18.1.1 Background

The multidisciplinary field of tissue engineering offers potentially exciting strategies for developing new and effective treatments for the repair and regeneration of damaged or diseased tissues. Such treatments, using living cells and biomaterials, exploit new methodologies in understanding principles in cell biology and materials science that control and, ultimately, direct target cell function. These potential applications in medicine are diverse, and are therefore likely to have a major impact in treatments for a variety of tissues, and in particular, those of the musculoskeletal system. Research techniques developed in tissue engineering involve a range of approaches, the key element of which is the use of biologically designed systems to achieve repair, healing and/or regeneration of trauma-affected or diseased tissues. The ultimate aim is to deliver an assembled three-dimensional biomimetic ‘tissue’ component, which clearly distinguishes this tissue engineering concept from the use of traditional medical devices, and would fill a void in situations where many tissues fail to repair or heal properly.

18.1.2 Articular cartilage: a ‘simple’ but complex tissue

This chapter will deal with the repair of articular cartilage, i.e. the cartilage that aids the smooth movement of synovial joints such as the hip, knee and elbow. Articular cartilage has long been recognised and described as a remarkable tissue, even though it is more often thought of as a simple component of the musculoskeletal system because of its seemingly basic structure compared to many other more metabolically active, multicellular and structurally complex tissues of the body.

For centuries, it has been recognised that cartilage is a difficult tissue to either repair or regenerate. Indeed, diseased cartilage has been described thus, ‘an ulcerated cartilage is universally allowed to be a very troublesome disease’ [1]. This is true today and degeneration of articular cartilage, as a result of osteoarthritis (the most prevalent disorder of the musculoskeletal system) affects approximately a third of adults over 65 years of age worldwide. Furthermore, damage to cartilage which occurs more frequently in younger individuals as a consequence of trauma, as well as disease states typical of older patients, such as inflammatory rheumatoid arthritis, can also have severe consequences on the successful regeneration of the tissue.

18.2.1 Implantation procedure in humans

The basic principle is that the techniques available for growing healthy in vitro chondrocytes can be employed clinically by harvesting cells from the patient, culturing the chondrocytes, and reimplanting them. The first step is an arthroscopic evaluation and biopsy. The defect, which can be located on the femoral condyle in a load-bearing area or the trochlea, is assessed. A biopsy of healthy articular cartilage is then taken from the medial or lateral femoral condylar ridge in a non-weight-bearing area. The biopsy specimen is used for laboratory culture of additional chondrocytes. Once chondrocytes have grown to about 2–5 million cells, they are ready for implantation, typically 11 to 21 days later. An arthrotomy of the knee is performed. The edges of the defect are trimmed to provide a healthy cartilage edge. The base of the lesion is also debrided so that only bone (no cartilage) is present. Next, a periosteal patch the same size as the defect is harvested from the anteromedial proximal tibia. The patch is then sutured onto the defect and the edge sealed with fibrin glue. Finally, the cells are injected under the patch and the injection point sealed.

18.2.2 Chondrocyte proliferation

Over time, the implanted cells begin to produce articular cartilage. Initially, the transplanted chondrocytes proliferate rapidly. A maturation phase follows with increased formation of matrix. At 2 to 6 months, the cartilage in the subchondral region forms an orderly transition into bone, resulting in an articular surface that closely resembles the host hyaline cartilage. Biopsies taken 1 to 2 years after implantation have revealed tissue described as ‘hyaline-like’. Clinical results have shown up to 80% good results at 5 to 10 year follow-up in some patients, with a complication rate of about 5% [20]. Most often, these complications have been in the form of surgical adhesions or fibrosis. In approximately 5% of patients, the graft failed outright. There are several practical limiting factors in the current techniques:

Some studies have considered randomised controlled trials. In 2003, a prospective comparative trial considered 2 year outcomes of 40 patients randomised to either osteochondral plug transplantation or ACI with 20 patients per group, yet no significant differences were found using several scoring systems [21]. A larger trial comprising 100 patients was presented where the patients were randomised to either mosaicplasty or ACI and again, no significant difference was calculated between the groups [22]. Furthermore, in 2006, a Cochrane Review was published that included four randomised controlled trials comprising almost 300 patients [23]. This study concluded that there was no evidence of a significant difference in the outcomes between ACI and other cartilage repair interventions. The authors reported that additional randomised controlled trials with long-term functional outcomes were required to provide clearer guidelines for clinical practice. It is clear that future reports of the longer-term results of such and other on-going studies will be of significant interest.

Other common treatments involve considerable surgery and removal of the affected tissue, e.g. total knee or hip replacement. This is a reasonably successful method to restore function to an affected joint, particularly in older patients, but total success is limited as a result of joint loosening and implant failure. Alternative more conservative methods include drilling, abrasion, debridement and microfracture through to the subchondral bone to encourage bleeding and release of relevant cells and growth factors [19]. The fibrocartilage formed, however, is less adequate and biochemically and mechanically inferior to the original tissue. Even less successful are defects that do not penetrate to the subchondral bone, more often than not failing to heal spontaneously and leading to further complications. With alternative graft use, material supply is very often limited.

18.3.1 Cartilage tissue in vitro

Various natural polymers have been investigated as scaffolds for cartilage repair. Poor mechanical properties, however, can be a problem. Furthermore, structuring and configuring these materials may be difficult with respect to design, manufacture/synthesis and integration to human tissues. Thus, it seems reasonable to develop a system whereby natural materials will be combined with synthetic biodegradable alternatives that may be rapidly cured within a 3D plotter to provide a stable layered chondrocyte scaffold.

The aim of this research ethos is to develop a scaffold that enables tissue engineering of neocartilage that is closer in structure and mechanical properties to original cartilage. The two most important facets of articular cartilage replacement therapy are (i) to fill any defect with appropriate tissue that has the same biochemical and mechanical properties as native tissue and (ii) to ensure the complete integration of the repair tissue and the native tissue [24]. Furthermore, it is not sufficient to consider only repair therapies for cartilage tissue alone, when many types of defects and degenerative diseases involve the subchondral bone, too. Thus, this chapter will also consider the role of recent therapies on osteochondral defects, those involving both cartilage as well as subchondral bone tissues. This presents another series of factors and considerations for in vitro engineering of such contrasting tissues.

During the initial stages of osteoarthritis, partial thickness defects of articular cartilage often consist of fissures which are limited to the articular cartilage above the calcified region. These defects do not heal sufficiently and there is considerable evidence to suggest it is because the injured tissue does not penetrate the calcified region and thus has no access to the progenitor and mesenchymal stem cells in the subchondral bone space [24]. Full-thickness defects on the other hand do pass through the subchondral bony layer and the repair response results in the relatively quick formation of fibrocartilage. Furthermore, many operative procedures in current practice make use of this approach to provide a mechanism of repair.

During the past 10–15 years, much progress and innovation have emerged regarding new strategies that have the potential to alleviate the symptoms of cartilage failure in patients. Combining innovation and new fundamental knowledge in the fields of cell and molecular biology, as well as materials engineering and biomechanics, and the post-genomic sciences, together have offered a more practical approach to tissue engineering. This has led to a focus on the joint delivery of appropriate cells, a scaffold matrix and relevant biological molecules that can promote cell differentiation.

18.3.2 Articular cartilage tissue engineering

Scaffolds are used in tissue engineering to incorporate chemical and mechanical signals to ‘guide’ the adhesion, migration and ultimate differentiation of the cells. This is important, as individual cell types require numerous mechanical, chemical, structural and spatial cues, which cells utilise to adapt to the extracellular environment and chondrocytes are no different [25, 26]. Typically, a scaffold on which cells are seeded also serves to strengthen the injured site mechanically [27].

3D tissue engineering methods are increasingly being developed and the evolution of scaffolds has led to the use of functional tissue substitutes in the treatment of cartilage defects. The 3D environment provides definitive gradients in which the cells ‘sense’ their environment, nutrients and external environmental cues; thus being in 3D surroundings is necessary for cell remodelling and other specific morphogenic events over time [28, 29]. It is therefore important to consider the morphogenic and 3D implications of seeding the cells and the effects these might have on cellular behaviour. Several studies have emphasised the importance of morphology in culturing and seeding chondrocytes in matrices [30–32], as 3D culture provides cells with an additional dimension for these cues, which has major effects on cell adhesion, matrix remodelling and integrin ligation, as well as the intracellular signalling associated with these processes [28].

18.4.1 Naturally derived biomaterials

Scaffolds made from natural polymers offer suitable compatibility in many cases and are excellent to support chondrocyte survival and cartilage matrix synthesis. Unsurprisingly, a gel-like environment is very well suited for articular chondrocytes since they are encased within and ultimately reside in an environment whereby the spherical morphology is maintained. Such hydrogels are three-dimensional, hydrophilic, polymeric networks capable of imbibing large amounts of water or biological fluids commonly used to mimic the chondrogenic environment. There are two main classes of hydrogels: (i) naturally derived hydrogels, such as collagen and alginate, and (ii) synthetic-based hydrogels [33].

Both animal and recombinant collagens are widely used biomaterials in cartilage tissue engineering, mainly because of their compatibility with cells and tissues, their abundance and conserved structure; but also due to the low immunogenic response they generate when implanted [34]. Collagens are readily isolated from many species and are relatively easy to purify with enzymes for use in tissue regeneration. However, currently the most widely used approach for synthesising a collagen scaffold is to make it into a hydrogel [35].

However, most hydrogels are mechanically inferior for long-term use and completely unsuitable for some tissue engineering purposes, such as joint regeneration due to a mismatch in the mechanical properties with the tissues of interest. Indeed, this is a problem with most tissue engineered structures used for repairing cartilage defects, which are often associated with mechanical instability, and in the worst case leads to further joint degeneration [36, 37]. When engineering cartilage replacements, it is especially important to consider the degradation of the scaffold, the degree of tissue remodelling of the scaffold and the by-products of this process. Kimura et al. first cultured and maintained chondrocytes in a collagen scaffold in 1984 [38]. Chondrocytes cultured in monolayers for prolonged periods of time lost their expression of collagen type II and expressed only collage type I, in addition to becoming elongated and losing their characteristic rounded morphology. Chondrocytes thus de-differentiate into cells with a fibroblast-like morphology which produces a predominantly non-glycosaminoglycan rich matrix [36, 39]. On the other hand, freshly isolated cells, which still maintained their rounded morphology showed expression of collagen type II [40]. Mature chondrocytes have a rounded shape when inside scaffolds, with evident sulphated glycosaminoglycan positive staining with Alcian blue cultured chondrocytes in collagen type I 3D matrix and managed to maintain the rounded cell morphology, with the chondrocytes synthesising collagen type II after 4 weeks in culture [41]. A cartilage-like structure has also been generated when maintaining the rounded chondrocyte morphology and found that these cells also synthesised collagen type II [42]. Some groups also advocate mimicking the zoned structure of cartilage in order to achieve the most realistic results and to increase both functionality and the long-term stability of the tissue [36, 43]. It is apparent that appropriately combining all these different extracellular signals will be the key for successful design of 3D scaffolds mimicking articular cartilage tissue for regeneration purposes.

Important considerations for designing biodegradable hydrogels for cell encapsulation have been reviewed [44]. It is well known that chondrocytes in monolayer cultures dedifferentiate to a more fibroblast-like phenotype and produce much less cartilage ECM than their 3D counterparts. In contrast, the cells grown in 3D collagen constructs often present a prolonged (up to 28 days) chondrocyte-specific gene expression and cartilage ECM synthesis during the total time in culture [45]. Other groups have shown that dynamic compression in cartilage applied in physiologically relevant conditions positively influenced the production of cell-secreted proteoglycans [46]. Similar results were obtained using an in vitro model based on type I collagen hydrogel scaffolds. Thus, in bovine articular chondrocytes grown in collagen sponges which had a physiologically relevant hydrostatic fluid pressure applied, synthesis of cartilage-specific matrix components was significantly enhanced [47].

In contrast to the need of an adequate perfusion of the medium through collagen sponges cultured to mimic bone tissue, perfusion conditions can inhibit chondrogenesis within scaffold systems. This is perhaps unsurprising as the physiology of cartilage tissue dictates it is in an environment of low oxygen tension of between 10% at the articular surface to 1% in the subchondral region. Such an oxidative stress will have profound effects on chondrocyte metabolism [48]. It has also been shown that differentiation of mesenchymal stem cells (MSCs) is directed to chondrocytes in such conditions [49]. Such results clearly emphasise the necessity of developing in vitro culture models where the oxygen gradient levels and hypoxic conditions in association with the mechanical forces are optimised in order to obtain a more physiologically relevant cartilage.

Collagen scaffolds have in the last decade become more prevalent as promising constructs to maintain cartilage cells. For example, human vertebrate disc cells have demonstrated a high proliferation rate and enhanced rate of proteoglycan synthesis in this type of matrix [50]. However, it has also been reported that the proteoglycan content of similar scaffolds cultured for 2 months never exceeded 10% of that present in the cartilage tissue, including that of the mature nucleus pulposus, i.e. a more inferior ECM compared with that of articular cartilage. Moreover, it is often seen that many more proteoglycans were actually lost to the culture medium than were retained in the cell/scaffold construct. This highlights the absolute necessity to optimise proteoglycan synthesis and ultimately retention by collagen constructs [51].

Several natural polysaccharides such as hyaluronic acid, alginate, chitosan, cellulose and dextran have also been explored for chondrocyte encapsulation as potential encapsulation media for chondrocytes and several of these will be considered here. Hyaluronic acid is a dominant component of the extracellular matrix found in developing embryonic mesenchymal tissues. It can be chemically and physically modified and, thus, can be fabricated into a large variety of physical forms [52]. It has also been shown that the chondroinductive properties of the high-molecular weight form suggest that it can be used as potential material or adjunct to develop an in vitro model of cartilage [53]. Indeed, there was a dose-dependent response to the exposure of hyaluronic acid to bovine articular chondrocytes in vitro; thus, low concentrations of hyaluronic acid (0.1 mg/mL and 1 mg/mL) significantly increase DNA, sulphated glycosaminoglycan and hydroxyproline synthesis. Immunohistology confirmed the maintenance of cell phenotype with increased matrix deposition of chondroitin-6-sulphate and collagen type II [54]. Numerous chemical modifications of hyaluronic acid have been designed to enable it to be highly functional and have led to its modification of biodegradation in vitro. One form of hyaluronic acid is a commercially available scaffold, Hyaff®-11, that is produced in different physical forms and has been shown promise for cartilage repair [55, 56].

The in vitro chondrogenic potential of this Hyaff®-11 scaffold with human MSCs up to 28 days in culture and in the presence of a high concentration of transforming growth factor-beta (TGF-β) has been investigated. As shown by the temporal expression of relevant chondrogenic genes such as Sox9, type I-, type II-, type IX-, type X-collagens and Aggrecan during chondrogenesis, the cultures of human MSCs into Hyaff®-11 were clearly characterised by a sequence of cellular and molecular events pointing to the in vitro formation of a neocartilage [57]. However, and almost predictably, the resulting morphology of the newly formed tissue was immature and histologically inferior to that which obtained from in vivo implantation. In this context, the use of an appropriate mechanical stimulation could be crucial for the development of a functional 3D in vitro cartilage model. Indeed, an enhancement of both type II collagen and aggrecan expression was observed when swine articular chondrocytes were mechanically stimulated for 5 days in vitro, confirming the importance of mechanical stimuli [58].

Alginate is a natural plant polysaccharide obtained from brown seaweeds that can gel in the presence of divalent cations by means of a simple ion crosslinking reaction. Remarkably, it does not degrade. Instead it dissolves when the divalent cations are replaced by monovalent ions. Although various alginate scaffolds do not necessarily promote cell–matrix interactions, this issue is resolved by incorporating the RGD sequence (Arg-Gly-Asp), a cell adhesion peptide motif [59]. However, alginate scaffolds without RGD peptides are still commonly used and recent studies have investigated chondrocyte differentiation within them [60, 61]. Such studies have indicated that MSCs showed a time-dependent accumulation of GAG, aggrecan and Type II collagen and that the resulting chondrocyte phenotype was clearly categorised into four distinct stages, which demonstrated a specific expression pattern of several putative novel marker genes for chondrogenesis. Upregulation of chondrocyte differentiation, somewhat tenuously confirmed by the synthesis of cartilage-like matrix, was also observed when porcine-derived cells were seeded in alginate scaffolds and cultured within a 3D perfusion system [60].

Chitosan, the structural component of the exoskeleton of crustaceans and fungal cell walls is another potential natural molecule linked with cartilage in vitro [62–64]. It is made by deacetylated chitin, i.e. high degrees of deacetylation lead to slower degradation times which increases hydrophobicity and thus markedly improved cell adhesion [65]. Chitosan hydrogels make suitable scaffolds to facilitate the entrapment of the highly negatively charged proteoglycans such as aggrecan owing to their cationic properties. Indeed, most proteoglycans produced by disc cells cultured within chitosan constructs were retained within the gel rather than released into the culture medium, unlike the scenario described for alginates [66].

A further natural biomaterial that has been considered for cartilage tissue engineering is silk. Silk fibroin hydrogels were explored for their potential to support chondrogenesis in vitro using lapine chondrocytes [67]. The pore sizes and the initial seeding density played significant roles in the type of cartilage tissue formed in vitro. Chondrocytes have also been shown to proliferate and maintain a differentiated phenotype within a silk-like sponge material at an enhanced rate compared with collagen sponges used as control [68]. Also, the mechanical properties of the regenerated cartilage tissue showed culture-dependent changes that were directly linked to the spatial and temporal deposition of cartilage-like ECM [69].

18.4.2 Synthetic polymers for subchondral bone repair

Synthetic polymers can be produced under a multitude of controlled and wide-ranging conditions. In so doing, very predictable and reproducible physical and mechanical properties such as degradation rate, tensile strength, elastic modulus and elasticity can be manipulated. Furthermore, those materials that can be formulated into 3D and biodegradable systems are of particular interest as their porosity, hydrophilicity and degradation properties can be partially controlled, and they can also be manufactured to a high degree of reproducibility [70, 71].

Commonly used biodegradable polymers for 3D scaffolds in such instances include the family of saturated poly-α-hydroxy esters, including poly(lactic acid) (PLA), poly(glycolic acid) (PGA), poly(ε-caprolactone) (PCL), as well as poly(lactic-coglycolide) (PLGA) copolymers [72–76]. Typically, cellular adhesion to PLGA is significantly higher compared to cells on PLA surfaces, and human osteoblasts grown on PLGA produced higher greater levels of ECM and developed a more mature cytoskeleton compared with PLA [77]. However, the deposition of ECM in such cases is rarely identical to the natural bone tissue. To improve such shortfalls in the quality of an ECM 3D perfusion systems with dynamic flow could be beneficial and several studies have shown the positive effects of perfusion upon cell distribution within scaffolds, an improved cell phenotype and even mineralised matrix synthesis within PLGA constructs compared to the static condition counterpart [78, 79]. While such in vitro systems offer an improved cell matrix deposition, maturation and appearance, the relative short culture times does not allow for highly organised and complex mineralised tissues to evolve.

PCL has proven biocompatibility and processability but its high hydrophobicity and low degradability in vivo make it less suitable for long term applications [80]. It is a promising material, however, and in particular with respect to a three-dimensional in vitro model of bone [81, 82]. PLA fibres reinforced with PCL allowed high proliferation of human MSCs and human osteoblasts, as well as the expression of alkaline phosphatase (ALP), although this was markedly less with respect to the control cultures.

In order improve the biological functionality of synthetic polymers intended predominantly for bone repair/regeneration, a variety of composite scaffolds have been developed which utilise hydroxyapatite (HA), biphasic calcium phosphate or tricalcium phosphate-type ceramic additions. For example, the development of heterogeneous scaffolds for articular cartilage and bone tissue engineering have been developed by production of rapid crosslinking, flexible, degradable polymers that can be fabricated into complex 3D structures that support production of new bone by osteoblasts. Utilising a variety of techniques, poly(propylene-co-ethylene) glycol of varying molecular weight has been used as initiators for ring opening polymerisation of lactide. Methacrylate groups were added to either chain end using methacryoyl chloride and the resultant purified PLDMA was characterised by nuclear magnetic resonance (NMR), Fourier transform infrared (FTIR) and Raman spectroscopy [83]. Commercial scale-up of production will be considered; poly(lead dimethacrylate) (PLDMA) crosslinking kinetics were assessed using attenuated total reflectance (ATR) FTIR [83, 84] and water sorption and mechanical properties were determined using gravimetric methods and dynamic mechanical analysis [84–88].

Other in vitro studies have demonstrated that similar composite scaffolds support the attachment, proliferation and differentiation of human MSCs [89–91]. Composite scaffolds including poly(DL-lactic acid), (PDLLA)/Bioglass©, PLA/calcium metaphosphate and PLGA/bioactive glass composites scaffolds have been developed and in vitro tested. For example, the increase of ALP activity in rat MSCs cultured on PLGA/bioactive glass was limited and insufficient to suggest full bone differentiation was evident and in addition, further studies are necessary to evaluate the expression of other early and late markers of bone-modulating genes, namely osteocalcin, type I collagen, bone sialoprotein and osteopontin are required [92]. Another study indicated that although nanofibrous scaffolds of PLA and nanocrystal demineralised bone powders supported bone formation some 12 weeks post-implantation, an upregulation of in vitro MSC differentiation was observed compared with the PLA scaffolds without bone. Indeed, after only 2 weeks in culture, the expression levels of type I collagen, ALP and Runx2 were similar in both types of scaffolds, perhaps indicating the absence of MSC differentiation enhancement in the composite scaffold [93].

Collagen sponges are inherently weak and thus have poor mechanical strength, and this invariably is a big disadvantage of these sponges if they are to be used as scaffolds for cell proliferation and differentiation. To overcome the inherent weakness of collagen sponges, combinations with other materials has been attempted. The incorporation of PGA fibres enabled collagen sponges to increase resistance to compression in vitro and in vivo [94]. In vitro culture tests revealed that the number of rat MSCs attached to the scaffolds increased with the incorporation of PGA fibres [95]. Moreover, the proliferation and the differentiation of MSCs cultured on PGA-reinforced collagen sponges were greatly influenced by a variety of culture conditions. Thus, appropriate dynamic perfusion conditions enabled MSCs to enhance the extent of proliferation and differentiation [95]. The presence of HA crystals within the collagen network in bone ECM has prompted the development of several scaffolds based on this structure.

A generation of scaffolds have been developed in order to mimic natural bone matrix and these were typically based on collagen and HA. Human MSCs seeded in collagen sponges reinforced with HA (ColHA scaffolds) and cultured for 28 days in both basal and osteogenic conditions revealed the infiltration of ALP positive cells throughout the constructs as well as the synthesis of new matrix [96]. Immunohistochemical staining showed osteocalcin was localized only in the periphery of the constructs, which may be indicative of limited diffusion of nutrient factors, or a particular hindrance of cell dispersion that does not allow for the formation of a mature ECM in the centre of the scaffolds. As in other 3D culture systems, the need of an appropriate perfusion of nutrient factors through the scaffolds is apparent. Although the ECM was composed of osteocalcin and type I collagen, the structure and the organisation of the in vitro neotissue were immature.

Mineralised type I collagen-based scaffolds have been also used in an attempt support human osteoclast-like cells and osteoblast cells in a co-culture system. Indeed, the osteoclast-like cells were able to invade and to degrade the scaffolds while osteoblasts proliferated, differentiated and produced mineralised ECM [97]. Such results confirmed the potential of these types of scaffolds in mimicking bone tissue, although further investigations are still needed to optimise, in particular, co-cultures and to initiate homogeneous in vitro bone tissue growth and organisation throughout a scaffold.

18.4.3 Bioceramic-based scaffolds for osteochondral repair

In order to develop an appropriate and successful full-thickness or osteochondral ‘plug’ for tissue engineering purposes, it is essential both bone and cartilage physiologies and mechanical properties are considered and combined, for example, by using engineered (bone–cartilage) composite scaffolds of predetermined pore dimensions and composition which promote both types of tissue formation. This is a big challenge because of the widely differing requirements of the component cells and hence tissue types. The bone aspect of the composite, made for example from a bioactive ceramic material, additionally needs to support anchoring of the graft within the defect since bone-to-bone interfaces bond more robustly and more readily than cartilage-to-cartilage interfaces [98].

Indeed, osteochondral tissue engineering can be considered a typical case of ‘interfacial tissue engineering’ which is becoming a discipline in its own right within tissue engineering [99, 100]. The tissue engineering of interfaces refers to the approaches being proposed to regenerate specialised tissue areas that intimately connect two different tissues of different biochemical and mechanical properties. The interface itself usually plays an important role in transferring mechanical load between tissues, as in this case; the osteochondral interface. Due to the complex biology and mechanics of the composite tissues, the challenge for osteochondral tissue models includes developing scaffolds that integrate with both the surrounding cartilage, and the underlying bone tissue. Many strategies for developing tissue engineering scaffolds for osteochondral repair consider the design of bilayered scaffolds that should be able to regenerate both cartilage and subchondral bone [99, 101, 102] and this approach involves the use of different combinations of materials, and specific properties in both regions of the scaffolds.

Several commonly used approaches have already been reported: (1) seeding autologous chondrocytes onto a scaffold creating a cell-seeded construct for in vivo implantation [103–106]; (2) two different cartilage and bone scaffolds assembled together either before or during implantation [107, 108] and (3) a completely integrated bilayered composite structure that leads to full integration of bone, cartilage layers and host tissues without needing a subsequent joining mechanism [109–111].

It is accepted that bilayered structures are more challenging to design and fabricate but they are ultimately more suitable for regenerating osteochondral defects. Such bilayered scaffolds should be able to incorporate and support the growth of different types of cells in favourable local environments requiring appropriate chemical signals and mechanical stimulation, leading to the growth of the two different tissues which are characterised by different biological requirements. The scaffold design is vital for the success of osteochondral tissue engineering, being necessary to consider the scaffold microstructure, surface topography, porosity, pore geometry and orientation, biodegradability and mechanical properties.

As a result of bone and cartilage having uniquely different biological compositions, and physiological requirements, tissue growth mechanisms, bilayered composite structures have been developed to exploit and combine the advantages of different biological materials. Bilayered scaffolds allow for establishment of optimised tissue-specific biological environments in each layer. Furthermore, they can be designed to mimic the native ECM for each tissue type, which may be more suitable than the fabrication of monolithic constructs with different functional requirements of both bone and cartilage in a single structure [112].

Bioactive ceramics and glasses are considered optimal candidates for the bone component because of their mechanical rigidity as well as high bone bonding properties. The polymeric phase confers toughness and plasticity, and it is suitable as substrate for providing integration with cartilage tissue. Therefore, the combination of bioceramic and polymeric phases generates suitable composite materials with adequate biological and mechanical properties attractive for osteochondral tissue engineering. A recent and comprehensive review describes the variety of approaches being investigated cartilage repair and encompasses the bilayered approach for osteochondral tissue engineering in which numerous challenges are highlighted and considered in depth [35].