Real-time imaging of intrinsic optical signals during early excitotoxicity evoked by domoic acid in the rat hippocampal slice.

Excitotoxicity involves neuronal depolarization and eventual cell death primarily through excess activation of glutamate receptors. Neuronal cell swelling is considered an early excitotoxic event mediated by ionic influx (mainly Na+ and Cl-) followed by water. Changes in the intrinsic optical signals of nerve tissue correlate with neuronal activity such that light transmittance (LT) increases across the brain slice as cells swell. The present study examined the effects of domoic acid, a potent excitotoxic food contaminant and glutamate analogue, on intrinsic optical signals in the rat hippocampal slice. A brief 1-min exposure to 10 microM domoate at 22 degrees C elevated LT by 58% in the apical dendritic region of CA1 and to a lesser extent in the molecular layer of the upper dentate gyrus. The responses peaked by 5 min and slowly reversed during a 30-min wash. The same responses were evoked by a 1-min application of 10 microM alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionate (AMPA) at 22 degrees C. Minor changes were observed in the CA3 region and the lower blade of the dentate gyrus. At 37 degrees C, exposure to 10 microM domoate for 10 min resulted in apparent irreversible neuronal damage in the CA1 and upper dentate regions. The Na+ channel blocker tetrodotoxin (1 microM) eliminated the evoked CA1 population spike but not the LT increase, indicating that the domoate signal is not associated with action potential discharge pre- or post-synaptically. However, the response to domoate at 22 degrees C was reversibly blocked by the nonspecific glutamate receptor antagonist kynurenate and the non-N-methyl-D-aspartate (non-NMDA) receptor antagonists 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 6,7-dinitroquinoxaline-2,3-dione (DNQX). The response was not blocked by the NMDA receptor antagonist 2-amino-5-phosphonovaleate (AP-5) nor the kainate receptor blocker gamma-D-glutamylaminomethyl sulfonate (GAMS). Relative tissue resistance (RREL) measured across the CA1 dendritic region increased rapidly in response to domoate and fell slowly over 30 min, which paralleled the LT response described above. The increase in RREL was blocked by kynurenate. We propose that domoate binding to AMPA receptors opens channels mediating ionic influx, presumably Na+ followed passively by Cl-. Water follows, producing prolonged postsynaptic swelling in the CA1 and dentate regions where AMPA receptors are most abundant. At higher temperature this swelling can progress to permanent neuronal injury. Imaging intrinsic optical signals allows a real-time view of early excitotoxic events and may prove useful in assessing potentially therapeutic agents that reduce damage induced by excitotoxic agents or ischemia.

Imaging NMDA- and kainate-induced intrinsic optical signals from the hippocampal slice.

1. Brain ischemia causes excess release and accumulation of glutamate that binds to postsynaptic receptors. This opens ionotropic channels that mediate neuronal depolarization and ionic fluxes that can lead to neuronal death. 2. The CA1 pyramidal cell region of the hippocampus is particularly susceptible to this neurotoxic process. Brain cell swelling is considered an early excitotoxic event, but remains poorly under stood and documented. As cells swell, light transmittance (LT) increases through brain tissue, so we hypothesized that brief exposure to glutamate agonists would elicit cell swelling that could be imaged in real time in the hippocampal slice. 3. A 1-min bath application of 100 microM N-methyl-D-aspartate (NMDA) or 100 microM kainate at 22 degrees C greatly increased LT, particularly in the dendritic regions of CA1. The response peaked by 2-3 min and slowly reversed over the subsequent 20 min following exposure. Peak LT increases were > 50% in CA1 stratum radiatum and > 20% in both CA1 stratum oriens and the dendritic region of the dentate gyrus, all areas with a high concentration of NMDA and alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptors. The CA3 stratum radiatum, which contains fewer of these receptors, showed a comparatively small LT increase. 4. The NMDA receptor antagonist 2-amino-5-phosphonovalerate (AP-5) [but not 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX)] blocked the CA1 response to NMDA, whereas the non-NMDA receptor antagonist CNQX (but not AP-5) blocked the response to kainate. The relative tissue resistance measured across CA1 stratum radiatum increased after NMDA or kainate exposure with a time course similar to the LT change described above. The increase in relative tissue resistance was blocked by kynurenate, a nonspecific glutamate antagonist. Increases in both LT and tissue resistance provide two independent lines of evidence that cell swelling rapidly developed in CA1 dendritic areas after activation of NMDA or AMPA receptors. 5. This swelling at 22 degrees C was accompanied by a temporary loss of the evoked CA1 field potential. However, at 37 degrees C the dendritic swelling rapidly progressed to an irreversible LT increase (swelling) of the CA1 cell bodies accompanied by a permanent loss of the evoked field. 6. We propose that dendritic swelling mediated by NMDA and AMPA receptors is an early excitotoxic event that can herald permanent damage to CA1 neurons, those cells most vulnerable to ischemic insult.

Intrinsic optical signaling denoting neuronal damage in response to acute excitotoxic insult by domoic acid in the hippocampal slice.

Using the seafood contaminant domoic acid (an AMPA/kainate receptor agonist), we demonstrate a distinct excitotoxic sequence of events leading to acute neuronal damage in the hippocampal slice as measured by (1) loss of the evoked CA1 field potential, (2) irreversible changes in light transmittance, (3) histopathology, and (4) lucifer yellow injection of single CA1 pyramidal neurons. Change in light transmittance (LT) through the submerged slice indirectly measures altered cell volume, both neuronal and glial. At 37 degrees C, a 1-min superfusion of 10 mu M domoate induced a prolonged reversible increase in LT, primarily in the dendritic regions of CA1 and dentate granule cells (GC), but not in the CA3 region. Spectral analysis (400-800 nm) revealed a wide-band transmittance increase, indicating cell swelling as a major source of the intrinsic signal. The evoked field potential recorded in the CA1 cell body region (PYR) was lost as LT peaked, but completely recovered upon return to the baseline LT level. Increasing domoate exposure to 10 min elicited a different and distinct LT sequence in CA1 and dentate regions. An initial LT increase in dendritic regions evolved in an irreversible decrease in LT. At the same time, LT irreversibly increased in cell body regions (CA1 PYR and GC) and the evoked field potential was irretrievably lost. Also, there was histological damage to cell body and dendritic regions of CA1 and granule cells. Injection of lucifer yellow into single CA1 neurons in slices displaying the irreversible LT sequence revealed extensive dendritic beading, whereas CA1 cells in control slices displayed a smoothly contoured arbor. Consistent with acute neuronal damage, the optical changes generated by domoate did not require extracellular Ca2+, and lowering the temperature protected the slice from irreversible damage to CA1 and GC regions. Although glial changes may also occur, we conclude that imaging light transmittance reveals dynamic and compartmentalized excitotoxic changes in neuronal volume. Beading of the dendritic arbor increases light scatter, thereby decreasing LT and highlighting damaged dendritic regions.

Topography of short portal vessels in the rat pituitary gland: a scanning electron-microscopic and morphometric study of corrosion cast replicas.

We applied scanning electron microscopy combined with imaging and morphometric techniques to analyze the dorsal topography and morphology of short portal vessels linking the capillary beds of the pituitary neural and anterior lobes in adult male albino rats. The pituitary microvasculature was replicated by intracarotid injection of Batson's No. 17 compound producing plastic casts that were advantageous for comprehensive morphometric analyses using an imaging device. The analysis revealed the existence of two types of portal vessels having quantitatively different morphological properties. The bilateral venular plexus of 3-4 vessels located at the base of the infundibular stalk (each venule measuring 300 microns in length and 32 microns in diameter) appears to be the major part of the short portal system in the dorsum of the rat pituitary gland. Narrower capillary-like shunt vessels (6.8 microns in diameter), of about the same length as the venules, were situated throughout other subregions of the intermediate lobe cleft. The short portal vessels of both types made direct anastomoses with the capillary networks in the neural and anterior lobes. The neural lobe capillaries were twice as numerous (1324 per mm2), and only half as wide (6.2 microns), as the sinusoidal capillaries in the anterior lobe (density of 637 per mm2; diameter of 13.7 microns). The topographical position of the portal venular system suggests that the caudolateral subregions of the pituitary neural and anterior lobes have a functional relationship dependent on rapid interlobe transfer of neurohumoral factors such as hormones via the portal blood. This process appears to be supplemented throughout the rest of the cleft between the two lobes by a small number of capillary shunts that supply the epithelial cell lobules of the intermediate lobe in situ. The findings collectively indicate that this portal system provides a constant stream of neurohumoral information that is shared moment-by-moment between the pituitary neural and anterior lobes.