The p75 neurotrophin receptor, semaphorins, and sympathetic traffic in the heart

the sympathetic and parasympathetic branches of the autonomic nervous system control cardiac function through the coordinated activity generated by neurons within the intracardiac autonomic ganglia and central nervous system (3). Activation of the sympathetic efferent projections to the heart increases heart rate, modifies the pattern and conduction velocity of the impulse through the heart, and augments both atrial and ventricular contractility while hastening myocardial relaxation. In the vasculature, the sympathetic nervous system (SNS) reduces venous capacitance and constricts resistance vessels, increasing blood pressure. Conversely, the parasympathetic nervous system (PSNS) decreases heart rate and cardiac impulse conduction. More recent evidence has suggested that the PSNS may also affect ventricular function, indirectly by countering β-adrenergic action or directly via inhibition of L-type Ca²⁺ channels (34). A transient imbalance between these two branches can lead to increased heart rate and blood pressure when cardiac output is increased. When subnormal cardiac PSNS activity concomitant with sympathetic hyperactivation persists, such as after myocardial infarction, this disequilibrium may increase the risk for cardiac adverse events, including arrhythmias, hypertension, cardiac hypertrophy, and, ultimately, heart failure. This condition is called “autonomic neural remodeling” and can greatly contribute to the induction of ventricular fibrillation, thus increasing the patient's susceptibility to sudden cardiac death (41, 47). It is plausible that neurotrophic factors, target organ-secreted proteins essential for the development and differentiation of neurons, are involved in the remarkable neuroplasticity exhibited by the SNS in the adult life, including autonomic remodeling after infarction and/or in the onset and progression of congestive heart failure (17, 30, 32).

The developmental formation of the SNS is intricately controlled by various growth factors, cytokines, and axon guidance molecules. As stated by Sossin and Barker (45):

The first few days of a neuron's life are tough. Within minutes of being born, it starts extending processes in an attempt to reach appropriate target cells before competitors do. It then has to integrate trophic and activity-dependent cues to determine whether target contact has been achieved.

This scenario is also true for sympathetic neurons. One of the primary regulators of postmitotic sympathetic neuron survival and target tissue innervation is nerve growth factor (NGF) (21). NGF is produced by the heart and vasculature and promotes sympathetic neuron survival and axonal growth and branching. NGF is the founding member of the neurotrophin (NT) family, which includes brain-derived neurotrophic factor (BDNF), NT3, and NT4. Growing evidence has suggested a critical role for NTs in governing cardiovascular functions such as the control of blood vessel and smooth muscle cell growth as well as heart development (9). Early studies have shown the ability of the embryonic chick heart to secrete neurotrophic factors such as NGF, which, in turn, stimulate neurite extension from sensory neurons in vitro (15). More recently, NGF has been shown to be present in neonatal rat myocytes, where it acts as an autocrine survival factor. In addition, increasing the expression of NGF protected the cells from ANG II-induced apoptosis, and, in a rat model of myocardial infarction, NGF gene transfer promoted cardiomyocyte survival (10). Similarly, BDNF gene deletion severely affects heart development, leading to premature death in mice (14, 16), and mice lacking the receptor for BDNF (TrkB) die immediately after birth (1). Thus, a deficit in NT signaling may occur in the diseased heart, affecting not only sympathetic transmission (35, 36) or catecholamine reuptake at the neuroeffector junctional areas (32) but also the viability of the cardiomyocytes themselves. Yet, overexpression of NGF in the heart from birth leads to cardiomegaly and pathological hypertrophy (24). Therefore, NTs may be required in ad hoc conditions in which their NT signaling is lost or hampered, but presumably their expression should not exceed physiological concentrations. Finally, different NTs may have different effects. NT3 mRNA is downregulated in the adult hypertrophic rat heart, and NT3 is likely necessary for countering maladaptive hypertrophy (28). Conversely, BDNF mRNA is upregulated in high salt-induced cardiomyopathy in rats (30), but whether this is a compensatory adaptation or a worsening feature remains to be established. Differing responses to NTs are also true for the growth of sympathetic neurons. BDNF produced by these same neurons (and/or by their target organs) has been shown to act through the p75 neurotrophic receptor (p75^NTR) to antagonize NGF-mediated growth and target innervation. Thus, it has been suggested that sympathetic target innervation is the result of a finely tuned balance of positively and negatively acting NTs present in developing and potentially mature targets (29).

The survival and axonal growth-promoting effects of NTs are primarily mediated by the activation of the Trk family of tyrosine kinase receptors; however, all of the NTs also bind to p75^NTR, a member of the TNF receptor family. This receptor is a pleiotropic signaling molecule and is able to induce a wide variety of biological effects (20). In sympathetic neurons, p75^NTR forms a high-affinity receptor complex with TrkA and facilitates the growth and survival effects of NGF (11). In addition, the interaction of p75^NTR with TrkA enhances its selectivity for NGF over NT3, which is critical as developing sympathetic neurons switch from their initial dependence on NT3 as they grow toward their targets to NGF upon reaching their destination (33). In contrast, BDNF, produced by these neurons, promotes apoptosis through binding to p75^NTR alone (4). This ability to induce cell death is crucial for establishing the correct balance between the size of the neuronal population and the tissue it is innervating (12). BDNF activation of p75^NTR has also been shown to induce axonal degeneration, selectively in the fiber exposed to the NT. The neuronal soma and other fibers remained intact as long as they were exposed to NGF (44). The absence of these regressive signals in p75^NTR-null mice is thought to underlie the sympathetic hyperinnervation observed in the eye (44) and pineal gland (4). However, analysis of tissue innervation patterns in p75^NTR-deficient (p75^NTR−/−) mice is further complicated by the fact that the receptor can also partner with the semaphorin 3A (Sema3A) receptors neuropilin-1 and plexin A4 and attenuate their ability to repel growing axons (5). In addition, p75^NTR interacts with the Nogo receptor complex and, in this context, actually mediates growth-repulsive signals activated by a family of myelin proteins (13). Hence, understanding the role of p75^NTR in the sympathetic regulation of cardiovascular function is a challenging and complex undertaking that may have implications in chronic cardiac disease such as heart failure (7).

Previous studies have revealed a developmental delay in cardiac sympathetic innervation in p75^NTR−/− mice (at embryonic day 16.5) (33) compared with wild-type (WT) mice, but 11–14 days after birth the arborization was grossly normal in mice lacking p75^NTR (27). However, through more careful analysis, Habecker and colleagues (22) demonstrated altered sympathetic axonal input to the right atria of adult p75^NTR-null mice. They found an increase in sympathetic density in p75^NTR−/− atria 4 wk after birth that was replaced by an overall reduction in the maintenance of the innervation in the adult. The arborization pattern was highly heterogeneous, such that some regions were hyperinnervated while others were devoid of sympathetic input. Conversely, parasympathetic innervation was normal at both stages despite the fact that cholinergic neurons of the intrinsic cardiac system also express p75^NTR (25). p75^NTR-null animals also exhibited a reduced basal heart rate and response to restraint stress (22). These altered responses seemingly resulted from the loss of sympathetic transmission rather than increased PSNS activity and/or direct effects of p75^NTR in cardiac cells. Indeed, p75^NTR-null atria exhibited threefold higher norepinephrine (NE) content compared with aged-matched controls (22). Thus, it is plausible that due to sympathetic remodeling in these atria, i.e., decreased density and altered distribution of sympathetic fibers, only a fraction of the NE present in p75^NTR−/− sympathetic fibers is accessible for release even when a direct releasing stimulus, such as tyramine, is applied. A close interplay between sympathetic innervation and p75^NTR in the heart is confirmed by the opposite condition. The induction of sympathetic hyperinnervation by a local infusion of NGF in the dog myocardium or left stellate ganglia is followed by positive p75^NTR immunolabeling in interstitial cells and the perivascular area of the myocardium as well as in axons, Schwann cells, and interstitial cells of sympathetic nerve fibers (48). Notwithstanding, the precise role of p75^NTR in the sympathetic innervation of target tissue and the functional consequences of its genetic ablation are not well understood.

In the current issue of American Journal of Physiology-Heart and Circulatory Physiology, Lorentz and coworkers (38) further explored the role of p75^NTR in regulating the sympathetic innervation of the heart and reported a dramatic loss of fiber input to the subendocardium of the left ventricle (LV) while that to the subepicardium and right ventricle was not significantly altered. This remarkable phenotype is reminiscent of that seen in Sema3A-overexpressing mice (26), prompting Lorentz and colleagues to consider the possibility that a p75^NTR interaction with the Sema3A system may explain this selective loss of innervation. To test this hypothesis, they cultured stellate ganglia from WT and p75^NTR−/− mice and demonstrated that the null neurons were much more sensitive to Sema3A-mediated inhibition of neurite outgrowth. This result suggests that as sympathetic neurons are innervating the heart, p75^NTR acts to blunt the repulsive effects of Sema3A, which is expressed in the subendocardium, thereby allowing axonal arborization. In the absence of p75^NTR, Sema3A potently repels the growing axons away from the subendocardium.

While it will be interesting to see if deletion of Sema3A can compensate for the loss of p75^NTR, major functional outcomes that differentiated p75^NTR−/− mice from WT mice emerged from the present study. When the authors examined the ventricular structure and function by echocardiography (under anesthesia), they found no major differences between the two genotypes in terms of LV dimensions, fractional shortening, ejection fraction, and cardiac output at rest. However, they discovered decreased mean arterial pressure in p75^NTR−/− mice accompanied by a drop in both systolic and diastolic pressure. These alterations are likely due to the peripheral (vascular) expression of Sema3A, which may decrease sympathetic innervation (37), thereby decreasing vascular tone, accounting for an overall reduction in afterload. The presence of concomitant afterload reduction (as suggested by present data showing reduction in mean, systolic, and diastolic pressure), unchanged LV dimensions (from echo data), and decrease in dP/dt_max and dP/dt_min in p75^−/− mice strongly hints at the possibility that these mice have an overall reduction in cardiac inotropy, a situation reminiscent to that observed with β-blockers such as verapamil. However, additional studies using a load-independent approach directly assessing LV inotropic status independent of any possible change in pre- and afterload, e.g., pressure-volume relationships, would definitely confirm this eventuality. Further to this, studies employing isolated myocytes from atria or ventricles (both right and left) would be also helpful, although available evidence seems to rule out the presence of p75^NTR in ventriculocytes (9, 48).

The other and likely more relevant finding is the higher incidence of premature ventricular complexes (PVCs) in p75^NTR−/− mice as well as lower heart rates found during the wake phase in these animals. The authors discovered an uneven distribution of sympathetic nerves in p75^NTR−/− mice that may account for the generation of a gradient of β₁-adrenergic receptors (β₁AR) across the LV wall. In p75^NTR-null mice, β₁ARs were significantly higher in the subendocardium than in the subepicardium. Therefore, in addition to the natural “anisotropy” of the cardiac tissue, this change in the distribution pattern may be the basis for the changes in both contractile and impulse generation/propagation in p75^NTR−/− animals. Furthermore, the denervated subendocardium of p75^NTR-null mice displayed a greater peak diastolic thickening velocity only upon stimulation with the β-agonist dobutamine compared with WT mice. Thus, some regional increase in the sensitivity to β-adrenergic signaling in face of an overall decrease in β₁AR expression may occur in p75^NTR−/− mice. As far as the possible higher propensity of p75^NTR−/− mice to develop ventricular (fatal) arrhythmias, it should be taken into account that the significance of the ectopic (ventricular) beats should be evaluated in the context in which they occur. In addition to the complexity and frequency of PVCs, the risk of developing fatal arrhythmias and sudden death, for instance, in patients that have had a myocardial infarction is related to a number of other factors, including, among others, reentrant conditions, increased sympathetic stimulation, and ongoing ischemia and reperfusion. Therefore, the present investigation paves the way for additional interesting studies in which p75^NTR−/− mice are subjected to coronary ligation and then monitored by ECG to evaluate whether they are more prone to these dreadful events.

In an attempt to start dissecting the mechanistic intricacies underlying the heart rhythm alterations seen in p75^NTR−/− mice, the authors proposed an intriguing yet not fully verified scenario commingling p75^NTR, NGF, the biosynthetic enzymes that catalyze catecholamine formation, and, ultimately, β-receptor expression in the heart. The lack of p75^NTR is known to enhance NGF-induced TrkA signaling in sympathetic neurons (23, 29), and NGF stimulates tyrosine hydroxylase (TH) expression, which is the rate-limiting step of catecholamine biosynthesis and NE release in the superior cervical ganglion (39). Here, the authors showed that NE content in p75^NTR−/− LVs is not different from that found in WT mice, but TH expression is more abundant in the LV of p75^NTR−/− mice. The authors suggested that the subepicardium has higher NE synthesis due to higher TH activity in sympathetic neurons in the subepicardium. Since high circulating (and local) NE levels are known to downregulate β-receptor density (19), this could explain the higher β₁AR levels in the denervated p75^NTR−/− endocardium that exhibited an augmented response to dobutamine. This same mechanism may underlie possible arrhythmogenesis, as most studies have suggested that adrenergic stimulation can alter, among other things, hERG currents, leading to the lengthening of repolarization and thus triggering arrhythmias (46). A complimentary hypothesis is that increased adrenoceptor activation promotes increases in Ca²⁺ currents without altering Ca²⁺ reuptake by the sarcoplasmic reticulum (2), resulting in elevated diastolic Ca²⁺, which, in turn, triggers oscillations in membrane potential and thus arrhythmias (6). The use of β-blockers in vivo will help in firmly establishing a major involvement of altered adrenergic signaling in the cardiovascular phenomena observed as a consequence of p75^NTR deletion. Again, electrophysiological and mechanical evaluations at the level of myocytes, possibly isolated from the different myocardial layers, should better define the mechanisms of mechanical and electrical abnormalities found in p75^NTR−/− hearts. Equally intriguing would be to test how all the vascular and basal/β-stimulated electromechanical alterations found in p75^NTR−/− animals conspire to affect the ability of these animals to cope with increased workload, such as during sustained exercise.

As with any exciting study, these findings raise a number of interesting molecular and signaling questions. For example, is the altered responsiveness to Sema3A responsible for the lack of subendocardial innervation? Through what other mechanisms does p75^NTR regulate cardiac sympathetic input, for instance, in the right atrium? After all, the right and left sides of the heart appear to be differently affected by p75 deletion (see the high NE content in the presence of normal sympathetic innervation density in the right ventricle). These questions combined with those raised in the preceding discussion highlight one more time the relevance of automonic, sympathetic neural remodeling in the onset and progression of acute and chronic cardiac disorders such as myocardial infarction (40), sudden cardiac death (6, 8), diabetes (43), hypertension (18, 42) and, not least, congestive heart failure (17, 49). One of the more provocative avenues for future research will be to establish whether patients affected by various cardiovascular diseases present with p75^NTR polymorphisms that correlate with such conditions. Certainly, deficiencies in p75^NTR, excesses in semaphorins, and their associated signaling may redirect the sympathetic traffic inside the heart, but it remains to be fully established whether and how this detour can cause more cardiovascular congestion and heart accidents.

GRANTS

This work was supported by National Institutes of Health Grants R01-NS-038220 (to B. D. Carter) and R011-HL-075265 and R011-HL-091923 (to N. Paolocci).

DISCLOSURES

No conflicts of interest are declared by the author(s).