Adaptive modulation of brain hemodynamics across stereotyped running episodes

Abstract

During locomotion, theta and gamma rhythms are essential to ensure timely communication between brain structures. However, their metabolic cost and contribution to neuroimaging signals remain elusive. To finely characterize neurovascular interactions during locomotion, we simultaneously recorded mesoscale brain hemodynamics using functional ultrasound (fUS) and local field potentials (LFP) in numerous brain structures of freely-running overtrained rats. Locomotion events were reliably followed by a surge in blood flow in a sequence involving the retrosplenial cortex, dorsal thalamus, dentate gyrus and CA regions successively, with delays ranging from 0.8 to 1.6 seconds after peak speed. Conversely, primary motor cortex was suppressed and subsequently recruited during reward uptake. Surprisingly, brain hemodynamics were strongly modulated across trials within the same recording session; cortical blood flow sharply decreased after 10–20 runs, while hippocampal responses strongly and linearly increased, particularly in the CA regions. This effect occurred while running speed and theta activity remained constant and was accompanied by an increase in the power of hippocampal, but not cortical, high-frequency oscillations (100–150 Hz). Our findings reveal distinct vascular subnetworks modulated across fast and slow timescales and suggest strong hemodynamic adaptation, despite the repetition of a stereotyped behavior.

Impaired neural replay of inferred relationships in schizophrenia

“…Neuroimaging studies in patients with early psychosis report abnormal DMN functional connectivity ( Whitfield-Gabrieli et al., 2009 ) and hippocampal hypermetabolism ( McHugo et al., 2019 ; Schobel et al., 2013 ) at rest. There is suggestive evidence the latter may relate to high-frequency (ripple) oscillations in hippocampus ( Bergel et al., 2020 ). These findings are intriguing, given the hypothesized role of offline hippocampal/DMN activity in imaginative (compositional) cognition ( Barron et al., 2020 , 2013 ; Buckner, 2010 ; Zeidman and Maguire, 2016 ), which we and others speculate may play a role in pathologies of belief seen in psychosis ( Buckner, 2010 ; Lewis et al., 2018 ; Suh et al., 2013 ).…”

Behavioral and physiological monitoring for awake neurovascular coupling experiments: a how-to guide

“…2(d) – 2(f) ]. In some brain areas, the increases in neural activity is accompanied by increase in blood flow, 8 , 40 , 41 , 51 , 56 but decrease in blood flow in other cortical regions. 8 , 41 In somatosensory areas, the changes induced by locomotion and movement are comparable to those generated by sensory stimulation.…”

“… 9 Locomotion drives increases in blood flow and volume in the cerebellum, 56 somatosensory 8 , 41 , 51 and visual cortex, 8 , 41 and hippocampus. 40 , 57 These increase in flow are not driven by increase in heart rate or blood pressure, 8 , 51 as they are blocked when local spiking activity is blocked. 7 – 9 Even short movements (twitches or brief postural adjustments) can drive robust hemodynamic signals [ Fig.…”

“…The behavioral and arousal-liked changes can be much larger than any stimulus-evoked changes. Furthermore, as movement and sleep drive bilaterally symmetric signals that show substantially higher correlations than during rest, 36 , 37 , 40 , 41 they will also contribute to “resting state” functional connectivity measures. Because of this, variations in the amount of a behavior will influence functional connectivity measures, and putative difference in connections across individuals or imaging sessions may simply due to differences in time asleep or moving.…”

Neural correlates of blood flow measured by ultrasound

“…Perhaps the apparently noisy fUSI signals reflect these structured fluctuations in neural activity. Indeed, fUSI signals approximately resemble simultaneously recorded local field potentials (Aydin et al, 2020;Bergel et al, 2018;Bergel et al, 2020;Sieu et al, 2015), which in turn reflect local neuronal firing (Buzsáki et al, 2012;Katzner et al, 2009). Moreover, it is not clear whether the neural component of fUSI signals reflect neuronal spiking through a simple linear relationship and if this relationship differs across brains and brain regions.…”