Characterizing single-channel behavior of GluA3 receptors

Abstract

AMPA receptors play a major role in excitatory neurotransmission in the CNS and are involved in numerous neurological disorders. Agonists bind to each of four bilobed LBDs of this tetrameric receptor, and upon binding, the lobes close to envelope the agonist, leading to channel activation. However, AMPA receptors exhibit complex activation kinetics, the mechanism of which has not yet been determined. We report here single-channel studies of a homomeric AMPA receptor (GluA3) activated by the full agonist, glutamate, and a partial agonist, fluorowillardiine. Both agonists activate the channel to the same three open conductance levels but with different open probabilities in each level. The closed probability (P(c)) varied within records, particularly at low agonist concentrations. By sorting discrete segments of the record according to P(c) using the X-means algorithm, we defined five modes of activity. The kinetic behavior could then be analyzed for both agonists over a range of agonist concentrations with a relatively simple model (three closed states and two open states for each open conductance level). The structural mechanism underlying the modal behavior is not clear; however, it occurs on a timescale consistent with hydrogen bonding across the lobe interface in the LBD.

Figure 1

(A) Initial kinetic model that was used as a starting point but was found not to describe the data adequately. (B) Final kinetic model. The notation for the equilibrium is as follows: $K_{C21}=k_{C2\to C1}/k_{C1\to C2}$, $K_{O12a}=k_{O1a\to O2a}/k_{O2a\to O1a}$, $K_{O1ab}=k_{O1a\to O1b}/k_{O1b\to O1a}$, etc.

Figure 2

(A) A representative 6-s segment of a single-channel record of GluA3 activated by 50 μM FW. In this and all subsequent figures, the channel opens in a downward direction. (B) Amplitude histograms for an entire record obtained with 50 μM FW at 100 mV (top) and another obtained with 5 mM glutamate at 80 mV (bottom). Above each histogram is a representative 500-ms segment of the single-channel record with each of the three conductance levels indicated. The inset gives the percentage of each conductance level determined from a fit of four Gaussians to the amplitude histograms. For both the FW and glutamate records shown, the channel only exhibited M mode.

Figure 3

Dwell time histograms for the same records used to generate Fig. 2B. The number in the upper-left corner indicates the number of events.

Figure 4

Representative segments of single-channel records obtained with (A) 500 μM, (B) 50 μM, and (C) 10 μM FW showing different modes of activity (VH, H, M, and L).

Figure 5

(A) P_o for segments of a single-channel record with 50 μM FW. Closed times greater than t_crit were removed, and the horizontal lines indicate the length of the segment at a given P_o (y axis). The plotting symbol at the beginning and end of a segment indicates the mode as defined in the legend. (B) The same segment as in panel A but classified according to mode instead of P_o. The inset is a modal transition matrix for the modes.