GPR124 facilitates pericyte polarization and migration by regulating the formation of filopodia during ischemic injury

Abstract

Prolonged occlusion of multiple microvessels causes microvascular injury. G protein-coupled receptor 124 (GPR124) has been reported to be required for maintaining central nervous system (CNS) angiogenesis and blood-brain barrier integrity. However, the molecular mechanisms by which GPR124 regulates pericytes during ischemia have remained elusive. Methods: A microsphere embolism-induced ischemia model was used to evaluate the expression of GPR124 following microsphere embolism. Immunocytochemistry and stochastic optical reconstruction microscopy imaging were used to assess the expression and distribution of GPR124 in human brain vascular pericytes (HBVPs) and after the treatment with 3-morpholino-sydnonimine (SIN-1) or oxygen-glucose deprivation (OGD). The effect of GPR124 knockdown or overexpression on HBVP migration was analyzed in vitro using wound healing assays and a microfluidic device. GPR124 loss-of-function studies were performed in HBVPs and HEK293 cells using CRISPR-Cas9-mediated gene deletion. Time-lapse imaging was used to assess dynamic changes in the formation of filopodia in an individual cell. Finally, to explore the functional domains required for GPR124 activity, deletion mutants were constructed for each of the N-terminal domains. Results: GPR124 expression was increased in pericytes following microsphere embolism. Morphological analysis showed localization of GPR124 to focal adhesions where GPR124 bound directly to the actin binding protein vinculin and upregulated Cdc42. SIN-1 or OGD treatment redistributed GPR124 to the leading edges of HBVPs where GPR124 signaling was required for pericyte filopodia formation and directional migration. Partial deletion of GPR124 domains decreased SIN-1-induced filopodia formation and cell migration. Conclusion: Taken together, our results provide the first evidence for a role of GPR124 in pericyte migration under ischemic conditions and suggest that GPR124 was essential for Cdc42 activation and filopodia formation.

Keywords: GPR124; directional migration; filopodia; focal adhesions; ischemia.; pericytes.

Figure 1

Changes in GPR124 expression after ME induction in the brain. (A) Schematic representation of the microsphere embolism (ME) model. (B-D) Real-time PCR analysis of GPR124 mRNA expression in the cortical areas of the ipsilateral hemisphere (B), isolated primary mouse pericytes (C) and endothelial cells (D) from Sham and ME mice. The cortical areas of the ipsilateral hemisphere were dissected from mice at 72 h after injection of microspheres. Sham: n = 4; ME: n = 4. (E) Real-time PCR analysis of CD13 and CD31 mRNA expression in the cortical areas of Sham and ME mice, indicating the ratio of pericytes to endothelial cells. Sham: n = 4; ME: n = 4. (F) Maximum intensity projections of confocal images from brain sections stained for GPR124 (green), CD13 (red) and DAPI (blue, nuclei) at 72 h following ME. The images on the right are magnifications of the boxed regions in the images directly to their left. Scale bar: 50 μm, magnified images, 20 μm. (G, H) Plots of the fluorescence intensity in arbitrary units (AU) along the solid lines indicated in the high-magnification images in F for Sham (G) and ME (H), respectively. (I) Signal intensities of GPR124 staining quantified from F (dot plot, Sham: n = 6; ME: n = 6). *P < 0.05, and **P < 0.01 vs. Sham. The data are presented as the mean ± s.e.m. Statistical significance was determined by unpaired Student's t-test. NS, not statistically significant.

Figure 2

GPR124 expression in focal adhesions in HBVPs. (A, B) Maximum intensity projections of confocal images from HBVPs stained for GPR124 (green), Vinculin (red, A), Paxillin (red, B), Phalloidin (magenta) and DAPI (blue, nuclei). The images on the right panel are magnifications of the boxed regions in the images. Scale bar: 40 μm (Left), magnified images, 5 μm. Representative stainings from three independent experiments are shown. (C, D) Isolated focal adhesions include the same proteins as focal adhesions found in intact HBVPs. Confocal fluorescence microscopy images of immunostained isolated focal adhesions demonstrating the presence of GPR124 (green), Vinculin (red, C) and Paxillin (red, D). Scale bar: 40 μm (Left), magnified images, 5 μm. (E) The localization of GPR124 and F-actin in HBVPs was observed by STORM. Note that there is a close interaction between GPR124 and the F-actin terminals. Scale bar: 10 μm. (F) Maximum intensity projections of confocal images from primary mouse brain pericytes stained for GPR124 (green), PDGF-β (red), Phalloidin (magenta) and DAPI (blue, nuclei). The images on the right panel are magnifications of the boxed regions in the images. Scale bar: 40 μm (Left), magnified images, 5 μm. (G, H) Western blot comparison of protein concentration in total cell lysate (T), isolated focal adhesion fractions (F) and cell body fractions (C). Equal total protein was loaded in each lane. The ratio shown on the lower (F/C) indicates the relative concentration of protein between the isolated focal adhesion and cell body fractions quantified from the intensities of bands from western blots. n =3 in each group. The data are presented as the mean ± s.e.m. Note that these only reflect the relative amount of protein in isolated focal adhesion fractions to cell body fractions. (I) Endogenous GPR124 binding to Vinculin in HBVPs incubated with or without SIN-1 (0.5 mM, 12 h). Immunoprecipitated Vinculin was immunoblotted with GPR124 and Vinculin. (J) Exogenous GPR124 binding to Vinculin in HBVPs. The HBVPs transfected with Flag-GPR124 and treated in the presence or absence of SIN-1 (0.5 mM, 12 h).

Figure 3

GPR124 distribution is altered following ischemia-like injury. (A) Representative images showing GPR124 (green, upper panel), Caveolin-1 (green, lower panel), Vinculin (red), Phalloidin (magenta) and DAPI (blue, nuclei) in the leading edge of indicated HBVPs at 4 h after scratch. The dashed line indicates the adjacent front of reorienting cells. Note that the expression of GPR124 is significantly increased in the pseudopodia of cells in the direction of migration. Scale bar: 20 μm, magnified images, 10 μm. (B-G) Fluorescence redistribution of GPR124 (B, C, D) and Caveolin-1 (E, F, G) in HBVPs treated with or without SIN-1 (0.5 mM, 12 h) and OGD (1 h). The i and ii shows higher-magnification images of the boxed regions in the image. Scale bar: 40 μm, 5 μm (i) and 10 μm (ii).

Figure 4

GPR124-dependent signaling controls pericyte migration. (A) Diagram of the knockdown system. (B) Immunoblot analysis of GPR124 protein level in control HBVPs and in HBVPs transduced with a lentivirus expressing an shRNA to silence endogenous GPR124. (C) Quantitative analyses of the results as shown in (B) and presented in the bar graph as the densitometry ratio of the control from four independent experiments. (D) Diagram of the overexpression system. (E) Immunoblot analysis of GPR124 proteins in HBVPs transfected with either control or gene construct encoding GPR124 (FLAG-GPR124, MOI = 50 and 100) to confirm overexpression of GPR124. (F) Quantitative analyses of the results as shown in (E) and presented in the bar graph as the densitometry ratio of the control from four independent experiments. (G) Cell viability was measured by the CCK8 assay in Ctrl-HBVPs and shGPR124-HBVP (n = 4). (H) Representative images of control HBVPs (control), shGPR124 HBVPs (shGPR124) and GPR124-overexpressing HBVPs (GPR124 OE) at the beginning (t = 0 h) and end (t = 24 h) of the wound healing assay. The dashed lines (black) indicate the edges of the wound immediately after scratch. The dashed lines (red) outline the area of cell migration. (I) Quantification of relative migration in a wound healing assay after 24 h (normalized to control values; n = 4 in each group). (J) Schematic representation of the microfluidic migration chamber used for observing the chemotactic response of cells exposed to chemical gradients. (K) Role of GPR124 on the migratory response of HBVPs to SIN-1. Control and GPR124 OE HBVPs exhibited migration backward gradients of SIN-1 (0.5 mM, b,f), but migrated randomly to equivalent concentration (no gradient, a,e). No directed migration was observed in shGPR124 HBVPs (c-d). Tracks of individual cells are shown. (L) Role of GPR124 on the migratory response of HBVPs to PDGF-BB. Control and GPR124 OE HBVPs exhibited migration toward gradients of PDGF-BB (100ng/mL, h,l), but migrated randomly to equivalent concentration (no gradient, g,k). No directed migration was observed in shGPR124 HBVPs (i-j). Tracks of individual cells are shown. (M, N) Dot plot of chemotactic indices of HBVP migration. *P < 0.05, **P < 0.01 and ***P < 0.001 vs. control. The data are presented as the means ± s.e.m. Statistical significance was determined by unpaired Student's t-test (for C, F, G, I) and one-way ANOVA with Dunnett's multiple comparisons test (for M, N).

Figure 5

GPR124 regulates filopodia formation in HBVPs in vitro. (A-C) Representative images and quantitative analysis of HBVPs lacking GPR124 treated with SIN-1 (0.5 mM). Cells were stained with Phalloidin (blue), VASP (red) and DAPI (cyan, nuclei). Right panels, magnification of boxed regions. Scale bar: 40 μm; magnified images, 10 μm. Quantification of VASP-labeled filopodia for mean filopodia number per cell (B, n = 9, 18, 9, 9) and mean length of filopodia (C, n = 11, 16, 11, 10) from experiments shown in A. (D-F) Representative images and quantitative analysis of HBVPs lacking GPR124 treated with PDGF-BB (100 ng/mL). Cells were stained with Phalloidin (blue), VASP (red) and DAPI (cyan, nuclei). Right panels, magnification of boxed regions. Scale bar: 40 μm; magnified images, 10 μm. Quantification of VASP-labeled filopodia for mean filopodia number per cell (E, n = 12, 12, 12, 12) and mean length of filopodia (F, n = 12, 30, 12, 11) from experiments shown in D. (G-I) Representative images and quantitative analysis of HBVPs lacking GPR124 treated with blebbistatin (20 μM). Cells were stained with Phalloidin (blue), VASP (red) and DAPI (cyan, nuclei). Right panels, magnification of boxed regions. Scale bar: 40 μm; magnified images, 10 μm. Quantification of VASP-labeled filopodia for mean filopodia number per cell (H, n = 18, 15, 12, 9) and mean length of filopodia (I, n = 12, 17, 11, 19) from experiments shown in G. *P < 0.05, **P < 0.01, and ***P < 0.001 vs. control. The data are presented as the means ± s.e.m. Statistical significance was determined by one-way ANOVA with Dunnett's multiple comparisons test. NS, not statistically significant.

Figure 6

GPR124 regulates filopodia formation in HEK293 cells upon nitrosative stress. (A) Time-lapse series of single confocal plane images taken of living normal HEK293 cells and GPR124-KO HEK293 cells. The cells were seeded on 35-mm glass-bottom dishes overnight and then stimulated with or without SIN-1 (0.5 mM) for 30 min. Scale bar: 20 μm. (B, C) Quantification of dynamic changes in the numbers of filopodia (B) and the lengths of filopodia (C) after SIN-1 (0.5 mM) treatment (n=30).

Figure 7

Essential domain of GPR124 required for SIN-1-induced pericyte migration. (A-H) Confocal images of Myc (green), VASP (red), Phalloidin (blue) and DAPI (gray) in cultured HBVPs. GPR124-KO HBVPs transfected with FL-GPR124-Myc or Myc-tagged GPR124 truncation mutants as indicated. The i and ii shows higher-magnification images of the boxed regions in the image. Plots of the fluorescence intensity in arbitrary units (AU) along the solid lines indicated in the high-magnification images in i. Scale bar: 40 μm, 5 μm (i) and 10 μm (ii). (I) 2D correlation coefficient between Myc and VASP signals (box plot; n = 26). (J) Quantification of the mean filopodia number per cell and mean length of filopodia (n = 20). ***P < 0.001 vs. FL-GPR124-Myc. The data are presented as the mean ± s.e.m. Statistical significance was determined by unpaired Student's t-test. NS, not statistically significant. SP, signal peptide; LRRs, leucine-rich regions; Ig, immunoglobulin (Ig)-like domain; Horm, hormone-binding domain; GPS, GPCR proteolytic site; 7TM, seven-pass transmembrane domains; PDZ, PDZ-binding motif.

Figure 8

Role of GPR124 in pericyte polarization and migration. (A) Immunoblot analysis of Cdc42-GTP (upper panel) and Cdc42-Total (Lower panel) in shCtrl-HBVPs and shGPR124-HBVPs cells treated with or without SIN-1. GAPDH used for normalization. (B) Densitometric analysis of active Cdc42 levels. For pull-down quantification, values were calculated as the ratio of Cdc42-GTP to total Cdc42 form and normalized to the values obtained with the control (Ctrl). Data are expressed as mean ± s.e.m. from four independent experiments (n = 3). (C) Exogenous GPR124 binding to ITSN, DOCK180 and ELMO in HBVPs transfected with Flag-GPR124 incubated with or without SIN-1 (0.5 mM) for 12 h. Flag was IP followed by IB for Flag, ITSN, DOCK180 and ELMO. (D) Schematic representation of the distribution of GPR124 in HBVPs under normal or nitrosative stress conditions. GPR124 localizes to focal adhesions and affects cell polarization by regulating polarized cytoskeletal rearrangements under conditions of cerebrovascular injury. A high-magnification view of the focal adhesion is shown. **P < 0.01 vs. control. Statistical significance was determined by one-way ANOVA with Dunnett's multiple comparisons test.