DD Kunkel

Multiple voltage-gated potassium (K) channel gene products are likely to be involved in regulating neuronal excitability of any single neuron in the mammalian brain. Here we show that two closely related voltage-gated K channel proteins, mKv1.1 and mKv1.2, are present in multiple subcellular locations including cell somata, juxta-paranodal regions of myelinated axons, synaptic terminals, unmyelinated axons, specialized junctions among axons, and proximal dendrites. Staining patterns of the two channel polypeptides overlap in some areas of the brain, yet each has a unique pattern of expression. For example, in the hippocampus, both mKv1.1 and mKv1.2 proteins are present in axons, often near or at synaptic terminals in the middle molecular layer of the dentate gyrus, while only mKv1.1 is detected in axons and synaptic terminals in the hilar/CA3 region. In the cerebellum, both channel proteins are localized to axon terminals and specialized junctions among axons in the plexus region of basket cells. Strong differential staining is observed in the olfactory bulb, where mKv1.2 is localized to cell somata and axons, as well as to proximal dendrites of the mitral cells. This overlapping yet differential pattern of expression and specific subcellular localization may contribute to the unique profile of excitability displayed by a particular neuron.

Electrophysiological and anatomical techniques were used to determine the role, in the hippocampal circuitry, of local circuit neurons located at the oriens/alveus border (O/A interneurons). Intracellular recording from these cells showed that their response characteristics were clearly nonpyramidal: high input resistance, short membrane time constant, short-duration action potential, pronounced, brief afterhyperpolarizations (AHP), and nondecremental firing during intrasomatic depolarizing current pulses. Intracellular Lucifer yellow (LY) injection and subsequent fluorescence microscopy confirmed their nonpyramidal nature. O/A interneuron somata were bipolar or multipolar; their dendrites projected mostly parallel to the alveus, except for 1 or 2 processes that turned perpendicularly, and ascended through stratum oriens and pyramidale and into radiatum. Their axons were seen to branch profusely in stratum oriens and pyramidale. Simultaneous intracellular recordings from O/A interneurons and CA 1 pyramidal cells showed that pyramidal cells directly excite these interneurons. Major hippocampal afferents also directly excited the O/A interneurons. In a small number of interneuron-pyramidal pairs, stimulation of the O/A interneuron directly inhibited pyramidal cells. In one case, reciprocal connections were observed: The pyramidal cell excited the interneuron, and the interneuron inhibited the pyramidal cell. In 1 interneuron-to-interneuron pair, an inhibitory connection from O/A interneuron to stratum pyramidale interneuron was also observed. With intracellular HRP injections into O/A interneurons and subsequent electron microscopy, we observed that O/A interneuron axons made contacts with pyramidal and nonpyramidal cells. HRP-filled symmetric synaptic contacts were found on pyramidal cell dendrites and somata. HRP-filled axons also made contacts with pyramidal cell initial segments. HRP-filled O/A interneuron axon contacts were also found on nonpyramidal cell dendrites in stratum oriens. These electrophysiological and anatomical results suggest that O/A interneurons make synaptic contact with pyramidal cells and may mediate feedforward and feedback inhibition onto CA 1 pyramidal cells.

Following kainate lesions of hippocampal subfield CA3, the remaining CA 1 pyramidal cells become hyperexcitable. This lesion is of interest because, morphologically, it resembles the damage often seen in cases of temporal lobe epilepsy; it may provide insight into the consequences of such cell loss in humans. The hyperexcitability in CA 1 is associated with a loss of both early and late IPSPs. At long postlesion latencies (2-4 months) inhibition shows partial recovery and the hyperexcitability subsides. The intent of the present work was to determine if alterations in CA 1 excitability and functional inhibition postlesion are correlated with changes in morphologic and physiologic indicators of inhibitory interneuron function or with alterations in binding sites for inhibitory transmitters. Using GAD immunocytochemistry, we found no acute or chronic lesion-induced decrease in numbers of CA 1 interneurons or in qualitative characteristics of the pericellular distribution of their terminals in CA 1 stratum pyramidale. Intracellular recordings from identified cells in CA 1 indicated that putative interneurons were viable in hyperexcitable tissue. It was further observed that "recovery" in tissue studied 2-4 months postlesion primarily involved the early IPSP; the late IPSP failed to reappear. Quantitative in vitro autoradiographic analysis of 3H-flunitrazepam--a marker for the early IPSP associated GABAA receptor complex--indicated that hyperexcitability was associated with an increase in GABAA receptor number in CA 1; receptor binding returned to normal at long postlesion latencies as the early IPSP returned and hyperexcitability subsided. Finally, hyperexcitable pyramidal cells were found to retain their responsivity to exogenously applied GABA. These data indicate that much of the cellular machinery necessary for inhibition is retained in CA 1, despite lesion-induced hyperexcitability. We suggest that the acute loss of the IPSP after kainate lesion is due to a transient disconnection between inhibitory and excitatory elements in CA 1 and/or to a loss of normal afferent drive from CA3 onto some CA 1 interneurons. We further suggest that incomplete recovery can be explained by abnormalities that occur as neuroplastic rearrangements in response to deafferentation of CA 1. The relevance of these studies to human hippocampal necrosis and to other models of focal epilepsy is discussed.

The granule cell population response to perforant path stimulation decreased significantly within seconds following release of endogenous dynorphin from dentate granule cells. The depression was blocked by the opioid receptor antagonists naloxone and norbinaltorphimine, suggesting that the effect was mediated by dynorphin activation of kappa 1 type opioid receptors. Pharmacological application of dynorphin B in the molecular layer was effective at reducing excitatory synaptic transmission from the perforant path, but application in the hilus had no significant effect. These results suggest that endogenous dynorphin peptides may be released from a local source within the dentate molecular layer. By light microscopy, dynorphin-like immunoreactivity (dynorphin-LI) was primarily found in granule cell axons in the hilus and stratum lucidum with only a few scattered fibers evident in the molecular layer. At the extreme ventral pole of the hippocampus, a diffuse band of varicose processes was also seen in the molecular layer, but this band was not present in more dorsal sections similar to those used for the electrophysiological studies. Electron microscopic analysis of the molecular layer midway along the septotemporal axis revealed that dynorphin-LI was present in dense-core vesicles in both spiny dendrites and unmyelinated axons with the majority (74%) of the dynorphin-LI-containing dense-core vesicles found in dendrites. Neuronal processes containing dynorphin-LI were observed throughout the molecular layer. The results suggest that dynorphin release from granule cell processes in the molecular layer regulates excitatory inputs entering the hippocampus from cerebral cortex, thus potentially counteracting such excitation-induced phenomena as epileptogenesis or long-term potentiation.

Immunocytochemical techniques were used to examine the synaptic relations of inhibitory interneurons in the developing rabbit hippocampus. Glutamic acid decarboxylase (GAD), the synthesizing enzyme for the inhibitory neurotransmitter GABA, was found in interneurons of immature (8 d old) as well as mature (30 d old) tissue. GAD-immunoreactivity was seen in somata, dendrites, and axon terminals of interneurons at both ages. Electron-microscopic examination revealed that GAD-positive "terminals" in immature tissue were often not associated with the usual synaptic specializations, but were rather in simple apposition to the "postsynaptic" element. In mature tissue, GAD-positive terminals made symmetric contacts primarily with pyramidal cell somata, initial segments, and proximal dendrites. In addition, GAD-positive terminals synapsed onto both GAD-positive and GAD-negative interneuron profiles.