Here, we propose a protocol for chondrogenic differentiation from cord blood mononuclear cell-derived human induced pluripotent stem cells.
Human articular cartilage lacks the ability to repair itself. Cartilage degeneration is thus treated not by curative but by conservative treatments. Currently, efforts are being made to regenerate damaged cartilage with ex vivo expanded chondrocytes or bone marrow-derived mesenchymal stem cells (BMSCs). However, the restricted viability and instability of these cells limit their application in cartilage reconstruction. Human induced pluripotent stem cells (hiPSCs) have received scientific attention as a new alternative for regenerative applications. With unlimited self-renewal ability and multipotency, hiPSCs have been highlighted as a new replacement cell source for cartilage repair. However, obtaining a high quantity of high-quality chondrogenic pellets is a major challenge to their clinical application. In this study, we used embryoid body (EB)-derived outgrowth cells for chondrogenic differentiation. Successful chondrogenesis was confirmed by PCR and staining with alcian blue, toluidine blue, and antibodies against collagen types I and II (COL1A1 and COL2A1, respectively). We provide a detailed method for the differentiation of cord blood mononuclear cell-derived iPSCs (CBMC-hiPSCs) into chondrogenic pellets.
The use of hiPSCs represents a new strategy for drug screening and mechanistic studies of various diseases. From a regenerative perspective, hiPSCs are also a potential source for the replacement of damaged tissues that have limited healing ability, such as articular cartilage1,2.
The regeneration of native articular cartilage has been a challenge for several decades. Articular cartilage is a soft, white tissue that coats the end of bones, protecting them from friction. However, it has limited regenerative ability when damaged, which makes self-repair almost impossible. Therefore, research focused on cartilage regeneration has been ongoing for several decades.
Previously, in vitro differentiation into the chondrogenic lineage was usually performed with BMSCs or native chondrocytes isolated from the knee joint3. Due to their chondrogenic potential, BMSCs and native chondrocytes have numerous merits supporting their use in chondrogenesis. However, because of their limited expansion and unstable phenotype, these cells face several limitations in the reconstruction of articular cartilage defects. Under in vitro culture conditions, these cells tend to lose their own characteristics after 3-4 passages, which eventually affects their differentiation abilities4. Also, in the case of native chondrocytes, additional damage to the knee joint is inevitable when obtaining these cells.
Unlike BMSCs or native chondrocytes, hiPSCs can indefinitely expand in vitro. With the proper culture conditions, hiPSCs have great potential as a replacement source for chondrogenic differentiation. However, it is challenging to change the intrinsic characteristics of hiPSCs5. Moreover, it takes several complicated in vitro steps to direct the fate of hiPSCs to a specific cell type. Despite these complications, the use of hiPSCs is still recommended due to their high self-renewal abilities and their capacity to differentiate into targeted cells, including chondrocytes6.
Chondrogenic differentiation is usually done with three-dimensional culture systems, such as the pellet culture or micromass culture, using MSC-like progenitor cells. If using hiPSCs, the protocol to generate MSC-like progenitor cells differs from the existing protocols. Some groups use monolayer culture of hiPSCs to directly convert the phenotype into MSC-like cells7. However, most studies use EBs to generate outgrowth cells that resemble MSCs8,9,10,11.
Various types of growth factors are used to induce chondrogenesis. Usually, BMP and TGFβ family proteins are used, alone or in combination. Differentiation has also been induced with other factors, such as GDF5, FGF2, and IGF112,13,14,15. TGFβ1 has been shown to stimulate chondrogenesis in a dose-dependent manner in MSCs16. Compared to the other isotype, TGFβ3, TGFβ1 induces chondrogenesis by increasing the pre-cartilage mesenchymal cell condensation.TGFβ3 induces chondrogenesis by significantly increasing the mesenchymal cell proliferation17. However, TGFβ3 is used more frequently by researchers than TGFβ17,10,18,19. BMP2 enhances the expression of genes related to the chondrogenic matrix components in human articular chondrocytes under in vitro conditions20. BMP2 increases the expression of genes critical to cartilage formation in MSCs in combination with TGFβ proteins21. It has also been shown that BMP2 synergistically enhances the effect of TGFβ3 through the Smad and MAPK pathways22.
In this study, CBMC-hiPSCs were aggregated into EBs using EB medium in a low-attachment Petri dish. Outgrowth cells were induced by attaching the EBs to a gelatin-coated dish. Chondrogenic differentiation using outgrowth cells was performed by pellet culture. Treatment with both BMP2 and TGFβ3 successfully condensed the cells and induced extracellular matrix (ECM) protein accumulation for chondrogenic pellet formation. This study suggests a simple yet efficient chondrogenic differentiation protocol using CBMC-hiPSCs.
This protocol was approved by the institutional review board of the Catholic University of Korea (KC12TISI0861). CBMCs used for reprogramming were directly obtained from the Cord Blood Bank of the Seoul St. Mary's Hospital.
1. Chondrogenic Differentiation from iPSCs
2. Chondrogenic Pellet Characterization by Staining
In this study, we generated chondrogenic pellets from CBMC-hiPSCs by inducing outgrowth cells from EBs. Chondrogenic differentiation was induced using CBMC-hiPSCs with confirmed high pluripotency11. A simple scheme of our protocol is shown in Figure 1A. Before differentiation, iPSC colonies were expanded (Figure 1B). The expanded iPSCs were assembled as EBs to initiate differentiation (Figure 1C). The generated EBs were attached to gelatin-coated dishes to induce outgrowth cells (Figure 1D). Then, the outgrowth cells were harvested and used to generate chondrogenic pellets. After 21 days of differentiation, small, bead-like chondrogenic pellets were obtained and used for further characterization (Figure 1E). The quality of the in vitro-generated chondrogenic pellets was confirmed through various assays.
We histologically evaluated the quality of the chondrogenic pellets on day 21. BMSC chondrogenic pellets were used as the positive control. The accumulation of ECM proteins secreted by the differentiated chondrocytes was confirmed by alcian blue and toluidine blue staining in Figure 2A. The major characteristics of healthy cartilage, such as lacuna and proteoglycan production, increased as the differentiation process continued over 21 days. Chondrogenic pellets generated from CBMC-hiPSCs expressed COL2A1, which is the major ECM component in healthy cartilage (Figure 2B). The COL2A1 expression of CBMC-hiPSC-derived pellets was higher than that of BMSC-derived pellets. The expression of collagen type I, a marker for fibrotic cartilage, was lower in CBMC-hiPSC-derived pellets compared to the expression of COL2A1.
The gene expression of cartilage ECM proteins in day-21 chondrogenic pellets was confirmed by real-time PCR. The aggrecan (ACAN) expression of CBMC-hiPSC-derived pellets was similar to that of BMSC-derived pellets (Figure 3A). The expression of COL2A1 was significantly higher in CBMC-hiPSC-derived pellets (Figure 3B). The expression of sex-determining region Y-box 9 (Sox9), a chondrogenic progenitor marker, was also evaluated. Pellets generated from CBMC-hiPSCs expressed high levels of Sox9 (Figure 3C). We confirmed that these genes were significantly upregulated in CBMC-hiPSC chondrogenic pellets compared to BMSC control pellets on day 21. The expression of a hypertrophic marker, COL1A1, was evaluated (Figure 3D). The expression of COL1A1 in CBMC-hiPSC-derived pellets was reduced compared to that in BMSC-derived pellets. The differentiation efficiency was analyzed by the ratio of COL2A1 to COL1A1 (Figure 3E). The increment of the ratio in CBMC-hiPSC-derived pellets demonstrated the relatively high expression of COL2A1 compared to COL1A1. In conclusion, we have confirmed the chondrogenic differentiation capacity of CBMC-hiPSCs. The quality of the generated CBMC-hiPSC-derived chondrogenic pellets had a quality compatible with that of BMSCs.
Figure 1: Chondrogenic differentiation of hiPSCs. (A) Scheme of chondrogenic pellet generation from hiPSCs. (B) Morphology of the generated CBMC-hiPSC. (C) Morphology of generated EBs. (D) Outgrowth cells derived from EBs attached to a gelatin-coated culture dish. (E) Image of the chondrogenic pellet after 21 days of differentiation. The units of the intervals are shown in millimeters. Scale bars = 200 µm. Please click here to view a larger version of this figure.
Figure 2: Histological analysis of the chondrogenic pellet. (A) Histological evaluation of chondrogenic pellets on day 21 using alcian blue and toluidine blue staining. (B) Immunohistochemistry image of chondrogenic pellets stained with COL1A1 and COL2A1 antibodies. Scale bars = 100 µm. Please click here to view a larger version of this figure.
Figure 3: Gene expression of the chondrogenic pellet. (A) Relative gene expression of ACAN. (B) Relative gene expression of COL2A1. (C) Relative gene expression of SOX9. (D) Relative gene expression of COL1A1. (E) Gene expression of COL1A1. (F) Gene expression of COL10A1. (G) The ratio of COL2A1:COL1A1 gene expression. Pellets were harvested and analyzed after 21 days of differentiation. Data were obtained using real-time PCR and displayed as the mean standard error of triplicate experiments per sample (n = 3). The gene expression was normalized to GAPDH as an internal control. (*p <0.05, **p <0.01, ***p <0.001). Please click here to view a larger version of this figure.
Cell number | Outgrowth cell yield | Pellet yield | |
hiPSCs | 2 x 106 | 2-5 x 107 | 70-150 |
MSC | 2 x 106 | – | 6 |
Table 1: Yield comparison of MSC and hiPSCs.
Target Name | Direction | Primer Sequence | Size |
SOX9 | Forward | TTCCGCGACGTGGACAT | 77 bp |
Reverse | TCAAACTCGTTGACATCGAAGGT | ||
ACAN | Forward | AGCCTGCGCTCCAATGACT | 107 bp |
Reverse | TAATGGAACACGATGCCTTTCA | ||
COL2A1 | Forward | GGCAATAGCAGGTTCACGTACA | 79 bp |
Reverse | CGATAACAGTCTTGCCCCACTTA | ||
COL1A1 | Forward | CCCCTGGAAAGAATGGAGATG | 148 bp |
Reverse | TCCAAACCACTGAAACCTCTG |
Table 2: Sequences of primers against chondrogenic markers in real-time PCR.
This protocol successfully generated hiPSCs from CBMCs. We reprogrammed CBMCs to hiPSCs using a Sendai viral vector containing Yamanaka factors24. Three cases were used in differentiation, and all experiments successfully generated chondrogenic pellets using this protocol. Numerous studies have reported protocols for the differentiation of hiPSCs into chondrocytes25,26,27,28. However, additional research is required to confirm the use of CBMC-hiPSCs a candidate for cartilage regeneration and recovery.
Chondrogenesis was confirmed using hiPSCs generated from various somatic cell types11,25,26,27,29. Many reports demonstrate that the origin of the somatic cell used in reprogramming can affect the differentiation outcome of the hiPSCs30,31,32,33. Cord blood is a source enriched with HSCs and MSCs. There is no report on the difference between cord blood-derived cells, such as blood cells or MSCs. However, previous studies have shown that articular chondrocyte-derived hiPSCs are more likely to go through chondrogenesis than hiPSCs generated from cord blood or skin fibroblasts29. Results from endothelial differentiation using hiPSCs generated from fibroblasts, cardiac progenitor cells, and endothelial cells demonstrated that the tissue-of-origin can reflect the tissue-specific somatic memory in terminally differentiated iPSC derivation, especially at early passages, between 10 and 2034. In early passages of hiPSCs, endothelial cell-derived hiPSCs differentiated more efficiently into the endothelial lineage. However, the significant difference between these cell lines disappeared in hiPSCs over 20 passages. Therefore, it is important to carefully consider the somatic memory carried by the hiPSCs in early passages when they are used in clinical research.
Recently, various procedures have been developed to repair defected articular cartilage. Microfracture is done by drilling multiple holes in the bone to induce the influx of bone marrow for natural repair35. Knee replacement, also known as knee arthroplasty, is used to replace the damaged cartilage. However, microfracture is not useful for severe articular cartilage damage, and the instrument used for knee replacements must be changed every 10-15 years.
These days, cell-based therapies are a promising alternative method for cartilage repair. Autologous chondrocyte implantation (ACI) is widely used for cell-based cartilage recovery. ACI is done by directly injecting autologous chondrocytes into the defect. However, autologous chondrocytes turn into fibroblast-like cells under in vitro cultivation conditions. Additional damage is also inevitable during the harvesting process of autologous chondrocytes. MSCs have been suggested as a cell source for cartilage recovery. Cord blood is a useful and accessible cell source of MSCs for engineering cartilage36. The success rate for isolating cord-blood MSCs, however, is controversial37,38. Numerous studies report the successful isolation of cord-blood MSCs. Several studies have shown that cord blood volume is a critical parameter that can affect the yield rate of MSC isolation36,39. To acquire high-quality MSCs, the passage of the cells is also critical. Previous studies indicate that MSCs have a restricted life span after a certain cell division number. By entering senescence, MSCs are characterized by decreased proliferation, as with any normal somatic cell40. Therefore, cord blood-derived MSCs must be used before the sixth passage to avoid cell senescence and chromosomal abnormalities41.
To avoid these limitations, human iPSCs opened a new possibility for personalized, cell-based therapy with high productivity11. Established hiPSCs can theoretically proliferate limitlessly. Also, with the same immune identity as the donor, they can avoid rejection and lower side effects when implanted in vivo. The transition of cord blood banks into hiPSC banks for allogeneic medical treatment has tremendous possibility and potential42. The screening of homozygous HLA-typed CBMCs before reprogramming can widely utilize the allogeneic iPSC lines for clinical treatment. Homozygous iPSC lines can avoid immunological reactions after allograft transplantation. This concept can also be applied to cartilage regeneration by generating HLA-homozygous cartilage11.
Usually, chondrogenic differentiation is performed through two critical steps: the induction of EB-derived outgrowth cells and pellet formation. The initial differentiation of hiPSCs into EB-derived outgrowth cells is critical to increase their chondrogenic differentiation potential. Outgrowth cells differentiated from hiPSCs are functionally and molecularly similar to native BMSCs7,43. Previous studies used monolayer culture or EB culture as a pre-differentiation step to induce mesenchymal-like cells10,43. However, direct differentiation by monolayer culture is time-consuming compared to the use of EBs, and chondrogenic pellets tend to differentiate into fibrocartilage with hypertrophic characteristics. It was reported that cell morphology and the chondrogenic differentiation potential of generated mesenchymal-like progenitor cells differ greatly according to the cell density of the cell monolayer9.
We used EBs to induce outgrowth cells, which is a relatively fast and simple method compared to the monolayer culture. Using this protocol, we were able to generate many pellets using a relatively small number of hiPSCs (Table 1). The size and number of EBs were critical for successful chondrogenic pellet mass production. For that reason, we enlarged the aggregated EBs in E8 medium, until more than half of the generated EBs were larger than 100 µm. After EB enlargement, we maintained the EBs in E8 medium without FGF2. Previously, researchers maintained generated EBs without FGF2 for mesodermal lineage induction9.
Even though we attempted to maintain the generated outgrowth cells at a high density for better quality, several limitations still remain. While we have confirmed that the generated outgrowth cells share a similar morphology, the cells may be still heterogenous. Previous studies have attempted to sort for a specific cell type; however, this procedure has resulted in the collection of a low number of cells. A new attempt to isolate homogenous outgrowth cells on a larger scale will be required to generate a large quantity of chondrogenic pellets with enhanced characteristics.
Various groups are inducing progenitor cells from hiPSCs by using monolayer or EB culture. The induction of progenitor cells with a high similarity to MSCs can be the key to obtaining chondrogenic pellets of higher quality. These progenitor cells derived from EBs have characteristics similar to those of MSCs. However, this might be because of their fibroblastic nature. Therefore, a detailed validation study on the outgrowth cells is required.
In this study, we suggest a protocol that can generate a relatively large quantity of chondrogenic pellets. We confirmed that the cartilage regenerated using CBMC-hiPSCs showed a healthy phenotype and can be used as material for tissue regeneration. However, an improved method with a shorter differentiation timeline is required for further application. The further development of quality-control standards to validate cartilage with a higher level of hyaline is also required for future applications of CBMC-hiPSCs as cell material for cartilage regeneration. The protocol is simple yet effective and does not require any additional sorting processes before pellet formation. In conclusion, using this protocol, high-quality chondrogenic pellets can be generated for studies on disease modeling, drug screening, and regenerative medicine to further our understanding of the nature of cartilage.
The authors have nothing to disclose.
This work was supported by a grant from the Korea Healthcare Technology R&D project, Ministry for Health, Welfare & Family Affairs, Republic of Korea (HI16C2177).
Plasticware | |||
100mm Dish | TPP | 93100 | |
6-well Plate | TPP | 92006 | |
50 mL Cornical Tube | SPL | 50050 | |
15 mL Cornical Tube | SPL | 50015 | |
10 mL Disposable Pipette | Falcon | 7551 | |
5 mL Disposable Pipette | Falcon | 7543 | |
12-well Plate | TPP | 92012 | |
Name | Company | Catalog Number | Description |
E8 Medium Materials | |||
DMEM/F12, HEPES | Life Technologies | 11330-057 | E8 Medium (500 mL) |
Sodium Bicarbonate | Life Technologies | 25080-094 | E8 Medium (Conc.: 543 μg/mL) |
Sodium Selenite | Sigma Aldrich | S5261 | E8 Medium (Conc.: 14 ng/mL) |
Human Transfferin | Sigma Aldrich | T3705 | E8 Medium (Conc.: 10.7 μg/mL) |
Basic FGF2 | Peprotech | 100-18B | E8 Medium (Conc.: 100 ng/mL) |
Human Insulin | Life Technologies | 12585-014 | E8 Medium (Conc.: 20 μg/mL) |
Human TGFβ1 | Peprotech | 100-21 | E8 Medium (Conc.: 2 ng/mL) |
Ascorbic Acid | Sigma Aldrich | A8960 | E8 Medium (Conc.: 64 μg/mL) |
DPBS | Life Technologies | 14190-144 | |
Vitronectin | Life Technologies | A14700 | |
ROCK Inhibitor | Sigma Aldrich | Y0503 | |
Name | Company | Catalog Number | Description |
Quality Control Materials | |||
18 mm Cover Glass | Superior | HSU-0111580 | |
4% Paraformaldyhyde | Tech & Innovation | BPP-9004 | |
Triton X-100 | BIOSESANG | 9002-93-1 | |
Bovine Serum Albumin | Vector Lab | SP-5050 | |
Anti-SSEA4 Antibody | Millipore | MAB4304 | |
Anti-Oct4 Antibody | Santa Cruz | SC9081 | |
Anti-TRA-1-60 Antibody | Millipore | MAB4360 | |
Anti-Sox2 Antibody | Biolegend | 630801 | |
Anti-TRA-1-81 Antibody | Millipore | MAB4381 | |
Anti-Klf4 Antibody | Abcam | ab151733 | |
Alexa Fluor 488 goat anti-mouse IgG (H+L) antibody | Molecular Probe | A11029 | |
Alexa Fluor 594 goat anti-rabbit IgG (H+L) antibody | Molecular Probe | A11037 | |
DAPI | Molecular Probe | D1306 | |
Prolong gold antifade reagent | Invitrogen | P36934 | |
4% Paraformaldyhyde | Tech & Innovation | BPP-9004 | |
Tween 20 | BIOSESANG | T1027 | |
Bovine Serum Albumin | Vector Lab | SP-5050 | |
Anti-Collagen II antibody | abcam | ab34712 | 1:100 |
Alcian blue | Sigma Aldrich | A3157-10G | |
Fast Green FCF | Sigma Aldrich | F7252-25G | |
Safranin O | Sigma Aldrich | 090m0039v | |
Nuclear fast red | Americanmastertech | STNFR100 | |
xylene | Duksan | 115 | |
Ethanol | Duksan | 64-17-5 | |
Mayer's hematoxylin solution | wako pure chemical industries | LAK7534 | |
DAP | VECTOR LABORATORIES | SK-4100 | |
Slide Glass, Coated | Hyun Il Lab-Mate | HMA-S9914 | |
Trizol | Invitrogen | 15596-018 | |
Chloroform | Sigma Aldrich | 366919 | |
Isoprypylalcohol | Millipore | 109634 | |
Ethanol | Duksan | 64-17-5 | |
RevertAid First Strand cDNA Synthesis kit | Thermo Scientfic | K1622 | |
Name | Company | Catalog Number | Description |
Chondrogenic Differentiation Materials | |||
DMEM | Life Technologies | 11885 | Chondrogenic media component (500 mL) |
Penicilin Streptomycin | Life Technologies | P4333 | Chondrogenic media component (Conc.: 1 %) |
Ascorbic Acid | Sigma Aldrich | A8960 | Chondrogenic media component (Conc.: 64 μg/mL) |
MEM Non-Essential Amino Acids Solution (100X) | Life Technologies | 11140-050 | Chondrogenic media component (Conc.: 100 mM) |
rhBMP-2 | R&D | 355-BM-050 | Chondrogenic media component (Conc.:100ng/ml) |
Recombinant Hman TGF-beta3 | R&D | 243-B3-002 | Chondrogenic media component (Conc.:10ng/ml) |
KnockOut Serum Replacement | Life Technologies | 10828-028 | Chondrogenic media component (Conc.: 1 %) |
ITS+ Premix | BD | 354352 | Chondrogenic media component (Conc.: 1 %) |
Dexamethasone-Water Soluble | Sigma Aldrich | D2915-100MG | Chondrogenic media component (Conc.:10-7 M) |
GlutaMAX Supplement | Life Technologies | 35050-061 | Chondrogenic media component (Conc.: 1 %) |
Sodium pyruvate solution | Sigma Aldrich | S8636 | Chondrogenic media component (Conc.: 1 %) |
L-Proline | Sigma Aldrich | P5607-25G | Chondrogenic media component (40μg/ml) |