Hoell, Zachariah J.

BET bromodomain inhibitors demonstrate significant promise as anti-inflammatory agents targeting a variety of inflammatory disorders. However, clinical data has demonstrated that the administration of non-selective BET bromodomain inhibitors has led to significant dose-limiting toxicity in clinical settings. More selective inhibitors of the second BET bromodomains, referred to as pan-D2 inhibitors, are better tolerated, however their efficacy varies significantly depending on disease indication. Here, we use three orally bioavailable inhibitors, 1–3, that are either selective for the first N-terminal bromodomain of BRD4 (BRD4-D1 selective), or pan-D1 biased, for assessing their cellular and in vivo efficacy and safety profile in two models of inflammatory liver disease. We compare their effects to known pan-D2 and pan-BET inhibitors. Our results show that pan-D1 biased inhibitor, 3, is as efficacious as the pan-BET inhibitor I-BET151 in reducing inflammation from LPS-induced and drug-induced toxicity, whereas pan-D2 inhibitors are less effective. BRD4-D1 selective inhibitors, 1 and 2, are also efficacious; however, new inhibitors with improved cellular target engagement will be necessary to more thoroughly assess their effects. Finally, BRD4-D1 selective inhibitors are better tolerated in a preclinical model of thrombocytopenia than pan-D1 biased inhibitors, while gastrointestinal (GI) toxicity may be a BRD4-driven effect. These results highlight the importance of assessing specific BET protein and bromodomain functions due to their varying roles in diverse disease models.