Activation of protein kinase A (PKA) signaling mitigates the antiproliferative and antiinvasive effects of alpha-difluoromethylornithine in breast cancer cells.

We have shown that alpha-difluoromethylornithine (DFMO), an inhibitor of ornithine decarboxylase, the first and rate-limiting enzyme in polyamine synthesis, has significant antiproliferative and antiinvasive effects in breast cancer cells. We have also reported that these antitumor effects are associated with activation of multiple signaling pathways, including STAT-3, STAT-1, Jun-N-Terminal kinase (JNK), and Mitogen activated protein kinase (MAPK), the latter being found to mediate its antiinvasive action in MDA-MB-435 cells. The present experiments were designed to test the effect of DFMO on the protein kinase A (PKA) pathway and determine its biological significance. We found that DFMO administration (1 mM) to MDA-MB-435 breast cancer cells significantly increased cAMP response element (CRE)-binding protein (CREB) phosphorylation as well as the transactivation of pCRE-luc, a CREB-dependent promoter activated by PKA. To determine the significance of this biochemical effect of DFMO, we used the PKA inhibitor H89 which, as expected, suppressed in a dose-dependent manner (1 and 10 microM) basal and DFMO-induced CREB phosphorylation in our system. Administration of H89 alone was able to suppress proliferation of MDA-MB-435 cells when used at a concentration (10 microM) shown to completely block basal CREB phosphorylation. At concentrations of 0.5 and 1 muM, H89 treatment, while having no antiproliferative effect of its own, potentiated in a dose-dependent fashion the growth inhibitory action of a suboptimal concentration of DFMO (0.01 mM). Ten micromoles of H89 reduced invasiveness of MDA-MB-435 cells in matrigel by approximately 40% (an effect similar to that of 1 mM DFMO). The combination treatment further reduced invasiveness by approximately 80% (P < 0.01 versus the individual treatments). H89 treatment (10 microM) partially reduced DFMO-induced phosphorylation of STAT-3 but not that of STAT-1, Extracellular regulated kinase (ERK), and JNK. In conclusion, our results indicate that PKA signaling exerts proproliferative and proinvasive effects in our experimental system. Therefore, its activation by DFMO represents a compensatory mechanism which should be blocked in order to maximize the antitumor action of the drug. Our data are also consistent with the notion that STAT-3 activation by DFMO is at least in part mediated through the PKA pathway.