Critical role of mitogen-inducible gene 6 in restraining endothelial cell permeability to maintain vascular homeostasis

Abstract

Although mitogen-inducible gene 6 (MIG6) is highly expressed in vascular endothelial cells, it remains unknown whether MIG6 affects vascular permeability. Here, we show for the first time a critical role of MIG6 in limiting vascular permeability. We unveil that genetic deletion of Mig6 in mice markedly increased VEGFA-induced vascular permeability, and MIG6 knockdown impaired endothelial barrier function. Mechanistically, we reveal that MIG6 inhibits VEGFR2 phosphorylation by binding to the VEGFR2 kinase domain 2, and MIG6 knockdown increases the downstream signaling of VEGFR2 by enhancing phosphorylation of PLCγ1 and eNOS. Moreover, MIG6 knockdown disrupted the balance between RAC1 and RHOA GTPase activation, leading to endothelial cell barrier breakdown and the elevation of vascular permeability. Our findings demonstrate an essential role of MIG6 in maintaining endothelial cell barrier integrity and point to potential therapeutic implications of MIG6 in the treatment of diseases involving vascular permeability. Xing et al. (2022) investigated the critical role of MIG6 in vascular permeability. MIG6 deficiency promotes VEGFA-induced vascular permeability via activation of PLCγ1-Ca²⁺-eNOS signaling and perturbation of the balance in RAC1/RHOA activation, resulting in endothelial barrier disruption.

Keywords: Endothelial cell barrier; MIG6; VEGFR2; Vascular permeability.

Fig. 1

Vascular permeability is increased in Mig6 knockout mice and endothelial barrier is compromised by MIG6 knockdown. (a) Representative images of WT (top) and Mig6^−/− (bottom) mouse showing Evans blue leakage, after treatment with 50 ng/ml of VEGFA (left) or BSA (right) for 1 h. (b) Images of Evans blue leakage in excised ears of WT and Mig6^−/− mouse. (c) Quantification of Evans blue leakage in ear tissues. Evans blue dye was extracted from ear skin of WT and Mig6^−/− mice to read an absorbance (n = 3–5). (d) Histamine (100 μM) or VEGFA (50 ng/ml)-induced EC permeability in siControl or siMIG6-treated HUVECs (n = 4). (e) Normalized resistance of siControl or siMIG6-treated ECs after stimulation with VEGFA. HUVECs were treated with siControl or siMIG6, and grown to confluence on gelatin-coated electrode arrays. After low serum starvation for 3 h, EC monolayer was stimulated with VEGFA (50 ng/ml) and TEER was measured by ECIS at a frequency of 4000 Hz. One-way ANOVA with Sidak multiple comparison test (c, d) and unpaired Student’s t-test (e) were performed. Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, NS: not statistically significant

Fig. 2

MIG6 expression is induced by VEGFA and suppressed by VEGFR2 inhibition in ECs. (a) VEGFA-induced MIG6 expression. Serum-starved HUVECs were stimulated with a different amount of VEGFA for 30 min. VEGFR2 activation was monitored by phosphorylation of Tyr1175. (b) MIG6 induction was analyzed by densitometry and normalized by tubulin. (c) Time kinetic assay of MIG6 induction in HUVECs. Serum-starved ECs were stimulated with VEGFA (20 ng/ml) for different time points. (d) MIG6 induction at different time points was measured by densitometry and normalized by tubulin. (e) HUVECs were treated with a VEGFR2 tyrosine kinase inhibitor Axitinib (10 nM) for 3 h prior to VEGFA treatment (20 ng/ml) for 30 min. The diminished VEGFA-induced MIG6 expression is shown by western blot. (f) MIG6 induction was quantified by densitometry and normalized by tubulin. Fold induction relative to the control is shown as the mean ± SEM (n = 3 for b, d, and f). Statistical significance was determined by one-way ANOVA with Sidak multiple comparison test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, NS: not statistically significant

Fig. 3

MIG6 binds to VEGFR2. (a) Association of MIG6 with VEGFR2 was assessed by GST pull-down assay, followed by western blot. (b) Schematic representation of VEGFR2 protein domains and the truncated proteins of GST-fusion VEGFR2 domains. Y951, Y1054, Y1059, Y1175, and Y1214 are major tyrosine phosphorylation sites located in the indicated domains of VEGFR2. TM, transmembrane; JX, juxtamembrane domain; KD1, kinase domain 1; KI, kinase insert domain; KD2, kinase domain 2; Cyto, whole cytoplasmic domain of VEGFR2; C-ter, C-terminus domain of VEGFR2 excluding JX, KD1, KI, and KD2. (c) Association of Flag-tagged MIG6 with truncated VEGFR2 proteins was assessed by GST pull-down assay, displaying that MIG6 binds to the cytoplasmic domain and kinase domain 2 of VEGFR2. Inputs are shown on the right

Fig. 4

Tyrosine phosphorylation and internalization of VEGFR2 are upregulated in MIG6 knockdown ECs. (a, c) MIG6 knockdown in HUVECs increases VEGFR2 phosphorylation on Tyr951 and Tyr1054/9 (a), and Tyr1175 and Tyr1214 (c) in response to VEGFA (50 ng/ml) at 5 and 10 min. (b, d) Tyrosine phosphorylation of VEGFR2 shown in (a) and (c) was analyzed and normalized by total VEGFR2. Fold induction relative to the control is shown as the mean ± SEM (n = 3 for b and d). (e) Antibody feeding assay of VEGFR2 internalization (red fluorescence) in response to VEGFA (20 ng/ml) for 30 min in ECs treated with siControl and siMIG6 (n = 6–7). (f) Quantification of internalized VEGFR2 fluorescence in (e). The internalized fluorescence intensity was analyzed and normalized by the surface level of VEGFR2. Data are presented as mean ± SEM. One-way ANOVA with Sidak multiple comparison test (b, d) and unpaired Student’s t-test (f) were used for statistical analyses. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, NS: not statistically significant

Fig. 5

MIG6 knockdown activates PLCγ1-Ca²⁺-eNOS signaling and VEGFA-induced increase of permeability by MIG6 knockdown is attenuated by PLCγ1 and eNOS inhibitors. (a) Western blot analysis of PLCγ1 phosphorylation on Tyr783 in siControl and siMIG6-treated ECs in response to VEGFA (50 ng/ml) at 10 min. (b) PLCγ1 phosphorylation on Tyr783 was analyzed and normalized by total PLCγ1 (n = 4). (c) Intracellular Ca²⁺ level in serum-starved ECs treated with siControl and siMIG6 was determined in response to VEGFA (20 ng/ml) at various time points (n = 3). (d) Western blot analysis of eNOS phosphorylation on Ser1177 in ECs treated with siControl and siMIG6, after treatment with VEGFA (50 ng/ml) for 10 min and 30 min. (e) eNOS phosphorylation on Ser1177 was analyzed and normalized by total eNOS (n = 3). (f) Western blot analysis of eNOS phosphorylation on Ser1177 in ECs treated with siControl, siMIG6, siPLCγ1, and siMIG6 + siPLCγ1, followed by treatment with VEGFA (50 ng/ml) for 10 min and 30 min. (g) eNOS phosphorylation on Ser1177 was analyzed and normalized by total eNOS (n = 3). (h, i) VEGFA-induced permeability was measured in siControl or siMIG6 knockdown ECs treated with a PLCγ1 inhibitor (U73122, 3 μM) for 30 min or an eNOS inhibitor (L-NAME, 300 μM) for 1 h prior to VEGFA (50 ng/ml) treatment (n = 4 for h; n = 5 for i). Statistical significance was determined by one-way ANOVA with Sidak multiple comparison test (b, e, and g), paired Student’s t-test (c), and two-way ANOVA with Tukey multiple comparison test (h, i). Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, NS: not statistically significant

Fig. 6

The balance of RAC1/RHOA GTPase activation is disrupted in MIG6 knockdown ECs. (a, c) Western blot analysis of RAC1-GTPase (a) and RHOA-GTPase (c) activity in siControl and siMIG6 knockdown ECs treated with VEGFA (50 ng/ml) for 10 and 60 min. (b, d) RAC1-GTPase (b) and RHOA-GTPase (d) activation was analyzed and normalized by total RAC1 and RHOA, respectively. (e) The proposed working model illustrates that MIG6 deficiency promotes VEGFA-induced vascular permeability via activation of PLCγ1 and eNOS signaling and perturbation of the balance in RAC1/RHOA activation, resulting in endothelial barrier disruption. One-way ANOVA with Sidak multiple comparison test was performed (b, d). Data are presented as mean ± SEM (n = 3 for b, d). * p < 0.05, ** p < 0.01