Human Platelet Lysate as Alternative of Fetal Bovine Serum for Enhanced Human In Vitro Bone Resorption and Remodeling

Introduction To study human physiological and pathological bone remodeling while addressing the principle of replacement, reduction and refinement of animal experiments (3Rs), human in vitro bone remodeling models are being developed. Despite increasing safety-, scientific-, and ethical concerns, fetal bovine serum (FBS), a nutritional medium supplement, is still routinely used in these models. To comply with the 3Rs and to improve the reproducibility of such in vitro models, xenogeneic-free medium supplements should be investigated. Human platelet lysate (hPL) might be a good alternative as it has been shown to accelerate osteogenic differentiation of mesenchymal stromal cells (MSCs) and improve subsequent mineralization. However, for a human in vitro bone model, hPL should also be able to adequately support osteoclastic differentiation and subsequent bone resorption. In addition, optimizing co-culture medium conditions in mono-cultures might lead to unequal stimulation of co-cultured cells. Methods We compared supplementation with 10% FBS vs. 10%, 5%, and 2.5% hPL for osteoclast formation and resorption by human monocytes (MCs) in mono-culture and in co-culture with (osteogenically stimulated) human MSCs. Results and Discussion Supplementation of hPL can lead to a less donor-dependent and more homogeneous osteoclastic differentiation of MCs when compared to supplementation with 10% FBS. In co-cultures, osteoclastic differentiation and resorption in the 10% FBS group was almost completely inhibited by MSCs, while the supplementation with hPL still allowed for resorption, mostly at low concentrations. The addition of hPL to osteogenically stimulated MSC mono- and MC-MSC co-cultures resulted in osteogenic differentiation and bone-like matrix formation, mostly at high concentrations. Conclusion We conclude that hPL could support both osteoclastic differentiation of human MCs and osteogenic differentiation of human MSCs in mono- and in co-culture, and that this can be balanced by the hPL concentration. Thus, the use of hPL could limit the need for FBS, which is currently commonly accepted for in vitro bone remodeling models.


Micro-computed tomography (µCT)
Bioreactors were scanned and analyzed with a µCT100 imaging system (Scanco Medical, Brüttisellen, Switzerland) after 4 weeks of culture. Scanning was performed at an isotropic nominal resolution of 17.2 µm, energy level of 45 kVp, intensity of 200 µA, integration time of 300 ms and with twofold frame averaging. To reduce part of the noise, a constrained Gaussian filter was applied with filter support 1 and filter width sigma 0.8 voxel. Filtered images were segmented to detect mineralization at a global threshold of 24% of the maximum grayscale value. Unconnected objects smaller than 30 voxels were removed through component labeling.

(Immuno)histochemistry
Scaffolds (N = 2) were soaked for 15 minutes in each 5% (w/v) sucrose and 35% (w/v) sucrose in phosphate buffered saline (PBS). Samples were embedded in Tissue Tek® (Sakura, Alphen aan den Rijn, The Netherlands) and quickly frozen with liquid N2. Cryosections were sliced with a thickness of 5 μm. Upon staining, sections were fixed for 10 minutes in 3.7% neutral buffered formaldehyde and washed twice with PBS.
To visualize collagen deposition, sections were stained with Picrosirius Red. Sections were soaked in Weigert's Iron Hematoxylin (HT1079, Sigma-Aldrich) solution for 10 minutes, washed in running tap water for 10 minutes, and stained in 1% w/v Sirius Red (36,554-8, Sigma-Aldrich) in picric acid solution (36011, Sigma-Aldrich) for one hour. Subsequently, sections were washed in two changes of 0.5% acetic acid and dehydrated in one change of 70% and 96% EtOH, three changes of 100% EtOH, and two changes of xylene. Sections were mounted with Entellan (107961 Sigma-Aldrich) and imaged with a bright field microscope (Zeiss Axio Observer Z1, 20x/0.8 Plan-Apochromat objective).
To study osteogenic differentiation, sections were stained with DAPI, CNA35, osteopontin and runtrelated transcription factor 2 (RUNX2). Briefly, sections were permeabilized in 0.5% triton X-100 in PBS for 5 min and blocked in 10% normal goat serum in PBS for 30 min. Primary antibodies were incubated overnight at 4 ºC, secondary antibodies were incubated with 0.1 µg/ml DAPI and 1 μmol/mL CNA35-mCherry (3) for 1 h at room temperature. Antibodies are listed in Table S1. Images were acquired with a laser scanning microscope (Leica TCS SP5X, 63x/1.2 HCX PL Apo CS objective). All images were prepared for presentation in Fiji (4).
Subsequently, 100 µl substrate solution (10 mM p-nitrophenyl-phosphate (71768, Sigma-Aldrich) in 0.75 M 2-amino-2-methyl-1-propanol) was added and wells were incubated at room temperature for 15 minutes. To stop the reaction, 100 µl 0.2 M NaOH was added. Absorbance was measured with a plate reader at 450 nm and these values were converted to ALP activity (converted p-nitrophenyl phosphate in µmol/ml/min) using standard curve absorbance values.

Receptor activator of nuclear factor κB ligand (RANKL) and osteoprotegerin (OPG) quantification
Secreted RANKL and OPG were quantified in cell supernatants from day 7 of 2 different bioreactors containing 4 scaffolds each (N = 2) with RANKL (ab213841, Abcam, Cambridge, UK) and OPG (EHTNFRSF11B, Thermo Fisher Scientific) enzyme-linked immunosorbent assays (ELISAs) according to the manufacturer's protocols. To measure RANKL, samples were added to anti-human RANKL coated microwells. After 90 min incubation at 37 ºC, samples were replaced by biotinylated antibody solution followed by 60 min incubation at 37 ºC. After thorough washing, avidin-biotinperoxidase complex (ABC) solution was added and plates were incubated for 30 min at 37 ºC. Wells were again washed and color developing agent was added followed by 15 min incubation in the dark at 37 ºC. To stop the reaction, stop solution was added and absorbance was measured at 450 nm in a plate reader. To measure OPG, samples were added to anti-human OPG coated microwells and incubated for 2.5 h at room temperature with gentle shaking. Wells were subsequently washed, biotinylated antibody solution was added followed by 60 min incubation at room temperature with gentle shaking. After washing, streptavidin-HRP solution was added and incubated in the wells for 45 min with gentle shaking. Wells were subsequently washed and incubated with substrate solution for 30 min in the dark with gentle shaking. The enzymatic reaction was stopped with stop solution and absorbance was measured at 450 nm in a plate reader. All absorbance values were converted to RANKL and OPG concentrations using standard curve absorbance values.

Supplementary Tables
The antibodies that were used for immunofluorescent stainings of MC-MSC co-cultures (nonstimulated and osteogenically stimulated) and three-dimensional osteogenically stimulated MSC mono-cultures are listed in Table S1. Table S1. List of antibodies that were used in this study.
Abbreviations: runt-related transcription factor 2 (RUNX2), receptor activator of nuclear factor κB ligand (RANKL), osteoprotegerin (OPG). To explore the protein content of hPL, a total of 21 proteins that have been reported to influence bone resorption, formation or remodeling were quantified using multiplex immunoassays. In addition, calcium and phosphate concentration were quantified as well. The concentrations from these analyses were compared to effective concentrations from in vitro experiments reported in literature. The results from these quantifications and the literature research are reported in Table S2.  (10) Can inhibit osteoclast formation (11) Can inhibit osteogenesis in adipose tissue derived MSCs, which can be counteracted by IL6 (12,13) Can inhibit osteoblast proliferation and promote ALP production 10,000 -100,000 32.72 pg/ml 100 -10,000 10,000 100 -10,000

IL6
Could enhance osteoclastic differentiation in co-culture by stimulating RANKL production by co-cultured cells (14 (46,47) Could enhance mature osteoclast activity and resorption (46) Could enhance osteogenic differentiation and bone-like matrix formation of bone marrow derived MSCs at low coating densities, and inhibit differentiation but promote proliferation at higher coating densities (

Supplementary Figures
To visualize the non-resorbed surface, osteo assay wells were stained with a modified Von Kossa. To capture the entire well, tile scans were made with a bright field microscope. Tile scans were stitched with Zen Blue software (version 3.1, Zeiss, Breda, The Netherlands). To enable segmentation and resorption quantification, scratches that were introduced by mechanical cell removal in co-cultures were manually masked whereafter image contrast was increased using Fiji (4). A clipping mask was created in Illustrator (Adobe Inc., San Jose, CA, USA) to remove the edges of the wells. Segmentation was performed in MATLAB (version 2019b, The MathWorks Inc., Natrick, MA, USA), using Otsu's method for binarization with global thresholding, where the threshold was kept constant throughout the entire image ( Figure S1) (57). Figure S1. Workflow osteo assay wells from raw data to image segmentation. Decellularized resorption wells (A) were stained with Von Kossa (B). Scratches were manually masked (C) and a clipping mask was used to remove the edges of the well (D). Lastly, images were segmented such that the resorbed surface could be quantified (E). To check whether osteogenically stimulated MC-MSC co-cultures showed differences in RANKL and OPG expression, cells were stained for these proteins ( Figure S3).