The scaffold RhoGAP protein ARHGAP8/BPGAP1 synchronizes Rac and Rho signaling to facilitate cell migration

Abstract

Rho GTPases regulate cell morphogenesis and motility under the tight control of guanine nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs). However, the underlying mechanism(s) that coordinate their spatiotemporal activities, whether separately or together, remain unclear. We show that a prometastatic RhoGAP, ARHGAP8/BPGAP1, binds to inactive Rac1 and localizes to lamellipodia. BPGAP1 recruits the RacGEF Vav1 under epidermal growth factor (EGF) stimulation and activates Rac1, leading to polarized cell motility, spreading, invadopodium formation, and cell extravasation and promotes cancer cell migration. Importantly, BPGAP1 down-regulates local RhoA activity, which influences Rac1 binding to BPGAP1 and its subsequent activation by Vav1. Our results highlight the importance of BPGAP1 in recruiting Vav1 and Rac1 to promote Rac1 activation for cell motility. BPGAP1 also serves to control the timing of Rac1 activation with RhoA inactivation via its RhoGAP activity. BPGAP1, therefore, acts as a dual-function scaffold that recruits Vav1 to activate Rac1 while inactivating RhoA to synchronize both Rho and Rac signaling in cell motility. As epidermal growth factor receptor (EGFR), Vav1, RhoA, Rac1, and BPGAP1 are all associated with cancer metastasis, BPGAP1 could provide a crucial checkpoint for the EGFR-BPGAP1-Vav1-Rac1-RhoA signaling axis for cancer intervention.

FIGURE 1:

BPGAP1 expression is elevated in breast cancer. (A) mRNA expression of BPGAP1 is up-regulated in breast cancer. BPGAP1 expression was analyzed by real-time PCR from the cDNA of 48 breast cancer tissues. *** represents P < 0.001, nonparametric Mann–Whitney test. (B) BPGAP1 protein expression is up-regulated in breast cancer. BPGAP1 protein expression was analyzed by immunohistochemistry in 167 breast cancer samples. BPGAP1 was detected in the cytoplasm of breast tumor epithelial cells (C). The mean IRS of cytoplasmic staining for BPGAP1 is 70. Breast cancer cells showed statistically significant higher BPGAP1 expression with IRS mean score (69.66 ± 4.57, n = 167) compared with adjacent normal breast tissues (21.50 ± 6.71, n = 20). *** represents P < 0.001, nonparametric Mann–Whitney test. (D) BPGAP1 expression is associated with cancer cells metastasized to the lymph node. * represents P < 0.05, Chi-square test. (Please refer to Table 1 for clinical data.) (E) BPGAP1 is up-regulated in breast invasive carcinoma (BRCA). Through TCGA data, BPGAP1 expression is up-regulated in primary tumors. The Mann–Whitney U test was used, and ** indicates P < 0.01. (F) BPGAP1 expression increases with the stages of breast cancer. Relative expression of BPGAP1 in normal, Stage 1, Stage 2, Stage 3, and Stage 4 breast cancer from TCGA data. The box-and-whisker plots represent median and interquartile ranges, and one-way ANOVA was used to make multiple comparisons against the normal group. Gaussian distribution was assumed, and the Greisser–Greenhouse correction was used to account for sphericity differences. * indicates P < 0.05, and **** indicates P < 0.0001. (G) Expression of BPGAP1 is higher in luminal and HER2-positive BRCA patients. Relative expression of BPGAP1 in normal, luminal, HER2-positive, and triple-negative BRCA patients in TCGA data. The box-and-whisker plots represent median and interquartile ranges, and one-way ANOVA was used to make multiple comparisons against the normal group. Gaussian distribution was assumed, and the Greisser–Greenhouse correction was used to account for sphericity differences. * indicates P < 0.05, and *** indicates P < 0.001.

FIGURE 2:

BPGAP1 is required for MCF7 breast cancer cell motility. (A) BPGAP1 promotes cell polarization and motility. MCF7 cells transfected with short hairpin RNA (shRNA) control or shRNA of BPGAP1 were seeded on collagen-coated glass bottom dishes. Time-lapse images of cells were acquired for 45 min. Representative images are shown. Scale bar: 30 μm. Red dashed lines demarcate the cell boundary. (B) Cell aspect ratio, (C) migration tracks (two representative tracks that were tracked over 2 h are illustrated in blue and red), (D) total distance migrated (over 2 h), and (E) speed of migration (tracked over 2 h) were quantified using ImageJ illustrated with in-house Matlab code. All the data above were obtained from three independent experiments and are represent as mean ± SEM. A two-tailed unpaired Student’s T test was used. *** represents P < 0.001, and * represents P < 0.05.

FIGURE 3:

BPGAP1 interacts with inactive Rac1 and promotes Rac1 activation. (A) BPGAP1 is required for Rac1 activation. MCF7 cells transfected with shRNA control or shRNA of BPGAP1 were spread on collagen-coated wells for 30 min. Cells were lysed and incubated with GST-PBD beads that captured active Rac. Active Rac1 and total lysates were analyzed by immunoblotting. Quantification on the right represents mean ± SEM of three independent experiments. A two-tailed Student’s T test was used, and **** represents P < 0.0001. (B) BPGAP1 interacts with dominant negative form of Rac1. HEK293T cells were cotransfected with Flag-tagged BPGAP1 and wild-type (WT), constitutive active (G12V), or dominant negative (T17N) forms of Rac1. The cell lysates were incubated with anti-Flag M2 beads. Bound HA-tagged proteins and their expression in WCL were detected with anti-HA, while the precipitated Flag-tagged proteins were detected with anti-Flag antibodies. Quantification on the right represents mean ± SEM of three independent experiments. One-way ANOVA was used to compare the different conditions to the control group. ** represents P < 0.01. (C) Schematic diagram of BPGAP1 truncation mutants: full-length (FL), N-terminus with proline-rich (NP), and proline-rich and C-terminus (PC). (D) BPGAP1 interacts with Rac1 via the region containing the RhoGAP domain. HEK293T cells coexpressing the indicated constructs were lysed and immunoprecipitated with anti-HA beads. Bound and WCL proteins were analyzed by immunoblotting. * indicates antibody light chains.

FIGURE 4:

BPGAP1 interacts with Vav1 upon EGF stimulation and during cell spreading. (A) Endogenous Vav interacts with BPGAP1. MCF7 cells were lysed and incubated with IgG control or anti-BPGAP1 for 6 h before adding beads conjugated with protein A. Both bound and WCL proteins were analyzed by immunoblotting with the indicated antibodies, n = 3. (B) BPGAP1 promotes interaction of Vav1 and Rac1. MCF7 cells were transfected with Flag-tagged Vav1 with or without HA-tagged BPGAP1 in control or BPGAP1 knockdown cells. Bound proteins and lysate proteins were analyzed by immunoblotting. Quantification on the right represents mean ± SEM of three independent experiments. One-way ANOVA was used to compare the different conditions to the control group. * represents P < 0.05. (C) Interaction of BPGAP1 and Vav1 is enhanced upon EGF stimulation of Ras/MEK/ERK. MCF7 cells were transfected with Flag-BPGAP1 and HA-Vav1 and stimulated with 100 ng/ml EGF for the times indicated. The cells were treated in the absence or presence of MEK inhibitor, U0126 (10 μM), for 1 h before EGF stimulation. Cells were lysed and immunoprecipitated with anti-Flag beads. Both bound and WCL proteins were analyzed by immunoblotting. Quantification on the right represents mean ± SEM of three independent experiments. One-way ANOVA was used to compare the different conditions to the control group. * represents P < 0.05, and ** represents P < 0.01. (D) Interaction of BPGAP1 and Vav1 is enhanced upon cell spreading. MCF7 cells expressing Flag-tagged BPGAP1 and HA-tagged Vav1 were seeded onto collagen-coated plates and lysed after the indicated times. Cell lysates were immunoprecipitated with anti-Flag antibodies. Both immunoprecipitated and total lysate proteins were analyzed by immunoblotting. Quantification on the right represents mean ± SEM of three independent experiments. One-way ANOVA was used to compare the different conditions to the control group. * represents P < 0.05. (E) Vav1 is localized to lamellipodia in the presence of BPGAP1. (i) MCF7 cells transfected with GFP-Vav1 in control, BPGAP1 knockdown cells, or BPGAP1 knockdown cells with reconstitution of HA-BPGAP1. Cells were seeded on collagen-coated glass coverslips for 30 min, fixed, permeabilized, and immunostained with BPGAP1 antibody, followed by secondary antibody conjugated with Alexa Fluor 555. Images were acquired using confocal microscopy. Line profiles of Vav1 from lamellipodia to lamella (as represented by the yellow dotted lines) were analyzed using ImageJ. At least 50 cells were quantified. Scale bar: 10 μm. (ii) Data sharing different letters are statistically significant at P < 0.001, Chi-square. (iii) The cytosol/nucleus ratios of Vav1 from experiments in E(i) and Supplemental Figure S4D were quantified with ImageJ as described in Materials and Methods. Scale bars: 10 μm. All data are represented as mean ± SEM of three independent experiments. A two-tailed Student’s T test was used. ** represents P < 0.01, and *** represents P < 0.001.

FIGURE 5:

BPGAP1 induces cancer cell motility, invadopodium formation, and extravasation in zebrafish larvae via the RacGEF activity of Vav1. (A, B) Vav1 is required for BPGAP1-induced Rac1 activation. MCF7 cells expressing vector control or BPGAP1 were (A) treated with Vav inhibitor, azathioprine (5 μM) or (B) transfected with siRNA specific for Vav1. Cells were seeded on collagen-coated plates for 30 min, followed by active Rac1 pull-down assay. All data are represented as mean ± SEM (n = 4 [A] and 3 [B] independent experiments). ** represents P < 0.01, and * represents P < 0.05. A one-way ANOVA test was used to compare the different conditions to the control group. (C) Vav1 is required for BPGAP1-induced cell migration. MDA-MB-231 cells stably expressing wild-type and mutants of BPGAP1 and Vav1 were seeded on collagen-coated dishes and imaged by time-lapse microscopy for 2 h. Three independent experiments were performed and are represented as mean ± SEM. Data sharing different letters are statistically significant at P < 0.05, two-way ANOVA test. (D) BPGAP1 promotes longer invadopodia via Vav. MDA-MB-231-mCherry or mCherry-BPGAP1 cells were seeded on the micropit topographic features for 6 h. Cells were treated with azathioprine 2 h after seeding. All data are represented as mean ± SEM (n = 3 independent experiments >1000 data points). ** represents P < 0.01. Unpaired two-tailed T test. (E) BPGAP1 promotes cancer cell extravasation in zebrafish larvae via Vav1. (i) MDA-MB-231-mCherry or mCherry-BPGAP1 cells were injected into the yolk of transgenic zebrafish (fli-1-EGFP) larvae 48 hpf and fixed at 70 hpf. Cancer cell extravasation is indicated by white arrows in the representative merged images. Scale bar: 100 μm. (ii) Cancer cell extravasation was quantified. Data sharing different letters are statistically significant at P < 0.05, two-way ANOVA test.

FIGURE 6:

BPGAP1 coordinates Rac1 and RhoA signaling. (A) Hypothetical working model depicting BPGAP1 as a regulator for RhoA/Rac1 coupling. The possible coupling between Rac and Rho is addressed in subsequent experiments. (B) Rac1 interaction with BPGAP1 is dependent on RhoGAP activity of BPGAP1. MCF7 cells coexpressing the constructs indicated were lysed and immunoprecipitated with anti-HA beads. Bound proteins were detected by the antibodies indicated. Quantification at the side represents mean ± SEM of three independent experiments. A two-tailed Student’s T test was used. * represents P < 0.05. (C) BPGAP1-induced Rac1 activation is dependent on its RhoGAP activity. MCF7 cells expressing the indicated constructs were treated with Rho activator CNF1 toxin (CN03 1 μg/ml) for 2 h before being seeded on collagen-coated plates for 30 min. Cell were then lysed and subjected to GST-PBD beads pulldown as described earlier. Quantification at the side represents mean ± SEM of four independent experiments. * represents P < 0.05, *** represents P < 0.001, and **** represents P < 0.0001. (D) Downstream signaling of RhoA regulates BPGAP1-induced Rac1 activation. MCF7 cells expressing HA-vector or BPGAP1 or BPGAP1-R232A were treated with Y-27632 (10 μM) for 1 h and subjected to GST-PBD beads pull down. Quantification at the side represents mean ± SEM of four independent experiments. ** represents P < 0.01, and *** represents P < 0.001. (E) Rac and Rho coupling in BPGAP1-induced cell motility. MDA-MB-231 cells stably expressing wild-type or inactive RhoGAP mutant of BPGAP1 were treated with or without azathioprine (10 μM) and/or Y-27632 (10 μM), seeded on collagen-coated dishes, and imaged for 2 h. Please refer to Supplemental Figure S7E for time-lapse images. Three independent experiments were performed and are represented as mean ± SEM. Data sharing different letters are statistically significant at P < 0.05, two-way ANOVA test.

FIGURE 7:

Schematic model depicting BPGAP1 as a switch that synchronizes Rac and Rho signaling during cancer cell motility and invasion. (A) Model for how BPGAP1 synchronizes Rho/Rac signaling during cell motility. The autoinhibited BPGAP1 is released upon growth factor stimulation and cell spreading (step 1). BPGAP1 recruits Rac1 and Vav1 (step 2; asterisk represents the “open conformation” of BPGAP1) to activate Rac1 (step 3) while inactivating RhoA, leading to reduced actomyosin contractility (step 4) and enhanced protrusive lamellipodia that collectively drive cancer cell motility and invasion (step 5). (B) BPGAP1 synchronizes Rac1/RhoA signaling. Upon growth factor stimulation and cell spreading, active BPGAP1 promotes Rac1 activation and RhoA inactivation (left panel), whereas the inactive BPGAP1 maintains a high level of RhoA activity that also impairs Rac1 binding to BPGAP1, thus preventing Rac1 from being activated (right panel). Please refer to the text for more details. “T” symbol refers to inhibition and the symbol refers to inactive RhoGAP.