Volumes of mitochondria in capillary endothelial cells were determined stereologically from electron micrographs of rat cerebellum, cerebral cortex, spinal cord, cauda equina, choroid plexus, anterior pituitary, median eminence of the hypothalamus, renal proximal tubules, skin, cardiac and skeletal muscle, lung, and renal glomerulus. The capillaries of the first four of these tissue types exhibit blood-brain barrier (BBB) characteristics of permeability and capillary ultrastructure and were found to have mitochondrial contents amounting to 8 to 11% of the endothelial cytoplasmic volume. Tissues from non-BBB regions were determined to have mitochondrial volumes of 2 to 5% of their respective cytoplasmic volumes, with a variety of capillary ultrastructures. The apparent excess metabloic work capability of the BBB suggested by this greater number of mitochondria may be related to maintenance of ion differentials between blood plasma and brain extracellular fluid, to extrachoroidal cerebrospinal fluid formation, or to maintaining the unique structural characteristics of central nervous system capillaries.

Volumetric cell densities in 13 different subfields of the temporal lobe were calculated to test various hypotheses about mesial and lateral temporal lobe sclerosis in patients with complex partial epilepsy. In patients benefitting (primary group) from anterior temporal lobectomy (ATL), sclerosis was greater (fewer cells) in anterior than in posterior hippocampus. By contrast, the patients lacking full benefit (nonprimary group) from ATL had decreased numbers of neurons equally distributed from anterior to posterior hippocampus, indicating that zones of mesial temporal cell loss are linked to zones of epileptogenicity. These data support a model of focal hippocampal epilepsy originating from zones of cell loss and synaptic reorganization that is epileptic. There were no differences in cell densities in gyrus hippocampi or in lateral temporal gyri when patients with temporal lobe epilepsy and controls were compared. Hippocampal cell densities in mesial temporal lobe were not reduced in psychomotor epileptic patients with extrahippocampal foci consisting of foreign tissue. Variables in seizure histories were not correlated with Ammon's horn cell densities, indicating that most of the sclerosis preceded the seizures, which did virtually no significant further damage to hippocampus with repeated partial or generalized seizures.

Histopathological studies were carried out on temporal lobe tissue from 25 patients with partial complex seizures who were studied by interictal positron computed tomography (PCT) with 18F-fluorodeoxyglucose and subsequently underwent anterior temporal lobe resection. Abnormalities were identified on x-ray computed tomographic scans in 7 patients, but none indicated the site of a pathologically confirmed structural lesion. Hypometabolic zones were observed on PCT scans of 22 patients and corresponded to focal pathological abnormalities in 19 (15 mesial temporal sclerosis, 2 small neoplasms, 1 angioma, 1 heterotopia). In 1 patient with a focally abnormal PCT scan and no pathological changes, the lesion may have been located posterior to the resection. In the remaining 2 patients, the hypometabolic zones later disappeared and may have represented a transient response induced by depth electrode implantation. Three patients with normal PCT scans had no pathological abnormalities in their resected tissue. The degree of relative hypometabolism measured by PCT correlated well with the severity of the pathological lesion, but the size of the hypometabolic zone was generally much larger than the area of pathological involvement. This discrepancy could not be considered an artifact of technique and must represent either structural abnormalities below the resolution of routine histopathological studies (e.g., loss of synapses) or functional inactivation of neuronal elements associated with the epileptogenic lesion.

No abstract

Pyramidal cell densities in various regions of the anterior and posterior hippocampal formation were measured from en bloc temporal lobe resections and compared with presurgical stereoelectroencephalography (SEEG) data derived from depth electrodes in 12 patients with temporal lobe epilepsy. These data were compared with cell densities observed in four nonepileptic control patients. Patients who consistently exhibited anterior focal changes in the SEEG accompanying onset of ictus had cell densities that were selectively reduced in the anterior hippocampal formation but normal with respect to controls in the posterior hippocampal formation. Patients who exhibited more regional changes in the SEEG at onset of ictus had reduced cell densities in both the anterior and posterior hippocampal formation. Patients who exhibited focal spike activity in the anterior hippocampal formation as their predominant interictal SEEG pattern also had selectively reduced cell densities in the anterior hippocampal formation, while patients with widespread spiking throughout the hippocampal formation had reduced cell densities both anteriorly and posteriorly. These data support the concept that epileptogenesis occurs in or near those areas of epileptic hippocampus that are most damaged. Hippocampal sclerosis must be viewed as related to adjacent hyperexcitable or epileptogenic neurons and not solely as a passive result of repeated anoxia or ischemia.

Light and electron microscopic analyses were carried out on the stimulated and unstimulated paravermal cortices of six rhesus monkeys that had electrodes implanted on their cerebella for 2 months. The electrodes and the stimulation regime (10 p.p.s.: 8 min on, 8 min off) were similar to those used to stimulate the human cerebellum for treatment of certain neurological disorders. Mere presence of the electrode array in the posterior fossa for 2 months resulted in some meningeal thickening, attenuation of the molecular layer, and loss of Purkinje cells immediately beneath the electrode array. There was no evidence of scarring. After 205 hours of stimulation (7.35 X 10(6) pulses) over 18 days, a charge of 0.5 muC/ph or estimated charge density of 7.4 muC/sq cm/ph resulted in no damage to the cerebellum attributable to electrical stimulation per se. Such a charge/phase is about five times the threshold for evocation of cerebellar efferent activity, and might be considered "safe" for stimulation of human cerebellum. Charge density/phase and charge/phase were directly related to increased cerebellar injury in the six other cerebellar cortices stimulated. Leptomeningeal thickening increased with increased charge density. Injury included severe molecular layer attenuation, ongoing destruction of Purkinje cells, gliosis, ongoing degeneration of myelinated axons, collagen intrusion, and increased levels of local polysaccharides. In all cases, even with damage that destroyed all conducting elements beneath the electrodes, there was no damage further than 1 to 2 mm from the edges of the electrode arrays.

We have employed the fluorescent dye nile red to distinguish between normal cells and cells containing lysosomal accumulations of phospholipids. When fibroblasts from an individual with a genetic deficiency in lysosomal sphingomyelinase activity (Niemann-Pick disease) were stained with nile red and visualized by fluorescence microscopy, orange-colored inclusions were observed throughout the cytoplasm. The orange fluorescent bodies could be distinguished from the neutral lipid droplets that fluoresce a brilliant yellow-gold in the presence of nile red. These inclusions were also observed in alveolar macrophages obtained from rats treated with amiodarone, an antiarrhythmic agent known to produce lysosomal phospholipidosis. Flow cytofluorometric analysis revealed that staining of these phospholipid-rich macrophages with nile red can distinguish them from control alveolar macrophages. These results demonstrate that nile red can be employed for the rapid staining of cellular phospholipid inclusions.

We have studied the distribution of gamma-aminobutyric acid (GABA) neurons, axons, and synapses in the rat and monkey hippocampal formation by using glutamate decarboxylase (GAD) immunocytochemistry together with Nissl stains, electron microscopy, and double-labeled retrograde transport of horseradish peroxidase. The numbers of GAD-containing (putative GABA) neurons and their percentages compared to all Nissl-stained neurons were calculated throughout all the various fields and strata of the mammalian hippocampus. Although their numbers are greatest in the polymorph region of the fascia dentata (FD) and in the principal cell layers stratum pyramidale (SP) and stratum granulosum (SG), GAD immunoreactive (GAD-IR) cells are numerous in other strata that contain mostly dendrites and scattered cells. These GAD-IR (putative GABA) neurons in dendritic regions may be involved in feedforward dendritic inhibition or may directly inhibit nearby neurons. We used a postmortem delay technique, which resulted in apparent diffusion of GAD into dendrites and axons and allowed better visualization of the extensive dendritic domain of GAD-IR neurons. Computerized image analysis of GAD-IR puncta indicated that putative GABA terminals were numerous on apical and basilar dendrites of all pyramidal cells but unexpectedly highest in the monkey presubiculum. In the rat, GAD-IR neurons projected axons ipsilaterally from every region to the fascia dentata and CA1; however, commissural GAD-IR axons to the fascia dentata arose from GAD-IR neurons in only the contralateral fascia dentata and subiculum. Electron microscopy of GAD-stained hippocampus identified GAD-IR neurons with non-GAD-IR (possibly excitatory) synapses and GAD-IR terminals on somata and dendrites, 80% being the symmetric type and 20% the asymmetric type. In contrast, non-GAD-IR terminals were asymmetric 80% of the time.