Protective paraspeckle hyper-assembly downstream of TDP-43 loss of function in amyotrophic lateral sclerosis

Protective paraspeckle hyper-assembly downstream of TDP-43 loss of function in amyotrophic... Background: Paraspeckles are subnuclear bodies assembled on a long non-coding RNA (lncRNA) NEAT1. Their enhanced formation in spinal neurons of sporadic amyotrophic lateral sclerosis (ALS) patients has been reported but underlying mechanisms are unknown. The majority of ALS cases are characterized by TDP-43 proteinopathy. In current study we aimed to establish whether and how TDP-43 pathology may augment paraspeckle assembly. Methods: Paraspeckle formation in human samples was analysed by RNA-FISH and laser capture microdissection followed by qRT-PCR. Mechanistic studies were performed in stable cell lines, mouse primary neurons and human embryonic stem cell-derived neurons. Loss and gain of function for TDP-43 and other microRNA pathway factors were modelled by siRNA-mediated knockdown and protein overexpression. Results: We show that de novo paraspeckle assembly in spinal neurons and glial cells is a hallmark of both sporadic and familial ALS with TDP-43 pathology. Mechanistically, loss of TDP-43 but not its cytoplasmic accumulation or aggregation augments paraspeckle assembly in cultured cells. TDP-43 is a component of the microRNA machinery, and recently, paraspeckles have been shown to regulate pri-miRNA processing. Consistently, downregulation of core protein components of the miRNA pathway also promotes paraspeckle assembly. In addition, depletion of these proteins or TDP-43 results in accumulation of endogenous dsRNA and activation of type I interferon response which also stimulates paraspeckle formation. We demonstrate that human or mouse neurons in vitro lack paraspeckles, but a synthetic dsRNA is able to trigger their de novo formation. Finally, paraspeckles are protective in cells with compromised microRNA/dsRNA metabolism, and their assembly can be promoted by a small-molecule microRNA enhancer. Conclusions: Our study establishes possible mechanisms behind paraspeckle hyper-assembly in ALS and suggests their utility as therapeutic targets in ALS and other diseases with abnormal metabolism of microRNA and dsRNA. Keywords: ALS, TDP-43, Paraspeckle, NEAT1 Background currently mutations in more than 20 genes are known to Amyotrophic lateral sclerosis (ALS), the most common cause fALS [1]. The complexity of the disease hinders form of motor neuron disease, is a severe adult-onset development of ALS therapeutics, and those two drugs neuromuscular disease affecting motor neurons in the that have been approved for the treatment of ALS so far, spinal cord, brainstem and motor cortex. Up to 90% of riluzole and edaravone, have very limited efficacy. ALS cases are sporadic (sALS), the rest 10% bear a A multifunctional RNA-binding protein TDP-43 strong genetic component (familial ALS, fALS), and encoded by TARDBP gene is believed to be the main culprit in ALS: TDP-43 pathology is typical for ~ 95% of sALS cases and for fALS cases caused by C9ORF72 gene * Correspondence: shelkovnikovat@cardiff.ac.uk mutation [2]; in addition, dozens of mutations in School of Biosciences, Cardiff University, Museum Avenue, Cardiff CF10 3AX, UK TARDBP have been identified in fALS and sALS patients Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 2 of 17 [3, 4]. Hallmarks of all these ALS cases include protein function of the miRNA machinery and those with ac- clearance from the nucleus, its cytoplasmic accumula- tivated dsRNA response. Finally, we show that enoxa- tion and aggregation [5, 6]. Therefore, both loss and gain cin, an enhancer of the miRNA pathway, promotes of TDP-43 function are implicated in ALS however the paraspeckle formation. relative contribution of these two mechanisms is still debated. The paraspeckle is a prototypical nuclear body localized Methods on the border of splicing speckles [7]. A long non-coding Stable cell line maintenance, transfection and treatments RNA (lncRNA) NEAT1 serves as a scaffold for para- SH-SY5Y neuroblastoma cells and MCF7 cells were speckles, spatially organizing a variety of proteins by direct maintained in 1:1 mixture of Dulbecco’s Modified Ea- binding or piggy-back mechanism [8–11]. The NEAT1 gle’s Medium and F12 medium supplemented with 10% locus produces two transcripts, NEAT1_1 and NEAT1_2. fetal bovine serum (FBS), penicillin-streptomycin and The longer NEAT1 isoform, NEAT1_2, is essential for glutamine (all Gibco, Invitrogen). For differentiation paraspeckle assembly [10, 12]. Functions of paraspeckles into neuron-like cells, SH-SY5Y cells were grown on described so far include nuclear retention of specific poly-L-lysine (Sigma) coated coverslips in advanced RNAs, including inverted Alu repeat-containing tran- DMEM/F12 (ADF)/Neurobasal A mixture supple- scripts; regulation of gene expression by sequestration of mented with 10 μM all-trans retinoic acid (Sigma), B27 transcription factors; and modulation of miRNA biogen- (Life Technologies) and BDNF (Miltenyi, 10 ng/ml) for esis [13–16]. 6 days. The following gene-specific siRNAs were used: There is an established association of paraspeckles and ADAR1; Dicer; Drosha; FUS; Ago2; IFNB1 (all Life their components with a variety of pathological states Technologies, Silencer®); TARDBP (Silencer Select®, and conditions, from cancer to neurodegeneration. Para- s23829 and EHU109221, Mission® esiRNA, Sigma); speckles protect cancer cells against DNA damage and NEAT1 (Silencer Select®, n272456). Scrambled negative replication stress, regulate hormone receptor signaling control was AllStars from Qiagen. Plasmids for expres- and hypoxia-associated pathways thereby increasing sion of TDP-43 dNLS and TDP-43 C-termical fragment their survival [17–19]. Paraspeckles become enlarged in are described elsewhere [28]. Cells were transfected cells primed by viral or synthetic double-stranded (ds) with siRNA (400 ng/well), plasmid DNA (200 ng/well) RNAs and play an important role in antiviral response or poly(I:C) (Sigma, 250 ng/well) using Lipofecta- [14]. An unusually tight association of paraspeckle com- mine2000 (Life Technologies) in 24-well plates. TDP-43 ponents with neurodegenerative conditions, and ALS in specific shRNA plasmid was from Sigma (MISSION® particular, has recently emerged. Firstly, enhanced para- SHCLNG-NM_007375). To delete the NLS of endogen- speckle formation has been reported in spinal motor ous TDP-43, Feng Zhang lab’sTarget Finder (http:// neurons of sALS patients [20]. This finding was surpris- crispr.mit.edu/) was used to identify guide RNA target ing because levels of the longer NEAT1 isoform, sequences flanking the genomic region of TARDBP NEAT1_2, essential for paraspeckle formation, are very gene encoding NLS. Respective forward and reverse oli- low in the adult nervous system [21]. Secondly, at least gonucleotides for two pairs of guides were annealed seven paraspeckle proteins, including TDP-43 and FUS, and cloned into pX330-U6-Chimeric_BB-CBh-hSpCas9 are genetically linked to ALS and a related condition, (pX330) vector provided by Feng Zhang (Addgene de- frontotemporal lobar degeneration (FTLD) [22–25]. posited plasmid) as described [29]. MCF7 cells were FUS, a protein structurally and functionally similar to transfected with plasmids encoding upstream and down- TDP-43, is required to build paraspeckles [8, 23]. stream guide RNAs (500 ng/well) using Lipofecta- TDP-43 association with paraspeckles has also been re- mine2000 and analysed after 72 h. Guide RNA sequences: ported [8]. TDP-43 directly binds NEAT1, and this inter- T1: 5’-TTATTTAGATAACAAAAGAAAAA-3′,T2: 5’-A action is increased in the brain of FTLD patients [26, ACATCCGATTTAATAGTGT-3′,T3: 5’-GGAATTCTG 27]. Overall, currently available data support the role of CATGCCCCAGATGC-3′,T4: 5’-ACATCCGATTTAAT paraspeckles in molecular pathology of ALS, however AGTGTT-3′. Cellulartreatments wereas follows: 1×10 the underlying mechanisms of their enhanced formation IU interferon beta-1a (IFNbeta), 0.5 μg/ml LPS, 100 μg/ml in spinal neurons are not understood. zymosan, 50 μM suramin, 500 nM TSA, 2 mM sodium In current study we show that loss of TDP-43 is butyrate, 10 and 50 μMenoxacin, 10 μM riluzole, 10 μM sufficient to stimulate paraspeckle formation – a edaravone (all Sigma). Human ES cell derived neurons phenomenon likely linked to the function of TDP-43 were transfected with 15 μg/well of poly(I:C) using in microRNA (miRNA) processing and as an RNA FuGENE®HD (Promega). Enoxacin, edaravone and riluzole chaperone. Furthermore, we provide evidence that toxicity was assessed using resazurin-based CellTiter-Blue paraspeckles are protective in cells with impaired Cell Viability Assay (Promega). Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 3 of 17 Primary culture of mouse neurons and Superscript IV (Invitrogen). For analysis of miRNA Primary cultures of mouse hippocampal neurons were levels, RNA was extracted with QIAzol (Qiagen) followed prepared from P0 CD1 mice as described [28] and main- by reverse transcription with Qiagen miScript II RT Kit. tained for 5–14 days. Real-time qPCR was conducted using SYBR green master mix as described [28]. For miRNA quantification, forward Differentiation of human ES cells into motor neuron miRNA-specific primers were used in combination with enriched cultures the universal reverse primer (unimiR). All primer se- Cultures of human neural precursor cells (NPCs) and quences are given in Table 1. motor neurons differentiated from H9 hES cell line were prepared as described previously [30]. Briefly, hES cells RNA immunoprecipitation (RIP) and PCR analysis were maintained in mTESR2 media (Stemcell Technolo- MCF7 cells were transfected with equal amounts of plas- gies) on Matrigel® (Corning) coated dishes. Confluent hES mids to express GFP-tagged FUS or NONO together H9 cultures were switched to differentiation medium with TARDBP siRNA or scrambled control siRNA. After composed of ADF supplemented with SB431542 (10 μM, 48 h, cells were scraped in RIP buffer prepared using Abcam). Purmorphamine (1 μM, Cayman Chemicals) and RNase-free water (1xPBS with 1% Triton-X100 and pro- retinoic acid (0.1 μM, Sigma) were added on Day 4. On tease inhibitors cocktail). Cells were left on ice for Day 8, cells were split in 1:2 ratio and on Day 16, NPCs 10 min with periodic vortexing, and the lysate was cen- were dissociated using Accutase®, plated onto Matrigel® trifuged at 13,000 rpm for 10 min. GFP-Trap® beads coated dishes and cultured in ADF with GlutaMAX, (Chromotek) were washed in RIP buffer 4 times and penicillin-streptomycin, B27 (12587–010) and N2 supple- added directly to cleared cell lysates with subsequent ro- ments (all Life Technologies) and BDNF (Miltenyi, 10 ng/ tation at + 4 °C for 3 h. Beads were washed 4 times in ml). On Day 23, Accutase® was used to re-plate neurons RIP buffer and RNA was eluted by resuspension in on dishes/coverslips at desired density. Neurons were cul- TRI-reagent (Sigma). RNA was purified according to tured in 50:50 mixture of ADF/Neurobasal A with the manufacturer’s protocol, and equal amounts of RNA above supplements until Day 40. were used for cDNA synthesis as described above. Immunocytochemistry and RNA-FISH on cultured cells Protein analysis Cells were fixed on coverslips with 4% paraformaldehyde Total cell lysates were prepared for Western blot by lys- on ice for 15 min and permeabilized in cold methanol ing cells in wells in 2× Laemmli (loading) buffer (or 70% ethanol in case of RNA-FISH). Coverslips were followed by denaturation at 100 °C for 5 min. Proteins incubated with primary antibodies diluted in blocking were resolved by SDS-PAGE and transferred to PVDF solution (5% goat serum in 0.1% Triton X-100/PBS) for membrane (Amersham) by semi-dry transfer. The mem- 1 h at RT or at 4 °C overnight. Secondary Alexa488- or brane was blocked in 4% non-fat milk in TBST and incu- Alexa546-conjugated antibody was added for 1 h at RT. bated in primary antibodies prepared in milk or 5% BSA For RNA-FISH, commercially available NEAT1 and overnight. Secondary HRP-conjugated antibodies were MALAT1 probes (Stellaris® FISH Probes against human from Amersham. For detection of proteins, Wester- NEAT1, middle segment or 5′ segment, or human nBright Sirius ECL reagent (Advansta) was used. β-actin MALAT1, all Biosearch Technologies) were used as per was used for normalisation. standard protocol. Fluorescent images were taken using BX61 microscope equipped with F-View II camera and Primary antibodies processed using CellF software (all Olympus). Paraspeckle The following commercial primary antibodies were used: quantification (number of individual paraspeckles per TDP-43 (rabbit polyclonal, 10782–2-AP, Proteintech and DAPI-visualised nucleus) was performed manually, by the mouse monoclonal, MAB7778-SP, R&D Biosystems); same person for all conditions, blinded to the experimen- FUS (rabbit polyclonal, Proteintech, 11570–1-AP); tal condition. Clusters of paraspeckles were counted as a p54nrb/NONO (rabbit polyclonal C-terminal, Sigma); single paraspeckle. For quantification of cleaved caspase 3 PSF/SFPQ (rabbit monoclonal, ab177149, Abcam); Tuj positive cells, ‘Analyze particles’ tool of Image J software (β-Tubulin III, mouse monoclonal, Sigma); dsRNA was used (8–10 fields were analysed per condition). (mouse monoclonal, J2, Kerafast); cleaved caspase 3 (rabbit polyclonal, 9661, Cell Signaling); NF-κB p65 RNA analysis (rabbit monoclonal, D14E12, Cell Signaling); IFIT3 Total cellular RNA was extracted using GenElute total (rabbit polyclonal, Bethyl); p-eIF2α (rabbit monoclonal, RNA kit (Sigma) and possible DNA contamination was re- ab32157, Abcam); p-PKR (rabbit polyclonal, Thr451, moved using RNase free DNase kit (Qiagen). First-strand ThermoFisher); PKR (mouse monoclonal, MAB1980-SP, cDNA synthesis were performed using random primers R&D Systems); eIF2α (rabbit monoclonal, D7D3, Cell Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 4 of 17 Table 1 Primers used in the study Target Forward Reverse GAPDH 5’-TCGCCAGCCGAGCCA-3′ 5’-GAGTTAAAAGCAGCCCTGGTG − 3′ NEAT1 total 5’-CTCACAGGCAGGGGAAATGT-3′ 5’-AACACCCACACCCCAAACAA-3′ NEAT1_2 5’-AGAGGCTCAGAGAGGACTGTAACCTG-3′ 5′-TGTGTGTGTAAAAGAGAGAAGTTGTGG-3’ TDP-43 5’-TCAGGGCCTTTGCCTTTGTT-3’ 5’-TGCTTAGGTTCGGCATTGGAT-3’ IL8 5’-ACACTGCGCCAACACAGAAA-3′ 5’-CCTCTGCACCCAGTTTTCCT-3’ ADARB2 ATATTCGTGCGGTTAAAAGAAGGTG ATCTCGTAGGGAGAGTGGAGTCTTG Alu RNA 5’-GAGGCTGAGGCAGGAGAATCG-3’ 5’-GTCGCCCAGGCTGGAGTG-3’ DICER 5’-TTAACCTTTTGGTGTTTGATGAGTGT-3’ 5’-GCGAGGACATGATGGACAATT-3’ DROSHA 5’-CGGCCCGAGAGCCTTTTAT-3’ 5’-TGCACACGTCTAACTCTTCCA-3’ ADAR1 5’-TTGTCAACCACCCCAAGGT-3’ 5’-CCATCAGCCAGACACCAGTT-3’ AGO2 5’-CACCATGTACTCGGGAGCC-3’ 5’-TCCCAAAGTCGGGTCTAGGT-3’ FUS 5’-GCGGGGCTGCTCAGT-3’ 5’-TTGGGTTGCTTGTTGGGTAT-3’ CHOP 5’-TTAAAGATGAGCGGGTGGC-3′ 5’-GCTTTCAGGTGTGGTGATGTA-3’ CXCL10 5’-TGCCATTCTGATTTGCTGCC-3’ 5’-ATGCTGATGCAGGTACAGCG-3’ IFNB1 5’-ACGCCGCATTGACCATCTAT-3’ 5’-AGCCAGGAGGTTCTCAACAA-3’ IFNA1 5’-TCTGCTATGACCATGACACGAT-3’ 5’-CAGCATGGTCCTCTGTAAGGG-3’ IFNA2 5’-AGGAGGAAGGAATAACATCTGGTC-3’ 5’-GCAGGGGTGAGAGTCTTTGAA-3’ MALAT1 5’-GGATCCTAGACCAGCATGCC-3’ 5′- AAAGGTTACCATAAGTAAGTTCCAGAAAA-3’ IFIH1 5’-GCATGGAGGAGGAACTGTTGA-3’ 5’-GCATGGAGGAGGAACTGTTGA-3’ CYCS 5’-TCGTTGTGCCAGCGACTAAA-3’ 5’-GCTTGCCTCCCTTTTCAACG-3’ STAT1 5’-CTGTGCGTAGCTGCTCCTTT-3’ 5’-GGTGAACCTGCTCCAGGAAT-3’ MYD88 5’-TGACCCCCTGGGGCAT-3’ 5’-AGTTGCCGGATCATCTCCTG-3’ Pri-miR-17–92 5’-CAGTAAAGGTAAGGAGAGCTC 5’-CATACAACCACTAAGCTAAAGAAT AATCTG-3’ AATCTGA-3’ Pri-miR-15a 5’-CCTTGGAGTAAAGTAGCAGCAC-3’ 5’-CCTTGTATTTTTGAGGCAGCAC-3’ miR-18a 5’-CATCATCGGTAAGGTGCATC-3’ 5’-GAATCGAGCACCAGTTACGC-3′ (unimiR) miR-92a 5’-GAGTCTATTGCACTTGTCCC-3’ unimiR miR-106a 5’-AAAAGTGCTTACAGTGCAGGTAG-3’ unimiR Signaling); β-actin (mouse monoclonal, A5441, Sigma). (Stellaris® FISH Probes against human NEAT1 5′ seg- Antibodies were used at 1:500–1:1000 dilution for all ment, Biosearch Technologies) diluted in hybridisation applications. buffer (10% formamide/2xSSC, 5 μl probe in 200 μl buf- fer per slide under a 24 × 60 mm coverslip) in a humidi- Analysis of human tissue samples fied chamber at 37 °C overnight. Nuclei were co-stained Human spinal cord paraffin sections from a panel of with DAPI. Paraspeckles were analysed using BX61 clinically and histopathologically characterised ALS cases microscope/F-View II camera (Olympus) at 100× magni- and neurologically healthy individuals were obtained fication. For RNAscope® ISH analysis, Hs-NEAT1-long from the Sheffield Brain Tissue Bank and MRC London (411541) probe (Advanced Cell Diagnostics) was used Neurodegenerative Diseases Brain Bank (Institute of according to manufacturer’s instructions. For qRT-PCR Psychiatry, King’s College London). Consent was ob- analysis, total RNA was extracted from thick frozen tained from all subjects for autopsy, histopathological as- spinal cord sections and cDNA prepared using sessment and research were performed in accordance Ready-To-Go You-Prime First-Strand Beads (GE Health- with local and national Ethics Committee approved do- care). For laser capture microdissection (LCM), frozen nation. Human spinal cord sections were 7 μm thick. spinal cord sections (total of 5 sections per patient/case) For conventional RNA-FISH, slides were boiled in citrate were cut into 5–10 μm thin sections using a cryostat, buffer for 10 min, washed in 2xSSC prepared with mounted on glass slides and fixed in cold acetone for DEPC-treated water and incubated with NEAT1 probe 3 min. Sections were stained using toluidine blue, Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 5 of 17 dehydrated in ascending alcohol series for 30 s and Thus, de novo paraspeckle formation is typical for spinal placed in xylene for 1 min. The PixCell® II Microdissec- motor neurons and glial cells of individuals affected by tion system (Applied Biosystems) was used for LCM. ALS with primary or secondary TDP-43 pathology. Motor neurons from the anterior grey horn were laser captured (300–500 or 30–80 per patient for healthy con- Loss of TDP-43 but not its cytoplasmic accumulation or trols and ALS cases respectively), with the Macro-LCM aggregation results in paraspeckle hyper-assembly cap films peeled and placed in test tubes with 50 μl ex- We next sought to determine possible mechanisms traction buffer (Pico Pure® RNA Isolation Kit; Thermo underlying paraspeckle hyper-assembly in ALS. TDP-43 Fisher Scientific) on ice. The extracted films were incu- pathology in the spinal cord is very common in ALS, be- bated at 42 °C with the extraction buffer for 30 min and ing present in almost all sALS cases, fALS cases caused frozen at -80 °C until RNA extraction. Total RNA purifi- by mutations in TARDBP gene itself as well as those cation was performed using the above kit as per manu- caused by C9ORF72 gene repeat expansions [2–4, 31]. facturer’s instructions. RNA samples were analysed TDP-43 has been identified as a paraspeckle protein using the Agilent RNA 6000 Pico Kit (Agilent Technolo- [20], thus we tested the possibility that TDP-43 dysfunc- gies®) and used for qRT-PCR. tion affects paraspeckle assembly. Hallmarks of TDP-43 proteinopathy are clearance of the protein from the nucleus and its accumulation and Statististical analysis aggregation in the cytoplasm [5, 6]. We first modelled GraphPad Prism software was used for statistical ana- loss of TDP-43 function in two stable cell lines. By using lysis. Statistical test used in each case is indicated in the specific siRNA, ~ 90 and 50% TDP-43 knockdown was figure legend. N indicates the number of biological repli- achieved in MCF7 and neuroblastoma SH-SY5Y cells, re- cates. On all graphs, error bars represent SEM. spectively (Fig. 2a; Additional file 1: Fig. S1a). TDP-43 depletion led to a significant increase of the number of Results paraspeckles per nucleus (Fig. 2b and c; Additional file Presence of paraspeckles in the spinal cord neurons is a 1: Fig. S1b). Consistently, the paraspeckle-specific hallmark of sALS and fALS NEAT1 isoform, NEAT1_2, was upregulated in cells Augmented paraspeckle assembly has been previously transfected with TDP-43 siRNA (Fig. 2a; Additional file reported in sALS spinal cord neurons as compared to 1: Fig. S1a). Similar results were obtained using an inde- non-ALS controls [20]. We sought to verify this result in pendent TDP-43 siRNA pool and an shRNA targeting a separate cohort of sALS cases as well as to extend this TDP-43 (Additional file 1: Fig. S1c and d). In contrast, we analysis to fALS. In total, 7 sALS cases, 2 cases with did not observe changes in the levels or distribution of an- TARDBP mutations and 4 cases with C9ORF72 muta- other abundant lncRNA, MALAT1, a component of spli- tions alongside with 6 healthy controls were examined cing speckles (Fig. 2a;Additional file 1: Fig. S1e). Levels of by RNA-FISH with NEAT1 probe. No neurons with core paraspeckle proteins SFPQ, NONO and FUS were paraspeckles were detected in the spinal cord of healthy also unaffected by TDP-43 knockdown (Additional file 1: individuals (97 neurons analysed), however such neurons Fig. S1f). We next examined whether TDP-43 knockdown were present in up to 40% of neurons in all ALS cases would result in enhanced association of core paraspeckle examined (Fig. 1a and b). We also confirmed the pres- proteins with NEAT1_2. Plasmids to overexpress ence of paraspeckles in ALS motor neurons using RNA- GFP-tagged FUS or NONO proteins were co-transfected scope® ISH (Fig. 1c). Consistently, qRT-PCR analysis of with scrambled or TDP-43 siRNA followed by RNA im- spinal cord tissue from four healthy controls and four munoprecipitation with GFP-Trap beads. Indeed, by PCR, ALS patients demonstrated elevated NEAT1 levels in the both FUS and NONO demonstrated increased association latter group (Fig. 1d). We further performed laser cap- with NEAT1_2 in TDP-43 depleted cells (Fig. 2d). To ver- ture microdissection (LCM) of spinal neurons in the ify that paraspeckles formed in TDP-43 depleted cells are ventral horn and analysed NEAT1_2 levels by qRT-PCR functional, we measured the expression of established (n = 3 for controls and n = 6 for ALS patients, including paraspeckle-dependent genes, IL8 and ADARB2, known three sALS and three ALS-C9 cases). NEAT1_2 levels to be positively and negatively regulated by paraspeckles, were indeed significantly upregulated in LCM neurons respectively [13, 14]. Indeed, IL8 mRNA was upregulated of ALS patients (Fig. 1e). Finally, using RNAscope® ISH, and ADARB2 mRNA decreased upon TDP-43 knock- we also analysed the presence of paraspeckles in down (Fig. 2e). non-neuronal cells. Wide-spread paraspeckle assembly Since nuclear clearance of TDP-43 in ALS is coupled in glial cells in ALS spinal cord was observed (n = 6 for to its cytoplasmic accumulation and aggregation, we controls and n = 4 for ALS patients, including two sALS, next evaluated the effect of cytoplasmic TDP-43 on one ALS-TDP and one ALS-C9 case) (Fig. 1f). paraspeckles. TDP-43 lacking nuclear localization signal Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 6 of 17 Fig. 1 Paraspeckles are formed in the spinal cord of sALS and fALS patients but not healthy controls. a and b Examples of spinal motor neurons with paraspeckles (a) and their quantification (b) in ALS patients with different disease aetiology. Paraspeckles were visualised in the spinal cord sections of a cohort of fALS and sALS patients as well as neurologically normal control individuals using RNA-FISH with a fluorescent (Quasar 570) probe mapping to the 5′ portion of NEAT1. Images were also taken in the FITC channel to distinguish between specific NEAT1 signal and green autofluorescence from lipofuscin (a). The fraction of neurons with identifiable paraspeckles in the spinal anterior horn of aetiologically different ALS cases and control individuals was quantified and plotted separately for fALS with TARDBP mutations (ALS-TDP), fALS with C9ORF72 repeat expansion (ALS-C9) and sALS cases (b). The top figure within each bar corresponds to the number of cases analysed and the figures below - to the number of individual neurons negative or positive for the presence of paraspeckles. Scale bars, 10 μm. c Examples of paraspeckle-containing neurons in the ALS spinal cord visualised with RNAscope® NEAT1_2 specific probe. In the bottom panel, a paraspeckle-positive (right) and a paraspeckle-negative (left) neurons, found adjacent to each other, are shown. Nuclei are circled. Scale bar, 10 μm. d NEAT1 levels in the total RNA samples extracted from transversely cut spinal cord blocks of ALS patients and healthy controls analysed by qRT-PCR (n = 4 for control and ALS patients, including two sALS and two ALS-C9 cases, Mann-Whitney U-test). e NEAT1_2 levels in neurons microdissected from the spinal anterior horn of ALS patients and healthy controls analysed by qRT-PCR (n = 3 for control and n = 6 for ALS cases, including three sALS and three ALS-C9 cases, Mann-Whitney U-test). f Paraspeckles in glial cells in the ALS spinal cord visualised with RNAscope® ISH using NEAT1_2 specific probe. Representative images of the spinal cord for a control individual and an ALS patient are shown. Scale bar, 20 μm Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 7 of 17 Fig. 2 TDP-43 depletion but not its cytoplasmic accumulation or aggregation stimulates paraspeckle assembly in stable cell lines. a TDP-43 siRNA- mediated knockdown upregulates NEAT1_2. MCF7 cells were transfected with scrambled or TDP-43 siRNA and analysed 48 h post-transfection by qRT- PCR (n = 6). **p < 0.01 (Mann-Whitney U-test). b and c TDP-43 depletion auguments paraspeckle assembly in MCF7 cells. Quantification (b)and representative images (c) are shown. RNA-FISH with NEAT1_2 probe (c, top panels) or anti-NONO staining (c, bottom panels) were used to visualise paraspeckles; arrowheads indicate clusters of paraspeckes. The number of cells analysed is indicated in the bottom of each bar (b)(***p <0.0001, Student’s t-test). d TDP-43 depletion enhances interaction of NEAT1_2 with core paraspeckle proteins NONO and FUS. GFP-tagged NONO or FUS was co-transfected into MCF7 cells together with scrambled siRNA or TDP-43 siRNA. NEAT1_2 and total NEAT1 were detected in GFP pull-down samples by RT-PCR. Arrowhead indicates the specific band for NEAT1_2 primer pair. e Expression of paraspeckle-regulated genes in MCF7 cells depleted of TDP-43 as measured by qRT-PCR (n = 6 or 8). *p < 0.05, ***p < 0.001 (Mann-Whitney U-test). f and g Expression of TDP-43 lacking nuclear localisation signal (TDP-43 dNLS) or TDP-43 C-terminal 25 kDa fragment (TDP-43 CT) does not affect paraspeckles or NEAT1 levels. Representative images of paraspeckles in transfected SH-SY5Y cells (f) and their quantiation (g, n =56 and n = 41 for GFP- and TDP-43 dNLS-expressing cells respectively) are shown. Scale bars are 10 μmin all panels (TDP-43 dNLS) and a C-terminal TDP-43 fragment corre- total cellular levels by CRISPR/Cas9 editing of the en- sponding to ~ 25 kDa TDP-43 cleavage product (TDP-43 dogenous TARDBP gene. Cells were transiently trans- CT, aa. 191–414), both characterized by predominantly fected with plasmids for expression of two cytoplasmic distribution, were transiently expressed in independent guide RNA pairs targeting upstream and neuroblastoma cells (Fig. 2f). However, neither paraspeckle downstream sequences encoding NLS of TDP-43 numbers nor NEAT1 levels were affected by these cytoplas- (Additional file 1: Fig. S1 g). For both guide RNA mic proteins (Fig. 2f and g). TDP-43 dNLS forms cytoplas- pairs tested, 15–20% transfected cells displayed partial mic aggregates in a fraction of cells, but their presence also TDP-43 redistribution to the cytoplasm which ne- did not affect paraspeckles (Fig. 2g, bottom panel). Finally, vertheless did not enhance paraspeckle assembly in order to recapitulate simultaneous nuclear depletion and (Additional file 1: Fig. S1 h), suggesting that substan- cytoplasmic accumulation of TDP-43, we targeted en- tial loss of nuclear TDP-43 is required to produce an dogenous TDP-43 out of the nucleus while preserving its effect on paraspeckles. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 8 of 17 Overall, a decrease in cellular TDP-43 levels results in (Fig. 3a). Consistent with the major role of Drosha in paraspeckle hyper-assembly. pri-miRNA processing, its knockdown led to the build-up of pri-miRNAs; both Drosha and TDP-43 knockdown also Compromising miRNA pathway results in enhanced resulted in significantly diminished levels of select mature paraspeckle assembly miRNAs (Additional file 2:Fig. S2). TDP-43 is known to contribute to miRNA biogenesis at RNA-FISH and paraspeckle quantification showed that two different levels, enhancing the activity of the Micropro- downregulation each of the above proteins is accompanied cessor in the nucleus and of the Dicer complex in the cyto- by enhanced paraspeckle assembly (Fig. 3b and c)and plasm [32, 33]. Recently, paraspeckles have been shown to upregulation of total NEAT1 and NEAT1_2 (Fig. 3d). contribute to pri-miRNA processing by spatially organizing Another ALS-linked protein, FUS, is structurally and the Microprocessor and enhancing its processivity [16]. We functionally similar to TDP-43 and also plays a role hypothesised that augmented paraspeckle assembly in cells in miRNA biogenesis [35], but it is a core paraspeckle depleted of TDP-43 might be a compensatory mechanism protein required for paraspeckle integrity [8, 23]. As to counterbalance the effect of TDP-43 loss of function on expected from its essential paraspeckle function, FUS miRNA processing. If this indeed is true, paraspeckle as- depletion resulted in decreased paraspeckle numbers sembly should be also increased in cells with compromised (Fig. 3b and c). Finally, a small molecule inhibitor of function of the miRNA pathway. To test this hypothesis, RISC loading, suramin [36], wasalsoabletoincrease we knocked down three core enzymes of the miRNA path- NEAT1_2 levels and promote paraspeckle assembly way, a Microprocessor component Drosha, the pre-miRNA (Fig. 3e). processing ribonuclease Dicer and the RISC endonuclease Thus, interfering with the function of the miRNA Ago2, in neuroblastoma cells. In our analysis we also in- pathway causes NEAT1 upregulation and enhanced cluded ADAR1 protein recently reported to promote paraspeckle formation in cultured cells. This may repre- miRNA processing [34]. Using specific siRNAs, we sent one of the mechanisms behind the effect of TDP-43 achieved at least 40% knockdown for each of these genes loss of function on paraspeckles. Fig. 3 Enhanced paraspeckle assembly in cells with compromised function of the miRNA pathway. a Downregulation of Drosha, Dicer, Ago2, ADAR1 and FUS in neuroblastoma cells after transfection of specific siRNA as analysed by qRT-PCR (n = 4–6). *p < 0.05, **p < 0.01 (Mann-Whitney U-test). b and c Knockdown of Drosha, Dicer, Ago2 or ADAR1 results in increased paraspeckle formation. Representative images of cells (b)and paraspeckle quantification (c) are shown. In c, the mean number of paraspeckles per cell and frequencies of such cells were plotted; the number of cells analysed is indicated at the bottom of each bar. *p < 0.05, **p < 0.01; ****p < 0.0001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). d NEAT1 is upregulated in cells after knockdown of Drosha, Dicer, Ago2 and ADAR1 (n = 4–6). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). e Suramin stimulates NEAT1_2 expression and paraspeckle formation. Cells were treated with suramin for 24 h before collection for RNA-FISH and qRT-PCR analysis (n = 4). *p < 0.05 (Mann-Whitney U-test). In a-d, cells were analysed 48 post- transfection. Scale bars are 10 μm in all panels Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 9 of 17 Type I IFN signaling is activated in cells depleted of TDP- Taken together, these data suggest that endogenous 43 or other miRNA factors and stimulates paraspeckle dsRNA accumulation and associated type I IFN response formation represent one of the mechanisms behind augmented para- TDP-43 is known to bind and regulate long transcripts speckle assembly caused by TDP-43 loss of function. [37], and its loss correlates with accumulation of tran- scripts prone to form double stranded (ds) RNA [38]. Paraspeckles are protective in cells with compromised Conspicuously, regulation of cellular response to viral miRNA biogenesis and activated dsRNA response dsRNA is one of the best characterized functions of We next examined whether paraspeckles confer protec- paraspeckles [14, 39]. Therefore, we considered abnor- tion to cells depleted of TDP-43 or stimulated with ex- mal accumulation of endogenous dsRNA in TDP-43 de- ogenous dsRNA. Neuroblastoma cells were transfected pleted cells as another mechanism underlying the effect with siRNAs targeting TDP-43 or another Microprocessor of TDP-43 loss of function on paraspeckles. Critically, component Drosha alone or in combination with NEAT1 miRNA pathway itself is the biggest cellular source of siRNA. Cytotoxicity was assessed after 36 h using cleaved dsRNA, and its factors Dicer, Drosha and ADAR1 are caspase 3 (CC3) and DNA damage-inducible transcript 3 known to limit the accumulation of transcripts with ex- also known as GADD153 or CHOP, a pro-apoptotic tran- tensive secondary structure [40–42]. scription factor [46, 47] as markers of apoptotic cell death. We used J2 antibody, a gold standard for dsRNA detec- NEAT1 siRNA-mediated knockdown was equally efficient tion [43], to study the presence of dsRNA species in cells alone and in cells co-transfected with TDP-43 or Drosha depleted of TDP-43, Drosha, Dicer or ADAR1. An in- siRNA, allowing 60% NEAT1_2 downregulation and loss crease in J2-positive signal was obvious after knockdown of paraspeckles in the majority of cells (Fig. 5a and data of each of these genes (Fig. 4a and b, Additional file 3:Fig. not shown). We found that while knockdown of TDP-43 S3a). In contrast, Ago2 or FUS knockdown did not cause or Drosha alone did not result in significant cell death, dsRNA accumulation (Fig. 4b). J2 antibody was reported simultaneous disruption of paraspeckles increased the to recognise Alu repeats especially well [44]. We next used rates of apoptosis for both genes studied (Fig. 5b and c). primers which specifically detect Alu-containing RNAs Although regulation of dsRNA response by paraspeckles [45]. Dicer and Drosha but not TDP-43 or ADAR1 knock- is well documented [14], their ability to modulate down resulted in increased levels of Alu-containing RNAs dsRNA-induced apoptosis has not been addressed. We as measured by qRT-PCR (Fig. 4c) indicating that a differ- transfected cells with scrambled siRNA or NEAT1 siRNA ent repertoire of dsRNA species accumulate after knock- and subsequently exposed them to a synthetic dsRNA down of each gene. analogue poly(I:C). Cultures depleted of paraspeckles had The build-up of dsRNA is known to trigger phosphor- increased CHOP mRNA levels and higher numbers of ylation of PKR and eIF2α and activation of type I inter- CC3-positive cells after 8 h and 24 h of poly(I:C) stimula- feron (IFN) signaling. In cells transfected with TDP-43 tion, respectively (Fig. 5d-f). siRNA, levels of phosphorylated PKR and eIF2α were el- How does paraspeckle deficiency promote apoptosis in evated (Fig. 4d; Additional file 3: Fig. S3b), and the ex- cells with compromised miRNA biogenesis and activated pression of IFNB1 and an IFN-stimulated gene (ISG) dsRNA response? A whole class of cytotoxicity-associated CXCL10 was increased (Fig. 4e). Since dsRNA response IFN-stimulated genes (ISGs) were reported to be negatively eventually converges on type I IFNs, we asked whether regulated by miRNAs [48]. We found that potentially these cytokines can contribute to paraspeckle response pro-apoptotic ISGs from this class, STAT1 and CYCS,were in TDP-43 depleted cells. Firstly, we showed that IFN- significantly upregulated in paraspeckle-deficient cells de- beta is the main type I IFN induced by dsRNA in neuro- pleted of TDP-43 and Drosha (Fig. 5g). Furthermore, three blastoma cells (Additional file 4: Fig. S4a). IFNbeta such ISGs, STAT1, MYD88 and IFIH1, were consistently simulation per se was sufficient to stimulate para- upregulated in NEAT1-depleted cells in the course of speckle assembly, although the effect was transient (Fig. poly(I:C) stimulation (Fig. 5h). 4f). In line with this, ligands of TLR3 (poly(I:C)) and Thus, paraspeckles are protective against apoptotic TLR4 (bacterial lipopolysaccharide, LPS) which stimu- death in cells with compromised miRNA machinery late IFNbeta expression, but not a TLR2 ligand zymo- and/or activated dsRNA response. san which does not affect IFNbeta production, were able to boost NEAT1 expression and paraspeckle assem- Cultured human neurons lack paraspeckles but their de bly (Additional file 4: Fig. S4b-d). Finally, co-transfection novo assembly can be triggered by dsRNA of IFNbeta siRNA was sufficient to reduce paraspeckle Normal postmitotic neurons in the brain or spinal cord abundance in cells transfected with TDP-43 siRNA al- of adult mice express very low levels of NEAT1_2 iso- though did not abrogate the hyper-assembly completely form and do not form paraspeckles in vivo [21]. How- (Fig. 4g). ever it was not clear whether neurons cultured in vitro Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 10 of 17 Fig. 4 Endogenous dsRNA response and type I interferon promote paraspeckle hyper-assembly in stable cell lines. a and b Depletion of TDP-43, Dicer, Drosha, ADAR1 but not Ago2 or FUS causes intracellular build-up of dsRNA. dsRNA was detected by immunocytochemistry using J2 antibody. Representative images of all conditions are shown. Scale bars, 50 and 10 μm for general plane and close-up panels respectively. c Levels of Alu-containing RNA as analysed by qRT-PCR using specific primers recognising Alu elements (n = 4). *p < 0.05 (Mann-Whitney U-test). d and e Markers of activated cellular reponse to dsRNA are upregulated in TDP-43 depleted cells. Levels of phosphorylated PKR and eIF2α were analysed by Western blot (d, representative blots are shown) and expression of IFNB1 and an IFN-stimulated gene CXCL10 -by qRT-PCR (e, n =6). *p <0.05 (Mann-Whitney U-test). f IFNbeta treatment stimulates NEAT1 expression and paraspeckle formation. NEAT1 levels were measured by qRT-PCR (n = 6). **p < 0.01 (Mann-Whitney U-test). Staining for an IFN-inducible protein IFIT3 was used as a positive control. Scale bar, 10 μm. g Simultaneous IFNbeta knockdown partially reverses the effect of TDP-43 depletion on paraspeckles. * and #p < 0.05, ***p < 0.001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). Scale bar, 10 μm. In all panels, cells were harvested for analysis 48 h post-transfection. Paraspeckles in panels f and g were visualised by NEAT1_2 RNA-FISH would acquire and preserve paraspeckles. First, we studied and Day 16 NPCs (Fig. 6b). However, paraspeckles were primary hippocampal cultures from newborn mice. While absent even in immature (Day 23) neurons and onwards glial cells have readily detectable paraspeckles already at (Fig. 6b). In contrast, retinoic acid/BDNF-induced dif- 5 days in vitro, neurons had no sign of paraspeckles even ferentiation of SH-SY5Y neuroblastoma cells for after 14 days in vitro (Fig. 6a). Next we examined para- 6 days did not affect their ability to form paraspeckles speckle assembly during differentiation of human ES cells (Additional file 5:Fig.S5),suggesting fundamental differ- into motor neurons. Human ES cells lack paraspeckles ences in the biology of neurons and neuroblastoma-derived but they appear during differentiation (days 4–5into dif- neuron-like cells. Thus, disappearance of paraspeckles ferentiation, trophoblast stage) [49]. Consistent with this marks the transition between human NPCs and neurons, study, we started observing paraspeckles in Day 4 neural whereas de novo paraspeckle assembly is not triggered by precursor cells (NPCs), and they were prominent in Day 8 in vitro conditions in primary neurons. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 11 of 17 Fig. 5 Loss of paraspeckles promotes apoptosis in cells with disturbed miRNA biogenesis and activated dsRNA response. a-c Disruption of paraspeckles in cells with downregulated TDP-43 or Drosha promotes apoptotic death in neuroblastoma cells. Efficiency of NEAT1_2 knockdown and levels of a proapototic protein CHOP mRNA were analysed by qRT-PCR (n = 3). * and #p ≤ 0.05 (Mann-Whitney U-test) (a and b). In c, representative images and quantification of cleaved caspase 3 (CC3) positive cells are shown. *p < 0.05, ** and ##p < 0.01 (one-way ANOVA). Scrambled siRNA or NEAT1 siRNA was co-transfected with an siRNA targeting TDP-43 or Drosha, and cells analysed 36 h post-transfection. * and # indicate statistically significant difference as compared to cells transfected with only scrambled siRNA or only NEAT1 siRNA, respectively. Scale bar, 100 μm. d-f Disruption of paraspeckles promotes apoptosis in dsRNA-stimulated cells. Cells were transfected with scrambled siRNA or NEAT1 siRNA and stimulated with poly(I:C) 36 h post-transfection. Induction of CHOP by poly(I:C) over time in normal cells (d) and CHOP mRNA levels in paraspeckle-deficient and paraspeckle-sufficient cells after 8 h of poly(I:C) stimulation (e) were analysed by qRT-PCR (n = 4). **p < 0.01 (Mann- Whitney U-test). In f, representative images and quantitation of CC3-positive cells in cultures transfected with scrambled siRNA or NEAT1 siRNA and treated with poly(I:C) for 24 h are shown. *p < 0.05 (Mann-Whitney U-test). Scale bar, 100 μm. g and h Expression of cytotoxicity-associated ISGs is potentiated by loss of paraspeckles in cells depleted of TDP-43 or Drosha (g) or stimulated by poly(I:C) (h). N = 3, *p < 0.05 (Mann-Whitney U-test) Finally, we tested whether inhibition of miRNA Paraspeckle assembly can be promoted by a small function, exposure to dsRNA or IFNbeta treatment molecule enhancer of miRNA biogenesis can initiate paraspeckle formation in neurons. Hu- Very few chemicals capable of stimulating paraspeckle man Day 40 motor neurons [30] were stimulated with assembly in cells with pre-existing paraspeckles have poly(I:C) or treated with suramin or IFNbeta for 24 h. In- been identified so far. Proteasome inhibitors are known duction of endogenous IFNbeta and NEAT1 was observed to promote NEAT1 synthesis and paraspeckle formation in poly(I:C)-treated neuronal cultures (Fig. 6c), and para- [13], however they are poor candidates as therapeutic speckles could be detected in a small fraction of poly(- molecules for ALS. Enoxacin is a small molecule enhan- I:C)-stimulated neurons (Fig. 6d). In contrast, neither cer of the miRNA pathway which stimulates the activity suramin nor IFNbeta induced paraspeckle assembly in of the Dicer complex [50, 51]. In doing so, enoxacin in- neurons. creases levels of mature miRNAs thereby depleting Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 12 of 17 Fig. 6 Post-mitotic neurons lack paraspeckles in vitro, but their assembly can be triggered by dsRNA. a Paraspeckles are present in glial cells but not in neurons in murine primary hippocampal cultures. Cultures were analysed at DIV14 by NEAT1_2 RNA-FISH. A representative image is shown. Neuronal nuclei are circled and paraspeckles in glial cells are indicated by arrowheads. b Paraspeckles are present in human neural precursor cells (NPCs) but disappear during their differentiation into motor neurons. Cultures were analysed at the indicated time-points by NEAT1_2 RNA-FISH. c and d Treatment of Day 40 cultures of human motor neurons with poly(I:C) leads to activation of IFN signaling, increased NEAT1 expression (c) and paraspeckle assembly in a fraction of cells (d, arrowheads). In c, gene expression was analysed by qRT-PCR (n = 4). *p < 0.05, **p < 0.01 (Mann-Whitney U-test). Scale bars, 10 μmin a, b and 20 μmin d miRNA precursors, pri-miRNA and pre-miRNA [51]. Re- reported to stimulate NEAT1 expression [53]and twoap- cently, enoxacin been shown to ameliorate pathology in proved ALS therapeutics, riluzole and edaravone. mouse models of ALS [52]. We studied the effect of enox- Treatment with 10 μM enoxacin for 24 h increased acin on paraspeckles in neuroblastoma cells. In addition, NEAT1_2 levels and paraspeckle assembly in neuroblast- in our analysis we included HDAC inhibitors which were oma cells (Fig. 7a and b); similar effect on paraspeckles Fig. 7 A small molecule enhancer of miRNA biogenesis stimulates paraspeckle assembly in neuroblastoma cells. a and b SH-SY5Y cells were treated with enoxacin, riluzole, edaravone and HDAC inhibitors trichostatin A (TSA) and sodium butyrate (NaB), and paraspeckle assembly was assessed by NEAT1_2 RNA-FISH (a) and qRT-PCR (n = 4–6, b). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Kruskal-Wallis test with Dunn’s correction for multiple comparisons). Cells were harvested for analysis after 4 h of TSA and NaB treatment (500 nM and 2 mM, respectively) and after 24 h of enoxacin, riluzole and edaravone treatment (all 10 μM). Scale bar, 10 μm Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 13 of 17 was observed with short (4 h) treatment and with a response. Given a significant crosstalk between miRNA higher enoxacin dose (Additional file 6: Fig. S6). Global and dsRNA response pathways [64], it is not surprising HDAC inhibitors trichostatin A (TSA) and sodium bu- that paraspeckles function as a regulatory platform for tyrate (NaB) also significantly increased NEAT1_2 levels both pathways. Although dsRNA response triggered by and led to the formation of large, elongated paraspeckles dysfunction of the miRNA pathway factors is mediated (Fig. 7a and b). Interestingly, edaravone but not riluzole via different molecular sensors, including TLR3 (for also promoted paraspeckle biogenesis (Fig. 7a and b). Drosha, our unpublished observations), MyD88 (for None of the compounds studied was able to trigger de Dicer) [65] and MDA5/RIG-I (for ADAR1) [66], it even- novo paraspeckles in hES cell derived motor neurons. tually converges on type I IFN. In contrast to the previ- ous study [14], we found that IFNbeta treatment alone Discussion can stimulate NEAT1 expression and paraspeckle forma- Nuclear bodies spatially organize and modulate various tion. This discrepancy is likely due to the transient effect cellular processes [54]. Therefore it is not surprising that of IFN treatment on paraspeckles which peaks at the 4-h these membraneless organelles and their components time-point, whereas in the previous work, the 24 h have been implicated in multiple human diseases. Prom- time-point was examined. It is possible that IFN levels inent examples are PML bodies and Gems linked to car- oscillate to maintain the dsRNA response active but at cinogenesis and motor neuron degeneration, respectively the same time preserve cellular viability [67]. Our in [55, 56]. Paraspeckles have recently come into the lime- vitro data are consistent with a recent in vivo study dem- light in the ALS field because of the extensive involve- onstrating that TDP-43 knockdown in the adult murine ment of paraspeckle proteins in ALS pathogenesis. In nervous system leads to widespread upregulation of im- the present study, we found that paraspeckle assembly mune and, more specifically, antiviral genes [27]. Loss of in the spinal cord is shared by ALS cases with different TDP-43 function in the nervous system might be suffi- aetiology and, as such, a hallmark of the disease. Using cient to trigger a chronic neuroinflammatory response. cell models, we identified two possible mechanisms In a previous study, siRNA-mediated TDP-43 knock- which may initiate paraspeckle assembly in the spinal down led to decreased paraspeckle numbers in HeLa cord cells of the majority of ALS cases – compromised cells [8]. One possible explanation for this discrepancy is miRNA biogenesis and activated dsRNA response – differences in cellular response to dsRNA and/or differ- both downstream of loss of TDP-43 function. ences in the miRNA pathway regulation between the cell In the CNS, miRNAs are highly abundant and are sub- lines. Indeed, in the study on TDP-43 functions as an ject to abnormal regulation in many neurodegenerative RNA chaperone, dsRNA was shown to be accumulated diseases, including ALS [57–61]. Levels of mature miR- only in the nucleus of HeLa cells, whereas in neuroblast- NAs in ALS spinal cord were reported to be globally re- oma M17 cell line it was mainly cytoplasmic [38], similar duced [52, 57, 62]. This dysregulation is consistent with to our study. Another possibility is the reliance of para- TDP-43 loss of function in the majority of ALS cases speckle assembly on some TDP-43 function(s) specific- since this protein is a known miRNA biogenesis factor ally in HeLa cells. [32]. An important role of paraspeckles in miRNA pro- Loss of TDP-43 function can explain paraspeckle cessing is supported by two recent studies [16, 63]. hyper-assembly in the majority of ALS cases, i.e. almost Thus, paraspeckle hyper-assembly in ALS motor neu- all sALS cases as well as fALS cases caused by mutations rons affected by TDP-43 loss of function may serve as in TARDBP and C9ORF72 genes. Recently, Drosha has one of the mechanisms to compensate for miRNA bio- been identified as a component of C9orf72 dipeptide in- genesis deficiency. We also show that not only compro- clusions in patient’s neurons [68]. Therefore in fALS-C9 mised function of the miRNA pathway but also its cases, loss of function for both TDP-43 and Drosha can pharmacological enhancement results in paraspeckle jointly contribute to paraspeckle response. In a subset of hyper-assembly. This suggests that paraspeckles can re- sALS patients, activation of an endogenous retrovirus spond to bi-directional changes in the activity of the (ERV), HERV-K, was reported [69]. Elevated expression miRNA pathway to either compensate for its compro- of ERVs can initiate dsRNA response [70, 71]. Activation mised function or to meet the demand for miRNA pre- of HERV-K may therefore contribute to paraspeckle cursors when its final step is over-active. hyper-assembly at least in some sALS cases. It still re- Another function of TDP-43 is acting as a chaperone mains to be established whether paraspeckle formation to control RNA secondary structure [38] and therefore is typical for other fALS cases such as those caused by cellular dsRNA response. Paraspeckles are known to re- mutations in genes encoding SOD1, FUS, TBK1 or spond to exogenous dsRNA (viral and its analogues) OPTN and, if so, the underlying mechanisms. Many of [14], and here we show that abnormal accumulation of ALS proteins function in miRNA and dsRNA metabol- endogenous dsRNA can also initiate paraspeckle ism, for example FUS is involved in miRNA biogenesis Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 14 of 17 and miRNA-mediated silencing [35, 72], whereas TBK1 scrambled siRNA or Silencer® TDP-43 siRNA and analysed by qRT-PCR (n =6). is one of the central factors in dsRNA response and type Mean number of paraspeckles per cell was also quantified using NEAT1_2 RNA FISH. **p <0.01, ***p < 0.001 (Mann-Whitney U-test in a and Student’s t- I IFN signaling. Thus compromised function of these test in b). c and d Downregulation of TDP-43 using an esiRNA (endoribonu- pathways may represent a common mechanism behind clease-prepared MISSION® esiRNA, c) or shRNA (d) stimulates paraspeckle paraspeckle response in different ALS cases. assembly. The efficiency of knockdown was analysed by TDP-43 immunocyto- chemistry, and paraspeckle assembly – by NEAT1_2 RNA-FISH. Representative Previously, paraspeckles were shown to be protective images are shown. Scale bars, 100 μm for left panels and 10 μmfor right against cell death caused by proteasomal inhibition [13]. panels. e Speckles visualised by MALAT1 RNA-FISH are not affected by TDP-43 In current study, we show that paraspeckles confer pro- knockdown. Representative images are shown. Scale bar, 10 μm. f Levels of core paraspeckle proteins NONO, SFPQ and FUS are not affected by TDP-43 tection to cells with compromised metabolism of miRNA knockdown in SH-SY5Y cells. Representative Western blots are shown. g Se- and activated dsRNA response. Intriguingly, many of quences and positions of gRNAs used for disrupting the NLS of the endogen- miRNA-controlled cytotoxicity-associated ISGs [48]were ous TDP-43 protein by CRISPR/Cas9-mediated editing. Two combinations of upstream and downstream gRNA sequences within TARDBP gene selected to also reported to be regulated by paraspeckles either by se- disrupt the NLS are shown. The sequence encoding for the NLS is given in questration of transcription factors or by nuclear retention blue and PAM sites are boxed. h Transient transfection of two combinations of edited RNAs [14, 49, 73] suggesting a multi-layered of plasmids encoding upstream and downstream gRNAs for targeting the NLS of TDP-43 results in partial redistribution of endogenous TDP-43 but does not control of cellular toxicity by paraspeckles. It should be lead to enhanced paraspeckle formation. Cells were analysed 72 h post- noted however that the effect of paraspeckle disruption on transfection. Representative images are shown, asterisks indicate cells with survival in stable cell lines was small both in this and in cytoplasmic TDP-43 redistribution. Scale bar, 10 μm. In a-f, cells were analysed 48 h post-transfection. (DOCX 517 kb) the previous [13] report, despite the use of cells com- Additional file 2: Figure S2. Drosha or TDP-43 downregulation affects pletely lacking NEAT1 and hence paraspeckles in the lat- miRNA processing. a Drosha knockdown leads to accumulation of miRNA pre- ter study. Such limited effect is in line with the fact that cursors, pri-miR-17-92a and pri-miR-15a (n =4–6). **p < 0.01 (one-way ANOVA NEAT1 knockout mice do not have an overt phenotype with Holm-Sidak correction for multiple comparisons). Note the absence of significant accumulation of these pri-miRNAs in TDP-43 depleted cells, in ac- [21] and further supports a modulatory role for para- cord with modulatory rather than essential function of this protein in miRNA speckles in cellular responses (such as miRNA biogenesis, processing in the nucleus. b Drosha or TDP-43 knockdown leads to downreg- gene expression, RNA retention) which only becomes ulation of mature miRNAs processed from pri-miR-17-92a (n =3). *p <0.05; **p < 0.01 (one-way ANOVA with Holm-Sidak correction for multiple compari- relevant under stressful/pathophysiological conditions. sons). Note that levels of all three mature miRNAs are significantly decreased However, such modulatory activities of paraspeckles might in Drosha depleted cells, and TDP-43 knockdown also negatively affects two be particularly important for neurons coping with of the three miRNAs measured. (DOCX 157 kb) neurodegeneration-inducing stresses. Additional file 3: Figure S3. Accumulation of dsRNA (a) and increased levels of p-eIF2α (b) in MCF7 cells depleted of TDP-43. Cells were analysed 48 h post-transfection. Scale bars, 100 μm and 10 μm for general plane and Conclusions close-up panels respectively. (DOCX 663 kb) Paraspeckle hyper-assembly might be broadly neuropro- Additional file 4: Figure S4. The effect of IFN-inducing ligands on NEAT1 and paraspeckles. a IFNbeta is robustly induced by poly(I:C) in tective in ALS. As a word of caution, however, these data neuroblastoma cells. Cells were analysed by qRT-PCR after 24 h of were obtained in stable cell lines with pre-existing para- poly(I:C) stimulation (n = 5). **p < 0.01 (Mann-Whitney U-test). b-d TLR3 speckles. Since normal post-mitotic neurons are free from and TLR4 ligands poly(I:C) and LPS, but not a TLR2 ligand zymosan, trigger IFNbeta response stimulating NEAT1 expression (b) and para- paraspeckles, the impact of their de novo formation on speckle assembly (c). Cells were treated with poly(I:C), LPS or zymosan for neuronal metabolism may be much more dramatic than 4 h and analysed by qRT-PCR (n = 3 or 4). *p < 0.05, **p < 0.01. NF-κB in stable cell lines. For example, it is coupled with changes nuclear translocation was examined in parallel to confirm the activity of the compounds (d, asterisks indicate cells with nuclear NF-κB). Scale bar, in the levels of the short NEAT1 isoform with diverse 10 μm. (DOCX 333 kb) paraspeckle-independent regulatory functions, including Additional file 5: Figure S5. Differentiation of human modulation of neuronal excitability [74–77]. Further stud- neuroblastoma cells into neuron-like cells does not lead to the loss ies, using neurons derived from human stem cells with ab- of paraspeckles. Differentiated SH-SY5Y cells develop extensive neurite network and are uniformly positive for a neuronal marker lated NEAT1_2 expression and from ALS patients’ iPS Tuj 1 (left panel) but preserve their ability to form paraspeckles cells, are required to understand whether paraspeckles are (right panel). SH-SY5Y cells were induced to differentiate into protective for motor neurons in the disease context and in neuron-like cells using retinoic acid/BDNF andanalysed6days into differentiation by immunocytochemitry and NEAT1_2 RNA-FISH. the long-term. This shall help us understand whether, how Representative images are shown. Scale bars, 100 μm(left panel) and when paraspeckles can be targeted for therapeutic and 10 μm (right panel). (DOCX 175 kb) purposes. Additional file 6: Figure S6. Dose-dependent toxicity of enoxacin, edaravone and riluzole. a SH-SY5Y cells were treated with correspond- ing doses of compounds for 24 h, and toxicity was assessed using Cell- Additional files Titer Blue® Cell Viability Assay. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as compared to control (non-treated) cells (Kruskal-Wallis test Additional file 1: Figure S1. The effect of TDP-43 dysfunction on para- with Dunn’s correction for multiple comparisons). b Enoxacin enhances paraspeckle assembly both with short (4 h) and prolonged (24 h) treat- speckles, speckles and paraspeckle proteins in MCF7 and SH-SY5Y cells. a and b TDP-43 siRNA-mediated knockdown upregulates NEAT1_2 (a) and enhances ment at a non-toxic concentration of 50 μM. Representative images are paraspeckle assembly (b) in SH-SY5Y cells. Cells were transfected with shown. (DOCX 298 kb) Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 15 of 17 Abbreviations 5. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, et (F)ISH: (fluorescent) in situ hybridisation; ALS: Amyotrophic lateral sclerosis; al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and dsRNA: double-stranded RNA; FTLD: Frontotemporal lobar degeneration; amyotrophic lateral sclerosis. Science. 2006;314:130–3. https://doi.org/10. HDAC: Histone deacetylase; IFN: Interferon; ISG: Interferon-stimulated gene; 1126/science.1134108. LPS: Lipopolysaccharide; miRNA: microRNA; NEAT1: Nuclear enriched 6. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, et al. TDP-43 is a abundant transcript 1; NPC: Neural precursor cell; component of ubiquitin-positive tau-negative inclusions in frontotemporal poly(I:C): Polyinosinic:polycytidylic acid; TDP-43: TAR DNA-binding protein 43; lobar degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res TSA: Trichostatin A Commun. 2006;351:602–11. https://doi.org/10.1016/j.bbrc.2006.10.093. 7. Fox AH, Lamond AI. Paraspeckles. Cold Spring Harb Perspect Biol. 2010;2: Acknowledgements a000687. https://doi.org/10.1101/cshperspect. a000687. We acknowledge the Sheffield Brain Tissue Bank and London 8. Naganuma T, Nakagawa S, Tanigawa A, Sasaki YF, Goshima N, Hirose T. Neurodegenerative Diseases Brain Bank for providing human materials. Alternative 3′-end processing of long noncoding RNA initiates construction of nuclear paraspeckles. EMBO J. 2012;31:4020–34. https://doi.org/10.1038/ emboj.2012.251. Funding 9. West JA, Mito M, Kurosaka S, Takumi T, Tanegashima C, Chujo T, et al. TAS is a recipient of a fellowship from Medical Research Foundation. The Structural, super-resolution microscopy analysis of paraspeckle nuclear body study was also funded by Research Grant from Motor Neuron Disease organization. J Cell Biol. 2016; https://doi.org/10.1083/jcb.201601071. Association to VLB (Buchman/Apr13/6096) and miRNA analysis was 10. Clemson CM, Hutchinson JN, Sara SA, Ensminger AW, Fox AH, Chess A, et al. supported by the Russian Science Foundation grant (18–15-00357). HA and An architectural role for a nuclear noncoding RNA: NEAT1 RNA is essential PD are recipients of Cardiff University/China Council PhD studentship and for the structure of paraspeckles. Mol Cell. 2009;33:717–26. https://doi.org/ Erasmus studentship, respectively. 10.1016/j.molcel.2009.01.026. 11. Hennig S, Kong G, Mannen T, Sadowska A, Kobelke S, Blythe A, et al. Prion- Availability of data and materials like domains in RNA binding proteins are essential for building subnuclear The datasets used and/or analyzed during the current study are available paraspeckles. J Cell Biol. 2015;210:529–39. https://doi.org/10.1083/jcb. from the corresponding author on reasonable request. 12. Sunwoo H, Dinger ME, Wilusz JE, Amaral PP, Mattick JS. Spector DL. MEN Authors’ contributions epsilon/beta nuclear-retained non-coding RNAs are up-regulated upon TAS conceived research; TAS, HA, MSK, PD, SA, OS and PRH performed muscle differentiation and are essential components of paraspeckles. experiments; TAS and VLB analysed data; TAS wrote manuscript with input Genome Res. 2009;19:347–59. https://doi.org/10.1101/gr.087775.108. from all authors. All authors read and approved the final version of the 13. Hirose T, Virnicchi G, Tanigawa A, Naganuma T, Li R, Kimura H, et al. NEAT1 manuscript. long noncoding RNA regulates transcription via protein sequestration within subnuclear bodies. Mol Biol Cell. 2014;25:169–83. https://doi.org/10. Ethics approval and consent to participate 1091/mbc.E13-09-0558. Human samples from clinically and histopathologically characterised ALS 14. Imamura K, Imamachi N, Akizuki G, Kumakura M, Kawaguchi A, Nagata K, et al. cases and neurologically healthy individuals were provided by the Sheffield Long noncoding RNA NEAT1-dependent SFPQ relocation from promoter Brain Tissue Bank and MRC London Neurodegenerative Diseases Brain Bank region to paraspeckle mediates IL8 expression upon immune stimuli. Mol Cell. (Institute of Psychiatry, King’s College London). Consent was obtained from 2014;53:393–406. https://doi.org/10.1016/j.molcel.2014.01.009. all subjects for autopsy, histopathological assessment and research were 15. Zhang Z, Carmichael GG. The fate of dsRNA in the nucleus: a p54(nrb)- performed in accordance with local and national Ethics Committee containing complex mediates the nuclear retention of promiscuously A-to-I approved donation. edited RNAs. Cell. 2001;106:465–75. http://www.ncbi.nlm.nih.gov/pubmed/ Competing interests 16. Jiang L, Shao C, Wu QJ, Chen G, Zhou J, Yang B, et al. NEAT1 scaffolds RNA- The authors declare that they have no competing interests. binding proteins and the microprocessor to globally enhance pri-miRNA processing. Nat Struct Mol Biol. 2017;24:816–24. https://doi.org/10.1038/ nsmb.3455. Publisher’sNote 17. Choudhry H, Albukhari A, Morotti M, Haider S, Moralli D, Smythies J, et al. Springer Nature remains neutral with regard to jurisdictional claims in Tumor hypoxia induces nuclear paraspeckle formation through HIF-2alpha published maps and institutional affiliations. dependent transcriptional activation of NEAT1 leading to cancer cell survival. Oncogene. 2015;34:4546. https://doi.org/10.1038/onc.2014.431. Author details 18. Chakravarty D, Sboner A, Nair SS, Giannopoulou E, Li R, Hennig S, et al. The School of Biosciences, Cardiff University, Museum Avenue, Cardiff CF10 3AX, oestrogen receptor alpha-regulated lncRNA NEAT1 is a critical modulator of UK. Institute of Physiologically Active Compounds Russian Academy of prostate cancer. Nat Commun. 2014;5:5383. https://doi.org/10.1038/ncomms6383. Sciences, 1 Severniy proezd, Chernogolovka, Moscow Region, Russian 19. Adriaens C, Standaert L, Barra J, Latil M, Verfaillie A, Kalev P, et al. p53 Federation142432. The Sheffield Institute for Translational Neuroscience, induces formation of NEAT1 lncRNA-containing paraspeckles that modulate 385A Glossop Road, Sheffield S10 2HQ, UK. replication stress response and chemosensitivity. Nat Med. 2016;22:861–8. https://doi.org/10.1038/nm.4135. Received: 13 March 2018 Accepted: 25 May 2018 20. Nishimoto Y, Nakagawa S, Hirose T, Okano HJ, Takao M, Shibata S, et al. The long non-coding RNA nuclear-enriched abundant transcript 1_2 induces paraspeckle formation in the motor neuron during the early phase of References amyotrophic lateral sclerosis. Molecular brain. 2013;6:31. https://doi.org/10. 1. Renton AE, Chio A, Traynor BJ. State of play in amyotrophic lateral sclerosis 1186/1756-6606-6-31. genetics. Nat Neurosci. 2014;17:17–23. https://doi.org/10.1038/nn.3584. 21. Nakagawa S, Naganuma T, Shioi G, Hirose T. Paraspeckles are 2. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, subpopulation-specific nuclear bodies that are not essential in mice. J Cell Rutherford NJ, et al. Expanded GGGGCC hexanucleotide repeat in Biol. 2011;193:31–9. https://doi.org/10.1083/jcb.201011110. noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. 22. Thomas-Jinu S, Gordon PM, Fielding T, Taylor R, Smith BN, Snowden V, et al. Neuron. 2011;72:245–56. https://doi.org/10.1016/j.neuron.2011.09.011. Non-nuclear pool of splicing factor SFPQ regulates axonal transcripts 3. Sreedharan J, Blair IP, Tripathi VB, Hu X, Vance C, Rogelj B, et al. TDP-43 required for normal motor development. Neuron. 2017;94:931. https://doi. mutations in familial and sporadic amyotrophic lateral sclerosis. Science. org/10.1016/j.neuron.2017.04.036. 2008;319:1668–72. https://doi.org/10.1126/science.1154584. 23. Shelkovnikova TA, Robinson HK, Troakes C, Ninkina N, Buchman VL. 4. Lattante S, Rouleau GA, Kabashi E. TARDBP and FUS mutations associated Compromised paraspeckle formation as a pathogenic factor in FUSopathies. with amyotrophic lateral sclerosis: summary and update. Hum Mutat. 2013; Hum Mol Genet. 2014;23:2298–312. https://doi.org/10.1093/hmg/ddt622. 34:812–26. https://doi.org/10.1002/humu.22319. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 16 of 17 24. Chesi A, Staahl BT, Jovicic A, Couthouis J, Fasolino M, Raphael AR, et al. 44. Kaneko H, Dridi S, Tarallo V, Gelfand BD, Fowler BJ, Cho WG, et al. DICER1 Exome sequencing to identify de novo mutations in sporadic ALS trios. Nat deficit induces Alu RNA toxicity in age-related macular degeneration. Neurosci. 2013;16:851–5. https://doi.org/10.1038/nn.3412. Nature. 2011;471:325–30. https://doi.org/10.1038/nature09830. 25. Kim HJ, Kim NC, Wang YD, Scarborough EA, Moore J, Diaz Z, et al. 45. Marullo M, Zuccato C, Mariotti C, Lahiri N, Tabrizi SJ, Di Donato S, et al. Mutations in prion-like domains in hnRNPA2B1 and hnRNPA1 cause Expressed Alu repeats as a novel, reliable tool for normalization of real-time multisystem proteinopathy and ALS. Nature. 2013;495:467–73. https://doi. quantitative RT-PCR data. Genome Biol. 2010;11:R9. https://doi.org/10.1186/ org/10.1038/nature11922. gb-2010-11-1-r9. 26. Tollervey JR, Curk T, Rogelj B, Briese M, Cereda M, Kayikci M, et al. 46. Marciniak SJ, Yun CY, Oyadomari S, Novoa I, Zhang Y, Jungreis R, et al. Characterizing the RNA targets and position-dependent splicing regulation CHOP induces death by promoting protein synthesis and oxidation in the by TDP-43. Nat Neurosci. 2011;14:452–8. https://doi.org/10.1038/nn.2778. stressed endoplasmic reticulum. Genes Dev. 2004;18:3066–77. https://doi. 27. Polymenidou M, Lagier-Tourenne C, Hutt KR, Huelga SC, Moran J, Liang TY, org/10.1101/gad.1250704. et al. Long pre-mRNA depletion and RNA missplicing contribute to 47. Matsumoto M, Minami M, Takeda K, Sakao Y, Akira S. Ectopic expression of neuronal vulnerability from loss of TDP-43. Nat Neurosci. 2011;14:459–68. CHOP (GADD153) induces apoptosis in M1 myeloblastic leukemia cells. FEBS https://doi.org/10.1038/nn.2779. Lett 1996; 395:143–147. doi: 0014–5793(96)01016–2. 28. Kukharsky MS, Quintiero A, Matsumoto T, Matsukawa K, An H, Hashimoto T, 48. Seo GJ, Kincaid RP, Phanaksri T, Burke JM, Pare JM, Cox JE, et al. Reciprocal et al. Calcium-responsive transactivator (CREST) protein shares a set of inhibition between intracellular antiviral signaling and the RNAi machinery structural and functional traits with other proteins associated with in mammalian cells. Cell Host Microbe. 2013;14:435–45. https://doi.org/10. amyotrophic lateral sclerosis. Mol Neurodegener. 2015;10:20. https://doi.org/ 1016/j.chom.2013.09.002. 10.1186/s13024-015-0014-y. 49. Chen LL, Carmichael GG. Altered nuclear retention of mRNAs containing 29. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome inverted repeats in human embryonic stem cells: functional role of a engineering using CRISPR/Cas systems. Science. 2013;339:819–23. https:// nuclear noncoding RNA. Mol Cell. 2009;35:467–78. https://doi.org/10.1016/j. doi.org/10.1126/science.1231143. molcel.2009.06.027. 30. Shelkovnikova TA, Dimasi P, Kukharsky MS, An H, Quintiero A, Schirmer C, et 50. Melo S, Villanueva A, Moutinho C, Davalos V, Spizzo R, Ivan C, et al. Small al. Chronically stressed or stress-preconditioned neurons fail to maintain molecule enoxacin is a cancer-specific growth inhibitor that acts by stress granule assembly. Cell Death Dis. 2017;8:e2788. https://doi.org/10. enhancing TAR RNA-binding protein 2-mediated microRNA processing. Proc 1038/cddis.2017.199. Natl Acad Sci U S A. 2011;108:4394–9. https://doi.org/10.1073/pnas. 31. Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, Gibbs JR, et 1014720108. al. A hexanucleotide repeat expansion in C9ORF72 is the cause of 51. Shan G, Li Y, Zhang J, Li W, Szulwach KE, Duan R, et al. A small molecule chromosome 9p21-linked ALS-FTD. Neuron. 2011;72:257–68. https://doi.org/ enhances RNA interference and promotes microRNA processing. Nat 10.1016/j.neuron.2011.09.010. Biotechnol. 2008;26:933–40. https://doi.org/10.1038/nbt.1481. 32. Kawahara Y, Mieda-Sato A. TDP-43 promotes microRNA biogenesis as a 52. Emde A, Eitan C, Liou LL, Libby RT, Rivkin N, Magen I, et al. Dysregulated component of the Drosha and dicer complexes. Proc Natl Acad Sci U S A. miRNA biogenesis downstream of cellular stress and ALS-causing mutations: 2012;109:3347–52. https://doi.org/10.1073/pnas.1112427109. a new mechanism for ALS. EMBO J. 2015;34:2633–51. https://doi.org/10. 15252/embj.201490493. 33. Buratti E, De Conti L, Stuani C, Romano M, Baralle M, Baralle F. Nuclear factor TDP-43 can affect selected microRNA levels. FEBS J. 2010;277:2268–81. 53. Schor IE, Lleres D, Risso GJ, Pawellek A, Ule J, Lamond AI, et al. Perturbation https://doi.org/10.1111/j.1742-4658.2010.07643.x. of chromatin structure globally affects localization and recruitment of 34. Ota H, Sakurai M, Gupta R, Valente L, Wulff BE, Ariyoshi K, et al. ADAR1 splicing factors. PLoS One. 2012;7:e48084. https://doi.org/10.1371/journal. forms a complex with dicer to promote microRNA processing and RNA- pone.0048084. induced gene silencing. Cell. 2013;153:575–89. https://doi.org/10.1016/j.cell. 54. Misteli T. Higher-order genome organization in human disease. Cold Spring 2013.03.024. Harb Perspect Biol. 2010;2:a000794. https://doi.org/10.1101/cshperspect. 35. Morlando M, Dini Modigliani S, Torrelli G, Rosa A, Di Carlo V, Caffarelli E, et a000794. al. FUS stimulates microRNA biogenesis by facilitating co-transcriptional 55. de The H, Le Bras M, Lallemand-Breitenbach V. The cell biology of disease: Drosha recruitment. EMBO J. 2012;31:4502–10. https://doi.org/10.1038/ acute promyelocytic leukemia, arsenic, and PML bodies. J Cell Biol 2012;198: emboj.2012.319. 11–21. doi: https://doi.org/10.1083/jcb.201112044. 36. Tan GS, Chiu CH, Garchow BG, Metzler D, Diamond SL, Kiriakidou M. Small 56. Yamazaki T, Chen S, Yu Y, Yan B, Haertlein TC, Carrasco MA, et al. FUS-SMN molecule inhibition of RISC loading. ACS Chem Biol. 2012;7:403–10. https:// protein interactions link the motor neuron diseases ALS and SMA. Cell Rep. doi.org/10.1021/cb200253h. 2012;2:799–806. https://doi.org/10.1016/j.celrep.2012.08.025. 37. Lagier-Tourenne C, Polymenidou M, Hutt KR, Vu AQ, Baughn M, Huelga SC, 57. Figueroa-Romero C, Hur J, Lunn JS, Paez-Colasante X, Bender DE, Yung R, et et al. Divergent roles of ALS-linked proteins FUS/TLS and TDP-43 intersect in al. Expression of microRNAs in human post-mortem amyotrophic lateral processing long pre-mRNAs. Nat Neurosci. 2012;15:1488–97. https://doi.org/ sclerosis spinal cords provides insight into disease mechanisms. Mol Cell 10.1038/nn.3230. Neurosci. 2016;71:34–45. https://doi.org/10.1016/j.mcn.2015.12.008. 38. Saldi TK, Ash PE, Wilson G, Gonzales P, Garrido-Lecca A, Roberts CM, et al. 58. Freischmidt A, Muller K, Ludolph AC, Weishaupt JH. Systemic dysregulation TDP-1, the Caenorhabditis elegans ortholog of TDP-43, limits the of TDP-43 binding microRNAs in amyotrophic lateral sclerosis. Acta accumulation of double-stranded RNA. EMBO J. 2014;33:2947–66. https:// Neuropathol Commun. 2013;1:42. https://doi.org/10.1186/2051-5960-1-42. doi.org/10.15252/embj.201488740. 59. Gascon E, Gao FB. The emerging roles of microRNAs in the pathogenesis of 39. Ma H, Han P, Ye W, Chen H, Zheng X, Cheng L, et al. The long noncoding frontotemporal dementia-amyotrophic lateral sclerosis (FTD-ALS) spectrum RNA NEAT1 exerts Antihantaviral effects by acting as positive feedback for disorders. J Neurogenet. 2014;28:30–40. https://doi.org/10.3109/01677063. RIG-I signaling. J Virol. 2017;91 https://doi.org/10.1128/JVI.02250-16. 2013.876021. 40. White E, Schlackow M, Kamieniarz-Gdula K, Proudfoot NJ, Gullerova M. Human 60. Goodall EF, Heath PR, Bandmann O, Kirby J, Shaw PJ. Neuronal dark matter: nuclear dicer restricts the deleterious accumulation of endogenous double-stranded the emerging role of microRNAs in neurodegeneration. Front Cell Neurosci. RNA. Nat Struct Mol Biol. 2014;21:552–9. https://doi.org/10.1038/nsmb.2827. 2013;7:178. https://doi.org/10.3389/fncel.2013.00178. 41. Heras SR, Macias S, Plass M, Fernandez N, Cano D, Eyras E, et al. The 61. Eitan C, Hornstein E. Vulnerability of microRNA biogenesis in FTD-ALS. Brain microprocessor controls the activity of mammalian retrotransposons. Nat Res. 2016;1647:105–11. https://doi.org/10.1016/j.brainres.2015.12.063. Struct Mol Biol. 2013;20:1173–81. https://doi.org/10.1038/nsmb.2658. 62. Campos-Melo D, Droppelmann CA, He Z, Volkening K, Strong MJ. Altered 42. Liddicoat BJ, Piskol R, Chalk AM, Ramaswami G, Higuchi M, Hartner JC, et al. microRNA expression profile in amyotrophic lateral sclerosis: a role in the RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as regulation of NFL mRNA levels. Molecular brain. 2013;6:26. https://doi.org/ nonself. Science. 2015;349:1115–20. https://doi.org/10.1126/science.aac7049. 10.1186/1756-6606-6-26. 43. Weber F, Wagner V, Rasmussen SB, Hartmann R, Paludan SR. Double- 63. Bottini S, Hamouda-Tekaya N, Mategot R, Zaragosi LE, Audebert S, Pisano S, stranded RNA is produced by positive-strand RNA viruses and DNA viruses et al. Post-transcriptional gene silencing mediated by microRNAs is but not in detectable amounts by negative-strand RNA viruses. J Virol. 2006; controlled by nucleoplasmic Sfpq. Nat Commun. 2017;8:1189. https://doi. 80:5059–64. https://doi.org/10.1128/JVI.80.10.5059-5064.2006. org/10.1038/s41467-017-01126-x. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 17 of 17 64. Heyam A, Lagos D, Plevin M. Dissecting the roles of TRBP and PACT in double-stranded RNA recognition and processing of noncoding RNAs. Wiley Interdiscip Rev RNA. 2015;6:271–89. https://doi.org/10.1002/wrna.1272. 65. Tarallo V, Hirano Y, Gelfand BD, Dridi S, Kerur N, Kim Y, et al. DICER1 loss and Alu RNA induce age-related macular degeneration via the NLRP3 inflammasome and MyD88. Cell. 2012;149:847–59. https://doi.org/10.1016/j. cell.2012.03.036. 66. Mannion NM, Greenwood SM, Young R, Cox S, Brindle J, Read D, et al. The RNA-editing enzyme ADAR1 controls innate immune responses to RNA. Cell Rep. 2014;9:1482–94. https://doi.org/10.1016/j.celrep.2014.10.041. 67. Ruggieri A, Dazert E, Metz P, Hofmann S, Bergeest JP, Mazur J, et al. Dynamic oscillation of translation and stress granule formation mark the cellular response to virus infection. Cell Host Microbe. 2012;12:71–85. https://doi.org/10.1016/j.chom.2012.05.013. 68. Porta S, Kwong LK, Trojanowski JQ, Lee VM. Drosha inclusions are new components of dipeptide-repeat protein aggregates in FTLD-TDP and ALS C9orf72 expansion cases. J Neuropathol Exp Neurol. 2015;74:380–7. https:// doi.org/10.1097/NEN.0000000000000182. 69. Li W, Lee MH, Henderson L, Tyagi R, Bachani M, Steiner J, et al. Human endogenous retrovirus-K contributes to motor neuron disease. Sci Transl Med. 2015;7:307ra153. https://doi.org/10.1126/scitranslmed.aac8201. 70. Hurst TP, Magiorkinis G. Activation of the innate immune response by endogenous retroviruses. J Gen Virol. 2015;96:1207–18. https://doi.org/10. 1099/jgv.0.000017. 71. Chiappinelli KB, Strissel PL, Desrichard A, Li H, Henke C, Akman B, et al. Inhibiting DNA methylation causes an interferon response in Cancer via dsRNA including endogenous retroviruses. Cell. 2016;164:1073. https://doi. org/10.1016/j.cell.2015.10.020. 72. Zhang T, Wu YC, Mullane P, Ji YJ, Liu H, He L, et al. FUS regulates activity of MicroRNA-mediated gene silencing. Mol Cell 2018; 69:787–801 e8. doi: https://doi.org/10.1016/j.molcel.2018.02.001. 73. Elbarbary RA, Li W, Tian B, Maquat LE. STAU1 binding 3' UTR IRAlus complements nuclear retention to protect cells from PKR-mediated translational shutdown. Genes Dev. 2013;27:1495–510. https://doi.org/10. 1101/gad.220962.113. 74. Barry G, Briggs JA, Hwang DW, Nayler SP, Fortuna PR, Jonkhout N, et al. The long non-coding RNA NEAT1 is responsive to neuronal activity and is associated with hyperexcitability states. Sci Rep. 2017;7:40127. https://doi. org/10.1038/srep40127. 75. Li R, Harvey AR, Hodgetts SI, Fox AH. Functional dissection of NEAT1 using genome editing reveals substantial localisation of the NEAT1_1 isoform outside paraspeckles. RNA. 2017; https://doi.org/10.1261/rna.059477.116. 76. Zhang F, Wu L, Qian J, Qu B, Xia S, La T, et al. Identification of the long noncoding RNA NEAT1 as a novel inflammatory regulator acting through MAPK pathway in human lupus. J Autoimmun. 2016;75:96–104. https://doi. org/10.1016/j.jaut.2016.07.012. 77. West JA, Davis CP, Sunwoo H, Simon MD, Sadreyev RI, Wang PI, et al. The long noncoding RNAs NEAT1 and MALAT1 bind active chromatin sites. Mol Cell. 2014;55:791–802. https://doi.org/10.1016/j.molcel.2014.07.012. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular Neurodegeneration Springer Journals

Protective paraspeckle hyper-assembly downstream of TDP-43 loss of function in amyotrophic lateral sclerosis

Free
17 pages
Loading next page...
 
/lp/springer_journal/protective-paraspeckle-hyper-assembly-downstream-of-tdp-43-loss-of-7l6ZZqdHfW
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Molecular Medicine
eISSN
1750-1326
D.O.I.
10.1186/s13024-018-0263-7
Publisher site
See Article on Publisher Site

Abstract

Background: Paraspeckles are subnuclear bodies assembled on a long non-coding RNA (lncRNA) NEAT1. Their enhanced formation in spinal neurons of sporadic amyotrophic lateral sclerosis (ALS) patients has been reported but underlying mechanisms are unknown. The majority of ALS cases are characterized by TDP-43 proteinopathy. In current study we aimed to establish whether and how TDP-43 pathology may augment paraspeckle assembly. Methods: Paraspeckle formation in human samples was analysed by RNA-FISH and laser capture microdissection followed by qRT-PCR. Mechanistic studies were performed in stable cell lines, mouse primary neurons and human embryonic stem cell-derived neurons. Loss and gain of function for TDP-43 and other microRNA pathway factors were modelled by siRNA-mediated knockdown and protein overexpression. Results: We show that de novo paraspeckle assembly in spinal neurons and glial cells is a hallmark of both sporadic and familial ALS with TDP-43 pathology. Mechanistically, loss of TDP-43 but not its cytoplasmic accumulation or aggregation augments paraspeckle assembly in cultured cells. TDP-43 is a component of the microRNA machinery, and recently, paraspeckles have been shown to regulate pri-miRNA processing. Consistently, downregulation of core protein components of the miRNA pathway also promotes paraspeckle assembly. In addition, depletion of these proteins or TDP-43 results in accumulation of endogenous dsRNA and activation of type I interferon response which also stimulates paraspeckle formation. We demonstrate that human or mouse neurons in vitro lack paraspeckles, but a synthetic dsRNA is able to trigger their de novo formation. Finally, paraspeckles are protective in cells with compromised microRNA/dsRNA metabolism, and their assembly can be promoted by a small-molecule microRNA enhancer. Conclusions: Our study establishes possible mechanisms behind paraspeckle hyper-assembly in ALS and suggests their utility as therapeutic targets in ALS and other diseases with abnormal metabolism of microRNA and dsRNA. Keywords: ALS, TDP-43, Paraspeckle, NEAT1 Background currently mutations in more than 20 genes are known to Amyotrophic lateral sclerosis (ALS), the most common cause fALS [1]. The complexity of the disease hinders form of motor neuron disease, is a severe adult-onset development of ALS therapeutics, and those two drugs neuromuscular disease affecting motor neurons in the that have been approved for the treatment of ALS so far, spinal cord, brainstem and motor cortex. Up to 90% of riluzole and edaravone, have very limited efficacy. ALS cases are sporadic (sALS), the rest 10% bear a A multifunctional RNA-binding protein TDP-43 strong genetic component (familial ALS, fALS), and encoded by TARDBP gene is believed to be the main culprit in ALS: TDP-43 pathology is typical for ~ 95% of sALS cases and for fALS cases caused by C9ORF72 gene * Correspondence: shelkovnikovat@cardiff.ac.uk mutation [2]; in addition, dozens of mutations in School of Biosciences, Cardiff University, Museum Avenue, Cardiff CF10 3AX, UK TARDBP have been identified in fALS and sALS patients Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 2 of 17 [3, 4]. Hallmarks of all these ALS cases include protein function of the miRNA machinery and those with ac- clearance from the nucleus, its cytoplasmic accumula- tivated dsRNA response. Finally, we show that enoxa- tion and aggregation [5, 6]. Therefore, both loss and gain cin, an enhancer of the miRNA pathway, promotes of TDP-43 function are implicated in ALS however the paraspeckle formation. relative contribution of these two mechanisms is still debated. The paraspeckle is a prototypical nuclear body localized Methods on the border of splicing speckles [7]. A long non-coding Stable cell line maintenance, transfection and treatments RNA (lncRNA) NEAT1 serves as a scaffold for para- SH-SY5Y neuroblastoma cells and MCF7 cells were speckles, spatially organizing a variety of proteins by direct maintained in 1:1 mixture of Dulbecco’s Modified Ea- binding or piggy-back mechanism [8–11]. The NEAT1 gle’s Medium and F12 medium supplemented with 10% locus produces two transcripts, NEAT1_1 and NEAT1_2. fetal bovine serum (FBS), penicillin-streptomycin and The longer NEAT1 isoform, NEAT1_2, is essential for glutamine (all Gibco, Invitrogen). For differentiation paraspeckle assembly [10, 12]. Functions of paraspeckles into neuron-like cells, SH-SY5Y cells were grown on described so far include nuclear retention of specific poly-L-lysine (Sigma) coated coverslips in advanced RNAs, including inverted Alu repeat-containing tran- DMEM/F12 (ADF)/Neurobasal A mixture supple- scripts; regulation of gene expression by sequestration of mented with 10 μM all-trans retinoic acid (Sigma), B27 transcription factors; and modulation of miRNA biogen- (Life Technologies) and BDNF (Miltenyi, 10 ng/ml) for esis [13–16]. 6 days. The following gene-specific siRNAs were used: There is an established association of paraspeckles and ADAR1; Dicer; Drosha; FUS; Ago2; IFNB1 (all Life their components with a variety of pathological states Technologies, Silencer®); TARDBP (Silencer Select®, and conditions, from cancer to neurodegeneration. Para- s23829 and EHU109221, Mission® esiRNA, Sigma); speckles protect cancer cells against DNA damage and NEAT1 (Silencer Select®, n272456). Scrambled negative replication stress, regulate hormone receptor signaling control was AllStars from Qiagen. Plasmids for expres- and hypoxia-associated pathways thereby increasing sion of TDP-43 dNLS and TDP-43 C-termical fragment their survival [17–19]. Paraspeckles become enlarged in are described elsewhere [28]. Cells were transfected cells primed by viral or synthetic double-stranded (ds) with siRNA (400 ng/well), plasmid DNA (200 ng/well) RNAs and play an important role in antiviral response or poly(I:C) (Sigma, 250 ng/well) using Lipofecta- [14]. An unusually tight association of paraspeckle com- mine2000 (Life Technologies) in 24-well plates. TDP-43 ponents with neurodegenerative conditions, and ALS in specific shRNA plasmid was from Sigma (MISSION® particular, has recently emerged. Firstly, enhanced para- SHCLNG-NM_007375). To delete the NLS of endogen- speckle formation has been reported in spinal motor ous TDP-43, Feng Zhang lab’sTarget Finder (http:// neurons of sALS patients [20]. This finding was surpris- crispr.mit.edu/) was used to identify guide RNA target ing because levels of the longer NEAT1 isoform, sequences flanking the genomic region of TARDBP NEAT1_2, essential for paraspeckle formation, are very gene encoding NLS. Respective forward and reverse oli- low in the adult nervous system [21]. Secondly, at least gonucleotides for two pairs of guides were annealed seven paraspeckle proteins, including TDP-43 and FUS, and cloned into pX330-U6-Chimeric_BB-CBh-hSpCas9 are genetically linked to ALS and a related condition, (pX330) vector provided by Feng Zhang (Addgene de- frontotemporal lobar degeneration (FTLD) [22–25]. posited plasmid) as described [29]. MCF7 cells were FUS, a protein structurally and functionally similar to transfected with plasmids encoding upstream and down- TDP-43, is required to build paraspeckles [8, 23]. stream guide RNAs (500 ng/well) using Lipofecta- TDP-43 association with paraspeckles has also been re- mine2000 and analysed after 72 h. Guide RNA sequences: ported [8]. TDP-43 directly binds NEAT1, and this inter- T1: 5’-TTATTTAGATAACAAAAGAAAAA-3′,T2: 5’-A action is increased in the brain of FTLD patients [26, ACATCCGATTTAATAGTGT-3′,T3: 5’-GGAATTCTG 27]. Overall, currently available data support the role of CATGCCCCAGATGC-3′,T4: 5’-ACATCCGATTTAAT paraspeckles in molecular pathology of ALS, however AGTGTT-3′. Cellulartreatments wereas follows: 1×10 the underlying mechanisms of their enhanced formation IU interferon beta-1a (IFNbeta), 0.5 μg/ml LPS, 100 μg/ml in spinal neurons are not understood. zymosan, 50 μM suramin, 500 nM TSA, 2 mM sodium In current study we show that loss of TDP-43 is butyrate, 10 and 50 μMenoxacin, 10 μM riluzole, 10 μM sufficient to stimulate paraspeckle formation – a edaravone (all Sigma). Human ES cell derived neurons phenomenon likely linked to the function of TDP-43 were transfected with 15 μg/well of poly(I:C) using in microRNA (miRNA) processing and as an RNA FuGENE®HD (Promega). Enoxacin, edaravone and riluzole chaperone. Furthermore, we provide evidence that toxicity was assessed using resazurin-based CellTiter-Blue paraspeckles are protective in cells with impaired Cell Viability Assay (Promega). Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 3 of 17 Primary culture of mouse neurons and Superscript IV (Invitrogen). For analysis of miRNA Primary cultures of mouse hippocampal neurons were levels, RNA was extracted with QIAzol (Qiagen) followed prepared from P0 CD1 mice as described [28] and main- by reverse transcription with Qiagen miScript II RT Kit. tained for 5–14 days. Real-time qPCR was conducted using SYBR green master mix as described [28]. For miRNA quantification, forward Differentiation of human ES cells into motor neuron miRNA-specific primers were used in combination with enriched cultures the universal reverse primer (unimiR). All primer se- Cultures of human neural precursor cells (NPCs) and quences are given in Table 1. motor neurons differentiated from H9 hES cell line were prepared as described previously [30]. Briefly, hES cells RNA immunoprecipitation (RIP) and PCR analysis were maintained in mTESR2 media (Stemcell Technolo- MCF7 cells were transfected with equal amounts of plas- gies) on Matrigel® (Corning) coated dishes. Confluent hES mids to express GFP-tagged FUS or NONO together H9 cultures were switched to differentiation medium with TARDBP siRNA or scrambled control siRNA. After composed of ADF supplemented with SB431542 (10 μM, 48 h, cells were scraped in RIP buffer prepared using Abcam). Purmorphamine (1 μM, Cayman Chemicals) and RNase-free water (1xPBS with 1% Triton-X100 and pro- retinoic acid (0.1 μM, Sigma) were added on Day 4. On tease inhibitors cocktail). Cells were left on ice for Day 8, cells were split in 1:2 ratio and on Day 16, NPCs 10 min with periodic vortexing, and the lysate was cen- were dissociated using Accutase®, plated onto Matrigel® trifuged at 13,000 rpm for 10 min. GFP-Trap® beads coated dishes and cultured in ADF with GlutaMAX, (Chromotek) were washed in RIP buffer 4 times and penicillin-streptomycin, B27 (12587–010) and N2 supple- added directly to cleared cell lysates with subsequent ro- ments (all Life Technologies) and BDNF (Miltenyi, 10 ng/ tation at + 4 °C for 3 h. Beads were washed 4 times in ml). On Day 23, Accutase® was used to re-plate neurons RIP buffer and RNA was eluted by resuspension in on dishes/coverslips at desired density. Neurons were cul- TRI-reagent (Sigma). RNA was purified according to tured in 50:50 mixture of ADF/Neurobasal A with the manufacturer’s protocol, and equal amounts of RNA above supplements until Day 40. were used for cDNA synthesis as described above. Immunocytochemistry and RNA-FISH on cultured cells Protein analysis Cells were fixed on coverslips with 4% paraformaldehyde Total cell lysates were prepared for Western blot by lys- on ice for 15 min and permeabilized in cold methanol ing cells in wells in 2× Laemmli (loading) buffer (or 70% ethanol in case of RNA-FISH). Coverslips were followed by denaturation at 100 °C for 5 min. Proteins incubated with primary antibodies diluted in blocking were resolved by SDS-PAGE and transferred to PVDF solution (5% goat serum in 0.1% Triton X-100/PBS) for membrane (Amersham) by semi-dry transfer. The mem- 1 h at RT or at 4 °C overnight. Secondary Alexa488- or brane was blocked in 4% non-fat milk in TBST and incu- Alexa546-conjugated antibody was added for 1 h at RT. bated in primary antibodies prepared in milk or 5% BSA For RNA-FISH, commercially available NEAT1 and overnight. Secondary HRP-conjugated antibodies were MALAT1 probes (Stellaris® FISH Probes against human from Amersham. For detection of proteins, Wester- NEAT1, middle segment or 5′ segment, or human nBright Sirius ECL reagent (Advansta) was used. β-actin MALAT1, all Biosearch Technologies) were used as per was used for normalisation. standard protocol. Fluorescent images were taken using BX61 microscope equipped with F-View II camera and Primary antibodies processed using CellF software (all Olympus). Paraspeckle The following commercial primary antibodies were used: quantification (number of individual paraspeckles per TDP-43 (rabbit polyclonal, 10782–2-AP, Proteintech and DAPI-visualised nucleus) was performed manually, by the mouse monoclonal, MAB7778-SP, R&D Biosystems); same person for all conditions, blinded to the experimen- FUS (rabbit polyclonal, Proteintech, 11570–1-AP); tal condition. Clusters of paraspeckles were counted as a p54nrb/NONO (rabbit polyclonal C-terminal, Sigma); single paraspeckle. For quantification of cleaved caspase 3 PSF/SFPQ (rabbit monoclonal, ab177149, Abcam); Tuj positive cells, ‘Analyze particles’ tool of Image J software (β-Tubulin III, mouse monoclonal, Sigma); dsRNA was used (8–10 fields were analysed per condition). (mouse monoclonal, J2, Kerafast); cleaved caspase 3 (rabbit polyclonal, 9661, Cell Signaling); NF-κB p65 RNA analysis (rabbit monoclonal, D14E12, Cell Signaling); IFIT3 Total cellular RNA was extracted using GenElute total (rabbit polyclonal, Bethyl); p-eIF2α (rabbit monoclonal, RNA kit (Sigma) and possible DNA contamination was re- ab32157, Abcam); p-PKR (rabbit polyclonal, Thr451, moved using RNase free DNase kit (Qiagen). First-strand ThermoFisher); PKR (mouse monoclonal, MAB1980-SP, cDNA synthesis were performed using random primers R&D Systems); eIF2α (rabbit monoclonal, D7D3, Cell Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 4 of 17 Table 1 Primers used in the study Target Forward Reverse GAPDH 5’-TCGCCAGCCGAGCCA-3′ 5’-GAGTTAAAAGCAGCCCTGGTG − 3′ NEAT1 total 5’-CTCACAGGCAGGGGAAATGT-3′ 5’-AACACCCACACCCCAAACAA-3′ NEAT1_2 5’-AGAGGCTCAGAGAGGACTGTAACCTG-3′ 5′-TGTGTGTGTAAAAGAGAGAAGTTGTGG-3’ TDP-43 5’-TCAGGGCCTTTGCCTTTGTT-3’ 5’-TGCTTAGGTTCGGCATTGGAT-3’ IL8 5’-ACACTGCGCCAACACAGAAA-3′ 5’-CCTCTGCACCCAGTTTTCCT-3’ ADARB2 ATATTCGTGCGGTTAAAAGAAGGTG ATCTCGTAGGGAGAGTGGAGTCTTG Alu RNA 5’-GAGGCTGAGGCAGGAGAATCG-3’ 5’-GTCGCCCAGGCTGGAGTG-3’ DICER 5’-TTAACCTTTTGGTGTTTGATGAGTGT-3’ 5’-GCGAGGACATGATGGACAATT-3’ DROSHA 5’-CGGCCCGAGAGCCTTTTAT-3’ 5’-TGCACACGTCTAACTCTTCCA-3’ ADAR1 5’-TTGTCAACCACCCCAAGGT-3’ 5’-CCATCAGCCAGACACCAGTT-3’ AGO2 5’-CACCATGTACTCGGGAGCC-3’ 5’-TCCCAAAGTCGGGTCTAGGT-3’ FUS 5’-GCGGGGCTGCTCAGT-3’ 5’-TTGGGTTGCTTGTTGGGTAT-3’ CHOP 5’-TTAAAGATGAGCGGGTGGC-3′ 5’-GCTTTCAGGTGTGGTGATGTA-3’ CXCL10 5’-TGCCATTCTGATTTGCTGCC-3’ 5’-ATGCTGATGCAGGTACAGCG-3’ IFNB1 5’-ACGCCGCATTGACCATCTAT-3’ 5’-AGCCAGGAGGTTCTCAACAA-3’ IFNA1 5’-TCTGCTATGACCATGACACGAT-3’ 5’-CAGCATGGTCCTCTGTAAGGG-3’ IFNA2 5’-AGGAGGAAGGAATAACATCTGGTC-3’ 5’-GCAGGGGTGAGAGTCTTTGAA-3’ MALAT1 5’-GGATCCTAGACCAGCATGCC-3’ 5′- AAAGGTTACCATAAGTAAGTTCCAGAAAA-3’ IFIH1 5’-GCATGGAGGAGGAACTGTTGA-3’ 5’-GCATGGAGGAGGAACTGTTGA-3’ CYCS 5’-TCGTTGTGCCAGCGACTAAA-3’ 5’-GCTTGCCTCCCTTTTCAACG-3’ STAT1 5’-CTGTGCGTAGCTGCTCCTTT-3’ 5’-GGTGAACCTGCTCCAGGAAT-3’ MYD88 5’-TGACCCCCTGGGGCAT-3’ 5’-AGTTGCCGGATCATCTCCTG-3’ Pri-miR-17–92 5’-CAGTAAAGGTAAGGAGAGCTC 5’-CATACAACCACTAAGCTAAAGAAT AATCTG-3’ AATCTGA-3’ Pri-miR-15a 5’-CCTTGGAGTAAAGTAGCAGCAC-3’ 5’-CCTTGTATTTTTGAGGCAGCAC-3’ miR-18a 5’-CATCATCGGTAAGGTGCATC-3’ 5’-GAATCGAGCACCAGTTACGC-3′ (unimiR) miR-92a 5’-GAGTCTATTGCACTTGTCCC-3’ unimiR miR-106a 5’-AAAAGTGCTTACAGTGCAGGTAG-3’ unimiR Signaling); β-actin (mouse monoclonal, A5441, Sigma). (Stellaris® FISH Probes against human NEAT1 5′ seg- Antibodies were used at 1:500–1:1000 dilution for all ment, Biosearch Technologies) diluted in hybridisation applications. buffer (10% formamide/2xSSC, 5 μl probe in 200 μl buf- fer per slide under a 24 × 60 mm coverslip) in a humidi- Analysis of human tissue samples fied chamber at 37 °C overnight. Nuclei were co-stained Human spinal cord paraffin sections from a panel of with DAPI. Paraspeckles were analysed using BX61 clinically and histopathologically characterised ALS cases microscope/F-View II camera (Olympus) at 100× magni- and neurologically healthy individuals were obtained fication. For RNAscope® ISH analysis, Hs-NEAT1-long from the Sheffield Brain Tissue Bank and MRC London (411541) probe (Advanced Cell Diagnostics) was used Neurodegenerative Diseases Brain Bank (Institute of according to manufacturer’s instructions. For qRT-PCR Psychiatry, King’s College London). Consent was ob- analysis, total RNA was extracted from thick frozen tained from all subjects for autopsy, histopathological as- spinal cord sections and cDNA prepared using sessment and research were performed in accordance Ready-To-Go You-Prime First-Strand Beads (GE Health- with local and national Ethics Committee approved do- care). For laser capture microdissection (LCM), frozen nation. Human spinal cord sections were 7 μm thick. spinal cord sections (total of 5 sections per patient/case) For conventional RNA-FISH, slides were boiled in citrate were cut into 5–10 μm thin sections using a cryostat, buffer for 10 min, washed in 2xSSC prepared with mounted on glass slides and fixed in cold acetone for DEPC-treated water and incubated with NEAT1 probe 3 min. Sections were stained using toluidine blue, Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 5 of 17 dehydrated in ascending alcohol series for 30 s and Thus, de novo paraspeckle formation is typical for spinal placed in xylene for 1 min. The PixCell® II Microdissec- motor neurons and glial cells of individuals affected by tion system (Applied Biosystems) was used for LCM. ALS with primary or secondary TDP-43 pathology. Motor neurons from the anterior grey horn were laser captured (300–500 or 30–80 per patient for healthy con- Loss of TDP-43 but not its cytoplasmic accumulation or trols and ALS cases respectively), with the Macro-LCM aggregation results in paraspeckle hyper-assembly cap films peeled and placed in test tubes with 50 μl ex- We next sought to determine possible mechanisms traction buffer (Pico Pure® RNA Isolation Kit; Thermo underlying paraspeckle hyper-assembly in ALS. TDP-43 Fisher Scientific) on ice. The extracted films were incu- pathology in the spinal cord is very common in ALS, be- bated at 42 °C with the extraction buffer for 30 min and ing present in almost all sALS cases, fALS cases caused frozen at -80 °C until RNA extraction. Total RNA purifi- by mutations in TARDBP gene itself as well as those cation was performed using the above kit as per manu- caused by C9ORF72 gene repeat expansions [2–4, 31]. facturer’s instructions. RNA samples were analysed TDP-43 has been identified as a paraspeckle protein using the Agilent RNA 6000 Pico Kit (Agilent Technolo- [20], thus we tested the possibility that TDP-43 dysfunc- gies®) and used for qRT-PCR. tion affects paraspeckle assembly. Hallmarks of TDP-43 proteinopathy are clearance of the protein from the nucleus and its accumulation and Statististical analysis aggregation in the cytoplasm [5, 6]. We first modelled GraphPad Prism software was used for statistical ana- loss of TDP-43 function in two stable cell lines. By using lysis. Statistical test used in each case is indicated in the specific siRNA, ~ 90 and 50% TDP-43 knockdown was figure legend. N indicates the number of biological repli- achieved in MCF7 and neuroblastoma SH-SY5Y cells, re- cates. On all graphs, error bars represent SEM. spectively (Fig. 2a; Additional file 1: Fig. S1a). TDP-43 depletion led to a significant increase of the number of Results paraspeckles per nucleus (Fig. 2b and c; Additional file Presence of paraspeckles in the spinal cord neurons is a 1: Fig. S1b). Consistently, the paraspeckle-specific hallmark of sALS and fALS NEAT1 isoform, NEAT1_2, was upregulated in cells Augmented paraspeckle assembly has been previously transfected with TDP-43 siRNA (Fig. 2a; Additional file reported in sALS spinal cord neurons as compared to 1: Fig. S1a). Similar results were obtained using an inde- non-ALS controls [20]. We sought to verify this result in pendent TDP-43 siRNA pool and an shRNA targeting a separate cohort of sALS cases as well as to extend this TDP-43 (Additional file 1: Fig. S1c and d). In contrast, we analysis to fALS. In total, 7 sALS cases, 2 cases with did not observe changes in the levels or distribution of an- TARDBP mutations and 4 cases with C9ORF72 muta- other abundant lncRNA, MALAT1, a component of spli- tions alongside with 6 healthy controls were examined cing speckles (Fig. 2a;Additional file 1: Fig. S1e). Levels of by RNA-FISH with NEAT1 probe. No neurons with core paraspeckle proteins SFPQ, NONO and FUS were paraspeckles were detected in the spinal cord of healthy also unaffected by TDP-43 knockdown (Additional file 1: individuals (97 neurons analysed), however such neurons Fig. S1f). We next examined whether TDP-43 knockdown were present in up to 40% of neurons in all ALS cases would result in enhanced association of core paraspeckle examined (Fig. 1a and b). We also confirmed the pres- proteins with NEAT1_2. Plasmids to overexpress ence of paraspeckles in ALS motor neurons using RNA- GFP-tagged FUS or NONO proteins were co-transfected scope® ISH (Fig. 1c). Consistently, qRT-PCR analysis of with scrambled or TDP-43 siRNA followed by RNA im- spinal cord tissue from four healthy controls and four munoprecipitation with GFP-Trap beads. Indeed, by PCR, ALS patients demonstrated elevated NEAT1 levels in the both FUS and NONO demonstrated increased association latter group (Fig. 1d). We further performed laser cap- with NEAT1_2 in TDP-43 depleted cells (Fig. 2d). To ver- ture microdissection (LCM) of spinal neurons in the ify that paraspeckles formed in TDP-43 depleted cells are ventral horn and analysed NEAT1_2 levels by qRT-PCR functional, we measured the expression of established (n = 3 for controls and n = 6 for ALS patients, including paraspeckle-dependent genes, IL8 and ADARB2, known three sALS and three ALS-C9 cases). NEAT1_2 levels to be positively and negatively regulated by paraspeckles, were indeed significantly upregulated in LCM neurons respectively [13, 14]. Indeed, IL8 mRNA was upregulated of ALS patients (Fig. 1e). Finally, using RNAscope® ISH, and ADARB2 mRNA decreased upon TDP-43 knock- we also analysed the presence of paraspeckles in down (Fig. 2e). non-neuronal cells. Wide-spread paraspeckle assembly Since nuclear clearance of TDP-43 in ALS is coupled in glial cells in ALS spinal cord was observed (n = 6 for to its cytoplasmic accumulation and aggregation, we controls and n = 4 for ALS patients, including two sALS, next evaluated the effect of cytoplasmic TDP-43 on one ALS-TDP and one ALS-C9 case) (Fig. 1f). paraspeckles. TDP-43 lacking nuclear localization signal Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 6 of 17 Fig. 1 Paraspeckles are formed in the spinal cord of sALS and fALS patients but not healthy controls. a and b Examples of spinal motor neurons with paraspeckles (a) and their quantification (b) in ALS patients with different disease aetiology. Paraspeckles were visualised in the spinal cord sections of a cohort of fALS and sALS patients as well as neurologically normal control individuals using RNA-FISH with a fluorescent (Quasar 570) probe mapping to the 5′ portion of NEAT1. Images were also taken in the FITC channel to distinguish between specific NEAT1 signal and green autofluorescence from lipofuscin (a). The fraction of neurons with identifiable paraspeckles in the spinal anterior horn of aetiologically different ALS cases and control individuals was quantified and plotted separately for fALS with TARDBP mutations (ALS-TDP), fALS with C9ORF72 repeat expansion (ALS-C9) and sALS cases (b). The top figure within each bar corresponds to the number of cases analysed and the figures below - to the number of individual neurons negative or positive for the presence of paraspeckles. Scale bars, 10 μm. c Examples of paraspeckle-containing neurons in the ALS spinal cord visualised with RNAscope® NEAT1_2 specific probe. In the bottom panel, a paraspeckle-positive (right) and a paraspeckle-negative (left) neurons, found adjacent to each other, are shown. Nuclei are circled. Scale bar, 10 μm. d NEAT1 levels in the total RNA samples extracted from transversely cut spinal cord blocks of ALS patients and healthy controls analysed by qRT-PCR (n = 4 for control and ALS patients, including two sALS and two ALS-C9 cases, Mann-Whitney U-test). e NEAT1_2 levels in neurons microdissected from the spinal anterior horn of ALS patients and healthy controls analysed by qRT-PCR (n = 3 for control and n = 6 for ALS cases, including three sALS and three ALS-C9 cases, Mann-Whitney U-test). f Paraspeckles in glial cells in the ALS spinal cord visualised with RNAscope® ISH using NEAT1_2 specific probe. Representative images of the spinal cord for a control individual and an ALS patient are shown. Scale bar, 20 μm Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 7 of 17 Fig. 2 TDP-43 depletion but not its cytoplasmic accumulation or aggregation stimulates paraspeckle assembly in stable cell lines. a TDP-43 siRNA- mediated knockdown upregulates NEAT1_2. MCF7 cells were transfected with scrambled or TDP-43 siRNA and analysed 48 h post-transfection by qRT- PCR (n = 6). **p < 0.01 (Mann-Whitney U-test). b and c TDP-43 depletion auguments paraspeckle assembly in MCF7 cells. Quantification (b)and representative images (c) are shown. RNA-FISH with NEAT1_2 probe (c, top panels) or anti-NONO staining (c, bottom panels) were used to visualise paraspeckles; arrowheads indicate clusters of paraspeckes. The number of cells analysed is indicated in the bottom of each bar (b)(***p <0.0001, Student’s t-test). d TDP-43 depletion enhances interaction of NEAT1_2 with core paraspeckle proteins NONO and FUS. GFP-tagged NONO or FUS was co-transfected into MCF7 cells together with scrambled siRNA or TDP-43 siRNA. NEAT1_2 and total NEAT1 were detected in GFP pull-down samples by RT-PCR. Arrowhead indicates the specific band for NEAT1_2 primer pair. e Expression of paraspeckle-regulated genes in MCF7 cells depleted of TDP-43 as measured by qRT-PCR (n = 6 or 8). *p < 0.05, ***p < 0.001 (Mann-Whitney U-test). f and g Expression of TDP-43 lacking nuclear localisation signal (TDP-43 dNLS) or TDP-43 C-terminal 25 kDa fragment (TDP-43 CT) does not affect paraspeckles or NEAT1 levels. Representative images of paraspeckles in transfected SH-SY5Y cells (f) and their quantiation (g, n =56 and n = 41 for GFP- and TDP-43 dNLS-expressing cells respectively) are shown. Scale bars are 10 μmin all panels (TDP-43 dNLS) and a C-terminal TDP-43 fragment corre- total cellular levels by CRISPR/Cas9 editing of the en- sponding to ~ 25 kDa TDP-43 cleavage product (TDP-43 dogenous TARDBP gene. Cells were transiently trans- CT, aa. 191–414), both characterized by predominantly fected with plasmids for expression of two cytoplasmic distribution, were transiently expressed in independent guide RNA pairs targeting upstream and neuroblastoma cells (Fig. 2f). However, neither paraspeckle downstream sequences encoding NLS of TDP-43 numbers nor NEAT1 levels were affected by these cytoplas- (Additional file 1: Fig. S1 g). For both guide RNA mic proteins (Fig. 2f and g). TDP-43 dNLS forms cytoplas- pairs tested, 15–20% transfected cells displayed partial mic aggregates in a fraction of cells, but their presence also TDP-43 redistribution to the cytoplasm which ne- did not affect paraspeckles (Fig. 2g, bottom panel). Finally, vertheless did not enhance paraspeckle assembly in order to recapitulate simultaneous nuclear depletion and (Additional file 1: Fig. S1 h), suggesting that substan- cytoplasmic accumulation of TDP-43, we targeted en- tial loss of nuclear TDP-43 is required to produce an dogenous TDP-43 out of the nucleus while preserving its effect on paraspeckles. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 8 of 17 Overall, a decrease in cellular TDP-43 levels results in (Fig. 3a). Consistent with the major role of Drosha in paraspeckle hyper-assembly. pri-miRNA processing, its knockdown led to the build-up of pri-miRNAs; both Drosha and TDP-43 knockdown also Compromising miRNA pathway results in enhanced resulted in significantly diminished levels of select mature paraspeckle assembly miRNAs (Additional file 2:Fig. S2). TDP-43 is known to contribute to miRNA biogenesis at RNA-FISH and paraspeckle quantification showed that two different levels, enhancing the activity of the Micropro- downregulation each of the above proteins is accompanied cessor in the nucleus and of the Dicer complex in the cyto- by enhanced paraspeckle assembly (Fig. 3b and c)and plasm [32, 33]. Recently, paraspeckles have been shown to upregulation of total NEAT1 and NEAT1_2 (Fig. 3d). contribute to pri-miRNA processing by spatially organizing Another ALS-linked protein, FUS, is structurally and the Microprocessor and enhancing its processivity [16]. We functionally similar to TDP-43 and also plays a role hypothesised that augmented paraspeckle assembly in cells in miRNA biogenesis [35], but it is a core paraspeckle depleted of TDP-43 might be a compensatory mechanism protein required for paraspeckle integrity [8, 23]. As to counterbalance the effect of TDP-43 loss of function on expected from its essential paraspeckle function, FUS miRNA processing. If this indeed is true, paraspeckle as- depletion resulted in decreased paraspeckle numbers sembly should be also increased in cells with compromised (Fig. 3b and c). Finally, a small molecule inhibitor of function of the miRNA pathway. To test this hypothesis, RISC loading, suramin [36], wasalsoabletoincrease we knocked down three core enzymes of the miRNA path- NEAT1_2 levels and promote paraspeckle assembly way, a Microprocessor component Drosha, the pre-miRNA (Fig. 3e). processing ribonuclease Dicer and the RISC endonuclease Thus, interfering with the function of the miRNA Ago2, in neuroblastoma cells. In our analysis we also in- pathway causes NEAT1 upregulation and enhanced cluded ADAR1 protein recently reported to promote paraspeckle formation in cultured cells. This may repre- miRNA processing [34]. Using specific siRNAs, we sent one of the mechanisms behind the effect of TDP-43 achieved at least 40% knockdown for each of these genes loss of function on paraspeckles. Fig. 3 Enhanced paraspeckle assembly in cells with compromised function of the miRNA pathway. a Downregulation of Drosha, Dicer, Ago2, ADAR1 and FUS in neuroblastoma cells after transfection of specific siRNA as analysed by qRT-PCR (n = 4–6). *p < 0.05, **p < 0.01 (Mann-Whitney U-test). b and c Knockdown of Drosha, Dicer, Ago2 or ADAR1 results in increased paraspeckle formation. Representative images of cells (b)and paraspeckle quantification (c) are shown. In c, the mean number of paraspeckles per cell and frequencies of such cells were plotted; the number of cells analysed is indicated at the bottom of each bar. *p < 0.05, **p < 0.01; ****p < 0.0001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). d NEAT1 is upregulated in cells after knockdown of Drosha, Dicer, Ago2 and ADAR1 (n = 4–6). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). e Suramin stimulates NEAT1_2 expression and paraspeckle formation. Cells were treated with suramin for 24 h before collection for RNA-FISH and qRT-PCR analysis (n = 4). *p < 0.05 (Mann-Whitney U-test). In a-d, cells were analysed 48 post- transfection. Scale bars are 10 μm in all panels Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 9 of 17 Type I IFN signaling is activated in cells depleted of TDP- Taken together, these data suggest that endogenous 43 or other miRNA factors and stimulates paraspeckle dsRNA accumulation and associated type I IFN response formation represent one of the mechanisms behind augmented para- TDP-43 is known to bind and regulate long transcripts speckle assembly caused by TDP-43 loss of function. [37], and its loss correlates with accumulation of tran- scripts prone to form double stranded (ds) RNA [38]. Paraspeckles are protective in cells with compromised Conspicuously, regulation of cellular response to viral miRNA biogenesis and activated dsRNA response dsRNA is one of the best characterized functions of We next examined whether paraspeckles confer protec- paraspeckles [14, 39]. Therefore, we considered abnor- tion to cells depleted of TDP-43 or stimulated with ex- mal accumulation of endogenous dsRNA in TDP-43 de- ogenous dsRNA. Neuroblastoma cells were transfected pleted cells as another mechanism underlying the effect with siRNAs targeting TDP-43 or another Microprocessor of TDP-43 loss of function on paraspeckles. Critically, component Drosha alone or in combination with NEAT1 miRNA pathway itself is the biggest cellular source of siRNA. Cytotoxicity was assessed after 36 h using cleaved dsRNA, and its factors Dicer, Drosha and ADAR1 are caspase 3 (CC3) and DNA damage-inducible transcript 3 known to limit the accumulation of transcripts with ex- also known as GADD153 or CHOP, a pro-apoptotic tran- tensive secondary structure [40–42]. scription factor [46, 47] as markers of apoptotic cell death. We used J2 antibody, a gold standard for dsRNA detec- NEAT1 siRNA-mediated knockdown was equally efficient tion [43], to study the presence of dsRNA species in cells alone and in cells co-transfected with TDP-43 or Drosha depleted of TDP-43, Drosha, Dicer or ADAR1. An in- siRNA, allowing 60% NEAT1_2 downregulation and loss crease in J2-positive signal was obvious after knockdown of paraspeckles in the majority of cells (Fig. 5a and data of each of these genes (Fig. 4a and b, Additional file 3:Fig. not shown). We found that while knockdown of TDP-43 S3a). In contrast, Ago2 or FUS knockdown did not cause or Drosha alone did not result in significant cell death, dsRNA accumulation (Fig. 4b). J2 antibody was reported simultaneous disruption of paraspeckles increased the to recognise Alu repeats especially well [44]. We next used rates of apoptosis for both genes studied (Fig. 5b and c). primers which specifically detect Alu-containing RNAs Although regulation of dsRNA response by paraspeckles [45]. Dicer and Drosha but not TDP-43 or ADAR1 knock- is well documented [14], their ability to modulate down resulted in increased levels of Alu-containing RNAs dsRNA-induced apoptosis has not been addressed. We as measured by qRT-PCR (Fig. 4c) indicating that a differ- transfected cells with scrambled siRNA or NEAT1 siRNA ent repertoire of dsRNA species accumulate after knock- and subsequently exposed them to a synthetic dsRNA down of each gene. analogue poly(I:C). Cultures depleted of paraspeckles had The build-up of dsRNA is known to trigger phosphor- increased CHOP mRNA levels and higher numbers of ylation of PKR and eIF2α and activation of type I inter- CC3-positive cells after 8 h and 24 h of poly(I:C) stimula- feron (IFN) signaling. In cells transfected with TDP-43 tion, respectively (Fig. 5d-f). siRNA, levels of phosphorylated PKR and eIF2α were el- How does paraspeckle deficiency promote apoptosis in evated (Fig. 4d; Additional file 3: Fig. S3b), and the ex- cells with compromised miRNA biogenesis and activated pression of IFNB1 and an IFN-stimulated gene (ISG) dsRNA response? A whole class of cytotoxicity-associated CXCL10 was increased (Fig. 4e). Since dsRNA response IFN-stimulated genes (ISGs) were reported to be negatively eventually converges on type I IFNs, we asked whether regulated by miRNAs [48]. We found that potentially these cytokines can contribute to paraspeckle response pro-apoptotic ISGs from this class, STAT1 and CYCS,were in TDP-43 depleted cells. Firstly, we showed that IFN- significantly upregulated in paraspeckle-deficient cells de- beta is the main type I IFN induced by dsRNA in neuro- pleted of TDP-43 and Drosha (Fig. 5g). Furthermore, three blastoma cells (Additional file 4: Fig. S4a). IFNbeta such ISGs, STAT1, MYD88 and IFIH1, were consistently simulation per se was sufficient to stimulate para- upregulated in NEAT1-depleted cells in the course of speckle assembly, although the effect was transient (Fig. poly(I:C) stimulation (Fig. 5h). 4f). In line with this, ligands of TLR3 (poly(I:C)) and Thus, paraspeckles are protective against apoptotic TLR4 (bacterial lipopolysaccharide, LPS) which stimu- death in cells with compromised miRNA machinery late IFNbeta expression, but not a TLR2 ligand zymo- and/or activated dsRNA response. san which does not affect IFNbeta production, were able to boost NEAT1 expression and paraspeckle assem- Cultured human neurons lack paraspeckles but their de bly (Additional file 4: Fig. S4b-d). Finally, co-transfection novo assembly can be triggered by dsRNA of IFNbeta siRNA was sufficient to reduce paraspeckle Normal postmitotic neurons in the brain or spinal cord abundance in cells transfected with TDP-43 siRNA al- of adult mice express very low levels of NEAT1_2 iso- though did not abrogate the hyper-assembly completely form and do not form paraspeckles in vivo [21]. How- (Fig. 4g). ever it was not clear whether neurons cultured in vitro Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 10 of 17 Fig. 4 Endogenous dsRNA response and type I interferon promote paraspeckle hyper-assembly in stable cell lines. a and b Depletion of TDP-43, Dicer, Drosha, ADAR1 but not Ago2 or FUS causes intracellular build-up of dsRNA. dsRNA was detected by immunocytochemistry using J2 antibody. Representative images of all conditions are shown. Scale bars, 50 and 10 μm for general plane and close-up panels respectively. c Levels of Alu-containing RNA as analysed by qRT-PCR using specific primers recognising Alu elements (n = 4). *p < 0.05 (Mann-Whitney U-test). d and e Markers of activated cellular reponse to dsRNA are upregulated in TDP-43 depleted cells. Levels of phosphorylated PKR and eIF2α were analysed by Western blot (d, representative blots are shown) and expression of IFNB1 and an IFN-stimulated gene CXCL10 -by qRT-PCR (e, n =6). *p <0.05 (Mann-Whitney U-test). f IFNbeta treatment stimulates NEAT1 expression and paraspeckle formation. NEAT1 levels were measured by qRT-PCR (n = 6). **p < 0.01 (Mann-Whitney U-test). Staining for an IFN-inducible protein IFIT3 was used as a positive control. Scale bar, 10 μm. g Simultaneous IFNbeta knockdown partially reverses the effect of TDP-43 depletion on paraspeckles. * and #p < 0.05, ***p < 0.001 (one-way ANOVA with Holm-Sidak correction for multiple comparisons). Scale bar, 10 μm. In all panels, cells were harvested for analysis 48 h post-transfection. Paraspeckles in panels f and g were visualised by NEAT1_2 RNA-FISH would acquire and preserve paraspeckles. First, we studied and Day 16 NPCs (Fig. 6b). However, paraspeckles were primary hippocampal cultures from newborn mice. While absent even in immature (Day 23) neurons and onwards glial cells have readily detectable paraspeckles already at (Fig. 6b). In contrast, retinoic acid/BDNF-induced dif- 5 days in vitro, neurons had no sign of paraspeckles even ferentiation of SH-SY5Y neuroblastoma cells for after 14 days in vitro (Fig. 6a). Next we examined para- 6 days did not affect their ability to form paraspeckles speckle assembly during differentiation of human ES cells (Additional file 5:Fig.S5),suggesting fundamental differ- into motor neurons. Human ES cells lack paraspeckles ences in the biology of neurons and neuroblastoma-derived but they appear during differentiation (days 4–5into dif- neuron-like cells. Thus, disappearance of paraspeckles ferentiation, trophoblast stage) [49]. Consistent with this marks the transition between human NPCs and neurons, study, we started observing paraspeckles in Day 4 neural whereas de novo paraspeckle assembly is not triggered by precursor cells (NPCs), and they were prominent in Day 8 in vitro conditions in primary neurons. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 11 of 17 Fig. 5 Loss of paraspeckles promotes apoptosis in cells with disturbed miRNA biogenesis and activated dsRNA response. a-c Disruption of paraspeckles in cells with downregulated TDP-43 or Drosha promotes apoptotic death in neuroblastoma cells. Efficiency of NEAT1_2 knockdown and levels of a proapototic protein CHOP mRNA were analysed by qRT-PCR (n = 3). * and #p ≤ 0.05 (Mann-Whitney U-test) (a and b). In c, representative images and quantification of cleaved caspase 3 (CC3) positive cells are shown. *p < 0.05, ** and ##p < 0.01 (one-way ANOVA). Scrambled siRNA or NEAT1 siRNA was co-transfected with an siRNA targeting TDP-43 or Drosha, and cells analysed 36 h post-transfection. * and # indicate statistically significant difference as compared to cells transfected with only scrambled siRNA or only NEAT1 siRNA, respectively. Scale bar, 100 μm. d-f Disruption of paraspeckles promotes apoptosis in dsRNA-stimulated cells. Cells were transfected with scrambled siRNA or NEAT1 siRNA and stimulated with poly(I:C) 36 h post-transfection. Induction of CHOP by poly(I:C) over time in normal cells (d) and CHOP mRNA levels in paraspeckle-deficient and paraspeckle-sufficient cells after 8 h of poly(I:C) stimulation (e) were analysed by qRT-PCR (n = 4). **p < 0.01 (Mann- Whitney U-test). In f, representative images and quantitation of CC3-positive cells in cultures transfected with scrambled siRNA or NEAT1 siRNA and treated with poly(I:C) for 24 h are shown. *p < 0.05 (Mann-Whitney U-test). Scale bar, 100 μm. g and h Expression of cytotoxicity-associated ISGs is potentiated by loss of paraspeckles in cells depleted of TDP-43 or Drosha (g) or stimulated by poly(I:C) (h). N = 3, *p < 0.05 (Mann-Whitney U-test) Finally, we tested whether inhibition of miRNA Paraspeckle assembly can be promoted by a small function, exposure to dsRNA or IFNbeta treatment molecule enhancer of miRNA biogenesis can initiate paraspeckle formation in neurons. Hu- Very few chemicals capable of stimulating paraspeckle man Day 40 motor neurons [30] were stimulated with assembly in cells with pre-existing paraspeckles have poly(I:C) or treated with suramin or IFNbeta for 24 h. In- been identified so far. Proteasome inhibitors are known duction of endogenous IFNbeta and NEAT1 was observed to promote NEAT1 synthesis and paraspeckle formation in poly(I:C)-treated neuronal cultures (Fig. 6c), and para- [13], however they are poor candidates as therapeutic speckles could be detected in a small fraction of poly(- molecules for ALS. Enoxacin is a small molecule enhan- I:C)-stimulated neurons (Fig. 6d). In contrast, neither cer of the miRNA pathway which stimulates the activity suramin nor IFNbeta induced paraspeckle assembly in of the Dicer complex [50, 51]. In doing so, enoxacin in- neurons. creases levels of mature miRNAs thereby depleting Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 12 of 17 Fig. 6 Post-mitotic neurons lack paraspeckles in vitro, but their assembly can be triggered by dsRNA. a Paraspeckles are present in glial cells but not in neurons in murine primary hippocampal cultures. Cultures were analysed at DIV14 by NEAT1_2 RNA-FISH. A representative image is shown. Neuronal nuclei are circled and paraspeckles in glial cells are indicated by arrowheads. b Paraspeckles are present in human neural precursor cells (NPCs) but disappear during their differentiation into motor neurons. Cultures were analysed at the indicated time-points by NEAT1_2 RNA-FISH. c and d Treatment of Day 40 cultures of human motor neurons with poly(I:C) leads to activation of IFN signaling, increased NEAT1 expression (c) and paraspeckle assembly in a fraction of cells (d, arrowheads). In c, gene expression was analysed by qRT-PCR (n = 4). *p < 0.05, **p < 0.01 (Mann-Whitney U-test). Scale bars, 10 μmin a, b and 20 μmin d miRNA precursors, pri-miRNA and pre-miRNA [51]. Re- reported to stimulate NEAT1 expression [53]and twoap- cently, enoxacin been shown to ameliorate pathology in proved ALS therapeutics, riluzole and edaravone. mouse models of ALS [52]. We studied the effect of enox- Treatment with 10 μM enoxacin for 24 h increased acin on paraspeckles in neuroblastoma cells. In addition, NEAT1_2 levels and paraspeckle assembly in neuroblast- in our analysis we included HDAC inhibitors which were oma cells (Fig. 7a and b); similar effect on paraspeckles Fig. 7 A small molecule enhancer of miRNA biogenesis stimulates paraspeckle assembly in neuroblastoma cells. a and b SH-SY5Y cells were treated with enoxacin, riluzole, edaravone and HDAC inhibitors trichostatin A (TSA) and sodium butyrate (NaB), and paraspeckle assembly was assessed by NEAT1_2 RNA-FISH (a) and qRT-PCR (n = 4–6, b). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Kruskal-Wallis test with Dunn’s correction for multiple comparisons). Cells were harvested for analysis after 4 h of TSA and NaB treatment (500 nM and 2 mM, respectively) and after 24 h of enoxacin, riluzole and edaravone treatment (all 10 μM). Scale bar, 10 μm Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 13 of 17 was observed with short (4 h) treatment and with a response. Given a significant crosstalk between miRNA higher enoxacin dose (Additional file 6: Fig. S6). Global and dsRNA response pathways [64], it is not surprising HDAC inhibitors trichostatin A (TSA) and sodium bu- that paraspeckles function as a regulatory platform for tyrate (NaB) also significantly increased NEAT1_2 levels both pathways. Although dsRNA response triggered by and led to the formation of large, elongated paraspeckles dysfunction of the miRNA pathway factors is mediated (Fig. 7a and b). Interestingly, edaravone but not riluzole via different molecular sensors, including TLR3 (for also promoted paraspeckle biogenesis (Fig. 7a and b). Drosha, our unpublished observations), MyD88 (for None of the compounds studied was able to trigger de Dicer) [65] and MDA5/RIG-I (for ADAR1) [66], it even- novo paraspeckles in hES cell derived motor neurons. tually converges on type I IFN. In contrast to the previ- ous study [14], we found that IFNbeta treatment alone Discussion can stimulate NEAT1 expression and paraspeckle forma- Nuclear bodies spatially organize and modulate various tion. This discrepancy is likely due to the transient effect cellular processes [54]. Therefore it is not surprising that of IFN treatment on paraspeckles which peaks at the 4-h these membraneless organelles and their components time-point, whereas in the previous work, the 24 h have been implicated in multiple human diseases. Prom- time-point was examined. It is possible that IFN levels inent examples are PML bodies and Gems linked to car- oscillate to maintain the dsRNA response active but at cinogenesis and motor neuron degeneration, respectively the same time preserve cellular viability [67]. Our in [55, 56]. Paraspeckles have recently come into the lime- vitro data are consistent with a recent in vivo study dem- light in the ALS field because of the extensive involve- onstrating that TDP-43 knockdown in the adult murine ment of paraspeckle proteins in ALS pathogenesis. In nervous system leads to widespread upregulation of im- the present study, we found that paraspeckle assembly mune and, more specifically, antiviral genes [27]. Loss of in the spinal cord is shared by ALS cases with different TDP-43 function in the nervous system might be suffi- aetiology and, as such, a hallmark of the disease. Using cient to trigger a chronic neuroinflammatory response. cell models, we identified two possible mechanisms In a previous study, siRNA-mediated TDP-43 knock- which may initiate paraspeckle assembly in the spinal down led to decreased paraspeckle numbers in HeLa cord cells of the majority of ALS cases – compromised cells [8]. One possible explanation for this discrepancy is miRNA biogenesis and activated dsRNA response – differences in cellular response to dsRNA and/or differ- both downstream of loss of TDP-43 function. ences in the miRNA pathway regulation between the cell In the CNS, miRNAs are highly abundant and are sub- lines. Indeed, in the study on TDP-43 functions as an ject to abnormal regulation in many neurodegenerative RNA chaperone, dsRNA was shown to be accumulated diseases, including ALS [57–61]. Levels of mature miR- only in the nucleus of HeLa cells, whereas in neuroblast- NAs in ALS spinal cord were reported to be globally re- oma M17 cell line it was mainly cytoplasmic [38], similar duced [52, 57, 62]. This dysregulation is consistent with to our study. Another possibility is the reliance of para- TDP-43 loss of function in the majority of ALS cases speckle assembly on some TDP-43 function(s) specific- since this protein is a known miRNA biogenesis factor ally in HeLa cells. [32]. An important role of paraspeckles in miRNA pro- Loss of TDP-43 function can explain paraspeckle cessing is supported by two recent studies [16, 63]. hyper-assembly in the majority of ALS cases, i.e. almost Thus, paraspeckle hyper-assembly in ALS motor neu- all sALS cases as well as fALS cases caused by mutations rons affected by TDP-43 loss of function may serve as in TARDBP and C9ORF72 genes. Recently, Drosha has one of the mechanisms to compensate for miRNA bio- been identified as a component of C9orf72 dipeptide in- genesis deficiency. We also show that not only compro- clusions in patient’s neurons [68]. Therefore in fALS-C9 mised function of the miRNA pathway but also its cases, loss of function for both TDP-43 and Drosha can pharmacological enhancement results in paraspeckle jointly contribute to paraspeckle response. In a subset of hyper-assembly. This suggests that paraspeckles can re- sALS patients, activation of an endogenous retrovirus spond to bi-directional changes in the activity of the (ERV), HERV-K, was reported [69]. Elevated expression miRNA pathway to either compensate for its compro- of ERVs can initiate dsRNA response [70, 71]. Activation mised function or to meet the demand for miRNA pre- of HERV-K may therefore contribute to paraspeckle cursors when its final step is over-active. hyper-assembly at least in some sALS cases. It still re- Another function of TDP-43 is acting as a chaperone mains to be established whether paraspeckle formation to control RNA secondary structure [38] and therefore is typical for other fALS cases such as those caused by cellular dsRNA response. Paraspeckles are known to re- mutations in genes encoding SOD1, FUS, TBK1 or spond to exogenous dsRNA (viral and its analogues) OPTN and, if so, the underlying mechanisms. Many of [14], and here we show that abnormal accumulation of ALS proteins function in miRNA and dsRNA metabol- endogenous dsRNA can also initiate paraspeckle ism, for example FUS is involved in miRNA biogenesis Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 14 of 17 and miRNA-mediated silencing [35, 72], whereas TBK1 scrambled siRNA or Silencer® TDP-43 siRNA and analysed by qRT-PCR (n =6). is one of the central factors in dsRNA response and type Mean number of paraspeckles per cell was also quantified using NEAT1_2 RNA FISH. **p <0.01, ***p < 0.001 (Mann-Whitney U-test in a and Student’s t- I IFN signaling. Thus compromised function of these test in b). c and d Downregulation of TDP-43 using an esiRNA (endoribonu- pathways may represent a common mechanism behind clease-prepared MISSION® esiRNA, c) or shRNA (d) stimulates paraspeckle paraspeckle response in different ALS cases. assembly. The efficiency of knockdown was analysed by TDP-43 immunocyto- chemistry, and paraspeckle assembly – by NEAT1_2 RNA-FISH. Representative Previously, paraspeckles were shown to be protective images are shown. Scale bars, 100 μm for left panels and 10 μmfor right against cell death caused by proteasomal inhibition [13]. panels. e Speckles visualised by MALAT1 RNA-FISH are not affected by TDP-43 In current study, we show that paraspeckles confer pro- knockdown. Representative images are shown. Scale bar, 10 μm. f Levels of core paraspeckle proteins NONO, SFPQ and FUS are not affected by TDP-43 tection to cells with compromised metabolism of miRNA knockdown in SH-SY5Y cells. Representative Western blots are shown. g Se- and activated dsRNA response. Intriguingly, many of quences and positions of gRNAs used for disrupting the NLS of the endogen- miRNA-controlled cytotoxicity-associated ISGs [48]were ous TDP-43 protein by CRISPR/Cas9-mediated editing. Two combinations of upstream and downstream gRNA sequences within TARDBP gene selected to also reported to be regulated by paraspeckles either by se- disrupt the NLS are shown. The sequence encoding for the NLS is given in questration of transcription factors or by nuclear retention blue and PAM sites are boxed. h Transient transfection of two combinations of edited RNAs [14, 49, 73] suggesting a multi-layered of plasmids encoding upstream and downstream gRNAs for targeting the NLS of TDP-43 results in partial redistribution of endogenous TDP-43 but does not control of cellular toxicity by paraspeckles. It should be lead to enhanced paraspeckle formation. Cells were analysed 72 h post- noted however that the effect of paraspeckle disruption on transfection. Representative images are shown, asterisks indicate cells with survival in stable cell lines was small both in this and in cytoplasmic TDP-43 redistribution. Scale bar, 10 μm. In a-f, cells were analysed 48 h post-transfection. (DOCX 517 kb) the previous [13] report, despite the use of cells com- Additional file 2: Figure S2. Drosha or TDP-43 downregulation affects pletely lacking NEAT1 and hence paraspeckles in the lat- miRNA processing. a Drosha knockdown leads to accumulation of miRNA pre- ter study. Such limited effect is in line with the fact that cursors, pri-miR-17-92a and pri-miR-15a (n =4–6). **p < 0.01 (one-way ANOVA NEAT1 knockout mice do not have an overt phenotype with Holm-Sidak correction for multiple comparisons). Note the absence of significant accumulation of these pri-miRNAs in TDP-43 depleted cells, in ac- [21] and further supports a modulatory role for para- cord with modulatory rather than essential function of this protein in miRNA speckles in cellular responses (such as miRNA biogenesis, processing in the nucleus. b Drosha or TDP-43 knockdown leads to downreg- gene expression, RNA retention) which only becomes ulation of mature miRNAs processed from pri-miR-17-92a (n =3). *p <0.05; **p < 0.01 (one-way ANOVA with Holm-Sidak correction for multiple compari- relevant under stressful/pathophysiological conditions. sons). Note that levels of all three mature miRNAs are significantly decreased However, such modulatory activities of paraspeckles might in Drosha depleted cells, and TDP-43 knockdown also negatively affects two be particularly important for neurons coping with of the three miRNAs measured. (DOCX 157 kb) neurodegeneration-inducing stresses. Additional file 3: Figure S3. Accumulation of dsRNA (a) and increased levels of p-eIF2α (b) in MCF7 cells depleted of TDP-43. Cells were analysed 48 h post-transfection. Scale bars, 100 μm and 10 μm for general plane and Conclusions close-up panels respectively. (DOCX 663 kb) Paraspeckle hyper-assembly might be broadly neuropro- Additional file 4: Figure S4. The effect of IFN-inducing ligands on NEAT1 and paraspeckles. a IFNbeta is robustly induced by poly(I:C) in tective in ALS. As a word of caution, however, these data neuroblastoma cells. Cells were analysed by qRT-PCR after 24 h of were obtained in stable cell lines with pre-existing para- poly(I:C) stimulation (n = 5). **p < 0.01 (Mann-Whitney U-test). b-d TLR3 speckles. Since normal post-mitotic neurons are free from and TLR4 ligands poly(I:C) and LPS, but not a TLR2 ligand zymosan, trigger IFNbeta response stimulating NEAT1 expression (b) and para- paraspeckles, the impact of their de novo formation on speckle assembly (c). Cells were treated with poly(I:C), LPS or zymosan for neuronal metabolism may be much more dramatic than 4 h and analysed by qRT-PCR (n = 3 or 4). *p < 0.05, **p < 0.01. NF-κB in stable cell lines. For example, it is coupled with changes nuclear translocation was examined in parallel to confirm the activity of the compounds (d, asterisks indicate cells with nuclear NF-κB). Scale bar, in the levels of the short NEAT1 isoform with diverse 10 μm. (DOCX 333 kb) paraspeckle-independent regulatory functions, including Additional file 5: Figure S5. Differentiation of human modulation of neuronal excitability [74–77]. Further stud- neuroblastoma cells into neuron-like cells does not lead to the loss ies, using neurons derived from human stem cells with ab- of paraspeckles. Differentiated SH-SY5Y cells develop extensive neurite network and are uniformly positive for a neuronal marker lated NEAT1_2 expression and from ALS patients’ iPS Tuj 1 (left panel) but preserve their ability to form paraspeckles cells, are required to understand whether paraspeckles are (right panel). SH-SY5Y cells were induced to differentiate into protective for motor neurons in the disease context and in neuron-like cells using retinoic acid/BDNF andanalysed6days into differentiation by immunocytochemitry and NEAT1_2 RNA-FISH. the long-term. This shall help us understand whether, how Representative images are shown. Scale bars, 100 μm(left panel) and when paraspeckles can be targeted for therapeutic and 10 μm (right panel). (DOCX 175 kb) purposes. Additional file 6: Figure S6. Dose-dependent toxicity of enoxacin, edaravone and riluzole. a SH-SY5Y cells were treated with correspond- ing doses of compounds for 24 h, and toxicity was assessed using Cell- Additional files Titer Blue® Cell Viability Assay. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as compared to control (non-treated) cells (Kruskal-Wallis test Additional file 1: Figure S1. The effect of TDP-43 dysfunction on para- with Dunn’s correction for multiple comparisons). b Enoxacin enhances paraspeckle assembly both with short (4 h) and prolonged (24 h) treat- speckles, speckles and paraspeckle proteins in MCF7 and SH-SY5Y cells. a and b TDP-43 siRNA-mediated knockdown upregulates NEAT1_2 (a) and enhances ment at a non-toxic concentration of 50 μM. Representative images are paraspeckle assembly (b) in SH-SY5Y cells. Cells were transfected with shown. (DOCX 298 kb) Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 15 of 17 Abbreviations 5. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, et (F)ISH: (fluorescent) in situ hybridisation; ALS: Amyotrophic lateral sclerosis; al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and dsRNA: double-stranded RNA; FTLD: Frontotemporal lobar degeneration; amyotrophic lateral sclerosis. Science. 2006;314:130–3. https://doi.org/10. HDAC: Histone deacetylase; IFN: Interferon; ISG: Interferon-stimulated gene; 1126/science.1134108. LPS: Lipopolysaccharide; miRNA: microRNA; NEAT1: Nuclear enriched 6. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, et al. TDP-43 is a abundant transcript 1; NPC: Neural precursor cell; component of ubiquitin-positive tau-negative inclusions in frontotemporal poly(I:C): Polyinosinic:polycytidylic acid; TDP-43: TAR DNA-binding protein 43; lobar degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res TSA: Trichostatin A Commun. 2006;351:602–11. https://doi.org/10.1016/j.bbrc.2006.10.093. 7. Fox AH, Lamond AI. Paraspeckles. Cold Spring Harb Perspect Biol. 2010;2: Acknowledgements a000687. https://doi.org/10.1101/cshperspect. a000687. We acknowledge the Sheffield Brain Tissue Bank and London 8. Naganuma T, Nakagawa S, Tanigawa A, Sasaki YF, Goshima N, Hirose T. Neurodegenerative Diseases Brain Bank for providing human materials. Alternative 3′-end processing of long noncoding RNA initiates construction of nuclear paraspeckles. EMBO J. 2012;31:4020–34. https://doi.org/10.1038/ emboj.2012.251. Funding 9. West JA, Mito M, Kurosaka S, Takumi T, Tanegashima C, Chujo T, et al. TAS is a recipient of a fellowship from Medical Research Foundation. The Structural, super-resolution microscopy analysis of paraspeckle nuclear body study was also funded by Research Grant from Motor Neuron Disease organization. J Cell Biol. 2016; https://doi.org/10.1083/jcb.201601071. Association to VLB (Buchman/Apr13/6096) and miRNA analysis was 10. Clemson CM, Hutchinson JN, Sara SA, Ensminger AW, Fox AH, Chess A, et al. supported by the Russian Science Foundation grant (18–15-00357). HA and An architectural role for a nuclear noncoding RNA: NEAT1 RNA is essential PD are recipients of Cardiff University/China Council PhD studentship and for the structure of paraspeckles. Mol Cell. 2009;33:717–26. https://doi.org/ Erasmus studentship, respectively. 10.1016/j.molcel.2009.01.026. 11. Hennig S, Kong G, Mannen T, Sadowska A, Kobelke S, Blythe A, et al. Prion- Availability of data and materials like domains in RNA binding proteins are essential for building subnuclear The datasets used and/or analyzed during the current study are available paraspeckles. J Cell Biol. 2015;210:529–39. https://doi.org/10.1083/jcb. from the corresponding author on reasonable request. 12. Sunwoo H, Dinger ME, Wilusz JE, Amaral PP, Mattick JS. Spector DL. MEN Authors’ contributions epsilon/beta nuclear-retained non-coding RNAs are up-regulated upon TAS conceived research; TAS, HA, MSK, PD, SA, OS and PRH performed muscle differentiation and are essential components of paraspeckles. experiments; TAS and VLB analysed data; TAS wrote manuscript with input Genome Res. 2009;19:347–59. https://doi.org/10.1101/gr.087775.108. from all authors. All authors read and approved the final version of the 13. Hirose T, Virnicchi G, Tanigawa A, Naganuma T, Li R, Kimura H, et al. NEAT1 manuscript. long noncoding RNA regulates transcription via protein sequestration within subnuclear bodies. Mol Biol Cell. 2014;25:169–83. https://doi.org/10. Ethics approval and consent to participate 1091/mbc.E13-09-0558. Human samples from clinically and histopathologically characterised ALS 14. Imamura K, Imamachi N, Akizuki G, Kumakura M, Kawaguchi A, Nagata K, et al. cases and neurologically healthy individuals were provided by the Sheffield Long noncoding RNA NEAT1-dependent SFPQ relocation from promoter Brain Tissue Bank and MRC London Neurodegenerative Diseases Brain Bank region to paraspeckle mediates IL8 expression upon immune stimuli. Mol Cell. (Institute of Psychiatry, King’s College London). Consent was obtained from 2014;53:393–406. https://doi.org/10.1016/j.molcel.2014.01.009. all subjects for autopsy, histopathological assessment and research were 15. Zhang Z, Carmichael GG. The fate of dsRNA in the nucleus: a p54(nrb)- performed in accordance with local and national Ethics Committee containing complex mediates the nuclear retention of promiscuously A-to-I approved donation. edited RNAs. Cell. 2001;106:465–75. http://www.ncbi.nlm.nih.gov/pubmed/ Competing interests 16. Jiang L, Shao C, Wu QJ, Chen G, Zhou J, Yang B, et al. NEAT1 scaffolds RNA- The authors declare that they have no competing interests. binding proteins and the microprocessor to globally enhance pri-miRNA processing. Nat Struct Mol Biol. 2017;24:816–24. https://doi.org/10.1038/ nsmb.3455. Publisher’sNote 17. Choudhry H, Albukhari A, Morotti M, Haider S, Moralli D, Smythies J, et al. Springer Nature remains neutral with regard to jurisdictional claims in Tumor hypoxia induces nuclear paraspeckle formation through HIF-2alpha published maps and institutional affiliations. dependent transcriptional activation of NEAT1 leading to cancer cell survival. Oncogene. 2015;34:4546. https://doi.org/10.1038/onc.2014.431. Author details 18. Chakravarty D, Sboner A, Nair SS, Giannopoulou E, Li R, Hennig S, et al. The School of Biosciences, Cardiff University, Museum Avenue, Cardiff CF10 3AX, oestrogen receptor alpha-regulated lncRNA NEAT1 is a critical modulator of UK. Institute of Physiologically Active Compounds Russian Academy of prostate cancer. Nat Commun. 2014;5:5383. https://doi.org/10.1038/ncomms6383. Sciences, 1 Severniy proezd, Chernogolovka, Moscow Region, Russian 19. Adriaens C, Standaert L, Barra J, Latil M, Verfaillie A, Kalev P, et al. p53 Federation142432. The Sheffield Institute for Translational Neuroscience, induces formation of NEAT1 lncRNA-containing paraspeckles that modulate 385A Glossop Road, Sheffield S10 2HQ, UK. replication stress response and chemosensitivity. Nat Med. 2016;22:861–8. https://doi.org/10.1038/nm.4135. Received: 13 March 2018 Accepted: 25 May 2018 20. Nishimoto Y, Nakagawa S, Hirose T, Okano HJ, Takao M, Shibata S, et al. The long non-coding RNA nuclear-enriched abundant transcript 1_2 induces paraspeckle formation in the motor neuron during the early phase of References amyotrophic lateral sclerosis. Molecular brain. 2013;6:31. https://doi.org/10. 1. Renton AE, Chio A, Traynor BJ. State of play in amyotrophic lateral sclerosis 1186/1756-6606-6-31. genetics. Nat Neurosci. 2014;17:17–23. https://doi.org/10.1038/nn.3584. 21. Nakagawa S, Naganuma T, Shioi G, Hirose T. Paraspeckles are 2. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, subpopulation-specific nuclear bodies that are not essential in mice. J Cell Rutherford NJ, et al. Expanded GGGGCC hexanucleotide repeat in Biol. 2011;193:31–9. https://doi.org/10.1083/jcb.201011110. noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. 22. Thomas-Jinu S, Gordon PM, Fielding T, Taylor R, Smith BN, Snowden V, et al. Neuron. 2011;72:245–56. https://doi.org/10.1016/j.neuron.2011.09.011. Non-nuclear pool of splicing factor SFPQ regulates axonal transcripts 3. Sreedharan J, Blair IP, Tripathi VB, Hu X, Vance C, Rogelj B, et al. TDP-43 required for normal motor development. Neuron. 2017;94:931. https://doi. mutations in familial and sporadic amyotrophic lateral sclerosis. Science. org/10.1016/j.neuron.2017.04.036. 2008;319:1668–72. https://doi.org/10.1126/science.1154584. 23. Shelkovnikova TA, Robinson HK, Troakes C, Ninkina N, Buchman VL. 4. Lattante S, Rouleau GA, Kabashi E. TARDBP and FUS mutations associated Compromised paraspeckle formation as a pathogenic factor in FUSopathies. with amyotrophic lateral sclerosis: summary and update. Hum Mutat. 2013; Hum Mol Genet. 2014;23:2298–312. https://doi.org/10.1093/hmg/ddt622. 34:812–26. https://doi.org/10.1002/humu.22319. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 16 of 17 24. Chesi A, Staahl BT, Jovicic A, Couthouis J, Fasolino M, Raphael AR, et al. 44. Kaneko H, Dridi S, Tarallo V, Gelfand BD, Fowler BJ, Cho WG, et al. DICER1 Exome sequencing to identify de novo mutations in sporadic ALS trios. Nat deficit induces Alu RNA toxicity in age-related macular degeneration. Neurosci. 2013;16:851–5. https://doi.org/10.1038/nn.3412. Nature. 2011;471:325–30. https://doi.org/10.1038/nature09830. 25. Kim HJ, Kim NC, Wang YD, Scarborough EA, Moore J, Diaz Z, et al. 45. Marullo M, Zuccato C, Mariotti C, Lahiri N, Tabrizi SJ, Di Donato S, et al. Mutations in prion-like domains in hnRNPA2B1 and hnRNPA1 cause Expressed Alu repeats as a novel, reliable tool for normalization of real-time multisystem proteinopathy and ALS. Nature. 2013;495:467–73. https://doi. quantitative RT-PCR data. Genome Biol. 2010;11:R9. https://doi.org/10.1186/ org/10.1038/nature11922. gb-2010-11-1-r9. 26. Tollervey JR, Curk T, Rogelj B, Briese M, Cereda M, Kayikci M, et al. 46. Marciniak SJ, Yun CY, Oyadomari S, Novoa I, Zhang Y, Jungreis R, et al. Characterizing the RNA targets and position-dependent splicing regulation CHOP induces death by promoting protein synthesis and oxidation in the by TDP-43. Nat Neurosci. 2011;14:452–8. https://doi.org/10.1038/nn.2778. stressed endoplasmic reticulum. Genes Dev. 2004;18:3066–77. https://doi. 27. Polymenidou M, Lagier-Tourenne C, Hutt KR, Huelga SC, Moran J, Liang TY, org/10.1101/gad.1250704. et al. Long pre-mRNA depletion and RNA missplicing contribute to 47. Matsumoto M, Minami M, Takeda K, Sakao Y, Akira S. Ectopic expression of neuronal vulnerability from loss of TDP-43. Nat Neurosci. 2011;14:459–68. CHOP (GADD153) induces apoptosis in M1 myeloblastic leukemia cells. FEBS https://doi.org/10.1038/nn.2779. Lett 1996; 395:143–147. doi: 0014–5793(96)01016–2. 28. Kukharsky MS, Quintiero A, Matsumoto T, Matsukawa K, An H, Hashimoto T, 48. Seo GJ, Kincaid RP, Phanaksri T, Burke JM, Pare JM, Cox JE, et al. Reciprocal et al. Calcium-responsive transactivator (CREST) protein shares a set of inhibition between intracellular antiviral signaling and the RNAi machinery structural and functional traits with other proteins associated with in mammalian cells. Cell Host Microbe. 2013;14:435–45. https://doi.org/10. amyotrophic lateral sclerosis. Mol Neurodegener. 2015;10:20. https://doi.org/ 1016/j.chom.2013.09.002. 10.1186/s13024-015-0014-y. 49. Chen LL, Carmichael GG. Altered nuclear retention of mRNAs containing 29. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome inverted repeats in human embryonic stem cells: functional role of a engineering using CRISPR/Cas systems. Science. 2013;339:819–23. https:// nuclear noncoding RNA. Mol Cell. 2009;35:467–78. https://doi.org/10.1016/j. doi.org/10.1126/science.1231143. molcel.2009.06.027. 30. Shelkovnikova TA, Dimasi P, Kukharsky MS, An H, Quintiero A, Schirmer C, et 50. Melo S, Villanueva A, Moutinho C, Davalos V, Spizzo R, Ivan C, et al. Small al. Chronically stressed or stress-preconditioned neurons fail to maintain molecule enoxacin is a cancer-specific growth inhibitor that acts by stress granule assembly. Cell Death Dis. 2017;8:e2788. https://doi.org/10. enhancing TAR RNA-binding protein 2-mediated microRNA processing. Proc 1038/cddis.2017.199. Natl Acad Sci U S A. 2011;108:4394–9. https://doi.org/10.1073/pnas. 31. Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, Gibbs JR, et 1014720108. al. A hexanucleotide repeat expansion in C9ORF72 is the cause of 51. Shan G, Li Y, Zhang J, Li W, Szulwach KE, Duan R, et al. A small molecule chromosome 9p21-linked ALS-FTD. Neuron. 2011;72:257–68. https://doi.org/ enhances RNA interference and promotes microRNA processing. Nat 10.1016/j.neuron.2011.09.010. Biotechnol. 2008;26:933–40. https://doi.org/10.1038/nbt.1481. 32. Kawahara Y, Mieda-Sato A. TDP-43 promotes microRNA biogenesis as a 52. Emde A, Eitan C, Liou LL, Libby RT, Rivkin N, Magen I, et al. Dysregulated component of the Drosha and dicer complexes. Proc Natl Acad Sci U S A. miRNA biogenesis downstream of cellular stress and ALS-causing mutations: 2012;109:3347–52. https://doi.org/10.1073/pnas.1112427109. a new mechanism for ALS. EMBO J. 2015;34:2633–51. https://doi.org/10. 15252/embj.201490493. 33. Buratti E, De Conti L, Stuani C, Romano M, Baralle M, Baralle F. Nuclear factor TDP-43 can affect selected microRNA levels. FEBS J. 2010;277:2268–81. 53. Schor IE, Lleres D, Risso GJ, Pawellek A, Ule J, Lamond AI, et al. Perturbation https://doi.org/10.1111/j.1742-4658.2010.07643.x. of chromatin structure globally affects localization and recruitment of 34. Ota H, Sakurai M, Gupta R, Valente L, Wulff BE, Ariyoshi K, et al. ADAR1 splicing factors. PLoS One. 2012;7:e48084. https://doi.org/10.1371/journal. forms a complex with dicer to promote microRNA processing and RNA- pone.0048084. induced gene silencing. Cell. 2013;153:575–89. https://doi.org/10.1016/j.cell. 54. Misteli T. Higher-order genome organization in human disease. Cold Spring 2013.03.024. Harb Perspect Biol. 2010;2:a000794. https://doi.org/10.1101/cshperspect. 35. Morlando M, Dini Modigliani S, Torrelli G, Rosa A, Di Carlo V, Caffarelli E, et a000794. al. FUS stimulates microRNA biogenesis by facilitating co-transcriptional 55. de The H, Le Bras M, Lallemand-Breitenbach V. The cell biology of disease: Drosha recruitment. EMBO J. 2012;31:4502–10. https://doi.org/10.1038/ acute promyelocytic leukemia, arsenic, and PML bodies. J Cell Biol 2012;198: emboj.2012.319. 11–21. doi: https://doi.org/10.1083/jcb.201112044. 36. Tan GS, Chiu CH, Garchow BG, Metzler D, Diamond SL, Kiriakidou M. Small 56. Yamazaki T, Chen S, Yu Y, Yan B, Haertlein TC, Carrasco MA, et al. FUS-SMN molecule inhibition of RISC loading. ACS Chem Biol. 2012;7:403–10. https:// protein interactions link the motor neuron diseases ALS and SMA. Cell Rep. doi.org/10.1021/cb200253h. 2012;2:799–806. https://doi.org/10.1016/j.celrep.2012.08.025. 37. Lagier-Tourenne C, Polymenidou M, Hutt KR, Vu AQ, Baughn M, Huelga SC, 57. Figueroa-Romero C, Hur J, Lunn JS, Paez-Colasante X, Bender DE, Yung R, et et al. Divergent roles of ALS-linked proteins FUS/TLS and TDP-43 intersect in al. Expression of microRNAs in human post-mortem amyotrophic lateral processing long pre-mRNAs. Nat Neurosci. 2012;15:1488–97. https://doi.org/ sclerosis spinal cords provides insight into disease mechanisms. Mol Cell 10.1038/nn.3230. Neurosci. 2016;71:34–45. https://doi.org/10.1016/j.mcn.2015.12.008. 38. Saldi TK, Ash PE, Wilson G, Gonzales P, Garrido-Lecca A, Roberts CM, et al. 58. Freischmidt A, Muller K, Ludolph AC, Weishaupt JH. Systemic dysregulation TDP-1, the Caenorhabditis elegans ortholog of TDP-43, limits the of TDP-43 binding microRNAs in amyotrophic lateral sclerosis. Acta accumulation of double-stranded RNA. EMBO J. 2014;33:2947–66. https:// Neuropathol Commun. 2013;1:42. https://doi.org/10.1186/2051-5960-1-42. doi.org/10.15252/embj.201488740. 59. Gascon E, Gao FB. The emerging roles of microRNAs in the pathogenesis of 39. Ma H, Han P, Ye W, Chen H, Zheng X, Cheng L, et al. The long noncoding frontotemporal dementia-amyotrophic lateral sclerosis (FTD-ALS) spectrum RNA NEAT1 exerts Antihantaviral effects by acting as positive feedback for disorders. J Neurogenet. 2014;28:30–40. https://doi.org/10.3109/01677063. RIG-I signaling. J Virol. 2017;91 https://doi.org/10.1128/JVI.02250-16. 2013.876021. 40. White E, Schlackow M, Kamieniarz-Gdula K, Proudfoot NJ, Gullerova M. Human 60. Goodall EF, Heath PR, Bandmann O, Kirby J, Shaw PJ. Neuronal dark matter: nuclear dicer restricts the deleterious accumulation of endogenous double-stranded the emerging role of microRNAs in neurodegeneration. Front Cell Neurosci. RNA. Nat Struct Mol Biol. 2014;21:552–9. https://doi.org/10.1038/nsmb.2827. 2013;7:178. https://doi.org/10.3389/fncel.2013.00178. 41. Heras SR, Macias S, Plass M, Fernandez N, Cano D, Eyras E, et al. The 61. Eitan C, Hornstein E. Vulnerability of microRNA biogenesis in FTD-ALS. Brain microprocessor controls the activity of mammalian retrotransposons. Nat Res. 2016;1647:105–11. https://doi.org/10.1016/j.brainres.2015.12.063. Struct Mol Biol. 2013;20:1173–81. https://doi.org/10.1038/nsmb.2658. 62. Campos-Melo D, Droppelmann CA, He Z, Volkening K, Strong MJ. Altered 42. Liddicoat BJ, Piskol R, Chalk AM, Ramaswami G, Higuchi M, Hartner JC, et al. microRNA expression profile in amyotrophic lateral sclerosis: a role in the RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as regulation of NFL mRNA levels. Molecular brain. 2013;6:26. https://doi.org/ nonself. Science. 2015;349:1115–20. https://doi.org/10.1126/science.aac7049. 10.1186/1756-6606-6-26. 43. Weber F, Wagner V, Rasmussen SB, Hartmann R, Paludan SR. Double- 63. Bottini S, Hamouda-Tekaya N, Mategot R, Zaragosi LE, Audebert S, Pisano S, stranded RNA is produced by positive-strand RNA viruses and DNA viruses et al. Post-transcriptional gene silencing mediated by microRNAs is but not in detectable amounts by negative-strand RNA viruses. J Virol. 2006; controlled by nucleoplasmic Sfpq. Nat Commun. 2017;8:1189. https://doi. 80:5059–64. https://doi.org/10.1128/JVI.80.10.5059-5064.2006. org/10.1038/s41467-017-01126-x. Shelkovnikova et al. Molecular Neurodegeneration (2018) 13:30 Page 17 of 17 64. Heyam A, Lagos D, Plevin M. Dissecting the roles of TRBP and PACT in double-stranded RNA recognition and processing of noncoding RNAs. Wiley Interdiscip Rev RNA. 2015;6:271–89. https://doi.org/10.1002/wrna.1272. 65. Tarallo V, Hirano Y, Gelfand BD, Dridi S, Kerur N, Kim Y, et al. DICER1 loss and Alu RNA induce age-related macular degeneration via the NLRP3 inflammasome and MyD88. Cell. 2012;149:847–59. https://doi.org/10.1016/j. cell.2012.03.036. 66. Mannion NM, Greenwood SM, Young R, Cox S, Brindle J, Read D, et al. The RNA-editing enzyme ADAR1 controls innate immune responses to RNA. Cell Rep. 2014;9:1482–94. https://doi.org/10.1016/j.celrep.2014.10.041. 67. Ruggieri A, Dazert E, Metz P, Hofmann S, Bergeest JP, Mazur J, et al. Dynamic oscillation of translation and stress granule formation mark the cellular response to virus infection. Cell Host Microbe. 2012;12:71–85. https://doi.org/10.1016/j.chom.2012.05.013. 68. Porta S, Kwong LK, Trojanowski JQ, Lee VM. Drosha inclusions are new components of dipeptide-repeat protein aggregates in FTLD-TDP and ALS C9orf72 expansion cases. J Neuropathol Exp Neurol. 2015;74:380–7. https:// doi.org/10.1097/NEN.0000000000000182. 69. Li W, Lee MH, Henderson L, Tyagi R, Bachani M, Steiner J, et al. Human endogenous retrovirus-K contributes to motor neuron disease. Sci Transl Med. 2015;7:307ra153. https://doi.org/10.1126/scitranslmed.aac8201. 70. Hurst TP, Magiorkinis G. Activation of the innate immune response by endogenous retroviruses. J Gen Virol. 2015;96:1207–18. https://doi.org/10. 1099/jgv.0.000017. 71. Chiappinelli KB, Strissel PL, Desrichard A, Li H, Henke C, Akman B, et al. Inhibiting DNA methylation causes an interferon response in Cancer via dsRNA including endogenous retroviruses. Cell. 2016;164:1073. https://doi. org/10.1016/j.cell.2015.10.020. 72. Zhang T, Wu YC, Mullane P, Ji YJ, Liu H, He L, et al. FUS regulates activity of MicroRNA-mediated gene silencing. Mol Cell 2018; 69:787–801 e8. doi: https://doi.org/10.1016/j.molcel.2018.02.001. 73. Elbarbary RA, Li W, Tian B, Maquat LE. STAU1 binding 3' UTR IRAlus complements nuclear retention to protect cells from PKR-mediated translational shutdown. Genes Dev. 2013;27:1495–510. https://doi.org/10. 1101/gad.220962.113. 74. Barry G, Briggs JA, Hwang DW, Nayler SP, Fortuna PR, Jonkhout N, et al. The long non-coding RNA NEAT1 is responsive to neuronal activity and is associated with hyperexcitability states. Sci Rep. 2017;7:40127. https://doi. org/10.1038/srep40127. 75. Li R, Harvey AR, Hodgetts SI, Fox AH. Functional dissection of NEAT1 using genome editing reveals substantial localisation of the NEAT1_1 isoform outside paraspeckles. RNA. 2017; https://doi.org/10.1261/rna.059477.116. 76. Zhang F, Wu L, Qian J, Qu B, Xia S, La T, et al. Identification of the long noncoding RNA NEAT1 as a novel inflammatory regulator acting through MAPK pathway in human lupus. J Autoimmun. 2016;75:96–104. https://doi. org/10.1016/j.jaut.2016.07.012. 77. West JA, Davis CP, Sunwoo H, Simon MD, Sadreyev RI, Wang PI, et al. The long noncoding RNAs NEAT1 and MALAT1 bind active chromatin sites. Mol Cell. 2014;55:791–802. https://doi.org/10.1016/j.molcel.2014.07.012.

Journal

Molecular NeurodegenerationSpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off