Exosomal lncRNA-H19 promotes osteogenesis and angiogenesis through mediating Angpt1/Tie2-NO signaling in CBS-heterozygous mice

Abstract

Rationale: Emerging evidence indicates that the growth of blood vessels and osteogenesis is tightly coordinated during bone development. However, the molecular regulators of intercellular communication in the bone microenvironment are not well studied. Therefore, we aim to investigate whether BMMSC-Exo promotes osteogenesis and angiogenesis via transporting lnc-H19 in the CBS- heterozygous mouse model. Methods: Using RT2 lncRNA PCR array screening, we identify a bone-specific, long noncoding RNA-H19 (lncRNA-H19/lnc-H19) in exosomes derived from bone marrow mesenchymal stem cells (BMMSC-Exo) during osteogenesis. Using bioinformatics analysis, we further discovered the seed sequence of miR-106a that could bind to lnc-H19. A luciferase reporter assay was performed to demonstrate the direct binding of miR-106a to the target gene angiopoietin 1 (Angpt1). We employed an immunocompromised Nude mouse model, to evaluate the effects of BMMSC-Exo on angiogenesis in vivo. Using a micro-CT scan, we monitored microstructural changes of bone in the experimental mice. Results: BMMSC-Exo possessed exosomal characteristics including exosome size, and typical markers including CD63, CD9, and TSD101. In vitro, BMMSC-Exo significantly promoted endothelial angiogenesis and osteogenesis. Mechanistic studies have shown that exosomal lnc-H19 acts as "sponges" to absorb miR-106 and regulate the expression of angiogenic factor, Angpt1 that activates lnc-H19/Tie2-NO signaling in mesenchymal and endothelial cells. Both of these effects on osteogenesis and angiogenesis are inhibited by antagonizing Tie2 signaling. Treatment of BMMSC-Exo also restored the bone formation and mechanical quality in vivo. Conclusion: These findings provide a novel insight into how the extracellular role of exosomal lnc-H19 affects osteogenesis and angiogenesis through competing endogenous RNA networks.

Keywords: Angiogenesis; Bone formation; Extracellular vesicles; lncRNA-H19 regulation; miRNA sponge.

Figure 1

Experimental strategies and studying the angiogenesis using conditioned media secreted during BMMSCs osteogenesis. (A) Experimental strategy for studying conditioned medium (CM) derived from osteogenic BMMSCs and examined its effect on EC angiogenesis. BMMSCs were cultured in the osteogenesis induction medium (OIM). After 14 days, the CM was harvested and examined for effects on EC angiogenesis. (B-C) CM was applied at different experimental conditions. HUVEC migration was analyzed using a Trans-well migration assay. NIM-CM denotes non-osteogenesis induction medium-CM. Scale bar, 200 μm (D-F) HUVECs tube formation assay was performed in the presence of CM derived from experimental conditions and was stained with DiI-acLDL. Images (X40) were taken using a fluorescence microscope. The scale bar: 200µm (D). Quantification of CM-induced capillary tube formation (tube length and branching point.) (E-F). (G) In vitro cell, proliferation assay was measured in HUVECs culture under the presence of CM derived from various conditions. (H) qPCR analysis of mRNA transcript of Ki67 in HUVEC culture under the presence of CM derived from various conditions. Results were repeated at least three times. All data are expressed as mean±SEM. n=5-7 mice for all groups. *p < 0.05 compared with the WT-CM.

Figure 2

Characterization of BMMSC and BMMSC-derived exosomes. (A) The isolated BM was cultured under osteogenic conditions for 14 days and allowed to differentiate into BMMSCs. Scale bar, 400 µm. (B) Immunofluorescent staining showed that mMSCs co-expressed CD73 and CD44. Scale bar, 100 µm. (C) Flow cytometry analysis of BMMSCs that co-expressed CD73, and CD44 on the 14^th day. (D) Western blot analysis of protein expression of CD63, CD9, TSG101 and CYC1 from cell lysates, supernatants and exosomal contents. (E) Average concentration/size distribution of isolated exosomes from CM of BMMSCs, as evaluated by Zetasizer Nano ZS analysis. The results are from three independent experiments. n=5 mice for the experimental group.

Figure 3

Long non-coding RNA (lncRNA) profiling and lnc-H19 regulated miR-106a expression by functioning as a molecular sponge. (A) Heat map of lncRNA PCR Array shows differentially expressed lncRNAs in a representative group of exosomes. The black arrow indicated the expression of lnc-H19. (B) The validation of lnc-H19 expression by qRT-PCR analysis. (C) lnc-H19 expression was performed in HUVEC culture using qPCR assay. (D) Schematic representation of the miR-106a binding sites in lnc-H19 using in silico analysis. (E) The expression level of miR-106a in HUVEC culture treated with CBS^+/--Exo by qRT-PCR analysis. (F) Schematic representation of the miR-106a binding site in Angpt1 3' UTR using Targetscan analysis. (G) The effect of miR-106a on the luciferase activities of a reporter containing Angpt1 3' UTR in HUVEC culture. (H) qPCR analysis of Angpt1 mRNA transcript in HUVEC culture by qRT-PCR analysis. (I) Angpt1 secretion levels were assessed in cultured HUVECs by mouse Angpt1 ELISA kit. (J) BMMSCs secreted Angpt1 levels were assessed by the mouse Angpt1 ELISA kit. Results were repeated at least three times. All data are expressed as mean±SEM. n=5-8 mice for all groups. *p < 0.05 compared with the WT-Exo and #p < 0.05 compared with the CBS^+/--Exo.

Figure 4

Angiogenic enhancement by MSCs-Exo in vitro and in vivo. (A) Experimental protocol in which primary CD31+ECs derived from both WT and CBS^+/- mice and were pre-treated with Exo derived from BMMSCs culture at day 14 and studied angiogenic phenotypes. (B) qRT-PCR analysis of lnc-H19 expression in ECs. (C-D) Wound healing cell migration study of ECs in the presence of Exo administration. (E-F) Trans-well migration (upper) and Matrigel angiogenesis (lower) study of ECS in presence of Exo administration. Migrating cells were represented as the % of cell migration (F). (E) 3D-matrigel tube formation assay of ECs was treated with Exo. The photographs (X20) were taken in five random fields. The scale bar represents 200 µm. (G-H) The tube length and branching point/field were quantified in the capillary-like structure by using AngioQuant image software. (I) ECs cell proliferation assay in ECs culture treated with Exo administration. (J-L) Exo stimulates angiogenesis in mouse metatarsal assays ex vivo. The area of metatarsal vessel length and branching points was quantified in vessel-like structures using AngioQuant image software. (M) lnc-H19 expression was analyzed in metatarsal tissues in vivo. (N) Matrigel plug assay shows that Exo is angiogenic in the nude (Foxn1nu/Foxn1nu) mice model in vivo. (O) Hemoglobin level in Matrigel blood contents were observed using the hemoglobin assay kit. (P-Q) Laser Doppler scanning of blood flow over hind limbs on Day 14 after Exo in experimental mice. Results were repeated at least three times. All data are expressed as mean±SEM. n=5-7 mice for all groups. *p < 0.05 compared with the WT-EC and #p < 0.05 compared with the CBS^+/--EC.

Figure 5

lnc-H19 mediated Tie2-NO signaling is induced in the promotion of endothelial angiogenesis by Exo. (A-C) Western blot analysis of Tie2 and eNOS phosphorylation in the ECs. Cells were treated with Exo (100µg), or Angpt1 (100 ng/ml) and phosphorylations of Tie2 (Y992), eNOS (S1179) were analyzed. (D-F) ECs were treated with various concentration of Exo (0, 50, 100 µg) or treated with engineered the EXO-H19 for 24h to study endothelial NO and nitrite production by DAF-FM imaging (D-E) and Griess assay method (F). (G-H) ECs were treated with Tie2 kinase inhibitor, Tie2-Ki (1.3 nM) and studied ECs migration stimulated by various Exo concentration (0, 50, 100 µg). Photographs (X40) were taken after 24h treatment. The scale bar represents 200µm. (I-K) ECs were treated with Exo (100 µg) and Angpt1 in the presence of Tie2 knockdown and studied the phosphorylation of eNOS (S1179) and Tie2 (Y992). (L-M) Angpt1 and lnc-H19 expression were analyzed using PCR assay. (N) qPCR analysis of miRNA-106 was performed in ECs. (O) Trans-well migration activity of ECs in the presence of Exo (100 µg), Angpt1 and overexpression of H19 plasmid. Results were repeated at least three times. All data are expressed as mean±SEM. n=5-7 mice for all groups. *p < 0.05 compared with the WT-EC and #p < 0.05 compared with the CBS^+/--EC, ##p < 0.05 compared with the CBS^+/--EC+Exo, ^$p < 0.05 compared with the CBS^+/--EC and n.s denotes not significance.

Figure 6

Exo promotes osteogenesis in the CBS-heterozygous mice in vitro and in vivo. (A) lnc-H19 expression in BMMSCs culture using qPCR assay. (B) qPCR analysis of miRNA-106 expression in BMMSCs culture. (C) BMMSCs were treated with various Exo concentration (0, 50, 100 µg) for 48 h and cell proliferation assay was performed. (D-E) BMMSCs were treated with different Exo concentration (0, 50, 100 µg) under osteogenic induction medium (OIM) and ALP and ARS were confirmed on Day 7 and day 21 respectively. (F) The bar diagram represents the quantification of cellular calcium was measured in BMMSCs treated with Exo (100 µg) treatment. (G-H) Representative western blot analysis for Runx2, Bglap and GAPDH protein control. Densitometry analysis of Runx2 and Bglap protein expression as represented in the bar diagram. (I) mRNA transcript expression of osteogenic marker genes (Runx2 and Bglap). (J-K) The isolated Exo was i.v. injected via the tail vein of CBS^+/- mice at 100µg/mL per mouse for 3 times/week for 8 weeks (a total of 24 injections). Before the last injection, Exo were labeled with the ExoGlow-Vivo dye and administered intravenously via the tail vein into CBS^+/- mice. Animals were imaged after 24 h using Biospace Lab Photon Imager in vivo and ex vivo. (L) lnc-H19 expression was analyzed in femoral bone tissue using qPCR assay. (M) Alcian Blue and Alizarin Red stained skeletons of a WT and CBS^+/- and Exo treated CBS^+/- mice (CBS^+/-+Exo) after 8 weeks of Exo transplantation. (N) Bodyweight of experimental mice. (O-P) The bar diagram represents the quantification of Plasma CTX (L) and P1NP level using ELISA assay. (Q) Representative µCT cross-sectional images of distal femurs in the experimental mice. (R-V) The bone phenotype parameters were observed: Bone mineral density (BMD), bone volume per tissue volume (BV/TV) (%), trabecular number (Tb.N) (1/mm), trabecular thickness (Tb.Th.) (mm) and trabecular separation (Tb.Sp.) (mm). (W) The graph represents the biomechanical quality of the femurs: ultimate load and stiffness. (X) Hematoxylin and eosin (H & E) staining of the trabecular bone volume of the femur. (Y-Z) ALP activity and ALP expression was analyzed in femoral bone tissue in vivo. Results were repeated at least three times. Data are expressed as mean ± SEM. n = 5-7 mice per group. *p < 0.05 compared with the wild-type (WT) mice, ^#p < 0.05 compared with the CBS^+/- mice. Fig. 6j-z: all the parameters are tested in the experimental mice after 8 weeks of Exo (100 µg) transplantation.

Figure 7

Exo promotes osteogenesis via lnc-H19 dependent Tie2-NO signaling. (A-C) Representative western blot analysis of Exo induced phosphorylation of Tie-2 (at the position Y992) (B) and eNOS (at the position S1177) in BMMSCs culture (C). (D-E) BMMSCs were treated with engineered the EXO-H19 and Exo (100 µg) and studied its effect on stimulated osteogenesis. ARS staining was performed at 21 days (D). Quantification of the amount of ARS staining (E). (F-H) BMMSCs were treated with engineered the EXO-H19. NO and its metabolite nitrite were measured by both DAF-FM imaging (F-G) and Griess method (H). (I-J) BMMSCs were treated with Tie-2 inhibitor (1.3nM) and studied osteogenesis and mineralization. (K-L) Tie-2 knockdown by siRNA (100ng/mL) and overexpression of lnc-H19 using H19 plasmid in BMMSCs. ARS staining was performed. (M) qPCR assay of mRNA transcript expression of osteogenic genes (Runx2 and Bglap) in BMMSCs culture. (N-O) NO production and nitrite level was measured in presence or absence of Exo using the DAF-FM fluorimetry and Griess method respectively. (P-R) Western blot analysis of phosphorylation of eNOS (S1179) and Tie2 (Y992) in the BMMSCs. Data are expressed as mean ± SEM. n = 5-6 mice per group. *p < 0.05 compared with the WT-BMMSC, #p < 0.05 compared with the CBS^+/--BMMSC, ^$p < 0.05 compared with the CBS^+/--BMMSC. Fig. 7a-r: Exo were treated at the concentration of 100 µg to obtain the data from an in vitro culture experiment.