS had an additional gain of function on the MN translatome by comparing the effect of mutant FUS expression to that of FUS knockout.…”

“…To explore how the translation landscape is affected by mutant FUS expression, we have sequenced ribosome-engaged transcripts from motor neurons. Importantly, FUS mutations induce transcript expression changes, and we have previously shown that these are comparable to, but weaker than FUS knock-out (Humphrey et al, 2020). However, when we compared the translatome datasets, we found specific changes in the ∆14 FUS compared to the knock-out translatome (Fig.5 D), indicating that mutant FUS affects the translatome through a gain of function mechanism.…”

“…We have recently shown that, unlike TDP-43, where ALS-associated mutations cause a nuclear gain of function (Fratta et al, 2018), mutations in FUS result in changes that are in line with a nuclear loss of function (Humphrey et al, 2020). However, in contrast to models expressing FUS mutants at physiological levels (Devoy et al, 2017;Scekic-Zahirovic et al, 2017), FUS knock-out animals do not show overt MN loss and functional motor impairment (Kino et al, 2015) supporting a toxic cytoplasmic gain of function to have a role in disease pathogenesis.…”

ALS-FUS mutation affects the activities of HuD/ELAVL4 and FMRP leading to axon phenotypes in motoneurons

“…This is particularly relevant for ALS-linked genes, genes encoding for cytoskeletal proteins and, notably, other RBPs [8,9]. Importantly, cellular levels of ALS-linked RBPs are tightly regulated by both auto-regulation mechanisms [10][11][12] and cross-regulatory mechanisms [9,[12][13][14][15][16]. The neural RBP HuD (encoded by the ELAVL4 gene), a novel component of cytoplasmic inclusions in FUS, TDP- 43 and sporadic ALS patients [9,14], represents an example of such cross-regulation.…”