Comparison of Human Platelet Lysate versus Fetal Bovine Serum for Expansion of Human Articular Cartilage–Derived Chondroprogenitors

Abstract

Purpose

Articular chondroprogenitors, a suitable contender for cell-based therapy in cartilage repair, routinely employ fetal bovine serum (FBS) for expansion and differentiation. The possibility of transplant rejections or zoonoses transmissions raise a need for xeno-free alternatives. Use of human platelet lysate (hPL), a nutrient supplement abundant in growth factors, has not been reported for human chondroprogenitor expansion thus far. Our aim was to compare the biological profile of chondroprogenitors grown in hPL versus FBS.

Methods

Chondroprogenitors were isolated from 3 osteoarthritic knee joints. Following differential fibronectin adhesion assay, passage 0 cells grown in (a) 10% FBS and (b) 10% hPL were considered for assessment of growth kinetics, surface marker expression, gene expression, and trilineage differentiation. Latent transforming growth factor–β1 (TGFβ1) levels were also measured for each culture medium used.

Results

Cellular proliferation was significantly higher in cells grown with hPL (P < 0.01). Surface marker expression was comparable except in CD-146 where hPL group had significantly higher values (P = 0.03). Comparison of mRNA expression revealed notably low values of collagen I, collagen X, aggrecan, and collagen II (P < 0.05). Trilineage differentiation was seen in both groups with higher alizarin red uptake noted in hPL. There were also significantly higher levels of latent TGFβ1 in the medium containing hPL as compared to FBS.

Conclusions

This is the first in vitro xeno-free study to affirm that hPL can serve as an optimal growth supplement for expansion of articular chondroprogenitors, although an in-depth assessment of resident growth factors and evaluation of different dilutions of hPL is required to assess suitability for use in translational research.

Introduction

Articular cartilage architecture is primarily composed of chondrocytes and chondroprogenitors embedded in extracellular matrix, providing a smooth surface, which aids in joint mobility.^1,2 Constant wear and tear compounded by avascularity may lead to damage, which when complicated by trauma or degenerative changes (as seen in osteoarthritis [OA]), require improved treatment strategies. Current practices in cell-based therapy for cartilage regeneration mainly involve the use of mesenchymal stem cells (MSCs) and chondrocytes. In many cases, repair seen with these cells is not ideal due to the formation of either fibrocartilage or a combination of fibro-hyaline cartilage.^3-5 This leaves a gap in management that may be bridged by chondroprogenitors, a cell type characterized extensively in the past decade for its application in cell-based therapy, due to its propensity for chondrogenesis and reduced hypertrophy, a combination best suited for generation of hyaline like cartilage. ⁶

Articular cartilage–derived chondroprogenitors exhibit stem cell–like characteristics and have been classified as an MSC (as per minimal International Society for Cellular Therapy criteria, 2006). They are isolated currently by employing differential adhesion assays,^2,7-10 explant cultures (migratory subpopulations) ¹¹ or sorting studies.^12,13 Standard expansion and characterization of these cells routinely involves the use of fetal bovine serum (FBS) as a supplement to stimulate growth and proliferation. ¹⁴ However, xenogeneic sera such as FBS could increase risk of transmission of zoonoses and constituent antigenic substrates could induce possible immunogenic reactions.^15,16 This necessitates use of xeno-free or human-derived alternatives to FBS for accentuation of growth and differentiation of chondroprogenitors.

Human platelet lysate (hPL) is a supplement derived from platelet-rich plasma, with extensive applications in experimental and clinical culture. ¹⁷ It is widely recognized that hPL can serve as a high-quality media supplement, due to constituents such as platelet-derived growth factor (PDGF), transforming growth factor (TGF), vascular endothelial growth factor (VEGF) among others, which aid in collagen synthesis, angiogenesis, and MSC recruitment and facilitating tissue regeneration.^18-21 hPL has been successfully employed in numerous studies for expansion of MSC and chondrocytes to overcome limitations of FBS use successfully, but there are no current reports suggesting its use for cartilage-derived chondroprogenitors. Thus, the aim of this study was to compare hPL versus FBS for expansion of human articular cartilage–derived chondroprogenitors. The assessment parameters undertaken for comparison included growth kinetics, senescence assay, surface marker expression, mRNA expression, and trilineage differentiation.

Materials and Methods

Study Design

The study was conducted after obtaining approval from the institutional review board, in accordance with the ethics committee guidelines. Prior to harvest, written informed consent was obtained. Human knee joints containing the cartilage were obtained from three subjects (mean age: 50±8.7 years) diagnosed with OA, requiring total knee replacement for their treatment. Our inclusion criteria comprised patients diagnosed with primary degenerative OA with a Kellgren-Lawrence radiological grade of 4. Patients requiring knee replacement following an infective, inflammatory, systemic pathology, secondary to metabolic diseases or posttrauma were excluded. Cartilage cells were isolated from articular cartilage and subjected to fibronectin adhesion assay in order to obtain chondroprogenitors. ⁷ Comparative assessment of growth kinetics parameters included cumulative population doubling up to passage 3 (CPD), cell cycle analysis, and galactosidase senescence assay. Flow cytometric assessment included positive MSC markers:CD105, CD73, CD90, negative MSC markers: CD34, CD45, and HLA-DR and markers of enhanced chondrogenic potential: CD146, CD166, CD49e and CD29.^2,13,22 mRNA expression was compared based on chondrogenesis markers (collagen type II, aggrecan, and SOX9), hypertrophic chondrocyte marker (collagen type I), and hypertrophy markers (RUNX2, MMP-13, and collagen type X). Potential for trilineage differentiation was also evaluated by comparing their adipogenic, osteogenic, and chondrogenic capacity ( Fig. 1 ).

Figure 1.

Algorithm depicting sequence of cellular isolation, culture condition using either FBS or HPL and its subsequent evaluation parameters. FBS, fetal bovine serum; hPL, human platelet lysate.

Preparation of hPL

hPL was prepared as per protocol described by Kølle et al. with minor modifications. ²³ Outdated therapeutic grade platelet-rich concentrate (PRC) prepared from O blood group donors, was collected and stored at −80°C, thawed at 37°C, and frozen again at −80°C. This freeze-thaw cycle was repeated 5 times to obtain the lysate. This was pooled, triturated, transferred into 50-mL tubes, and centrifuged at 2600 × g for 25 minutes at room temperature. The supernatant was carefully removed and recentrifugation (2600 × g for 25 minutes) was done to avoid contamination with platelet fragments. The obtained supernatant was stored at −20°C after addition of 0.1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) and 2.0 IU/mL heparin.

Isolation and Culture of Chondroprogenitors

Full depth cartilage shavings were harvested (nonweightbearing areas), minced, and washed in Dulbecco’s Modified Eagle/Nutrient Mixture F12 Ham (DMEM/F12: Himedia) medium. Cartilage slices were subjected to sequential enzymatic digestion using Pronase (Roche: 12 IU/mL) for 180 minutes followed by collagenase type II (Worthington: 100 IU/mL) for 16 hours. The released cells were loaded at a concentration of 4000 cells/well (9.3 cm²) on precoated fibronectin plates (Sigma: 10 µL/mL containing 1 mM MgCl₂ and 1 mM CaCl₂) for 20 minutes. The components of the standard growth medium included Glutamax DMEM-F12, 0.1 mM ascorbic acid, 10 mM HEPES, and penicillin-streptomycin (Gibco: 100 IU/mL) and amphotericin-B (Gibco:2 μg/mL). Postincubation, the nonadherent cells were removed, and the wells from respective groups were replaced with standard growth medium containing either 10% FBS or 10% hPL. The adherent cells were cultured for a period of 12 days till they reached a growth of about 32 cells/clone (5 population doublings). The enriched chondroprogenitors were further expanded in monolayer cultures till sub confluence in standard medium additionally containing transforming growth factor–beta 2 (TGFβ2: 1 ng/ml) and fibroblast growth factor–2 (FGF2: 5 ng/mL), with medium change every 3 days.

Growth Kinetics

Cumulative Population Doubling

Chondroprogenitors grown with either 10% FBS or 10% hPL were expanded in monolayer cultures at a seeding density of 5000 cells/cm² when 85% to 90% confluence was reached. Cell count by tryphan blue exclusion technique was performed using an automated counter (Countess, Invitrogen) and population doubling time (PDT) for the groups, was calculated using the formula:

$$\mathrm{PD}=\frac{\log (N)-\log (N_0)}{0.301}$$

where N₀ is the initial number of cells seeded (day 0) and N is the number of cells at confluence. Cumulative population doubling (CPD) was compared from p1 to p4.

Cell Cycle Analysis

The percentage of cells in different phases of cell cycle was analyzed using DAPI (4′,6-diamidino-2-phenylindole). Cells were trypsinized on reaching 70% confluence. Passage 1 cells were harvested, washed, and fixed with ice cold 70% ethanol for 48 hours. DAPI (1 µg in 200 µL of phosphate buffered saline [PBS]) was used for incubation for a duration of 30minutes. After a wash, ice-cold PBS was used to resuspend the cells for flow analysis. Flo-Jo software was used to analyze the percentage of cells in various phases using the Watson algorithm.

Senescence Assay: Galactosidase Assay

Chondroprogenitors grown in FBS and hPL culture conditions were comparatively analyzed for presence of presenescent and senescent cells at the end of passage 1. In brief, the cells were cultured for 5 days (5,000 cells/cm²) and were fixed and stained using β-galactosidase senescence staining kit (Biovision), as per manufacturer’s instructions. The fixed cells were incubated with the staining solution overnight and observed for stain uptake.

Quantitative Cytokine Array: Latent human TGFβ1 levels

Standard culture media containing either 10% FBS or 10% hPL were analyzed for latent human TFGβ1 levels in duplicates using a commercially available kit (ELISA development kit, Mabtech) according to manufacturer’s instruction. Values were denoted in ng/mL.

Fluorescence-Activated Cell Sorting

The presence of MSC positive, MSC negative, and markers of enhanced chondrogenesis were characterized by fluorescence-activated cell sorting (FACS). Passage 1 chondroprogenitors cultured in the 2 conditions were studied against the following human surface antigen: positive MSC markers—CD105-FITC (fluorescein isothiocyanate), CD73-PE, and CD90-PE (phycoerythrin); negative MSC markers—CD34-PE, CD45-FITC; and markers of enhanced chondrogenesis—CD29-APC (allophycocyanin), CD49e-PE, CD166-VioBright FITC, and CD146-PE (Supplementary Table S1). The staining method followed protocols provided for individual antibodies. BD FACS ARIA III flow cytometer was used for data acquisition. Gating was applied, and isotype conjugate controls were run and analyzed using BD FACS Diva v 5.0.2 software.

Reverse Transcription–Polymerase Chain Reaction

Total RNA was isolated from passage 1 chondroprogenitors using TRIzol reagent (Sigma) and quantified using nano spectrophotometry, as per manufacturer’s instructions. RNA was reverse transcribed to cDNA using Primescript first strand cDNA synthesis kit (Takara-bio). Using QuantStudio 6K Flex (Applied Biosystem) quantitative reverse transcription–polymerase chain reaction (10 ng/10 µL reaction) was performed with Takyon Rox SYBR Master Mix (Eurogenetec). Relative expression for each gene (ΔCt) was normalized to the reference gene (GAPDH). Sequences of the primers used are listed in Supplementary Table S2.

Trilineage Differentiation Potential

Chondroprogenitors from the hPL and FBS groups were assessed for multilineage differentiation potential. Potential for adipogenesis and osteogenesis was evaluated in monolayer cultures where chondroprogenitors were cultured to 50% confluence prior to differentiation. Chondroprogenitors in standard expansion culture medium were also run as negative controls. Half a million cells were centrifuged and left for 48 hours to stabilize and form a 3-dimensional pellet prior to chondrogenic differentiation. Complete differentiation into adipocytes, osteocytes and chondrocytes was performed using StemPro (Thermo Fischer Scientific) adipogenesis, osteogenesis, and chondrogenesis differentiation kits. The medium was refreshed every 3 days for a period of 21 days and differential staining was performed.

Differential Stains

Oil Red O and Alizarin Red

To order to confirm differentiation, the adipogenic differentiated cells were fixed with 10% buffered formalin for 60 minutes, washed, and stained with oil red O (Sigma) and the osteogenic differentiated cells were fixed with 70% ethanol for 60 minutes, washed, and stained with alizarin red (Sigma). The control cells were also stained with oil red O and alizarin red. Images were captured using Olympus virtual slide microscope.

Histological Staining: Alcian Blue and Collagen II

To visualize glycosaminoglycan deposition, differentiated progenitors from monolayer cultures and chondrogenic pellets were subjected to a short fixation, stained with Alcian blue and counterstained with neutral red. Pellets were also subjected to fixation, paraffin embedding, antigen retrieval using pronase and hyaluronidase and further analyzed for collagen type II (5 µg/mL, mouse monoclonal antibody; DSHB II-II6B3) by immunohistochemistry using chromogen 3,3′-diaminobenzene and counterstained with hematoxylin. ¹⁵

Statistical Analysis

Offline data analysis was performed using Microsoft Excel and graphical data representation by IGOR Pro Version 5.0.4.8 (Wave Metrics Inc.). Statistical Package for Social Sciences (SPSS) version 21.0 software was used for statistical analysis. Comparison of groups was done using unpaired Student t test. Data were expressed as mean ± SD. A P value <0.05 was considered significant.

Results

Growth in Culture

Chondroprogenitors maintained clonality in FBS and hPL on fibronectin-coated plates up to 7 days, following which cells in the hPL group exhibited honeycomb like morphology, which was maintained even on monolayer expansion ( Fig. 2 ). Cells grown in hPL showed increased ease of detachment with 0.25% trypsin-EDTA requiring only 30 seconds as compared with standard 4 minutes for the other group. Delayed attachment (≃3 days) following trypsinization was also observed with the hPL group.

Figure 2.

Representative clonally derived human articular CPs following FAA cultured with FBS/hPL. (A, E): A clone at day 3 forming a cluster of 2 to 3 cells (10×). (B, F): Clone cluster at day 7 (20×). (C) Clonal growth >5 population doubling. (G): Nonclonal growth at day 11 (10×). (D, H): Passage 0 cultures with honeycomb growth pattern observed with CPs grown in hPL (10×). CP, chondroprogenitor; FAA, fibronectin adhesion assay; FBS, fetal bovine serum; hPL, human platelet lysate.

Growth Kinetics

Cumulative Population Doubling

CPD was assessed from cells of passage (P) 0 to P3 with an average time in culture for each passage calculated as 7 days. It was noted that at each passage CPD was significantly higher in the hPL group as compared with the FBS group (P1-hPL: 3.5 ± 0.1 and FBS: 0.9 ± 0.2 with P < 0.001; P2-hPL: 7.4 ± 0.4 and FBS: 2.7 ± 0.2 with P < 0.001; P3-hPL: 11 ± 0.2 and FBS: 5 ± 4 with P < 0.001 and P4-hPL: 13.6 ± 1.7 and FBS: 6.1 ± 0.5 with P = 0.004) ( Fig. 3 ).

Figure 3.

Data representing cumulative population doubling (CPD) values for 2 subgroups (n = 3 human samples) across time in culture (p0-p3). Data expressed as mean ± SD. FBS, fetal bovine serum; hPL, human platelet lysate. (*P < 0.05, **P < 0.01, ***P < 0.001).

Cell Cycle Analysis

Cell cycle analysis was performed with day 3 P1 cells. Results demonstrated that cells grown in hPL had a higher subset seen in S phase (30.97% ± 2.05%), G2M phase (5.93% ± 1.17%), and a lower number in the G1 phase (59.47% ± 2.74%) as compared with FBS (S, 24.17% ± 6.52%; G2M, 2.54% ± 1.81%; and G1, 73% ± 4.35%), although not significant ( Fig. 4 ).

Figure 4.

CP cell cycle analysis with DAPI using Watson model. (A, B). Representative FACS plots of phase distribution. (C, D) Pie chart depicting phase distribution of mean scores (n = 3 human samples). CO, chondroprogenitor; FBS, fetal bovine serum; hPL, human platelet lysate.

Galactosidase Staining

Senescence assay showed comparable results in both groups with the minimal uptake of β-galactosidase ( Fig. 5 ).

Figure 5.

β-Galactosidase assay comparing the number of presenescent and senescent cells between the chondroprogenitors grown with FBS and hPL. Positive uptake indicated light blue stain. Magnification: 10×. FBS, fetal bovine serum; hPL, human platelet lysate.

Latent Human TGFβ1 levels

It was seen that latent TGFβ1 levels were notably higher in the media containing 10% hPL (7.32 ng/mL) than media containing 10% FBS (0.30 ng/mL).

Fluorescence-Activated Cell Sorting

FACS was performed at P1 and analysis revealed comparable expression in positive and negative MSC markers, that is, both groups exhibited high expression of CD105, CD73, and CD90 (positive MSC markers) and minimal expression of CD34, CD45, and HLA-DR (negative MSC markers). When markers of enhanced chondrogenesis were compared both groups showed high expression except for CD146, where hPL group exhibited significantly higher levels than FBS group (hPL, 68.7 ± 3.4; FBS, 49.6 ± 7.7; P = 0.033) ( Fig. 6 ).

Figure 6.

Comparison of percentage expression of CD markers between FBS and hPL at passage 1 (n = 3 human samples). Data expressed as mean ± SD. *P < 0.05. FBS, fetal bovine serum; hPL, human platelet lysate.

Reverse Transcription–Polymerase Chain Reaction

Comparison of gene expression was performed at P1, which demonstrated high expression of SOX9, aggrecan, and collagen type I, moderate expression of MMP13 and collagen type X followed by low expression of collagen type II and RUNX2 in both groups ( Fig. 7 ). hPL group displayed significantly lower values than FBS for aggrecan (P = 0.005) and collagen type II (P = 0.025), known markers of chondrogenesis and also for collagen type X (P = 0.022) and collagen type I (P = 0.005), known markers of hypertrophy and hypertrophic chondrocyte, respectively.

Figure 7.

Relative expression of SOX9, collagen type II, aggrecan, collagen type I, RUNX2, MMP-13, and collagen type X between FBS and hPL groups. ΔCt values normalized to GAPDH are expressed as mean ± SD (*P < 0.05, **P < 0.01). FBS, fetal bovine serum; hPL, human platelet lysate.

Trilineage Potential

Both hPL and FBS groups displayed multilineage differentiation potential. Qualitative analysis for osteogenic potential showed that hPL group exhibited more calcified matrix deposition as evident by higher uptake of alizarin red ( Fig. 8D and J ). Regarding chondrogenic differentiation, when estimated by both groups showed comparable glycosaminoglycan deposition as evident by Alcian blue uptake although intensity for staining of collagen type II staining was comparatively higher in the FBS group ( Fig. 8F and L ).

Figure 8.

Trilineage differentiation of chondroprogenitors from osteoarthritic joints between FBS and hPL groups. Representative microscopic images of oil red O (A-B, G-H) and Alizarin red (C-D, I-J) staining to confirm adipogenic and osteogenic differentiation (10×). Chondrogenic differentiation was confirmed by Alcian blue (E, K) and collagen II (F, L) staining of formed cell pellets (10×).

Discussion

Due to the increased interest in chondroprogenitors owing to their properties of inherent chondrogenesis and reduced hypertrophy as seen in various reports, this cell type may be a successful contender in cell-based therapy for cartilage-related pathologies.^11,24,25 Since chondroprogenitors have shown promise in studies performed on animal models and have been successfully labeled for implantation, translational studies employing these cells are the next endeavor.^6,26 Most studies with reported evidence employ the use of FBS as a culture additive, which stands as a point of concern, since it has been demonstrated that its use can lead to antigenic reactions, which can be a hindrance to translational therapy. In order to accomplish clinical grade expansion, xenofree alternatives to routinely used sera are required.

hPL is a relatively cost-efficient alternative, which may also be used in an autologous setting for cell proliferation and is uniquely suited for chondroprogenitors due to its high growth factor content, namely TGFβ and PDGF, important signaling molecules involved in the sequence for chondrogenesis. ²⁷ Interestingly, in this study we found that latent TGFβ1 levels in medium containing hPL were almost 14 times higher than that of medium containing FBS (Supplementary Fig. S1). Several studies have reported use of hPL for growth and differentiation of bone marrow MSC and chondrocytes among other prominent cell lines.^20,28,29 This was the first in vitro study to assess suitability of hPL as an alternative to FBS for chondroprogenitor expansion and influence on phenotype. In accordance to previous reports, our study demonstrated significantly high proliferative capacity and comparable viability parameters in hPL group as opposed to the FBS group. Evaluation of viability and senescent cells was also crucial, since growth rate seen with hPL was significantly high and data pertaining to cell phenotype in early passages will be useful for translational research.^15,16 It is of importance to note that the dilution of hPL used was 10%, comparable to that of FBS and the results seen could be attributed to this concentration, although further evaluation of different dilutions is required to establish dose dependence.

We also observed that hPL did not compromise the stemness of chondroprogenitors as was evident by its retention of specific MSC markers. Moreover, significantly high level of CD146, a known marker of enhanced chondrogenesis was also observed with cells grown in hPL.^13,22,30 Analysis of gene expression showed high levels of SOX9 and aggrecan, although aggrecan levels were seen to be higher in the FBS group. Results also revealed low levels of hypertrophy in hPL group suggesting that certain constituents may prevent terminal maturation. This finding is of value since fibrocartilage formation is a hinderance in long-term management following cell-based therapy. Retention of multipotency was reiterated when results of trilineage differentiation were examined. hPL exhibited comparable adipogenic potential in addition to showing notable higher deposition of calcified matrix when differentiated toward an osteogenic lineage. Chondrogenic potential corroborated with the gene expression profile where the glycosaminoglycan deposition was seen to be similar, whereas collagen type II was of a relatively lower intensity as compared with cells grown in FBS. Since nutrient supplement requirement is implicit for cell growth and expansion and use of FBS and similar supplements raise concerns regarding transmission of possible zoonoses and precipitation of immunogenic response, this study proposes hPL as an optimal xeno-free alternative for chondroprogenitors in preclinical and clinical evaluation.

Conclusion

This is the first in vitro study to establish xeno-free conditions for expansion of human articular cartilage derived chondroprogenitors and establishes hPL as an effective growth supplement for characterization of these cells when compared with FBS. Since chondroprogenitors maintain comparable MSC characteristics and differentiation potential in addition to showing significantly higher proliferative rate, consideration of hPL as an efficient candidate for supplementation needs to be explored; although standardization of concentration used for optimal growth and differentiation, in-depth analysis of the specific growth factors present in hPL and their individual effects on chondrogenesis, are recommendations that warrant further evaluation.