Contribution of retinal ganglion cells to the mouse electroretinogram.

PURPOSE: To quantify the direct contribution of retinal ganglion cells (RGCs) on individual components of the mouse electroretinogram (ERG). METHODS: Dark- and light-adapted ERGs from mice 8 to 12 weeks after optic nerve transection (ONTx, n=14) were analyzed through stimulus response curves for a- and b-waves, oscillatory potentials (OPs), positive and negative scotopic threshold response (p/n STR), and the photopic negative response (PhNR) and compared with unoperated and sham-operated controls, as well as to eyes treated with 6-cyano-7-nitroquinoxaline-2,3-dion (CNQX). RESULTS: We confirmed in mice that CNQX intravitreal injection reduced the scotopic a-wave amplitude at high flash strength, confirming a post-receptoral contribution to the a-wave. We found that ONTx, which is more specific to RGCs, did not affect the a-wave amplitude and implicit time in either photopic or scotopic conditions while the b-wave was reduced. Both the pSTR and nSTR components were reduced in amplitude, with the balance between the two components resulting in a shortening of the nSTR peak implicit time. On the other hand, amplitude of the PhNR was increased while the OPs were minimally affected. CONCLUSION: With an intact a-wave demonstrated following ONTx, we find that the most robust indicators of RGC function in the mouse full-field ERG were the STR components.

Glutamate-induced intracellular calcium changes and neurotoxicity in cortical neurons in vitro: effect of chemical ischemia.

To study the role of calcium in neuronal death during ischemia, we examined the characteristics of intracellular Ca2+ ([Ca2+]i) changes in single rat forebrain neurons exposed for 5 min to glutamate (3 microM + 1 microM glycine), NMDA (30 microM + 1 microM glycine), kainate (100 microM) or high K+ (50 mM), under both normal and ischemic conditions. The parameters of [Ca2+]i change measured included peak [Ca2+]i level, plateau [Ca2+]i level, area under the [Ca2+]i response curve and time taken by [Ca2+]i to recover to 10% of the peak response. Under normal conditions, all the agonists studied produced similar [Ca2+]i changes. Chemical ischemia simulated by application of 5 mM KCN in glucose-free buffer had no effect on the basal level of [Ca2+]i, but significantly enhanced and prolonged the [Ca2+]i changes produced by all the agonists. However, in toxicity studies, chemical ischemia significantly potentiated the toxicity of only glutamate and N-methyl-D-aspartate. In correlation studies, all the neurons which died displayed an irreversible secondary [Ca2+]i load prior to loss of viability. These studies suggest that while Ca2+ entry may play a critical role in neuronal death, the magnitude of initial [Ca2+]i change does not predict the toxicity of an agonist in cortical neurons.

NMDA receptors are involved at the ventrolateral nucleus tractus solitarii for termination of inspiration.

The purpose of the present study was to determine whether blockade of excitatory amino acid receptors at the ventrolateral nucleus of the tractus solitarius would influence respiratory activity. This was done by microinjecting excitatory amino acid receptor antagonists into the ventrolateral nucleus of the tractus solitarius of alpha-chloralose-anesthetized animals while monitoring respiratory activity using a Fleisch pneumotachograph and arterial blood pressure and heart rate. Bilateral microinjection of the NMDA receptor antagonist, 3-[(R)-carboxypiperazin-4-yl]-propyl-1- phosphomic acid (CPP), 5.62 nmol per side, produced an increase in inspiratory duration (+4 +/- 1.6 s, n = 8) which progressed to an apneustic pattern of breathing. Similar results were obtained with CPP microinjected into the ventrolateral nucleus of the tractus solitarius of three vagotomized animals. Bilateral microinjection of a second NMDA receptor antagonist, 2-amino-7-phosphono-heptanoic acid (AP7), 562 nmol per side, produced qualitatively similar effects on respiration as seen with CPP. In contrast, blockade of non-NMDA receptors with 6-cyano-7-nitroquinoxaline-2,3-dione (CNXQ), 0.125 nmol per side, had very little effect on respiration. Activation of NMDA receptors at the ventrolateral nucleus of the tractus solitarius with bilateral microinjection of NMDA, 39 pmol, produced a large increase in expiratory duration (+11 +/- 3 s, n = 8), and apnea during the expiratory phase of the respiratory cycle in half of the animals studied. Similar results were obtained with D,L-alpha-amino-3-hydroxy-5-methyl-4-isoxazol-proprionate (AMPA). These results indicate that an endogenous excitatory amino acid released at the ventrolateral nucleus of the tractus solitarius and acting at the NMDA receptor, plays a significant role in respiratory timing.

Effects of parallel fiber stimulation on neurons of rat dorsal cochlear nucleus.

We have compared the effects of parallel fiber stimuli on extracellularly recorded neurons showing regular or bursting spontaneous activity patterns in the dorsal cochlear nucleus of rat brainstem slices. Ninety percent of regular neurons failed to respond to stimulus currents (1.4 +/- 0.28 mA, mean +/- SEM) significantly greater than those (0.4 +/- 0.07 mA) that elicited responses from 96% of bursting neurons. Responses of bursting neurons were elicited from widely separated loci along the molecular layer. Kynurenic acid and CNQX or DNQX blocked both spontaneous firing and responses to parallel fiber stimuli of bursting neurons. The same agents also blocked responses of regular neurons but had little or no effect on their spontaneous firing rates. AP-5 caused small decreases in spontaneous rates of both bursting and regular neurons but did not appear to affect responses to stimuli. The data support the hypothesis that the responses of both regular and bursting neurons to parallel fiber stimulation are mediated by glutamate, acting mainly through non-NMDA receptors. Spontaneous activity of bursting, but not regular, neurons also requires non-NMDA glutamatergic transmission, suggesting that the spontaneous firing of bursting neurons, consisting largely of cartwheel cells, may depend upon granule cell activity.