Concurrent regulation of LKB1 and CaMKK2 in the activation of AMPK in castrate-resistant prostate cancer by a well-defined polyherbal mixture with anticancer properties

Background: Zyflamend, a blend of herbal extracts, effectively inhibits tumor growth using preclinical models of castrate-resistant prostate cancer mediated in part by 5'-adenosine monophosphate-activated protein kinase (AMPK), a master energy sensor of the cell. Clinically, treatment with Zyflamend and/or metformin (activators of AMPK) had benefits in castrate-resistant prostate cancer patients who no longer responded to treatment. Two predominant upstream kinases are known to activate AMPK: liver kinase B1 (LKB1), a tumor suppressor, and calcium-calmodulin kinase kinase-2 (CaMKK2), a tumor promotor over-expressed in many cancers. The objective was to interrogate how Zyflamend activates AMPK by determining the roles of LKB1 and CaMKK2. Methods: AMPK activation was determined in CWR22Rv1 cells treated with a variety of inhibitors of LKB1 and CaMKK2 in the presence and absence of Zyflamend, and in LKB1-null HeLa cells that constitutively express CaMKK2, following transfection with wild type LKB1 or catalytically-dead mutants. Upstream regulation by Zyflamend of LKB1 and CaMKK2 was investigated targeting protein kinase C-zeta (PKC) and death-associated protein kinase (DAPK), respectively. Results: Zyflamend's activation of AMPK appears to be LKB1 dependent, while simultaneously inhibiting CaMKK2 activity. Zyflamend failed to rescue the activation of AMPK in the presence of pharmacological and molecular inhibitors of LKB1, an effect not observed in the presence of inhibitors of CaMKK2. Using LKB1-null and catalytically-dead LKB1-transfected HeLa cells that constitutively express CaMKK2, ionomycin (activator of CaMKK2) increased phosphorylation of AMPK, but Zyflamend only had an effect in cells transfected with wild type LKB1. Zyflamend appears to inhibit CaMKK2 by DAPK-mediated phosphorylation of CaMKK2 at Ser511, an effect prevented by a DAPK inhibitor. Alternatively, Zyflamend mediates LKB1 activation via increased phosphorylation of PKCζ, where it induced translocation of PKCζ and LKB1 to their respective active compartments in HeLa cells following treatment. Altering the catalytic activity of LKB1 did not alter this translocation. Discussion: Zyflamend's activation of AMPK is mediated by LKB1, possibly via PKCζ, but independent of CaMKK2 by a mechanism that appears to involve DAPK. Conclusions: Therefore, this is the first evidence that natural products simultaneously and antithetically regulate upstream kinases, known to be involved in cancer, via the activation of AMPK.