The Control of Microglial Neurotoxicity by Fractalkine and CX3CR1

Richard Ransohoff, M.D.

The Cleveland Clinic Foundation

Funded in June, 2004: $300000 for 5 years


back to top

Examining How Brain Cells Signal Inflammatory Microglial Cells to Limit "Collateral" Damage

Researchers will learn how brain cells signal resident inflammatory microglial cells to modify their attack, so that the inflammatory reaction does not inadvertently damage brain cells. The findings could provide a better understanding of how inflammatory microglial cells both protect and sometimes inadvertently damage brain cells.

Microglial cells are the only inflammatory cells residing in the brain to fend off invaders. Current evidence suggests that microglial cells help repair brain tissue that has been damaged by trauma, but it is less clear whether they are effective in warding off invading microbes. Some data suggest that when microglial cells attack an invader, brain cells simultaneously attempt to moderate the response, to prevent damaging nearby brain cells. There also is evidence that microglial cells may exacerbate the damage caused by low-intensity chronic injury, such as occurs in Alzheimer's and Parkinson's diseases, autoimmune multiple sclerosis (MS), and HIV dementia. Scientists do not know how to differentiate the factors that determine whether microglial cells protect against or promote destruction.

The Cleveland Clinic investigators have been studying microglial cells' actions to identify the circumstances associated with destructive outcomes. The researchers have identified a substance, called fractalkine, which brain cells secrete to communicate with microglial cells. Scientists also have identified a receptor, called CX3CR1, which is located on the surface of microglial cells and receives this fractalkine message. The researchers hypothesize that brain cells release fractalkine to signal moderation, and that when the signal is picked up by the microglial CX3CR1 receptor, the microglial cells mount only a modest attack. When a malfunction occurs in either the brain cells' release of fractalkine or in the microglial cells' CX3CR1 receptor, however, the message is lost and microglial cells attack unabated, destroying brain cells in the area.

The investigators will compare mice with intact CX3CR1 receptors on microglial cells and mice that have been genetically engineered to lack this receptor. The microglial cells were engineered to produce a fluorescent signal, enabling the researchers to see the cells' actions when challenged with a systemic bacterial infection or with the animal model of autoimmune MS, both of which result in chronic brain disease. The investigators will compare the number of brain cells that have died in both groups of mice, hypothesizing that the mice with the intact CX3CR1 receptors on microglial cells will have fewer dead brain cells. If so, this will provide indirect evidence that brain cell signaling mutes an attack by microglial cells.

Significance: This study will help demonstrate how brain cells and inflammatory microglial cells communicate. The research also will help elucidate how microglial cells damage brain cells when their signals to moderate an inflammatory response are blocked by a chronic brain infection. These findings may help to explain how inflammatory microglial cells exacerbate brain cell damage in the presence of chronic brain infections that block this signal.


back to top

The Control of Microglial Neurotoxicity by Fractalkine and CX3CR1

Microglia, the resident immune cells of the central nervous system (CNS), are widely regarded as double-edged swords. In particular, microglia exert neuroprotective functions, but can also produce bystander damage to neurons (neurotoxicity) when they are activated. The mechanisms that switch between neuroprotection and neurotoxicity of microglia are poorly understood. Microglial neurotoxicity is proposed to play a primary role in diseases of major public health interest including multiple sclerosis (MS) and HIV-1-associated dementia. In particular, microglia are the predominant inflammatory elements in the cortical pathology of MS. Furthermore, activated microglia are found in proximity to dying neurons in virtually all neurodegenerative disorders, including Alzheimer's disease (AD), Parkinson's disease (PD), and motor neuron diseases. It is therefore extremely important to understand how microglia become either toxic or protective and to dissect the signaling that underlies these dual functions.

Our prelilminary data indicate that mice lacking CX3CR1, the receptor for fractalkine, exhibit dramatically increased microglial reactivity and neurotoxic effects, in models of CNS insult. We conducted these studies in CX3CR1+/GFP and CX3CR1GFP/GFP mice, a unique genetic model that allows us to monitor microglial reaction in vivo and also to establish the function of CX3CR1. Our proposed research will be conducted in these mice, in which the CX3CR1 coding region was replaced by green fluorescence protein (GFP), so that GFP is present in cells that are destined to express CX3CR1. All CNS microglia constitutively express CX3CR1 and microglia are therefore fluorescence labeled in both CX3CR1+/GFP heterozygotes (CX3CR1+/-  mice) and CX3CR1GFP/GFP knockouts (CX3CR1-/-  mice). Our analysis does not indicate any CNS cells aside from microglia that express the CX3CR1/GFP marker in vivo. Fractalkine is constitutively produced by neurons in the healthy CNS, so effects of tonic release of fractalkine are observed primarily in forebrain gray matter structures.

These results lead us to propose that fractalkine and CX3CR1 constitute a major neuron/microglial communication pathway. Specifically, we hypothesize that microglial activation is inhibited by neuronally derived fractalkine, signaling to CX3CR1, its receptor on microglia. We propose further that microglial activation, in the absence of restraint by CX3CR1 signaling, is deleterious in the settings of innate or autoimmune inflammation. Our preliminary data indicated disease models in which to address these hypotheses. We found that induction of CNS innate immunity in CX3CR1-/- mice led to apoptosis of both neurons and microglia. We studied experimental autoimmune encephalomyelitis (EAE) in mice lacking CX3CR1 and found prominent cortical microglial activation, along with altered expression of surface markers of activation. We will extend our preliminary findings and define how CX3CR1 regulates outcomes of inflammatory CNS pathology through our Specific Aims.


back to top
Richard Ransohoff, M.D.

Richard M. Ransohoff is Professor of Molecular Medicine at the Cleveland Clinic Lerner College of Medicine, as well as Staff Scientist in the Dept. of Neurosciences of the Lerner Research Institute and Staff Neurologist in the Mellen Center for MS Treatment and Research, both at the Cleveland Clinic Foundation (CCF), Cleveland, OH. Dr. Ransohoff graduated with honors from Bard College, Annandale, NY, with a B.A. in Literature and received the M.D. degree with honors from Case School of Medicine, Cleveland, OH. He completed residencies in Internal Medicine (Mt. Sinai Medical Center, Cleveland, OH; Board Certified 1981) and Neurology (CCF; Board Certified 1985). From 1984 until 1989, Dr. Ransohoff was a post-doctoral fellow in the laboratory of Dr. Timothy Nilsen, Dept. of Molecular Biology and Microbiology, Case School of Medicine.

Among other honors and awards, he received a Physician's Research Training Award from the American Cancer Society (1984-86); a Harry Weaver Neuroscience Scholarship from the National Multiple Sclerosis Society (NMSS; 1987-1992); a Clinical Investigator Development Award from the National Institutes of Health (NIH; 1988-1993); and the John and Samuel Bard Award in Science and Medicine, 2002. He has been cited from 1996 through 2004 in the "Best Doctors in America" for his expertise in the clinical care of patients with multiple sclerosis (MS).

Dr. Ransohoff served as regular member on NIH and NMSS Study Sections and on numerous Special Emphasis Panels, and has served as Chair of the NMSS Peer Review Committee B starting in October, 2004. He is a member of the Editorial Boards of the Journal of Immunology, (where he is presently Section Editor); Trends in Immunology; and the Journal of Neuroimmunology. From 1998-2000, Dr. Ransohoff was a member of the NINDS Director's Planning Panel on "The Neural Environment." He is a member of the Steering Committee for the NIH Therapeutic Development Program in Spinal Muscular Atrophy; the International Advisory Boards for the 7th (2004) and 8th (2006) Congresses on Neuroimmunology; and the Scientific Advisory Board for Chemocentryx, San Carlos, CA. He serves on External Advisory Boards for CHARTER (CNS HIV Anti-Retroviral Therapy Effects Research; MH22005); a Program Project on Alexander's Disease (NS 42803); the MS Lesion Project (NMSS RG 3185); the University of Nebraska's Center for Neurovirology & Neurodegenerative Disorders (NS43985) and is the External Advisor for the European Union's Project on "Mechanisms of Brain Inflammation" (QLG3-00612). He is a member of the National MS Society's Medical Advisory Board. He is a Co-director of the Marine Biological Laboratory's special topics course on "Pathogenesis of neuroimmunological disease," held biennially at Wood's Hole, MA.

For the past decade, Dr. Ransohoff's research has focused on the functions of chemokines and chemokine receptors in development and pathology of the nervous system. He also has a longstanding and continuing interest in the mechanisms of action of interferon-beta. Dr. Ransohoff has received continuous research support from the NIH and the NMSS since 1988. He has published more than 130 scientific reports, more than 35 reviews and book chapters, and three edited books.

Dr. Ransohoff is a member of the American Academy of Neurology, the American Neurological Association, the American Association for the Advancement of Science and the American Association of Immunologists.


back to top

Specific Aim 1.
We will define mechanisms of microglial neurotoxicity associated with CNS innate immune responses, induced by systemic LPS injections. Intraperitoneal (i.p.) injections of LPS induce CNS innate immunity, without blood brain barrier disruption or hematogenous leukocyte infiltrates. We found that CNS innate immunity was associated with striking levels of neuronal and microglial TUNEL staining in the hippocampus and cortex of CX3CR1-/- mice but not in CX3CR1+/- mice. We next transferred in vivo activated microglia from the CNS of LPS-injected CX3CR1+/- or CX3CR1-/- mice by intracranial injection to the cortex of wild type recipients. Remarkably, activated microglia from CX3CR1-/- but not CX3CR1+/- mice mediated widespread injury to neurons in the cortices of wild-type adoptive recipients, assayed by in situ TUNEL reaction. To provide descriptive data relevant to our hypothesis, selected microglial gene expression will be monitored. To determine function of components highly expressed in CX3CR1-/- cells, microglia will be pre-treated before transfer, either with modifiers of cytokine action (such as IL1RA) or chemical inhibitors of critical enzymes (such as iNOS, NADPH oxidase components or myeloperoxidase) to suppress individual redox stress pathways.

Specific Aim 2.
We will determine how CX3CR1 regulates microglial function in EAE

Specific Sub-Aim 2a.
We will determine how CX3CR1 regulates parenchymal microglia in EAE, by comparing disease in WT→CX3CR1+/GFP and WT→CX3CR1GFP/GFP radiation chimerae. The focus of our research is to examine how CX3CR1 regulates microglial function in EAE. Because the trafficking and function of circulating leukocytes is regulated in part by CX3CR1, we will prepare radiation chimerae in which both CX3CR1+/- and CX3CR1-/- mice contain wild type bone marrow so that the only variable in these EAE experiments will be the presence or absence of CX3CR1 on parenchymal microglia. Based on our preliminary observations, we predict that excessive activation of microglia during EAE in CX3CR1-/- mice will lead both to demyelination and neuronal injury, particularly in the cortex and hippocampus, which are the sites of maximal fractalkine expression and paracrine action.

Specific Sub-Aim 2b.
We will generate a novel model of restricted cortical demyelination by inducing EAE in MCP-1-/-→CX3CR1+/GFP and MCP-1-/-→CX3CR1GFP/GFP radiation chimerae. We showed that EAE in MCP-1-/- mice is characterized by extensive T cell infiltrates and microglial activation, but absence of infiltrating monocytes, accounting for the mild neurobehavioral phenotype. We hypothesize that over-activation of microglia in MCP-1-/-→CX3CR1GFP/GFP radiation chimerae with EAE may mediate "pure" cortical demyelinating pathology as observed in MS, with cortical demyelination, neuronal apoptosis, neuritic damage, and microglial activation but limited white matter pathology. This model would be invaluable for examination of mechanisms of this poorly understood aspect of MS.


back to top

Cardona A.E., Huang D., Sasse M.E., and Ransohoff R.M.  Isolation of murine microglial cells for RNA analysis of flow cytometry.  Nat Protoc. 2006;1(4):1947-51.

Cardona A.E., Pioro E.P., Sasse M.E., Kostenko V., Cardona S.M., Dijkstra I.M., Huang D., Kidd G., Dombrowski S., Dutta R., Lee J.C., Cook D.N., Jung S., Lira S.A., Littman D.R., and Ransohoff R.M.   Control of microglial neurotoxicity by the fractalkine receptor.   Nat Neurosci. 2006 Jul;9(7):917-24.