ElectroGenOmics: Integrierte räumlich aufgelöste Elektrophysiologie und Transkriptomics zur Exploration von Tumor-Host Kommunikation bei Glioblastomen

Cancer neuroscience is emerging, emphasizing the interaction of cancer cells with their neuronal environment. Recent insights have redefined glioblastoma, transitioning from an isolated system to a bidirectional consortium integrated into neural circuits. However, technical obstacles have made integrating electrophysiology and transcriptomic data challenging. Here, we present ElectroGenOmics, a solution for spatially-integrated, multi-omic analysis merging array-based electrophysiology and transcriptomics.

Using patient-derived tumor cultures introduced into human cortical slices, we employed a multi-channel electrode array system. After recording for 2 hours, which involved electrical and pharmacological stimulation, the sections underwent PFA fixation and paraffin embedding followed by HE staining and spatial transcriptomic profiling using the 10X Visium kit. Data postprocessing was performed by space-ranger and a novel developed package "SPATAElectro", an R-based toolbox built upon the SPATA framework.

Quality checks on human cortical slices post-electrophysiological recordings showed high RNA integrity in 84% of the samples, with a DV>200 exceeding 45%. Image co-registration post-paraffin embedding demonstrated high accuracy for the multi-modal integration, enabling effective data overlay. we identified highly correlative transcriptomic modules that are associated with high tumor network activity and a high firing rate. These modules were linked to enriched genes, providing new insights into the molecular underpinnings of glioblastoma activity. By examining the spatial context, we explored the directionality of information transport in different zones of the tumor and adjacent brain tissue. Looking into the composition of the microenvironment in active malignant nodes, revealed how these areas interact with and influence their surroundings.

ElectroGenOmics overcomes previous multi-omic data integration challenges, offering a comprehensive view of the transcriptional context in relation to glioblastoma electrophysiological communication patterns.