An Integrative Workflow to Prioritize Toxicity Initiation Events From Air-liquid-interface Exposure of Human Lung Cells to Combustion Aerosols: First Application to Emissions From a Heavy Fuel Oil or Diesel Fuel Operated Ship Engine

Abstract BackgroundAdvancement in the instrumentation techniques made it possible to identify the compositions and concentrations of various chemicals, such as organic compounds or metals in combustion aerosols. However, the prediction of combustion aerosol-induced toxicity end points in lung epithelium cells is difficult due to the large number of non-linear chemical-biological interactions. It is evident that some chemicals present in the combustion aerosols upregulate certain genes/ proteins while other chemicals downregulate them, making the prediction of toxicity of a mixture of chemicals a challenging task. Also, the presence of large numbers of feedback and feedforward regulatory loops makes the entire prediction process highly dynamic. MethodsHere we present an integrative workflow to construct and analyze combustion aerosol-type specific chemical-gene regulatory network. For this, we develop an algorithm to estimate the combined regulatory effect of chemicals present in the combustion aerosol on each of the genes/proteins in the network. Further we rank the nodes by combining various network topological and non-topological parameters using a multi-objective optimization function. The top ranked nodes were used to identify aerosol-type specific key regulators that contributes toward the understanding of adverse outcomes due to the exposure of combustion aerosols. ResultsThe integrative workflow is evaluated using transcriptomics analysis carried out on human bronchial epithelium cell line BEAS-2B exposed at the air liquid interface (ALI) to the combustion aerosols generated by ship engine running either on distillate Diesel Fuel (DF) or on Heavy Fuel Oil (HFO) along with chemometric analysis. Based on the prioritized chemicals, we prepared DF and HFO combustion aerosols-specific chemical-gene regulatory network. Our study reveals the large differences in the observed biological effects caused by operating the same ship engine on the different fuels.ConclusionThe presented workflow can be used to investigate key regulatory processes associated with the toxicity outcomes of mixture of chemicals and also for the hazard classification and assessment of various combustion aerosols.