Serotonin excites fast-spiking interneurons in the striatum

Abstract

Fast-spiking interneurons (FSIs) control the output of the striatum by mediating feed-forward GABAergic inhibition of projection neurons. Their neuromodulation can therefore critically affect the operation of the basal ganglia. We studied the effects of 5-hydroxytryptamine (5-HT, serotonin), a neurotransmitter released in the striatum by fibres originating in the raphe nuclei, on FSIs recorded with whole-cell techniques in rat brain slices. Bath application of serotonin (30 microm) elicited slow, reversible depolarizations (9 +/- 3 mV) in 37/46 FSIs. Similar effects were observed using conventional whole-cell and gramicidin perforated-patch techniques. The serotonin effects persisted in the presence of tetrodotoxin and were mediated by 5-HT(2C) receptors, as they were reversed by the 5-HT(2) receptor antagonist ketanserin and by the selective 5-HT(2C) receptor antagonist RS 102221. Serotonin-induced depolarizations were not accompanied by a significant change in FSI input resistance. Serotonin caused the appearance of spontaneous firing in a minority (5/35) of responsive FSIs, whereas it strongly increased FSI excitability in each of the remaining responsive FSIs, significantly decreasing the latency of the first spike evoked by a current step and increasing spike frequency. Voltage-clamp experiments revealed that serotonin suppressed a current that reversed around -100 mV and displayed a marked inward rectification, a finding that explains the lack of effects of serotonin on input resistance. Consistently, the effects of serotonin were completely occluded by low concentrations of extracellular barium, which selectively blocks Kir2 channels. We concluded that the excitatory effects of serotonin on FSIs were mediated by 5-HT(2C) receptors and involved suppression of an inwardly rectifying K(+) current.

Fig. 1

Effects of serotonin on FSIs in the presence of TTX. (A) Electrophysiological properties of an FSI revealed by current injections (± 100 pA). (B) Rhythmic intermittent firing was observed in another FSI depolarized by a steady current injection. (C) Serotonin (30 μm) reversibly depolarized an FSI in the presence of TTX (1 μm). Input resistance was monitored using current pulses (15 pA, 1 s) applied every 10 s. During the depolarization induced by serotonin, the FSI was manually repolarized to control level, to measure input resistance. Expanded traces are averages of 20 voltage deflections elicited by consecutive current pulses in each condition (control, serotonin and washout). No significant changes in input resistance were observed.

Fig. 2

Serotonin effects are mediated by 5-HT₂ receptors. (A) Serotonin depolarized an FSI in the presence of TTX (1 μm). Subsequent addition of ketanserin (10 μm) caused the FSI to hyperpolarize to a level more negative than control. During the depolarization induced by serotonin, the FSI was transiently repolarized to control level to measure input resistance (not significantly different from control). (B) In this experiment, an FSI (recorded without steady current injection) did not display any spontaneous firing activity in control solution. Serotonin induced a depolarization accompanied by spontaneous bursts of action potentials. Subsequent application of the 5-HT₂ receptor antagonist ketanserin (10 μm) repolarized the FSI and abolished spontaneous firing. (C) Another FSI, with a resting membrane potential of −69 mV, was depolarized (∼9 mV) by the 5-HT₂ receptor agonist α-methyl-5-HT (30 μm). Subsequent addition of ketanserin reversed these effects. All vertical deflections in the slow time-scale trace are depolarizations induced by current steps (50 pA, 1 s), which were delivered to monitor the excitability of the FSI. Three representative responses to such current steps in different pharmacological conditions are shown below the slow time-scale trace. (D) In this experiment, bath application of serotonin caused a depolarization (∼12 mV) from a resting membrane potential of −74 mV in an FSI; in the presence of serotonin, application of the selective 5-HT_2C receptor antagonist RS 102221 caused the membrane to repolarize to levels slightly more negative (−76 mV) than that observed in control solution. Washout of RS 102221 (still in the presence of serotonin) caused the membrane potential to depolarize again to a level similar to that observed in the presence of serotonin before RS 102221 application; subsequent reapplication of RS 102221 (still in the presence of serotonin) caused the FSI to repolarize again at −76 mV.

Fig. 3

Effects of serotonin on FSI excitability. (A) In a representative FSI, suprathreshold depolarizing current pulses were applied in control solution, in the presence of serotonin (which had caused a 4 mV depolarization) and after serotonin washout. Five consecutive traces (interval 10 s) are superimposed for each condition. Serotonin caused a large decrease in the latency of the first spike and these effects were reversed on washout. Black traces were recorded in control solution, before (left panel) and after (right panel) serotonin application; grey traces were recorded in the presence of serotonin and are truncated earlier to avoid overlapping of the second spike with the first spike of the traces recorded in control solution. (B) These traces, taken from the same experiment as in A, show that in the presence of serotonin the first inter-spike interval (ISI) was strongly reduced; it can also be appreciated that the temporal scattering of the second spike evoked by a current step is greatly reduced in the presence of serotonin, resulting in a much lower variability of the first ISI. (C) Average results for seven FSIs to which the protocol illustrated in A and B was applied; the average first ISI and the average latency of the first spike for each FSI are represented by a point in a scatter plot in control solution (open circle) and another point in serotonin (black circle). The points in the two conditions for each FSI are connected by a line. A strong decrease in both parameters in the presence of serotonin is apparent for each cell. (D) The half-width of the action potentials evoked by current steps in control solution and in the presence of serotonin did not differ significantly. Average of the first spikes elicited by 10 consecutive steps (delivered at 10 s intervals) for each condition. Dashed lines indicate the threshold and peak of each action potential; double arrowed vertical bars represent the action potential amplitude; double arrowed horizontal bars represent the action potential half-width.

Fig. 4

Membrane conductances modulated by serotonin revealed by voltage-clamp experiments. (A) A representative example of the voltage-clamp experiments carried out in the presence of TTX. Voltage steps (1 s) were applied to an FSI from the holding value (−80 mV) to levels between −100 and −40 mV (in 10 mV increments), in control solution and in the presence of serotonin. (B) Steady-state currents (recorded at the end of each voltage step) plotted vs. voltage in control solution (black) and in the presence of serotonin (grey) for the FSI of A. (C) Voltage dependence of the serotonin-suppressed currents in the same FSI. The steady-state current induced by serotonin was calculated, for each voltage, by subtracting the membrane current measured in control solution from that measured in the presence of serotonin. Serotonin-suppressed currents reversed around −100 mV and displayed prominent inward rectification.

Fig. 5

Serotonin-suppressed current has small slope conductance at FSI resting potential. (A) Diamonds represent the average steady-state current suppressed by serotonin at different voltages for four FSIs in which the protocol illustrated in A was carried out (error bars represent SEs). Superimposed (in grey) is a curve of best fit for these points (see Materials and methods for details). The area shaded in grey is centred around the average resting membrane potential of FSIs and spans two SDs (−69 ± 5 mV). (B) Estimated slope conductance associated with the serotonin-suppressed current, as a function of voltage. The curve of best fit shown in A was used to obtain this slope conductance. As in A, the area shaded in grey is centred around the average resting membrane potential of FSIs and spans two SDs (−69 ± 5 mV). The two horizontal dashed lines enclose the levels (between −0.15 and 0.25 nS) attained by the average slope conductance within this voltage range, where most resistance measurements were carried out.

Fig. 6

Barium (100 μm) occludes the excitatory effects of serotonin. (A) In this FSI, application of serotonin caused a reversible 10 mV depolarization, from a resting membrane potential of −70 mV; this depolarization was accompanied by spontaneous action potentials. After washout of serotonin effects, application of barium (100 μm) caused a depolarization (∼ 18 mV), also accompanied by spontaneous action potentials. In the continuous presence of barium, the FSI was manually repolarized to −70 mV (as indicated by the bottom trace depicting the injected current); under these conditions, subsequent application of serotonin failed to affect the FSI membrane potential. (B) Expanded traces (from the experiment in A) showing the spontaneous firing activity observed in the presence of serotonin (left) and barium (right).