Vimentin modulates regulatory T cell receptor-ligand interactions at distal pole complex, leading to dysregulated host response to viral pneumonia Authors: Ruihua Ma, Andrew D Prigge, Tatiana P Ortiz Serrano, Yuan Cheng, Jennifer M Davis, Karen F Lou, Walter A Wood, Hanh Chi Do, Ziyou Ren, McKenzie M Fulcer, Mary J Lotesto, Benjamin D Singer, Bria M Coates, Karen M Ridge Published: 2024 ABSTRACT SUMMARYForkhead box P3 (Foxp3)+ regulatory T cells (Tregs) resolve acute inflammation and repair the injured lung after viral pneumonia. Vimentin is a critical protein in the distal pole complex (DPC) of Tregs. This study reveals the inhibitory effect of vimentin on the suppressive and reparative capacity of Tregs. Treg-specific deletion of vimentin increases Helios+interleukin-18 receptor (IL-18R)+ Tregs, suppresses inflammatory immune cells, and enhances tissue repair, protecting Vimfl/flFoxp3YFP-cre mice from influenza-induced lung injury and mortality. Mechanistically, vimentin suppresses the induction of amphiregulin, an epidermal growth factor receptor (EGFR) ligand necessary for tissue repair, by sequestering IL-18R to the DPC and restricting receptor-ligand interactions. We propose that vimentin in the DPC of Tregs functions as a molecular switch, which could be targeted to regulate the immune response and enhance tissue repair in patients with severe viral pneumonia. INTRODUCTION Viral pneumonia damages the lower respiratory tract and causes acute respiratory distress syndrome (ARDS) due to persistent inflammation. Innate and adaptive immune responses are necessary for the successful clearance of pathogens, but an excessive immune response, which can persist for weeks or months after the initial infection, contributes to tissue damage and aberrant repair.1 Shortly after influenza A virus (IAV) infection, tissue-resident alveolar macrophages are depleted, and monocyte-derived alveolar macrophages (MoAMs) are recruited into the lung.2–4 Recruited MoAMs are associated with excessive or dysregulated cytokine and inflammatory cell responses in humans and mice,1,5,6 which limits lung tissue repair. During IAV infection, regulatory T cells (Tregs) infiltrate and proliferate in the lung tissue, where they play a critical role in inflammation resolution and tissue repair.7–9 Tregs, a distinct population of CD4+ lymphocytes expressing the lineage-specifying transcription factor forkhead box P3 (Foxp3), are essential for maintaining self-tolerance and immune homeostasis. Tregs suppress immune responses by regulating immune cell activation or eliminating pathologically activated immune cells, such as MoAMs.10,11 In the context of viral infection, Tregs maintain immune homeostasis by preventing exaggerated and detrimental inflammation, constraining tissue damage, and thereby indirectly promoting tissue repair.12,13 Recently, multiple studies have unveiled novel reparative functions of Tregs. Tregs directly influence nonimmune cells, including epithelial, endothelial, and mesenchymal cells, promoting the healing of injured tissue by producing various growth factors in a tissue-specific manner. In mice, Tregs promote lung tissue regeneration and repair by generating reparative mediators, including keratinocyte growth factor (KGF) and the epidermal growth factor receptor (EGFR) ligand amphiregulin (Areg).7,14,15 Treg production of Areg is interleukin (IL)-18 dependent but T cell receptor (TCR) independent. Notably, Tregs are a major early source of Areg during influenza infection and are required to restore normal lung function after viral infection.7,16 The interaction between Tregs and antigen-presenting cells initiates the recruitment of proteins to the immunological synapse, leading to the formation of the distal pole complex (DPC) on the opposite side of the immunological synapse. Key components of the DPC include CD43, a cell-surface mucin that has been proposed to function as a negative regulator of T cell signaling, and the ERM proteins (ezrin, radixin, and moesin).17 Recent studies showed that protein kinase C θ (PKC-θ) is also uniquely localized to the DPC in Tregs.18,19 The DPC is important for T cell activation because it functions as an independent signaling complex.17 Nonetheless, the exact role of the DPC in Tregs has remained largely unknown. In a graft-versus-host disease model, Tregs transfected with a vimentin small interfering RNA (siRNA) disrupted the DPC and displayed enhanced suppressive activity.20 However, the role of vimentin in the DPC of Tregs and its potential role in regulating Treg function during viral pneumonia remain unknown. In this study, we generated Treg-specific vimentin knockout (KO) mice, and using a murine model of influenza A pneumonia, our findings indicate that vimentin constrains Treg function. Treg-specific deletion of vimentin enhances the expression of suppressive molecules and augments the suppressive function of Tregs. Moreover, in vimentin conditional KO (cKO) mice, vimentin deficiency in Tregs induces the proliferation of Helios+IL-18 receptor (IL-18R)+ Tregs in the lung and augments IL-18/IL-18R/Areg signaling, contributing to increased survival and lung repair after influenza infection. Mechanistically, vimentin sequesters IL-18R at the DPC of Tregs. Disruption of the DPC leads to the rearrangement of IL-18R, rendering Tregs more susceptible to binding their ligands, ultimately promoting IL-18/IL-18R/Areg-mediated lung repair. RESULTS Vimentin limits Treg suppressive function in vivo and in vitro To determine whether vimentin modulates Treg-mediated suppressive function in vivo, we used the Rag1−/− mouse model of T cell-induced colitis.21–23 The efficacy of wild-type (WT) and Vim−/− Tregs to suppress conventional T (Tcon) cell-induced colitis was assessed by adoptively transferring WT or Vim−/− Tregs together with Tcon cells into Rag1−/− mice. We used high (1:4) and low (1:8) titers of Treg:Tcon cells. Rag1−/− mice receiving only Tcon cells exhibited significant weight loss and severe disease characterized by epithelial damage, focal crypt injury, goblet cell depletion, and inflammatory cell infiltration in the colon (Figures 1A–1C). Rag1−/− mice receiving a high titer (1:4) of either WT or Vim−/− Tregs and Tcon cells displayed reduced crypt damage and inflammation (Figures S1A–S1C). Vimentin limits Tregs’ suppressive function in vivo and in vitro (A) Weight change of Rag1−/− mice following adoptive transfer of 4 × 105 Tcon cells (n = 5) alone or with 0.5 × 105 WT or Vim−/− Tregs. Data are mean ± SEM. (B) H&E staining of distal colon from Rag1−/− mice 45 days post-transfer. Scale bar, 100 μm. (C) Histological scores of colon tissue from (B). (D) Vimentin expression in WT and Vim−/− nTregs. Scale bar, 5 μm. (E) In vitro suppression assay by co-culturing Tcon cells with WT or Vim−/− nTregs, showing proliferating cell percentage (left) and suppression percentage (right) (n = 3). Data are mean ± SD, n = 3. (F) Expression of Foxp3, CD25, Helios, and CD73 on CD4+Foxp3+ cells from WT and Vim−/− spleens. Data are mean ± SD, n = 3–6 mice per group. p values were calculated via one-way ANOVA (C), two-way ANOVA (E), and two-tailed unpaired Student’s t test (F). NS, not significant (p > 0.05), *p < 0.05, **p < 0.005, and ***p < 0.0005. In contrast, Rag1−/− mice receiving a low titer (1:8) of WT Treg:Tcon cells exhibited significant weight loss and severe disease, similar to Rag1−/− mice receiving only Tcon cells (Figures 1A–1C). Rag1−/− mice receiving Vim−/− Tregs at the same low titer (1:8) maintained their body weight, displayed normal tissue histology, and had significantly reduced disease severity scores (Figures 1A–1C). We found that as few as 0.5 3 105 (1:8) Vim−/− Tregs were sufficient to attenuate Tcon cell-induced colitis in Rag1−/− mice, while 2-fold (1:4) more WT Tregs were required to achieve similar suppression (Figures 1A–1C and S1A–S1C). These data suggest that vimentin limits Treg-mediated suppression in vivo. To explore vimentin’s role in Treg function, we assessed the consequences of disrupting vimentin expression in Tregs. Splenic natural Tregs (nTregs) were isolated from WT and Vim−/− mice. Vimentin formed a cap-like structure in WT nTregs, which was absent in Vim−/− nTregs (Figure 1D). Next, we performed a standard contact-dependent, in vitro suppression assay, which showed that Vim-deficient Tregs suppress Tcon cell proliferation more effectively than WT Tregs (Figure 1E). To determine whether vimentin regulates in vitro induced Treg (iTreg) generation and function, we cultured naive CD4+ T cells from WT and Vim−/− mice under Treg-skewing conditions to produce WT and Vim−/− iTregs. We observed that vimentin also formed a cap-like structure in WT iTregs, which was absent in Vim−/− iTregs (Figure S1D). Notably, the proportion of CD4+CD25+Foxp3+ iTregs was comparable in WT and Vim−/− groups (Figure S1E), suggesting that vimentin is dispensable for iTreg generation in vitro. Similarly, Vim−/− iTregs exhibited enhanced suppressive activity, effectively inhibiting Tcon cell proliferation compared with WT iTregs (Figure S1F). Flow cytometric analysis of splenic nTregs and iTregs showed that WT and Vim−/− Tregs exhibited similar expression levels of Foxp3 and CD25 (Figures 1F and S1G). The upregulated expression of suppressive markers in Vim−/− Tregs contributes, at least in part, to the augmented suppressive function in Vim−/− Tregs both in vivo and in vitro.24,25 Indeed, we found that Vim−/− Tregs exhibited significantly higher expression of Helios, CD73, neuropilin 1 (NRP1), and lymphocyte-activation gene (LAG3) compared to WT Tregs (Figures 1F and S1G). Treg-specific deletion of vimentin protects the host from IAV-induced lung injury and mortality To unequivocally define the role of vimentin in Tregs, we generated a Treg-specific vimentin-deficient cKO mouse (Vimfl/flFoxp3YFP-Cre) by crossing a LoxP-flanked vimentin mouse with a Foxp3YFP-Cre mouse (Figure S2A). The WT control mice harbored Foxp3YFP-Cre containing a WT vimentin gene. qPCR, western blot, and immunostaining analysis confirmed the complete loss of vimentin in splenic nTregs from cKO mice (Figures S2B–S2D). Confocal imaging confirmed the formation of a cap-like structure by vimentin in WT nTregs, which was absent in vimentin cKO nTregs (Figure S2D). Similarly, splenic nTregs from vimentin cKO mice had increased Helios and CD73 expression, while no differences were observed in Foxp3 and CD25 expression compared to splenic nTregs from WT mice (Figure S2E). To evaluate the host response to viral pneumonia, WT and vimentin cKO mice were infected with a lethal dose of IAV. At 10 days post-infection (dpi), 80% of WT mice had died, while ~90% of vimentin cKO mice remained alive (Figure 2A). Vimentin cKO mice exhibited >50% survival, and surviving cKO mice exhibited mild lethargy, coat ruffling, febrile shaking, and eye watering under lethal dose infection. Treg-specific deletion of vimentin protects the host from IAV-induced lung injury and mortality WT (Vim+/+Foxp3YFP-Cre) and vimentin cKO (Vimfl/flFoxp3YFP-Cre) mice were intratracheally infected with lethal (60 plaque-forming units [PFUs]) (A) or a sublethal (30 PFUs) (A–K) dose of IAV. (A) Survival curve for WT (n = 15, lethal; n = 9, sublethal) and vimentin cKO (n = 9, lethal; n = 8, sublethal) mice using log rank (Mantel-Cox) test. (B) Percentage of weight loss from baseline in WT (n = 8) and vimentin cKO (n = 10) mice, analyzed by mixed-effects model with Sidak’s multiple comparisons test. Data are mean ± SEM. (C) Representative H&E lung staining at 21 dpi. Scale bar, 100 μm. (D) Lung injury score from H&E sections. (E) CD45 staining of lungs at 21 dpi. Scale bar, 100 μm. (F) Flow cytometry quantification of CD45+ cells from whole lung at 21 dpi. (G) Krt5 and Foxp3 staining of lungs at 21 dpi. Scale bars, 100 μm (bright field) and 50 μm (fluorescence). (H and I) Quantification of Krt5 expression (H) and frequency (I) using TissueGnostics HistoQuest. (J) IFN-β levels in BALF by multiplex immunoassay (n = 3–8 mice per group). (K) Viral titer of whole-lung homogenates at 5 dpi. Data are mean ± SD, n = 4–6 mice per group. p values were calculated by two-way ANOVA (B and J) and two-tailed unpaired Student’s t test (D, F, H, I, and K). NS, not significant (p > 0.05), *p < 0.05, **p < 0.005, and ****p < 0.0001. Next, we wanted to determine whether WT mice with vimentin-sufficient Tregs were more susceptible to influenza-induced lung injury due to impaired recovery. We infected WT and vimentin cKO mice with a sublethal dose of IAV. At 21 dpi, only 55% of WT mice survived, whereas all vimentin cKO mice were alive (Figure 2A). WT mice lost 30% of their body weight and failed to recover their total body weight compared with vimentin cKO mice following a similar nadir (Figure 2B). At 21 dpi, WT mice displayed more severe lung injury (Figures 2C, 2D, and S2F) and increased lung infiltrates, which were mainly composed of immune cells identified by the pan-hematopoietic marker CD45 by immunostaining (Figures 2E and S2F) and flow cytometry (Figure 2F). In contrast, vimentin cKO mice had fewer immune cell infiltrates (Figures 2E, 2F, and S2F). These results suggest that vimentin cKO mice resolve tissue inflammation during recovery from severe influenza-induced injury, but WT mice do not. In previous studies, investigators demonstrated that following influenza-induced lung injury, a population of cytokeratin 5+ (Krt5+) basal-like cells expand and migrate to the distal airspaces to form a provisional barrier.26 These cells cannot differentiate into functional type 2 alveolar epithelial cells, resulting in a dysplastic response that contributes to a dysregulated and incomplete repair phenotype after injury.8,27,28 At 21 dpi, vimentin cKO mice exhibited a significant reduction in Krt5+ cells compared with WT mice (Figures 2G–2I and S2F). Enhanced survival and recovery from influenza-induced lung injury in vimentin cKO mice was observed despite having similar increases in interferon (IFN)-β levels in the bronchoalveolar lavage fluid (BALF) and no difference in viral titer in the lung homogenates compared with WT mice (Figures 2J and 2K). In summary, vimentin cKO mice displayed an enhanced capacity to restore lung tissue architecture and function during the recovery phase of influenza-induced lung injury. Treg-specific deletion of vimentin promotes the resolution of inflammation after influenza infection Next, we tested the hypothesis that an augmented Treg suppressive function in vimentin cKO mice conferred protection from viral pneumonia by regulating the influx of immune cells and accelerating the resolution of inflammation. We assessed the recruitment of immune cells into the influenza-infected lung by flow cytometry29 (Figure S3A). We observed ~30% and 110% increases in Treg numbers in the lungs of vimentin cKO mice compared to WT mice on 5 and 10 dpi, respectively (Figures 3A and 3B). There were no discernible differences in the overall T cell population, including CD4+ and CD8+ T cells, under steady-state and influenza-infected conditions in either genotype (Figures S3B–S3D). Treg-specific deletion of vimentin promotes the resolution of inflammation after influenza infection (A and B) Treg frequency and number in WT and vimentin cKO lungs following influenza infection. (C and D) MoAM frequency within the CD45+ population and MoAM absolute number in WT and vimentin cKO lungs following influenza infection. (E–I) IL-18, TNF-α, IL-6, IFN-γ, and IL-33 levels in BALF measured by multiplex assay. Data are mean ± SD, n = 3–8 mice per group. p values were calculated by two-way ANOVA. NS, not significant (p > 0.05), *p < 0.05, **p < 0.005, ***p < 0.0005, and ****p < 0.0001. Tissue-resident alveolar macrophages rapidly decreased after influenza infection in both genotypes (Figures S3E and S3F). Neutrophil recruitment to the influenza-infected lung tissue was similar in WT and vimentin cKO mice (Figure S3G and S3H). MoAMs are typically not present in the alveolar compartment in normal, healthy lung tissue at steady state. During influenza infection, monocytes are recruited and rapidly differentiate into MoAMs in the lung parenchyma, which is associated with excessive or dysregulated cytokine and inflammatory cell responses in humans and mice.1,5,6 WT and vimentin cKO mice displayed increased MoAMs in the infected lung, but vimentin cKO mice had 40% fewer MoAMs than WT mice 5 dpi (Figures 3C and 3D). Despite the reduced number of MoAMs recruited to vimentin cKO lungs, the level of inflammatory cytokines in the BALF of vimentin cKO mice was similar to that observed in WT mice at 5 dpi. Nevertheless, at 10 dpi, vimentin cKO mice exhibited significantly lower IL-18, tumor necrosis factor alpha (TNF-α), IL-6, and IFN-γ levels than their WT counterparts (Figures 3E–3I). These results suggest a delay in the resolution of inflammation and a decrease in the number of proinflammatory MoAMs and that vimentin deficiency in Tregs promotes the resolution of inflammation after influenza infection. Vimentin deficiency induces enhanced CD73-adenosine signaling in Tregs Influenza infection results in high levels of ATP in BALF,30 which serves as a danger-associated molecular pattern to induce innate and adaptive immune responses. The ectonucleotidases CD39 and CD73 convert extracellular ATP into ADP and AMP metabolites, which are then metabolized to adenosine. The mRNA levels of Nt5e and Entpd1, the genes encoding CD73 and CD39, respectively, were significantly upregulated in vimentin-deficient splenic nTregs relative to WT Tregs (Figure 4A). Vimentin cKO mice had an ~2-fold greater number of CD73+ Tregs and elevated CD73 expression in lung Tregs compared to WT mice (Figure 4B and 4C) at steady state. Vimentin-deficient iTregs and Tregs from the spleen or lymph nodes of vimentin cKO mice also expressed higher levels of CD73 (Figures 4D and S4A–S4D). The CD39 surface suppressive marker was upregulated in Vim−/− iTregs compared to WT iTregs (Figure S4E) but was similar in lung Tregs from WT and vimentin cKO mice (Figures S4F and 4G). We observed a significant increase in adenosine levels in Vim−/− iTregs treated with ATP compared to WT iTregs (Figure 4E). Vimentin deficiency induces enhanced CD73-adenosine signaling in Tregs (A) Nt5e, Entpd1, and Vim expression in WT and Vim-deficient splenic nTregs (n = 3). (B) CD73+ Treg frequency among lung Tregs after influenza infection. (C) CD73 expression on lung Tregs after influenza infection. (D) CD39+CD73+ Treg frequency in WT or Vim−/− iTregs (n = 6). (E) Adenosine levels in supernatants from WT and Vim−/− iTregs treated with 5 mM ATP for 1 h (n = 3). (F and G) WT and vimentin cKO mice were intratracheally infected with IAV and vimentin cKO mice received daily anti-CD73 (100 μg/mouse, intraperitoneal) or PBS from 1 dpi for 1 week. (F) Weight loss. (G) Representative H&E lung staining at 21 dpi. Scale bar, 500 μm. Data are mean ± SD, n = 3–5 mice per group. p values were calculated by two-way ANOVA (A–C) and two-tailed unpaired Student’s t test (D and E). *p < 0.05, **p < 0.005, ***p < 0.0005, and ****p < 0.0001. Surprisingly, the elevated CD73 expression observed in vimentin cKO lung Tregs was not maintained following influenza infection (Figures 4B and 4C). To determine whether this nucleotide signaling pathway contributes to the protection from influenza-induced lung injury observed in vimentin cKO mice (Figure 2), mice infected with a sublethal dose of IAV were treated with either anti-CD73 or vehicle control. Influenza-infected vimentin cKO mice treated with anti-CD73 and WT mice lost 30% of their body weight and showed delayed recovery compared with vehicle-treated vimentin cKO mice (Figure 4F). At 21 dpi, vimentin cKO mice treated with anti-CD73 and WT mice had severe lung injury compared to vehicle-treated vimentin cKO mice, as evaluated by histological analysis of H&E-stained lung tissue (Figure 4G). These results suggest that the activated CD73 signaling in vimentin cKO Tregs may be sufficient to suppress the activation and function of effector immune cells31,32 and promote the activity of the Tregs early in the host response to viral infection. Vimentin deficiency augments the IL-18/IL-18R/Areg signaling axis in Tregs Several studies have shown that Tregs are required for lung tissue regeneration and repair by directly affecting nonimmune cells.7,14,15,33–35 Arpaia et al. found that IL-18 and IL-33 signaling drives Treg production of Areg, an EGFR ligand,7 which limited the severity of lung damage and exerted a tissue-protective function.7 Influenza-infected vimentin cKO mice displayed significantly enhanced recovery at 10 dpi compared to the WT mice (Figure 2B), leading us to investigate whether vimentin deletion in Tregs also influences their reparative function. We observed a significant increase in the frequency of IL-18R+ Tregs in the lungs of vimentin cKO mice, at steady state and 5 and 10 dpi, compared to WT mice (Figure 5A). Furthermore, we observed elevated IL-18R expression levels in lung Tregs of vimentin cKO mice compared to WT mice across all the aforementioned conditions (Figure 5B). The absolute number of IL-18R+ Tregs in the lungs was higher in the vimentin cKO mice compared to the WT mice, both at steady state and after influenza infection (Figure 5C); vimentin cKO mice had an ~3-fold greater number of IL-18R+ Tregs compared with WT mice at 10 dpi. At steady state and after influenza infection, IL-18R+ Treg frequency and IL-18R expression in Tregs of vimentin cKO mice were upregulated in the spleen and lymph nodes compared to WT mice (Figures S5A and S5B). Vimentin deficiency augments IL-18/IL-18R/Areg signaling in Tregs (A) Flow cytometry contour plots and quantification of IL-18R+ Treg frequency among lung Tregs of WT and vimentin cKO mice after influenza infection. (B) IL-18R expression on lung Tregs after influenza infection (n = 4–5 mice per group). (C) IL-18R+ Treg number in whole lung after influenza infection. (D) BALF Areg levels measured by ELISA (n = 3–8 mice per group). (E) Areg expression in lung Treg by qPCR at 10 dpi. (F) CD4+CD25+ lung Tregs cultured with IL-2, CD3/CD28 beads, and 100 ng/mL IL-18 for 4 days, with Areg levels assessed in supernatant by ELISA. (G–I) WT and vimentin cKO mice infected with IAV; vimentin cKO mice received anti-Areg (50 μg/mouse, intraperitoneal) at 5 dpi, 4 times in 3 day intervals, or vehicle control. Lung histology at 21 dpi. Scale bar, 500 μm. (J and K) Representative contour plots (J) and Ki67+IL-18R+ Treg frequency (K) among lung Tregs after influenza infection. Data are mean ± SD, n = 3–5 mice per group. p values were calculated by two-way ANOVA (A–F and K) and two-way ANOVA (I). *p < 0.05, **p < 0.005, ***p < 0.0005, and ****p < 0.0001. IL-18 drives Tregs to produce Areg after influenza infection.7 A relatively high level of IL-18 was detected in BALF samples from both WT and vimentin cKO mice following influenza infection (Figure 3E). We observed higher levels of Areg in BALF samples from vimentin cKO mice on 10 and 15 dpi compared to WT mice (Figure 5D). Additionally, 10 dpi lung Tregs from cKO mice showed increased Areg expression compared to WT lung Tregs (Figure 5E), implying a potentially enhanced reparative environment in vimentin cKO mice. Neither WT nor vimentin cKO mice had increased KGF levels following influenza infection (Figure S5C), suggesting that Areg, not KGF, drives the reparative capacity. To determine whether the increased IL-18R+ Tregs in vimentin cKO mice induced more Areg production in response to IL-18 stimulation, WT and vimentin lung Tregs were cultured in vitro under a T cell medium with or without IL-18. IL-18 stimulation resulted in notably higher Areg levels in vimentin-deficient lung Tregs than in WT cells (Figure 5F). These data directly support the idea that vimentin cKO mice generate more Areg in response to influenza infection due to the elevated IL-18R levels in lung Tregs compared to WT animals. Additionally, we examined the effects of Areg inhibition using an anti-Areg antibody36,37 on the protective effects of Treg-specific vimentin deletion during influenza-induced lung injury. Vimentin cKO mice treated with anti-Areg exhibited more severe lung injury than vehicle-treated vimentin cKO mice, as assessed by histological analysis of H&E-stained sections on 21 dpi (Figures 5G–5I). Furthermore, IL-18R+ lung Tregs exhibited striking proliferative capacity compared to their IL-18R− counterparts (Figures 5J, 5K, and S5D–S5F), consistent with a prior study.7 The frequency and absolute number of IL-18R+Ki67+ Tregs in vimentin cKO mice were significantly higher than WT mice at baseline and 5 and 10 dpi (Figures 5J, 5K,and S5F). Moreover, vimentin cKO mice demonstrated an increase in the non-proliferating IL-18R+Ki67− Tregs in the lung compared to WT mice (Figures 5J and S5E). In summary, these results demonstrate that vimentin-deficient Tregs increase IL-18R expression and produce more Areg when exposed to IL-18. In contrast, vimentin-sufficient Tregs fail to upregulate the expression of IL-18R following influenza infection, resulting in lower production of Areg and delayed recovery from viral pneumonia. Vimentin deficiency in Tregs leads to the accumulation and persistence of Helios+IL-18R+ Tregs in the lung after influenza infection Helios, a member of the Ikaros transcription factor family, is essential for the suppressive function of Tregs. Tregs lose their suppressive function following the genetic deletion or pharmacological degradation of Helios.25,38–41 We tested the hypothesis that the increased vimentin cKO Tregs observed after influenza infection (Figures 3A and 3B) and enhanced suppressive activity (Figures 1 and 2) were associated with increased Helios expression. We assessed the frequency of Helios+ lung Tregs in influenza-infected WT and vimentin cKO mice using flow cytometric analysis of single-cell suspensions from whole lung tissue. We observed an increased frequency and absolute number of Helios+ lung Tregs and increased Helios expression in vimentin cKO mice at steady state and on 5 and 10 dpi but not in WT mice (Figures 6A–6C). These findings indicate that influenza infection selectively induces the accumulation of Helios+ Tregs in the lungs of vimentin cKO but not WT mice. Vimentin deficiency in Tregs leads to accumulation and persistence of Helios+IL-18R+ Tregs in the lung after influenza infection (A) Flow cytometry contour plots and quantification of Helios+ Treg frequency among lung Tregs in WT and vimentin cKO mice after influenza infection. (B) Helios expression on lung Tregs of WT and vimentin cKO after influenza infection. (C) Helios+ Treg number in whole lung after influenza infection. (D and G) Heatmap of differentially expressed genes (p < 0.05). (E, F, H, I, and N) Normalized expression of Nt5e, Entpd1, Il18r1, Areg, and Prkcq in Helios+ and Helios− Tregs. (J) Frequency of Helios+IL-18R+, Helios+IL-18R−, and Helios−IL-18R+ cells among lung Tregs after influenza infection. (K and L) Frequency and absolute number of Helios+IL-18R+ Tregs in the whole lung tissue after influenza infection. (M) Relative mRNA expression of Il18r1, Areg, and Vim in WT and Vim-deficient splenic nTregs. Data are mean ± SD, n = 4–5 mice/group. p values were calculated by two-way ANOVA (A–C and K–M) and two-tailed unpaired Student’s t test (E, F, H, I, and N). *p < 0.05, ***p < 0.0005, and ****p < 0.0001. Thornton and colleagues generated HeliosGFPFoxp3RFP double reporter mice, sorted Helios+ and Helios− Tregs, and processed these cells for bulk RNA sequencing (RNA-seq).25 We reanalyzed their RNA-seq data for Helios+ and Helios− Tregs and found that Helios+ Tregs displayed an increase in the expression of genes related to Treg activation and suppression, including Nt5e and Entpd1 (Figures 6D–6F and S6A–S6C). These Helios+ Tregs were also marked by increased expression of genes related to extracellular matrix remodeling and tissue repair (Figure 6G). Notably, Helios+ Tregs exhibited increased expression of Il18r1 and Areg compared to Helios− Tregs (Figures 6H and 6I). Transcriptomic analysis of IL-18R+ Tregs demonstrated an enhanced gene expression profile associated with extracellular matrix remodeling and tissue repair.7 Therefore, we performed additional flow cytometric analyses of Tregs based on Helios and IL-18R expression and identified four distinct lung Treg populations in WT and vimentin cKO mice (Figure 6J). The frequency and absolute number of Helios+IL-18R+ Tregs were increased in vimentin cKO mice compared to WT mice under steady-state conditions and following influenza infection (Figures 6K and 6L). Helios+IL-18R− and Helios−IL-18R+ Tregs exhibit consistent levels following influenza infection in both genotypes (Figures S6D and S6E). Intriguingly, the transcriptional levels of Il18r1 and Areg were upregulated in the vimentin-deficient nTregs relative to their WT counterparts (Figure 6M), endowing vimentin-deficient Tregs with an increased capacity for tissue recovery and repair. Through the retrieval and analysis of the published single-cell assay for transposase-accessible chromatin with sequencing (scATAC-seq) dataset for colon Tregs,42 we identified reduced chromatin accessibility at the Il18r1 and Areg locus in Ikzf2-deficient Tregs compared to WT Tregs (Figures S6F and S6G). Chromatin accessibility at the vimentin locus was comparable between Ikzf2-deficient Tregs and WT Tregs (Figure S6H). Similarly, vimentin expression showed no significant difference between Helios+ Tregs and Helios− Tregs (Figure S6I). Interestingly, the transcript of Prkcq, encoding PKC-θ, recognized for inhibiting Tregs’ suppressive function, displayed a notable downregulation in Helios− Tregs compared to Helios+ Tregs (Figure 6N). In summary, Helios+ Tregs, characterized by elevated gene signatures associated with suppressive and reparative functions, accumulate and persist in the lungs of vimentin cKO mice after influenza infection, potentially contributing to an expedited recovery from viral pneumonia. Vimentin spatially sequesters IL-18R at the DPC of Tregs To investigate the molecular mechanism by which vimentin regulates the function of Tregs, we performed immunofluorescent confocal microscopy on iTregs stained with vimentin and DPC markers CD43 and ezrin. Vimentin was localized to a cap-like structure beneath the CD43 surface marker and colocalized with ezrin (Figures 7A and 7B, top), confirming that vimentin is part of the DPC in Tregs.20 Withaferin A (WFA) is a pharmacological agent known to bind to and disassemble vimentin intermediate filaments.43–45 Treatment of WT iTregs with WFA resulted in the collapse of the cap-like structure formed by vimentin (Figures 7A and 7B, middle, and 7E). Strikingly, Vim-deficient iTregs displayed a complete absence of the DPC structure (Figures 7A and 7B, bottom). These results suggest that vimentin intermediate filaments are essential for the DPC structure in Tregs. Vimentin spatially sequesters IL-18R at the DPC of Tregs (A and B) Confocal images of vimentin and CD43 or ezrin in WT iTregs treated with DMSO (vehicle control) or WFA and Vim−/− iTregs. Scale bar, 5 μm. (C and D) Confocal images showing uniform IL-18R and CD73 distribution on vimentin-deficient nTregs. Scale bar, 5 μm. (E) Confocal images of vimentin and IL-18R in WT nTregs treated with vehicle control or 1 μM WFA for 24 h. Scale bar, 5 μm. (F) Vimentin expression in WT iTregs treated with 1 μM WFA or vehicle control for 24 h. Scale bar, 5 μm. (G) Carboxyfluorescein succinimidyl ester (CFSE) dilution in CD4+CD25− Tcon cells co-cultured with WT iTregs treated with WFA or vehicle control (n = 3). (H) CD4+CD25+ lung Tregs from WT mice were cultured with IL-2, CD3/CD28 beads, and 100 ng/mL IL-18 for 3 days, followed by 1 μM WFA treatment for 24 h. Supernatants were collected for Areg measurement by ELISA (n = 6–7 per group). Data are presented as mean ± SD. p values were calculated by two-way ANOVA (G) and two-tailed unpaired Student’s t test (H). **p < 0.005, ***p < 0.0005, and ****p < 0.0001. Intriguingly, confocal imaging revealed a distinct pattern of IL-18R and CD73 expression. These proteins were exclusively expressed within the vimentin DPC in WT nTregs (Figures 7C and 7D, top) and iTregs (Figure S7A). In contrast, vimentin deficiency prompted a uniform redistribution of the Treg surface proteins IL-18R and CD73 across the cell membrane in vimentin-deficient nTregs (Figures 7C and 7D, bottom, and S7B). This implies that vimentin intermediate filaments play a crucial role in spatially sequestering signaling molecules, such as IL-18R and CD73, within the DPC area of Tregs. Treatment of WT iTregs with WFA also led to the redistribution of IL-18R on the Treg surface (Figure 7E), with an increased total expression level (Figure S7C). Functionally, WFA treatment enhanced the suppressive capacity of WT iTregs compared to untreated controls (Figures 7F and 7G). The WFA-mediated redistribution of surface receptor IL-18R was associated with the increased production of Areg in IL-18-treated WT lung Tregs (Figure 7H). In parallel to the complete loss of the DPC structure and redistribution of IL-18R and CD73 on the cell membrane, vimentin-deficient iTregs demonstrated enhanced suppressive functionality in vitro compared to vimentin-sufficient WT iTregs (Figures S1D–S1F). These results suggest that the redistribution of receptors on the cell membrane of vimentin-deficient Tregs facilitates enhanced receptor-ligand interactions, such as IL-18R/IL-18, thereby contributing to Areg production, compared to vimentin-sufficient WT Tregs. FOXO3a is a transcription factor that plays a crucial role in promoting Treg function. The mammalian target of rapamycin complex 2 (mTORC2) reduces FOXO3a activity by promoting phosphorylation-dependent nuclear exclusion, and the PKC-θ complex associated with vimentin may play a role in facilitating this process via IkB kinase (IKK).20 We observed that PKC-θ was localized within the DPC along with vimentin in WT iTregs (Figure S7D). Remarkably, vimentin deficiency decreased PKC-θ-mTORC2-FOXO3a signaling, which was demonstrated by a reduction in phosphorylated (p)-FOXO3a/FOXO3a levels (Figure S7E). The expression of Nrp1 and Lag3, driven by Foxo3a signaling, was upregulated in vimentin-deficient Tregs compared to WT Tregs (Figure S7F), consistent with a previous report.20 Interestingly, the expression of Nt5e (Figure 4A), Il18r1, and Areg (Figure 6M) was upregulated in vimentin-deficient Tregs compared to WT Tregs. This may be attributed, at least in part, to the decreased PKC-θ-mTORC2-FOXO3a signaling in vimentin-deficient Tregs. DISCUSSION Tregs are primarily recognized for their ability to dampen the response of effector immune cells, leading to the resolution of inflammation.46 In addition, Tregs can directly regulate tissue repair by generating pro-repair mediators, such as Areg and KGF.7,15,33,34 In the context of severe viral pneumonia, including coronavirus disease 2019 (COVID-19), characterized by lung inflammation and respiratory failure due to an overly aggressive immune response, Tregs play a pivotal role as key orchestrators in both the inflammatory response and the repair processes in the lungs during viral infections.10,47 However, the specific cytoskeleton regulator governing Treg suppressive and reparative functions during viral pneumonia has remained unclear. In this study, using murine IAV infection as a model for virus-induced immunity, inflammation, and tissue damage, we demonstrate that vimentin constrains Tregs’ suppressive and reparative capabilities. Treg-specific vimentin-deficient mice exhibit an increase in Helios+IL-18R+ Tregs in the lungs, a reduction in proinflammatory MoAM populations in the lungs, a less inflammatory cytokine profile, and an increase in the pro-repair factor Areg in the BALF. As a result, Tregs deficient in vimentin display a more prominent suppressive and reparative phenotype, protecting the mice from influenza-induced lung injury and mortality. McDonald-Hyman and colleagues used siRNA to knock down vimentin in Tregs, creating a sparse filament network, and found that vimentin knockdown enhanced Treg suppression in vitro. Adoptive transfer of vimentin-knockdown Tregs significantly ameliorated disease severity in a graft-versus-host disease model.20 Our study improves the rigor of this prior research by utilizing a genetically engineered mouse model to knock out vimentin, specifically in Tregs. We found that vimentin deficiency and pharmacological inhibition enhance Treg suppressive function in vitro, with vimentin-deficient Tregs also showing increased suppression in vivo. Notably, Treg-specific vimentin deficiency protected mice from influenza-induced mortality. Although WT and vimentin cKO mice experienced maximum weight loss at 10 dpi, vimentin cKO mice lost less than 20% body weight, while WT mice lost over 30%. This was attributed to the enhanced suppressive function in vimentin-deficient Tregs, characterized by increased expression of suppressive molecules such as Helios, and enhanced inflammation resolution, as demonstrated by significantly lower inflammatory cytokines in the BALF of vimentin cKO mice (Figures 3E–3I). In conclusion, these findings underscore vimentin’s crucial role as an inhibitory signaling molecule in Treg function. Helios was initially identified as a marker for distinguishing naturally occurring thymic-derived Tregs from peripherally induced Tregs.48 However, subsequent studies raised concerns about Helios being a definitive marker for nTregs, given its expression in conventional human T cells under specific activation conditions49 and mouse iTregs.50 Moreover, Helios− Tregs were found within the nTreg subsets in humans.50 Despite these inconsistencies, Helios is crucial for maintaining Foxp3+ Treg stability and suppressive activity. Its ectopic expression enhances human-engineered Treg suppression,51 and genetic deletion in Foxp3+ Tregs led to their conversion into effector T cells and a compromised ability to inhibit self-reactive T cells.25,38–40 Pharmacological degradation of Helios in human Tregs led to diminished suppressive function, as indicated by a reduced ability to inhibit effector T cell proliferation.41 Recently, Lu and colleagues reported that influenza virus infection selectively induces the accumulation and persistence of Helios+Foxp3+ Tregs in the lungs.52 Our study observed that vimentin deficiency in Tregs results in an augmented accumulation of Helios+ Tregs in the lungs after influenza. Helios+ Tregs have demonstrated an enhanced capability to suppress effector T cell activation compared to Helios− Tregs in a scurfy cell transfer model.25 Additionally, Helios+ Tregs display increased efficacy in suppressing influenza virus-specific CD8+ T cell activation, cytokine production, and proliferation in vitro.52 Our findings also revealed that Treg-specific vimentin-deficient mice exhibited a decreased proinflammatory MoAM population in the lungs and a corresponding reduction in inflammatory cytokines in the BALF following influenza infection. Given the superior ability of Helios+ Tregs to suppress effector immune cell responses, we posit that the decreased inflammatory response in Treg-specific vimentin-deficient mice is partly attributable to the increased accumulation of Helios+ Tregs in the lung. Influenza infection induces ATP release into the extracellular space, acting as a danger signal for immune cells.30 CD39 dephosphorylates ATP/ADP to AMP, which is then converted to adenosine by CD73.53 Adenosine signals through A1AR, A2AAR, A2BAR, and A3AR receptors, which support tissue protection in various injury models54–59 and are critical for resolving acute lung inflammation.31,60 Upon contact with an antigen-presenting cell, Tregs form an immunological synapse and a DPC, resulting in the spatial reorganization of signaling molecules and increased CD39/CD73 activity.53,61 In our studies, vimentin cKO Tregs displayed a significant reorganization of Treg surface proteins, including CD73, enhancing their suppressive activity (Figures 1E, 1F, S1F, S1G, and 4C–4G). Enhanced CD39/CD73 activity has been shown to promote Tregs to enter inflamed tissue.53 We reason that vimentin cKO Tregs, which express increased CD39/CD73 at homeostasis, may be better suited to infiltrate infected and inflamed lung tissues earlier than WT Tregs (Figures 3A and 3B). We posit that Tregs’ immune-suppressive and anti-inflammatory functions occur in distinct tissue microenvironments in the virus-infected lung tissue. CD73-mediated adenosine production occurs within microdomains around Tregs.62 Therefore, Treg-derived adenosine must be close to cells expressing adenosine receptors. Romio et al. revealed that CD73-produced adenosine in Tregs dampens nuclear factor kB (NF-kB) activation in nearby effector T cells expressing the A2B receptor, leading to decreased secretion of various proinflammatory cytokines and chemokines.62 We observed similar results in our vimentin cKO Tregs, and vimentin cKO mice exhibited fewer MoAMs and reduced IL-18, TNF-α, and IL-6 levels (Figure 3) compared to WT mice following viral infection. Additionally, Helios+ Tregs had higher Nt5e and Entpd1 expression (genes encoding CD73 and CD39) than Helios− Tregs (Figures 6E and 6F). These findings indicate that increased CD73-adenosine signaling in vimentin cKO mice partially contributes to the accelerated resolution of inflammation in the infected lung. Tregs can directly exert a tissue repair function, at least in part, through producing Areg, distinct from their conventional suppressive role.7,16,36,63 Throughout the course of influenza infection, we and others identified a relatively high proportion of IL-18R+ Tregs in the lungs. This subset of Tregs displayed elevated proliferative activity following infection compared to IL-18R− Tregs.7 We also confirmed previous reports indicating the importance of IL-18R signaling in Tregs for tissue protection by inducing Areg production following IL-18 stimulation.7 Notably, vimentin deficiency in Tregs led to a substantial increase in IL-18R expression, correlating with Areg production in the presence of IL-18. These findings suggest that IL-18R+ Tregs represent a notable reparative subset, conferring a tissue-protective role. Helios+ Tregs demonstrate increased expression of Areg, and the genetic deletion of Ikzf2 in Tregs reduces chromatin accessibility at the Areg promoter region compared to their WT counterparts,42 suggesting that Helios+ Tregs are involved in promoting tissue repair. Our findings reveal an enrichment and increase of Helios+IL-18R+ Tregs in the lungs of Treg-specific vimentin-deficient mice under both steady-state conditions and following influenza infection. Furthermore, the transcriptional level of Areg was upregulated in vimentin-deficient Tregs. This indicates that vimentin deficiency in Tregs creates a reparative environment in the lungs, characterized by the accumulation and persistence of Helios+IL-18R+ Tregs, ultimately contributing to enhanced recovery after influenza infection. Moreover, the increased proliferation and stability of Tregs may contribute to the enrichment of Helios+IL-18R+ Tregs in the lungs of Treg-specific vimentin-deficient mice, as IL-18R+ Tregs are more proliferative compared to IL-18R− Tregs. Helios, recognized for stabilizing Tregs during inflammation,38,39 is also associated with upregulated CD83 expression in Helios+ Tregs, which is crucial for Treg stability.42,64 In summary, vimentin deficiency in Tregs manifests as a more pronounced reparative phenotype. Vimentin is a type III intermediate filament protein that maintains cellular integrity and serves as a scaffolding for signaling.65–67 Our study corroborates previous findings that vimentin intermediate filaments are in the DPC of Tregs.20 Vimentin deficiency or disruption of the intermediate filaments by WFA results in the loss of the DPC superstructure and is associated with enhanced suppressive function. This suggests that vimentin restrains Treg function by forming the DPC structure. Furthermore, vimentin spatially sequesters IL-18R and CD73 within the DPC region of Tregs. Vimentin deficiency induces uniform expression of these proteins on the cell surface, making them more accessible to their ligands. Our findings present a novel perspective on the structural regulation of Treg functions, shedding light on the intricate interplay between protein spatial distribution and function. In summary, our findings reveal that vimentin intermediate filaments limit the localization of IL-18R and CD73 within the DPC region. However, the absence or disassembly of vimentin intermediate filaments triggers the redistribution of IL-18R and CD73 across the Treg surface, bolstering their capability to bind to ligands and subsequently augment Treg function. Limitations of the study The precise sorting of Helios+IL-18R−, Helios−IL-18R+, and Helios+IL-18R+ Treg subsets for transcriptional analysis is limited when using surrogate markers for the transcription factor Helios. Future studies will utilize Helios-GFP reporter mice to directly sort these subsets for single-cell transcriptional analysis. Further investigation is needed to unravel the molecular mechanisms through which vimentin regulates the expression of Helios and IL-18R. We reanalyzed intestinal Treg ATAC-seq data42 to highlight that Ikzf2-deficient Tregs displayed reduced chromatin accessibility at the Il18r1 and Areg loci. It is important to note that it is unknown whether intestinal Tregs are similar to lung Tregs. As such, generating an scATAC-seq dataset for lung Tregs will be necessary.42 Acknowledging a significant limitation of antibody therapy used in our study is important. Neither anti-CD73 nor anti-Areg therapy selectively targets specific cell types in vivo. Even though we demonstrated that these therapies exacerbated lung injury after viral infection, this does not necessarily imply that the effect is mediated solely through the impairment of Tregs. Finally, while pharmacological inhibitors such as WFA exhibit potential in enhancing Treg function for potential anti-inflammatory therapy, there is a need to develop and translate cell-specific delivery systems, such as nanoparticles. This is crucial to enable targeted anti-inflammatory interventions exclusively at sites where the respective mediators are deemed detrimental. STAR★METHODS RESOURCE AVAILABILITY Lead contact Further information and requests for resources should be directed to and will be fulfilled by the lead contact, Karen M. Ridge (kridge@northwestern.edu). Materials availability This study did not generate new, unique reagents. Data and code availability Data reported in this paper will be shared by the lead contact upon request. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS Ethics statement All procedures complied with United States federal guidelines and were approved by the Institutional Animal Care and Use Committee at Northwestern University (protocol IS00017822). Mice C57BL/6 Vimfl/fl mice, which harbor LoxP sequences flanking exon 2 (ENSMUSE00000139530) of the Vim gene (ENSMUSG00 000026728), were generated by Ozgene. Foxp3YFP−Cre mice (strain 016959) were originally sourced from The Jackson Laboratory. The vimentin-deficient mice (Vim−/−) on a 129/Sv6 background were generously provided by Albee Messing (University of Wisconsin, Madison, WI), and later backcrossed to the C57BL/6J background for >10 generations. Rag1−/− (strain 002216) and B6 CD45.1 (strain 002014) mice are all on the C57BL/6J background and were obtained from The Jackson Laboratory. All animals were genotyped using services provided by Transnetyx, with primers provided by The Jackson Laboratory or Ozgene (C57BL/6 Vimfl/fl mice). The appropriate littermate mice were used as WT controls (Vim+/+Foxp3YFP−Cre, WT). Ten-to 12-week-old female and male mice were used for experiments, with randomization and blinding procedures applied. Animals received water ad libitum, were housed in a controlled environment (20°C–23°C, 14-h light/10-h dark cycles), and received standard rodent chow (Envigo, 500 mg/kg of chow, approximate daily dose of 40–80 mg/kg) at the Northwestern University Center for Comparative Medicine animal facility. METHOD DETAILS Influenza viruses and viral infection Influenza virus strain A/WSN/1933(H1N1) (WSN) was grown for 48 h at 37°C with 55–60% humidity in the allantoic cavity of 10–11-day-old embryonated chicken eggs (obtained from Department of Animal Sciences, University of Illinois Urbana-Champaign, Urbana, IL). Viral titers were measured by plaque assay in Madin-Darby canine kidney epithelial (MCDK) cells (ATCC). Virus aliquots were stored in liquid nitrogen and freeze-thaw cycles were avoided. Ten-to 12-week-old WT and vimentin cKO mice were anesthetized with isoflurane and infected intratracheally with 60 plaque-forming units (PFU; lethal dose) or 30 PFU (sublethal dose) of WSN in 50 μL of sterile PBS per mouse. Viral plaque assay MDCK cells were cultured in 6-well plates until reaching 100% confluence. Lung homogenates were obtained from mice 5 days after infection. MDCK cells were washed twice with 2 mL of 1× Dulbecco’s PBS (DPBS) containing Ca2+ and Mg2+ then incubated with serial dilutions of lung homogenates or control IAV in Dulbecco’s modified Eagle medium (DMEM) containing 1% bovine serum albumin (BSA) at 37°C for 1 h. The homogenate suspension was then removed and replaced with 3 mL of overlay comprising 50% 23 replacement medium (2× DMEM, 0.12 M NaHCO3, 2% penicillin-streptomycin, and 1% HEPES), 50% Avicel (2.35%), and N-acetyl trypsin (1.5 μg/mL) for 3 days at 37°C. Subsequently, the medium was removed, the plates were washed twice with 1× DPBS with Ca2+ and Mg2+, and plaques were visualized using naphthalene blue-black stain (0.1% naphthalene black, 6% glacial acetic acid, and 13.6% anhydrous sodium acetate). Bronchoalveolar lavage fluid and cytokine measurement A 22-gauge angiocatheter was inserted and secured in the trachea, followed by 2 rounds of lung lavage with 500 μL of sterile PBS. An aliquot of the bronchoalveolar lavage fluid was taken out to count the total cell number using a Nexcelom K2 cellometer. Subsequently, the remaining BALF was centrifuged at 300 × g for 10 min. The supernatant was then collected to quantify cytokine levels. Areg levels were measured using a mouse Areg ELISA Kit (Invitrogen) according to the manufacturer’s instructions. IL-18, TNF-α, IL-6, IFN-γ, and IL-33 levels were determined using a customized multiplex assay (Life Technologies Corporation) according to the manufacturer’s instructions. Single-cell suspension preparation For the flow cytometric analysis of immune cells in the lung, mice were euthanized, and the lungs were perfused with 10 mL of Hank’s buffered salt solution (HBSS) containing Ca2+ and Mg2+ through the pulmonary artery. Lung lobes were then removed and inflated with 2.5 mL of enzyme solution (0.2 mg/mL DNase I and 2 mg/mL collagenase D in HBSS with Ca2+ and Mg2+) using a 30G needle. The tissue was minced and then dissociated using a GentleMACS dissociator (Miltenyi Biotec) according to the manufacturer’s instructions. Mixtures were incubated on a shaker at 37°C for 30 min. Processed lungs were filtered through a 70-μm cell strainer, and any remaining red blood cells (RBCs) were lysed with 1× BD PharmLyse RBC lysis buffer. Subsequently, the cells were washed and resuspended in MACS buffer (Miltenyi Biotec) for subsequent staining. To prepare single-cell suspensions from spleens and lymph nodes, tissues were mechanically dissociated with the back of a syringe plunger and filtered through a 70-μm cell strainer. After the lysis of red blood cells, the resulting cells were washed and resuspended in MACS buffer for subsequent staining. Cell staining and flow cytometry Cell counts of single-cell suspensions were determined by straining cells in a 1:1 ratio with AOPI staining solution (Nexcelom Bioscience) and counting with a K2 cellometer (Nexcelom Bioscience) before preparation for flow cytometry staining. Dead cells were stained with Fixable Viability Dye eFluor 506 (eBioscience) in 1× DPBS for 15 min at 4°C. Subsequently, cells were incubated with Fc Block (553142, clone 2.4G2; BD Pharmingen) for 10 min at 4°C. Cell surface antigens were then stained for 30 min at 4°C in the dark using a mixture of fluorophore-conjugated antibodies (see the key resources table). Cells were fixed and permeabilized using an eBioscience Foxp3/Transcription Factor staining buffer set (ThermoFiser, 00–5523-00), following the manufacturer’s instructions, before intracellular staining for Foxp3, cytotoxic T-lymphocyte-associated protein 4 (CTLA4), Helios, and Ki67. Cells were then resuspended in MACS buffer (Miltenyi Biotec) and filtered through a 40-μm cell strainer before either data acquisition on a BD Symphony A5 instrument with FACSDiva software (BD Biosciences) or sorting on a 4-way purity setting on a BD Biosciences FACSAria SORP instrument with FACSDiva software. CD4+ T cells were enriched before sorting using a Mouse CD4+ T cell Isolation Kit (Stemcell Technologies, 19852) according to the manufacturer’s protocol. FlowJo v10.8.1 software was utilized for data analysis. All antibodies used in this study are listed in the key resources table. KEY RESOURCES TABLEREAGENT or RESOURCESOURCEIDENTIFIERAntibodiesBrilliant Violet 421 anti-mouse CD326 (Ep-CAM) AntibodyBioLegendCat#118225; RRID:AB_2563983BD OptiBuild BUV395 Rat Anti-Mouse Siglec-FBDCat#740280; RRID:AB_2740019PE/Cyanine7 anti-mouse CD11c AntibodyBioLegendCat#117318; RRID:AB_493568Invitrogen eBioscience Fixable Viability Dye eFluor 506InvitrogenCat#65–0866-14; RRID:AB_3668689APC/Cyanine7 anti-human/mouse/rat CD278 (ICOS) AntibodyBioLegendCat#313529; RRID:AB_2566127Brilliant Violet 650 anti-mouse/human CD44 AntibodyBioLegendCat#103049; RRID:AB_2562600APC anti-mouse/human Helios AntibodyBioLegendCat#137221; RRID:AB_10662535APC/Cyanine7 anti-mouse CD25 AntibodyBioLegendCat#101917; RRID:AB_2650981PE/Cyanine7 anti-mouse CD357 (GITR) AntibodyBioLegendCat#126317; RRID:AB_2563385BD OptiBuild BUV395 Rat Anti-Mouse CD62LBDCat#740218; RRID:AB_2739966FOXP3 Monoclonal Antibody (FJK-16s), PE, eBioscienceLife TechnologiesCat#12–5773-82; RRID: AB_3668665PE anti-mouse Notch4 AntibodyBioLegendCat#128407; RRID:AB_1133997Brilliant Violet 421 anti-mouse IL-33Rα (IL1RL1, ST2) AntibodyBioLegendCat#145309; RRID:AB_2565634PE/Cyanine7 anti-mouse CD126 (IL-6Rα chain) AntibodyBioLegendCat#115814; RRID:AB_2565575APC anti-mouse CD218a (IL-18Rα) AntibodyBioLegendCat#157906; RRID:AB_2860735BD Horizon BUV395 Rat Anti-Mouse CD4BDCat#563790; RRID:AB_2738426FOXP3 Monoclonal Antibody (3G3), FITCThermoFisherCat#MA1–41628; RRID:AB_1074471Brilliant Violet 421 anti-mouse CD4 AntibodyBioLegendCat#100438; RRID:AB_11203718PE/Cyanine7 anti-mouse CD39 AntibodyBioLegendCat#143806; RRID:AB_2563394PE/Cyanine7 anti-mouse CD223 (LAG-3) AntibodyBioLegendCat#125226; RRID:AB_2715764FITC anti-mouse CD279 (PD-1) AntibodyBioLegendCat#135214; RRID:AB_10680238APC anti-human CD304 (Neuropilin-1) AntibodyBioLegendCat#354506; RRID:AB_11219600FOXP3 Monoclonal Antibody (FJK-16s), FITC, eBioscienceeBioscienceCat#11–5773-82; RRID:AB_465243.APC/Cyanine7 anti-mouse CD25 AntibodyBioLegendCat#101918; RRID:AB_2650982CD45 Monoclonal Antibody (30-F11), FITC, eBioscienceeBioscienceCat#11–0451-82; RRID:AB_465050.BD Pharmingen Alexa Fluor 488 Rat Anti-Mouse CD73BDCat#561545; RRID:AB_10714516PE/Cyanine7 anti-mouse CD357 (GITR) AntibodyBioLegendCat#126318; RRID:AB_2563386BD OptiBuild BV650 Rat Anti-Mouse IL-33R (ST2)BDCat#745403; RRID:AB_2742958PE/Cyanine7 anti-mouse CD194 (CCR4) AntibodyBioLegendCat#131214; RRID:AB_2244410Alexa Fluor 647 anti-mouse CD198 (CCR8) AntibodyBioLegendCat#150304; RRID:AB_2565935APC/Cyanine7 anti-mouse CD69 AntibodyBioLegendCat#104526; RRID:AB_10679041BD OptiBuild BV650 Hamster Anti-Mouse CD54BDCat#740462; RRID:AB_2739815BD OptiBuild BUV395 Mouse Anti-Human CD73BDCat#742636; RRID:AB_2740929PE/Cyanine7 anti-mouse CD3 AntibodyBioLegendCat#100219; RRID:AB_1732068FITC anti-mouse CD8a AntibodyBioLegendCat#100705; RRID:AB_312744PE anti-mouse CD45 AntibodyBioLegendCat#147711; RRID:AB_2563597APC anti-mouse CD152 AntibodyBioLegendCat#106310; RRID:AB_2087653Brilliant Violet 650 anti-mouse CD223 (LAG-3) AntibodyBioLegendCat#125227; RRID:AB_2687209BD Horizon BUV395 Rat Anti-Mouse CD39BDCat#567264; RRID:AB_2916524BD OptiBuild BUV395 Rat Anti-Mouse CD103BDCat#740238; RRID:AB_2739985BD OptiBuild BUV395 Hamster Anti-Mouse CD69BDCat#740220; RRID:AB_2739968PE/Cyanine7 anti-mouse Ki-67 AntibodyBioLegendCat#652426; RRID:AB_2632693eBioscience Fixable Viability Dye eFluor 506eBioscienceCat#65–0866-18; RRID: AB_3668691APC MS LY-6G (1A8)TonboCat#20–1276-U100; RRID: AB_3668692APC/Cyanine7 anti-mouse CD3 AntibodyBioLegendCat#100222; RRID:AB_2242784PE/Cyanine7 anti-mouse CD45 AntibodyBioLegendCat#103114; RRID:AB_312979FITC anti-mouse CD45RB AntibodyBioLegendCat#103305; RRID:AB_313012APC/Cyanine7 anti-mouse CD73 AntibodyBioLegendCat#127232; RRID:AB_2800627BD Pharmingen Purified Rat Anti-Mouse CD16/CD32 (Mouse BD Fc Block)BDCat#553142; RRID:AB_394657BD Horizon PE-CF594 Rat Anti-Mouse Siglec-FBDCat#562757; RRID:AB_2687994BD Pharmingen Alexa Fluor 700 Rat Anti-Mouse Ly-6GBDCat#561236; RRID:AB_10611860Ly-6C Monoclonal Antibody (HK1.4), eFluor 450, eBioscienceInvitrogenCat#48–5932-80; RRID:AB_10805518PE/Cyanine7 anti-mouse CD3 AntibodyBioLegendCat#100220; RRID:AB_1732057PE/Cyanine7 anti-mouse CD25 AntibodyBioLegendCat#102016; RRID:AB_312865PE/Cyanine7 anti-mouse CD152 AntibodyBioLegendCat#106314; RRID:AB_2564238APC anti-mouse/human Helios AntibodyBioLegendCat#137222; RRID:AB_10662900CD223 (LAG-3) Monoclonal Antibody(eBioC9B7W (C9B7W)), PE-Cyanine7, eBioscienceThermofisherCat#25–2231-82; RRID:AB_2573428APC anti-mouse CD304 (Neuropilin-1) AntibodyBioLegendCat#145206; RRID:AB_2562032CD25 Monoclonal Antibody (PC61.5), PE, eBioscienceeBioscienceCat#12–0251-81; RRID:AB_465607BD OptiBuild BV650 Rat Anti-Mouse CD279 (PD-1)BDCat#748266; RRID:AB_2872694PE/Cyanine7 anti-mouse/human Helios AntibodyBioLegendCat#137236; RRID:AB_2565990APC/Cyanine7 anti-mouse CD45 AntibodyBioLegendCat#103116; RRID:AB_312981BD Pharmingen FITC Rat Anti-Mouse CD73BDCat#567701; RRID:AB_2916705Anti-CD45 antibodyAbcamCat#ab10558; RRID: AB_3668696Purified anti-Keratin 5 Polyclonal Chicken AntibodyBioLegendCat#905904; RRID:AB_2721743FoxP3 Antibody - BSA FreeNovus BiologicalsCat#NB100–39002; RRID:AB_838249Vimentin Polyclonal AntibodyInvitrogenCat#PA1–10003; RRID:AB_2216267CD43 Polyclonal AntibodyInvitrogenCat#PA5–80067; RRID:AB_2747182Ezrin Rabbit pAbAbclonalCat#A0703; RRID:AB_2757351IL18R1/IL-18 R alpha Rabbit pAbAbclonalCat#A2706; RRID:AB_2764561CD73 Monoclonal Antibody (7G2)InvitrogenCat#41–0200; RRID:AB_2533492PKCq (E1I7Y) Rabbit mAbCell Signaling TechnologyCat#13643; RRID:AB_2798282Goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 568InvitrogenCat#A-11011; RRID:AB_143157Goat anti-Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 568InvitrogenCat#A-11004; RRID:AB_2534081Goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488InvitrogenCat#A-11008; RRID:AB_2534085Goat anti-Chicken IgY (H + L) Secondary Antibody, Alexa Fluor 488InvitrogenCat#A-11039; RRID:AB_2534096Goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 647InvitrogenCat#A-21244; RRID:AB_143165Goat Anti-Chicken IgY Antibody (H + L), BiotinylatedVector LaboratoriesCat#BA-9010; RRID:AB_2336114CD3e Monoclonal Antibody (145–2C11), Functional GradeeBioscienceCat#16–0031-82; RRID:AB_468847CD28 Monoclonal Antibody (37.51), Functional GradeeBioscienceCat#16–0281-82; RRID:AB_468921Purified anti-mouse IL-4 AntibodyBioLegendCat#504102; RRID:AB_315316Purified anti-mouse IFN-γ AntibodyBioLegendCat#517902; RRID:AB_10900809CD4 Monoclonal Antibody (GK1.5), PEeBioscienceCat#12–0041-81; RRID:AB_465506CD8a Monoclonal Antibody (53–6.7), PerCP-Cyanine5.5eBioscienceCat#45–0081-82; RRID:AB_469323CD39 Monoclonal Antibody (24DMS1), PE-Cyanine7eBioscienceCat#25–0391-80; RRID:AB_1210766FITC Rat Anti-Mouse CD3 Molecular Complex, Clone 17A2BD BiosciencesCat#555274; RRID:AB_395698CD19 Monoclonal Antibody (eBio1D3 (1D3)), APCeBioscienceCat#17–0193-80; RRID:AB_657650Anti-β-Actin antibody, Mouse monoclonal (clone AC-15)Sigma-AldrichCat#A1978; RRID:AB_476692CD304 (Neuropilin-1) Monoclonal Antibody (3DS304M), APCeBioscienceCat#17–3041-80; RRID: AB_3668713CD223 (LAG-3) Monoclonal Antibody (eBioC9B7W (C9B7W)), PE-Cyanine7eBioscienceCat#25–2231-80; RRID:AB_2573428CD279 (PD-1) Monoclonal Antibody (J43), FITCeBioscienceCat#11–9985-82; RRID:AB_465472CD152 (CTLA-4) Monoclonal Antibody (UC10–4B9), APCeBioscienceCat#17–1522-82; RRID:AB_2016700CD103 (Integrin alpha E) Monoclonal Antibody (B-Ly7), FITCeBioscienceCat#11–1038-42; RRID:AB_10598799Phospho-PKCδ/θ (Ser643/676) AntibodyCell Signaling TechnologyCat#9376; RRID:AB_2168834Mouse Amphiregulin AntibodyR&D SystemsCat#AF989; RRID:AB_2060663PI3 Kinase p110 δ (D1Q7R) Rabbit mAbCell Signaling TechnologyCat#34050; RRID:AB_2799043Purified Rat Anti-Mouse CD43, Clone S7BDCat#552366; RRID:AB_394368FOXO3A Polyclonal AntibodyInvitrogenCat#PA1–805; RRID:AB_2294274Anti-FOXO3A (phospho S253) antibodyAbcamCat#ab47285; RRID:AB_869819Anti-beta Actin antibody - Loading ControlAbcamCat#ab8229; RRID:AB_306374IRDye 680RD Goat anti-Mouse IgG Secondary AntibodyLi-CorCat#926–68070; RRID:AB_10956588IRDye 800CW Goat anti-Rabbit IgG Secondary AntibodyLi-CorCat#926–32211; RRID:AB_621843Brilliant Violet 421 anti-mouse CD326 (Ep-CAM) AntibodyBioLegendCat#118225; RRID:AB_2563983Bacterial and virus strainsInfluenza A/WSN/1933(H1N1) virusDr. Robert Lamb68N/AChemicals, peptides, and recombinant proteinsWithaferin AChromaDexASB-00023250; CAS: 5119–48-2NSC668394Sigma-AldrichCat#341216Recombinant mouse IL-2 (carrier-free)BioLegendCat#575402Recombinant human TGF-β1 (carrier-free)BioLegendCat#580702Recombinant Murine IL-6PeprotechCat#216–16Recombinant Murine IL-4PeprotechCat#214–14Recombinant Murine IFN-γPeprotechCat#315–05Adenosine 5′-(α,β-methylene)diphosphateSigma-AldrichM3763; CAS: 3768–14-7Critical commercial assaysVectastain ABC-HRP Kit, Peroxidase (Rabbit IgG)Vector LaboratoriesCat#PK-4001DAB Substrate Kit, Peroxidase (HRP), with NickelVector LaboratoriesCat#SK-4100Avidin/Biotin Blocking KitVector LaboratoriesCat#SP-2001EasySep Mouse Naive CD8+ T cell Isolation KitStem CellCat#19858Mouse IL-17 Quantikine ELISA KitR&D SystemsCat#M1700Mouse Amphiregulin ELISA KitSigma-AldrichCat#RAB0020–1KTMouse Amphiregulin DuoSet ELISAR&D SystemsCat#DY989Mouse Keratinocyte Growth Factor, KGF/FGF-7 ELISA Kit MyBioSource.com Cat#MBS701004Deposited dataHelios+ and Helios- Treg TranscriptomeThornton AM et al.25GEO: GSE117989Ikzf2-deficient or sufficient Tregs chromatin accessibilityRamanan D et al.42GEO: GSE240657Experimental models: Organisms/strainsVim−/− miceThis studyN/AVimfl/fl miceThis studyN/AFoxp3YFP-Cre miceThe Jackson LaboratoryStrain #:016959RRID:IMSR_JAX:016959Rag1−/− miceThe Jackson LaboratoryStrain #:002216RRID:IMSR_JAX:002216B6 CD45.1The Jackson LaboratoryStrain #:002014RRID:IMSR_JAX:002014OligonucleotidesGapdh qPCR Forward AGGTCGGTGTGAACGGATTTGIntegrated DNA TechnologiesN/AGapdh qPCR Reverse TGTAGACCATGTAGTTGAGGTCAIntegrated DNA TechnologiesN/ANt5e qPCR Forward TCCTGCAAGTGGGTGGAATCIntegrated DNA TechnologiesN/ANt5e qPCR Reverse AGATGGGCACTCGACACTTGIntegrated DNA TechnologiesN/AEntpd1 qPCR Forward GGATGCGGGGTCATCTCACIntegrated DNA TechnologiesN/AEntpd1 qPCR Reverse TTCCAGGACCTTTCACTTGGCIntegrated DNA TechnologiesN/AVim qPCR Forward CGGCTGCGAGAGAAATTGCIntegrated DNA TechnologiesN/AVim qPCR Reverse CCACTTTCCGTTCAAGGTCAAGIntegrated DNA TechnologiesN/AAreg qPCR Forward GGTCTTAGGCTCAGGCCATTAIntegrated DNA TechnologiesN/AAreg qPCR Reverse CGCTTATGGTGGAAACCTCTCIntegrated DNA TechnologiesN/ANrp1 qPCR Forward GCTGTGAAGTGGAAGCACCTIntegrated DNA TechnologiesN/ANrp1 qPCR Reverse CTGTGGCCAGGACAGTGGIntegrated DNA TechnologiesN/ALag3 qPCR Forward CCAGGCCTCGATGATTGCTAIntegrated DNA TechnologiesN/ALag3 qPCR Reverse ACGCGGTGAGTTGTAGACAGIntegrated DNA TechnologiesN/ASoftware and algorithmsAxioVision 4.8.2Zeiss https://www.zeiss.com/microscopy/en/service-support/support/discontinuedproducts.html Zen Lite 3.4Zeiss https://www.zeiss.com/microscopy/en/products/software.html NIS-Elements AR 4.50.00Nikon https://www.microscope.healthcare.nikon.com/products/software NIS-Elements AR 5.11.02Nikon https://www.microscope.healthcare.nikon.com/products/software NIS-Elements Viewer 5.21Nikon https://www.microscope.healthcare.nikon.com/products/software TissueFAXSTissueGnostics https://tissuegnostics.com/ HistoQuest 7.0.1.0153TissueGnostics https://tissuegnostics.com/ PrismGraphPad https://www.graphpad.com/scientific-software/prism/ Odyssey 3.0Li-Cor https://www.licor.com/bio/ FlowJo 10.8.1BD Biosciences https://www.flowjo.com/ BioRenderBioRender https://www.biorender.com KEY RESOURCES TABLE Histology For histopathological preparation of lung tissues, mice were euthanized, and the lungs were perfused through the pulmonary artery with 10 mL of HBSS with Ca2+ and Mg2+. A 22-gauge angiocatheter was sutured into the trachea. Subsequently, the lungs were removed en bloc and were inflated with up to 0.8 mL of 4% paraformaldehyde (PFA; Electron Microscopy Sciences 15713-S) at a pressure not exceeding 16 cm H2O. The lungs were then fixed in 4% PFA. For histopathological preparation of the colon, the entire length, starting from the anus, was meticulously excised. It was then dissected free of the mesentery, from below the cecum to just above the anal opening. Subsequently, the colon was cleansed of fecal material by injecting PBS. A small (0.5 cm) segment from the distal portions of the colon was acquired and subsequently fixed in 4% PFA. Tissue samples were fixed in 4% PFA for 48–72 h at 4°C. Tissue processing, embedding, sectioning (4 μm thickness), and staining with hematoxylin and eosin (H&E) were performed by the Mouse Histology and Phenotyping Laboratory at Northwestern University Feinberg School of Medicine. Images were acquired by means of a TissueGnostics TissueFAXS automated slide imaging system at the Center for Advanced Microscopy at Northwestern University Feinberg School of Medicine. Lung pathology scoring was performed independently on specimens from 4 mice per group by each of 3 blinded investigators. The scores in each group were pooled for analysis. The scoring system was based on that described by Semaniakou et al.69 to assess immune cell infiltration and thickening of the alveoli and peribronchial areas. Scores ranged from 0 (none) to 5 (>45%) of lumens with involvement in each whole-lung section. The severity of the inflammation in colitis was assessed and scored in a blinded manner.22 The score ranged from 0 (no inflammation) to 5 (severe ulceration, diffuse transmural inflammation, and crypt loss). Immunohistochemistry Immunohistochemistry for CD45 was performed by the Mouse Histology and Phenotyping Laboratory at Northwestern University Feinberg School of Medicine. Immunohistochemistry for Krt5 was performed as follows. Tissue sections were deparaffinized in xylene and rehydrated in a series of graded ethanol solutions. Antigen retrieval was performed by incubating the sections with 10 mM citric acid trisodium salt dihydrate (Sigma-Aldrich C8532) at 95°C–99°C for 20 min. After blocking, the sections were incubated with Krt5 primary antibody (BioLegend 905903/905904, 1:10,000) overnight at 4°C, followed by 30 min of incubation at room temperature with biotinylated secondary antibodies and 30 min of incubation with ABC reagent (Vector Laboratories PK-4001) at room temperature, as per the manufacturer’s instructions. Immunolabeling was visualized using a DAB (3,3’diaminobenzidine) substrate kit (Vector Laboratories SK-4100), and the sections were counterstained with Gill 3 hematoxylin (Epredia 72604) before mounting. Images were acquired by means of a TissueGnostics TissueFAXS automated slide imaging system, and quantification was performed by means of TissueGnostics HistoQuest software, at the Center for Advanced Microscopy at Northwestern University Feinberg School of Medicine. The segmentation method used was Nuclear Segmentation, and the color separation method used was Single Reference Shade V2. For each tissue section, regions of interest (ROIs) were drawn around lobes of lung parenchyma, and within ROIs exclusion areas were drawn around large airway lumens and debris. For each specimen, the area is the sum of the areas of the ROIs (minus any exclusion areas), the total cell count is the sum for all ROIs of the counts in the upper right quadrants in the scattergrams of nuclear area versus nuclear mean intensity, and the Krt5-positive cell count is the sum for all ROIs of the counts in the upper right quadrants in the scattergrams of Krt5 mean intensity versus nuclear mean intensity for all ROIs. Immunofluorescence Rehydrated tissue sections were quenched in quenching buffer with 10 mM glycine (Sigma-Aldrich G8898) in PBS for 30 min at room temperature. Subsequently, antigen retrieval was performed by incubating the slides with 10 mM citric acid trisodium salt dihydrate (Sigma-Aldrich C8532) at 95°C–99°C for 20 min. Following blocking in 5% normal goat serum (NGS; Jackson ImmunoResearch 005–000-001), slides were treated with primary antibodies against Krt5 (BioLegend 905903/905904, 1:500) and Foxp3 (Novus Biologicals NB100–39002, 1:1000) overnight at 4°C. Afterward, secondary antibodies (Invitrogen A-11039 and A-21244) were applied for 1 h at room temperature, along with Hoechst 33342 nuclear dye (Invitrogen H3570) for 5 min at room temperature. The slides were mounted using ProLong Diamond Antifade Mountant (Invitrogen P36965 or P36970) and stored at −20°C. Immunofluorescence images were acquired with a Zeiss Axioplan 2 microscope. Immunocytochemistry Tregs were counted and adjusted to a density of 106 live cells/mL, then cytospun onto charged slides (Fisherbrand 12–550-15) at 1000 rpm for 3 min using a Shandon Cytospin 3. The cells were fixed in 2.7% PFA for 10 min, permeabilized in 0.10% Triton X-100 for 3 × 3 min, and blocked for 1–1.5 h in 5% NGS at room temperature. Subsequently, cells were incubated with primary antibodies (vimentin [Invitrogen PA1–10003, 1:500]; Foxp3 [Novus Biologicals NB100–39002, 1:500]; CD43 [Invitrogen PA580067, 1:100]; ezrin [Abclonal A0703, 1:100]; IL18R1 [Abclonal A2706, 1:100]; CD73 [Invitrogen 41–0200, 1:50]; and PKC-θ [Cell Signaling Technologies 13643, 1:400]) either for 2 h at room temperature or overnight at 4°C. After primary antibody treatment, cells were incubated for 1 h at room temperature with secondary antibodies (Invitrogen A-11039, A-11011, A-11004, A21235, or A-11008) and Hoechst 33342 nuclear dye (Invitrogen H3570, 1:10,000). Finally, cells were mounted using ProLong Diamond Antifade Mountant (Invitrogen P36965 or P36970) and stored at −20°C. Images were acquired by a Zeiss Axioplan 2 microscope or Nikon A1+ laser scanning confocal system. Mouse in vitro Treg differentiation The spleens of mice were mechanically dissociated using the back of a syringe plunger and filtered through a 70-μm cell strainer to obtain a single-cell suspension. Mouse CD4+ T cells were then isolated following the manufacturer’s protocol using a Mouse CD4+ T cell Isolation Kit (Stemcell Technologies, 19852). Subsequently, these cells were stimulated for 5 days with a 1:1 ratio of Dynabeads Mouse T-Activator CD3/CD28 (Invitrogen, 11452D) along with polarizing cytokines (5 ng/mL IL-2, 5 ng/mL TGF-β, 2 ng/mL anti–IFN-γ, and 2 ng/mL anti–IL-4), to generate iTregs. WFA (ChromaDex ASB-00023250) was used at a concentration of 1 μM in dimethyl sulfoxide (DMSO; Sigma-Aldrich D2650). In vitro Treg suppression assays Mouse T cells were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 1× nonessential amino acids, 2 mM L-glutamine, 1 mM sodium pyruvate, and 50 μM β-mercaptoethanol (complete RPMI 1640). In vitro Treg suppression assays were performed as described elsewhere.70 In brief, CD4+CD25− conventional effector T cells labeled with CellTrace carboxyfluorescein diacetate succinimidyl ester (CFSE; Invitrogen, C34554) served as responder T cells. These cells were cultured with a 1:1 ratio of Dynabeads Mouse T-Activator CD3/CD28 (Invitrogen, 11452D) in the presence of varying numbers of Tregs for 72 h. The division of responder T cells was assessed by dilution of CellTrace CFSE, and the suppression percentage was calculated using FlowJo v10.8.1 software. In vivo Treg suppression assay Splenocyte samples were enriched for CD4+ T cells by negative selection with a Mouse CD4+ T cell Isolation Kit (Stemcell Technologies, 19852). Cells were then stained and sorted into CD4+CD45RBhiCD25− or CD4+CD45RBloCD25+ populations. Rag1−/− recipient mice were intravenously injected with 4 × 105 CD45.1+CD4+CD45RBhiCD25− Tcon cells, either alone or in the presence of 1 × 105 or 0.5 × 105 CD45.2+CD4+CD45RBloCD25+ Tregs. Mice were weighed at the time of injection and monitored 3 times weekly until tissue harvest. RNA isolation and real-time PCR RNA was isolated using a TRIzol Plus RNA Purification Kit (Invitrogen 12183555) following the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized using an iScript cDNA Synthesis Kit (Bio-Rad 170–8891). Real-time PCR was performed on a CFX384 Real Time System (Bio-Rad) using iQ SYBR Green Supermix (Bio-Rad) and 200 nM (final concentration) primers. The following primers were used: Nt5e forward, TCCTGCAAGTGGGTGGAATC; Nt5e reverse, AGATGGGCACTCGACACTTG; Entpd1 forward, GGATGCGGGGTCATCTCAC; Entpd1 reverse, TTCCAGGACCTTTCACTTGGC; Vim forward, CGGCTGCGAGAG AAATTGC; Vim reverse, CCACTTTCCGTTCAAGGTCAAG; Areg forward, GGTCTTAGGCTCAGGCCATTA; Areg reverse, CGCTTATGGTGGAAACCTCTC; Nrp1 forward, GCTGTGAAGTGGAAGCACCT; Nrp1 reverse, CTGTGGCCAGGACAGTGG; Lag3 forward, CCAGGCCTCGATGATTGCTA; Lag3 reverse, ACGCGGTGAGTTGTAGACAG; Gapdh forward, AGGTCGGTGTGAA CGGATTTG; and Gapdh reverse, TGTAGACCATGTAGTTGAGGTCA. Data were analyzed using the comparative Ct method (ΔΔCt method).71 Western blot Tregs were lysed in cell lysis buffer (Cell Signaling), resolved on 12% polyacrylamide gels (Bio-Rad), and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat milk for 1 h at room temperature, then incubated with primary antibodies against vimentin (ThermoFisher PA1–10003, 1:1000), PKC-θ (clone E1I7Y, Cell Signaling Technology 13643, 1:1000), phospho-PKC-δ/θ (Ser643/676, Cell Signaling Technology 9376, 1:1000), FOXO3a (Invitrogen PA1–805, 1:1000), phospho-FOXO3a (S253, Abcam ab47285, 1:1000), and β-actin (Abcam ab8229, 1:1000) overnight at 4°C. After incubation with a goat anti-mouse IRDye 680RD (Li-Cor C50721, 1:5,000) or goat anti-rabbit IRDye 800CW (Li-Cor C60321, 1:10,000) for 1 h at room temperature, the membranes were directly imaged using an Odyssey Fc Analyzer (Li-Cor). QUANTIFICATION AND STATISTICAL ANALYSIS All figures and statistical analyses were performed using GraphPad Prism 9 software. Unpaired Student’s t-tests were used to compare the means of two groups. A one-way or two-way analysis of variance (ANOVA) was used to compare the different groups. p-values are represented in figures as follows: NS, not significant (p > 0.05); *p < 0.05; **p < 0.005; ****p < 0.0001. Error bars on all figures represent standard deviation (SD) unless otherwise specified as standard error of mean (SEM). Sample size (n) is indicated in the figure legend. Supplementary Material Highlights Vimentin limits the suppressive function of Tregs Vimentin limits the suppressive function of Tregs Vimentin sequesters Treg receptors to DPC, limiting receptor-ligand interactions Vimentin sequesters Treg receptors to DPC, limiting receptor-ligand interactions Vimentin decreases Helios+IL-18R+ Tregs in the influenza-infected lung Vimentin decreases Helios+IL-18R+ Tregs in the influenza-infected lung Vimentin suppresses IL18R-mediated increase in Areg and impairs lung tissue repair Vimentin suppresses IL18R-mediated increase in Areg and impairs lung tissue repair ACKNOWLEDGMENTS We thank the Robert H. Lurie Comprehensive Cancer Center of Northwestern University for the use of the Flow Cytometry Core Facility. Histology services were provided by the Northwestern University Mouse Histology and Phenotyping Laboratory. Bright-field imaging was performed at the Northwestern University Center for Advanced Microscopy. Confocal microscopic imaging was kindly supported by Robert Goldman and Vladimir I. Gelfand, Department of Cell and Developmental Biology, Northwestern University. R.M. was supported by a David and Christina Cugell Fellowship; A.D.P. was supported by the Gorter Family Foundation, NIH NHLBI 5T32HL076139-18, the ATS Research Program, and the Parker B. Francis Fellowship Program; T.P.O.S. was supported by NIH 5T32HL076139; B.D.S. was supported by NIH R01HL149883, R01HL153122, P01HL154998, P01AG049665, U19AI135964, and U19AI181102; B.M.C. was supported by the American Lung Association and NIH R01HL168672 and P01HL154998; and K.M.R. was supported by NIH P01 HL154998, R01HL166807, and P01AG049665. Footnotes DECLARATION OF INTERESTS B.D.S. holds United States patent no. US 10,905,706 B2, “Compositions and Methods to Accelerate Resolution of Acute Lung Inflammation,” and serves on the scientific advisory board of Zoe Biosciences. SUPPLEMENTAL INFORMATION Supplemental information can be found online at https://doi.org/10.1016/j.celrep.2024.115056. REFERENCES Associated Data Supplementary Materials Data Availability Statement Data reported in this paper will be shared by the lead contact upon request. This paper does not report original code. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.