Fig 2: Colonic epithelial ablation of Tnfr2 disrupts tissue repair in acute injury(A) whole- mount images of distal colon at day 12 after DSS exposure was shown in Vil1-Cre; Tnfr2flox/flox and Tnfr2flox/flox mice (8weeks). TNFR2-deficient crypts labeled with methyl green had altered crypt morphology (A′) with enlarged lumen (A′-A″, um2; n = 3) and crypt circumference (A″, p = 0.008), and increased proliferation (A′-A‴; pH-H3, red; n = 4) compared to control (A-A‴; n = 4). (a-a’) Rotation of en face images in A and A′ showed enlarged lumen (arrow) and larger injured area (star) in Vil1-Cre; Tnfr2flox/flox (a’) compared to Tnfr2flox/flox (a). Black arrow shows the apical side in a. (∗∗p = 0.001; mean ± SD, unpaired t test). Scale bar: 200um.(B) Transcriptome sequencing was performed for sorted EpCAM+ (epithelial) cells from the injured regions of distal colon at day 12 after DSS exposure and day 0 (control) in Vil1-Cre; Tnfr2flox/flox and Tnfr2flox/flox mice. (B) PCA plot showed clusters of water-versus DSS-treated samples in both groups. (B′-B‴) Data analysis compared the differential activation of injury-induced pathways at day 12 between Vil1-Cre; Tnfr2flox/flox and Tnfr2flox/flox mice. It showed the enhancement of proliferation and regenerative signaling (B′), elevated fetal signaling (B″, p < 0.05), downregulation of differentiated cells, and enrichment of progenitor cells (B‴) in Vil1-Cre; Tnfr2flox/flox compared to control.(C–F) In situ hybridization with Ly6a (brown) targeted probes showed elevated level of Ly6a in impaired (F′-F″) and healed (F‴; arrow) crypts of TNFR2-deficient epithelium at day 12 compared to Tnfr2flox/flox (E-E‴, arrow) and control (day0, C–D). (C–F) The zoom-out views of Swiss roll colons were shown. Red squares with assigned numbers show the location of zoom-in images. Scale bars: 100 μm, 50 μm.(G and H) Ly6a transcript level remained high in distal colon and healed crypts at day 35 in TNFR2-ablated epithelium (H′-H”; brown) compared to Tnfr2flox/flox (G′-G″). (G-H) The zoom-out views of Swiss roll colons were shown. Red squares with assigned numbers are showing the location of zoom-in images. Scale bars: 100um, 50um.(I) The density of Ly6a signal was significantly increased at day 12 (∗p < 0.02; n = 5) and day 35 (∗∗p = 0.001; n = 5) in Vil1-Cre; Tnfr2flox/flox compared to Tnfr2flox/flox mice. Ly6a density was not elevated at day 0 and 8 in TNFR2-ablated mice. (mean ± SD, unpaired t test).(J) Injured areas were significantly larger at day 12 (∗p < 0.02; n = 4) and day 35 (∗∗p = 0.001; n = 3), but not at day 8, in TNFR2-deficent epithelium compared to Tnfr2flox/flox mice. (mean ± SD, unpaired t test).(K and L) Immunohistochemistry staining of MUC-2 and REG-4 proteins showed reduced number of goblet cells (K′, brown) and deep secretory cells (L′, brown) at day 12 in Vil1-Cre; Tnfr2flox/flox mice compared to Tnfr2flox/flox (K, L).(M) Statistical analysis showed the significant reduction of goblet cells (MUC-2) and deep secretory cells (REG-4) at day 12 in TNFR2-deficient epithelium. (∗∗p = 0.001, ∗∗∗p = 0.0005; mean ± SD, unpaired t test). Scale bar: 50 um.

Fig 3: Adar1ΔCd11c mice promotes increased resistance to early SARS-CoV-2 infectionAdar1ΔCd11c (red) or control (gray) mice were infected intranasally with 106 PFUs of SARS-CoV-2 (B.1.351 strain) and sacrificed 3 days post-infection (dpi).(A) Lung viral titer at 3 dpi measured by plaque assay.(B) Weight loss dynamics. Representative results from 3 independent experiments with n = 15–18 per group, represented as mean ± SEM. Unpaired t test, ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001.(C) Representative H&E-stained sections of lung from naive and infected mice.(D) Gating scheme for isolation of lung CD45+ immune cells, EpCAM+ epithelial cells, and CD31+ endothelial cells.(E) Heatmap of differential Reactome pathways between Adar1ΔCd11c versus wild type (WT; FDR < 0.05, normalized enrichment score [NES] > 1.8) across lung cell types.

Fig 4: Biodistribution and cellular uptake of nebulized BIEXO@IPA. (a-d) In vivo biodistribution of DiR-labeled BIEXO@IPA following intravenous (i.v.) administration (a, b) or inhalation (inh.) (c, d), visualized by an IVIS imaging system at 2, 6, 12, 24 and 48 h post-administration (n = 3). (e, f) Dose-dependent pulmonary accumulation 12 h post-inhalation (n = 3). (g) Total lung cell uptake of BIEXO@IPA, IEXO@IPA, or Lipo@IPA by flow cytometry at 24 h post-inhalation (n = 3). (h-j) Comparative analysis of BIEXO@IPA, IEXO@IPA, and Lipo@IPA uptake distributions across different cell populations, presented as percentages of total uptake-positive cells (n = 3). (k-n) Flow cytometric analysis showing the uptake of BIEXO@IPA, IEXO@IPA, or Lipo@IPA by specific cell populations, including LLC-MSLN tumor cells (k), macrophages (F4/80+CD11b+) (l), epithelial cells (CD45−CD31−EpCAM+) (m), and dendritic cells (DCs, CD45+CD11c+CD24+) (n), expressed as a percentage of total lung cells at 24 h post administration (n = 3). The data are presented as the mean ± SD. Statistical significance was determined via one-way ANOVA followed by Tukey’s multiple comparisons test (g, k-n). *p < 0.05, **p < 0.01, ***p < 0.001. ns, Not significant

Fig 5: Adar1ΔCd11c mice promotes increased resistance to early SARS-CoV-2 infectionAdar1ΔCd11c (red) or control (gray) mice were infected intranasally with 106 PFUs of SARS-CoV-2 (B.1.351 strain) and sacrificed 3 days post-infection (dpi).(A) Lung viral titer at 3 dpi measured by plaque assay.(B) Weight loss dynamics. Representative results from 3 independent experiments with n = 15–18 per group, represented as mean ± SEM. Unpaired t test, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.(C) Representative H&E-stained sections of lung from naive and infected mice.(D) Gating scheme for isolation of lung CD45+ immune cells, EpCAM+ epithelial cells, and CD31+ endothelial cells.(E) Heatmap of differential Reactome pathways between Adar1ΔCd11c versus wild type (WT; FDR < 0.05, normalized enrichment score [NES] > 1.8) across lung cell types.