Network analysis of endogenous gene expression profiles after polyethyleneimine-mediated DNA delivery.

BACKGROUND: DNA delivery systems, which transport exogenous DNA to cells, have applications that include gene therapy, tissue engineering and medical devices. Although the cationic nonviral DNA carrier polyethyleneimine (PEI) has been widely studied, the molecular factors and pathways underlying PEI-mediated DNA transfer remain largely unknown, preventing the design of more efficient delivery systems. METHODS: HEK 293 T cells were treated with polyplexes formed with PEI and pEGFPLuc encoding for green fluorescent protein (GFP). Transfected cells expressing GFP were flow-separated from treated, untransfected cells. Gene expression profiles were obtained using Affymetrix HG-U133 2.0 microarrays and differentially expressed genes were identified using R/Bioconductor. Gene network analysis using EGAN (exploratory gene association network) bioinformatics tools was then used to find interaction among genes and enriched gene ontology (GO) terms related to transfection. Genes identified by this method were perturbed using pharmacologic activators or inhibitors to assess their effect on DNA transfer. RESULTS: Microarray analysis comparing transfected cells to untransfected cells revealed 215 genes to be differentially expressed, with the majority enriched to GO processes including metabolism, response to stimulus, cell cycle, biological regulation and cellular component organization or biogenesis pathways. Gene network analysis revealed a coordinated induction of RAP1A, SCG5, PGAP1, ATF3 and NEB genes implicated in cell stress, cell cycle and cytoskeletal processes. Altering pathways with pharmacologic agents confirmed the potential role of RAP1A, SCG5 and ATF3 in transfection. CONCLUSIONS: Microarray and gene network analyses of the sorted, transfected cell population can identify potential mediators of transfection, providing a basis for the design of improved delivery systems.