95: Seed Oils and Insulin Resistance: What the Science Really Says with Dr. Ben Bikman

Hello, everyone, and welcome back to the Metabolic Classroom. I am Professor Benjamin Bickman, a professor of cell biology and biomedical scientist. Today I am your honorary teacher. Thanks for being my guest student. The mini lecture today is exploring one of the more relevant and hotly debated topics within metabolism. Specifically, we are discussing the effects of seed oils and how they influence insulin resistance. If you've caught my talks in the past, you know that I've got a lot of thoughts, certainly about insulin resistance in general, but also about seed oils. Although I don't really consider myself a seed oil guy. I am an insulin guy, an insulin resistance guy. But nevertheless, this is an important topic. A lot of people want to learn more about it, and I thought I'd provide some of my hopefully very informed views on it. So today we'll focus on linoleic acid, which is the kingpin of seed oils. Before we get into the science of it, I thought it might just be fun for those of you, the few of you perhaps, who haven't learned of its history in the human diet, but also how it went from being nothing to the star of the dietary show, really being a kind of involved in a hostile takeover. And then, of course, we will get into the cell biology of the mechanisms that may connect it to insulin resistance. But let me grease the skids a little bit with some seed oils here. There's a little bit of a twist, as you're going to see, and I regret that some people are going to become very inflamed at this perspective. But perhaps linoleic acid itself isn't the bad guy, but rather what it turns into, what it gives birth to. Now, that may be splitting hairs, which a guy like me isn't inclined to do very often, but nevertheless, I think it's an important distinction and it does help us understand the evidence. Why is it that some studies show no effect with linoleic acid, but others do show an effect? Well, let's sort of tease that out. But before we move forward, let's go back. In fact, way back, we have to go back about a century to really explore the introduction of seed oils. Things like soybean, corn, sunflower, canola oil, these weren't a thing in the human diet. We might have used them industrially as a lubricant, but we didn't use them to lubricate our intestines by consuming them. At the time, people largely just used butter or lard or tallow, natural fats that are rich in saturated and monounsaturated fats. Now, this isn't to say they are devoid of the villain of the story. Linoleic acid, the omega 6 polyunsaturated fat, it has always been there, just about 1 to 2% of the fats we ate. So just to really make that clear, even natural fats that come from animals, which again, humans have been eating since time immemorial, and that's one of the reasons, among others, that I'm a great defender of them, they do have linoleic acid. And linoleic acid has been considered an essential fat. That's fine. We get it from those natural sources again, about 1 to 2% perhaps. But then came the industrial food shift, for lack of a better term. By the 20th century, the refining technology made seed oils not only cheap and abundant, but then rather only being used in industrial capacities. They led the charge in this dietary takeover, as I mentioned earlier, and became a very prominent part of the global diet now. But at the time it was things like margarine and used in cooking all over in homes and in restaurants more and more. In a 2011 paper published in the American Journal of Clinical Nutrition, a group at the NIH is the really. It's the group that we primarily can thank for exploring the historic trends. But they showed that linoleic acid intake skyrocketed. Soybean oil went from essentially nothing in the human diet in 1909 to up to 20 pounds per person in 1999. And indeed it's continued since then to increase to the point where soybean oil is the single most commonly consumed source of calories from fat. So more people are getting their fat calories from soybean oil than any other fat. A 2015 Advances in Nutrition study confirmed the trend by actually tracking the amount of linoleic acid in fat cells. And they found that it had gone up over 136% since the 1950s. That's important because linoleic acid is not a fat that we're going to make. So insofar as this study was able to determine an increase in the amount of stored linoleic acid, it is further evidence of just how much more we are consuming now. Why is this relevant? Part of it is relevant because saturated fats got demonized. That explains partly why seed oils became so such darlings of the dietary world post World War II. The scientific professionals really pushed these so called heart healthy oils, considered heart healthy because they have an effect of reducing LDL cholesterol. Now that is a topic for another time, but you could look up the Minnesota Coronary Experiment in the Sydney Diet Heart study to just get an appreciation of very big long term studies that use these seed oils and found yes, LDL cholesterol went down, but unfortunately and inconveniently mortality went up, so they actually were dying more. But again, that's a bigger topic and certainly relevant. I encourage people to go beyond perhaps just what I'm focusing on in order to appreciate just how harmful seed oils can be. My intention in this mini lecture is to just focus on the metabolic aspect with a specific emphasis on insulin resistance. So let's come back to that. Well, there is obviously a trend. As much as nutrition science likes to rely on big global surveys and trends, this one's too obvious to ignore. Insulin resistance has become the common, the most common health disorder worldwide at the same time that refined seed oils have become the most commonly consumed fats. This is a remarkable coincidence and indeed I do think it goes beyond coincidence. I think there is some important connections here. Insulin matters, of course, insulin resistance does so much because not only it's so common, but it contributes to essentially every non infectious chronic disease, either directly causing it or or exacerbating it. So think problems like type 2 diabetes, Alzheimer's disease, heart disease, infertility and more. Of course, I wrote an entire book on the topic of why insulin resistance is such a problem. But let's come back to linoleic acid. The main culprit in why refined seed oils are, I would say, appropriately vilified. Is linoleic acid itself the culprit or is the things it gives birth to, those double bonds in that polyunsaturated fat certainly make it unstable. And the more unstable the fat is, the more likely it is to undergo peroxidation. So peroxidation can be this process of manipulating the fat, whether it is in a fryer or in our body. We end up converting the linoleic acid into its production. The peroxidation products, like some of the big ones, 4h and e or 13h o d e 13 hold and other oxy lipids, those are the main peroxidation products or the little villains that are actually potentially the problem. But I'm getting a little ahead of myself. Let's explore the evidence now. As I do so, I hope you will appreciate the progression that we're going to take. I want to start with the simplest of biological models that we biomedical scientists use, namely cell cultures. So we're going to look at the evidence in cells, then we're going to move on to a little bit of animal Evidence. Now, the nice thing about both of those models is while they are stepping up in complexity, going from isolated cells to animals is certainly a step up with humans being the most complex, not just because of our biology, but because of our, well, the variety and the complication of perfectly trying to control every, in every aspect. Whereas the animal studies, they're all basically ident. But as we with the simpler model comes a little more clarity. All right, well, let's look at that clarity. Firstly, a 2016 molecular metabolism paper by Sassen et al. Tested fat cells or adipocytes with unoxidized or what they intended to be, just straight linoleic acid. When they did so, there was no effect of insulin resistance that the fat cells treated with linoleic acid. They had normal insulin signaling. Glucose uptake as measured in the study was perfectly fine. However, when they treated these adipocytes with the OX, with the peroxidation product 4H&E, things didn't work out so well. They found that the 4h and e was compromising a particular protein called the Insulin Receptor Substrate 1 or IRS1. In order for normal insulin insulin signaling to occur. When insulin binds to its insulin receptor, the most immediate next step is going to be that IRS one gets phosphorylated or activated. They found that that went down by about 50%. No surprise there was a reduction in glucose transport. This already in this one single study is highlighting an important variable, that linoleic acid didn't directly cause a problem, but its peroxidation product did. Similarly, a 2023 study in Geroscience by Gutierrez Mariscal et al. They hit visceral cells with 4Hne. Not only did they see some substantial mitochondrial damage, which is extremely relevant, indeed relevant in multiple other disorders that I'm not discussing, perhaps including cancer, but they also found that insulin sensitivity dropped, but linoleic acid alone had no real effect. And finally, just to round out some evidence on cells, a 2019 Redox Biology paper by Zhang et al. Saw 4H compromise muscle cell insulin signaling, this time by inhibiting the activation of a protein called AKT. I just mentioned IRS1. Well, go a few steps down in the insulin signaling pathway and you get to akt. AKT is a critical activation point. If it is compromised, insulin signaling is compromised. So in my view, based on these studies and a handful of others from the based with cell culture used using cell culture, the pattern appears clear. Linoleic acid is kind of neutral in this regard, but its peroxidation products are not. Now, the animal data Kind of scales it all up a bit. A 2015 Journal of Biological Chemistry study by Beam et al. Fed rats linoleic acid rich diets. Insulin resistance was observed. And in particular the. In particular they looked at insulin resistance in the heart and found that there was evidence of it. A 2020 paper published in the journal Scientific Reports gave mice soybean oil obesity. Insulin resistance went up and there was also importantly a threefold increase in some of the peroxidation products like 13HOD and 9HOD. That was the Hode Oxylipin that I mentioned earlier in the liver. So huge increases in the level of these peroxidation products in the liver of these mice. But just to compare it to another group, mice in the same study that were fed a diet enriched with coconut oil fared much better, not only gaining less weight, but maintaining better insulin signaling. So things are getting a little more complicated now when we get to humans, it actually gets even more complicated. So the cell evidence suggested with, with all of the clarity that is largely only capable with cell cultures, that linoleic acid itself wasn't, didn't appear to be a problem, but its peroxidation products were. When you were feeding the animals diets high in seed oils enriched with linoleic acid, of course, things appeared to be a little more consistent that the animals were developing insulin resistance. Now with humans, again, it does get complicated. You can, with a lazy search of databases, get a lot of correlational studies. Correlational studies, of course, are based on questionnaires and they are so unreliable, either because of the nature of the questionnaire itself, the nature of the response from the test subject, effects in their ability to recall what they eat perfectly, or even, unfortunately, a bias of the scientific group. They're so unreliable that I am honestly not even going to waste my time or your time to review them. I don't think any nutrition decision should ever be based on a correlation study. And I have very low regard for such tools. So I only want to rely on clinical studies with direct interventions and clear outcomes. But that of course narrows it down. And there's not a lot of studies that we can rely on. And every one of them is confounded by the fact that they use diets that are always high in carbohydrates as well. So to put a fine point on this and make it clear, I'm unaware of any diet that has used primarily a low carbohydrate approach and then compared different fats. So what would it look like if you had a group of humans eating ketogenic diets that was in one group perhaps enriched with soybean oil as the main fat, and another enriched with, you know, animal fats or fruit fats like coconuts or olives. I'm unaware of any study that has ever done this. So we have to consider this and I'm going to come back to that consideration after I review some of these studies. And I can already imagine the gnashing of teeth that's going to come from reviewing these. But I hope I'm going to be able to provide some nuance and context at the end of it all. So, firstly, let's consider a 2002 study from Diabetologia. That's a very good journal in the realm of metabolism, and it was published by Summers and other authors at Oxford. In this trial, they used a fairly small group, which itself is a problem. Only 17 participants. And with just these few participants, they had quite a variety of subjects with type 2 diabetes, people who were obese and non obese, also non diabetics. So you can see that they have multiple groups within just a group within a study population that only included 17 people. That is very low. However, there aren't a lot of studies for us to choose from. Now, they had these study subjects follow two five week diets, one rich in saturated fats from dairy and of course, processed foods, and another emphasizing polyunsaturated fats from spreads and oils like sunflower or safflower. Now, as I noted earlier, all of these diets are confounded by fairly high carbohydrate levels. And so the carbohydrates were almost up to 300 grams a day, which is pretty heavy. That is an important consideration. These are high carb diets. Now, nevertheless, with all of those limitations in mind, they used a gold standard to measure insulin resistance, namely the hyperinsulinemic euglycemic clamp. And the team found that the PUFA rich diet improved insulin sensitivity, particularly in the women in the non diabetic subject, compared to the saturated fat rich diet. Now, I'm going to revisit this in a moment and provide some, I hope, helpful context. But again, compared to the saturated fat rich diet, the PUFA rich diet fared better with regards to insulin signaling. Now, let's move on from this study and explore other trials that again specifically investigate linoleic acid rich polyunsaturated fats and their effects on on insulin resistance. A 2018 study by Miller et al. In the Journal of nutrition randomized 29 healthy subjects to a diet providing 15% of energy from omega 6 PUFAs, of course predominantly linoleic acid rich from seed oils versus a saturated fat rich diet. For four weeks they observed a slight reduction or improvement in the HOMA insulin resistance score indicating a subtle improvement in insulin sensitivity on the PUFA rich diet. Similarly, a 2007 trial by Gough et al In a good journal metabolism involved 35 overweight men, in this case consuming 10% of energy from linoleic acid via sunflower. And this was for six weeks and they found about a 10% enhancement in insulin sensitivity based on their tests. Now this isn't always unanimous. There is A study in 2015 published in the journal Diabetes which is a very good journal again and they looked at 61 obese insulin resistant adults and they had a linoleic acid rich diet which was 8% of energy from sunflower oil versus a saturated fat rich diet for 10 weeks. They found no significant changes in insulin sensitivity. So these trials suggest that linoleic acid from seed oils doesn't inherently cause insulin resistance. Now as another sort of. Well, a brief tangent, I don't intend and didn't when I put some thoughts together for this lecture to focus on saturated fats. I have made it very clear in the past that I'm a great defender of natural fats which are often rich in saturated fats. But I'm cognizant of the fact that I just highlighted some studies that suggest saturated fats are worse on insulin resistance than polyunsaturated fats fats. I do think context matters, which I'm going to come back to in just a moment. But before going on to that, I wanted to highlight some of the biochemistry that would indicate or reveal how it is that linoleic acid could be contributing to insulin resistance. Because I did show some studies. As much as the human studies get a little unclear, the animal studies are a little clearer and the cell culture based studies suggest that there's some critical nuance. Namely that the linoleic acid in its unadulterated state may be fine, but it's often adulterated, it is often altered. So first, how might we reconcile all of this? Peroxidation products like 4H&E sabotage insulin signaling. I'd mentioned the IRS one effect in that SA paper. Also, there's evidence that the peroxidation products can hit AKT in a paper in 2019 and then a 2018 paper in the journal Biochemica A Biophysica Acta. Some of those titles are sort of a Greek origin, and that was a study by Ramsden and they fed mice linoleic acid versus its oxidized derivatives. And peroxidation tripled the amount of oxylipins that accumulated. Linoleic acid really didn't. So there's this direct effect again of the peroxidation product. Second, there's inflammation. These oxylipins will activate strongly a primary critical mediator in inflammation, namely NF kappa B, which when activated, dramatically increases the production of pro inflammatory cytokines. Well, anytime you're increasing inflammation, which I've discussed abundantly in the past, you are promoting insulin resistance. So if these peroxidation products are the problem, and I do think they are, part of the connection to insulin resistance could be via inflammation. So in that regard we would consider it where these peroxidation products are acting as a secondary cause because they're relying on a primary pathway, namely the inflammation aspect. But third, I can't really talk about mediators of insulin resistance without invoking my favorite, which is the sphingolipid ceramides. Now, why is it my favorite? It's because that is the in fact specific topic I've focused on the most as a direct cause of insulin resistance throughout my career, particularly during my fellowship studies and early years of having my own lab here at byu. Linoleic acid itself is not a backbone or a directly feeding into ceramide biosynthesis, but its peroxidation products might activate that process. A 2019 Redux Biology study linked four H&E to ceramide accumulation via the activation of stress pathways in muscle tissue. Similarly, a 2017 paper published in the Journal of Lipid Research suggests that linoleic acid's arachidonic acid metabolites could stoke inflammation and then that is boosting ceramide accumulation. So linoleic acid is perhaps less and less the primary culprit, but its peroxidation products more and more appear to be the problem. Now, to come back to some of the nuance that I'd mentioned earlier as we near the end of this mini lecture, I presented some inconvenient data where in the humans, the diets that were higher in saturated fats appeared to worsen insulin resistance compared to the diets high in the polyunsaturated fat from seed oils. Here's the context that I believe is relevant and perhaps helps us reconcile some of this and perhaps allows a lot of various camps to still remain friendly in this discussion. Perhaps saturated fats are a problem when consumed with carbs, because when Carbs are consumed, the insulin elevation will directly promote the saturated fats like palmitate to go into the ceramide biosynthesis pathway. That is important because palmitate is not only the most commonly consumed saturated fat, but it is the backbone of ceramides. Again, the lipid that I've focused on the most as a cause of insulin resistance. But the saturated fat would need to be directed into that pathway. Insulin does that. We've published that paper. I've done that myself, finding that chronically elevated insulin does stimulate ceramide accrual by shunting palmitate into ceramide biosynthesis. So perhaps if a diet is high in carbs, combining it with high in saturated fats might be more problematic, specifically with regards to insulin resistance. And then similarly with linoleic acid, perhaps it's primarily a problem when it's consumed without antioxidants, or rather consumed in a body that has a low antioxidant capacity. If peroxidation is allowed to occur, then the otherwise benign linoleic acid, at least in the context of insulin resistance, is enabled to wreak havoc on insulin signaling via the peroxidation product production. Let me just state all of that again, based on the human evidence primarily, and even going through the animal and the cell culture work, perhaps saturated fats are confounded or become problematic when consumed with carbohydrates in a way that polyunsaturated fats aren't, because the high insulin coming from the glucose spike will direct saturated fats into ceramide biosynthesis. But linoleic acid is not a substrate, it's not a building block for ceramides, again, which I believe to be the main mediator of insulin resistance. And so the high insulin level isn't directly making the linoleic acid promote ceramide accrual. In contrast, or sort of in parallel, the linoleic acid becomes more problematic, less benign, when there is a greater tendency for it to undergo peroxidation, which is happening in the context of a high oxidative stress environment, and then the high oxidative stress, and then the inflammation that comes with that will then promote insulin resistance. But if the linoleic acid is defended from undergoing peroxidation, then perhaps it's less of a problem. Now, importantly, linoleic acid undergoes peroxidation so readily that even when you are getting. Even if you were to drink straight vegetable oils, which is really seed oil, some of it is already, has under. Already undergone peroxidation. That's a very important consideration. Also, it's important to consider that a lot of the seed oils that people consume come with a lot of things that have been added in order through the chemical industrial process of taking something like soybean, which doesn't naturally give an oil, but then extracting an oil from it. So what's the takeaway? As we come to the end now, I believe at least some conclusion could be that linoleic acid is a naturally occurring fat that you cannot completely avoid. And indeed, you shouldn't try to avoid it completely. You can't. It exists in nature, in all of these natural fats, particularly very healthy animal fats. But as we flood our diet, which we have done globally, we eat too much of it, perhaps in the presence of an oxidative stress environment, we've already stressed it through frying it or et cetera. Then we're brewing the accumulation of all these harmful peroxides, like 4H and E and 13HOD and more. And then those will, I believe, the evidence strongly suggests, directly compromise insulin sensitivity via the distinct mechanisms that I mentioned earlier. So as much as I am a defender of dietary fat, I do suggest that we enjoy natural fats more than refined seed oils, especially those that come from animals and fruits, the fruit fats being coconuts and olives. Thanks for listening to this short lecture. I hope you'll join me next time. And until then, more knowledge, better health.