Fibroblast-like synoviocytes (FLS) play a critical role in the pathogenesis of rheumatoid arthritis (RA). Chronic inflammation induces transcriptomic and epigenetic modifications that imparts a persistent catabolic phenotype to the FLS, despite their dissociation from the inflammatory environment. We analyzed high throughput gene expression and chromatin accessibility data from human and mouse FLS from our and other studies available on public repositories, with the goal of identifying the persistently reprogrammed signaling pathways driven by chronic inflammation. We found that the gene expression changes induced by short-term tumor necrosis factor-alpha (TNF) treatment were largely sustained in the FLS exposed to chronic inflammation. These changes that included both activation and repression of gene expression, were accompanied by the remodeling of chromatin accessibility. The sustained activated genes (SAGs) included established proinflammatory signaling components known to act at multiple levels of NF-kappaB, STAT and AP-1 signaling cascades. Interestingly, the sustained repressed genes (SRGs) included critical mediators and targets of the BMP signaling pathway. We thus identified sustained repression of BMP signaling as a unique constituent of the long-term inflammatory memory induced by chronic inflammation. We postulate that simultaneous targeting of these activated and repressed signaling pathways may be necessary to combat RA persistence. on a Hi-Seq 2500 System (University of Georgia Genomics Core) to generate data in FastQ file format. In order to study chromatin accessibility changes due to chronic inflammation, we downloaded FastQ files from a DNase I hypersensitivity sequencing (DNase-seq) experiment of wild-type FLS and inflamed FLS from inflammatory arthritis mouse model (Table S1). FastQ files from the ATAC-seq and DNase-seq experiments were analyzed using the Strand NGS sequence analysis pipeline first by aligning the sequences to the mouse genome, mm10 build, followed by removal of duplicate reads.Peak calling and gene annotation were performed by MACS2 algorithm using a padding distance of 5 kb from the transcription start sites of genes. Pathway analysis. Ingenuity Pathway Analysis tool was used to predict the canonical regulatory pathways and diseases and functions associated with the SAGs and SRGs (38). Functional protein-protein analysis was performed using STRING protein-protein interaction plugin on Cytoscape, an open source software platform for visualizing complex networks pathway analysis tool (39, 40). Open-source tool, ChEA3 that uses ChIP-seq experiments from the ENCODE database was used to predict transcription factors associations with SAG and SRGs (17).