Brain-specific angiogenesis inhibitor-1 signaling, regulation, and enrichment in the postsynaptic density

Abstract

Brain-specific angiogenesis inhibitor-1 (BAI1) is an adhesion G protein-coupled receptor that has been studied primarily for its anti-angiogenic and anti-tumorigenic properties. We found that overexpression of BAI1 results in activation of the Rho pathway via a Gα(12/13)-dependent mechanism, with truncation of the BAI1 N terminus resulting in a dramatic enhancement in receptor signaling. This constitutive activity of the truncated BAI1 mutant also resulted in enhanced downstream phosphorylation of ERK as well as increased receptor association with β-arrestin2 and increased ubiquitination of the receptor. To gain insights into the regulation of BAI1 signaling, we screened the C terminus of BAI1 against a proteomic array of PDZ domains to identify novel interacting partners. These screens revealed that the BAI1 C terminus interacts with a variety of PDZ domains from synaptic proteins, including MAGI-3. Removal of the BAI1 PDZ-binding motif resulted in attenuation of receptor signaling to Rho but had no effect on ERK activation. Conversely, co-expression with MAGI-3 was found to potentiate signaling to ERK by constitutively active BAI1 in a manner that was dependent on the PDZ-binding motif of the receptor. Biochemical fractionation studies revealed that BAI1 is highly enriched in post-synaptic density fractions, a finding consistent with our observations that BAI1 can interact with PDZ proteins known to be concentrated in the post-synaptic density. These findings demonstrate that BAI1 is a synaptic receptor that can activate both the Rho and ERK pathways, with the N-terminal and C-terminal regions of the receptor playing key roles in the regulation of BAI1 signaling activity.

Keywords: 7-Helix Receptor; Adhesion; Arrestin; Brain; G Protein-coupled Receptors (GPCR); G Proteins; Rho GTPases; Scaffold Proteins; Signaling; Synapses.

FIGURE 1.

Association between the N-terminal and 7TM regions of BAI1. A, schematic drawing showing full-length BAI1 and the BAI1-ΔNT construct. Rectangles represent the thrombospondin-like repeats on the BAI1-NT, and the circle represents the GPS motif. B, lysates from HEK293T cells transiently transfected with either full-length BAI1 or BAI1-ΔNT were probed with anti-BAI1-CT antibody. The size of molecular mass markers is indicated in kDa. C, HEK293T cells were transiently transfected with a Myc-tagged expression vector for the N terminus of BAI1 (Myc-BAI1-NT) in the absence or presence of a second expression vector for BAI1 lacking the N terminus (BAI1-ΔNT). Immunoprecipitation (IP) was performed with an anti-Myc antibody bound to protein A/G beads. Co-immunoprecipitation of BAI1-ΔNT was detected via immunoblot (IB) with the anti-BAI1-CT antibody. Input lysates were examined as controls for protein amount and integrity. WB, Western blot; NT, N terminus.

FIGURE 2.

BAI1 activates the Rho pathway via coupling to Gα_12/13. A, quantification of active RhoA via pulldown with GST-RBD, a recombinant GST fusion protein corresponding to the Rho binding domain of rhotekin (*, comparison with mock-transfected; #, p < 0.05 (n = 6), one-way ANOVA). Top, Western blot analysis of active RhoA after pulldown with GST-RBD beads from HEK293T cells transiently transfected with empty vector (Mock), or expression vectors for full-length BAI1 or BAI1-ΔNT. Bottom, Western blot analysis of total RhoA levels in the input fraction of the same cells. B, quantification of active RhoA via pulldown with GST-RBD in the presence of RGSp115 (N.S., not significant; n = 3, Student's t test). Top, Western blot analysis of active RhoA via pulldown with GST-RBD beads from HEK293T cells co-transfected with RGSp115 and either empty vector (Mock), full-length BAI1, or BAI1-ΔNT. Bottom, Western blot analysis of total RhoA levels in same cells. Refer to Fig. 1B for representative total expression levels of BAI1 and BAI1-ΔNT in HEK293T cells. N.S., not significant; NT, N terminus.

FIGURE 3.

BAI1 C terminus peptide binds selectively to PDZ domains from multiple synaptic proteins. A, a recombinant GST fusion protein corresponding to the last 30 amino acids of BAI1 (GST-BAI1-CT30) was overlaid at 100 nm onto a proteomic array containing 96 distinct PDZ domains. The data shown are representative of three independent experiments. B, list of PDZ domains that interact with the GST-BAI1-CT30 fusion protein. The complete list of the PDZ proteins on this array has been described previously (23). C, co-immunoprecipitation (IP) of BAI1 and PSD-95 from mouse brain homogenates. Crude membrane fractions were collected by homogenization and centrifugation. Membrane proteins were solubilized in 2% dodecyl-β-d-maltoside, immunoprecipitated with protein A/G beads ± anti-PSD-95 antibodies, and probed via Western blot analyses with anti-BAI1-CT (Thermo Scientific) and anti-PSD-95 antibodies. IB, immunoblot.

FIGURE 4.

Truncation of the PDZ-binding domain on the BAI1-CT disrupts binding to PDZ domains and differentially regulates signaling to Rho versus ERK. A, a recombinant GST-MAGI-2 PDZ4 (GST-M2-PDZ4) fusion protein pulls down transfected BAI1-ΔNT but not the BAI1-ΔNTΔPDZ mutant from solubilized HEK293T cell lysates. B, quantification of active RhoA via pulldown with GST-RBD (n = 4). Top, immunoblot (IB) analysis of active RhoA following pulldown with GST-RBD beads from HEK293T cells transfected with empty vector (Mock) or expression vector for BAI1-ΔNTΔPDZ. Bottom, Western blot analysis of total RhoA levels in the same cells. C, quantification of phosphorylated ERK (pERK) to total ERK (tERK) from HEK293T cells transiently transfected with expression vectors for pcDNA3.1, BAI1-ΔNT, or BAI1-ΔNTΔPDZ with or without an expression vector for HIS-V5-MAGI-3 (*, comparison with mock-transfected; #, p < 0.05 (n = 7); one-way ANOVA). Top panel, immunoblot analysis of pERK. Second panel, immunoblot analysis of tERK. Third panel, immunoblot analysis of BAI1 from whole cell lysates. Bottom panel, immunoblot analysis of MAGI-3 from whole cell lysates. NT, N terminus.

FIGURE 5.

Truncation of the PDZ-binding domain on the BAI1-CT enhances β-arrestin2 association but does not affect receptor ubiquitination. A, HEK293T cells were transiently transfected with expression vectors for full-length BAI1, BAI1-ΔNT, or BAI1-ΔNTΔPDZ in the absence or presence of FLAG-β-arrestin2 (βArr2). Immunoprecipitation (IP) was performed with anti-FLAG antibody coupled to agarose beads. Co-immunoprecipitated BAI1 was detected by Western blot with anti-BAI1-CT antibody. B, expression vectors for full-length BAI1, BAI1-ΔNT, or BAI1-ΔNTΔPDZ were transiently co-transfected into HEK293T cells with an HA-ubiquitin (HA-Ub) expression vector. Immunoprecipitation was performed with anti-HA antibodies coupled to agarose beads. Co-immunoprecipitated BAI1 was visualized with anti-BAI1-CT antibody. C, HEK293T cells were transiently transfected with full-length BAI1, BAI1-ΔNT, or BAI1-ΔNTΔPDZ expression vectors and incubated with 10 mm Sulfo-NHS-ss-Biotin (Biotin). Biotinylated proteins, which represent proteins at the cell surface, and whole cell lysates (WC lysates) were visualized by Western blot using the anti-BAI1-CT antibody. IB, immunoblot; NT, N terminus; CT, C terminus; Strep, streptavidin.

FIGURE 6.

BAI1 is enriched in the postsynaptic density. A, synaptosome and postsynaptic density fractions were prepared from adult mice brains via a Percoll/sucrose gradient and 1% Triton X-100 extraction. Lysed crude membrane (MEM), synaptosome (SYN), and PSD fractions were probed via Western blot with antibodies for BAI1-CT (Thermo Scientific), PSD-95, and synaptophysin. Molecular masses of markers are indicated in kDa. B, schematic of the proposed role of BAI1 in the postsynaptic density. BAI1 is enriched in the PSD, signals through Gα_12/13 to activate the Rho pathway, and can also stimulate signaling to Rac and ERK. These signaling pathways can be regulated by BAI1 binding to a variety of PDZ domain-containing synaptic proteins. This model provides a mechanistic basis for BAI1-mediated regulation of synaptogenesis, dendritic spine integrity, and synaptic plasticity.