Credit: PHOTOALTO

Non-small-cell lung cancer (NSCLC) includes several subtypes that, although different in histology, all have a poor prognosis. A study in Nature now shows that inactivation of the tumour-suppressor gene serine/threonine kinase 11 (also known as LKB1 ) underlies some of these variants, and has multiple effects on lung tumorigenesis.

Although LKB1 inactivation occurs in lung adenocarcinomas along with KRAS activation, mouse embryonic fibroblasts lacking Lkb1 are resistant to transformation with oncogenic Ras. To address this conundrum Nabeel Bardeesy, Norman Sharpless, Kwok-Kin Wong and colleagues studied the effect of Lkb1 inactivation in a mouse model of Kras-induced lung cancer.

Mice expressing oncogenic Kras (KrasG12D) in lung cells developed non-metastatic tumours after a long latency. The concomitant loss of the tumour suppressors Trp53, Ink4a or both Ink4a and Arf, promoted KrasG12D-induced lung tumorigenesis; but, surprisingly, Lkb1 loss had a stronger effect than any of these tumour suppressors, causing more and larger tumours, which had a shorter latency and more frequently metastasized. In the absence of KrasG12D, Lkb1 inactivation does not cause lung cancer, suggesting that LKB1 is crucial for suppressing KrasG12D-induced tumorigenesis.

Interestingly, Lkb1 inactivation expanded the range of histological tumour variants because, in addition to the adenocarcinomas observed in KrasG12D mice (with or without inactivation of Trp53, Ink4a or both Ink4a and Arf), KrasG12D; Lkb1−/− mice also developed squamous cell carcinomas (SCCs), adenosquamous carcinomas and large-cell carcinomas (LCCs), indicating that LKB1 has a role in cell differentiation. Consistent with these findings, the authors found inactivating mutations of LKB1 in all the histological subtypes of human NSCLC.

How does LKB1 exert its anti-tumour activity? LKB1 is known to function through activation of p53 and/or INK4a and ARF, but KrasG12D;Lkb1−/− tumours are more aggressive than those lacking p53 or INK4a and ARF, indicating that LKB1 has anti-tumour activity that is independent of these other tumour suppressors. Consistent with this, LKB1 expression in a human NSCLC cell line that lacks both LKB1, and INK4a and ARF function inhibited anchorage-independent growth and metastasis formation in mice after tail-vein injection, without affecting the expression of p53 and its targets.

Gene-expression analysis of KrasG12D mouse tumours and human NSCLC cells with different LKB1 status identified several metastasis-promoting genes as LKB1 targets. Inactivation of one of these, NEDD9 , by RNA interference, inhibited migration and invasion of NSCLC cells in vitro, indicating that the NEDD9 upregulation observed following LKB1 loss could be crucial in promoting metastases.

Altogether these data show that LKB1 suppresses lung cancer by inhibiting tumour initiation, differentiation and metastasis, and therefore LKB1 loss could represent a useful prognostic marker of aggressive disease. It will be important to determine whether LKB1 targets might be exploited for new therapeutic strategies.