161: Why Your Body Chooses Sugar Over Fat

What determines whether your body burns sugar or fat? It's not just about which fuel is available. One of the biggest signals is insulin. When you eat carbohydrates, blood glucose rises and insulin rises with it. That insulin does 2 important things. It tells your fat cells to stop releasing stored fat, and it makes it harder for fatty acids to enter the mitochondria where they would normally be burned for energy. So as insulin rises, your body shifts toward burning more glucose. And less fat. When insulin falls, that changes. Stored fat can be released again, enter the mitochondria, and be used for fuel. That's one reason lowering carbohydrate intake can dramatically increase fat oxidation. Your body already knows how to burn fat. Sometimes it just needs the hormonal signal to let it. This is Lecture 161 of The Metabolic Classroom. Welcome to the Metabolic Classroom. I'm Ben Bikman, metabolic scientist and professor of cell biology. The idea that glucose and fats compete for the same oxidative or metabolic pathways is a long one. This idea that there's this competition between the 2 fuels in the body, whether the body's burning sugar or burning fats, is one that's become in fact far more popular and prominent than I ever imagined when I first learned about it in graduate school 20 years ago. But the idea of this competition between burning substrates or burning fuels originates from a paper that was published in 1963. At the time, the investigators described what they called the glucose fatty acid cycle. It was later then referred to as a Rand— as the Randle cycle, but I learned about it as the glucose fatty acid cycle. Their claim was that the two fuels interfere with one another. A cell oxidizing a large amount of fat will find it harder to oxidize glucose or blood sugar. Glucose breakdown happens in a sequence of steps, so there's a series of biochemical events, and the mechanism they proposed inhibits that sequence starting from the end. And so if the burning of glucose is getting disrupted, it's starting at the back end of that fat oxidation or fat burning actually inhibits the final step of glucose burning. That last step is controlled by an enzyme called pyruvate dehydrogenase. When that step slows, intermediates will build up and slow the steps that precede it until ultimately glucose uptake into the cell itself is reduced. So fat oxidation reduces both glucose uptake and glucose oxidation. In fact, I should say that the other way around, because that's the more sequential, properly sequential series of events. Fat oxidation inhibits glucose oxidation, which starts to lead to an inhibition or reduction in glucose uptake into the cell. The original observations that led to this cycle were made in isolated rat muscle and as well as rat heart. Of course, those are not the same thing as a whole human. In humans, much of what we know about this competition comes from clamp studies where insulin and glucose are infused and held at chosen or clamped at chosen levels while they're doing all kinds of measurements in the people. This gives the researchers precise control. Instead of feeding someone a meal and watching several things change at once, they can hold very specific variables steady and then see what the body does. In several of the studies that I'm going to review with you over the course of the mini lecture, you're going to— I'm going to also introduce other studies where they also held fats or fatty acid levels constant by infusing directly into the body a particular level of fats. What, what's called a lipid emulsion. So the human version of the experiment that was done in rat muscle, it came decades after the first observations. One of the seminal studies in this took healthy volunteers and they went through 2 insulin clamps. On one occasion, fatty acids were kept elevated with a lipid infusion, so an infusion of a lipid emulsion. So there's just this liquid slurry of fats. On the other, they were allowed to fall the way insulin normally makes them fall. Now, what do I mean by that? Normally, if insulin is high, then fatty acids are going to be down. And this is a point we're going to come back to. So what they were able to do was keep fatty acids available even when insulin was high, and putting both of those up— so keeping fatty acids up and insulin up. It cut glucose uptake into the leg muscle by about a third, and it also changed what the leg was using for fuel. With fatty acids present, glucose made up a little over half of the energy use. Without them, then the use of glucose went up over to about 3 quarters of total fuel use, which is no surprise. If there was more fat, the, the muscle was burning less glucose. Now let's go back to that enzyme. That I mentioned at the last step of glucose burning, pyruvate dehydrogenase, they found that pyruvate dehydrogenase activity was significantly down. Now remember, this is in human muscle, in whole living humans, but it complemented the work from that early initial 1963 paper. So they found the same. It supported what that 1963 paper would've predicted to happen. So the competition holds true in human muscle. And it works through the same step that the original work identified. Now, one measurement in these human studies shows up again and again, and it's called the respiratory quotient, the RQ, the respiratory quotient. That's simply the ratio of carbon dioxide that is produced off compared to the oxygen that is consumed or used. So think about a working cell. The cell is pulling in oxygen and then it's breathing out, or, or I shouldn't say breathing out, but then it's, it's giving out carbon dioxide. When you burn carbohydrate, that ratio is about 1, and when you burn fat, it drops to about 0.7 because fat has, uh, more hydrogen in it relative to carbon. It actually needs a little more oxygen. So if someone If, if a muscle tissue has a respiratory quotient above, say, 0.9, it means that the tissue is running almost entirely on carbohydrate. But if the respiratory quotient is about 0.7, it means that it's running mostly on fat. Now, whether it's measured across a, a limb or measured at the level of the whole body breathing, which is commonly called the respiratory exchange ratio, RER, but that's just sort of the whole body version of RQ, regardless of where it's measured. That number is how these studies can tell us which fuel is being used. But let me just reiterate a point, and it is one that we're going to revisit a bit. We need to pay attention to what the experiment I mentioned earlier actually created— the, the metabolic milieu. It had high insulin and high fatty acids at the same time. In a healthy person, those 2 things don't usually happen together. Because insulin at high levels, or even modestly elevated levels, will shut down the release of fatty acids from fat tissue. So the researchers had to infuse fats in order to override what the hormone insulin was trying to do on its own. So what they created is a situation where the fat cell's no longer responding to insulin. Now, we're going to come back to that. As I said, when fat cells still respond normally to insulin, We, we would see the opposite. That, as I said earlier, if insulin's high, fatty acids will be down. If insulin's down, fatty acids will be high. That becomes important as we seek to understand this glucose fatty acid cycle. Okay, so the work that I just mentioned now, the initial work found that fat oxidation can suppress glucose oxidation. The more relevant question, I think, for how most of us eat is whether it works the other way, whether glucose oxidation suppresses fat oxidation. Now, when I say how most of us eat, that's just simply built on the fact that statistically speaking, most calories that most people consume worldwide are coming from carbohydrates. In fact, based on a nutrient database, about 70% of all calories globally come from carbohydrates. All right. So this view was tested in healthy volunteers under a clamp that raised both insulin and blood glucose well above normal levels. Fatty acid levels in the blood were held constant the whole time by continuous infusion. So if fat oxidation fell, it couldn't be because there was less fat available. So they were putting all of these up. Glucose oxidation, or glucose burning, rose more than threefold And fat oxidation fell. The investigators concluded that what mainly determines which fuel is oxidized is how much glucose is available inside the cell, not how much fat is available. Now there was a companion or a follow-up experiment that showed us more precisely where all of this biochemistry is changing. They actually infused two labeled or tagged. Fatty acids. So these are fatty acids that they could measure as they were getting burned. So two fatty acids that were labeled were infused at the same time. One of them was oleate, which is a long-chain fatty acid that has to use normal entrance into the mitochondria through through this carnitine shuttle, which I'll mention again in a moment. And so just think of it as a long-chain fat that requires normal steps to enter the mitochondria. But they also infused octanoate. Which is an MCT. It's a medium-chain fat. And you may recall from previous mini lectures that as the fats get shorter, they get— there's less regulation as to how readily they can enter the mitochondria. Medium-chain fats are short enough, and octanoate is one of those, so it can enter the mitochondria on its own. It does not go through that carnitine shuttle system that I just mentioned that the longer-chain fats have to use. All right, so All this to say, they also infused these 2 specific fats, a long-chain fat and a medium-chain fat, and they held the fats constant again. So they were constantly infusing them in under the lipid, uh, under the insulin and glucose clamp. So as they were also pushing up insulin and glucose, they found that oleate oxidation, the long-chain fat burning, it dropped by almost half, but the burning of the medium-chain fat didn't change at all. This comparison is very helpful because it lets us understand where the regulation of the fat burning is occurring. Medium-chain fats don't need carnitine as this sort of transporter to get into the mitochondria. They were unaffected. Long-chain fats— and these are the fats that of course get released from fat to burn and the fats that we eat for the most part— they need this specific shuttle called the carnitine palmitoyltransferase 1, or CPT1. And, and they, we, they found that the regulation, uh, CPT1 was more blocked during this by elevating glucose and insulin, and that prevented the long-chain fats from burning as well. So the fat oxidation of the long-chain fats fell because transport into the mitochondria was inhibited by the glucose and the insulin signal, not because the fatty acids weren't there. Now, after a normal carbohydrate-containing meal, things are pretty straightforward. Glucose will enter the blood, insulin rises in, in response, glucose uptake into muscle and fat tissue increases, and the release of fatty acids from the fat tissue is shut down. At the same time, Rising glucose inside the muscle cell inhibits fatty acid transport into the mitochondria, like I just mentioned a moment ago with the carnitine shuttle system being turned down. More glucose is being delivered while both the supply of fatty acids and their entry into the mitochondrion are reduced, so the cell will burn glucose. Now, none of this depends on whether total calories are in deficit or surplus. It's just following a hormonal signal, and that signal is insulin. Insulin matters when you want to understand the Randle cycle. It, it matters enormously. Insulin regulates fat oxidation at 2 different points. One is the transport step into the mitochondria, and the other is further upstream at the release of the fatty acids from the fat tissue. That upstream, or we could call it the first step, is highly Sensitive to insulin. Lipolysis, which is the release of the stored fat into the blood, and that, that fat comes out as free fatty acids, or what I, or maybe what I've just called fatty acids. Sometimes it's called non-esterified fatty acids. They're all the same thing. So lipolysis is what determines whether fatty acids are available in the first place or whether they're in the blood. In a clamp that independently controlled for insulin and some other signals like growth hormone and glucagon. It took less than 2 microunits of insulin per milliliter to reduce fat release by half in, in healthy people. A healthy adult usually has somewhere between 5 and 10 microunits per milliliter after an overnight fast. So fat release is already substantially reduced even before breakfast. So again, they just found that if they bumped the insulin Up to about just 2 microunits, there was already, they were already seeing a reduction in fat burning. So fat release is down even with modest increases in insulin. And, and so the authors, to use some of their own language, they described that the adipose tissue lipolysis is exquisitely sensitive to insulin. Yeah, you could say that again. These 2 control points are not equally sensitive. Remember the control point of fat getting released from the fat cell to be burned, and then the earlier process I mentioned of that carnitine shuttle. Mm-hmm. which is the process of pulling the fat into the mitochondria. Insulin is blocking that too. In a separate study, insulin was infused at 4 increasing rates while both fatty acid release and fat oxidation were measured. The concentration needed to cut fat release in half was roughly 1/3 the concentration needed to cut fat oxidation in half. Insulin reduces the supply of fatty acids at much lower levels than those required to inhibit mitochondrial oxidation. To say all that again in a slightly different way, insulin is better at blocking the fat cell from breaking down its fats than it is from blocking the mitochondria from burning those fats. And we do see this is relevant even with ordinary meals. In one particular study, They had 6, uh, 6 active men cycled for an hour under 3 conditions: after an overnight fast, after an hour drinking, an hour after drinking a glucose solution, and an hour after drinking a fructose solution. And remember, fructose is unique because, yes, it's a carbohydrate, but it doesn't increase— and it doesn't increase insulin. Average insulin in the 50 minutes before exercise was 8 microunits per mL when fasted. It went to 17 after the fructose drink, so it went up a bit, no doubt, but then it went up to 38 after the glucose meal. These are very normal post-meal numbers. 25 minutes into the exercise, fat release was roughly cut in half after the glucose drink compared with the fasted trial, and fat oxidation fell right along with it. They then ran a fourth trial that separated the two control points. The subjects drank the glucose again, but this time fatty acid supply was restored by infusing lipid directly throughout the rest period and the exercise period. They found that restoring the supply of fat raised fat oxidation about 30% above the glucose-only trial, but it still didn't bring it back. up to the level that they saw in the fully fasted state. Part of the reduction came from less fatty acid being released. The rest came from blocked entry into the mitochondria. So a single carbohydrate drink produced both effects. Exercise is used in these studies because it raises fuel use high enough to measure cleanly the fuel use, but the same mechanisms do still matter at rest. Someone eating several carbohydrate-containing meals across the day keeps insulin elevated, for much of those waking hours. Now, you might wonder whether this is really just about the fat in the bloodstream, whether insulin is simply keeping fat locked away so the muscle can't reach it. And I've already touched on this a little bit, but there's still more evidence worth exploring. In another study, 6 endurance-trained subjects cycled for 40 minutes, either fasted or after a large glucose drink, while labeled or tagged Long chain and medium chain fats were infused. The glucose drink raised insulin and reduced total fat oxidation by 34%, so about a third. About a half of that reduction came from circulating fatty acids, and half from the fat already stored inside the muscle itself. That second point is important: intramuscular triglyceride never goes through the blood, so the fat that is within the muscle already. That muscle has its own reserve within its own fat cells. So this is fuel that's already sitting inside the muscle cell. Even still, its oxidation fell, which further points to the inhibition that insulin's able to elicit at the level of the mitochondria directly. All right, so a lot of this is focused on insulin, appropriately. If insulin regulates fuel selection so tightly after a single meal, then sustained differences in carbohydrate intake. should produce sustained differences in fat oxidation, right? And fat burning. And they do. Here's some studies. 5 well-trained cyclists ate a diet that was about 2/3 of all the calories from carbohydrate for 1 week, then switched to a very low-carb ketogenic diet with fewer than 20 grams of carbohydrate per day for 4 weeks. The diets were matched for calories and matched for protein. During steady-state endurance exercise, the respiratory quotient Remember, the measurement of what fuel's being burned fell from 0.83 to 0.72, and glucose oxidation dropped roughly threefold, and muscle glycogen use fell roughly fourfold. So they were burning a lot more fat. Maximal oxygen uptake, so the kind of VO2 max measurement stayed the same. So they— and the time to exhaustion. was unchanged. So they still got as much work out of the workout as they did, but, you know, between both, both diets. So same amount of work was being done with a very different mix of fuels. We see even larger differences, differences in athletes who have been eating this, these differences, these different ways for a longer time. 20 elite ultramarathoners and Ironman distance athletes were trained in, in, in, or used in one study. They were matched for their age, their body size, the competition distance, and even their performance times. Half of these studies ate a conventional high-carb diet and half ate a low-carb diet. So about, um, 10% carbohydrate on, in this last study and 70% fat in, in the low-carb group. And this was for an average of 20 months they found across these subjects. And some of them had been doing this for several years. Peak fat oxidation during exercise was more than twice as high in the low-carb group, and it occurred at a higher percentage of their VO2 max. There was no overlap at all. Interestingly, every single athlete on the low-carb diet had higher peak fat oxidation than every single athlete on the high-carb diet. And then they found that across 3 hours of running, fat provided 88% of the energy in the low carbohydrate athletes versus 56% in the others. Resting muscle glycogen and how much glycogen was used during the run and restored after did not differ between the groups. So as much as people want to say, if you're on a low carb diet, your muscle glycogen is going to get depleted. Well, it wasn't. It was the same in both groups of athletes used the same amount of muscle glycogen. All right, everything so far describes a healthy system responding to the insulin signal as we talk about this Randle cycle or the glucose fatty acid cycle. The problem starts to show up when that response is lost. A specific scientific review examining the entire glucose fatty acid framework found that insulin-resistant muscle. Gets it wrong. In fact, it gets it wrong in, in both directions, and I'll elaborate. Between meals and overnight, when it should be burning mostly fat, the body's burning more carbohydrate. And when insulin rises and it should shift toward burning carbohydrate, that shift doesn't happen as strongly as it does in healthy people. This is what the authors call— this is the birth of the idea metabolic inflexibility. Their verdict in this famous work on, based on the older framework, the Randle cycle, is that they're, they state that the Randle cycle is a valid physiological principle, but it may not explain insulin resistance in skeletal muscle. And this is what's important, where you have a lot of people who say if you're eating more fat, then you're developing insulin resistance. Their work suggests that is just not true. And there is further studies to support this. Researchers sample blood, um, going into the leg. So one of the techniques that scientists can use is sampling blood going into a set of muscle or to a leg muscle and sample the blood coming out, looking at the balance of what's going in versus what's coming out. And based on the difference, they can work out what the muscle is burning, not what the whole body burned, but just that muscle specifically. In obese study subjects, the legs respiratory quotient after an overnight fast sat almost at 0.85, which means the muscle was leaning on carbohydrate at the very time of day when you think after an overnight fast it should have been— it should have shifted over to fat. And when they took muscle biopsies from those same people, they found lower carnitine palmitoyltransferase activity. That's the enzyme that carries the fats into the mitochondria. Along with lower activity of the enzymes that do the burning of that fat, so the machinery for taking up fat and burning it is diminished on top of a hormonal environment that's already pushing the muscle toward burning more carbohydrate. Chronic high insulin creates opposite problems in two tissues. In fat tissue, insulin is no longer effective at restraining lipolysis. So again, this is an insulin resistance. So fatty acids keep leaving. Leaving their storage site, even though insulin is high. This is a combination that you would never see in a healthy person: high insulin and high fatty acids. In muscle, insulin is still effective at blocking fatty acid entry into the mitochondria, so the muscle keeps oxidizing carbohydrate. Fatty acids are being delivered to the muscle, but the muscle can't burn it. Because insulin isn't letting it. And so they accumulate in the muscle, or they accumulate in the liver, or in the pancreas, or any other places that were never meant for long-term storage of fat. Everything to this point measured— that, that's been measured, it's looking at the fuel that's being burned. And that's not necessarily going to be the same thing as how much fat a person is burning or losing. So does this— the next question is through the mini lectures, does the shift in fuel selection show up as differences in body composition? And this is not— there's more nuance here than most people want to admit, and I'm going to make sure you know it. The most direct test of this put 28 overweight and obese adults on 2 calorie-restricted diets. So low— they cut calories, but they put them on 2 different diets in sequence. So each person served as their own control. One version of the diet dropped— one version of the diet dropped carbohydrate below 10% of energy to induce ketosis. The other, the other version of the low-calorie diet followed conventional recommendations at about carbs at 60%. The study subjects ate more on the low-carbohydrate diet. They ate almost 1,900 calories a day as opposed to about 1,500 calories per day, and they lost More weight, more total fat, and more truncal fat or belly visceral fat, roughly 300 extra calories a day, and they still lost more fat. The food— now, this is a kind of a limitation. The food wasn't provided and intake was self-reported. Even though there was a strength that each person was acting as their own control by studying over, there's a potential for error there. Over longer periods, the advantage persists perhaps with low-carb diets, but it gets a little narrower. A meta-analysis of 13 trials running at least 12 months found greater weight loss on ketogenic than low-fat diets, but the difference was as modest as just about a kilogram. Now, I have a problem with all of these trials, everyone including those in the meta-analysis that lumped into that meta-analysis, because every one of these trials that has ever been done. To compare low carbohydrate versus high carbohydrate diets, they all cut calories in both groups. So think about this: the person in the high carbohydrate arm is still eating several hundred calories less than they were before the study began, which means they're eating fewer carbohydrates. So their insulin also comes down. We aren't comparing high insulin against low insulin; we're comparing. somewhat lowered insulin against more lowered insulin. So both groups end up getting kind of the benefit here where they're both having lower levels of carbohydrate relative to where they started, to where they started, even though you'd say, well, one group is going lower. Yeah, but they're both going lower than where they started. So what we need to do— so what ends up happening is that the study gets designed on accident To wash out these differences. So let me—I hope I'm articulating that well because I think it is a significant criticism of every one of these trials that's low carb versus low fat, high carb. They've all been lower calorie, so they're calorie restricted. So even the high carb group is eating fewer carbs than they did before the study started. So they also benefit. Every study that has ever looked at insulin has found that both. Groups experience a drop in insulin. The proper way to actually test the effect and the difference between these types of studies is to actually run it calorically in the opposite direction. We need to raise calories in both groups instead of cutting them. Then nobody is eating less than before, and whatever difference that we see has to come from a strict difference in the composition of the diet. Now There is one single study in the history of all nutrition science that's done this, but it was done on only one person. So this was a case study that was published with a single subject, Mr. Sam Feltham in the UK, and he published this work with, um, Eric Westman at Duke. He ate 5,800 calories a day for 21 days at a time, so 3 different, um, Uh, 3 weeks at a time on 3 different diets, and he would give himself 3 months between each one to go back to baseline body weight. So he had a low carbohydrate diet where only 6% of calories were coming from carbs. Then he shifted over after washing out and getting back to normal to a low fat diet. Then he went to a very low fat plant-based vegan diet. Now again, this was 5,800 calories, so it was hypercaloric. And what was the difference between them? On the low-carb version of eating 5,800 calories, he gained a little over 1 kilogram in the 3 weeks. On the low-fat version of 5,800 calories, he gained over 7 kilograms in those 3 weeks. And then lastly, on the vegan plant-based version of the 5,800-calorie hypercaloric diet, he gained about 5 kilograms. Pretty interesting, isn't it? Again, it's one single study, one case study on one single person, but it is the only one that has ever been done, as far as I'm concerned, the right way. All right, so taken together, the human data support a handful of points: that carbohydrate restriction increases fat burning; it produces at least as much fat loss as fat restriction, and usually more; it does this without requiring as large of of a caloric deficit. And relatively more of that fat that's being lost is coming from the visceral and abdominal fat. Now, in closing, as we try to make some practical sense of this Randle cycle and the glucose fatty acid cycle, I think, is there any truth to the statement when people say, well, their body's just not burning fat? You know, maybe there is. Maybe they're not that far from the truth. Which fuel gets used isn't a matter of which one is more abundant in the blood. The clamp studies, remember, held fatty acid levels constant, and fat oxidation fell anyway if glucose and insulin were up, especially insulin. Insulin decides; it controls the release of fats from fat tissue, and it controls their entry and burning into the mitochondria. That puts carbohydrate in a position that the other dietary variables don't occupy. It's alone. Carbohydrate is the main dietary driver of insulin, and insulin is the main controller of which fuel gets. Burned. So lowering carbohydrate acts on the control points themselves rather than somewhere else downstream. And in conclusion, lowering carbohydrate intake will lower that insulin signal that keeps the body from burning fat. So one way to interpret all of this, the complicated glucose fatty acid cycle, is you burn what you eat. Class dismissed. Until next time, more knowledge, better health.