Fig. 1

Flow chart of the study. After donor inclusion, post-mortem in situ 3D T1 and multi-shell diffusion MRI images were collected and (pre)processed: the LC was segmented [44] and the LC tracts to the anterior cingulate cortex (ACC, blue tract), the dorsolateral prefrontal cortex (DLPFC, purple tract), the primary motor cortex (M1, green tract) and the hippocampus (yellow tract) were reconstructed, deriving the FA and the MD of the LC and its tracts. Autopsy and brain dissection were performed after the in situ MRI scans. Brain tissue blocks were formalin-fixated for 4 weeks, then dissected and paraffin-embedded. Next, 20-µm-thick sections were processed for immunohistochemistry for dopamine-beta hydroxylase (DBH), phosphorylated Ser129 α-synuclein (pSer129-αsyn), phosphorylated-tau (p-tau) and amyloid-β (Aβ), and imaged using a whole-slide scanner. Immunoreactivity in regions of interest were analyzed using Qupath, deriving the LC-noradrenergic cell density and fiber load, Lewy body (LB) density and Lewy neurite (LN) load. The group comparisons between AD, PD and controls were investigated with general linear models and the association between MRI and pathology markers was investigated with linear regression models, including age, sex and post-mortem delay as covariates. Abbreviations: DWI, diffusion-weighted tensor imaging; LC, locus coeruleus; NA, noradrenergic; IHC, immunohistochemistry; FA, fractional anisotropy; MD, mean diffusivity; LB, Lewy body; LN, Lewy neurite; PMD, post-mortem delay