Editors' ChoicePharmacology

Kinase inhibitors that block bromodomain interactions

See allHide authors and affiliations

Science Signaling  25 Oct 2016:
Vol. 9, Issue 451, pp. ec248
DOI: 10.1126/scisignal.aal2473

Among the mitogen-activated protein kinases (MAPKs), extracellular signal regulated kinase 5 (ERK5) is unique in having a large noncatalytic C-terminal domain in addition to its kinase domain. Compounds with ERK5 kinase inhibiting activity, such as XMD8-92, are candidates for anti-inflammatory and antitumor therapies. Lin et al. found that ERK5-inhibiting small molecules that inhibit cellular proliferation and inflammatory signaling exert these effects independently of their inhibition of ERK5 catalytic activity. In an initial screen of XMD8-92 derivatives for kinase inhibition in cell lysates, four new compounds specifically inhibited ERK5 activity, but only two of these exhibited anti-inflammatory activity in a cell culture–based assay, suggesting that ERK5 kinase inhibition was not responsible for this anti-inflammatory activity. Many kinase inhibitors also inhibit the binding of bromodomains (BRDs), which are not present in ERK5, to proteins with ε-N-acetylated lysine–containing motifs. Bromodomains are present in proteins that regulate transcription and the response to DNA damage. In an assay for compounds that inhibit BRD binding, only the compounds identified in the ERK5 inhibitor screen that exhibited anti-inflammatory activity inhibited the interaction between BRD4 and acetyl-lysine interaction. Furthermore, specific BRD-inhibiting compounds I-BET762 and JQ1 also exhibited anti-inflammatory activity in the same cell culture–based assay. BRD inhibition by either the BRD-specific or the dual BRD and ERK5 inhibitors suppressed interleukin-6 (IL-6) and IL-8 production in two different cell-based assays of anti-inflammatory activity. The ERK5 inhibitors without BRD-inhibiting ability were not effective in any of the assays for anti-inflammatory activity. However, siRNA knockdown of ERK5 in the cells reduced the induction of inflammatory cytokines, consistent with ERK5 indeed being involved in mediating inflammatory signals. However, the lack of an effect of the catalytic inhibitors suggested that noncatalytic functions of ERK5 may be involved. Specific ERK5 kinase inhibitors also failed to prevent proliferation in acute myeloid leukemia cell line MV4-11, whereas the dual ERK5 and BRD inhibitors and specific BRD inhibitors did. The two different classes of inhibitors also induced different transcriptional profiles. In addition to identifying a class of pharmacological compounds, dual ERK5 and BRD inhibitors, this study reveals the hazards of inferring the mechanism of drug action even when genetic knockdown results apparently agree with pharmacological assays. Screens to assay effects on protein interaction motifs and protein domains should be included in pharmacological profiling.

E. C. K. Lin, C. M. Amantea, T. K. Nomanbhoy, H. Weissig, J. Ishiyama, Y. Hu, S. Sidique, B. Li, J. W. Kozarich, and J. S. Rosenblum, ERK5 kinase activity is dispensable for cellular immune response and proliferation. Proc. Natl. Acad. Sci. U.S.A. 113, 11865–11870 (2016). [PubMed]

Stay Connected to Science Signaling