Intermittent Thoughts On Intermittent Fasting - Exercise (3/3): How Training Solves the AMPK/mTOR Antagonism.

Image 1: Just like Two-Face, a character from the Batman comic books, AMPK turns out to have two faces,... ah I mean isoforms the differential expression of which explain why exercise, contrary to starving yourself, maintains or even builds muscle mass while reducing your love handles (img batman.wikia.com).

In the last installment of the Intermittent Thoughts on Intermittent Fasting series, we have revisited the idea of different training modalities, i.e. endurance and strength training, for the promotion of AMPK-related reductions in body fat and mTOR-dependent increases in muscle mass. We have also busted the long-standing myth of the "anabolic window of opportunity", which, upon closer examination, turned out to have the size of a barn door (>24h) that is unlocked with the key of exercise and nutrition sciences. Related findings showed that even in the absence of additional nutritional stimuli a single intense strength training session led to a profound and (>24h sustained) increase in mTOR phosphorylation in 24 untrained, young, healthy, male subjects (Vissing. 2011). In conjunction with the results of Burd et al. (Burd. 2011), who found that the beneficial effects of strength training on the subsequent response to protein feeding depend on exercise intensity and volume and last for >24h, these results further underline the synergistic effects the fasting, training, feeding cycle of classical intermittent fasting regimens had and still has on the health and physiqueof its practitioners.

Unfortunately, both the concept of "fat loss", as well as that of "muscle gain" are still largely associated with notion of what is commonly referred to as "energy balance". If you read my recent blogpost on the  "High(er) Reps for Fat Loss"-Myth, you will be aware of the fallacy behind the idea of "going to the gym to burn fat". And while more and more trainees (also thanks to the educational work of BodyRX Radio ;-) are getting the idea that you have already lost the fight against your love handles, when you go to the gym solely "to burn calories", the notion that you go to the gym to either "pump up" or "totally exhaust", "damage" and "break down" muscle tissue is similarly illusive. Contrary to what the more is more mentality of the western society may suggest, simple linear causality is nothing you will ever see as the underlying "reason" for the success of a given exercise regimen.

Gain muscle or lose fat? AMPK vs. mTOR and the unique effect of exercise
 

Image 2: "Immunocytochemistry/ Immunofluorescence - AMPK alpha 1 + AMPK alpha 2 (phospho S485 + S491) antibody (ab39400)" ... and if you do not understand this lingo, what you see here is nothing else but one of the unspecific markers for both isoforms of AMPK that is used in most of the studies (img abcam)

Regardless of whether you intend to lose fat, to build muscle or strength, the previous installments should have made it pretty clear that you will always be dealing with two-way processes, or I should say cycles. Now, interestingly enough, exercise, contrary to dieting or overeating, appears to have the unique quality of driving both at the same time - fat loss and protein synthesis, AMPK and mTOR. This works, and this is going to be the main message of this concise piece of the Intermittent Thoughts series, because the exercise induced muscular(!) AMPK-response differs from the one your brain and many other organs will exhibit, when you starve yourself during a diet. Actually we have been knowing for quite some time that the predominant isoform of AMPK that is expressed during exercise is AMPK-alpha2. Back in 2000, already, Wojtaszewski et al. found that "high" (in this case >70% of the individual VO2max) intensity exercise for 60min selectively increased AMPK-alpha2 activity almost threefold (Wojtaszewski 2011). Similar to the results of previously discussed studies, the increased AMPK levels returned to baseline within 3h after exercise-cessation.

Unfortunately, only few of the subsequent studies, which investigated the effects of different exercise regimen, used iso-form specific tests to determine which of the two AMPK isoforms was expressed consequent to the respective training protocols. According to the ground-laying work of Stapleton et al. (Stapleton. 1996) and supported by a study by Stephens et al., it is yet likely that the relative exercise-induced expression of AMPK-a1 in human muscle tissue is negligable.

Figure 2: AMPK-a2 expression (arbitrary units measured in the absence of AMP) and fat oxidation in g/min in 7 healthy individuals during 30 minutes cycling at 62.8% of VO2Max (data adapted from Stephens. 2002).

Moreover, the results of Stephens et al. underline that the exercise-induced increase in AMPK-alpha2 does not only increases fatty acid oxidation, but that both exhibit an excellent correlation with exercise induced glucose depletion (Stephens. 2002).

Figure 2: Glycogen content (mmol/kg) and phosphorylation of AMPK (arbitrary units) in human vastus lateralis muscle before (0 min) and at the cessation of 120 min of one-legged knee-extensor exercise, while consuming either a glucose containing drink or a placebo drink.  (data adapted from Thorbjorn. 2006)

It is thus not surprising that Thorbjorn et al. were able to show that the ingestion of 0.7 g of glucose/kg of body weight/hour did not only blunt the exercise induced AMPK-a2 response but also reduces its beneficial effects on fat oxidation by -47% (cf. figure 2)!

The results of older studies sometimes begin to shine in the light of novel findings 

Now, you probably knew all that before - after all we have been talking about this effect, its beneficial effects on fatty acid oxidation and glucose uptake, as well as its supposedly negative impact on protein synthesis in previous installments of this series. And in fact, these results begin to shine only, in the light of the results of a a recently published study by Mounier et al., who were able to show that only the increased expression of the alpha1 isoform of AMPK, but not AMPK-alpha2 does impair mTOR signalling. Against that background, the systemic antagonism of AMPK-alpha1 (expressed in liver, brain, and other organs) and mTORc1 mediated protein synthesis stands in stark contrast to the metabolically highly beneficial synergism of concomittant exercise-induced AMPK-alpha2 and mTORc1 expression.

To make a long story short: Exercise is unique in its ability to help you shed fat and build muscle "at the same time", because it activates a specific isoform of the "starvation sensor" AMPK, which does not block the concomitant increase in protein synthesis subsequent to the (likewise) exercise-induced increase in mTOR phosphorylation. On that note, my schedule forces me to end this abbreviated version of the Intermittent Thoughts, yet not without the promise that I am finally going to tie all the knots together in the next installments of this series.

Intermittent Thoughts On Intermittent Fasting - Exercise (2/3): Opening the "Anabolic Barn Door" With the Key of Exercise and Nutrition Science!

Image 1: The "anabolic window" turns out to be more of a barn door, which is unlocked by the key of exercise and nutrition science (Random House Books)

Looking back, the main take-aways from the last installment were the dependence of exercise performance on adequate and not so much constant energy supply, as discussed in the context of the Ramadan fasting soccer players, the increased AMPK response to fasted training on a hypercaloric diet, which would suggest that things like "fasted cardio" in the morning could well have it's place in an intermittent fasting regimen even when you are bulking (in order to ward off fat gains), and, last but not least, the differential AMPK- and p70S6K protein synthetic response of cyclists and powerlifters to unaccustomed training stimuli. Accordingly, a versatile training routine that is timed in a way that allows you to train fasted or semi-fasted training, i.e. having your first easily digestible high protein meal / supplement ~30min-1h before you hit the gym, will certainly help with lean gains and muscle-sparing fat loss.

How to train if someone "just wants to look good naked"?

While the observations of the Coffey study (Coffey. 2005) did underline the importance of versatility, or, I should say constant "novelty", or at least modification of the training stimuli, they did not really provide any clues on how someone, who "just wants to look good naked" (and I assume this applies to the majority of non-athletes, today) should train to transform his formerly at best non-obese physique to the cover-model'ish look everybody is aspiring these days.

Figure 1: Study design of the Vissing study with its 10-week preconditioning phase for the strength and endurance training groups (generated based on information from Vissing. 2011)

In regard to this question, a similar, yet more recent study on non-athletes comes to mind. In the course of the latter, K. Vissing and his colleagues from Aarhus, Denmark, and Geelong, Australia, took a closer look at the response of the "AMPK/mTOR seesaw" to either endurance or strength training (Vissing. 2011) after a comparatively brief per-conditioning period of 10 weeks (cf. illustration 1) - a scenario of which we can expect more reliable results than from its "highly trained recreational athletes" counterpart from the Coffey study, where the participants have been focusing on training for their respective sport (cycling or powerlifting) for years. Accordingly, Vissing et al. expected to see that...

[...] mTORC1 signaling would be selectively activated by SE [strength training], whereas AMPK signaling would be activated by both types of exercise but to a relatively higher degree after EE [endurance exercise] compared with SE [...]

Thus, their research hypothesis was in accordance with the publicly accepted idea that only strength training builds muscle (obviously the role of mTOR-activation in this process is widely unknown in the general public), while endurance exercise would be the better form to train if one wanted to lose fat - as a diligent reader of the SuppVersity, you will obviously be aware that the reduction in adipose tissue you will hopefully observe, when you are dieting, is primarily a result of the depletion of muscular (and hepatic) ATP stores, which brings the AMPK energy emergency police on the scene which will concomitantly tell your muscles to suck up all extra (i.e. more than your brain needs) glycogen from your blood stream and kick your adipocytes' asses, so that they release some of their fatty energy reserves as metabolic firewood for your mitochondria.

I hope you remember "The 'hungry' side of neuronal AMPK activation", i.e. the differential effects of AMPK phosphorylation in reaction to energy shortage in muscle or liver tissue vs. its effects in the brain. If not, I suggest you (re-)read the respective passage in "AMPK III/III: Natural Rythmicity for Maximum Fat & Minimal Muscle Loss", as a thorough understanding of this difference if of utmost importance if you want to be able to compare and interpret the data from various studies correctly.

The Coffey study (discussed in the last installment) did however show that this assumption, i.e. both endurance, as well as strength training will always increase AMPK, does not hold true, when we are talking about highly trained athletes - neither in the cyclists nor in the powerlifters from the Coffey study did engaging in their respective discipline produce statistically significant increases in AMPK phosphorylation.

Figure 2: AMPK phosphorylation (0, 2.5, 5 and 22h post) and approximate area under the respective curces (small graph) during post-exercise recovery from single-bout exercise, conducted with an exercise mode to which the exercise subjects were accustomed through 10 weeks of prior training (data calculated based on Vissing. 2011)

Conversely, in the Vissing study, AMPK phosphorilation did transiently increase in both the strength and endurance trained groups immediately post (at 0h) exercise (cf. figure 2). However, with the subsequent drop of the phosphorylated AMPK (pAMPK) below the values of the control groups, the estimated area under the curve (AUC; I simply used weighed averages for the calculation), i.e. the absolute AMPK phosphorylation over the whole 22h post-exercise window, for which the scientists have data (cf. figure 2, right), was -12% and -17% lower in the strength training group than in the control and endurance group, respectively.  

Without the AMPK elevation of an intermittent fast (or calorie reduction), it is thus unlikely that strength training alone is going to trigger significant AMPK responses.

Interestingly, the scientists state that the protein expression "of any of the reported signaling proteins" was "not altered" by the 10 weeks of pre-training, which would indicate that, contrary to years of competitive endurance exercise (cf. cyclists in illustration 1 in previous installment), 10 weeks with three weekly sessions of combined steady-state and interval exercises on stationary bikes do not blunt AMPK phosphorylation in response to 120 min of bicycle exercise at 60% of the individual VO2 max.

The induction of mTOR phosphorylation is and will remain the real strength of strength training

Likewise, the protein synthetic response (as evidenced by mTOR and p70S6K expression) did not change in response to a 10-week pre-conditioning phase comprising 30 leg workouts (3 exercises; 3-5 sets; 10 reps in the first 15 sessions, 4-6 reps in the last 15 sessions). Interestingly, and contrary to the often heard assertion that mTOR phosphorylation would be a strength training exclusive, figure 3 shows that there is still a minor, yet over the course of the post-exercise period, non-negligible increase in mTOR phosphorylation in the endurance trained subjects, whose 45min cycling session effectively blunted the mTOR dephosphorylisation the control group, who, just like all of the previously (before the preconditioning) 22 untrained healthy male subjects (79.1 kg; 182 cm; 23.3 years), fasted for the first 5h "post exercise" (their exercise consisted of sitting on the couch, doing nothing ;-).

Figure 3: mTOR phosphorylation (0, 2.5, 5 and 22h post) and approximate area under the respective curces (small graph) during post-exercise recovery from single-bout exercise, conducted with an exercise mode to which the exercise subjects were accustomed through 10 weeks of prior training (data calculated based on Vissing. 2011)

Even without looking at the data in figure 3 it should be obvious that the meager increase in mTOR phosphorylation in the endurance group cannot compete with what we see in the strength trained subjects, whose p-mTOR ( = phosphorylated mTOR) levels skyrocket in the post exercise phase, peaking at +218% (control: 56%; endurance: 130%) not immediately or maybe 1h post exercise but 5h after. Thus, the purported "anabolic window" of 1-2h after a workout turns out to be a barn door, in the real world - a barn door which is wide open right in the middle of your intermittent fasting feeding window!

Strength training = opening the "anabolic barn door"

Yet, while we do now know how to unlock the barn door, we still do not know if there ain't a way to push it open even further / faster, and how to keep it wide open for as long as possible. In this context, a study by Burd et al. from Steward Phillips group at the Department of Kinesiology of  McMaster University in Hamilton, Ontario (Burd. 2011) could provide further clues into the "optimal" way(s) to push the "anabolic barn door" open, as wide as possible.

After all that has been said about the over-expression of mTOR in our current society in the previous installments, it should be said that the problem does not lie with mTOR itself, as it is not the latter which inhibits AMPK, but the energy abundance that triggers the mTOR response in our western obesity scenario. This chronic nutritionally induced suppression of AMPK is something we need to distinguish from both the training-induced increase in mTOR phosphorylation and the temporary and strategically used dietary stimuli that are so characteristic of intermittent fasting.

Figure 4: If we disregard the nutritional component, the training induced "anabolic barn door" does not only coincide with the feeding window, it would also keep you nicely "anabolic" in the course of the fasting period.

In figure 4, I have extrapolated the missing two hours to complete a 24 hour intermittent fasting period, in the course of which you would do your training session early in the morning, head towards the gym at 8:00am, change your clothes, warm up, training for about an hour and break the fast at 10:00am. Thereafter, you would have a pretty long feeding window of about 6 hours, to then begin another fast... in that, your meal pattern would differ profoundly from the one of the study subjects, because the latter had to fast for the first 5 hours post exercise, so that the mTOR response was not augmented and the study results distorted by meal ingestion (afterwards they were allowed to eat whatever they wanted until 22:00pm and had to report back for the 8:30am blood draw (mTOR still +89% elevated) on the next morning. Due to these differences it is difficult to predict how your overall (i.e. exercise + food induced) mTOR response would look like on the above regimen.

Will the "anabolic barn door" stay open in the course of the fast and thusly prevent muscle breakdown?

This is where the data from the Burd study comes into play (Burd. 2011). In their study, Bird et al. had measured the fractional protein synthesis rate in response to feeding (15g of whey protein) and feeding and exercise (unilateral leg raises) at different intensities, i.e. 90% 1RM to failure, 30% 1RM with matched work-load and 30% 1RM to failure. What they found was that

regardless of condition, rates of mixed muscle protein and sarcoplasmic protein synthesis were similarly stimulated at FED and EX-FED (Burd. 2011)

- an observation, the scientist attribute to the fact that the sarcoplasmic constituents of the muscle may be more susceptible to hydration flux, so that the results may not adequately represent the "actual" protein synthetic response.Thusly, the researchers rely in their interpretation of the data mainly on the myofibrillar protein synthesis rate (cf. figure 5).

Figure 5: Changes (% per hour) in absolute myofibrillar protein synthesis (adapted from Burd. 2011)

As you would expect and actually can see in figure 5, the latter did respond to the additional exercise stimulus. Pumping away at 30% of your 1RM max without going to failure, is yet not enough to augment the statistically hardly significant increase in fractional protein synthesis that was triggered by protein ingestion, alone. It takes some effort, or, in other words, heavy weights and training to failure to trigger elevations in AKT phosphorylation (90% 1RM to failure) or mTOR phosphorylation (30% 1RM to failure) to get that done (note: neither of the two, i.e. protein kinase B = AKT or mTOR was significantly elevated by feeding, alone).

[...] protein ingestion stimulated rates of myofibrillar protein synthesis above fasting rates by 0.016 ± 0.002%/h and the response was enhanced 24 h after resistance exercise, but only in the 90FAIL and 30FAIL conditions, by 0.038 ± 0.012 and 0.041 ± 0.010, respectively. Phosphorylation of protein kinase B on Ser473 was greater than FED at EX-FED only in 90FAIL, whereas phosphorylation of mammalian target of rapamycin on Ser2448 was significantly increased at EX-FED above FED only in the 30FAIL condition.(Burd. 2011)

Moreover, and this may be of even greater importance in the context of exercising on an intermittent fast, muscle protein synthesis stayed elevated way beyond what is usually considered the <4h "anabolic window".

Our results suggest that resistance exercise performed until failure confers a sensitizing effect on human skeletal muscle for at least 24 h that is specific to the myofibrillar protein fraction. (Burd. 2011)

While this is obviously important for everyone who wants to accrue as much muscle muss as possible, any elevations in protein synthesis will also help a dieter to keep is hardly earned muscle, because in essence our muscles are continuously build up and broken down  - proteolysis, i.e. the breakdown of muscle tissue, and protein synthesis are going hand in hand and it is the ratio of one to the other, which decides whether we are in an "anabolic" (synthesis > breakdown) or catabolic (breakdown > synthesis) state. Consequently, any elevation in protein synthesis will ameliorate muscle loss - no matter how proteolytic a dieter may become during the fasting phase.

It takes >24h for the barn door to close itself - use this time to get rid of fat, not muscle

Fine, we unlocked the "anabolic barn door", it stays open for "at least 24h"... blah blah... wtf! how does all that translate from the metaphorical into the real world of intermittent fasting? Well, the answer is pretty simple, as hundreds of trainees have been practicing exactly that with extreme success over the past couple of months:

1. fast until min. 1h before your training
2. spike your protein synthesis with a protein shake (~20g of whey), EAAs (~10g) or BCAAs (~8g)
3. train semi-fasted and heavy
4. feast within a 5-8h window
5. repeat the same litany again

Now, the sheer size of the barn door, ahm... sorry, the long-lasting anabolic and thusly anti-catabolic effect of intense strength training should allow you to either skip or replace "3. train semi-fasted and heavy" with "3. passive or active recovery" (in that case you also do not want to ingest the protein shake / EAA / BCAA) or even some "3. semi-fasted cardio" (see notes in red box) if you feel that your conditioning or weight loss will benefit from that, every other day without running the risk of either gaining too much fat weight.

Image 2: Your "anabolic barn" is huge enough to accommodate one or two steady state, low intensity or high intensity "cardio" sessions per week.

If you want to incorporate "cardio" training into your routine, the pre-conditioning protocol from the Vissing study could actually be a very good, since diversified, regimen. In that, you would cycle between doing "standard" steady state conditioning work, longer medium-intensity interval training and short, but intense HIIT sessions. The result would be a very complete "cardio" protocol, of which the Vissing study showed that it will help you ramp up your AMPK levels pretty profoundly, even if you are only sitting on one of those cycle ergometers pedaling away jovially at 60% of your VO2 max. And in case you are now concerned about possibly shutting the barn door - look at figure 3 again, the mTOR response to this kind of exercise may not be earth-shattering, but a plus of 25% @5h post exercise is better than what you would get if you just lay around lazily, as the control group in the Vessing study did.

With these insights into why that of which you already knew that it works actually works, I conclude this week's installment of the Intermittent Thoughts and hope that I did not bore you so much that you do not come back next Sunday for another installment of this series ;-)

Intermittent Thoughts On Intermittent Fasting - Exercise (1/3): Cycling, Powerlifting and Lean Gaining ;-)

Image 1: It may be more effective than your usual "eat half as much diet", but even with intermittent fasting exercise is compulsory, not facultative.

In the last installments of this series we have analyzed the natural interplay between AMPK and mTOR, have learned that chronic over-expression of either of the two can be detrimental to the way you look, feel and perform and have scratched on the surface of how intermittent fasting and the use of AMPK and/or mTOR promoting supplements can restore and amplify the natural up and down on the AMPK/mTOR seesaw and thus promote fat loss and and gains in lean body mass - not at the same time, but cyclically. In this episode it is high time to take a closer look on how exercise, the one and only true "body recompositioning agent", plays into this.

Tell me who you are and I tell you how your body will respond to exercise.

What we already know is that the exercise induced depletion of intra-cellular ATP and the corresponding increase in ADP and AMP levels will produce profound(!) increases in AMPK phosphorylation. In a recent study on the effects of a 30s Wingate test (a sprinting test on a cycle ergometer with breaking loads equivalent to 10 and 8% of body weight for men and women), for example, scientists from Gran Canaria found increases in AMPK phosphorylation vs. baseline of +495% - 98% for the ten women in the study and +278% - 33% for the 17 (cf. figure 1)

Figure 1: Relative changes in AMPK phosphorylation in response to 30s Wingate sprint test in 17 male and ten female subjects (data calculated based on Fuentes. 2011)

As you can see there is a huge (and statistically significant) gender difference in the initial AMPK(-alpha) response to sprinting, the difference at 30min and 120min post exercise on the other hand did not reach statistical significance (p<0.05). In an aerobic exercise scenario (90min at 60%VO2max), however, Roepstorff et al. PK came up with the exact opposite results (Roepstorff. 2006):

A 198% increase (P < 0.001) was observed from rest to 90 min of exercise in men, whereas in women the exercise-induced 74% increase in αAMPK Thr¹⁷² phosphorylation was only borderline-significant.

The different fiber-composition of the male and female subjects could provide an adequate explanation for this phenomenon. With a +23% higher ratio of slow twitch type I to fast twitch type II fibers, the women in the Roepstorff study were, on a pound per pound base, more effective endurance athletes than the men. Consequently, they did not run out of fuel so fast and thusly there was no need for their bodies to ramp up AMPK by the same 198% as the bodies of the men did.

Figure 2: Fat oxidation (in kcal/kg per min) calculated based on respiratory exchange ratio in male and female participants of a 90min cycling bout at 60% of their individual VO2Max (Roepstorff. 2006)

It is important to note that despite lower AMPK activiation in the female participants of the Roepstorff study, the women (due to their high ratio of type I fibers with +25% higher capillarization) had statistically significantly higher fatty acid oxidation rates (i.e. they burnt more fat) than their male counterparts (cf. figure 2)! This goes to show you that the metabolic scene, which is set by AMPK, is not the sole determinant of substrate metabolism. In the end, the capacity of the mitochondrial furnaces decides how much energy from fat you will be able to burn. For subjects with a high amount of type I fibers whose primary objective is to actively burn additional fat calories, aerobic exercise may thus well be a viable alternative for intense HIT regimens, of which Astorino et al. have recently shown that the rate of fatty acid oxidation in recreationally active men and women is identical within the statistical margins (Astorino. 2011).

Conversely, the higher type II to type I fiber ratio of men makes them better sprinters. That and their overall larger muscle mass could have allowed them to perform the 30s sprint on the cycle ergometer without having to resort to extra-muscular energy stores, which would explain why we did not see a significant increase of AMPK immediately after sprinting. Its occurrence 30 minutes after the sprinting exercise does yet go to show that even very short exercise bouts can trigger pretty profound AMPK responses, of which I would speculate that they facilitate post-exercise glycogen repletion via "energy-repartitioning".

Why cyclists should powerlift and powerlifters should cycle

Illustration 1: Differential response of cyclists and power lifters to strength and endurance training; statistically significant increases are highlighted in green, statistically significant decreases in red (data compiled from Coffey. 2005)

With regard to the differential response to different exercise modalities we also know from a 2005 study by Coffey et al. (Coffey. 2005) that muscle from strength-  and endurance-trained individuals respond very differently to endurance (1 h cycling at 70% VO2peak) or resistance training (8 sets of 5 maximal repetitions of isokinetic leg extensions).

I've gone to all the bother of compiling the extensive data on muscle protein synthesis and related signalling proteins from the study into a single chart (cf. illustration 1), where statistically highly significant increases are highlighted in green and statistically significant decreases are highlighted in red. Thusly, you should be able to see that if the goal is to increase AMPK, cyclists have to strength train, while powerlifters will have to get into the saddle of an elliptical or ergometer.

What appears paradoxical at first, is the result of adaptation processes: Only novel and unaccustomed stimuli trigger further adaptation... and "novelty" is such a profound trigger of adaptational responses that - under the assumption that p70S6K phosphorylation is a reliable measure of the protein synthetic training response - cycling causes almost comparable increases in protein synthesis as resistance training in powerlifters, a group of athletes who are not exactly known for doing large amounts of "cardio" training.

Since this is not exactly "intermittend fasted" related, I leave it up to you to interpret the rest of the data. Before I "think on", I do yet want to caution you against getting stuck in doing the same type of exercise over and over again - there is a reason that 99% percent of the figure athletes, bodybuilders, fitness models or whatever other athletes and celebrities you think have an aesthetic body, incorporate some form of aerobic training into their regimens, as well (if you read the latest posts on HIIT training, you will be familiar that "aerobic" does not always mean steady state endurance training ;-)

Training fasted? Maybe, for athletes and performance oriented amateurs.

While the previously discussed studies showed that AMPK/mTOR responses to different exercise regimes largely depend on who you are and what type of training you have conditioned your body to, it did not answer the question that appears to be preying on everyone's mind, which is "Do I do my aerobic and or resistance training in a fasted (no food at all), semi-fasted (only protein and maybe some fat), or fed state?" Or in other words: "Do I break the fast before or after training?"

Image 2: Ramadan fasting can serve as a relatively particularly well studied "model" of intermittent fasting. You can find more information about the strengths and limitations of this model, please read Part 2 and Part 3 of this series.

A brief reminder for all who of you who missed the first installments of this series and may now wonder why I am, without further explanations, referring to studies on Ramadan fasting as if it was intermittent fasting - in essence it is! This is why I have already discussed a handful of studies that investigated the effects of Ramadan fasting on Muslim athletes, in the initial installments of this series. In that context, I have also pointed out why the Ramadan "protocol" is an acceptable model of intermittent fasting and where it deviates from what we are seeing in the dietary regimens with which Adelfo and Duong get to grips with the little fat that is still left on their athletic bodies. For more information on that I would like to refer you to Part 2 and Part 3 of this series, in particular.

If you are a competitive athlete, who follows the advice of the establishment, the answer is easy - EAT, EAT, EAT! And do not even think of fasting! On the other hand, even experts openly admit that despite the fact that it is (Maughan. 2010).

often automatically assumed that intermittent fasting will lead to decrements in exercise performance. [...t]he available evidence does not entirely support this view, but there is little or no information on the effects on elite athletes competing in challenging environments.

With respect to the lack of data, we are in the fortunate position that the Olympic Games 2012 coincide with the Ramadan period from July 21 to August 20, 2012, i.e. right in the heart of the Games. Meanwhile we do yet have to resort to the little reliable data there is and of which Maughan et al. writes in another article that it "suggests that effects of Ramadan-style fasting on exercise performance are generally small." And a pretty recent study which investigated the effects of Ramadan fasting on performance and body composition of 16 young soccer players (17.4±1.2 years, 175.4±3.6 cm, 69.6±4.3 kg and 5.1±1.3 years of training experience) corroborates this assertion.

Study shows: You can improve body composition and performance if you train intermittendly fasted, but not in a fasted state 

Alpay Güvenc from the School of Physical Education and Sports at the Akdeniz University in Antalya, Turkey, assessed body composition, hydration status, dietary intake and sleep duration of his 16 male subjects, who continued their regular pre-season soccer training during the four weeks of Ramadan, on four occasions: before Ramadan, at the beginning of Ramadan, at the end of Ramadan and 2 weeks after the end of Ramadan. The training sessions were yet postponed, so that the soccer players could have a snack or meal before they took to the field - they were thus intermittendly fasting, but not training in a fasted state!

Figure 3: Relative changes in RD: running distance, RT: running time, RV: running velocity and
RV4.0: running velocity at 4.0mmol.L-1 lactate concentration due to Ramadan fasting during the pre-season preparations in 16 male soccer players (data calculated based on Güvenc. 2011)

As figure 3 goes to show, there was an initial decline in exercise performance in the first week of Ramadan (=intermittent) fasting. In the last week of Ramadan, the maximal running distance, the running time and velocity and the RV4 (running velocity at 4.0mmol.L-1 lactate concentration) had improved - only by 3%, 3%, 1% and 2% over baseline, but nevertheless statistically significantly. Now, what may be even more interesting for professional athletes is that these beneficial effects continued well into the post-Ramadan phase - how much of this has yet to be ascribed to the training regimen (remember the soccer players were in their pre-season preparation) could only be determined if half of the kids had been Christians and had served as a non-intermittendly-fasted control.

Figure 4: Changes in total body water (TBW in L), fat free mass (FFM in kg), body fat (in kg) and  the sum of skin-fold measures (in mm) due to Ramadan fasting during the pre-season preparations in 16 male soccer players (data calculated based on Güvenc. 2011)

In a similar vein, we cannot say for sure, whether there had been comparable improvements in body composition (as evidenced by the statistical significant reduction in skinfold measures, i.e. -2.2% by week 4 of Ramadan fasting; cf. figure 4), if the players had just continued their usual pre-season training without fasting intermittendly. What we can say for sure though, is that they achieved the latter without any major changes in their overall caloric intake or macronutrient composition (cf. figure 5)
And that the fasting had no negative effects on the subjects sleep duration (~8.7h) or their hydration status. So that, it would appear that during a metabolically demanding pre-season training a non-specific intermittent fast works just / at least as good as a normal diet, as long as the athletes meet their training induced caloric demands.

Training fasted? Yes, for lean gains.

Image 3: For some "lean gains" happen only in their heads (img muscle.iuhu.org)

Now, while "exercise performance" obviously is an important variable, I assume most of you who are toying around with the idea of doing an intermittent fast, are more interested in its effect on body composition and would tolerate a dip in "exercise performance" (whatever type of exercise that may be in your case) if only those love handles finally disappeared and allowed your ever-increasing muscle mass to shine... or are you interested in the potential (largely AMPK-related) health benefits intermittent fasting has to offer in a world, where nutritional abundance is a 24/7 phaenomenon and the world "bulking" is often misinterpreted as taking advantage of the former as often as possible?

In both cases the results of a 2010 study from the Human Performance Laboratory in Leuven, Belgium (not the one in Canada!) would be relevant for you. In that study (Van Proeyen. 2010), Van Proeyen et al. had 27 healthy male volunteers consume a hypercaloric high-fat diet (∼+30% kcal/day; 50% of kcal from fat) for 6 weeks. Additionally 20 of the subjects had to participate in 4 training sessions per week (2x90min and 2x60min) consisting of cycling at 70-75% of the individual VO2Max and running at 85% of the maximal heart rate. 10 of the subjects (CHO; n=10) had yet had breakfast (~90min before training; 675 kcal, 70% carbohydrates, 15% fat, 15% protein), the rest (fasted; n= 10) reported to the lab after an overnight fast.

Figure 4: GLUT4 and AMPK expression in 10 healthy male subjects before and after 5 weeks on a hyper-caloric high-fat diet with or without (control) exercise in the fasted or fed (CHO) state  (data calculated based on Van Proeyen. 2010).

As the asterisk in figure 4 indicates, the exercise induced increase in GLUT4 (responsible for muscular glucose uptake) and AMPK expression is significant (p<0.05, i.e. chances that this is only coincidence <5%) only in the subjects which trained in a fasted state. Moreover, only the group which trained in the fasted state had neither statistically significant weight increases, nor statistically significant increases in the sum of the skinfold measurements (a relative reliable marker of body fat levels). The unexercised controls and the CHO group (training in fed state), on the other hand, gained 3kg and 1.4kg body weight (both p<0.05). Interestingly, though only the control group experienced a statistical significant increase in the sum of their skinfold measures of +15.1%! (+1.1% in fasted; +5.4% in CHO).

This study for the first time shows that fasted training is more potent than fed training to facilitate adaptations in muscle and to improve whole-body glucose tolerance and insulin sensitivity during hyper-caloric fat-rich diet. (Van Proeyen. 2010).

Obviously, we are dealing with a very different situation, when an intermittent fast is combined with a caloric deficit - yet in view of the idea to use intermittent fasting as dietary strategy on a "lean bulk", the results of the Van Proeyen study could be of great importance. Not only in view of keeping the fat gains at bay, but also with regard to potential negative health effects of deliberate overeating and subsequently compromised insulin sensitivity.

A pros pos "lean bulk", I suggest you do come back next week if you want to know more about when and what to eat right after what type of workouts in order to maximize muscle and minimize fat gains when you train intermittendly fasted. For now, I wish all of you a sunny Sunday (here it is one) and an intense week at the gym, regardless of whether you train (intermittendly) fasted or not ;-)

Intermittent Thoughts On Intermittent Fasting - AMPK III/III: Natural Rythmicity for Maximum Fat & Minimal Muscle Loss

Image 1: The quest for fat loss, muscle size, health and longevity reminds me of the famous egg-laying wool-milk-sow.
(img austria-lexicon)

At the end of last weeks installment, I provided you with an extensive, yet obviously "incomplete" list of AMPK-"promoters", among which the  organosulfur compound alpha lipoic acid (ALA) turned out to be of chief interest in the subsequent enjoyable and inspiring "intellectual intercourse" I had with Banga, Dr. J, Bomb Jack and of course Lerner in the comments section of the post. Now, the fundamental question that arose from this discussion was how supplementation with ALA (or agents with similar effects on 5' adenosine monophosphate-activated protein kinase expression) could be used as a tool to promote fat, not weight loss while preserving or even building lean muscle mass and optimizing health and longevity - basically it's the quest for the egg-laying wool-milk-sow, an imaginary animal and a commonly used German metaphor for a jack-of-all-trades device. And I can assure you that an ultimate solution to this problem is just as hard to find as this mythical animal an image of which you can see on the right.

When AMPK is the good guy, mTOR must not necessary be the bad guy

We have learned in the past couple of installments that its modulatory effect on the AMPK/mTOR seesaw, which, even in your average citizen of the affluent Western hemisphere, is oftentimes imbalanced towards the "anabolic" mTOR side, these days. Against that background I have pointed towards the myriad of beneficial metabolic effects of AMPK activation, which could in fact ameliorate, if not reverse, many of the ailments that have befallen our fat-anabolic (remember without mTOR, fat cells cannot differentiate, cf. Bell. 2000, where blockage of the mTOR pathway with rapamycin inhibited adipocyte differentiation) society.

Illustration 1: Lifestyle factors like nutrition, nutrient timing, sleep etc. determine the fundamental balance between AMPK and mTOR, but supps can skew / tweak the balance

I hope that highlighting the merits of AMPK in the lat two installments of the series did not give the impression that mTOR its anabolic counterpart on the seesaw (cf. illustration 1) was "useless" or even dangerous, as the sheer amount of recently published studies on the implication of mTOR in the etiology of cancer could make you believe.

Image 2: A breast cancer cell (img from The Guardian); if you want it (not yourselves!) to live longer, don't feed it with leucine, Ladies!

The mTOR-cancer connection: In the course of these first 10 months of 2011 more than 16,000 studies have been published that focus or at leas mention the involvement of an (over-)stimulation of the mTOR pathway in cancer development. It is yet plain short-sighted stupidity to believe that measures like dietary leucine restriction (and subsequent down-regulation of the mTOR pathway) would protect you, let alone cure cancer. It is certainly correct that mTOR promotes proliferation in cancer cells, but it does so in about every other cell in your body. Moreover, studies on the effect of leucine deprivation on mTOR signalling in breast cancer patients showed that not only was "leucine restriction is not sufficient to inhibit mTOR signaling in most breast cancer cell lines", but was "associated with activation of survival molecule Akt, making leucine deprivation an undesirable approach for breast cancer therapy" (Singh. 2011). And with regard to the use of "true" mTOR inhibitors such as rapamycin, David Sabanti remarks in a recent review:

As induction of apoptosis rather than cytostasis is increasingly considered a prerequisite for an effective anticancer agent, it will be crucial to understand when rapamycin has such effects and where it does not, and to learn how to trigger apoptosis with additional therapies.

Now, the use of novel pharmacological mTOR-inhibitors aside, any dietary approach (such as leucine deprivation) would obviously prevent apoptosis, because it would act via mTORC1 inhibition to increase Akt-expression and (cancer-)cell survival (Sun. 2005; O'Reilly. 2005). Would you risk your "metabolic currency", i.e. your muscle, and sacrifice quality of life for the futile hope that you could thusly keep cancer "at bay"? I hope not, because in that case you still have not grasped the fundamental idea that in life its not about black and white, but about black and white and a balance between the two.

This takes us back to the egg-laying wool-milk sow or, without the metaphorical ornamentation, the issue of losing fat weight, while gaining muscle weight. From all you have learned before, it appears quite obvious that - intermittendly fasted or not - gaining muscle and losing fat literally at the same time is virtually impossible. On the other hand, you could obviously first milk and fleece your egg-laying wool-milk sow and then collect the eggs, it has been laying before you grabbed, milked and fleeced it. I assume you will notice that, again, this is a cycle: milk, fleece, collect eggs,... milk, fleece, collect eggs, ... fast, train, feed... fast, train, feed. As you see, nature is not very inventive in her fundamental concepts. She will always rely on her tried and proven cyclicality. Now, if we cannot escape that, we can at least try to tweak it depending on our goals. While for Mr. Supp in illustration 1 this would be only a step to the left to help Mr. AMPK to get the better of Mr. mTOR) or a step to the right to help Mr. mTOR gain control over the AMPK/mTOR seesaw. For us, in the real world, the most obvious thing we could resort to in order to achieve similar results are supplements.

When to take what to optimize fat loss and minimize muscle loss in the AMPK phase

In the comments related to the last installment, Bomb Jack suggested that taking ALA during a fast to jack the naturally high AMPK level up even more, would theoretically make sense, but he mentions that he had seen "studies about [ALA] causing lean tissue loss on the long run". One of these studies was conducted by Yi Wang et al. in 2010 on 24 month old male C57BL/6 mice (Wang. 2010). Half of the mice received 0.75% alpha-lipoic acid in their drinking water for one month, the rest of the mice served as a unsupplemented control.

Image 3: 22 month old C57BL/6 mouse, in view of the fact that mice live ~2.5 years this is already a granny; in other words, the 24 month old mice in the Wang study were really old.

How much is 0.75% in the drinking water of mice in human equivalents? To calculate that you need to know that the mice in the study weighed ~30g, that the average mouse drinks ~5.8ml per day (Bachmanov. 2002) and that a 0.75% solution means that 100ml contain 0.75g of the given solute. Now you calculate the absolute dose per day, which would be 5.8ml/day * 0.75g /100ml = 0.435mg/day divide that by 30 to get the dose per gram of body weight and multiply it by 1000 to get the dose per kg => 14.5mg/kg. To get the human equivalent dose (HED) you then use the formula I explained before and your calculator will tell you that the HED of 14.6mg/kg in a mouse would be roughly 1.2mg/kg in a human being (~95mg for the average adult).

Although the human equivalent (~95mg/day; cf. red box above) of the dose the mice received was way less than most commercially sold supplements contain and by far less than the 800mg - 1,200mg of alpha lipoic acid that has been used with positive results in many trials that involved diabetic patients, the results the Wang et al. observed were pretty pronounced. In the ALA group...

• food consumption decreased: -18% - 4.50 ± 0.30 g/d vs. 5.50 ± 0.30 g/d in control mice
• body weight decreased: -15.8% - 5.27 ± 0.62 g total body weight loss
• energy expenditure increased:  +25% - 7.64 ± 0.10 kcal/kg^0.75/h vs. 5.90 ± 0.10 kcal/kg^0.75/h
• glucose utilization increased: +10% judged by the respiratory quotient
• insulin sensitivity increased: -47% area under the curve in glucose tolerance test
• mitochondrial biogenesis increased: +138% relative abundance of mtDNA content
• PGC-1α* in skeletal muscle increased: + 80.0% 
• GLUT-4 in skeletal muscle increased: + 105.0%
  
  * PGC-1α ramps up thermogenesis, stimulates mitochondrial biogenesis, promotes the remodeling of muscle tissue, controls lipid and glucose metabolism (Liang. 2006)

And all that was accompanied by significant increases in AMPK and decreases in mTOR and P70S6K phosphorylation. In view of the results of the Wilson study, which showed that a much smaller (~40%) increase in AMPK activity went hand in hand with decrease in protein synthesis we can safely assume that the latter is (unfortunately) responsible for both the desirable increases in energy expenditure, glucose metabolism and mitochondrial biogenesis, as well as the not so desirable loss of appetite (you do not need to eat, when you run on stored substrate) and body mass (cf. figure 2).

Figure 1: Relative changes in phosphorylation status of AMPK, mTOR, p70S6K and 4E-BP1 in old mice supplemented with 0.75% ALA in their drinking water for 1 month compared to unsupplemented control (data calculated based on Wang. 2010)

On the other hand, the decrease of the amount of phosphorylated Eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1), of which Anthony et al. have shown in 2002 that its phosphorylation (which usually occurs consequent to the activation of the mTOR pathway) is not necessary for the increase in protein synthesis that occurs upon leucine administration (Anthony. 2002), was not statistically significant. This could be important, as dephosphorylation of 4E-BP1 and not decreases in p70S6K appear to be hallmark features of cancer cachexia, i.e. profound muscle loss in cancer patients (Tisdale. 2008).

Figure 2: Absolute (left) and relative (right) changes in body composition in old mice supplemented with 0.75% ALA in their drinking water for 1 month compared to unsupplemented control (data calculated based on Wang. 2010)

And indeed, if you, as a diligent student of the SuppVersity who knows about the subtle differences between absolute and relative data, do not focus on the absolute (figure 3, left) but rather on the relative (firgure 3, right) changes in body composition, you will have to concede that the mouse grandpas underwent a transformation you would usually expect to see in the course of the contest preparation for a bodybuilding show. While the mice "on ALA" did lose ~20% of their body weight, they also cut down from 18% to 9% body fat and (-60% fat mass) and increased the relative amount of lean mass from 74% to 87%. For a 300lbs bodybuilder this would be like cutting 32lbs of fat while losing "only" 16lbs of muscle, to end up 48lbs lighter and with a body fat percentage of 9% at 252lbs - not yet really stage ready, but certainly not bad, given the fact that he would not even have to diet to achieve that result (remember: his appetite would have decreased and his energy expenditure increased).

Rythmicity is the key when it comes to seesawing and supplementation 

The mouse vs. human issue aside, the Wang study has another caveat in terms of drawing conclusions regarding the use of ALA and other AMPK promoters during an intermittent fast - the mice did not fast ;-) Reason would dictate, though, that ingesting ALA with drinking water 24/7 on a dietary regimen with constant food availability would actually be a disadvantage compared to taking ALA specifically at the onset of the fast (i.e. when your last meal would be digested) to keep AMPK, which would already been raising at that point (and thus mTOR declining) maximized in the course of the fasting period. On the other hand, taking ALA or any other AMPK promoter after your workout and before your meal would appear to be plain out stupid, as intermittent fasting does only make sense if you really reap the maximal anabolic benefit from the short feeding window. Against that background, popping alpha lipoic acid post workout and thus potentially increasing AMPK appears (remember that we do not have any studies explicitly investigating the effects of taking ALA post-workout vs. at other times of the day, so that these are just yet to be validated hypotheses!) to be counter-indicated, as it would potentially blunt the mTOR and p70S6K response to exercise and food-intake.

Image 2: R- and S- isomer of alpha lipoic acid (Shay. 2010)

"Alpha lipoic acid, yeah! But can I buy the cheap racemic mixture or is it worth paying the extra bucks for the R-ALA?" Actually, it was Daniel Spasic who posted a similar question on my Facebook pinwall and thus made me look into the purported advantages of R-ALA, again. 

I mean we all "know" that R-ALA is the preferable form, but do you remember where you know that from? 

Well, me neither and so I dug back into the host of studies BASF was doing back in the 1990s, until their business department finally realized that a natural anti-diabetes drug is not only non-patentable, but could also compromise the sales of patentable pharmacological drugs like Chlorpropamide (patented in the mid-1980s; cf. DrugPatentWatch) which happens to be made of Propylamine, which in turn - you guessed it - is produced by BASF and sold to BigPharma and the Agrobusiness who use the same ingredient to produce of Prochloraz, and other well-known fungicide... but I am getting taken away, here.

Image 2: The scientific
evidence pro R-ALA
is quite conclusive;
the S-isomer, however,
could potentially negate
its benefits and is
useless, at best!

The main point is that there is clear-cut evidence that the R-isomer has not only a longer half-life and a greater bioavailability (Herrmann. 1996), but is also the "true" anti-oxidant: In a 1997 study by Streeper et al., for example, S-lipoic acid had no effect on insulin-mediated glucose uptake in obese Zucker rats, while R-lipoic acid increased the latter by +64%. In the same study chronic intake of R-lipoic acid reduced plasma insulin and free fatty acid levels. S-lipoic acid, on the other hand, increased insulin levels and had no effect on free fatty acids - probably a direct consequence of the reduced expression of glucose-transporter protein (GLUT4) the scientists observed in the S-ALA supplemented rats. This may now sound like the racemic mixture, which is in fact the ~50/50 R-ALA/S-ALA mixture you can buy for a few bucks at every GNC would be toxic - this certainly ain't the case. On the other hand, would you buy an anticoagulant with an with vitamin K in it, i.e. a product with two active ingredients with partly antagonistic effect? I don't think so...

And for those of you who like to think in images: Taking ALA right at the beginning of your short feeding window would be as if Mr. Supps from illustration 1 took a step to the left, just when Mr. mTOR turn on the right of the seesaw was about to gain momentum - Mr. Supps with his alpha lipoic acid would be a real spoilsport, then, wouldn't he? If, on the other hand, Mr. Supps knows how to play the game, he will follow the AMPK/mTOR seesaw's natural rythm, grab some alpha lipoic acid and step to the left, when, at the beginning of the postprandial phase (~2h after the last meal), it actually is Mrs. AMPK turn, wait until she has had her share and then jump to the right in order to hand Mr. mTOR a leucine-rich protein shake. By doing this, i.e. taking ALA at the onset of the fast, ~2h after the last meal and a leucine rich (post-workout-)protein shake to ramp up protein synthesis right at the beginning of the feeding phase, the natural rythm would not only be preserved, it would also be amplified. And what would translate into a more energetic seesawing in our metaphorical world or Mr. Supps, Mrs. AMPK and Mr. mTOR, should translate to improved fat loss during the fast and a pronounced muscle anabolic response to the daily "refeeds", as you may well call your meals if you compress the feeding window to <6h. If the mice in the Wang study had gotten this rythm right, they would probably have made it to the Mouse-Bodyduilding Masters ;-)

The "hungry" side of neuronal AMPK activation

Before I end, this pretty epic (in terms of the details we have covered) yet not very productive (in terms of how much ground we have made) installment of the Intermittent Thoughts, I want to briefly mention a very important and actually completely logical, since natural difference between the effects of increased AMPK phosphorylation in skeletal muscle and increased AMPK phosphorylation in the brain. While the former triggers all the fat-loss friendly adaptations you read about in this, as well as in previous Intermittent Thoughts, the later will have you forage through your fridge in no time - regardless of whether it's feeding time or not ;-)

Illustration 2: The differential role of hypothalamic, liver and skeletal muscle AMPK expression (Long. 2006)

You will certainly remember from previous installments that, in our eukarocyte ancestors, AMPK was the major cellular energy sensing mechanism and has been preserved by humans (and all other mammals) even in these days of nutritional abundance. AMPKs activation is thus a signal for your body that the respective cell is starving (to be precise that the intracellular ratio of ADP+AMP to ATP is rising) - if your brain cells are not already starving, it should therefore be obvious that, contrary to AMPK signalling from the muscle tissue, which usually means "Hey, if you want me to do some work for you, you better provide me with some energy! I see those lazy adipocytes over there have more than enough stored energy...", AMPK signaling from the brain indicates an acute emergency - after all, the biological equivalent of "women and children first" is "brain first! Who cares about the rest?" It should thus not surprise you that

• an increase in cerebral AMPK phosphorylation results in increased food intake (Andersson. 2004), while dephosphorylation decreases food intake (Kim. 2004); 
• genetically modified mice with no AMPK activity in the AgRP neurons were leaner and had an increased energy expenditure compared to wild-type mice (Claret. 2007)
• ghrelin, the hunger hormone, triggers AMPK phosphorylation in the brain and thus increases food intake (Kola. 2008)

With all that being said, you are probably happy to hear that alpha lipoic acid decreases or at least counters potential increases in AMPK activity in the hypothalamus (Kim. 2004).

Update (10/29/2011) - The two AMPK-isoforms: As Mounier et al. report in a very recent paper the two isoforms of AMPK, i.e. AMPKα1 and AMPKα2 have very distinct effects on the mTOR induced increase in muscle protein synthesis, and thusly, muscle size (Mounier. 2011). As the scientists point out, AMPKα1 plays "a predominant role in the control of muscle cell size" (meaning it prevents exuberant hypertrophy), while AMPKα2 mediates "muscle metabolic adaptation" (increased glucose uptake, mitochondrial biogenesis, etc.), of which we have learned that they are so vital for our metabolic health.

Close your eyes and lean out

On that note, I will close today's lesson with an some interesting information from a study Jonathan P. has brought back onto my radar, recently. It's a study by Dworak et al. on ATP changes during sleep (Dworak. 2010), which underlines the importance of sleep, specifically when you want to lose weight, because at the onset of sleep, the reduction in neuronal activity goes hand in hand with a surge in brain ATP levels, which (as you have learned today) will reduce cerebral AMPK phosphorylation and its negative metabolic consequences. Prolonged waking, on the other hand, has been shown to increase AMPK activity in the brain (Wigren. 2009). So, what are you waiting for? That's all for today... lights out ;-)!

Intermittent Thoughts On Intermittent Fasting - AMPK II/III: Leucine, HMB and a Glimpse on Other AMPK Modulators

Image 1: You pick a health, diet or diabetes supplement and I find the study that shows that in one way or another its effect is related to AMPK ;-)

I ended yesterday's installment of the Intermittent Thoughts on Intermittent Fasting Series on a pretty bold statement about the benefits of preworkout BCAA supplementation that would, at first sight, contradict common sense, or rather what common sense would dictate based on all you have read about the beneficial effects of BCAA supplementation on mTOR-related muscle protein synthesis (MPS) and the complementarity of mTOR and AMPK as regulators of anabolic (e.g. MPS, adipogensis ,etc.) and non-anabolic "scraping, rebuilding, recycling and repairing" processes. Since, after all, Bomb Jack, who posted a comment on last weeks installment of this series, is right: It would be logical that supplementation with BCAAs (he mentions HMB specifically) during the fast should result in dephosphorylation (~deactivation) of AMPK and thus negate its desirable effect on (metabolic) health.

And in fact, in the Wilson study I wrote about on Saturday the postprandial increase in AMPK phosphorylation, was blunted by the provision of carbohydrates, leucine or a combination of both (cf. yesterday's news) and you would assume that HMB supplementation would do the same, but the latter is - at least for chronic supplementation with low amounts (320mg/kg in rats ~ 52mg/kg in humans) of HMB - not the case (Pimentel. 2011), as the data I plotted in figure 1 clearly shows:

Figure 1: Effect one month of saline (control) or 80mg/day HMB on mTOR and AMPK phosphorylation and GLUT-4 expression in extensor digitorum longus (EDL) muscle of rats (Pimentel. 2011).

In the Pimentel study, there was, if anything, a non-significant increase in the AMPK and its purported downstream effect on GLUT-4 mediated glucose uptake  - both of which common sense would have told us to be compromised by HMB supplementation. While the lack of information on the "timing" or, more specifically, the interval between the last feeding and the intragastric administration (gavage) of 320 mg/kg body weight of HMB is a drawback in view of the significance of these results in an intermittent fasting context, rats usually eat at night and thus the administration of the 80mg of HMB (the rats weighed only 250g) "daily at the same time (during the light period)" will probably have coincided with a "fasting" period.

How can we explain that mTOR expression increased, while AMPK remained constant?

Are the different result a consequence of the metabolic magic of HMB? Well, before we analyze that in detail, there is another significant difference, we have to account for - in fact, a much more obvious one, which the amount of amino acids the rats were given in the Wilson and the Pimentel study, respectively (cf. figure 2).

Figure 2: Dosage, not type of supplement would be the most probable explanation for the different effects of leucine and HMB supplementation on AMPK phosphorylation in the Wilson vs. the Pimentol study.

I hope you did not already forget that, the main function of AMPK is to prevent that your cells run out of fuel or, to be precise, to avoid the ratio of "used" energy ADP and AMP (adenosine di- and monophospate) to ATP (adenosine triphospate) to continue to rise beyond a tolerable level. I further assume that you will be familiar with the fact that branched-chain amino acids bypass oxidation in the liver and thus become readily available energy sources for skeletal muscle (Renny. 2011). Now, if you put one and one together the answer seems pretty obvious: If the dosage of amino acids is sufficient (remember that those 270mg leucine are 4x more leucine than the the rats in the Wilson study got for "breakfast") to restore ATP levels to "appropriate" levels, the decrease in the ADP/ATP ratio will allow part of the AMP-activated protein kinase to be dephosphorylated.

According to our current understanding, BCAAs in general and leucine in particular trigger the ATP related decrease in AMPK and the complementary increase in mTOR by two distinct pathways, of which Tokunaga et al. write (Tokunaga. 2004)

[...]leucine stimulates p70α phosphorylation via mTOR pathway, in part, by serving both as a mitochondrial fuel through oxidative carboxylation and an allosteric activation of glutamate dehydrogenase. This hypothesis may support an idea in which leucine modulates mTOR function, in part by regulating mitochondrial function and AMPK.

In plain English: Leucine increases ATP when it is "burned" as fuel and it docks directly to the the non-active site of glutamate dihydrogenase enzyme and thusly increases the conversion of glutamate to alpha-ketoglutarate which in turn can be fed into the citric cycle to ultimately produce ATP.

Is it all about (cellular) energy ...

Figure 3: AMPK phosphorylation in Escherichia coli at different ADP/ATP ratios (data adapted from Xiao. 2011)

In April 2011 Xiao et al. published a study in Nature with some interesting quantitative data on the ADP/ATP ratio, on the one hand, the phosphorylation status of AMPK, on the other (Xiao. 2011). As my plot of the data in figure 3 shows, with increasing ATP levels (at constant ADP levels of 30µM) the phosphorylation of AMP-activated protein kinase in Escherichia coli BL21 cells declines by roughly -20% from 44% at a 30/0 ADP/ATP ratio to 22% at a 30/800 ADP/ATP ratio.

Yet, although these results would confirm the hypothesis that the main reason for the discrepancy is dose, or rather, energy related, and each and every nutrient that could potentially raise ATP levels, would eventually decrease AMPK, this still does not explain the increase in mTOR Pimentel et al. observed, despite (statistically non-significant) increases in AMPK.

... or is there a place for the "magic" of HMB?

As you probably know, beta-hydroxy-beta-methylbutyrat (HMB) is an oxidation product of leucine and / or its keto-acid alpha-ketoisocaproate (KIC) (Koevering. 1992). In 1998 Lembert et al. found that even KIC is not a direct substrate for ATP production, instead "KIC must transaminate with glutamate or glutamine to yield alpha-ketoglutarate and leucine" (Lembert. 1998). We may thus assume that similarly HMB cannot be used (directly) to restore cellular ATP pools. Moreover, HMB is thought to be the second (non-energetic) pathway by which leucine acts on protein synthesis / breakdown. According to a 2011 review of the literature by Zanchi et al. (Zanchi. 2011)

Nissen et al. (1996) suggested that HMB or some other metabolite (since there is no specific inhibitor to BCAT) is the main component responsible for the anti-catabolic effects of HMB because when adopting inhibitors of BCAA transamination, the only BCAA capable of anti-proteolytic effects is leucine, which undergoes a process capable of generating HMB (Slater and Jenkins 2000). Such effects were not observed when other BCAAs were tested (isoleucine and valine), suggesting that HMB or some metabolite may be the key element in promoting the [anticatabolic] effects.

When usually 5% of the dietary leucine is metabolized into HMB (Wilson. 2008), and these 5% are responsible for the non-ATP dependent effects on phosphorylation of mTOR, p70^S6k, and 4E-BP1 of leucine (Eley. 2007), it is no wonder that chronic intake of 80mg of HMB did stimulate mTOR in the absence of increased ATP levels (which would obviously have led to a decrease in AMPK expression that was not present in the Pimentol study), while 270mg leucine, yielding only 13.5mg HMB, did not stimulate mTOR, but was (ab-)used as a substrate to increase cellular ATP levels, thusly reduced AMPK levels and increased protein anabolism - different pathways, similar results: an increase in net protein synthesis.

Figure 4: Simplified illustration of the two distinct pathways by which leucine can work its muscle protein synthetic (MPS) magic and a hint on the compensatory (/) / amplifying (+) effects of exercise.

There is however, a third major pathway to the metabolic effects that are brought about by common intermittent fasting programs and this third player makes things even more complicated (cf. figure 4) - it's exercise! You probably remember from yesterday's installment that

1. during exercise in the fasted state temporarily AMPK increases and the energetically costly muscle protein synthesis (MPS) is reduced, while
2. after exercise (regardless of whether it was performed fasted or not, cf. "Glycogen-Free Growth") muscle protein synthesis increases due to an exercise-induced stimulation of the mTOR protein synthetic cascade

Before we dig deeper into this modulatory effects of different modes of exercise in the next installment of the Intermittent Thoughts on Intermittent Fasting, however, I want to conclude today's thoughts with a preliminary list of supplements / medications that have been shown to modulate the phosphorylation state of 5' AMP-activated protein kinase.

Image 2: If you insist on trying HMB, don't be stupid and buy a capped products, the prices for bulk HMB powder have lately been crushed - a major European carrier, for example, sells 250g at <13€ atm; HMB is thus cheaper than BCAAs, which cost 16Euros in the small 250g pack - did you hear me say that even 13€ is too much, no - you must be mistaken ;-)

"Should you prefer HMB over leucine as a dietary supplement to promote lean mass gains and prevent muscle loss during the fast?" I assume this is a question many of you will now be pondering about. My answer to this question would be "NO!" Firstly, if you are no construction worker or pursue a similar physically demanding profession, the fear of losing muscle (which is different from "feeling flat", my bodybuilding friends ;-) during a ~16h fast is hilarious, which means that BCAA, Leucine or HMB supplementation, while you sitting fasted at your desk in the office is simply unwarranted. Secondly, when you are exercising the increased energy demand will negate / compensate the negative effect the increase in ATP has on AMPK activity. And thus, thirdly, a large bolus of leucine (or a complete BCAA or EAA product) taken pre-workout will not only ward off proteolysis (as HMB would do) it will also provide the necessary energy to train harder and thus help to increase the exercise induced stimulus on protein synthesis.

All that and the absence of conclusive scientific evidence that would demonstrate the superiority of HMB supplementation over the provision of adequately dosed BCAA or EAA mixtures (it stands to reason that you cannot compare 3g of HMB to 3g of BCAA) are arguments against the use of β-Hydroxy β-methylbutyric acid. If you wanted to try it, anyway (and have no problem swallowing a powder that tastes like poison), the prices for bulk-powders have gone through the floor, lately ;-)

How to modulate AMPK "artificially" -  supplements, medications, hormones and more

In view of the fact, that the discussion of the effects of leucine (BCAAs and HMB) alone took much longer than I had expected and this whole episode took a different turn than I would have expected, the following list is more a preliminary overview than a comprehensive explanation of the effects of various supplements, medications, hormones and hormone-like substances on the AMPK. The latter will follow, as promised, but for today, you will have to content yourselves with what I would like to call a sneak peak on the AMPK-mTOR modulation handbook of which I hope that it will be one of the outcomes of all the past and future work that is going into this series ;-)

AMPK promoters:

• Adiponectin (Yamauchi. 2002; Handy. 2010; Iwabu. 2010)
• Alpha lipoic acid (Lee. 2005; Wang. 2010; Packer. 2011)
• Artemisia sacrorum Ledeb (Yuan. 2010)
• Aspirin (Sung. 2011)
• Astragalus (Zhou. 2009)
• Berberine (Ma. 2011)
• Brain-derived neurotrophic factor (Matthews. 2009)
• Caffeine (Jensen. 2007; Spirydopulos. 2008; Egava. 2009) 
• Capsaicin (Kim. 2010)
• Catechins / green tea (Huang. 2009; Murase. 2009; Reiter. 2010)
• Chromium (Wang. 2009)
• Ciliary neurotrophic factor (CNTF) (Watt. 2006)
• Cinnamon (Huang. 2011)
• Conjugated Linolic Acid (CLA) (HSU. 2011)
• Coptidis Rhizoma (CR), the dried rhizomes of Asian herbs (Egawa. 2011) 
• Cordiceps sinensis (Wong. 2010)
• Creatine (Ceddia. 2004)
• Curcumin (Kim. 2010)
• DHA (Jing. 2011)
• Ecklonia Cava (Kim. 2010)
• Garlic (Lee. 2011)
• Ginseng (Juang. 2010)
• Glucosamine (Kong. 2009)
• Grape seed extract (Meeprom. 2011)
• Hibiscus sabdariffa extract (Yang. 2010)
• Interleukin-6 (Kelly. 2009)
• Kidney bean extract (Lee. 2009)
• Malva verticallata seeds (Jeong. 2011)
• Metformin (Zhou. 2001; Freyer. 2002; Ohira. 2009)
• Mulberry extract (Tsuduki. 2009)
• Nicotine (Cheng. 2007)
• Prostaglandin E2 (Zhu. 2011) 
• Pu-erh tea (Way. 2009) 
• Quercitin (Jung. 2010)
• Raishi Mushroom (Lane. 2011)
• Resveratrol (Fullerton. 2010; Lin. 2010; Breen. 2008)
• Tiozolidinediones (diabetes meds. like rosiglitazone) (Freyer. 2002; LeBrasseur. 2005; Yang. 2011) 
• Vanadium (Hwang. 2011)

I still have two things to add to this list, firstly, this list is the result of a VERY cursory and 100% random search and is not even intended to be complete (at this time ;-). The intention (at least for in this installment) is to show you that an overwhelmingly large percentage of purported health supplements, diabetes and obesity treatments work via the AMPK pathway. And, secondly, I decided to limit the references to 1-3 per compound, even if in cases such as Metformin, ALA & Co the number of relevant studies is probably >500. Therefore you better consider the given references as evidence that I did not make up any associations between compound X and AMPK phosphorylation - and, if you want to know more before the release of the next installment, I suggest you go to PubMed and enter the respective keywords and do some digging on your own (your SuppVersity homework of the day - so to say ;-)

I hope you do not mind that I did not manage to tackle the effects of sleep and exercise in this installment, as I had originally intended. It is, after all, the central characteristic of this series that I sit down in front of the computer and start thinking at point "A", then I dig, here, get distracted there and follow up on "A1" to "A743", so that the output is by no means as structured and straight forward as my lectures and seminars or my SuppVersity blogposts on isolated topics... so, I can only hope that you enjoyed the turn this installment took (at best, because you learned something new) and in the unfortunate case that you did not enjoy what you have just read, you can at least look forward to the next episode of the Intermittent Thoughts on Intermittent Fasting Series ;-)