Turnover of Rat Brain Perivascular Cells Ingo Bechmann,* ,1 Erik Kwidzinski,* Adam D. Kovac,* Eva Simbu ¨ rger,* Tamas Horvath,† Ulrike Gimsa,* Ulrich Dirnagl,‡ Josef Priller,‡ and Robert Nitsch* *Institute of Anatomy, ‡Department of Cell and Neurobiology and Department of Neurology, Humboldt-University Hospital Charite ´, Berlin, Germany, and †Section of Neurobiology and Department of OB/GYN, Yale University School of Medicine, New Haven, Connecticut Received May 1, 2000; accepted November 17, 2000 Brain perivascular spaces harbor a population of cells which exhibit high phagocytic capacity. Therefore, these cells can be labeled by intraventricular injection of tracers. Such perivascular cells at the interface be- tween blood and brain are believed to belong to the monocyte/macrophage lineage and to be involved in an- tigen presentation. Currently, it is unclear whether these cells undergo a continuous turnover by entering and leaving the bloodstream. Using bone-marrow-chi- meric animals, migration of donor macrophages into brain perivascular spaces has been reported. On the other hand, following intracerebral injection of india ink into nontransplanted animals, ink-labeled perivas- cular cells were still found 2 years after injection, sug- gesting a high stability of this cell pool. Thus, the turn- over of perivascular cells observed in chimeras might be a result of bone marrow transplantation rather than a physiological occurrence. To address this issue, we mon- itored de novo invasion of macrophages into perivascu- lar spaces of apparently healthy adult rats by applying techniques other than bone marrow transplantation, (i) consecutive injections of different tracers and (ii) ex vivo isolation of macrophages from the blood, cell labeling, and reinjection into the same animal to avoid MHC mis- match. Both approaches revealed vivid de novo invasion of macrophages into perivascular spaces, but not into brain parenchyma, rendering untenable the concept of perivascular cells forming a stable population of macrophages in the brain. Thus, brain perivascular spaces are under permanent immune surveillance of blood borne macrophages in normal adult rats. © 2001 Academic Press Key Words: immune privilege; microglia; meninges; antigen presentation; neuroimmunology. INTRODUCTION Vessels of the brain are surrounded by leptomenin- geal tissue forming perivascular spaces (2). These perivascular spaces are partly filled with cerebrospinal fluid (CSF) (13) and separate the blood compartment from the brain parenchyma. That cells are located within the perivascular spaces was an early notion (25). This population differs from pericytes and perivascular microglia by their location, morphology and immune phenotype (9, 10, 12). Due to their loca- tion at the interface between blood and brain, they were addressed as perivascular cells (9). These cells exhibit a high phagocytic activity (1, 4, 5, 19, 23, 24, 30, 35) and share the immune phenotype of peripheral macrophages (10, 11), including MHC-II expression (12, 19, 29, 30), suggesting a role in antigen presenta- tion (1, 2, 3, 16, 20, 22). Thus it is important to deter- mine whether these perivascular cells form a resident population within the brain or undergo continuous turnover by entering and leaving perivascular spaces. Indeed, using bone-marrow-chimeric animals to study migration of blood-borne cells into the central nervous system (CNS), a slow turnover of perivascular cells was reported by several investigators (3, 8, 14, 16, 18, 21, 22, 33). However, this was challenged by the observa- tion that perivascular cells labeled by intracerebral injection of india ink were still found in situ after 2 years in rats which were not subjected to bone-marrow transplantation (19). On the basis of this finding, it was suggested that perivascular cells form a stable popula- tion of macrophages within the CNS. Thus, migration of blood-derived macrophages into perivacular spaces observed in chimeric animals might reflect an engraft- ment following bone marrow transplantation, but not a normal physiological occurrence. Therefore, we reeval- uated the migration of macrophages into the perivas- cular spaces of adult rats using approaches other than bone marrow transplantation: (i) Repetitive injections of fluorescent dextran amines of different colors were performed to follow the sequential invasion of new macrophages into perivas- cular spaces. Such tracers diffuse along perivascular spaces (17) and are subsequently taken up by perivas- cular cells following intracerebral and intraventricular injection (1, 4 – 6, 19, 24, 30, 35), leading to a stable 1 To whom correspondence should be addressed. Fax: 0049-(0)30- 2802-1460. E-mail: ingo.bechmann@charite.de. Experimental Neurology 168, 242–249 (2001) doi:10.1006/exnr.2000.7618, available online at http://www.idealibrary.com on 242 0014-4886/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.