A new mechanism for LKB1 activation

ABSTRACT

Liver kinase B1 (LKB1, also known as serine/threonine kinase 11, STK11) has been thought to be a constitutively active tumor suppressor that is activated by forming an active complex. Very recently, a new post-translational modification on LKB1 was identified that can regulate LKB1 activation and LKB1-mediated cancer cell survival under energy stress.

Abbreviations

ACC

acetyl Co-A carboxylase

AMPK

AMP-activated protein kinase

BRAF

v-RAF murine sarcoma viral oncogene homolog B1

CAB39

calcium binding protein 39

ERK

extracellular signal-regulated kinase

HCC

hepatocellular carcinoma

K63

lysine 63

LKB1

liver kinase B1

MEK

mitogen-activated protein/extracellular signal-regulated kinase kinase

MO25

mouse protein 25

p90RSK

90-kDa ribosomal protein S6 kinase

RAS

rat sarcoma

SCF

SKP1-Cullin-F-box

STK11

serine/threonine kinase 11

SKP

S-phase kinase-associated protein

STRAD

STE20-related kinase adaptor

UBD

ubiquitin-binding domain

Liver kinase B1 (LKB1), also known as serine/threonine kinase 11 (STK11), has commonly been considered a tumor suppressor, as evidenced by genetic studies of the cancer-prone Peutz-Jeghers syndrome.¹ LKB1 functions as a master upstream kinase by directly phosphorylating and activating AMP-activated protein kinase (AMPK) and 12 AMPK-related kinases,² and is thereby involved in a wide range of cellular events such as energy metabolism, proliferation, apoptosis, and cell polarity.³ Among those substrates, AMPK, a key sensor of intracellular energy status, was the first to be identified and is the best characterized substrate of LKB1, and many functions of AMPK account for the roles of LKB1 in tumor suppression.⁴ However, emerging evidence indicates that LKB1 may also exhibit previously unrecognized pro-oncogenic functions. Regardless of what roles LKB1 may play, it remains largely unclear how LKB1 activity is regulated. In our recent paper published in Molecular Cell, we identified a new regulatory mechanism for LKB1 activation that reveals the role of LKB1 in cancer cell survival during energy stress.⁵

LKB1 activation is governed by a phosphorylation-independent allosteric mechanism in which it forms a heterotrimeric complex with 2 accessory proteins, STE20-related kinase adaptor (STRAD) and mouse protein 25 (MO25, also known as calcium binding protein 39, CAB39). Within the LKB1-STRAD-MO25 heterotrimer there are considerable interactions between the 3 proteins. STRAD, displaying an active conformation stabilized by association with MO25 and ATP,⁶ binds to LKB1 as a pseudo-substrate, and binding of MO25 to LKB1, stabilized by the presence of STRAD, properly positions the LKB1 activation loop in an optimally active conformation competent for phosphorylating substrates.⁷ Hence, both STRAD and MO25 are important for LKB1 activity. However, it is poorly understood how the LKB1-STRAD-MO25 complex is maintained and regulated. We uncovered a new regulator of LKB1 activation—S-phase kinase-associated protein 2 (SKP2), a component of the S-phase kinase-associated protein 1 (SKP1)-Cullin-F-box (SCF) ubiquitin ligase complex—and demonstrated that SKP2-driven lysine 63 (K63)-linked polyubiquitination of LKB1 is indispensable for LKB1-MO25 interaction, LKB1 activity, and downstream AMPK signaling. The extended conformation of K63-linked ubiquitin chains has been thought to provide a unique platform allowing proteins containing ubiquitin-binding domains (UBDs) to bind to ubiquitinated proteins. As most UBDs bind to a hydrophobic patch of ubiquitin using α-helical structures, the amphipathic nature of the MO25 helices, with conserved hydrophobic patches within the α-helical repeats, implicates the potential existence of unidentified UBD(s) in MO25 that can interact with the ubiquitinated LKB1. Further experiments will be needed to test this assumption and determine how LKB1 ubiquitination mechanistically modulates the LKB1-MO25 interaction.

By screening different upstream stimuli, we further identified oncogenic rat sarcoma (RAS) protein as an upstream activator of the SKP2/LKB1/AMPK axis. Unexpectedly, the oncogenic insult of RAS hyperactivation promoted assembly of the SKP2-SCF complex, leading to elevated K63-linked polyubiquitination of LKB1 and subsequent activation of LKB1/AMPK signaling. It should be noted that RAS/v-RAF murine sarcoma viral oncogene homolog B1 (BRAF)/mitogen-activated protein/extracellular signal-regulated kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) signaling has previously been shown to affect LKB1 function via ERK-mediated Ser325 phosphorylation and 90-kDa ribosomal protein S6 kinase (p90RSK)-mediated Ser428 phosphorylation of LKB1.⁸ However, in our study, phosphorylation of LKB1 at either or both of these sites did not significantly affect SKP2-mediated LKB1 ubiquitination and activation (unpublished data). Interestingly, we found that RAS and SKP2 cooperated to induce LKB1 ubiquitination and activation. While the synergistic effect of RAS and SKP2 on LKB1 ubiquitination and activation is compelling, the molecular mechanism by which the RAS pathway promotes SKP2-SCF activity toward LKB1 remains to be defined. Future investigations will thus be required to address this important question.

The LKB1/AMPK pathway mediates multiple cellular mechanisms for proper responses to intracellular energy status. It switches metabolic pathways between anabolism and catabolism in response to distinct energy insults and coordinates cell growth and autophagy in order to maintain energy and nutrient homeostasis in cells. The LKB1/AMPK pathway also performs a critical function in cell survival following metabolic stress.^9,10 We demonstrated that RAS/SKP2-dependent ubiquitination of LKB1 is essential for protecting cancer cells from metabolic stress-induced cell death. Ablation of SKP2 or LKB1 in RAS-overexpressing cells compromised the RAS-mediated protective effect on cell survival. These results underscore the role of the RAS/SKP2/LKB1 axis in regulating cancer cell survival under energy stress, and may provide a new therapeutic target for cancer treatment. Treating RAS-driven cancers is a challenging task because of the extreme complexity of the RAS signaling networks, which often leads to unwanted therapeutic outcomes such as drug resistance and off-target effects. We found that interrupting the RAS/SKP2/LKB1 cascade by pharmacologic inhibition of SKP2 further sensitized the RAS-overexpressing cells to the metabolic drug phenformin. Our data therefore provide a proof of principle that combination treatment with SKP2 inhibitor and metabolic drugs may be highly effective for treating RAS-driven cancers.

In addition to our in vitro observations, we also highlighted the in vivo importance of the SKP2/LKB1 pathway in cancer regulation. In hepatocellular carcinoma (HCC) patient samples, we found that SKP2 expression, LKB1 expression, and phosphorylation of the AMPK downstream effector acetyl Co-A carboxylase (ACC) are all upregulated and significantly related to disease stage. Remarkably, not only do their levels positively correlate with one another, but their high expression levels can also independently serve as prognostic markers for poor survival outcome. Mouse xenograft tumor models also showed that overexpression of wild-type LKB1, but not its ubiquitination-deficient or kinase-dead mutants, promotes HCC tumor growth. These data suggest that LKB1 exhibits oncogenic activity in HCC. However, resolution of the paradoxical and complex roles (including oncogenic, tumor-suppressive, and tissue-, context- and stage-specific functions) of LKB1 in cancer development will require generation of appropriate mouse models with conditionally manipulatable Lkb1.

As the master regulator LKB1 participates in a variety of cellular signaling pathways, understanding how it is activated has been an intriguing question. We have provided convincing evidence supporting K63-linked ubiquitination of LKB1 as a new post-translational modification event that regulates LKB1 activation and is important for its function in cancer cell survival (Fig. 1). Identification of the role of the RAS/SKP2/LKB1 axis also suggests a potential strategy for cancer therapy.

Figure 1.

RAS/SKP2-driven K63-linked ubiquitination of LKB1 modulates LKB1 activation. We demonstrated that SKP2-SCF is a direct ubiquitin ligase for liver LKB1 and that oncogenic RAS protein is an upstream regulator of the LKB1/AMPK axis, in which AMPK exists in a heterotrimeric complex (α/β/γ subunits) and is activated by LKB1-mediated Thr172 (T172) phosphorylation of AMPKα. Mechanistically, RAS/SKP2-mediated K63-linked polyubiquitination of LKB1 regulates LKB1 activity by maintaining the integrity of the LKB1-STRAD-MO25 complex and thus the function of LKB1 in cancer cell survival following energy stress. Interrupting the RAS/SKP2/LKB1 pathway by administration of a SKP2 inhibitor may be a potential strategy for cancer treatment, and further studies are required to fully understand how this pathway relates to cancer development. AMPK, AMP-activated protein kinase; K63, lysine 63; LKB1, liver kinase B1; MO25, mouse protein 25; RAS, rat sarcoma; SKP2-SCF, S-phase kinase-associated protein 2 (SKP2)-S-phase kinase-associated protein 1 (SKP1)-Cullin1-F-box; STRAD, STE20-related kinase adaptor.

Funding Statement

This study was supported by CPRIT grant, NIH RO1 grants, R. Lee Clark Award, and MD Anderson prostate moonshot program to H.-K.L.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.