66 Bosma, M. M. and Hille, B. (1989) Proc. NatlAcad. Sci. USA 86, 2943-2947 67 Bosma, M. M., Bernheim, L., Leibowitz, M. D., Pfaffinger, P. J. and Hille, B. (1990) in G Proteins and Signal Transduction (Nathanson, N. M. and Harden, T. K., eds), pp. 43-59, Rockefeller University Press 68 Kirkwood, A., Simmons, M. A., Mather, R. J. and Lisman, J. (1991) Neuron 6, 109-1014 69 Akasu, T. eta/. (1993) Neuron 11, 1133-1145 70 Selyanko, A. A. and Brown, D. A. Soc Neurosci. Abstr. (in press) 71 Meriney, S. D., Gray, D. B. and Pilaf, G. R. (1994) Nature 369, 336-339 72 Divers~-Pierluissi, M. hnd Dunlap, K. (1993) Neuron 10, 753-760 73 Selverston,A. I. (1993) Int. Rev. CytoL 147, 1-24 Cellularand synapticlocalizationof NMDA and non-NMDA receptorsubunitsin neocortex:organizational features related to corticalcircuitry, function and disease George W. Huntley, James C. Vickers and John H. Morrison George W. Huntley andJohnH. Mort~son areat the Fishberg Research Center for Neurobiology, Box 1065, The Mount Sinai Schoolof Medicine, One Gustave L. Levy Place,NY 10029, USA, andJames C ricketsis at the Dept of Pathology, Clinical School, University of Tasmania,Hobart, Tasmania. Excitatory amino acid (EAA) receptors are an import- ant component of neocortical circuitry as a result of their role as the principal mediators of excitatory syn- @tic activity, as well as their involvement in use- dependent modifications of synaptic efficacy, excito- toxicity and cell death. The diversity in the effects generated by EAA-receptor activation can be attributed to multiple receptor subOrpes, each of which is composed of multimeric assemblies of functionally distinct recep- tor subunits. The use of subunit-specific antibodies and molecular probes now makes it feasible to localize individual receptor subunits anatomically with a high level of cellular and synaptic resolution. Initial studies of the distribution of immunocytochemically localized EAA-receptor subunits suggest that particular subunit combinations exhibit a differential cellular, laminar and regional distribution in the neocortex. While such patterns might indicate that the functional heterogeneffy of EAA-receptor-linked circuits, and the cell types in which they operate, are based partly on differential subunit parcellation, a definitive integration of these anatomical details into current schemes of cortical circuitry and organization awaits many further studies. Ideally, such studies should link a high level of mol- ecular precision regarding subunit localization with synaptic details of identified connections and neuro- chemical features of neocortical cells. Glutamate and aspartate are the excitatory amino acid (EAA) neurotransmitters of the majority of intrinsic cerebral cortical neurons (pyramidal cells and spiny stellate cells) and of thalamocortical relay cells ~. The excitatory synaptic activity conveyed by these cells is mediated by three pharmacologically defined subtypes of ionotropic glutamate receptors: D,L-o~-amino-3- hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/ low-affinity k.ainate (AMPA receptors), high-affinity kainate (kalnate receptors) and N-methyl-D-aspartate (NMDA) receptors. Molecular cloning and functional expression techniques have demonstrated that each of the pharmacologically defined EAA-receptor sub- types is composed of multimeric assemblies of subunit proteins, details of which have been summarized in several recent reviews (for example, see Refs 2-4). Based on common electrophysiological and pharma- cological properties and similar gene sequences, distinct subunit families are recognized: subunits GIuR1-4 compose the AMPA receptors; subunits GIuR5-7 and the kainate-binding proteins KA1 and KA2 compose the kainate receptors; and subunits NMDAR1 and NMDAR2A-D compose the NMDA receptors. A further level of molecular and functional diversity arises from the existence of many subunits in several alternatively spliced forms. In addition, many subunit transcripts are subject to RNA editing, a mechanism that appears to be particularly important for determining subunit properties such as per- meability to Ca2+, and rectification characteristics 3. Although the precise subunit composition of EAA receptors at any specific postsynaptic site is largely unknown, different combinations of subunits yield functionally distinct receptor-channels when ex- pressed in Xenopus oocytes or mammalian cells. For example, receptors composed of only NMDAR1 subunits yield functional channels displaying low- amplitude elicited currents, whereas receptors con- sisting of any of the NMDAR2 subunits without NMDAR1 do not yield functional channels. However, combining NMDAR1 with any of the NMDAR2 sub- units greatly potentiates elicited currents, suggesting that NMDAR1 might be an obligatory component of all NMDA receptors4. Moreover, different combinations of NMDA2A-D with NMDAR1 yield channels that vary in sensitivity to Mg 2+ block, glycine potentiation, and pharmacological properties 4. In addition, channels assembled from combinations of GluR1, GluR3 or GIuR4 display strong inward rectification and are permeable to Ca2+ when activated, while the incor- poration of GluR2 into any combination of AMPA- receptor subunits yields channels displaying linear or outwardly rectifying current-voltage (l-V) relation- ships, and also prevents channel permeability to Ca2+ (Ref. 3). The anatomical localization of EAA-receptor sub- types has been principally by autoradiographic methods5, which are now limited in that individual ~ubnnits or splice variants cannot be distinguished, nor is resolution at the cellular or synaptic level possible. In situ hybridization studies in rats using subunit-specific probes were the first studies to suggest that the emerging molecular complexity of EAA subunits and their splice variants is paralleled by a high degree of anatomical specificity in the distri- bution of subunits across functionally ,distinct cell types and regions6-8. The generation of subunit- specific antibodies and molecular probes makes it 536 © 1994, ElsevierScience Ltd TINS, Vol. 17, NO. 12, 1994