Year: 2012
Type: article
Abstract: Malignant melanoma is characterized by early and widespread metastasis, which is the major cause of morbidity and mortality in this cancer. In keeping with the highly migratory behavior of the melanocyte, melanomas can metastasize to virtually every organ often with limited primary tumor burden. An intriguing recent study by Sharpless and colleagues (Liu et al., 2012) now establishes the LKB1 tumor suppressor as a critical gatekeeper for this aggressive metastatic behavior. Using genetically engineered mice with melanocyte-specific expression of activated KRAS and deletion of LKB1, they show that LKB1 inactivation promotes both the initiation of melanoma and its extensive metastasis. Studies in cell lines revealed that LKB1 acts to inhibit the SRC family kinase YES and consequently restrain the expansion of a highly invasive subpopulation of melanoma cells expressing the CD24 stem cell marker. These findings point to targeting YES kinase as a promising strategy for the treatment of metastatic melanoma. Of additional significance, although multiple genetically engineered mouse models have been developed that incorporate mutations seen in the human disease, most fail to demonstrate the characteristic metastatic propensity that is faithfully recapitulated in the LKB1–KRAS mice. These findings also suggest insights into the pathophysiology of the abnormal mucocutaneous pigmentation in patients with germline LKB1 mutations (i.e., patients with Peutz–Jeghers syndrome) and more broadly into the biology of normal melanocytes. LKB1 was first identified by positional cloning approaches as being mutationally inactivated in patients with the PJS polyposis, hyperpigmentation, and cancer syndrome [reviewed in (Shackelford and Shaw, 2009)]. This gene encodes a serine–threonine kinase with well-established functions in both maintaining metabolic homeostasis in response to decreased cellular energy supply and in the establishment of epithelial cell polarity. These actions are mediated by members of the AMPK-related kinase family, which are directly phosphorylated and activated by LKB1. How these various functions relate to LKB1-mediated tumor suppression remains an area of active investigation. A series of mouse models have been developed to study the role of LKB1 in different tissues and genetic contexts. Whereas the data clearly confirm key requirements for LKB1 in tumor suppression, they also suggest that there is important tissue specificity in terms of whether LKB1 inactivation results in benign or malignant tumors, or alternatively causes cell death (Ollila and Mäkelä, 2011). The study by Sharpless and colleagues is the first to focus on genetically dissecting LKB1 function in melanocytes, and is of particular interest in regard to the hyperpigmentation in PJS, and the documentation of somatic inactivating LKB1 mutations in ~5–10% of sporadic human melanomas. The authors first generated a series of mice in which an activated KRAS allele (KRASG12D) and/or null alleles for LKB1 can be induced specifically in melanocytes through the use of tamoxifen-inducible Tyrosinase-CRE-ERT2 transgene. The authors find that WT, LKB1−/−, and KRASG12D-expressing primary melanocytes grow poorly in vitro, whereas KRASG12D/LKB1−/− cells exhibit unimpeded immortal growth. Correspondingly, no overt defects were observed when these individual genetic alterations were induced in vivo, while combined KRASG12D/LKB1 mutations resulted in melanocytic hyperproliferation and darkened coat color, and subsequently in rapid development of melanoma. These mice also showed a prominent metastatic phenotype, with gross lesions in the lymph nodes, liver, lung, and spleen. By contrast, other models with combined KRASG12D activation and deletion of either p53 or INK4A had comparable latency in primary melanoma formation, but these tumors failed to metastasize. Collectively, these findings demonstrate that LKB1 suppresses early stages of melanoma initiation and progression and also point to a specific and potent role for LKB1 in restraining metastasis. The extent of metastasis is particularly impressive even when compared with other models that exhibit a metastatic phenotype—for example, BRAFV600E and PTEN deletion in melanocytes results in metastatic melanoma—but these lesions are largely limited to the regional lymph nodes and the lungs (Dankort et al., 2009). Thus, the KRASG12D/LKB1 mice should serve as a valuable model of this fundamental aspect of the human disease. The authors found that restoration of LKB1 suppressed migration and invasion of KRASG12D/LKB1−/− melanoma cell lines, and reciprocally, that LKB1 knockdown in KRASG12D–p53−/− murine melanoma cells or in a human melanoma cell line resulted in enhanced motility. These findings thereby establish that LKB1 acts in a direct manner to control metastasis. To investigate the mechanism of elevated migratory capacity, Liu and colleagues performed a series of elegant experiments to show that the SRC family kinase, YES, was constitutively activated in LKB1 mutant cells. Pharmacologic or RNAi-mediated inhibition of YES reduced migration in vitro, and YES knockdown effectively reduced metastasis in vivo. Interestingly, the metastatic lung cancers provoked by combined KRASG12D/LKB1 mutation are associated with a marked elevation in SRC signaling (Carretero et al., 2010), suggesting a broad importance of SRC family kinases in mediating metastasis downstream of LKB1, but with selective roles for the individual family members in different tissues. To further identify programs downstream of LKB1 that contribute to the metastasis of KRASG12D/LKB1 melanomas, Liu et al. performed mRNA expression analysis in mouse melanoma tumor tissue or cell lines with or without functional LKB1. These analyses suggested that LKB1 deficiency was associated with a significant enrichment in the expression of transcriptional targets of LEF1, a key transcription factor in the WNT/beta-catenin pathway. Additionally, their results indicated that CD24 (CD24a in mice), a glycophosphatidylinositol-anchored surface protein gene associated with metastatic behavior and acquisition of stem cell properties, was consistently elevated in LKB1 mutant melanomas. Whereas melanomas with intact LKB1 exhibited heterogeneous expression of CD24, with only ~1–13% of the cells detected as CD24+ by flow cytometry, LKB1 deficiency was associated with an increase in the CD24+ population (ranging up to 30% of the cells), which was decreased by restoration of LKB1. Analysis of human melanoma cell lines also showed inverse correlation between LKB1 and CD24 levels. Interestingly, CD24-high subpopulations displayed strikingly elevated levels of activated YES kinase compared with CD24-low cells in the same cell line. Furthermore, the authors showed that CD24+ cells were significantly more metastatic in vivo than CD24− melanoma cells. Consistent with a critical role for YES kinase in the expansion of CD24+ melanoma cells downstream of LKB1, the authors show that genetic or pharmacologic YES inactivation strongly reduces the proportion of CD24+ cells. Based on this demonstration of a physiologically relevant LKB1–YES–CD24 pathway, it will be important to elucidate the precise signaling mechanisms that are at play, and to further define the molecular program regulating growth of the highly invasive CD24-positive subpopulation and understand its lineage relationships to other melanoma cells. Furthermore, in addition to its role in restraining cell motility, whether this signaling pathway is responsible for the initiation of melanoma and the in vitro immortalization of melanocytes that result from LKB1 inactivation remains to be determined. The findings of Liu et al. should motivate a detailed examination of the extent to which LKB1 inactivation contributes to metastasis in human melanoma patients. Key questions include establishing whether there is an association between LKB1 mutations in the primary tumor and metastasis in human patients and whether LKB1 inactivation correlates with YES activation in this setting. Such results could thus potentially establish the LKB1 mutant subset as a distinct clinical entity with specific clinicopathologic features and sensitivity to specific therapies, as appears to be the case in lung, uterine, and cervical cancers. In terms of the frequency of LKB1 defects in melanoma, the observed haploinsufficient tumor phenotypes in some mouse models, and the inverse correlation between wild-type LKB1 expression levels and activation of the YES–CD24 migration program suggest that single-copy deletion may be an additional mode of LKB1 inactivation in melanoma beyond intragenic mutation. It is possible that lesions in upstream signaling components may also compromise LKB1 activity. For example, activated BRAF has been reported to phosphorylate and inactivate LKB1 (Zheng et al., 2009), although the presence of concurrent BRAF and LKB1 mutations in the same cell lines suggests that this may not be a major contributor to pathway inactivation. In conclusion, the study by Li et al. uncovered powerful functions for LKB1 in restraining both melanoma initiation and metastasis. Their novel mouse model recapitulates the pronounced metastatic potential of the corresponding human malignancy and will be a valuable system to further tackle questions regarding the underlying mechanisms and to test experimental therapeutics. The link between LKB1–YES–CD24 provide intriguing leads to start unraveling the role of this elusive tumor suppressor and define novel therapeutic modalities tailored to specific subsets of tumors.
Source: Pigment Cell & Melanoma Research
DOI: 10.1111/pcmr.12009
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2021125583