Ebola virus (EBOV) causes severe human disease. During late infection, EBOV virions are on the skin’s surface; however, the permissive skin cell types and the route of virus translocation to the epidermal surface are unknown. We describe a human skin explant model and demonstrate that EBOV infection of human skin via basal media increases in a time-dependent and dose-dependent manner. In the dermis, cells of myeloid, endothelial, and fibroblast origin were EBOV antigen–positive whereas keratinocytes harbored virus in the epidermis. Infectious virus was detected on the apical epidermal surface within 3 days, indicating that virus propagates and traffics through the explants. Purified human fibroblasts and keratinocytes supported EBOV infection ex vivo and both cell types required the phosphatidylserine receptor, AXL, and the endosomal protein, NPC1, for virus entry. This platform identified susceptible cell types and demonstrated dynamic trafficking of EBOV virions. These findings may explain person-to-person transmission via skin contact.

Ebola virus (EBOV), the causative agent of Ebola virus disease, remains one of the World Health Organization’s top 10 threats to global health. Infectious EBOV virions can be found on the surface of skin late in infection and may be transmitted to others through skin-to-skin contact. We investigate in vivo EBOV tropism and the kinetics of virus movement to and from the skin. Increasing viral loads were detected over time in the skin of EBOV-infected non-human primates and mice, with antigen detected in dermal stromal and immune cells. Epidermal cells within and surrounding hair follicles also harbored viral antigen, suggesting a novel mechanism of virus egress to the epidermal surface. During late infection, proinflammatory responses were elevated in infected visceral organs but minimal in the skin despite significant viral loads. We observed similar viral trafficking and cell tropism in the skin of mice intraperitoneally infected with a low containment EBOV model virus, rVSV/EBOV GP, allowing more detailed mechanistic studies. Sites of virus infection in the skin were patchy, with intense focal areas of infection surrounded by uninfected areas. To investigate virus entry into the body through skin, rVSV/EBOV GP was applied to the surface of gently abraded skin to remove the stratum corneum; epidermal keratinocytes were robustly infected with subsequent systemic viral dissemination observed in some mice. Optimal levels of infection within the skin required expression of the phosphatidylserine receptor, AXL. Collectively, our data demonstrate that skin serves as an important organ targeted by EBOV, facilitating virus entry into and egress from the body. IMPORTANCE: Ebola virus (EBOV) remains one of the World Health Organization’s top 10 threats to global health, despite the availability of a U.S. Food and Drug Administration-approved vaccine. EBOV spreads through human-to-human contact, yet the role of skin in viral transmission remains unclear. Here, we identify skin as a site of EBOV infection, serving as a potential portal for entry into and egress from the body. In vivo , infectious virions and viral RNA increased in the skin over time, localizing to dermal myeloid and stromal cells and to cells within and surrounding hair follicles, suggesting a novel mechanism for viral shedding. Skin infection was patchy and associated with minimal inflammation, despite significant viral loads. Using a surrogate EBOV model, we demonstrate that systemic infection can occur following topical administration through abraded skin and requires phosphatidylserine receptor, AXL, for optimal infection of skin. These findings redefine the role of skin in EBOV pathogenesis, with implications for barrier-targeted interventions.

Macrophages are critical in the pathogenesis of a diverse group of viral pathogens, both as targets of infection and for eliciting primary defense mechanisms. Our prior in vitro work identified that CD40 signaling in murine peritoneal macrophages protects against several RNA viruses by eliciting IL-12, which stimulates the production of interferon gamma (IFN-γ). Here, we examine the role of CD40 signaling in vivo. We show that CD40 signaling is a critical, but currently poorly appreciated, component of the innate immune response using two distinct infectious agents: mouse-adapted influenza A virus (IAV, PR8) and recombinant VSV encoding the Ebola virus glycoprotein (rVSV-EBOV GP). We find that stimulation of CD40 signaling decreases early IAV titers, whereas loss of CD40 elevated early titers and compromised lung function by day 3 of infection. Protection conferred by CD40 signaling against IAV is dependent on IFN-γ production, consistent with our in vitro studies. Using rVSV-EBOV GP that serves as a low-biocontainment model of filovirus infection, we demonstrate that macrophages are a CD40-expressing population critical for protection within the peritoneum and T-cells are the key source of CD40L (CD154). These experiments reveal the in vivo mechanisms by which CD40 signaling in macrophages regulates the early host responses to RNA virus infection and highlight how CD40 agonists currently under investigation for clinical use may function as a novel class of broad antiviral treatments.

Ebola virus (EBOV) within the Filoviridae family causes severe human disease. At late stages of infection, EBOV virions are found on the surface of patients’ skin; however, the permissive cell types within the skin and how infectious virus translocates to the apical skin surfaces is not known. Here, we describe a human transwell skin explant culture model and show that EBOV infection of human skin tissues via the basal media results in a time- and dose-dependent increase in infectious virus in dermal and epidermal tissue. Infectious virus was detected on the apical epidermal surface within 3 days, indicating that the virus propagates within and traffics through the tissue. In the dermis, EBOV-infected cells were of myeloid, endothelial and fibroblast origins, whereas keratinocytes harbored virus in the epidermis. Complementary studies showed that both purified skin fibroblasts and keratinocytes supported EBOV infection ex vivo and that both cell types required the phosphatidylserine receptor, Axl, and the endosomal protein, NPC1, for virus entry. Our experimental platform identified new susceptible cell types and demonstrated dynamic trafficking of EBOV virions that resulted in infectious virus on the skin surface; findings that may explain person-to-person transmission via skin contact. Teaser: Using a human skin explant model, these studies identify and characterize skin cell populations that support Ebola virus infection.

Ebola virus (EBOV), the causative agent of Ebola virus disease (EVD), remains one of WHO’s top ten threats to global health. Infectious EBOV virions can be found on the surface of skin late during systemic infection and passed from the deceased through skin-to-skin contact. Here, we assess viral load and antigen expression in the skin of EBOV-infected non-human primates (NHP) and mouse adapted-EBOV (ma-EBOV) - infected mice and use the low containment viral model, rVSV/EBOV GP, to mechanistically define skin infection in mice. Viral RNA peaked within the skin proximal to the site of injection in EBOV-infected NHPs on day 6. In contrast, mouse skin sites distal to the site of ma-EBOV injection achieved maximal viral loads by day 3. At late times of infection, viral antigen-positive cells co-localized with markers for endothelial, stromal, and immune cells in the dermis. Epidermal cells within and surrounding hair follicles also harbored viral antigen, suggesting a potential mechanism of virus trafficking to the epidermal surface. Despite robust viral infection, distal skin sites of ma-EBOV-infected mice had low expression of proinflammatory stimulated genes. A similar cellular tropism was observed in the skin of mice infected with rVSV/EBOV GP, with discrete focal areas of intense infection. When virus was applied to the surface of gently abraded skin to remove the stratum corneum, epidermal keratinocytes were robustly infected, followed by systemic viral dissemination. To define cell surface receptors critical for virus trafficking to and replication within the skin, mice lacking the phosphatidylserine receptors were infected intraperitoneally with rVSV/EBOV GP. At day 3 of infection, skin distal to the site of infection of TIM-1 knock out (KO) mice had significantly lower levels of infectious virus than the control mice, suggesting that TIM-1 is essential for efficient distribution of virus to the skin. Our findings reveal that EBOV targets specific skin cell populations at late times of viral infection and that the host receptor TIM-1 is required for optimal viral dissemination.