The lateral paragigantocellular nucleus modulates parasympathetic cardiac neurons: a mechanism for rapid eye movement sleep-dependent changes in heart rate

Abstract

Rapid eye movement (REM) sleep is generally associated with a withdrawal of parasympathetic activity and heart rate increases; however, episodic vagally mediated heart rate decelerations also occur during REM sleep. This alternating pattern of autonomic activation provides a physiological basis for REM sleep-induced cardiac arrhythmias. Medullary neurons within the lateral paragigantocellular nucleus (LPGi) are thought to be active after REM sleep recovery and play a role in REM sleep control. In proximity to the LPGi are parasympathetic cardiac vagal neurons (CVNs) within the nucleus ambiguus (NA), which are critical for controlling heart rate. This study examined brain stem pathways that may mediate REM sleep-related reductions in parasympathetic cardiac activity. Electrical stimulation of the LPGi evoked inhibitory GABAergic postsynaptic currents in CVNs in an in vitro brain stem slice preparation in rats. Because brain stem cholinergic mechanisms are involved in REM sleep regulation, we also studied the role of nicotinic neurotransmission in modulation of GABAergic pathway from the LGPi to CVNs. Application of nicotine diminished the GABAergic responses evoked by electrical stimulation. This inhibitory effect of nicotine was prevented by the alpha7 nicotinic receptor antagonist alpha-bungarotoxin. Moreover, hypoxia/hypercapnia (H/H) diminished LPGi-evoked GABAergic current in CVNs, and this inhibitory effect was also prevented by alpha-bungarotoxin. In conclusion, stimulation of the LPGi evokes an inhibitory pathway to CVNs, which may constitute a mechanism for the reduced parasympathetic cardiac activity and increase in heart rate during REM sleep. Inhibition of this pathway by nicotinic receptor activation and H/H may play a role in REM sleep-related and apnea-associated bradyarrhythmias.

Fig. 1.

Schematic drawing of the experimental preparation. Cardiac vagal neurons (CVNs) were identified by retrograde fluorescent labeling. Slices of 400-μm thickness were obtained and individual identified. CVNs in the nucleus ambiguus (NA) were studied using whole cell patch-clamp technique. The location of the lateral paragigantocellular nucleus (LPGi) was identified using stereotaxic coordinates in addition to the location relative to fluorescently identified CVNs in the NA. Square wave current injections of 0.1–0.5 mA and 1-ms duration were applied to evoke GABAergic pathways from the LPGi to CVNs.

Fig. 2.

A: stimulation of the LPGi evoked a postsynaptic current in CVNs. Application of the N-methyl-d-aspartate (NMDA) and non-NMDA receptor blockers AP-5 and CNQX, respectively, did not alter the evoked inhibitory postsynaptic current (IPSC) as shown in a typical example (A) and in the summary data (B). However, application of strychnine, a glycinergic receptor antagonist, significantly decreased the evoked postsynaptic current (A, typical experiment; B, summary data). Finally, gabazine, a GABAergic receptor antagonist, nearly abolished the activation of this synaptic pathway (A, typical experiment; B, summary data). As shown in C, paired pulse stimulations with a delay of 10 ms evoked postsynaptic GABAergic responses (isolated by focal application of strychnine, AP-5, and CNQX) in CVNs with a constant latency. +P < 0.05. In this and all subsequent figures, arrow indicates electrical stimulation (unless other stated), ■ with asterisks indicates statistically significant differences, *P < 0.05, **P < 0.01, and ***P < 0.001.

Fig. 3.

Photoexcitation of LPGi neurons with uncaged glutamate elicited a significant increase in the frequency of GABAergic IPSCs in the CVNs. A typical experiment is shown in A, whereas the summary data are shown in B. Arrow indicates 5-ms stimulation of the LPGi with uncaged glutamate.

Fig. 4.

As shown in A (left), in a typical experiment and in the histogram from 11 cells, the frequency of IPSCs followed by the initial short-latency GABAergic current was not altered by the LPGi stimulation (42% of CVNs tested). However, in other CVNs tested, there was either decrease (42%, A, right, typical experiment and histogram from 11 cells) or increase (16%, A, bottom, typical experiment and histogram from 4 cells) in the frequency of IPSCs followed the initial GABAergic response. Burst stimulation evoked GABAergic current (B, right) significantly larger than GABAergic currents evoked by single stimulation (B, left, and B, bottom, the summary data from 10 cells). This initial GABAergic current evoked by burst electrical stimulation was followed by an increase in the IPSC frequency with long latencies, as shown in the histogram obtained from 8 cells (B, right).

Fig. 5.

As shown in a single experiment (A, top) and in the summary data (A, bottom), application of nicotine at concentrations of either 1 or 10 μM did not significantly alter the peak amplitude of the GABAergic current evoked by LGPi electrical stimulation. However, nicotine, at a concentration of 100 μM, significantly inhibited the evoked GABAergic current (A, top, single experiment; A, bottom, summary data). As shown in a single experiment (B, top) and in the summary data (B, bottom), application of the non α7 receptor antagonist DHβE did not alter the evoked GABAergic response and DHβE also did not alter the inhibitory effect of nicotine on the evoked GABAergic current. However, the inhibitory effect of nicotine was antagonized with α-Bgtx, an α7 receptor blocker, whereas α-Bgtx alone did not evoke any effect on the inhibitory GABAergic current evoked by LGPi stimulation, as shown (C, top, single experiment; C, bottom, summary data).

Fig. 6.

As shown in a typical experiment (A, top) and in the summary data (A, bottom), hypoxia/hypercapnia (H/H) elicited a significant and reversible decrease in the peak amplitude of GABAergic current evoked by electrical stimulation of the LPGi. This inhibition of evoked GABAergic response was prevented by application of the α7 receptor antagonist α-Bgtx in periods of both 1–2 and 4–5 min, whereas in a period of 9–10 min, α-Bgtx did not alter the H/H-elicited inhibition in evoked GABAergic current, as shown in a typical experiment (B, top) and in the summary data (B, bottom).

Fig. 7.

Hypothetical framework for rapid eye movement (REM) sleep-dependent changes in heart rate. REM sleep-related activation of the LPGi neurons evokes GABAergic pathway to CVNs in the NA, which results in withdrawal of parasympathetic activity to the heart and increases in heart rate (right). H/H that typically occurs during apnea-associated events during REM sleep induces an α7 nicotinic receptor–mediated attenuation of the inhibitory pathway to parasympathetic CVNs that results in CVN disinhibition and activation of vagal tone, which may cause life-threatening bradyarrhythmias (left).