Wednesday, May 12, 2010

Arteriosclerosis and sulphation factor

The alpha granules of platelets contain many substances which are useful for repairing damaged tissue, especially damaged blood vessels. That's no surprise.

The addition of sulphate to sugar and sugar derived molecules is essential for the production of many glycosaminoglycans. Most of the work with sulphation has centred on cartilage development but an equally interesting location of sulphated GAG production is the arterial wall which leads to the formation of fibrous plaques, via proteoglycans, in stressed areas.

In ancient days of old Salmon and Daughaday (1957 from the look of this title) wanted to assay Growth Hormone. They developed GH deficient rats (better not ask how) then took cartilage samples and added GH to see if they could get sulphate uptake as a method of assaying GH levels in serum, a bioassay. The idea failed and they eventually found out that "Sulphation Factor" was needed for sulphate uptake and SF was produced in response to GH by the liver, GH per se was not active on isolated chondrocytes. To confuse matters further, at some stage SF was re named Somatomedin. To make matters even more confusing somatomedin was even later discovered to be ILGF-1. There is a nice summary of the story here.

Even today we still use ILGF-1 as a surrogate for measuring GH in cats with GH secreting tumours.

So the sulphation of proteoglycans is mediated through ILGF-1. A brief scan (Wiki) of the contents of alpha granules in platelets lists ILGF-1 as the first ingredient. Platelets deliver ILGF-1 to damaged areas. It's not exactly surprising that ILGF-1 promotes proteoglycan sulphation as well as, or as part of, tissue repair.

Human beings are not designed to have fasting insulin levels over 200pmol/ml with a rise of up to 1000pmol/ml (in extreme cases) post prandially. If we do things to ourselves which produce this effect we can expect to have insulin interacting of ILGF-1 receptors which were never designed to see such astronomical insulin levels. Even if their true ligand is at minimal concentration, arterial wall cells will behave as if platelets are delivering ILGF-1 in "damage repairing" amounts. Even in the absence of damage.

It does not take a great leap of faith to suggest that the normal physiological response to normal vascular injury might be driven beyond an appropriate level by chronic hyperinsulinaemia acting on ILGF-1 receptors.

Perhaps type two diabetics should be at increased risk of heart disease? I wonder if anyone has checked this...


BTW anyone who has followed the homocystine hypothesis of CVD will find the title (all I can get) of this paper of some interest.


Ned Kock said...

Hi Peter.

The problem seems to be that IGF-1 and insulin are similar in molecular structure, thus physiologically binding to their own receptors and pathologically cross-binding to each other's receptors. Right?

Possibly another example of evolution creating a semi-redundant mechanism in the absence of certain factors (hyperinsulinemic foods), and the "smart" ape then later engineering those factors.

I wonder if insulin is the real culprit though. Beef, for example, leads to an average insulin level of 65 (pmol/l) postprandially (120 min), whereas white bread to 145. These were calculated based on these tables:

Yes, it is a big difference, but I wonder if it is enough to cause all of the problems that we see today.

Maybe this difference is underestimated. The numbers from those tables were obtained based on isocaloric portions of white bread and beef. I guess most people will tend to eat a lot more white bread in one sitting than beef ...

But I still wonder.

Peter said...

I guess the other aspect is that these are the numbers for a slim 21 year old. If you have type two diabetes and need to shift glucose which won't go anywhwere, would you need a much higher insulin response to white bread which provides glucose compared to beef which provides insulinogenic amino acids? I'd doubt the effect of beef would be as bad as bread but I've not seen that for IR people. Bernstein certainly does not worry about using extra insulin to cover beef (vs any other protein). I'd like to see Holt's results for chicken (I couldn't see it there) as this provokes a zero insulin response, according to Dandona. Be nice to check Holt's numbers... Has anyone done this that you know of?


shawn57187 said...

Insulin has many other functions aside from simply regulating blood sugar. I don't think the beef and white bread insulin index comparison is a fair one. For instance, one will see a higher glucagon response (insulin’s counter-regulatory hormone) from beef than white bread.

Ned Kock said...

There are quite a few citations to Holt's article on Google Scholar:

I couldn't find one that actually checked the numbers; there must be one somewhere in that group.

Of course shawn57187 is correct about the hormone mix for beef vs. white bread.

I would be surprised if chicken led to a zero insulin response, unless we are talking about the insulin response in the dead chicken ...

John said...

The glucagon response is an important consideration. Obviously the percent fat in the beef is factors in as well. Is this detail available? The lack of insulin response from eggs [compared to beef] would lead me to think the beef used was lean.

shawn57187 said...

I would supposition that increasing the fat content of the beef might potentiate the insulin reaction.

GLP-1 release is controlled by L-cells in the gut, and is highly sensitive to nutrient content. Secretagogues for GLP-1 include fat, protein, and carbohydrate, with fat enhancing the GLP-1 response from the latter two.

The significance of GLP-1 is that it increases insulin secretion, decreases glucagon secretion, and increases beta cells mass and insulin gene expression.

GLP-1 also controls satiety, which may be why people eat less on low-carb.

Glucagon-like peptide 1 promotes satiety and suppresses energy intake in humans.
"In conclusion, the results show that GLP-1 enhanced satiety and reduced energy intake and thus may play a physiological regulatory role in controlling appetite and energy intake in humans."

Ghrelin and glucagon-like peptide 1 concentrations, 24-h satiety, and energy and substrate metabolism during a high-protein diet and measured in a respiration chamber
"An HP diet, compared with an AP diet, fed at energy balance for 4 d increased 24-h satiety, thermogenesis, sleeping metabolic rate, protein balance, and fat oxidation. Satiety was related to protein intake, and incidentally to ghrelin and GLP-1 concentrations, only during the HP diet."

Something to think about I guess?

Peter said...

Plenty to think about, yes.

Here's where I pulled the chicken insulin response from. Doesn't look like they controlled for calories per portion though

Off to Skye for a week now, hoping for sun between the rain...


blogblog said...

I'm thinking that the insulin response to meat would be far less in long-term low-carbers. I assume that the initial insulin levels in most people is actually a conditioned response to an (anticipated) high carbohydrate meal.

shawn57187 said...

blogblog, the anticipation of the meal could certainly exacerbate the insulin spike. However, insulin is required to shuttle amino acids to the muscle cells and start protein synthesis. Leucine in particular is known to be very insulinotropic.

donny said...

I wonder if the same idea relates to the degeneration of cartilage in the joints? If IGF-1 is necessary to proper cartilage maintenance, excess insulin signalling could mess things up there, too. Cartilage is sort of at the end of the supply line, not much blood flow. Excess insulin, promoting glucose-wasting, localized hypoglycemia?

webster said...

The energy density of the beef used in the study is equal to that of egg, so from there you can draw your own conclusions of its fattiness.

For beef, I figure it has to be either fatty or amino acids per se eliciting the insulin reaction since the glucose score is the least of the protein foods, though I don't understand why...palmitic acid?

This may explain why it's easier for me to reach satiety on eggs vs 73% ground beef. These results are surprising to me as egg yolks contain 1g carb each.

I still don't understand why it's easier to have "enough" melted butter than solid butter. (Anybody know?)

I must eat over 3,000 calories of Walmart factory fat per day, and I wonder if supplementing with omega-3s would even put a dent in my 3:6 ratio.

Peter where is the number zero for the chicken-insulin response coming from? Is it really zero? I see 73% lower which is low but not exactly zero, but 27 is still lower than eggs so. Peter may I ask you (I've been wondering for a while), have you been successful at maintaining such a lean weight because of a high-fat diet, or have you always been on the thin side? I gather you don't count calories (but maybe euros?)

webster said...

oops didn't look close enough at the fat:protein ratio earlier. Actually the 1000 kj steak used was only 7.7g fat : 42g protein. The insulin AUC of beef was 7910 vs 9350 for fish. Still probably need to look out for the insulinogenic bovine somatotropin though.

webster said...

Maybe it is the palmitic acid. Peter could you please have a look at these and give us an interpretation?

Does protein and GNG make GSIS relevant here? Is there enough reason to start increasing our MUFAs vs SFAs?

Peter said...

Hi Webster,

The key phrases from the first two links were these:

"They are observed only at fatty acid/albumin ratios higher than those normally occurring in plasma"


"The results establish that in the rat, the high circulating concentration of FFA that accompanies food deprivation is a sine qua non for efficient GSIS when a fast is terminated"

The first says that this paper is not physiology (haven't checked the full text) and the second I really like. On a fast (or a high sat fat LC diet, they are metabolically very similar) you will have large amounts of palmitic acid in your muscles and they will be physiologically insulin resistant, to keep you alive. When glucose arrives suddenly, because your wife found a tuber above the permafrost, your body does not want hyperglycaemia while it settles down to clear the palmitate to allow glucose in to muscles, so under high palmitate you need a little extra insulin to to maintain normoglycaemia when carbs arrive. Modest increased insulin appears preferable to increased glucose. On a chronic basis palmitic acid appears to blunt the GSIS, ie once you are established in to starvation or high fat eating longer term your pancreas stops anticipating glucose, lowers insulin response to what glucose there is and allows blood glucose to drift up a little way. Adapting out of this state is probably what takes the three days of modest carb loading to get a normal OGTT if you are hard core LC/high sat fat.

Loading up on oleic acid is an experiment any of us can try but the suggestion from an old secondary intervention trial is not encouraging. Olive oil is as healthy as corn oil and both are worse than no intervention change to a diet which was associated with a first heart attack.

Re the chicken-insulin response, the graph is here

The insulin level after chicken may only be 73% of that after a HC Powerbar, but it's the same as it was during fasting at the start of the experiment, ie the insulin-time line is horizontal after chicken. BTW the fasting insulin is a bit on the high side in all of the subjects the experiment...

I've never been fat, I think I peaked at about 68kg but I was slim with a small pot belly I could never get rid of and ALWAYS hungry a few hours after eating. I was still able to put muscle on with hard exercise but it went when I stopped. I think you could suggest that I was a skinny but metabolically obese person. Certainly the need to eat NOW coupled with the intense soporific effect of carbs suggest I was hyperinsulinaemic with reactive hypoglycaemia. I never went overboard on sucrose through my 20s and 30s so probably missed damaging my liver too much.


Yes, I was definitely thinking along those lines but Art Sayers didn't think chondrocytes were listening to insulin too much. But I'm still suspicious that there is a link between arthritis and insulin resistance. I don't buy "wear and tear" for the cause of osteoarthritis...