Q&A Report: Molecular Signals Mediating Increases in Muscle Size and Function

Bert Blaauw, PhD
Associate Professor/PI
Department of Biomedical Sciences/Venetian Institute of Molecular Medicine
University of Padova

Good question, and also a very difficult one. In my opinion, I would say repeated bouts of muscle activity would be best to modulate nerve remodeling and potentially even improving CR. A paper from Jorge Ruas’ lab, now more than a decade ago, linked pgc1-alpha levels and aerobic exercise to changes in kynurenine metabolism in muscle, which then feeds back on the mood. So really suggesting that metabolic alterations in muscle could alter hippocampal levels of metabolites connected to tryptophan metabolism. Perhaps similar effects could be observed when inducing muscle growth (thinking along the lines of metabolic competition), but my guess would be that aerobic exercise (and therefore not muscle hypertrophy) would be the most effective type of exercise.

Yes we do when we activate Akt. We showed this in figure 4B in Marabita et al., Cell Reports 2016.

In the past, we checked this (FASEB J 2009), and while we saw an increase in proliferating interstitial cells, we never found them inside the muscle fibers. So I think there is no evident proliferation of satellite cells.

Good question, I think a big part of the increased mTORC1 signaling in aged muscle was observed in western blotting. In this case, you take a whole muscle lysate and therefore a mix of all fibers. Marcus Ruegg showed that the increased mTORC1 signal is coming mainly from a few fibers with very elevated mTORC1 signaling. We have seen the same, these fibers are very likely the denervated fibers which increase with aging. So in my mind, increased P-S6, increased denervation, decreased function.

One of the recent papers from Troy Hornberger shows this re-expression nicely in the ‘classical’ functional overload approach. With the HSA-CreER approach, it is critical not to induce/damage/regeneration, as this will lead to re-introduction, unless one continues giving tamoxifen to the animals (not sure this is a great idea). With the AAV-approach, this problem is not so relevant as the is continuously ‘on’ and only in the fibers.

We never checked VEGF (even though it is likely increasing), but capillarization is clearly increased as we showed in (Blaauw et al., FASEB J 2009).

We see both transversal and longitudinal staining; however, mitochondrial structure as seen by EM is significantly impaired in skinned fibers.

We have found that activation of Akt affects the regulation of anti-oxidant markers in dystrophic muscles (Hum Mol Genetics 2008); however, if oxidative stress affects Akt, I wouldn’t know.

No, we don’t see any difference in the hypertrophic effect of Akt in different fiber types. I think part of this is also due to the fact that this is really a very strong activation of Akt.

Very good question, one I am still struggling with myself as well. Clearly, rapamycin doesn’t block all mTORC1 signaling; however, it does hit all parts of the fiber at the same time. Raptor knockout is efficient (by western blotting and RTPCR), but it is not clear to me how much residual mTORC1 signaling remains. I feel that Raptor knockout lowers overall mTORC1 signaling a lot (but not completely, at least not in the beginning as all nuclei need to recombine), possibly affecting the 4E-BP1 ‘branch’ quite significantly. Rapamycin on the other hand inhibits almost completely part of mTORC1 signaling (for example S6K), while not affecting much of the mTORC1 targets. These different levels of inhibition create a situation where the combined inhibitory effect is much stronger than the single one.

Under basal conditions we don’t observe major differences. There was a paper in drosophila from 2007, I believe, where they even suggested that S6K is involved in the induction of autophagy (to counterbalance the inhibitory effect of mTOR). LC3 levels are not different between wildtype and KO animals

We didn’t check calcium transients, but we also found a strong reduction in the strength of skinned muscle fibers, where calcium is not a limiting factor. So, while we cannot exclude an effect of calcium release, it is clearly not the major factor for the loss in muscle force in vivo. When we give rapamycin in control animals, we don’t see any reduction in specific force (I think the dosage required for affecting the RyR is 5-10 folds higher).

I have no experience with this myself, but for morphology I think it is a very nice model, being somewhat transparent. I know there are some methods to measure muscle force, but how reproducible these are and how they correlate with human muscle I don’t know.

Great question – I’ve asked myself this as well multiple times. Temperature in the muscle goes up a lot during exercise (exceeding 40 degrees), and in humans, temperature is known to affect recovery after exercise as well. I think many simple questions need answers in this field, so potentially a very promising avenue.

I believe the p62 aggregates in the aged /S6K background is due to a mix of autophagy inhibition and increased rates of protein synthesis. Rapamycin acts on both, so it is hard to untangle them. There has been in vitro work showing that TSC1/2 ablation leads to impaired protein quality due to high translation rates. That being said, in our model my guess would be that it is a mix of both.

I know this was reported for multiple weeks of rapamycin treatment, but we don’t see this in muscle. It might be a tissue specific thing. When we have the genetic overexpression, we actually see a reduction in the activation levels of the endogenous Akt, making it very difficult in these conditions to evaluate this negative feedback loop.

Good question, and one we don’t have an answer for. What is very clear is that activation of Akt leads to a very quick and significant glucose uptake. For example, fibers transfected with caAkt can also be identified by staining for glycogen content. How the intracellular signaling feeds back on the insulin sensitivity we don’t know though.

The short answer to this question is ‘we don’t know’. It is also important to point out though that there are other mediators of adult muscle mass, so particularly under homeostatic conditions, perhaps basal mTORC1 signaling is compensated in part by other signals. That being said, chronic mTORC1 reduction and subsequent hyperactivation of Akt clearly have a strong effect on FoxO.

You are correct, muscle function is so much more than just force. I feel that our assays for measuring force decline after electrical stimulation and are not great to look at fatigue, a very complex phenomenon. Also, in part due to its complexity, we have not given the right attention to these aspects yet. We are now working on a story linking mTOR signaling more to fiber type, but we don’t know yet how this works exactly.

We have started to look into this, but so far without any clear results. There are some nice studies from the group of Ken Walsh who showed some Akt-dependent secretome which could be of your interest.

Good question, but I don’t know to be honest. I think that part of the force reduction is more pronounced when working with genetic mutants (so from birth). These mice also show hyperplasia, something never seen in the adult (as far as I know). If you intervene on the myostatin/activin axis in the adult, for example by using a soluble activin receptor decoy, you get growth with an increase in force (at least partial), suggesting that timing is very important in stimulating growth and its effect on increased force production.

Definitely – we found, for example, that overexpression of Akt in muscle fibers can prevent/reduce the number of denervated fibers in cancer cachexia, suggesting that there is very active muscle-to-motor neuron communication. Rapamycin, being administered systemically, clearly has beneficial effects on the motor neuron.

When we look at muscles without S6K or Raptor, the functional deficit occurs, at least in part, on the level of the contractile apparatus (reduced force in skinned fibers). So while we cannot exclude a loss of innervation as part of the problem, particularly in the hypertrophic Raptor KO, it is clearly not the main problem. We never observed polyinnervation.

We didn’t check any mTORC2 targets because we didn’t really focus on endogenous Akt activation (which as I mentioned is a bit altered in the transgenic mouse, independently from rapamycin).