Electrophysiological and morphological properties of neurons in the rat basolateral amygdala (BLA) were assessed using intracellular recordings in brain slice preparations. The vast majority of cells studied were identified as pyramidal cells on the basis of their accommodation response and by a prominent afterhyperpolarization that followed a current-evoked burst of action potentials. The second class of cells consisted of late-firing neurons that were distinguished electrophysiologically by their very negative resting membrane potential (-82 mV) and conspicuous delay in the onset of spike firing in response to depolarizing current injection. The third class of cells, termed fast-firing neurons, possessed many of the features of intrinsic inhibitory interneurons found elsewhere in the brain. These included very brief action potentials (0.7 msec), a relatively depolarized resting membrane potential (-62 mV), and spontaneous firing at a high rate and the absence of spike frequency accommodation. Intracellular labeling with Lucifer yellow of electrophysiologically identified pyramidal and late-firing cells showed them to have pyramidal to stellate cells bodies and spine-covered dendrites. Although having an overall pyramidal-like morphology, late-firing neurons possessed cells bodies and dendritic fields that were smaller than those of pyramidal cells. Lucifer yellow-labeled fast-firing neurons had a nonpyramidal morphology, with somata that were spherical to multipolar in shape and spine-sparse or aspiny dendrites. The morphological features of these cells corresponded closely to those of GABA-containing interneurons that have been described previously in the rat BLA using immunohistochemical techniques (McDonald, 1985b). Thus, it seems likely that activation of fast-firing neurons underlies inhibitory synaptic events that are recorded in the rat BLA. Our data support the conclusion derived from previous anatomical studies that pyramidal neurons constitute the predominant cell type in the BLA and function as projection neurons in this region of the amygdala. The determination of whether late-firing cells constitute a unique class of projection neurons distinct from pyramidal cells must await the outcome of studies in which the anatomical terminations of this cell type are specified.

Intracerebroventricular administration of dynorphin produced potent and long-lasting effects on motor function and the electroencephalogram in rats. In addition, local iontophoretic or pressure ejection of dynorphin consistently inhibited hippocampal unit activity. None of these effects were significantly affected by naloxone even at high doses. Moreover, a fragment of dynorphin that failed to displace any of a number of tritiated narcotics from rat brain homogenates produced similar effects on these physiological measures in vivo. On the basis of a variety of criteria for "opiate action," the results suggest that a second biologically active site within the dynorphin sequence is capable of quite potent but nonopiate effects.

Whole-cell patch-clamp recordings were used to characterize calcium channel types that are modulated by mu-opioid receptor activation in rat dorsal root ganglion (DRG) neurons. Five distinct components of high-threshold calcium current were isolated on the basis of their sensitivity to the selective channel blockers omega-conotoxin GVIA, nifedipine, omega-conotoxin MVIIC, or omega-agatoxin IVA. The mu-opioid selective agonist Tyr-Pro-NMePhe-D-Pro-NH2 (PLO17) routinely suppressed high-threshold currents and this effect was always reduced by omega-conotoxin GVIA. A fraction of PLO17-sensitive current remained after omega-conotoxin GVIA that was eliminated by application of omega-agatoxin IVA alone or in combination with omega-conotoxin MVIIC. Nifedipine had no effect on mu-opioid responses nor did PLO17 affect the slow component of tail current induced by Bay K 8644. These data suggest that mu-opioid receptors are negatively coupled to three types of calcium channels in rat DRG neurons, including an omega-conotoxin GVIA-sensitive (N-type) channel, an omega-agatoxin IVA-sensitive (P-type) channel and an omega-conotoxin MVIIC-sensitive, nifedipine/GVIA/omega-Aga IVA-resistant (presumptive Q-type) channel.

SUMMARY1. Intracellular recordings were obtained from pyramidal-type neurons in the basolateral amygdaloid nucleus (BLA) in slices of rat ventral forebrain and used to compare the actions of exogenously applied cholinomimetics to the effects produced by electrical stimulation of amygdalopetal cholinergic afferents from basal forebrain.2. Bath application of carbachol depolarized pyramidal cells with an associated increase in input resistance (Ri), reduced the slow after-hyperpolarization (AHP)that followed a series of current-evoked action potentials and blocked spike frequency accommodation. All of these effects were reversed by the muscarinic antagonist atropine but not by the nicotinic antagonist hexamethonium. 3. Electrical stimulation of amygdaloid afferents within the external capsule evoked a series of synaptic potentials consisting of a non-cholinergic fast excitatory postsynaptic potential (EPSP), followed by early and late inhibitory postsynaptic potentials (IPSPs). Each of these synaptic potentials was reduced by carbachol in an atropine-sensitive manner.4. Local application of carbachol to pyramidal cells produced a short-latency hyperpolarization followed by a prolonged depolarization. The hyperpolarization and depolarization to carbachol were blocked by atropine but not hexamethonium.5. The carbachol-induced hyperpolarization was associated with a decrease in Ri and had a reversal potential nearly identical to that of the early IPSP. The inhibitory response was blocked by perfusion of medium containing tetrodotoxin (TTX), bicuculline or picrotoxin, while the subsequent depolarization was unaffected. On the basis of these data, it is concluded that the muscarinic hyperpolarization is mediated through the rapid excitation of presynaptic GABAergic interneurons in the slice.6. The findings that the carbachol-induced depolarization was associated with an increase in Ri, often had a reversal potential below -80 mV, was sensitive to changes in extracellular potassium concentration and was blocked by intracellular ionophoresis of the potassium channel blocker caesium suggest that it resulted from a muscarinic blockade of one or more potassium conductances. M. S. WASHBURN AND H. C MOISES evoked a series of fast EPSPs followed by IPSPs. These non-cholinergic potentials were followed by a slow EPSP that lasted from 10 s-4 min. The slow EPSP was enhanced by eserine and blocked by atropine. It was also blocked by TTX or cadmium, indicating that it was dependent on spike propagation and calciumdependent release of acetylcholine (ACh).8. Stimulation of cholinergic afferents in the slice mimicked other effects produced by carbachol including blockade of the slow AHP and accommodation of action potential discharge and these actions were potentiated by eserine and blocked by atropine.9. The present results provide the first demonstration of muscarinic cholinergic actions in an area of the amygdala known to receive cholinergic innervation. These effects appear to involve both direct and indirect mechanisms.

Whole-cell patch-clamp recordings were used to examine the regulation of voltage-dependent calcium channels by mu- and kappa-opioid receptors in acutely isolated rat dorsal root ganglion (DRG) sensory neurons. Agonists selective for either mu- (Tyr-Pro-NMePhe-D-Pro-NH2, PLO17) or kappa-opioid receptors (dynorphin A, U69,593) inhibited high-threshold calcium currents in a reversible and naloxone-sensitive manner, whereas administration of D-Pen2,5-enkephalin, a delta-selective agonist, was without effect. However, none of the opioids reduced low-threshold T- type currents. The inhibitory effects of PLO17 were blocked by the irreversible mu-opioid antagonist beta-funaltrexamine but not the kappa- opioid antagonist nor-binaltorphimine, while responses to kappa-opioid agonists showed the opposite pattern of antagonist sensitivity. In addition, many cells responded to both PLO17 and dynorphin A (or U69,593), and in these neurons the inhibitory response to one agonist was occluded when tested in the presence of the other. These data suggest that mu- and kappa-opioid receptors are coexpressed on at least some DRG neurons and appear to be functionally coupled to a common pool of calcium channels. Both rapidly inactivating (transient) and sustained components of high-threshold current, arising from pharmacologically distinct types of calcium channels, were identified in our neurons. Activation of mu-opioid receptors selectively reduced the transient component of currents evoked at +10 mV from Vh = -80 mV, while sparing the sustained component. The transient component was irreversibly blocked by the N-type channel antagonist omega-conotoxin GVIA (omega-CgTx), and in one-half of the neurons there was a concomitant loss of the response to PLO17. In the remaining neurons, PLO17 continued to reduce a small fraction of omega-CgTx-insensitive current and subsequent administration of the L-type channel blocker nifedipine in saturating concentrations failed to reduce the opioid- induced inhibitory effect. These data demonstrate that mu-opioid receptors are negatively coupled to several pharmacologically distinct types of calcium channels in DRG sensory neurons, one that was blocked by omega-CgTx and thus likely to be N-type, and a second that was resistant to blockade by N- and L-type channel blockers.