Sema3a

Human female MSCs (donor 127, Ossium Health, Indianapolis, IN, USA) were cultured in GM comprised of αMEM without nucleosides, supplemented with 2 mM l-glutamine and 10% heat inactivated FBS (Gemini Bioscience) at 37 °C in 5% CO₂ and 100% humidity and cultured to 80% confluence in T75 flasks (Corning Inc.) before plating on the surfaces. Cells were plated at a density of 5000 cells/cm² at 37 °C, 5% CO_2, and 100% humidity for all experiments. Cultures were treated for 7 days with vehicle or soluble recombinant human semaphorin 3a (sema3a) (1 μg/mL), Wnt5a (50 ng/mL), or Wnt11 (50 ng/mL) (R&D Systems, Minneapolis, MN), based on previously defined concentrations [7 (link), 42 (link), 50 (link)]. 24 hours after plating, GM and supplemented GM were changed with subsequent media changes every 48 hours. Cells were incubated for 24 h on day 7 with fresh GM before harvest. At harvest, conditioned media were collected from the cultures and stored at − 80 °C, and MSCs were rinsed twice with 1x PBS and placed in 0.5 mL of Triton-X100 and stored at − 80 °C for biological assays. OCN (R&D Systems) and Wnt16 (Biomatik, Pittsburgh, PA) were measured in the conditioned media.

Both flanks of adult anaesthetized mice were shaven. The next day, 100 µl of 1% Evans Blue (Sigma-Aldrich) in sterile saline (wt/vol) was injected intravenously through the lateral tail vein. In some experiments, mice received an intraperitoneal injection of pyrilamine maleate (4 mg/kg body weight in 0.9% saline; Sigma-Aldrich) before Evans Blue injection to inhibit release of endogenous histamine. 30 min after Evans Blue injection, 20 µl of PBS or PBS containing 50 ng VEGF164 (PeproTech), 300 ng SEMA3A (R&D Systems), or 50 ng histamine (Sigma-Aldrich) were injected intradermally each at three sites into the flank skin of anaesthetized mice. After 20 min, mice were culled, the skin was separated from the underlying muscle, and the tissue surrounding the injection sites excised (circled in Fig. 1 A) and dried overnight at 55°C. Evans Blue was extracted by incubation in formamide at 55°C overnight and quantified by spectrophotometry at 620 nm after subtraction of background absorbance at 740 nm. Data from the three sites injected with the same agent (ligand or PBS) were averaged and expressed as fold change relative to PBS control per each mouse. In some experiments, the inner side of the skin was imaged on an MZ16 stereomicroscope (Leica) equipped with a Micropublisher camera (Perkin-Elmer). In other experiments, a sample of liver or skin tissue was retained for immunoblotting.

Recombinant human sNRP1 (1.5 nmol/L) spiked in sNRP1‐depleted serum matrix was tested in the novel sNRP1 ELISA with and without GuHCl sample pre‐treatment. The sample was further tested with and without prior addition of a molar surplus of the NRP1 ligands SEMA3A (R&D Systems) or VEGF‐A₁₆₅ (Enzo Life Sciences).
Interferences of SEMA3A or VEGF‐A₁₆₅ with sNRP1 binding of the monoclonal anti‐human NRP1 detection antibody employed in the sandwich ELISA were further tested in bio‐layer interferometry measurements. First, dissociation rates of these two NRP1 ligands were tested. Therefore, biotinylated sNRP1 (4 µg/mL) diluted in PBS was loaded to two streptavidin sensors (ForteBio). Sensors were then either incubated for 10 minutes with 10 µg/mL VEGF‐A₁₆₅ or SEMA3A. Dissociation was performed in PBS for 15 minutes. To further investigate if SEMA3A or VEGF‐A₁₆₅ binding to sNRP1 interferes with antibody binding, four additional sensors were loaded with sNRP1 as described. Two sensors were then either incubated with 10 µg/mL VEGF‐A₁₆₅ or SEMA3A for 10 minutes, while two other sensors were incubated with PBS alone acting as reference. After a one‐minute stabilization phase, association of the monoclonal antibody (2 µg/mL in PBS) was performed for 10 minutes for all four sensors, followed by a 15 minutes dissociation phase.