Human adipose stromal cells derived from 3 healthy donors were kindly provided by Prof. Pojda Z., Department of Regenerative Medicine, Maria Sklodowska-Curie Natio-nal Research Institute of Oncology. Adipose tissue was collected for the routine clinical purposes (aesthetic liposuc-tion) after obtaining patient’s informed consent for the use of the leftover material for research purposes. All the procedures were approved by Institutional Review Board and were conducted in agreement with the guidelines of Helsinki Declaration.

The protocol of human adipose tissue MSCs isolation as well as identification were previously described in detail by our group (20). For growth, cells were seeded into plastic flasks at a density of 3.5×10³ cells/cm² in standard growth medium (GM) composed of DMEM (Dulbecco’s modified Eagle’s Medium; Sigma-Aldrich) supplemented with fetal bovine serum (FBS; 15%; Sigma-Aldrich) and antibiotic–antimycotic solution (Penicillin-Streptomycin- Amphotericin; 1.5%; Sigma-Aldrich) and incubated under standard cell culture conditions (37℃, 5% CO₂, 95% humidity). Cells were cultured up to passage 3, then underwent flow cytometry analysis and multilineage differentia-tion tests, the rest were cryopreserved. Refrozen hASCs from 3 donors at passages 4-5 were used in the study.

Phenotype of cells at the third passage was determined with the use of Human MSCs Analysis Kit (BD Bio-sciences). The expression of positive (CD73, CD90, CD105, CD44) and negative (CD34, CD45, CD11b, CD19, HLA- DR) cell surface markers were investigated. Cells were stained following manufacturer’s instructions and analyzed using BD FACSCanto II flow cytometer equipped with BD FACSDiva Software.

In order to confirm capacity of isolated cells to differentiation towards bone, cartilage and fat tissue cells, tests with the use of differentiation culture media were perfor-med.

In vitro osteogenic differentiation: After reaching 70% of confluence, cells were cultured in medium containing 0.1 μM dexamethasone, 10 mM β-glycerophosphate, 50 μM AA-2P (ascorbic acid 2-phosphate) in GM with 10% FBS (all the supplements were purchased from Sigma- Aldrich). After 2 weeks cells were stained with Alizarin Red (Sigma-Aldrich) in order to visualize calcium deposits.

In vitro adipogenic differentiation: After reaching 90∼100% of confluence, cells were cultured in medium containing 0.5 μM IBMX (isobutyl methyl xanthine), 1 μM dexamethasone, 10 μg/ml insulin, 60 μM indomethacin in GM with 10% FBS (all the supplements were purcha-sed from Sigma-Aldrich). After 3 weeks cells were fixed for 10 min with 4% paraformaldehyde and stained with Oil Red O (Sigma-Aldrich) in order to visualize lipid droplets.

In vitro chondrogenic differentiation: Human ASCs suspension in GM (3×10⁵ of cells) were put into 15 ml tube and centrifuged (1500 rpm, 10 min, RT) in order to form a cell pellet. Unsuspended cell pellet was cultured for 3 weeks in chondrogenic medium composed of 1% FBS, 1% ITS (insulin-transferrin-selenium), 1% Penstrept, 10 ng/ml TGF-β2 (transforming growth factor), 0.1 mM AA-2P, 0.1 μM dexamethasone, 100 μg/ml sodium pyruvate in DMEM (all the supplements purchased from Sigma-Aldrich). Medium were changed every 3 days. For histological analysis, pellet was immersed in paraffin, sectioned and stained with Toluidin blue (Sigma-Aldrich).

Human ASCs were seeded onto 96-well plate (5×10³ cells per well) and cultured in standard conditions for 24 h. Then, various AA-2P concentrations (250 μM, 500 μM, 1000 μM) were applied. Non-treated cells (GM) represented negative control. Cells from 3 independent donors were used. Two independent experiments were performed, each in triplicate. Culture media were changed once (after 48 h from the beginning of the experiment). After 5 days, cell culture media from the last day of experiment were immediately frozen in a liquid nitrogen to preserve them for a next experiment described below. After collection of media, 100 μl of MTT solution (1 mg/ml MTT in GM) was added to each well. The plate was incubated for 2 h in 37℃. After that, supernatants were removed and 100 μl of DMSO was added to each well. Then, the plate was shaken for 5 min in order to mix wells contents. Quantity of formed chromatic product was assessed colorimetrically with the use of plate reader (BioTek PowerWave XS) at a wavelength 570 nm.

Samples collected during the course of cell viability test (see above) were used to perform semi-quantitative ascorbyl radical assay. Within the experiment probes remained frozen, they were thawed separately just before a measure-ment. Approximately 50 μl of a sample was drawn into a capillary tube (hematocrit capillaries BLAUBRAND^Ⓡ, intraMARK, volume 50 μl) and incubated in ESR spectrometer sample cavity for 3 min (37℃). MiniScope MS200 X-band ESR system (Magnettech) was used for spectra collection. The settings for the detection of ascorbyl radical were as follows: center field - 330.4 mT; sweep width - 6.6 mT; sweep time - 30 s; number of scans - 8; time constant - 15 ms. In order to quantify AFR concentration, spectrum of a standard (DTBN, Sigma-Aldrich) with known concentration (0.5 μM) was registered with the same experimental parameters. Spectra simulations were performed in MATLAB Software with Easy Spin toolbox (version 5.1.11) using the “garlic” function, in order to estimate the signal intensity, as the low signal- to-noise ratio made it impossible to perform direct spectrum integration. Results from “ITS” and “no ITS” group were compared with the use of Wilcoxon signed-rank test.

The following treatments were applied to test differentiation ability of hASCs:

In all the experiments the same cells were cultured in parallel in standard GM and served as internal negative control (named CTRL).

To select the optimal experiment length for subsequent analyzes we conducted a preliminary experiment, during which cells were analyzed after 3, 5 and 7 days of culture under experimental conditions. Cells from 3 donors were seeded on μclear^Ⓡ, black 96 well culture plates (Greiner, CELLSTAR^Ⓡ), 5×10³/well. The next day, cells were exposed to above described experimental conditions: 1) AA- 2P; 2) 2% FBS; 3) AA-2P＋2% FBS. Cells cultured in GM served as internal negative control (CTRL). All samples were done in 3 technical repetitions. After 3, 5 or 7 days cells were fixed with ice-cold 75% methanol (15 min., 4℃), treated with blocking solution composed of 5% normal donkey serum (NDS), 1% bovine serum albumin (BSA) and 0.3% Triton X-100 in PBS (30 min, RT) and incubated overnight at 4℃ with anti-SCLERAXIS antibody (Thermofisher Scientific, PA5-23943, 1：50) diluted in 5% NDS, 1% BSA in PBS. Then, antibodies were rinsed 3 times with PBS and incubated with secondary antibody (donkey anti-rabbit Alexa Fluor 488, Jackson Immuno-Research) at concentration 1：150 for 1 h, RT. Cells were rinsed 3 times and after that the nuclei were stained with DAPI (20 ng/ml of DAPI solution for 4 min, RT). The cells were visualized with automated imaging reader Cytation^TM 1 (BioTek) and analyzed with Gen5 3.04 software. Two parameters were analyzed in obtained data. Firstly, the mean translocation ratio was calculated. It constituted the ratio of mean green fluorescence in the nucleus to the mean green fluorescence in the cytoplasm in certain object (a cell). At least 500 cells were included in each sample for analysis. Additionally, the mean green fluorescence in the nuclei, which refers to the nuclear expression of SCLERAXIS was compared between groups.

Human ASCs from 3 donors (cells from each donor treated separately) were seeded onto cell culture dishes (Ø 60 mm) in a density of 5×10³ cells/cm² and cultured in standard conditions. After 48 h, cells were exposed to above described experimental conditions: 1) AA-2P; 2) 2% FBS; 3) AA-2P＋2% FBS. The untreated cells cultured in parallel constituted internal control (CTRL). Experiments lasted 5 days, media were changed once after 48 h. Tested hASCs were then detached by scratching in RLT buffer (Qiagen) with β-mercaptoethanol and frozen in (−80℃). The next step was total RNA isolation with the use of RNeasy Mini Kit (Qiagen). The procedure was performed according to manufacturer’s instructions. RNA concentration and purity were then assessed with the use of Nano-Drop system (ND-1000 Spectrophotometer) at wavelengths 260 and 280 nm. Reverse transcription was performed using PrimeScript RT reagent Kit (Perfect Real Time) (TaKaRa Bio). cDNA template was synthesized using 850 ng of total RNA in 20 ul of reaction mixture using Oligo dt Primer and Random 6 mers according to manufacturer’s instructions. Quantitative PCR was conducted on 7500 Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific) using Premix Ex Taq (Perfect Real Time) (TaKaRa Bio). 2 ul of a 1：25 dilution of the appropriate reversed transcription reaction product served as template in real time qPCR. The following primers were used (all purchased from Applied Biosystems): collagen, type I, alpha 1 (COL1) Hs00164004_m1, collagen, type III, alpha 1 (COL3) Hs00943809_m1, scleraxis (SCX) Hs03054634_g1, mohawk homeobox (MKX) Hs00543190_ m1, decorin (DCN) Hs00370385_m1, Glyceraldehyde 3- phosphate dehydrogenase (GAPDH) (4333764 T) gene served as a reference gene. Each cDNA sample from each donor was analyzed for all the genes in two independent rounds. Amplification efficiency for each primer was determined using a standard curve determined by a series of five-fold dilutions of pooled cDNA from all analyzed samples. Relative gene expression changes were calculated with the use of 2^ΔΔCt method, where dCt was calculated as Ct_REF - Ct_GOI and ddCt as dCt_TREATED - dCt_CTRL. For statistical analysis ddCT values were used.

Human ASCs (cells from each donor treated separately) were seeded onto cell culture dishes (Ø 100 mm) in a density of 5×10³ cells/cm² and cultured in standard conditions. After 48 h, cells were exposed to the experimental conditions: 1) AA-2P; 2) 2% FBS; 3) AA-2P/2% FBS. Negative internal control (CTRL) was included for cells from each donor. Experiments lasted 5 days, media were changed once after 48 h. Cells from 3 donors were used in the analysis. Tested hASCs were then detached mechanically and centrifuged (3000 rpm, 5 min, 20℃). The next step was the isolation of total protein using RIPA lysis buffer supplemented with protease and phosphatase inhibitor cocktails (all the components purchased from Sigma- Aldrich). Lysis was conducted for 30 min at 4℃. Suspen-sion was then centrifuged (10,000 g, 20 min, 4℃) and supernatant containing proteins was collected. Total protein concentration was assessed colorimetrically using Bio-Rad protein assay dye reagent according to the producer’s instructions (Bio Rad Laboratories Inc). Reading was performed at a wavelength 595 nm with the use of plate reader BioTek PowerWave XS. Proteins (35 μg per well) were resolved by SDS-PAGE and transferred onto PVDF membrane. Membranes were blocked for 1 h with 5% non-fat dry milk in TBS (20 mM Tris-HCl, 500 mM NaCl) containing 0.5% Tween 20. Primary antibodies: rabbit polyclonal anti-collagen I antibody (Abcam, ab34710, 1：500), rabbit polyclonal anti-collagen III antibody (Abcam, ab7778, 1：500), rabbit polyclonal anti-mohawk antibody (LSBio, aa46-75, 1：1000), rabbit polyclonal anti-scleraxis antibody (Thermofisher Scientific, PA5-23943, 1：1000) or rabbit monoclonal anti-β-actin antibody (Santa Cruz Biotechnology, sc-47778, 1：1000) were added for overnight incubation at 4℃. After that, blots were washed with TBST (3×10 min) and incubated (1 h, RT) with appropriate secondary antibodies conjugated with IR fluorochromes: IRDye 680RD or IRDye 800CW (Li-COR Biosciences) at 1：5000 dilution. The protein synthesis was analyzed using ChemiDoc MP Imaging System (Bio- Rad Laboratories Inc.) and the integrated optical density (IOD) was quantified in relation to β-actin IOD with the Image Lab Software (Bio-Rad Laboratories Inc). Samples from each donor were processed independently at least 3 times.

Cells were seeded on Nunc^TM Lab-Tek^TM II CC2^TM 8-Chamber Slide System and after 24 h cultured for 5 days with or without an addition of 500 μM AA-2P and 1% ITS. Cells from 2 donors were used in this part of the study. After the treatment, cells were fixed with 4% paraformaldehyde (10 min, RT), permeabilized with 70% methanol (15 min, −20℃), treated with blocking solution composed of 5% normal donkey serum (NDS), 1% bovine serum albumin (BSA) and 0.3% Triton X-100 in PBS (30 min, RT) and incubated overnight on a vortex mixer at 4℃ with 2 primary antibodies solution (antibodies dissolved in 5% NDS, 1% BSA in PBS): goat polyclonal anti-actin antibody (Santa Cruz, C-11, sc1615, 1：50) and rabbit polyclonal anti-collagen I antibody (Abcam, ab34710, 1：300) or rabbit polyclonal anti-collagen III antibody (Abcam, ab7778, 1：150). After incubation, antibodies were rinsed with PBS (3×15 min). Cells were then incubated with 2 appropriate secondary antibodies conjugated with fluorochromes (Alexa Fluor 594 and Alexa Fluor 647, Jackson ImmunoResearch) at concentration 1：150 for 1 h RT. Antibodies were rinsed with PBS (3×15 min) and after that the nuclei were stained with DAPI (20 ng/ml of DAPI solution for 4 min, RT). At the end, DAPI solution was rinsed with PBS (3×5 min) and the slide was preserved. In parallel, each step was performed for a control slide (control of slide and cells autofluorescense and non-specific secondary antibody binding). Images were obtained with confocal microscope Olympus Fluoview FV1000 with FV10-ASW 3.1. imaging software and captured with 10x objective. Image acquisition was divided into 2 modules: 1) analysis of total collagen; 2) analysis of extracellular collagen. Modules differed only with parameter settings applied for detection of collagen immunostaining signal with higher sensitivity of signal detection applied for extracellular collagen staining. Signal detection parameters were established at the beginning of experiment and remained constant. For each of 8 parts of slide, 6 fields of view (possibly next to each other) with relative mean cell density were collected. Focus plane was set every time using actin staining signal on detector for actin. For each field, 2 images (one for each experiment module) were collected (first one with higher sensitivity of detection). In the first module, fluorescence signal from DAPI was also collected. In order to define cell borders in the second module, images of actin were additionally collected. On the beginning and after experiment, images from the control slide were collected using parameters for both modules. Images processing was performed with the use of ImageJ software (21). Images were analyzed in 16-bit space. Threshold value of actin fluorescence for cell border defining was determined empirically and remained constant for all the images. Extracellular area defined with the use of actin image was marked as a region of interest (ROI) and put onto collagen image. Values of mean fluorescence per cell for total and extracellular collagen were collected and analyzed statistically with the use of Mann- Whitney U test. It was assumed that this parameter reflects linearly the quantity of collagen distributed intracellularly and on the plate. Calculations were performed for each type of collagen separately. Fluorescence values of the control slide were considered.

All the statistical analyzes were conducted using STATISTICA software (StatSoft^Ⓡ-Polska). Due to the size of the samples and the need to reject the null hypothesis of the origin of analyzed data from a normal distribution (based on Shapiro-Wilk test) we performed nonparametric tests for almost all the analyzes. The groups of paired two-sample data were analyzed by Wilcoxon matched- pairs signed-ranks test. For the analysis of not-related data U Mann-Whitney test was used. Student’s t test was used to compare groups with confirmed normal distribution. Significance was set at p<0.05 and graphs are presented as mean±standard error of the mean (SEM).

RT-PCR data were analyzed in 2 step protocol. Firstly, comparison was perform to evaluate if there are significant differences between treatment and control groups. The control group constituted untreated cells from the same donors. The control group was not subject to any treatment (e.g. administration of PBS instead of the agent), hence the one-sample Student’s t-test was performed on dCt differences (confirmed normal distribution) obtained by subrtaction dCt_treated-dCt_control. In the second part of the analysis only the effect of individual treatment on gene expression was compared with each other on groups created by subtraction dCt_TREATED - dCt_CTRL:

ΔΔ AA-2P= dCt _[AA-2P]- dCt_[CTRL]

ΔΔ 2% FBS= dCt _{[2% FBS]}- dCt_[CTRL]

ΔΔ AA-2P＋2%FBS= dCt [AA-2P＋2% FBS]- dCt_[CTRL]

One way ANOVA was used for such a comparison (p<0.05). Bonferroni post-hoc test was performed for the detected differences with a changed significance level of α_critical=0.017, including the accumulation of the 1st type error in multiple comparisons. If the assumption about homogeneity of variance was not met (MOHAWK), Kruskal- Wallis test was performed. Additionally, size effects were calculated as Cohen’s d for repeated measures taking into account the correlation between the paired groups using online calculator (https://www.psychometrica.de/effect_size.html).

Each concentration starting from 500 μM of AA-2P used in this study caused a statistically significant increase of a relative cell viability (Fig. 2A). Moreover, the additional treatment with 1% ITS (v/v) resulted in a further increase in cell viability in each tested concentration compared to the corresponding cells cultured without the addition of ITS (Fig. 2A). Comparing ITS positive and negative group (n=30), the difference was significant with p=0.019. Considering results from above mentioned MTT tests and literature data, 500 μM with 1% ITS was chosen as a most optimal combination.

Spectra were successfully collected and a signal of ascorbyl radical was identified with the use of available data (g=2.0054, a_H=0.188 mT) (22). An exemplary spectrum is presented in Fig. 2B. AFR concentration in control, 250μM AA-2P and 500 μM AA-2P groups were mostly under detection limit (10 nM). The concentration estimation was performed for 1000 μM AA-2P and 2000 μM AA-2P groups only. AFR mean concentrations were lower in “ITS” comparing to “no ITS” group (15.9±0.50 nM vs. 18.5±0.71 nM in 1000 μM AA-2P group and 19.4±0.93 nM vs. 20.6±1.37 nM in 2000 μM AA-2P group), but no statistical significance was observed (p=0.4102). Percent of AA-2P oxidation to AFR was higher in 1000 μM AA-2P group than in 2000 μM AA-2P group (0.0017% vs. 0.0010%).

Immunocytochemical analysis of SCLERAXIS, tenogenesis-associated transcription factor, showed that hASCs cultured in growth medium displayed higher expression of this protein in the nuclei than in the cytoplasm - the mean translocation ratio (MTR) was 2.65, indicating that this transcription factor is active even in untreated hASC. However, the MTR increased in AA-2P treated cells as well as in AA-2P＋2% FBS conditions in comparison to control cells. This increase was statistically significant for both treatments already in the 3^rd day of experiment, but in AA-2P＋2% FBS group the translocation ratio was further increased on the 5^th day of experiment (p<0.05, Fig. 3A). There was no further increase noticed in the 7^th day of the experiment. Additionally, the mean green fluorescence was evaluated in this experiment. This value corresponds with the expression of active form of SCX. The intensity of green fluorescence in nuclei increased significantly after 3 days of treatment with AA-2P and AA-2P＋2% FBS in comparison to control cells. The increase was sustained on the 5^th day of experiment but on the 7^th day, the nuclear SCX expression was no longer higher in AA-2P＋2% FBS group than in the control cells (Fig. 3B). These results indicate that under the above described experimental conditions, the fifth day is the optimal time point for the analysis of tenogenic differentiation in hASCs.

Tenogenic differentiation in response to tested agents was evaluated based on the expression changes of tenocyte-lineage markers at both mRNA and protein levels. Culturing hASCs with reduced serum resulted in a statistically significant change only for SCX (mean 1.58-fold increase in comparison to control, p=0.036). Comparing to the untreated control AA-2P treatment led to increased gene expression of all analyzed ECM related genes. For COL1 it was 2.01-fold change, p=0.001, effect size Cohen’s d for Repeated Measures, (d_RM)=2.76, for COL3 it was 2.7-fold increase, p=0.0003, Cohen’s d_RM=5.58 and for DCN – 1.8-fold change, p=0.004, Cohen’s d_RM=5.72. Moreover AA-2P addition affected expression of genes encoding analyzed transcription factors: SCX – 1.68-fold, p=0.01, Cohen’s d_RM=1.78, whereas MKX was decreased and mean expression in AA-2P group amounted 0.32 of a control value, p<0.001, Cohen’s d_RM=4.20. The culture of hASCs in reduced serum with addition of AA-2P resulted in a statistically significant increase of all analyzed genes expression in comparison to the control cells cultured in GM: SCX – 1.95-fold, p=0.03, MKX – 3.15-fold, COL1 – 3.43-fold, COL3 – 6.92 fold and DCN – 10.81-fold, for all genes p<0.001, Cohen’s d_RM >7 and for SCX >1 (Fig. 4).

Comparing the effects of the treatments with ANOVA, a statistically significant difference between the treatment effects was demonstrated for MKX (p=0.0003) between AA-2P (resulting in decreased expression) and AA-2P＋2% FBS (increased expression). Similarly, the effects differences were significant for DCN expression (p=0.000002) although here both treatments resulted in increased gene expression. In addition, for the DCN the effect caused by serum reduction only also differs significantly from that caused by combining factors (p=0.000002). In the case of COL1 and COL 3 all treatment effects differed significantly (p<0.005) from each other (Fig. 4).

The analysis of protein expression using Western blot method demonstrated that serum reduction induced an increase of MKX synthesis in hASCs comparing to control (1.77-fold, p=0.002). The combination of 2% FBS and AA-2P caused a similar change in MKX expression in comparison to the control (1.53-fold, p=0.004). The treatment with AA-2P only did not result in a significant change in MKX synthesis (Fig. 5). In regard to SCX expression, the most prominent change was noticed in cells exposed to AA-2P - this treatment caused 5.3-fold increase in expression of this transcription factor (p=0.00006) in comparison to control. The combination of reduced serum and AA-2P treatment also induced significant increase in SCX synthesis, but the difference was less prominent (1.77-fold in comparison to control, p=0.004). The production of COL1 analyzed by Western blot was decreased in cells cultured in reduced serum and in the combination of tested factors (AA-2P＋2% FBS) and amounted 60% and 69% of control values, respectively (p<0.01 for both groups). The synthesis of COL3 was significantly decreased in cells from all the treatments in comparison to control – it amounted 82%, 60% and 68% of control values for AA-2P, 2% FBS and AA2P＋2% FBS groups, respectively; (p<0.01 for all groups) (Fig. 5).

Each of investigated cell populations displayed COLLAGEN I and III expression. Analysis of collagen distribution was reasoned by results from WB study indicating changes of collagen synthesis in response to applied treatments. Exemplary set of images used for analysis is presented in Fig. 6. Significance of changes between AA-2P and the control group (regardless of a donor) was assessed with the use of U-Mann Whitney’s test. All the changes are statistically significant (p=0.0001 for extracellular COLLAGEN I, p=0.0023 for extracellular COLLAGEN III, p=0.0077 for total COLLAGEN I, p=0.0086 for total COLLAGEN III, Fig. 7). Extracellular collagen distributed on a plate remained invisible in most of the microphotographs. Although, distribution of vesicular collagen and fibrillar collagen was possible to observe within cell area (fibrillar collagen at the cell boundaries).