SFN Neuroblogging: Got Type 2 Diabetes on the Brain

As some of my readers from WAY back (all two of you, hi guys!) may know, diabetes is one of Sci’s favorite things. It’s one of those things that, if she could start her entire little sciency life over, would be something she would heavily consider as a focus. Heck, there’s always another post-doc, right?
Anyway, you might think that diabetes would not be one of the things generally discussed at Society for Neuroscience meetings. But you would be wrong. The symptoms of diabetes, type I or II, stem from not enough insulin, whether that is because you don’t produce any (type I) or you don’t have enough and aren’t sensitive enough to what you have (type II). Insulin isn’t just limited to the gut, pancreas, and muscles, however. It’s also important in the brain. Normally, your brain is pretty responsive to blood levels of glucose, no matter what, because you want your brain to be the last thing to go when your blood sugar levels drop. But insulin still plays an important role, and insensitivity to insulin, like that seen with type II diabetes extends to the brain as well.
This study taught Sci a lot of things that she didn’t necessarily know. First, it taught her that insulin sensitivity is affected by free fatty acid levels. And it taught her that both of these together could have major effects on cognitive impairment. Suddenly the major increases in type II diabetes are looking a little more scary.
V. E. COTERO, E. C. MCNAY “Effect of intrahippocampal FAs with varied saturations on spatial memory in adult Sprague-Dawley rats”
Doesn’t sound like anything to do with type II diabetes, does it? You would be surprised. :)


Type II diabetes can be induced in rats using a high-fat diet-induced obesity model.
fat rat1.jpg
(Fat rats are exceptionally adorable)
The animals show all the symptoms of type II diabetes, and respond to treatments for type II diabetes just like humans do, making them a very good model indeed. What the scientists in this study wanted to look at, however, was not just the physical issues with this disease. They wanted to look at the cognitive effects. It’s not very well publicized, but type II diabetes is not only associated with insulin resistance and weight problems, it’s associated with cognitive declines. The question is, why?
In this study, the authors looked at the relationship between insulin resistance, fatty acids, and spatial memory in an area of the brain called the hippocampus (Greek for seahorse because it’s all curly-like), an area strongly associated with memory. They found before that a high fat diet, with a lot of free fatty acids hanging around, increases insulin-resistance in the hippocampus, and saturated fatty acids have been shown to be associated with aspects of cognitive decline. In particular, fatty acids can change neuronal activity in the hippocampus, which might be responsible for some of the issues they cause. So here, the authors injected fatty acid solutions, either saturated or unsaturated, into the rat hippocampus, and determined the behavioral effects on memory.
(Can we all take a moment to be totally mind-boggled that a single injection of something extracted from olive oil into your brain can completely change behavior?! Srsly…thinking about this too hard boggles Sci’s mind in a major way, and I DO stuff like this all the time!!!)
The authors found that the TYPE of fatty acid made a big difference in how well rats did in a series of mazes. If they used oleic acid in the hippocampus, an unsaturated fatty acid found in olive oil, the rats did BETTER in the maze. If they used palmitic acid, a saturated fatty acid from palm oil, the rats did significantly worse in the maze, indicating that increases in saturated fatty acids in the brain can impair memory function. And the palmitic acid inhibited many of the neurotransmitter changes associated with successful maze completion, suggesting that high fatty acids levels can impair maze performance by inhibiting neuronal function.
This could mean that insulin and saturated fatty acid levels could interact (as found in their previous work), and that the high saturated fatty acid levels could be partially responsible for some of the cognitive declines seen in people with type II diabetes. This could give us other options for treatment of type II, treating not just the endocrine aspects of the disease, but possibly allowing us to find ways to treat the cognitive effects. Also, it’s another great excuse to use more olive oil in your cooking. :)

11 Responses

  1. Ultra-cool experiment, but not nearly enough. What are some possibilities for mechanism?

  2. Interesting, it reminded me of a study published earlier this year which shows that insulin signalling can reduce the number of binding sites to Amyloid beta oligomers in hippocampal neurons in vitro, thus providing another possible explanation of the associaion between thye II diabetes and Alzheimer’s disease.
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2634809/?tool=pubmed
    It’s quite an interesting picture that’s developing!

  3. becca: fair point, but this is a pretty new phenomenon. I imagine they’ll get there.

  4. Wow – very interesting research! The complications from diabetes are indeed very scary – and this is yet another example. More reason to work harder on prevention strategies, particularly in educating the public and improving accessibility of healthy foods (which is a more pressing need for aboriginal groups in remote areas, whom I work with).

  5. Disclaimer: I’m McNay, EC :). Very cool to see Sci blog this! We were expecting to be further off in left field at SfN than turned out to be the case – not only is there a lot of interest in the T2DM-AD link (which was the topic of another poster from the lab), but there were a couple of folks starting to look at FA as a neural and cognitive modulator; we got several nice comments and suggestions. [I missed the DrugMonkey meetup, alas, but Chicago overall was a very good meeting, and I love the city.]
    Becca: the poster mentions a couple of possibilities, but let me offer several (!). First are metabolic effects, both direct and indirect. We showed a while back (McNay et al 2000, PNAS) that cognitive demands of difficult* memory tasks cause drainage of glucose within the hippocampus that are reversed by a dose of glucose that also enhances memory performance. More recently (presented various places but currently under review; positive comments in, top of my to-do list to get it back to the editor) we showed that not only (i) can performance also be enhanced by intrahippocampal insulin administration, but (ii) that blockade of _endogenous_ hippocampal insulin signalling causes marked cognitive impairment (this is a key finding: otherwise it could just be that the insulin story is one of pharmacology, but the fact that rats *require* insulin signalling to perform memory tasks optimally, combined with the fact that the dose of insulin that enhances performance is similar to that produced by a largeish meal**, strongly suggests that this pathway is physiologically relevant and important).
    We also showed that the hippocampus is metabolically responsive to insulin: very rapid increases in glucose metabolism. We and others (notably Larry Reagan’s group down in South Carolina) have shown that as in the periphery, insulin is able to cause translocation of the glucose transporter GluT4 to the cell surface within the hippocampus, which is likely the mechanism for increasing glucose metabolism and we suggest may directly lead to the memory boost.
    Metabolism is not the only possible mechanism, though. Vickie Cotero (the postdoc whose work this is, and the first author) helped demonstrate, in her graduate lab, that FAs can directly affect neuronal firing rates and excitability; this is also true of insulin, which affects both acute firing and also plasticity (Izumi et al did the best study of LTP and insulin). There are also possibilities around membrane composition and hence fluidity/synaptic effects, but my bet would be on one or both of the first two possibilities above. The poster does show, as Sci mentioned, that palmitic acid impaired not only performance but also neurotransmitter (ACh) release, and we know that ACh release is linked to both memory performance and glucose treatment (Mike Ragozzino’s work, and then Mark Stefani had a follow-up paper, both in Paul Gold’s lab [where I did my doctoral work]). Currently that’s being wrapped up, GABA and glutamate measurements being completed, and measurements of insulin signalling being performed (e.g. pAkt).
    [* 'difficult' varies by group, e.g. AD patients versus college students have very different demands placed by a MiniMental state exam]
    [** we measure this using in vivo microdialysis for insulin; also not yet published, but as far as I know unique to us. At SfN there were other posters looking at microdialysis for beta amyloid, which is something we have in yet another manuscript on my desk, but our sensitivity and results are thus far still better :)]
    Paul Browne: you hit on a key interest in the lab. OK, so, back to T2DM for a second. One of the reasons for doing Vickie’s study was our finding that a high-fat diet (where ‘high-fat’ is actually lower fat than the average US diet these days…), which makes about half of the rats*** who are fed it type-2 diabetic, as Sci mentions, causes marked cognitive impairment, and that’s accompanied by altered lipid composition in plasma etc. Not only are these animals (called diet-induced obese, or DIO) impaired at baseline, their response to insulin treatment is also markedly blunted with respect to both cognitive and metabolic effects: putting insulin into their hippocampi has no impact at all on local metabolism. So the insulin-insensitivity extends to the brain. OK, now we looked at amyloid: the DIO guys have higher total amyloid but *lower* soluble amyloid – there’s a very complex but fascinating story that came out after we collected these data, suggesting that the soluble form may not be harmful and may in fact be protective, but the agglomeration into oligomers is the problem – AND they do not respond to insulin by removing amyloid from the interstitial space as control animals do. That’s one finding in a study which is – yeah, I know – on my desk for writing up, but it looks solid.
    The other poster I mentioned above looked at the effects of direct intrahippocampal elevation of oligomeric beta amyloid: we see impaired cognition, lowered insulin signalling, and possibly impaired metabolism. The insulin-amyloid T2DM-AD story is a main focus of my lab right now (but this comment is way long enough already so I will not go into details), and it was neat to see these two initially disparate studies begin to converge****.
    [*** The other half of the rats adjust to the new diet and do not become obese, nor hyperinsulinemic; they make a great control group for the diet itself.]
    [**** Anyone looking for a good lab to do their doctoral work in? :)]
    OK, enough. But I’d be delighted to answer any further Qs.

  6. Thanks for the discussion Ewan, I’m curious about the fact that soluble amyloid levels are lower in the DIO rats than in the controls, if they are clearing amyloid more slowly than the controls because of insulin insensitivity I’d have expected both soluble and oligomeric forms to be higher. This seems to suggest that something is driving the self-association of the soluble monomers into oligomers in the DIO rats, but what?
    Forgive me if I’ve missed something, my knowledge of the Amyloid beta literature is a bit sketchy. I certainly look forward to reading your next papers.

  7. Paul -
    - there’s a deficit in exporting soluble amyloid from the neurons, linked to insulin insensitivity. My guess is that the prolonged higher intracellular concentration leads to agglomeration just due to increased likelihood of binding; there’s also a competition effect for breakdown via IDE in the hyperinsulinemic animals (same protease breaks down both insulin and amyloid, but has higher affinity for insulin; hence when lots of insulin around, amyloid is undigested). Complex story for sure, but I am not surprised at the lower soluble form in the DIO guys.

  8. Yay for answers!
    Ewan- I can understand from the first part of your answer that it’s extremely important to have proper insulin signaling for hippocampal metabolism and memory formation. The second part of your answer is closer to what I was wondering about- are unsaturated fatty acids signaling like insulin, while saturated fatty acids have a somewhat inverse function?
    (as an aside, has anyone done any comparative proteomics on the rats that develop diabetes on the high fat diet vs. those that don’t?)
    Isn’t it possible that instead of something driving self-association, there is an impairment of disassociation?
    What about the effects of hyperinsulinemia on autophagy inhibition? Wouldn’t that make it harder to degrade a multitude of malformed proteins, including things like amyloid beta?

  9. Becca -
    * Yes, Barry Levin (whose model this variety of DIO is, really) is doing the gene- and protein-level analyses of the DIO vs DR (diet-resistant) rats. One can now buy the two strains – i.e. guaranteed DIO or DR – but I don’t believe it’s yet known what the differences are.
    * I/we don’t know yet whether e.g. omega-3s actually activate insulin signalling paths; but we should by the end of the month (and ditto on saturated fat). Even money on that vs. some other explanation.
    * The cleavage and secretion of beta-amyloid is reasonably well-understood; there are several points (e.g. at glycogen-synthase-kinase 3) where insulin appears to have modulatory activity. I don’t know of any mechanism by which amyloid actually ever really gets out of agglomeration: there’s an equilibrium w/ monomers, but I think it’s pretty tilted kinetically toward staying in the oligomerised state; conversely, in the non-AD brain the level of oligomers is pretty low, so we don’t think that the dissociation step is normally involved.
    * Yes, it’s possible that other proteins – both tau and non-AD – might also get ‘caught up’ in the hyperinsulinemic effects; given that IDE has other targets, that’s almost certainly true in fact.
    [For another twist, consider that the hippocampus and olfactory bulb are two of the few brain areas rich in insulin receptor, and also the only two known to have adult neurogenesis. Which appears to be impaired in AD. That's another ten person-years of work right there :)]

  10. I hate to be the one asking a non-science question here, but I have a fat pet rat who isn’t very bright. What are the symptoms of Type II diabetes that she would display if she had it? What are the health risks for her if she has it?

  11. Olivia: sorry, I just noticed this comment! I hope you come back.
    Symptoms: the main symptom for type II diabetes that you want to look out for is excessive urine production. Rat will pee a LOT. That’s usually how people catch it (same in cats, and in humans for that matter). You can also take him in the to the vet and have him tested, but he’s probably ok unless he starts to get lethargic and pee a lot.
    Also, if you’re worried, you might consider a running wheel if you don’t already have one. Rats LOVE to run and will do it all night if you let them. Might run some of that fat off. :)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 35 other followers

%d bloggers like this: