Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 190 (11), 1301-10

Optogenetic Stimulation of Adrenergic C1 Neurons Causes Sleep State-Dependent Cardiorespiratory Stimulation and Arousal With Sighs in Rats

Affiliations

Optogenetic Stimulation of Adrenergic C1 Neurons Causes Sleep State-Dependent Cardiorespiratory Stimulation and Arousal With Sighs in Rats

Peter G R Burke et al. Am J Respir Crit Care Med.

Abstract

Rationale: The rostral ventrolateral medulla (RVLM) contains central respiratory chemoreceptors (retrotrapezoid nucleus, RTN) and the sympathoexcitatory, hypoxia-responsive C1 neurons. Simultaneous optogenetic stimulation of these neurons produces vigorous cardiorespiratory stimulation, sighing, and arousal from non-REM sleep.

Objectives: To identify the effects that result from selectively stimulating C1 cells.

Methods: A Cre-dependent vector expressing channelrhodopsin 2 (ChR2) fused with enhanced yellow fluorescent protein or mCherry was injected into the RVLM of tyrosine hydroxylase (TH)-Cre rats. The response of ChR2-transduced neurons to light was examined in anesthetized rats. ChR2-transduced C1 neurons were photoactivated in conscious rats while EEG, neck muscle EMG, blood pressure (BP), and breathing were recorded.

Measurements and main results: Most ChR2-expressing neurons (95%) contained C1 neuron markers and innervated the spinal cord. RTN neurons were not transduced. While the rats were under anesthesia, the C1 cells were faithfully activated by each light pulse up to 40 Hz. During quiet resting and non-REM sleep, C1 cell stimulation (20 s, 2-20 Hz) increased BP and respiratory frequency and produced sighs and arousal from non-REM sleep. Arousal was frequency-dependent (85% probability at 20 Hz). Stimulation during REM sleep increased BP, but had no effect on EEG or breathing. C1 cell-mediated breathing stimulation was occluded by hypoxia (12% FIO2), but was unchanged by 6% FiCO2.

Conclusions: C1 cell stimulation reproduces most effects of acute hypoxia, specifically cardiorespiratory stimulation, sighs, and arousal. C1 cell activation likely contributes to the sleep disruption and adverse autonomic consequences of sleep apnea. During hypoxia (awake) or REM sleep, C1 cell stimulation increases BP but no longer stimulates breathing.

Keywords: EEG; hypoxia; medulla oblongata; respiration; rostral ventrolateral medulla.

Figures

Figure 1.
Figure 1.
Characterization of neurons transduced by ChR2-EYFP adeno-associated virus type 2. (A) Transduced cells are catecholaminergic. (A1) Merged photomicrograph of tyrosine hydroxylase (TH) immunoreactive (ir) (revealed by Cy3, red) (A2) and ChR2-EYFP (i.e., virally transduced; revealed by Alexa Fluor 488, green) (A3). Note the complete overlap of TH and EYFP (arrows). (B) Virally transduced cells are spinally projecting C1 neurons. (B1) Photomicrograph of phenylethanolamine-N-methyl transferase (PNMT) ir (revealed by DyLight 649, blue) (B2) with ChR2-EYFP-transduced cells (revealed by Alexa Fluor 488, green) (B3) that are retrogradely labeled with cholera toxin B (CTB) from thoracic spinal cord injections (revealed by Cy3, red) (B4). Arrows point to triple-labeled neurons. Asterisks indicate non-PNMT spinally projecting neurons. Arrowheads indicate nontransduced C1 bulbospinal neurons. (C) Example of catecholaminergic (TH positive, red) neuron transduced with ChR2-EYFP (green). (D) Example of catecholaminergic (TH positive, red) bulbospinal (CTB, blue) neuron transduced with ChR2-EYFP (green).
Figure 2.
Figure 2.
Rostrocaudal distribution of ChR2-EYFP-transduced neurons. The number of transduced catecholaminergic neurons (EYFP+TH) and transduced neurons without detectable tyrosine hydroxylase (EYFP only) were counted per section in a one-in-six series of 30-μm coronal sections in eight rats. Expression of ChR2-EYFP in presympathetic C1 neurons (immunoreactive for both PNMT and CTB) was determined in three other rats with spinal injections of CTB. Error bars show SEM. FN shows the location of the facial motor nucleus.
Figure 3.
Figure 3.
Photostimulation of baroinhibited neurons in anesthetized rats. (A) Example of single RVLM neuron that generated one action potential per light pulse up to 40 Hz. (B) Example of a single baroinhibited RVLM neuron. This neuron could be silenced by a moderate rise in blood pressure (BP) (left excerpt). Photostimulation at 20 Hz could drive unit activity at 20 Hz even while receiving inhibitory inputs from the baroreceptors. PE = phenylephrine (5 μg/kg i.v.).
Figure 4.
Figure 4.
Photostimulation of C1 neurons produced network activation of RTN neurons and respiratory neurons in anesthetized rats. (A) Example of a single CO2-sensitive RTN neuron that was indirectly (i.e., synaptically) activated by photostimulation of C1 neurons at low and high levels of end-tidal CO2 (ETco2). End-expiratory CO2 was changed by adding variable concentrations of this gas to the breathing mixture. Photostimulation of C1 neurons (gray bars) occurred at 20-Hz blue light with 5-ms pulse width for 10–20 s. Top trace: Arterial blood pressure (BP). RTN neuronal activity is unaffected by ramp increases in blood pressure with phenylephrine (PE, 5 μg/kg i.v.). Middle traces: ETCO2 and extracellular action potentials. Lower trace: Integrated rate histogram (bin size = 1 s) shows the increases in RTN unit firing rate from baseline by C1 stimulation at various levels of ETco2. (B) Example of a respiratory neuron that increased firing rate and cycle frequency (i.e., respiratory frequency [fR]) with photostimulation of C1 neurons (20 Hz for 20–30 s). (C) Average discharge frequency of eight RTN neurons at baseline (4–5% ETco2) and during C1 cell photostimulation. (D) Mean discharge frequency in the active phase of 13 respiratory neurons at rest (4–5% ETco2) and during C1 cell photostimulation. (E) Average respiratory cycle frequency at rest and during C1 cell photostimulation (n = 9). Paired Student’s t test. *P < 0.05, **P < 0.01, ****P < 0.001.
Figure 5.
Figure 5.
State-dependent effects of C1 cell stimulation on breathing and BP. Left panel: Photostimulation of C1 neurons (20 Hz for 20 s; blue bar) during non-REM sleep (nREM) increased breathing (fR; respiratory frequency) and BP and produced arousal (EEG desynchronization and reduced δ power) with a sigh (black arrowheads). Middle panel: Photostimulation of C1 neurons during REM sleep elevated BP, but had no effect on respiratory frequency and did not produce sighs or arousal from REM sleep. Right panel: Photostimulation of C1 neurons in a quiet wake state (wake) increased BP and breathing and evoked a sigh. fR trace is capped at 150 breaths/min. HR = heart rate.
Figure 6.
Figure 6.
Group data for light pulse frequency- and state-dependent effects of C1 cell activation on blood pressure, breathing, and state of vigilance. (A) Average change in cardiorespiratory parameters elicited during non-REM (nREM) sleep by increasing the photostimulation frequencies of C1 neurons (2, 10, and 20 Hz for 20 s; one-way repeated measures [RM] analysis of variance [ANOVA], Dunn’s post hoc test). (B) Average cardiorespiratory parameters at rest (open columns) and during the 20-Hz, 20-s photostimulation trials (shaded columns) in nREM sleep, REM sleep, and quiet wake states (two-way RM ANOVA, Bonferroni’s post hoc test). (C) Probability of arousal (1-s bins, mean ± SE, n = 7 rats) during photostimulation of ChR2-transduced C1 neurons (20 s at 2, 10, or 20Hz) in nREM sleep. The control rats ([ChR2(−), n = 6]) received a 20-Hz light stimulus, but had no transduced neurons. (D) Cumulative probability of arousal from nREM sleep and sighs as a function of photostimulation frequency in ChR2(+) rats (one-way ANOVA, Dunn’s post hoc test). (E) Cumulative probability of arousal from nREM or REM sleep by 20-Hz photostimulation of C1 neurons. Probability of a sigh related to the arousal from sleep, or evoked in the quiet wake state, with 20-Hz photostimulation of C1 neurons (two-way RM ANOVA, Bonferroni’s post hoc test). *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001.
Figure 7.
Figure 7.
Spontaneous and C1-evoked arousal from non-REM sleep in rats resulted in a similar sequence of cortical desynchronization, tachycardia, and sighs. (A) Spontaneous arousal from non-REM sleep (gray arrow) is accompanied by tachycardia (asterisk) and sighs (black arrow). (B) Probability of tachycardia (100%) and sighs (51%) during spontaneous EEG desynchronization in non-REM sleep (recorded in n = 6 rats; >20 events/rat). (C) During spontaneous arousal, onset of tachycardia followed onset of EEG desynchronization by <1 s on average) and sighs by less than 6 s on average (n = 6 rats; >30 events/rat). (D) 20-Hz photostimulation of C1 cells in non-REM sleep produced the same stereotypic sequence of EEG desynchronization (gray arrow) with tachycardia (asterisk) and sighs (black arrow). Arousal tachycardia was blunted by C1-evoked pressor response that produced a reflex bradycardia throughout the 20-s stimulus period. (E) xy scatterplot of the latency to sighs relative to the onset of arousal produced by C1 cell stimulation. Sighs follow arousal. HR = heart rate.
Figure 8.
Figure 8.
Hypoxia, but not hypercapnia, occluded the increase in breathing by C1 cell stimulation. (A) Photostimulation of C1 neurons (20 Hz, 20 s) at rest (left panel; 21% O2 balanced in N2) or during hypercapnia (middle panel; 6% CO2, 21% O2 balanced in N2) increased breathing rate, elevated BP, and caused bradycardia. In hypoxia (right panel; 12% O2 balanced in N2), photostimulation of C1 neurons still elevated BP and caused bradycardia, but no longer stimulated breathing. Group data summarizing the effects of selective C1 cell stimulation on blood pressure (B), heart rate (C), respiratory frequency (D), and minutes of ventilation (E) under hypoxic or hypercapnic conditions (n = 7). xy plots show the average change from rest with C1 cell stimulation (y) plotted against the resting baseline value (x) in each condition.

Similar articles

See all similar articles

Cited by 31 PubMed Central articles

See all "Cited by" articles

Publication types

Feedback