Can your food make you fit?
A paper recently appeared in the scientific literature demonstrating an enormous and previously unknown benefit of low-carbohydrate dieting. (1) I haven’t seen this study mentioned anywhere, so, important though it is, it seems to have flown under the radar of all the low-carb experts and the press. It is a study about how restricting carbohydrates – even in the face of overfeeding – drives profound signaling protein changes. This is an exciting study, but before we get to the specifics, I want to digress a bit.
First, a story about my own peripheral involvement in a similar study many years ago.
At a scientific meeting in the late 1980s I found myself in a hotel bar with a couple of young anti-aging researchers, who were affiliated with a major government nutritional research facility. The brunt of their research focused on the enzymatic changes in lab animals undergoing caloric restriction. In the course of their work, they found that calorically-restricted rodents exhibited the same changes as did primates and, they assumed, humans. But they had no human data to confirm. They, along with everyone else in caloric-restriction research, knew that reducing the calories fed to rodents improved their health and extended their lives. And at that time, it was looking as if the primate research would show the same thing. (It ultimately did.) What these guys wondered about, though, were humans. Did humans undergoing caloric restriction display the same enzymatic changes?
No one knew because at that time there were no human caloric-restriction studies.
During the course of our conversation, I told these researchers about my practice and about the success I was having with patients on low-carb diets. I explained how my patients lost weight fairly easily and experienced significant and rapid changes in blood pressure, lipids, fasting insulin and blood sugar levels. They became intrigued since these changes pretty much mirrored those seen over time in caloric-restriction studies on lab animals. It set them to wondering whether humans following low-carb diets would manifest the same enzymatic changes as calorically-restricted animals. They proposed an experiment.
Before we plunge ahead, though, let’s take a minute to review enzymes. Enzymes are such an integral part of everything that makes life possible that I often forget that they are not really common knowledge. This was brought home to me with great clarity when I was having a conversation a couple of weeks ago with someone who, though not a doctor or a scientists, has a pretty firm grasp of the overall workings of the metabolic system. I was rambling on about the phenomenal results of the study we’ll soon be delving into and mentioned enzymes. My friend fixed me with a quizzical look and asked, “What is an enzyme?” I figure if this guy didn’t know, there are many people out there in the same boat. If you know all about enzymes, feel free to skip on down to where I pick up on happy hour.
What are enzymes? Where do enzymes come from and what do they do?
Enzymes give us life. If it were not for enzymes catalyzing the unimaginable number of reactions required for us to live, we would all be giant lumps of fairly inert chemicals instead of the moving, breathing creatures we are. Every impulse we have, every beat of the heart and blink of the eye, every breath we take is the end result of a series of chemical reactions, none of which would take place without enzymes driving them. Every time you see a chemical reaction that takes place in the body, each step is catalyzed by a specific enzyme. Take a look at the partial pathway of glycolysis from Wikibooks. Each of the steps shown is catalyzed by a specific enzyme.
Below right is another well known biochemical pathway, the synthesis pathway for cholesterol. The most important enzyme in this pathway, hydroxyl-methylglutaryl coenzyme A reductase (HMG CoA reductase), is the rate-limiting enzyme in the cholesterol synthesis pathway. And it is the enzyme that statin drugs inhibit. Inhibiting HMG CoA reductase decreases the body’s production of cholesterol. Unfortunately, there are some downstream effects of this inhibition that the statin folks don’t like to talk about. The next molecule in the pathway, mevalonic acid, is catalyzed by yet another enzyme into farnesyl phosphate, which is the precursor to Coenzyme Q10, an important substance found throughout the body. So inhibiting HMG CoA reductase in an effort to reduce cholesterol often ends up also depleting the body of Coenzyme Q10. Graphic came from this paper on statins and Coenzyme Q10.
Enzymes are large proteins coded for in the DNA and transcribed from an RNA template. Enzymes have varying degrees of activity, typically controlled by other enzymes upstream in the reaction cascade. The synthesis of new enzymes takes a while (which is why there is an adaption period when switching to low-carb dieting) but the activation of existing enzymes can take place almost instantly. Activation typically involves adding or removing a chemical group from one of the amino acid chains of the inactive enzyme.
Some enzymes catalyze only one reaction, while others catalyze many. A few are master regulator enzymes in that they catalyze many reactions and activate or deactivate multiple other enzymes and enzyme pathways. These enzymes, often called signaling proteins, drive and/or inhibit multiple systems. It is these enzymes that scientist measure the activity levels of to determine what effect various drug treatments or nutritional therapies bring about.
So, back to my happy hour conversation.
The two researchers proposed an experiment in which I saved a little blood from the regular labwork that I did on all my new patients. They wanted me to then send them a bit of blood from these same patients when they came in for their regular six-week blood tests. (I drew blood on all patients during their initial visit and again at six week intervals to monitor progress.) They sent me special collection vials and a coding system so that patients would remain anonymous. I asked the patients if they minded, and no one did, so the experiment started.
I dutifully collected the blood and sent if off. After enough patients had cycled through their first six weeks, I got an enthusiastic call from the researchers. And enthusiastic is a mild word for their state of mind. They were practically gibbering with excitement, because they had found the same enzymatic changes in the blood from the patients who weren’t trying to restrict calories but were simply following a low-carb diet. They told me they were going to report their preliminary data to their boss to see if they could pursue funding to continue research with my patients.
When they called me back their enthusiasm was gone. In its place was an overwhelming glumness. They presented the data to their boss, who was the head of the research institute at which they worked, and the response was not what they expected. He told them that they had not followed academic protocol, they hadn’t gone through an institutional review board, and had no business doing an off the books experiment using the institute’s equipment. They got the lecture about how they put the institution at risk and how they could be fired if and on and on. He told them to wash their hands of the whole thing. Which they did.
They didn’t share their data with me, so I never learned what changes had taken place. I did figure, however, that the changes must have been pretty good given the degree of excitement they generated. At that time, I was simply a clinician taking care of patients and hadn’t started my deep dive into the scientific literature, so I really had no idea what the enzymes they looked at were.
Which brings us 25 years later back to the paper just published online ahead of print in Hormone and Metabolic Research. This paper discusses the activity level of adenosine monophosphate-activated protein kinase (AMPK) as a function of carbohydrates in the diet. AMPK is an enzyme with all manner of downstream effects and can be considered as the Queen Mother of all enzymes, a powerful signaling protein that drives multiple metabolic pathways. You can see a photo of a 3D model of AMPK at the top right of this post from Wikipedia.
The ‘K’ in AMPK stands for kinase, which means that it adds a high-energy phosphate group to other downstream compounds. AMPK is itself activated by an upstream kinase called, appropriately enough, an AMP kinase kinase.
AMPK: what does it do?
AMPK basically monitors the energy levels inside the cells, and when it finds them low, it kicks off several chains of reactions directed toward energy repletion.(2)
As most of you know, ATP is the body’s energy currency. It is made from its precursors ADP and AMP. The food we eat ends up as high-energy electrons that drive the process designed to keep the cells filled with their high-octane energy molecules ATP. You can think of how a battery works. When it is charged, it is at the ready to discharge current to run an iPad, flashlight, cellular phone, whatever. After a time the battery needs recharging or it ceases to provide current. Human cells operate much the same way, but they have an advantage over an iPhone. Human cells have the ability to constantly recharge as their supplies of ATP are consumed. The ratio of the ATP precursors ADP and AMP to ATP signals whether the body should be in the energy discharging or energy storing state.
The signaling protein AMPK monitors this cellular ‘battery’ and sends the appropriate signals to ramp up the forces required to restore the ATP balance to the fully charged state.(3)
If we eat less or don’t eat, we discharge our cellular batteries, because we continue to use ATP but aren’t providing the energy via food to make more. If we exercise, we discharge our batteries, because we are consuming large amounts of ATP quickly, and unless we’re eating on the run – literally – we are not getting the food energy needed to replace our depleted ATP. So when we eat less and/or exercise, we put ourselves in a cellular battery discharge state.
Numerous studies have shown health and longevity benefits from eating less and exercising, though these prescriptions are tough to stick with for the long haul. If this is true, and I believe it is, then the body is better off health-wise to be in a battery-draining state more of the time than in a battery-charging state.
Just for clarity’s sake, a distinction should be made so that confusion doesn’t set in. By cellular battery, I’m referring to the ratio of ATP to ADP and AMP in the cells. I’m not talking about stored fat and sugar. In this model, stored fat and stored sugar would be considered the wall charger or power company where the power comes from to charge the batteries in our devices.
Whether the body is exercising or resting, eating or fasting, the cells need to have their ATP levels pretty much constant. But caloric restriction and/or exercise deplete ATP quickly, so this ATP needs to be restored just about as quickly as it is being depleted.(4) The new ATP needed to top off the tank comes from high energy electrons thrown off from burning fat and/or glucose. If there is no fat or glucose coming in via the mouth, it must come from stores socked away in the body. From glycogen (stored glucose) and body fat.(5)
When ATP levels fall as the batteries discharge, signals go out to the parts of the metabolic system that are responsible for harvesting the energy from stored sugar and fat to create the high energy electrons required to make more ATP.
AMPK is one of the primary signaling proteins that monitors the ATP levels in the cells and signals for more energy when levels drop. When AMPK is activated indicating our cellular energy tanks are depleted, all kinds of good things happen. Here is a short list of metabolic efforts all kicked into action by activated AMPK and why they’re important. (adapted from ref #6)
- Increases glucose uptake: We want to get glucose out of the blood and into the cells to burn.
- Increases glycolysis: We need to break down glycogen (stored sugar) to get the glucose to burn.
- Increases fatty acid oxidation: An obvious one. We want to start burning fat to replenish the depleted energy stores.
- Increases mitochondrial biogenesis: we want to make more mitochondria to burn fat and generate as much ATP as possible.
- Inhibits gluconeogenesis: We don’t want to spend energy making more sugar – we want to burn it.
- Inhibits glycogen synthesis: Same thing – we don’t want to store sugar, we want to burn it.
- Inhibits fatty acid and cholesterol synthesis: We don’t want to spend energy making fat and cholesterol.
- Inhibits insulin secretion: We want insulin to be low, so that we can move stored fat and sugar to where it needs to be burned.
When our ATP tanks are filled to the bursting, as when we eat and are stuffed with food (especially carbohydrates) and/or we don’t exercise, all the above pathways go in the opposite direction. If we chronically overeat the wrong foods, our metabolic systems end up sending all the above pathways in the opposite direction most of the time.
When viewed from this perspective, it’s pretty easy to see why AMPK activated by a calorically-restricted diet and/or exercise brings about many healthful changes. It also might make one wonder why drugs haven’t been developed to increase the activity of AMPK to provide these same benefits to people who suffer from obesity, high blood sugar, diabetes and all the other disorders caused by overnutrition. A drug designed to activate AMPK would be diet and exercise in a pill. Who wouldn’t want that?
Well, there are several such drugs. Most have probably heard of one of them: metformin (trade name Glucophage.) Metformin, derived from an ancient herbal remedy, is used by doctors to treat diabetes and insulin resistance and works by activating AMPK. In 2010, physicians wrote some 100 million prescriptions for metformin to treat type 2 diabetes.(5) Some use it to treat obesity, and many folks who can get access to it, take metformin in hopes of increasing longevity.
Drugs that increase the activity of AMPK, when used over time, along with all the effects mentioned above tend also to increase the number of mitochondria, which increases the capacity to burn fat and turn it into ATP. More mitochondria leads to improved endurance, and, consequently, many of these drugs have been placed on the banned list of the World Anti-Doping Agency, the regulating body that deals with drug abuse in sports.(7)
Up to this point in this post, your take away message should be that activated AMPK is a very good thing. If your AMPK is activated much of the time, it would indicate you are eating less, exercising more and making mitochondria. All to be desired. Plus, though it has no bearing on the study we’re about to discuss, it looks like AMPK activation modulates the immune response in a positive way (8) and may even prevent some kinds of cancer.(9) More good things.
At last, to the paper at hand.
Now that I’ve set the stage, lets get to the exciting study that kicked off this blog post.
The researchers knew that AMPK was activated with calorically-restricted diets. They wanted to test whether or not macronutrient composition had an effect on AMPK activity.
These folks from the medical school at the University of Colorado presented two different studies in this paper. I’m going to discuss them in reverse order.
In the second study, eighteen obese subjects (8 men – 10 women; avg age 32.4 years and avg wt 227.3 lbs) went on a eucaloric (enough calories to match energy output) diet of 30% fat, 50% carbohydrate and 20% protein for five days to establish a baseline. After the five days, the subjects were randomized to receive either five days of a low-fat, high-carb diet (20% fat, 60% carb, 20% protein) or five days of a high-fat, low-carb diet (50% fat, 30% carb, 20% protein). Both of these diets were restricted to 30% of the calories in the baseline diet. On the night of the fifth day of the study, patients were hospitalized and after an overnight fast, the researchers performed insulin clamp studies and muscle biopsies.
What did study #2 show?
After five days on the calorically-restricted diet, obese subjects in both groups experienced modest weight loss. There were no significant changes in the subjects in either group in any parameters of insulin sensitivity. In both groups, as would be expected on a calorically-restricted diet, fasting insulin levels fell.
Since both diets were calorically restricted, it was expected that the activity levels of AMPK would be increased. And in the low-carb/high-fat diet, AMPK levels were significantly increased. The big surprise, however, was that the activation of AMPK in those subjects on the low-fat/high-carb diet was basically unchanged.
As the study authors put it, this change as a function of carb restriction
suggest[ed] that high carbohydrate intake prevents activation of AMPK… in skeletal muscle that otherwise would have been induced by caloric deprivation.
In other words, these subjects were rowing one way by reducing calories while rowing in the other direction by increasing carbs. Maybe this is the explanation of why it’s so difficult to lose weight and improve health parameters on a low-calorie, high-carbohydrate diet.
In the first study – which I think is the much more interesting – the researchers recruited 21 lean (11 men, 10 women; avg age 27.8 yrs; avg wt 147.4 lbs), healthy, non-diabetic subjects, all of whom were started on the same five-day eucaloric baseline diet as the subjects in the other study (30% fat, 50% carb, 20% prot). The subjects were then randomized into two different groups, both of which consumed 40% more calories as compared to baseline. These five-day overfeeding diets were either low-carb (50% fat, 30% carb, 20% prot) or low-fat (20% fat, 60% carb, 20% prot). As before, on the fifth day of the overfeeding study, subjects were hospitalized so that insulin clamp studies and muscle biopsies could be done the next morning. Then, unlike with the other study, this same group of subjects came back a month later and went through the process again except in a cross-over fashion so that each subject could act as his/her own control.
What did study #1 show?
Interestingly, despite the 40% caloric overfeeding, no significant changes in body weight occurred in either diet. And there were no changes in insulin sensitivity, glucose, lipids or other parameters measured. What was truly amazing, however, was what happened to AMPK activation. Low-fat/high-carb overfeeding did not produce any effect of AMPK activity as compared to baseline. But low-carb/high-fat overfeeding produced a significantly increased activation of AMPK.
In referring to these two studies, the researchers noted:
We observed that caloric restriction with [a low-fat/high-carb] diet did not alter the AMPK [activation], suggesting that increased dietary carbohydrate content even in the face of caloric restriction prevented activation of AMPK… in skeletal muscle of obese individuals. In contrast, overfeeding with [a low-carb/high-fat] diet increased the activity of this pathway [AMPK] indicating that low carbohydrate content may be sufficient for its activation.
And in summary, they commented:
Our data indicate that a relative deficiency in carbohydrate intake or, albeit less likely, a relative excess of fat intake even in the absence of caloric deprivation is sufficient to activate this network and increase fat oxidation.
These studies may provide an answer as to why most weight-loss studies comparing low-fat/high-carb diets to low-carb/high-fat diets almost always find the low-carb diet to bring about the greatest loss.
And this activation of AMPK even in the face of overfeeding may explain why it is difficult for most people to gain weight on a true low-carbohydrate diet. Even one with a large dollop of extra calories.
This paper is exciting because, at least in this case, researchers are looking at what macronutrient differences do to signaling proteins, even in the face of overfeeding. Up until now, most researchers have written if off to a caloric-restriction phenomenon. It’s nice to see that this group has tried to tease out what is really doing the heavy lifting in terms of AMPK activation. It appears to be the carb restriction.
If these findings hold up in other studies, then it would seem that, in a manner of speaking, you could have your cake and eat it, too. Wouldn’t it be nice to be able to go on a diet that truly allowed you to eat all you wanted and have that diet not only not put excess avoirdupois on you but even make your body think it had been exercising?
Before I get too carried away here, there are a few questions this study raises.
First, as anyone who looked at the different diets critically would notice, the low-carb arm of the diet wasn’t really all that low-carb. Reducing carbs to 30 percent of calories is a semi-sort-of low-carb diet. Most people following true low-carb diets get their carbs down to anywhere from 5-15% of total calories. If would be nice to know if these findings held with subjects following a really low-carb diet and not the minimal restriction studied. I would bet the findings would be even better. I base this on the enthusiasm of the researchers years ago who studied the blood of my own patients on very-low-carb diets. But that kind of data won’t feed the whippet, so we’ll have to wait until further studies are done with lower carb restriction to know for sure.
Second, these studies were of only five days duration. We don’t know if AMPK activity runs up and hits a max at five days only to begin to decline thereafter and end up lower on a low-carb diet after two weeks or a month or six months. We simply don’t know based on the data from this study.
Third, there weren’t a huge number of subjects in these studies, so, once again, though the data looks tantalizing, it might not hold if several hundred subjects were evaluated.
Given my bias, installed by several decades of using low-carb diets to treat all kinds of problems, I suspect the data will hold up and get even better with more carb restriction and longer study periods. But we’ll have to wait and see until we know for sure.
Such an exciting study as this one should drive a fair amount of research in this direction quickly. If you do a PubMed search for AMPK, you will find there is plenty of interest. I hope we don’t have to wait long. But until new research comes along definitively overriding this paper, I’m going to continue my own regimen of restricted carb dieting and recommend you to do the same.
1. Draznin B, et al. Effect of Dietary Macronutrient Composition on AMPK and SIRT1 Expression and Activity in Human Skeletal Muscle. Horm Metab Res. 2012 Aug;44(9):650-5.
2. Hardie DG, et al. Management of cellular energy by the AMP-activated protein kinase system. FEBS Lett. 2003 Jul 3;546(1):113-20.
3. Kahn BB, et al. AMP-activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab. 2005 Jan;1(1):15-25.
4. Hardie DG, et al. AMPK: a nutrient and energy sensor that maintains energy homeostasis.
Nat Rev Mol Cell Biol. 2012 Mar 22;13(4):251-62.
5. Hardie DG. Organismal carbohydrate and lipid homeostasis. Cold Spring Harb Perspect Biol.
2012 May 1;4(5).
6. Hasenour, C.M., et al. Emerging role of AMP-activated protein kinase in endocrine control of metabolism in the liver. Molecular and Cellular Endocrinology 2012, Epub ahead of print. http://dx.doi.org/10.1016/j.mce.2012.06.0185.
7. Hardie DG. AMP-activated protein kinase: an energy sensor that regulates all aspects of cell function. Genes Dev. 2011 Sep 15;25(18):1895-908.
8. Lui TF, et al. Fueling the flame: bioenergy couples metabolism and inflammation. J Leukoc Biol. 2012 May 9. [Epub ahead of print]
9. Fogarty S & Hardie DG. Development of protein kinase activators: AMPK as a target in metabolic disorders and cancer. Biochim Biophys Acta. 2010 Mar;1804(3):581-91.