Silent plateau potentials, rhythmic bursts, and pacemaker firing: three patterns of activity that coexist in quadristable subthalamic neurons

Abstract

Subthalamic neurons display uncommon intrinsic behaviors that are likely to contribute to the motor and cognitive functions of the basal ganglia and to many of its disorders. Here, we report silent plateau potentials in these cells. These plateau responses start with a transient burst of action potentials that quickly diminish in amplitude because of spike inactivation and current shunt. The resulting interruption of spiking reveals a stable depolarization (up state) that clamps the cell membrane potential near -40 mV for several seconds. These plateau potentials coexist in single subthalamic neurons with more familiar patterns of burst and pacemaker firing. Within a narrow range of baseline membrane potentials (-67 to -60 mV), depolarization abruptly switches single cells from bistable to rhythmic bursts or tonic firing modes, thus selecting entirely distinct algorithms for integrating cortical and pallidal synaptic inputs.

Fig. 1.

Subthalamic plateau potentials in whole-cell recordings. (A) Prototypic plateau potentials elicited by a brief depolarization (Left; bias current, –6 pA) or hyperpolarization (Right; bias current, +2 pA). (B1–B3) Plateau potential elicited by pairing a current step and a train of three EPSPs (B3 Right) in a cell that did not show plateau responses to negative (B1) or positive (B2) current steps nor to synaptic stimulation alone (B3 Left).

Fig. 2.

Silent all-or-none plateau potentials. (A) Responses of a subthalamic neuron to current steps of increasing intensity (10–30 pA; duration, 50 ms; bias current, –12 pA). (B) In another cell, responses to single EPSPs (0.5 ms, 0.5 mA) in control conditions (eight superimposed traces) and after addition of CNQX (10 μM) and d-APV (25 μM; bias current, –8 pA). (C) In another subthalamic neuron, measures of input resistance during the down state (▴) (Inset) by using small (–1 pA) 2-s-long current steps (four averaged traces) and during the up state using larger current steps (+10, +20, and +30 pA; 2-s duration). For this experiment, plateau potentials were triggered repeatedly by a 200-ms, 50-pA current step and terminated with a 300-ms, –30-pA current step (bias current, –9 pA).

Fig. 3.

Plateau potentials and burst firing. (A) Responses to depolarizing currents (A1–A2) applied with the same (–13.5 pA) bias current for different duration (A1, 40 pA and 200 or 100 ms) or intensity (A2, 90 or 70 pA and 100 ms) and to hyperpolarizing currents applied with the same (+4 pA) bias current for different duration (A3, –20 pA and 1 or 0.8 s) or amplitude (A4, –25 or –15 pA and 1 s). Negative step currents (A1–A2, –50 pA and 200 ms; A3–A4, –120 pA and 60 ms) were used to terminate the plateau responses (see the occasional failure in A4). (B) Responses to EPSPs (500 μs and 0.5 mA) in another subthalamic neuron recorded with bias currents of –12, –10, or 0 pA. (C) Rebound responses to a train of IPSPs (50 μs, 100 μA, and 20 Hz) recorded in another cell with a bias current of –10 or 0 pA. (D) Plateau potentials, burst firing, and pacemaker activity in a subthalamic neuron recorded at ≈35°C. Current stimuli were adjusted so that depolarizing step currents (100 ms long) reached the same absolute value of +50 pA when the bias current was set at –10, –15, or 0 pA (Left). The 2-s-long hyperpolarizing current steps reached –20 or –30 pA (Right).

Fig. 4.

Quadristability in subthalamic neurons. The two cells shown here were challenged by positive step currents (100 ms) applied with a –13, –9, or –5 pA bias current (A) or by EPSPs (one or two 0.5-ms-long stimuli applied with a –3.5, –1, or +2 pA bias current) (B).

Fig. 5.

Subthalamic firing patterns in perforated-patch recordings. (A) Prototypic responses to current steps. (A1) The spontaneous spikes sampled for 40 min at the onset of the recording session (A1 Left, 10 s and 20 mV calibration bars) and the Lucifer yellow distribution at the end of the experiment before (A1 Center) and after (A1 Right) rupturing the patch. (A2 and A3) Cell responses to positive (A2 Left) and negative (A2 Right) current steps and to an EPSP (0.5-ms-long stimulus, 100-μA intensity) (A3; bias current of –22.5 pA). (B) In another cell, plateau potentials evoked by single or train of EPSPs (0.5-ms stimuli, 30-ms apart, bias current of –15 pA). (C) Rebound plateau response to a hyperpolarizing current step (–10 pA, 2 s long, bias current of –0.5 pA). (D) Bias currents required to silence subthalamic neurons in perforated-patch (n = 17) or whole-cell (n = 25) recordings (P < 0.001).

Fig. 6.

Plateau potentials in cell-attached recordings. Traces show the responses to hyperpolarizing voltage-steps [–150 mV for 6 s (A1) or –100 mV for 3 s (A2)]. In A1, the distribution of Lucifer yellow is shown before (a) and after (b) rupturing the patch at the end of the recording session. (Insets) Gray bars indicate the plateau responses, the asterisk indicates the initial burst of action potentials, the filled triangle indicates the action potential at the end of the plateau response, and the open triangle indicates the action potentials triggered by low-threshold Ca spikes.