Systemic dexmedetomidine augments inhibitory synaptic transmission in the superficial dorsal horn through activation of descending noradrenergic control: an in vivo patch-clamp analysis of analgesic mechanisms

Abstract

α2-Adrenoceptors are widely distributed throughout the central nervous system (CNS) and the systemic administration of α2-agonists such as dexmedetomidine produces clinically useful, centrally mediated sedation and analgesia; however, these same actions also limit the utility of these agents (ie, unwanted sedative actions). Despite a wealth of data on cellular and synaptic actions of α2-agonists in vitro, it is not known which neuronal circuits are modulated in vivo to produce the analgesic effect. To address this issue, we made in vivo recordings of membrane currents and synaptic activities in superficial spinal dorsal horn neurons and examined their responses to systemic dexmedetomidine. We found that dexmedetomidine at doses that produce analgesia (<10 μg/kg) enhanced inhibitory postsynaptic transmission within the superficial dorsal horn without altering excitatory synaptic transmission or evoking direct postsynaptic membrane currents. In contrast, higher doses of dexmedetomidine (>10 μg/kg) induced outward currents by a direct postsynaptic action. The dexmedetomidine-mediated inhibitory postsynaptic current facilitation was not mimicked by spinal application of dexmedetomidine and was absent in spinalized rats, suggesting that it acts at a supraspinal site. Furthermore, it was inhibited by spinal application of the α1-antagonist prazosin. In the brainstem, low doses of systemic dexmedetomidine produced an excitation of locus coeruleus neurons. These results suggest that systemic α2-adrenoceptor stimulation may facilitate inhibitory synaptic responses in the superficial dorsal horn to produce analgesia mediated by activation of the pontospinal noradrenergic inhibitory system. This novel mechanism may provide new targets for intervention, perhaps allowing analgesic actions to be dissociated from excessive sedation.

Keywords: Alpha2 adrenoceptor; Dexmedetomidine; Noradrenaline; Patch-clamp analysis; Spinal cord.

Figure 1. Dose-dependent anti-nociceptive and sedative action of systemic dexmedetomidine

A) Paw withdrawal thresholds to mechanical stimuli were measured in conscious animals using the Dynamic Plantar Aesthesiometer 20 min after intraperitoneal administration of dexmedetomidine (DEX, 0.01 to 10 μg/kg, n = 13). Dexmedetomidine significantly increased the withdrawal threshold at doses of 1 μg/kg (*p < 0.05) and 10 μg/kg (**p < 0.01) compared control prior to drug administration (n = 13, one-way ANOVA and Dunnett’s post hoc test). B) The sedation rating scores were assessed 20 minutes after intraperitoneal dexmedetomidine (DEX, 0.01 to 10 μg/kg, n = 6). Median scores were significantly decreased at 1 μg/kg (*p < 0.05) and 10 μg/kg (**p < 0.01) compared with pre-administration control (Kruskal-Wallis test followed by the Stell-Dwass test) although most of the animals were considered to be still awake or only mildly sedated in this dose range. Only those animals receiving 30 μg/kg showed evidence of moderate to heavy sedation which was a significantly greater degree of sedation than that produced by either 1 or 10 μg/kg (p < 0.05).

Figure 2. Effect of systemic dexmedetomidine on SG neuron excitability

A) To look for α₂-adrenoceptor mediated activation of potassium currents, we voltage clamped SG neurons (holding potential of −70mV with a potassium-based intracellular solution). The application of low doses of dexmedetomidine (1 μg/kg) did not elicit any detectable outward currents (left trace). In contrast, much higher doses of dexmedetomidine (30-75 μg/kg) were required to produce any sign of an outward current (right trace). B) In the presence of tetrodotoxin (TTX), systemic administration of dexmedetomidine (1 μg/kg) did not enhance miniature EPSCs. The traces below are shown on an expanded time base to demonstrate the resolution of individual miniature EPSCs. The cumulative probability plots for the inter-event interval and amplitude of miniature EPSCs (from the trace in B), showing dexmedetomidine has no effect on either the frequency or amplitude (p > 0.05, Kolmogorov-Smirnov test).

Figure 3. Systemic dexmedetomidine dose-dependently enhances spontaneous IPSCs in the SG

A) Systemic dexmedetomidine (left trace, 0.1 μg/kg) and clonidine (right trace, 40 μg/kg) elicited barrages of spontaneous IPSCs (holding potential of 0 mV with a cesium-based based solution). The traces below are shown on an expanded time base to demonstrate the individual IPSCs. B) Dexmedetomidine dose-dependently elicits barrages of spontaneous IPSCs. C) Summary plot showing the enhancement of IPSCs (open circles) at dexmedetomidine doses (<10 μg/kg) that are lower than those required to induce outward currents (closed circles). D) The cumulative probability plots for the inter-event interval and amplitude of spontaneous IPSCs (from the left trace in A). Dexmedetomidine at a dose of 0.1 μg/kg significantly increased the frequency (p < 0.01) and amplitude (p < 0.01) of spontaneous IPSCs (Kolmogorov-Smirnov test).

Figure 4. Dexmedetomidine does not enhance IPSCs by a direct spinal action

A) Spinal superfusion of dexmedetomidine (10 μM) had no effect on spontaneous IPSCs. In the same neuron, subsequent systemic application of dexmedetomidine (1 μg/kg) enhanced IPSCs. B) Recordings from SG neurons in rats spinalized at the cervical level showed that systemic application of dexmedetomidine (1 μg/kg) no longer changed IPSC frequency or amplitude. However, spinal application of noradrenaline (NA, 50 μM) was able to dramatically facilitate IPSCs in the same neuron. C) Summary showing effects of direct spinal superfusion of dexmedetomidine (0.1 – 10 μM), and systemic dexmedetomidine application in spinalized rats on normalized synaptic charge of spontaneous IPSCs.

Figure 5. Spinal noradrenaline and systemic dexmedetomidine facilitate IPSCs in the same SG neurons

A) In this neuron the facilitation of IPSCs by local superfusion of noradrenaline (NA, 50 μM, shown in upper trace) and systemic dexmedetomidine is blocked by spinal application of the α₁-adrenoceptor antagonist prazosin (10 μM, lower trace). Data in A were obtained from the same neurons shown in Fig. 4A. B) There was a strong correlation between the facilitatory actions of spinal noradrenaline and systemic dexmedetomidine on spontaneous IPSCs evoked in the same SG neurons (linear regression R² = 0.96). C) Summary chart showing the facilitatory action of systemic dexmedetomidine on spontaneous IPSCs which is blocked in the presence of spinal prazosin (an α₁ antagonist, 10 μM) but not by yohimbine (an α₂ antagonist, 4 μM).

Figure 6. Activation of LC neurons by low-dose systemic dexmedetomidine

A) The characteristic biphasic response with excitation followed by inhibition of LC neuron firing by cutaneous pinch stimulation applied to the contralateral hind paw (cell-attached patch recording). B) Extracellular recording showing several discriminated spontaneous LC units (right panels). For each cell the firing rate was increased by low dose systemic dexmedetomidine (1 μg/kg). However, higher doses of dexmedetomidine (30 μg/kg) strongly inhibited the firing of the same neurons. C) Normalized firing rates of LC neurons after systemic dexmedetomidine (1-30 μg/kg). The firing rate was increased in 5 of 7 LC neurons at 1 μg/kg. D) Normalized LC firing rates after superfusion of Dexmedetomidine over the floor of the 4^th ventricle. Firing rates were increased in a substantial proportion of the cells (5/11) at 1nM Dexmedetomidine but the majority of the cells were inhibited by the 1uM dose. Dashed lines in C and D show 20% increase or decrease in normalized firing frequency.

Figure 7. Schematics summarizing the proposed mechanisms of action of systemic dexmedetomidine

A) Low dose systemic dexmedetomidine (DEX) facilitates spinal IPSCs by disinhibiting a noradrenergic descending pathway. The spinal noradrenaline (NA) acts via pre- and postsynaptic α₁ adrenoceptors on the GABAergic and glycinergic inhibitory neurons to facilitate inhibitory synaptic transmission onto SG neurons. In contrast, high dose dexmedetomidine inhibits the LC, perhaps leading to sedation, and also directly activates α₂-adrenoceptors on the SG neurons to induce outward potassium currents that hyperpolarize and inhibit the cells producing analgesia by a different mechanism. B) Diagram showing the multiple actions of systemically administered dexmedetomidine at different doses. The analgesia and the facilitation of spontaneous IPSCs appear with low doses of systemic dexmedetomidine (0.1-1 μg/kg). At these doses dexmedetomidine enhanced the activity of LC neurons. Higher doses (10-75 μg/kg) of dexmedetomidine were needed to induce outward currents in SG neurons while also directly inhibiting the LC perhaps producing the known sedative effect.