Tumor necrosis factor-induced ArhGEF10 selectively activates RhoB contributing to human microvascular endothelial cell tight junction disruption

Abstract

Capillary endothelial cells (ECs) maintain a semi-permeable barrier between the blood and tissue by forming inter-EC tight junctions (TJs), regulating selective transport of fluid and solutes. Overwhelming inflammation, as occurs in sepsis, disrupts these TJs, leading to leakage of fluid, proteins, and small molecules into the tissues. Mechanistically, disruption of capillary barrier function is mediated by small Rho-GTPases, such as RhoA, -B, and -C, which are activated by guanine nucleotide exchange factors (GEFs) and disrupted by GTPase-activating factors (GAPs). We previously reported that a mutation in a specific RhoB GAP (p190BRhoGAP) underlays a hereditary capillary leak syndrome. Tumor necrosis factor (TNF) treatment disrupts TJs in cultured human microvascular ECs, a model of capillary leak. This response requires new gene transcription and involves increased RhoB activation. However, the specific GEF that activates RhoB in capillary ECs remains unknown. Transcriptional profiling of cultured tight junction-forming human dermal microvascular endothelial cells (HDMECs) revealed that 17 GEFs were significantly induced by TNF. The function of each candidate GEF was assessed by short interfering RNA depletion and trans-endothelial electrical resistance screening. Knockown of ArhGEF10 reduced the TNF-induced loss of barrier which was phenocopied by RhoB or dual ArhGEF10/RhoB knockdown. ArhGEF10 knockdown also reduced the extent of TNF-induced RhoB activation and disruption at tight junctions. In a cell-free assay, immunoisolated ArhGEF10 selectively catalyzed nucleotide exchange to activate RhoB, but not RhoA or RhoC. We conclude ArhGEF10 is a TNF-induced RhoB-selective GEF that mediates TJ disruption and barrier loss in human capillary endothelial cells.

Keywords: Rho guanine nucleotide-exchange factor; capillary leak; inflammation; permeability; transendothelial electrical resistance; vascular dysfunction.

FIGURE 1

Differentially expressed guanine nucleotide exchange factors (GEFs) alter permeability in human dermal microvascular endothelial cell (HDMEC). A, Heatmap of the expression levels (Regularized log-transformed counts) for 57 GEF genes (rows) replicate human dermal microvascular endothelial cell (HDMEC) samples stimulated with tumor necrosis factor (TNF, columns). Only GEF genes uniformly significantly differentially expressed (q-value < 0.05) between treatments are considered. Expression values are scaled by row (Z-scores). B, Trans endothelial electrical resistance (TEER) screen of short interfering RNA knockdown of the 17 significantly upregulated GEFs in HDMEC stimulated with and without TNF (10 ng/mL). The y-axis is relative TEER, which is calculated as a ratio of TEER measurement taken post-TNF to the initial TEER level before adding TNF. Non-targeting (red) and TNF-receptor 1 (green) siRNA serve as the negative and positive controls respectively. This screen demonstrates that depletion of ARHGEF10 (blue) and TRIO (purple) diminish TNF-induced leak, whereas loss of ARHGEF18 (orange) promotes more rapid recovery of barrier function after 6 hours. Results are expressed as the means relative to control with standard deviation (n = 3)

FIGURE 2

ArhGEF10 knockdown blocks and reverses endothelial permeability and improve barrier function induced by TNF. A, Human dermal microvascular endothelial cells (HDMECs) transfected with siRNA (non-targeting, si-NT or ARHGEF10, si-ARHGEF10) demonstrate efficient knockdown of ARHGEF10 under saline (PBS) and tumor necrosis factor (TNF, 10 ng/mL, 6 hours) stimulation. mRNA levels are expressed as the means relative to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) with a standard deviation (n = 3). Concordantly, HDMECs transfected with ARHGEF10 siRNA demonstrate depleted ArhGEF10 (representative western blot B, quantification C, results are expressed as the means relative to control with standard deviation, n = 3). D, E, TEER changes in HDMECs transfected with non-targeting siRNA (si-NT), siRNA for ARHGEF10 (si-ARHGEF10) or siRNA for RHOB was recorded between 48 and 72 hours of transfection stimulated in the presence and absence of TNF (10 ng/mL, 6 hours) without replacing the growth medium at 24-hour post-feeding. The y-axis shows normalized TEER calculated as a ratio of TEER measurement taken post-TNF to the basal TEER level before adding TNF, which is set at 1.0. HDMECs depleted of ArhGEF10, Rho or both demonstrate dramatically less TNF-induced leak. Error bars represent standard deviation (n = 3). Non-parametric Mann-Whitney test was used to compare groups (n = 3)

FIGURE 3

ArhGEF10 knockdown lessens TNF-induced tight junction disorganization. A, Immunofluorescence staining was performed to assess the organization of tight junction proteins—claudin-5 and ZO-1. Human dermal microvascular endothelial cells (HDMECs) transfected with non-targeting siRNA (siNT) or ARHGEF10 siRNA (siARHGEF10) were stimulated with or without TNF (10 ng/mL, 6 hours) were analyzed for claudin-5 (green) ZO-1 (red). Both groups of HDMECs treated with PBS display smooth, confluent circumferential staining of claudin-5 or ZO-1 (A, first panels #1-Green and #2-Red, white arrows). HDMEC transfected with NT siRNA and TNF displays the characteristic disrupted “saw-tooth” staining (A, third column of panels, yellow arrows). However, HDMECs depleted of ArhGEF10 and stimulated with TNF display maintained smooth junctional staining (A, fourth column of panels, white arrows). Results are typical of three separate experiments (scale bar = 25 μm). Quantification of confluent smooth staining of claudin-5 (B) and ZO-1 (C) reveal dramatically disrupted staining in the NT siRNA group with preserved junctional staining in the ArhGEF10-depleted cells

FIGURE 4

ArhGEF10 knockdown Significantly reduces the total amount of GTP-bound active RhoB but not, RhoA or RhoC in whole-cell lysates. A-C, Human dermal microvascular endothelial cells (HDMECs) transfected with non-targeting or ARHGEF10 siRNA demonstrate similar expression of RHOB and RHOC with slightly increased RHOA. Stimulation with TNF (10 ng/mL, 6 hours) reveals no changes in RHOC expression, a similar pattern but significantly less RHOB upregulation and a blunted increase in RHOA expression. mRNA levels are expressed as the means relative to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) with a standard deviation (n = 3). D-F, HDMECs transfected with non-targeting siRNA (si-NT) and stimulated with TNF (10 ng/mL, 6 hours) demonstrate increased levels of total (relative to the loading control β-actin) active RhoA and RhoB but not RhoC. In HDMEC depleted of ArhGEF10, there are significant increases in total and active RhoA, but decreases in total and active RhoB (quantification G, J, H, K, I, L). The density of the proteins in each control (NT-PBS) group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. Non-parametric Mann-Whitney test was used to compare groups (n = 3)

FIGURE 5

ArhGEF10 is a specific GTP exchange factor for RhoB in cell-free assay. We are able to immunoprecipitate ArhGEF10 protein, which was performed separately immediately prior to each cell-free assay (A, D, J, quantification relative to IgG pulldown F, H, N, n = 3). Immunoprecipitated ArhGEF10 was incubated with recombinant Rho proteins, which contained very little pre-formed active Rho, as assessed by pulldown assays (B, E, K). Pulldown of active Rho after incubation revealed no changes in GTP loading for RhoA (C) or RhoC (L), but a dramatic increase in GTP loaded RhoB (I, quantification relative to GTP loading in the IgG group G, M, O). Results are expressed as the means of three different experiments. The density of the proteins in each control (NT-PBS) group was used as a standard (1 arbitrary unit) to compare the relative density of the other groups. Non-parametric Mann-Whitney test was used to compare groups (n = 3)