Controlling and tuning the mechanical properties of fabricated scaffold via AM techniques that are comparable to native tissue is another pivotal challenge. For example, complex tissues, such as heart, muscle, cartilage, skin, etc., are soft and flexible structures, but they are tough enough to withstand high stresses without any destruction.
4.1. Reinforcement Mechanism in Melt Electrowritten Fiber-Hydrogel Composites
Hydrogels are excellent candidates for scaffold fabrication with the potential for resembling the microenvironments of human body and encapsulation of different cells in their highly hydrated structures. Their tunable physicochemical properties, such as growth and differentiation factor ingredients, could strongly affect cellular behavior [104
]. However, their structures are not as mechanically strong as the ECM of soft tissues, including fibrous proteins. Their lack of mechanical instability also limits the proper cellular functionality. The enhancement of mechanical strength can be achieved by increasing the concentration of polymer content or the crosslinking degree in the hydrogel, which may negatively affect cell viability, proliferation, migration, and differentiation [104
]. The incorporation of a hydrogel within geometrically varied micro-fibers produced by MEW has fulfilled the required mechanical properties to mimic the function of fibrous ECM of soft tissues [84
]. Mainly, the construction procedure of hybrid hydrogel-MEW composites has two steps, which are the fabrication of fibers via MEW and infiltration of hydrogel in manufactured fibers [84
The mechanism behind reinforcement of hydrogel matrices by melt electrowritten fibers was investigated in different studies with several hypotheses. Visser et al. [109
] elucidated the reinforcement mechanism while using gelatin methacrylate (GelMA) hydrogel infiltrated into highly arranged networks of PCL microfibers. The composite was manufactured for musculoskeletal tissue engineering application, and the reinforced structure showed a significant increase in stiffness compared to hydrogel structure. It was revealed that the reinforced structure’s mechanical strength was similar to that of native cartilage tissue. The hydrogel reinforced with fiber structure showed higher stiffness, rather than hydrogel scaffold and, interestingly, higher than the fiber scaffold without hydrogel. This result demonstrated the synergistic effect of reinforcing the composites. Mainly, lateral expansion of the hydrogel leads to the conversion of axial loads into lateral loads, which can be covered by fiber networks as tension in hybrid composites. Therefore, an increase in stiffness of the composite was correlated with horizontal expansion of the hydrogels while applying stress to the neighboring fibers under tension. Moreover, comparing the compressive loading responses also assessed the effect of fiber diameter on the stiffness of the composites. Low and high fiber diameter networks were manufactured via MEW and FDM, respectively. The composite structure with high fiber diameter showed similar stiffness with the structure without hydrogel. This means that the axial loads did not cause an elongation of the thick fibers. On the other hand, for the composites with thin fibers, the fibers that were elongated as a response to axial loading and hydrogel supported the structural integrity. In addition, the stiffness difference was observed between the groups of fiber networks with and without hydrogels. These observations indicated that a synergistic reinforcement effect was only observed only in the composites with small fiber diameters that were manufactured via MEW. The stiffness of the composite after compressive loading was further demonstrated through a mathematical model [109
]. To calculate construct stiffness, fiber radius, the number of fibers and elastic modulus of the polymer were used as directly proportional variables, while the axial strain of fiber and composite and construct radius were used as reversely proportional ones. The mathematical model revealed that hydrogel expands with axial compressive loading and causes exposure of the MEW fibers to tensile loads. However, theoretical stiffness value was calculated larger than the experimental one. It demonstrates the complexity of theoretical modeling of polymer-hydrogel composites, which demands more in-depth studies.
Bas et al. stated a similar hypothesis for reinforcement mechanism of hydrogels by MEW fiber network [48
]. As stiffness of the composite relies on the MEW fiber networks, a detailed study was performed by controlling fiber spacing of 400 µm and 800 µm, and grid patterns with 0–90° and 0–60–120° orientations. The constructed melt electrowritten PCL fiber networks were infiltrated with GelMA, and GelMA/hyaluronic acid-methacrylamide (HAMA) hydrogels and the stiffness of the structures were evaluated. It was proposed that the hydrogels had significant lateral to axial strain ratio due to high Poisson’s ratio values. However, the composites showed low Poisson’s ratio, since highly organized fiber networks suppressed lateral deformation of the hydrogels. This synergistic reinforcement mechanism was the same as that reported by Visser et al. [109
In another study, the high order finite element method was used to simulate the mechanical characteristics and elastic modulus of composites [110
]. Composites with varying fiber spacing were used for the analysis. Simulation analysis revealed that decreasing the fiber spacing increased the compressive moduli of the composite due to higher reinforcing filler ratio, which was a similar synergistic reinforcement mechanism with the aforementioned studies [48
]. The simulation results presented higher stiffness value for fiber networks as compared to experimental data, although the experimental and theoretical data for hydrogel alone and fiber-reinforced hydrogel samples were similar.
Castilho et al. performed two different finite element (FE) analyses in order to investigate the mechanism behind the reinforcement of hydrogel through fiber networks [111
], as summarized in Figure 4
. In this regard, melt electrowritten PCL network with different fiber spacing and GelMA hydrogel were manufactured. Subsequently, a compression test was performed to obtain stress-strain data that were used as an input for FE analysis. Afterward, a melt electrowritten fiber network was manufactured, and GelMA hydrogel was infiltrated into its gaps, and the FE analysis results were validated with experimental data. In the first analysis, continuum FE model was examined by employing the idealized geometry of the composite, which is unidirectional lamina. Continuum FE model exhibited the expansion of hydrogel inside the composite. Besides, the diminishing effect of fiber network on hydrogel movement was observed. For individual hydrogel and fiber network structures, continuum the FE model results were similar to the experimental data in terms of compressive stress-strain behaviors. However, theoretical stiffness of the composite with higher fiber volume fraction was significantly lower than the experimental data in this model, although the stiffness values for composites with lower fiber volume fraction were similar.
In second analysis, micro-FE model was performed by employing the micro-computed tomography (µ-CT) images of the composite’s geometry. The micro-FE model presented similar results with the experimental data in terms of deformation of fiber scaffold and composites. The results indicated that addition of hydrogel into fiber scaffold increased stiffness of the overall composite several folds due to the prevention of fiber network buckling through the resistance of hydrogels. Several inferences were made from those FE models. Continuum FE model stated that the reinforcement of the composites with low fiber fraction volume was governed by lateral expansion of the hydrogel, which put the fibers under tension. This hypothesis was similar to the previously mentioned studies. However, those studies did not consider different fiber spacing while explaining the mechanism. On the other hand, micro-FE model underlined the significance of load transfer through the interconnecting regions of the fibers. Reinforcement of the composites with high fiber volume fraction through buckling inhibition by the resistance of hydrogels was also highlighted. Thus, FE analysis would be effective in optimizing the structure’s architecture based on desired mechanical properties.
As an alternative to the heuristic approach, which relies on experimental trials, numerical modeling could be employed for design of the architecture and optimization of manufacturing parameters. In this regard, Bas et al. introduced the design of soft network composite with different mechanical and biological characteristics by modeling and manufacturing the composite, accordingly [112
]. By provided compressive modulus and Poisson’s ratio value of the hydrogel as an input to numerical model, compressive modulus of the composite was calculated. Based on the numerical results, 0–90° grid patterns with varying pore size and fiber thickness were determined for the fiber network design. Theoretical compressive modulus value obtained from the numerical model was validated by comparing the experimental results with that of different zones of the articular cartilage tissue model having different mechanical features. For this aim, the PCL fiber network with different fiber thicknesses and pore size was manufactured via MEW, and GelMA hydrogel was filled within the scaffold gaps, as shown in Figure 5
. The initial layers of scaffold were printed with the PCL fibers, including hydroxyapatite nanoparticles (nHA), in order to mimic calcified zone, which is present between the native articular cartilage tissue and subchondral bone. Reinforcement of the hydrogels by fiber network was determined through uniaxial compression test. Compressive modulus values obtained by numerical modeling were in agreement with the experimental mechanical testing results.
4.2. Biological and Mechanical Aspects of Reinforced Composites in Different Tissue Engineering Applications
Depending on mechanical requirements of the target tissue, such as stress-strain relations, anisotropy, viscoelasticity, and flexibility, different improvements have been made on the MEW fiber-hydrogel composite. Within this framework, fiber networks with varying polymer types, fiber thicknesses, fiber spacing, and geometries have been designed and combined with several hydrogels for different tissue engineering applications.
In a simple approach, MEW PCL scaffold’s porosity and crosslinking degree of GelMA was evaluated based on the composite stiffness and recovery for articular cartilage tissue [109
]. Chondrocytes that were encapsulated within GelMA hydrogel were homogeneously distributed throughout the PCL construct. The cells kept their spherical morphology and showed enhanced viability within the reinforced GelMA hydrogels. According to quantitative reverse transcriptase-polymerase chain reaction (PCR) analysis, a physiological compressive loading of 20% strain and 1 Hz induced the up-regulation of expression of genes encoding the ECM proteins of chondrocyte.
In another study, to recapitulate viscoelastic and stress relaxation characteristics of the articular cartilage, high negatively charged star-shaped poly(ethylene glycol)/heparin (sPEG/Hep) was used as hydrogel and then reinforced with the PCL fiber network having a 0°–90° grid pattern with different pore sizes that emulated collagen fibers in terms of anisotropic and nonlinear features [110
]. Treating the surface with NaOH treatment increases the wettability of the fiber network.
When compared to alone sPEG/Hep hydrogel and MEW fiber network, the compressive modulus was several folds higher in fiber-reinforced hydrogel composite. Since the compressive modulus of the composite was measured as being higher than the summation of compressive modulus of hydrogel and fiber network alone, which indicated the synergistic reinforcement effect. Moreover, composite structure exhibited similar viscoelastic nature of the articular cartilage. Similar to the results of compressive modulus, only the fiber-reinforced hydrogel composite exhibited similar stress-relaxation behavior with the human articular cartilage, and the enhancement of ECM protein expression under hydrostatic pressure was observed [110
Another study that was related to articular cartilage tissue engineering was conducted by employing reinforcement of hydrogels through melt electrowritten bi-layered microfiber network [113
]. Different zones of articular cartilage tissue were resembled by two layers designed with different fiber patterning strategy in order to obtain zonal mechanical characteristics of the native cartilage tissue. The GelMA hydrogel was cast into the PCL fiber network, which was made up of dense structure with 0–45–90–135° crossed diagonal pattern mimicking a superficial tangential zone (STZ) and uniform 0–90° box structure mimicking middle and deep zone (MDZ) of the articular cartilage. A significant mechanical strength difference was observed between the reinforced and non-reinforced hydrogel structures. The presence of a STZ layer enhanced the compressive modulus several folds as compared to the composite without STZ layer as the presence of STZ layer provided distribution of compressive load to the whole composite uniformly. Human chondrocytes encapsulated in GelMA hydrogel reinforced with bi-layered PCL fibers were tested under two different conditions: static conditions with chondrogenic differentiation factor and mechanical strength induced condition without addition of any differentiation factor. The results revealed that both conditions showed cartilage differentiation by the similar production of sulfated glycosaminoglycan (GAG) and collagen II, which indicated the activation of signaling factors under proposed mechanical condition.
The reinforcement of hydrogel by fiber networks was also used for a variety of other soft tissues. Fiber networks with different geometries have been constructed to meet the requirements of different mechanical properties of soft tissues. For instance, since soft tissues function under high tensile loads, a flexible soft tissue bearing high tensile loads were manufactured by embedding stretchable curvy PCL fiber networks into a hydrogel for tensile load-bearing, mechanical anisotropy, and flexibility [114
]. A combination of poly (ethylene glycol) diacrylate (PEGDA), GelMA, and alginate were employed as hydrogel matrix. Since the transfer of the applied loads depends on the interaction between soft matrix and reinforcing fibers, the surface of the fibers was functionalized by photo-crosslinkable acryl groups, which provided covalent bonding between fiber network and PEGDA and GelMA hydrogels. The maximum tensile strain value that the composite could handle before irreversible deformation was recorded as being similar to the values of native soft tissues. After yielding point, manufactured fiber network elongated rather than being ruptured. Stiffness of the composite was several folds higher than the stiffness of hydrogel without reinforcement. Encapsulating human bone-marrow-derived mesenchymal precursor cells (hBMPCs) into the hydrogel assessed the biocompatibility of the soft network system. The encapsulated cells infiltrated into the curvy PCL MEW fibers were found to be viable after 72 h of incubation period.
Cardiac tissue is another complex soft tissue with a highly organized fibrillar structure and mechanical strength. MEW manufacturing method offers great potential to establish an environment with a highly organized structure for cardiac tissue engineering by providing sufficient mechanical strength. Based on this idea, Castilho et al. systematically investigated the effect of hydrophobicity and the morphology of melt electrowritten PCL fibers on collagen reinforcement and the response of cardiac progenitor cells (CPCs) for potential therapeutic applications [84
]. Due to the drawbacks of PCL, such as hydrophobic nature and slow degradation, hydroxyl-functionalized polyester, (poly (hydroxymethylglycolide-co-ε-caprolactone) (pHMGCL) was employed. Fiber scaffolds with rectangle and square geometries were manufactured while using two different formulations of polymers, which are PCL and pHMGCL-PCL blend. The fibrous scaffolds were evaluated with and without hydrogel forms. In case of fibrous networks without hydrogel, the tensile modulus of the blended-PCL scaffold was lower than that of scaffolds manufactured from PCL, but both types of scaffolds were in the similar range with the tensile modulus of human myocardium. However, those scaffolds showed different tensile modulus values that were based on their shapes. The rectangular shape exhibited anisotropic behavior of the native cardiac tissue. Even though tensile modulus and anisotropy of the scaffold showed almost similar mechanical characteristics to the native cardiac tissue, long-term mechanical features need to be investigated by employing cyclic tensile loading and other mechanical tests. In the case of composite structures, both blended and bare-PCL scaffold allowed collagen matrix infiltration in a uniform distribution, while collagen hydrogel with no support of PCL fiber scaffold remodeled into a clumped form. The analysis of CPCs in collagen hydrogel within both of the fiber scaffolds that were made up of PCL with and without blending with pHMGCL showed that the cells were 99% viable, regardless of the geometry and hydrophobicity of the fibers. However, cellular alignment and morphology changed, depending on geometry of the scaffold. Cells were randomly arranged in squared geometry scaffolds while the alignment of the cells was more regular, particularly in a preferential direction on rectangular ones. The number of aligned cells on the blended-PCL scaffold was found more than PCL scaffold, as shown in Figure 6
. The nucleus of CPC’s in the rectangular blended-PCL composites promoted less circular shape when compared to those in the square-shaped ones. In addition, it was observed that the interconnected porous structure also allowed the interaction of the cells between the compartments. This study revealed that hydrophobic nature of fibers and geometry of the scaffold strongly affect the interference between matrix and reinforcing fibers, which is reflected by cellular behavior. The authors also noted that pHMGCL might change the electrical stimulation on the fibers, which can also have a positive effect on CPC’s response.
A physiologically relevant 3D environment that was fabricated from melt electrowritten fibrous composite could be considered for treatment of myocardial regeneration. Generally, the treatment can be followed by replacement of the died cells during myocardial infarction with the healthy cells. Since induced-pluripotent stem cells (iPSCs) have the capability of generating large numbers of human cardiomyocytes, they are the target for direct transplantation, which showed promising results in non-human primates. However, they are not effective due to limited contractility and electrical instability.
Complex fibrillar architecture and mechanical characteristics of myocardial tissue were intended to be recapitulated for the production of heart patch in another study of Castilho et al. [118
]. For this purpose, fiber networks with the hexagonal and rectangular structure were manufactured via MEW and then seeded with human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CM), which was encapsulated within a collagen-based hydrogel matrix. Under in-plane tensile loading conditions, biaxial deformation characteristic of the hexagonal fiber scaffold was found to be superior to the rectangular-shaped scaffold. Although the results were better than the rectangular fiber scaffold, deformation, and fatigue behavior of the hexagonal fiber network was poor in the y-direction. The fatigue behavior of the hexagonal fiber scaffold after cyclic tensile loading was more promising than the one in the case of the rectangular fiber network. IPSCs encapsulated in a collagen-based hydrogel were used to monitor the effect of different geometrical composite structures on cellular behavior. The cells were aligned through the fibers and they showed synchronous contractions throughout the scaffold. The cells in the hexagonal fiber network showed faster contractions and maturation of contractile myocytes. In vivo studies followed by injection of the maturated cardiac patch onto a beating, the porcine heart did not reveal harmful effects on cell viability and on integrity of the engineered construct. It was proposed that fabrication of PCL fiber network with hexagonal microstructure improved the biaxial deformation and compliance.
Another different biomimetic design and fabrication strategy was proposed for heart valve tissue engineering (HVTE) in the study carried out by Saidy et al. [116
]. The biomechanical characteristics of heart valve leaflets were aimed to be achieved by highly organized wavy-like PCL melt electrowritten fiber scaffold to mimic the collagen fibers that are present in the structure of the heart valve. Human vascular smooth muscle cells (HUVSMCs) were seeded to the scaffold either directly onto the scaffold or by encapsulating in fibrin hydrogel, followed by casting to the scaffold. The mechanical features of the manufactured construct showed similar results with native tissue in terms of J-shaped stress-strain curve behavior, anisotropy, and viscoelasticity. HUVSMC in the homogenously formed composite sustained the viability and exhibited the biochemical and mechanical properties of the main ECM component of native heart valve leaflet with the enhanced synthesis of collagen type I and type III. In the aforementioned studies on cardiac tissue engineering, uniaxial tensile tests were only performed on the fiber scaffolds without hydrogel matrix. Although it would be speculated that the main contributor to the mechanical properties of the composite is fiber network, investigation of the fiber network with hydrogel matrix due to the synergistic effect of the hybrid structure would reveal valuable understanding.
There are other reports that consist of melt electrowritten fibers and hydrogel composites. In those studies, although their potential to recapitulate tissue functions was reported, their mechanical properties were not explored. For example, Hutmacher et al. presented an approach for periosteum regeneration [118
]. Periosteum, which is the reservoir for vascular components, and bone-forming cells has a critical role in the regeneration of complex multiphasic system. The periosteum is mechanosensitive tissue that is exposed to shear and traction loads during movement and based on the subjected force it regulates cell proliferation and differentiation. This multiphasic structure was mimicked as in a composite of melt electrowritten PCL tubular scaffold seeded with human bone marrow mesenchymal stem cells (BM-MSCs) and sPEG/Hep hydrogel system that was loaded with human umbilical vein endothelial cells (HUVECs) as a target for large bone defect repair. BM-MSCs were used as bone-forming cells, while the HUVECs cells were utilized for vessel formation. sPEG/Hep hydrogels were modified with RGD and loaded with VEGF. After the hydrogel infiltration in the tubular PCL scaffold, the cells were cultured in angiogenic conditions for seven days of in vitro culturing. Subsequently, the composites were implanted in femur side of mice. The explanted constructs after 30 days from implantation showed that HUVEC cells were not proliferated, but formed capillary-like structures and BM-MSC kept proliferation without differentiation. In addition, it was observed that the implanted human cells were gradually replaced by the host cells. This study demonstrated the efficiency of multiphasic hybrid design with a combination of different human cell types in the periosteum tissue engineering concept.
In another study, the effect of the composition of soft matrix component and photoinitiator type for crosslinking within PCL microfibers on the re-differentiation of chondrocyte was investigated [119
]. GelMA type A or type B that were mixed with hyaluronic acid (HA) were photocrosslinked with either lithium acylphosphinate (LAP-visible light; 405 nm) or Irgacure 2929 (IC2929-at UV light; 365 nm). Interestingly, it was found that PCL fiber reinforcement increased the expression of ECM components of chondrocytes in hydrogel constructs. GelMA from type B photocrosslinked with IC2929 enhanced the formation of cartilage-like tissue as compared to others through promoting increased GAG production, which showed similar compressive strength as native articular cartilage.
The MEW-hydrogel hybrid manufacturing approach has also allowed for the development of 3D in vitro tissue models, such as neural and tumor cultures [117
]. For example, Glycine receptor transfected (GTR; ligand-gated ion channel) Ltk-11 cells encapsulated in Matrigel reinforced with melt electrowritten PCL scaffold. This model provided information regarding the functioning of the ion channels with electrophysiological measurements from a physiologically similar 3D system [120
]. In another study, patient-derived ovarian cancer cells were encapsulated in PEG hydrogel infiltrated into PCL fiber scaffold. It was used to mimic the tissue environment for the investigation of malignant behavior and tumor-promoting signals [117
]. Designed models showed clinically similar gene expression of ovarian cancer with the samples that were obtained from a high-grade serous ovarian cancer patient. These models may be helpful for screening the possible medicines or a combination of therapies and design for other clinical trials.
It is clear that the MEW-hydrogel hybrid system is a promising technique for RM, TE, and disease models, yet two-step fabrication of the hybrid structures limits the control over-structure design and fiber writing. To overcome this limitation, Ruijter et al. assessed a single-step construction approach by MEW of PCL and extrusion bioprinting of eMSC-laden GelMA hydrogel [105
]. This approach allowed for the precise deposition of cells in a mechanically stable composite with controlled porosity and pore shape. It was confirmed that the applied high voltage during construction did not affect either the cell viability and metabolic activity or the ability to differentiate toward multiple lineages of eMSC.