Identification of dengue virus (DENV) NS1 protein residues involved in its cellular secretion and host factors involved in NS1 protein N-glycosylation

Abstract

Dengue virus (DENV) is a Flavivirus of the Flaviviridae family of (+) RNA viruses. It is known to cause approximately 390 million infections and 25,000 deaths annually in tropical and sub-tropical areas across the world. DENV infection has a wide spectrum of clinical manifestations ranging from mild fever to severe forms of dengue; formerly referred to as dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS). The viral NS1 protein is secreted from DENV-infected cells as a soluble hexameric lipoparticle that can induce vascular leakage; an established hallmark of DENV disease pathogenesis. Despite widespread acknowledgement of the importance of NS1 secretion in DENV pathogenesis, the exact molecular features of NS1 that are critical for its secretion from infected cells are not fully characterised. In this study, we employed random point mutagenesis and a luminescent ‘HiBiT’ peptide tag assay approaches to identify NS1 residues that are essential for its efficient secretion. From these approaches, we identified 10 point mutations that correlated with impaired NS1 secretion, with subsequent in silico analyses demonstrating that the majority of these mutations are located within the β-ladder domain of NS1. More detailed studies on two of these mutations V220D and A248V, revealed they also abrogated viral RNA replication and infectious virus production. Subsequent analyses of these mutants by confocal microscopy in the context of a DENV non-structural protein expression system revealed a similar but more reticular NS1 localisation pattern, while these mutant NS1 proteins could not be detected by immunoblotting using a conformation specific NS1 monoclonal antibody. Together, these studies demonstrated that the V220D and A248V mutations of NS1 may disrupt multiple functions and properties of the protein, as might occur if it is improperly folded and/or cannot form critical interactions with other proteins or membranes. Next, we sought to identify novel host factors associated with NS1 N-glycosylation, an essential post-translational modification of NS1, using an APEX2-based proximity biotinylation labelling approach and quantitative proteomics. Towards this goal, appropriate virus constructs were generated for wildtype and N-glycosylation NS1 mutants. Following the comprehensive testing of a variety of constructs and optimisation of transfection and proximity biotinylation conditions, large scale experiments were conducted, and lysates were prepared for future streptavidin pulldowns and quantitative proteomics analyses. Taken together, these studies employed a combination of random point mutagenesis and sensitive luminescent peptide assays to identify a panel of mutations that impair NS1 secretion activity. A novel proximity labelling-based experimental workflow was also developed and applied towards proteomics-mediated identification of NS1 proximal proteins and how the NS1 protein microenvironment is altered by mutational disruption of NS1 N-glycosylation. We propose that NS1 secretion and N-glycosylation may represent novel and viable targets for future antiviral drug and attenuated vaccine development.