Homosynaptic and heterosynaptic changes in driving of dentate gyrus interneurons after brief tetanic stimulation in vivo.

This study addressed changes in interneuron driving in the dentate gyrus (DG) of urethane-anaesthetized rats in response to tetanic stimulation of the perforant path (PP) or the converging dentate commissural pathway (CP). Using an extracellular tungsten electrode, we recorded from putative interneurons in the DG that fired to stimulation of both the PP and the CP. Conditioning trains (400 Hz, 17.5 ms) were delivered to each pathway individually and to the two pathways together. The primary measure of synaptic drive was the latency of interneuron discharge. High-intensity PP tetany, CP tetany, and paired tetany consistently reduced firing latency to CP driving (P < .05 for all three), indicating an LTP-like increase in synaptic activation through the CP. High-intensity PP tetany decreased latency to PP driving in only two of seven cases. Heterosynaptic changes occurred frequently in individual experiments. Activity-mediated adjustments in synaptic driving of inhibitory interneurons could play a role in normal physiological function.

Synaptic inhibition regulates associative interactions between afferents during the induction of long-term potentiation and depression.

The induction of long-term potentiation and depression depends upon associative interactions between synapses that converge on individual dendrites. The distance over which these associative interactions occur is limited. The present study evaluates whether this limitation is regulated by synaptic inhibition. We evaluated the associative interactions between two inputs that terminate on different proximo-distal locations along the dendrites of dentate granule cells in the presence of the gamma-aminobutyric acid (GABA) antagonist bicuculline methiodide. Local blockade of GABAergic inhibition enhanced associative interactions between nonoverlapping inputs, compared to within-animal control sites, where inhibitory transmission was intact. The results suggest that synaptic inhibition limits interactions between excitatory synapses by creating current shunts that limit the spread of depolarization within the dendritic tree.

Activity-mediated changes in feed-forward inhibition in the dentate commissural pathway: relationship to EPSP/spike dissociation in the converging perforant path.

1. We tested the hypothesis that long-term synaptic potentiation (LTP)-associated excitatory postsynaptic potential (EPSP)/spike dissociation in the dentate gyrus (DG) is determined, in part, by changes in the feed-forward inhibition evoked by perforant path (PP) stimulation. The dentate commissural pathway (CP) and the PP activate a common pool of interneurons. Therefore a change in synaptic efficacy in the inhibitory circuit due to activation of one pathway could lead to changes in inhibitory efficacy in the other. The relationship between changes in feed-forward inhibition in the CP and EPSP/spike (E-S) functions in the PP should provide information about the site(s) of synaptic modification. 2. In urethan-anesthetized rats, we measured the inhibition of evoked PP population spikes by the CP at interstimulus intervals of 6 and 12 ms. This measure of commissural inhibition and conventional E-S functions for the PP input to the DG were obtained before and after 1) PP tetany (400 Hz, 8-pulse trains) at low, medium, and high stimulus intensities, and 2) CP tetany (200 Hz, 7-pulse trains). 3. Low-intensity PP conditioning (just above population spike threshold) led to a decrease in CP inhibition and large left shifts of the E-S function. High- and medium-intensity PP conditioning yielded increases in commissural inhibition and smaller leftward E-S shifts. 4. Commissural conditioning led to increases in commissural inhibition and inconsistent changes in the E-S functions.(ABSTRACT TRUNCATED AT 250 WORDS)

LTP-associated EPSP/spike dissociation in the dentate gyrus: GABAergic and non-GABAergic components.

The induction of long-term potentiation (LTP) in the dentate gyrus (DG) leads to a change in the firing characteristics of the dentate granule cells. This phenomenon, termed EPSP/spike dissociation, is seen in field potential studies as a shift to the left of the E-S curve, in which population spike amplitude is plotted against pEPSP slope at various stimulus intensities. It has been suggested that EPSP/spike dissociation reflects a decrease in feed-forward inhibition. To test this hypothesis, we blocked GABA-A neurotransmission in a circumscribed area of the DG in urethane-anaesthetized rats by inserting a micropipette filled with 8 mM bibuculline methiodide in saline. We then recorded E-S curves from 9 such electrodes and from 8 control electrodes before and after inducing LTP in the perforant path. Bicuculline prevented the LTP-associated leftward shift of the E-S curves. Instead, the E-S curve showed a consistent shift to the right at the bicuculline sites after LTP, reflecting potentiation of the pEPSP without corresponding increases in the population spike amplitude. The results indicate that the EPSP/spike relationship is controlled largely by GABAergic input, and that potentiation of the population spike in the DG depends largely on a change in the EPSP/spike relationship.

Injection of tetrodotoxin into the entorhinal cortex suppresses cell firing in the dentate gyrus.

Ablation of the entorhinal cortex (EC) of rats induces a reorganization of afferents and dendrites in the denervated dentate gyrus (DG). The signal which triggers these events is unknown, but one candidate is the reduction of granule cell firing which follows the EC lesion. Testing this hypothesis requires eliminating activity in the perforant path without destroying the EC. In the present study, we evaluated whether injecting tetrodotoxin (TTX) into the EC could reduce neuronal activity in the DG to the same extent as EC ablation. Using microelectrode recording techniques, we recorded the activity of single cells in the DG before and up to 8 h after TTX injection. Transmission over the perforant path was monitored before and up to 24 h after TTX injections by stimulating the EC and recording evoked responses in the DG. TTX injections into the EC consistently reduced the firing rate of neurons in the DG by about 80%. Neither firing rate nor temporodentate-evoked responses recovered during the observation period. Saline injections did not alter either physiological measure. The results suggest that the postlesion decreases in neuronal activity in the DG reflect lost synaptic drive rather than an effect dependent upon early degenerative events. Because TTX injection reduces postsynaptic activity to the same extent as does a lesion, the technique can be used to determine whether a loss of afferent drive is sufficient to induce the biochemical and morphological sequelae of denervation.

A new technique for interpreting the BAER in cochlear disease.

We found, by studying 20 normal ears and 34 ears with cochlear disease, that the brainstem auditory evoked response (BAER) wave I latency was a good predictor of the I-V interval (r = -0.67, p less than 0.001), whereas hearing loss had little predictive value. The normal and hearing-loss groups generated regression lines (wave I latency vs I-V latency) that did not differ significantly from each other. A normal range of I-V intervals can be established for any wave I latency, increasing the sensitivity of the BAER.

Aberrant conduction in human peripheral nerve: ephaptic transmission?

Later motor responses were recorded from the foot muscles of patients with neuropathy after stimulation of the posterior tibial nerve at the ankle. The latencies were too short to involve the spinal cord, but latencies were reduced by more proximal stimulation, indicating that the pathway begins with proximal conduction. The response differed from previously reported "axon reflexes," because it appeared on supramaximal stimulation. It was attributed to reflection of an impulse at a discontinuity of the myelin sheath. In 2 of 32 subjects, stimulation of the medial plantar nerve in the great toe resulted in reproducible motor responses with latencies of 37 and 38 msec in the flexor hallucis brevis. Ephaptic transmission was implied.