Nuclear import, mediated in part by karyopherin-α (KPNA)/importin-α subtypes, regulates transcription factor access to the genome and determines cell fate. However, the cancer-specific changes of KPNA subtypes and the relevancy in cancer biology remain largely unknown. Here, we report that KPNA4, encoding karyopherin-α4 (KPNA4), is exclusively amplified and overexpressed in head and neck of squamous cell carcinoma (HNSCC). Depletion of KPNA4 attenuated nuclear localization signal-dependent transport activity and suppressed malignant phenotypes and induced epidermal differentiation. Mechanistically, KPNA4-mediated nuclear transport of Ras-responsive element-binding protein (RREB1), which sustains Ras/ERK pathway signaling through repressing miR-143/145 expression. Notably, MAPK signaling enhanced trafficking activity of KPNA4 via phosphorylation of KPNA4 at Ser60. These data reveal that KPNA4 establishes a feed-forward cascade that potentiates Ras/ERK signaling in HNSCC.

Summary Biomolecules may undergo liquid–liquid phase separation (LLPS) to spatiotemporally compartmentalize and regulate diverse biological processes. Because the number of tools to directly probe LLPS is limited (ie. FRAP, FRET, fluorescence microscopy, fluorescence anisotropy, circular dichroism, etc.), the physicochemical traits of phase-separated condensates remain largely elusive. Here, we introduce a light-switching dipyrene probe (Pyr-A) that forms monomers in either hydrophobic or viscous environments, and intramolecular excimers in aqueous solutions. By exploiting their distinct fluorescence emission spectra, we used fluorescent microscopic imaging to study phase-separated condensates formed by in vitro protein droplets and membraneless intracellular organelles (centrosomes). Ratiometric measurement of excimer and monomer fluorescence intensities showed that protein droplets became hydrophobic and viscous as their size increased. Moreover, centrosomes became hydrophobic and viscous during maturation. Our results show that Pyr-A is a valuable tool to characterize LLPS and enhance our understanding of phase separation underlying biological functions.

Epigenetic deregulation plays an essential role in colorectal cancer progression. Bromodomains are epigenetic “readers” of histone acetylation. Bromodomain-containing protein 4 (BRD4) plays a pivotal role in transcriptional regulation and is a feasible drug target in cancer cells. Disease-specific elevation of nucleoporin, a component of the nuclear pore complex (NPC), is a determinant of cancer malignancy, but BRD4-driven changes of NPC composition remain poorly understood. Here, we developed novel aminocyclopropenones and investigated their biological effects on cancer cell growth and BRD4 functions. Among 21 compounds developed here, we identified aminocyclopropenone 1n (ACP-1n) with the strongest inhibitory effects on the growth of the cancer cell line HCT116. ACP-1n blocked BRD4 functions by preventing its phase separation ability both in vitro and in vivo, attenuating the expression levels of BRD4-driven MYC. Notably, ACP-1n significantly reduced the nuclear size with concomitant suppression of the level of the NPC protein nucleoporin NUP210. Furthermore, NUP210 is in a BRD4-dependent manner and silencing of NUP210 was sufficient to decrease nucleus size and cellular growth. In conclusion, our findings highlighted an aminocyclopropenone compound as a novel therapeutic drug blocking BRD4 assembly, thereby preventing BRD4-driven oncogenic functions in cancer cells. This study facilitates the development of the next generation of effective and potent inhibitors of epigenetic bromodomains and extra-terminal (BET) protein family.