Abstract
Many pathologies of the brain are characterized by both devastating daily effects for patients and daunting challenges for treatment for physicians. The endothelial cells of the brain serve many important functions (in both normal physiology and in disease), including comprising the BBB and regulating blood flow and nutrient exchange, and maintaining cerebral homeostasis. Treatment of many pathologies of the brain could be improved markedly by the development of non-invasive therapeutic approaches that elicit robust, endothelial cell-selective, gene expression in specific regions of the brain. Focused ultrasound (FUS) in conjunction with gas-filled microbubbles (MBs) has emerged as a non-invasive modality for MR image-guided gene delivery to the brain.
In the first aim of this dissertation, we introduce a new MR image-guided FUS method that elicits endothelial-selective transfection of the cerebral vasculature (i.e. “sonoselective” transfection), without opening the BBB. We demonstrate that activating circulating, cationic plasmid-bearing, MBs with pulsed very low-pressure FUS facilitates sonoselective gene delivery to the endothelium without MRI-detectable disruption of the BBB. This approach permits targeted gene delivery to blood vessels and could be used to facilitate gene therapy for a variety of brain pathologies where BBB disruption is contraindicated.
Next, in the second aim of this thesis, we seek to better characterize the effects of FUS-mediated BBB opening, and the ways in which different peak-negative pressures (PNPs) of FUS affect transfected cell populations. Following plasmid delivery to and across the BBB with FUS, we use single cell RNA-sequencing to identify the different populations of transfected cells, which prove to be highly dependent on PNP. Pressure-dependent differential gene expression is observed for multiple cell types, with cell stress genes upregulated proportionally to PNP, independent of cell type. These results underscore how FUS may be tuned to bias transfection toward specific brain cell types in-vivo and predict how those cells will respond to transfection.
Although many signaling pathways in brain endothelial cells have been implicated in disease, optimal molecular targets for endothelial cell-based drug or gene therapy can be difficult to determine. To address this need, in the third aim of this dissertation, we develop a large-scale computational model of brain endothelial cell signaling, capable of identifying the most influential molecules for pharmaceutical targeting to promote therapeutic changes in the endothelial cell phenotype. We use the model to identify influential and sensitive nodes under different physiological or pathological contexts. We then identify nodes (or combinations of nodes) with the greatest influence over combinations of desired model outputs as potential druggable targets for these disease conditions.
