March 19, 2019

A New Look at Insulin, Glucagon & the Pancreas (a.k.a. ITIS part 9)

“Contrary to popular belief, insulin is not needed for glucose uptake and utilization in man.” (Source)

What? Insulin is not needed for glucose uptake? Did I just blow your mind a little?  If so, hang on to your hat. Lots more of that to come.

As I mentioned in my previous post on the personal fat threshold concept, what I enjoy most about writing my blog is that I get to share with you the fascinating and surprising things I learn. And one thing I can say with certainty is, the more I learn, the less I know. It seems like I barely hit publish on a new blog post before coming across a bunch of papers that teach me even more about the subject in question, or make me rethink what I wrote about it in the past.

One subject I’ve learned more about since I last wrote about it is insulin. If you’re new here, I recommend digging into my 8-part series on insulin. If I do say so myself, it’s some of the most important and educational stuff I’ve written.  But you don’t need to have read that to understand today’s post.

If you’re accustomed to thinking about insulin as a “blood sugar hormone,” you’re about to have your world turned upside downWhat I’ve learned about insulin over the past couple of years makes me think that lowering blood glucose might be one of the least important and impressive things it does.

Another very long post coming your way here, so grab a coffee or some pork rinds, and happy reading!

Before you dive in, though, I recommend scrolling way down to the bottom of this post where it says “End.” You might want to spend a while reading the whole thing, or you might not…that will help you decide. 

Glucose Transporters (GLUTs)

Owing to the chemical structure of glucose, glucose is water-soluble and can travel freely in the bloodstream. It can not, however, pass freely through cell membranes, which are lipid (fat)-based. In order to get inside cells, glucose is escorted through via glucose transporters (GLUTs). There are several different kinds of GLUTs, and different cell types use different GLUTs, although there’s some overlap. (For example, neurons use GLUT3, fat cells and skeletal muscle cells use GLUT-4, and cells at the blood-brain barrier use GLUT1. Sperm cells use GLUT5, which is actually a fructose transporter.)

The thing to know here is, most GLUTs do not require insulin to get glucose into cells. 

“Contrary to popular belief supported by the leading physiology and biochemistry textbooks, there is sufficient population of glucose transporters in all cell membranes at all times to ensure enough glucose uptake to satisfy the cell's respiration, even in the absence of insulin. Insulin can and does increase the number of these transporters in some cells but glucose uptake is never truly insulin dependent.” (Source)


If the primary role of insulin was stimulating glucose uptake into cells, then people with type 1 diabetes should have impaired glucose uptake. But they don’t. In fact, glucose uptake is actually increased sometimes—even in the context of insufficient insulin. (So why then, is their blood glucose so high? We’ll get to that soon.)

“…in the face of hyperglycaemia, tissue glucose uptake is usually increased above normal even when insulin deficiency is severe. This cannot be reconciled with the concept that insulin is required for glucose uptake by insulin-sensitive tissues. Indeed it proves beyond question that insulin is not required. We now know the detailed mechanisms involved and can explain this. Glucose uptake by all cells is by means of a specific transport protein (glucose transporter) of which at least six isomers (Glut 1 to Glut 6) are known. Glucose is a highly polar substance, being freely soluble in water but insoluble in fat. It cannot enter cells except through the specific transport system utilizing Glut 1–6. Glut 4 is the transport protein present in muscle and adipose tissue, which is known to be ‘insulin sensitive’. This means that, in addition to the transporters resident in the cell membrane at any given moment, there is a pool of glucose transporter molecules in the cytoplasm of the cell which can be recruited in response to a rise in plasma insulin, to join those already in the cell membrane in the fasting state. […] even in the fasting state or in a state of absolute insulin deficiency, there are sufficient glucose transporters already in place in the cell membrane to allow glucose uptake to exceed that of a normal individual when the gradient of glucose concentration across the cell membrane is sufficiently high.

This ‘mass action’ effect accounts for the observations which show unequivocally that tissue glucose uptake can exceed normal even in the face of severe insulin deficiency such as occurs in uncontrolled diabetes mellitus.” (Type 1.) (Source)

Most glucose transporters operate just fine whenever a certain blood glucose concentration is reached. (Each of the GLUTs has a different threshold, so to speak, at which they’ll start accepting glucose. Some of them will take glucose in even when BG is very low; others need BG to get pretty high before they’ll let glucose in. As I discussed way back in my series on the metabolic theory of cancer, some cancer cells express GLUTs that let glucose in when BG is very low, which explains why ketogenic diets, by themselves, aren’t a magical cure for cancer—because even when BG is very low, those wily cancer cells will still be able to suck it in like champs.  [NB: I still think keto is a fabulous adjunct to conventional treatment and should absolutely be offered as an additive option. I’m just saying there’s a reason why keto alone isn’t a slam dunk. Plus, cancer cells can also ferment the amino acid glutamine. If you’d like to learn more about this, one of the all-time best podcasts I’ve ever heard is this interview with cancer researcher Thomas Seyfried. It gets my highest recommendation.])

Insulin stimulates GLUT4s to move to the cell membranes of muscle cells, including the cardiac muscle cells in your heart. This is why it’s said that GLUT4 is an “insulin sensitive” GLUT. BUT: insulin isn’t the only stimulus for glucose uptake into muscle cells.  Muscle contraction and stretching (exercise, general physical movement) will facilitate GLUT4 action even in the absence of insulin, which is why exercise is said to stimulate “non-insulin mediated glucose uptake,” at least with regard to GLUT4s. 

Aaaaanyway, the point is, if the majority of glucose uptake into cells is not insulin-dependent, why have we spent so many decades surrounded by the notion that facilitating cellular glucose uptake is insulin’s most important job?

The Dark Ages

I’m going to quote liberally from a paper called Insulin: understanding its action in health and disease. This is a must-read if you want to nerd out and really dig into this stuff. (You don’t need a PhD to understand it, though. It’s very well-written and not overly dense with jargon.)

Insulin was discovered in 1921. A Scottish scientist, Sir Edward Schafer, had written about an unidentified substance that he called “insuline,” at least as early as 1913.

“His description of how he thought the hypothetical substance ‘insuline’ acted in the body is remarkable because the passage of time has shown him to be correct almost word for word. Things have been confused, however, by a 20 yr ‘black age’ of endocrinology (between approximately 1960 and 1980), where leading scientists—through extrapolating beyond their new discoveries—confused scientific thinking and teaching. They formulated new hypotheses based for the first time on hard scientific evidence but they got it badly wrong through extrapolating (incorrectly) from in vitro experimental data in rat tissues to in vivo metabolism in humans.

The effects of this ‘black age’ are still with us because these incorrect hypotheses have, with the passage of time, been turned into dogma and become cast into ‘tablets of stone’ in undergraduate textbooks. They are also carried forward into postgraduate teaching. For example, even in well respected texts it is still common to find statements such as ‘The basic action of insulin is to facilitate glucose entry into cells, primarily skeletal muscle and hepatocytes.’” (Source)

In case you didn’t catch what they said, it’s incorrect that the basic action of insulin is to facilitate cellular glucose uptake.

“Current dogma would have us believe that administration of insulin to somebody with severely deranged diabetes suddenly and miraculously allows the cells in the body to breathe again and be restored to their former healthy state. This is, as we have seen, untrue, so it is amazing how long this dogma has persisted.” (Ibid)

“These actions of insulin in vitro were discovered in the late 1950s when it was also shown that insulin stimulated glucose uptake by rat muscle. It was extrapolation of this last observation in rat muscle to explain the pathophysiology of diabetes that was erroneous. The consequence of this error was the (fallacious) concept of insulin being ‘required’ for glucose entry into cells rather than just accelerating glucose uptake. The hyperglycaemia of diabetes was interpreted as a ‘damming back’ of glucose in the blood stream as a consequence of a lack of insulin. This became established teaching and, although the concept was shown to be erroneous in the mid-1970s, the teaching has not changed. Consequently, therapy has been based on a flawed concept.” (Ibid)

So what’s the right concept, then? I don’t claim to know, but I’m happy to share some ideas with you. 

Blood Glucose Regulation, Diabetes, etc.

“It is understandable, but nevertheless troubling, that the historic dimensions of the discovery of insulin in 1922 have distorted scientific and clinical perspectives of hormonal dysregulation in diabetes for so long.” (Source)

When blood sugar is high, especially chronically—like in type 2 diabetes, for example—we tend to think it’s because cells can’t or won’t take in any more glucose. And this is likely at least part of the problem, but only part of it. What if the body itself is producing too much glucose?

When your kitchen sink is filling with water and water is spilling onto the floor, what’s the problem? Is the drain clogged so the water can’t exit normally, or is the faucet turned on too high, so too much water is coming out too quickly and overwhelming the drain’s ability to clear it? Maybe both, right? It could be too much water output and too little water disposal. So in chronic hyperglycemia, is the problem one of disposal (cells can’t take glucose in fast enough) or one of output (the body is making too much glucose)? Maybe both:

“Endogenous glucose production is excessive before eating and fails to appropriately suppress after eating in people with type 2 diabetes. This is due in part to impaired insulin-induced suppression of endogenous glucose production…” (Source)

An elevated fasting blood glucose is not the result of decreased cellular uptake of glucose, but rather, overproduction of glucose. Glucose can be produced a few different ways, but let’s focus on only a couple of them for now. One way is glycogenolysis, which is the breakdown of stored carbohydrate in the liver. (Glycogen is also stored in muscle tissue, but glucose from muscle glycogen can only be used in those muscle cells; it is not released into the bloodstream, so it doesn’t affect blood glucose.)

The other way is gluconeogenesis. Yes, GNG, the dreaded bogeyman of keto zealots everywhere. The thing that scares people away from eating adequate protein on ketogenic diets. *Sigh.* (Worried about GNG? I wrote a detailed post on it a while back. Check that out here.) GNG is making glucose out of things that are not glucose, such as amino acids.

GNG is stimulated by multiple things, but the one we’re most concerned with here is the hormone glucagon. Insulin is secreted by cells in the pancreas called beta cells (β-cells), while glucagon is secreted by alpha cells (α-cells), which are in close physical contact with the β-cells. The endocrine part of the human pancreas contains about 48-59% β-cells and 33-46% α-cells, with the rest comprised of δ-cells (delta cells), which produce a hormone called somatostatin. (These are not the only cells in the pancreas. There are other cells with ­non-­endocrine functions, such as producing digestive enzymes.)

As a bit of an aside, but something not totally unrelated, research done on insulin and glucagon secretion in animals (rodents, specifically) might not translate 100% to humans, because the physical structure of the pancreas differs somewhat: “Although the islets have a similar cellular composition among different species, that is, human, rat and mouse, their cytoarchitecture differs greatly. Although islets in rodents are primarily composed of β-cells located in the center with other cell types in the periphery, human islets exhibit interconnected α- and β-cells.” (Source)

People have become afraid of glucagon because they think it’s going to raise blood glucose to pathological levels. In a healthy person with a functioning pancreas and good insulin sensitivity, this does not happen. It does keep blood glucose from tanking too low by raising it a little bit, when it needs to be raised. (In fact, the name glucagon comes from glucose agonist.) We like glucagon. If you’re following a ketogenic diet to lose body fat, you love glucagon. Glucagon is your best friggin’ friend.

Think of glucagon as a fuel mobilizer. Glucagon stimulates release of things from storage, such as glucose (from liver glycogen), amino acids (from body protein), and fatty acids (from your fatty *ss from your adipose tissue). If you want to burn fat, you first have to get some fat out of your fat cells. Glucagon is one of the things that signals your adipose tissue to release these fatty acids. (If you want to use cash, you have to go to the ATM and get some first). Glucagon isn’t the only hormone that does this, but it’s the only one we need to focus on here. Glucagon stimulates autophagy and activation of brown adipose tissue, while inhibiting the synthesis of new fat. Glucagon also stimulates ketogenesis, by the way. If you want to be in ketosis, glucagon is a good thing to have around. If you’re fasting, or on a ketogenic diet, you have glucagon to thank (in part) for keeping you alive.

I did say that glucagon stimulates release of amino acids (AAs) from body protein. This is mainly so these AAs (more specifically, their carbon skeletons) can be used in the Krebs cycle to produce ATP, or so they can be converted to glucose for various purposes, one of which is also generating ATP. There’s no need to fear muscle breakdown from glucagon unless you have type 1 diabetes and do not have access to insulin. (More on this in a bit.) If you have a healthy, functioning pancreas and decent insulin sensitivity, glucagon is your friend.

(If you’re worried about glucagon or gluconeogenesis on a low-carb or keto diet, I beg you to watch this video from Dr. Ben Bikman. Worth every second of your time. The effect of glucagon on blood glucose is very different depending on how full of glycogen your liver is. The higher your liver glycogen, the higher your BG will be under the influence of glucagon, but if you’ve fasted or you’re on a low carb or ketogenic diet, glucagon isn’t going to put your BG into the stratosphere.)

So, glucagon mobilizes things. Gets them out of storage and into the bloodstream. Insulin does the opposite. Insulin takes things out of the bloodstream and puts them in storage. Insulin stimulates muscle tissue uptake of amino acids, liver and muscle uptake of glycogen, and adipose tissue uptake of fatty acids. (As Dr. Roger Unger said, “Insulin’s role as a lipogenic hormone is underplayed. We know you can’t get fat without insulin.”) 

Insulin inhibits all the things glucagon stimulates: lipolysis (breaking down fat), proteolysis (breaking down muscle), glycogenolysis, gluconeogenesis, and ketogenesis. And insulin increases the things glucagon decreases, like glucose uptake and lipid (fat) synthesis.

Physiological Actions of Insulin and Insulin Counter-Regulatory Hormones

Looking at this chart, there are three hormones that raise BG and only one that lowers it. This suggests that, evolutionarily speaking, we likely had a more pressing, dire, and probably more frequent need to raise BG rather than to lower it. After all, we might have had to chase prey down, or run as fast we could to keep from becoming prey, and we would have needed to mobilize glucose fast. But we didn’t have corn syrup, cheez doodles, cream-filled cookies, jelly donuts, and all you can eat pasta and breadstick buffets. So what else does insulin do besides lower BG—which can happen without insulin anyway?

Insulin is an anabolic hormone. It stimulates/facilitates growth. Most of us know only too well that it stimulates growth of fat tissue like mad. It also appears to stimulate growth in other parts of the body, like having skin tags or an enlarged prostate gland. Chronically elevated insulin might also be a contributing factor (one of many) to facilitated growth of cancerous tumors, and inhibiting insulin action might be a beneficial adjunct to other therapies.  

Insulin also prompts the kidneys to retain sodium (a major driver hypertension) and to retain uric acid (a major driver of gout).

The bottom line is, insulin does a whole lot more than lower blood sugar. And since most cells can take up glucose just fine even without insulin, does insulin have some other actions with regard to managing BG? Might it have an effect on the output, and not just the disposal? To return to the sink analogy, could insulin be affecting the spigot, rather than just the drain?

Prof. Roger Unger, Type 1 & Type 2 Diabetes

Much of what I’m about to share comes from Dr. Roger Unger. He’s co-author on the paper, Glucagonocentric restructuring of diabetes: a pathophysiologic and therapeutic makeover, and his video lecture from 2014 is absolutely, 100% a must-watch. I’ve watched it 5 times and I learn something new every time.

If we think of insulin as an anabolic hormone, then we can also think of it as anti-catabolic. It prevents things from being released and broken down. In this sense, insulin doesn’t lower BG by clearing glucose out of the blood; it does it by signaling the liver to stop releasing glucose.

“Impaired insulin-mediated suppression of hepatic glucose production (HGP) plays a major role in the pathogenesis of type 2 diabetes (T2D).” (Source)

Let’s think about type 1 diabetes. In T1D, autoimmune destruction of β-cells means these folks produce little to no insulin. If lowering blood glucose by pushing glucose into cells was the main action of insulin, then the only problem we would see in T1D is very high BG. But we see much more than that, right? Without insulin injections, people with T1D don’t just have high blood sugar. They have super high ketones (diabetic ketoacidosis) because they’re breaking down fat tissue like crazy, and they also waste away in general, because they’re breaking down muscle tissue like crazy as well. Without adequate insulin, the actions of glucagon are left to proceed unchecked. So it’s not so much that insulin has magical powers of its own; the thing it’s doing—the thing that doesn’t happen in type 1—is that it puts the brakes on glucagon. 

Apart from the complications of chronic hyperglycemia, the pathology of T1D is essentially the result of glucagon run amok. So one of insulin’s big jobs is to keep glucagon in check. (Which is why these two are “counter-regulatory” hormones: they counter the effects of the other.)

“Glucagon concentrations tended to be higher in people with either mild or severe diabetes before glucose ingestion and did not suppress and, if anything, paradoxically increased after glucose ingestion.” (Source)

“Thus, people with type 2 diabetes have excessive rates of hepatic glucose release, impaired hepatic glucose uptake, decreased hepatic glycogen synthesis, and decreased uptake of extracellular glucose.” (Ibid)

“When insulin is administered to people with diabetes who are fasting, blood glucose concentrations falls. It is generally assumed that this is because insulin increases glucose uptake into tissues. However, this is not the case and is just another metabolic legend arising from in vitro rat data. It has been shown that insulin at concentrations that are within the normal physiological range lowers blood glucose through inhibiting hepatic glucose production.” (Source)

“…suppression of endogenous glucose production was markedly impaired in the subjects with type 2 diabetes whether measured during the nondiabetic or the diabetic insulin profiles. Thus, a delay in insulin secretion results in higher peak glucose concentrations, and a decrease in insulin action results in prolonged hyperglycemia, which is particularly severe when decreased glucose clearance is accompanied by excessive rates of endogenous glucose production.” (Source)

This line appears to be saying the problem is both glucose output and glucose disposal. (The faucet is stuck in the on position, and the drain is clogged. More water [glucose] is pouring out, and it has nowhere to go. Water spills onto your floor; glucose builds up in your bloodstream.)

“People with type 2 diabetes have excessive rates of endogenous glucose production that fail to appropriately suppress after eating. Rates of gluconeogenesis and perhaps glycogenolysis are increased early in the evolution of diabetes.” (Ibid)

You’re getting message loud and clear, right? The postprandial (after meal) surge in glucose in diabetics comes less from the food they just ingested and more from the liver continuing to release glucose even when it’s not “supposed to,” because insulin should be suppressing this. It appears to be a failure of the liver to respond to insulin by stopping glycogenolysis. It could also be the pancreas, itself—the α-cells, specifically—failing to respond to insulin by stopping glucagon secretion. More on this in a bit, ‘cuz it’s a fascinating issue.

“Insulin-induced stimulation of hepatic glucose uptake is impaired in people with type 2 diabetes. This leads to lower rates of hepatic glycogen synthesis primarily due to reduced uptake of extracellular glucose presumably because of inadequate activation of hepatic glucokinase.” (Source)

In plain English, the above quote says: in people with T2D, the liver doesn’t take up glucose like it’s supposed to, because of inadequate activation of one of the key enzymes involved in this process (hepatic glucokinase). My question is, could it also be a personal fat threshold thing? If liver glycogen is already full, then the liver can’t take up any more glucose, so the glucose has to stay in the bloodstream for a while.

α-Cells, β-cells, and their Crucial Juxtaposition

It’s an understatement to say I don’t envy people with type 1 diabetes. I cannot imagine how difficult it must be to figure out how much insulin is appropriate to dose at mealtimes or as a basal amount. It’s not just about the amount of carbohydrate you eat, y’know. When you have type 1 diabetes and you have to deliver insulin to your body, rather than having it produced on demand at the right time, and in the right amount, you also have to account for your protein intake and other things that affect insulin sensitivity and glucose disposal, such as your sleep the night before, your stress levels, whether or not you exercised, and more. If you don’t have type 1 diabetes, and most of your β-cells are intact and functioning properly, count your blessings. This is not something to take for granted!

There’s a reason it can be so difficult for type 1 diabetics to manage their blood sugar, and also why type 2 diabetics, who have plenty of insulin, might have some of the same difficulty. 

Before I go on, I want to be clear that many people with T1D have accomplished remarkable BG stability, with A1c measurements in the 4-5% range—better than literally millions of non-diabetics. T1 diabetics on low carb or ketogenic diets use substantially less insulin than those eating high carb diets. To be clear, they will always need some insulin. Unlike people with type 2 diabetes, people with type 1 will never be able to completely discontinue insulin injections. But by consuming very little carbohydrate, they’re able to manage BG with much lower doses, and this results in far fewer occurrences of hypo- or hyperglycemia. (If you know the work of Dr. Richard Bernstein, he calls this “the law of small numbers” – the less insulin you need to use, the less volatility there’ll be in your blood sugar. When you need to use huge boluses, then you have a greater risk for huge ups and downs.)

(OT: if you’ve heard that low-carb or ketogenic diets are not safe for type 1 diabetics, that is patently false. I wrote about this here.)

Getting back on message:

Much of the following comes from Dr. Roger Unger’s 2014 talk, “A New Biology For Diabetes.” (Watch this mindblowing video here.)

Dr. Unger would agree with the author of the paper talking about the “dark ages” of knowledge on insulin. There is a terrible inertia when it comes to displacing scientific dogma, even when the dogma is obviously incorrect. Unger said, “The flood of new information and facts in all branches of science far exceeds the ability of the scientist to assimilate and use the new information to displace dogma that are no longer valid.”

In his lecture, Unger used the example of the geocentric universe – the idea that Earth was at the center of everything. This false idea persisted for 450 years after being disproven. The same has happened with the role of insulin, but hopefully it won’t take over four centuries to come to our senses.

The pancreas is a fascinating organ. It has endocrine, exocrine, autocrine, and paracrine functions. Say what? Here’s what this means in plain English: 
  • Endocrine: hormones are secreted into the bloodstream (ex: insulin, glucagon)
  • Exocrine: substances are secreted into a duct (ex: digestive enzymes like pancreatic amylase and lipase)
  • Paracrine: hormones act on cells near the ones that secreted them
  • Autocrine: hormones feed back to and affect the cell that secreted them (a cell acting upon itself via a hormone or other signaling molecule)

We are not concerned with the exocrine action of the pancreas. (If you’re interested in that, check out my series on digestion from way back in 2013—some of my first blog posts! The post on the pancreas is here.)

The paracrine function is where the gold is. As I mentioned earlier, insulin-secreting β-cells and glucagon-secreting α-cells are in close contact with each other: 

Source: Hormones: The Messengers of Life, by Lawrence Crapo

Here’s a better look at this, via a screenshot from the Unger video:

As you can see, the β-cells and α-cells are interspersed. On the left, as we would expect in type 1 diabetes, there are only α-cells and no β-cells. This is not a small matter. It’s a big deal that these cells are so close together.

Unger’s group did an experiment in which they destroyed (most of) the β-cells of some poor, unsuspecting mice, and here’s what they found:

In this screenshot from the video, you can see that after an oral glucose dose, the glucose levels are very similar in mice with intact β-cells (left) and mice with no β-cells (right). They’re not identical, but certainly if insulin was playing a big role in lowering blood glucose after an oral glucose load, we would expect the mice with no β-cells to have much higher BG, right? But we don’t see this.

And here’s what the insulin levels looked like:

In the mice with intact β-cells (left), we see the expected sharp rise and then decrease over the next couple of hours. In the mice with most β-cells gone (right), insulin barely moves. (Not all the β-cells were destroyed, just enough to make the experiment viable, in case you’re wondering how there was any measurable insulin at all.) That the insulin curves are so dramatically different, but the glucose curves are so similar, tells us that insulin is not in control of the glucose level.

From Dr. Unger: “We had eliminated most of the beta cells, and yet, the glucose tolerances were virtually the same. This suggests once again that glucagon is determining the shape of the glucose tolerance curve by its actions on the liver, and that insulin has very little influence on this curve, even though it may be increasing peripheral uptake of glucose.” […] “These results told us that the function of the juxtaposition of beta and alpha cells was to restrain glucagon secretion, and that insulin was in control of glucagon.”

Glucagon is high in every form of diabetes, whether type 1 or type 2 in humans or experimental animals, and even in the case of total pancreatectomy—that is, when the pancreas is removed. What? Yes, even when you remove the pancreas altogether, glucagon is still high, which led to the discovery of glucagon being produced by other tissues, but that’s just kind of gee-whiz info for our purposes.

As you can see from the screenshot above, hyperglucagonemia (high blood level of glucagon) occurs in all forms of diabetes.

The plot thickens…big time

One of the things that blows my mind the most from the Unger video is the visual on the different concentrations of insulin different organs are exposed to. Endogenous insulin—that is, insulin secreted by intact, functioning β-cells—reaches different tissues at different concentrations. It is almost impossible to mimic this with injected insulin, which likely explains the frequent highs and lows in BG that T1 and T2 diabetics experience. (Particularly when they’re on high-carb diets and are slamming their bodies with industrial-strength doses of insulin. As I mentioned earlier, people with T1D using much smaller doses of insulin have these complications much, much less frequently.)

This was one of the most stunning things for me in the Unger video:

The concentration of insulin that acts upon the pancreas itself—both the β-cells (autocrine function) and the α-cells (paracrine function)—is much higher than the concentration that reaches the liver via the portal circulation, and higher still than concentration that reaches skeletal and cardiac muscles. You cannot reliably replicate this with injections.

“Indeed, in patients with diabetes treated with insulin who have ‘tight glycaemic control’, there can be ‘over-control’ of other metabolic processes. This is explained as follows. Exogenous insulin is given peripherally and the dose adjusted to control hepatic glucose production, not proteolysis. With subcutaneous or intravenous administration of insulin, the concentration in the portal [liver] circulation will always be less than in the systemic circulation. This is a reversal of the normal situation. Good glycaemic control thus invariably results in peripheral hyperinsulinaemia.” (Source)

In plain English, what they’re saying is, if you were to inject the amount of insulin the liver and pancreas require to suppress glucagon release and/or action, it would be way more than the muscles and other tissues are equipped to handle, and you would probably end up with hypoglycemia. According to Dr. Unger: “If you try to suppress glucagon by giving enough insulin to reach the paracrine levels inside the islet, you would overwhelm the tissues outside the islets and cause serious hypoglycemia.” 

And isn’t this indeed what so many type 1 and insulin-dependent type 2 diabetics experience? Massive, industrial doses of insulin are used, and they sometimes suffer massive industrial volatility in blood glucose.

However, it doesn’t have to be this way. Following a low carb/ketogenic diet appears to make this extra layer of complication almost irrelevant. RD Dikeman, who runs the Type One Grit Facebook page and who was co-author on this paper about type 1 diabetics having great success using LCHF/keto to improve BG management, shared this with me in a personal communication, regarding his son Dave’s use of insulin (Dave has T1D):

“The majority of the insulin Dave injects is about glycogen release suppression, not about ‘shoving carbo-glucose into the cells.’ Dave injects basal insulin to keep the liver ‘in check’ and he injects bolus insulin to (mostly) counteract the glucagon release provoked by dietary amino acids at the location of the liver. Injected insulin can work on the liver even if not injected into the portal vein.”

“Exogenous insulin can never be physiological.”

The above was stated in this post by Petro “Peter” Dobromylskyj, a veterinarian who knows more about cellular metabolism and energetics than the majority of doctors who work with humans. (His blog, Hyperlipid, is not for the faint of heart. It’s a gold mine of fascinating biochemistry insights you’ll find nowhere else, but I admit I can understand only about 50% of his posts!)

He addressed this shortcoming of injected insulin compared to natural endogenous insulin quite eloquently. He entire following indented section is an excerpt from his post: Metformin (6) – Insulin-induced insulin resistance is real:

“Recall that the hyperglycaemia in T1DM has little to do with the lack of insulin per se. The hyperglycaemia is caused by an excess of glucagon from the alpha cells of the pancreas. Insulin starts its control of hyperglycaemia by the suppression of pancreatic glucagon secretion, it's a local action within the islets. How high this concentration of insulin is under normal physiological conditions is quite hard to determine but it is likely to be a lot higher than the diluted insulin concentration in the portal vein, heading towards the liver.

The diluted insulin within the portal vein arrives at the liver where its next job is to suppress hepatic glucose output, again in antagonism to glucagon.

Finally, if glucose from the liver continues to enter the systemic circulation, the function of insulin here is to push that glucose in to any cells that will take it. Muscle and adipose tissue being two major targets.

So under normal physiology there is a gradient of insulin concentrations from very high within the Islets of Langerhans, to significantly lower at the hepatocytes, down to much lower in the systemic circulation.

Exogenous insulin produces no such gradient. It drains from its injection site into the systemic veins and is then redistributed, at a single concentration, throughout the body.

This will never effectively suppress alpha cell glucagon secretion and will only do a modestly effective job of suppressing hepatic glucose output. So glucose will be continuously secreted in to the systemic circulation. The dose of detemir used has to be enough to mop up this excess glucose supply, and it can only put it in to cells sensitive to insulin throughout the body. Muscle cells. Adipocytes.

Aside: Except, of course, under deeply ketogenic eating where only minimal insulin is ever secreted, very little is metabolised, the gradients between alpha cells, hepatocytes and adipocytes flattens out and the correct physiology is for glucagon to be elevated with minimal insulin. I've posted this before. T1DM patients have no choice, ketosis is the only physiological state which can be fairly well mimicked using very low doses of exogenous insulin. End aside.

I would never suggest that exogenous insulin has no effect on pancreatic glucagon secretion or elicits no suppression of hepatic glucose output. It will always have some effect, but there will always be an abnormal emphasis of its effect on systemic tissues.”

Pancreatic Fat Buildup, Type 1, and Type 2 Diabetes

We know that type 1 and type 2 diabetes are very different entities. Type 1 diabetes is the result of autoimmune attack on the beta cells. Type 2 diabetes initially involves over-secretion of insulin from perfectly intact beta cells. However, over time, the pancreas of a type 2 diabetic loses its ability to function properly, apparently due to a buildup of fat within the pancreas itself. (Learn more about this in my post on the Personal Fat Threshold concept.) Let’s see how this works:

This is in mice, but we can still see what’s happening: the graph with red boxes is a strain of mice bred to become easily obese. The graph with blue circles is a lean mouse strain. The top boxes are showing the amount of tryglyceride (TG) building up inside the pancreatic islets—the islets where beta & alpha-cells reside. The bottom is showing the actual buildup of fat. You can see things really take a turn for the worse somewhere between 9-10 weeks of mouse lifespan.  

Now, let’s look at the relationship between this buildup of fat inside pancreatic islets and deterioration of blood glucose control:  

Buildup of islet triglyceride is shown in the upper boxes; blood glucose level is shown in the lower boxes. As you can see, when the buildup of fat in the pancreas reaches a certain point—around week 9-10—that’s when BG starts rising. The lean mice, who don’t have the pancreatic fat accumulation, do not develop diabetes. Their BG remains normal, while the poor mice with fat being socked away in their pancreatic islets have BG rising precipitously. This is the mouse version of the personal fat threshold: the buildup of fat in the pancreas appears to precede the elevation of BG. Fat being stored in places it’s no bueno to be stored (such as your blood sugar regulating organs—liver & pancreas) is a primary driver of diabetes.

Okay, that’s all well and good for mice. What about humans? Below is a chart showing pancreatic fat buildup in humans—lean people, obese (but non-diabetic) people, and people with impaired glucose tolerance (IGT). I guess they had no data for people with type 2 diabetes, which is a bummer, but we can probably infer what it would look like. IGT is basically pre-diabetes—where your fasting glucose (or response to an oral glucose tolerance test) is higher than “normal,” but not high enough to be diagnosed with full-on type 2 diabetes. And look at the dramatic difference between the level of pancreatic fat in a non-diabetic obese individual and an individual with IGT. It’s that big buildup of fat in the pancreas—not total body weight or even total body fat—that correlates with the rising blood glucose.

I said earlier that it was noteworthy that the pancreas is constructed with alpha and beta cells in such close contact. (I also said we’d come back to that, and here we finally are, only a few thousand words later.) I also said it’s not negligible that rodent and human islet architecture are slightly different. This might have more profound implications for research than we realize, and might make it even more treacherous than it already is to extrapolate what happens in rodents to what happens in humans. We can still study things in rats and learn about mechanisms, but we should never assume that findings in rodents will translate exactly the same to humans.

I’ve been saying that the way insulin controls blood sugar is not on the uptake side (not by forcing glucose into cells, although it facilitates this in some cells), but on the output side – by suppressing the release of glucose from the liver. And the way it does this is by countering the action of glucagon. Well, if insulin is going to suppress release of glucagon, then the alpha cells have to be insulin sensitive. And it appears that this is not the case in type 2 diabetes:

In T2D, alpha cells don’t respond to insulin the way a non-diabetic’s do. Glucagon is not suppressed as much per unit of insulin.

 In Dr. Unger’s words: “The very first rise in insulin is associated with a profound reduction in glucagon. Insulin keeps on suppressing glucagon and this insulin glucagon ratio that results is over 7.0, which is very high. And this insulin: glucagon ratio tells the liver that it better get busy and suck up all the incoming glucose and store it as glycogen. This is the normal relationship of insulin and glucagon to one another and to their actions on the liver.”

But what happens in T2D?

“You do the same experiment in pancreas isolated from type 2 diabetic animals and the first thing you note is that the dramatic spike in insulin is gone. BUT, insulin levels are higher than normal. Because the insulin spike is gone, it’s not surprising that there is no suppression of glucagon. And then, despite the high levels of insulin, glucagon levels are also high, because they’re insulin resistant. The net result is to give an insulin:glucagon ratio of less than 1.0. What this tells the liver is, DON’T store incoming glucose; continue to manufacture glucose, keep producing glucose, because we’re starving, even though a big meal has been ingested. In other words, the wrong information is being transmitted to the liver, and this is why glucose tolerance is so abnormal and hyperglycemia persists. The liver continues to produce glucose even though it doesn’t need any glucose.”

Dr. Ben Bikman knows the deal. Here’s what he said in a tweet:
“…glucagon usually increases over time with #insulinresistance; the alpha cells become resistant to the inhibitory effects of insulin and make glucagon when they shouldn’t.

By the way, this is one of the main mechanisms of the drug metformin: suppression of hepatic glucose output.

“…paracrine insulin reaches the α cells before insulin reaches any other targets in the body in concentrations far above the endocrine levels delivered to peripheral insulin targets. […] In human islets, there is extensive juxtaposition of β cells and
α cells that should permit insulin to reach α cells across their shared interstitium in a paracrine relationship.” (Source)

The physical proximity of alpha and beta cells is no coincidence: it “facilitates instantaneous insulin control of glucagon secretion via the interstitial space separating the two cells.” (Ibid)

Instantaneous insulin control of glucagon. This is exactly what’s absent in people with type 1 diabetes, and it seems to be impaired in those with type 2. Seems like there are two possibilities in type 2: 1) Beta cell failure or “burnout” – beta cells that were overworked for years have simply called it quits and no longer produce sufficient insulin to counteract glucagon at the level of the alpha cell. 2) The beta cells are still making plenty of insulin, but the alpha cells are resistant to it; that is, they don’t respond to insulin by stopping glucagon secretion, so they continue pumping out glucagon, acting as if the insulin isn’t there.

All I know is, just about every insulin-dependent type 2 diabetic I’m aware who adopts and adheres to a low carb or ketogenic diet is able to reduce their doses and, in many cases, stop insulin treatment entirely. So I honestly don’t like the phrase “beta cell burnout,” because I don’t think it’s a real thing. I mean, maybe it is, but this can be easily tested with a C-peptide test. (C-peptide is a leftover fragment from endogenous insulin – insulin the body makes, itself. It does not come from injected insulin, so a C-peptide test is a way to measure whether a type 1 or type 2 diabetic is still producing their own insulin.) If beta cell burnout is a thing, maybe it’s only temporary, because the fact that so many T2 diabetics are able to stop their insulin entirely and maintain normal blood glucose on low carb proves that those cells were not dead, but just needed to rest a while before they could get back to work.

I’m willing to acknowledge that it’s not always so straightforward, though. For some people, just going keto isn’t enough. They might have to add in some fasting, exercise, and yes, maybe even some insulin-sensitizing medications. Maybe there is some degree of “burnout” that can’t be overcome. Maybe some beta cells do actually just straight-up die in type 2. I dunno. Either way, a low carb diet is still a dynamite place to start on the way to recovery.


Having written almost 8000 words in this post, it’s a bit embarrassing to admit here at the end that I’m not quite sure what to do with all of it. What are the implications? What does it all mean?  Honestly, maybe nothing.

I like writing about things that matter – things that have an impact on what we eat or how we live. But once in a while, sometimes it’s just information. Stuff that feels enriching and is worth knowing even if it doesn’t make much difference in the way any of us would implement our own preferred version of low carb/keto.

And I think that’s the case here. None of what I’ve learned in the past few months or that I’ve written about here is paradigm-shifting in terms of what I eat, how I exercise, the supplements I take, etc. I do think it reinforces that low carb is best for me, as if I wasn’t already 100% confident about that. That’s the interesting thing, isn’t it? Despite the incredible complexity of pancreatic and hepatic physiology, and the weird things that happen in disease states, low carb/keto seems to make most of this irrelevant. So we can sit around and debate and argue endlessly over the minutia, or we can eat a very low carb diet and take most of blood glucose and insulin “management” out of the equation. Take it out of the equation.

And that being said, I still stand by what I’ve always said—not everyone needs keto. Some people can eat a surprising amount of carbs and still have low glucose and insulin and remain metabolically healthy. For those people, great! Awesome. Eat all the carbs they want. For the rest of us, not so much.

And remember, people with type 1 diabetes will always need at least some insulin. They won’t be able to completely discontinue their injections. But by going low carb/keto, they’ll be able to significantly reduce the amount they require, which can save them a big chunk of money, but even more important, can have a dramatic impact on quality of life in both the short term (far fewer hypos/hypers) and the long term (lower risk for diabetic complications, many of which actually come from the high insulin doses, not just from the high blood sugars). (Interested in keto for T1D? Check out this article I wrote for my day gig.)

If you’re interested in really digging into the details on this new perspective on insulin action, as well as insulin’s relationship with glucagon and the relevance of this to both type 1 and type 2 diabetes, the following papers are must-reads and the video is a must-watch:


Want to support more blog posts like this? (Or shorter ones? Yes, I know ones like this are ridiculous.) All the time I spend reading relevant papers, watching videos, and brushing up on biochem & physiology is unpaid. I’d love to get back to writing blog posts more regularly. If you’d like to help make that happen, consider making a monthly contribution via Patreon. Any and all amounts are welcome and appreciated; no amount of support is too small! Check out my Patreon page here:


Disclaimer: Amy Berger, MS, CNS, NTP, is not a physician and Tuit Nutrition, LLC, is not a medical practice. The information contained on this site is not intended to diagnose, treat, cure, or prevent any medical condition and is not to be used as a substitute for the care and guidance of a physician. Links in this post and all others may direct you to, where I will receive a small amount of the purchase price of any items you buy through my affiliate links.


  1. I scrolled to the end. Read End and thought, "there is no way I want to read 8000 words on insulin." But started on it anyway and could NOT PULL MYSELF AWAY. This was riveting. Thanks. And super glad you are on Patreon.

  2. Okay… I have to echo Robin's comments. I love the way you write and one can never tell when one "geeks out" on material where you will find a nugget of understanding or at least avoid misunderstanding something important. Clearly I'm going to have to come back to this blog post and spend a little time "studying" and going back to look at some of the other references that you so kindly pointed my nose too. Just when I thought I could go read some lazy novel… Something interesting comes my way. I don't know if you have thought about another book… But perhaps a translation book between dieting and physiology… And possible practical implications (there is always that strange issue of fasting and autophagy and its related cell death which I can't spell). You might take on the persona of Isaac Asimov who wrote some "popular" science books in addition to his science fiction… Something along the lines of "The Human Body: Its Structure and
    Operation" only more focused and related to low-carb and diet in general. In any event thank you again for the delightful blog post and I will get back to it and watch the videos and read the papers and put my novel aside for a little while. I never thought that when I wanted to figure out how to lose weight I would get sucked up into human physiology as well.

  3. I'm glad you're a voice of reason in the low carb/keto community. I bought a continuous glucose monitor to test what happened to my blood sugar when eating protein. I was convinced I'd find a massive blood sugar rise when eating 80+, 120+, 160+ grams of protein in a single meal. What did I find? Nothing. My blood sugar seems unaffected by the proteins I was eating (from real food, not isolated in cans). Moreover, I felt better. So, now, I eat a higher protein and lower fat keto diet, and my HbA1c has continued to decrease.

    There are still locations on the 'Net where keto people are terrified of protein. The 2 Keto Dudes forum is one of them.

    1. Thanks for writing. Yes, far too much fearmongering about protein in the keto world, although I like a lot of Carl & Richard's work, so I'll keep silent on that particular issue. I think they both understand that different things work for different people and they generally don't give blanket recommendations for "everyone."

  4. Great, thank you Amy!

    Would you be interested to tackle nutrition in pancreatitis in T2Ds? The big three (Chris, Robb, and Mark) do not have specific posts about this condition, only forums or comments. General advice is low-fat, high fiber. With quick googling, only Dr. Dahlman recommends keto.

    1. I know approximately zero about pancreatitis, so I can't write about it right now, but if I had to take a wild guess, it's probably the same story with kidney function -- higher protein is probably fine, because I would imagine it's sugar (glucose/fructose) that does the most damage, not amino acids. But I can't write on it at the moment. Maybe sometime down the line.

  5. It makes logical sense to me that placing less "pressure" on the pancreas would be better for pancreatitis. Thus, I would think keto could would well. Furthermore, if the pancreatitis is caused at all by extra fat, intermittent or longer term fasting should help. See, eg:

    If you're specifically looking at T2 diabetics, and they're taking drugs including insulin, you have to do this under supervision, as drug requirements can plummet.

  6. Excellent post! Content like this is sorely needed. I need to pour over the references, they look fascinating.

    Question: At one point there's discussion about fat in the pancreas causing issues between the alpha and beta cells. Then later there's a bit about either the beta cells burning out and not putting out enough local insulin or the alpha cells being resistant to the insulin.

    Any thoughts on which it is? Or a combination of both? Or have I misread these parts?


    1. Thanks for reading. I don't know which it is. Maybe both. But I have a hard time believing that most people with T2D have beta cell "burnout." If you measure their insulin, it's usually still sky high, so they are still pumping out of plenty of insulin. There might be a very end-stage kind of thing where the cells actually atrophy or die and are not able to pump out inordinate amounts of insulin, but I think to actually test this is very rare. It would be C-peptide, and it's hardly ever ordered in T2 diabetics. (They use it with T1D to see if someone is still producing any of their own insulin.) I'm honestly not sure of the answers here, though. It could be both -- beta cells not secreting enough insulin, *and* localized "resistance" (hate that word) at the level of the alpha cell. *Shrug.*

  7. RE: " _World Turned Upside Down" the author states that calling sugar addiction trivializes real addictions like alcohol, and other hard drugs — kidney failure, blindness, amputations et al. are _really serious_. Dr. Lustig says the long term metabolic consequences of sugar and alcohol are the same.

    1. Just because I link to something doesn't mean I agree with everything the author has ever said or written. I've met Dr. Feinman a few times in person and he's a good guy. But I disagree about sugar addiction.l It can absolutely be devastating in many ways, and not just to health. I would definitely classify sugar as a "real" addiction. Hard to comprehend if one has never experienced it, just like compulsive overeating, binge eating, and *any* food addiction, for that matter.

  8. This post was amazing!! Keep up the long form (as you are inspired to do so!). I love all the data and primary source citations embedded. As a research scientist I appreciate all the time to find, read, write about, and cite this wealth of knowledge! You are a rock star! Thank you for caring enough to make the effort to publish these extensive posts!! Heading over to Patreon now! :D

  9. It is so refreshing to see someone write about how the more they learn the less they know. This is how I feel most of the time. How lucky are we you are still learning and sharing you information? Thank you for that. You are providing very helpful information, especially for those of us who are trying to regain our health and have a long, healthy life.