Science Spotlight

MAVS signaling effects on Tregs

From the Lund Laboratory, Vaccine and Infectious Disease Division

The immune system is a complex, highly regulated system that helps protect us from infection and autoimmunity. The fine balance between self-recognition and immunopathology versus foreign protein (antigen) identification is controlled through the immune response. One way this response is controlled is by identification of foreign antigens by recognition of pathogen-associated molecular patterns (PAMPs) through pattern recognition receptors (PRRs). An example of a PRR is RIG-I, which is a cytoplasmic RNA sensor that, once engaged, induces interferons and thus the innate immune response. RIG-I is mediated by its downstream adaptor mitochondrial antiviral-signaling protein or MAVS. Most identification of foreign antigens and downstream signaling after pathogen recognition is attributed to antigen presenting cells, however recent work has shown that T cells have the capacity to respond through innate immune mediators such as Toll-like receptors (TLRs) alone. Researchers at Fred Hutch in Jenifer Lund’s laboratory (Vaccine and Infectious Disease Division) recently published a paper in Science Reports looking at West Nile virus (WNV) and factors that mediate a protective immune response against it. In a previous study published by the lab, researchers found that mice lacking MAVS infected with WNV showed uncontrolled expansion of effector cells, suggesting a lack of regulatory T-cell function. That finding led to the current follow-up study performed by graduate student Andreia Costa. “Increasing evidence over the last decade has highlighted the important role that regulatory T cells (Tregs) play in the development of an effective immune response to infection. Tregs can be thought of as an immunological fulcrum, allowing for the expansion of conventional effector T cells but also serving to constrain potential collateral damage to host tissues. Our work extends the understanding of the utility of one mechanism of infection sensing by Tregs. Our current paper investigated the potential role of MAVS infection sensing by Tregs during WNV infection while retaining intact MAVS-function in the rest of the immune system.” said Costa. The group hypothesized that Tregs may directly detect infection using MAVS associated signaling pathways, altering the suppressive capacity of the Tregs and thus changing the anti-viral response.

To investigate the role of Tregs in a MAVS free system the group again used Mavs-/- (knockout) mice, infected them with WNV and compared their responses to those from WNV-infected wild-type (Mavs+/+) mice. They confirmed the phenotype of increased pathological signs in the Mavs-/- compared to wild type mice. Pathology included dramatic weight loss, lethargy, inflammation and swollen lymph nodes. This suggested that the Mavs-/- mice had a dysregulated inflammatory response. Looking closer at the Treg frequency, the knockout mice showed lower frequency of Tregs after infection compared to controls. The Tregs that were present in the knockout mice had reduced Foxp3 (a Treg marker) expression. Past research has shown that the amount of Foxp3 expressed is associated with suppressor function. Interestingly in this study the Mavs-/- Tregs had higher expression of suppressor markers (CD73, CTCA-4, and ICOS). Activation markers including CD44, Ki67 and CXCR3 were also elevated on knockout Tregs, similar to controls. One reason for decreased Foxp3 and reduced numbers with normal activation and activity could be that Tregs are converting to inflammatory Th17 cells. To support this idea, high levels of cytokines, which are needed for this conversion, were found in infected mice. This could suggest that MAVS proteins are needed to stabilize the Foxp3 Treg population.

Schematic of bone marrow chimera and of effects of WNV on weight loss and replication in mice.
A. Schematic of bone marrow chimera generation and WNV infection. B. Graphic of the three phases of WNV infection, this work focuses on the late visceral phase. C. Graphs of weight loss and viral load (D) in the wildtype and MAVS-/- Tregs animals. Image provided by DR. Costa

To further characterize WNV infection on Tregs the group looked at cell proliferation and suppressor effect in vitro. MAVS-deficient Tregs were stimulated with or without WNV, resulting in similar suppression and proliferation as control Tregs. This suggests the MAVS signaling is not required for TCR-stimulated proliferation or suppressive functions. Given the conflicting results between the decreased frequency in vivo and lack of effect in vitro, further testing was preformed. This time they focused on knocking out MAVS in Tregs alone, not the whole mouse, to study Treg-intrinsic MAVS effects. In order to accomplish this, an asymmetric mixed bone marrow chimera mouse model was used (see figure). In brief, mice with the diphtheria toxin receptor (DTR) inserted in the Foxp3 locus were irradiated and then bone marrow containing 90% Foxp3DTR/ 10% Mavs-/- (or 90% Foxp3DTR/ 10% WT) was transferred. After recovery, toxin was given to deplete Foxp3 and allow for expansion of either wildtype or Mavs-/- Tregs. Animals were then infected subcutaneously in the footpad with WNV and monitored for clinical signs of infection and viral load. Results were similar in both the wildtype and Mavs-/- Treg mice, supporting the idea that MAVS signaling in Tregs does not play a role in WNV disease outcome. After infection of the mice, the group found no difference by flow cytometry between the wildtype and Mavs-/- Tregs frequency, functional markers or in total T cell numbers. Together this data suggests that Treg-intrinsic MAVS signaling is not required for stable expression of Foxp3 or Treg function. In regards to the finding presented here Dr. Costa said, “We found that although MAVS deletion lead to a down-regulation of the Treg lineage-defining transcription factor, FoxP3, it did not contribute to the increased mortality seen in the full knock-out mouse model. Through our investigation of the role of viral sensing in Tregs via the RLR pathway we've furthered the understanding of Treg function in this system as well as addressed a long standing question in the field. It is often just as scientifically fruitful to rule out a cell type's involvement as to rule it in”.

Funding was provided by the National Institutes of Health and the Disease and Public Health Importance Training Grant.

Da Costa A,Garza E,Graham JB,Swarts JL,Soerens AG,Gale M,Lund JM. 2017. Extrinsic MAVS signaling is critical for Treg maintenance of Foxp3 expression following acute flavivirus infection. Sci Rep, 7, 40720.