Novel Functional View of the Crocidolite Asbestos-Treated A549 Human Lung Epithelial Transcriptome Reveals an Intricate Network of Pathways with Opposing Functions

BMC Genomics. 2008 Aug 7;9(1):376. [Epub ahead of print] [Link]

Hevel JM, Olson-Buelow LC, Ganesan B, Stevens JR, Hardman JP, Aust AE.


Background: Although exposure to asbestos is now regulated, patients continue to be diagnosed with mesothelioma, asbestosis, fibrosis and lung carcinoma because of the long latent period between exposure and clinical disease. Asbestosis is observed in approximately 200,000 patients annually and asbestos-related deaths are estimated at 4,000 annually[1]. Although advances have been made using single gene/gene product or pathway studies, the complexity of the response to asbestos and the many unanswered questions suggested the need for a systems biology approach. The objective of this study was to generate a comprehensive view of the transcriptional changes induced by crocidolite asbestos in A549 human lung epithelial cells.

Results: A statistically robust, comprehensive data set documenting the crocidolite-induced changes in the A549 transcriptome was collected. A systems biology approach involving global observations from gene ontological analyses coupled with functional network analyses was used to explore the effects of crocidolite in the context of known molecular interactions. The analyses uniquely document a transcriptome with function-based networks in cell death, cancer, cell cycle, cellular growth, proliferation, and gene expression. These functional modules show signs of a complex interplay between signaling pathways consisting of both novel and previously described asbestos-related genes/gene products. These networks allowed for the identification of novel, putative crocidolite-related genes, leading to several new hypotheses regarding genes that are important for the asbestos response. The global analysis revealed a transcriptome that bears signatures of both apoptosis/cell death and cell survival/proliferation.

Conclusions: Our analyses demonstrate the power of combining a statistically robust, comprehensive dataset and a functional network genomics approach to 1) identify and explore relationships between genes of known importance 2) identify novel candidate genes, and 3) observe the complex interplay between genes/gene products that function in seemingly different processes. This study represents the first function-based global approach toward understanding the response of human lung epithelial cells to the carcinogen crocidolite. Importantly, our investigation paints a much broader landscape for the crocidolite response than was previously appreciated and reveals novel paths to study. Our graphical representations of the function-based global network will be a valuable resource to model new research findings.