Wednesday, October 31, 2007

Do you believe in MRI scanners?

I was thinking about mitochondria, but I got sidetracked when I remembered Ling and his opinion about ATP. Here's how it went:

The vast majority of the energy production in our bodies occurs in our mitochondria. These are the tiny powerhouses which generate most of the high energy adenosine tri phosphate (ATP) on which we ultimately run our metabolism.

NB an aside; Gilbert Ling does not believe in ATP as an energy source for cells (though he thinks it is a crucial molecule), any more than he believes in the lipid bi-layer cell membrane or the sodium/potassium ATP pump. He is truly out on a limb with his "Association Induction Hypothesis" of life at cell and below cell level. A fruitcake, obviously. Except his ideas on the localisation of cell water around proteins led to the invention of the MRI scanner. If MRI scanners work, that cookie Ling is right. And almost all of modern physiology is wrong.

It is, for me personally, MUCH easier to think within the "normal" physiology of my education. But, because MRI scanners work, this is the equivalent of thinking that the table in front of me is "solid" when quantum mechanics tells me that it is anything but solid...

Now there's a thought.


Wednesday, October 24, 2007

Dietary sins of Swedish children

Dr Haglund Garemo is a paediatric nutritionist at Göteborgs University in Sweden. She defended her PhD in December 2006. The abstract is available here. She studied the socioeconomic status, diet, weight and health markers in a group of 4 year old children.

The relevant quotes from the abstract are

"The energy intake/kg was according to Nordic nutrition recommendations"

"Most children had a higher intake of saturated fat and sucrose than recommended"

"A higher BMI was associated with lower fat and higher sucrose intake"

"A lower fat intake was associated with higher BMI and higher HOMA ß-cell function"

If you had to summarise these findings they would be, with a little license, as follows: Eating fat makes you thin. Eating sugar makes you fat. Eating sugar makes you hyperinsulinaemic.

Here are some interesting interview quotes from Dr Garemo in addition to her thesis abstract:

Food and Drink Europe quote her in an interview as commenting:

"analysis of the children's body build showed that weight increases was a result of the body storing more fat, but those who ate the most fat were not the ones who weighed most. Instead, children who ate less fat had higher BMIs"

News-medical have this quote. Odd way of phrasing it, "less heightened"!

"Insulin production was less heightened in girls who ate more fat"

"Such results would go against the common perception that fat causes increased insulin production as a result of insulin resistance"

Overall, eating fat came out rather well. Most of it was saturated. That is an understatement. Interestingly that last quote above is the explanation of the bizarre medical idea that eating fat raises your blood sugar level.

It's wrong, excepting the ideas in my post on physiological insulin resistance. I doubt any of these children were eating enough fat to be in ketosis! BTW if you ever come across a diabetic person having a severe hypoglycaemic episode, please do NOT give them butter to try and raise their blood sugar, they might well die. Try sugar.

That should tell you something. Duh.


Tuesday, October 23, 2007

Physiological insulin resistance

Back in mid summer 2007 there was this thread on the Bernstein forum. Mark, posting as iwilsmar, asked about his gradual yet progressively rising fasting blood glucose (FBG) level over a 10 year period of paleolithic LC eating. Always eating less than 30g carbohydrate per day. Initially on LC his blood glucose was 83mg/dl but it has crept up, year by year, until now his FBG is up to 115mg/dl. Post prandial values are normal. He wanted to know if he was developing diabetes. I've been thinking about this for some time as my own FBG is usually five point something mmol/l whole blood. Converting my whole blood values to Mark's USA plasma values, this works out at about 100-120mg/dl. Normal to prediabetic in modern parlance. However my HbA1c is only 4.4%, well toward the lower end of normality and healthy. That's always assuming that I don't have some horrible problem resulting in very rapid red blood cell turnover. I don't think so... I spend rather a lot of my life in mild ketosis, despite the 50g of carbs I eat per day. So I can run a moderate ketonuric urine sample with a random post-chocolate blood glucose value of 6.5mmol/l. What is happening? Well, the first thing is that LC eating rapidly induces insulin resistance. This is a completely and utterly normal physiological response to carbohydrate restriction. Carbohydrate restriction drops insulin levels. Low insulin levels activate hormone sensitive lipase. Fatty tissue breaks down and releases non esterified fatty acids. These are mostly taken up by muscle cells as fuel and automatically induce insulin resistance in those muscles. There are a couple of nice summaries by Brand Miller (from back in the days when she used her brain for thinking) here and here and Wolever has some grasp of the problem too. This is patently logical as muscle runs well on lipids and so glucose can be left for tissues such as brain, which really need it. Neuronal tissue varies in its use of insulin to uptake glucose but doesn't accumulate lipid in the way muscle does, so physiological insulin resistance is not an issue for brain cells. However, while muscles are in "refusal mode" for glucose the least input, from food or gluconeogenesis, will rapidly spike blood glucose out of all proportion. This is fine if you stick to LC in your eating. It also means that if you take an oral glucose tolerance test you will fail and be labelled diabetic. In fact, even a single high fat meal can do this, extending insulin resistance in to the next day. Here's a reference for this. The general opinion in LC circles is that you need 150g of carbohydrate per day for three days before an oral glucose tolerance test. I did this carb loading thing, then performed my own OGTT. It came out very normal except for mild reactive hypoglycaemia. So, I often walk around with a fasting blood glucose of 5.9mmol/l and in mild ketosis, yet have normal pancreatic and muscle function, provided I carb load before the test. BTW my FBG dropped to 4.3mmol/l after three days of carb loading. That then raises the question as to whether Mark "iwilsmar" and myself are typical of LC eating people, or an oddity or two. This brought to mind the self selected macronutrient study performed on mice by Ortman, Prinzler and Klause. They allowed mice to select their own diet and, lo and behold, the mice chose (by calories, not weight!) 82% fat and 5.6% carbohydrate. Sensible mice. NB These German mice should each be given Professorships of Nutrition at medical schools in the most obese nations of the world. Quite what we should do with the current professors I'm not sure, but I bet the mice could think of something. Anyway, these mice are cool. The only thing that bugged me when I first read the paper was that they had a higher fasting blood glucose than those poor mice fed the normal junk which passes for laboratory mouse "chow". This now fits in to an overall pattern. Elevated non esterified fatty acids induce physiological insulin resistance and a higher than expected FBG level. A simple switch to higher carbohydrate eating (in myself) allows the normal underlying pancreatic and muscle function to show. It also fits in with the FBG of 3.5mmol/l found in the carbohydrate fuelled natives in the Kitava studies. So do I worry about a FBG of over 5.5mmol/l? Not while my HbA1c is 4.4%. Peter

Wednesday, October 17, 2007

Familial Hypercholesterolaemia

Familial hypercholesterolaemia (FH) is a common genetic disease. That says something quite profound to me. If it is both genetic and common there cannot be too strong a disadvantage to it on an evolutionary basis. There are also lots and lots and lots of variants. All affect the LDL-C receptor. If you get one copy of any one of a large selection of defective receptor genes, you have quite high levels of cholesterol in your blood and... well, you know the story about clogged arteries and heart attacks. In fact, if LDL-C is such lethal stuff you would have thought that evolution would not have been so careless with the stability of the crucial piece of DNA coding for its receptor. But it wasn't careful and there are a myriad of FH variant genes. Oddly enough people in the 1800s with FH lived as long as, or even longer than, "normal" people, check out the Dutch pedigree and Utah pedigree studies.

Now this is interesting. In 1807 LDL-C at above normal levels was beneficial, particularly in Holland. In 2007 it is an automatic statin deficiency and don't try to get life insurance. What's going on?

If you ask a cardiologist about the function of the LDL-C receptor you will find that it is used by the liver to mop up that horribly artery clogging LDL-C before the patient secumbs. However, there are LDL-C receptors in other places than the liver. Such as on the endothelial cells lining the arteries. If FH sufferers have non functional receptors in their livers I am willing to bet they have poor receptors on their endothelial cells too. But wait a minute, this should be good! If you have reduced receptors you should have reduced stickiness and LESS clogging of the arteries.

But there again, the receptor is normally present and evolution does not go to the effort of building a complex structure just for the fun of wasting protein resources. No, that receptor is there for a reason, and that reason is to stick LDL-C to endothelial cells when they need it. Interestingly, isolated endothelial cells produce lots of LDL-C receptors, where as endothelial cells in contact with their brethren don't. The conclusion from this is that in areas of damage, the endothelial cells are isolated and express lots of receptors which cry out for cholesterol. In FH they don't get the cholesterol they need because the receptors don't work. Cell repair is difficult without a handy supply of lipids.

So, if you had to suggest what sort of problem might be caused by a defective cholesterol receptor, then vascular problems might be a good guess. Possibly heart attack.

NB What do you think might happen if you got two genes for a defective LDL-C receptor? This is homozygous FH. It's bad. You can see why.

So why wasn't heart attack common in FH carriers in 1807? How common was sugar in 1807? Not very. Perhaps there is a link here.

Just a minute. Where did all of that cholesterol come from in the arteries of modern sufferers? Well, you can stick LDL-C to vascular tissue with things other than the apoB-100 receptor. Try the LOX1 receptor for a start. This binds oxidised LDL-C rather nicely. Oxidised cholesterol is seriously nasty stuff. What oxidises cholesterol in your bloodstream? Try fruit and vegetables. Try sugar. Try a low fat diet. My guess is that these were thin on the ground in 1807 in Utah or Holland.


Thursday, October 04, 2007

Niacin and beta hydroxybutyrate

There is nothing about antioxidants in general that affect lipoprotein levels. Niacin undoubtedly does have significant action on blood lipids so undoubtedly is more than an antioxidant. I find that interesting. Why should that be? Many biological processes act through receptors. The niacin effect on lipids, and the niacin flush (ever tried this? It's fun fun fun, not!), are receptor mediated. Clever people have found at least a couple of receptors. The first was HM74, now joined by HM74A. These receptors were not sitting there waiting for either psychiatrists or cardiologists to prescribe multigram doses of nicain, useful though that may be.

No, they have a natural ligand, beta hydroxybutyrate. Beta hydroxybutyrate is a substance dear to my heart, and your heart too, see here. It's a ketone body, naturally manufactured by the liver in times of starvation or carbohydrate restriction. Plus times of coconut fat consumption too, as medium chain triglycerides (MCTs) will produce ketone bodies even in the presence of carbohydrate. By the way, this is not my preferred fat, and it worries me a little that humans break down MCTs as fast as the liver can do so, plus they are shunted down the hepatic vein rather than the thoracic duct, minimising their access to the general circulation. It just reminds me of our metabolism's approach to fructose a bit. Still, coconuts have an excellent track record as a human food in the tropics, so I'm probably just being a bit paranoid here.

Low carbohydrate diets naturally produce ketone bodies. They will certainly elevate HDL cholesterol levels too. To which you may ask, "so what?". Well, elevated HDL cholesterol appears to be a marker of a high fat, low carbohydrate diet and its associated beta hydroxybutyrate. So it is a marker of good things happening in the metabolism. As such I welcome it, but not if it is an effect of some drug. Of course if your primary protection against heart disease is normal blood sugars and low insulin levels the elevated HDL-C is not essential, as seen on Kitava. On western food supply a little exercise plus LC eating seems to be the easiest way to maintain normal blood sugars, and coincidentally elevated beta hydroxybutyrate and so HDL-C. Statins and niacin may elevate HDL cholesterol but my guess is that their benefits (however small from the statins) are unrelated to their effect on HDL-C. And of course, once you are on a drug to elevate HDL-C, there is no way of telling if your metabolism is doing well or badly in terms of insulin sensitivity. If you combine a statin plus niacin plus the dreaded AHA Coco Krispies based low fat diet, you may well be on the road to a cardiovascular disaster, despite having the appearence of a "good" "good" cholesterol level...


Wednesday, October 03, 2007

Niacin and adrenochrome

We know that torectrapib was excellent at raising "good" HDL cholesterol, while simultaneously increasing the human death rate from heart attack to the point where it couldn't be ignored. We know that the Kitavans have low "good" HDL cholesterol and no heart disease. So we should forget good and bad cholesterol and concentrate on the things that really matter, which are probably hyperglycaemia, hyperinsulinaemia and wheat. Plus a few other things like trans fats. But the HDL as "good" cholesterol hypothesis will not lie down, although it seems to have been dead for some time. The statins raise HDL, but can be ignored due to their host of confounding pleiotropic effects, inseparable from their cholesterol effects. The current flavour of the month for raising HDL, since the demise of torcetrpib, appears to be niacin, vitamin B3, in multigram doses.

There is no doubt that niacin elevates HDL cholesterol levels and, according to those who follow these things, decreases overall mortality. Now niacin is a very interesting drug. Oh, at 2.0g a day niacin is a drug, not a vitamin. Its career began with Abraham Hoffer, who is still going strong, back in the 1950s. On the basis that "vitamin" doses of niacin had half emptied the psychiatric wards when the problem of pellagra was solved by Goldberger, Hoffer tried massive doses of B3 on schizophrenic patients. The idea was to see if they had a poor response to "normal" doses of B3 which could be overcome by high doses. It worked.

Hoffer's publication in 1954 was last-authored by Smythies, so I assume he was Hoffer's supervisor.

Smythies is also still going strong and has recently summarised the adrenochrome theory of schizophrenia in one of his many publications.

My guess is that niacin may be working by reducing adrenochrome in the brains of schizophrenics. I've no idea of how this works, but it is interesting to note that adrenochrome is an oxidation product and niacin is an effective antioxidant.

That dynamo of genuine cardiovascular research, Kummerow, has looked at adrenochrome in the blood of hypertensive patients. It's there and it mangles the vascular epithelium. Now, does niacin reduce adrenochrome in the blood stream as well as in the brain? It is, after all, a significant antioxidant and anti inflammatory agent. No one has looked at this as far as I know. I certainly don't know if this is the case, but I'm suspicious.

If it does, who cares about the HDL cholesterol effect? Like the statins, niacin appears to be a useful drug, possibly a very useful drug, chosen for the wrong reason. Unlike the statins, niacin has relatively little "badness" attached to it. The problem with torcetrapib was that it ONLY adjusted cholesterol levels. Whatever its associated "badness" or "goodness" might have turned out to be.

That's what happens when you spend millions or maybe billions of dollars developing a drug based on a wrong hypothesis.