Pulmonary epithelium is divided into three histopathologically and biologically distinct compartments: bronchi, bronchioli, and alveoli. Two of the main types of human lung cancers (squamous cell and small cell carcinoma) are typically centrally located and originate from bronchi. In contrast, the most common lung cancer type adenocarcinoma is typically a peripheral tumor which originates from the peripheral airway cells (PACs) composed of bronchiolar Clara cells or alveolar type 2 cells. Our long term goal is to characterize the morphological changes, aberrant cellular differentiation and genetic damage associated with premalignant changes and carcinogenesis in each compartment. One of our main tools is the development of relevant mouse models for human lung carcinogenesis.
A. Peripheral airway cell (PAC) and neuroendocrine (NE) differentiation We have previously shown that field cancerization in human lung is associated with alterations in the expression patterns of both PAC and NE markers. We are now developing experimental models in which the pulmonary changes parallel those seen in man. Exposure to naphthalene, a component of cigarette smoke, kills lung airway epithelial (Clara) cells in mice, but is rapidly followed by Clara cell reconstitution coincident with proliferation of pulmonary NE cells (PNECs). The reconstituting progenitor cells have been suggested to enter a transient NE differentiation phase before differentiating to Clara cells. Furthermore, these progenitors were suggested to be the target population for transformation to a NE tumor; small cell lung cancer (SCLC) whose precursor lesions remain unknown. To determine the relevance of the NE phenotype to post-naphthalene reconstitution, we examined post-naphthalene reconstitution in Gfi1-/- mice, which display a marked (70%) reduction of PNECs. Interestingly, our data from these experiments suggest that neither Gfi-1 nor the NE phenotype play a dominant role in the regeneration process. However, the few Gfi1-/- cells capable of NE differentiation showed a previously proposed role for Gfi1 in controlling NE cancer growth. Approximately one third of all human lung cancers are characterized by NE differentiation. We have previously shown that a neural transcription factor from Drosophila, achaete-schute homolog-1 is expressed at high levels in NE lung cancers and is essential for NE differentiation in neoplastic and non-neoplastic lung. We have also shown that constitutive expression of achaete-schute homolog-1 under CC10 promoter together with SV40 caused massive tumors with NE differentiation. Notably, constitutive expression of the achaete-schute homolog-1 transcription factor alone under CC10 promoter also caused marked bronchialization of alveoli, which provides a model for a potential premalignant lesion for human lung adenocarcinomas. We have previously shown that such lesions occur in 12% of human lung cancer resection specimens being part of the field cancerization phenomenon. Using our mouse model we have now demonstrated that the lesions are due to sustained cellular proliferation in terminal bronchioles and resistance to apoptosis. The molecular mechanisms involved increased expression of anti-apoptotic Bcl-2 and cMyb as well as the activation of Akt/mTOR pathway. We were also able to show that forced expression of hASH1 in immortalized human bronchial epithelial cells decreased apoptosis. The significance of our findings is two-fold: First, we show that constitutive expression of hASH1 in lung epithelium promotes remodeling through the same pathways that are commonly activated during lung carcinogenesis and provides a novel model for the potential premalignant lesion of human lung adenocarcinomas. Second, our results suggest that the impact of hASH1 is not limited to neuroencocrine cells, but is extended to epithelial cells as well. We are currently assessing further the role of specific molecular pathways in the progression of bronchiolization of alveoli and formation of adenocarcinomas.
B. Mouse model for human small cell lung cancer (SCLC) The prototype of human NE lung cancers is small cell lung cancer (SCLC), the most common and virulent of all NE cancers in man. The precursor lesions of SCLC are unknown. One hundred percent human SCLCs contain alterations of RB1 and/or P53 tumor suppressor genes. We established a mouse model by conditional inactivation of these genes in mouse lung epithelial cells. Mice carrying conditional alleles for both RB1 and P53 developed highly aggressive lung tumors with striking similarities with human SCLC in their morphology, immunophenotype and extrapulmonary metastatic spread. In our model, inactivation of both RB1 and p53 was a prerequisite for the pathogenesis of SCLC. We are using this model together with the bitransgenic CC10-SV40Tag-hASH1 mouse model for non-SCLC with NE features to identify precursor lesions, progenitor and stem cells for SCLC.
C. Generating and characterizing a mouse model for squamous cell carcinogenesis in the lung. Squamous cell carcinoma composes approximately 20% of all lung cancers in the United States. It is a central tumor arising from bronchi, and the only human lung carcinoma for which the precursor lesions are well established. However, squamous cell carcinoma in mouse is rare, and there is a critical need to have a good pre-clinical model for chemoprevention studies. Our goal is to establish a mouse model for lung squamous cell carcinoma. We postulate that keratin 5/14 (K5/K14) containing cells are necessary for the formation of squamous cell carcinomas. We have shown that they are scanty or non-existent in mouse intrapulmonary airways. We have generated a novel transgenic mouse which constitutively expresses human K14 in airway epithelial cells under the CC10 promoter. This mouse developed normally and, at the molecular level, expressed the squamous differentiation program. Histologically, airway epithelium in some animals displayed multifocal hyperplasia, squamous metaplasia and dysplasia, similar to the premalignant lesions seen in humans. Older mice had a 30% increased incidence of tumorigenesis, but overt squamous maturation was rare. Moreover, we discovered that the constitutive expression of human K14 in airways significantly enhanced tumorigenesis in chemical and viral/genetic lung carcinogenesis models. We are also generating in vitro cell lines that express human K14 under CC10 promoter for functional studies. The significance of this project is that it will generate a preclinical model for testing novel chemotherapeutic agents.
The significance of our research is that the results will provide a rational basis for early detection and molecular targets for therapeutic interventions for human lung carcinogenesis by identifying specific markers and pathways as well as distinct models of multistep epithelial carcinogenesis
Showing the most recent 10 out of 21 publications
