Exploring the role of astrocytes in the pathophysiology of multiple sclerosis

Universidade do Minho Escola de Medicina Sofia Pereira das Neves junho de 2020 Exploring the role of astrocytes in the pathophysiology of multiple sclerosis Sofia Pereira das Neves Exploring the role of astrocytes in the pathophysiology of multiple sclerosis UMinho|2020 Sofia Pereira das Neves junho de 2020 Exploring the role of astrocytes in the pathophysiology of multiple sclerosis Trabalho efetuado sob a orientação da Doutora Fernanda Marques e da Doutora Rita Teodoro Tese de Doutoramento Doutoramento em Envelhecimento e Doenças Crónicas Universidade do Minho Escola de Medicina ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ iii Agradecimentos “Look up… and climb the steps for what you want to achieve” (Charlie Chaplin) Começo por agradecer a orientação, o apoio e ajuda que a minha orientadora Fernanda me deu durante a realização deste trabalho. Agradeço também o apoio da minha orientadora Rita. De seguida, este trabalho não poderia ter sido realizado sem o apoio incondicional das pessoas mais importantes na minha vida – Ricardo, Carla, mãe e pai. Agradeço também toda a ajuda laboratorial do Joe DeYoung, Fuying Gao, Giovanni Coppola, Ana Falcão, Cláudia Nóbrega, Carlos Bessa, João Cerqueira, Liliana Santos, Goreti Pinto, Eduarda Correia, Cláudia Pereira, Joana Palha, Nuno Sousa, Margarida Correia-Neves, João Sousa, Cláudia Miranda, Patrício Costa, João Oliveira, Gisela Santos, Catarina Barros, Diana Pereira, Ricardo Ferreira, Cristina Mota, Susana Monteiro, Adelaide Fernandes, toda a equipa do biotério. Aos meus (ex-)companheiros de bancada: Sandro, Vítor, Inês, Marta, Joana, e em especial, Catarina, aprendi muito convosco, o vosso apoio foi muito importante e vou guardar-vos sempre no meu coração. Aos meus companheiros e amigos para a vida Margarida, Madalena e Eduardo. À Sónia Gomes, que para além de colega de bancada, foi uma parceira de grandes aventuras. Aos grandes amigos que levo do ICVS: Teresa, Dinis, Bárbara, Sónia Borges, Fátima, Liliana Amorim, Gabriela, Francisca, Belina. Aos colegas do PhDOC, e seus orientadores, pelas discussões científicas em todos os retiros. Em especial à Raquel, à Rita e ao Marco pelo companheirismo durante estes anos. A todos os NeRDs em geral pela ajuda e discussão científica, e em especial à Sara Silva pelo sorriso constante e disponibilidade para com os outros. This work was supported by Foundation for Science and Technology (FCT) and COMPETE through the project EXPL/NEU-OSD/2196/2013 and by The Clinical Academic Center (2CA-Braga) through the project EXPL/001/2016. The work at ICVS/3B’s has been developed under the scope of the project NORTE-01-0145-FEDER-000013, supported by the Northern Portugal Regional Operational Programme (NORTE 2020), under the Portugal 2020 Partnership Agreement, through the European Regional Development Fund (FEDER), and funded by FEDER funds through the Competitiveness Factors Operational Programme (COMPETE), and by National funds, through the Foundation for Science and Technology (FCT), under the scope of the project POCI-01-0145-FEDER-007038. FM is an assistant researcher and recipient of an FCT Investigator grant with the reference CEECIND/01084/2017. SN is a recipient of a Ph.D. fellowship with the reference PD/BD/114120/2015 from MCTES national funds. iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. v Exploring the role of astrocytes in the pathophysiology of multiple sclerosis Abstract The presence of reactive astrocytes in multiple sclerosis (MS), a chronic inflammatory disease of the central nervous system, has been described for a long time, however its contribution for disease pathophysiology is still not fully understood. Therefore, in this work we proposed to further explore the contribution of astrocytes in MS using different experimental approaches. We started by performing a temporal transcriptomic analysis of astrocytes in the chronic experimental autoimmune encephalomyelitis (EAE) MS animal model. For that we isolated astrocytes from the cerebellum in 3 disease time points: before the symptom’s appearance (pre-symptomatic phase), at the onset/peak of the disease and at the chronic phase of EAE. Particularly at the onset phase of disease, astrocytes overexpressed genes associated with a neurotoxic phenotype (known as A1 phenotype), along with the overexpression of genes involved in metabolic pathways, like glycolysis and tricarboxylic acid cycle. This suggested that astrocytes could be undergoing metabolic reprogramming, similarly to what happens in macrophages/microglia in response to different inflammatory stimuli. On the second part of this work, we induced EAE in a model that presents “silent” astrocytes, i.e. astrocytes that present minimal calcium elevations in the soma and main processes, due to the ablation of the inositol 1,4,5-triphosphate receptor type 2 (IP3R2-null). IP3R2null mice and wild-type littermates presented similar disease clinical scores, however IP3R2-null mice had decreased lesion burden in the cerebellum white matter, which could be associated with the increased astrocyte reactivity observed in these mice near the lesions, compared to the normal appearing white matter. Next, we studied the effects of a neuroprotective drug, dimethyl fumarate (DMF), on astrocyte activation and on the cognitive function in the EAE model. DMF treatment, started at the symptomatic phase of disease, was able to reduce astrocyte reactivity, evaluated by a decrease in the number of cells, and demyelination in the fimbria, which is the most important white matter pathway in the hippocampus, possibly contributing to the prevention of cognitive deficits observed in DMF-treated EAE animals. Finally, we focused our attention on a protein produced by astrocytes during EAE, lipocalin-2 (LCN2), an acute phase protein that was previously suggested as a possible disease biomarker. After the quantification of LCN2 concentration in cerebrospinal fluid samples from MS patients, we observed that increased LCN2 levels were associated with faster disease progression. However, this association was no longer statistically significant after controlling for the patients’ age and the presence of oligoclonal bands. Keywords: astrocytes; dimethyl fumarate; lipocalin-2; metabolic reprogramming; multiple sclerosis. vi Investigação do papel dos astrócitos na patofisiologia da esclerose múltipla Resumo A presença de astrócitos reativos na esclerose múltipla (EM), uma doença inflamatória crónica do sistema nervoso central, já é conhecida há muito tempo, porém a sua contribuição para a patofisiologia da doença não está totalmente esclarecida. Assim, neste trabalho propomos explorar a contribuição dos astrócitos na EM usando diferentes abordagens experimentais. Começámos por fazer uma análise temporal do transcriptoma dos astrócitos no modelo animal crónico de EM de encefalomielite autoimune experimental (EAE). Para isso, isolámos astrócitos do cerebelo em 3 momentos experimentais: antes do aparecimento de sintomas (fase pré-sintomática), no pico da doença e na fase crónica. Em particular no pico da doença, os astrócitos sobre-expressaram genes associados a um fenótipo neurotóxico (conhecido como fenótipo A1), e genes envolvidos em diferentes vias metabólicas, tais como a glicólise e o ciclo dos ácidos tricarboxílicos. Estas alterações sugeriram que os astrócitos poderiam estar a passar por um processo de reprogramação metabólica, semelhante ao que acontece com macrófagos/microglia em resposta a diferentes estímulos inflamatórios. Na segunda parte deste trabalho, induzimos EAE num modelo de astrócitos “silenciados”, i.e. astrócitos que apresentam elevações de cálcio mínimas no soma e processos principais, devido à deleção do recetor tipo 2 do inositol 1,4,5-trifosfato (IP3R2-null). Os animais IP3R2-null e do tipo selvagem das mesmas ninhadas desenvolveram a doença de modo semelhante, no entanto os animais IP3R2-null apresentaram uma diminuição da carga de lesão na substância branca do cerebelo, que poderá estar associada ao aumento da reatividade dos astrócitos observada perto destas regiões de lesão, comparativamente à substância branca aparentemente normal. De seguida, estudámos os efeitos de um fármaco neuroprotetor, o dimetil fumarato (DMF), na ativação dos astrócitos e na função cognitiva no modelo de EAE. O tratamento com DMF, iniciado na fase sintomática da doença, reduziu a reatividade dos astrócitos, avaliada através de uma diminuição do seu número, e a desmielinização na fimbria, que é o feixe de substância branca mais importante no hipocampo, o que poderá ter contribuído para a prevenção de défices cognitivos nos animais EAE tratados com DMF. Por último, focámos a nossa atenção numa proteína produzida pelos astrócitos em resposta à EAE, a lipocalina-2 (LCN2), uma proteína de fase aguda que foi anteriormente sugerida como um possível marcador de doença. Após a quantificação dos níveis de LCN2 no líquido cefalorraquidiano de doentes com EM, observámos que níveis elevados de LCN2 estavam associados a uma progressão mais rápida da doença. Porém, esta associação perdeu a significância estatística após controlar para a idade dos doentes e para a presença de bandas oligoclonais. Palavras-chave: astrócitos; dimetil fumarato; esclerose múltipla; lipocalina-2; reprogramação metabólica. vii Table of contents Abstract............................................................................................................................................... v Resumo.............................................................................................................................................. vi Abbreviations list ................................................................................................................................ ix Figures list......................................................................................................................................... xii Tables list .......................................................................................................................................... xiv Thesis layout ..................................................................................................................................... xvi CHAPTER 1 ........................................................................................................................................ 1 General introduction ....................................................................................................................... 1 1. Multiple sclerosis........................................................................................................................ 2 1.1. Pathophysiology of MS ............................................................................................................ 3 1.2. Genetic and environmental susceptibility factors ...................................................................... 6 1.3. Pathological hallmarks of disease ............................................................................................ 8 1.4. MS symptoms ......................................................................................................................... 9 1.4.1. Cognitive deficits in MS patients ..................................................................................... 11 1.5. MS diagnosis ........................................................................................................................ 12 1.6. MS biomarkers ...................................................................................................................... 13 1.6.1. Lipocalin-2 ..................................................................................................................... 22 1.7. Current therapeutic approaches ............................................................................................ 24 1.7.1. Dimethyl fumarate ......................................................................................................... 26 2. Astrocytes in the pathophysiology of MS ................................................................................... 27 2.1. Synthesis of metabolic substrates .......................................................................................... 30 2.2. Formation and maintenance of BBB integrity ......................................................................... 32 2.3. Synapse maintenance and refinement ................................................................................... 34 2.4. Synthesis of immunomodulatory molecules and growth factors .............................................. 34 2.5. Interaction between astrocytes and T cells ............................................................................. 36 xiv Tables list Chapter 1 Table 1. 1 – 2017 McDonald criteria for diagnosis of MS. Table 1. 2 – Biological function or pathway correspondent to differentially expressed genes in MS or EAE samples. Table 1. 3 – Therapeutic agents approved in the European Union for the treatment of MS. Table 1. 4 – Identification of astrocytic sub-populations by single cell analysis of samples from EAE animals or MS patients. Chapter 2 Supplementary table 2. 1 – Primers sequence and annealing temperature. Supplementary table 2. 2 – List of genes specifically associated with CNS cells (astrocytes, Bergmann glia, microglia, oligodendrocytes and neurons), endothelial cells and different immune cell populations (B cells, T cells and granulocytes). Supplementary table 2. 3 – Pathways significantly altered in the comparison between EAE and noninduced animals at the pre-symptomatic phase. Supplementary table 2. 4 – Pathways significantly altered in the comparison between EAE and noninduced animals at the onset phase. Supplementary table 2. 5 – Pathways significantly altered in the comparison between EAE and noninduced animals at the chronic phase. Supplementary table 2. 6 – Pathways significantly altered in the comparison between EAE animals at the onset and pre-symptomatic phases, after normalization for the respective non-induced group. Supplementary table 2. 7 – Pathways significantly altered in the comparison between EAE animals at the chronic and pre-symptomatic phases, after normalization for the respective non-induced group. Supplementary table 2. 8 – Pathways significantly altered in the comparison between EAE animals at the chronic and onset phases, after normalization for the respective non-induced group. Chapter 4 Supplementary Table 1 – Mean ± SEM of data compared using the parametric two-tailed one-way ANOVA with Tukey’s multiple comparison post-hoc test. Supplementary Table 2 – Median ± IQR of data compared using the non-parametric two-tailed KruskalWallis with Dunn’s multiple comparison post-hoc test. xv Chapter 5 Table 5. 1 – Sample demographic information. Table 5. 2 – Cox regression analysis of time for progression until EDSS 3. Annex 1 Table S1. 1 – Summary of effects observed in astrocytes of different animal models after treatment with therapeutic agents. xvi Thesis layout CHAPTER 1 comprises a general introduction that covers general aspects of multiple sclerosis (MS). An overview of the main functions of astrocytes both in physiological and pathological conditions is also performed. CHAPTER 2 includes the transcriptomic and morphological study performed in astrocytes from EAE animals, at different disease time points. In this work, we observed that astrocytes overexpressed genes associated with a neurotoxic phenotype (A1 phenotype) and also genes involved in metabolic pathways, including glycolysis and tricarboxylic acid (TCA) cycle, particularly in the onset/peak phase of disease. Also, astrocytes near lesion regions, in EAE animals, presented increased length and complexity, compared to astrocytes from the white matter of non-induced animals. CHAPTER 3 presents the results obtained after EAE induction in a model of impaired global astrocytic calcium signaling, the type 2 inositol 1,4,5-triphosphate receptor (IP3R2)-null mice. Even tough no differences were found between genotypes regarding clinical score, the IP3R2-null animals presented a decreased lesion burden in the cerebellum white matter, at the onset phase of disease. Moreover, cerebellum’ astrocytes from IP3R2-null mice, but not wild-type animals, presented increased length and complexity near lesion sites compared to the normal appearing white matter, which could be associated with an increased ability to limit immune cell spreading in the IP3R2-null mice. CHAPTER 4 comprises the results obtained after the symptomatic treatment of EAE animals with dimethyl fumarate (DMF). DMF treatment was able to reduce astrocytic activation in the fimbria of the animals and revert the cognitive deficits observed in vehicle-treated animals, which was accompanied by a decrease in demyelination. Lipocalin-2 (LCN2) is highly expressed by astrocytes in both MS and EAE and it was suggested as a possible disease biomarker. In CHAPTER 5 we performed a follow-up study to explore the prognostic value of LCN2 in MS patients, and found that higher cerebrospinal levels of this protein were associated with a faster disease progression. However, this association was no longer significant after controlling for the age of the patients and presence of oligoclonal bands. Nevertheless, and considering the confidence intervals, we suggest that LCN2 could be useful for disease prognosis if used in a wider biomarker panel. CHAPTER 6 presents a general discussion of the results and future experiments that would help to further explore the findings obtained in this work. CHAPTER 1 General introduction 2 1. Multiple sclerosis Multiple sclerosis (MS) clinical and pathological characteristics were first described more than 100 years ago by Charcot, Carswell, Cruveilhier and others (Lucchinetti et al., 2005; Noseworthy et al., 2000). Nowadays, it is defined as a chronic immune-mediated disorder of the central nervous system (CNS), that occurs in genetically susceptible people, associated with an environmental stimulus, however its etiopathology remains largely unknown (Compston and Coles, 2008; Nicholas and Rashid, 2013; Noseworthy et al., 2000; Nylander and Hafler, 2012). Approximately 80-85% of MS patients initially present with a relapsing-remitting (RR) disease course, characterized by the appearance of clinical symptoms followed by a recovery period, spontaneous or in response to treatment. After a variable period of time irreversible damage accumulates and patients begin to experience a steady state of deterioration in neurologic functions, independently of any acute attack. This stage of disease is termed secondary progressive MS (SPMS) (Faissner et al., 2019; Goodin, 2014; Lassmann, 2014; Noseworthy et al., 2000; Nylander and Hafler, 2012). The remaining 15-20% of MS patients develop a progressive clinical course from the beginning of disease [primary progressive MS (PPMS)], experiencing neurologic decline without ever presenting acute clinical attacks (Faissner et al., 2019; Goodin, 2014; Noseworthy et al., 2000; Nylander and Hafler, 2012). MS is the most common cause of neurologic disability in young adults (Duffy et al., 2014; Nicholas and Rashid, 2013; Stys et al., 2012), with the appearance of symptoms occurring between 15 and 45 years of age, in the majority of cases (Goodin, 2014; Nicholas and Rashid, 2013). RRMS presents a female:male ratio of approximately 2:1, while PPMS presents similar incidence among men and women (Correale et al., 2017; Goodin, 2014; Noseworthy et al., 2000). The world prevalence of MS varies considerably, however there is a strong correlation between prevalence and distance from the equator (Figure 1. 1), which is believed to be associated with sunlight exposure and vitamin D levels (Azary et al., 2018; Kearns et al., 2018; Reich et al., 2018). In the district of Braga in Portugal, the incidence of MS was reported to be 2.74/100.000 inhabitants and the prevalence 39.82/100.000 inhabitants (Figure 1. 1) (Figueiredo et al., 2015). 3 Figure 1. 1 – Epidemiology of multiple sclerosis. Estimated MS prevalence worldwide and in the district of Braga. Data obtained or adapted from (Figueiredo et al., 2015; Goodin, 2014). 1.1. Pathophysiology of MS There are two main hypothesis that try to explain the pathophysiology of MS. On one hand, MS is believed to be caused by the activation of peripheral autoreactive T cells that migrate into the CNS and initiate the disease process (“outside-in” hypothesis) (Baecher-Allan et al., 2018). On the other hand, the “insideout” hypothesis proposes that MS occurs within the CNS, as a primary neurodegenerative disorder. Namely, cytodegeneration, possibly focused on the oligodendrocyte-myelin complex, can be the initial event, leading to the release of highly antigenic constituents and, consequently, to an inflammatory and autoimmune response, in a predisposed host (Duffy et al., 2014; Stys et al., 2012). In this case, what distinguishes MS from other progressive neurodegenerative disorders is the propensity of the hosts’ immune cells to react against the highly autoantigenic components, released after cytodegeneration (Stys et al., 2012). In contrast, the “outside-in” hypothesis suggests that mechanisms of molecular mimicry, meaning crossreactivity between self and non-self antigens, are responsible for the development of the autoimmune response in MS patients (Nylander and Hafler, 2012; Rojas et al., 2018). Of interest, potential crossreactivity between myelin epitopes with microbial antigens has been demonstrated (Wucherpfennig et al., 1997). Nevertheless, the antigen specificity of the immune response in MS is unresolved (Compston and Coles, 2008), but the general consensus is that several pathological antigens are involved in the disease (Nylander and Hafler, 2012). The search for candidate autoantigens has concentrated on myelin components, since demyelination is a hallmark of MS lesions. In fact, myelin-specific antibodies and reactive cells have been found in MS patients (Berger et al., 2003; Cruz et al., 1987; Navikas et al., 1995; 4 O'Connor et al., 2005). However, a recent study has identified a possible non-myelin target antigen, the guanosine diphosphate (GDP)-L-fucose synthase (Planas et al., 2018). Interestingly, human and bacterial GDP-L-fucose synthase peptides present high similarity. Moreover, a significant association was found between the recognition of GDP-L-fucose synthase peptides and myelin basic protein (MBP) peptides, however it remains unclear if there is cross-recognition between these two peptides, whether the recognition of one of the peptides facilitates the exposure of the other one and the de novo activation of autoreactive T cells against this last one, or if the two responses occur simultaneously (Planas et al., 2018). Likewise, post-translational modifications, such as citrullination (namely MBP citrullination), glycosylation and glycation, have been suggested to have an impact on the generation of autoimmune antigens (Opdenakker et al., 2016; Stys et al., 2012). The MS pathological process includes increased BBB permeability, multifocal inflammation, demyelination, oligodendrocyte death, reactive gliosis, axonal degeneration and neuronal death (BaecherAllan et al., 2018). Of relevance, immune cell migration to the CNS can occur through the brain barriers. There are two main barriers that protect the brain: the well-known blood-brain barrier (BBB), which is formed by the presence of tight junctions between the brain endothelial cells and by the astrocytic endfeet, and the blood-cerebrospinal fluid (CSF) barrier which is composed by the choroid plexus (CP) epithelial cells. The CP is the brain structure responsible for CSF production, and the presence of tight junctions between CP epithelial cells form the blood-CSF barrier. This barrier prevents leukocyte entry into the CSF, even tough they can easily pass the fenestrated blood vessels of the CP parenchyma (Lopes Pinheiro et al., 2016). However, in MS, this barrier was found to be compromised by the specific loss of epithelial tight junction proteins, like claudin-3 (Kooij et al., 2014). Moreover, CP epithelial cells may express high levels of adhesion molecules and chemokines. After crossing this barrier, leukocytes now have access to the CSF, where they interact with resident dendritic cells (epiplexus cells), triggering the release of large amounts of inflammatory mediators, like pro-inflammatory cytokines. These mediators are believed to induce the expression of adhesion molecules and chemokines in brain endothelial cells. In turn, this will induce a massive entry of immune cells into the brain parenchyma through the BBB (Figure 1. 2) (Lopes Pinheiro et al., 2016). The migration process across the endothelial cells of the BBB is better understood than the one across CP epithelial cells. The first point of contact occurs between lymphocyte selectin ligands (e.g. P selectin glycoprotein ligand-1) and endothelial cell selectins (e.g. Eand P-selectin). These interactions allow the tethering and rolling of lymphocytes along the vascular wall, where they will encounter chemokines, upregulated by endothelial cells during neuroinflammation. The binding to these chemokines deliver a 5 signal to very late antigen-4 (VLA-4) and leukocyte function-associated antigen-1 (LFA-1) integrins, on the lymphocyte surface, that induces their conformational change and clustering, increasing their affinity for the endothelial ligands intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule1 (VCAM-1) (Lopes Pinheiro et al., 2016). During the inflammatory response, tumor necrosis factor alpha (TNFα) and interferon gamma (IFNγ) promote VCAM-1 expression by endothelial cells (Steinman, 2001). Interestingly, reactive astrocytes can also express VCAM-1 (Gimenez et al., 2004; Lee and Benveniste, 1999; Ponath et al., 2018), and this expression is important for lymphocytes to effectively enter the CNS parenchyma (Gimenez et al., 2004). Lymphocytes are then arrested on the vessel wall, and may begin to polarize and crawl, migrating to the brain parenchyma by diapedesis. Lymphocyte diapedesis can occur in two different ways: migration across adjacent endothelial cells (paracellular route) or migration through a single endothelial cell (transcellular route), in which case a channel or pore has to be formed (Lopes Pinheiro et al., 2016). After leukocytes have extravasated, they do not enter directly in the CNS parenchyma, but are first trapped in perivascular spaces formed by the parenchyma basement membrane, that surrounds blood vessels, and by retracted astrocytic end-feet. Passage beyond this barrier into the parenchyma requires active recruitment and degradation of extracellular matrix (ECM) proteins (Sofroniew, 2015; Steinman, 2001). Both astrocytes and immune cells are sources of matrix metalloproteinases (MMPs; family of enzymes involved in the degradation of the ECM), which are important for this migration (Gerwien et al., 2016; Miljkovic et al., 2011; Song et al., 2015). Once inside the CNS, myelin-specific T cells are reactivated through an interaction with antigen-presenting cells (APCs), in the presence of the antigen and adequate co-stimulation (Chastain et al., 2011; Lucchinetti et al., 2005; Noseworthy et al., 2000). Microglial cells, infiltrating macrophages, B cells and astrocytes may function as APCs (Lucchinetti et al., 2005). Then, T cell start producing cytokines that induce the activation of macrophages, microglia and astrocytes, that then produce nitric oxide (NO), osteopontin, oxygen free radicals, among others (Ortiz et al., 2014; Steinman, 2001). B cells will start producing not only antibodies against proteins and lipids of the myelin sheath, but also pro-inflammatory and regulatory cytokines (Kasper and Shoemaker, 2010; Steinman, 2001). The combined effect of antibodies, complement, NO and pro-inflammatory cytokines induces myelin phagocytosis by macrophages, leading to demyelination and axonal and neuronal damage (Figure 1. 2) (Ortiz et al., 2014; Steinman, 2001). Importantly, B cells can also play a protective role in the CNS, via downregulation of inflammation and opsonization of myelin debris, facilitating their clearance by phagocytosis (Duffy et al., 2014). 6 Even tough MS is viewed as a T cell-mediated disease, it has become increasingly clear that several other innate and adaptive immune cells, as well as glial cells, like microglia and astrocytes, are also relevant in the disease context (Duffy et al., 2014). Microglial activation occurs not only in lesions, but also in the normal appearing white matter (NAWM) and gray matter. Moreover, these cells are able to produce reactive oxygen and nitrogen species (ROS and RNS, respectively), that induce direct neuronal damage, via mitochondrial dysfunction. However, microglia is also involved in debris phagocytosis and clearance, and can also produce growth factors (Correale et al., 2017; Reich et al., 2018). Astrocytes proliferate and become activated within demyelinating lesions, suggesting these CNS cells can also play critical roles in oligodendrocyte injury and axonal degeneration (Correale et al., 2017). The main physiological and pathological roles of astrocytes will be further discussed in a following section. Figure 1. 2 – Immune cell entry to the CNS during MS. Immune cells enter the CNS of MS patients after the blood-CSF and blood-brain barriers have been compromised. Images obtained and adapted from Servier Medical Art and (Lopes Pinheiro et al., 2016). CNS – central nervous system; CP – choroid plexus; ICAM-1 – intercellular adhesion molecule 1; LFA-1 – lymphocyte functionassociated antigen 1; MHC – major histocompatibility complex; RNS – reactive nitrogen species; ROS – reactive oxygen species; TCR – T cell receptor. 1.2. Genetic and environmental susceptibility factors As mentioned before, MS can arise in environmentally and genetically susceptible individuals. Regarding the genetic susceptibility, human leukocyte antigen (HLA) genes exert the largest contribution for MS susceptibility, however it is not fully clear how it affects the risk of developing MS (Alcina et al., 2012). 7 The HLA system is a cluster of gene complex, encoding the major histocompatibility complex (MHC) proteins in humans, which are involved in antigen presentation. Disease associated MHC alleles may confer susceptibility by interfering with the T cell repertoire selection in the thymus and peripheral immune system and/or by presenting self-peptides against which an autoimmune response is developed (Smith et al., 1998). The specific allele that presents the higher genetic risk for MS patients is the HLADRB1*1501 (increased risk, odds ratio ≈ 3) (Baecher-Allan et al., 2018; Goodin, 2014; Hafler et al., 2005; Kearns et al., 2018; Paul et al., 2019; Reich et al., 2018), which explains between 14-50% of the genetic risk (Hafler et al., 2005). Nevertheless, only a very small fraction of HLA-DRB1*1501 carriers (<5%) are susceptible to develop MS (Goodin, 2014). Genome wide association studies have revealed other polymorphisms associated with higher risk for developing MS, most of them related to immune pathway genes, although most of them showed only a modest effect. In this regard, polymorphisms in the interleukin-2 receptor subunit alpha and interleukin-7 receptor genes seem to be the most commonly associated with MS, after MHC (Baecher-Allan et al., 2018; Nylander and Hafler, 2012; Paul et al., 2019). Epidemiologic data has identified several infectious and non-infectious factors associated with increased risk to develop MS. Potential environmental triggers include trauma, stress, vaccinations, typhoid, smallpox, chickenpox, Epstein-Barr virus (EBV) and human herpesvirus 6 infection, tobacco, vitamin deficiencies, low sunlight exposure, living with domesticated animals, occupational hazards, dietary habits and toxic exposure (Celarain and Tomas-Roig, 2020; Goodin, 2014; Reich et al., 2018). Alterations in gut microbiota composition have also been identified not only in MS patients compared to healthy controls, but also among RRMS patients during the remission or active states of disease (Chen et al., 2016), and were shown to possibly play a role in T helper (Th) 17 cell differentiation (Cosorich et al., 2017). Interestingly, recent publications from the Quintana lab have demonstrated that different environmental factors, like commensal bacteria metabolites and environmental chemicals (e.g. herbicide linuron), are able to modulate CNS-resident cells, contributing to disease pathogenesis in a MS animal model, the experimental autoimmune encephalomyelitis (EAE) model (Rothhammer et al., 2018; Rothhammer et al., 2016; Wheeler et al., 2019). Furthermore, epigenetic alterations have been described in MS patients. The hypermethylation of Ecadherin and ICAM-1 in MS patients may contribute to increased BBB permeability and leukocyte invasion, respectively (Celarain and Tomas-Roig, 2020), contributing to the CNS invasion by peripheral immune cells. 14 therapeutic response in MS patients (Harris et al., 2017; Paul et al., 2019; Ziemssen et al., 2019). The identification of disease biomarkers is important for all these steps. A biomarker is a characteristic that can be objectively measured and assessed, and that can be used as an indicator of normal biological processes, pathological processes or pharmacologic response to therapies (Comabella and Montalban, 2014; Lycke and Zetterberg, 2017; Paul et al., 2019; Ziemssen et al., 2019). Ideally, this characteristic should be present in patients with a specific disease and absent in other patients or healthy people, or vice-versa. Sample collection for biomarker analysis should be performed using a safe procedure for the patient, and its detection should be performed accurately and reproducibly, while also being fast, simple and cost-effective to ensure wide implementation (Paul et al., 2019; Ziemssen et al., 2019). Hence, the detection methods used should be reliable enough to not be influenced by factors like sample collection, sample processing and sample storage (Ziemssen et al., 2019). In addition, good biomarkers should correlate with the disease biology or pathogenesis, such as inflammatory activity, degree of neurodegeneration, demyelination or remyelination (Paul et al., 2019). It is also important to take into consideration that the usage of a biomarker in clinical practice should include multiple validations using independent cohorts of patient (Paul et al., 2019). Currently, MRI acquisitions provide important information regarding the size, number, age and development of CNS lesions, and play an important role in MS diagnosis and therapy monitoring. Namely, in the case of CIS patients, those that present an abnormal MRI have a long term risk of progression to clinically definite MS of 60-82%, compared to a 8-25% risk in CIS patients with a normal MRI (Brownlee and Miller, 2014). The usage of gadolinium (Gd) as a contrast agent in MRI is useful to detect acute lesions characterized by BBB disruption (Lycke and Zetterberg, 2017). In the recent 2017 revisions of the McDonald criteria, dissemination in time can be demonstrated in MRI by the simultaneous presence of Gd-enhancing and non-enhancing lesions at a given time point, or by the presence of a new T2hyperintense or Gd-enhancing lesion on a follow-up MRI, having as baseline a scan taken previously. Dissemination in space can be demonstrated by the presence of one or more T2-hyperintense lesions characteristic of MS in at least two of the following areas of the CNS: periventricular, cortical or juxtacortical and infratentorial brain regions, and spinal cord (Thompson et al., 2018). Importantly, Gdenhancing lesions are rarely identified in progressive MS, suggesting a reduced breakdown of the BBB in these patients (Baecher-Allan et al., 2018; Lassmann, 2018). As for molecular biomarkers, the existence of oligoclonal bands (OCBs) in the patients’ CSF, but not in the serum, is an indicator of intrathecal antibody synthesis and is present in more than 87% of MS patients (Dobson et al., 2013; Ziemssen et al., 2019). OCBs are defined as at least two immunoglobulin (Ig) bands 15 present in the CSF, with no corresponding band present in the serum (Wahed, 2019). Agarose gel electrophoresis with isoelectric focusing, followed by immunoblotting or immunofixation for IgG, is the most sensitive approach, at present, to demonstrate the presence of CSF-specific OCBs. Importantly, CSF and serum samples always have to be processed together, to confirm CSF specificity (Thompson et al., 2018). Even tough OCBs are not specific of MS, and can be found in several other disorders, such as systemic lupus erythematosus and neurosyphilis, their presence presents a strong prognostic value for conversion from CIS to clinically definite MS (Deisenhammer et al., 2019; Harris et al., 2017; Reich et al., 2018; Ziemssen et al., 2019). The IgG index can also be used as a marker of intrathecal production of Igs (Ziemssen et al., 2019), but this method is less reliable (Thompson et al., 2018). Besides the presence of OCBs, other biomarkers have also been suggested to present prognostic value for CIS conversion to clinically definite MS. High chitinase-3-like-1 (CHI3L1) concentration in the CSF is an independent risk factor for this conversion (Harris et al., 2017; Lycke and Zetterberg, 2017; Ziemssen et al., 2019), and is also associated with a faster disease progression in MS patients (Martinez et al., 2015). CHI3L1 is a glycosidase secreted by monocytes, microglia and activated astrocytes, whose physiological role in the CNS in unknown (Ziemssen et al., 2019). Moreover, the presence of antibodies against measles, rubella and varicella-zoster in the CSF was found to be significantly more frequent in CIS patients that converted to clinically definite MS (Brettschneider et al., 2009). Maveskar and co-workers (2019) have recently reported the overexpression of two clusters of proteins in MS patients compared to controls, one enriched in astrocytes and the other in microglia (astrocyte cluster 8: MMP7, Serpin Family A Member 3, granzyme A and chloride intracellular channel 1; microglia cluster 2: desmoglein-2 and TNF receptor superfamily member 25). Moreover, these clusters were significantly correlated with disease severity scales (Masvekar et al., 2019). Considering MS is an inflammatory disease, the cytokine and chemokine profile of these patients has also been studied. Compared to controls, MS patients presented increased serum levels of interleukin (IL)-6, IL-12, IL-17, IL-23, IL-4 and IFNγ (Kallaur et al., 2013; Li et al., 2017), and CSF levels of IL-6, IL4, IL-12, chemokine C-C motif ligand (CCL) 2, CCL5 and chemokine C-X-C motif ligand (CXCL) 13 (Alvarez et al., 2013; Bartosik-Psujek and Stelmasiak, 2005; Deisenhammer et al., 2019; Domingues et al., 2017; Malmestrom et al., 2006). Moreover, increased CSF CXCL13 levels were found in CIS patients converting to clinically definite MS compared to non-converters (Lycke and Zetterberg, 2017; Paul et al., 2019). On the other hand, Tejera-Alhambra and colleagues (2015) observed a decreased serum concentration of CCL11, CCL2 and CCL5 in RRMS, compared with progressive MS patients and healthy controls (TejeraAlhambra et al., 2015). Complement system activation has also been demonstrated in MS, having 16 increased complement component 1q (C1q) and complement component 3 (C3) levels been found in the CSF of MS patients (Hakansson et al., 2020; Lindblom et al., 2016). In addition, CSF C3a presented prognostic value, since patients who developed new T2 MRI lesions during follow-up presented significantly higher levels of this protein at baseline (Hakansson et al., 2020). However, increased CSF cytokine levels are also found in other neuroinflammatory disorders (Komori et al., 2015; Lepennetier et al., 2019; Lycke and Zetterberg, 2017), so their quantification alone does not present good diagnostic or prognostic value in MS, but their inclusion in a broader biomarkers panel could be useful. Biomarkers associated with neurodegeneration, that reflect different pathological processes like axonal or glial degeneration, astrogliosis and oxidative stress, have also been measured in the MS context (Lycke and Zetterberg, 2017). Namely, MS patients presented increased CSF levels of neurofilament light chain (NF-L), compared to controls, and its levels were found to be a good prognostic biomarker for CIS conversion to clinically definite MS, and RRMS conversion to SPMS (Hakansson et al., 2017; 2018; Harris et al., 2017; Lycke and Zetterberg, 2017; Martinez et al., 2015; Paul et al., 2019; Ziemssen et al., 2019). NFs are the major structural components of the axonal and dendritic cytoskeleton, and during axonal injury they are released into the extracellular space (Harris et al., 2017). In addition, total Tau protein levels were increased in MS patients compared to controls (Bartosik-Psujek et al., 2011). Nevertheless, these biomarkers are not diagnosis-specific, since higher levels of NF-L and total Tau are a general marker of axonal injury, and are present in other neurological disorders associated with this process (Deisenhammer et al., 2019). As for markers associated with glial degeneration, studies have found increased CSF glial acidic fibrillary protein (GFAP) levels, an intermediate filament of the astrocytic cytoskeleton, in MS patients compared to controls (Domingues et al., 2017; Lycke and Zetterberg, 2017), and high levels of this protein were associated with earlier disability progression and more severe disability (Comabella and Montalban, 2014; Martinez et al., 2015; Petzold et al., 2002). Recently, GFAP levels were found to be increased in the serum of SPMS patients, compared to controls and RRMS, and the levels of this protein were positively correlated with the serum NF-L levels, even tough no significant differences were found between groups regarding NF-L levels. Moreover, elevated serum GFAP and NF-L levels were associated with increased disability, evaluated by higher EDSS score, and disease duration (Hogel et al., 2020). Similarly, the CSF and serum levels of S100 calcium-binding protein β (S100β), a commonly used astrocytic marker (Wang and Bordey, 2008), were found increased in MS patients (Barateiro et al., 2016; Bartosik-Psujek et al., 2011; Petzold et al., 2002; Rejdak et al., 2007). 17 Other biomarkers could be useful for differential diagnosis, since they are indicative of other demyelinating diseases. This is the case of anti-aquaporin 4 (AQP4) antibodies, that are detectable in approximately 75% of neuromyelitis optica (NMO) patients but not in MS patients (Ziemssen et al., 2019). Additional biomarkers have been used to monitor therapy response. Namely, the presence of neutralizing antibodies against IFNβ and natalizumab is used to identify patients non-responders to treatment (Ziemssen et al., 2019). The presence of John Cunningham virus antibodies is also monitored in patients treated with Natalizumab, since the reactivation of this virus could cause progressive multifocal leukoencephalopathy (PML) (Paul et al., 2019). In conclusion, a single biomarker will most likely not be effective in the diagnosis of such a complex and heterogeneous disease like MS, and future efforts must be done to integrate different types of data, including clinical, radiological and biological, into predictive models for diagnosis, prognosis and treatment response (Comabella and Montalban, 2014; Paul et al., 2019). In an attempt to find novel disease biomarkers, several studies have used samples of MS patients and animal models to identify differentially expressed genes and proteins, along with their biological functions and signaling pathways (Table 1. 2). Table 1. 2 – Biological function or pathway correspondent to differentially expressed genes in MS or EAE samples. Sample Analysis performed Biological function of altered genes/ Altered pathways Reference Peripheral blood of MS patients Microarray Tand B-cell activation; degradation of extracellular matrix; chemokine receptors; apoptosis; humoral immune responses (Ramanathan et al., 2001) MS patients brain lesions Microarray Adhesion/structure/transport; growth factor-related; myelin formation; signaling; cell cycle/homeostasis/unknown; immune-related (Whitney et al., 1999) MS patients brain lesions Microarray Immune response genes; adhesion molecules; macrophage-related genes; complement activation; proinflammatory response; stress related genes; myelin proteins; neuron-specific genes (Lock et al., 2002) MS patients brain lesions Microarray Differences between lesion margin and center: cytokines; growth factors; receptors/surface molecules; signalling molecules; transcription factors; cell cycle proteins (Mycko et al., 2003) Endothelial cells from MS patients Laser capture microdissection and microarray analysis Endothelial cell activation; permeability and vessel tone; cell injury; matrix metalloproteinases (Cunnea et al., 2010) (Continues) 18 (Continues) Table 1. 2 (Continued). Sample Analysis performed Biological function of altered genes/ Altered pathways Reference Astrocytes from the NAWM of MS patients Immunoguided laser capture microdissection RNA processing; immunity and defense; intracellular signaling cascade; protein metabolism and modification; cytoskeleton; neuronal development; response to metal ions; cellular cation homeostasis; response to hypoxia (Waller et al., 2016) MS patients brain lesions Laser capture microdissection and proteomics Active plaque: oxidative phosphorylation; regulation of actin cytoskeleton; IL-4, IL-6, IL-2 signaling pathways; transcriptional regulation by STAT; catenin signaling pathway; calcium signaling; oxidative phosphorylation Chronic-active plaque: focal adhesion; cell communication; ECM-receptor interaction; purine metabolism; inflammation mediated by chemokine and cytokine signaling pathway; integrin signaling pathway; muscarinic acetylcholine receptor 1 and 3 signaling pathway; nicotinic acetylcholine receptor signaling pathway; PI3K signaling pathway; IL-4, IL-6, IFNα/β signaling pathways; transcriptional regulation by STAT; calcium signaling; hepatic fibrosis and hepatic stellate cell activation; urine metabolism; actin cytoskeleton signaling; oxidative phosphorylation Chronic plaque: focal adhesion; regulation of actin cytoskeleton; oxidative phosphorylation; cell communication; integrin signaling pathway; IL-4, HGF, TCRα/β, IL-6 signaling pathways; biosynthesis of steroids; actin cytoskeleton signaling; ubiquinone biosynthesis; axonal guidance signaling; integrin signaling (Han et al., 2008) (Satoh et al., 2009) CSF from RRMS patients Proteomics Cell adhesion; ECM proteins; complement factors; ephrins and other proteins involved in synaptic plasticity; aminoglycan processes; coagulation and inflammation (Kroksveen et al., 2017) Spinal cord of C57BL/6 mice immunized with MOG35-55 Microarray Antigen processing and presentation; immune-related; extracellular matrix; cell adhesion and matrix degradation; signal transduction; transcription; cell structure; movement and secretion; CNS-related; cell division and death (Ibrahim et al., 2001) Spinal cord of C57BL/6 mice immunized with MOG38-50 Microarray Acute phase: injury response factors; inducers of neurite outgrowth and differentiation; ion channels and ion transporters; regulators of synaptic vesicle targeting/fusion; synapse structure/formation Recovery phase: growth factors; regeneration-related proteins; immunoglobulins; myelin genes (Carmody et al., 2002) Spinal cord of Lewis rats immunized with gp-MBP Microarray Ion homeostasis; Ca2+ homeostasis; vesicular function; exocitosis; mitochondrial function; impulse conduction (Nicot et al., 2003) 19 Table 1. 2 (Continued). Sample Analysis performed Biological function of altered genes/ Altered pathways Reference Spinal cord of Dark Agouti rats immunized with MOG1-125 Microarray Ion channel and transporter proteins; regulators of synaptic transmission; myelin genes; cholesterol biosynthesis pathway (Mueller et al., 2008) Spinal cord of SJL/J mice immunized with MOG35-55 Microarray Immune/defense responses; cytokine signaling; antiviral response; extracellular matrix organization; iron metabolism; arginine and proline metabolism. (Gresle et al., 2014) Spinal cord of Lewis rats immunized with recombinant MBP Microarray Immune-related: antigen presentation; Th cell differentiation; CTL-mediated apoptosis of target cells; Fcγ-mediated phagocytosis in macrophages; interferon signaling; dendritic cell maturation; B cell development; TREM signaling; innate/adaptive immune cells; complement system; pathogenesis of MS CNS-related: synaptic long-term potentiation; dopamineDARPP32 in cAMP signaling; ALS signaling; CREB signaling in neurons; endothelin-1 signaling; biosynthesis of steroids; taurine/hypotaurine metabolism (Inglis et al., 2012) Spinal cord of SJL/J mice immunized with PLP and TMEVIDD mice Microarray Downregulated only in EAE: myelination; brain development; axogenesis1 Upregulated only in EAE: chemotaxis; leukocyte migration; cellular response to interleukin-1; T cell differentiation involved in immune response; T helper 1 cell differentiation; adaptive immune response Upregulated only in TMEV-IDD: positive regulation of B cell activation; B cell receptor signaling pathway; complement activation-classical pathway; innate immune response Upregulated in both models: antigen processing and presentation; Myd88-dependent toll-like receptor signaling pathway; microglial cell activation; response to interferon-beta; Toll-like receptors signaling pathways; astrocytes development (DiSano et al., 2019) Brain of C57BL/6 mice immunized with MOG35-55 LncRNA microarray Glutamatergic synapses; calcium signaling pathways; retrograde endocannabinoid signaling; adherens junctions; leukocyte transendothelial migration; salivary secretion; insulin signaling pathway; gastric acid secretion; amphetamine addiction; long-term potentiation; renin-angiotensin system; african trypanosomiasis; malaria; neuroactive ligand-receptor interaction; complement and coagulation cascades; basal cell carcinoma; glutathione metabolism; GABAergic synapse (Liu et al., 2018) (Continues) 20 Table 1. 2 (Continued). Sample Analysis performed Biological function of altered genes/ Altered pathways Reference Spinal cord of Lewis rats immunized with gpMBP Proteomics Energy metabolism; cell growth and/or maintenance; transport; protein metabolism; cell communication and signaling; regulation of nucleobase; nucleoside; nucleotide; and nucleic acid metabolism (Farias et al., 2012) CSF from Lewis rats immunized with gp-MBP Proteomics Class I acute phase proteins; components of the complement system; serine protease inhibitor; antioxidant enzymes; iron metabolism; immunoglobulins; calcium homeostasis (Rosenling et al., 2012) Astrocytes from spinal cord and cerebellum of GFAP-Cre RiboTag mice immunized with MOG35-55 High throughput sequencing Spinal cord: superpathway of cholesterol biosynthesis; cholesterol biosynthesis I; cholesterol biosynthesis II (via 24;25dihydrolanosterol); cholesterol biosynthesis III (via desmosterol); hepatic fibrosis/hepatic stellate cell activation; clathrin-mediated endocytosis signaling; superpathway of geranylgeranyldiphosphate iosynthesis I (via mevalonate); axonal guidance signaling; antigen presentation pathway; IFN signaling; IL-8 signaling Cerebellum: superpathway of cholesterol biosynthesis; cholesterol biosynthesis I; cholesterol biosynthesis II (via 24;25dihydrolanosterol); cholesterol biosynthesis III (via Desmosterol); LXR/RXR activation; antigen presentation pathway; IFN signaling; complement system; epoxysqualene biosynthesis; mevalonate pathway I; superpathway of geranylgeranyldiphosphate biosynthesis I (via mevalonate) (Itoh et al., 2018) Astrocytes from spinal cord of C57BL/6 mice immunized with MOG35-55 FACS and RNAseq Signal transduction; interferome type I interferon responsive genes; reactome interferon signaling; interferon-beta response up; interferon-beta 1 targets; IkB-kinase-NF-κB cascade; positive regulation of cell proliferation; regulation of cellular metabolic processes; cytokine activity; cell migration; chemokine activity; inflammatory response (Rothhammer et al., 2016) Astrocytes form the optic nerve of GFAP-Cre RiboTag mice immunized with MOG35-55 RNAseq Complement system; antigen presentation; Th cell differentiation; Th1 and Th1 activation; OX40 signaling; Ca2+-induced T lymphocytes apoptosis; ICOS-ICOSL signaling in Th cells; communication between innate and adaptive immune system; dendritic cell maturation; autoimmune thyroid disease signaling; superpathway of cholesterol biosynthesis; cholesterol biosynthesis I; cholesterol biosynthesis II; cholesterol biosynthesis III; LXR/RXR activation; neuropathic pain signaling; oxidative ethanol degradation III; TCA cycle II (eukaryotic); ethanol degradation IV; epoxysqualene biosynthesis (Tassoni et al., 2019) (Continues) 21 Table 1. 2 (Continued). Sample Analysis performed Biological function of altered genes/ Altered pathways Reference Endothelial cells of RosatdTomato; VECadherinCreERT2 mice immunized with MOG35-55 FACS and RNAseq Acute phase: immune response; inflammatory response; cellular response to interferon beta; TNF signaling pathway; malaria; cytokine-cytokine receptor interaction; cell adhesion; cellular response to extracellular stimulus; positive regulation of endothelial cell proliferation; ECMreceptor interaction; protein digestion and absorption; focal adhesion Subacute phase: cell adhesion; cell division; mitotic nuclear division; ECM-receptor interaction; Pi3K-AKT signaling pathway; focal adhesion; negative regulation of transcription from RNA polymerase II prmtr; negative regulation of sequence-specific DNA binding transcription factor activity; cell differentiation; MAPK signaling pathway; amphetamine addiction; pathways in cancer Chronic phase: cell adhesion; ossification; collagen fibril organization; ECM-receptor interaction; PI3K-AKT signaling pathway; focal adhesion; muscle organ development negative regulation of ERK1 and ERK2 cascade; intracellular signal transduction; cGMP-PKG signaling pathway; thyroid hormone signaling pathway; oxytocin signaling pathway (Munji et al., 2019) Spinal cord of Biozzi ABH mice immunized with spinal cord homogenate RNAseq Superpathway of cholesterol biosynthesis; dendritic cell maturation; T helper cell differentiation; hepatic fibrosis/hepatic stellate cell activation; altered T cell and B cells signaling in rheumatoid arthritis; acute phase response signaling; antigen presentation pathway; LXR/RXR activation; complement system; role of pattern recognition receptors in recognition of bacteria and viruses; iron transport; cell chemotaxis; cell adhesion; cell differentiation; proliferation (Sevastou et al., 2016) AKT – Protein kinase B; ALS – amyotrophic lateral sclerosis; Ca2+ – calcium; cAMP – cyclic adenosine monophosphate; cGMP-PKG – cyclic guanosine monophosphate dependent protein kinase G; CNS – central nervous system; CREB – cAMP-response element binding protein; CTL – cytotoxic T cell; DARPP32 – dopamineand cAMP-regulated neuronal phosphoprotein; ECM – extracellular matrix; ERK – extracellular-signal-regulated kinase; GABA - gamma-aminobutyric acid; gp-MBP – guinea pig myelin basic protein; HGF – hepatocyte growth factor; ICOS – inducible costimulator; ICOL – ICOS ligand ; IFN – interferon; IkB – inhibitory proteins of kB family; IL – interleukin; LXR – liver X receptor; MAPK – mitogen activated protein kinase; MBP – myelin basic protein; MOG – myelin oligodendrocyte glycoprotein; NF-κB – factor nuclear kappa B; Pi3K – phosphoinositide 3-kinase; PLP – proteolipid protein; RXR – retinoid X receptor; STAT – signal transducer and activator of transcription; TCA – tricarboxylic acid cycle; TCR – T cell receptor; Th – T helper; TMEV-IDD - Theiler’s murine encephalomyelitis (Continues) 22 virus-induced demyelinating disease; TNF – tumor necrosis factor; TREM – triggering receptor expressed on myeloid cells. Of interest, recently our group has identified one astrocytic protein with potential as an additional MS biomarker, which is lipocalin-2 (LCN2), that we will further explore. 1.6.1. Lipocalin-2 LCN2 levels were found to be increased in serum and CSF samples of MS patients, compared to controls (Al-Temaimi et al., 2017; Al Nimer et al., 2016; Berard et al., 2012; Marques et al., 2012). One study, however, reported decreased LCN2 CSF and serum levels in MS patients compared to controls (Khalil et al., 2016). However, these authors identified high LCN2 CSF levels as an independent factor predicting CIS conversion to clinically definite MS, during the follow-up period (Khalil et al., 2016). In addition, Al Nimer and co-workers (2016) reported higher LCN2 CSF levels in PPMS and SPMS patients compared to RRMS (Al Nimer et al., 2016). Interestingly, in the SPMS patients, LCN2 index was positively correlated with the NF-L levels in the CSF (Al Nimer et al., 2016). As for animal studies, Lcn2 was found to be highly upregulated in the CP and spinal cord of two different EAE mice models, at the onset phase of disease (Berard et al., 2012; Marques et al., 2012; Nam et al., 2014). Of interest, LCN2 levels were shown to decrease after natalizumab treatment, both in the EAE model (Marques et al., 2012) and in MS patients (Al Nimer et al., 2016). LCN2 is an acute phase protein that acts as a potent bacteriostatic agent. More specifically, during an infection, bacteria acquire iron, that is important for its replication and growth, from the host by secreting and re-uptaking iron chelators, called siderophores. Siderophores present higher affinity for iron than the mammalian proteins involved in their transport, like lactoferrin, transferrin, and ferritin. LCN2 released by neutrophils at sites of infection and inflammation bind to bacterial siderophores with high specificity and affinity, thus preventing iron sequestering by bacteria (Flo et al., 2004; Goetz et al., 2002). Moreover, the ability of LCN2 to deliver or withdraw iron from cells is important for the regulation of cell apoptosis or proliferation, respectively (Devireddy et al., 2005). The modulation of the homeostatic functions of LCN2 is highly dependent on its interaction with cell-surface receptors, namely the 24p3 cell receptor (24p3R) and megalin (Devireddy et al., 2005; Ferreira et al., 2015). In the CNS, LCN2 is probably involved in similar mechanisms as those described in the periphery, like the modulation of the innate immune response, through siderophore binding, the balance between proand anti-inflammatory responses, cellular activation and migration. Of interest, previous work from our 23 lab has demonstrated that LCN2 is a key modulator of neurogenesis, and that LCN2-null mice present impaired hippocampal-dependent memory and anxious-like behavior (Ferreira et al., 2013; Ferreira et al., 2018). LCN2 expression in the CNS is described to occur mainly in response to an injury or inflammation, while in physiological conditions it was observed by some authors but not by others (Ferreira et al., 2015). LCN2 was shown to be highly expressed in astrocytes, but not in microglia, neurons or oligodendrocytes (Marques et al., 2012; Nam et al., 2014), while the 24p3R is expressed in almost all CNS cell types (Ferreira et al., 2015). Cell culture studies have demonstrated that Lcn2 expression was strongly enhanced in astrocytes stimulated by lipopolysaccharide (LPS) and TNFα (Jang et al., 2013; Lee et al., 2009). Peripheral injection of LPS also increased LCN2 expression in astrocytes and CP cells of wild-type (Wt) mice (Jang et al., 2013; Marques et al., 2008). In addition, cultured astrocytes exposed to LCN2 became reactive, demonstrating hypertrophy of cellular processes, increased GFAP expression, increased NO production, and increased expression of Il-1b , Tnfa , inducible nitric oxide synthase ( Inos ) and Cxcl10 (Jang et al., 2013; Lee et al., 2009). Since astrocytes express the 24p3R, the LCN2 secreted by these cells under inflammatory conditions can act in an autocrine manner to induce morphological alterations (Lee et al., 2009). Two studies so far have used LCN2-null animals to induce EAE, however the results are contradictory. Berard and colleagues (2012) reported that some LCN2-null and Wt animals developed RR EAE, while others presented with chronic disease. For the animals with RR EAE, LCN2-null mice presented increased disease severity, evaluated by a higher mean clinical score at disease peak and delayed mean day of disease remission. However, LCN2-null animals with chronic EAE only presented a significantly higher mean clinical score at the end-stage of disease. Moreover, LCN2-null mice presented an increase in total lesioned area and number of immune cells per spinal cord section, and increased expression levels of TFNα and IFNγ in the spinal cord (Berard et al., 2012). Overall, this study suggests that LCN2 plays a protective role in EAE. On the other hand, Nam and co-workers (2014) observed a decreased disease severity in LCN2-null mice immunized with 200 µg of myelin oligodendrocyte glycoprotein 35-55 (MOG3555), compared to Wt mice, while disease symptoms were not significantly different in the two genotypes under mild disease conditions (50 µg of MOG35-55). LCN2-null animals, immunized with the highest MOG dosage, presented a decrease in infiltration of inflammatory cells, histological score and demyelination area percentage in their spinal cord, in accordance with the lowest disease score observed. Moreover, the increase observed in the number of GFAP+ astrocytes and ionized calcium (Ca2+)-binding adaptor 30 The beneficial roles of reactive astrocytes and the glial scar are evidenced by several studies in EAE animals, in which reactive astrogliosis has been prevented, resulting in increased disease severity associated with increased parenchymal immune-cell infiltration (Haroon et al., 2011; Liedtke et al., 1998; Toft-Hansen et al., 2011; Voskuhl et al., 2009). Hence, ablating astrocyte reactivity completely seems to be detrimental for CNS repair, as astrocytes are important for the confinement of lesions and the restoration of CNS homeostasis (Nair et al., 2008). Astrocytic reactivity is regulated by key canonical signaling cascades, among which the NF-κB pathway is essential for the establishment of neuroinflammation. This signaling pathway is directly activated by stimulation with pro-inflammatory cytokines, among them TNFα and IL-1β (Ponath et al., 2018). Accordingly, EAE animals with selective ablation or inactivation of NF-κB in astrocytes presented a reduced clinical score, usually associated with reduced demyelination, inflammation and neuronal loss (Brambilla et al., 2014; Brambilla et al., 2009; Brosnan and Raine, 2013; Gupta et al., 2019; Sofroniew, 2014). 2.1. Synthesis of metabolic substrates Astrocytes present several metabolic functions essential for brain homeostasis, including the transport of nutrients and metabolic precursors to neurons via the malate-aspartate shuttle; the transport of extracellular glutamate and conversion to glutamine, that can be shuttled back to neurons (glutamateglutamine cycle); the uptake of glucose and synthesis of lactate (glucose-lactate shuttle); glycogen storage; and cholesterol synthesis (Brambilla, 2019; Eilam et al., 2018; Maragakis and Rothstein, 2006; Ponath et al., 2018; Sofroniew and Vinters, 2010). Interestingly, transcriptomic analysis of astrocytes revealed an enriched expression of genes involved in several metabolic pathways (Cahoy et al., 2008; Lovatt et al., 2007). Moreover, Lovatt and colleagues (2007) found that both neurons and astrocytes expressed enzymes mediating aerobic metabolism, however, many pathways entering or exiting glycolysis and the tricarboxylic acid cycle (TCA) cycle, such as those involved in lactate and glutamate metabolism, were compartmentalized, with some reactions being more catalyzed in neurons and others in astrocytes (Lovatt et al., 2007). Specifically, astrocytes were found to overexpress, relatively to neurons, enzymes involved in glycolysis and the TCA cycle (Cahoy et al., 2008; Lovatt et al., 2007). However, in pathological conditions, the stimulation with cytokines for example, can impair the astrocytic metabolic functions (Ponath et al., 2018). 31 The astrocytic location at the interface between blood vessels and neurons, allows them to uptake glucose from circulation and use it to provide energy supply to neurons (glucose-lactate shuttle) (Brambilla, 2019; Ponath et al., 2018; Sofroniew and Vinters, 2010). More specifically, astrocytes degrade glucose to pyruvate which is then converted to lactate and released via astrocytic monocarboxylate transporters (MCT1 and 4) into the extracellular space. Once there, neurons uptake the lactate, via MCT2 and use it as energy substrate, following its intracellular conversion to pyruvate (Zeis et al., 2015). Interestingly, the expression levels of lactate dehydrogenases (Ldh) in astrocytes and neurons support a compartmentalized relationship of lactate metabolism between the two cell types. Namely, Ldhb catalyzes the conversion of pyruvate to lactate and is enriched in astrocytes, while Ldha that catalyzes the reverse reaction is enriched in neurons (Lovatt et al., 2007). Lactate levels were found to be increased in the CSF of CIS and MS patients, in comparison to control subjects, and positively correlated with the number of inflammatory plaques and number of CSF mononuclear cells (Lutz et al., 2007; Simone et al., 1996). Importantly, it is well established that cerebral lactate concentration is directly dependent on its rate of production in the brain, and, in accordance with this, no correlation was found between serum and CSF lactate levels. However the authors hypothesize that the main source of lactate production are the infiltrating macrophages (Lutz et al., 2007; Simone et al., 1996), and not astrocytes. In fact, the astrocytic lactate production seems to be decreased in MS. Microarray analyses of MS brains detected a downregulation of genes involved in the glucose-lactate shuttle, such as MCT1 , while LDHA was increased (Zeis et al., 2015). Moreover, the activation of cultured murine astrocytes with TNFα/IFNγ or TNFα/IL-1β decreased lactate release (Chao et al., 2019; Gavillet et al., 2008). Another source of energy supply to neurons, especially during hypoglycemia or during periods of high neuronal activity, results from glycogenolysis in astrocytes, since these cells are the main storage site of glycogen in the CNS (Brambilla, 2019; Santello et al., 2019; Sofroniew and Vinters, 2010). Actually, astrocytes express enriched levels of all glycogen-metabolizing enzymes, in comparison to neurons (Lovatt et al., 2007). Through the glutamate-glutamine cycle, astrocytes uptake glutamate from the synaptic cleft and convert it to glutamine, that is released back to synapses for reconversion into active neurotransmitters by neurons (Sofroniew and Vinters, 2010). In fact, antisense knock-down studies have shown that the astrocytic glutamate transporters GLAST and glutamate transporter solute carrier family 1 member 2 (SLC1A2/GLT-1) are responsible for over 80% of glutamate uptake in the brain (Maragakis and Rothstein, 2006). Similarly to what is observed for Ldh expression, astrocytes express glutamate synthetase (GLUL), 32 that catalyzes the conversion of glutamate into glutamine, whereas neurons express glutaminase and glutamate decarboxylase 1, involved in glutamate and gamma-aminobutyric acid (GABA) synthesis from glutamine (Lovatt et al., 2007). Following injury, astrocytes initially upregulate the expression of glutamate transporters, like GLT-1, and glutamine synthetase, as a protective mechanism activated by the increased glutamate levels (Kostic et al., 2013; Nair et al., 2008). However, in the case of MS and EAE, in the next phase of disease this protective mechanism seems to be inactivated, since GLT-1 protein levels decrease below control levels (Kostic et al., 2013). Actually, increased CSF glutamate levels were reported to occur in MS patients, compared to individuals lacking objective signs of neurological disorders (Stover et al., 1997a; Stover et al., 1997b), and GLUL expression levels were found to be decreased in the brain of MS patients (Zeis et al., 2015). In vitro studies have shown that inflammatory mediators, like TNFα and IL-1β, can potentiate glutamate neurotoxicity by inhibiting its uptake by astrocytes (Kostic et al., 2013; Musella et al., 2016; Tilleux and Hermans, 2007). Moreover, glutamate excitotoxicity causes damage to both neurons and oligodendrocytes (Pitt et al., 2000). In order to maintain membrane homeostasis and myelin synthesis, neurons and oligodendrocytes are dependent on the cholesterol synthesized by astrocytes, which are the main cells producing cholesterols (Brambilla, 2019; Itoh et al., 2018). However, cholesterol synthesis seems to be impaired in MS and EAE, since the expression of genes involved in this pathway were found to be decreased in both conditions (Itoh et al., 2018; Lavrnja et al., 2017; Lock et al., 2002; Mueller et al., 2008; Sevastou et al., 2016; Tassoni et al., 2019). 2.2. Formation and maintenance of BBB integrity It was already mentioned that the BBB integrity is compromised during MS (Baecher-Allan et al., 2018). In physiological conditions, the BBB separates the brain tissue from the circulating blood, and is constituted by epithelial cells with tight junctions, pericytes, the basement membrane, and astrocytic endfeet (Ortiz et al., 2014; Prat et al., 2001; Sofroniew and Vinters, 2010). This barrier is responsible for the regulation of the import and export of key substances between the peripheral blood and the CNS, including gases, glucose, water, metabolites and cells (Abbott et al., 2006; Broux et al., 2015; Prat et al., 2001). More importantly, the BBB protects the brain from fluctuations in ionic composition, that occur after meals and exercise, for example, and that could disturb the synaptic and axonal signaling (Abbott et al., 2006). Some of the functions performed by astrocytes at the BBB are the production of vasoactive 33 molecules, including prostaglandins and NO, that regulate CNS blood flow. Furthermore, they present specialized water and ion channels, like AQP4, that control the composition of extracellular fluid and the pH (Brambilla, 2019; Ludwin et al., 2016; Ponath et al., 2018; Sofroniew, 2015). It was shown that, during EAE, astrocytes retract their processes, thereby decreasing perivascular coverage (Figure 1. 5) (Eilam et al., 2018). In addition, the distribution of AQP4 channels is described to be highly polarized towards astrocytic foot processes, however this polarization is loss in astrocytes located near EAE lesions (Figure 1. 5) (Wolburg-Buchholz et al., 2009). IL-6, TNFα, IL-1β, immunity-related GTPase family M member 1, vascular endothelial growth factor (VEGF), and CCL2 released by astrocytes during inflammation increase BBB permeability, by acting on endothelial cells and their tight junctions (Brosnan and Raine, 2013; Broux et al., 2015; Nair et al., 2008; Wang et al., 2013a). VEGF has the additional characteristic of being chemotactic for monocytes, so its release by astrocytes also increases leukocyte extravasation (Brosnan and Raine, 2013; Proescholdt et al., 2002; Sofroniew, 2015). Of notice, a high number of VEGF+ astrocytes were found in or adjacent to MS lesions, while no VEGF+ cells were observed in control brains or unaffected regions of MS brains (Proescholdt et al., 2002). In accordance with these deleterious functions, the conditional deletion of VEGF-A in astrocytes (Gfap-Cre:Vegfafl/fl) improved the clinical signs of EAE, associated with the maintenance of claudin-5 expression (tight junction protein) and a restriction in BBB breakdown (Argaw et al., 2012). Endothelin-1 (ET-1) is a potent vasoconstrictor predominantly produced by endothelial cells and also by reactive astrocytes. ET-1 overexpression both in endothelial cells and astrocytes, and posterior EAE induction, resulted in more severe disease, increased demyelination and increased inflammatory cell infiltration. Furthermore, plasma and CSF levels of ET-1 are increased in MS patients (Guo et al., 2014). Other proteins released by reactive astrocytes during EAE have an opposite effect, being involved in the BBB repair and inhibition of leukocyte migration. These include annexin-1, retinoic acid and sonic hedgehog synthesis (Brambilla, 2019; Huitinga et al., 1998; Mizee et al., 2014; Sofroniew, 2015). The basement membrane is composed of ECM proteins, including glycoproteins and fibrous proteins, that are synthetized by endothelial cells and astrocytes (Miljkovic et al., 2011). As mentioned before, MMPs produced by astrocytes, namely MMP2 and MMP9, are able to degrade these ECM proteins (Figure 1. 5) (Miljkovic et al., 2011; Song et al., 2015). However, astrocytes also have the ability to produce tissue inhibitors of MMPs (TIMPs), namely TIMP-1, which inhibit the action of MMPs (Miljkovic et al., 2011; Pagenstecher et al., 1998; Teesalu et al., 2001). 34 Remarkably, studies performed in GFAP-luciferase animals showed bioluminescence, indicative of GFAP transcriptional response, in the CNS of mice before CD4+ T cell infiltration and the presence of clinical manifestations, suggesting that astrocytes become reactive before CNS inflammation and could contribute to BBB destabilization and permeability (Broux et al., 2015; Luo et al., 2008). 2.3. Synapse maintenance and refinement The ramified processes of astrocytes make contact with preand post-synaptic terminals, forming the tripartite synapse. Within this structure, astrocytes are involved in different processes, like regulation of synaptogenesis, synaptic plasticity and stability, uptake of glutamate and buffering of extracellular potassium (K+) (Brambilla, 2019; Ponath et al., 2018; Santello et al., 2019). There is now accumulating evidence that astrocytes directly participate in synaptic transmission. More specifically, alterations in neuronal synaptic activity induce the release of neurotransmitters, such as glutamate and GABA, that are sensed by astrocytes. Astrocytes respond by increasing its intracellular Ca2+ concentration, and posteriorly releasing synaptically active molecules (gliotransmitters), like glutamate, adenosine triphosphate (ATP), adenosine and D-serine (Barres, 2008; Sofroniew and Vinters, 2010). 2.4. Synthesis of immunomodulatory molecules and growth factors In EAE and MS, astrocytes can synthesize several immunomodulatory factors: chemokines involved in increased BBB permeability and immune cell migration, including CCL2, CCL3, CCL4, CCL5, CCL8, CCL20, CCL19, CX3CL1, CXCL8, CXCL10, CXCL12 and CXCL1; proand anti-inflammatory cytokines, such as TNFα, IL-1β, IL-6, IL-10, IL-12, IL-23, IL-27, IL-15, IL-33, IFNα and TGFβ; complement proteins, like complement component 5a (C5a), C3a, C1q; B-cell activation factor, involved in B cell survival and proliferation; and ROS and RNS involved in oxidative stress (Ambrosini et al., 2005; Berman et al., 1996; Brambilla, 2019; Chastain et al., 2011; Columba-Cabezas et al., 2003; Correale and Farez, 2015; Glabinski et al., 1997; Glabinski et al., 1995; Lopes Pinheiro et al., 2016; Miljkovic et al., 2011; Miyagishi et al., 1997; Nair et al., 2008; Nataf et al., 1998; Ponath et al., 2018; Ransohoff et al., 1993; Shrestha et al., 2014; Sunnemark et al., 2005; Villarroya et al., 1996; Williams et al., 2007). The astrocytic targeted deletion of some of the immunomodulatory factors mentioned was able to improve disease symptoms in EAE animals. Namely, the conditional deletion of IL-6 in astrocytes was able to significantly decrease the clinical score in EAE female mice. Moreover, these mice presented a significant 35 decrease in the number of infiltrates and in astrocyte reactivity, and an increased percentage of myelinated area (Erta et al., 2016). The deletion of CXCL10 in astrocytes delayed EAE onset and decreased the severity of cumulative clinical deficits, compared to control animals. However, during the chronic phase of disease there were no differences between the experimental groups, indicating that the effect of astrocyte-derived CXCL10 is exerted during the initial phases of disease (Mills Ko et al., 2014). The conditional deletion of CCL2 decreased the severity of EAE clinical neurological deficits (Moreno et al., 2014; Paul et al., 2014). This was accompanied by a significant decrease of CNS leukocyte accumulation (Moreno et al., 2014; Paul et al., 2014) and by a higher preservation of claudin-5 staining (Paul et al., 2014). Similarly, the astrocytic targeted expression of a complement inhibitor, complement receptor-related protein y, improved the clinical scores of EAE animals, that presented decreased immune cell infiltration in the CNS (Davoust et al., 1999). On the contrary, the overexpression of C3a in astrocytes increased the EAE severity and mortality rate. This was associated with an increased CNS infiltration with macrophages and CD4+ T cells (Boos et al., 2004). The overexpression of C5a in astrocytes did not affect disease development (Reiman et al., 2005). IFNγ signaling in astrocytes was described as being important for protection against autoimmune-mediated neurological disability, since EAE induction in mice expressing a signaling deficient dominant negative IFNγ receptor I specifically in astrocytes presented increased disease severity, associated with higher expression levels of CCL5, IL-1 and TNF, loss of axons and increased demyelination (Hindinger et al., 2012). Some of the chemokines produced by astrocytes are also chemoattractants for OPCs, including CXCL1, CXCL8 and CXCL10, allowing them to migrate towards the demyelinated plaques (Williams et al., 2007). Surprisingly, CXCL12 and CCL2 also seemed to be involved in the attraction of transplanted neural precursor cells, in the EAE model, suggesting that, besides attracting inflammatory cells, they could also play a role in the migration of precursor cells to lesions (Cohen et al., 2014). Astrocytes were found to be central contributors for oxidative stress in EAE (Mossakowski et al., 2015). iNOS, endothelial nitric oxide synthase and neuronal nitric oxide synthase expression was found to be upregulated in astrocytes during EAE, contributing to NO production (Brambilla, 2019; Cross et al., 1997; Kim et al., 2000; Shin, 1999; 2001; Tran et al., 1997). In addition, in situ hybridization and immunohistochemistry studies have shown extensive iNOS reactivity in hypertrophic astrocytes in acute but not chronic MS lesions (Brosnan and Raine, 2013). ROS and RNS cause oxidative stress and have been implicated in mechanisms underlying lesion pathogenesis (Brambilla, 2019; Emerson and LeVine, 2000). They are produced not only by microglia and immune cells, but also by astrocytes (Brambilla, 2019). The main local effects of increased ROS and RNS are the modulation of BBB function and 36 oligodendrocyte damage (Brambilla, 2019; Miljkovic et al., 2011). Oligodendrocytes are particularly vulnerable to oxidative stress possibly because their antioxidant levels are low (Miljkovic et al., 2011). Contrariwise, and in an attempt to decrease oxidative-mediated cellular damage, astrocytes overexpress heme-oxygenase 1, involved in the protection against oxidant-mediated injury (Emerson and LeVine, 2000), and antioxidant enzymes, like superoxide dismutase 2, NAD(P)H Quinone Dehydrogenase 1, and peroxiredoxins (Brambilla, 2019; Qi et al., 1997; Yun et al., 2015). Reactive astrocytes also protect neurons from oxidative stress via glutathione dependent mechanisms (Gavillet et al., 2008; Pekny and Nilsson, 2005). Of interest, a possible beneficial role has also been suggested for increased NO levels, namely in the apoptosis of inflammatory cells (Kim et al., 2000). Besides immunomodulator molecules, astrocytes also increase the production of growth factors in the disease context, like brain-derived neurotrophic factor (BDNF) and nerve growth factor (Brambilla, 2019; Chastain et al., 2011; Micera et al., 1998; Nair et al., 2008; Ponath et al., 2018; Stadelmann et al., 2002). The increased levels of these growth factors could represent an attempt to prevent or reduce neurite damage in the lesioned regions (Micera et al., 1998). In accordance with these protective properties, a worsening of EAE symptoms could be observed in GFAP-Cre:BDNFfl/fl mice, which present a specific deletion of BDNF in astrocytes. These animals additionally presented increased mortality and axonal damage (Linker et al., 2010). 2.5. Interaction between astrocytes and T cells Activated anti-myelin T cells that have infiltrated the CNS will not induce injury unless they are reactivated against their specific antigen, through its presentation by APCs (De Keyser et al., 2003). Astrocytes have been pointed as possible APCs of myelin antigens to T cells, in the context of EAE and MS (Correale and Farez, 2015; De Keyser et al., 2003; Ponath et al., 2018). Namely, astrocytes have been shown to process and present myelin-related antigens, such as MBP, proteolipid protein (PLP) and MOG, to encephalitogenic CD4+ T cells (Miljkovic et al., 2011; Myers et al., 1993). Importantly, the necessary molecules for efficient antigen presentation to and activation of CD4+ T cells, including MHC class II and co-stimulator molecules, like B7-1 (CD80) and B7-2 (CD86), have been shown to be present on astrocytes (Issazadeh et al., 1998; Lee et al., 1990; Miljkovic et al., 2011; Nair et al., 2008; Nikcevich et al., 1997; Ponath et al., 2018; Villarroya et al., 2001). The astrocytic interaction with T cells is also important for T-cell infiltration in the CNS. More specifically, receptor activator of NF-κB (RANK) ligand (RANKL) expression on T cells was shown to induce the 37 expression of CCL20 on astrocytes, and both the T-cell-specific inactivation of RANKL and the astrocytespecific deletion of RANK decreased the infiltration of T cells into the CNS of EAE animals, significantly improving their clinical score (Guerrini et al., 2015). Furthermore, T cell apoptosis can be induced by astrocytes, for example via Fas ligand (FasL) (Correale and Farez, 2015; Hara et al., 2011; Wang et al., 2013b; Xie and Yang, 2015). FasL is expressed in astrocytes and increases in response to cytokines, like IL-1, IL-6, TNFα and IFNγ (Xie and Yang, 2015). A previous study has demonstrated that astrocytic FasL is crucial for the termination of the autoimmune T cell response in the CNS, which allows clinical recovery of symptoms in EAE. Specifically, GFAPCre:FasLfl/fl mice induced with EAE presented a similar day of first symptoms and peak of disease, however control mice recovered after the peak while the conditional knock-outs did not. Furthermore, these animals presented more widespread inflammation, characterized by an increased number of CD4+ and CD8+ T cells, and more severe demyelination, compared to controls. they also presented a decreased percentage of 7-Aminoactinomycin D positive CD4+ T cell, suggesting that the elimination of these cells by apoptosis was impaired (Wang et al., 2013b). 2.6. Astrocyte subpopulations Astrocytes are a very heterogenous population of cells in terms of gene expression profile, physiological properties and response to injury (Batiuk et al., 2020; Bribian et al., 2018; Zhang and Barres, 2010). Differences have even been observed between astrocytic subpopulations regarding spontaneous and behavior-evoked Ca2+ activities. Being such a diverse group of cells, it is not surprising that different astrocytic subpopulations respond to disease and injury in different manners (Zhang and Barres, 2010). In order to better understand astrocyte heterogeneity, several laboratories have recently taken advantage of transcriptomic tools to perform molecular classifications of astrocyte subpopulations, both under physiological and pathological conditions (Boisvert et al., 2018; Chai et al., 2017; Clarke et al., 2018; John Lin et al., 2017; Pekny and Pekna, 2016). Interestingly, these studies have shown that astrocytes exhibit broad and diverse changes in gene expression profiles in response to different types of stimulation (Sofroniew, 2014), however, common altered pathways induced in reactive astrocytes include antigen presentation and regulation of inflammatory-mediators production. In vitro studies have shown major differences in the gene expression profile of astrocytes stimulated with different mediators of reactive astrogliosis, such as LPS, IL-1β, TNFα, IFNγ or TGFβ (Anderson et al., 2014; Hamby et al., 2012). In vivo studies comparing neuroinflammation and ischemia models found 38 two different types of reactive astrocytes, which were recently termed “A1” and “A2”, in parallel to the “M1” and “M2” macrophage nomenclature (Liddelow and Barres, 2017; Liddelow et al., 2017). A1 astrocytes were found to lose the ability to promote neuronal survival, outgrowth and synaptogenesis. Instead, A1 astrocytes highly upregulate classical complement cascades previously shown to be destructive to synapses, thus acquiring neurotoxic functions, inducing the death of neurons and oligodendrocytes. By contrast, A2 astrocytes, present in ischemia models, upregulate the expression of genes involved in repair and protective functions, such as neurotrophic factors, therefore presenting a neuroprotective phenotype (Anderson et al., 2014; Liddelow et al., 2017; Zamanian et al., 2012). Of interest, C3+ A1 astrocytes were observed in MS demyelinating lesions (Liddelow et al., 2017). It should be noted that this nomenclature is too simplistic, and more than two polarization states have been described for macrophages (M1, M2a, M2b, M2c) (Liddelow and Barres, 2017). The same is probably true in the case of astrocytes, since recent studies have identified several astrocytic clusters both in MS and EAE (Table 1. 4). Table 1. 4 – Identification of astrocytic sub-populations by single cell analysis of samples from EAE animals or MS patients. Samples Methodology Results Pathway analysis Reference Brain and spinal cord of C57BL/6 mice immunized with MOG35-55 Single-cell RNAseq Identification of 8 clusters of astrocytes, being cluster 4 the most expanded in EAE Cluster 4: mitochondrial dysfunction; NRF2 oxidative stress response; unfolded protein response; production of NO and ROS; IL-6 signaling; neuroinflammation signaling; leukocyte extravasation signaling; GM-CSF signaling; NF-κB signaling; pathogenesis of MS; DNA methylation signaling (Wheeler et al., 2020) Brain and spinal cord of TdTomatoGfap mice immunized with MOG35-55 Single-cell RNAseq Identification of 17 clusters of astrocytes, being cluster 5 the most expanded in EAE Cluster 5: mitochondrial function; NRF2 oxidative stress response; unfolded protein response; hypoxia signaling; chemokine signaling; S1P signaling; ceramide signaling; ER stress pathway; IL-6 signaling; GMCSF signaling (Wheeler et al., 2020) MS patients brain lesions Single-cell RNAseq Identification of 16 clusters of astrocytes, being cluster 1 the most expanded in MS Cluster 1: neuroinflammation signaling; chemokine signaling; mitochondrial dysfunction; NRF2 oxidative stress response; GM-CSF signaling; IL-6 signaling; iNOS signaling; NF-κB signaling (Wheeler et al., 2020) (Continues) 39 Table 1. 4 (Continued). Samples Methodology Results Pathway analysis Reference One demyelinating lesion and one postdemyelinating lesion from two MS patients Imaging mass cytometry and single-cell RNAseq Identification of 5 clusters in the demyelinating lesion and 2 clusters in the postdemyelinating lesion Not performed (Park et al., 2019) Cortical gray matter and adjacent subcortical white matter from lesions of MS patients Single-nucleus RNAsequencing Differentiation between protoplasmic and fibrous/reactive astrocytes Glutamate homeostasis; potassium homeostasis (Schirmer et al., 2019) ER – endoplasmic reticulum; GM-CSF – granulocyte-macrophage colony-stimulating factor; IL – interleukin; iNOS – inducible nitric oxide synthase; MS – multiple sclerosis; NF-κB - Nuclear factor kappa B; NO – nitric oxide; NRF2 – nuclear factor erythroid 2-related factor 2; ROS – reactive oxygen species; S1P – sphingosine-1-phosphate. Future studies, that help to define astroglial subtypes based on their molecular, functional and structural properties will greatly improve our understanding of their specific roles in MS pathophysiology (Correale and Farez, 2015). 2.7. Astrocyte Ca2+ signaling Astrocytes do not elicit or propagate action potentials along their processes. Instead, they exhibit regulated increases in intracellular Ca2+ concentrations, that represent a form of astrocyte excitability (Scemes and Giaume, 2006; Sofroniew and Vinters, 2010). Ca2+ waves are defined as localized intracellular Ca2+ increases, that are followed by a succession of similar events in a wave-like fashion, and they can be restricted to one cell (intracellular) or transmitted to neighboring cells (intercellular) (Scemes and Giaume, 2006). Intercellular Ca2+ waves can propagate by cell-to-cell diffusion, through gap junction channels, or by the release of factors like ATP (Abbott et al., 2006; Scemes and Giaume, 2006; Volterra and Meldolesi, 2005). Astrocytes are linked and communicate with each other by gap junctions, formed by connexin (Cx) 43 and Cx30 (Correale and Farez, 2015), thus allowing the transfer of ions and metabolites throughout the astrocytic network (Brand-Schieber et al., 2005), including K+ dissipation and intercellular Ca2+ waves 46 Thesis aims The role of astrocytes in the pathophysiology of MS is not fully understood. Recent studies have provided information regarding astrocytic genes and pathways possibly altered at specific disease time points, however an analysis performed throughout disease development is still missing, which we propose to perform in this work. With this analysis we expect to unravel novel possible disease biomarkers or therapeutic targets. One protein that was found to be overexpressed by astrocytes in the EAE context, and that is increased in samples from MS patients, is LCN2. Nevertheless, LCN2 is also increased in the context of other diseases, so its diagnostic value alone is poor, but its prognostic value in MS is currently unknown. As already mentioned, alterations in intracellular Ca2+ levels are important for astrocytic function, and represent a form of astrocyte excitability. “Silencing” astrocytes, by impairing global astrocytic Ca2+ signaling, is beneficial in some disease animal models, but detrimental in others, but its impact on EAE remains to be addressed. A growing body of data indicates that astrocytes are a good therapeutic target to treat many neurological conditions. In the case of MS, it could be particularly important for progressive forms of disease, since astrocytes seem to play an important role during this stage of the disease. Moreover, astrocytes also seem to play a role in the mechanisms leading to cognitive deficits in MS patients. In this regard, DMF could be a good drug to target astrocytes and improve cognition in MS, since it presents immunomodulatory and neuroprotective effects, and was shown to modulate the inflammatory response of astrocytes. Taking all this into consideration, in this work we intended to: - Explore the transcriptomic and morphological alterations occurring in astrocytes during EAE, at distinct time points along disease development; - Assess the impact of “silencing” astrocytes for EAE development and progression; - Study the impact of dimethyl fumarate on the cognitive performance of EAE animals and its effect on astrocytes; - Evaluate the prognostic value of LCN2 for MS progression. 47 References Abbott NJ, Ronnback L, Hansson E. Astrocyte-endothelial interactions at the blood-brain barrier. Nature Reviews Neuroscience. 2006;7:41-53. Agarwal A, Wu PH, Hughes EG, Fukaya M, Tischfield MA, Langseth AJ, et al. Transient opening of the mitochondrial permeability transition pore induces microdomain calcium transients in astrocyte processes. Neuron. 2017;93:587-605 e587. Agulhon C, Petravicz J, McMullen AB, Sweger EJ, Minton SK, Taves SR, et al. What is the role of astrocyte calcium in neurophysiology? Neuron. 2008;59:932-946. Al-Temaimi R, AbuBaker J, Al-Khairi I, Alroughani R. Remyelination modulators in multiple sclerosis patients. Experimental and Molecular Pathology. 2017;103:237-241. Al Nimer F, Elliott C, Bergman J, Khademi M, Dring AM, Aeinehband S, et al. Lipocalin-2 is increased in progressive multiple sclerosis and inhibits remyelination. Neurology: Neuroimmunology and NeuroInflammation. 2016;3:e191. Alcina A, Abad-Grau Mdel M, Fedetz M, Izquierdo G, Lucas M, Fernandez O, et al. Multiple sclerosis risk variant HLA-DRB1*1501 associates with high expression of DRB1 gene in different human populations. PloS One. 2012;7:e29819. Allen IV, McQuaid S, Mirakhur M, Nevin G. Pathological abnormalities in the normal-appearing white matter in multiple sclerosis. Neurological Sciences. 2001;22:141-144. Alme MN, Nystad AE, Bo L, Myhr KM, Vedeler CA, Wergeland S, et al. Fingolimod does not enhance cerebellar remyelination in the cuprizone model. Journal of Neuroimmunology. 2015;285:180-186. Alvarez E, Piccio L, Mikesell RJ, Klawiter EC, Parks BJ, Naismith RT, et al. CXCL13 is a biomarker of inflammation in multiple sclerosis, neuromyelitis optica, and other neurological conditions. Multiple Sclerosis. 2013;19:1204-1208. Alvarez JI, Saint-Laurent O, Godschalk A, Terouz S, Briels C, Larouche S, et al. Focal disturbances in the blood-brain barrier are associated with formation of neuroinflammatory lesions. Neurobiology of Disease. 2015;74:14-24. Ambrosini E, Remoli ME, Giacomini E, Rosicarelli B, Serafini B, Lande R, et al. Astrocytes produce dendritic cell-attracting chemokines in vitro and in multiple sclerosis lesions. Journal of Neuropathology and Experimental Neurology. 2005;64:706-715. Anderson MA, Ao Y, Sofroniew MV. Heterogeneity of reactive astrocytes. Neuroscience Letters. 2014;565:23-29. 48 Araque A, Parpura V, Sanzgiri RP, Haydon PG. Glutamate-dependent astrocyte modulation of synaptic transmission between cultured hippocampal neurons. European Journal of Neuroscience. 1998;10:2129-2142. Argaw AT, Asp L, Zhang J, Navrazhina K, Pham T, Mariani JN, et al. Astrocyte-derived VEGF-A drives blood-brain barrier disruption in CNS inflammatory disease. Journal of Clinical Investigation. 2012;122:2454-2468. Azary S, Schreiner T, Graves J, Waldman A, Belman A, Guttman BW, et al. Contribution of dietary intake to relapse rate in early paediatric multiple sclerosis. Journal of Neurology, Neurosurgery and Psychiatry. 2018;89:28-33. Baecher-Allan C, Kaskow BJ, Weiner HL. Multiple sclerosis: mechanisms and immunotherapy. Neuron. 2018;97:742-768. Barak Y, Achiron A. Effect of interferon-beta-1b on cognitive functions in multiple sclerosis. European Neurology. 2002;47:11-14. Barateiro A, Afonso V, Santos G, Cerqueira JJ, Brites D, van Horssen J, et al. S100B as a potential biomarker and therapeutic target in multiple sclerosis. Molecular Neurobiology. 2016;53:3976-3991. Barres BA. The mystery and magic of glia: a perspective on their roles in health and disease. Neuron. 2008;60:430-440. Bartosik-Psujek H, Psujek M, Jaworski J, Stelmasiak Z. Total tau and S100b proteins in different types of multiple sclerosis and during immunosuppressive treatment with mitoxantrone. Acta Neurologica Scandinavica. 2011;123:252-256. Bartosik-Psujek H, Stelmasiak Z. Correlations between IL-4, IL-12 levels and CCL2, CCL5 levels in serum and cerebrospinal fluid of multiple sclerosis patients. Journal of Neural Transmission. 2005;112:797-803. Batiuk MY, Martirosyan A, Wahis J, de Vin F, Marneffe C, Kusserow C, et al. Identification of regionspecific astrocyte subtypes at single cell resolution. Nature Communications. 2020;11:1220. Berard JL, Zarruk JG, Arbour N, Prat A, Yong VW, Jacques FH, et al. Lipocalin 2 is a novel immune mediator of experimental autoimmune encephalomyelitis pathogenesis and is modulated in multiple sclerosis. Glia. 2012;60:1145-1159. Berger T, Rubner P, Schautzer F, Egg R, Ulmer H, Mayringer I, et al. Antimyelin antibodies as a predictor of clinically definite multiple sclerosis after a first demyelinating event. The New England Journal of Medicine. 2003;349:139-145. 49 Berman JW, Guida MP, Warren J, Amat J, Brosnan CF. Localization of monocyte chemoattractant peptide-1 expression in the central nervous system in experimental autoimmune encephalomyelitis and trauma in the rat. Journal of Immunology. 1996;156:3017-3023. Bjugstad KB, Flitter WD, Garland WA, Su GC, Arendash GW. Preventive actions of a synthetic antioxidant in a novel animal model of AIDS dementia. Brain Research. 1998;795:349-357. Boisvert MM, Erikson GA, Shokhirev MN, Allen NJ. The aging astrocyte transcriptome from multiple regions of the mouse brain. Cell Reports. 2018;22:269-285. Boos L, Campbell IL, Ames R, Wetsel RA, Barnum SR. Deletion of the complement anaphylatoxin C3a receptor attenuates, whereas ectopic expression of C3a in the brain exacerbates, experimental autoimmune encephalomyelitis. Journal of Immunology. 2004;173:4708-4714. Brambilla R. The contribution of astrocytes to the neuroinflammatory response in multiple sclerosis and experimental autoimmune encephalomyelitis. Acta Neuropathologica. 2019;137:757-783. Brambilla R, Morton PD, Ashbaugh JJ, Karmally S, Lambertsen KL, Bethea JR. Astrocytes play a key role in EAE pathophysiology by orchestrating in the CNS the inflammatory response of resident and peripheral immune cells and by suppressing remyelination. Glia. 2014;62:452-467. Brambilla R, Persaud T, Hu X, Karmally S, Shestopalov VI, Dvoriantchikova G, et al. Transgenic inhibition of astroglial NF-kappa B improves functional outcome in experimental autoimmune encephalomyelitis by suppressing chronic central nervous system inflammation. Journal of Immunology. 2009;182:2628-2640. Branco M, Ruano L, Portaccio E, Goretti B, Niccolai C, Patti F, et al. Aging with multiple sclerosis: prevalence and profile of cognitive impairment. Neurological Sciences. 2019;40:1651-1657. Brand-Schieber E, Werner P, Iacobas DA, Iacobas S, Beelitz M, Lowery SL, et al. Connexin43, the major gap junction protein of astrocytes, is down-regulated in inflamed white matter in an animal model of multiple sclerosis. Journal of Neuroscience Research. 2005;80:798-808. Brennan MS, Matos MF, Li B, Hronowski X, Gao B, Juhasz P, et al. Dimethyl fumarate and monoethyl fumarate exhibit differential effects on KEAP1, NRF2 activation, and glutathione depletion in vitro. PloS One. 2015;10:e0120254. Brennan MS, Matos MF, Richter KE, Li B, Scannevin RH. The NRF2 transcriptional target, OSGIN1, contributes to monomethyl fumarate-mediated cytoprotection in human astrocytes. Scientific Reports. 2017;7:42054. 50 Brettschneider J, Tumani H, Kiechle U, Muche R, Richards G, Lehmensiek V, et al. IgG antibodies against measles, rubella, and varicella zoster virus predict conversion to multiple sclerosis in clinically isolated syndrome. PloS One. 2009;4:e7638. Bribian A, Perez-Cerda F, Matute C, Lopez-Mascaraque L. Clonal glial response in a multiple sclerosis mouse model. Frontiers in Cellular Neuroscience. 2018;12:375. Brosnan CF, Raine CS. The astrocyte in multiple sclerosis revisited. Glia. 2013;61:453-465. Broux B, Gowing E, Prat A. Glial regulation of the blood-brain barrier in health and disease. Seminars in Immunopathology. 2015;37:577-590. Brownlee WJ, Miller DH. Clinically isolated syndromes and the relationship to multiple sclerosis. Journal of Clinical Neuroscience. 2014;21:2065-2071. Burness CB, Deeks ED. Dimethyl fumarate: a review of its use in patients with relapsing-remitting multiple sclerosis. CNS Drugs. 2014;28:373-387. Cahoy JD, Emery B, Kaushal A, Foo LC, Zamanian JL, Christopherson KS, et al. A transcriptome database for astrocytes, neurons, and oligodendrocytes: a new resource for understanding brain development and function. Journal of Neuroscience. 2008;28:264-278. Calabrese M, Agosta F, Rinaldi F, Mattisi I, Grossi P, Favaretto A, et al. Cortical lesions and atrophy associated with cognitive impairment in relapsing-remitting multiple sclerosis. Archives of neurology. 2009;66:1144-1150. Cambron M, D'Haeseleer M, Laureys G, Clinckers R, Debruyne J, De Keyser J. White-matter astrocytes, axonal energy metabolism, and axonal degeneration in multiple sclerosis. Journal of Cerebral Blood Flow and Metabolism. 2012;32:413-424. Carmody RJ, Hilliard B, Maguschak K, Chodosh LA, Chen YH. Genomic scale profiling of autoimmune inflammation in the central nervous system: the nervous response to inflammation. Journal of Neuroimmunology. 2002;133:95-107. Ceccarelli A, Rocca MA, Falini A, Tortorella P, Pagani E, Rodegher M, et al. Normal-appearing white and grey matter damage in MS. A volumetric and diffusion tensor MRI study at 3.0 Tesla. Journal of Neurology. 2007;254:513-518. Celarain N, Tomas-Roig J. Aberrant DNA methylation profile exacerbates inflammation and neurodegeneration in multiple sclerosis patients. Journal of Neuroinflammation. 2020;17:21. Chai H, Diaz-Castro B, Shigetomi E, Monte E, Octeau JC, Yu X, et al. Neural circuit-specialized astrocytes: transcriptomic, proteomic, morphological, and functional evidence. Neuron. 2017;95:531549 e539. 51 Chao CC, Gutierrez-Vazquez C, Rothhammer V, Mayo L, Wheeler MA, Tjon EC, et al. Metabolic control of astrocyte pathogenic activity via cPLA2-MAVS. Cell. 2019;179:1483-1498 e1422. Chastain EM, Duncan DS, Rodgers JM, Miller SD. The role of antigen presenting cells in multiple sclerosis. Biochimica et Biophysica Acta. 2011;1812:265-274. Chen J, Chia N, Kalari KR, Yao JZ, Novotna M, Paz Soldan MM, et al. Multiple sclerosis patients have a distinct gut microbiota compared to healthy controls. Scientific Reports. 2016;6:28484. Chiaravalloti ND, DeLuca J. Cognitive impairment in multiple sclerosis. Lancet Neurology. 2008;7:1139-1151. Choi JW, Gardell SE, Herr DR, Rivera R, Lee CW, Noguchi K, et al. FTY720 (fingolimod) efficacy in an animal model of multiple sclerosis requires astrocyte sphingosine 1-phosphate receptor 1 (S1P1) modulation. Proceedings of the National Academy of Sciences of the United States of America. 2011;108:751-756. Ciccarelli O, Barkhof F, Bodini B, De Stefano N, Golay X, Nicolay K, et al. Pathogenesis of multiple sclerosis: insights from molecular and metabolic imaging. Lancet Neurology. 2014;13:807-822. Clarke LE, Liddelow SA, Chakraborty C, Munch AE, Heiman M, Barres BA. Normal aging induces A1like astrocyte reactivity. Proceedings of the National Academy of Sciences of the United States of America. 2018;115:E1896-E1905. Cohen ME, Fainstein N, Lavon I, Ben-Hur T. Signaling through three chemokine receptors triggers the migration of transplanted neural precursor cells in a model of multiple sclerosis. Stem Cell Research. 2014;13:227-239. Colombo E, Di Dario M, Capitolo E, Chaabane L, Newcombe J, Martino G, et al. Fingolimod may support neuroprotection via blockade of astrocyte nitric oxide. Annals of Neurology. 2014;76:325-337. Colotta F, Jansson B, Bonelli F. Modulation of inflammatory and immune responses by vitamin D. Journal of Autoimmunity. 2017;85:78-97. Columba-Cabezas S, Serafini B, Ambrosini E, Aloisi F. Lymphoid chemokines CCL19 and CCL21 are expressed in the central nervous system during experimental autoimmune encephalomyelitis: implications for the maintenance of chronic neuroinflammation. Brain Pathology. 2003;13:38-51. Comabella M, Montalban X. Body fluid biomarkers in multiple sclerosis. Lancet Neurology. 2014;13:113-126. Compston A, Coles A. Multiple sclerosis. Lancet. 2008;372:1502-1517. Cornell-Bell AH, Finkbeiner SM, Cooper MS, Smith SJ. Glutamate induces calcium waves in cultured astrocytes: long-range glial signaling. Science. 1990;247:470-473. 52 Correale J, Farez MF. The role of astrocytes in multiple sclerosis progression. Frontiers in neurology. 2015;6:180. Correale J, Gaitan MI, Ysrraelit MC, Fiol MP. Progressive multiple sclerosis: from pathogenic mechanisms to treatment. Brain. 2017;140:527-546. Cosorich I, Dalla-Costa G, Sorini C, Ferrarese R, Messina MJ, Dolpady J, et al. High frequency of intestinal TH17 cells correlates with microbiota alterations and disease activity in multiple sclerosis. Science Advances. 2017;3:e1700492. Cross AH, Manning PT, Stern MK, Misko TP. Evidence for the production of peroxynitrite in inflammatory CNS demyelination. Journal of Neuroimmunology. 1997;80:121-130. Cruz M, Olsson T, Ernerudh J, Hojeberg B, Link H. Immunoblot detection of oligoclonal anti-myelin basic protein IgG antibodies in cerebrospinal fluid in multiple sclerosis. Neurology. 1987;37:1515-1519. Cunnea P, McMahon J, O'Connell E, Mashayekhi K, Fitzgerald U, McQuaid S. Gene expression analysis of the microvascular compartment in multiple sclerosis using laser microdissected blood vessels. Acta Neuropathologica. 2010;119:601-615. D'Amelio FE, Smith ME, Eng LF. Sequence of tissue responses in the early stages of experimental allergic encephalomyelitis (EAE): immunohistochemical, light microscopic, and ultrastructural observations in the spinal cord. Glia. 1990;3:229-240. Darwish H, Haddad R, Osman S, Ghassan S, Yamout B, Tamim H, et al. Effect of vitamin D replacement on cognition in multiple sclerosis patients. Scientific Reports. 2017;7:45926. Davoust N, Nataf S, Reiman R, Holers MV, Campbell IL, Barnum SR. Central nervous system-targeted expression of the complement inhibitor sCrry prevents experimental allergic encephalomyelitis. Journal of Immunology. 1999;163:6551-6556. De Keyser J, Zeinstra E, Frohman E. Are astrocytes central players in the pathophysiology of multiple sclerosis? Archives of neurology. 2003;60:132-136. Deisenhammer F, Zetterberg H, Fitzner B, Zettl UK. The cerebrospinal fluid in multiple sclerosis. Frontiers in Immunology. 2019;10:726. Devireddy LR, Gazin C, Zhu X, Green MR. A cell-surface receptor for lipocalin 24p3 selectively mediates apoptosis and iron uptake. Cell. 2005;123:1293-1305. Ding S, Fellin T, Zhu Y, Lee SY, Auberson YP, Meaney DF, et al. Enhanced astrocytic Ca2+ signals contribute to neuronal excitotoxicity after status epilepticus. Journal of Neuroscience. 2007;27:1067410684. 53 DiSano KD, Linzey MR, Royce DB, Pachner AR, Gilli F. Differential neuro-immune patterns in two clinically relevant murine models of multiple sclerosis. Journal of Neuroinflammation. 2019;16:109. Djedovic N, Stanisavljevic S, Jevtic B, Momcilovic M, Lavrnja I, Miljkovic D. Anti-encephalitogenic effects of ethyl pyruvate are reflected in the central nervous system and the gut. Biomedicine and Pharmacotherapy. 2017;96:78-85. Dobson R, Ramagopalan S, Davis A, Giovannoni G. Cerebrospinal fluid oligoclonal bands in multiple sclerosis and clinically isolated syndromes: a meta-analysis of prevalence, prognosis and effect of latitude. Journal of Neurology, Neurosurgery and Psychiatry. 2013;84:909-914. Domingues RB, Fernandes GBP, Leite F, Tilbery CP, Thomaz RB, Silva GS, et al. The cerebrospinal fluid in multiple sclerosis: far beyond the bands. Einstein (Sao Paulo). 2017;15:100-104. Dorr J, Paul F. The transition from first-line to second-line therapy in multiple sclerosis. Current Treatment Options in Neurology. 2015;17:354. Dubey D, Kieseier BC, Hartung HP, Hemmer B, Warnke C, Menge T, et al. Dimethyl fumarate in relapsing-remitting multiple sclerosis: rationale, mechanisms of action, pharmacokinetics, efficacy and safety. Expert Review of Neurotherapeutics. 2015;15:339-346. Duffy SS, Lees JG, Moalem-Taylor G. The contribution of immune and glial cell types in experimental autoimmune encephalomyelitis and multiple sclerosis. Multiple Sclerosis International. 2014;2014:285245. Dutra RC, Moreira EL, Alberti TB, Marcon R, Prediger RD, Calixto JB. Spatial reference memory deficits precede motor dysfunction in an experimental autoimmune encephalomyelitis model: the role of kallikrein-kinin system. Brain, Behavior, and Immunity. 2013;33:90-101. Eilam R, Segal M, Malach R, Sela M, Arnon R, Aharoni R. Astrocyte disruption of neurovascular communication is linked to cortical damage in an animal model of multiple sclerosis. Glia. 2018;66:10981117. Emerson MR, LeVine SM. Heme oxygenase-1 and NADPH cytochrome P450 reductase expression in experimental allergic encephalomyelitis: an expanded view of the stress response. Journal of Neurochemistry. 2000;75:2555-2562. Eng LF, D'Amelio FE, Smith ME. Dissociation of GFAP intermediate filaments in EAE: observations in the lumbar spinal cord. Glia. 1989;2:308-317. Erta M, Giralt M, Jimenez S, Molinero A, Comes G, Hidalgo J. Astrocytic IL-6 influences the clinical symptoms of EAE in mice. Brain Sciences. 2016;6. 54 European Multiple Sclerosis Platform. MS Facts | MS Treatments. 11/03/2020. http://www.emsp.org/about-ms/ms-treatments/. Evangelou N, Esiri MM, Smith S, Palace J, Matthews PM. Quantitative pathological evidence for axonal loss in normal appearing white matter in multiple sclerosis. Annals of Neurology. 2000;47:391-395. Faissner S, Plemel JR, Gold R, Yong VW. Progressive multiple sclerosis: from pathophysiology to therapeutic strategies. Nature Reviews: Drug Discovery. 2019;18:905-922. Farias AS, Martins-de-Souza D, Guimaraes L, Pradella F, Moraes AS, Facchini G, et al. Proteome analysis of spinal cord during the clinical course of monophasic experimental autoimmune encephalomyelitis. Proteomics. 2012;12:2656-2662. Ferreira AC, Da Mesquita S, Sousa JC, Correia-Neves M, Sousa N, Palha JA, et al. From the periphery to the brain: lipocalin-2, a friend or foe? Progress in Neurobiology. 2015;131:120-136. Ferreira AC, Pinto V, Da Mesquita S, Novais A, Sousa JC, Correia-Neves M, et al. Lipocalin-2 is involved in emotional behaviors and cognitive function. Frontiers in Cellular Neuroscience. 2013;7:122. Ferreira AC, Santos T, Sampaio-Marques B, Novais A, Mesquita SD, Ludovico P, et al. Lipocalin-2 regulates adult neurogenesis and contextual discriminative behaviours. Molecular Psychiatry. 2018;23:1031-1039. Figueiredo J, Silva A, Cerqueira JJ, Fonseca J, Pereira PA. MS prevalence and patients' characteristics in the district of Braga, Portugal. Neurology Research International. 2015;2015:895163. Fischer JS, Priore RL, Jacobs LD, Cookfair DL, Rudick RA, Herndon RM, et al. Neuropsychological effects of interferon beta-1a in relapsing multiple sclerosis. Multiple Sclerosis Collaborative Research Group. Annals of Neurology. 2000;48:885-892. Flo TH, Smith KD, Sato S, Rodriguez DJ, Holmes MA, Strong RK, et al. Lipocalin 2 mediates an innate immune response to bacterial infection by sequestrating iron. Nature. 2004;432:917-921. Galloway DA, Gowing E, Setayeshgar S, Kothary R. Inhibitory milieu at the multiple sclerosis lesion site and the challenges for remyelination. Glia. 2020;68:859-877. Galloway DA, Williams JB, Moore CS. Effects of fumarates on inflammatory human astrocyte responses and oligodendrocyte differentiation. Annals of Clinical and Translational Neurology. 2017;4:381-391. Gavillet M, Allaman I, Magistretti PJ. Modulation of astrocytic metabolic phenotype by proinflammatory cytokines. Glia. 2008;56:975-989. 55 Gerwien H, Hermann S, Zhang X, Korpos E, Song J, Kopka K, et al. Imaging matrix metalloproteinase activity in multiple sclerosis as a specific marker of leukocyte penetration of the blood-brain barrier. Science Translational Medicine. 2016;8:364ra152. Geurts JJ, Barkhof F. Grey matter pathology in multiple sclerosis. Lancet Neurology. 2008;7:841-851. Gimenez MA, Sim JE, Russell JH. TNFR1-dependent VCAM-1 expression by astrocytes exposes the CNS to destructive inflammation. Journal of Neuroimmunology. 2004;151:116-125. Giovannoni G, Butzkueven H, Dhib-Jalbut S, Hobart J, Kobelt G, Pepper G, et al. Brain health: time matters in multiple sclerosis. Multiple Sclerosis and Related Disorders. 2016;9 Suppl 1:S5-S48. Glabinski AR, Tani M, Strieter RM, Tuohy VK, Ransohoff RM. Synchronous synthesis of alphaand beta-chemokines by cells of diverse lineage in the central nervous system of mice with relapses of chronic experimental autoimmune encephalomyelitis. The American Journal of Pathology. 1997;150:617-630. Glabinski AR, Tani M, Tuohy VK, Tuthill RJ, Ransohoff RM. Central nervous system chemokine mRNA accumulation follows initial leukocyte entry at the onset of acute murine experimental autoimmune encephalomyelitis. Brain, Behavior, and Immunity. 1995;9:315-330. Goetz DH, Holmes MA, Borregaard N, Bluhm ME, Raymond KN, Strong RK. The neutrophil lipocalin NGAL is a bacteriostatic agent that interferes with siderophore-mediated iron acquisition. Molecular Cell. 2002;10:1033-1043. Gonzalez-Cabrera PJ, Cahalan SM, Nguyen N, Sarkisyan G, Leaf NB, Cameron MD, et al. S1P(1) receptor modulation with cyclical recovery from lymphopenia ameliorates mouse model of multiple sclerosis. Molecular Pharmacology. 2012;81:166-174. Goodin DS. The epidemiology of multiple sclerosis: insights to disease pathogenesis. Handbook of Clinical Neurology. 2014;122:231-266. Graumann U, Reynolds R, Steck AJ, Schaeren-Wiemers N. Molecular changes in normal appearing white matter in multiple sclerosis are characteristic of neuroprotective mechanisms against hypoxic insult. Brain Pathology. 2003;13:554-573. Gresle MM, Schulz K, Jonas A, Perreau VM, Cipriani T, Baxter AG, et al. Ceruloplasmin gene-deficient mice with experimental autoimmune encephalomyelitis show attenuated early disease evolution. Journal of Neuroscience Research. 2014;92:732-742. Guerra-Gomes S, Cunha-Garcia D, Marques Nascimento DS, Duarte-Silva S, Loureiro-Campos E, Morais Sardinha V, et al. IP3 R2 null mice display a normal acquisition of somatic and neurological development milestones. European Journal of Neuroscience. 2020. 62 Ludwin SK, Rao V, Moore CS, Antel JP. Astrocytes in multiple sclerosis. Multiple Sclerosis. 2016;22:1114-1124. Luo J, Ho P, Steinman L, Wyss-Coray T. Bioluminescence in vivo imaging of autoimmune encephalomyelitis predicts disease. Journal of Neuroinflammation. 2008;5:6. Lutz NW, Viola A, Malikova I, Confort-Gouny S, Audoin B, Ranjeva JP, et al. Inflammatory multiplesclerosis plaques generate characteristic metabolic profiles in cerebrospinal fluid. PloS One. 2007;2:e595. Lycke J, Zetterberg H. The role of blood and CSF biomarkers in the evaluation of new treatments against multiple sclerosis. Expert Review of Clinical Immunology. 2017;13:1143-1153. Maimone D, Gregory S, Arnason BG, Reder AT. Cytokine levels in the cerebrospinal fluid and serum of patients with multiple sclerosis. Journal of Neuroimmunology. 1991;32:67-74. Malmestrom C, Andersson BA, Haghighi S, Lycke J. IL-6 and CCL2 levels in CSF are associated with the clinical course of MS: implications for their possible immunopathogenic roles. Journal of Neuroimmunology. 2006;175:176-182. Maragakis NJ, Rothstein JD. Mechanisms of Disease: astrocytes in neurodegenerative disease. Nature Clinical Practice: Neurology. 2006;2:679-689. Marques F, Mesquita SD, Sousa JC, Coppola G, Gao F, Geschwind DH, et al. Lipocalin 2 is present in the EAE brain and is modulated by natalizumab. Frontiers in Cellular Neuroscience. 2012;6:33. Marques F, Rodrigues AJ, Sousa JC, Coppola G, Geschwind DH, Sousa N, et al. Lipocalin 2 is a choroid plexus acute-phase protein. Journal of Cerebral Blood Flow and Metabolism. 2008;28:450-455. Marques KB, Scorisa JM, Zanon R, Freria CM, Santos LM, Damasceno BP, et al. The immunomodulator glatiramer acetate influences spinal motoneuron plasticity during the course of multiple sclerosis in an animal model. Brazilian Journal of Medical and Biological Research. 2009;42:179188. Marrie RA. Comorbidity in multiple sclerosis: implications for patient care. Nature Reviews: Neurology. 2017;13:375-382. Martinez MA, Olsson B, Bau L, Matas E, Cobo Calvo A, Andreasson U, et al. Glial and neuronal markers in cerebrospinal fluid predict progression in multiple sclerosis. Multiple Sclerosis. 2015;21:550-561. Mastronardi FG, Min W, Wang H, Winer S, Dosch M, Boggs JM, et al. Attenuation of experimental autoimmune encephalomyelitis and nonimmune demyelination by IFN-beta plus vitamin B12: treatment to modify notch-1/sonic hedgehog balance. Journal of Immunology. 2004;172:6418-6426. 63 Masvekar R, Wu T, Kosa P, Barbour C, Fossati V, Bielekova B. Cerebrospinal fluid biomarkers link toxic astrogliosis and microglial activation to multiple sclerosis severity. Multiple Sclerosis and Related Disorders. 2019;28:34-43. Matsumoto Y, Watanabe S, Suh YH, Yamamoto T. Effects of intrahippocampal CT105, a carboxyl terminal fragment of beta-amyloid precursor protein, alone/with inflammatory cytokines on working memory in rats. Journal of Neurochemistry. 2002;82:234-239. Mattioli F, Stampatori C, Bellomi F, Scarpazza C, Capra R. Natalizumab significantly improves cognitive impairment over three years in MS: pattern of disability progression and preliminary MRI findings. PloS One. 2015;10:e0131803. Mattioli F, Stampatori C, Capra R. The effect of natalizumab on cognitive function in patients with relapsing-remitting multiple sclerosis: preliminary results of a 1-year follow-up study. Neurological Sciences. 2011;32:83-88. Micera A, Vigneti E, Aloe L. Changes of NGF presence in nonneuronal cells in response to experimental allergic encephalomyelitis in Lewis rats. Experimental Neurology. 1998;154:41-46. Miljkovic D, Blazevski J, Petkovic F, Djedovic N, Momcilovic M, Stanisavljevic S, et al. A comparative analysis of multiple sclerosis-relevant anti-inflammatory properties of ethyl pyruvate and dimethyl fumarate. Journal of Immunology. 2015;194:2493-2503. Miljkovic D, Timotijevic G, Mostarica Stojkovic M. Astrocytes in the tempest of multiple sclerosis. FEBS Letters. 2011;585:3781-3788. Mills Ko E, Ma JH, Guo F, Miers L, Lee E, Bannerman P, et al. Deletion of astroglial CXCL10 delays clinical onset but does not affect progressive axon loss in a murine autoimmune multiple sclerosis model. Journal of Neuroinflammation. 2014;11:105. Miyagishi R, Kikuchi S, Takayama C, Inoue Y, Tashiro K. Identification of cell types producing RANTES, MIP-1 alpha and MIP-1 beta in rat experimental autoimmune encephalomyelitis by in situ hybridization. Journal of Neuroimmunology. 1997;77:17-26. Mizee MR, Nijland PG, van der Pol SM, Drexhage JA, van Het Hof B, Mebius R, et al. Astrocyte-derived retinoic acid: a novel regulator of blood-brain barrier function in multiple sclerosis. Acta Neuropathologica. 2014;128:691-703. Moreno M, Bannerman P, Ma J, Guo F, Miers L, Soulika AM, et al. Conditional ablation of astroglial CCL2 suppresses CNS accumulation of M1 macrophages and preserves axons in mice with MOG peptide EAE. Journal of Neuroscience. 2014;34:8175-8185. 64 Moriguchi K, Miyamoto K, Fukumoto Y, Kusunoki S. 4-Aminopyridine ameliorates relapsing remitting experimental autoimmune encephalomyelitis in SJL/J mice. Journal of Neuroimmunology. 2018;323:131-135. Mossakowski AA, Pohlan J, Bremer D, Lindquist R, Millward JM, Bock M, et al. Tracking CNS and systemic sources of oxidative stress during the course of chronic neuroinflammation. Acta Neuropathologica. 2015;130:799-814. Muckschel M, Beste C, Ziemssen T. Immunomodulatory treatments and cognition in MS. Acta Neurologica Scandinavica. 2016;134 Suppl 200:55-59. Mueller AM, Pedre X, Stempfl T, Kleiter I, Couillard-Despres S, Aigner L, et al. Novel role for SLPI in MOG-induced EAE revealed by spinal cord expression analysis. Journal of Neuroinflammation. 2008;5:20. Munji RN, Soung AL, Weiner GA, Sohet F, Semple BD, Trivedi A, et al. Profiling the mouse brain endothelial transcriptome in health and disease models reveals a core blood-brain barrier dysfunction module. Nature Neuroscience. 2019;22:1892-1902. Musella A, Mandolesi G, Mori F, Gentile A, Centonze D. Linking synaptopathy and gray matter damage in multiple sclerosis. Multiple Sclerosis. 2016;22:146-149. Mycko MP, Papoian R, Boschert U, Raine CS, Selmaj KW. cDNA microarray analysis in multiple sclerosis lesions: detection of genes associated with disease activity. Brain. 2003;126:1048-1057. Myers KJ, Dougherty JP, Ron Y. In vivo antigen presentation by both brain parenchymal cells and hematopoietically derived cells during the induction of experimental autoimmune encephalomyelitis. Journal of Immunology. 1993;151:2252-2260. Myhr KM, Mellgren SI. Corticosteroids in the treatment of multiple sclerosis. Acta neurologica Scandinavica Supplementum. 2009:73-80. Nair A, Frederick TJ, Miller SD. Astrocytes in multiple sclerosis: a product of their environment. Cellular and Molecular Life Sciences. 2008;65:2702-2720. Nam Y, Kim JH, Seo M, Kim JH, Jin M, Jeon S, et al. Lipocalin-2 protein deficiency ameliorates experimental autoimmune encephalomyelitis: the pathogenic role of lipocalin-2 in the central nervous system and peripheral lymphoid tissues. Journal of Biological Chemistry. 2014;289:16773-16789. Nataf S, Davoust N, Barnum SR. Kinetics of anaphylatoxin C5a receptor expression during experimental allergic encephalomyelitis. Journal of Neuroimmunology. 1998;91:147-155. Navikas V, Link J, Palasik W, Soderstrom M, Fredrikson S, Olsson T, et al. Increased mRNA expression of IL-10 in mononuclear cells in multiple sclerosis and optic neuritis. Scandinavian Journal of Immunology. 1995;41:171-178. 65 Nedergaard M, Rodriguez JJ, Verkhratsky A. Glial calcium and diseases of the nervous system. Cell Calcium. 2010;47:140-149. Nicholas R, Rashid W. Multiple sclerosis. American Family Physician. 2013;87:712-714. Nicot A, Ratnakar PV, Ron Y, Chen CC, Elkabes S. Regulation of gene expression in experimental autoimmune encephalomyelitis indicates early neuronal dysfunction. Brain. 2003;126:398-412. Nikcevich KM, Gordon KB, Tan L, Hurst SD, Kroepfl JF, Gardinier M, et al. IFN-gamma-activated primary murine astrocytes express B7 costimulatory molecules and prime naive antigen-specific T cells. Journal of Immunology. 1997;158:614-621. Noseworthy JH, Lucchinetti C, Rodriguez M, Weinshenker BG. Multiple sclerosis. The New England Journal of Medicine. 2000;343:938-952. Nylander A, Hafler DA. Multiple sclerosis. Journal of Clinical Investigation. 2012;122:1180-1188. O'Connor KC, Appel H, Bregoli L, Call ME, Catz I, Chan JA, et al. Antibodies from inflamed central nervous system tissue recognize myelin oligodendrocyte glycoprotein. Journal of Immunology. 2005;175:1974-1982. Okubo Y, Kanemaru K, Suzuki J, Kobayashi K, Hirose K, Iino M. Inositol 1,4,5-trisphosphate receptor type 2-independent Ca(2+) release from the endoplasmic reticulum in astrocytes. Glia. 2019;67:113124. Olesen J, Gustavsson A, Svensson M, Wittchen HU, Jonsson B, group Cs, et al. The economic cost of brain disorders in Europe. European Journal of Neurology. 2012;19:155-162. Opdenakker G, Proost P, Van Damme J. Microbiomic and posttranslational modifications as preludes to autoimmune diseases. Trends in Molecular Medicine. 2016;22:746-757. Ortiz GG, Pacheco-Moises FP, Macias-Islas MA, Flores-Alvarado LJ, Mireles-Ramirez MA, GonzalezRenovato ED, et al. Role of the blood-brain barrier in multiple sclerosis. Archives of Medical Research. 2014;45:687-697. Pagenstecher A, Stalder AK, Kincaid CL, Shapiro SD, Campbell IL. Differential expression of matrix metalloproteinase and tissue inhibitor of matrix metalloproteinase genes in the mouse central nervous system in normal and inflammatory states. The American Journal of Pathology. 1998;152:729-741. Park C, Ponath G, Levine-Ritterman M, Bull E, Swanson EC, De Jager PL, et al. The landscape of myeloid and astrocyte phenotypes in acute multiple sclerosis lesions. Acta Neuropathologica Communications. 2019;7:130. Paul A, Comabella M, Gandhi R. Biomarkers in multiple sclerosis. Cold Spring Harbor Perspectives in Medicine. 2019;9. 66 Paul D, Ge S, Lemire Y, Jellison ER, Serwanski DR, Ruddle NH, et al. Cell-selective knockout and 3D confocal image analysis reveals separate roles for astrocyte-and endothelial-derived CCL2 in neuroinflammation. Journal of Neuroinflammation. 2014;11:10. Pekny M, Nilsson M. Astrocyte activation and reactive gliosis. Glia. 2005;50:427-434. Pekny M, Pekna M. Reactive gliosis in the pathogenesis of CNS diseases. Biochimica et Biophysica Acta. 2016;1862:483-491. Petravicz J, Fiacco TA, McCarthy KD. Loss of IP3 receptor-dependent Ca2+ increases in hippocampal astrocytes does not affect baseline CA1 pyramidal neuron synaptic activity. Journal of Neuroscience. 2008;28:4967-4973. Petzold A, Eikelenboom MJ, Gveric D, Keir G, Chapman M, Lazeron RH, et al. Markers for different glial cell responses in multiple sclerosis: clinical and pathological correlations. Brain. 2002;125:14621473. Pitt D, Werner P, Raine CS. Glutamate excitotoxicity in a model of multiple sclerosis. Nature Medicine. 2000;6:67-70. Planas R, Santos R, Tomas-Ojer P, Cruciani C, Lutterotti A, Faigle W, et al. GDP-l-fucose synthase is a CD4(+) T cell-specific autoantigen in DRB3*02:02 patients with multiple sclerosis. Science Translational Medicine. 2018;10. Ponath G, Park C, Pitt D. The role of astrocytes in multiple sclerosis. Frontiers in Immunology. 2018;9:217. Popescu BF, Pirko I, Lucchinetti CF. Pathology of multiple sclerosis: where do we stand? Continuum (Minneap Minn). 2013;19:901-921. Prat A, Biernacki K, Wosik K, Antel JP. Glial cell influence on the human blood-brain barrier. Glia. 2001;36:145-155. Proescholdt MA, Jacobson S, Tresser N, Oldfield EH, Merrill MJ. Vascular endothelial growth factor is expressed in multiple sclerosis plaques and can induce inflammatory lesions in experimental allergic encephalomyelitis rats. Journal of Neuropathology and Experimental Neurology. 2002;61:914-925. Prokopova B, Hlavacova N, Vlcek M, Penesova A, Grunnerova L, Garafova A, et al. Early cognitive impairment along with decreased stress-induced BDNF in male and female patients with newly diagnosed multiple sclerosis. Journal of Neuroimmunology. 2017;302:34-40. Qi X, Guy J, Nick H, Valentine J, Rao N. Increase of manganese superoxide dismutase, but not of Cu/Zn-SOD, in experimental optic neuritis. Investigative Ophthalmology and Visual Science. 1997;38:1203-1212. 67 Rafiee Zadeh A, Askari M, Azadani NN, Ataei A, Ghadimi K, Tavoosi N, et al. Mechanism and adverse effects of multiple sclerosis drugs: a review article. Part 1. International Journal of Physiology, Pathophysiology and Pharmacology. 2019a;11:95-104. Rafiee Zadeh A, Ghadimi K, Ataei A, Askari M, Sheikhinia N, Tavoosi N, et al. Mechanism and adverse effects of multiple sclerosis drugs: a review article. Part 2. International Journal of Physiology, Pathophysiology and Pharmacology. 2019b;11:105-114. Rakers C, Petzold GC. Astrocytic calcium release mediates peri-infarct depolarizations in a rodent stroke model. Journal of Clinical Investigation. 2017;127:511-516. Ramanathan M, Weinstock-Guttman B, Nguyen LT, Badgett D, Miller C, Patrick K, et al. In vivo gene expression revealed by cDNA arrays: the pattern in relapsing-remitting multiple sclerosis patients compared with normal subjects. Journal of Neuroimmunology. 2001;116:213-219. Ransohoff RM, Hamilton TA, Tani M, Stoler MH, Shick HE, Major JA, et al. Astrocyte expression of mRNA encoding cytokines IP-10 and JE/MCP-1 in experimental autoimmune encephalomyelitis. FASEB Journal. 1993;7:592-600. Reich DS, Lucchinetti CF, Calabresi PA. Multiple sclerosis. The New England Journal of Medicine. 2018;378:169-180. Reichenbach N, Delekate A, Breithausen B, Keppler K, Poll S, Schulte T, et al. P2Y1 receptor blockade normalizes network dysfunction and cognition in an Alzheimer's disease model. Journal of Experimental Medicine. 2018;215:1649-1663. Reiman R, Campos Torres A, Martin BK, Ting JP, Campbell IL, Barnum SR. Expression of C5a in the brain does not exacerbate experimental autoimmune encephalomyelitis. Neuroscience Letters. 2005;390:134-138. Rejdak K, Petzold A, Kocki T, Kurzepa J, Grieb P, Turski WA, et al. Astrocytic activation in relation to inflammatory markers during clinical exacerbation of relapsing-remitting multiple sclerosis. Journal of Neural Transmission. 2007;114:1011-1015. Rojas M, Restrepo-Jimenez P, Monsalve DM, Pacheco Y, Acosta-Ampudia Y, Ramirez-Santana C, et al. Molecular mimicry and autoimmunity. Journal of Autoimmunity. 2018;95:100-123. Rosenling T, Stoop MP, Attali A, van Aken H, Suidgeest E, Christin C, et al. Profiling and identification of cerebrospinal fluid proteins in a rat EAE model of multiple sclerosis. Journal of Proteome Research. 2012;11:2048-2060. 68 Rossi S, Motta C, Studer V, Barbieri F, Buttari F, Bergami A, et al. Tumor necrosis factor is elevated in progressive multiple sclerosis and causes excitotoxic neurodegeneration. Multiple Sclerosis. 2014;20:304-312. Rothhammer V, Borucki DM, Tjon EC, Takenaka MC, Chao CC, Ardura-Fabregat A, et al. Microglial control of astrocytes in response to microbial metabolites. Nature. 2018;557:724-728. Rothhammer V, Kenison JE, Tjon E, Takenaka MC, de Lima KA, Borucki DM, et al. Sphingosine 1phosphate receptor modulation suppresses pathogenic astrocyte activation and chronic progressive CNS inflammation. Proceedings of the National Academy of Sciences of the United States of America. 2017;114:2012-2017. Rothhammer V, Mascanfroni ID, Bunse L, Takenaka MC, Kenison JE, Mayo L, et al. Type I interferons and microbial metabolites of tryptophan modulate astrocyte activity and central nervous system inflammation via the aryl hydrocarbon receptor. Nature Medicine. 2016;22:586-597. Rungta RL, Bernier LP, Dissing-Olesen L, Groten CJ, LeDue JM, Ko R, et al. Ca(2+) transients in astrocyte fine processes occur via Ca(2+) influx in the adult mouse hippocampus. Glia. 2016;64:20932103. Sa MJ, Kobelt G, Berg J, Capsa D, Dalen J, European Multiple Sclerosis P. New insights into the burden and costs of multiple sclerosis in Europe: results for Portugal. Multiple Sclerosis. 2017;23:143154. Sabatino JJ, Jr., Probstel AK, Zamvil SS. B cells in autoimmune and neurodegenerative central nervous system diseases. Nature Reviews Neuroscience. 2019;20:728-745. Salmen A, Gold R. Mode of action and clinical studies with fumarates in multiple sclerosis. Experimental Neurology. 2014;262 Pt A:52-56. Santello M, Toni N, Volterra A. Astrocyte function from information processing to cognition and cognitive impairment. Nature Neuroscience. 2019;22:154-166. Satoh JI, Tabunoki H, Yamamura T. Molecular network of the comprehensive multiple sclerosis brainlesion proteome. Multiple Sclerosis. 2009;15:531-541. Scannevin RH, Chollate S, Jung MY, Shackett M, Patel H, Bista P, et al. Fumarates promote cytoprotection of central nervous system cells against oxidative stress via the nuclear factor (erythroidderived 2)-like 2 pathway. Journal of Pharmacology and Experimental Therapeutics. 2012;341:274-284. Scemes E, Giaume C. Astrocyte calcium waves: what they are and what they do. Glia. 2006;54:716725. 69 Schilling S, Goelz S, Linker R, Luehder F, Gold R. Fumaric acid esters are effective in chronic experimental autoimmune encephalomyelitis and suppress macrophage infiltration. Clinical and Experimental Immunology. 2006;145:101-107. Schirmer L, Velmeshev D, Holmqvist S, Kaufmann M, Werneburg S, Jung D, et al. Neuronal vulnerability and multilineage diversity in multiple sclerosis. Nature. 2019;573:75-82. Schrempf W, Ziemssen T. Glatiramer acetate: mechanisms of action in multiple sclerosis. Autoimmunity Reviews. 2007;6:469-475. Sevastou I, Pryce G, Baker D, Selwood DL. Characterisation of transcriptional changes in the spinal cord of the progressive experimental autoimmune encephalomyelitis Biozzi ABH mouse model by RNA sequencing. PloS One. 2016;11:e0157754. Sharp AH, Nucifora FC, Jr., Blondel O, Sheppard CA, Zhang C, Snyder SH, et al. Differential cellular expression of isoforms of inositol 1,4,5-triphosphate receptors in neurons and glia in brain. Journal of Comparative Neurology. 1999;406:207-220. Shields DC, Tyor WR, Deibler GE, Banik NL. Increased calpain expression in experimental demyelinating optic neuritis: an immunocytochemical study. Brain Research. 1998;784:299-304. Shin T. Enhanced expression of constitutive endothelial nitric oxide synthase by astrocytes in the spinal cords of rats with experimental autoimmune encephalomyelitis. Immunological Investigations. 1999;28:381-390. Shin T. Increased expression of neuronal nitric oxide synthase in astrocytes and macrophages in the spinal cord of Lewis rats with autoimmune encephalomyelitis. Journal of Veterinary Science. 2001;2:195199. Shrestha B, Ge S, Pachter JS. Resolution of central nervous system astrocytic and endothelial sources of CCL2 gene expression during evolving neuroinflammation. Fluids and Barriers of the CNS. 2014;11:6. Simone IL, Federico F, Trojano M, Tortorella C, Liguori M, Giannini P, et al. High resolution proton MR spectroscopy of cerebrospinal fluid in MS patients. Comparison with biochemical changes in demyelinating plaques. Journal of the Neurological Sciences. 1996;144:182-190. Smith KJ, Pyrdol J, Gauthier L, Wiley DC, Wucherpfennig KW. Crystal structure of HLA-DR2 (DRA*0101, DRB1*1501) complexed with a peptide from human myelin basic protein. Journal of Experimental Medicine. 1998;188:1511-1520. Smith PA, Schmid C, Zurbruegg S, Jivkov M, Doelemeyer A, Theil D, et al. Fingolimod inhibits brain atrophy and promotes brain-derived neurotrophic factor in an animal model of multiple sclerosis. Journal of Neuroimmunology. 2018;318:103-113. 70 Sofroniew MV. Multiple roles for astrocytes as effectors of cytokines and inflammatory mediators. Neuroscientist. 2014;20:160-172. Sofroniew MV. Astrocyte barriers to neurotoxic inflammation. Nature Reviews Neuroscience. 2015;16:249-263. Sofroniew MV, Vinters HV. Astrocytes: biology and pathology. Acta Neuropathologica. 2010;119:7-35. Song J, Wu C, Korpos E, Zhang X, Agrawal SM, Wang Y, et al. Focal MMP-2 and MMP-9 activity at the blood-brain barrier promotes chemokine-induced leukocyte migration. Cell Reports. 2015;10:1040-1054. Srinivasan R, Huang BS, Venugopal S, Johnston AD, Chai H, Zeng H, et al. Ca(2+) signaling in astrocytes from Ip3r2(-/-) mice in brain slices and during startle responses in vivo. Nature Neuroscience. 2015;18:708-717. Staats KA, Humblet-Baron S, Bento-Abreu A, Scheveneels W, Nikolaou A, Deckers K, et al. Genetic ablation of IP3 receptor 2 increases cytokines and decreases survival of SOD1G93A mice. Human Molecular Genetics. 2016;25:3491-3499. Stadelmann C, Kerschensteiner M, Misgeld T, Bruck W, Hohlfeld R, Lassmann H. BDNF and gp145trkB in multiple sclerosis brain lesions: neuroprotective interactions between immune and neuronal cells? Brain. 2002;125:75-85. Steinman L. Multiple sclerosis: a two-stage disease. Nature Immunology. 2001;2:762-764. Stover JF, Lowitzsch K, Kempski OS. Cerebrospinal fluid hypoxanthine, xanthine and uric acid levels may reflect glutamate-mediated excitotoxicity in different neurological diseases. Neuroscience Letters. 1997a;238:25-28. Stover JF, Pleines UE, Morganti-Kossmann MC, Kossmann T, Lowitzsch K, Kempski OS. Neurotransmitters in cerebrospinal fluid reflect pathological activity. European Journal of Clinical Investigation. 1997b;27:1038-1043. Stys PK, Zamponi GW, van Minnen J, Geurts JJ. Will the real multiple sclerosis please stand up? Nature Reviews Neuroscience. 2012;13:507-514. Sunnemark D, Eltayeb S, Nilsson M, Wallstrom E, Lassmann H, Olsson T, et al. CX3CL1 (fractalkine) and CX3CR1 expression in myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalomyelitis: kinetics and cellular origin. Journal of Neuroinflammation. 2005;2:17. Tan W, Pu Y, Shao Q, Fang X, Han D, Zhao M, et al. Insulin-like growth factor-binding protein 7 is upregulated during EAE and inhibits the differentiation of oligodendrocyte precursor cells. Biochemical and Biophysical Research Communications. 2015;460:639-644. 71 Tassoni A, Farkhondeh V, Itoh Y, Itoh N, Sofroniew MV, Voskuhl RR. The astrocyte transcriptome in EAE optic neuritis shows complement activation and reveals a sex difference in astrocytic C3 expression. Scientific Reports. 2019;9:10010. Teesalu T, Hinkkanen AE, Vaheri A. Coordinated induction of extracellular proteolysis systems during experimental autoimmune encephalomyelitis in mice. The American Journal of Pathology. 2001;159:2227-2237. Tejera-Alhambra M, Casrouge A, de Andres C, Seyfferth A, Ramos-Medina R, Alonso B, et al. Plasma biomarkers discriminate clinical forms of multiple sclerosis. PloS One. 2015;10:e0128952. Thompson AJ, Banwell BL, Barkhof F, Carroll WM, Coetzee T, Comi G, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurology. 2018;17:162-173. Tilleux S, Hermans E. Neuroinflammation and regulation of glial glutamate uptake in neurological disorders. Journal of Neuroscience Research. 2007;85:2059-2070. Toft-Hansen H, Fuchtbauer L, Owens T. Inhibition of reactive astrocytosis in established experimental autoimmune encephalomyelitis favors infiltration by myeloid cells over T cells and enhances severity of disease. Glia. 2011;59:166-176. Tran EH, Hardin-Pouzet H, Verge G, Owens T. Astrocytes and microglia express inducible nitric oxide synthase in mice with experimental allergic encephalomyelitis. Journal of Neuroimmunology. 1997;74:121-129. Treumer F, Zhu K, Glaser R, Mrowietz U. Dimethylfumarate is a potent inducer of apoptosis in human T cells. Journal of Investigative Dermatology. 2003;121:1383-1388. Ure DR, Rodriguez M. Polyreactive antibodies to glatiramer acetate promote myelin repair in murine model of demyelinating disease. FASEB Journal. 2002;16:1260-1262. Van Doorn R, Van Horssen J, Verzijl D, Witte M, Ronken E, Van Het Hof B, et al. Sphingosine 1phosphate receptor 1 and 3 are upregulated in multiple sclerosis lesions. Glia. 2010;58:1465-1476. Vardjan N, Zorec R. Excitable astrocytes: Ca(2+)- and cAMP-regulated exocytosis. Neurochemical Research. 2015;40:2414-2424. Vercellino M, Trebini C, Capello E, Mancardi GL, Giordana MT, Cavalla P. Inflammatory responses in multiple sclerosis normal-appearing white matter and in non-immune mediated neurological conditions with wallerian axonal degeneration: a comparative study. Journal of Neuroimmunology. 2017;312:49-58. Villarroya H, Klein C, Thillaye-Goldenberg B, Eclancher F. Distribution in ocular structures and optic pathways of immunocompetent and glial cells in an experimental allergic encephalomyelitis (EAE) relapsing model. Journal of Neuroscience Research. 2001;63:525-535. 78 Introduction The term astrogliosis has been use for a long time and was first applied to designate pronounced structural changes occurring in astrocytes in response to CNS damage and disease (Sofroniew, 2014). Some astrogliosis features are transversal to several pathologies, like upregulation of glial fibrillary acidic protein (GFAP) and cellular hypertrophy, however these alterations can occur in different degrees, varying from small to very intense changes (Sofroniew, 2014). Of interest, the presence of highly abnormal astrocytes in multiple sclerosis (MS) lesions is known for a long time, namely, hypertrophic astrocytes were found both in acute and chronic active lesions, while chronic silent lesions display astroglial scar tissue (Correale and Farez, 2015; Frohman et al., 2006; Williams et al., 2007; Wu and Alvarez, 2011). MS is a chronic inflammatory neurodegenerative disease of the central nervous system (CNS) with unknown cause and diverse pathophysiological mechanisms (Negrotto and Correale, 2017; Noseworthy, 1999; Noseworthy et al., 2000). Several studies have also shown the presence of reactive astrocytes in the experimental autoimmune encephalomyelitis (EAE) model, which is an animal model of immunemediated demyelination (Aquino et al., 1990; Luo et al., 2008; Smith and Eng, 1987; Smith et al., 1983; Tani et al., 1996). Still, the role of astrocytes, and the alterations occurring in these cells throughout disease development are not fully understood. Previous transcriptomic studies have demonstrated variability in astrocytic gene expression among different CNS regions in response to EAE induction, at the chronic phase of disease (days 45-50 post-disease induction) (Itoh et al., 2018; Tassoni et al., 2019). Nevertheless, commonly upregulated genes are associated with immune-related pathways, like antigen presentation and interferon (IFN) signaling pathways, while cholesterol synthesis pathways were downregulated, compared with control animals (Itoh et al., 2018; Tassoni et al., 2019). Moreover, Chao and colleagues (2019) recently described that alterations in immunometabolic pathways of astrocytes contribute to a pro-inflammatory profile of these cells in EAE animals (Chao et al., 2019). In the present work, we studied the alterations occurring in astrocytes along disease development, by looking at different disease time points, namely the pre-symptomatic, the onset (peak of disease) and chronic phases. To do so, we studied the morphology and transcriptomic profile of astrocytes from the cerebellum, a CNS region widely affected in MS and EAE (das Neves et al., 2018; Du et al., 2019; Itoh et al., 2018; MacKenzie-Graham et al., 2006; Saab et al., 2004; Schreck et al., 2018). Interestingly, at the onset phase of disease, astrocytes presented a significant overexpression of genes associated with a neurotoxic phenotype (Liddelow et al., 2017) and others involved in metabolic pathways. 79 Methods Animals and EAE induction All experiments were reviewed and approved by the Portuguese national authority for animal experimentation, Direcção Geral de Veterinária (ID: DGV9458). Animals were housed and handled in accordance with the guidelines for the care and handling of laboratory animals in the Directive 2010/63/EU of the European Parliament and the Council. Animals were housed under specific-pathogen-free conditions and maintained under standard laboratory settings: 12 h light/dark cycles (lights on at 8 a.m.), relative humidity of 55%, temperature between 22– 24ºC, and fed with regular rodent chow (4RF21, Mucedola SRL) and tap water ad libitum . Disease was induced in 10-weeks old female C57BL/6J mice, purchased from Charles River Laboratories (France) using a commercial kit (EK-2110; Hooke Laboratories, Lawrence, MA, USA) according to the manufacturer’s instructions. Briefly, animals were immunized subcutaneously with 200 μg of myelin oligodendrocyte glycoprotein (MOG)35−55, emulsified in complete Freund’s adjuvant (CFA), at the upper and lower back. Pertussis toxin (PTX) in phosphate buffered saline (PBS) was administered intraperitoneally 2 and 24 h after immunization [136 ng of PTX per injection (lot #1006)]. Non-induced age-matched littermate females were used as controls and were injected subcutaneously with a control emulsion and PTX (CK-2110; Hooke Laboratories) at the same concentration and time points as the EAE animals. Animals were daily weighted and monitored for clinical symptoms of disease. Disease severity was assessed daily as follows: 0 = no clinical symptoms; 0.5 = partially limp tail; 1 = paralyzed tail; 1.5 = at least one hind limb falls through consistently when the animal is placed on a wire rack; 2 = loss in coordinated movement, wobbly walk; 2.5 = dragging of hind limbs; 3 = paralysis of both hind limbs; 3.5 = hind limbs paralyzed and weakness of forelimbs; 4 = complete hind limbs paralysis and partial forelimbs paralysis; 4.5 = animal is not alert, no movement; 5 = moribund state or death. Paralyzed mice, with clinical scores above 3, were offered easier access to food and water. For biological sample collection, groups of EAE and non-induced animals were sacrificed, at the light phase of the diurnal cycle, at day 6 post-immunization (p.i.) (pre-symptomatic phase), on the first day of a clinical score of 3 (onset/peak phase; days 9-12), and on days 20-21 p.i. (chronic phase). Animals were anesthetized with an intraperitoneal injection of ketamine hydrochloride (150 mg/kg, Imalgene 1000) plus medetomidine hydrochloride (0.3 mg/kg, Dorben). Under deep anesthesia, mice were transcardially perfused with cold 0.9% saline solution, and the brain was dissected. For histological analysis, the brain was immediately embedded in Tissue-Tek O.C.T. compound (Sakura Finetek, Japan), snap-frozen and kept frozen (-20ºC) until further sectioning. 80 Astrocyte isolation The cerebellum was macrodissected for astrocyte isolation using magnetic-activated cell sorting (MACS) (Holt and Olsen, 2016). Briefly, the cerebellum was mechanically dissociated using a scalpel, followed by enzymatic dissociation using the Neural Tissue Dissociation kit (P) (Miltenyi Biotec, Cologne, Germany), according to the manufacturer’s instructions. Myelin and cell debris were removed using the Myelin removal kit (Miltenyi Biotec), followed by microglia removal using Cd11b microbeads (Miltenyi Biotec). Finally, astrocytes were isolated using anti-astrocyte cell surface antigen (ACSA)-2 beads (Miltenyi Biotec). All the separation steps were performed using the AutoMACS Pro Separator equipment (Miltenyi Biotec). Gene expression analysis by RNA-sequencing Total RNA was extracted from isolated astrocytes using the RNeasy Plus Micro kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA quality and quantification were performed using the Experion RNA HighSens Analysis kit (Bio-Rad, CA, USA) according to the manufacturer’s instructions. RNA samples were sequenced at the UCLA Neuroscience Genomics Core. RNA-sequencing (RNAseq) was carried out using Nugen Ovation RNA Ultra Low Input and Kapa Hyper. The RNA samples were made into barcoded cDNA library and then sequenced at 2x75bp paired end reads output with Illumina HiSeq 4000. No read trimming or filtering was done with this dataset, because the quality distribution and variance appeared normal. Short reads were aligned using STAR to the mouse (mm10), with default parameters. Differential expression analysis was performed using observation based-model (limmavoom). Genes with counts per million (CPM) > 0.5 in at least 3 non-induced samples and adjusted pvalues < 0.05 were considered statistically significant and were used for differential expression analysis. Pathway analysis was performed using the ConsensusPathDB-mouse tool (Kamburov et al., 2011; Kamburov et al., 2009). A total of three non-induced and four EAE animals were sacrificed per experimental time point for the RNAseq experiments. Gene expression analysis by qRT-PCR Total RNA was extracted from isolated astrocytes using the RNeasy Plus Micro kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA quality and quantification were performed using the Experion RNA HighSens Analysis kit (Bio-Rad, CA, USA) according to the manufacturer’s instructions. 750 pg of total RNA from each sample were reverse transcribed into cDNA using the iScript cDNA synthesis kit (Bio-Rad), according to the manufacturer’s instructions. qRT-PCR was performed on 81 a CFX96 real-time instrument (Bio-Rad) using the SsoFast EvaGreen Supermix (Bio-Rad). For each reaction, 5 µL of reaction mix, 0.5 µL of each primer (initial concentration 10 µM), 3 µL of RNase/DNase free water and 1 µL of cDNA were used. The cycling parameters were 1 cycle at 95ºC, for 1 minute (min), followed by 40 cycles at 95ºC for 15 seconds (s), annealing temperature (primer specific) for 20 s and 72ºC for 20 s, finishing with 1 cycle at 65ºC to 95ºC for 5 s (melting curve). Product fluorescence was detected at the end of the elongation cycle. All melting curves exhibited a single sharp peak at the expected temperature. Adenosine Triphosphate subunit 5 beta ( Atp5b ), Heat Shock Protein 90 alpha family class B member 1 ( Hspcb ) and TATA binding protein ( Tbp ) were used as reference genes. Primers used to measure the expression levels of selected mRNA transcripts by qRT-PCR were designed using the PrimerBLAST tool of NCBI (Bethesda, MD, USA) on the basis of the respective GenBank accession numbers, or were used as described by Liddelow and colleagues (2017) (Liddelow et al., 2017). Primers DNA sequences and annealing temperatures are provided in Supplementary table 2. 1. GFAP immunofluorescence and 3-dimensional reconstruction of astrocytes Serial 20 µm sections of cerebellum were fixed in 4% paraformaldehyde in PBS for 30 min at room temperature (RT). After antigen retrieval, with pre-heated citrate buffer (Sigma-Aldrich) in the microwave for 20 min, tissue slices were permeabilized with PBS-triton 0.3% at RT for 10 min, and subsequently blocked with 10% fetal bovine serum in PBS-triton 0.3% at RT for 30 min. Slides were incubated overnight with rabbit anti-mouse GFAP antibody (1:200; Dako, Denmark), diluted in blocking solution. Afterwards, slides were incubated with Alexa Fluor® 594 donkey anti-rabbit (1:500; Fisher Technologies, Thermo Fisher Scientific), diluted in PBS-triton 0.3%, for 2 h at RT. After incubation with 4′,6-diamidino2-phenylindole (DAPI; 1:200; Invitrogen, Thermo Fisher Scientific), for 10 min at RT, slides were coverslipped with Immumount (Fisher Scientific, Thermo Fisher Scientific) and examined under fluorescent light. To perform the 3-dimensional reconstruction of astrocytes, 3-4 photographs per animal were taken from the cerebellum white matter, using a confocal microscope (FV1000, Olympus) and the following parameters: 40x objective, 1024x1024 resolution, 1 µm increment. In the case of EAE animals, photographs were acquired both near lesion regions and in regions of normal appearing white matter (NAWM), except for animals at the pre-symptomatic phase which did not present lesions. The confocal images were then used to performed the morphological reconstruction using the Fiji plugin “Simple Neurite Tracer” (Longair et al., 2011; Schindelin et al., 2012), as previously described (Tavares et al., 2017). For Sholl analysis, concentric circles were superimposed on astrocytes, with origin in the cell 82 soma and with 4 µm distance from each other. Results are presented as the average of 4-5 animals per experimental group, and 7-8 astrocytes were reconstructed per animal. Statistical analysis Statistical analysis was performed using SPSS software (version 23, IBM, USA) and GraphPad Prism (version 8, La Jolla, California, USA). The number of biological replicates (n) are specified in the legend of each figure. Results are presented as mean ± standard error of the mean (SEM), or only as mean for astrocytic sholl analysis. Statistical significance was considered for p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****). The partial eta squared value (ηp2) was calculated as a measure of effect size (Lakens, 2013). 83 Results Altered astrocytic gene expression in EAE animals To explore the alterations occurring in astrocytes throughout disease development, we performed transcriptomic analysis, by RNAseq, in astrocytes isolated from the cerebellum at different disease time points (Figure 2. 1A). The cerebellum is known to be affected both in the human disease and in the EAE model, and ACSA-2 was shown to be highly expressed in this region (Kantzer et al., 2017). To validate the specificity of our isolation protocol, in our RNAseq data we looked for the expression levels of genes associated with different CNS cell types and with different types of immune cells (Supplementary figure 2. 1 and Supplementary table 2. 2). We confirmed that this sorting method provided a cell population enriched in astrocytes, since the majority of astrocyte and Bergmann glia associated genes were highly expressed in our samples, while the majority of genes associated with other cell types were not expressed or presented low expression values. The ATPase Na+/K+ Transporting Subunit Beta 2 ( Atp1b2 ) gene, which was identified as being the ACSA-2 epitope (Batiuk et al., 2017), also presented high expression levels. PTX has been suggested to facilitate EAE development by increasing the BBB permeability, thereby facilitating the migration of pathogenic T cells to the CNS, among other important biological effects (Hofstetter et al., 2002; Turley and Miller, 2007). Considering that astrocytes play an important role in the maintenance of BBB permeability (Abbott et al., 2006; Allen and Barres, 2009), and that non-induced animals were also injected with PTX, we also sacrificed a group of non-induced animals at each of the experimental time points. In fact, we observed that several astrocytic genes were significantly altered in non-induced animals, when comparing the animals of the three experimental time points (Supplementary figure 2. 2A). Next, we compared gene expression values between EAE and non-induced animals, for each time point analyzed, and observed a total of 804 genes significantly altered at the pre-symptomatic phase, 470 genes altered at the onset phase and 296 genes altered at the chronic phase of disease (Figure 2. 1B). To explore the alterations occurring along disease development, we also compared the three EAE time points among each other (Supplementary figure 2. 2B). However, some of the differences found were also present in the comparison between time points in non-induced. For that reason, we next normalized each disease time point for the respective non-induced group, and verified which differences remained significant when comparing the EAE groups (Figure 2. 1C). Of interest, the chronic and pre-symptomatic phases of disease were more similar among each other than with the onset phase. 84 Figure 2. 1 – Sample collection for RNAseq analysis and differentially expressed genes. (A) Schematic overview of the methodology used to isolate astrocytes from the cerebellum and posterior RNAseq analysis; average clinical score of the animals used in the study. (B) Volcano plot and Venn diagram depicting the differentially expressed astrocytic genes after comparing EAE and non-induced animals for each experimental time point. (C) Volcano plot and Venn diagram depicting the differentially expressed genes after comparison between disease time points in EAE animals, normalized for the respective non-induced group. In the Volcano plots, significant underexpessed genes are represented in blue; significant overexpressed genes are represented in red; genes whose expression was not significantly altered are represented in gray. Images obtained and adapted from Servier Medical Art. CFA – complete Freund’s adjuvant; C – chronic time point; MACS – magnetic-activated cell sorting; MOG – myelin oligodendrocyte glycoprotein; O – onset time point; PBS – phosphate buffered saline; P – pre-symptomatic time point. Over-representation pathway analysis was performed for the differentially expressed genes between EAE and non-induced animals (Figure 2. 2A-C, Supplementary table 2. 3, Supplementary table 2. 4, 85 Supplementary table 2. 5) and between EAE time points (Figure 2. 2D-F, Supplementary table 2. 6, Supplementary table 2. 7, Supplementary table 2. 8). Interestingly, several pathways related with extracellular matrix organization were significantly altered in EAE animals at the pre-symptomatic phase of disease, while at the onset phase, the mostly represented pathways were involved in metabolic pathways. 86 Figure 2. 2 – Pathway analysis results. (A) Top 20 pathways altered in EAE animals at the pre-symptomatic, (B) onset and (C) chronic phases of disease, compared to non-induced animals. (D) Top 20 pathways altered at the onset vs . pre-symptomatic phases. (E) Top 20 pathways altered at the chronic vs . pre-symptomatic phases. (F) Top 20 pathways altered at the chronic vs . onset phases. 87 Astrocytic metabolic reprogramming at the onset phase of disease As already mentioned, several metabolic pathways were significantly altered at the onset phase of disease. Among these were genes involved in glycolysis and the tricarboxylic acid cycle (TCA) cycle (Figure 2. 3A). The overexpression of some of these genes was confirmed by qRT-PCR, in a different group of animals (Figure 2. 3B. Two-Way ANOVA – between subject factors: disease and sacrifice time point. Phosphofructokinase muscle ( Pfkm ): disease – F(1,22) = 0.0212, p = 0.8854; sacrifice time point – F(2,22) = 4.858, p = 0.0179, η2 = 0.297, Bonferroni’s multiple comparison test – p < 0.05 for onset vs . chronic phase; interaction - F(2,22) = 0.4698, p = 0.6312; fructose-biphosphate aldolase C ( Aldoc ): disease – F(1,22) = 0.3990, p = 0.5341; sacrifice time point – F(2,22) = 2.832, p = 0.0805; interaction - F(2,22) = 4.161, p = 0.0293, η2 = 0.229; isocitrate dehydrogenase 3 (NAD+) gamma ( Idh3g ): disease – F(1,22) = 6.639, p = 0.0172, η2 = 0.089; sacrifice time point – F(2,22) = 17.140, p < 0.0001, η2 = 0.462; interaction - F(2,22) = 5.664, p = 0.0104, η2 = 0.153; succinate dehydrogenase complex subunit A flavoprotein ( Sdha ): disease – F(1,22) = 0.9211, p = 0.3476; sacrifice time point – F(2,22) = 5.358, p = 0.0127, η2 = 0.309, Bonferroni’s multiple comparison test – p < 0.05 for onset vs . chronic phase; interaction - F(2,22) = 0.5214, p = 0.6009). Recently, Liddelow and colleagues (2017) characterized a population of reactive astrocytes, termed A1, induced by classically activated neuroinflammatory microglia (Liddelow et al., 2017). In our samples, we observed an overexpression of A1-specific genes in EAE animals, that was more evident at the onset phase of disease, while A2-specific genes did not present major alterations (Figure 2. 4A-B. Two-Way ANOVA – between subject factors: disease and sacrifice time point. Fibulin 5 ( Fbln5 ): disease – F(1,22) = 21.370, p = 0.0001, η2 = 0.214; sacrifice time point – F(2,22) = 12.670, p = 0.0002, η2 = 0.253; interaction - F(2,22) = 15.670, p < 0.0001, η2 = 0.313; adhesion molecule with Ig like domain 2 ( Amigo2 ): disease – F(1,22) = 15.070, p = 0.0008, η2 = 0.209; sacrifice time point – F(2,22) = 10.150, p = 0.0008, η2 = 0.281 , η2 = ; interaction - F(2,22) = 7.406, p = 0.0035, η2 = 0.205; FK506 binding protein 5 ( Fkbp5 ): disease – F(1,22) = 34.860, p < 0.0001, η2 = 0.326; sacrifice time point – F(2,22) = 9.893, p = 0.0009, η2 = 0.185; interaction - F(2,22) = 15.170, p < 0.0001, η2 = 0.284). Classically activated (M1) macrophages are characterized by an induction of aerobic glycolysis, that results in lactate production, and increased levels of TCA cycle intermediates (Galvan-Pena and O'Neill, 2014). Considering that similar metabolic alterations occur in astrocytes of EAE animals, and that these cells acquire a neurotoxic phenotype, it is possible that astrocytes could also undergo metabolic reprogramming in the EAE context. 94 Discussion Herein, we showed an astrocytic upregulation of glycolytic and TCA cycle regulatory genes in the MS animal model. Together with these alterations, we also detected an upregulation of genes associated with an astrocytic neurotoxic phenotype. How these metabolic alterations are linked with a pro-inflammatory and neurotoxic astrocytic phenotype and disease progression is unknown, but these results are in line with data from other fields, showing that a TCA cycle rate increase in monocytes is associated with a stronger immune response. Indeed, metabolic reprogramming is known to occur in macrophages and microglia in response to different stimuli. M1 (classically activated) macrophages/microglia are activated by bacterial-derived products, like LPS, and infection-associated signals, such as IFN gamma, and are usually part of the first line of defense of the innate immune system (Galvan-Pena and O'Neill, 2014; Orihuela et al., 2016). M1 polarized cells then produce pro-inflammatory cytokines and high levels of NO in order to kill the foreign pathogen and activate T cells to mount an adaptive immune response. A similar response can occur in the absence of microorganisms, as a result of trauma, ischemia-reperfusion injury or chemical exposure (Orihuela et al., 2016). On the other hand, M2 (alternatively activated) macrophages play an important role in the resolution phase of inflammation (Galvan-Pena and O'Neill, 2014; Orihuela et al., 2016), by producing anti-inflammatory factors that switch off pro-inflammatory cell phenotypes and re-establish homeostasis (Orihuela et al., 2016). Notably, the cells metabolic profile is a reflection of these functions. Namely, in M1 cells aerobic glycolysis is induced, to provide the cell with rapid energy, and the pentose phosphate pathway and respiratory chain functions are induced and attenuated, respectively, to increase the production of ROS and RNS (Galvan-Pena and O'Neill, 2014; Orihuela et al., 2016). M2 polarized cells need to be sustained for longer periods of time, so oxidative metabolism and fatty acid oxidation are induced (Galvan-Pena and O'Neill, 2014; Orihuela et al., 2016). Interestingly, the polarization for one of these phenotypes can be induced not only by the inflammatory factors already mentioned, but also by the modification of the cells metabolic state. Specifically, blocking oxidative metabolism drives macrophage polarization to an M1 state, while forcing it in M1 macrophages potentiates the M2 phenotype (Galvan-Pena and O'Neill, 2014). Of relevance, astrocytes were recently classified into “A1” (neurotoxic) and “A2” (neuroprotective), in analogy to the M1/M2 macrophage nomenclature (Liddelow et al., 2017). Liddelow and colleagues (2017) showed that neuroinflammation induced an A1 phenotype, characterized by the loss of normal astrocytic functions and the gain of a neurotoxic role, which was deleterious for both neurons and oligodendrocytes. Conversely, A2 astrocytes were induced by ischemic conditions and presented a neuroprotective phenotype (Liddelow et al., 2017). Additionally, C3+ A1 astrocytes were found to be 95 present in demyelinating lesions of MS patients (Liddelow et al., 2017). In accordance, we observed that astrocytes isolated from the cerebellum of EAE animals presented an A1 phenotype, particularly at the onset phase of disease. Moreover, several genes involved in metabolic pathways were altered during this time point, including glycolytic and TCA cycle genes. Previous studies had also reported alterations in other metabolic pathways, namely sphingolipid metabolism and cholesterol biosynthesis, in astrocytes during EAE, which we also observed here (Chao et al., 2019; Itoh et al., 2018; Mayo et al., 2014; Mueller et al., 2008; Sevastou et al., 2016; Tassoni et al., 2019). Considering all these data, and the fact that metabolic alterations are associated with M1/M2 polarization, we hypothesize that similar metabolic alterations could occur in A1/A2 astrocytes. To further explore this hypothesis, we are currently using primary astrocytic cultures, isolated by MACS from the cerebellum of P5-P7 pups, that are being stimulated with IL-1α, TNFα and C1q, to induce an A1 phenotype. Then, we will move to the characterization of their metabolic profile, focusing on the expression levels of genes involved in glycolysis and TCA cycle. Interestingly, a recent in vitro study has demonstrated differential metabolic profiles in astrocytes exposed to LPS for short or long periods of time. Namely, 30 min of LPS treatment increased the astrocytic glycolytic rate, but did not affect their oxidative phosphorylation rate, while treatment with LPS for 24 hours decreased the glycolytic capacity and increased the mitochondrial respiration of astrocytes (Robb et al., 2020). This study suggests that proinflammatory stimuli are able to modulate the metabolic profile of astrocytes, supporting the findings observed in our work. In this work we also performed for the first time a morphological analysis of astrocytes of the cerebellum white matter in different disease time points. As already expected, astrocytes near lesion regions, at the onset and chronic phases of disease, were more reactive than astrocytes from non-induced animals, as evaluated by increased total length and number of ramifications. Regarding the NAWM, previous studies reported the presence of gliosis in MS patients (Allen and McKeown, 1979), however, Graumann and colleagues (2003) did not observe significant differences in GFAP expression between the NAWM of MS patients and control subjects (Graumann et al., 2003). In accordance, we also did not find significant differences in the astrocytic morphology between the NAWM of EAE animals and non-induced controls. Altogether, in this work we propose that, in response to EAE, astrocytes showed an increased expression of enzymes of the glycolysis and TCA cycle suggesting a metabolic shift to oxidative phosphorylation. Those alterations are concomitant with the development of an A1 neurotoxic phenotype. Hence, we hypothesize that TCA cycle upregulation in MS astrocytes is contributing to exacerbate brain inflammation, leading to myelin damage, which needs further investigation. 96 Acknowledgments The authors would like to acknowledge Sara Duarte-Silva (ICVS) for sharing primers; Claudia Nobrega (ICVS) for help with the AutoMACS Pro Separator equipment. Funding This work was supported by Foundation for Science and Technology (FCT) and COMPETE through the project EXPL/NEU-OSD/2196/2013 and by The Clinical Academic Center (2CA-Braga) through the project EXPL/001/2016. The work at ICVS/3B’s has been developed under the scope of the project NORTE-01-0145-FEDER-000013, supported by the Northern Portugal Regional Operational Programme (NORTE 2020), under the Portugal 2020 Partnership Agreement, through the European Regional Development Fund (FEDER), and funded by FEDER funds through the Competitiveness Factors Operational Programme (COMPETE), and by National funds, through the Foundation for Science and Technology (FCT), under the scope of the project POCI-01-0145-FEDER-007038. FM is an assistant researcher and recipient of an FCT Investigator grant with the reference CEECIND/01084/2017. SN is a recipient of a Ph.D. fellowship with the reference PD/BD/114120/2015 from MCTES national funds. Author contributions SN performed the experimental procedures, statistical analysis and wrote the manuscript; FG, JD and GC produced the RNAseq data results; JC, NS, JP an JC critically revised the manuscript; FM supervised the study and edited the manuscript. Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationship that could be construed as a potential conflict of interest. 97 References Abbott NJ, Ronnback L, Hansson E. Astrocyte-endothelial interactions at the blood-brain barrier. Nature Reviews Neuroscience. 2006;7:41-53. Allen IV, McKeown SR. A histological, histochemical and biochemical study of the macroscopically normal white matter in multiple sclerosis. Journal of the Neurological Sciences. 1979;41:81-91. Allen NJ, Barres BA. Neuroscience: Glia - more than just brain glue. Nature. 2009;457:675-677. Aquino DA, Shafit-Zagardo B, Brosnan CF, Norton WT. Expression of glial fibrillary acidic protein and neurofilament mRNA in gliosis induced by experimental autoimmune encephalomyelitis. Journal of Neurochemistry. 1990;54:1398-1404. Batiuk MY, de Vin F, Duque SI, Li C, Saito T, Saido T, et al. An immunoaffinity-based method for isolating ultrapure adult astrocytes based on ATP1B2 targeting by the ACSA-2 antibody. Journal of Biological Chemistry. 2017;292:8874-8891. Chao CC, Gutierrez-Vazquez C, Rothhammer V, Mayo L, Wheeler MA, Tjon EC, et al. Metabolic control of astrocyte pathogenic activity via cPLA2-MAVS. Cell. 2019;179:1483-1498 e1422. Correale J, Farez MF. The role of astrocytes in multiple sclerosis progression. Frontiers in neurology. 2015;6:180. das Neves SP, Serre-Miranda C, Nobrega C, Roque S, Cerqueira JJ, Correia-Neves M, et al. Immune thymic profile of the MOG-induced experimental autoimmune encephalomyelitis mouse model. Frontiers in Immunology. 2018;9:2335. Du XF, Liu J, Hua QF, Wu YJ. Relapsing-remitting multiple sclerosis is associated with regional brain activity deficits in motorand cognitive-related brain areas. Frontiers in neurology. 2019;10:1136. Frohman EM, Racke MK, Raine CS. Multiple sclerosis--the plaque and its pathogenesis. The New England Journal of Medicine. 2006;354:942-955. Galvan-Pena S, O'Neill LA. Metabolic reprograming in macrophage polarization. Frontiers in Immunology. 2014;5:420. Graumann U, Reynolds R, Steck AJ, Schaeren-Wiemers N. Molecular changes in normal appearing white matter in multiple sclerosis are characteristic of neuroprotective mechanisms against hypoxic insult. Brain Pathology. 2003;13:554-573. Hofstetter HH, Shive CL, Forsthuber TG. Pertussis toxin modulates the immune response to neuroantigens injected in incomplete Freund's adjuvant: induction of Th1 cells and experimental autoimmune encephalomyelitis in the presence of high frequencies of Th2 cells. Journal of Immunology. 2002;169:117-125. 98 Holt LM, Olsen ML. Novel applications of magnetic cell sorting to analyze cell-type specific gene and protein expression in the central nervous system. PloS One. 2016;11:e0150290. Itoh N, Itoh Y, Tassoni A, Ren E, Kaito M, Ohno A, et al. Cell-specific and region-specific transcriptomics in the multiple sclerosis model: Focus on astrocytes. Proceedings of the National Academy of Sciences of the United States of America. 2018;115:E302-E309. Kamburov A, Pentchev K, Galicka H, Wierling C, Lehrach H, Herwig R. ConsensusPathDB: toward a more complete picture of cell biology. Nucleic Acids Research. 2011;39:D712-717. Kamburov A, Wierling C, Lehrach H, Herwig R. ConsensusPathDB--a database for integrating human functional interaction networks. Nucleic Acids Research. 2009;37:D623-628. Kantzer CG, Boutin C, Herzig ID, Wittwer C, Reiss S, Tiveron MC, et al. Anti-ACSA-2 defines a novel monoclonal antibody for prospective isolation of living neonatal and adult astrocytes. Glia. 2017;65:9901004. Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Frontiers in Psychology. 2013;4:863. Liddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer L, et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature. 2017;541:481-487. Longair MH, Baker DA, Armstrong JD. Simple Neurite Tracer: open source software for reconstruction, visualization and analysis of neuronal processes. Bioinformatics. 2011;27:2453-2454. Luo J, Ho P, Steinman L, Wyss-Coray T. Bioluminescence in vivo imaging of autoimmune encephalomyelitis predicts disease. Journal of Neuroinflammation. 2008;5:6. MacKenzie-Graham A, Tinsley MR, Shah KP, Aguilar C, Strickland LV, Boline J, et al. Cerebellar cortical atrophy in experimental autoimmune encephalomyelitis. Neuroimage. 2006;32:1016-1023. Mayo L, Trauger SA, Blain M, Nadeau M, Patel B, Alvarez JI, et al. Regulation of astrocyte activation by glycolipids drives chronic CNS inflammation. Nature Medicine. 2014;20:1147-1156. Mueller AM, Pedre X, Stempfl T, Kleiter I, Couillard-Despres S, Aigner L, et al. Novel role for SLPI in MOG-induced EAE revealed by spinal cord expression analysis. Journal of Neuroinflammation. 2008;5:20. Negrotto L, Correale J. Amino acid catabolism in multiple sclerosis affects immune homeostasis. Journal of Immunology. 2017;198:1900-1909. Noseworthy JH. Progress in determining the causes and treatment of multiple sclerosis. Nature. 1999;399:A40-47. Noseworthy JH, Lucchinetti C, Rodriguez M, Weinshenker BG. Multiple sclerosis. The New England Journal of Medicine. 2000;343:938-952. 99 Orihuela R, McPherson CA, Harry GJ. Microglial M1/M2 polarization and metabolic states. British Journal of Pharmacology and Chemotherapy. 2016;173:649-665. Robb JL, Hammad NA, Weightman Potter PG, Chilton JK, Beall C, Ellacott KLJ. The metabolic response to inflammation in astrocytes is regulated by nuclear factor-kappa B signaling. Glia. 2020. Saab CY, Craner MJ, Kataoka Y, Waxman SG. Abnormal Purkinje cell activity in vivo in experimental allergic encephalomyelitis. Experimental Brain Research. 2004;158:1-8. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature Methods. 2012;9:676-682. Schreck L, Ryan S, Monaghan P. Cerebellum and cognition in multiple sclerosis. Journal of Neurophysiology. 2018;120:2707-2709. Sevastou I, Pryce G, Baker D, Selwood DL. Characterisation of transcriptional changes in the spinal cord of the progressive experimental autoimmune encephalomyelitis Biozzi ABH mouse model by RNA sequencing. PloS One. 2016;11:e0157754. Smith ME, Eng LF. Glial fibrillary acidic protein in chronic relapsing experimental allergic encephalomyelitis in SJL/J mice. Journal of Neuroscience Research. 1987;18:203-208. Smith ME, Somera FP, Eng LF. Immunocytochemical staining for glial fibrillary acidic protein and the metabolism of cytoskeletal proteins in experimental allergic encephalomyelitis. Brain Research. 1983;264:241-253. Sofroniew MV. Astrogliosis. Cold Spring Harbor Perspectives in Biology. 2014;7:a020420. Tani M, Glabinski AR, Tuohy VK, Stoler MH, Estes ML, Ransohoff RM. In situ hybridization analysis of glial fibrillary acidic protein mRNA reveals evidence of biphasic astrocyte activation during acute experimental autoimmune encephalomyelitis. The American Journal of Pathology. 1996;148:889-896. Tassoni A, Farkhondeh V, Itoh Y, Itoh N, Sofroniew MV, Voskuhl RR. The astrocyte transcriptome in EAE optic neuritis shows complement activation and reveals a sex difference in astrocytic C3 expression. Scientific Reports. 2019;9:10010. Tavares G, Martins M, Correia JS, Sardinha VM, Guerra-Gomes S, das Neves SP, et al. Employing an open-source tool to assess astrocyte tridimensional structure. Brain Structure and Function. 2017;222:1989-1999. Turley DM, Miller SD. Peripheral tolerance induction using ethylenecarbodiimide-fixed APCs uses both direct and indirect mechanisms of antigen presentation for prevention of experimental autoimmune encephalomyelitis. Journal of Immunology. 2007;178:2212-2220. 100 Williams A, Piaton G, Lubetzki C. Astrocytes--friends or foes in multiple sclerosis? Glia. 2007;55:13001312. Wu GF, Alvarez E. The immunopathophysiology of multiple sclerosis. Neurologic Clinics. 2011;29:257278. 101 Supplementary material Supplementary figure 2. 1 – Expression levels of cell-type specific markers. Expression levels of genes associated with CNS cells (astrocytes, Bergmann glia, microglia, oligodendrocytes and neurons), with endothelial cells and different immune cell populations (B cells, T cells and granulocytes). The expression levels of Atp1b2 , which was identified as the ACSA-2 epitope, used in the MACS technique to isolate the astrocytes, is highlighted. Heatmaps obtained using the Morpheus APP tool of CLUE. C – chronic time point; FPKM – fragments per kilobase of exon per million fragments mapped; NI – non-induced; O – onset time point; P – pre-symptomatic time point. 102 Supplementary figure 2. 2 – Volcano plots of differentially expressed genes. (A) Differentially expressed astrocytic genes for the comparisons between experimental time points in non-induced animals. (B) Differentially expressed genes for the comparisons between disease time points in EAE animals, before normalization to the respective non-induced group. Genes whose expression was not altered are represented in gray; significant underexpessed genes are represented in blue; significant overexpressed genes are represented in red. C – chronic time point; O – onset time point; P – pre-symptomatic time point. 103 Supplementary table 2. 1 – Primers sequence and annealing temperature. Gene Primer sequence (5’→3’) Annealing temperature (ºC) Amigo2 Fw: GAGGCGACCATAATGTCGTT 60 Rv: GCATCCAACAGTCCGATTCT Fbln5 Fw: CTTCAGATGCAAGCAACAA 58 Rv: AGGCAGTGTCAGAGGCCTTA Fkbp5 Fw: TATGCTTATGGCTCGGCTGG 60 Rv: CAGCCTTCCAGGTGGACTTT Atp5b Fw: GGCCAAGATGTCCTGCTGTT 60 Rv: GCTGGTAGCCTACAGCAGAAGG Hspcb Fw: GCTGGCTGAGGACAAGGAGA 60 Rv: CGTCGGTTAGTGGAATCTTCATG Tbp Fw: GGGAGAATCATGGACCAGAA 55 Rv: TTGCTGCTGCTGTCTTTGTT Idh3g Fw: GGCAATGCTCAAGCCAACTC 60 Rv: TGGAGGAATTGTTTGTTGTGAGG Aldoc Fw: GCCTGTTTGGTTAGGAGAGGA 60 Rv: CATGCTGCCTACGGACTCAT Sdha Fw: TGATGCTGTGGTTGTAGGCG 60 Rv: GATACCTCCCTGTGCTGCAA Pfkm Fw: GGTTTGGAAGCCTCTCCTCC 60 Rv: GGGTCATGATCCACTCTTGTAGT 110 Supplementary table 2.4 (continued). p-value q-value Pathway 0.001065 0.02173 Valine degradation I 0.001169 0.022541 Estrogen signaling pathway - Mus musculus (mouse) 0.001169 0.022541 Adipogenesis genes 0.001243 0.023315 Dopaminergic synapse - Mus musculus (mouse) 0.001327 0.02423 Thyroid hormone signaling pathway - Mus musculus (mouse) 0.001389 0.024696 Salivary secretion - Mus musculus (mouse) 0.001423 0.024696 Long-term depression - Mus musculus (mouse) 0.001564 0.025896 Acetylcholine regulates insulin secretion 0.001567 0.025896 Endochondral ossification 0.001773 0.028622 PPAR signaling pathway 0.001988 0.031282 Type II interferon signaling (IFNG) 0.002066 0.031282 Glucagon signaling pathway - Mus musculus (mouse) 0.002073 0.031282 VEGFA-VEGFR2 Pathway 0.002254 0.032592 Metallothioneins bind metals 0.002254 0.032592 Response to metal ions 0.002463 0.034891 Long-term potentiation - Mus musculus (mouse) 0.002599 0.036077 Antigen presentation: folding, assembly and peptide loading of class I MHC 0.002853 0.038077 Synthesis of IP3 and IP4 in the cytosol 0.002853 0.038077 Fatty acid biosynthesis 0.002948 0.038605 The role of GTSE1 in G2/M progression after G2 checkpoint 0.003429 0.044072 Adipocytokine signaling pathway - Mus musculus (mouse) 0.003693 0.046411 Signaling by VEGF 0.003756 0.046411 Mitochondrial fatty acid Beta-oxidation 0.003851 0.046411 Leucine catabolism 0.004009 0.046411 Inositol phosphate metabolism - Mus musculus (mouse) 0.004325 0.046411 Gastric acid secretion - Mus musculus (mouse) 0.004426 0.046411 Valine, leucine and isoleucine biosynthesis - Mus musculus (mouse) 0.004426 0.046411 Astrocytic glutamate-glutamine uptake and metabolism 0.004426 0.046411 Neurotransmitter uptake and metabolism in glial cells 0.004426 0.046411 Lysine degradation II 0.004426 0.046411 γ-linolenate biosynthesis II (animals) 0.004426 0.046411 Cholesterol biosynthesis via desmosterol 0.004426 0.046411 Cholesterol biosynthesis via lathosterol 0.004515 0.046411 Proteoglycans in cancer - Mus musculus (mouse) (Continues) 111 Supplementary table 2.4 (continued). p-value q-value Pathway 0.004547 0.046411 Apoptosis - Mus musculus (mouse) 0.005523 0.055241 Herpes simplex infection - Mus musculus (mouse) 0.005572 0.055241 Insulin signaling pathway - Mus musculus (mouse) 0.005817 0.056861 Inositol phosphate metabolism 0.00614 0.058494 Fatty acid β-oxidation I 0.006153 0.058494 Choline metabolism in cancer - Mus musculus (mouse) 0.006399 0.060011 Platelet activation, signaling and aggregation 0.007241 0.063615 Activation of RAS in B cells 0.007241 0.063615 Sodium-coupled phosphate cotransporters 0.007241 0.063615 The fatty acid cycling model 0.007241 0.063615 The proton buffering model 0.007241 0.063615 Mitochondrial Uncoupling Proteins 0.007489 0.064161 Isoleucine degradation 0.007489 0.064161 Syndecan interactions 0.007594 0.064273 SLC-mediated transmembrane transport 0.008984 0.075118 Circadian rhythm - Mus musculus (mouse) 0.009326 0.077051 Epstein-Barr virus infection - Mus musculus (mouse) 0.009742 0.079211 Collagen biosynthesis and modifying enzymes 0.009816 0.079211 Extracellular matrix organization 0.009942 0.079305 Necroptosis - Mus musculus (mouse) 112 Supplementary table 2. 5 – Pathways significantly altered in the comparison between EAE and noninduced animals at the chronic phase. p-value q-value Pathway 1.29E-12 7.84E-10 Adar1 editing deficiency immune response 4.83E-11 1.47E-08 Epstein-Barr virus infection - Mus musculus (mouse) 4.66E-10 7.96E-08 Th17 cell differentiation - Mus musculus (mouse) 5.24E-10 7.96E-08 Cell adhesion molecules (CAMs) - Mus musculus (mouse) 2.20E-09 2.68E-07 Antigen processing and presentation - Mus musculus (mouse) 2.71E-09 2.74E-07 Immune System 6.28E-09 5.44E-07 Human T-cell leukemia virus 1 infection - Mus musculus (mouse) 1.01E-08 7.67E-07 Viral myocarditis - Mus musculus (mouse) 3.14E-08 2.10E-06 NOD-like receptor signaling pathway - Mus musculus (mouse) 3.47E-08 2.10E-06 Influenza A - Mus musculus (mouse) 6.42E-08 3.54E-06 Toxoplasmosis - Mus musculus (mouse) 7.19E-08 3.64E-06 Type II interferon signaling (IFNG) 8.24E-08 3.85E-06 TNF signaling pathway - Mus musculus (mouse) 2.05E-07 8.88E-06 Th1 and Th2 cell differentiation - Mus musculus (mouse) 2.46E-07 9.95E-06 Herpes simplex infection - Mus musculus (mouse) 2.64E-07 1.00E-05 TCR signaling 5.81E-07 2.07E-05 Apoptosis 1.25E-06 4.23E-05 Kaposi sarcoma-associated herpesvirus infection - Mus musculus (mouse) 1.38E-06 4.40E-05 TYROBP Causal Network 2.31E-06 7.02E-05 Osteoclast differentiation - Mus musculus (mouse) 6.22E-06 0.000175 Translocation of ZAP-70 to Immunological synapse 6.34E-06 0.000175 Measles - Mus musculus (mouse) 7.34E-06 0.000194 Staphylococcus aureus infection - Mus musculus (mouse) 8.33E-06 0.000211 Tuberculosis - Mus musculus (mouse) 9.94E-06 0.000241 Allograft rejection - Mus musculus (mouse) 1.06E-05 0.000248 Downstream TCR signaling 1.33E-05 0.000288 Inflammatory bowel disease (IBD) - Mus musculus (mouse) 1.33E-05 0.000288 Graft-versus-host disease - Mus musculus (mouse) 1.50E-05 0.000305 Human cytomegalovirus infection - Mus musculus (mouse) 1.51E-05 0.000305 Phosphorylation of CD3 and TCR zeta chains 1.56E-05 0.000306 Generation of second messenger molecules 1.71E-05 0.000323 Death Receptor Signalling 2.02E-05 0.000372 Human immunodeficiency virus 1 infection - Mus musculus (mouse) (Continues) 113 Supplementary table 2.5 (continued). p-value q-value Pathway 2.10E-05 0.000374 C-type lectin receptor signaling pathway - Mus musculus (mouse) 2.30E-05 0.000398 Natural killer cell mediated cytotoxicity - Mus musculus (mouse) 2.61E-05 0.000441 Type I diabetes mellitus - Mus musculus (mouse) 2.99E-05 0.000491 NF-kappa B signaling pathway - Mus musculus (mouse) 3.16E-05 0.000505 PD-1 signaling 3.32E-05 0.000511 B Cell Receptor Signaling Pathway 3.36E-05 0.000511 Leishmaniasis - Mus musculus (mouse) 4.28E-05 0.000633 Innate immune system 5.87E-05 0.000848 Adaptive immune system 6.79E-05 0.000959 Autoimmune thyroid disease - Mus musculus (mouse) 9.10E-05 0.001255 Regulation of actin cytoskeleton - Mus musculus (mouse) 0.000115 0.001546 Adipogenesis genes 0.000133 0.001755 Endosomal/Vacuolar pathway 0.000142 0.001832 Apoptosis - Mus musculus (mouse) 0.000145 0.001838 TRAIL signaling 0.000158 0.001961 Osteoclast 0.000166 0.001992 Cytokine signaling in immune system 0.000169 0.001992 Downstream signaling events of B cell receptor (BCR) 0.000174 0.001992 XPodNet - protein-protein interactions in the podocyte expanded by STRING 0.000174 0.001992 Hepatitis C - Mus musculus (mouse) 0.000188 0.002117 Rheumatoid arthritis - Mus musculus (mouse) 0.000198 0.002187 T cell receptor signaling pathway - Mus musculus (mouse) 0.000287 0.003111 Neutrophil degranulation 0.000396 0.004118 Phagosome - Mus musculus (mouse) 0.000407 0.004118 RIPK1-mediated regulated necrosis 0.000407 0.004118 Regulated necrosis 0.000407 0.004118 Interleukin-4 and 13 signaling 0.000468 0.004653 T cell receptor signaling pathway 0.000492 0.004821 Viral carcinogenesis - Mus musculus (mouse) 0.000546 0.005264 Intestinal immune network for IgA production - Mus musculus (mouse) 0.000568 0.005389 IL-2 signaling pathway 0.000608 0.005676 Chemokine signaling pathway 0.000674 0.006092 TNF-alpha NF-κB Signaling Pathway 0.000677 0.006092 Salmonella infection - Mus musculus (mouse) (Continues) 114 Supplementary table 2.5 (continued). p-value q-value Pathway 0.000682 0.006092 Activation of NF-kappaB in B cells 0.000719 0.006324 Pathways in cancer - Mus musculus (mouse) 0.000744 0.006407 MHC class II antigen presentation 0.000749 0.006407 JAK-STAT signaling pathway - Mus musculus (mouse) 0.000776 0.006539 Inflammatory response pathway 0.00082 0.006622 Cytokine-cytokine receptor interaction - Mus musculus (mouse) 0.000821 0.006622 Costimulation by the CD28 family 0.000829 0.006622 AGE-RAGE signaling pathway in diabetic complications - Mus musculus (mouse) 0.000829 0.006622 Immunoregulatory interactions between a lymphoid and a non-lymphoid cell 0.001001 0.007888 Chemokine signaling pathway - Mus musculus (mouse) 0.001053 0.008191 Apoptosis - multiple species - Mus musculus (mouse) 0.001273 0.00978 Regulation of actin cytoskeleton 0.001397 0.010598 CLEC7A (Dectin-1) signaling 0.001461 0.010945 Necroptosis - Mus musculus (mouse) 0.00159 0.011271 CASP8 activity is inhibited 0.00159 0.011271 Regulation of necroptotic cell death 0.00159 0.011271 Dimerization of procaspase-8 0.00159 0.011271 Regulation by c-FLIP 0.001597 0.011271 Antigen presentation: folding, assembly and peptide loading of class I MHC 0.001724 0.012027 C-type lectin receptors (CLRs) 0.001838 0.012536 B cell receptor signaling pathway - Mus musculus (mouse) 0.001838 0.012536 Adipocytokine signaling pathway - Mus musculus (mouse) 0.001994 0.013451 Hematopoietic cell lineage - Mus musculus (mouse) 0.002091 0.013949 Leukocyte transendothelial migration - Mus musculus (mouse) 0.002531 0.016701 Interferon signaling 0.002613 0.016982 Antigen processing-cross presentation 0.002658 0.016982 Asthma - Mus musculus (mouse) 0.002658 0.016982 Interleukin-6 family signaling 0.002809 0.017764 TNFR1-induced proapoptotic signaling 0.002963 0.018543 Hepatitis B - Mus musculus (mouse) 0.003266 0.020232 Nucleotide-binding domain, leucine rich repeat containing receptor (NLR) signaling pathways 0.003365 0.020629 IL-3 Signaling Pathway 0.003555 0.020992 TFs Regulate miRNAs related to cardiac hypertrophy (Continues) 115 Supplementary table 2.5 (continued). p-value q-value Pathway 0.003555 0.020992 CLEC7A/inflammasome pathway 0.003577 0.020992 Chagas disease (American trypanosomiasis) - Mus musculus (mouse) 0.003597 0.020992 Novel Jun-Dmp1 Pathway 0.003597 0.020992 TNFR1-induced NFkappaB signaling pathway 0.003905 0.022576 IL-4 signaling Pathway 0.004141 0.023713 NOD1/2 Signaling Pathway 0.004457 0.025282 IL-7 signaling pathway 0.005252 0.02952 Cellular senescence - Mus musculus (mouse) 0.005499 0.030624 Ligand-dependent caspase activation 0.006101 0.032953 DAP12 signaling 0.006101 0.032953 Interferon alpha/beta signaling 0.006101 0.032953 Amine compound SLC transporters 0.006135 0.032953 Platelet degranulation 0.006485 0.03453 TNFR2 non-canonical NF-κB pathway 0.006645 0.034591 EDA Signalling in Hair Follicle Development 0.006667 0.034591 IL-5 signaling pathway 0.006667 0.034591 RIG-I-like receptor signaling pathway - Mus musculus (mouse) 0.007746 0.039844 Response to elevated platelet cytosolic Ca2+ 0.007923 0.040078 Complement activation, classical pathway 0.007923 0.040078 RHO GTPases activate PAKs 0.008596 0.043124 Toll like receptor signaling 0.009086 0.045208 Chemokine receptors bind chemokines 116 Supplementary table 2. 6 – Pathways significantly altered in the comparison between EAE animals at the onset and pre-symptomatic phases, after normalization for the respective non-induced group. p-value q-value Pathway 6.46E-08 4.69E-05 Extracellular matrix organization 1.30E-07 4.74E-05 Metabolism 1.19E-06 0.000279 Metabolism of amino acids and derivatives 1.53E-06 0.000279 ECM-receptor interaction - Mus musculus (mouse) 8.06E-06 0.001172 HSP90 chaperone cycle for steroid hormone receptors (SHR) 3.18E-05 0.003851 Laminin interactions 4.26E-05 0.00442 Branched-chain amino acid catabolism 6.58E-05 0.005499 Valine degradation 6.81E-05 0.005499 Proteoglycans in cancer - Mus musculus (mouse) 8.16E-05 0.00558 Thyroid hormone synthesis - Mus musculus (mouse) 8.44E-05 0.00558 ECM proteoglycans 0.000134 0.008114 PodNetprotein-protein interactions in the podocyte 0.000177 0.009913 XPodNet - protein-protein interactions in the podocyte expanded by STRING 0.000215 0.011175 Long-term potentiation - Mus musculus (mouse) 0.000241 0.011697 Amphetamine addiction - Mus musculus (mouse) 0.000295 0.013128 PPAR signaling pathway - Mus musculus (mouse) 0.000313 0.013128 Valine, leucine and isoleucine degradation - Mus musculus (mouse) 0.000325 0.013128 Gap junction - Mus musculus (mouse) 0.000374 0.013698 Leucine catabolism 0.000382 0.013698 Non-integrin membrane-ECM interactions 0.000397 0.013698 Fatty acid Beta-oxidation (streamlined) 0.000415 0.013698 Amoebiasis - Mus musculus (mouse) 0.000451 0.013918 Focal Adhesion-PI3K-Akt-mTOR-signaling pathway 0.000459 0.013918 Gastric acid secretion - Mus musculus (mouse) 0.000563 0.016364 Focal adhesion 0.00068 0.018331 Salivary secretion - Mus musculus (mouse) 0.000683 0.018331 Vascular smooth muscle contraction - Mus musculus (mouse) 0.000706 0.018331 Calcium signaling pathway - Mus musculus (mouse) 0.000896 0.019522 PPAR signaling pathway 0.000896 0.019522 Estrogen signaling pathway - Mus musculus (mouse) 0.000958 0.019522 Dopaminergic synapse - Mus musculus (mouse) 0.000958 0.019522 Syndecan interactions 0.000965 0.019522 Cysteine formation from homocysteine (Continues) 117 Supplementary table 2.6 (continued). p-value q-value Pathway 0.000965 0.019522 Cysteine biosynthesis/homocysteine degradation 0.000965 0.019522 Cysteine biosynthesis II 0.000967 0.019522 Collagen biosynthesis and modifying enzymes 0.001032 0.020268 Cellular responses to external stimuli 0.001082 0.020268 Melanogenesis - Mus musculus (mouse) 0.001087 0.020268 A tetrasaccharide linker sequence is required for GAG synthesis 0.001246 0.022343 Mitochondrial LC-fatty acid Beta-oxidation 0.00126 0.022343 Glucagon signaling pathway - Mus musculus (mouse) 0.001311 0.0227 MAPK signaling pathway 0.001404 0.023513 Focal adhesion - Mus musculus (mouse) 0.00146 0.023513 Post-translational protein phosphorylation 0.001488 0.023513 Leucine degradation I 0.001488 0.023513 Valine degradation I 0.001563 0.024169 Cellular responses to stress 0.001749 0.026351 PI3K-Akt signaling pathway - Mus musculus (mouse) 0.001776 0.026351 Inflammatory mediator regulation of TRP channels - Mus musculus (mouse) 0.002127 0.030928 Inflammatory response pathway 0.00218 0.031081 Acetylcholine regulates insulin secretion 0.002261 0.031612 Histidine, lysine, phenylalanine, tyrosine, proline and tryptophan catabolism 0.00274 0.036155 Long-term depression - Mus musculus (mouse) 0.002835 0.036155 Metallothioneins bind metals 0.002835 0.036155 Response to metal ions 0.002835 0.036155 Fatty acid β-oxidation III (unsaturated, odd number) 0.002835 0.036155 Inositol transporters 0.003005 0.03767 Chondroitin sulfate biosynthesis 0.003313 0.040827 Signaling by receptor tyrosine kinases 0.003752 0.044715 Fatty acid Beta-oxidation 0.003752 0.044715 COPI-independent Golgi-to-ER retrograde traffic 0.004089 0.046348 The role of GTSE1 in G2/M progression after G2 checkpoint 0.004089 0.046348 Dermatan sulfate biosynthesis 0.0043 0.046348 Regulation of Insulin-like Growth Factor (IGF) transport and uptake by Insulin-like Growth Factor Binding Proteins (IGFBPs) 0.00432 0.046348 Triacylglyceride synthesis 0.004355 0.046348 Chondroitin sulfate/dermatan sulfate metabolism (Continues) 118 Supplementary table 2.6 (continued). p-value q-value Pathway 0.004355 0.046348 Fatty acid degradation - Mus musculus (mouse) 0.004355 0.046348 Arginine and proline metabolism - Mus musculus (mouse) 0.004399 0.046348 Pathways in cancer - Mus musculus (mouse) 0.004526 0.047001 Peroxisome - Mus musculus (mouse) 0.004864 0.049804 Insulin secretion - Mus musculus (mouse) 0.004943 0.049907 Protein processing in endoplasmic reticulum - Mus musculus (mouse) 0.005044 0.050232 Fluid shear stress and atherosclerosis - Mus musculus (mouse) 0.005297 0.051637 One carbon metabolism and related pathways 0.005327 0.051637 CDP-diacylglycerol biosynthesis II 0.005553 0.053116 Valine, leucine and isoleucine biosynthesis - Mus musculus (mouse) 0.005982 0.056482 MAPK signaling pathway 0.006116 0.057002 Mitochondrial fatty acid Beta-oxidation 0.006413 0.059019 Primary focal segmental glomerulosclerosis FSGS 0.006767 0.059995 Phosphatidylglycerol biosynthesis I (plastidic) 0.006767 0.059995 CDP-diacylglycerol biosynthesis I 0.006767 0.059995 Regulation of KIT signaling 0.006855 0.060047 GnRH signaling pathway - Mus musculus (mouse) 0.007805 0.067547 Small cell lung cancer - Mus musculus (mouse) 0.008416 0.071147 Phosphatidylglycerol biosynthesis II (non-plastidic) 0.008416 0.071147 Omega-9 FA synthesis 0.008618 0.072013 Fatty acid metabolism 0.009065 0.072417 Disinhibition of SNARE formation 0.009065 0.072417 The fatty acid cycling model 0.009065 0.072417 The proton buffering model 0.009065 0.072417 Mitochondrial uncoupling proteins 0.009174 0.072496 Fatty acid Beta-oxidation I 119 Supplementary table 2. 7 – Pathways significantly altered in the comparison between EAE animals at the chronic and pre-symptomatic phases, after normalization for the respective non-induced group. p-value q-value Pathway 0.000164 0.028341 Cilium assembly 0.000384 0.033188 Organelle biogenesis and maintenance 0.001049 0.056976 Cell junction organization 0.001634 0.056976 Adherens junctions interactions 0.002353 0.056976 Acyl-CoA hydrolysis 0.002811 0.056976 Translocation of ZAP-70 to Immunological synapse 0.002811 0.056976 TNFR1-induced proapoptotic signaling 0.002868 0.056976 Cell-cell communication 0.002964 0.056976 Intraflagellar transport 0.003594 0.060421 Cell-cell junction organization 0.003842 0.060421 Phosphorylation of CD3 and TCR zeta chains 0.006249 0.089271 Phagosome - Mus musculus (mouse) 0.006708 0.089271 Tuberculosis - Mus musculus (mouse) 0.009364 0.108426 Huntington disease - Mus musculus (mouse) 126 Abstract Growing evidence suggest that peripheral inflammatory cells alone cannot explain the neurodegenerative processes observed in multiple sclerosis (MS). Rather, evidence suggests that disease progression and disability is better correlated with the maintenance of a low-grade persistent inflammation inside the CNS, driven by local glial cells, such as astrocytes. Considering that the modulation of intracellular Ca2+ levels is the major modulator of astrocyte signaling, in this work we explored the role of astrocytic calcium signaling in MS. To do so, we induced EAE in type 2 inositol 1,4,5-triphosphate receptor (IP3R2)-null animals, whose astrocytes present minimal Ca2+ elevations in the soma and main processes. No differences were observed regarding disease progression, as evaluated by the clinical score, however IP3R2-null animals presented a decreased lesion burden in the cerebellum, compared to Wt littermates. Also, astrocytes from IP3R2-null animals had increased length and complexity near lesions, compared to the non-lesioned areas. We hypothesize that in IP3R2-null animals, at a more inflammatory phase of the disease, astrocytes become more reactive near the lesions, and so are able to control better immune cell infiltration, resulting in smaller lesions. 127 Introduction Multiple sclerosis (MS) is characterized by an autoimmune reaction against the myelin sheath that surrounds central nervous system (CNS) axons, leading to demyelination and, consequently, to axonal and neuronal damage (Sospedra and Martin, 2005). However, growing evidence suggest that T cellmediated inflammatory mechanisms alone cannot explain the neurodegenerative processes observed. In this regard, it was proposed that the innate immune response of resident CNS cells, including microglia and astrocytes, also plays a role in oligodendrocyte injury and axonal degeneration (Correale and Farez, 2015). Regarding astrocytes, in the disease context, they become hypertrophied and upregulate GFAP and vimentin, and gliosis is a characteristic feature of demyelinated plaques (Williams et al., 2007). Moreover, astrocytes may contribute for leukocyte migration to the CNS by secreting both cytokines and metalloproteinases (Brosnan and Raine, 2013). On the other hand, astrocytes may also play a beneficial role in MS by secreting factors involved in lesion repair (Voskuhl et al., 2009). Alterations in intracellular calcium (Ca2+) levels ([Ca2+]i) are the major modulator of astrocyte signaling (Guerra-Gomes et al., 2017; Petravicz et al., 2008; Strokin et al., 2011). Appropriate Ca2+ signaling is important for several astrocytic functions, including regulation of gene expression, gliotransmitter release, local blood flow and morphological plasticity of astrocytic processes (Strokin et al., 2011; Wu et al., 2019). The most widely accepted mechanism for astrocytic Ca2+ increase is the phospholipase C (PLC)/inositol 1,4,5-triphosphate (IP3) pathway. In response to Gq G protein-coupled receptor activation, PLC hydrolyses phosphatidylinositol 4,5-biphosphate to diacylglycerol and IP3, which is going to activate IP3 receptors (IP3Rs) and result in Ca2+ release from the endoplasmic reticulum (Agulhon et al., 2008; Petravicz et al., 2008; Vardjan and Zorec, 2015). Immunohistochemistry and transcriptomic analysis showed that, in the CNS, type 2 IP3R (IP3R2) is mainly expressed in astrocytes (Hertle and Yeckel, 2007; Sharp et al., 1999). In addition, previous in vitro and in vivo studies showed that genetic deletion of IP3R2 resulted in complete loss of spontaneous and agonist-evoked IP3R-dependent Ca2+ increases in astrocytes (Agulhon et al., 2008; Petravicz et al., 2008). Of interest, Staats and colleagues (2016) observed an overexpression of the Itpr2 gene in the spinal cord of experimental autoimmune encephalomyelitis (EAE) animals (Staats et al., 2016), and our transcriptomic results also demonstrated an increased astrocytic expression of this gene, in the cerebellum of EAE animals, at the onset phase of disease (unpublished results). Thus, considering that EAE animals overexpress Itpr2 in regions affected by the disease and that proper Ca2+ signaling in astrocytes is crucial for their function, in this work we studied the functional outcome of “silencing” astrocytes in the EAE mice model. 128 Methods Animals and EAE induction All experiments were reviewed and approved by the Portuguese national authority for animal experimentation, Direção Geral de Veterinária (ID: DGV9458). Animals were housed and handled in accordance with the guidelines for the care and handling of laboratory animals in the Directive 2010/63/EU of the European Parliament and the Council. Animals were housed and maintained in a specific pathogen free environment at 22-24ºC and 55% humidity, on 12 hours (h) light/dark cycles (lights on from 8 a.m. to 8 p.m.) and fed with regular chow and tap water ad libitum . Paper towels were used as environmental enrichment. Paralyzed mice, with clinical scores above 3, were offered easier access to food and water. IP3R2-null mice were kindly given by Prof. Alfonso Araque (University of Minnesota, USA) (Navarrete et al., 2012), under agreement with Prof. Ju Chen (University of San Diego, USA) (Li et al., 2005). Mice were backcrossed to C57BL/6 for at least five generations in our lab. IP3R2-null and wild-type (Wt) littermates were obtained in our lab by crossing IP3R2+/- mice. Toe clipping was performed on P6-P7 pups for genotyping and identification. IP3R2-null female mice and Wt littermates, with 9-15 weeks of age (18.1-26.0 g) were induced with EAE using a commercially available kit (EK-2110; Hooke Laboratories, USA). Briefly, animals were immunized subcutaneously with 200 µg of myelin oligodendrocyte glycoprotein 35-55 (MOG35-55), emulsified in complete Freund’s adjuvant (lot# 0114 and #0126 for experiments 1 and 2, respectively), at the upper and lower back. Pertussis toxin (PTX) in phosphate buffered saline (PBS) was administered intraperitoneally 2 and 24h after immunization (227 ng of PTX from lot# 1001 and 90.8 ng per injection from lot#1007, for experiments 1 and 2 respectively). Animals were daily weighted and monitored for clinical symptoms of disease. Disease severity was evaluated as previously described (das Neves et al., 2018), in a blind manner regarding the animals’ genotype. Only one animal, an IP3R2-null, died during the experiment, at day 17 post-induction, and was excluded from all analysis. Tissue sample collection For biological sample collection, EAE animals were sacrificed at the light phase of the diurnal cycle, at day 16 (active phase of disease, onset phase, experiment 1) and 30 (chronic phase, experiment 2) postdisease induction. Animals were anesthetized with an intraperitoneal injection of ketamine hydrochloride (150 mg/kg, Imalgene 1000; Merial, USA) plus medetomidine hydrochloride (0.3 mg/kg, Dorben Vet; 129 Pfizer, USA). Under deep anesthesia, mice were transcardially perfused with cold 0.9% saline solution, and the brain and spinal cord were dissected. For histological analysis, the brain was immediately embedded in Tissue-Tek O.C.T. compound (Sakura Finetek, Japan), snap-frozen and kept frozen until further sectioning. Luxol Fast Blue staining Serial 20 µm sections of frozen cerebellum were cut in the cryostat and collected to SuperFrost Plus slides (ThermoFisher Scientific, USA), and were posteriorly stained with Luxol Fast Blue as reported previously (das Neves et al., 2018). The quantification of the total lesioned area was done in 5 nonconsecutive sections per animal, representative of the entire cerebellum, in a blind manner regarding the animals’ genotype. The sections were visualized with an Olympus BX51 stereological microscope, and the quantification of the areas was performed with the Visiopharm integrator system software (version 2.12.3.0, Hoersholm, Denmark). The total white matter area was drawn using a 4x objective, and the lesion areas using the 10x objective. The percentage of lesioned area, for each section, was calculated by dividing the sum of the lesioned areas by the total white matter area. Statistical analysis was performed using the average of all sections of the animal. GFAP immunofluorescence and 3-dimensional reconstruction of astrocytes Serial 20 µm sections of cerebellum (coronal sections) were stained for GFAP. Sections were fixed in 4% paraformaldehyde (PFA) in PBS for 30 minutes (min) at room temperature (RT). After antigen retrieval, with pre-heated citrate buffer (Sigma-Aldrich) in the microwave for 20 min, tissue slices were permeabilized with PBS-triton 0.3%, for 10 min, and subsequently blocked with 10% fetal bovine serum in PBS-triton 0.3%, for 30 min. Sections were incubated over-night with rabbit anti-mouse GFAP antibody in blocking solution (1:200; Dako), and, afterwards, with Alexa Fluor 594 goat anti-rabbit diluted in PBStriton 0.3% (1:500; Fisher Technologies, Thermo Fisher Scientific), for 2 hours. After incubation with 4′,6diamidino-2-phenylindole (DAPI; 1:200; Invitrogen, Thermo Fisher Scientific), for 10 min, slides were coversliped with Immumount (Fisher Scientific, Thermo Fisher Scientific). All steps were performed at RT, except for antigen retrieval, and were followed by a washing step using PBS or PBS-triton 0.3%. To perform the 3-dimensional reconstruction of astrocytes, 2-4 photographs per animal were taken from the cerebellum white matter, near lesion regions and in normal appearing white matter (NAWM) regions, using a confocal microscope (FV1000, Olympus) and the following parameters: 40x objective, 1024x1024 resolution, 1 µm increment. The confocal images were then used to performe the 130 morphological reconstruction using the Fiji plugin “Simple Neurite Tracer” (Longair et al., 2011; Schindelin et al., 2012), as previously described (Tavares et al., 2017). For Sholl analysis, concentric circles were superimposed on astrocytes, with origin in the cell soma and with 4 µm distance from each other. 5-6 astrocytes were reconstructed per animal, and their average was used for statistical analysis. Statistical analysis Statistical analysis was performed using SPSS software (version 23, IBM, USA). Sample normality was assessed using the Shapiro-Wilk normality test. EAE animals of both genotypes were compared using a two-tailed unpaired t-test or the Mann-Whitney U test, respectively, for samples with and without a normal distribution. For the percentage of lesioned area, comparison between genotypes and disease time points was performed with a two-way ANOVA. The repeated measures ANOVA was used to analyze the results of the weight, clinical score, total astrocytic length and astrocytic Sholl analysis. The Bonferroni’s multiple comparison test was used as post-hoc test. The partial eta squared value (ηp2) was calculated as a measure of effect size. Results are presented as mean ± standard error of the mean (SEM) for parametric statistical tests, or median ± interquartile range (IQR) for non-parametric statistical tests. For Sholl analysis, results are presented only as mean. The number of biological replicates (n) are specified in the legend of each figure. Statistical significance was considered for p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****). 131 Results Similar disease development in IP3R2-null animals and Wt littermates IP3R2-null animals and Wt littermates immunized with MOG35-55 developed disease in a similar fashion (Figure 3. 1A. Time factor: F(30,540) = 92.033, p < 0.0001, ηp2 = 0.836; genotype factor: F(1,18) = 1.036, p = 0.3222; interaction: F(30,540) = 0.459, p = 0.9945). The animals reached the peak of disease by days 1516 post-induction, and this average clinical score was maintained until the end of the experiment, at day 30 post-induction. The weight of both genotypes was also similar along disease development. As expected, we observed a weight loss in the days following symptoms appearance (Figure 3. 1B. Time factor: F(30,510) = 41.698, p < 0.0001, ηp2 = 0.710; genotype factor: F(1,17) = 0.859, p = 0.3669; interaction: F(30,510) = 0.730, p = 0.8528). On average, IP3R2-null and Wt littermates presented the first clinical symptoms of disease at the same day post-induction (Figure 3. 1C. t(18) = 0.521, p = 0.6088), and they also progressed similarly to a score of 3 (paralysis of both hind paws) (Figure 3. 1D. U = 32.50, p = 0.5016). Moreover, both genotypes presented similar clinical scores on the first day of disease (Figure 3. 1E. U = 48.00, p = 0.8703). Figure 3. 1 – IP3R2-null EAE animals presented a disease course similar to their Wt EAE littermates. (A) IP3R2-null and Wt littermates developed EAE in a similar way until the end of the experiment (day 30), presenting no differences regarding the clinical score. (B) Also, the weight variation was similar between genotypes, having both groups lost weight after the appearance of clinical symptoms of disease. (C) IP3R2-null and Wt littermates presented the first symptoms of disease and (D) reached a clinical score of 3 on average on the same day. (E) The (Continues) 132 score given on the first day of disease was similar between genotypes. Data from one representative experiment is presented as mean ± SEM or median ± IQR. nWt = 10, nIP3R2-null = 8 in (D); nWt = 10, nIP3R2-null = 10 for remaining graphs. Decreased lesion burden in the cerebellum of IP3R2-null animals at the onset phase of disease Even though no differences were observed between genotypes regarding their clinical score, which is mainly dependent on lesion burden at the spinal cord, we next wanted to evaluate if the lesion burden was altered in the cerebellum, which is also affected in EAE and MS. To do so, we quantified the percentage of lesioned area in the cerebellum white matter (Figure 3. 2A), at a more active phase of the disease, the onset phase (day 16), and later at the chronic phase (day 30). As previously published by our group (das Neves et al., 2018), we found a decreased percentage of lesioned area at the chronic phase of disease compared to the onset phase. Interestingly, at the onset phase, the IP3R2-null animals showed a significant decrease in lesion burden compared to their Wt littermates (Figure 3. 2B. Time point factor: F(1,23) = 54.259, p < 0.0001 , ηp2 = 0.702; genotype factor: F(1,23) = 10.636, p = 0.0034, ηp2 = 0.316; interaction: F(1,23) = 20.308, p = 0.0002, ηp2 = 0.469). Figure 3. 2 – IP3R2-null presented a decreased percentage of lesion area compared to Wt animals, at the onset phase of disease. (A) Both groups of EAE animals presented lesioned areas, characterized by inflammatory infiltrates and paler blue staining with LFB [scale bars indicate 1 mm for lower magnification (left) and 100 µm for higher magnification (Continues) 133 (right)]. (B) The percentage of white matter area occupied by these lesions was decreased in IP3R2-null animals compared to Wt littermates, at the onset phase of disease. Data presented as mean ± SEM. nWt onset = 4; nIP3R2-null onset = 4; nWt chronic = 10; nIP3R2-null chronic= 9. *** p < 0.001, **** p < 0.0001. GL – granular layer; LFB – Luxol Fast Blue; ML – molecular layer; WM – white matter. Increased astrocytic complexity near lesion regions in IP3R2-null animals Since we observed differences between genotypes regarding the total white matter area occupied by lesions, at the onset phase of disease, we next reconstructed astrocytes near these lesion regions and compared with astrocytes from regions where the white matter appeared normal (NAWM) (Figure 3. 3A). Representative images of the astrocytes reconstructed are presented in Figure 3. 3B. Regarding the total astrocytic length, we observed that astrocytes located near lesion regions were significantly longer than astrocytes from the NAWM, but this was only significant for the IP3R2-null animals (Figure 3. 3C. Region factor: F(1,6) = 11.820, p = 0.0138, ηp2 = 0.663; genotype factor: F(1,6) = 0.657, p = 0.4485; interaction: F(1,6) = 7.682, p = 0.0324, ηp2 = 0.561). Moreover, IP3R2-null astrocytes were also more complex, presenting more ramifications, near lesion regions compared to NAWM regions (Figure 3. 3D. Region factor: F(1,6) = 17.413, p = 0.0059, ηp2 = 0.744; genotype factor: F(1,6) = 0.192, p = 0.6765; radius factor: F(18,108) = 261.762, p < 0.0001, ηp2 = 0.978; region*genotype interaction: F(1,6) = 9.401, p = 0.0221, ηp2 = 0.610; radius*genotype interaction: F(18,108) = 1.049, p = 0.4129; region*radius interaction: F(18,108) = 0.814, p = 0.6799; region*radius*genotype interaction: F(18,108) = 0.521, p = 0.9429). At the chronic phase of disease, no differences were observed between genotypes or cerebellum regions (Supplementary figure 3. 1. Total length: region factor: F(1,6) = 3.914, p = 0.0952; genotype factor: F(1,6) = 1.037, p = 0.3478; interaction: F(1,6) = 1.942, p = 0.2129. Sholl analysis: Region factor: F(1,6) = 2.643, p = 0.1551; genotype factor: F(1,6) = 0.673, p = 0.4434; radius factor: F(23,138) = 251.819, p < 0.0001, ηp2 = 0.977; region*genotype interaction: F(1,6) = 1.510, p = 0.2651; radius*genotype interaction: F(23,138) = 0.566, p = 0.9436; region*radius interaction: F(23,138) = 1.647, p = 0.0419, ηp2 = 0.215; region*radius*genotype interaction: F(23,138) = 0.700, p = 0.8392). 134 Figure 3. 3 – At the onset phase of disease, astrocytes from IP3R2-null animals were more complex near lesion regions. (A) Representative images of cerebellum sections immunostained for GFAP (scale bar represents 20 µm). (B) Representative drawings of astrocytes reconstructed using the Simple Neurite Tracer plugin of Fiji (scale bar represents 20 µm). (C) At the onset phase of disease, IP3R2-null mice presented longer astrocytes near lesion regions compared to NAWM regions, while Wt animals’ astrocytes were similar in both regions. (D) IP3R2-null animals also presented astrocytes more complex near lesion regions, compared to NAWM regions, as evaluated by the Sholl analysis. Data presented as mean ± SEM for total length and as mean for Sholl analysis. n = 4 per experimental group, 5-6 astrocytes reconstructed per animal. ** p < 0.01. L – lesion, NAWM – normal appearing white matter, WM – white matter. 135 Discussion It is well known that astrocytes play diverse and important roles both in physiological and pathological conditions. In this regard, astrocytic intracellular Ca2+ signaling has gained increased attention since these cells are electrically not excitable, but present robust spontaneous and stimulus-induced Ca2+ transients (Okubo et al., 2019; Shigetomi et al., 2016). In fact, alterations in glial Ca2+ signaling were suggested to be associated with different neuropathologies (Nedergaard et al., 2010; Shigetomi et al., 2016). In stroke, the propagation of the infarction was associated with aberrant astroglial Ca2+ waves, which in turn induced waves of astroglial glutamate release that resulted in distant excitotoxicity (Nedergaard et al., 2010). In vivo imaging studies have also suggested that astrocytic Ca2+ homeostasis and signaling is altered in an Alzheimer’s disease mice model, particularly in astrocytes surrounding amyloid beta plaques (Kuchibhotla et al., 2009; Nedergaard et al., 2010; Shigetomi et al., 2016). In this work we sought to explore the role of astrocytic Ca2+ signaling in MS. To do so, we induced EAE in IP3R2-null animals, which present “silent” astrocytes, with minimal Ca2+ elevations in the soma and main processes. Regarding the clinical score of the animals, we did not find differences between genotypes, however IP3R2-null animals presented a decreased lesion burden in the cerebellum, compared to Wt littermates. In the IP3R2-null animals, this was accompanied by a significant increase in astrocytic length and complexity near lesions, compared to the NAWM. We hypothesize that in IP3R2-null animals, at a more active phase of the disease, astrocytes become more reactive near the lesions, and so are able to control better cell infiltration, resulting in smaller lesions. Our results go along with studies performed in other disease models, which have shown a phenotype improvement after IP3R2 ablation. In particular, IP3R2 ablation in the APPPS1 mice model of Alzheimer’s disease (Appps1+/- x IP3R2-/-) was able to improve spatial memory deficits, in comparison with diseased animals presenting functional astrocytic Ca2+ signaling (Appps1+/- x IP3R2+/+) (Reichenbach et al., 2018). In another study, using a model of brain ischemia, IP3R2-null animals exhibited reduced brain damage, neuronal death and tissue loss, compared to Wt mice, which resulted in improved functional behavioral. However, in this study the authors refer an attenuation of the reactive astrogliosis, characterized by a decreased density of GFAP positive cells (Li et al., 2015), while we suggest that the decrease in the lesioned area observed in IP3R2-null animals is accompanied by an increased astrocytic reactivity. It is conceivable that different pathologies present differences in the inflammatory mediators produced, resulting in different states of astrocytic reactivity, and also that the same activation state could be beneficial in the context of a disease, but deleterious in another. Even in the context of EAE model, different results were observed after targeting astrogliosis. In one study, EAE mice treated with MW01-5-