Dual platform spatial transcriptomics reveals parvalbumin interneuron subtype vulnerability in mouse models of Alzheimer's disease.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline and synaptic dysfunction. Among the earliest regions affected is the retrosplenial cortex (RSC), where parvalbumin-expressing (PV + ) interneurons are particularly susceptible to AD-related pathology. To understand the molecular alterations within these vulnerable neurons we employed a dual-platform spatial transcriptomics approach, integrating GeoMx Digital Spatial Profiler (DSP) and Xenium In Situ. We analyzed the transcriptomic profiles of PV+ and NeuN+ neurons in the RSC of female 5xFAD mice. We leveraged the individual strengths of each platform to generate a robust and comprehensive dataset. Using non-negative matrix factorization and k-means clustering, we identified disease-associated metagenes and examined their spatial distribution. Our analysis revealed distinct transcriptional subpopulations within PV+ interneurons, with specific metagenes differentially expressed in RSC. Dner, Gad1, and Pvalb exhibited significant down-regulation in TG mice, suggesting impairments in PV+ interneuron function and GABAergic signalling. Cross-validation between GeoMx DSP and Xenium In Situ as well as RNAscope and immunohistochemistry confirmed the reproducibility and robustness of these findings. This study provides insights into the heterogeneity and molecular vulnerabilities of PV+ interneurons in AD and demonstrates the power of integrating spatial transcriptomic platforms to uncover disease-associated neuronal subtypes and molecular markers.