Functional neuroimaging, such as fMRI, is based on coupling neuronal activity and accompanying changes in cerebral blood flow (CBF) and metabolism. However, the relationship between CBF and events at the level of the penetrating arterioles and capillaries is not well established. Recent findings suggest an active role of capillaries in CBF control, and pericytes on capillaries may be major regulators of CBF and initiators of functional imaging signals. Here, using two-photon microscopy of brains in living mice, we demonstrate that stimulation-evoked increases in synaptic activity in the mouse somatosensory cortex evokes capillary dilation starting mostly at the first- or second-order capillary, propagating upstream and downstream at 5-20 µm/s. Therefore, our data support an active role of pericytes in cerebrovascular control. The gliotransmitter ATP applied to first- and second-order capillaries by micropipette puffing induced dilation, followed by constriction, which also propagated at 5-20 µm/s. ATP-induced capillary constriction was blocked by purinergic P2 receptors. Thus, conducted vascular responses in capillaries may be a previously unidentified modulator of cerebrovascular function and functional neuroimaging signals.

Cortical spreading depression (CSD) is associated with release of arachidonic acid, impaired neurovascular coupling, and reduced cerebral blood flow (CBF), caused by cortical vasoconstriction. We tested the hypothesis that the released arachidonic acid is metabolized by the cytochrome P450 enzyme to produce the vasoconstrictor 20-hydroxyeicosatetraenoic acid (20-HETE), and that this mechanism explains cortical vasoconstriction and vascular dysfunction after CSD. CSD was induced in the frontal cortex of rats and the cortical electrical activity and local field potentials recorded by glass microelectrodes, CBF by laser Doppler flowmetry, and tissue oxygen tension (tpO 2 ) using polarographic microelectrodes. 20-HETE synthesis was measured in parallel experiments in cortical brain slices exposed to CSD. We used the specific inhibitor HET0016 (N-hydroxy-NЈ-(4-n-butyl-2-methylphenyl)formamidine) to block 20-HETE synthesis. CSD increased 20-HETE synthesis in brain slices for 120 min, and the time course of the increase in 20-HETE paralleled the reduction in CBF after CSD in vivo. HET0016 blocked the CSD-induced increase in 20-HETE synthesis and ameliorated the persistent reduction in CBF, but not the impaired neurovascular coupling after CSD. These findings suggest that CSD-induced increments in 20-HETE cause the reduction in CBF after CSD and that the attenuation of stimulation-induced CBF responses after CSD has a different mechanism. We suggest that blockade of 20-HETE synthesis may be clinically relevant to ameliorate reduced CBF in patients with migraine and acute brain cortex injuries.

Cerebral blood flow (CBF) is regulated by the activity of neurons and astrocytes. Understanding how these cells control activity-dependent increases in CBF is crucial to interpreting functional neuroimaging signals. The relative importance of neurons and astrocytes is debated, as are the functional implications of fast Ca changes in astrocytes versus neurons. Here, we used two-photon microscopy to assess Ca changes in neuropil, astrocyte processes, and astrocyte end-feet in response to whisker pad stimulation in mice. We also developed a pixel-based analysis to improve the detection of rapid Ca signals in the subcellular compartments of astrocytes. Fast Ca responses were observed using both chemical and genetically encoded Ca indicators in astrocyte end-feet prior to dilation of arterioles and capillaries. A low dose of the NMDA receptor antagonist (5R,10s)-(+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine-hydrogen-maleate (MK801) attenuated fast Ca responses in the neuropil and astrocyte processes, but not in astrocyte end-feet, and the evoked CBF response was preserved. In addition, a low dose of 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP), an agonist for the extrasynaptic GABA receptor (GABA R), increased CBF responses and the fast Ca response in astrocyte end-feet but did not affect Ca responses in astrocyte processes and neuropil. These results suggest that fast Ca increases in the neuropil and astrocyte processes are not necessary for an evoked CBF response. In contrast, as local fast Ca responses in astrocyte end-feet are unaffected by MK801 but increase via GABA R-dependent mechanisms that also increased CBF responses, we hypothesize that the fast Ca increases in end-feet adjust CBF during synaptic activity.

Effective treatments of neurodegenerative diseases require drugs to be actively transported across the blood-brain barrier (BBB). However, nanoparticle drug carriers explored for this purpose show negligible brain uptake, and the lack of basic understanding of nanoparticle-BBB interactions underlies many translational failures. Here, using two-photon microscopy in mice, we characterize the receptor-mediated transcytosis of nanoparticles at all steps of delivery to the brain in vivo. We show that transferrin receptor-targeted liposome nanoparticles are sequestered by the endothelium at capillaries and venules, but not at arterioles. The nanoparticles move unobstructed within endothelium, but transcytosis-mediated brain entry occurs mainly at post-capillary venules, and is negligible in capillaries. The vascular location of nanoparticle brain entry corresponds to the presence of perivascular space, which facilitates nanoparticle movement after transcytosis. Thus, post-capillary venules are the point-of-least resistance at the BBB, and compared to capillaries, provide a more feasible route for nanoparticle drug carriers into the brain.

Rises in local neural activity trigger local increases of cerebral blood flow, which is essential to match local energy demands. However, the specific location of microvascular flow control is incompletely understood. Here, we used two-photon microscopy to observe brain microvasculature in vivo. Small spatial movement of a three-dimensional (3D) vasculature makes it challenging to precisely measure vessel diameter at a single x–y plane. To overcome this problem, we carried out four-dimensional (x–y–z–t) imaging of brain microvessels during exposure to vasoactive molecules in order to constrain the impact of brain movements on the recordings. We demonstrate that rises in synaptic activity, acetylcholine, nitric oxide, cyclic guanosine monophosphate, ATP-sensitive potassium channels, and endothelin-1 exert far greater effects on brain precapillary sphincters and first-order capillaries than on penetrating arterioles or downstream capillaries, but with similar kinetics. The high level of responsiveness at precapillary sphincters and first-order capillaries was matched by a higher level of α-smooth muscle actin in pericytes as compared to penetrating arterioles and downstream capillaries. Mathematical modeling based on 3D vasculature reconstruction showed that precapillary sphincters predominantly regulate capillary blood flow and pressure as compared to penetrating arterioles and downstream capillaries. Our results confirm a key role for precapillary sphincters and pericytes on first-order capillaries as sensors and effectors of endothelium- or brain-derived vascular signals.