Controlled Fabrication Methods for Tissue Engineering Constructs

David B. Gehlen, 1, Laura De Laporte, 1

DWI – Leibniz-Institute for Interactive Materials, Aachen, Germany

Material Matters, 2018, 13.3


Organ failure is a major health issue that affects millions of patients annually and costs of hundreds of billions US Dollars. For the last 30 years, scientists have been combining the tools, methods, and molecules of engineering, biology, chemistry, and physics in order to fabricate new tissues for the restoration or replacement of functional organs.1 This field, known as tissue engineering, has three main subfields: i) ex vivo generation of implantable tissue using material scaffolds with cells and growth factors, ii) biohybrid materials with or without cells to trigger regeneration in vivo, and iii) ex vivo tissue models to study tissue formation and pathological processes in combination with drugs (Figure 1).

Strategies for designing tissue engineering constructs

Figure 1. Strategies for designing tissue engineering constructs: Different fabrication techniques can be used to combine molecular and nanomicrometer size building blocks to form 3D constructs for implantable scaffolds, injectable applications, or ex vivo tissue models.

Material Selection for Tissue Constructs

Over the last few decades, researchers have developed two major paths for the fabrication of tissue constructs in different sizes and complexities: implantable scaffolds and injectable hydrogels. Scaffolds for implantation or ex vivo tissue models can be built with intricate architectures, since the materials are processed outside the body. For example, one technique is decellularization of natural tissues to create structures very similar to the original (Figure 2A). In the first animal experiment in 1995, decellularized small intestinal submucosa from pigs were implanted in dogs, resulting in improved Achille’s tendon repair. In 2010, a decellularized trachea was successfully implanted in a 10-year-old child. In initial experiments, tissue was decellularized by immersing it in a detergent solution. More recent methods use the native vascular network for perfusion and recellularization,2 allowing for the creation of whole liver and other types of grafts.3 Even though decellularization techniques have improved significantly and now maintain most of the essential extracellular matrix (ECM) components, many challenges remain. These include inflammatory responses, better preservation of the biochemical and physical integrity entire ECMs, and optimized bioreactors for efficient recellularization.

To study material/cell interactions in a systematic way, scientists are building constructs using both natural and synthetic materials, two components with important differences. While natural materials, such as collagen, fibrin, and Matrigel® inherently contain many biological signals, artificial synthetic ECMs (aECM) are prepared with a small number of well-defined building blocks. To form functional tissue, these matrices mimic the mechanical, biochemical, and structural properties of the ECM, including degradability. In order to incorporate live cells during preparation, physiological conditions and biocompatible chemistry must be maintained. Nowadays, most materials are biohybrid systems that combine the controllability of synthetic materials with the biological activity of natural compounds.

Fabrication Methods for Solid Implantable Scaffolds

In addition to composition, material properties have to be designed at multiple scales, from the physical, chemical, and biochemical properties of the molecules to the nano- and micrometer scale of the porosity and structural elements, all the way up to the macroscopic architecture. Solvent casting is one of the earliest methods used for the preparation of implantable scaffolds, where a polymer is dissolved in an evaporating organic solvent using leachable porogens such as salt particles.4 However, the cytotoxicity of organic solvents has made it impossible to add cells or proteins. Subsequently, more biocompatible methods such as freeze-drying and gas foaming particulate leaching have been developed that enable both the incorporation of bioactive molecules and the creation of more complex architectures. For example, aligned channels can be fabricated to mimic oriented tissues, such as the spinal cord (Figure 2B). Another method is fiber spinning, which allows the creation of fibrous mats with random or oriented fibers in which the diameter, fiber density, and topography can be controlled. The most popular spinning method is electrospinning, where electrostatic forces induce a Taylor cone, accelerating the solution towards a collector with the opposite polarity, producing synthetic fibers as the solvent evaporates (Figure 2C).5 The first electrospun mats for tissue engineering were used as a vascular prosthesis in 1978. Another method, solvent-assisted spinning, does not apply an electric field, but the fibers are mechanically pulled and collected by a rotating drum. This enables the formation of fibers with larger diameters and more controllable inter-fiber distances, fiber orientation, and fiber topography. In wet spinning, polymers are dissolved in a non-volatile solvent and extruded inside a solution, which washes the solvent out, resulting in rapid fiber production. Thermostable polymers can be molten without using a solvent and pressed out of a nozzle during melt-spinning, then cooled by air flow between the nozzle and the collector. This technique has the advantage of not requiring further washing steps.

 Fabrication of tissue constructs in different sizes and complexities

Figure 2. A) Decellularization of tissues and subsequently recellularization for the production of vascularized tissue engineering constructs for potential implantation. B) Production of multiple channel spinal cord bridges using a gas foaming/particulate leaching technique. C) Electrospinning of nano- and micrometer sized fibers for implantable scaffolds.

Fabrication Methods for Hydrogel-based Scaffolds

The previously mentioned fabrication methods create solid implantable scaffolds, but natural tissues are often softer and viscoelastic. Hydrogels are water-containing networks that mimic these properties, and are prepared by crosslinking hydrophilic natural or synthetic proteins, polymers, or sugars. Crosslinks are achieved through physical interactions and reversible bonds, and/or by chemical reactions and covalent bonds. For example, bonds in gelation can be triggered by ionic interactions, pH, temperature, or enzymes.6 Synthetic polymers, glycosaminoglycans (GAGs), and recombinant proteins are used to prepare hydrogels in order to minimize batch-tobatch variability and avoid potential pathogenic contaminations. In order to create synthetic hydrogels, building blocks such as poly(ethylene glycol) (PEG) or polyacrylamide, bioactive peptides, ECM fragments, and proteins are coupled or mixed within the network. If the ability to induce degradation is desired, the linkers between the molecules or the molecules themselves can be designed with ester bonds amenable to hydrolysis, or with matrix metalloproteinase (MMP) cleavage sites. Combining an understanding of biocompatible chemical reactions, biochemistry, and physical chemistry, a large toolbox of hydrogels with different mechanical and biological properties has been established for a wide variety of applications.

The remaining challenge is to decouple the effects of different parameters, such as the mesh size, stiffness, degradation rate, ligand spacing, and topography, on cell behavior. Another limitation is that most conventional hydrogels have nano-size pores, which prevent cell migration when covalent and nondegradable crosslinks limit efficient nutrient perfusion. When cells are mixed inside the hydrogels, either matrix degradation or reversible crosslinks are required to enable the cells to spread and migrate. Therefore, hydrogels with dynamic bonds have recently been designed and studied to better mimic the viscoelastic, strain-stiffening, and fibrous properties of the native ECM. To further improve infiltration of endogenous cells in vivo (which is challenging as cells choose the least resistant path around the hydrogel), macroporous hydrogels have been produced. Here, strategies differ for implants and injectable materials. For implants, sacrificial porogens or templates can create macroscopic pores and hydrogels with shape-recovering complex structures (Figure 3A), 7 but may not allow for simultaneous cell encapsulation into the hydrogel.

To address this problem, bioprinting has emerged as a way to print 3D hydrogel constructs in combination with cells and proteins. There are three methods of bioprinting: inkjet printing, micro-extrusion, and laser-assisted printing (Figure 3 B–D).8 While inkjet printing is a cheap and accessible method that deposits the material droplet-wise, micro-extrusion prints continuous lines and patterns, allowing for higher viscosities and cell densities. However, extrusion is slower, and maintaining cell viability is challenging due to high shear stresses. In the case of laser-assisted bioprinting, a focused, pulsed laser shoots micro-beats from a ribbon onto specific positions on a receiving substrate. Laser-assisted bioprinting is more complex, elaborate, and costly, and requires fine-tuning for each application and cell type. However, laser-assisted bioprinting operates over a large range of viscosities and high cell densities, and cells remain viable during the process. While exceptional progress has been made in these printing techniques, a new generation of bioinks is needed to mimic the ECM properties without sacrificing printability in the presence of cells. The ideal bioink would be a non-cytotoxic liquid, that crosslinks only on demand (to prevent clogging of the nozzle), enables high cell densities, and avoids subjecting the cells to thermal and mechanical stress. The crosslinking trigger must not damage the cells and the crosslinking kinetics must be controllable and rapid enough to achieve high resolution. The resulting hydrogel must be cell-adhesive and degradable with a tunable stiffness. As an alternative to bioink, the Kenzan method prints cell spheroids directly without using a supporting hydrogel-scaffold (Figure 3E). Here, cell spheroids are robotically placed on micro-needles to achieve a specific structure — for example vascular tubes after fusion of the spheroids.9

Fabrication methods for hydrogel-based scaffolds

Figure 3. A) Sacrificial templates to create macroscopic hydrogels. B) Droplet-wise inkjet bioprinting. C) Continuous line printing using micro-extrusion. D) Laser-assisted bioprinting using focused pulsed laser to shoot micro-beats on a receiving substrate. E) Kenzan method to print cell spheroids directly without using supporting hydrogel-scaffolds.

As previously mentioned, hydrogel precursor solutions can be injected as a liquid and then cross-linked in situ under physiological conditions using a minimally invasive procedure. This enables adaptation to irregular shapes inside the body, and forms a close interface with surrounding tissue provided that the crosslinking mechanism is fully optimized — fast enough to avoid leakage but slow enough to allow handling and precise injection. There are two major shortcomings with this method: i) the lack of macroscopic pores to facilitate infiltration of endogenous cells and ii) the formation of hierarchical and oriented structures to direct cell growth.

One approach to overcoming the first problem is to inject a pre-crosslinked microgel containing some amount of remaining reactive groups; after injection the microgels are linked together via a secondary crosslinking mechanism. Macroscopic pores of various size can be created depending on the diameter of spherical microgels. These structures are called interconnected microporous annealed particle scaffolds (MAPs), and they significantly enhance cell infiltration and tissue healing compared to conventional hydrogels with the same polymer composition (Figure 4A).10 Microgels for this purpose are prepared via microfluidics using different materials, such as PEG or alginate, and linked together covalently either by enzymatic reactions or via specific interactions.11 To obtain shear-thinning properties inside these MAPs, microgels are crosslinked using guest-host chemistry, enabling this material to be used as a bioink.12 An alternative method for obtaining larger pores is the creation of ECM-like fibrous structures via self-assembly of highly defined molecules using supramolecular chemistry. The pore size and strain-stiffening properties of the structure vary depending on the properties and lengths of the molecular building blocks (Figure 4B).13 These microporous structures have the advantage that cells do not interact with the ECM, but can interact with each other. These cell-cell interactions are crucial for many biochemical processes, including cell organization and maturation. For example, the stemness of neural progenitor cells in the absence of differentiation factors is highly dependent on the degradation rate and mechanism of the hydrogel, and thus on the ability of the cells to remodel the matrix over time and exert cadherin-mediated cell-cell interactions.14 Furthermore, during the development of the mesenchyme, cells initially exhibit significant interactions with the ECM, but over time they start to interact more with each other via cell-cell interactions, a process that is not yet completely understood.15

The second problem is the often isotropic character of injectable hydrogels. To induce hierarchy and orientation into these hydrogels, many different technologies are being developed, including photopatterning, oriented self-assembling nanofibers, and the formation of guiding structures using an external magnetic field. Photo-patterning uses light to create high resolution structures of biochemical and/or mechanical patterns inside a hydrogel that can act as guiding cues. In this technique, a focused laser beam activates photosensitive molecules, resulting in local crosslinking and thus stiffening, degradation, exposure of functional groups, or post-modifications (Figure 5A).16 Two-photon lithography can achieve better resolution in the z-direction but, except for laser ablation, only relatively thin hydrogel layers of several millimeters can be photopatterned, so the fabrication of larger tissues, like heart or liver are currently not possible using this method. In the case of self-assembly, peptide amphiphiles turn into high aspect ratio nanofibers, which can be aligned over a centimeter range by manually dragging the fibrous structure into salty media (Figure 5B). These supramolecular filaments can be functionalized with bioactive peptides, and support aligned neuronal cell growth.17 Even though this technique allows for a minimally invasive in vivo application, variation of the nanofiber dimensions is limited, and the alignment of the filaments depends on the direction of the flow inside the needle. As an alternative, magnetic fields can be applied to form strings of magnetic particles inside hydrogels (Figure 5C). These strings then function as guiding elements to orient cells, but their shapes and dimensions are difficult to control and a high concentration of cytotoxic iron oxide particles is required.

Interconnected microporous annealed particle scaffolds (MAPs)

Figure 4. A) Assembly of spherical microgels using different interactions to promote cell infiltration into the scaffold and cell-cell interactions. B) Stressstiffening fibrous structures using polyisocyanopeptides. Different polymer lengths can be used to control the stress-stiffening properties.

In the interest of obtaining more control of the anisotropy of injectable hydrogels with a minimal amount of iron oxide, the Anisogel (Anisotropic hydrogel) concept was developed. Anisogel is a hybrid hydrogel that consists of two components: magnetoresponsive, micron-scale, rod-shaped guiding elements and a precursor solution. After injection, the guiding elements orient in the presence of an external magnetic field in the millitesla range within one minute, while the surrounding precursor solution crosslinks to fix the alignment of the oriented elements within approximately 2–3 minutes. The elements can be rodshaped microgels, produced either by in-mold polymerization or microfluidics, or short polymeric fibers, fabricated using a spinning/microcutting combinatorial method (Figure 5 D–F). The magneto-responsiveness is achieved by incorporating small amounts of superparamagnetic iron oxide nanoparticles (SPIONs) inside the micro-objects during fabrication. This technology allows for high degree of control over individual parameters like stiffness, bioactivity, topography, dimensions, and concentration of the guiding micro-elements, as well as the stiffness and bioactivity of the surrounding hydrogel (Figure 5G).18

Technologies to induce hierarchy and orientation into hydrogels

Figure 5. A) Photopatterning resulting in local crosslinking or degradation, exposure of functional groups, or post-functionalization. B) Self-assembly of peptide amphiphiles. C) Aligned strings of magnetic particles in a magnetic field. D) In mold polymerization method and E) Microfluidics for preparation of rod-shaped microgels (PEG-SH: PEG-thiol and PEG-VS: PEG-vinylsulfone). F) Spinning/microcutting combinatorial method for short fiber production. G) Concept of the Anisogel.

Summary and Outlook

While most of the technologies described are still being developed and optimized, they are also being used to identify the factors that control cell behavior and tissue formation, and to turn biopsy-samples or cells into organoids to study the formation of mini-tissues, the onset of pathologies, and the effect of specific drugs (Figure 6). While Matrigel® is still the most efficient hydrogel material to create organoids, it is not well characterized, making a reductionist approach impossible.19 Therefore, synthetic approaches are under investigation to control and guide cell behavior in a more robust and reproducible manner.20

Formation of organoids by embedding stem cells in synthetic or natural hydrogels and differentiating/maturating them by using differentiation factors.

Figure 6. Formation of organoids by embedding stem cells in synthetic or natural hydrogels and differentiating/maturating them by using differentiation factors.

While organoids are already a useful tool for ex vivo tissue models, they are still on the millimeter scale, limiting their use as implantable tissue engineering constructs for the replacement of damaged tissue. However, as our understanding grows of tissue formation and the role of cell-ECM and cell-cell interactions during the stages of stem cell differentiation and maturation, these techniques hold promise for generating complex tissue constructs through natural organizational processes, without the need to guide each cell individually. By identifying the minimal amount of “outside” triggers required in an artificial environment, we may be able to target and activate natural healing and regeneration processes, which will be essential for generating entire organs in the future.

Place ES, Evans ND, Stevens MM. 2009. Complexity in biomaterials for tissue engineering. Nature Mater. 8(6):457-470.
Ott HC, Matthiesen TS, Goh S, Black LD, Kren SM, Netoff TI, Taylor DA. 2008. Perfusion-decellularized matrix: using nature's platform to engineer a bioartificial heart. Nat Med. 14(2):213-221.
Uygun BE, Soto-Gutierrez A, Yagi H, Izamis M, Guzzardi MA, Shulman C, Milwid J, Kobayashi N, Tilles A, Berthiaume F, et al. 2010. Organ reengineering through development of a transplantable recellularized liver graft using decellularized liver matrix. Nat Med. 16(7):814-820.
Choudhury M, Mohanty S, Nayak S. 2015. Effect of Different Solvents in Solvent Casting of Porous PLA Scaffolds?In Biomedical and Tissue Engineering Applications. j biomater tissue eng. 5(1):1-9.
Sill TJ, von Recum HA. 2008. Electrospinning: Applications in drug delivery and tissue engineering. Biomaterials. 29(13):1989-2006.
Rose JC, De Laporte L. 2018. Regenerative Medicine: Hierarchical Design of Tissue Regenerative Constructs (Adv. Healthcare Mater. 6/2018). Adv. Healthcare Mater.. 7(6):1870026.
Torres?Rendon JG, Femmer T, De Laporte L, Tigges T, Rahimi K, Gremse F, Zafarnia S, Lederle W, Ifuku S, Wessling M, et al. 2015. Bioactive Gyroid Scaffolds Formed by Sacrificial Templating of Nanocellulose and Nanochitin Hydrogels as Instructive Platforms for Biomimetic Tissue Engineering. Adv. Mater.. 27(19):2989-2995.
Murphy SV, Atala A. 2014. 3D bioprinting of tissues and organs. Nat Biotechnol. 32(8):773-785.
Moldovan NI, Hibino N, Nakayama K. 2017. Principles of the Kenzan Method for Robotic Cell Spheroid-Based Three-Dimensional Bioprinting. Tissue Engineering Part B: Reviews. 23(3):237-244.
Griffin DR, Weaver WM, Scumpia PO, Di Carlo D, Segura T. 2015. Accelerated wound healing by injectable microporous gel scaffolds assembled from annealed building blocks. Nature Mater. 14(7):737-744.
Hu Y, Mao AS, Desai RM, Wang H, Weitz DA, Mooney DJ. Controlled self-assembly of alginate microgels by rapidly binding molecule pairs. Lab Chip. 17(14):2481-2490.
Mealy JE, Chung JJ, Jeong H, Issadore D, Lee D, Atluri P, Burdick JA. 2018. Injectable Granular Hydrogels with Multifunctional Properties for Biomedical Applications. Adv. Mater.. 30(20):1705912.
Kouwer PHJ, Koepf M, Le Sage VAA, Jaspers M, van Buul AM, Eksteen-Akeroyd ZH, Woltinge T, Schwartz E, Kitto HJ, Hoogenboom R, et al. 2013. Responsive biomimetic networks from polyisocyanopeptide hydrogels. Nature. 493(7434):651-655.
Madl CM, LeSavage BL, Dewi RE, Dinh CB, Stowers RS, Khariton M, Lampe KJ, Nguyen D, Chaudhuri O, Enejder A, et al. 2017. Maintenance of neural progenitor cell stemness in 3D hydrogels requires matrix remodelling. Nature Mater. 16(12):1233-1242.
Cosgrove BD, Mui KL, Driscoll TP, Caliari SR, Mehta KD, Assoian RK, Burdick JA, Mauck RL. 2016. N-cadherin adhesive interactions modulate matrix mechanosensing and fate commitment of mesenchymal stem cells. Nature Mater. 15(12):1297-1306.
Kloxin AM, Kasko AM, Salinas CN, Anseth KS. 2009. Photodegradable Hydrogels for Dynamic Tuning of Physical and Chemical Properties. Science. 324(5923):59-63.
Zhang S, Greenfield MA, Mata A, Palmer LC, Bitton R, Mantei JR, Aparicio C, de la Cruz MO, Stupp SI. 2010. A self-assembly pathway to aligned monodomain gels. Nature Mater. 9(7):594-601.
Rose JC, Cámara-Torres M, Rahimi K, Köhler J, Möller M, De Laporte L. 2017. Nerve Cells Decide to Orient inside an Injectable Hydrogel with Minimal Structural Guidance. Nano Lett.. 17(6):3782-3791.
Dutta D, Heo I, Clevers H. 2017. Disease Modeling in Stem Cell-Derived 3D Organoid Systems. Trends in Molecular Medicine. 23(5):393-410.
Gjorevski N, Sachs N, Manfrin A, Giger S, Bragina ME, Ordóñez-Morán P, Clevers H, Lutolf MP. 2016. Designer matrices for intestinal stem cell and organoid culture. Nature. 539(7630):560-564.