Coupled aging of cyto- and myeloarchitectonic atlas-informed gray and white matter structural properties

A key aspect of brain aging that remains poorly understood is its high regional heterogeneity and heterochronicity. A better understanding of how the structural organization of the brain shapes aging trajectories is needed. Neuroimaging tissue “types” are often collected and analyzed as separate acquisitions, an approach that cannot provide a holistic view of age-related change of the related portions of the neurons (cell bodies and axons). Because neuroimaging can only assess indirect features at the gross macrostructural level, incorporating post-mortem histological information may aid in better understanding of structural aging gradients. Longitudinal design, coupling of gray and white matter (GM, WM) properties, and a biologically informed approach to organizing neural properties are needed. Thus, we tested aging of the regional coupling between GM (cortical thickness, surface area, volume) and WM (fractional anisotropy, mean, axial, and radial diffusivities) structural metrics using linear mixed effects modeling in 102 healthy adults aged 20-94 years old, scanned on two occasions over a four-year period. The association between age-related within-person change in GM morphometry and the diffusion properties of the directly neighboring portion of white matter were assessed, capturing both aspects of neuronal health in one model. Additionally, we parcellated the brain utilizing the histological-staining informed von Economo-Koskinas atlas to consider regional cyto- and myelo-architecture. Results demonstrate several gradients of coupled association in the age-related decline of neighboring white and gray matter. Most notably, gradients of coupling along the heteromodal association to sensory axis were found for several areas (e.g., anterior frontal and lateral temporal cortices, vs pre- and post-central gyrus, occipital, and limbic areas), in line with heterochronicity and retrogenesis theories of aging. Further effort to bridge across data and measurement scales will enhance understanding of the mechanisms of the aging brain.