Researchers at the Mayo Clinic in Jacksonville have published a spatial map of gene activity in the temporal cortex of people with Lewy body dementia, reporting that some of the largest molecular disturbances lie outside the cortical spots that actually contain Lewy bodies. The study, by Yunjung Jin, Kai Chen, Alexander Q. Wixom and colleagues in the laboratory of Na Zhao, with Dennis W. Dickson and Owen A. Ross, appeared in Acta Neuropathologica on 26 January 2026 (volume 151, article 10).
The design was small and postmortem. Formalin-fixed temporal cortex from four autopsy-confirmed controls and six people with Lewy body dementia (four sporadic cases, two early-onset cases carrying a triplication of the SNCA gene) was transferred onto 10x Genomics Visium slides using CytAssist, with an equal split of APOE3 and APOE4 genotypes and minimal Alzheimer's co-pathology. Each tissue spot was assigned to a cortical layer (1 to 6) or to white matter by its gene signature, and annotated as containing a Lewy body, surrounding one, or free of them. Layer 5 stood out: it showed the highest SNCA expression, matched the layer where most Lewy bodies were counted, and carried the most pronounced synaptic and metabolic disruption, more so in APOE4 carriers. All the tissue came from the Brain Bank for Neurodegenerative Disorders at Mayo Clinic, Jacksonville.
The part that does not track with the aggregates is the point of interest. Reelin signalling, running from the RELN ligand to the LRP8 and VLDLR receptors, was reported as disrupted across cortical layers rather than confined to Lewy body positive spots; in the authors' analysis of cell-to-cell signalling, neuronal and lipid pathway activity was higher in Lewy-surrounding and Lewy-negative regions than in the spots holding the aggregates themselves. White matter, which does not carry cortical Lewy bodies, showed a separate problem again: disrupted myelination pathways, with more myelin debris and glial response in APOE4 carriers than in non-carriers. The authors backed the Reelin finding with protein work in separate brain series (16 controls against 16 cases by western blot, with larger APOE-stratified sets for further blotting and immunostaining) and with iPSC-derived cortical organoids carrying SNCA triplication, which is laboratory tissue and not a person.
The counterpoint is size and timing. Ten brains, one cortical region, one moment at the end of life cannot establish what came first: whether failing Reelin signalling helps aggregates form, or follows them. Visium spots hold several cells rather than one, so cell populations were inferred by deconvolution using single-nucleus sequencing from the same cases, not counted directly. Two of the six cases carry a rare genetic form, and the paper does not separate dementia with Lewy bodies from Parkinson's disease dementia clinically.