ALS-Associated FUS Mutations Result in Compromised FUS Alternative Splicing and Autoregulation

The gene encoding a DNA/RNA binding protein FUS/TLS is frequently mutated in amyotrophic lateral sclerosis (ALS). Mutations commonly affect its carboxy-terminal nuclear localization signal, resulting in varying deficiencies of FUS nuclear localization and abnormal cytoplasmic accumulation. Increasing evidence suggests deficiencies in FUS nuclear function may contribute to neuron degeneration. Here we report a novel FUS autoregulatory mechanism and its deficiency in ALS-associated mutants. Using FUS CLIP-seq, we identified significant FUS binding to a highly conserved region of exon 7 and the flanking introns of its own pre-mRNAs. We demonstrated that FUS is a repressor of exon 7 splicing and that the exon 7-skipped splice variant is subject to nonsense-mediated decay (NMD). Overexpression of FUS led to the repression of exon 7 splicing and a reduction of endogenous FUS protein. Conversely, the repression of exon 7 was reduced by knockdown of FUS protein, and moreover, it was rescued by expression of EGFP-FUS. This dynamic regulation of alternative splicing describes a novel mechanism of FUS autoregulation. Given that ALS-associated FUS mutants are deficient in nuclear localization, we examined whether cells expressing these mutants would be deficient in repressing exon 7 splicing. We showed that FUS harbouring R521G, R522G or ΔExon15 mutation (minor, moderate or severe cytoplasmic localization, respectively) directly correlated with respectively increasing deficiencies in both exon 7 repression and autoregulation of its own protein levels. These data suggest that compromised FUS autoregulation can directly exacerbate the pathogenic accumulation of cytoplasmic FUS protein in ALS. We showed that exon 7 skipping can be induced by antisense oligonucleotides targeting its flanking splice sites, indicating the potential to alleviate abnormal cytoplasmic FUS accumulation in ALS. Taken together, FUS autoregulation by alternative splicing provides insight into a molecular mechanism by which FUS-regulated pre-mRNA processing can impact a significant number of targets important to neurodegeneration.     

Recruitment into stress granules prevents irreversible aggregation of FUS protein mislocalized to the cytoplasm

Fused in sarcoma (FUS) belongs to the group of RNA-binding proteins implicated as underlying factors in amyotrophic lateral sclerosis (ALS) and certain other neurodegenerative diseases. Multiple FUS gene mutations have been linked to hereditary forms, and aggregation of FUS protein is believed to play an important role in pathogenesis of these diseases. In cultured cells, FUS variants with disease-associated amino acid substitutions or short deletions affecting nuclear localization signal (NLS) and causing cytoplasmic mislocalization can be sequestered into stress granules (SGs). We demonstrated that disruption of motifs responsible for RNA recognition and binding not only prevents SG recruitment, but also dramatically increases the protein propensity to aggregate in the cell cytoplasm with formation of juxtanuclear structures displaying typical features of aggresomes. Functional RNA-binding domains from TAR DNA-binding protein of 43 kDa (TDP-43) fused to highly aggregation-prone C-terminally truncated FUS protein restored the ability to enter SGs and prevented aggregation of the chimeric protein. Truncated FUS was also able to trap endogenous FUS molecules in the cytoplasmic aggregates. Our data indicate that RNA binding and recruitment to SGs protect cytoplasmic FUS from aggregation, and loss of this protection may trigger its pathological aggregation in vivo.     

Nucleic acid-binding specificity of human FUS protein

FUS, a nuclear RNA-binding protein, plays multiple roles in RNA processing. Five specific FUS-binding RNA sequence/structure motifs have been proposed, but their affinities for FUS have not been directly compared. Here we find that human FUS binds all these sequences with K_d^app values spanning a 10-fold range. Furthermore, some RNAs that do not contain any of these motifs bind FUS with similar affinity. FUS binds RNA in a length-dependent manner, consistent with a substantial non-specific component to binding. Finally, investigation of FUS binding to different nucleic acids shows that it binds single-stranded DNA with three-fold lower affinity than ssRNA of the same length and sequence, while binding to double-stranded nucleic acids is weaker. We conclude that FUS has quite general nucleic acid-binding activity, with the various proposed RNA motifs being neither necessary for FUS binding nor sufficient to explain its diverse binding partners.     

Phosphorylation of the FUS low‐complexity domain disrupts phase separation,aggregation, and toxicity

Neuronal inclusions of aggregated RNA‐binding protein fused in sarcoma (FUS) are hallmarks of ALS and frontotemporal dementia subtypes. Intriguingly, FUS's nearly uncharged, aggregation‐prone, yeast prion‐like, low sequence‐complexity domain (LC) is known to be targeted for phosphorylation. Here we map in vitro and in‐cell phosphorylation sites across FUS LC. We show that both phosphorylation and phosphomimetic variants reduce its aggregation‐prone/prion‐like character, disrupting FUS phase separation in the presence of RNA or salt and reducing FUS propensity to aggregate. Nuclear magnetic resonance spectroscopy demonstrates the intrinsically disordered structure of FUS LC is preserved after phosphorylation; however, transient domain collapse and self‐interaction are reduced by phosphomimetics. Moreover, we show that phosphomimetic FUS reduces aggregation in human and yeast cell models, and can ameliorate FUS‐associated cytotoxicity. Hence, post‐translational modification may be a mechanism by which cells control physiological assembly and prevent pathological protein aggregation, suggesting a potential treatment pathway amenable to pharmacologic modulation.     

FUS contributes to mTOR-dependent inhibition of translation

The amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD)–linked RNA-binding protein called FUS (fused in sarcoma) has been implicated in several aspects of RNA regulation, including mRNA translation. The mechanism by which FUS affects the translation of polyribosomes has not been established. Here we show that FUS can associate with stalled polyribosomes and that this association is sensitive to mTOR (mammalian target of rapamycin) kinase activity. Specifically, we show that FUS association with polyribosomes is increased by Torin1 treatment or when cells are cultured in nutrient-deficient media, but not when cells are treated with rapamycin, the allosteric inhibitor of mTORC1. Moreover, we report that FUS is necessary for efficient stalling of translation because deficient cells are refractory to the inhibition of mTOR-dependent signaling by Torin1. We also show that ALS-linked FUS mutants R521G and P525L associate abundantly with polyribosomes and decrease global protein synthesis. Importantly, the inhibitory effect on translation by FUS is impaired by mutations that reduce its RNA-binding affinity. These findings demonstrate that FUS is an important RNA-binding protein that mediates translational repression through mTOR-dependent signaling and that ALS-linked FUS mutants can cause a toxic gain of function in the cytoplasm by repressing the translation of mRNA at polyribosomes.     

BCR-ABL prevents c-jun-mediated and proteasome-dependent FUS (TLS) proteolysis through a protein kinase CbetaII-dependent pathway

The DNA binding activity of FUS (also known as TLS), a nuclear pro-oncogene involved in multiple translocations, is regulated by BCR-ABL in a protein kinase CbetaII (PKCbetaII)-dependent manner. We show here that in normal myeloid progenitor cells FUS, although not visibly ubiquitinated, undergoes proteasome-dependent degradation, whereas in BCR-ABL-expressing cells, degradation is suppressed by PKCbetaII phosphorylation. Replacement of serine 256 with the phosphomimetic aspartic acid prevents proteasome-dependent proteolysis of FUS, while the serine-256-to-alanine FUS mutant is unstable and susceptible to degradation. Ectopic expression of the phosphomimetic S256D FUS mutant in granulocyte colony-stimulating factor-treated 32Dcl3 cells induces massive apoptosis and inhibits the differentiation of the cells escaping cell death, while the degradation-prone S256A mutant has no effect on either survival or differentiation. FUS proteolysis is induced by c-Jun, is suppressed by BCR-ABL or Jun kinase 1, and does not depend on c-Jun transactivation potential, ubiquitination, or its interaction with Jun kinase 1. In addition, c-Jun-induced FUS proteasome-dependent degradation is enhanced by heterogeneous nuclear ribonucleoprotein (hnRNP) A1 and depends on the formation of a FUS-Jun-hnRNP A1-containing complex and on lack of PKCbetaII phosphorylation at serine 256 but not on FUS ubiquitination. Thus, novel mechanisms appear to be involved in the degradation of FUS in normal myeloid cells; moreover, the ability of the BCR-ABL oncoprotein to suppress FUS degradation by the induction of posttranslational modifications might contribute to the phenotype of BCR-ABL-expressing hematopoietic cells.     

A yeast model of FUS/TLS-dependent cytotoxicity

FUS/TLS is a nucleic acid binding protein that, when mutated, can cause a subset of familial amyotrophic lateral sclerosis (fALS). Although FUS/TLS is normally located predominantly in the nucleus, the pathogenic mutant forms of FUS/TLS traffic to, and form inclusions in, the cytoplasm of affected spinal motor neurons or glia. Here we report a yeast model of human FUS/TLS expression that recapitulates multiple salient features of the pathology of the disease-causing mutant proteins, including nuclear to cytoplasmic translocation, inclusion formation, and cytotoxicity. Protein domain analysis indicates that the carboxyl-terminus of FUS/TLS, where most of the ALS-associated mutations are clustered, is required but not sufficient for the toxicity of the protein. A genome-wide genetic screen using a yeast over-expression library identified five yeast DNA/RNA binding proteins, encoded by the yeast genes ECM32, NAM8, SBP1, SKO1, and VHR1, that rescue the toxicity of human FUS/TLS without changing its expression level, cytoplasmic translocation, or inclusion formation. Furthermore, hUPF1, a human homologue of ECM32, also rescues the toxicity of FUS/TLS in this model, validating the yeast model and implicating a possible insufficiency in RNA processing or the RNA quality control machinery in the mechanism of FUS/TLS mediated toxicity. Examination of the effect of FUS/TLS expression on the decay of selected mRNAs in yeast indicates that the nonsense-mediated decay pathway is probably not the major determinant of either toxicity or suppression.     

Aggregation of ALS-linked FUS mutant sequesters RNA binding proteins and impairs RNA granules formation

Protein aggregate/inclusion is one of hallmarks for neurodegenerative disorders including amyotrophic lateral sclerosis (ALS). FUS/TLS, one of causative genes for familial ALS, encodes a multifunctional DNA/RNA binding protein predominantly localized in the nucleus. C-terminal mutations in FUS/TLS cause the retention and the inclusion of FUS/TLS mutants in the cytoplasm. In the present study, we examined the effects of ALS-linked FUS mutants on ALS-associated RNA binding proteins and RNA granules. FUS C-terminal mutants were diffusely mislocalized in the cytoplasm as small granules in transiently transfected SH-SY5Y cells, whereas large aggregates were spontaneously formed in ∼10% of those cells. hnRNP A1, hnRNP A2, and SMN1 as well as FUS wild type were assembled into stress granules under stress conditions, and these were also recruited to FUS mutant-derived spontaneous aggregates in the cytoplasm. These aggregates stalled poly(A) mRNAs and sequestered SMN1 in the detergent insoluble fraction, which also reduced the number of nuclear oligo(dT)-positive foci (speckles) in FISH (fluorescence in situ hybridization) assay. In addition, the number of P-bodies was decreased in cells harboring cytoplasmic granules of FUS P525L. These findings raise the possibility that ALS-linked C-terminal FUS mutants could sequester a variety of RNA binding proteins and mRNAs in the cytoplasmic aggregates, which could disrupt various aspects of RNA equilibrium and biogenesis.     

Liquidizing FUS via prion‐like domain phosphorylation

FUS is an RNA‐binding protein (RBP) with a prion‐like domain (PrLD) that condenses into functional liquids, which can aberrantly phase transition into solid aggregates comprised of pathological fibrils connected to neurodegenerative disease. How cells prevent aberrant phase transitions of FUS and other disease‐linked RBPs with PrLDs is poorly understood. In this issue of The EMBO Journal, Monahan et al ( 2017 ) establish that phosphorylation of specific serine and threonine residues in the FUS PrLD inhibits aberrant phase separation and toxicity.     

The FUS about arginine methylation in ALS and FTLD

EMBO J (2012) 31 22, 4258–4275 doi:10.1038/emboj.2012.261; published online September112012In a time where links between amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD) neurodegeneration are becoming increasingly clear, it is important to establish the convergent and divergent mechanisms responsible for this. Accordingly, Dormann et al (2012) have identified that methylation of the Fused in sarcoma (FUS) RGG3 domain is involved in the cytoplasmic mislocalisation of ALS-FUS mutants, through a transportin-dependent mechanism. By contrast, hypomethylation in this domain may play a role in the aberrant accumulation of FUS in FTLD-FUS. This work showcases arginine methylation as a phenomenon to watch out for in neurodegenerative pathology.The nuclear RNA/DNA-binding protein, FUS, first burst onto the ALS scene in 2009 when it was discovered that mutations in this protein are causative for familial ALS (Kwiatkowski et al, 2009; Vance et al, 2009). It is now thought to be responsible for 4% of familial (and rare sporadic) ALS cases. In ALS patients with FUS mutations (ALS-FUS), the FUS protein is deposited in abnormal protein inclusions in neurons and glia and nuclei often show a reduced FUS staining (Lagier-Tourenne et al, 2010). Fascinatingly, this abnormal FUS deposition is also observed in several subtypes of FTLD, subsequently termed as FTLD-FUS. However, unlike in ALS-FUS, there are no known FUS mutations in this disease (Da Cruz and Cleveland, 2011). Dormann et al (2012) now broaden our knowledge of how ALS-causing mutations in FUS lead to its abnormal cytoplasmic deposition in neurons and glia, and for the first time suggest a mechanism through which this could also occur in FTLD-FUS, in the absence of FUS mutations.The majority of pathogenic mutations identified in ALS-FUS are located at the C-terminus of the protein within a region identified to be a proline-tyrosine nuclear localization signal (PY-NLS) (Figure 1). The PY-NLS binds to the nuclear import receptor transportin (TRN), which facilitates FUS transport into the nucleus. Pathogenic FUS mutations affect key residues of the PY-NLS or completely delete the signal sequence and thus impair nuclear import of FUS (Dormann et al, 2010). This nuclear transport defect is likely directly involved in pathogenesis, as mutations that cause a very severe nuclear import deficit (e.g., FUS-P525L), are associated with earlier disease onset and a rapid disease course (Dormann et al, 2010).Open in a separate windowFigure 1(A) Schematic diagram showing the domain structure of FUS. SYGQ-rich, serine, tyrosine, glycine, glutamine-rich domain; RRM, RNA recognition motif; ZnF, zinc finger; PY, proline-tyrosine nuclear localization signal (PY-NLS). (B) Schematic diagram summarizing the divergent mechanisms by which arginine methylation (or absence thereof) may bring about FUS mislocalization and accumulation in ALS-FUS and FTLD-FUS. In ALS-FUS caused by FUS mutations, neuronal cytoplasmic inclusions contain methylated FUS, but are negative for EWS, TAF15 and TRN. Dormann et al (2012) propose that these FUS-specific inclusions result from the combination of a genetic defect (point mutations in the PY-NLS that impair transportin binding) and post-translational modification (arginine methylation in RGG3 domain that also impairs transportin binding). By contrast in FTLD-FUS, neuronal cytoplasmic inclusions contain all three FET proteins and transportin, but are not immunoreactive with meFUS-specific antibodies. It is therefore possible that hypomethylation of the FET proteins and thus increased transportin binding may be involved in the co-deposition of these proteins in FTLD-FUS.FUS and other related PY-NLS-containing FET proteins such as Ewing sarcoma (EWS) protein and TATA-binding protein-associated factor 15 (TAF15) have been described to undergo extensive asymmetric dimethylation in their arginine-glycine-glycine (RGG) domains (Araya et al, 2005; Hung et al, 2009; Jobert et al, 2009). In recent times, it has been established that this arginine methylation can affect their nuclear localization (Araya et al, 2005; Jobert et al, 2009; Tradewell et al, 2012). In this issue, Dormann et al (2012) sought to delve further into the mechanism by which this occurs. They began their studies in a similar fashion to Tradewell and colleagues, by confirming that inhibiting global arginine methylation using the general methylation inhibitor, adenosine-2,3-dialdehyde (AdOx) could restore the lost nuclear localization of both HA-tagged cytoplasmic ALS-causing mutants and cytoplasmic EWS and TAF15 point mutants in HeLa cells. As AdOx is capable of inhibiting protein, DNA and lipid methylation, they also investigated the effect of specifically preventing protein methylation on the localization of the severe FUS-P525L ALS mutant. In these experiments, siRNA-mediated silencing of PRMT1, the protein arginine methyltransferase responsible for the majority of cellular protein arginine dimethylation, successfully restored the nuclear localization of FUS-P525L, thus confirming the importance of arginine methylation in the cytoplasmic localization of ALS-FUS mutants. After performing this groundwork, Dormann and colleagues then began an elegant set of experiments to unravel the mechanism behind their results. They first wanted to identify whether TRN was involved. Co-expression of GFP-tagged TRN inhibitor peptide with HA-tagged FUS-P525L in HeLa cells lead to a complete prevention of the normal nuclear accumulation of FUS-P525L upon AdOx treatment. Thus, identifying a critical role for TRN in the observed nuclear import of ALS-FUS mutants upon demethylation.Dormann et al (2012) next sought to determine how arginine methylation may impact the nuclear import of FUS by TRN. Their first step was to determine which arginine residues in FUS were involved. With this came the ground-breaking finding that it is actually arginine residues in the RGG3 motif N-terminal to the PY-NLS (Figure 1), rather than in the PY-NLS itself that can modulate TRN-dependent nuclear import of mutant FUS. Through NMR spectroscopy, Dormann and colleagues were then able to show that arginine residues in the FUS RGG3 motif can bind directly to TRN. These observations were taken further via studies of the interaction of recombinant FUS-RGG3 domain or synthetic FUS peptides with TRN by isothermal titration calorimetry. Here, they showed that the FUS-P525L mutant bound weakly or not at all to TRN, but in the presence of the unmethylated RGG3 domain the binding affinity of the mutant was rescued to WT-like levels. These observations were validated by the fact that the unmethylated RGG3 domain alone was able to bind TRN in the absence of a C-terminal PY-NLS with an affinity similar to that of the WT PY-NLS. Methylation of the RGG3 peptide completely prevented this binding. Through these experiments Dormann et al (2012) were able to show for the first time that residues outside of the PY-NLS can be involved in FUS nuclear import. Furthermore, they were able to establish a working model by which this occurs: they propose that in normal situations, the PY-NLS anchors the FUS C-terminus to TRN and the adjacent RGG repeats stabilize the interaction. Methylation of the RGG repeats interferes with TRN binding, but in WT FUS the affinity of the PY-NLS for TRN is sufficient to allow nuclear import to continue. By contrast, in the methylated P525L mutant, weak binding of both the methylated RGG domain and the c-terminus of the PY-NLS to TRN abrogates FUS nuclear import, thus causing its cytoplasmic accumulation.On the basis of their model, Dormann et al (2010) predicted that cytoplasmically mislocalized ALS FUS mutants would be methylated in their RGG3 domains. To test this hypothesis, they generated two monoclonal antibodies specific to the methylated RGG3 domain (meFUS antibodies). Cytosolic FUS mutants, such as FUS-P525L, were recognized by the meFUS-specific antibodies in HeLa cells, thus strongly suggesting that methylation of FUS mutants is a contributing factor to their cytoplasmic retention.Having generated the valuable tools of meFUS-specific antibodies, Dormann et al (2010) next turned their attention to the analysis of whether the intriguing shared property of cytoplasmic FUS accumulation in ALS-FUS and FTLD-FUS could be related to arginine methylation. In direct support of their cell-culture experiments, they discovered that there was a strong and consistent co-labelling of all FUS-positive cytoplasmic neuronal and glial inclusions with the meFUS antibody in post-mortem tissue from ALS-FUS patients. However, the surprise came when they began analysing post-mortem tissues from various subtypes of FTLD-FUS patients. Here, they could not see any labelling of FUS-positive neuronal and glial cytoplasmic and intranuclear inclusions with the meFUS-specific antibody. This strongly suggests that a different FUS cytoplasmic retention mechanism is in play in FTLD-FUS patients compared to ALS-FUS patients. In support of this, inclusions in ALS-FUS patients have recently been identified to contain only the FUS protein, while inclusions in FTLD-FUS show a clear co-deposition of all FET proteins (FUS, EWS, and TAF15) along with TRN itself (Neumann et al, 2012). From this, Dormann et al (2012) conclude that mislocalization of FUS in ALS is likely a phenomenon resulting directly from mutations in the nuclear localization signal that is exacerbated by arginine methylation in the RGG3 domain, whereas FUS mislocalization in FTLD-FUS may result from a more general defect in TRN-mediated nuclear import of FET proteins (Figure 1). Indeed, they speculate that in FTLD, hypomethylation of the FET proteins by a yet to be determined mechanism may lead to excessively tight binding of these proteins to TRN. This in turn may lead to impaired dissociation of FET-TRN complexes, which could over time lead to the co-deposition of FET proteins and TRN in cytoplasmic and nuclear inclusions in FTLD patients. The results of future experiments performed to test this hypothesis will doubtless be of huge significance for the FTLD field. Furthermore, this study raises many questions about the normal cellular role of FUS arginine methylation. Is it for example involved in the fine tuning of the shuttling of FUS into and out of the nucleus and if this is the case could arginine methylation in the RGG1 and RGG2 domains, one of which (RGG1) is close to the FUS nuclear export signal, also be involved in regulating FUS localization? All this remains to be seen…     

FUS Interacts with HSP60 to Promote Mitochondrial Damage

FUS-proteinopathies, a group of heterogeneous disorders including ALS-FUS and FTLD-FUS, are characterized by the formation of inclusion bodies containing the nuclear protein FUS in the affected patients. However, the underlying molecular and cellular defects remain unclear. Here we provide evidence for mitochondrial localization of FUS and its induction of mitochondrial damage. Remarkably, FTLD-FUS brain samples show increased FUS expression and mitochondrial defects. Biochemical and genetic data demonstrate that FUS interacts with a mitochondrial chaperonin, HSP60, and that FUS translocation to mitochondria is, at least in part, mediated by HSP60. Down-regulating HSP60 reduces mitochondrially localized FUS and partially rescues mitochondrial defects and neurodegenerative phenotypes caused by FUS expression in transgenic flies. This is the first report of direct mitochondrial targeting by a nuclear protein associated with neurodegeneration, suggesting that mitochondrial impairment may represent a critical event in different forms of FUS-proteinopathies and a common pathological feature for both ALS-FUS and FTLD-FUS. Our study offers a potential explanation for the highly heterogeneous nature and complex genetic presentation of different forms of FUS-proteinopathies. Our data also suggest that mitochondrial damage may be a target in future development of diagnostic and therapeutic tools for FUS-proteinopathies, a group of devastating neurodegenerative diseases.