The role of lymphotoxin ligand-receptor interactions in regulating epithelial cell fate

LTβR and HVEM are non-death domain-containing TNFRs that can induce cell death via possible recruitment of TNFR-associated factors (TRAF), thus may share similarities to other TNFR members (e.g. CD40). This thesis aimed to investigate the effects of soluble LT agonists on a panel of carcinoma cells of colorectal (CRC) and bladder (UCC) origins and to compare the ability of these agonists to induce cell death against membrane-bound LIGHT (mLIGHT), and to unravel for the first time the cell signalling pathways responsible for mLIGHT-mediated cell death.

Due to the complexity of some of the approaches used, a significant part of the experimental work involved optimisations involving not only soluble LT agonists, cytokines, specific pharmacological inhibitors but mainly optimisation for the first time of a co-culture system for the delivery of the mLIGHT signal to epithelial cells (involved co-culture of target cells with growth-arrested third-party L cells expressing surface mLIGHT). Several assays were also optimised for detection of cell viability, cell death (based on protease release, caspase activation and DNA fragmentation) and for detection of pro-inflammatory cytokine secretion. Moreover, immunoblotting techniques were optimised and utilised for detection of proteins associated with intracellular LTβR and HVEM-signalling. Transfection experiments using specific small interfering RNAs (siRNAs) were also employed to knockdown the expression of LTβR and HVEM proteins in CRC and UCC cells.

This project revealed for the first time that normal human urothelial cells (NHU), CRC and UCC cells express LTβR and HVEM, and that the activation of LTβR and HVEM by mLIGHT, in the absence of IFN-γ, is pro-apoptotic in carcinoma cells, whereas mLIGHT appeared to be cyto-protective in NHU cells. By contrast, soluble LT agonists were weakly pro-apoptotic and required IFN-γ to kill HT29 cells, yet this combination did not kill other, well-characterised carcinoma cell lines, in particular HCT116 and EJ cells. Moreover, mLIGHT caused some DNA fragmentation in HCT116, yet little DNA fragmentation was detected in HT29 and EJ cells. It was also found that mLIGHT caused IL-8 and GM-CSF secretion. mLIGHT triggered TRAF1 and TRAF3 induction and caused little detectable differences in phospho-ERK, -JNK and -p38 expression in CRC and UCC cells. Functional inhibition experiments showed that blockade of MEK/ERK abrogated death in all cell lines tested, and JNK inhibition attenuated death in HCT116 and EJ (but not HT29 cells) and p38 inhibition significantly attenuated, but not fully, mLIGHT-mediated cell death in CRC and UCC cells. Moreover, an NF-κB inhibitor partially reduced mLIGHT-mediated death in CRC cells and potentiated it in UCC cells, whereas, inhibition of AP-1 partially blocked mLIGHT-mediated death in HCT116 and EJ cells. By contrast, AP-1 blockade did not cause any statistically significant effect in mLIGHT-mediated death in HT29 cells. Moreover, mLIGHT-mediated death is ROS dependent in CRC and UCC cells as the antioxidant NAC attenuated death. The current work has also provided evidence for the first time that a role for NOX enzyme in cell death of HCT116 and EJ cells as it was found that mLIGHT induced the phosphorylation of p40phox (a subunit of NOX). Importantly, despite observing that ASK1 was activated in HCT116 cells, but not other cells, mLIGHT caused downregulation of Thioredoxin-1 expression in CRC and UCC. siRNA experiments for LTβR and HVEM knockdown showed some preliminary evidence that LTβR and HVEM might signal cooperatively in the context of LIGHT-mediated cell death.

Collectively, this thesis has demonstrated for the first time that triggering cell death in CRC and UCC is clearly dependent on signal quality, cell-type specificity and death is tumour cell-specific. The current study has also provided some mechanistic insight into how cell death induced by mLIGHT-LTβR/HVEM interactions occurs, which may involve a novel pathway of receptor-TRAF3-MAPK-NOX interactions, which utilise ROS for the activation of cell death pathways in CRC and UCC cells. These findings have not only improved our understanding of how TNFRs induce carcinoma cells death, but may also help in the design of better therapeutic strategies in the future.