Ruscogenin alleviates LPS-triggered pulmonary endothelial barrier dysfunction through targeting NMMHC IIA to modulate TLR4 signaling

Abstract

Pulmonary endothelial barrier dysfunction is a hallmark of clinical pulmonary edema and contributes to the development of acute lung injury (ALI). Here we reported that ruscogenin (RUS), an effective steroidal sapogenin of Radix Ophiopogon japonicus, attenuated lipopolysaccharides (LPS)-induced pulmonary endothelial barrier disruption through mediating non-muscle myosin heavy chain IIA (NMMHC IIA)‒Toll-like receptor 4 (TLR4) interactions. By in vivo and in vitro experiments, we observed that RUS administration significantly ameliorated LPS-triggered pulmonary endothelial barrier dysfunction and ALI. Moreover, we identified that RUS directly targeted NMMHC IIA on its N-terminal and head domain by serial affinity chromatography, molecular docking, biolayer interferometry, and microscale thermophoresis analyses. Downregulation of endothelial NMMHC IIA expression in vivo and in vitro abolished the protective effect of RUS. It was also observed that NMMHC IIA was dissociated from TLR4 and then activating TLR4 downstream Src/vascular endothelial cadherin (VE-cadherin) signaling in pulmonary vascular endothelial cells after LPS treatment, which could be restored by RUS. Collectively, these findings provide pharmacological evidence showing that RUS attenuates LPS-induced pulmonary endothelial barrier dysfunction by inhibiting TLR4/Src/VE-cadherin pathway through targeting NMMHC IIA and mediating NMMHC IIA‒TLR4 interactions.

Keywords: Acute lung injury; Endothelial barrier; Interaction; Lipopolysaccharide; Non-muscle myosin heavy chain IIA; Ruscogenin; TLR4; VE-cadherin.

Graphical abstract

Figure 1

RUS attenuates LPS-induced pulmonary endothelial barrier dysfunction. C57BL/6 mice were pre-treated with RUS (0.1, 0.3, and 1 mg/kg) for 1 h followed by LPS intratracheal instillation for 6 h. (A, B) Paraffin embedded lung tissue sections were stained with hematoxylin and eosin (H&E) and lung injury was analyzed. Scale bar = 50 μm. (C, D) After exposed to LPS for 6 h, mice were sacrificed and lungs were lavaged with PBS. The total protein content and white blood cells in bronchoalveolar lavage fluid (BALF) were analyzed. (E, F) At 4 h after LPS administration, Evans blue albumin (EBA) was given by tail intravenous injection. After 2 h, the dye was extracted from lung tissues and quantified. n = 6. (G) RUS (1 μmol/L) was administrated 1 h prior to LPS exposure in murine lung vascular endothelial cells (MLECs). After LPS treatment for 6 h, immunofluorescence of VE-cadherin and phalloidine was employed to detect the effect of RUS on LPS induced MLECs barrier disruption. The asterisks mean disruptive adherens junctions. Scale bar = 20 μm. (H, I) MLECs seeded on the gelatin-coated transwell inserts for 7 days to form a monolayer, RUS (1 μmol/L) was administrated 1 h prior to LPS challenge for 6 h, the effect of RUS on LPS-induced MLECs hyperpermeability was determined by transendothelial electrical resistance (TER) and EBA assays. n = 3. Data are presented as mean ± SD. ^###P < 0.001 vs. control group; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 vs. model group. RUS, ruscogenin; LPS, lipopolysaccharides; DEX, dexamethasone; WBC, white blood cell.

Figure 2

Non-muscle myosin heavy chain IIA (NMMHC IIA) is identified as a target protein of RUS. (A) Serial affinity chromatography was used to capture the specific binding proteins of RUS and NMMHC IIA was identified as a RUS targeting protein. A comparison of proteins bound to the first resins (A1, B1, and C1) and second resins (A2, B2, and C2) showed that only the protein marked by arrow decreased significantly in amount, indicating that it was a specific binding protein. M is for the protein marker. (B) Molecular docking analysis revealed that RUS binds to the head domain of NMMHC IIA. (C) Residues that form the binding pocket of RUS are indicated. (D) Microscale thermophoresis (MST) assay was performed to detect the binding kinetics of RUS and blebbistatin (BLE) on full length NMMHC IIA.

Figure 3

RUS directly binds to N-terminal and head domain of NMMHC IIA. (A‒D) NMMHC ⅡA is divided into four major domains: N-terminal (aa. 29‒69), head domain (aa. 83‒764), IQ motif (aa. 775‒835), and tail domain (aa. 842‒1921) according to the analysis of Myh9 cDNA sequence. HEK293T cells were transfected with expression plasmids for GFP-tagged NMMHC ⅡA domains and the lysate was collected as assay buffer for MST assays. A and B showed that RUS bound with N-terminal and head domain of NMMHC IIA in a dose-dependent manner, C and D showed that there was no binding between RUS and IQ motif and tail domain. (E) Biolayer interferometry (BLI) analysis showed that there was a directly interaction between RUS and NMMHC IIA head domain.

Figure 4

LPS stimulation induces an elevated expression of NMMHC IIA and downregulation of VE-cadherin in pulmonary endothelium. (A‒D) NMMHC IIA, TLR4, p-Src (Tyr416), and VE-cadherin expression in pulmonary endothelium after LPS stimulation was assessed based on immunofluorescence with anti-CD31 (red), anti-NMMHC IIA (green), anti-TLR4 (green), anti-p-Src (green), and anti-VE-cadherin (green), and visualized using confocal microscopy. Nuclei were counterstained with DAPI (blue). Scale bar = 10 μm. (E‒H) Fluorescence intensity of NMMHC IIA, TLR4, p-Src, and VE-cadherin in vessels was quantified by “Colocalization Analysis” using Image J. All data are expressed as mean ± SD, n = 3. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 vs. 0 h group.

Figure 5

Endothelium-specific NMMHC IIA monoallelic deletion attenuates LPS-induced lung vascular hyperpermeability. Myh9^fl/wt; Tek‒Cre and Myh9^+/+; Tek‒Cre mice were challenged with LPS (5 mg/kg) via intratracheal instillation for 6 h. (A) Endothelial-specific monoallelic deletion of NMMHC IIA was analyzed by immunofluorescent co-staining of NMMHC IIA (green) and the endothelial cell marker CD31 (red), nuclei were stained with DAPI (blue). Arrows indicate the co-localization of NMMHC IIA and CD31 (n = 3). Scale bar = 20 μm. (B) NMMHC IIA expression levels in MLECs isolated from lung tissues were detected by Western blot (n = 3). (C) Paraffin-embedded lung tissue sections were stained with H&E and lung injury was analyzed (n = 5‒8). Scale bar = 50 μm. (D) Lung myeloperoxidase (MPO) levels were determined using specific kits after LPS treatment (n = 5‒8). (E) At 4 h after LPS administration, mice were i.v. injected with EBA. After 2 h, the dye was extracted from lung tissues and quantified (n = 6). (F) Immunofluorescence for extravascular albumin (green) in lung frozen sections, nuclei were stained with DAPI (blue) (n = 4). Scale bar = 50 μm. (G, H) Western blot analyses of VE-cadherin and p120-catenin expression in lung tissues (n = 3). All data are expressed as mean ± SD. ∗P < 0.05, ∗∗∗P < 0.001.

Figure 6

NMMHC IIA knockdown abolishes the protective effect of RUS on LPS-induced endothelial barrier destruction through activation of TLR4 signaling. (A, B) NMMHC IIA was knockdown via small interfering RNA in HUVECs. TER and EBA leakage assays were used to determine the effect of RUS on LPS-induced hyperpermeability. (C‒F) Western blot analyses of TLR4, p-Src (Y416), VE-cadherin, and p120-catenin expression in HUVECs. Data are expressed as mean ± SD, n = 3; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Figure 7

NMMHC IIA is dissociated from TLR4 after LPS treatment. (A) Visualization of interactions between NMMHC IIA and TLR4 in pulmonary endothelium induced by LPS using the proximity ligation assay (PLA) (red), CD31 (green), and DAPI (blue). The interactions of NMMHC IIA and TLR4 in pulmonary endothelium decreased after LPS treatment (arrows). Scale bar = 50 μm. (B) Fluorescence intensity of PLA signals in vessels was quantified using Image J. (C) Confocal microscopy of MLECs challenged with LPS for 0, 5, 10, 20, 40, and 60 min, followed by PLA reaction for NMMHC IIA‒TLR4 interactions (red signal) and DAPI (blue) staining. Scale bars = 20 μm. (D) PLA signals were quantified using Image J. Data are expressed as mean ± SD, n = 3. ∗∗P < 0.01, ∗∗∗P < 0.001.

Figure 8

NMMHC IIA binds to TLR4 through its N-terminal and head domains. (A) Binding mode analysis of NMMHC IIA and BB loop of TLR4 TIR domain via molecular docking (blue represents NMMHC IIA and red represents the binding probability of BB loop and NMMHC IIA). (B) Homology modeling of the complex of TLR4 TIR domain and NMMHC IIA. (C) Co-IP analysis was used to determine interactions between different domains of NMMHC IIA and TLR4. (D) BLI analysis was employed to detect the direct interaction between NMMHC IIA head domain and TLR4. (E, F) HEK293T cells were co-transfected with plasmids encoding Myc-tagged TLR4 and GFP-tagged Myh9 N-terminal or head domain mutants. Cell lysates were precipitated with anti-GFP antibody and immunoprecipitates detected via Western blot using anti-Myc and anti-GFP antibodies.

Figure 9

RUS suppresses LPS-induced NMMHC IIA dissociation from TLR4 and activation of TLR4/Src/VE-cadherin pathway. (A) Visualization of interactions between NMMHC IIA and TLR4 in pulmonary endothelium detected by PLA (red), CD31 (green) and DAPI (blue). Scale bar = 10 μm. (B) Immunofluorescence of NMMHC IIA (green) and TLR4 (red) in MLECs. Scale bar = 20 μm. (C) Visualization of interactions between NMMHC IIA and TLR4 in MLECs induced by LPS using the PLA (red) and nucleus was stained with DAPI (blue). Scale bar = 20 μm. (D) Fluorescence intensity of PLA signals in vessels in (A) was quantified using Image J. (E) PLA signals in MLECs in (C) were quantified using Image J. (F) NMMHC IIA was immunoprecipitated from whole cell lysates and immunoblotted for TLR4. (G‒J) Western blot analyses of NMMHC IIA, TLR4, p-Src (Y416), and VE-cadherin expression in MLECs. All data are expressed as mean ± SD, n = 3; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Figure 10

The graphic illustration of the mechanism of RUS ameliorating LPS-induced pulmonary endothelial barrier dysfunction. NMMHC IIA binds to TLR4 through its N-terminal and head domains as an inactive complex under physiological conditions. Upon LPS binding to TLR4, NMMHC IIA is dissociated from TLR4, initiating downstream signaling and leading to endothelial barrier dysfunction by disruption of VE-cadherin junctions. RUS effectively prevents LPS-induced pulmonary endothelial barrier disruption by targeting NMMHC IIA and modulating NMMHC IIA‒TLR4 interactions. PMECs, pulmonary vascular endothelial cells; p120, p120-catenin.