Comments on: How brains work, how brain imaging works: Astrocytes

By: bsci

bsci — Mon, 23 Jun 2008 09:54:48 +0000

A lot of this is recent. This all builds decades old literature on blood flow and metabolism, but the focus on exactly how neural activity changes blood flow wasn’t as interesting a question compared to how neurons fire until the explosion of PET, fMRI, and optical imaging which all require a better understanding of this link. PET started its rapid growth in the early 80’s and fMRI in the early 90’s.I can probably dig up some older ones, or you can using the references of the above references, but our knowledge has changed so rapidly in this area, that much of that stuff really isn’t up-to-date anymore.

By: WotWot

WotWot — Mon, 23 Jun 2008 00:33:48 +0000

Very much appreciated. Exactly what I was looking for.Feel free to throw in a bunch more refs. No rush. I’ll keep an eye on this thread for another couple of weeks (ahh, the wonders of bookmarking).One thing I notice about those refs is that (with one exception) they are all relatively recent. Is that because you are deliberately picking recent ones to be more up to date, or because most work on this stuff has only been done relatively recently?Thanks again.

By: bsci

bsci — Sun, 22 Jun 2008 21:44:15 +0000

WotWot,There’s a lot of stuff on this.Attwell & Laughlin 2001, J Cereb Blood Flow Metabolism 21(10) p1133 is a good general article on the energy budget of cells in the brain. (Note that it’s not a completely solved model and you’ll be able to find papers that reference this that disagree with their findings. One thing that makes these types of measures hard is that the ratios of different types of cells vary across species so the energy budgets might also vary.Pellerin & Magistretti 1994 PNAS 91(22) talks about astrocytes and vasodilationChatton & Pellerin 2003 PNAS 100(21)Attwell & Iadecola 2002 Trends Neurosci 25(12) talk about some Ca2+ effectsOther ionic controls of vasodilation are discussed in Lauritzen 2005 Nat Review Neurosci 6(1)Buerk & Ances 2003 Neuroimage 18(1)Other discussion of astrocytes are:Zonta, Angulo 2003 Nat Neurosci 6(1)Mulligan & MacVicar 2004 Nature 431(7005)Peppiatt & Attwell 2004 Nature 431(7005) contrasts the models of the above two papersVolterra & Meldolesi 2005 Nat Review Neurosci 6(8)Patel, de Graff et al 2004 J Cereb Blood Flow Metab 24(9)Parri & Crunelli 2003 Nat Neurosci 6(1)I can write more, but since I was away for the weekend and don’t know if you’re still reading this thread, I figured I’d wait for a reply before putting in any more work.

By: WotWot

WotWot — Fri, 20 Jun 2008 22:36:21 +0000

From the metabolism front, astrocytes have been shown to dominate much of the energy usage in the brain since they are heavily involved in the active transport of glutamate and other molecules across membranes. They also have direct involvement in causing vasodilation through Ca2+ levels and other factors.Posted by: bsciVery interesting. Can you recommend some papers on this stuff?Thanks

By: Ryan Morehead

Ryan Morehead — Fri, 20 Jun 2008 14:18:44 +0000

I meant to link it: http://www.nature.com/nature/journal/v453/n7197/abs/nature06976.html

By: Ryan Morehead

Ryan Morehead — Fri, 20 Jun 2008 14:17:31 +0000

I meant to link it: http://www.nature.com/nature/journal/v453/n7197/abs/nature06976.html

By: Ryan Morehead

Ryan Morehead — Fri, 20 Jun 2008 13:50:22 +0000

There’s also an excellent (I think) article in Nature by Nikos Logothetis titled “What we can do and what we cannot do with fMRI.”It goes into detail on what higher and lower BOLD levels do and don’t mean.

By: bsci

bsci — Fri, 20 Jun 2008 10:29:54 +0000

caynazzo. You’re not quite right. fMRI signal changes are dominated my the levels of deoxyhemoglobin in the blood. This level is affected by two factors. First, the more metabolism (i.e. the more energy being used in a region) the more oxygen is used and the higher the deoxyhemoglobin levels. Second, the faster the blood flows to a region (more vasodilation) the higher the oxyhemoglobin levels and thus the lower the deoxyhemoglobin levels.From the metabolism front, astrocytes have been shown to dominate much of the energy usage in the brain since they are heavily involved in the active transport of glutamate and other molecules across membranes. They also have direct involvement in causing vasodilation through Ca2+ levels and other factors.As for the hurdles of fMRI, background vs. signal is a challenge, but, like all science, a well designed study can answer questions within the limits of the measurement method.

By: caynazzo

caynazzo — Fri, 20 Jun 2008 10:20:55 +0000

Correct me if I’m wrong, but technically fMRIs measure blood indirectly by measuring spikes in oxygen levels, a function of metabolism and not necessarily tans-neuronal electric activity. In talking with cognitive psychologists, one of the biggest hurdles to interpreting fMRI data is deciding what counts as background and what counts as signal. Interesting post though.

By: bsci

bsci — Fri, 20 Jun 2008 10:07:11 +0000

The research linking astrocytes to control blood flow has been active and widely accepted for over a decade. (In fact that’s the first sentence of their abstract). In fact, many molecular aspects of the mechanism linking neural activity to vasodilation using astrocytes is already known.Astrocytes aren’t the only thing that links neural activity to blood flow, but they are a key factor. This is not the unique aspect of your linked article.The unique part is showing a unique way to image astrocyte activity to get spatial patterns of astrocytes and to use this method to demonstrate that glutamate transport is a major contributor to astrocytes and blood flow changes (something that has been shown using different methods before)