The human hepatitis B e antigen targets and suppresses toll-like receptor signalling pathways

HBV infection is characterised by a prolonged ‘immunotolerant hepatitis B e antigen (HBeAg)-positive phase’ where patients are persistently infected with high viral loads, yet they demonstrate no overt inflammation or liver damage. This represents a highly evolved host-parasite relationship. Viruses are known to target the host innate immune system as a means of establishing infection and maintaining chronicity. In humans, the innate immune system recognises pathogens via pattern recognition receptors such as the Toll-like receptors (TLRs), initiating a pro-inflammatory responses. TLRs and associated signalling proteins are characterised by their transmembrane and cytoplasmic Toll-IL1 receptor (TIR) domain. The TIR domain is critical for homotypic TIR:TIR interactions with downstream adaptor proteins leading to activation of downstream transcription factors Nuclear Factor-kappa B (NF-κB) and interferon regulatory factor (IRF)-3. Previous studies have demonstrated that both TLR expression and pro-inflammatory cytokine production is reduced in HBeAg-positive patients compared to HBeAg-negative patients following TLR stimulation. HBeAg is a secreted, non-particulate version of the HBV nucleocapsid protein, the hepatitis B core antigen (HBcAg). It is transcribed, translated and secreted very early in the HBV replication cycle, although importantly, 20-30% of HBeAg is retained in the cytoplasm. No known role is ascribed to this cytosolic HBeAg. The aim of this study was to investigate interactions between TLR signal transduction pathways and the mature HBeAg protein localised in the cytosol. In this thesis I have provided a novel molecular mechanism illustrating how HBeAg inhibits TIR-mediated inflammatory response through direct interaction with a sub-set of TIR-containing proteins. Bio-informatics analysis of HBeAg identified a novel N-terminal 10 amino acid sequence referred to as the Precore Specific Sequence (PSS) distinct from HBcAg, as homologous to the TIR motif found in TLR and associated signal transduction proteins. I have demonstrated by co-localisation and immunoprecipitation that HBeAg, but not HBcAg, is able to interact with a sub-set of membrane proximal TIR-proteins including MyD88 adaptor-like (Mal) and Toll-like receptor adaptor molecule-2 (TRAM). This specificity of this interaction between HBeAg and TIR-containing proteins is demonstrated by the requirement of three critical amino acids “LCL” within the PSS, as mutation of these amino acids ablates all interaction and exerts no inhibitory properties. Critically interactions between HBeAg and TIR-containing proteins were specific as no interaction was observed with MyD88 as illustrated by co-immunoprecipitation or co-localisation studies. Luciferase reporter assays were employed to demonstrate the biological consequences of HBeAg interaction with TIR-containing receptor and adaptor proteins. Co-expression of HBeAg with TLR-2, Mal, TRAM and TIR domain containing adaptor inducing interferon (TRIF) resulted in the significant suppression of both NF-κB and Interferon-β (IFN-β) activation. Consistent with previous findings, NF-κB activation was unaffected by co-expression of HBeAg with MyD88. Importantly, HBcAg which lacks the PSS and the HBeAg mutant had no effect on downstream activation of NF-κB or IFN-β. HBeAg appears to function by sequestering TIR-containing proteins from homotypic TIR:TIR signalling complexes as co-expression. This was illustrated by co-expression of HBeAg inhibiting homotypic TIR:TIR interactions between Mal and MyD88, therefore disrupting downstream signalling. There are several genotypes of HBV found globally and previous studies have reported variation in responsiveness to IFN-β treatment between genotypes. I wished to determine if there were differences in modulation of innate responses amongst genotypes. Interestingly it appears that genotypes A and B suppress predominantly NF-κB mediated responses, whereas genotypes C and D are were found to significantly inhibit both NF-κB and IFN-β mediated responses. Furthermore, TLR-2-induced Interleukin-6 (IL-6) expression is reduced in HBV genotypes C and D compared to genotypes A and B following treatment with IFN-β and TLR-2 agonist Pam3Cys. Critically my findings demonstrate that HBV genotype significantly influences the degree of immuno-modulation on TIR-mediated responses. These findings correlate with previous clinical studies which show individuals with genotypes A and B have higher response rates to IFN treatment facilitating seroconversion compared to genotypes C and D. Critically I have provided a mechanism describing how administration of IFN enhances host immunity in CHB individuals which may explain differences in responsiveness to IFN treatment between HBV genotypes. Overall HBV genotype is an important predictor of response to IFN-based therapy in CHB infection. For best clinical practice and treatment outcome it is strongly recommend that determination of HBV genotype is established before commencing treatment. In conclusion findings presented in this thesis provide a novel molecular mechanism as to how HBeAg interacts with TLR signal transduction pathways to immuno-modulate TIR-mediated innate responses. Importantly the discovery of the PSS, which acts to negatively regulate TLR signalling, may be used in future designs for a novel therapeutic against not only HBV but other TLR-inflammatory diseases circumventing chronicity and poor clinical outcome.