Functional rhythmogenic domains defined by astrocytic networks in the trigeminal main sensory nucleus

Abstract

Stimuli that induce rhythmic firing in trigeminal neurons also increase astrocytic coupling and reveal networks that define the boundaries of this particular population. Rhythmic firing depends on astrocytic coupling which in turn depends on S100β. In many nervous functions that rely on the ability of neuronal networks to generate a rhythmic pattern of activity, coordination of firing is an essential feature. Astrocytes play an important role in some of these networks, but the contribution of astrocytic coupling remains poorly defined. Here we investigate the modulation and organization of astrocytic networks in the dorsal part of the trigeminal main sensory nucleus (NVsnpr), which forms part of the network generating chewing movements. Using whole-cell recordings and the dye coupling approach by filling a single astrocyte with biocytin to reveal astrocytic networks, we showed that coupling is limited under resting conditions, but increases importantly under conditions that induce rhythmic firing in NVsnpr neurons. These are: repetitive electrical stimulation of the sensory inputs to the nucleus, local application of NMDA and decrease of extracellular Ca . We have previously shown that rhythmic firing induced in NVsnpr neurons by these stimuli depends on astrocytes and their Ca -binding protein S100β. Here we show that extracellular blockade of S100β also prevents the increase in astrocytic coupling induced by local application of NMDA. Most of the networks were small and remained confined to the functionally distinct area of dorsal NVsnpr. Disrupting coupling by perfusion with the nonspecific gap junction blocker, carbenoxolone or with GAP26, a selective inhibitor of connexin 43, mostly expressed in astrocytes, abolished NMDA-induced rhythmic firing in NVsnpr neurons. These results suggest that astrocytic coupling is regulated by sensory inputs, necessary for neuronal bursting, and organized in a region specific manner.

A Neuron–Glial Perspective for Computational Neuroscience

“…Stimulation of glutamatergic sensory fibers that project onto the trigeminal sensory-motor circuit for mastication can indeed activate astrocytes, triggering Ca 2+ -dependent release of the astrocyte-specific Ca 2+ -binding protein S100β. In turn, this results in a reduction of Ca 2+ in the ECS which promotes rhythmic bursting at frequencies that are compatible with those observed for voluntary chewing Condamine et al, 2018). In addition, medial basal hypothalamic astrocytes can control feeding behavior bidirectionally by purinergic gliotransmission, regulating appetite under both favorable and unfavorable conditions (Yang et al, 2015).…”

Astrocytic modulation of information processing by layer 5 pyramidal neurons of the mouse visual cortex

“…For astrocyte recordings, pipettes (resistance 4-6 MΩ) were filled with (in mM) 140 Kgluconate, 5 NaCl, 10 HEPES, 0.5 EGTA, 2 Tris ATP salt, 0.4 Tris GTP salt, pH 7.2-7.3, 280-300 mOsmol/kg and 0.2% biocytin. To identify astrocytes, Sulforhodamine 101 (SR101) labeling was performed in the holding chamber before the beginning of the July 9, 2020 8/79 experiment by adding to the aCSF solution 1 mM of SR101 for min at 34°C, and then min of wash out in aCSF (Condamine et al 2018;Kafitz et al 2008). Patched cells were considered astrocytes on the basis of their morphological characteristics (cell body ≤ ~10 µm), and when showing a linear current increase following application of a 600 ms voltage ramp from -120 to +110 mV in voltage-clamp mode and no action potential following step current injections in current-clamp mode (Condamine et al 2018;Panatier et al 2011;Serrano et al 2008).…”

“…Patched cells were considered astrocytes on the basis of their morphological characteristics (cell body ≤ ~10 µm), and when showing a linear current increase following application of a 600 ms voltage ramp from -120 to +110 mV in voltage-clamp mode and no action potential following step current injections in current-clamp mode (Condamine et al 2018;Panatier et al 2011;Serrano et al 2008). Additionally, astrocytes with a resting membrane potential above -60 mV in current clamp mode were rejected (Condamine et al 2018).…”

“…In some experiments, the following drugs were bath applied: the specific Nav 1.6 channel blocker 4,9-anhydrotétrodotoxine (4,9-TTX, 0.1 µM); the AMPA/kaïnate receptor antagonist 6-cyano-7nitroquinoxaline-2,3-dione (CNQX, 10 μM), the NMDA receptor antagonist (2R)-amino-5-phosphonovaleric acid (AP5, 26 μM or 50 µM) and the GABAA receptor antagonist gabazine (20 µM), the non-selective mGluR antagonist LY 341495 (100 µM) (Schoepp et al 1999), and the purinergic receptor antagonist suramin (50 or 100 µM) (Torres et al 2012). In some experiments, antibodies were applied as previously done (Condamine et al 2018;Morquette et al 2015) using larger pipettes (tip diameter 10-20 µm) that were carefully lowered to the surface of the tissue near the recorded cells. Each antibody of interest was slowly injected by applying a small pressure (0.1 psi) during 8-30 min before testing its effect on the recorded neuron.…”

“…Two antibodies were used: a monoclonal mouse anti-S100β antibody (Sigma S2532, lot 123M4881, 40 μg/mL, sodium azide salt 140 μM) or a mouse anti-myosin heavy chain type IIA antibody as a control injection (Developmental Studies Hybridoma Bank SC-71, lot 4-23-15, 477 μg/ml, sodium azide salt 340 μM). S100β was obtained from Inixium (Laval, Québec, Canada) in aliquots at a concentration of 14.0 mg/mL supplemented with glycerol to a final concentration of 50% and stored at -80°C as previously described (Condamine et al 2018). Before use, to eliminate the glycerol, aliquots of the protein were thawed and microdialysed overnight using a dialysis device (Float-A-Lyzer from Spectra/Por) floating in a 20 mM HEPES buffer containing 140 mM of NaCl at pH 7.4.…”