Modelling and Analysis of TLR3 Signalling Pathway and Protein-Protein Interaction Studies to Understand the Mechanism of Dengue Virus Mediated Interferon Inhibition.

Abstract

Dengue virus (DENV) infection which is characterized by dengue fever (DF), dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS), is a pathophysiological condition which tends to arise due to failure of various immune mechanisms, including interferon (IFN) dependent immune response. Numerous studies have now strongly established the role of Toll like receptor (TLR)3 dependent IFN response in early clearance of DENV to avert pathogenesis. Additionally, the role of IFN in switching over of innate immune response to adaptive immune response and maintenance of pro and anti-inflammatory cytokine homoeostasis is also a key factor in the determination of a course of pathogenesis of DENV infection and is also extensively reported in literature. In this thesis, the inhibitory mechanism of Toll like receptors (TLR)3 dependent type I interferon (α & β) during dengue virus infection has been investigated by applying systems-level computational approaches in order to get insights into the possible mechanisms and to elucidate the drug intervention point which can help to curtail the progression of disease. Furthermore, non-structural (NS) proteins of DENV also plays a crucial role in pathogenesis. The JAK/STAT pathway as an allied signalling cascade of TLR dependent IFN production is also very important for curtailment of DENV pathogenesis. Literature extensively reported the role of DENV NS proteins in abruption of JAK/STAT pathway at various points. STAT1 and STAT2 are important adopters in the JAK/STAT signalling and any impediment in their signalling pattern ultimately leads toward decrease IFN and various other resultant allied responses which can be catastrophic for immunity. In the first part of the this study qualitative formalism of René Thomas has been utilized to model the biological regulatory network of Toll like receptor 3 initiated signalling pathway related to the dengue pathogenesis. Logical parameters for the qualitative modelling were deduced, utilizing a model checking method applied 1

in SMBioNet. In order to incorporate the stimulation and suppression time delays in the qualitative model, a linear hybrid model, parametric linear hybrid automaton was developed. The qualitative model was able to demonstrate all the potential expression dynamics of proteins as paths, some of which was found as abstract cycles (showing homeostasis) and diverge cycles leading to stable states. The analysis of the qualitative model illustrates the significance of SOCS proteins in enhancing the propagation of dengue virus through suppression of type 1 interferons. Further studies with HyTech tool gave real-time constraints (delay constraints) of the proteins related to cyclic paths of the regulatory network supporting the evidence given by the qualitative analysis. The delay constraint analysis demonstrated that the regulation of SOCS is important in homoeostasis. The non structural proteins of DENV inhibit IFNα and β in human cells. Various studies implicated the development of resistance against interferon (IFN) mediated antiviral effect due to proteasomal degradation of Signal Transducer and Activator of Transcription (STAT) 2 which is mediated by the Non-Structural protein 5 (NS5) of dengue virus. Seven in Absentia Homologue (SIAH) 2 is a host protein, which has been reported to mediate the ubiquitination of various proteins, it has also been implicated to interact with NS5. In the second part of this thesis a comprehensive computational analysis, including docking and physicochemical studies, has been carried out for the characterization of protein-protein interactions between NS5, SIAH2 and STAT2. The objective of the analysis is to gain an insight into the residues and sites of interaction between these proteins. Comprehensive docking and physicochemical analysis revealed that NS5 of DENV may first interact with the host SIAH2 and this complex then bind with STAT2 via SIAH2. It has been observed that the regions of NS5 comprised of β-hairpin and amino acids with the range 740-747 offer major interactions with both SIAH2 and STAT2. Additionally the most significant interaction between NS5 and SIAH2 has been observed within residues Cys160 and His64 of SIAH2 with Trp474 and Ser747 of NS5 (strong charge complementarity), which helps to hold on the com- 2 plex strongly. These implications have been reported for the first time and further require validations by web-lab studies. 3