FEATURED ARTICLE

A Shared Neural Substrate by Diverse Anesthetics and Sleep

Luping Yin, PhD
Li-Feng Jiang-Xie, PhD
Fan Wang, PhD
Department of Neurobiology
Duke University Medical Center
Durham, North Carolina

luping Yin, PhD
Luping Yin, PhD
Li-Feng Jiang-Xie, PhD
Li-Feng Jiang-Xie, PhD
Fan Wang, PhD
Fan Wang, PhD

Background
General anesthetics have been widely used in surgery since 1846, when William T.G. Morton (1819 –1868) performed the first successfully modern general anesthesia (GA) in Massachusetts General Hospital.1 The discovery of general anesthesia has revolutionized surgical procedures in medicine over the past decades.1,2 Each year, millions of patients worldwide receive GA for medical treatments. GA is a drug-induced reversible and global brain/body state characterized by unconsciousness, amnesia, analgesia, and immobility along with stability of basic physiological functions.2,3 Despite centuries of scientific endeavor, the neural circuit mechanisms by which structurally diverse groups of GA drugs induce unconsciousness remain unclear.3-5

The process of GA can be divided into three periods based on clinical signs and electroencephalogram (EEG) patterns: induction, maintenance, and emergence. EEG patterns during the maintenance period are characterized by the progressive decrease in the power of higher frequency oscillations (13-30 Hz) and the strong appearance of low-frequency, high-amplitude delta oscillation (0.5-4 Hz). Interestingly, the increased delta oscillation resembles that which occurs during natural slow-wave sleep.2,6,7 Delineating the identities of neurons and the precise neural circuitry shared between GA and sleep should enrich our basic understandings and clinical applications of GA.

Molecular Targets of General Anesthetics
Many different receptors and molecular targets have been discovered for different GA drugs. GA state is thought to be the consequence of either enhancement of inhibitory or suppression of excitatory neurotransmission. Among the known molecular targets of GA drugs, ion channels are the predominant candidates. Many drugs, such as gaseous anesthetic isoflurane and intravenous anesthetics barbiturates/propofol, have been shown to potentiate GABA induced Cl- current; in higher doses, they can directly activate GABAA receptors even without anesthetics.2 Other non-GABA anesthetics, like ketamine, have been shown to reduce excitatory or glutamate mediated synaptic transmission by blocking NMDA receptors on the postsynaptic membrane, and dexmedetomidine, which binds to alpha-2 (a2) adrenergic receptors and inhibits the norepinephrine release from locus coeruleus.2 Sodium/potassium channels and other ion channels such as hyperpolarization activated cyclic nucleotide (HCN) channels are also involved in this process. Collectively, these results lead to the common belief that distinct GA drugs work by exerting distinct inhibitory impacts on the nervous system. In recent years, a few studies have suggested the existence of anesthetic-activated cells in several brain regions and challenged this rationale using either immediate early gene markers c-fos, or ex vivo brain slice recording,8-11 but direct in vivo extracellular recording evidence had been lacking.

Technologies Used in Our Study
Using the immediately early gene Fos, we have discovered a population of neurons located in the region immediately above the optic tract: the supraoptic nucleus and its nearby regions. We named this distinct population of neurons ‘anesthesia-activated neurons (AANs)’. Our novel hypothesis is that AANs play essential roles in the GA process and may also be involved in regulating sleep. To label and manipulate activated Fos+ neurons so as to test this hypothesis, we used a state-of-the-art technology called ‘CANE’ for ‘capturing activated neuronal ensembles’ recently developed in our lab 12 (Figure 1). CANE technology consists of two parts: (1) a knock-in mouse in which a destablized foreign receptor (dsTVA) is expressed in a similar spatial and temporal pattern as that of the immediate early gene Fos (FosTVA mouse), and (2) designer pseudotyped viruses (CANE-lentivirus or CANE-rabies virus) that can carry desired transgenes and only infect neurons expressing the dsTVA receptor, hence Fos+ neurons. We show that CANE can be used to specifically and efficiently capture Fos+ neurons and express any desired genes in activated neurons. Therefore, CANE provides us an unprecedented tool to achieve this goal with accurate temporal resolution.

Figure 1

Figure1: Schematic diagram of CANE technology.

Main Findings of Our Study
In our recently published work,13 we showed that the majority of AANs sat within a tiny nucleus of the brain, called the supraoptic nucleus (SON) (Figure 2A), which is known as the hub for neuroendocrine cells that release hormones to regulate carious bodily homeostasis and consistently these AANs express multiple neuropeptides and hormones, including vasopressin, dynorphin, and galanin. We used in-vivo electrophysiology recording to reveal that subsets of neurons within and near SON significantly increased their firing in vivo in response to isoflurane anesthesia. After using CANE to capture and label AANs, we further demonstrated that AANs can be activated by multiple distinct general anesthetics widely used in clinical trials, including isoflurane, propofol, ketamine/xylazine, and dexmedetomidine, using Fos staining (Figure 2B) and slice electrophysiology.

Figure 2

Figure 2: A Shared Neuronal Population is Activated by Different Anesthetics.
(A) Upper, viral construct and injection site in FosTVA mice. Lower, illustration of the SON and paraSON. (B) Representative images of CANE-captured isoflurane-activated neurons (red) and Fos+ neurons (green) induced by re-exposure to either isoflurane again, or to propofol, ketamine (plus xylazine) or Dex.

We wanted to selectively manipulate these AANs and observed the consequence on animal behaviors. To achieve this goal, we employed the CANE technology to specifically express either chemogenetic (hM3Dq-DREADD) or optogenetic (Channelrhodopsin, ChR2) actuators into these AANs, or control fluorescent proteins. Remarkably, once we activated AANs, using either the artificial ligand CNO that works on the hM3Dq-DREADD receptor, or with a pulse of blue light to stimulate ChR2-expresed in AANs, mice stopped moving around. Concomitant EEG recordings showed that their brain state changed from wakefulness into a deep sleep state characterized by strong slow-wave oscillation. Even a brief AAN activation is sufficient to result in subsequent prolonged sleep, consistent with the fact that neuropeptides released from AANs likely have longer-lasting signaling after light pulse was turned off. On the other hand, if we conditionally ablated these AANs, mice reduced total sleep time, have fragmented sleep, and even if they sleep, their brain showed diminished slow oscillation. Some AAN-ablated mice died from sleep deprivation.

Furthermore, we also investigated the functional role of AANs during GA. By selectively expressing either ChR2 to activate these cells, or by expressing a light-gated proton pump, eArch3.0 to turn off the cells with light, we bidirectionally manipulated AAN activity during anesthesia. Photo-activation of AAN did not alter the GA induction but significantly delayed the emergence from GA. In other words, animals would stay longer under GA once we activated AAN. Conversely, inhibition of AAN did not alter GA induction, but significantly shortened the duration of GA. These results further underscore the importance of the slow but long-lasting peptidergic (hormonal) signaling from AANs in maintaining the GA state. In summary, AANs have a crucial role in the maintenance but not the induction of GA-induced unconsciousness state.

Contribution to the Fields of GA and Sleep
AANs discovered here are activated by multiple different GA drugs and may represent a shared substrate between GA and sleep. In freely behaving mice, optogenetic14 and chemogenetic15 activation of AANs were sufficient to strongly potentiate slow-wave sleep (SWS) and GA. Importantly, conditional ablation of AANs resulted in significant decline of slow-wave power and loss of both SWS and rapid eye movement (REM) sleep, while acute inhibition of AANs shortened the duration of GA. Together, our results revealed a previously unrecognized function of SON neuroendocrine cells (primarily vasopressin/dynorphin/galanin producing cells in SON), which are known for their role of releasing hormones, in regulating both GA and natural sleep.

Significance
AANs represent a common neural substrate for GA and sleep, and function to maintain or prolong the unconscious or sleep state. The most surprising discovery is that most of AANs are peptide-releasing neuroendocrine cells in the SON. It turns out that these cells not only release peptide hormones into the circulation through their projections to the posterior pituitary, they also release large amounts of peptides through massive dendritic release into the hypothalamus and to the cerebrospinal fluids. We speculate that neuropeptide signaling can last longer and, through the cerebral spinal fluid, spread across the whole brain, and therefore may be ideally suited to maintaining the global brain state in an unconscious state.

Acknowledgements
This work is supported by an NIH DP1MH103908 to F.W., a Brain Research Foundation SIA to F.W., the W. M. Keck Foundation grant to F.W. and K.D, and Human Frontier Science Program (LT000038/2018-L) to L.Y.

References

  1. Robinson, D. H. & Toledo, A. H. Historical development of modern anesthesia. J Invest Surg 25, 141-149, doi:10.3109/08941939.2012.690328 (2012).
  2. Franks, N. P. General anaesthesia: from molecular targets to neuronal pathways of sleep and arousal. Nat Rev Neurosci 9, 370-386, doi:10.1038/nrn2372 (2008).
  3. Brown, E. N., Lydic, R. & Schiff, N. D. General anesthesia, sleep, and coma. N Engl J Med 363, 2638-2650, doi:10.1056/NEJMra0808281 (2010).
  4. Alkire, M. T., Hudetz, A. G. & Tononi, G. Consciousness and anesthesia. Science 322, 876-880, doi:10.1126/science.1149213 (2008).
  5. Koch, C., Massimini, M., Boly, M. & Tononi, G. Neural correlates of consciousness: progress and problems. Nat Rev Neurosci 17, 307-321, doi:10.1038/nrn.2016.22 (2016).
  6. Akeju, O. & Brown, E. N. Neural oscillations demonstrate that general anesthesia and sedative states are neurophysiologically distinct from sleep. Curr Opin Neurobiol 44, 178-185, doi:10.1016/j.conb.2017.04.011 (2017).
  7. Rudolph, U. & Antkowiak, B. Molecular and neuronal substrates for general anaesthetics. Nat Rev Neurosci 5, 709-720, doi:10.1038/nrn1496 (2004).
  8. Gelegen, C. et al. Excitatory Pathways from the Lateral Habenula Enable Propofol-Induced Sedation. Curr Biol 28, 580-587 e585, doi:10.1016/j.cub.2017.12.050 (2018).
  9. Lu, J. et al. Role of endogenous sleep-wake and analgesic systems in anesthesia. J Comp Neurol 508, 648-662, doi:10.1002/cne.21685 (2008).
  10. Moore, J. T. et al. Direct activation of sleep-promoting VLPO neurons by volatile anesthetics contributes to anesthetic hypnosis. Curr Biol 22, 2008-2016, doi:10.1016/j.cub.2012.08.042 (2012).
  11. Zhang, Z. et al. Neuronal ensembles sufficient for recovery sleep and the sedative actions of alpha2 adrenergic agonists. Nat Neurosci 18, 553-561, doi:10.1038/nn.3957 (2015).
  12. Sakurai, K. et al. Capturing and Manipulating Activated Neuronal Ensembles with CANE Delineates a Hypothalamic Social-Fear Circuit. Neuron 92, 739-753, doi:10.1016/j.neuron.2016.10.015 (2016).
  13. Jiang-Xie, L.-F. et al. A Common Neuroendocrine Substrate for Diverse General Anesthetics and Sleep. Neuron 102, 1053-1065.e1054, doi:https://doi.org/10.1016/j.neuron.2019.03.033 (2019).
  14. Boyden, E. S., Zhang, F., Bamberg, E., Nagel, G. & Deisseroth, K. Millisecond-timescale, genetically targeted optical control of neural activity. Nat Neurosci 8, 1263-1268, doi:10.1038/nn1525 (2005).
  15. Roth, B. L. DREADDs for Neuroscientists. Neuron 89, 683-694, doi:10.1016/j.neuron.2016.01.040 (2016).

Back to top