ORCID
https://orcid.org/0000-0002-8081-7867
Date of Award
12-19-2024
Degree Name
Doctor of Philosophy (PhD)
Degree Type
Dissertation
Abstract
Ross River virus (RRV) is a mosquito transmitted virus endemic to Australia that infects humans and causes acute symptoms of rash, fever, myalgia, and fatigue. Although systemic viral infection subsides within a few weeks, joint pain lingers in many patients for months. Active viral replication is not readily detected in this chronic phase, although persistent viral RNA has been found in the synovial fluid of patients. This can be studied in mice, which can be productively infected with RRV when inoculated in the foot. Like in humans, in the chronic phase of disease, viral RNA is still present in the distal joints of mice even 4 or more months after initial infection. The immune response against RRV and other globally relevant related arthritogenic alphaviruses, such as Chikungunya virus (CHIKV), is of great interest. Although the immune system controls active viral infection within about a week, viral RNA is not cleared and persists. Initial replication is controlled by the type I interferon (IFN) response after innate sensing of viral patterns like double stranded RNA. Type I IFN stimulates cells to increase expression of antiviral genes that hinder replication and virus production. Mice lacking the type I IFN receptor succumb to alphavirus infection within just 4 days. This innate immune-mediated control allows B cells time respond to viral antigen and start making virus-specific antibodies that neutralize free virus and help prevent infection of new cells. Without B cells, mice are unable to control virus infection, and infectious virus can be detected in their blood over a year after initial infection. Although antibodies are effective at neutralizing free virus, they do not target infected cells whose antigen may only be inside the cell where antibodies are unable to access. Since virus is not actively replicating in the chronic phase of RRV infection, other immune cells are needed to remove cells harboring viral RNA, which still can act as a trigger for inflammation. Cells that express viral protein can be killed by cytotoxic CD8+ T cells, which survey for non-self peptides generated within cells and displayed on the MHC class I molecules expressed on all nucleated cells. Essential for control of other RNA viruses such as lymphocytic choriomeningitis virus (LCMV) and other mosquito-transmitted viruses including flaviviruses, CD8+ T cells are usually an important part of the adaptive immune response. However, during RRV and CHIKV infection, CD8+ T cells appear to have little function in reducing viral titers in infected foot or in lymphoid tissue. Mice lacking CD8+ T cells do not have higher titers in these tissues during either the acute or chronic phases. This unusual evasion of CD8+ T cells likely contributes to viral RNA persistence and chronic symptoms. The CD8+ T cell response to RRV has been previously investigated, and RRV-specific CD8+ T cells were found to expand in the spleen and draining lymph node (DLN). These cells expressed markers of antigen experience and produced cytokines including IFNγ and TNF. However, CD8+ T cell depletion only resulted in a maximum of 10-fold increase of RRV infection in quadriceps muscle and no difference in the foot or spleen. Adoptive transfer of both CD8+ cytotoxic T cells and CD4+ helper T cells can reduce titers in the feet of mice lacking adaptive immune systems, but CD8+ T cells alone had no effect. Given that CD8+ T cells were present, expressed functional cytokines, and had the ability to affect titers in limited tissues, we wondered if the numbers of T cells in RRV infection were similar to those in other virus infections. Perhaps the response was too small to affect titers in musculoskeletal tissues. To address this question, we compared the CD8+ T cell response to RRV to the response to the well-characterized model virus LCMV. We inoculated 3- to 4-week-old mice in the footpad with the same dose of either LCMV or an RRV strain that had been genetically modified to express the immunodominant CD8+ T cell epitope from LCMV. We found that LCMV stimulated the expansion of about 10-fold more virus-specific CD8+ T cells, which significantly reduced LCMV titers in the spleen between days 5 and 7 and in the foot between days 7 and 10 post infection. As expected, RRV-specific CD8+ T cells had no effect in these tissues. Single cell RNA sequencing analysis revealed that antigen-specific CD8+ T cells in RRV infection only achieved a less-proliferative intermediate activation state by day 5 post infection when compared with LCMV. Although both LCMV and RRV replicated to high titers in the foot, only LCMV efficiently infected lymphoid tissues. In an attempt to correct for a lack of RRV antigen in the DLN where CD8+ T cells become activated, we treated mice with local administration of anti-IFNAR1 blocking antibody to allow for greater infection in the foot and draining lymph node. This resulted in an RRV-specific CD8+ T cell response comparable to LCMV in both number and activation state. Priming of these CD8+ T cells was dependent on direction infection of dendritic cells (DCs), which had been inhibited by the type I IFN response. Despite the reliance on direct infection for maximal priming, we found that RRV also promoted DC presentation of exogenous antigen. Even with limited direct infection, splenic CD8+ T cells, while relatively low in number, were surprisingly efficient at killing cells pulsed with viral peptide. We also found that after adoptive transfer, RRV-specific CD8+ T cells trafficked successfully to sites of infection. However, while we were able to kill peptide pulsed cells ex vivo with CD8+ T cells from RRV-infected mice, we were unable to show killing of RRV-infected cells. Even the increased CD8+ T cell response after anti-IFNAR1 blockade failed to affect titers in the foot or spleen. This points to a defect both in the efficient generation of the response and in the interaction of the CD8+ T cells with bona fide infected cells. Overall, we find that the CD8+ T cell response to RRV is lower and delayed compared to another RNA virus that infects lymphoid cells. By enabling direction infection of dendritic cells, the response to RRV in dramatically increased. However, boosting the CD8+ T cell response did not directly lead to decreases in viral titers. There appears to be additional defects, as infected cells are not killed efficiently like peptide-pulsed cells. Although there are multiple challenges to overcome, this work highlights antigen availability and virus tropism as possible targets for immunomodulation against RRV and related arthritogenic alphaviruses, which evade CD8+ T cells to establish persistence.
Language
English (en)
Chair and Committee
Michael Diamond
Committee Members
Brian Edelson; Deborah Lenschow; Mark Miller; Robert Schreiber; Siyuan Ding
Recommended Citation
Bullock, Christopher, "Mechanism of Impaired CD8+ T Cell Activation During Ross River Virus Infection" (2024). Arts & Sciences Electronic Theses and Dissertations. 3349.
https://openscholarship.wustl.edu/art_sci_etds/3349