8x Increase in "Mitochondria Building" Protein PGC1-Alpha W/ Medium Intensity Exercise in Glycogen Depleted Elite(!) Cyclists: Training Revolution or Recipe for Disaster?

With only 2-7x increases in PGC1-alpha expression HIIT seems to lag behind compared to this "eat low, train low, gain high" strategy, but not every protein essay that glitters in the petri dish will turn into Olympic Gold in the real world ;-)
As a diligent student of the SuppVersity you should by now have at least a preliminary understanding of how the adaptive machine you call your "body" adapts to the various nutritional and physical challenges most people subsume under the all-encompassing and pretty nondescript terms "diet" and "exercise". Against that background it should not really come as a surprise that researchers from the The Swedish School of Sport and Health Science are soon going to publish the data of an experiment that shows that even (you could probably also say, in particular, although respective evidence is still missing) highly trained athletes can benefit from exercising in a glyocogen depleted state - at least if the yardstick you use to measure the "benefits" is an increase in mitochondrial biogenesis (Psilander. 2012).

Train high, eat low (carb), train low and...?

To elicit the differential effects of 6x10 min bouts of cycling at 60% of the individual VO2max (4min of active rest in between) with normal vs. depleted skeletal muscle glycogen stores, Psilander et al. had their 10 highly trained male national elite level competitive road cyclists and mountain bikers (27.8±1.6 years, 74.7±2.0 kg, 183±2 cm, and 4.9±0.1 l/min VO2Max) perform an 8x4min interval training at 88% of their individual VO2Max ~16.5h before they had to report back at the laboratory on the actual testing day (the intervals were seperated by 4min of active rest, i.e. cycling at 100W+).
Figure 1: Graphical outline of the experimental protocol and its effect on the glyocogen stores of the from the vastus lateralis muscle (based on Psilander.. 2012)
The protocol (see figure 1) was repeated twice, with adequate time in-between and in random order, with the subjects consuming water only and low carbohydrate meals
  • low carb meals (LC) were eggs and bacon (0.02 g CHO, 0.6 g protein and 0.8 g fat/kg bw) for dinner and breakfast, providing a total of of <0.04 g CHO, 1.2 g protein and 1.6 g fat/kg bw
before the glycogen depleted trial (LC) and high carbohydrate beverages (maltodextrin-dextrose powder Carbo 134 w/ 1.0g CHO/kg bw) + high carbohydrate meals
  • high carb meals (HC) were pasta with meat sauce and lemonade for dinner (1.83 g CHO, 0.53 g protein and 0.14 g fat/kg body weight bw) and oatmeal and orange juice (1.54 g CHO, 0.31 g protein and 0.12 g fat/kg bw) for breakfast and additional bananas with beverage 3,5,7 and 8 for a total of 12.6 g CHO, 0.9 g protein and 0.3 g fat/kg bw
before the glycogen repleted trial (HC).

... get impressive increases in PGC1-alpha, but no AMPK response at all!

As the data in figures 1 & 2 goes to show you the nutritional intervention was not without effect the factual glycogen levels (figure 1, right) and the glucose, insulin and fatty acid levels before and after the workout (figure 2) - and, as you would expect it, the corresponding changes in gene and protein expression in the muscle samples the researchers collected approximately 15 min before the depletion (S1) and test exercise (S2), as well as 3 h after the test exercise:
Figure 2: Free fatty acid levels before depletion (S1) and before (S2) and after (S3) exercise trial, as well as PGC1-alpha and p-AMPK expression (data calculated based on Psilander. 2012)
Now what you probably won't have anticipated, though is the absence of the expected p-AMPK response to exercise in the low glycogen (LC) trial.
"The mRNA content of the master regulator of mitochondrial biogenesis (PGC-1a) was not changed 14 h after depletion exercise (pre-test exercise) but was significantly increased 3h after the test exercise in both conditions (Fig.2). The increase was, however, much more pronounced in LG than in NG (8.1-fold vs. 2.5-fold, P<0.01). The mRNA content of two other regulators of mitochondrial biogenesis (PRC and Tfam) also increased significantly but with no difference between conditions (time-dependent effect, Py0.01; [not shown in my graph]). The mRNA content of genes for oxidative metabolism enzymes (PDK4 and COX I) only increased after LG with a significant difference between the two conditions. The mRNA content of CS, Sirt1, NRF1 and PPAR[-delta] did not change under any conditions." (Psilander. 2012)

Almost 8x elevated levels of PGC-alpha but no change in the "fat burning, GLUT-4 pomoter" AMP-activated protein kinase? How can that be? The answer to this question is actually pretty simple: If the phosphorlyation of AMPK changes in response to changes in the ATP to ADP ratio (the name is misleading, here as scientists have initially believed that the main determinant was the ATP to AMP ratio, which is yet not the case), it should be obvious that it won't change, if the ATP levels are already so low that at most the ADP to AMP, but not the already rock bottom ATP do AMP ratio will be changing.

What happens if your body senses that it cannot fuel his energetic demands with glucose?

In the presence of borderline hypoglycemic glucose levels (the normal range starts at 4.4 mmol/L; after the depleted test the subject were at 4.3 mmol/L!) your body would be ill advised to increase glucose uptake. So if this is not an option the only way to make up for the lack of energy are fatty acids. Unfortunately the amount of fatty acids your skeletal muscle can oxidize is strictly rate-limited by your mitochondrial capacity ... now, I am asking you what's the "natural", the logical and in the case of the 10 cyclists in the study at hand also the factual reaction that will get you out of this mess? Right! To build more powerful mitochondria and thus widen the "bottle neck"! And what's going to do just that? Yeah! The ~8x increase in PGC-1alpha expression you see in figure 2. 

Practical implications: From protein essays to results?  

 Now that we have gotten the mechanisms straight, there is but one question we have to answer - what does that mean for you? When and for whom does it make sense to train with depleted gycogen stores? And in an even broader context - what does that tell us about low-carbing and (intermittent) fasting?
  1. Before you even consider making this a staple of your regimen, I would encourage you to read the whole SuppVersity Athlete's Triad Series
    Even (or especially?) for trained athletes competing in largely aerobic sports, training in a state of depleted glycogen store can serve as a viable tool to elicit even higher (2-7x; cf. Gibala. 2009, Nordsorg. 2010. Psilander. 2010) increases in increases (8x!) PGC1-alpha and (allegedly) mitochondrial biogenesis as you would see them in response to high intensity interval training at much lower intensities (but correspondingly longer durations). 
  2. Training in a fasted state does not per se guarantee / put you at risk of being glycogen depleted, neither does intermittent fasting and or "training on empty". As long as you replete your glycogen stores after your workouts you won't see similarly pronounced increases in PGC1-alpha in response to "regular" aerobic training at a low intensity. You will, on the other hand, still see increases in AMPK and, what's even more important, you will be able to perform at much higher intensities! A fact that is particularly important for the strength trainees out there.
  3. While it may make sense on occasion, and merely based on it's beneficial effects on purported  mitochondrial biogenesis (I don't have to remind you that we don't have any information on whether the increase in PGC1-alpha did even translate into an increase in mitochondrial biogenesis in the absence of adequate glycogen / ATP levels!), I want to reemphasize the scientists very hint that "[l]ongitudinal studies examining protein levels and performance are required" before it can be recommended to include this practice as a staple into your routine!
  4. Life is to complex for black-and-white thinking, and so are AMPK, mTOR & co! Learn more in the Intermittent Thoughts.
    Long-term exercise in a glycogen-depleted state without adequate carbohydrate intake and thus glycogen repletion is not for nothing one of the causative factors of the athlete's triad (see Part I & II of the SuppVersity Athlete's Triad Series). I would therefore be very surprised if the long-term outcomes of low-carbing + (intermittent) fasting w/out regular glycogen repletion would be anything but negative, regardless of its beneficial effects on PGC1-alpha. After all, the study at hand clearly shows that you will also be missing out on the benefificl effects of increased p-AMPK expression of which you know based on what you have read in the Intermittent Thoughts on Intermittent Fasting Series that it is one of the, if not the central argument in favor of intermittent fasting.
The practical take home message of this study is therefore that exercise + diet induced targeted glycogen depletion before a workout (not via an overnight fast, only; that would leave your muscle glycogen stores largely intact, while your body is burning fat and tapping into your hepatic glycogen reserves) can become one among a whole host of tools in your workout-toolbox. You can use it sporadically, but you should not need another study to be able to predict that the downsides of chronic use are going to outweigh (purported - again, we are measuring markers only, here!) short term benefits.

On a last note: I guess you know that the SuppVersity is the place where you will hear about respective longitudinal data first, right? To make sure you don't miss that I suggest you go to www.facebook.com/SuppVersity like the page or register for updates at twitter.com/ProfDrAndro!

References:
  • Gibala MJ, McGee SL, Garnham AP, Howlett KF, Snow RJ, Hargreaves M. Brief intense interval exercise activates AMPK and p38 MAPK signaling and increases the expression of PGC-1alpha in human skeletal muscle. J Appl Physiol. 2009 Mar;106(3):929-34.
  • Nordsborg NB, Lundby C, Leick L, Pilegaard H. Relative workload determines exercise-induced increases in PGC-1alpha mRNA. Med Sci Sports Exerc. 2010 Aug;42(8):1477-84.
  • Psilander N, Wang L, Westergren J, Tonkonogi M, Sahlin K. Mitochondrial gene expression in elite cyclists: effects of high-intensity interval exercise. Eur J Appl Physiol. 2010 Oct;110(3):597-606. Epub 2010 Jun 23.
  • Psilander N, Frank P,  Flockhart M, Sahlin K. Exercise with low glycogen increases PGC-1agene expression in human skeletal muscle. Eur J Appl Physiol. 02 Oct 2012 [ahead of print]