Dengue virus infectie van endotheelcellen: Pathogenese en glycosaminoglycanen als antiviraal doelwit

Dengue viruses (DENV) are single-stranded positive-sense RNA viruses that belong to the family of the Flaviviridae, together with other important human pathogens (e.g. yellow fever virus and West Nile virus). Dengue is currently considered to be the most important arthropod-borne vial disease in humans. Four genetically distinct, but serologically related serotypes (DENV 1-4) are transmitted to humans by mosquitoes of the Aedes genus in more than 100 countries in tropical and subtropical regions in South East Asia, the Americas and Africa. The principal dengue vector is Aedes aegypti, whereas Aedes albopictus functions as a secondary vector. Over 2.5 billion people live in areas that are affected by dengue and it is estimated that each year 390 million DENV infections occur. The DENV disease spectrum ranges from a self-limiting febrile illness called dengue fever (DF), to the severe and potentially life-threatening dengue hemorrhagic fever (DHF)/dengue shock syndrome (DSS). While DF symptoms include a sudden, high fever (>39°C), headache, nausea, vomiting, rash, and joint and muscle pains, DHF/DSS are characterized by increased vascular permeability, leading to plasma leakage, thrombocytopenia and coagulation disorders. Despite decades of extensive research, the mechanisms responsible for DENV-induced plasma leakage remain to be elucidated. The ease of collecting blood samples has prompted scientists to focus on the involvement of immune cells and soluble mediators. These studies have identified dendritic cells, monocytes and macrophages as the primary DENV target cells. Although endothelial cells (ECs) line the inner layer of the endothelium and regulate the exchange of gasses, fluid and solutes between the blood vessels and the surrounding tissues, an active role of the endothelium in the pathogenesis of DENV has not been considered until recently. Immunohistological studies and in situ hybridization performed on tissues from fatal DHF/DSS patients have demonstrated the presence of DENV antigens and viral RNA in ECs of various organs (e.g. lung, spleen, liver, heart and brain), indicative of viral infection and replication in ECs. Therefore the aims of this study were to (i) investigate the infection of microvascular endothelial cells by DENV; (ii) compare the response of different EC types to infection with DENV 1-4; and (iii) evaluate compounds for their ability to inhibit DENV infection of ECs. In Chapter 2 we investigated the susceptibility of primary microvascular ECs (HMVEC-d) and the immortalized EC line HMEC-1 to DENV infection, since DENV-induced plasma leakage is believed to occur at the level of the microvasculature. We showed for the first time that primary microvascular ECs are susceptible to DENV infection and replication in the absence of the well-described receptors DC-SIGN (on dendritic cells) or the mannose receptor and Fcγ-receptor (on monocytes/macrophages). The ability of natural glycosaminoglycans (GAGs), including heparin and heparan sulfate, to inhibit DENV-2 infection of ECs suggested the involvement of heparan sulfate-containing proteoglycans (HSPGs). Flow cytometric analysis revealed high expression levels of HSPGs on these cells and moreover, enzymatic removal of heparan sulfate moieties from the cellular surface resulted in a strong decrease in the number of infected cells. Taken together, these results indicated that HSPGs mediate DENV-2 infection of microvascular ECs. The clinical use of heparin and heparin analogues is limited because of their well-known anticoagulant activity. Therefore, we evaluated the ability of sulfated derivatives of the K5 capsular polysaccharide of Escherichia coli, which has the same structure as the natural precursor of heparin and is devoid of anticoagulant activity, to inhibit DENV-2 infection of HMEC-1 and HMVEC-d cells. The highly sulfated K5-OS(H) and K5-N,OS(H) proved to be very active compounds with EC50 values in the nanomolar range. Time-of-addition and attachment/entry assays revealed that these compounds inhibit the early steps in the DENV infection cycle. We showed that the K5 derivatives inhibit the fusion of DENV-infected C6/36 mosquito cells by interacting with the DENV envelope (E) protein. Using surface plasmon resonance (SPR) analysis we demonstrated that K5-OS(H) and K5-N,OS(H) bind to DENV envelope protein domain III (E DIII), which contains the putative receptor-binding domain. Furthermore, the highly sulfated K5 derivatives inhibited the binding of DENV E DIII to immobilized heparin, which mimics the infection process in microvascular ECs. We also demonstrated that K5-OS(H) and K5-N,OS(H) inhibit DENV-2 infection of dendritic cells. Together, these results indicate that highly sulfated K5 derivatives may provide a new class of anti-DENV compounds, although their antiviral activity against the other serotypes should still be confirmed. In Chapter 3 we evaluated the antiviral activity of chemokine peptides. As DENV infection of ECs is dependent on HSPGs, positively charged peptides may compete with DENV for binding to HS. Chemokines are dependent on GAG binding on ECs to create a chemokine gradient for the recruitment of immune cells to the site of production. Chemokine binding to GAGs is thought to be mediated by basic amino acid clusters in the carboxy-terminus and the negatively charged sulfated/carboxyl groups of HSPGs. The CXC chemokines, CXCL9 and CXCL12γ, have a characteristic basic COOH-terminus and contain multiple putative HS-binding motifs. Therefore, the binding of different synthetic COOH-terminal CXCL9 and CXCL12γ peptides to heparin was characterized using SPR analysis. These data revealed high binding affinities of the positively charged peptides to immobilized heparin, except for CXCL9(86-103) in which 2 putative HS-binding motifs are missing. Furthermore, the CXCL9 and CXCL12γ peptides were able to compete for heparin binding to CXCL8, a strong heparin binder. Since several viruses have been shown to use HSPGs as attachment receptors on target cells, the antiviral activity of the chemokine peptides was investigated against a range of such viruses. This antiviral screen showed marked inhibitory activity of the heparin-binding peptides against herpes simplex virus type 1 (HSV-1), respiratory syncytial virus (RSV) and DENV-2 in the micromolar range. SPR analysis demonstrated that CXCL9(74-103) binds faster to heparin than DENV E DIII and inhibits the interaction of DENV E DIII with heparin, hence explaining its anti-DENV-2 activity. These results indicate that the positively charged COOH-terminal CXCL9 and CXCL12γ chemokine peptides may be lead compounds for the development of a new class of antiviral agents targeting virus-HSPG interaction. In Chapter 4 we aimed to compare the effects of DENV 1-4 infection in ECs from different origin. Contradictory findings have been reported regarding the DENV infection rate and EC response. The use of ECs from different origin further complicates the comparison of the obtained results. Furthermore, most data involve DENV-2 and to a lesser extent serotype 4, whereas very few studies reported the effects of EC infection with DENV-1 and -3. Therefore, we infected the microvascular EC line HMEC-1 and primary microvascular (HMVEC-d) and macrovascular (HUVEC) ECs with four DENV serotype strains: DENV-1 Djibouti, DENV-2 NGC, DENV-3 H87 and DENV-4 Dak. DENV-4 proved to be the most infectious serotype, followed by DENV-2, DENV-1 and DENV-3. We found the highest percentage of DENV infection in HMEC-1 cells, whereas infection of primary HMVEC-d and HUVEC cells resulted in a comparable, lower infection rate. The observed differences in the infection rate between the EC types were found to correlate with the expression of HS on the surface of these cells. Using a custom-made Bio-Plex assay, we simultaneously quantified the levels of 18 cytokines involved in the regulation of EC permeability or the recruitment and/or stimulation of immune cells. DENV 1-4 infection of ECs resulted in the strong upregulation of IL-6, MCP-1, RANTES, CXCL9, CXCL10 and CXCL11. Furthermore, we observed a time-dependent cytopathogenic effect in all EC types after DENV 1-4 infection. Specific apoptosis assays revealed the translocation of phosphatidylserine to the extracellular side of the plasma membrane in HMEC-1 cells infected with DENV 1-4 and the activation of caspase-3 in DENV-4 infected HMEC-1 and HUVEC cells. To investigate the effect of direct infection of ECs on vascular permeability, we monitored the integrity of a HUVEC monolayer during DENV-4 infection in real-time. We found that DENV-4 infection alone did not alter the cell index in comparison to uninfected control cells. However, when DENV-4 infected ECs were stimulated with TNF-α 24 h after infection, a strong decrease in the cell index was observed, when compared to uninfected ECs treated with TNF-α. These results suggest that DENV infection sensitizes ECs to the permeability-enhancing effects of TNF-α, although the underlying mechanisms remain to be elucidated.

Publication year:2015

Accessibility:Closed