Comparative impact of platelet rich plasma and transforming growth factor-β on chondrogenic differentiation of human adipose derived stem cells

Cell isolation, expansion, and characterization

Liposuctioned samples were collected from healthy patients after obtaining their written consent according to the ethics of Stem Cell Technology Research Center (Tehran, Iran). Adipose-derived stem cells (ADSCs) were isolated according to a previously published procedure.¹⁰ Isolated cells were cultured in DMEM (Gibco, Cat. No. 21885025) supplemented with 10% FBS (Gibco, Cat. No. 10270106), 1% pen/strep (Gibco, Cat. No. 15070063), and 1% amphotericin (Gibco, Cat. No. R01510). Cells used for differentiation experiments were collected after 3 to 6 subcultures.

To confirm the multipotent characteristics of the isolated cells, osteogenic and adipogenic differentiations were conducted and then treated cells were stained by Alizarin Red (Fisher Chemical, CAS Number: 130-22-3) and oil Red (Acros Organics, CAS Number: 1320-06-5), respectively. Moreover, expression levels of CD₉₀, CD₁₀, CD₃₄, and CD₄₅ were analyzed by flow cytometry and the histograms were drawn using color codes.

PRP preparation

Blood samples with heparin as anti-coagulant factor centrifuged at 1500 rpm for 10 minutes. Plasma (without the buffy coat) was collected and centrifuged at 4000 rpm for 15 minutes. The sediment of platelets was dissolved in platelet-poor plasma (PPP), and activated using 10% CaCl₂ at 4˚C overnight and finally centrifuged at 3000 rpm for 15 minutes. Platelet components (e.g. growth factors, differentiators) that were released via CaCl2 osmotic shock, were pooled in a couple of containers and stored in stocks at -70°C. This extract remained applicable for up to three months. In order to retain PRP concentration at 5% (v/v) in the culture media, the DMEM supplemented with PRP was replaced every two days up to 2 weeks.

Chondrogenic differentiation

The study was performed in three groups of control (no differentiation inducer added), TGF-β (Invitrogen RP8600), and PRP. Next, in each group, 2.5×10⁵ cells were centrifuged (1200 rpm, 5 minutes) to form a pellet. Control medium consisting of proline (400 µg/mL, Sigma P3350000), pyruvate (1 mg/mL, Sigma Cas No.: 127-17-3‎), ascorbic acid (50 µg/mL, Sigma A92902-25G), and ITS (insulin-transferrin-selenium) 1% (Sigma, I3146) was used for the control group. Standard chondrogenic differentiation medium, which contains TGF-β, included TGF-β 110 ng/mL, dexamethasone 10^-7M, proline (400 µg/mL), pyruvate (1 mg/mL), ascorbic acid (50 µg/mL), and ITS 1%. For PRP–based chondrogenic differentiation, DMEM was supplemented with 5% PRP. This procedure repeated exactly for 14 days. PRP concentration of 5% (v/v) was selected for this study since it was reported that proliferation rate of ADSCs was higher in the 2.5%, 5%, and 7.5% PRP groups.¹³

Alcianblue staining and immunochemical analyses

Samples were prepared with five-micrometer thickness sections and used for immunocytochemistry (ICC) and Alcian blue 8GX (Roth) staining. The immune staining was performed on deparaffinized sections using Col II and X rabbit anti-human antibodies (Abcam, 1:20 and 1:80 dilutions), respectively. Col stands for collagen. The PE-conjugated secondary antibody (eBioscience, 1:80 dilution, goat anti-rabbit) was used for fluorescence visualization. For proteoglycan production assays, sections were stained with Alcian blue 8GX and examined using Genucam digital camera.

Extraction of miRNA and total RNA, cDNA synthesis, and RT Q-PCR

At 14th day of differentiation, total RNA was extracted using RNA extraction kit (CinnaGen, Iran). Then, cDNA was synthesized using cDNA synthesis kit and random hexamers according to the manufacturer’s instructions (Fermentas, Waltham, USA).

The miRNA first-strand cDNA was synthesized according to the manufacturer’s instruction (BONmiR, Iran). Briefly, total RNA was polyadenylated by poly (A) polymerase and reverse transcribed using general RT primer, according to the manufacturer’s protocol (BONmiR, Iran).

Finally, relative fold changes of miRNA expression in PRP and TGF-β groups were normalized against the control group using the comparative CT (2^−ΔΔCT) method with U6 small nuclear RNA (snRNA) as an internal control. The primers used in this study are presented in Table 1.

Monitoring the hypertrophic markers expressions

Chondrogenic differentiation assay by glycosaminoglycan (GAG) production

The GAG production in TGF-β and PRP treatments was assigned to the increment of blue color according to the RGB measurement tool of Image J software (Fig. 2). In this regard, the blue color intensity of the control sample (144.88 ± 4.44) significantly increased to 170.17 ± 7.61 and 166.78 ± 7.72 in TGF-β and PRP treatments, respectively.

Fig. 2

Alcian blue staining of cultured chondrogenic differentiated cells.

As presented in Fig. 3, Col-II deposition in the extracellular matrix of differentiated pellets was observed in TGF-β and PRP treatments. All samples were simultaneously stained with DAPI (4′, 6-diamidino-2-phenylindole) to confirm their vitality. The deposition of Col-X in the extracellular matrix, as a marker of hypertrophic maturation, was also observed for both TGF-β and PRP treatments. The comparative intensity of red color against the background (black color) was measured by RGB analysis tool of ImageJ software. Then, intensity of Col-II and X deposition was normalized to the intensity of DAPI (blue color). In this regard, 91.53 % ± 3.71 of living ADSCs that were treated with TGF-β, deposited Col-II in their extracellular matrix (ECM), while 47.11 % ± 3.08 of them deposited Col-X in their ECM. In the case of cells that were treated with PRP, 98.1 % ± 10.21 of living ADSCs deposited Col-II in their ECM while 47.60 % ± 3.08 of them deposited Col-X in their ECM.

Fig. 3

Expressions of collagen II and X in chondrogenically differentiated pellets were visualized by PE-immunostaining. From left to right, first and second columns represent immunocytochemistry (ICC) staining for PE-conjugated anti-Col II and their corresponding DAPI stained micrograph, respectively. The third and fourth columns, from left to right, represent immunostaining with PE-conjugated anti-Col X and their corresponding DAPI stained micrograph, respectively.

mRNA expression of chondrogenic markers

Normalization of real-time Q-PCR expression in respect to the control showed that application of either TGF-β or PRP to the differentiation media considerably upregulated SOX9 and Col-II while had no significant impact on the expression of Col-X (Fig. 4).

Fig. 4

Comparative normalized gene expression in chondrogenically differentiated pellets after 14 days. The vertical axis represents relative gene expression, which is normalized against control. Relative quantification is based on the expression levels of SNORD gene. The star on the bar stand for differences with P value less 0.05.

microRNA expression of chondrogenic markers

At 3rd day of differentiation, PRP/TGF-β relative level of miR-146a expression upregulated significantly while PRP/TGF-β relative expression level of other miRNAs remained unchanged (Fig. 5). At 14th day of differentiation, PRP/TGF-β relative level of all five miRNAs upregulated (Fig. 5).

Fig. 5

Expressions of miRNAs at (A) 3rd and (B) 14th days in chondrogenically differentiated cultures. Relative quantification is based on the expression level of SNORD gene. Present demonstration is based on normalizing PRP gene expression against TGF-β using REST software.

Alkaline phosphatase and calcium content assay

The ALP activity of PRP treatment was considerably higher than that of TGF-β (Fig. 6A). Based on calcium content assay (Fig. 6B), calcium deposition in the PRP group was significantly lower than that in TGF-β.

Fig. 6

Expressions of angiogenic and inflammatory factors by chondrogenically differentiated cells at 14th day. (A) The alkaline phosphatase (ALP) assay, (B) Calcium deposition assay, (C) Expression profiles of VEGF in different groups, and (D) Expression profile of TNFα. The “*” demonstrate significant difference in respect to TGF-β.

VEGF and TNFα secretion

When ADSCs were treated with PRP, the least VEGF secretion was observed (Fig. 6C). Treatment of ADSCs with TGF-β resulted in higher TNFα secretion compared to the PRP group (Fig. 6D).

How GAG production, Col II deposition, and Col X expression are affected by our treatments?

The accelerated chondrogenesis is evidenced by an increase in GAG content.¹⁵ Treatment of ADSCs with TGF-β and PRP results in considerable GAG production (Fig. 2). Besides, based on the intensity of the deposited Alcian blue stain, one can tell that PRP induces GAG production as well as TGF-β.

During embryogenesis, chondroblasts synthesize collagen type II and proteoglycans while post-mitotic hypertrophic chondrocytes express collagen type X and matrix metalloproteinase.⁸ Both TGF-β and PRP induce major Col-II deposition in ADSCs’ ECM while inducing moderate Col-X production (Fig. 3), which is in agreement with previous reports.⁸

What we learn from the gene expression profile?

The SOX9 as a key transcription factor involved in the chondrogenesis inhibits ECM degradation and enhances the chondrogenesis process.^16,17 After 14 days (late stage of chondrogenesis), both of chondroprotective markers (Col II and SOX9) significantly upregulated after the application of either TGF-β or PRP to the culture media (Fig. 4). The ICC assay also confirmed that both TGF-β and PRP equally promoted chondrogenesis and ECM stabilization.

What can we learn from variation of miRNA expression profile?

Unlike various transcription factors and cytokines that control multiple phases of chondrogenic differentiation, a few miRNAs intermediate chondrogenic differentiation of stem cells.¹⁶ The miR-146 is confirmed to be expressed in osteoarthritis (OA) cartilage in non-physiological conditions which comes up with cartilage lesions, inflammation, and neoangiogenesis.¹⁸ In fact, miR-146a and miR-199a promote the degradation of the cartilage matrix.¹⁹ At the early stage of chondrogenesis (3rd day), PRP resulted in overexpression of miR-146 compared to TGF-β (Fig. 5), which remained upregulated until late chondrogenesis (14th day).

Early or late co-expression of miR-101 with miR-145 removes suppression of MSC condensation, chondrocytes proliferation, and differentiation.^20,21 The miR-140 upregulation is positively correlated with SOX9 overexpression,²² which in turn controls chondrocyte proliferation.^23,24 Upregulation of miR-199 promotes degradation of cartilage matrix as well as miR-146.¹⁹ The PRP considerably induced upregulation of miR-101, 140, and 145 in late chondrogenesis (14th day), compared to the TGF-β (Fig. 5). Although upregulation of miR-101, 140, and 145 promotes chondroprotective effects and inhibits inflammatory side effects, miR-199 and miR-146 exert anti-chondrogenic roles by targeting SMAD1 and SMAD2/3, respectively.²⁵ Therefore, at early chondrogenesis, PRP stimulates angiogenesis and ECM neogenesis via the upregulation of miR-146 while at late chondrogenesis, facilitates ADSCs differentiation to mature chondrocytes via upregulation of miR-101, 140, and 145, as well as tuning their stemness ability.

What do ALP and calcium deposition assays reveal?

Increment of either alkaline phosphatase (ALP) activity or calcium deposition in the ECM results in hypertrophy, matrix degradation, and osteogenesis.⁸ Although enhanced ALP activity was observed at late chondrogenesis, following the application of PRP over TGF-β (Fig. 6A), calcium deposition eliminated at late chondrogenesis (Fig. 6B).

How the applied treatments affect neovascularization and TNFα secretion?

The presence of PRP in ADSC differentiating media instead of TGF-β for 14 days, considerably suppressed VEGF secretion (Fig. 6C). Hence, PRP guarantees ADSC fate to chondrogenesis rather than ossification.

All of the chondrocytes secret TNFα at the basal level.²⁶ However, the IL-1β boosts TNFα production. TNF‐α mediates both chondrocyte apoptosis and expression of endochondral resorption mediated cytokines. Application of TGF-β triggered secretion of TNFα while the addition of PRP to the media suppressed it (Fig. 6D).

The miR-146a does not target TNFα directly but up-regulation of this miRNA results in the downregulation of stimuli-mediated TNFα.²⁶ Our results are in line with this fact that higher miR-146a expression is accompanied with the diminished secretion of TNFα in PRP groups.