Multiple forms of short-term plasticity at excitatory synapses in rat medial prefrontal cortex

Short-term synaptic plasticity, in particular short-term depression and facilitation, strongly influences neuronal activity in cerebral cortical circuits. We investigated short-term plasticity at excitatory synapses onto layer V pyramidal cells in the rat medial prefrontal cortex, a region whose synaptic dynamic properties have not been systematically examined. Using intracellular and extracellular recordings of synaptic responses evoked by stimulation in layers II/III in vitro, we found that short-term depression and short-term facilitation are similar to those described previously in other regions of the cortex. In addition, synapses in the prefrontal cortex prominently express augmentation, a longer lasting form of short-term synaptic enhancement. This consists of a 40-60% enhancement of synaptic transmission which lasts seconds to minutes and which can be induced by stimulus trains of moderate duration and frequency. Synapses onto layer III neurons in the primary visual cortex express substantially less augmentation, indicating that this is a synapse-specific property. Intracellular recordings from connected pairs of layer V pyramidal cells in the prefrontal cortex suggest that augmentation is a property of individual synapses that does not require activation of multiple synaptic inputs or neuromodulatory fibers. We propose that synaptic augmentation could function to enhance the ability of a neuronal circuit to sustain persistent activity after a transient stimulus. This idea is explored using a computer simulation of a simplified recurrent cortical network.

Spatio-temporal dynamics of cyclic AMP signals in an intact neural circuitm

The functional properties of neuronal networks can be reconfigured by a variety of modulatory neurotransmitters, which may alter the excitable properties of neurons or the strengths of synaptic connections. Many of these neuromodulators act via the intracellular second messenger cyclic AMP, but their effects on the spatial distribution of cAMP concentration have never been examined in an intact neural circuit. We therefore used the cAMP-indicator dye FICRhR (refs 1, 2) to investigate the effect of several neuromodulators (octopamine, dopamine, acetylcholine, serotonin and proctolin) on cAMP distribution in identified neurons of the lobster stomatogastric ganglion (STG). When added to the bath solution, each of these neuromodulators produced a unique pattern of cAMP transients among the different neurons of the STG. Electrical stimulation of neurons innervating the STG causes synaptic release of endogenous modulators, leading within a few seconds to local increases of cAMP in fine neurite branches, the site where many modulators are thought to act. After prolonged stimulation, cAMP diffuses from the site of production to throughout the neuritic tree and eventually to the cell body. Diffusion of cAMP may explain how transient localized inputs to a neuron can produce long-range effects such as long-term changes in gene expression.

Crossed reticular formation connections that mediate the startle reflex in rats

The startle response is a bilateral response even when elicited by unilateral acoustic or tactile stimuli. Similarly, unilateral electrical stimulation of the reticular formation also elicits a bilateral startle-like response. To examine whether crossed reticular formation connections can distribute the effects of unilateral stimulation across the midline, we delivered one pulse to the caudal pontine (RPC) or medullary reticular formation (MRF) and a second pulse to the opposite side of the brain, at various interpulse intervals. The symmetric collision effects suggest that axons which produce at least 37% (range 23-53%) of the startle response efficacy cross from RPC to RPC with a mean conduction velocity of 13 m/s. Similar collision effects were observed between RPC and MRF sites but at shorter conduction times. To examine which axons might cause these collision effect, the axonally transported label DiI was injected post mortem into 37 RPC sites. Many coarse axons were observed to cross in fascicles between bilateral RPC sites and then separate in the contralateral RPC. The fiber diameters and trajectories of these DiI-labelled axons are consistent with the conduction velocities and trajectories of the substrates mediating the startle-like response determined in collision tests.

Axons and synapses mediating startle-like responses evoked by electrical stimulation of the reticular formation in rats: symmetric and asymmetric collision effects

A new method for determining the locations, directions of transmission and transmission times of synapses mediating electrically evoked responses is proposed here. Electrical stimulation of pontine or medullary reticular formation with one 0.1-ms pulse evokes a short-latency startle-like response. Two pulses were delivered to single sites at various interpulse intervals and the currents required to evoke a criterion startle response were measured. The results suggest that the startle-evoking substrates have absolute refractory periods that range from 0.25-0.6 ms. When one pulse was delivered to a caudal pontine site and a second pulse was delivered to a an ipsilateral medulla site, decreases in required current were observed as interpulse interval increased from +0.4 to +0.8 ms or as interpulse interval decreased from -0.4 to -0.8 ms. These collision-like effects, being symmetric around an interpulse interval of 0, suggest that electrically evoked startle is mediated by fast axons that pass longitudinally through medulla. When one pulse was delivered to the rostral pons and a second pulse to the ipsilateral medulla, however, required currents decreased sharply as interpulse intervals increased from +0.4 to 1.0 ms and as interpulse intervals decreased from +0.2 to -0.2 ms. These asymmetric collision-like effects suggest that strong synapses in the caudal pons, transmitting from pons to medulla, mediate electrically evoked startle. The 0.3-ms asymmetry suggests that the transmission time (i.e., from presynaptic stimulus to postsynaptic action potential) averaged 0.3 ms via monosynaptic connections. The short duration of collision (0.7 ms) suggests that only one postsynaptic action potential was produced with high probability for each presynaptic action potential. From the localization of these effects and the short refractory periods, we estimate that < 60 giant cells on each side of the ventral pontine reticular formation mediate the startle reflex in the rat.