Nrf2 Regulates Inflammation by Modulating Dendritic Cell-T Cell Crosstalk during Viral-Bacterial Superinfection

Abstract

Every year millions of people are infected with influenza, which can be complicated by secondary bacterial pneumonia. One factor that may contribute to increased susceptibility to secondary bacterial infection is the modulation of inflammatory cytokines. NF erythroid 2-related factor 2 (Nrf2) has been shown to be a master regulator of the antioxidant response and various inflammatory cytokines. To test the role of Nrf2 during viral-bacterial superinfection, we used a mouse model of influenza-Staphylococcus aureus superinfection with wild-type (WT) or Nrf2-deficient (Nrf2-/-) mice. Loss of Nrf2 reduced influenza burden and increased S. aureus burden during superinfection. Additionally, Nrf2-/- mice had increased abundance of conventional type 1 dendritic cells (DCs). We then tested the interaction between DCs and T cells using an in vitro model of bone marrow-derived DCs with OVA and OT-II T cells. In this system, Nrf2-/- DCs promoted a Th2/regulatory T cell response as opposed to a Th1/Th17 response by WT DCs. This was recapitulated in vivo with superinfected Nrf2-/- mice having increased regulatory T cell populations. We also observed an increased median survival time of Nrf2-/- superinfected mice, due at least in part to increased IL-10 signaling, as anti-IL-10R Ab treatment reduced median survival time to levels seen in WT mice. Overall, these data suggest that loss of Nrf2 promotes differential T cell skewing mediated by DCs that promote a regulatory phenotype, increasing superinfection survival time, despite increased bacterial burden.

Figure 1:. Loss of Nrf2 alters immune responses to microbial infection.

(A) WT or Nrf2−/− mice were inoculated with PR8 influenza and monitored over a time course for weight loss. (B) Viral burden was assessed via qRT-PCR at 6dpi. (C) Formalin-fixed, paraffin-embedded lung tissue at 6dpi was processed and stained for influenza-HA with Fast Red in WT and Nrf2−/− mice, counterstained with hematoxylin. Images are representative, with inset scale bars measuring 200µM and arrows highlighting areas of HA-positive staining. (D) QuPath quantification of influenza-HA positive cells. (E) Airway immune cell recruitment quantified in the BAL of WT and Nrf2−/− mice. (F) Cytokine profile of WT and Nrf2−/− mice at 6dpi of influenza. (N=7). WT and Nrf2−/− mice were inoculated with USA300 MRSA and monitored at 24hpi for (G) weight loss, (H) lung bacterial burden, (I) immune cell recruitment in BAL and (J, K) changes in cytokine secretion. (N=4–13) *p≤0.05, **p≤0.01, ***p≤0.001

Figure 2:. Super-infected Nrf2−/− mice have higher bacterial burden, immune cell recruitment and inflammatory cytokines than WT mice.

(A) Schematic of super-infection for WT and Nrf2−/− mice. (B) Weight change over the course of infection for mock- and super-infected mice. (C) Lung bacterial burden for super-infected mice at 7dpi. (D) Total immune cell recruitment in BAL of super-infected mice at 7dpi. (E) Cytokine profile of mock- and super-infected mice WT and Nrf2−/− mice at 7dpi. (F) Chemokine profile of super-infected WT and Nrf2−/− mice at 7dpi. *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001 (N=8–16)

Figure 3:. Immune cell populations in the lung of WT and Nrf2−/− infected mice.

Lungs were dissociated and stained for cell markers as described in the methods. Cell populations were determined using (A) Cytobank and (B, C) traditional gating strategies. (D) Median fluorescence intensity (MFI) of indicated markers was calculated using FlowJo. (N=16–20) *p≤0.05, **p≤0.01

Figure 4:. Nrf2−/− phagocytic cells are less efficient at killing bacteria and more inflammatory.

Bone marrow derived (A) macrophages and (B) dendritic cells were inoculated with USA300 MRSA at an MOI of 10 for the indicated amount of time. After gentamycin treatment to kill extracellular bacteria, intracellular bacterial load was quantified and measured over time. (N=5–10) (C) Cytokine profile of BMDCs after inoculation with heat-killed USA300 MRSA for 24 hours. (N=10–12) *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001

Figure 5:. Nrf2−/− DCs alter T cell responses to infection.

(A) MHC-II expression in cDC1 and cDC2 cells in mock infected WT and Nrf2−/− mice. (N=16) (B) Cytokine profile of WT or Nrf2−/− BMDCs co-cultured with naïve OT-II T cells over 5 days in unstimulated, OVA stimulated, or OVA and heat-killed USA300 MRSA stimulation (N=10–12). Flow cytometry of WT or Nrf2−/− super-infected mice at 8dpi assessing regulatory T cell markers. (C) FoxP3, (D) IL-10 or (E) total numbers of regulatory T cells. (N=8) *p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001

Figure 6:. IL-10 signaling increases median survival time of super-infection in Nrf2−/− mice.

(A) Abundance of IL-10+ CD45+ cells at 8dpi of super-infection. (B) WT or (C) Nrf2−/− mice were super-infected as previously and i.p. injected with IgG control or αIL-10R antibody at 2- and 5dpi and monitored for survival. Dashed line represents 50% probability of survival and is graphed in (D) (Ns: WT IgG = 12, WT IL-10R = 15, Nrf2−/− IgG = 11, Nrf2−/− IL-10R = 15). (E) Percentage of CD4⁺ T cells that display a regulatory phenotype (FoxP3⁺IL-10⁺) against body weight at time of mouse euthanasia. (N=8). (F) Lung leak as assessed by total protein concentration in BAL. (N=3–5). *p≤0.05