Tension on JAM-A activates RhoA via GEF-H1 and p115 RhoGEF

Abstract

Junctional adhesion molecule A (JAM-A) is a broadly expressed adhesion molecule that regulates cell-cell contacts and facilitates leukocyte transendothelial migration. The latter occurs through interactions with the integrin LFA-1. Although we understand much about JAM-A, little is known regarding the protein's role in mechanotransduction or as a modulator of RhoA signaling. We found that tension imposed on JAM-A activates RhoA, which leads to increased cell stiffness. Activation of RhoA in this system depends on PI3K-mediated activation of GEF-H1 and p115 RhoGEF. These two GEFs are further regulated by FAK/ERK and Src family kinases, respectively. Finally, we show that phosphorylation of JAM-A at Ser-284 is required for RhoA activation in response to tension. These data demonstrate a direct role of JAM-A in mechanosignaling and control of RhoA and implicate Src family kinases in the regulation of p115 RhoGEF.

FIGURE 1:

Overview of tension models. Cells were grown on fibronectin-coated substratum, and anti-JAM-A–coated paramagnetic beads were added at approximately a 3:1 bead-to-cell ratio. For application of pulsatile forces, the pole tip of a magnetic tweezers was lowered to 25 μm above the bead and force applied using a defined regimen. Bead displacement was imaged at 30 frames/s, and bead movement was tracked using custom software as described in Materials and Methods. For application of continuous force, a magnet was suspended parallel to the apical surface of the cells for the determined time. Cells were then lysed and processed for biochemical analysis as needed.

FIGURE 2:

Tension on JAM-A activates RhoA to increase cell stiffness. RhoA activity was measured using RBD-pull-down assays on untreated HUVECs, HUVECs treated with anti-JAM-A–coated beads, HUVECs with the same beads plus 3 min of continuous force (A), or the same regimen with PLL-coated beads (B). Data are representative of at least three separate experiments and are quantified as means ± SEM in C and D. *p < 0.01 vs. untreated as determined by t test. To determine cell stiffness, HUVECs were untreated (E) or treated with 1 mg/ml C3 transferase for 60 min (F) or 10 μM Y-27632 for 30 min (G) before force application on anti-JAM-A–coated magnetic beads with magnetic tweezers. *p < 0.01 vs. pulse 1 as determined by t test.

FIGURE 3:

JAM-A activates PI3K upstream of RhoA. HUVECs were transfected with GFP or GFP-Akt-PH and incubated with anti-JAM-A– or PLL–coated magnetic beads in the presence or absence of 1 min of continuous force. Cells were fixed with paraformaldehyde, and enrichment of GFP to the area around the bead was determined. (A) Representative images. (B) Quantification. Data are mean ± SEM of >25 cells/experiment from three independent experiments. *p < 0.05 vs. control by t test. HUVECs were incubated with anti–JAM-A beads, and force was applied for 0–10 min. Akt phosphorylation, used as a marker of PI3K activation, was determined by Western blot (C). RhoA activity in response to force on JAM-A–coated beads was measured in HUVECs with or without incubation with the PI3K inhibitor LY294002 (10 μM, 30 min). (D) Representative blots. (E) Means ± SEM from four experiments. *p < 0.05 vs. control by t test.

FIGURE 4:

Tension on JAM-A activates GEF-H1 and p115 RhoGEF via PI3K. HUVECs were exposed to anti-JAM-A–coated beads, and tension was imposed with a permanent magnet for 3 min. Activation of RhoA GEFs was determined using the GST-RhoA^G17A pull-down assay. (A) Representative Western blots. (B) Means ± SEM from at least three experiments. White bars are untreated control, gray bars are bead only, and black bars are beads plus 3 min of tension. To test for a role of PI3K in activation of GEFs, some cells were treated with LY294002 (10 μM, 30 min) before addition of beads. GEF activity was assessed using GST-RhoAG17A pull-down assay as described as methods (C). (D) White bars are bead only, gray bars are beads plus 3 min of tension, black bars are LY294002 plus beads, and checkered bars are LY29004 plus beads with tension for 3 min. Statistical analysis was conducted by t test.

FIGURE 5:

JAM-A–mediated RhoA activation requires both GEF-H1 and p115 RhoGEF. HUVECs were transfected with control siRNA or oligos designed against GEF-H1or p115 Rho GEF. Transfected cells were incubated with anti-JAM-A–coated magnetic beads, and some cells were exposed to force for 3 min. RhoA activity was determined by GST-RBD pull down. Representative Western blots (A) and quantifications (B) from six separate experiments. *p < 0.05 vs. no force sample for each condition as determined by t test.

FIGURE 6:

GEF-H1 is activated downstream of FAK/ERK in response to tension on JAM-A. (A) HUVECs were incubated with anti-JAM-A–coated magnetic beads, and tension was applied for 0–5 min. Cells were lysed and analyzed for ERK and FAK phosphorylation by Western blot analysis. (B) HUVECs were incubated with anti-JAM-A–coated magnetic beads, and some cells were pretreated with the MEK inhibitor U0126 (25 μM) or FAK inhibitor 14 (2 μM) for 30 min before addition of beads, with some samples experiencing 3 min of force. RhoA GEF activity was assessed by GST-RhoA^G17A pull-down assay. (C, D) Activation of GEF-H1 and p115 RhoGEF, respectively. Data are mean ± SEM from at least three experiments. *p < 0.05 vs. no-force control for each condition by t test.

FIGURE 7:

Src family kinases activate p115 RhoGEF in response to tension on JAM-A. (A) HUVECs were incubated with inhibitors against JAK2 (AG 490, 25 μM), PKCα (Gö6976, 10 μM), or Src family kinases (su6656, 5 μM) for 30 min, followed by addition of anti-JAM-A– coated magnetic beads. Some cells also experienced force for 3 min. p115 RhoGEF activity was determined by RhoA^G17A pull-down assay. (B) Means ± SEM of at least three separate experiments *p < 0.05 vs. no-force control for each condition by t test.

FIGURE 8:

JAM-A S284 phosphorylation is regulated by mechanical forces. (A) HUVECs were incubated with anti-JAM-A–coated magnetic beads and exposed to force for 0–10 min. (B) HUVECs were exposed to shear stress for 0–60 min. For both experiments, cells were lysed, and Western blot analysis of total and phosphorylated JAM-A was conducted. Blots are representative of at least three independent experiments

FIGURE 9:

JAM-A S284 phosphorylation is required for RhoA activation and cell stiffening in response to force on JAM-A. (A) CHO cells were transfected with empty vector, JAM-A, or JAM-A S284A, and expression of JAM-A and phosphorylated JAM-A was determined by Western blot. (B) Barrier function of CHO cells expressing empty vector (EV), JAM-A, or JAM-A S284 was determined by FITC-dextran flux. Data are representative of three experiments run in triplicate. (C) CHO cells expressing EV, JAM-A, or JAM-A S284A were incubated with anti-JAM-A–coated magnetic beads, and some cells were exposed to force for 5 min. RhoA activity was determined by the RBD pull-down assay. (D) Means ± SEM of four independent experiments, with *p < 0.05 vs. no-force control by t test. Cell stiffness was determined in cells expressing (E) JAM-A, (F) JAM-A S284A, (G) JAM-A and incubated with 10 μM Y-27632, or (H) expressing JAM-A and incubated with 1 mg/ml C3-transferase as described in Materials and Methods.

FIGURE 10:

Force on JAM-A regulates PKCζ to activate RhoA. HUVECs were incubated with anti–JAM-A beads in the presence or absence of forces, and PKCζ phosphorylation was determined by Western blot (A). HUVECs were incubated with anti-JAM-A–coated magnetic beads, and some cells were exposed to force for 3 min. Some cells had been pretreated with PKCζ inhibitory peptide. RhoA activity was determined by the RBD pull-down assay (B). Cell stiffness was determined in HUVECs after tension was imposed on anti-JAM-A–coated beads in the presence of PKCζ inhibitor as described in Materials and Methods (C).

FIGURE 11:

Working model of RhoA activation in response to force on JAM-A. Anti-JAM-A–coated magnetic beads engage JAM-A on the cell surface. In response to force, SFKs dissociate from the protein’s C-terminus. In short order, PI3K likely signals for the activation of PKCζ, as well as of GEF-H1 and p115 RhoGEF. Activation of GEF-H1 and p115 RhoGEF also requires FAK/ERK- and SFK-mediated pathways, respectively. Finally, RhoA is activated by either GEF to regulate actomyosin-based cellular contractility and cell stiffness.