Activation of glutamate transporters in the locus coeruleus paradoxically activates descending inhibition in rats

Abstract

Descending noradrenergic inhibition is an important endogenous pain-relief mechanism which can be activated by local glutamate signaling. In the present study, we examined the effect of glutamate transporter activation by riluzole in the regulation of activity of locus coeruleus (LC) neurons, which provide the major inhibitory descending noradrenergic projection to the spinal cord. Local injection of riluzole into the LC dose-dependently reduced hypersensitivity in rats after L5-L6 spinal nerve ligation (SNL). This anti-hypersensitivity effect of LC-injected riluzole was blocked by intrathecal administration of the alpha2-adrenoceptor antagonist idazoxan and intra-LC co-injection of the AMPA antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and the gap-junction blockers, carbenoxolone (CBX) and meclofenamic acid (MEC). In brainstem slices from normal rats, riluzole increased phosphorylated cAMP response element binding protein (pCREB) expressing nuclei in dopamine-beta-hydroxylase (DbetaH) containing cells in the LC. This riluzole-induced pCREB activation in LC neurons was also blocked by CNQX and CBX. In the primary astrocyte culture, riluzole enhanced glutamate-induced glutamate release. Contrary to expectations, these results suggest that activation of glutamate transporters in the LC results in increase of extracellular glutamate signaling, possibly via facilitation of glutamate release from astrocytes, and activation of LC neurons to induce descending inhibition, and that this paradoxical action of glutamate transporters in the LC requires gap-junction connections.

Fig 1

Intra-LC injected riluzole reduces mechanical hypersensitivity after SNL surgery. SNL surgery reduced withdrawal threshold to pressure applied to the paw ipsilateral to injury. After development of hypersensitivity, rats received intra-LC injection of vehicle (n=7) or riluzole (0.2-5 μg, n=7-8). *p<0.05 vs. time 0 by one-way ANOVA. Riluzole 1 μg and 5 μg groups significantly differ from the vehicle group by two-way repeated measures ANOVA (p<0.05).

Fig 2

Effects of spinally and intra-LC injected antagonists on the anti-hypersensitive effect of LC-injected riluzole. Vehicle, riluzole (Ril, 1 μg), CNQX (0.3 μg), CBX (1 μg), MEC (1 μg), or their mixture were injected in the LC 15 min prior to the post-drug measurement in SNL rats (n=6-9). Intrathecal idazoxan (IT Ida, 30 μg) was administered 15 min prior to intra-LC injection of vehicle (n=7) or Ril (n=7), and the measurement was obtained 15 min later. * p<0.05 vs. vehicle. # p<0.05 vs riluzole alone.

Fig 3

Photomicrographs of DβH- and pCREB-immunoreactive neurons in brainstem slices from normal rats. Brainstem slices treated with vehicle (A-C) or riluzole (1 μM, D-F) for 30 min were stained with antibodies for DβH (Green) and pCREB (Red). Scale bar = 100 μm.

Fig 4

Quantification of riluzole concentration response on the number of pCREB expressing cells in brainstem slices from normal rats. Data are presented as percentage of pCREB-immunoreactivity in DβH-immunoreactive neurons. Brainstem slices were treated with vehicle (n=4) or riluzole (0.1 - 10 μM, n=4-5) for 30 min. *p<0.05 vs. vehicle.

Fig 5

Effects of AMPA and Gap-junction antagonists on riluzole-induced pCREB activation in brainstem slices from normal rats. (A, B, and C) Photomicrographs of DβH (Green)- and pCREB (Red)-immunoreactive neurons in brainstem slices treated with riluzole (Ril, 1 μM) alone (A) or riluzole plus CNQX (50 μM, B) and CBX (50 μM, C) for 30 min. Scale bar = 50 μm. (D) Quantification of effects of CNQX and CBX on riluzole-induced pCREB activation in brainstem slices. Data are presented as percentage of pCREB-immunoreactivity in DβH-immunoreactive neurons. Brainstem slices were treated with vehicle or riluzole (Ril, 1 μM) in the presence or absence of CNQX and CBX for 30 min. *p<0.05 vs. vehicle. # p<0.05 vs. riluzole alone.

Fig. 6

Effect of riluzole in glutamate-induced glutamate release from primary astrocyte culture. After loading with 1 μM glutamate (combination of both tritiated and unlabeled glutamate) for 1 hr, astrocytes were treated with glutamate (1-100 μM) with or without riluzole (1 μM) for 10 min. [3H]-glutamate (Glu) release is presented as a percentage of total radioactivity. [3H]-glutamate release in riluzole-treated group significantly differ from the vehicle group by two-way ANOVA (p<0.01). *p<0.05 vs. 0 μM.