ª 2015 John Wiley & Sons A/S.
Scand J Med Sci Sports 2015: 25 (Suppl. 4): 74–82 doi: 10.1111/sms.12600
Published by John Wiley & Sons Ltd
Review
Reflex control of the circulation during exercise P. J. Fadel Department of Kinesiology, College of Nursing and Health Innovation, University of Texas at Arlington, Arlington, Texas, USA Corresponding author: Paul J. Fadel, PhD, Department of Kinesiology, College of Nursing and Health Innovation, University of Texas at Arlington, 500 W. Nedderman room #147, Arlington, Texas 76019, USA. Tel: +1-817-272-3288, Fax: +(817) 2723233, E-mail: [email protected] Accepted for publication 1 October 2015
Appropriate cardiovascular and hemodynamic adjustments are necessary to meet the metabolic demands of working skeletal muscle during exercise. Alterations in the sympathetic and parasympathetic branches of the autonomic nervous system are fundamental in ensuring these adjustments are adequately made. Several neural mechanisms are responsible for the changes in autonomic activity with exercise and through complex interactions, contribute to the cardiovascular and hemodynamic changes in an intensity-dependent manner. This short review is from a presentation made at the Saltin Symposium June 2–4, 2015 in Copenhagen, Denmark. As such, the focus
will be on reflex control of the circulation with an emphasis on the work of the late Dr. Bengt Saltin. Moreover, a concerted effort is made to highlight the novel and insightful concepts put forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans. Thus, the multiple roles played by adenosine triphosphate (ATP) including its ability to induce vasodilatation, override sympathetic vasoconstriction and stimulate skeletal muscle afferents (exercise pressor reflex) are discussed and a conceptual framework is set suggesting a major role of ATP in blood flow regulation during exercise.
Profound cardiovascular and hemodynamic adjustments occur during dynamic exercise to match oxygen supply to the metabolic needs of active skeletal muscle. Alterations in the sympathetic and parasympathetic nervous system are responsible for many of the key cardiovascular responses to exercise including increases in cardiac output, skeletal muscle blood flow, and arterial blood pressure (BP) all of which are designed to meet the metabolic demands of exercising skeletal muscle (Mitchell, 1990; Kaufman & Forster, 1996; Laughlin et al., 2012; Fisher et al., 2015). Several neural mechanisms working in concert are responsible for these reflex adjustments and through complex interactions control the cardiovascular and hemodynamic changes in an intensity-dependent manner. Indeed, central command (a feed forward mechanism originating from the cerebral cortex and/or subcortical nuclei), the exercise pressor reflex (a feedback mechanism originating from skeletal muscle), the arterial baroreflex (a negative feedback mechanism originating from the carotid sinus and aortic arch), and the cardiopulmonary baroreflex (a negative feedback mechanism originating from low pressure mechanically sensitive stretch receptors located in the heart, great veins and blood vessels of the lungs) are all involved in mediating the characteristic cardiovascular and hemodynamic adjustments
to exercise. This review will focus on these neural mechanisms responsible for the reflex control of the circulation during exercise. Brief descriptions of each mechanism and its contributions will be discussed with the work of Dr. Bengt Saltin in each area highlighted, when applicable. Next, the novel and insightful concepts put forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans will be discussed (Mortensen & Saltin, 2014). Thus, the multiple roles played by adenosine triphosphate (ATP), including its ability to induce local vasodilatation, override local sympathetic vasoconstriction and stimulate skeletal muscle afferents (exercise pressor reflex) are discussed and a conceptual framework is set suggesting a major role of ATP in the regulation of blood flow during exercise. Because of limited space, not all research in the areas discussed can be covered in detail and therefore, additional review articles are cited throughout to direct the reader to further work in these important areas.
74
Central command The concept of descending signals from higher brain centers evoking cardiovascular responses to exercise has been well documented since Zuntz and Geppert
Reflex control of the circulation (Zuntz & Geppert, 1886) first hypothesized that exercise hyperpnea at the onset of exercise resulted, in part, from a central cortical control mechanism that simultaneously activated respiratory centers in conjunction with voluntary locomotor pathways. This feed forward parallel activation of locomotor and respiratory neural circuits was further advanced to incorporate the cardiovascular and circulatory responses to exercise (Johansson, 1895; Krogh & Lindhard, 1913). Collectively, these investigators established one of the seminal concepts in neural reflex control of the circulation. Originally termed “cortical irradiation,” this concept is now referred to as “central command,” referring to descending neural signals from motor centers in the brain that activate somatomotor, respiratory and cardiovascular systems in parallel (Goodwin et al., 1972). Some of the key studies identifying a role for central command in mediating the cardiovascular and hemodynamic responses to exercise in humans have used neuromuscular blockade with the majority of this work under the direction of Dr. Niels Secher in Copenhagen Denmark (Leonard et al., 1985; Secher et al., 1988; Victor et al., 1995) with some collaboration with Dr. Saltin (Leonard et al., 1985). The use of neuromuscular blockade to enhance central command input has been shown to increase the heart rate (HR), BP, and muscle sympathetic nerve activity (SNA) responses to both dynamic and static exercise (Leonard et al., 1985; Secher et al., 1988; Victor et al., 1995). However, central command may only influence muscle SNA at higher intensities of exercise (Victor et al., 1995). Studies employing additional strategies to modulate central command input such as electrical stimulation of resting skeletal muscle, hypnosis, muscle tendon vibration, and patients with unilateral limb weakness or sensory neuropathies have also demonstrated a role for central command in mediating the cardiovascular responses to exercise. For more details on these studies and others examining central command, several excellent reviews are available (Waldrop et al., 1996; Williamson et al., 2006; Green & Paterson, 2008; Fisher et al., 2015). More recent studies using hypnotic suggestion to selectively alter perceived levels of exertion or effort sense have suggested that the magnitude of a central command mediated cardiovascular response during exercise can be independent of force production (i.e., imagined exercise) and dictated more by an individual’s perception of effort (Thornton et al., 2001; Williamson et al., 2001). However, it is important to note that effort sense is a complex variable that can be influenced by numerous stimuli (e.g., pain), which provide afferent feedback to cardiovascular control centers of the brain (Williamson et al., 2006; Williamson, 2010). Nevertheless, in addition to the
traditional feed forward concept, it is likely that central command also involves feedback mechanisms (Williamson et al., 2006; Williamson, 2010). This is an area in which additional studies are warranted. Exercise pressor reflex Early physiologists recognized the potential for a reflex originating in skeletal muscle to contribute to cardiovascular control during exercise (Zuntz & Geppert, 1886; Johansson, 1895). However, it was the observations of Alam and Smirk (1937) (Alam & Smirk, 1937) that established the importance of neural afferent feedback from skeletal muscle. Since this seminal work, much research has been devoted to understanding this feedback mechanism emanating from skeletal muscle including its afferent and efferent components and the mechanism(s) by which reflex information is processed within the central nervous system (Mitchell, 1990; Kaufman & Forster, 1996; Potts, 2001; Smith et al., 2006). Briefly, skeletal muscle afferents are comprised of mechanically and metabolically sensitive sensory fibers that provide feedback to cardiovascular control areas in the brain stem in response to mechanical and metabolic stimuli, respectively (Mitchell, 1990; Kaufman & Forster, 1996; Potts, 2001; Smith et al., 2006). The afferent fibers of the muscle mechanoreflex are mainly comprised of thinly myelinated group III afferent neurons whose receptors are primarily activated by mechanical deformation as induced by changes in pressure or stretch. Thus, the muscle mechanoreflex is set-up to respond at the immediate onset of muscle contraction as the muscle is being mechanically distorted. On the other hand, the afferent fibers of the muscle metaboreflex are mainly composed of unmyelinated group IV afferent neurons whose receptors are chemically sensitive and stimulated by metabolites produced by working skeletal muscle. As such, the muscle metaboreflex does not respond at the immediate onset of muscle contraction requiring a sufficient period of time for the production of metabolites by contracting skeletal muscle (Mark et al., 1985). However, this separation in activation of these exercise pressor reflex afferent neurons is not absolute as some Group III fibers respond to metabolic changes within the muscle while Group IV fibers may respond to mechanical distortion. Aside from these polymodal qualities, it has also been demonstrated that both fiber types can be sensitized by changes in the chemical milieu of the muscle interstitium such that their responsiveness can be altered (Kaufman & Forster, 1996; Smith et al., 2006). Overall, during muscle contraction, activation of both mechanically and metabolically sensitive afferent fibers contributes to the exercise pressor reflexmediated neural cardiovascular responses. Dr. Saltin
75
Fadel was involved in a seminal paper in collaboration with Drs. Niels Secher and Jere Mitchell that demonstrated the importance of the exercise pressor reflex to the BP response to dynamic exercise in humans (Strange et al., 1993). Subjects performed electrically evoked contractions, to eliminate central command input, with and without epidural anesthesia (Lidocaine or Bupivacaine administered through L3-L4 catheter) to block skeletal muscle afferent feedback. During epidural anesthesia, the increase in BP with electrically evoked contractions was abolished demonstrating that the feedback from skeletal muscle was essential for the normal BP response to dynamic exercise. The majority of studies attempting to isolate the mechanoreflex, particularly in humans, have demonstrated rather minimal cardiovascular responses suggesting a small contribution to the overall neural cardiovascular response to exercise (Gladwell & Coote, 2002; Gladwell et al., 2005; Cui et al., 2006). Technical difficulties may contribute as it is very challenging to isolate the stimulation of mechanoreceptors, particularly in humans. In contrast, studies have shown a rather robust ability of the muscle metaboreflex to induce neural cardiovascular responses. The majority of studies have used postexercise ischemia to trap the exercise-induced metabolites within the previously active muscle, thus preserving their stimulatory effect on metabolically sensitive skeletal muscle afferents in the absence of the muscle mechanoreflex and central command. This maneuver has been consistently shown to maintain a major portion (~85%) of the exercise-induced increase in BP and muscle SNA (Mark et al., 1985) in an exercise intensity-dependent manner. Although the HR response returns toward baseline values during post-exercise ischemia, a contribution of the mus(a)
cle metaboreflex to the control of HR should not be excluded (O’Leary, 1993, 1996; Fisher et al., 2010). Despite the robust nature in which the metaboreflex stimulates BP and muscle SNA, the specific chemical product(s) that activate metabolically sensitive muscle afferents remains unclear (Mitchell, 1990; Kaufman & Forster, 1996; Smith et al., 2006; Kaufman, 2012). Several of the metabolites produced by muscular work have been targeted as candidates and these include, but are not limited to lactic acid, potassium, ATP, adenosine, arachidonic acid, diprotonated phosphate, prostaglandins, and hydrogen ion. Equivocal results have been reported for most of these candidates with studies supporting and refuting the involvement of each substance to some degree. However, the work of Saltin and Colleagues has indicated that adenosine likely does not contribute (MaClean et al., 1997). Indeed, intraarterial injection of adenosine did not directly result in an increase in muscle SNA when it was trapped in the leg by inflation of a supra-systolic pressure cuff. Rather it was not until it was released into the circulation that increases in muscle SNA were observed (Fig. 1). These data challenged a role for adenosine in directly stimulating skeletal muscle afferents in the leg of humans. Nevertheless, the primary substance(s) that activate metabolically sensitive muscle afferents remains to be elucidated. Interestingly, despite the uncertainty of metabolites involved, Dr. Saltin found that the lower cardioventilatory response following exercise training is, in part, due to a reduced signaling from skeletal muscle, likely involving traininginduced changes in potassium, lactate, ATP, and/or pH (Mortensen et al., 2013). This work not only highlights the importance of skeletal muscle afferent signaling in exercise training-induced adaptations but also suggests that it is likely a multitude of substances (b)
Fig. 1. Adenosine does not appear to directly stimulate skeletal muscle afferents. (a) Femoral arterial adenosine injection resulted in a threefold increase in muscle sympathetic nerve activity (MSNA) burst frequency (bursts/min). *Significant difference from baseline before adenosine injection P < 0.05. (b) Femoral arterial adenosine injection distal to a thigh cuff inflated to supra-systolic pressure had no effect on MSNA burst frequency. However, a rapid increase in MSNA bursts/min was observed after cuff release that was similar in magnitude to the increase observed with adenosine injection alone. *Significant difference from baseline before cuff release, P < 0.05. Adapted with permission from (MaClean et al., 1997).
76
Reflex control of the circulation that are involved in stimulating metabolically sensitive skeletal muscle afferents.
(a)
Arterial baroreflex Arterial baroreceptors located in the carotid artery and aorta play a key role in the rapid reflex adjustments to acute cardiovascular stressors (Mancia & Mark, 1983; Sagawa, 1983). These mechanically sensitive receptors function as the sensors in a negative feedback control loop that responds to beat-to-beat changes in BP by reflexively altering autonomic neural outflow to adjust cardiac output and total vascular conductance. A branch of the glossopharyngeal nerve, the Hering nerve, carries impulses from the carotid baroreceptors, while small vagal branches carry impulses from the aortic baroreceptors. These afferent signals converge centrally within the nucleus tractus solitarius of the medulla oblongata, a key central regulatory site for autonomic neural outflow. Although there was initial debate, a primary role of the arterial baroreflex to the autonomic adjustments to exercise is now well established. Both animal and humans studies have demonstrated that the arterial baroreflex continues to regulate BP during exercise by resetting to operate around the exercise-induced elevation in BP (Mancia & Mark, 1983; Potts, 2001; Fadel et al., 2003; Raven et al., 2006). Numerous studies have shown that baroreflex resetting occurs in direct relation to the intensity of exercise from low-to-moderate intensity but it was the work of Dr. Saltin in collaboration with Dr. Peter Raven that extended these findings to show resetting continues to occur up to maximal-intensity exercise (Fig. 2) (Norton et al., 1999). Further research has demonstrated that arterial baroreflex resetting during exercise is due to the interactive effects of central command (Gallagher et al., 2001b; McIlveen et al., 2001) and the exercise pressor reflex (Gallagher et al., 2001a; McIlveen et al., 2001) along with inputs from the cardiopulmonary baroreceptors. Overall, there is now convincing evidence that a properly functioning arterial baroreflex is requisite for an appropriate neural cardiovascular response to exercise. Studies in both animals and humans have investigated the relative contribution of changes in cardiac output and total vascular conductance to baroreflexmediated changes in BP and have demonstrated that the capacity of the carotid baroreflex to regulate BP depends almost exclusively on its ability to alter total vascular conductance both at rest and during exercise (Collins et al., 2001; Ogoh et al., 2002, 2003). In fact, although at rest approximately 25% of the carotid baroreflex-mediated BP response could be attributed to changes in cardiac output, during exercise the ability of the carotid baroreflex to control
(b)
Fig. 2. Progressive arterial baroreflex resetting to maximum dynamic exercise. Representation of carotid baroreflex resetting during dynamic exercise in one subject demonstrating intensity-dependent resetting to maximal exercise without a change in maximal gain. (a) resetting of carotid-cardiac baroreflex curve. (b) Resetting of carotid-mean arterial pressure baroreflex curve. L, leg-only exercise; L + A, combined leg and arm exercise. Reprinted with permission from (Norton et al., 1999).
BP was solely reliant on reflex-mediated changes in the vasculature (Collins et al., 2001; Ogoh et al., 2003). Thus, the ability of the arterial baroreflex to regulate BP is critically dependent on alterations in vascular tone both at rest and during exercise. Such changes in vascular conductance are mediated by a maintained ability of the baroreflex to regulate SNA during exercise. Indeed, multiple studies have reported a progressive resetting of the baroreflex control of SNA to operate around the exerciseinduced elevations in BP with a maintained or increased sensitivity (Fadel et al., 2001; Keller et al., 2004; Ogoh et al., 2007; Ichinose et al., 2008). However, an important paper by Dr. Saltin and colleagues indicated that although an activation of the carotid baroreceptors during moderate intensity (55% VO2max) exercise could withdraw sympathetic vasoconstrictor tone in active skeletal muscle, during high intensity (88% VO2max) exercise no effect of carotid baroreflex stimulation was observed (Strange et al., 1990). These data indicated that the baroreflex may not be able to restrain the elevated sympathetic
77
Fadel outflow to skeletal muscle with high intensity exercise. Nevertheless, a maintained ability of the carotid baroreflex to regulate BP was observed suggesting that other “inactive” vascular beds become important for the baroreflex control of BP during high intensity exercise.
in central venous pressure and cardiopulmonary baroreceptor load. This effect would then be overcome by greater stimulation of skeletal muscle afferents and the increased central command input as exercise intensity increases contributing to the elevations in muscle SNA to active and inactive skeletal muscle during exercise.
Cardiopulmonary baroreflex Cardiopulmonary baroreceptors are mechanically sensitive stretch receptors located in the heart and lungs that sense changes in central blood volume and pressure (Mark & Mancia, 1983; Hainsworth, 1991; Ray & Saito, 2000) and reflexively modulate SNA. Although not as well recognized, the cardiopulmonary baroreflex also plays an important role in the neural cardiovascular adjustments to exercise. Indeed, the limited information available suggests that the cardiopulmonary baroreflex plays a role in modulating SNA and BP responses during dynamic exercise along with contributions to arterial baroreflex resetting, as mentioned above. Initial observations in humans indicated that the removal of the inhibitory influence of the cardiopulmonary baroreceptors with lower body negative pressure facilitated the exercise BP response to static handgrip (Walker et al., 1980; Arrowood et al., 1993). Likewise, studies in rabbits (O’Hagan et al., 1994) and rats (Collins & DiCarlo, 1993) suggested that cardiac afferents were capable of modulating the increase in renal SNA during dynamic exercise. In humans, Ray et al. (1993) and Saito et al. (1993) demonstrated a clear role for the cardiopulmonary baroreflex in regulating the muscle SNA responses to dynamic exercise. Ray et al. (1993) used postural changes to change the cardiopulmonary load during one legged kicking demonstrating a reduction in muscle SNA below resting values when exercise was performed in the upright position and central venous pressure was increased due to enhanced venous return. In contrast, no changes in muscle SNA were observed during supine exercise when increases in central venous pressure were absent. In agreement, Saito et al. (1993) reported a decrease in muscle SNA from rest during light intensity leg cycling in the upright position suggesting that the increases in central blood volume associated with muscle pumping activates the cardiopulmonary baroreceptors inhibiting SNA. This effect would be overcome during higher intensity exercise by the greater stimulation of skeletal muscle afferents and central command. Collectively, these studies demonstrate that the cardiopulmonary baroreflex is indeed capable of modulating the muscle SNA response to exercise. In this regard, the initial inhibition of muscle SNA at the onset of dynamic leg exercise is likely due to the increase in venous return and subsequent elevations
78
Integrative neural cardiovascular responses and control of blood flow The aforementioned neural mechanisms are all involved in the reflex control of the circulation during exercise contributing to the neural, cardiovascular, and hemodynamic adjustments needed to meet the metabolic demands of working skeletal muscle. The increase in blood flow to active skeletal muscle requires up to a fivefold increase in cardiac output (Saltin et al., 1998; Thomas & Segal, 2004; Mortensen & Saltin, 2014; Fisher et al., 2015; Holwerda et al., 2015). A reduction in parasympathetic nerve activity and activation of cardiac SNA and epinephrine release from the adrenal medulla contribute to exercise-induced increases in heart rate, ventricular contractility, stroke volume and thus, cardiac output. The redistribution of cardiac output to active skeletal muscles is facilitated by a sympathetically mediated vasoconstriction in non-exercising muscles and visceral organs (e.g., splanchnic circulation). In addition, the ability of SNA to cause vasoconstriction is attenuated in the active muscles, in part due to an effect of muscle metabolites to diminish the vasoconstrictor response to a-adrenergic receptor activation (Remensnyder et al., 1962). This modulation has been termed functional sympatholysis; however, sympatholysis is not complete and the increased SNA to active muscles does have a restrictive effect on blood flow, which is important for the maintenance of BP during exercise (Thomas & Segal, 2004; Fisher et al., 2015; Holwerda et al., 2015). Indeed, the importance of the sympathetic nervous system to the cardiovascular adjustments to exercise can be readily appreciated from the finding that patients who have undergone a thoracolumbar sympathectomy for treatment of hypertension cannot maintain the lightest loads of dynamic exercise, even if performed in the supine position to enhance central blood volume and venous return (Marshall et al., 1961). In comparison to resting values of ~15% of cardiac output, active skeletal muscle can receive upwards to ~85% of cardiac output during maximal exercise (Laughlin et al., 2012). Indeed, during high intensity dynamic exercise, active skeletal muscle has a tremendous capacity to increase blood flow (Andersen & Saltin, 1985; Radegran & Saltin, 1998; Saltin et al., 1998; Mortensen & Saltin, 2014). Many of the studies demonstrating the massive capacity of
Reflex control of the circulation skeletal muscle to increase blood flow in humans have come from the Copenhagen Muscle Research Center and have been conducted under the guidance of Dr. Saltin using the one legged knee extensor exercise model. The initial studies by Saltin and Colleagues indicated that when cardiac output is not limiting, skeletal muscle blood flow can increase 100-fold from rest reaching perfusion rates of 300– 400 mL/min/kg (Andersen & Saltin, 1985; Radegran & Saltin, 1998; Saltin et al., 1998). Many subsequent studies have attempted to identify the vasoactive compound(s) that contribute to the intensity-dependent increases in skeletal muscle blood as well as the potential compounds that are responsible for functional sympatholysis, which further facilitates the increase in blood flow to active skeletal muscle. A companion review will specifically detail these studies investigating the regulation of skeletal muscle blood flow and readers are directed to this review for more details (Hellsten, 2015). Here, a focus will now be placed on ATP and the conceptual framework set forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans (Mortensen & Saltin, 2014). ATP: A major regulator of neural vascular control during exercise? Despite the short half-life of ATP (
Scand J Med Sci Sports 2015: 25 (Suppl. 4): 74–82 doi: 10.1111/sms.12600
Published by John Wiley & Sons Ltd
Review
Reflex control of the circulation during exercise P. J. Fadel Department of Kinesiology, College of Nursing and Health Innovation, University of Texas at Arlington, Arlington, Texas, USA Corresponding author: Paul J. Fadel, PhD, Department of Kinesiology, College of Nursing and Health Innovation, University of Texas at Arlington, 500 W. Nedderman room #147, Arlington, Texas 76019, USA. Tel: +1-817-272-3288, Fax: +(817) 2723233, E-mail: [email protected] Accepted for publication 1 October 2015
Appropriate cardiovascular and hemodynamic adjustments are necessary to meet the metabolic demands of working skeletal muscle during exercise. Alterations in the sympathetic and parasympathetic branches of the autonomic nervous system are fundamental in ensuring these adjustments are adequately made. Several neural mechanisms are responsible for the changes in autonomic activity with exercise and through complex interactions, contribute to the cardiovascular and hemodynamic changes in an intensity-dependent manner. This short review is from a presentation made at the Saltin Symposium June 2–4, 2015 in Copenhagen, Denmark. As such, the focus
will be on reflex control of the circulation with an emphasis on the work of the late Dr. Bengt Saltin. Moreover, a concerted effort is made to highlight the novel and insightful concepts put forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans. Thus, the multiple roles played by adenosine triphosphate (ATP) including its ability to induce vasodilatation, override sympathetic vasoconstriction and stimulate skeletal muscle afferents (exercise pressor reflex) are discussed and a conceptual framework is set suggesting a major role of ATP in blood flow regulation during exercise.
Profound cardiovascular and hemodynamic adjustments occur during dynamic exercise to match oxygen supply to the metabolic needs of active skeletal muscle. Alterations in the sympathetic and parasympathetic nervous system are responsible for many of the key cardiovascular responses to exercise including increases in cardiac output, skeletal muscle blood flow, and arterial blood pressure (BP) all of which are designed to meet the metabolic demands of exercising skeletal muscle (Mitchell, 1990; Kaufman & Forster, 1996; Laughlin et al., 2012; Fisher et al., 2015). Several neural mechanisms working in concert are responsible for these reflex adjustments and through complex interactions control the cardiovascular and hemodynamic changes in an intensity-dependent manner. Indeed, central command (a feed forward mechanism originating from the cerebral cortex and/or subcortical nuclei), the exercise pressor reflex (a feedback mechanism originating from skeletal muscle), the arterial baroreflex (a negative feedback mechanism originating from the carotid sinus and aortic arch), and the cardiopulmonary baroreflex (a negative feedback mechanism originating from low pressure mechanically sensitive stretch receptors located in the heart, great veins and blood vessels of the lungs) are all involved in mediating the characteristic cardiovascular and hemodynamic adjustments
to exercise. This review will focus on these neural mechanisms responsible for the reflex control of the circulation during exercise. Brief descriptions of each mechanism and its contributions will be discussed with the work of Dr. Bengt Saltin in each area highlighted, when applicable. Next, the novel and insightful concepts put forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans will be discussed (Mortensen & Saltin, 2014). Thus, the multiple roles played by adenosine triphosphate (ATP), including its ability to induce local vasodilatation, override local sympathetic vasoconstriction and stimulate skeletal muscle afferents (exercise pressor reflex) are discussed and a conceptual framework is set suggesting a major role of ATP in the regulation of blood flow during exercise. Because of limited space, not all research in the areas discussed can be covered in detail and therefore, additional review articles are cited throughout to direct the reader to further work in these important areas.
74
Central command The concept of descending signals from higher brain centers evoking cardiovascular responses to exercise has been well documented since Zuntz and Geppert
Reflex control of the circulation (Zuntz & Geppert, 1886) first hypothesized that exercise hyperpnea at the onset of exercise resulted, in part, from a central cortical control mechanism that simultaneously activated respiratory centers in conjunction with voluntary locomotor pathways. This feed forward parallel activation of locomotor and respiratory neural circuits was further advanced to incorporate the cardiovascular and circulatory responses to exercise (Johansson, 1895; Krogh & Lindhard, 1913). Collectively, these investigators established one of the seminal concepts in neural reflex control of the circulation. Originally termed “cortical irradiation,” this concept is now referred to as “central command,” referring to descending neural signals from motor centers in the brain that activate somatomotor, respiratory and cardiovascular systems in parallel (Goodwin et al., 1972). Some of the key studies identifying a role for central command in mediating the cardiovascular and hemodynamic responses to exercise in humans have used neuromuscular blockade with the majority of this work under the direction of Dr. Niels Secher in Copenhagen Denmark (Leonard et al., 1985; Secher et al., 1988; Victor et al., 1995) with some collaboration with Dr. Saltin (Leonard et al., 1985). The use of neuromuscular blockade to enhance central command input has been shown to increase the heart rate (HR), BP, and muscle sympathetic nerve activity (SNA) responses to both dynamic and static exercise (Leonard et al., 1985; Secher et al., 1988; Victor et al., 1995). However, central command may only influence muscle SNA at higher intensities of exercise (Victor et al., 1995). Studies employing additional strategies to modulate central command input such as electrical stimulation of resting skeletal muscle, hypnosis, muscle tendon vibration, and patients with unilateral limb weakness or sensory neuropathies have also demonstrated a role for central command in mediating the cardiovascular responses to exercise. For more details on these studies and others examining central command, several excellent reviews are available (Waldrop et al., 1996; Williamson et al., 2006; Green & Paterson, 2008; Fisher et al., 2015). More recent studies using hypnotic suggestion to selectively alter perceived levels of exertion or effort sense have suggested that the magnitude of a central command mediated cardiovascular response during exercise can be independent of force production (i.e., imagined exercise) and dictated more by an individual’s perception of effort (Thornton et al., 2001; Williamson et al., 2001). However, it is important to note that effort sense is a complex variable that can be influenced by numerous stimuli (e.g., pain), which provide afferent feedback to cardiovascular control centers of the brain (Williamson et al., 2006; Williamson, 2010). Nevertheless, in addition to the
traditional feed forward concept, it is likely that central command also involves feedback mechanisms (Williamson et al., 2006; Williamson, 2010). This is an area in which additional studies are warranted. Exercise pressor reflex Early physiologists recognized the potential for a reflex originating in skeletal muscle to contribute to cardiovascular control during exercise (Zuntz & Geppert, 1886; Johansson, 1895). However, it was the observations of Alam and Smirk (1937) (Alam & Smirk, 1937) that established the importance of neural afferent feedback from skeletal muscle. Since this seminal work, much research has been devoted to understanding this feedback mechanism emanating from skeletal muscle including its afferent and efferent components and the mechanism(s) by which reflex information is processed within the central nervous system (Mitchell, 1990; Kaufman & Forster, 1996; Potts, 2001; Smith et al., 2006). Briefly, skeletal muscle afferents are comprised of mechanically and metabolically sensitive sensory fibers that provide feedback to cardiovascular control areas in the brain stem in response to mechanical and metabolic stimuli, respectively (Mitchell, 1990; Kaufman & Forster, 1996; Potts, 2001; Smith et al., 2006). The afferent fibers of the muscle mechanoreflex are mainly comprised of thinly myelinated group III afferent neurons whose receptors are primarily activated by mechanical deformation as induced by changes in pressure or stretch. Thus, the muscle mechanoreflex is set-up to respond at the immediate onset of muscle contraction as the muscle is being mechanically distorted. On the other hand, the afferent fibers of the muscle metaboreflex are mainly composed of unmyelinated group IV afferent neurons whose receptors are chemically sensitive and stimulated by metabolites produced by working skeletal muscle. As such, the muscle metaboreflex does not respond at the immediate onset of muscle contraction requiring a sufficient period of time for the production of metabolites by contracting skeletal muscle (Mark et al., 1985). However, this separation in activation of these exercise pressor reflex afferent neurons is not absolute as some Group III fibers respond to metabolic changes within the muscle while Group IV fibers may respond to mechanical distortion. Aside from these polymodal qualities, it has also been demonstrated that both fiber types can be sensitized by changes in the chemical milieu of the muscle interstitium such that their responsiveness can be altered (Kaufman & Forster, 1996; Smith et al., 2006). Overall, during muscle contraction, activation of both mechanically and metabolically sensitive afferent fibers contributes to the exercise pressor reflexmediated neural cardiovascular responses. Dr. Saltin
75
Fadel was involved in a seminal paper in collaboration with Drs. Niels Secher and Jere Mitchell that demonstrated the importance of the exercise pressor reflex to the BP response to dynamic exercise in humans (Strange et al., 1993). Subjects performed electrically evoked contractions, to eliminate central command input, with and without epidural anesthesia (Lidocaine or Bupivacaine administered through L3-L4 catheter) to block skeletal muscle afferent feedback. During epidural anesthesia, the increase in BP with electrically evoked contractions was abolished demonstrating that the feedback from skeletal muscle was essential for the normal BP response to dynamic exercise. The majority of studies attempting to isolate the mechanoreflex, particularly in humans, have demonstrated rather minimal cardiovascular responses suggesting a small contribution to the overall neural cardiovascular response to exercise (Gladwell & Coote, 2002; Gladwell et al., 2005; Cui et al., 2006). Technical difficulties may contribute as it is very challenging to isolate the stimulation of mechanoreceptors, particularly in humans. In contrast, studies have shown a rather robust ability of the muscle metaboreflex to induce neural cardiovascular responses. The majority of studies have used postexercise ischemia to trap the exercise-induced metabolites within the previously active muscle, thus preserving their stimulatory effect on metabolically sensitive skeletal muscle afferents in the absence of the muscle mechanoreflex and central command. This maneuver has been consistently shown to maintain a major portion (~85%) of the exercise-induced increase in BP and muscle SNA (Mark et al., 1985) in an exercise intensity-dependent manner. Although the HR response returns toward baseline values during post-exercise ischemia, a contribution of the mus(a)
cle metaboreflex to the control of HR should not be excluded (O’Leary, 1993, 1996; Fisher et al., 2010). Despite the robust nature in which the metaboreflex stimulates BP and muscle SNA, the specific chemical product(s) that activate metabolically sensitive muscle afferents remains unclear (Mitchell, 1990; Kaufman & Forster, 1996; Smith et al., 2006; Kaufman, 2012). Several of the metabolites produced by muscular work have been targeted as candidates and these include, but are not limited to lactic acid, potassium, ATP, adenosine, arachidonic acid, diprotonated phosphate, prostaglandins, and hydrogen ion. Equivocal results have been reported for most of these candidates with studies supporting and refuting the involvement of each substance to some degree. However, the work of Saltin and Colleagues has indicated that adenosine likely does not contribute (MaClean et al., 1997). Indeed, intraarterial injection of adenosine did not directly result in an increase in muscle SNA when it was trapped in the leg by inflation of a supra-systolic pressure cuff. Rather it was not until it was released into the circulation that increases in muscle SNA were observed (Fig. 1). These data challenged a role for adenosine in directly stimulating skeletal muscle afferents in the leg of humans. Nevertheless, the primary substance(s) that activate metabolically sensitive muscle afferents remains to be elucidated. Interestingly, despite the uncertainty of metabolites involved, Dr. Saltin found that the lower cardioventilatory response following exercise training is, in part, due to a reduced signaling from skeletal muscle, likely involving traininginduced changes in potassium, lactate, ATP, and/or pH (Mortensen et al., 2013). This work not only highlights the importance of skeletal muscle afferent signaling in exercise training-induced adaptations but also suggests that it is likely a multitude of substances (b)
Fig. 1. Adenosine does not appear to directly stimulate skeletal muscle afferents. (a) Femoral arterial adenosine injection resulted in a threefold increase in muscle sympathetic nerve activity (MSNA) burst frequency (bursts/min). *Significant difference from baseline before adenosine injection P < 0.05. (b) Femoral arterial adenosine injection distal to a thigh cuff inflated to supra-systolic pressure had no effect on MSNA burst frequency. However, a rapid increase in MSNA bursts/min was observed after cuff release that was similar in magnitude to the increase observed with adenosine injection alone. *Significant difference from baseline before cuff release, P < 0.05. Adapted with permission from (MaClean et al., 1997).
76
Reflex control of the circulation that are involved in stimulating metabolically sensitive skeletal muscle afferents.
(a)
Arterial baroreflex Arterial baroreceptors located in the carotid artery and aorta play a key role in the rapid reflex adjustments to acute cardiovascular stressors (Mancia & Mark, 1983; Sagawa, 1983). These mechanically sensitive receptors function as the sensors in a negative feedback control loop that responds to beat-to-beat changes in BP by reflexively altering autonomic neural outflow to adjust cardiac output and total vascular conductance. A branch of the glossopharyngeal nerve, the Hering nerve, carries impulses from the carotid baroreceptors, while small vagal branches carry impulses from the aortic baroreceptors. These afferent signals converge centrally within the nucleus tractus solitarius of the medulla oblongata, a key central regulatory site for autonomic neural outflow. Although there was initial debate, a primary role of the arterial baroreflex to the autonomic adjustments to exercise is now well established. Both animal and humans studies have demonstrated that the arterial baroreflex continues to regulate BP during exercise by resetting to operate around the exercise-induced elevation in BP (Mancia & Mark, 1983; Potts, 2001; Fadel et al., 2003; Raven et al., 2006). Numerous studies have shown that baroreflex resetting occurs in direct relation to the intensity of exercise from low-to-moderate intensity but it was the work of Dr. Saltin in collaboration with Dr. Peter Raven that extended these findings to show resetting continues to occur up to maximal-intensity exercise (Fig. 2) (Norton et al., 1999). Further research has demonstrated that arterial baroreflex resetting during exercise is due to the interactive effects of central command (Gallagher et al., 2001b; McIlveen et al., 2001) and the exercise pressor reflex (Gallagher et al., 2001a; McIlveen et al., 2001) along with inputs from the cardiopulmonary baroreceptors. Overall, there is now convincing evidence that a properly functioning arterial baroreflex is requisite for an appropriate neural cardiovascular response to exercise. Studies in both animals and humans have investigated the relative contribution of changes in cardiac output and total vascular conductance to baroreflexmediated changes in BP and have demonstrated that the capacity of the carotid baroreflex to regulate BP depends almost exclusively on its ability to alter total vascular conductance both at rest and during exercise (Collins et al., 2001; Ogoh et al., 2002, 2003). In fact, although at rest approximately 25% of the carotid baroreflex-mediated BP response could be attributed to changes in cardiac output, during exercise the ability of the carotid baroreflex to control
(b)
Fig. 2. Progressive arterial baroreflex resetting to maximum dynamic exercise. Representation of carotid baroreflex resetting during dynamic exercise in one subject demonstrating intensity-dependent resetting to maximal exercise without a change in maximal gain. (a) resetting of carotid-cardiac baroreflex curve. (b) Resetting of carotid-mean arterial pressure baroreflex curve. L, leg-only exercise; L + A, combined leg and arm exercise. Reprinted with permission from (Norton et al., 1999).
BP was solely reliant on reflex-mediated changes in the vasculature (Collins et al., 2001; Ogoh et al., 2003). Thus, the ability of the arterial baroreflex to regulate BP is critically dependent on alterations in vascular tone both at rest and during exercise. Such changes in vascular conductance are mediated by a maintained ability of the baroreflex to regulate SNA during exercise. Indeed, multiple studies have reported a progressive resetting of the baroreflex control of SNA to operate around the exerciseinduced elevations in BP with a maintained or increased sensitivity (Fadel et al., 2001; Keller et al., 2004; Ogoh et al., 2007; Ichinose et al., 2008). However, an important paper by Dr. Saltin and colleagues indicated that although an activation of the carotid baroreceptors during moderate intensity (55% VO2max) exercise could withdraw sympathetic vasoconstrictor tone in active skeletal muscle, during high intensity (88% VO2max) exercise no effect of carotid baroreflex stimulation was observed (Strange et al., 1990). These data indicated that the baroreflex may not be able to restrain the elevated sympathetic
77
Fadel outflow to skeletal muscle with high intensity exercise. Nevertheless, a maintained ability of the carotid baroreflex to regulate BP was observed suggesting that other “inactive” vascular beds become important for the baroreflex control of BP during high intensity exercise.
in central venous pressure and cardiopulmonary baroreceptor load. This effect would then be overcome by greater stimulation of skeletal muscle afferents and the increased central command input as exercise intensity increases contributing to the elevations in muscle SNA to active and inactive skeletal muscle during exercise.
Cardiopulmonary baroreflex Cardiopulmonary baroreceptors are mechanically sensitive stretch receptors located in the heart and lungs that sense changes in central blood volume and pressure (Mark & Mancia, 1983; Hainsworth, 1991; Ray & Saito, 2000) and reflexively modulate SNA. Although not as well recognized, the cardiopulmonary baroreflex also plays an important role in the neural cardiovascular adjustments to exercise. Indeed, the limited information available suggests that the cardiopulmonary baroreflex plays a role in modulating SNA and BP responses during dynamic exercise along with contributions to arterial baroreflex resetting, as mentioned above. Initial observations in humans indicated that the removal of the inhibitory influence of the cardiopulmonary baroreceptors with lower body negative pressure facilitated the exercise BP response to static handgrip (Walker et al., 1980; Arrowood et al., 1993). Likewise, studies in rabbits (O’Hagan et al., 1994) and rats (Collins & DiCarlo, 1993) suggested that cardiac afferents were capable of modulating the increase in renal SNA during dynamic exercise. In humans, Ray et al. (1993) and Saito et al. (1993) demonstrated a clear role for the cardiopulmonary baroreflex in regulating the muscle SNA responses to dynamic exercise. Ray et al. (1993) used postural changes to change the cardiopulmonary load during one legged kicking demonstrating a reduction in muscle SNA below resting values when exercise was performed in the upright position and central venous pressure was increased due to enhanced venous return. In contrast, no changes in muscle SNA were observed during supine exercise when increases in central venous pressure were absent. In agreement, Saito et al. (1993) reported a decrease in muscle SNA from rest during light intensity leg cycling in the upright position suggesting that the increases in central blood volume associated with muscle pumping activates the cardiopulmonary baroreceptors inhibiting SNA. This effect would be overcome during higher intensity exercise by the greater stimulation of skeletal muscle afferents and central command. Collectively, these studies demonstrate that the cardiopulmonary baroreflex is indeed capable of modulating the muscle SNA response to exercise. In this regard, the initial inhibition of muscle SNA at the onset of dynamic leg exercise is likely due to the increase in venous return and subsequent elevations
78
Integrative neural cardiovascular responses and control of blood flow The aforementioned neural mechanisms are all involved in the reflex control of the circulation during exercise contributing to the neural, cardiovascular, and hemodynamic adjustments needed to meet the metabolic demands of working skeletal muscle. The increase in blood flow to active skeletal muscle requires up to a fivefold increase in cardiac output (Saltin et al., 1998; Thomas & Segal, 2004; Mortensen & Saltin, 2014; Fisher et al., 2015; Holwerda et al., 2015). A reduction in parasympathetic nerve activity and activation of cardiac SNA and epinephrine release from the adrenal medulla contribute to exercise-induced increases in heart rate, ventricular contractility, stroke volume and thus, cardiac output. The redistribution of cardiac output to active skeletal muscles is facilitated by a sympathetically mediated vasoconstriction in non-exercising muscles and visceral organs (e.g., splanchnic circulation). In addition, the ability of SNA to cause vasoconstriction is attenuated in the active muscles, in part due to an effect of muscle metabolites to diminish the vasoconstrictor response to a-adrenergic receptor activation (Remensnyder et al., 1962). This modulation has been termed functional sympatholysis; however, sympatholysis is not complete and the increased SNA to active muscles does have a restrictive effect on blood flow, which is important for the maintenance of BP during exercise (Thomas & Segal, 2004; Fisher et al., 2015; Holwerda et al., 2015). Indeed, the importance of the sympathetic nervous system to the cardiovascular adjustments to exercise can be readily appreciated from the finding that patients who have undergone a thoracolumbar sympathectomy for treatment of hypertension cannot maintain the lightest loads of dynamic exercise, even if performed in the supine position to enhance central blood volume and venous return (Marshall et al., 1961). In comparison to resting values of ~15% of cardiac output, active skeletal muscle can receive upwards to ~85% of cardiac output during maximal exercise (Laughlin et al., 2012). Indeed, during high intensity dynamic exercise, active skeletal muscle has a tremendous capacity to increase blood flow (Andersen & Saltin, 1985; Radegran & Saltin, 1998; Saltin et al., 1998; Mortensen & Saltin, 2014). Many of the studies demonstrating the massive capacity of
Reflex control of the circulation skeletal muscle to increase blood flow in humans have come from the Copenhagen Muscle Research Center and have been conducted under the guidance of Dr. Saltin using the one legged knee extensor exercise model. The initial studies by Saltin and Colleagues indicated that when cardiac output is not limiting, skeletal muscle blood flow can increase 100-fold from rest reaching perfusion rates of 300– 400 mL/min/kg (Andersen & Saltin, 1985; Radegran & Saltin, 1998; Saltin et al., 1998). Many subsequent studies have attempted to identify the vasoactive compound(s) that contribute to the intensity-dependent increases in skeletal muscle blood as well as the potential compounds that are responsible for functional sympatholysis, which further facilitates the increase in blood flow to active skeletal muscle. A companion review will specifically detail these studies investigating the regulation of skeletal muscle blood flow and readers are directed to this review for more details (Hellsten, 2015). Here, a focus will now be placed on ATP and the conceptual framework set forth by Dr. Saltin in his last published review article on the regulation of skeletal muscle blood flow in humans (Mortensen & Saltin, 2014). ATP: A major regulator of neural vascular control during exercise? Despite the short half-life of ATP (
