Microvascular architecture and physiological fluctuations constrain the control of cerebral microcirculation.

Brain vasculature is a multiscale network that actively regulates cerebral blood flow to maintain homeostasis. A systematic understanding of how this network enables robust and precise flow control has been hindered by the lack of understanding of flow in networks, as opposed to single vessels. To address this gap at the conceptual level, we theoretically studied nonperturbative, network-level flow responses to hydrodynamic conductance changes in individual vessels. We show vasodilation can either increase or decrease flow in the neighboring branches, yet selectively positioning the "controller" in the inflow branch of diverging nodes guarantees downstream increases in flow, regardless of surrounding network topology. Moreover, the effect of an individual vasodilation is small, so coordinated vasodilation is essential for effective regulation. To validate and refine our theoretical analysis, we developed a computational framework to analyze individual blood cell motion captured by confocal light field microscopy. This approach enabled tracking over one million cell detections across a network of more than 3,000 interconnected branches, with 2 µm spatial and 14 ms temporal resolution. Network-based analysis uncovered significant flow fluctuations, exhibiting long-range anticorrelation in spatially separated segments. The prevalence of diverging nodes within three branches of penetrating arterioles suggests that ensheathing pericytes are optimally positioned for fine-scale flow regulation. Finally, we quantified a phase separation of blood serum and cells at diverging nodes. This revealed a stochastic partition ratio with a nonlinear dependence on local hemodynamics. Collectively, our work highlights principles of organization for the control of blood flow among the seemingly random connectivity of brain microvessels.