N-Methyl-D-aspartate receptors contribute to excitatory postsynaptic potentials of cat lateral geniculate neurons recorded in thalamic slices ( excitatory amino acid / corticogeniculate / low-threshold calcium spike / optic tract / retinogeniculate )
H. Scharfman,S.-M. Lut,W. GUIDOt,P. Adams,S. M. SHERMANt
Abstract:Neurons of the cat's dorsal lateral geniculate nucleus were recorded intracellularly to study the contribution of N-methyl-D-aspartate (NMDA) receptors to excitatory posty tic potentials (EPSPs) and low-threshold calcium spikes. EPSPs were evoked by stimulation of retinogeniculate axons in the optic tract and/or corticogeniculate axons in the optic radiations; EPSPs from both sources were similar. These EPSPs had one or two components, and the second component had several characteristics of NMDA receptor-mediated events. For example, EPSP amplitude decreased when neurons were hyperpolarized and increased when stimulus frequency was increased; these EPSPs could also be blocked reversibly by application of the selective NMDA receptor antagonist DL2-amino-5-phosphonovaleric acid (APV). We also studied the influence ofNMDA receptors on low-threshold calcium spikes, which are large, voltageand calcium-dependent depolarizations that are often accompanied by high-frequency action potential discharge. APV blocked synaptically activated lowthreshold calcium spikes, but APV had no effect on lowthreshold calcium spikes that were elicited by current i 'ection. Therefore, APV does not appear to have a direct effect on the T-type calcium channel that is involved in generation of low-threshold calcium spikes. The voltage and frequency dependence of the NMDA receptor-mediated component of the EPSPs, as well as its ability to trigger low-threshold calcium spikes, provide for complex signal processing in the lateral geniculate nucleus. To examine mechanisms underlying retinogeniculate and corticogeniculate transmission, we used a slice preparation to record from geniculate neurons intracellularly while stimulating the retinal or cortical input. Of specific interest was the possible role of N-methyl-D-aspartate (NMDA) receptors in generating excitatory postsynaptic potentials (EPSPs), since activation of this receptor is associated with complex events (5-7). Thus, we compared EPSPs of geniculate neurons with the distinctive characteristics of NMDA-mediated EPSPs that have been described in other preparations. For example, in rat neocortex (8, 9) and hippocampus (10-12), NMDA receptor-mediated EPSPs contribute a component to the EPSP that peaks at a relatively long latency after the stimulus. This late component decreases in amplitude upon membrane hyperpolarization, a phenomenon attributed to the voltage-dependent block by magnesium ions of the ion channel that is coupled to the NMDA receptor (13, 14). It has also been shown that NMDA receptor-mediated EPSPs are enhanced by increasing frequency of stimulation (8, 12). In addition, NMDA receptor-mediated events may be identified by use of the selective NMDA receptor antagonist DL2-amino-5-phosphonovaleric acid (APV). Our results suggest that NMDA receptors contribute to retinogeniculate and corticogeniculate transmission and may be one of the factors that provides the lateral geniculate nucleus with the capacity for complex information processing in response to both retinal and cortical inputs. MATERIALS AND METHODS Brain slices were made from the lateral geniculate of cats that were deeply anesthetized (0.5-1.0% halothane in a 70:30 mixture of N2O/02), systemically paralyzed (5 mg of gallamine triethiodide followed by 3.6 mg of gallamine triethiodide per hr and 0.7 mg of d-tubocurarine per hr), artificially respired via a tracheal cannula, and positioned in a stereotaxic apparatus. To ensure that the cat was adequately anesthetized and maintained in stable condition, electrocardiograms, electroencephalograms, end-tidal CO2, and body temperature were continuously monitored. A rectangular area of skull overlying the right lateral geniculate nucleus was removed, and a block of tissue containing the lateral geniculate nucleus (-=1.5 cm anteroposteriorly and 3 cm mediolaterally) was excised and immediately placed in ice-cold buffer (124 mM NaCI/5 mM KCl/2 mM CaCI2/1.0 mM MgSO4/1.25 mM NaH2PO4/26 mM NaHCO3/10 mM dextrose, pH 7.4). Coronal slices (400 Am thick) were cut from the block of tissue with a vibratome while the entire block was immersed in cold, oxygenated buffer. Slices were immediately placed on a nylon mesh in a recording chamber at an interface of warmed (34°C), oxygenated (95% 02/5% C02) air and buffer. Neurons in layers A and A1 ofthe lateral geniculate nucleus (Fig. 1A) were impaled with glass microelectrodes (capillaryfilled borosilicate glass, 50-120 MfQ resistance, and filled with 4 M potassium acetate or 1-2% Lucifer yellow CH in 1 M LiCI). We saw no differences in electrophysiological properties or in synaptic potentials when using these different electrolytes. A high-input impedance amplifier with a bridge circuit was used (Axoclamp-2A). Stimulation of afferent pathways consisted of brief current pulses (0.1-0.5 mA; 20-200 ,s; typically 0.3 Hz) using a bipolar stimulating electrode (twisted, Teflon-coated, stainless steel wires; each wire diameter, 50 ,m). The stimulating electrode was placed on the slice surface, in the area of the optic tract or optic radiations, >100 ,um from the lateral geniculate nucleus and >500 ,um from the impaled cell (Fig. 1A). APV (Sigma) was applied in small droplets (<100 ,ul of 1 mM APV dissolved in 0.9% NaCl to the slice surface, or added to the perfusate as 25 ,M or 50 ,uM, dissolved in buffer). Abbreviations: NMDA, N-methyl-D-aspartate; EPSP, excitatory postsynaptic potential; APV, DL-2-amino-5-phosphonovaleric acid; IPSP, inhibitory postsynaptic potential. 4548 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Proc. Natl. Acad. Sci. USA 87 (1990) 4549 iN Optic A stim\ radiations