| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
REVIEW ARTICLES |
From the Department of Psychology, University of Pittsburgh (K.F.H.), and the Departments of Psychiatry, Epidemiology, and Psychology, University of Pittsburgh School of Medicine, (K.A.M.), Pittsburgh, PA.
Address reprint requests to: Karen A. Matthews, PhD, Cardiovascular Behavioral Medicine, Department of Psychiatry, WPIC, 3811 O’Hara Street, Pittsburgh, PA 15213. E-mail: matthewska{at}umpc.edu
ABSTRACT
OBJECTIVES: Endothelial dysfunction is a new pathway in cardiovascular disease (CVD) development. Psychosocial factors have been little studied in relation to endothelial function, although they may interact via associations with the autonomic nervous system (ANS). The purpose of this review is to propose a model by which psychosocial factors are related to CVD development through interactions between the ANS and vascular endothelium.
METHODS: The literature supporting an interaction between the ANS and endothelium in healthy and disease states is reviewed. Potential mechanisms linking the two systems are explored as a pathway for CVD development.
RESULTS: Endothelial dysfunction and impaired cardiovascular ANS regulation are both markers for increased CVD risk. Sympathetic nerves and vascular endothelial cells share a functional antagonism in healthy states to maintain appropriate blood vessel tone. Alterations in sympathetic activity and endothelial cell function are both observed early in the development of CVD and may result from an inability to maintain the functional antagonism. Impairments in either ANS regulation or endothelial function may contribute to further disease development by evoking maladaptive changes in the opposing system.
CONCLUSIONS: Although interactions between cardiovascular ANS regulation and endothelial function are likely involved in CVD development, further research is needed to determine whether ANS and endothelium interactions are a plausible pathway linking psychosocial factors with increased CVD risk.
Key Words: endothelial function, • autonomic nervous system, • sympathetic nervous system, • pathophysiology, • cardiovascular disease, • psychosocial stress.
Abbreviations: CVD = cardiovascular disease;; ANS = autonomic nervous system;; SNS = sympathetic nervous system;; PSNS = parasympathetic nervous system;; NO = nitric oxide;; CAD = coronary artery disease;; FMD = flow-mediated dilation;; HRV = heart rate variability;; vWF = von Willebrand factor;; LDL = low-density lipoprotein;; NOS = nitric oxide synthase;; HPA = hypothalamic-pituitary-adrenal.
INTRODUCTION
Cardiovascular disease (CVD) development is a complicated process. Fortunately, new technologies are providing fresh insights on the mechanisms of CVD pathogenesis. Endothelial function is one new measure that is broadening our understanding of CVD development. Endothelial cells play an important role in maintaining the structural and functional integrity of the vasculature. Inability of the endothelial cells to stimulate vasodilation properly is referred to as endothelial dysfunction. Endothelial dysfunction is believed to be one of the earliest stages of atherosclerosis and can be observed in healthy people with risk factors for heart disease (1). Endothelial dysfunction is correlated with subclinical measures of CVD and is prospectively associated with an increased risk for clinical CVD events (2).
Interactions between autonomic nervous system (ANS) regulation and endothelial function may provide a testable model for exploring a novel mechanism relating psychosocial factors to CVD development (Figure 1). Studies on psychosocial stress and negative psychosocial traits demonstrate associations with heightened sympathetic nervous system (SNS) activity (3, 4). Recent studies have suggested that similar psychosocial factors may also be linked with suppressed parasympathetic nervous system (PSNS) activity (5). Heightened SNS activity and suppressed PSNS activity impair the ability of the ANS to regulate the cardiovascular system (6). Impaired ANS regulation may be linked with CVD development through impairments in endothelial function. The endothelium shares a functional antagonism with SNS efferents in maintaining blood vessel tone. Research on the interrelationship between these two systems is growing. Interactions between the SNS and endothelial function provide an exciting new pathway for exploring mechanisms linking psychosocial factors with CVD.
|
Various methods for measuring endothelial function (7, 8) and ANS activity (9, 10) have been discussed in detail elsewhere. In this review, studies examining endothelial function in humans were selected for inclusion based on adherence to standard procedures. Included studies primarily focused on examining vasodilatory responses in either the coronary or brachial artery. Brachial artery responses were measured in response to flow-mediated dilation because this measure is closely correlated with coronary artery endothelial function (11). Coronary endothelial function was primarily measured in response to acetylcholine or sympathetic stimulation via the cold pressor task. Variations of standard methodology are noted in the text where appropriate.
WHAT IS THE VASCULAR ENDOTHELIUM?
The endothelium is a single layer of cells that line the blood vessel lumen. These cells regulate vascular tone by responding to signals in the blood to release either vasodilating or vasoconstricting factors. Endothelial function is usually measured by exposing blood vessels to an endothelium-dependent vasodilating stimulus and then measuring the extent of vasodilation in response to the stimulus. Greater vasodilation reflects better endothelial function.
The primary vasodilator released by the endothelium is nitric oxide (NO) (12). Other relaxing factors released by the endothelium include endothelium-derived hyperpolarizing factor, prostacyclin, C-type natriuretic factor, 5-hydroxytryptamine serotonin (5-HT), adenosine triphosphate (ATP), substance P, and acetylcholine (13, 14). Basal blood flow maintains a continual release of endothelium-derived relaxing factors, and an increase in blood flow increases the release of relaxing factors (15). Increased release of NO from the endothelium results from changes in the shear stress on the vascular wall (16).
The endothelium also releases contracting factors, such as endothelin-1, angiotensin II, thromboxane A2, prostaglandin H2, superoxide anions, and ATP (13, 14). Stimuli that initiate release of these endothelium-dependent vasoconstrictors include norepinephrine, thrombin, hypoxia, and stretch (13). Normally, endothelin levels in the plasma are very low, likely because it is primarily released abluminally (17). Endothelin causes concentration-dependent arterial contractions that increase in magnitude in the absence of a functional endothelium (18). Higher levels of endothelin have been reported in some disease states, such as hypertension, although the role endothelin plays in these states is unclear (17). NO can eliminate endothelin-induced arterial constriction and inhibit further release of endothelin from the endothelium (14, 18).
Endothelial cells throughout the body are heterogeneous in function and may operate under different influences depending on the vascular bed (19). This is important to remember when comparing results of functional studies conducted on different vascular beds. Different stimuli also use different pathways of signal transduction to promote the release of vasodilating substrates (20). These multiple pathways may help to explain stimulus-specific impairments in endothelium-dependent vasodilator function.
IS ENDOTHELIAL FUNCTION RELATED TO RISK FOR CARDIOVASCULAR DISEASE?
Endothelial function is important because of its role in CVD development. In addition to regulating vessel tone, endothelial cells help to prevent the build-up of lipids and platelets that initiate the atherosclerotic process. Additionally, endothelial dysfunction may contribute to ischemic or arrhythmic events by preventing needy cells, such as those in the heart, from receiving the blood supply and nutrients required for proper functioning.
The endothelium releases a variety of substrates that play an important role in the inhibition of atherogenesis. Both NO and prostacyclin inhibit platelet activation and aggregation, whereas the release of tissue plasminogen activator enhances fibrinolysis (21). Other anticoagulant factors, such as thrombomodulin, are presented on endothelial cell surfaces to prevent cellular adhesion to the vascular wall (21). All of these mechanisms help to prevent the build-up of atherosclerotic plaque along arterial walls.
Inflammation is widely thought to play a role in the cardiovascular disease process. Damage to the endothelium can enhance inflammatory responses through the presentation of chemoattractants that promote cellular adhesion and uptake of lipids and macrophages into the abluminal space (22). Damage to the endothelium also results in the release of growth factors, such as platelet-derived growth factor, which can promote connective tissue growth and enhance smooth muscle cell proliferation, resulting in vascular wall thickening (22, 23). Loss of atheroprotective influences and activation of thrombosis, inflammation, and vascular growth all emphasize the negative effect of endothelial dysfunction on vascular health.
Numerous studies have linked impairments in endothelial function to CVD. When exposed to an endothelium-dependent vasodilator, such as acetylcholine, normal coronary arteries dilate, whereas irregular or stenosed vessels constrict (24). A similar response pattern has been demonstrated in the coronary arteries of patients with coronary artery disease (CAD) in response to the cold pressor test (25, 26). Even smooth coronary arterial segments of patients with CAD may constrict when exposed to cold (26).
Deficits in endothelial function can be evidenced in people with risk factors for CVD even in the absence of overt CVD. Among angiography patients with smooth coronary arteries, coronary diameter change in response to acetylcholine is lower in those who are older, are hypertensive, are male, have high cholesterol, or have a family history of early CVD (27, 28). Antony et al. (29) examined coronary vasomotor responses to the cold pressor test in two groups of angiography patients without evident CVD: patients with recently diagnosed hypertension and patients without risk factors for CVD. They found that all vessels in patients without hypertension dilated on exposure to cold, whereas all vessels in patients with hypertension either showed no change or constricted on exposure to cold (29). This pattern has also been demonstrated in the carotid artery: whereas carotid arteries from patients with CAD constrict, arteries from high-risk patients without CAD demonstrate mixed responses, and arteries from average-risk controls dilate in response to the cold pressor task (30).
Zeiher et al. (31) examined coronary diameter change with different endothelium-dependent vasodilators among angiography patients with different stages of disease . Patients with normal arteries and low risk for CAD dilated in response to all endothelium-dependent stimuli, including acetylcholine, cold exposure, and increased blood flow. Patients with normal arteries and high cholesterol dilated in response to cold exposure and increased blood flow but constricted in response to acetylcholine. Normal coronary arteries of patients with CAD in other vessels dilated in response to increased blood flow, whereas cold exposure or acetylcholine resulted in vasoconstriction. Abnormal vessel segments in patients with CAD constricted during cold exposure or in response to acetylcholine and failed to dilate in response to increased blood flow. These findings suggest that a progressive impairment of endothelial function is evident with progressive stages of CAD (31).
Endothelial function can be measured in the brachial artery of healthy samples using ultrasonography. Brachial artery endothelial function correlates well with the presence of CAD (11, 32, 33). Impaired brachial vasodilation in response to increased blood flow is evident in patients with coronary heart disease and patients with greater carotid artery intima-media thickness (32, 34). Healthy people with CVD risk factors have impaired brachial endothelial function even in the absence of clinical CVD. Celermajer et al. (1) showed that flow-mediated dilation (FMD) of either the brachial or femoral artery is impaired in adult smokers, children with hypercholesterolemia, and patients with CAD. Additional risk factors, such as high blood pressure, family history of early CVD, older age, male gender, and a higher composite risk factor score, are related to impaired FMD among apparently healthy people (35).
Prospective studies of patients with CAD show that impaired endothelial function predicts subsequent cardiac events (2, 36–38) . Additionally, non-CAD angiography patients who have abnormal coronary responses to acetylcholine are at an increased risk for subsequent CVD events (36). Perticone et al. (39) also showed that among hypertensive patients without known CVD, smaller increases in forearm blood flow in response to acetylcholine, a marker of microvascular endothelial function, predict subsequent clinical cardiovascular events. There are currently no studies examining the predictive value of endothelial function in an initially healthy sample.
IS ENDOTHELIAL FUNCTION RELATED TO PSYCHOSOCIAL FACTORS?
Few studies have examined endothelial function in relation to psychosocial traits. Harris et al. (40) tested correlations between psychosocial risk factors for CVD and endothelial function among a sample of postmenopausal women. They reported that women with higher anxiety scores had less FMD than women with lower anxiety scores. Among women not using any hormone replacement, higher scores on Type A behavior, trait anger, and depression were also associated with less FMD, indicating poorer endothelial function (40). However, because no endothelium-independent vasodilator was used in this study, impairment in vasodilatory function at the level of the smooth muscle cannot be ruled out. A second study demonstrated that patients with current major depression also demonstrated impairments in FMD relative to nondepressed controls (41). These studies suggest that psychosocial risk factors for CVD are associated with impairments in endothelial function.
Stress reactivity protocols have been used as stimuli for measuring endothelial function. As mentioned, the cold pressor test produces coronary vasomotor responses similar to those observed after administration of acetylcholine and increases in blood flow (26). Other stress reactivity protocols have also contributed to abnormal vasomotor responses in the coronary artery. Mental arithmetic results in a similar pattern of differential responses in diseased and healthy coronary vessels, where healthy vessels dilate and diseased vessels constrict during stress (42). High levels of experienced anger reported during an anger recall interview were associated with coronary vasoconstriction in narrowed segments of patients with CAD (43). Mental stress induced by performing a public speaking task during cardiac catheterization resulted in coronary vasoconstriction in CAD and non-CAD patients alike (44). It is important to point out that endothelial function was not specifically tested in these studies of mental stress. Therefore, although these studies suggest that vascular changes during mental stress may be used to measure endothelial function, more specific studies on endothelial function responses to mental stress are needed.
Relationships between psychosocial stress and endothelial function have also been examined in healthy people using brachial artery ultrasonography. Healthy men who performed a speech stress task for 5 minutes had lower brachial FMD responses at 30 and 90 minutes after the task relative to FMD responses before the task (45). This finding suggests that brief exposure to mental stress may induce temporary declines in endothelial function that remain after other physiological concomitants of stress exposure, such as alterations in blood flow and pressure, have returned to normal.
The mechanisms responsible for the effects of psychological stress on the endothelium have not been completely elucidated. Possible explanations for the endothelium-dependent dilatory response to stress include activation of 
2-adrenergic receptors on the endothelium resulting in release of NO or hemodynamic alterations leading to endothelium-dependent FMD. In one sample of patients undergoing diagnostic cardiac catheterization, ß-blockade with propanolol only slightly attenuated the vasodilating response to the cold pressor test (25). In another similar study, propanolol administration failed to alter the vasomotor response of coronary arteries to the cold pressor test (26). However, both studies reported significant increases in heart rate and blood pressure with exposure to cold (25, 26), suggesting that changes in blood flow hemodynamics may stimulate endothelium-dependent vasodilation during cold stress.
The relationship between stress-induced sympathetic stimulation and endothelial function suggests that psychosocial factors may be related to endothelial function via ANS pathways. Explanations of associations between psychosocial factors and CVD often involve differences in ANS regulation (46), especially heightened SNS activity (47). Psychosocial factors, such as stress, anger, and hostility, have been linked with greater sympathetic and lower parasympathetic activity, suggestive of imbalances in ANS regulation (3, 5). To assess potential connective links between psychosocial factors and endothelial function via ANS regulation, it is first important to understand the relationship between the ANS and the endothelium.
HOW DO THE ENDOTHELIUM AND THE AUTONOMIC NERVOUS SYSTEM INTERACT? A MODEL FOR CARDIOVASCULAR DISEASE RISK
Healthy State
In a normal state, the ANS and the endothelium work together to maintain vascular tone. There is a tonic balance between the release of vasodilating factors from the endothelium and vasoconstricting factors from sympathetic nerve terminals (Figure 2). The balance between these opposing forces acts on the vascular smooth muscle cells to maintain the appropriate vessel tone (48).
|
1-adrenergic, and to a lesser extent, 2-adrenergic receptors on smooth muscle cells contributes to smooth muscle cell contraction, resulting in vasoconstriction (49–51). Parasympathetic stimulation of muscarinic receptors on smooth muscle cells also results in contraction and therefore vasoconstriction (52). In contrast, release of vasodilating factors, such as NO, from the endothelium stimulates smooth muscle cell relaxation, resulting in vasodilation.
The ANS may directly influence the endothelium. Endothelial cells possess both 2-adrenoceptors and ß-adrenoceptors (49, 53). In the coronary and large conduit arteries, activation of endothelial 2-adrenergic receptors releases NO, resulting in vasodilation (20, 49, 54). This endothelium-dependent vasodilatory effect of 2-adrenergic receptor activation counteracts the vasoconstricting effect of 1-adrenergic receptor stimulation of the vascular smooth muscle (50, 53). The function of ß-adrenoceptors on the endothelium is unclear, although a possible role in endothelium-dependent NO-mediated vasorelaxation has been speculated (49). In addition to vasodilators, the SNS can stimulate the release of endothelium-derived contracting factors, such as endothelin (48, 55). Endothelial cells also possess receptors for the primary PSNS neurotransmitter, acetylcholine. Activation of these receptors produces endothelium-dependent relaxation of human coronary arteries (52).
Although endothelial cells in the microvasculature may receive ANS inputs, endothelial cells along the major conduit vessels do not receive direct neural innervation from the ANS (48). Therefore, the effects of neurotransmitters on endothelial function must be exerted by circulating levels or by diffusing through the smooth muscle cell layer without degradation. Some evidence suggests that ANS neurotransmitters are able to diffuse across the vascular smooth muscle cells to act on the endothelium (14). Animal studies show that alterations in endothelial function occur after ANS denervation (48). In isolated rat mesenteric vessels, vasomotor responses to electrical stimulation of perivascular nerves are similar to responses after perfusion of the lumen with norepinephrine (16), suggesting that ANS efferents can directly influence endothelial cells. In contrast, others suggest that endothelial cells respond to changes in smooth muscle tone after sympathetic stimuli and not direct neural stimulation. When the vasoconstriction after sympathetic stimulation is blocked, relaxing factors are not released from endothelium, despite the presence of SNS-derived substrates (56). This suggests that sympathetic nerves do not directly increase release of endothelium-derived relaxing factors. Instead, vasoconstriction resulting from the release of norepinephrine from sympathetic nerve terminals may stimulate the increases in endothelium-derived NO (56).
The endothelium may affect the extent to which the ANS regulates vascular tone. Basal release of NO attenuates sympathetic vasoconstriction (51). In the presence of arterial shear stress, the endothelium can further depress vasoconstrictive responses to sympathetic nerve norepinephrine release in rabbit carotid arteries, likely because of increased release of NO (57). The endothelium may also inhibit SNS activity by contributing to the metabolic breakdown of norepinephrine and by providing a physical barrier to prevent diffusion of the neurotransmitter into the bloodstream (56, 58) . NO released from the endothelium decreases the sensitivity of smooth muscle cells to the vasoconstrictor effects of the SNS and inhibits central and peripheral SNS activity (14, 18, 59)
Other endothelial substrates also influence ANS function. In rat mesenteric arteries, low levels of endothelin inhibit norepinephrine release from sympathetic nerve terminals, resulting in a slightly lower arterial pressure (60, 61). However, at higher levels, endothelin can increase the sensitivity of smooth muscle cells to norepinephrine, resulting in increases in arterial pressure (60, 61). Therefore, endothelin at normal levels may suppress SNS-induced vasoconstriction, whereas excess endothelin release may enhance vasoconstrictive responses to SNS stimulation. In support of this, endothelin can enhance norepinephrine-induced vasoconstriction in isolated human mammary arteries (18). However, caution must always be exercised when interpreting studies using animals or isolated vessel segments because the results may not carry through to intact human samples.
In general, the endothelium and ANS act in opposition to maintain the appropriate vessel tone. This tone is a reflection of local and systemic factors. To serve both purposes, the endothelium may be more responsive to local factors, whereas the ANS may be more responsible for integrating systemic factors (62). In this way, the endothelium and ANS work together to create an ideal balance between the needs of local tissues and those of other systems.
What Happens When Things Go Wrong?
What Are the Direct Links Between the Autonomic Nervous System and Endothelial Function?
Some studies show correlations between impaired endothelial function and greater resting systolic blood pressure, diastolic blood pressure, and heart rate (30, 69), although these correlations are not consistent (30, 38, 69). Few studies have directly examined correlations between cardiovascular changes to stress and endothelial function. One study showed that impaired femoral artery FMD was related to greater systolic blood pressure reactivity to physical challenges in children (70). In another study, impaired brachial FMD correlated with greater increases in vascular resistance during mental stress in adults (69). This finding raises the possibility that cardiovascular reactivity may be a crude marker of endothelial function.
Indirect evidence suggests a potential association between heart rate variability (HRV) and endothelial function. Decreased HRV correlates with greater arterial stiffness among patients with diabetes (71). Endothelin levels correlate negatively with some measures of HRV among patients with congestive heart failure (72). Among patients with diabetes, lower HRV is associated with higher levels of von Willebrand factor (vWF), an indirect marker of endothelial function (73).
Can Impaired Autonomic Function Contribute to Endothelial Dysfunction?
Some of the best evidence for a negative effect of SNS activity on endothelial function comes from studies examining psychosocial stress in nonhuman primates. The psychosocial stress is induced by exposing monkeys to a novel social environment (ie, the test monkey is placed in a cage with several unfamiliar monkeys). This stimulus is a very potent stressor because the test monkey is forced to establish a new position along the dominance hierarchy through repeated aggressive interactions with other monkeys. Monkeys exposed to unstable social environments demonstrated coronary vasoconstriction to acetylcholine, indicating that endothelial function was impaired (77). After psychosocial stress, monkeys treated with saline demonstrated more endothelial cell death and replication than monkeys treated with ß-blockade (78). These results indicate that SNS activation may contribute to impaired endothelial function, possibly because of activation of ß-adrenergic receptors.
As noted, mental stress may be linked with reduced endothelial function. Studies in nonhuman primates suggest that ß-adrenergic receptor blockade may protect endothelial cells from the negative effect of heightened SNS activity. However, do the beneficial effects of ß-blockade occur in humans as well? Hypertensive men treated for 22 weeks with either the ß-adrenergic receptor blocker atenolol or the
Additionally,
Can Endothelial Dysfunction Contribute to Impairments in the Autonomic Nervous System?
Inability of the endothelium to counteract sympathetic vasoconstricting factors may contribute to mental stress-induced ischemia (88). Inasmuch as impaired endothelial function contributes to ischemic episodes, it may also exert an influence on neuronal reuptake of norepinephrine. Sobey et al. (89) demonstrated that ischemia followed by reperfusion of rabbit thoracic aorta enhanced sensitivity to norepinephrine because of an impaired ability of sympathetic neurons to reuptake norepinephrine. The direct effect of endothelial dysfunction on neuronal reuptake requires further investigation.
Endothelial dysfunction enhances the contractile effect of catecholamines. Sympathetic nerve stimulation contributed to greater carotid artery constriction in rabbits fed a high cholesterol diet than in rabbits fed a normal diet, and removal of the endothelium eliminated this difference (90). This result suggests that enhanced constriction to sympathetic stimulation in high-cholesterol rabbits is caused by endothelial dysfunction. Indeed, removal of the endothelium from rabbit carotid arteries increased norepinephrine-induced smooth muscle cell contraction (54). Similar findings have been reported in humans. Angiography patients with coronary endothelial dysfunction demonstrate enhanced coronary vasoconstriction to sympathetic stimulation with the
Only one study has prospectively examined the effect of endothelial dysfunction on ANS activity in humans. In recently diagnosed patients with diabetes followed for 3 years, higher vWF levels predicted a subgroup of patients with diabetes who subsequently developed deficits in lower limb nerve conduction velocity (91). This finding suggests that endothelial dysfunction may predispose patients with diabetes to impairments in peripheral neural conduction. Because none of the patients in this sample developed deficits in ANS function, further follow-up is expected to clarify whether elevated vWF will similarly predict future autonomic neuropathy (91).
What Are the Indirect Links Between the Autonomic Nervous System and Endothelial Function?
Autonomic regulation also differs between premenopausal and postmenopausal women (92). Some suggest that estrogen may influence ANS function by upregulating PSNS activity and downregulating SNS activity (92). However, studies examining the influence of estrogens on cardiovascular responses to stress have produced mixed results, with some showing a protective effect (99) and some showing no effect of estrogen (100).
Most studies show a positive influence of estrogen on the vascular endothelium, and some suggest that estrogen helps to regulate ANS function. Sex hormones are important modulators of vascular activity and may provide one link between alterations in ANS and endothelial function. The role of sex hormones as a communicative link between the ANS and endothelium has not yet been explored.
Oxidative Stress
Oxidative stress also contributes to impairments in ANS function. Oxidation induces neuronal cell death, including apoptosis of SNS neurons (106). Peroxynitrite, formed from the interaction of superoxide anion with NO, is a principal oxidizing agent of the ANS (107). Administration of the antioxidant vitamin E to patients with type II diabetes mellitus reduced circulating epinephrine and norepinephrine levels and improved HRV (108). Overall, changes in ANS function with vitamin E supplementation reflect an upregulation of the PSNS with a concomitant downregulation of the SNS, possibly by attenuating the negative effect of oxidative stress on ANS function (108).
Platelet Activation
Renin-Angiotensin System
Hypothalamic-Pituitary-Adrenal Axis
Insulin Resistance
Aging
DIRECTIONS FOR FUTURE RESEARCH
Interactions between ANS regulation and endothelial function are evident in healthy and diseased states. For this reason, studies exploring vascular reactivity need to consider both the ANS and endothelium in interpreting physiological responses to physical and psychosocial stress. Currently, much is unknown about the effect of alterations in one of these systems on the other. Evidence suggests that abnormalities in either ANS or endothelial function may negatively affect the function of the other system. However, this hypothesis has not been directly tested in humans. Future research should focus on uncovering both the effect of ANS dysregulation on endothelial function and the effect of endothelial dysfunction on ANS regulation.
Exploring the relationship between ANS and endothelial function provides a testable model for examining processes involved in CVD development. ANS regulation is related to CVD survival and predicts CVD incidence in healthy samples. Endothelial dysfunction is one of the earliest phases of atherosclerosis, and endothelial function can be measured noninvasively in healthy people before the onset of disease. For these reasons, examining endothelial function is useful for determining factors that influence the early development of CVD. Therefore, examining associations between the ANS and vascular endothelium may provide insight on the nature of CVD development and progression. However, many questions remain regarding interactions between ANS and endothelial function as a pathway of CVD development. First, can endothelial dysfunction predict CVD among an initially healthy sample? Are common measures of ANS regulation, such as pre-ejection period, muscle sympathetic nerve activity, and HRV, correlated with endothelial function? Do other third party factors associated with CVD risk mediate associations between ANS and endothelial dysfunction? Future research should address these questions to gain a better understanding of the role ANS and endothelial cell interactions play in the development of CVD.
As mentioned, interactions between ANS regulation and endothelial function may provide a testable model for exploring a novel mechanism relating psychosocial factors to CVD development (Figure 1). Psychosocial factors have been rarely studied in relation to endothelial function, providing a wide-open arena for research on this new pathway of disease development. Psychosocial factors are highly associated with measures of cardiovascular ANS regulation (4). Psychosocial stress and negative psychosocial traits may therefore predispose some people to patterns of ANS regulation that are detrimental to endothelial function. In this way, psychosocial risk factors for CVD may play a role in the disease process through the negative effect of ANS regulation on the endothelium. This hypothesis remains to be tested.
CONCLUSIONS
In summary, alterations in cardiovascular ANS regulation and endothelial function are highly implicated in the development of CVD. In normal arteries, the ANS and endothelium share a functional antagonism that maintains vascular tone. However, alterations in the balance between these two systems may lead to detrimental effects on the cardiovascular system. Overactivation of the ANS may directly impair endothelial function. Similarly, endothelial dysfunction may contribute to abnormalities in ANS regulation through alterations in neurotransmitter release, reuptake, or receptor sensitivity. Other factors, including sex hormones, oxidative stress, and many others, may also partially explain the co-occurrence of abnormalities in ANS regulation and endothelial function. Associations between impairments in ANS regulation and endothelial function provide a mechanistic link by which psychosocial factors may be related to the development of CVD.
Although there is much evidence linking the ANS and endothelium in healthy and diseased states, there are many questions remaining about the physiological processes involved in these links. Gaining a better understanding of this physiological relationship will provide new and exciting opportunities for determining the role of psychosocial processes in CVD development. Currently, there is much conflict within the literature on psychosocial factors and CVD. Further understanding the physiological pathways of psychosocial factors in disease can help bring clarity to this perplexing puzzle.
ACKNOWLEDGMENTS
Funding for this research was supported by the Pittsburgh Mind-Body Center Research Training Program HL07560.
Received for publication June 10, 2003.
REFERENCES
Disease states or negative lifestyle characteristics may predispose susceptible people to impairments in ANS regulation or endothelial function. Although ANS dysregulation and endothelial dysfunction co-occur, it is difficult to determine the driving force behind the associations. Impairments in ANS regulation may contribute to abnormal changes in endothelial cells, resulting in endothelial dysfunction, or impairments in endothelial function may lead to maladaptive alterations in ANS regulation. Additionally, many other factors are related to both ANS and endothelial function. These third party factors may link alterations in ANS and endothelial function. It is important to identify which factors are responsible for alterations in ANS and endothelial function to understand better the role these factors play in the development of CVD.
Correlations Between Impairments in the Autonomic Nervous System and Endothelial Function
Evidence for potential correlations between ANS and endothelial function comes from disease states associated with impairments in both systems. In addition to CVD, diabetes, hypertension, and congestive heart failure are associated with abnormalities in ANS regulation of the cardiovascular system (63–65). All of these disease states are similarly associated with impairments in endothelial function (66–68). It is unclear whether the ANS and endothelial systems are negatively affecting one another, or whether both systems undergo dysregulation as a consequence of the disease process.
Currently, the best evidence of alterations in endothelial function resulting from ANS-related mechanisms comes from experimental data. Exaggerated SNS activity may impair endothelial function and enhance endothelium-mediated atherogenic processes. Chronic stimulation of perivascular nerves in the rabbit ear artery contributes to structural changes in the endothelial cells along the arterial wall (74). Endothelial cells exposed to chronic stimulation by perivascular nerves express greater immunoreactivity (74). High levels of circulating catecholamines upregulate the interferon-
–induced expression of class II major histocompatibility complex (MHC) receptors in bovine cerebral capillary endothelial cells, indicating that catecholamines may contribute to the induction of macrophages into the abluminal space (75). Circulating catecholamines may also increase the uptake of low-density lipoproteins (LDLs) by endothelial cells (76). -adrenergic receptor blocker doxazosin had lower levels of circulating vWF after treatment than before treatment (79). The men treated with atenolol also demonstrated declines in circulating endothelin (79). Treatment with the ß-blocker propanolol may also improve vWF antigen levels among patients with hyperthyroidism (80). Treatment with the ß-blocker carvedilol for 4 months improved brachial FMD among patients with CAD (81). Studies examining the effects of ß-adrenergic receptor blockers on resistance vessel endothelial function have generally failed to show a beneficial effect (82–84). However, different types of ß-adrenergic receptor blockers may influence resistance vessel endothelial function in different ways (85). These studies indicate that although the effects of ß-adrenergic receptor blockade on resistance vessels are unclear, ß-blockers may improve deficits in endothelial function among the larger conduit vessels. -receptors may also be an important link between SNS activity and impaired endothelial function. Hijmering et al. (86) measured FMD in healthy people at rest and while undergoing sympathetic stimulation using lower body negative pressure. Sympathetic stimulation resulted in a significant reduction in FMD that was abolished in subjects who underwent -receptor blockade before the SNS stimulus (86).
Impaired endothelial function may influence ANS activity through alterations in neurotransmitter release, reuptake, or receptor sensitivity. Removal of the endothelium increased the release of norepinephrine from sympathetic nerve terminals in rabbit carotid arteries (54). Alterations in neurotransmitter release from SNS nerve terminals may result from decreased NO release. Among 15 healthy men, reduced NO release caused by the inhibition of nitric oxide synthase (NOS), a catalyst for NO production, enhanced muscle sympathetic nerve activity, suggesting that basal release of NO inhibits SNS activity (87). -adrenergic agonist phenylephrine (27). This enhanced vasoconstriction occurs in the absence of any changes in heart rate, blood pressure, or blood flow (27). Therefore, impaired endothelial function increases the contractile effects of SNS stimulation and may increase neurotransmitter release while decreasing reuptake.
In addition to the direct influence of ANS regulation and endothelial function on one another, other factors may indirectly link deficits in these two systems (Figure 3). Each of these third party factors shows some evidence of being linked with both ANS regulation and endothelial function. It is currently unclear whether these factors may also mediate associations between impairments in the ANS and endothelium.
View larger version (16K):
[in a new window]
Figure 3. Indirect associations between autonomic nervous system regulation and endothelial function.
Reproductive Axis
Women and men differ in relation to ANS regulation and endothelial function (35, 92–95). Gender differences in vascular regulation may result from differences in sex hormones. Estrogen stimulates the release of NO from the endothelium by increasing NOS and enhancing NOS activity (93, 96). Estrogen is also associated with lower levels of endothelin and decreased sensitivity to the vasoconstrictive effects of the peptide (93). Premenopausal women and postmenopausal women using hormone replacement have greater brachial artery FMD than postmenopausal women not using hormone replacement therapy (97, 98).
Evidence exists for a negative influence of oxidative stress on both ANS and endothelial function. Oxidative stress may impair endothelial function by two mechanisms: free radicals, such as superoxide anion, and lipid peroxidation (101). Oxygen radicals can inhibit NO by combining with it to form peroxynitrite (101). Oxidized LDL inactivates NO and NOS, increases the release of oxidized anions from the endothelium, and is directly cytotoxic to endothelial cells (101, 102). Oxidative stress promotes the release of endothelin from the endothelium and can enhance endothelin activity (103). Exposure to oxidants causes morphological changes in endothelial cells, and these changes decrease the ability of the endothelium to provide a barrier between the blood and subendothelial matrix (104). Oxidative stress promotes leukocyte adhesion to endothelial cells (104). The combination of increased leukocyte adhesion and increased permeability of the endothelial layer implicates oxidative stress as an important event in the initiation of the atherosclerotic process (104). Indeed, in comparing asymptomatic patients with moderate carotid artery stenosis to healthy controls, the patients had higher levels of oxidized LDL, greater evidence of oxidative stress, and more circulating endothelin than controls (105).
Impaired ANS regulation is associated with greater platelet activation, contributing to enhanced aggregation and adhesion to vessel walls (109). In contrast, endothelial cells may act in opposition by inhibiting platelet activation, adhesion, and aggregation (110). Enhanced SNS activity may influence the interplay between the endothelium and circulating platelets, enabling platelet aggregation to commence at lower than normal levels of shear stress (111). Platelet activation can also negatively affect endothelial cell function (110, 112). Heightened SNS activity may therefore contribute to deficits in endothelial function because of enhanced platelet activation.
The connection between ANS regulation and the renin-angiotensin system is evident in healthy people and patients with CAD (113, 114) . Angiotensin II is a powerful vasoconstrictor synthesized by angiotensin-converting enzyme and may contribute to activation of central SNS processes and suppression of PSNS activity (115, 116)
Sympathetic nervous system activity may play a role in activating the hypothalamic-pituitary-adrenal (HPA) axis (4). In contrast, glucocorticoids may suppress sympathetic activity (118). The complex relationship between ANS and HPA axis function, especially in regard to blood pressure regulation, may be better understood if the influence of another important blood pressure regulator, the endothelium, is also considered. However, few studies have examined the relationship between the HPA axis and endothelial function. Human endothelial cells possess adrenocorticotrophic hormone (ACTH) receptors, suggesting that interactions between these two systems do occur (119). Glucocorticoids such as cortisol can reduce adhesion of inflammatory molecules to endothelial cells, suggesting enhancement of endothelial function (120).
Insulin resistance and SNS activity are correlated. Insulin resistance is associated with impairments in HRV and may contribute to enhanced SNS activity (65). SNS activity may similarly increase insulin resistance (121). Insulin produces vasodilatory responses that are at least partially endothelium-dependent and may help to protect endothelial function (122). Indeed, people with insulin resistance have poorer endothelial function than those without insulin resistance (66). Some have hypothesized that endothelial dysfunction is the primary cause of insulin resistance and accounts for associations between insulin resistance and other risk factors for CVD, including SNS overactivity (123, 124).
Autonomic nervous system nerves and endothelial cells both retain a great deal of plasticity even into adulthood (13). However, neither of these systems escapes eventual aging-related alterations. There is a progressive decline in brachial FMD with aging (125). Declines in NO, as measured by the metabolic by-product nitrate in the blood, and increases in endothelin are apparent in older people (126, 127). The decline in endothelial function with age is stimulus-dependent in that brachial vasodilator responses to the cold pressor test, a sympathetic stimulus, are blunted before brachial vasodilator responses to increased blood flow (94). Similar to the decline in endothelial function, ANS activity appears to be altered with aging, primarily reflecting enhanced SNS and suppressed PSNS activity (95, 127). Therefore, increased SNS-induced vasoconstrictive forces may overpower the declining endothelium-dependent vasodilatory forces with age.
