Endothelial ARHGEF26 is an angiogenic factor promoting VEGF signalling

Abstract

Aims: Genetic studies have implicated the ARHGEF26 locus in the risk of coronary artery disease (CAD). However, the causal pathways by which DNA variants at the ARHGEF26 locus confer risk for CAD are incompletely understood. We sought to elucidate the mechanism responsible for the enhanced risk of CAD associated with the ARHGEF26 locus.

Methods and results: In a conditional analysis of the ARHGEF26 locus, we show that the sentinel CAD-risk signal is significantly associated with various non-lipid vascular phenotypes. In human endothelial cell (EC), ARHGEF26 promotes the angiogenic capacity, and interacts with known angiogenic factors and pathways. Quantitative mass spectrometry showed that one CAD-risk coding variant, rs12493885 (p.Val29Leu), resulted in a gain-of-function ARHGEF26 that enhances proangiogenic signalling and displays enhanced interactions with several proteins partially related to the angiogenic pathway. ARHGEF26 is required for endothelial angiogenesis by promoting macropinocytosis of Vascular Endothelial Growth Factor Receptor 2 (VEGFR2) on cell membrane and is crucial to Vascular Endothelial Growth Factor (VEGF)-dependent murine vessel sprouting ex vivo. In vivo, global or tissue-specific deletion of ARHGEF26 in EC, but not in vascular smooth muscle cells, significantly reduced atherosclerosis in mice, with enhanced plaque stability.

Conclusions: Our results demonstrate that ARHGEF26 is involved in angiogenesis signaling, and that DNA variants within ARHGEF26 that are associated with CAD risk could affect angiogenic processes by potentiating VEGF-dependent angiogenesis.

Keywords: angiogenesis; atherosclerosis; coronary artery disease; endothelial cell.

Figure 1

Phenome-wide association results for CAD-risk lead variant rs12493885 at ARHGEF26 locus. ICD code-based 1403 phenotypes in 17 disease categories in UK Biobank were queried for association with rs12493885 and plotted by −log₁₀ (P-value). Phenotypes were declared as significantly associated with rs12493885 if they achieve a Bonferroni-corrected P-value < 0.0000356 (0.05/1403 traits) or -log₁₀ (P-value) > 4.448 (dashed line). Besides coronary atherosclerosis, rs12493885 is associated with additional phenotypes exclusively within the circulatory system, including significant association with ischaemic heart diseases, and less significant association with unstable angina and varicose veins.

Figure 2

GOF variant Val29Leu of ARHGEF26 differentially interacts with angiogenic factors and enhances pro-angiogenic MAPK signalling. (A) Scatterplot showing relative enrichment of proteins with FLAG-ARHGEF26-29Leu (Quadrant I) compared with FLAG-ARHGEF26-WT (Quadrant III) identified by immunoprecipitation and quantitative mass spectrometry from HEK293 cells. Biological replicates from two independent experiments are plotted on the x- and y-axes, respectively. (B) Confirmatory western blot of proteins IP with anti-FLAG antibody from HEK293 cells expressing FLAG-ARHGEF26-29Leu compared with FLAG-ARHGEF26-WT. Stronger band intensity was observed for MAPK1/3, MTA1, and RBBP4/7 by immunoblot (IB). Each blot represents one out of three biological replicates conducted for each IP. (C) Overexpressing FLAG-ARHGEF26-29Leu led to significantly higher pro-angiogenic phosphorylation of MAPK1/3 than FLAG-ARHGEF26-WT. HUVEC receiving an empty vector or FLAG-tagged ARHGEF26-WT or ARHGEF26-29Leu was stimulated with VEGF for 0, 5, 15, or 30 min and subjected to immunoblotting analysis using antibodies against MAPK1/3 (phosphorylated or total). Each blot represents one of four independent experiments. Bar graphs below show quantification of the IBs (n = 4 blots per column). Multiplicity adjusted P < 0.05 (‡, FLAG-WT vs. Vector; §, FLAG-29Leu vs. Vector; #, FLAG-29Leu vs. FLAG-WT) by two-way ANOVA (Tukey); error bars, mean ± SD.

Figure 3

Endothelial ARHGEF26 promotes angiogenesis in vitro. (A) Significantly reduced wound healing in HAEC receiving siRNA against ARHGEF26 (siARHGEF26) compared with control (siControl). Representative of three independent experiments, with two images captured per experiment per time point (n = 6 independent measurements). Scale = 200 μm. *Multiplicity adjusted P < 0.05 by Tukey post-hoc tests from two-way ANOVA [F = 3.341, degree of freedom (DF) = 3]. (B) Capillary tube formation in Matrigel was impaired in HAEC receiving siARHGEF26, measured by number of junctions per field and total mesh area. Representative of two independent experiments, with five images captured per experiment per time point (n = 10 independent measurements). Scale = 100 μm. *P < 0.05 by two-tailed, unpaired Student’s t-test (t = 4.374, DF = 18 for number of junctions per field; t = 3.309, DF = 18 for total mesh area). (C) 3D tube formation assay in fibrin gel using HAEC-coated beads at Days 1 and 3. Cells were treated with siControl, siARHGEF26, or siRNA against RhoG (siRhoG). Representative of three independent experiments. Scale = 200 μm. (D) Quantification of endothelial sprout number per bead (Day 1) and mean sprout length (Day 3) showing that transient knockdown of ARHGEF26 or RhoG by siRNA impaired vasculature formation in the 3D tube formation assay. Five images were captured per experiment per time point from three independent experiments (n = 12 independent measurements). *Multiplicity adjusted P < 0.05 by Dunnett post-hoc tests from one-way ANOVA (F = 9.425, DF = 2 for sprout number per bead; F = 15.36, DF = 2 for mean sprout length). Error bars, mean ± SD.

Figure 4

ARHGEF26 interacts with an angiogenic transcriptional program in EC. (A) Sanger sequencing of genomic DNA from CRISPR/Cas9-edited TeloHAEC. An 11-nucleotide (nt) fragment was deleted from Exon 2 of ARHGEF26 in KO cells. (B) Multiplex western blot of lysate from parent (P; no gene editing), WT (receiving gene editing and clonal selection with no cleavage) and KO (receiving gene editing and clonal selection with 11-nt homozygous deletion) TeloHAEC using antibodies against ARHGEF26, RhoG, and Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH, loading control). (C) Heatmaps showing the top 30 DEGs (excluding ARHGEF26) by RNA-seq of WT vs. KO TeloHAEC at resting and TNF-ɑ-treated conditions, ranked by adjusted P-value (n = 3 biological replicates per condition per genotype). Genes known to participate in angiogenesis were shown in red symbols. (D) Top 10 most significantly enriched GO terms among all DEGs from RNA-seq at TNF-ɑ-treated condition, ranked by −log₁₀ (adjusted P). Note: GO terms related to angiogenesis (red) were the most enriched pathways besides those previously known to ARHGEF26 (blue).

Figure 5

ARHGEF26 promotes VEGF signalling in EC. (A) VEGF-dependent macropinocytosis of VEGFR2 requires ARHGEF26. HUVEC receiving siControl or siARHGEF26 was labelled with a mouse anti-VEGFR2 extracellular domain antibody, and incubated at 37°C to allow VEGFR2 internalization in the presence of VEGF and 70-kDa dextran. After washing off unbound antibodies, internalized VEGFR2 was fluorescently labelled with an anti-mouse secondary antibody (green) and inspected using confocal microscopy for co-localization with 70-kDa dextran (red), a marker for macropinosome. Quantification shows significantly reduced yellow macropinocytosomes containing both VEGFR2 and dextran (arrows) by siARHGEF26 treatment (n = 20 cells per treatment). Scale = 25 μm. *P < 0.05 by two-tailed, unpaired Student’s t-test (t = 2.952, DF = 19). (B–E) Western blot analysis of VEGFR2 internalization following cell surface biotinylation, VEGF stimulation, and streptavidin-pull-down of internalized surface-biotinylated proteins. HUVEC was treated with siControl, siARHGEF26 or siRhoG (B and C). Alternatively, HUVEC was transfected with an empty vector or FLAG-tagged ARHGEF26-WT or ARHGEF26-29Leu (D and E). (B and D) Blots show internalized surface VEGFR2, as well as proteins in total lysate at the indicated times. Each blot represents one of eight independent experiments. (C and E) Quantification of VEGFR2 internalization at the indicated times following VEGF stimulation (n = 8 independent blots per time point). Multiplicity adjusted P < 0.05 (*, siARHGEF26 vs. siControl; †, siRhoG vs. siControl; ‡, FLAG-WT vs. Vector; §, FLAG-29Leu vs. Vector; #, FLAG-29Leu vs. FLAG-WT) by two-way ANOVA [Tukey; F = 45.66, DF = 6 for (C); F = 3.174, DF = 6 for (E)]. Error bars, mean ± SD.

Figure 6

Deletion of ARHGEF26 in EC, but not VSMC, reduces atherosclerosis in mice. (A) Schematic of the murine atherosclerosis model created by a single injection of GOF PCSK9-AAV8 followed by high-fat diet. (B) Study design to characterize atherosclerosis susceptibility with global or tissue-specific Arhgef26 deletion in mice. (C) Representative images of en face aortic plaque (scale = 2 mm) and aortic root sections stained with ORO (scale = 200 μm) or the monocyte/macrophage marker MOMA2 (scale = 200 μm) from mice receiving AAV-induced atherosclerosis. Note significant reduction of plaque area relative to WT in global_KO and EC_KO aortas, but not in SMC_KO aorta. Representative staining of multiple animals (exact n below and in Figure 7C). (D and E) Quantification of the en face aortic lesion area (D; n = 13, 22, 25, 24, 25, and 23, respectively) and cross-sectional aortic root lesion area (E; n = 10, 22, 22, 25, 24, and 25, respectively). All mice are male. *P < 0.05 by two-tailed, unpaired Student’s t-test [t = 3.929, 3.206, and 0.5842; DF = 33, 47, and 46 for (D); t = 3.516, 3.726, and 0.1610; DF = 30, 45, and 47 for (E); ns, non-significant]. Error bars, mean ± SD.

Figure 7

ARHGEF26 regulates ex vivo angiogenesis and plaque stability in mice. (A) Ex vivo mouse aortic ring assay. Aortic rings from WT and Arhgef26−/− (KO) mice were embedded in collagen and stimulated with PBS (control) or VEGF. The microvessel outgrowth was labelled with CellTracker Green and imaged under bright field and fluorescence on Day 4. Scale = 2 mm. (B) Quantification of relative sprout length, calculated as vessel length (a) divided by ring diameter (b) n = 18 rings per treatment (6 rings/mouse × 3 mice/genotype). *P < 0.05 by two-tailed, unpaired Student’s t-test (t = 2.912, DF = 34). (C) quantification of the content of monocyte/macrophage by MOMA2+ area on aortic root sections of mice receiving AAV-induced atherosclerosis (n = 15, 20, 23, 20, 20, and 20, respectively). *P < 0.05 by two-tailed, unpaired Student’s t-test (t = 2.974, 2.906, and 0.07233, DF = 33, 41, and 38); ns, non-significant. (D) Collagen content (blue) stained with Masson’s Trichrome and smooth muscle cell (SMC) content (yellow) stained with α-SMA on aortic root sections of EC_WT and EC_KO mice receiving AAV-induced atherosclerosis. Scale = 200 μm. Representative of 17 mice each. (E) Quantification of ORO (lipid), MOMA2, collagen, and SMA staining allows comparison of plaque stability by vulnerability index between EC_WT and EC_KO mice. Vulnerability index = (monocyte/macrophage content % + lipid content %)/(collagen content % + SMC content %) (n = 17). *P < 0.05 by two-tailed, unpaired Student’s t-test (t = 4.204, DF = 32). All mice are male. Error bars, mean ± SD.