Citations

Abstract Orthoebolaviruses, including Ebola virus (EBOV) and Bundibugyo virus (BDBV), cause sporadic outbreaks and devastating hemorrhagic infections in humans. A significant challenge that impedes the study of filoviruses is that the infectious virus can only be handled in biosafety level 4 (BSL4) facilities due to their extreme virulence. Life cycle modeling systems that use minigenomes and transcription/replication-competent virus-like particles (trVLPs) are powerful tools for studying these virus’ biology and to explore antiviral therapies under BSL2 conditions. To date, these powerful tools have only been utilized in vitro in cell culture systems. In this work, we demonstrate the use of EBOV trVLPs in vivo. We report the fine specificity of first infection events in a model of EBOV injection in the skin. Cre recombinase expressing monocistronic EBOV trVLPs were packaged and used to inject Cre reporter mice intradermally (ID). Initial infection events were monitored in living animals by luciferase imaging and then by flow cytometry to track cell-specific expression of Cre-activated GFP. By this approach, EBOV trVLPs were shown to directly infect T cells, monocytes, dendritic cells (DCs) with notably high first infections of skin-resident Langerhans cells (LCs) in the skin barrier and its draining lymph node. EBOV trVLPs were next tested as an in vivo tool to assess antiviral therapies by testing the ability of an EBOV vaccine to reduce trVLP luciferase levels after infection in the skin. This work demonstrates first in vivo use of EBOV trVLPs and provides proof of concept for their use as an in vivo BSL2 model to more rapidly develop antiviral agents and vaccines.

Uncontrolled systemic inflammatory responses are a critical pathological feature of fatal Ebola virus (EBOV) infection. While some inflammatory responses may originate from mononuclear phagocytes (MNPs), nonimmune cells vastly outnumber MNPs and may be an important source of inflammation. Here, we demonstrated that highly virulent EBOV induced a high and sustained pro-inflammatory response compared to less virulent ebolaviruses in non-MNPs through TLR4-independent NF-κB activation. We identified the EBOV matrix protein VP40 as a potent activator of NF-κB in non-MNPs, whose intrinsic inflammatory activation ability is higher than VP40 proteins from less virulent ebolaviruses. This suggests that VP40 is a virulence determinant inducing distinct degrees of pro-inflammatory responses among ebolaviruses. Mechanistically, VP40 activated the NF-κB signaling pathway, primarily via TNFR1 using a ligand-independent mechanism. These findings reveal mechanisms that may drive systemic inflammation and promote EBOV pathogenesis, suggesting potential therapeutic strategies to mitigate immune dysregulation in severe EBOV infections.

Ebola virus (EBOV) causes severe disease outbreaks in humans with high case fatality rates. EBOV requires adaptation to cause lethal disease in mice by acquiring single mutations in both the nucleoprotein (NP) and VP24 genes. As an attempt to model mouse-adapted EBOV (MA-EBOV), we engineered novel pentacistronic minigenomes (5xMG) containing a reporter gene, VP40, and glycoprotein genes as well as the NP and VP24 genes from either EBOV or MA-EBOV. The 5xMGs were constructed and optimized, and the produced transcription- and replication-competent virus-like particles (trVLPs) were demonstrated to infect several cell lines. Introduction of the mouse-adaptation mutations did not significantly impact the replication and transcription of the 5xMG or the relative infectivity of the trVLPs in vitro. This work demonstrates the development of the 5xMG system as a new versatile tool to study EBOV biology.

Abstract Most adenovirus vaccines are replication-defective RD-Ads. In contrast, replication-competent adenovirus (RC-Ad) replicate antigen genes 10,000-fold in each cell to amplify immune responses. While RC-Ads vaccines are more potent than RD-Ad, they risk causing adenovirus infections in humans. We engineered “single-cycle” Ad (SC-Ad) vectors that still amplify transgenes, but that do not produce infectious progeny. RD-, SC-, and RC-Ads were tested by single immunization against influenza, Ebola virus, HIV-1, Zika virus, HCV, CMV, Clostridium difficile, and cancers in mice, hamsters, and macaques and antibody and T cell responses were measured. SC-Ad expressing GFP, influenza HA, or HIV gp140 generated antigens with 33 to 100-times less virus than RD-Ad. SC-Ad generated vaginal antibodies that persisted beyond 6 months after single intranasal immunization. Intramuscular injection of 1010 vp of SC-Ad6 expressing HIV-1 gp160 generated env antibodies in sera within 4 weeks of a single immunization in rhesus macaques. SC-Ad expressing a C. difficile toxin A and B immunogen generated toxin-binding antibody reciprocal titers above 105 that persisted in excess of a half of a year after a single immunization. Single immunization with SC-Ad protected vaccinated animals against toxin and against challenge with C. difficile spores. When SC-Ads were used to amplify the expression of immune stimulatory proteins in the context of cancer, single intratumoral injection of SC-Ad expressing 4-1BBL or GMCSF mediated significant reductions in tumor growth and extension of survival without the need for exogenous cancer antigen expression. These data suggest that SC-Ads may have utility against infectious diseases and cancer.

Oncolytic viruses (OVs) can have utility for direct killing of cancer cells, but may also serve to activate the immune system against cancer cells. While viruses alone can serve as immune stimulators, there is great interest in arming OVs with genes encoding immune stimulatory proteins to amplify their effects. In this work, we have tested the efficacy of conditionally-replicating adenoviruses (CRAds) with and without selected immunostimulatory payloads, murine CD40L (mCD40L) or 4-1BBL (m4-1BBL), in an immune competent mouse model of melanoma. When CRAd657-m4-1BBL and CRAd657-mCD40L were injected into B16-hCAR murine melanoma tumors, both single vectors delayed tumor growth and prolong survival compared to empty CRAd657. However, combined injection of both CRAd-m4-1BBL and CRAd-mCD40L mediated significantly better control of tumor growth. All of the payloads increased immune cell infiltration into tumors and notably reduced expression of PD-1 exhaustion marker on T cells. Tumor volumes were negatively associated with total infiltrating T cell population. We found that the payloads increased immune cell infiltration into tumors with some specificities: recruitment of CD8 + T cells was higher with m4-1BBL expression, while mCD40L expression induced more CD4 + T cell infiltration. Importantly, the combination of CRAd657-m4-1BBL and CRAd657-mCD40L induced the highest immune cells/T cell infiltration and the highest anti-TRP-2 tumor-associated antigen T cell responses than empty or single gene vector. This combination also caused depigmentation in areas adjacent to the tumor sites in more animals. These data indicate that driving two axes of the immune system with combined immune stimulatory payloads can lead to improved anticancer immune responses and better tumor control in an immune competent model of cancer.