Found in translation: neural feedback from exercising muscles

Perhaps the most universal observations in physiology are that heart rate, respiration, and blood pressure rise during exercise (Krogh & Lindhard, 1913). These simple observations have been the focus of investigation since the dawn of experimental medicine and physiology in the 1800s, and they were discussed on an observational basis much earlier. By the late 1700s and early 1900s there were two ‘competing’ ideas about the neural signals that contribute to the acceleration of the circulation and respiration during exercise. One line of thinking was that signals related to motor commands also activate the cardiovascular and respiratory centres (Krogh & Lindhard, 1913; Goodwin et al. 1972). This mechanism has had multiple names, but is currently termed ‘central command.’ Most recently central command has been shown both in conscious humans and in animal preparations to play a key role in the baroreceptor resetting that occurs with exercise (Gallagher et al. 2001).

The second main mechanism thought to contribute to the cardiovascular and respiratory adjustments to exercise was feedback from the contracting muscles. There were a number of early observations related to this topic, but perhaps the most dramatic came from Alam and Smirk who showed in the late 1930s that post-exercise muscle (leg) ischaemia caused a sustained rise in blood pressure after exercise stopped as long as the muscle ischaemia continued. This response was absent when post-exercise muscle ischaemia was studied in the insensitive leg of a patient with a unilateral sensory deficit in the lower extremity (Alam & Smirk, 1938). These and other studies raised questions about the nature of the signals from skeletal muscle that contribute to the cardiovascular and respiratory responses to exercise (Kaufman & Forster, 1996). Are the signals related to muscle force or muscle movement? Are muscle spindles and/or tendon organs responsible? Are there chemosensitive sensory afferents in skeletal muscle that sense the metabolic byproducts of contraction?

These questions were all in play in the 1960s and early 1970s when Ian McCloskey and Jere Mitchell published a paper in The Journal of Physiology entitled ‘Reflex cardiovascular and respiratory responses originating in exercising muscle’ (McCloskey & Mitchell, 1972). In this paper, isometric contractions of hind limb muscles in cats elicited by stimulating L7 and S1 ventral roots were associated with a rise in arterial pressure and small increases in heart rate and ventilation. These responses were abolished by cutting the dorsal roots.

Importantly, the blood pressure raising effects of the contractions (but not heart rate or ventilation) were augmented by superimposition of ischaemia during and after contractions. Similar physiological responses were also caused by injection of potassium chloride into the arterial vessels supplying the muscles of interest. Finally, studies with anodal current block and selective application of varying concentrations of local anaesthetics demonstrated that the sensory fibres involved were the thinly myelinated and unmyelinated group III and IV afferents and that the large myelinated afferents did not contribute to the heart rate, respiratory and pressor responses to muscle contraction.

So, what did we know as a result of this paper that was unknown before its publication? First, feedback from contracting muscles can evoke powerful blood pressure-raising reflexes, and to a lesser extent evoke a rise in heart rate and ventilation. Second, the studies with potassium suggested that some substance released by the active muscles stimulated the afferents. Third, the fine group III and IV afferents emerged as the prime neural substrate for these responses. These findings have withstood more than 30 years of rigorous investigation; they have been extended and together they form the basis for a large body of work on the reflex cardiovascular and respiratory responses to exercise.

Where have these studies led? First, there has been a multifaceted attempt to understand what substances from the contracting muscles stimulate the chemosensitive muscle afferents. A host of factors have been identified as candidates, with major contributions coming from hydrogen ion and perhaps adenosine or related compounds (Kaufman & Forster, 1996). Second, there has been an impressive effort to trace the neural pathways in the spinal cord and brainstem that contribute to these responses and their modulation. The role of substance P has a key neurotransmitter in the spinal cord is one of the main highlights of this work, and the potential for the endogenous opiate system to modify it has also been intriguing. More recently, the way the afferent information converges onto barosensitive brainstem areas has been ‘mapped’ (Kaufman & Forster, 1996; Potts et al. 2003). Third, what do these afferent signals do? One idea is that the primary function of these afferents in ‘real life’ is to sense a mismatch between skeletal muscle metabolism and blood flow and then evoke a rise in arterial pressure that helps relieve this mismatch and improve perfusion in the active skeletal muscles (Augustyniak et al. 2001). Fourth, are any of these small either thinly myelinated or unmyelinated afferents mechanosensitive? It was once thought that the group III (thinly myelinated) afferents were primarily mechanosensitive and the unmyelinated group IV afferents were primarily chemosensitive. However, this is not completely straightforward and under some conditions chemosensitive afferents can become sensitized and behave in a mechanosensitive manner (Li et al. 2004a). Fifth, do these afferent systems play an important role in any pathophysiological responses? Emerging evidence (including ongoing work by Jere Mitchell) indicates that when patients or animals with heart failure exercise, the thinly myelinated mechanosensitive afferents become engaged early on and evoke a large increase in sympathetic outflow to skeletal muscle and perhaps the kidney. These events could have very negative consequences under a variety of circumstances (Middlekauff et al. 2001, 2004; Joyner, 2004; Li et al. 2004b; Smith et al. 2005).

Finally, how vanniloid receptors operate in these sensory afferents and how the receptors are altered by conditions like heart failure is another new subplot to this long running story (Li et al. 2004a; Smith et al. 2005). So, simple observations first made in humans have led to progressively reductionist approaches that are ultimately explaining why heart rate, ventilation and blood pressure rise during exercise. Is this a case of the physiological role feedback from muscle being found in translation – found in translational research?

Original classic paper

The original classic paper reviewed in this article and published in The Journal of Physiology can be accessed online at: 10.1113/jphysiol.2005.092411 http://jp.physoc.org/cgi/content/full/ jphysiol.2005.092411/DC1

This material can also be found as part of the full-text HTML version available from http://www.blackwell-synergy.com