Regulation of epithelial tight junction assembly by AMPK and GSK-3beta

AMP activated protein kinase (AMPK), an eukaryotic cellular energy sensor, is activated by LKB1-dependent phosphorylation. Recent studies revealed that the tumor suppressor LKB1 is a kinase involved in the polarization of both epithelial cells and neurons. In a Drosophila model, the role of LKB1 in epithelial cell polarization is mediated by AMPK. I wished to determine whether AMPK also contributes to the assembly of tight junctions in the epithelial cell polarization process. I found that AMPK is activated during Ca2+-induced tight junction assembly. Activation of AMPK by AICAR facilitates tight junction assembly under conditions of normal extracellular Ca2+ concentrations, and initiates relocalization of tight junction components under conditions of low extracellular Ca 2+. Expression of a dominant negative AMPK construct inhibits tight junction assembly, and this defect in tight junction assembly can be partially ameliorated by rapamycin. These results suggest that AMPK plays a role in the regulation of tight junction assembly.;In a non-polarized neuron, LKB1 is selectively activated only in neurites destined to become axons, and this activation precedes the selective inhibition of glycogen synthase kinase (GSK)-3beta at the same locations. GSK-3beta, however, has not yet been identified to participate directly in epithelial cell polarization. I found that in MDCK cells, the inhibition of GSK-3beta by either SB216763 or LiCl is capable of relocalizing junction proteins to the sites of cell-cell contact independent of extracellular Ca2+. I also determined that the effects of AMPK activation and GSK-3beta inhibition on the relocalization of junction proteins are synergistic and occur via different pathways. I found that knocking down E-cadherin does not affect the Ca 2+-independent relocalization of junction proteins induced by either AMPK activation or GSK-3beta inhibition, and that knocking down afadin disrupts the relocalization of junction proteins induced by AMPK activation, but not that induced by GSK-3beta inhibition. Using an in vitro kinase assay, I identified afadin as a substrate for AMPK. I conclude that AMPK activation and GSK-3beta inhibition independently induce the relocalization of epithelial junction proteins under conditions of low extracellular Ca 2+ and that afadin appears to act as a substrate of AMPK to mediate this process.