Human mesenchymal stem cells (hMSCs) are multipotent stem cells that can be derived from multiple sources within the adult body. Because of their unique roles in immune regulation and wound healing, hMSCs are under much investigation for therapeutic applications involving tissue engineering, treating protein deficiencies, and cancer treatment. All of these applications would be greatly aided by the introduction of exogenous DNA to genetically guide stem cell differentiation, induce production of therapeutic proteins, or to secrete anti-tumorigenic proteins. Non-viral gene delivery, although considered much safer than viral genetic modification, remains especially inefficient in hMSCs, with efficiencies as low as 1% of cells expressing transgene. Previous studies within our lab have shown up to 13-fold increases in transgene expression by hMSCs when cells were primed by treatment with a glucocorticoid (GC) prior to transfection with DNA complexes. Using BSA-Cortisol, a protein-conjugated GC incapable of crossing the cell membrane, we have been able to confirm that this effect is mediated by binding of the GC to the cytoslic GC receptor, which through unknown mechanisms inreases total nuclear plasmid translocation.
The promising priming effects of the ubiquitously prescribed GCs on transfection efficiency of hMSCs has also led us to search for other clinically approved pharmacological agents that might have similar enhancements to gene delivery. We began this search by performing a high-throughput, high-content screen of the 727 compounds of the NIH Clinical Collection for priming effects on the transfection efficiency of HEK293T cells by both polyethyleneimine /DNA and chitosan/DNA complexes, identifying several compounds for both formulations whose priming significantly increased or decreased transfection relative to vehicle controls, including Diazepam, which was shown to increase transfection, and Epigallocatechin gallate, which was shown to decrease transfection. The screen was performed in a 96 well plate format, with each well of the compounds under test assessed for effects on transfection efficiency, cell proliferation, and cytotoxicity, utilizing high-content fluorescence microscopy imaging. With these data, in combination with annotations of each compound’s effects from literature and databases, we are able to hypothesize the mechanisms by which observed increases or decreases to transfection are being achieved, including the induction of mitosis, reduction of cytotoxicity, or by exclusion, a compound specific interaction with the mechanisms of transfection. In addition, we employ bio/chem-informatics data mining and clustering approaches to correlate the data from this screen to our previous microarray studies, which compiled the gene expression profiles of successful and unsuccessful transfection, as well as existing databases and models of drug, protein, and DNA interactions, to improve our general understanding of how genomic and proteomic targets can be potentially modulated to improve non-viral gene delivery. This combinatorial screening, informatics, and drug re-purposing approach, which began with the non-obvious identification of glucocorticoid transfection priming effects, has been fruitful in identifying compounds that: illuminate pharmacologically tunable mechanisms in transfection, may enable the efficient nonviral transfection of various stem cell types, and could potentially be instrumental as adjuvants that boost transfection efficiency of nonviral gene delivery therapies in vivo to clinically relevant levels.