April 11, 2012

Nutritional biochemistry

Gravity and insulin: the dynamic duo

Read Time 9 minutes

Last week I wrote about how carbohydrates are effectively a performance-enhancing substance, at least for certain performances in certain people.  I received many great questions, including some challenging this suggestion, which really pleases me because it demonstrates folks are thinking about tradeoffs and questioning everything.

Is carbohydrate reduction or outright restriction “right” for everyone?  I doubt it.  Besides oxygen and water the list of “universal truths” for human health is pretty short.  Parenthetically, one can still overdose on both oxygen and water – in other words, even these two completely essential compounds can be toxic outside of their ideal ranges.

If you’ve been following this blog and/or this general discussion, you’ve probably asked yourself the question I’m about to pose.  If you haven’t asked it yourself, you’ve likely been asked by someone as you’ve had the discussion with friends or family.  Here’s the question:

If insulin is so important in regulating fat metabolism, why do some people eat whatever they want and not get fat? Conversely, why do some people following the strictest carbohydrate-reduced diet remain fat?

This is a paramount question, and I’m sorry it’s taken me four months to finally get to it.  Before we do get to it let me digress (seemingly) to discuss gravity.  If you know everything about gravitational forces, feel free to skip this section.


What is gravity?


For the purpose of simplicity I will limit this discussion to Newton’s law of universal gravitation, as it applies to virtually any “gravity scenario” we encounter in our lives, and it’s the law everyone thinks of when they hear the word “gravity.”  Newton’s law of universal gravitation states that every mass in the universe attracts every other mass with a force that is directly proportional to the product of their masses and inversely proportional to the square of the distance between them.

For those, like me, who love equations, here it is:



In this equation, m1 and m2 are the masses of the two spheres, and r is the distance between them.  G is a really, really, really tiny number called the gravitational constant which, as its name suggests, is a constant.

In other words, objects in the universe have attractive forces between them. These attractive forces depend on two things: the masses of the objects (the more the mass, the more the force, since mass is in the top of the equation) and the distance between them (the more the distance between them, the less the force, since distance – squared – is in the bottom of the equation).  No other amount of detail is really necessary to understand the point I’m going to make.

For most objects on earth (excluding the earth), this force of gravitation is actually not particularly dominant.  When I’m standing next to someone, neither of us can feel this force for two reasons: 1) our masses are not high enough, and 2) the distance between us is too great to overcome the really, really, really tiny gravitational constant.  For example, two 80 kg people standing even 1 cm apart only exert 0.004 Newtons on each other, which is completely unnoticeable.  [A Newton is the unit we use to measure force.]

It turns out for most objects on earth that we interact with, gravitational force is irrelevant (actually, it’s what we’d call a “higher order term” if you remember my discussion of ordered terms, or relative importance from this post).  But there is one enormous exception: the earth!  Because the earth weighs so much (about 6 trillion trillion kilograms; no, that’s not a typo – 6 followed by 24 zeros) for pretty much any object on earth ranging from a feather to a 747 to a skyscraper, the force the earth exerts on us (and us back on her) is equal to our own mass multiplied by 9.8 meters per second per second (this number is the acceleration we experience due to gravity).  You can verify this for yourself using the equation above (plug in your mass, the mass of the earth, and the distance from you to the center of the earth).

This is a long way of getting to one of the most often experienced applications of Newton’s Second Law: the net force that acts on a body is equal to the mass of that body multiplied by the change in the mass’s velocity (referred to as acceleration).   When you drop an object, the earth is pulling it towards itself with a force equal to the mass of object multiplied by 9.8 meters per second per second (this unit of m/s/s is the unit for acceleration – the rate of change of velocity).

Here’s a more visceral example: When you are in an airplane taking off on the runway you feel an enormous force on your back, though you’re traveling much slower than when in full flight, when you feel nothing.  Why? In full flight acceleration is zero – velocity, albeit high, is constant – so you feel no force.  On the runway, the plane is accelerating (i.e., changing velocity), so you feeeeel it.

The Apollo 11 mission, the first manned lunar mission, launched from the Kennedy Space Center, Florida via the Saturn V launch vehicle on July 16, 1969 and safely returned to Earth on July 24, 1969. Leaving a wide plume of flame as it climbed into the thin atmosphere of high altitude, the 363 foot tall, 6,400,000 pound Saturn V rocket hurled the spacecraft into Earth parking orbit and then placed it on the trajectory to the moon. Nasa.gov

Why did I just explain all of this?

We live in a world that is governed by physical laws, of which Newton’s laws are but an important subset.  Do you need to know them?  Nope.  Does understanding them change them?  Nope.  Can you “will” your way out of them? Nope.  Is it better to know them or be completely oblivious of them?  I guess it depends on your personality.  I prefer to know as much as possible about the physical world I live in.  It keeps me safe.

Some of you know everything about gravity that I just explained, and then some.  Some of you just heard it for the first time.  Regardless, the laws of gravitation are universally true and apply to every single one of us, whether we like or are aware of it or not.  Let’s consider a few examples.

Look at this fellow:



Photo by Karina Carvalho on Unsplash

Or this fellow:


Photo by Sergio Rao on Unsplash

Are they “defying gravity?”

Are they experiencing a “different” form of gravity than, say, this fellow?


By http://www.flickr.com/photos/theeerin/4112368718/in/photostream/ [CC BY-SA 2.0], via Wikimedia Commons
Or this fellow?


Photo by Robert and Pat Rodgers is licensed under CC by 2.0

Not at all.  In fact, each of these four men is actually subject to the exact same universal truth of gravity.  So, why do the first two appear to challenge the law of gravity while the second two appear to exaggerate the law of gravity?

Lots of reasons, probably, but the two most obvious are differences in genetics and differences in conditioning.  What about the guy belly flopping into the pool? Not quite the same genes and nowhere near the same level of conditioning as someone who seemingly defies gravity.  As a result, the impact of gravity is more obvious on them.

Here’s another example:  Consider a ledge 10 feet off the ground.  Let’s line these three gentlemen up and ask each of them to jump off the exact same ledge.


Different types

The same law of gravity applies to each of them. No exceptions. Will they all experience the law of gravity in the same way?  Of course, not.  First guy – no problem.  Second guy – probably breaks his leg.  Third guy – probably dies.

Let’s add one more layer of complexity to this.  When I was 16 years old, I could jump off a 10 foot ledge with zero difficulty.  Today, at 39, it would take a bit of practice and a lot of concentration to avoid spraining my ankle and tearing a ligament in my knee.  When I’m 65, I’ll be lucky not to break my leg.  When I’m 85, I’ll be lucky not to break my neck.  Why?  Same gravity, right?  Same genes, right? Yet over time, I will experience gravity differently.  Changes in my body over time lead to a differential experience of the same physical laws.


How does insulin fit into this discussion?

Insulin is the most important hormone in our body when it comes to fat mobilization (breakdown) and fat storage.  This is a fact.  There is not one person who studies the endocrine system who will not acknowledge the following quote from Lehninger’s Principles of Biochemistry (the “bible” of biochemistry).

“High blood glucose elicits the release of insulin, which speeds the uptake of glucose by tissues and favors the storage of fuels as glycogen and triglycerides, while inhibiting fatty acid mobilization in adipose tissue.”

In other words, eating glucose (carbohydrates) increases insulin levels in our body.  Insulin drives glucose into liver and muscle cells as glycogen (in small, finite amounts) and into fat cells as triglycerides (in unlimited amounts). Insulin also inhibits the breakdown and utilization of fat, as shown here.

Insulin does not act alone, and the story of fat storage and breakdown is complex if you want to understand every single detail, but the “first order term” is insulin.  I will spend time in the future writing about insulin’s “dance partner,” leptin.  But insulin is probably the General when it comes to determining how the body partitions fat.

So, insulin is sort of like gravity.  It’s in your body whether you know about it or not.  It’s acting on your cells whether you like it or not.  It’s playing a major role in determining your ability to mobilize versus store fat if you believe me or not.

Does this mean insulin has the same effect on everyone?  Does this mean insulin has the same effect on any given person over time? Of course not. Contrast me and my wife.  I look at carbohydrates and start to store fat. If you want a reminder of what I looked like on an “athlete’s diet” of complex carbs and little saturated fat, coupled with 3 to 4 hours of exercise a day, look here, here, and here.  On the other hand, my wife can eat a bag of Oreo cookies for dinner every night, coupled with all the pasta, bread, and rice the world has to offer and not put on one pound (she has weighed about 110 pounds her entire adult life).  How is this possible?  Does this mean insulin doesn’t control fat metabolism?  No, it means we have an entirely different genetic make-up.  Her grandmother is 86 years old, eats bread all day long, is healthy as a horse, and weighs 100 pounds. Conversely, I come from a family where every single man has died of heart disease and looked like the Pillsbury Dough Boy prior to doing so.  I’m genetically programmed to lean towards metabolic syndrome, but I’ve been able to reverse it through strict attention to my eating habits.

This isn’t unique to me and my wife. There is an entire spectrum – a distribution across the population – of people with varying degrees of susceptibilities to the effects of carbohydrate on insulin levels and the commensurate effects of insulin levels on fat storage and breakdown.

And like gravity, the effect of insulin on our metabolism of fat changes over time at the individual level, usually for the worse.

Consider, again, my example: When I was 16 years old I weighed 160 pounds, had between 4 and 5% body fat, a 28-inch waist and a 44-inch chest.  Breakfast consisted of a box (not a bowl) of cereal.  Lunch consisted of 7 turkey and tuna sandwiches (yes, 14 pieces of bread), a gallon of apple juice, and a plate of fries and gravy.  Dinner was a pound of pasta and half a chicken.  Despite eating over 1,000 gm of carbohydrate per day, I was quite resistant to them (i.e., I was very insulin sensitive) and remained exceptionally lean.

Fast forward to three years ago, at the age of 36, I weighed 200 pounds, had 25% body fat, a 36-inch waist and a 44-inch chest.  What changed over 20 years?  I was actually eating considerably less – in both absolute amounts and total carbohydrates – and yes, I was exercising a bit less (3 to 4 hours per day at 36 versus 6 hours per day when I was 16). But, is that the only thing that changed?  What changed in me, and what changes in most people over the same period of their lives, is that I became progressively more insulin resistant.  Most people casually observe that their “metabolism slows down” as they age, but what really happens?

I wish I could definitively tell you why this happens. I can’t. What I can tell you is how it happens. There are many factors, and they certainly vary by person and by individual significance.  The list below is a bit simplified and by no means complete.

  1. Over time, endogenous production of sex hormones (e.g., testosterone, estrogen) becomes altered, and this seems to play a role in fat metabolism.  In addition to sex hormones, other non-sex steroid hormones (e.g., cortisol), which have a strong effect on fat metabolism, may be altered for a variety of reasons including sleep reduction and stress.
  2. Perhaps (at least partially) related to this, the cellular distribution and density of lipoprotein lipase (LPL) also changes.  [Recall, LPL is a very important enzyme on the surface of muscle cells and fat cells.  On muscle cells, it fosters fat oxidation (good). On fat cells, it fosters fat accumulation (bad).]  As we age, we tend to have less LPL on muscle cells and more LPL on fat cells, both of which contribution to fat accumulation rather than fat oxidation.
  3. The membranes of our cells tend to change in fatty acid composition, which may result in more difficulty getting the GLUT-4 transporter into cell membranes to foster glucose flux into cells.  The more difficult it is to get glucose into cells, the more insulin the pancreas must secrete to exert its eventual effect, the more exposure all cells have to circulating insulin levels.  Higher levels of insulin also exacerbate the phenomenon of more LPL on fat cells and reduced fat oxidation.

These changes are all linked, and probably play a different role of importance in different people at different times in our lives.


So what do gravity and insulin have in common?

The forces of gravity and effects of insulin are natural phenomena.  Sure, the comparison is not perfect, but it serves a very important purpose in making the following case:

  1. These forces act on us whether we know it or not and whether we like it or not.
  2. The net impact of these forces on you is highly dependent on your genes, your age, and the choices you make (e.g., practicing gymnastics versus siting on the couch, changing your eating habits versus eating the same old standard foods).
  3. Just because some people seem to “defy” these forces does not negate their existence.  Michael Jordan dunking from the top of the free throw line doesn’t mean I can and doesn’t mean gravity is irrelevant.
  4. What matters most is how these forces act on you. Be less concerned with the folks who lie to either side of you on the population distribution curve (i.e., those more or less impacted by gravity or insulin).  Figure out what works for you and be ready to modify the plan over time, because it will likely get less and less easy to maintain and improve your performance over time.  We may not be able to outrun Father Time, but we can keep him at bay.


Astronaut Robert L. Stewart, Mission Specialist (MS), uses his hands to control his movement in space while using the nitrogen propelled Manned Maneuvering Unit (MMU). He is participating in a EVA, a few meters away from the cabin of the Shuttle Challenger. MS Stewart is centered in a background of clouds and Earth in this view of his EVA. He is floating without tethers attaching him to the Shuttle. Nasa.gov

Disclaimer: This blog is for general informational purposes only and does not constitute the practice of medicine, nursing or other professional health care services, including the giving of medical advice, and no doctor/patient relationship is formed. The use of information on this blog or materials linked from this blog is at the user's own risk. The content of this blog is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Users should not disregard, or delay in obtaining, medical advice for any medical condition they may have, and should seek the assistance of their health care professionals for any such conditions.


  1. Hi Peter,
    I was wondering if there was a way to find out (of course won’t be precise) about IR by monitoring glucose levels in blood?

    I’m taking around 6 grams of salt a day most of which by diluting in water, would it be better to take sodium supplement instead?
    If I’m taking enough sodium, do I still need to supplement megnesium?

    Many Thanks

  2. Hi Peter,
    Love your blog and the depth of information. Have been devouring it for weeks now! I’m committed to doing what I can to become more insulin sensitive. A few questions that keep hanging around for me (hopefully not stupid questions!) — 1) I understand that the only way to truly measure insulin is with a fasting insulin test, and I understand that a high blood glucose level suggests the pancreas is pumping out increased insulin to bring it back down, but is it also then reasonable to assume that a low blood glucose level implies a low insulin level? 2) I recently had my fasting insulin and glucose levels tested. I see there is a “HOMA-IR calculator” available on the web. Would inputting my numbers in this give an accurate result? (when I input my numbers the result was 3.21!)

  3. Hi Peter, thoroughly enjoyed the Gravity revision. I really expected some conclusion about the speed at which fat moves in gravity vs muscle, but I think i know the answer to that. But anyways, I also have a question about blood glucose levels when in ketosis after exercise and after meals. I don’t have a ketone monitor but I can measure glucose.

    Typical suggested levels of blood glucose for diabetes patients is between 7.7 and 4.5mmol/L. The low range of blood glucose is from 4.4 to less than 2.8mmol/M.

    Have your experiments provided some baseline data on your blood glucose levels while in ketosis.

    Is there a blood glucose level mmol/M at which ketosis is interrupted?
    thanks for a great resource

  4. Hi Peter,
    I was wondering if you could please comment on higher than expected HbA1c (5.5-5.9) among low carb/ketotic people. Is there a discordance that exists in this regard; less glucose leads to longer life span of red blood cells, hence higher HbA1c?

  5. Hi Peter,
    in the following study (table 4)
    they say that eggs, cheese and beef make our body release more insulin than pasta (for the same grams of food). Am I mistaken?

    I am sorry to bother you, but if that is true, shouldn’t it be advisable to eat pasta instead of beef, cheese or eggs? It is a honest question. I don’t get it. I don’t eat wheat and that is not going to change, but I would like to understand the data on that study.

    • Hi Peter,
      thank you for your answer.

      I believe the authors of the study also do that interpretation:
      “However, some protein and fat-rich foods (eggs, beef, fish, lentils, cheese, cake, and doughnuts) induced as much insulin secretion as did some carbohydrate-rich foods (eg, beef was equal to brown rice and fish was equal to grain bread).” (pages 1273-4)

      I did a google search looking for opinions on that study and I found Mark Sisson’s talking about the glucagon release compensating for the insulin release in high-protein foods:
      “For people without diabetes, the insulin and glucagon responses mitigate each other, and we’re looking at a healthy picture.”

      Gary Taubes also talks about glucagon when asked the same question:
      “Glucagon works opposite insulin on the fat tissue itself and so would be expected to mute or counteract entirely any fat accumulation stimulated by the insulin”. He also says that indeed lean meat can have that effect and it is advisable to eat meat with a high fat percentage. Moreover, instead of using only the AUC values he suggests the effect of that insulin release “is muted because it takes considerably longer to digest protein from meat”. I couldn’t find any graph showing the curve of serum insulin for meat compared with a high-carb food.

      Three more comments:
      1) I haven’t found another study corroborating/updating/expanding the insulin release data. Weird…
      2) I suppose when you are keto-adapted the results are completely different. That would be interesting to see.
      3) Are 120 min response curves too short? Would we see something different in longer experiments?

      Thanks again for your time!

  6. Peter,although Vincente seems satisfied with your answer and his followup research re the ajcn article table concerning the insulinogenic effect of various foods, I’m still confused. I interpreted it the same way Vincente originally did. And I have seen other articles speaking to the significant insulinogenic effect of protein such as whey and beef. If proteins and many other non carbohydrate rich foods trigger significant insulin release and, therefore, at least temporarily halt fat burning, don’t they have essentially the same effect as some higher glycemic foods?

    I know you have noted the substrates have different fates. But if de novo lipogenesis is minimal unless carb loads are extreme, it would seem that as far as fat loss/gain the most significant effect of raised insulin is the duration of the elevation and the halt to fat burning, So wouldn’t we would want to watch the timing and mix of consumption of any highly insulinogenic foods ?
    Maybe you can’t respond to this in a comment but it seems to deserve more discussion. thanks

    • I believe Vincente was responding to the column that noted the insulin response per g of CHO, which will certainly be higher in the protein meals. I’m looking at the AUC columns. Also, this is for a fixed kcal dose of each food, so it’s a bit misleading, since most people eat more of the energy from CHO than from protein and fat combined.
      I’m not saying protein doesn’t stimulate insulin, I’m saying this table is easy to misunderstand, glucagon notwithstanding.

    • Hi Peter and larry,
      the data that surprised me is from the third column (starting from the right side) of table 4:

      “Insulin AUC per g serving weight:”
      — White pasta: 22
      — Beef: 50

      Almost everybody says protein is less insulinogenic than carbs and that column came as a surprise to me. Surprising but I suppose not that important because I don’t think the absolute amount of protein in the diet is a variable I want to play with. The question for me is “more fat and less carbs or more carbs and less fat?” If protein is making the body release insulin, it doesn’t matter because I don’t want to eat less (nor more) of it. I don’t even think glucagon’s role is important for the same reason.

      Table 4 from the study doesn’t kill the insulin hypothesis, as long as low-carb diets do reduce the overall insulin release (see figure 1A from http://ajcn.nutrition.org/content/91/3/578.full.pdf+html).

      From the same figure 1A, I am not sure 120 min is enough to compute the insulin AUC. Some insulin release curves seem to reach as far as 4h.

      And I insist, may be we need a “table 4” for keto-adapted people.

  7. The Wikipedia article on insulin resistance has a few problems, especially the “Diet” session. It is quite disorganized, unpleasant to read, and some conclusions are misleading. The sources cited need to be scrutinized and their validity criticized when necessary, and new sources added, so that on balance the conclusions of the whole article are in sync with the best evidence we have today.

    One reason this is VERY important is that Wikipedia is one of the primary sources for information and for quick conclusions, and people use it to make decisions.

    One thing I’d like to read more often is:

    “Many studies seem to claim that X. Those studies are flawed, however, because (observational, clinically irrelevant, not replicated, not blinded etc). Better studies (prospective, blinded, independently replicated) have shown that actually Y, and this is what should guide our decisions.”

    Given the many equivocated conclusions, my impression is that they are being based on bad studies, but even most above-average readers are not educated enough to see that from the sources.

    Unfortunately, I’m not the best person to do this (not knowledgeable enough to accurately criticize the sources and add better ones), or I’d do it myself.

    Peter, do you agree this is relevant? Would you like to have a look and maybe edit the most damaging claims (or add updated relevant conclusions)? Or guide someone else in doing this accurately? Your bad-studies-unwarranted-conclusions detector is much sharper than that of most of us, and even a cursory reading of yours would quickly pick nonsense.


  8. “Consider, again, my example: When I was 16 years old I weighed 160 pounds, had between 4 and 5% body fat, a 28-inch waist and a 44-inch chest. Breakfast consisted of a box (not a bowl) of cereal. Lunch consisted of 7 turkey and tuna sandwiches (yes, 14 pieces of bread), a gallon of apple juice, and a plate of fries and gravy. Dinner was a pound of pasta and half a chicken. Despite eating over 1,000 gm of carbohydrate per day, I was quite resistant to them (i.e., I was very insulin sensitive) and remained exceptionally lean.”

    Omg, you should have never said that. Now I wont be able to sleep after reading that. I wish so so so much that this was my case, but unfortunately it is a complete opposite.

  9. Dr. Attia was kind enough to answer this question by another medium, so I will summarize it here. Dr. Attia suggested I look at the work of Drs. Phinney, Dashti, Westman and others. He also suggested there might be some new data on the subject in the next year or so. Thank you Peter!

    At the same time, I received a report from a client today whose C-peptide did indeed improve with 9 months of a VLKD.
    – Slow weight loss estimated around 40 pounds (but another 100 to go).
    – A1c 11.3 to 6.0
    – Stopped Victoza and Glyburide
    – Decreased metformin from 2,500 daily to 500 daily
    – Stopped HCTZ and Amlodipine

    …… and c-peptide 5.33 to 2.25 (ha ha! so c-peptide, at least, can improve!)

    This is not necessarily saying that his insulin resistance is improved or resolved. In fact, I believe he is still insulin resistant, although not formally tested with method such as insulin clamp. But I can say, at least, he is not as “hyperinsulinemic.”

    The endocrinologist indicated that the diabetic patients remained “insulin resistant” then reference a measure, which taken by itself, is only a measure of “hyperinsulinemia” (the c-peptide). Point being that a better, although not perfect, estimate of insulin sensitivity would probably be the QUICKI rather than a stand alone c-peptide.

    Thank you again Peter for responding!


    Justin E. Anderson, MD

  10. Dr. Attia. I cannot find much information regarding fasting insulin. I recently had a fasting insulin test done and it came back very low (1.2). My glucose was normal (95) and my A1c normal (4.6). Insulin had me very concerned but I cannot find much information about this test on the web. If I am eating low carb (75-125g/day) and was fasting approx 14hrs at the time of the test, isn’t this still a very low number?

Leave a Reply

Facebook icon Twitter icon Instagram icon Pinterest icon Google+ icon YouTube icon LinkedIn icon Contact icon