Glucose and NEFA: From Dysfunctional Metabolism to Toxicity
This post started out as a comment in response to Paul Jaminet's mention of a post on this blog in his Around the Web post yesterday. It got rather long so I decided to move it here. In the interest of more rapid publication, I'm not going to be doing a whole lot of referencing in my discussion here, but if you're interested in a particular statement please indicate so in the comments section and I'll try to track down the reference(s) I have in mind. Paul writes:
Every once in a while someone writes to ask me if they should fear a high-fat diet because of CarbSane’s writings on lipotoxicity. I reply that lipotoxicity only appears after metabolic syndrome has developed and, while it may drive the transition from obesity to diabetes, it is not a cause of obesity, and not a danger to people who don’t have metabolic syndrome. Also, the implications for diet are not obvious, since carb intake suppresses NEFA clearance from the blood and enhances glucotoxicity. The literature commonly speaks of “glucolipotoxicity” to describe this compounded toxicity problem. CarbSane hasn’t always been clear on these points, so it’s good to see an excellent post from her covering the basics.
Thanks for the shout out Paul!
So the statement that got my mind in motion this rainy Sunday was this: "carb intake suppresses NEFA clearance from the blood" . I'm not sure if this is a typo or not, but it is actually the opposite that seems to be true. And here's where Gary Taubes gets things right (if taken woefully out of context). Carbohydrates stimulate insulin release and insulin's action on the fat cell is two fold - it stimulates NEFA uptake by stimulating esterification (it will also further stimulate ASP -- thus ASP-stimulated esterification -- in addition to that stimulated by dietary fat/chylomicrons). Secondly, insulin will suppress NEFA release from the fat cells.
Insulin is, after all, mostly just a traffic cop in the metabolic system. It does not appear to have much impact on metabolic rate, rather it regulates which fuel the cells are "burning" for energy at any given time. To simplify matters, let's just deal with oxidation (burning) of glucose and free fatty acids (NEFA) and ignore protein and ketones (and alcohol).
Circulating glucose levels are regulated by insulin as follows:
- When glucose levels are high, by facilitating clearance from plasma/uptake into cells and "disposing" of it. Glucose disposal is through one of three paths:
- glycolysis - burning for energy, otherwise known as oxidative disposal
- glycogen synthesis - otherwise known as non-oxidative disposal
- de novo lipogenesis - where the products of excessive glycolysis are converted to fatty acids.
In those with impaired glucose tolerance, the defect appears to be in path (2) and perhaps it is as simple as having glycogen stores chronically "topped off". We have limited glycogen stores in which to stuff large quantities of glucose.
- Insulin suppresses gluconeogenesis and glycogenolysis (breakdown) in the liver and release of glucose into circulation when circulating levels are high.
It is because of the rather limited storage buffer capacity that glucose is given preference in the heirarchy of fuel usage. One of insulin's primary jobs in policing fuel selection is to stimulate uptake and both oxidative and non-oxidative disposal. But it also acts to suppress beta-oxidation, aka the burning of fatty acids for fuel. This is where the whole LC canard of shutting down fat burning comes from.
All of this is fine and dandy until the NEFA get out of control. The supply of fatty acids to the cells is not supposed to come from dietary fat. Dietary fat is packaged in chylomicrons with distinct signalling proteins so that cells (especially fat cells it appears) can distinguish it from other triglycerides (e.g. VLDL). Fat is constantly being released from adipocytes in the form of NEFA and re-cycled back into triglyceride form both by peripheral tissues or re-uptake into fat cells themselves. The level of NEFA is regulated primarily by insulin's suppressive action on lipolysis by hormone sensitive lipase (HSL) in fat cells. In the fasted state, NEFA are higher, but in normal people, following a mixed meal, they are knocked down to "baseline" rather quickly. Keith Frayn phrases this as release clamped down to near-zero in normal individuals. In this graphic from the post of mine Paul was referring, we see this occurring after a large breakfast and smaller lunch five hours later.
Following a mixed meal, the metabolic state of fuel delivery is shifted towards one of almost entirely glucose. As glucose is disposed of, NEFA release ramps up to meet fuel needs as metabolism switches towards fat burning.
I believe the evidence is overwhelming for Keith Frayn's (and many others') hypothesis for the etiology of insulin resistance initiating in the fat tissue. As adipocytes fill and grow in size they lose insulin sensitivity. This leads to failure to suppress NEFA release following a mixed meal, and as Frayn has also shown, in the obese, there appears to be a considerable contribution to NEFA from dietary fat due to impaired trapping. Fatty acids liberated by lipoprotein lipase (LPL) intended for storage in the adipocyte essentially escape into circulation in larger amounts. The result of these two in concert leads to heightened delivery of NEFA to the cells at a time when they need most to be accepting the glucose delivery.
Here's where things get worse. Our cells can refuse the glucose delivery but they are limited in their ability to refuse the fatty acid delivery. This is because fatty acids flow into cells by a facilitated diffusion mechanism. There does appear to be some sort of transport facilitation going on (and it may be cell specific), but it is not required nearly as much as glucose requires it's transporters (the various GLUT's) for passage in. The excessive delivery of fatty acids to the cells forces them to switch to burning them for fuel. They also suppress glycolysis and non-oxidative glucose disposal which means that the glucose channels are essentially "clogged" and the cells refuse it. When this happens, you get more elevated glucose spikes and lower rate of return to fasting levels ... in other words, impaired glucose tolerance (IGT). This is peripheral IR.
The other thing that chronically elevated NEFA do is that they seem to suppress the ability of pancreatic beta-cells to mount an appropriate acute insulin response to carbohydrate. This exacerbates the hyperglycemia. However this action of NEFA appears to be contributing to metabolic dysfunction and not exactly exerting a toxic effect on the beta cells. Glucose also plays a roll in metabolic dysfunction before becoming overtly toxic in that it causes the IR individual to secrete more insulin to compensate for the hyperglycemia. Apparently this can go on for years, perhaps even indefinitely, in those genetically gifted individuals with the capability to continue this compensation.
Glucose does not appear to be very toxic in and of itself until you get to a frank diabetic state where glucose levels are chronically elevated and glycation wreaks its havoc. The NEFA do appear to have somewhat toxic effects on the cells directly especially in the sedentary or hypercaloric state during this "metabolic mayhem" phase. Let me explain what I mean. In the normal person consuming an energy balanced diet the metabolism switches almost effortlessly between fat and carb burning, although in reality there's always some of each going on, and certain cells are given preference to available fuel to burn. However in the hypercaloric or obese IR state (worsened further if still gaining weight on hypercaloric diet) cells are forced to burn fat and it seems that the buildup of diacylglycerols and ceramides as a result and ROS formation exert lipotoxic effects. It appears that esterification to form lipid droplets of triglycerides in the cells (often referred to as intramyocellular triglycerides, IMTG, in muscle) is protective in nature. These in and of themselves are also not detrimental as conditioned athletes tend to have high levels of IMTG yet remain exquisitely insulin sensitive. This is explained by their high energy expenditure so when these fatty acids are burned, the DAG's and ceramides aren't produced to do their damage. The exact mechanism by which DAG's and/or ceramides impact cell function have yet to be elucidated, but their content in the cells is correlated with chronic peripheral IR. Still, at this point the toxicity is more "mucking up the works" than totally destructive.
So all of these dysfunctions lead to the truly toxic mix of hyperglycemia + elevated NEFA = glucolipotoxicity. It appears that either can do damage separately, but the mix of the two is generally implicated in beta-cell apoptosis (cell death) and apoptosis of various other critical tissue cells (e.g. liver, vascular endothelium, heart, kidney) . It's like a car that runs rough vs. one that won't start anymore. The cell death is the ultimate manifestation of toxicity!
Thus, you want to avoid this deadly duo at all cost. However I cautiously disagree with Paul's assessment that a high fat diet cannot be problematic in the absence of hyperglycemia. It could be causing your cells to "run rough" to use the car analogy. I don't see a problem with PHD if it establishes and sustains a suitably lean condition. And the failure of VLC diets to suppress postprandial NEFA release is not an issue, IMO, on a diet containing 100 or so grams of starch. However, since NEFA are rarely measured in lab tests, and neither is pancreatic or liver fat content routinely measured, we are all flying a bit by the seats of our pants on this issue and knowing exactly what's going on.
I'm not a doctor or dishing out medical advice, but based on what I've been researching in this area, here's a couple of things that make sense to me.
- In one study on VLC diets, elevated fasting (and postprandial) NEFA were correlated with LDL. So ... if you have normal LDL you're probably fine. And by normal, I mean a bit higher than causes most lipophobic doctors to get concerned, but not the dramatic increases some experience eating VLC.
- Fasting or going on a VLCal diet a few days a month can't hurt to clear out any backlogged lipids. Where there is rapid turnover, it's less likely for ROS production and such. A mere 8 weeks of severe caloric restriction restored normal beta cell function attributed to reduced pancreatic fat in that recently published study.
- If you're not losing weight on VLC and/or remain significantly overweight it seems prudent to err on the side of caution. Middle of the road approaches seem far less effective than the extremes, but high carb/low fat seems the more logical approach to long term restoration of insulin sensitivity.
- Live an active life or exercise. Don't be a couch potato.
- Perhaps do periodic "carb ups" where you ramp up carb intake slowly over a period of a week or so then test for glucose tolerance. I don't know, the more I learn about insulin and its myriad functions in the body aside from glucose trafficking, the more I believe having it flowing and signaling properly is the optimal state.
- If you struggle with blood pressure control, NEFA appears to interfere with insulin's vasodilatory effects and this may be the problem.
- Listen to your body.
It seems to me that while eating a high fat diet will not cause diabetes in those not predisposed, a ketogenic diet has been shown to hasten the progression of diabetes in a genetic diabetic rat model. I don't think this result should be dismissed lightly especially since we have no real human cultures to study in this regard. And, after all, though diabetes tends to have a heavy genetic component, it has been known to happen to folks with no family history of the disease. I, personally, have no concerns over VLC/VHF (though I prefer a leaner version of the diet) in the context of a reducing diet. Why? Because the elevated NEFA are also being utilized for energy.
But the health risks of obesity seem quite closely tied to the degree of insulin resistance the obese state creates. I'm not entirely sure that a normal weight person consuming a high fat diet doesn't run the risk of being one of those metabolically obese thin people, with most of the inherent risks of IR (IR is a CVD risk factor independently of IGT/diabetes) to go with. Just my opinion.