Correlative Ultrastructural Mapping of Lewy Pathology Reveals Regional Diversity in Parkinsons and Dementia with Lewy bodies

Lewy body diseases, including Parkinsons disease (PD) and dementia with Lewy bodies (DLB), are defined by neuronal accumulation of misfolded α-synuclein (α-Syn), yet the ultrastructural diversity of these inclusions across brain regions and disease contexts remains unclear. Here, we applied large-scale correlative light and electron microscopy (CLEM) to map α-Syn pathology across cortical regions (entorhinal cortex, ENT; anterior cingulate cortex, AC; hippocampal CA2 region) and substantia nigra (SN) in clinically and pathologically confirmed PD and DLB donors. We identified pronounced regional heterogeneity in Lewy pathology, with cortical inclusions showing diverse maturation stages at the ultrastructural level, ranging from low-density fibrils interspersed with organelles to highly compact fibrillar inclusions. In the SN of DLB donors, we observed the full range of classical nigral LB morphologies previously described in PD. We additionally characterized diverse neuritic α-Syn pathologies in DLB and identified a distinct population of electron-dense, degenerating, α-Syn-positive cortical neurons not previously reported. Importantly, we found no significant difference in LB ultrastructure between PD and DLB in either cortical or nigral pathology. In contrast, quantitative analysis of >10,000 mitochondria revealed disease- and region-specific signatures of altered mitochondrial homeostasis. Specifically, the density of mitochondria increased in both PD and DLB, but mitochondrial swelling occurred exclusively in PD and was confined to the SN. These findings indicate that LB ultrastructure alone does not distinguish PD from DLB; instead, mitochondrial morphology provides a more sensitive readout of disease identity and regional susceptibility. Overall, we provide a high-resolution framework for human Lewy pathology in PD and DLB, revealing that ultrastructural responses to α-Syn pathology are driven primarily by neuronal identity and regional vulnerability. Our results highlight the need for disease- and region-specific models that capture human phenotypes to advance mechanistic understanding and therapeutic targeting of synucleinopathies.