This is the question asked by Joyce Osanyingbemi-Obidi and colleagues from Johns Hopkins in an interesting recent paper published in Molecular Cancer Research. The cancer stem cell hypothesis proposes that there is a small proportion of tumor cells with stem cell-like characteristics, including self-renewal capability, as well as a much larger proportion of "differentiated" cells with limited proliferative capability. While stem cell-like cells have been characterized in a variety of malignancies, such a population for NSCLC has not been reported. I previously blogged on a murine model for evidence of lung stem/progenitor cells. The authors cite the emerging body of data (perhaps not surprisingly) suggesting that the same key signaling pathways in embryonic and fetal development (e.g. Notch, Wnt, and Hedgehog) are also involved in the regulation of cancer stem cells. The intracellular domain of Notch receptors (NICD) is released upon ligand binding and translocates to the nucleus where it converts the CSL complex from a transciptional repressor into an activator. However, the NICD-CSL complex requires mastermind-like (MAML) proteins to recruit and bind coactivators that drive expression of downstream targets that promote cell growth and proliferation. Aberrant Notch signaling has been described in a variety of malignancies (albeit with both activating and inactivating Notch mutations) and, indeed, elevated expression of Notch family members and a key downstream target, "hairy and enhancer of split 1" (Hes1), have been reported in NSCLC.
Using NSCLC cell lines showing overexpression of Notch3 and Hes1, the authors demonstrated that pharmacologic inhibition of Notch was associated with suppression of Hes1 expression and inhibition of clonogenic growth. The authors used a clever LSL-KRASG12D mouse model in which an oncogenic KRAS allele is silenced by a transcription stop site flanked by lox P sites; inhalation of an adenovirus vector containing the Cre recombinase promotes rearrangment of the lox sites and deletion of the transcriptional stop and selective overexpression of mutant KRAS in bronchial and alveolar cells. The authors showed a transient induction of Hes1 expression by immunohistochemistry in airway epithelial cells 1 week postinfection with adenovirus and gradually diminshing Hes1-positive cells in tumors over time. They then used an novel inducible double-negative MAML knock-in murine model to show that inhibition of Notch signaling in vivo was not sufficient to suppress carcinogenesis or tumor progression in the context of mutant KRAS-driven carcinogenesis.
So--these data suggest that Notch signaling may have a role in early lung adenocarcinogenesis through the transformation of a tumorigenic stem cell-like cell but is not likely required for adenocarcinogenesis. The authors provide several notable caveats and qualifications of this conclusion but, I think the key consideration is that Notch signaling is considerably more complex that we are given to understand. Notch signaling in development can maintain stem cell characteristics in certain cell populations or drive cell differentiation in other. Likewise, Notch pathway activation in neoplasia appears to have tumor-promoting and tumor-suppressing effects. Also, not all Notch family members appear to equally contribute to lung neoplasia. Finally, individual Notch isoforms may have opposing effects on carcinogenesis.
This paper may not be of general interest, especially for the busy practicing pathologist, but is worthwhile to reflect upon and ruminate over. Intriguing stuff.