Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 4, 60
eCollection

Deubiquitinase Function of A20 Maintains and Repairs Endothelial Barrier After Lung Vascular Injury

Affiliations

Deubiquitinase Function of A20 Maintains and Repairs Endothelial Barrier After Lung Vascular Injury

Dheeraj Soni et al. Cell Death Discov.

Abstract

Vascular endothelial cadherin (VE-cad) expression at endothelial adherens junctions (AJs) regulates vascular homeostasis. Here we show that endothelial A20 is required for VE-cad expression at AJs to maintain and repair the injured endothelial barrier. In endothelial cell (EC)-restricted Tnfaip3 (A20) knockout (A20∆EC ) mice, LPS challenge caused uncontrolled lung vascular leak and persistent sequestration of polymorphonuclear neutrophil (PMNs). Importantly, A20∆EC mice exhibited drastically reduced VE-cad expression in lungs compared with wild-type counterparts. Endothelial expression of wild-type A20 but not the deubiquitinase-inactive A20 mutant (A20C103A) prevented VE-cad ubiquitination, restored VE-cad expression, and suppressed lung vascular leak in A20∆EC mice. Interestingly, IRAK-M-mediated nuclear factor-κB (NF-κB) signaling downstream of TLR4 was required for A20 expression in ECs. interleukin-1 receptor-associated kinase M (IRAK-M) knockdown suppressed basal and LPS-induced A20 expression in ECs. Further, in vivo silencing of IRAK-M in mouse lung vascular ECs through the CRISPR-Cas9 system prevented expression of A20 and VE-cad while augmenting lung vascular leak. These results suggest that targeting of endothelial A20 is a potential therapeutic strategy to restore endothelial barrier integrity in the setting of acute lung injury.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Fig. 1
Fig. 1. A20 deficiency augments LPS-induced signaling in endothelial cells.
a A20 expression was determined by immunoblot (IB) analysis in lung endothelial cells (LECs) and bone marrow-derived macrophages (BMDMs) from A20fl/fl and A20∆EC mice. **p < 0.001, four separate cell preparations. b Immunostaining of lung sections from A20fl/fl and A20∆EC mice with antibody specific to A20 and endothelial marker vWF. c LECs from A20fl/fl and A20∆EC mice treated with LPS (1 µg/ml) were used for immunoblot to determine A20 protein expression. **p < 0.001, n = 4 experiments. d LECs from A20fl/fl and A20∆EC mice treated with LPS (1 µg/ml) for different time intervals were used for immunoblot analysis to determine phosphorylation of TAK1, IKKβ, and p38. Representative data from three experiments are shown in a-d
Fig. 2
Fig. 2. A20∆EC mice exhibit heightened sensitivity to LPS-induced septic shock.
a Survival of age- and weight-matched A20fl/fl and A20∆EC mice after administration of indicated LPS doses (mg/kg, i.p.). (n = 10 per genotype per LPS dose). b Hematoxylin-and-eosin staining of lung sections from A20fl/fl and A20∆EC mice challenged with LPS (2.5 mg/kg; i.p.) for different time points. Scale bars, 100 μm. Brbronchi, V vessel. c Myeloperoxidase (MPO) activity in lung tissue from A20fl/fl and A20∆EC mice challenged with LPS (2.5 mg/kg, i.p) for 0, 6, 12, and 24 h. n = 5 per genotype per time point. *p < 0.01; **p < 0.001; ***p < 0.0001; A20fl/fl versus A20∆EC mice. d Concentrations of cytokines TNF-α and MCP-1 in lungs of A20fl/fl and A20∆EC mice challenged with LPS (5 mg/kg, i.p) for 0 and 6 h. n = 5 per genotype. *p < 0.01, A20fl/fl versus A20∆EC mice. e TUNEL staining of lung tissue sections from A20fl/fl and A20∆EC mice shows apoptotic cells in A20∆EC mice. Representative data from three experiments are shown. **p < 0.001, A20fl/fl versus A20∆EC mice. f Pulmonary microvessel liquid filtration coefficient in lungs of A20fl/fl and A20∆EC mice after LPS challenge (5 mg/kg, i.p.). n = 5 per genotype per time point. *p < 0.01; **p < 0.001; A20fl/fl versus A20∆EC mice. g A20fl/fl and A20∆EC mice challenged with LPS (5 mg/kg, i.p.) for 0 and 24 h were used to assess in vivo lung vascular leak by measuring Evans blue dye-conjugated with albumin (EBA) uptake. Left, representative photographs of the lungs are shown; right, quantified results are shown; n = 4 per genotype per time point. **p < 0.001; A20fl/fl versus A20∆EC mice
Fig. 3
Fig. 3. A20 deficiency in endothelial cells decreases VE-cad expression at AJs to impair endothelial barrier integrity.
a Immunoblot analysis of VE-cad in lungs of A20fl/fl and A20∆EC mice. n = 4, in each group; **p < 0.001. b Quantitative RT-PCR analysis of VE-cad mRNA in lungs of A20fl/fl and A20∆EC mice. n = 4, in each group. c LECs from A20fl/fl and A20∆EC mice grown on coverslips were stained with VE-cad pAb (green) and analyzed by confocal microscopy. d HLMVECs were transfected with either scrambled-siRNA (Sc-siRNA) or A20- siRNA (100 nM). At 72 h after transfection, cells were challenged with LPS (1 µg/ml) for indicated times. The cell lysates were then used to measure LPS-induced A20 and VE-cad expression by IB. e HLMVECs transfected with either Sc-siRNA or A20- siRNA (100 nM) were immunostained with VE-cad pAb (green), and analyzed as in b. f HLMVECs were transfected as in d after pretreatment with proteasomal inhibitor MG132 (10 µM) for 2 h, challenged with LPS (1 µg/ml) for indicated time intervals were then used to determine VE-cad phosphorylation using antibody specific to phospho-Y685 and phospho-Y731. c-f Representative data from three experiments are shown. g HLMVECs transfected with Sc-siRNA or A20- siRNA as in e. At 24 h after transfection, cells plated on gold electrodes were challenged with LPS (1 µg/ml) in the presence of medium containing 10% FBS, and endothelial AJ integrity was monitored by TER (n = 4 per group). *p < 0.01, compared with respective controls
Fig. 4
Fig. 4. DUB activity of endothelial A20 restores endothelial AJs integrity after LPS-induced endothelial injury.
a HMECs were co-transfected with an HA-tagged ubiquitin-expressing plasmid and with either A20WT or A20C103A plasmid. At 24 h after transfection, cells were infected with recombinant adenovirus-expressing VE-cad (Adeno-VE-cad). At 72 h, cell lysates were used for IB to determine A20 expression. Prior to lysis, cells were pretreated with MG132 (10 µM) for 2 h. b, As in a, lysates were immunoprecipitated (IP-ed) using VE-cad antibody. The IP-ed samples were used for IB analysis to determine ubiquitination of VE-cad. Top panel, blotted with pan-ubiquitin-specific antibody; Second panel, blotted with ubiquitin K63-linkage-specific antibody; Third panel, blotted with ubiquitin K48-linkage-specific antibody; Bottom panel, blotted with VE-cad-specific antibody. Representative data from three experiments are shown. c A20∆EC mice were injected i.v. with liposome-A20 plasmid complexes containing either wild-type A20 (A20WT) or DUB-inactive A20-C103A mutant (A20C103A). At 96 h after injection, lungs harvested were used to determine A20 and VE-cad expression by IB. n = 4 mice each group; **p < 0.001, A20 expression-A20∆EC vs A20WT or A20C103A; **p < 0.001, VE-cad expression: A20fl/fl vs A20∆EC, A20∆EC injected with A20WT vs A20∆EC, A20∆EC injected with A20WT vs A20∆EC injected with A20C103A. d As in c A20fl/fl, A20∆EC, A20∆EC injected with A20WT construct, and A20∆EC injected with A20C103A construct were challenged with LPS (2.5 mg/kg) for 24 h and then in vivo lung vascular leak was determined. n = 4 mice per group; ***p < 0.0001; A20∆EC mice versus A20∆EC mice injected with A20WT construct; A20∆EC mice injected with A20WT construct versus A20∆EC mice injected with A20C103A construct
Fig. 5
Fig. 5. IRAK-M–IKKγ–NF-κB axis activation downstream of TLR4 induces A20 expression in ECs and signals VE-cad expression at AJs.
a Lung endothelial cells (mLECS) from wild-type (C57BL/6) mice treated with LPS (1 µg/ml) for different time intervals were used for immunoblot to determine IRAK-M and A20 expression. b Control mLECs, mLECs transfected with Sc-siRNA or mIRAK-M-siRNA (100 nM) for 72 h were treated with LPS (1 µg/ml) and used for IB to determine IRAK-M expression. c Control mLECs, mLECs transfected with Sc-siRNA or mIRAK-M-siRNA (100 nM) for 72 h were treated with LPS (1 µg/ml) and the cell lysates were used for IB to determine phosphorylation of IKKγ (Ser376) and IKKβ (S177/181). Blots were re-probed with antibodies specific to either IKKγ or IKKβ. d mLECs treated with either Sc-siRNA or IRAK-M siRNA as in b were challenged with LPS (1 µg/ml) for different time periods and used for IB to determine VE-cad and A20 expression. e LECs from TAK1 (Map3k7)-floxed (TAK1fl/fl) mice infected with (25 pfu/cell) control adenovirus (vector) or adenovirus expressing Cre-recombinase (Adeno-Cre) for 48 h were treated with or without LPS (1 µg/ml) challenge and used for IB to determine TAK1 expression (left panel) or ICAM-1 and A20 expression (right panel). a-e representative data from three experiments are shown. f HLMVECs transfected with Sc-siRNA or IRAK-M-siRNA as b. At 24 h after transfection, cells plated on gold electrodes were challenged with 1 µg/ml LPS in the presence of 10% FBS containing medium, and endothelial AJs integrity was monitored by TER (n = 4/group). *p < 0.01, compared with respective controls
Fig. 6
Fig. 6. CRISPR/Cas9 system-mediated silencing of IRAK-M in mouse lung vascular ECs impaired the repair of endothelial barrier after LPS-induced lung injury.
a WT (C57BL/6) mice were injected i.v. with liposome-CRISPR/Cas9 plasmid constructs containing two different guide RNA (gRNA1, gRNA2) against mouse IRAK-M gene. At 96 h after injection, lungs harvested were used for western analysis. n = 4 mice each group; *p < 0.01, vector vs gRNA1. b, Mice were i.v. injected with liposome-CRISPR/Cas9-IRAK-M (gRNA1) plasmid complex. At 96 h after injection, mice were challenged with LPS (5 mg/kg) for 24 h and then lungs harvested were used for immunoblot to determine expression of IRAK-M, A20, and VE-cad. n = 4 mice each group; *p < 0.01, vector vs gRNA1. c Mice injected with liposome-CRISPR/Cas9-IRAK-M (gRNA1) plasmid complex as in b were used to assess in vivo lung vascular leak. n = 4 mice per group. *p < 0.01; Vector injected mice versus IRAK-M-gRNA1 injected mice
Fig. 7
Fig. 7. Model for TLR4-induced endothelial AJs disassembly, A20 expression in ECs, and A20-mediated repair of endothelial barrier.
ROS-mediated Src activation downstream of TLR4 induces VE-cad phosphorylation to disassemble AJs. Subsequently, internalized phosphorylated VE-cad ubiquitinated and degraded via proteasomal pathway. A20 induced by IRAK-M–IKKγ–NF-κB axis downstream of TLR4 in ECs prevents ubiquitination of VE-cad to replenish VE-cad at AJs to restore the endothelial barrier integrity. ROS reactive oxygen species

Similar articles

See all similar articles

References

    1. Dejana E. Endothelial cell-cell junctions: happy together. Nat. Rev. Mol. Cell Biol. 2004;5:261–270. doi: 10.1038/nrm1357. - DOI - PubMed
    1. Dejana E, Orsenigo F, Lampugnani MG. The role of adherens junctions and VE-cadherin in the control of vascular permeability. J. Cell Sci. 2008;121:2115–2122. doi: 10.1242/jcs.017897. - DOI - PubMed
    1. Breviario F, et al. Functional properties of human vascular endothelial cadherin (7B4/cadherin-5), an endothelium-specific cadherin. Arterioscler. Thromb. Vasc. Biol. 1995;15:1229–1239. doi: 10.1161/01.ATV.15.8.1229. - DOI - PubMed
    1. Corada M, et al. Vascular endothelial-cadherin is an important determinant of microvascular integrity in vivo. Proc. Natl. Acad. Sci. USA. 1999;96:9815–9820. doi: 10.1073/pnas.96.17.9815. - DOI - PMC - PubMed
    1. Gotsch U, et al. VE-cadherin antibody accelerates neutrophil recruitment in vivo. J. Cell Sci. 1997;110(Pt 5):583–588. - PubMed

LinkOut - more resources

Feedback