PerspectiveCancer

Tumor Suppression by LKB1: SIK-ness Prevents Metastasis

See allHide authors and affiliations

Science Signaling  01 Sep 2009:
Vol. 2, Issue 86, pp. pe55
DOI: 10.1126/scisignal.286pe55

Abstract

The LKB1 serine-threonine kinase is a tumor suppressor that is inactivated in a large number of sporadic human lung non–small cell carcinomas (NSCLCs) and cervical cancers. Genetic deletion of LKB1 in various mouse tissues results in tumorigenesis, and loss of LKB1 increases metastasis in a mouse model of NSCLC. LKB1 directly activates a family of 14 kinases related to AMPK [adenosine monophosphate (AMP)–activated protein kinase] to control cell metabolism, growth, and polarity, though which of these are critical to its tumor suppressor functions remain undefined. The LKB1-dependent kinase SIK1 (salt-inducible kinase 1) has now been identified as a key modulator of anoikis (apoptosis induced by cell detachment) and transformation in culture, and its modulation of the tumor suppressor p53 controls metastasis in transplanted tumor cells. Reduced SIK1 expression is correlated with poor prognosis in two large human breast cancer data sets. These findings suggest that SIK1 is a key upstream regulator of p53-dependent anoikis that may be targeted in tumorigenesis.

The gene encoding the LKB1 tumor suppressor is located on human chromosome 19p13, and its inactivation is responsible for the inherited cancer disorder Peutz-Jeghers syndrome (1). LKB1 is also one of the most commonly mutated genes in sporadic human lung cancer, particularly in multiple subtypes of non–small cell lung carcinoma (NSCLC), where at least 15 to 35% of cases have this lesion (2, 3). LKB1 is somatically mutated in 20% of cervical carcinomas, making it the first known recurrent genetic alteration for this tumor type (4).

LKB1 is a highly conserved serine-threonine kinase that directly phosphorylates a threonine residue in the activation loop of 14 kinases in the AMPK [adenosine monophosphate (AMP)–activated protein kinase] family in mammals (5), resulting in their activation (Fig. 1). Thus, in tumor cells acquiring LKB1 mutations, up to 14 different downstream kinases become inactivated. Which of these kinases are most critical for the role of LKB1 as a tumor suppressor remains an unanswered question. Currently, there is minimal mutational data from human tumors to specifically support any single LKB1-dependent kinase as the critical target for LKB1 in tumorigenesis. However, there appears to be a great deal of redundancy among them, suggesting that, in many tissues, loss of any one kinase may be compensated for by other family members (6).

Fig. 1

LKB1 tumor suppressor signaling. LKB1 directly phosphorylates and activates AMPK and 12 related kinases, including the SIK, MARK (microtubule affinity-regulating kinase), SAD [synapses of the amphid defective; also known as BRSKs (brain-specific kinases)], and NUAK subfamilies. These kinases in turn directly phosphorylate a number of downstream effectors to regulate cell growth, metabolism, and cell polarity. Many of the known AMPK substrates can also be phosphorylated by SIK family members—and perhaps also by MARKs under different cellular conditions.

Though originally characterized for its roles in glucose and lipid metabolism, AMPK has been connected to growth control over the past 5 years because of reports that it activates the TSC2 (tuberous sclerosis 2) and p53 tumor suppressors (79). Consequently, many have assumed that the tumor suppressor activity of LKB1 must be partially mediated through AMPK-dependent increases in p53 activity and decreases in mTOR (mammalian target of rapamycin) activity. In contrast to AMPK, little is known about many of the 12 LKB1-dependent kinases, which are all structurally related to AMPK. On the basis of sequence homology, the salt-inducible kinases SIK1 (also known as SNF1LK), SIK2 (also known as SNF1LK2) and SIK3 constitute one subfamily of AMPK-related kinases that are activated by LBK1. The best-established role for SIK family members relates to their ability to directly phosphorylate many of the same metabolic transcriptional regulators as AMPK in muscle or liver, including the CREB (cAMP response element–binding protein)–regulated transcription coactivator (CRTC) family (1012) and class II histone deacetylases (HDACs) (13, 14). Despite this functional overlap in metabolic control and downstream substrates with AMPK, a role for SIK family members in growth control had not been examined.

Enter the power of an unbiased short hairpin RNA (shRNA) screen for cancer-related genes. In a screen for kinases whose loss promoted anchorage-independent growth in soft agar when combined with an activated PI3KCA allele, Zhao and colleagues discovered that multiple shRNAs directed against SIK1 were top hits (15). Consistent with a previous study from this lab which revealed that loss of the p53 tumor suppressor synergizes with phosphoinositide 3-kinase to promote anchorage-independent growth (16), SIK1 shRNA-expressing cells exhibited decreased p53 abundance under anchorage-independent conditions. A direct examination revealed that indeed SIK1 was capable of modulating p53 phosphorylation and abundance in culture only under conditions of cell detachment and not in response to other cellular stresses, including glucose deprivation, which activates AMPK modulation of p53. Immunoprecipitated SIK1 kinase activity toward p53 was increased only under conditions of cell detachment (15), perhaps suggesting that its activation by LKB1 or its localization or association with downstream substrates may be stimulated by cell detachment. Future studies will be needed to dissect how cell detachment controls SIK1 kinase activity.

Unlike SIK1, for which there was no previous indication of a link to anoikis or metastasis, p53 deficiency accelerates the rate of metastasis in mouse models of several different cancer types (including liver, lung, and pancreas) (2, 1719), and p53 regulates anoikis and invasion in culture (20, 21).

Given the previous connection between loss of LKB1 and increased metastasis in the K-ras mouse model of NSCLC (2), the authors directly examined whether SIK1 modulates experimental metastasis. Strikingly, loss of SIK1 promoted the development of micrometastases in the absence of primary tumors in immunodeficient mice (15). Finally, reconstitution of LKB1-deficient NSCLC cells with a constitutively activated allele of SIK1 restored anoikis and inhibited growth in suspension and the ability of injected tumor cells to metastasize, effects that were abolished by shRNA directed against p53 (15). Taken together, these data suggest that SIK1 is an unappreciated key regulator of p53 downstream of LKB1 that may play a unique role in suppressing growth under anchorage-independent conditions.

Is SIK1 involved in human cancer? Though a few cases of somatic mutation in SIK1 have been observed in cancer genomic sequencing efforts (22), the authors found that SIK1 abundance was significantly reduced in primary breast tumors compared with normal breast tissue. Moreover, expression profiling of breast cancer samples revealed that reduced SIK1 expression was associated with a reduced interval before the appearance of distal metastasis (15). Future studies using genetically engineered mouse models and further analysis of human mutational data are required to define how broad a role SIK1 plays in the suppression of metastasis in different tumor types.

One interesting possibility that arises from these findings is that different members of the AMPK kinase family may target many of the same downstream substrates but in response to distinct stress stimuli. Determining the molecular basis for the stress-specific regulation downstream of LKB1, and the factors that dictate whether one of these kinases is more important than the others in growth or metastasis suppression within a given tissue, are critical goals in investigating this exciting and expanding signaling pathway.

References

View Abstract

Navigate This Article