Serum-Free Culture Conditions for Promoting Chondrogenic Differentiation
The major challenge in attempting to culture stem cells under serum-free conditions is that cells generally tend to have a lower mitotic index, become apoptotic, and display poor adhesion in the absence of serum . The removal of serum has been reported to slow down the proliferation rate of MSCs [51, 52]. Nevertheless, the osteochondral potential of MSCs can still be maintained in a chemically defined serum-free medium [50, 53]. Presently, most studies with MSCs use serum-free in vitro culture conditions so as to avoid the arte-factual and pleiotropic effects of serum on the experimental data. However, for optimal ex vivo proliferation of MSCs, the presence of serum is still preferred . With primary chondrocytes, the absence of serum has also been reported to be detrimental to the proliferation rate . In the case of ES cells, serum-free in vitro culture condition for differentiation into the chondrogenic lineage was recently reported by Nakayama et al. .
To prevent transmission of pathogens, the patient's own serum could be used for the in vitro culture of stem cells. However, for clinical applications, there are several reasons that make it preferable to eliminate serum from the in vitro culture milieu. First and foremost, the composition of serum is poorly defined, with a considerable degree of interbatch variation, even when obtained from the same patient or manufacturer. This impedes good quality control in the laboratory. Serum is also not completely physiological, because it is essentially a pathological fluid formed in response to blood clotting. Additionally, it may also contain uncharacterized growth and differentiation factors, which may result in the uncontrolled spontaneous differentiation of stem cells into divergent multiple lineages other than the chondrogenic lineage.
A step toward serum-free culture conditions is the development of chemically defined synthetic serum substitutes. At present, there are several such commercially available synthetic serum substitutes [49, 50]. Most notable of these is knockout serum replacement (KSR), which was specifically developed for the maintenance of ES cells in an undifferentiated state within in vitro culture . The exact chemical composition of KSR is not available, because it is a protected trade secret. However, it has been reported to be completely devoid of any undefined growth factors or differentiation-promoting factors . This would be extremely useful for achieving controlled differentiation of stem cells into the chondrogenic lineage in vitro.
Use of Exogenous Cytokines and Growth Factors to Promote Chondrogenic Differentiation
The use of exogenous cytokines and growth factors is another step forward in the development of a defined culture milieu for directing the chondrogenic differentiation of stem cells. Indeed, numerous cytokines and growth factors have been implicated in chondrogenesis (Table 1), and many of these display a high degree of functional overlap. Because the process of chondrogenesis is so closely intertwined with osteogenesis, many of the cytokines and growth factors that promote chondrogenic differentiation are also somewhat implicated in osteogenic differentiation [56, 57]. Hence, the challenge is to find an optimized subtle combination of these various cytokines and growth factors that would bias differentiation specifically toward the chondrogenic lineage. An added complication is that many of these cytokines and growth factors exert nonspecific pleiotropic effects on stem cell differentiation, which is most likely attributable to the activation of multiple intracellular signaling pathways by each individual cytokine or growth factor.
Among the most potent inducers of chondrogenic differentiation are members of the transforming growth factor beta (TGF-β) family. These consist of more than 40 polypeptide growth factors that share a high degree of homology, in particular the seven conserved residues in their C-terminal region . Besides the various isoforms of TGF-β [41, 59–61] and bone morphogenetic protein (BMP) [36, 57, 62], other members of the TGF-β superfamily include activin , osteogenic protein-1 , and GDF-5 . The effects of cytokines of the TGF-β superfamily on chondrogenic differentiation are transduced through two major intracellular signaling pathways. The first of these involves the SMAD family of signaling molecules , whereas the second pathway involves mitogen-activated protein kinase (MAPK) signaling . Interestingly, both signaling cascades are activated by the same TGF-β receptor complex. This comprises two distinct transmembrane protein components (type I and II receptor) that undergo heterodimerization upon binding to cytokines of the TGF-β superfamily. More recently, there is also evidence of regulatory cross-talk between TGF-β–activated signaling cascades and the Wnt pathway during chondrogenic differentiation [41, 68]. An interesting aspect of the TGF-β superfamily is the modulation of its biological activity through binding interactions with extracellular matrix proteins such as β-glycan and endoglien [69, 70]. These are thought to alter the presentation of these cytokines to their corresponding receptors on the cell surface.
Another family of cytokines that plays an important role in chondrogenesis is comprised of the various isoforms of fibroblast growth factor (FGF). Numerous studies have implicated the FGF family of cytokines in limb bud cartilage formation . The effects of FGF on chondrogenic differentiation are transduced primarily through MAPK signaling . This leads to increased expression of the transcription factor Sox9, which is the master regulator of chondrogenesis. Additionally, there is also evidence of regulatory cross-talk between FGF-activated signaling cascades and the signaling pathways initiated by parathyroid hormone–related peptides and Indian hedgehog [73, 74].
Besides the TGF-β and FGF family of cytokines, insulin-like growth factor-1 (IGF-1) [75–77] has also been shown to have potent stimulatory effects on chondrogenic differentiation. It is believed that the chondrogenic-promoting activity of IGF-1 is primarily transduced through a phosphatidyl inositol 3-kinase signaling pathway [78, 79]. Because the combination of TGF-β1 with either IGF-1  or FGF-2  was reported to have a synergistic effect on chondrogenic differentiation in vitro, it is likely that the IGF-1–activated signaling cascade could have regulatory cross-talk with the signaling pathways initiated by the TGF-β and FGF family of cytokines. Other proteins that have also been shown to have a stimulatory effect on chondrogenesis include prolactin , interleukin-1β , Cyr61 , HB-GAM , and growth hormone . All of these are summarized in Table 1.
Use of Nonproteinaceous Chemical Compounds to Promote Chondrogenic Differentiation
In addition to protein-based cytokines and growth factors, several nonproteinaceous chemical compounds have also been shown to promote chondrogenic differentiation in vitro (Table 2). Such chemicals tend to be less labile, with a longer active half-life in solution, compared with protein-based cytokines and growth factors. This is advantageous for prolonged in vitro cell culture over several days or even weeks. Moreover, unlike proteins that have to be synthesized in living cells and subjected to complex post-translational modifications (i.e., glycosylation, peptide splicing, conformational folding), non–protein-based chemical compounds can be manufactured by chemical reactions in the laboratory and hence are more structurally and chemically defined compared with proteins.
Table Table 2.. Nonproteinaceous chemical factors that promote chondrogenesis
|Chemical factors||Key references|
|Prostaglandin E2||Miyamoto et al. , Biddulph et al. |
|Thyroid hormone||Siebler et al. |
|1,25-dihydroxy vitamin D||Tsonis , Harmand et al. |
|Ascorbic acid||Farquharson et al. |
|Dexamethasone||Johnstone et al. , Mackay et al. |
|Ethanol||Kulyk et al. , Hoffman et al. |
|Staurosporine||Kulyk , Kulyk et al. |
|Dibutryl cAMP||Revillion-Carette et al. |
|Concanavalin A||Mikhailov et al. |
|Vanadate||Kato et al. |
|FK506||Nishigaki et al. |
Among the nonproteinaceous chemical compounds that are known to promote chondrogenic differentiation in vitro (Table 2) are dexamethasone, thyroid hormone, 1,25-dihydroxy vitamin D3, prostaglandin E2, ascorbic acid, ethanol, staurosporine, dibutryl cAMP, concavalin A, vanadate, and FK506. Dexamethasone is a synthetic glucocorticoid that is a potent inducer of chondrogenic differentiation in human-derived mesenchymal stem cells [87, 88]. Thyroid hormones are steroid derivatives of cholesterol metabolism that have also been implicated in chondrogenic differentiation [89, 90]. In an interesting study by Locker et al. , it was reported that the sequential addition of dexamethasone and thyroid hormone (triiodothyronine) under serum-free conditions permitted full chondrogenic differentiation of the pluripotent mesoblastic C1 cell line. By contrast, the addition of exogenous cytokines (IGF-1 and TGF-β), as well as the cell-intrinsic activation of the BMP autocrine signaling pathway, failed to elicit full chondrogenic differentiation . 1,25-dihydroxy vitamin D3, also known as calcitriol, is the active form of vitamin D. It has been reported to stimulate chondrogenesis of embryonic limb bud mesenchymal cells  and promote expression of differentiated chondrocyte function within in vitro culture . Prostaglandin E2 (PE2) is a naturally occurring eicosanoid that is derived from arachidonic acid metabolism. Numerous studies have implicated PE2 in limb cartilage formation [94, 95] and chondrocyte differentiation . Ascorbic acid, better known as vitamin C, has been shown to stimulate chondrogenic differentiation by promoting 1,25-dihydroxy vitamin D3 synthesis and cartilage matrix production . The teratogenic effects of ethanol on vertebrate development are well-known. It has been demonstrated to have potent stimulatory effects on the chondrogenic differentiation of both embryonic limb  and embryonic facial  mesenchyme cells. Staurosporine, a protein kinase C inhibitor, has also been demonstrated to have stimulatory effects on the chondrogenic differentiation of both embryonic limb  and embryonic facial  mesenchyme cells. Dibutryl cAMP, the cell-permeable form of cAMP, was reported to have a stimulatory effect on embryonic limb bud cartilage formation . Concanavalin A is a member of the lectin family of carbohydrate-based compounds. It has been reported to induce both neural and cartilage tissue formation in amphibian early gastrula ectoderm . Vanadate, an inorganic vanadium salt, has been demonstrated to promote cartilage-matrix proteoglycan synthesis in rabbit coastal chondrocyte cultures . FK506 is an immunosuppressive drug that has been shown to stimulate chondrogenic differentiation of a clonal murine embryonic carcinoma cell line, ATDC5 .
Use of Naturally Occurring and Artificially Synthesized Extracellular Matrix Substratum to Promote Chondrogenic Differentiation
The in vitro culture milieu for directing chondrogenic differentiation should also incorporate naturally occurring and artificially synthesized extracellular matrix (ECM) substratum to optimize cell attachment, growth, and differentiation. Indeed, the presence of ECM components would more closely replicate the physiological environment that supports chondrogenesis. In situ, cartilage tissue consists of chondrocytes embedded within an avascular ECM that serves to maintain its structural integrity. This is primarily comprised of collagenous proteins  and proteoglycans  that account for the bulk of its dry weight (60%–90%), together with smaller quantities of noncollagenous proteins [108, 109] and hyaluronan . Within cartilaginous tissue, the major collagen isoform is collagen type II , whereas the predominant proteoglycan is aggrecan . In addition to its structural role, cartilage ECM plays a physiological role by influencing the immediate microenvironment of the chondrocytes embedded within it. Histological studies have reported extensive remodeling of cartilage ECM during growth and development  as well as under various pathological conditions [114, 115].
Hence, the introduction of appropriate ECM substratum within in vitro culture would certainly enhance the directed differentiation of stem cells into the chondrogenic lineage . These can either be based on naturally occurring components of cartilage ECM or may use synthetic materials. Additionally, composite matrix scaffolds of both natural and synthetic materials have also been fabricated.
It is important to note the various properties of ECM substratum that would be favorable for chondrogenesis. For tissue engineering applications, it is imperative that the synthesized matrix substratum is biocompatible and has no cytotoxic properties. Biodegradability is also another preferred characteristic , although this is not absolutely critical. Culture on 3D matrix scaffolds has been reported to be superior to conventional 2D monolayer culture for maintaining the differentiated phenotype of chondrocytes . Porosity of the ECM substratum is essential for 3D tissue growth. Hence, matrix scaffolds for cartilage tissue engineering are often fabricated as porous foams or granules. At the present moment, there are as yet no reported studies on the use of ECM substratum to promote the chondrogenic differentiation of ES cells. Hence, the following discussion will focus solely on MSC and primary chondrocytes.
The attachment, proliferation, and subsequent chondrogenic differentiation of MSC and chondrocytes on the ECM substratum are dependent on several interrelated properties, namely chemical composition, electrostatic charge, surface texture/roughness, and geometrical configuration . Obviously, the chemical composition of the ECM substratum is the most critical factor in determining its ability to influence chondrogenesis. These should preferably incorporate naturally occurring constituents of cartilage ECM, so as to provide a more physiological environment for chondrogenic differentiation. Indeed, matrix scaffolds fabricated for cartilage tissue engineering are often based on either collagen [120–122], hyaluronan [123, 124], chondroitin sulfate [125, 126], mineralized calcium [127, 128], fibrin [129, 130], or composites of these materials [131, 132], all of which are constituents of cartilage ECM. Other naturally occurring materials that are not found in cartilage ECM have also been used. These include gelatin , chitosan , and alginate [135–138]. Additionally, matrix scaffolds based completely on synthetic materials such as poly (lactic-co-glycolic acid), poly (ethylene glycol), poly (epsilon-caprolactone) have also been fabricated [139–144]. Nevertheless, it would be preferable to use these as composites with naturally occurring materials. The electrostatic charge on the surface of the ECM substratum is another important factor. The presences of negative charges have been reported to be advantageous for the adhesion and spreading of MSCs on ECM substratum [145, 146]. Furthermore, there is evidence that variations in the electrostatic charge content of the peri-cellular matrix synthesized by chondrocytes have a profound influence on the biomechanical properties of cartilaginous tissue . Surface texture/roughness seems to have differential effects on chondrocyte proliferation, differentiation, and matrix production, depending on the maturational state of the chondrocyte itself [148, 149]. Bhardwaj et al.  reported that geometric configuration of the matrix scaffold, as defined by average pore size, had profound effects on chondrocyte proliferation and matrix synthesis.
As mentioned earlier, collagen type II is the most abundant collagen isoform within cartilage ECM. Indeed, it has been reported that cartilage-specific collagen type II is superior to collagen type I for primary culture of chondrocytes [119, 120]. Although chondrocytes appeared to maintain high proliferative capacity when cultured in collagen type I, there was a gradual loss of differentiated phenotype . To enhance their structural and chondroinductive properties, collagen-based matrices are often fabricated as composites that may incorporate a variety of naturally occurring and synthetic materials such as mineralized calcium phosphate , agarose , proteoglycan aggregates , demineralized bone powder , fibronectin , chondroitin sulfate [125, 126], hyaluronan , poly-L-lactic acid , and polylactide-coglycolid .
Another class of bone ECM molecules that is known to play an integral role in chondrogenic differentiation is the glycosaminoglycan (GAG), which is essentially a long-chain sugar molecule. With the exception of hyaluronan, all of GAGs (heparin sulfate, chondroitin sulfate, keratin sulfate, and dermatin sulfate) are sulfated branch-chained molecules that are conjugated to proteins in the form of proteoglycans. Chondroitin sulfate is the major sulfated glycosaminogly can found in cartilage (predominantly conjugated to aggrecan). Indeed, chondroitin sulfate–based matrix scaffolds have been reported to enhance chondrocyte proliferation and maturation [121, 131]. Additionally, hyaluronan-based matrix scaffolds are also widely utilized in cartilage tissue engineering [123, 124, 159, 160].
At present, the development of ECM substratum for application in cartilage tissue engineering is progressing rapidly. It is anticipated that more novel types of composite matrix scaffolds incorporating a variety of natural and synthetic materials will be developed in the near future for the controlled differentiation of MSC and even ES cells into the chondrogenic lineage.