Targeting Parasympathetic Activity to Improve Autonomic Tone and Clinical Outcomes

Abstract

In this review we will briefly summarize the evidence that autonomic imbalance, more specifically reduced parasympathetic activity to the heart, generates and/or maintains many cardiorespiratory diseases and will discuss mechanisms and sites, from myocytes to the brain, that are potential translational targets for restoring parasympathetic activity and improving cardiorespiratory health.

In this review we will briefly summarize the evidence that autonomic imbalance, more specifically reduced parasympathetic activity to the heart, generates and/or maintains many cardiorespiratory diseases and will discuss mechanisms and sites, from myocytes to the brain, that are potential translational targets for restoring parasympathetic activity and improving cardiorespiratory health. More specifically, we describe and compare the advantages and disadvantages of stimulation of carotid sinus baroreceptors, identified parasympathetic cholinergic neurons or fibers within the vagus nerve or cardiac ganglia, and the activation of brainstem cardiac vagal neurons via activation of an upstream hypothalamic oxytocin neuronal network as novel approaches to increase parasympathetic activity to the heart. We also discuss how the clinical implementation for stimulating these targets may eventually involve innovative technologies such as electronic microarrays, chemogenetics, and/or optogenetics. New basic science and clinical translation opportunities for targeting the powerful cardioprotective effects of the cardiac parasympathetic system are rapidly leading to exciting new avenues for this important field.

Diminished Parasympathetic Activity to the Heart in Cardiorespiratory Diseases

It is estimated that heart failure (HF) affects nearly 23 million people worldwide. A distinctive hallmark of cardiac hypertrophy, heart failure, and associated abnormalities in the cardiac conduction system is autonomic imbalance: more specifically increased sympathetic activity and decreased parasympathetic tone (1). Despite available treatments HF has a high mortality rate, with a 5-yr survival of only 25% in men and 38% in women (2, 3). As heart failure progresses the increased sympathetic and depressed parasympathetic activity may be beneficial to maintain cardiac output and support myocardial performance initially (4). As cardiac dysfunction increases, however, the altered autonomic balance contributes to deterioration of ventricular performance, a higher risk of electrical instability, structural remodeling with increased apoptosis of myocytes, ventricular dilatation, and risk of arrhythmias (4, 5).

An altered balance of autonomic activity occurs in the early stages of heart failure and often precedes other stereotypical changes. Parasympathetic tone is decreased 3 days after the development of cardiac dysfunction, and this depression of vagal activity to the heart typically precedes the enhancement of sympathetic activity (6, 7). Reduced cardiac vagal activity presents early in patients with heart failure and occurs with even mild left ventricular impairments, and the adverse consequences of diminished parasympathetic activity happen independently from increased sympathetic activity (8–10).

Another highly prevalent cardiovascular disease, sudden cardiac death (SCD) is responsible for 15 to 20% of all deaths and ∼50% of all cardiovascular deaths in the United States (11). The most common cascade of events leading to SCD in the Western world is acute coronary syndrome (ACS) that progresses to acute myocardial ischemia and/or inflammation that triggers electrical instability and lethal arrhythmias, including ventricular tachycardia, asystole and fibrillation (11–13). The absolute risk of a fatal arrhythmic event is greatest within the first few hours and days immediately after a myocardial infarction (MI) and declines significantly thereafter, reaching a steady state after ∼1 yr (14, 15). Myocardial ischemia and infarction are accompanied by a marked autonomic imbalance with sympathetic over activity and reduced parasympathetic cardiac vagal activity (16–18). Cardiac acetylcholine levels remained preserved in infarcted hearts, and the firing and function of postganglionic parasympathetic neurons remain viable post MI, indicating the parasympathetic system is intact but the generation of parasympathetic activity to the heart is diminished post-MI (19). This reduced vagal drive to the heart is a strong independent risk factor for life-threatening arrhythmias and SCD (20). In contrast, studies suggest if parasympathetic cardiac activity can be maintained post-MI then the generation of fatal ventricular arrhythmias and risk of SCD is significantly reduced (16–21).

Obstructive sleep apnea (OSA) occurs in ∼24% of men and 9% of women in the United States (22, 23), and it is estimated that OSA increases cardiovascular mortality threefold (24). Unfortunately, however, there are no Food and Drug Administration (FDA)-approved pharmacological treatments for OSA, leaving mainly continuous positive airway pressure (CPAP) or similar devices as treatment options. Unfortunately, CPAP is too often discontinued because it is intrusive and poorly tolerated. Approximately 50% of all patients with OSA discontinue CPAP use entirely or use it for <4 h each night (25). Autonomic dysfunction, particularly the withdrawal of parasympathetic tone, is a dominating cause of the adverse cardiorespiratory events that occur with OSA (26–28). In both patients with OSA, and animal models of OSA using chronic intermittent hypoxia (CIH), there is decreased baroreflex sensitivity, elevated blood pressure, elevated sympathetic activity, and diminished parasympathetic activity to the heart (29–35).

While it is well established that parasympathetic activity to the heart is diminished in many cardiorespiratory diseases, the sites and mechanisms for this dysfunction are not fully understood. As examples, while CIH reduces reflex control of heart rate and parasympathetic tone to the heart (35, 36), these changes are not due to changes in parasympathetic innervation of the sinoatrial node or function within the cardiac ganglia (33, 37, 38). Heart rate responses to vagal efferent stimulation are not diminished but rather enhanced following CIH (33, 37, 38). However, in mice exposed to CIH, preganglionic cardiac vagal axons and terminals were found to be swollen with fewer close contacts to postganglionic cardiac neurons (39). Following a MI, cardiac acetylcholine levels remained preserved, and the firing and function of postganglionic parasympathetic neurons remain viable, indicating the generation and transmission of parasympathetic activity from the central nervous system (CNS) is diminished (19). These results indicate a central brainstem dysregulation of premotor cardiac vagal neuronal activity, with or without additional changes in cardiac ganglia transmission, function, or cardiac innervation, is likely most responsible for the impaired parasympathetic control of heart rate that occurs in cardiovascular diseases.

Potential Sites and Mechanisms to Restore Parasympathetic Activity to the Heart

Interventions, discussed more fully below, have targeted each of the sites within the parasympathetic system from cardiac muscle to the brain. These sites and therapeutic targets include the efficacy of acetylcholine within the myocardium, enhanced activity within the cardiac ganglia, activation of efferent vagal nerve fibers, and finally activation of preganglionic cardiac vagal neurons and other sites within the parasympathetic circuit in the brainstem and hypothalamus.

Use of Acetylcholinesterase Inhibitors to Improve Autonomic Imbalance

Postganglionic parasympathetic nerve fibers release acetylcholine (ACh), which activate M2 muscarinic receptors in the myocardium and sinoatrial and atrioventricular nodes. This muscarinic receptor activation reduces heart rate and contractility and maintains tonic dilation of coronary arteries independent of left ventricular preload, afterload, and heart rate (40). Acetylcholinesterase inhibitors have been studied as possible treatments to increase parasympathetic activation in the heart as these agents slow the breakdown of ACh at the cholinergic nerve endings, prolonging its availability and activation of M2 receptors. Pyridostigmine and donepezil are among the most clinically used acetylcholinesterase inhibitors.

Acetylcholinesterase inhibitors have shown benefit in animal models of hypertension, myocardial ischemia and heart failure (see review Ref. 41). In addition, a few clinical studies have examined their potential benefit in patients with cardiovascular diseases. In a double-blinded, randomized, placebo-controlled crossover study in 15 patients with exercise induced myocardial ischemia, pyridostigmine increased peak oxygen consumption, improved peak exercise tolerance, and increased the exercise intensity at which myocardial ischemia occurred (42). In 21 patients with HF, pyridostigmine significantly reduced heart rate and improved exercise capacity and reduced inflammatory markers (43). However, elimination of pyridostigmine occurs relatively quickly (half-life of 20 to 30 min) and the significant adverse effects of acetylcholinesterase inhibitors, likely because of their nonselective enhanced activation of both nicotinic and muscarinic receptors in autonomic ganglia, as well as the tissue innervated by autonomic nerves, have generally precluded their widespread study and potential use (44). Adverse effects of acetylcholinesterase inhibitors are often severe and can include increased peristalsis, stomach pain, nausea, vomiting, diarrhea, muscle cramps, sweating, increased salivation, and blurred vision.

Parasympathetic Cardiac Ganglia Stimulation to Mitigate Adverse Cardiovascular Events

Postganglionic cholinergic cardiac ganglia neurons, located within the fat pads at the base of the heart, receive and integrate input from preganglionic cardiac vagal neurons that originate in the brainstem with descending fibers situated in the vagus nerve. Activation of cardiac ganglia neurons decrease heart rate and contractility as well as the rate of spread of cardiac depolarization within the heart (45). In rodent animal models, cholinergic cardiac ganglia neurons that express choline acetyltransferase (ChAT) can be selectively targeted in vitro and in vivo using a Cre-Lox recombination recombinase approach (46–48). In this paradigm, floxed viruses that express light sensitive ion channels, such as channelrhodopsin (ChR2), or chemogenetic G protein-coupled receptors, such as designer receptors exclusively activated by designer drugs (DREADDs), can be focally injected into the cardiac ganglia to induce selective expression of those proteins within cholinergic parasympathetic ganglia neurons of transgenic ChAT-Cre mice or rats (46). With the use of this approach in rats, acute chemogenetic stimulation of cardiac ganglia ChAT neurons elicited robust reductions in blood pressure (∼20 mmHg) and heart rate (∼100 beats/min) (46). Chronic chemogenetic activation of cardiac cholinergic neurons during the development of heart failure (HF) in rats improved electrophysiological adaptations to increases in pacing rate (49). Furthermore, in that same animal model of HF, chronic chemogenetic stimulation of only the cardiac ganglia ChAT neurons reduced left ventricular systolic dysfunction (reductions in ejection fraction, fractional shortening, stroke volume, and cardiac output), reduced cardiac autonomic dysfunction (reduced HR recovery postpeak effort), and reduced mortality by 30% (46).

However, while it is possible to selectively stimulate the ChAT neurons of the cardiac ganglia in transgenic animal models, as discussed above, such cell-specific approaches are not currently available clinically. There are complex clusters of ganglia in the human heart, many with diverse populations that include sensory and motor neurons and local neuronal networks comprised of cholinergic and noradrenergic neurons (50–52). There are also competing right atrial pacemakers that may preferentially control fast and slow heart rates (53). This complexity makes selective electrical stimulation of parasympathetic cardiac ganglia neurons seemingly infeasible using traditional bipolar electrode approaches. However, advances in the field of three-dimensional microelectrode arrays could provide promising new avenues for selective microelectrical stimulation of ChAT neurons in cardiac ganglia (54). Likewise, clinical translation of cardiac optogenetics, including new optical therapies for rhythm control, are promising new avenues for improving autonomic balance within the heart (55). Advantages of both microelectrode array and optogenetic stimulation include high spatial and temporal control, and optogenetics offers the additional benefit of remote stimulation independent of contact with the targeted tissue.

Vagal Nerve Stimulation Treatment

In 1991 a landmark study conducted by Vanoli and colleagues (56) showed that in unanesthetized dogs vagal nerve stimulation (VNS) prevented ventricular fibrillation induced by a myocardial infarction. Those studies, and many others, suggested that implantable vagal nerve stimulators could improve left ventricular function, reduce infarct size and decrease mortality after an infarction, and that VNS could be a promising novel treatment for heart failure patients (4).

Unfortunately, however, a recent large clinical study [Neural Cardiac Therapy for Heart Failure (NECTAR-HF)] provided mostly negative results for chronic vagal nerve stimulation in HF patients. This work found there was no improvement with VNS in left ventricular systolic dimensions or other echocardiographic parameters or biomarkers, although patients in this study did show an improvement in quality of life (10). Additional clinical trials with vagal nerve stimulation demonstrated improved heart rate variability and six-minute walk distance in HF patients but also indicated adverse effects that include dysphonia, cough, and throat pain, likely due to the nonspecificity of electrical vagal nerve stimulation (9, 57). There are many potential reasons for the lack of efficacy and adverse effects seen in this and other clinical studies of vagal nerve stimulation. These include ineffective and/or nonspecific stimulation parameters, as well as the inherent disadvantage of activating not only noncardiac parasympathetic efferent fibers but also the sensory afferent fibers in the vagus nerve, and their secondary responses (57). This lack of specificity is significant and not surprising as ∼70% of the fibers in the vagus nerve are sensory, not parasympathetic efferent nerve fibers (58).

Carotid Baroreflex Activation Therapy

The paradigm of baroreflex activation therapy (BAT) includes restoring autonomic balance by electrical stimulation of sensory carotid sinus baroreceptors, thereby evoking the baroreceptor reflex with resultant increases in parasympathetic and decreases in sympathetic activity. Stimulation of carotid sinus baroreceptors is accomplished by surgical implantation of electrodes to the outside of the carotid artery close to the carotid sinus. In preclinical studies in dogs, BAT improved left ventricular (LV) ejection fraction and decreased LV end-systolic volume and reduced interstitial fibrosis and cardiomyocyte hypertrophy (59). However, in another preclinical study in dogs, BAT improved survival and decreased plasma norepinephrine and angiotensin II levels but did not significantly improve arterial pressure, resting heart rate, or left ventricular function (60). A recent clinical study examined the efficacy and safety of BAT in the Baroreflex Activation Therapy for Heart Failure (BeAT-HF) trial that compared patients that received only guideline-directed medical therapy (GDMT) alone (control group) or BAT plus GDMT. The BeAT-HF trial did not assess left ventricular structure or function but BAT improved 6-min walk distance, Quality of Life, and reduction in NH₂-terminal pro b-type natriuretic peptide (NT-proBNP) levels in men and women (61). The BeAT-HF trial thus far has not examined endpoints in morbidity and mortality or HF hospitalization. Similar to VNS there is uncertainty concerning the selectivity of carotid sinus baroreceptor stimulation with BAT as bipolar stimulating electrodes implanted close to the carotid sinus would also likely activate carotid body chemoreceptors critical for responses to hypoxia and respiratory function.

Targeting Parasympathetic Activity to the Heart Within the CNS

Preganglionic cardiac vagal neurons (CVNs) located in the brainstem (nucleus ambiguus and dorsal motor nucleus of the vagus) are intrinsically silent and hence synaptic activation of these neurons is crucial in generating and modulating parasympathetic activity to the heart (62). CVNs receive three major synaptic pathways: glutamatergic, GABAergic, and glycinergic (63). CVNs are inhibited by GABA and glycine neurotransmission arising from inspiratory neurons in the brainstem and locus coeruleus. Respiratory sinus arrhythmia, in which heart rate increases during inspiration and postinspiration, is mediated by increases in inhibitory GABAergic and glycinergic transmission to CVNs during inspiration (64). CVNs are inhibited by photoactivation of locus coeruleus neurons that express ChR2, indicating that this pathway likely serves as the substrate for the increased heart rate that occurs with alertness (65). One major source of excitatory transmission to CVNs originates from hypothalamic paraventricular nucleus (PVN) oxytocin (OXT) neurons (66, 67). This pathway, and the receptors involved, have recently been targeted to increase cardiac parasympathetic activity in animal models of cardiorespiratory diseases, as well as some clinical trials (68–70).

In an animal model of HF, the optogenetic synaptic release of OXT from PVN synapses onto CVNs is reduced, but can be restored with chronic chemogenetic activation of PVN OXT neurons (71–73). Restoration of PVN OXT neuron activity reduces mortality, cardiac inflammation and fibrosis, and improves critical longitudinal indices of autonomic balance and cardiac function in increased afterload models of HF (71–73). It also provides beneficial outcomes for ventricular electrophysiology during increased afterload HF, including lower steady-state action potential duration as well as reduced variance of conduction loss and improved action potential duration versus diastolic interval dispersion during high heart rates (49). Furthermore, in an animal model of obstructive sleep apnea (OSA), where animals were exposed to 3 wk of CIH for 8 h a day, chronic chemogenetic activation of PVN OXT neurons prevented the hypertension that occurred in untreated animals (74). Consistent with this work PVN OXT mRNA expression is lower in spontaneously hypertensive rats (75), and there is reduced OXT receptor mRNA density in the nucleus tractus solitarius in hypertensive rats that can be significantly increased with exercise training (76).

In an initial study of patients with OSA, intranasal (IN) oxytocin decreased the incidence of arousals that accompanied hypopnea events, diminished the frequency of hypopnea events, and shortened the duration of both apneas and hypopneas (70). In addition, IN oxytocin increased parasympathetic activity in OSA patients, as shown by the increase in the high-frequency component of heart rate variability and short-term variability in Poincaré plots (70). In a larger and randomized, double-blinded, and placebo-controlled study, IN oxytocin significantly decreased the duration of obstructive events, as well as the oxygen desaturations and incidence of bradycardia that were associated with these events (69). Although the site of action of IN oxytocin in these clinical studies is unknown, it is worth noting that intranasal (IN) application of oxytocin is more effective in increasing the levels of oxytocin within the cerebrospinal fluid than the plasma (77).

In addition to the pathway from OXT neurons in the PVN to brainstem parasympathetic CVNs, other pathways in the CNS are possibly involved in the beneficial effects of IN oxytocin in OSA patients. For example, oxytocin microinjected into the brainstem pre-Bötzinger complex, a nucleus critical for generating inspiratory rhythm, increased respiratory frequency and diaphragm EMG activity (78). Intranasal oxytocin may also increase respiratory rate via activation of neurons in the rostral ventrolateral medulla (RVLM) since microinjection of oxytocin into the RVLM increased both respiratory rate and muscle diaphragm activity (79). Oxytocin could also excite hypoglossal (XII) motorneurons thereby increasing protruder and/or retractor tongue muscle activity, upper airway tone, and patency. Chemogenetic excitation of XII neurons increases genioglossus muscle activity and improves upper airway patency (80, 81). Inhibiting hypoglossal motorneurons leads to sleep-disordered breathing (80, 81). In agreement with these findings, topical anesthesia applied to the upper airway in OSA subjects increased the duration of apneas (82). An inward depolarizing current is elicited by administration of oxytocin in some, but not all, hypoglossal motor neurons in rodents (83). Chemical stimulation of the PVN augments genioglossus nerve activity, which occurs, at least in part, via activation of oxytocin receptors (84).

There are only limited well-established clinical protocols for the use of inhaled oxytocin as it is not an FDA-approved drug for a clinical application. However, the use and benefits of utilizing oxytocin in subjects with autism spectrum disorder (ASD) and posttraumatic stress disorder (PTSD) are ongoing and exciting areas of research. None of the most common adverse events, such as nasal discomfort, tiredness, or irritability, was statistically associated with oxytocin treatment in a recent meta-analysis (85) of intranasal oxytocin in 223 ASD participants. In a study comprised of subjects with PTSD, there were no reported adverse effects with oxytocin self-administered at a dose of 40 IU, once a week for 10 wk (86). Similarly, no study-related adverse events occurred with intranasal oxytocin administration (40 IU, twice daily for 2 wk) in individuals treated with methadone for opioid use disorder (87). Clinical trials are currently underway to test if chronic IN oxytocin provides long-term benefits in OSA patients (ClinicalTrials.gov Identifier: NCT03860233).

In summary, it is clear that autonomic imbalance, more specifically reduced parasympathetic activity to the heart, is critically involved in both the initiation and maintenance of many cardiorespiratory diseases. There are many exciting new targets of opportunity to selectively stimulate and restore parasympathetic activity to the heart, with their pros and cons summarized in Table 1. These include stimulation of carotid sinus baroreceptors, identified parasympathetic cholinergic neurons or fibers within the vagus nerve or cardiac ganglia, and the activation of brainstem cardiac vagal neurons via activation of an upstream hypothalamic oxytocin neuronal network. The clinical implementation for stimulating these sites may eventually involve innovative technologies such as electronic microarrays, chemogenetics, optogenetics, or novel pharmaceuticals that have yet to be discovered or repurposed. New basic science and clinical translation opportunities for targeting the powerful cardioprotective effects of the cardiac parasympathetic system are leading to exciting new avenues for this important field.

TABLE 1.

Comparisons of approaches used for increasing parasympathetic activity to the heart

Approach	Pros	Cons
Acetylcholinesterase inhibitors	Easy administration, strong responses	Short half-life with significant adverse off-target effects
Carotid baroreflex therapy	Strong acute responses in preliminary clinical trials	Requires surgery to implant, undetermined long-term efficacy and off-target effects
Cardiac ganglia neuron activation	Selective parasympathetic cardiac activation with strong, long acting responses in animal models	Significant translational barriers
Vagus nerve stimulation	Significant acute responses	Requires surgery to implant device, modest long-term efficacy and significant adverse off-target effects
Oxytocin excitation of brainstem cardiac vagal neurons	Preclinical animal work promising. In pilot clinical trials intranasal oxytocin treatment well tolerated, no adverse effects, easy administration	Long-term clinical efficacy unknown