Vascular endothelial growth factor expression and their action in the synovial membranes of patients with painful knee osteoarthritis

This study was approved by the Institutional Review Board for Clinical Research and Treatment in Kitasato University (approval No. B13–113). Sample size was determined with a power analysis for an alpha of 0.05 and power of 0.80 using G*POWER3. Patients scored their pain on a 0 to 10 cm visual analog scale (VAS). Power analysis showed that 44 SM samples of patients with VAS < 6 and 58 SM samples of patients with VAS ≥ 6 were required to identify a difference in VEGF expression between the two groups. SM samples were harvested from 102 patients undergoing total knee arthroplasty. The study enrolled 22 men and 80 women (age 46–89 years, mean ± standard error (SE) = 73.2 ± 0.8 years; body mass index (BMI) range 18.4–36.7, mean ± SE = 26.0 ± 0.4 kg/m²) with radiographic evidence of KOA (unilateral Kellgren/Lawrence [K/L] grades 2 (3/102, 2.9%), 3 (36/102, 35.3%) and 4 (63/102, 61.8%)). All patients provided informed consent for participation in this study 1 day before surgery. SM samples were harvested intraoperatively from the suprapatellar pouch of each operated knee and immediately stored frozen in liquid nitrogen at − 80 °C until required for extraction of RNA. SM samples obtained from six patients were used for cell culture. The remaining samples were fixed in 4% paraformaldehyde phosphate-buffered solution (Nacalai Tesque, Kyoto, Japan) for 72 h for use in histological analysis.

Extraction of total RNA from SM and cultured SM cells and cDNA synthesis was conducted as reported previously [45]. PCR primer pair sequences for use in qPCR analysis were: VEGF-Forward (5′- TTGCCTTGCTGCTCTACCTC-3′) and VEGF-Reverse (5′- AGCTGCGCTGATAGACATCC-3′) for VEGF amplification (product size: 117 bp); apelin-Forward (5′- GAATCTGCGGCTCTGCGT-3′) and apelin-Reverse (5′- CATCAGGGACCCTCCACACA-3′) for apelin amplification (product size: 76 bp); and GAPDH-Forward (5′-TGTTGCCATCAATGACCCCTT-3′) and GAPDH-Reverse (5′-CTCCACGACGTACTCAGCG-3′) for GAPDH amplification (product size: 223 bp). Specificity of the qPCR products was evaluated using melting curve analysis. Relative mRNA expression levels of VEGF and apelin were evaluated using qPCR (CFX-96®, Bio-Rad, Richmond CA, USA). Expression levels of VEGF and apelin mRNA were normalized to the expression of the housekeeping gene, GAPDH.

Expression levels of VEGF mRNA were compared between the strong/severe (VAS ≥ 6) and mild/moderate pain (VAS < 6) groups (Table 1), using VAS = 6 as a cutoff based on previous reports [39, 46, 47]. The correlation between VAS levels and VEGF mRNA expression was also determined. Relative VEGF expression was calculated based on the mean of all samples of the VAS < 6 group.

To investigate the relationship between VEGF and K/L grades, the 102 knee OA patients were divided into three groups based on their K/L grade (2, 3, or 4). The clinical characteristics of patients in each of these groups are shown in Table 2. Relative VEGF expression was calculated based on the mean of all samples of the K/L2 group.

Following fixation, SM samples were embedded in paraffin, sectioned at 3 μm thickness, then deparaffinized (Clear Plus®, FALMA, Tokyo, Japan) and pretreated with sodium citrate buffer (pH 6.0) containing 0.1% polyoxyethylene sorbitan monolaurate (Nacalai Tesque, Kyoto, Japan) at 98 °C for 20 min for antigen retrieval. The sections were subsequently washed three times with phosphate-buffered saline for 5 min and incubated with rabbit polyclonal anti-VEGF antibody (1:100 dilution; Santa Cruz Biotechnology Inc., Santa Cruz CA, USA) and mouse monoclonal anti-apelin antibody (1:100 dilution; Santa Cruz Biotechnology) for 4 h at 4 °C. The sections were additionally incubated with Alexa 488 Fluor®-conjugated goat anti-rabbit IgG antibody (1:100 dilution; Thermo Fisher Scientific, Waltham MA, USA) and Alexa 594 Fluor®-conjugated goat anti-mouse IgG antibody (1:100 dilution; Thermo Fisher Scientific) for 1 h at room temperature. The distribution of fluorescence in SM sections was analyzed using a fluorescence microscope (Axiovert 200®, Zeiss, Jena, Germany).

Synovial membrane cells (SMCs) were isolated from 500 mg SM using 40 mL of a 1 mg/mL collagenase solution. The SMCs were incubated in α-minimal essential media (α-MEM; Nacalai Tesque) containing 10% fetal bovine serum in six-well plates. After 1 week, the SMCs were incubated with vehicle (serum free α-MEM) or 10 or 100 ng/mL human recombinant VGEF (Biolegend, San Diego CA, USA) for 24 h. Subsequently, total mRNA and protein were extracted and used in western blotting and qPCR analysis. Relative expression was calculated based on the mean of all samples of the vehicle-treated group.

To investigate the regulation of apelin by VEGF, SMCs harvested from six patients were stimulated with 1 ng/mL or 10 ng/mL VEGF for 24 h. Using methodology described elsewhere [48], SMCs were lysed in radioimmunoprecipitation buffer (Thermo Fisher Scientific) containing a protease inhibitor cocktail (Sigma-Aldrich, St. Louis MO, USA). Protein concentration was determined for each cell extract using a bicinchoninic acid (BCA) assay (Thermo Fisher Scientific). A total of 30 μg of each protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electrophoretically transferred to polyvinyl difluoride membranes. These membranes were then blocked with polyvinylidene fluoride (PVDF) blocking reagent (DS Pharma Biomedical, Suita, Japan) for 1 h and incubated overnight at 4 °C with mouse monoclonal primary antibody against apelin (1:200 dilution; Santa Cruz Biotechnology Inc.). The membranes were washed with Tris-buffered saline containing 0.05% Tween and then incubated with horseradish peroxidase-conjugated anti mouse IgG (1:1000 dilution; GE Healthcare, Piscataway NJ, USA). Apelin proteins were visualized by chemiluminescence using an enhanced chemiluminescence detection system (GE Healthcare) and exposure of the membranes to x-ray film. Bands were quantified by densitometric scanning using ImageJ software (NIH, Bethesda MD, USA). Densitometry levels of apelin proteins were normalized against that of β-actin.

Differences in VEGF expression between the VAS < 6 and VAS ≥ 6 groups were compared using the Mann-Whitney U-test. Differences in VEGF expression among K/L2, 3 and 4 subjects were compared using the Kruskal-Wallis test. Tukey’s multiple comparisons test was used to compare vehicle control and VEGF-treated cells. The relationship between VEGF expression and VAS was evaluated using Spearman’s correlation coefficient. All statistical analyses were conducted using SPSS software (v. 19.0; SPSS, Chicago IL, USA), with a P value < 0.05 considered statistically significant for all analyses.

The VAS ≥ 6 and VAS < 6 groups did not differ with regard to patient age, male/female ratio, BMI, or KL 2/3/4 ratio (Table 1). qPCR analysis showed that VEGF expression in the SM was significantly higher in the VAS ≥ 6 group than the VAS < 6 group (Fig. 1a, P < 0.05). VEGF levels were also positively correlated with VAS (Fig. 1b, ρ = 0.346, P < 0.05).

The three K/L grade groups did not differ with regard to patient age, male/female ratio, BMI, or VAS (Table 2). There was no difference in synovial VEGF expression among the K/L2, 3 and 4 groups (Fig. 2).

Immunohistochemical analysis was conducted to investigate the distribution of VEGF and apelin in the SMs of KOA patients (Fig. 3). Immunostaining showed that VEGF and apelin protein were both expressed in the synovial lining layers (Fig. 3).

qPCR analysis showed that the expression of apelin mRNA increased significantly in SMCs following 10 and 100 ng/mL VEGF stimulation (1.73-fold and 1.69-fold, respectively, P < 0.05; Fig. 4). Western blotting analysis showed that the expression of apelin was significantly increased in SMCs in the presence of exogenously added 10 and 100 ng/mL VEGF (1.85-fold and 1.56-fold, respectively, P < 0.05; Fig. 5a and b).