NATURE NEUROSCIENCE VOLUME 15 | NUMBER 4 | APRIL 2012 543 ARTICLES Multiple sclerosis is a disease of the CNS with largely unknown etiology. The symptoms are diverse, but include tremor, fatigue and paralysis 1 . Histopathologically, it presents with large, multifocal demyelinated sclerotic plaques scattered throughout the CNS 2 . These lesions are characterized by oligodendrocyte (ODC) death, axonal degradation, gliosis, activation of microglia, and perivascular and parenchymal accumulation of lymphocytes and macrophages 1,2 . Such inflammatory infiltrates are thought to represent an immune response against the CNS, leading to demyelination. However, histopathologi- cal analyses suggest that neurodegenerative mechanisms could be responsible for some aspects of multiple sclerosis, possibly by genetic predisposition 3 . Other studies have even proposed ODC death as the first pathological event in the formation of demyelinating plaques 4,5 . ODC death and reactive microglia are present in such ‘early stage’ plaques, but blood-derived inflammatory cells are not 4,5 . Accordingly, magnetic resonance imaging studies have found subtle focal changes in the white matter weeks before new lesions are formed 6,7 . Multiple sclerosis was therefore proposed to be of neurodegenerative origin, with adaptive immune involvement being a secondary phenomenon and not a cause. The proposed sequence of events commences with autochthonous ODC death, resulting in microglia-macrophage acti- vation and antigen leakage into CNS-draining lymph nodes (Fig. 1a). Myelin-derived antigens presented by danger signal-activated antigen- presenting cells (APCs) lead to clonal expansion of CNS-reactive T cells. These then invade the CNS, amplify myelin damage and lead to sustained local inflammation. This model seems to be supported by some transgenic mice, in which ODC impairment results in scat- tered lymphocyte infiltration. However, no actual autoimmunity 8,9 is found and, in contrast with the observations made in early sclerotic plaques 4,5 , these models do not show substantial ODC death. To test ODC death as a trigger of inflammation, we investigated initiation of a multiple sclerosis–like pathology in a mouse model (oDTR; Supplementary Fig. 1a) in which diphtheria toxin receptor (DTR) is expressed specifically in ODCs 10 (Supplementary Fig. 1b). Following injection of diphtheria toxin (DTx), we observed wide- spread ODC death, extensive myelin damage and accumulation of myelin components in CNS-draining lymph nodes. However, even after application of immune challenges mimicking pathogenic stimuli, we did not induce any anti-self immunity. Thus, ODC death does not appear to support development of CNS inflammation even under autoimmune-prone experimental conditions. RESULTS Motor dysfunction and demyelination following ODC death To investigate whether ODC death could result in an inflammatory disease of the CNS, we generated the oDTR model, in which DTx- mediated cell ablation was achieved by combining ODC-specific Cre recombinase expression (MOGi-cre) with a Cre-inducible DTR strain (iDTR) 10,11 . DTx-induced clinical disease was dose dependent with progressive pathology, clinically characterized by exacerbating ataxia, tremor, kyphosis and cachexia 10 . We quantified disease severity by weight loss, tremor and motor coordination (see Online Methods 1 Institute of Experimental Immunology, University of Zürich, Zürich, Switzerland. 2 Institute for Molecular Medicine, University Medical Center of the Johannes Gutenberg, University of Mainz, Mainz, Germany. 3 Institute for Medical Microbiology, Immunology and Hygiene, Technische Universität München, Germany. 4 Institute of Pathology, Campus Mitte, Charité – Universitätsmedizin Berlin, Germany. 5 Institute of Anatomy, University of Leipzig, Leipzig, Germany. 6 Department of Neuropathology, Campus Mitte, Charité – Universitätsmedizin Berlin, Germany. 7 These authors contributed equally to this work. 8 These authors jointly directed this work. Correspondence should be addressed to S.W. (woertge@uni-mainz.de) or T.B. (thorsten.Buch@mikrobio.med.tum.de). Received 11 November 2011; accepted 27 January 2012; published online 26 February 2012; doi:10.1038/nn.3062 Primary oligodendrocyte death does not elicit anti-CNS immunity Giuseppe Locatelli 1,7 , Simone Wörtge 2,7 , Thorsten Buch 1,3,7 , Barbara Ingold 4 , Friederike Frommer 2 , Bettina Sobottka 1 , Martin Krüger 5 , Khalad Karram 2 , Claudia Bühlmann 1 , Ingo Bechmann 5 , Frank L Heppner 6 , Ari Waisman 2,8 & Burkhard Becher 1,8 Anti-myelin immunity is commonly thought to drive multiple sclerosis, yet the initial trigger of this autoreactivity remains elusive. One of the proposed factors for initiating this disease is the primary death of oligodendrocytes. To specifically test such oligodendrocyte death as a trigger for anti-CNS immunity, we inducibly killed oligodendrocytes in an in vivo mouse model. Strong microglia-macrophage activation followed oligodendrocyte death, and myelin components in draining lymph nodes made CNS antigens available to lymphocytes. However, even conditions favoring autoimmunity—bystander activation, removal of regulatory T cells, presence of myelin-reactive T cells and application of demyelinating antibodies—did not result in the development of CNS inflammation after oligodendrocyte death. In addition, this lack of reactivity was not mediated by enhanced myelin-specific tolerance. Thus, in contrast with previously reported impairments of oligodendrocyte physiology, diffuse oligodendrocyte death alone or in conjunction with immune activation does not trigger anti-CNS immunity. npg © 2012 Nature America, Inc. All rights reserved.