Calpains in muscle: selective and protective?


Calpains, Ca2+-activated neutral proteases, have a bit of an image problem, often being portrayed as mediators of non-specific cell damage. However, there is much evidence showing that calpains act only on specific targets and in a limited and highly directed manner, facilitating various important processes (Goll et al. 2003). In fact, there is increasing evidence that calpains play a vital role in protecting cellular integrity, as discussed in a paper in this issue of The Journal of Physiology by Gailly et al. (2007) in regard to skeletal muscle fibres.

Skeletal muscle fibres contain both ubiquitous calpains, μ-calpain and m-calpain, and also a muscle-specific isoform, calpain-3, which is poorly understood although it is known that its absence or dysfunction results in a type of limb girdle muscular dystrophy. Gailly et al. (2007) measured total calpain proteolytic activity and intracellular [Ca2+] ([Ca2+]i) in single enzyme-dissociated skeletal muscle fibres in various circumstances. They found that there was a small ongoing level of Ca2+-dependent proteolytic activity even when a muscle fibre was at rest. Importantly, this proteolytic activity was not increased when [Ca2+]i rose during normal muscle activation. It did increase markedly however, in non-physiological circumstances where there was substantial influx of extracellular Ca2+, such as when large-scale store-operated Ca2+ entry (SOCE) was triggered by artificially emptying the sarcoplasmic reticulum (SR) of Ca2+ or when subjecting a fibre to a large hypo-osmotic stimulus causing it to swell considerably, stretching the surface membrane. The entry of extracellular Ca2+ evidently occurred through stretch-activated channels, as it was blocked by the specific spider toxin, GsMTx4.

It is significant that in these cases of extracellular Ca2+ influx, the increase in calpain activity occurred even though the measured [Ca2+]i did not reach as high a level as during normal muscle activation. How can this be explained? It likely reflects that increases in both [Ca2+]i and calpain activity are not uniform throughout a fibre but instead occur in specialized local domains. Firstly, it would be expected that during the influx of extracellular Ca2+, the [Ca2+] just beneath the surface membrane at the sites of entry would be considerably higher than the level reached in the bulk of the cytoplasm. It may reach many micromolar there, even though the average in the cytoplasm as a whole only reaches ∼300 nm (Gailly et al. 2007). Secondly, it has been shown recently that even though most of the μ-calpain in a resting muscle fibre is freely diffusible, ∼15% of the total is bound at the inside of the surface membrane (Murphy et al. 2006). A large fraction of this pool of μ-calpain had undergone autolysis from the full-length 80 kDa form to the 78 kDa form, a process in which the calpain proteolyses itself and thereby increases its Ca2+ sensitivity more than 10-fold. In this form it is proteolytically active to some extent even at ∼300 nm Ca2+ (Goll et al. 2003; Murphy et al. 2006).

The localization of a pool of ‘preactivated’μ-calpain under the surface membrane is not unexpected, because it has been found that in the presence of micromolar Ca2+ the freely diffusible, full-length μ-calpain binds within seconds, probably at least in part to cell membranes, and stays bound for some time even if the [Ca2+] decreases (Murphy et al. 2006). Furthermore, the Ca2+ sensitivity of μ-calpain autolysis is increased in the presence of phospholipids (Goll et al. 2003). Together this could readily lead to accumulation of autolysed calpain in places where the local [Ca2+] rises, such as beneath the surface membrane, and also at the triad junctions, the sites of intracellular Ca2+ release.

Gailly et al. (2007) point out that this pool of preactivated calpain seems ideally placed for facilitating rapid membrane resealing if there is localized damage, a process known to be dependent on dysferlin and local remodelling of the cytoskeleton by calpain. It is important to note that this pool of sensitized μ-calpain, being bound, would be prevented from causing widespread uncontrolled proteolytic cleavage. It is targeted where it is needed. Its activity was detected in the experiments of Gailly et al. because they measured activity with a diffusible calpain substrate. So the substrate reached the protease rather than vice versa, and the observed resting proteolytic activity was not indicative of a deleterious situation, but rather of a protective mechanism.

As proteolytic activity in fibres from calpain-3 knockout mice was similar to that in normal fibres, Gailly et al. (2007) concluded that calpain-3 was not responsible for the observed activity. A role for m-calpain was also considered unlikely because it requires very high [Ca2+] for proteolytic activity, > 200 μm according to in-vitro data (Goll et al. 2003), but this possibility could not be completely discounted. It does seem however, that m-calpain may be regulated in vivo by factors other than Ca2+, possibly by phosphorylation.

Finally, Gailly et al. found that resting calpain activity was higher in fibres from mdx mice, which lack dystrophin, than it was in normal fibres, and it was increased proportionately more by hypo-osmotic stretch. They suggest such stretch may mimic the situation when mdx fibres are given eccentric contractions. Here, the calpains may be acting deleteriously, as such contractions cause excessive Ca2+ influx and lead to muscle weakness and damage (Allen et al. 2005). Interestingly, this weakness is largely due to reduced Ca2+ release from the SR, which likely results from Ca2+-mediated disruption of some triad junctions, possibly due to proteolysis by calpain (Verburg et al. 2005). Perhaps this is simply reflecting excessive activation of a process which in normal muscle fibres is actually another calpain-mediated protective mechanism, one which acts to reduce SR Ca2+ release at any triad where the [Ca2+] becomes excessive and potentially deleterious (Murphy et al. 2006).