Journal: bioRxiv

Article Title: IL-1β and TNF drive endothelial dysfunction and coagulopathy in acute COVID-19

doi: 10.64898/2026.03.21.713333

Figure Lengend Snippet: (A-D) NHBE/HMVEC-L co-cultures were infected with SARS-CoV-2 (MOI 1) or mock-infected, and cells were fixed at 24, 48, and 72h post-infection. A) Immunofluorescence staining of ICAM-1 (green), F-actin (phalloidin; grey), and nuclei (DAPI; blue) in HMVEC-L in the basal compartment of the co-culture. Images are representative of three independent experiments. B) Quantification of HMVEC-L ICAM-1 intensity analysed by 2-way ANOVA with Sidak’s multiple comparison’s test. C) Immunofluorescence staining of VE-cadherin (magenta) in HMVEC-L in the basal compartment of the co-culture. Images are representative of three independent experiments. D) Quantification of the percentage of gaps in the endothelial monolayer was analysed by 2-way ANOVA with Sidak’s multiple comparison’s test. E-H) NHBE/HMVEC-L co-cultures were infected with SARS-CoV-2 (MOI 1) or mock-infected, and cells were fixed at 72h post-infection. E) Immunofluorescence staining of Zombie Red (magenta) in HMVEC-L in the basal compartment of the co-culture. Images are representative of four independent experiments. F) The percentage of Zombie Red-positive cells between conditions was analysed using the Kruskal-Wallis test with Dunn’s multiple testing correction. G) Immunofluorescence staining and imaging of CellTrace Yellow-labelled platelets (yellow), VE-cadherin (magenta), and nuclei (DAPI; blue) in HMVEC-L in the basal compartment of the co-culture. Images are representative of three independent experiments. H) Gaps in the cellular monolayer were outlined relative to VE-cadherin staining, and only the platelets present in gaps were quantified. I) The number of platelets (CellTrace Yellow positive particles larger than 2 mm) in gaps was quantified and analysed by the Kruskal-Wallis test with Dunn’s multiple testing correction. Scale bar for all images = 50 µm. 5 ROIs per experiment were quantified (small dots) and are colour-coded per experiment. The average of the 5 ROIs is represented by the large dot (colour-coded by experiment), and the data show the mean ± SEM. Asterisks indicate statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001.

Article Snippet: Sections were then incubated overnight at 4 °C with primary antibodies against mouse ICAM-1 (BioXCell; BE0020-1) and mouse CD31 (New England Biolabs; 77699S).

Techniques: Infection, Immunofluorescence, Staining, Co-Culture Assay, Imaging

Journal: bioRxiv

Article Title: IL-1β and TNF drive endothelial dysfunction and coagulopathy in acute COVID-19

doi: 10.64898/2026.03.21.713333

Figure Lengend Snippet: (A-E) NHBE/HMVEC-L co-cultures infected with SARS-CoV-2 (MOI 1) and then treated with 100 mg/mL dexamethasone or media alone immediately post-infection. Cells were fixed at 72h post-infection. A) Immunofluorescence staining of ICAM-1 (green) in HMVEC-L in the basal compartment of the co-culture. Images are representative of 2 independent experiments. B) Quantification of HMVEC-L ICAM-1 intensity, where data shows mean + SD. C) Immunofluorescence staining of VE-cadherin (magenta) in HMVEC-L in the basal compartment of the co-culture. Images are representative of three independent experiments. D) Quantification of the percentage of gaps in the endothelial monolayer was analysed by one-way ANOVA with Sidak’s multiple comparison’s test. E) Viral titres from the apical compartment of SARS-CoV-2-infected NHBE/HMVEC-L co-cultures, untreated or treated with 100 mg /mL dexamethasone, at 72h post-infection. n = 3 independent experiments, analysed by unpaired, two-way t -test. F) Schematic of supernatant transfer experiment. G-J) NHBE monocultures were infected with SARS-CoV-2 for 48h. The supernatant from the basal compartment was then transferred onto HMVEC-L. 100 mg /mL dexamethasone or PBS was added to the NHBE basal supernatants before they were transferred onto HMVEC-L. After 24h, HMVEC-L were fixed for immunofluorescence staining. G) Immunofluorescence staining of ICAM-1 (green) in HMVEC-L. Images are representative of 3 independent experiments. H) Quantification of HMVEC-L ICAM-1 intensity analysed by one-way ANOVA with Sidak’s multiple comparison’s test. I) Immunofluorescence staining of VE-cadherin (magenta) in HMVEC-L. Images are representative of three independent experiments. J) Quantification of the percentage of gaps in the endothelial monolayer was analysed by one-way ANOVA with Sidak’s multiple comparison’s test. Scale bar for all images = 50 µm. 5 ROIs per experiment were quantified (small dots) and are colour-coded per experiment. The average of the 5 ROIs is represented by the large dot (colour-coded by experiment), and the data show the mean ± SEM. Asterisks indicate statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001.

Article Snippet: Sections were then incubated overnight at 4 °C with primary antibodies against mouse ICAM-1 (BioXCell; BE0020-1) and mouse CD31 (New England Biolabs; 77699S).

Techniques: Infection, Immunofluorescence, Staining, Co-Culture Assay

Journal: bioRxiv

Article Title: IL-1β and TNF drive endothelial dysfunction and coagulopathy in acute COVID-19

doi: 10.64898/2026.03.21.713333

Figure Lengend Snippet: (A-H) NHBE/HMVEC-L co-cultures were infected with SARS-CoV-2 (MOI 1) or mock-infected and treated with 10 mg/mL anti-TNF (Adalimumab) immediately post-infection. Cells were fixed at 72h post-infection. A) Immunofluorescence staining for ICAM-1 (green), F-actin (Phalloidin; grey), and nuclei (DAPI; blue). Images are representative of n = 3 independent experiments. B) ICAM-1 intensity between conditions was analysed by one-way ANOVA, with Sidak’s multiple comparison test. C) Immunofluorescence staining for VE-cadherin (magenta), F-actin (Phalloidin; grey), and nuclei (DAPI; blue). Images are representative of n = 3 independent experiments. D) Quantification of gaps in the endothelial monolayer under different conditions was determined by calculating the percentage of the image area covered by gaps, and analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. E) Zombie Red-stained cells (magenta) indicate cells (containing F-actin and nuclei) undergoing cell death. Images are representative of n = 3 independent experiments. F) The percentage of Zombie Red positive cells was analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. G) Immunofluorescent staining of CellTrace Yellow-labelled platelets incubated with HMVEC-L. Images are representative of n = 3 independent experiments. H) The number of platelets (CellTrace Yellow positive particles larger than 2 mm) in gaps was quantified and analysed by the Kruskal-Wallis test with Dunn’s multiple testing correction. I) NHBE monocultures and NHBE/HMVEC-L co-cultures were infected with SARS-CoV-2 (MOI 1) or mock-infected, and TNF levels in the apical and basal supernatants were analysed at 24, 48, and 72h post-infection. Data show the mean ± SEM of 3 independent experiments, analysed by 2-way ANOVA. (J-M) NHBE monocultures were infected with SARS-CoV-2 for 48h. The supernatant from the basal compartment was then transferred onto HMVEC-L. Anti-TNF (10 mg/mL) or PBS was added to the NHBE basal supernatants before they were transferred onto HMVEC-L. After 24h, HMVEC-L were fixed for immunofluorescence staining. J) Immunofluorescence staining of ICAM-1 (green) in HMVEC-L. Images are representative of 3 independent experiments. K) Quantification of HMVEC-L ICAM-1 intensity was analysed by one-way ANOVA with Sidak’s multiple comparison’s test. L) Immunofluorescence staining of VE-cadherin (magenta) in HMVEC-L. Images are representative of three independent experiments. M) Quantification of the percentage of gaps in the endothelial monolayer was analysed by one-way ANOVA with Sidak’s multiple comparison’s test. Scale bar for all images = 50 µm. 5 ROIs per experiment were quantified (small dots) and are colour-coded per experiment. The average of the 5 ROIs is represented with the large dot (colour-coded per experiment), and the data shows mean ± SEM. Asterisks indicate statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001.

Article Snippet: Sections were then incubated overnight at 4 °C with primary antibodies against mouse ICAM-1 (BioXCell; BE0020-1) and mouse CD31 (New England Biolabs; 77699S).

Techniques: Infection, Immunofluorescence, Staining, Comparison, Incubation

Journal: bioRxiv

Article Title: IL-1β and TNF drive endothelial dysfunction and coagulopathy in acute COVID-19

doi: 10.64898/2026.03.21.713333

Figure Lengend Snippet: (A-F) NHBE/HMVEC-L co-cultures were infected with SARS-CoV-2 (MOI 1) or mock-infected and treated with 10 mg/mL Anakinra immediately post-infection. Cells were fixed at 72h post-infection. A) Immunofluorescence staining for ICAM-1 (green), F-actin (Phalloidin; grey), and nuclei (DAPI; blue). Images are representative of n = 5 independent experiments. B) ICAM-1 intensity between conditions was analysed by one-way ANOVA, with Sidak’s multiple comparison test. C) Immunofluorescence staining for VE-cadherin (magenta), F-actin (Phalloidin; grey), and nuclei (DAPI; blue). Images are representative of n = 3 independent experiments. D) Quantification of gaps in the endothelial monolayer under different conditions was determined by calculating the percentage of the image area covered by gaps, and analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. E) Zombie Red-stained cells (magenta) indicate cells (containing F-actin and nuclei) undergoing cell death. Images are representative of n = 3 independent experiments. F) The percentage of Zombie Red positive cells was analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. G) IL-1β levels in the apical supernatant of NHBE monocultures infected with SARS-CoV-2 (MOI 1) or mock-infected, at 24, 48, and 72h post-infection (n = 3 independent experiments). H) TNF levels in the apical (left panel) and basal (right panel) supernatant of SARS-CoV-2-infected NHBE/HMVEC-L co-cultures, untreated or treated with Anakinra, at 24, 48, and 72h post-infection (n = 2 independent experiments). Scale bar for all images = 50 µm. 5 ROIs per experiment were quantified (small dots) and are colour-coded per experiment. The average of the 5 ROIs is represented with the large dot (colour-coded per experiment), and the data shows mean ± SEM. Asterisks indicate statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001.

Article Snippet: Sections were then incubated overnight at 4 °C with primary antibodies against mouse ICAM-1 (BioXCell; BE0020-1) and mouse CD31 (New England Biolabs; 77699S).

Techniques: Infection, Immunofluorescence, Staining, Comparison

Journal: bioRxiv

Article Title: IL-1β and TNF drive endothelial dysfunction and coagulopathy in acute COVID-19

doi: 10.64898/2026.03.21.713333

Figure Lengend Snippet: A) Representative immunohistochemistry (IHC) images of lungs from SARS-CoV-2-infected (10 4 TCID50) or mock-infected wild-type (WT), Tnf -/- , Il1b -/- , and Tnf -/- IL1b -/- mice, harvested at 3 days post-infection (dpi). Tissues were stained with CD31 (magenta), ICAM-1 (green), and DAPI (blue). Scale bar = 50 µm. Images are representative of 5 images per mouse, 3 mice per group. B) Quantification of ICAM-1 intensity in areas of CD31 staining, analysed by 2-way ANOVA with Sidak’s multiple testing correction. 5 ROIs per mouse were quantified (small dots) and are colour-coded per mouse. The average of the 5 ROIs is represented with the large dot (colour-coded per mouse), and the data shows mean ± SEM. C-F) Aged K18-hACE c57BL/6□J mice infected with SARS-CoV-2 (10 3 PFU) and treated with an isotype control or anti-IL-1β antibody at 1h or 3 days post-infection. Lungs were harvested at day 6 post-infection. C) Lung tissues were stained with CD31 (magenta), ICAM-1 (green), and DAPI (blue). Scale bar = 100 µm. Images are representative of 5 images per mouse, 4-6 mice per group. D) Quantification of ICAM-1 intensity in areas of CD31 staining, analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. 5 ROIs per mouse were quantified (small dots) and are colour-coded per mouse. The average of the 5 ROIs is represented by the large dot (colour-coded per mouse), and the data show the mean ± SEM. E) Lung tissues were stained for fibrinogen. Scale bar = 50 µm. Images are representative of 4-6 mice per group. F) Quantification of fibrinogen staining intensity analysed by one-way ANOVA, with Sidak’s multiple comparison’s test. Asterisks indicate statistical significance: * p < 0.05, ** p < 0.01, *** p < 0.001.

Article Snippet: Sections were then incubated overnight at 4 °C with primary antibodies against mouse ICAM-1 (BioXCell; BE0020-1) and mouse CD31 (New England Biolabs; 77699S).

Techniques: Immunohistochemistry, Infection, Staining, Control

Journal: bioRxiv

Article Title: Spatial polarization of endothelial ICAM-1 governs T-cell exclusion in melanoma

doi: 10.64898/2026.03.19.712709

Figure Lengend Snippet: (A) Flow cytometry analysis showing frequencies of CD31⁺ endothelial cells expressing ICAM-1, VCAM-1, P-selectin, or E-selectin from the periphery vs. core of large (∼900 mm³) YUMMER1.7 tumors. Statistical analysis was performed using a paired two-tailed Student’s t-test. (B) Expression of CD11a (LFA-1 α-subunit; Δ geometric MFI) on the indicated tumor-infiltrating immune populations. (C, D) Representative immunofluorescence showing ICAM-1 (green) on CD31⁺ vessels (red) in small (∼100 mm³) and large (∼900 mm³) tumors at periphery and core (C) and corresponding quantification of ICAM-1⁺/CD31⁺ area (%) (D). Scale bar in C, 100 µm. (E-H) Representative images of extravasated fibrin(ogen) (red) (E) and pericyte (desmin, red) (G) relative to CD31 (green) in small and large tumors at periphery and core and corresponding quantification of fibrin/CD31 area (%) (F) and desmin/CD31 area (%) (H), respectively. Scale bar for E and G, 100 µm. Statistical analysis was performed using paired or unpaired two-tailed Student’s t-test. Each point represents one mouse; bars show ± SEM.

Article Snippet: Mice received intraperitoneal injections of anti-mouse ICAM-1 (CD54) monoclonal antibody (BioXcell, BE0020, clone YN1/1.7.4; 2.5 mg/kg), anti-PD-1 (BioXcell, BP0146, clone RMP1-14; 5 mg/kg), or rat IgG2b isotype control (BioXcell, BE0090) every other day for a total of seven doses.

Techniques: Flow Cytometry, Expressing, Two Tailed Test, Immunofluorescence

Journal: bioRxiv

Article Title: Spatial polarization of endothelial ICAM-1 governs T-cell exclusion in melanoma

doi: 10.64898/2026.03.19.712709

Figure Lengend Snippet: (A) Schematic diagram of treatment. 2.5 × 10 5 YUMMER1.7 cells were subcutaneously inoculated in C57BL/6 mice. When tumors reached ∼70 mm³ (days 8–10), mice were treated with anti-ICAM-1 or rat IgG isotype control (2.5 mg/kg, i.p.) every other day for seven doses; tumors were collected on days 21–24. (B-C) Tumor growth curves showing group means ± SEM (B) and individual trajectories (C) for IgG (black, n = 10) and anti-ICAM-1 (blue, n = 10) pooled from three independent experiments. P -value at the final time point by two-way ANOVA. (D) Representative immunofluorescence showing CD8⁺ T-cell distribution (green) across periphery to core in IgG- and anti-ICAM-1–treated tumors. Solid line marks the tumor boundary, and dashed line indicates the periphery–core boundary. Scale bar, 200 µm. (E) Quantification of CD8⁺ T-cell counts in periphery vs core (IgG, n = 6; anti-ICAM-1, n = 6). Statistical analysis was performed using unpaired or paired student two-tailed Student’s t-test. ( F-J ) Flow cytometric analysis of whole tumors comparing the frequency of CD8 + T cells among live cells (F), Granzyme B + cells among CD8 + T cells (G), and CD45 + immune cells among live cells (H), CD4 + T cells (I) among live cells, and regulatory T cells (Treg; CD4 + Foxp3 + ) among CD4 + T cells (J) between IgG (black, n = 14) and anti-ICAM-1 (blue, n =14). Statistical analysis was performed unpaired two-tailed Student’s t-test. Each point represents one mouse; bars show mean ± SEM.

Article Snippet: Mice received intraperitoneal injections of anti-mouse ICAM-1 (CD54) monoclonal antibody (BioXcell, BE0020, clone YN1/1.7.4; 2.5 mg/kg), anti-PD-1 (BioXcell, BP0146, clone RMP1-14; 5 mg/kg), or rat IgG2b isotype control (BioXcell, BE0090) every other day for a total of seven doses.

Techniques: Control, Immunofluorescence, Two Tailed Test

Journal: bioRxiv

Article Title: Spatial polarization of endothelial ICAM-1 governs T-cell exclusion in melanoma

doi: 10.64898/2026.03.19.712709

Figure Lengend Snippet: (A) Schematic diagram of treatment. 5 × 10 5 YUMM1.7 cells were subcutaneously inoculated in C57BL/6 mice. When tumors reached ∼70mm³ (days 9–12), mice were intraperitoneally treated with rat IgG isotype control, anti-ICAM-1 (2.5 mg/kg), anti-PD-1 (5mg/kg), or combination (anti-ICAM-1 [2.5mg/kg] and anti-PD-1 [5mg/kg]) every other day for seven doses; tumors were collected on days 22-25. (B) Tumor growth curve showing group means ± SEM for IgG (black, n = 10), anti-ICAM-1 (blue, n = 10), anti-PD-1 (red, n = 10), and combined treatment (anti-ICAM-1 plus anti-PD-1; green, n = 10). P -value at the final time point by two-way ANOVA. (C-F) Flow cytometry analysis of whole tumors comparing frequency of CD8 + T cells among live cells (C) and proportions of activated CD8 + T cells with granzyme B + (D), CD62 - CD44 + (effector memory cells) (E), and CD69 + (F) among CD8 + T-cells (IgG, n = 14; anti-ICAM-1, n = 14; anti-PD-1, n =14; anti-ICAM-1 plus anti-PD-1, n = 14). For C-F, each point represents one mouse; bars show ± SEM. Statistical analysis was performed using one-way ANOVA.

Article Snippet: Mice received intraperitoneal injections of anti-mouse ICAM-1 (CD54) monoclonal antibody (BioXcell, BE0020, clone YN1/1.7.4; 2.5 mg/kg), anti-PD-1 (BioXcell, BP0146, clone RMP1-14; 5 mg/kg), or rat IgG2b isotype control (BioXcell, BE0090) every other day for a total of seven doses.

Techniques: Control, Flow Cytometry