CrossTalk opposing view: The diaphragm muscle does not atrophy as a result of inactivity

Gary C. Sieck (right) is the Vernon F. and Earline D. Dale endowed Professor and Chair of the Department of Physiology and Biomedical Engineering at Mayo Clinic. He served as President of the American Physiological Society and President of the Association of Chairs of Departments of Physiology. He is currently the editor-in-chief of Physiology and served as editor-in-chief of the Journal of Applied Physiology. Carlos B. Mantilla (left) is Professor of Anesthesiology and Physiology at Mayo Clinic. He is actively engaged in clinical practice, research and education working also in the Department of Physiology and Biomedical Engineering. Together, Dr Sieck and Dr Mantilla study the neuromotor control of breathing across the life span, and neuroplasticity following spinal cord injury.

The hypothesis that inactivity induces muscle fibre atrophy is widely accepted, and indeed this hypothesis has essentially become physiological dogma in the literature, forming the basis for conclusions reached by many investigators when interpreting their results. Certainly, there is no better muscle to explore the impact of inactivity than the diaphragm. Breathing persists day-in and day-out for our entire life, and the diaphragm muscle is the primary muscle for inspiration, responsible for moving air into our lungs. As such, the diaphragm muscle has a duty cycle (fraction of time active) ranging from 25 to 40% (Sieck et al. 1984; Kong & Berger, 1986; Mantilla et al. 2011; Mantilla & Sieck, 2011), which exceeds that of almost all other skeletal muscles in the body and is comparable to that of the heart. Thus, inactivity imposed on the diaphragm muscle should result in fibre atrophy if the hypothesis of inactivity-induced atrophy is correct. However, as with any hypothesis, support must come from direct experimentation and alternative hypotheses must be considered. In this respect, a number of observations strongly indicate that inactivity per se does not result in diaphragm muscle fibre atrophy. Unfortunately, many of these basic observations have been largely ignored by those who simply support the dogma of inactivity-induced muscle fibre atrophy.

First, one must consider the physiological activation of diaphragm muscle fibres by the central nervous system. The essential element of neural activation is the motor unit consisting of a motor neurone and the muscle fibres it innervates. These motor units comprise different muscle fibre types (Enad et al. 1989; Sieck et al. 1989) and are activated in an orderly fashion based on their mechanical and fatigue properties (Dick et al. 1987; Seven et al. 2013). It is not surprising that fatigue-resistant motor units are recruited to accomplish sustained motor behaviours (Sieck & Fournier, 1989; Mantilla et al. 2010). There is no better example than activation of diaphragm motor units during inspiration where recruitment of highly fatigable motor units would not be practical. Only during short-duration, expulsive motor behaviours would recruitment of more fatigable diaphragm motor units become necessary. These expulsive behaviours such as coughing and sneezing are relatively infrequent especially in comparison to the inspiratory-related activation of fatigue-resistant motor units. Thus, fatigue-resistant diaphragm motor units comprising type I and IIa fibres are clearly much more active than more fatigable fast-twitch motor units comprising type IIx and/or IIb fibres. If activity is the major determinant of diaphragm muscle fibre size, then those fibres that are most active (i.e. type I and IIa fibres) should be much larger than those fibres that are least active (i.e. type IIx and/or IIb fibres). Accordingly, type I and IIa fibres should be most sensitive to inactivity. However, type I and IIa diaphragm fibres are much smaller (less than half the size) than type IIx and/or IIb fibres across a range of species. Furthermore, type IIx and/or IIb fibres seem to be more susceptible to a variety of conditions inducing atrophy compared to type I and IIa fibres (Lewis & Sieck, 1990; Lewis et al. 1992; Farkas et al. 1994; Gosselin et al. 1994, 1996; Miyata et al. 1995; Zhan et al. 1997; Verheul et al. 2004; Greising et al. 2013; Mantilla et al. 2013). It is important to note that many of these conditions do not involve changes in activity, indicating that fibre type may be the main determinant of responses to activity, inactivity or other factors influencing fibre size (Greising et al. 2012). In conditions that directly impose inactivity of the diaphragm muscle, i.e. unilateral denervation (phrenicotomy), tetrodotoxin-induced nerve conduction block and upper cervical spinal cord hemisection (Miyata et al. 1995; Zhan et al. 1997; Mantilla et al. 2013), type IIx and/or IIb fibres also display the greatest extent of atrophy even though these fibres are almost certainly the least active within the diaphragm (Sieck & Fournier, 1989; Mantilla et al. 2010). Clearly, something other than activity must be the major determinant of diaphragm muscle fibre size (Enad et al. 1989; Sieck et al. 1989).

A clear definition of muscle inactivity is a useful staring point. Simply, inactivity is the absence of neural activation of muscle with consequent absence of excitation–contraction coupling of muscle fibres and force generation against an external load. Unfortunately, many studies remove the influence of external loading (e.g. reducing gravitational loading of muscles – unloading) without assessing the impact on neural activation (e.g. via electromyographic (EMG) recordings). Investigators often interpret their results showing muscle fibre atrophy as reflecting the impact of inactivity even if inactivity was not actually confirmed. In limb muscles that are unloaded by casting, bed rest or hind limb suspension, fibre atrophy is observed even though EMG activity persists (Blewett & Elder, 1993; Haddad et al. 2006; Roy et al. 2007). Even in situations where hind limb muscle inactivity is confirmed, the impact of gravitational unloading is commonly marginalised. All of these results are often interpreted only in the context of inactivity-induced muscle fibre atrophy.

The diaphragm muscle is not influenced by gravitational loading and works against resistance and elastic loading of the airways and chest wall. There is negligible passive loading of the diaphragm muscle across a range of lung volumes (Zhan et al. 1995; Shrager et al. 2002). Inactivity can be imposed on the diaphragm muscle by removal of neural activation via denervation (phrenicotomy), blockade of axonal conduction (e.g. tetrodotoxin nerve blockade) or removal of descending excitatory input to phrenic motor neurones (e.g. with upper cervical spinal cord injury (SCI)). In each of these cases, the imposed inactivity of the diaphragm muscle is confirmed by the absence of EMG activity (Miyata et al. 1995; Zhan et al. 1997; Prakash et al. 1999; Mantilla et al. 2007, 2013). If the hypothesis of inactivity-induced diaphragm muscle fibre atrophy is correct, a greater impact on type I and IIa fibres comprising highly active, fatigue-resistant motor units would be expected especially in comparison to type IIx and/or IIb fibres comprising far less active, more fatigable motor units. With phrenicotomy and TTX nerve blockade, effects on the diaphragm muscle are fibre-type selective, with an initial transient hypertrophy of type I and IIa fibres and an eventual atrophy of type IIx and/or IIb fibres, but not until nearly 7 days afterwards (Fig. 1). These results have been confirmed in a number of studies and do not support the hypothesis that inactivity induces muscle fibre atrophy. Similarly, following diaphragm muscle inactivity imposed by upper cervical SCI, there is no change in fibre size for at least 2 weeks (Miyata et al. 1995; Zhan et al. 1997; Prakash et al. 1999) and selective atrophy of type IIx and/or IIb fibres by 6 weeks (Mantilla et al. 2013). Again, the most active diaphragm muscle fibres do not display atrophy following inactivity. One could argue that under these conditions of hemidiaphragm inactivity, contraction of the contralateral hemidiaphragm imposes a mechanical strain (lengthening) on the inactivated hemidiaphragm. However, in a direct examination of this possibility, sonomicrometers were positioned on the sternal and midcostal regions of the inactivated (denervated) hemidiaphragm and strain imposed by contraction of the contralateral hemidiaphragm was measured (Zhan et al. 1995). Fibres in the sternal regions of the two hemidiaphragms are parallel to each other and during contraction of the intact side there is a slight shortening of fibres in the inactivated hemidiaphragm. Fibres in the midcostal region of the two hemidiaphragms are in series with each other; thus, contraction of the intact side imposes slight strain on fibres in the inactivated hemidiaphragm but without imposing any appreciable stress (loading) on these fibres. Most importantly, despite these differences in mechanical strain, denervation induced comparable changes in muscle fibre size in both the sternal and midcostal regions of the diaphragm muscle (Zhan et al. 1995). However, the diaphragm muscle is most certainly under mechanical loading prior to denervation or other conditions of inactivity; thus, mechanical unloading may still be a major contributor to changes in muscle fibre size. Regardless, based on these results, the hypothesis that inactivity induces diaphragm muscle fibre atrophy should be rejected.

Figure 1. Cross-sectional area (mean ± SEM) of type-identified diaphragm muscle fibres in adult rats across various conditions resulting in diaphragm inactivity.

In adult male rats, 2 weeks of diaphragm muscle inactivity was induced by unilateral spinal hemisection at C₂ (C₂ hemisection), unilateral denervation (phrenicotomy) or tetrodotoxin (TTX) axonal conduction blockade of the phrenic nerve. *Statistically significant difference from control group. Data adapted from Miyata et al. (1995), Zhan et al. (1997) and Prakash et al. (1999).

An alternative hypothesis that should be explored is the impact of trophic influences on diaphragm muscle fibres. Some of these trophic influences emanate from the nerve and vary across motor unit types. A predominant influence of innervation on muscle mechanical properties was clearly demonstrated in the classic cross-innervation study by Buller, Eccles and Eccles (Buller et al. 1960). In this study, innervation to the slow-twitch soleus and fast-twitch flexor digitorum longus muscles in the cat were crossed, and after re-innervation, the mechanical properties of the soleus became faster while those of the flexor digitorum longus became slower, suggesting fibre-type conversion. Insightfully, these investigators concluded that this transformation did not result from differences in neural activation per se but rather reflected a motor neurone-derived trophic influence. In a previous study (Miyata et al. 1995), we explored different models in which neurone-derived trophic effects are either removed (denervation) or not impaired (spinal hemisection) while the diaphragm muscle is inactive. With denervation, neurone-derived trophic influences are totally removed after a few days subsequent to Wallerian degeneration. However, with spinal cord hemisection the phrenic nerve remains intact and axoplasmic transport of neurone-derived trophic influences persists. Even though diaphragm muscle inactivity was present in both models, changes in muscle fibre size varied, clearly indicating that inactivity per se is not the main determinant of muscle fibre atrophy. These results clearly suggest that a neurone-derived trophic influence is important and in this respect there are many candidates (Mantilla & Sieck, 2008; Argadine et al. 2009, 2011). For example, neuregulin-1 acting via ErbB2-ErbB3 receptors increases diaphragm muscle protein synthesis and would promote maintenance of muscle fibre size (Hellyer et al. 2006). Neuregulin-1 is expressed in phrenic motor neurones and ErbB2-ErbB3 receptors are expressed in diaphragm muscle fibres (Mantilla & Sieck, 2008). Thus, alternatives such as altered neuregulin-1/ErbB2-ErbB3 signalling should be explored.

In conclusion, the preponderance of evidence does not support the hypothesis that inactivity induces diaphragm muscle fibre atrophy. On face value alone, this appears to be a very reasonable hypothesis given observations made predominantly in limb muscles where immobility quickly leads to substantial muscle atrophy. However, it is likely that this immobility results in unloading of limb muscles rather than frank inactivity. As with any hypothesis, when well-designed studies do not provide support there comes a time when the hypothesis should be rejected and alternative hypotheses considered. We believe that time has come.

Call for comments

Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief comment. Comments may be posted up to 6 weeks after publication of the article, at which point the discussion will close and authors will be invited to submit a ‘final word’. To submit a comment, go to http://jp.physoc.org/letters/submit/jphysiol;591/21/5259

Additional information

Competing interests

None declared.

Funding

This work is supported by NIH grants R01-HL096750 R01-AG044615 and T32-HL105355, and the Mayo Clinic.