The heart-brain axis in the context of cardiovascular disease

Abstract

The heart-brain axis forms an important physiological network, which is increasingly gaining recognition due to its involvement in cardiac function under steady-state conditions and pathological modifications of the heart in cardiovascular disease. Neurological disorders are known to affect cardiac function by propagating structural alterations in the heart. On the other hand, cardiovascular events have detrimental effects on the central nervous system affecting several brain regions, such as the hippocampus, which is important for cognition. Several anatomical regions of the brain, such as cortical and subcortical forebrain structures, regulate cardiovascular functions via the autonomic nervous system. The sympathetic and parasympathetic nervous systems, which are parts of the autonomic nervous system, play a crucial role in cardiovascular health. Cardiovascular disease, such as myocardial infarction (MI), activates the sympathetic nervous system, leading to exaggerated cardiac remodeling and subsequent arrhythmias. MI also alters afferent sensory neurons affecting nociceptive neurotransmission. This review focuses on the significance of the heart-brain axis and summarizes recent studies in this arena.

Introduction

The central nervous system (CNS) maintains physiological functions based on a regular influx of information from both external and internal sources. The afferent signals received from peripheral organs are processed by the CNS and transduced into efferent outputs directed towards target organs. This communication between the CNS and periphery is classified as the brain-body axis (1). This axis facilitates the homeostasis of the immune response and different systems such as respiratory, cardiovascular, renal, gastrointestinal, neuroendocrine, and musculoskeletal systems (1).

Considering this physiological control by the CNS, a complicated network of interactions exists between the heart and brain. The cortical and sub-cortical brain areas regulate cardiovascular function through the autonomic nervous system (ANS) (2). Understanding how the CNS and ANS affect heart physiology will benefit various disciplines including cardiology, neurology, and psychiatry. The physiological association between the heart and brain is bidirectional where dysfunction in one affects another and vice versa. The neurological disorders and cardiovascular diseases (CVD) are global burdens and leading causes of mortality and morbidity worldwide (3). The term ‘neurocardiology’ was coined during the 1980s to emphasize the importance of understanding the neurological mechanisms in CVD. Report suggests a steady rise in cases of depression-related illnesses and CVD due to their interrelation (4, 5). For example, patients with cardiac disease are more prone to develop cognitive decline, depression, cerebrovascular disorders, and other neurological disorders (3). For example, chronic heart failure (HF) patients develop cognitive decline (6), and coronary diseases affect the lateral and frontal occipital regions, which are crucial for cognitive functions (3). Several studies have demonstrated that HF leads to changes in the peripheral afferent networks and vascular connections. These alterations further disrupt the cardiac afferent system, the central processing of afferent signals, and the efferent innervation of target organs.

On the other hand, neurological disorders, such as stroke, epilepsy, and emotional disturbances, can cause CVD such as HF, atrial fibrillation, and MI. (7, 8). ANS disorders are major underlying causes of HF (9). The treatment strategies for HF, involving the stimulation of neurohumoral system, are considered effective (9). Neuroactive peptides stimulate sensory neurons in the forebrain and hindbrain, triggering the release of mediators that cross the blood-brain barrier and modulate various brain regions, ultimately altering cardiovascular function. Neuropeptides are produced in excess quantity in HF, resulting in regulation of ANS by brain (9).

As mentioned above, understanding the heart-brain axes is crucial for developing therapies to treat patients with neurological and cardiovascular diseases. Therefore, this review highlights recent studies examining the anatomical connections between the heart and brain, the effects of cardiac disease on different brain regions, and the role of the brain in cardiac healing after MI. We also discuss the clinical significance of the heart-brain axis to identify novel therapeutic targets.

Anatomical connections between the heart and brain

The cardiovascular and nervous systems display a complex set of interactions (2) (Figure 1). These interactions have been studied in the early 20^th century, when Gaskell and Langley described the anatomy of the ANS and its effects on the cardiovascular system (1). Specific regions of the brain, such as the hippocampus, amygdala of subcortical forebrain structures, hypothalamus, periaqueductal gray, ventrolateral medulla, and parabrachial region of the brain stem, are involved in the regulation of cardiovascular function (10). The dorsal cingulate and orbitofrontal cortex, as higher-order cortical regions, help in the processing of afferent and efferent information required for cardiovascular function (2). The thalamus, which is anatomically connected to the insular cortex, relays afferent cardiac input to posterior insula, where it integrates with the information from higher cortical regions in the rostroventral insula (11, 12). The orbitofrontal and prefrontal areas send signals to the central part of the amygdala, which connects the brain stem nuclei and hypothalamus to modulate cardiac responses to emotional stimuli (13-15). Some of the notable nuclei of the hypothalamus, such as paraventricular, lateral, and dorsomedial nuclei, alter the autonomic signals received from higher cortical regions. The periaqueductal gray, nucleus tractus solitaries, rostral ventrolateral medulla (RVLM), parabrachial region, and vagal dorsal motor nucleus in the brain stem play critical roles in this process (16) (17, 18). The hemodynamic signals from afferent neurons are received by the nucleus tractus solitarius present in the posterior medulla, which transmits efferent excitatory and inhibitory signals to the vagal dorsal motor nucleus and the RVLM, modulating parasympathetic and sympathetic outflow (19).

Figure 1: Various components of the heart-brain axis.

Schematic representation of the nerves interconnecting the brain and heart to form the heart-brain axis is depicted. In the heart-brain axis, sensory signals are transported from the heart to the brain via spinal afferents and the vagal nerve through the dorsal root ganglion, spinal cord or nodose ganglia. Motor signals from the brain are relayed through sympathetic and parasympathetic efferent fibers. The intrinsic cardiac nervous system interconnects with the heart and the sympathetic and vagus nerves. Biorender was used to generate this figure.

Preganglionic cholinergic sympathetic neurons innervating the heart originate in the intermediolateral cell column of the spinal cord. Neurons of the RVLM provide excitatory glutamatergic input to the sympathetic neurons. The sympathetic neurons synapse with noradrenergic neurons of cervicothoracic (stellate), superior, and middle cervical ganglia (19). The innervations are asymmetrical and variable, producing heterogeneous effects on the heart’s electrophysiological functions (19). Further, the ventricle, atria, and cardiac conduction systems are innervated by noradrenergic postganglionic sympathetic axons (2).

Additionally, the atrial conduction system, myocardium, and ventricles are innervated by the parasympathetic system through cardiovagal fibers, which originate from the vagal dorsal motor nucleus in the medulla and nucleus ambiguous (2, 20). Cholinergic parasympathetic neurons extend to cardiac ganglia via the superior and inferior cervical thoracic rami. These neurons then synapse with cardiac sympathetic nerves to form the cardiac plexus (19). Therefore, abnormalities in the ANS contribute to the pathogenesis of HF (21).

The cardiac afferent neurons and the intrinsic cardiac nervous (ICN) system also affect the cardiovascular system. The ICN system comprises interconnected ganglionated plexi present in epicardial fat tissue, which innervate the atrioventricular and sinoatrial nodes as well as the pulmonary vein–left atrial junction. This junction is also innervated by parasympathetic and sympathetic neurons (22, 23). The interneurons of the ICN synapse with local ganglionic neurons, and parasympathetic and sympathetic neurons interact with other neurons, altering autonomic efferent fibers (24).

The heart supplies blood to the brain through two pairs of arteries: vertebral arteries perfuse the posterior brainstem, cerebrum, and cerebellum and the internal carotid arteries supply the anterior cerebrum (25). The anterior and posterior circulations are interconnected by the Circle of Wills, which give rise to paired cerebral arteries, which further branch into small arteries to penetrate brain parenchyma. The smallest vessels, known as the pial arterioles, form smaller blood vessels that pass through Virchow Robin space after entering the brain surface (26) (25). The blood vessels become less dense after penetrating deep inside the brain tissue. The arterioles are straight in cortical gray matter whereas they form loops after reaching the subcortical white matter (26, 27). As a result, during ischemic conditions when global cerebral blood flow is compromised, the subcortical areas are severely affected (26).

Sympathetic neuronal activation in cardiovascular disease

The sympathetic nervous system (SNS) is important for cardiovascular health since it is involved in the regulation of arterial blood pressure, cardiac output, heart rate, and systemic vascular resistance in coordination with the parasympathetic nervous system (28) (29). These effects are mediated by several mechanisms including cardiac and vascular sympathetic activation, epinephrine and norepinephrine secreted by the adrenal medulla, and sympathetic activation of renal juxtaglomerular cells that stimulate the renin-angiotensin-aldosterone system (RAAS) (29). Further, elevated plasma norepinephrine levels are correlated with increased neural activity. In this regard, Leimbach et al reported that sympathetic nerve activity in muscle was higher in individuals with HF compared to healthy patients and correlated with plasma norepinephrine concentration (30). Exaggerated activity of sympathetic nerves was subsided after heart transplant surgery in patients with HF, suggesting a bidirectional association between the SNS and heart function (31). Additionally, arterial and cardiopulmonary baroreceptors control central sympathetic outflow. Reduced baroreceptor activity due to increased aldosterone or reduced cardiac function decreases the afferent signals, which generally prevent sympathetic efferent activity, resulting in activation of the SNS (32) (33). MI is often associated with the activation of the RAAS and SNS, which can exacerbate cardiac remodeling by increasing stress on the myocardium, decreasing the expression of the β₁ adrenergic receptor, and exerting toxicity on the myocardium (34). Following MI, microglia and proinflammatory cytokines within the paraventricular nucleus enhance the activation of the SNS and the hypothalamic-pituitary-adrenal axis (35). Hyperactivation of the sympathetic system advances HF phenotype (31) by increasing ouabain-like components in the brain, which enhance the RAAS (36) and escalate norepinephrine discharge by the kidney and heart (34) (37). Acute MI stimulates the release of local nerve growth factor (NGF) and growth-associated protein 43, resulting in the sprouting of cardiac nerves in the infarct (5). For neuromodulation, several techniques are being considered. For example, non-invasive transcranial ultrasound stimulation to induce the reparative phenotype of microglia can thwart neuroinflammation and cardiac remodeling (35). Moreover, low-intensity focused ultrasound improves imbalance of the ANS after MI, preventing ventricular arrhythmias. Sympathetic activation has also been linked with myeloid cell production. Our lab reported that norepinephrine levels are correlated with myeloid cell numbers in patients with diabetes (38). Surprisingly, catecholamines are produced by tyrosine hydroxylase-expressing splenic leukocytes in diabetic mice due to enhanced sympathetic neuronal activity. Tyrosine hydroxylase-expressing leukocytes express higher levels of neuropeptide Y receptors, which facilitate the neuroimmune connection. The catecholamines bind to the β₂ adrenergic receptor present in granulocyte macrophage progenitors, triggering their proliferation and myeloid cell production. Diphtheria toxin-mediated ablation of tyrosine hydroxylase-positive leukocytes, genetic ablation of the β₂ adrenergic receptor, and surgical splenic sympathectomy attenuated myeloid progenitor proliferation, myeloid cell production, and atherosclerosis in diabetic mice (38). To summarize, the SNS has a well-defined role in cardiac remodeling after MI. Therefore, sympathetic modulation after MI can curb the adverse remodeling outcomes.

The role of sensory neurons in cardiac function

The heart is electromechanically regulated by the ANS. The cardiac nervous system is composed of efferent parasympathetic and sympathetic neurons that innervate cardiomyocytes and coronary arteries as well as afferent sensory neurons that convey information to the CNS to modulate heart function. Sympathetic, parasympathetic, and sensory neurons arise from neural crest cells. Specifically, sympathetic and sensory neurons are derived from trunk neural crest cells whereas parasympathetic neurons originate from cardiac neural crest cells (39). The cell bodies of afferent cardiac sensory neurons are located in the dorsal root ganglia (DRG), nodose ganglia, and intrathoracic, extracardiac, and intrinsic cardiac ganglia. The neurons of the dorsal root and nodose ganglia are somatic neurons that respond to pain stimuli (40) (41). These pain-sensitive neurons often induce cardioprotective actions after MI. The conduction of sensory cues from the heart to the CNS occurs through cardiac afferents consisting of myelinated A fibers and unmyelinated C fibers, which project to the upper thoracic dorsal horn via the DRG (39). The DRG harbor the cell bodies of sensory neurons that can detect and transmit various physical impulses. Sensory neurons have distinct electrophysiological and histochemical characteristics (42) and have been reported to maintain immune homeostasis in the lung (43). JAK1 present in vagal sensory neurons regulates the expression of the calcitonin gene-related peptide β (CGRPβ) gene, which suppresses allergic inflammation in the airway. Sensory neurons also play a key role in regulating immune cells that control tissue regeneration and repair after injury (44). Lu et al demonstrated that Na_V1.8 nociceptor ablation disrupted skin wound repair and muscle regeneration. CGRP released by sensory neurons binds to the receptor activity-modifying protein 1 (RAMP1) on myeloid cells and promotes efferocytosis and macrophage polarization towards a pro-reparative phenotype. MI releases different metabolites, such as prostaglandins and bradykinin, which activate sensory neurons. Transient receptor potential vanilloid 1 (TRPV1)-expressing cardiac afferent fibers, which are sensitized in HF, activate the cardiac sympathetic afferent reflex (CSAR). Accordingly, the inhibition of TRPV1 using resiniferatoxin decreases CSAR activation in rats (45). Epicardial administration of resiniferatoxin after MI in rats resulted in decreased CSAR, subsided sympathetic activation, and enhanced baroreflex sensitivity (46). Furthermore, resiniferatoxin attenuated lung edema, left-ventricular end-diastolic pressure, and cardiac hypertrophy mediated by β-adrenergic receptor excitation via isoproterenol administration in rats with chronic heart failure. Molecular experiments revealed resiniferatoxin-mediated diminution of apoptosis, TGF-β receptor I activation, and fibrotic marker expression, leading to reduced cardiac fibrosis and improved end-diastolic pressure volume. In contrast, TRPV1 afferent neurotransmission enhances fibrosis, worsens cardiac remodeling, and exerts electrical dysfunction, leading to ventricular arrhythmogenesis following MI (47). A separate study demonstrated that the cardiac sympathoexcitatory reflex is moderated by capsaicin-sensitive afferent nerve fiber activation (45). The afferent nerve fibers expressing the vanilloid receptor 1 (VR1) are found on the epicardial surface of the rat heart. Resiniferatoxin administration can decrease the expression of this receptor and lower capsaicin-induced arterial blood pressure and renal sympathetic nerve firing (45). Chronic pain following MI induces changes in the DRG and may also modulate the nodose ganglia (48). The functional analysis of nociceptive neurons in a porcine model of MI indicated that MI enhances nociceptive neuron numbers but reduces the neural signaling of nociception. MI also raised the abundance of GABA-expressing CGRP⁺ neurons in the nodose ganglia, resulting in a reduction of nociceptive neurotransmission and vagal retreat (48).

Alterations in various brain regions in cardiac diseases

Sudden cardiac alterations, such as structural changes and abnormality in electrocardiography including T wave inversion and prolong ST and QT segments, are often associated with neurological disorders, which can be asymptomatic or life-threatening conditions (2). CVD has severe effects on brain regions such as the frontal cortex, hippocampus, thalamus, cingulate gyrus, and precuneus (49). Higher CVD risk factors are often linked to reduced volumes of brain regions such as the hippocampus. Even patients without any cardiac disease but with significant CVD risk factors have alterations in brain structures (50). Magnetic Resonance Imaging (MRI) of the brain indicates that patients with HF display atypical brain changes (51). Additionally, Zheng et al demonstrated that HF patients display increased apparent diffusion coefficient (ADC) in different brain regions such as the cingulate cortex, cerebellum, hippocampus, deep nuclei, and frontal, temporal, parietal, and occipital lobes. ADC measures the magnitude of water molecule diffusion in tissues and is derived from diffusion-weighted imaging in MRI (52). Impaired water movement within tissues due to pathological changes results in lower ADC values. Therefore, micro changes in brain structure, as indicated by altered ADC, can lead to cognitive impairment as well as myocardial fibrosis (51). Park et al reported that HF patients exhibit anomalies in resting-state brain connectivity, resulting in deficits in cognitive and autonomic functions. The investigators employed resting state functional MRI to evaluate brain network topology and functional connectivity of the brain in healthy and HF patients. HF patients exhibit reduced functional connectivity in the cerebellar and precentral gyri, olfactory sites, and medial, frontal, and vermis regions whereas the functional connectivity in areas such as the orbito/medial frontal and superior parietal gyrus regions, cerebellar lobe, and fusiform gyrus remains unaffected. These alterations in brain connectivity of HF patients interfered with the organization of brain network, resulting in exacerbated neurological deficits (53).

HF patients develop lateralized neural injury with mood disorders and poor cognition (54). Vascular cognitive impairment is the term used to describe cognitive impairment due to cerebrovascular disease including cerebral small-vessel disease, cerebral amyloid angiopathy, cerebrovascular artery occlusion, parenchymal lesions such as atrophy of white and gray matter, inflamed perivascular spaces, brain infarcts, brain microbleeds, and profound hemorrhages (55) (56) (57). Brain dysfunction and cerebral injury in HF can lead to cognitive impairment (58, 59). Cognitive impairment accompanied by alterations in brain anatomy due to unknown causes in HF patients is known as cardiocerebral syndrome (60-62). MRI in these patients displays white matter hyperintensities and atrophy of the frontal cortex and hippocampus (63, 64). Almeida et al conducted a study in 45 years old individuals with HF and ischemic heart disease as well as control patients and compared their MRI images to analyze region-specific changes in cerebral gray matter volumes. This experiment indicated that HF patients show signs of short- and long-term memory loss due to impaired psychomotor function compared to the controls (63). A study by Wang et al compared patients with HF displaying signs of cognitive impairment and patients with HF without cognitive dysfunction (58). The patients with HF with cognitive impairment showed decreased local and global brain connections, a reduction in in white matter networks, and weaker local networks in important brain regions. The degree of cognitive impairment is often correlated with the severity of HF. Indeed, Frey et al (65) reported that 41% of HF patients with a higher degree of medial temporal lobe atrophy, as evidenced by MRI, showed defects in reaction time and 46% of them had defective verbal memory. Moreover, the extent of medial temporal lobe atrophy was proportional to the severity of cognitive dysfunction in these patients (65). Another study comparing control and MI patients demonstrated that patients had no detectable acute effects on executive function, memory, and global cognition immediately after MI but later exhibited cognitive decline over years (66).

The mechanisms behind HF-associated cognitive impairment are not well understood. Dridi et al underscored the effects of dysfunctional ryanodine receptor type 2 (RyR2) on cardiomyocytes of HF patients with cognitive decline (67). These disorders are referred to as channelopathies caused by abnormal ion channels in cellular membranes and organelles (68). The experiments by Dridi et al showed that intracellular Ca²⁺ channels and RyR2 are more permeable and undergo post-translational modifications in HF. The RyR2 undergoes protein kinase A activation, oxidation, nitrosylation, phosphorylation, and other post-translational modifications, such as depletion of calstabin 2, which is a stabilizing subunit of RyR2. The alterations in RyR2 are partly due to TGF-β pathway activation and increased adrenergic signaling. When mice with HF were treated with the beta blocker propranolol, a TGF-β inhibitor, and an RyR2 stabilizing drug, they exhibited cognitive improvement. Genetically engineered mice resistant to RyR2 Ca²⁺ leak displayed a similar phenotype. Additionally, a higher resting heart rate accelerates the chances of cognitive decline and possibility of dementia (59, 69). Arteriosclerosis, characterized by thickening, hardening, and loss of elasticity of arterial walls, is a risk factor for developing dementia and cognitive decline (70, 71), which could be attributed to structural alterations of small blood vessels in the brain (71). Studies have associated systemic arteriosclerosis markers with smaller volumes of brain regions, cerebral microbleeds, and white matter hyperintensities (72) (71, 73). Chronic hypertension, which is associated with high cardiovascular mortality, leads to impaired cerebral blood flow and is a significant risk factor for Alzheimer’s disease (AD) (74). Uncontrolled hypertension also contributes to thickening of vessel walls, thereby affects their elasticity (75, 76), and alters blood-brain barrier integrity, resulting in cerebral edema (74, 77). Santos et al demonstrated a moderate correlation between higher cardiac workload at rest and cortical amyloid-β burden in the preclinical phase of AD in middle-aged patients (78).

Furthermore, impaired cardiac output causes abnormal cerebral blood flow (CBF), affecting both white and gray matter. When the CBF in control and HF patients was analyzed using pseudo-continuous arterial spin labeling (ASL), the analysis uncovered decreased CBF in HF patients in the bilateral prefrontal, thalamus, corona radiata, cerebellum, corpus callosum, amygdala, hippocampus, and temporal, frontal, and occipital cortices (79). Muller et al, using MRI, reported a negative correlation between HF and gray matter density (GMD) using voxel-based morphometry and HF markers specifically elevated levels of N-terminal pro-B-type natriuretic peptide (NTpro-BNP) and reduced ejection fraction (EF). They studied three groups of patients. Two of the groups had coronary artery disease (CAD) and higher levels of NTpro-BNP, and the control group had normal NTpro-BNP concentrations in the blood. The results indicated that CAD patients with increased levels of NTpro-BNP displayed diminished GMD in the precuneus and posterior cingulate cortex. Even decreased EF resulted in lower GMD in the right and left orbitofrontal cortex. Overall, the study suggests that lower EF and higher levels of NTpro-BNP in HF patients can lead to brain damage (80).

Atherosclerosis is characterized by the accumulation of lipid in the arterial walls followed by immune cell infiltration, leading to plaque formation (81) (82). Recent studies highlight the significance of neuroimmune interactions in atherosclerosis (83) (82). Autonomic imbalance induces atherosclerosis, activates systemic inflammation, and propagates tissue ischemia by constricting blood vessels (84). Gils et al reported the importance of netrin-1 in macrophage migration induced by CCL19 and CCL2 (81). Netrin-1 deletion in macrophages caused their retention in the arterial wall, resulting in reduced atherosclerosis burden (81).

While CVD can alter the CNS structurally and functionally, neurological abnormalities can also affect cardiovascular disease and heart structure. For example, depression has been shown to increase the risk of coronary artery disease. In a study by Bremner et al, CAD patients with depression showed enhanced stimulation of the parietal cortex and diminished activation of the anterior and prefrontal cingulate compared to CAD patients without depression. Additionally, CAD patients with MI and depression had activation of the rostral area of the anterior cingulate (85). Epilepsy is another neuronal condition caused by asynchronous neuronal activity in the brain (86, 87). These patients exhibit relatively higher occurrence of cardiac arrhythmias compared to control patients without epilepsy (88) (87). During the ictal phase, the ongoing seizure of epilepsy patients is concomitant with increased cardiac arrhythmia (89, 90). However, due to the absence of constant electroencephalograph (EEG) surveillance, it is unclear how the occurrence of arrythmias is causally associated with seizures (91). Some epilepsy patients also suffer from ictal bradycardia (92). Analyses of the UK Biobank data showed that epilepsy patients can also develop bradyarrhythmias, atrial fibrillation, and ventricular arrhythmias (88). These findings indicate crucial interactions between the brain and heart, which can be targeted for the treatment of CVD.

The contribution of the brain in cardiac healing after MI

Ischemic myocardial disease is a major cause of morbidity and mortality (93). Myocardial ischemia is associated with remodeling and pathophysiological changes of the heart, such as enhanced oxidative stress, exaggerated immune responses, fibrosis, and scar formation (93). Patients with CVD display neurological alterations that include higher stress levels and emotional and motivational changes. The activation of the ventral tegmental area (VTA), which is involved in dopaminergic neurotransmission and plays an important part in emotions, motivation, and expectation, enhances cardiac remodeling. Mice with MI and experimental VTA inactivation had less cardiac fibrosis, altered immune responses, enhanced angiogenesis, higher left ventricular ejection fraction, and lower end systolic and end diastolic volumes (93). An interesting study by Huynh et al demonstrated that MI mobilizes monocytes into the brain to increase sleep, which reduces sympathetic outflow and lower cardiac inflammation and ventricular remodeling (94). After MI, microglia draw monocytes into the thalamic lateral posterior nucleus (LPN) of the brain through the choroid plexus to produce tumor necrosis factor (TNF), which stimulates glutamatergic neurons expressing Tnfrsf1a to enhance slow-wave sleep pressure. Disturbance in sleep following MI disrupts cardiac function, resulting in ventricular tachycardia, reduced heart rate, and elevated susceptibility towards secondary cardiovascular occurrences (94). The cardiac remodeling after MI is also reported to be regulated by the renin–angiotensin–aldosterone system (95). The heart is exposed to aldosterone which may come from the circulation or local production after MI (95, 96). Aldosterone locally produced in the brain induces mineralocorticoid receptors in the CNS, causing enhanced sympathetic outflow (97, 98), which increases the release of vasopressin and arginine (99). To understand the impact of systemic and CNS aldosterone, Lal et al administered spironolactone orally or through the intracerebroventricular route in rats (95). Treatment via both routes reduced sympathetic tone, diminished the internal circumference of the right and left ventricles, and suppressed perivascular and interstitial fibrosis. The laminin and fibrillar collagens were less abundant in the spironolactone-treated rats compared to controls. Therefore, the study suggests that aldosterone present in the CNS is involved in cardiac remodeling (95). In another study, Ziegler et al. demonstrated that denervation of sympathetic axons in pineal glands of mice renders them susceptible to cardiac disease (100). Cardiac disease expands inflammatory macrophage numbers in the superior cervical ganglia (SCG), resulting in fibrosis and loss of sympathetic neurons innervating the pineal gland. Depletion of macrophages in the SCG inhibited the denervation of the pineal gland and restored melatonin secretion (100).

Clinical significance of the heart-brain axis

The studies discussed above demonstrate that CVD, including MI and HF, affects the communication between the heart and the brain. Many therapeutic approaches are being considered to maintain the physiological function of the heart-brain axis (Figure 2). Pharmacological mediators are used to maintain homeostasis of the ANS in cardiometabolic diseases and HF (82). The three major pharmacological strategies are: 1) prevention of SNS activity using sympathoplegic drugs, 2) inhibition of transmission at the ganglion, and 3) blockade of peripheral nerve transmission and adrenergic receptors (82). However, some ganglionic blockers, such as nAChR inhibitors, have limited clinical applicability due to severe side effects (101). Similarly, sympathoplegic drugs have moderate effectiveness (102). Inhibitors of angiotensin-converting enzyme (ACE), which is crucial for the heart-brain axis, are used in patients with HF (103). The RAAS and SNS can regulate each other. Stimulation of the β1-adrenergic receptor causes the release of renin from the kidney, activating the RAAS, which stimulates the SNS and CNS (104, 105). Therefore, a decrease in Ang II secretion inhibits sympathetic activity and the release of norepinephrine (106) (107). Of note, sodium-glucose cotransporter 2 inhibitors reduce cardiovascular mortality. Although the mechanisms of action of these compounds are not clear, some reports show that they also act by altering the SNS (108) (109). The paraventricular nucleus (PVN), present in the hypothalamus, regulates the SNS and fluid homeostasis (110). The forebrain enhances PVN neuronal activity via the renin-angiotensin network, leading to HF. Administration of ACE inhibitors, angiotensin type 1 (AT1) receptor antagonists, and mineralocorticoid receptor antagonists decreases the activity of PVN neurons in HF (110). Mineralocorticoid receptor antagonists are standard therapy in HF as they prevent mortality and morbidity (111). This treatment suppresses the activity of the renal sympathetic system, maintains fluid homeostasis, and decreases TNF-α in HF by reducing sympathetic drive and enhancing baroreflex activity. Hence, mineralocorticoid receptor antagonists are crucial in regulating volume and sympathetic drive in HF with reduced ejection fraction (111). β-adrenergic receptor antagonists, such as bisoprolol and metoprolol, have demonstrated a reduction in mortality in HF (112) (113).

Figure 2: Therapeutic strategies targeting the heart-brain axis to modulate cardiovascular disease.

The therapeutic interventions of the heart-brain axis include modulation of neurohormonal activity, suppression sympathetic activation, maintenance of fluid homeostasis, and stimulation of the vagus nerve, resulting in lower heart rate, reduced LVESV, and improved cardiac ejection fraction. Additionally, pharmacological inhibitors decrease sympathetic activation, prevent ganglionic transmission, and obstruct adrenergic signaling. In various recent clinical trials, such as the BeAT-HF trial, spinal cord stimulation was carried out to attenuate HF pathogenesis. Exosome transplantation is also considered as a therapeutic strategy for cardiovascular disease to improve patient survival. Biorender was used to generate this figure.

Exosomes have been proposed to be a crucial biomarker for early diagnosis of cardiovascular and pulmonary diseases (114) (115). Although exosome abundance increases in most diseases, the cargo contents of exosomes, including RNA and proteins, are specific to a disease (116, 117). The exosome bilayer made of cholesterol, sphingolipids and ceramides preserves and protects the cargo materials (118). miR-192, miR-194, and miR-34a levels in exosomes were elevated after MI in patients who subsequently developed HF, suggesting the significance of these miRs in adverse cardiac remodeling (119). Additionally, this study reported a correlation between left ventricular end-diastolic dimension and miR-34a and miR-194 contents at one year following MI (119). In acute ischemic stroke, patients exhibited augmented levels of circulating exosomes containing brain-specific miRNAs such as miR-124 and miR-9 (120, 121). The increase in these miRs was correlated with infarct size, highlighting the significance of these brain-derived miRs in stroke prognosis (120) (121).

Bommel et al reported that HF patients after cardiac resynchronization therapy showed improved peak systolic velocity and end diastolic velocity resulting in better left ventricular systolic function as well as cerebral blood flow (CBF) (122). Another study illustrated that captopril treatment restored CBF in HF patients (123). Heart transplantation surgery also enhanced CBF, which was calculated by computed tomography using ¹³³Xe as a tracer and transcranial Doppler ultrasound to measure blood flow in the middle cerebral artery, within the first month following surgery (124). Continuous-flow left ventricular assist devices (CF-LVADs) also improved CBF. Cornell et al compared cerebral autoregulation in healthy controls, patients with CF-LVADs, and patients with pulsatile LVADs patients using the parameters cerebral blood flow velocity and mean arterial pressure (125). CBF in CF-LVADs was comparable to that in healthy patients indicating that lower pulsatility in CF-LVADs does not affect autoregulation (125).

Advanced technologies are employed to develop refined electrodes to activate the nerves innervating the tissue of interest (82). Several clinical studies have used neuromodulation to alter the functions of autonomic nerves (82). In this context, vagus nerve stimulation (VNS) has been reported to potentially reduce HF-related mortality (21, 126). VNS has been experimentally shown to improve autonomic imbalance, decreasing the risk of HF related issues (21, 126). Preclinical studies conducted in different canine and rat models have indicated improved cardiac function in experimental HF. For example, Li et al demonstrated, in a rat model of MI-induced HF, that increased electrical stimulation of the vagal nerve suppressed cardiac remodeling and improved blood circulation (127). In this study, MI was induced in rats via left coronary artery ligation, after which an implantable miniature radio-controlled electrical stimulator was used for right vagal nerve stimulation for 6 weeks. This resulted in reduced end diastolic pressure of the left ventricle with enhanced cardiac contractility (127). Similarly, Hamann et al demonstrated that canines with intracoronary microembolization-induced HF displayed improved cardiac conditions with VNS therapy (128). Implantation of a pulse generator and a bipolar cuff electrode around the right cervical vagus nerve improved left ventricle parameters such as ejection fraction and end systolic volume, thwarted left ventricle remodeling, and reduced HF biomarkers (128). Further, clinical studies were conducted to examine the therapeutic benefits of VNS therapy in patients. VNS was approved for patients with drug-resistant depression (2005) and epilepsy (1997). Schwartz et al conducted a pilot study on eight HF patients with Cardiofit implantation for VNS (129). The patients exhibited reduced left ventricular end-systolic volume after VNS. However, VNS neuromodulation affects may have off-target effects on the systems such as the immune and gastrointestinal systems (21). Bioelectronic medical devices that induce neural reflexes can cause adverse side effects on the vagus nerve. Moreover, clinical studies have shown inconsistencies in the required stimulation intensity of the vagus nerve and in patients tolerance (21). Therefore, preclinical experiments are being conducted to reduce the off-target effects of VNS and increase its safety by stimulating branches of the vagus nerve located near specific organs. For instance, the celiac efferent branches of the vagus nerve can be activated to stimulate the splenic nerve (130). This treatment has the potential to modulate splenic inflammatory cell production, thereby avoiding any direct adverse effects on other internal organs.

Baroreceptors present in the aortic arch and carotid sinus aid in maintaining blood pressure by altering ANS tone (82, 131). In the Baroreflex Activation Therapy for Heart Failure (BeAT-HF) trial, baroreflex activation therapy (BAT) improved quality of life in patients although HF-associated morbidity was not reported (132). There are several recent studies examining the benefit of spinal cord stimulation, which is achieved by placing electrodes in the posterior epidural cavity, to prevent HF after MI. Although the exact mechanisms of this therapy are still unclear, the treatment may prevent efferent sympathetic nerve overactivation (82). Although a preliminary trial in patients suffering from systolic HF demonstrated improved left ventricular function following spinal cord stimulation (133), double blinded clinical trials with a larger number of patients are required to confirm the therapeutic benefits of this neuromodulatory approach.

Conclusion

In summary, there exists a complex bidirectional crosstalk within the heart-brain axis involving the CNS, ANS, and cardiovascular system. The discovery of HF patients showing signs of cerebral abnormalities and neurological disease has revealed interconnections between the heart and brain. There are specific brain regions such as the hippocampus, amygdala of the forebrain, and brain stem, which are known to regulate cardiovascular function. The neurocardiac axis has emerged as a crucial link to CVD. Additionally, cardiovascular disease can affect cognitive functions, which requires thorough evaluation. Following a cardiovascular injury, the brain signals through different pathways to precipitate cardiac remodeling. The SNS working with the parasympathetic nervous system controls cardiovascular health. Moreover, sensory neurons have also been reported to influence cardiac recovery following MI. Recent studies have expanded our knowledge of the heart-brain axis and have revealed several therapeutic targets. Future clinical trials will validate the benefits of these molecular targets. Furthermore, as discussed above, neuromodulation is a potential treatment to quell inflammation and cardiac remodeling. Neuromodulation strategies need to be improved to enhance efficacy and limit off-target effects. In summary, understanding the mechanisms of neuromodulations and the interactions between the heart and the brain will unravel new therapeutic avenues for patients with CVD.