Friday, May 21, 2010

Alzheimer's and RAGE

Something I wrote in an email a while ago . . .

Advanced glycation endproducts (AGEs) are the endpoints of some complicated chemistry that occurs when simple sugars (glucose, fructose, etc.) react with proteins (and apparently fats too). They’re toxic for a variety of reasons, and trigger an inflammatory response via the receptor for advanced glycation endproducts, or RAGE.

It turns out that RAGE binds to a whole bunch of things, and amongst them is the amyloid beta peptide, which is implicated in the development of Alzheimer’s. Amyloid beta is apparently produced via neural activity. I can’t figure out if it has a function or is just a by-product. I suspect it has some function, because the body has a mechanism for achieving a balance in the central nervous system (CNS). One kind of receptor (LRP) causes active transport out of the CNS to the blood, while RAGE triggers transport from the blood to the CNS across the blood-brain barrier. More RAGEs means you’ll have more amyloid beta in your brain. I couldn’t verify this, but I would guess that insulin drives the formation of RAGE. It makes sense, as your body would be preparing for glycation damage (more AGEs) from increased blood sugar, whether the source was food or glucose released due to stress. And indeed, diabetics have higher concentrations of RAGE (as do the blood vessels in the brains of Alzheimer’s victims).

We learned today that stress actually increases amyloid beta production in the brain, via the action of corticotrophin releasing factor, or CRF. I got in contact with one of the authors of that study and he was nice enough to send me a reprint of the paper. It’s a pretty solid piece of research. Amongst other things, they showed that the more you stress mice, the more amyloid beta is produced. They could introduce CRF directly into the brain, and observe increased amyloid beta production. They could block the action of CRF, stress the mice, and see that less amyloid beta was produced. And finally they could directly block neural activity, and either stress the mice or introduce CRF, and again would see reduced amyloid beta. So it was a pretty solid case, albeit in mice. It would be surprising if humans turned out to be much different, though it’s certainly possible. CRF is released as part of the stress response. It is also released as a result of insulin-induced hypoglycemia, i.e. insulin goes up, blood sugar crashes, CRF pumps out.

One last piece of the puzzle: by itself, amyloid beta is soluble, and shouldn’t form solid plaques (or at least do so slowly). But test-tube experiments show that formation of solid “fibrillar aggregates” of amyloid beta are accelerated if you provide seeds of altered amyloid beta. And what’s one form of the alteration? Glycation damage from sugar.

So, less than surprisingly my hypothesis is that the route to Alzheimer’s mirrors that of heart disease. A high-carbohydrate diet leads to the following effects:
  1. Increase in density of receptors for advanced glycation endproducts, which leads to increased amyloid beta concentrations in the brain.
  2. Release of CRF, which increases production of amyloid beta in the brain.
  3. Damage to amyloid beta, which increases the formation rate of solid aggregates, which may be contributory toward the formation of the plaques associated with Alzheimer’s.
And of course, there’s the usual feedback between stress and diet: psychosocial stress makes you want to eat more carbohydrates, which makes you more stressed, etc.

5 comments:

Mrs. Ed said...

This backs up what I have been suspecting about stress. I have theory, something I call the "God Dammit Theory". That stress is so bad for you that God gave us a built in mechanism (nuture's programming) to warn us when we're bringing on ourselves. Thus, why it's called the "God Dammit Theory" and not the "Dammit Janet Theory".

Dave said...

@Mrs. Ed

An excellent book on the stress response is "Why Zebras Don't Get Ulcers". The punch-line is exactly as you note: the stress response is bad for you. We evolved it because in a natural environment, having a stress response (fight/flight/etc.) is better than getting eaten. The stress response is evolved to be relatively short-lived, something that you activate in a crisis, but which is usually suppressed. Stress as we experience it in modern life (like road rage) is pretty much all downside, since you get all the negative effects, but aren't actually utilizing the response to avoid a more dangerous threat.

As noted in the blog, your food can cause the stress response as well. See this paper for a great example of stress hormone release following high-glycemic meals:

http://pediatrics.aappublications.org/cgi/content/abstract/103/3/e26

LeonRover said...

Hi Dave

I saw your reference to Jayne's Probability text in the comments of Doc Eades' latest post. I was intrigued so I ended up downloading and reading the ‘free intro’. As you know this consists of the 1st 3 Chapters plus the Introduction. Really quite Cool, for those of us interested in Scientific Method and Statistics. In particular, the introduction discusses Safety and the Linear Response Model with and without Threshold.The particular example used there, artificial sweeteners, was of interest as I had been researching sucralose and wondering how close I might be to some 'long-term' damaging effect. I was taken with his notion of methodologically assuming some non-zero threshold and estimating that level by means series of time delayed studies. FYI, serendipitously, I had a few day’s earlier read about a Co-60 study mentioned in stan-heretic’s blog of May 21 ,2010 – Gamma radiation protects against cancer, in low doses? Cancer mortality reduction by 97% – huge! Is it true? Hormesis?. This impressed me even more, as I remember trying to allow for background radiation while doing particle counting from a radiative source while a physics' undergrad.

A blow to the Linear No Threshold Model, I trust.

Once I have gone thro' those first 3 chapters in detail, I may consider buying it. Why? While I had a little exposure to Bayesian Methods when in Statistics Grad School, I never came across Maximum Entropy procedures there, tho' I had earlier when studying Statistical Mechanics - Boltzmann, Bose-Einstein and Fermi-Dirac.

The only really new post-school idea that blew me away when I came across it was Chaos and Non-Linear Dynamics.

I appreciate finding out about this thro' your comment.

Anonymous said...

Nice post Dave. I wrote a lighter version for a "bigger" audience. http://bit.ly/cW3jpD VBR Hans

DancinPete said...

Hi Dave,
What are your thoughts on the fact that hypoglycemia from too few carbs can cause the release of stress hormones (cortisol, epinephrine etc) to increase blood sugar. How do you burn excess body fat without the negative aspects of the associated stress hormones?