Computational design and application of endogenous promoters for transcriptionally targeted gene therapy for rheumatoid arthritis.

The promoter regions of genes that are differentially regulated in the synovial membrane during the course of rheumatoid arthritis (RA) represent attractive candidates for application in transcriptionally targeted gene therapy. In this study, we applied an unbiased computational approach to define proximal-promoters from a gene expression profiling study of murine experimental arthritis. Synovium expression profiles from progressing stages of collagen-induced arthritis (CIA) were classified into six distinct groups using k-means clustering. Using an algorithm based on local over-representation and comparative genomics, we identified putatively functional transcription factor-binding sites (TFBS) in TATA-dependent proximal-promoters. Applying a filter based on spacing between TATA box and transcription start site (TSS) combined with the presence of over-represented nuclear factor kappaB (NFkappaB), AP-1, or CCAAT/enhancer-binding protein beta (C/EBPbeta) sites, 382 candidate murine and human promoters were reduced to 66, corresponding to 45 genes. In vitro, 9 out of 10 computationally defined promoter regions conferred cytokine-inducible expression in murine cells and human synovial fibroblasts. Under these conditions, the serum amyloid A3 (Saa3) promoter showed the strongest transcriptional induction and strength. We applied this promoter for driving therapeutically efficacious levels of the interleukin-1 receptor antagonist (Il1rn) in a disease-regulated fashion. These results demonstrate the value of bioinformatics for guiding the selection of endogenous promoters for transcriptionally targeted gene therapy.