Premium
Co‐cultures of microglia and astrocytes from kainic acid‐lesioned adult rat hippocampus: Effects of glutamate
Author(s) -
Bowman Charles L.,
Swann John W.,
Severin Charles M.,
Romanowski Marcus R.
Publication year - 1992
Publication title -
glia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.954
H-Index - 164
eISSN - 1098-1136
pISSN - 0894-1491
DOI - 10.1002/glia.440050406
Subject(s) - microglia , biology , neuroglia , cytoplasm , lipofuscin , kainic acid , astrocyte , microbiology and biotechnology , hippocampal formation , glial fibrillary acidic protein , ultrastructure , glutamate receptor , pathology , neuroscience , central nervous system , anatomy , biochemistry , immunology , medicine , inflammation , immunohistochemistry , receptor
The long‐standing question concerning the direct actions of glutamate on the membrane potential of astroglial cells in the central nervous system was addressed using the in vitro kainic acid‐lesioned hippocampal slice preparation and primary cell co‐cultures of astrocytes and microglia derived from such lesions. The ultrastructure of the lesioned hippocampus was examined to aid in the identification of the cells appearing in culture. In culture, microglia appeared as flat cells, less than 1 μm in thickness at the edge of the cell, but thicker (about 5 μm) near the nucleus. The cytoplasm was packed with granular inclusions. Microglia appeared in two morphological forms, amoeboid and ramified. The amoeboid form was characterized by a cell body with a single process, and was always observed 1 day after starting the cell culture. Such cells became less frequent after 1 week in culture. The ramified form appeared as a rounded cell, devoid of processes, and were frequently observed in older cultures (> 1 week). Microglia did not round up after exposure to dibutyrylcyclic adenosine monophosphate (cAMP), and did not stain for glial fibrillary acidic protein (GFAP). An ultrastructural examination of the lesion demonstrated that microglia were present and that they contained many cytoplasmic granules similar to lipofuscin‐containing granules. No filaments were observed in the cytoplasm of microglia. By contrast, the cytoplasm of astrocytes in culture had far fewer granules, rounded up to dibutyryl‐cAMP, exhibited multiple processes, and stained for GFAP. In slices, astrocytes had no lipofuscin‐containing granules, but numerous cytoplasmic filaments were present. The membrane potential of the rounded‐up cells in culture varied between −10 mV and −95 mV (while electrically unresponsive cells in the lesion varied between −47 and −95 mV). In culture, glutamate evoked a depolarization or a hyperpolarization of the membrane potential of 87% of the rounded‐up cells tested. Ionophoretic application of glutamate depolarized the membrane potential of 50% of the cells tested in the lesion. The results show that co‐cultures of microglia and astrocytes can be obtained routinely from lesions within the adult central nervous system, and that glutamate can affect the membrane potential of astrocytes in cultures derived from such adult tissue. These findings suggest that responses to glutamate, recorded from unresponsive cells (presumptive astrocytes) in the lesioned slice preparation, are likely the result of direct action of this neurotransmitter on these cells. © 1992 Wiley‐Liss, Inc.