Glymphatic Optimal Mass Transport with Lagrangian Workflow Reveals Advective and Diffusion Driven Solute Transport

Abstract

Acknowledgements: NIH R01AG048769, RF1 AG053991, R01AG057705 and Leducq Foundation (16/CVD/05) Author contributions:AT and HB conceived the study; AT and RE developed the rOMT algorithm and Lagrangian framework analysis; RE performed all rOMT processing; SN provided key suggestions for processing the Lagrangian analysis; HB performed all rOMT post-processing and designed figures with contributions from RE; HB performed kinetic analysis; SK, SS and XL performed all glymphatics experiments. SK executed all morphometric analysis including brain atlas segmentation; HL designed all pulse-sequences and other hardware for the MRI experiments and computational pipeline for the volumetric analysis. SS, FX, WVN performed the immunohistochemistry. SC performed qPCR analysis. YX performed the quantitative AQP4 analysis. HB, RE and AT wrote the manuscript. JW advised on the SHRSP animal model and cerebral small vessel disease. MN advised on the glymphatic system. All authors posed scientific questions, read and revised the manuscript. All authors edited and reviewed the paper. AbstractThe presence of advection in neuropil is contested and solute transport is claimed to occur by diffusion only. To address this controversy, we implemented a regularized version of the optimal mass transport (rOMT) problem, wherein the advection/diffusion equation is the only a priori assumption required. rOMT analysis with a Lagrangian perspective of glymphatic system (GS) transport revealed that solute speed was faster in cerebrospinal fluid (CSF) compared to grey and white matter. rOMT analysis also demonstrated 2-fold differences in regional particle speed within the brain parenchyma. Collectively, these results imply that advective transport dominates in CSF while diffusion and advection both contribute to transport in parenchyma. In rats with chronic hypertension, solute transport in perivascular spaces (PVS) and PVS-to-tissue transfer was slower compared to normotension. Thus, the analytical framework of rOMT provides novel insights in local variation and dynamics of GS transport that may have implications for neurodegenerative diseases.

Bulk flow of cerebrospinal fluid observed in periarterial spaces is not an artifact of injection

“…However, conservation of mass implies that fluid flowing through pial PVSs must continue through whatever regions are contiguous, and we hypothesize that Virchow-Robin spaces are contiguous. That hypothesis, which we hope to test directly in future work, is supported by data collected using several complementary methodologies in mice (Iliff et al, 2012;Mortensen et al, 2019;Koundal et al, 2020) and by recent MRI studies in humans (Ringstad et al, 2017(Ringstad et al, , 2018,…”

Diffuse white matter loss in a transgenic rat model of cerebral amyloid angiopathy

“…Another possibility for extensive perivascular Aβ accumulation in the thalamus may be related to patterns of glymphatic transport flow and waste drainage which has been shown to be less efficient in deep regions such as the thalamus. 68–70…”

Disparate volumetric fluid shifts across cerebral tissue compartments with two different anesthetics

“…The importance of the glymphatic and lymphatic systems for brain health is supported by studies demonstrating glymphatic clearance of amyloid beta (Aβ) [1] and tau [4] and declining waste clearance with aging [5][6][7]. Furthermore, glymphatic transport is reduced in rodent models of Alzheimer's disease (AD) [8], cerebral small vessel disease [9][10][11][12], and in humans with normal pressure hydrocephalus [13,14].…”

“…PDW images were first corrected for B1 inhomogeneity, followed by spatial registration and tissue segmentation into three tissue compartments: grey matter (GM), white matter (WM), and CSF. Spatial registration was performed using DARTEL image processing pipelines and custom-made tissue probability maps [10,42], and the spatially normalized images were subsequently smoothed by a Gaussian smoothing kernel of 0.6 mm. Voxel-wise statistical analysis was performed to identify local morphological differences using a paired t-test and statistical significance was reported at p < 0.05 after correcting for multiple comparisons via a false discovery rate (FDR) algorithm [43].…”