Understanding glucagon (and somatostatin 28)


By Grant Schofield and George Henderson

Say what? OK, we’re always on about insulin the so called “master hormone” that manages our blood sugar and fat burning . But hormones almost never act alone. There are more complex bits and pieces going on. Let’s look at two others involved the process – glucagon and somatostatin 28.

Warning: This is reasonably complex physiology, not general banter about poor nutrition guidelines!

What is Glucagon?

Glucagon is the hormone that normally dominates metabolism when you’re fasting, or when you eat fat and protein. It tells the body to make energy from fat and protein, and to supply glucose from amino acids and glycerol (part of the triglyceride [fat] molecule) to keep up the steady supply of blood glucose.

For people without insulin resistance or diabetes glucagon is high when insulin is low and vice versa.

Glucagon tells the liver to release stored glucose (glycogen) to the bloodstream. If you’re fasting or undereating, glucagon also triggers the breakdown of stored fat and protein. It’s important that this energy be released carefully, and insulin serves as a counterbalance in this phase. It only takes a little insulin to counterbalance glucagon in someone eating a very low carb, high fat diet, and this is why blood glucose and lipids can normalise quickly when someone with diabetes eats this way, and insulin requirements are usually reduced.

Like insulin, glucagon is produced by the pancreas (by alpha cells for glucagon, beta cells for insulin). But there are also alpha cells in the gut. Carbs (glucose) in the stomach stimulates the release of glucagon from the gut alpha cells.

Why does eating glucose stimulate the release of glucagon? No-one really knows – it seems counter-intuitive – it is certainly counter-productive for people with diabetes. However, in people with normal insulin metabolism, it isn’t maladaptive at all – glucagon and glucose both stimulate the pancreas to produce insulin very promptly, so everything runs as it should.

Eating too much carbohydrate

If you are someone who has a problem processing carbs well – if you are insulin resistant/Type 2 diabetic or have Type 1 diabetes – then eating more carbs than you can handle has consequences.

First, glucagon from the alpha-cells or the gut rises excessively. One of two things can then happen – either insulin is forced higher than is desirable, in an insulin-resistant person, which is part of the vicious cycle of type 2 diabetes and obesity. I guess this is what we call “metabolic dysregulation”. Or, in an insulin-deficient person, this rise in glucagon triggers an excessive release of amino acids and fatty acids. When this happens, the liver makes too much glucose, and this excessive production of glucose can also lead to the over-production of ketones from fatty acids. (The excessive production of glucose has the same metabolic effect in liver cells as a deficiency of glucose, resulting in the incomplete oxidation of fatty acids and formation of ketones).

This stimulation of glucagon by dietary carbohydrate could thus trigger diabetic ketoacidosis – a very high, very toxic level of both glucose and ketones – if it wasn’t controlled by insulin. In the days before insulin was available, the diet that was shown to have the lowest risk of diabetic ketoacidosis was very low in carbohydrate, low in protein (so that there was as little material as possible to make glucose from), and high in fat (which tended to spare protein from being broken down).[1] This was not only a low-glucose diet, it was also, very importantly, a low gluconeogenic (glucose-forming) diet. Nowadays, fortunately, most people deficient in insulin have access to effective medical forms of this hormone, but you can see how a low-carbohydrate diet still makes the control of diabetes easier than it would otherwise be.

Glucagon helps us understand why insulin resistance, and diabetes, are an intolerance to carbohydrate

Normally the peak in glucagon production when you eat carbs doesn’t last long. But in someone with insulin deficiency or insulin resistance this means that glucagon will maintain gluconeogenesis and lipolysis exactly when it’s least required.

For example, in this study, glucagon and gluconeogenesis are not decreased when people with type 2 diabetes are fed a low-fat, high-carbohydrate diet, and glycogenolysis (the release of glucose from glycogen) is actually increased after a high-carb meal – which is the opposite of what should happen.[2]

Now we have new research showing that even in people with no pancreas there’s a significant release of glucagon from the gut in response to glucose.[3] This strengthens the earlier finding. Diabetes is an intolerance to carbohydrate.

Roger Unger has researched the role of glucagon in diabetes since the 1950’s and has shown that a number of compounds that inhibit glucagon (including leptin and somatostatin, as well as specifically engineered antibodies), can restore normal glucose tolerance in animals without insulin-producing beta-cells.[4]

What is Somatostatin 28?

Eating fat stimulates the release of somatostatin 28 from the gut (initially, from delta cells in the pyloric antrum, at the bottom of the stomach).[5] In normal metabolism somatostatin 28 inhibits the insulin and glucagon response to glucose and protein, but it has no effect on basal insulin.[6] Importantly, somatostatin 28 inhibits lipolysis – the release of fatty acids from fat storage cells.[7, 8] Somatostatin 28 also has a protein-sparing effect.[9, 10] This means fewer amino acids are released to be converted to glucose in the liver.

Insulin tells the liver to stop releasing glucose when glucose begins to enter the bloodstream from the gut; somatostatin 28 seems to play a similar role in balancing dietary and stored fat.

This could provide the answer to an old paradox. There’s no insulin response to pure dietary fat to speak of, so if ketones are made from fats, how could a high-fat diet (one that we would call a ketogenic diet today) prevent ketoacidosis?

Somatostatin 28 – primarily a regulator of fat homeostasis

When you eat lots of fat, high amounts become available in the bloodstream, and your body needs to cut down the release of fats (as free fatty acids) from fat cells to make room for it. Otherwise you’ll be flooded with energy and with ketones. (Obviously you can still lose weight if you need to by eating fat – the control of lipolysis is fairly flexible – but not if you eat loads of it past the point where all the mechanisms which regulate hunger are over-ridden.)

The action of somatostatin 28 – inhibiting glucagon, inhibiting lipolysis, and sparing protein – is a useful backup to the similar effects of basal insulin. Basal insulin – the low background insulin level – supports a feedback loop by which ketones themselves inhibit lipolysis.[11] Together, this helps to explain why high fat diets (very low in carbohydrate, essentially ketogenic diets) were as effective, and somewhat safer, than fasting for reducing blood sugar and preventing diabetic ketoacidosis in the pre-insulin days, in people with low, but not zero, insulin production.[1]

No hormone acts alone, and the regulatory mechanisms described above have been simplified. However, for people interested in high fat diets, somatostatin 28 is a potent hormone that specifically responds to dietary fat, and has actions which are desirable in a low-insulin context.

A note on dietary guidelines and pure research

Note that the research into somatostatin 28 that we’ve cited is somewhat limited – the evidence on lipolysis, for example, comes from two studies in chicken adipocytes. The most recent of these, published in a poultry journal, is twenty years old. We could also find only a few studies into the effects of pure fat feeding on insulin levels (in these there is little or no insulin response to dietary fat). Experimentation into the protein-sparing effect of somatostatin 28 seems to have trailed off when it became obvious that it wouldn’t be useful as a drug.

It seems to have been the case that, because high fat diets were thought to have no benefits, and diabetics and others have been instructed to avoid them until recently, research into some of their mechanisms has been very limited. Somatostatin 28 researchers may not have realised that their work imitated the effects of a ketogenic diet, and ketogenic diet researchers don’t seem to have measured somatostatin 28 in their studies.

Pure research isn’t always that pure – preconceptions about what will be important (in this case, managing high-carbohydrate diets, and developing drugs) can sometimes create a blind spot where research is neglected, and this seems to have happened with the somatostatin 28 response to fat-feeding.


[1]  Newburgh LH, Marsh PL. Further observations on the use of a high fat diet in the treatment of diabetes mellitus. Arch Intern Med. 1923;31(4):455-490.

[2]  Allick G, Bisschop PH, Ackermans MT et al. A low-carbohydrate/high-fat diet improves glucoregulation in type 2 diabetes mellitus by reducing postabsorptive glycogenolysis.  J Clin Endocrinol Metab. 2004 Dec;89(12):6193-7.

[3]  Lund A, Bagger JI, Wewer Albrechtsen NJ et al. Evidence of Extrapancreatic Glucagon Secretion in Man. Diabetes. 2015 Dec 15. pii: db151541. [Epub ahead of print]

[4]  Raskin P, Unger RH. Hyperglucagonemia and Its Suppression — Importance in the Metabolic Control of Diabetes. N Engl J Med 1978; 299:433-436.

[5]  Ensinck JW, Vogel RE, Laschansky EC, Francis BH. Effect of Ingested Carbohydrate, Fat, and Protein on the Release of Somatostatin-28
in Humans. Gastroenterology. 1990;98:633-638.

[6]  Ensinck JW, Vogel RE, Laschansky EC et al. Endogenous Somatostatin-28 Modulates Postprandial Insulin Secretion. Immunoneutralization Studies in Baboons. J Clin Invest. 1997;100(9):2295–2302.

[7]  Strosser M-T, Di Sclala-Guenot D, Kock B, Mialhe P. Inhibitory effect and mode of action of somatostatin on lipolysis in chicken adipocytes. Biochimica et Biophysica Acta, 1983;763:191-196.

[8]  Oscar TP. Prolonged in vitro exposure of broiler adipocytes to somatostatin enhances lipolysis and induces desensitization of antilipolysis. Poultry Science 1996;75:393-401.

[9]  Shaw JHF, Wolfe RR. Metabolic Intervention in Surgical Patients: An Assessment of the Effect of Somatostatin, Ranitidine, Naloxone, Diclophenac, Dipyridamole, or Salbutamol Infusion on Energy and Protein Kinetics in Surgical Patients Using Stable and Radioisotopes. Ann Surg. 1988;207(3):274-82.

[10]  Heindorff H, Billesb0lle P, Ligard Pedersen S, Hansen R, Vilstrup H. Somatostatin prevents the postoperative increases in plasma amino acid clearance and urea synthesis after elective cholecystectomy. Gut. 1995;36:766-770.

[11]  Coppack SW, Jensen MD, Miles JM. In vivo regulation of lipolysis in humans. J Lipid Res. 1994;35:177-193.



  1. Brilliant info Thanx

  2. The fact that a very low carb, moderate protein, high fat diet is still not enough to suppress lipolysis and subsequently reduce gluconeogenesis to normal levels in type one diabetics shows that insulin is indeed the master hormone of nutrient regulation. Also, it takes several weeks or even months for gut alpha cells to proliferate after a total pancreatectomy and even then these patients are just mildly diabetic. (In contrast to dogs that develop full diabetes and to mice that develop no diabetes at all.) I believe their contribution to physiological glucagon levels in normal people is minuscule. I believe the physiological role for basically all glucagon-like peptides (incretins and glucagon) released from the gut is to stimulate insulin secretion and prepare the body for the fed state by other means.
    Last year three different papers were published on the importance of inhibiting lipolysis in maintaining normal blood glucose in a diabetic context. If insulin can successfully exert this effect on white adipose tissue, hyperglycemia does not develop. All this by limiting substrate to pyruvate carboxylase, the first enzymatic step in gluconeogenesis. I believe this discovery is of huge importance and potentially can also give an explanation to why there are different phenotypes in the course of the same disease. It’s an interaction mainly between insulin resistance developing in the liver and in white adipose tissue. Order and strength determine metabolic health, obesity and when WAT becomes very IR, diabetes.
    This discovery also proves Unger wrong about the need for 2-3 different insulin doses. It is enough to inhibit lipolysis in WAT, i.e. peripherally and there are no diabetic symptoms. In real life that’s exactly what happens when somebody goes on a low carb, high fat diet: WAT insulin sensitivity is restored and insulin levels are greatly reduced. Regardless of residual insulin resistance in the liver and maybe in alpha cells, both glycemia and circulating FFA levels normalize.

  3. Good argument erdoke.

    “to stimulate insulin secretion and prepare the body for the fed state by other means” – by other means includes the role of SST28 described here.
    Control of lipolysis (rather than proteolysis) may do little to decrease pyruvate as glycerol GNG only contributes about 6% of fasting glucose; glycerol 2-carbon seems to be mainly conserved in TCA cycle to maintain fatty acid oxidation, not extracted as oxaloacetate.
    However acetoacetate is also a source of pyruvate via acetoacetate decarboxylase > acetone > methylglyoxal > pyruvate > oxaloacetate

    “On the basis of our specific activity data, we have calculated that 4-11% of plasma glucose production could theoretically be derived from acetone”.

    The 11% was calculated for 21 day starved humans.


    Another interesting point about the fed state is that insulin, or an increase in insulin, is not required to store dietary fatty acids in adipocytes, as DGAT1 merely requires basal glucose level (5mM).

    This explains why Inuit don’t waste away with zero carb diets. Our point is not that insulin isn’t required for metabolic control – I doubt Unger’s mice did well over the long term – but that there are these insulin-independent countercurrents that we’re also riding when we eat LCHF diets – and that may well be useful for those with poor insulin control.

    1. I was referring to this (and a few similar) recent study:
      It seems that, although insulin has direct actions in the liver as well, it is insulin’s action on adipose tissue which is critical to maintain glucose homeostasis. In other words, insulin resistance in adipose tissue is central in metabolic diseases. I go further and claim that most modern metabolic diseases start with oxidative stress and subsequent insulin resistance in subcutaneous adipose tissue. All other phenomena can be explained as parts of the subsequent cascade, including visceral (also a kind of ectopic) adipose accumulation, NFALD, T2DM, (most) CVD, etc.

  4. Hi Erdoke,

    That’s a good theory. If acquired metabolic disease is the inability to cope with excess energy from food, then the biggest energy deposit in the body is also likely too be involved.
    If the acetyl-CoA from fat is flooding mitochondria, then acetyl-CoA from glucose and aminos is likely to be kept out (thus turn to lipids) and pyruvate preferred (thus more substrate for glucose).

    The immediate effect of a glucagon injection is glycogenolysis. Diets with zero carb completely inhibit post-prandial glycogenolysis in subjects with and without type 2 diabetes. This is prima facie evidence for a significant effect of carbohydrate-stimulated glucagon release.

    If you drink a cup of olive oil, there will be zero insulin response. Certainly, there will not be an insulin response even remotely commensurate with the intake of energy. Yet metabolism is not overwhelmed, even in diabetic subjects. This changes when fat and carbohydrate are consumed together. But there are obviously mechanisms independent of insulin for preparing the body for the fed state when fat is the only factor, and it seems likely that these remain important when protein is kept low and carbohydrate is minimal.


    1. First to your second thought about accommodating fuel sources. Yes, it seems that when carbohydrates and fats are consumed separately, the body is perfectly prepared for them. Even when slowly acting carbs are consumed with fats there is no big stress provoked. The real problem is when different stressors come in regular waves and converge upon mitochondrial stress in adipocytes. In rodents it takes only 5-6 days to induce chronic inflammatory responses in subcutaneous by a high fat, high sugar diet. Acute immunomodulation results in stronger responses, e.g. in beta cells, but the problem, as always, arises when these signaling pathways become chronic.
      Regarding “energy deposit”, I must say that it is still a very common mistake to underestimate the role adipose tissue plays in energy homeostasis. After deeply digging into this, for me it’s clear that WAT is the main regulator of metabolism together with the immune system. It makes a lot of sense evolutionarily, because adipose tissue preceded huge complex brains. So what basically happened was that the brain was built upon the existing system and tried to chip in and exert as strong control as it was possible to secure. However, the brain still relies on old signaling systems, and sitting rather separated this translates into rather limited options.
      By now I strongly believe that all modern metabolic diseases start with chronic adipocyte (oxidative) stress and that insulin resistance is an antioxidant defense system in adipose tissue. Just as it is downstreams in other tissues/organs. It is very important to understand that insulin’s single most important action in the body is suppression of lipolysis. It doesn’t really matter what glucagon does in the liver until insulin can suppress fat flow from adipose tissues. Normoglycaemia is maintained. That’s why very low carber type 1 diabetics can achieve excellent glycemia with way lower insulin doses than what is required in the liver, not to mention alpha cells. And that’s where Unger fails with the glucagon’s centric approach, although he he had not been aware of the importance of lipolysis until early last year (see study below). Until you can adequately suppress lipolysis, everything runs just fine.
      I keep linking to this landmark study from Perry et al. in different Facebook groups: http://www.sciencedirect.com/science/article/pii/S0092867415000148
      It is a must read.
      Another one on the roots of metabolic stress:

  5. Recent confirmation of above study pointing to impaired suppression of lipolysis (i.e. adipose IR) as the cause of diabetic hyperglycemia:
    A landmark study published in Nature 2 months ago finally deciphers how HFD causes obesity in rodent models. It’s mediated by gut flora changes and overproduction of fermented acetate. A parasympathetic feedback from the brain stimulates insulin hypersecretion. Hyperinsulinemia therefore – at least in rodents – precedes insulin resistance. It’s a kind of supporting evidence in favor of the insulin-obesity hypothesis.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Richard David Feinman

Richard Feinman, the Other

The Science of Human Potential

Understanding how to be the best you can be. Professor Grant Schofield.


A topnotch WordPress.com site


LCHF, Diet & Health


Eat real food. Enjoy real health.

%d bloggers like this: