miR-126 regulates angiogenic signaling and vascular integrity

Abstract

Precise regulation of the formation, maintenance, and remodeling of the vasculature is required for normal development, tissue response to injury, and tumor progression. How specific microRNAs intersect with and modulate angiogenic signaling cascades is unknown. Here, we identified microRNAs that were enriched in endothelial cells derived from mouse embryonic stem (ES) cells and in developing mouse embryos. We found that miR-126 regulated the response of endothelial cells to VEGF. Additionally, knockdown of miR-126 in zebrafish resulted in loss of vascular integrity and hemorrhage during embryonic development. miR-126 functioned in part by directly repressing negative regulators of the VEGF pathway, including the Sprouty-related protein SPRED1 and phosphoinositol-3 kinase regulatory subunit 2 (PIK3R2/p85-beta). Increased expression of Spred1 or inhibition of VEGF signaling in zebrafish resulted in defects similar to miR-126 knockdown. These findings illustrate that a single miRNA can regulate vascular integrity and angiogenesis, providing a new target for modulating vascular formation and function.

Figure 1. Identification of microRNAs enriched in endothelial cells

(A) Gene expression changes were monitored by qRT-PCR during differentiation of mouse ES cells in an embryoid body (EB) model. Flk1 is expressed in vascular progenitors and mature endothelial cells, while CD31 and Tie2 are markers of mature endothelial cells. Expression was normalized to Tata-binding protein (Tbp) levels. The average of multiple experiments is shown. (B) Endothelial cells were isolated from day 7 (d7) EBs by cell sorting with anti-CD31 antibodies and microRNA arrays were performed. microRNAs enriched more than 1.5-fold relative to miR-16 in CD31⁺ cells compared to CD31⁻ cells are shown. (C) Enrichment of microRNAs identified in (B) in CD31⁺ endothelial cells sorted from E10.5 mouse embryos assayed by qRT-PCR. (D) Expression of Egfl7, miR-126 and miR-126* in EBs assayed by qRT-PCR. (E) miR-126 was enriched in sorted vascular progenitors (Flk1⁺) from d4 EBs and in endothelial cells (CD31⁺) from d7 EBs compared to ES cells. Tie2 expression was used to assess mature endothelial gene expression in sorted cells used in (B) and (E).

Figure 2. miR-126 regulates endothelial phenotype in vitro

(A) Relative levels of mature miR-126, spliced EGFL7 mRNA across the miR-126-containing intron, and EGFL7 protein (immunoblot) were measured 72 h after HUVECs were electroporated with miR-126 morpholinos (MOs) or control (con) MO. Densitometric analysis of EGFL7 protein levels is indicated above. (B) The effect of miR-126 knockdown (MO) or over-expression (mimic) on endothelial cell migration in response to VEGF in the presence or absence of a VEGF receptor inhibitor was determined by generating a “scratch” in a confluent monolayer of endothelial cells and measuring the degree of “wound closure” after 8 and 24 h. Dashed lines indicate width of “wound”. Percent wound closure is shown as the mean ± SEM of five scratches from one representative experiment. * p < 0.05 compared to control. (C) Actin cytoskeletal structure was observed by phalloidin staining of control and miR-126 knockdown HUVECs. (D) Actin staining revealed a decrease in cytoskeletal reorganization and cytoplasmic extensions into the denuded area in a scratch assay. (E) Capillary tube formation of endothelial cells transfected with control or miR-126 MOs and seeded onto Matrigel. Endothelial cells transfected with miR-126 MOs formed capillary tubes, but these tubes were unstable, exhibiting frequent dissociation of tubular structures compared to control. Shown is a representative image at 24 h after plating. Arrows indicate examples of broken tubes. Quantification of the average number of tubes per field of view at 4 h and the number of complete and broken tubes at 24 h from a representative experiment is shown below. (F) VEGF-mediated changes in cell number and cell death by TUNEL were assessed in control and miR-126 knockdown cells. VEGF treatment increased total cell number and decreased the percent of serum-starved cells that were TUNEL positive, but this effect was absent in cells with reduced levels of miR-126. * denotes a significant difference between VEGF treated and non-treated cells (p<0.05).

Figure 3. miR-126 regulates vascular integrity and lumen maintenance in vivo

(A) miR-126 and miR-126* enrichment (qRT-PCR) in GFP⁺ endothelial cells from 72 hpf Tg(flk1:GFP)^s843 zebrafish compared to GFP⁻ cells. (B) Relative levels of miR-126 and miR-126* in 72 hpf zebrafish embryos. (C) miR-126 expression monitored by qRT-PCR during zebrafish development. (D) Levels of miR-126/126* or egfl7 (across intron containing miR-126) quantified by qRT-PCR in 72 hpf zebrafish injected with miR-126 MOs relative to control. (E) Lateral views of control and miR-126 MO-injected Tg(flk1:GFP)^s843; Tg(gata1:dsRed)^sd2 zebrafish (72 hpf). Brightfield microscopy (top) revealed no major changes in gross morphology, while flk1:GFP showed normal blood vessel patterning (middle). Presence of blood cells (gata1:dsRed) in the intersomitic vessels (isv), dorsal aorta (da) and primary cardinal vein (pcv) was greatly reduced in morphants (bottom). y, yolksac; h, head; ht, heart. (F) miR-126 morphants had normal vessel patterning (flk1:GFP), but developed cranial hemorrhages (gata1:dsRed; arrow) in the head. baa, branchial arch arteries. (G) Ventral view of baa suggested smaller luminal size (indicated by arrow) in miR-126 morphants. (H) Transverse section of control or miR-126 MO-treated zebrafish revealed that the da and pcv of morphants had a smaller lumen than controls; higher magnification (right panels) of boxed area shows collapsed da and small pcv in morphants. flk1:GFP, green; ZO-1, an epithelial marker, red.

Figure 4. Identification of miR-126 mRNA targets

(A) Relative luciferase activity of constructs containing the 3′ UTR of potential miR-126 targets introduced into HeLa cells in the presence of miR-1 or miR-126. The 3′ UTR was also inserted in the antisense orientation as a control (control 3′ UTR). Firefly luciferase activity for each construct was normalized to the co-transfected Renilla luciferase construct and then normalized to the change in pGL3 luciferase in the presence of microRNA. For each 3′ UTR construct, normalized luciferase activity in the absence of microRNA was set to 1. * p < 0.05 compared to pGL3. (B) Relative luciferase activity of select constructs in (A) in HUVECs upon inhibition of miR-126 with MOs. Normalization was performed as in (A). (C) mRNA levels of potential targets in HUVEC cells transfected with miR-126 mimic or MO quantified by qRT-PCR. Values are relative to transfection controls. (D) Immunoblot of SPRED1, VCAM1 and PIK3R2 protein in control or miR-126 MO-transfected HUVECs 72 h post-transfection. GAPDH is shown as a loading control. (E) Sequence complementarity of a potential miR-126 binding site in the zebrafish spred1 3′ UTR. (F and G) Luciferase assays (as in (A) and (B), respectively) using the zebrafish spred1 3′ UTR.

Figure 5. miR-126 positively regulates VEGF signaling in endothelial cells by repressing SPRED1 and PIK3R2

(A) Immunoblot of lysates from HUVECs transfected with control or miR-126 MOs in the presence or absence of VEGF (10 ng/mL, 10 min). VEGF induced phosphorylation of ERK (p-ERK) and AKT (p-AKT), which was blocked by miR-126 inhibition. Total levels of ERK and AKT were not affected. Phosphorylation of SRC was not affected by miR-126 knockdown. Densitometric analysis of normalized protein levels are indicated above. (B) PIK3R2 mRNA was knocked-down by RNAi in HUVECs transfected with control or miR-126 MOs (qRT-PCR). Western analysis indicates a decrease in PIK3R2 protein by introduction of siRNA, even in the presence of miR-126 MOs. Knockdown of PIK3R2 rescued the defect in VEGF-dependent phosphorylation of AKT in miR-126 MO-treated cells. (C) Western analysis shows reduced SPRED1 levels by transfection of a MO that blocks SPRED1 translation, even in the presence of miR-126 MO. SPRED1 MO rescued the defect in VEGF-induced phorphorylation of ERK in miR-126 MO-transfected cells. (D) Quantification of percent (%) wound closure of endothelial cells in a “scratch” assay reveals rescue of miR-126 MO effects by knocking down SPRED1. * p < 0.05 compared to control MO. (E) Treatment of 48 hpf embryos with a VEGF receptor inhibitor (Vatalanib, 5 µM, 18 h) resulted in reduced circulation of blood cells (gata1:dsRed). ISVs (flk1:GFP) appeared to lack a lumen after VEGF inhibition. (F) Hemorrhages also occurred in some embryos treated with VEGF receptor inhibitor (arrow).

Figure 6. Increased Spred1 causes vascular instability and hemorrhage similar to miR-126 knockdown

(A) Levels of phosphorylated Erk were reduced in zebrafish embryos (8 hpf) injected with spred1 mRNA compared to control mRNA (hadp1). Densitometric analysis is shown above. (B) Zebrafish Spred1, but not Hadp1, expression also reduced p-ERK in COS cells as observed by immunoblot. (C) Lateral view of trunk region of 48 hpf embryos after injection of control (hadp1) or 100 pg of spred1 mRNA. flk1:GFP reveals normal vascular patterning but gata1:dsRed shows diminished blood cells in the dorsal aorta (da), primary cardinal vein (pcv) and intersomitic vessels (isv). (D) Injection of spred1 mRNA also resulted in pericardial (2^nd panel from right) and cranial (far right) hemorrhages (arrows) visualized by gata1:dsRed marking of blood cells, and edema (2^nd panel from left). (E) Confocal analysis of transverse sections of 48 hpf embryos revealed collapsed lumens of da and primary cardinal vein (pcv) in spred1-injected embryos.

Figure 7. Putative model of miR-126 function in endothelial cells

miR-126 represses SPRED1 and PIK2R2, which negatively regulate VEGF signaling (and possibly other growth factor signaling pathways) via the MAP kinase and PI3 kinase pathways, respectively. Thus, miR-126 promotes VEGF signaling, angiogenesis and vascular integrity by inhibiting protein production of endogenous VEGF repressors within endothelial cells.