I think there is a mix up in the time scales we are talking about. First,
on the millisecond range when neurons fire they probably act
anaerobically (do you have a ref for this?), but as they recover their
resting potential via the na/k atpase pump (tens to hundreds of
millisecs), they work aerobically. this is when the oxygen
use peaks and thus deoxyHg peaks (by 1 sec after activity). This peak has
been documented for years spectroscopically (Frostig et al, 1990;
revisted by Malonek and Grinvald, 1996) and only recently by fMRI (Li??et
al, 1997). And if Magistretti (1996) is correct, then glial cells are the
ones working anaerobically AND using up local glucose. Neurons rely on
pyruvate shuttled over from the glial cells and predominantly oxygen from
the capillaries for metabolism.
This increased O2 consumption is followed by increased blood flow, the
positive BOLD signal. Unfortunately most fMRI and all PET studies lack the
spatial resolution to distinguish brain from arteries and veins. That is
why the spectroscopy data is more convincing as to what is happening.
On Sat, 6 Dec 1997, Ron Blue wrote:
|My understanding supports your comments but in a backward direction.
|Paradoxically the neurons use up energy and function anaerobically
|while working. Then the increase in oxygen levels occur after a thought
|to resupply the nerons. This results in a delayed picture
|of neuro activity but a relatively accurate one for functional magnetic
||>|ablab at usa.net wrote:
|>|: I've seen there are some physicists out there also. Can someone explain
|>|: how Functional Magnetic Resonance (hope it's correct) works? It should
|>|: be one of the methods used in the brain function scanning.
|>||>|...and it is.
|>||>|In a nutshell:
|>|Local increases in neural activity lead to
|>|local increases in local blood flow (outpacing demand for O2), leads to
|>Mike, What do you mean by this? My understanding of the situation now
|>(Malonek & Grinvald, 1996; and several others I don't have in front of me
|>right now) is that there is an initial increase in O2 demand, as seen
|>clearly by imaging spectroscopy studies, and now also by fMRI (with higher
|>temporal resolution than most previous work). So the "classic" BOLD signal
|>does indeed reflect local decreases in deoxyHg, but there is an initial
|>dip in the signal (with the proper temporal resolution) that reflects the
|>initial increase in deoxyHg.
|>|>Why do I bring this up??? Because this O2-demand signal is probably a
|>better spatial indicator of neuronal activity (i.e. from the capillary
|>beds) that is uncontaminated by blood vessel signals (i.e. arteries and
|>|>|>|local decreases in the concentration of deoxyHg, leads to
|>|more homogenous local magnetic field, leads to
|>|less T2* transverse spin dephasing, leads to
|>|increased MR signal return in T2*-weighted images
|>||>| __Mike Worden mworden at neurocog.lrdc.pitt.edu|>| o/ 630 LRDC University of Pittsburgh
|>| <\__,Pittsburgh, PA 15260 412 624-5279
|>| "> http://neurocog.lrdc.pitt.edu/~mworden|>| ` -climb
|>Department of Psychobiology, College of Medicine