3D Bioprinting for Cartilage and Osteochondral Tissue Engineering
3D生物打印技术在骨软骨组织工程中的应用

2.1.1 General Bioink Requirements

A wide range of bioinks, with suitable rheological behaviour, have been developed for microextrusion and inkjet bioprinting. For example, alginate,53-56 GelMA,57, 58 agarose,59, 60 collagen,61, 62 fibrin,63-65 silk,66-68 forms of poly(ethylene) glycol (PEG)69-71 and hyaluronic acid72, 73 have all been shown to be compatible with such bioprinting technology. A number of criteria must be met when developing bioinks for microextrusion. First, the bioink must have suitable rheological properties for controlled microextrusion, and second, it must also be capable of supporting cell growth and tissue development post-printing. Developing bioinks that can satisfy these two requirements is challenging. Cells benefit from lower polymer concentrations and cross linking densities where they can more readily proliferate and differentiate toward a target tissue.74-76 However, higher polymer concentrations often result in hydrogels with suitable rheological behaviour for extrusion as the viscosity is increased. In addition, higher polymer concentrations are typically associated with improved mechanical properties. These opposing requirements form what has been coined the biofabrication window.77 The most important rheological parameters to consider when designing bioinks are viscosity, yield stress and shear thinning behaviour.77 Higher polymer densities are more suitable for microextrusion since they are more viscous and possess a higher yield stress.

However, it should be noted that higher viscosity inks, along with higher extrusion pressures and smaller diameter nozzles, increase shear forces experienced by cells during extrusion, which can lead to cell death.78, 79 For example, shear stresses greater than 60 kPa have been shown to kill greater than 35% of cells during microextrusion.80 As a result, the resolution of microextrusion based bioprinting is limited by shear induced cell death. Typically filament diameters of 150–2000 µm can be achieved by controlling the bioink viscosity, extrusion pressure and nozzle diameter.57, 79, 81 The lowest resolution nozzle typically used for microextrusion of cells is 30 Gauge (159 µm diameter). The needle geometry can also be controlled to reduce shear forces experienced by cells during extrusion.57 For example, at lower inlet pressures cell viability can be increased by using conical rather than cylindrical needles. Interestingly however, at higher inlet pressures conical needles supported lower cell viability. The authors hypothesised that both the magnitude of shear stress along with the exposure time to the shear stress are important. For more general information on bioink development the reader is directed to a number of excellent reviews.77, 82, 83

2.2.1 Bioink Development for Cartilage Tissue Engineering

A number of different bioinks have been explored for CTE. In an early study, it was demonstrated how bioprinting technology could be used to engineer osteochondral tissue constructs using an alginate bioink.51 Bi-layered implants were printed with chondrocytes encapsulated in the cartilage layer and osteogenic progenitors encapsulated in the bone layer (Figure 1b). Distinctive tissue formation was observed both in vitro and in vivo using this system. Cartilage matrix components such as hyaluronic acid and chondroitin sulphate can be combined with alginate to create more biomimetic hydrogels capable of supporting superior neocartilage formation.84 GelMA has also been shown to support cartilage tissue formation using chondrocytes and MSCs,36 although this bioink appears to support the development of a more fibrocartilaginous type tissue with higher levels of collagen type I production.85 Again however, the addition of hyaluronic acid and chondroitin sulphate to GelMA can enhance chondrogenesis.36, 86 In addition, the incorporation of these components has the added benefit of increasing the bioink viscosity which can enhance printability.36 In a similar approach, a synthetic thermosensitive hydrogel composed of a base methacrylated polyHPMA-lac-PEG triblock copolymer was chemically combined with either methacrylated chondroitin sulphate (CSMA) or methacrylated hyaluronic acid (HAMA) for cartilage bioprinting.87

Novel shear thinning bioinks have also been developed for CTE by combining alginate and nanofibrillated cellulose.88 These inks are highly printable, capable of supporting cell proliferation and the subsequent synthesis of cartilage matrix components.88, 89 The chondrogenic capacity of these inks can also be enhanced by sulphating the alginate component.89 The sulfation of the alginate has been shown to maintain the phenotype of chondrocytes through activation of FGF (Fibroblastic Growth Factor) signaling.90 However, care must be taken when utilizing nanocellulose as a thickening agent. The resultant inks are highly viscous and it has been demonstrated that after extrusion through smaller diameter nozzles, excess shear stress can cause chondrocytes to lose their capacity for proliferation and synthesis of ECM components.89

Recently, a number of groups have started to explore the use of extracellular matrix (ECM) based bioinks.52, 91-93 It is believed that the incorporation of tissue specific ECM components into a bioink can provide environments more conducive to supporting specific cellular phenotypes. In one example, heart, fat and cartilage tissue were first decellularised and then solubilised at neutral pH at 4 °C to create an extrudable thermosensitive bioink.91 The solubilised ECM bioink could be stored at 4 °C in an extrudable form. Post-printing, the bioink filaments could be solidified by raising the temperature to 37 °C. It was found that the encapsulated cells remained viable post-printing and were capable of differentiating toward the lineage specific to the ECM used to form the ink. It was also possible to deposit the bioink within a polymeric framework to generate three dimensional tissue templates. An identical approach has been used by the same authors using a skeletal muscle derived bioink, further demonstrating the versatility of the approach.92 In another study, a chondroinductive bioink was developed by combining gellan, alginate and cartilage ECM particles.52 Highly accurate anatomical shapes could be printed with the ink and chondrocytes were capable of proliferating and producing matrix components within it (Figure 1c). In a novel scaffold-free bioprinting approach, modular cartilage tissue strands, which were fabricated by fusing tissue spheroids in a confining mold, were capable of being printed into 3D constructs using a robotic dispensing system.94 A summary of bioinks used in CTE to date is provided in Table 1 below.

2.2.2 Bioprinting of Heterogeneous Cartilage Tissues

Recently, a number of studies have explored whether bioprinting can be used to engineer cartilage tissues with regional distinctions in their composition. Spatially heterogeneous, anatomically shaped constructs have been bioprinted using a high density collagen bioink for CTE.61 The constructs were able to support cell growth and it was possible to spatially localise two different cell populations. In addition, it was also possible to modulate the regional mechanical properties of the construct by varying the bioink density. This is an important consideration for CTE as the stiffness of the underlying substrate can influence the chondrogenic capacity of MSCs, with stiffer matrices supporting a more transient hypertrophic phenotype.75 As mentioned previously, the composition of articular cartilage varies greatly with depth. In an attempt to recapitulate this structure, Ren et al. bioprinted a density gradient of chondrocytes in a single construct using a collagen type II bioink.34 The cell density gradient resulted in a graded distribution of ECM components, demonstrating the promise of such biomimetic approaches. A wide range of alternative approaches could be explored in order to engineer zonally organized cartilage tissues using bioprinting technology. These will be further discussed in Section 3.1.2.

2.3.1 Developing Bioinks for Bone Tissue Engineering

One of the most common natural materials used for hydrogel based BTE constructs is alginate.101, 102 As discussed previously, alginate is commonly used as a bioink, making it a promising candidate for bioprinting implants for bone regeneration.103, 104 For example, when implanted in conjunction with BMP-2, bioprinted alginate constructs can promote bone formation after 12 weeks in vivo.104 However, these studies used a high molecular weight alginate which is easier to print with as it has a higher viscosity than other molecular weight alginates. This increase in molecular weight also significantly decreases its degradability, which is a key consideration for BTE. Both studies saw significant amount of alginate still present after 12 weeks in vivo. Moreover, there was no tissue formation or vessel infiltration within the gel itself, only in the surrounding area. Other studies using alginate based constructs for bone regeneration have reported similar problems.105-108 This motivates the development of lower molecular weight alginate bioinks with faster degradation rates for BTE applications.108

In general, the leading limitation to the use of hydrogels alone for bone regeneration is their limited osteoinductivity and poor mechanical properties. This has motivated the development of composite bioinks with enhanced mechanical properties and osteo-inductivity. Wang et al. found that a bioink of both gelatin and alginate seeded with human adipose derived stem cells could induce bone matrix formation when subcutaneously implanted for 8 weeks in nude mice.109 Another study created a methacrylated gelatin bioink with encapsulated bone marrow MSCs and collagen microfibers bound to BMP-2. When compared to a bioink encapsulated with bone marrow MSCs alone and cultured in osteogenic medium, the bioink with bone marrow MSCs and BMP-2 bound collagen microfibers induced faster osteogenesis of MSCs compared to those cultured in the presence of osteogenic growth factors after 14 days in vitro.110 Campos et al. investigated a composite bioink of collagen and agarose, and found that by combining the thermos-responsive agarose hydrogels with collagen type 1, the mechanical stiffness significantly improved compared to collagen type 1 alone.111 Other studies have looked into adding bioglass particles to increase the mechanical stability of the hydrogel.112, 113 These studies have found that the addition of the bioglass can significantly improve the mechanical properties of the bioink113 whilst also producing a printable, porous material that can be used for generating BTE scaffolds.112, 113 Micro-carriers or small particles (100–400 µm) of polymers such as PLA (poly(lactic) acid) or polyethylene have also been shown to increase the mechanical strength of a hydrogel.114, 115 Leveto et al. investigated the effect that adding micro-carriers to GelMA had on the mechanical properties of the hydrogel. This enhanced the mechanical properties of the construct and the differentiation potential of human MSCs was also improved.116 Wüst et al. investigated a combination of gelatin, alginate, and hydroxyapatite and found that the Young's modulus was significantly increased if 8% hydroxyapatite was added to the hydrogel.117 However, even with a composite hydrogel the natural based hydrogels will still have a relatively low stiffness, in the kPa range, compared to bone which is in the GPa range.

Other investigators have explored the use of synthetic polymers within bioinks for BTE applications. Inkjet printing GelMA with added PEG significantly improved the resultant mechanical properties compared to GelMA constructs alone. Other studies investigated the addition of hydroxyapatite within a gelatin118 and polymer119 based bioink to increase the mechanical properties of printed scaffolds. These studies found that the mechanical properties of such scaffolds increased to within the range of trabecular bone of the same density (3.8–4 MPa). Although these printed constructs have improved mechanical properties, they are still lower than scaffolds produced using other rapid prototyping strategies such as fused deposition modelling (FDM). PLGA-PEG-PLGA,120 PLGA alone,121 PLA alone,122 PCL alone,123 and PCL-PLGA-TCP124 printed constructs have also been shown to have increased stiffness with moduli within the range of 57.4–244 MPa respectively, which is closer to the range for cortical bone.125, 126 This had led to increased interest in strategies where FDM can be integrated with microextrusion bioprinting to engineer composite reinforced tissues for BTE. This will be discussed in Section 2.3.

2.3.2 Bioprinting and Vascularisation in Bone Tissue Engineering

Ensuring the successful vascularisation of TE constructs is arguably the most challenging hurdle to overcome in the field of BTE today.127 As a result, there has been increased interest in using bioprinting technology to accelerate vascularisation of tissue engineered constructs for BTE. For example, a number of studies have explored the importance of the architecture of TE constructs on vascularisation and subsequent bone formation in vivo.56, 103, 128 For example, it has been demonstrated that the incorporation of microchannels into MSC seeded alginate hydrogels, achieved using bioprinting technology, could enhance vessel infiltration compared to a solid non-porous controls after 14 days in vivo.103 Other studies, have used bioprinting to allow for spatial presentation of specific growth factors to enhance angiogenesis. Park et al. printed a PCL fiber based construct, with dental pulp stem cells and BMP-2 encapsulated in type 1 collagen deposited on the periphery, and a composite of gelatin and alginate with VEGF deposited in the center (Figure 2a). Microvessels were newly formed in the center of the printed construct, with angiogenesis from the host tissue was also observed. Interestingly, vascularisation was enhanced by localizing VEGF presentation to central regions of the construct (Figure 2b). In another study, hierarchical vascularised bone biphasic constructs were bioprinted using a novel dual bioprinting approach.129 Regional localisation of VEGF and BMP-2 was achieved using a novel thiol-ene click reaction which resulted in formation of a vascular network within the implant. In another approach, it was demonstrated that bioprinted scaffolds that facilitated the sustained long-term release of VEGF from a matrigel/alginate bioink could be used to enhance vascularisation in vivo.130

2.4.1 Fused Deposition Modelling (FDM)

FDM is an additive manufacturing technology used for production, prototyping and modelling applications. As cells cannot be incorporated during the printing process due to high processing temperatures FDM is not technically considered bioprinting, however it is commonly used for producing porous scaffolds for TE.133 FDM printers use a thermoplastic filament which is heated above its melting point and extruded onto a platform in a layer-by-layer process. The key advantage of FDM is that scaffolds with highly interconnected pore geometries and channels sizes can be fabricated rapidly. By varying process parameters such as the extrusion pressure, nozzle diameter and deposition speed it is possible to print scaffolds with a wide range of filament diameters and porosities.134, 135 Another key advantage of these scaffolds is that they are mechanically strong, possessing mechanical properties in the range of articular cartilage and cancellous bone.25, 134-138

2.4.2 Bioprinting of Reinforced Constructs for Cartilage and Osteochondral Tissue Engineering

It has been shown that PCL and cell containing bioinks can be co-printed in a layer-by-layer fashion to build 3D constructs for CTE (Figure 3a).153 Even though the PCL component is heated to approximately 60 °C to facilitate extrusion, cells in the bioink phase remain viable post-printing.85 The mechanical properties are significantly improved with the incorporation of PCL filaments and by modulating the percentage of reinforcing polymer, constructs with compressive equilibrium moduli in the range of articular cartilage can be achieved.85 A subcutaneous analysis of a PCL reinforced alginate construct embedded with chondrocytes and supplemented with chondrogenic growth factor TGF-β3 reported collagen type II and sGAG deposition.54 In a similar approach, bi-layered constructs with spatially distinct regions of ECM materials and growth factors were created for osteochondral tissue engineering.154 Human MSCs were combined with atelocollagen and BMP-2 in the bone region, and hyaluronic acid and TGF-β3 in the cartilage region. The construct was reinforced using a PCL frame and demonstrated heterogeneous tissue development in an osteochondral defect model. With such approaches there is little or no mechanical interaction between the reinforcing component and the bioink network. This can be a problem, particularly at higher strains, which can lead to mechanical disintegration of the implant upon the application of mechanical loads. Recently, it has been demonstrated that by covalently crosslinking the hydrogel and thermoplastic phases together it is possible to increase the interface strength.132 This resulted in improved resistance to repetitive rotational and axial loading. It should be noted here that co-printing approaches can influence the mechanical properties of the final construct. For example, layer-by-layer bioink deposition can interfere with the adhesion of PCL fibers during co-printing which can significantly reduce the resultant mechanical properties of the engineered construct.155

The degradation rate of PCL, which can be up to 2–3 years,157, 158 is a potential limitation with such multi-material approaches, as residual filaments can act as a barrier to tissue formation. Recently alternative polymers based on PCL have been developed to overcome these potential limitations. A poly (hydroxymethylglycolide-co-caprolactone) (PHMGCL) polyester which has a greater hydrophilicity than unmodified PCL due to the addition of hydroxyl groups has been developed which loses ≈60% of its initial weight after 3 months in vivo.147, 159 An extensive review on the use of PCL for additive biomanufacturing and other biomedical applications is available elsewhere.158 An alternative approach is to use a faster degrading polymer such as PLGA. PLGA co-polymers composed of PLA and Poly(glycolic acid) (PGA) can be developed which degrade more rapidly due to the hydrophilic nature of PGA which facilitates hydrolysis of the polymer backbone.160 The rate of degradation can be controlled by varying the ratio of lactide and glycolide in the copolymer, with higher percentages of glycolide resulting in accelerated degradation rates.160 One drawback of using PLGA is that the build-up of acidic by-products that occur following its degradation can cause adverse inflammatory responses.161, 162

Osteochondral constructs have also been fabricated using a co-printing approach with PLGA rather than PCL.163 Here, the chondral layer was formed using an alginate bioink combined with cartilage derived ECM and an alginate/hydroxyapatite bioink was used for the osseous layer. Another way to overcome limitations with residual PCL material is to reduce the amount of the reinforcing polymer used by increasing the porosity of the reinforcing phase. This has led to an increased interest in the use of melt-electrowriting (MEW) which is an emerging technology that combines key aspects of melt-electrospinning and FDM. The process is similar to FDM except that a voltage is applied between the nozzle tip and the printing platform to draw filaments of material from the nozzle tip onto the platform. By carefully controlling the relative motion between the printhead and the collector it is possible to print the fibers with a high degree of accuracy. Highly organized networks of fibers with diameters down to 0.8 µm have been printed with this technology.164-166 This is a major advantage over traditional FDM printing where it is difficult to print fibers with diameters lower than 100 µm. MEW has recently been used to reinforce cell laden GelMA hydrogels with highly porous PCL scaffolds (porosity 98%–93%, fiber diameter 19–50 µm).33 The stiffness of the resultant composites were within the native cartilage range and importantly the yielding strains (≈25%–40% strain) of the composites were much higher than equivalent scaffolds produced using FDM (≈8%). This is an important consideration for CTE as strains of greater than 10% are repeatedly experienced during locomotion.167 In addition, cyclic mechanical tests demonstrated that the scaffold could recover after 20 cycles at 20% strain. Composite soft cartilage constructs have also been fabricated by combining PCL microfibers produced using MEW and a highly negatively charged star-shaped poly(ethylene glycol)/heparin hydrogel (sPEG/Hep)156 (Figure 3c). The hydrogel network mimicked the function of the cartilage proteoglycan network and the PCL fibers mimicked the function of the cartilage collagen fiber network. The resultant constructs exhibited mechanical anisotropic, nonlinear and viscoelastic behaviour analogous to native cartilage.

Other hybrid strategies include incorporating inkjet printing with electrospinning.99 A PCL electrospun mat was alternated with a fibrin-collagen solution containing chondrocytes for 5 layers until a thickness of 1 mm was reached (Figure 3b). The fabricated constructs formed a cartilage matrix both in vitro and in vivo as evidenced by the deposition of type II collagen and sGAG.

2.4.3 Bioprinting of Reinforced Constructs for Bone Tissue Engineering

Co-printing approaches have also been explored for BTE.56, 128 In one approach, a cell laden bioink containing adipose derived stem cells was co-deposited alongside a mixture of PCL and tricalcium phosphate (TCP).128 The constructs were implanted in cranial defects in rats and demonstrated good integration and evidence of bone repair. In a similar approach, vertebrae shaped constructs were co-printed using PCL and alginate bioink containing MSCs56 (Figure 4a). The constructs were chondrogenically primed in vitro and once implanted in vivo, were capable of forming vascularised bone organs via endochondral ossification (Figure 4b). The study presents a novel biofabrication strategy for engineering whole bone organs by bioprinting developmentally inspired templates with the capacity to undergo endochondral ossification over time following implantation. Rather than trying to print the complex final bone organ, the relatively simple developmental precursor was printed. This concept of bioprinting developmental precursors could also be used to engineer other complex solid organs.

3.1.1 Biofabrication of Mechanically Functional Cartilage Tissues

Since bioinks are often hydrogel based they are generally mechanically weak and not suitable for deployment in load bearing locations. While significant mechanical reinforcement can be achieved by co-printing thermoplastic polymers,85 it is not clear whether such implants will be sufficiently tough to withstand long term repetitive loading in a joint environment. For example, scaffolds produced using FDM typically permanently deform at strains levels above 8–10%.33 In addition, it has been demonstrated that polymer reinforced hydrogels disintegrate at the boundary between the two phases under physiological loads.132 This problem can be at least partially addressed by improving the interface binding through chemical modification of the materials.132 However, it is still unclear whether these modifications will be sufficient to withstand repetitive long term loading in a joint environment. Typically, the mechanical properties of tissue engineered cartilage are only measured in compression, and other potential failure modes like shear, are usually ignored. High levels of shear, tension and compression will be experienced in the joint and implants that can withstand repetitive combinations of these loading patterns will be required. Bioreactors models of the joint, that can mimic joint loading patterns, should be used as a checkpoint before attempting to progress into large animal models.132, 175

Another promising approach to developing mechanically functional implants for joint resurfacing could involve the development of tough bioinks based on interpenetrating network (IPN) hydrogels. IPNs are a class of materials formed by combining multiple polymer networks. By combining a suitably contrasting primary and secondary network that can support IPN entanglement and energy dissipation, it is possible to engineer extremely tough and flexible hydrogels with mechanical properties comparable to high load bearing tissues.176, 177 Recently, a double network bioink, formed by combining PEG and alginate, was combined with a nano-silicate clay to create a mechanically tough ink with mechanical properties in the range of native articular cartilage.178 These approaches will likely find utility in the future of cartilage tissue bioprinting. The challenges will arise in balancing the mechanical and biological functionality of the implant as IPN networks are typically dense making it difficult for embedded cell populations to produce de novo matrix components.

3.1.2 Bioprinting of Stratified Cartilage Tissues and Osteochondral Tissue Interfaces

As mentioned previously, articular cartilage is a highly anisotropic tissue and its composition and organization varies greatly with depth. Surprisingly, few studies have explored the possibility of bioprinting gradients of biological factors in order to recreate the zonal properties of articular cartilage and its interface with the underlying bone. Now that a wide range of suitable bioinks are available, future work should explore how combinations of biological cues can be used to engineer biomimetic tissue gradients for CTE. During tissue development, repair and homeostasis cells experience and respond to gradients of chemical and physical cues. These gradients can influence a number of different cellular behaviors including proliferation, migration and differentiation. In a tissue engineering context, physical gradients such as pore size and substrate stiffness, or biochemical gradients such as growth factor presentation, can be introduced into a construct. It has been demonstrated that biochemical gradients, such as the graded presentation of growth factors, can lead to the development of heterogeneous engineered tissues. For example, an overlapping graded presentation of recombinant human bone morphogenic protein 2 (rhBMP-2) and insulin-like growth factor (rhIGF-I) has been shown to direct localized osteogenic and chondrogenic differentiation of MSCs in silk scaffolds.179 In addition, it has been demonstrated that gradients of TGF-β in MSC seeded hydrogels, which result from a combination of high-affinity binding interactions and a high cellular internalization rate, can lead to the development of highly heterogeneous cartilage tissues.180

These studies all highlight the benefits of incorporating biochemical and physical gradients into cartilage and osteochondral tissue engineering strategies. Surprisingly, given the additive layer-by-layer nature of biofabrication technology, relatively few studies have explored the possibility of developing gradients using bioprinting approaches. In an early approach, overlapping gradients of IGF-II and BMP-2 have been deposited onto a fibrin substrate and shown to spatially direct the fate of C2C12 cells toward an osteogenic lineage.181 Future work should look to expand on these approaches to direct the development of both stratified cartilage tissues and the osteochondral interface. For example, combinations of physical and biochemical gradients could be used to engineer more native like cartilage tissues. As an example of this, a novel dual syringe system was developed to produce a combination of physical and chemical gradients, including substrate stiffness, RGD ligand presentation and growth factor concentration, to control differentiation of MSCs along chondrogenic and osteogenic lineages.182 In another multifactorial approach, it was demonstrated how combinations of growth factors (BMP-7, IGF-1 and hydroxyapatite), substrate stiffness values (80 KPa, 2.1 MPa and 320 MPa) and nanofiber alignments (horizontal, random, and perpendicular to the gel surface) could be used to direct the zone specific chondrogenic differentiation of MSCs.183 Bioprinting technology could be used to further expand on these ideas to engineer cartilage tissues with a native like organization. Future advances in bioprinting technology will likely make it possible to present multiple gradients of physical and chemical cues, across multiple length scales. For example, in a recent ground breaking study, a microfluidic bioprinting system capable of simultaneous spatial deposition of 7 distinct bioinks was developed.184 The system was able to generate complex gradient structures highly suitable for tissue engineering applications.

The choice of cell source is also critical for cartilage tissue engineering with chondrocytes and MSCs being the most widely used.185 An alternative approach could involve bioprinting gradients of cell types/sources in order to engineer stratified cartilage tissues. For example, co culture systems, composed of these cell types, have shown promise in cartilage tissue engineering and could be combined in graded ratios in an attempt to engineer more stratified tissues.173, 186, 187 Alternatively, zonal specific chondrocytes could be isolated and printed in a layered fashion to recapitulate the depth-dependent properties of the tissue.21, 188

3.1.3 Towards Biofabrication of Anatomically Accurate Osteochondral Tissues

The layer-by-layer nature of bioprinting technology makes it easy to create anatomically accurate constructs for tissue engineering applications.189 Imaging techniques such as magnetic resonance imaging (MRI) ad computed tomography (CT), that are widely used in the clinic, can be easily converted into formats compatible with biofabrication technology. Patient specific tissue engineered implants could therefore become a reality for sufferers of joints diseases such as OA. However, a number of challenges exist with current bioprinting software that must be addressed. Traditional biofabrication technology creates 3D constructs by continually adding 2D x–y patterns into the z plane. As a result, fibers are orientated parallel to printing platform and the outer surfaces of curved constructs are jagged and discontinuous in nature. This can make it challenging to create constructs with smooth outer surfaces. This is a key consideration for CTE as the main function of the tissue is to support smooth, pain-free articulation. Future work should explore strategies to orientate printed fibers in anatomically relevant orientations. Such capabilities could also lead to more biomimetic mechanical behaviour of resultant tissue engineered implants.

3.3.1 Engineering Zonally Organized Interface Tissues Using Nucleic Acid Delivery

The development of spatially controlled gene delivery systems in combination with biofabrication techniques could also offer a very tunable strategy for interface TE. The cell-mediated expression of the gene of interest could generate precisely controlled gradients of the gene product to enhance local cell differentiation or maintenance of a desired phenotype (Figure 6d), avoiding the use of supra-physiological concentrations of recombinant growth factors and associated side-effects.217 Bi-phasic non-viral gene delivery in an oligo poly(ethylene glycol) fumarate (OPF) scaffold loaded with pDNA encoding for RUNX2 in the osteo layer and for the SOX trio in the chondro layer, showed increased subchondral bone formation in the osteo layer but failed to develop a well-integrated cartilage layer in an rat osteochondral defect model.218 Spatial distribution of pDNA encoding for either TGF-β1 or BMP-2 in a MSC-laden, multiphasic collagen-based scaffold showed simultaneous cartilage and subchondral bone regeneration in a rabbit osteochondral defect.15 Although the described studies highlight the potential of spatial gene delivery for the regeneration of complex interface tissues, 3D bioprinting might solve the limitations of traditional tissue engineering associated with poor layer integration, the scalability of the approach and the tissue organization present in the repair tissue. In addition, more sophisticated patterns of genetic material could be achieved using bioprinting technology in order to recapitulate the body's natural developmental and regenerative pathways.