Identification of a tight junction-associated guanine nucleotide exchange factor that activates Rho and regulates paracellular permeability

Abstract

Rho family GTPases are important regulators of epithelial tight junctions (TJs); however, little is known about how the GTPases themselves are controlled during TJ assembly and function. We have identified and cloned a canine guanine nucleotide exchange factor (GEF) of the Dbl family of proto-oncogenes that activates Rho and associates with TJs. Based on sequence similarity searches and immunological and functional data, this protein is the canine homologue of human GEF-H1 and mouse Lfc, two previously identified Rho-specific exchange factors known to associate with microtubules in nonpolarized cells. In agreement with these observations, immunofluorescence of proliferating MDCK cells revealed that the endogenous canine GEF-H1/Lfc associates with mitotic spindles. Functional analysis based on overexpression and RNA interference in polarized MDCK cells revealed that this exchange factor for Rho regulates paracellular permeability of small hydrophilic tracers. Although overexpression resulted in increased size-selective paracellular permeability, such cell lines exhibited a normal overall morphology and formed fully assembled TJs as determined by measuring transepithelial resistance and by immunofluorescence and freeze-fracture analysis. These data indicate that GEF-H1/Lfc is a component of TJs and functions in the regulation of epithelial permeability.

Figures

Figure 1.

Subcellular localization of the B4/7 antigen. (A and B) MDCK cells cultured on polycarbonate filters for 5 d were fixed with methanol and stained with mAb B4/7 (A1 and B1) and polyclonal anti–ZO-1 (A2 and B2) antibodies. Shown are confocal xy (A) and xz (B) sections. A3 and B3 show overlays of the two labelings (B4/7, FITC; anti-ZO-1, Texas red). Bars, 5 μm. (C and D) Ultrathin cryosections of MDCK cells were labeled with B4/7 followed by a rabbit anti–mouse and 5 nm protein A conjugated to colloidal gold (C, apical region of a cell; D, larger magnification of a TJ). The arrows indicate the position of TJs. Bars: (C) 150 nm; (D) 100 nm.

Figure 2.

Immunostaining of the B4/7 antigen in mouse blastocysts. Embryos were collected and cultured for up to 2 d to the blastocyst stage before removal of the zona pellucida, fixation in 1% formaldehyde, and double staining with mAb B4/7 (A1, B1, and C1) and α-catenin (A2) or ZO-1 (B2 and C2). A3, B, 3 and C3 are overlays of the two labelings. In A and C, midplane optical sections through the trophectoderm wall of the spherical blastocyst are shown, whereas in B, a tangential plane is shown. Arrows in A indicate distinct sites of B4/7 and α-catenin. Bar, 20 μm.

Figure 3.

Molecular identification of the B4/7 antigen. (A) Total MDCK cell extracts or an immunoprecipitate generated with mAb B4/7 were separated on a 5–15% SDS-PAGE gel and then immunoblotted with B4/7. The asterisk marks the IgG heavy chain. (B) Domain structure of cGEF-H1. Domains identified by PFAM (sequence data available from GenBank/EMBL/DDBJ under accession no. AF494096/AF494097) are shown. (C) B4/7 immunoprecipitates were immunoblotted with antipeptide antibodies generated against both cGEF-H1 termini, the alternative domain, or a control antibody.

Figure 4.

cGEF-H1 associates with intercellular junctions and microtubules in MDCK cells. (A) MDCK cells were transiently transfected with HA-tagged cGEF-H1. After 3 d the cells were permeabilized, fixed, and labeled with anti-HA antibodies. Bar, 5 μm. (B and C) Low confluent MDCK cells transiently transfected with HA-cGEF-H1 were fixed with methanol and labeled with anti-HA and anti–α-tubulin antibodies. B shows the anti-HA staining of a strongly expressing cell, and C is a larger magnification of a peripheral region of the cell shown in B (C1, anti-HA; C2, anti–α-tubulin). Bars: (B) 5 μm; (C) 2 μm. (D) Transiently transfected cells were processed as in C. Cells with a moderate expression level (D1, anti-HA; D2, anti–α-tubulin; D3, DNA) are shown. Bar, 5 μm. (E and F) High density wild-type MDCK cells were stained with a peptide antibody against the NH2 terminus of cGEF-H1 (E1) and an mAb specific for ZO-1 (E2). Bar, 3 μm. (F) Low confluent MDCK cells were stained with antibodies against the NH2 terminus of cGEF-H1(F1) and anti–α-tubulin (F2). Bar, 3 μm. (G) Low confluent MDCK cells were extracted with Triton X-100 in the presence of Taxol before fixation and then processed for immunofluorescence with mAb B4/7. G1 and G2 show two serial sections of the same cell. Note, a fraction of cGEF-H1 remained associated with junctions during mitosis. Bar, 3 μm.

Figure 5.

cGEF-H1 associates with different actin-based cytoskeletal structures in different cell types. Low confluent MRC-5 fibroblasts (A), MDCK (B), and Caco-2 (C) cells were fixed with PFA, permeabilized with Triton X-100, and fluorescently labeled with B4/7 (A1, B1, and C1) and TRITC-phalloidin (A2, B2, and C2). Bars, 6 μm.

Figure 6.

Activation of Rho by cGEF-H1. (A) Total cell extracts of wild-type (wt) or transfected MDCK cells were immunoblotted with mAb B4/7. 3A3 and 3A4, clones overexpressing cGEF-H1; control, clone derived from control transfection. (B) Wild-type and transfected MDCK cells expressing truncated cGEF-H1 (HA-GEF-T) or wild-type cGEF-H1 were extracted and tested for the presence of activated Rho or Rac1 by performing pull-down assays with recombinant fusion proteins lacking (GST) or containing a GTPase-binding domains of (GST-GBD). The pulled-down fractions were analyzed by immunoblotting for the presence of Rho or Rac1. Increasing amounts of total cell extracts were loaded on the gels to normalize the amount of GTPases in different cell extracts (1X Total ex and 2X Total ex). Immunoblots were quantified by densitometric scanning, and the values obtained for the activated GTPase were divided by those obtained in total cell extracts. All ratios were then normalized to wild-type cells and expressed as fold activation (mean ± SD of three experiments; both increases in active Rho are significant, p < 0.05 using the Student's t test).

Figure 7.

Functional analysis of TJs of cells overexpressing cGEF-H1. (A) Wild-type (wt) and cGEF-H1–overexpressing (3A4) cells, cultured for 1 wk on polycarbonate filters, were fixed and processed for double immunofluorescence using mAb B4/7 and anti–ZO-1 antibodies. Images obtained with constant microscope sensitivity settings are shown. Bar, 10 μm. (B and C) MDCK cells were cultured on filters for 1 wk and were then analyzed first by measuring TER (B) and, after a time of recovery, by determining paracellular permeability of 4 kD FITC-dextran (C) or 400 kD FITC-dextran (D). At the end, the cultures were analyzed for overexpression of cGEF-H1 by immunoblotting (examples are shown in Fig. 3 A). All values of transfected cells represent means ± SD of several independent clones: four clones were analyzed for cGEF-H1+I, three clones for cGEF-H1-I, and three control clones that were derived from a transfection of the empty vector. The increases in 4 kD FITC-dextran permeability of overexpressing cells are significant (cGEF-H1+I, p < 0.02 and cGEF-H1-I, p < 0.05 using the Student's t test) (E) Fluid phase transcytosis was measured after labeling cells for 10 min at 37°C with HRP from the apical side. Cells were then cooled on ice and extensively washed. Transcytosis was then allowed to proceed for 2 h at 37°C. Basolateral media were then collected, and transcytosed HRP was measured. Averages obtained from two different clones for each type of transfection are shown.

Figure 8.

Freeze-fracture analysis of TJs. (A) MDCK cells were fixed and processed for freeze-fracture. Images of wild-type cells and cells overexpressing cGEF-H1-I and cGEF-H1+I, respectively, are shown. Bar, 150 nm. (B). Images obtained from freeze-fracture replicas were quantified over a total distance of 44.2 μm for wild-type cells, 28.0 μm for cGEF-H1-I, and 50.8 μm for cGEF-H1+I–overexpressing cells. The total number of segments counted for each cell line was set to 100%.

Figure 9.

Analysis of MDCK cells with reduced expression of GEF-H1. (A) MDCK cells transfected with a plasmid driving the expression of a control RNA duplex (control-RD c1) or RNA duplex targeting different regions of cGEF-H1 (GEF-RD-I c6 and GEF-RD-II c4) were harvested, and expression was analyzed by immunoblotting with mAb B4/7. ZO-1 and α-tubulin were immunoblotted as controls. The reduction of GEF-H1 expression in all clones used for the functional analysis was at least 50%. (B–D) MDCK cells were cultured on filters for 9 d and were then analyzed by immunofluorescence (B shows labelings with mAb B4/7 and a polyclonal anti–ZO-1 antibody; Bar, 8 μm), by measuring TER (C), and by determining paracellular permeability of 4 kD FITC-dextran (D). The values represent means ± SD of four clones for cells with reduced cGEF-H1 expression (GEF-RD) and two clones for control transfections (control-RD). The decrease in paracellular permeability of GEF-RD cells is significant (p <0.01 using the Student's t test).