Tuesday, December 7, 2010

Your Brain Loves Cholesterol (Don't Go Too Low)

Sometimes this blog writes itself.  Today I was sitting around minding my business, spreading happiness and serenity, when I got an email from the Amazing Jamie Scott, who sent me a link to this paper: Diabetes and Insulin in Regulation of Brain Cholesterol Metabolism.  And then my receptionist handed me my mail, and the top story of this month's Psychiatric Times is "Statins, Cholesterol Depletion, and Mood Disorders:  What's the Link?"

I'm beginning to feel less small and alone in the world.

Let's start with the Psychiatric Times article.  Statins, as most biochem nerds will know, are pharmacologic inhibitors of HMG-CoA reductase, which is the key rate-limiting enzyme in the biosynthesis of cholesterol.  So our livers, doing their best to kill us off with heart disease, make cholesterol like mad fiends, while a statin will slow that pesky liver down, lowering serum cholesterol, and allowing us to live forever.*  Or something like that.

But throwing a monkey wrench into the cholesterol machinery has some... issues.  For one thing, it seems to ruin the binding and G-protein coupling (total random aside - at my medical school we had several professors of biochemistry who had Nobel Prizes to their name - among them Brown and Goldstein for their elucidation of the metabolism of cholesterol, and Gilman who discovered the G-protein.  And here they are, together at last on my heretical blog) to the serotonin IA receptors (1).  That's probably not the best thing to do - decreasing the ability of the serotonin IA receptors to work can lead to anxiety and irritability.  At the same time, there seem to be other changes in the actions of receptors in the context of "chronic cholesterol depletion" (I bet you would never find that phrase in 'Cardiology Times.').  As we know, low serum cholesterol is associated with violence, accidents, and suicide.

Now it is my pleasure to introduce Dr. James Lake, a psychiatrist and chair of the APA's Caucus on Complementary and Integrative Medicine (that's mainstream medicine talk for "woo."  I emailed one of my residency mentors about my blog a few months ago as I had some questions for him - he wrote back after reading some of the entries and said I was the "alternative Harvard Mental Health Letter" - I'm still not sure if that was a positive or negative comment).  Anyway, Dr. Lake seems to have drunk some of the same kool-aid that I have, as he recommends that depressed patients with elevated cholesterol aim not to go lower than a total cholesterol of 160.  How very reasonable!

He's also been hanging out with Dr. Beatrice Golomb, who has studied data from a few websites and run some surveys of her own.  In an analysis of 324 emails of people taking statins who were bothered enough to go out of their way to email "https://www.statineffects.com/" or "http://www.askapatient.com/," 30% reported mood changes such as depression, irritability, and anxiety.  When patients who complained about statin side effects were asked survey questions, 65% of 843 endorsed increased anxiety or irritability and 32% reported an increase in depressive symptoms.  Golomb published a case series of 6 patients who self-referred with irritability or short temper on statins (including "homicidal impulses, threats to others, and road rage") - in 100% of cases, stopping the statin cured the symptoms, and 4 of the 6 had renewal of the symptoms with a statin rechallenge. 

Granted, these are all people who complained of symptoms in the first place, so it is hardly a random sampling, and the case series could represent the nocebo effect.  But when I looked at the PDR for Crestor a few months ago (I can't seem to find it again on the internet, but if I do I will link it), I didn't find anything on irritability or anxiety, and depression was only mentioned briefly as an "aftermarket" side effect - meaning the crestor folks didn't find those to be side effects in their carefully controlled studies, but there are now some reports of depressed mood in the general public after release of the medicine.  To me that just doesn't quite add up to my own experience.  I've had several situations clinically where withdrawing the statin resulted in immediate improvement of anxiety, depression, and/or irritability for some treatment resistant patients.  Of course that could be nocebo too, but nocebo and placebo effects tend to wear off after about 3 months, and I've seen patients improve after 2 years, and the biological mechanisms seem plausible.  Well, definitive answers will wait for another day.

Statins improve mortality for middle-aged men who have known heart disease, have had a stroke, or have high levels of inflammatory markers.  If you don't meet those particular criteria, statins will give you no mortality benefit.

But let's not be so negative - a literature review involving statins and mental health "found no statistically significant effect" of low cholesterol on psychological well-being.  However, there may be a difference among the different statins.  Simvastatin can readily cross the blood brain barrier, whereas pravastatin really can't.  Golomb tested 1016 healthy men and women for 6 months with simvastatin, pravastatin, or placebo.  Those on simvastatin reported significantly worse sleep, and, if sleep was impaired, worsening aggression.  It was felt that statins that cross the blood brain barrier inhibited serotonin production.  Interesting.

Now onto Jamie's paper - which is a study of diabetic mice.  The researchers found that insulin-deficient diabetic mice had a reduction of a major regulator of cholesterol metabolism, leading to a reduction in brain cholesterol synthesis and lower synaptic cholesterol content (that's bad).   The decline in brain cholesterol production happened in cases of insulin depletion OR hyperglycemia in various mouse models of diabetes type I and II, but not in obese (but normoglycemic) or insulin resistant (but normoglycemic, meaning high levels of circulating insulin were needed to keep the blood sugars normal) mice.  Diabetes type I and II can both lead to CNS complications, including delirium and faster than "normal" decline in cognitive functioning.  Diabetics, as we know, have higher rates of depression and Alzheimers.  The brain contains 25% of the cholesterol in the body, and much of it is made right in the brain.  Therefore diabetes produces a "global suppression of the enzymes of cholesterol synthesis and their master transcriptional regulator, SREB-2 in the brain... [which alters] neuronal and physiological function."  A lot of this action occurs in the hypothalamus, which is a major point of control of the endocrine system, appetite, and energy balance. 

Sometimes it all comes together.  The bottom line?  Eat whole real food, not tons of sugar and linoleic acid or wheat.  Don't get diabetes if you can help it.  Don't let anyone or anything suck the cholesterol out of your brain.  Once things get out of whack, they can continue to be out of whack in all sorts of disastrous ways for quite a while.

* not really


  1. Emily, I don't consider you "alternative," and that's a compliment. You are in the time-honored mainstream of logical, scientific exploration. You're a pioneer, a little ahead of the pack.

  2. And this is exactly why I sent you the link. A mind immeasurably superior to my own was able to decipher all of that and turn it into something I can understand!

    Thanks! Oh, and congrats on getting a front page header on PaleoHacks!

  3. Another wonderful post! As a side note, I've seen Dr. Lake speak, and he's a brilliant and very kind man. You are in good company, Emily (and based on what I've read on your blog, so is he!) Keep up the fantastic work....

  4. Steve - thanks! Psychiatrists as a group, especially on the East Coast, are conservative and, to some extent, snarky (too much time spent with people, I guess ;), however, as we do tend to deal with the "whole" patient and lifestyle, I've found most psychiatrists I've talked to in person to be surprisingly open and interested in my ideas. The primary care doctors (having been burned too many times, perhaps, by studies of vitamins) tend to tell me that vitamin D is a fad, that low fat has to be the way to go, etc. I don't have a lot of contact with cardiologists (and though I make fun of them, they save lives every day!! I just wish they would release their death grip on the lipid hypothesis.)

    Jamie - my mind is hardly "immeasureably superior" to most, certainly not to your own - pretty much I'm just as smart as the latest research paper I'm reading... and I greatly appreciate the papers and links you pass along!

    Joyful - No one gets to be the chair of APA anything without being an exceptional clinician, so I'm certain Dr. Lake must be an amazing fellow.

  5. Oh great trying to get dinner on and I see this. It strongly reminds me of this post on my blog.


    This thing drove me around the bend and now, with this citation, it's back!

    If dinner is late, I'm blaming you guys.

  6. Michael - I need to take another long look at the paper and then think on it.

  7. Okay - so if you think of diabetes as a fat storage disease, you've got straight up type ii where you store fat too easily, then you have type I where immune attack on the beta cells means you can't store fat, then you've got type ii kta-prone where no one knows exactly what is going on, but you're more on the not able to store fat side than storing fat too easily. So if *somehow* you can manage to pack on some pounds, you can shift yourself to becoming more like a straight up type ii, which ironically enough is curative for your fat storage problem and will normalize blood sugars, because you're able to bypass the whatever-it-is fat storage problem by storing fat too easily as more like a type ii. That's my stab at you previous post paradox, Michael. It is a different issue than cholesterol making in the brain, and the wording of the paper makes me think it is low low insulin that is the problem - so type I diabetes or advanced type ii with beta cell burn out.

  8. Emily, this is why your blog got me stewing.
    For the past month or so I’ve searching around for an endocrine mechanism that can be suppressed by hyperglycemia. This is because by the sheerest of luck, I found a little trick that allows me to turn on my 1st phase insulin response and conversely, suppress it. I don’t know if this will work on anyone else but it does give me a window into this type of diabetes.
    I won’t go into all the gory details but if you wish to read more details about it. You can read it here.
    The 1st phase would get depressed by hyperglycemia. My readings would then match my early readings before I took insulin. (100 to 160)This led me to take another look at the idea that it was the loss of beta cells and the subsequent insulin insufficiency which was responsible for my blood sugar numbers.
    If, in the space of a week, I could reduce the numbers down to normal blood sugar numbers between 80 and 120, this is hard to ascribe to beta cell growth. It is known that the KPD’s who recover to near normal blood sugars have been shown to have c-peptide numbers that were low normal even though they had just went DKA. Also buried in the pile, I remember them challenging the body with drugs to see if they could get a glucagon response, again normal.
    What seems to be the case to me is that rather than system deterioration, what happens in the case of sudden onset type 2 diabetes is rather a metabolic derangement. This makes sort of sense when you think of how abrupt this comes on and also how quickly it can be resolved, especially by insulin. Nothing matches insulin in the resolution of KPD. It could be viewed as a intact system malfunctioning because some control element has been lost and the parts of the system no longer coordinate. This was my thinking anyway.
    I’ve been digging about looking for something that effects glucose metabolism that is sensitive to the effects of hyperglycemia, can be fixed by insulin and can be affected as well by a PGE2 inhibitor. This is why I showed such interest in Jamie’s article. This article seems to suggest that the hypothalamus is a good candidate. What’s your take?

  9. Michael, the hypothalamus is an obvious place to look, as one small thing can go wrong there and throw the whole metabolism out of whack. Something that also randomly occurs to me would be some sort of issue with the insulin receptors. I believe I read somewhere that they can be differentially spliced depending on the environment and maybe some signal goes wrong in type ii kta where some of the receptors get screwed up, or possibly there is an autoimmune attack that ebbs and flows on the receptors? It isn't bad enough to make them completely nonfunctional, but more insulin could therefore be nearly curative. I've only had a change to quickly scan Jamie's paper a couple of times - Ill look more closely at their discussion of insulin in the hypothalamus and see what I can make of it.

  10. Dear Emily, is it possible to change cholesterol levels in the brain with statins? This recent research says the brain is shield from external colesterol and produces all it needs. Please correct me if I'm wrong. Love your blogue!


  11. O Primitivo - most of the brain's cholesterol is made by the brain. In fact, the neurons seem to forgo making their own cholesterol because they need all the energy they can get keepi g the ions flowing - they outsource to the astrocytes - it is clear that some lipophilic statins (like simvastatin) readily cross the blood-brain barrier. The most study showing statins interfering with serotonin receptors was done in hamster ovary cells, however. Also, according to some autopsy studies I linked in a long-ago post, they found the free cholesterol types and amounts to be rather similar between the brain and the rest of the body, though e brain has a lot of cholesterol in the synapses and myelin also. I think putting all thAt info together would suggest that the brain and body aren't entirely separate when it comes to lipids - after all we spend a lot of time thinking about ApoE, too, which is how some bodily cholesterol gets to the brain.

  12. One more thing - one of the links you kindly left discusses how the overall turnover of cholesterol in the brain is very slow - this is true - most of the cholesterol is in the myelin and the turnover is slow, however, turnover in the synapses is quite fast.

  13. Thanks for these explanations. Here is a great cholesterol article: http://raypeat.com/articles/articles/cholesterol-longevity.shtml

  14. Although I knew that eggs are healthy. But now, I'm going to eat more eggs to reduce some of the burden of the liver and brain in making the essential cholesterol. Not sure if I'm right. Does eating more eggs give more cholesterol to the brain?

  15. Hi Emily, I remember the eggs are bad for you hype as a kid (I'm almost 70), and my Mom said, "How can eggs be bad for you when they make a whole chick? Nonsense!" So I ate eggs for breakfast with whole wheat toast for the rest of my life, and little did I know, the whole wheat toast wasn't such a good idea. Paleo (with fat from dairy) is working great for my Ironman training, and health in general. It's the old big lie phenomenon all over again. Great blog. Cheers, Kevin aka FitOldDog