Macrophage-specific NF-κB activation dynamics can segregate inflammatory bowel disease patients

The heterogeneous nature of inflammatory bowel disease (IBD) presents challenges, particularly when choosing therapy. Activation of the NF-κB transcription factor is a highly-regulated, dynamic event in IBD pathogenesis. We expressed the human NF-κB/p65 subunit in blood-derived macrophages, using lentivirus. Confocal imaging of p65 activation revealed that a higher proportion of macrophages from Crohn’s patients responded to lipid-A compared to controls. In contrast, cells from ulcerative colitis (UC) patients exhibited a shorter duration of p65 nuclear localisation compared to healthy controls and Crohn’s donors. Using a similar lentivirus approach, NF-κB-regulated luciferase was expressed in patient macrophages, isolated from frozen peripheral blood mononuclear cell samples. Following activation, samples could be segregated into three clusters based on the NF-κB-regulated luciferase response. The majority of UC samples appeared in hypo-responsive cluster 1, with Crohn’s patients representing the majority of hyper-responsive cluster 3. A positive correlation was seen between NF-κB-induced luciferase activity and cytokine levels released to medium from stimulated macrophages, but not in serum or biopsy. Analysis of macrophage cytokine responses and patient metadata revealed a strong correlation between Crohn’s patients who smoked and hyper-activation of p65. These in vitro dynamic assays of NF-κB activation in blood-derived macrophages segregate IBD patients into groups with different phenotypes and therefore may help determine response to therapy. Significance statement This manuscript describes two dynamic assays of NF-κB activation in blood-derived macrophages that can segregate IBD patients into groups with different phenotypes. For the first time we introduce the use of dynamic measurements of a transcription factor activation as a method to stratify patients and we are confident that our approach will lead in future to early patient stratification and prediction of treatment outcome.