C9orf72 ALS/FTD dipeptide repeat protein levels are reduced by small molecules that inhibit PKA or enhance protein degradation
2021; Springer Nature; Volume: 41; Issue: 1 Linguagem: Inglês
10.15252/embj.2020105026
ISSN1460-2075
AutoresNausicaa Valentina Licata, Riccardo Cristofani, Sally Salomonsson, Katherine Wilson, Liam Kempthorne, Deniz Vaizoglu, Vito D’Agostino, Daniele Pollini, Rosa Loffredo, Michael Pancher, Valentina Adami, Paola Bellosta, Antonia Ratti, Gabriella Viero, Alessandro Quattrone, Adrian M. Isaacs, Angelo Poletti, Alessandro Provenzani,
Tópico(s)Cholinesterase and Neurodegenerative Diseases
ResumoArticle18 November 2021Open Access Source DataTransparent process C9orf72 ALS/FTD dipeptide repeat protein levels are reduced by small molecules that inhibit PKA or enhance protein degradation Nausicaa V Licata Nausicaa V Licata orcid.org/0000-0003-0750-0692 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy These authors contributed equally to this work Search for more papers by this author Riccardo Cristofani Riccardo Cristofani orcid.org/0000-0003-2719-846X Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano, Milan, Italy These authors contributed equally to this work Search for more papers by this author Sally Salomonsson Sally Salomonsson orcid.org/0000-0001-6717-3369 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Katherine M Wilson Katherine M Wilson Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Liam Kempthorne Liam Kempthorne orcid.org/0000-0002-9790-8968 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Deniz Vaizoglu Deniz Vaizoglu orcid.org/0000-0001-6572-1544 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Vito G D'Agostino Vito G D'Agostino orcid.org/0000-0003-3379-2254 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Daniele Pollini Daniele Pollini orcid.org/0000-0001-7782-7960 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Rosa Loffredo Rosa Loffredo orcid.org/0000-0001-7981-9227 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Michael Pancher Michael Pancher orcid.org/0000-0002-3783-6069 HTS Core Facility, Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Valentina Adami Valentina Adami orcid.org/0000-0002-0617-9393 HTS Core Facility, Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Paola Bellosta Paola Bellosta orcid.org/0000-0003-1913-5661 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Department of Medicine, NYU at Grossman School of Medicine, NY, USA Search for more papers by this author Antonia Ratti Antonia Ratti orcid.org/0000-0002-4264-6614 Department of Neurology, Stroke Unit and Laboratory of Neuroscience, Istituto Auxologico Italiano, IRCCS, Milan, Italy Dipartimento di Biotecnologie Mediche e Medicina Traslazionale, Università degli Studi di Milano, Milan, Italy Search for more papers by this author Gabriella Viero Gabriella Viero orcid.org/0000-0002-6755-285X Institute of Biophysics, CNR Unit at Trento, Trento, Italy Search for more papers by this author Alessandro Quattrone Alessandro Quattrone orcid.org/0000-0003-3333-7630 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Adrian M Isaacs Adrian M Isaacs orcid.org/0000-0002-6820-5534 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Angelo Poletti Corresponding Author Angelo Poletti [email protected] orcid.org/0000-0002-8883-0468 Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano, Milan, Italy Search for more papers by this author Alessandro Provenzani Corresponding Author Alessandro Provenzani [email protected] orcid.org/0000-0003-1652-3415 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Nausicaa V Licata Nausicaa V Licata orcid.org/0000-0003-0750-0692 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy These authors contributed equally to this work Search for more papers by this author Riccardo Cristofani Riccardo Cristofani orcid.org/0000-0003-2719-846X Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano, Milan, Italy These authors contributed equally to this work Search for more papers by this author Sally Salomonsson Sally Salomonsson orcid.org/0000-0001-6717-3369 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Katherine M Wilson Katherine M Wilson Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Liam Kempthorne Liam Kempthorne orcid.org/0000-0002-9790-8968 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Deniz Vaizoglu Deniz Vaizoglu orcid.org/0000-0001-6572-1544 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Vito G D'Agostino Vito G D'Agostino orcid.org/0000-0003-3379-2254 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Daniele Pollini Daniele Pollini orcid.org/0000-0001-7782-7960 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Rosa Loffredo Rosa Loffredo orcid.org/0000-0001-7981-9227 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Michael Pancher Michael Pancher orcid.org/0000-0002-3783-6069 HTS Core Facility, Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Valentina Adami Valentina Adami orcid.org/0000-0002-0617-9393 HTS Core Facility, Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Paola Bellosta Paola Bellosta orcid.org/0000-0003-1913-5661 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Department of Medicine, NYU at Grossman School of Medicine, NY, USA Search for more papers by this author Antonia Ratti Antonia Ratti orcid.org/0000-0002-4264-6614 Department of Neurology, Stroke Unit and Laboratory of Neuroscience, Istituto Auxologico Italiano, IRCCS, Milan, Italy Dipartimento di Biotecnologie Mediche e Medicina Traslazionale, Università degli Studi di Milano, Milan, Italy Search for more papers by this author Gabriella Viero Gabriella Viero orcid.org/0000-0002-6755-285X Institute of Biophysics, CNR Unit at Trento, Trento, Italy Search for more papers by this author Alessandro Quattrone Alessandro Quattrone orcid.org/0000-0003-3333-7630 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Adrian M Isaacs Adrian M Isaacs orcid.org/0000-0002-6820-5534 Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK Search for more papers by this author Angelo Poletti Corresponding Author Angelo Poletti [email protected] orcid.org/0000-0002-8883-0468 Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano, Milan, Italy Search for more papers by this author Alessandro Provenzani Corresponding Author Alessandro Provenzani [email protected] orcid.org/0000-0003-1652-3415 Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy Search for more papers by this author Author Information Nausicaa V Licata1, Riccardo Cristofani2, Sally Salomonsson3,4, Katherine M Wilson3,4, Liam Kempthorne3,4, Deniz Vaizoglu3,4, Vito G D'Agostino1, Daniele Pollini1, Rosa Loffredo1, Michael Pancher5, Valentina Adami5, Paola Bellosta1,6, Antonia Ratti7,8, Gabriella Viero9, Alessandro Quattrone1, Adrian M Isaacs3,4, Angelo Poletti *,2 and Alessandro Provenzani *,1 1Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy 2Dipartimento di Scienze Farmacologiche e Biomolecolari, Università degli Studi di Milano, Milan, Italy 3Department of Neurodegenerative Disease, UCL Queen Square Institute of Neurology, London, UK 4UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, London, UK 5HTS Core Facility, Department of Cellular, Computational and Integrative Biology, University of Trento, Trento, Italy 6Department of Medicine, NYU at Grossman School of Medicine, NY, USA 7Department of Neurology, Stroke Unit and Laboratory of Neuroscience, Istituto Auxologico Italiano, IRCCS, Milan, Italy 8Dipartimento di Biotecnologie Mediche e Medicina Traslazionale, Università degli Studi di Milano, Milan, Italy 9Institute of Biophysics, CNR Unit at Trento, Trento, Italy *Corresponding author. Tel: +39 02 50318215; E-mail: [email protected] *Corresponding author. Tel: +39 0461 283176; E-mail: [email protected] The EMBO Journal (2022)41:e105026https://doi.org/10.15252/embj.2020105026 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract Intronic GGGGCC (G4C2) hexanucleotide repeat expansion within the human C9orf72 gene represents the most common cause of familial forms of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) (C9ALS/FTD). Repeat-associated non-AUG (RAN) translation of repeat-containing C9orf72 RNA results in the production of neurotoxic dipeptide-repeat proteins (DPRs). Here, we developed a high-throughput drug screen for the identification of positive and negative modulators of DPR levels. We found that HSP90 inhibitor geldanamycin and aldosterone antagonist spironolactone reduced DPR levels by promoting protein degradation via the proteasome and autophagy pathways respectively. Surprisingly, cAMP-elevating compounds boosting protein kinase A (PKA) activity increased DPR levels. Inhibition of PKA activity, by both pharmacological and genetic approaches, reduced DPR levels in cells and rescued pathological phenotypes in a Drosophila model of C9ALS/FTD. Moreover, knockdown of PKA-catalytic subunits correlated with reduced translation efficiency of DPRs, while the PKA inhibitor H89 reduced endogenous DPR levels in C9ALS/FTD patient-derived iPSC motor neurons. Together, our results suggest new and druggable pathways modulating DPR levels in C9ALS/FTD. Synopsis GGGGCC repeat expansion in the C9orf72 gene causes amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) in part via neurotoxic dipeptide-repeat proteins (DPRs) produced by repeat-associated non-AUG translation. Here, a high-throughput drug screen identifies novel positive and negative small-molecule modulators of DPR levels. cAMP-elevating compounds increase DPR levels by enhancing protein kinase A (PKA) activity. PKA inhibition reduces DPR levels in cells and in C9ALS/FTD patient-derived iPSC motor neurons. Inhibition or knockdown of PKA ameliorates motility and survival in a Drosophila model of C9orf72 ALS/FTD. Knockdown of PKA catalytic subunits correlates with reduced DPR translation efficiency. HSP90 inhibitor geldanamycin and aldosterone antagonist spironolactone reduce DPR levels by promoting protein degradation. Introduction Repeat-associated non-AUG (RAN) translation is an unconventional translation mechanism associated with several nucleotide-repeat expansion disorders. The hexanucleotide repeat expansion GGGGCCn, also known as (G4C2)n, is localized in the first intron of the C9orf72 gene (DeJesus-Hernandez et al, 2011; Renton et al, 2011) and it is the most common genetic cause of familial forms of ALS and FTD (hereafter C9ALS/FTD) (Gijselinck et al, 2012). The pathogenic mechanisms proposed for C9ALS/FTD suggest that sense (G4C2)n- and antisense (C4G2)n-containing transcripts cause two different mechanisms of toxicity. The first is mediated by the formation of RNA foci that bind and sequester RNA-binding proteins, thereby leading to impairment of RNA metabolism (Donnelly et al, 2013; Lee et al, 2013; Mori et al, 2013b; Xu et al, 2013; Cooper-Knock et al, 2014; Haeusler et al, 2014; Wen et al, 2014; Zhang et al, 2015; Conlon et al, 2016; Swinnen et al, 2018); and the second mediated by their unconventional RAN translation into five different dipeptide-repeat proteins (DPRs: poly-GA, poly-GP, poly-GR, poly-PA, poly-PR) (Ash et al, 2013; Mori et al, 2013a, 2013c). In addition, pathological expansions of (G4C2)n reduce C9orf72 transcription and translation with decreased C9orf72 protein levels (DeJesus-Hernandez et al, 2011; Renton et al, 2011); this latter event can also be associated with endosomal trafficking, autophagy dysfunction, which synergizes with repeat-associated toxicity (Shi et al, 2018; Boivin et al, 2020; Zhu et al, 2020). DPR-induced toxicity has been shown in several cell lines, in iPSC-derived neurons (May et al, 2014; Su et al, 2014; Yamakawa et al, 2015; Yang et al, 2015a; Westergard et al, 2019), in Drosophila (Mizielinska et al, 2014; Wen et al, 2014; Chew et al, 2015; Freibaum et al, 2015; Yang et al, 2015a; Boeynaems et al, 2016) and in mouse models (Zhang et al, 2016, Zhang et al, 2018, Zhang et al, 2019; Schludi et al, 2017; Choi et al, 2019; Hao et al, 2019). Multiple studies demonstrated proteasome dysfunction due to the sequestration of proteasomal proteins by poly-GA in both in vitro (May et al, 2014; Yamakawa et al, 2015) and in vivo models (Zhang et al, 2016). RAN translation of (G4C2)n-RNAs has been recently shown to require a near-cognate start codon upstream of the repeat in the +1 frame (Green et al, 2017; Tabet et al, 2018) and to be triggered by stress conditions in a cap-dependent (Kearse et al, 2016; Green et al, 2017; Tab et et al, 2018) or cap-independent way (Cheng et al, 2018; Sonobe et al, 2018). However, the mechanisms regulating RAN translation have not yet been completely elucidated. Antisense oligonucleotides (ASOs) (Jiang et al, 2016; Gendron et al, 2017) and small molecules targeting the (G4C2)n (Su et al, 2014; Simone et al, 2018; Wang et al, 2019) and/or r(CGG)n RNAs (Yang et al, 2015b, 2016; Green et al, 2019) have been proposed as possible therapeutic approaches, but no clinically approved drugs are known to selectively modulate RAN translation. The small molecules available at present alter RNA secondary structures providing a proof of principle as to how their binding to (G4C2)n can inhibit both RNA foci formation and RAN translation. A genetic screen recently identified the ribosomal protein RPS25 as a regulator of RAN translation of different repeat sequence expansions (Yamada et al, 2019). Other modifiers of DPR production were identified by two independent genome-wide CRISPR-Cas9 screens (Kramer et al, 2018; Cheng et al, 2019; Wilson et al, 2019). In addition, the RNA helicase DHX36 has been shown to favour (G4C2)n and FMR1-associated RAN translation (Tseng et al, 2021), whereas the RNA helicase DDX3X inhibits (G4C2)n RAN translation (Cheng et al, 2019) but it promotes FMR1-associated RAN translation (Linsalata et al, 2019). Here, we used a chemical genomic approach to identify small molecules and relative molecular targets. These small molecules modulate DPR levels by either increasing protein clearance or inhibiting translation of (G4C2)n-containing RNAs. Among these small molecules, we found that Geldanamycin (GELD, an inhibitor of Heat Shock Protein 90, HSP90) increases proteasome activity and that Spironolactone (SPL, an aldosterone antagonist) modulates DPR autophagy degradation. Moreover, we found for the first time that cAMP-elevating compounds increase DPR levels by boosting protein kinase A (PKA) activity, while PKA silencing, or inhibition reverted these effects. This suggests a novel mechanism in which PKA is involved in pathways that aberrantly enhance the translation of C9orf72 (G4C2)n mRNA to neurotoxic DPRs. Results Development of a HTS assay for identifying modulators of C9-DPR levels We set up a cell-based high-throughput screen (HTS) (full HTS protocol in Appendix Materials and Methods), to find small molecules capable of modulating DPR levels. In the HTS we used an artificial reporter containing 58 G4C2 repeats outside of the native C9orf72 sequence, and with the GFP sequence in the GP frame (Freibaum et al, 2015) (hereafter polyGP-GFP) (Appendix Tables S1 and S2). We obtained a consistent signal of polyGP-GFP across experimental repeats using a reverse transfection approach in HEK293T cells (Fig EV1A and B). In the HTS we also co-transfected an AUG-RFP plasmid (Fig 1A) to report on AUG-mediated translation. In the screen, Cycloheximide (CHX) was used to model general translation inhibition. We used the variation in the number of GFP- or RFP-positive cells as the read-out of the assay. Due to the lack of a positive control, small molecule RAN translation inhibitor, the variability and the robustness of the assay were optimized to perform a HTS based on the effect of CHX on RFP-expressing cells (Z′-factor = 0.5) (Zhang et al, 1999) (Fig EV1C). We observed that CHX did not decrease the fluorescent intensity or the number of cells expressing polyGP-GFP (Fig EV1D), consistent with a recent report showing non-AUG translation to be resistant to elongation inhibitors (Kearse et al, 2019). Click here to expand this figure. Figure EV1. Set up of a High-Throughput Screen (HTS) for identifying modulators of DPR levels SH-SY5Y and HEK293T cells were co-transfected with AUG-RFP and polyGP-GFP plasmids by standard (SH-SY5Y and HEK293Ta) or reverse (HEK293Tb) transfection method. Images were acquired 24 h after using Operetta High-Content Imaging System and transfection efficiency were calculated on the ratio of cells expressing RFP/tot and cells expressing GFP/tot. Data are mean ± SD from three biological replicates. Images in HEK293Tb are from DMSO in Fig 2A. Scale bars 200 μm. Immunoblot for testing the level of polyGP-GFP and AUG-RFP in lysates from HEK293T cells co-transfected with both plasmids or with polyGP-GFP or AUG-RFP and mock. Distribution of cells expressing AUG-RFP (upper inset) or AUG-RFP fluorescent intensity (lower inset) in negative (DMSO) and positive (CHX, 5 μM) controls. Data used to calculate Z′-factor. Data are mean ± SD from 45 (upper inset) and 35 (lower inset) technical replicates. Distribution of cells expressing polyGP-GFP (upper inset) and polyGP-GFP fluorescent intensity (lower inset) in negative (DMSO) and positive (CHX, 5 μM) controls. Data are mean ± SD from 45 (upper inset) and 35 (lower inset) technical replicates. HTS. HEK293T cells were co-transfected with the constructs shown in Fig 1A. Scatter plot shows the distribution of negative (DMSO, orange dots) and positive (CHX 5 μM, purple dots) controls added 3 h after reverse co-transfection. On the Y-axis reported the Z-score values of cells expressing AUG-RFP and on the X-axis the Z-score values of cells expressing polyGP-GFP. Grid lines represent the thresholds arbitrarily set up around the DMSO distribution (polyGP-GFP ± 1.5 on the X-axis and ± 1.5 for AUG-RFP on the Y-axis). Source data are available online for this figure. Download figure Download PowerPoint Figure 1. Primary and confirmatory screening for identifying modulators of C9orf72-derived DPR levels A. Schematic representation of the constructs utilized in (B) and (C). The first construct contains 58 (G4C2) repeats outside of the native C9orf72 sequence and GFP in the GP frame (polyGP-GFP). The start codon of GFP was removed. The second construct AUG-RFP is used as a positive control of canonical translation. B. HTS. HEK293T cells were co-transfected with the constructs in (A) and, negative (DMSO, orange dots) and positive (CHX 5 µM, purple dots) controls added 3 h after reverse co-transfection with compound-libraries (5 µM). Images and data acquisition collected after about 30 h of treatment. Scatter plot shows the distribution of compounds. On the Y-axis reported the Z-score values of cells expressing AUG-RFP and on the X-axis the Z-score values of cells expressing polyGP-GFP. Grid lines represent the thresholds arbitrarily set up around DMSO distribution (polyGP-GFP ± 1.5 on the X-axis and ± 1.5 for AUG-RFP on the Y-axis) to select compounds for the confirmatory screening and eliminate the ones without effect (orange square). C. Confirmatory screening performed as described above. Schematic distribution of compounds based on Z-score values of cells expressing polyGP-GFP and AUG-RFP (above) and on Z-score values of the fluorescent intensity of the two reporters per each compound (below). Data are from four technical replicates, boxes are the Z-score mean value and whiskers represent ± SD. Baseline of Z-score = 0 indicates that Z-score is identical to the mean score. Asterisk (*) represents the selected compounds: 1 Forskolin (FSK), 2 Erysolin (ERY), 3 Geldanamycin (GELD), 4 Helenin (HLN) and 5 Spironolactone (SPL). Number sign (#) represents cellular stress inducers: #1 Thapsigargin and #2 Tunicamycin. D, E. Dose-response analysis of ERY, HLN, SPL, FSK and GELD. Cells were co-transfected with AUG-RFP and polyGP-GFP plasmids and treated with two concentration ranges 0.5, 1, 5 and 10 µM (D) and 20, 40 and 60 µM (E) for 24 h. CHX used only low dosages. Data are mean ± SD from three biological replicates. One-way ANOVA followed by Dunnett's multiple comparison tests: *P < 0.5; **P < 0.01; ***P < 0.001. Download figure Download PowerPoint We screened approximately 2,500 compounds with biological activity from different chemical libraries (see Materials and Methods). The compounds were added 3 h after plasmid reverse co-transfection. GFP and RFP reporter signals were measured approximately 30 h later. Plotting the Z-score of the number of cells expressing AUG-products (RFP, Y-axis) versus the Z-score of the number of cells expressing DPR-products (polyGP-GFP, X-axis), we obtained a graphical representation of the simultaneous effect of the small molecules on both AUG and RAN translation-dependent products (Fig 1B). The majority of tested compounds did not modify the levels of the reporters. Their signals overlapped with the distribution of negative controls (Figs 1B and EV1E), indirectly proving the assay quality. We selected effective compounds using an arbitrary threshold of ± 1.5 Z-score for cells expressing polyGP-GFP and ± 1.5 for AUG-derived positive cells. These thresholds gave a significant separation of DMSO treated from CHX-treated samples (see Appendix for full details on thresholding calculations). We excluded highly toxic compounds using a threshold Z-score nuclei ≤ −2, indicating that < 50% of cells survived. A confirmatory screen was next performed as described above but increased the number of replicates from one to four. As expected, while only a few compounds were able to reduce levels of the RAN products, many others had the opposite effect (Fig 1C). This comes as no surprise because many RAN-increasing compounds were cellular stressors (Thapsigargin, (Green et al, 2017; Westergard et al, 2019) and Tunicamycin (Green et al, 2017; Westergard et al, 2019)) present in the chemical library (Fig 1C). We selected three small molecules according to their capability to specifically reduce or increase the number of cells expressing polyGP-GFP and/or the fluorescent intensity of GFP (Fig 1C, Table 1 and Dataset EV1). GELD and SPL reduced RAN products, whilst cAMP-elevating compounds, with Forskolin (FSK) being the most potent one, increased them. Interestingly, FSK, which activates adenylyl cyclase (AC) and enhances intracellular cAMP levels, triggers a multitude of PKA-dependent and/or -independent pathways resulting in pleiotropic effects on cells. These events include the activation of many intracellular signalling cascades and of the cAMP Response Elements Binding (CREB) family of transcription factors that, upon phosphorylation, regulate the expression of genes containing cAMP Response Elements (CREs) in their promoters (Seamon et al, 1981; Sapio et al, 2014; Kanne et al, 2015). We also identified two phytochemicals with undefined mechanisms of action, Erysolin (ERY) and Helenin (HLN) that effectively reduced RAN products. These results were then validated in the confirmatory screen (Dataset EV1). Table 1. List of the small molecules selected from the confirmatory screening. Small molecules Cells expressing polyGP-GFP Cells expressing AUG-RFP polyGP-GFP intensity AUG-RFP intensity Number of cells DMSO −0.06 −0.2 0.3 −0.12 −0.2 Erysolin (ERY) −3.5 −0.2 −1.34 −0.7 −1.1 Forskolin (FSK) 20.3 −2.4 27 −0.8 0.05 Geldanamycin (GELD) −5 1.6 0.002 3 −2.6 Helenin (HLN) −1.05 −0.7 −2.8 −1.1 −1.3 Spironolactone (SPL) −0.5 0.7 −2.3 0.6 −0.9 Mean of compounds from four biological replicates. All the compounds were added after 3h of reverse co-transfection and used at the final concentration of 5 µM for approximately 30 h. Mean of DMSO from 64 biological replicates (32 replicates per each 384-well plate). Data are reported as Z-score values. The entire list of small molecules used is reported in Dataset EV1. To obtain a confidence interval of safe utilization of the selected compounds, we then performed dose–response experiments by treating cells with two concentration ranges for each compound (Fig 1D and E), simultaneously checking their toxicity and the effects on DPR levels. All the compounds confirmed their activity in modulating the number of polyGP-GFP-positive cells, although to various extents. The most potent compound was FSK that selectively increased polyGP-GFP-positive cells compared to AUG-RFP-positive cells. HLN decreased both products, whereas ERY, GELD and SPL decreased more efficiently the polyGP-GFP products than the AUG-RFP. All these compounds were moderately toxic at concentrations higher than 40 μM, with HLN the most toxic. Therefore, we excluded HLN due to its toxicity and proceeded with the other four molecules to gain information about their molecular mechanism of action. GELD, SPL and FSK modulate DPR levels independently of the near-cognate CUG start codon To understand whether the four selected DPR modulators (Fig 2A) affect general transcription and translation, we used the incorporation of the modified nucleoside 5‑ethynyl uridine (EU) to evaluate general RNA transcription. In parallel, we took advantage of O-propargyl-puromycin (OPP) incorporation assay to evaluate de novo protein synthesis. GELD marginally induced general transcription (Fig 2B). None of the compounds modulated translation (Fig 2C). The molecular mechanism of RAN translation initiation (Kearse et al, 2016) is still a matter of debate and a near-cognate CUG start codon within C9orf72 first intron 1A has been suggested to play a key role in (G4C2)n RAN translation (Green et al, 2017; Tab et et al, 2018). The polyGP-GFP reporter used in the HTS did not contain the native sequence upstream of the repeat. Therefore, to evaluate whether the effect of these compounds was CUG independent, we used the (G4C2)x66 construct (hereafter 66R) that contains repeats within the native C9orf72 sequence and a specific C-term tag for each frame (Gendron et al, 2013). To prove the efficacy of GELD and SPL in motor neuron-like cells, we used NSC34 cells (Appendix Fig S1A). To ensure the proper evaluation of the total amounts of DPRs produced in cells, we quantified both the soluble DPR levels and the PBS-insoluble DPR aggregate fraction by immunoblot analysis and filter retardation assay, respectively. GELD and SPL significantly reduced the accumulation of DPRs, while FSK significantly increased poly-GA in HEK293T (Fig 2D), poly-GP and its PBS-insoluble fraction in NSC34 cells (Fig 2E and F), despite the presence of the upstream CUG codon. In contrast, ERY did not show any effect in modulating DPR levels in either cell line (Fig 2D–F). We further challenged the selected compounds using C9RAN NLuc reporters (in the GA frame) with the native (CUG) or mutated (CCC) start codon (Green et al, 2017). C
Referência(s)