Search This Blog

Thursday, 26 July 2012

Fructose, Calories, Carbohydrate and de novo lipogenesis: What does "Hypercaloric" mean?

What does “hypercaloric” actually mean?

One possible objection to the prediction that fructose (and glucose from higher carbohydrate intakes) will enhance HCV replication by inducing DAGT1 and VLDL expression is the claim that is sometimes made that fructose only exerts strong effects on de novo lipogenesis in hypercaloric states.
Otherwise, it will be used to replenish glycogen stores or converted to energy.
The chemistry textbooks put this another way; glucose is converted to fat when it cannot be used to replenish glycogen or generate ATP. No mention of calories. The possibility is left open that other factors - micronutrient deficiencies, toxins, mitochondria defects, hormonal sensitivities - could influence the process.
(Note: I will skip between glucose and fructose in this essay; I think that anyone familiar with this topic will recognize that there is no sleight of hand going on. Fructose is a subset of carbohydrate, of which glucose is the main representative.)
The whole calories in, calories out controversy would take too long to review here. Suffice it to say that, in my opinion, counting calories in provides a both over-complicated and over-simplified way to do something that is more quickly and usefully done by counting (or estimating, rather) grams of fat, carbohydrate, and protein in a meal.
And counting calories out with any hope of accuracy is pretty much impossible unless you spend time in a laboratory or a high-tech gym, and the results there won’t tell you much about calorie expenditure in other conditions. That’s what your appetite ought to be doing…
The most sensible definition of a hypercaloric state is provided by Chris Kresser at Healthy Skeptic; basically, if you are gaining weight, your diet is hypercaloric. (Dr Kresser summarizes the case against fructose as a driver of DNL )

Unfortunately the loss and gain of body weight is a very protracted process and the weight of most people fluctuates from hour to hour and day to day.
Is there such a thing as a hypercaloric meal?
If I were to graph calories out in the course of a day, the baseline of expenditure (basal metabolic rate) would be quite high, with many peaks above it where I exerted my body or my mind; any troughs where the BMR dropped would be shallow.
If I were to graph calories in, you would see two huge curves around my two meals; the graph would be at zero at least half the time. There would only be four “isocaloric” points on the graph, two hypercaloric peaks (of perhaps 4 hours duration each) and the rest of the time, calories in would be less than calories out.
To confuse things even more, if you fast for a week no single day's eating can restore you to a "hypercaloric" state, but carbohydrate will be converted to fat as soon as glycogen stores are full...
Simply because that is how your body stores most of the energy it gets from carbohydrate. Glycogen stores cannot be expanded beyond about 150% of normal (and even that takes some doing - see the next link), whereas fat stores, as everyone knows by now, are pretty much capable of expanding indefinitely.

This makes it very hard to rely on the result of any particular piece of research unless we know a great deal about the surrounding conditions.
Lucas Tafur has done a worthwhile analysis of one DNL study.
Showing how things may not always be as they appear.

A further point is that fructose studies often use pure fructose, when what we should be concerned about is a) the combined effect of fructose and glucose, at the ratios similar to those found in sugar, HFCS, and fruit juice; and b) the combined effects of sugar and dietary carbohydrate from starch.
Recently Dr Peter Attia has published
some tables showing the difference in triglycerides after feeding of glucose, fructose, and HFCS.

A further complication is, that in the case of type 2 diabetes (a common complication of Hepatitis C), there is already a high blood glucose level due to increased hepatic glucose production.
Does this elevated blood sugar then mimic a “hypercaloric” state whenever additional sugars are fed? Any fed sugars have to compete with blood glucose for conversion to glycogen or ATP.
And DNL, according to the textbooks, occurs when sugars are surplus to amounts that can be converted to glycogen or ATP (and if DNL cannot take care of the sugars, they will be dumped through the kidneys, a convenience which is not conducive to good kidney function in the long haul, hence the connection between diabetes and kidney disease).
Calories, of course, can also come from fat or protein. Is a meal of fat likely to prevent replenishment of glycogen?
R. D. Feinman points out that fructose is easily converted to glycogen when total dietary carbohydrate is restricted.

That’s where we want to be. That’s the sweet spot, if you’ll pardon the expression. Restriction of carbohydrate means we don’t need to worry about the occasional fruit we eat, or the sugar in the pickle we put on our bacon and eggs, or the sugar in our dark chocolate, within reason. There’s fructose in beetroot, carrot, potato, and onion that no-one ever mentions.
Of course, carbohydrate can, at least in theory, cause fatty liver and high TG without any DNL whatsoever; if the liver is preoccupied with burning carbohydrate, it may be unable to convert all the fat you eat into ATP; in which case the fat will be recycled into triglycerides, to be stored in the liver or released as VLDL (or IDL).
You really do want to be primarily a fat-burner, either way.

The problem with the “hypercaloric” version of fructose-driven DNL is that it promises that “if you don’t eat too much, it doesn’t matter”. But the science doesn’t tell us how much to eat and when to achieve this miracle.
And restricting calories overall tends to increase appetite and mess with our best intentions.

Whereas R. D. Feinman’s prescription “if you don’t eat carbs (much) it doesn’t matter (much)” is more practical.
Or, in Professor Feinman's own words, "as carbohydrate and calories are reduced, any effect of fructose will be minimized. In the extreme, if you are on a very low carbohydrate diet, any fructose you do eat is likely to be turned into glucose".
Glucose, for glycogen storage or conversion to ATP, rather than fat.

A philosophical digression

There is a critical philosophical distinction between the type of notions that calories (from physics) and macronutrients (from chemistry) represent.

A sugar molecule is real, a gram is abstract, but you can look at, handle and taste a gram of sugar.
A calorie is abstract and amorphous (it could be applied to food, petrol, wood, coal, uranium, movement, noise, heat, radiation and so on).
There are glucose receptors, metabolites and so on. There is no receptor or enzyme that deals with calories. Saying "calories (in)" is often just a lazy and potentially misleading way of saying "so many grams (or ounces etc.) of macronutrients in such and such a ratio" (just as carbohydrate is a lazy way of saying "starch and/or sugar" which is itself a lazy way of saying something even more informative).
(If I say "a whale 100metres long" I am giving you extra information about the whale. You are not meant to be focusing on the implication that a metre is 1/100th the length of my whale.)

A calorie does not have a "nature". There is no natural history of the calorie as there is a natural history of every nutrient.
So why bother? Calories (or joules) are the only way to measure calories out for comparison with nutrition taken in. This is useful for calculations of work and food allowance - as in rationing, for example, or for planning calorie restricted or overfeeding diets.
But it does lead to a false assumption - that what is the only useful measurement of metabolic output must therefore be the only measurement of input worth considering.


Manythings said...

Thanks for clarifying the position on fructose - and calories, for that matter. This makes a lot of sense.

ProudDaddy said...

Been reading and appreciating your comments in other blogs for a long time, so I'm very pleased to have stumbled onto your blog! Regards, Wayne.

feinman said...

I am not sure that calories are so amorphous as that it is not used accurately, that is, the calorie in physics is well defined in terms of the specific heat of water and this carries over into biochemistry in the energy of chemical reactions. The calorie in nutrition, however, refers to the energy of a specific reaction, namely, complete oxidation to CO2 and water. Other reactions are related to this but the calories go with the reaction not with the substance. It is not like particle physics where mass is given in energy units because of E=mc2. What this means practically is that, when you carry out life processes, making cell material, making protein, the calories you eat do not apply directly.

Puddleg said...

I agree that the use of the calorie is way to measure what can be measured, and this is always good science. There are many life processes that consume energy but produce or reduce complexity (more or less elaborate structures). When nutrients are used to synthesise cholesterol, for example, this has mass but cannot be converted back to metabolic energy.
The "hypercaloric" state that increases hepatic DNL from fructose is specifically a "hypercarbohydrate" state, as far as I can see. (This seems to be the take-home message from your last post, Proff. Feinman In which case it is more informative to describe it as such when describing the relationship.
There are what seem to be common genetic variations in individual metabolism of fructose in the literature and I will be posting some information about this in the future.