Activating an anterior nucleus gigantocellularis subpopulation triggers emergence from pharmacologically-induced coma in rodents

Abstract

Multiple areas within the reticular activating system (RAS) can hasten awakening from sleep or light planes of anesthesia. However, stimulation in individual sites has shown limited recovery from deep global suppression of brain activity, such as coma. Here we identify a subset of RAS neurons within the anterior portion of nucleus gigantocellularis (aNGC) capable of producing a high degree of awakening represented by a broad high frequency cortical reactivation associated with organized movements and behavioral reactivity to the environment from two different models of deep pharmacologically-induced coma (PIC): isoflurane (1.25%-1.5%) and induced hypoglycemic coma. Activating aNGC neurons triggered awakening by recruiting cholinergic, noradrenergic, and glutamatergic arousal pathways. In summary, we identify an evolutionarily conserved population of RAS neurons, which broadly restore cerebral cortical activation and motor behavior in rodents through the coordinated activation of multiple arousal-promoting circuits.

Fig. 1

Pharmacological activation of aNGC neurons restores arousal. a Schematic depicts the loss of right reflex. Inset: electrode location in the cingulate cortex (Cg. Ctx) and aNGC for LFPs and single units during bicuculline (Bic) microinjections in aNGC. b Schematic shows awake animal after Bic. c Cortical LFPs and normalized power spectrogram show increased power seconds after microinjection of 10 mM Bic in aNGC (purple drops represent Bic injection) under constant isoflurane 1.25%. vol. (pink line). Firing rate of high-active (HA) aNGC neurons (bottom panel) increased before changes in cortical activity. Firing rate remained high during movement onset (dashed line). d Average spectrogram of five animals before and after aNGC disinhibition with Bic. Color bar (power in decibels; dB). e Average firing rate of high (n = 17) and low (n = 3) active neurons before and after Bic microinjection. The data shown as mean ± s.e.m. **p = 0.01 and *p = 0.05; t test. f Graph illustrates sensor responses before (n = 24) and after Bic microinjection. n = 24. The data represented as mean ± s.e.m. g Prominent increase in the relative gamma frequency band (30–100 Hz) at the moment of awakening (onset of movement) after Bic injection (n = 24). h Loss of righting reflex caused by insulin overdose (yellow line) resulting in a coma state. i Recovery of the righting posture after Bic injection in aNGC of hypoglycemic-comatose mice. j LFP and spectrogram shows the deep state of cortical activity suppression prior to Bic microinjection and transition to increased power in low- and high-frequency oscillations. Onset of movement occurred few seconds after injection. Firing rate increased in aNGC cells and decreased in LA active neurons. k Average spectrogram shows power increase after drug injection (n = 5). l Firing rate of HA cells (n = 5) and LA (n = 5) before and after aNGC disinhibition in comatose mice, paired t test, *p = < 0.05. m Sensor response before (n = 3) and after (n = 4) Bic injection and after noxious stimulation (n = 17). Kolmogorov–Smirnov test (KST), ***p = < 0.001. n Gamma oscillations increased during onset of movement after Bic microinjection. Time includes 5 s before and 10 s after onset of movement (n = 114). Source data are provided as a Source Data file

Fig. 2

Optogenetic stimulation of aNGC-Vglut2 + cells induces arousal. a Schematic shows injection area of AAV-packaged Cre-inducible light-gated cationic ion channel in aNGC and location of LFP electrodes in the cingulate cortex and optic fiber placement for stimulation in aNGC in a sagittal mouse brain section. b Schematic depicts an anesthetized mouse (isoflurane 1.25% vol) optogenetically stimulated. c Fluorescence image of aNGC showing location of optic fiber in a Vglut2-Cre mouse injected with AAV expressing ChR2-mCherry (red). Scale bar: 200 µm. 7n, seven nerves; Pyr, pyramidal tract. ChR2-mCherry was expressed within ~600 µm from injection site. d LFP (raw data) and spectrogram plotted as the difference from the temporal mean showing increases in high frequencies and decreases in low frequency cortical oscillations (< 4 Hz) in phase with photostimulation (5s-30 Hz stimulation). Color bar represents power in decibels. LFP transitions abruptly to a desynchronized state with each photostimulation. Note, the recruitment of firing of a Vglut2 + aNGC neuron occurred during photostimulation in PIC (isoflurane 1.25%). e Average spectrogram of 135 laser pulses aligned to laser onset. f Mean firing rates (mean ± s.e.m.) of aNGC cells (n = 21) before, during photostimulation (on) and after laser ceased (off) ***p = < 0.001 (one-way ANOVA). g Quantification of arousal score and tail movement. h Before and after laser stimulation in animals expressing ChR2 or mCherry. Kolmogorov–Smirnov test (KST), *p = < 0.05. i, j Schematic of hypoglycemic-comatose mouse before and after a 5 s laser stimulation. A prolonged stimulation promoted a full right up posture. k LFP and spectrogram while performing optogenetic stimulation. Each laser pulse increased power in all frequencies with movement of tail and limbs (see Supplementary Video 2) and modulated glutamatergic cells (k, m). l Power spectral density using total power from over 33 laser pulses aligned to laser onset. m Firing rate (mean ±  s.e.m.) of aNGC cells (n = 3) before and after laser onset ***p = < 0.001 (one-way ANOVA). n Arousal score and sensor responses (o) obtained after laser stimulation using yellow light (n = 32) and blue light (n = 30). Animals show strong responses after noxious stimulation (pinch; n = 6) and smell (odor; n = 5) 2 s after laser stimulation ended. KST, ***p < 0.001and **p < 0.01. Source data are provided as a Source Data file

Fig. 3

aNGC recruits ventral and dorsal pathways to promote arousal. a Depicts coronal projections of reference annotation, averaged registered density maps, and p-value maps of areas with significantly higher c-Fos + cell densities after bicuculline microinjections (n = 4; mice) in aNGC under constant anesthetic (Isoflurane 1.25% vol). Same areas show c-Fos + cell densities in mice injected with saline (n = 3; mice). b Shows automated segmentation of cell counts where c-Fos + cell densities were significantly higher than controls. The data represented as mean ± s.e.m. Reported p-values derived from two-tailed Mann–Whitney U test per brain area. - LC Locus coeruleus, PAG periaqueductal grey, PPN pedunculopontine tegmentum, MeA medial amygdala, BLA basolateral amygdala, PIR piriform cortex, NDB diagonal band nucleus, CLA claustrum. Scale bars are 500 µm. Source data are provided as a Source Data file

Fig. 4

Photoinhibition of GABAergic aNGC neurons generates cortical and behavioral arousal. a Schematic depicts area of injection of AAV-packaged Cre-inducible light-gated proton pump, ArchT in aNGC and location of electrodes in the cingulate cortex and the optic fiber in aNGC. b Drawing depicts optogenetic stimulation of an anesthetized mouse (isoflurane 1.25% vol; pink line). c Photoinhibition of a Gad65 + aNGC neuron (bottom panel) and concurrent desynchronized cortical LFP (raw data). d Average spectrogram of 48 laser pulses obtained from three animals. Color bar represents power in decibels. There is a consistent decrease in slow cortical oscillations and an increase in high frequencies in pulses with effective GAD65 + aNGC cell photoinhibition. e Average instantaneous firing rate of GAD65 + cells shows decreased firing during photoinhibition (mean± s.e.m.) ***p = < 0.001; one-way ANOVA. Source data are provided as a Source Data file

Fig. 5

Synergistic activity of aNGC neurons promotes emergence from anesthesia. a Schematic shows experimental configuration and location of electrodes in the cingulate cortex (Cg Ctx) and aNGC (nucleus gigantocellularis) to record LFPs and single units during emergence from anesthesia. b Representative trace of raw LFP recorded in the cingulate cortex (top) and normalized spectrogram (deviation from mean) showing the predominant frequencies observed during the transition between a coma-like state and wakefulness under isoflurane concentrations of 1% and 0.75% until the onset of movement (dashed lines). Color bar represents power in decibels. Bottom panel shows the average firing rate observed in a representative HA and LA neuron simultaneously recorded as the animal emerges from anesthesia. c Displays the average spectrogram during the transition from low to high states (n = 14; 100 s before and after transition) and the average firing rate (error in light colors) of HA and LA cells located in aNGC. d Average spectrogram of detected states, low state (n = 17), high cortical state I (n = 11), and high-state II-Mov (n = 3) in six animals. e Average firing rate (mean± s.e.m.) from HA (n = 6) and LA neurons (n = 5) in low cortical state and high state I and II. HA (n = 8) and LA (n = 3); nonparametric analysis; Kolmogorov–Smirnov **p < 0.01. Source data are provided as a Source Data file