FMRI: Biological Basis and Experiment Design Lecture 3 Cell metabolism Vascular architecture Blood flow regulation Harrison, Harel et al., Cerebral Cortex.

fMRI: Biological Basis and Experiment DesignLecture 3

• Cell metabolism• Vascular architecture• Blood flow regulation

Harrison, Harel et al., Cerebral Cortex 12:225 (2002)

Oxidative vs. anaerobic metabolism

http://personal.nbnet.nb.ca/trevgall/biology/

Non-oxidative (glycolysis)

TCA

Nucleus

mitochondrion

Oxidative (16 times more ATP)

glc glc

pyr

lac

Fox and Raichle

• Surprise finding suggests that neuronal activity elicits anaerobic metabolism

Fox and Raichle, 1986: CBF >> CMRO2

CBF CMRO2 OEF

The Magistretti Hypothesis

• Astrocytes anaerobically metabolize glucose to lactate• Neurons aerobically metabolize lactate/pyruvate

Magistretti (2000) Brain Research 886:108

Capillary pre-capillary arteriole endothelium

endothelium

sm muscle

Neurons and astrocytes are cells

Astrocyte

Neuron

2 LAC + 2 ATP

TCA

glucose

2 LAC + 2 ATP

TCA

glucose

32 ATP 32 ATP

Capillary pre-capillary arteriole endothelium

endothelium

sm muscle

Magistretti hypothesis: an explanation for Fox and Raichle

Astrocyte

Neuron

2 LAC + 2 ATP

TCA

glucose

2 LAC + 2 ATP

TCA

glucose

32 ATP

Metabolism in astrocytes and neurons

Pellerin: put back in an arrow that went missing (too much disagreement about what role lactate plays for neurons)Attwell & Laughlin

(2001). JCBFM 21: 1133-1145.

Continued debate about whether (approximately) stoichiometric coupling indicates that glucose uptake is driven by glutamate cycling

Continued debate about compartmentalization of oxidative and non-oxidative metabolism in neurons and glia

Evidence for compartmentalization of metabolism

• Kasischke, K. A., Vishwasrao, H. D., Fisher, P. J., Zipfel, W. R. & Webb, W. W. (2004). Neural activity triggers neuronal oxidative metabolism followed by astrocytic glycolysis. Science, 305, 99-103.

Mintun, Vlassenko, Rundle, Raichle (2004). Increased lactate/pyruvate ratio augments blood flow in physiologically activated human brain. PNAS, 101 (2), 659-664.

Brains are not muscles

Pediatric patient (with fungal infection of liver)Adult (showing scar tissue following hernia repair)

18-FDG PET images from Abouzied et al. (2005). J. Nuc. Med. Tech. 33(3):145

Capillary pre-capillary arteriole endothelium

endothelium

sm muscle

Neurovascular coupling: why energy budgets and oxidative metabolism matter

Astrocyte: Inc Ca++, uptake of glutamate --> (release of NO, EET), increased glucose metabolism (non-oxydative)?

Interneuron - inc Ca++even w/o spikes - release of NO, EETs … --> dilation - release of NPY, SOM(?) --> contstriction - inc. glc metabolism?

Neuron - inc Ca++ when spiking - release of NO, EETs … - inc. glc metabolism (oxidative)?

propagation of dilatory signals

autoregulation

Harrison, Harel et al., Cerebral Cortex 12:225 (2002)

Harrison, Harel et al., Cerebral Cortex 12:225 (2002)

100m

50m

On the scale of a voxel

• Blood is supplied to and drained from the cortex by the pial network

– ~100 – 500 micron diameter

• ~half the blood volume is in intracortical veins and arteries (2% gray matter vol.)

– ~10 – 50 micron diameter– diameter depends on depth

• ~half the blood volume is in the capillary network (2% gray matter vol.)

– ~8 micron diameter– density correlates with neural demand

• White matter is supplied by transcortical arteries and veins

Human temporal cortex

Reina de la Torre et al (1998) Anatomical Record

251:87

375 m

The Plumbers and the Electricians

• There is no such thing as constant flow– Pulse– Vasculature is highly responsive; can autoregulate

• The vascular network is not a fixed entity– Flow can switch directions in small vessels and capillaries– Capillaries can grow to match metabolic demand

• Bottom-up regulation is more practical than top-down

5m

Balloon Model, Part I: CBF and CBV

• CBF = cerebral blood flow– increased CBF increases signal strength

• CBV = cerebral blood volume– increased venous blood volume decreases signal strength

Fout(t)Fin(t)

Filling the balloon

Fout(t)Fin(t)

( ))()(1)(

0

tFtFdt

tdvoutin −=

τ ⎥⎦⎤

⎢⎣⎡ −−≅ α

ττ 1

)()()()( tvtFtFtF inv

oinout

where τ0 is mean transit time through balloon, resting stateτv is mean transit time through expanded balloonv(t) is volume of balloon