(422e) Regulatory Association of Cell Responses Induced By Metal and Metal Oxide Nanoparticles
- Conference: AIChE Annual Meeting
- Year: 2013
- Proceeding: 2013 AIChE Annual Meeting
- Group: Topical Conference: Environmental Aspects, Applications, and Implications of Nanomaterials and Nanotechnology
- Time: Wednesday, November 6, 2013 - 10:10am-10:35am
A data mining approach was developed and applied to discover relationships among fourteen different types of cell responses (including ten signaling pathway activities and four cytotoxicity effects) for two different cell lines in response to exposure to six different metal and metal oxide nanoparticles. Association rules were identified for a High-Throughput Screening (HTS) data developed for murine macrophage (RAW264.7) and transformed bronchial epithelial (BEAS-2B) cells. The dataset consisted of a total of 55,296 data readings for nanoparticle-regulated pathway activities over an exposure period up to 24 h and exposure concentration range of 0.39-200 µg/mL. HTS data were processed for outlier removal and standardized via inter/intra-plate normalization with subsequent hit-identification (for significantly regulated pathways), which provided a reliable multivariate (multi-pathways) toxicity profiles for the nanoparticles. The association rules identified from the multivariate toxicity profiles were then pruned to remove redundant rules. The final set of non-redundant association rules revealed pathways that are co-regulated such that “significant regulation” of one or more pathways implied the existence of other (associated) specific pathways that are also significantly regulated. In order to explore the validity of data-driven hypothesis generation, specific pathway-activity studies were carried out with ZnO nanoparticles confirming the association rule identified for the p53 and Mitosox pathways. This pathway relationship revealed that blocking of the transcriptional activity of p53 lowered the Mitosox signal (i.e. p53 leads to mitochondrial dysfunction). The present approach to data-driven hypothesis generation (i.e., association rules discovery) has important implications for streamlining multi-parameter HTS assays, support for improved understanding of toxicity mechanisms, and selection of endpoints for the development of nano-Structure Activity Relationships (nano-SAR).