LC-derived excitatory synaptic transmission to dorsal raphe serotonin neurons is inhibited by activation of alpha2-adrenergic receptors

Abstract

In the central nervous system, noradrenaline transmission controls the degree to which we are awake, alert, and attentive. Aberrant noradrenaline transmission is associated with pathological forms of hyper- and hypo-arousal that present in numerous neuropsychiatric disorders often associated with dysfunction in serotonin transmission. In vivo, noradrenaline regulates the release of serotonin because noradrenergic input drives the serotonin neurons to fire action potentials via activation of excitatory α1-adrenergic receptors (α1-A_R). Despite the critical influence of noradrenaline on the activity of dorsal raphe serotonin neurons, the source of noradrenergic afferents has not been resolved and the presynaptic mechanisms that regulate noradrenaline-dependent synaptic transmission have not been described. Using an acute brain slice preparation from male and female mice and electrophysiological recordings from dorsal raphe serotonin neurons, we found that selective optogenetic activation of locus coeruleus terminals in the dorsal raphe was sufficient to produce an α1-A_R-mediated excitatory postsynaptic current (α1-A_R-EPSC). Activation of inhibitory α2-adrenergic receptors (α2-A_R) with UK-14,304 eliminated the α1-A_R-EPSC via presynaptic inhibition of noradrenaline release, likely via inhibition of voltage-gated calcium channels. In a subset of serotonin neurons, activation of postsynaptic α2-A_R produced an outward current through activation of GIRK potassium conductance. Further, in vivo activation of α2-A_R by systemic administration of clonidine reduced the expression of c-fos in the dorsal raphe serotonin neurons, indicating reduced neural activity. Thus, α2-A_R are critical regulators of serotonin neuron excitability.

Fig. 1. Selective activation of LC system noradrenergic axons in the dorsal raphe produces an α1-A_R-EPSC.

A Schematic of coronal brain sections outlining the subregions of the LC system and observed expression of ChR2-eYFP (also see Fig. S2). B Representative maximum intensity projection confocal image of noradrenaline neurons in the LC of TH-Cre mice (TH immunostaining, cyan) expressing ChR2-eYFP (ChR2, magenta) after viral transduction following microinjection of AAV₅-Ef1α-DIO-ChR2-eYFP into the LC; scale bar, 100 μm. C Representative maximum intensity projection confocal image of two neurobiotin-filled dorsal raphe serotonin neurons immunostained with streptavidin (SA, yellow) surrounded by LC system-derived ChR2-eYFP-expressing axons (ChR2, magenta); scale bar, 50 μm. D Representative whole-cell current-clamp recording from an LC neuron expressing ChR2 demonstrating that ChR2 activation (10 pulses at 60 Hz, 5 ms, 473 nm) drove action potential firing. E Representative whole-cell voltage-clamp recording from a serotonin neuron in the dorsal raphe. Activation of ChR2 expressed in LC system noradrenergic axons (train of 5 or 10 pulses at 60 Hz, 5 ms, 473 nm, n = 4 mice) was sufficient to produce an α1-A_R-EPSC. Line and error bars represent mean ± SEM.

Fig. 2. Activation of α2-A_R inhibits the α1-A_R-EPSC via a presynaptic mechanism.

A Representative whole-cell voltage-clamp recording of the α1-A_R-EPSC in control conditions and in UK-14,304. B Representative whole-cell voltage-clamp recording of the α1-A_R-EPSC in an α2-A_R antagonist, idazoxan, and in UK-14,304 with idazoxan. C Plot of the percent remaining of the α1-A_R-EPSC after application of UK-14,304 in control conditions and in idazoxan (p = 0.0001, n = 15 and 5, Mann–Whitney test). D Representative whole-cell voltage-clamp recording of the α1-A_R-dependent inward current produced by focal iontophoretic application of noradrenaline (ionto) in control conditions and in UK-14,304. E Plot of the amplitude of the inward current produced by iontophoretic application of NA (I_NA) in control conditions (ctrl) and in UK-14,304 (UK, p = 0.46, n = 8), Wilcoxon matched-pairs signed rank test. F Time course of the inhibition of the α1-A_R-EPSC (black circles) by UK-14,304, shown in comparison to the lack of effect of UK-14,304 on I_NA (white circles). Line and error bars represent mean ± SEM. ns denotes not significant, * denotes statistical significance.

Fig. 3. Inhibitory effect of α2-A_R activation does not depend on membrane potential of innervating axon terminals.

A Representative whole-cell voltage-clamp recording of the α1-A_R-EPSC in BaCl₂ and in UK-14,304 with BaCl₂. B Plot of the amplitude of the α1-A_R-EPSC in BaCl₂ and in BaCl₂ with UK-14,304 ( + UK, p = 0.03, n = 6, Wilcoxon matched-pairs signed rank test). C Plot of the inhibition of the α1-A_R-EPSC by UK-14,304 in control conditions (ctrl), BaCl₂, 4-AP, and 10.5 mM extracellular potassium ([K⁺]_o) (ctrl vs. BaCl₂ p > 0.99; ctrl vs. 4-AP, p = 0.0013, ctrl vs. 10.5 [K⁺]_o, p < 0.99^, n = 15, 6, 7, and 18 respectively, Kruskal-Wallis test). D Representative whole-cell voltage-clamp recording of the α1-A_R-EPSC in 4-AP and in UK-14,304 with 4-AP. E Plot of the amplitude of the α1-A_R-EPSC in 4-AP and in 4-AP with UK-14,304 ( + UK, p = 0.016, n = 7, Wilcoxon matched-pairs signed rank test). F Time course of the inhibition of the α1-A_R-EPSC by UK-14,304 in control conditions (gray circles with black outline), in 10.5 mM [K⁺]_o (black circles) and in 4-AP (gray circles). Depolarizing innervating axon terminals by increasing [K⁺]_o from 2.5 to 10^.5 mM did not prevent the inhibition of the α1-A_R-EPSC by activation of α2-A_R with UK-14,304; shown in representative traces (G, H) and in grouped data (I, p < 0.0001, n = 18, Wilcoxon matched-pairs signed rank test). Line and error bars represent mean ± SEM. * denotes statistical significance.

Fig. 4. Activation of α2-A_R produces GIRK-mediated outward current in a subset of serotonin neurons.

A Representative whole-cell voltage-clamp recordings of the outward current produced by UK-14,304 in a subset of neurons (responders) while other neurons were unaffected (non-responders). We note that the inhibition of the α1-A_R-EPSC (evoked at each arrow) to UK-14,304 was observed in both responders and non-responders. B Plot of the amplitude of the UK-14,304-induced outward current in responders (res) and non-responders (non). C UK-14,304 produced a small decrease in membrane resistance, indicative of opening of ion channels, in responders (res, p = 0.015, n = 17, Wilcoxon matched-pairs signed rank test) but not non-responders (non, p = 0.71, n = 17, Wilcoxon matched-pairs signed rank test). D Representative whole-cell voltage-clamp recordings of the current produced by voltage ramps (8 mV/10 ms) that were used to determine the current-voltage relationship of the UK-14,304-induced current by subtracting the current in control conditions from the current in UK-14,304. Capacitive transient has been truncated for clarity. E Current-voltage relationship of the UK-14,304-induced current measured in 10.5 mM [K⁺]_o in responders (res) and non-responders (non). Line and error bars (B, C) and gray shading (E) represent mean ± SEM. ns denotes not significant, * denotes statistical significance.

Fig. 5. In vivo activation of α2-A_R reduces c-fos expression in dorsal raphe serotonin neurons.

A Representative maximum intensity projection confocal image of the dorsal raphe. Sections were immunostained for tryptophan hydroxylase (TPH2, gray) labeling the serotonin neurons, scale bar: 100 μm. The box indicates the midline area where analyses were performed from ×40 images. B–E A single i.p. injection of clonidine reduced the expression of c-fos in serotonin neurons shown in (B, C) representative maximum intensity projection confocal ×40 images of the dorsal raphe. Sections were immunostained for tryptophan hydroxylase (TPH2, cyan) labeling the serotonin neurons and c-fos (magenta), scale bar: 50 μm; (D) in grouped data (p = 0.032), n = 18 mice/group, n = 17,233 & 15,717 TPH2⁺ neurons in saline and clonidine respectively; (E) in both male and female mice (Two-way ANOVA: main effect of treatment, p = 0.022; main effect of sex, p = 0.37; interaction, p = 0.93; n = 9 mice/group). Line and error bars represent mean ± SEM. * denotes statistical significance.