The role of heparan sulfate in poxvirus infections

Abstract

The heparan sulfate (HS) component of heparan sulfate proteoglycans (HSPGs) has been implicated in the initiation of several viral infections, including vaccinia virus (VACV). A cell infected with VACV releases two different forms of VACV, namely the mature virus (MV) released following the death of infected cells and which infects neighbouring cells, and the enveloped virus (EV) ejected from infected cells for long-range dissemination. The relative role of HS in the infectivity of the different forms of VACV is unclear. Furthermore, there is little known about the fine specificity of the VACV-HS interactions. Therefore, in order to develop HS based molecules that could potentially have antiviral properties against HS-dependent viral infection, VACV was used as a prototype virus to understand the structural and functional consequences of the interaction between VACV and HS. ELISA studies described in Chapter 3 were used to evaluate the specificity of the MV form of VACV for heparin, differentially sulfated HS, chondroitin sulfate (CS) A-D and hyaluronic acid (HA). Lack of appropriate EV specific antibodies meant that similar ELISA studies could not be performed for the EV form of VACV. Nevertheless, the MV form of VACV bound to immobilized heparin and highly sulfated HS (HShi) with high avidity, compared to lowly sulfated HS (HSlow). The MV particles also bound to CS A-D, however, very weakly. Furthermore, the ability of the MV rich Western Reserve (WR) strain of VACV to form plaques in vitro was affected by soluble heparin, WR plaque numbers being reduced 5-fold with an incremental increase in plaque size. The formation of plaques by the EV rich International Health Department-J (IHD-J) strain was also affected in the presence of heparin, there being a 10-fold reduction in plaque numbers, an incremental increase in plaque size and the disappearance of the trademark ‘comet’ shaped plaques. These data suggest that HS recognition plays a significant role in both MV and EV infectivity, with this recognition being more important for EV infectivity. To better understand the interaction between heparin/HS and the two forms of VACV, green fluorescent protein (GFP) expressing recombinant strains of VACV were constructed, as described in Chapter 4. Subsequent inhibition of infectivity assays, performed using soluble glycosaminoglycans (GAGs), suggested that sulfated GAGs more easily inhibited EV infections than the MV infections, with heparin and HShi being highly potent inhibitors of infection. Furthermore, the ability of the EV form of VACV to establish an infection was significantly reduced in cells treated with the HS-degrading enzyme heparanase and in cells genetically deficient in HS production, compared to the MV form of VACV which appeared largely unaffected. These findings confirmed that recognition of cell surface HS is vital for EV infectivity but less important for the infectivity of the MV form of VACV. In Chapter 5, the ability of soluble heparin/HS molecules and HS mimetics to inhibit VACV infections was further investigated to identify structural features of these molecules that are responsible for their interaction with VACV particles. The study also aimed to determine whether HS-based molecules could be used as possible antivirals against VACV and potentially against other HS-dependent viral infections. It was observed that the 2-O-sulfate of uronic acid and the 6-O and N-sulfate groups of glucosamine residues were important for VACV infectivity, with 6-O-sulfate being particularly crucial and EV infections being more dependent on these groups than MV infections. Furthermore, the length of heparin chains did not affect their ability to interact with and inhibit VACV infectivity. However, the linkages of different D-glucose-based HS mimetics had a profound effect on the ability of the sulfated molecule to inhibit VACV infections, with the order of potency being β(1g4) > α(1g6) > α(1g4) > β(1g3). Interestingly, however, a D-mannose-based sulfated oligosaccharide mixture (PI-88, Muparfostat) was identified as the only HS mimetic that was a more potent inhibitor of MV infections than of EV infections, in fact it was a stronger inhibitor of MV infections than unfractioned heparin (UFH). These data suggest that the EV and MV forms of VACV interact with different structural aspects of HS chains and that synthetic HS-based molecules could be designed with the ability to inhibit both EV and MV forms of VACV. In Chapter 6 studies are described that attempted to identify the proteins on the surface of the EV form of VACV that are responsible for the interaction of the VACV with cell surface HS. Thus, EV and MV outer membrane proteins were solubilised and identified by Western blotting using polyclonal anti-VACV antibodies. Four potential heparin-binding proteins were identified in the EV outer membrane extracts, being 150 kDa, 85 kDa, 60 kDa and 25 kDa proteins. The 150 kDa heparin binding protein was further analysed using 1D nanoLC ESI MS/MS and was found to be a poxvirus DNA directed RNA polymerase, with sequence similarity to the 65 kDa VACV F12 protein, a protein important in EV formation. Bioinformatic searches were also performed to determine possible HS-binding motifs in VACV proteins and identified the 64 kDa VACV B4R protein and a 78 kDa RNA helicase as likely candidates. Overall, it was concluded that there are multiple HS binding proteins on the outer EV envelope and that it is likely that in many instances the heparin/HS binding site(s) of these EV proteins may not be composed of linear amino acid sequences. Since both MV and EV forms of VACV bind HS, in Chapter 7 experiments are described that examined the role of heparanase in VACV spread, it being predicted that heparanase may aid spread by releasing VACV from cell surface and extracellular matrix (ECM) HS. Wild type (WT) and heparanase deficient (HPSE -/-) mice were inoculated with the WR strain of VACV via intranasal (i.n.) and intramuscular (i.m.) routes to evaluate the spread of infection in the two groups of mice. The WR strain of VACV was inoculated via the i.n. route when there was a 24 hr delay in weight loss in the HPSE -/- mice compared to the WT mice. Furthermore, this delay in weight loss correlated with a delay in the onset of disease with there being a 24-48 hr delay in the spread of infection from the primary site of inoculation to distant organs like the ovaries. Similarly, when VACV was delivered by the i.m. route, there was a 24-48 hr delay in the infection of the ovaries, although there was a similar delay in infection of the spleen, despite there being no weight loss difference. Overall, the results suggest that VACV depends on host-derived heparanase to aid its spread. Since heparanase mediated degradation of HS aids the infiltration of leukocytes with antiviral activity into sites of infection, the results obtained from the current study are contrary to the prevailing immunological paradigm. In conclusion, VACV like several other viruses interacts with cell surface HS prior to infecting cells. Furthermore, VACV relies on host-derived heparanase to degrade cell surface and ECM HS to aid its spread. Thus, synthetic HS-based molecules could be designed that could inhibit EV and MV forms of VACV from infecting cells and may simultaneously act as heparanase inhibitors and consequently prevent VACV spread. Show more