Cucurbitacin I inhibits cell motility by indirectly interfering with actin dynamics

Cucurbitacins are plant natural products that inhibit activation of the Janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3) pathway by an unknown mechanism. They are also known to cause changes in the organization of the actin cytoskeleton.

We show that cucurbitacin I potently inhibits the migration of Madin-Darby canine kidney (MDCK) cell sheets during wound closure, as well as the random motility of B16-F1 mouse melanoma cells, but has no effect on movement of Dictyostelium discoideum amoebae. Upon treatment of MDCK or B16-F1 cells with cucurbitacin I, there is a very rapid cessation of motility and gradual accumulation of filamentous actin aggregates. The cellular effect of the compound is similar to that observed when cells are treated with the actin filament-stabilizing agent jasplakinolide. However, we found that, unlike jasplakinolide or phallacidin, cucurbitacin I does not directly stabilize actin filaments. In in vitro actin depolymerization experiments, cucurbitacin I had no effect on the rate of actin filament disassembly at the nanomolar concentrations that inhibit cell migration. At elevated concentrations, the depolymerization rate was also unaffected, although there was a delay in the initiation of depolymerization. Therefore, cucurbitacin I targets some factor involved in cellular actin dynamics other than actin itself. Two candidate proteins that play roles in actin depolymerization are the actin-severing proteins cofilin and gelsolin. Cucurbitacin I possesses electrophilic reactivity that may lead to chemical modification of its target protein, as suggested by structure-activity relationship data. However, mass spectrometry revealed no evidence for modification of purified cofilin or gelsolin by cucurbitacin I.

Cucurbitacin I results in accumulation of actin filaments in cells by a unique indirect mechanism. Furthermore, the proximal target of cucurbitacin I relevant to cell migration is unlikely to be the same one involved in activation of the JAK2/STAT3 pathway.