Integrated in vivo multiomics analysis identifies p21-activated kinase signaling as a driver of colitis.
Jesse LyonsDouglas K BrubakerPhaedra C GhaziKatherine R BaldwinAmanda EdwardsMyriam BoukhaliSamantha Dale StrasserLucia Suarez-LopezYi-Jang LinVijay YajnikJoseph L KissilWilhelm HaasDouglas A LauffenburgerKevin M HaigisPublished in: Science signaling (2018)
Inflammatory bowel disease (IBD) is a chronic disorder of the gastrointestinal tract that has limited treatment options. To gain insight into the pathogenesis of chronic colonic inflammation (colitis), we performed a multiomics analysis that integrated RNA microarray, total protein mass spectrometry (MS), and phosphoprotein MS measurements from a mouse model of the disease. Because we collected all three types of data from individual samples, we tracked information flow from RNA to protein to phosphoprotein and identified signaling molecules that were coordinately or discordantly regulated and pathways that had complex regulation in vivo. For example, the genes encoding acute-phase proteins were expressed in the liver, but the proteins were detected by MS in the colon during inflammation. We also ascertained the types of data that best described particular facets of chronic inflammation. Using gene set enrichment analysis and trans-omics coexpression network analysis, we found that each data set provided a distinct viewpoint on the molecular pathogenesis of colitis. Combining human transcriptomic data with the mouse multiomics data implicated increased p21-activated kinase (Pak) signaling as a driver of colitis. Chemical inhibition of Pak1 and Pak2 with FRAX597 suppressed active colitis in mice. These studies provide translational insights into the mechanisms contributing to colitis and identify Pak as a potential therapeutic target in IBD.
Keyphrases
- mass spectrometry
- ulcerative colitis
- electronic health record
- network analysis
- big data
- oxidative stress
- multiple sclerosis
- mouse model
- ms ms
- genome wide
- endothelial cells
- liquid chromatography
- single cell
- healthcare
- type diabetes
- social media
- transcription factor
- machine learning
- binding protein
- protein protein
- genome wide identification
- insulin resistance
- gas chromatography
- tandem mass spectrometry