Cell type-specific effects of ICMS on brain cells: new approaches integrating spatial transcriptomics, modeling, and behavior - SUMMARY Penetrating microelectrode arrays can electrically stimulate cortex at high spatial resolution, offering the promise of addressing a wide range of neurological conditions. However, optimizing stimulation protocols, and predicting therapeutic and off-target effects, remain major challenges in the field. The present guidelines for safe cortical stimulation are based on decades old studies, and intracortical microstimulation (ICMS) can induce unexpected effects including plasticity, desensitization, glial activation, and changes in hemodynamics. Understanding genetic correlates of these phenomena can provide critical guidance for the design of long-term cortical neural interfaces. The goals of this proposal are to: (1) modernize the biomarkers used for stimulation safety, (2) reveal cell type-specific mechanisms of action of ICMS, and (3) answer fundamental questions about the relationship between electrical stimulation, behavioral outputs, electrophysiology, and quantified biomarkers (gene and protein-level). Further, we will deliver a new computational model to predict stimulation effects, which extends beyond a simple binary damage/no-damage threshold. The overarching hypothesis guiding this proposal is that the high-throughput, spatially precise, and comprehensive data sets provided by spatial transcriptomics will enable identification of new biomarkers to guide the design of electrode arrays and stimulation paradigms for cortical neural interfaces. Spatial transcriptomics will be used to identify differentially expressed (DE) genes associated within the region local to electrodes delivering distinct stimulation paradigms, and gene ontology will be used to map changes in expression to known pathways reflecting cell death and stress. We will employ multiple assays to capture both whole-transcriptome effects on a multicellular spatial scale, as well as single- and sub-cellular effects using a focused gene panel. Genes associated with neural damage will be benchmarked based on their expression level relative to traditional immunohistochemical markers within the same post-mortem brain slice. Gene expression quantified by spatial transcriptomics will be overlaid with physiological responses to provide insights into desensitization of neural circuits when subject to repetitive electrical stimulation. Low discrepancy sequences will be integrated into our experimental design strategy to efficiently and effectively survey the entire multi-dimensional stimulation parameter space while minimizing the time, cost, and numbers of animal subjects needed. Genes indicative of important cellular processes including hypoxia, metabolic stress, apoptosis, and necrosis will be assessed in individual neuronal and glial subtypes. This proposal will greatly expand upon the basic science understanding of the effects of electrical stimulation on brain cells, while producing a new quantitative model to guide the development of safer and more effective cortical interfaces.
