FUS/TLS undergoes calcium-mediated nuclear egress during excitotoxic stress and is required for Gria2 mRNA processing

Excitotoxic levels of glutamate represent a physiological stress that is strongly linked to amyotrophic lateral sclerosis (ALS) and other neurological disorders. Emerging evidence indicates a role for neurodegenerative disease linked RNA-binding proteins (RBPs) in the cellular stress response. However, the relationships between excitotoxicity, RBP function and pathology have not been explored. Here, we found that excitotoxicity induced the translocation of select ALS-linked RBPs from the nucleus to the cytoplasm within neurons. RBPs affected by excitotoxicity include TAR DNA-binding protein 43 (TDP-43) and, most robustly, fused in sarcoma/translocated in liposarcoma (FUS/TLS). FUS translocation occurs through a calcium-dependent mechanism and coincides with striking alterations in nucleocytoplasmic transport. Further, glutamate-induced upregulation of Gria2 in neurons was dependent on FUS expression, consistent with a functional role for FUS under excitotoxic stress. These findings reveal a link between prominent factors in neurodegenerative disease, namely excitotoxicity, disease-associated RBPs and nucleocytoplasmic transport.


42
Glutamate is the major excitatory neurotransmitter in the central nervous system. Upon release 43 from pre-synaptic terminals, relatively low levels of glutamate activate metabotropic glutamate  experiments, 100mM glutamate (MilliporeSigma G5889) was freshly prepared in neurobasal 120 media and diluted using primary neuron cultured media to achieve 0.1-10µM solutions. To apply 121 stress, neuronal media was replaced with glutamate-containing primary cultured media or primary 122 cultured media alone (glutamate-free control) for 10 minutes. After 10 minutes, treatment media 123 was replaced with primary cultured media for 30 minutes or longer depending on the experiment 124 prior to fixation or lysate collection. Kainic acid (Abcam ab144490) was diluted from 10 mM/ml to 125 300µM/ml in primary cultured media and added to motor neurons for 10 minutes followed by a 126 replacement with glia-conditioned media for one hour. Stock solutions of 5mM Ionomycin 127 (MilliporeSigma I9657) or 1M sodium arsenite (MilliporeSigma 71287) prepared in prepared in 128 dimethyl sulfoxide (Corning 25-950-CQC) or water were diluted to 10µM or 1 mM in primary cultured media, respectively prior to addition to neurons for one hour. Sorbitol

159
were combined in an eppendorf tube and brought up to a 50µL volume using neurobasal media.

160
The DNA mixture was allowed to incubate for 20 minutes before addition to neurons. Upon

191
Primary motor neurons images were acquired as Z-stacks (0.2µm step size) using a 60x lens. As     Table 1). For the analysis of puromycin immunostaining upon FUS knockdown, 229 a 20x20 pixel region was placed in the soma of GFP-positive cells. Using MetaMorph, the integrated intensity of this region was obtained and used to quantify relative puromycin levels as

273
To investigate a potential relationship between excitotoxicity and neurodegenerative disease- and physiologically relevant levels of glutamate 4,28 (10µM; hereon referred to as Glu excito ) for 10 279 minutes followed by a 30-minute washout period (Fig. 1A). Immunofluorescence was then used 280 to assess the effect of Glu excito on the cytoplasmic to nuclear (C:N) ratio of the endogenous RBPs ( Fig. 1B-I). Strikingly, the FUS C:N ratio significantly increased ~15-fold from 0.04±0.05 to 0.6±0.3 282 in response to Glu excito (Fig. 1B,F). This increase is likely due to a rapid egress of FUS from the 283 nucleus into the cytoplasm, as a Western analysis revealed total FUS protein levels are 284 unchanged before and after stress (Fig. S1). Glu excito likewise induced a significant increase in the 285 C:N ratio of TDP-43 (Fig. 1C,G) and hnRNPA1 (Fig. 1D,H) without altering protein expression 286 (Fig. S1). Conversely, Glu excito did not significantly alter the C:N ratio (Fig. 1E,I) or protein 287 expression ( Fig. S1) of TAF15.

289
In light of the robust response of FUS to Glu excito , we focused our attention on the properties of 290 FUS translocation in more detail. First, endogenous FUS translocation in response to Glu excito was 291 confirmed using a panel of different anti-FUS antibodies ( Fig. S2 A,B). We then examined the 292 relationship between FUS translocation and glutamate concentration. With 10µM glutamate, the 293 vast majority of neurons (91.3±11.5%) exhibited FUS egress ( Fig. 2A,B), whereas <5% neurons 294 exhibited translocation at £1µM, revealing a dependence of FUS localization on glutamate 295 concentration (Fig. 2B). Within the time course of our experiment (Fig. 1A), a significant

299
Given the toxicity of Glu excito on neurons 28 , we interrogated whether the rapid and robust 300 accumulation of FUS outside the nucleus was simply a consequence of cell death and/or loss of 301 nuclear envelope integrity. The extent of cell death was assessed using the LDH cytotoxicity 302 assay, which detects the activity of LDH upon its release into the media from dead or dying cells.

303
In contrast to neurons treated with lysis buffer, there was no evidence of cell death for neurons 304 treated with Glu excito (Fig. 2E). Further, Lamin A/C staining revealed an intact nuclear envelope in 305 neurons exposed to Glu excito (Fig. 2F). These observations support the premise that cytoplasmic FUS accumulation represents a cellular response to Glu excito , rather than a non-specific 307 consequence of cell death. Moreover, RBP translocation appears selective, as TAF15 (Fig. 1E,I) 308 and the cytoplasmic protein, fragile X mental retardation protein (FMRP; Fig. S2C), did not 309 change localization following excitotoxic insult. It is noteworthy that Glu excito affects neuron 310 morphology at 30 minutes (Fig. 1A), potentially indicative of a stressed state. Anti-MAP2 staining 311 revealed a rearrangement of the cytoskeleton; staining was more pronounced around the nucleus 312 and indicated dendritic fragmentation (Fig. 1,2). Likewise, the nuclear lamina appeared thickened 313 and the size of nuclei smaller in stressed neurons (Fig. 2F). As expected, neurons exposed to 314 excitotoxic stimuli (10 µM, but not 1 µM glutamate) eventually undergo cell death within 24 hours 315 of the initial insult 28 (Fig. S2D,E).

317
Excitotoxic stress induces egress of predominately nuclear ALS-linked FUS variants.

318
The majority of ALS-linked mutations are located within the nuclear localization sequence (NLS) 319 and, as such, these variants exhibit varying degrees of cytoplasmic mislocalization 15 . Given that 320 both ALS-mutations and Glu excito influence the subcellular localization of FUS, we investigated the 321 relationship between these two factors. A series of FLAG-HA-tagged FUS variants were 322 transiently expressed in neurons and the C:N ratio of exogenous FUS was determined in the 323 absence and presence of Glu excito (Fig. 3). In addition to wildtype (WT) FUS, we examined:

329
Similarly, the C:N ratio of FLAG-HA-FUS H517Q and R521G, both of which exhibit a 330 predominately nuclear localization under basal conditions ( Fig. 3 and 16,26 ), also increased 331 significantly with Glu excito . The C:N ratio of FLAG-HA-FUS R495X, which already exhibits a high degree of cytoplasmic localization under basal conditions ( Fig. 3 and 26 ), did not change with 333 Glu excito . This observation may be indicative of a 'ceiling effect', in that the normal 334 nucleocytoplasmic distribution of R495X-FUS is equivalent to that of 'maximally' redistributed 335 endogenous FUS following acute excitotoxic insult.

337
Nucleocytoplasmic transport is disrupted in response to excitotoxic stress.

338
To understand the mechanism(s) underlying endogenous FUS egress in response to Glu excito , we 349 Surprisingly, KPT-330 also failed to fully restrict NLS-tdTomato-NES to the nucleus under 350 conditions of Glu excito (Fig. 4A,B). Although there was a significant decrease in the percentage of 351 cells with cytoplasmic NLS-tdTomato-NES in the presence of both KPT-330 and Glu excito 352 (60.1±8.0%) compared to Glu excito alone (98.3±2.6, p=<0.0001), these results suggest that CRM1-353 mediated export is dysregulated under conditions of stress (Fig. 4B). Moreover, while 354 endogenous CRM1 predominately localizes to the nucleus, Glu excito induced a significant increase 355 the number of neurons exhibiting a cytoplasmic localization of this protein (Fig. 4D,E). This finding 356 prompted us to examine another critical nucleocytoplasmic transport factor, Ras-related nuclear 357 protein (Ran). Ran is a GTPase that shuttles between the nucleus and cytoplasm and, depending on its nucleotide bound state, facilitates nuclear export or import 34 . Indeed, Glu excito also induced 359 a significant change in the nucleocytoplasmic distribution of Ran (Fig. 4F,G). Taken together,

360
Glu excito caused the redistribution of critical transport factors and attenuated the effects of KPT-361 330 on CRM1 export.

364
Knowing that calcium influx is a critical component of excitotoxicity 1 , we sought to determine 365 whether this signaling molecule is required for the response of FUS to excitotoxicity. To this end, 366 the calcium chelator, EGTA, was included in the neuronal media during the experimental time 367 course. Indeed, EGTA completely prevented Glu excito -induced FUS egress (Fig. 5A,B). Further 368 application of the calcium ionophore, Ionomycin, was sufficient to induce FUS translocation in the 369 vast majority (89.0±5.6%) of neurons (Fig. 5C,D). In light of our previous finding that hyperosmotic 370 stress induces nuclear FUS egress 12 , we wondered whether calcium also mediated this response.

371
In contrast to Glu excito , there was no effect of EGTA on FUS egress in neurons treated with 372 hyperosmotic levels of sorbitol (Fig. 5E,F) (Fig. 5G,H). Application of the glutamatergic agonist, kainic acid, to motor neurons 379 also induced a significant increase in the C:N ratio of FUS (Fig. 5I,J). Kainic acid is known to 380 induce motor neuron excitotoxicity 35 and was used here to avoid confounding effects of glutamate 381 uptake by astroglia present in the motor neuron cultures 36 . We noted a relatively wide range in 382 the C:N ratio of FUS in kainic acid treated neurons; a sub-population of cells exhibited near complete egress of nuclear FUS (Fig. 5I,J) (Fig. 6B). The degree of translational repression induced by Glu excito was comparable to 399 treatment with the translational inhibitor, cycloheximide ( Fig. 6A-D), and did not promote FUS 400 egress (Fig. 6B). Global translational repression was confirmed by a Western analysis (Fig. 6C, 401 D), and was found to occur independently of eukaryotic translation initiation factor 2 alpha 402 (EIF2a)-phosphorylation (Fig. S3B,C) 38 . However, endogenous FUS does not appear to play a 403 vital role in regulating global translation, as puromycin levels were unaffected by FUS knockdown, 404 both in the presence and absence of Glu excito (Fig. S3D-I). RBPs such as FUS play crucial roles in mRNA processing 15 . Although FUS expression did not 409 affect global protein synthesis (Fig. S3E-I), this analysis would not necessarily detect differences 410 in the translation of specific transcripts, especially those targeted to dendrites for local 411 translation 44 . Therefore, we investigated whether FUS modulates mRNA metabolism following 412 excitotoxic insult and focused on Gria2, a transcript that is directly bound by FUS 45 . Gria2 mRNA 413 encodes the GluR2 protein subunit of the AMPA receptor and has been implicated in calcium 414 dyshomeostasis in both ALS 1 and FTD 46 . Following depolarization, dendritic GluR2 expression is 415 enhanced 47 . Under excitotoxic conditions, we uncovered a significant increase in Gria2 transcript 416 density by FISH in both the soma and dendrites of cortical neurons (Fig. 7, S4A-C). To examine 417 whether this increase in Gria2 mRNA density is FUS dependent, endogenous FUS levels were 418 knocked down using two shRNAs targeting distinct sequences within FUS 48 (Fig. S3D-F) prior to 419 excitotoxic treatment (Fig. 7). Consistent with previous findings 45 , reduced FUS expression did 420 not have a significant effect on the levels of Gria2 under basal conditions, as determined by FISH 421 within the neuronal soma and dendrites (Fig. 7B,C). In contrast, Glu excito -induced changes to 422 Gria2 were significantly attenuated upon FUS knockdown. Dendritic expression of Gria2 was 423 particularly sensitive to FUS levels under Glu excito , as knockdown of FUS restored dendritic Gria2 424 levels to baseline (Fig. 7C, D). Within the time course of the analysis, we were unable to detect 425 significant changes in GluR2 protein levels by Western blot analysis of whole cell lysates ( Fig.   426   S4D,E). Taken together, these data show that FUS expression is required for Glu excito -induced 427 changes to Gria2 processing in neuronal dendrites (Fig. 8).

430
This study uncovered an association between disease-linked RBPs and excitotoxicity, a stress 431 that has particularly profound effects on the nucleocytoplasmic distribution of FUS in both cortical 432 (Fig. 1, 2) and motor neurons (Fig. 5). There is a compelling body of evidence linking glutamate-433 induced excitotoxicity to neurodegenerative diseases, including ALS 1-3 . For instance, elevated levels of glutamate were detected in biological samples from ALS patients 5,49,50 . Cell death caused 435 by glutamate and calcium dysregulation has also been documented in multiple animal and cellular 436 models 1,2,5-7,51 . The outcomes of this study shed new light on the excitotoxicity cascade and 437 implicate, for the first time, a role for the ALS/FTD-linked protein FUS in this process.

439
Our results are consistent with a functional role for FUS in response to glutamatergic signaling 52 440 rather than a non-specific effect of cell death. First, FUS egress precedes cell death (Fig. 2).

441
Second, there is selectivity with respect to proteins that undergo a change in cellular localization; 442 the response of FUS is particularly robust compared to the other proteins assessed in this study 443 ( Fig. 1, S2C, S3A). Third, the effects of excitotoxicity on Gria2 depend on FUS expression (Fig.   444   7). FUS binds Gria2 mRNA within introns and the 3' untranslated region, and Gria2 splicing is 445 effected by FUS expression under basal conditons 45 . Under Glu excito , Gria2 density was enhanced 446 in neuronal dendrites in a FUS-dependent manner (Fig. 7). Gria2 encodes the GluR2 protein 447 subunit of the AMPA receptor. Normally, GluR2 is post-transcriptionally edited and GluR2-448 containing AMPA receptors are calcium impermeable. As such, the calcium permeability of AMPA 449 receptors and the susceptibility of neurons to excitotoxicity is dependent on GluR2 1,8 . We 450 speculate that the enhanced dendritic density of Gria2 may serve to increase the number of 451 calcium impermeable (GluR2-containing) AMPA receptors and thereby offset calcium influx 452 caused by existing calcium permeable (GluR2-lacking) receptors (Fig. 8). In ALS, this process 453 could be compromised as a result of dysregulated Gria2 editing and/or GluR2 expression 8,53 , 454 particularly in motor neurons that rely heavily on AMPA receptor signaling 1,2 . The effect of FUS 455 on dendritic Gria2 density following Glu excito (Fig. 7B,C) is novel and consistent with a role of FUS 456 in modulating Gria2 processing. The exact nature of this role however remains to be fully

483
In fact, nucleocytoplasmic transport is an emerging area of therapeutic development and the 484 CRM1 inhibitor KPT-350 is advancing towards ALS clinical trials. Partial inhibition of CRM1 is 485 expected to offset defects in nuclear import. CRM1 inhibitors have had a therapeutic effect in some 65,66 , but not all 57,64 , models of neurodegeneration. Collectively, the available data, including 487 our own (Fig. 4), support CRM1-mediated nucleocytoplasmic transport as a viable therapeutic 488 target for neurodegenerative disorders. However, a combination therapy addressing additional 489 effects of stress-induced nuclear pore degradation (i.e., calpain inhibitors 64 ) may be required for 490 a significant therapeutic outcome.

492
The calcium-mediated response of FUS to Glu excito has additional implications for  (Fig. 3). As ALS-linked variants 499 R521G and H517Q translocate further into the cytoplasm under Glu excito (Fig. 3), we predict these 500 and other variants with impaired binding to nuclear import factors will accumulate in the cytoplasm

805
Enhanced levels of edited Gria2 transcript may represent a mechanism to offset the toxic effects 806 of calcium influx. Prolonged or severe stress could manifest in the pathological aggregation of