Over the last decade, Extracellular Vesicles (EVs), including Exosomes, emerged as an important vector of intercellular communication. EVs have been proposed to transfer cargoes such as lipids, nucleic acids and proteins from donor to acceptor tissues or cells. EVs have been associated with several physiological functions and diseases. Today, EV biology is an intense and high impact research topic, which promises to revolutionize translational research through the development of EV-mimetics designed for targeted delivery of therapeutics. Tremendous progresses have been made to better understand the mechanisms that regulate EV secretion. However, our knowledge of EV uptake and content delivery within the acceptor cells is still very limited. In this project, we propose a combination of cell biology experiments coupled with in vitro studies and structural replicate electron microscopy to study how EVs fuse with their target membranes. In a first aim, we will capitalize on our previously published cell-based assays and imaging methods (optical-and electron-microscopy) to 1) further characterize EV uptake and content delivery 2) identify new proteins involved in the fusion process through a candidate approach, 3) determine basic parameters such as cargo size/ type that condition EV delivery through fusion. We will test several proteins candidates that are suspected to control EV uptake and delivery. We will focus our work on a family of proteins name hEnvs, derived from ancestral viral envelop proteins that have been integrated in human genome, and also IFITM1 and 3 proteins. We suspect that hEnv might be involved in pH-dependent EV delivery whereas IFITM1 and 3 would inhibit EV-content delivery. The major actors of endocytosis pathways (clathrin, dynamin, caveolin) will also be tested. Our bulk assays will be complemented with morphological analysis (fluorescent microcopy and classical electron microscopy), to analyze the distribution of EV markers on cells In a second aim, we plan to develop novel fluorescent-based assays to image EV membrane fusion in real time using fluorescent microscopy and correlative EM to capture all the intermediates. Our main goal is to formally establish that membrane fusion is the mechanism responsible for EV content delivery. This mechanism has never been proved so far. Our work will provide an important breakthrough in understanding and proving this delivery content. We will adapt a cell-free assay that used acceptor Plasma Membrane (PM) sheets in suspension coupled with single particle tracking diffusion. Using fluorescent microscopy, we will follow in real-time and on single EV the diffusion of EV, the docking and the fusion reactions. In addition, the same in vitro assay will be used to directly visualize the fusion process at the ultrastructural level. We propose to image the samples through platinum-replica electron microscopy (PREM). EVs will be loaded on the top of the deposited PM sheets and fixed. Initially we expect to dissect the fusion reaction and capture all the fusion-intermediates (including the fusion pore itself) by rapidly fixing the samples at different time points under different conditions Note that all candidates mentioned in Aim1 can be quickly tested within this cell-free imaging assay to directly demonstrate if cell phenotypes observed in Aim 1 correspond to perturbation of the fusion reaction. In other term we will establish causality and not just correlation.

Clathrin-mediated endocytosis is the best characterized process for the entry of proteins and lipids at the plasma membrane of eukaryotic cells. It is also the Achilles’ heel for human diseases such as neurodegenerative diseases and congenital myopathies. While neuronal axons form small clathrin-coated pits for fast endocytosis at synapses, muscle cells form large flat clathrin-coated plaques for cell adhesion at costameres, a cornerstone of mechanotransduction in muscle through their role as physical and functional links between the extracellular matrix and the cytoskeleton. We have demonstrated that the formation of actin networks capable of scaffolding intermediate filaments around clathrin plaques requires dynamin 2 (DNM2), another member of the endocytosis machinery crucial for the formation and release of clathrin-coated vesicles. Mutations in DNM2, amphiphysin 2 (BIN1) and myotubularin (MTM1) lead to centronuclear myopathy (CNM) and we showed that clathrin plaques are altered in CNM cells and mouse models. One common denominator in all three forms is a defective actin and cortical cytoskeleton leading to disorganized costameres, suggesting a defect of mechanotransduction as a common pathomechanism in CNMs. Our preliminary results strongly indicate that this alteration is associated with defective YAP/TAZ mechanosensitive pathway leading to altered transcription of their target genes. The first objective of CLASS is to decipher the function of clathrin plaques in YAP/TAZ mechanotransduction and to highlight involvement of this function in the pathomechanisms of centronuclear myopathy. We will (1) analyse clathrin plaque response to mechanical cues, (2) dissect mechanotransduction signaling in CNM and (3) reveal the benefit of releasing YAP/TAZ nuclear sequestration on CNM myotubes. While our results point towards a specialized function for clathrin plaques at muscle costameres, one fundamental question remains: Why do some cells exclusively form coated pits while others form plaques? We recently achieved the proof of concept that formation and plasticity of CCS are controlled by tissue-specific alternative splicing events through investigation of clathrin heavy chain's exon 31, which is aberrantly spliced in myotonic dystrophy (DM1), a "spliceopathy" affecting in particular muscle and central nervous system (CNS) functions. In DM1, functional loss of RNA binding proteins MBNL1 and CELF1 disturbs the developmentally regulated splicing program resulting in aberrant expression of embryonic isoforms in adult tissues. The second objective is to reveal the mechanisms by which muscle cells form plaques using alternative splicing isoforms of the endocytosis machinery and provide novel insight to DM1 pathophysiology. We will (4) reveal the effect of skipping key exons of developmentally regulated splicing variants of endocytosis proteins in neurons and myotubes, (5) decipher the effect of mis-regulated splicing of the endocytosis machinery in DM1 cells and (6) force exon skipping of developmentally regulated splicing variants in mice CLASS offers a genuinely innovative opportunity to push beyond the actual limitations concerning the fundamental biology of endocytosis: We will unravel how alternative splicing can modulate clathrin polymerization and induce formation of mechano-signaling platforms. This will be a major conceptual turning point in the endocytosis field by proving with unambiguous visual evidence that the clathrin machinery is genetically regulated to produce functional diversity. Our technical capabilities and leading expertise in the clathrin field, CNM and DM1 pathophysiology, our unique expertise in producing genetically modified hiPSC clones and differentiating them in neurons and muscle cells, and our powerful expertise on RNA splicing, put us in a unique position to undertake the dissection of genetic determinants in clathrin structural plasticity in healthy and diseased states.

Intracellular organization and transport is fundamental for the function of eukaryotic cells. Dynamins are a prototype class of mechano-enzymes and key regulators of membrane fission. These large GTPases regulate the scission of nascent vesicles from the plasma membrane during endocytosis; they also participate in the formation and trafficking of vesicles from intracellular compartments, and fusion processes. Moreover, dynamins bind to both microtubules and actin and potentially regulate these cytoskeletons. While tremendous efforts have been made to understand dynamin-dependent membrane fission, the cellular function of dynamins in specific tissues and their physiological role remain poorly understood, especially for the ubiquitous member of the family, dynamin 2. Describing the physiological role of dynamin 2 (DNM2) is of particular importance as DNM2 mutations cause two unrelated neuromuscular disorders for which no treatments are available and pathomechanisms are largely unknown: dominant Charcot-Marie-Tooth neuropathies and dominant centronuclear myopathy. Centronuclear myopathies (CNM) are rare neuromuscular disorders characterized by muscle weakness and intracellular disorganization of muscle fibres. A crucial role for DNM2 in muscle homeostasis was supported by our studies in mice, however data from cell culture showed these mutations impact on only some membrane trafficking routes controlled by dynamins, suggesting that specific functions of DNM2 are important in normal and diseased muscle, and that additional functions remain to be discovered in a tissue-specific context. Moreover, we have shown DNM2 represents a therapeutic target for several forms of CNM. Marc Bitoun (Myology Institute, Paris) and Jocelyn Laporte (IGBMC, Illkirch)’s teams have been involved for many years in the identification of the functional consequences of mutations in DNM2 and in the development of animal models for different forms of centronuclear myopathies. It is now crucial to define the physiological role of DNM2 in muscle, and this will be instrumental to understand the pathomechanisms linked to DNM2 mutations and develop future therapies for patients. We will thus synergize to decipher the roles of dynamin 2 in skeletal muscle under normal and pathological conditions, and will tackle the following relevant questions: - what is the basis for the muscle–specific function of the ubiquitously expressed DNM2 ? (Task 1). We will identify and validate tissue-specific isoforms and binding partners. - what is the impact of DNM2 alteration on its muscle-specific functions ? (Task 2). We will study the impact of patient mutations on localization and binding partners, and characterize novel mouse models for DNM2-loss in muscle to decipher its physiological importance. - what are the cellular pathways controlled by DNM2 in muscle ? (Task 3). We will especially investigate costamere formation and maintenance, mitochondria and triads that are key structures for muscle metabolism and contraction, and autophagy flux. This will also lead to a better understanding of CNM pathomechanisms. - what are the best pre-clinical approaches to rescue DNM2-related myopathy ? (Task 4). Based on preliminary data and a better understanding of the role of DNM2 in muscle, we will test genetic modulation, allele specific inhibition and autophagy modulators to improve and revert the CNM phenotypes. By sharing tools and expertise available in our 2 teams, we expect to achieve important steps toward preclinical development of therapies for DNM2-related CNM. The two partners bring together a set of unique and complementary knowledge, expertise, models and tools, and have access to patient samples for validation. Overall, this multidisciplinary project aims to detail for the first time the poorly studied physiological roles of dynamin in muscle under both normal and disease conditions, to ultimately provide novel and innovative therapeutic strategies leading to clinical trials.

Centronuclear myopathies (CNM) are rare congenital myopathies characterized by abnormal centralization of nuclei in muscle fibers. Three main forms of CNM are distinguished: The X-linked recessive CNM (also called X-linked myotubular myopathy, XLMTM) due to mutations in myotubularin (MTM1), the autosomal recessive CNM (AR-CNM) due to mutations in amphiphysin 2 (BIN1), and the autosomal dominant CNM (AD-CNM) due to mutations in dynamin 2 (DNM2). There is no therapies available for patients affected by CNMs. The teams of Marc Bitoun (UMRS974, Myology Institute, Paris) and Belinda Cowling/Jocelyn Laporte (IGBMC, Illkirch) have been involved for many years in the identification and the functional consequences of mutations in CNMs, in the development of animal models and therapeutic strategies for the CNMs. During the last years, common efforts of the two teams supported by the “DynaMuscle” ANR program led to essential breakthrough for the development of innovative therapeutic strategies based on modulation of the DNM2 expression in two forms of CNM. First, proof of concept for allele-specific silencing therapy was achieved in AD-CNM by targeting the most frequent DNM2 mutation in the Knock-in-Dnm2R465W mouse model (KI-Dnm2) and in patient-derived cells. Second, cross-therapy by reduction of DNM2 was established as efficient therapeutic strategy for XLMTM in which DNM2 is over-expressed. DynaTher aims at “converting the try” by pursuing preclinical developments for these two “DNM2 therapies” by tackling the following questions: - What preclinical developments are required for allele-specific therapy for the p.R645W DNM2 mutation in AD-CNM? We will investigate the long-term maintenance of efficacy and benefit of systemic treatment in young KI-Dnm2 mice. We will also optimize the therapeutic benefit in old mice. - Can allele-specific silencing be extended to other AD-CNM DNM2 mutations and is it possible to develop “pan-mutations” tools to avoid a mutation-based strategy? We will screen for allele-specific siRNA able to silence 5 other DNM2 mutations in patient-derived cells. We will also develop “pan-mutations” allele-specific-siRNA against the two versions of frequent heterozygous non-pathological single nucleotide polymorphisms present in the DNM2 mRNA. - Is the cross-therapy strategy also efficient for other forms of CNM, thus increasing the number of patients that can be treated by this approach? Based on our previous proof of principle for XLMTM, we will extend this therapy in mouse models of AR-CNM and AD-CNM. - Do siRNAs against DNM2 have “off-target effects”? That will be answer by associating transcriptome and proteome analyses in human cells from patients affected by the three main forms of CNM allowing also to uncover common and specific altered pathways in CNMs and readouts for future clinical trials. - Can DNM2 therapy be improved by targeting specific DNM2 isoform or by other delivery methods and be extended to other neuromuscular disorders? Therapeutic benefit of reduction of the DNM2 muscle-specific isoform will be determined in Knock-out-Mtm1 and KI-Dnm2 mouse models. Non-viral delivery methods will be also developed. Finally, we will identify novel applications for DNM2 therapy through a screening for elevated DNM2 expression in a panel of neuromuscular disorders. The ambition of DynaTher is to accelerate the preclinical development of these approaches to ultimately provide the most efficient and safe molecular tools targeting DNM2 required for the first gene therapy clinical trial for autosomal CNMs.

Duchenne muscular dystrophy (DMD) is one of the most severe pediatric degenerative myopathies. DMD muscle is constantly exposed to cycles of degeneration and regeneration; over time, regenerative potential is exhausted and necrosis prevails. As of today, the cellular and molecular determinants responsible for this functional exhaustion remain largely uncharacterized. Recently, I have integrated single-cell proteomics and transcriptomics to obtain a cellular atlas of skeletal muscle. In Model_DMD I now propose to capitalize on this expertise to elucidate the determinants interfering with regeneration in the dystrophic muscle. The overall goal of this proposal is to gain critical missing knowledge on the evolution and crosstalk between the different cell populations during DMD and use it to identify therapeutic targets to improve regeneration providing an immediate impact on the quality life of patients.