Chronic elevation of SNS activity in the myocardium is a key component of the altered signaling pathways that accompany cardiac remodeling resulting from hypertension. While there is evidence that the SNS mediates cardiac fibrosis in hypertension via α-adrenergic receptor stimulation, 2
the exact mechanisms have not been fully elucidated. However, it is likely to be more complex than direct activation of α-adrenergic receptors on cardiac fibroblasts. To this end, inflammation is also an important component in the development of hypertension-induced cardiac fibrosis. 3-5,7,8,18
More specifically, we recently demonstrated a causal role for MCs in regulating myocardial cytokines, macrophage recruitment and development of fibrosis in the hypertensive heart. 6
Given the critical roles of both SNS up-regulation and MCs in hypertension-induced myocardial remodeling, the current study aimed to explore the effects of surgical sympathectomy on cardiac MCs as well as specific cytokines in the hypertensive heart.
Following sympathectomy in SHR, we observed a dramatic almost 5-fold increase in cardiac MC density. While surgical removal of the superior cervical ganglion likely does not result in complete denervation of the heart, nerve fibers from this ganglion do project to the heart 19
, and this striking change in MC density clearly demonstrates an impact of its removal on the heart. This same relationship is not present in the normal heart since such a change did not occur in sympathectomized WKY. Facoetti et al. 20
also found cardiac MC density to be unchanged in normal Sprague-Dawley rats following 10 weeks of chemical sympathectomy induced by 6-hydroxydopamine. These disparate effects on MC density are likely indicative of underlying differences in the local environment of normal versus hypertensive hearts. That fibrosis was attenuated in sympathectomized SHR despite MC density being elevated, suggests that either sympathectomy altered MCs to that of a non-fibrotic phenotype, or alternatively, MCs signal the SNS, but can no longer induce a pro-fibrotic response from the damaged SNS. Interestingly, in our isolated cardiac MC experiments, incubation with norepinephrine failed to induce degranulation, and we were unable to detect α-adrenergic receptors on cardiac mast cells using flow cytometry. This is consistent with a previous report that norepinephrine was not the mechanism by which contact with sympathetic neurons decreased membrane resistance in immortalized RBL-2H3 cells, a widely used model for MCs. 13
However, some afferent nerve fibers projecting from the heart, travel through the superior cervical ganglia, 21
and substance P, which is primarily located in afferent nerve fibers, has been demonstrated to be present in superior cervical ganglia. 22
Substance P-containing nerves are associated with numerous areas of the heart including the ventricle, atria, valves and connective linings. 23
Here we show that substance P was a strong inducer of isolated cardiac MC degranulation, consistent with what we have demonstrated previously 15
and suggestive of afferent activation of cardiac mast cells.
Suzuki et al., 24
using a co-culture system, showed that RBL-2H3 mast cells activated sympathetic neurons following IgE stimulation. Activation of sympathetic neurons did not occur if RBL-2H3 cells were not present. The most obvious mechanism for MC activation of neurons to occur is through release of MC tryptase acting on protease activated receptor-2 (PAR-2). Steinhoff et al. 25
recently detected PAR-2 on a large number of neurons containing substance P, and activation of these receptors by tryptase in dorsal horn slices initiated the release of substance P. However, cardiac MCs are also capable of producing renin and are, assumed to be involved in the local generation of angiotensin II in the heart. 12,26,27
Silver et al. 12
postulated that MC generated angiotensin II could then activate nerves through angiotensin receptors. We isolated a mixed population of inflammatory cells including lymphocytes, macrophages and MCs from rat hearts and found that incubation with substance P resulted in angiotensin II production. Thus, while in sympathectomized SHR, superficially it appears that the SNS is regulating cardiac MC density, MCs may actually be upstream of SNS activation and, therefore, responding to an underlying stimulus (e.g., increased wall stress) by unsuccessfully attempting to up-regulate SNS activity. That is, the increase in MC density may be an attempt to achieve increased SNS activity that cannot be elicited because of the sympathectomy. Therefore, we speculate, that a self-perpetuating relationship may exist, where cardiac MC tryptase could activate PAR-2 on afferent nerve fibers causing the release of substance P, which in turn induces angiotensin II production by inflammatory cells, that then acts on fibroblasts and also sympathetic nerves to release norepinephrine (). Angiotensin II has been shown to induce myocardial necrosis through stimulation of sympathetic neurons. 28,29
In fact, a similar situation has been described in the rat knee joint, where it appears that sensory nerve activation of mast cells led to stimulation of the SNS and subsequently plasma extravasion. 30
Elaboration of norepinephrine by the SNS in response to stimulation by MCs would explain how α-adrenergic receptors mediate fibrosis 2
when cardiac MCs do not degranulate in response to norepinephrine in our study. This may also explain why sympathectomy does not result in increased cardiac MC density in the normal heart where there is no underlying pathologic stimulus driving MCs to up-regulate sympathetic activity.
Figure 5 Schematic depicting a possible series of events in the development of fibrosis in the hypertensive heart. Cardiac mast cell tryptase may activate protease activated receptor-2 (PAR-2) on afferent nerve fibers resulting in the release of substance P, which (more ...)
With this clear SNS-MC interaction, we sought to determine the effect of sympathectomy on selected myocardial cytokines that we have previously shown to be regulated, either directly or indirectly, by MCs in the hypertensive heart. 6
Both IFN-γ and IL-4 were elevated in the SHR, with sympathectomy normalizing IFN-γ, but not IL-4. IFN-γ is a marker of TH1
T cell phenotype 31
and its elevation in hypertension is consistent with the supposition advanced by Yu et al. 9
cells may be pro-fibrotic. Interestingly, while sympathectomy had no effect on IFN-γ in the WKY, sympathectomy increased IL-4 levels in the WKY. Combined with our previous studies, 6
it would appear that, in the normal heart, the SNS has a role in regulating IL-4, but that role is shifted to MCs in the hypertensive heart. Although the levels of IL-6 and IL-10 were lower in the SHR myocardium compared to the WKY, sympathectomy returned them to normal. What is intriguing about these results is that, with the exception of IL-4, these changes are almost identical to what we have observed previously in the SHR following MC stabilization. 6
This similarity, together with the close spatial relationship between the nervous system and MCs, 11,12
is more evidence of an interplay between the SNS and MCs in the regulation of the inflammatory process in the hypertensive heart.
In addition to attenuating cardiac fibrosis, sympathectomy in the SHR also prevented myocardial hypertrophy. However, our previous finding that MC stabilization did not prevent hypertrophy in the hypertensive heart indicates that MCs are not involved in this modulation of hypertrophy.6
Instead, rather than acting via inflammatory cells to induce hypertrophy, the SNS most likely has direct effects on cardiomyocytes via β-adrenergic receptors. While it cannot be ruled out that the changes in these parameters and the assayed cytokines were the result of the sympathectomy-related attenuation of blood pressure, we believe the following considerations support the argument that the changes in cytokines and fibrosis were due to direct interactions between the SNS and MCs. Firstly, attenuation of hypertrophy by sympathectomy has been shown to be independent of systolic blood pressure, 2,32
and Perlini et al. 2
further demonstrated that chemical sympathectomy in hypertensive rats prevented cardiac fibrosis independent of blood pressure. Secondly, cardiac MC density was dramatically with sympathectomy in the SHR but not in the normal non-hypertensive heart. Thirdly, our in vitro
studies demonstrate that substance P, found in afferent nerves, can stimulate cardiac MC activation and angiotensin II production from a mixed population of cardiac inflammatory cells. Finally, the cytokine profile reported herein following sympathectomy in the SHR is almost identical to that following MC stabilization in the SHR where blood pressure was unaffected. 6
The results of this study reiterate the importance of the SNS in the development of myocardial remodeling and provide evidence of a link between the neurohormonal and inflammatory paradigms of hypertension, by demonstrating the importance of SNS interactions with cardiac MCs and cytokine production. Furthermore, even though a number of inflammatory cell types have been implicated in the development of fibrosis, 7,9,10
our findings are consistent with the hypothesis that MCs not only trigger and orchestrate inflammatory responses, 33
but are also responsible for cardiac fibrosis. It is important to note that not all of the cardiac remodeling effects of the SNS are mediated via mast cells. Clearly hypertrophy does not appear to involve SNS interactions with mast cells, but instead is likely the result of direct stimulation of cardiomyocyte β-adrenergic receptors. The findings herein indicate the need for further investigation regarding the mechanisms mediating SNS, inflammatory cell and afferent nerve interactions in the hypertensive heart, with particular focus on the role of substance P.