Adopting a gene regulatory network approach to investigate toxicity through the adaptive stress response in teleost fish species.
Date: 23 April 2019
University of Exeter
PhD in Biological Sciences
Given current risks of pollutant exposures in aquatic environments, there is a growing need to generate reliable computational risk assessment methods to establish how adverse outcomes can be produced across exposure organisms. The adaptive stress response is widely targeted by pollutants of concern and includes transcription factors ...
Given current risks of pollutant exposures in aquatic environments, there is a growing need to generate reliable computational risk assessment methods to establish how adverse outcomes can be produced across exposure organisms. The adaptive stress response is widely targeted by pollutants of concern and includes transcription factors including nuclear factor (erythroid- derived 2)-like 2 (Nrf2), hypoxia inducible factor (HIF-1α), heat shock factor (HSF1), nuclear factor kappa-light-chain-enhancer of activated -B cells (NFkB), metal transcription factor 1 (MTF1), the aryl hydrocarbon receptor (AhR) and tumor protein P53 (P53). While these TFs are known to be activated by distinct inducers, less is understood about the regulatory links between factors, particularly at the transcription factor (TF) DNA-binding level. In this thesis, a gene regulatory network (GRN) of adaptive-stress response factors that are key targets of chemical toxicity was constructed based on experimental evidence from mammalian cell-lines. The GRN was modeled using boolean logic and this identified a number of response outcomes that could be attributed to the activation of pathways including antioxidant defence processes and glucose metabolism. The GRN model illustrated that the activation of Nrf2, HIF-1α, AhR, MTF1 and HSF1 led to the same adverse outcomes, suggesting canalisation in stress response pathways. The ability to use GRNs across different species is widely supported by the identification of TF binding sites (TFBS) within target genes. To assess the efficiency of using the mammalian GRN across teleost fish species, a comprehensive analysis of validated binding sites for the AhR, MTF1, HIF-1α and Nrf2 was conducted across fish-species in comparison to the mammalian consensus binding sequence. This showed variations in binding site composition across validated TFBS for HIF-1α and Nrf2 in fish compared to the mammalian consensus, preventing the identification of the functional sequences for these factors using traditional methods. To establish if such changes affected the efficiency to predict positive downstream target genes for Nrf2 and HIF-1α in mammals and across teleost fish species, random 1 forest classification models were used to compare the efficiency of multiple positional weight matrices (PWM) motifs of TFBS for Nrf2 and HIF-1α. Whilst the result from this analysis identified discrepancies in the ability to predict target genes based on the mammalian motif file used, mammalian motifs were able to predict target genes across fish species. Validated binding sites in fish species were then aligned to generate PWM motifs and sites were predicted across shared target genes hsp70 and hmox1 using both fish based and mammalian based models. This showed that whilst there was some overlap in identified sites across species, fish-specific motifs identified unique sites from mammalian models. To validate the GRN, gene-expression responses across exposures traditionally associated with activating distinct adaptive stress response factors were collated across the literature. This showed support for some of the key responses identified in the model. Chemical exposure studies were then undertaken in vivo in embryo-larval zebrafish (2 and 4 dpf) to identify potential connectivity between the TFs NFkB, MTF1 and HIF-1α with Nrf2, a key factor in the adaptive stress-response and a regulator of antioxidant response processes. The inducer of Nrf2, tert-butylhydroquinone (tBHQ), was used to determine if there was a change in transcriptional output of mtf1, hif1a and nfkb1 over time and with exposure concentration. This showed a significant difference in expression for nfkb1 and alterations in expression of mtf1 over prolonged exposure scenarios. In addition, the developmental expression of nrf2a, mtf1, hif1α and nfkb1 from 2 hpf to 96 hpf showed differences between transcript levels with hif1α and nfkb1 having the highest levels of expression compared to nrf2a and mtf1. Overall, the research presented in this thesis provides a novel approach to assess the initiation of adaptive stress-response factors from molecular interactions. The research goes some way in establishing the feedback loops and connections between NFkB, MTF1, Nrf2, AhR, HIF-1α, HSF1 and P53. In doing so, the model generated in this thesis provides a novel approach of establishing outcomes under toxicant exposures.
Item views 0
Full item downloads 0