Brain Reseurch Bull~~rin, Vol. 25. pp. 351-354. C Pergamon Press pk. 1990. Printed in the lJ.S.A
BRIEF COMMUNICATION
Characterization of the Dynamic Nature of Microglial Cells W. ERIC THOMAS D~~~rtrne~t of UraE Biology, College of Dentists, Pestle Hail, 305 W. 12th Avenue The Ohio State University, Columbus, OH 43210
Received 26 March 1990 THOMAS, W. E. Charffcte~i~~tjo~ of tlzedynamicnatitre of ~j~r~gliui cells. BRAIN RES BULL 2512) 351-353. 199O.-Dynamic properties of ramified microglia were investigated in tissue cultures of rat brain cells. Microglia were identified in viable cultures based on previously determined aspects of their morphology and location, and this identification was confirmed through immunohistochemical staining for type-three complement receptors. When living microglial cells were monitored using time-lapse video recording, an extremely dynamic nature was revealed. These cells exhibited morphological alteration, surface and cytoplasmic activity, and migration. All of these properties were both constant and rapid, with individual cells being capable of assuming a completely different overall morphology within a period as short as 5-10 min. These dynamic properties were consistently observed in microglia and appeared specific to this cell type. Thus, ramified microgha appear to possess a unique and highly dynamic nature, and this feature is proposed as a potential marker for these cells. Additionally, the significance of microglial celluiar dynamics in the functional role of this cell population in vivo is considered. Microplia CR3 receptors
Time-lapse
video microscopy
Tissue culture
Dynamic
MICROGLIAL cells correspond to a highly distinctive cell type or class within the mammalian CNS. While the functional role of these cells has remained incompletely resolved for some time, microgha have currently come to be proposed as macrophages; however, detailed aspects of their cellular function and properties have not been extensively revealed. Recently, the identification in culture of cells equivalent to the ramified microglia of in vivo has been reported (4). These cells were identified on the basis of morphological criteria and selective histochemical staining for thiamine pyrophosphatase enzyme activity. The morphological features and typical location of the ramified microglial cells proved sufficiently distinctive to permit their direct identification visually in living cultures. This ability thus allows the study of individual viable cells. When living microglia were observed under tight microscopy. it was noted that these cells appeared to be plastic with respect to their overall morphology. Therefore, in an initial characterization of viable microglia in vitro, dynamic properties of the cells were investigated. In the present study time-lapse video microscopy was empIoy~d to reveal how microghal cells change their mo~hology over specific time periods. The findings demonstrate that microglia exhibit a unique and stereotypic set of highly dynamic properties.
properties
Plasticity
Migration
scribed ( 13). Immunohistochemical staining was performed on fixed cultures using an indirect fluorescence procedure according to established methods (5,6). The monoclonal antibody MAC-I (M1170) was used in the primary incubation. This antibody has previously been demonstrated to selectively recognize type-three complement receptors (CR3) (1). Binding of the primary antibody was detected with a rhodamine-conjugated goat anti-mouse antibody. Living cerebral cortical cultures were observed with phasecontrast optics on a Zeiss IM-35 inverted microscope. This same microscope was equipped for epifluorescence. photography, and time-lapse video recording. In all cases. cultures were maintained at 37°C through the use of a heated stage (model 5000, Technology Inc.. Whitehouse Station, NJ) and the pH of the culture medium was regulated to 7.4 using a gas lamina flow attachment (model 5010. Technology Inc.). Under these conditions cultures could be maintained for extended periods (4-h hr) without any detectable cellular deterioration. Microglial cells were readily recognized based on their morphology in culture fields under low magnification power. Individual microglia or several cells in close proximity were then selected for study at higher magnification, where they were analyzed via time-lapse recording. For monitoring of living ceils, a video camera @AGE-MT1 65) par focal with the microscope objective and connected to a black and white monitor (Sony PVM-122) via a time-lapse video recorder (Panasonic AC-0650) was employed. Individual identified cells were continuously monitored for 4-6 hr periods: recording rate was reduced from the real-time speed of 60 framesisec to between 1-4 seciframe. Photographs of recorded images were then obtained
METHOD
Primary cultures of dissociated cerebral cortical cells obtained from fetal rat were prepared and maintained as previously de-
351
FIG. 1. Immunofluorescent staining of ramified microglia for CR3 receptors. (A) A microglial celi identified in phase-contrast on the basla iit morphology. (B) Subsequent fluorescence view of the same cell in A showing intense CR3 staining. (C and D) Fluorescence photomicrographs (It individual microglial cells stained for CR3. The fluorescent staining reveals the overall morphology and provides some reflection of the granular nature of the cell soma (indicated by armw in C). Scale bar= IO pm.
through the use of a video printer (Sony UP-81 1) or by direct photo~aphy of the video monitor. RESULTS When viable cortical cultures were observed under light microscopy, microglial cells could readily be identified based on previously determined features of morphology and location (4). These cells were characterized by a small (5-1.5 @) pleomorphic cell body with multiple thin branching processes (Fig. 1A). This overall appearance gives them a resemblance to the ramified type of microglia in vivo (8). When living microglial cells were observed at higher mag~~cation, the nucleus could be seen to fill a significant portion or most of the cell soma. The cytoplasm, as well as the cell membrane surface, invariably displayed a highly granular or vesicular appearance (see Fig. 1A and C). In further support for the identified cells as microglia, i~uno~st~hemical staining for CR3 was employed. The CR3 antigen, as recognized by MAC-l antibodies, is a specific component of various macrophages and phagocytic cells (11,12), and has been shown to be selectively associated with microglia in the CNS (3,9). The ceils identified as microglia in culture exhibited select CR3 immunostaining (Fig, 1). Staining was most intense in the area of the cell body, but processes were also labelled. In some instances, the staining also revealed or reflected the granular or vesicular nature of the cell soma (see arrow in Fig. 1C). Thus, this feature was
consistent between living and stained cells, and has been previously observed in hist~hemically stained preparations (4). When immunostained preparations were viewed initially under phasecontrast, the morphological identification of microglial cells could be assessed via subsequent viewing with fluorescence. In all cases. the visual identification of microglia correlated with CR3 immunofluorescence (see Fig. 2A and B). Thus, previous findings, the present immunohistochemical staining, and cellular morphology all appear to confirm the identifed cells as ramified microglia. During direct observation of living microglia, it was noticed that the cells appeared to possess unique dynamic properties. Within just a few minutes, the cell would change in both shape and position. In order to investigate the dynamic nature of microglia in the cortical cultures, time-lapse video recording was implemented. Both identified fields of microglia and individual cells were monitored for periods of several hours. Individual microglial cells were indeed observed to exhibit a high degree of dynamic activity (Fig, 2). This activity was characterized by a constant change in overall morphology through alterations in the shape and size of the cell soma and through the retraction and extension of processes. As well as being continuous, the change in mo~hoiogy was also rapid, occurring in just a few minutes (5-10 min maximum). Thus, within the space of every few minutes, the same single microglial cell would have a completely different appearance. This mo~hological alteration was accompanied by motility in the granular or vesicular structures associated with the soma; these
DYNAMIC PROPERTIES
OF MICROGLIA
FIG. 2. Demonstration of the dynamic properties of ramified microglial cells. Nine separate frames at sequential time intervals from a time-lapse video recording of a single microglial cell are shown in A-I. The overall sequence covers 128 min and the interval between frames ranges from 7 to 40 min. The cell is indicated (arrow) in each frame and the time and date shown (lower left). A line drawing at actual size of the cell at each time, prepared from the enlarged negative, has been placed in each frame as an insert (upper left). Note the constant change in overall morphology and particularly the rapidity of change indicated by shorter time intervals (e.g., 9 min: A-B. IO min: D-E and G-H). In order to facilitate an assessment of movement or migration, an asterisk has been placed at the same absolute position in the culture field of each frame. All frames are presented at the came total magnification ( x 1.633).
structures moved in a whirling or circulating fashion. Simultaneous with all this activity, the microglia also displayed migration (see asterisks in Fig. 2). While some cells migrated in a specific direction and often wandered into or out of the microscope field, the usual pattern of migration appeared nondirectional or random. Each cell typically migrated about in a spatial area several times the cellular diameter. This migration, just as with the changes in overall morphology and motility of somal vesicular structures, transpired on a continuous and rapid basis. The dynamic properties described for individual microglia were also observed in fields containing multiple cells. Each ramified cell in a field displayed morphological changes and migration. Thus, within a short time period (often 30-60 min), microglia in a culture field could present an entirely different appearance. From these observations, it also appeared that each microglial cell existed somewhat in isolation, or separate from other microglia as opposed to being clustered. In addition, microglia appeared rather evenly or uniformly distributed in most areas. These features combined with the dynamic properties gave the impression that the microglial population formed a network surveying the biological environment.
The dynamic properties attributed to microglia here appear to be unique and consistent. In all of the time-lapse studies, microglia exhibited a much higher level of dynamic activity than other cell types. although other cells exhibited some movement or motility involved with growth or functionality. In over 30 recordings of individual cells or fields, microglia always exhibited these features. Thus. the microglial cell type has been characterized to possess a highly dynamic nature. DISCUSSION
Brain microglia, which can be estimated to constitute IO-20% of the cellular population in this tissue (2, 7. lo), have been proposed as macrophages (7,8). In the present study, microglial cells in cerebral cortical cultures have been demonstrated to possess unique and highly dynamic properties as revealed through video microscopy. Such properties would appear to be consistent with the potential role of these cells as macrophages, since macrophages are known to be highly active and continuously interact with their cellular environment (7). Also. the unique and selective possession of these properties may make dynamic nature
354
‘THOMAS
useful as a cellular marker for microglia, in addition to biochemical components, functional processes, and morphological features. This type of marker would prove particularly useful for confirming identification of living cells. While the microglia in vitro are characterized by a highly dynamic nature, tissue culture certainly represents a less confining cellular environment than the intact tissue. Therefore, it is not possible to say to what degree the full complement of properties reported here are expressed in vivo, and how much dynamic activity might be constrained or inhibited by the physical proximity of adjacent cells. However, it seems at least possible that microglia, and especially their processes, could be able to move somewhat within the extracellular space of brain tissue, since other cells such as direct blood-derived macrophages and lymphocytes migrate through the tissue. It is intriguing to speculate that this may explain some of the basis for their
relatively small size. Assuming some degree of dynamic activity in vivo and given the tissue distribution of microglial cells, this presents an image of a population network patrolling or surveying the interstitial expanse of the tissue. This provides an attractive model for a system of intrinsic macrophages. ACKNOWLEDGEMENTS The skillful technical assistance of Mr. Paul Booth is greatly appreciated. Gratitude is also expressed to Drs. F. Jordan and M. Whitehead for
critical review of the manuscript, and to Dr. B. Zwilling for the donation of Ml/70 hybridomas. This work was supported by NSF Grant Rll8896145, ONR contract NOOOl4-88-K-0528, a Klingenstein Fellowship Award in the Neurosciences, an Ohio State University Seed Grant, and the OSU College of Dentistry.
REFERENCES 1. Beller, D. I.; Springer, T. A.; Schreiber, R. D. Anti-Mac-l selectively inhibits the mouse and human type three complement receptor. J. Exp. Med. 156:1000-1009; 1982. 2. Cammermeyer, J. Morphological distinctions between oligodendrocytes and microglia cells in the rabbit cerebral cortex. Am. J. Anat. 118:227-248; 1966. 3. Giulian, D.; Baker, T. J. Characterization of amoeboid microglia isolated from developing mammalian brain. J. Neurosci. 6:21632178; 1988. 4. Glenn, J. A.; Jordan, F. L.; Thomas, W. E. Further studies on the identification of microglia in primary cultures of mixed cerebral cortical cells. Brain Res. Bull. 22:104O-1052; 1989. 5. Jordan, F. L.; Rieke, G. K.; Thomas, W. E. Presence and development of ependymal cells in primary tissue cultures derived from embryonic rat cerebral cortex. Dev. Brain Res. 35:97-l 10; 1987. 6. Jordan, F. L.; Thomas, W. E. Identification of somatostatin-containing neurons in primary cultures of rat cerebral cortex. Neurosci. Lett. 77:249-254; 1987. 7. Jordan, F. L.; Thomas, W. E. Brain macrophages: questions of origin
and interrelationship. Brain Res. Rev. 13:165-178; 1988. 8. Perry, V. H.; Gordon, S. Macmphages and microglia in the nervous system. Trends Neurosci. 11:273-277; 1988. 9. Perry, V. H.; Hume, D. A.; Gordon, S. Immunohistochemical localization of macrophages and microglia in the adult and developing mouse brain. Neuroscience 15:313-326; 1985. 10. Peters, A.; Palay, S. L.; Webster, H. D. F. The fine structure of the nervous system: The cells and their processes. New York: Harper and Row; 1970. Il. Springer, T. A. Monoclonal antibody analysis of complex biological systems. Combination of cell hybridization and immunoabsorbents in a novel cascade procedure and its application tu the macrophage cell surface. J. Biol. Chem. 2_56:3833-3839; 1981. 12. Springer, T.; Galfre, G.; Secher, D. S.; Milstein, C. Mac-l: a macrophage differentiation antigen identified by monoclonal antibody. Eur. J. Immunol. 9301-306; 1979. 13. Thomas, W. E. Synthesis of acetylcholine and y-aminobutyric acid by dissociated cerebral cortical cells in vitro. Brain Res. 332:79-89; 1985.
BRIEF COMMUNICATION
Characterization of the Dynamic Nature of Microglial Cells W. ERIC THOMAS D~~~rtrne~t of UraE Biology, College of Dentists, Pestle Hail, 305 W. 12th Avenue The Ohio State University, Columbus, OH 43210
Received 26 March 1990 THOMAS, W. E. Charffcte~i~~tjo~ of tlzedynamicnatitre of ~j~r~gliui cells. BRAIN RES BULL 2512) 351-353. 199O.-Dynamic properties of ramified microglia were investigated in tissue cultures of rat brain cells. Microglia were identified in viable cultures based on previously determined aspects of their morphology and location, and this identification was confirmed through immunohistochemical staining for type-three complement receptors. When living microglial cells were monitored using time-lapse video recording, an extremely dynamic nature was revealed. These cells exhibited morphological alteration, surface and cytoplasmic activity, and migration. All of these properties were both constant and rapid, with individual cells being capable of assuming a completely different overall morphology within a period as short as 5-10 min. These dynamic properties were consistently observed in microglia and appeared specific to this cell type. Thus, ramified microgha appear to possess a unique and highly dynamic nature, and this feature is proposed as a potential marker for these cells. Additionally, the significance of microglial celluiar dynamics in the functional role of this cell population in vivo is considered. Microplia CR3 receptors
Time-lapse
video microscopy
Tissue culture
Dynamic
MICROGLIAL cells correspond to a highly distinctive cell type or class within the mammalian CNS. While the functional role of these cells has remained incompletely resolved for some time, microgha have currently come to be proposed as macrophages; however, detailed aspects of their cellular function and properties have not been extensively revealed. Recently, the identification in culture of cells equivalent to the ramified microglia of in vivo has been reported (4). These cells were identified on the basis of morphological criteria and selective histochemical staining for thiamine pyrophosphatase enzyme activity. The morphological features and typical location of the ramified microglial cells proved sufficiently distinctive to permit their direct identification visually in living cultures. This ability thus allows the study of individual viable cells. When living microglia were observed under tight microscopy. it was noted that these cells appeared to be plastic with respect to their overall morphology. Therefore, in an initial characterization of viable microglia in vitro, dynamic properties of the cells were investigated. In the present study time-lapse video microscopy was empIoy~d to reveal how microghal cells change their mo~hology over specific time periods. The findings demonstrate that microglia exhibit a unique and stereotypic set of highly dynamic properties.
properties
Plasticity
Migration
scribed ( 13). Immunohistochemical staining was performed on fixed cultures using an indirect fluorescence procedure according to established methods (5,6). The monoclonal antibody MAC-I (M1170) was used in the primary incubation. This antibody has previously been demonstrated to selectively recognize type-three complement receptors (CR3) (1). Binding of the primary antibody was detected with a rhodamine-conjugated goat anti-mouse antibody. Living cerebral cortical cultures were observed with phasecontrast optics on a Zeiss IM-35 inverted microscope. This same microscope was equipped for epifluorescence. photography, and time-lapse video recording. In all cases. cultures were maintained at 37°C through the use of a heated stage (model 5000, Technology Inc.. Whitehouse Station, NJ) and the pH of the culture medium was regulated to 7.4 using a gas lamina flow attachment (model 5010. Technology Inc.). Under these conditions cultures could be maintained for extended periods (4-h hr) without any detectable cellular deterioration. Microglial cells were readily recognized based on their morphology in culture fields under low magnification power. Individual microglia or several cells in close proximity were then selected for study at higher magnification, where they were analyzed via time-lapse recording. For monitoring of living ceils, a video camera @AGE-MT1 65) par focal with the microscope objective and connected to a black and white monitor (Sony PVM-122) via a time-lapse video recorder (Panasonic AC-0650) was employed. Individual identified cells were continuously monitored for 4-6 hr periods: recording rate was reduced from the real-time speed of 60 framesisec to between 1-4 seciframe. Photographs of recorded images were then obtained
METHOD
Primary cultures of dissociated cerebral cortical cells obtained from fetal rat were prepared and maintained as previously de-
351
FIG. 1. Immunofluorescent staining of ramified microglia for CR3 receptors. (A) A microglial celi identified in phase-contrast on the basla iit morphology. (B) Subsequent fluorescence view of the same cell in A showing intense CR3 staining. (C and D) Fluorescence photomicrographs (It individual microglial cells stained for CR3. The fluorescent staining reveals the overall morphology and provides some reflection of the granular nature of the cell soma (indicated by armw in C). Scale bar= IO pm.
through the use of a video printer (Sony UP-81 1) or by direct photo~aphy of the video monitor. RESULTS When viable cortical cultures were observed under light microscopy, microglial cells could readily be identified based on previously determined features of morphology and location (4). These cells were characterized by a small (5-1.5 @) pleomorphic cell body with multiple thin branching processes (Fig. 1A). This overall appearance gives them a resemblance to the ramified type of microglia in vivo (8). When living microglial cells were observed at higher mag~~cation, the nucleus could be seen to fill a significant portion or most of the cell soma. The cytoplasm, as well as the cell membrane surface, invariably displayed a highly granular or vesicular appearance (see Fig. 1A and C). In further support for the identified cells as microglia, i~uno~st~hemical staining for CR3 was employed. The CR3 antigen, as recognized by MAC-l antibodies, is a specific component of various macrophages and phagocytic cells (11,12), and has been shown to be selectively associated with microglia in the CNS (3,9). The ceils identified as microglia in culture exhibited select CR3 immunostaining (Fig, 1). Staining was most intense in the area of the cell body, but processes were also labelled. In some instances, the staining also revealed or reflected the granular or vesicular nature of the cell soma (see arrow in Fig. 1C). Thus, this feature was
consistent between living and stained cells, and has been previously observed in hist~hemically stained preparations (4). When immunostained preparations were viewed initially under phasecontrast, the morphological identification of microglial cells could be assessed via subsequent viewing with fluorescence. In all cases. the visual identification of microglia correlated with CR3 immunofluorescence (see Fig. 2A and B). Thus, previous findings, the present immunohistochemical staining, and cellular morphology all appear to confirm the identifed cells as ramified microglia. During direct observation of living microglia, it was noticed that the cells appeared to possess unique dynamic properties. Within just a few minutes, the cell would change in both shape and position. In order to investigate the dynamic nature of microglia in the cortical cultures, time-lapse video recording was implemented. Both identified fields of microglia and individual cells were monitored for periods of several hours. Individual microglial cells were indeed observed to exhibit a high degree of dynamic activity (Fig, 2). This activity was characterized by a constant change in overall morphology through alterations in the shape and size of the cell soma and through the retraction and extension of processes. As well as being continuous, the change in mo~hoiogy was also rapid, occurring in just a few minutes (5-10 min maximum). Thus, within the space of every few minutes, the same single microglial cell would have a completely different appearance. This mo~hological alteration was accompanied by motility in the granular or vesicular structures associated with the soma; these
DYNAMIC PROPERTIES
OF MICROGLIA
FIG. 2. Demonstration of the dynamic properties of ramified microglial cells. Nine separate frames at sequential time intervals from a time-lapse video recording of a single microglial cell are shown in A-I. The overall sequence covers 128 min and the interval between frames ranges from 7 to 40 min. The cell is indicated (arrow) in each frame and the time and date shown (lower left). A line drawing at actual size of the cell at each time, prepared from the enlarged negative, has been placed in each frame as an insert (upper left). Note the constant change in overall morphology and particularly the rapidity of change indicated by shorter time intervals (e.g., 9 min: A-B. IO min: D-E and G-H). In order to facilitate an assessment of movement or migration, an asterisk has been placed at the same absolute position in the culture field of each frame. All frames are presented at the came total magnification ( x 1.633).
structures moved in a whirling or circulating fashion. Simultaneous with all this activity, the microglia also displayed migration (see asterisks in Fig. 2). While some cells migrated in a specific direction and often wandered into or out of the microscope field, the usual pattern of migration appeared nondirectional or random. Each cell typically migrated about in a spatial area several times the cellular diameter. This migration, just as with the changes in overall morphology and motility of somal vesicular structures, transpired on a continuous and rapid basis. The dynamic properties described for individual microglia were also observed in fields containing multiple cells. Each ramified cell in a field displayed morphological changes and migration. Thus, within a short time period (often 30-60 min), microglia in a culture field could present an entirely different appearance. From these observations, it also appeared that each microglial cell existed somewhat in isolation, or separate from other microglia as opposed to being clustered. In addition, microglia appeared rather evenly or uniformly distributed in most areas. These features combined with the dynamic properties gave the impression that the microglial population formed a network surveying the biological environment.
The dynamic properties attributed to microglia here appear to be unique and consistent. In all of the time-lapse studies, microglia exhibited a much higher level of dynamic activity than other cell types. although other cells exhibited some movement or motility involved with growth or functionality. In over 30 recordings of individual cells or fields, microglia always exhibited these features. Thus. the microglial cell type has been characterized to possess a highly dynamic nature. DISCUSSION
Brain microglia, which can be estimated to constitute IO-20% of the cellular population in this tissue (2, 7. lo), have been proposed as macrophages (7,8). In the present study, microglial cells in cerebral cortical cultures have been demonstrated to possess unique and highly dynamic properties as revealed through video microscopy. Such properties would appear to be consistent with the potential role of these cells as macrophages, since macrophages are known to be highly active and continuously interact with their cellular environment (7). Also. the unique and selective possession of these properties may make dynamic nature
354
‘THOMAS
useful as a cellular marker for microglia, in addition to biochemical components, functional processes, and morphological features. This type of marker would prove particularly useful for confirming identification of living cells. While the microglia in vitro are characterized by a highly dynamic nature, tissue culture certainly represents a less confining cellular environment than the intact tissue. Therefore, it is not possible to say to what degree the full complement of properties reported here are expressed in vivo, and how much dynamic activity might be constrained or inhibited by the physical proximity of adjacent cells. However, it seems at least possible that microglia, and especially their processes, could be able to move somewhat within the extracellular space of brain tissue, since other cells such as direct blood-derived macrophages and lymphocytes migrate through the tissue. It is intriguing to speculate that this may explain some of the basis for their
relatively small size. Assuming some degree of dynamic activity in vivo and given the tissue distribution of microglial cells, this presents an image of a population network patrolling or surveying the interstitial expanse of the tissue. This provides an attractive model for a system of intrinsic macrophages. ACKNOWLEDGEMENTS The skillful technical assistance of Mr. Paul Booth is greatly appreciated. Gratitude is also expressed to Drs. F. Jordan and M. Whitehead for
critical review of the manuscript, and to Dr. B. Zwilling for the donation of Ml/70 hybridomas. This work was supported by NSF Grant Rll8896145, ONR contract NOOOl4-88-K-0528, a Klingenstein Fellowship Award in the Neurosciences, an Ohio State University Seed Grant, and the OSU College of Dentistry.
REFERENCES 1. Beller, D. I.; Springer, T. A.; Schreiber, R. D. Anti-Mac-l selectively inhibits the mouse and human type three complement receptor. J. Exp. Med. 156:1000-1009; 1982. 2. Cammermeyer, J. Morphological distinctions between oligodendrocytes and microglia cells in the rabbit cerebral cortex. Am. J. Anat. 118:227-248; 1966. 3. Giulian, D.; Baker, T. J. Characterization of amoeboid microglia isolated from developing mammalian brain. J. Neurosci. 6:21632178; 1988. 4. Glenn, J. A.; Jordan, F. L.; Thomas, W. E. Further studies on the identification of microglia in primary cultures of mixed cerebral cortical cells. Brain Res. Bull. 22:104O-1052; 1989. 5. Jordan, F. L.; Rieke, G. K.; Thomas, W. E. Presence and development of ependymal cells in primary tissue cultures derived from embryonic rat cerebral cortex. Dev. Brain Res. 35:97-l 10; 1987. 6. Jordan, F. L.; Thomas, W. E. Identification of somatostatin-containing neurons in primary cultures of rat cerebral cortex. Neurosci. Lett. 77:249-254; 1987. 7. Jordan, F. L.; Thomas, W. E. Brain macrophages: questions of origin
and interrelationship. Brain Res. Rev. 13:165-178; 1988. 8. Perry, V. H.; Gordon, S. Macmphages and microglia in the nervous system. Trends Neurosci. 11:273-277; 1988. 9. Perry, V. H.; Hume, D. A.; Gordon, S. Immunohistochemical localization of macrophages and microglia in the adult and developing mouse brain. Neuroscience 15:313-326; 1985. 10. Peters, A.; Palay, S. L.; Webster, H. D. F. The fine structure of the nervous system: The cells and their processes. New York: Harper and Row; 1970. Il. Springer, T. A. Monoclonal antibody analysis of complex biological systems. Combination of cell hybridization and immunoabsorbents in a novel cascade procedure and its application tu the macrophage cell surface. J. Biol. Chem. 2_56:3833-3839; 1981. 12. Springer, T.; Galfre, G.; Secher, D. S.; Milstein, C. Mac-l: a macrophage differentiation antigen identified by monoclonal antibody. Eur. J. Immunol. 9301-306; 1979. 13. Thomas, W. E. Synthesis of acetylcholine and y-aminobutyric acid by dissociated cerebral cortical cells in vitro. Brain Res. 332:79-89; 1985.
