Membrane-binding and activation of LKB1 by phosphatidic acid is essential for development and tumour suppression

Abstract

The serine/threonine kinase LKB1 regulates various cellular processes such as cell proliferation, energy homeostasis and cell polarity and is frequently downregulated in various tumours. Many downstream pathways controlled by LKB1 have been described but little is known about the upstream regulatory mechanisms. Here we show that targeting of the kinase to the membrane by a direct binding of LKB1 to phosphatidic acid is essential to fully activate its kinase activity. Consequently, LKB1 mutants that are deficient for membrane binding fail to activate the downstream target AMPK to control mTOR signalling. Furthermore, the in vivo function of LKB1 during development of Drosophila depends on its capacity to associate with membranes. Strikingly, we find LKB1 to be downregulated in malignant melanoma, which exhibit aberrant activation of Akt and overexpress phosphatidic acid generating Phospholipase D. These results provide evidence for a fundamental mechanism of LKB1 activation and its implication in vivo and during carcinogenesis.

Phospholipids of the Plasma Membrane – Regulators or Consequence of Cell Polarity?

“…Thereby, the apical-basal polarity of phospholipids [PI(4,5)P2 apical, PI(3,4,5)P3 basolateral] is established and regulated. In contrast, there are increasing reports of phospholipids regulating the localization and function of polarity regulators suggest that phospholipid polarity is not only a simple consequence of protein polarity but both modules are affecting each other: Disturbed or manipulated phospholipid polarity results in mislocalization of proteins and impaired apical-basal polarity (e.g., Gassama-Diagne et al, 2006;Martin-Belmonte et al, 2007;Claret et al, 2014) and deletion/mutation of phospholipid-binding motifs in polarity regulators leads to mislocalization of these proteins and disturbed polarity (Moravcevic et al, 2010;Bailey and Prehoda, 2015;Dong et al, 2015;Dogliotti et al, 2017;Kullmann and Krahn, 2018b). Taken together, both mechanisms, phospholipid-regulated protein localization and -function and protein-directed phospholipids accumulation, are essential for cell polarization and probably affect each other reciprocally so that it is rather a positive feedback circuit than a simple cause-consequence relationship.…”

“…For the later, taking in account the charge density/concentration as well as its abundance in the plasma membrane (Balla, 2013), PI(4)P and PI(4,5)P2 are obviously the most targeted phosphoinositides (Gambhir et al, 2004). However, even short polybasic motifs might be as specific for one particular phospholipid as a specific lipid binding domain, e.g., the binding of the polybasic tail of LKB1 to PA is magnitudes higher than to PI(4,5)P2 and PI(3,4,5)P3 (Dogliotti et al, 2017).…”

“…Apart from a C-terminal farnesylation, which facilitates only transient membrane association, LKB1 contains a C-terminal polybasic motif adjacent to the farnesylated Cysteine, which directly binds to PA with a high specificity (Dogliotti et al, 2017). Mutation of this motif abolishes membrane association and severely decreases the kinase activity of LKB1 and the addition of PA-enriched micelles to recombinant LKB1 kinase complex increases its activity in vitro, and overexpression of Phospholipase D, one of the enzymes producing PA, results in an enhanced LKB1 activity in cell culture (Dogliotti et al, 2017). Notably, PDZGEF, a guanine exchange factor for Rap, also binds to PA (and other phospholipids) via a polybasic motif (Consonni et al, 2014).…”

Structural Basis for MARK1 Kinase Autoinhibition by Its KA1 Domain

“…An analogous second step of activation may be driven by MARK1 binding to peptide ligands of the KA1 domain, such as GAB1 (Yang et al, 2012). Given that the KA1 domain inhibits both T215E and WT kinase domain (Emptage et al, 2017a) and recent findings that LKB1 is also activated by anionic phospholipids (Dogliotti et al, 2017), phosphorylation at T215 may not necessarily precede relief of autoinhibition, and the order of MARK1 activation may depend on cellular context.…”

AMP-activated protein kinase – not just an energy sensor

“…As well as these findings of binding of LKB1 to lysosomes, it has recently been reported that LKB1 can associate with specific plasma membrane compartments in cells from humans and Drosophila melanogaster . This appears to be due to basic regions in the C-terminal tail of LKB1 that cause its binding to phosphatidic acid and other phospholipids; mutation of these regions interferes with AMPK activation when LKB1 is expressed in HeLa cells 44 .…”