Dengue virus non-structural protein NS1: Structure, intracellular trafficking and role as a diagnostic marker

David Muller (2010). Dengue virus non-structural protein NS1: Structure, intracellular trafficking and role as a diagnostic marker PhD Thesis, School of Chemistry & Molecular Bioscience, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s364309_PhD_finalthesis.pdf Final thesis application/pdf 38.47MB 49
Author David Muller
Thesis Title Dengue virus non-structural protein NS1: Structure, intracellular trafficking and role as a diagnostic marker
School, Centre or Institute School of Chemistry & Molecular Bioscience
Institution The University of Queensland
Publication date 2010-10
Thesis type PhD Thesis
Supervisor Paul Young
Total pages 251
Total colour pages 50
Total black and white pages 201
Subjects 06 Biological Sciences
Abstract/Summary The dengue virus protein NS1 is the first of seven non-structural proteins encoded for by the viral genome. It is a 46 kDa glycoprotein that contains 12 completely conserved cysteines that form 6 disulfide bonds. NS1 exists in several biophysical forms. It is secreted as a soluble hexamer, is present on the surface of infected cells and is thought to form part of the viral replication complex. Work presented in this thesis extends our current understanding of the dynamics of NS1 kinetics and associated immune responses during dengue infection, provides a structural model of secreted NS1 and determines the pathways involved in NS1 synthesis and trafficking. The dengue virus NS1 protein is secreted during the natural course of infection with elevated levels of free secreted NS1 found in patient plasma during the early phase of disease. High levels of NS1 early during infection have been suggested as a predictor of progression to more severe disease in secondary infected patients. To investigate the immune responses associated with dengue virus infection, a prospective study was performed on a cohort of 52 patients from Vietnam, infected during epidemic outbreaks occurring between 2005-2007. Multiple samples were taken from each patient during disease development (1-9 days) and convalescent serum samples were taken several weeks after resolution of illness. Patients had symptoms ranging from mild dengue fever through to DHF/DSS. Each sample was examined in an ELISA format for the presence of NS1, anti-NS1 IgG, anti-E IgM and anti-E IgG. NS1 was detected in patient samples from the first day of illness with levels peaking between days 1 and 4. The maximum level of NS1 detected in this study was 7.8 µg/ml. Patients experiencing a secondary dengue infection showed a strong anamnestic anti-NS1 IgG response that was often associated with a loss in detection of NS1, possibly suggesting the formation of immune complexes. Attempts of dissociate theses immune complexes failed to increase NS1 suggesting that NS1 bound in immune complexes maybe cleared rapidly or may be deposited on capillary endothelial walls where this may contribute to disease pathogenesis. Anti-E IgM was detected in most patients from the third day of illness with detection sensitivity increasing thereafter for both primary and secondary infected patients. Anti-E IgG was rarely detected in primary patients. However in secondary infection, anti-E IgG was detected from the third day of illness. When results were compiled, it was clear that the NS1 capture ELISA is at its most effective early in infection, from the onset of symptoms, whereas IgM and IgG ELISAs rely on the patients induced immune response, which in most cases appears from day 3 or 4. By the time IgM ELISAs are performing at a high diagnostic rate, NS1 detection is rapidly declining. The conclusion from this study was the need for complementary testing of both NS1 and antibody levels, as this testing regime was found to cover the disease period with greater than 80% successful diagnosis. As the crystal structure of full-length dengue virus or any Flavivirus NS1 remains elusive, several attempts were performed at crystallizing a stable, trypsin resistant carboxy-terminal NS1 domain. The trypsin resistant NS1 fragment was produced by expressing full length NS1 in a recombinant baculovirus system followed by trypsin digestion. Although micro crystals were obtained on several occasions, unfortunately these crystals did not diffract and attempts at using them to seed further crystallization failed. In the absence of a crystal structure, single particle analysis (SPA) was performed using negative stained images of secreted baculovirus derived NS1 imaged by electron microscopy. Several 3D models were reconstructed from 2778 NS1 particles with different symmetries implied. Following careful scrutiny a final self-consistent model with 3-fold rotational symmetry at a resolution of 23 Å was constructed, providing an insight to the overall size, shape and structure of NS1. The final 3D model suggests that secreted NS1 is composed of a trimer of dimers this is also supported by chemical cross-linking with BS3 which it was observed that the major species of cross-linked NS1 corresponded to the dimer, tetramer and hexamer. Whether it is destined for secretion or for insertion in the infected cell plasma membrane, little detailed information is known about the cellular pathways along which NS1 is trafficked. This study has shown that membrane-associated NS1 appears to traffic to the plasma membrane of mammalian cells (Vero cells) as a high mannose, dimeric species via an alternative pathway that bypasses the Golgi. This conclusion was made based on the effects of several classical secretion pathway blockers such as BFA, monensin and nocodazole as well as the effects of the cholesterol depleting agent MβCD. Passage of NS1 to the cell surface could not be completely blocked by any of the inhibitors tested. However, NS1 secretion was totally blocked by BFA and MβCD while monensin and nocodazole reduced secretion to varying degrees. The loss of NS1 secretion from infected Vero cells treated with MβCD resulted in a corresponding increase in NS1 at the plasma membrane suggesting that cholesterol may be involved in the formation of the secreted NS1 species. Upon investigation of NS1’s amino acid sequence a cholesterol recognition/interaction amino acid sequence (CRAC) motif was found. As a result we propose that dimeric NS1 associates with cholesterol though the CRAC motif. This interaction causes a localized concentration effect of NS1 which causes NS1 dimers self associate resulting in hexamer formation. The results of this study have shown that NS1 has an intimate association with cholesterol, a finding which is under further investigation.
Keyword flavivirus
dengue virus
ns1
lipid rafts
patient immune responses
ns1 structure
cholesterol
methyl-beta-cyclodextrin
single particle analysis
Additional Notes Colour pages 31, 37, 44, 45, 54, 88, 90-97, 100,105-115, 117, 119-121, 132, 137, 138, 146, 156, 164, 166, 174-176, 178, 179, 181-183, 185, 186, 202, 204

 
Citation counts: Google Scholar Search Google Scholar
Access Statistics: 1295 Abstract Views, 49 File Downloads  -  Detailed Statistics
Created: Thu, 07 Apr 2011, 15:18:09 EST by Mr David Muller on behalf of Library - Information Access Service