Apelin Affects the Progression of Osteoarthritis by Regulating VEGF-Dependent Angiogenesis and miR-150-5p Expression in Human Synovial Fibroblasts

Figure 1. The expression of apelin (APLN) is positively correlated with the expression of vascular endothelial growth factor (VEGF) among osteoarthritis (OA) patients. (A) Expression levels of APLN were retrieved from the Gene Expression Omnibus (GEO) database records (accession code: GSE46750), which compares non-inflammatory areas to inflammatory areas of the synovium from 12 patients with OA. (B,C) IHC staining revealed higher levels of VEGF expression in OA synovial tissues (n = 8) compared with non-OA tissues (n = 5). (D) The ELISA assay revealed higher levels of serum VEGF expression among OA patients (n = 21) compared with controls (n = 21) and also (E) higher levels of serum APLN expression among OA patients compared with controls. (F) A positive correlation between VEGF and APLN expression in serum OA samples. NL = normal controls. * p < 0.05 compared with normal group.

Figure 2. APLN stimulates VEGF expression and angiogenesis capacity in osteoarthritis synovial fibroblasts (OASFs). (A) APLN (0, 1, 3, or 10 ng/mL) was administered to OASFs for 24 h, and the extent of VEGF transcription was examined using real-time quantitative PCR (RT-qPCR) (n = 4). (B) OASFs were incubated under various APLN concentrations for 24 h, and VEGF translation levels were examined using Western blot (n = 3). (C) OASFs were cultured with APLN for 24 h, and excreted VEGF protein levels were assayed using ELISA assay (n = 5). The culture medium was subsequently collected, and culture medium from OASFs incubated with APLN was further treated with VEGF antibody and IgG. All culture medium (with and without VEGF/IgG antibody treatment) were subjected to endothelial progenitor cell (EPC) tube formation (n = 5) (D) and EPC migration (E) assays (n = 5). * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 3. Focal adhesion kinase (FAK) phosphorylation is involved in APLN-induced VEGF synthesis. (A,B) OASFs were pretreated with a FAK inhibitor for 30 min or FAK small interfering RNAs (siRNA) for 24 h, then incubated with APLN (10 ng/mL) for 24 h. (A) VEGF levels were examined using RT-qPCR (n = 4) and ELISA assay (n = 5). (C,D) OASF culture medium collected after APLN incubation was subjected to EPC (C) tube formation (n = 5) and (D) migration assays (n = 5). (E) OASFs were incubated with APLN for the indicated time intervals, and the extent of FAK phosphorylation was examined using Western blot analysis. * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 4. Src and Akt phosphorylation are involved in APLN-induced VEGF synthesis. (A,B) OASFs were pretreated with Src/Akt inhibitors for 30 min or Src/Akt siRNAs for 24 h, then incubated with APLN for 24 h. (A) VEGF mRNA levels were examined using RT-qPCR (n = 4), and (B) excreted VEGF protein levels were quantified using ELISA assay (n = 5). (C,D) After pretreatment with Src/Akt inhibitors or Src/Akt siRNA and incubation with APLN, the culture medium was collected and subjected to EPC (C) tube formation (n = 5) and (D) migration assays (n = 5). (E) OASFs were incubated with APLN, and the extent of Src/Akt phosphorylation was examined using Western blot analysis (n = 3). (F) OASFs pretreated with a FAK inhibitor before APLN incubation were assessed for the extent of Src/Akt phosphorylation, while OASFs pretreated with a Src inhibitor before APLN incubation were assessed for Akt phosphorylation (n = 3). * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 5. Apelin suppression of miR-150-5p enhances VEGF production. (A) Open-source software (TargetScan, miRMap, RNAhybrid, and miRWalk) was employed to determine which miRNAs could potentially inhibit VEGF transcription. (B) OASFs were incubated with APLN. miR-150-5p expression levels were examined using RT-qPCR assay (n = 4). (C,D) OASFs were transfected with miR-150-5p mimic and then stimulated by APLN. (C) VEGF transcription and (D) translation levels were examined using RT-qPCR (n = 4) and ELISA assays (n = 5). (E) OASFs were transfected with the mt-VEGF-3′UTR plasmid with or without miRNA-150-5p mimic, then incubated under APLN. Relative luciferase activity reflected VEGF promoter activity (n = 6). (F) OASFs were treated with FAK, Src, or Akt inhibitor then incubated with APLN. miR-150-5p expression levels were examined using RT-qPCR assay (n = 4). * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 6. APLN knockdown in OASFs decreases VEGF-dependent angiogenesis. (A,B) Culture medium from control-shRNA and shAPLN-transfected OASFs was applied to EPCs for 24 h; EPC capillary-like structure formation and cell migration was examined using tube formation (n = 5) and migration assays (n = 5). (C) The harvested culture medium was applied to 6-day-old fertilized chick embryos for 4 days. Chorioallantoic membranes (CAMs) were then examined using microscopy and photographed, and vessels counted (n = 3). (D) Matrigel plugs containing the harvested culture medium were subcutaneously injected into the flanks of nude mice (n = 6). (E) After 7 days, the plugs were photographed, and hemoglobin levels were quantified. (D) Plug specimens were immunostained with antibodies against CD31, CD34, CD133, and APLN and (F) quantified using the IHC score. * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 7. shAPLN ameliorates histologic severity of OA. (A) Micro-CT images of the control knee, anterior cruciate transection (ACLT) knee, and shAPLN-transfected ACLT knee (n = 8). (B–E) OARSI scores (B), micro-CT parameters, including trabecular separation (C), trabecular thickness (D), and volumetric bone mineral density (E), of the control knee, ACLT knee, and shAPLN-transfected ACLT knee. (F) Specimens from the control knee, ACLT knee, and shAPLN-transfected ACLT knee were immunostained with Safranin-O, VEGF, CD31, CD133, and CD34. * p < 0.05 compared with control; # p < 0.05 compared with the APLN-treated group.

Figure 8. Schematic diagram summarizes the mechanism whereby Apelin upregulates VEGF expression in human synovial fibroblasts. APLN induces VEGF expression and angiogenesis through the FAK/Src/Akt signaling cascade and reduces miR-150-5p expression in OASFs. FAK: focal adhesion kinase; VEGF: vascular endothelial growth factor; EPCs: Human endothelial progenitor cells.