Is it really as simple at that? I got a tweet from a reader yesterday pointing me to an article in the LA Times. The article was covering a study from the University of South Florida on whether caffeine, and more specifically, coffee, can stave off Alzheimer's disease. The reader was skeptical, and so am I, seems a bit too good to be true, with all this talk of "unidentified ingredients" in the coffee, etc, etc. Sure, it's coffee, and I DO love me some coffee. And any excuse to drink it is going to be just fine with me. But I'm not sure we want to put out feelings of faith and hope in the great brown bean prematurely.
So let's go on through this paper, and through a little bit on Alzheimer's disease, and we'll see what comes out the other end. Hope you've got your coffee ready!
(Source, licensed under Creative Commons)
Cao et al. "Caffeine Synergizes with Another Coffee Component to Increase Plasma GCSF: Linkage to Cognitive Benefits in Alzheimer's Mice" Journal of Alzheimer's Disease, 2011.
We're going to start with Alzheimer's disease. Very basically, it's incurable, degenerative, and terminal. The most common symptoms are things like confusion, mood swings, and memory loss, and the disease is known for its massive loss of neurons in the brain. There is also buildup of things called amyloid plaques and neurofibrillary tangles, both of which are proteins that end up clumping together. The plaques accumulate outside, and the tangles happen inside the neurons. This actually happens in a lot of older patients, but those with Alzheimer's Disease have it way worse. And of course, we have no idea what causes it still, and there's a lot of research going on to find out. In the meantime, scientists like the ones in this article start looking for preventatives, things that people who don't have Alzheimer's tend to do much more or less than those who suffer from the disease. By finding out what these things are, we might be able to help prevent some cases, and by finding out how these things WORK, we might be able to find out more about the disease itself.
One of the first possible preventatives that popped up in the literature was caffeine. Caffeine has a lot of both known and unknown properties, and it's in really common use. And it turns out that several labs have shown decreases in Alzheimer's Disease risk and in Alzheimer's symptoms in animal models using caffeine.
Problems with the previous studies: for the ones done in humans, how do we know it's the coffee consumption specifically? Do people who drink a lot of coffee have, say, more intellectually stressful, and thus possibly neuroprotective jobs than people who don't? Do people who have a tendency to LIKE coffee maybe have some other sort of protection? Being humans, we are TERRIBLE at controlling the variables in our lives. There's also the inherant problem of asking people with a neurodegenerative disorder which affects MEMORY about their coffee drinking habits throughout their lives. You're going to need an animal model here.
But why coffee SPECIFICALLY? After all, caffeine is caffeine no matter the source. Well, one lab showed some neuroprotection in flies with decaf coffee. The lab that wrote this paper has shown previously that caffeine may be able to suppress the formation of beta amyloid, leading to fewer amyloid plaques. But they note that caffeine and coffee are two very different animals. Coffee contains a lot of things like antioxidants (which I believe, but I'm not so sure of the benefits from antioxidants taken orally. It sounds awfully nice, but I'd like to see the research on it first), as well as other, as yet unidentified compounds. So the authors wanted to compare the effects of caffeine and COFFEE on cognitive impairment.
(Side Note: I'm of course having my AM coffee while I write this, and now staring somewhat suspiciously into the depths. WTF is IN there...?)
So the authors took a mouse model of Alzheimer's Disease, known as the ABPPsw mice (The B is actually a beta but I haven't figured out how to do that one on here yet). Believe it or not, but we actually have a whole pile of Alzheimer's Disease mouse models now, of various types. This one is an early-onset model that you can order from Jackson. The mice get beta-amyloid plaque development around 7 months of age, and so can be useful for looking at the disease's development over time. I do want to note, though, that these are not a perfect model. They do display cognitive impairment like Alzheimer's, and they have a bunch of other similarities, but it just isn't the human disease. That's ok though, you've gotta start somewhere.
The first thing they did was look at levels of inflammatory cytokines, which are markers released from neurons and glia to help control the immune system. High levels of inflammatory cytokines are characteristic of cell damage and have been shown in Alzheimer's Disease. So they took a bunch of mutant Alzheimer's mice (we're going to call them AD mice rather than ABPPsw), and gave them injections of concentrated coffee, decaf coffee, caffeine, or saline.
(Click to embiggen)
The topmost picture here is the AD model mice and their cytokine levels following a single injection of coffee, decaf, etc. The bottom is the controls. You can see that both models had major increases in cytokine levels following the coffee SPECIFICALLY. Not the caffeine, not the decaf.
This one here is a little more complex. They didn't find any difference in beta-amyloid (the amyloid that primarily makes up the amyloid plaques characteristic of Alzheimer's) in the mouse model following different injections of coffee, decaf, caffeine, etc. But if they looked ONLY at the mice with high levels of blood caffeine (not whether it was in coffee or not) they saw a decrease in beta-amyloid.
Here's the graph where we get to the meat of the matter, cognitive behavior. They tested for cognitive impairment using two radial arm mazes, A and B. The radial arm maze is a maze designed like the spokes of a wheel. At the end of each spoke could be a food prize. The mouse has 8 possible options for the prize, and has to find it. They take great care to ensure the mouse can't SMELL where the prize is, so he has to try all the spokes and find which one, and then REMEMBER that for the next trial. In this case, they did this in two mazes, so the mice had to remember one, learn the other, and remember the first one again, tasks which are very similar to those used to test Alzheimer's Disease.
(Source, via wikipedia)
In the graph above you can see how the AD mouse model did compared to the normal controls (which aren't shown the graph, the data is presented as percent of the control's error behavior). They showed that AD-mice given COFFEE (this time orally, twice a week in a dose about of about 2.5 "cups" per mouse per day) made fewer errors in the behavioral task than those given decaf.
From all of this, they conclude that coffee, the combination of caffeine and some UNKNOWN THING in the coffee, is reducing the cognitive impairment in the mice and doing so via an increased elevation of inflammatory cytokines, specifically GCSF, which is shown in the top two graphs and known to be reduced in early stage Alzheimer's.
You'll have to excuse me, but my skepticism is prickling. And I shall tell you why.
They got their biggest elevations in inflammatory cytokines when they gave a single injection of coffee, both in the disease model and the controls. But I'm not sure about this. It's i.p., and that bypasses the gut. I'd try to get the mice to drink it myself, figure if you sweeten it and cream it up enough they should take it. Also make it cold, like a mouse caramel machiatto (whipped cream option, now offered in double-venti size!). Not only that, it's a single injection, they didn't look at the long term effects on cytokines following this. And they DIDN'T look at the effects of chronic oral administration of coffee on the inflammatory cytokines in the mice. Do the effects of a single injection match the effects of repeated oral administration? I kind of doubt it. I mean, you are injecting COFFEE. The body might not put up a fight for injected caffeine in saline, but concentrated coffee? Somehow I think a mouse's immune system might be a little...annoyed.
Secondly, in the behavioral tests, we're only seeing coffee vs decaf (though they did do oral administration this time). Where's the caffeine alone? Does caffeine alone result in fewer errors like coffee does? I want to see that before I call it. Also, they only gave the coffee twice week in the oral long term study. And it was only 2.5 "cups" per mouse per day. But the best effects in the ACUTE injection study were seen at the highest caffeine doses...did the mice in the behavioral study even get caffeine doses this high?
Finally...an unknown compound? Caffeine is interacting with an UNKNOWN COMPOUND?
I'm not convinced.
Here's what I'd like to see. Take some models. Take some controls. Given them little mousey machiattos, decaf, caffeine, or saline, every day, testing their blood levels for caffeine, every day, for however long deemed necessary (2 weeks? 4 weeks?). Look at the cytokines before, during, and after the administration. Test the mice on behavior and see what you get.
And then, if you've got an interaction, and coffee worked differently and better than caffeine alone...SHOW ME THE MONEY. Show me the COMPOUND. What is interacting here? Start adding the various chemicals in coffee to the caffeine alone. What potentiates the responses? What reduces them? What does nothing? Because I'm not going to believe it til I see it.
So, according to the Twitter question I received yesterday: is the link between COFFEE specifically and Alzheimer's weak? So far, you bet it is, in fact I'll say I think it's basically non-existant. The link between caffeine and prevention of Alzheimer's...less so. But any reason to NOT drink your coffee? I don't think so. :) *sssiiiipppp*
Cao C, Wang L, Lin X, Mamcarz M, Zhang C, Bai G, Nong J, Sussman S, & Arendash G (2011). Caffeine Synergizes with Another Coffee Component to Increase Plasma GCSF: Linkage to Cognitive Benefits in Alzheimer's Mice. Journal of Alzheimer's disease : JAD, 25 (2), 323-35 PMID: 21422521