Identification and characterization of dengue virus target cells in human and murine skin

Dengue virus (DENV) is the causative agent of dengue, which affects approximately 96 million people per year. DENV is an arbovirus that is transmitted by Aedes mosquitoes; thus, host cells in the skin are the first point of contact with the virus. Using normal human skin, we identified antigen-presenting cells (APCs) as the major target cells for DENV infection. A detailed analysis of all known skin APC subsets revealed that epidermal Langerhans cells (LCs), CD14+ dermal dendritic cells (DCs), CD1c+ DCs and dermal macrophages, but not CD141+ DCs, are infected by DENV upon exposure in a single-cell suspension. Blocking of the previously described DENV attachment factor DC-SIGN did not inhibit viral replication in any of the subsets. When comparing the ability of infected DC subsets to produce viral progeny, we identified LCs as the major source of viral load. This efficient production of viral progeny correlated with an early onset of infection and a reduced sensitivity to IFNβ, compared to the other infected APC subsets, which showed slower infection kinetics, lower viral progeny and higher sensitivity to IFNβ Gene expression analysis showed that exposure to DENV caused a significant upregulation of IFNβ, STAT-1 and CCL5 in infected DC subsets. Infection had no impact on their ability to migrate towards CCL19, a chemokine expressed in lymph nodes (LNs) attracting CCR7-expressing cells –including skin DCs- from the periphery. This suggests that infected skin DCs are able to transport virus from the periphery to the skin-draining lymph nodes. Addressing the role of skin antigen-presenting cells in antibody-dependent enhancement (ADE), a phenomenon that is well described for DENV, we observed pronounced ADE only in CD14+ and CD1c+ DCs but not in LCs, correlating with the expression of FcγRIIA/CD32A. To confirm our human skin results in an in vivo model, we infected IFNAR-/- mice with DENV intradermally. We detected viral replication in skin-resident DCs as well as in migratory DCs found in the draining LN. Additionally, we observed a pronounced infiltration of monocytes into the infected skin tissue. These cells became targets for the virus upon differentiation into monocyte-derived DCs in the skin. We further dissected the role of skin DCs in vivo by using double knock out (DKO) mice lacking both the IFNAR and the CCR7 receptor, the latter being indispensable for skin DC migration to the lymphoid system. DKO mice showed higher levels of viremia accompanied by increased morbidity and lethality compared to IFNAR-/- mice. Rates of DC infection in the skin of these mice were similar to IFNAR-/- mice. However, skin DCs were absent in the draining LN. Despite the lack of skin DCs, DKO mice developed higher DENV-specific T cell responses than IFNAR-/- mice, possibly leading to immunopathological effects reflected in the increased morbidity. T reg cells, which control effector T cell proliferation, were slightly decreased in the LN of DKO mice, potentially accounting for an uncontrolled T cell activation. In summary, DC and/or monocyte/monocyte-derived DC infection in the skin might initiate systemic infection while at the same time providing a way to counteract virus dissemination through antigen presentation and activation of the adaptive immune response, which seem to be disregulated in DKO mice due to effects that have to be further addressed. Collectively, the findings obtained from this thesis provide insight into the early events during DENV infection both in its natural host and in a murine animal model.​Doctor of Philosophy (SBS